摘要:

ii. 133 General and Physical Chemistry. Continuous Spectrum of Hydrogen. H. B. LEMON (Nature 1924 113 127-128).-The continuous spectrum of hydrogen is developed with unusual intensity when discharge takes place through tubes provided with hot cathodes and operated at low potentials of the order of 100 v. The emitted light extends to the yellow-green region. This continuous spectrum seems to have no sharp limits; it bears no relation to the secondary or to the series lines and in this respect differs entirely from similar spectra in the X-ray region. A. A. E. Ultra-violet Line Spectrum of Chlorine. E. V. ANGERER (2. wiss. Phot. 1924 22 200-209).-The ultra-violet line spectrum of chlorine has been measured over the range 35224- 20704 8. Some 322 lines have been measured and tabulated and compared with Jevons’ value (A.1923 ii 274). The spectrum obtained by Jevons contained oxygen lines due to impurities in the discharge tube. To obviate this in the present work the author has adopted a streaming method. The spectrum measurements were made with a Steinheil spectrograph which gave a dispersion of 15 8. at 3000 8. and 4.4 A. at 2000 8. The line 2300 $. of the iron arc was used as a standard. A group of six lines lying between 2093 and 2086 8. is found to give four pairs of lines with constant frequency difference. Attention is directed to the fact that a very large number of lines between 2400 and 2100 8. recorded by Jevons are absent in the author’s photographs. Extension of the Spark Spectraof Lead Bismuth Antimony and Thallium to the Extreme Ultra-violet.L. BLOCH and E. BL~CH (Cornpt. rend. 1924 178 472474).-In continuation of previous work (cf. A. 1924 ii 4) the authors have measured the wave-lengths of new lines in the spectra in question. The measure- ments have been extended to wave-lengths of about 1300 A. Results have been obtained for 23 new lines for lead 11 for bismuth 20 for antimony and 14 for thallium Most of the new lines are situated in the extreme region; others of feeble intensity are situated amongst lines previously observed. H. N. RUSSELL (Xature 1924 113 163).-A number of the strongest lines in the spark spectrum of ahminium which do not fit into Paschen’s system (A. 1923 ii 672) are considered. Five of the seven lines of intensity 10 or more are found to show a close corre- spondence with lines in the arc spectrum of magnesium.A. ,4. E. Mass-spectrum of Indium. F. W. ATON (Nature 1924 113 192).-The mass-spectrum of indium shows only one line VOL. CXXVI. ii. 6 J. F. S. A. B. H. Singlet Series in the Spark Spectrum of Aluminium.ii. 134 ABSTRACTS OF CHEMICAL PAPERS. a t 115 no deviation from the whole number rule being observed although its intensity was not sufficient to exclude the possibility of the presence of a small percentage of a second isotope. J. J. HOPFIELD and S. W. LEIFSON (Astrophys. J. 1923 58 59-63).- A careful determination has been made of the wave-lengths of thirty- seven characteristic lines between A 800 and h 1800 which occur in condensed discharge spectra of impure hydrogen oxygen nitrogen helium and air.The results which are accurate to h 0 . 1 A. are intended to serve as reference standards. Resonance and Ultimate Lines in the Flash Spectra obtained at different Levels in the Solar Atmosphere. F. CROZE (Compt. rend. 1924 178 200-202).-Resonance lines which represent combinations derived from lines belonging to one and the same series are remarkable for their persistence. Such lines can be detected at relatively very high levels in the solar atmosphere. On the other hand resonance lines which represent combinations derived from lines belonging to different series are much less per- sistent. The resonance line ? 4571 of magnesium and the lines 5432.6 and h 5394.7 of manganese are of the latter type and are only shown by flash spectra obtained a t relatively low elevations.Whether resonance lines are ultimate lines depends therefore on the character of the resonance lines (cf. this vol. ii 1). The Mass of the Particles which Emit the several Band Spectra attributed to Nitrogen. M. DUFFIEUX (Compt. rend. 1924 178 4 7 4 4 7 6 ) .-The limiting orders of interference fringes produced by the various band spectra emitted by nitrogen under the electric discharge have been studied to determine the masses of the particles giving rise to the emission. The second positive group has been previously (Hamy Compt. rend. 1913 157 142) attributed to the nitrogen atom. The bands of the second negative group and those said to be due to cyanogen have now also been attributed to the atom whilst those of the first positive group must be due to the nitrogen molecule. Observations of the Doppler Effect in Canal Ray Line and Band Spectra.H. RAU (Ann. Phgsik 1924 ivy 73 266-271).- Particulars are given of a method of observation of the Doppler effect in the canal ray discharge in various gases. In oxygen the effect was clearly observed in the case of the lines 4368 and 3947. In front of the cathode all the " negative '' bands in the nitrogen spectrum included between 5000 and 3900 A. showed the effect a t all values of the exciting potential drop employed ranging from 3000 to 35,000 v. Behind the cathode the Doppler effect was first observed at a somewhat higher voltage drop of about 9000 v. The observed effect increased with increase of voltage drop and decrease of pressure in the discharge tube.Theresults indicate that the " negative '' nitrogen bands are attributable to the presence of positively-charged nitrogen molecules ( N2+) in accordance with the result obtained by Wien (A. 1923 ii 349). The effects observed A. A. E. Wave-length Standards in the Extreme Ultra-violet. A. A. E. E. E. T. A. B. H.GENERAL AND PHYSICAL CHEMISTRY. ii. 135 with the nitrogen arc lines 4110 agreed with the assumption that the arc lines are attributable to the presence of neutral nitrogen atoms. The Doppler effect was not observed in the case of the positive nitrogen bands. In the multilinear spectrum of hydrogen the effect was observed behind the cathode in the case of the lines 4723 4634 4573 4568 4213 4177 4171-5 and 4063B. The results indicate that the multilinear spectrum is attributable to neutral hydrogen molecules.A relatively weak Doppler effect was observed behind the cathode in the case of the helium line 4686 but none in front of the cathode. The Zeeman Effect of the Multiplets of Molybdenum. M. A. CATALAN (Anal. Fis. Quim. 1923,21 527-532).-Calculated values for the Zeeman effects of the lines of the multiplets of molyb- denum show complete agreement with observed values. The Theory of Multiplets and their Zeeman Effects. A. SOMMERFELD (Ann. Physik 1924 iv 73 209-227).-1n continua- tion of previous work (A. 1923 ii 355) the author discusses mathematically and principally from the arithmetical and geo- metrical point’s of view the character of the inner quantum numbers associated with the various classes of multiplet spectral lines.It is shown inter ulia that in the case of the even terms in the multiplet series i.e. doublets quartets etc. the inner quantum number associated therewith is an odd multiple of 0.5 whilst in the case of the odd terms the associated inner quantum number is a whole number. Landk’s formula relating to anomalous Zeeman effects (2. Physik 1923 15 189) is applicable as a special case to the magneto-mechanical anomaly investigated amongst others by Barnett. Rydberg’s rule concerning Dhe character of the multi- plicity of the lines in the spectrum of an element is re-stated as follows. In the spectra of elements arranged according to the periodic classification odd and even number spectral terms occur alternately and the maximum multiplicity of the lines in any given case is greater by unity than the number expressing the maximum valency of the element concerned.Magneton numbers characteris- ing the elements are in like manner when expressed in Bohr quantum units alternately odd and even and the maximum number characterising any element is equal to the number expressing the maximum valency of the element. It is assumed that the s-con- dition represents the normal or fundamental state of an atom. J. S. G. T. G. W. R. J. S. G. T. Arrangement of the Vanadium Lines in Multiplets. 0. LAPORTE (Physikul. Z . 1923 24 510-515).-Lines additional to those already classified in multiplet series by the author (Natzirwiss. 1923,11 779) are classsed as quartets and sextets and a combination of these two systems. The existence of the system of sextets is in accordance with anticipation based on the application of Rydberg’s rule to the position of vanadium in the periodic classification.The quartet system is characterised by two groups of terms between which a cyclical combination relation exists such that the frequencies of the lines in a fourth quartet may be calculated from the frequencies 6-2ii. 136 ABSTBBCTS OF CHEMICAL P&PEflS. of the lines composing three quartets. An approximate evaluation of terms in the series is effected by consideration of two series of the type d-np in the respective sextet and quartet systems The occurrence of widely separated super-Bergmann terms F* in the series representative of the spectrum of vanadium is confirmed by the existence of a multiplet which must according to its structure and characteristic Zeeman effect be classified as belonging to the combination FP*.The Relation between the Spectra and the Sizes of the Alkali Metal Atoms. L. A. TURNER (Astrophys. J. 1923 58 176-194).-1t is shown that the differences between the spectra of the different alkali metals can be related to the differences of the sizes of the structures immediately underlying the valency electron. Atomic models in accord with modern theories of atomic structure are assumed the nucleus (so far as the outer orbits are concerned) being regarded as having a charge +9e and being surrounded by eight electrons and a valency electron; the configuration of eight is assumed to have a mean field of force which is geometrically similar but differs in size for the various atoms.The application of the theory developed on these lines leads to a fairly good co- ordination of the spectra of the alkali metals in terms of the assumed (and apparently reasonable) sizes of the atomic kernels. Results are also obtained which confirm the qmntum numbers of the p d and b terms as given by Bohr. The radii of some of the electron orbits and the approximate mean field of force in the outer parts -of the orbits of the valency electron have been calculated. The radii ( x 10-8) are found to be as follows lithium 2.38 ; sodium 2.72 ; potassium 3.45 ; rubidium 3.61 ; czsium 3.94. The results also indicate that in the crystalline form the atoms are present as positive ions the electrons fitting into the structure rather than following their usual orbits.The application of the method to the 4b terms of the alkaline-earth metal spectra shows that the kernel of each must be much smaller than that of the corresponding alkali metal. Intensity Distribution Series Formulae and Excitation Function in reference to the Spectra of the Alkalis. H. BARTELS (2. Physik 1924 20 398-412).-A theoretical paper in which the relative intensities of series lines are discussed with reference to the so-called excitation function F( V ) . This affords a measure of the percentage of inelastic impacts which occur in the collisions between atoms and electrons and its value depends on the electron velocity (measured in volts by V ) . On the assumption that for these inelastic impacts azimuthal quantum numbers can be determined according to the principles of selection which are applicable to the emission of radiation the author deduces a relation between the excitation functions characterising the various energy levels of the atom and the various terms in the series formuh for the corresponding spectrum.The result is shown to be in approxi- mate agreement with observations on the distribution of intensities in the spectra of the alkali metals. J. S. G. T. A. 9. E. J. S. G. T.GENERAL AND PHYSICAL CHEMIS!L’RP. ii. 137 The Characteristic Vibration Frequency of an Element. J. E. P. WAGSTAFF (Phil. Mug. 1924 vi 47 84-90).-A simple method of deducing Einstein’s relation between the characteristic frequency V the compressibility K the atomic volume V and the atomic weight M of an element in the form v = PV1’6/~1”M1!2 where p is theoretically a constant equal to 2-56 x lo7 is given.Values of v calculated by means of this equation are shown to be in fair agree- ment with those given by Einstein and Lindemann in the case of fourteen elements. It is shown that substances for which the Einstein law holds are those in which the atomic energy is due to a state of strain. Elements characterised by values of ( ~ K V ~ V ~ ~ large compared with 2.54 x lo7 possess atomic energy in excess of that attributable to strain energy. By plotting values of ( ~ K V ~ V ~ ’ ~ ) as ordinates against the respective values of the atomic numbers of the elements as abscissae a periodic curve closely following the Mendelgev classification of the elements is obtained. In this curve in contrast x-ith the atomic volume curves of Meyer the halogen elements occupy the peaks.A well-defined line drawn through the positions indicating the alkali metals separates the non-metals above the line from the metals. Silicon is the only exception amongst the elements considered and lies below the line. Energy of the Characteristic K-radiation from certain Metals. E. C. UNNEWEHR (Physical Rev. 1923,22,529-538).- The author has investigated the K-characteristic emission from silver rhodium copper and chromium in order to ascertain the dependence of energy of emission on applied voltage and upon the atomic number the relation of the energy of the K-characteristic emission to the energy a t that wave-length in the continuous spectrum and the ratio of the energies of the a and the p character- istic emission.Suitably corrected values for the intensities were linear functions of the square of the applied voltage. The ratios of the intensities of the a and p radiations were silver 4-85; rhodium 4-75 ; copper 6.53 ; chromium 7.36. The absorption coefficient of rhodium for its K-radiation is 222 per cm. (Ra) and 178 per cm. (KP). The absorption coefficients of mica for K - radiation from chromium copper rhodium and silver are 227 89.2 7.7 and 5.1 per cm. for Ka and 171 66 5.5 and 3.6 per cm. for KP respectively. A Re-investigation of the p-Ray Spectrum of Radium-B and Radium-C. C. D. ELLIS and H. W. B. SKINNER (Proc. Roy. Xoc. 1924 A 105 165-184).-In continuation of previous work (A.1924 ii SS) the authors have determined the energy values and approximate relative intensities of thirty-one lines in the P-ray spectra of radium-B and of sixteen lines in the p-ray spectra of radium-C‘ for values Hp up to 2390 H denoting the strength of the magnetic field and p the curvature of the rays in the field. The results are tabulated and confirm the conclusion that the groups are due to the conversion of y-rays in the various electronic levels. Although the majority of the p-ray lines from radium-B (atomic number 82) are due to conversion in atoms of this atomic number J. S. G. T. A. A. E.ii. 138 ABSTRACTS OF CHEMICAL PAPERS. a group of lines of low energy has been found tlo be due to conversion in atoms of atomic number 83. The relative absorbing powers of the different L sub-groups are found to depend on the frequency of the absorbed y-ray.The Interpretation of @-Ray Spectra. C. I>. ELLIS and H. W. B. SKINNER (Proc. Roy. Xoc. 1924 A 105 lS5-198).- Measurements of the frequencies of 7-rays emitted by radium-B and radium4 (see preceding abstract) confirm the conclusion that these rays are due to transitions between stationary states in the nucleus. The probable energy levels which are involved in these transitions in the two substances are discussed. The two sets of levels exhibit striking similarities and the evidence on t'he whole is strongly in favour of the hypothesis that the dynamics of the nucleus can be interpreted in terms of the quantum theory. The change in the relative absorbing powers of the L sub-groups is shown to be in agreement with Robinson's results derived from analysis of the electrons ejected from a target by homogeneous X-rays in a magnetic field (A.1923 ii 818). S. BARRATT (Proc. Roy. Xoc. 1924 A 105 221-225).-A new series of bands each consisting of fine lines separated in the middle by intervals of a few A. but much more closely packed near the heads of the bands has been observed in the absorption spectrum of a mixture of sodium and potassium vapours. The respective wave-lengths of the heads of these bands are 5955.0 59314 5867.0 5846.0 5824.6 5803.7 5783.5 5764.3 5746.1 5728-6 5712.1 5696.0 5680.9 5667-4 5654.1 and 5645.2 a. The intensity of absorption decreases considerably a t the short wave-length end of the group of bands.It is suggested that the bands are attributable to sodium-potassium molecules present in the mixed vapour. Absorp- tion of the magnesium line 4571 8. which is excited by low-voltage electrons was consistently observed in a column of mixed vapours in which a little magnesium vapour was present (cf. NcLennan Distribution of Electrocs between the L Levels of Elements. A. DAUVILLIER (Compt. rend. 1924,178,476479).-From measure- ments of the coefficients of absorption of gold as obtained with a Bragg spectrometer it is concluded that in the gold atom the electrons in the L levels are 2L 2L2 and 4-4. Other results using different methods will be published later and are in agreement with the above. E. E. T. The Influence of Variation in Concentration on the Absorp- tion Spectra of Dye Solutions.W. C. HOLMES (Ind. Eng. Chem. 1924 16 3540).-The variation in the wave-length of maximum absorption in the spectra of more than one hundred dyes with varia- tion in concentration of the solution was studied. In the majority of cases displacement of this wave-length with increasing dilution towards the red end of the spectrum was observed. 9 shift towards the blue occurred in a few cases. An example of the dependence of this phenomenon on chemical constitution is given by the triamino- J. S. G. T. J. S. G. T. The Absorption Spectra of Mixed Metallic Vapours. A. 1916 ii 591). J. S. G. T.GENERAL AND PHYSICAL CHEBUSTRY. ii. 139 derivatives of triphenylmethane in which it appears only in cases of substitution within the amino-groups. In such cases measure- ments of solutions of intermediate concentration show that two widely separated absorption bands exist the relative intensity of which depends on concentration.It was shown that variations in hydrogen-ion concentration temperature or chemical purity could not account for these results. Whilst change in the state of molecular aggregation or similar colloidal phenomena may explain cases of symmetrical shifting of the absorption band in very dilute solution the discontinuous modification referred to above appears rather to be due to molecular tautomerism. Absorption Spectra of Amino-acids. L. MARCHLEWSKI and A. MoRoz.-(See i 272.) Absorption of Ultra-violet Light by Organic Com- Absorption Spectra of Indole Derivatives. F. W. WARD.- (See i. 314.) Absorption Spectra of Kynuretic Acid and Related Quinoline Derivatives. Ultra-violet Absorption Spectrum of Gaseous Ammonia.FERRI~ES (Compt. rend. 1924 178 202-205).-The absorption spectrum of gaseous ammonia shows a large band beginning sharply a t 2265 8. and extending with increasing intensity towards the region of shorter wave-length. Narrower bands are superposed on the broad band and the first five of these have been resolved into the following triplets (in each of which the two rays of shorter wave-length are the most intense) 2247 2245 and 2239; 2213 2211 and 2206; 2171,2167 and 2163; 2129,2126 and 2122 ; 2086 2083 and 2079 8. Excepting the first pair the differences between the mean frequencies of the triplets bands are about 900. E. E.T. Chemical Constitution and K Absorption Spectra. I. The Investigation of certain Compounds of Phosphorus. 0. STELLING (2. anorg. Chem. 1923 131 48-56).-The author has examined the K Rontgen absorption spectrum of phosphorus in its various modifications and in combination as hypophosphite phosphite and phosphate. The wave-lengths of the limits of K absorption for the black and violet modifications of phosphorus are identical within the limits of error of the experiment but are greater than for yellow phosphorus. In the hypophosphites phos- phites and phosphates a progressive displacement takes place in the direction of lower wave-length. The salts of any one acid and the acid itself give identical results which are however different from those given by organic derivatives of the same acid in which different numbers of hydrogen atoms are replaced.These results are compatible with the following formule for the respective acids H H OH HO-kO HO*kO HO*h:O c. I. pounds. F. W. m'ARD.-(See i 274.) F. W. WAm.-(See i 315.) I?[ 6H 6Hii. 140 ABSTRACTS OF CHEMICAL PAPERS. The gradual displacement of the absorption limit corresponds with the replacement of hydrogen atoms by hydroxyl groups directly The Absorption of X-Rays by Iron Cobalt Nickel and Copper. F. K. RICHTMYER and F. W. WARBURTON (Physical Rev. 1923 22 539-545).-ThinY uniform sheets of the metals were examined bj- means of a spectrometer and the X-ray absorption coefficients of iron cobalt nickel and copper for wave-lengths below the K absorption limit were measured.The sequence of values of the atomic coefficients is found to be that not of the atomic weights but of the atomic numbers according to the relation p=22*4 > 10-27N4h3+~ which holds within the limits of experi- mental error for all the ten elements for which results have been obtained. except iron the experimental value for which is 3% too low. A. A. E. Action of Red and Infra-red Radiations on Photo- luminescent Substances. M. CURIE (J. Chim. physique 1923 20 437463; cf. A. 1923 ii 812).-A continuation of previous work in which the author has investigated the influence of temper- ature on the luminescence of fluorite and other substances subjected to the action of light from a quartz mercury lamp which has passed through a nickel oxide light-filter. It is found that some varieties of fluorite are fluorescent whilst others particularly the colourless and rose-eoloured varieties are non-fluorescent.Most varieties commence to show a feeble luminescence a t 50° which increases in intensity slowly to a maximum and then decreases with increasing temperature. The maximum is often only reached a t temperatures above 200'. The luminescence of fluorite shows two bands; a t the ordinary temperature a violet band only is observed but the range of this band depends on the intensity of excitation. At low temperatures below -loo" an intense band extending from red to green is observed. The change of intensity of the two bands with temperature has been determined and plotted and the curves hare been compared with similar curves obtained for zinc sulphide zinc oxide calcium and barium sulphide willemite and synthetic ruby The curves are extremely diverse the phosphorescence disappearing a t very different temperatures.The interpretation of the curves is discussed and it is pointed out that there are indi- cations of a common electronic mechanism operating in fluorescence phosphorescence and thermoluminescence. The index of refrac- tion of the phosphorescent alkaline-earth sulphides has been deter- mined for light of different wave-lengths and it is shown that the relation >.,'l/K=A,=constant cannot be confirmed. Corpuscular Radiation Excited by X-Rays. C. G. BARKLA and ,4. E. &I. &I. DALLAS (Phil. Mag. 1924 vi 47 1-23).-1nvestig- ation of the energy of the corpuscular radiation emitted by metal plates exposed to X-radiation of various frequencies has afforded results differing substantially from those obtained by Sadler (A 1910 ii 2511 Bragg and Porter (A.1911 ii 683) and Beatty (A. attached to the phosphorus. W. H.-R. J. F. S.GENERAL AND PHYSICAL CHEMISTRY. ii. 141 1911 ii 245). Two independent discontinuities are indicated in corpuscular emission due to a- and p-constituents of the K radiation when this was used to excite corpuscular emission. Only few K electrons i.e. those emitted in exclusive association with the K characteristic radiation whatever their origin ha\-e energy of the order hn where n is the frequency of the exciting radiation. From silver and tin these electrons are insufficient in number to account either for the whole K ionisation observed when the sub- stance is in the gaseous form or for the full energy of characteristic radiation on the basis of one quantum of characteristic radiation for one high-speed electron. The results are interpreted as indicat- ing the importance of the part played by electrons with energy equal to h(n-nn,) as found by de Broglie (Solvay Conference Brussels 192 1).J. X. G. T. C. B. BAZZONI and A. T. WALDIE ( J . Franklin Inst. 1924 197 57-72).- Theoretical and practical difficulties connected with the inter- pretation of experimental data for the ionisation and resonance potentials of gases are briefly referred to. The authors show that at pressures ranging from 0.1 mm. to 10 cm. of mercury nitric oxide is not dissociated either by heat radiation or by ultra-violet radiation.Hot metals e.g. nickel cause dissociation of the gas the oxygen liberated combining a t once with the metal. It is shown that the virtual free path of electrons in nitrogen must be several times that predicted by the kinetic theory and that the dimensions of the apparatus employed in the ionisation experiments more especially the grid distance determine the appearance of the higher critical voltages of nitrogen. AU the critical potentials observed by Brandt for nitrogen (A 1922 ii 186) have been confirmed in the present experiments. The evidence suggests that the resonance potential at about 8 v. and the lowest ionisation potential 17.7 v. are characteristic of the nitrogen atom whilst the higher potentials particularly the critical potential a t 24 v.are connected with the ionisation of the molecule. The results Impact Effects in Nitrogen and Nitric Oxide. are in accord with those obtained by Smyth (A. 1923 ii 602). J. S. G. T. Radiation Potentials of Atomic Hydrogen. P. S. OLMSTEAD and K. T. COMPTON (Physical Rev. 1923,22,559-565).-Radiation potentials of atomic hydrogen produced by thermal dissociation of molecular hydrogen with a tungsten furnace which could be maintained at 2800" K were observed a t 10.15 12-05 12-70,13-00 13-17 and 13.27 v. with additional radiation at the ionising potential 13-54 v. The results which are liable to an error of less than 0-05 v. are in agreement with the Lyman spectral series and Bohr's theory. A. A. E. The Protective Action of Anti-oxygenisers against the Changes due to Light on Dyed Fabrics.A. GILLET (Bull. Soc. chim. 1923 33 [iv] 1602-1606; cf. A. 1923 i 676 also Gillet and Giot A. 1923 ii 527).-A claim to priority over Sisley (A. 1923 ii 717). H. J. E. 6*ii. 142 ABSTRACTS OF CHEMICAL PAPERS. The Validity of Ohm’s Law in the Case of Electrolytes subjected to very strong Fields. 31. WIEN (Ann. Physilc 1924 iv 73 161-181).-Experiments are described showing that Ohm‘s law is valid to within the limit of experimental error (esti- mated as about 1%) in the case of conduction by elecfrolytes in which the potential gradient is as high as 500,000 v. per em. This result confirms qualitatively the conclusions of Lenard relating to electrolytic conduction (A. 1920 ii 349). The Magnetons of Weiss and of Bohr and the Constitution of the Atom.B. CABRERA (Anal. Pis. Qzcim. 1923,21,505-526). -The author discusses the relationship between the magnetic moments and magnetons in the case of ions of elements of the iron group criticising in particular the conclusions of Sommerfeld (A. 1923 ii 734). I n the case of the rare earth metals the magnetic moments are multiples of the Weiss magneton and not of the Bohr magneton calculated by Pauli’s formula. The latter formula is compared with that of Langevin. The hypothesis that changes can occur in the orbits of electrons may justify Langevin’s law. J. S. G. T. G. W. R. Discovery and Separation of the Isotopes of Chlorine and the Whole Number Rule. W. D. HARKINS and T. H. LIGGETT ( J . Physical Chern. 1924 28 74-82).-Using a process of diffusion of hydrogen chloride into air through a membrane of porous porcelain and making a cut of 8000 the atomic weight of chlorine was increased by 0.044 unit.If the process of separation had an efficiency of loo% the increase of atomic weight should be 0.086 unit from which it follows that the efficiency of the actual separ- ation was 51 yo. This agrees well with the efficiency of 60 % obtained by Harkins and Broeker (Nature 1920 105 230) with the same porous material but is higher than that of 35% obtained by Harkins and Hayes (A. 1922 ii 140) who used matlerial with larger pores. The efficiency obtained in a cut of 400 was also 51% which is in perfect agreement with that for a cut of 8000. Such an agreement could not be attained unless the process described had actually separated the isotopes for the material separated after each diffusion was purified before the next diffusion was commenced and in consequence any impurities could not have a constant effect.The atomic weights were determined by pre- cipitation of the chlorine as silver chloride and are in good agree- ment with those obtained in independent separations by Harkins and Broeker and by Harkins and Hayes who used a density method for the atomic-weight determinations. A n Experimental Determination of the Rate of Decay of the Short-life Product Radium-C’. J. C. JACOBSEN (Phil. Mag. 1924 vi 47 23-31).-A method of preparation of pure radium- C’ is described. By observation of the numbers of scintillations produced on a zinc sulphide screen placed at various distances from the pure radiozctive material the value of the transformation constant A of radium-C” was found to be 8.4 x lo5 sec.-l A com- parison of this result with the value 5 x lo7 sec.-l deduced from J.F. S.GENERAL AND PHYSICAL CHEMISTRY. ii. 143 t’he formula log A=a+b log R suggested by Geiger and Nuttall (A. 1911 ii 953) in which R denotes the range of the cc-rays and a and b are constants and the consideration of results relating to other products of the uranium-radium sequence suggest a modi- fication of this formula. J. S. G. T. Researches on Cellular Disintegration. Remarks on the Preceding Note. D. Rare Earths. XV. Search for Element i 347.) i 345.) YxTEiu.-(See ii 185.) A. NoDoN.-(See BERTHELOT.-(See Sixt.y-one. L. F. Chemical Valency and Luminous Radiation. I.The Pictet-Trouton Rule. J. DUCLAUX ( J . Chim. physique 1923 20 4 0 3 4 1 0 ) .-A theoretical paper in which the Pictet-Trouton rule is discussed. The deviations from the normal value of &/T exhibited by associated liquids are interpreted by application of the conception of quanta to vaporisation. E. D. EASTMAN ( J . Azncr. Chem. Xoc. 1924 46 3943).-A theoretical paper in which it is argued that there is in the ideal case no sharp dividing line between the crystalline and the (supercooled) liquid states. If the conclusion of Lewis and Gibson (A. 1920 ii 585) is accepted that glasses possess finite entropy a t the absolute zero it is necessary to believe that complicated crystals also do. Foliow- ing Lewis and Gibson ( ( randomness ” is regarded as the source of the entropy differences among glasses and crystals at the absolute zero and the minimum number of atomic positions necessary to define the state of any system is adopted as a measure of its random- ness.From the results of certain plausible assumptions as fo the form of the relationship between entropy and randomness it appears possible that measurable entropy effects may be found in real crystals of the more complex types. Finally a statement of the third law of thermodynamics in the following terms is in keeping with the above ideas. “At the absolute zero entropy changes in all reactions of the same type are equal and in reactions in which the system does not change in type the entropy changes are zero.” J. F. S. The Systematic Doctrine of Affinity.XXTV. W. BILTZ. XXV. w. BILTZ and E. ME1NECKE.-(See ii 168 165.) Crystallisation [Crystal Structure] of Sodium Chlorate and Sodium Chloride. J. BECKENKAMP (2. Physik 1923,20,414).- The crystal structure suggested by Kiby (A. 1923 ii 687) for sodium chlorate requires modification in order that the optical activity of this substance may be accounted for. According to the amended structure the two oxygen atoms in the formula NaOClKg which are connected by double bonds with the chlorine atom are separated in the crystal from the chlorine atom by a J. F. S. Statement of the Third Law of Thermodynamics. 6*-2ii. 144 ABSTRACTS OF CHEMICAL PbPEBS. distance which differs from the distance between the chlorine atom and the third oxygen atom.This difference which is responsible for the optical activity of sodium chlorate cannot be detected by X-ray measurements. Determination of the Vapour Pressures of Ca!sium and Rubidium and a Calculation of their Chemical Constants. D. H. SCOTT (Phil. Maq. 1924 vi 47 32-50).-The respective vapour pressures of casium and rubidium at various temperatures between 50" and 130" have been determined by means of a slightly modified form of Haber's vibrating quartz-fibre manometer (A. 1914 ii 534) with which observations were made of the time t>aken for the amplitude of vibration of the fibre to diminish to a definite fraction of its original value. The following empirical relations were deduced between p the vapour pressure measured in mm. of mercury and T the absolute temperature for caesium log p = 7.256 - 3753/T; for rubidium log p = 7-331 - 4209/T.The corresponding values of the respective chemical constants C are caesium 1-84 5 0.16,; rubidium 2-36 & 0.18,. The values of the absolute chemical constant Co calculated from the equation C = 1.5 log M + C where M is the molecular weight of the element are cmium - 1.54 5 0.16,; rubidium - 1-33 0.18,. Possible reasons why these values are greater than the theoretical value -1.608 are discussed. The respective values of Trouton's coefficient LTB/TB TB denoting the boiling point of the element are caesium 18.2; rubidium 19.9. J. S. G. T. Vapour Pressures of Nitric Oxide. F. -4. HEXGLEIN and H. KRuGER.-(See ii 170.) Coagulation and Surface Tension. J. AMAR (Compt. rend.1924 178 522-525).-Investigation is made of the effect of time on the coagulum obtained by the addition of citric acid to goat's milk and of hydrochloric acid to sodium silicate solution. In both cases dehydration proceeds according to an asymptotic curve. A study of the effects of various coagulants shows that the velocity of coagulation is proportional to the lowering of surface tension by the coagulant. 9. B. H. The Molecular Association of Liquids and Highly-com- pressed Gases. I and 11. E. E. WALKER (Phil. ,Mag. 1924 vi 47 lll-l26).-A theoretical paper in which it is shown that the critical coefficient RT,/P,V of non-associated substances is a function of the length of the molecule. A Dieterici equation of state is proposed which is modified by the introduction of a function of Z the length of the molecule.The constants cx and p in this equation are additive quantities both for associated and non- associated substances. The effect of association on the critical constants may be calculated with the aid of the modified equation. The Velocity of Sound in Gases and Vapours and the Ratio of the Specific Heats. H. B. DIXON and G . GREESM-OOD (Proc. Roy. Xoc. 1924 A 105 199-221).-The authors have determined J. S. G. T. W. E. G.GEYERAL AND PHYSICAL CHEMISTRY. ii. 145 directly the respective values of the velocity of sound in various gases and vapours contained in a lead pipe of 25 mm. internal diameter and have deduced therefrom the corresponding values of the velocity in the free gas or mpour. From the results the values Cp - C have been calculated by means of the equation Cp - C = R(l + 27tr3/16) in which the symbols have the customary significance and the values of y by means of Berthelof's equation Cp/C = P M [ 1 - 9 ~ ~ ( 1 - 6~~)/64]/RT. Combination of these data gives Cp and C,.The following are the respective values of 7 Cp - C Cp and C recorded a t different temperatures ethyl ether SO" 1.086 2.202 27.81 25.61 ; pentane 86" 1-086 2.202 29.50 27.31 ; hexane 80° 1.080 2.323 31.36,29.03 ; ethyl alcohol go" 1.129 2.138 18-71" 16.57 ; methyl alcohol 77" 1.203 2-120 12-56 10.44; benzene go" 1.097 2.245 25-39 23.15. Values of y C"p - C and C for ammonia and ethylene a t the temperatures stated are as follows ammonia 16.5" 1.3144 2.062 6.56; 41" 1.3095 2.045 (3.61; 63.5" 1.3062 2.034 6.64; 86" 1.3021 2.025 6-70; ethylene 12" 1.2562 2.0494 8-20; 4 1 * 5 O 1.2380 2.0327 8.45 ; 63*5" 1.2295 2.0246 8-82 ; S9" 1.2213 2.0162 9-12. For ammonia the values of C at temperature to agree very closely with those calculated by means of the formula C = 6.525 + 0.002t.The corresponding relation for ethylene is C = 8.06 + 0-0119t. The Ratio of the Specific Heats of Nitrogen and of Oxygen. J. R. PARTIKGTON and A. B. HOWE (Proc. Roy. Sm. 1924 A 105 225-243).-1n continuation of previous work (A. 1914 ji 529 ; 1921 ii 621; 1922 ii 191) measurements of the specific heat ratio y for nitrogen and oxygen have been made by the method of adiabatic expansion. The results corrected for radiation and for deviation from the ideal gaseous state are nitrogen 1.4045 & 0.0003; oxygen 1.3936 & O.OOc)2 a t 20" and 760 mm.The value for nitrogen is in close agreement Ivith that found by DIxon Relation between Vapour Pressure and Vapour Com- position in Binary Mixtures of Volatile Liquids. W. K. LEWIS and E. V. AIORPHEE ( J . Amer. Chem. Xoe. 1924 46 1-7).-4 theoretical paper in which an approximate integration is presented for the Duhem-Margules equation. With the aid of this it is possible to calculate the vapour-liquid composition curve from the isothermal total pressure curve. The method outlined has been tested in the case of the binary mixtures methyl alcohol-water ethyl also hol-w a t er carbon disulphide-wa tier and ace t one-chlor o - form. For only one of these mixtures is the divergence between the experimentally determined composition and the calculated value as great as 2%.T. W. JONES (Chemistry and Industry 1924 43 4-6).-A theoretical paper in which after outlining J. J. Thomson's theory of solution and ionisation the author shows in what respects this is in agreement with the views of H. E. Armstrong and Lowry. The poinfs of J. S. G. T. Campbell and Parker (A. 1921 ii 621). J. 8. G. T. J. F. S. Attempt to Correlate certain Theories of Solution.ii. 146 ABSTRACTS OF OHEMICAL PAPERS. divergence are also indicated and it is shown that the three theories are agreed on the point that the active constituents of solutions do not exist unattached but form complexes with the molecules of the solvent. Two Attempts to Replace the Ionic Theory by a Theory based on the Molecular Condition of Water.V. AUGER (Compt. rend. 1924 178 330-332).-The author points out certain obvious difficulties in connexion with the anti-ionic theories advanced respectively by Armstrong (cf. A. 1923 ii 542) and by Kling and Lassieur (ibid. 543). He regards Armstrong's two main premises as hypothetical and arbitrary and directs attention to the exiguity of detail in connexion with the other authors' formuk for the two modifications of water which are postulated. J. F. S. E. E. T. F. SAuERWALD.-(See ii 189.) Mixed Crystals of Alloys. The Law of Definite Proportions in the Light of Modern Research. U. R. E v a ~ s (Trans. Faraday Xoc. 1923 19 420- 429) .-The author reviews briefly metallographic evidence in reference to the law of definite proportions paying special attention to the hardness and stabili5y of alloys and re-states the present metallographic position concerning the variable composition of alloys from a point of view somewhat different from that adopted by Kurnakov (2.anorg. Chenz. 1914 83 109). It is claimed that Kurnakov's contention which is tantamount to a revival of Berthollet's doctrine of the variable composition of homogeneous bodies may be extended to compounds between metals and non- metals. A " compound " is in many cases merely a particularly stable member of a continuous series of solid solutions; the fact that maximum stability within the series occurs a t a composition expressible by a simple formula may often be explained from the physical point of view. Mixed Crystals.G. T~MMANN and W. KRINGS.-(SW ii 175.) The Structure of Solid Solutions. A. WESTGREN and G. PHRAGMBN (Nature 1924,113 122-124).-By the use of a precision camera the main features of which are indicated it is found that the face-centred elementary cube of steel quenched a t 1100" has an edge of 3.629 hi. whilst that of steel quenched a t 750" has an edge of 3.601 8. The lattice thus appears to increase in size with increasing content of carbon although the carbon atoms are smaller than the iron atoms ; the solid solution is hence not formed by a simple substitution. In the case of a homogeneous austenitic manganese steel it is evident that the lattice is formed only by iron and manganese atoms the carbon atoms being distributed uniformly between them. If less than 8% of aluminium is dissolved in copper the copper lattice is unchanged in type but it increases in size with increasing content of aluminium.These solid solutions would appear to be formed by simple substitution of copper atoms by aluminium atoms. In the range 16-25y0 Al the lattice dimensions decrease with increasing content of aluminium. The edge of the elementary cube varies from 8.701 8. to 8.656 8. and it J. S. G. T.GENERAL AXD PHYSICAL CHICMISTBY. fi. 147 appears that three copper atoms in the lattice are replaced by two aluminium atoms. The groups of atoms present in the elementary cube correspond very nearly with the formulae Cu&,6 C U ~ ~ A ~ ~ The powder photogram of a zinc-copper alloy containing 65% Zn bears an unexpectedly close resemblance to those of aluminium- copper alloys containing 16-25% Al; these would appear to be formed by simple substitution of one metal for the other in the lattice framework.A. A. E. Double Decomposition in Absence of Solvents. A. G. BERGMANN.-(See ii 178.) Displacement of Acids by Diffusion. E. DEMOUSSY (Compt. rend. 1924 178 208-211).-The displacement of a strong acid (hydrochloric) by a weaker one (formic) in contact with a membrane (cf. Li Shou Houa Ann. Physique 17 475) is shown to be due to the greater speed of diffusion of the strong acid. The phenomenon is observed in the absence of a membrane. If a solution containing barium chloride (1 mol.) and formic acid (2 mols.) is introduced into the lower part of a tall vessel containing water hydrochloric acid diffuses upwards more rapidly than formic acid and the original state of equilibrium is continuously displaced.Displacement of Acids by Diffusion. P. GIRARD (Compt. rend. 1924 178 389-391).-The simple explanation offered by Demoussy (cf. supra) of results obtained by Li Shou Houa and by Girard (cf. A. 1914 ii 718) is shown to be insufficient. The inter- position of a membrane between a salt solution and water inverts but does not create the electrostatic field (at the junction of t3he two media) and it is the latter which is responsible for the effects observed by all three authors. E. E. T. Method of Determining Micromiscibilities. N. PERRAKIS and A. MASSOL (Compt. rend. 1924 178 322-323) .-Micromisci- bilities are readily determined by a titration method the end-point being determined by the appearance of cloudiness.Two mixtures of water alcohol and an oil were examined the water or the oil being present in traces only (cf. A. 1923 ii 836). E. E. T. G. EDGAB and W. H. SCHUYLER ( J . Amer. Chenz. Xoc. 1924,46 66-75).-The equilibrium in the gaseous state between ethyl alcohol acetic acid ethyl acetate and water has been examined. The method of pro- cedure adopted consisted in determining the composition of the vapour which distilled a t measured temperature and pressure from the liquid equilibrium mixture. The temperature varied from 52.2' to 77.6" and the pressure from about 0.5 to 1.0 atmosphere. The manipulation at atmospheric pressure is simpler than a t reduced pressure and in consequence the results of the individual experiments (12) at this pressure are in better agreement.The results show that the equilibrium in the gaseous phase is in general different from that in the liquid phase and in the present case it is widely dif€erent. From the result's it is calculated that a t 150" cu&so and cu,,AI,,. E. E. T. Esterification Equilibria in the Gaseous Phase.ii. 148 ABS!CRACTB OF CHEMICAL PAPERS. the equilibrium corresponds with about 95% esterification. The results obtained by Reid and Mulliken were in all cases less than this and the suggestion that the experiments in question afford evidence of the displacement of the equilibrium by the contact catalyst can therefore not be accepted. Dissociation Constant of Boric Acid. E. B. R. PRIDEAUX and A. T. ~T7a~~.-(See ii 171.) Faraday’s Law for Lithium Hydride.K. PETms.-(See ii 174.) Electrolytic Conductivity of Molten Scandium Chloride. -(See ii 184.) Determination of the Degree of Dissociation of a Binary Electrolyte from Conductivity [Measurements]. E. CHER- BTTLIEZ ( J . Chim. ph!psique 1922 20 464-470).-A theoretical paper in which the author discusses the probable errors in the value of the degree of ionisation calculated from the equation cc = A/Am. These errors are much reduced with increasing ionisation and with increasing ratio between the two dilutions. More accurate values of x are obtained by determining the variation of the conductivity with the dilution and equations are given showing the relationship between these. J. 37. S. Influence of Gelatin on the Transport Numbers of Hydro- chloric Acid.W. G. PRAKCE and W. H. MORAN ( J . Amer. Chem. SOC. 1924,46 19-30) .-The transport numbers of 0.1N- and 0.01N- solutions of hydrochloric acid have been determined a t 25” by an E.1II.F. method in the presence of gelatin. It is shown that the transport number of the anion varies from 0.1699 in the absence of gelatin to 0-77718 in the presence of 20% of gelatin. The addition of gelatin reduces the effective hydrogen-ion concentration of 0.1127- and O-OlX-hydrochloric acid solutions and alters the con- ductivity. The relative change in the conductivity of 0-01N- solutions is greater than for O.lX-solutions. In the case of a 10% gelatin solution the conductivity of a 0-lN-solution is reduced almost to zero; a similar change cccurs in 0.OlN-hydrochloric acid with 2-3% of gelatin.These facts suggest that not only is the hydrogen-ion concentration reduced but that the acid is removed as a whole by the gelatin. Jn agreement with this view evidence has been obtained that gelatin reacts with acids to form a dissociable adsorption complex or additive chemical compound. This is in agreement with the hypothesis put forward by Loeb. J. F. S. Hydrofluoric Acid Solutions. C. 137. DAVIES and L. J. Influence of Neutral Salts on the Potential of Metal Electrodes. G. POMA (2. physikal. Chem. 1923 107 329-332). -A claim for priority o ~ e r &kadiev (A. 1923 ii 369) Sachanov (Diss. Odessa 1915) Harned (-4. 1916 ii S) and Akerlof (A. 1922 ii 134) in connexion with experimental work on the influence of neutral sa1t)s on the potential of metal electrodes and of hypotheses put forward to explain the effects obserl-ed (‘4.1912 ii 323; J. P. S. HcDLEsTos.-(See ii 167.)GEXERAL AND PHYSICAL CHEMISTRY. ii. 149 1914 ii 410). The matter contained in the previous papers is briefly discussed J. F. S. Effect of a Magnetic Field on the Potential of Hydrogen Occluded in Iron. T. W. RICHARDS and W. T. RICHARDS ( J . Aher. Chem. Xoc. 1924 46 89-104).-The potential of pure iron in ferrous sulphate in mixtures of ferrous and ferric sulphate and in mixtures of ferrous sulphate and sulphuric acid of various concentrations bas been measured against it normal calomel electrode a t 25". The effect of a magnetic field on the potential of pure iron and of spongy iron charged with hydrogen in ferrous sulphate solutions has also been examined.It is found that the patential of iron in ferrous sialphate is changed only very slightly (0-1.0 millivolt) by the application of a magnetic field. This small alteration is probably to be attributed in part to diffusion in the electrolyte under magnetic influence. On the other hand the potentid of hydrogen occluded in iron is found to be lowered dis- tinctly (0.5- 5.1 millivolts) the change being roughly proportional to the strength of the magnetic field employed. With a given magnetic field the decrease of " overvoltage " is approximately proportional to the " overvoltage " itseIf amounting with the strongest magnet used to about 5% of the " overvoltage " value. The results appear to indicate that occluded hydrogen is in the atomic condition but not ionised and that a magnetic field affects its potential not because of any change in affinity between hydrogen and iron but rather because of some change in the texture of the occluding metal due to magnetostriction.Effect of Various Preparations of Mercurous Sulphate on the Electromotive Force and Hysteresis of Weston Standard Cells. W. C. VOSBURGH and M. EPPLEY ( J . Amer. Chem. Soc. 1924 46 104-lll).-Weston cells have been prepared with mercurous sulphate of variously sized particles (3p - 360p) and the E.M.F.'s measured a t 25" after varying periods of time. It is shown that cells with finely divided mercurous sulphate have E.M.P. values which are from 40 to 100 microvolts higher than those of cells made with coarser mercurous sulphate.The hysteresis of the cells varied greatly with the characteristics of the mercurous sulphate being largest in the case of a coarse-grained white mercur- ous sulphate and practically zero in the case of a fine-grained grey mercurous sulphate. Both the smallness of the particles and the presence of finely divided mercury appear to influence the prevention of hysteresis. The addition of acid to cells lowers the hysteresis considerably. In setting up a standard cell it is important there- fore that the mercurous sulphate should be sufficiently large-grained. Digestion with sulphuric acid solution at the boiling point is a wise precaution but this cannot be depended on to correct a very fhely divided uniform mercurous sulphate. Low hysteresis cannot be taken as indicating trustworthiness in cells of low acid content.On the contrary a cell with a fairly high hysteresis is more trustworthy for use as a primary standard. When it is more important that a cell should have low hysteresis rather than that the E.N.P. should J. F. S.ii. 150 ABSTBACTS OF CHEMICAL PAPERS. have the established value a higher acidity and a fine-grained mercurous sulphate may be used. J. GRANT and J. R. PARTINGTON (Trans. Faraduy Xoc. 1923 19 414417) .-Contrary to the conclusion of Wilson (A. 1906 ii 144) the authors find that the values obtained for the E.M.F.'s of concenration cells in which solutions of silver nitrate in methyl alcohol with a concentration range from 3/10 to N/640 were contained agree with the corresponding values calculated by means of Nernst's formula.Wilson's discordant results are attributed to experi- mental error. N. EOUMAN (Rec. trav. chim. 1924 43 1-29).-The determination of the potential of chromium in acid solution is attended with great difficulty on account of the tendency of the metal to become passive. Many previous workers have experimented with metal prepared by Goldschmidt's method which is now shown not to give reproducible results on account of the presence of traces of impurity. Constant results have however been obtained with electrolytic chromium obtained by Sargent's method (Trans. dmer. Electro- chem. Soc. 1920 37) activated in 24% hydrochloric acid a t 100". The potential measured in 1-ON-sulphuric acid is - 0-546 volt. This potential is independent of the metal on which the chromium is deposited and also of the method of activation.It is also independent of the ratio of Cr"- to Cr"'-ions in solution. It is curious that with increasing hydrogen-ion concentration the potential becomes more positive in sulphuric acid but more negative in hydrochloric acid solution. Chromium remains active only when the acidity is above a certain limit about 0.001N. I n the passive state the potential of chromium in potassium chloride solution depends on the previous treatment which it has received and this is shown to be equally true of other metals including platinum. Consequently no conclusions can be drawn from such measurements relative to the state of the metal The potential of passive chromium varies with the metal on which it is deposited.When chromium is polarised anodically the potential varies in the same way with the acidity as the potential of the unattackable electrode. The polarisation tension is therefore governed by the reaction Cr"' + 4H,O = CrO," + 8H' + 3 0 . The phenomenon of " after-anodic polarisation " was investigated and was found to be less general than Aten supposes (A. 1919 ii 8). The behaviour of chromium is explicable by the theory of allotropy. The Electromotive Behaviour of Magnesium and Magnesium Amalgams. R. MULLER and W. KNAUS (2. anory. Chem. 1923 130 173-180).-The potentials of the metal and its amalgams have been measured in solutions of magnesium chloride potassium chloride sodium hydroxide and hydrochloric acid in dilutions from 1.ON to 0*0001N using the calomel normal electrode The values for the amalgams are in all cases considerably higher than for the metai.The values for the latter do not alter when it J. F. S. Concentration Cells in Methyl Alcohol. J. S. G. T. The Electrochemical Behaviour of Chromium. E. H. R.BENERBL AND PEYSICAL CHEMISTRY. ii. 151 is filed under the solution except in the case of the sodium hydroxide solutions in which a skin of hydroxide forms very rapidly. The values in acid solution are of no significance since both metal and amalgam react vigorously. The potential of the metal in N - magnesium chloride solution calculated with reference to the hydrogen electrode is 1.49 v. and of the liquid amalgam (10-20 atomyo magnesium) 1.8 v. but the latter value is uncertain as the amalgams react vigorously with the solutions.Determinations were therefore made with saturated solutions of magnesium bromide in anhydrous pyridine measuring against the half-element Agl AgNO 0.1N in pyridine; the potential for pure magnesium is 1.33 v. which value remains constant for amalgams up to 50 atoms% of mercury; at this point a sudden rise to 1.72 v. occurs. The amalgams rich in magnesium are solid very brittle and con- siderably harder than pure magnesium. Above 50 atoms% of mkrcury the potential remains a t 1.72 v. until only 10 atoms% of magnesium is present. The potential falls slightly for less than 10 atoms% of magnesium then suddenly for 1 atomoh the potential for pure mercury being 0.427 v. The alloys very poor in magnesium are liquid.The potential curve is exceptional the sudden change in- dicating the compound HgMg which is formed with liberation of heat. Magnesium may be separated electrolytically from the pyridine solution of the bromide on a platinum or mercury cathode the potential for platinum being 1.35 and for mercury 1.65 v. Amphoteric Elements. J. KASARNOWSKY (2. anorg. Chem. 1923,130 l40).-A note explaining that the difference in the values of the potential o ~ h given in an earlier paper (A. 1923 ii 824 828) is due to alternative methods of calculation. Intermittent Current Electrolysis. 111. The Measurement of Overvoltage. S. GLASSTONE ( J . Chem. SOC. 1924 125 250- 260).-The true potential of the electrode at the instant of switching off the polarising current is determined by measuring the rate of fall of the potential of a polarised electrode after the current is switched off and extrapolating the P.D.-time curve.The overvoltage a t the electrode is determined from these potentials obtained a t various current densities. The results show that for small values of polarising current the overvoltages measured by the direct method and by the extrapolation method are almost the same hence the existence of transfer resistance is very doubtful. For very low current densities the direct method is the only method available since the repeated make and break of the polarising current prevents the electrode from reaching its equilibrium value. At greater current densities the results obtained by the direct method are appreciably higher than the extrapolated values. This difference is probably due to the resistance of a metal-gas-electrolyte system at the surface of the electrode.Variation in the size of the electrode and of the dilution and temperature of the electrolyte give results which strongly support this view. The “surface resistance” at a lead electrode 1 sq. cm. area in a N/1-solution of sodium hydroxide or sulphuric acid is not greater than 4 ohms. S. I. L. s. I. L. J. B. F.ii. 152 ABSTRACTS OF CHEMICAL PAPERS. The Dependence of Polarisation-overvoltage on Hydroxyl and Hydrogen-ion Concentration. I. Polarisation-over- voltage of an Antimony Cathode in Aqueous Alkaline Solution. H. J. S. SAND and E. J. WEEKS ( J . Chern. Xoc. 1924 125 160-168).-The polarisation overvoltage o of an antimony cathode in alkaline solution is detlermined by the commutator met hod and obeys with considerable accuracy the empirical eqllation o =a-2h in which h (the so-called pE voltage) is the potential difference between a standard calomel electrode and a hydrogen electrode in the solution examined and a is a constant depending only on the temperature.The relation holds for most varied electrolytes and hence such factors as sodium-ion concen- tration are without measurable influence. I n explanation of these results it is suggested that ions of the formula H(OH)," are formed in solution m-itlh a very high velocity. When these ions which have a definite solubility and also a definite supersolubility limit have accumulated in a sufficient number (in the liquid or solid state) they undergo decomposition and furnish thereby the energy required for the formation of a gas bubble. It is suggested further that polarisation overvoltage is determined by the concentration of hydrogen dissolved in or adsorbed by the surface of the electrode that the hydrogen dissolved in the working cathode is in thermo- dynamic equilibrium with hydrogen dissolved in the layer of electrolyte immediately outside the double layer separating the metal from solution and that hydrogen is de-ionised simultaneously on the electrode throughout the double layer and immediat'ely bepiid it i.e.free electrons travel directly from the cathode into the solution over this range during the passage of the current. The conclusion is d r a m that identical relations should be found with other metals which act as ir,different cathodes in alkaline s o h t icn.J. B. F. Free Energy of Antimony Trioxide and the Reduction Potential of Antimony. R. SCHUHMAKX ( J . Amer. Chem. Xoc. 1924 46 52-58).-The E.N.F. of the cells of the type Sb,Sb,O,(s)I HClO,(c)lH has been measured at 25' for concentrations of per- chloric acid lying between O * l J l and 1.OM. The solubility of antimony trioxide in perchloric acid solutions of this range of concentrations was also measured a t 25". The value of the E.II1.F. is found to be -0.152 volt for the cell. From this value the free energy of formation of antimony trioxide from its elements.at 25" is found to be - 148,600 cal. with an error which probably does not exceed +300 cal. The results of the solubility determinations indicate that the dissolved antimony exists mainly in the form of SbO' a t the concentrations of perchloric acid employed.From the solubility data and the E.M.F. of the cell the molecular reduction potential corresponding with the electrode reaction Sb(.s) +H,O+ 3@=SbO'+2H' is found to be -0.212. J. F. S. Activity of Perchloric Acid in Aqueous Solution. R. SCHUHMANN ( J . Amer. Chern. Xoc. 1924 46 58-60).-The E.M.F.GENERAL AND PHYSICAL CHEMISTBY. ii. 153 of cells of the type H2(HCIIHC1041H has been measured a t 25' for concentrations of the two acids between 0.1N and 1-ON. In all combinations the concentrations of the two acids were very nearly the same. The liquid junction potential resulting from the slight difference between the mobilities of the chloride and the perchlorate ions was very small and was estimated as 0-00047 v.The corrected E.M.F. values obtained are so small as to be negligible in determin- ing the relative activities of the two acids at the same concentration. Thus the largest value 0-00017 v. would correspond with a difference of only 0.7% in the value of the activity coefficient of the hydrogen- ion. The conclusion may be drawn therefore that the activity coefficient of perchloric acid is substantially the same as that of hydrochloric acid for concentrations up to l . O N and the values de- rived for the latter acid (see Lewis and Randall Thermodynamics ") may therefore be employed in mass action and thermodynamic expressions relating to perchloric acid. This conclusion is of interest since it indicates that the behaviour of these acids includ- ing the pronounced minimum in their activity coefficients in the neighbourhood of 0 4 N is primarily determined by the hydrogen- ion the effects of the two anions being apparently relatively small and nearly equal. Free Energy of Aqueous Sulphuric Acid.D. F. SMITH and J. E. M-AYER (J. Amer. Chem. Soc. 1924 46 7&83).-The equili- brium conditions of the reversible reaction H2S0,(5~4-6.8N) + 6HI(O.l-l.ON) =312+4H,0+S(rhom) have been determined a t 80". The E.M.P.'s of cells of the type Pt,H,1H,S04(C,)+ HC1( C,) I H,S04( Cl) +HI( C,) +I,( C,) I Pt were also measured a t the same temperature. By combining the data obtained the partial free energy of sulphuric acid (5.441 molar) was calculated to be -169,850 cal.at 80". Making use of already known heat data the corresponding value for 25" was calculated as- 175,010 cal. By means of the activity data of Lewis and Randall (" Thermodynamics ") the free energy of the sulphion SO:' at 1Jl concentration WM found to be -176,540 cal. at 25" thus confirming the provisional value of -176,500 cal. derived by Lewis and Randall. By combining this new value for the free energy of the sulphion with the values given by Lewis and Randall for other sulphur compounds the reduction potentials of various combinations which have considerable practical significance have been worked out for 25" and tabulated. J. F. S. The Variation of Surface Tension with Temperature and some Related Functions. S. SUGDEN ( J .Chern. Soc. 1924 125 3 2 4 1 ) .-Equations put forward by various observers showing the variation of surface tension with temperature are discussed. If B is given the value 1.20 then from van der Waals' equation y=kl@;~(I -m)B=k26,*pc~(l -m)B and Katayama's modification of the Ramsay and Shields equation,. y[M/D-d)]~=At9,(1 -m) it is possible to derive Macleod's relation between surface tension and density. The results of Ramsay and Shields (A. 1894 ii 179) are too low owing to inadequate correction for capillary rise and are corrected as also are the results of Walden and J. F. S.ii. 154 ABSTRACTS OF CHEMICAL PAPERS. Swinnc (A 1912 ii 628). From the data thus obtained for six non-associated liquids it is shown that (a) the variation in surface tension with temperature is represented accurately by the equation y=yo(1-8)1'2 where 8 is the reduced temperature and yo a con- stant (b) Macleod's relation Y = C ( D - - ~ ) ~ where C is a constant and D and d are the densities of liquid and vapour respectively holds accurately for non-associated liquids up to 40" below their critical temperature.In the van der Waals equation the universal constants k and k2 vary several units yo among the different liquids whilst the relations connecting yo and the critical constants are shown to hold only approximately. The method of correcting the data of Ramsay and Shields is described. The Determination of Surface Tension from the Maximum Pressure in Bubbles. 11. S . XUGDEN (J. Chem. SOC. 1924 125 27-31).-An improved form of apparatus is described which consists essentially of three parts the vessel holding the liquid under examination (called the bubbler) into which pass two tubes one a capillary and a tube of about 4 mrn.diameter a pressure gauge and a small mercury aspirator. To avoid direct measurementl of the radius of the capillary tube the surface tension of pure benzene which is known with sufficient accuracy may be taken as a reference liquid for the purpose of calibration. A table giving the surface tension of benzene €or every degree between 10" and 30" is given. The equation y=AP(I +0.69r2gD/P) may be used for the sizes of tubes generally employed. A is a constant for the particular apparatus P the difference between the pre'ssures required t o liberate bubbles from the two tubes D the density of the liquid and r2 the radius of the nide tube.Surface Tension at the Surface of Separation of Water and an Organic Liquid in Presence of Aliphatic Acids and of Alkalis. R. DUBRISAY and P. PICARD (Compt. rend. 1924 178 205-208; cf. A. 1923 ii 741).-Solutions of aliphatic acids in benzene have been examined by the drop method (Donnan A. 1900 ii 201) the volume of 10 drops (of benzene solution) formed in (a) iV/1250-sodium hydroxide ( b ) N/25OO-sodium hydroxide and ( c ) N/2500-sodium hydroxide containing 1 yo of sodium chloride being measured. This volume decreases in the order (a) lauric myristic oleic palmitic erucic ricinoleic stearic acid ; ( b ) lauric myristic ricinoleic oleic erucic palmitic stearic acid ; ( c ) lauric ricinoleic myristic oleic erucic palmitic stearic acid.Similar experiments have been carried out with different concentrations of stearic oleic and ricinoleic acids. With increasing concentration the volume of the 10 drops a t first diminishes and then increases. Permeability of Precipitated Copper Ferrocyanide Mem- branes for Non-electrolytes. R. COLLANDER (KoZLChem. Beihefte 1924 19 72-104) .-The permeability of precipitated copper ferrocyanide membranes for non-electrolytes has been investigated. It is shown that these membranes behave towards non-electrolytes in the same way as an ultra-filter. Compounds of which the molecules do not exceed a certain size pass more or less J. B. 3'. J. B. F. E. E. T.GENEUL AND PHYSICAL CHEMISTRY. ii. 155 easily through such membranes whilst larger molecules are held back. I n the present experiments the upper limit of size at which the molecules are able to pass through the membrane lies a t a mole- cular volume of 80-100.Adsorption and solution processes in the membrane if they are present a t all are of small importance in comparison with this dominating sieve action. Electrification by Adsorption of Membranes and Colloids. F. CHOUCROUW ( J . Chim. physique l923,20,411436).-The author gives a ~dsurne' of the present state of knowledge on electrification by adsorption and he points out that in all cases of electro-osmosis an opposing electromotive force of filtration is set up. The slectro- endosmosis of distilled water dilute acids and alkalis across mem- branes of ferric oxide arsenic sulphide ferric ferrocyanide and copper ferrocyanide has been investigat.ed and it is shown that the phenomena observed are the result of two different actions the one generally observed depending on the action of acids and alkalis and the other a specific action due to superficial ionisation of the walI.Attention is directed to a group of liquids with properties which suggest that they are neither suspensions nor true colloidal solutions; these are the opalescent liquids in which the particles are in static equilibrium with a liquid in which they are soluble. It is shown that the micellz when travelling in an electric field draw with them the intermicellar liquid in certain cases. Method of Determining the Detergent Action of Soaps. J. W. MCBAIN R. S. HARBORNE and A.M. KING ( J . Xoc. Chem. I d . 1923 42 373-378~).-A method has been developed and standardised for the direct and rapid determination of the amount of finely divided carbon which various soaps carry through iilter- paper. This gives a " carbon number " characteristic of each soap solution which may be taken as a measure of the detergent action of the particular material. The carbon number is defined as the number of grams of carbon carried through by 1 kg. of solution under standard conditions. The method of determination consists of the following One g. of " Auk " carbon black is weighed to within 1% placed in a large test-tube containing 20 C.C. of soap solution thoroughly shaken and placed in a thermostat for twenty-three hours shaken again and kept for a further hour in the thermostat and then poured on an 11.3 cm.No. 31 Whatman flter-paper. The filtration is carried out in the thermostat and when about 10 C.C. of filtrate have been collected this is analysed and the amount of carbon determined. It is shown that by a slight variation of the conditions the detergent action of a soap may be enormously in- creased. There is an optimum concentration in moderate dilution for which the effect is a maximum; very slight addition of either acid or preferably alkali greatly enhances the detergent power. With rise of temperature the detergent power diminishes rapidly a t first and then more slowly. There is surprisingly little difference between soaps as different as oleate and myristate. Colloidal Vanadium Pentoxide .-A.V. Dummsm.-(See ii 195.) J. F. S. J. F. S. J. F. S.ii. 156 ABSTRACTS OF CiXEMICAL PAPERS. Colloidal Nature of Clay. A. FODOR and B. SCHOENFELD (Ko2Z.-Chem. Beihefte 1924,19,1-46) .-The behaviour of electrolytes towards suspensions of clay has been investigated with particular reference to the rate of sedimentation coagulation peptisation cataphoresis adsorption and swelling. It is shown that the behaviour of different alkalis is fundamentally different. In small concentrations ammonia increases the stability of clay suspensions although in larger quantities it causes an ultramicroscopic aggreg- ation of the particles without effecting a macroscopic coagulation. Sodium hydroxide however in small quantities effects the coagu- lation of the coarser particles and in large quantities causes quanti- tative precipitation.The critical alkali concentrations are not only dependent on the hydrogen-ion concentration but also on the quantity of clay contained in unit volume. Alkali hydroxide is adsorbed by clay and the adsorption isotherm is found to be of normal type. Relatively small quantities of calcium hydroxide bring about considerable coagulation ; with increasing concentration of calcium hydroxide the volume of the coagulate increases. This the authors attribute to peptisation and subsequent adsorption. In support of this is the fact that in general lime behaves as a dehydrating agent and applied to the soil has a disintegrating action. The form of the adsorption isotherm of calcium hydroxide by clay is in keeping with this view and further support is afforded by the rapidity with which equilibrium is reached.Clay which has adsorbed calcium hydroxide never regains its previous power of adsorbing this sub- stance even after the most thorough washing. The precipitation of clays by alkalis differs noticeably from that produced by acids. This difference is seen in the relatively greater density of the coagula produced on the addition of acids. Small amounts of acid increase the stability of the suspension but large amounts bring about immediate coa,gulation. Salts in small quantities increase the stability of suspensions; in larger quantities coagulation is accelerated. Humus-acids and phosphates exercise a protective action on clay suspensions. The peptising action of phosphate is independent of the hydrogen-ion concentration and in keeping with its adsorption by clay ; phosphates tend to stabilise the suspensions.The above results are all obtained from experiments of a qualitative character. Composition of Micella?. 11. Colloidal Iron Oxide. R. WINTGREN and M. BILTZ (2. physikal. Chern. 1923,107,403-422 ; cf. A. 1922 ii 78).-With the object of controlling the method previously described (Zoc. cit.) for determining the composition and the equivalent aggregation that is the number of molecules in a single colloidal particle per unit of electric charge the authors have measured the electrical conductivity of a series of ferric oxide hydrosols and that of the liquid obtained from them by ultra-filtration. I n this way a value is obtained for the conductivity of the micell=.In addition the migration of the sols in an electrical field has been studied. The results show that the composition of the micellae is not changed by ultra-filtration. The amount of chlorine present in the sols which mere prepared from a commercial dialysed J. F. S.GENERAL AND PHYSICAL CHEMISTRY. ii. 157 product was determined gravimetrically and the chlorine-ion electro- metrically. Using the method of calculation previously described it is shown that in a sol containing chlorine and iron in the ratio 0.0605 and containing 1-601 g. of iron in 100 c.c. a single micelle contains 7.63 x lo5 molecules of ferric oxide whilst in a sol with the ratio C1 Fe = 0,1236 and 2.869 g. of iron in 100 C.C. the micelle contains 3.31 x lo5 molecules of ferric oxide.On dividing the number of molecules in the ionic micelle by the equivalent aggre- gation the valency is found to be 10,230 and 6494 in the two cases mentioned. Assuming that the micelle is Fe,O of density 5.1 the radius of the particle is 21.1 and 16*Opp respectively. This assump- tion is probably unjustifiable in that the colloidal particles may have a spongy structure in which there is a considerable amount of water. J. F. S. F. G. TRYHORN and S. C. BLACKTIN (Trans. Faraday Soc. 1923 19 433-441).-The authors find that the slow diffusive reaction in gelatin gels between mercuric chloride and potassium iodide and between potassium dichromate and silver nitrate results in the production of anomalous Liesegang bands in which the normal bands are accompanied by secondary bands similar to those described by Hatschek for the reaction between lead ace6ate and potassium chromate or dichromate (Proc.Roy. Xoc. 1921 A 96 496). The substances formed in these three reactions crystallise from the respective gels more readily in the light than in the dark. This observation affords an adequate explanation of the formation of the anomalous bands and a tentative suggestion according to which the possibility of the formation of Liesegang bands depends on the stability of the ion-gelatin complex taking part in the reaction involved is advanced to explain the mechanism of the process. The Formation of Anomalous Liesegang Bands. J. S. G. T. Rhythmic Formation of Precipitates. Liesegang’s Rings. C. K. JABLCZYNSEI (Bull.Xoc. chim. 1923 33 [iv] 1592-1602). -Measurements by Morse and Pierce (A. 1904 ii 14) of the rate of formation of Liesegang’s rings have been interpreted by these authors in support of Ostwald’s theory. This method of inter- preting the results is criticised on the ground that these can be explained by reference to the rate of diffusion of one of the reacting substances. The author’s treatment of the problem leads to the relation (r$ - rI3)/(t2 - tl) = 3kDC,/~p where rl and rz are the radii of two rings in order of formation t and t2 the corresponding times of formation D is the coefficient of diffusion of silver nitrate in the gelatin and C its concentration p being the thickness of the gelatin layer. If the last two quantities are constant the expression on the right assumes a constant value. New measurements have given results which are in agreement with the calculated values.The mechanism of the ring-formation is interpreted on the assumption that the growth of the larger particles of silver chromate is a t the expense of the smaller. This view is preferred to Ostwald’s theory of metastable conditions on the ground that the changes observedii. 158 ABS!I?RAOTS OF CHEMICAL PAPERS. are gradual at all stages. The analogy between ring-formation in precipitates and the annular structure of tree-trunks suggested by Kuster is shown to be superficial only but Liesegang's application t o ring-formation in agates is valid. Formation of Periodic Precipitates. 11. N. R. DHAR and A. C. CHATTERJI (J. Physical Chem.1924 28 41-50; cf. A. 1922 ii 627) .-A continuation of the investigation of periodic precipitates. It is shown that gelatin previously mixed with dichromate has a better peptising effect on silver chromate than that which is pre- viously mixed with silver nitrate. In the formation of lead chloride rings in silicic acid jelly by t)he action of lead nitrate on sodium chloride the rings are more closely packed as the concentration of the jelly is decreased and other conditions being constant the rings are less closely packed the lower the concentration of the lead nitrate. Similar results are obtained for silver chloride rings in silicic acid. Rings of silver chloride and iodide are affected by diffused light whilst those of lead chloride and iodide are not). Tubes containing mercuric iodide rings show more rings when exposed to diffused daylight than in the dark.The theories explaining the formation of Liesegang rings put forward by Ostwald and Bradford are found to be unsatisfactory. It has been frequently observed that sols can be more or less completely adsorbed and coagulated by freshly precipitated solids. Based on the foregoing observations a hypo- thesis of the formation of Liesegang rings is put forward in which it layer of coagulated material is assumed to be followed by a space free from the precipitated material. Mercuric iodide rings in gelatin and agar consist of a layer of coagulated and crystalline mercuric iodide which is red followed by a layer of colloidal mercuric iodide which is yellow. An explanation of the formation of rings consisting of a layer of coagulated and crystalline material followed by a layer of the peptised substance has been put forward in which the adsorp- tion of the second product of the chemical change is the dominant factor.It is pointed out that there is the possibility of the occurrence of two distinct classes of Liesegang rings. I n the one class a layer of precipitate is followed by a zone practically free from the sub- stance whilst in the other class the rings consist of a layer of coagulated sol which in course of time may crystallise followed by a layer of the peptised sol. The authors are of the opinion that under suitable conditions all sparingly soluble substances are capable of forming Liesegang rings. Calorimeter €or Heats of Mixing at Elevated Temperatures.B. H. CARROLL and J. H. MATHEWS (J. Amer. Chena. SOL 1924 46 30-36).-A calorimeter is described which is suitable for the measurement of the heat of mixing a t temperatures up to the boiling point of the mixture concerned. This calorimeter has been employed for measuring the heat of mixing of carbon tetrachloride with ethyl acetate benzene with ethyl alcohol acetone with ether and chloro- form respectively and ethyl alcohol with water. The curves obtained by plotting the heat of mixing per. g.-mol. against the molar composi- tion deviate in all cases appreciably from a straight line. The curve H. J. E. J. 3'. S.GENERA5 AND PHYSICAL CHEMISTBY. ii. 169 for benzene-ethyl alcohol shows a strong minimum at -360 cal. at 0.351 mol. benzene.The curve for chloroform-acetone shows a well-deked maximum at 400 cal. for the mol. fraction 0-5. In this case the curve is taken to indicate the formation of an equimolecular compound. The absorption of heat on dilution of ethyl alcohol with benzene may be attributed to the decrease in the association of the alcohol. The heat changes observed with the other mixtures do not afford conclusive evidence of any clearly defined molecular change. J. F. S. Rapid Method for the Determination of Heats of Adsorption and some Values for Hydrogen on Nickel and Copper. R. A. BEEBF and H. R. TAYLOR ( J . Amer. Chem. Soc. 1924,46,43-52).- A rapid method for measuring heats of adsorption of gases on metallic catalysts is described. The catalyst mass is placed in the inner tube of a vacuum vessel and an insulated platinum-iridium wire dis- tributed evenly throughout the mass serves as a resistance ther- mometer.A wide tube carrying a Beckmann thermometer is placed in the middle of the catalyst mass and this with the aid of cement closes the inner tube of the vacuum vessel. The hydrogen is admitted by a suitable tap and the amount adsorbed measured by the method of Taylor and Burns using nitrogen as reference gas (A 1921 ii 630). The integral value for the heat of adsorption of hydrogen on nickel varies from 13,500 to 20,500 cal. depending on the past history of the adsorbent. Corresponding values for the same samples are in good agreement and although the authors do not claim an accuracy greater than about lo% yet the results generally are much closer than this.The magnitude of the heat of adsorption show8 that the phenomenon is not a simple condensation. The differential values of the heat of adsorption remain nearly constant with increasing partial pressure and this suggests that a specific adsorption effect is involved. With copper as adsorbent of hydrogen the results are more concordant. The magnitude of the heat of adsorption is in this case 9600 cal. Formation of Ozone at Low Temperature and Pressures. A. EUCKEN (2. physikal. Chem. 1923 107 436-452).-When the concentration of oxygen in discharge tubes is below 5 millimol. per litre a very much smaller yield of ozone is obtained a t high temperatures than at the temperature of liquid air. A similar variation is found when the gas is subjected to ultra-violet light of short wave-length.These observations indicate that the mean persistence of the primarily excited oxygen molecules decreases rapidly with increase of temperature. The decreased formation of ozone is accompanied by a change in the band spectrum. The Mechanism of Chemical Reaction. A. JOB (Bull. SOC. chim. 1923 33 [ivl 1561-1591).-A lecture delivered before the Soci6t6 Chimique de France 10th March 1923. The Causation of Organic Reactions by Alumina and Theories of Catalysis. H. ADEINS and B. H. NISSEN (J. Amer. Chern. Soc. 1924 46 130-145; cf. A. 1922 i? 422; ii 834).- Two series of experiments have been conducted using alumina J. 3'. S. J. F. S. H. J. E.ii. 160 ABSTRACTS OF CHEMICAL PAPE&s. prepared either from hydrated alumina aluminium hydroxide or various aluminium alkoxides.In one series of experiments the behaviour of esters alcohols etc. has been studied when passed over the various aluminas a t temperatures from 350" to 490". It is believed that the aluminas differ in the spacial relationships of the '' active points " of the alumha. I n another series of experiments the effect on the decomposition of an ester of introducing water alcohols heptane ethylene acetic acid acetone or other esters has been observed. The relative stabilities or reactivities of the members of even a homologous series cannot be determined by comparing the speeds of reaction or decomposition of the members of the series because even the relative speeds are a function of the surface at which the reaction occurs.The formation of ketones from acids of nitriles from amides and of olefines from alkyl halides is not affected by modifications in the spacial configuration of the alumina catalysts. The reactions of esters and alcohols are very sensitive to such mofications. KO experimental evidence has been obtained to support the idea that the function of the catalyst is to remove one member of an equilibrium or that the catalyst reacts with the substance catalysed to form compounds such as aluminium acetate or ethoxide. It is believed that in every case the substance catalysed unites chemically with the catalyst with a resultant rearrangement of the electrons of the substrate followed by vaporisation of the products. The nature of the rearrangement and subsequent reaction is determined primarily by energy relationships and electronic configuration and in most cases by the spacial configurat>ion of the catalyst.Esters attach themselves directly to the catalyst without any preliminary rupture of the molecule and then rearrange and react with adj acent molecules to give various proportions of acid ketone olefine alcohol carbon dioxide and water depending on the spacing of the atoms of the catalyst. Alcohcl and ether are equally stable towards alumina and react by a mechanism similar to that of the esters. Acetic acid attaches itself to the surface of the catalyst and then reacts with a second molecule imFinging on it to give acetone carbon dioxide and water. Formic acid rearranges as do the esters to give either water and carbon monoxide or carbon dioxide and hydrogen.It also gives formaldehyde and carbon dioxide by the mechanism suggested for the formation of acetone from the acetic acid. Amides must be affixed to some surface before they give nitriles and water but the character of this surface is relatively unimportant. Alkyl halides probably give olefines and halogen acid by a similar mechanism. It appears that the rate of reaction of the two latter classes of conipounds is determined by their rates of adsorption. F. A . 31. (i) In the Presence of Salts of Cerium. (ii) In the Presence of Salts of Iron. A. K. GOAED and E. K. RTDEAL (Proc. Roy. So:. (i) 1924 A 105 135- 145; (ii) 1924 A 105 148-164).-(i) The theories of induced Catalytic and Induced Reactions.GENERAL AND PEYSICAL CHE-WTRY.ii. 161 oxidation proposed respectively by Schonbein (Pogg. Ann. 1857 100 1) and Bach (Cornpi. rend. 1892 16 411) are briefly referred to and in connexion with the conditions under which an induced react’ion is transformed into a continued or catalytic reaction details are given of an experimental investigation of the actim of cerous salts in effecting (1) the induced oxidation of potassium arsenite dissolved in a concentrated solution of potassium carbonate (2) the catalytic oxidation of five reducing sugars viz. dextrose lactose arabinose galactose and laxulose. The oxidising powers of the solutions employed were deduced from measurements of oxidation potentials and a special type of electrode is described for measuring the potentials of substances which behave irreversibly towards the platinum electrode.In the case of the oxidation of potassium arsenite the observed potentials arranged in the order of increasing “ oxygen pressure ” are cerous salt (inductor)-arsenite (acceptor) -equilibrium mixture-perceric salt. In the case of the catalytic oxidation of the sugars the corresponding order is reducing sugar- cerous salt-perceric salt. (ii) Schonbein’s reaction viz. the separation of iodine from a solution of potassium iodide in the presence of hydrogen peroxide and ferrous salts has been similarly studied from the point of view of oxidation potentials and the results are shown to accord with the authors’ views on the nature of coupled and catalytic reactions. Direct evidence has been obtained of the formation of a peroxide of iron Fe,05 by the action of hydrogen peroxide on ferrous sulphate in neutral solution and its decomposition which takes place in agreement with the equation for a unimolecular change has been followed by means of potential measurements.The Influence of the Support on the Activity of Catalysts. K. W. ROSENMTTND and P. LANGER (Ber. 1923 56 [Bl 2262- 2264).-In order to study the effect of varying the nature of the support on which a catalyst is deposited a series of experiments was carried out on the rate of catalytic reduction of cinnamic acid by hydrogen with the aid of a palladium catalyst deposited on various supporting materials in presence or absence of arsenic oxide or carbon monoxide which act as “ poisons.” Kieselguhr-palladium catalysts show the least activity and greatest sensitiveness to poison- ing of all the catalysts examined whilst blood-charcoal and palladium represents the most active and most resistant catalyst.In both cases the activity of the catalyst runs parallel to its resistance to poisoning but this behaviour is not invariably observed. The general conclusion drawn from the experiments is that the influence of a third substance on a catalyst varies according to the nature of the supporting material on which the catalyst rests. The following figures indicate the results obtained using as carriers (a) crude kieselguhr (b) purified kieselguhr (c) purified pumice powder (d) barium sulphate (e) purified bone charcoal (f) purified blood- charcoal respectively.(The numbers represent the average rate at which hydrogen is absorbed) I. Normal experiments 1.90; 0.70; 2.05; 2-30; 1-25; 6-20. 11. Arsenious oxide poisoning J. 8. G. T.ii. 162 ABSTRACTS OF CHEXICAL PAPERS. 0.00; 0-00; 0.40; 0.23; 0.43; 4-25. 111. Carbon monoxide poisoning 0.80; 0.25; 1.65; 0-45; 1.00; 5.45. Velocity of the Hydrogen Electrode Reaction on Platinum Catalysts. L. P. HAMMETT ( J . Amer. Chem. Xoc. 1924,46,7-19). -The polarisation of the hydrogen-hydrogen-ion reaction on platinum electrodes of a wide range of catalytic activity has been investigated. An equation has been developed on general grounds which represents the data satisfactorily for small polarisations. The results are in keeping with the hypothesis that the reaction proceeds in two stages a rapid electrochemical reaction consisting in the discharge of hydro- gen-ions to atoms followed by a much slower combination to form molecules and similarly for the reverse reaction The deviations from the theoretical equation have been discussed and certain possi- ble causes eliminated.An empirical equation has been formulated to represent the data. The equation used by Haber and Russ (A. 1904 ii 309) to represent their results on the polarisation of the quinone-quinol electrode has been shown to be unsatisfactory in that it is incapable of giving both positive and negative polaris- ations with a single equation. A more satisfactory equation is proposed and it is suggested that the intermediate hydrogen mechanism for the quinone reduction is incorrect.Catalytic Activity. C. 0. HENKE and 0. W. BROWN (J. Physical Chem. 1924 28 71-73) .-The authors discuss Rideal’s statement (A. 1920 ii 220) that metals with low overvoltages are catalytically active whilst metals with high overvoltages are catalytically inactive. The authors’ experimental results (A. 1922 i 586 1196 ; this vol. ii 31) show that tin and lead have high over- voltages and are excellent catalysts whilst the well-known catalysts nickel and platinum have relatively low overvoltages. This differ- ence indicates that there is no relationship between catalytic activity and overvoltage. In the reduction of nitrobenzene the catalysts act specifically. Their activity and even the product obtained may be varied if only within certain limits. . There seems to be a relationship between the atomic weight of a metal and its capacity t o act as a catalyst in the formation of azobenzene.The behaviour of iron and antimony catalysts suggests oxidation of the metal by the nitrobenzene and subsequent reduction of the oxide by hydrogen. J. F. S. Photochemical Sensitisation of the [Molecular] Trans- formation of Maleic Esters effected by Bromine. J. EGGERT (Physilcal. Z. 1923 24 504-506) .-The author has investigated the mechanism of the photochemical transformation of maleic acid into fumaric acid employing for this purpose the ethyl esters of maleic acid in place of the free acid used by Wislicenus in his investigation of this phenomenon (Ber. d . Such. Akad. d . Wiss. 1895 47 491). The transformation was effected by bromine in the presence of carbon tetrachloride and the radiation employed was of wave-length 557 436 or 365 pp. Although the amount of ester transformed was under these conditions proportional to the total number of quanta of radiation absorbed the photochemical F. A. M. J. F. S.WNERALOQICAL CHEMISTRY. ii. 163 equivalence law was not obeyed. The transformation is attribut- able to the activation of bromine by the absorption of radiation. The fact that the amount of substance transformed increases exponentially with the temperature a rise of 10" doubling the amount transformed is attributable to the circumstance that the activated bromine molecule can only transform such molecules of the ester as possess more than a certain critical amount of energy. The Radiation Hypothesis of the Velocity of Chemical Reaction. 11. Theory and Experiment in Gas Reactions. A. KISS (Chem. Weekbhd 1924 21 26-31).-A discussion of the radiation hypothesis giving a bibliography of recent work. J. S. G. T. s. I. L. Extraction Apparatus for Liquids. E. MURMANN (Oesterr. Chem.-Ztg. 1924 27 3).-An apparatus is described for use in the continuous extraction of a liquid with light solvents such as ether and benzene. The extracting liquid should be immiscible with the liquid to be extracted. The apparatus containing the latter is inserted between the flask containing the boiling solvent and a reflux apparatus and consists of a glass cylinder into which two tubes lead one from the flask and the other from the reflux apparatus. The tube from the reflux apparatus down which the condensed solvent flows dips below the liquid and reaches nearly to the bottom of the extraction apparatus. The vapour of the solvent passes up the other tube which projects 5-8 mm. above the level of the liquid to be extracted and reaches the condenser through a per- foration at the top of the other tube. The extracting solvent after passing up through the liquid forms a layer on the top which continuously overflows back into the flask. ANON. (Chemistry and Industry 1924 43 87).-A summary of views expressed by Ruska (Arch. Gesch. Med. 1923,15) in general agreement with those of Holmyard (Proc. Roy. Xoc. Med. 1923 16 46). It is definitely held that Geber is Jabir ibn Hayyan the great Muslim chemist of the eighth century and that although the Latin works ascribed to Geber may not be authentic they represent for the most part the state of Arabic chemistry at the time they were written. (Cf. A. 1923 ii 148 314 628 683.) G. W. R. Geber (Jabir ibn Hayyan). A. A. E.

ISSN:0368-1769    

DOI:10.1039/CA9242605133

出版商:RSC    

年代:1924

数据来源: RSC

摘要:

WNERALOQICAL CHEMISTRY. Mineralogical Chemistry. ii. 163 The Analysis of the Interior of the Earth. G. TAMMANN (2. anorg. Chern. 1924 131 96-100).-From our knowledge of the composition of the earth’s crust and assuming that the com- position of the outer silicate layer has not changed greatly during solidification it is possible to draw conclusions with regard to the composition of the middle sulphide layer d 5.6 and of the metallicii. 1f-X ABSTBACXS OF CHEMICAL PAPERS. core d 9.6. Since the three layers must have been in equilibrium in the liquid state it is possible to apply the proposition that the division of metals between a fluid silicate layer and a liquid metal layer occurs in such a way that the less noble more electropositive metals dissolve in the silicate the more noble in the metallic layer.The order of metals from least to most noble (most to least electro- positive) is the order of the heats of formation of their chlorides; when the metals are so arranged and against each is placed the percentage extent to which it occurs in the earth's crust it is obvious that of the metals more positive (less noble) than iron only traces can occur in the earth's metallic core whilst the metals more noble than iron may occur to a much greater extent in the core than in the crust. Nickel is the next most common metal to iron in the crust but being more positive must occur in relatively greater proportion in the core; the composition of meteorites (90% Fe; loo,( Ni) may thus correspond approximately with the composition of the earth's core for the proportion of noble metals (lead bismuth silver mercury gold pIatinum etc.) in the crust is so small that if it were a thousand times greater in the core it would still be below 1% of the core. The density of the middle sulphide layer may correspond with a mixture Fe 20% FeS 70% FeO and silicate lo% which has d 5.2 a t 20" and 1 atm. The small sulphur content of eruptive rocks may be due to the presence of phosphorus silicon etc. in the middle layer which would reduce the sulphur content of the silicate layer (slag). s. I. L.

ISSN:0368-1769    

DOI:10.1039/CA9242605163

出版商:RSC    

年代:1924

数据来源: RSC

摘要:

ii. 1f-X ABSTBACXS OF CHEMICAL PAPERS. Inorganic Chemistry. Attempted Separation of the Isotopes of Chlorine by Adsorption on Charcoal. J. SAMESHIMA K. ~ H A R A and T. SHIEAI ( J . Chem. SOC. Japan 1922,43,761-766).-Dry hydrogen chloride is passed into a tube med with bamboo charcoal (A). When the gas is absorbed by A the tube is heated at 110" and the evolved gas is absorbed in a second tube filled with the same charcoal (B). A is heated further and the gas evolved between 150" and 220" is absorbed in a third tube ( H ) . The same process is repeated until B is saturated t,hen B is heated and the gas evolved below 110" is absorbed in a fourth tube then to the fifth tube (D). The gas evolved between 150" and 220" from H is absorbed in a seventh tube (L) through the sixth tube.The gases in D and L are then transferred to two tubes containing sodium hydroxide solution a t room temperature and converted into sodium chloride. The atomic weight of the chlorine in the two samples of sodium chloride (D and L) and in ordinary sodium chloride was determined by a modification of Richards' method (Carnegie Inst. Washington Publ. No. 25 etc.) the results being 35.4612 35.4588 and 35.4608 respectively. K. K. Isotopes of Chlorine. W. D. HARKINS and T. H. LIGCIETT.- (See ii 143.)INORGANIC CHEMISTRY. ii. 165 Solubility of Iodine in Chloroform. L. GRIMBERT M. w and G. Pontcfi (J. Pharm. Chim. 1924 29 5-9).-The weight of pure iodine dissolved by 100 g. of pure chloroform is 1.314 g. at 0" 1.483 g. a t 5" 1.805 g. a t lo" 2.176 g. a t 15" 2.630 g.at 20" and 3.200 g. a t 25". Traces of alcohol contained in the anaesthetic have no appreciable influence on the solubility of iodine whilst the impurities present in commercial rectified chloroform tend to increase it slightly. Kinetics of the Formation of Iodine from Iodide and Periodate. E. ABEL and A. FURTH (2. physikal. Chem. 1923 107 313-328).-The kinetics of tthe reduction of potassium periodate by potassium iodide in an acetic acid-sodium acetate solution have been investigated a t 25". The reaction which may be represented 104'+21'+2H'=I0,'+12+H,0 follows the laws of bimolecular reactions namely -d[&O,"]/dt- -0*5d[I']/dt= d[I,]/dt=k[IO,'J . [I']. The concentrations of the various substances were varied in the following proportions in different experiments periodate 4 1 iodide 20 1 acetate 480 1 acetic acid 3000 1.It is shown that with a low concentration of acetate (0.1N) the value of Ic is independent of the acetic acid (H') concentration but it depends on the concentration of the acetate itself in the sense that with increasing acetate concentration the value increases. The value of the constant for the unrealisable acetate-free solution is about 380 and for 0-1N-acetate about 545 (time in minutes and concentrations in mols. per litre). With higher concentrations of acetate a dependence on the acetic acid concentration is found and this is the more noticeable the more concentrated is the acetate. This dependence is of such a kind that the constant k with increasing acetic acid concentration a t first increases and then decreases that is it passes through a maximum.In the range of concentrations examined the value of E varies between 239 and 1024 and for the concentrations acetic acid 0.1N acetate 0.4N it has the value 707. The mechanism of the reaction IO,'+I' on the basis of the above results is shown to be IO,'+I'=IO,'+IO' (determined by time) followed by 10'+I'+2H'=I,+H20. The Systematic Doctrine of Affimity. XXV. The Behaviour of certain Halides with respect to Halogen. W-. BILTZ and E. MEINECKE (2. anorg. Chem. 1923 131 i 1-21).- The action of liquid halogen on some halides of elements of the fourth and fifth groups has been examined to see whether complex com- pounds are formed when the halide and halogen are completely miscible. The general conclusion is drawn that no chemical action takes place.Freezing-point diagrams have been constructed on the basis of thermal analysis (cooling curves) supplemented in some cases by vapour-pressure determinations. Lead dichloride cerium trichloride zirconium tetrachloride and thorium tetrachloride are all insoluble in liquid chlorine whilst tungsten hexachloride is very slightly soluble. Arsenic trichloride and the tetrachlorides of silicon titanium and tin are all completely miscible with liquid chlorine forming simple eutectic systems with W. T. K. B. J. F. S. ' VOL. CXXVI. ii. v 7ii. 166 ABSTRACTS OF CHEMICAL PAPERS. qo evidence of the formation of any compound. Carbon tetra- chloride is completely miscible with liquid chlorine and the equili- brium data suggest that a compound may be formed.It is possible however that the observed results may be due to the transformation of carbon tetrachloride which is found to exist in two forms with a transition temperature a t -44". The system carbon tetrabromide- bromine is similar to that of carbon tetrachloride-chlorie. Lead tetrachloride is completely miscible with liquid chlorine and a method is described for its preparation by means of a new substance The Formation of Periodate by the Action of Permanganate on Various Iodine Compounds. R. LANG (2. anorg. Chem. 1923 130 141-150).-Excess of potassium permanganate in the cold in presence of 5N-sulphuric acid converts iodic acid quanti- tatively into periodic acid. The excess of permanganate is destroyed by addition of nitrite and the latter by carbamide after which the periodic acid formed may be determined by reduction with hydriodic acid and titration of the liberated iodine.The results are about 1% too low since the nitrite effects slight reduction of the periodate as does also oxalic acid under these conditions although it does not reduce periodic acid alone even a t the boiling- point. The oxidation of iodates by permanganates is influenced by hydrogen-ion and chlorine-ion concentrations ; the latter inhibits the oxidation at normal concentrations (X/l-Cl') and renders it incomplete at lower concentrations. Hydriodic acid is also oxidised t o periodic acid by sufficient excess of permangsnate. Iodine monochloride is incompletely oxidised ; the chlorine-ions formed prevent complete oxidation of the iodic acid.Iodine cyanide is also oxidised to periodic acid although the reaction is not quite complete. The permanganate in these reactions is reduced only to the manganic stage as is to be expected from the normal potentials and from the observation that periodic acid will oxidise manganous salts. If however the solutions be very strongly acid periodic acid is not reduced by manganous salts although on the other hand it is not oxidised by manganic salts; the effect of hydrogen-ion concentration on the potentials involved is discussed. At the boiling point the oxidation is completely reversed manganese salts in boiling strongly acid solutions being oxidised by periodic acid to permanganate; this observed result is contrary t o the result to be expected from the reaction constants at low and high tem- peratures calculated from the potentials and the heats of formation. An explanation of this discrepancy is suggested in the formation of ozone in hot solutions of periodic acid which might be accelerated by the presence of an aoceptor such as a manganese salt.Kinetics of the Reduction of Periodate by Arsenious Acid. E. ABEL and A. FURTH (2. physikul. Chem. 1923,107,305-312).- The reduction of potassium periodate by arsenious acid in the presence of acetic acid has been investigated at 25". It is shown that the reduction takes place according to the equation -lead pyridine hexachloride. W. H.-R. S. I. L.INORUANIC CHEMIEiTRY. ii. 167 -d[IO,']/dt= -d(A~O,')kEt=5~5[10,'1(As0,') where [IO,'] is the periodate-ion concentration (AsO,'! the analytically determined total concentration of arsenious acid rn mols per litre and t the time in minutes.The velocity is independent of the concentration of hydrogen-ions. The initial concentrations in the various experi- ments were varied in the proportions of 4.5 1 for the periodate 6 1 for the arsenious acid and 4000 1 for the hydrogen-ion concentra- tion. J. F. S. The Early History of Hydrofluoric Acid. J. R. PARTINGTON (Mem. Munchester Lit. and Phil. SOC. 1922-3 67 73-87). Transference Numbers and Ionic Complexity of Hydro- fluoric Acid Solutions. C. W. DAVIES and L. J. HUDLESTON ( J . Chem. Soc. 1924 125 260-268).-Transference measurements were made a t 25" for hydrofluoric acid solutions varying from 0.016 to 2-32?.The transport number of the anion increases uniformly from 0.13 in the most dilute to 0.338 in the most concentrated solutions thereby indicating the formation of a complex ion. The results thus obtained were combined with the conductivity values obtained by Deussen (A. 1905 ii 311) and the mobilities of the hydrogen-ion and fluoride-ion by Kohlrausch. For solutions of great dilution the equilibrium constant for HF-H*+F' has a value K = 7 * 4 ~ 10-4 whilst for the reaction HF,'=F'+HF the equilibrium constant K2=4-7 which increasea slightly with dilution. The mobility of the hydrofluoride-ion has an approximate value of 75. The concentrations of these three ions (He F' and HF'2) in solutions over the range 0-03-2N were also determined. J. B. F. Formation of Ozone at Low Temperatures and Pressures.A. EucKEN.-(See ii 159.) Disulphur Difluoride. M. CENTNERSZWER and C. STRENK (Ber. 1923 56 [B] 2249-2253).-When a mixture of well-dried silver fluoride (1 g . ) and sulphur (4 g.) is heated in a vacuum a gas (about 100 c.c.) is evolved which does not condense a t the ordinary temperature and is believed to be disulphur difluoride S,F,. The surface of the mercury becomes blackened owing to the form- ation of mercuric sulphide. On exposure to the atmosphere sulphur is deposited from the gas. The molecular weight of the gas calculated from the density was found to be about 97.33 (mean of two experiments) and on analysis by absorption in a solution of potassium hydroxide and hydrogen peroxide the following figures were obtained S 64-04% F 35.17%.It is assumed that the discrepancies were due to the presence of a little sulphur dioxide as the gas has not yet been obtained quite pure ; it is also formed on heating mercurous fluoride with sulphur. The gas is colourless and has an odour similar to but even more objectionable than that of disulphur dichloride. It i s at once decomposed by moisture with deposition of sulphur. F. A. M. The Systematic Doctrine of Affinity. Xxnr. The Power of Crystallised Salts to Combine with Ammonia. W. BILTZ (2. anorg. Chem. 1923 130 93-139).-The results - obtained in 7 2ii. 168 ABSTRACTS OF CHEMICAL PAPERS. the twenty-three previous investigations are examined and com- pared in the present paper. An exhaustive table has been drawn up showing for the ammoniates obtained by direct combination of ammonia with the crystalline halides of the alkali and alkaline- earth metals magnesium zinc tin lead copper silver thallium manganese iron cobalt and nickel (1) the decomposition temper- atures at 100 mm.pressure (2) the heat of formation per mol. of ammonia from the next lower compound (3) the total heat of form- ation from the ammonia-free salt per mol. of ammonia and (4) the affinity at 300° i.e. the work produced when 1 mol. of ammonia a t that temperature and normal pressure combines with the pure salt to form the respective ammoniate. Curves showing the disso- ciation pressures a t various temperatures are also given. From the tables and curves for the 167 ammoniates considered it appears that apart from 34 monoammoniates the commonest compounds have the Werner co-ordination numbers 2,4 6 and 8.By plotting the total heats of formation against the co-ordination numbers the valency isobars are obtained the inclination of which is a measure of the decomposition temperature and therefore of the stability ; compounds with a small ammonia-affinity form one or few ammoni- ates although where one only is formed it may have a high ammonia content as in barium octamminecliloride. As a general rule the heat of formation (2) becomes smaller as the saturation with succes- sive molecular increments of ammonia increases. The decom- position interval i.e. the region enclosed by the dissociation curves of the highest and lowest ammoniates also brings out the relations between the various groups.The constitutions of the ammoniates formed from solid salts are by no means simple and they fall into various classes e.g. those in which the anion is included in the complex the molecular ammoniates ; the normal kationammoniates ; the double-shell ammoniates mixed compounds etc. It is impossible to deduce laws of complete application although general rules may apply for various groups. The ability to form ammoniates is also considered with regard to the general physical and chemical properties of the salts. The molecular volumes of the halides of the metals considered obey fairly generally the linear relationship ; the atomic voliimes of barium and strontium have been redetermined and are given as 39 and 33 respectively. Silver thallium zinc lead and univalent copper do not obey the linear rule.The atomic volumes of the halogens deduced by applying Kopp's law agree fairly well amongst t'hemselves and with the values at absolute zero calculated by Herz (A. 1919 ii 220). With the group I1 metals and tin and lead the values indicate a contraction on combination ; with manganese zinc copper and thallium there is an expansion. When the mole- cular volumes are compared with the heats of formation of the halogen salts it appears that the iodides and bromides which have the lowest heats of formation occupy the greatest space; with the alkali and alkaline-earth halides the heat of formation is greater as the contraction is greater but in other groups this is reversed,MORaANIC CHEMISTRY. ii. 169 The melting points show no general regularity and only rules of limited application can be applied t o the solubility.The space lattice measurements are of great interest the closeness of the atoms in the copper and silver halides explaining their anomalous solubilities melting points etc. Applying these tabulated data for physical properties of the halides to the collected data for their ammoniates the following conclusions are drawn (1) I n the group of the alkali and alkaline-earth metals and magnesium the affinities in the formation of the higher ammoni- ates are smaller the greater are the weight atomic volume and elec- trolytic potential of the metal-ion the more strongly negative and the smaller is the anion and the greater the heat of formation and volume contraction in formation of the salt.(2) The iron group obeys the same rules but the effects of the weight of the kation and volume contraction are reversed. (3) The copper zinc and lead compounds obey the same rules in some cases but in others inverse rules. (4) High melting point and low solubility of the ammonia- free salt influence the stability of the ammoniates. Examination of the lattice structure of the ammoniates by X-ray spectrum methods shows that for the hexammine-compounds of the halides of the iron metals manganese and zinc the ammonia molecules are very closely packed there being practically no free space in the metal-hexammine complex; the spaces occupied by the halogen atoms are in agreement with those observed for normal polar binary compounds. The formation of the ammonia compound is in these cases accompanied by a separation of the metal from the halogen-ion involving mechanical work E which may be calculated.The total heat of formation of the ammoniate involves this value E and the energy concerned in the association of the ammonia molecules with the kation; where E is small i.e. where the kation and anion are already somewhat separated as in the iodides the ammoniate will be more stable for the same kation. The differences in affinity in the formation of the ammoniates are therefore bound up with space considerations; the instability of the ammoniates of the alkali metals for example depends not so much on the low energy involved in association of the kation with ammonia mole- cules as on the high value of E.s. I. L. Catalytic Oxidation of Ammonia by Air in Contact with pure Palladium. E. DECARRIBRE (BuEZ. Xoc. chim. 1924 [iv] 35 48-58; cf. A. 1923 ii 631).-The catalytic oxidation of am- monia by air in contact with pure palladium has been.studied in a manner similar to that used in the case of platinum (A. 1919 ii 463). The palladium is a more effective catalyst as foil than a3 wire wire in turn being more effective than sponge. As with platinum oxidation attains a maximum for a certain temperature range and an ammonia concentration range the temperature being the more important factor. Palladium is more efficient than platinum with high concentrations of ammonia 90 yo oxidation being obtained with concentrations up to 11%. With platinum foil oxidation is as efficient a t 450" as with palladium foil at 650-if.170 ABSTRAOTS OF CFIEMTCAL PAPERS. 700". Maximum oxidation (920/6) occurs in the case of palladium a t 740-780" and in that of platinum at 630" (95%). Increase in the curvature of the surface of the palladium catalpt gives an optimum oxidation temperature of 760" the effect of surface thus being much less than in the case of platinum. When used as a catalyst as above palladium disintegrates and loses it9 activity as a result. E. E. T. Measurements of the Vapour Pressures of Nitric Oxide. F. A. HENGLEIN and H. KRUGER (2. unorg. Chem. 1923 130 181-187).-Pure nitric oxide was obtained by the action of nitrosylsulphuric acid on mercury and after several distillations the solid was pure white the liquid having a faint blue colour.The temperature control for pressures of 30 to 800 mm. of mercury was solid mercury which when preserved in a vacuum rose in temperature O * O W - l O o per minute; for pressures from 0-01 to 3.0 mm. the manometer arm was immersed in a liquid air bath. Temperatures were measured by an oxygen vapour-pressure ther- mometer. The results were plotted and are expressed for solid nitric oxide by the formula log p= -867.4/T+10.1466 and for the liquid by logp=-681.1/T+8.4440 p being measured in millimetres of mercury and T on the absolute scale; the m. p. is 109.4" Abs. a t a pressure of 165.7 mm. from which and the critical pressure (64 atm.) the heats of sublimation and vaporis- ation a t the melting point are calculated as 3980 and 3080 cal. respectively; the latent heat of fusion is therefore 900 cal.The b. p. a t 1 atm. is 122.4" Abs. a t which temperature the latent heat of vaporisation is 3024 cal. so that Trouton's coefficient A/T is 24.7 a high value indicating association in the liquid state. The Clausius-Clapeyron function Sp/p .8T is relatively great ; the substance therefore shows a very great change of vapour pressure for a relatively small change of temperature and is especially suitable for temperature measurements over the range -164" to -145"; a t -150" a change of 1" corresponds with a change of 90 mm. pressure. s. I. L. Catalysis in Homogeneous Gas Reactions. 11. Catalysis of Nitrosyl Chloride Formation by Nitrogen Peroxide. A. KISS (Rec. truv. chim. 1924 43 68-79; cf. A. 1923 ii 237).- The interaction of nitric oxide and chlorine is no$ appreciably accelerated by small concentrations (1-2 yo) of nitrogen peroxide but the acceleration is considerable with about 10% of that com- pound.Measurements were made of the reaction velocity with varying concentrations of nitrogen peroxide a t 18" in the manner previously described (loc. cit.). There is no doubt that the reaction is truly catalytic with intermediate formation of the compound NO,Cl thus 2N0 +C1,= 2N02C1 ; 2N02C1 +2NO = 2NOC1 +2N02. The former reaction is rapid; the latter is slower and W a true gas reaction of the second order. The reactions involved are independent of the surface of the reaction vessel and are not influenced by light. E. H. R,MORGAMC CIIEMISTRY. ii. 171 The Metaphosphates.P. PASCAL (Bull. Soc. chim. 1923 [iv] 33 1611-1627).-A republication in fuller detail of work previously described (A. 1923 ii 489 563). H. J. E. " Insoluble " Alkali Metaphosphates. P. PASCAL (Compt. rend. 1924,178,211-213).-The decomposition by heat of sodium dihydrogen phosphate or of sodium ammonium hydrogen phos- phate affords the acid pyrophosphate NazH2P20 some meta- phosphate also resulting if the temperature is allowed to rise too rapidly. With control of heating definite products are always obtainable. At 250" the acid pyrophosphate first formed begins to pass into an insoluble metaphosphate this change being complete in an hour a t 355". The meta hosphate is stable up to about 477" passing into trimetaphosp !I ate apparently irreversibly a t higher temperatures.At 510" 97 yo of metaphosphate passes into trimetaphosphate in an hour. As shown previously (this vol. ii 39) hexametaphosphate is formed a t still higher temperatures. The metaphosphate is scarcely affected by boiling water or by saline solutions. A second metaphosphate is formed on heating sodium methyl (or ethyl) phosphate to redness the friable product being washed with water to remove trimetaphosphate. Crystals so obtained are used to inoculate metaphosphate prepared as follows hexa- metaphosphate is superfused at 550" inoculated with the above crystals and cautiously heated to aid crystallisation. In this way a completely crystalline insoluble metaphosphate is obtained (m. p. 804-811"). It is only affected by heat if it contains tri- or hexa-metaphosphate when fusion affords a soluble salt.The metaphosphate dissolves in water in presence of traces of soluble pyro- or meta-phosphates to give the viscous colloidal liquids described previously (loc. cit.). The elasticity observed with hexa- penta- and deca-metaphosphates by various authors is due to the presence in these salts of the metaphosphate. The latter appears to be closely related to the hexametaphosphates since in the solid or liquid states it gives the same complexes as the latter with iron and uranyl salts with simultaneous loss of viscosity. E. E. T. The Dissociation Constant of Boric Acid. E. B. R. PRIDEAUX and A. T. WARD ( J . Chem Soc. 1924 125 69-71).- The apparent dissociation constant of boric acid (uncorrected for incomplete ionisation) is calculated from the hydrogen-ion con- centrations given in the literature supplemented with determin- ations in the more dilute solutions.The constant varies for borax solutions from 5x10-lo up to nearly 1 ~ 1 0 ~ in 0.25M- and 0~02M-solutions respectively. The constant diminishes with in- crease of the degree of neutralisation. Thus for 0~05M-solutions 12% and 64% neutralised the constants are 8.4 x and 7.1 x 10-lo respectively. Substituting ionic activities for ionic concentrations a .limiting true dissociation constant should be obtained which f a r dilute borate solutions is found to be nearly independent of theii. 172 ABSTRACTS OF CHEMICAL PAPERS. degree of neutralisation and has a value 5 x 10-lo which agrees closely with the apparent constant in 0.2M and 0-25M solutions.J. B. F. The Nature of Graphite and Amorphous Carbon. G. ASAEARA (Sci. Papers Inst. Phys. Chem. Research 1922 1 23-29 and Japan. J . Chem. 1922,1,35-41).-The author has established the essential identity of graphite and amorphous carbon by the X-ray analysis of thirty-four different forms of carbon from different sources. The material was ground to a fine powder compressed to a layer of about 1 mm. thickness and fixed in front of a small window of a wooden box containing a photographic plate. The interference figure of the material was recorded on the plate as concentric rings the plate cutting the mantles of cones of the diffracted rays. Six of the cameras,. in which the wooden box was placed were arranged conically with the vertex in the source of the X-rays facing the X-rays bulb in a shielded box.The interference figures thus obtained constitute a continuous series gradually altering from those with six comparatively sharp concentric rings of intensity maxima (Type I) to those with an indistinct halo of distributed intensity (Type 111) those where the intensity maxima are not sharply defined but broadened t o ill- defined bands (Type 11) lying between the two extremes. Certain forms of graphite such as Ceylon and Korean graphites etc. gave figures with radial streaks (Type IV) the intensity maxima being common to graphites appearing intact. The streaks are to be attributed to the fact that these graphites are only ground along their flat cleavage faces and the incident rays fall chiefly and almost perpendicularly on these cleavage faces. The angles corre- sponding with the six concentric rings which the diffracted rays make with the incident rays are approximately 14" 18" 50' 22" 20' 25" 40' 31" and 36" 30' respectively.The natural and artificial graphites however finely they may be powdered have a crystalline structure as they plainly show sharply defined interference rings. Amorphous carbons also gave interference figures but no distinct maxima were observed owing to the distributed intensity. The estimated maxima are however invariably a t about 14" and 23". Carbons produced by the decom- position of certain gases or vapours such as carbon monoxide acetylene or carbon disulphide or iron carbide and coal gave no indication of definite crystalline form but they gave definite interference figures which establish their crystalline nature.The word " amorphous " must therefore be replaced by " very minutely crystalline " (cf. Debye and Scherrer A. 1917 ii 437). The Adsorption of Air by Charcoals at Low Temperatures. K. HAYASHI ( J . Chem. SOC. Japan 1922 43 535-544).-The volume of air absorbed a t 1 atm. pressure a t -185" -79" 0" and the ordinary temperature by charcoal from twenty-two different aources and by silicic acid gel was measured. The results together with apparent densities are tabulated in the original. K. K. K. K.INORGANIC CHEMISTRY. ii. 173 The Decomposition of Carbon Monoxide. G . FESTER and G. BRUDE (Ber. 1923 56 [B] 2245-2249).-By the use of a catalytic mass consisting of palladium deposited either on activated charcoal or on activated silicic acid the authors have shown that the decomposition SCO=C+CO occurs at quite moderate tem- peratures either with a mixture of equal volumes of hydrogen and carbon monoxide or with the pure monoxide alone.Using palladium and charcoal and passing a mixture of hydrogen and carbon monoxide over it a t loo" up to 16.1% of carbon dioxide was obtained ; using pure carbon monoxide 1 yo of carbon dioxide was obtained a t 65" and 11% a t 110" ; by allowing the monoxide to remain in contact with the catalyst for sixteen hours the per- centage of dioxide obtained was at 35" 0.6y0 ; loo" 12.0y0 ; 240" 55%. With palladium deposited on activated silica and dried at 60" the following amounts of carbon dioxide were obtained (using pure carbon monoxide) at 35" 0.7% ; 65" 3.6% ; loo" 14.6% and 8-87;.I n all cases however a few per cent. of hydrogen were obtained due to adsorbed moisture in the silica. The authors contradict the statement of Orlov (A. 1909 i 77) that a mixture of hydrogen and carbon monoxide when passed. over a nickel-palladium catalyst yields up to 8.3% of ethylene ; they regard the reactions obtained as being probably due to small amounts of saturated hydrocarbons together with nickel carbonyl. F. A. M. Gases extractable from Heated Steels and Reducibility of the Oxides of Carbon. N. PARRAVANO and C. R. DEL TURCO (Atti R. Accnd. Lincei 1923 [v] 32 ii 373-376).-As a general rule de-oxidation of steel is found to result in diminution of the proportion of carbon monoxide extractable from the hot metal in the usual may.Further the deoxidising agents commonly used in the steel industry are able to educe both carbon monoxide and dioxide a t high temperatures. [Cf. B. 1924,176.1 The Influence of Catalysts on the Production of Potassium Perchlorate by the Action of Heat on Potassium Chlorate. W. FARMER and J. B. FIRTH (J. Chem. Xoc. 1924 125 82-87).- Potassium chlorate with which 15% of catalyst is intimately mixed was heated in glass unglazed porcelain and quartz tubes respectively. For temperatures between 480" and 550" the results were simila.r the period of heating being determined by the temperature of the experiment. The yield of potassium perchlorate is not increased by the addition of silver oxide the whole of the silver oxide being decomposed before any appreciable amount of perchlorate is formed. The addition of cerium and thorium oxides results in almost complete decomposition of the chlorate into chloride and oxygen whilst zirconium oxide and glass considerably reduce the yield of perchlorate but increase the yield of chloride and oxygen.Aluminium silicate and potassium dichromate considerably reduce the percentage of chlorate decomposed. The yield of perchlorate is influenced by the nature of the react'ion vessel being greatest for quartz vessels. A mixture of potassium chlorate with 15% T. H. P. 7"ii. 174 ABSTRACTS OF CHEMICAL PAPERS. of powdered quartz heated in a quartz tube at 500" for seventy five minutes resulted in the conversion of 59.3% of chlorate into perchlorate 3% of the chlorate being unchanged.Revision of the Atomic Weight of Sodium. E. MOLES and J. M. CLAVERA (2. physikal Chem. 1923 107 423435).-The authors have converted sodium azide into sodium nitrate and as the result of eight experiments the ratio NaNO NaN is found to have a value which lies between the extreme values 1.30731 and 1.30738. From this the atomic weight of sodium is calculated to 22*998&0.002 on the basis that nitrogen has the atomic weight 14.008. The present value is in full agreement with the value found by Richards and Wells (A. 1905 ii 450) and all other modern determinations. The ratio NaNO NaN confirms the value 14.008 accepted for the atomic weight of nitrogen. The present work confirms once more the statement that the atomic weights of carbon (12.005) and sulphur (32.060) derived on the basis Na=22-995 by Richards and Hoover and adopted by the International Atomic Weight Commission since 1916 are incorrect.The values C=12-000 S=32.070 are to be preferred. C. MATIGNON (Bull. 8%. chim. 1924 [iv] 35 29-31).-The heat of formation of sodium silicate has been determined from the interaction of sodium silicate and dilute hydrochloric acid the following equation being obtained N?,SiO (solid) +2HC1 (dissolved) = SiO,,aq. (precipitate) +2NaC1+ H,O+32.8 cal. whence SiO,,aq. (precipitate) +Na,0=Na2SiOs+ 51 *I cal. ; SiO&aq (precipitate) +Na,C0,=Na,Si03+C02-24~7 cal. ; and Si (crystallme)+30+2Na=Na2Si0,+331*4 cal. (Cf. Mulert A. 1912 ii 626.) E. E. T. A Cmfhmation of Faraday's Law for Lithium Hydride.K. PETERS (2. anorg. Chem. 1924 131 140-172).-The work of .Moers (A. 1921 ii 200) which showed that on electrolysis of lithium hydride hydrogen is evolved a t the anode has been repeated and extended quantitatively the results being in agreement with Faraday's law. The m. p. of the hydride wm found to be 697'; it diffused readily through the walls 3 mm. thick of the steel tube used for the cell and mechanical difficulties were experienced from the great adhesive ower of the molten hydride from attack of the iron vessel a n l from dissociation a t the high temperature. fially it was found that the solid crystalline hydrate can be electrolysed a t temperatures much below the melting point and measurements of pressure increases due to hydrogen evolved at 558" with a current of 2 amperes at 10 v.after allowance for the volume formed by dissociation showed that Faraday's law is obeyed for a short time although the evolution rapidly feu off for a number of reasons. By an ingenious arrangement in which the hydrogen evolved was allowed to escape into a relatively large space the pressure in which gave by reference to curves previously constructed the volumes of gas it was possible to obtain truly quantitative results a t temperatures of 630-675" the average of J. B. F. J. F. S. Heat of Formation of Sodium Silicate.mORGAKIC CHEMISTRY. ii. 175 many experiments giving a yield on the current of 99.5% and many giving the exact figures demanded by theory. Metallic lithium free from hydride could not be obtained from the cathode but analyses of the mixtures of hydride and metal obtained from the cathode rod showed percentages of lithium varying from 55% to 74%.The acidic character of hydrogen in lithium hydride is thus proved confirming the work of Moers ( 7 0 ~ . cif.) and the results of the X-ray spectrum examination by Bijvoet and Karssen (A. 1922 ii 499 569; 1923 ii 857). In the course of the work it was observed that after remaining for a few minutes in contact with iron molten lithium very rapidly penetrates i t ; hence the dissociation of the hydride a t high tem- peratures in iron vessels is greatly increased by removal of the lithium set free a phenomenon attributed by Ephraim and Michel (A 1922 ii 58) to the supposed reaction LiH+Fe=LiFe+H. After remaining for some time in contact with iron melted lithium or its hydride completely removes carbon sulphur silicon phos- phorua etc. from the iron which accounts for the difEculties encountered by earlier workers who observed abnormal evolution of hydrogen from the hydride heated in steel tubes.From observ- ations carried out with knowledge of these facts it is clear that the dissociation pressures observed are abnormal being influenced by Lithium Perborate. (WE.) R. BEZNER L ~ W Y (Bd. 8oc. Chim. Romdnia 1923 5 81-82> .-Lithium perborate has been prepared both by the hydrogen peroxide and the electrolytic metho&. In the former lithium metaborate LiB0,,8H20 (pre- pared by fusing lithium carbonate with boric acid and dissolving the mass in boric acid solution) is treated in aqueous solution with an excess of hydrogen peroxide a t 0" ; absolute alcohol is then added to the solution.Evaporation in a vacuum a t 40" yields the per- borate as a white amorphous powder. In the electrolytic method the electrolyte is a solution of 4.5 g. of lithium metaborate and 13.5 g. of potassium carbonate in 100 C.C. of water the anode a platinum spiral and the cathode a tube of tin through which water a t 15" is circulated. A current density of 2 amps./cm.2 at 6 v. is employed. The maximum yield is obtained after one and a half hours. Analysis shows the perborate to possess the formula Li,B,05,2.5H,0. It is stable a t the ordinary temper- atures but decomposes above 50". It is soluble to the extent of 10.13 g. in 100 C.C. of water a t the ordinary temperature.The Differences between Mixed Crystals obtained from Melted Salts and from Solutions. G. TAMPUNN and w. KRINGS (2. aizorg. Chem. 1923 130 229-245).-The study of the differences in properties of mixed crystals according to whether they are obtained from mixtures of the pure salts in the melted state or from solutions (Tammann A. 1919 ii 398) has now been extended to the investigation of the heats of solution. The methods of obtaining and analyshig the mixed crystals and the the solubility of lithium in its melted hydride. s. I. L. J. W. B. 7*-2ii. 176 ABSTRACTS OF CHEMTCAL PAPERS. calorimeter devised for the measurements are described mixtures of potassium chloride and bromide potassium and rubidium chlorides and barium and lead nitrates being employed.By interpolation of the results to allow of comparison of mixtures of the same compositions the following conclusions are drawn (1) Mixed crystals obtained from the melted salts (melt cryshls) have appreciably lower heats of solution than those of the same composition from solutions (solution crystals) ; for mixtures of potassium chloride and bromide the differences are 0-4-44 cal. /g. (1-1.5% of the total) for potassium and rubidium chlorides 0*5-2*0 cal./g. (14%). For the pure salts potassium chloride and potassium bromide fusion has no effect on the heats of solu- tion. (2) The heats of solution of mechanical mixtures of the salts do not agree with those calculated from the simple mixture rule but are appreciably lower ; they are equal to the value for the solution crystals for mixtures of potassium and rubidium chlorides and lead and barium nitrates but 1.3 cal./g.smaller for mixtures of potassium chloride and bromide. (3) The anomalous double refraction of mixed crystals of lead and barium nitrates is not connected with the heats of solution for the latter remain unaltered by heating at 340° whilst the optical anomaly disappears after heating at 340° and does not reappear on cooling. From the fact that the solution crystals were generally cloudy and from results obtained by other workers it appears probable that such crystals are not homogeneous in structure this view is strengthened by a consideration of the conditions under which separation from solution occurs especially in view of the fact that the composition both of the crystals and of the solution alters during the separation. The various examinations which have been made of the X-ray spectra of mixed crystals including hitherto unpublished investigations by Wever show that interference- spectra are observed which are identical not only for solution and melt crystals but also for mechanical mixtures of the component,s SO that these give no information on the differences under dis- cussion.The specific gravities of mixed crystals from the melted potassium chloride and bromide are considerably lower than those calculated from the simple mixture rule (Retger’s rule); owing to inclusions this property has not been d-etermined for the solution crystals. The m. p. curves have been constructed for the systems potassium chloride and bromide and potassium and rubidium chlorides ; the latter system shows a distinct crystallisation interval the former definite solidifying points; the curves for the latter fall continuously; that.for the former has a minimum a t 740° 70-80% of potassium bromide. Curves have also been con- structed to show the wide variations in composition between the mixed crystals and the solutions from which they separate. s. I. L. Henry’s Law as Applied to Aqueous Ammonia Solutions and the Hydrolysis of Ammonium Salts. E. KLARMANN (2. anorg. Chem. 1924 132 289-300) .-The partial pressures ofINORGANIC CHEMISTRY. ii. 177 aqueous ammonia solutions were determined by a modification of the dynamic method for concentrations ranging from 0.5N to 0.0083N.The results are in agreement with Henry's law. The formation of an aqueous solution of ammonia is represented accord- ing to the " quinquevalent " and <' quadrivalent " theories respectively by the equations (!) NH,+H,?=NH,OH ; NH,OH=NH,'+OH'. (u) NH3+H =NH,' ; H'+OH'=H,O. The two theories are developed from the point of view of the law of mass action and the degree of hydrolysis of ammonium acetate is calculated. From conductivity measurements of dilute aqueous ammonia and dilute acetic acid a t O" the dissociation constants are Koc,a,o,=l.68 x and K 0 ~ ~ o ~ = 1 - 5 2 x The conductivity and freezing point of aqueous solutions of ammonium acetate were @lso determined. The former is determined exclusively by the ions present in solution whilst the depression in the freezing point is determined by the total effect of uiidissociated salt elect'rolytic and hydrolytic products.The results for the depression of the freezing point exhibit anomalies and it is not possible to combine the results in order to establish the degree of hydrolysig. Partial pressure measurements of ammonium acetate solution a t 0" show that Henry's law applies for the hydrolytic ammonia formed. Electrolytic Preparation of Ammonium Persulphate. J. SALAUZE (BvZZ. Soc. chim. 1923 33 [iv] 1738-1761).-An extensive investigation of the electrolytic preparation of ammonium persulphate by different methods. The preparation may be effected either in acid or in neutral solution (chromate method Miiller A. 1912 ii 895 etc.). The second method gives the higher yield (80-85%) but has many disadvantages.The chromium hydroxide diaphragm formed on the cathode (which must be of platinum) is very fragile and dissolves if the solution becomes slightly acid. Temperature has little effect on the yield which is as good a t 22" as at 8" whilst the yield is improved if chlorides are added in traces. The process can only be followed by measuring (1) the gas evolved or (2) the cathode potential which necessitates con- tinued attention. Another drawback to the method is due to the great solubility of ammonium persulphate in neutral or alkaline solution. Prolonged electrolysis is needed before the salt separates. Electrolysis in acid solution does not require much attention and the ammonium persulphate separates readily.Electrolysis may be effected in presence or in absence of potassium ferrocyanide addition of the latter increasing the yield but giving a product slightly contaminated with Prussian-blue which is difficult to remove. The conditions for a good yield (70-75%) are as follows (1) a low temperature (2) a concentration of sulphuric acid corre- sponding with the presence of ammonium hydrogen sulphate (3) a fresh platinum anode (anodic current density 50 amps. per sq. dm.) (4) a platinum cathode and (this being very important) the largest possible cathodic current density and (5) the presence of a trace of chloride-ion. J. B. F. E. E. T.ii. 178 ABSTRACTS OF CHEMICAL PAPERS. Double Decomposition in the Absence of Solvents. V. A. G. BERGMANN ( J . Russ. Phps. Chem.Soc. 1924 54 625-637; cf. A 1923 ii 568 636 761 764).-1. The system silver chloride- mercuric iodide. The systems previously investigated bcluded those formed by mercury halides with the nitrates of it variety of metals; such an investigation is rendered difficult by the fact that mercuric nitrate decomposes on melting and its behaviour towards the chlorides of other metals cannot be studied. The system now investigated does not suffer from this disadvantage ; the thermal effect of the double decomposition is practically zero and more- over both mercuric iodide and silver iodide are coloured and thus facilitate observation. It has been found that the system does not form complex compounds; there is a simple eutectic at 136". The componentas appear to interact below this temperature in the solid state as evidenced by the change of colour from reddish to orange- yellow.The reason for the absence of interaction is to be sought in the low ionisation of mercuric iodide; this is proved by the results obtained with the system silver iodide-mercuric chloride. It is found that the liquidus and solidus curves are in this cake identical with those previously obtained with silver chIoride and mercuric iodide as also is the colour etc. Carefully dried silver iodide and mercuric chloride interact on mixing in the cold the colour gradually changing from pale yellow to orange-red. 2. A new type of reciprocal system. Four-component systems can be divided into three categories of which the most common is formed by four independent components (the simplest of these is one of four elements).The diagram of state of such a system is represented by a regular tetrahedron (Janecke A. 1908 ii 808 841; 1912 ii 750 762). The second category includes reciprocal systems in which there is an equilibrium U + N Y MY+NX in the liquid state and also apparently in the solid state. These systems can be formulated as three-phase systems (Janecke Eoc. cit.); the best-known examples are those of potassium chloride- magnesium aulphnte and sodium chloridepotassium sulphate. It is now proposed to distinguish a third category the irreversibly reciprocal systems. The difference from the second category con- sists in the fact that the reaction is complete in one or other direc- tion the system 2AgI+HgCl -+ 2AgC1+Hg12 afiording a good example.The systems formed by mercury halides with silver and thallium nitrates can also be referred to this group although they could not be examined completely owing to the instability of mercuria nitrate. A method of graphic representation of such systems is given and it i~ pointed out that the three categories have the following morphological characteristics the first gives rise to one quaternary four ternary and six binary systems ; the second to one quaternary and four binary and the third to two ternary and five binary systems. The Energy of Crystallisation of Ignited Gypsum. M. VON GLASENAPP (2. amrg Chern. 1923 130 246-252).-The author has already shown that the setting of ordinary gypsum which has G. A. R. K.INORCIAXIC CHEMISTRY. ii.179 beeh ignited a t temperatures of 130-200" is due finally to the recrystallisation of the soluble hemihydrate CaSO4,O.5H,O in the form of the relatively insoluble dihydrate CaS0,,2H20. Burnt gypsum in the form of small plates or cubes allowed to stand with one surface in water recovered in two hours about one-third more than the water originally driven off by heating; after fully drying the increase was exactly equal to the amount originally driven off. Microscopical examination before ignition and after regeneration (burning and then keeping in water and finally drying) showed that the relatively coarse crystalline st,ructure before .jgnition had been converted into a fine leaf structure whilst in the powdered con. dition the addition of water results in fine needles.Material regenerated in lump form is much harder than the original but has a lower specific gravity; originally clear lumps become cloudy. If alum solution is used instead of water for the regeneration the final material has a hardness approaching that of marble. These changes are attributed to the diflicult conditions of recrystallisation in the solid lump. Small lumps of ignited material exposed to an atmosphere saturated with water vapour in a closed space required from forty to seventy-five days to recover the original weight. Microscopical examination showed that recrystallisation occurs in this case in extremely fine needles with a greater change in the specific gravity. The hardening of pavement gypsum a variety obtained by heating to 1000° which is accompanied by some loss of sulphur trioxide is very different from the above the alteration in form of crystal being less and complication is introduced by the formation of carbonate from the free lime.s. I. L. Thermal Effects on the Heating Curves of Barium Strontium and Calcium Oxides and their Carbonates in Graphite Tubes. G. TAMMANN and K. I?. GREVEMAYER (2. anorg. Chem. 1923 130 205-208).-During the heating of the oxides in graphite tubes an acceleration in the rate of rise of the temperature was observed between 700" and 900" with barium and strontium oxides followed in each case by a marked reduction in the rate a t higher temperatures; the first effect is a t temperatures at which carbon monoxide and dioxide are formed by oxidation of the graphite tube but the second effect occurs at temperatures 100" or more below those a t which the respective carbonates have a dissociation pressure of 1 atm.If however the carbonates be mixed with soot or charcoal the temperature at which the reduc- tion in the rate of rise occurs is considerably reduoed i.e. the carbonate decomposes at lower temperatures by reason of the diminution of the partial pressure of the carbon dioxide due to the formation of carbon monoxide ; industrially charcoal is added to barium carbonate in order to causticise it at lower temperatures. The acceleration and subsequent decrease in the rate of rise of temperature on heating the oxides in graphite tubes are therefore due to formation and subsequent dissociation of the Carbonates. s. I. L.ii.180 ABSTRACTS OF CHEMICAL PAPERS. Hydrates in Aqueous Solution. I. The Beryllium-ion. R. FRICKE and H. SCH~~ZDEILER (2. anorg. Chem. 1924,131,130- 139) .-Examination of solutions of beryllium chloride confirms that of all bivalent metallic ions in aqueous solution the beryllium-ion is the most highly hydrated. Solutions prepared by neutralising the pure hydroxide with the equivalent quantity of hydrochloric acid have abnormally high conductivities possibly from the presence of some uncombined hydroxide in colloidal solution. The measure- ments were therefore made with solutions prepared from the sulphate by double decomposition with the equivalent quantity of barium chloride. Measurements of the viscosity at 25" of solutions from 0.053N to 0-526N gave value greater than for solutions of magnesium zinc and ferrous salts.Similarly the molecular depression of the freezing point for aqueous solutions has a pro- nounced minimum at a concentration of about 0-103N which is considerably less than the corresponding concentrations for solu - tions of ferrous and magnesium chlorides. The mobility of the beryllium-ion at 25" as determined from conductivity measure- ments is about 30 the lowest value yet observed for a metallic ion with the possible exception of the thorium-ion. The great degree of hydration of the beryllium-ion indicated by these measurements is in agreement with the marked deliquescence of the chloride its great heat of solution and the stability of the tetrahydrate BeC1,,4H20 which was found to suffer no loss in weight even after remaining for six weeks over phosphoric anhydride.On the other hand the volume contraction on dissolution of the chloride in water is very small as calculated from determinations of the specific gravity of the anhydrous chloride and of the aqueous solutions ; in this property the beryllium-ion is closely analogous to the lithium-ion. s. I. L. Properties of Magnesite. T. NISHIMURA and M. MURACHI (Rikwuguku Kenlcyujo 1 h6 1923 2 408419).-The material con- tained MgCO 97*43y0 CaCO 0.68% mineral matter insoluble in acids 1.06y0; the salt MgCO ,3H,O was also used. The mode of hydrolysis of magnesite in water can be observed under a micro- scope by the adsorption of fluorescent dye such as fluorescein eosin or curcumin. Crystallised magnesite adsorbs the dye and emits fluorescence whilst the hydrolytic product does not.With the synthetic compound the diflerence is not distinct. The solubili- ties of the natural and synthetic compounds respectively in water a t 20" saturated with atmospheric air (I) and carbon dioxide (11) at 1 atm. pressure are (I) 0.084 g. and 1.56 g. per litre of solu- tion and (11) 0.156 g. and 25.70 g. per litre. The solubilities of the magnesite in 0-530N-hydrochloric acid O.507N-nitric acid and O647N-sulphuric acid were measured a t 16--19"; the pro- ducts of the velocity constant and a definite surface of the sample are 0.0193 0.0095 and 0.0014 respectively. For 0.589 1.034 1.526 and 2*585N-hydrochloric acid the products are 0.0182 0.0102 0.0073 and 0.0056 respectively. The velocity of hydrolysis of the magnesite is less than its velocity of solution in about 06N-INORGANIC CHEMISTRY.ii. 181 hydrochloric and nitric acids but larger than that in sulphuric acid of almost the same concentration. When the magnesite is covered with the hydrolytic product the solubility velocity in 1.034N- hydrochloric acid ~ one-half to one-third of thatl of the unhydrolysed substance. The presence of 2 mols. of sodium chloride or ammohium chloride or 1 mol. of calcium chloride only slightly decreases the solubility velocity in about O6N-hydrochloric acid. In the case of magnesium chloride if the concentration of the salt is less than 27% the solubility velocity of the magnesite in about 0-SN-hydrochloric acid is moderately decreased ; if however the concentration is above 28% a double salt is formed as in the case of the synthetic salt (cf.A. 1923 ii 689) which is practicdly insoluble in about O6N-hydrochloric and sulphuric acids. K. K. (MLLE.) A. DRESCHER (But!. Soc. Chim. Rorndnia 1923 5 $O-Sl).--A saturated solution of chemically pure zinc sulphate is treated with sodium amalgam at 81"; the zinc amalgam obtained becomes on cooling a semi- solid crystalline mass. The crystals are washed and dried and the mercury is removed by slow distillation at 400" in a vacuum. Zinc of 99-95y0 purity remains. The Action of Hydrogen Peroxide on the Precipitation of Zinc Salts with Sodium Metasilicate. A. H. ERDENBRECHER (2 anorg. Chern. 1924,131,119-129).-Solutions of sodium meta- sihcate in 30% hydrogen peroxide may be obtained containing very high proportions of sodium silicate but cannot be made to deposit crystals.If sodium silicate be added to concentrated solutions of zinc salts in presence of 30% hydrogen peroxide products are obtained in which the proportion of oxygen varies according to the concentration from 2.3 to 7.9% ; they are t o be regarded as mixtures of zinc peroxide and silicic acid admixed with zinc silicates and other products of reaction as shown by the similarity of the dis- sociation curves with those of zinc peroxide and silicic acid mixtures made up for comparison. Precipitation with sodium carbonate gave products containing up to 12.1% of oxygen. Since precipitated zinc silicate does not combine with hydrogen peroxide the latter may be used to distinguish zinc oxide in presence of the silicate.s. I. L. Preparation of Chemically Pure Zinc. J. W. B. Combining Weight of the Lead of a Vesuvian Cotunnite. A. PIUTTI and D. MIGLIACCI (Atti R. Accad. Lincei 1923 [vj 32 ii 468472).-If sufficiently pure lead chloride is phenylated in accordance with the procedure given by Hofmann and Wolfl (A. 1907 ii 521) the resulting lead tetraphenyl may be perfectly purified by alternate crystallisations from benzene and chloroform and treatment with absolute alcohol. Treatment with bromine vapour alone suffices to convert the lead tetraphenyl quantitatively into lead bromide. By this means ibis found that the combining weight of ordinary lead is 207-192 and that of the lead obtained from a sample of Vesuvian cotunnite ori,hating in tohe eruption of 1906 207.050.From these results together with those of ROSSii. I S2 ABSTRAOTS OF CHEMICAL PAPERS. (A. 1908 ii 9) the conclusion is drawn that the lead of the cotun- nite consists of a mixture of ordinary lead with about 12% of uranium-lend. No explanation is advanced for the slight divergence of the value faund for the combining weight of ordinary lead from the accepted value namely 207.20 (cf. A. 1915 ii 455 456). The Solubility of Lead Chloride and Lead Bromide in Aqueous Solutions of the Chlorides of the Alkali and Alkaline- earth Metals. W. HERZ and M. HELLEBRANDT (2. anorg. Chem. 1923 130 188-198).-As a preliminary to the work various methods for the determination of lead in presence of alkali and alkaline-earth metal halides were examined and precipitation with sodium cyanide in the cold was found most suitable ; the precipitate was dissolved on the filter with cold nitric acid and the metal precipitated as sulphide by ammonium sulphide and weighed as sulphate.The solubility of lead chloride in solutions of the chlorides and bromides of sodium potassium calcium strontium and barium in concentrations up to about 4N was found to obey the same general rule declining to a minimum in solutions of concentrations about or below N/1 and then rising again (cf. von Ende A. 1901 ii 241); it is greater in solutions of the bromides than in those of the chlorides of equivalent concentrations. The curves plotted from the results lie very close together for the interval of diminish- ing solubility i.e.the kation plays only a small part but as the solubility increases again the curves diverge. The effect of the kation in solutions of the chlorides up to 3-6N is in the order K Ca Na Sr Ba and above 3.6N K Ca Sr Na Ba the last being most effective; for solutions of the bromides the effect is the same as in the second order but greater so that the solubility of lead chloride in a strong bromide solution maybe greater than in pure water. The solubility of lead bromide in these solutions is very similar. . s. I. L. The Equilibrium between Lead Chloride and Iodide and some Alkali Chlorides and Iodides in Aqueous Solution. (MME.) N. DEMASSIEUX (Ann. Chim. 1923 [ix] 20 233-296).- The systems previously studied lead chloride-potassium chloride- water (A 1914 ii 185) lead chloride-ammonium chloride-water (A.1913 ii 409) lead chloride-sodium chloride-water (A. 1914 ii 271) lead iodide-potassium iodide-water and lead iodide- ammonium iodide-water (A. 1923 ii 565) are dealt with in detail and supplemented by a description of the system lead chloride- lithium chloride-water which behaves in an analogous manner to the system lead chloride-sodium chloride-water (Em. cit.). A general survey of the results obtained shows that from the point of view of double salt formation the alkali chlorides form two distinct groups. Potassium and ammonium chlorides each form two compounds with lead chloride one series having the general formula 2PbCI,,MCl. The aecond pair of compounds do not possess the same type of composition their formulae being PbCl,,KCI,+H,O and PhC$,SNH,Cl respectively.Moreover the latter substance is T. fl. P.INORGANIC CHEMISTRY. ii. 183 €table only above 75" whilst the former can exist a t a considerably lower temperature. The potassium analogue of the compound PbC1,,2NH4CI has been prepared by other methods (cf. Loren2 and Ruckstuhl A 1906 ii 853) but was not detected in the course of the work described. The chlorides of sodium and lithium are differentiated from those of potassium and ammonium in theb action on lead chloride in that they do not .yield wi6h that substance compounds capable of detection by solution methods. However the solubility of lead chloride increases considerably in the presence of a concentrated solution of sodium or lithium chloride a possible explanation being the formation of complexes in solution.The compounds of lead iodide with the iodides of potassium and ammonium do not give the characteristic reactions of lead salts or of iodides with the ordinary reagents; in this respect they differ from the double chlorides described which behave towards reagents as mixtures of chlorides. H. J. E. . Evolution of the Molecule of Cupric Hydroxide in Aqueouq Suspension. (WE.) S. VEIL (Cmpt. rend. 1924,178,329-330). -The molecular coefficients of magnctisation of the following four substances have been compared (1) Freshly prepared blue oupric hydroxide; (2) the oxide obtained by igniting (1) ; (3) the brown hydroxide obtained by warming (1) in presence of water; and (#) the oxide obtained by igniting (3). The coefficients of (3) and (4) are approximately equal and less than that of (2) which is about a third of that of (1).The change of colour from blue to brown is therefore not due as usually supposed to simple dehydration. m m m The Reaction between Copper and Nitrogen Peroxide J. R. PARK and J. R. PARTINCTON ( J . Chem. Soc. 1924,125 72- 82) .-When nitrogen peroxide is passed over specially prepared copper the latter is oxidised to cuprous oxide with the evolution of heat the nitrogen peroxide being reduced to nitric oxide 2Cu + NO = Cu,O + NO. The cuprous oxide then adsorba about 33% of nitrogen peroxide. The actual amount of nitrogen peroxide taken up is variable although the mean value of the afomio ratio O/N is practically 3. The results are consistent with the formation of cuprous oxide and a subsequent adsorption complex yN02,xCu20.The properties of the substance are not in agree- ment with the formula Cu,NO $s suggested by Sabatier and Senderens (A. 1893 ii 374) or Cu,(NO,) as suggested by Tartar and Semon (A. 1921 ii 336). No oxygen or nitrogen is evolved on the decomposition of the nitro-copper by heat. With water nitric oxide is evolved and a solution of cupric nitrate with a little nitrite is left a result which may be explained by the following reactions (i) 2N0 + H,O HNO + HNO,; (ii) 3HN0 HNO $- 2NO+H,O ; (iii) 3&,0 + 14HN O,= 6Cu(N08) + 2N0 + 7H,O. A little cupric nitritB will be formed by the action of undecom-ii. 184 ABSTRACTS OF CHEMICAL PAPERS. posed nitrous acid. The residue of copper found by previous workers had evidently escaped reaction since the product only contained 27% of the nitrogen and oxygen as a maximum.At 65-70" the whole of the nitrogen peroxide is removed by an inert solvent such as carbon tetrachloride leaving a residue of cuprous oxide. This indicates that the nitrogen peroxide is adsorbed and not chemically combined. The Reaction between Solutions of Complex Cupric Salts and Potassium Cyanide. R. LANG (2. anorg. Chem. 1923,130 151-16O).-Two chief factors are involved in the reaction the first being a reduction of the cupric salt to the cuprous state and the second the formation of a cyanide complex in which the ratio CN/Cu may vary from less than 2 to 4 ; both these factors are strongly influenced by the cyanide-ion concentration which is in turn effected by the temperature and by the hydroxyl- hydrogen- ammonium- etc.ion concentration so that no stoichiometric relation exists. The decolorisation of an ammoniacal solukion of a cupric salt by addition of potassium cyanide is certainly due to the formation of a complex cuprous cyanide but no definite complex formula can be deduced; the conclusion of Treadwell (A 1904 ii 172) that the variation in the amount of cyanide required observed in the presence of ammonium salts is due to back-form- ation of cyanide varying with the alkalinity of the solution is not supported. Even in strongly alkaline solutions it is shown that this back-formation of cyanide is extremely slight and its effect on the quantitative aspect of the reaction is entirely masked by the factors discussed.s. I. L. J. B. F. Power of Spontaneous Transformation of Yellow Mercuric Iodide. A. DAMIENS (Cmpt. rend. 1924 178 326-328).- Crystals of red mercuric iodide with an area of about 0.7 sq. cm. were heated a t 155" until completely yellow chilled to definite temperatures and observations made of (a) the time in seconds required for the appearanceof the first redparticle and (b) thenumber of red particles visible three seconds later. The lower the temper- ature of chilling (for temperatures down to 40°) the smaller the value of a and the larger that of b. Chilling to temperatures a little below 40" made b too large to measure. Chilling from 155" to - 80" was not accompanied by an appearance of the red form but on suddenly warming to the ordinary temperature general conversion to the red form rapidly occurred this effect being intensified if chilling was effected to liquid air temperatures (cf. also A.1923 ii 864). E. E. T. The Electrolytic Conductivity of Molten Scandium Chloride. W. BILTZ and W. KLEMM (2. anorg. Chem. 1923 131 22-26).- The scandium chloride was melted under pressure in a quartz apparatus in order to avoid loss by sublimation. The m. p. was 939". The specific conductivity was found to be 0.51 a t 959" 0.55 at 969" 0.57 a t 981" and 0-59 a t 991". A marked decrease in conductivity was observed as the molten salt solidified on cooling.INORGANIC CHEMISTRY. ii. 186 The observed results resemble those of thorium chloride rather than those of aluminium chloride the conductivity of the last- named being from 10-5 to times as much.W. H.-R. Cerium Mixed Cerium Metals and Hydrogen. A. SIEVERTS and G. M~~LLER-GOLDECW (2. anorg. Chem. 1924 131 65-94).-The metals examined were (1) cerium containing 94.9y0 of Ce 4.2% of other cerium metals and the remainder iron arid silicon; (2) a mixed metal containing 53.8% of Ce about 44% of other cerium metals mostly lanthanum and the remainder aluminium; (3) a mixed metal containing 10.2% of Ce 84.3% of other cerium metals (mostly lanthanum) some iron aluminium and silicon and (4) a cerium-manganese alloy Ce L a Mn= 83.5 9.0 7-5. These samples absorb nitrogen extremely slowly even when molten. Hydrogen is not absorbed until the temper- ature exceeds 300"; if however the metal has been previously melted in a vacuum it absorbs hydrogen very rapidly even at normal temperatures.The mixed metal (3) absorbs hydrogen very rapidly even without previous heating ; after previous heating the absorption is even more rapid ceasing only when the metal is cooled in mixed ether and solid carbon dioxide and starting again if the temperature is allowed to rise to -10". Traces of nitride in the metals reduce the speed of absorption and larger quantities are inhibitory. The products saturated with hydrogen a t 20" are greyish-bIack powders that from (1) being spontaneously inflammable. Samples (1) and (2) gave products containing 1*S9y0 of hydrogen ; (3) gave 8 product containing l%yO of hydrogen; the formulae CeH and LaH3 require 2-11 and %13y0 respectively. By plotting the pressures as ordinates against the volumes of gas absorbed at 800" as abscissie curves are obtained consisting of a slowly rising portion continuing in an almost horizontal line and finally rising very sharply; a t 600" and lower temperatures the horizontal portion disappears.The volumes absorbed vary with the previous history of the sample being continuously lessened by repeated heating and cooling. The final sharply rising part of the curves is of the form m=Kl+K,dp where m is the volume absorbed by 1 g. p the pressure and K and K are constants depending on the temperature and previous treatment. The curves are similar to those obtained by Hoitsema and Rooze- boom for the system palladium-hydrogen and suggest the form- ation of two non-miscible solid solutions ; they furnish no support for the view that hydrides CeH and LaH are formed.S. I. L. Rare Earths. XV. Search for Element Sixty-one. L. F. YNTEMA ( J . Amer. Chem. Xoc. 1924 46 37-39).-An account of experiments undertaken to detect the presence of an element of atomic number 61 which was shown by Moseley (A 1914 ii 326) to lie between neodymium and samarium. The materials investi- gated were pure neodymium oxide pure samarium oxide the oxides of intermediate fractions between the two piwe oxides and2. 186 ABSTRACTS OF CHEMICAL PAPERS. intermediate oxides from a gadolinite fractionation and a fergusonite fractionation. The arc spectrum showed that the five faint lines 3305.8 3329-1 3342.5 3378.0 and 3379.2 d. were found in both pure neodymium and pure samarium oxides.The X-ray absorp- tion spectra were found to give the K absorption Limit for samarium easily but on mixing with 5% of neodymium oxide the correspond- ing value for this element was not obtained. Consequently the method was abandoned since the missing element is not likely to be present to so large an extent as 5%. With the primary X-ray emission spectrum it is shown that 0.1% of impurity can be detected in the case of the pure oxides of neodymium and samarium and that if the element 61 is present in these substances it is present to 8 still smaller extent than 0.1%. Hence whilst the spectroscopic evidence indicates that the element of atomic number 61 may be associated with neodymium and samarium X-ray analysis of material from various sources has given no evidence of the presence of this element. Constitution and Evolntion of Precipitates of Alumina.P. PASCAL (Compt. rend. 1924 178 481-483).-From measure- ments of the magnetic susceptibilities of specimens of alumina precipitated under various conditions the author concludes that (1) the precipitate (a gel) formed by the addition of aqueous ammonia to an aluminium salt consists originally of anhydrous alumina; this a t length is converted into the hydrate A1,0,,3H20 which is unstable and is not identical with the crystalline hydrate ; (2) the aluminium in solutions of aluminate is largely present in the colloidal state; (3) in no case does hydrated alumina consist of the true hydroxide Al(OH),. The conclusion is drawn that it is impossible to have more than a certain number of hydroxyl groups attached to the same atom without the occurrence of dehydration as with carbon compounds.A. B. H. Clays. V. Action of Heat. 0. BOUDOUARD and J. LEFRANC (Bull. Xoc. chim. 1923 [iv] 33 1627-1640).-The authors have studied the dehydration of twenty-seven specimens of clay from nine distinct sources. The temperatures of experiment ranged from 182" to 800° and the results obtained are tabulated to show both percentage and molecular loss of water. Graphic represent- ation of the dehydration as a function of temperature indicates that the loss of combined water occurs between 450" and 500". The Formation of the Green Manganous Sulphide. A. MICKWITZ and G. LANDESEN (2. anorg. Chem. 1924 131 101- 118).-The work of Villiers (A. 1914 ii 658) and of Fischer (A.1915 ii 462) whilst confirming that the rose-coloured sulphide is always precipitated first and is not converted into the green form if precipitation has been effected with sodium or potassium sul- phides has not removed $he uncertainty left by earlier workers as to the conditions for obtaining the green form. It is now found that the sole condition is the presence of free ammonia before the J. 8. S. H. J. E.fNOkOANIC CHEBflSTRY. ii. 187 addition of the ammonium sulphide. If no free ammonia be present the rose-coloured sulphide precipitated by ammonium hydrogen sulphide contains more sulphur than is required by the formula MnS and its composition is better expressed by the formula 4,Mn3S4 ; in presence of free ammonia the formula is (NH4)HMn3S4 and this rose-coloured precipitate slowly passes into the green form MnS giving up (NHJSH to the mother liquor.Mechanism of the Reduction of Permanganate and its Physico-chemical Basis. VIII. Kinetics and the Hydroxyl- ion Influence in the Stepped Reaction Manganate-Formalde- hyde-formate. J. HoLLnm (2. physikal. Chem. 1023,107,333- 346; cf. A. 1923 ii 864; this vol. ii 115).-A continuation of previous work. The velocity equations have been deduced for the stepped reaction manganate-formaldehyde-formate for the special case where a tenfold excess of formate is used. The consequence of this condition on the secondary reaction has been deduced; this allows of a simple mathematical treatment and in consequence of the ease with which the second partial reaction of the whole process can be isolated (MAO,“+HCO,’=M~O,+CO~”+OH‘) also allows of the calculation of the velocity constant of the fist partial reaction (Mn0,”+H*CHO=Mn02+H.C0,’S-OH’) from the con- centration of the manganate and from the known velocity constants of the reduction of manganate by formate. The velocity equatioas obtained have been tested in a series of measurements with different hydroxyl-ion concentrations and found to represent the facts In addition to the retardation previously found which is propor- tional to the square root of the hydroxyl-ion concentration a second retardation has been found which is directly proportional to the hydroxyl-ion concentration.The two retardations take place side by side. It is probable that the cause of the second retardation is the formation of the anion H,C(OH)O’ which is produced by a neutralisation process with the formaldehyde and reacts more slowly with the manganate than formaldehyde itself.The temper- ature coefficient of the reduction of mangannte by formaldehyde a t an alkalinity of 0.1N is found to be 1.85. J. F. S. Temperature Periods in the Emission of Occluded Gases from Iron. G. BORELIUS and F. GUNNESON (Nature 1923,113 82-83).-1n addition to properties known to exhibit a periodicity as a function of temperature the speed with which occluded gases are removed from iron on rise of temperature is also periodic. Experiments with electrolytic iron containing occluded hydrogen or nitrogen showed that an accelerated emission of gas takes place more or less in the neighbourhood of the hundreds on the centi- grade scale.A. A. E. Decomposition of [Iron] Pyrites by Meat. (MLLE.) G. MBRCHAL (BUZZ. SOC. chim. 1924 [iv] 35 4347).-A study of the dissociation of iron pyrites (into ferrous sulphide and sulphur) in a vacuum and in an atmosphere of nitrogen. In a vacuum dissociation begins at 500° is much more rapid a t 550° and a t S. I. L.ii. 188 ABSTRACTS OF CHEMICAL PAPERS. 670-680" is complete in eight hours large pieces of pyrites under- going disintegration during the process. At 700-800" the sulphur condenses in tt well-defined form and possibly could so be extracted from pyrites on the large scale. At 850" dissociation is complete in two hours. At higher temperatures (1200") the residue is a mixture of ferrous sulphide and (a little) iron.In an atmosphere of nitrogen dissociation is very rapid at 850". E. E. T. Definitions of Steel and Cast .Iron. K. HONDA (Rikwagah Kenkpjo Ih6 1923 2 405407).-The author has defined steel and cast iron from the content of carbon in them. The maximum amount of carbon contained in iron as a solid solution at the ordinary temperature is 0.035~0 and that of carbon dissolved in iron a t high temperature is 1.7%; this figure has a distinct meaning on the equilibrium diagram of iron and carbon. Steel is therefore defined to be an alloy of iron and carbon which contains 0.035-1*7~0 of carbon. Cast iron is defined to be an alloy of iron and carbon which contains more than 1.7% of carbon. The Solidus Curve of Austenite. G. ASAHARA (RikwaquEu Kenkyujo Ih6 1923 2 420-425).-The author has determined the solidus curve of austenite by the kuenching method.The sample was prepared from " Armco " iron and sugar carbon the content of carbon being 0.46 0-48 0.58 0.95 and 1.44%. The sample was quenched a t 1134-1408" cooled rapidly by throwing into cold water and microscopically examined. The curve occupies a higher position than that of Gutowsky (MetaZZurgie 1909 731) and far lower than that of Carpenter and Keeling ( J . Iron and Steel Imt. 1904 1 224). Derivatives and Reactions of Iron Pentacarbonyl. H. FREUNDLICH and E. J. CUY (Ber. 1923 56 [B] 2264-2267).- The observations of Dewar and Jones (A. 1906 ii 89; 1907 ii 266) on iron pentacarbonyl iron tetracarbonyl and iron nonacarbonyl were confirmed except that the pentacarbonyl was found to react with iodine on exposure to light or on heating in alcoholic solution. All three carbonyls are diamagnetic the values for their mag- netic susceptibilities being Fe(CO) K = -0.51 x lo* ; Fe,(CO) ~=-0*37 x lo* ; Fe(CO) ~=-1.3 x (approx.).The best method for preparing iron tetracarbonyl was found t o be by heating the pentacarbonyl for some time in absence of air with a slight excess of concentrated sodium hydroxide or ammonia solution; the carbonyl dissolved to a brown solution which was treated with excess of ammonium sulphate and extracted with ether. The red ethereal solution was dried with copper sulphate and the solvent removed by distillation the characteristic green crystals of iron tetracarbonyl remaining. The brown solution of the pentacarbonyl in alkali was found to contain a number of complex iron compounds which possess strong reducing properties.A red substance was isolated in dichroic needles but has not yet been obtained sufficiently pure for analysis. It possesses powerful reducing properties. F. A. hl. K. K. K. K.INORGANIC CHEMISTRY. ii. 189 The Magnetisation of Alloys of Nickel and Electrolytic Chromium. J. SAFRANEK (Compt. rend. 1924 178 479480). -The author has studied the magnetic properties of alloys of pure nickel and electrolytic chromium. Accepted values of the con- stants for nickel were confkmed. For chromium the susceptibility was found to be independent of the magnetic field between 2000 and 14,000 gauss and of the temperature between 100" and WOO and to have a value of 4-31 x lo*.The reciprocal of the suscepti- bility of the alloys plotted against temperature gives a straight line becoming concave towards the axis of T a t higher temper- atures. The various magnetic constants are found to be linear functions of the composition. The specific magnetisation at the absolute zero for different alloys has been found by extrapolation and was zero for 13% chromium content corresponding with CrNi,. Such a compound has not been isolated and microscopical observations have shown these alloys to be solid solutions. A. B. H. The Hardness of Mixed Crystals of Copper-Nickel and Iron-Nickel Alloys at Temperatures up to the Melting Point. F. SAUERWALD [with K. KNEHANS] (2. anorg. Chem. 1923 131 57-64) .-Metallic solid solutions are characterised by an increase in hardness as one metal dissolves in another the relation between hardness and composition being expressed by a graph with a well- defined maximum in the region of middle composition.The author's experiments show that this rule also applies to the hard- ness of solid solutions a t high temperatures in the case of copper- nickel and iron-nickel alloys the hardness being tested up to 910" for the copper-nickel and up to 1230" for the iron-nickel alloys in both of which cases a single series of solid solutions is present at high temperatures. The hardness is measured by an indentation method a steel ball attached to a bar being allowed to fall on to the specimens. At low temperatures the iron-nickel results are complicated by the presence of more than one constituent.W. H.-R. Nickel Hydride and the Mechanism of Hydrogenation using a Nickel Catalyst. W. SCHLENK and T. WEICHSELFELDER (Ber. 1923 56 [B] 2230-2234).-Anhydrous nickel chloride on treat- ment with an ethereal solution of a Grignard reagent such as mag- nesium phenyl bromide reacts rapidly forming a dark brown sol containing nickel in colloidal solution. This solution absorbs hydrogen readily the nickel being precipitated as a flocculent deep black sediment leaving a clear solution. Measurement of the volume of hydrogen absorbed indicates that a compound of the formula NiH is formed but on separating the black product and drying it in a current of hydrogen it was found to possess exactly the formula NiH2. On treatment with alcohol the black precipitate at once begins to evolve hydrogen although it is quite stable in contact with ether.This is accepted as further evidence that the usual processes of catalytic reduction in presence of nickelii. 190 ABSTRACTS OF CHEMICAL CAPERS. are essentially chemical due to the formation of nickel hydride and are not merely adsorption phenomena. Chromium-Copper-Nickel Alloys. E. SIEDSCHLACI (2. anorg. Chem. 1924 131 173-190).-0f the three binary systems those of copper-nickel and chromium-nickel have already been investi- gated; the first shows complete miscibility in both the liquid and solid states the melting and freezing-point curves running continuously between the melting points of the pure metals whilst the system chromium-nickel is homogeneous in the liquid phase with a series of mixed crystals in the solid state the curves showing a deep minimum a t 1300" 4 0 4 2 % of nickel. The system copper- chromium has now been examined thermally and microscopically ; there is only partial miscibility in the liquid state with a eutectic near the co$per end (1.5% Cr 1076") the limits within which a mixture of two liquids is formed being 37% to 93% chromium above 1470".In the solid state only heterogeneous mixtures of two crystal species are obtained chromium and the eutectic rich in copper. The triangular and solid diagrams for the ternary system have been worked out and as might be expected from the binary systems the heterogeneity of the copper-chromium system is partly removed by addition of nickel. Mixed crystals were obtained from alloys in the following proportions Cr Ni Cu=(l) 10 80 10; (2) 10 60 30; (3) 20 60 20; (4) 30 :'60 10 and ( 5 ) 40 55 5.With proportions of nickel below 55% only heterogeneous systems were obtained. Of the alloys only (1) showed magnetic properties. The alterations in composition of the liquid and solid phases during crystallisation are worked out in great detail in a number of cases. Microscopical examination of the ternary mixed crystals showed dendritic inclusions which may be due t o the change from @- to a-nickel in the solid state. The alloys which solidify as mixtures show under the microscope beautifully-formed chromium-nickel crystals in a cupreous matrix. Alloys in the region of non-miscibility in the liquid state solidify as shown under the microscope in separate layers more or less emiilsified.Alloys of composition (2) and (5) were examined for resistance to acids but were found to be resistant only to concentrated nitric acid also to sodium hydroxide; the influence of chromium in the ternary system therefore is not apparent in this respect. The alloys forming mixed crystals may be mechanically useful but as the samples obtained appeared under the microscope to contain inclusions partly of slag and partly of gas no detailed measure- ments were made. These alloys have a high silvery lustre even after exposure to moist air but on heating become tarnished with Chromium-Molybdenum and Chromium-Molybdenum- Copper Alloys. E. SIEDSCHLAG (2. anorg. Chem. 1924 131 191-202) .-Molybdenum dissolves readily in molten chromium until the mass contains 25% of the former ; in investigating alloys richer in molybdenum the finely powdered metals were mlxed F.A. M. a dark green oxide. s. I. L.INORGANIC CHEMISTRY. ii. 191 and compressed into briquettes which were then heated but for proportions cf molybdenum above 50 yo the initial solidification temperature could not be measured. The curve shows a eutectic :it 1460" 22.7y0 Mo. The liquid is homogeneous but no mixed crystals are formed below 22.7% Mo the solid consisting of chromium crystals and the eutectic mixture ; above this proportion the solid consists of molybdenum crystals containing 2.5% Cr and the eutectic. The alloys are not likely to be mechanically useful; those rich in chromium cut glass with remarkable ease.The system copper-molybdenum has already been found t o be heterogeneous in both the liquid and solid states whilst chromium and copper form no mixed crystals and are only partly miscible in the liquid state. In the ternary system the three metals remain separate in the solid state. Electrolytic copper mas melted with powdered chromium and molybdenum and the thermal analysis carried out in the usual way; no mixed crystals are formed tlie solids consisting of tlie three metals with the eutectics Cr-Mo and Cr-Cu. The diagram shows that alloys of copper with up to 38% of chromium should dissolve up to 8% of molybdenum whilst chromium-molybdenum mixtures should dissolve up to 7 yo of copper. s. I. L. The Ageing of Chromium Hydroxide.Alkali Chromites and their Solutions. R. FRICKE and 0. WINDHAUSEN (2. anorg. Chem. 1924 132 273-288).-The solubility of pure chromium hydroxide (approximately Cr20,,9H,0) in sodium hydroxide solution was determined at the ordinary temperature. From the solubility curve and the analysis of the solid phase the existence of sodium chromite in the liquid is concluded. The formation of chromite precedes the formation of the hydroxide. With sodium hydroxide below 10N primary sodium chromite is formed whilst above 10N the solution contains also tertiary sodium chromite. Potassium chromite is similarly produced; below 8N-alkali only the primary chromite is formed whilst above 8N the solution contains also secondary chromite. Prom solutions of potassium chromite which have stood for a long time needle-shaped crystals of the formula Cr2O3,3&O,8H2O are obtained.The ageing of chromium hydroxide was also studied. It is more rapid under alkali solution than under water and the rate of ageing increases with increase of hydroxyl-ion concentration. Increase of temperature also accelerates the ageing. The ageing ia not the result of dehydration and is analogous to the behaviour of beryllium hydroxide and the so-called crystalline aluminium hydroxide (cf A. 1920 ii 387). The development of crystalline structure is not revealed by examination with the micropolariscope and the ageing is attributed to an increase in the size of the particles. The subsequent separation of chromium hydroxide from clear chromite solutions is the result of the ageing of chromium hydroxide already present in the solution in the colloidal state.From cryoscopic measurements the lowering of the freezing point by potassium chromite in l.5N-solution increases with the time of standing andii. 192 ABSTRACTS OF CIIEMTCAL PAPERS. is due t o hydrolysis resulting in the formation of secondary chromite colloidal hydroxide and fhally precipitated hydroxide. The existence of chromite in solution is also demonstrated by potential measurements. J. B. F. The Action of Chromisulphuric and Ferrisulphuric Acids on the Decomposition of Hydrogen Peroxide. J. POIZAT (Bull. SOC. chim. 1923 [iv] 33 1606-1611).-The action of chromi- sulphuric chromidisulphuric chromitrisulphuric and ferrisulphuric acids (cf. Recoura A. 1892 ii 783; 1893 ii 470; 1903 ii 599 600) on hydrogen peroxide of various concentrations was studied and from the results obtained the corresponding values of the velocity constants of the reaction were calculated. The values thus found decreased with the progress of the reaction; this is ascribed to the dissociation of the acids into sulphuric acid and metallic sulphate with subsequent partial hydrolysis of the latter.I n the case of ferrisulphuric acid an additional effect is produced by reason of the catalytic activity of the liberated ferric sulphate. H. J. E. The Chlorides of Bivalent Molybdenum Tungsten and Tantalum. 11. Tri-molybdenum Hexachloride [" Molyb- denum Dichloride "1 and its Derivatives. K. LINDNER E. HALLER and H. HELWIG (2. anorg. Chem. 1923,130 209-228).- The " dichloride," Mo,Cl is obtained in good yield by heating molyb- denum powder in carbonyl chloride a t 610-620" ; approximately 90% of the metal is obtained as dichloride in one operation.Other organic chlorine compounds may be used but the reaction is less smooth. The powdered mass on treatment with ether containing 5% of ethyl alcohol in an extraction apparatus and subsequent evaporation in a vacuum yields the compound Mo,CI,,EtOH as a light yellow powder stable in air ; the alcohol cannot be removed by heating in an inert atmosphere without decomposition. Addition of alcoholic silver nitrate t o the alcoholic solution precipitates silver chloride ; the mother-liquor deposits the compound Mo3C1,(N03),,EtOH on keeping but on immediate treatment of the fdtrate with ether the nitrate Mo3CI4(NO3) is obtained. The compound Mo3C1,(NH3),,2EtOH is obtained by precipitating a saturated solution of the dichloride with ether and passing in dry ammonia; it is a bright yellow powder decomposing a t 100".The compound H[Mo3C17,H,0],3H,0 already described (A. 1922 ii !jog) when heated in a stream of hydrogen chloride yields the compound Mo3Cl,,H,0 ; both these substances may be hydrolysed in stages yielding compounds in which the ratio Mo C1 is 3 6 3 5 and 3 4 ; the complex [Mo3C14] is stable and not ionisable. The chloro-acid H[Mo3C17,H,0],3H,0 when treated with pyridine and excess of hydrogen chloride in alcoholic solution gives pyridinium salts containing alcohols which separate as yellow crystalline solids ; the compounds C,H5N,H[Mo,C1,,H,0],MeOH and C5H5N,H[Mo3C1,,H,0],EtOH are described the former being dimorphous.If the reaction is carried out in amyl alcohol theINORGANIC CHEMISTRY. ii. 193 alcohol-free pyridinium salt C5H5N,H[Mo3Cl7,H2O] is obtained ; recrystallisation from the lower alcohols gives the alcoholate salts. Other alcohols give analogous products. From the mother- liquors and by varying the reaction conditions a series of corn- pounds in the anion of which the ratio Mo C1 varies from 3 7 to 3 9 may be obtained but these were not more closely examined. In presence of a large excess of hydrogen chloride the compounds ( C5H5N),H,Mo,C1 3MeOH ( C5H5N),H3MoCla,3EtOH and (C5H5N),H3MoC19,C5Hl10H were obtauied from which on heating the compound C5H5N,H(Mo,C17) is formed.By treating solutions of the chloro-acid in the minimum quantity of hydrochloric acid with concentrated solutions of metallic chlorides the salts K[Mo3C1,],3H,0 an'd NH4[Mo3C17J,1-5H20 and their anhydrous compounds were obtained ; the pyridine compound C,H5N,H[Mo3C17] readily forms alcohol-additive products and simdar compounds are obtained with aniline carbamide ethylene- diamine and triaminopropane. The hydroxide Mo,Cl,(OH) described by Blomstrand (J. pr. Chem. 1859 77 96) on treatment with hydrobromic acid yields an mid H[Mo,C14Br3,H,0],3H,0 which readily passes into the compound Mo,C1,13r2,H20. With a large excess of hydrobromic acid the compound H2[Mo3C14Br4,4H,0],2H,0 is obtained. Salts of these bromine-substituted chloro-acids are obtained by the methods described above ; the pyridine compound C5H5N,HMo,C14Br3,2C5H5N,3Et OH and its derivative C5H5N,H[Mo,Cl,Br3] and the potassium salt K,Mol2Ch5Br3,1 6H,O and the corresponding anhydrous salt are described.Blomstrand's crystalline hydroxide (Zoc. cit.) has now been shown to have the formula [Mo,C1,,3H,0](OH),,6H20; a t 35" it loses 5H,O a t 67-lOO" 6H,O and at 200-300" 7H20 ; the final product which resists further dehydration has the composition and is dark brown; it dissolves completely without residue in aqueous ammonia. An analogous series of amorphous hydroxides is described which yield the same end-product on dehydration. Modified hydrolysis of the chloro-acid yields hydrated products of the compound Mo,Cl,*OK ; hydrolysis with very little water yields a chloride [Mo3C1,,2H,O]Cl2,2H,O which on dehydration yields [Mo3C14,2H,0]CI in which compound two chlorine atoms are ionisable and replaceable by other anions.Determinations were made of the molecular weight of the pyridine salt C5H,N,H[Mo,C17,H20] from the depression of the freezing point of nitrobenzene; the average of three readings gave 717 the Compounds of Tervalent Molybdenum and Tungsten. A. ROSENHEIM and T. H. LI (Ber. 1923 56 [B] 2228-2230).- Double chlorides of tervalent molybdenum were obtained by electrolysing strongly acid solutions of molybdic acid using a mercury cathode adding suitable chlorides t o the deep purplish-red [Mo,CI4,2H 01 0 formula requiring 634. s. I. L.ii. 194 ABSTRACYTS OF CHEMICAL PAPERS. solution and evaporating the double chlorides of the general formula R,MoCl separating in red needles.The followin chlorides (C5H6N),MoC]1,,2H,O ; and the ethylenediammonium salt (C,H,N2),(MoC1,),,7H,0. The crystalline aniline salt was not obtamed rn a pure state. The following double bromides were obtained as deep blcod-red crystals by similarly electrolysing a solution of molybdic acid in hydrobromic acid (NH4),MoBr5,2H20 ; ( C5H6N),M0Br6,4H,O. Two purplish-red double fluorides prepared analogously had the following formulae (NH4),Mo,F,,2H,0 ; K,Mo2F,,2H,O. Compounds of tervalent tungsten are under investigation ; the following compounds are noted KWF4,H,0 ; (NH,)WF,,H,O. F. A. M. The Precipitation of Tungstic Acid. J. A. M. VAN LIEMPT (Rec. trav. chirn. 1924 43 30-35; cf. A. 1923 ii 867).-A com- parison was made of the rate of precipitation of tungstic acid from sodium tungstate solution by different acids using an optical method similar to that used by Lottermoser for a similar investi- gation (A.1915 ii 267). The increasing opacity of the solution is measured by passing a beam of light from a constant source through the solution under examination to a light-sensitive cell which is connected to a galvanometer. The results obtained are influenced considerably by the method of mixing the two solutions. When the tungstate solution is added drop by drop to the acid the rate of precipitation appears to depend on the degree of dissociation of the acid. Thus 0.25N-nitric perchloric and hydrochloric acids showed very nearly identical precipitation curves whilst 0-25N- sulphuric acid showed a much more extended period of induction before precipitation started.This last curve however was very similar to that given by O.18N-hydrochloric acid the dissociation of which is about equal to that of O-25N-sulphuric acid. When the acid was added to the tungstab solution the slower the addition of the acid (up to the limit of twelve minutes) the slower was the subsequent precipitation of the tungstic acid. The form of the curves indicates that the action is autocatalytic but this deduction is probably incorrect since the precipitate is not simply the com- pound WO but contains water and processes of hydration and dehydration probably occur during the precipitation. were isolated and analysed K,M0Cl6 ; (NH4)3M~C f ,,2*5H,O ; E.€3. R. Simultaneous Presence of Celtium and the Rare Earth [Yttrium] Metals in some Zirconium Minerals. E. URBAIN and G. URBAIN (Compt. rend. 1924 178 265-266).-The atomic weights of the zirconium and of the rare earth fractions of a number of zirconium minerals have been determined. The zirconium fractions of the different minerals (purified through the oxychloride ZrOC12,8H,0 which does not came removal of celtium) gave atomic weights varying from 90.3 to 95.0 as determined from the conversion of sulphate into oxide. The higher atomic weights corresponding with a higher celtium content were obtained with1NOkGANlC CHEMISTRY. ii. 195 minerals of higher rare earth (yttrium metals atomic weight 96-93) content . E. E. T. A. 1‘. DUMANSKI (J. Russ.Phys. Chem. SOC. 1924 64 703-733).- Vanadium pentoxide sols were prepared by the method of Biltz (A. 1904 ii 324) and their properties investigated. It is shown that the red sols so obtained contain negatively charged particles which gradually become smaller as the solution ages. The electrical conductivity a t first gradually falls and reaches a certain minimum value after some days whilst the viscosity of the solution increases ; there is however no constant relationship between these two properties. Dilution causes a gradual rise in the conductivity which Gnally assumes a constant value; this is doubtless due to the hydrolysis of the colloidal particles with the production of ions. The red colour of the sol h a l l y changes to yellow and yellow solutions are found to contain no colloidal particles when viewed in the ultramicroscope.The effect of temperature on the conductivity has also been studied. A rise of temperature is shown to produce an increase in conductivity; on cooling the conductivity of the solution does not immediately regain its former value. The sol shows the usual behaviour towards electrolytes; thus it is readily coagulated by the addition of barium chloride a small quantity of barium being adsorbed a t the same time. On keeping for some hours in contact with excess of reagent the amount of barium adsorbed by the colloid increases and the composition of the solid coagulum approximates to that of barium hexavanadate BaH,V OI7 the conductivity of the solution rising at the same time. h o r n the amount of barium adsorbed in the first stage an approximate formula for the particles of the sol is deduced namely [H2V,01,,(V20,),]”+2H’ giving a molecular weight of 1678 ; cryoscopic measurements point to a similar formula.The equivalenb of thc particle of the sol is found to be approximately 1000. The phenomenon of double refraction observed in vanadium pentoxide sols (Freundlich A. 1916 ii 442) is shown to be absent when the sol is quite clear but is well marked when it becomes turbid or “ gels ” ; gels therefore consist of particles arranged or oriented in a deiinite manner. Finally the effect of reducing agents wm studied and it was found that hydrazine produced an unstable green sol ; the reduction proceeds to the V,O stage or to V,O if an excess of reducing agent be present.Free Energy o€ Antimony Trioxide and the Reduction Potential of Antimony. R. SCHUHMANN ( J . Amer. C‘hern. Soc. 1924 46 52-58).-(See ii 152). The System Antimonious Oxide-Hydrochloric Acid-Water. C. LEA and J. K. WOOD ( J . Chem. Em. 1924 125 137-14#3).- The behaviour of hydrated antimonious oxide towards solutions of hydrochloric acid of varying concentrations and also the action of water on antimonious chloride are described. The transition point a t which the compound SbOCl undergoes conversion into The Properties of Colloidal Vanadium Pentoxide. G. A. R. K.ii. 196 ABSTRACTS OF CHEMICAL PAPERS. Sb,O,Cl occurs when the total concentration of chlorine is approxi- mately 8N and this latter oxychloride only ceases to exist when the total concentration of chlorine is reduced fo 0-1N.These results are not in agreement with those of other observers; thus according to Le Chatelier (A. 1885 630) the transit,ion point of SbOCl to Sb40,Cl occurs a t a chlorine concentration of 2.12N. The amorphous precipitate produced on the addition of water to antimonious chloride is not SbOC1 but has a variable composition and is probably an adsorption product formed by the adsorption of hydrochloric acid by a highly hydrated form of antimonious oxide. The adsorbed acid slowly interacts with the hydrated oxide forming that oxychloride which most nearly approximates to its own composition. One of these metastable oxychlorides is Cooke's oxychloride Sb,O,( OH),Cl whilst from more concentrated solutions a crystalline compound is deposited to which the formula Sh,O,( OH),CI3 is assigned.This crystalline compound can apparently change into either SbOCl or Sb,O,Cl according to the conditions ; Sb,O,( OH),Cl,+HCl= 4SbOC1+ 2H20 ; Sb,O,( OH),Cl = Sb40,C12+H20 +HCl. J. B. F. A New Compound of and Test for Palladium. V. N. IVANOV ( J . Russ. Phys. Chem. SOC. 1924 54 701-702).-0n adding a solution of sodium nitroprusside to a solution of palladium chloride the mixture sets to a stiff jelly; on keeping the jelly shrinks and the solid residue after washing and drying consists of palladium nitroyurusside PdFe(CN)5N0 in the form of a greenish-grey powder. The formation of the jelly can serve as a test for palladium but it is less sensitive than the glyoxime test. A New Method for the Preparation of Soluble Ruthenium Compounds from Ruthenium. F. KRAUSS and H. KUKENTEAL (2. anorg. Chem. 1924 132 315-317).-Four g. of ruthenium are fused in a silver crucible with 30 g. of potassium hydroxide and 4 g. of potassium nitrate and heated until the melt is quite fluid. On cooling the melt is extracted with 200 C.C. of water and the precipitate collected. The precipitate is then stirred with 100 C.C. of alcohol a t 40". The residue recovered either by decantation or filtration is washed with water and dissolved in dilute hydro- chloric acid. The resulting solution is evaporated to dryness on a water-bath and the residue treated with water. To the dark brown solution potassium hydroxide or alkali carbonate is cautiously added and the ruthenium hydroxide precipitated the liquid being still slightly acid. The precipitate is collected washed with water until the filtrate remains clear drained a t a suction pump and dried. The resulting black compound is free from alkali and chlorine and is the starting material for subsequent preparations Alkali Ruthenates. F. KRAUSS (2. anorg. Chem. 1924,132,301- 314) .-Rubidium and cssium ruthenates have been prepared G. A. R. K. J. B. F.AXALYTICAL CHEMISTBY. ii. 197 having the composition Rb,RuO Rb,Ru04,H,0 Ck$uO and Cs,Ru04,H,0 respectively ; they resemble in properties the cmre- sponding potassium salts. Solutions of potassium and rubidium ruthenates on treatment with concentrated ammonia solution yield a salt having the composition (NH,),RuQ,. In the case of the cesium salt a product of less definite composition is obtained. From the properties of this compound it is not a true ammonium salt but exists as ‘’ dioxy-dihydroxy-diammine-ruthenium ” hav- ing the formula Ru02(OH),,2NH,. When treated with hydro- chloric acid it yields * ’ diox y-diaquo -diammine -rut henon-chloride ,” Ru20,2NH3,2H,0,Cl which is reconverted into the dihydroxy- compound on treatment with ammonia solution. On treat- ing a solution of potassium ruthenate with a 10% solution of trimethylamine a precipitate is obtained having a composition corresponding with trimethylammonium ruthenate (NHMe3)eRu0,. J. B. F.

ISSN:0368-1769    

DOI:10.1039/CA9242605164

出版商:RSC    

年代:1924

数据来源: RSC

摘要:

AXALYTICAL CHEMISTBY. Analy tidal Chemistry. ii. 197 Electrometric Titration. Its Methods and Application to certain Metallurgical Analyses. G. A. SHIRES (J. Chem. Met. Min. &c. S. Afr. 1923 24 129-145).-The paper contains a brief account of the theory of electrometric titrations a descrip- tion of an inexpensive apparatus for carrying out the work and examples showing the application of the method to the deter- mination of iron in ores and of chromium vanadium and manganese in steels. For oxidimetric titrafions the titration cell was pro- vided with a platinum wire electrode connected through a moving coil mirror galvanometer to one end of the bridge wire and a normal calomel electrode connected to the sliding contact of the wire. A 2-volt accumulator was connected across the ends of the potentiometer.The stirrer consisted of a bent glass rod sup- ported in a greased glass tube and driven through a cork pulley and a belt of thin cord by means of a small motor a t 200 revolu- tions per minute. In using the apparatus the galvanometer is kept steady by moving the slider with each small change in E.M.F. until a small addition of solution produces it sudden big deflection in the spot of light thus indicating the end-point. In titrating ferrous chloride solutions with dichromate the hquid hould be strongly acid with hydrochloric acid; otherwise a definite end- point is not obtained. Ferric chloride solutions inay be titrated with stannous chloride with great accuracy by the use of the above a paratus as the end-point is marked by a chaa e i n the 8' $the E.N.F.and the spot of light is violently ieilected. [% B. 1924 220.1 A. R. P. VOL. CXXVI. ii. 8ii. 198 ABSTRACTS OF CHEMICAL PAPERS. Substitution of Borchers' Metal for Platinum in Electro- analysis. M'. BLEESEN (2. anal. Chem. 1923 63 209-228).- Borchers' metal is an alloy containing 65 to 60% of nickel 30 to 35% of chromium 2 to 5% of molybdenum and 0.2 to 1% of silver. It becomes passive after use as an anode with low-voltage currents but for use as a cathode it should preferably be rendered passive by treatment with strong nitric acid. The most suitable form of electrode was found to be a sand-blasted gauze or spiral. With cathodes of this material smooth adherent deposits of copper from nitric acid and ammonia solutions of nickel from ammonia solutions of antimony from sodium sulphide solutions and of tin from oxalate solutions were obtained.The deposits are readily removed by suitable treatment with acids in the cold without any serious loss in weight although several mg. may be lost if hot acids are used. Borchers' metal cannot be used as anodes and hence is not suitable for the deposition of lead or manganese peroxides. A. R. P. R. V. STANFORD (Biochem. J . 1923 17 839-843).-Details are given of various instrumental improvements in the dilution colorimeter already described by the author (A. 1913 ii 856) in which the concentration and not the thickness of the layer examined is altered. The improvements make for greater convenience and ease of mani- Improvements in Colorimetry.pulation. e J. P. Magnesium Perchlorate Trihydrate as a Drying Agent for Steel and Organic Combustion Analysis. G. F. SMITH M. BROWN and J. F. Ross (Ind. Eng. Chenz. 1924 16 20-22; cf. Willard and Smith A. 1922 ii 850).-Magnesium perchlorate may be prepared by adding magnesium oxide or carbonate to perchloric acid of 68-70% concentration. The solution is boiled until crystallisation begins cooled and the crystalline mass centrifuged. The hexahydrate crystals so obtained when heated a t 138-140" for fifteen to twenty hours in a good vacuum yield the trihydrate in which state it retains its original crystalline form and is very porous. No decomposition occurs in dehydration under these conditions. At 250" the anhydrous salt can be prepared but in this case slight traces of impurity cause decomposition and the trihydrate is therefore to be preferred as a drying agent. It is a trustworthy substitute for phosphoric oxide in the analysis of steel by combustion and it may equally well replace calcium chloride for organic combustions. It is easily handled offers little resistance to the passage of gas and when exhausted can be readily The Preservation of Standard Solutions of Oxalic Acid.S. ISHIMARU (J. Chem. Xoc. Japan 1922 43 767-772).-Decom- position of 0.1N-solntions of oxalic acid when preserved in the dark is negligible after one hundred and forty-eight days and no greater when kept in transparent bottles. In the latter case regenerated. c. I.ANALYTICAL CHEMISTRY. ii. 199 contrary to Riegler (A. 1896 ii 676) the addition of sulphuric acid increased the amount of decomposition.Quantitative and Qualitative Analysis with the Aid of Rontgen Rays. D. COSTER (Chem. Weekblad 1924 21 59-42). -The X-ray spectrum has great advantages for identification purposes since it is purely a property of the individual atom however combined or admixed the elcment examined; it is simple and characteristic and can detect 1% of a specific element in 1 mg. of substance. It can unfortunately only be employed for solid elements heavier than sodium. The use of X-rays in quantitative determinations of hafnium by addition of tantalum and comparison of the intensity of the tantalum and hafnium lines in the spectra obtained is described. Additions of tin and antimony were also made in some determin- ations.The opinion is expressed that by careful elimination of present sources of error very accurate determinations of concen- trations of specific elements may be made by this method. K. K. s. I. L. The Colorimetric Determination of the Hydrogen Exponent of Soil. I. M. KOLTHOFF (Chem. Weekblad 1923,20,675-677).- The determination of the hydrogen-ion concentration of soils by the electrometric method does not give accurate results owing to the great resistance. Colorimetric methods are also unsatisfactory because of the dBculty of obtaining clear colourless solutions. The colorimetric method gives much better results if the mixture of soil and water is dialysed using a parchment membrane for Application of the Mercury Electrode in the Electrometric Titration of Halides Cyanides Sulphides and Thiosulphates.I. M. KOLTHOFF and E. J. A. H. VERZVL (Re. traz;. china. 1923 42 11 11) .-Several titration curves omitted from the previous paper on this subject (A 1923 ii 873) are now published. twenty-four hours. s. I. L. 5. F. S. Formation of Periodate. Bromometry as a Substitute for Iodometry. R. LBNa.-(See ii 165.) W. MANCHOT and 3’. OBERHAUSER (2. anorg. Chem. 1923 130 161-167).- Bromine can be used in place of iodine in many volumetric deter- minations. It is best used in solution in 20-22% hydrochloric acid such solutions being stable for many weeks but care must be taken to avoid volatilisation losses. Arsenious acid is rapidly and accurately titrated using indigo-carmine as indicator and the method has the advantage over the iodine titration that it can be carried out in acid solution hydrobromic acid not acting as a reducing agent in solutions containing less than 26% of free hydrochloric acid.[With E. B ~ ~ ~ ~ . ] - B r o m i n e oxidises thiosulphate quantitatively forming sulphuric acid; 1 mol. of thiosulphate requires 8 atoms of bromine. 8-2ii. 200 ABSTRALOTS OF CHEMICAL PAPERS. Bleaching powder may be analysed by adding it to an acidifled bromide solution and titration of the liberated bromine with arsenious acid ; hypochlorite solutions and chlorine water may be analysed in the same manner. Manganese dioxide lead dioxide and potassium dichromate may be determined by heating with hydrochloric acid the liberated chlorine being led either into bromide solution with subsequent titration of the liberated bromine by means of arsenious acid or into excess of arsenite solution the excess being titrated with bromine solut,ion.Permanganate may be added directly to an acidified bromide solution. s. I. L. The Bromometric Determination of Ozone. W. MANCHOT and F. OBERHAUSER (2. anorg. Chem. 1923 130 168-172).-The work of Treadwell (L4. 1906 ii 123) has been repeated and his conclusion that bromides may be used in place of iodides either in acid or in neutral solution confirmed ; under suitable conditions however the reaction may be carried out much more quickly. The gas is introduced into a nitrometer and shaken with 10-15 C.C. of 0-2N-potassium bromide in acid solution ; if hydrochloric acid is used the solution must be not weaker than N/1 with respect to this whereas if sulphuric acid is used it must be not less than 2N.The mixture is cooled and shaken then washed into standard arsenite solution and the excess of the latter determined with a standard solution of bromine in hydrochloric acid (see preceding abstract). With less acid solutions the results are too low. No hydrogen peroxide is formed with concentrations of ozone of less than 10%. s. I. L. The reaction is complete in ten minutes. Determination of Dissolved Air in Small Quantities of Water. H. G . BECKER and W. E. ABBOTT ( J . Soc. Chem. Ind. 1923 42 484486~).-The method depends on the fact that when potassium hydroxide is dissolved in water the dissolved air is expelled and may be collected and measured. The apparatus used consists of a graduated measuring tube 2.5 mm.in diameter provided at the top with a three-way mercury-sealed tap and having a t its lower end a bulb of about 30 C.C. capacity; this bulb is connected with a larger cylindrical bulb provided with a side tube and a tube a t its lower end connected with a mercury reservoir. A piece of solid potassium hydroxide is introduced through the side tube; this is securely closed with 8 rubber stopper the whole apparatus is filled with mercury and the desired volume of the water (e.g. 50 c.c.) is admitted through the tap. The mercury level is then lowered so that the water comes into contact with the potassium hydroxide and when evolution of gas ceases the volume of the bubble is beasured in the graduated tube. The gas is then passed over into a small absorption 'vessel containing alkaline pyrogallol solution and the residual volume of nitrogen is measured subsequently.A number of other substances such as ammonium sulphate ammonium chloride etc. may be used in place of the potassium hydroxide. w. P. s.ANALYTICAL CHELMSTRY. ii. 201 Volumetric Determination of Dithionic Acid in presence of Sulphurous and Thiosulphuric Acids by Oxidation with Bromine in the Nascent State. C. MAYR and I. SZENTPALY- PEYFUS~ (2. anorq. Chem. 1924,131,203-208) .-Whilst dithionic acid is not affected by bromine in the cold it is completely oxidised to sulphuric acid by bromates in boiling hydrochloric acid solution. A measured quantity of bromate is employed the excess of liberated bromine being distilled over into iodide solution and determined in the usual manner.The method allows of a high order of accuracy. If sulphurous and thiosulphuric acids be present in addition to the oxidation at the boiling point by bromine a second oxidation with bromine in the cold is carried out (this oxidises the latter two acids only forming sulphuric acid in each case) and an iodine oxidation in the cold which oxidises sulphurous to sulphuric acid and thiosulphuric to tetrathionic acid. From the quantities of halogen required in each of the three cases the quantities of dithionic sulphurous and thiosulphuric acids present may be calculated. s. I. L. Method for the Rapid and Quantitative Removal of Ammonia from Solutions Especially Applicable to the Micro- quantitative Determination of Nitrogen and Urea in Products of Living Origin.R. V. STANFORD (Biochem. J . 1923,17 847- 850).-An apparatus is described for the quantitative removal of ammonia from small volumes of fluid and its determination by vacuum distillst'ion into sulphuric acid followed by nesslerisation. Nesslerisation and the Avoidance of Turbidity in Nesslerised Solutions. R. V. STANFORD (Biochem. J. 1923 17 844-846).-J?olin and Youngburg's method of '. direct nessleri- sation " (A. 1919 ii 304) does not give clear solutions. Turbidity may be avoided in pure solutions of ammonium salts containing not more than 0.03 mg. of nitrogen per c.c. if the Nessler reagent be added drop by drop with constant shaking to the ammonia solution. Excess of alkali or of Nessler reagent and the presence of urease produce turbidity. J.P. The Determination of Hydroxylamine with Permangan ate. A. KURTENACKER and R. NEUSSER (2. nnorg. Chem. 1923 130 199-204) .-A comprehensive survey of the attempts desck5bed in the literature to carry out the oxidation quantitatively in acid neutral and alkaline solution shows that a succeseful method has not been evolved. I n acid solution the results vary with the concentrations of hydroxylamin; acid and permanganate and with time temperature etc. and under the most favourable con- ditions not more than 2.7 atomic proportions of oxygen are used. In neutral solutions all these factors as well as the nature of the acid used for neutralisation affect the results. Similar variations were obtained in alkaline solution and in no case could conditions be found which would allow of quantitative determination.J. P. s. I. L.ii. 202 ABSTRACTS OB CHEMICAL PAPERS. Determination of Phosphine in Acetylene. T. E. PERKS (Analyst 1924 49 32-33) .-Lunge and Cedercreutx’s method for determining phosphine in acetylene (A. 1898 ii 54) gives inaccurate results but is rendered satisfactory when certain modifications are introduced. [CE. B. 1924 161.1 T. H. P. The Separation of Arsenic from Antimony and Tin. W. STRECKER (2. anal. Chem. 1923 63 252).-Sulphur dioxide or a sulphite is a more suitable reducing agent for quinquevalent arsenic than sodium thiosulphate as recommended by Jairvjnen (A. 1923 ii 254 ; cf. Moser ibid. 758). Thionyl chloride is more satisfactory still as it provides not only the necessary reducing agent but also the hydrochloric acid required to prevent the hydrolysis of the antimony chloride. A.R. P. Reduction of Arsenic Acid by Sulphurous Acid in Presence of Vanadic Acid. V. AUGER and (MLLE.) L. ODINOT (Compt. rend. 1924 178 213-214).-1f sulphur dioxide is passed into a boiling solution of arsenic and vanadic acids in 10% aqueous sulphuric acid very little reduction of arsenic occurs the result being scarcely affected by the presence or absence of vanadium (cf. Trautmann A. 1911 ii 544). If as recommended by Gooch (“ Methods in Chem. Anal.,” 1912 350) the solution saturated in the cold with sulphur dioxide is heated in SL closed vessel for an hour on the water-bath complete reduction to arsenious acid and vanadyl salt takes place. If a trace of potassium iodide is added to a warm solution contain- ing arsenic vanadic and sulphuric acids reduction by sulphur dioxide may be effected in fiftecn minutes in an open vessel.After removal of sulphur dioxide by boiling and of iodide by adding silver nitrate the usual titration with permanganate may be carried out. E. E. T. The Iodometric Determination of Carbon Disulphide. E. ANDR~ (Bull. SOC. chim. 1923 [ivj 33 1678-1681).-The method in general use is not trustworthy as the reaction on which it is based is not quantitative. A series of experiments showed that the quantity of carbon disulphide which corresponds with 1 C.C. of standard iodine solution varies with the proportion of the former with respect to the amount of alcoholic potash present. Further the time occupied in the attainment of equilibrium may introduce another source of error.The Perchlorate Method for the Determination of Potassium in Soils Fertilisers etc. H. J. PAGE (J. Agric. Sci. 1924 14 133-138) .-Samples of perchlofie acid were found to contain considerable amounts of chloric acid and as a consequence figures obtained by their use for the determination of potassium were very untrustworthy. It is also indicated that in Neubauer’s method for the analysis of soil extracts the addition of 0.1 g . of calcium carbonate (instead of the usual 0.5 g.) to soils deficient in carbonates suffices. Smaller amounts of perchloric acid may then be used in the subsequent analysis. H. J. E. A. G. P.ANALYTICAL CHEMISTRY. ii. 203 A New Method for the Volumetric Determination of Barium.B. N. ANGELESCTJ (Bul. Soc. Chirn. Romdnia 1923 5 72-74).- A volumetric method for the determination of barium is described based on the precipitation of barium pyroborate by the addition of a solution of sodium pyroborate to the solution of a barium salt and the subsequent hydrolysis in dilute solutions first to the metaborate and then to the hydroxide BaB,0,+3H,0=Ba(B02)2 +2H3B0 ; Ba(B0,)~+4H,0=Ba(OH),+2H3B03. The barium solution (containing 0.20-0.25 g . of barium) is treated with 40 C.C. of alcohol and 40 C.C. of O.1N-sodium pyroborate solution. A white precipitate is formed and after fifteen minutes on the water- bath precipitation of the barium is complete. After cooling the solution is diluted to 100 C.C. and 25 C.C. are titrated with 0-1N- hydrochloric acid to determine the excess of pyroborate dimethyl- aminoazobenzene being used as an indicator.The maximum error recorded is 0.46%. The Accuracy of the Potentiometric Titration of Zinc with Ferrocyanide. I. M. KOLTHOFF and E. J. A. H. VERZYL (2. anorg. Chem. 1924 132 318-320) .-The accuracy of the method was tested with specially purified materials. Zinc sulphate solution containing about 1 g . of the salt in 100 C.C. of distilled water was titrated with 0.025M-potassium ferrocyanide solution containing O.lyo of ferricyanide. In neutral solution at 15" and 6 5 O respec- tively approximately 1% too little reagent was required. If 3 C.C. of O.25N-sulphuric acid were added the error at 15" was reduced to 0.7y0 whilst when 3 C.C.of 0~25N-sulphuric acid and 3 g. of potassium sulphate were added the error was about 06y0. For all other propor- tions the amount of reagent required was too small. (Cf. A. 1922 ii 580.) J. €3. F. J. W. B. Schoenbein's Test Applied to the Micro-investigation of Copper-ions. H. IMBERT R. IMBERT and P. %GRAIN (Bull. Sm. chim. 1924 riv] 35 60-64).-The Schoenbein test for cyanides has been adapted to the detection and determination of copper in aqueous solution at concentrations as loy as five parts in ten million. The copper solution is treated with a few drops of freshly prepared guaiacum resin solution and dilute potass. ium cyanide solution (0.15yo for neutral and 1.5% for acid copper solutions) added when a blue coloration appears. If excess of potassium cyanide is added the brown coloration produced by the action of alkali on guaiacum masks the blue colour which similarly does not appear in presence of an excess of acid.The new test for copper is about twenty times as sensitive as that afforded by ammonia ferrocyanide or iodide and considerably more sensitive than the electrolytic test. Electrolysis of copper solutions followed by testing the cathode deposit (molybdic reagent) barely allows of the detection of copper a t concentrations of one part in a million. Distilled water (from a copper still) after being concentrated from 6 1. to 250 c.c. gave a positive test for copper by the newii. 204 ABSTBACTS OF CHEMICAL PAPERS. method which has also been applied to the detection of co per in preserved vegetables and in blood. Micro-gravimetric Determination of Copper as Cop er Benzoinoxime.R. STREBINGER (Mikrochemie 1923 1 72-74). -The copper solution containing a few mg. of copper is rendered slightly ammoniacal heated and treated with 1 yo alcoholic benzoin- o d e solution. The mixture is heated in a boiling water-bath for twenty minutes the precipitate then collected on a small a t e r washed with very dilute ammonia and alcohol dried and weighed. Use of Hypophosphorous Acid in Gravimetric Analysis. 11. Determination of Mercury Gold and Palladium and a Method of Separating them from Other Metals. L. MOSER and M. NIESSNER (2. anal. Chern. 1923 63 240-252):-For the determination of mercury the solution of chloride or sulphate free from nitric acid is heated on the water-bath with an excess of O.5.N-hypophosphorous acid; 8 to 10 C.C.of concentrated hydro- chloric acid are added and heating is continued until the precipitated mercury collects into a globule. This is collected on a Gooch crucible washed free from acid with warm water then with alcohol and finally witlh ether and dried a t 30". The presence of lead zinc cadmium or iron in the solution does not affect the results. For the determination of gold either alone or in the presence of platinum the faintly acid solution of the chloride is treated with 2 g . of sodium chloride and heated to boiling. A 0.25M-solution of hypophosphorous acid is added together with a little filter-pulp and heating is continued until the precipitate coagulates. The gold is collected on a hardened paper washed free from chlorine with 1% acetic acid ignited and weighed.Any platinum in the filtrate may be separated by boiling with sodium formate. Palladium may be determined in a similar may to that employed for gold except that no filter pulp is necessary as the precipitated metal readily coagulates after boiling for a few minutes. If platinum is present in the solution the precipitate will be seriously contamin- ated with this element. Bismuth and copper are incompletely precipitated as metal by hypophosphorous acid. A. R. P. The Determination of Manganese. I. The Bismuthate Method. T. R. CUNNINGHAM and R. W. COLTMAN (Ind. Eng. Chem. 1924 16 58-64).-1t has been believed that the accurate determination of manganese by oxidation with sodium bismuthate was impracticable if the weight of manganese present exceeded 50 mg.As this limitation rendered the method unsuitable for the analysis of the metal or of alloys rich in manganese the reactions involved in the method were examined in detail. It was found that a 60% excess of sodium bismuthate over the theoretical quantity was necessary to render oxidation complete. the quantity recommended being 2.6 g. of 80% sodium bismuthate for each 100 mg. of manganese present. A further possible s o m e of error is the liability of the permanganic acid formed to decompose before it can be determined. An investigation of the stability of per- E. E. 5. The precipitate contains 22.02% of copper. w. P. s.ANALYTICAL CHEMISTRY. ii. 206 manganic acid in different concentrations in contact with 11- 22% nitric acid 'solution showed that the rate of decomposition fell with increasing dilution to a negligible amout a t 0.05N.If nitric acid mom dilute than 11% is used oxidation may be incom- plete. A standard time of oxidation of one minute is recorn- mended. With such precautions accurate and concordant analyses wing up to 500 mg. of manganese were obtained Examples with working details of the application of the method to metallic manganese manganese ore and ferro-manganese are given. C. I. Simplified Qualitative Analysis of the Ammonium Sulphide Group. E. RUPP (Ber. Deuts. phamn. Ges. 1923 33 258-262). -The filtrate from the hydrogen sulphide tpeatment is boiled oxidised with nitric acid and treated with an excess of ammonia. The precipitate is collected and washed; a portion is tested for iron by the ferrocyanide test a second portion is fused with sodium nitrate and carbonate on the end of a magnesia rod when the presence of chromium is indicated by the production of a yellow stain and the remainder of the precipitate is boiled with sodium hydroxide and hydrogen peroxide the insoluble material filtered off and the solution acidified then rendered ammoniacal when any gelatinous precipitate indicates aluminium.The atrate from the original treatment with ammonia is treated with ammonium sulphide and the washed precipitate is stirred with 5% hydrochloric acid. The insoluble material is collected and dissolved in dilute aqua regia. The solution is divided into two parts one of which is treated with an excess of potassium nitrite which gives a charac- teristic yellow precipitate of the double nitrite if cobalt is present.A drop is taken from the second portion after addition of ammonia in slight excess and placed on a piece of filter-pBper previously soaked in a 1 yo alcoholic dimethylglyoxime solution ; a pink spot proves the presence of nickel. Zinc i~ tested for in the solution of the sulphidcs soluble in acid by treatment with sodium acetate and hydrogen sulphide and manganese in the same solution by means of sodium hydroxide and by the fusion test. I f phosphoric acid is present in the original sample it is removed by evaporation with nitric acid and metastannic acid; oxalic acid is removed by evaporation and ignition of the residue at 300'. A. R. P. The Nickel-Dimethylglyoxime Reaction by Reflected Illumination. F.KIRSCHNER (Mikrochemie 1923 1 8S).-The dimethylglyoxime reaction will detect the presence of nickel in a drop of solution containing 1 mg. of nickel and 300 mg. of iron per 3 litres of water. When examined by reflected light (dark ground) the crystals have an emerald-green colour the usual red colour being seen only where the crystals are agglomerated. Critical Studies on Methods of Analysis. I. Nickel. L. A. GONGDON and C. H. BELGE (Chem. News 1924 128 67-68). 11. Chromium. L. A. CONGDON and R. K. GURLEY (ibid. 68-70) 111. Manganese. L. A. CONGDON and J. L. NEAL jun. (ibid. 70-71).-I. Comparative tests on a standard nickel solution showed that equally good results were obtained by the dimethyl- W. P.S.ii. 206 ABSTRACTS OF CHEMICAL PAPERS. glyoxime gravimetric method and the cyanide volumetric method but that Carnot’s method (A. 1918 ii 138) gave results that were appreciably low. 11. The iodometric determination of chromium as chromate and the gravimetric determination as barinm chromate in acetic acid solution appear to be the only methods in common use that yield trustworthy results. Precipitation of the chromate with mercurous nitrate gives low figures as a small part of the chromate is reduced to chromic salt and escapes precipitation. The chromic phosphate method gives low and discordant results. Addition of an excess of ferrous salt to a chromate solution followed by titration of the excess with permnnganate yields figures that are about 0.5% too low.111. The determination of manganese in steel may be performed with great accuracy by the persulphate bismuthate or lead peroxide methods; the first-named method was found to be the most rapid. Reactions of Chromium Iron and Aluminium in the Presence of Tartaric Acid and of Glycerol. S. HAKOMORI ( J . Chem. SOC. Japan 1922 43 629-647) .-The hydroxides of chromium iron and aluminium are not precipitated by ammonia in the presence of tartaric acid or of glycerol. The cause has been studied by the aid of the absorption spectrum of their solutions their electrical conductivities the ultramicroscope and by dialysis. In the presence of a large quantity of tartaric acid the cause is attributed to the formation of complex salts from 1 atom of the metal and 1 mol. of the acid.When the quantity of the acid is small it is attributed mainly to the formation of colloidal solutions of the metallic hydroxide. I n the case of chromium the velocity of the formation of the complex salt is not large but is increased by heating i.e. by increasing the concentration of the free acid. In the case of iron the ferric salt is reduced to ferrous salt by long contact with tartaric acid. I n the presence of glycerol the cause is mainly attribnted to the formation of a colloidal solution. A. R. P. K. I<. Qualitative Micro-analysis. 3’. FEIGL (Xikrochemie 1923 1 74-78) .-Characteristic angular crystals are obtained when a drop of a solution of a tervalent antimony salt in dilute hydro- chloric acid is treated with a small quantity of solid pyrogallol; tartaric acid may be added to prevent formation of basic antimony compounds. Copper salts yield a green amorphous precipitate when treated with an alcoholic solution of benzoinoxime.Very small quantities of manganese may be detected by placing a drop of the solution and a drop of potassium hydroxide solution on a microscope slide partly evaporating the mixture and then adding a drop of a solution of benzidine in acetic acid. Blue particles or granules develop in the mixture. The sensitiveness of this test is 1 in 5 x lo7. Benzidine may also be used for the detection of phosphoric acid a blue coloration being obtained when a drop of phosphate solution is treated successively with molybdic acidANALYTICAL CHEMISTRY. ii. 207 solution (in nitric acid) benzidine solution (in acetic acid) and ammonia.w. P. s. Determination of Antimony by Means of Permanganate in Hydrochloric Acid Solution. 0. COLLENBERG and G. BAKEE (2. anal. Chem. 1923 63 229-240).-For the titration of antimony with permanganate the metal is usually obtained in sulphuric acid solution and a certain amount of hydrochloric acid is added before titration. If the latter is a t least 10 C.C. and not more than 19 C.C. of strong acid (d 1.19) per 100 c.c. the end-point is very sharp and the rose colour is persistent for a t least a minute. The method gives results which agrse closely with those obtained gravimetrically if the atomic weight of antimony is taken as 121.8 (cf. Knop this vol. ii 54). Petriccioli and Reuter’s modification in which the solution is diluted until a turbidity forms which is discharged with a few drops of hydrochloric acid (A.1902 ii 177) gives low results. The addition of tartaric acid to prevent hydrolysis leads to high figures unless not more than 1 mol. of the acid is added for each atom of antimony and the concentration of tartaric acid does not exceed O.O5o/d. At the same time a t least 12 C.C. of hydro- chloric acid per 100 C.C. must be present. The Determination of Niobium Taatalwn and Titanium in Minerals. E. W. TODD (Uniz,. Toronto Studiee Geol. Ser. 1923 No. 16 4045).-Metzger and Taylor’s method (A. 1909 ii 702) is the most trustworthy and the easiest to manipulate providing the error introduced by the presence of titanium is corrected. The method with modifications is as follows.Five g. of the mineral are fused with potassium hydrogen sulphate and the mass is disintegrated by boiling with 300 C.C. of water and 5 C.C. of sulphuric acid. Sulphurous acid is added until its odour is plainly perceptible and the precipitate (which contains niobium tantalum silicon and much of the titanium contaminated with tin zirconium iron aluminium calcium uranium and the rare earths) is allowed to settle in the hot solution. After removal of silica with hydrofluoric acid the residue is fused with potassium carbonate the mass extracted with 100 C.C. of hot water filtered and washed. The fusion extraction and filtration are repeated using less potassium carbonate. The solution which contains niobium tantalum and some of the titanium and tin is a c i a e d with sulphuric acid boiled for ten minutes and treated with sulphurous acid.Ammonium sulphide is added tin removed by washing with water and the precipitate ignited in an atmosphere of ammonium carbonate. Its titanium content is determined colorimetrically after fusion with potassium hydrogen sulphate. I n the remainder the reduced niobium and tantalum oxides are titrated with 0-1N-potassium permanganate according to Metzger and Taylor’s method in order to determine the niobium a deduc- tion being made for the reduced titanium. A. R. P. CHEMICAL ABSTRACTS. Micro-electrolytic Determination of Gold. K. FUCHS (Mikro- chernie 1923 1 86-87).-Small quantities of gold (0.25 to 5.0 mg.)ii. 208 ABSTRACTS OF CHEMICAL PAPERS. m‘ay be deposited from a potassium cyanide solution using a current of 3 volts and 3 amperes; the solution is boiled during the electro- lysis which requires ten minutes then cooled rapidly and the current passed through the solution for a further five minutes.The electrodes are removed before the current is interrupted and the cathode ia washed successively with water alcohol and ether and dried over a flame. w. P. s. Test for Palladium. V. N. 1vAwov.-(See ii 196.) The Evaluation of the Purity of Various Organic Products by the Dichromate Method. E. C. GREY (Biochem. J . 1923,17 768-771).-Aliphatic organic substances may be oxidised by a mixture of potassium dichromate and sulphuric acid to carbon dioxide unless methyl groups are present when acetic acid is also formed. By volumetric measurement of the amount of dichromate used in the oxidation the substance oxidised may be quantitatively determined Tables are given whereby the method may be applied to mixtures of various aliphatic alcohols and acids.Succinic acid is exceptional in that it resists oxidation by the method described. Analysis of Naphthalenesulphonic Acids and Naphthalene. W. S. CALCOTT F. L. ENGLISH and P. B. DOWNING (Ind. Eng. Chem. 1924 16 27-3O).-An examination of the oxidation of naphthalenesulphonic acids by means of a solution of vanadic acid in 70% sulphuric acid shows that the conversion to phthalic or sulphophthalic acids proceeds smoothly and quantitatively the position and number of the sulphonic acid radicals having no influence on the rate of oxidation or of course on the oxygen required.In the case of the mono- and di-sulphonic acids the oxidation is quantitative within the accuracy of the methods of determination; in the case of the trisulphonic acids the conversion is 97.5-98.0%. By titrating the vanadium reduced in the reaction with O*lN-permaRganate solution the naphthalene content of a mixture of naphthalenesulphonic acids can therefore be deter- mined and the method may be extended to the determination of the naphthalene content of crude naphthalene. The sulphonic acid content of a similar mixture may be determined by boiling with barium carbonate to remove sulphuric acid filtering and precipitating the barium present in the filtrate as soluble sulphonate with dilute sulphuric acid. A rapid approximate method is also described for determining small proportions of moisture in refhed naphthalene.[Cf. B. 1924 167.1 C. I. The Determination of Reducing Sugars by means of Alkaline Copper Solution. L. MAQUENNE (Bull. Xoc. chim. 1923 [iv] 33 1681-1692).-A criticism of the work of Boutot (‘‘ Contribution h 1’8tude du dosage des sucres reducteurs au moyen des liqueurs cuproalcalines,” Paris 1922) and of Fleury and Boutot (A 1922 ii 879) in which it is stated that the suggested modi- fications of the author’s method (A. 1922 i 920) are unnecessary and have in addition certain disadvantages. J. P. H. J. E.ANALYTIOAL UHE~STBY. ii. 209 Application of the Iodometric Method to the Analysis of Sng.ar Products. C. L. HINTON and T. Mac- (Ana;lyst 1924 49 2-24).-The authors have investigated the effects on the reduction by dextrose lactose sncrose and lsevulose of iodine in alkaline solution of varying the proportions of iodine and alkali used the time and the temperature.Under suitable conditions dextrose and lmtose are oxidised quantitatively to monobasic acids; aucro~e and lsvulose exert slight but under de'finite con- ditions constant reducing powers. A standard method of pro- cedure is suggested and the corresponding iodine values are given for the different sugars. Iodometric Method for the Determination of Nitrogen in Osazones. D. R. NANJI (Biochem. J . 1923,17 761-763).-For the determination of nitrogen in osazones 5-60 mg. are heated on the boiling water-bath with 10 C.C. of N-hydrochloric acid for one hour. The mixture is neutralised with 10 C.C.of N-sodium hydroxide acid%ed with dilute acetic acid and made alkaline with pure sodium hydrogen carbonate. The alkaline solution is added to a known excess of O.02N-iodine and after completion of the reaction with the phenylhydrazine the residual iodine is determined by titration with standard thiosulphate. A table is given from which the amount of nitrogen is read off from the volume of iodine solution used. J. P. A New Method for the Determination of Acetic Acid in Acetic Anhydride. J. H. WALTON and L. L. WITHROW (J. Amer. Chem. Soc. 1923 45 2689-2693).-Measurernents are made of the inhibitory effect of from 1 to 5% of acetic acid on the catalysis by strychnine of the unimolecular decomposition of formic acid in the presence of acetic anhydride (cf. Schierz A.1923 ii 231). The results which are exhibited graphically may be used as the basis of a method for determining acetic acid in acetic anhydride when present in concentrations within the limits mentioned. Very pure acetic anhydride may be obtained from the commercial 97% material by treating it with sodium first for several days a t the ordinary temperature and then for several hours at the boiling point (under reduced pressure) and finally distilling over a mixture of sodium and sodium acetate. Gold-beater's Skin Test for Tannins. P. H. PRICE (Analyst 1924 49 25-29).-Atkinson and Hazleton's test (A. 1922 ii 793) is modified as follows. The gold-beater's skin is successively soaked in 2% hydrochloric acid solution instead of water washed thoroughly to eliminate the acid treated for thirty minutes with 1 C.G.of the solution to be tested washed for fifteen minutes treated with 1 C.C. of 1% ferrous sulphate or chloride solution for fifteen minutes and washed for two minutes. Staining of the skin indicates the presence of tannin. If the liquid is to be tested for phlobaphens the gold-beater's akin treated as above is then left covered with 1 O.C. of 5% hydrochloric acid solut.ion for two minutes and after- wards washed for two minutes. [Cf. B. 1924 189.1 T. H. P. W. S. N. T. H. P.ii. 210 ABSTRACTS OF CHEMICAL PAPERS. A Source of Error in Employing MagentaSulphurous Acid Solution as a Test for Formaldehyde in Ethyl Alcohol. J. L. MAYER ( J . Amer. Pharm. ASSOC. 1923 12 698-700).- Undiluted ethyl alcohol not containing formaldehyde gives a positive colour reaction with magenta -sulphurous acid solution in the presence of sulphuric acid but the reaction is not given with 5-10~o aqueous ethyl alcohol.The test will detect 0.00005 C.C. of 37% formaldehyde in 5 C.C. of solution. The test employed by the author consists in the development of a purple colour a t the surface of contact of sulphuric acid with 1 C.C. of a solution contain- ing lo* C.C. of 37% formaldehyde and five drops of 035% morphine sulphate solution on gentle rotation. Dodsworth and Lyons’s modification of Hepner’s test (using iron sulphuric acid and albumin) and the morphine sulphate test; are both very sensitive detecting 10-6 C.C. of 37% formaldehyde in 1 C.C. of solution; they are easily performed and trustworthy.The Qualitative Tests for Acetone Bodies their Signi- ficance and Value. E. J. BIGWOOD and W. S. LADD (J. Biol. Chm. 1923 58 347-361) .-The sodium nitroprusside colour reaction is given by both pure acetone solutions and solutions of acetoacetic acid as free as possible from acetone (cf. Harding and Ruttan A. 1913 ii 79). Only the acetoacetic acid solution however gives a colour reaction with ferric chloride. The sodium nitroprusside reaction is intensified by the presence of sodium chloride and of other electrolytes present in wine. Neither colour S. SCHMIDT-NIELSEN and A. W. OWE (Videnukap. S k f t e r 1923 No. 15 5-77) .-A systematic investigation of various methods for determining the iodine value of oils and of the conditions govern- ing the reactions involved.Preference is given to the Hub1 method since substitution of halogen does not take place as in the Wijs method; the latter method is however more rapid. The Waller Hanus and Winkler methods yield less trustworthy results. CHEMICAL ABSTRACTS. reaction serves as a quantitative test. E. s. Determination of the Iodine Value [of Fats and Oils]. w. P. s. Determination of Butter Fat and Coconut Fat in Fat Mixtures. G . H. LEOPOLD and W. J. DE Moov (Rec. truv. chim. 1924 43 103-124).-A graphic method is described for deter- mining butter fat and coconut fat in mixtures such as margarine containing in addition only neutral fats. [Cf. B. 1924 222.1 E. H. R. The Quantitative Preparation of Urea from Human Urine. W. 0. MOOR (Biochem. Z. 1923 143 423432).-Urea may be quantitatively recovered as the oxalate from the vacuum-dried residue of urine by extraction with methyl alcohol and the addition of oxalic acid to the extract dissolved in amyl alcohol. The urea may be determined by Kjeldahl’s method either directly on the oxalate or in the solution obtained by treating the oxalate with calcium carbonate. Comparing the results obtained by thisANALYTICAL CHEMISTRY.ii. 211 method with those given by the Morner-Sjoquist method the latter is found to give high results which run parallel with the capacity of the alcoholic extract to decolorise permanganate. J. P. Iodometric Determination of Thiosinamine Comparison with Other Methods and Applications. MORVILLEZ and R. MEESEMACKER (J. Pharm. Chim. 1923 [vii] 28 44245).-To 10 C.C.of the thiosinamine solution acidified with sulphuric acid are added A C.C. (excess) of 0-lN-iodine. After keeping in the dark for @teen minutes the unchanged iodine is titrated back with 0-05N-sodium thiosulphate in the presence of starch and 10 C.C. of chloroform. The titre of the thiosinamine solution is given by (A-.n/2)0.0058 x 10 where n=no. of C.C. of O*OSN-thiosulphate requued. The method which gives results in close agreement with those found by the methods of Volhard and of Denighs is applicable to the estimation of allylthiocarbimide in oil of mustard. Volumetric Method for the Determination of the Alka- loids. A. IONESCU and (MLLE.) E. SPIRESCU (Bul. Xoc. Chim. Rorn&nia 1923 5 74-80).-The authors have extended their '' mercurimetric " method previously described for acetone (A.1923 ii 666) to the determination of several alkaloids. The procedure is as previously described 1-10 C.C. of the solution of the alkaloid (containing 0.01-0-03 9.) being employed the final dilution with water before the addition of the sodium nitroprusside solution being with 200 C.C. of water. Each C.C. of O.1N-sodium chloride solution is equivalent t o the following quantities of alka- loids quinine 0.0066 g. ; strychnine 0.014 g. ; morphine 0.0083 g. ; codeine 0.01 g.; cocaine 0.009 g. The corresponding theoretical figures for the precipitate (Hg12)3,(Alk.HI)2 are quinine 0.0127 g. ; strychnine 0.011 g.; morphine 0.0102 g . ; codeine 0.0107 g.; cocaine 0.0102 g. Beckurts' Method €or Determining the Alkaloids in Cinchona Extract.G. RECORDATI (Boll. Chim. Farnz. 1923 62 707).-When 5 C.C. of liquid cinchona bark extract are extracted with chloroform the residue from the chloroform solution should be dissolved in 100 C.C. of 0-O1N-hydrochloric acid and the excess of the latter determined by titration with sodium hydroxide solution. [Cf. B. 1924 150.1 T. H. P. A. KOSSEL and R. E. GROSS (Sitxungsber. Heidelberg Akad. Wiss. [B] 1923 1-6; from Chem. Zentr. 1923 iii 1151-1152).-Protein is hydro- lysed with hydrochloric acid and the products after removal of the excess of acid and neutralisation with sodium carbonate are treated with " Naphthol-yellow-S " or 2 3-dinitro-1-naphthol- 7-sulphonic acid (m. p. 150-151"). From the arginine salt thus obtained the dye acid may be liberated by boiling with 33% sulphuric acid and serves as a means of determining the arginine.Data are given for the solubilities of the arginine histidine lysine W. T. K. B. J. W. B. Preparation and Determination of Arginine.ii. 212 ABSTRACTS Or CHEMICAL PAPERS. guanidine and ammonium salts. The arginine salt has m. p. above 260"; histidine salt 2 2 6 2 2 6 " (decornp.) ; lgsine salt 213" (decornp.); guanidine salt m. p. 274" (decomp.); the sdts oonta,in equimolecular proportions of base and acid except in the case of the ammonium salt which contains two molecules of base to one of acid. The dye acid aLso gives crystalline precipitates with agmatine methylguanidine creatinine tetramethylenediamine and purine derivatives. With carbamide a crystalline salt of the composition CO (NH,),,C,,H,O,N,S is obtained.The Determination of Fibrinogen. 11. W. STARLINGER (Biochem. Z. 1923 143 179-185).-A further survey of the methods available for the determination of fibrinogen (cf. A. 1923 ii 890). The addition of sodium citrate either as solid or in solution to native plasma has a purely additive effect on the refractive index. The following readily applicable method for the determination of fibrinogen by merential refractometry is given. The native plasma ( N P ) obtained by centrifuging 4 C.C. of blood is collected in paraffied vessels. A known volume of this-approximately 1 c.c.-is treated with citrate and the refrac- tive index of the native salt plasma (NSP) so obtained is measured. The residue of the N P is allowed to coagulate spontaneously and the native serum (NS) from this compared in the refractometer with the salt serum obtained by heating the NSP at 65" for five minutes. The dserences between the N P (obtained by a formula from NSP) and NS and between the NSP and salt serum give respectively the native fibrinogen and the salt fibrinogen. J. P. A Substitute for Acid Haematin as the Standard in Sahli's Haemoglobinometer. L. D. PELTON (Bull. Johns Hopkins Hmp. 1923 34 357-359).-Sufficient 40% sodium hydroxide solution is added t o a 10% solution of pyrogallol in 75% ethyl alcohol to yield it solution containing 10% of sodium hydroxide. The mixture is oxidised by bubbling air through it over-night or as long as a precipitate is formed. The supernatant liquid is decanted and the tarry residue dissolved in water. Precipitation with 75% alcohol and re-dissolution are continued until the product is no longer strongly alkaline. After being washed with 95% alcohol the precipitate is left in the air until the odour of alcohol has disappeared and is then dried in a desiccator over sulphuric acid. The standard liquid prepared by dissolving the powder in a Sdrensen phosphate buffer mixture of pH 6 matches acid haematin very closely down to a value representing 60%. G. W. R. CHEMICAL ABSTRACTS.

ISSN:0368-1769    

DOI:10.1039/CA9242605197

出版商:RSC    

年代:1924

数据来源: RSC

摘要:

ii. 213 General and Physical Chemistry. Spectroscopic Confirmation with Caesium of Bohr’s Distribution Numbers. E. FUES (2. PhysiE 1924 21 265- 280).-A development of previous work (A 1923 ii 1 103 353) in which it was shown how the potential gradient inside the atom could be computed from the spectral terms. The mathematical treatment has been extended from the point of view of relativity. The underlying assumption that a single value for the potential gradient can be employed for all possible orbits is only approximately correct so that the numerical results can agree only to a few units per cent. Within these limits it is possible to calculate a potential gradient within the caesium atom giving orbits which correspond in their energy values on the one hand with the terms of the caesium spectrum and on the other with the requirements of the quantum theory.From this are deduced the dimensions of the atom shells and further by the introduction of Bohr’s distribution numbers those of the individual groups of orbits. This gives a “model” from which together with very probable assumptions as to the distribution of the electrons a second computation of the potential gradient within the atom is made. This is in good agreement with the first which is considered to be a confirmation of Bohr’s distribu- tion numbers. The atomic field of magnesium has also been examined. L. J. H. Interpretation of the Quantum Theory. R. MECKE (2. Physik 1924 21 26-37).-A discussion of the quantum theory in the light of the principles of relativity. Theory of Radiation Transformation.G. KORNFELD (2. physikal. Chem. 1924 108 1 lS-l20).-An experiment designed to test Weigert’s theory of the transformation of radiation in chemical reactions (A. 1923 ii 3 813) is described. According to this hypo- thesis in a mixture of ozone and chlorine the decomposition of ozone by light is effected by radiation absorbed by the chlorine and re-emitted in a form which is absorbed by the ozone. To test this the author has examined the formation of carbon dioxide in the presence of carbonyl chloride. It is known that carbon monoxide and oxygen combine under the influence of light in the presence of carbonyl chloride to form carbon dioxide. A quartz tube containing carbonyl chloride carbon monoxide and ,oxygen was surrounded by a second tube with quartz ends containing carbonyl chloride and submitted to light.It was found that no combination of oxygen and carbon monoxide occurred in the inner tube thus demonstrating an error in the assumptions made in Weigert’s hypothesis and furnishing further proof of the sensitisation of the reaction between oxygen and carbon monoxide by kinetic means. J. F. S. J. S G. T. VOL. CXXVI. ii. 9ii. 214 ABSTRACTS OF CHEMICAL PAPERS. Behaviour o€ the Many-lined Spectrum of Hydrogen at Low Temperatures. L. CITRON (Ann. Physik 1924 [iv] 73 470476) .-The intensities of the lines in the spectrum of hydrogen cooled in liquid air were compared with those a t the ordinary tem- perature errors due to impurities which might be condensed or to the change in density on cooling being eliminated as far as possible.The greater number of the lines in the spectrum exhibited no change in intensity but others showed variations ranging from -65% to +140%. It was not found possible as a result of these experiments to distinguish groups of related lines. Origin of certain Spectral Lines hitherto attributed to Oxygen. W. JEVONS (Phil. Mag. 1924 [vi] 47 586-590).-A re-examination of the 0 lines of Schniederjost (2. wiss. Phot. 1904 2 283) shows that of the twenty-five lines of this series twenty-one are not due to oxygen but may be ascribed to carbon impurity since they are successfully identified with Deslandre’s carbon band-heads (A. 1888 637 ; Compt. rend. 1903 137 460). The line 2883.93 8. is confirmed as a genuine oxygen line although it is possibly an enhanced 0 rather than an 0 line.The remaining three. 2895.37. 2858.81. and 2708-18A.. mav still be M. S. B. regarded& oxygen lines pending further investigation of ihe bands. A. E. M. Spectrum of Manganese. S. GOUDSMIT (Nature 1924 113 238).-With reference to Catal6n’s work (A. 1922 ii 726) the values of some of the multiple terms are calculated. New Rays in the Arc Spectrum of Scandium in Air between 3200 and 2200 A. S. P ~ A DE RUBIES (And. I%. Quim. 1924 22 49-55).-Scandium oxide was prepared from the oxalate by way of the acetylacetone salt. Using this material a series of arc and spark spectrographs were obtained. The data are tabulated. R. A. M-~UIKAN and I. S. BOWEN (Physical Rev. 1924 23 1-34).-By the measurement and comparison of thirty plates obtained by the use of the vacuum apparatus and explosive spark previously described numerous lines between 136 8.and 1862 A. have been identified as belonging to one or other of the elements hydrogen helium lithium beryllium sodium boron carbon nitrogen oxygen fluorine magnesium aluminium silicon phosphorus sulphur chlorine potassium calcium chromium and copper the results being tabulated. A progression of spectra with atomic number in this region is revealed which is similar to that observed by Moseley for hard X-rays. The method is capable of revealing the full M-spectra of all the. elements up to copper (atomic number 29) ; with a few omissions above 20 these have been observed. In the case of copper this spectrum extends to about 155& Interpretation of the results in terms of Bohr’s theory gives values of constants of the 1; and M levels of the atoms as follows for sodium L(I) v/R=2-826; for magnesium L(I) 4.298 L(II) 3-402 L(III) 3.381 ; for aluminium A.A. E. G. W. R. Extreme Ultra-violet Spectra.GENERAL AND PHYSICAL CHEMISTRY. ii. 215 L(I) 6.045 ; L(II) L(IlI) 5.008 ; for calcium -M(II) v/R=1.839 M(III) 1.810. There is evidence that in some cases the strongest lines are emitted by atoms from which all the valency electrons have been stripped. Measurements of the Intensity and Direction of the Penetrating Radiation [occurring in the Earth's Atmosphere]. W. KOHLHORSTER (Sitzungsber. Preuss. Akad. Wiss. Berlin 1923 34 366-377) .-The intensity of the penetrating radiation increases with the height above sea-level; at 3000 4000 and 5000 metres it liberates 4,9 and 17 ions cmr3 seer1 respect,ively.The radiation is shown to be 7-rays with an absorption coefficient pE,0=2.5~ 10-3 cm.-l as compared with that of y-rays from Th-D pHzO=3.3 x 10-2 cmrl. It is thus harder than the y-radiation from any known radioactive element. From measurements made in a glacier on the Jungfrau at 3550 metres it is shown that the maximum intensity occurs between the hours 8 and 9 in the morning and evening when the milky way passes through the zenith. It is concluded that the radiation is emitted from the red stars of t,he milky way and is due to the disintegration of radioactive elements unknown on the earth. There is no evidence supporting the view that either the sun or the earth is the source of this radiation.A. A. E. W. E. G. Soft X-Rays from Carbon. P. I. LUKIRSKY (Phil. Mag. 1924 vi 47 466470).-A contribution to the study of the gap between " optical " and " X-ray " spectra. The photoelectric effect has generally been utilised for the detection of radiation in this dBcult range. A variant of the usual method is here developed in which the wave-length of the radiation is deduced from the velocity of the secondary electrons it releases from a target. These velocities are determined by an electrostatic method. The method is applied to the spectrum of carbon and a value 48.9 A. (&l yo) is obtained for the K line. S. B. Soft X-Rays from Tungsten. C. B. BOZZANI and C. T. CHU ( J . Franklin Inst. 1924 197 183-197).-An analysis of the pro- duction of X-radiation from the point of view of the Bohr theory calling attention to the fact that when the energy contents of successive orbits are plotted against atomic numbers these curves cross one another in the region of soft X-rays indicating that for any particular atomic number either of two exterior orbits is equally probable.With the apparatus of Richardson and Bozzani (A. 1922 ii 14) a critical impact of 356 volts ident'ified with the M line has been confirmed for molybdenum. With tungsten critical values have been obtained at 33.6 34.6 39.2 36.6 and 71 volt impacts (& 1 v.) and these figures are identified with the N N 03-4 and 0 levels respectively and are in substantial agreement with the figures of Bohr and Coster (A. 1923 ii 110) and of Hjalmar (A.; 1923 ii 448).Thus the five-quantum levels 0 and OH have greater negative energies than the four-quantum levels N and N which is in accordance with the Bohr theory A. E. M. 9-2ii. 216 ABSTRACTS OF OEEMICAL PAPERS. Spectral Series in the Soft X-Ray Region. G. K. ROLLEFSOH (Physical Rev. 1924 23 3545).-In the range 40-175 volts nine critical radiation potentials (of the M series) were observed photoelectrically for iron values for Y / R being 3.46 6.03 7-05 8.21 9.63 10-40 10.88 11-33 and 11.83 respectively. Corre- sponding values in the range 600-700 volts for the L series were 456 47.0 and 51.5. All but the first and third of the first series of points fall on a straight line when plotted against 1/N2 (N=7); the convergence limit of the M series is found to be 13.1.For the L series L,( v/R) =59.0. The critical potentials are considered to correspond with emission lines of wave-length 264 151 129.6 111.0 94-6 87.8 83.9 80-5 and 77.08 respectively and (for the L potentials) with 20.0 19.4 and 17.7 8. The N-Series of X-Ray Spectra. V. DOLEJSEK (2. Physik 1924 21 11 1-1 19) .-The identification and classification of lines in the N series of X-ray spectra of elements of atomic numbers ranging from 46 to 92 are critically discussed by means of the Bohr- Coster scheme of classification and the Moseley diagram and the results compared with those found by Hjalmar (A. 1923 ii 448). J. S. G. T. A. A. E. Rhtgenographic-chemical Investigations. 111. Quanti- tative Chemical Analysis by means of Rontgen Radiation.H. STINTZING (2. physikal. Chem. 1924 108 51-69; cf. this vol. ii 83).-The possibilities of using X-ray spectrum analysis as an aid to chemical analysis arediscussed. It is shown that with a sufficiently long exposure of the photographic film X-ray spectrum analysis is capable of detecting even smaller quantities of elements than optical spectrum analysis. A method is put forward the principle of which has been experimentally tested which is capable of quantitatively determining the amount of a given element in a mixture. This method which is not applicable to the lighter elements hydrogen to sodium consists in embedding the substance to be analysed together with a comparison substance in suitable quantity and state of subdivision in an indifferent substance The quantitative relationships are deduced from the ratio of the intensities of the lines of the substance to be analysed and the comparison substance.The indifferent substance used was graphite and this mixed with powdered material was compressed with water- glass into pastilles. The method will only give usable results when a very efficient mercury vapour pump is used in the system. It is shown that the lighter elements may also be examined by the use of mixed crystals of suitable superposed space lattices whereby in vacuum interference occurs between wnve-lengths of double the normal value. This increase in the value of the lattice constant is sufficient to make it possible to obtain the X-ray spectra of the elements lighter than sodium and so determine them.Rontgenographic-chemical Investigations. IV. Measuring Instruments for High Vacua. H. STINTZING (2. physikal. Chem. 1924 108 70-81 ; cf. preceding abstract).-Two methods of J. F. S.GENERAL AND PHYSICAL CHEMISTRY. ii 217 measuring the pressure in high vacua are described. The method of compressing the gas has been modified so that it may be used to measure pressures down to 10-6 mm. of mercury. By combining this method with the discharge method it is possible to measure pressures down to mm. of mercury. J. F. S Variation of the Band Spectrum of Nitrogen produced by the Rare Gases. W. STEUBING and M. TOUSSAINT (2. Physik 1924 21 128-137).-Very marked variations in the respective intensities of the various series of bands in the nitrogen spectrum are produced by the presence of large proportions of ionised argon (2.Physik 1923 19 271). The authors find that a similar effect less marked both qualitatively and quantitatively is pl'oduced by the presence of strongly ionised neon. The effect is not attributable t o a variation of temperature of the discharge and is not produced by the presence of other electronegative or electropositive elements of high or low atomic weights. It is suggested that the effect produced by the ionised rare gases is attributable to an alteration in their respective electronic configurations occurring due to ionisa- tion and that the effect on the nitrogen band spectrum is produced in the case of argon when this gas becomes singly ionised whilst multiple ionisation is necessary in order that the effect may be produced by neon.J. S. G. T. Continuous Spectrum of Hydrogen. 0. W. RICHARDSON and T. TANAKA (Nature 1924 113 192-193).-Using as a source of electrons a filament either of tungsten or of platinum coated with barium oxide about 2 cm. from a flat circular nickel anode of 2.5 cm. diameter provided with a central rectangular slit the whole being enclosed in an evacuated silica tube observations have been made of electron discharges in hydrogen which appear to supplement those described by Lemon (this vol. ii 133). Over the range examined except at the lowest pressures there are two stable types of luminous discharge; a t pressures from 0.05 to 0.1 mm. on a higher range of voltage a third type is observed. The continuous spectrum is observed in the case of the low-voltage type. Details of the work are promised in a subsequent corn- munication.A. A. E. Continuous Spectrum of Hydrogen. F. HORTON and A. C. DAVIES (Nature 1924 113 273-274; cf. A. 1923 ii 820).-In view of the results previously obtained by the authors it is probable that the change from a line spectrum to the continuous spectrum which Lemon (this vol. ii 133) found to occur on increasing the electron emission from the filament was due to an accompanying decrease in the electric intensity between the filament and the part of the luminous discharge under observation. A. A. E. Properties of Dyed Materials. H. WALES ( J . Amer. Ckm. Boc. 1923 45 2420-2430).-The position of the maximum in the absorption spectrum of a solution of a dye determines the generalii.218 ABSTRACTS OF CHEMICAL PAPERS. colour but not the particular shade both of the solution and of wool dyeings produced from it. A logical method of comparing dyes is to map the reflection spectrum of a dyed fabric. This has been done by a method similar to that of Meek and Watson (Chem. Xoc. Trans. 1916 109 544) using wool coloured by means of a series of xanthene dyes having absorption maxima between 4875 and 5550 A. The spectrophotometric curves from wool dyeings are much broader than those from aqueous or alcoholic solutions. Dyeings made on filter-paper have reflection curves identical with those of the dyed wool. The reflection spectra of thin gelatin films containing Eosin BN and erythrosin are the same as those of the wool dyeings and the transmission spectra of these films correspond with those of the aqueous solutions.These results indicate that the dyes are actually dissolved in the wool. A discussion follows of the '' simple solution " theory of dyeing. w. s. N. Variations in the Specific Refraction of Gases with Pressure below one Atmosphere. V. POSEJPAL ( J . Phys. Radium 1923 [vi] 4,451460) .-Measurements of the variations in the refractive index of oxygen and hydrogen with pressure have been made at pressures below 760 mm. Using data for the refractive indices and the compressibilities of the gases the change in their specific refraction is calculated over a wider pressure range. The specific refraction of oxygen (as of air and carbon dioxide) increases rapidly with pressure up to two or three atmospheres above which it is nearly constant whilst the values for hydrogen diminish at first with rise of pressure becoming nearly constant above one atmosphere.A qualitative explanation of the behaviour of these gases is given based on the Bohr theory. S. B. Formula for Rotation-dispersion. F. BURKI (Helv. Chim. Acta 1924 7 163-167).-From the formula [a]=k/h2-Aq2 it is shown that the relation between wave-length and rotation for optically active substances is given by [cr]=C/&. epld\' where C and This may be regarded as a generalised form of Boltzmann's formula [a]=A/h2+B/h4. The formula may also be written log h2[ a] =log C+ Pv2 where Y is the frequency ; log h2[ 71 is a linear function of the square of the'frequency. This graphic relation is found to hold in the examples tested.E. H. R. are constants. Harmonic Frequency Relations in the Infra-red Absorp- tion Spectra of Liquids and Solids. J. W. ELLIS ( J . Opt. Xoc. Arner. 1924 8 1-9).-A determination of the respective frequencies of the bands in the infra-red absorption spectra of various substances including cottonseed oil a petroleum distillate lard oil kerosene mesitylene pentane heptane and water suggests that a pair of chemically combined atoms e.g. carbon and hydrogen or oxygen and hydrogen responds not only to an incident electro- magnetic wave of a certain frequency but also to waves the fre- quencies of which are whole multiples or approximately whole multiples of the fundamental frequency. J. S. G. T.GENERAL AND PHYSICAL CHEMISTRY.ii. 219 Near Infra-red Absorption Spectra of some Organic Liquids. J. W. ELLIS (Physiml Rev. 1924 23 48-62).- Observations with a self-recording prism spectrograph were made in the following cases methyl ethyl propyl butyl methylene and ethylene halides butyl nitrite benzene toluene pentane hexane heptane octane mesitylene p-xylene chloroform and bromoform. The wave-lengths of the absorption maxima accurate to about 0.01 p are tabulated up to 2-4 p. Visual observations were also made in the case of chloroform benzene toluene acetone methyl iodide pentane hexane heptane ethyl bromide and ethyl ether. Besides common bands at about 1.17 1-38 and 1.7 p bands near 0.9 and 1.0 p appear in many of the compounds; their presence is ascribed to the carbon-hydrogen linking since they are not observed with carbon disulphide or carbon tetrachloride or with compounds containing hydrogen without carbon.Shifts (up to &0.03p) in the positions of the bands for different liquids may be due to a change in the relative intensity of the two components of the band. A. A. E. Absorption of Potassium Vapour at High Temperatures and Satellites accompanying the Members of the Principal Series. A. I;. NARAYAN and D. GUNNAYYA (Physical Rev. 1924 23 46-47).-The existence of the lines 5340 5300 5205 5100 4970 and 4831 8. is confirmed. Fifteen satellites observed in the absorption spectrum are tabulated. Absorption Spectra of some Derivatives of Phenol and other Substances. J. E. PURVIS ( J . Chem. Xoc. 1924 125 406-418).-A continuation of previous work (Chem.Soc. Trans. 1923 123 2515). Absorption spectra of solutions (in some cases vapours) of many phenolic substances have been investigated. None of the substances exhibited any of the narrow vapour bands characteristic of phenol or of its simpler derivatives. The type position and intensity of absorption in the ultra-violet regions are regulated by the nature of the unsaturated elements or groups of elements acting as oscillatory centres. With increasing com- plexity the vapour bands and eventually the solution bands also disappear. It is probable that completely saturated substances do not exhibit specific absorption in the ultra-violet. The general absorption depends on the translatorv motion of the molecule. and A. A. E. this &ll be cohxolled by the weight (Gf.A. 1913 ii 363). S. K. T. Fluorescence Spectra. 11. Phenol and Phenolic Ether Vapours. J. K. MARSH ( J . Chem. Xoc. 1924 125 4 1 8 4 2 3 ) . - All the substances examined in continuation of previous work (Citem. Soc. Truns. 1923 123 3315) had similar spectra beginning with wave-lengths of about 2750 A attaining a maximum at about 2900 B. and dying out at about 3700 A. Some substances (veratrole methyl salicylate benzyl alcohol etc.) show a second spectrum in the blue region and phenyl mercaptan shows this spectrum alone. The weight of the side chain the residual affinity in the side chain,ii. 220 Al3STBACTS OF CEEMICAL PAPERS. and ortho-substitution may influence the production of this second region. In the case of cresols the fluorescence increases in the order ortho meta para.Phenol and anisole show a number of fine fluorescence bands and p-cresol and its methyl ether each show a few. 8. K. T. Sensitised Fluorescence of Lead and Bismuth Vapours. R. KOPFERMANN (2. Physik 1924 21 316-325).-The method of sensitising fluorescence in vapours by the action of light absorbed bymercuryvapour used by Cario and Franck (A. 1922 ii 809) is applied to lead and bismuth vapours at 950". Lead gave the lines 4057.93 3683.60 3639.71 2873.40 and 2833.17 8. which agree with the series scheme of Grotrian (A. 1923 ii 710). Those of bismuth 4722.72 and 3067.81 8. suggest further regularities in the arc spectra of this element. E. BAYLE and R. FABRE (Compt. rend. 1924 178 632-635).-The dominant wave-length of the fluorescence spectra of a number of organic compounds has been measured using as source of excitation the Hg,,,q line.The intensities of fluorescence are compared with that of sodium salicylate which possesses a high fluorescence under these conditions. Novocaine shows an intense fluorescence which permits of its ready detection in mixtures also containing cocaine and stovaine. These substances exhibit no fluorescence. E. E. T. Polarised Light of Fluorescence. A. CARRELLI (Atti R. Accud. Lincei 1924 [v] 33 i 21-23).-The author confirms the results obtained by Weigert (Physikul. Z. 1922 23 232) and by Schmidt (ibid. 233) with reference to the influence of the solvent on the polarisation of the fluorescent light emitted by a solution. It is also shown that the emission is polarised when i t leaves the particles so that the polarisation depends neither on dichroism of the solution nor on a phenomenon of luminous diffusion.T. H. P. Phosphorescence of Fused Transparent Silica. D. L. CHAPMAN and L. J. DAVIES (Nature l924,113,309).-When oxygen or hydrogen is driven into fused quartz by the electric discharge the quartz acquires the property of phosphorescing. After the glow has ceased the quartz can again be made to phosphoresce by heating it ; the phosphorescence caused by heating can be observed with no apparent diminution in intensity 24 hours after the stimu- lating discharge has been stopped. Phosphorescence of Fused Transparent Silica. E. B. LUDLAM and W. WEST (Nuture 1924 113 389).-The observation of Chapman and Davies (preceding abstract) is confirmed.The phosphorescence exhibited by fused silica after exposure to the light from an aluminium spark has already been attributed to the fact that silica begins to absorb strongly in the neighbourhood of lines at 1860 and 1760 A. The colour of the phosphorescence differs with the specimen of silica and is probably due to impurity. A strong line at 2060 A. is observed in the case of iodine. L. J. H. Fluorescence of some Organic Compounds. A. A. E. A. A. E.GENERAL AND PHYSICAL CH3lMISTBY. ii. 221 Method of Determining Radiating and Ionisation Potentials not necessitating Low Pressures. W. J. CLARK (Proc. Durham Phil. Soc. 1922-23 6 312-322).-A stream of electrons from a cathode coated with lime passes through an adjacent gauze screen which can be charged positively to give varying accelerating poten- tials.The stream then traverses the gas and falls on an insulated conductor connected to an electroscope. Working with air a t pressures of the order of 1 mm. of mercury the author has deter- mined the form of the curves showing the variation of the potential of the conductor and the current to it with varying accelerating potentials. The potential curves are influenced considerably by perturbing secondary effects but the current curves under suitable conditions exhibit discontinuities from which approximate values of the radiating potentials of nitrogen and oxygen and of the Statistical Equilibrium and Mechanism of Ionisation by Electronic Impacts. R. H. FOWLER (Phil. Mag. 1924 [vi] 47 257-277) .-A discussion of the mechanism of the equilibrium between ions electrons and atoms in a dissociating gas with special reference to the relative importance of collision and radiative processes.Kleiii and Rosseland's theory of non-radiative collisions has been amplified and extended to ionisation by electronic impact. It is deduced that radiationless (collision) processes are negligible in the reversing layer of the sun (10-4 atm.) but are of greater importance than radiative captures a t a pressure of 1 atm. Other astrophysical applications are mentioned. Theory of the Electron Current in Ionisation by Collision. L. HEIS (Physikul. Z. 1924 25 2541).-The theory of Townsend on the ionisation of gases by collision (London Constable 1910) is not in good agreement with experiment for the rare gases and breaks down completely under the experimental conditions of Franck (Physikul.Z. 1921 22 358). The author extends this theory and deduces relationships which are in agreement with the experimental work of both Townsend and Franck. J. S. TOWNSEND and T. L. R. AYRES (Phil. Mug. 1924 [vi] 47 4 0 1 4 1 5 ; cf. A. 1923 ii 366 72l).-Measurements were made of the currents passing between electrodes in helium with variation of gas pressures potential differences and electrode distances. From these it is deduced that the energy necessary to ionise a helium atom does not exceed that of a 10-volt electron but that the energy of the electrons must be twice this value for appreciable ionisation to occur. Previous work on ionisation and resonance potentials is criticised in the light of these results which it is claimed show that electrons moviiig with velocities corresponding with resonance potentials rarely lose their energy by molecular collisions. ionisation potential of oxygen have been derived.s. s. S. B. A. E. M. Ionisation by Collision in Helium. S. B. Motions of Electrons in Neon. V. A. BAILEY (Phil. Mag. 1924 [vi] 47 379-384 ; cf. A. 1923 ii 721 and preceding abstract). -Measurements have been made of the velocity and energy of 9"ii. 222 ABSTRACTS OF CHEMICAL PAPERS agitation of electrons in neon under the influence of an electric field. The values deduced for the mean free paths of the electrons in neon and in other gases are in disagreement with those obtained from viscosity determinations.S. B. Re-combination of Gaseous Ions Chemical Combination of Gases and Monomolecular Reactions. J. J. THOMSON (Phil. Hag. 1924 [vi] 47 337-378).-The f i s t section of this paper is devoted to the theory of the rate of re-combination of positive and negative ions the rate of formation of complex ions and kindred subjects. I n the second section the results are applied to problems of chemical combination. With the assumption that complex formation must precede true chemical combination it is deduced that charged ions or “ polar ” molecules can accelerate reactions (a) by promoting the formation of additive compounds (b) by assisting atomic rearrangements within the molecule. An expression resemb- ling Arrhenius’ empirical equation is obtained for the rate of dis- sociation of gaseous compounds on the hypothesis that the decomposition is due to molecular collisions and the usual objections to this view are discussed.Finally an expression is developed for the rate of unimolecular reactions in solids such as sublimation on the assumption that the “critical energy increments” are supplied by pulses running through the space lattice of the solid. The influence of the walls of the containing vessel on reaction velocities is considered in the light of this theory. Potential Gradient in the Sodium-Potassium Vapour Arc Lamp. F. H. NEWMAN (Phil. Mag. 1924 [vi] 47 6 0 3 4 8 ) . - The fall of potential along the arc of a vapour lamp with each electrode of sodium-potassium alloy (2 parts Na to 1 part K) has been investigated by means of a probe electrode for currents of 1-2 amp.The potential gradient along the arc column is uniform. The potential drop at the cathode is greater than that a t the anode but both are small compared with the results for other metal electrodes. Application of Poisson’s law shows no preponderance of ions of either sign. The small potential difference between the terminals of the lamp when working is explained by the photo- electric effect exhibited by the alloy. S. B. A. E. M. Calculation of the Normal Cathode Drop of Potential in Mixtures of Gases. A. GUNTHER-SCHULZE (2. Physik 1924 21 50-59).-An expression for the normal cathode drop of potential V, in a mixture of gases is derived in the form Vm2= X ~ Y ~ ~ V ~ ~ ] where Vil and Via are the ionisation potentials of the respective gases Vl and V2 the respective normal cathode falls A and A the values of the mean free atomic paths in the gases at N.T.P.P1 and P2 the respective mean number of ions produced by an electronic impact ccl and u2 are coefficients rela,ting to the increase of the cathode fall due t.0 inelastic impacts and x and 1-x the respective proportions in which the gases 1 [(A2/11) + (1 - ~ ) P a ~ z z ~ 2 / ~ p l ~ ~ ~ ~ l l / [ I ( X 2 / X l V12) 3- (1 -x)Pz‘vit~21GENERAL AND PHYSICAL CHEMISTRY. ii. 223 and 2 are present in the mixture. The formula is shown to be in close agreement with experimental results for mixtures of monatomic gases e.g. mixtures of argon and mercury vapour. In the case of mixtures containing a diatomic or other multi- atomic gas or vapour the measured values of the normal cathode drop of potential are in all cases greater than the calculated values throughout the greater part of the range of mixtures employed.This discrepancy between experimental and calculated values is attributed to the removal by a process of chemical combination of cations produced in the glow by the primary electrons. The discrepancy is exhibited to a slight extent by mixtures of argon and oxygen and to a considerably more marked degree by mixtures of helium and oxygen and the author concludes that in both these cases in consequence of ionisation compounds of the respective gases argon and helium with oxygen are formed. Theory of Emission of Positive Ions and Electrons from Glowing Metals.M. VON LAUE (Xitxungsber. Preuss. Akud. Wiss. Berlin 1923 32 334-348).-A theoretical paper in which the thermodynamic equilibrium between a metal and its ionised vapour is employed as a basis for the deduction of expressions for the positive and negative saturation currents from the surface of the metal. It is deduced that the saturation currents at the melting points of the metals will be the same for the liquid and solid phases. This is in agreement with the experiments of Goetz (Physilcal. Z. 1923 24,377) at the melting points of iron copper and manganese. J. S. G. T. A. E. M. Law and Mechanism of the Emission of Electrons from Hot Bodies. S. C. ROY (Phil. Mug. 1924 [vi],47,561-569).-Assuming that the electrons within a metal form ideal Born-K&rman space lattices and utilising a quantum theoretical value for their energy content and identifying the thermally emitted electrons with photo-electrons an expression is deduced giving electron emission as a function of temperature.This is essentially of the same form as that of Dushman (Physical Rev. 1922,20 109). Intrinsic potentials for alkali and alkaline-earth vapours are calculated from thermionic and from photoelectric data and are in fair agreement with thermionic observations. Comparison of intrinsic potential with observed resonance potential suggests that the mechanism of emission is a resonance process. Normal and Selective Photoelectric Effects associated with the Alkali Metals. K. FARWIG (2. Physik 1924 21 3845).- Measurements of the photoelectric effects exhibited by sodium potassium and rubidium illuminated by radiation between 2000 and 6000A.indicate that the photoelectric effect on the long wave-length side of the selective region characterising these metals is continuous with the normal effect exhibited on the short wave- length side of this region. The curve representing the selective photoelectric effect is superposed on the continuous curve repre- senting the normal effect as suggested by Pohl and Pringsheim A. E. M. 9"-2ii. 224 ABSTRACTS OF CHEMlCAL PAPERB. (Verh. d I). Phys. Ges. 1910,12,682 and A. 1910 ii 922). Pranck'a interpretation of the maximum selective effect (2. Physik 1923 13 241) is shown to be untenable. Photoelectric Effects in Rock Salt Crystals. J. BINGEL (2. Physik 1924,21,229-241) .-Several rock-salt crystals natural and artificially coloured (by exposure to X-rays) were exposed to monochromatic illumination of intensities 5 x 10-5-5 x cal./ sec.cm.2 and the photoelectric currents measured. There is a linear relationship between current and applied voltage and no evidence of saturation up to field strengths of 50,000 volt/cm. The quantum relationship between the energy absorbed and output tends to show that the phenomenon is that of photoelectric con- duction rather than of dielectric polarisation. The displacement of the crystal lattice can be of only molecular dimensions. J. S. G. T. A. E. M. Electrical Conductivity associated with the Excitation of and Emission of Light from [various] Phosphors. B. GUDDEN and R. POHL (2. Physik 1924 21 1-8).-In continuation of previous work (cf.A. 1922 ii 680) the authors have correlated the photo-excitation of various phosphors e.g. crystals of Zn-S-aCu ZnS-Mn Zn-S-yCu and the emission of light accompanying the excitation with the phenomena of the primary and secondary photoelectric currents in the crystals. Employing a string electro- meter the authors were able to detect transport of electricity within the excited crystal independently of whether the returning electron caused emission of light or not. The conditions under which the quantum equivalence law holds for the absorption and emission processes are briefly discussed. Change in the Temperature Coefficient [of Electrical Conductivity] of pure Platinum by Mechanical Treatment. 0. FEUSSNER (2. Physik 1924 21 163-167).-A number of experiments are described in which the resistance of a thin pure platinum wire (0.1 mm.diam.) is compared with that of a similar wire which is subjected to the pull of various weights from 0 to 130 g. The measurements were made at 0" and 100". I n most cases the change in the temperature coefficient scarcely falls outside the experimental error. J. F. S. M. R. LUCAS (J. Phys. Radium 1923 [vi] 4 491492).-Pyro-electric phenomena have been observed with p-dimethylaminobenzylidene- camphor and dibenzylcamphor. When crystals of these substances are suspended by fibres and heated at loo" they are attracted or repelled by charged bodies. The direction of polarisation is perpendicular to the long axis of the crystals whereas tourmaline is Hall Effect and Specific Resistance of CathodicaUy De- posited Films of Gold. S.S . MACKEOWN (Physical Rev. 1924 23 85-93).-The Hall coefficient for sputtered films of gold is the J. S. G. T. Pyro-electricity of Two Derivatives of Camphor. polarised in the direction of this axis. s. B.ii. 225 GENERAL AND PHYSICAL CHEMISTRY. same as for bulk metal namely 643 x lo*( &l yo). The results of experiments on the resistance and change of resistance of such films are in agreement with work already published and are expressed in graphical form. E. WEDEKIND (2. angew. Chem. 1924 37 87-89).-Whilst the compounds of iron with the exception of the oxide Fe304 are much less strongly magnetic than the metal itself the reverse is true of manganese. I n the latter case the magnetic effect is a maximum when the metal is present in the tervalent state (e.g.in the boride arsenide etc.) and is considerably less in the bivalent state (e.g. the sulphide). Chromium is feebly magnetic but yields two strongly magnetic oxides Cr,O and Cr,O which may be regarded as derived from the tri- and sesqui-oxide and are thus analogues of Fe304. Vanad- ium also exhibits latent magnetism the monoxide VO having the highest susceptibility and the other oxides V?O3 VO and V 0 following in order of diminishing susceptibihty; for the sdpkdes however this order is reversed. The susceptibility of titanium compounds is still less than that of vanadium compounds the maximum being found here as in the chromium series in the salt-like oxide Ti30,. Whilst the susceptibility decreases in passing down the series iron manganese chromium vanadium titanium the effect of the second component (e.g.oxygen sulphur) varies. I n general the susceptibility of the oxides decreases with increasing oxygen content with the exception of the mixed salt- like oxides (e.g. Fe304.). Copper itself feebly diamagnetic is magnetic in many cupric compounds but generally diamagnetic in cuprous compounds. Attention is directed to the periodic nature of the susceptibility and the relation between the latter and colour in the case of the rare earths. Mass Spectra of Chemical Elements. V. Accelerated Anode Rays. F. W. ASTON (Phil. Mag. 1924 [vi] 47,3S5-400).- The extension of mass-spectrum analysis has been hampered by difficulties in the production of the mass rays of many metallic elements.Unsuccessful attempts to produce them by drawing out the ions from a vacuum arc are described. Details are given of a method using the anode rays from a composite anode of graphite and metallic salts which has proved of very general application. With this apparatus the principal isotopes of lithium beryllium magnesium and calcium are found to have atomic weights closely approximating to whole numbers (cf. A. 1923 ii 748). Capture and Loss of Electrons by a-Particles. E. RUTHER- FORD (Phil. Mag. 1924 [vi] 47 277-303).-The frequency of the capture and loss of electrons by a-particles is discussed from observations on the number of singly charged and neutral helium atoms present in a pencil of or-rays. The experimental procedure ia to observe the magnetic deflection in a high vacuum of @-rays from radium-B and radium-C counting the particles by the scintilla- tion method.The neutral atoms remain undeflected and the singly A. A. E. Magneto-chemical Investigations. W. T. K. B. S. B.ii. 226 ABSTRACTS OF CHEMICAL PAPERS. charged atoms are deflected half as much as the doubly-charged atoms. The velocity of the a-particles is varied by placing mica screens before the source. With decreasing velocities the number of particles which have captured electrons becomes relatively larger. It is found that the number of singly charged and neutral atoms falls off almost exponentially with increase of the gas pressure in the apparatus. The mean free paths of a particle for capture and loss of an electron are deduced from this effect of the presence of gas.For example with a-particles of velocity 0.85 x lo9 cm./sec. the mean free path for loss is 0.005 mm. and for capture 0.037 mm. in air a t N.T.P. The ratio of the mean free path for capture to that for loss varies as PG where V is the velocity of the a-particles. The theoretical difficulties involved in the capture of electrons by high-speed a-rays are considered and i t is concluded that the process is intimately connected with that taking place in the ionisation of a gas by the rays. S. B. Application of the Law of Probability to the Radioactive Emission of Polonium. W. KUTZNER (2. Physik 1924 21 281-298).-The a-ray emission of polonium was investigated to see whether i t conformed to Bateman’s law (A.1910 ii 917) and was found to show sub-normal dispersion that is the proportion of the time fractions during which approximately the average number of particles was emittled was found greater than theory requires. The cause is discussed and i t is suggested that the atoms are influenced by emissions from other atoms. L. J. H. A. BECKER (8. Physik 1924 21 304-315).-An instrument for the compara- tive measurement of radium emanation and a method of calibrating it are described. Precision Measurement of Radium Emanation. L. J. H. Experiments with a Glow Lamp as a Counting Chamber [for the Detection of a- p- and 7-Rays etc.]. 0. v. BAEYER and W. KUTZNER (2. Physik 1924 21 46-49).-When a direct- current voltage of from 150 to 200 volts is applied to a circuit comprising a high resistance in series with an electric glow (micro- phone) lamp shunted with a suitable capacity the lamp glows intermittently.With suitable choice of the applied voltage the intermittent glow disappears but may be restored by bringing near the lamp a source of a- p- or y-radiation each lighting-up of the lamp indicating the incidence of an a- p- or 7-ray. The phenomenon may also be produced by the incidence of light of shorter wave-length than about 3800 8. on the iron anode employed in the lamp or as the result of electrostatic effects produced in the neighbourhood of the lamp. The frequency of the intermittence appears to be proportional to the intensity of the incident light. J. S. G. T. Number of a-Particles expelled by Radium. I. Counting of Scintillations.H. GEIGER and A. WERNER (2. Physik 1924 21 187-203).-The number of a-particles emitted by 1 g. of radiumGENERAL AND PHYSICAL CHEMISTRY. ii. 227 (without disintegration products) per sec. has been determined by counting the scintillations produced in a zinc sulphide screen and the value 3.40 x 1O1O obtained. The method used was designed to eliminate the more obvious causes of error such as fatigue of the eye and imperfections of the screen. This value is compared with Rutherford and Geiger’s value of 3.57 x 1O1O and Hess and Lawson’s value 3-72 x 1O1O and the reasons for the discrepancy are discussed. Making use of the above figure and of Millikan’s value of the ele- mentary quantum the half period of radium is calculated to 1730 years and the evolution of helium to 159 cu.mm. per year per gram of radium in equilibrium with emanation radium- A and radium-C’. The heat evolution of 1 g. of radium (free from disintegration pro- ducts) is calculated to 22.25 cal./hour per gram of radium. Com- paring this value with t,he experimental values of Rutherford and Hess namely 25.2 and 25.1 cal./hour it appears likely that in the radioactive changes of the radium atom in addition to the kinetic energy of the or-particles and the recoil atoms other quantities of energy must be set free in consequence of the changes in the atomic nucleus. J. P. S. Employment of Normal Radium and Uranium Solutions for Emanation Measurements. 11. P. LUDEWIC and E. LORENSER (2. Yhysik 1924 21 25S-263).-Using the apparatus previously described (A.1923 ii 453) further measurements have been made with normal radium solutions from the Physilcalisch- Technische Reichsanstalt and with several uranium solutions of the E’reiberg Radium Institute. The influence of the pressure in the ionisation chamber on the ionisation effects of y- and mays has been investigated between 400 and 1000 mm. The y-Fay ionisation is a linear function of pressure and x-ray ionisation is a maximum a t 715-735 mm. The extension of the measurements over a further year and the application of the pressure corrections has served to prove the constancy of the solutions and to substantiate further the use of the method as a standard. A. E. M. Artificial Disintegration of Atoms. G. KIRSCH and H. PETTERSSON (Phil. Mag. 1924 [vi] 47 500-512).-A detailed account of previous work (A.1923 ii 675 819). A. E. M. Electronic Theory of Valency. B. FLURSCHEIM (Phil. Mag. 1924 [vi] 47 569-576; Trans. Farachy Xoc. 1923 19 531- 535).-A reply to criticisms by Lowry (A. 1923 ii 848) of con- ceptions advanced by the author. A. E. M. Secondary Valency in the Light of Recent Researches. 11. Alkali Polyhalides. G. L. CLARK (Amer. J. Xci. 1924 7 109- 120; cf. A. 1923 ii 856).-A detailed consideration of the facts which must be taken into account by any constitutional formula for alkali polyhalides leads t o the conclusion that there is only one possible mechanical formulation for these substances which is in agreement with experimental facts. This is the constitutionalii. 228 ABSTRACTS OF CHEMIClAL PAPERS.formula suggested by Knorr (2. anorg. Ch.em. 1923,129,109) accord- ing to which cssium iodide e.g. is represented by This represents the halogen atoms in a line as a singly acting unit with two atoms sharing a single electron which describes an eccentric orbit around them. Each iodine atom is surrounded by the rare gas configuration of eight electrons. The arrow shows that a valency electron has been removed from the czsium atom with a resultant polar linking between the halogen group and the metal ion. Consequently the halogen atom next to the metal will invar- iably be the strongest halogen accounting for the fact that in changes brought about by increased temperature the strongest halogen remains bound to the metal. J. F. S. Specific Subsidiary Valency Linkings. Internally Complex Salts of Bivalent Iron.H. LEY C. SCHWARTE and 0. MUNNICH (Be?. 1924,57:[B] 349-356).-Although glycine and its homologues a- and p-aminopropionic acids piperidinoacetic acid and aromatic amino-acids such as anthranilic acid do not give characteristically coloured ferrous salts this is not the case with compounds obtained from acids containing the group C:N*C02H in which the C N - group is a component of a cyclic system; these vary in colour from reddish-yellow to dark violet and are regarded as internally complex salts having the group [.F.<which is analogous to the ferronitric oxide complex [Pe<TH2)j (cf. Manchot A. 1914 ii 557 567). They show the ultra-violet absorption spectrum of the acid from which they are derived but possess a character- istic band in the neighbourhood of 1!~ = 2000 which is considered to be characteristic of the attachment of ferrous iron to nitrogen by a subsidiary valency.The tendency of the acids to form complex copper and ferrous salts does not appear to be parallel. The following new salts have been prepared Ferrous pyridine-2- curboxylate (C6H,0,N)2Fe,411[20 brown crystals which in aqueous solution in the absence of air and light passes into basic ferric p yridine-2 -carboxylate ( C6H402N) J?e OH a microcrystalline yellow substance ; ferrous pyridine-2 3-dimrboxylate (+ H,O) brownish-black lustrous crystals and the salt C,H,O,,NFe,ZH,O yellow crystals ; f errou.~ p yraxine-2 3 -d imrboxylate C ,H,0,N2Fe,2H20 small prisms ; copper quinoxaline-2 3-dicarboxylate ib pale green substance ; ferrous quinoxaline-2 3-dicarboxg,?ate C oH404N2Ele 3H,O dark prisms.A neutral solution of quinoxaline-2 3-dicarboxylic acid is a sensitive reagent for ferrous iron with which it gives a violet color- ation; in the presence of salts of ferric iron nickel cobalt or copper the quinoxaline-2 3-dicarboxylates of these metals are first C,()H404N2Cu>H20,GENERAL AND PHYSICAL CHEMISTRY. ii. 229 precipitated after which the coloration appears. The sensitiveness is of the same order of magnitude as that of Turnbull’s blue reaction. H. COLLINS (Chem. News 1924 128 145-147).-The author’s views (a) that the atom of phosphorus has the constitution expressed by the formula Na-H-H3-H-H3 and ( b ) that the heat of formation of an element is proportional to the product of the atomic weight and the change of volume are developed.Constitution and Structure of an Atom of Sodium. H. COLLINS (Chem. News 1924 128 161-163).-Various numerical values are considered to support the author’s view that an atom of sodium has been formed by the union of an atom of carbon with one of boron. A. A. E. Chemical Nature of the Purple of Cassius. A. HUBER (Physikal. Z. 1924 25 45-47).-An X-ray analysis of a sample of purple of Cassius by the Debye-Scherrer powder method gave thirteen interference lines for eleven of which the corresponding lattice constant is that of gold. The remaining two are also given by a preparation of colloidal stannic acid and correspond with the strongest reflections from crystalline stannic oxide.It is concluded therefore that the gold is not chemically combined but that the substance is merely a mixture of colloidal gold and colloidal stannic acid. A. E. M. Structure of Vanadium. H. COLLINS (Chem. News 1924 128 100-102).-Evidence based on the relative volumes and the heats of formation of their compounds is adduced in an attempt t o prove that vanadium has been formed by the union of an atom of magnesium with an atom of aluminium. Solid Solutions and Inter-atomic Relationships. A. L. NORBURY (Nature 1924 113 271).-Polemical with Rosenhain (ibid. 1923 112 832). Experimental data show that there is no connexion between the sizes of the solute atoms and their effects in increasing the electrical resistivity of a metallic solvent. The diminution of conductivity of metals with increasing temperature is probably due to the effect of heat on the electrons and not on the atoms.A. A. E. W. ROSEN- HAM (Nntzrre 1924,113,271-272 ; cf. Norbiiry preceding abstract). -Polemical and explanatory. The effect of the contraction or expansion of a lattice on the electrical properties of a metallic system is discussed. F. SIMON and C. VON SIMSON (Z. Physik 1924 21 168-1’i7).-An apparatus is described whereby the crystal structure of condensed gas may be determined using the X-ray method. By means of t h h apparatus the structure of the two modifications of hydrogen chloride has been H. W. Structure of Phosphorus. A. A. E. A. R. P. Solid Solutions and Inter-atomic Relationships. A. A. E. Crystal Structure of Hydrogen Chloride.ii.230 ABSTRACTS OF CHEMICAL PAPERS. examined. The variety stable above 98" abs. has a cubic face- centred lattice with a lattice constant of 5.50 &- 0.05 A. Accurate measurements could not be made with the second variety which is stable below 98" abs. but sufficient information was obtained to show that it possesses a smaller symmetry than the other variety. Hydrogen chloride is shown to possess a molecular lattice. The density of both modifications of hydrogen chloride has been deter- mined; for the variety stable above 980aba d = 1.469 for that stable below 9Qoabs Using the first of these values the lattice constant of the variety stable above 98" abs. is calculated to 5-47 A. The density of hydrogen chloride a t the melt'ing point is 1.277.It contracts therefore 15% on freezing and 2.5% on transition at 98" abs. = 1.507. J. P. S. Halogen Hydrides. H. BELL (Phil. Mag. 1924 [vi] 47 549-560).-The energy formula of Born and Huckel is used in it theoretical analysis of the infra-red absorption data of hydrogen fluoride chloride and bromide. Evaluation of t!he constants in the formula has rendered possible the determination of the variation with distance of the force acting on the hydrogen nucleus as it vibrates along the nuclear line. The relative range of vibration of hydrogen fluoride is larger than those of the other two compounds owing to the larger relative size of its quantum. The restoring forces for HF HCL and HBr are in the proportion 1-84 1 0.80 and the corresponding relative variation with distance is 1-33 1 0.98.It is deduced that the halogen atoms have all increased in radius by a constant quantity at the expense of other atoms. The radius of C1-ions has been increased from 1.05 8 . to 1-72 A. which value is in agreement with those of Bragg James and Bosanquet (A. 1922 ii 703) and of Davey (A. 1923 ii 17). It is concluded that the hydrogen nucleus is well within the halogen shell and is buried to the same depth in all three and from the argon electron " economy " of Kramers (Naturwiss. 1923 27 550) it is deduced that the oscillations of the hydrogen nucleus are always well outside the etable configuration of the neon shell. A. E. M. Difference between the Specific Heats of Liquid and Vapour at Constant Volume. A. BRANDT (Ann. PhysiE 1924 [iv] 73 412-414).-On applying to carbon dioxide Planck's formula for c1 - c2 where c is the specific heat of the liquid at constant volume and c2 that of the saturated vapour a t the same temperature it is found that a t - 5" + lo" and + 25" c1 - c2 = 0.Similar results are obtained with the so-called normal substances pentane isopentane etc. It must not be inferred from this that the specific heat a t constant volume depends only on temperature and not on volume. There is in fact a considerable amount of experimental evidence which points to the contrary. The approximate equality of c1 and c2 for substances which undergo dissociation with rise of temperature can be explained by the assumption that the influence of temperature on the degree of dissociation is the same for both liquid and saturated vapour.M. S. B.GENERAL AND PHYSICAL CHEMISTRY. ii. 231 Specific Heat of Graphitic Carbon and Coke. P. SCH~PFER and P. DEBRUNNER (Helv. Chim. Acta 1924,7,31-58).-The specific heat of carbon in the form of (a) Ceylon graphite containing 0.35% of ash; (b) retort graphite practically free from ash; (c) carefully purified beech charcoal; (d) a number of samples of gas coke was determined over the temperature range 20-1200" by making use of a mixing calorimeter the details of which are described and illustrated. It is found t'hat all these varieties of carbon have the same specific heat over the temperature range examined. The mean specific heat between 0" and to is given by the formula cat = 0.1517 + 3.9043 . lO-*t - 2.1908 . 10-7t2 + 0.596 .10-11t3 +- 2.8850. 10-14t4. The specific heats of gold and cast quartz were also determined since containers made of these substances were used for holding the samples in the calorimeter. The mean specific heat of gold between 20" and to is 0.0308 + 3.481 . 10-6(t - 20) and of cast quartz between 0" and t3$ 0.18324 + 0.10321 . lO-3t - 0.13210 . 10-7t2 - 0.99987. 10-11t3 - 0.889275 . 10-14t4. S. LEES (Phil. Mag. 1924 [vi] 47 431-446).-A new empirical equation of state is derived of the form ( p + ap1/3/TV513)( V - b ) = R T the correction term for p being made a function of three variables. It is claimed that the agreement with experiment on certain points is closer than with previously proposed formule Simple Derivation of van der Waals' Vapour Pressure Equation with a Note on Molecular Diameters.S. P. OWEN (Proc. Durham Phil. Soc. 1922-23 6 308-311).-The van der Waals vapour pressure equation log pC/p = B,/R x (6 - 0)/6 (in which B is a constant p the vapour pressure at temperature 6 pc the critical pressure and 8 the critical temperature) is deduced from kinetic considerations using the Maxwell law of distribution of velocities. It is assumed that all the molecules striking the surface of the liquid are absorbed whilst the only molecules which escape from the liquid are those which have a kinetic energy greater than a critical value 8mu,2 which is assumed to be proportional to If the critical kinetic energy is taken to be equal to the cross- sectional area of the molecule multiplied by the surface tension then by the aid of the Eotvos equation for the variation of molecular surface energy with temperature it is shown that the diameter of the molecule is given by d = djB(M~)~/~/O.4343?rk where f = B/B log,,e and I% = 2.12 is the temperature coefficient of molecular surface energy.The molecular diameters for seven substances calculated from this formula are in substantial agreement with the values Critical Constants of Various Gases. S. F. PICRERING ( J . Physical Chem. 1924 28 97-124).-An account is given of the methods used by different investigators in determining the critical constants of a large number of gases including oxygen ozone air nitrogen nitric and nitrous oxides ammonia hydrogen carbon E. H. R. Proposed Empirical Equation of State for Fluids. S. B. 6 - 6'.deduced from viscosity limiting density etc. s. s.ii. 232 ABSTRACTS OF CHEMICAL PAPERS. monoxide and dioxide the inert gases chlorine the halogen hydrides saturated and unsaturated hydrocarbons of low molecular weight methyl and ethyl chlorides carbonyl chloride sulphur dioxide and hydrogen sulphide. A selected list of these constants has been compiled and reasons are given for the choice. Those values about which there is still cbnsiderable uncertainty are indicated. M. s. B. Degradation of the Ideal Gaseous State and the [Mean] Free Path. E. SCHODINGER (Physikal. Z. 1924 25 4145).-In the theories of the degradation of a gas from the ideal gaseous state (which occurs for a perfect gas in the neighbourhood of the absolute zero) the characteristic temperature 0 a t which the gas begins to depart from the gas laws is given by the relation O = h2/8m12k where h is the Planck constant and m the mass of the molecule; I has the dimensions of length and in the different theories has different meanings.These theories lead to values of 0 in the neighbourhood of 1 OK which conflict with the experimental equation of state. The author develops a theory in which the distance I is taken as the mean free path of the molecule A. It is shown that o for a gas under normal conditions is of the order of 1/10,000" K . This does not conflict with the experimental data for the equations of state and the thermal behaviour of gases. For the electrons within a metal on the other hand 0 has a value between 4500 and 18,000" K. Thus at a moderate temperature the electrons in a metal will be degraded from the ideal gaseous state.Connexion between van der Waals' Equation and Trouton's Rule. A. BRANDT (Ann. Physik 1924 [iv] 73 403405).-The author deduces a value for the ratio of the molecular heat of vaporis- ntion to the absolute boiling point using van der Waals' equation and Guldberg's relationship T/Tc = 0.6 where T and Tc are the boiling point and critical temperature respectively. The value found however is only half that given by Trouton's rule. M. S. B. W. E. G. Cohesion Pressure. A. BRANDT (Ann. Physik 1924 [iv] 73 415-425).-Van der Waals' equation is discussed with reference to the view first put forward by E. and U. Duhring in 1886 that all substances liquid and gaseous exhibit association the extent of which depends on the pressure temperature and volume and may become zero.The negative value found by Amagat for the cohesion pressure of hydrogen is explained as being due to the increased association of the molecules which results when the volume is increased at high pressures. Latent Heat of Evaporation and Pressure of Saturated Vapour at very low Temperatures. A. BRANDT (Ann. Physik 1924 [iv] 73 406-40S).-A criticism of two papers by Aries (Compt. rend. 1917 164 343; A 1917 ii 192). Molecular Association of Liquids and Highly Compressed Gases. 111. Molecular Association of Liquids. E. E. WALKER (Phi2. Mag. 1924 [vi] 47 513-525).-A discussion of the M. S. B. M. S. B.GENERAL AND PHYSICAL CHEMISTRY. ii. 233 theory of corresponding states leads to the conclusion that only those substances the molecules of which have the same mean collision area are truly corresponding substances.The modified law is applied to associated liquids after making the corrections indicated by the author (cf. this vol. ii 144) for the effect of mole- cular association.on the critical constants. The association factors of twenty-four substances at the boiling point have been determined from selected experimental data. The results show no anomalies and calculations based on temperature-vapour pressure relation- ships a.gree substantially with those calciilated from reduced volumes and apparent reduced temperatures. Normal paraffins are used as comparison substances since they show the least (if any) evidence of association.The data for isomeric esters indicate that the acetates are more highly associated than the formates propionates or butyrates and it is suggested that the methyl group exerts a specific Suence. The degree of association is in all cases greater at the boiling point than at the critical point. Relation between the Fluidity and the Temperature of Liquids. H. J. M. CREIGHTON (Proc. Nova Xcotiun Inst. Sci. 1918-22 15 16&168).-The author finds that the absolute temperatures at which two liquids e.g. methyl alcohol and water or octane and benzene have the same fluidity (reciprocal of viscosity) are connected by an equation which is similar to the isobaric temper- ature formula of Ramsay and Young for substances which are not closeIy related (Phil. Mag. 1885 20 515; 1886 21 33).This relation may be expressed in the form R’ = R + c(t’ - t ) where R’ is the ratio of the absolute temperatures corresponding with two values of the fluidity for one of the two liquids and I2 is the corre- sponding ratio for the second liquid. The term (t‘ - t ) is the temperature difference for one of the liquids concerned. A. E. M. J. S. G. T. Viscosities of Liquids experimentally correlated to Pendulum Dampings. E. H. BARTON and H. M. BROWNING (Phil. Mag. 1924 [vi] 47,495-500).-The appearance of some of the exponentially-damped vibrations obtained with a lathe and bob pendulum (Phil. Mug. 1923 46 399406) suggested that these damping effects might be utilised for comparative measurements of the viscosities of liquids. A method has been devised for the determination in which the pendulum carries a pair of planes dipping into the liquid under examination and moving parallel to and at memured distances from the inner side of the containing vessel.Mixtures of glyeerol and water were used in standardising the apparatus which may be modified so as to be suitable for oils or for liquids with viscosities as low as that of water. Binary and Ternary Molybdenum Alloys. DREIBHOLZ (2. physikal. Chem. 1924 108 1-5O).-Binary systems of molybdenum with copper silver gold platinum nickel and cobalt respectively and the ternary systems copper-nickel-molybdenum and copper-cobalt-molybdenum have been investigated. Copper A. E. M.ii. 234 ABS!l!RAC!CS OF CHEMICAL PAPERS. silver and gold form no alloys with molybdenum. Platinum is capable of forming alloys with .molybdenum and in the neighbour- hood of its melting point takes up a t least 16% Mo into solid solution.On lowering the temperature however the molybdenum separates from the solid solution. Baar's diagram for molybdenum- nickel (A. 1911 ii 611) is confirmed and has been amplified in some respects. The Brine11 hardness of a series of mixed crystals of molybdenum and nickel increases with increasing molybdenum content. The miscibility gap in the system molybdenum-copper is closed by the addition of nickel a t the composition 35.5% Cu 15.5% Mo 49% Ni. Copper-nickel alloys are not improved in their electrical properties by tJhe addition of molybdenum. Numerous phase diagrams and photographs of sections of the alloys are included in the paper.J. F. S. Piezochemical Studies. XXIV. Experimental Determin- ation of the Fictitious Volume Change in Solution Equilibria. 11. E. COHEN W. A. T. DE MEESTER and A. L. T. MOESVELD (2. physikd. Chern. 1924 108 103-117; cf. A 1923 ii 746).- Making use of methods previously published (loc. cit.) the authors have determined experimentally the fictitious volume change in solution equilibria for several systems. The following values are recorded for naphthalene-tetrachloroethane Av = 0.1313 c.c./g. ; benzene-naphthalene 0.128 c. c. /g. ; ethyl alcohol-p-nitrotoluene 0.089 c.c./g. ; ethyl acetate-resorcinol 0.011 c.c./g. The influence of constitution on the value of Av is discussed (cf. Bridgeman Phys. Rev. 1915 6 94; Block A. 1912 ii 128). J. I?.S. Fractional Distillation of Small Quantities. G. WIDMER (HeZu. Chim. Acta 1924 7 59-61).-A fractionating column for small quantities of liquid (about 20 c.c.) consists of three concentric tubes. The vapours pass upwards between the outside and middle tubes then downwards between the middle and inside tubes and finally up the middle tube which contains a glass rod on which is wound a glass spiral. From the top of this tube the vapours pass to the side tube. All the tubes return condensed liquid to the flask. The whole is surrounded by a wide glass tube to protect it from temperature fluctuations. At the ordinary pressure tlhe apparatus can be used up to 170" and at low pressures up to 190". A mixture containing 10 C.C. of ethylaniline (b. p. 204") and 10 C.C. of diethylan- iline (b. p.213") was separated with this apparatus in 22 hours into 8-6 C.C. of ethylaniline 2.0 C.C. of a middle fraction and the rest diethylaniline. For still smaller quantities of liquid use is made of an apparatus fitted only with the inner tube the glass rod and the spiral. E. H. R. Electrical Resistance of Phenol-Water Systems near the Critical Solution Temperature. M. H. FISCHER (KoZZoid-2. 1924 34 97-99).-As a preliminary to work on soap solutions measurements have been made of the electrical resistance of mixtures of phenol and water above and below the critical solution temperature. S. B.GENERAL AND PHYSICAL CHEMISTRY. ii. 235 The Equilibrium Hydrogen-ion-Methyl Alcohol-Water. H. GOLDSCEMIDT and P. DAHLL (2. physikaZ. Chem. 1924 108 12 1-127) .-The electrical conductivity of hydrogen chloride in absolute methyl alcohol and aqueous methyl alcohol containing 0-01 to 3 moles of water per litre has been determined a t 25".The equivalent conductivity in absolute methyl alcohol a t infinite dilution a t 25" is found to be 204.2. The conductivity decreases with increasing concentration of water reaching the value 108.4 in 3N-solution. The equivalent conductivity in solutions up to 0.5 mole per litre is given by the formula h = 204.2 - 98-9n(r + n) where n is the number of moles of water per litre and r is the constant for the equilibrium bet,ween hydrogen-ion methyl alcohol and water. The value of r is 0.235; this value is in satisfactory agree- ment with the value 0.21 determined from esterification experiments.Determination of the Equilibria involving Calcium- Hydrogen- Carbonate- Bicarbonate- and Primary Secondary and Tertiary Phosphate-ions. I. N. KUGELMASS and A. T. SHOHL ( J . Biol. Chem. 1924 58 649-666).-1n con- nexion with the physiology of bone calcification and of nervous irritability determinations have been made of a number of constants which are characteristic for the conditions of ionic equilibrium in the blood at 38". For the calcium-carbonate-bicarbonate equi- librium both in presence and in absence of sodium bicarbonate [Ca"][HC0,']2/[H2C0,] = 4-14 & 0.14 x and [Ca"][HC03']/[H'] = 133 & 3; for the equilibria in which phosphate plays a part the carbon dioxide pressure being variable [Ca"J[HPO,"] = 67 & 7 x 10-8 [Ca*g]2[HC0,']2[HP04'']/[H~][H2P0,'] = 4.0 rt 0-4 x lO-3and [Ca"]2[HCO,'][HPO4'']/[H'] = 7.6 & 0-6 x These and other derived equilibrium constants are compared with previously determined data.G. M. B. Physico-chemical Basis of Vital Permeability. IV. Capillary Activity of Oxygen and of Carbon Dioxide at the Interface Light Petroleum-Water. R. BRINKMAN and A. VON SZENT-GYORGYI (Biochem. Z. 1924 144 47-51).-The interfacial tension between light petroleum and water is lowered by oxygen carbon dioxide or ammonia although oxygen and ammonia have no effect on the surface tension a t the interface water-air. Import- ance is attached to this observation in its bearing on tlhe selective permeability of the cell membrane. J. P. C. E. DAVIS H. M. SALISBURY and M. T. HARVEY (Id. Eng.Chem. 1924,16,161-163). -The variation of the surface tension of gelatin solutions with concentration temperature hydrogen-ion concentration and age of solution has been studied by Morgan's drop weight method. The concentration of the solutions could be raised to nearly 10% if the temperature was kept above 38". Further evidence of a transition point at this temperature was found by these measure- ments the drop weight-temperature curves showing a marked J. P. S. Surface Tension of Gelatin Solutions.ii. 236 ABSTRACTS OB CHEMICAL PAPER-8. change of s l o p at this point. Increasing concentration cause8 G decrease in the drop weight. In the more concentrated solutions the drop weight increases with rise of temperature but a maximum is reached and then the drop weight falls as the temperature ap- proaches the transition point.Above this temperature the drop weight remains nearly constant. As the pE value increases gelatin solutions of all concentrations afford evidence of a minimum value of the drop weight at the neutral point. The drop weight changes slightly with the age of the solution. L. LASCARY (KoEZoid-Z. 1924 34 73-83) .-The surface tension of solutions of sodium salts of thirteen of the fatty acid series between formic and cerotic acids has been determined by the drop number method. The undissociated salt and fatty acid molecules have considerable influence in diminishing the surface tension of the solution whilst the colloids present only affect it secondarily through adsorption of ions and molecules. The surface tension-concentration curves for the salts ranging from sodium caprate to laurate show a break or inflexion at the concentration at which the solution becomes turbid.This is correlated with a corresponding break in the concentration of fatty acid molecules in solution. The effect of these salts on surface tension increases with molecular weight to sodium myristate above which it decreases. Theory of Colloid Phenomena. U. R. EVANS and L. L. BIRCUMSHAW (Kolloid-Z. 1024 34 65-72).-A discussion of the forces determining colloid phenomena. I n addition t o surface tension there must be forces which act at right angles to the surface of the colloidal particles. These are treated as being electrical in nature. The idea of the selective adsorption of ions is used to explain coagulation peptisation etc.S. B. Formation of Colloids by the Electrolysis of Dilute Solutions. P. N. PAVLOV (Kolloid-Z. 1924 34 100-102) .- Observations have been made on the formation of colloidal particles by electrode disintegration during the electrolysis of water and dilute salt solutions with E.M.P.’s ranging from 2 to 16 volts. Both electrodes were observed to emit streams of particles which showed electrophoresis phenomena. The particles from the cathode are metallic those from the anode are composed of the oxide of the metal. Experiments were made with electrodes of silver iron gold bismuth and copper The cathodic disintegration investigated in these experiments appears to be of the same character as that which occurs in the production of colloids by the arc discharge under water.S. B. Viscosimetric Investigations on Lyophilic Colloids. W. R. HESS (Rec. trav. chim. 1923,42 1097-1103).-A reply to de Jong’s criticisms (A. 1923 ii 132) of bhe author’s work on the viscosity of lyophilic colloids particularly of blood. It ig shown that de Jong has misunderstood the nature of the author’s work and has mis- H. C. R. Surface Tension of Soap Solutions. S. B.GENERAL AND PHYSICAL CHEMISTRY. ii. 237 interpreted the results. A short re'aume' is given of the essential points of the work which deals mainly with the influence of pressure on viscosity. J. F. S. 0. F. TOWER ( J . Physical Chern. 1924 28 176-178; cf. Tower and Cooke A. 2922 ii 853).-The most favourable conditions for obtaining nickel hydroxide gel by the action of N/1 alcoholic solution of potassium hydroxide on a glycerol solution of nickel acetate have been studied in detail.Additional proof was obtained that the peptisation of nickel hydroxide in the formation of a colloidal solution is due to the presence of traces of potassium chloride in the solution. Experi- ments were also carried out to determine whether nickel hydroxide as it is ordinarily precipitated from solution is really Ni(OH) or NiO with adsorbed water. Samples of the precipitate from both hot and cold solutions were air-dried pulverised and left in a vacuum desiccator over sulphuric acid of increasing concentration until the weight became constant. The percentage of nickel was determined electrolytically and indicated that t'he final stable product was Ni(OH) and not the oxide or any intermediate hydroxide.31. S. B. Colloidal Silica. F. DIENERT and F. WANDENBULCKE.-(See ii 253.) Effect of the Concentration of Colloidal Clay upon its Hydrogen-ion Concentration. R. BRADFIELD (J. Physical Chem. 1924 28 170-175).-The colloidal clay separated from soils has an acid react'ion whilst the non-colloidal portion is neutral. If the former is extracted diluted to different concentrations with water and the hydrogen-ion concentration determined by the hydrogen electrode the pR value is found to vary with concentration in a similar manner to acetic acid. This is regarded as confirming the opinion that the acidity of soils is due to a true acid and not to the preferential adsorption of OH ions by soil colloids.[Cf. R. 1924 269.1 M. S. B. Colloidal Nickel Hydroxide. Technique of Ultra-filtration. Preparation of Membranes with Uniform Distribution of Pores. F. E. BARTELL and M. VAN Loo ( J . Physical Chem. 1924 28 161-165; cf. Bartell and Carpenter A. 1923 ii 221 298).- During the drying and setting of collodion membranes containing volatile and non-volatile constituents and also visible particles in suspension the cooling of the surface by evaporation of the volatile portion gives rise to vortex action; at first there is a large number of small vortices of varying strengths but later a more uniform arrangement of a small number of more powerful ones of similar strength is reached By the mutual thrust of these on one another a condition of closest packing results Le.each one is in contact with six others and the surface appears to be made up of hexagonal cells. The centre of each vortex or cell forms a pore a t first con- C. WHA.-(See i 458.)ii. 238 ABSTRACTS OF CHEMICAL PAPERS. taining liquid with a high concentration of volatile matter. By adding water after a certain interval of time the volatile constituent is removed and an opening or pore is left. The size of the pore and therefore also the permeability of the membrane depends on the amount of volatile matter left when the water is added and hence on the time the vortex action. is allowed to continue. The longer the time the smaller is tlhe pore. G. C. SCHMIDT and F. DURAU (2. physikal. Chern. 1924 108 128-150).-Experiments have been made on the adsorptive properties of sheet glass and glass powders with the object of determining whether the adsorbed substances form a single layer or a series of superposed layers.The solubility of glass both massive and powdered in a solution of sodium hydroxide N / 2 and sodium carbonate N was determined and the surface of the glass powder was measured by the Wartenberg-Wolff method (2. anqew. Chem. 1922 35 138). It is shown that the amount of glass dissolved whether massive or powdered is independent of the rate of stirring the solution and the concentration of the solvent; it increases with the time and also with the temperature a t first slowly and then rapidly. The similarity of the dissolution processes is indicated by the validity of the equation ulo=b/x where a is the loss of weight of a sheet of glass o its surface b the loss of weight of a powdered glass and x its surface.If the dyes examined have a cubic molecular structure the results of the experiments indicate that two adsorption layers are formed but if the molecule has a parallelepiped structure and if the narrower face lies on the glass there is only one adsorption layer. J. P. S. M. S. B. Adsorption. Adsorption as a Preliminary to Chemical Combination and Adsorption Measurements on Zirconium Oxide Gels. -E. WEDEKIND and H. WILKE (Kolloid-Z. 1924 34 83-96).- The adsorption of arsenious acid by zirconium oxide gels is normal and reversible equilibrium being attained in a few hours; the amount of arsenic acid removed from solution continues to increase for some days and only part of the acid can be recovered from the gel by washing.It is suggested that the zirconium oxide and the adsorbed arsenic acid undergo slow chemical combination and a formula Zr(HAsO,) is ascribed to the compound. R61e of Adsorption in the Determination of Residual Nitrogen. J. H. CascAo DE ANcLaco.-(See ii 273.) S. B. Hydration of Natural and Artificial Glasses. G. SCHOTT and G. LINCK (Kolloid-Z. 1924 34 113-116).-Measurements have been made of the rate a t which obsidian pitchstone and artificial glasses take up water when the finely powdered substances are heated with water at about 250" and 35 atm. pressure. A Jena borosilicate glass for example absorbed more than 7% of water in 3 weeks. The results suggest that the process of taking up water is not usually one of chemical combination but of imbibition by aGENERAL AND PHYSIUAL CHEMISTRY.ii. 239 gel. The view that the obsidians contain chemically combined water is supported by observations on the dehydration of these substances. S. B. Rhythmic Banding of Precipitates (Liesegang’s Rings). H. MCGUIGAN and G. A. BROTJGH ( J . Biol. Chem. 1923 58 415- 423) .-Mainly a reiteration of McGuigan’s theory of the Liesegang phenomenon (A. 1922 ii 38). Experiments most of which are taken from the literature are adduced to show that ring formation occurs in the absence of a gel provided that the factors which produce turbulence or assist diffusion are reduced to a minimum. A gel when i t is the medium in which precipitation takes place further fixes the precipitate a t the position of formation.Rhythmic Crystallisation of Undecoic Acid. W. E. GARNER and P. C. RANDALL ( J . Chem. Soc. 1924 125 369-372).-The rhythmic crystallisation of thin films of undecoic acid has been examined microscopically by means of ordinary and polarised light. The troughs and crests of the wave forms are permanent and the distance between the crests varies with the thickness of the film. Rhythmic forms of the same kind are obtained when the glass surface of the microscope slide is replaced by a silver plate or by the crystal faces of selenite calcite or quartz. The velocity of crystallisation diminishes with rise of temperature and observations near the melting point showed that a solid skin possessing a greater area than the glass surface is produced on the surface of the acid which is still in the liquid condition. This skin wrinkles after the passage of the wave of crystallisation and gives rise thereby to the rhythmic structure.The formation of the solid skin is attributed to differences between the solidification temperatures of the acid at the liquid-air and liquid-glass interfaces. Other acids e.g. myristic lauric decoic etc. exhibit the pheno- menon. S. K. T. E. S. Absorption of Gases by Colloidal Oxides and the Mode of Action of Electric Accumula+,ors. P. BARY (Rev. Gdn. Coll. 1924 2 33-37).-The mechanism involved in the production of current by lead storage batteries and by accumulators of the Edison and the Lalande-Chaperon types is considered to be analogous to that of the gas cells described by Cailletet and Collardeau (Compt.rend. 1894 119 830). Calorific Action of Radiation on Metals immersed in Solutions of their Salts. G. ATHANASIU (Cmpt. rend. 1924 178 561-563).-The author has investigated the effect of illu- minating by meam of a quartz mercury arc one of two metal plates immersed in a solution of a salt of the metal. With mercury lead and platinum the E.M.P. generated between the two surfaces may be entirely ascribed to the heating effect of the radiation whilst with silver copper cadmium and zinc a secondary effect due to chemical alteration of the illuminated surface is observed. E. E. T. J. S. G. T.ii. 240 ABSTRACTS OF CHEMICAL PAPEM. Measurement of SmalI Quantities of Heat. Use of the Compensating Microcalorimeter. A. TIAN (Cmpt. rend. 1924,178 705-707).-An apparatus is described which the aubhor has designed for the measurement of the relatively small heat changes which are associated with enzyme and similar reactions.It consists of a thermally insulated vessel serving as the calorimeter immersed in a constant-temperature bath. The two vessels are connected by a t(hermopi1e and a galvanometer which serve to detect and measure the variation in temperature difference between the two. Further details are given for which the original should be consulted. E. E. T. F. KIRCHHOF (Chem.- Ztg. 1924,48,113-114).-For saturated hydrocarbons from hexane onwards the heat of combustion is expressed by the formula H=53,000(2n+2)+50,50O(n-l) the value 53,000 corresponding with the C*H linking and the value 50,500 with the C*C linking.For the lower members the values are somewhat higher in con- formity with their gaseous condition. Using these values for the unsaturated series the C:C bond has values varying from 2.0 to 2.3 times the value of the single bond the CiC linking values from 3.3 to 4.0 times this value. For cyclic hydrocarbons the single C*C linking has the value 63,000 cal. for cyclopropane falling to 51,160 for cydohexane. In the benzene series taking the number of C*C linkings in benzene itself as 9 the value for this is 51,660 cal. and in the polycyclic benzene compounds counting the number of linkings in the same way the value remains approximately constant. From a consideration of the heats of combustion of the hydrogen- ated benzenes the prism formula of Ladenburg is favoured reduc- tion to cydohexadiene and cyclohexene leading to formation of cydobutane and cyclopentane rings respectively.Incomplete combustion of benzene yields amorphous carbon which should therefore possess the same type of C*C linking; since the molecule of the latter contains about 1000 atoms the number of C*C linkings may be taken as double the number of atoms and as the heat of combustion per g. atom of amorphous carbon is 96,960 the value for a C*C linking is 48,480 cal. which is considered as a limiting value for benzene derivatives in the solid state of aggregation. Constitution and Heat of Combustion. s. I. L. Thermodynamic Surface of Water. A. BRANDT (Ann. Physik 1924 [iv] 73,409-411).-Diagrams are given to show the relation- ship between pressure temperature and volume in the equilibria which characterise the behaviour of water as a one-component system.M. S. B. Theory of the Velocity of Chemical Reactions. N. BJERRUM (2. physikal. Chem. 1924 108 82-100).-The author discusses Bronsteds formula (A. 1922 ii 699) for the influence of salts on the velocity of chemical change. In a new method of deducing Brijnsted's formula the assumption is made that the velocity is proportional to the concentration of a complex intermediate sub-QHNERAL AND PHYSICAL CHEMISTRY. ii. 241 stancu forrncd by collision between the molecules of the reactants. The conclusion is drawn that the concentrations of the reacting substances cannot be replaced by either their osmotic pressures or their aotivities. J.F. S. [Theory of the Assimilation of Carbon Dioxide.] 0. WARBTJRG and E. NEGELEIN (2. physikal. Chern. 1924,108 101- 102).-A criticism of the interpretation given to the authors' work by Weigert (A. 1923 ii 1271) in the development of a theory relating to the assimilation of carbon dioxide. J. F. S. Gas Kinetics. C. S. LIND (J. Physical Chem. 1924 28 65-58) .-The reaction of hydrogen with bromine is photosensitive at 250". The reaction velocity under arc lamp illumination is twelve times that under the dark thermal conditions employed by Bodenstein and Lind (A 1907 ii 76) and the experiment affords quantitative confirmation of the results obtained by Kastle and h t t y (A. 1898 ii 214). Calculation shows that the observed reaction velocity does not exceed the value required by photo- chemical equivalence and also that the Nernst hypothesis (A.1919 ii 208) is not violated by the reaction. Discussion of the mechanism of this reaction and of that of hydrogen and chlorine affords support for the view of Nernst and Noddack (A. 1923 ii 526) that photochemical partial reactions are not amenable to thermodynamic treatment. The multicyclical theory of Nernst seems to offer the most reasonable explanation. I n view of the results of Bodenstein and Dux (A. 1913 ii 1039) it is held that the negative results of Amato (Kastle and Beatty Zoc. cit.) for the reaction of hydrogen with chlorine a t - 12" may be due to the presence of oxygen or some other inhibitor. It is suggested that inhibitors for thermal and photochemical reactions are identical in nature and that the mechanism is the same in both cases.Attempts to arrest the propagation of explosion in mixtures of hydrogen and oxygen by means of an electric field gave negative results. A. E. M. Velocity of Oxidation of Nitric Oxide and its Importance in Nitrogen Fixation. G. W. TODD (Proc. Durham Phil. SOC. 1922-23 6 291-307).-The equations derived by the author (A. 1918 ii 102 190) are discussed with reference to t,he experi- mental data obtained by Lunge and Berl (A. 1907 ii 863) for the reactions 2NO+02=2N0 and 2NO+02=N,0,. It is shown that these data give sensibly constant values for the velocity coefficient a t 20" when pure oxygen is added to nitric oxide but with air as oxidising agent higher values of the coefficient are found in the earlier stages of the reaction.This is ascribed to the time required for mixing the gases and a mean value for the velocity coefficient is calculated from the data for later stages in the reaction. This value is used to derive the optimum conditions for the chemical change in question in the ammonia oxidation and atro procesl3es for the fixation of nitrogen. s. s.ii. 242 ABSTRACTS OF CHEMICAL PAPERS. Starch-Iodine Reaction. N. R. DHAR ( J . Physical C'hem. 1924 28 12&130).-The nature of the starch-iodine reaction has been explained on the one hand by assuming the formation of a definite chemical compound to which thirteen dif€erent formulae have been ascribed and on the other by the assumption that the blue colour is due to the adsorption of iodine by colloidal starch.The author considers that adsorption plays an important part in the formation of the blue substance. When an alcoholic iodine solution is added to an aqueous starch solution the electrical conductivity of the resulting solution is dehitely greater than the sum of the conductivities of the separate constituents. It is suggested that a colloidal adsorption product is formed which conducts by means of micellar ions. It behaves rather like an unstable iodide and is readily destroyed by oxidising agents. The blue colour is immediately developed on the addition of freshly prepared alcoholic iodine solution to starch paste or colloidal starch and in these circumstances the iodide ion can only be formed by hydrolysis of iodine which reaction being non-ionic is probably slow.The effect of added electrolytes appears to consist in the intensi- fication of the blue colour in consequence of the coagulation of the colloid. M. S. B. Simplicity of Mechanism of Reaction as One of the Factors Conditioning Catalysis. E. R. BULLOCK ( J . PhysicaZ Chem. 1924 28 179-181).-The reaction between silver in gelatin and a solution containing potassium dichromate and a bromide slightly acidified with acetic acid only occurs in presence of potassium ferricyanide or copper sulphate which appear to act as catalysts. The author accounts for the action of the catalyst by the greater simplicity of the initial ionic reaction Fe(CN),"'+Ag =Ago+ Fe(CN) '"' or Ag+Cu"=Ag'+Cu* as compared with SAg+Cr"'"'= 3Ag'+&"' which involves four ions instead of two.Experiment also showed that 0.0001N-ferric sulphate increases by ten times the rate of " bleaching " of a silver image by O.05N-bromic acid and simpmcation of the mechanism is again suggested as the explanation. The only other known instance of a catalysed reaction of this type is afforded by the reaction between zinc and a dilute solution of potassium permanganate and sulphuric acid in presence of a trace of pure nitric acid. This may be explained by the relative simplicity of the reaction Zn+N""'=Zn**+N'". It is admitted however that further instances should be sought before the prin- ciple is accepted as valid. Negative Catalysis. J. A. CRRISTLANSEN ( J . Physical Chem. 1924 28 145-148).-The author is of opinion that Taylor's explanation (A. 1923 ii 399) of negative catalysis in homogeneous systems namely that the formation of molecular compounds with the negative catalyst leads to a reduction in the number of the reactive molecules cannot be true in all cases for the proportion of such non-reactive compounds must be very small where the number of molecules of catalyst necessary is only a small fraction M.S. B.GENERAL AND PHYSICAL CHEMISTRY ii. 243 of the total number of molecules of reacting gases. The molecules of the reaction products possess immediately after the reaction an amount of available energy which is greatly in excess of the mean energy a t the temperature considered since not only do they contain the critical energy necessary for the reaction to occur but often also the kinetic or potential energy which is represented by the heat evolved in the reaction.These molecules are able to activate more molecules of the reactants by collision and these in turn produce fresh activators as in the so-called " chain reactions " (Bodenstein 2. physikal. Chem. 1913 85 346; A. 1916 ii 422). The author suggests that in such cases the negative catalyst acts by taking up the energy from the active molecules and thus breaking the chain (Gohring A. 1922 ii 9). This theory indicates a method by which chain reactions may be detected. They are reactions which can be inhibited by a small quantity of a foreign substance. There are other instances of negative catalysis where the mechanism must be different. For example the decomposition of oxalic acid by sulphuric acid is retarded by the presence of small quantities of water probably in consequence of the displace- ment of the equilibrium H2S0,~H2O+SO3 since addition of small quantities of SO greatly increases the rate of reaction.M. S. B. Catalytic Hydrogenation of Organic Compounds with Common Metals at the Temperature of the Laboratory. VI. Influence of Oxygen on Nickel Catalysts. Influence of Water on Nickel Catalysts at Elevated Temperatures Temperature of Reduction and Activity. C. KELBER (Ber. 1024 57 [B] 136-141).-1t has been shown previously (Kelber A. 1921 ji 630) that contrary to the statement of Willstatter and Wald- Schmidt-Leitz (A. 1921 ii 187) the presence of oxygen is not necessary for hydrogenation in the presence of metallic nickel. A comparison of the behaviour of the pure metal with that of the metal deposited on a carrier shows that the latter catalyst recovers its activity more rapidly than the pure metal when after inactiv- ation has been induced by relatively short contact with oxygen the two forms are placed in hydrogen.After prolonged contact with oxygen at 20" both types of catalyst regain activity more slowly in the presence of hydrogen and do not become fully active after many hours. Catalytic activity seems to be destroyed permanently by exposure of the catalysts to oxygen a t 60-70". The nickel catalyst deposited on a carrier suffers a marked loss in activity when its suspension in water is heated in an atmosphere of hydrogen whereby it is caused to react with the water yielding hydrogen and apparently nickelous oxide which forms a coating over the unchanged metal.Under similar conditions the pure nickel catalyst does not become oxidised or suffer loss in catalytic activity. In contrast with the pure nickel catalyst (cf. Kelber A. 1916 ii 309) the metal deposited on a carrier exhibits an optimumii. 244 A~S'I!RACTS OF CHBBf1CA.L PAPERS. activity when prepared at 350° but is not completely inactive when reduction is effected a t 1000". On the other hand pure nickel prepared a t 550° is practically inactive. Catalytic Hydrogenation of Organic Compounds with Common Metals at the Temperature of the Laboratory. VII. Hydrogenations with Elementary Nickel in the Absence of Oxygen. C. KELBER (Ber. 1924 57 [B] 142-l43).-The possibility of effecting catalytic hydrogenation with the aid of nickel in the absence of oxygen (cf.Kelber A 1921 ii 630; Willstatter and Waldschmidt-Leitz A. 1921 ii 187 ; Willstatter and Seitz A. 1923 i 771) is established by the observations that diphenyldiacetylene and azobenzene dissolved in hexane are very readily reduced in the presence of nickel prepared from nickel cyanide. The activity of the catalyst thus obtained resembles that of the product derived from nickel carbonate since the opti- mum results are secured when reduction is effected a t about 300" and less active specimens are formed a t higher temperatures. Support is thus afforded for the author's view that loss in activity is due to surface changes in the metal. Photolysis and the Law of the Photochemical Equivalent. VOLMAR (Compt.rend. 1924,178 697-700).-The wave-lengths of the radiant energy which is responsible for the photolysis of ketones aldehydes and aliphatic acids have been calculated from the law of photochemical equivalence in the form & ~ = 3 x 104 (where & is the energy in cals. required to effect photolysis). Indirect methods were used to calculate &. It is found that h=0*27p for ketones 0.28p for aldehydes and 0.21,~ for aliphatic monocarboxylic acids these values being in good agreement with experiment. It is further shown that the active wave-length is the same for dXerent compounds which contain the same group and that this wave- length corresponds with the position of maximum absorption. H. W. H. W. E. E. T. Photographic Action of Canal Rays. M. JAKOBSON ( A m .Physik 1924 [iv] 73,326-338 ; cf. Konigsberger and Kutschewski Physikul. Z. 1910 11 666).-By observation of the blackened traces produced on photographic plates by the action of canal rays which have passed through a combined magnetic and electric field a gas mixture can be analysed qualitatively. The author thought it might be possible by measuring the degree of blackness produced to make a quantitative analysis also. To do this it was first necessary to determine under what conditions such measurements might be regarded as trustworthy. Yerutz silver chloride-bromide diapositive plates were found to be most satisfactory. The light was found to produce only a small proportion of the blackening from a short exposure but its relative importance increases with length of exposure.When both act together however the canal rays apparently reduce the effect of the light rays still further. Three stages have been observed as the length of exposure is gradually increased. (1) A blackening is obtained which increasesQENERAL AJSD PEYSICAL CHEXISTRY. ii. 245 to a limiting value. This effect is not fundamentally Werent from that due to the action of light and it is from this that it should be possible to make quantitative determinations. (2) A white circular patch appears in the centre of the black spot comparable with t,he effect obtained by over-exposure to light. (3) Black islets appear in the white patch and these are probably crystals of silver surrounded by gelatin. For the last two stages an explanation is suggested which depends on the disintegration of the gelatin and its subsequent re-hardening under the prolonged bombardment by the canal rays.Influence of Gum Arabic on the Hydrolysis of Methyl Acetate. J. N. PEARCE and J. V. O’LEARY ( J . Physical Chem. 1924,28,51-54).-Gum arabic solutions absorb iafra-red radiation and if it is capble of emitting infra-red radiation which is absorb- able by methyl acetate then according to the radiation hypothesis gum arabic might be expected to accelerate the hydrolysis of methyl acetate. Measurements of the rate of hydrolysis of methyl acetate in presence of hydrochloric acid at 25” have ahown that the velocity of the reaction is diminished in the presence of gum arabic. The decreased velocity is shown to be due to 8 diminution in the hydrogen-ion concentration brought about by the adsorption of the hydrochloric acid by the gum.The fall in the hydrogen-ion concentration increases with the amount of gum arabic present in the solution. J. P. S. Accurate Calibration of Capillary Tubes. K. J. ISAAU and I. MASSON (J. Physical Chem. 1924 28 166-169).-The tube closed a t the top is mounted vertically in a jacket at constant temperature and the calibration is made with mercury. Horizontal calibration is untrustworthy since the meniscus sags and its volume cannot be determined. The weights of successive 2 cm. mercury threads are determined; a second calibration with intervals over- lapping those of the first set is then made and finally as a check a single “ overall ” determination. Corrections for the mercury meniscus which should always be a rising one are made by the aid of Schalkmijk’s data [Comm.Php. Lab. Leiden No. 67 (1901); Vershg. Kon. Akad. Wet. Amst. 1900 462 1901 5121. The mean calibration curve showing mean cross-section or volume per linear cm. is plotted. The aggregate volume of the tube is calcu- lated and compared with the “overall” determination. The difference which is not usually more than 1 in 6000 is distributed proportionally over the length of the tube w being probably due to a cumulative error in the meniscus correction. The accuracy of the calibration is 1 part in several thousand parts. [Substitute €or Ground Joints in Vacuum Technique. 1 G. LANDESEN (Ber. 1924 57 [B] 183).-A claim for priority against von Antropoff (A.1923 ii 850). Apparatus for Drying Gases. V. T. JACKSON ( I d . Eng. Chem. 1924 16 163).-A glass tube 45 cm. x 5 cm. is fitted with a gas inlet at the lower end and an outlet near the top. In the VOL. CSYVI. ii. 10 M. S. B. M. S . B. H. W.ii. 246 ABSTRACTS OF CHEMICAL PAPERS. bottom is placed a layer of glass wool 1 em. thick and on this glass beads to a depth of 10 cm The tube is filled to this depth with concentrated sulphuric acid. Openings are also provided for filling and emptying. If charged with 50% potassium hydroxide solution the apparatus serves for removing carbon dioxide from air. Constant-level Regulator for Water Distillation. c. I. L. JEND- RASSIK (Biochem. Z. 1924 144 285-286).-A simple constant- level apparatus for continuous water distillation is described.The. principle consists in passing the condenser water through a constant - level vessel from which a siphon tube terminating in a capillary reaches to the bottom of the still. Prevention of ‘ I Bumping ” during Vacuum Distillation. H. G. BECKER ( J . Chem. Xoc. 1924,125,460461).-‘* Bumping ” is prevented by the use of a flask inclined a t 45” which is rotated rapidly about its axis (150-200 revs. per min. for small flasks) and is connected with the rest of the apparatus by means of a well- greased ground joint Device for Gas-heated Thermostats. W. MURR-~Y (J. Chem. Xoc. 1924 125 461-462).-A device for obviating the sticking of mercury in a thermo-regulator is described. Two calcium chloride towers are arranged in U-tube form by connecting the lower tubulures and are partly filled with water.The upper end of one of these towers is closed by a cork through which pass the gas supply tube and the exit tube leading to the regulator. The ball- shaped head of a glass float fits into the supply tube and when the pressure exceeds the normal the water in the tower is driven down-and the float closes the entry tube. When the excess of gas thus trapped is used up the water rises and the process is repeated. The flame of t-he regulator pulsates in step with the float (adjusted by the quantity of water to 30 or 40 pulsations per minute) and the mercury surface in the regulator is in constant but almost imperceptible motion. S. K. T. L. E. DAWSON (Ind. Eng. Chem. 1924,16 160-161).-The device described which is capable of maintaining for days a constant difference of pressure between a partly exhausted system and the atmosphere consists of a trap connected on t.he one hand to the system under suction and on the other by two vertical tubes of different diameters to a mercury reservoir open to the atmosphere and placed below the trap.The narrower of the two connecting tubes dips less deeply into the reservoir than the wider one. Should the pressure in the trap fall mercury is drawn up into it uncovering the end of the narrower tube and permitting air to be momentarily sucked up restoring equilibrium. The lengths of the two tubes are adjusted to the pressure difference which is to be maintained and their diameters to the capacity of the air-pump used. The apparatus can readily be modified to work with a pressure greater than atmospheric. J. P. s. K. T. Automatic Pressure Regulator. c. I.INORGANIC CHEMISTRY. ii. 247 Soxhlet Extraction Apparatus. P. H. PRAUSN~TZ (2. ungew. Chem. 1924 37 50).-The usual paper extraction thimble is replaced by a filter-plate made of porous glass (cf. following abstract) which is fused into the extraction vessel and is preferably concave upwards. The rest of the apparatus is as usual except that an extra tube connecting the parts of the apparatus above and below the filter-plate is necessary for equalising the pressure or removing displaced air. The apparatus can be used for extracting with such solvents as acetic acid or 20% hydrochloric acid. Use of Filter-plates made of Sintered Glass. G. F. HUTTIC (2. angew. Chem. 1924 37 48-5O).-Glass filter-plates of various degress of porosity prepared by sintering finely powdered and sieved glass are sealed into glass crucibles filter funnels etc. Crucibles fitted with such plates are found to be preferable to Gooch crucibles for quantitative analytical work in that they filter more quickly dry more readily show less variation in weight when left exposed to the atmosphere and do not require packing with asbestos They cannot of course be heated to redness nor are they resistant to strong alkalis. Rapid Analysis of Sugars. Purification and Concentration of m y m e Solution. Apparatus for the Determination of Colour in Terms of Dominant Wave-length Purity and Brightness. I. G. PRIEST (J. Opt. Soc. Amer. 1924 8 173-200).-The author describes a monochromatic colorimeter and accessory apparatus for the determination of the dominant wave-length in heterogeneous light for the investigation of the purity of white light and the changes produced by transmission and reflection under specified conditions. Transport Apparatus. A. JANEK (Kolloid-Z. 1924 34 103-10a) .-An apparatus is described suitable for the rapid determination of the sign of the charge on a dispersoid. The colloidal solution is placed in a U-tube the ends of which are closed by colloidal membranes and the electrodes are supported above the membranes in wider tubes which are attached to the limbs of the U-tube. The movement of the particles towards one film and away from the other is observed either visually or by analysis of the solution. S. B. W. T. K. B. W. T. K. B. F. w. REYxoLDs.-(See i 464.)

ISSN:0368-1769    

DOI:10.1039/CA9242605213

出版商:RSC    

年代:1924

数据来源: RSC

摘要:

i. 236 Physiological Chemistry. The Influence of Various Carbohydrates and Amino-acids on the Blood and Urinary Sugar of the Healthy Organism. A. SCHXTTI (Biochem. Z. 1923 143 201-220).-1n general the ingestion by normal individuals of 20 g. of the following carbo- hydrates produces a hyperglyczemia the intensity of which diminishes in the order given dextrose laevulose sucrose lactose galactose. The same amounts of starch glycine and alanine produce no hyper- glycsemia but after the ingestion of 100 g. of starch the rise in blood-sugar is as great as but more delayed than that produced by 100 g . of dextrose. Peptone and coffee give no hyperglycaemia. Lsevulose is excreted more quickly and in greater amount than dextrose and the sugar excreted after ingestion of sucrose consists chiefly of lawulose.Galactose seems to be excreted largely un- changed by the kidneys. Glycosuria following ingestion of dextrose increases with increasing hyperglyczemia but the former shows a more delayed maximum than the latter. Glycosuria is not de- pendent on diuresis but is regarded as a secretory process. J. P. Blood-clotting as an Agglutination Process. E. HEKMA (Biochem. Z. 1923 143 105-110).-A theoretical paper in which the view is advanced as a result of the author's pretious work that fibrin formation is a crystallisation process involving the dehydration and agglutination of fibrinogen amicrons. Thrombin is regarded not as a fibrin ferment but as an agglutinin and fibrin consists of fibrinogen plus agglutinin. J. P.PHYSIOLOGICAL ClHEMISTRY. i.237 The Influence of Chemicals on Erythrocyte Membranes by Changes in Corpuscular Volume. B. S. NEUHAUSEN and J. E. BRESLJN (Bull. Johns Horpkim H o q . 1923 34 199-201).-A study of the effects of the salts of the blood and of dextrose on the erythrocyte membranes as shown by a comparison of cell volumes in iso-osmotic solutions. Ions cause swelling in the order C a t Na < K. The contracting action of calcium as opposed to the swelling action of potassium is specially pointed out. In the case of blood-corpuscles the membranes are very complex colloidal mixtures the swelling or precipitation of any component of which will loosen the whole structure. CHEMICAL ABSTRACTS. The R61e of Pancreatic Juice in the Digestion of Proteins. Relative Importance of Trypsin and Erepsin.E. F. TERROINE and ST. J. PRZYLECKI (Arch. internut. physsiol. 1923 20 377- 396; Ber. ges. Physiol. 1923 20 64; from Chem. Zentr. 1923 iii 1290).-Pancreatic juice not activated by kinase attacks the normal products of gastric digestion. With activation by kinase the action is increased. The relative importance of erepsin and trypsin depends on the degree of previous gastric digestion; where this is small trypsin plays the more important part but is less important where gastric digestion is more complete. A.s pancreatic digestion proceeds the effect of trypsin recedes relative to that of erepsin. G . W. R. Effect of Magnesium Sulphate and Metallic Magnesium on Metabolic Exchanges. R. REDING and A. SLOSSE (Corn$. r e d . Csoc. biol. 1923 88 -6; from Chem.Zentr. 1923 iii 1108). -Intramuscular injection of 70 C.C. of 10% magnesium sulphate solution decreases the output of total nitrogen uric acid and sodium chloride whilst the output of magnesium is increased. In a rectal carcinoma injection of 20 C.C. of 10% magnesium sulphate solution was followed by decreased output of total nitrogen amino-nitrogen and sodium chloride and increased output of uric acid and creatine. Similar effects were observed in a rabbit after introduction under the skin of a piece of metallic magnesium. G. W. R. The Internal Secretion of the Genital Organs. I. The Genital Organs and Carbohydrate Metabolism. S. TSUBURA (Biochem. Z. 1923 143 24&-290).-Castration has no effect on the blood-sugar of fasting dogs but in the sexually mature animal the operation produces a lowered sugar tolerance.The same effect is produced by ligature of one spermatic cord and extirpation of the opposite testicle by ligature of both cords or by exposing the genital glands to X-rays. These operations produce a degener- ation of the sperm-cells but do not affect the interstitial cells. The lowered sugar tolerance may be temporarily raised by trans- plantation of the sexual organs of another animal but not by feeding the organs or by injection of extracts. Glycogen formation glycogenolysis and glycolysis of the blood-sugar in vitro are some- what delayed after castration but the blood amylase is unchanged.i. 238 ABS!l!RACTS OB' CHEMICAL PAPERS. After cmtration the reactions of the blood-sugar to adrenaline diuretin and pituitrin are increased but thyroid feeding has no effect although it raises the low sugar tolerance of the castrated animals.The kidneys of the latter are more permeable to sugar but the excretion of phenolsulphonephthalein is delayed as com- pared with the normal condition. In general parallel experiments on male and female dogs gave similar results. The other endocrine organs show alterations after castration and the observed results may not be wholly due to disturbances of the internal secretory functions of the genital organs. The Internal Secretion of the Genital Organs. 11. The Genital Organs and Respiratory Gaseous Metabolism. S. TSUBURA (Biochem. Z. 1923 143 291-322).-Castration produces a lowered gaseous metabolism which becomes apparent about one month after the operation. A similar effect is produced by ligature of one spermatic cord and removal of the opposite testicle.Trans- plantation of the testes of another animal raises the gaseous exchange whilst feeding experiments are ineffective. Dogs in heat show no increased gaseous metabolism. Thyroid feeding raises the low gaseous exchange of castrated animals. The combined results of thyroid feeding on the sugar tolerance (cf. preceding abstract) and gaseous metabolism of castrated dogs are explained as being due to a moderate stimulation of the sluggish carbohydrate meta- bolism by the thyroid hormone. The Ferments in the Placenta. K. MAEDA (Biochem. Z. 1923 143 347-364).-Extracts of the placenta contain diastase in considerable amount in part derived from the parental blood.Carboxylase and small quantities of lactase and invertase are present but glycolytic enzymes are not found. Of proteolytic enzymes erepsin a weak trypsin and traces of pepsin but no rennin are found. An asparagine deaminase is present. Mono- and tri- butyrases are found and the latter being stable to both quinine and atoxyl is thus merentiated from serum lipase. Histozyme a feeble salicylase and an oxydase acting on pyrocatechol adrenaline and dihydroxyphenylalanine but not on tyrosine are present. J. P. J. P. J. P. The Site of Origin of Bilirubin. A. R. RICH (Bull. Johns Hopfins Hosp. 1923 34 321-329).-Haemoglobin introduced into the blood-stream of a dog without a liver and circulating actively for as long as five and a half hours is not transformed into bilirubin.Since however the production of bilirubin from cir- culating hzemoglobin takes place rapidly and readily in a dog with a liver through which the blood-stream passes it is concluded that the liver is necessary for the transformation of circulating haemo- globin into bilirubin. The conclusions are applicable only to hzemo- globin circulating in the blood-stream not to hemoglobin liberated into the tissues from extravasated blood ; the conversion of haemor- rhage hzemoglobin into bilirubuin (haematoidin) is a purely local phenomenon the exact mechanism of which is uiiknown. CHEMICAL ABSTRACTS.PHYSIOLOGICAL CHEMISTRY. i. 230 [Fish] Liver Oils. M. TSUJIMOTO ( J . Chem. I d . Jupun 1923 26 482486).-The oil from tlie Jewfish (Stereolepis ischinagi (mg.) contains a large amount of peculiar unsaponifiable matter (up to 50%) which is a viscous substance of reddish-orange colour having a resinous odour and showing much resemblance to colo- phony when dried at 100” for several hours.The name “liver- resin ” is proposed for it. Liver oil from Erilepk zonifer (Lock.) also contains a similar substance (about 4%). These oils and the unsaponifiable matter (especially in carbon disulphide solution) give a deep violet coloration with sulphuric acid. With a sample of 8. ischinagi Liver oil the coloration was observed to be about one hundred times as strong as that of a cod-liver oil. The Highly Unsaturated Acids in Cod-liver Oil. M. TSUJIMOTO and K. KIMURA ( J . Chem. Iyzd. Japan 1923,26,1162).- By the potassium salt-ether method and the Lithium salt-acetone method a mixture of highly unsaturated acids was obtained.By converting it into the methyl ester a fraction (about 4% of the oil) boiling a t 220-226”/5 mm. was isolated. The analysis of the ester and the fatty acid from the ester gave C,,Hs02Me and C,,HaO respectively and behenic acid was produced from the latter by hydrogenation. Cod-liver oil therefore contains a large amount of clupanodonic acid C,,H,,O,. The part of the methyl ester boiling below 220°/5 mm is supposed to contain a large amount of a highly unsaturated acid containing eighteen or twenty carbon atoms K. K. K. K. Further Studies on Autoxidations and Oxido-reduction Processes. V. E. ABDERIIALDEN and E. WERTHEIMER (Pfliiger’s Archiv 1923 200 176-193; from Chem.Zentr. 1923 iii 1290).- The reducing power of muscle in which thermal rigor has been induced is three or four times that of normal muscle. In chemical rigor and rigor mortis there is a similar increase in reducing power. An increase also occurs during normal or tetanic contraction. It is supposed that such increases are associated with the presence of a thiol group in the muscle fibres. Cysteine serves as a hydrogen carrier and acceptor. A scheme for a stable oxido-reduction system is given. G. W. R. A Methylation Function of the Thyroids and the Biological Significance of Iodine. B. STUBER A. RUSSMAN and E. A. PROEBSTINC (Biochem. Z . 1923 143 221-234) .-If guanidine- acetic acid be administered intravenously to young dogs an increase of blood creatine-creatinine follows.No such increase occurs in thyroidectomised animals but if these be given dried thyroid or potassium iodide per os before the injection of guanidineacetic acid they react like the normal animals. The blood of normal dogs but not of thyroidectomised dogs is also effective in this respect. It is concluded that the capacity of the organism to methylate guanidineacetic acid and so form creatine and creatinine is depen- dent on the integrity of the thyroids and that iodine compounds are necessary for this methylation process. J. P.1. 240 ABSTRACTS OF CHEMICAL PAPERS. The Oxytocic-Pressor-Diuretic Principle of the Infundi- bular Portion of the Pituitary Gland. J. J. ABEL C. A. ROUILLER and E. M. K. GEILPNG ( J . Pharm. exp.Ther. 1923 22 289-316).-After a purification of the gland material by precipitation with mercuric chloride phosphotungstic acid and tannic acid in succession the residue is dissolved in an alcoholic solution of tartaric acid and re-precipitated with ether. The result- ing precipitate is called the tartrate of the active principle. Its action on the isolated uterus of the guinea-pig may be as much as 1,250 times as great as that of histamine phosphate whilst it also retains the diuretic and pressor action characteristic of the pituitary extract. C. R. H. Anaerobic Processes Involved in Muscular Activity. W. HARTREE and A. V. HILL ( J . Physiol. 1923 58 127-137).-1n muscle O-lyo of lactic acid can be produced so rapidly by exercise or by stimulation that no appreciable escape of carbon dioxide or of acid is possible.Hence if the hydrogen-ion concentration inside a muscle is not to rise to an excessive degree during exercise there must be some buffer in it much more effective than a bicarbonate solution. Meyerhof (PfEGger's Archiw 1922 195 22) has empha- sised that in a frog's muscle the absolute amount of bicarbonate present as determined from the carbon dioxide driven out from the muscle by excess of acid is quite inadequate to neutralise the lactic acid liberated in severe stimulation. Even if all the carbon dioxide were driven out only one-seventh to one-tenth of the lactic acid could be neutralised. The authors further calculate that the 0.3% inorganic phosphate in muscle (a mixture of KH,PO and GHPO,) is also quite inade- quate to keep the P within the extreme limits observed viz.a change from 7.5 to 6.9). Bicarbonate and phosphate together are insufficient to neutralise half the lactic acid formed. There must be as Meyerhof assumes some other powerful buffer doubtless an alkali-protein salt. The addition of lactic acid to shed blood also produces only a fraction of the change in the hydrogen-ion which would result in a bicarbonate solution of the same coiicentration ; here also alkali protein must play a considerable part. The authors confirm the existence of a secondary maximum in the production of heat in a muscle in the absence of oxygen; this maximum occurs after about two and a half minutes (see the follow- ing abstract for its interpretation). A " balance sheet " is given of the heat evolved in the different phases of muscular contraction; it shows that during the recovery process something between 1/4*7 and 1/6 of the lactic acid removed is oxidised the rest being reconverted into glycogen.G. B. Heat of Combustion of Glycogen in Relation to Muscular Contraction. W. K. SLATER ( J . Physiol. 1923 53 163-167).- Glycogen (from MytiZus edulis a convenient source) only becomes anhydrous at 100" in a vacuum over phosphoric oxide (cf. Harden and Young T. 1902,81,1224). Dried by the method of Atkins and Wilson (T. 1915 107 916) it has the composition (C,H,,,O,,H,O),.PErYsIoLo(3IaAL (3EE&frS!rRY. i. 241 The heat of combustion seems previously t o have been determined with slightly hydrated material. The author h d s the heat of combustion of hydrated glycogen to be 3,883 cal.per gram which is about 100 cal. higher than that calculated from Stohmann and Schmidt's value 4,190 (A 1895 ii 102) for (presumably imperfectly) dehydrated glycogen for this would give for the hydrated form 162/180 x 4,190 = 3,771 cal. The heat of wetting ia found to be about 9 cal. hence the heat of combustion of hydrated glycogen in solution is 3,874 cal. The heat of combustion of lactic acid is according to Meyerhof 3,601 cal. leaving 273 cal. for the conversion of 1 g. of glycogen into 1 g. of lactic acid. The heat of neutralisation of lactic acid by acid salts is according to Meyerhof (Zoc. cit.) 19 cal. per g. which leads to the value 273 + 19 = 292 cal. for the heat liberated during contraction and relaxation of a muscle per g.of lactic acid formed. The total initial heat given by Hartree and Hill in their balance-sheet (as the result of physical measure- ment see preceding abstract) is 296 cal. in close agreement with that calculated above. The salt buffering of the lactic acid is however probably only of a temporary character and is replaced by the more efficient buffering by alkali-protein. Meyerhof has shown that the neutralisation of 1 g . of lactic acid by alkali-protein produces 138 cal. or 119 more than that by salts. The whole of these 119 cal. are however not produced; according to Hattree and Hill the delayed anaerobic heat production is 74 cal. per g. of lactic acid so that only about 60% of the lactic acid formed is neutralised by alkali protein (the rest remains neutralised by salts).The delayed heat production presumably due t o a transfer of alkali from protein to acid phosphate and carbonic acid is the cause of the secondary maximum in the heat production occurring after two and a half minutes (preceding abstract). The velocity of this reaction is independent of the temperature and thus probably depends on a physical process (rate of diffusion of acid phosphate- and carbonic acid-ions) through the muscle-tissue. This theory demands some special distribution of phosphate and carbonate molecules in the muscle substance and may be represented by the following scheme Contraction (a) Glycogen -+lactic acid; (b) lactic acid and con- tra ctile mechanism produce mechanical response. Helacatiotz (a) Lactic acid + K,HPO and KHCO + K lactate + KH,PO and H,CO,.Anaerobic recovery (a) KH,PO and H,CO + K protein + qHPO + KHCO + H protein. Products of the Catalytic Hydrolysis of Horse Hair. V. S. SADIEOV (Biochem. Z. 1923 143 504-511).-Fractionation of the product obtained by the catalytic hydrolysis of horse hair under pressure yielded the following substances (1) A peptide anhydride (m. p. 258") of alanine and leucine to which the formula C2,H,,04N4 or C,,H,,O,N is ascribed. (2) A peptide anhydride of alanine and leucine (m. p. 260") of the formula C,H,,0$T2. (3) A peptide anhydride of leucine and valine (m. p. 246") of the formula G. B.i. 242 SBSTRbCTS OF CHEMICAL PAPERS. C11€&,0$2. (4) Butylalanyl-leucine anhydride Cl,,Hl8O2N2 or its unsaturated derivative Cl,Hl,O,N (m.p. 224"). (5) Dlketo- methylpiperidine C,H 0,N (m. p. 196"). (6) A cyclic nitrogen substance Cl,H&N o! an alkaloidal nature. (7) A substance of the formula C13Hm03N (m. p. 170"). Various other less well characterised derivatives of the nature of diketopiperazines and keto- and diketo-piperidines were obtained. Calcium Fixation by Animal Tissues. IX. E. FREUDEN- BERG and P. GYORGY (Biochem. Z . 1923,142 407-416).-A study has been made of the extent to which calcium phosphate and car- bonate are formed when solutions of casein and egg-albumin are treated with sodium phosphate or carbonate a t different pE and dialysed against calcium chloride. After dialysis the distribution of calcium phosphate and fixed carbon dioxide was determined both externally and internally and the bearing of the results on the process of calcification occurring in animal tissues is discussed.The conclusion is drawn that a reversible fixation of calcium by the tissue proteins takes place. With diminishing metabolic activity the fixed calcium reacts with phosphate and carbonate fo give rise to complex protein-calcium-phosphate-carbonate compounds from which the carbonate is gradually eliminated by the acid nature of the proteins under tissue conditions. The excess of calcium phosphate is then split off leaving the original active group of the protein free to combine with more calcium (cf. also Gyijrgy this vol. i 120). The Basis of the Oxidation Theory of Wieland. 0. WARBURG (Biochem. Z. 1923 142 518-523).-A criticism of Wieland's views on the mechanism of oxidation.In place of the theory that organic oxidations proceed in two phases first hydration followed by abstraction of hydrogen by interaction with molecular oxygen the theory advanced by the author supposes that molecular oxygen is first activated by combination with a catalyst (iron platinum or organic catalyst) and then reacts directly with the oxidisable substance. It is argued that Methylene-blue quinone etc. do not react like molecular oxygen when substituted for the latter (Wieland A. 1913 i 1304) but as activated oxygen i.e. molecular oxygen + catalyst. The action of hydrocyanic acid in inhibiting oxidations is regarded as opposing Wieland's theory whilst it receives a rsady interpretation on the author's views since it combines with the catalyst (e.g.iron) and prevents the activ- ation of the oxygen. F. BATTELLI and L. STERN (Compt. rend. Xoc. Biol. 1923 88 575-577; from Chem. Zentr. 1923 iii 1037).-The constituent of animal tissues to which the taste is due is named by the author carlaisupidin. The amount present in different tissues was determined approximately by dilution of extracts until the taste was just perceptible. The highest content was found in muscles liver spleen and kidneys lower amounts were found in the thymus brain and lungs. J. P. J. P. J. P. Carnisapidine in Animal Tissues. G. W. R.PHYSIOLOGICAL CHEMISTRY. i. 243 Sarcodhromogen in Animal Tissues. F. BATTELLI and L. STERN (Compt. rend. Soc. BioZ. 1923 88 6 7 9 4 8 1 ; from Chem. Zentr. 1923 iii 1037 ; cf.preceding abstract).-Aqueous extracts of animal tissues contain a sarcochromogen which differs from sarcochrome in not being precipitated by metaphosphoric acid. Sarcochromogen is not changed into sarcochrome on evaporation to dryness if the temperature remains below 80". The R61e of Carnisapidin and Sarcochromogen in Animal Tissues. L. STERN and F. BATTELLI (Compt. rend. Xoc. BioZ. 1923,88,681-683 ; from Chern. Zentr. 1923 iii 1037 ; cf. preceding abstracts) .-The occurrence of carnisapidin and sarcochromogen in animal tissues appears to have no direct correlation with metabolic exchange. Certain proteins such as casein give on treatment with pepsin substances analogous to sarcochromogen. Carnisapidin and sarcochromogen are easily dialysable and are not attacked by diges- tive ferments. Carnisapidin administered intravenously is elimin- ated in the urine but when administered orally or subcutaneously it is retained or decomposed probably in the liver.The Extractives of Actiniu equina. D. ACKERMANN I?. HOLTZ and H. REINWEIN (2. BioZ. 1924 80 131-136; cf. A . 1923 i 1155).-The extractives of Actinia equina were subjected to frac- tionation by Kossel and Kutscher's method. The histidine and arginine fractions have not yet been worked up. I n the purine fraction there was obtained adenine .isolated in the form of its picrate. From the lysine fraction there was obtained (1) tetra- methylammonium hydroxide (as picrate); (2) a base identified as pyridylmethylammonium hydroxide (chloroaurate m. p. 252-253" and chbroplatinate) ; (3) a base of unknown constitution actinine C,,H,O,N,; from it there were prepared a chlorouurute m. p.169" a chbroplatinate dark red crystalline nodules decomp. 209" and a hydrochzoride m. p. 207-208'. The Chemical Differential Diagnosis of Transudates and Exudates. K. HIRUMA (Biochem. Z . 1923 142 506-517).- Exudates contain an albumin precipitable by an acetate mixture a t C 0.36 x lo-* by a phosphate mixture at C 0-214 x and by a citrate mixture a t C 1.13 x Inflammatory cerebro- spinal fluids contain the same type of albumin as the exudates formed in pleurisy and in peritonitis. The average sugar content of transudates is 0.115% and of exudates 0.07% the former being in general above the plasma sugar value of the patient and the latter somewhat lower.The residual nitrogen of transudates and the amino-nitrogen and ammonia of both transudates and exudates are approximately the same as those of the blood whilst the residual nitrogen of exudates is higher than that of transudates. In nephritis the residual nitrogen is greater the more marked is the cedematous condition. J. P. The Distribution of Nitrogen in the Urine of Young Dogs and its Dependence on Diet. P. SERIO (Biochem Z . 1923,142 440-453).-The distribution of nitrogen in the urine of young dogs G. W. R. G. W. R. C. R. H.i. 244 ABSTRACTS OF CHEMTOAL PbpERs. kept on a diet rich in fat but poor in or free from nitrogen k~ similar to that of starved animals. The urea is diminished and the ammonia increased due to starvation acidosis. The excretion of amino- acids is not influenced by a low nitrogen diet but it is lowered on a milk diet.The " urease " method for estimating urea gives lower results than the Morner-Sjoquist method a Werence which is more marked the smaller the ratio of urea nitrogen to total nitrogen becomes. During the period of pre-mortal increase in the nitrogen excretion the dserence disappears and it is less apparent on a milk diet. The suggestion is made that the daerence is due to allantoin and creatinine being estimated along with urea in the Morner-Sjoquist method and that the excretion of these urinary constituents varies under the conditions studied ceasing during the pre-mortal period. J. P. Sulphatase. 111. The Enzymatic Fission of the Ethereal Sulphates in the Urine of the Horse Camel and Dog.J. NOWCHI (Biochem. Z . 1923 143 186-189).-By treating horse and camel urine with preparations of sulphatase at 37" hydrolysis of the ethereal sulphates varying from GO to 90% was obtained. Similar extensive hydrolysis of the organic sulphates occurred in the urine of a dog to which phenol had been administered. J. P. Acetonuria and Acidosis. D. ADLERSBERG (Biochem. Z. 1923,143,527-532).-From a study of the effect of orally adminis- tered ammonium dihydrogen phosphate on normal subjecfs kept on a diet free from carbohydrate and on diabetic subjects it is concluded that acidosis diminishes the urinary excretion of ketone substances in these conditions. J. P. K. TAKAO (2. physiol. Chem. 1923 131 304-306).-According to Blum (A. 1892 1116) thymol is not excreted by the dog as thymolglycuronic acid although it is so in man.It is now shown however that a very small fraction of it is so excreted by the dog. By treatment with sodium hypochlorite and fuming hydrochloric acid thymolglycuronic acid in the urine was converted into dichlorothymolglycuronic acid C16H2208C12 colourless needles m. p. 118-119" [c~ftO-66.46~ (c = 309497%) (barium salt [C,,H,,O,Cl&Ba). W. 0. K. K. MORINAKA (Biochem. Z. 1923 142 381-384).-The phosphorus content of the livers of avitaminosed rats to which sodium phosphate lecithin or phosphoprotein has been administered does not differ appreciably from that of the livers of similarly fed normal rats. The conclusion is drawn that in avitaminosis the soft tissues do not lose their power of phosphorus retention.J. P. The Kephalin and Lecithin Content of the Brain in Avitaminosis. H. NAITO (Biochem. Z. 1923 142 385-392).- The percentage content of kephalin and lecithin in the brains of avitaminosed rats and guinea-pigs does not depart appreciably from normal although there is a diminution in the total mass of brain-tissue. J. I?,- Thymqlglycuronic Acid. Phosphorus Metabolism in Avitaminosis.PHPSIOLOGCIOAL CHEMISTRY. i. 245 The Lecithin Content of the Brain and Liver of Normal and Avitaminosed Pigeons after Forced Lecithin Feeding. H. NAITO (Biochem. Z. 1923 142 393-397).-Lecithin feeding slightly increases the amount of the lipoid found in the livers of normal but not in the livers of avitaminosed pigeons. The lecithin of tho brain is not affected in either case.It is concluded that in the avitaminosed liver phosphorus compounds other than lecithin increase a t the expense of the latter (cf. Morinaka above). Fat Metabolism in Avitaminosis. Iv. The Gaseous Meta- bolism of Starved Avitaminosed Rats during Digestion and after Adrenaline Injection. K. ASADA (Biochem. Z. 1923,143 387-398) .-The oxygen consumption of partly and completely avitaminosed rats is sub-normal both during a period of starvation and while a vitamin-free meal is being digested. The extent of the increase observed under the latter conditions compared with the requirements of the starved animals vanes inversely with the severity of the avitaminosis. Adrenaline injections produce a greater increase in oxygen consumption in avitaminosed than in normal rats.A discussion is appended of the earlier results obtained in the study of carbohydrate and fat metabolism in avitaminosis. J. P. Uric Acid and Allantoin Metabolism in Avitaminosis. J. P. A. ADACHI (Biochem. Z. 1923 143 408-422).-In avitaminosis the excretion of allantoin in dogs shows no notable variations whilst that of uric acid is variable and may show transitory increases. In long-continued avitaminosis the uric acid excwtion is markedly increased and may reach values three times those obtained in normal animals. It is concluded that an increased consumption of nucleins occurs in the later stages of avitaminosis. The Pharmacology of the Rare Earths. I. Cerium. S. HARA (Arch. expt. Path. Pharm. 1923 100 217-253).-h general cerium bears a close resemblance to aluminium and to the heavy metals in its pharmacological action. Proteins are precipitated by moderate but not by strong concentrations of its salts.C. R. H. The Pharmacology and Toxicology of Carbon Tetra- chloride. I?. D. LAMSON G. H. GARDNER R. K. GUSTAFSON E. D. WE A. J. MCLEAN and H. S. WELLS ( J . Pham. a p t . Ther. 1923 22 215-288).-A comprehensive study of the path- ological effects following administration to dogs of carefully purified carbon tetrachloride. Oral administration produced only slight and transitory symptoms unless accompanied or preceded by administration of fats or alcohol. When given intravenously it is lethal in doses of 0.154 C.C. per kg. body-weight. The pathology of carbon tetrachloride poisoning consists essen- tially (apart from local irritant effects such as bronchitis following inhalation) in damage done to the liver one of the first indications of which is the appearance of abnormal amounts of bilirubin in the blood.C. R. El. J. P.i. 246 ABSTRACTS OF CHEMIClAL PAPEBS. The RGle of certain Carbohydrates in the Organism. E. 0. FOLKMAR (Bibliotek I&eger 1923 115 120-125; from Chem. Zentr. 1923 iii 1291) .-With continued intravenous injection of pentoses in small amounts at a rate comparable to intestinal absorp- tion approximately one-half is excreted in the urine. Galactose can be assimilated at the daily rate of 1-2 g. per kg. of live weight without glycosuria or appreciable increase in blood-sugar. Maltose is well utilised after becoming hydrolysed by maltase.There is an appreciable retention of sucrose in certain circumstances ; invertase is however not present in blood. Raffinose is completely excreted. The small amounts of ethyl alcohol normally present in blood are attributed to casual decomposition of carbohydrates and not to the action of an alcoholase. G. W. R. The Influence of Oral Administration of Dextrose on the Blood-sugar and on Glycosuria in Healthy Individuals. G. CONSTAM (Biochem. Z. 1923 143 7&104).-The elimina- tion of sugar in the urine is best followed by estimating the amount excreted in unit time. Observations based purely on changes in concentration do not give dependable results. In normal individuals a mixed meal is followed by a dehite glycosuria. Tap water produces a dilution glycosuria which reaches a maximum in about one hour.The assimilation limit for sugar in the same individual is variable and is higher after five hours' than after fourteen hours' abstention from food whilst the assimilation capacity varies in the opposite sense. In the cases quoted the tolerance lay between 100 and 150 g. o%dextrose in 500 C.C. of water after five hours and between 20 and 30 g. in the same volume after fourteen hours' starvation ; less concentrated solutions caused dilution glycosuria which diminished with increasing concentration of dextrose until the assimilation limit was reached when a true glycosuria super- vened. The blood-sugar reaches itsmaximum about half an hour before the urinary sugar. The suggestion is made that the power of the organism to raise the sugar tolerance is a protective mechanism which prevents loss of sugar during periods of diminished rate of assimilation. J. P. A New Method of Preparing s-Diphenylguanidine. Its Pharmacological Effects. 0. RIESSER (2. physiol. Chem. 1923 131 204-213).-Diphenylguanidine is formed together with 8 small amount of triphenylmelamine when dicyanodiarnine is heated with aniline hydrochloride at 190-200". It crystallises from alcohol in white needles m. p. 145-148' and forms a chloroaurate. The pharmacological action of this substance has been investigated. It increases the reflex irritability and also causes central paralysis. In warm-blooded animals it causes death by arrest of the respiration. It produces progressive paralysis when applied to isolated frog's muscle. W. 0. K.

ISSN:0368-1769    

DOI:10.1039/CA9242600236

出版商:RSC    

年代:1924

数据来源: RSC

摘要:

INORGANIC CHEMISTRY. Inorganie Chemistry. ii. 247 Halogen Hydrides. H. BELL.-(See ii 230.) Crystal Structure of Hydrogen Chloride. Determination of the Boiling Point of Bromine. F. SIMON and C. VON SmsoN.--(See ii 2 2 9 . ) A. BOUZAT and G. LELUAN (Compt. rend. 1924 178 635-637).-The b. p. of 10'ii. 245 ABSTRACTS OF CHEMICAL PAPERS. pure bromine is found to be 58.80"/760 mm. (referred to the hydrogen scale). The result is thought to be correct to less than 0.1" (cf. Ramsay and Young T. 1886 49 453). E. H. RIESENFELD and &I. BEJA (2. anorg. Chern. 1923,132,179-200).- The vapour pressures of pure ozone between -169" (4-8 mm. mercury) and -108.5" (€340.8 mm.) have been determined with an accuracy estimated to be within about 1%. From the results the Nernst equation for ozone is found to be logp=-742*4/T+ 1-75 log T-0.01169T+5.4769 T being the absolute temperature and p mm.of mercury. The heat of vaporisation a t absolute zero is calculated as 3500 cal. per mol. and values are given for each 10" rise from 111" to 165" Abs. The chemical constant c calculaked from the relation observed by Nernst with the van der Waal constant a namely c=l*la using the average value of a for ozone at about half an atmosphere pressure is 2.90; employing this and the value 3500 for the heat of vaporisation a t absolute zero the Nernst equation becomes log p=-765*7/T+l-75 log T -0-0126817 +5*783 which also agrees well with the observed vapour pressures. The b. p. calculated from this equation is -110-7" the actual value previously observed by Riesenfeld and Schwab (A 1922 ii 761) being -112.3" ; the true temperature may therefore be assumed on resent knowledge to be -111~5"&0~5".The molecular heat of F. FOERSTER and E. T. MOMNSEN (Ber. 1924 57 [B] 258-263).-The reaction between sulphurous acid and hydrogen sulphide can be represented by the equilibrium H,S+ H2SO3 =+= H2S,0,+H20 (cf. Foerster and Hornig A. 1923 ii 23) ; the Intermediate product is converted by an excess of sulphurous acid into sulphoxylic acid which becomes polymerised to thiosulphuric acid H2S202+H,S03+H,0 + 3H2so and 2H2S02 -+ H2S203+ H,O. The production of thiosulphuric acid may therefore be expected to be the end-point of t'he changes if the reaction can be conducted under conditions such that the hydrogen-ion concentra- tion is insufficient to affect the stability of the thiosulphate anion.These conditions are realised experimentally when aqueous solutions of sodium hydrogen sulphide and sodium hydrogen sulphite in the molecular proportion 1 2 are mixed whereby sodium thiosulphate is obtained readily and in a high degree of purity 2NaHS+ 4NaHS03 + 3Na,S20,+3H20. The preparation of sodium thio- sulphate from sodium sulphide and sulphur dioxide or from sodium sulphite and hydrogen sulphide occurs very smoothly and almost without separation of sulphur when alkali hydroxide is added initially to the solution in such amount as to lead ultimately to the presence of sodium hydrogen sulphide and sodium hydrogen sulphite in the correct proportion. Maxted's Synthesis of Ammonia at High Temperatures.FV. MOLDENHAUER (Chem-Ztg. 1924 48 73-74).-Haber and Nernst have shown that ammonia is an exothermic compound and E. E. T. Vapour-pressure Measurements for Pure Ozone. [quid ozone is calculated to be about 18. s. I. L. Thiosulphates. H. W.mORGANIU CHhMISTRY. ii. 249 that the quantity in equilibrium with its constituent gases at temperatures above 1000" is vanishingly small. Maxted (A. 1918 ii 165) however has synthesised ammonia by injecting nitrogen and hydrogen into an oxy-hydrogen flame and the equilibrium constant of the reaction calculated from his figures is shown to increase with the temperature suggesting that the reaction is endothermic. Haber's results show a maximum heat of formation at 1100" and a zero value above 2000". It is suggested that at the temperature of the flame nitric oxide is formed and on cooling the gases (which contain a large amount of unburnt hydrogen) this is reduced to ammonia.A. R. P. Wuence of Temperature on the Decomposition of Ammonia by Ultra-violet Light. W. KUHN (Compt. rend. 1924 178 708-709; cf. A. 1923 ii 815).-The rate of photo- chemical decomposition of ammonia (20-500") is independent of the pressure between 30 and 300 mm. It rises with increase of temperature but to a less extent than for most chemical changes. The rate falls off as decomposition proceeds is decreased in presence of hydrogen but is unaffected by the presence of nitrogen. This is due to the fact that molecular hydrogen but not molecular nitrogen is capable of reacting with activated ammonia molecules.E. E. T. Catalysts for the Oxidation of Ammonia. 11. Sand Iron Briquette as a Catalyst. 111. Effect of Impurities in the Catalyst. K. INABA (Rikwagaku Kenkyujo I@ 1923,2,426431 431434; cf. ibid. 2 222).-11. Grains of " sand iron briquette " (Yamaguchi J . Chem. SOC. Japan 42 317) were used as a catalyst for the oxidation of ammonia. The briquette when reduced by the cyanide process immersed in nitric acid containing bismuth nitrate and heated at red heat gave a 72.23% yield (961% concentration of ammonia). When thorium nitrate was added to a solution of nitric acid and bismuth nitrate about 72% yields were obtained (10% concentration of ammonia). A gap in the yield-curve was observed at about 7% ammonia which may be attributed to the presence of titanium (5%) in the briquette.111. The addition of 1% of calcium oxide t o the iron-bismuth- thorium catalyst described above gives about 3% increase of the yield. Titanium oxide has practically no efIect but a break in the yield curve was observed at about 7% concentration of ammonia. With 1% of zinc or magnesium oxide decreases of 17-1804 and 743% respectively were observed in the yield of ammonia. K. K. Oxidation of Hydroxylarnine. A. KURTENACKER and R. NEUSSER (2. anorg. Chem. 1923 131 2747).-The oxidation of hydroxylamine by excess of permanganate bromate or ferric salt is greatly influenced by the acidity or alkalinity of the solutions and to a much smaller extent by the concentrations. With perman- ganate in acid or strongly alkaline solutions the products of oxidation are nitrous oxide and nitrio acid or nitrite; in slightly alkaline solutions the oxidation products also contain nitrogen. With 10"-2ii.250 ABSTRACTS OF CHEMICAL PAPERS. bromate the oxidation takes place only in neutral or acid solutions and occurs in two stages (1) a slow oxidat’ion by the bromate itself yielding nitrogen only followed by (2) the liberation of free bromine with rapid oxidation to nitrous oxide and nitric acid the proportion of the latter increasing with the acidity. Reaction (2) is greatly accelerated by acids and in strongly acid solution the sole oxidation product is nitric acid. With ferric salt in acid solution the sole oxidation product is nitrous oxide but the action is not complete unless the ferric salt is in considerable excess.In neutral solution a little nitrogen is produced but in alkaline solution the ferrous hydroxide formed is itself oxidised by the hydroxylamine. These results are discussed and accounted for by the following scheme of reactions (1) NH,*OH+O=NOH+H,O (2) 2NOH= H2N202=N20+H,0 (3) NOH + O,( 0) =HN03(HN02) (4) NOH + NH,*OH=N2+2H,0. Reaction (1) is greatly accelerated by acids and reaction (4) cannot take place in acid solution. Peculiar Catalytic Decomposition of Hydroxylamine. A. KURTENACKER and R. NEUSSER (2. anorg. Chem. 1923,131 310- 320) .-The decomposition of hydroxylamine in presence of a vanadate in acid solution yields nitrogen and nitrous oxide the proportion of the latter first decreasing and then increasing with decreasing acidity. Simultaneous reduction of the vanadate tiakes place.In neutral solution nitrous oxide only is produced but in alkaline solution catalytic decomposition to nitric and nitrous oxides and nitrogen takes place the proportions of the first two decreasing with increasing alkalinity. The oxygen in these gases must be derived from the hydroxylamine and the reduction of the latter to ammonia was proved 100 mols. of hydroxylamine giving 48.8 mols. of ammonia in weakly alkaline solutions and 2-3 mols. in acid solutions. When the vanadate is replaced by ammonium molybdate nitric oxide nitrogen and nitrous oxide are produced the first-named decreasing in amount as the solution passes from acid to alkaline. The gas evolution corresponds with about half the hydroxylamine used ; the molybdate behaves like the vanadate in strongly alkaline solutions.S. K. T. Velocity of Oxidation of Nitric Oxide and its Importance in Nitrogen Fixation. G. W. ToDD.-(See ii 241.) Structure of Phosphorus. Relation between the Glow of Phosphorus and the Form- ation of Ozone. W. E. DOWNEY ( J . Chem. Soc. 1924 125 347-357) .-Pure dry oxygen was passed over purified phosphorus in non-actinic light the oxides of phosphorus was afterwards frozen out and the ozone absorbed in 2% potassium iodide solution the iodine liberated being determined colorimetrically. The amount of ozone formed was proportional to the intensity of the glow although a perfectly fresh phosphorus surface did not show this until i t had become “ acclimatised.” The action of the glow on a stream of oxygen through windows of quartz or fluorite 2 nim.W. H.-R. H. C~LLINS .-(See ii 229.)INORGANIC CHEMISTRY. ii. 251 thick caused the formation of ozone similar stains being produced on potassium iodide-starch paper in 10 minutes with a window of quartz and 1 minute with one of fluorite. The use of phosphorus trioxide in place of phosphorus gave similar results but of a much lower degree of intensity e.g. an hour elapsed before ozone was detected in the second experiment (fluorite window). The glow was shown to be capable of ionising air and hence (Hughes Proc. Camb. Phil. Soc. 1910,15,453) its spectrum extends into the region A 1200-1800 A. which supports the idea that the ozone is formed by the glow of the phosphorus. C. N. HINSHEL- WOOD and B. TOPLEY ( J . Chem.Soc. 1924 125 393406).-The rate of thermal decomposition of pure phosphine in a silica bulb was investigated under varying conditions. The bulb seemed to exert some activating influence which after a time decreased to a constant minimum; porcelain was no less active than silica. The addition of powdered fused silica of an approximately known surface area caused a velocity increase corresponding with the increased surface ; the reaction was thus shown to be heterogeneous (" wall-reaction " as distinct from gas-reaction) up to 1044" Abs. at least in con- tradiction to the results of Trautz and Bhandarkar (A. 1919 ii 277). The decomposition is unimolecular over the pressure range 376-707 mm. The heat of activation determined from the ex- perimental results was approximately constant between 40,000 and 50,000 cals.the average calculated value for a homogeneous (gas-) reaction between 940" and 956" being 86,719 cals. Thus no evidence is afforded that the wall-reaction changes its nature a t higher temperatures. S. K. T Dissociation Constants of Phosphoric Acid. E. B. R. PRIDEAUX and A. T. WARD (J. Chem. SOC. 1924,125,42346).- From the various published dissociation constants of phosphoric acid it is shown that the most probable values of k2 and k3 are about 6 x 10-8 and 1.1 x Preparation of Metallic Arsenides by the Action of Arsenic Hydride on Metallic Salt Solutions. A. BRUKL (2. anorg. Chem. 1923 131 236-246).-Arsenides of the heavy metals are prepared by dropping an aqueous solution of a salt of the appro- priate metal into an atmosphere of arsenic hydride in the complete absence of air the reaction vessel being frequently shaken; preci- pitation by the usual method is accompanied in the cases of copper mercury lead gold and silver by a secondary reaction with the excess metallic ions in the solution M3As+3M'+3H20=6M+ 3H'+&(OH) and in the cases of metals of higher valency by the production of the lowest oxides of these metals.With metals of the iron group alcoholic solutions of the salts were required aqueous solutions giving a product contaminated with arsenic. The heavy metal arsenides are black readily oxidisable sub- stances. On keeping in air they are converted into the metal and arsenious acid whilst in the dry state they ignite spontaneously. S. K. T. Unimolecular Decomposition of Phosphine. respectively.S. K. T.ii. 252 ABSTRACTS OF UHEMICAL PAPERS. At higher temperatures in the absence of air the arsenides of the noble metals lose nearly all their arsenic the other heavy metal arsenides being converted to lower arsenides. Metallic arsenides and phosphides (cf. A. 1022 ii 393) are of similar struoture the former being the more stable. Arsenides of the following metals were prepared copper (CU,As) lead (Pb,As,) mercury (Hg,As) gold (AuAs) silver (Ag,As) bismuth (BiAs) cadmium (Cd3As2) iron (Fe3As2) and nickel (Ni,As,). Cobalt arsenide could not be prepared free from elementary arsenic. Of the above arsenides those of lead and cadmium give arsenic hydride with cold dilute acids and those of copper gold and bismuth are stable towards dilute acids and alkalis.Strong nitric acid decomposes gold arsenide yielding spongy gold whilst hot water and alkalis decompose lead arsenide yielding spoiigy lead. Zinc and manganese arsenides are readily hydrolysed and so could not be prepared by the above method. Formation of Arsenate Jellies. H. B. WEISER and A. P. BLOXSOM ( J . Phy8icaZ Chem. l924,28,2640).-A typical inorganic jelly is made up of a network of minute particles that adsorb liquid strongly; hence jellies may be prepared by coagulating a colloidal solution of a highly hydrated substance in the absence of an electro- lyte that possesses an appreciable solvent action. The effect of salts on the formation of jellies is determined in large measure by the precipitating and stabilising action of the ions.A relatively slow rate of precipitation is essential such as may be realised in the immediate vicinity of the precipitation concentration of an elec- trolyte. A little below this value no precipitation or only a slight precipitation occurs whilst above it coagulation is usually so rapid that a gelatinous precipitate is formed instead of a jelly. The formation of jellies by dialysis of a colloidal solution of a hydrated substance is a special case of coagulation of a colloid by decreasing the adsorption of the stabilising ion below a critical value. Colloidal ferric asmenate and aluminium arsenate have been prepared which owe their stability to adsorbed hydrogen ions. Jellies result by reducing the concentration of the stabilising ions by dialysis.Similar results have been obtained by coagulating the colloids with various electrolytes having strongly adsorbed cations such as citric sulphuric phosphoric and oxalic acids ammonia sodium sulphate and acetate. Since jellies are formed by coagulation of the colloid by acids it is evident that the neutralisation of the adsorbed hydrogen ion and not the hydrogen-ion concentration in the solution is important. Jellies may be formed by mixing quite dilute solutions that react to form a hydrated precipitate provided the precipitation can be delayed until the interacting solutions are thoroughly mixed and then can be made to proceed at a suitable rate. These conditions have been realised with the arsenates of manganese cobalt ferrous iron cadmium and zinc by mixing solutions of their salts of strong acids with potassium dihydrogen arsenate in the cold and keeping them at the ordinary temperature S.K. T. or warming to a suitable temperature J. F s.INORaANIU CHEMISTRY. ii. 253 Colloidal Silica. F. DIBNERT and F. WANDENBULCKE (Compt. rend. 1924 178 564-566) .-The conversion in aqueous solution of colloidal into non-colloidal silica is effected by dissolved salts and is accelerated by rise of temperature. Alkali carbonates are more active catalysts than neutral salts whilst salts with an acidio reaction (e.g. aluminium sulphate) hinder the conversion. The change is so much more rapid in quartz than in platinum vessels that in the former case it proceeds in a short time in the absence of salts under which conditions no change occurs in platinum vessels.The dialysis of a solution obtained by mixing sodium silicate and hydro- chloric acid solutions has been followed by the molybdate method (A. 1923 ii 507). A solution of sodium (or potassium) silicate was kept in a glass vessel for several months. On then submitting it to the action of a current of carbon dioxide it showed an increased silica content (as determined .by the molybdate method) the original silica content being observed however a day later when the carbon dioxide had disappeared. E. E. T. Hydration of Natural and Artificial Glasses. G . SCHOTT and G. LmcK.-(See ii 238.) Soft X-Rays from Carbon. Fusibility of Carbon in the Heat of the Electric Arc. P. I. LUEIRSKY.-(See ii 215.) A. THIEL and F. RITTER (2. anorg. Chem.1923 132 153-158).- The presence of apparently solidified drops of graphite in the crater of the carbon electrode allowed to cool after forming the arc is confirmed by photomicrographs but from a study of the arc in being it is definitely concluded that the spherical formation is not a result of fusion. When no chemical action takes place in the vicinity of the arc a cloud of carbon dust is formed which settles partly on the electrodes especially on the anode forming crystalline dendritic structures which glow brightly and disappear by sublimation. It is intended to obtain cinematographic pictures of the arc in being in s. I. L. A. THIEL and F. RITTER (2. anorg. Chem. 1923,132,125-152).-Two methods which might be expected to indicate temperatures corresponding with definite vapour pressures have been examined.In the f i s t a filament of 0.4 mm. diameter purified to contain only 0*02~0 of ash by a preliminary heating for 2 minutes at 1700-1800" in a high vacuum was heated for short periods to definite temperatures in an atmosphere of pure helium at definite pressures the rate of loss of weight at each temperature and pressure being recorded (cf. Ruff and Bergdahl A. 1919 ii 265). The rate of loss-temperature curve for 2.0 mm. pressure showed no sudden break but a rapid change in direction; that for 0.03 mm. was not very definite although a very sharp break was indicated. The curve for 0.08 mm. was very well defined and showed a sudden break a t 2415" Abs. which is accepted as the temperature at which the vapour pressure order to decide if momentary fusion occurs.Determination of the Vapour Pressure of Carbon.ii. 254 ABSTRACTS OF CHEMICAL PAPERS. of the carbon filament became equal to the external pressure of the argon atmosphere. In the second method a cylindrical length of pure carbon was heated under constant pressures of pure argon by means of increas- ing electric currents; measurements were made at 100 10 and 1.0&0*1 mm. of mercury and curves constructed for each pressure to show the relation of temperatures to electric power. These showed no definite maximum temperatures but that for 1 mm. indicated approach to a maximum at a temperature estimated to be about 2400" Abs. ; experimental difficulties prevented the completion of the curves. Measurements of the absolute temperature of the positive crater of a pure carbon arc under various pressures of air and argon were also made but the former were some 500" lower than those given by Kohn (A.1921 ii 302) ; in argon at a pressure of 1 mm. the observed temperature was 2300" Abs. It is doubtful if the arc system rea.lly corresponds with a vaporisation equilibrium although it is recognised that the results obtained by the other two methods cannot be Graphite as a Metallic Modification of Carbon. E. RYSCHKEWITSCH ((?hem.-Ztg. 1924 48 10l).-The atomic con- figuration of graphite is characterised like that of the metaIs by the presence of free electrons which make it a good conductor of heat and electricity and confer on it its metallic lustre. It is a better conductor of electricity than mercury and a better conductor of heat at high temperatures than many of the pure metals.It resembles metals in not being soluble without change in any solvents except liquid metals and it yields the positive C"" ion in sufficient concentration to be electromotively active. G. FESTER and G. BRUDE (Brennstog-Chem. 1924 5 49-53; cf. this vol. ii 173).- Carbon monoxide is oxidised to carbon dioxide by oxygen even a t the atmospheric temperature in the presence of a palladium catalyst. The latter does not bring about preferential oxidation of carbon monoxide in admixture with hydrogen but given a sufficient partial pressure of carbon monoxide such preferential oxidation occurs with vanadic acid as catalyst. Carbon monoxide is also slowly oxidised a t the atmospheric temperature by an acid solution of vanadium pentoxide (5 g .in 50 C.C. of concentrated sulphuric acid containing 0.1 g. of mercury). Absorption Spectrum of Potassium Vapour at High Temperatures. A. L. NARAYAN and D. GUNNAYYA.-(Seeii 219.) Constitution and Structure of an Atom of Sodium. H. CoLLrm.-(See ii 229.) Determination of Molecular Weights of Sodium Salts by Cryoscopy in Fused Hydrated Sodium Thiosulphate. A. BOUTARIC E. CHAUVENET and ( ~ L E . ) Y. NABOT (Comppt. rend. 1924 178 571-572; cf. A. 1911 ii 1060 etc.).-Sodium thio- regarded as final. s. I. L. A. R. P. [Oxidation of] Carbon Monoxide. W. T. K. €3.INORGANIC CHEMISTRY. ii. 255 suiphate pentahydrate used as a cryoscopic solvent gives normal values for the molecular weights of most sodium salts (chloride bromide iodide carbonate nitrate nitrite phosphate acetate and arsenate).Sodium sulphate (anhydrous) had M=303 a value attributed to polymerisation whilst borax had N=140 (anhydrous salt) no doubt owing to hydrolysis. No experimental details are given. E. E. T. Thermal Analysis of the System Sodium Thiosulphate- Water. M. PICON (Compt. rend. 1924 178 566-568).-By studying the re-heating curves of sodium thiosulphate-water systems obtained by spontaneous crystallisation (in nuclei-free closed vessels) two new hydrates (10 and 12 [ ?]H,O) of sodium thiosulphate have been detected. The decahydrate is isolable at low temperatures the dodecahydrate however undergoing transformation during attempts to isolate it. A third new hydrate is suggested but not proved by the results obtained. The author concludes that the hydrates of sodium thiosulphate do not exist in solution.E. E. T. Hydrates of Sodium Thiosulphate. M. PICON (Compt. rend. 1924 178 700-703 ; cf. previous abstract).-A continuation of previous work the concentration of the salt being now carried to the limit (for 1 atm. pressure) of 23 mols. per 100 mols. of water. The existence of tertiary quaternary and quinternary hydrates (described by Young and Burke) could not be confFrmed above 13 mols. of salt to 100 mols. of water. All hydrates are either primary or secondary. E. E. T. R. AUFSCHIAGER (2. ges. Xchiess- ZG. Xpengsto#w. 1923 18 117-120).-Determinations of the heat of explosion of ammonium nitrate a t constant volume and the estimated temperature of explosion are tabulated.The rate of detonation as measured by Dautriche's method under various conditions of confinement and diameter of cartridge was found to lie between 1270 and 2400 m. per sec. There is no doubt that this compound must be looked upon as a high explosive although of a very insensitive nature. Like other explosives it is more di6cult to detonate in the free than in the confined state but even in the free state it can be repeatedly detonated if sufficient initial impulse is given. H. C. R. Properties of Ammonium Nitrate. VII. The Reciprocd Salt Pair Ammonium Nitrate and Sodium Sulphate. E. P. PERMAN and W. R. HARRISON ( J . Chem. Soc. 1924 125 364- 369; cf. T. 1923 123 2128).-This system exhibits two ternary points ; the main one is a t 118" a t which sodium nitrate sodium sulphate and mixed crystals of ammonium nitrate and ammonium sulphate are in equilibrium. The other is at 117" and consists of sodium nitrate mixed crystals (as above) and ammonium nitrate.The latter is represented by a very small area on the equilibrium diagram due apparently to the conversion of ammonium nitrate into the &form. Explosibility of Ammonium Nitrate. S. K. T. lo**ii. 256 ABSTRMTS OF CHEMICAL PAPERS. Susceptibility of Silver Bromide to Reduction. A. REYCHLER (BUZZ. Soc. chim. Bdg. 1923 32 404407).-Silver bromide can be reduced by alkaline quinol in absence of gelatin not only when exposed to light but also in darkness. The initiation of the action depends on the presence of quinol but is complicated in photographic developers by the fact that the sodium sulphite present also plays a part in the reduction.The author interprets his results as showing that in the presence of sodium carbonate quinol reacts with 2 mols. of silver bromide but on addition of sodium sulphite the latter is activated by the quinol and if present in sufficient quantity the two substances together react with 4 mols. of silver bromide. Any excess of sulphite is inactive as also is sulphite in absence of quinol. Suggested equations representing the oxidation of quinol are given. Solubility of Silver Brom-ate in Solutions of other Salts and the Corresponding Activity Relations. R. H. DALTON R. POMEROY and L. E. WEYMOTJTH ( J . Amer. Chem. SOC. 1924 46 60-64).-Determinations of the solubility of ailver bromate at 25" show that Werent salts of the same valency type (namely potassium nitrate and perchlorate potassium and sodium sulphates magnesium and barium nitrates) have effects on the activity co- efficient product of silver bromate that differ by amounts (3 or 10% at 0-5N) which indicate specific effects of the separate ions that are pronounced enough to merit recognition as important secondary factors.The salts of different valency types have on the average in conformity with the rule of Lewis and Randall nearly equal activating effects when S(cv2) has the same value where c is the molecular concentration and Y the valency of the ions. S. OCHI (J. Chem. Ind. J a p n 1923 26 l-l4).-The curve of the heat of solution of bleaching powder indicated the existence of a monohydrate OCl*CaCl,H,O the heat of solution of which is 7530 cal.whilst that of OC1-CaC1 is 9830 cal. giving the heat of hydration 2300 cal. The vapour pressure of the water in bleaching powder containing 0.82 mol. of water for 1 mol. of OC1-CaC1 (determined by passing pure dry air through it) was found to be 7.99 mm. at 30" 3.93 mm. a t 20" and 1-80 mm. at lo" and the heat of hydration calculated from these data is 2260 cal. (from 10" and 20") or 2000 cal. (from 20" and 30"). Bleaching powder moistened with tetrachloroethane exhibits double refraction which is most distinct when the material contains a quantity of water corresponding with the formula OCl*CaCl,H,O and vanishes when the substance is dried or moistened with water ; some amorphous masses are considered to be unchanged calcium hydroxide.The author concludes that bleaching powder consists essentially of the compound OCl.CaCl,H,O mixed with calcium hydroxide or oxide. High - temperature Researches. XVI. Synthesis and Decomposition of Calcium Carbide. 0. RUFF and E. FOERSTER (2. anorg. Chem. 1923 131 321-347; cf. A. 1922 ii 818).- Attempts were made to prepare pure calcium carbide. A 92.3% H. J. E. J. F. S. Constitution of Bleaching Powder. K. K.INORGANIC CHEMISTaY. ii. 257 product resulted by softening commercial carbide (74.9y0 CaC,) in hydrogen a t 2000-2100" and squeezing out with a carbon rod the process being repeated on the expressed material. Very pure calcium carbide was also obtained by heating at above 1000" powdered carbon or soot (0*2-1*2% ash) with metallic calcium (9705% Ca) in either open or closed crucibles of carbon or iron and in an atmosphere of hydrogen or argon.In this way carbide of 90-93% purity wits prepared. The product obtained in an open crucible is always of a loose nature and very sensitive to moisture but by careful re- melting in argon at 780 mm. it is obtained as a greyish-black crystalline mass in which form it was used in the experiments described below. In a closed iron crucible carbides of 90% purity were obtained (in one case 94%). No evidence was found to support the supposed existence of a carbide CaC. Vaporisation of calcium carbide in nitrogen at 1100" (cf. A. 1919 ii 265) yields calcium cyanamide which vaporises with partial decomposition a t about 1300". In hydrogen traces of acetylene are formed above 2200° and in carbon monoxide calcium oxide and carbon at 1700".From the curves representing these processes and also for the dissociation in argon the dissociation pressure of calcium carbide was found to be 1 atm. at 2500"&50" Le. 1 mm. at 1825"& 50". By comparing the theoretical weight of the residue in the argon experiment with the weight observed it was shown that the vapour is mostly that of calcium 4.5% being the maximum amount of carbide vapour at 1 atm. The " softening temperature " of mixtures of calcium carbide and calcium oxide decreases with increasing calcium oxide content up to about 30% CaO after which it increases again. Extrapolation of the " softening temperature "-composition curve showed that pure calcium carbide melts at about 2300".The equilibrium of the reaction CaO (liquid)+3C CaC (liquid)+CO depends on the partial pressure of the carbon monoxide as well as on the ratio [CaC,]/[CaO]. In an atmosphere of carbon monoxide when the eutectic solidifies this reaction is rapidly and completely reversed. It should be possible on theoretical grounds to prepare pure carbide from calcium oxide and carbon below 2300" for a certain carbon monoxide pressure. This was not realised experimentally. S. K. T. Chemical Effects with Crystals. VI. Etching and Solution Phenomena of Aragonite. F. RINNE and W. KRUGER (Ber. Stich. Ges. Wiss. Math.-Phys. Kl. 1923 74 253-268; from Chem. Zentr. 1924 i 154).-Etching and solution experiments with aragonite using hydrochloria acid of d 1.12 showed that the re- sultant form was rhombic bipyramidal.The Dolomite System. H. L. J. BACKSTROM ( J . Chem. SOC. 1924 125 430434).-Mitchell's figures (T. 1923 123 1055) do not give the true dissociation pressures of calcite magnesite and dolomite the agreement with the Nernst approximation formula in the case of calcite being limited to a very restricted range. G. W. It. 1 o**-2ii. 258 ABSTRACTS OF CHEMICAL PAPERS. Mitchell's conclusion that dolomite dissociates in one step is based on the coiitinuity of the pressure-temperature curve; it is now pointed out that stepwise dissociation only makes itself apparent in the pressure-composition curve and dolomite dissociates in two steps (cf. A. 1913 ii 51 6). Moreover the method used for determin- ing the specific heats of magnesite etc.is liable to a 10% error on the final result. Mitchell's thermochemical equations lead to the con- clusion that calcium carbonate and magnesium oxide can form a compound with a heat of formation from its constituents of about 17 Gals. Objections are also raised to the method used for finding the heat of solution of dolomite. Position of Beryllium and Magnesium in the Periodic System of the Elements. P. PFEIFFER (2. angew. Chern. 1024 37 41) .-The author whilst agreeing with the long-period classi- fication of Panet,h (ibid. 1923 36 407) prefers to place beryllium and magnesium in the zinc-cadmium group. The alkaline-eartlh metals form salt-like hydrides XH and metal-ammonia com- pounds X(NH,) and (in their salts) give a characteristic coloration to the Bunsen flame but none of these properties is shown by beryllium or magnesium.On the other hand the halides and oxalates of the latter like those of zinc and cadmium show a tendency to form double halides and double oxalates which is not possessed by the corresponding alkaline-earth salts. The former also give hexammine salts whereas the halides of the calcium group form octammine salts. The relationship of beryllium and magnesium with zinc and cadmium is particularly clearly seen from their hydrated sulphates some of which are isomorphorus as well as from their ability to form double sulphates which also show iso- morphism ; these characteristics are not shown by calcium strontium and barium. Equilibrium between Metals and Salts in the Molten State.I. The System Lead Cadmium Lead Chloride Cadmium Chloride. R. LORENZ W. FRAENKEL and J. SIZBER- STEIN (2. anorg. Chern. 1923,131,247-265).-Equilibrium accord- ing to the equation Cd+PbCl e CdCl,+Pb is attained in ten minutes when lead cadmium lead chloride and cadmium chloride are fused together a t 600-700". The system consists of three phases and with each salt phase of given composition there corresponds a whole series of metal phases. The composition of the equilibrium mixture was independent of the absolute quan- tities of the phases present and of the direction of the reaction according to the above equation and nearly independent of the temperature over the interval 600-700". The value of [Pb]/[Cd] . [CdCl,] /[PbCl,] a t constant temperature decreased steadily as the cadmium content decreased.Addition of lithium chloride and of mixtures of sodium and potassium chlorides favour the forward reaction in the above equation. Bis- muth has no influence on the equilibrium but antimony favours the reaction from right to left in the equation above. Explanations are attempted on the basis of the formation of compounds in both S. K. T. W. T. K. B.INORGANIC CHEMISTRY. ii. 259 salt and metal phases and on the variation of surface tension at the metal-salt boundary. S. K. T. Fluorescence of Lead and Bismuth Vapours. H. KOPFER- Lead Tetrachloride. E. KRAUSE (Ber. 1924 57 [B] 318).- Lead tetrachloride which has been purified by distillation in a high vacuum gives intensely brownish-red solutions in benzene and its homologues ; the colour is discharged by the addition of carbon tetrachloride chloroform dibromoethane glacial acetic acid or ether without conversion of the lead tetrachloride into the dichloride.Cold dilute solutions may be preserved unchanged in the dark during several hours after which gradual decomposition occurs with separation of lead dichloride. Concentrated solutions decompose after some time with explosive violence without any apparent cause. H. VC'. Colloidal Lead Arsenate. F. J. BRINLEY ( J . Agric. Res. 1923 36 373-374).-A mixture of 311-96 g. of disodium arsenate and 17-35 g . of gelatin is dissolved in a small quantity of hot water and diluted to 10 litres and a solution of 331.4 g. of lead nitrate in 10 litres of water is added slowly with constant stirring avoiding excess of lead. The resulting lead arsenate is colloidal and does not separate for several days. Stronger solutions than 0.1M yield a curdy precipitate and in 0.01M solutions there is a tendency to form fine needle-shaped crystals.J. MEYER and H. WILK (2. anorg. Chem. 1923 132 239-259).-Thallium is readily dissolved in the thallic condition by the action of bromine in hydrobromic acid solution but not by the action of chlorine in hydrochloric acid solution. The solution is freed from halogen and the hydroxide precipitated by means of ammonia; the washed hydroxide is converted into thallic oxide by long boiling under water. For analysis the hydroxide is precipitated by means of ammonia converted into oxide in the same way and weighed; titration with permanganate after reduction also gives very accurate results.The system thallic oxide sulphur trioxide water yields two compounds HT1(S04!,,4H20 and OH*T1S04,2H,0 which are stable only in strongly acid solutions and are readily hydrolysed by water ; the transition at 30" wit'h a concentration of 30% H2S04 requires many days and both compounds adsorb sulphuric acid. The transition temperature rises from 20" a t 42% acid to 90" at about 60% acid. The basic salt exists a t 25" in concentrations from 10 to 40% sulphuric acid and the acid salt in concentrations from 42 to 75%; in acid stronger than this the oxide does not dissolve. No other salts could be observed; the normal hepta- hydrate of Crookes and of Strecker is thought to have been a molecular mixture of these two.The transition temperatures in acid of varying concentrations were determined as a result of eighty-nine analyses of the equilibrium systems. Analogous thallic selenates HT1(Se0,),,2H2O and OH*TISeO,,H,O MANN.-(See ii 220.) A. G. P. Thallic Sulphates and Thallic Selenates.ii. 260 ABSTRACTS OF CHEMICAL PAPERS. have been found to be the odyselenates capableof separate exist- ence. They are likewise very sensitive to water hydrolysing even in moist air; the solubility relations are almost exactly expressed by the diagrams for the sulphates. Thallic hydroxide dissolves in a saturated solution of ammonium sulphate forming & solution from which addition of ammonia and alcohol at 0" precipitates a compound Tl2(SO4),,12NH3,12H20 stable only in an atmosphere of ammonia; the constitutions for this and the analogous selenate compound.A thllo-thallic setenate T1,SeO4,Tl,( SeO,) which is colour - less and a yellow double saZt 5T12Se0,,Tl,(Se04) exactly anal- ogous to the known double sulphates have been obtained and a salt of the composition Tl~TlICl,(SO,) obtained by the reaction 2HTI(S0,),,4H2O + 4HC1= TlLT1T11C12(S04) +C1,+3H2S04+8H20 at 200". The cmpounds TFIITlIBr,( SeO,) and Tln1TPC12(Se04) are also described. Copper-Nickel-Lead and Copper-Iron-Lead Alloys. IX. W. GUERTLER and F. MENZEL (2. anorg. Chem. 1923 132 201- 208).-Whilst copper and nickel mix completely in the liquid state and form mixed crystals in the solid state in all proportions the binary systems copper-lead and nickel-lead show very limited miscibility in the liquid state the limits for non-miscibility being 36-86-5% of lead in the first and 30-84% in the second ; never- theless the ternary system shows complete miscibility in the liquid state in almost all proportions.Mixtures of the German coinage nickel (Ni 25% Cu 75%) with lead in all proportions yielded clean homogeneous alloys of excellent lustre and appearance the micro- scopical examination showing mixed crystals of copper-nickel in a ground mass of almost pure lead. Alloys containing 60% of lead with varying proportions of copper and nickel show complete miscibility in the liquid phase down to very small proportions of either nickel or copper (2.5% Ni 6% Cu); this is attributed to the fact that a small proportion of either copper or nickel is sufficient to make an otherwise immiscible fluid layer of the other with lead completely miscible.The alloys promise to be of considerable value. The immiscibility of the binary systems iron-lead and copper- lead on the other hand persists to a great degree in the ternary system. The cooled mixtures show two or three layers the three consisting essentially of the three metals each showing inclusions of the other two ; where two layers are formed one is iron containing inclusions of copper and lead the other consists of copper crystals in a ground of lead although in the liquid state the copper is com- H. SCHACK (2. anorg. Chem. 1923 132 265-272).-Of the three binary systems which have already been worked out lead and antimony and copper and antimony are miscible in the liquid state but the latter form a compound Cu,Sb which decomposes on cooling into Cu,Sb and Cu,Sb ; the system lead-copper is not completely homogeneous in [n ( ~ 2 ~ ) wH3) 612 ( 0 4 ) ~ and r H,O) 6 ( NH,) 612 (se 0,) are suggested s.I. L. pletely miscible with the iron. s. I. L. Ternary System Copper-Lead-Antimony.MORGANIC CHEMISTRY. ii. 261 the liquid state. The limits for incomplete miscibility in the ternary system are worked out and expressed in the usual triangular diagram; in this region the solidified alloys show two layers the upper consisting of both the copper-antimony compounds the lower of Cu,Sb and lead embedded in a ground mass of the ternary eutectic. The alloys rich in copper have the general structure of the binary copper-antimony alloys with a ground mass of lead; those rich in antimony are too brittle for technical use but those rich in lead appear likely to be valuable industrially the addition of copper tending to improve the technical lead-antimony Action of Ozonised Oxygen on Mercury.V. 0. J. HODGSON ( J . Chem. A'oc. 1924 225 462463).-The prolonged action of ozonised oxygen on dry mercury at the ordinary temperature and in the dark yields mercurous oxide only; in the light photochemical decomposition of this oxide produces a trace of mercuric oxide as well. S. K. T. Arc Spectrum of Scandium. 8. PIRA DE RmIEs.-(See ii 214.) Baskerville and Catlett 's Lanthanates. F. ZAMBONINI and G. CAROBBI (Guzzettu 1924 54 46-52).-The so-called sodium tetralanthanate described by Baskerville and Catlett (A.1904 ii 260) is found to be a mixture composed of lanthanum oxide various indefinite compounds and the double carbonate La,(C0,),,Na,C03,12H20 ; this double carbonate results from the action of water on the product formed by fusing lanthanum oxide with sodium carbonate. The other supposed lanthanates described by BaskerviUe and Catlett (Zoc. cit.) are evidently mixtures similar to the above. Double Carbonates of Sodium and Metals of the Cerium Group. F. ZAMBONINI and G. CAROBBI (Gazxetta 1924 54 53-59).-The double carbonates of sodium and metals of the cerium group prepared under the conditions indicated by Meyer (A. 1904 ii 734) have compositions corresponding with the formula R"',(C0,),,Na,C0,,12H20. The lanthanum cerium praseo- dymium neodymium and samarium compounds are described (cf.Cleve A. 1885 636). Carbonates of the Rare Earths. I. Conditions of Form- ation and Hydrolysis of the Cerite-earth Carbonates. J. PREISS and A. DTJSSIK (2. anorg. Chem. 1923 131 275-286).- Precipitation of 2 yo solutions of the chlorides of cerium lanthanum neodymium and praseodymium with hot normal alkali carbonate furnishes strongly basic carbonates the hydroxyl content increasing in the order La Ce Pr Nd. Excess of alkali raises the hydroxyl content in the case of lanthanum and decreases it in the case of neodymium. When cold solutions are used brightly coloured nearly normal carbonates are precipitated in the amorphous state ; they become crystalline on keeping. A solution of sodium hydrogen carbonate saturated with carbon dioxide precipitates brightly alloys.s. I. L. T. H. P. T. H. P.ii. 262 ABSTRACTS OF CHEMICAL PAPERS. coloured normal carbonates in the amorphous state ; they rapidly become crystalline. The precipitates after drying during eight days in air had the formula E20,,3C0,,8H,O. Drying in a vacuum over sulphuric acid for 50 days reduced this to the dihydrate which lost a further molecule of water when heated a t 100" for 50 hours. It was necessary to heat the monohydrate a t 200" in a stream of carbon dioxide for a week to form the anhydrous carbonate. Hydrolysis of the normal carbonates by heating with excess of water leads to the formation of a basic carbonate probably OHdXCO,. The hydrolysis is complete after about 3 hours (cf.fol- lowing abstract). I n the hydrolysis one-third of the content of carbon dioxide is eliminated and this is regarded as a purely chemical stoicheiometric proof of the t,ervalency of the rare earfhs. In the case of neodymium the temperature coefficient of the velocity of hydrolysis is 2.6 at 70". The strongly basic substances precipitated by hot sodium carbonate solution are mixtures pf the normal carbonate with OH*ECO,. S. K. T. Carbonates of the Rare Earths. 11. Thermal Decom- position of the Cerite-earth Carbonates. J. PREISS and N. RATNER (2. anorg. Chem. 1923,131 287-298).-When the normal carbonates are heated in a slow stream of dry carbon dioxide loss of carbon dioxide occurs first at 300" all the water being expelled below 200". The loss is rapid a t 400" and in the cases of lanthanum praseodymium and neodymium the basic carbonate E203,C02 is formed without any intermediate steps.The products are readily soluble in dilute acids; the lanthanum compound is white and stable between 600" and 900" the praseodymium com- pound is dirty yellow stable between 550" and 815" and the neodymium compound bluish-violet stable between 550" and 800". When the respective temperature limits are exceeded the lanthanum and praseodymium compounds are rapidly converted with no intermediate stages into the oxides La,O and Pr,O,( ?Pr40,) respectively. In the case of neodymium however an inter- mediate compound Nd,O,,$CO is formed which is stable over the range 810-870"; above this range it is completely converted into the oxide.The curves for dissociation in air are of similar form to those in carbon dioxide; the decomposition temperatures are lower however and the temperature-range of stability of the intermediate products is smaller. The curve for cerium carbonate in carbon dioxide exhibits only one step ; the Ce20,,3C0 probably first forms Ce20,,3C0 which loses the greater part of its carbon dioxide at a lower temperature than the compound of CeIn and yields Ce,O4,+CO2 a lemon-yellow compound which a t goo" is converted into CeO,. The compound [Ce,(C0,)3]2[Ce,04,~CO~] was also prepared. These results correspond with the formula (CO,:E*),CO for the normal cerite carbonates. In mixtures the carbonates of these metals lose their individuality as far as thermal dissociation is concerned. Mixtures not containing cerium exhibit mean sharp dissociation temperatures for each step ;INORGANIC CHEMISTRY.ii. 263 mixtures containing cerium however follow the behaviour of the carbonate of this metal. The thermal dissociation curve of the compound OH-NdCO corresponds with that of neodymium carbonate and suggests that the structure of the compound is (OH),:Nd*CO,*Nd:CO (cf. preceding abstract). S. K. T. V. Fuss (2. MetuZZk. 1924 36 24-25).-1n the fifteen possible ternary systems comprising aluminium and two of the elements iron copper nickel zinc magnesium and silicon the aluminium corner of the ternary diagram is intersected by a quasibinary line only in the cases in which the addition of the two elements t o aluminium has a specifically new influence on its properties namely in the systems Al-Mg-Si Al-Mg-Zn and Al-Mg-Cu.Aluminium decom- poses the silicides of iron nickel and copper liberating free silicon and forming intermetallic compounds and addition of any of these elements or of zinc to aluminium results in the formation of these compounds. [Cf. B. 1924 259.1 A. R. P. Range of the Mixed Crystal Phase rich in Aluminium in the Quaternary System Aluminium-MagnesiumSilicon- Zinc. W. SANDER and K. I;. MEISSNER (2. MetaZZE. 1924 16 12-17).-The amounts of zinc magnesium and silicon which aluminium can hold in solid solution increase from S% 2% and 04% respectively a t 20" to 65% 12y0 and 04% respectively a t 450" ; consequently all aluminium alloys containing quantities of these elements between the above limits will deposit them partly on cooling slowly.If the cooling is rapid this action is suppressed but takes place slowly a t the ordinary temperature and is accom- panied by a gradual increase in volume hardness and tensile strength. The equilibrium-diagram of the quaternary system Al-Zn-Mg-Si is developed from a consideration of those of the four ternary systems involved. Hydrates and Hydrogels. 111. Aluminium Hydroxides Poor in Water. R. WILLSTATTER and H. KRAUT (Ber. 1924 57 [B] 58-63 ; cf. A. 1923 ii l67,493).-Specimens of aluminium hydroxide A prepared by the precipitation of aluminium sulphate by concentrated ammonia and prolonged digestion of the precipitate with ammonia exhibit a totally different behaviour when desiccated in a current of air according to whether they are still moist or have been subjected to a rapid preliminary drying in a high vacuum over phosphoric oxide.I n the latter case the weight remains almost constant below 150" thus showing the water to be united chemically to the aluminium oxide. Comparison of the desiccation curves with those of aluminium hydroxide C proves that the dry specimens of composition varying between A120,,2H20 and Al,O,,H,O cannot in the moist condition have the composition Al(OH) but are to be regarded as compounds or mixtures of compounds of the type Al,(OH!,I-xH,O. Desiccation of the precipitate to the ultimate composition AlO(0H) has not yet been Constitution of Ternary Aluminium Alloys. [Cf. B. 1924 259.1 A. R. P.ii. 264 ABSTRAUTS OF OBEMICAL PAPERS. effected by means of ammonia.Comparison of the desimation curves of the dried and moist specimens in a current of air satumted with aqueous vapour at 18" shows that new hydrates poorer in water are produced by an increase in temperature whereas this change is not observed in a dry atmosphere. The desiccation of moist gels is similar in its result to digestion with water or ammonia. The observations explain the failure of attempts to establish the presence of definite hydrates in the hydrogels of various hydroxides. Precipitation of Hydrated Aluminium Oxide by means of Ammonia. G. JANDER and B. WEBER (2. unorg. Chem. 1923 131 266-274) .-Hydrated aluminium oxide is appreciably soluble in slightly ammoniacal solutions of ammonium salts the solubility being less in cold tha'n in hot solutions.The solutions are probably not colloidal since multivalent and univalent anions have the same influence on the solubility organic solvents (e.g. toluene) exert no precipitating effect and the membrane filter (A. 1919 ii 520) allows the solution to pass readily. Equilibrium of Colloidal Aluminium and Lanthanum Hydroxides with Dilute Acids and Bases. H. VON EULER and R. NIISSOX (Ber. 1924 57 [I?] 217-222).-An extension of previous work to aluminium hydroxide which has undergone " ageing " (cf. von Euler and Nilsson this vol. ii 23). Solutions of acetic acid are never completely neutralised by suspensions of aluminium hydroxide A (cf. Willstatter and Kraut A. 1923 ii 493) neutralisation being less complete as the con- centration of the acid in the mixture is increased; an alteration in pH does not occur with an initial acidity of pE 2.6.The addition of small quantities of acid to a suspension of lanthanum hydroxide results in complete neutralisation and the alkalinity of the hydroxide suspension itself ( p H 9.37) is almost attained. The difference between the results with the two hydroxides is not attributed to the differing properties of aluminium and lanthanum but t o the differing age of the specimens. Observations of the behaviour of freshly pre- cipitated aluminium hydroxide towards solutions of hydrochloric acid and sodium hydroxide show the substance to be equally strong as acid or base. Aluminium hydroxide does not appear to be stable in media in which the acidity lies beyond pH 5.5-7.1; the narrower limits depend on the type of the hydroxide. Tervalent Manganese.IV. J. MEYER and W. SCHRAMM (2. anorg. Chem. 1923 132 226-238).-The reaction of sulphur dioxide with manganic hydroxide suspended in water proceeds in two ways the manganic sulphite first formed decomposing according to the equations Mn,(SO,),=.MnSO +lslnS,O and Mn,(SO,),= MnSO +MnSO +SO ; there 1s no reduction to manganous salts nor any decomposition of the manganic salt into manganous salts and the dioxide. No definite results could be obtained connecting the relative extent to which the two reactions occur with any H. W. [Cf. B. 1924 254.1 S. K. T. H. W.INORGANIC CHEMISTRY. ii. 265 external factor. With manganese dioxide two reactions also occur according to the equations 2Mn0 +4SO2=Mn2(SO,) +SO,= MnSO,+SO +MnS20 and MnO +SO,=MnSO ; the latter is favoured and the former repressed by increase of t,emperature.s. I. L. Effect of Silicon on the Equilibrium Diagram of the Iron- Carbon System in the Vicinity of the Eutectic Point. H. A. SCHWARTZ H. R. PAYNE and A. F. GORTON (Metal Ind. 1923 23 226 ; from Chem. Zentr. 1924 i 237).-In the iron-carbon diagram the A eutectic points of the stable and metastable systems are coincident in the absence of silicon. The a- transformation of the stable solid solution of a given composition takes place at a lower temperature than in the metastable system. At a given temperature and with a definite silicon content ferrite is more soluble in " boydenite " than in austenite. Boydenite is the name given to the solid solution in the stable system corresponding with austenite in the metastable system.Chemical Effects with Crystals. VII. Etching and Solution Phenomena of Magnetite. F. RINNE and H. MIELKE (Ber. Sachs. Ges. Wiss. Math.-Phys. Kl. 1923 74 319-329; from Chem. Zentr. 1924 i 154) .-Etching and solution experiments with magnetite using hydrochloric acid (d 1-21) showed that it belongs to the hexakisoctahedral class. The form resulting from the solution of a sphere of the material was a rounded cube. G. W. R. Constitution of the Compounds Formed by the Action of Acetic Acid and its Salts on Ferric Chloride and Ferric Nitrate. R. WEINLAND K. KESSLER and A. BAYERL (2. anorg. Chem. 1923 132 209-225).-By the action of glacial acetic acid on ferric chloride under appropriate conditions a series of cornpunds of the hexa-acetato-triferric cation and the tetrachloro-ferric anion (11) [Fe,( Oh),]( OAc),[FeC1,],CH,*C02H,2H,0 ; has been obtained.By treatment with alcohol the compound (V) [Fe3(OAc),( OH),][FeCl4],5EtOH,H20 is obtained. All these compounds are obtained as well-formed red or reddish-brown crystals and by treatment with sodium chloroplatinate all give the same characteristic compound [Fe,( OH),( OAc),~,[PtCl,],10H20 an orange-red crystalline substance. A further compound (IV) [Fe3(OAc),]C~,,CH,~C~,H,5H2~ is obtained from the mother- liquor from compound (111); by treatment of hydrated ferric chloride with glacial acetic acid the closely related compound (VI) [Fe3(OAc),]C1,,3H,0 is obtained ; both these give the characteristic chloroplatmate.By treatment of one molecular proportion of ferric chloride in solution with two and one proportions of sodium acetate respec- tively the compounds (VII) [Fe,( OH),( OAc),]C1,7H20 and (VIII) [Fe,( OAc) ( OH)2][FeC1,],GH,0 are obtained ; the action of alcohol on the jatter gives the compound (V). G. W. R. (I) [Fe,( OAc),]( OAc),[FeC14],0~5CH,*C0,H,3H20 ; (Ir1) [Fe3(oAc),I(OAc),[FeC14],5H,0,ii. 266 ABSTRACTS OF CHEMICAL PAPERS. By treatment of ferric nitrate with acetic acid the compound (IX) [Pe,(OAc),(OH)]OAcN03,3(4)H,0 is obtained ; sodium acetate gives (X) [Fe3(OAc),( OH),]OAc,4H20 with large excess of acetate (XI) [Fe,( Oh),( OH),]NO,,GH,O with two molecular proportions of acetate and (XII) [Fe3(0Ac),(0H),](N0,),,12H,0 with one molecular proportion of acetate.The latter is interesting as the first salt of a penta-acetato-triferric cation. A series of four compounds obtained by interaction of ferric acetate and ferric nitrate are rega'rded as double compounds of salts of the hexa-acetato- and penta-acetato-triferric base ; they are deep red compounds crystallising unchanged from cold water and the ratios NO OAc for each 9 atoms of ferric iron are respectively 3 17 3.5 17.5 4 17 and 4.5 16.5; a further double compound [Be,( Oh),( OH),]NO [ Fe,( Oh),( OH)]( NO,) 10H,O separates from a solution of one molecular proportion each of ferric hydroxide and nitric acid and three molecular proportions of acetic acid. -411 these nitrate-acetates decompose on exposure to air losing nitric and acetic acids. s.I. L. 0. F. TOWER.-(See ii 237.) Colloidal Nickel Hydroxide. Binary and Ternary Molybdenum Alloys. QREIBHOLZ- (See ii 233.) Soft X-Rays from Tungsten. C. B. BOZZANI and C. T. CHU.- (See ii 215.) Hydrates and Hydrogels. IV. Stannic Acids. R. WILL- ST~TTER H. KRAUT and W. FREMERY (Ber. 1924,57 [B] 63-72).- The stannic acids have been regarded as hydrates with differing water content and more recently as colloidal modifications of hydrated tin dioxide in which the differing properties are due to differing sizes of the particles. It is shown that the latter conception is inadequate to account for the observed facts and that it is neces- sary to assume the existence of more or less definite hydrates. The following stannic acids are described which are most sharply differentiated by determination of the concentration of hydro- chloric acid necessary to bring them into solution as the tin base or to convert them into compounds which are soluble in water Unstable monostannic acids of the composition Sn( OH),? Sn( OH) or SnO( OH) obtained from stannic ammonium chloride ammon- ium chloride and ammonia or from cc-stannate and carbon dioxide in cold solution; they are soluble in hydrochloric acid (lye) nitric acid (1% and more concentrated) and sulphuric acid (1 yo) ; the other varieties of stannic acid are insoluble in the two latter acids.a-Stannic acids oligo- and poly-orthostannic acids are formed by intermolecular dehydration from several molecules of Sn( OH) ; the group includes stannic acids soluble in hydrochloric acid and hence not precipitated by this reagent; it is not possible to decide whether a particular individual is a higher member of this series or a lower member of the series p-stannic acids which are derived from compounds of the cc-group by more or less extensive intra-XNORGANIC CHEMISTRY.ii. 267 molecular dehydration ; the group includes those stannic acids insoluble in hydrochloric acid or transformed thereby into poly- stannyl chlorides soluble in water. (For the preparation and properties of the individual members of each group the original communication must be consulted.) Desiccation of these highly sensitive substances can be accom- plished only by short exposure to the air. Under these conditions the specimens retain the solubilities of the suspensions but dissolve rather more slowly.When dried over phosphoric oxide in a high vacuum or when dried in air and subsequently heat,ed at 150° they lose their original solubility and yield compounds resembling those of the @-group. H. W. Electrolytic Preparation of Tin Sulphides and Thio-salts. G. Tocco and N. JACOB (Gaxxettcc 1924,54,32-38).-Tin sulphides and thio-salts are obtained by the alternating current electrolysis with tin electrodes of sodium thiosulphate solutions. [Cf. B. 1924 255 .] T. H. P. Chemical Nature of Purple of Cassius. A. HnBER.-(See ii 229.) Hafnium Content of some Historical Zirconium Prepara- tions. G. HEVESY (Nature 1924 113 384-385).4pecimens of zirconium compounds prepared by various investigators including those submitted to fractional purification and prepared for atomic weight determinations have been found to contain hafnium in quantities up to 7%.In certain rare minerals containing only small amounts of zirconium and a large number of other elements an even larger amount of hafnium than of zirconium has been detected. The failure to detect hafnium in zirconium by atomic weight methods is ascribed entirely to the untrustworthiness of all the earlier methods employed. A. A. E. Normal Thorium Molybdate Th(MoO,),. F. ZAMBONINI (Gaxxetta 1924 54 39-45) .-Anhydrous normal thorium molybdate prepared by fusing partly dehydrated thorium chloride with excess of anhydrous sodium molybdate forms microscopic bipyramidal crystals belonging to the tetragonal system a c=l 0.73565 d7'5 4.92.This value of a c is approximately double that found for molybdates of the rare earth metals and of the metals of the calcium group. The molecular volume of the molybdate is 112.2 so that the value 56.1 for (O-STh)MoO agrees well with the values for the molybdates of the alkaline-earth metals whereas the value 168.3 for (1.5Th)(MoO4) falls into line with the values for the molybdates of the rare earth metals. Cerous and thorium molybdates exhibit mutual miscibility in the solid state. Confirmation is thus furnished of Urbain and Dauvillier's statements that the rare yttria earths are often found in zirconia that minerals of the rare earths contain elements of every species and that quadrivalent thorium always accompaniesii. 268 ABSTRACTS OF CHEWICAL PAPERS.the tervalent rare earth metals in minerals. Coster and Hevesy's conclusion that the element 72 is chemically homologous with zirconium because the latt,er most probably contains at least 0.01-4-1 yo of this element is regarded as unauthorised. T. H. P. Structure of Vanadium. Electrolytic Preparation of Antimony Sulphides and Thio- salts. G. Tocco (Guxxettct 1924 54 23-32) .-Alternating current electrolysis of sodium thiosulphate solutions with antimony electrodes yields the sulphides and fhio-sa,lts of this metal iiz H. Coums.-(See ii 229.) proporti0ns"varying with che conditions. [Cf. B. 1924 255.1 T. H. P. Hardness of Lead-Bismuth and Cadmium-Bismuth Alloys. c'. DI CAPUA and M. ARNONE (Atti B. Accad. Lincei 1924 [v] 33 i 28-31 ; cf. this vol. ii lll).-As is the case with lead-antimony lead-tin and zinc-cadmium alloys (cf. Kurnakov and Schemt- schuschny A. 1908 ii 932) the hardness curves of lead-bismuth and cadmium-bismuth alloys exhibit maxima corresponding with the compositions of the eutectics. Further as Glasunow and Matweeff found to be the case with zinc-cadmium alloys (Int. 2. MetaZEog. 1914,5,113) these maxima disappear when the alloys are subjected to prolonged re-heating. These results are regarded as confirmatory of the view that the marked hardness of the eutectic alloys is due to the very fine grain and great homogeneity of the alloys and to the internal stresses generated therein by the rapid separation of fine crystals during the solidification. T. H. P. Complex Sulpho-acids of Platinum. L. TSCHUGAEV and s. I(RASSIKOBE' (2. anorg. Chem. 1923,131,299-302).-The action of sulphur dioxide on a hot solution of the chloride of the first Reiset base [Pt,4NH3]Cl gave a precipitate of pale yellow needles [Pt,4NH3][Pt ( 2NH3) (SO,),] ; evaporation of the mother-liquor gave white plates [ Pt (2NH,)Cl (S O,)],[Pt ,4NH3]. Potassium chloro - platinate with the latter gave [Pt(2NH3)C1(S0,)]K together with the Magnus green salt [Pt,4NH3]PtC1,. The formation of the white plates is explained by assuming the 2NH3 is replaced by 2SO this then being replaced by C1 and SO,. Alternatively the following reaction might take place 2[Pt(2NH3)(S0,),][Pt,4NHglf 4HC1=2H,S03+ [Pt ( 2NH3)C1( S03)]2[Pt,4NH3]+ [Pt,4NH3]Cl,. The structures of the compounds were determined by warming with hydrochloric acid and by reaction with potassium chloroplatinate. According to VC7erner both these compounds are of the trans-form S. K. T.

ISSN:0368-1769    

DOI:10.1039/CA9242605247

出版商:RSC    

年代:1924

数据来源: RSC

摘要:

i. 253 Organic Chemistry. The Combustion of Methane to Formaldehyde. H. TROPSCH and 0. ROELEN (Brennstofl-Chem. l924,5,3742).-A gas mixture containing methane and oxygen was passed through a quartz tube (without a catalyst) heated in an oven and the formaldehyde produced absorbed in water and determined titrimetrically. The effect of varying the methane oxygen ratio in the gas the diameter of the tube the reaction temperature and the gas velocity through the tube was studied. The yield of formaldehyde h t increases with increasing gas velocity (the other factors being kept constant) reaches a maximum and then decreases a characteristic curve being obtained by plotting the results graphically. In the experi- ment in which the highest yield of formaldehyde was obtained the gas velocity was as high as 20 m.per sec. The tubes employed had cross-sectional areas of 63.5 13-5 and 1.4 sq. mm. respectively and the highest yields were obtained with the second of the three. If the oxygen content of the gas used is maintained constant (about 16-19y0) and the methane content is varied the highest concentration of formaldehyde in the resulting gas is obtained with a methane oxygen ratio of 1 1 as required by theory. When working a t high temperatures the maximum is only obtained by starting with a high gas velocity and gradually diminishing it much lower results being obtained by working conversely. With increasing reaction temperature the absolute quantity of formaldehyde produced increases but the yield per unit of methane decomposed diminishes. The highest yield of formaldehyde viz.5-96y0 of the methane treated was obtained with a gas containing 1.2% of methane and 19.4% of oxygen whilst a maximum yield of 70% of the methane decomposed was obtained a t a reaction temperature of 700" showing that formaldehyde is formed as the principal product in the action of oxygen on methane. Discrep- ancies between the present results and those of previous workers are discussed. W. T. K. B. The Influence of some Non-inflammable Vapours of Organic Liquids on the Limits of Inflammability of Mixtures of Methane and Air. W. P. JORISSEN and J. V ~ E K (Rec. truv. chim. 1924 43 80-86).-The limits of inflammability of mixtures of methane and air are found to be 5.4% and 14.1y0 of methane. These limits are considerably modified by the pregence of vapours of non-inflammable liquids.In presence of 0.7 to O-SyO of perchloroethylene the limits are 7.35 and 10-15y0; with 0-7 to O-SyO of tetrachloroethane 7-15 and 9*15y0; with 1% of pentachloroethane 5-95 and 10.3y0. In presence of 20% of di- chloroethylene 5% of trichloroethylene or of 12.2 % of carbon tetrachloride explosion is entirely inhibited whilst with 8*5% c;f carbon tetrachloride the limits me 9.0 and 9.9% of methane. VOL. CXXVI. i. k E. H. R.i. 254 ABS!CRACTS OB CHEMICAL PAPERS. The Cracking Process of Burton. H. I. WATERMAK and H. J. W. REUS (Rec. trav. chim. 1924 43 87-92).-1t has been claimed that condensation under pressure in the oil-cracking process leads to the production of distillates free from unsaturated hydro- carbons. The authors are unable to verify this claim.[Cf. B. 1924 207.1 E. H. R. T. HISAIYIURA and Y. IMAOKA (Kwaken HGEoku 1923 No. 7 1-21).-Ethylene is prepared by passing alcohol vapour over " adsol " electrically heated a t 390400" the yield being about 83%. (Adsol is a product used for drying air and is prepared from Japanese acid clay by pressing and burning; cf. Japan. Pat. 42368.) The rate of passing alcohol is 60-80 C.C. (in liquid) per hour the thickness of the layer of the catalyst is 7 cm. and the resulting gas contains 98% of ethylene. After ten hours' passage of the vapour the yield decreases to 71% ; the catalyst may however be re-activated by heating it for about thirty minutes a t 650-700" in a current of air. K. K. The Additive Power of Iodine with regard to Ethylenic Compounds.E. ANDRB (Bull. Xoc. chim. 1923 33 [iv] 1641- 1647).-The author has studied the action of iodine in various solvents on olive sesame poppy and cod-liver oils also on iso- heptene and phenylbutylene. The results obtained are compared with the iodine values furnished by Hanus' method and shorn that the element itself is fairly active. Least activity is shown by the violet solutions in carbon tetrachloride and carbon disulphide ; the brown solution in acetic acid and the violet-red chloroform solution exhibit much greater activity the latter in some cases fixing 97% of the amount used in an iodine value determination. A second series of experiments in which the iodine concentration was in each case N/10 gave similar results from the comparative point of view alt,hough the ratio of iodine fixed to iodine value decreases considerably on dilution.The additive action in respect of ethylenic compounds seems to be capricious m the same reactants under apparently similar conditions yielded discordant results in many cases. H. J. E. Aliphatic Nitro-compounds. XIV. The Preparation of Nitroacetic Esters and the Alkylation of their Silver Salts. W. STEIKEOPF [and in part E. A. HAUGEN A. SCHKADE T. HOPNEB and B. No-] (Annalen 1923 434 21-34).-Ethyl nitroacetate b. p. 97"/13 mm. is prepared by passing dry hydrogen chloride into a solution of the acid or a suspension of its potassium salt in absolute alcohol at 0" or by the gradual addition of concentrated sulphuric acid to a solution of the acid in alcohol at -15".Its ammonium salt Iong needles has m. p. 124" (cf. Bouveault and Wahl A. 1901 i 4). The silver salt if prepared from the pure ammonium salt may be kept for a week before it becomes dis- coloured; it darkens at 40° and decomposes a t 117-119". The following esters are similarly prepared methyl b. p. 106- 110"/31 mm.; propyl b. p. 105"/18 mm.; isopropjyZ b. p. 92.2- The Generation of Ethylene from Alcohol.ORGANIC CHEMISTRY. i. 255 93-2"/12 mm. (ammonium salt m. p. 107-109"); isobutyl b. p. 111~5-112~5"/15 mm. (ammonium salt m. p. 110-112") ; isoamyl b. p. 122-123"/16 mm. (ammonium salt m. p. 112-115'). The silver salt of ethyl nitroacetate (1 mol.) is gradually added to methyl iodide (2 mols.) at or below 0". The product contains ethyl a-nitropropionate (cf.Steinkopf and Supan A 1911 i 4) and ethyl nitrometate acimethyl ether C0,Et-CH:NO,Me a colour- less liquid b. p. 84"/2.5 mm. which gives a red coloration with ferric chloride and is decomposed by means of concentrated hydro- chloric acid at 70" into formaldehyde and isonitrosoacetic acid. Similarly the use of ethyl iodide leads to the formation of ethyl a-nitrobutyrate (cf. Schmidt and Widmann A. 1909 i 453) and ethyZ nitrometate aciethyl ether a colourless liquid b. p. 81"/3 mm. These minitroethers in which in addition to the nitro group there is only one other negative radical are quite stable and do not show any tendency to undergo isomerisation to the C-alkyl deriv- ative. The view of Hantzsch and Voigt (A. 1912 i El) that two such radicals are necessary to stabilise an acinitroet'her is therefore incorrect.It is probable that the compound m. p. 114" obtained by Ratz (A. 1906 i 238) from silver nitroacetamide and ethyl iodide is an minitroether of the formula C&ON*CH:NO,Et. During the action of methyl iodide on silver ethyl nitroacetate (above) the following phenomena are observed. If the salt i~ added a t about -20° it dissolves immediately to give a deep brown solution the colour of which rapidly gives place to a white opales- cence; silver iodide is then precipitated. If the addition is made at -W" there is no reaction; as the mixture is allowed to become warm the solid is gradually dissolved to give a colourless solution. This is stable over quite a wide range of temperature. Suddenly however the solution becomes dark.The brown colour then disappears as before and silver iodide makes its appearance. It is supposed that in the colourless solution there is a molecular '. / IR BgI '... OR (1.1 (11.) (111.) compound (I); the appearance of the brown coloration is due to a kind of double decomposition resulting in the formation of a halo- chromic complex (11) from which the removal of silver iodide gives stereoisomeric acinitroethers O:N-OR RO*N:O' These are then stabilised the former by isomerisation to the C-alkyl derivative and the latter by " conjugation " (111) (Hantzsch and Voigt Zoc. cit.). > 0 is obtained by the addition of concentrated mercurio chloride solution to an aqueous solution containing sodium acetate and the miammonium salt of isopropyl nitroacetate ; the isobutyl and k 2 C02Et*EH and CO,Et*gH NO Hg isopropyl mercuriaci-nitroacetate anhydride C02PrW<i.256 ABSTRACI'S OR' CHEMICAL PAPERS. isoamyl derivatives are prepared in a similar manner. These compounds resemble ethyl nitroacetate acimercurianhydride (cf . Prager A. 1913 i 5 ) ; they are white solids which sublime when heated and are soluble in dilute sodium hydroxide solution potassium cyanide solution or hydrochloric acid. The solution in alkali does not react with potassium iodide. The action of mercuric chloride on potassium or ammonium nitroacetate gives apparently a heterogeneous product which does not sublime on being heated but decomposes violently. Diammonium nitroacetate NH,O*CO*CH:NO,*NH a white solid is obtained from ammonia and nitroacetic acid rn dry ethereal solution ; the disilver salt is also prepared.Condensation could not be effected between ethyl nitroacetate and benzaldehyde piperonal or anisaldehyde using methylamine ethylamine amylamine or piperidine as catalysts. W. S. N. The Formation of Ethyl Chloride from Ethylene and Hydrogen Chloride. E. BERL and J. BITTER (Ber. 1924 57 [B] 95-99).-The union of ethylene with hydrogen chloride in the presence of aluminium chloride is effected most advantageously a t 130-170" but the slowness of the reaction renders prolonged contact with the catalyst necessary for the production of good yields of ethyl chloride; the latter substance is absorbed from the products of the reaction by active charcoal. Ethyl chloride is analysed by passing it through a very narrow quartz tube heated a t 900" absorption of the hydrogen chloride by sodium hydroxide and titration of the chloride by N/lO-silver nitrate.The process is applicable to dichloroethylene but not to chloroform or benzyl chloride. Chlorination of Normal Butyl Alcohol. H. GAULT and R. GUILLEMET (Bull. SOC. chim. 1923 33 [iv] 1792-1801).-The chlorination a t definite tem- peratures in diffused daylight of n-butyl alcohol has been studied using iron (steel wire) as a catalyst and different methods of pro- cedure viz. either bubbling chlorine through the alcohol or (counter-current method) allowing the latter to run in a helical path down a vertical tube up which the chlorine passed. A com- parative chlorination was carried out in each case in absence of the catalyst.The latter does not affect the yield of main product but diminishes the amounts of by-products. These are produced in larger amount in cold than in hot chlorinations owing to the formation of acetals in the latter case and resulting restriction of action. Chlorination at temperatures not higher than lo" by the bubbling method gave in presence of iron a single product (corre- sponding with 18-20% of the alcohol used) shown to be the dibutyl acetctl of a chlorobutaldehyde C,H,Cl( OBu) a colourless liquid b. p. 125"/15 mm. The latter rn presence of hydrogen chloride in a closed tube a t 120-130" was partly converted into an unstable liquid (b. p. 80-90"/15 mm. with loss of hydrogen H. W. Chlorination of Carbon Chains.OItaA.NI(3 CHEMISTRY. i.257 chloride) which reduced Fehling's solution and was converted by alcoholic hydrogen chloride into an mtal CHMeCl*CH,*CH( OEt),(?). Chlorination by the bubbling method (using the catalyst) a t 40-80" gave the dibutyl metal of a dichlmobutuldehyde (yield 50% of alcohol taken; b. p. 140-142'115 mm.). This acetal although unaffected by diluted mineral acids a t the ordinary pressure was partly converted on heating at 125" in presence of hydrogen chloride into a liquid (b. p. 105-110"/15 mm.) which reduced Fehling's solution and lost hydrogen chloride on being distilled. The acetal on being heated with acetic anhydride in a closed tube a t 2W0 was partly converted into n-butyl acetate and a dichlorobutylidene diucetate (b. p. 124-129"/15 mm.) (?) and with chromic acid yielded butyric acid and a butyl dichloro- butyrate (b.p. 210-215'). The counter-current method gives less homogeneous products than are given by the bubbling method. Oleic Alcohol. Y. TOYAMA (Chem. Umschau 1924,31,13-16; cf. A. 1922 i 895).-The acetyl ester of the oleic alcohol obtained from rabukazame oil gave on oxidation with potassium perman- ganate nonylic acid and acetylhydroxynonylic acid. This result coupled with the previously established fact that the same substance gives the acetyl ester of normal octadecyl alcohol on hydrogenation establishes the formula CH,*[CH,].,*CH:CH*[CH,],*CH,*OH for oleic alcohol. It is accordingly identical with the alcohol obtained from ethyl oleate. It has di5 0.8523 d? 0.8491 djo 0.8367; ng 1.4626 ng 1.4607 ng 1-4530 b.p. 333-335'1760 mm. 197'/10 mm. 209'/15 mm. 228"/30 mm. Elaidic alcohol has m. p. 35-0- 35-5" dr 0.8338 ng 1.4522 b. p. about 333"/760 mm. 198"/10 mm. 210"/15 mm. 229-230"/30 mm. These two isomeric alcohols are very stable and show no tendency to change into one another on heating or acetylation. They are sharply differentiated from one another by their melting points. Equilibria between some Glycols and their Acetone Com- pounds. J. BOESEKEN and P. H. HERMANS (Rec. trav. chim. 1923 42 1 104-1 110) .-The authors have prepared the acetone derivatives [isopropylidene ethers] of ethylene glycol propane- a@- diol propane-ay-diol glycerol and monochlorohydrin 1 3-ciscyclo- pentanediol and 1 2-ciscyclohexanediol. The method of preparation used was to heat suitable quantities of the glycol and acetone with sulphuric acid.The following new physical data are recorded iso- propylidene ether of ethyleneglycol b. p. 92.5-92.7" di7 0.9469 n" 1.40024 solubility in water 1 vol. in 3- vols. a t 18"; iso- propylidene ether of propane-ap-diol b. p. 98-99" ng'5 1.40191 cI?,s'5 0.9090 solubility in water 1 vol. in 9-10 vols. ; isopropylidene ether of propane-ay-diol b. p. 123-125" ~2',8'~ 1.4252 di6'5 0.9587 ; isopropylidene ether of a-monochlorohydrin b. p. 116'11 1 mm. d:7'5 1-322 7~5'~ 1.4820. The molecular weight of these compounds has been determined and is found to show no association of the molecules. The equilibrium constants of the reaction acetone+ E. E. T. H. C. R.i. 255 ABSTRACTS OF CHEMICAL PAPERS.diol isopropylidene ether+water have been ascertained in each case and the values compared with the difference in electrical conductivity of a 0.5 mol. solution of the diol containing 0-5 mol. of boric acid and the sum of the electrical conductivitiw of a 0-5 mol. solution of boric acid and a 0.5 mol. solution of the diol. The two sets of figures are parallel which is taken to indicate that the glycols combine with the same facility with acetone as they do with boric acid. A modification of the Le Bel-Henniger fractionating column was used in the purification of the materials the bulbs of the original apparatus being connected with capillary tubes instead of the one opening; this causes the outside lateral tubes continuously to contain liquid. J. F. S.i"t The Condensation of Mannitol with Olive Oil. J. C. IRVINE and H. S. GILCERIST (J. Chem. Soc. 1924,125,lO-15).--" Mannitol fat " (cf. Lapworth and Pearson A. 1919 i 570) prepared by the interaction of mannitol and olive oil in the presence of sodium ethoxide consists essentially of mannitan dioleate which is formed by the dehydration of munnitol dioleute the first product of the reaction. Further dehydration (at 200" under reduced pressure) results in the formation of isomannide dioleate. Treatment of mannitan dioleate with excess of methyl iodide and silver oxide results only in the formation of rnonometh@nunnitan dioleate together with a trace of monmethylisomnnide (% 1.4506) from isomannide originally present. On boiling with 0.5% alcoholic hydrogen chloride monomethylmannitan dioleate gives ethyl oleate and monomethylmannitan a viscous syrup.The latter was methyl- ated to trimethylmannitan a clear mobile syrup b. p. 115-120"/ 0.18 mm. nD 1-4518. Structural considerations based on the reactions described lead to the formula CH,*CHOX*CHOX*CH*CHOH*CH,OH L-O--I where X is the oleyl residue for "mannitol fat." No more than two acyl groups can be introduced into the mannitol molecule. 3'. G. W. The ay-Dichloropropyl Ethers and the Corresponding Acetals. W. DULI~RE (Bull. SOC. chim. 1923,33 [iv] 1647-1654; cf Wohl A. 1898 i 555).-The ethers were prepared by saturating an equimolecular mixture of acraldehyde and aliphatic alcohol with dry hydrogen chloride at -lo" the products being purified by distillation under reduced pressure.They are colourless liquids of characteristic odour decomposing with liberation of hydrogen chloride when heated a t atmospheric pressure. They are not miscible with water but react with it hydrogen chloride being formed. With aliphatic alcohols particularly those containing few carbon atoms they readily produce acetals by reaction with the a-chlorine atom and with magnesium alkyl halides mono- chloroethers are obtained. The following are described metAy2 ay-dichloropropyl ether b. p. 45"/12 mm. 1.187 nz 1.44777 yielding with methyl alcohol the acetal y-chZoro-~~-dimethoxy-ORQAXIC CHEMISTRY. i. 259 proparte b. p. 45'/12 mm. d20 1.059 n? 1.41631 ; ethyl ccy-dichloro- propyl ether has b. p. 56"/12 mm. dzo 1.122 n20 1.44235 and y-chloro-a-diethoxypropane has dae 0.991 (cf.Wohl Zoc. cit.) ; popyl ay-dichlmopropyl ether b. p. 65"/12 mm. d2O 1.100 nz 1.44576 and y-chloro-acc-di~rop~ypropane has b. p. 87"/20 mm. dm 0,972 ng 1.42871; ay-dichloropropyl isobutyl ether b. p. S9"/21 mm. d2" 1.067 9.1; 1.44305 and y-chloro-a-diisobutoxypropane b. p. 105"/12 mm. d'?' 0.936; ay-dipropyl isoumyl ether b. p. 103"/20 mm. d ? O 1.027 turns brown and decomposes on being kept. )I. J. E. The Alkyl Titanates. F. BISCHOFF and H. A D E ~ S ( J . Amerr. Chem. Soc. 1924 46 256-259).-l'etramethyl titunate prepared from methyl alcohol sodium and titanium tetrachloride crystal- lising point 209-210° b. p. 243"(corr.)/52 mm. is rapidly hydro- lysed in air. Tetraethyl titanate prepared similarly from ethyl alcohol had b. p. 205"/156 mm. 145"/8-5 mm.1.107. Tetra- isopropyl titanate obtained mixed with 4% of isopropyl alcohol as a constant-boiling mixture has b. p. 230"/740 mm. Tetra-n-butyl titanate has b. p. 185-187"/16 mm. dz 0.993. On boiling the higher esters with methyl alcohol it quantitative yield of the tetramethyl derivative is obtained. All these esters are colour- less. I?. A. M. H. KAHN (Bul. Soc. Chim. Romdnia 1923 5 70-72) .-Pure octane obtained by the repeated fractionation of benzene from Ploesti and removal of aromatic hydrocarbons by repeated action of a mixture of nitric and sulphuric acids is converted into octyl chloride which when treated with a saturated alcoholic solution of potassium hydrogen sulphide yields octyl mercuptan a colourless liquid of disagreeable odour b.p. 198- 200". When heated it readily decomposes into octyl sulphide and hydrogen sulphide. J. W. B. Valeric Acid. S. KOIZUMI and H. ICHINOSE (Japan. Pat. 4191O).-Valeric acid is prepared from amyl alcohol by electro- lytic oxidation in 5-20% sulphuric acid solution a t the ordinary temperature. Lead or lead peroxide is used as anode and platinum copper iron or nickel as cathode without diaphragm. The addition of 1 g. of oxides of vanadium chromium or manganese to 1 litre of the sulphuric acid promotes the reaction. The electro- lyte is vigorously agitated during the operation the current density being 1-5 amp. per sq. dm. The product separates in a nearly pure state the current yield being above 80%. D. HOLDE J. RIPPER and F. ZADEK (Ber. 1924 57 [B] 103-104).-Since the constants recorded in the literature for these substances are very discordant they have been redetermined with material prepared by Albitzky's method as modified by Holde and Tacke (A.1920 i 811). Palmitic anhydride has m. p. 64' dy 043383 whence cP5 0.8832 and n1O0 1.4679. Stearic anhydride has m. p. 72" dy 0.855 whence d4 0.897 and n'," 14284. Octyl Mercaptan. K. K. Palrnitic and Stearic Anhydrides. H. W.i. 260 A B S ~ O T S OR' CHlClldICAL PAPERS. The Composition and Constitution of Elseostearic Acid. L. MAQUENNE (Bull. Xoc. chim. 1923 33 [iv] 1654-1655).- The author claims priority over Vercruysse (A. 1923 i 533) and refers to his own work (A. 1903 i 62) in support of his view that the extreme readiness with which the acid undergoes oxidation has led to errors with regard to its formula and m.p. which are C18H3002 and 48" respectively. Elaidic Acid and its Anhydride. D. HOLDE and K. RIETZ (Ber. 1924 57 [B] 99-102).-Elaidic acid obtained by the action of nitrous acid on oleic acid crystallises in colourless lustrous leaflets m. p. 44.4" (corr.) mbw 1.4308. Elaidic anhydride pre- pared from the acid and acetic anhydride has m. p. 46.4" (corr.) The electrolytic conductivities of oleic and elaidic acids and their anhydrides in very highly purified acetone have been measured. Little differences are observed between the constants of the acids and their anhydrides but those of elaidic acid and its anhydrides are somewhat higher than those of the oleic corn- pounds. H. W. The Relative Solubilities of the Calcium and Magnesium Salts of the Higher Fatty Acids.W. BILTZ and W. ROHRS (2. angew. Chem. 1923 36 609-611).-A study of the equilibria resulting from the action of solutions of calcium chloride on solid magnesium stearate and oleate. The mixture was heated for one hour cooled and filtered the extent of double decomposition being found from analysis of the filtrate. When equimolecular proportions were used the amounts of calcium stearate and oleate produced were 88 and 81 yo respectively ; these values increased to 98% and 90% on doubling the molecular ratio of calcium chloride. Fractional precipitation of a mixture of equivalent quantities of calcium and magnesium chlorides with potassium stearate and oleate yielded mainly calcium salt (Ca 6G4y0 Mg o.03y0). Experiments on commercial soap are also described. The general conclusion is drawn that calcium stearate is consider- ably and calcium oleate appreciably less soluble than the corre- sponding magnesium salts.The Composition of Whale Oil. C. H. MILLIGAN C. A. KNUTH and A. S. RICHARDSON ( J . Amer. Chem. Xoc. 1924 46 157-166).-Whale oil is shown to contain a complicated mixture of fatty acids having 14 to 22 carbon atoms probably with a small amount of C acids. The highly unsaturated acids are chiefly those containing 20 and 22 carbon atoms. The percentage com- position of the mixture of fatty acids is approximately (Cl& myristic 4.5 ; (CJ palmitic 11.5 palmitoleic 17.0 ; (CJ stearic 2.5 unsaturated (nearly all oleic) 36.5; (CzO) unsaturated 16; (Czz) unsaturated 10 ; (Cz4) unsaturated 1-5 ; unsaponifiable 0.7.F. A. M. The Catalytic Hydrogenation of some Unsaturated Com- pounds especially Oils by Means of Nickel at the Ordinary Temperature. M. TANAKA (Chem.-Ztg. 1924,48,25-26) .-The H. J. E. 043476 $ 0.8396 dioo 0.8338 np 1.4339. H. J. E.ORGANIC CHEMISTRY. i. 261 author has extended the work of Kelber (A. 1916 ii 309 609) by using nickel precipitated on kieselguhr for reduction a t the ordinary temperature of ethyl cinnamate cinnamaldehyde cinn- amyl alcohol azoxybenzene (to azobenzene hydra<zobenzene and aniline) ricinosiearol di-iodide perilla oil hemp-seed oil chrysalis oil and shark oil. Details of the amounts of hydrogen fixed are given and also of similar experiments carried out with a palladium catalyst for comparison.[Cf. B. Mar.] H. J. E. Reaction of Zinc on Mixtures of Ethyl a-Bromoisobutyrate and Aldehydes or Ketones. I. COUCOULESCO (Bul. Xoc. Chim. Rodnia 1923,5 63-70).-The action of zinc on mixtures of ethyl a - bromoisobu t yra te with isovaleraldehyde isopropyl ketone and acetophenone respectively has been studied. With isovaleralde- hyde the reaction takes the normal course and yields ethyZ p-hydroxy- aa6-trimethylhexoate (b. p. 173-175"/140-145 mm.) which is hydrolysed to p-hydroxy-or as-trimethylhexoic acid colourless tetragonal or hexagonal crystals m. p. 81" a=0-00147. The acid is soluble in alcohol and in ether. The barium silver potassium zinc and lead salts are described. The action either of dilute sulphuric acid (1 10) or of hydrogen iodide on this acid yields the y-lactone CHMe2*CH*CH2*CMe2*C0 b.p. 221-222"/742 mm. probably owing to the initial formation of the py-unsaturated acid. Oxid- ation of the acid with potassium permanganate yields only unchanged material carbon dioxide and water. With the two ketones the main products isolated are ethyl isobutyrate unchanged ketones and a small quantity of oil which is probably the expected ester of the hydroxy-acid. This on hydrolysis yielded an acid oil from which a pure substance could not be isolated. -0-J J. W. B. Synthesis of the Higher Monoalkylmalonic Acids. (Mis.) G. M. ROBINSON ( J . Chem. Soc. 1924 125 226-231).-A method for the preparation of higher monoalkylmalonic acids suitable for use with considerable quantities of material in one operation and in which the formation of the corresponding dialkyl acids is minimal is as follows.A mixture of the higher alkyl iodide (1 mol.) ethyl cyanoacetate (2 mols. or more) and potassium carbonate (1.5 mols.) is boiled under a reflux condenser and at a reduced pressure until the reaction is completed. The unchanged cyanoacetate is then recovered by distillation the whole residue hydrolysed with hot 20% sodium hydroxide and the product precipitated with dilute sulphuric acid. Under these conditions the hydrolysis of a-cyanocyclohexylacetic acid yielded only the malonamic acid CH2<g2:g2>CH-CH( CO*NH,):CO,H a result ascribed to steric hindrance. The remaining cyanoacetic acids examined were completely hydrolysed to the corresponding malonic acids. cyclo- Hexylmalonamic acid colourless needles m.p. 184" was unattacked on heating for four hours with alcoholic potassium hydroxide. k*i. 262 ABSTRACTS OF CHEMICAL PAPERS. Heated in a tube it loses carbon dioxide and forms cyclohexylacet- amide . The preparation of n-heptylmalonic acid n-octylmalonic acid rhombic prisms m. p. 115" n-undecyzmdonic acid rectangular needles m. p. 108" and n-hexadecylmalonic acid m. p. 110" is described in detail. n-DecyZdim&yZcra;rbimZ obtained from methyl undecoate and magnesium methyl iodide is a mobile liquid b. p. 145"/10 mm. F. G. W. The Chlomsulphides of Carbon. Decomposition in Presence of Iron. M. DEL~EPINE and J. GIRON (Bull. SOC. chim. 1923 33 [iv] 1785-1792).-The chlorosulphides of carbon when left in contact with metallic iron are converted into sulphur and carbon tetrachloride.The action is catalytic the iron suffering very little alteration. The chlorosulphides examined were thio- carbonyl chloride perchloromethyl mercaptan (also examined in presence of carbon disulphide and/or sulphur chloride) and the chlorosulphides represented by the formuls (1) CCl,-S*CS*Cl (2) CCl,*S,*CCl and (3) C,CI,S (obtained from the residue in the distillation of perchloromethyl mercaptan and having b. p. 150- 154"/26 mm.). E. E. T. Condensation of Citral with Ketones and Synthesis of some New Ionones. H. HIBBERT and L. T. CANNON ( J . Amer. Chem. Xoc. 1924 46 119-129).-Citraly purified by a modification of Tiemann's method (A. 1899 i 622) was condensed with acetone by means of sodium ethoxide giving a 55% yield of pure $-ionone.The best condensing agent for converting $-ionone into a-ionone was found to be 85% phosphoric acid. $-Ethyliouone prepared from citral and methyl propyl ketone is a pale yellow oil b. p. 155-158"/8 mm.; its structure is probably Me,C:CH*CH,*CH,-CMe:CH*CH:CH*CO*Pr. It is converted by means of 85% phosphoric acid into ethylionone a pale yellow oil b. p. 138-140°/8 mm. possessing a pleasant odour of violets. Citral condenses with acetophenone in presence of sodium ethoxide to form phenyl-$-ionone an odourless viscous pale yellow oil b. p. 182-185"/4 mm. Phenylionone prepared from it by the action of 85% phosphoric acid is a pale yellow viscous oil b. p. 172-175"/4.5 mm. with a very faint but rather dis- agreeable odour the violet fragrance being almost entirely suppressed.Attempts to condense citral with diethyl ketone or dipropyl ketone were unsuccessful. Ring-chain Tautomerism. IX. The Mutarotation of the Sugars. J. W. BAKER C. K. INGOLD and J. F. THORPE ( J . Chem. Xoc. 1924 125 268-291).-A proof from a dynamical investig- ation and survey of structural evidence supported by data from two series of crucial experiments that the mutarotation of the mgars is a case of ring-chain tautomerism in the sense F. A. M. r-1 [H lo*( C),&O 0 (C),*C*O[ HIORGANIC CHEMISTRY. i. 263 on lines suggested by Jacobson and Stelzner (Meyer and Jacobson’s “ Organischen Chemie,” 2 Aufl. 1 2 886 910 915 927) and not dependent on the intermediate formation of hydrates as suggested by Lowry (T. 1903 83 1316) and Armstrong (T. 1903 83 1309).It is postulated that the phenomenon is of the same nature in all cases and not of varying kinds as suggested by Irvine and Steele (T. 1915 107 1239-1240). The full mathematical derivation of the form and spacing of the mutarotation-time curves on the mutarotation-time-concentration diagram according to the tauto- meric-hydrogen theory is given. It is deduced that whilst the form of the mutarotation-time curves can be accommodated by the intermediate-hydrate theory the spacing of curyes obtained from measurements of mutarotation in presence of varying initial concentrations of water is zero in direct contradiction of the con- sequences of the hydrate theory and in agreement with the experi- mental results. From structural considerations it is pointed out that whilst d-fructose should not be capable of mutarotation according to Armstrong’s theory its mutarotation has actually been measured in water pyridine and formamide solutions and that d-glucoseoxime d-glucoseanilide d-glucoseimine d-glucose- phenylhydrazone and ethylamino-d-glucose all shorn mutarotation contrary to expectations based on Lowry’s views.Also no sub- stance which according to all three theories should be incapable of mutarotation has been found to exhibit this phenomenon although interconversion of isomeric forms can sometimes be effected by reagents. This is shown to be at variance with Arm- strong’s views and it is suggested that the mechanism of such con- versions is fundamentally different from that of mutarotation.Experimentally the mutarotation of tetra-acetyl-d-glucose in ethyl acetate solution at 44.8” was measured under conditions such that water was rigorously excluded or present in small accurately known concentrations of 0.025 0.05 and 0.1 yo respectively these amounts being small in proportion to that possibly present in the dried materials. The velocity constants were respectively 0.0222 0.0221 0-0209 and 0.0208 i.e. independent of the small initial concentration of water. The mutarotation of a-d-glucose in methyl alcohol was also measured a t the same temperature first with dried materials and then in presence of 0.5 1.0 and 2.0% respectively of added water these concenfrations of water being relatively large in comparison with that which might have been present originally.The initial velocities were respectively 0.165 0.178 0.246 and 0.379. These values are not in proportion to the initial concentration of the water as the hydrate theory demands. Oxidation of Dextrose by Iodine in the Presence of Insulin. G. A. ALLES and H. M. WINEGARDEN ( J . Biol. Chem. 1923 58 225-234).-The rate of oxidation of dextrose by iodine is nof altered by the presence either of insulin alone or of insulin plus liver extract blood-serum or oxalated blood. Since the rates of oxidation of the various sugars differ greatly i t is evident that under the conditions employed insulin does not transform dextrose F. G. W. k* 2i. 261 ABSTRACTS OF CHEMICAL PAPERS. into a more reactive form (cf. Winter and Smith A. 1923 i 513). E. S. A Quantitative Study of the Interaction of Dextrose and Phenylhydrazine. E.KNECHT and F. P. THOMPSON ( J . Chem. Xoc. 1924 125 222-226).-In view of the strong reducing pro- perties of phenylhydrazine and its resistance to reduction by titanous chloride the reaction between phenylhydrazine and dextrose was studied to confirm quantitatively the formation of aniline and ammonia simultaneously with the osazone and to discover the fate of the reactants in view of the low yields of osazone usually obtained. Pure dextrcse (1 mol. 3.6 g.) and phenyl- hydrazine (3 mols. 6.48 g.) were dissolved in a mixture of water (190 c.c.) and glacial acetic acid (10 c.c.) and heated for three hours on the water-bath. A parallel experiment was made using p-nitrophenylhydrazine but at twice the dilution of dextrose and phenylhydrazine.The osazone was filtered off and weighed. The unchanged phenylhydrazine was determined by boiling with Fehling's solution and measuring the nitrogen evolved and the ammonia by a distillation during which tarring was avoided. The amount of ammonia produced in both cases was in agreement with that calculated from the weight of osazone obtained in accord- ance with Fischer's equation but it is inferred that a portion of the dextrose remains in solution as phenylhydrazone. The addition of aniline acetate or ammonium acetate to the reaction mixture increases the yield of osazone. By heating 3.6 g. of dextrose and 6.48 g. of phenylhydrazine with 20 C.C. of glacial acetic acid diluted to 100 C.C. with water for three hours at water-bath temperature a yield of osazone corresponding with 84% of the theoretical may be obtained.F. G. W. A Synthetic Fat containing a Methylglucoside Residue. J. C. IRVINE and H. S. GILCHRIST ( J . Chem. Xoc. 1924 125 1- lo).-When olive oil (1 mol.) is heated with a-methylglucoside (3 mols.) in presence of sodium ethoxide a t 140" methylglucoside mono-oleate is formed and this on further heating a t 220"/12 mm. loses water and forms anhydro-methylglucoside mono-oleate a viscous yellow oil [a]g + 38" in chloroform. Attempts to introduce more than one oleyl residue failed. Treatment of the anhydro-mono- oleate with methyl iodide and silver oxide gave monomethyl- anhydro-methylglucoside mono-oleate a mobile yellow oil ; this on boiling with methyl-alcoholic hydrogen chloride gave methyl oleate and monomethyl-anhydro-methylglucoside a colourless viscous syrup.The latter was further methylated to dimethyl-anhydro- methglqlucoside b. p. 115-120"/0~2 mm. n 1.4419 and hydrolysed with barium hydroxide to monomethyl methylglucoside a colourless viscous syrup. Further hydrolysis with 5% hydrochloric acid a t 100" gave a syrupy monomethylglucose. Consideration of the reactions described leads to the formula CHOMe*CHOH*CHOX*CH*CH*CH where X=CO*[CH,I7*CH:CH*[CH,],*CH3 for the methylglucoside fat. F. G. W. 1 o _ I '0'ORGANIC CHEMISTRY. i. 265 Fluoroacetyl Derivatives of Sugars. 11. Optical Rotation and Atomic Dimension. D. H. BRAUNS ( J . Amer. Chem. Xoc. 1923 45 2381-2390).-In comparing the optical rotations of monohalogen-acetyl derivatives of dextrose cellose xylose and lamdose the differences F-Cl C1-Br Br-I are found to be approximately proportional to the differences in atomic diameter.This simple relation holds for the specific but not for the molecular rotation. Flwwotetra-acetyl-lcewlose is obtained from p-penta-acetyl-levulose. It is readily soluble in chloroform and has m. p. 112" [a] -90.43" in chloroform. It is stable colourless and odourless. Brmtetra-acetyl-lcevulose ws s prepared by the action of a solution of hydrogen bromide in glacial acetic acid on a solution of p-penta-acet,yl-lavulose in the same solvent. It is a very unstable crystalline golid [a]g -189.1" in chloroform m. p. 65". When preserved it is transformed into p-tetra-acetyl- hvulose. F. G. P. Unsaturated Reduction Products of the Sugars and their Transformations.VIII. M. BERGMANN [and in part H. SCHOTTE E. RENNERT S. LUDEWIG and M. KOBEL] (Annalen 1923,434,79-1lO).-Fischer and Curme (A. 1914 i 931) obtained a substance which they described as lactal by the hydrolysis of hexa-acetyl-lactal by means of barium hydroxide. This product had 1 mol. of water of crystallisation [.ID +26.77" to +26-95" m. p. 184-186" or when anhydrous m. p. 165-170". Lactal prepared by using methyl-alcoholic ammonia instead of barium hydroxide is anhydrous and has [a]'; +27.66" to +27.67" m. p. 192" (corr.) decomp. about 212". A sample so prepared is not changed on standing with barium hydroxide solution a t 37" for two or three days. Lactal prepared by means of ammonia is reconverted by the action of acetic anhydride and pyridine into the hexa-acetate m.p. 114" but the product from " baryta- lactal" has m. p. 10&109". It is thought that Fischer and Curme's material was simply impure lactal. When lactal is boiled with water the product has m. p. 198" (corr.) [a] +36.43" but the other properties are in general unchanged. Hexa-acetyl-lactal is converted by boiling with water into $-lactal penta-acetate a heavy snow-white powder (flat tablets small thin prisms or elongated hexagonal crystals) m. p. 123-124" (con.) after previously softening decomp. 190". The structure of this compound is diBerent from that of lactal or its hexa-acetate because on treatment with acetic anhydride and pyridine it gives $-lactaZ hexa-acetate aggregated prisms or needles m.p. 127-128" (con.) [a] +32.24" which is reconverted by boiling with water into the penta-acetate. Both the penta- and the hexa-acetate of $-lactal have a bitter taste ; the hexa-acetate does not give the pine- shaving reaction; and does not react with bromine but it reduces ammoniacal silver nitrate solution and gradually gives a red coloration with magenta-sulphurous acid. In its sensitiveness to even dilute mineral acids it resembles lactal and the deoxy-sugars.i. 266 ABSTRACTS OF OaElldnCAL PAPERS. When +-lactal penta-acetate is treated successively with cold methyl-alcoholic hydrogen chloride sodium hydroxide solution (with cooling) and sodium hydroxide and methyl sulphate a yellow syrup is obtained b. p. 178-180"/0~3-0~4 mm. nE'6 1-4661; this is the penturnethyl ether of a compound C1,H2,0 isomeric with lactal.This series of reactions may involve a change of structure. When +-lactal penta-acetate is treated at 0" with a saturated solution of hydrogen bromide in glacial acetic acid two molecular proportions of hydrogen bromide are taken up giving a dibromide C,,H,,0,,Br2 small prisms or needlm m. p. 124" (corr.) (decomp.) [a]? +69.6". The latter is converted by means of silver carbonate in moist acetone solution into a hydroxybromide C,H,,O,,Br( OH),&H,O snow-white glistening flakes m. p. 87-88' after sintering a t 80" decomp. 110". In dry methyl- alcoholic solution the product is the corresponding methoxybromide m. p. 147-148" (corr.) with subsequent decomposition which is hydrolysed by means of methyl-alcoholic ammonia to the free bromomethoxy-derivative snow-white aggregated needles m.p. 119" (decomp.). q-Lactal penta-acetate which need not be isolated is converted by the action of cold barium hydroxide solution into isolactaZ short slender needles which is very difficult to purify. It has a vigorous reducing action on Fehling's solution and gives a green coloration with a pine shaving. With warm alkali it becomes yellow or brown whilst concentrated mineral acids give a red solution; it is not attacked by perbenzoic acid. isoLactal has a slightly sweet taste resembling that of lactose. The action of acetic anhydride and pyridine on isolactal gives isoZactaZ hexa-acetate rhombic prisms or leaflets m. p. 166-167" after sintering at 163" [ar +55.30" which gives the pine-shaving test and has an insipid taste.By the action of boiling water one acetyl group is eliminated. isoLactal may be acetalised by means of methyl-alcoholic hydrogen chloride or methylated by means of methyl sulphate; it is rapidly oxidised by means of aqueous bromine. The action of perbenzoic acid in warm ethyl acetate solution on an aqueous solution of lactal gives 5-gaZactosidomannose needles m. p. 196-197" (corr.) after slight softening. This compound has initially [a];; +23" (about) rising to 3-30' after eighty minutes. It reduces Fehling's solution immediately and reacts gradually with ammoniacal silver nitrate solution or magenta and sulphurous acid. With phenylhydrazine it gives lactosazone ; it is decom- posed by emulsin. The diacetate obtained by boiling triacetylglucal with water is converted by heating with acetic anhydride and sodium acetate a t 100" into +-gZucaZ triacetate a colourless liquid b.p. 150- 165"/02-0*3 mm. which has a bitter taste and a pungent dour. It is reconverted by the action of boiling water into the diacetate which is described as J/-glucal diacetate. The triacetate is highly sensitive to the action of acids and immediately becomes brownish- black when treated with cold alkali. It gives an intensely greencoloration with a pine shaving and reacts with magenta-sulphurous acid Fehling's solution or ammonictcal silver nitrate. With bromine in chloroform solution at 18" two atoms of halogen are absorbed one of which may then be removed by boiling with silver carbonate and dry methyl alcohol.The action of 1% methylc alcoholic hydrogen chloride on t,b-glucal diacetate gives a paler yellow syrup from which the following fractions are isolated. (1) A colouless mobile liquid b. p. 68--6S0/0*3 mm. %g 1.4763 [a ]$? + 1 -2" apparently C7H1003 and perhaps the methylcycEo- acetal of a hydroxyaldehyde with two bonds and a methylene group adjacent to the acetalised aldehyde group. (2) A viscous slightly yellow oil b. p. 88-9O"~O~l mm ng 1.4984 [a] +1540- -/-16.1" evidently C8H1204. (3) A viscous colourless liquid appar- ently C7H1204 b. p. 120-121"/0~2 mm. n'," 1.4860 [a]'$ +71-7- +72.2". Each of these fractions gives a green pine-shaving reaction and is sensitive to acids; Fehling's solution is not reduced. The action of cold barium hydroxide solution on $-glucal diacetate givea isoglucal a syrup b.p. 120-130"/0~2-0=3 mm. (but sub- limes at about 60" a t this pressure) [a]g +43*15" which is isolated as its phenylbenzylhydrazone microscopic needles m. p. 121-122" [aE -22.38" from which it is regenerated by boiling with aqueous benzaldehyde or concentrated formaldehyde solution the former being preferable. isoGluca1 is readily decomposed by means of acids or of alkali. It gives a very pronounced green coloration with a pine shaving and vigorously reduces Fehling's solution. On keeping the syrupy material in a vacuum crystals are gradually deposited m. p 49-50" [a] +45-6" which in contrast with the original material do not give a coloration with magenta- sulphurous acid. The syrupy isoglucal rapidly reacts with bromine but in chloroform solution the reaction is apparently incomplete.Neutral or alkaline pemanganate solution is immediately de- colorised but perbenzoic acid is not attacked. Acetobromomaltose is reduced by means of zinc dust and 50% acetic acid giving a poor yield (50%) of impure acetomaltal. The latter when boiled with water gives penta-ucetyZmaZM hydrate needles m. p. 173-174" (corr.); when acetylated it gives a hem- acetate m. p. 155-157'. The elimination of the acetyl groups from acetomaltal followed by oxidation using perbenzoic acid gives a disaccharide containing mannose perhaps 6-glucosido- mannose (details will be given later). When a solution of rhamnal in N-sulphuric acid is continuously extracted with ether at 35" a yellow liquid is removed ; the aqueous solution then contains 1-rhammdeose (2-deoxy-l-rhamnose) a colour- less liquid.This substance is very readily decomposed by the action of mineral acids. It gives the pine-shaving reaction and reduces Fehling's solution. It is converted by keeping a t 20" with methyl-alcoholic hydrogen chloride into its methylcyclocccetcal b. p. 120-130"/0~2 mm. which undergoes fission when boiled with aqueous hydrochlorio acid. When rhamnodeose is oxidiaed by means of aqueous bromine solution the product is rhumnodaonici. 268 ABS!CRAC?TS OF CHEMICAL PAPERS. acid which is isolated as its barium salt colourless needles or as the phenylhydrazide m. p. 172-1725” (corn.). Kiliani’s digitoxose is therefore not identical with I-rhamnodeose. It is proposed (cf.Bergmann Schotte and Lechinsky A. 1922 i 227) that in naming a reducing disaccharide the name of the sugar containing the intact aldehydic group should end in “ose,” whilst the second constituent should be named as a glucosidic substituent e.g. 5-galactosidomannose. If both the reducing groups are glucosidically bound as in trehalose a name such as glucosido- glucoside is advocated. Thus sucrose becomes glucosidofructoside or alternatively fructosidoglucoside but raffinose is called galactosido- (glucosidofruc toside). The terms anhydro-sugar anhydrobiose etc. are only employed when the aldehyde group or its FH A half-acetal form is not involved in the anhydrisation ; otherwise one speaks of monose - anhydride disaccharide - anhydride VH*O*C,H,,O etc.The points of attachment of the oxygen CH,OH bridge are indicated by means of bracketed numbers. Thus the substance having the annexed formula is designated 5-galactosidomannose-anhydride The Action of Iodine on Several Carbohydrates. J. VINTILESCU and D. FALTIS (Bul. Xoc. Chim. Romcinia 1923 5 59-63).-When powdered iodine is heated (in a steam bath) in a sealed tube with solutions of sucrose invert-sugar dextrose lzevulose lactose dextrin glycogen or gum arabic the iodine passes slowly into solution during the first period of heating and is completely converted into hydriodic acid the solution becoming colourless. On prolonged heating decomposition of the carbo- hydrate occurs wit,h the formation of formaldehyde formic acid and a brown oxidation product of the sugar.The iodine is trans- formed more easily by sugars containing Itevulose and progressively slowly by dextrose lactose dextrin glycogen and gum arabic ; the period of heating necessary to cause the same amount of h a 1 decomposition follows the same order. The reaction is more rapid the greater the concentration of the sugar solution. The action of hydriodic acid on the carbohydrates is found to be similar but the destruction of reducing sugars formed is slower than is the case when iodine is added directly. [Cf. B. 1024 190.1 J. W. B. E. HEUSER and S. S. AIYAR (2. angew. Chem. 1924 37 27-28).-The hypothesis of the identity of cellulose obtained from different sources is supported by experiments with cotton cellulose and wood cellulose. Both specimens give the same yield of pure cellulose triacetate on acetylation with glacial acetic acid acetic anhydride and sulphuryl chloride (cf.Barnett A. 1921 i 164) and on further treatment with methyl alcohol and hydrogen chloride (cf. Irvine and others T. 1920,117 1489 ; 1922 121,1585) both triacetates give the same yield of the same a-methyl- glucoside. Hydrolysis of both samples of cellulose with 72% OdH1 FHOH I [l 41 [l 21. w. s. N. Wood Cellulose.ORGANIC CH.EM.l.STRY. i. 269 sulphuric acid (cf. Ost and Wilkening A. 1910 i 364) and recovery of the dextrose (cf. Monier-Williams T. 1921 119 803) gives the same yield of crystalline dextrose. The Constitution of Polysaccharides. VII. Esparto Cellulose. J. C. IRVINE and E. L. HIRST (J. Chem. Xoc. 1924 125 15-25).-The amount of furfuraldehyde produced by the action of hydrochloric acid on esparto cellulose corresponds with the presence of 18.5y0 of pentosan in the dry bleached material.This pentosan can be completely extracted with boiling 12% sodium hydroxide and was identified as xylan by hydrolysis in 2% sulphuric acid a t go" to xylose. The total yield of xylose based on optical measurements was 75% and the nett yield of crystal- line material 35-50% m. p. 143-144". The latter yielded on methylation the same trimethyl p-methylxyloside as has previously been obtained from xylose. The formation of furfuraldehyde during the hydrolysis of the xylan could not be prevented. The remaining 81-5y0 of the esparto cellulose after complete extraction of the pentosan with alkali gave on acetylation cellobiose octa- acetate and on methylation a trimetlhyl derivative from which 2 3 6-trimethylglucose was obtained on hydrolysis thereby establishing the identity of the alkali-extracted material with cotton cellulose.The original 4sparto cellulose is more resistant to acetylation than cotton cellulose ; the acetate was prepared by incorporating the material with glacial acetic acid containing a little chlorine adding acetic anhydride and then passing in a little sulphur dioxide and stirring when a clear solution was obtained after twenty-four hours. The yield of acetate corre- sponds with 97.2% of the theoretical assuming that the pentosan forms a diacetate and the remainder a triacetate. The acetate is softer and more sensitive to high temperature than that of cotton cellulose.On treatment with 1-5 yo methyl-alcoholic hydrogen chloride a t 130" for 120 hours the acetate yields a mixture of a- and p-methylglucoside and a- and p-methylxyloside from which pure a-methylglucoside was isolated by crystallisation from alcohol. It is concluded that esparto cellulose is a mixture of 81.5y0 of hexose cellulose and 18.5% of pentosan probably in the form of a solid solution. F. G. W. Starch Iodide. A. LOTTERMOSER (Oesterr. Chem. Ztg. 1924 27 13; cf. A. 1922 i lo).-Experiments are mentioned on the adsorption of tri-iodion iodion and iodine by starch. The curves obtained are typical adsorption isotherms. In the reaction between starch and iodine in the presence of potassium iodide there is probably an initial adsorption followed by a further slow corn- bination. G.W. R. Lignosulphonic Acid and the Lactone of Waste Sulphite Liquors. S. V. HINTIKKA (Cellubsechemie 1923 4 93-94).- Salts of a-lignosulphonic acid with primary amines were prepared by separating the sodium lignosulphonate by salting out from waste sulphite liquors and precipitating the aqueous solution of W. T. K. B.i. 270 ABSTRAmS OB UHEMICAL PAPERS. the product with or-na hthylamine sulphate p-naphthylamine has considered the a-naphthylamine compound to be a normal ammonium salt of the sulphonic acid and the @-naphthylamine compound a cyclic internal salt coupling the sulphonic acid and the aldehyde groups. The behaviour of these compounds towards 5% sodium hydroxide at 50" has been studied the liberated amine bemg removed by extraction with ether and the lignosulphonic acid reprecipitated by acidification.No difference in the behaviour of these amine salts could be detected; the regenerated ligno- sulphonic acids were similar in appearance and contained only small residues of nitrogen. The crystalline substance isolated by Holmberg (A. 1921 i 849) from waste sulphite liquors and de- scribed as a lactone has been further investigated. It occurs in the liquors from the digestion of spruce wood but could not be detected in those from pine wood or from spruce bark. In the digestion of spruce wood this lactone can be isolated in substantial quantity from the liquor in the early stages of digestion but the quantity decreases towards the end of the process. M. SOMMELET (Cmpt.rend. 1924 178 217-219).-Methylamine hydrochloride prepared by the methods of Brochet and Cambier or of Werner (cf. T. 1917 111 844) is found to contain ammonium chloride and trimethyl- trimethylenetriamine hydrochloride. It may be obtained pure by rendering alkaline extracting with benzaldehyde distilling the resulting benzylidenemethylamine (b. p. 180°) hydrolysing the latter with hydrochloric acid washing the methylamine hydrochloride so obtained with alcohol to remove traces of the triamine salt and finally cryst'allising twice from 88% alcohol. Pure methylamine hydrochloride has m. p. (in closed tube) 23%- 233.5" (corr.). E. E. T. F. D. CHATTAWAY and F. L. GARTON ( J . Chem. Xoc. 1924 125 183-188).-Stable tetrachloro-iodides of most organic bases can be prepared by dis- solving the base in strong hydrochloric acid adding an equivalent quantity of iodine and passing excess of chlorine into the solution.The iodine gradually dissolves with evolution of heat and the tetrachloro-iodide separates either immediately or on allowing the solution to cool in a current of chlorine. The yields are generally quantitative. The tetrachloro-iodides are all golden-yellow crystalline compounds soluble in water with slight decomposition. They may be recrystallised from suitable solvents to which the addition of a little iodine trichloride is sometimes necessary to prevent dissociation. They have sharp melting points at which decomposition generally takes place but appear to soften about 20" below the melting point. Heated at higher temperatures they evolve chlorine iodine monochloride and iodine and leave a residue of the chloride of the base.They can be stored unchanged for at least a month in dry air or in a vacuum over lime. The iodine atom appears to act as the central atom of the negative ion hydrochloride and o-touidine P salt Klason (A. 1922 i 324) J. F. B. Preparation of Methylamine. Tetrachloro-iodides of Organic Bases.ORaANIa CEEMISTRY. i. 271 the four chlorine atoms being arranged tetrahedrally around it the general formula thus becoming similar to that usually assigned to the chloroaurates. Cruanidim tetrmhloro-iodide M€:C( NH&NH,IC14 golden-yellow flattened prisms m. p. 163O. Carbamide tetrachloro- iodide NH2=CO*NH3*IC1,,H2O long pale yellow needles m. p. 73". The tetrachloro-iodides of pyridine quinoline etc.are so sparingly soluble that for their preparation it is necessary to dissolve the iodine in hydrochloric acid by means of a current of chlorine before adding the base. Pyridinium tetrachloro-iodide pale yellow flattened prisms m. p. 205" (decomp.). a-Picolinium tetrachloro- iodide bright yellow flattened prisms showing multiple twinning m. p. 166" (decomp.). CoUidinium tetrachZoro-iodide C5H2Meg,HIC14 orange-yellow long flat rhombic prisms m. p. 94". Quinolinium tetrachbro-iodide C9H,N,HIC14 pale yellow prisms m. p. 195" (decomp.) . 8-Methylquinolinium tetrachloro-iodide C,H6MeN,H1C1 long pale yellow hair-like crystals m. p. 152" (decomp.). 6-Nitro- quinolinium tetrachloro-iodide N02*CgH6N,HICI dull yellow elongated plates m.p. 131" (decomp.). 2-8hloroquinolinium t et rachloro - iodide C,H @" HICl pale yellow flattened prisms m. p. 143" (decomp.). 6-Chloroquinoliniurn tetrachloro-iodide dull yellow crystalline powder m. p. 131" (decomp.). Quinaldinium tetrachloro-iodide CgH,MeN,HIClp dull yellow dendriform crystals m. p. 149" (decomp.). Caffeine tetrachloro-iodide C,H1,0~4,HIC1 pale yellow crystalline powder m. p. 134" (decomp.). Piperidinium tetrachloro-iodide C5H&H,HIC14 exists in two polymorphic forms ; it crystallises from acetic acid in canary-yellow fern-like crystals which change on standing in contact with the mother-liquor into compact six-sided orange-yellow prisms. Both forms melt a t 102" (decomp.). Piperazine bis-tetrachloro-iodide bright yellow compact prisms tmth domed ends darkens on heating above loo" and loses iodine trichloride at 160"; iodine is liberated at 195" a black charred residue remaining.Pyridine-betaine tetrachloro-iodide C5H,N(IC14)*CH2CO&€ bright yellow compact six-sided prisms m. p. 133" (decomp.). Semicarbazide does not form a tetrachloro-iodide but is decomposed by a solution of HICl into nitrogen carbon dioxide and ammonium tetrachloro- iodide. Methylammonium tetrdloro-iodide CH,*NR,,HICl long golden-yellow hexagonal prisms with domed ends m. p. 96" (decomp.). Dimethylammonium tetrachloro-iodide (CH,),NB[,HICI compact orange-yellow prisms m. p. 82" (decomp.). Trim&&- ammonium tetrachloro-iodide canary-yellow plates m. p. 1 8 2 O (decomp.). Ethylammonium tetrachloro-iodide C&&*m2y=(&y hygroscopic orange-yellow compact rhombs m.p. 45". Diethyl- ammonium tetrachlmo-iodide thin four-sided pale yellow plates m. p. 79" (decomp.). Ethylenediammonium bis-tetrachloro-iodide 7QNH2,H1C14 CH,em2,HIC1,} 2H20 golden-yellow long hexagonal prisms H1CI&C4H,6N,,HIC1&i. 272 ABSTRACTS OF CHEMICAL PAPERS. m. p. 105" (decomp.). Benxylumrnonium tetrachloro-iodide C6-H5*~2*hTH,,HIC1 flattened yellow prisms m. p. 97" (decomp.). Dzbenzyluminonium tetrachloro-iodide (C,H,*CH,)&E&HICl,,. h e yellow needles m. p. 165". Tribenxylammonium tetrachloro-zodide (C6H5*~,)~,HIC1 yellow rhombs m. p. 100" (decomp.). F. G. W. The Constitution of Galegine. G. BARCER and F. D. WHITE (Biochem. J . 1923 17 827-835) .-The unsaturated amine C,H,,N first obtained by Tanret by hydrolysis of galegine from the seeds of Gulega oficimlis (A.1914 i 721) gives Hofmann's carbylamine reaction forms a toluenesulphonamide soluble in sodium hydroxide and decolorises acid potassium per- manganate. It has di 0.779 and forms a chloroplutimte m. p. 194-197" a chloroaurate m. p. $lo and a picrate m. p. 138.5- 139.5". It is identified as an aminoamylene. The urea obtained by Tanret from galegine arises from a guanidine residue. Galegine sulphate gives Weyl's reaction and the diacetyl reaction and in the presence of a palladium catalyst takes up 1 mol. of hydrogen to give dihydrogalegine sulphute (c6H15N3)2,H2s04 colourless prisms m. p. 270". The nitrate long needles m. p. 75-76" and the picrate long narrow plates m. p.172" are also described. On distilling the sulphate with quicklime isoamylamine was obtained. Dihydrogalegine (isoamylguanidine) synthesised from cyanamide and isoamylamine gave the same salts as the product obtained by reduction of galegine from natural sources. The oxidation of galegine sulphate with barium permanganate yielded acetone and glycocyamine. On being boiled with dilute sulphuric acid galegine takes up 1 mol. of water to form hydroxydihydrogalegine sulphate (C6H,,0N3),,H,S0 m. p. 205-206" which yields on hydrolysis hydroxyisoamylamine. Hydroxydihydrogulegine picrate rhomb- shaped crystals has m. p. 153-154". On the basis of these results the formula Me,C:CH*CH,*NH*C( :NH)*NH or CH,:CMe*CH,*CH,*NH*C( :NH)*NH is ascribed to galegine. The cyclic structures suggested by Tanret (h.cit.) are therefore both incorrect. The Absorption Spectra of some Amino-acids. The Possible Ring Structure of Cystine. F. W. WARD (Biochem. J. 1923 17 898-902) .-Phenylalanine tryptophane and tyrosine show marked absorption bands whereas alanine histidine glutamic acid and cystine give general absorption. Cystine is the only amino-acid having any marked absorption in the region of the solar ultra-violet and it shows absorption of greater intensity than t'hat of the other aliphatic amino-acids examined this being as strong as that of phenylalanine. A possible cyclic structure for cystine is suggested. J. P. The Reaction between Nitriles and Organo-magnesium Compounds. Ethyl Cyanoacetate. R. BRECIIPOT (Bull. Soc. chim. Belg. 1923 32 386-397).-The yield of ketone obtained from the reaction product of nitriles and the Grignard reagent is practically nil in the case of acetonitrile and phenylaceto- J.P.ORGANIC (3HEJ6ISTRY. i. 273 nitrile. Bruylants (A. 1923 i 310) postulates the exist- ence of an equilibrium between the tautomeric forms of a nitrile :CH*CN :C:C:NH the normal form yielding a ketone by the Blaise reaction the pseudo-acid form on the other hand yielding only condensation products. The formation of the condensation prodncts is the result of the intermediate production of the deriv- ative :C:C:NMgBr. With ethyl cyanoacetate and ethyl magnesium bromide a violent reaction takes place; a certain amount of the ketone is obtained but the main portion of the reaction mixture consists of tarry products whilst considerable quantities of gas (ethylene) are evolved.Using four molecular proportions of the Grignard reagent to one of ester the amount of ethylene obtained shows that the ester contains two labile hydrogen atoms which are almost completely eliminated by the Grignard reagent. If the operation is carried out by adding the Grignard reagent to the nitrile much ethylene is liberated until slightly more than one gram-molecular proportion has been added when the reaction stops abruptly owing to the formation of CN-CH(MgBr)*CO,Et. Addition of water to this compound regenerates the nitrile. The ketimine of ethyl propionylacetate CH,-CH,*C( :NH)*CH,*CO,Et b. p. 100-103"/12 mm. is a colourless liquid of musty odour dp 1.0169 n 1.4938. The Relation between the Structure of Organic Halides and the Speed of their Reaction with Inorganic Iodides.I. The Problem of Alternating Polarity in Chain Compounds. J. B. CONANT and W. R. KERNER (J. Amer. Chem. Soc. 1924,46 232- 252).-A method has been developed for comparing the reactivities of the halogen atoms in organic compounds consisting in measuring the rate of reaction between the organic halide and potassium iodide in acetone solution. The reactivity of the chlorine atom in three series of compounds of the type A(CH,),Cl has been measured and compared with the reactivity of the chlorine atom in n-butyl chloride. The halogen atom is more reactive in the compounds A*CH,Cl than in AC1 or A*CH,*CH,Cl in the three series studied in which A is either benzoyl phenyl or carbethoxy.The influence of the group on the chlorine atom in the compounds of the type A*CH,Cl is roughly proportional to the effect of the same group on the activity of the hydrogen atom in the compounds of the type A*CH,. In the benzoyl series the compound A*CH2*CHzCI is eighty times more reactive than a simple normal alkyl chloride; in the phenyl and carbethoxy series the activity of the corre- sponding compound is of the same order as that of the alkyl chlorides. The compounds A*CH2*CH,*CH,C1 in the benzoyl and phenyl series are more reactive than the compounds A*CH,*CH,Cl this increase being large in the benzoyl series; in the carbethoxy series no such increase is apparent. The chlorine atom in the next higher homologue in the phenyl and carbethoxy series is only as reactive as that in butyl chloride; no higher homologues were available in the benzoyl series.Phenylamyl phenylhexyl and phenylheptyl chlorides do not differ significantly in their reactivity from one another or from 72-butyl or n-amyl chlorides. F. G . P.i. 274 ABSTRACTS OF CHEMICAL PAPERS. The preparation and properties of the following compounds are described P-chloropropiophenone from p-chloropropionyl chloride and benzene in presence of aluminium chloride m. p. 49-50" (previous investigators have given it as 57-58'). y-Chlorobutyro- phenone m. p. 19-20". p-Phenylethyl chloride from the corre- sponding alcohol and fuming hydrochloric acid a t 140" for four to five hours b. p. 68-5-69'/4 mm. PhenylpropyZ chloride from phenylpropyl alcohol and concentrated hydrochloric acid a t loo" b.p. 85-87'/9 mm. 6-PhenylbzctyE chloride b. p. 100-101"/6 mm. r-Phenylamyl chloride b. p. 111-112'/6 mm. [-Phenylhexyl chloride b. p. 115-116"/4 mm. Phenylheptyl chloride was obtained only about S8y0 pure from the crude alcohol and hydrochloric acid b. p. 120-140'/3 mm. Absorption of Ultra-violet Light by Organic Compounds. 11. L. M~RCHLEWSKI and A. MOROZ (Bull. SOC. china. 1934 [iv] 35 3740).-The results of studying the ultra-violet absorp- tion spectra of alcoholic solutions of nitrobenzene azoxybenzene azobenzene hydrazobenzene and aniline are given in tables and curves. The extinction coefficients are calculated (cf. this vol. A. ii 7). E. E. T. The Constitution of Dicyclopentadiene. H. STAUDIKGER and A. RHEINEB (Helv.Chim. Ada 1924 7 23-31).-Commercial dicyclopentadiene can be separated by fractional distillation at a low pressure into two isomerides ; the higher-boiling fraction the p-form has m. p. 19-5' whilst the a-form the known dicyclo- pentadiene has m. p. 32.5". These two forms probably corre- spond with the formulze F. A. M. Each of these isomerides should exist in two-stereoisomeric forms. The two isomerides which were isolated cannot be stereoisomerides since by partial reduction they both give the same dihydrodicyclo- pentadiene white crystals m. p. 57". This compound can be dist>illed a t atmospheric pressure (b. p. 183-5-184.5") and is there- fore more stable than dicyclopentadiene which under these con- ditions is completely depolymerised. Dihydrodicyclopent'adiene is decomposed into cyclopentadiene and cyclopentene when its vapour is passed over a red-hot platinum spiral.Further reduction of dihydrodicycbpentadene gives the known tetrahydrodicyclo- pentadiene which is still more stable and is with difiiculty decom- posed to cyclopentene. The disruption of the cyclobutane ring in dicyclopentadiene is attributed to the influence of the double bond in the p-position to the carbon linking. A similar cause is respon- sible for the ready decomposition of limonene caoutchouc and hexaphenylethane. Tricyclopentadiene is obtained when dicyclo- pentadiene is heated in a sealed tube a t 170" for some hours. It forms colourless crystals m. p. 60" b. p. 90-92"/0.06 mm. It is more stable to heat than dicyclopentadiene or tetracyclopentadiene.The latter is obtained by heating dicyclopentadiene for it longerORGAN10 CHEMISTRY. i. 275 time a t 180,". It has m. p. 188-190" and when heated at 180- 200" decomposes smoothly into cyclopentadiene. When reduced with hydrogen and platinum black it gives tetrahydrotetracyclo- pentadiene m. p. 200-202" which can be distilled unchanged. Still further polymerisation of dicyclopentadiene results in the production of polycyclopentadiene an insoluble infusible white crystalline powder decomposing a t 285". The Constitution of the Disulphoxides. I. S. SMILES and D. T. GIBSON ( J . Chem. Xoc. 1924 125 176-183).-The various forrnuh given to the disulphoxides at different times are discussed and an attempt is made to decide between the symmetrical and unsymmetrical f ormulz R*SO,*SR and R-SO*SO*R.Considered as a whole the evidence a t present available favours the unsym- metrical type. The reaction of mercaptans with disulphoxides is studied and using a mercaptan R'*SH where R' is different from the group R in the disulphoxide it is found that high yields of tlhe sulphinic acid R*SO,H are obtained with it mixed disulphide R'S*SR but no sulphinic acid R'*SO,H. 2 5-DichZorobe7zxene- sulphinic acid C,H,CJ-SO,H forms crystals m. p. 122". 2 5 2' 5'-TetrachlorodiphenyZ disulphoxide (C6H3Cl,)2S,0 crys- tallises in needles m. p. 128"; 3 3'-dimrboxydiphenyl disulph- oxide (C,H,*C02H),S202 has m. p. 229" ; 2 5-dichloro-4'-methyZ- diphenyl disulphide forms needles m. p. 71-72' ; phenyl-2-mphthyl disulphide m.p. 74-75" ; 2 5-dichloro-3'-nitrodip~enyl disulphide needles m. p. 89-91 O ; O-anthryl-4-tolyl disulphide bright yellow needles m. p. 98-99" ; 3-carboxyphenyl-4-tolyl disulphide crystals m. p. 140". C. H. LOWE and C. JAMES (J. dnier. Chem Xoc. 1923 45 2666-2669).-When using the apparatus previously described ( J . Amer. Chem. Soc. 1917 39 933) for the preparation of diphenyl diiliculties arise owing to local heating and sagging of the filament of the heating element with resulting short-circuiting. A filament support is now described in which these defects are eliminated. It is claimed that by maintaining the filament at st yellowish-red heat very good yields of diphenyl may be obtained from commercially pure benzene which need not be anhydrous. The addition of more water does not hasten the reaction.w. s. N. Anthracenemonosulphonic Acids. Sulphonation of Hydro- carbons in a Basic or Neutral Medium. M. BATTEGAY and P. BRANDT (Bull. SOC. chim. 1923 33 [iv] 1667-1678; cf. A. 1922 i loOl).-In the direct sulphonation of anthracene by means of fuming sulphuric acid or chlorosulphonic acid the separation of the two isomeric anthracenemonosulphonic acids is effected by means of the barium in preference to the sodium salts. The position of the entering sulphonic group was determined by con- version of the acids into the corresponding a- and p-chloroanthra- quinones under the influence of nascent chlorine. The two acids are formed in almost equal proportions; elevation of temperature does not appear to favour a greater production of the P-isomeride E.H. R. F. G . P. Preparation of Diphenyl.i. 276 ABSTRACTS OF CHEMICAL PAPERS. as in the case of naphthalene (Fierz and Weissenbaqh A. 1920 i 430) although accurate observation is diflticult by reason of the disulphonation of anthacene at temperatures above 100". When however anthracene- a-sulphonic acid was heated in 96% sulphuric acid a t temperatures ranging from 150-180" and subsequently transformed into dichloroanthraquinone the product appeared to consist wholly of the 1 5- and 1 8-isomerides thus indicating that no shifting of the sulphonic group from the a- to the p-position had occurred. Sulphonation in presence of mercury mercurous sulphate or mercurous oxide resulted in a slight increase in yield without any appreciable effect on the relative proportions of the isomerides.Attempts at sulphonation in a neutral or basic medium (cf. Battegay and Brandt A. 1922 i 1001) were made pyridine being used as the medium in place of the acetic acid em- ployed in t'he work described above. Experiments were carried out over a temperature range of 95-115" and with varying pro- portions of the solvent. The best yields were obtained when a portion of the pyridine was replaced by a neutral solvent of higher b. p. In the presence of petroleum at 165-175" almost the whole of the resulting anthracenesulphonic acid is the a-isomeride and the addition of mercury compounds has no appreciable influence on the yield or the course of the reaction. When nitrobenzene is used the yield is considerably smaller.A brief note on the applic- ation of Wagner's theory (A. 1886 708) to the mechanism of the reaction is appended. Anthracene-a-sulphonic chloride m. p. go" was prepared but probably in an impure condition; on being heated with alcoholic ammonia it yields the corresponding sulphonamide brown crystals m. p. 205" (cf. Heffter A 1895 i 671). H. J. E. Reactions of Strongly Electropositive Metals with Organic Substances in Liquid Ammonia Solution. Tv. Action of the Alkali Metals on Triphenylmethyl and its Compounds. C. A. KRAUS and T. KAWAMURA ( J . Amer. Chem. Soc. 1923 45 2756- 2763) .-The strongly electropositive metals react with triphenyl- methyl chloride in liquid ammonia forming the compound of the metal with the triphenylmethyl group e.g. CPh,-Na. The sodium compound has been obtained in red needles.With ammonium chloride in liquid ammonia the sodium derivative gives an immediate pink precipitate which turns white on standing owing to production of triphenylmethane CPh,*Na + CPh,-NH + CPh,H+NH,. The sodium compound is unstable but the potassium derivative is stable even at 100". Triphenylmethyl is obtained by the action of triphenylmethyl chloride on sodium triphenylmethyl in the presence of toluene; this reaction does not take place if the sodium compound is replaced by the potassium derivative. Bromo- benzene converts sodium triphenylmethyl into tetraphenylmethane. The mechanism of the reduction of organic halides by means of the alkali metals in liquid ammonia solution is discussed. F. G. P.OROA.N'X(3 CHEMISTRY. i.277 The Preparation and Properties of Organic Stannichlorides. VI. The Action of Sulphuric Acid on certain Stannichlorides. The Formation of Stannisulphates. J. G. F. DRUCE (Chem. News 1924 128 33-34; cf. A. 1922 i 1206).-Stannisulphates of some organic bases can be obtained by the action of concentrated sulphuric acid on the corresponding stannichloride. The stanni- sulphate crystallises from the sulphuric acid after the addition of ether. Aniline stannimlphute ( PhNH2),,H2Sn( SO,) forms grey deliquescent crystals. m-Phenylenediamine stannisulphate does not melt below 300". forms colourless deliquescent needles m. p. 124". Potassium and calcium stannisulphates can be prepared in the same way from the s tannichlorides. E. H. R. The Mechanism of the Hofmann Rearrangement of Methyl- aniline Hydrochloride.J. W. HOWARD and C . G. DERICK (J. Amer. Chem. Xoc. 1924 46 16&177).-The investigation was undertaken to confirm Hofmann's original views as fo the nature of the change (Ber. 1871 4 742). It was found that in the re- arrangement methyl chloride dissociates hom the methylaniline hydrochloride and combines with a second molecule forming phenyl- trimethylammonium chloride ; this compound then undergoes re- arrangement. No rearrangement takes place at 220-250' unless the heating is continued for ninety-six hours. At 300" however rearrangement takes place and is somewhat affected by time but temperature is the chief factor. Neither methylaniline nor dimethylaniline rearranges. A quantitative separation and deter- mination of the three classes of amine by means of benzenesulphonyl chloride has been- worked out. pBromophenyltrimethylammonium Perhalides.T. H. READE ( J . Chem. Soc. 1924 125 148-157; cf. T. 1923 123 141) .-The degradation of perhalides by acetone has yielded analogous quaternary ammonium salts in the case of this series the corresponding p-iodo series and the simple phenyltrimethyl- ammonium series itself the part removed by acetone being either iodine chloride or bromide or chlorine. The points of Werence between the p-bromo-series and the unsubstituted series are that p-bromophenyltrimethylammonium bromide di-iodide yielded the p-bromophenyltrimethylamxnonium bromide on treatment with acetone whereas its analogue in the unsubstituted phenyl series was not attacked and that well-defined double perhalides have been obtained in several reactions.Chlorine replaces N-halogen in iodides and periodides replaces N-bromine only when iodine is also present and replaces added halogens in perbromides. Both bromine and iodine replace N-chlorine to an appreciable extent and iodine replaces N-bromine in one instance. p-Bromophenyl- trimet h y hmmonium met h y 1 sulphate C ,H,Br*NMe,*SO,Me white crystals m. p. 208O when treated with bromine in hydrobromic C,H*(NH2),,H,Sn(SO,) (C9H,N)2,H8n(SO,) Quinoline stannisulphute F. A. M.i. 278 ABSTRBCTS OF CHEMICAL PAPERS. acid solution gives the perbromide C,H4Br*NMe3Br3 orange-red leaflek M. p. 175". When warmed with acetone this perbromide yields the bromide c,H,Br*Nl\de,Br m. p.210". Iodine in acetic acid gave C6H4BreNMe,Br,~r ; chlorine similarly yielded C,H,Br*NMe,Br*C~ lemon-yellow plates m. p. 158". The bromo-dichloride the per- bromide and the bromodi-iodide are obtained by the action of halogens on the simple bromide of m. p. 210". Chlorine with the p-bromophenyltrimethylammonium iodide gives the chloride iodo- trichloride C,H4Br*NMe3C1,1C1 m. p. 187" bromine gives the bromoiodobromide m. p. 185" iodine the tri-iodide m. p. 172". The chloride iodochloride C,H,Br*NMe,a,Ia m. p. 177" is obtained by the action of chlorine on the simple iodide and on treatment with bromine gives a yellow double perhalide C6H4Br*NMe3C1 Ic1+ c ,H4Br*NMe3Br,ICl. Acetone converts the chloride iodochloride into the simple chloride m. p. 199"; this with iodine gives the double compound C6H4Br*NMe313 + C6&Br*me3I,.The bromide iodobrornide c6H4Br*me3Br,'IBr obtained by the action of bromine on the iodide crystallises in orange plates m. p. 189" (decomp.). With chlorine it yields C,H,Br*NMe3c1,1c?13 ; bromine has no action iodine gives a red double perhalide C,H,Br*NMe,Br,IBr + C ,H4Br *NMe,BrI,. The bromide d i -iodide C 6H4Br *NMe,BrI obtained by the action of iodine on the bromide has m. p. 172" (decomp.). With chlorine it yields yellow needles of c ,H,Br*NMe,a Ia m. p. 186" with bromine the bromide iodobrornide m. p. 189" and with iodine the tri-iodide. The bromide iodochloride C,H4Br*NMe3Br,IC1 is obtained by the action of iodine mono- chloride on the bromide in the form of yellow crystals m. p. 178- 179". With chlorine it yields C,HqBr*NMe3C1,1C1 m.p. 186" ; with bromine a mixture of the bromide iodobromide and bromide iodochloride and with iodine a mixture of bromide di-iodide and bromide iodochloride. The bromide dichloride C,H,Br*NMe3Br,C1 results by the action of chlorine on the bromide. It forms greenish- yellow crystals m. p. 158". Neither chlorine nor bromine has any action on i t ; with iodine it yields the bromide iodochloride m. p. 178". p-Iodophenyltrimethylamrnonium Perhalides.? T. H. READE and S. A. SIM ( J . Chem. Soc. 1924 125 157-160).- Chlorine converts p-iodophenyltrimethylammonium iodide into the chloride iodotrichloride C,H,INRle3C1*ICl and this can be degraded by ethyl malonate yielding the dichloro-iodide C,H,I*NMe,Cl*IC1. With hot glacial acetic acid the dichloro- iodide yields a double perhalide C,H,I*NMe3~I+ C,H41*NMe,C112.p-Iodophenyltrimethylammonium chlorade iodoirichloride C,H,I*NMe,C1eIC1 crystallises in yellow needles m. p. 187"; on treatment with ethyl malonate under reduced pressure it gives the chloride-iodochbride yellow crystals m. p. 184'. The tri-iodide crystallises in brow11 LF. G. I?.ORGANIC CHEMISTRY. i. 279 leaflets m. p. 189". A method for determining the ionisable halogen in the perhalides is given Synthesis of Indanylamine and of its N-Substituted Derivatives. C. COURTOT and A. DONDELINGER (Cmpt. re?zd. 1924 178 493495; cf. A. 1923 i 1090).-1-Chloroindane is conrerted by dry liquid ammonia in a closed vessel a t the ordinary temperature into 1-indanylamine b. p. 96-97"/8 mm.and a little di-indanylamine m. p. 84-85'. Methyl- dimethyl- ethyl- and diethyl-amines similarly convert chloroindane into methyl- indanylamine b. p. 106-107"/15 mm. dimethylindanylamine b. p.1 00°/10 mm. ethybindanylamine b. p. 106-107"/7 mm. and diethyZindanyZami?ze b. p. 112-113"/8 mm. respectively all four amines being colourless liquids. Indene on hydrogenation in presence of platinum black affords indane (in 97% yield) which does not react with magnesium ethyl bromide at 135" (cf. Grignard and Courtot A. 1911 i 193). The reactivity of the methylene group in indene is therefore due to the presence of unsaturation in the 5-membered ring. Preparation of 3 3'-Diaminodiphenylsulphone by Electro- lytic Reduction of 3 3'-Dinitrodiphenylsulphone. J. LACROIX (Compt.rend. 1924 178 483486) .-The electrolytic reduction of 3 3'-dinitrodiphenylsulphone to the diamino-deriv- ative is best effected (yield 94.5%) by the use of an alcoholic solution of the dinitro-compound acidified wth hydrochloric acid as the cathode liquid with a cathode consisting of nickel foil covered with spongy tin The anode is of lead immersed in 10% sulphuric acid. The cathode liquid is maintained at a temperature of 68-72' and well agitated. A. B. H. J. TAYLOR and A. E. DIXOX ( J . Chem. Soc. 1924 125 243-250).-Neither 0- nor p-chloronitrobenzene nor 2 5-dichloronitrobenzene combines with thiocarbamide. 1 -Chloro-2 4-dinitrobenzene unites with thiocarbamide and with phenylthiocarbamide to form additive salts from which alkali hydroxides eliminate hydrogen chloride the dinitrophenyl residue breaking off as dinitrophenyl mercaptan.With thiocarbanilide chlorodinitrobenzene gives mainly phenyl- thiocarbimide and dinitrodiphenylamine (NO2),C6H3-NHPh. Picryl chloride unites with thiocarbamide to form an additive compound from which a picrate and a sabicylate were obtained Phenylthiocarbamide gave little additive salt thiocarbanilide none. The picryl compounds are unstable decomposing readily into dipicryl sulphide and a dark-coloured powder. In no case was an N-substituted thiocarbamide obtained the behaviour of dinitro- chlorobenzene and picryl chloride thus more resembling that of an alkyl halide than of an acid chloride. The picrates of the compounds obtained from 1-chloro-2 4-dinitrobenzene with thio- carbamide and phenylthiocarbamide are yellow compounds m.p. 192-193" and 160-161" respectively. The picrate derived from picryl chloride and thiocarbamide has m. p. 149-150" the salicylate m. p. 76". F. G. P. E. E. T. Chloronitrobenzenes and Thiocarbamides. F. G. P.i. 280 ABSTRACl!S OF CHEMICAL PAPERS. The Decomposition of Substituted Carbamyl Chlorides by Hydroxy-compounds. I. The Reaction between Phenyl- methylcarbamyl Chloride and Ethyl Alcohol at Different Temperatures. T. W. PRICE ( J . Chem. Soc. 1924 125 115- 129) .-The reaction between phenylmethylcarbamyl chloride (m. p. 87.5") and alcohol is very slow a t temperatures below the boiling point of alcohol and was consequently studied in sealed tubes at loo" W" 80" and 70". Under these conditions the carbamyl chloride first reacts with the alcohol to form phenylmethylurethane together with hydrogen chloride which then reacts further with alcohol to form ethyl chloride and water.The first reaction was studied separately by adding dimethylaniline to the reaction mixture to combine with the hydrogen chloride and prevent the formation of ethyl chloride. The velocity of the reaction between the alcohol and the carbamyl chloride was found to vary with the amount of dimethylaniline added being slower in presence of more base. Three series of experiments were therefore made at 100" with different amounts of dimethylaniline and the velocity constant of the reaction in absence of dimethylaniline was found by extra- polation to be k,=0.0426 a t 100". As this value was in good agreement with the value obtained by calculation from the study of the combined react,ion measurements in presence of varying quantities of dimethylaniline were omitted a t the other temper- atures.The position of the reaction under these conditions was determined by titration of the dimethylaniline hydrochloride present a t any given moment the initial concentration of the carbamyl chloride in the alcohol being in all cases N / 2 . The action of hydrogen chloride ( N / 2 ) on ethyl alcohol was investigated separately. The velocity const4ant fell steadily a t each temperature the means of the first three values being k = 0.0102 a t lOO" 0-00443 a t 90" 0.00162 at 80" and 0.000521 at 70". The whole reaction was then studied by determining the amounts of carbamyl chloride hydrogen chloride and ethyl chloride present at any instant.The hydrogen chloride was determined by direct titration and the carbamyl chloride by hydrolysis with excess of potassium hydroxide after removal of the ethyl chloride followed by titration with silver nitrate. The velocity constants of the first reaction calculated from the amounts of carbamyl chloride present were k =0.0427 a t loo" 0.0196 a t go" 0.00873 a t 80° and 0.00329 a t 70" Expressions deduced for the amounts of the various reactants present at any time from the values of k and k2 as well as for the maximum amount of hydrogen chloride present and the time taken for this maximum to be reached gave satisfactory agreement with experimental results and the variation of temper- ature coefficient for each reaction was shown to be in accordance with Lewis' deductions.F. G. W. Esterification of cycZoHexano1 and of some of its Homo- logues. (MILE.) G. CAUQUIT (Gmpt. rend.,i1924 178 323- 326).-cycZoHexanol 2-methylcyclohexanol or 3 4-dimethylcyclo- hexanol was heated at 95"; with an equivalent weight of aceticORGANIC CHEMISTRY. i. 281 acid until equilibrium was attained when it was found that 55-6% 49*8% and 47.5% of the three alcohols respectively had under- gone esterification. E. E. T. The Catalytic Hydrogenation of Organic Fluoro-derivatives. III. The Hydrogenation of a-TrifluorcMn-cresol. F. SWARTS (Bull. Xoc. chim. Belg. 1923 32 367-376; cf. A. 1921 i 656).- The hydrogenation of w-trifluoro-m-cresol in the presence of platinum black proceeds in two directions one resulting in the formation of w-trifluoromethylcyclexanol the other in the pro- duction of w-trifluoromethylcycbhexane and water the group CF3- being unaffected. The relative proportions of the cyclo- hydrocarbon and cycloalcohol are not constant increase of tem- perature favouring the production of the hydrocarbon. In the case of w-trifluoro-m-cresol a thermal coefficient of acceleration is observed whilst the velocity of absorption of the hydrogen by the trifluoromethylcyclexane varies little with the temperature.An attempt is made to deduce a formula by means of which the thermal coefficient of the reaction can be calculated. TriJluoromethylcyclohexan-3-ol CF3*C6H,,*OH is a viscous liquid b. p. 183-184"; its benxoyl derivative crystallises in prisms m.p. 55" d17 1.2611. The acetyl derivative has b. p. 1924--194". TriJEuoromethylcycloh~un-3-one is obtained by oxidising the come. sponding cyclohexane with chromic acid; it boils a t 173-174"/ 746 mm. d16 1.242. Its semicurbaxone melts at 190". F. G. P. C. A. TAYLOR and W. H. RINKENBACH (U.S. Bur. Mines Repts. Investigatim 1923 No. 2533).-The lead derivative of trinitroresorcinol is obtained in 94.4% -@eld as fine light orange-coloured crystals d 3-09 by mixing bodmg solutions of lead nitrate trinitroresorcinol and sodium carbonate to which glacial acetic acid has been added. Its explosion temperature is 293" and rate of detonation 2097- 2209 m. per second. The Influence of Hydrogen Concentration on the Auto- oxidation of Quinol. A Note on the Stability of the Quinhydrone Electrode.V. K. LAMER and E. K. RIDEAL ( J . Amer. Chem. Xoc. 1924 46 223-231).-The rate at which quinol reacts with molecular oxygen has been followed by shaking borate buffers 0.01M with respect to quinol with oxygen. The rate does not become appreciable until the solution is more alkaline than pH 7-3 or 7-8 for air. The rate of reaction is highly sensitive to further increase of pE value being proportional to [H+I3B where [IT+] is the hydrogen-ion concentration for the region investigated. A mechanism involving complex formation of the primary and secondary ions is suggested to explain this relation- plex ion; complex+20 + 2 quinone+H,O,. F. A. M. Arylazides. 111. Quinol Ethers from p-Methylated Aryl- azides. E. BAMBERGER and J.BRUN (Helv. Chim. A&; 1924,7 112-122).-A detailed account of experiments the results of which have already been recorded (A. 1921 i 721). The Lead Derivative of Trinitroresorcinol. CHEMICAL ABSTRACTS. Ship such as 2H,C,H,O + 3H++HC6H~O-/-C6H,0n + com- E. H. R.i. 282 ABSTRACTS OF CHEMICAL PAPERS. Arylazides. IV. Iminoquinol Ethers from p-Methylated Arylazides. E. BAMBERBER J. BRUN and A. HARTMANN (Helv. Chim. Achy 1924 7 123-131).-A detailed account of work already published (A. 1921 i 721). Electrochemical Oxidation of Alkyl Ethers of Phenol and of the Three Isomeric Dihydroxybenzenes. F. FICHTER and W. DIETRICH (Helv. Chim. Acta 1924 7 131-143).-Anodic oxidation of anisole in suspension in dilute sulphuric acid using a lead peroxide anode results in a high yield of benzoquinone up to 71% of theory under favourable conditions.This result is obtained using a diaphragm; in absence of a diaphragm a mixture of quinhydrone and quinol is obtained. Quinol monomethyl ether is probably formed as an intermediate product since this is converted practically quantitatively into benzoquinone by anodic oxidation. The methyl group is oxidised to methyl alcohol. When a platinum anode is used the yield of benzoquinone is very small; some formaldehyde is formed but the principal product is carbon dioxide. Phenetole and phenyl isoamyl ether give smaller yields of benzoquinone whilst the alkyl group is in each case oxidised to the corresponding acid. Quinol dimethyl ether gives benzoquinone the best yield being 48.6% of theory.By the oxidation of guaiacol using a diaphragm a t first only fumaric acid could be obtained; this in absence of a diaphragm wills reduced to succinic acid. By stopping the oxidation a t an early stage however an oily product was obtained from which a diphenyl derivative diguaiacol was isolated. This is probably 4 4'-dihydroxy-3 3'-dimethoxydiphenyl; i t crystallises in slender needles m. p. 166" and gives an orange colour with ferric chloride. Its diacetyl derivative crystallises in leaflets m. p. 198" and its dibenxoyl derivative in white needles m. p. 203". By demethyl- ation it gives a tetrahydroxydiphenyl ( ? 3 4 3' 4') needles m. p. 134" giving an emerald-green coloration with ferric chloride. Its tetra-acetyl derivative has m. p. 161". Further methylation of diguaiacol converts it into diveratrole or tetramethoxydiphenyl colour- less tablets or leaflets m.p. 130". Anodic oxidation of veratrole gives rise to some diveratrole but in addition partial demethyl- ation occurs with formation of a hydroxytrimethoxydiphenyl which gives an acetyl derivative pale yellow leaflets m. p. 116O. Anodic oxidation of resorcinol monomethyl ether gives an amorphous dihydr- oxydimethoxydiphenyl which when demethylated gives 2 4 2' 4'- tetrahydroxydiphenyl (2 4 2' ; 4'-diresorcinol). The Catalytic Condensation of Acetylene with Phenols. H. H. WENZKE and J. A . NIEUWLAND ( J . Amer. Chem. Soc. 1924,46,177-181).-Acetylene condenses with phenols and cresols in 95% alcoholic solution in presence of sulphuric acid and mercury salts to form uncrystallisable resins.The di- and tri- hydroxybenzenes react even more readily. With p-naphthol acetylene condenses to form an acetal and some ethylidenedi- p-naphthol. Resorcinol condenses with acetylene to form ethyl- E. H. R. E. H. R.ORGANIC CHEMISTRY. i. 283 idenediresorcinol (as-2 4 2' 4'-tetmhydroxydiphenykthum) which loses water spontaneously to form 3 6-dihydr~y-g-rnetItylxanthene soluble in water and alkalis (decomp. 230-240"). On heating with stannic chloride a t 160-180" it oxidises to form 6-hydroxy- 9-medhyZJEuorone the solutions of which resemble that of fluorescein. The presence of nitro- and sulphonic groups prevents the absorption of acetylene by a phenolic compound. No cases have been noted where the acetylene substitutes in a meta-position to another group.Phenetole does not react with acetylene under the above conditions. F. A. M. The Hydrobenzoin Rearrangement. I. Hydroanisoin. M. 'J~FFENEAU and A. OR~KEIOFF (BUZZ. Xoc. chim. 1923 33 [iv] 1832-1838).-According to Russell (AnnuZen 1869 151 25) hydroanisoin on dehydration affords deoxyanisoin (m. p. 95"). Irvine and Moodie (T. 1907,91,536) found for the latter compound the m. p. 108-109". It is now shown that Russell's compound is dknisyZacetuZdehyde (OMe*C,H,),CH*CHO as would have been expected (cf. Orekhoff and Tiffeneau A. 1921 i 566). The alde- hyde (needles m. p. 10&105"; semimrbamne m. p. 140-141") is also obtained by treating magnesium p-anisyl bromide with ethyl ethoxyacetate and the product of this interaction with formic acid. On oxidation the aldehyde affords dianisyl ketone and with magnesium phenyl bromide gives a-phenyl- PP-di-p- anisyZdhyZ alcohol (needles m.p. 139-140"). Hydroanisoin (m. p. 170-171") on heating with 50% sulphuric acid affords the above aldehyde isohydroanisoin (m. p. 125-126") behaving similarly. Dianisyl ketone with magnesium methyl iodide is converted into diunisylethykne (OMe*C,H,),C:CII[ (m. p. 142-143") t,he latter when treated with mercuric oxide in presence of water and iodine affording an iodohydrin (not isolated) which is converted by silver nitrate into deoxyanisoin. E. E. T. Semi-pinacolic Transformation. Necessity for the Presence of the Phenyl Group in Semi-pinacolic Transform- ations. (-.) J. L h y (Bull. Xoc. chim. 1923 33 [iv] 1655- 1666; cf.Tiffeneau and LBvy A. 1923 i 788 789).-In the aromatic series the stability of the secondary hydroxyl group in respect of semi-pinacolic transformations appears to be conditioned by the nature of the aryl group linked to the secondary carbon atom. The author has studied the dehydration of six tri-substituted glycols and draws the conclusion that in addition to transpositions involving only phenylic groups in which the shifting of a group occurs only when the group capable of migration is arylic others take place only in the cyclic series but this is limited to cases in which the aryl group occupies one definite position. In these cases the migratory group may be either cyclic or acyclic. Experi- mental details of the dehydration of the glycols are given (cf.A 1921 i 860). The following substances are described By-dz-i. 284 ABST&boTs OF OHEMICAL PAPERS. hydroxy-a-phenyl-y-methylbutane needles m. p. M 5 " ; ethyl benzylgZywlZute b. p. 170-173"/30 mm.; benzyl isopopyZ ketone b. p. 234-235" ; semhm-bame m. p. 140-141" ; a-phenyl-y-ethyl- pentane-By-diol small needles m. p. 79-80" yielding on dehydration by means of sulphuric acid an oil b. p. 160-170"/50 mm. which forms a semicarbazone m. p. lf30-161" ; as-diphenyl-y-benxyl- butane- py-diol crystals transformed by concentrated sulphuric acid into a substance C,,H m. p. 94-95" of unknown con- stitution ; a-methy2hexane-p-ol b. p. 140-142"/760 mm. prepared by the action of magnesium butyl bromide on acetone yields on dehydration a methyl-AP-hexene b.p. 9&98"/760 mm. which on treatment with hypoiodous acid yields an iodohydrin; this is converted into a ketone on treatment with silver nitrate without undergoing any transposition. 3 4D;hydroxyphenylmethylaminoethanol (Adrenaline). I. HOSHINO and D ~ C H I SEIYAKU KABUSHIKI KAISHA (Japan. Pat. 42351).-Adrenaline is prepared from methylaminoacetyl-3 4- dihydroxybenzene by reducing with hydrogen using as a catalysf colloidal solutions of metals of the platinum group the finely powdered metals or nickel or carbon bearing these metals. The product is purer than that obtained by Stolz's method and the yield almost theoretical. K. K. H. J. E. Cyclic Trimethylene [cycZoPropane] Compounds. P. BRUYLANTS (Bull. SOC. chim. Belg. 1923 32 358--367).-Bromin- ation (in direct sunlight) of cycEopropane-1 -carboxylic chloride until the liquid boils a t 135-140" followed by treatment of the product with methyl alcohol gives a mixture of methyl cyclo- propane-1-carboxylate methyl 1-bromocyclopropane-1-curboxykcte (a colourless mobile liquid b.p. 168"/702 mm. @ 1.545 nz 1-4784) and methyl ay-dibromobutyrate. The latter (b. p. 113-114'114 mm.) is the only product of bromination in the dark and was identified by conversion into ay-dibromobutyramide (m. p. 80-82"). cyclo- Propane-1 -carboxylic anhydride which on bromination gives only open-chain derivatives is a colourless liquid b. p. 232"/752 mm. diJ 1.1132 n!$' 1-4628. The corresponding ethyl ester has b. p. 178"/758 mm. dy 1.428 ng 1.4678. The esters are converted by ammonia into the amide (needles m.p. 108") and by concentrated sodium hydroxide into the acid m. p. 77". Mhgnesium methyl bromide converts the methyl ester into the bromhydrin (b. p. 171-172"/762 mm.) the latter with acetic anhydride giving the broww-acetin (b. p. 180-182" clr 1.327 ng 1.4731) and with alkali hydroxide the oxide FH2>C-CMe2 b. p. 121-125" d 0.886 nz 1.419. Digestion of the oxide with dilute sulphuric acid followed by addition of semicarbazide gives a semicurbccxone ( ?) m. p. 115- 140" (very indefinite). Electrolysis of potassium cyclopropane- carboxylate yields the free acid and a little cyclopropane. CH \o/ F. G. P.o w m a OHEMISTRY. i. 285 Preparation and Properties of some Organic Uranyl Salts and in ppticular of Uranyl Benzoates axid Sdlicylates.G. COURTOIS (Bull. Soc. chim. 1923 33 [iv] 1761-1773).-A continuation of previous work (A. 1914 i 799 802). When the hydroxide U03,H20 is dissolved in a hot aqueous solution of benzoic acid and the solution evaporated uranyl benzoate (Ph*CO,),UO (yellow crystals stable at looo) separates together with benzoic acid which may be extracted with benzene. On further evaporation the acid salt (Ph*CO,),UO,,Ph-CO&€ separah the latter salt also being obtained if uranyl hydroxide is dissolved in boiling benzoic acid. Uranyl benzoate is sparingly soluble in water the solution depositing a basic salt on standing. Below Go the saturated aqueous solution deposits the dihydmte which becomes anhydrous on keeping at the ordinary temperature and dissolves in ether fo give a solution which rapidly deposits a yellow crystalline compound (Ph~C0,)2U02,2Et,0. The latter is also formed when a concentrated aqueous solution of the benzoate is ehaken with ether and slowly loses ether of crystallisation on keeping (rapidly at 100").Uranyl hydroxide dissolves in a hot alcoholic solution of benzoio acid the compound (Ph*CO,),UO,,EtOH (pale yellow crystah from which ether displaces the alcohol of crystallisation) separating on cooling. This salt on cryetallisation from water or from chloro- form or on heating a t lOO" loses its alcohol of crystallisation. When aqueous solutions of urmyl benzoate are concentrated by protracted (hot) evaporation a yellow crystalline basic salt (PhC0,),U0,,U0,,H20 separates and may be freed from accom- panying uranyl benzoate by extracting the latter with alcohol.Prolonged heating with water converts the biasio salt into uranyl hydroxide. Aqueous solutions of uranyl benzoate are unaffected by sun- light whilat alcoholic solutions afford a violet precipitate of uranoso-uranic oxide the latter becoming yellow on washing with water containing dissolved air. Uran yl sulk yhte ( OH*C,H,*CO,),UO separates as orange-yellow crystals readily soluble in methyl or ethyl alcohol from the warm solution obtained by heating uranyl hydroxide with aqueous salicylic acid. The mother-liquors on cooling deposit the dihydrate of the above salt in the form of yellow crystals giving red solutions in methyl or ethyl alcohol and becoming anhydrous a t 140-150" with slight decomposition. Aqueous solutions of uranyl salicylate on boiling acquire a red colour.Neither aqueous nor alcohoh solutions of the sdt are affected by sunlight a fact that distin- guishes the salt from all other uranyl salts studied. E. E. T. Stability of Solutions of some Organic Uranyl Salts in Absence or in Presence of Light. G. COURTOIS (Bull. Xoc. chirn. 1923 33 {iv] 1773-1785).-Previous results (A. 1914 i 799 and 802; and preceding abstract) are re-stated and slightly extended. It is shown that saturated aqueous solutions of the VOL. CXXVI. i. ii. 286 ABS!CRACTS Or CHEMICAL PAPEBS. following uranyl salts are unstable even in absence of light giving crystalline basic salts (already described) Formate acetate benzoate glycollate and citrate. The other uranyl salts examined (propionate butyrate lactate quinate salicylate oxalate and tartrate) are stable in the dark in aqueous solution.Sunlight in absence of air causes the reduction of all the above salts with the exception of the benzoate and salicylate. The stability of the latter may be due to the " antioxygen " property of the salicylate residue. Contrary to the view of Aloy and Rodier (A 1920 ii 182) decomposition by sunlight does not depend on the hydroxylic character of the salt as in the case of the tartrate since uranyl propionate butyrate etc . are decomposed in aqueous solution under the influence of sunlight. I n these cases the violet hydrated uranoso-uranic oxide (converted by washing with water into the oxide U0,,2H20) is precipitated. Alcohol (or ether) accelerates these decompositions but does not initiate them.In the decomposition of uranyl salts of monobasic aliphatic acids carbon dioxide and a hydrocarbon result (the formate giving carbon monoxide and dioxide). The other salts examined give rise to carbon dioxide except the oxalate which also affords the mon- oxide. In the case of the glycollate (lactate) formic acid and formaldehyde (acetaldehyde) are first formed. The Isoelectric Point of m-Aminobenzoic Acid -and its Equilibrium with Water Acetic Acid and Sodium Acetate. J. BARBAUDY (Bull. SOC. chim. 1924 [iv] 35 31-37).-The equilibria in the systems (1) m-aminobenzoic acid acetic acid and water and (2) m-aminobenzoic acid acetic acid sodium acetate and water have been studied at 25' by determinations of hydrogen ion concentration density and viscosity in addition to solubility.From the solubility of m-aminobenzoic acid in pure water the pn of the isoelectric point for this acid is found to be 3.886. System (1) gave concordance between calculated and found iso- electric points whilst in system (2) concordance was less good. [Cf. Michaelis and Davidsohn A. 1911 ii 192.1 Preparation of Benzyl Diphenylalkylacetates and the Corresponding Acids by Means of Sodamide and Alkyl Iodides. (MME.) P. RAnlART ( h n p t . rend. 1924,178,396-397 ; cf. this vol. i 17l).-Benzyl diphenylacetate was dissolved in ether and heated with powdered sodamide for several hours an alkyl halide then being added and so on. The following new compounds (in addition to some previously known) have been prepared by this method Benxyl aa-diphenylpropionate prisms m.p. 71-72' b. p. 230-233"/10 mm.; benzyl aa-diphenyl- P-methylbutyrate m. p. 73-74' b. p. 225'12 mm.; aa-diphenyl- Ay-pentenoic acid prisms m. p. 142" (benzyl ester b. p. 240- 242'jlO mm.; arnide m. p. 90-92'). E. E. T. The Methods of Preparation of 6-Chloro-3-hydroxybenoic Acid. V. I. M~NAEV and K. M. R~PPER ( J . Russ. Phys. Chem. Soc. 1924 54 673-679).-A further account of work already published (A. 1922 i 162). E. E. T. E. E. T. G. A. R. K.ORGANIC CHEMISTRY. i. 287 p-Phenoxypropionic Acid and some of its Derivatives. Chromanone. S. G. POWELL ( J . Amer. Chem. Soc. 1923 45 2708-271 1) .-In preparing y-phenoxypropyl alcohol Ph* O*[ CH,],*CH,OH (Rindfusz A. 1919 i 342) an aqueous solution of sodium phenoxide in which trimethylenechlorohydrin is completely soluble may be used in place of an absolute-alcoholic solution; in fact a slightly better yield (80%) is actually obtained by this method. p-Phenoxy- propionic acid Ph*O*[CH,],*CO,H may be obtained in yield varying between 32% and 45% by the gradual addition of potassium permanganate solution to a suspension of p-phenoxypropyl alcohol in a solution of magnesium sulphate. Its ethyl ester a colourless oil has b.p. 17Oo/aO mm. The acid chloride a liquid of irritating odour is obtained when the acid is gently warmed for a few seconds with only a slight excess of thionyl chloride ; it reacts with aqueous ammonia giving the amide glistening needles m. p. 119". The action of phosphoric oxide in boiling benzene solution on the acid * .L v gives chromanone C6H4< O-gH2 a colourless highly refracting COO H,' liquid b. p. 160°/50 mm. with pleasant lemon-like odour; it solidifies on keeping to a mass of crystals m. p. 38.5". Chromanone is also formed together with a fraction of b. p. 237"/50 mm. (which is free from halogen) by the prolonged action of an excess of thionyl chloride on p-phenoxypropionic acid. Chromanone forms an oxime glistening needles m. p. 140" and a benzylidene derivative long The Isomeric Esters of p-Methoxybenzoylacrylic Acid. G. P. RICE ( J . Amer. Chem. Soc. 1924 46 2 1 6 2 2 3 ; cf. A. 1923 i 218).-In continuation of previous work on the isomeric esters of benzoylacrylic acid the isomeric methyl and ethyl esters of p-methoxybenzoylacrylic acid have been examined.The stable yellow forms when exposed to sunlight in benzene solution are converted into colourless isomerides the reverse change taking place on exposing solutions of the colourless solids containing a crystal of iodine to sunlight. p-Methoxybenzoyiacrylic acid itself is transformed by sunlight to a polymeric acid. The yellow unsaturated esters are much more readily hydrolysed by sodium carbonate than the colourless isomerides the product being methoxybenzoylacrylic acid. With cold concentrated or boiling dilute hydrochloric acid both isomeric esters give a mixture of unsaturated acid and a-hydroxy-p-methoxybenzoylacrylic acid. The esters of the p-methoxy-acid do not behave in the same way as the esters of the unsubstituted acid on treatment with semi- carbazide.p-Methoxybenzoylacrylic acid itself yields a semi- carbazone ; the colourless esters react with semicarbazide to form additive products and semicarbazones of these additive products. Methyl p-methoxybenzoylacrylate yellow needles m. p. 71-72" ; wlourless isomeride needles m. p. 62". Ethyl p-methoxybenzoyZ- acrykte large yellow crystals m. p. 4 2 4 3 " b. p. 210-215"/11 mm.; colourless isomeride m. p. 46-47" Methyl a@-dibromo- yellow needles m. p. 112". w. s. N. 12i. 288 ABSTRACTS OF CHEMICAL PAPERS. p-methoxybenxoylpropionate from the yellow ester m. p. 115". Ethyl ap-d~bromo-p-methoxybenxoylpropio~te from the yellow ester 1st isomeride m. p. 51"; 2nd isomeride m. p. 66". Bromination of p-methoxybenzoylacrylic acid yielded two isomeric forms of up-dibrorm-p-methoxybenxoyljpropionic acid m.p. 140" and 164" respectively. The polymeric acid C,oH,oO from p-methoxybenzoylacrylic acid has m. p. 180". a-Hydroxy-p-methoxybenxo ylacrylic acid (from hydrolysis of un- saturat,ed acid) has m. p. 122". The yellow methyl p-methoxybenzoylacrylate reacts with 1 mol. of semicarbazide and yields two isomeric semicarbames yellow needles m. p. 165" and colourless plates m. p. 178". The same yellow methyl ester on treatment with two molecular proportions of semicarbazide yields the semicarbaxone of methyl a-semicccrbaxido- p-methoxybenzoybpropionute m. p. 178" (decomp.). Methyl a-semi- mrbaxido-p-methoxybenxoylpropionate m. p. 161-162" is obtained as a by-product in preparation of the semicarbazone m. p. 178" and also by treating the semicarbazone of methyl a-semicarbazido- p-methoxybenzoylpropionate with concentrated hydrochloric acid.Treatment of the solid m. p. 161" with nitrous acid gave methy,? nitroso- a-semicarbazido-p-methoxybenxo ylpropimte m. p. 132 " (decomp.). Ln a similar manner the yellow ethyl ester of p-methoxybenzoyl- acrylic acid yields two isomeric semimrbames colourless crystals m. p. 165" and yellow crystals m. p. 124". The semicurbaxone of ethyl u-semicarbaxido-p-mt~xybenxoyl- propionate m. p. 182" (decomp.) and ethyl u-semicarbaxido-p- methoxybenmylpropionate m. p. 173" and the nitroso-compound m. p. 12A125" (decomp.) are described. Methoxybenzoylacrylic acid itself yields a semicurbaxone pale yellow solid m. p. 216" (decomp.). Constitution of Phthalonic Acid.CORNILLOT (Compt. rend. 1924 178 4 9 0 4 9 2 ) .-Phthalonic anhydride is rapidly converted by acetic. acid a t 100" into the acetyl-$-phthalonic acid described by Kuroda and Perkin (T. 1923 123 2094) who concluded that phthalonic acid has the phthalide structure. With this conclusion the author disagrees since the anhydride possesses the normal structure. Acetyl chloride or acetic anhydride converts phthalonic acid into a mixture of phthalonic anhydride and acetyl-+-phthalonic acid the proportions of the two products varying with the con- ditions of experiment. In boiling ethyl acetate solution (with acetic anhydride) or in suspension in acetyl chloride almost pure phthalonic anhydride results whilst the latter on boiling for an hour with acetic anhydride is entirely converted into the acetyl- +-acid.In any case (as in the work of Kuroda and Perkin) the secondary reaction is responsible for the production of the acetyl-+- acid. Condensations of Aldehydes with Resorcinol and some other Aromatic Hydroxy Compounds. R. N. SEN and N. N. SINHA ( J . A m r . Chem. XOC. 1923 45 2984-2996).-The benzein F. A. M. [See also Tcherniac T. 1916 109 1236.1 E. E. T.i. 289 OEQANICJ CHEXISTRY. type of compound has been prepared by condensing aldehydes with resorcinol and similar substances in the presence of con- centrated sulphuric acid a t 100-130°. The intensity of the fluores- cence of alkali solutions of these compounds varies with the number and the position of hydroxyl and carboxyl groups the influence of hydroxyl groups being greater when they occur in the benzene nuclei associated with the pyrone ring than when they occur in the phenyl residue whereas the opposite is the case with the carboxyl group.Resorcinol-1'-hydrmybenxein Cl9HI2O4 has m. p. above 260". Its tetrahomo-derivative forms red crystals. Resor- cinol-2'-methoxy-3'-hydroxybenzein C2&1405 is obtained from vanillin and resorcinol as a greyish-black powder rn. p. above 295". Gallic acid benzein C,,H,,O from gallic acid and benzaldehyde is a green microcrystalline powder. Resorcinolethein C,,H,,O from resorcinol and paraldehyde is a black microcrystalline powder m. p. above 290". Resorcinolphenetoleaxobenzeinsulphonic acid C,,H,,O,N,S from phenetoleazobenzaldehydesulphonic acid and resorcinol does not fluoresce.Resorcinol-p-hydroxybenzein-m- carboxylic acid CzoHla06 is a yellow powder m. p. above 295;". p-Resorcylic acid benzezn C,lH,,O from p-resorcylic acid and benzaldehyde is a yellow powder. Resorcinol-3'-methom~benzein C20H,404 from anisaldehyde and resorcinol gives a red tetrabromo- derivative decomposing a t 200". Reduction Products of some Acids Derived from Camphor. L. PALFRAY (Ann. Chim. 1923 20 [ix] 297-385).-A republic- ation of work previously described (A. 1921 i 418; 1922 i 548; also Haller and Palfray A. 1923 i 577) with additional matter. Reduction of the dimethyl ester of carboxycamphoracetic acid by means of sodium and absolute alcohol resulted in the formation of a mixture of borneol camphoracetic glycol C,H,,< CH*CH,*CH20H I an oil b. p. 195-198"/14 mm.a +l9" 57' yielding a phenyl- urethane colourless spangles and the mononzethyl ester F. G. P. CH*OH (- (C 0 ,Me) *CH,*C 0,H C81-I,4<&H.*H 9 a +49" 22'. Similar reduction of the diethyl ester also yields borneol the same glycol and a substance of the formula C,,H,O probably a lactone crystals m. p. 50-5lo,.a~ +62" 22'. A dis- cussion on the molecular refraction of the substances dealt with is appended. The following additional data are given. Ethyl hydrogen camphorate has m. p. 56-56.5"; the diethyl ester of carboxycamphoracetic acid has b. p. 202-204"/18 mm. Gg 1.09635 nz 1.47354 nz 1.47595 n; 1.48708; the dimethyl ester $fl 1.14245 n:'l 1-48086 ng'l 1.48347 n?.l 1.49440. To obtain any considerable yield of reduction products from cyanocamphoric and homocamphoric esters the reaction should be carried out at a temperature approaching 150".This is also the case with the esters of carboxycamphoracetic acid but only the ketonic group and the a-carbalkoxy-group are completely hydrogenated the h a 1 product being the glycol. H. J. E.i. 290 ABSTRACTS OF CHEmCAL PBPEBS. Derivatives of Camphoronic Acid. W. F. GOEBEL and W. A. NOYES ( J . Amer. Chem. Soc. l923,45,3064-3070).-The investig- ation was undertaken in order to establish the structure of the various derivatives of camphoronic acid. The attempts to prepare either a dibasic amide or an anhydro-amide which would react with sodium hypobromite failed. Isomeric imide-amides were however obtained from the corresponding imides and have been found to react with sodium hypobromite.p-Anhydrocamphoronyl chloride reacts with ammonia fo give transparent crystals of an imide C,H,,O,N m. p. 234.5-235" [ayi -5-40". A bromo-imide m. p. 220° is similarly obtained from a-anhydrobromocamphoronyl chloride. a-Camphoronamide-imide is obtained by the action of phosphorus pentachloride and then anhydrous ammonia on a-camphoronimide. It forms transparent crystals m. p. 235- 236" +11*87". The acid chloride of p-camphoronimide is obtained from the imide and phosphorus pentachloride. It is a crystalline solid m. p. 1756-176-5" [a]z,6 -18.32". p-Camphoron- imide-amide obtained by the action of ammonia on the p-imide chloride is a crystalline solid m. p. 205" -13.73". It reacts readily with sodium hypobromite. Potassium hydroxide fails to open the ring of either a- or p-camphoronimide partial hydrolysis to the parent acid being the only reaction involved.F. G. P. Oxidations with Ozone. I. The Preparation of Vanillin. E. B R ~ E R R. PATRY and E. DE LUSERNA (Helw. Chim. Acta 1924 7 62-74).-The possibility of utilising ozone for technical oxidations is discussed and experiments on the oxidation of iso- eugenol to vanillin are recorded. The oxidations were carried out by conducting a current of ozonised air a t varying concentrations of ozone (5% to 19%) into a solution of isoeugenol in carbon tetra- chloride a t temperatures from -15" to +50". The highest oxid- ation efficiency obtained in terms of ozone consumed was 20% a t the lowest temperature. At higher temperatures t.he efficiency falls off rapidly and the quantity of resinous by-product formed increases.[Cf. B. Apr.] E. H. R. Phthalaldehyde. 11. L. SEEKLES (Rec. traw. chim. 1924 43 93-102 ; cf. A. 1923 i 931).-Attempts were made unsuccessfully to ut4ilise the blue colour obtained by the reaction between phthal- aldehyde ammonia and acetic acid for the determination of small quantities of ammonia. The reaction product appears to contain at least three different coloured substances and the tint produced varies with the proportions of the reactants. The dioxime of phthalaldehyde has not hitherto been prepared. It is now obtained in small yield by the action o f hydroxylamine on phthalaldehyde in cold alcohol. Phthahldioxime forms light yellow crystals containing 2H20 m. p. 105" (indistinct) decomposing a t 150".The di-p-nitrophenylhydruxone of phthalaldehyde forms an orange precipitate (a-form) m. p. 244-245" (decomp.). By treat- ment with gaseous hydrogen chloride in chloroform solution it is converted into a red isomeride (p-form) decomposing violently a ti. 291 OWANIO CHETt¶IS!t'RY. 208'. Attempts t o repare isoquinoline by the interaction of methylamine and phthp,laldehyde were not successful. R. M. HIXON ( J . Amer. Chem. Soc. 1923 45 2333-2341).-1t is pointed out t'hat quinones may be expected to add ammonia (or an amine) (a) at the carbonyl group by analogy with the formation of aldehyde-ammonia or (b) at the ethylenic bond as do certain unsaturated aliphatic acids and that actually both reactions probably occur with the quinones. The intermediate additive compound derived from an amine and a quinone is structurally analogous to an ammonium salt ; this undergoes rearrangement the amine becoming attached in either of the two ways mentioned.Such an intermediate complex if formed from a tertiary amine would be incapable of rearrangement. Actually Jackson and Clarke (A. 1905 i 908) have reported that dimethylaniline yields coloured additive products with quinones. This is now qualitatively confirmed since an additive compound appears t o be formed when thymoquinone is dissolved in dimethylaniline (2 mols.). This is apparently stable when air and solventls are excluded but otherwise a purple tar is produced which on hydrolysis by means of 50 % sulphuric acid gives both mono- and di-hydroxythymo- quinone. This indicates to some extent that tertiary amines may react with quinones just as do primary and secondary amines.Directions are given for the purification of nitrosothymol and of nitrosocarvacrol. Their benzoates are in effect hydrolysed when treated in dry heptane solution with dry methylamine gas. The reaction which cannot be an ordinary hydrolysis is explained by assuming the formation of an intermediate compound structurally similar to aldehyde-ammonia which undergoes rearrangement to give the nitroso-derivative and N-methylbenzamide. These benzo- ates form unstable additive compounds with concentrated nitric acid having the formulz and m. p. respectively C17H17?2N,3HN03 53" and C,7H,70~,2HN0 65". Thymoquinone &omme may be prepared more conveniently by the action of hydroxylamine hydro- chloride on nitrosocarvacrol yield 70 yo than by using nitroso- thymol yield &loyo.It forms a dibenxoate colourless needles The Addition of Mercuric Salts to ap-Unsaturated Ketones. E. B. MIDDLETON ( J . Amer. Chem. Soc. 1923 45 2763-2769).- @-Unsaturated ketones such as benzylideneacetophenone (phenyl styryl ketone) react with but one molecule of mercuric acetate whilst dibenzylideneacetone (distyryl ketone) containing two double linkages reacts with 2 mols. of mercuric acetate. The products obtained in alcoholic solution show that the groups *OR and HgOCOCH have been added. The reactions of the additive compounds show that the mercury compounds of phenyl styryl ketone have the struct'ure Ph*CH( OR)*CH(HgX)*COPh whilst the derivatives of distyryl ketone are PhCH ( OR) *CH (HgX) *CO *CH (HgX)CH( OR)*Ph .Substitution in the nuclei does not affect the reaction of the E. H. R. Amino- and Oximo-derivatives of Thymoquinone. m. p. 199-200" (becomes discoloured a t 170"). w. s. N.i. 292 ABSTRACTS OF cHE116ICAL PAPERS. unsaturated ketones with mercuric acetate. Ketones of the cycb- propane series do not yield additive compounds. A comparison is made between the mercuric acetate additive compounds and the compounds obtained by Vorlander and Eichwald (A. 1923 i 684) using the mercury halides. a-Acetoxymercuri- p-methoxy-p-phenylpro- piophenone C,,H,,O,Hg obtained by the action of mercuric acetate on phenyl styryl ketone in methyl alcohol has m. p. 115". The corre- sponding ethoxy-derivative has m.p. 134". a-Bromomercuri-p- methoxy- p-jp~~enylpr021iophenone,C,~H,~O,BrHg m. p. 141" is prepared by the action of potassium bromide on the acetoxy-compound. The ethoxy-compound has m. p. 138". ~-Bromo-p-methoxy-p-phenyZ- propiophenone m. p. 76" is obtained by adding bromine to the acetoxymercuric-methoxy-derivative of phenyl stgryl ketone. The ethxg-compound has m. p. 60-61" and the corresponding iodo- compounds 96" and 75-76". Sodium methoxide converts a-bromo- p-methoxy- p-phenylpropiophenone into an oil b. p. 200-205"/ 6 mm. addition of mercuric acetate to the oil yielding diacetoxy- mercuridibenmylmethune C,,H,,O,Hg m. p. 237". The bromo- methoxyphenylpropiophenone reacts mth ethyl magnesium bromide to yield p-phenylvalerophenone C,,H,,O m. p. 63".Distyryl ketone reacts with mercuric acetate to form the dimercurated compound containing two methoxy- and two mercuri-groups m. p. 183". a-Acetoxymercuri-~-methoxy-~-jphenyl-jp-chlorop~opiophenone m. p. 124" is obtained from mercuric acetate and p-chlorophenyl styryl ketone in methyl alcohol solution. Preparation of 2 4 2' 4'-Tetrahydroxybenzophenone. J. B. SHOESMITH and J. HALDANE ( J . Chem. Xoc. 1924,125 113- 115) .-2 4-Diacetoxybenzonitrile is condensed with resorcinol by means of zinc chloride and dry hydrogen chloride the resulting imine hydrochloride ( Ac 0 ),C 6H3-C (XH HC1) *C 6H3 ( OH) 2 being readily hydrolysed to the required benzophenone in acid solution. A certain amount of resacetophenone and of p-resorcylamide may be recovered from the reaction mixture.2 4-Diacetoxy-2' 4'- dihydroxybenxophenoneimine hydrochloride gives crystals decompos- ingat 195". F. G. P. The Benzil Rearrangement. V. Cannizzaro's Reaction. A. LACHMAN ( J . Amer. Chem. Soc. 1923 45 2356-2363).-1n the Cannizzaro reaction benzyl benzoate is formed from benzaldehyde even in the presence of water provided an excess of alkali is avoided and any considerable rise in temperature is prevented. It is suggested that dihydroxydibenzyl ether [CE€Ph( OH)],? is formed as an intermediate stage and by a glycol rearrangement is converted into the benzoate. In the course of the investigation it was shown that by heating benzyl alcohol in a sealed tube a t 210-215" dibenzyl ether is obtained and that bromine reacts with dibenzyl ether to produce benzyl bromide and benzoyl bromide. F.G. P. The Isomerism of the Oximes. XV. The Alleged Fourth Benzildioxime. 0. L. BRADY and F. P. DUNN ( J . Chem. Soc. 1924 125 291-296).-The work of Atack and Whinyates (T. F. G. P.ORQANIC CHEMIS!l%Y. i. 293 1921 119 1184) has been repeated but the authors have been unable to confirm the results no evidence of the existence of the fourth dioxime being obtained. The purely structural theory of the isomerism of the oximes suggested by Atack is criticised and the principal objections to it are stated particularly the existence of two 0-ethers. F. G. P. Dioximes. XVI. G. PONZIO and V. BERNARDI (Gazzetta 1923 53 813-817; cf. this vol. i 54 56).-The formation of the two benzildioximes from the two forms of phenylglyoxime by the action of phenyldiazonium chloride (A. 1923 i 856) proves to be a reaction of general application constituting a new method of synthesising symmetrical and unsymmetrical disubstituted glyoximes. Not only a glyoxime of the type NOH:CAr*CH:NOH in which Ar is an aryl radical but also methylglyoxime reacts with many aryldiazonium chlorides with liberation of nitrogen and formation of the glyoximes NOH:CR*CR,:NOH.This reaction which gives a yield of about 45% is not so simple as is represented by the scheme since there is entire lack of experimental evidence of the formation of an intermediate unstable compound with the NzAr group in direct union with a carbon atom. Participation in the reaction of the oximino-groups may however be excluded since these pass with unaltered properties into the final products.Of special interest is the behaviour of methylglyoxime which yields only the p-forms of glyoximes NOH:CMe*CAr:NOH ; the conclusion that it is a p-modification is thus confirmed. Thus p-phenyl- methylglyoxime (A. 1922 i 17) may be obtained by the action of phenyldiazonium chloride on methylglyoxime in 10% sodium hydroxide solution cooled with ice. 8-0 - Tol ylmeth yZglyoxime NOH:CMe*C(NOH) *C 6H4Me prepared from methylglyoxime and o-tolyldiazonium chloride crystallises in white needles m. p. 236" (slight decomp. and sublimationy; its nickel salt (C1,,H1102N2)2Ni forms red prisms m. p. 222" and its dibenxoyl derivative CzaHz0O4N2,. prisms m. p. 144". p-p-Tolylmethylglyoxime crystalllses rn white laminae m. p. 234-235"; Borsche (A.1907 i 326) gave m. p. above 230" (decomp.). Its nickel salt (CloHl102N2)2Ni forms blood-red prisms m. p. 250-251" (decomp.) mth prevlous browning and its dibenxoyl compound white needles m. p. 148.5-149.5". p -p - Anis ylmet h ylgl y oxime prepared from p -me t ho x yp hen y 1 - diazonium chloride and methylglyoxime forms white prisms m. p. 213" (slight decomp.). Boeris (A. 1894 i 72) gave m. p. 206" and Tschugaev (A. 1908 i 554) m. p. 207"; cf. also Wieland (A. 1903 i 837). Its nickel salt C,H,,O,hT,Ni crystallkes in scarlet needles m. p. 224" (cf. Tschugaev Zoc. cit.) and its dibenxoyt derivative in needles m. p. 148.5". 8 -Phenyl- p - tolylglyoxime N0H:CPh.C (NOH)*C cH4Me prepared by the interaction either of p-tolyldiazonium chloride and p-phenyl- I" NOH:CR*CH:NOH+Ar*N2*OH -qO + NOH:CR*C(NOH)*N&r-" + NOH:CR*CAr:NOH,i. 294 ABSTRACTS OF CHEMICAL PAPERS.glyoxime or of phenyldiazonium chloride and p-p-tolylglyoxime crystallises in lustrous laminae m. p. 223-224" (partial sublimation). Its nickel salt (C1sH1302N2)2Ni forms microscopic brick-red needles m. p. 295-297" (decomp.) and its dibenxoyl derivative laminae m. p. 184" (slight decomp.). p -Phen yl- p - unis ylg l yoxime C 5H 1403N2 prepared similarly cr ys - tallises in white laminae m. p. 223" (slight decomp.). Its nickel salt forms a microcrystalline blood-red powder m. p. 264" (decomp.) and its dibenxoyl derivative laminae m. p. 162". T U P Dioximes. XVIII. L. AVOGADRO (Guxxettcc 1923 53 824- 828).-The dehydrogenating action of nitrogen tetroxide on the a-modifications of arylglyoximes proceeds similarly to that of the same reagent on a-phenylglyoxime (A.1923 i 1019 1137). Thus a-p-tolylglyoxime yields a compound which must be regarded as the oxide of p-toluoyl cyanide oxime since (1) with acetic anhydride it yields a monoacetyl derivative (2) it is reduced by means of zinc and acetic acid giving p-toluoyl cyanide oxime (3) when heated with xylene it undergoes isomerisation to 5-hydroxy-3-p-tolyl- 1 2 4-oxadiazoleY (4) it unites with a molecule of ammonia to form amino-a-p-tolylglyoxime and (5) it reacts with magnesium methyl iodide to form p-tolylmethylglyoxime. The oxide of p-toluoyl cyanide oxime NOH:C(C,H4Me)*C< I or NOH:C(C6H,~le)*CiN:0 crystabes in silky white needles m. p. 108" with previous softening and is highly stable towards acids but is readily decomposed by bases or even by boiling water giving p-toluonitrile and substances having the odour of carbylamine.Its acetyl derivative forms white prisms m. p. 157-158". Amino-a-p-tolylglyoxime C6H4Me*C( :NOH)*C( :NOH)*NH crystal- lises in white laminae m. p. 173-174" and when treated in aqueous alcoholic solution with nickel acetate gives a greyish-green pre- cipitate soluble in dilute acetic acid. When boiled with dilute acetic acid it slowly undergoes isomerisation into amino- @-p-tolyl- glyoxime which has not been investigated further. Ketens. XLVII. The Constitution of Dimeric Ketens a Contribution to the Valency Problem of Organic Chemistry. H. STAUDINGER (Helv. Chim. Acta 1924 7 3-8).-A reply to Schroeter (A.1920 i 852) who criticises the author's conception of the dimeric ketens as cyclobutan-1 3-diones or 1 3-cyclo- butenolones and considers them to be molecular compounds. The conception of dimeric aldoketens as 1 3-cycbbutenolone derivatives is supported by their physical properties and absorption spectra. Schroeter considers the more stable dimeric ketens to be true cyclobutandiones the less stable to be molecular compounds but it is pointed out that cycbbutane derivatives differ much in their stability to heat ; substitution including the keto-group weakens the ring and heavily substituted compounds such as ethyl diethyl- cycbbutandionedicarboxylate are especially labile. The author's 0 N T. H. P.ORGAh4C CHEMISTRY. i. 295 view is also supported by preparation of cyclobutane derivatives by t'he combination of diphenylketen with ethylenic compounds (A.1920 i 556 and following abstract). E. H. R. Ketens. XLVIII. cgctoButane Derivatives from Diphenyl- keten and Ethylenic Compounds. H. STAUDIKGER and A. RHEINER (Helv. Chirn. A c t 1924 7 8-lS).-The compound formed by diphenylketen with styrene is now found to be 1 1 4- I triphenylcyclobutan-2-mae ?Hph'(?ph2 instead of the 1 1 3- CH,-CO ' compound as previously statcd (Staudinger and Suter A. 1920 i 556). It follows that the acid obtained by treating it with sodium hydroxide in alcoholic solution is ccap-tri~henyl-n-butyric acid instead of the aq-compound. Evidence for the constitution of the tri- phenylcyclobutanone was obtained through the action of methyl magnesium iodide by means of which the tertiary alcohol l-methyl- 2 2 3-triphenylcyclobutan-1-ol was obtained colourless crystals m.p. 132"; the yield was S70/ of the theoretical. When rapidly distilled a t the ordinary pressure the tertiary alcohol decomposes to the extent of 80% into as-diphenylacetone and styrene and 20% into acetone and triphenylethylene. These products establish its constitution. When distillation is carried out in a high vacuum decomposition does not take place but an ismeride is formed m. p. 137"; in all probability the ring has opened with formation of a triphenyl-n-propyl methyl ketone. By the action of phenyl magnesium bromide on the triphenylcyclobutanone there is obtained not the expected tertiary alcohol but benxylidenediphenyletme CHPh2*CO*CH:CHPh [benxhplryl styryl ketone] m.p. 112" b. p. 250°/18 mm. The same compound was synthesised from diphenyl- acetone and benzaldehyde (cf. Rack A. 1923 i 6S1). As previously described (loc. cit.) diphenylketen combines with cvclonentadiene to form a comnound which can be reduced to a V L * saturated compound for which the formula CH2< CH,-FH-I-FO CH,*CH-CPh,' is now proposed. When this is submitted to prolckged boilin'' with sodium hydroxide hydrolysis takes place a t the dotted line with formation of diphenylcyclopentylacetic acid m. p. 85" ; anilide m. p. 139". The compound of diphenylketen with cyclopentadiene reacts with methyl magnesium iodide annexed formula designated 6 6-di- phenyl-7-methyl-0 2 3-bicyclo-3 4( 2) - hepten-7-oZY m.p. 130". When boiled with alcoholic potassium hydroxide the heptane ring opens probably at the dotted line forming a ketone ; its semicarbazone has m. p. 115". By treatment with hydrogen and platinum black the above heptenol is reduced to the corresponding heptunol m. p. 99". By the action of phenyl magnesium iodide on the diphenylketen-cyclopentadiene compound a substance was obtained m. p. 214" having the same empirical composition as the original compound. It may be a 3 cH2.yH-i-@e(oH) 2 1 to give the tertiary alcohol having the cH'cH-cH-7ph2 4 5 l* 2i. 296 ABSTRACTS OF CHEMICAL PAPERS. dimeride. Diphenylketen and dicyclopentadiene form a compound m. p. 95". Ketens. XLIX. cycZoButane Derivatives from Dimethyl- keten and Ethylenic Compounds.H. STAUDINGER and P. J. MEYER (Helv. Chim. Acta 1924 7 19-22) .-Dimethylketen com- bines less readily with ethylenic compounds than diphenylketen but the compounds when formed are more stabIe than those from diphenylketen. The four-membered ring is relatively stable and the ketonic group has normal properties. Dimethylketen does not combine with styrene but with vinyl ethyl ether it forms 3(or 4)- ethoq-1 l-dimethylcyclobutan-2-one a colourless oil b. p. 166- 169'; the phnyZhydraxone forms white needles m. p. 70° and quickly decomposes in the air. With cyclopentadiene dimethylketen CH,*CH*FO forms a compound probably having the formula CH< CR-bH-CMei but the position of the double bond is uncertain; it is an oil with a camphor-like odour b. p. 175-178"; when superheated it decomposes into dimethylketen and cyclopentadiene.Its serni- curbaxone has m. p. 215". When reduced with hydrogen and platinum black it gives the saturated compound 6 6-dimethyl- 0 2 3-bicyc1oheptan-7-oney also having the odour of camphor b. p. 176-178"; it gives dimethylketen when superheated. Its semicarbazone has m. p. 224". Reaction Products from Aldehydes and 1 2-Diketones. 0. DIELS [with R. W. POINDEXTER K. ILBERG and W. GADKE] (Annulen 1923 434 1-2O).-When a mixture of benzaldehyde and diacetyl is treated a t - 10" with 25% methyl-alcoholic potassium hydroxide solution the diketone first polymerises to its dimeride 1 4-dihydroxy-1 4-dimethylcyclohexan-3 6-dione (Diels Blan- chard and v. d. Heyden A. 1914 i 1052); this then undergoes condensation with the aldehyde (3 mols.) 1 mol.of water being eliminated with the formation of the same compound (I) slender white needles m. p. 198" (decomp.) which is obtained if the dimeride itself is used. The structure of this substance is evident from the following reactions. Although probably owing t o steric hindrance the ketone groups are not reactive the hydroxyl yH*CHPh*oH group next to the phenyl radical can be CHPh*CH CO esterified. The benxmte forms thick transparent prisms m. p. 192"; the acetate four-sided tablets or prisms has m. p. 197-198". But if in preparing the CHPh-0 acetate the action of acetic anhydride is continued water is eliminated with the production of the unsaturated acetate -OC*CMe:C(C€€Ph*OAc)*CO- m. p. 237". This compound reacts immediately with potassium permanganate in glacial acetic acid solution or with bromine in cold chloro- form solution ; the subsequent spontaneous elimination of hydrogen bromide from the dibromide produced in the latter reaction gives two bronzo-derivatives C,,H,,O,Br (a) colourless needles or E.H. R. E. H. R. CNe* OH /\ oy I \/ (I.) 0 $!fileOIWANIC CHEMISTRY. i. 297 small elongated plates m. p. 193" ( b ) small transparent prisms m. p. 210". The action of phenylhydrazine in methyl-alcoholic glacial acetic acid solution at 65" on the unsaturated acetate gives the corresponding unsaturated alcohol m. p. 19&-203" +xEtOH long prisms which is readily converted into the compound (I) by boiling with hydrochloric acetic or propionic acid and gives the saturated methyl ether -OC*CMe(OMe)CH(CHPh*OH)*CO- (felted needles m.p. 240° after decomposing at 218") when boiled with methyl alcohol containing a trace of hydrochloric acid. Similarly the action of acids on the unsaturated acetate gives the saturated acetate which like the analogous benzoate is readily hydrolysed to the compound (I) by means of boiling alcoholic potassium hydroxide solution. By the action of a boiling mixture of glacial acetic acid and concentrated nitric acid on the compound (I) the secondary alcohol group is oxidised to a ketone group -COCMe( OH)*CH( C0Ph)-CO- with the formation of a compound m. p. 203-204" (decomp.) which crystallises from acetonitrile with lMeCN in rhombic crystals readily soluble in alkali. In 25 % methyl-alcoholic potassium hydroxide solution benzaldehyde is rapidly eliminated with the production of the compound (11) thin felted needles m.p. 151-153" after FH.cOPh sinteriiig at about 146". The latter is soluble in aqueous alkali but soon resinifies more rapidly on warming with the elimination of benzaldehyde. It is converted by the action of methyl alcohol containing hydrogen chloride into its methyl ether com- pact snow-white crystals m. p. 169-171" (slight decomp.) ; methyl hydrazinecarboxylate gives a salt felted needles M. p. 130-132" (decomp.) in the formation of which presumably the benzoyl group has taken part in its enolic form. The ketonic groups are not reactive. The compound (11) forms an additive compound with cold formic acid but the boiling anhydrous reagent causes the elimination of water to give 2-benzoyl-1 4-dihydroxy-5- benzylidene-1 4-dimethylcyclohexan-3 6-dione (111) a bright sul- phur-yellow compound m.p. 238-243" one of the ketone groups of which is. reactive; oc; C;HmCOph the oxime m. p. 221-222" semicarbamne long colourless needles decomp. about 235" CHPh:C CO m. p. 245-248" and the curbomethozyhydr- axone long colourless prisms decomp. 220" CJ&oH. m. p. 240-243" are described. The com- pound (111) is decomposed a t 250-270" into benzoyldiacetyl and benzylidenediacetyl. Compounds are also prepared by condensing diacetyl with furfurol and anisaldehyde. The first forms thick colourless needles m. p. 203" (decomp.) after sintering a t 180"; the second forms small needles and has m. p. 203-205" (decomp.). With concentrated sulphuric acid they give respectively a dark reddish-purple color- aTeOH (11.1 /\ oy OH*CHPh*CH CO \/ CNeoH (111.1 /\ \/i.298 ABSTRACTS OF CHEMICAL PAPERS. ation which changes to brown and an intense carmine coloration Trihydroxymethylanthraquinones. I. G. D. GRAVES and R. ADAMS ( J . Amer. Chem. Xoc. 1923 45 2439-2455).-3 6-Di- methoxyphthalic anhydride is prepared by the action of boiling 98% sulphuric acid on the dimethyl ether of 2 3-dicyanoquinol which is obtained by the action of potassium hydroxide solution and methyl sulphate on 2 3-dicyanoyuinol. The action of an excess of p-cresol on 3 6-dimethoxyphthalic anhydride in the presence of aluminium chloride at 70" gives 3 6-dimethoxy-2- (2'-h ydrox y-5'-meth yl benxo yl) benzoic white plates m.p. 218" (corr.) together with 12 15-dimethoxy-2 7-di- methyl;fuoran white crystals m. p. 290" (corr.). When the benzoic acid derivative is boiled with potassium hydroxide solution and methyl sulphate the product is 3 6-dimethoxy-2-(2'-methoxy- 5'-methyZbenzoyl)benzoic acid white needles or colourless truncated prisms m. p. 203-204" (corr.) which is identical with the product from 3 6-dimethoxyphthalic anhydride and p-cresyl methyl ether (cf. following abstract). When the substituted benzoic acid and concentrated sulphuric acid are heated a t 100" with a small quantity of boric acid the colour changes from green to bluish-red owing to the formation of l-hydroxy-5 8-dimethoxy-4-methylanthraquinne red needles m. p. 224" (corr.). The latter is demethylated by boil- ing with glacial acetic acid and hydrobromic acid (constant-boiling) giving 1 5 8-trihydroxy-4-methyEanthraqt~inone. When 3 6-di- methoxyphthalic anhydride is heated a t 75" with an excess of o-cresol in the presence of aluminium chloride the products are 3 6-dimethoxy-2- (2'-hydroxy-3'-methylbenzoyl)benzoic acid white needles m.p. 194" (corr.) and o-cresol-3 6-dimethoxyphthaZein white needles m. p. 258" (corr.). The latter is an indicator chang- ing from colourless to red a t pE 9.1. The methylation of the benzoic acid gives 3 6-dimethoxy-2-(2'-methxy-3'-methylbenzoyZ)- benxoic acid white needles m. p. 178" (corn.) which is not identical with the acid formed by condensing 3 6-dimethoxyphthalic anhydride with o-cresyl methyl ether ; in the latter reaction there is also formed the dimethyl ether of o-cresol-3 6-dimethoxyphthalein white needles m. p.202" (corr.) which is likewise produced by the methylation of the phthalein. The action of sulphuric acid containing boric acid on the benzoic acid derivative gives l-hydroxy- 5 8-dimethoxy-2-methylanthraquinone red needles m. p. 165" (corr.) which may be demethylated to give 1 5 8-trihydroxy- 2-methylanthruquinone red needles which sublime a t 250-260". When o-eresol-3 6-dimethoxyphthalein and 3 6-dimethoxy- phthalic anhydride are heated at 125" with concentrated sulphuric acid the product is 2 5 8-trihydroxy-l-methylanthrquinone dark red needles m. p. 270" (corr.) which is identical with the product obtained by condensing 3 6-dimethoxyphthalic anhydride with o-cresyl methyl ether dehydrating the benzoylbenzoic acid thus obtained and hydrolysing the resulting anthraquinone.The int'eraction of 3 6-dimethoxyphthalic anhydride and m-cresol a t which persists during twenty-four hours. w. s. N. mid,OBaANIC CHEMISTRY. i. 299 75" in the presence of aluminium chloride gives 3 6-dimethoxy- 2-(2'-hydroxy-4'-methylbelexoyl)benzoic mid white needles m. p. 233" (corr.) and 3-hydroxy-6-methoxy-2- (2'-hydroxy-4'-methylbenx- oy1)benxoic m i d white needles m. p. 189" (corr.). A third substance formed in the above reaction is m-cresol-3 6-dimethoxyphthalein white needles m. p. 271" (corr.) a solution of which changes from colourless to red a t pH 9.8. The conversion of 3 6-dimethoxy- 2-(2'-hydroxy-4'-methylbenzoyl)benzoic acid into l-hydroxy-5 8-di- methoxy-3-methylanthraquinone red needles m.p. 172" (corr.) can only be accomplished by the use of fuming sulphuric acid at 100" in the presence of boric acid. The demethylation of this dimethoxyanthraquinone by means of glacial acetic acid and hydrobromic acid gives 1 5 8-trihydroxy-3-meth ylanthrquinone red needles m. p. 227" (corr.). The condensation of 3 5-dimethoxyphthalic anhydride with m-cresol by heating with aluminium chloride a t 70" gives 3 5-di- methoxy-2 - (2'- h ydrox y-4'-meth ylbenzoyl) benzoic mid white needles m. p. 233" (corr.) and m-cresol-3 5-dimethoxgphthalein which cannot be purified. The former gives on methylation 3 5-di- methoxy-2 - (2 ' -methoxy-4'-methylbenmyl) benzoic acid white needles m.p. 219" (corr.) whereas the condensation of 3 5-dimethoxy- phthalic anhydride with m-cresyl methyl ether gives 3 5-dimethoxy- 2-(4'-methoxy-2'-methylbenxoyl)benxoic acid m. p. 233" (corr.). The action of fuming sulphuric acid and boric acid at 100" on 3 5-di- methoxy-2-(2'-hydroxy-4'-methylbenzoyl)benzoic acid leads to sul- phonation a poor yield of the expected anthraquinone yellow needles m. p. about 185" being obtained. This substance gives a red solution in alkali which reacts with methyl sulphate a t 100" to give a trimethoxy compound yellow needles m. p. about 175- 180" which gives a red solution in concentrated sulphuric acid and should be the trimethyl ether of emodin if the latter has the structure suggested (Arch. Pharrn. 1912 250 301). The action of concentrated nitric acid in glacial acetic acid solution a t 50-60" on 3 5-dimethoxy-2-(2'-hydroxy-4'-methylbenzoyl)benzoic acid gives the 5'-nitro-derivative pale yellow needles m.p. 241" (corr.) but the product of the condensation of 3 5-dimethoxyphthalic anhydride and o-nitro-m-cresol is 3 5-dimethoxy-2-(4'-hydroxy- 5'-nitro-2'-methylbenzoyl)beiazoic acid white needles m. p. 181" (corr .) . 3 5-Dimethoxyphthalic acid is obtained by the action of cold alkaline potassium permanganate solution on 3 5-dimethoxy- phthalide which is prepared by heating its carboxylic acid. The latter is made by the action of warm sodium hydroxide solution on 3 5-dimethoxytrichloromethylphthalide which is formed when a mixture of chloral hydrate and ethyl 3 5-dimethoxybenzoate is treated with somewhat concentrated sulphuric acid.Trihydroxymethylanthraquinones. 11. J. H. GARDNER and R. ADAMS ( J . Amer. Chem. Xoc. 1923 45 2455-2462; cf. pre- ceding abstract).-The action of an excess of p-cresyl methyl ether on 3 6-dimethoxyphthalic anhydride a t 70" in the presence of W. S. N.i. 300 ABSTRACTS OF CHEMICAL PAPERS. aluminium chloride gives 3 6-dimethoxy-2-(2'-methoxy-5'-methyl- benzoy1)benzoic acid together with the dimethyZ ether of p-cresol- 3 6-dimethoxyphthaleinY colourless needles m. p. 187.5" (corr.). When the trimethoxybenzoic acid is rapidly heated with concen- trated sulphuric acid a t 150° and then cooled the products are 1-hydroxy-5 8-dimethoxy-4-methylanthraquinone and 1 5 8-tri- hydroxy-4-methylanthrapuinone slender red needles m.p. 276- 278" (corr.) (wjth sublimation); the latter is the sole product if the reaction mixture is kept at 145-155" for twenty minutes. When a mixture of 3 6-dimethoxyphthalic anhydride and an excess of o-cresyl methyl et'her is treated a t 65" with aluminium chloride there are formed 3 6-dimethoxy-2-(4'-methoxy-5'-methyZ- benzoyZ)benzoic acid a white powder m. p. 192" (corr.) and the dimethyl ether of o-cresol-3 6-dimethoxyphthalein colourless needles m. p. 202-203" (corr.). When the acid is rapidly heated with concentrated sulphuric acid a t 1 50" and immediately cooled the chief product is 5 8-dihydroxy-2-methoxy-1 -methylanthrapuinone dark red needles m. p. 249-2496" (corr.) which is accompanied by a small quantity of 2 5 8-trihydroxy-1-methylanthraquinone. The latter may more conveniently be prepared by heating a mixture of concentrated sulphuric acid 3 6-dimethoxyphthalic anhydride and 0-cresol-3 6-dimethoxyphthalein or its dimethyl ether a t 125".The structure of the original benzoic acid is proved by dis- tilling the trihydroxyanthraquinone with zinc dust the resulting methylanthracene being then oxidised to 1 -methylanthraquinone. 3-H ydrox y-6-methoxy-2- (4'-methoxy- 1 '-meth ylbenxo yZ) benzoic mid colourless prisms m. p. 194-195" (corr.) and not the dimethoxy- acid is obtained when 3 6-dimethoxyphthalic anhydride is treated with an excess of m-cresyl methyl ether and aluminium chloride at 75"; a phthalein is not produced. A pure derivative of anthra- quinone has not been obtained by the dehydration of this acid.w. s. N. Reduction Products of the Hydroxyanthraquinones. IV. J. W. E. HALLEREL~~A. G. PERKIN ( J . Chem. Xoc. 1924,125,231- 239).-Muction of 2-hydroxyanthraquinone with glucose and alka.li in the presence of sodium nitrate yields 2-hvdroxvbenzanthrone- co co Carbogylic acid a,nd two compounds which are now shown to be a 3 6'-dihvdroxvhelian- throne ~28H1i04 (I) and 3 6'-dhydroxy- m eaon aDhthadianthr- obtaingd by the o&da- tion of dihydroxydianthrone with potassium ferricyanide. (11) may be obtained by exposing the diacetyl derivative of (I) (in benzene or acetic acid solution) to sunlight for some weeks when an almost quantitative yield of the diacetoxynaphthadianthroneORGANIC CHEMISTBP. i. 301 results. Oxidation of the acetyl compound of (I) with chromic acid yields diacetoxy-1 1 '-dianthraquinonyl C,,Hl,06Ac2 small plates m.p. 277". The corresponding dihydroxy compound crystallises in prisms or leaflets m. p. >350". The dimethyl ether C3OH1806 crystallises in yellow prisms m. p. 343-344". With iodine in pyridine solution (I) yields 5-iodo-3 6'-diacetoxyhelianthrone CS2Hl7O6I yellow needles m. p. 282" and from this by hydrolysis the dihydroxy compound is obtained in the form of scarlet needles m. p. >350". 3 6'-Dihydroxynuphthadianthrolze C28H1204 is an orange-brown powder m. p. > 350". Its diacetyl derivative crystallises in yellow needles. Camphorylcarbamates and their Physiological Action. H. E. FIERZ-DAVID and W. MULLER (J. Chem. Soc. 1924 125 26-27).-Carnphorylcarbamates may be obtained in almost quantitative yield by the interaction of aminocamphor and,:alkyl chloroformates.Their pronounced poisonous action renders them unsuitable for practical application. isoAmyl and allyl camphoryl- curbarnates are viscous oily liquids b. p. 199"/11 mm. and 186"/ 10 mm. respectively with [.ID +34.2" +34-3". Methyl cumphoryl- curbamate m. p. 110" b. p. l69"/11 mm. [.ID +39.4". The ethyl iso- propyl and isobutyl derivatives have respectively m. p. 88" 73" 83" ; b. p. 178"/13 mm. 170"/10 mm. 184"jll mm. with [aID +35.1" 35*3" and 33.9". Piperitone. VII. The Constitution of Piperitone. J. READ H. G. SMITH and R. S. HUGHESDON ( J . Chern. Xoc. 1924 125 129-137).-Benzyl-dl-isomenthone obtained by hydrogen- ation of benzylidene-dl-piperitone is distinct from 2-benzyl-dl-iso- menthone obtained from dl-piperitone.It is concluded from this observation that piperitone is Al-p-menthen-3-one and benzylidene- piper it one is 6 - b enz ylidene -A1 -p -ment hen - 3 -one. 6 - Benz y 1-dl- iso- menthone reacts with benzaldehyde to yield 2-benzylidene-6-benzyl- dl-isomenthone and it also forms a semicarbazone whilst 2-benzyl- dl-isomenthone is incapable of so reacting. 2-Benxylidene-dl-jso- menthone hydrochloride forms colourless needles m. p. 108-1 10". 2-Benzyl-dl-isomenthol C,oH,SO*CH2Ph is a colourless viscid liquid b. p. 194-196"/20 mm. Its phenylurethane melts a.t 153- 154". 2-Benzylidene-dl-isomenthone CloHl,O:CHPh is a colourlesa liquid b. p. 193-195"/15 mm. The corresponding benzyl com- pound CloH170*CH2Ph is a colourless viscid liquid b.p. 195- 197"/20 mm. Go 0-9957 nz 1.5230 and [&ID 74.92. 6-Benzyl- dl-isomenthone CloH170*CH2Ph is a colourless viscid liquid b. p. 202-203"/14 mm. dy 0.9884 ng 1.5176 and [&ID 74.84. Its semicurbawne crystrtllises in colourless prisms m. p. 170-171". 6-Benzyl-dl-isomenthone distils at 213-220"/20 mm. and has &? 0.9905 ng 1.5239 and [&ID 75.42. Its semicurbawne has m. p. 170-171 O. 6-Benzyl-2-benzylidene-dl-isomenthone is a yellow mobile oil b. p. 21%-225"/15 mm. Reduction of this yields 2 6-dibenzyl-dl-isomenthol a pale yellow viscid liquid b. p. 282- 287"/23 mm. 6-Benzyl-dl-isomenthol CloH,,0*CH2Ph distils at 203-207"/16 mm. Its phenylurethune crystallises in needles and F. G. P. 3'. G. P.i.302 ABSTRACTS OF CHEMICAL PAPERS. has m. p. 138-139". A crystalline variety of the isomenfhol m. p. 66-68' has also been obtained. Higher Terpene Compounds. XVIII. The Constitution of Cadinene. L. RUZICKA and M. STOLL (Helv. Chirn. Acta 1924 7 84-94).-The carbon framework of cadinene has already been established by its dehydrogenation with sulphur to cadaline (Ruzicka and Meyer A. 1921 i 573). This result has now been confirmed by the catalytic dehydrogenation of cadinene with platinum black in a high vacuum at 300". It remains therefore to determine the positions of the two double bonds That one of these is pro- bably adjacent to the isopropyl group I Seidel A. 1922 i 562). It is now shown by a variety of arguments that cadinene is probably to be repre- sented by one or both of the two annexed formula. The products obtained by the action of ozone on cadinene followed by decomposition of the ozonide first formed render it extremely improbable that both double bonds are contained in the same ring.Although great difficulty was experienced in obtaining any identifiable products from this reaction an ethyl dicarboxyhte was obtained b. p. 190-200'/0~2 mm. having the composition C19H3005 or C,,H,,06 in which all ?fe the carbon atoms of cadinene are intact. The CH CH absence of conjugated double bonds in cadinene /\ /\ is shown by the failure of attempts a t reduction CH CH CH with sodium and amyl alcohol. The oxidation of 6~ 6~ &c1 cdinene with manganese dioxide and sulphuric ,> \/ Me acid as described by Ruzicka Schinz and Meyer (this vol.i 171) resulted in the formation of /'\ CH2 trimellitic acid and mellophanic acid ~ t s primary products and benzenepentacarboxylic acid as a C1 PrS secondary product. Cadineiie dihydrochloride can be defkitely stated to have the annexed formula. Higher Terpene Compounds. XIX. Cadinol from Galbanurn Oil. L. RUZICKA and M. STOLL (Hdv. Chim. Actu 1924 7 94-105).-A sesquiterpene alcohol cadinol C,,H,,O which by loss of a molecule of water gives cadinene was discovered by Semmler and Jonas in galbanum resin (A. 1915 i 573). Having elucidated the constitution of cadinene (preceding abstract) the authors now attempt to discover that of cadinol. The compound does not react a t a high temperature with phthalic anhydride and must therefore be a tertiary alcohol.To determine the position of the double bond and of the hydroxyl group the action of ozone was studied on cadinol itself and on the dihydrocadinene obtained by hydrogenating cadinol and then removing water. An attempt to reduce cadinol to a dihydrocadinene by heating with zinc dust was unsuccessful ; only cadinene was obtained by dehydration of the cadinol. Reduction of cadinol to dihydrodinol was effected I?. G. P. \/\/- has already been shown (Ruzicka and /\ (11.) I /\ (1.1 E. H. R.o w m o cErEMIsTRY. €. 303 with some difficulty by hydrogen and platinum black; it is a colour- less oil b. p. 150-153"/12 mm. [.ID +23" n'," 1.4948 di6 0.9579. By boiling with 90% formic acid dihydrocdinol is converted into dihydrmdinene (annexed formula) a colourless oil b.p. 129- 131"/12 mm. [.IZ +19.8" ng 1-4952 di3 0.8999. By )\,\ the action of ozone on this compound and decom- osition of the ozonide there mere obtained a keto- ' ' I- %dehyde and a keto-acid by the disruption of the \\/\' hydrocarbon at the double bond; neither however was obtained in a pure condition. By ozonisation of cadinol itself there was obtained a neutral product Ci5H2402 probably a ketone+kohoZ b. p. 180-134"/12 mm.; it contains one double bond. In addition a doubly unsaturated ketone C15H,2O probably formed from the previous compound by loss of water was obtained b. p. 140-145"/12 mm. Oxid- ation of this ketone with manganese dioxide and sulphuric acid gave only mellophanic acid. The higher-boiling fractions of the neutral products from the ozonisation of cadinol gave besides mellophanic acid trimellitic acid. These observations can be explained on the assumption that cadinol consists of a t least two forms o! and p with probably a small proportion of a third form y having the formulae /\ a-Cadinol.fi-Cadinol. ycadinol. Structural formulae are suggested for the products obtained by the ozonisation of cadinol based on the above formulae for cadinol. The cause of the formation of trimellitic acid by oxidation of higher fractions of these products is traced back to the small pro- portion of 7-cadinol probably present in the cadinol. G. G. HENDERSON and C. A. KERR ( J . Chem. Xoc. 1924 125 102-106).-p-Pinene on treat- ment with a very dilute aqueous solution of hypochlorous acid gives mainly a mixture of three isomeric crystalline dichbrohydm'm C,,H180,C1 of m.p. 135" 166" and 131" respectively. They are isomeric but not identical with the corresponding compounds prepared from a-pinene. The dichlorohydrin of m. p. 135" is con- verted info a rnonochbrohydrrin C,&€,,O,CL m. p. 128-129" by the action of dilute potassium hydroxide solution; this is an iso- meride of pinol glycol chlorohydrin. Further action of caustic alkali converts i t into an oily liquid which is apparently somewhat similar to pinol oxide in properties. The dichlorohydrin m. p. 166" also yields an oily liquid on treatment with caustic alkali but a solid monochlorohydrin could not be obtained in this ca-se. The third dichlorohydrin m. p. 131" is unaffected by caustic alkali in the cold.E. H. R. The Chlorohydrins of p-Pinene. F. G. P.i. 304 ABSTRACTS OF CHEMICAL PAPERS. Some Oxidation Products of p-Pinene. G. G. HENDERSON and D. CHISHOLM ( J . Chew. h c . 1924,125,107-113).-f3-Pinene unites with chromyl chloride to form a solid compound Cl,,H,,*2CrO,Cl2 which on decomposition by water yields as chief product the ketone CgH,40 previously prepared in a similar fashion from a-pinene. A trace of an aldehyde and a small quantity of the inactive form of trans-pinol glycol were also obtained. Oxidation of p-pinene in glacial acetic acid solution by hydrogen peroxide gives a mixture of volatile and non-volatile products the chief constituents of the volatile portion being borneol and a little fenchyl alcohol. The non-volatile portion could not be obtained sufficiently pure for identification.F. G. P. Diacetyl Compound of Desmotropo-Artemisin. P. BERTOLO (Gazzettu 1923 53 867-869).-It is now found that desmotropo- artemisin forms a diacetyl derivative so that in the transform- ation of artemisin into desmotropo-artemisin the fourth oxygen atom retains its alcoholic character (cf. A. 1923 i 1108). DiacetyEdesnwtropo-artemisin C1,H,,06 crystallises in hard lustrous white needles often grouped in radiating masses m. p. 149-150" [a] -50.75" (in alcohol). When hydrolysed by means of alcoholic potassium hydroxide it yields not artemisin but a moderately stable acid compound m. p. above 200° which is apparently identical with that obtained by hydrolysis of the acetyl derivative of artemisin (Zoc. cit.). T.H. P. The Formation of Free Radicals by Reduction with Vanadous Chloride. J. B. CONANT and A. W. SLOAN ( J . Amer. Chem. Soc. 1923 4-5 2466-2472).-The addition of vanadous chloride solution to a solution of triphenylpyrylium chloride or of its ferric chloride double salt in the absence of air causes the formation of a red precipitate which is thought to be a free radical. With chloroform or benzene it gives a red solution the colour of which disappears when it is shaken with air. The substance is reoxidised to the pyrylium salt when it is treated in aqueous suspension with a solution of a ferric salt; if this process is conducted in the absence of oxygen using freshly prepared material a maximum of 1 mol. of ferric salt is required but much less than this is needed if the compound which is evidently unstable has been kept for a few hours even in an atmosphere of carbon dioxide.The action of an excess of vanadous chloride on 9-phenylxanthyl- ium chloride in hydrochloric acid solution in the absence of air gives Gomberg and Cone's free radical phenylxanthyl which is identified by conversion into its peroxide. Triphenylcarbinol in concentrated hydrochloric acid or con- centrated sulphuric acid solution is reduced by means of vanadous chloride in an atmosphere of carbon dioxide to triphenylmethyl which is likewise identified by means of the peroxide. The action of vanadous chloride on a very dilute solution of xanthylium chloride in concentrated hydrochloric acid gives a pink solid which may be kept for a few hours in air a t the ordinaryORGANIC CHEMISTBY. i.305 temperature but gradually changes to the colourless dixanthyl. Attempts to dry it in an atmosphere of nitrogen without loss of colour have been unsuccessful. When the bright pink solid is washed with water drained and dissolved in chloroform in an atmosphere of nitrogen the solution is momentarily red; this colour disappears in the course of a few seconds. It is evident from the last experiment that polymerisation and not oxidation causes the change to dixanthyl. There is therefore strong evidence that the pink solid is the free radical xanthyl. The addition of vanadous chloride to an aqueous solution of benzylpyridinium chloride does not cause the formation of a pre- cipitate although the first portions of vanadous chloride undergo a change of colour; after making the solution alkaline however N N'-dibenzyltetrahydrodipyridyl may be extracted from the solution by means of ether.Chloro- and Amino-anthracoumarins. V. I. MINAEV (J. Russ. Phys. Chern. Soc. 1924 54 680484).-A short abstract of this work has already appeared (A. 1922 i 162); the preparation of the following additional compounds is now described 8-amino- anthracoumurin from m-aminocinnamic and m-hydroxybenzoic acid silvery crystals resembling p-naphthol softens a t 285" carboiiises a t about 300" ; 8-~hZoroa~zthracoumarin obtained in a similar way from m-chlorocinnamic acid forms brown crystals m. p. 241" ; gives on sublimation golden-yellow needles m. p. 329" ; 9-aminoanthracoumurin obtained by condensing p-aminocinnamic acid and nz-hydroxybenzoic acid and the corresponding chloro- compound were also prepared but their properties are not recorded.G. A. R. K. A Synthesis of Pyrylium Salts of Anthocyanidin Type. 111. A New Synthesis of Pelargonidin Chloride. D. D. PRATT and R. ROBINSON (J. Chem. SOC. 1924 125 18&199).- Phloroglucinol dimethyl ether on treatment with formanilide and phosphoryl chloride followed by subsequent decomposition yields 2-hydroxy-4 6-dimethoxybenzaldehyde and a smaller relative amount of 4-hydroxy-2 6-dimethoxybenzaldehyde. When hydro- gen chloride is passed into a cold solution of the 2-hydroxy-corn- pound in the presence of ~0-4-dimethoxyacetophenone a good yield of tetramethylpelargonidin chloride (annexed formula) is obtained. On demethylatiqn pelargonidin chloride is produced.4-Hydroxy- 2 6-dimethoxybenzaldehyde on Meo/\/O /-\OMe similar treatment gives a trimethyl- pelargonidin chloride. It is sug- I I lo&\-/ gested that the blue coloration characteristlic of pelargonidin and other anthocyanidins is dependent on the presence of at least four hydroxyl groups as well as on their positions. 3 7-Dihydroxy-2-phenylbenzopyrylium chloride crystal- lises in orange-brown prisms decomposkg a t 184". 3 7 4'-Tpi- hydroxy-2-phenylbenxopyrylium chloride C15H110,Cl,H,0 crystallises W. S. N. c1 +7 \/\/ OMei. 306 ABS!CFUCTS OF CHEMICAL PAPEm. in needles and decomposes at 262". 4-Hydroxy-2 6-dinzethoxy- benzaldehyde crystallises in colourless prisms m. p. 190". Tri- rnethylgalungidin ferrichloride (3 5 7-trimethoxy-2-phenylbenxo- pyrylium ferrichloride) forms yellowish-brown prisms m.p. 174'. The trimethylgalungidin chloride crystallises in orange-red needles. F. G. P. A Synthesis of Pyrylium Salts of Anthocyanidin Type. IV. Flavylium Salts related to Chrysin Apigenin and Luteolin. D. D. PRATT R. ROBINSON and P. N. WILLI~LMS ( J . Chem. Soc. 1924 125 199-207 ; cf. previous abstract).-A series of compounds related t o the flavones has been obtained and a system of nomenclature based on names suggesting their relation- ship to particular flavones and having C1 OH the termination -idin suggested. Thus ,-% (I) is luteolidin chloride. The dimethyl HO/\ '\-/-\OH -ether of chrysinidin chloride is obtained I I \-/ by the action of hydrogen chloride on \/\/ acetophenone in the presence of 2-hydr- OH oxy-4 6-dimethoxybenzaldehyde and has been isolated as a ferrichloride.With phloroglucinol p-anisyl P-hydroxyvinyl ketone reacts in the presence of hydrogen chloride to give acacetidin chloride (11). Trimethyl- apigenidin ferrichloride has been c1 prepared from 2-hydroxy-4 6-di- 6 methoxybenzaldehyde and acetyl- /' \-/-\OMe anisole. Dimethylchrysinidin f erri- chloride Ci,H1,O,CI4Fe crystallises in reddish-brown prisms m. p. 178". p- Anis yl (3-h ydroxyvin yl ketone (11.1 OMeeC6H4*CO*CH:CH*OH can be isolated as a copper salt green prisms m. p. 206-207". 7-Hydroxy- 4'-methoxy-2-phenylbenxopyrylium chloride Ci,H1,03Cl,ZH20 brown- ish-orange needles decomposes at 182-183" ; its picrate melts at 219-221". Acacetidin chloride forms brown prisms m.p. > 360" ; trimethykpigenidin ferrichloride orange needles ; 2-hydroxy- 4 6-dimethoxystyryl 3 4-dirnethoxyphenyl ketone yellow prisms m. p. 178-179" ; tetramethyl-luteolidin ferrichloride salmon-red needles m. p. 206-207". (1.1 Ho() I \-/ \/\/ OH F. G . P. Some Derivatives of Benzopyrylium. R. ROBINSON and (in part) H. G. CRABTREE C. R. DAS W. LAWSON R. W. LUNT B. H. ROBERTS andP. N. WILLLBMS ( J . Chem. Xoc. 1924,125,207- 214).-The oxonium salts described have been prepared by the method of synthesis which depends on the condensation of a derivative of salicylaldehyde and a substance containing the group -CO*CH2- under the influence of hydrogen chloride. 4-Methoxy- phenyl2-hydroxy-5-rnethylstyryl ketone forms yellow plates m. p.151-152" ; 4'-rnethoxy-2-phenyl-6-methyl- 0H-C ,H,&fe*CH :CH*CO *c6H4*oMe,ORGCBNIC CHEMISTRY. i. 307 benmpyrylium ferrichloride Cl,Hl,O,C1,Fe obtained from t,he ketone crystallises in crimson needles m. p. 165-166". PhenyE 2-hydroxy-3-methoxystyryl ketone MeO*C,H,(OH)*CH:CH*COPh crystallises in yellow prisms m. p. 112". 8-Methoxy-2-phenylbenzo- pyrylium ferrichloride from the ketone crystallises in reddish-brown needles m. p. 157". 8 4'-Dimethoxy-2-phenylbenzopyrylium ferri- chloride obtained from o-vanillin and p-methoxyacetophenone crystallises in brown needles m. p. 180". 2 3 4-IPrimethoxyphenyZ 2-hydroxy-3-methoxystyryl ketone obtained from o-vanillin and gall- acetophenone trimethyl ether forms bright yellow needles m. p. 145". 2'-Hydroxy-8 3' 4'-trimethoxy-2-phenylbenzopyrylium ferrichloride forms dark red needles m.p. 194"; the picrate red prisms m. p. 164" ; the periodide purple needles m. p. 133" ; the merczcrichloride yellow needles m. p. 143". 2-Phenyl-3 4'-dimethylbenxopyrylium ferrichloride from salicylaldehyde and p-tolyl ethyl ketone forms yellow needles m. p. 165-166". 8-Methoxy-2-phenyl-3 4'-di- methylbenzopyrylium ferrichloride from o-vanillin and p-tolyl ethyl ketone forms orange-brown needles m. p. 172" ; 8-methoxy-2-a- naphthylbenxopyrylium ferrichloride from 0-vanillin and a-naphthyl methyl ketone brownish-red needles m. p. 230" ; 2'-hydroxy- 3'-methoxy-2-benzylidene-l-hydrindone from o-vanillin and or-hydr- indone yellow needles m. p. 198" ; 8-methoxy-2 S-indeno-( 1 2)- benzopyrylium ferrichloride red needles m.p. 187-188". F. G. P. Some Benzopyrylium Salts. L. R. RIDGWAY and R. ROBINSON (J. Chem. Soc. 1924 125 214-222).-Attempts to prepare a benzopyrylium salt in which a halogen atom occurs in position 3 have not so far succeeded. The attachment of nitrogen to the pyryliurn nucleus has been successful and 3-benzoylamino- 8-ethoxy-2-phenylbenzopyrylium ferrichloride (I) has been prepared from 2-hydroxy-3-ethoxy- EtO +- benzaldehyde and w-benzoylaminoacetophen- /\/O\Ph one. The removal of the benzoyl group from I l m c o p h this salt was not possible. An account is \/ / included of some compounds which have been prepared in order to throw light on the property of fluorescence in relation t,o structure in this group. 8- Ethoxy -2-phenylbenxop yrylium ferrichloride C ,H1502C14Fey forms brown plates m.p. 139-140" ; 3-benxoylamino-8-ethoxy- 2-phenylbenxopyrylium ferrichloride (I) orange-red needles m. p. 3 - benzoylamino - 4'- methoxy - 8 - Ho OEt ethoxy - 2 - phenylbenzopyrylium ferri - /-\ chloride C2,H,,04~Cl,Fe reddish-brown Fea4 gH-\-/ plates m. p. 201 . 5-Ethoxy-4-(2-hydr- oxy-3-ethoxybenzylidene)-l 2 3 4-tetra- hydroxanthylium ferrichloride (11) is ob- tained from cyclohexanone and 2-hydr- (11.1 CH oxy-3-ethoxybenzaldehyde. It is a chocolate-coloured powder m. p. 197". The chloride obtained from this when boiled with water containing 8 FeCl (1.1 - 198" ; EtO e ci. 308 ABSTRACTS OF CHEMICAL PAPERS. little hydrochloric acid is converted into d i e t h o x ~ ~ ~ e t h y ~ ~ ~ b e n x OEt spiropymn (111) colourless needles ,,/ \c/ m.p. 155". u-Naphtliylrnethoxymethyl ketone C,,H,*CO~CH,*OMe is ob- I ~-C€€:C/\C:CH-I 1 tained from methoxyacetonitrile \/ and magnesium u-naphthyl bromide. It is a pale yellow viscid oil " h CH b. p. 184-186"/13 mm. It con- denses with 2-hydroxy-5-methyl- benzaldehyde t o 3-methoxy-2-~-mphthyl-6-methy7benmpy~lium chloride C,,H1~02Cl,3H20 orange needles m. p. 84". The ferri- chloride crystallises in plates m. p. 184". With 2-hydroxy-3-ethoxy- benzaldehyde the above ketone yields c1 3-methoxy-8-ethoxy-2-u-naphth ylbenmpyrylium 6 chloride C2,H,,0,C1,2H,0 orange-red plates m. p. 115-116" whilst with p-hydroxy- a-naphthaldehyde 3-methoxy-2-a-naphthyl- naphtha-(2 1)-pyrylium chloride (IV) is ob- tained in red microscopic plates m.p. 94". The ferrichloride forms bright red plates m. p. 183". 6 4'-Dimethoxy-2-phenyl6enxo- pyrylium chloride forms orange-red plates m. p. 135"; the ferri- chloride crystallises in brownish-red plates m. p. 192-193". 8-Methoxy-p-naphtha- f 'I cournarono-(b a-2 3)-benzopyrylium chlmide FeCI /\,& /\/ (V) is prepared from o-vanillinand p-naphtha- I I I 1 coumaranone. It crystallises in green \o/\/ prisms m. p. 236-240". When warmed with ethyl alcohol and a trace of sodium acetate it is decolorised and the colourless ethgl ether of the pyrnnol is obtained in bhe form of needles m. p. 165". F. G. P. EtO /O o\ I \/ WI.) a 2 A'o\CloH,(a) UV.) v Me0 F\ (V. 1 " Tervalent " Carbon. I. The Tetra-arylallyl Radicals and their Derivatives. K. ZIEGLER [with G.BRERIER F. THIEL and F. THIELMANN] (AnnaZen 1923 434 34-78; cf. A. 1922 i 1047).-This investigation was undertaken to determine the factors responsible for the dissociation of poly-substituted ethane derivatives. For this purpose compounds of the type of octaphenyldiallyl were studied. Assuming that the theoretical considerations developed by Thiele and Werner are correct the unsaturated p p-diphenylvinyl group would possess a markedly higher degree of " valency require- ment " than the relatively saturated phenyl residue. Comparison of the dissociation capacity of the diallyl derivatives with that of the hexaphenylethanes provides a test for the accuracy of the theoretical considerations advanced by various authors. The first results of the work are given in an eklier paper (Zoc.cit.). 9-pp- Diphenyl~~inylxanthenol o<$H~>c<CH:Cph C H OH has now been obtained in needles which ol'; seating lose wkter and pass into the allene derivative. Characteristic salts are obtained particularlyo m m o OHEMISTBY. i. 309 the perchbrute as well as acid halogenides such as the chloride- hydruchhide 0<2%>C( Cl,HCl)*CH:CPh red needles m. p. 156-157' and the bromide hydrobromide m. p. 130-131"; per- halogenides namely the chloride tetrubromide and bromide tetra- bromide reddish-brown powders decomp. 123-124' and 128-129" respectively. Double salts of the xanthyl chloride are obtained by addition of heavy metal salts to solutions of the chloride hydrochloride in glacial acetic acid. The zinc chloride salt is a red crystalline solid decomposing at 165-166'.Hydrolysis of the salts regenerates the xanthenol whilst tertiary amines such as pyridine and dimethylaniline convert them into the allene deriv- atives. Methyl and ethyl alcohols convert the hydrohalogenidea into 9-met7wxy- and 9-dhoxy-9- pp-diphenylvinylxunthen crystallising in needles m. p. 133-134" and 136-137" respectively. These ethers are readily decomposed by acids into alcohol and the coloured salts. The salts particularly the chloride hydrochloride dissolve in saturated alcoholic hydrogen chloride unaltered and on reduction by zinc the parent substance 9-BB-di~henyZvinyZxunthe.iz has been obtained in large colourless crystals m. p. 16G165". This has also been obtained from 9-xanthylacetic acid by the Grignard reaction.Aryl and alkyl derivatives of 9- pp-diphenylvinylxauthen are obtained by the action of the Grignard reagent on the xanthyl salts particularly on the perchlorate. The xanthen and its alkyl and aryl derivatives are easily reduced by hydriodic acid the pp-diphenylvinyl group probably being eliminated as aac-diphenyl- ethane. 9-EthyZ-9-pp-diphenyZvinyZxanthen crystallises in colourless crusts m. p. 129-130". These xanthens when acted on by bromine yield perbromides such as O<c6H4>C(Br*Brz)H C H and brominatd compounds of the type of a~p-tribromo-ua-diphenyl- ethune Ph,CBr*CHBr2 m. p. 88-89". Di-p-anisyl ketone and 2 4 4'-trimethoxybenzophenone on treatment with the mag- nesium compound of pp-diphenylvinyl bromide yield alcohols (MeO*C,H4),CR*CH:CPh2 and (MeO),C6H,(MeO*CGH4)*CR*CH:CPh from which crystalline perchlorates may be prepared. The tri- medhozy-compound forms h e red crystals yielding violet-red solutions.Pyridine converts it into 2 4 4'-trimethoxydiphenyl- y y -cliphenylallene ( MeO)2CG€€,*C( C6~40Me):c:CPh2 colourk%s crystals m. p. 13A135". Tetraphenylpropene derivatives have been prepared aayy-tetruphenykUyZ ethyl ether Ph,C(OEf)*CH:CPh resulting from the action of alcoholic sulphuric acid on the corre- sponding alcohol. It forms. colourless crystals m. p. 106-107". TetraphenyZuUyZ perchbate obtained from tetraphenylallyl alcohol is an unstable chocolate-brown powder decomposing a t 63-64'. On keeping it is converted into the colourless 1 1 3-triphenyl- indene. Results show that the coloured compounds of the sub- stituted tetra-allylallenes are true tetra-arylallyl salts that tetra- arylallyl alcohols are pseudo-bases ; the chlorides were not isolated losing hydrogen chloride and giving tetra-allylallenes ; these allenes readily add acid 60 as to form salts.Magnesium aryl 6 4 6 4i. 310 ABSTRACTS OF CHEMICA-L PAPERS. haloids convert the xanthen perchlorate into the free di-9-pp- diphenylvinylxanthyl O<c6$>C*CH:CPh2 C H a colourless crystalline 6 4 powder m. p. 146145" spanngly soluble in most reagents. This dissolves in benzene with a yellow colour the solution becoming brown on warming turning yellow again on cooling. Air com- pletely decolorises the solution a peroxide [O<C,H c6H4>C(O-)*cII:CPh2] 2 being obtained. Bromine converts the coloured solution into 9-p P-diphenylvinylxanthyl bromide tetrabromide.Hydrazobenzene reduces it to 9-PP-diphenylvinylxanthen (cf. triphenylmethyl). Sodium and potassium give coloured metal derivatives the potass- ium derivative crystallising in the form of a copper-red sludge. Dry air decolorises the solutions of the alkali metal derivatives water also decomposes them with production of the xanthen whilst ethyl bromide converts them into the ebhylxanthene. Tetra- phenylhydrazine converts the free radical into 9-pp-diphenylvinyl- xanth yldiphenylumine 0 <::2>C( NPh,)*CH:CPh crystallising in colourless plates m. p. 196196" (cf. triphenylmethyl). These analogies with hexaphenylethane show that the xanthyl compound in solution exhibits radical dissociation.Molecular-weight deter- minations in naphthalene show that the xanthyl is decomposed to the extent of about 60% into 2 mols. of radical-like form. The free tetra-arylallyl radicals are also obtained by the decomposition of the corresponding perchlorates with the Grignard reagent. Dianisyldiphenylullyl (MeO°C6H4),*C*CH:CPh2 crystallises in dark green plates 2 4 4'-trimethoxydiphenyl-yy-diphenyhllyl is similar but the colour is deeper. uuyy-Tetraphenyhllyl the parent substance is a green crystalline powder molecular-weight determinations in benzene showing a degree of dissociation of 80%. These three compounds are exceedingly stable. The corresponding peroxides are very unstable. The sodium compounds of the ally1 radicals are readily converted into the corresponding propenes.In the triarylmethyl series the union of the two benzene nuclei through oxygen opposes the association of the radicals for 9-phenylxanthyl in naphthalene is present to the extent of about 50-70% in the unimolecular form and about 30-50"/9 in the bimolecular form,' whilst hexaphenylethane under like conditions is only dissociated to the extent of 15-25y0. Tetraphenylallyl on the other hand in benz- ene solution at 4" is present in the bimolecular form to the extent of a t most 20 yo whilst 9- pp-diphenylvinylxanthyl in naphthalene at 80" is present perhaps to the extent of 40%; a t 4" therefore it is still more strongly associated. If according to Thiele and Werner the high valency requirement of a residue is explained by its unsaturated character based on the principle of a distribution of aEnity the existence of tetra-arylallyl radicals is a weighty argument in favour of this principle.ota-Diphenylethylene is essentially more unsaturated than benzene so that it must be (MeO),*C,H,*C( C ,€€,OMe)*C*CH:CPh,,ORUANIC CHEMISTRY. i. 311 assumed that the p p-diphenylvinyl residue requires more valency than the phenyl residue and in agreement with this it is found that on replacing two phenyl residues in hexaphenylethane by two such residues the tendency to dissociate is very much increased. F. G. P. Atropine Phosphates. L. DEBUCQUET (J. Pharm. Chim. 1924 29 15-32).-The salt C17H,03N,H3P04 is obtained in 92 yo yield when an alcoholic solution containing equimolecular quantities of atropine and phosphoric acid is heated for half an hour in a water-bath at 90-95" and then left to crystallise.Small lustrous crystals (m. p. 196') are obtained which are soluble in rather less than five times their weight of water (giving a solution acid to litmus) sparin ly soluble in cold alcohol and almost in- (C17H,0,N),~H3P04 gave only the monobasic phosphate (in diminished yield) and unchanged atropine. soluble in ether and c % loroform. Attempts to prepare the salt W. T. K. B. Stereoisomerism and Local Anaesthetic Action in the p-Eucaine Group. Resolution of p- and iso-p-Eucaine. H. KING ( J . Chem. Soc. 1924 125 41-57).-0n benzoylation a- and p-vinyldiacetonalkamines yield 0-benzoyl derivatives and for convenience the name of iso-p-eucaine is proposed for the deriv- ative of the p-compound.P-Eucaine has been resolved by means of camphorsulphonic acid. With d-camphor-10-sulphonic acid the partial racemate dl- p-eucuine d-camphor-10-sulphonate was obtained working between the ordinary temperature and zero and it has been found possible to resolve this salt by means of the more soluble active forms which are unstable in respect of the partial racemate a total yield of 54% of 1-p-eucaine d-mmphor- sulphonute being obtained. Owing to the formation of a continuous series of mixed crystals between the two active salts the more soluble d- p-eucaine d-camphorsulphonate could not be isolated but the isolation of d- p-ewine 1-mmphorsuEphcmute was finally accomplished by the use of E-camphor-10-sulphonic acid. iso- p- Eucaine was resolved by the use of d- a-bromocamphor-r-sulphonic acid.In this case the salts being of approximately equal solubility separate alternately from the solution on fractional crystallhtion. On the rabbit's cornea r-p-eucaine r-iso- p-eucaine and their optic- ally active components have approximately equal local anaesthetic action; on the sciatic nerve of the frog however p-eucaine is a more powerful anzesthetic than .iso- (3-eucaine. dl- p-Eucaine d-camphorate forms plates m. p. 211-212"; the 1-wzulate has m. p. 212-213" and [.ID -2.6"; the d-tartrate crystallises in tablets which have m. p. 257" and [.ID +10-1". The d-camphor- 10-sulphonate (mono-alcoholate) forms tablets m. p. 228-229" [m]D +10.3' (anhydrous salt). Z- p-Eucaine d-camphorsulphonate forms leaflets m. p. 248-249" [.ID 4-78' and 4-83" whence the I-p-eucainium-ion has [.ID - 11".d-p-Eucaine I-camphorsulphon- ate has [.ID -5.09" whence the d-p-eucainium-ion has [mb +lO*S". I-p-Eucaine forms prisms m. p. 57-55'. The hydrochloride forms .i. 312 ABSTRACTS OF CXiEMICAL PAPERS. plates (m. p. 244-245" [.ID -11-3") the picrate (small prisms) having m. p. 198-199". d-p-Eucuine has m. p. 57-58" the hydrochloride having [.ID +11.5". r-p-Eucuine hydrochloride forms tablets m. p. 277-270"; the free base has m. p. 70-71"; the picrate (plates) has m. p. 230.5-231.5". d-a- Vinyldiacetonalkamine hydrochloride (tablets m. p. above 300") has [.ID +13.3"; the l-hydrochloride has m. p. above 300" and [a]54sl -118.5". The l-base f o m prisms m. p. 79-81" (picrate needles m. p. 242- 244").N-Benxoyl- p-vinyZdiacetonallmine obtained by benzoylation of the p-alkamine base forms leaflets m. p. 121-122"; the O-benzoyl derivative obtained by benzoylating the hydrochloride forms needles or plates m. p. 269-271". Its picrate has m. p. 256-258". I-iso- p-Eucaine d- u-bromocamphor-R-s~p~~te forms needles m. p. 236-238" and has +52*1° and [MI5461 +290.9" whence for the Z-iso-p-eucainium-ion is -55.6". d-iso- p-Eucaine d-u-bromocamphor-R-~lpho~te forms needles con- taining 0-5-1 mol. H,O the anhydrous salt having [ct]5461 +70" +390" whence the d-iso- p-eucainium-ion has [ +17-8". d-iso-P-Eucuine hydrochloride forms needles m. p 271- 273" [a15461 +14.9". The l-hydrochloride has m. p. 271-273" and [p]5461. -114.25". The 1-picrate forms needles m. p. 280". d-P- Vznyldzacetonulkamine hydrochloride tablets has m.p. 217- 219" ['x]5461 +22*85". The base has m. p. 121-123"; the pkrate (needles) has m. p. 181-182". ON-Dibenzoyl-a-vinyldiaceton- aZkumine obtained by benzoylation of 8-eucaine base crystallises in plates m. p. 142-143". The similar dibenzoyl derivative from the iso-p-eucaine base forms prisms m. p. 114-115". 3'. G. P. Complex Thiocyanates of Tervalent Metals. 111. G. SCAGLIARINI and G. TARTARINI (Gazzetta 1923 53 876-878).- Unlike variadium (A.y 1923 i 547 1225) molybdenum forms complex thiocyanates which are highly stable and resemble indeed those of chromium. Pipem'dine 2.anadih~athiocyanate (C,H,,N*H),[V( SCN) (i] forms deep red acicular crystals ; basic pyradzne 2.anaditetrathaocyanate (C5H,N.H),[V [ ~ ~ ~ ) ~ ] small brick-red crystals ; hexamethylene- tetramine moly6dihe.~~thiocyanate (C6H,,N4*H),IMo(SCN)16 straw- yellow crystals ; basic pyridine molybdipentathiocyanate orange-yellow crystals ; and basic p-peridine molybdipe?? tathio- cyana-ley ( C5H l,N*H)s[Mo (SCN),OH] brick-red crystals.(C,H,N'H),rMo(SCN),OH1 T. H. P. Galegine. BARGER and WHITE.-(See i 272.) The Synthesis of Pyridine from Aldehydes and Ammonia. A. E. TSCHITSCHIBABIN and (MLLE.) 31. P. OPARMA (J. Russ. Phys. Chem. Soc. 1924 54 601-606; cf. A. 1923 i 1121 et seq.).- Pyridine can theoretically be formed by the interaction of ammonia either with one molecule of acraldehyde and one of acetaldehvdeORGANICI CHEMISTRY. i. 313 (scheme I) or with two molecules of acetaldehyde and one of formaldehyde (scheme 11) thus Either scheme would account for the formation of small quantities of pyridine in the experiments of Stohr (A.1891 579; 1892 628) because it is well known that acraldehyde is formed from formaldehyde and acetaldehyde and that the process is reversible. It is now shown that mixtures of acraldehyde and acetaldehyde or of acet'aldehyde and ethylal can be made to react with ammonia under the influence of alumina (loc. cit.) yielding pyridine as the main product. In the former case some p-picoline and a small quantity of a-picoline were also obtained whilst in the second case a- p- and 7-picoline and 3 5-dimethylpyridine were isolated in small quantities. The formation of p-picoline from ethylal acetaldehyde and ammonia is doubtless due to the intermediate formation of acraldehvde which then reacts with ammonia whilst pyridine can be prodbced according t,o either scheme (I) or (11).G. A. R. K. 4Methyl-2-ethylpyridine. A. E. TSCHITSCHIBABIN ( J . Buss. Phys. Chem. SOC. 1924 54 607-610) .-Pure y-picoline regen- erated from its picrate was heated with ethyl iodide in sealed tubes at 310-325" and the new base isolated in the form of its picrate short yellow prisms m. p. 120-121"; the base regenerated from this picrate boiled at 172" and was characterised by the formation of the chloroplutinate short sparkling orange prisms m. p. 176" (decomp.) and the chlwmurate yellow prismg m. p. 80". The synthesis of 4-methyl-2-ethylpyridine has since been described by Eckert and Loria (A. 1918 i 79) but the properties of the derivatives prepared by'them differ somewhat from those given above.The Condensation of Acetylene with Ammonia in the Presence of Alumina. A. E. TSCHITSCHIBABIN and I?. A. M O S ~ (J. Russ. Phys. Chew. SOC. 1924 54 611-624).-A preliminary account of this work has already been published (A. 1915 i 638); a detailed description is now given of the. method of synthesis and of the separation of the pyridine bases formed by the fractional precipitation of their picrates. Pyrrole bases were found to be present; of the crude condensation product some 3074 consisted of picolines which were isolated in the form of their picrates and by crystallising the latter from acetone were shown to consist of the a-compound with small quantities of the y-isomeride. The collidine fraction constituted about 10% of the original mixture.The separation was again effected through the picrates which mere crystallised from ethyl acetate. The presence G . A. R. K.i. 314 ABS'I'RACTS OF CHEMICAL PBPERS. of p-collidine (4-methyl-3-ethylpyridine chloroaurate m. p. 140- 141" chloroplatinate m. p. 215" [decomp.] picrate m. p. lag") identical with that obtained from cinchonine and %methyl- 5-ethylpyridine (picrate m. p. 164" chloroaurate m. p. 88") was confirmed. In addition two other collidines were isolated one of b. p. 17&-182"/756.7 mm. (picrate orange rods m. p. 143"; chloroaurate m. p. 119-121" ; chloroplutinate m. p. 207" [decomp.]) and another of b. p. 185.5-186.5"/756.7 mm. (picrate short flattened orange needles m. p. 134"; chloroaurate plates m.p. 116-118" ; chloroplatinate orange crystals m. p. 210" [decomp.]). These collidines do not appear to be identical with 4-methyl- 2-ethylpyridine (preceding abstract) or with either 2 3 4- or 2 3 6-trimethylpyridine which might possibly have been pro- duced; but it is thought that the compounds may not be pure. G. A. R. K. The Absorption Spectra of some Indole Derivatives. F. W. WARD (Biochem. J. 1923 17 891-897).-Using the Hilger ultra- violet spectrum photometer the absorption spectra of the follow- ing compounds have been charted indole indole-3-carboxylic acid indole-3-aldehyde indole-3-ethyl alcohol indole-3-propionic acid indole-3-alanine (tryptophane) 2 3-dihydroxyindole 2-hydroxy- 3-indole-aldehyde isatin and sodium indigotinsulphonate.The effect of substitution in the indole nucleus is discussed. J. P. Certain Quinoline4carboxylic Acids. S. BERLINGOZZI and C . M~RZELLA (Atti R. Accud. Lincei 1923 [v] 32 ii 403-406; cf. A 1923 i 482 847).-!J%e action of acetonyl- or phenacyl- phthalimide on isatoic acid gives rise in the usual way to com- pounds of the type ~ When heated beyond their melting points all these quinoline- 4-carboxylic acids lose carbon dioxide yielding the corresponding S-amino- and 3-hydroxy-quinolines ( b c . cit.). Those of the above acids which have a phenyl group in the 2-position may be regarded as derivatives of 2-phenylquinoline-4-carboxylic acid so largely used in therapeutics. The mpound (annexed f)-C( C0,H):~*N€€*CO*C6H4*C02H formula) obtained from isat- \/-N=CMe oic acid and acetonylphthal- imide crystallises in minute white needles m.p. 267-268" (decomp.). 3 -Amino- 2-rnethylpuinoline-4-carbox y lic acid C 11H1,0~2 ob- tained together with phthalic acid when the preceding compound is boiled with concentrated hydrochloric acid forms lemon-yellow0mANIo OHEMISTRY. i. 315; prisms m. p. 232-233" (decomp.) yields solutions which develop fluorescence on addition of a mineral acid and gives %amino- 2-methylquinoline when heated at 233-235". Its acetyl derivative forms a white microcrystalline powder m. p. 272-273". 3-Hydroxy-2-methylqui~wline-4-mrboxylic acid prepared by the action of nitrous acid on the preceding amino-compound forms minute pale yellow crystals m. p. 229-230" (decomp.) and is converted into 3-hydroxy-2-methylquinoline when fused ; its alcoholic solutions are fluorescent and are coloured red by ferric chloride.The compound formed by condensation of isatoic acid with phenacylphthalimide crystallises in lustrous straw-yellow scales m. p. 260" (decomp.). 3 - Amino-2 -phenyZquimline-4-mrboxylic acid forms pale yellow needles and prismatic plates m. p. 226" (decomp.) exhibits fluores- cence in acid solution and gives 3-amino-2-phenylquinoline when heated. Its acetyl derivative crystallises in lustrous white scales m. p. 257-258". 3-Hydroxy-2-phenylquinoline-4-carboxyZic acid is obtained as a lemon-yellow microcrystalline powder m. p. 206-207" forms fluorescent alcoholic solutions which are reddened by ferric chloride and gives 3-hydroxy-2-phenylquinoline when heated.T. H. P. The Absorption Spectra of Kynurenic Acid and some Related Quinoline Compounds. F. W. WARD (Biochem. J. 1923 17 903-906).-The absorption spectra of quinoline quin- aldine l e p i h e quinaldinic acid 2 6-dimethylquinoline kynurenic acid and of quinolylacetaldehyde have been mapped. The last- mentioned compound gives an abnormal spectrum and doubts are *expressed as to its constitution. The effects of substituents in t,he quinoline nucleus are discussed. Some Heterocyclic Derivatives of Substituted o-Amino- phenols. A. KORCZYNSKI and ST. OBARSKI (Bull. SOC. chim. 1923 33 [iv] 1823-1832).4-Chloro-2-nitrophenol on electrolytic reduction gave 4-chloro-2-aminophenol (m. p. 185" ; diacetyl derivative m. p. 201"). The following were prepared in it similar manner 4-chloro-6-bromo-2-aminophenol (metyl derivative m.p. 150") 4- bromo-2-a.minopheno1 4 6-dibromo-2-aminopheno1 and 4 6-di-iodo-2-aminophenol. 4-Chloro-2-aminopheno1 on heating with acetamide (cf Niemen- towski A. 1898 i 210) gave 4-chloro-l-methylbenxoxazole (m. p. 63"). The following compounds were obtained in a similar manner 4-homo-l-methylbenzoxaxole (m. p. 70°) 4-chloro-6-bromo- l-methyl- benxoxamle (m. p. 117") and 5 6-dibromo-l-meth~lbenxoxaxole (m. p. 127"). 4-Chloro-2-aminophenol was converted (cf. Chefmicki A. 1887 477) by thiocarbonyl chloride in presence of pyridine into 4-chloro- l-thiolbenxoxaxole (m. p. 283") and the following compounds were prepared in a similar manner 4-chloro-6-bromo- 1 -thiolbenzoxaxole (m. p. 228") and 4 6-dibromo-l-thiolbenxoxazole (m.p. 235"). All the above thiol derivatives form yellow needles soluble in alkali. J. P. All the above oxazoles form yellow needles.i . 316 ABSTRACTS OF C H W ( I B L PAPERS. 4-Bromo-2-aminophenol (2 mols.) on boiling in alcoholic solution with c h l o r o d (1 mol.) aff or& 8-bromo-4-ccminophenxaxone (giving a blue solution in concentrated sulphuric acid) the following com- pounds being similarly prepared 8-&oro-2 lO-d~bromo-4-amino- phenoxazune (reddish-violet solution in concentrated sulphuric acid) 2 8 10-tribromo-4-aminop~nomzone and 2 8 lO-tri-~od0-4-amino- phenoxazone (blue solution in concentrated sulphuric acid). The above phenoxazones form red needles not melting at 300". Chloroanil converts 4-nitro-2-aminophenol into a substance which forms orange-red needles.E. E. T. Diphenylyl Phenylenediamines and the Colour of their Oxidation Products. J. PICCARD and F. ABOUCHY (Helv. Chirn. A& 1924 7 75-83).-To compare the auxochromic effect of the diphenylyl group Phec6H4- with those of phenyl and alkyl groups the following derivatives of p-phenylenediamine were prepared for oxidation into meri-quinonoid salts s-di-p-diphenylyl-p- phenylenediamine (I) ; s-diphenyldi-p-diphenylyl-p-phenylene- diamine (11) and tetra-p-diphenylyl-p-phenylenediamine (111) . To prepare the first compound (I) p-dibromobenzene was condensed with s - diacet y M i - p-diphen y Ey 1 -p - phenylenedkmine colourless crystals m. p. 245". By alkaline hydrolysis this gave s-di- p - diphen yl yl -p -phen yl enediarnine long colourless spangIes m.p. 256". The second base (11) was pre- pared in two ways (a) by condensing (I) with iodobenzene and ( b ) by condensing p-iododiphenyl with s-diphenyl-p-phenylene- diamine. s- Diphenyldi- p-diphen yl yl -p-phenylenediamine forms small crystals m. p. 211". The third base (111) was also obtained in two ways (a) by condensing (I) with p-iododiphenyl and (b) by condensing p - di -iodo b e m ene with di-p - dip henyly lamine. The product tetra_p-diphenylyl-p-phenylenediamine forms small colour- less crystals m. p. 280". Of these three bases only the first can be oxidised to a quinonedi-imine. By oxidation with chromic acid it gives di-p-d~~~nylylbenzoquinonedi-irnine Ph*C,H,N:C,fl,:N*C,H,*Ph a red amorphous powder m. p. 266O. It is less basic than diphenyl- benzoquinonedi-imine and the colours of its salts are deeper ; its mono-acid salts are blue di-acid salta violet and mri-quinonoid salts green.meri Di- p -diphen yl yl benxoquinonedi-imoniurn picrate (Ph *C 6H4NH :C,H,:NH*c ,H4*Ph::: Ph*C6H4NH*C6H,*NH*C6H,Ph]rC,H,( NO,),O$. forms a bluish-green powder m. p. 208'. A picrate of the meri- quinonoid salt of the second base was also isolated; i t forms t~ yellowish-green powder m. p. 169". The colours of the meri- quinonoid salts of the three bases in solution are respectively bluish-green green and yellowish-green. The colours of the holo-quinonoid salts are only developed in concentrated sulphuric acid solution and are respectively violet bluish-violet and blue. It is noted that glacial acetic acid dissolves the quinonedi-imine compounds as bases the colour of the salts being developed only on addition of water.The leuco-compounds are weaker bases p - acetamidodi p heny 1 givingORGANIC CHEMISTRY. i. 317 than the hZo-quinonoid compounds but the meri-quinonoid com- pounds are stronger bases than either of these. The auxochromic effect of the diphenylyl group in meri-quinonoid salts in the phenyl- enediamine benzidine or fuchsine series is less than that of two phenyl groups but considerably greater than that of one group. The phenylated salts in the benzidine series are deeper in colour than the diphenylylated salts in the phenylenediamine series. The difference between the effect of the phenyl and diphenylyl groups is more marked in the hoZo-quinonoid than in the meri-quinonoid salts.It may be stated ~ E I a general rule that the weight of a group has more effect in the active chromophore than in an auxo- chromic group. E. H. R. The Reaction of Carbon Disulphide on p-PhenyIenediamine. C. E. BOLSER and E. B. E~RTSHORN (J. Amer. Chem. Xoc. 1923 45,2349-2355) .-When p-phenylenediamine and carbon disulphide are boiled in alcoholic solution (Pawlewski A. 1899 ii 405) a small quantity of pp'-diaminodiphenylthiocarbamide is produced together with an insoluble compound whioh has been described m T T 1 as p-phenylenethiocarbamide C6H4<~~>C:S (Lellmann A. 1883 185). This compound is now shown to be di-p-phenylene- dithiocarbamide C6Ha<m.cs.m hTecs'NH>c6H4. In the &st place its formation may be almost entirely suppressed the yield of diamino- diphenylthiocarbamide being correspondingly increased by con- ducting the process in aqueous solution by keeping the temperature of reaction low and by decreasing the concentration of carbon disulphide.It is advisable frequently to remove by filtration the diaminodiphenylthiocarbamide which is less soluble in warm alcohol than in hot and still less soluble in water. This suggesfs that the simple .thiocarbamide is an intermediate stage in the formation of the insoluble compound; that this is actually so is shown by the formation of the latter when carbon disulphide and pp'-diaminodiphenylthiocarbamide react in boiling alcoholic solu- tion. The identity of the product with that formed directly from carbon disulphide and p-phenylenediamine is shown (although in neither reaction is an absolutely pure substance obtained) by the temperature of decomposition (285O) by the formation of p-phenylenediamine by the action of hot concentralxd acids and by the formation of p-phenylenedithiocarbimide by the pro- longed action of boiling acetic anhydride.The latter reaction proves that the compound has the above structure since p-phenyl- enethiocarbamide could not give the dithiocarbimide but should be converted into p-acetamidolphenylthiocarbimide white needles m. p. 195-196'. The latter is produced by the action of boiling acetic anhydride on di-p-acetumidophenylthiocarbamide a white solid m. p. 235-237" which is formed by the action of an excess of carbon disulphide on p-aminoacetanilide in boiling alcoholic solution.It was hoped that the hydrolysis of this derivative which requires the use of hot concentrated hydrochloric acid VOL. CXXVI. i. 172i. 318 ABSTRACTS OF CHEMICAL PAPERS. might give di-p-aminophenylthiocarbamide (thus avoiding the concurrent formation of the insoluble compound) but actually phenylenediamine hydrochloride is the sole product. p The reduction by means of stannous chloride and a mixture of glacial acetic acid and concentrated hydrochloric acid of axobenxene-p-thiocarbimide Ph*N:N*C,H,*NCS salmon-coloured needles m. p. 94-95' which is obtained by heating thiomrbonyl chloride and p-aminoazobenzene in carbon tetrachloride solution at W O O gives diphenylenedithiocarbamide but the latter does not appear until the solution is made alkaline.It is probable that the acid solution contains in equilibrium with the dimeric compound p-aminophenylthiocarbimide (hydrochloride). In fact when the acetyl- derivative (above) of the latter compound is boiled with glacial acetic acid in the presence of this weaker acid diphenylene- dithiocarbamide is precipitated more of the same material being thrown down on dilution. It is evident that the tendency to the formcation by what is practically a tautomeric change of this substance is so great that free p-aminophenylthiocarbimide cannot exist but on liberation from its salts immediately undergoes N e w Methods of Splitting Pyrimidines. 111. The Action of Iodine Solution on Pyrimidines. L. W. BASS and 0. BAUDISCE ( J . Amer. Chem. Xoc. 1924 46 181-183; cf.A. 1922 ii 328).-Pylunidines with unsubstituted nitrogen atoms and with a double bond between carbon atoms 4 and 5 absorb iodine from solution in the presence of sodium bicarbonate. On distillation of the reaction mixtures after removal of excess iodine carbamide is found in the residues as a scission product. No carbamide can be detected before the distillation. Substitution by an alky€ group in position 5 retards the rapidity of decolorisation of iodine solution. Hydrouracil decolorises iodine solution but no carb- amide ia formed when the reaction mixture is distilled. 1 3-Di- methylthymine does not decolorise iodine solution. Thymine &If yielda carbamide and acetol; no pyruvic acid is formed. When bromine is substituted for iodine the pyrimidine ring is not split under the conditions of the experiments.N e w Mathods of Splitting Pyrimidines. IV. A Study of the Mechanism of the Decomposition of Thymine. 0. BAUDISCEI and L. W. BASS (J. Anzer. Chem. Soc. 1924 46 1 8 A 189; cf. preceding abstract).-The action of vazious oxidising agents on thymine has been studied; the primary products of the reactions still contain the carbamide residue. The final products cazbdde scetol and pyruvic acid are formed only when the primary products are heated in aqueous solution with sodium hydrogen carbonate. Acetol is formed as a direct hydrolytic product of thymine whilst pyruvic acid is formed by the complete hydrolysis of an intermediate oxidation product. A mechanism is suggested to explain the decomposition of thymine under the conditions used.F. A. M. polymerisation. w. s. N. F. A. M.ORGANIC CHEMISTRY. i. 319 New- Methods of Splitting Pyrimidines. V. The Action of Oxygen plus Ferrous Salts on Thymine under the Influence of Light. L. W. BASS ( J . Amer. Chem. SOC. 1924,46 190-192; cf. preceding abstracts).-When in an atmosphere of oxygen aqueous solutions of thymine are illuminated using a quartz mercury arc the pyrimidine is split to some extent into carbamide and pyruvic acid; no acetol is formed. The reaction is accelerated but other- wise not affected by ferrous sulphate or to a smaller extent by potassium ferrocyanide. No scission occurs in an atmosphere of nitrogen even in the presence of the above salts. F. A. BK Catalysis. 111. Reduction of Uracil to Hydrouracil. E.B. BROWN and T. B. JOHNSON ( J . Amer. Chem. Soc. 1923,45 2702- 2708; cf. A. 1921 i 806; 1922 i 709).-hther experimental details are given relating to the reduction of uracil to hydrouracil by means of hydrogen in the presence of colloidal platinum. Col- loidal palladium may also be used. Sulphur compounds even when present only as traces completely inhibit the reaction but the chlorides of sodium potassium calcium and magnesium are quite without effect. Synthetic Researches on the Constitution of Bile-pigments. 11. H. PISCHER and J. M&ER (2. physwl. Chem. 1924 132 73-103).-If dry ammonia is passed through a cold alcoholic solution of ethyl a-acetyl- p-methylsuccinate crystals are obtained m. p. 77" (decornp.) apparently of ethyl p-amino-AP-pentene-yS-di- carboxylate which on keeping or on warming in a vacuum a t 40-50" completely change into ethyl 2-hydroxy-3 6-dimethyl- pyrrole-4-carboxylate m.p. 127" a compound which is also found in the mother-liquors from the crystals. This pyrrole derivative when warmed in 60% sulphuric acid yields a-methyl-laevulic acid and when treated with hydrazine hydrate is converted into the hgdrazide of 2-hydroxy-3 5-dimethylpyrrole-4-carboxylic acid sharp-angled leaflets decomposing at 212". This compound may be converted by nitrous acid into the cczide of 3 5-dimethyl-2- hydroxypyrrole-4-carboxylic acid flat prisms which decompose explosively a t 114" and this compound in solution in boiling absolute ethyl alcohol yields ethyl 2-hydrow-3 6-dimethyl.pywl- 4-carbamte white needles decomposing at 148" whilst if methyl alcohol is used the corresponding methyl curbamute is obtained as white glistening needles decomposing at 164".From ethyl acetylsuccinate the following corresponding series of compounds is obtained ethyl p-amino-A@-butene-y8-dicurboxylute m. p. 122- 124" (decomp.) ; ethyl 2-hydroxy-5-methylpyrrole-4-carboxylate m. p. 134"; (hgdraxide leaflets decomposing a t 232"; axide needles which decompose explosively at 135" ; methyl curbamute needles m. p. 175"). Ethyl 2-hydroxy-5-methylpyrrole-4-carboxylate condenses with aldehydes in alcoholic solution in presence of hydrochloric acid and the following compounds are described ethp? 2-keh-3-benzyl- idene-5-~thyE-2:3-dihydropyrrole-4-mrboxykcte yellow leaflets W. S. N. rn2i.320 ABSTRACTS OF CHEMICAL PAPERS. m. p. 184" ; ethyl 2- keto- 3 -p -d imethylaminobenxylide~-5-~thyl- 2 3-d&ydropyrrole-4-curboxylate red needles from alcohol m. p. 231" (hydrochloride yellow needles unstable in air decomposing a t 210" perch,lorate yellow needles stable in air m. p. 150" [decornp.]) ; ethyl 2-keto-3-o-nitrobenzylidene-5-methyl-2 3-dihydro- pyrrok-4-carboxyZute felted yellow needles m. p. 201" ; 4-curb- ethoxy-2-hydroxy-5-methylp yrrolidyl- 3'- carbethoxy - 2 4'- dimeth ylpyr- rylmethene (annexed formula) orange- f?Co2Et)*(?:m*$?-~H yellow prisms m. p. 210" with which CMe co CMe CMe is also formed a compound (deep orange- yellow prisms m. p. 162-163") ; and 4- C*COzEt mrbethoxy-2-hydroxy- 5 - methylpyrrolid- yZ-5'-carbethoxy-2' 4'-dimethyl~yrrylmethene needles m.p. 266". Ethyl 2-hydroxy-5-methylpyrrole-4-carboxylate couples with benzene-diazonium chloride to give a dye C $&o& (glistening pure yellow felted needles m. p. 233") whici forms a sodium salt Cl4HI4O3N3Na red needles and a potussium salt orange-yellow needles. When treated with ammoniacal copper solution the dye is converted into an isomeric compound (large orange-red to dark red transparent needles decomposing a t 213") which on crystallis- ation from alcohol yields the original dye. The latter when reduced with zinc dust in hot alcoholic solution or with hydrogen in presence of platinum black yields ethyl 3-amino-2-hydroxy- 5-methylpyrrole-4-carboxylate m. p. 244" (cf. Fischer and Herr- mann A. 1922 i 1054). Dyes obtained similarly from m-nitro- benzene-diazonium chloride and from a-naphthalene-diazonium chloride form thin yellow bunched needles m.p. 233" and red needles decomposing a t 234" respectively. The application of the Gattermann reaction to ethyl 2-hydroxy-5-methylpyrole- 4-carboxylate yields besides a by-product which decomposes at 262" the aldimine C,H120,N2 colourless leaffets m. p. 241" which is not hydrolysed by water to the aldehyde .and is decomposed by alkalis with the liberation of ammonia. The &y& however may be obtained when sodium is added to a boiling mixture of ethyl 2 - h ydr ox y - 5 -met h ylp yrrole - 4- car b ox ylat e and ethyl formate and forms large flat needles m. p. 206". It yields a semicarbazone bunched needles m. p. 252" (with previous browning and sintering) and a hydrazone small rods m.p. 190". It condenses with ethyl 3-hydroxy-5-methylpyrrole- 4-carboxylate to yield 4-mrbethxy- 3-hydroxy- 5-methylpyrrolidyl- 4'-carbethxy-2'-hydroxy-5'-methylpyrrylmethene yellow needles m. p. 250" and with ethyl 2-hydroxy-5-methylpyrrole-4-carboxylate to yield bis- (4-mrbethoxy-2- hydroxy- 5-meth ylp yrryl)methene brownish- yellow prisms m. p. 254". 4-Carbethxy-3-hydroxy-5-methylpyrrole- 2-aldehyde faintly coloured needles m. p. 188" is obtained when ethyl 3-hydroxy-5-methylpyrrole-4-carboxylate is condensed with ethyl formate in presence of sodium. Ethyl oxalate similarly condenses with ethyl 2-hydroxy-5-methylpyrroIe-4-carboxylate to form ethyl 4-carbethox y -2 - h ydrox y - 5-meth ylp yrry 1-3 -gl yox ylate glistening prisms m.p. 180" and with ethyl 3-acetyl-2 4-dimethyl- pyrrole-5-carboxylate to yield ethyl 5-carbethoxy-2 4-dimethyl- \/ \NH/ORUANIC CHEMIS!CRY. i. 321 pyrryl-3-ethanonoxalate fine colourless felted needles m. p. 150.5". The latter is readily decomposed by alkalis to yield 3-acetyl- 2 4-dimethylpyrrole-5-carboxylate and forms a semicurbaxone m. p. 201" and a ketuzine-hydrazide CI3H1,O3N5 needles m. p. 239" (decomp.). Ethyl 2 4-dimethylpyrryl-3-ethanonoxalate forms a semicurbaxone m. p. 198" (decomp.). Ethyl 2 4-dimethyl- pyrrole-3 5-dicarboxylatq when heated with hydmzine hydrate a t 200" in a closed tube yields 2 4-dimethylpyrrole. If 1'-p-toluoyl- 4-carbethoxy- 3 - hydroxy - 5 - methylpyrrolidyl-3' - carbethoxy-2' 6'- dimethylpyrrylmethene is treated with benzene-diazonium chloride decomposition of that compound takes place simultaneously with coupling to give a mixture of 3-carbethoxy-2 5-dimethyl-1-p- toluylpyrrole-4-aldehyde and 2-benzeneazo-4-carbethoxy-3-hydroxy- 5-methylpyrrole.W. 0. K. Oxidation-Reduction. IV. Electrode Potentials of Indigo- tinsulphonates. M. X. SULLIVAN B. COHEN and W. M. CLARK (U.S. Public Health Repts. 1923 38 1669-1718).-A survey of the literature shows conflicting statements regarding the behaviour of indigotin and its sulphonates in vitro and in vivo and a lack of satisfactory theories due t o lack of quantitative data for the oxid- ation-reduction equilibrium. Indigotin-mono- -di- -tri- and -tetra-sulphonates were prepared by Bloxam's method (A. 1906 ii 819) and the electrode potentials at 30" were determined by elaboration of methods already described (A.1923 ii 726 etc.). The experimental results were in accord with the rational equation where Eh is the observed electrode potential referred to the hydrogen standard 8 and So are the concentrations of total reductant and total oxidant (H+) is the hydrogen-ion concentration of the solu- tion and K the acid dissociation constant of one of the two hydrogen atoms of the hydrogenated leuco-compound (dissociation of the sulphoiiic acids not being considered). The number n of electrons in the reversible oxidation-reduction is 2. The second acidic hydrogen atom of the leuco-compound has a dissociation constant below 10-12. The values of Eo and K were as follows indigotin- monosulphonate s leuco-compound +0-262,1-6 x ; indigotin- disulphonate e leuco-compound +0-291 4.9 x 10 ; indigotintri- sulphonate + leuco-compound +0-332,7.7 x ; indigotintetra- sulphonate =+= leuco-compound +0.365 11.2 x lo*. Indigotin - indigo-white is not adapted to accurate experiments in aqueous solution.Approximate data indicate K,=lO-s and Eo near that of the monosulphonate. Anomalies with borate buffers are shown and evidence is given of a " salt effect " with other buffer systems of the acidic nature of the leuco-compounds and of the ionisation of only one hydrogen atom a t moderate alkalinities. An electro- metric method of analysis of mixtures of the sulphonates is developed. The data cover all conditions for the reversible oxidation-reduction of an indigotinsulphonate and give their properties for use in colorimetric determinations of oxidation-reduction intensities of biological solutions.CHEMICAL ABSTRACTS. E n = Eo - [(RT/nP) loge (Sr/flo)I+ (RT/nP) loge [Kl(H+) + (33+)21,Constitution of Isatan and Isatide. A. WAHL and W. HANSEN (Compg. rend. 1924 178 393-396; cf. following abstract).- Isatan (I) C,,H,,O& on heating in boiling naphthalene solution affords indine (cf. Laurent Ann. Chim. Phys. 1840 [iii] 3 469); it results from the condensation of oxindole and isatin in presence of piperidine (cf. Stoll6 A. 1914 i 992) and is the product called isatide by Erdmann ( J . pr. Chem. 1840 [i] 24 l) who treated isatin with excess of ammonium hydrosulphide. Isatan was obtained by Laurent (ZOC.cit.) by reducing disulphoisatide with ammonium hydrogen sulphite. Isatide Cl,H1204N2 (11) &which C H w ( O H ) ,C( OH)---C(OH) m m NH NR (1.1 c&&\ / \ /co co( >c$& c,H,/\ >co co( >C&4 (19. does not give indine on heating is the product obtained by Laurent by treating isatin with a slight excess of ammonium hydrosulphide and by Heller (A. 1904 i 416) by reducing isatin with zinc and acetic acid. Isatide is formed rapidly when a mixture of dioxindole and isatin is treated in alcoholic solution with piperidine. E. E. T. Transformation and Constitution of Disdphoisatide. A. WAHL and W. HANSEN (Compt. rend. 1924 178 214-216).-The formula Cl,H,,0~2S2 given by Laurent (Ann. Chim. Phys. 1840 [iii] 3 469) to disulphoisatide has been confbmed.The formation of indine by the action of alkali on disulphoisatide is shown to depend on the nature and concentration of the alkali used and not to occur if air is excluded. The conversion is best effected by adding aqueous sodium carbonate to an aqueous-alcoholic solution of disulphoisatide. Isatin is produced simultaneously (55% of that obtainable on the basis of the equation 2C16H,,0~2S2 +2H20=C),6HloO$2+4H,S +2C,H,02N). Sulphur is initially C(SH)-.C(SH- formed and reduces indine to leucoiso- indigotin which is oxidised if air is ',II,< >'' co( )'GH4 present. The annexed formula (cf. NH NH Kohn and Ostersetzer A. 1916 i 607 and Lefhae ibid. 430) is proposed for disulphoisatide a disul- phide formula being excluded,lsince :on reduction it gives leucoiso- indigotin (A.1923 i 607). 111. The Nomenclature of Four-membered Rings and the Formation and Properties of some Derivatives of Methylene- 1 2 4oxadi-imine. C. K. INGOLD ( J . Chem. SOC. 1924 125 87-102).-1t has been found that the formation of a four-membered ring by the additive union of double bonds (I) is of considerable generality and thisl confims the prediction (T. 1922 121 2793) made from the analogy with the tautomeric change or internal addition reaction shown in (11) A=B A-B /A=B /A-33 -+ LA \C=h -+ \C-D C=D E. E. T. The Additive Formation of Four-membered- Rings. 1 I (11.) (1.1 +ORGANIC CHEMISTRY. i. 323 A system of nomenclature is proposed for the new heterocyclic ring types that have been already produced. Part of the object of the work is to discover which of these types are the most stable.It has been shown earlier (Zoc. cit. and T. 1923 123 2745) that certain types of azomethines and nitroso-compounds pass into stable cyclic bimerides. It is now shown that an azomethine and a nitroso compound of the particular type combine ad&- yR2-YR tively to give a stable 1 2 4-oxadi-imine (annexed n7 R-o formula). Hethylene-p-chloraniline CH&N*C,H,C1 forms needles m. p. 142" the corresponding p-bromo-derivative needles having m. p. 169". p-Chloronitrosobenxene has m. p. go" the corres- ponding bromo-derivative having m. p. 94". The following compounds are also described 2-phenyl-4-p-tolylmethylene-1 2 4-oxadi-imine (needles m. p. 150") ; the similar 4-p-chlorophenylrnethylene com- pound (needles m.p. 162") ; the 4-p-bromophenyt compound (needles m. p. 171") ; 4-p-chlorophenyl-2-p-tolylmethylene-oxadi-imine (needles m. p. 163") ; tJhe 4-p-bromophenyl derivative (m. p. 155") the 2 4-di- chloro-derivative (m. p. 171") and the 2 4-dibromo-derivative (m. p. 175"). p-Chlorophenylcarbylamilze forms needles (m. p. 71") the corresponding bromocurbylamine having m. p. 99". Both have an odour somewhat less powerful than that of phenylcarbylamine itself but the taste of the vapour is considerably stronger. 2-p- Chlorophe7iyl-4-p-bro.omophen ylmethylene-1 2 4-oxadi-imine forms needles m. p. 171'. Pher; yl-p-tolykrbodi-imide C,H,Me*N:C:NPh (b. (converted by aqueous-alcoholic hydrochloric aci into phenyl- p- tolylcarbamide) phenyl-p-c hlorophenykarbodi-imide (b.p. 175 I1 1 O mm.) p-bromodi.phenylcarbodi-imide (b. p. lS5/llo mm.) p-chloro- phenyll-p-tolylcarbodi-iwzide (b. p. 175'/10 mm.) p-bromphenyl- p-tolylcarbodi-imide (b. p. 188"/10 mm.) pp'-dichlorodiphenyl- carbodi-imide (b. p. 190"/10 mm.) pp'-dibromodiphenylcal.bodi- imide (b. p. 202"/10 mm.) and p-chloro-p'-bromodiphenylcarbodi- iemide (b. p. 198-200"/11 mm.) are like the phenyl-p-tolyl deriv- ative viscous liquids their constitutions being proved by conversion into the corresponding carbamides. p-Chlorophenyl-p-tdylcarbamide (needles m. p. 297-299' [decomp .I) p-bromophenyl -p-tolylmrbamide (needles m. p. 3 16-3 18' [decomp. I) and p-chloro-p'-bromodiphenylcarbamide (needles m. p. about 300" [decomp.]) were obtained (see ante) from the corre- sponding carbodi-imides.F. G. .P. The Additive Formation of Four-membered Rings. IV. The Influence of Temperature on the Tendency towards Self-addition of the Nitroso-group. C. K. INGOLD and H. A. PIBMTT ( J . Chem. SOC. 1924 125 168-176).-0wing to the chromophoric character of the nitroso-group the use of the colori- to be meter enables the reversible reaction 2 k 0 s followed in greater detail than is the case with the reactions previously studied (see preceding abstract). The influence of temperature on the equilibrium is determined for the case of nitroso- f;' 170P2" mm.) -N-? I U-N-i. 324 ABSTRACTS OF CHEMICAL PAF'ERS. mesitylene in benzene solution. Using these results the order of the reaction is calculated from certain equations and shown to be in accordance with the chemical evidence.Nitrosomesitylene the preparation of which (and of mesidine) from dimethylaniline is described has m. p. 122-123" (cf. Pechmann and Nold A. 1898 i 310). The heat absorbed during the dissociation of 1 g.-mol. of dimeric nitrosomesitylene is calculated to be about 13,700 g.-cal. In 3% solution in benzene dissociation occurs to the extent of about 18% ; in a 2% solution 21.6%. This corresponds with a molecular weight 245 agreeing fairly well with Bamberger's value 24.6 obtained (A. 1901 i 141) cryoscopically in benzene solution. Dioximes. XVII. G. PONZIO and B. ZANARDI-LAMBERTI (Gazzetta 1923 53 818-824).-1n some of their properties 5-hy*oxy-3-aryl-1 2 4-oxadiazoles (or hydroxyazoximes) F. G. P. ()<N==QAr - ()<N=yAr C(0H):N CO*NH ' which are formed by withdrawal of two atoms of hydrogen from the a-modifications of monoarylglyoximes differ considerably from the 3 5-diaryl-1 2 4-oxadiazoles (or azoximes) which result on elimination of a molecule of water from a s-diarylglyoxime.Thus the azoximes are either completely destroyed or slowly converted into nitriles by the action of reducing agents whereas the hydroxy- azoximes are very easily reduced to amidines NH,*CR:NH by means of hydriodic acid in presence of red phosphorus. Further the hydroxyazoximes exhibit acid properties the mean values of the ionisation constant at 25" being 0.879 x 0.022 x and 0 . 1 4 0 ~ for 5-hydroxy-3-phenyl- p-tolyl- and benzyl-1 2 4- oxadiazoles ; thus the p-tolyl compound is slightly weaker than carbonic acid and the other two compounds are stronger.Benzenylamidine obtained by reduction of 5-hydroxy-3-phenyl- 1 2 4-oxadiazole gives a picrate m. p. 235"; Dieckmann (A. 1892,705) gave m. p. 228". According to Pinner (" Die Imidoaether," 167)' 2 4 6-triphenyl-I 3 5-triazine is formed when benzenyl- amidine hydrochloride is heated for six hours at 120-140" with benzoyl chloride. This reaction occurs at the ordinary temperature when benzenylamidine carbonate is treated with 20% sodium hydroxide solution and benzoyl chloride. 5-Hydroxy-3-p-tolyl-1 2 4-oxadiazole prepared from p-tolenyl- amidoxime and ethyl chlorocarbonate (cf. Schubart A. 1890 47) is the product of the isomerisation of the oxide of p-tolyl cyanide oxime (cf. Avogadro this vol. i 294). By hydriodic acid and red phosphorus it is converted into p-tolenylamidinc which forms a picrade C6H4Me*C( :NH)*NH2,C,H,07N crystallising in yellow needles m.p. 224-2225' (decornp.) and a carbona-te crystallising in white laminze m. p. 127-128" (gas) ; the corresponding nitrate was described by Pinner (Zoc. cit. 184). 5-Hydroxy-3-p-tolyl-1 2 4- oxadiazole forms the following derivatives sodium salt white powder ; silver salt white crystals stable towards light ; metfiyl ether slender white needles.ORGANIC CHEMISTRY i. 325 O<N=-Q *CH,Ph 5-Hydroxy-3-benxyl-1 2 4-oxadiaxole 9 pre- C(0H):N pared by the action of sodium hydroxide on the ethyl pheizyl- ethenylamidoximecarboxylate obtained by the interaction of ethyl chlorocarbonate and phenylethenylamidoxime CH,Ph*C( :NOH)*NH crystallises in large lustrous lamina m.p. 115". By hydriodic acid and red phosphorus it is converted into phenylethenylamidine CH,Ph*C(:NH)*NH the picrate of which forms yellow needles m. p. 224" (decomp.). 5-Hydroxy-3-benzyl-1 2 4-oxadiazole forms a sodium salt white powder ; an insoluble silver salt white crystalline powder stable towards light and a methyl ether crystallising in needles m. p. 112-113". 5-Hydroxy-3-a-naphthenyl-1 2 4-oxadiazole has m. p. 196" and not 189" as stated by Richter (A. 1890 62). a-hTaphthenyZamidine CloH.,*C(XH)*NH crystallises in white laminae m. p. 154" and its pzcrate in yellow needles m. p. 223" (decomp.). 5-Hydroxy-3-p-naphthenyl-1 2 4-oxadiazole m. p. 219-220" (Richter loc. cit. gave 216") yields on reduction p-naphthenyl- amidine the picrate of which forms reddish-yellow needles m.p. 240" (decomp.). T. H. P. K. VON AUWERS W. BUSCHMANN and R. HEIDENREICH (Annalen 1924 435 277-321).-A study of the tetrahydroindazoles obtained by condensing hydrazine derivatives with hydroxymethylenecyclohexanone and 1 -methyl-3-hydroxy- methylenecyclohexan-2-one. In the preparation of the latter compound (cf. A. 1915 i SlS) acetone sodium (Freer A. 1894 i 65) and 1 -methylcycZohexan-2-01 are formed as by-products. 1 - Methyl-3- benzoxymethylenecyclohexan-2- one (needles m. p. 84-85") is hydrolysed more readily by acids than by alkalis. Semicarbazide reacts with hydroxymethylenecyclohexanone to give tetrahydroindazole-2-carbonamide (Wallach and Steindorff A 1904 i 104) the constitution of which is now definitely estab- lished by its formation by the hydrolysis of the semicurbaxone of benzoxymethylenec yclohexanone (needles /\/CH:N*NH*Co"H2 m.p. 193" if bath previously heated a t the hydroxymethylene ketone into the disemimrbaxone (I) which in hot dilute sulphuric acid solution affords the above carbonamide. The latter as shown by Wallach (Zoc. cit.) occurs in two forms although the latter author's m. p. observations are incorrect. The carbon- amide if obtained in the labile form (as in the fkst method given above) has m. p. 158" but unless the m. p. bath is previously heated to within a few degrees of this temperature the labile form passes before its m. p. is reached into the stable form (m. p. 186-188") (see infra). Wallach's second m. p. (220") actually indi- cated the decomposition of the carbonamide t o give cyanuric acid.The above disemicarbazone (I) had m. p. 193" when first prepared but after six months m. p. 203". The lower m. p. is that of the labile form which was not subsequently obtained. The substance Tetrahydroindazoles. vaN.NH.C0.NH2 I I'D 190"). Excess of semicarbazide converts (1.) m"i. 326 ABSTRACTS OF CHEMICAL PAJ?EBS. becomes solid after melting a t 203' re-melting with decomposition at 234-236". Methyl-hydroxymethylenecycbhexanone aff or& (with semicarb- azide in cold acetic acid solution) the normal semicurbaxone (white needles m. p. 193" or 203" according t o the rate of heating) the benxoyt derivative of the latter being identical with the semicurbazone (m. p. 186") of methyl-benzoxymethylenecyclohexanone. Excess of semicarbazide converts the methyl hydroxymethylene ketone into the disemicarbaxone (prisms m.p. 215" if heated suddenly and 237 O [decomp.]) together with a little 7-methyltetrahydro- indazoZe-:!-carbonamide the latter also being formed in small quan- tities in the preparation of the monosemicarbazone. The carbon- amide exists in two forms labile form monoclinic needles m. p. 138" (if plunged suddenly into bath just below 138") then solidifying t o give the stable form tetragonal crystals m. p. 163". The labile form can be kept at the ordinary temperature but passes into the stable form on heating in a toluene vapour-bath or on boiling in alcoholic solution. The labile form results when the semicarbazone is treated in cold acetic acid solution with a little sulphuric acid with subsequent addition of water whereas if the solution is heated before addition of water the latter precipitates the stable form.Again when the parent ketone is treated with semicarbazide in presence of mineral acid the carbonamide is formed the production of the labile or of the stable form depending on the hydrogen-ion concentration. The carbonamide is converted by ammoniacal silver nitrate solution into the silver salt of 7-methyltetrahydro- indazole this salt with carbamyl chloride affording a mixture of the two forms of the carbonamide. The.latter are shown to be stereoisomerides (see an- nexed formulze) and not / B structural isomerides. ,,'.NJN o"N- CO*xHz They are readily inter- convertible ; moreover under conditions when a 1-derivative might be expected t o result (as when the parent ketone is treated with semicarbazide hydro- chloride) the labile form already proved to be a 2-derivative is obtained.Polymorphism is also excluded as an explanation of the isomerism on experimental grounds. Tetrahydroindazole (m. p. 84") &st obtained by Wallach (Zoc. cit.) is prepared by boiling tetrahydroindazole-2-carbonamide (or the parent semicarbazones) with mineral acids and gives the following derivatives picrate m. p. 155-156" ; silver salt ; 2-met$ derivative b. p. 130"/15 mm. ; and 2-o-nitrobenxoyt derivative white needles m. p. 129-130". 7-d.17eth,yttetrahydroindazole is a viscous oil b. p. 152.8"/12 mm. becoming solid at the temperature of mixtures of ether and solid carbon dioxide.The following derivatives are described hydro- chloride white prisms ; picrate m. p. 151-152" ; mercuric chloride additive compound white needles ; silver nitrate additive compound Ag(C,H,zN2)2E0 white needles m. p. 132" ; eilver salt C,H,,N,Ag (two forms one soluble and one insoluble in ether); 2-phenyl- ~ C H ~O*LNH /\,pCH\ Me f i ' MeORGANIC CHEMISTRY i. 327 carbonamide white prisms m. p. 70" (from the indazole and phenyl- carbimide) ; and 2-o-nitrobenzoyl derivative leaflets m.p. 104-107°. The parent indazole is obtained either by treating methyl-hydr- oxymethylenecyclohexanone semicarbazone with hot dilute sulphuric acid or the ketone itself with hydrazine. Hydroxymethylenecyclhexanone when treated with phenyl- hydrazine gives a mixture of 1- and 2-phenyltetrahydroindazoles which are not interconvertible and are separable as the picrates.The constitution of the 2-phenyl derivative follows from its form- ation from the benzoylated ketone and phenylhydrazine. The intermediately formed phenyzhydrume of the benzoylated ketone has only a transitory existence (white needles m. p. about 80") and rapidly passes into the above 2-phenylindazole derivative. 1-PhenyZ-4 5 6 7-tetruhydroi~zdazole forms leaflets m. p. 58-59" b. p. 178"/10 mm. (picrate m. p. 125-126" and perchlorute are described). Z-PhenyZtetruhydroj~u~z~ forms prisms m. p. 486- 49*5" b. p. 177"/10 mm. (picrute m. p. 126.5-127*5' and perchlorute are described). In the interaction of hydroxymethylenecyclohexanone and phenylhydrazine the higher the concentration of hydrion the smaller the proportion of 2-phenyl derivative formed particularly if at the same time the temperature is kept low.Thus when the ketone is treated a t 0" with phenylhydrazine hydrochloride only the 1-phenyl derivative is formed. This is explained as being due to the driving back by hydrion of the dissociation of the hydroxy- methylene group with resulting increase in the proportion of the aldehydic form from which the 1-compound would be formed (cf. Claisen A. 1895 i 62). In this connexion the nature of the hydrazine derivative affects the proportion of 1- to 2-derivative obtained this being under comparable conditions as follows phenylhydrazine 1 0 m-phenylmethylhydrazine 2 1 benzyl- hydrazine 3 4 methylhydrazine 4 1 and semicarbazide 0 1.In order to throw light on the constitution of the intermediate compounds formed in the interaction of phenylhydrazine and hydroxymethylenecyclohexanone the interaction of the latter with phenylmethylhydrazine has been studied. The main product (m. p. 170-171") is insoluble in alkali and may be either thephenyl- methylhydrazone of 2-ketocyclohexyEformccldehyde (11) or l-methyl- phenyZ,'cydmzidomethenecyclohexcm-2-one (111). A second product soluble in alkali is the hydrate (needles m. p. 123-124') of the acid produced by the spontaneous oxidation of IV. There is also formed a third product (under slightly different conditions) viz. the hydrate of the phenylmethylhydrazone of l-forrnylcyclohexcsn- 2-one (needles m. p. 125-126"). (111.) i IV.)i.328 ABS!CRACTS OF CHEMICAL PAPERS. Methyl-hydroxymethylenecyclohexanone with phenylhydrazine affords a mixture of the 1- and 2-indazole derivatives low tem- peratures and high concentrations favouring the production of the 1-derivative. The latter is a stronger base than the 2-derivative and forms a stable perchlorate (m. p. 214-215") which allows of its separation from it's isomeride. The benzoyl derivative of the methylcydohexanone affords a phenylhydrazone (m. p. 125" with previous sintering) which rapidly passes in solution into 2-phenyl- 7-methyl-4 5 6 7-tetrahydroindoole (a colourless oil b. p. 181.4"/ 10 mm. The methiodide m. p. 172" gives 2-phenyltetrahydro- indazole when heated). 1 -Phenyl-7-methyltetrakydroindazole is a colourless oil b.p. 175"/10 mm.; its methiodide m. p. 154" on heating gives 1 -phenyltetrahydroindazole. The products of the alkylation of tetrahydroindazole depend on the nature of the alkyl group introduced and on the experimental conditions. Tetpahydroindazole when heated in alcoholic solution with benzyl chloride in presence of alkali or when heated alone with the chloride at water-bath temperatures affords a mixture of the 1- and 2-benzyl derivatives this mixture also being obtained by condensing benzylhydrazine with hydroxymethylenec yclohexanone. The benzoyl derivative of the latter ketone is converted by benzylhydrazine into 2-benzylindaxole (colourless oil b. p. 177"/ 10 mm.). 1-Benzyltetrahydroindazole is a colourless oil b. p. 172"/10 mm. (picrate m.p. 128-129" and perchlorate are described). Tetrahydroindazole when heated at 100" with ethyl bromide gives 2-ethyltetrahydroindazole (a colourless oil b. p. 11S0/15 mm. ; picrate has m. p. 150~5-151~5") whilst when similarly heated in presence of alkali it affords 1-ethyltetrahydroindazole (a colourless oil b. p. 111"/14 mm.; picrate has m. p. 103-104"). The methylation of tetrahydroindazole leads in presence or in absence of alkali to the formation of the 1- and 2-methyl deriv- atives which cannot be separated. 2-Methyltetrahydroindazole obtained from benzoxymethylenecyclohexanone is a colourless oil b. p. 93-94"/9 mm. or 103"/13 mm. (picrate m. p. 1664-167.5"). 7-Methyltetrahydroindazole on methylation ethylation or benzyl- ation in presence or in absence of alkali always affords 2-derivatives which may be obtained from the appropriate hydrazine and methyl- benzoxymethylenecyclohexanone. 2 7-Dimethyl-4 5 6 7-tetra- hydroindazole is a colourless oil with a feeble blue fluorescence has b.p. 111"/13 mm. and forms a picrate m. p. 123-125". 7-Methyl- 2-ethyZtetrahydroindazole a colourless oil b. p. 108-5"/10 mm. gives a picrate m. p. 101-102" and a methiodide m. p. 102-103" the latter affording 7-methyl-2-ethyltetrahydroindazole when heated. 7-Methyl-2-benzyltetrahydroindazole a colourless oil b. p. 181- 183"/12 mm. or 152"/3 mm. forms a methiodide (leaflets m. p. 175" [decomp.]) which is also obtained from 2 7-dimethyltetra- hydroindazole and benzyl iodide and on heating affords the two last-named substances. The fact that the 7-methyl group prevents alkylation in position 1 is to be ascribed to electrochemical (notORGANIC CHEhnSTXY.i. 329 stereochemical) effects (cf . attempts to prepare 1 5-dialkylpyrazoles by Auwers and Broche A. 1923 i 151). The formation of quaternary salts from the reduced indazoles is not altogether parallel to their formation in the ordinary series. The 2-methyl- and 2-ethyl-tetrahydroindazoles follow the general rule. Methyl iodide converts either 1 - or 2-methyltetrahydro- indazole into 1 2-dimethyl-4 5 6 7-tetrahyd~oinduxolium iodide (m. p. 152-153"). Ethyl iodide converts either 1- or 2-ethyl- tetzahydroindazole into 1 2-diethyltetrahydroindazolium iodide (m. p. 98-99'). Since these quaternary salts afford mixtures of 1- and 2-alkyl derivatives when heated they are given formuh V or VI or better VII W.) (VI.1 WI.) 2-Benzyltetrahydroindazole when heated with ethyl iodide affords the compound (m.p. 138-139") obtained by heating 2-ethyltetrahydroindazole with benzyl iodide. The compound on heating affords pure 2-ethyltetrahydroindazole and therefore is written as [C,H,*CH(N2Et[Ph*CH,*])(N1)]I both alkyl groups being attached to the same nitrogen atom. Similarly the compound [C6H,*CH(N2)(N1Et[Ph*CH,*])]I (m. p. 127-128') is formed from 1 -benzyltetrahydroindazole and ethyl iodide or from the 1-ethyl derivative and benzyl iodide. The 1 1-quaternary salt on heating affords not 1-ethyl- but 2-ethyl-tetrahydroindazole. 7-Methyltetrahydroindazole gives quaternary salts in which the two akyl groups are attached to the same nitrogen atom (see ante) The methiodides of 1 - and 2-phenyltetrahydroindazoles afford when heated the parent phenyl derivatives.7-Methyltetrahydroindazole reacts with ethyl bromoacetate to give ethyl 7-methyl-4 5 6 7-tetrahydroi?adazyl-2-acetate (colourless oil b. p. 166.5-167.5'/14 mm.; picrate m. p. 127-130" [indef.]). The corresponding acid (also obtainable from the indazole and chloroacetic acid) obtained by hydrolysis forms white needles m. p. 168-169" (silver salt described) and has unlike the indazole bases a sweet taste. E. E. T. uic.-2 N-Aryltriazolephthalonic Acids. G. CHARRIER (Gaxzetta 1923 53 829-843).-1t has been shown by Dimroth (A. 1902 i 403) that starting from the three isomeric N-phenyl derivatives of 1 2 3-triazolemonocarboxylic acid and eliminating the phenyl group by oxidation after nitrating and converting into the amino-derivative by reduction one and the same acid is obtained; the desmotropy of the triazole ring is thus establhhed.Oxidation of naphthalene by means of alkaline permanganate solution yields mainly phthalonic acid and the author finds similarly that the 2-arylnaphthatriazoles give under the samei. 330 ABSTRACTS OF CHEMICAL PAPEBS. conditions principally the corresponding vicinal 2-aryltriazole- phthalonic mids Since these acids are highly stable towards oxidising agents and are prone to form monometallic salts with the alkali metals prob- ably owing to the formation of an internal anhydride between t’he *CO*CO,H group and the triazole ring and further on account of the analogy existing between them and the product of the oxid- ation of naphthalic acid by permanganate in alkaline solution (cf.Graebe and Bossel A. 1893 i 593) formula (I) is regarded as preferable to formula (11). The constitution r y \ N A r of these acids is shown also by their oxidation CO,H()-N/ to the corresponding 2-aryltriazolephthalic acids (annexed formula) and by their abilityto react as keto-acids analogously to phthalonic acid with phenylhydrazine hydrochloride to form phenylhydrazones. The latter are far more stable than those of phthalonic acid since only by heating above their melting points are they transformed like phenylhydrazones of a keto-acid into derivatives of the corre- sponding ketotetrahydropyridazines (pyridazinones) ; a carboxyl group is eliminated a t the same time so that from a dibasic acid a neutral heterocyclic derivative is obtained CO,H\/ These acids behave as comparatively strong dibasic acids ; they decompose carbonates form well crystallised salts containing water of crystallisation and readily yield normal and acid esters and amid-.When subjected to prolonged energetic oxidation they give the corresponding phthalic acids. [With A. ODIFREDDI A. MANFREDI and E. MARs.]-vic.-2- Phenyltm’axolephtha-lonic acid C,,H,05N3 obtained by oxidising 2-phenylnaphthatriazole crystallises rn white or lustrous needles m. p. 242O and has an intense acid reaction towards litmus methyl orange and phenolphthalein. It yields the following derivatives monosodium salt C,,H,05N,Na,4H,0 m.p. 324-325” gives aqueous solutions acid to litmus and phenolphthalein. The normalORGANIC CHEMISTRY. i. 331 barium salt crystallises with 4H20 and also with 5H,O and when subjected t o dry distillationin the anhydrous Nph/y-\ state yields vic .-2-phenyltriaxokphthalic an- \ /\co hydride (annexed formula) which crystallises Ico>O in slender white needles m. p. 183-154" and when treated with m-diethylamino- phenol and zinc chloride gives a deep red crystalline colouring matter ; the hydrochloride of the latter dissolves in boiling alcohol to an intense reddish-violet markedly fluorescent solution and cryst al- lises in slender metallic green needles similar to the crystals of rhodamine base B hydrochloride. Basic barium 2-phenyltriazole- phthalonate [PhN3:C,H,<C02- cO.cO,>,a] ,Ba( OH),,12H20 forms rosettes of slender silky needles and decomposes before melting.The strontium salt Cl5H1,O5N3Sr74~H2O forms lustrous white needles and decomposes before melting. The methyl ester Cl,H,0N,(C02Me) forms minute white prisms in. p. 87-S8" and the methyl hydrogen ester C,,H1105N3 colourless prismatic crystals m. p. 201-202". The ethyl ester crystallises in colourless prisms m. p. 169-170". The diamide C1,H,ON3(CO*NH,) forms lustrous colourless prisms m. p. 233-234" and when boiled with 10% ammonia solution yields the ammonium salt of the mononmide C15H1304N5 which crystallises in slender white needles m. p. 230" (decomp.). The phenylhydruxone probably Ph*N3X6H,(C0,H)*C( CO,H):N*NHPh crystnllises in white silky needles m.p. 206-207" (decomp.) and acts as an energetic acid. The monocarboxylic ketotetrahydropyridazinic acid that is the pyridazinone of vic. -2 -phenyltriazolephthalonic acid is not formed when the phenylhydrazoiie is boiled in acetic acid bEt when heated at 250" the phenylhydrazone T\TPh/y-\ loses carbon dioxide and water to \~-/)-cH:P give the decarboxylated pyridazimize (,-,TO .Nph (annexed formula) which crystahes in colourless prisms m. p. about 200". When the phenylhydrazone is treated in boiling water with barium carbonate phenylhydrazine separates and the pentahydrated normal barium 2-phenyltriaxolephthalonute C15H,05N,Ba,5H,0 and the dodecahydrated basic barium salt are formed. The great resistance offered by vic.-2-phenyltriazolephthalonic acid to oxidising agents may be due to the existence of an an- hydridic linking between the triazole nucleus and the COC0,H group which is thus protected from the action of nascent oxygen.The acid is however attacked by chromic acid with formation of a compound which crystallises in aggregates of radiating leaflets m. p. 250" acts as an energetic acid and has the nitrogen content of a zic. -2-phenyltriazolephthalic acid. ICT-/ \/ 3 C,.5H,,o 1 SNgBa3 ,Ba( OH )2 7 2H20,i. 332 ABSTRACTS OF CHEMICAL PAPERS. ric .-2-p-Chlorophen yltriazolephthalonic acid crystallises in slender white needles m. p. 264-265"; its mono- sodium salt (+2H20) does not melt a t 360". 2-p-Bromophenylnaphthatriaxole C10H6.N3-C6H4Br prepared by oxidising p-bromophenylazo- P-naphthylamine by means of chromic acid in acetic acid solution crystallises in white silky needles m.p. 200". vic .-2-p-BromophenyEtriaxol~phthalonic acid obtained by the action of alkaline permanganate solution on the preceding com- pound forms flat white needles m. p. 270"; its monosodium salt (+2H,O) does not melt a t 320". vic. -2-p-Carbox yphen yltriaxolephtha lo?& acid C16H,0 ,N3 crystal- lises in slender white needles m. p. 287-288" (decomp.) and is an energetic tribasic acid; its monosodium salt (+44H,O) does not melt a t 350". C,H,CbN,:C 6H2( CO,H)*CO*CO,H T. H. P. ap-Naphthatriazoles. G. CHARRIER and M. GALLOTTI (Gaxxetta 1923 53 851-861).-The authors have prepared 2-p- chloroacetylphenyl-or P-naphthatriazole by means of the following series of reactions Ph-NHAc +- CH,CI*CO-/~NHAc + CH,Cl*CO/-)NH ++ CH2C1*CO/-'N2Cl-+ \-/ CloH,-NH -+ \-/ L.NH / \ \-/ \/ The mobile chlorine atom of the chloroacetyl group is readily replaced giving t'he corresponding iodo- and cyano-derivatives. Hydrolysis of the latter gives the unstable ketonic acid which loses carbon dioxide to form the corresponding ketone Z-p-acetyl- phenyl-ap-naphthatriazole. Treatment of the chloroacetyl com- pound with silver acetate yields the acetyl derivative of the corresponding ketocarbinol a~-naphthatriazole-2-p-benzoylcarbinol which may be obtained either by hydrolysis of the acetyl deriv- ative or by the action of moist silver oxide on the chloroacetyl compound itself. The latter reacts normally with ammonia yielding a~-naphthatriazole-2-p-benzoylmethylamine whilst with primary aliphatic amines both substitution and addition occur t,he unstable compound thus formed losing water and alkylamine when treated with an acid or even with a solvent the resulting compound consists of the corresponding 2-p- p-alkyl- amino-or-alkyliminoethylphenyl-o:p-naphthatriazole. Aniline acts similarly to ammonia the chlorine atom of t.he p-chloroacetyl group being replaced by the anilino residue.ORGAN10 UHEMISTRY.i. 333 [With R. SALA Q. MINGOIA and P. To~~zz~.]-p-Chloroacetyl- benzeneazo-p-naphthylamine NH2*CloH6*N:N*CGH4*CO*CH2Cl crys- tallises in thin lustrous garnet-red leaflets m. p. 172" and dis- solves in concentrated sulphuric acid t o a magenta-red solution. Its phenylhydraxone forms a red crystalline powder with metallic green reflex m.p. 180" (decomp.). 2 -p-ChloroacetyZphenyl-u~-napht~triaz~le prepared by oxidising the preceding compound by means of chromic acid forms thin white leaflets m. p. 19So and dissolves in con- centrated sulphuric acid to a yellow solution. Its phenylhydraxone crystallises in thin pale yellow leaflets m. p. 156". 2 -p-lodoacet ylphen yl- u p-naphthatriaxole separates in lustrous white needles m. p. 195". 2 -p- Cyanoctcetylphen yl- u p -naphthatriaxole C19H12ON4 crystallises in slender yellow needles m. p. 185" with prevlous contraction and on protracted boiling with 50% sulphuric acid gives 2-p-acetyl- phenyl-ap-naphthatriazole (cf. A. 1922 i 771). ap-~Ta~hfhairiaxole-2-p- be?axoylcarbinol C,,H 6 < ~ > ~ * C 6 H 4 0 ~ ~ * C H 2 ~ ~ ~ crystallises in silvery leaves m.p. 191" and its acetyl derivative in slender white needles m. p. 187". a P-Na~h f~ntriazole-2-p- benxoylmethylami?ze c' 6<~>~c6H4'co*cH2'NH2 forms minute almost colourless crystals m. p. 192" and gives a pale yellow hydrochloride m. p. 184". 2 - p - p -Methylamino -a- methyliminoethylphenyl - ap- naphthatriaxole c loH 6< $>N c sH4' C ( :We) *CH,*NHMe crys t allises in yellow needles m. p. 160" and forms a yellowish-brown hydrochloride. 2 - p - p - Ethylamino - a - ethyliminoethylphenyl - ap - naphthatriamle C,,H,N separates in slender pale yellow needles m. p. about 150" mth previous contraction up-hTaphthatriaxole-2-p- benxoylmethylaniline C,,H ,<~>N'C6H4'CO*~,'~Ph crystallises in lustrous sulphur-yellow leaflets m. p.277" and forms benzene solutions exhibiting pronounced greenish-yellow fluores- cence ; its phenylhydrazone separates in slender lemon-yellow needles m. p. about 270" (decomp.). T. H. P. Oxidation of Uric Acid by Iodine in Alkaline Solution. J. MORE (Compt. rend. 1924 178 49&-501).-Oxidation of uric acid (1 mol.) by iodine (2 atoms) in potassium hydrogen carbonate solution affords a substance (not isolat,ed) which is converted by acetic acid into allantoin and (cf. Biltz and Max A. 1921 i 893) is further oxidisable by iodine in acid solution. Oxidationi. 334 ABSTRACTS OF CHEMICAL PAPERS. of uric acid by means of iodine in sodium hydroxide solution (cf. Kreidl A. 1893 ii 558) gives first a substance which affords a uroxanate on evaporation in alkaline solution or allantoin on evaporating in acetic acid solution and in presence of alkali may also pass into a less readily oxidised isomeride.The latter how- ever yields the same products as above on evaporation in alkaline or acid solution and on oxidation with iodine gives allantoxalic arnide. The latter when treated with concentrated potassium hydroxide affords potassium allantoxalate ; the following salts pPhenylene-1 2-dinaphthaditriazole. G. CHARRIER (Gazzetta 1923 53 S62-867).-This compound has been prepared by means of the following series of reactions C,H,*NHAc -+ NO,*C,H,*NHAc -f NH,*C,H,*NHAc -+ ~ ~ I * C H ~ N H ~ c + ~ l o ~ * ~ ~ 2 -+ NH / \ / \ L/ \/ It is a highly stable compound and as is usual with ap-naphtha- triazoles is oxidised by permanganate in alkaline solution giving &.-p-phenyleneditriazolediphthalonic acid. Derivatives of the p-phenylene- 1 2-dinaphthaditriazole have not previously been prepared (cf.Morgan and Chazan A. 1922 i 181). It is found possible to pass directly by oxidation by means of alkaline per- manganate solution from 1 2-naphthatriazole-2-p-benzeneazo- p-naphthylamine to the above diphthalonic acid preparation of the corresponding naphthatriazole as an intermediate product being unnecessary. Similarly vic.-2-phenyltriazolephthalonic acid may be prepared directly from 1 -benzeneazo-2-naphthylamine.ORGANIC CHEMISTRY. i. 335 [With G. BONOMI and T. B~~~rxnzz~.]-p-~cety~aminobenzeiie- azo- p-naphthylamine C,,H,,ON prepared by the interaction of p-acetylaminophenyldiazonmm chloride and p-naphthylamine in alcoholic solution cryst allises in slender orange-red needles m.p. 208". 2-p-Acetylaminophengl-1 %naphthatrinzole CI8Hl4.ON4 obtained by oxidising the preceding compound in acetic acid solution by means of chromic acid forms slender white needles m. p. 228" and exhibits marked blue fluorescence even in T-ery dilute solutions. 2-p-Aminophenyl- 1 2-naphthatriaxole C16H12N4 crj-stallises in lustrous silky white needles m. p. 200". 1 2-Xaphthatriazole- p- benzen eaxo- p-naphthyhinine C',,H 18x6 separates in bright red nedes m. p. 250". p-Phenylene-1 2-dinnphthaditriazole C,,H,,N forms thin pink laminze m. p. 333". vic.-p-Pheny~cizeditriaxol~i~~~~~alonic acid (annexed formula) co.co2H crystallises in slen- der white needles /\C02H m.p. 322-323" with slight brown- ing a t 320". COT 0,H co2H''-+N_/-\-N<y - I IJ-" \-/ N-\/ 2-PhenyltriazoZephthulonic acid Ph*N3:C6H2(C02H)*Co*co2H [N3Ph C02H CO*C02H=12 4 31 prepared by oxidising benzeneazo- p-naphthylamine by means of alkaline permanganate solution crystallises in lustrous white needles in. p . 242". s-Trinitroarylazodiarylamines. 9. BERETTA (Gaxetta 1923 53 S70-876).-By the action of picryl chloride on amino- azo-derivatives in alcoholic solution in presence of sodium acetate the author has prepared the s-trinitrophenyl derivatives of various 0- m- and p-aminoazo-compounds of the benzene and naphthdene series. (N0,),C,H,*NK*C6H,oN:NPh obtained from m-aminoazobenzene crystallises in slender yellow needles m. p. 212" and dissolves in concentrated sulphuric acid to a yellow solution.4'-Benzeneazo-2 4 6-trinitrodiphenylamine (cf. Walther and Lehmann A. 1904 i 352) has m. p. li8O. 3' -Net h yl-6' - p- tolu en ea.20- 2 4 6- trinitrodiphn ylumine prepared from p-tolueneazo-p-toluidine forms slender slightly orange red crystals m. p. ZOl" and gives a red solution in concentrated sulphuric acid. 2'-~~ethyZ-4'-0-tolueneaxo-2 4 6-tri- nitrodiphenylamine crystallises in red needles m. p. 194" and dis- solves in concentrated sulphuric acid to a red solution. 3'-Arni?zo- 4'-benzeneazo-2 4 6-trinitrodiphenylamine C18H1306N prepared from 2 4-diaminoazobenzene (chrysoidine) forms minute red crystals m. p. 254" (decornp.) and gives a red solution in concentrated sulphuric acid. 3'-Amino-6'-~nethyl-4'-benseneaxo- LT.13. P. 3'-Benxeneaxo-2 4 6-trinitrodiphenylamine C6H4Me*N:N-C,H3Me*NH*C6H2( NO,) ,i. 336 ABSTRACTS OF CHEMICAL PAPERS. 2 4 6-trinitrodiphenylcine C,,H1,O6N7 prepared from 2 4-di- aminoazomethylbenzene (methylchrysoidine) forms slender lustrous prisms with metallic reflexion m. p. 228" (decomp.) and dissolves in concentrated sulphuric acid to a red solution. 3'- Acetyhmino-6'-methyl-4'-benzeneazo-2 4 6-trinitrodiphenylamine crystallises in slender orange-yellow prisms m. p. 215" (decomp.). 3'- Amino- 6'- methyl - 4'-p-chlorobenzeneazo -2 4 6 - trinitrodiphenyl- amine Cl,H,40,N7C1 prepared from 4'-chloro-2 4-diamino- 5-methylazobenzene crystallises in slender metallic brown prisms m. p. 235" (decomp.) and forms a red solution in concen- trated sulphuric acid.The corresponding 3'-acetylamino-compound separates in slender orange crystals m. p. 228" (decomp.). 3'-Amino- 6'-methyl-4'-p-bromobenxeneazo-2 4 6-trinitrodiphenylamine crystallises in slender lustrous brown needles m. p. 233" (decomp.) and dissolves in concentrated sulphuric acid to a red solution. The corresponding acetylamino-derivative forms slender orange needles m. p. 227" (decomp.). Ph*N2*CloH,*NH*C,H,(N02)3 prepared from benzeneazo- p-naphthylamme crystallises in slender reddish-brown needles m. p. 221" (decomp.) and dissolves in con- centrated sulphuric acid to a blue solution. ~-Benxeneuxo-cc,- naphthyl-2 4 6-trinitrophenylamine forms slender reddish-brown needles m. p. 193" (decomp.) and yields a blue solution in concen- trated sukphuric acid.R. N. SEX and B. SETT ( J . Amer. Chem. SOC. 1924 46 lll-l19).-Three types of azo-triphenylmethane dyes have been studied containing respectively one two and three azo-groups in the p-position to the central carbon atom. It has been found that the introduction of one azo-group deepens the colour (compare T. 1912,101 1113) of two azo-groups lightens the colour and of three makes it still lighter. The effect of the multiplication of chromophores has also been studied by introducing simultaneously an azo-methine group and an azo-group into chrysoidine rosaniline safranine and Congo-red and in each case the colour becomes lighter. The direct-dyeing property of Congo-red also disappears. The introduction of the azo-group has been found to deepen the colour in the case of an indamine dye as well as in the case of fluorescein.The preparation and properties of the following dyes which appear to be new substances are described phenetoleazosulpho- phenyltetramethyldiaminodiphenylcurbinol (from phenetoleazobenz- nldehydesulphonic acid and dimethylaniline) dyes wool green ; dimet h y lamino benzeneaxop h enyltet ramethy Id iaminod iphen ylcurb in01 (from benzeneazodimethylaniline and tetramethyldiaminobenz- hydrol) dyes wool blue ; salicylylazo~henylcEisalicylylcarbino1 (from meth ylenedisalic ylic acid and benzeneazosalicylic acid) dyes wool red ; phenyldi(dimethyluminobenzeneazopheny1)carbinol (from benz- C19H1406N7Br a- Benxeneazo- fi -nnphthyl-2 4 6-trinitrophen ylamine T. H. P. Studies in Dyes with Multiple Chromophores.ORG ANlC CHEMISTRY.i. 337 aldehyde and benzeneazodimethylaniline) dyes wool violet ; phenet- oleazosulphophenyldi(dirneth ylarninobenxeneaxophenyl)mrbinol (from phenetoleazobenzaldehydesulphonic acid and benzeneazodimethyl- aniline) dyes wool yellow ; diphenetoleuxosulphobenxylidenechrysoidine (from chrysoidine hydrochloride and phenetoleazobenzaldehyde- sulphonic acid) dyes wool golden-yellow ; diphenetoleaxosulpho- benxylidene-rosaniline hydrochloride (from rosaniline and phenetole- azobenzaldehydesulphonic acid) dyes wool bluish-red ; diphenetole- axosulphobenxylidene-safranine hydrochloride (from safranine and phenetoleazobenzaldehydesulphonic acid) dyes wool red ; dimethyl- arninobenxeneuxophenyldimethylquinoneimine hydrochloride (from nitrosodimethylaniline hydrochloride and benzeneazodimethyl- aniline) dyes wool olive-green ; sulphobenxeneuxoJlurescein (from diazotised sulphanilic acid and fluorescein) dyes wool orange.Amine Oxidation. VIII. Oxidation of Aliphatic Amines. S. GOLDSCHMIDT and V. VOETH (Annalen 1924 435 265-277).- Benzylamine is oxidised by potassium permanganate in aqueous or in acetone solution to give benzaldehyde no intermediate com- pound being isolable even if oxidation be effected a t -60". Since however benzaldehyde under similar conditions is rapidly oxidised to benzoic acid it cannot be the primary product in the oxidation of benzylamine which probably proceeds thus PhCH,*NH + PhCH,*N -+ PhCH:NH + PhCH:O+NH,. Dibeiizylamine is oxidised by permanganate in acetone solution to give benxoyltribenxylhydraxine (white needles m.p. 181") the constitution of which was established by synthesis (from tribenzoylhydrazine and benzoyl bromide) ; its formation is explained by the scheme 2(PhCH,),NH -3 2(PhCH,),Ni -t (PhCH,),N*N( CH,Ph) -+ (PhCH,),N-NBz*CH,Ph. Confirm- ation of this explanation was obtained tetrabenzylhydrazine on oxidation with permanganate gives benzoyltribenzylhydrazine which on reduction with zinc dust and acetic acid yields benzyl- benzamide (and traces of benzaldehyde and dibenzylamine) and on treatment with mineral acids gives (per mol.) benzaldehyde (I mol.) and hydroxytribenxylhydraxine hydrochloride. The latter (m. p. 220-240" [indef.]) could not be obtained in a pure state and was found to contain dibenzylamine hydrochloride whilst on treating it with alkali benzamide and tribenzylhydrazine were isolated.The oxidation of di-n-propylamine by means of permanganate gives a p dipropion yl- o! p -di -n-prop ylh ydraxine C OE t*NPra*NPrCI*C OE t . This substance evideiitly formed by the oxidation of the first- formed tetrapropylhydrazine was only obtained in a slightly impure state as a yellow oil possessing an odour like that of menthol and beginning to decompose at 40" (more readily in presence of alkali) with formation of propaldehyde. The dipropionyl deriv- ative on being distilled affords a mixture of propaldehyde n-propyl- propionamide NHPr.*COEt (pale yellow oil b. p. 64-65"/i. 335 ABSTRACTS OF CHEMICAL PAPERS. 0.4 mm.) and n-propyZidenepropionamide CHEt:N*COEt (viscous greenish-yellow oil b.p. 87"/0.4 mm.). Aqueous permanganate with piperidine gave no isolable products but in acetone solution oxidation proceeded much more rapidly than in the cases described already t o give N-pi(perid.yZ-N-Z-p~(per- idone as a yellowish-red oil which on distillation was converted into a mixtke of tetrahydropyridine and 2-piperidone (b. p. 64- 65" ,'04 mm .) . E. E. T. Certain Nitroarylhydrazines. M. GIUA (Gazzettcc 1923 53 844-4351) .-The results obtained when various aromatic nitro- compounds are treated with semicarbazide show that the latter serves to replace by the residue -NH*NH*CO*NH not only the nitro-group but also labile halogen atoms. Thus when heated on a water-bath in alcoholic solution containing sodium acetate 1 -chloro-2 4-dinitrobenzene and semicarbazide react in accord- ance with the equation C,H,C1(NO,),+2NH2*NH*CO*NH2 = C6H3(N02)2.NH*NH.C0."H,.CO*~H*~H2,HCl.I n the case of nitro-compounds containing a labile nitro-group the formation of nitrous acid renders the reaction somewhat complicated and although the yields of nitrophenylsernicarbazides are good various secondary products are also formed. With p- and y-trinitrotoluenes the first phase of the reaction is probably expressed by the equation C,H,Me(N0,),+2NH,*NHCO*NH2 = C6H,Me(N0,),~NH~~~C0."H,.C0.N,S-2H,0 ; the azide of carbamic acid has not been observed among the secondary products but small proportions of cyanuric acid which is one of the many transformation products of this azide have been detected (cf Curtius and Schmidt A.1923 i 1080). 4 6-Dinitro-m-toEylsemicarbazide obtained by the action of semicarbazide on y-trinitrotoluene forms crystals m. p. 237- 235" (decomp.). 2 4-Dinitro-m-toEyZsemicarbazide similarly ob- tained from @-trinitrotoluene forms crystals m. p. 185" (decomp.) and in alcoholic solution yields a reddish-brown coloration with alkalis. 5-Chloro-2 4-dinitrophenylsemicarbaxide prepared from 1-chloro-2 4 5-trinitrobenzene forms yellow crystals m. p. 2 1 7-2 1 8 ' (dec omp . ) . 2 4 - Dinit rop hen y Zs emicarbazid e C7H 0 5N 5 prepared from 1 -chloro-2 $-dinitrobenzene forms pale yellow crystals m. p. 205-207" (decomp.). CI(NO,),C ,H,*NH*NiMePh obtained by the interaction of as-methylphenylhydrazine and 1-chloro-3 4 6-trinitrobenzene crystallises in orange-yellow lamellz m.p. 120-121" and dissolves in concentrated sulphuric acid to a reddish-brown solution. Acetone 4 6-dinitro-m-toZylhydrazone C,,H1,O4N4 crystallises in yellow needles m. p. 144-145" and in alcoholic solution gives a dark red coloration with an alkali. Other 4 6-dinitro-m-tolyl- hydrazones are those of acetophenone C,,H140,N4 crystaking in orange-yellow needles m. p. 215-216"; ethyl phenyl ketone C16H1604N4 forming a red crystalline powder m. p. 183-184'; u-JIethyl-sc-phenyl-5-chloro-2 4-dinitrophenylhydrazine,ORGANIC CHEMISTRY. j. 339 benzaldehyde CI4H,,O4N4 crystallising in reddish-yellow needles m. p. 224-225" ; anisaZdehyde separating in garnet-red prisms m. p. 238-239" (decomp.) ; p-tolzcaldehyde C15H1404N4 forming lustrous deep red needles m. p.237-238"; cinnamaldehyde C,,H,O,N forming bright red needles m. p. 263-264" (decomp.). T. H. P. Additive Reactions of the Phosphorus Halides. VIII. Kinetic Evidence in Regard to the Mechanism of the Reaction. J. B. COXAXT and V. H. WALLINGFORD ( J . Amer. Chern. SOC. 1924 46 192-202; cf. A. 1923 i 498).-The reaction between benzaldehyde and phosphorus trichloride proceeds t o a definite equilibrium both in the absence of a solvent and in benzene solution. The composition of the equilibrium mixture accords with the law of mass action. An earlier explanation of the mechanism of the reaction between carbonyl compounds phosphorus trichloride and acetic acid or anhydride assumed that this addition reaction was an intermediate step; this explanation is now abandoned.The gradual addition of small amounts of water to a mixture of benz- aldehyde and the trichloride causes an irreversible reaction to take place. The product appears to be an open-chain phosphorus compound which readily loses hydrogen chloride to form a phostonic acid derivative (cf. Conant and Macdonald A. 1921 i 69). The reaction between a trichloride an aldehyde and acetic acid must be either a direct interaction of three molecules or a reaction in which some open-chain additive product of an unstable nature is an intermediate J the authors favour the latter alternative. F. A. M. Action of Halogen Derivatives of Phosphorus on some Phenolic Bases. ill. BOURNEUF (Bull. Xoc. chim. 1923 33 [iv] 1808-1823).-When dimethylaniline (2 mols.) is heated a t 70" with phosphorus trichloride (1 mol.) p-dimethylaminophenyl- dichlorophosphine NMe,*C,H:,*PC~ is primarily formed shce sodium carbonate converts the product into a mixture (with dimethylaniline) of tetramethyldiarninodiphenylphosphine oxide (NMe,*C,H,),PHO m.p. 169" (insoluble) and dimethylamino- phenylphosphinic acid (soluble and stable towards boiling water ; cf. Michaelis and Schenk A. 1888 834). The above oxide in presence of water and air gives tetramethyldiaminodiphenylphos- phinic acid (m. p. 249-250" on Maquenne block or 199" [decomp.] in capillary tube) the latter being formed together with hem- methyltriarninotriphenylphosphine dihydroxide (NMe,*C6H4)3P(O13[)2 (m. p. 321" on Maquenne block) in the original condensation. Diethylaniline and phosphorus trichloride give non-crystalline products .Dimethylaniline (4 mols.) when heated with 1 mol. of phosphorus oxychloride at 130" the product then being treated with sodium hydroxide gives a mixture of (1) tetramethyldiaminodiphenyl- phosphinic acid affording dimethylaniline when heated with calcium carbonate or with mineral acids; sodium salt 1 (NMe,-C,H4),P( ONa) 0 ,H,O ;i. 340 ABSTRACTS OF CHEMICAL PAPERS. (2) tetramethyldiaminodiphenylmethane and (3) hexamethyltri- amhotriphen ylphosphine oxide m. p. 262" and the corresponding hydroxide (cf . supra). Diethylaniline is converted by phosphorus oxychloride into tetraethyldiaminodiphenylphosphinic acid (m. p. 195" [decomp.] or 253-254" on Maquenne block) which is decomposed on heating with calcium carbonate or with mineral acids to give diethylaniline ; it forms a dihydrochloride (hygroscopic m.p. 185" on Maquenne block) a sodium salt (crystallising with 6 or 7H,O) and a barium salt. Orthophosphoric acid and aniline give only aniline pyrophosphate. T i l W T u. u. 1. Studies Concerning the Direct Preparation of Organo- beryllium Halides. H. GILMAN ( J . Amer. Chem. Xoc. 1923 45 2693-2695) .-Metallic beryllium even when freshly cleaned by treatment with very dilute pure nitric acid does not react with methyl iodide ethyl iodide benzyl bromide bromobenzene chloro- benzene or a-bromonaphthalene using as solvents anhydrous ether anisole benzene or p-ethoxynaphthalene and as catalysts iodine bromine dimethylaniline ethyl acetate or freshly prepared Grignard reagent.Experiments in which freshly sublimed beryllium iodide was used and one in which the reaction mixture was exposed to X-rays were also unsuccessful. W. S. N. Action of Polyhalogenated Derivatives of Methane and Ethane on Magnesyl [Magnesium Alkyl] Compounds. 111. R. BINAGHI (Guzzettu 1923 53 879-887; cf. A. 1922 i 313 1002) .-Towards magnesium ethyl bromide tetrachloromethane behaves similarly t o chloroform giving only ethylene and methane. With magnesium phenyl bromide however it yields not tetra- phenylmethane but hexaphenylethane triphenylmethyl peroxide and triphenylcarbinol. This reaction exhibits marked analogy to the formation of hexaphenylethane by the action of zinc on triphenyl- chloromethane (cf. Gomberg A. 1901 i 77 638; 1913 i 259) ehich is yielded by the first phase of the reaction 3MgPhBr+CCI,= CPh,Cl+SMgBrCl. The latter compound then reacts with another molecule of magnesium phenyl bromide giving hexaphenylethane and bromobenzene which also occurs among the products of the reaction 2CPh3C1+MgPhBr =MgCI,+PhBr+CPh,=CPh That the reaction proceeds thus is confirmed by the simultaneous formation of triphenylcarbinol which under the experimental conditions employed could not arise except by the hydrolysis of the triphenyl- chloromethane either by the water added to precipitate the basio magnesium salt or by the current of steam used to remove the bromobenzene and diphenyl.Atmospheric oxidation of hexa- phenylethane explains the formation of triphenylmethyl peroxide which is not obtained when an atmosphere of hydrogen is employed.In the reaction between magnesium ethyl bromide and tetra- chloromethane the first phase consists of the addition of four molecules of the magnesium alkyl compound to one molecule of the methane derivative evolution of gases occurring only whenORGANI(3 CHEMISTRY. i. 341 the product of the reaction is either left for some time or heated CCll-!-4MgEtBr=4MgClBr+4C,H,+ CH4. Traphenylmethyl peroxide O:O(CPh3) forms a lustrous white microcrystalline powder m. p. 186" and yields rather low values for the molecular weight in freezing naphthalene. By dissolution in cold concentrated sulphuric acid and subsequent addition of the solution to ice triphenylcarbinol is obtained. Its hexanitro- derivative O:O[C(c,H,*NO,),] has m. p. 208-210". T.H. P. C. W. RODEWALD and R. AD- (J. Amer. Chem. Soc. 1923 45 3102-3105).-Amino-arylarsinic acids on condensation with p-chloroethyl and y-chloropropyl chloro- f ormates yield o-chloroalkyl- (amino-ary1)carbarnates. On treat- ment with two molecular equivalents of aqueous alkali the p-chloro- ethyl compounds are converted into amino-aryl oxazolidones and the 7-chloropropyl compounds into 3-amino-aryl-tetrahydro-1 3 2- oxazones. Excess of alkali hydrolyses the oxazolidones to p-arsino- aryl amino-ethanols and the oxazones to y-arsino-aryl-aminopro- panols. C6H4(AsH,03)NH*C02*CH,*CH,CI has m. p. above 250" the corresponding (o-arsanophenyl)curbamate having m. p. 156-157". p-Chloroethyl (2-methyl-5-arsinoplphenyl)carb- amute has m. p. 193-195". The y-chloropopyl (~-ccrsinophenyl)carb- amate C,H4(AsH,0,)NH-C02*[CH,],*C1 and the corresponding o-ar- sinophenykarbamate melt at 245-246" and 130-132" respectively.y -Chloropropyl (2 -met hy 1-5 -arsinophen y1)carbamate has m . p . 160- 162 ". 3 -p- Arsi nophenyl-2 -o~~zolidone,C~H,( hH203) *N*CO,*CH,*CH 3-0-Arsinophenyl-2-oxazolidone crystallises in colourless plates m. p. 212-213". 3-p-Arsinophenyl-1 3 2-oxaxone crystallises in white plates m. p. 245-247". p-(p-Arsinophenyl)- aminoethanol C,H,(AsH,O,)NH*CH,*CH,OH has m. p. 173-174" ; the corresponding o-arsanophenyl derivative has m. p. 1&146". p-(2-MethyZ-5-arsinophenyl)aminoethanol has m. p. 146146". y-(p-Arsinophenyl)amino~o~nol has m. p. 167-168" ; the similar o-arsinophenyl compound has m. p. 84-85". y-(2-MethyZ-5-arsino- pheny1)aminopropanoZ melts at 142-143'.p-Arsinophenyl- p-hydr- oxyethyl nitrosoarnine C,H,(AsH,O,)N(NO) -CH,*CH,OH is obtained from t i e corresponding ammoethanol and nitrous acid. It crystal- lises in bright yellow needles m. p. 236". P-Arsino~henyl-y-hydro~~ propyl nitrosoamine C,H,(AsH,03)N(NO)CH2*CH,-CH,0H is pre- pared similarly. It forms yellow crystals m. p. 142-143'. The arsino-aryl-amino-alcohols are much less toxic than the correspond- ing arsanilic acids. ' F. G. P. Arsino-arylamino-alcohols. p- Chloroet hyl (p-arsinophen yZ)carbamte forms colourless plates m. p. above 280". u C 6H4( AsH20,)*N*C02*CH2*CH,*CH L J The Protein Matter of Bile. J. F. LOGAN ( J . Biol. Chem. 1923 58 17-32).-Attempts have been made by various methods to isolate a pure protein from ox bile and elementary analyses have been made of several of the preparations obtained.Thei. 342 ABS'J!RACl'S OF C'EEMIOAL PAPERS. results however indiwte that in each case the product W ~ S probably a mixture of glucoprotein with a small amount of nuch- protein. Direct acid hydrolysis did not liberate a reducing carbo- hydrate but when this followed preliminary trwtment with alkali Proteins of Wheat Bran. I. Isolation and Elementary Analyses of a Globulin Albumin and Prolamine. D. B. JONES and C. E. F. GERSDORFF ( J . Biol. Chem. 1923 58 117-131).- What bran as free as possible from other portions of the wheat kernel has been found to contain 17*25y0 of protein (Nx6.25). By extracting the bran successively with water 4% sodium chloride 70% alcohol and 0.5% sodium hydroxide 86.61% of the total protein has been removed and an albumin a globulin and an alcohol-soluble protein (prolamine) have been isolated from the &st three extracts. Elementary analyses gave the following results albumin C=53.21 H=6.71 N=15*42 S=1-35y0 ; globulin C=53-43 H=7-40 N= 17.76 S=O-91 yo ; prolamine Theory of Vegetable Tanning. 11. Dehydration of the Gelatin Sol by Tannic Acid Crystalline Tannins and Simpler Phenols. H. G. B. DE J o ~ a (Rec. traw. chim. 1924 4& 35-47; cf. A. 1923 i 821).-The stability of a gelatin sol is governed by two factors electric charge and hydration. At the isoelectric point pR ~ 4 . 7 the gelatin sol is uncharged but retains its stability above &he gelation temperature (about 40") on account of its hydration. Increase of the hydrogen-ion concentration causes the particles to assume a positive charge up to a maximum beyond which the charge decreases. In the same way decrease of the hydrogen-ion concentration imparts a negative charge to the particles up to a point beyond which discharge takes place. Isoelectric or discharged sols can be coagulated by addition of dehydrating substances such as alcohol. Hydration changes in a sol can be followed by viscosity measurements although the charge can also influence the viscosity (the quasi-viscous effect). This view of gelatin sols it is claimed expresses their behaviour better than the view which regards t.hem as ion disperse systems. It is shown by viscosimetric measurements that gelatin sols at 40" obey Poiseuille's law and that the viscosity minimum is situated at 40° at the isoelectric point. The effect of tannic acid and phenols on gelatin sols was studied viscosimetrically at this temperature at different hydrogen-ion concentrations. At a small concentration of hydrochloric acid (0.003N) tannic acid has a marked dehydrat- ing action indicated by a lowering of the viscosity but no precipi- tation occurs. At the isoelectric point tannic acid has both a dehydrating and a positive charging action on account of its own slight acidity. Consequently the viscosity minimum of a gelatin sol containing tannic acid lies at a small nominal concentration of alkali which serves to restore the medium to the pH of the iso- electric point. li'urt'her addition of alkali causes a powerful the product had slight reducing properties. E. s. C=54.25 H=6*75 N=15*35 S=1.35%. E. s.BIOOHEMISTRY. i. 343 rehydration. The dehydration is relatively greatest at the iso- electric point and if pushed far enough causes coagulation. The primary action of tannic acid on gelatin sols is therefore the same as on agar sols (loc. cit.). Phenol pyrocatechol resorcinol quinol pyrogallol phloroglucinol and ethyl gallate all have a similar dehydrating action on gelatin sols the dehydrating action increasing rapidly with the number of phenolic groups. In the case of the two crystalline tannins d-catechin and hamameli tannin the dehyd- ration is more powerful than with the simpler phenolic compounds. In the case of digalloylglucose however the positively charging influence of the tannin tending to increase the viscosity exceeb the dehydrating action tending to lower the viscosity at the isoelectric point so that the viscosity is actually raised. However a little sodium hydroxide in this case causes complete coagulation. Picric acid added to an isoelectric gelatin sol exerts at fbst a positively charging action but with greater concentrations dehydra- tion becomes very powerful the positively charging action decreases and finally precipitation occurs. Sodium picrate has no dehydrating action. E. H. R. Gradual Decomposition of Proteins. Partial Hydrolysis of Keratin (Hog Bristles). E. ABDERHALDEN and E. KOMM (2. physiol. Chem. 1924 132 1-ll).-Hog bristles (400 g.) were heated with 1% hydrochloric acid (2 litres) in an autoclave at 120" for five hours the product was neutralised with ammonia and then evaporated t o dryness mixed with sand and extracted with et'hyl acetate. From the extract there were obtained d-alanylglycine anhydride m. p. 250" (decomp.); prolyl-leucine anhydride m. p. 158" [a]" -46.3" (in alcohol) (apparently partly racemised) ; 1-prolyl-d-valine anhydride m. p. 250-252" and a compound CI4H,O3N3 m. p. 180-182" which yields on hydrolysis proline leucine and alanine and is formed from these by the loss of 3 mols. of water. From the product of the hydrolysis of hog bristles using 70% sulphuric acid for seven days the only pure compound isolated was 1-leucylserine [a];; -44.3" (in water) and from the product of the hydrolysis by 70% sulphuric acid for fourteen days a vulylvaline m. p. 255-260° [a]2,0 +3.5" (in water) was obtained. In both cases however other scission products were shown to be present. T7. 0. K.

ISSN:0368-1769    

DOI:10.1039/CA9242600253

出版商:RSC    

年代:1924

数据来源: RSC

摘要:

ANALYTICAL CHEMISTRY. ii. 269 Mineralogical Chemistry. Traversella Scheelite. G. CAROBBI (Gaxxetta 1924 54 59- 64).-Analysis of a crystal of scheelite from Traversella gave the following percentage results H,O 0.01 ; WO 79.51 ; Moo 0-58 ; Nb,O 0.08 ; Ta20 0.02 ; CaO 19.57 ; BaO trace ; SrO trace ; MgO 0.05 ; yttria earths 0.06 ; Ce,O 0.03 ; La,0,,Nd20,,Pr,0 0.05. T. H. P. Analcitic Lavas of North Africa and Classification of Lavas containing Analcite. A. LACROIX (Cmpt. rend. 1924 178 529-535; cf. Washington A. 1915 ii 275).-The basaltic lavas of Djebel Guerrien and Ben Ganah (N. Africa) are analcitic and not leucitic since dilute hydrochloric acid extracts Na,O? 5.11 and 4.07% and K20 0.76 and 0.72% in the two cases respectively. The results of numerous analyses (by Raoult) of analcites are given and the general conclusion is reached that the analcites of N.Africa are magmatic in origin. Mineralogical Constitution of Bauxites and the Limestones found in Contact with them. J. DE LAPPARENT (Cmpt. rend. 1924 178 581-583).-A general discussion of the composition of French bauxites and the limestones associated with them. E. E. T. E. E. T. Analytical Chemistry. Quantitative Chemical Analysis by RGntgen Radiation. H. STrmzma.-(See ii 216.) Automatic Burette with Improved Guard System. F. J. CONSIDINE (Chem. News 1924 128 149-150) .-The liquid is drawn into the burette from below through a stopcock the caout- chouc tubing connecting the reservoir with the mouth of the burette being detachable in the middle a t a short piece of glass tubing in order that a closed system may be maintained.Chemical Analysis with Membrane Filters. IV. Nature of the Filter and its Application in Analytical Chemistry. G. JANDER and W. JANDER [in part with H. k s ] (2. anal. Chem. 1923 63 273-291).-The rate of filtration through membrane filters is entirely unaffected by repeated use with all neutral salt solutions acid solutions weaker than N strong hydrochloric or acetic acid or N solutions of sodium hydrogen carbonate or acetate. Strong solutions of nitric or sulphuric acid alkali carbonate solu- tions stronger than N alkali hydroxide solutions stronger than A. A. E.ii. 270 ABSTRACTS OP CHEMICAL PBPERS. 0.2N or ammonium sulphide solutions stronger than N rapidly cause the pores to close and filtration then ceases.The ash of a 15-cm. filter weighs from 6 to 12 mg. ; most of this can be removed by soaking the filter for some time in N hydrochloric acid and nothing will be removed from the filter during the course of a sub- sequent analysis. The membranes do not adsorb even traces of salts during use either from acid or alkaline liquids and precipitates like antimony or manganese sulphide or hydroxide may be com- pletely removed from the filter by means of a jet of water. The amount of precipitate that adheres so strongly as not to be removed by this treatment rarely exceeds 0.2 mg. Indicators. VI. Theory and Practice of the Rational Use of Indicators in Acidimetry. A. THIEL (Z. anorg. Chern. 1923 132 159-178).-A critical historical survey of the recent work of Noyes Kolthoff Bjerrum Sorensen and the author.The author’s nomenclature given in earlier papers is elaborated. The term “ acidity level ” originally proposed by Friedenthal to express the logarithm of the hydrogen-ion concentration (Sorensen’s pE) is supported and definitions are given for ths indicator-constant characteristic etc. The chief practical object of the study is t o allow for errors as functions of the end-point with a particular indicator of the concentrations and temperatures of the reagents of concentrations of foreign substances etc. and an exhaustive list is given of the points on which systematic knowledge must be available in order that the full possibilities of quantitative stoi- Indicators €or Use in Determining the Acidity of Soils. E. RAMANN and H.SALLTNGER (Z. anal. Chem. 1923,63,292-301).- Dyes made by a German firm for use in determining the acidity of soils by Clark and Lubs’ method ( J . Wash. Bmd. Sci. 1915 5 609; A. 1916 ii 571) are shown to give equally good results as those manufactured in America. The theory construction and application of Bjerrum’s double wedge colorimeter for determining the hydrogen-ion concentration is explained. Potassium Carbonate and Potassium Hydrogen Carbonate as Standards [in Acidimetry] G. BRUHNS (Chern.-Ztg. 1924 48 89-90).-!I!he use of separate portions of potassium carbonate obtained by heating the hydrogen carbonate instead of sodium carbonate is recommended. If methyl-orange is used as the indicator in the subsequent titration the sharpness of the end- point is improved by removing the greater part of the liberated carbon dioxide by agitation or a6rat’ion of the solution before the last few drops of acid are added.Neutralisation of Mixtures of Acids and a Universal Buffer Mixture. E. B. R. PRIIIEAUX and A. T. WARD ( J . Chem. XOC. 1924 125 426-429).-A buffer solution consisting of phos- phoric phenylacetic and boric acids is described. The mixture is 0.02N with respect to each hydrogen ion and the whole is 0.l” By adding alkali or strong acid of the same normality a series A. R. P. cheiometry may be realised. s. I. L. A. R. P. H. C. R.ANALYTICAL CHEMISTRY. ii. 271 of solutions giving a pH range of 26-11-5 is obtained. The neutralisation curve (cf. A. 1915 ii 677) agrees well with that obtained by the hydrogen electrode.Determination of the Moisture Content of Expressed Plant tissue Fluids. R. A. GORTNER and W. F. HOFFLUN (Bot. Gaz. 1922 74 30&-313).-The moisture content of plant-juices was calculated from the refractive index determined by a Abbe refracto- meter fitted with the ordinary “ sugar ” scale. The refractive indices of solutions of inorganic salts and proteins in the concen- trations occurring in plant saps were sufficiently near to those of similar concentrations of sugars. There was close agreement between results so obtained and those determined by drying in a vacuum over sulphuric acid at the ordinary temperature. S. I<. T. A. G. P. Determination of the Hydrophilic Colloid Content of Expressed Plant-tissue Fluids. R. NEWTON and R. A. GORTNER (Bot.Qaz. 1922 74 442446).-The freezing point of the expressed plant-juice is first determined. The total solids are determined by the refractometric method (see preceding abstract) a quantity of sucrose sufficient to make a molar solution in the total water present is added to the plant-juice and the freezing point again determined. The magnitude of the excess depression over the calculated figure in the second determination is taken as a measure of the “ bound ” water not available as solvent for the sugar. The “ bound ” water is shown to bear a definite relation- ship to the hydrophilic colloids of the plant-fluid. Influence of certain Colloids on the Titration of Chlorides by Volhard’s Method. B. VAN DER BURG and C. A. KOPPEJAN (Chem. Weekblad 1924 21 66-67).-The presence of colloids Modification of Isaacs’ Colorimetric Determination of Blood Chlorides.M. DUPRAY ( J . Biol. Chem. 1924 58 675- 679).-Isaacs’ method (A. 1922 ii 716) depending on the con- version of silver chromate into sodium chromate by the chloride and tintometric determination of the chromate solution is made more sensitive by adding potassium iodide and 10% sulphuric acid and determining the iodine colorimetrically. The greater depth of colour makes it possible to use only 0.5 C.C. of blood. A. G. P. appears to make the process untrustworthy. s. I. L. G. M. B. Determination of Chlorides in Blood and Tissues. D. D. VAN SLYKE [with J. SENDROY jun.] ( J . Biol. Chem. 1923 58 523-529) .-Organic matter is destroyed and the chlorides are precipitated as silver chloride by digesting the blood or tissue on a water-bath with standard silver nitrate solution prepared in concentrated (d 1.4) nitric acid.The excess of silver nitrate is then titrated by Whitehorn’s modification (A. 1922 ii 272) of Volhard’s method. The method may be used either on a macro- or a micro-scale. E. s.ii. 212 ABSTRACTS OF CHE;MICAL PAPERS. Analysis of Alkali Bromides. L. W. WINKLER (Phnm. 2entr.-h. 1924 65 37-39).-0ne g. of potassium bromide or sodium bromide is dissolved in 500 C.C. of water; 100 C.C. of this solution are acidified with 10 C.C. of concentrated sulphuric acid and the boiling solution is titrated with O.1N-permanganate solution until a permanent red coloration is obtained. Each 1 C.C. of the permanganate solution is equivalent to 0.0119 g.of potassium bromide. In the case of ammonium bromide the solution must be boiled with the addition of sodium hydroxide to expel all the ammonia before the solution is acidified and titrated. To deter- mine chlorides in alkali bromides 1 g. of the sample is dissolved in 120 C.C. of water the solution acidified with 20 C.C. of sulphuric acid boiled and titrated with O6N-permanganate solution ; one drop of oxalic acid solut’ion is then added and the chloride is deter- mined by addition of silver nitrate solution and titration of the excess. w. P. s. Potassium Permanganate as a Standard for Iodometry. J. M. HENDEL (2. unal. Chem. 1923 63 321-324).-Results obtained by the use of potassium permanganate previously standardised against pure sodium oxalate as a standard for deter- mining the strength of sodium thiosulphate solutions for iodometric work agree to within O.lyo of those obtained by the use of iodine or potassium dichromate provided that the acidity of the solution lies between 0.1 and 0-7N.The test is preferably made by adding 35 C.C. of the permanganate solution to 50 C.C. of 0-7N-sulphuric or hydrochloric acid and 3 g. of potassium iodide contained in a glass-stoppered flask. After keeping during 5 minutes in the dark the solution is diluted to 450 C.C. and titrated with the thio- sulphate solution. A. R. P. Application of Potassium Permanganate in Oxidimetry. I. M. KOLTHOBF (Pharm. Weekblad 1924 61 133-137).-1n acid solution the reaction may be written MnO,’+SH’ + Mn”+ 4H20+5e (e=charge of a univalent cation) ; the oxidation potential is E = 0.0002T/sx log ([MnO~][H”Js/[Nn“])+E where E is a constant.8. I. L. Iodometric Determination of Sodium Sulphide. E. BERL and W. PFANNMULLER ((%ern.-Ztg. 1924 48 115-116).-Accurate results in this titration using the hydrolytic precipitation method of Jellinek and Krebs (A. 1923 ii 871) to check the sodium sulphide solutions are obtained by adding the diluted sulphide solution to an aoidified iodine solution; with a neutral solution the results are too high. In alkaline solution hydrolysis of the sulphide occurs. s. I. L. Determination of Sulphur Dioxide in Sodium Hydrogen Sulphite Solution. L. DEBUCQUET (J. Pharm. Chim. 1924 [vii] 29 66-65).--0ne C.C. of the solution is diluted with 20 C.C. of water and 25 C.C.of 5% mercuric chloride solution are added; the mixture is then titrated with 0-1N-sodium hydroxide solution,ANALYTICAL CHEMISTRY. ii. 273 using methyl-orange as indicator. The number of C.C. of the alkali solution multiplied by 0.0064 gives the amount of sulphur dioxide Determination of Incoagulable Nitrogen (Residual Nitrogen). F. FONSECA (Biochem. Z. 1924 144 175-178).- Comparative determinations have been made of the residual nitro- gen of human blood sera after deproteinisation by colloidal iron by sodium tungstate and sulphuric acid by trichloroacetic acid and by sodium tungstate and trichloroacetic acid (Pincussen). The colloidal iron method gives the lowest and the two last- mentioned methods the highest results from which it is concluded that in the latter cases adsorption of the residual nitrogen com- pounds is a t a minimum.On a protein diet an increase in the residual nitrogen may be demonstrated by using any of the pre- cipitants mentioned with the exception of colloidal iron. Carbc- hydrate and fat have no influence on the residual nitrogen. present. w. P. s. J. P. R6le of Adsorption in the Determination of the Residual Nitrogen. J. H. CASCAO DE ANCI~~ES (Biochem. Z. 1924 144 179-189).-Urea is not adsorbed by fibrin or by placental tissue; creatinine amino-acids and uric acid are adsorbed in amounts varying from 5 to S% whilst peptone is adsorbed to the extent of 25 to 30%. I n the presence of 2-5y0 of sodium chloride the adsorption of peptone is markedly reduced. Using the various methods of deproteinisation described by Fonseca (preceding abstract) the adsorption of amino-acids and more especially of peptone is considerable even in the presence of sodium chloride but by washing the coagulum with 5% sodium chloride or 1.OM acetone the adsorbed nitrogen compounds may be partly recovered.It is concluded that in determining residual nitrogen considerable loss of compounds of high molecular complexity such as peptones occurs. 5. P. Determination of Nitrogen by Kjeldahl’s Method and its Modifications. P. FmmY and H. LEVALTIER ( J . Phurm. Chim. 1924 [vii] 29 137-147).-Using casein the effect of introducing the following modifications and combinations thereof into the original process was tried catalysts (metallic sulphatea mercuric chloride stannic oxide) reducing agents (oxalic acid and oxalates) and substances raising the boiling point (potassium sulphate phosphoric acid).The digestion of 0-25 g. of casein with 5 C.C. of sulphuric acid mixed with 15 C.C. of phosphoric acid and 5 g. of potassium sulphate gave a theoretical yield of ammonia and the digestion was complete in fifteen minutes. The use of sulphates of copper mercury or iron as catalysts reduced the period of digestion at the expense of accuracy. The use of oxalic acid however gave satisfactory results. The boiling points of the various mixtures of acids and potassium sulphate were determined at various periods of the digestion; the best results were obtained with the mixtures of highest boiling point. In the case of mixtureaii. 274 ABSTRACTS OF CHEMICAL PAPERS.of acids and potassium sulphate of higher boiling point the presence of catalysts although it accelerates the decolorisation of the solution retards the complete transformation of nitrogen into ammonia. The digestion should be continued for fifteen to thirty minutes after the solution has become colourless. The onlv dis- advantage of phosphoric acid is that it attacks glass and silka. H. C. R. Bromoinetric Determination of Ammonia Sulphurous Acid Hydrogen Sulphide and Chromates. W. MANCHOT and F. OBERHAUSER (Ber. 1924 57 [B] 29-32; cf. this vol. ii 199 200) .-The solution of bromine in hydrochloric acid (about 20 yo) may be replaced by a similar solution in approximately N - potassium bromide which preserves its titre better. For the determination of ammonia a measured volume of 0.1- O.2N-bromine solution in potassium bromide is treated with approximately twice the calculated volume of about O4N-sodium hydroxide solution and the ammonium salt is added ; when after about 1-2 minutes the evolution of nitrogen is complete the solution is acidified with hydrochloric acid and treated immediately with an excess of arsenious acid.Unused arsenious acid is titrated witth bromine in the presence of indigo-carmine as indicator. Sulphurous acid is determined by allowing its solution to flow under the surface of a solution of bromine in hydrochloric acid and titration of the excess of halogen with arsenious oxide solution. The reaction does not require careful regulation of the concentration of the reactants as is the case with the iodometric determination of sulphurous acid.Solid sulphites and hydrogen sulphites are added directly to the solution of the halogen. Hydrogen sulphide is determined analogously by adding its solution to an excess of bromine dissolved in hydrochloric acid. Reaction occurs in accordance with the equation H,S +4H,O + SBr=SHBr +H,SO,. The troublesome separation of sulphur is not observed in sufficiently dilute solution. For the determination of chromate the solution is acidified with hydrochloric acid and treated with a sufficient quantity of potassium bromide and excess of arsenious acid solution. Unused arsenious acid is titrated with bromine in the presence of a solufion of indigo- carmine (0.2 g.) and trinitroresorcinol (0.2 g.) in water (100 c.c.).The method can be extended to the indirect determination of lead and:other metals which yield insoluble chromates. H. W. Determination of Nitrates in Soils. Phenoldisulphonic Acid Method. H. J. HARPER ( I d . Eng. Chern. 1924 16 180- 183).-fill details of a simple procedure for accurately determining nitrates in soils by the phenoldisulphonic acid method are given. Perfectly clear and colourless soil extracts are obtained by using copper hydroxide precipitated in the soil suspension from copper sulphate and calcium hydroxide as the decolorising agent. This substance removes. inappreciable quantities of nitrate from the solution by adsorption whereas animal charcoal adsorbs appreci-ANALYTICAL CHEMISTRY. ii. 275 able quantities of nitrate. Losses of nitrate can be prevented by keeping the solution alkaline on evaporation by removing the chlorides with silver sulphate and by flooding the dry residue with 3 C.C.of phenoldisulphonic acid solution. The procedure described embodies the best points of several methods with certain modifications due to the author. H. C. R. Direct Determination of Secondary Phosphate. I. N. KUGELMASS and C. ROTHWELL ( J . Biol. Chem. 1924 58 643- 648) .-An excess of saturated aqueous calcium sulphate pre- cipitates secondary phosphates quantitatively according to the equation 4&HP0,+3CaS04 =+= Ca,(P0,)2 +2KH2P0 +3K,SO whilst primary phosphates are unaffected. In direct determin- ations the phosphate in the precipitate is determined colorimetric- ally (A. 1922 ii 718). In the presence of twenty times the amount of primary phosphate the results are subject to an error of &5%.G. M. B. Use of Bromine for the Volumetric Determination of Arsenic and Antimony and in Winkler’s Method for the Determination of the Bromine-Iodine Value (of Oils). J. PRESCHER (Phrm. 2entr.-h. 1924 65 61-63).-Antimony tri- chloride in hydrochloric acid solution may be titrated with 0.W- potassium bromate solution using methyl-orange or indigo as indicator. The indicator is decolorised as soon as all the antimony has been oxidised. I n the titration of arsenic trioxide an indicator k not used; potassium bromide is added and the end-point of the titration denoted by the appearance of a faint yellow coloration. The following modification of Winkler’s method is suggested with t’he object of avoiding the use of potassium iodide.The weighed quantity of oil is dissolved in 10 C.C. of carbon tetrachloride 50 C.C. of 0-1N-potassium bromate solution 1 g. of potassium bromide and 10 C.C. of 10% hydrochloric acid are added; after two hours 10 C.C. of 0-1N-sodium arsenite solution and 20 C.C. of hydrochloric acid (d 1.19) are added and the mixture is titrated with potassium bromate solution. w. P. s. Photosynthesis. An Electrometric Method of Determining Carbon Dioxide. H. A. SPOEHR and J. M. MCGEE (Id. Eng. Chem. 1924,16 128-13O).-In determining changes in the carbon dioxide content of an air stream containing O%-0.6~0 by volume with an accuracy of 0.00274 the carbon dioxide was absorbed in a solution of barium hydroxide and the conductivity of the latter measured.The electrolytic cell employed was in the shape of a two-bulbed pipette with electrodes 0.25 sq. cm. in area. When charged with O.1N-potassium chloride solution this had a resist- ance of about 6000 ohms and errors due to polarisation and self- induction were found to be negligible. A resistance-concentration curve for the barium hydroxide solution was constructed by passing through it volumes of outdoor air taken to have the fixed carbon dioxide content of 00031% by volume. c. I.ii. 276 AkSTRACTS OF (IHEMIUAL PAPERS. Detection of Sodium and Potassium by Wet Analysis. G. G. LONGMESCU and (MLLE.) G. CHABORSKI (Bul. SOC. R d n a Xtiin. 1923 26,21-26).-Details are given of a method for detect- ing sodium and potassium by precipitation as pyroantimonate and hydrogen tartrate respectively the remaining metals being removed by precipitation with barium carbonate.N. TARUGI and G. GASPERINI (Boll. Chim. Furm. 1924,63,33-38,65-70).-Boudron and Boudet’s modification of Clark’s method for determining the hardness of water although in common use is an empirical method and presents many defects. The authors discuss the numerous suggestions which have been made with the object of overcoming such defects and propose the use of a palmitate solution for the purpose. [Cf. B. 1924 310.1 T. H. P. Ash Determination by Aid of Oxygen. D. J. K. WETSELAAR (Phurm. Weekblad 1924 61 212).-In determining ash in organic materials a great saving in time is effected if a gentle stream of oxygen is directed on to the carbon skeleton formed by heating Alkalinity of the Ash of Foodstuffs.11. Simultaneous Titration of a Number of Components of the Ash. B. PFYL and W. SAMTER (2. Unters. Nahr. Genus.sm. 1923,46,241-275).- An improved method of determining phosphate consists in pre- cipitating the tricalcium phosphate with sodium hydroxide in the boiling solution in the presence of excess of phosphate when a finely crystalline precipitate is obtained which is very little hydro- lysed on keeping. The influence of the following constituents of the ash of foodstuffs on the alkalinity and phosphate determination was systematically studied and methods were worked out for preventing their disturbing effects manganese iron aluminium zinc lead tin silicic boric and hydrosilicofluoric acids.Volu- metric methods depending on acidimetric and alkalimetric titrations with phenolphthalein and methyl-orange as indicators are described for the determination of manganese iron aluminium zinc lead and boric acid in the ash in conjunction with the determination of the alkalinity and phosphate content. A rapid volumetric method for determining calcium in the ash is also given. [Cf. B. 1924 229.1 H. C. R. Determination of Copper as Cuprous Iodide. L. W. WINRLER (2. anal Chem. 1923 63 32&33O).-The determin- ation of copper as cuprous iodide by the following procedure yields results more accurate than those obtained by the gravimetric oxide or sulphide processes. The boiling neutral or faintly acid solution containing 0.01 to 0.1 g. of copper in a volume of 100 C.C.is treated in succession with 1 g. of ammonium chloride 10 C.C. of 10% potassium iodide solution and sufficient of a 10% sodium sulphite solution to remove the liberated iodine and t o came the liquid to have a slight odour of sulphur dioxide. After boiling during 1-2 minutes the solution is set aside over-night in a dark place. The precipitate is collected on cotton wool washed with 50 C.C. W. H.-R. Determination of Hardness of Water. in a porcelain or platinum boat. s. I. L.ANALYTICAL CHEMISTRY. ii. 277 of cold water dried a t 130° and weighed. As a correction for the slight solubility of cuprous iodide 1 mg. is added to the weight found. The process is applicable to the separation of copper from the alkali metals magnesium zinc manganese cobalt and nickel.Cadmium salts exert a slight solvent action on the pre- cipitate which may be suppressed by the addition of more iodide. Even very small quantities of lead are co-precipitated and impart a pale red colour to the cuprous iodide. Iron does not interfere if an excess of sodium sulphite is added before the potassium iodide. Critical Studies on Methods of Analysis. IV. Aluminium. L. A. CONGDON and J. A. CARTER. V. Cadmium. L. A. CONGDON and R. C. CANTER. VI. Molybdenum. L. A. CONGDON and L. V. ROHNER (Chem. News 1924 128 98-100; 116-118 ; 118-120).-1V. Duplicate determinations of aluminium in aluminium chloride were made by seven different methods from which it appears that hydrolysis of the solution by boiling with ammonium nitrite yields the most accurate results whilst pre- cipitation with ammonia in the presence of ammonium chloride gives high results and the iodide-iodate method low results.Volumetric methods involving determination of the free acidity after elimination of the aluminium as fluoride or alum followed by direct titration of the total acid in a second sample yield results comparable with those obtained by the nitrite method. V. Precipitation of cadmium as sulphide and weighing as such after drying at loo" or evaporation of the cadmium solution with sulphuric acid followed by ignition of the residue and weighing as cadmium sulphate gave results with pure cadmium chloride closely approximating to the theoretical whereas the figures obtained by precipitation with potassium carbonate and ignition to oxide or by precipitation as cadmium ammonium phosphate followed by conversion to the pyrophosphate were decidedly low.Of the volumetric methods tested titration of the sulphide with iodine and titration of cadmium oxalate with permanganate gave good results whilst indirect titration of the arsenate or sulphide gave slightly low results. VI. In determining the molybdic acid in pure ammonium molybdate precipitation with lead acetate followed by weighing the ignited precipitate or distillation of the solution with hydro- chloric acid and potassium iodide gave results very close to the theoretical. Low figures were obtained by precipitation as sulphide followed by roasting to and weighing as oxide and high results by weighing as oxide after precipitation as mercurous molybdate.A. R. P. A. R. P. Simultaneous Electrometric Determination of Iron and Manganese. E. MULLER and 0. WAHLE (2. anorg. Chem. 1923 132 260-264) .-The solution is titrated in presence of sulphuric acid at normal temperature with 0-1N-potassium dichromate solution using a normal calomel electrode and a back E.M.P. of 0.57 volt; the whole is then heated at SO" potassium fluorideii. 278 ABSTRAOTS OF CHEMICAL PAPERS. added and the mangaiious salt titrated to the manganic state with 0-1N-permanganate solution as previously described (ibid. 1923 129 278; this vol. ii 64). Critical Studies on Methods of Analysis. VII. Cobalt. L. A. CONGDON and T. H. CHEN (Chem. News 1924 128 132- 134).-In determining cobalt by precipitation as cobalt ammonium phosphate accurate results are obtained only if the small quantity that escapes precipitation is recovered as sulphide and converted into oxide for weighing. The nitroso- @-naphthol method yields more consistent and satisfactory results than any other method whilst precipitation of a complex cobaltic molybdate gives high and inconsistent figures.Co-precipitation of Cobalt and Nickel by Stannic Sulphide. Colorimetric Determination of Cobalt. V. AUGER and (MLLE.) L. ODINOT (Compt. rend. 1924 178 710-711).-The stannic sulphide precipitated from acid (hydrochloric) solutions containing cobalt or nickel contains these metals the amount adsorbed vary- ing inversely as the concentration of the acid. Stannous sulphide does not adsorb these metals. The determination of tin in such solutions is best effected by using cupferron.The blue colour obtained by dissolving cobalt salts in an excess of concentrated hydrochloric acid may be used for the colorimetric determination of cobalt. The influence of concentration of acid on the blue colour has been investigated. Detection of Chromium in the Presence of Manganese by Wet Analysis. (MLLE.) G. CHABORSKI (Bul. SOC. Romdna Xtiin. 1923 26 83-85).-The mixed hydroxides of iron chromium and manganese remaining after the separation of aluminium and zinc by sodium hydroxide solution are treated with 1 to 2 C.C. of sodium hypochlorite solution heated to the boiling point without actually being allowed to boil and filtered. If chromium is present the filtrate is coloured yellow. I n the absence of nickel a black residue obtained in the test for chromium indicates the presence of manganese but if nickel was originally present the test is not conclusive and a fresh portion of the mixed hydroxide precipitate is treated with 5-10 C.C.of sodium hypochlorite solution boiled for 2-3 mins. and the remaining solid allowed to settle when if manganese is present the liquid will be coloured Determination and Separation of Bismuth by Hydrolysis. G. LUFF (2. anal. Chem. 1923 63 330-348).-Bismuth may be quantitatively separated from lead copper and cadmium by precipitation as the basic nitrate 2Bi2O,,N,0,,2H,O. The solution (80 c.c.) is neutralised with sodium hydrogen carbonate the pre- cipitate dissolved by adding N nitric acid and the liquid boiled under a condenser. Precipitation of the basic nitrate may be brought about by the addition of either sodium nitrite or sodium hydrogen carbonate solution.The precipitate is ignited to s. I. L. A. R. P. E. E. T. violet owing to the formation of permanganate. W. H.-R.ANALYTICAL CHEMISTRY. ii. 279 and weighed as oxide. I n carrying out the oxychloride precipit- ation it is best to neutralise the nitrate solution add ammonium chloride and hydrochloric acid heat to boiling and slowly add boiling water. To obtain the basic sulphate the neutralised nitrate solution is treated with 1 10 sulphuric acid until the precipitate dissolves and boiling water added to the boiling solution; the bismuth is completely precipitated as microscopic colourless needles of the composition Bi,O,,SO,,H,O.This compound remains un- changed when dried a t 100"; it contains 74.02y0 of bismuth and is quantitatively converted to oxide by prolonged ignition over a blast burner. A. R. P. Reduction of Bismuth Salts by Hydrazine Hydrate and the Titration of Metallic Bismuth. J. HA NU^ and A. JILEK (Chew. Listy 1924 18 8-12).-The bismuth salt solution is boiled with the addition of hydrazine hydrate the resulting metallic bismuth is collected on a filter and washed with water containing dissolved carbon dioxide ; the filter and precipitate are then trans- ferred to a vessel containing ferric chloride solution acidified with hydrochloric acid a current of carbon dioxide is passed through the mixture and when the bismuth has dissolved the ferrous chloride produced is titrated with O.1N-permanganate solution. The presence of a small quantity of lead (less than 5% of the bismuth) does not interfere with the determination.W. P. S. The Resorcinol Test for Methyl Alcohol. A. B. LYONS ( J . Amer. Pharm. ASSOC. 1923 12 323-327).-The presence of ethyl alcohol influences the resorcinol-sulphuric acid (formaldehyde) test but rarely renders it invalid. CHEMICAL ABSTRACTS. Detection of Methyl Alcohol in Presence of Ethyl Alcoho?; C. H. LAWALL (Trans. Wagner Free Inst. Xci. Philadelphia 10 55-62; A m r . J. Phnrm. 1923 95 812-820).-To 5 C.C. of the mixture diluted to contain about 5 vol.-yo of ethyl alcohol are added 5-6 drops of phosphoric acid and 2 C.C. of 3% aqueous potassium permanganate solution the liquid being then allowed to remain for 10 minutes.Oxalic acid (1 C.C. of a 10% aqueous solution) is then added and the mixture kept until it is trans- parent brown. Dilute sulphuric acid (1 3 mols.) then U.S.P. magenta-sulphurous acid test solution (5 c.c.) are added well mixed and allowed to remain for 10 minutes. The presence of methyl alcohol is indicated by a blue or violet colour observed against a white background. The test is sensitive t o 1 part of rnet,hyl alcohol in 500 parts of ethyl alcohol; preliminary fraction- ation increases the sensitiveness t o 1 in 10,000. The record of the test may be made permanent by dyeing white wool. Increasing t'he period of oxidation increases the sensitiveness but heating after addition of permanganate produces formaldehyde from ethyl alcohol; increasing the volume of permanganate from 2 to 5 C.C. also caused ethyl alcoliol to give a positive reaction.CHEMICAL ABSTRhCTS.ii. 280 ABSTBACTS OF CHEMICAL PDERs. Determination of the Copper Number of cellulosic Materials. The Fontas-Thivolle Molybdomanganimetric Method. H. GAULT and B. C. MUKERJI (Compt. rend. 1924,178 711-713).-About 1 g. of the cellulose is added to 50 C.C. of Fehling’s solution diluted with 100 C.C. of water and previously heated a t 120” (calcium chloride bath). Heating is continued for exactly 15 minutes the mixture then filtered the cuprous oxide (or the copper obtained therefrom by reduction) dissolved in the phosphomolybdic reagent and determined by titration with permanganate. E. E. T. Rapid Analysis of Sugars. Purification and Concentration of m y m e So~utions.F. w. REYNoLDs.-(8ee i 464.) Table for the Determination of Sugar in Blood. G. A. WETSELAAR (Pharm. Weekblad 1924 61 213-218).-A complete table has been prepared for determinations by Shaffer and Hart- mam’s method (A. 1921 ii 417) as modified by Cohen and Tervaert (Nederl. Tijdsch. Geneesk. 1921,65 ii No. 7). The figures were obtained by determinations after addition of known quantities of dextrose and are accurate to the third signXcant figure. From the volume of 0-005N-thiosulphate solution used under specifled conditions the weight of dextrose (mg.) in 1 C.C. of blood is read I. Folin and Wu’s Method for the Determination of Blood Sugar. 11. A Modification of the Method. V. E. ROTHBE~G and F. A. EVANS ( J . Biol. Chem. 1923 58 435-442 443451).- I. The results obtained with Folin and Wu’s method (A. 1920 ii 337) are trustworthy only when the concentrations of dextrose in the standard solution and that under examination are approxim- ately the same. 11. When the concentrations in the two solutions do not differ by more than loo% this source of error can be over- come by diluting the blood filtrate after addition of the phospho- molybdate-phosphotungstate reagent until the colour is approxim- ately the same as that of the standard solution and introducing a volume factor into the calculation. All the precautions emphasised in the original method must be observed. Preparation of Dried Urease and the Determination of Urea in Normal and Pathological Urines. G. REVOLT=.- (See i 474.) Fluorescent Oxidation Products of Bilirubin and their Importance as Sources of Error in the Detection of Urobilin. A. ADLER (Biochem. Z. 1924 144 64-65).-A reply to the criticisms by Barrenscheen and Weltmann (A. 1923 ii 800) of the author’s work on the detection of urobilin in human fistula bile. directly. s. I. L. E. S. J. P.

ISSN:0368-1769    

DOI:10.1039/CA9242605269

出版商:RSC    

年代:1924

数据来源: RSC

摘要:

ii. 281 General and Physical Chemistry. Foundations of the Quantum Theory. I. H. A. SENFT- LEBEN (2. Physik 1924 22 127-156).-A discussion of the principles underlying the quantum theory and its applications to intra-atomic changes. The mechanism of atomic radiation is considered. S. B. Quantvm Defect md the New Bohr Theory of Atcjmic Structure. F. S. BRACKETT and R. T. BIRGE ( J . Opt. SOC. Arner. 1924 8 213-230).-According to the new Rohr theory successive increase in the atomic number is associated with the conversion of virtual orbits into orbits which are permanently occupied (cf. A. 1923 ii 110 478 679). Direct calculations of t’he energies of orbits are only possible for atoms possessing one orbital electron (e.g. H and ionised He) but this theory makes possible a correlation of experi- mental data for more complex atoms.In the present paper spectroscopic and X-ray data are used to calculate the energy values of the various atomic states of elements up to atomic number 56 and quantum numbers are assigned to each of these atomic states. The relation between the energies of corresponding levels in different elements is illustrated by curves which show many regularities. These diagrams may be used to predict the values of unknown orbits. The points on the curves obtained from “optical” and “ X-ray ” spectra show good agreement. The assignment of orbits differs from Bohr’s in a few instances.. R. T. BIRGE ( J . Upt. SOC. Amer. 1924 8 233-243).-New series of triplets found by Hopfield in oxygen and sulphur are assigned to a new energy level OP.The deviations from Rydberg’s formula found in the same type of sequence of related elements are inter-related so that it is possible to predict approximate values of unknown spectral terms (see pre- ceding abstract). The missing 0s level of calcium and zinc is probably identical numerically with the I8 level but has a different inner quantum number. It is suggested that any given type of orbit can exist in two states when a virtual orbit or an X-ray orbit but in only one possible state as a valency orbit. Simultaneous Action of Crossed Electric and Magnetic Fields on the Hydrogen Atam. 0. KLEIN (2. Physik 1924 22,109-1 18).-The disturbances of the hydrogen electron in crossed electric and magnetic fields are calculated by a method similar to that used by Bohr for the Stark effect; these results are employed t o find the stationary states for such a system.Hydrogen Balrner Series and the Impossibility of Further Corrections to the Quantising of Hydrogen Atoms. A. E. RUARK (Astrophp. J . 1923 58 46-55).-A series of plates VOL. CSXVL ii. 11 S. B. Spectral Series of Divalent Elements. S. B. S. B.ii. 282 ABSTRACTS OF CHEMICAL PAPERS. obtained by Wood has been subjected to measurement in order (i) to furnish better wave-lengths for the higher members of the series (ii) to determine the Rydberg constant and (iii) to test the necessity for corrections arising in the asymmetry of the nucleus or electron or in the existence of an intrinsic magnetic moment in either or in the possibility that part of the potential energy of the atom may be considered to move with the electron so as to change its apparent mass These three corrections are shown to be invalid; tthe relativity correction only is necessary to explain existing observations.Very accurate values of 1 from H7 to H18 are tabu- lated. The mean value of R (computed for the lines 3 to 12 inclusive) is 109677.26 & 0.23 whence R = 109736.45 & 0.23. A. A. E. Doublet Separation of the Balmer Lines. G . M. SHRUM (Proc. Roy. Xoc. 1924 A 105 259-270).-The doublet separations in the first five lines of the Balmer series of hydrogen have been determined a t the temperature of liquid air and found to be in almost exact quant'itative agreement with the theory of Sommerfeld this being the first direct proof of the theory in the case of hydrogen.The intensit;y of t3he Balmer lines relatively to that of the secondary spectrum may be increased many times if the inner walls of the quartz tube in which the discharge takes place are coated with a very thin and even film of ice which together with adsorbed nitrogen shields the atomic hydrogen from the catalytic action of the quartz and so diminishes the intensity of the secondary spectrum which is due to hydrogen molecules. No evidence of the fine structure was observed even at the temperature of liquid hydrogen. Intensities of Complex Spectral Lines. H. B. DORZELO (2. Fhy& 1924 22 170-177; cf. ibid. 1923 18 206) -The relative intensities of wide doublets and triplets have been measured. The intensity ratio of the alkali metal doublets was found to be 2 1 and the ratios for the alkaline-earth triplets to be 5 3 1.The intensity ratios of the components of simple doublets and triplets are fixed by the inner quantum numbers of the spectral terms which correspond to the series limits (LandB's J-values). W. H.-R. S. B. Fine Structure of Certain Spectral Lines. E. LAU (Physikal. Z . 1924 25 60-68).-A critical discussion of methods of spectral measurement with special reference to the fine structure of the hydrogen and ionised helium lines. It is shown that measurements involving the use of step or line gratings or Lummer-Gehrcke plates in conjunction with photographic observations are subject to errors producing an apparent broadening 01 doublets of 40%. Interfero- meter methods or those involving photometric examination are held to be less liable to error.The most probable value for AvH is 0.298 cm.-l whereas t$he theory of Sommerfeld requires AvH = 0-365 cm.-l. In the Fowler and the Pickering series of the helium spectrum only the triplet 4686 8. gives the theoretical A= values ; thus the relativity correction in the Bohr equation for the heliumGENERAL AND PHYSICAL CHEMISTRY ii. 283 spectrum should not be regarded as confirmed. The fine structure of the X-ray spectra is discussed. Direct Determination of the Relative Intensities of [the Components of ] Multiple [Spectral] Lines. W. GERLACH and 0. BREZINA (2. Physik 1924 22 215-221).-The slit of a spectro- graph is illuminated by light which has traversed a pair of crossed Nicols.Interposed between the Nicols is a thin doubly refracting plate of mica quartz etc. cut so that the planes of vibration in the plate are inclined at 45" to the planes of vibration in the Nicols. The plate comprises a small central portion which can be rotated relatively to the outer and is separated therefrom by a narrow gap. The relative intensities of lines in the visible region are compared by rotating the inner portion of the plate until the intensity of the one line as viewed through the inner portion is adjusted to equality with the infensity of the other line as seen through the outer portion of the thin plate. Lines outside the visible region of the spectrum are photographed and an interpolation method is used. The results obtained are not vitiated by any possible unsteadiness of the light source employed. Contrary to the result obtained by Wood (A.1914 ii 398) the authors find that even in flames only slightly coloured by sodium the relative intensities of the D and D lines are in the ratio 1 2. The mathematical theory of the method is given. J. S. G. T. Width of the Lines in the Spark Spectrum of Hydrogen. M. HANOT (Compt. rend. 1924,178,1071-1073).-The " width " of a line is defined as the interval of wave-length corresponding with a diminution of the maximum intensity by one-half. The character of the spectrum is independent of the damping of the discharge but increase of the period decreases the width of the lines HP and Hy which were investigated. The variation in width is attributed to A. E. 81. the effect of the peFiod on the temperature and pressure of the gas. A.B. H. Excitation of the Helium Spectrum by Electronic Bombard- ment. A. UDDEN and J. C. JACOBSEN (Physical Rev. 1924 23 322-326).-At the higher voltages used the ortho and parhelium spectra were of about equal intensity but a t low voltages the former was relatively much the stronger. In explanation the supposition of the existence of the metastable state of the helium atom is favoured. Experimental work to test this view has so far been negative. A. A. E. Excitation of Spectral Lines by Electron Impact. I. G. HERTZ (2. Physik 1924 22 18-26).-1t is still uncertain whether lines of it spectral series can be successively excited by gradual increase of the energy of electron impact or whether they appear together (except the resonance lines) a t the ionisation potential. In the present paper small refinements of the hot cathode and grid anode apparatus usually employed for these investigations are described.Errors due to the space field in the ionising chamber we eliminated by using very small current densities and by 11-2ii. 284 ABSTRACTS OF CHEMICAL PAPERS. introducing small quantities of a gas or vapour (usually mercury) of ionisation potential lower than the resonance potential of the gas under investigation. The successive development in accordance with the Bohr theory of lines in the mercury neon helium and zinc spectra has been observed with this apparatus. Excitation- of the Spectra of Argon of Krypton and of Xenon. G. DBJARDIN (Corn@. rend. 1024 178 1069-1071).- A continuation of previous work on the rare gases (Compt. rend.1921 172,1347 1482 ; A. 1923 ii 282). A study of the develop- ment of the spectra of the neutral atom the monovalent ion El the doubly E and triply charged ions E as the potential in- creases from 10 to 100 volts. The arc-spectra appear when the potential of the electrons exceeds the ionisation potential and the spark spectra of the first and second order appear a t higher potentials. The excitation potentials of the different spectra of argon krypton and xenon are proportional amongst themselves. A. B. H. S. B. Intensity Relations in the Helium Spectrum. C. B. BAZZOXI and J. T. LAY (Physical Rev. 1924 23 327-336).-The crossed orbit lines 4437 4387 4922 and 5016 A. increased in relative in- tensity with voltage although not uniformly whilst the co-planar lines 4713 5875 and 4471 A.diminished to a minimum at 50 volts. The results are in general agreement with those obtained by Hughes and Lowe (ibid. 1923 21 714; -4 1923 ii 804). A. A. E. Excitation Stages in the Open Arc-light Spectra. 111. Lead Mercury Thallium Magnesium. IV. Hydrogen Air Water-vapour Pressure Effect Mixed Elect.rodes. B. E. MOORE (Astrophys. J . 1923 58 86-103 104-112; cf. A. 1922 ii 242) .-111. Photographic methods were applied in the observation of the lines which are tabulated. At the pole tip the temperature and velocities of the electrons were always high enough to produce appreciable dissociation of the atoms. In the case of lead the lines represent very easy excitations and probably fall in stage I ; two stages and possibly a third are found in the development of the lines of mercury two in the case of thallium and four for magnesium corresponding respectively with resonance first ionisation subse- quent resonance and second ionisation.The substitution of hydrogen for air or nitrogen reduces the intensity of the spectra of the metals in the middle of the arc. It is concluded that hydrogen is completely dissociated into atoms within the polar illurninatlion m d that the " second hydrogen spectrum " is the first hydrogen excitation. Decreased pressure is accompanied by a decrease in temperature and results in less contrast in t'he series lines. The effect of volatility of the electrodes is also considered. The changes due to pressure and temperature together with the variations in electron velocity are considered to be sufficient to explain all of the differences which have been noted in the spectra. IV.A. A. E.GENERAL AND PHYSICAL CHEMISTRY. ii. 285 Series Spectrum of Ionised Carbon (CII). A. FOWLER (Proc. Roy. Xoc. 1924 A 105 299-310; cf. A. 1923 ii 447).-The spectrum of carbon has been examined under a variety of experi- mental conditions and the discovery of previously unrecorded lines has led to the identification of the leading members of the series of ionised carbon (CII or C+) in the ordinary region of observation. Tkre ma'in series however lie in the extreme ultra-violet and have been identified in the observations of Simeon (A. 1923 ii 709) and of Millikan (A. 1921 ii 3).The series of CII form a doublet system as would be expected from the spectroscopic displacement lam and previously unrecorded pairs of lines have been found a t 7236-19 and 7231-12 2174.14 and 2173.86 2137-93 and 2137.45 The highest terms are the common limits of the sharp and diffuse series 196612 and 196670 from which it may be deduced that the second ionisation potential of carbon is 24-28 volts representing the energy necessary to remove the second electron when the first has already been removed. In the normal state of the atoms tlwo of the outermost electrons in ionised carbon probably move in 2 orbits whilst the electron which generates the spectrum traverses a 2 orbit as suggested by Bohr for neutral boron. W. H.-R. Arc Spectrum of Silicon in Relation to Spectrographic Analysis. c.P0RLEzzA.-(See ii 345.) Arc Spectrum of Chromium. H. GIESELER (2. Physik 1924 22 228-244).-Lines belonging to the quintet and septuplet systems characterising the spectrum of chromium are classified into series and the values of the series terms calculated. Details are given of the magnitudes of the Zeeman effect exhibited by lines of (notation of Land6 and Paschen). The results confirm those obtained by Catalan (A. 1923 ii 519). The normal orbit of the chromium atom is a 4s orbit. A number of terms relating to the systems of quintets and septuplets may be calculated by means of Ritz formule. The system of quintets are classified into twelve series including four triplets of p terms. The f and d terms are of the order of magnitude of the p terms.Particulars are given of combination lines the frequencies of which may- be calcul- ated by a combination of terms relating to the quintet and septuplet systems respectively. Details are given of the Zeeman effect exhibited by the triplet 5208 5206 5204. J. S. G. T. F. M. WALTERS ( J . Opt. Xoc. Amer. 1924 8 245-274).-Twenty-two polyfold energy levels (of the same total and azimuthal quantum numbers and nearly the same energy value but with different inner quantum numbers) have been found in the arc spectrum of iron producing 54 or more multiplets by intercombination. From the inner quantum numbers and Landd's scheme for the Zeeman effect several polyfold levels have been identified with P D and F terms. Lines in an iron multiplet show similar pressure effects and behave alike in the electric furnace.S. B. the series 4 - 424,2:.,22 49 13 - 45" 3,,23,22 and of the series 472 - 4&,21 Regularities in the Arc Spectrum of Iron.ii. 286 ABSTRACTS OF CHEMICAL PAPERS. Higher Order Spark Spectra. L. BLOCH E. BLOCH and G. D~JARDIN (Cornpi. rend. 1924,178 766-769).-Previous work (A. 1923 ii 350) has been applied to argon krypton and xenon. Three spark spectra are observed the first (brightest) extends from the extreme red to the extreme ultra-violet; the second begins a t the end of the violet and does not enter the visible spectrum ; the Ohird is mainly in the ultra-violet region. Secondary p-Rays produced in a Gas by X-Rays. I?. AUGER (Cornpt. rexd. 1924 '178 929-931).-Hydrogen saturated with water vapour has been investigated as previously described (*4.1923 ii 601). The 163 p-rays studied were sufficiently recti- linear to allow of measurement to within a few degrees of the angles made with the X-rays. The rays projected forward are virfually double as numerous as those projected backward more than 85% of the electrons being emitted a t inclinations to the direction of the X-rays ranging from 45" to 115"; the emission is approximately symmetrical about the middle position 80". A theoretical interpretation of these facts is given. Direction of Ejection of Photo-electrons by Polarised X-Rays. Symmetry of Incident and Emergent Photo-electronic Velocities. R. J. PIERSOL (Physical Rev. 1924 23 144-152). Scattering of X-Rays by Hydrogen. G. A. SCHOTT (Phgsical Rev.1924 23 119-127).-Measurements of the absorption and scattering of X-rays by A u r h (A. 1919 ii 89; 1921 ii 367) and by Olson Dershem and Storch (A. 1923 ii 365) have been recalcul- ated from the actual results instead of from smoothed curves using the method of least squares and avoiding Aurbn's approximations. By suitable choice of the quantity go the ring-electron theory and Compton's quantum theory can be made to fit the curve approxi- mately the former better at large and the latter a t small wave- lengths. Sufficiently accurate experimental results could be secured only by using elementary hydrogen either liquid or gaseous under high compression. Rotatory Dispersion of Quartz. F. BURKI (Helv. Chim. Bcla 1924 7,328-330).-The author's rotatory dispersion formula [a] = CIA2.@/*' (this vol. ii 218) has been used to calculate the rotation-dispersion of quartz. The calculated values accord more closely with experiment than those derived from Drude's formula [a] = kl/(A - Ao2) + k'fA2. The values of the constants are C = 7.1505 p = 0.01864. The sum of Drude's constants El = 12.200 and E l = - 5.046 is practically equal to the value of C. E. H. R. Molecular Absorption Spectra of Hydrogen Cyanide. E. I?. BARKER (Physical Rev. 1924 23 200-204).-Two maxima very nearly symmetrical and with nearly equal intensities were observed a t about 6.94 p and 7.23 p ; three others unsymmetrical were found a t 4.756 p 4.723 p and 4.79 p and a single narrow E. E. T. T. H. P. F. W. BUBB (Physical Rev. 1924 23 137-143). A. A. E.GENERAL AND PHYSICAL CHEMISTRY.ii. 287 maximum at 3664 p. Estimates of the moment of inertia of the molecule from various bands do not agree; measurements a t $p give a value of about 13.2 x 10-40. A. A. E. H. GIESELER and W. GROTRIAN (2. Physik 1924 22 245-260).- In continuation of previous work (A. 1923 ii 106) the authors have observed the absorption spectra of chromium and iron vapours respectively a t a temperature of about 1230". For chromium vapour absorption was observed in the case of the triplets 4289.92 4275.01 4254.52 A. and 3605.49 3593-64 3578.81 A. constituting the principal series of the septet system. Contrary to anticipation absorption was not observed in the case of the combination lines 4pi and 4pi0 viz. 3732.18 3730.95 8. and 3351.97 3379.18 A.The 4s term of the septet system corresponds with the normal condition and the 24 electrons of the chromium atom are arranged in a 4 orbit. With iron vapour absorption was observed a t 1230-1250" iu the case of the lines 3886029 3859.91 3745.56 3737.14 3719.96 3047.61 ( ?) 3021.08 and 3020.64 (not separated) 2973.24 2966.90 and 2719.04. Absorption was most marked in the case of the line 3859.91 and was only just discernible in the case of the lines 2719.04 and 3047.61. The twenty multiplets classified by Walters in the ion arc spectrum (A. 1923 ii 519) are arranged according to the energy levels associated with the respective quintet and triplet terms. The lines in the iron spectrum correspond with an electronic transition from the lower group of levels 3d4,392,1,0 comprised within this arrangement of levels.The existence of triplet and quintet systems in the iron arc spectrum is in accord with Sornmerfeld's extension of Rydberg's rule of alternation to multiplets (A. 1923 ii 734). It is concluded that most probably the term 3 4 represents the normal condition of the iron atom and that the 26 electrons of the iron atom are arranged in a 3 orbit. At about 1600° the iron lines previously observed and their multiplets are shown to be present as absorption lines. In addition marked absorption was observed in the region 2447 to 2491 8. The components of this multiplet are classified. A few additional unclassified lines in the region 5500- 2300A. were feebly absorbed at about 1600". F. G . WICK (Proc. Amer. Acad. Arts Xci.1923 58 557-573).-With the aid of a modified form of Bridgman's apparatus (ibid. 1911-12 47 321; 1913-14 49 627) experiments have been performed with aqueous solutions of neodymium praseodymium erbium (with yttrium) uranium and cobalt compounds and with s-ynthetic ruby and samples of coloured glass a t pressures up to 3500 atm. In general the changes which take place in the absorption spectra on increase of pressure are similar to those observed on lowering the temperature or decreasing the concentration. There is however a slight shift (in most cases towards the violet but for the band 4272 towards the red) in the position of the neodymium bands on increase of pressure but not under the latter conditions; only Absorption Spectra of Chromium and Iron Vapours. J.S. G. T. Effect of Pressure on Optical Absorption.ii. 288 ABSTRACTS OF CHEMICAL PAPERS. neodymium solutions containing a dehydrating agent show an apprsciable shift with rise of temperature. Bands barely visible under normal conditions are brought out more strongly by either increase of pressure or lowering of temperature but not by dilution ; the effect of the highest pressures used with neodymium solutions was however much less than that produced by lowering the temperature to that of liquid air and was possibly equivalent to that observed at about -60". The pressure effect cannot be explained by any resulting changes of temperature or concentration ; there must be more than one simple type of mechanism involved in the production of the absorption bands since the three agencies affect only some of the bands and those in different ways.Since absorption bands in solution are made sharper by pressure those of the crystalline neodymium salts also being sharper it is suggested that the structure of solutions under pressure tends towards that of the solid state. S. J. WAWILOW (2. Physik 1924 22 266-272). -Employing a spectrophoto- metric method similar to that used by Wood and Dunoyer for the investigation of the resonance spectrum of sodium vapour (2nd Solvay Conf. 1921 p. 307) the author has determined the value K of the ratio of the energy of the secondary fluorescent radiation emitted by solutions of various dyes to t'he total energy absorbed. The strengths of the solutions employed ranged from to low6 g./c..c.The error in K due to the small absorption of fluorescent radiation by the solution and to reflection by the glass walls of the containing vessel was 10 to 15%. The following values of I< were determined aqueous solutions of fluorescein 0.80 ; rhodamin Scarlet G 0-50; rhodamin 5 G 0.37; rhodamin B 0-25; eosin bl. 0.15; acridine red 0.08; erythrosine <0.02 ( ?) ; solution of fluorescein in methyl alcohol 0.74 ; solution of fluorescein in ethyl alcohol 0.66; solution of Magdala red in ethyl alcohol 0.54; solution of uranium in uranium glass 0.23. The approxim- ation of K to unity in the case of a typically fluorescent dye such a; fluorescein would not be anticipated from the classical theory of thermal absorption. The respective maxima of the incident and fluorescent radiation were separated in all cases by about 230 A.The value of K diminished rapidly as the strength of the solutions employed increased. The aut'hor considers that Wood's conclusion that in the case of sodium vapour K is approximately equal to unity is unjustified. Cathodo-luminescence of Solid Solutions of Forty-two Metals. T. TANAKA ( J . Opt. SOC. Amer. 1924 8 287-318).- Extensive observations on the cathodo-luminescence of mixtures of various salts are described. The materials were excited by cathode rays in a vacuum tube actuated a t about 4000 volts by a Holst machine. The luminescence was examined by a Hilger spectro- photometer. Between 0.1 and 1% of the substance under examin- ation was mixed with a " base " (calcium carbonate fluoride sulphate barium sulphate cadmium sulphate etc.).A series of A. A. E. Fluorescent Efficiency of Solutions of Dyes. J. S. G. T.GENERAL AND PHY SICBL CHEMISTRY. ii. 289 the luminescent sulphides prepared by Lenard and Klatt was also examined. Three types of luminescence were observed characterised by (1) fine line-like bands (2) groups of broad bands the members of each group overlapping and (3) a whitish colour without any distinct maxima. Whatever base is used the metal of the added salt shows a characteristic series of bands although a change of base may produce a slight shift of the bands. Each metal has from one to four characteristic band series; most frequently there are two such series. For each metal there exist two characteristic frequency intervals totally inde- pendent of the base and the mean of these intervals decreases quite regularly with increase in the atomic weight of the metal. Of the metals examined the active metals of Lenard and IClatt's sulphides and the rare earths gave the brightest luminescence.S. B. The following general conclusions were drawn. Phosphorescence of Fused Transparent Silica. W. E. CURTIS (Nature 1924 113 495 ; cf. Chapman and Davies this vol. ii 220).-Experiments on the production and destruction of the phosphorescence of silica discharge tubes indicate that it is due to the ultra-violet radiations (constituting the " Lyman " series) of atomic hydrogen. A. A. E. Cherni-luminescence and Energy Changes in the Oxidation of Phosphorus. A. PETRIKALN (2. Physik 1924 22 119- 126; cf. A.1912 ii 709).-The ultra-violet bands in the glow of phosphorus have been photographed. Five groups of bands are described with "cent'res of gravity" a t 3270 2600 2530 2460 and 2390 A. These agree in wave-length and intensity with bands in the arc spectrum of phosphoric oxide. The production of ozone which accompanies this oxidation is discussed from a thermocllemical point of view and it is concluded that the phenomenon cannot be photochemical. S. B. Spectral Emissive Power and Melting Point of Tungsten. A. G. WORTHING (2. Physik 1924 22 9-17).-The emissive power of tungsten as determined by the reflection method is 0.465 and 0.453 for 6650 A. a t 300" and 1330" abs. respectively and 0.507 and 0.493 for 4670 A. a t the same temperatures. From these data the melting point of tungsten can be calculated by Wien and Planck's law.Previous determinations of this constant are tabulated and the most probable mean is given as 3645" or 3665" abs. (according to the value assumed for the constant c2 in the Wien-Planck formula) with an experimental error of 20". S. B. Angular Momentum and Electron Impact. P. M. S . BLACKETT (Proc. Camb. Phil. Xoc. 1924 22 56-66).-4 theory of the detailed mechanism of ionisation or excitation by electronic impact. Assuming that the total energy of the system consisting of one atom and one impacting electron is the same after as before fhe collision and that there is an exact conservation of both energy 11"ii. 290 ABSTRACTS OF CHEMICAL PAPERS. and linear angular momentum an upper limit is arrived a t for the probability of excitation by electronic impact for energies just greater than the excitation energy.The probabilities so predicted are in rough agreement with some results of Franck and Knipping (A 1921 ii 150) and of Franck and Einsporn (2. Physik 1920 2 18). A. E. M. Critical Potentials of Hydrogen in the Presence of Nickel Catalyst. A. W. GAUGER (J. Amer. Chem. Soc. 1924 46 674- 6SO) .-The ionisation and resonance potentials of the nickel- hydrogen complex are measured by methods previously described (Franck 2. Physik 1920 2 18; Olson A. 1923 ii 455). A specially prepared nickel plate with surface-adsorbed hydrogen is bombarded with electrons the radiation generated falls on a platinum plate and a photoelectric current is produced which is studied as a function of the accelerating potentials applied to the electrons producing the radiation.The graphs obtained by plotting the electrometer deflection against the accelerating potential reveal seven critical potentials (at voltages of 104 12.0 13.6 16.4 18.2 20.5 and 22.6). The first four of these are attributed to atomic hydrogen the fifth to dissociation of the nickel-hydrogen complex together with ionisation of one hydrogen atom and the last two to the presence of molecular hydrogen. It is considered that the hydrogen is activated (atomic) a t least momentarily and is concentrated in this condition a t or near the surface of the metal. Normal Potential Gradient in the Discharge through Gases. A. G~NTHER-SCHVLZE (2. Physik 1924 22 70-97).-The potential gradient in the uniform part of the discharge away from electrode disturbances is characteristic of the gas in the path of the dis- charge.Its value for a gas a t 1 mm. pressure a t 27" in volts per em. is called the '' normal gradient " for the gas. To obtain satis- factory measurements of this constant large plane parallel electrodes must be used in a vessel so large that the walls have no influence and the temperature of the gas and the cathode must be controlled. Measurements were made for the rare gases oxygen nitrogen hyhogen and water-vapour and for various gas mixtures. The iiormal gradient' was smallest in neon (0.493 volt/cm.) and largest for watsr (50 volts/cm.). From humps in the gradient-composition curves it is suggested that oxygen forms temporary associations with argon and helium in the path of the discharge.The mobility of the electrons at atmospheric pressure in helium neon and argon is calculated to be 1880 1330 and 590 cm.2/volt-sec. respectively. The degree of dissociation of the gases in the discharge is of the order of 10-8. s. 33. Mobilities of Electrons in Helium [Hydrogen and Nitrogen]. L. B. LOEB (Physical Rev. 1924 23 157-168; ef. Townsend A. 1923 ii 721).-For helium K=[7.57 x 108j (166+E .76O/p)]i whence for E,=O K,=22,000 ; hence the mean free path seems to have the value given by the kinetic theory. S. I<. T.GENERAL AND PHYSICAL CHEMISTRY. ii. 291 Corrected empirical equations (cf. ibid. 1922 19 24; 20’ 398) are respectively for nitlrogen K=[3-637 x lo5/( ll.9+E0. 760/p)] and for hydrogen K=4-32 x lO5/[55.2+E,(760,/p~~J. The quantity assumed to be the mobility is possibly au improperly averaged figure.A. A. E. Motion of Electrons in Nitrogen. H. B. WAHLIN (Physical Rev. 1924 23 169-177).-For low fields the mobility of electrons in nitrogen is 18,000 cm./sec. but the value decreases rapidly for higher fields. The results are in conditional agreement with Compton’s equation. The mean free path of electrons in nitrogen a t 760 mm. and in thermal equilibrium with the molecules is 0.000131 cm. or 2.7 times the value required by the kinetic theory. V. KONDRAT~EV and N. SEMENOV (2. Physik 1924 22 1-8).-Thermionic electrons with a defmite velocity were caused to collide with a stream of the salt molecules. The minimum ionisation potential for mercuric chloride and iodide is about 11 volts.The only negative ions detected were singly charged halogen atoms but it is stated that in HgI vapour HgI,’ HgI’ Hg’ and possibly I’ and Hg” are produced. A. A. E. Ionisation of Salt Vapours. S. B. Ionisation of Cesium Vapour and the Mobility of Electrons in the Bunsen Flame. B. T. BARNES (Physical Rev. 1924 23 178-188).-1n experiments to test Saha’s theory of the ionisation of metallic vapours at high temperatures (A 1920 ii 659) a dilute solution of a caesium salt was sprayed into an air-gas flame the conductivity being determined as a function of the concentration of the solution. The concentration of vapour in the flame was computed and the black body temperature measured. A com- putation of Saha’s equilibrium constant K and comparison with that given by Saha’s equation supported that author’s theory.Selected values for the mobility of electrons in the flame are 75 and 82 m./sec. for 1 v./cm. Electrical Properties of Flames containing Salt Vapours for High Frequency Alternating Currents. A. B. BRYAK (Physical Rev. 1924 23 189-194).-A study of the dielecttric properties of salted flames. Conductivity of [Salted] Flames for Rapidly Alternating Currents. H. A. WILSON and A. B. BRYAN (Physical Rev. 1924 Electron Emission from Metals as a Function of Temper- ature. 0. W. RICHARDSON (Physical Rev. 1924 23 153-155).- Polemical and mainly concerning priority. Electron Emission from Metals. S. DUSHMAN (Physical Rev. 1924,23,156).-A reply to the above. The value of Richard- son’s constant A in i=AT2e-botT is valid only for pure crystalline substances being less in the case of certain covered filaments. A.A. E. A. A. E. 23 195-199). A. A. E. ll*-2ii. 292 ABSTRACTS OF CHEMICAL PAPERS. This is in agreement with G. N. Lewis' interpretation of the third law of thermodynamics. A. A. E. 0. GOSSBIASN (2. Physik 1924 22 273-274).-1n continuation of the work of Schmidt and Walter (this vol. ii 13) the author finds that man- ganese chloride antimony chloride sodium fluoride sodium nitrite potassium iodide potassium chloride and cobalt chloride do not emit ions a t temperatures below 450". Within the same range of temperatures the iodide bromide chloride and nitrate of cadmium the iodide bromide and chloride of zinc the chloride and fluoride of aluminium ferric chloride calcium fluoride the iodide bromide and nitrate of ammonium sodium nitrate and potassium nitrate emit positive ions only.The conductivity of the respective vapours attains a maximum value after a short time and then decreases rapidly a t first and then slowly attaining finally an approximately constant value. The form of the time-conductivity curve is the same whether the salts are heated by a platinum or a nickel wire but ionic emission occurs a t a temperature about 30" lower with the former. Addition of iodine or bromine to the halide salts increases the emission whilst an excess of cadmium or zinc reduces the emission. The curve showing the relation of the applied E.M.F. to the current in the vapour has the customary form and positive ions appear to be first emitted when the applied E.M.F. attains a constant value independent of the salt employed of about 1.3 volts. V.P. BARTON (Physical Rev. 1924 23 337-344).-The change of elec- trical resistance produced by light was measured for cuprous oxide for the range 3900 to 6000 8. a maximum decrease being observed a t 4600 A. and for selenium between 4300 and 9300 8. the maxi- mum decrease being a t 7000 A. The change in potential of the substance produced by light when cuprous oxide or selenium formed one of the plates in an electrolytic cell (photoelectrolytic effect) measured over the ranges respectively of 2700-5500 8. and 2700-9000 pi. is for cuprous oxide a t a maximum a t 3500 A. (upper limit about 5600 8.) and for selenium at a maximum a t 5700 A.(upper limit about 9500 A.). The effect is positive and proportional in the former case to the intensity I and in the latter to I $ . The photoelectric wave-length limit is 2570 A. for cuprous oxide and 2260 A. for selenium. Determination of Dielectric Constants of Gases by High Frequency Method. E. C. FRITTS (Physical Rev. 1924 23 345-356) .-By the application of Hyslop and Carman's met hod (ibid. 1920 15 243) of heterodyne beats for the measurement of small capacities the following values reduced to 1 atm. a t 0" and accurate to about 1% of the difference from vacuum were obtained for the dielectric constants hydrogen 1.000263 ; oxygen 1-000507 ; nitrogen 1.000555 ; air 1-000540 ; carbon dioxide 1 *000948. A. A. E. Emission of Positive Ions by Heated Salts.J. 8. G. T. Light Sensitivity of Cuprous Oxide and Selenium. A. A. E.GENERAL AND PHYSICAL CHEMISTRY. ii. 293 Electrical Resistance and Thermo-electric Power of the Alkali Metals. C . C . BIDWELL (Physical Rev. 1924 23 35i- 376) .-Determinations of the electrical resistance and thermoelectric power of the alkali metals between - 183" and 250" indicate trans- formations in each case from an a- to a p-form occurring a t approxi- mately the following temperatures lithium 50"; sodium -20" to +20" ; potassium -120" ; rubidium -35" ; caesium -80". In all cases a sharp rise in thermoelectric power and resistance occurs near the m. p. Extrapolated values for the temperature coefficient of resistance a t 0" are lithium 0.01404; sodium 0.00758 ; potassium 0.00604 ; rubidium 0-00470 ; caesium 0.00311 ; the values thus decrease smoothly with increasing atomic weight.The atomic heat of electrons in the alkali metals is computed to be approximately 0.24 cal. for sodium or potassium and 0.58 cal. for rubidium or cmium. E. WILSON and E. F. HERROUN (Proc. Roy. Xoc. 1924 A 105 334-3&).-The effects of impressed voltage temperature compressive stress and mag- netisation on the electrical conductivity of some varieties of magnetits have been examined. I n general the resistance was found to vary with the impressed voltage the resistance decreasing as the potential difference was increased. The effect of heating the specimens and subsequently cooling them to the ordinary temperature was to diminish both the resistance and the temper- ature coefficient whilst the application of magnetic force or com- pressive stress also caused a decrease in resistance.A specimen with high magnetic retentivity behaved exceptionally its resistance being scarcely affected by variation in the applied potential difference or by alternating changes of temperature ; furthermore the applic- ation of a magnetic field made relatively little difference in the resistance although this was diminished by compressive stress as in the case of the other specimens. Effect of Changes in Total Carbon and in the Condition of Carbides on the Magnetic Properties of Steel. A. W. SMITH E. D. CAMPBELL and W. L. FINK (Physical Rev. 1924 23 3'77-385) .-Normal magnetisation curves were obtained by the magnetic potentiometer method using the bar and yoke apparatus the carbon content of the steel being controlled by heating a t 930" in dry hydrogen with or without the presence of steel of a diEerent content of carbide.The minimum reluctivity (reciprocal of the maximum permeability) is a linear function of the carbon contient to 0.8% for hardened steels whereas for annealed carbon steels its rate of increase is somewhat greater. For annealed chromium steels the value is maximal a t about 0.5% and minimal at about O - S ~ . [Cf. B. 1924 425.1 A. A. E. Magnetic Susceptibility of Oxygen Hydrogen and Helium. A. P. WILLS and L. G. HECTOR (Physical Rev. 1924,23,209-220).- Values for K~ a t 1 atm. and 20" were obtained as follows oxygen +0.1447 x ; hydrogen - 1.64 x 10-lO ; helium -0.807 x 1G-lo.A. A. E. Electrical Conductivity of Magnetite. W. H.-R. A. A. E.ii. 294 ABS!CRACTS OF CHEMICAL PAPERS. Magnetic Researches. XXVI. Measurements of Magnetic Permeabilities of Chromium Chloride and Gadolinium Sulphate at the Boiling Point of Liquid Hydrogen in Alternat- ing Fields of Frequency 369,000 per Second. G. BREIT and H. K. ONNES (Proc. K. AEad. Wetensch. 1923 26 840-849).-1t has been suggested that at very low temperatures paramagnetic substances may show phenomena of hysteresis. To test this point the authors have measured the magnetic susceptibility of anhydrous chromic chloride and gadolinium sulphate a t the boiling point of liquid hydrogen and have compared the results with those obtained by the use of a direct field. The order of magnitude of the sus- ceptibility is unchanged when the frequency of alternation of the field is increased to 3*69x/lO5. The values obtained for chromic chloride and gadolinium sulphate in alternating fields are 0-75 and 0.51 respectively of the values for a direct field. The authors consider that the observed differences may be due to causes other than the change in the character of the applied field.J. F. S. Characteristics of Iron in High Frequency Rotating Mag- netic Fields. Electrostatic Nature of the Molecular Field. P. WEISS (Compt. rend. 1924 178 739-742).-Assuming an electrostatic moment to be superposed on the magnetic moment of an atom the electrostatic moment of the combination is calculated to be 0.86 x 10-18 for the atom of nickel and 1-25 x lO-l4 for the atom of cobalt.In the case of p-iron an ambiguity arises the moment being 1.09 x 10-l8 or 1-90 x lO-l* according as each atom or every third atom has a moment. For magnetite the moment is 1.74~10-~*. Debye’s calculation of this moment by a different method gave values between 0-34 x for many liquids (water lower aliphatic alcohols acetone etc.). The close agreement between the last figures and those for ferromagnetic substances supports the theory of an electrostatic origin for the molecular field. E. E. T. Magnetisation of Nickel-Chromium AlIoys above the Curie- Isotope Effects in the Band Spectra of Boron Monoxide and Silicon Nitride. R. S. MTTLLIKEN (Nature 1924,113 423- 424; cf. this vol. ii 3).-The spectrum of boron nitride observed by Jevons (A. 1915 ii 33) consists of two superposed spectra of BlOO and Bl10 respectively the bands being of the same pattern but the scale being larger for the lighter isotope.The possibility that the bands may be due to boron nitride is not however entirely excluded. Jevons’ observations (A. 1913 ii 813) on the bands of silicon nitride were also examined in order t o detect possible isotopy. Complete agreement exists between the experimental and theoretical values for the ratio of the vibration frequencies of Si29N and SimX with respect to Si2W; Si29 appears to be a little more abundant L. TONKS (Physical Rev. 1924 23 221-238). and 1.18 x point. J. gAFRANEK.-(See ii 342.)GENERAL AND PHYSICAL CHEMISTRY. ii. 295 than Sim and there is no evidence of isotopes other than Si2* in appreciable amounts.A. A. E. Spectroscopic Evidence of Isotopic Elements. H. NAGAOKA and Y. SUGIURA (Nature 1924 113 532-534),-1f as has been assumed there may be formation of pairs between atoms especially in the ionised state leading to the emission of non-series spectral lines a method of calculating the difference of wave-length due to isotopes is available. Application to known isotopes gives satis- factory results ; preliminary calculations also indicate that calcium contains a small quantity of Ca44 sodium a trace of Na21 whilst barium consists of Ba135 Ba13' BaI38 and Ba139 principally the last three. A. A. E. Isotopes of Merczry and Bismuth revealed in the Satellites of their Spectral Lines. H. NAGAOKA Y . SUGIURA and T. MISHIMA (Nature 1924 113 4 5 9 4 6 0 ) .-Accurate measurements have been made of the position of the satellites of mercury and bismuth lines.The remarkable resemblance of non-series lines in gold and mercury indicates that the nuclei may have something in common-possibly a hydrogen-proton is slightly detached from the central nucleus and quasi-elastically connected with it so that when excited by electro-magnetic waves they may make coupled vibrations. The difference in mave-lengths of such nuclear vibrations arising from different isotopes is calculated and the correspondence with the observed results is sufficient to show that the satellites in mercury and bismuth lines are due to the different isotopes. A. A. E. Isotope Effect as a Means of Identifying Emitters of Band Spectra Metal Hydrides.R. S. MULLIKEN (Nature 1924 113 489-490; cf. this vol. ii 3).-The considerations (a) that the spacing of bands in a band system should be less for the heavier isotope in approximately the ratio p of the respective molecular vibration frequencies and ( b ) that the spacing of the lines in a given band should be less in approximately the ratio p2 for the heavier isotope are applied to observations made by Frerichs (this vol. ii 77) on copper and by Kratzer (Ann. Physili 1923 71 102) on zinc cadmium and mercury. Satisfactory interpretation requires the supposition that the bands are of hydride origin; similar bands of silver gold and aluminium are probably due to hydrides whilst magnesium and calcium hydrides are known emitters of band spectra. Detection of Thorium Emanation (Thorom) in Thermal Springs by the Method of Induced Activity. A.LEPAPE (Compt. rend. 1924 178 931-934).-A number of springs in the Pyrenees and in the Central Plateau of France have been examined to ascertain their content of radium and thorium emanations. The former occurs in various proportions in all the waters but thorium emanation is found in traces only in those of the Central Plateau. Even in those cases in which thorium emanation is not detectable it is probable that the water comes into contact with this ernanation A. A. E.ii. 296 ABSTRACTS OF CHEMICAL PAPERS. but that the time elapsing between the termination of this contact c?nd the appearance of the water a t the outlet of the spring exceeds 10 minutes which is the duration of the life of thorium emanation.Radium Emanation in Air from the Schneeberg Mines. P. LUDEWIG and E. LORENZER (2. Physik 1924 22 178-185).- The appreciable-amount of emanation present in the air in various mines in the Schneeberg-Oberschlema district has been measured. It is the probable cause of a local lung disease. L. F. BATES and J. S. ROGERS (Proc. Roy. Xoc. 1924 A 105 360-369).-Three different sources of polonium have been examined by the scintilla- tion method for the emission Df long-range particles and it has been found that for every 107 cc-rays of range 3.93 em. emitted there are also present 98 51 and 26 particles of ranges 6.130.1 10.0&0.1 and 13.1 h0.2 em. respectively together with about 7 part'icles of longer range which are probably long-range hydrogen particles.Prom the brightness of the scintillations it is concluded that these particles are a-rays and from a consideration of the materials used it is shown that the rays must be emitted by a product of radium-D i.e. either by polonium or by a substance chemically similar to it. As the relative number of long-range particles agreed closely in all three specimens it is concluded that these particles are emitted by polonium itself as no B-ray evidence of branch products has been noted and no evidence of long-lived products similar to polonium has been obtained. W. H.-R. E. RUTHER- FORD and J. CHADWICK (Nature 1924 113 457).-On the assump- tion that the particles of disintegration are emitted in all directions relative to the incident a-rays a simple method has been devised whereby the disintegration of an element can be observed with certainty when the ejected particles have a range of only 7 cm.in air ; the presence of hydrogen in t'he bombarded material has then no effect nor do the long-range particles found by Bates and Rogers interfere. I n addition to the elements boron nitrogen fluorine sociium aluminium and phosphorus which give hydrogen particles of masimum range in the forward direction between 40 and 90 cm. t,he following give particles of range above 7 cm. neon magnesium silicon sulphur chlorine argon and potassium ; preliminary experi- ments with beryllium indicated a small effect. Hydrogen helium lithium carbon and oxygen give no detectable effect beyond 7 cm. The sulphur nucleus is clearly nat built up solely of helium nuclei.Preliminary experiments with indefinite results have been carried out with the elements from calcium to iron; no effects were observed with nickel copper zinc selenium krypton molybdenum pa.lla- dium silver tin xenon gold and uranium. Explanation of the Theory of the Rotation of the Atomic Nucleus. V. H. HENSTOCK (Chem. News 1924 128 84-67; cf. A 1933 ii 400 477 679 845).-With the aid of diagrams T. H. P. S. B. Particles G€ Long Range from Polonium. Bombardment of Elements by or-Particles. A. A. E.GENE= AND PHYSICAL CHEMISTRY. ii. 297 the author's view of the constitution of ozone sulphur dioxide and trioxide and phosphorus oxychloride is depicted. Consider- ations of polarity lead to the following gmeralisations (i) When two or more atoms unite the bonds uniting them are of opposite polarities in each atom.(ii) Normal stable compounds are formed only where each atom is united to its neighbours by single bonds which are all positive or all negative throughout each atom. (iii) Normal stable compounds are formed only where atoms are united by double or triple bonds having positive and negative bonds corresponding in each atom which may be in conjunction with single bonds normal to ea9h atom. (iv) Neutral atoms like carbon may unite normally by either positive or negative bonds but the tendency will be for these compounds to be the more stable where bonds are all of one polarity in each atom. K. vox H. COLLINS (Chem. News 1924 128 230-235) .-From considerations of the relative volume and heat of formation of organic compounds containing oxygen an attempt is made to show that the atom of oxygen has the constitution Li-H-H3-H-H3-H.The presence of two " free " helium portions causes it to be a non-metallic element and since the mass of the metallic portion is not greater than that of the rest there are left only t'wo electro-negative valencies emanating from the two metallic portions. Neutral oxygen is formed from acidic oxygen by the complete absorption of the two helium portions by the metallic part and alcoholic oxygen by the withdrawal of only the H port ions. A. R. P. Structure of Neon and Argon. H. COLLINS (Chem. News 1924 128 81-83).-Numerical considerations are held to indicate fhat neon and argon atoms are both constructed of lithium and hydrogen atoms in a manner described. Periodic Law of Atomic Radii.W. 9. DAVEY (Physical Rev. 1924 23 318-321).-"The ratio of the radii of any tm7o simple atoms belonging to the same vertical column in the periodic table is the same as tlhe ratio of the radii of any other two simple atoms on the same horizontal lines of the table provided that these atoms also belong to a common vertical column." The periodic relation- ship is conveniently expressed in the form of a curve obtained by plotting the atomic radii as a function of the number of electrons in the outer shell. Atomic radii are predicted (to -J= 2%) as follows phosphorus 1-08 ; sulphur 1.03 ; scandium 1-73 ; rubidium 2.49 ; strontium 2.16 ; yttrium 1.91 ; niobium 1.45 ; czsium 2.85 ; barium 2.48 ; lanthanum 2-19.A. A. E. New Theory relative to Molecular Constitut.ion of Chemical Compounds. C. BULOW (Cmpt. rend. 1924 178 770-771).- The author states that in 1919 he put forward views similar to those of Armstrong and of Kling and Lassieur (A. 1923 ii 542 543; cf. Auger this vol. ii 146). A. A. E. Physical Constants of Stereoisomeric Compounds. Structure of Oxygen. ACWERS and B. OTTENS.-(See i 513.) A. A. E. E. E. T.ii. 298 ABSTRACTS OF CHEMICAL PAPERS. Atomic Structure and the Relationship of the Chemical Elements. P. P. WORLEY (Rept. Austral. Assoc. Adw. Sci. 1923 16 21%219).-An electronic theory of valency. The four valency electrons in the carbon atom are supposed to revolve in planes parallel to the faces of a tetrahedron two revolving in a clockwise and two in an anti-clockwise direction.The valency bonds are largely magnetic. The external structure of other elements is arrived a t by modifying the different faces of the carbon atom. Nitrogen oxygen fluorine neon and other atoms are discussed. M. S. B. Calculation of Electrostatic Potential and Energy for Di- and Quadri-polar Lattices. H. KORNFELD (2. Physilc 1924 22 2743).-General formulce are deduced for the electrostatic poten- tial and energy of di- and quadri-polar space lattices and the energy of two special cubic lattices. 31. YAMADA (Physikal. Z. 1924 25 52-56).-A continuation of a previous paper (ibid. 1923 24 364) in which the surface energy is calcu- lated from the work necessary to overcome the electrostatic forces across the cleavage planes.Surfaces inclined to the crystal axes are dealt with in this paper. H. MARK and M. P~LBNYI (2. Physik 1924 22 200).-Corrections to the diagrams and letter- press of a previous paper (A. 1923 ii 768). H. OTT (2. Physik 1924 22 201-214) .-Examination of crystals of aluminium nitride by the Laue method of X-ray analysis indicates that the crystals belong to the hexagonal system. Investigation by the rotating crystal and powder methods shows that the aluminium and nitrogen atoms are separately distributed in close spherically packed hexagonal lattices slightly compressed in the direction of the six-fold axis of symmetry and displaced in the same direction by an amount determined by the parameter p =0.38. The basal axis a is of length 3.11 B. and the value of the axial ratio c is 1-60,.The basal group comprises two molecules and each atom of either kind is surrounded tetrahedrally by four atoms of the other kind arranged a t a uniform distance of 1.89,8. The respective intensi- ties of the lines in the X-ray spectra afforded no evidence of a possible polar structure of the crystals. G. E. M. JAUNCEY and H. L. MAY (Physical Rev. 1924 23 128-136).- X-Rays from a molybdenum target were allowed to pass through a thin crystal slab of rock salt. The curve obtained by plotting the scattering angle against the intensity has two maxima a t 15" and 30° respectively. Two maxima are also exhibited by the curve showing the relation between the spectra of scattered radiations and the electrometer deflection ; the scattered rays are much harder than the primary rays.The experimental values correspond fairly mell with Compton and Jauncey's quantum theory of scattering. S . B. Surface Energy of Crystals and Crystal form. A. E. 31. Space Lattice of White Tin. s. B. [Crystal] Lattice of Aluminium Nitride. J. S. G. T. Intensity of the X-Rays scattered from Rock Salt.CEXERAL AND PHYSICAL CHEMISTRY. ii. 299 The mass scattering coefficient for rock salt for the primary rays used (04-0.7 A.) is 0.23 & 0.02 as compared with Thornson’s value of 0.193. Refraction of X-Rays in Pyrites. B. DAVIS and R. VON NORDOFF (Proc. A’at. Acud. Sci. 1924 10,&63).-The bending of the molybdenum K,- and KB-rays by refraction has been measured for pyrites. The effect has been greatly increased by grinding and polishing the crystal so that the surface makes an angle with the natural cleavage planes. The mean value of the bending of the rays 6 is 3.33 x 10-6 for the K line and 2.82 x 10-6 for the KB line which are in good agreement with those calculated from the disper- sion formula of Lorentz (“ Theory of Electrons,” 1909 159).It is impossible to determine however whether the resonance effects required by the Lorentz theory are present or not. A. E. 31. Lipid Crystals Soap Solutions and X-Rays. J. W. MCBAIN (Nature 1924 113 534).-Terminological. A. A. E. The Scattering of X-Rays and Bragg’s Law. G. E. M. JAUNCEY (Proc. Nut. A d . Xci. 1924 10 57-59) .-i-4 theoretical paper in which it is concluded that the experimental deviations from Bragg’s law cannot be explained by a change of wave-length on reflection but that the theory of Darwin (Phil.Hug. 1914 27 318) requiring that a crystal should have a refractive index for X-rays provides the only available explanation of these deviations. A. E. 11. Excitation Reflection and Utilisation in Crystal-structure Analyses of Characteristic Secondary X-Rays. G. L. CLARK ( J . Amer. Chem. Soc. 1924 46 372-384; cf. A 1923 ii 468 469 856) .-A continuation of previous work. The characteristic secondary X-rays of iodine from the 100 planes of potassium iodide and the 0001 planes of iodoform have been measured through several orders. Uranium nitrate hexahydrate is found to reflect secondary L-series uranium rays. The wave-lengths of these secondary spectra are identical with those of characteristic lines in the spectra of target elements but the relative intensities of the lines vary con- siderably from the characteristic primary rays.The reflection is only partly in accordance with the Bragg law nX=2d sin 8. The secondary spectrum furnishes an easy way of obtaining accurately the distances between successive planes of atoms. In the case of iodine in KI d=3.52x10-8 cm. and in iodoform 3-737 x 10-8 em. The excitation and reflection of the characteristic secondary X-rays is considered not to be due to resonance but to the transfer of radiation momenta in quanta {cf. Mie 2. Physik 1923 15 56). By using finely divided crystalline materials as scattering radiators no change in the wave-length of the scattered rays was observed (cf. Compton A. 1923 ii 280). On the other hand with highly polished thin sheets of met,als e.g.copper or silver tertiary X - radiations due to photo-electrons from both the K- and L-rings were noticed. These tertiary rays may account for the phenomena A. A. E.ii. 300 ABSTRACTS OF CHEMICAL PAPERS. observed by Compton and Ross. They have a frequency which is the difference between that of a primary ray and of the critical absorption frequency characteristic of the element of the secondary radktor. H. T. G. L. CLARK and W. DUANE (Xcience 1923 58 400-402).-1n view of the experiments of Wyckoff (A 1923 ii 743) who failed to obtain any indication with potassium iodide of '( abnormal X-peak reflection," the authors have reinvestigated the phenomenon. Using both the ionisation spectrometer and the photographic method they have confirmed their previous conclusions.The success is ascribed to the use of a constant voltage of 75,000 volts instead of an alternating voltage of 50,000 volts to the employment of a thicker crystal and to the use of a single pin-hole instead of the ordinary arrangement of two to define the beam of X-rays incident on the crystal. H. S. ROBERTS (PhysicaZ Rev. 1924 23 38&395).-After recrystal- lisation the following m. p. were determined with calibrated thermo- couples potassium dichromate 397.5"; 45 KC1+55 Na2S0 by weight 517.1" ; 30.5 NaC1+69.5 Na,SO by weight 627-0" ; potass- ium chloride 770.3" ; sodium chloride 800-4" ; sodium sulphate 884.7" ; potassium sulphate 1069.1" (a11&0.5") ; calcium borate Ca,B,O 130~4."&5~.The transformation point of potassium sulphate is 583"&1". A. A. E. M. M. GREEN ( J . Amer. Chern. Xoc. 1924 46 544-545).-Assuming that A the latent heat of vaporisation is a function of temperature only and has the same form for all non-associated liquids then Trouton's rule leads to Guldberg's relation (A. 1890 1043) according to which the ratio of the boiling point to the critical temperature is constant. Within the limits given by the boiling point and critical temperatures the dependence of L on T is therefore probably the same for all non-associated liquids. S. K. T. R. LORENZ and W. HERZ (2. anorg. Chem. 1924,132 36-40).-The values of p=(n2y l)!(n2+2) have been calculated for thirty-one substances in the liquid and gaseous states a t their respective boiling points.There are no obvious relations between the values of qe (liquid) and qq (gas) for the substances examined. Pressure-Volume-Temperature Relations for Gaseous Ethyl Ether. J. A. BEATTIE ( J . Amer. Chern. Xoc. 1924 46 342- 357).-The isometrics for ethyl ether over the temperature range 175" to 325" and from specific volumes of 2.5 to 35 C.C. per gram have been accurately measured. There are practically straight lines from the largest specific volumes down to 15 C.C. per gram; a t greater densities marked curvature becomes apparent presum- ably due to association or dissociation of the moleculajr species. The equations of Keyes (Proc. Nat. Acnd. Xci. 1917 3 323) hold Abnormal Reflection of X-Rays by Crystals. A. A. E. Standard Melting Points at High Temperatures.Heat of Vaporisation a Function of Temperature. Volume in the Liquid and Gaseous States. H. T.GENERAL AND PHYSICAL CHEMISTRY. ii. 301 with great accuracy a t lower pressures but do not give satisfactory results in the regions of greater density. Equilibrium in the Liquid-Saturated Vapour System Heat of Vaporisation Law of Rectilinear Diameter and Molecular Attraction. V. KARPEN (Bull Acad. Xci. Roumaine 1923 8 17&180).-The slow vaporisation of a liquid is considered as the diffusion of liquid molecules a t pressure p9 through a thin diffusion layer a t the surface of the liquid into the vapour a t pressure p2. By assuming that the molecular heat of vaporisation L is propor- tional to Ad the difference in the densities d and d of the liquid and vapour phases and adopting p(w-b)-RT as the equation of state the expression L=RT .Ad/dc(l-bdc)2 is deduced the suffix c referring to the critical state. This equation is simplified to Ldc=5T .Ad or Ad/L=const. by taking 2-7dC=b-l. Values of L calculated from this formula agree with the observed values within about 2% but greater divergences are shown by methyl and propyl alcohols. The value of Ad/L for water increases with temperature. Assuming 2dC=b-l the law of rectilinear diameter is deduced in the form ~(d,+-d2)=dc+dc[(17,-T)/Tc] for the region of the critical point. Observed results for 22 substances out of 28 (e.g. argon ether benzene chlorobenzene ethyl acetate) conform to the formula fairly well but hydrogen shows a large deviation. When d is negligible in comparison with d and the latter can be considered constant d 2 = e [ ~ @ T ~ 1 a and b being constants a formula which has been verified exmrimentalIv for water (30- H.T. 150") to within -J=7:/,. 105 734.) S . K. T. (Cf. Chapkan and AYppleby T. 1914 Solution and Molecular Attraction. V. KARPEN (Bull. Acad. Xci. Roumaine 1923 8 180-185) .-Considerations similar to those outlined in the preceding abstract have been applied to solutions of gases in non-volatile solvents. Assuming Boyle's law to hold in the layer separating the gaseous and liquid phases Henry's law is derived in the form p2=p1 e-L'RT (!) (L=molar heat of solution of gas ; pl p2=pressures of free and dissolved gas respectively). If L is independent of T t?hen T.log (p1/p2)=const. ; this is shown to hold approximately for ammonia but not for sulphur dioxide. The theory is applied to solutions of solids in liquids and the limiting heats of solution are calculated from equation (l) p1 and p referring to the osmotic pressures of the dissolved substance.For lead chloride (25-100") L=3360 cals. and for potassium chloride (20-100") L=920 cals. The expression p1,/p27e4Z RT is obtained for the partition coefficient (pl,. p2=osmotic pressures of the solute in the two sol- vents; AL=dlfference in molar heats of solution of the solute in the solvents). Boyle's law is assumed for each solution as a first approximation The relation shows that T.log p1/p2 should be constant if the composition of the phases is independent of the temperature. This has not been verified experimentally although p l / p 2 is constant a t constant temperature. S.K. T.ii. 302 ABSTRACTS O F CHEMICAL PAPERS. Constitution of Solids. V. KARPEN (Bull. A d . Sci. Roumaine 1923 8 190-197).-Solids are regarded its hetero- geneous mixtures of a solid and a liquid phase the latter occupying pores disseminated throughout the former. An equilibrium similar to that between a liquid and its vapour (cf. preceding abstracts) is set up between the molecules of the solid crystalline phase and those of the liquid phase. It is shown that many well- known phenomena can be satisfactorily interpreted in terms of the author’s hypothesis e.g. increase of ductility with temperature supra-cpnductirity etc. It permits of the calculation of latent heats of fusion the value obtained for ice being 85.7 as compared with the experimental value of 80. This is regarded as in favour of the theory.F. ECKERT (Juhrb. Radio- nctiv. Electronik 1923 20 93-275) .-A comprehensive survey of the results obtained in the whole field of research on the physical properties of glass. Generally speaking the subject is treated with reference to the effect of the constituent oxides and where the properties are approximately additive in character (specific heat thermal expansion density etc.) the factors for the individual oxides obtained by various workers are quoted. Deviations from the additive relationship are considered in the light of constitutive effects. In conclusion a short summary is given of the prevailing theories as to the nature of glass; of interest in this connexion are the data obtained for glasses over the softening range.[Cf. B. 1924 335.1 A. COUSEN. S. K. T. Physical Properties of Glass. Solutions Considered as Binary Mixtures of Liquids. G. G. LONGINESCU and I. N. LONGINESCU (BuE. SOC. Romdncr. Xtiin. 1923 26,3-18) .-Solutions of gases liquids and solids may all be considered as binary mixtures of liquids a dissolved gas being regarded as liquefied by the pressure to which it is subjected in the solution whilst a dissolved solid undergoes fusion. The most soluble gases dissolve in water with evolution of heat which is regarded as due to the latent heat liberated on liquefaction and it is stated that no gas is soluble above its critical temperature or below the critical solution temperature for the liquefied gas and the liquid solvent.If such solutions are thus considered as binary mixtures of liquids it is shown that the density of a solution may be expressed by (cd+c’d’)/100 where c and c‘ are the percentage concentrations cl and d’ the densities of the two liquids. A more accurate expression can be obtained if it is assumed that the density of the dissolved substance varies with the concentration so that the term d may be replaced by a term dO(l+K1c+K2c2) where do K and Kz may be determined empirically from the densities of three solutions. It is shown that this equation affords accurate t-alues for the densities of aqueous solutions of ammonia hydro- fluoric hydrochloric hydrobromic and hydriodic acids sulphur dioxide glycerol sugar sodium sulphate and copper sulphate ; it is also applicable to mixtures of alcohol and ether.Tables show-GENERAL AND PHYSICAL CHEMISTRY. ii. 303 ing the densities of the above solutions and the values of do K and K2 are given for fifty-two substances. The equation may be employed to calculate the densities of substances in the free condi- tion from the densities of their solutions. W. H.-R. Internal Friction of Liquid Systems. W. HERZ and E. IC1Imm (2. anorg. Chem. 1924 132 41-53).-The relative merits of the formulae +=a+bT (Meyer and Rosencranz 2. physilcal. Ghem. 1909 66 238) +=a+bv (Batchinsky A. 1913 ii 928) and +=m-n log (Tc-T) (van Aubel A. 1921 ii 575) for the repre- sentation of the internal friction of liquid systems have been com- pared by reference to the data for a large number of organic liquids.This comparison shon7s that the formulz of Batchinsky and van Aubel correspond more exactly with experiment the latter giving slightly better agreement than the former. I n addition the authors have investigated the influence of temperature on the viscosity of (1) a number of aqueous salt solutions (2) water-alcohol mixtures and aqueous-alcoholic solutions of potassium iodide and (3) pyridine and pyridine solutions. The applicabilit'y of the above formulz to these observations has been examined H. T. Aqueous Solubilities in Homologous Series. H. FUHNER (Ber. 1924 57 [B] 510-515).-As a result of the examination of the solubilities in water of normal aliphatic hydrocarbons alkyl halides alcohols esters etc. the general rule is put forward that in homologous series of organic compounds the molecular solubility in water diminishes as we ascend the series so that each member is three to four times less soluble than the previous member.Figures are given for the approximate solubilities of a number of aliphatic derivatives in water. F. A. M. Solubility of Picric Acid in Mixtures of Liquids. E. ANGELESCU and D. DUMITRESCU (Bul. SOC. R d n a Xtiin. 1923 26 71-82).-The empirical formulz put forward by Bodlander (2. physilcal. Chem. 7 308) and by Hertz and Knoch (2. anorg. Chem. 46 193) are only applicable to a very limited range of con- centration. The authors have determined the solubility of picric acid in mixtures of alcohol and carbon tetrachloride alcohol and chloroform and of alcohol and benzene a t 12".The results obtlained are shown to be in agreement with the empirical equation Sc-Xo= KCP where if one liquid is considered as the solvent 8 is the quantity of picric acid which dissolves in 100 g . of the pure solvent 8 is the quantity which dissolves in 100 g. of the solvent plus c g. of the second liquid Le. in (lOO+c) grams of the mixture and K and p are constants. The range of validity of the equation depends on the nature of the components of the mixed solvent; it extends from pure carbon tetrachloride chloroform and benzene to mixtures of these cont.aining respectively 40,60 and 73 g. of alcohol in 100 g. of mixed liquid. The equation has been applied to results obtained by other workers and is found to be more satisfactory than other empirical equations.W. H.-R.ii. 304 -4BSTRACTS O F CHEMICAL PAPERS. Graphical Solution of certain Ratios in Temperature- Concentration Diagrams. C. 0. SWANSON (Amer. J . Xci. 1924 [v] 7 233-238).-The author gives a geometrical proof of the rules for determining from the f reezing-point diagrams of binary and ternary systems characterised by a single eutectic mixture the relative proportions of the solid and liquid phases. I n addition for ternary systems the ratio of the two constituents in the separated solid when the systems considered are cooled from a higher to a lower temperature is derived geometrically. Gibbs' method of representing the ternary system is employed and it is assumed that no complications resulting from the formation of compounds or solid solutions are involved.111. Absolute Measurements of the Specific Conductance of certain Potassium Chloride Solutions. H. C. PARKER and (MISS) E. W. PARKER ( J . Amer. Chem. Xoc. 1924 46 312-335).- The values obtained for solutions containing 1 0.1 and 0.01 equivalent per c.dm. a t 0" are 0.065098 0.0071295 and 0~000'774"84 ohm-1 cm.-l respectively. The conductivities for temperatures up to 30" have been calculated and compared with the older values of Kohlrausch and Maltby (A. 1900 ii 61 ; 1901 ii 82) and Kohlrausch Holborn and Diessel- horst (A. 1898 ii 366). The former are 0.145% and the latter about 0.15% too high. It is considered that the sum of all the errors in the new measure- ments does not exceed 0.02 %. Dilution Law for Uni-univalent Salts.B. SZYSZKOWSKI ( J . PhysicaZ Chem. 1924 28 193-2ll).-The author seeks to show that by adopting Ghosh's hypothesis (A. 1922 ii 125) it is possible to establish a theory of electrolytic dissociation which is in accord- ance with experiment'al data. The theory is developed from a thermodynamic point of view. I n deducing the expression for the intrinsic energy of the solution Ghosh takes into account only the attractive forces between oppositely charged ions because he looks on the ions as points or as volumes carrying uniform positive or negative charges. Such a view is permissible a t high dilutions but not when the ions are more closely packed for regarding the ion as a central positive nucleiis surrounded by negative electrons it is evident that the relative strengths of the repulsive and attractive forces coming into play between two ions will vary as the distance between the ions varies.The authors' formula takes into consideration both these forces and is found to be applicable to the conductivity data of Kohlrausch and his collabor- ators for uni-univalent electrolytes over a wide range of concen- trations. Salts undergoing hydrolysis exhibit deviations at high dilutions . According to the author's theory the decrease of equivalent conductivity with increase of concentration is exclusively due to the decrease of the ionic mobility. The theory of complete ionisation applies only to salts and not to 31. S. B. Calibration of Cells €or Conductance Measurements. H. T.GENERAL AND PHYSICAL CHEMISTRY. ii. 305 acids which must be assumed to be only partly ionised even when strong.This behaviour is attributed to the peculiar constitution of the hydrogen ions which have no exterior ring of negative electrons to exercise a repulsive force on the anion and so prevent combination. M. S. B. Electrolytic Dissociation of some Salts in Furfuraldehyde. F. H. GETMAN ( J . Physical Chem. 1924 28 212-220).-Experi- ments have been carried out to test the Nernst-Thomson rule in the case of furfuraldehyde. The electrolytes employed were the iodides of lithium sodium potassium and rubidium and also ammonium and tetraethyl ammonium iodides. By plotting the equivalent conductivity against the cube root of the concentration the curves obtained for the alkali iodides are approximately parallel ; the equivalent conductivity falls off far less rapidly with increase of concentration than is the case for the ammonium and substituted ammonium iodides.Walden’s relationship (A. 1921 ii 423) con- necting the viscosity and the dielectric constant of the solvent is found to be inapplicable to furfuraldehyde solutions. The fact that the electrical conductivities of the four alkali salts are in the inverse order of the molecular weights suggests that there is ionic solvation. The solvation of the cations has been calculated by Walden’s method (A. 1920 ii 230) on the assumption that the iodide ion is not solvated. The values are of the same order as those calculated for the same ions in aqueous solution (Washburn and Millard A. 1915 ii 311). The variation with dilution of the equivalent conductivity of the alkali iodides in furfuraldehyde is accurately expressed by Storch’s equation (A.1896 ii 288). I. hl. KOLTHOFF (Rec. tra??. ch im. 1924 43 207-215) .-Previously recorded values of the second dissociation constant of sulphuric acid K2 show little agreement. The hydrogen-ion concentration in a mixture of alkali sulphate and alkali hydrogen sulphate is given by [H’] = K,[HSO,’]/[SO,’’]. Its value has been determined colorimetrically by comparison with buffer solutions. The SO,”-ion concentration was obtained from conductivity measurements. The HS0,’-.ion concentration was determined on the assumption that the dissociation of sodium hydrogen sulphate is equal to that of uni-univalent salts. The average value obtained for K is 3.0 x 10-2 (with methyl-orange 2.8 x with tropaeolin 00 It is shown that methyl-red gives the most accurate results when strong acids are titrated in presence of sulphates. Strength of the Acid Function of the Oxygen Acids of Sulphur.I. M. KOLTHOFF (Rec. trav. chim. 1924 23 216- 220).-Jellinek has stated (A. 1911 ii 362) that the second dis- sociation constant K of thiosulphizric acid H,S20 is approxi- mately the same as that of sulphuric acid. This view is criticised and a redetermination of K has been made on the lines already hl. S. B. Second Dissociation Constant of Sulphuric Acid. 3 4 x 10-2). S. K. T.ii. 306 ABSTRACTS OF CHEMICAL PAPERS. described (cf. preceding abstract). Decomposition of the thio- sulphate had no perceptible influence on the measurement the mean value obtained being K = 5.7 x This result leads to the conclusion that sulphur has a stronger negative action than oxygen on the acid molecule.Similar measurements have been made with tetrathionic acid. The results which are not quoted indicate that this acid is one of the strongest dibasic acids. An experiment is described which shows qualitatively that the intro- duction of siilphur into the hydrogen sulphide molecule increases the acidity. S. K. T. at & 15". Activity Coefficients of Ions in Very Dilute Solutions. J. N. BRONSTED and V. I<. LAMER ( J . Amer. Chem. SOC. 1924,46 555-573).-The expressions given previously (A. 1922 ii 482) for the osmotic and activity coefficients have been extended to include salts of higher valence types.The ratio of the logarithms of two activity coefficients is very nearly equal to the square of the ratio of the corresponding valences; this is only exact in the ideal region. The equations have been confirmed experimentally by measuring the change in solubility of four sparingly soluble cobalt- ammines on the addition of electrolytes of various valence types. I n general the deviations do not exceed the probable experimental error. S. K. T. Activity of Water in Hydrochloric Acid. H. J. E. DOBSON and I. Missox ( J . Chem. SOC. 1924 125 668-676).-The partial vapour pressures of water vapour and hydrogen chloride in equi- librium with solutions of hydrochloric acid from 3N to 10.5N at 25" have been determined by a modification of the method used by Bates and Kirschman (A.1920 ii 88). No great accuracy is claimed for the partial pressures of the hydrogen chloride as the experimental conditions were designed primarily for the measure- ment of the aqueous vapour. The results are compared with those calculated from E.M.F. data by means of the Duhem equation and good agreement is obtained. L. J. R. Vapour Pressure of Hydrochloric Acid. J. S. DTJNN and E. K. RIDEAL ( J . Chem. SOC. 1924 125 676-684; compare precedingabstract and A. 1920 ii 88).-The partial vapour pressures of hydrogen chloride over its aqueous solutions from 0.3 to 5.3M at 25" have been measured by a distillation method. The con- centrations of the distillates were measured by conductivity and the ratio of the vapour pressures calculated on the assumption that both constituents obey the gas laws in the vapour state.The values of p H 2 0 were obtained from the data of Linhart (A. 1918 ii 28). fiom these measurements the activity of hydrogen chloride in its aqueous solutions was calculated and compared with the values deduced from E.M.P. measurements. Agreement is obtained up to 2M above which considerable divergence is found. The results are considered antagonistic to the theory of the total dissociation of strong electrolytes. L. J. H.GENERAL AND PHYSICAL CHEMISTRY. ii. 307 Reactions at the Interface of Two Immiscible Liquids and the Part Played by the Vapour of Each. Reaction between Water and Benzyl Chloride. G. HARKER ( J . Chem. Soc. 1924 125 500-511).-An attempt was made to compare the hydrolytic action of steam and liquid water on benzyl chloride in the hope of comparing the reactivity of hydro1 (simple H,O molecules) in the pure state (as steam) with that in complex liquid water.It was found however that under the conditions of the experiments steam impinging on the surface of the oil produced no hydrolysis. On the other hand the mixed vapour of benzyl chloride and steam impinging on the surface of water with due precautions against condensation was found to result in an amount of hydrolysis comparable with that obtained at the two liquid surfaces probably indeed identical with it after making certain corrections. The conclusion is drawn that at the interface of the two liquids the mixed saturated vapours are present. Relation between the Diffusion of a Gas Mixture and its Pressure.R. Wussow (Brennstofl-Chem. 1924 5 65-67).- The author utilises experimental results recorded by Fischer Schrader and Jager (A. 1923 ii 742) to test his formula giving the connexion between the partial pressures of the components in mixtures of gases before and after diffusion. The experimental data are in fairly close agreement with the calculated numbers. L. J. IT. T. S. W. Determination of Coefficients of Diffusion in Gels by means of Chemical Analyses and a Comparison of Results Obtained with those Yielded by the Indicator Method. C. E. T. MANN (Proc. Roy. Xoc. 1924 A 105 270-281).-The coefficients of diffusion of sodium potassium lithium ammonium magnesium calcium and barium chlorides in 2% agar-agar gel have been determined at 20" and 30".The chlorides were allowed to diffuse from normal aqueous solution into a cylinder of the gel contained in a glass tube successive layers of the gel being analysed after the experiment. The coefficients of diffusion calculated from these data are generally lower than those obtained by the indicator method. This difference is attributed to the fact that free ions diffuse more rapidly than undissociated molecules and consequently the coefficient calculated from the indicator method which measures the rate of penetration of ions is greater than the coefficient calcu- lated on the basis of chemical analysis in which the total quantity of salt dissociated and undissociated is considered. A relationship apparently linear has been found to exist between the coefficient of diffusion and the degree of dissociation of the solute the coefficient of diffusion being greater as the dissociation is bcreased.Con- sequently the coefficient of diffusion of electrolyte tends to increase with the dilution. W. H.-R. Rhythmic Formation of Layers. Liesegang 's Phenomenon. P. D. ZACHARIAS (KoZZ.-Z. 1924 34 37-40).-The author reviews the various explanations which have been put forward toii. 308 ABSTRACTS OF CHEMICAL PAPERS. explain the Liesegang phenomenon. As the result of observations on a natural water containing sulphur the author advances the view that the formation of Liegegang rings is to be attributed to the diffusion of particles of different sizes which have almost the same density as the dispersive medium. Although the formation of rings is affected by a number of factors the jelly structure appears to be of predominant import'ance.J. F. S. Adsorption and Distribution between Two Inmiscible Solvents. E. ANGELESCU and D. DUMITRESCU (Bul. SOC. Romtina Stiin. 1923 26 27-38).-Equilibrium is rapidly attained when a substance is adsorbed from solution and the solute may be regarded as distributed between two solutions the one existing only a t the surface of the adsorbing solid and the other comprising the main bulk of the solution the concentrations of the two solutions being connected by the equation C =. aC2n. Experiments show that a similar relation holds for the distribution of iodine between an aqueous solution of potassium iodide and organic solvents although in dilute solutions the partition of the iodine is in agreement with the simple law of distribution. If a given solution of iodine in potassium iodide is in equilibrium with solutions of iodine in carbon tetra- chloride chloroform and carbon disulphide of concentration C Cz and C3 respectively it is found that C C and C3 are in the same ratio as the solubilities of iodine in the three organic solvents concerned. Consequently it is possible to express the distribution of iodine between a solution of potassium iodide and the three organic solvents by one and the same equation CKl = AC1/n provided that the concentrations in the organic solvents are expressed as fractions of the corresponding solubilities.The experiments of Herz and Fischer (Ber. 1903 38 1138) indicate that the same equation applies to the distribution of acetlic acid betwcen water and aromatic hydrocarbons. W.H.-R. Adsorption of BinaryMixtures by Animal Charcoal and a Comparative Study of the Adsorptive Power of Different Varieties of Charcoal. N. A. YAJNIK and T. C. RANA ( J . Physical Chem. 1924 28 267-27S).-The adsorption of the follow- ing electrolytes by purified charcoal has been studied the chlorides of sodium potassium lithium magnesium strontium and calcium potassium and sodium hydroxides sodium carbonate and hydro- chloric nitric and oxalic acids. Changes of temperature produce very little change in the adsorption but increase of dilution causes increase of the relative adsorption. Experiments with pairs of the above electrolytes in solutions of varying concentrations show that there are threed possibilities in the adsorption of binary mixtures ( a ) the tm70 substances may be quite independent of one another indicating that adsorption depends very little on ionisation ; (b) the adsorption may be reduced by the presence of another substance this no doubt being the action in the poisoning of a catalyst; (c) the adsorption may be increased that of the other substance a t the same time either increasing or decreasing.No general rules can beGENERAL AND PHYSICAL CHEMISTRY. ii. 309 given by which the effect of one substance on another may be pre die t ed . The adsorptive powers of three varieties of wood charcoal (talia acacia and mulberry) have been compared with that of animal charcoal. For most substances mulberry charcoal appears to be the best of the three and for a few substances the values for adsorption by animal charcoal are approached and may even be exceeded by all three.M. S. B. Adsorption by Precipitates. VI. H. B. WEISER (J. Physicat Chem. 1924 28 232-244; cf. A. 1922 ii 262).-Experiments have been made to compare for different types of sols the precipita- ting values of mixtures of two electrolytes of similar and also of widely varying precipitating power with the precipitating values of the separate electrolytes. In addition the adsorption of the different ions from mixtures of electrolytes has been studied. In the presence of a smaller quantity of potassium chloride than is required for complete precipitation of the positive hydrated chromic oxide sol the amount of added potassium sulphate or oxalate necessary for complete coagulation is less than the calculated amount by about 20%.Similar results were obtained with negative hydrated stannic oxide sol when precipitated by mixtures of lithium chloride with barium or magnesium chloride. They are not in agree- ment with the conclusion of Freundlich and Scholz (A. 1922 ii 828) that the hydration of a colloid and of the precipitating ions results in ionic antagonism to such an extent that the behaviour of a colloid with mixtures of electrolytes may be used to determine whether it is hydrated or not. The precipitating values of mixtures of potassium sulphate and oxalate which have similar precipitating power are practically additive. I n the case of t,he negative arsenic sulphide sol which is regarded as not being hydrated very marked antagonistic action may be observed where the two electrolytes differ widely in coagulating powers e.g.lithium and barium chlorides lithium and magnesium chlorides and hydrochloric acid and magnesium chloride. As found for other colloids however the effect is almost additive when the precipitating powers are similar e.g. barium and magnesium chlorides. These phenomena are explained by the influence of each precipita- ting ion on the adsorption of the other by the sol. A strongly adsorbed ion will naturally affect the extent to which another ion in the same solution is adsorbed and hence the coagulating power of the latter. The reduction below the additive values where no antagonistic action is observed is accounted for by the well-known fact that adsorption is proportionately greater a t lower concentra- tions of the zdsorbed ion.The influence of the stabilising action of the anion in the precipi- tation of negative colloids or of the cation in the precipitation of positive ones is discussed. Contrary to the conclusions of Tartar and Gailey (A. 1932 ii,ii. 310 ABSTRACTS OF CHEMICAL PAPERS. 829) it has been shown that magnesium chloride may have the same precipitation value for widely different hydrogen-ion concentrations. J. BOHM and H. NICLASSEN (2. anorg. Chem. 1924 132 l-g).-Further experimental investigation has verified the conclusions reached by Haber (A. 1922 ii 553) with regard to the conditions under which crystalline or amorphous precipitates or sols are obtained.I n the case of substances with a low velocity of orientation e.g. the hydr- oxides of zirconium thorium and cerium the sols as well as the precipitated forms are amorphous but with the exception of precipitated Zr(OH) these become crystalline under the more favourable conditions existing a t higher temperatures. The cold precipitated hydroxides of aluminium scandium beryllium lanthanum and neodymium are completely amorphous but the sols produced under more favourable conditions i.e. a t low concentrations are crystalline. Strongly bipolar substances such as the halides of silver the fluorides of the alkaline-earths etc. are crystalline in both the precipitated and the sol forms. Osmotic Pressure of some Electrolytes of High Molecular Weight.E. JORPES and E. G. HELLGREN (Biochem. Z . 1924 145 57-62) .-The osmotic pressures of the ammonium sodium trimethylammonium triethylammonium and tripropylammonium salts of Congo red increase with increasing molecular volumes. This is explained by assuming that the large negative ion adsorbs on its surface the small positive ions and that this process becomes less complete with increase in the size of the latter e.g. in the homologous series of aliphatic amine salts investigated. Viscosity and Refractive Researches on the Coagulation of Lyophilic Colloids. N. SCHOORL (Rec. trav. chim. 1924 43 203-206) .-The viscosities of a 5 % solution of air-dried egg- albumin mixed with 24y hydrochloric acid were measured a t 15" in an Ostwald viscosimeter a t progressively increasing time intervals after mixing.The refractivities mere also measured. The object was to determine whether variations in the colloidal state of blood- serum could be followed by refractivity changes the colloid used in these experiments being similar t o blood-serum although the changes are slower. The decrease in refractivity during coagulation runs parallel with increase in viscosity. Both properties in their variation wit'h time afford evidence of an autocatalytic process and it is probable that either property may be used to detect vzriations in the colloidal state of blood-serum the refractivity being convenient when only small quantities are available. Hydrophilic and Hydrophobic Colloids and the Influence of Electrolytes on Membrane Potentials and Cataphoretic Potentials.J. LOEB ( J . Gen. Physiol. 1924 6 307-32S).-The depressing effect of neutral salts on the cataphoretic P.D. of hydro- phobic colloids (mastic Acheson's graphite or collodion particles) which have been stabilised by the presence of N/256-sodium chloride M. S. B. Amorphous Precipitates and Crystalline Sols. H. T. J. P. S. K. T.GENERAL AND PHYSICAL CHEMISTRY. ii. 311 and also of denatured egg-albumin in N/5000-sodium hydroxide solution depends on the valency and molar concentration of the cation and not on its nature. This does not hold for heavy metals e.g. silver nitrate has a greater depressing effect on the cntapho- retic P.D. than sodium nitrate. The influence of dilute acids on the cataphoretic P.D. of gelatin-coated collodion particles depends only on the pE and the valency of the acid used.These experi- ments agree with the view that the cataphoretic potential difference of protein particles is determined largely a t least by the existence of a Donnan equilibrium between the medium and thc surface layer of the particle and except where chemical changes are possible as in the case of heavy metals they accord with IVilson’s suggestion (“ The Chemistry of Leather Manufacture,” 1923 128) that in the case of hydrophobic colloids the effect may be due first of all to the attraction of the particle for the anion causing stabilis- ation followed by the operation of the Donnan equilibrium between the surface layer containing excess of anion and the bulk of the medium. W. 0. K. Process of Filtration. K.SCHAUM A. MOELLER and T. MARX (KoZl.-Z. 1924 34 l-l2).-The time required for a liquid to flow through a filter plate is approximately proportional to the thickness of the plate. I n order to obtain satisfactory support for the filter it is best to use a thin plate with large pores covered by a fine-pored overlayer instead of a thick plate with fine pores. The arrangement of filter layers of different pore size one after the other is not efficient since the time required for filtration by such a system is considerably greater than the sum of the times for the individual layers. The quantity of liquid flowing through a filter in unit time is proportional to the pressure above the liquid. For liquids of high viscosity the total time of filtration may be reduced by dilution of the solution.Apart from the fact that the thickness of the cake of filtered material increases it is found that the impermeability of a cake of given thickness increases with the time occupied by the atration. The retardation may be partly prevented by stirring during filtration or by removing the upper portion of the cake by means of a scraper. Filtration after sedimentation when the latter takes place rapidly is recommended as an efficient method. J. F. S. General Colloid Chemistry. VIII. Analysis and Con- stitution of Colloidal Gold. 11. M. ADOLF and W. PAULX (KoU.-Z. 1924 34 29-34; cf. A. 1923 ii 636).-A carefully dialysed gold sol contains particles the mean size of which corre- sponds with the cube edge = 27pp. Using Scherrer’s X-ray data this gives approximately 1,444,000 atoms to each colloidal gold particle and since there are 25 gold atoms to each charge there must be 57,000 negative charges on the surface of the cube.The number of gold atoms in the surface of the average cubic particle is therefore somewhat greater than 76,000. J. F. S. F. V. V. HAHN (KolL-Z. 1924 34 162-169),-The solubility of night blue in a number of Colloid Chemistry of Night Blue.ii. 312 ABSTRACTS OF CHEMICAL PAPERS. solvents is found to be connected with the dielectric constant of the solvent. When this constant is less than 5 the dye dissolves to a small extent only and gives green solutions; in solvents with a higher dielectric constant it is readily soluble and the solutions are blue. The solubility in mixed liquids is determined in a similar manner by the dielectric constant.Yet it is found that toluene in which the dye is only sparingly soluble will completely extract it from an aqueous solution. The residues obtained by evaporating solutions in different solvents show differences in solubility which indicate that the structure of the colloid has been modified by the solvent. Ultra-microscopical examination affords evidence of molecular dispersion in the green solutions whilst the blue solutions contain colloidal aggregates which vary in size in different solvents and are largest in an aqueous solution. Eight blue precipitated from aqueous solution by sodium hydroxide dissolves in toluene in the same way as the pure dye and does not appear to be a sodium salt.The surface tensions of toluene and of water are slightly raised by the addition of night blue and the interfacial tension between these liquids is doubled when the dye is added. Colloid Chemistry of Cellulose. H. WISLICENUS and W. GIERISCH (KoZZ.-Z. 1924 34 169-181) .-Purified cellulose (filter- paper) disintegrated by long-continued milling in the dry state in a closed porcelain ball mill and in a Dreef’s steel plate mill was found on microscopical examination to have lost all traces of fibrous structure and to have the form of minute irregular particles which especially when moistened tend to collect into spherical aggregates. The disintegrated cellulose is more soluble in water than the original fibre adsorbs methylene-blue more readily and is more rapidly hydrolysed by sulphuric acid.Its reaction with iodine is much the same as that of the start>ing material but the copper number of the heavily milled product is much higher suggesting that some chemical Influence of Electrolytes on the Swelling of Agar. S. DOKAN (KoZZ.-Z. 1924 34 155-161).-1n dilute solutions cations (hydrogen excepted) influence the swelling of agar according to the valency rule. In stronger solutions specific effects appear and in normal solutions of the chlorides the influence of the cation increases in the order Li Mg < Ca < Na < A1 < Ba < K < H. Anions in dilute solutions do not affect the swelling of agar but in stronger solutions exhibit specific effects. The order of action of the anions in normal solutions is I < CNS <NO < OH < Br < C1< OAc < SO < citrate.The influence of an electrolyte on the swelling of agar is regarded as the resultant of two effects (a) the lowering of the charge on the colloid by the cation and ( b ) the specific dehydrating effect of the Electrical Resistance of Soapwater Systems at the Setting- point. M. H. FISCHER (KoZL-Z. 1924 34 140--145).-Previous observations (A. 1923 ii 725 823) suggested a sharp increase in electrical resistance a t the temperature at which an emulsion of S. S. change has taken place. s. s. ions which becomes effective a t higher concentrations. s. s.GENERAL AND PHYSICAL CHEMISTRY. ii. 313 soap in water changes into an emulsion of water in soap. Experi- ments on N / 2 solutions of sodium and potassium soaps confirm this conclusion.The temperatures a t which the sudden increase in resistance takes place are for the sodium soaps stearate 60"; palmitate 62" ; myristate 34.5" ; laurate 16.5" ; oleate 7" and for the potassium soaps stearate 44"; palmitate 23.5". The curves for potassium myristate laurate and oleate indicate that the inversion points lie below 0". The solutions were allowed to cool without shaking since mechanical treatment would destroy the outer phase of hydrated soap and produce an emulsion of less hydrated soap in an aqueous outer phase. s. s. Physical Properties of Mixtures of Alcohol and Serum. R. FORTH and 0. B~ijrr (KoZZ.-Z. 1924 34 129-135).-With increasing additions of ethyl alcohol the dielectric constant optical activity and electrical conductivity of bovine serum a t first increase attain maxima a t 2 3 and 7 parts of alcohol per 1000 by volume respectively and thereafter decrease continuously.Conversely the viscosity curve exhibits a minimum a t 2 volumes of alcohol per 1000 whilst the surface tensions of alcohol-serum mixtures lie on a curve parallel to that for water and alcohol mixtures. The refractive index of the serum is at first unaltered by small additions of alcohol but increases linearly after 2 volumes per 1000 have been added. The results are discussed in terms of Drude's dipole theory. s. s. Electro-osmotic Experiments on the Reversal of the Electric Charge of Colloids and Precipitates and the Pre- paration of Stable Sols with a Charge Opposite in Sign to that Commonly Obtained.J. N. MUKHERJEE and B. C. ROY ( J . Chem. Soc. 1924 125 476-488).-Two types of adsorption by colloids are distinguished (a) primary adsorption of ions due to chemical affinity and ( b ) electrical adsorption of oppositely charged ions. It is supposed that the original charge of the surface is due to primarily adsorbed ions fixed on the surface as point charges and that oppositely charged ions can be held by electrical attraction just opposite these. Consequently a reversal of the charge of the surface is only possible by electrical adsorption when the valency of the oppositely charged ion is greater than that of those primarily adsorbed but if the oppositely charged ion is affected by chemical affinity of the surface atoms a reversal of charge is possible independent of the valency.It follows that the alkali-metal cations can never reverse the charge of colloidal sur- faces for these ions are univalent and their chemical behaviour is summed up in their tendency to exist as ions. Similarly the alkaline-earth metal ions are not likely to reverse the charge although they may do so by electrical adsorption if the primarily adsorbed ions are univalent. These views are confirmed by cata- phoretic and electro-endosmotic experiments dealing with the effects produced by the nitrates and chlorides of different metals and the hydrochlorides of organic bases on various sols. L. J. H. VOL. CXXVT. ii. 22ii. 314 ABSTRACTS OF CHEMICAL PAPERS. Effect of Hydrophilic Colloids on Size and Distribution of Particles in Electrolytic Precipitation.I. Gelatin and Basic Lead Carbonate. W. G. FRANCE and D. MCBURNEY ( J . Amer. Chem. Xoc. 1924 46 540-544).-The average size of basic lead carbonate particles electrolytically precipitated a t 20" is reduced to approximately one-third by the presence of 0.013y0 of gelatin. The decrease after the first addition is not so large being approximately proportional to the concentration of the gelatin added. The particles are of more uniform size when gelatin is present. Samples of dry commercial white lead contain particles which are larger and less uniform than those in the electrolytic product obtained in presence of gelatin Gelatinisation of Nitrocellulose Solutions. A. SZEGVARI (Roll.-Z. 1924 34 34-37).-Solutions of nitrocellulose when diluted with amyl acetate and benzine gelatinise on elevation of the temperature but 011 lowering the temperature again they re-liquefy.The temperature of gelatinisation depends on the composition of the liquid medium. This phenomenon is attributed to solvation and to variations in the relative affinity of the nitro- cellulose for the solvent and the diluent with change of temperature. Electro-viscous Effect in Solutions of Soluble Starch. H. G. B. DE JONG (Rec. trav. chim. 1924 43 189-202).-Specially prepared water sols of soluble starch (d20 1.0039) obey Poiseuille's law. The addition of small quantities of barium chloride causes a rapid decrease in the rate of flow of the sol in the viscosimeter (cf. A. 1923 ii 132); this is attributed to the disappearance of the electro-viscous effect.At higher concentrations a more gradual change occurs which can be considered as a lyotropic influence on hydration. The lowering of the initial viscosity is independent of the anion but increases with the valency of cation. These results are comparable with those obtained with agar sols (A. 1922 ii 357) from which it is inferred that the starch sol possesses a negative charge which is apparently below the critical value since the sol is so readily coagulated by excess of alcohol or by small amounts of tannic acid. Addition of hexaethylenediamine-hexoltetracobaltic nitrate [Cof( OH),Co en,),](N0,),,3H20 causes reversal of the charge of the starch sol and a t the isoelectric point hydration keeps the sol stable. With a partly dehydrated sol (addition of alcohol) neutral electrolytes cause complete precipitation but the above complex salt behaves irregularly a coagulation region (0.05-0.3 mg.equiv. per litre) being followed by a region of stability (0.3-2 mg.) and finally a t higher concentrations by coagulation. It was con- firmed by cataphoresis experiments that in the stable region the starch sol was partly hydrated and positively charged. The behaviour of the starch sol towards hydrochloric acid and sodium hydroxide confirmed the results of Samec (A. 1913 i 17 1155). The behaviour of the acid resembles that of neutral salts that of sodium hydroxide points to starch sols being true colloidal solutions. It appears as if the hydration increases considerably with increasing S. K. T. J. F. S.GENERAL AND PHYSICAL CHEMISTRY.ii. 315 concentration of sodium hydroxide (or calcium hydroxide) the phenomenon resembling the lyotropic effect of neutral electrolytes. In explanation it is supposed that the weakly acid starch sols are neutralised by sodium hydroxide the mobile hydrogen ions being replaced by the comparatively slow sodium ions. In this process the charge remains constant and hence the electro-viscous effect increases. Further addition of sodium hydroxide brings into play other factors which tend to remove the quasi-viscous properties. Ultramicroscopical Investigations with Unidirectional Illumination. The Ultramicroscopic Mie-effect. A. SZEG- VARI (2. Physik 1924 21 34S-357).-When a beam of light falls on a particle which is smaller in diameter than half a light-wave length the light is scattered in all directions but the intensity is greatest in the directlion of the primary beam.The latter phe- nomenon becomes more pronounced with increasing size of the particle. The author calls this the " Mie-effect '' (see Mie Ann. Physik 1908 [iv] 25 37). It has been shown that if unidirectional illumination is obtained in the ultramicroscope by means of the azimuth screen which is a screen with a small aperture allowing only a thin beam of light to fall on the colloidal solution to be examined (Szegvari Physikal. Z . 1923 24 91) one portion of the circular image obtained displays as a result of the Mie-effect a more intense brightness than the rest. By rotating the screen rotation of the luminous portion of the image is also obtained but no change in intensity is observed if the particle is spherical.This may be applied to the study of immovable colloidal particles such as those of a precipitate. In a similar way the appearance of scintillation phenomena when a colloidal solution showing Brownian movement is illuminated through an azimuth screen indicates that the particles are not spherical. Along with the azimuth screen a screen with variable aperture for the objective should also be used. By this means disturbing effects due to particles of larger size or to excessive density of particles in suspension may be eliminated. In order to take full advantage of the Mie-effect the direction of illumination should not be a t right angles to the direction of observ- ation but should approach a coaxial position.Contraction [of Volume] accompanying Formation of Compounds. F. EPHRAIM (Hell;. Chim. Acta 1924 7 298- 307; cf. Ephraim and Michel A. 1919 ii 268).-For the chlorides bromides and iodides of manganese iron (ferrous) nickel copper zinc cadmium and mercury the percentage contraction is nearly independent of the halogen although the iodides generally show a somewhat smaller contraction. Mercuric chloride is exceptional The chlorides and bromides of univalent copper silver and thallium show a similar relationship but the iodide is in each case different from the chloride and bromide on account of its different crystalline structure. A similar regularity appears to hold for the halogen compounds of phosphorus boron and antimony. The halide salts S.K. T. M. S. B. 12-2ii. 316 ABSTRACTS OE' CHEMICAL PAPERS. of magnesium calcium strontium barium lead and tin show no such regularities. The hexammines derived from the halide salt,s of nickel cobalt iron manganese and zinc show practically the same contraction as the corresponding halide salts of the alkali metals. It is sig- nificant that all these compounds in which the influence of the halogen predominates belong to the regular system. When the contraction is calculated from the volumes of the salt and ammonia the ammines of the zinc and manganese salts give values entirely different from the others. It follows that the lattice structure of the ammines must be entirely different from that of the free salts. Some of Biltz's conclusions regarding the formation and structure of compounds of salts with ammonia (this vol.ii 167) are criticised. There is no contradiction between Kopp's rule as applied to homo- logous series of organic compounds and the author's percentage contraction rule as Kopp's factors refer to already contracted radicals. E. H. R. Electromotive Forces of Cells and Molecular Attraction. V. KARPEN (Bull. Sci. Acad. Roumaine 1923 8 185-190). -By the introduction of the author's diffusion-layer theory (cf. this vol. ii 301) the Nernst electrode-potential formula is modified to E=(AZ/n)-(RT/n) log p l / p 2 AZ being the difference between the attractions of the metal ions in the metal and in the solution. No application of the second law of thermodynamics is made in the deduction of this equation which gives comparable values for the solution pressures of metals. The application of the revised formula to the E.M.F.of cells of the type Pt/saturated amyl alcohol solution of caustic soda/saturated. aqueous solution of caustic soda/Pt is discussed. A cell of this type with carbon electrodes has been found to yield a current uninterruptedly for 7 months with no apparent internal change. Influence of Salts on the Constants of the Silver Bromide and Silver Electrode. I. M. KOLTHOFF (2. anorg. Chem. 1924 132 117-124).-The value given in an earlier paper (A. 1922 ii 20) for the normal potential of the silver electrode referred to the normal hydrogen electrode at 18" has been corrected to ,Eh=0.801 volt. The dependence of the potential referred to the null electrode on the concentration can be expressed by -log[Ag']=pa,= (1-0780-x)/0.05768 a t 18" and in the case of the silver-silver bromide electrode -log[Br'] =pBr= (T-0.3570) /0.05768 a t 18".From these values it follows that the ionic solubility product for silver bromide = 3.4 x 1 O-13. The influence of uni- and multi-valent cations on the potential of the silver electrode and of uni- and multi-valent anions on the potential of the silver-silver bromide electrode have been inves- tigated. Ferro- and ferri-cyanides have a very marked effect but the others examined exert no perceptible influence. Oxidation Potentials at Mercury Electrodes. 111. Ter- and Bi-valent Titanium. G. S. FORBES and L. P. HALL (J. Amer. Chem. Soc. 1924 46 385-390).-A mixture of titanium S.K. T. H. T.GENERAL AND PHYSICAL CHEMISTRY. ii. 317 di- and tri-chlorides obtained by heating finely divided titanium of 99.9% purity in a stream of dry hydrogen chloride was intro- duced in the absence of air into a half cell with a mercury electrode and the E.M.F. of this combined with a calomel electrode was determined. By subsequent analyses the ratio of di- and ter- valent titanium was determined. The average value of gh for the reaction Ti"'+@ z Ti'' is -0-369 volt. This value is inde- pendent of the acid concentration provided the latter does not exceed 0.1N. At higher concentrations the dihalide is rapidly decomposed. By combining the above value with that for the reaction Ti""+@ =Ti"' (A.' 1908 ii 350) it follows that for the reaction Ti""+2@ Ti" &h=-0.16,t0-01 volt.H. T. Polarisation of Electrodes. 111. J. E. VERSCHAFFELT (Rec. truv. chim. 1924 43 125-134; cf. A. 1923 ii 115 369).-A theoretical paper in which mathematical reasoning is adduced to show (1) that the concentrations of anions and cations in a solution may be considered as equivalent; (2) the variation of electrode- potential with current density ; (3) the relationship between current density and electromotive force in an electrolyte which is dissociated into two ions the liquid being separated into three portions anodic cathodic and intermediate the last being continually stirred and (4) that the osmotic theory of the electromotive forces leads to the generalisation that the product of the ionic concentrations must be constant for saturated solutions.Overvoltage Phenomena. 11. Connexion between the Overvoltage of Hydrogen as Pure Metal Surface and certain Properties of the Metals. A. THIEL and W. HAMMERSCHMIDT (2. unorg. Chem. 1924 132 15-35).-The overvoltage of hydrogen at the surfaces of a large number of pure metals has been determined by a method similar to that previously used (A.' 1913 ii 104). There appears to be a connexion between the overvoltage of hydrogen and the position of the metal in the periodic classification metals of the first and eighth groups having low and those of the second third and fourth high overvoltage values. Of the metals inves- tigated manganese and vanadium both of which are difficult to obtain pure are the only "misfits.~y The so-called solubility theory for the explanation of overvoltage phenomena is rejected as untenable.Tafel's theory (A. 1900 ii 588) according to which the metal is supposed to accelerate or retard the spontaneous change 2H' + H2,. is considered the most satisfactory. An attempt is made to link up this theory with the results obtained by Sabatier according to which the most active metals in the process of hydrogenation are those with a low overvoltage and also with the work of R. W. Wood on the decomposition and re-formation of molecular hydrogen under the influence of an electric discharge. H. T. E. NEWBERY ( J . Chem. Xoc. 1924 125 51 1-518).-The velocity of decomposition of the overvoltage hydrides of certain metals (cf. A. 1917 ii 64 H. J. E. Life Period of Overvoltage Compounds.ii.318 ABSTRACTS OF CHEMICAL PAPERS. 290) has been determined using an intermittent current with variable commutator speeds. Those of zinc and chromium appear to be comparatively stable those of graphite silver and platinum less so whilst those of cadmium copper nickel and lead are either very unstable or variable in their composition or proportion on the electrode surface. The “ direct ” method of measuring over- voltage is criticised and the use of alternating current ammeters with intermittent unidirectional currents shown to be inadmissible. I n part polemical (cf. Glasstone A. 1923 ii 531). Depolarisation of the Chlorine Electrode by Organic Com- pounds. M. A. YOUTZ ( J . Amer. Chem. Xoc. 1924 46 545- 554) .-The influence of organic compounds on anode decomposition potentials has been examined.With carbon anodes in potassium chloride solution ethylene maleic oleic and 10% acetic acids show slight depolarising action. The current-anode potential curves do not rise smoothly as is the case when no depolariser is present. The phenomenon is reproducible with solutions of ferrous chloride but does not occur with hydrochloric acid solutions of the organic compounds and is often influenced by stirring. Phenol is oxidised to benzoquinone and 70% acetic acid in l.2N-hydrochloric acid gives chloroacetic acid although ordinary chlorine under the same conditions gives no trace of this substance. With carbon anodes toluene m-xylene mesitylene amylene or ether exerts no de- polarising action ; crotonic acid depolarises slightly and aniline very strongly.Strong solutions of acetone (50”/ and above) depolarise; weaker ones do not!. This is shown to be due to the differences in the free energies of hydrochloric acid in water and acetone solutions. The potential of a chlorine electrode (platinum) with an overvoltage of 0.5 is not reduced by addition of acetone acetic acid chloroacetone or chloroacetic acid. Neither method is suitable for deciding whether organic compounds can or cannot be chlorinated. S. K. T. L. J. H. Method for Studying the Rapid Absorption of Gases by Liquids. P. G. LEDIG and E. R. WEAVER (J. Amer. Chem. Xoc. 1924 46 650-657).-The rate of absorpt>ion in caustic soda solu- tions of small reproducible bubbles of carbon dioxide has been measured in a special apparatus.The absorption was measured by registering photographically on a rotating drum the movement of a mercury thread contained in a capillary tube connected with the absorption vessel and by recording simultaneously the vibra- tions of a tuning fork. The absorption of a fresh bubble proceeds in three stages and the maximum rate is not attained immediately. The rate of absorption depends on the size cf the bubble and the viscosity of the solution is shown also to have a great effect. Jn presence of an inert gas (air) the rate varies inversely as the square root of the carbon dioxide pressure. Sodium carbonate solution (2-53‘) absorbs carbon dioxide a t less than half the rate for pure water and less rapidly t,han a sucrose solution of the same viscosity. Mass action may exert some effect here.S. K. T.GENERAL AND PHYSICAL CHEMISTRY. ii. 319 Electrical Method of Determining the Velocity of Deton- ation of Explosives. J. E. P. WAGSTAFF (Proc. Roy. Soc. 1924 A 105 282-298).-For the measurement of the velocity of detonation of explosives a method has been devised in which the length of the column of the explosive need not be greater than 15-60 cm. The method involves the measurement of time intervals of the order sec. and depends on the rate of discharge of a condenser during the short interval required for the detonation of a column of explosive. The method has been tested by finding the time of impact of steel balls and the conditions necessary for high accuracy in the case of explosives have been investigated. The velocities of detonation of tetryl trinibrotoluene and other explosives have been determined the results obtained agreeing closely with those found with much longer lengths of material in the Mettegang apparatus (Mettegang Rep.of 5th Cong. Appl. Chem. vol. 2 327 and Kast " Spreng und Zundstoffe," 1025). W. H.-R. Explosion Limits and Ignition Temperatures of Com- bustible Liquids. H. STRACHE (Oesterr. (?hem.-Ztg. 1924 27 19-24) .-The paper contains vapour-pressure curves of several combustible organic compounds and a table showing their upper and lower ignition temperatures explosion limits with air and the calorific value of the air-vapour mixtures a t the ignition tem- perature and discusses the bearing of the composition and ignition temperature of petrol on its storage.[Cf. B. 1924 364.1 A. R. P. Influence of Cooling [Due to the Walls of the Container] on the Measurement of Explosive Pressures. H. MURAOUR (Mdm. Poudres 1923 20 323-329).-The influence of the cooling effect of the walls of the closed vessel on the pressure recorded in the combustion of propellants has been investigated by introducing a plate of gun steel bent into the form of an incomplete hollow cylinder into the closed vessel and so varying the extent of the cooling surface in contact with the hot gases. The effect on the pressures generated by propellants of different rates of Combustion is given in a table and by means of a graph in which the loss of pressure in kg. due to the cooling effect of the interior surface of the closed vessel is plotted agninst the duration of combustion expressed in thousandths of a second.For the most rapidly burning powders this loss of pressure amounts to about 50 kg. per sq. crn. and increases rapidly with the duration of combustion until a loss of 190 kg. is reached with the most slowly burning powders. Given equal periods of combustion the loss of pressure varies very little with the temperatures of the gases This curious result is ascribed to the fact that the products of combustion of the hotter pro- pellants (such as ballistite) consist largely of carbon dioxide and water the specific heats of which are much greater especially a t high temperatures than the specific heats of carbon monoxide and hydrogen which form a large proportion of the products of combustion of the cooler propellants containing large proportions of mineral jelly.For a given quantity of hea.t given to the walls,ii. 320 ABSTRACTS OF CHEMICAL PAPERS. the fall of temperature and pressure will therefore be less in the case of the ballistite than in that of the cooler propellants. The fall of pressure due to the cooling effect of the walls of the closed vessel may amount to 10% of the total pressure in the case of slow-burning propellants. H. C. R. Addition Reactions Molecular Compounds and the Hydro- lysis of Tantalum Pentachloride. K. LENDNER and H. FEIT (2. nnorg. Ghem. 1924,132 10-14).-Pure tantalum pentachloride dissolved in a suitable solvent e.g. carbon disulphide gives rise to additive products with a number of organic substances. Some of these can be separated in t,he crystalline condition. The pyridine compounds TaC1,,2C5H,N 4Ta02C1,7C,H,N and 2TaOCl,,3C5H,NHCI,2C,H,*OH and the piperidine compound TaCl 6C5H1,N,2C,H ,*OH have been isolated. Physical Theory of Chemical Reactions.P. LASAREFF (Compt. rend. 1924 178 773-775).-A qualitative theory of chemical reaction based on the Bohr atom. Chemical Aspect of Theories of Quanta and Thermo- dynamics of Photochemical Reactions. D. BERTHELOT (BUZZ. SOC. china. 1924 iv 35 241-302).-A lecture delivered at hhe College de France. Catalysis by the Action of Subdivided Metals. 11. Thermal Effect of the Hydrogenation of Ethylene at the Surface of Finely-divided Nickel. B. FORESTI (Gaxzetta 1924 54 132- 146; cf. A. 1923 ii 747).-Theories of catalysis are discussed with special reference to the hydrogenation of ethylene in presence of nickel The various phases involved in this reaction consist of (a) adsorption of the gases by the nickel ( b ) combination of the adsorbed molecules to yield adsorbed ethane and (c) evaporation of the ethane from the surface of the metal.Consideration of the thermal changes accompanying these phases indicates that the catalyst does not diminish the critical increment of the total reaction but renders possible a certain number of intermediate reactions with lesser critical increments without excluding the possibility that the sum of such increments may exceed the critical increment of the total reaction. Conclusions analogous to these were reached by Pratolongo from the results of his studies on chemical kinetics.Rideal's views are criticised adversely by the author. The mean experimental values obtained for the molecular heats of adsorption at 0" of ethylene and ethane a t constant volume by reduced nickel are 5845.6 cal. and 5004.7 cal. respectively. H. T. E. E. T. T. H. P. Hemoglobin and Methsmoglobin as Oxidative Catalysts. M. E. ROBINSON (Biochem. J. 1924 18 255-264).-H~~moglobin methEmoglobin and hzmin are efficient catalysts of the autoxid- ation of linseed oil. Hzmatoporphyrin does not have such action. The catalysis is most probably due to the presence of iron in the molecule. The concentration of iron in the form of bloodQENERAL AND PHYSICAL CHEMISTRY ii. 321 pigment capable of producing this catalytic effect is very much smaller than that required if the iron is present in the form of an inorganic salt.The addition of 'neutralised potassium cyanide in an amount equivalent to that of the iron present has no inhibitory Promoting Action of Palladium on Copper. I. Catalytic Combustion. W. W. HURST and E. K. RIDEAL ( J . Chem. Soc. 1924 125 685-694).-A mixture of hydrogen carbon monoxide and oxygen approximately in the proportions 8 1 1 was passed over catalysts of pure copper or copper with small amounts (up to 5%) of palladium and the ratio CO/H burnt was determined over the temperature range 130-270". For constant composition of the catalyst and varying temperature it was found in all cases that this ratio increases up to about 175" and t'hen diminishes rather rapidly up to the highest temperatures investigated.This tem- perature where the ratio is a maximum is identical with that a t which cupric oxide instead of cuprous oxide is formed by contact of copper with oxygen. At constant temperature increasing palladium content (up to 0.2%) increases the ratio after which it steadily diminishes. With 1.7% of palladium in the catalyst the ratio has the same value as with pure copper. The results support the theory that the promoting action must be due to a peculiarity of the Pd-Cu interface but on continued addition of palladium the augmented activity is decreased apparently by some specific action of that metal. L. J. H. influence on the catalytic action. s. s. z. Promoting Action of Palladium on Copper. 11. Adsorp- tion of Hydrogen and Carbon Monoxide. W. W. HURST and E.K. RIDEAL ( J . Chem. Xoc. 1924,125,694-705).-The adsorption of carbon monoxide and hydrogen separately a,nd mixed (1 1 yo CO) by (a) copper and ( b ) copper containing 0.2% of palladium was determined a t 0" room temperature and 100". This was made possible by the development of a method of analysis of the gas mixture giving an accuracy of 0.1 yo with only 1-3 c.mm. of the gas. The method depends on the measurement of the thermal con- ductivity. The adsorption of hydrogen is independent of the pres- sure down to about 150 mm. indicating that the surface is com- pletely saturated probably with a unimolecular layer whilst the adsorption of considerably more carbon monoxide suggests a multimolecular layer. The adsorption from the mixed gases is far less than that calculated from the assumption of proportionality and these results are discussed in the light of the authors' hypothesis concerning the adsorption of polar and non-polar molecules.From the mixed gases both constituents are always adsorbed but the composition of the adsorbed layer varies continuously with the composition and pressure of the surrounding gas and also with the composition of the solid. The ratio CO/H in the adsorbed layer obtained with promoted copper compared with that for pure copper is in the proportion of 1.20 to 1 a t 13" and 760 mm. pressure (cf. preceding abstract 1.32 1 a t 175"). L. J. H. 12*ii. 322 ABSTRACTS OF CHEMICAL PAPERS. Formation of Hydrogen Iodide by the Silent Electrical Discharge. E. WARBURG (2. Physik 1924 21 372-380).- Experiments have been carried out to determine the rate of form- ation of hydrogen iodide in the Siemens ozone tube at different temperatures namely room temperature 30° and 40".The alternating current used had a frequency of about 50. A stream of pure hydrogen was passed through a U-tube containing iodine kept at constant temperature. It thus became charged with iodine vapour and was next passed through the Siemens tube. The hydrogen iodide formed and part of the remaining iodine were ab- sorbed by water and the rest of the iodine by potassium iodide. The fraction of iodine transformed into hydrogen iodide increases with increasing strength of current and with diminishing rate of flow of the gas but reaches a limit a t approximately 90% when the velocity of formation of the iodide equals the velocity of decom- position.The fraction of a gram-molecule of gas formed per coulomb of electricity has been calculated and is found to be of the same order as the amounts of oxygen (to form ozone) ammonia and carbon dioxide which are decomposed per coulomb. The number of collisions between positive ions of molecular size and the gas molecules in the formation of ozone is found to be of the same order as the number of oxygen molecules decomposed. M. S. B. Yield of Photochemical Reactions with Complex Light. M. PADOA and N. VITA (Gaxxetta 1924 54 147-156).-The authors have investigated (1) the oxidation of hydrogen iodide and (2) the reaction 2FeC1 + H,C,O = 2FeC1 + 2HC1 + 2CO when these reactions are effected by the light from an arc lamp both before and after passage through various light-filters.For reaction (1) the sum of the effects produced by the blue red green and violet zones is 3.17 times the effect produced by the unfiltered light. For reaction (2) in the absence of a catalyst the spectral region over which the action takes place is somewhat limited but the sum of the effects obtained in the blue and violet regions is 1.123 times that of the white light. I n presence of quinine hydrogen sulphate the zone of active radiations is widened and the total action in blue green and violet lights is 2.125 times that in the white light. Experiments with Eder's solution give analogous results. Possible explanations for the above observations are discussed. T. H. P. Metal to Glass Joint.H. N. RIDYARD ( J . Anher. Chem. Soc. 1924 46 287-290).-By coating a glass silica or porcelain tube with a mirror of platinum and then dipping this platinised end momentarily into a molten lead bath a closely adhering film of lead is obtained on the glass. This lead-coated glass tube can readily be soldered into a suitable metal tube. Joints made in this way will hold a vacuum of 10-4 mm. a t temperatures up to 150" for some days and up to mm. for short periods. H. T. R. SCHLEIPER (Chew.-Ztg. 1924 48 160-161).-A laborat,ory gasometer adapted for continuous work- Laboratory Gasometer.GENERAL AND PHYSICAL CHEMISTRY. ii. 323 ing consists of two 3500 C.C. graduated Woulfe's bottles each having three necks and connected in such a way that the gas can be passed into either bottle as desired.The gas displaces the water from one bottle and measures it in the other and the same water is used continually so that no error is introduced by reason of the solubility of the gas and the pressure is always that of the atmosphere. A. R.. P. Bifilar Quartz-fibre Manometer. A. S. COOLIDGE ( J . Arner. Chem. Xoc. 1924 46 680-681).-The behaviour of the bifilar quartz-fibre manometer (A. 1923 ii 612) a t low pressures has shown the ineffectiveness of the gold-leaf trap. The substitution of a plain trap cooled to - 25" reduces the damping by condensation of mercury vapour. The temperature of the gauge need not be known when it is used in the lower range although when used for measuring pressures in systems at different temperatures a correction for thermal transpiration must be made.At higher pressures the viscosity of the gas influences the results but a calibration curve for one tem- perature can be adapted to another by simple displacement. In the previous work (Zoc. cit.) the pressure was never high enough to be affected by this corwction. A Six-prism Glass Spectrograph and a Neutral Wedge. J. S. FOSTER ( J . Opt. SOC. Amer. l924,8,373-379).-Constructional details are given of a spectrograph of which the dispersion can be varied by using any number of prisms from one to six. A maximum dispersion is attained of 2 A. per mm. in the violet. A neutral wedge is described which is used with the spectrograph for comparing the intensities of spectral lines. Methods of determining the density of the wedge for different wave-lengths are discussed and owing to difficulties encountered in maintaining a constant discharge through a vacuum tube a method utilising a continuous spectrum is adopted.S. B. Tri-colour Mixing Spectrometer. F. ALLEN ( J . Opt. SOC. Amer. 1924 8 339-341).-A description of a new instrument combining features of the Helmholtz colour-mixing and the Hilger wave-length spectrometers. Mixtures of two or three colours can be made and the wave-lengths and intensities of the components controlled and measured. S. B. Iron-Mercury Calorimeter. F. H. SCEOFIELD ( J . Scient. In&. 1924 1 141-144).-The calorimeter is designed for specific heat measurements on materials of low conductivity which react with water and consists of an iron block with a hole into which the hot specimen is dropped from a furnace. Rapid interchange of heat between the block and specimen is secured by flooding the hole Radium Balance. L. H. CLARK ( J . Scient. Instr. 1924 1 137-141) .-An apparatus for simple and rapid determinations of radium. Two insulated chambers are maintained at a constant difference of potential. An insulated electrode entering both S. K. T. with mercury. s. B. 12*-2ii. 324 ABSTRACTS OF CHEMICAL PAPERS. chambers makes electrical connexion between them. The chambers are exposed simultaneously one to the y-radiation from the radium under examination the other to the a-radiation from a very small mass of radium. The two consequent ionisation currents cause the potential of the insulated electrode to rise to a steady value when a balance is set up. A gold leaf indicator is attached to the electrode. The accuracy attained is of the order of 2%. s. B. Lecture Experiments to Show the Inflammability of Hydriodic Acid and Hydrogen Sulphide. G. G. LONGINESCU and C. N. THEODOSIU (Bul. Xoc. Rom&nu Xtiin. 1923 26,19-20).- A small portion of a thick paste of potassium permanganate and sulphuric acid on the end of a glass rod is introduced into a vessel completely full of hydriodic acid when the latter immediately takes f i e with incandescence and liberation of violet fumes of iodine. A similar method may be used to illustrate the inflammability of hydrogen sulphide. An alternative experiment with hydriodic acid consists in passing a stream of chlorine through a glass tube into the vessel containing the hydriodic acid when the latter a t once takes fire. W. H.-R. Diffnsion Gas Electrode. A. SCHMID (Helv. Chim. Acta 1924 7 370-373).-A gas electrode for use as a hydrogen electrode in a hydrogen-chlorine cell consists of a hollow porous carbon support on which is deposited a skin of electrolytic platinum. The platinum although having a smooth metallic appearance has microscopic pores impervious to liquids but permeable to gases. The gas passes from the inside of the carbon support through the platinum skin by capillary diffusion and forms a true solution at the active metal surface. The activity of such an electrode is so high that it is not polarised even by free halogen in the solution. Using this electrode for hydrogen and a similar carbon electrode without the platinum surface for chlorine hydrochloric acid free from chlorine can be produced continuously. The potential of the hydrogen-chlorine cell measured with this apparatus is 1.360 volt of the hydrogen-bromine cell 1.080 and of the hydrogen-iodine cell 0.54 volt at 20" against a normal solution of the corresponding acid. The temperature coefficient of the potential between 0" and 80" for these three cells is - 0.0010 except in the interval 20-40" for the bromine and iodine cells when it has only one-fourth of this value. T. S. PATTERSON (Chenz. and Ind. 1924 43 196-198).-2eise discovered mercaptan about the year 1824 and gave this name to the substance; he derived it from corpus mercurium captans ( J . pr. Chem. 1834 1 E. H. R. Derivation of the Word "Mercaptan." 1861 and not from corpus mercurio upturn. w. P. s.

ISSN:0368-1769    

DOI:10.1039/CA9242605281

出版商:RSC    

年代:1924

数据来源: RSC