摘要:

FEBRUARY, 1916. Vol. XL., No. 467. THE ANALYST. PROCEEDINGS OF THE SOCIETY OF PUBLIC ANALYSTS AND OTHER ANALYTICAL CHEMISTS. CORRECTIONS IN BOMB CALORIMETRY. BY G. NEVILL HUNTLY. (Read at the Meeting, December 2, 1914.) THIS present note is supplementary to a paper previously published (J. SOC. Chem. Ind.,*1910, 29, 917 ; ANALYST, 1910, 35, 444). I n a determination of the calorific value of a fuel with the bomb calorimeter, corrections are required for-( 1) Errors of the thermometer tube (calibration) ; (2) heat given out by the fuse ; (3) cooling (radiation); (4) change in water equiva- lent with temperature ; (5) sulphuric acid formed ; (6) nitric acid formed ; (7) heat loss by evaporation ; (8) heat gained from stirrer ; (9) incomplete combustion.Many instruments are used in a more or less empirical manner, especially by engineers, and it appears to be customary to'make a rough allowance for cooling and to neglect the other corrections.Among chemists, in addition to the cooling correction, an allowance is generally made for the sulphuric acid and nitric acid formed by the combustion. For some purposes the neglect of small corrections may be justified by the nature of the sampling; but it seems worth while to estimate the magnitude and direction of all sources of error, so that their possible cumulative effect in the most unfavourable case may be known, and the number of significant figures given in the final calorific value made to correspond.Thermometer Tube Errors (CaZibration) .--This question is discussed in a previous paper (Zoc.cit.). If a good thermometer is used without a calibration table-and this must not be confused with a table of corrections obtained;by comparison with a standard, which is not good enough for the present purpose-the errors will not exceed, but may amount to, one unit of division; that is, 0.02O on a thermometer divided into fiftieths (8 per thousand)," O*0lo (4 per thousand) if the divisions are in O*0lo.Fuse.-This may be iron, nickel, or platinum wire, either embedded in the briquette or resting on it, or the wire may be arranged not to touch the fuel, the flame being led down by a cotton thread or collodion wool. The cotton thread used with fine platinum wire is convenient, and rarely fails to ignite. The wire should be very fine (0.002 inch), and unnecessarily high battery voltage avoided.The heat of * The rise of 25" figures 0. in brackets represent throughout the error in parts per 1,000 on an assumed average42 NEVILL HUKTLY : fusion even of this fine wire is just measurable, amounting to about 5 calories, to which is added about 20 calories for the cotton. The use of thick wires and high voltages will lead to an error which only vanishes when the rise of temperature for the fuel happens to coincide with the rise obtained during the calibration with a standard substance.Cooling Correctiolz.-This has been dealt with in detail in the earlier paper, in which the meaning of each of the terms in the Regnault-Pfaundler expression is discussed. The author is of opinion that this formula should always be applied in work of any preteneions to accuracy, and its application is much facilitated by the use of the form given below : Date : November 23, 1914.Capsule +fuel = 7.8750. Capsule = 7,0200. Temperature (final) : 15.0. Thermometer: No. 1. Water : 2.160. Weight = 0.8550. W.E.: 2,500. Number: : 585B. Substance : Oil. Ash: -. Preliminary Period. Time.0 1 2 3 4 5 80 9 nv V .- - - - - -. Temperature. 0.026 0.030 0.033 0.037 0.04 1 0.045 0.035 - 0.0038 - 0 019 -- --- After Period. I Chief Period. I ___-. --I__ Temperature. Time. 1 Temperature. 2 .oo 3.47 3.60 3.599 = 12.67 = 1.82 - - -0.17 = 14-32 -- ---.- 10 8, 11 12 13 14 15 3.598 3.594 3.590 3.586 3.583 3.678 8'= 3,588 ?I*= +0~0010 Cooling correction = nv + kS = - 0.019 + 0-0314 = + 0.0124.3.560" Rise per grm. = ___ 0.855" Calibr. - @0069" -- R ... = 4.164" Uncorrected rise (3.553' + 0.0069" Thread ... ... - 0-0079° R x 4500= 18,740 B.T.U. per pound Cooling ... ... + 0-0124" Total correction = +0*007" R x 2500 = 10,410 cal. per grm. Nitric acid ... - 0.004B0 - Sulphur . . . ... Corrected rise ... = 3.560" According to this the correction is- d - v .nv + 2; (6) + (6, + O5) - ite }. The calculation of the correction should be obvious from the form. Sixteen readings are taken at intervals of one minute, and in the intervals between theCORRECTIONS IN BOMB CALORIMETRY 43 readings the values of 8, F, nu, 2:(8), 9 (80++5), no, S, (the large bracket) the thermometer calibration errors, and the uncorrected rise, are readily calculated and inserted, so that after the last thermometer reading only 0', v', k, and kS, have to be inserted, and the time required for this averages two minutes.At the risk of repetition, it may he emphasised that this is the only expression which takes into account errors due to evaporation from the calorimeter vessel and the heat evolved by the friction of the stirrer. A Shortened Form of the Regnutilt-Pfaundler Formula.The expression 1 V f 8'- o\ Correction = nv + ~ v l q e ) + 8 (8, + 05) - ne represents the approximate integration of the time-temperature curve, and if con- ditions are arranged so that the shape of this curve remains similar, the formula can be reduced to a very simple expression without loss of accuracy. The shape of this curve depends on the heat conductivity of the walls of the bomb, the rate of stirring, and the velocity of combustion of the fuel.The predominating factor is the rate of stirring, The first obviously remains constant for a given instrument, the second can be kept constant if the stirrer is driven by a suitable motor, and the third depends on the nature of the fuel burnt. From the examination of over 200 determinations, the author has found that for his own instrument (Kroeker, enamel- lined), with the stirrer working at 120 complete oscillations per minute, &-r8 is constant, and equals 4.2 k 0.1 for coals (in briquettes), and 4-45 * 0.05 for oils or pure organic substances.This leads to an expression of extreme simplicity, which for this instrument, working under these fixed conditions, retains the accuracy of the original lengthy formula.Numbering the temperature readings to, t,, up to t15, the formula may be written : S S 5 (6' - e) \ Correction = (to - t5) + h 1 0 - tl5) - (to - t5) } S 4.2 - Taking the coal figure, -----=---084, the correction reduces to 0.16 (to - t5) + 0.84 (tlo - t,,), due regard being paid to the signs of the two tempera- ture differences. If a small table of two columns be made, giving the products 0.16 (to - t5) and 0-84 (tlo- t15) for such temperature differences as are found to occur in practice, the whole correction is complete within thirty seconds.5(ef - 8) 5 For oils or pure organic compounds, the corresponding expression is 0.11 (to - ts) + 0.89 ( t i 0 - t15). The average error of this approximate method is under O-O0lo, and rarely amounts to 0.002°.The saving of time over the full expression, if the form be used as above described, is not more than two minutes, but the liability to error is reduced; even if the full calculation be preferred, the shortened form is a valuable check on the calculation. Two points of interest can be deduced from the above.If - (to - t5)= + (tlo - t15)44 NEVILL HUNTLY : --that is, if the rate of gain of temperature during the preliminary period is exactly equal to the rate of cooling during the after-period-it has been assumed by many workers that the cooling correction is nil. In this case we get as the correction (to - tJ (0.84 - 0.16). Taking the rise of temperature during the preliminary period as 0.020, the correction is +0*014O (5.6 per thousand) instead of O*OOOo.Another interesting application is the deduction of the effects of thermometer errors (reading or calibration or both combined) on the final corrected rise. The latter is (tlo - ta) + 0.16 (to - t,) + 0.84 (tlo - t,,), and this reduced to approximately 1*84t10 - 1*16t5+ 0.16t0 - 0*84t,,, so that an assumed error of 0.001 on each reading will give the total error on the corrected rise.The maximzcm error will be obtained if the errors on the four readings-to, t,, tlO, t,,-are alternately plus and minus, as shown below : &ssumed Effect on the Corrected Rise. Reading. + *0Ol0 + *00016° - - 0 0 1 O + *00116" + -O0l0 + *00184O - *001" + *00084O + *004" From this it appears that an error on the first reading is comparatively unim- portant, but that an error on the first reading of the after period has double weight. Also the error on the corrected rise may be, as a maximum, four times the possible error of reading.This means 22 B.T.U. on a coal of 14,000 B.T.U. per pound, or 30 B.T.U. on an oil of 20,000 B.T.U. per pound. If readings are taken to 0.005" only, as appears to be frequently the case, the possible error due to this cause alone will be 110 B.T.U.on a coal of 14,000. This result is general, and is not limited by the particular method of applying the cooling correction. Whilst dealing with the cooling correction, it may be of interest to compare the method used by Mahler with the Regnault-Pfaundler correction. Taking an example given by Mahler (Bull.de la Socidtd d'Encouragement p o w 1'Industrie Nationale, 1892, 310), the corrected rise given in the paper is 3.615O; by the longer method, 30598~. Mahler deduces a heating value of 9,69108 oalories, instead of 9,646, a difference of about 0.5 per cent. I t is clear that, if a rough-and-ready cooling correction is deemed sufficient, the number of significant figures in the resulting calorific value must be limited to correspond. J.A. Riche (J. Amer. Chem. Xoc., 1913, 35, 1747) abolishes the cooling correction by the use of a glass Dewar tube. In the form proposed both the radiation and evaporation errors are abolished, the heat evolved by stirring still remains un- accounted for. From the description given, it would appear that the time required to fit the bomb into place would exceed that required for the full correction.This vacuum jacket method is to be preferred, in principle, to the method of adiabatic calorimetry proposed by T. W. Richards. The Temperature Coe$icient.-The water equivalent of the bomb, water and con- taining vessel, varies slightly with the temperature. Hence, unless the room in which the calorimetric determinations are carried out can be kept st a uniformCORRECTIONS IN BOMB CALORIMETRY 45 temperature throughout the year, this correction must be measured and allowed for.As regards the change in the specific heats, each 4' from 15' C. causes a variation of 1 part in 1,000 in the W.E. If a thermometer of the Beckmann type with variable zero is employed, the value of the degree varies with the temperature-that is, with the weight of mercury in the bulb.These two corrections are conveniently combined by determining the water equivalent at regular intervals by combustion of a pure substance, benzoic acid for preference. A set of figures obtained with such a ther- mometer is appended. The quantity of water taken was constant (2,160 grms.), and corresponded with a water equivalent of 2,500 grms.at 15' C. The variation was found to be linear, and could be expressed by the relation W.E. = 2,500 + (T - 1 5 O ) where T was the final temperature of the water in the calorimeter : - I I Water Equivalent. I Temperature : 0" c. Difference. I 1 Calculated. 1 Found. 1 Grms. Water. I Parts per 1,000. 10 12 14 14 15 22 22 23 23 2,495 2,497 2,499 2,499 2,600 2,507 2,607 2,508 2,508 2,491 2,496 2,500 2,500 2,500 2,504 2,508 2,510 2,509 - 4 - 1 +1 + 1 - - 3 +1 + 2 4-1 - ~ -1.6 ~ -0.4 l +0*4 ~ +0.4 - 1.2 . +04 + 0.8 + 0.4 - Average error from the mean, 0.6 per thousand, or 4 calories. Greatest error between extreme results, 2.4 per thousand, or 15 calories.Maximum error due to assumed error of 0*001" on the four critical thermometer readings, 0*004°, or 10 calories.The Corrections for SulphzLr and Nitric Acid.-For the determination of this correction it is usual to place 10 C.C. of water in the bomb, and, after allowing time for the fog to settle (thirty minutes from the time of firing), estimate the nitric and sulphuric acids formed, either by a volumetric method (baryta, and sodium oar- bonate) or by a titration of the total acidity foliowed by a gravimetric estimation of the sulphuric acid. In the latter method, preferred by the author, there are two sources of error to be noted.In a bomb lined with enamel, minute transparent particles of the enamel are apt to be detached, and if these are weighed as barium sulphate the result will be high.With a leaden washer, appreciable quantities of lead nitrate and sulphate are formed. If a measured excess of & sodium carbonate is added to the bomb washings, the liquid boiled and filtered, the excess of alkali can be accurately determined by back titration with acid, with a, sharp end-point. Direct titration with alkali gives a rather indefinite end-point in presence of the lead salt, and the total acidity will be too low, this error falling on the nitric acid.The filtration also removes any particles of enamel, and the filtrate serves for the46 NEVILL HUNTLY: precipitation with barium chloride. Direct experiment shows that no sulphate has been carried down by the lead carbonate. The correction for nitric acid is fairly constant for a given bomb, and rarely exceeds 10 to 12 calories.One C.C. sodium carbonate=0.143 calorie, or 0.00057° for a water equivalent of 2,500. As regards the sulphur, it is usual to subtract 22.5 calories for each 1 per cent. present, 01, as an alternative, 0*00144° for each C.C. of & alkali used by the sulphuric acid (W.E. 2,500). This is based on two assumptions: that the sulphur is com- pletely oxidised to sulphuric acid in the bomb, and that in practical use of the fuel the sulphur is burnt entirely to sulphur dioxide.The first of these assumptions is substantially accurate, the second is not. In a paper* on the influence of the volatile constituents of solid fuel on the efficiency of internally-fired boilers, Constan1 and Schlapfer have investigated the question of the distribution of sulphur in fuel between the ash and the flue gas, and the proportion of sulphur dioxide in the latter.In one set of four experiments, after deducting the sulphur found in the ash, the amounts of SO, found in the flue gas only amounted to from 15 to 25 per cent. of that calculated on the assumption that all the volatile sulphur was burnt to SO,. Another set of four experiments by the same authors gave an average of 40 per cent.of the expected sulphur dioxide, and this would be still lower if calculated on the total sulphur in the coal. Some sulphur trioxide is well known to be formed during the combustion of sulphur compounds in air, and it is probable that under the catalytic action of traces of oxides of nitrogen, always present in a, flame, and of ferric oxide on the boiler tubes, a further conversion into SO, takes place. In internal combustion engines of the Diesel type the conditions of combustion approach those obtaining in the bomb, and consequently for fuel oils this correction may become misleading.Thus, direct analysis of the exhaust gas from a Diesel engine burning a fuel oil containing 3.6 per cent.of sulphur showed that only 5 per cent. of the total sulphur was given off as SO,. Here the same set of experimental figures could be interpreted to give a gross calorific value of 18,445 or 18,300 B.T.U. per pound, depending on the omission or application of the sulphur correction. In this case the omission of the sulphur correction would correspond nearer to the facts, although there is still uncertainty, as 90 per cent.of the sulphur of the fuel remained in the cylinder, and only a, small proportion of this was present as sulphuric acid. To avoid dispute, therefore, a certificate of analysis should state whether the sulphur correction has been applied or not, together with a statement of the assumptions upon which the difference is based.Heat Loss by Evaporation.-Mahler (hc. cit.) gives the loss by evaporation as 1 grm. per hour. This will obviously vary with the instrument, but, accepting this figure as an approximation, the heat loss will be about 10 calories per minute, or 0*004O per minute for a calorimeter with water equivalent 2,500. This is in the same direction as the cooling correction. This error is eliminated by the Regnault- Pfaundler formula, but is not eliminated in methods of adiabatic calorimetry unless the calorimeter vessel is enclosed in an air-tight manner.The heat developed by * ‘ I Yitteilungen iiber Forschungsarbeiteii auf dem Gebiete des Ingenieurwesens, ” Yercin dezclschen Ingenieure, Heft 103. Berlin, 1911.CORRECTIONS IN BOMB CALORIMETRY 47 the stirrer acts in the opposite direction.Richards and Burgess ( J . Amer. Chem. Soc., 1910, 32, 448) give some data on this point, and show that with 110 oscillations per minute the temperature rise is of the order of 0-006" for five minutes. The im- portance of efficient stirring does not always appear to be realised. Inefficient stirring magnifies the errors due to the lag of the bomb and thermometer, and vitiates the radiation correction, as the latter is based on the assumption that the reading of the thermometer represents the temperature of the outer wall of the calorimeter vessel.The only drawback of rapid stirring is the development of heat by the friction, and this is eliminated completely by the full correction, provided the rate of stirring is uniform throughout, a condition which can only be fulfilled by using a motor.One hundred and twenty osoillations per minute is a convenient rate. Incomplete Combustiort,-Ash-free solids, or liquids with low vapour pressures, are burnt completely without difficulty in the bomb. Volatile liquids are also com- pletely burnt, provided the precautions indicated by Berthelot are followed. With coals the combustion is not always complete, and these cases may be conveniently classified under three heads : (1) Fuels with low hydrogen content, such as anthra- cite, coke, graphite; (2) fuels with high, non-fusible ash; and (3) coals with fusible ash.For substances such as anthracite and coke briquettes cannot be formed, and the substance must be burned as a fine powder. Various methods may be adopted.The use of a sheet of mica or a pad of asbestos or other non-conducting material prevents the ash from being prematurely cooled by the platinum capsule, and this device, together with a higher oxygen pressure (40 atmospheres instead of 25) has given satisfactory results with several workers. Another device, which gives very good results, is to moisten the coal, after weighing out, with a few drops of a high boiling oil (B.P.paraffin oil does well) of known calorific value. The oil is weighed and its heat of combustion deducted. Another device which the author has proved to be satisfactory is to weigh out the coal in a small square of dry, ash-free filter- paper, the calorific value of the paper being determined in a separate experiment, and of course deducted as before.This works better than the oil when the coal ash happens to be fusible. For bituminous or semi-bituminous coals (over 15 per cent. volatile matter) with high non-fusible ash, combustion is complete if a mica plate is inserted under the briquette. With olinker, flue dust, or similar products, the addition of oil is neces- sary for satisfactory combustion.The third class is the most difficult to handle, coals giving fused globules in the platinum capsule. These almost invariably contain enclosed particles of unburnt oarbon. The error can be reduced by using a mica plate, or, better, ignited quartz sand, as numerous small globules are thus obtained instead of two or three large ones. Breaking up the briquette into six or eight pieces also reduces the size of the globules, and using 0.5 grm.instead of 1 grm., with a lower oxygen pressure, also gives higher results. But the only really satisfactory method of dealing with such cases is the determination of the unburnt carbon in the ash. The use of quartz sand facilitates removal of the ash from the capsule, and also renders its pulverisation in a steel mortar more complete,48 CORRECTIONS IN BOMB CALORIMETRY A combustion of this is made in a miniature closed tube after mixing with lead chromate and potassium dichromate, the GO, being either weighed or, more con- veniently, caught in baryta solution and titrated. For a given coal of this class, the amount unburnt is curiously constant, so that two or three consecutive determina- tions of calorific value may agree well, and yet be from 1 to 4 per cent.low. One Scotch coal gave results 8 per cent. low when burnt in platinum; with quartz sand this error was reduced to between 1 and 3 per cent. This was exceptional, but deficiencies up to 3 per cent. may be expected whenever large globules of ash are found. This is the most serious error in coal calorimetry when the instrument is used empirically. There is an unfortunate tendency, .frequently commented upon, for chemists to give their analytical results to more significant figures than is justified by the experi- mental methods employed. This is especially the case with determinations of calorific value. The calorific value of ordinary fuels, expressed in B.T.U. per pound, is of the order 10,000 to 20,000. Hence has arisen the custom of reporting to the last figure, such as 12,831 B.T.U. per pound. The last figure is purely arithmetical, and even under the best conditione of working has no meaning. Results agreeing closer than 30 B.T.U. must be regarded as accidental; indeed, having regard to the possibilities of constant errors indicated above, the agreement of duplicates, taken probably on the same thermometer and upon the same portion of the scale, is apt to give completely false ideas of the accuracy of the measurement. With uncalibrated thermometers reading to 0.005, errors of 200 B.T.U. per pound are quite possible. In such cases the accuracy of the measurement will be fully expressed by three significant figures only.

ISSN:0003-2654    

DOI:10.1039/AN9154000041

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

48 CORRECTIONS IN BOMB CALORIMETRY A NEW CHEMICAL HYGROMETER. BY ERIC K. RIDEAL PH.D. B.A. AND A. HANNAH B.A. THE important measurement of the relative humidity of the air is usually carried out by means of the “wet and dry bulb thermometer” called Mason’s hygrometer or psychrometer. This instrument first suggested by Dr. Hutton of Edinburgh and investigated by Sir John Leslie (on The Relations of Air to Heat and Moisture,” 1813 p. 39) is in a standardised form adopted for the application of the humidity regulations of the Workshop and Factory Act for meteorology and indeed for most other purposes. I t is well known that the instrument as generally used is very inaccurate; under unfavourable conditions-for instance when the air is dry or at even fairly fairly low temperatures-the error may be as much as 30 per cent.; and it is a curious fact that in saturated air when both wet and dry thermometers should read alike the wet bulb will sometimes read slightly higher than the dry bulb. These errors have long been recognised and find expression in the large number of different tables and formulae that have been devised to correlate the readings of the thermometers with the humidity of the air. The tables of Glaisher used i A NEW CHEMICAL HYGROMETER 49 England are derived from simultaneous measurements conducted over thirteen years (from 1841) with the wet and dry bulb thermometer and Daniell’s hygrometer, both at Greenwich and in India combined with Regnault’s quantity figures. No theoretical considerations therefore enter into his tables but all the errors (which are known to be considerable) of Daniell’s hygrometer are of course reproduced in his figures leaving the inaccuracy of measurement of the wet and dry bulb ther-mometers out of consideration.According to Daniel1 (“ Meteorological Essays,” 1856 ii. 100) the results obtained by Apjohn’s original formula are of greater accuracy than those of Glaisher’s tables. This formula (1834) is the basis of most of the modern ones used on the Continent and elsewhere. In its simplest form it may be represented by the equation F= f - Adh, where F= the vapour pressure at the dew point; f = the vapour pressure at the temperature of the wet bulb ; A = a psychrometric constant ; d = the difference between the wet and dry bulb readings in OF.; 12 = height of the barometer in millimetres. According to Apjohn the value of A is approximately 0.00096. It is evident fhat since the lowering of the wet-bulb thermometer is due to the beat absorbed from the bulb the wet cotton and the surrounding air by the evaporation of the water on the wick the velocity of the dry air over the bulb and wick to a great extent influences the rate of evaporation and consequently the reading. Most modifications of Apjohn’s formula attempt to recognise this fact. Jelinek’s well-known formula adopted in Austria takes it into consideration in the following way : The psychrometric constant A is given the following values : 0.0012 in still air (0 - 0.5 metre per sec.). 0.000800 for a light wind (1 - 1.5 metres per sec.). 0.000656 for a strong wind (more than 2.5 metres per sec.).According to curves drawn by the Meteorological Office at South Kensington, there is considerable divergence between the results obtained by Jelinek’s formula and Glaisher’s tables especially at low humidities and in dry air. Similar modifications of Apjohn’s formula are used in other countries among which may be mentioned those of Haeghens Wild and Guyot. Owing to the fact fhat these corrections for varying wind velocities are not sufficiently exact Assrnann’s principle of a constant draught with a fixed psychrometric constant is now often adopted. An air current of from 2 to 3 metres per second controlled by an electric motor is passed through the apparatus which is thus further complicated. d simpler artifice much in favour in the United States is the L‘sli~g” ther-mometer used in conjuction with Ferrel’s formula.This instrument is more suitable than Assmann’s for field work but requires some skill in manipulation. The obvious errors which occur in readings of this kind and the empirical nature of the result when obtained have led several investigators to the subject of chemical hygrometry 50 RIDEAL AND HANNAH : Gravimetric determinations although absolute and leaving nothing to be desired in point of accuracy are tedious require a large and complicated apparatus and at the best only give an average value for the humidity of the air over a long period. In consequence efforts were made to design types of apparatus for measuring the contraction which a known volume of air underwent on drying.In 1878 and 1879 several attempts were made to produce such a chemical-hygrometer. A simple and most ingenious one was that of Rudorf He fitted & three-necked Wolff bottle with an oil manometer and a burette containing concen-trated sulphuric acid; the air was titrated until the pressure returned to that of the atmosphere. The volume of sulphuric acid used was then equal to that of the aqueous vapour in the bottle. Neesen fitted a second Wolff’s bottle to the open limb of the manometer to compensate for any change of temperature which might take place during the experiment and allowed the operator to do two determinations successively. t - U C Edelmanns (1879) apparatus was also designed to use sulphuric acid in this manner.Schwackhofer designed an apparatus which was used by a few meteorologists. for some years. The pipette A into which a sample was drawn had a long graduated stem connected to the mercury reservoir B. After the sample was measured and the sulphuric acid raised in the absorber C by the rubber bulb the plunger in the mercury reservoir was screwed down transferring the sample to the absorber. Upon returning the sample the contraction in volume was noted on the scale of the pipette. The pressure was adjusted on the gauge D. Dines’s vapour meter was designed on lines very similar to that of Schwackhofer but more complicated and possessing the 8ame disadvantages. W. N. Shaw in his report on hygrometric methods writes “There is no hygrometric method of which it has been proved that an observer following out definite written instructions with due care and skill can obtain measurements o A NEW CHEMICAL HYGROMETER 51 vapour pressure which are accurate to within 1 per cent .. . There is no evidence to show that the various absolute hygrometers such as those of Schwackhofer, Edelmann Neesen and Dines and others can be relied upon to that degree of accuracy whatever may be done by one particular observer after laborious trial and testing with one particular instrument.” S. Rideal (“The Relative Hygienic Values of Electric Light and Gas 1908,” J. Roy. Sun. Inst. 1908 29) wishing to measure humidity quickly and accurately, also noted the unsatisfactory nature of these instruments. He constructed a hygro-meter on the lines of Schwackhofer improving it by the addition of a temperature and pressure compensator simpler absorber and means of adjustment ; but though completion of the inquiry put an end to the experiments he was able to point out that this type of instrument could hardly be successful for when the dried air was returned to the measuring pipette it took up a further quantity of moisture owing to the hygroscopic nature of the glass.He found that the sample had to be passed over at least eight or ten times before being thoroughly dried. Thia observation, together with the fact that sulphuric acid will completely dry air on one contact, which is not the case with solid dehydrators forms the basis of the instrument which is now described. I t consists essentially of two bulbs A and B each of the same oapacity, conveniently 20 C.C.The bulbs are connected at their lower ends to two movable mercury reservoirs (suspended on ratchets to insure easy manipulation) E and F. Above the mercury in the bulb A there floats a few C.C. of strong sulphuric acid, occupying the bulb D when A is full of mercury. The upper stem H of the bulb B is calibrated and can be marked either in grains per oubic foot or vapour pressure in millimetres of mercury. The upper end of the calibrated tube and the stem of th 52 RIDEAL AND HANNAH: aulphuric acid bulb are connected together by means of a three-way cock X and a horizontal capillary tube Y which is also connected to an aniline U-gauge fitted with a bulb C to compensate for any small changes in temperature or pressure taking place during the experiment A NEW CHEMICAL HYGROMETER 53 .' 59.0 ' 2-82 I - ' 1 3.87 4-27 4.87 . 58.8 2.96 3.17 8.96 . ' * 66.0 - 4.83 5 a03 - 4.25 3.60 i I May 28 . June 8 June 25 . June 30 . June 4 . 63.6 1 -The method of taking a reading is as follows : The reservoir E is raised until the bulb A is full of mercury and the sulphuric acid above it all but fills the bulb D. A sample of fresh air is drawn into B through X by lowering F. The compensator and gauge taps M and N are opened thus insuring that the sample in B is at the same temperature and pressure as the surrounding air. The taps M and N are then closed and the bulbs B and A con-nected by turning the tap X ; the air in B is transferred to the sulphuric acid bulb D by raising the reservoir F and lowering E.When the mercury in B rises to the mark 0 and the level in A has reached P the tap N is opened connecting the air inside the bulbs D and A to the gauge. Since the volume of mercury entering B is exactly equal to the volume leaving A if the air were perfectly dry no change in pressure would be observed but with ordinary moist air a contraction in volume is noticed and the liquid is no longer level in the two arms of the U-tube. It is then necessary to raise the level of mercury in the calibrated stem H at the same fiine maintaining the mercury level at P in the other arm until the gauge comes into equilibrium. The actual quantity of moisture in the air can then be directly read off on the stem H. If the measurement adopted is the vapour-pressure in millimetres of mercury, no corrections are necessary; but when the more usual notation of grains per cubic foot is used the usual gas temperature pressure corrections must be applied.The calibration of the instrument may be made when it is remembered that 1 grain of moisture per cubic foot of air is represented by a .contraction of 5-69 hundredths of a C.C. on 20 C.C. of air when corrected for temperature and pressure. If calibrated at 60" F. and 760 mm. of mercury the error in omitting this correction will be small, and for practical purposes need not be considered. It is advisable to mount the apparatus in a water-bath and it hae been found convenient to adopt the form used by Dr. Haldane in his apparatus for determining the carbon dioxide in the air.I n order to prevent the inclusion of the sulphuric acid between the mercury and the glass by too violent movement of the mercury it is advisable to insert a short fength of capillary tubing between the reservoir E and the bulb A . No difficulty is experienced in manipulation while less than two minutes are sufficient for a reading; the sulphuric acid lasts for many hundreds oE deter-min ation 6. 2.88 4.18 3.02 4-80 4.30 Date. An apparatus of this description has been in constant use for some months past, and has given excellent results. Daily determinations throughout the month of Jun 54 ABSTRACTS OF CHEMICAL PAPERS with this instrument wet and dry bulb thermometers in duplicate a Regnault’s dew-point apparatus together with frequent gravimetric determinations have shown that the agreement between this hygrometer Regnault’s apparatus and the grsvimetric method was excellent but marked and sometimes very wide discrepancies were observed in the figures obtained by the wet and dry bulb instruments. The figures given in table on p. 53 tabulrtte some of the results obtained

ISSN:0003-2654    

DOI:10.1039/AN9154000048

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

54 ABSTRACTS OF CHEMICAL PAPERS ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOOD AND DRUGS ANALYSIS. Examination of Cheese. S. Tijmstra. (Chem. Weekblad., 1914,11, 90%)- Sumpling : Cheeses over three or four weeks old are sufficiently uniform to allow sampling by boring; new cheeses, on account of the pickling process, require the cutting of segment, as the rind contains more salt and less fat than the interior.Fut : In Van Gulik's method sulphuric acid of sp. gr. 1-53, and not 1.50, as suggested in the Codex Alimenturius, should be used. The following procedure is recom- mended : Three grms. of the sample are dissolved in sulphuric acid of sp. gr. 1-53 in the butyrometer, warmed to 68' C., and vigorously shaken; 1 C.C. of amyl alcohol is added, the butyrometer is shaken, filled with sulphuric acid, and left for five minutes at 6 8 O C,, centrifuged for five minutes, warmed to 68" C.for five minuties, and read off. The results agree with those obtained by the petroleum-ether extraction method of Bondzyski-Ratzlaff. Moisture : Satisfactory results are not given by distillation with petroleum or by heating to 105" C., but are obtained by drying a mixture of 5 grms.cheese with 25 grms. fine sea sand and 5 C.C. alcohol, warming to 70" C. and thoroughly mixing. 0. E. M. Grape-Seed Oil. G. Dell' Acqua. (Annali Chim. Applic., 1914, 2, 295-301.) -The increasing cost of olive oil has led to the introduction of grape-seed oil as a commercial product. A pure extracted sample had a greenish-yellow colour similar to that of inferior olive oil, and gave the following values : Sp.gr. at 15" C., 0.9226 ; Valenta test, 83' C. ; refractometer reading (Zeiss), 78.8 at 15" C., 62.9 at 40" C.; iodine value (Hubl), 140.25 ; acetyl value, 17.84 ; unsaponifiable matter, 0.32 per cent. Fatty Acids.-Sp. gr. at 25' C., 0.8988 ; melting-point, 25" to 28.5' C. ; solidifying point, 21' to 18" C. ; refrwtometer reading, 62.8 at 20" C.; 51.8 at 40' C. ; and iodine value, 141. The oil thus resembled Soya-bean oil in many of its constants, and, like that oil, gave a lemon-yellow emulsion in the uranium nitrate test (ANALY~T, 1913, 38, 36). The oils could be distinguished, however, by heating 10 C.C. with 3 C.C. of an ethereal 2 per cent. solution of uranium nitrate for two minutes in a boiling-brine bath (102" C.).Soya-bean oil assumes an olive-green colour changing to garnet red within twenty minutes, while grape-seed oil becomes yellowish-green in two minutesFOOD AND DRUGS ANALYSIS 55 and golden-yellow within twenty minutes. The reactions of grape-seed oil with aqueous or ethereal solutions of uranium nitrate are not masked by the presence of a large amount of mineral oil.Hauchecorne's nitric acid test will also distinguish between the two oils. Soya-bean oil heated for ten minutes in water at 60" C. with nitric wid gives an orange-brown coloration changing to chocolate-brown, while grape-seed oil assumes an orange-brown colour changing to reddish-orange. C. A. M. Determination of the Freezing-Point of Milk. M. G. Hummelincka Chem.Weekblad, 1914, 11, 207-209.)-The sample of whole milk is cooled in a freezing mixture at -4' C. to such an extent that it is super-cooled by 1' C. when transferred to Beckmann's apparatus. A little of the ice from the side of the tube is introduced by means of a platinum wire, and the point to which the temperature rises is read. The zero- point of the thermometer is obtained with recently boiled distilled water.The control samples of milk thus examined during the last few years showed freeeing- points ranging from - 0.57" to - 0-542' C. Duplicate determinations should agree within 0*003' C. C. A. M. Freezing-Point of Milk. van Raalte. (Chm. Weekblad, 1914, 11, 206-207.) -According to the Dutch Codex Alimelztarius, the presence of added water may be assumed in milks that have a higher freezing-point than -0.54' C.In the case of 155 samples of genuine milk examined by the author, the freezing-point ranged from - 0.54" to - 0*57O C. C. A. M. Freezing-Point of Milk. L. T. Reieher. (Chem. Weekblad, 1914, 11, 323- 324.)-Of 467 samples of milk of undoubted purity examined since 1909, 92 per cent. showed freezing-points ranging from - 0.580" to - 0*550° C., and not one gave a higher reading than - 0 ~ 5 4 1 ~ C.These samples of milk were the mixed products of herds of eighteen to thirty cows. C. A. M. Study of the Saponins. A. W. van der Haar. (Chm. NeekbZad, 1914,11, 214.)-Owing to the difficulty of separating the saponins, the work of previous investigators is open to the objection that it was carried out on impure material. From the leaves of Hedera heZix the author has separated a specimen of a-hederin of m.-pt. 256' to 257O C., differing considerably in its properties from that investigated by previous workers : on hydrolysis it gives equimolecular proportions of a- hedera- genin, arabinose, and rhamnose. To a-hederagenin he assigns the structure On reduction with zinc in a current of hydrogen, it yields a sesquiterpene or mixture of sesquiterpenes giving a violet coloration with sulphuric acid, and a non-volatile solid. 0. E. M.

ISSN:0003-2654    

DOI:10.1039/AN9154000054

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

56 ABSTRACTS OF CHEMICAL PAPERS BACTERIOLOGICAL, PHYSIOLOGICAL, ETC. Estimation of 6’-Hydroxybutyric Acid. A. L. Kennaway. (Biochem. J., 1914, 8,230-246.)-The /3-hydroxybutyric acid is oxidised with potassium dichromate and sulphuric acid, and the acetone formed estimated according to the method of Scott-Wilson. The method may be illustrated by the following example : Twenty- five C.C. of diabetic urine giving a moderately strong ferric chloride reaction were placed in a 250 C.C.flask, diluted with water, and mixed with 50 C.C. basic lead acetate solution and 10 C.C. strong ammonia. A drop of dilute sulphuric acid in the supernatant fluid showed that excess of .lead was present. The mixture was made up to 250 c.c., filtered, and 100 C.C. of the filtrate made faintly acid with 9 C.C.20 per cent. sulphuric (by vol.) acid and made up to 250 C.C. One hundred C.C. of this solution were placed in the distilling flask with 400 C.C. water, 15 C.C. concentrated sulphuric acid and talc, and 300 C.C. distilled off at constant volume. The distillate was made up to 500 c.c., and two portions of 50 C.C. precipitated with 20 C.C. Scott- Wilson reagent (see below). Eight hundred C.C.were then distilled off at constant volume, with addition of 400 C.C. 0.2 per cent. potassium diohromate solution ; the distillate was made up to 1 litre, and two portions of 50 C.C. precipitated witb 20 C.C. Scott-Wilson reagent. Three hundred C.C. were then distilled off at constant volume with addition of water, and made up to 500 c.c.; 100 C.C.of this with 15 C.C. reagent gave no precipitate on standing overnight. I t is most convenient to have from 0.5 to 1 mgrm. of acetone in 25 to 50 C.C. of distillate. One mgrm. acetone requires 17.3 O.C. + potassium thiocyanate. The amount of acetone in urine can be judged roughly by colour tests. If Rothera’s reaction gives a strong colour within a few seconds, probably more than 0.25 per cent.aceto-aoetic acid is present. The ferric chloride reaction is not obvious unless more than 0.07 per cent. aceto-acetic acid is present. In order to form some idea of the amount of 6-hydroxybutyric acid, it may be remeh- bered that in well-marked cases of acetonuria this acid forms from 60 to 80 per cent. of the total ketonic bodies. I t is best to make up the Scott-Wilson reagent at least a week before it is required, and to filter through glass-wool before use.This reagent contains mercuric cyanide, silver nitrate, and caustic soda (J. Physiol., 1911,42,444). The titration is carried out with i& potassium thiocyanate solution, which can be standardised against silver nitrate. The first faint brown colour is the end-point ; three more drops of thiocyanate produce an obvious reddish-brown tint, and it is well always to obtain this second colour in order to be sure of the first.E. W. Yield and Composition of Indian Milk, and Errors in Milk Tests. J. W. Leather and A. C. Dobbs. (Mem. Dep. Agric. in India, 1914, 3, 147-175.) -The cows whose milk yield was investigated belonged to the Montgomery herd at Pusa, where the practice is to hand-milk one side of the udder whilst the calf sucks the other.I t was found in practice that under such conditions exactly half the total milk was yielded on each side, thus making it possible to calculate with accuracy the total yields, etc., of a series of trials extending over a year. TheBACTERIOLOGICAL, PHYSIOLOGICAL, ETC'. 57 probable error in the estimation of the percentage of fat in individual samples was found to be about k0.3.The maximum yield w a , ~ found to vary from 7 t o 20 pounds, and 12 pounds may be taken as the average figure; of this the calf took about 2 pounds when milking in the ordinary way for profit. The fat percentage varied from 3.5 to 5 per cent. Tests of the milk from the several quarters of the udder indicated that there axe differences in quality characteristic of the cow.Thus, one cow yielded milk from the fore-quarter which was systematically richer than that from the hind-quarter, while in another case the converse of this was observed. It would appear that the quality of the milk is a function of the tissue of that region of the cow's udder in which it is produced, and that, though the opposite sides are symmetrical in this respect, the udder is in other directions no more uniform in function than it is in shape. A very large amount of tabulated data is included in the paper, covering the following points amongst others: (1) Mean yield of milk morning and evening, (2) mean yield of milk from each half-udder, (3) error of the mean, (4) error of the individual milking, ( 5 ) total production of milk of each cow, (6) mean percentage in fat, (7) error of the mean, (8) error of the individual milking, (9) difference of percentage fat from different sides and different quarters of ihe udder, (10) difference in percentage fat in morning and evening milk, and (11) effect of meteorological conditions.The main interesii of the paper lies in the arithmetical treatment and analysis of the statistical data collected as applied to the checking of percentage errors.H. 3'. E. H. Standardisation of Commercial Papain. F. W. Heyl, C. R. Caryl, and J. F. Staley. (Amer. J. Pharm., 1914, 86, 542-550.)-Dried papaw juice was found to be capable of digesting at least 40 per cent. of egg albumin at 80" to 100" C., and no commercial samples of '' papain" had a higher proteolytic activity under those conditions. Twelve out of twenty-six samples examined had been diluted to such an extent with starch, sugar, or dextrin, that their digestive capacity was below that standard.The following method is recommended: The egg albumin solution is prepared by slightly beating the whites of six new-laid eggs, adding 2 volumes of 1 per cent.sodium chloride solution, filtering, and diluting the filtrate so that 15 C.C. contain 0.4 grm. of coagulable protein. Of this standardised solution 15 C.C. are placed in a 50 C.C. Erlenmeyer flask, and 1 C.C. of papain solution (prepared by taking up 1 grm. of the powder with salt solution, diluting to 100 c.c., and leaving for exactly thirty minutes), and 9 C.C.of 1 per cent. sodium chloride solution added. The flask is transferred to a thermostat at 80" C., and after fifteen minutes' digestion 1 C.C. of acetic acid is added, and the flask transferred in exactly one minute to a water-bath at 100" C., where it is kept for ten minutes. The undigested protein is collected on a tared filter, washed free from chlorides, and then successively with 10 C.C. of 95 per cent.alcohol and 10 C.C. of ether, and dried at 100' to 105" C. until constant in weight. Simultaneously a control estimation is made of the amount of protein coagulable by heat in 15 C.C. of the egg-white solution mixed with 10 C.C. of salt solution (or 9 C.C. of salt solution and 1 C.C. of papain solution in which the enzyme has been destroyed by boiling for fifteen minutes), the operations being carried out as described.The percentage of protein rendered non-coagulable under58 ABSTRACTS OF CHEMICAL PAPERS Per Cent. 43.4 to 58.9 5.6 to 50.i 22.3 to 31'5 nil these conditions is calculated. Test for Pepsin ; A mixture of 16 C.C. of the egg- white solution, 2 C.C. of 1 per cent. salt solution, 3 C.C. of : hydrochloric acid, and 5 C.C.of 1 per cent, papain solution (with 0.5 ch. toluene to prevent putrefaction), is digested for fifteen hours at 40° C. Twenty-five C.C. of 10 per cent. trichloracetic acid are then added, the liquid boiled for ten minutes, and the precipitate collected on a tared filter, washed with water, alcohol, and ether, and dried at looo to 105' C. until constant in weight.A control estimation is also made of the amount of coagulable protein present, and the percentage of protein digested under those conditions is calculated. Tryptophane Test No. 1 : Five C.C. of 5 per cent. Witte's peptone in 1 per oent. salt solution are mixed with 5 C.C. of 1 per cent. papain extract in 1 per cent. salt solution (toluene 1 drop), 1.2 C.C. of 1.71 per cent.hydrocyanic acid solution, and 1.2 C.C. of FT hydrochloric acid, and the mixture digested in glass- stoppered bottles for seventeen hours at 36O to 40' C., after which bromine-water is added drop by drop. Tryptopham Test No. 2 : The mixture of 5 C.C. of the Witte's peptone solution, 5 C.C. of the papain solution, and 1 C.C. of 1-71 per cent. hydro- cyanic acid solution, is digested for fifteen minutes at 80' C., and then tested with bromine-water. The intensity of the coloration is reported as faint, distinct, marked, strong, or deep.The following results are typical of those obtained (cf. ANALYST, 1914, 39, 444) : Per Cent. nil to 2 9 nil to 8.5 47.0 to 49.3 61.0 Dried papaw latex (8even samples) Commercial papain con- taining starch (fifteen samples) Papain containing sugar (tl?ree samples) Fictitious sample ...Marked to deep None to deep Per Per Per Per 1 2 i Per Cent. Cent. Cent. Cent. I Cent. Cent. 9.51 to nil 6'56 to 5-10 to, nil 9.S9 13-87 11-50 1-14 to 15.1 to 6-07 to 1-40 to nil to nil to i'S1 57% ll%Q S-62 1 3'26 1 11'19 Strang to deep Trace to strong Distinct to marked nil Faint 5-79 to nil 4'14 4-11 to 24'5*to - 1 - nil 1 ~ i 5 1 nil I 9-59 1 :E 1 * Calculated as lactose.C. A. M. Protein Substances of Barley,. Malt, and Wort. H. Schjerning. (Comptes Rendzcs des TravazLx du Lab. de Carlsberg, 1914, 11, 45-105.)-The present paper is a summary.of the author's work, published in the Comptes Rendw of the Carlsberg Laboratory in 1906, 1910, 1913 (ANALYST, 1910,35, 350; 1913, 38, 496), and includes a description of the analytical methods employed, some of which have been modified from time to time.By the employment of stannous chloride, mercuric chloride, ferric acetate, uranium acetate, and magnesium sulphrtte, as precipitating agents, taken in conjunction with total nitrogen (Kjeldahl) and ammoniacel nitrogen (distillation under diminished pressure with magnesium oxide), it is found possible to obtain values for the following nitrogenous substances present in barley, melt, and malt wort: (1) albumin I.(leucosin), (2) albumin 11. (edestin), (3) denuclein, (4) proteoses (albumoses), (5) peptone, (6) amino compounds and amides, and (7) ammonia. Of these, Nos. 3,4, end 5, represent the products of pepticBACTERIOLOGICAL, PHYSIOLOGICAL, ETC. 59 cleavage, and 6 and 7 those of tryptic cleavage.A large amount of data, including many tables and curves, is included, for which the original paper should be consulted. H. F. E. H. Determination of the Critical Moisture Content of Soils. R. 0. E. Davis. (J. Ind. and Eng. Clzem., 1914, 6, 1008-1010,)-The critical or optimum moisture con- tent of a soil is the moisture percentage at which the physical properties of the soil attain either a maximum or minimum value; it corresponds to that physicalcondition of the soil which is most suited to cultivation and plant growth.Previous methods for the determination of this constant consisted of long and tedious measurements of specific gravity of the soil at different moisture contents, or investigations into the relative penetrability of the soil as measured by the weight required to force a pointed instru- ment into it at different moisture contents.In this latter case it so happens that the minimum penetration moisture values coincide with the critical moisture content. The method recommended is based on the capacity of soils for holding moisture as a surface film by capillary attraction.A brass tube 12 inches long and 14 inches in diameter is split longitudinally down the middle, one-half being fitted with a cslluloid window. The two halves are kept together in position when in use by binding with copper wire; the tube is filled with the soil under examination, which is shaken well down after being freed from lumps in 8 mortar, and closed at one end with a piece of cotton cloth secured by a rubber band.The closed end is then immersed in water. When the water has advanced several inches, the tube is removed from the water, laid down horizontally, and the wet end covered with a piece of rubber sheeting to prevent evaporation. The length of the soil column should be such that the length of dry soil beyond the extreme point to which the moisture has advanced is at least 2 inches.The tube is then carefully opened and a moisture determination made of the moist column of soil in the inch at the extreme end t o which water has advanced. The percentage of moisture so found is the critical water content. H. F. E. H. Estimation of Urea, and Indirectly of Allantoin, in Urine by Means of Urease. R. H. A. Plimmer and R.F. Skelton. (Biochenz. J., 1914, 8, 70-74.) -The estimation of urea in urine is quickly and accurately made by decomposing it with urease (1 grm. powdered soya-bean) at 35" to 40" C. for one hour. By fitting together three or four cylinders and Allihn bottles in series with a sulphuric acid bottle at the end, duplicate estimations of ammonia and urea in urine can be carried out simultaneously.In the cylinders for the urea estimations are placed 50 to 60 C.C. of water, 1 grm. of finely ground soya-bean, and 5 or 10 C.C. of urine. These cylinders are kept in a water-bath at a temperature of 35' to 40' C., an air current being drawn through the series. After about an hour the cylinders and bottles are disconnected, and 1 grm. of anhydrous sodium carbonate is dropped into the cylinders; they are then connected together again, and the air current drawn through for another hour.To prevent frothing liquid paraffin (B.P.) is used; it is superior to petroleum or toluene, 8s it does not evaporate, and it obviates the necessity of using a tube containing cotton-wool between the cylinder and Allihn60 ABSTRACTS OF CHEMICAL PAPERS bottle.It is unnecessary to carry out a control experiment with soya-bean alone, since no ammonia was evolved by two different samples of the bean which were tested several times. The Allihn bottles are charged with excess of fg sulphuric aoid (25 to 50 c.c.), which is titrated with i: alkali, using alizarin red as indicator. The method was tested on solutions containing different amounts of pure urea, and the results were found to agree with a nitrogen determination by Kjeldahl's method.Urease does not decompose allantoin; and since both allantoin and urea, are quanti- ltatively decomposed by the magnesium chloride method of Folin, the amount of allantoin in those urines which contain both compounds is readily estimated by difference. E. W. Enzymes of Washed Zymin and Dried Yeast (Lebedeff).111. Peroxy- dase, Catalase, Invertase, and Maltase. A. Harden and S. S. Zilva. (Biochem. J., 1914, 8, 217-227.)-Fresh English brewery yeast gives the peroxydase reaction with hydrogen peroxide end p-pbenylene-diamine. When, however, the yeast is dried for seventeen hours at a temperature of 37" C., it no longer gives the reaction ; but on washing the dried yeast the presence of peroxydase can again be detected.Dried Munich yeast behaves like dried English yeast, and does not react for peroxy- dase when unwashed, but regains that activity on being washed, The addition of the washings to washed dried yeast inhibits the action of peroxydase. It is evident that washing removes some substance from dried yeast which has an inhibiting effect on the peroxydase of the yeast. The washings not only inhibit the action of the peroxydase in yeast, but also exert a deterrent effect on that of milk. Washed and unwashed Munich dried yeast, washed and unwashed fresh English brewery yeast, washed and unwashed English dried yeast, give no reaction with p-phenylene dirtmine and with benzidine in absence of hydrogen peroxide, which proves the absence of an oxydase. Similarly, the above-mentioned yeasts under the same con- ditions give no reaction for tgrosinase. The catalase in dried Munich yeast is capable of liberating the entire available oxygen of a considerable quantity of hydrogen peroxide (0.1 grm. yeast completely decomposed 25 C.C. hydrogen peroxide i n five minutes). The washing of the yeast makes no difference either to the volume .of the oxygen liberated or to the rate of evolution. The power of hydrolysing cane- sugar is partially, but not entirely, removed from zymin and from dried Munich yeast by washing at ordinary room temperature, whilst the power of hydrolysing maltose is .not affected. E. W.

ISSN:0003-2654    

DOI:10.1039/AN9154000056

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

60 ABSTRACTS OF CHEMICAL PAPERS ORGANIC ANALYSIS. Methods for Determining the Melting-Point of Asphalts. J. G. Miller ,,and P. P. Sharples. (Chm. Engiwer, 1914, 20, 217-218.)-1n the case of coal-tar pitches with melting-points UP to 170' F., the " &-inch cube method in water " is recognised in America as the standard, but the present investigation shows that the attempts to apply the method to asphalts of higher melting-points, by substituting a bath of cotton-seed oil for the water, do not give satisfactory results.Perfectly satisfactory values were obtained by Hubbard's method, in which the cube is heated in a bath of air in a beaker of 300 C.C. capacity enclosed in another, of 800 c.c.,ORGANIC ANALYSIS 61 'U containing cotton-seed oil. I n all these cases the &-inch cubes are suspended 1 inch from the bottom of the apparatus by a copper wire 0.0808 inch thick (No.12, B. and S. gauge) and heated at the rate of 9" F. ( 5 O C.) per minute; the melting-point is taken when the cube touches the bottom of the apparatus. Many manufacturers in America have abandoned the cube methods in favour of the '( ring and ball " method. This latter would prove an acceptable and easy laboratory method if it were rigidly standardised. At present, however, the dimensions of the brass rings and steel balls employed in the various laboratories differ so widely that concordant results are impossible, a difference of 25' F.being observed between observations made with two of the principal types of this apparatus. The melting-points determined by the cube method are not comparable with those found by the ring and ball method, even with the use of a factor, since the relative difference depends on the kind of asphalt under examination (cf.ANALYST, 1914, 39, 264). J. F. B. Some Properties of Solutions of the Boric Acids in Alcohol. A Modified Boiling-Point Ap- paratus. J. B. Firth and J. E. Myers.(J. Chem. SOC., 1914, 105, 2887-2892.)-The apparatus shown in the figure was devised in the course of a, study of the influence of boric acid upon the boiling-point of dry alcohol. It oonsists of an ordinary Beckmann boiling- point tube enclosed in a Dewar vacuum vessel. All exposed glass surfaces are covered with a layer of eotton-lint to shield the heated liquid and vapour from draughts.The liquid is heated by an electric current of 2.5 to 3 ampikes, regulated by a rheostat with an ammeter in the circuit. The current is carried by platinum wires sealed through glass tubes containing mercury to make contact with the leads, The platinum wire is stout until it dips below the level of the liquid, whence it is continued in the form of a thin wire which is passed through an open coil of glass tubing 2 mm.in diameter. The glass coil is broken halfway along its length, so as to leave a, gap ( A ) between the upper and the lower halves about 4 mm. wide. As the wire becomes heated a circulation of the liquid is set up through the narrow coil, and when the boiling-point is reached streame of bubbles should issue from the openings at the top, bottom, and middle, of the coil.Ebullition at all three points is essential in order to avoid superheating, and can be insured by slightly tapping the apparatus. With pure liquids, the boiling-point, registered by a Beckmsnn thermometer, remains steady within 0*01" or even 0.005' C., so long as the barometric pressure is steady. In experiments made with orthoboric and metaboric acids and boron trioxide, it was found that small proportions of these bodies lowered the boiling-point of pure ethyl62 ABSTRACTS OF CHEMICAL PAPERS alcohol, while larger proportions raised it.The maximum lowering of the boiling- point was produced by the least volatile solute (boron trioxide), which also had the greatest subsequent effect in raising the boiling-point as the proportions were increased, The distillates were found to contain the original solutes.J. F. B. Modification of Kj eldahl's Method of Estimating Nitrogen in Organic Compounds. M. Wunder and 0. Lascar. (J. Pharm. Chzm., 1914, 19, 389-333.) -The following modification of Kjeldahl's method gives good results, even in the case of compounds, such as antipyrin or isatin, which show low results with other modifications.From 0.1 to 0-6 grm. of the substance, according to the proportion of nitrogen, is weighed in a small thin glass.cylinder (15 mm. x 10 mm.), which is sealed up in the case of volatile compounds. This cylinder is introduced into the Kjeldahl flask, in which have already been placed 3 grms. of oxalic wid, 2 grms. of sodium oxalate, and 0.5 grm. of powdered recently calcined vanadium oxide (V,O,).A mixture of 5 C.C. of phosphoric acid (sp. gr. 1.71) and 25 C.C. of strong sulphuric acid is added, and the flask gently heated, with frequent agitation, until the organic matter is destroyed and the liquid is of a greenish-yellow colour. I t is then cooled, its contents diluted with 100 to 200 C.C. of water, and, after the introduction of two or three pieces of iron wire (0.5 to 0.75 grm.), gently heated for thirty minutes, after which the liquid is distilled with caustic alkali in the usual way.When the sub- stance contains nitrogen united to oxygen, it is necessary to add 2 to 3 grms. of sodium oxalate or formate and a few pieces of iron wire (0.75 grm.) during the initial digestion with sulphuric acid, and a further 0.5 grm.of wire after the destruction of organic matter, while the quantity of sulphuric acid should be increased to 40 C.C. I t is essential also that the destruction should be carried out very gradually. C. A. M. Notes on Essential Oils. (Schimmel and Co.'s Semi-Annual Report, April, 1914.)-Ambrette Seed Oil.-A series of these oils derived from seed from the West Indies and Java gave the following values: Sp.gr. at 15O C., 0.9088 to 0.9161; [+= +lo 19' to -2O 24'; 1.47421 to 1.48013; acid value, 0.8 to 2.4; ester value, 137.7 to 180.5; and solubility in 80 per cent. alcohol, 1 in 2.5 to 6 vols. Ashanti Pepper Oil.-Two samples of the fruit of Piper guiwernse from West Africa yielded 11.5 and 10.96 per cent. of oil with the following charaoters : Sp.gr., at 15' C., 0.8733 and 0.8788 ; [a],,= - 3' 43' and - 5' 34' ; [N]D ~OOC., 1.48905 and 1.48847 ; acid value, 0.6 and 0.9 ; ester value, 5.5 and 4.2 ; and solubility in 90 per cent. alcohol, 1 : 8. Both samples gave a pronounced phellandrene reaction. Oil ojSzoeet Basil..-A sample from Central Africa gave results in agreement with those of European oils. Cathetus Fasciczdata Oil.-A sample from Annam (where the plant is known as Bruydre d'Annam) was of a bluish-green colour, and had an odour resembling cajuput Oil.Its constants were-sp. gr. at 15O c., 0.8897 ; [a]D= - 4' 34' ; "ID 200 c., 1.4754 ; acid value, 1.9 ; ester value, 3.7 ; ester value after scetylation, 44.8 = 12.7 per cent.ORGANIC ANALYSIS 63 at Sp. 15" Gr. C, Cl,H,,O ; and solubility in 90 per cent.alcohol, 1 in 0.5 vol. It contained cineol and apparently cymene and linalool. Citronella Oil.-Samples of Java oils contained from 26 to 44-4 per cent. of geraniol and 23.4 to 50.1 per cent. of citronellal, while Ceylon oils containod from 26.7 to 38.8 per cent. of geraniol and 5-5 to 10.5 per cent. of citronellal. The Java oils containing a high proportion of citronellal had a delicate aroma; they were also characterised by a very low sp.gr. Helichr~sum Angustijiolium Oil. -Two samples from Dalmatia gave the following values : Sp. gr. at 15' C., 0.8923 and 0.8964 ; [uID= - 9' 40' and - go 38' ; "ID 200 c., 1.48490 and 1.48422; acid value, 2.8 and 1.9; ester value, 39.2 and 56.2; and solubility in 90 per cent. alcohol 1 in 8 to 9, and 1 in 10 vols.(cf. AXALPST, 1915, 17). Lemon Grass Oil.-Three samples of oils from the Seychelles had the following characters : Aldehyde Content. [.ID 2OOC. Solubility. Bisul hite Sulphite Met%od. Method, - - _- - - - -- .. . 0.8954 0.9101 - . . - - - . . -_ - Per Cent. -0' 15' 1.48905 79 -0' 25' 3.49240 77 0-9008 1 -2'40' Per Cent. 76 - 86 70 - - 81 -_ . . . .. - .-- - In 2 vols.90 per cent. In 4 vols. 90 per cent. In 2.3 vols. 70 per cent. dcohol. alcohol. alcohol. I I I 1 I The difference in solubility of one of the samples suggests that they may have been derived from different species of plants, Cympobogon$?exzmsus (yielding a readily soluble oil) and C. citratus (yielding a sparingly soluble oil). Machilus Oil.-A sample from Japan was a solid yellowish mass, melting above 58' C.to a yellow liquid. I t had an odour somewhat recalling that of guaiacum- wood oil. Its constants were as follows : Sp. gr. at 50' C., 0.9482; [a]== + 27' 50' (in 10 per cent. alcoholic solution); acid value, 1.2; ester value, 5.4; ester value after acetylation, 155.5; and solubility in 70 per cent. alcohol, 1 : 3. It contained B sesquiterpene alcohol of m.-p.82' C. Ocimcum Canum Oil from Central Africa had the following characters : Sp. gr. at 30' C., 1.0431 ; [ u ] ~ = - 2' 25' ; [N]D 3oo c., 1.55611 ; acid value, 0.6 ; ester value, 289.8; and solubility in 80 per cent. alcohol, 1 : 1. I t contained methyl cinnrtmltte; but d-camphor, which was a constituent of Ocimum canum oil from the island of Mayotte, was not present.Per% Balsam Oil,-A sample had the following characters : Sp. gr. at 15' C., 1-1200; [alD= +Oo 55'; [N]D200C., 1.57177; mid value, 36.4; ester value, 228-2; and solubility in 90 per cent. alcohol, 1 in 0.5 vol. It contained a sesquiterpene alcohol, Cl,H,,O (b.-p. 125' to 127O C. ; sp, gr., 0.8987 ; [u],, = + 12' 22' ; [N]D 403 c., 1*48982), identical with Hesse and Zeitschel's nerolidol, but the '( peruviol " of Thoms was not found.C. A. M.64 ABSTRACTS OF CHEMICAL PAPERS Iodine Value of Linseed and Petroleum Oils. W. H. Smith and J. B. Tuttle. (J. Ind. and Eng. Chem., 1914, 6, 994-998.)-The iodine values of raw, boiled, and burnt linseed oils were estimated by the Hanh method (ANALYST, 1902, 27, 15; 198) with the object of determining tbe influence of varying excess of iodine, time o€ absorption, and temperature.The results obtained show that con- cordant figures can be obtained only by adhering to an exact method of procedure, particularly in the case of burnt linseed oil (oil which has been heated or ignited until it has acquired a viscosity rendering it suitable for use in the manufacture of printing ink). With raw linseed oil, however, a constant value is reached within comparatively wide limits of weight of oil and excess of iodine.The substitution of iodine which takes place in the reaction might be diminished by working at a low temperature, and it might be feasible to improve the Raniis method in this respect. As regards the estimation OF the iodine value of mineral oils, the authors suggest that not less than 1 grm.of the oil should be used with 25 C.C. of the iodine solution; the method is particularly in need of standardisation when used for these oils. w. P. s. Estimation of Naphthalene in Spent Oxide. W. C. Davis. ( J . Sot. C J M ~ . liad., 1914, 33, 1120-1121.)-The naphthalene is separated from the spent oxide by steam-distillation and estimated as picrate.Ten grms. of the sample are placed in a Wurtz flask through which a, current of steam is passed; the steam and vapours of the distillate are passed through a second small Wurtz flask heated to 100" C., and containing 50 C.C. of sulphuric or citric acid, then into a condenser, and are finally collected in a receiver provided with an inlet and outlet tube. The latter is sealed by connecting it with a small wash-bottle containing 25 C.C.of saturated picric acid solution. At the end of the distillation, which takes about ten minutes, the water in the condenser is regulated so as to allow any naphthalene in the condenser to melt and run into the receiver. The picric acid solution in the wash-bottle is then washed into the receiver, a further quantity of 125 C.C.of saturated picric acid solu- tion is added, and the receiver is closed with a rubber stopper and heated in a water- bath until a clear solution is obtained. On cooling, the naphthalene picrate crystallises out, and the excess of picric acid is estimated by titration with & sodium hydroxide solution, using lacmoid as indicator. The purity of the naphthalene picrate obtained is ascertained by determining the melting-point of the naphthalene separated from the picrate by steam-distillation; in experiments carried out by the author, the melting-point was from 1" to 3" below that of pure naphthalene, which melts at 79" C., showing that the precipitate contained only a small quantity, if any, of other aromatio picrates.Slightly larger quantities of naphthalene are found if, after the picric acid solution has been washed into the receiver, a quantity of solid picric acid is added sufticient to saturate the whole solution.For example, a spent oxide at first found to contain from 2.57 to 2.63 per cent. of naphthalene yielded from 2-77 to 2.92 per cent. after the distillate had been saturated with picric acid. W. P. S. Analysis of Paper.H. A. Bromley. (Chem. Engineer, 1914, 20, 253-258.)- In the estimation of fibrous composition, purely chemical methods may be used inORGANIC ANALYSIS 65 two cases-esparto and mechanical wood. The percentage of esparto may be calculated from the yield of furfural by distillation with hydrochloric acid, after the manner of pentosans, on the assumption that 12.5 parts of furfural correspond to 100 of esparto.A number of chemical methods for the estimation of mechanical wood are noted-e.g. , colorimetric esbimation with dimethyl-p-phenylenediamine ; reduction of gold chloride (Godeffroy and Coulon) ; absorption of phloroglucinol ; titration of the hydrochloric acid formed on chlorination. In the estimation of mineral matters, the ash of those specimens containing calcium or barium sulphate should be moistened with sulphuric acid and re-incinerated ; corrections may be applied for the moisture expelled from the minerals (20 per cent.for calcium sulphate, 12 per cent. for china clay). Quantitative analysis of the ash includes estimations of aluminium, calcium, barium, magnesium, sulphuric acid, and silica. Nitrogenous .sizing materials are estimated by Kjeldabl’s method.The estimation of resin is best carried out by Sammet’s method: Five grms. of paper are extracted with acidified alcohol (100 C.C. of 95 per cent. alcohol and 15 C.C. of 5 per cent. aqueous acetic acid), and the extract, evaporated to B small bulk, is taken up with ether and shaken with water. For the estimation of starch, the author recommends the con- version of the starch into soluble sugars either by diastase or boiling acid, and estimation of the reducing sugar with Fehling’s solution.Pigments which resist incineration may be estimated colorimetrically in the ash by comparison with etandard specimens mixed with china clay; other pigments may be estimated by determination of their main constituents.Salicylic acid, present as a preservative, may be extracted with light petroleum and titrated in alcoholic solution. Carbolic acid is extracted with alcohol, the extract treated with 10 per cent. sodium hydroxide solution, the tar oils and naphthalene separated by filtration, and the tar acid8 liberated by acidification in presence of brine ; the resin remains suspended in the liquid, which is then drawn off as completely as possible, and the layer of tar acids taken up with ether or light petroleum.Soluble chlorides, left in the paper through imperfect removal of bleach residues, are extracted by hot water and estimated as silver chloride. Alum is extracted by hot water acidified by 1 to 2 drops of sulphuric acid. In an aqueous extract the total acidity of the paper is determined by titration in presence of litmus and soluble sulphur compounds by titration with & iodine.Arsenic in fly-papers may be titrated with iodine after extraction with sodium bicarbonate ; small quantities are estimated by the Marsh Bereelius method. J. F; B. Estimation of Pentosans. J. van Haarst and S. C. J. Olivier. (Chm. Weekblad, 1914, 11, 918.)-Furfural is decomposed by hydrochloric acid of the strength used (sp.gr. 1-06), so that rapid distillation, to remove the furfural as fast as it is formed, is advantageous. The method of Tollens gives higher and more accurate results than that of the Codez Alimenturiw. A standard rate of distilla- tion is necessary. The furfural is weighed as the phloroglucinol compound, which is also better as a qualitative test than the aniline acetate compound.Arabinose and pentosans yielding arabinose require rapid distillation with larger quantities of acid, on account of their slow decomposition. 0. E. M.66 ABSTRACTS OF CHEMICAL PAPERS Hysteresis Tests for Rubber. E. L, Davies. ( J . Id. and Ertg. Chem., 1914,6, 985-986.)-The expert usually judges a piece of rubber by means of a crude hysteresis test, which he performs by stretching a small strip with his fingers.Experience enables him to judge closely, but by no means accurately, small differences between two samples; it does not enable him to standardise his tests, nor to make his results available to others. Several machines have been devised to perform and record these tests graphically, but they have not come into their full usefulness, due to the difficulty encountered in translating the graph into terms which are intelligible and comparable.The object of the present paper is to point out and explain some of the relationships between the mathematical equation for the curve and the properties of the rubber being tested, and to point out the close relationships which some of the tests, made in the laboratory, show between the theoretical curve and actual curves made by the machine.A typical form of curve produced by the machine is reproduced, the ordinates representing tension and the abscissae stretch. Cheneveau and Heim (Compt. Rend., 1911, 152, 320) have shown that this curve has the equation x = cy + a sin2 by, and that, when OB is drawn tangent to the curve through the origin, c=the tangent of the angle contained between OB and the axis of Y.I t is shown (1) that c is dependent solely upon the initial resistance of the rubber to stretching; (2) that a will be relatively large for pure gum stock in comparison with I‘ tread ” stock, which will be strong and cause the entire curve to be fairly steep, but will be small in comparison with a “whiting stock,” which is charaoterised by an initial stiffness, after which it offers compara- tively little resistance to stretching; and (3) that b will be large for pure gum and tread” stocks.The ease with which these values could be standardised in specifications for a given stock is apparent, and the value of such standmds, both from the point of view of the manufacturer and the buyer, is said to be obvious by the author, who is himself concerned with the manufacture of tyres and other rubber goods.Experimental work has been conducted to find how closely the curve computed from the equation agrees with the actual curve, of which a typical example is given. Test specimens of a well-known make of inner tube were cut and tested with a Schwatz rubber-testing machine, and from the curves drawn in testing these speci- mens constants were obtained which gave the equation x= 1.17 y+ 2-06 sin2 34.6 y.Another tube made by the same formula, was selected at random, and a curve drawn from a sample of this tube. Assuming various values for y, the corresponding values of x were computed from the equation, and compared with the same values as measured on the curve.The agreement is remarkable, the maximum difference between any of the computed and actual values of x being less than 2 per cent. G. C. J. Identification of Artificial Silks. L. J. Matos. (Chem. Ertgirneer, 1914, 20, 209-210.) -The following simple chemical tests are described for distinguishing between the various commeroial artificial silks : A small tuft of the silk is heated in a dry tube, and the vapours are tested with litmus-paper.Alkane vapours indicate gehtin silk, acid vapours silks of cellulose basis. With iodine-zinc chloride andINORGANIC ANALYSIS 67 iodine-sulphuric acid, the reagents being made at suitable concentration,pellulose acetate silk is stained yellow, and the cellulose silks reddish-violet or blue. In cold concentrated sulphuric acid the collodion and viscose cellulose silks dissolve rapidly, cuprammonium silks only slowly. In warm 40 per cent. caustic potash solution the cellulose and cellulose acetate silks are swollen, while gelatin silk dissolves rapidly and completely. I n cuprammonium reagent the cellulose silks swell and dissolve, cellulose acetate silk swells without, dissolving, and gelatin silk takes a bluish-violet coloration without dissolving. The nickel-oxide-ammonia reagent, both cold and warm, causes swelling of the cellulose and cellulose acetate silks without dissolving ; gelatin silk takes a brown coloration without dissolving. Alkaline copper glycerol solution has no action on the cellulose and cellulose acetate silks on boiling, but gelatine silk dissolves in a short time. Diphenylamine-sulphuric acid gives a blue coloration with collodion silk, and serves to distinguish this from the other varieties of cellulose silk. Dyed samples should be ‘‘ stripped ” by means of hydrosulphite before applying the reagents. The tests should be controlled by comparison with type samples of known origin. J. F. B.

ISSN:0003-2654    

DOI:10.1039/AN9154000060

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

INORGANIC ANALYSIS 67 INORGANIC ANALYSIS. Estimation of Cuprous and Cupric Sulphides in Mixtures of Both. E. Posnjak. (J, Amer. Chem. Soc., 1914, 36, 2475-2479.)-The method depends on the fact that cuprous sulphide reacts with silver salts, with the formation of equivalent amounts of metallic silver and silver sulphide, whilst under similar conditions cupric sulphide gives rise only to silver sulphide.The metallic silver in the mixture of silver and silver sulphide is determined by solution in ferric nitrate solution and subsequent estimation as chloride. The substance (0.4 to 0.5 grm.) is ground to pass a sieve of 200 meshes per lineal inch, and treated with 50 C.C. of 5 per cent. silver nitrate solution. The reaction is rapid at first, even in the cold, but is only quantitative if allowed to proceed for some three hours on the steam-bath with frequent agitation, the products of the reaction tending to protect the material from the reagent.The precipitate is washed by decantation until the filtrate no longer gives a precipitate with hydro- chloric acid, and it is then extracted two or three times at about 70' C. with 40 to 50 C.C. of a 6 per cent. solution of ferric nitrate (anhydrous).The extract is filtered each time through a hot-water funnel, and the residue finally washed.with hot water. To the filtrate nitric acid is added until it becomes colourless, and the silver is then precipitated and weighed as chloride. 2Ag = Cu&. The residual silver sulphide is roasted, dissolved in nitric acid, and the silver precipitated as chloride. If the original sample contained cuprous sulphide, but not cupric sulphide, the weight of silver chloride in this second experiment will equal that in the first, within the limits of experimental error of the method, about 1.5 per cent.If the second precipitate is appreciably heavier than the first, then the cupric sulphide in the sample is calculated from the difference between the weights of silver chloride obtained in the two experiments.2Ag- GUS. G. C. J.68 ABSTRACTS OF CHEMICAL PAPERS Volqmetric Estimation of Copper in its Salts and Alloys, G. Zuccari. (Annuli Chim. JppZic., 1914, 2, 287-290.)-The solution of the copper salt is titrated with a standard solution (48.866 grms. per litre) of pure crystalline sodium nitro- prusside, the end-point of the reaction being found by a spotting test with an alkaline sulphide applied to a few drops of the filtered liquid.Each C.C. of the reagent; corresponds to 0.01 grm. of metallic copper. The presence of ferric salts, zinc (up to 50 grms. per litre), alkaline earth metals, manganese, aluminium, tin, or lead, does not affect the results, but ferrous salts cause aturbidity and eventual precipitation of the reagent.I t is therefore necessary to boil the copper solution with nitric acid to convert all iron salts into the ferric condition before the titration, the free nitric acid not interfering with the process. The method is also directly applicable to alloys of copper not containing cadmium or nickel, provided that the solution is not of greater concentration than 2 to 3 per cent.The standard solution will keep unaltered for months if protected from light and air, especially if slightly acidified with sulphuric acid. The results thus obtained with solutions of the principal salts of copper containing commercial impurities were in agreement with those found by electrolytic methods. C. A. M. Estimation of Hydrogen in Gas Mixtures by Means of Colloidal Palladium. G.A. Burrell and G. G. Oberfell. (J. Ind. and Eng. Chem., 1914, 6, 992-994.)-The authors have examined the method described by Paal and flartmrtnn (ANALYST, 1910, 35, 139), and express the opinion that its advantage over the use of palladium asbestos or palladium sponge lies in the fact that it never fails to work satisfactorily, which cannot be said of methods which make use of the other substances.The reagent is best prepared according to Brunck’s formula, (ANALYST, 1911, 36, 84), as it then has a longer life, Theoretically, 1 C.C. of the solution should absorb 40 C.C. of hydrogen, but when a little more than one quarter of this amount has been absorbed it becomes too sluggish for use in an ordinary Orsat pipette packed with glass tubes.In their experience, as long as twenty-five minutes may be required for the absorption of 30 C.C. of hydrogen by a tolerably fresh solution, but they agree that a shorter time would probably suf3tice when working, according fo Paal and Hartmann’s original directions, with an ordinary Hempel pipette, which permits of agitation, [Note by Abstractor.-Against this must be set the tendency to froth under such conditions, as noted by Hempel (ANALYST, 1912, 37, 518).] In contradiction to Hempel (Zoc.cit.), the authors find the reagent does not deteriorate with age if stored in the dark-at any rate, not appreciably in eight months. The sole disadvantages of the method are its slowness and the somewhat troublesome method of regeneration of the reagent ; but these are more than outweighed by the advantages above-stated, and by the fact that the reagent can be placed in an ordinary pipette and does not need to be heated.G. C. J. Modification of Jager’s Method for Estimating Hydrogen and Methane. J. P. Wibaut. (Chern. Weekbld, 1914, 11, 498.)-Cerium oxide is added in sufficient quantity to produce a greyish-white mixture with the copper oxide used inINORGANIC ANALYSIS 69 Jiiger's method of fractional combustion.The combustion is more rapid and the end-point sharper. Analyses of a mixture by this method and by absorption of hydrogen in palladium black, and explosion of methane, show satisfactory agreement. 0. E. M. Corrosion of Iron, and its Application to Determine the Relative Strengths of Acids.J. A. N. Friend and C. W. Marshall. (J. Chem. SOC., 1914, 105, 2776-2782.)-Experiments were made to determine the minimum concentration of alkali required to inhibit the corrosion of iron by solutions of sodium salts, and trace the possible relationship to the relative strengths of the acids combined in the salts. The quantitative results with hydroxides were irregular owing to the presence of carbon dioxide in the air, but with carbonates and borates of the alkali metals they were quite consistent.The experiments were carried out in a series of tubes charged with 5 C.C. of the salt solution and increasing quantities of sodium carbonate solution; the volume was made up to 10 c.c., and a piece of clean iron-foil 1 cm. square placed in each tube.The results were appreciable after two to three days, the tube being rotated in a strong illumination for the detection of minute spots of corrosion. By repeating the tests with intermediate quantities of alkali, the quantity of the latter required to prevent corrosion could be estimated with an error of about 5 per cent. In a series of sodium salts the order of inhibiting concentrations of carbonate was the same as that of the electrical conductivities of the free acids of* the salts, and the same relative values were approximately maintained.With increasing dilution of the salt solutions there was a tendency towards equalisation of the inhibiting oarbonate concentrations, just as the relative strengths of the various acids approach equality with increasing dilution, in accordance with the ionic theory.With sodium borate as the inhibitor, the results showed a reasonably close similarity to those obtained with the carbonate, with the exception that the bromide and iodide, both close together, exchanged places. In the table below, the amounts of sodium carbonate, expressed as C.C. of N-solution per 10 C.C.of liquid, required to inhibit corrosion by salt solutions at different concentrations are set forth : Sodium Salt: 1 N I Chloride ... 1 - i - Nitrate ... ; 10.0 7.6 Sulphate ... 4.8 3-2 Fluoride ... / - 1 1-4 Sulphite ... . - - - ! - -.. 1 Acetate I 1 ' zG i i& 2.70 1 1.05 1.45 1 0.75 1.40 1 0.80 1.05 0.75 0.05 0.07 0.24 i 0.14 Certain salts of feeble acids-e.g., sodium acetate, sulphite, arsenate, phosphate, carbonate, and borate-which are capable of inhibiting corrosion when fairly concentrated (see zero values for acetate and sulphite in table), cannot do so when dilute ; the limits of this '' auto-inhibition " were determined in several cases, but the interpretation of the values is not perfectly clear.J. F. B.70 ABSTRACTS OF CHEMICAL PAPERS Estimation of Nitrogen in Steel.L. E. Barton. ( J . I d . and Eng. Chem., 1914, 6, 1012-1013.)-This is usually effected by the method of A. H. Allen, as subse- quently modified by J. W. Langley. The method is given in Blair’s t‘ Chemical Analysis of Iron,” and the present author follows the directions there given, he does not restate them. The method depends on the use of Nessler’s solution, and its practical diEculty depends on the fact that standards prepared in the ordinary way never match the assay solutions satisfactorily, being orange-yellow in comparison with brownish-yellow.The modification now recommended overcomes this difficulty. About 500 C.C. of water and 40 C.C. of sodium hydroxide solution (cf. Blair) are distilled from a large distilling flask until the distillate gives no reaction with Nessler solution.A solution of 5 grms. of the steel in 40 C.C. of ammonia-free hydrochloric acid is now added to the contents of the still, and 150 C.C. of distillate, which will contain all the nitrogen in the sample, is collected. About 150 C.C. of ammonia-free water and exactly 25 C.C. of standard ammonium chloride (0.01 rngrm.N per c.c.) are next added to the contents of the still, and 150 C.C. of distillate again collected. As before, all the ammonia will pass over in the first 150 c.c., to which 6 C.C. of Nessler’s reagent is added; whilst 1 C.C. of Nessler’s reagent is added to 30 C.C. of the distillate from the assay distillation, and comparison made at the end of one minute or within ten minutes.The full colour develops within about a minute, and after ten minutes a cloudiness may arise. In amplification of Blair’s directions for the preparation of ammonia-free hydrochloric acid, the author says this is readily obtained by diluting acid to sp. gr. 1.1, distilling without the addition of sulphuric acid, rejecting the first tenth of the distillate, which will usually contain all the ammonia, and collecting the rest, testing each portion, of course, for freedom from ammonia. G.C. 5. Estimation of Peroxide in Commercial Lead Oxide. L. S. Dean. (Chem. News, 1915, 111, 2.)-The method described is not new, but is essentidly that of Andrews (J. Amer. Chenz. SOC., 1903, 23, 792). I t depends on the decomposition of the peroxide by hydrochloric acid in presence of potassium iodide, using chloroform or carbon tetrachloride as an indicator, and titrating back the excess of potassium iodide with potassium iodate.The iodate solution may be made up accurately by weighing out the pure salt, and the iodide solution is standardised against the iodate. The equations involved are- 4KI + 2K103+ 12HCl= 6KCl+ 61Cl+ 6H20 and 4KI + 2Pb0, + KIO, + 14HC1= 2PbC1, + 5KC1+ 5IC1+ 7H20 ; SO that KIO, = 2Pb0,.The advantages of the method are that it requires no special apparatus, and is available in presence of organic matter and of undissolved matter. This last is a great advantage in analysing litharge containing very small quantities of peroxide, since large portions must be taken for analysis, but, when using this method, need not be wholly dissolved.The results agree well with those obtained by Bunsen’s method. G. C. J.INORGANIC ANALYSIS 71 Systems : (A) Water, Magnesium Carbonate, and Carbonic Acid ; and @) Water, Calcium Carbonate, Magnesium Carbonate, and Carbonic Acid. J. w. Leather and J. N. Sen. (Mern, Dept. Agric. in Ind., 1914, 3, 205-234.) ( A ) The concentration of solution of Mg(HCO,), in respect of any specified partial pressure of carbonic acid is from five to fifteen times greater than that of the corres- ponding calcium salt.The ratio of MgCO, : H,CO, in the solution is much more nearly unimolecular than in the case of the calcium salt ; in fact, the solution contains only a slight excess of the quantity of H2C0, required by the bicarbonate formula.The solution of Mg(HCO,), is much more stable than that of the calcium salt. (B) When a mixture of calcium and magnesium carbonates is subjected to the action of carbonic acid and water, the calcium carbonate is largely or wholly pre- vented from dissolving, a fact in accordance with the indication of the law of mass action. Dolomite was found by the authors, as by previous workers, to dissolve in carbonic acid as such, but apparently the concentration obtainable does not depend only on that of the carbonic acid.Dolomite is protected from the action of calcium carbonate to a great extent by the presence of calcium carbonate, while magnesium carbonate protects it completely. No explanation can be offered of the observation that the addition of calcium carbonate to a solution of dolomite in presence of excess of carbonic acid seems to cause a slight precipitation of magnesium carbonate.The addition of magnesium carbonate in any material quantity to agricultural land may readily cause an almost complete precipitation of calcium carbonate, and hence render the soil infertile. This fact probably explains in part the crop failure which has so often accompanied the application of magnesium carbonate to land in experi- ments on Loew’s (‘ lime-magnesia ratio.” H.F. E. H. Qualitative Test for Water by Use of Acetylene-Cuprous Chloride Reaction. E. R. Weaver. (J. Amer. Chew&. SOL, 1914, 36, 2462-2468.)-The principle of the method is treatment of the substance under examination with calcium carbide and an anhydrous solvent of acetylene, preferably but not necessarily a solvent of the substance itself also, followed by decanting or distilling the solution into an ammoniacal solution of cuprous chloride.When the substance to be examined is a liquid, no other solvent for acetylene is necessary, acetylene being appreciably soluble in nearly all organic liquids. In such cases the calcium carbide, which always contains some occluded acetylene, is deprived of this by repeated treatment with anhydrous ether, which is then boiled away.The liquid to be examined is next poured over the carbide, the mixture shaken in a closed tube, the carbide allowed to settle, and the liquid then poured on to the surface of a small quantity of a solution made by dissolving 0.75 grm. cupric chloride, 1.5 grms. ammonium chloride, 2 C.C. strong ammonia, and 3 grms. hydroxylamine hydrochloride, in 50 C.C. of water. Solids are treated with boiled-out calcium carbide and an anhydrous solvent, the test being most sensitive when the solvent is not miscible with water, as the pre- cipitate is then concentrated at the zone of contact of the solvent with the cuprous solution. Calcium carbide settles well, and distillation is never really necessary, but may be resorted to in doubtful cases. If carbon disulphide is employed, the copper solution must not contain hydroxylamine or other reducing agent, but it need72 ABSTRACTS OF CHEMICAL PAPERS not be colourless, the reaction showing up well against solutions containing enough cupric salt to give a, pronounced blue colour. Only strong acids interfere. The sensitiveness of the reaction lies between 1 in 10,000 and 1 in 3,000, according to the solvent employed. G. C. J.

ISSN:0003-2654    

DOI:10.1039/AN9154000067

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

72 ABSTRACTS OF CHEMICAL PAPERS APPARATUS, ETC. Modified Preeision Barometer. A. F. 0. Germann. (J. Amer. Chem. SOC., 1914,36,2456- 2462.)-In the determination of the densities of gases by the most refined methods, the values obtained for the pressure are always less accurate than the other data, which are susceptible of measurement with very great precision. Chief among the errors to which barometers are sub- ject is the residual pressure in the ‘‘ vacuum.” Errors due to variations in the meniscus are reduced by the use of wide-bore tube, but the author describes curious anomalies in some in- L’ struments of this type which he has traced to a change in the glass whilst heating to boil air out of the mercury.The instrument now de- scribed is free from these objections. Essentially it is a modification of the mercury pump of Carduso and Germann (J.Chim. Phys., 1912, 10, 306), the main reservoir being omitted. The tube L, is 110 cm. x 15 mm. ; L, is 20 cm. x 15 mm., B is a capillary of 0.2 mm. bore, and must not be shorter than the maximum local barometric height, but should not greatly exceed this. C, and C, are lubricated with Acheson graphite, and, as this has no adhesive power, are held in place by brass clamps.Into the clean, dry barometer freshly distilled mercury is intro- duced by opening Cl and C4, closing C2 and C3, and raising M or applying suction through the drying tubes by means of the water-pump. Mercury is allowed to fill L, and run over into R. Then M is lowered, the mercury in L, subsides to a point at which it is in equilibrium with the air enclosed in A,, and the level of the mercury in the latter may be adjusted to the coloured-glass point P by opening Cs and adjusting the height of N.The air admitted with the mercury collects at E, and may be expelled through Cz. If the length of B has been chosen as described, any great excess of mercury in R will be forced back into V by the pressure of the atmosphere, but some will always remain to maintain the vacuum.After several days air will have disengaged itself from the glass walls of V, as may be seen by raising the bulb Muntil the mercury rises in Yand To WATER PUMP TO APPAWTUSAPPARATUS, ETC. 73 approaches the mercury thread in the capillary. The test is extremely delicate as the ~- air in question is under a very small pressure. factorily by turning on the water-pump, any attempt to drive it over being likely to cause it to adhere to the capillary walls of B.I n practice, however, expulsion is unnecessary, as the volume of air is negligibly small, cor- responding to a pressure of the order of 0*00001 mm. In use the gaseous pressure to be measured is admitted through C,.The paper includes suggestions for facilita- ting corrections for capillarity, thermal ex- pansion of mercury and glass scales, and for altitude and latitude. G. C. J. Freezing-Point Apparatus. M. C. Dekhuizen. (Chem. Weekblad, 1914, 11, 126-131.)-The modification of Beckmann’s apparatus here shown is useful for very accurate cryoscopic determinations. The jar containing the freezing mixture is covered with an ebonite lid through which pass the two tubes S and L.The former, containing a little mercury, is intended for rapidly cooling the observation tube, while the latter is a Dewar-vacuum tube. To prevent access of the warmer air of the room, the top of the Beckmann tube is surrounded by an air-chamber, LK, in which is an ice-box, EK. The air-chamber is constructed in two parts, which are clamped together and form a handle for transferring the Beckmann tube from S to L. For determining the freezing-point of milk, a suitable freezing mixture (- 2.5’ C.) is obtained by adding I t s expulsion can only be effected satis- 1 C r 1.8 kilos of large lumps of ice and 0.5 kilo f of finely crushed ice to 3 litres of an 8 per cent. solution of salt. C. A. M. Chemical Reactions in Colloidal Media. H. C. Jacobsen. (Chem. Weekblad, 1914, 11, 588.)-A summary of the observations of various experimenters on precipitations in colloids 0. E. M.

ISSN:0003-2654    

DOI:10.1039/AN9154000072

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

74 REVIEW REVIEW. THE ELEMENTS OF PHYSICAL CHEMISTRY, By J. LIVINGSTON R. MORGAN. Fifth New York : John Wiley and Sons ; London : Chapman and Hall, IF it were possible to take high price as a guarantee of excellence, the volume under review could be at once welcomed as a first-class elementary textbook of physical chemistry. This, however, one cannot do, for Professor Morgan’s work is open to many serious criticisms.There is a notable failure to introduce the student to the experimental aspects of physical chemistry. In the discussion of the specific gravity of gases, for example, there is no reference to Regnault’s and Rayleigh’s method of determining this con- stant, or to Victor Meyer’s method of finding vapour density. Again, the author talks of the index of refraction of a liquid, and of Gladstone and Dale’s formula for specific refraction, but says nothing about the principle and application of the refractometer-an instrument so largely used both by the physical chemist and by the analyst.Indeed, only one page is devoted to the subject of refraction, whilst eighteen pages are occupied with a discussion of surface tension. Professor Morgan makes a great point of defining eaah concept in terms of .experiment; this is done, for instance, with molecular weight.Farther on he points out that Avogadro’s “Law,” as stated by him, is simply expressive of the definition of molecular weight already given, and maintains that the hypothetical character of this relation has been lost. This is surely a travesty of Avogadro’s contribution to the foundations of modern chemistry.Further criticisms might be offered, but quotation of the two following passages will illustrate the need there is for drastic correction and revision. On p. 203 it is stated that ‘‘ the solution for which AZ is found also gives the maximum osmotic pressure.” On p. 197 we read : (‘It will be seen that any liquid can be used as .the solvent in the freezing-point method so long as it is not a solution, judged by the formation of a solid phase-Lo., so long as the solid separating from the pure solvent has the same composition as the liquid it leaves behind.” One pities the student who has to wrestle with these and similar passages.Edition. Ltd. 1914. Pp. xiv + 506. Price 12s. 6d. net. JAMES C. PHILIP.* * * + * INSTITUTE OF CHEMISTRY. PASS LIST : JANUARY (1915) EXAMINATIONS. OF four candidates who presented themselves for the Intermediate Examination, two passed : G. T. Bray and E. G. Macintyre, B.Sc. (Glas.). Of thirteen candidates who presented themselves for the Final (A.I.C.) Examination, five passed; in the Branch of Mineral Chemistry : R. L. Amoore and E. G. G. Wheeler ; in the Branch of Organic Chemistry: W. J. S. Naunton, M.A. (Cantab.), B.Sc. (Lond.), and H. Shulman, B.Sc. (Lond.); in the Branch of the Chemistry (and Microscopy) of Food and Drugs, Fertilisers and Feeding Staffs, Soils, and Water : W. S. Clark,

ISSN:0003-2654    

DOI:10.1039/AN9154000074

出版商:RSC    

年代:1915

数据来源: RSC