140 ABSTRACTS O F CHEMICAL PAPERS ORGANIC ANALYSIS. Simultaneous Estimation of Carbon and Halogen by the Chromic Acid Method. P. W. Robertson. ( J . Chenz. SOC., 1916, 109, 215-221.) - The satisfactory results obtained with the chromic acid method in the estimation of chlorine and bromine in organic compounds (ANALYST, 1915, 40, 413) have led the author to extend the process to the simultaneous estimation of carbon, and his method of procedure may be applied with advantage to the estimation of carbon alone, particularly in compounds which are combustible with difficulty in the dry way.The apparatus, when used for the estimation of carbon only, may be adapted for either gravimetric or volumetric measurement of the carbon dioxide. I n either case, in order to complete the combustion of traces of carbon monoxide, a short silica tube containing platinised asbestos is attached to the delivery tube from the reaction flask, and between this and the absorption apparatus is interposed a small U-tube containing glass-wool well moistened with sulphuric acid containing chromic acid, to retain the oxides of sulphur.The substance to be analysed is weighed from a small tube into the reaction flask, and 3 grms.of chromic acid added. For the gravimetric procedure the absorption apparatus ccmsists of a calcium chloride tube and two soda-lime tubes, the latter being previously weighed full of oxygen. A slow current of oxygen, free from carbon dioxide, is passed through the whole apparatus, and about 30 C.C. of redistilled sulphuric acid added to the reaction flask.The re- action is controlled by cooling if necessary, and the combustion completed in seventy to ninety minutes. The results are checked by a blank experiment and corrected for the amount of carbon dioxide produced by the reagents alone. The numbers compare favourably with those obtained by dry combustion; even in the case of trinitrophenol no error is introduced by nitrous fumes.By the volumetric procedure the method may be applied to much smaller quantities of material. I n this case the absorption apparatus is replaced by an evacuated flask of 1,500 C.C. capacity containing a measured quantity of standard baryta solution. A pressure regulator is attached to the oxygen inlet tube, consisting of a T-piece with its end dlpping in water, and a gentle stream of bubbles is allowed to escape.When the sulphuric acid has been added, the tap of the evacuated flask is slightly opened, and the combustionORGANIC ANALYSIS 141 allowed to proceed for about two hours, The baryta solution is slightly agitated at intervals, and finally titrated with & acid. For the simultaneous estimation of carbon and bromine, two absorption tubes are used, the first cmtaining 10 C.C.of 1.3 N-sodium hydroxide and 2 c c. of 2 N-sodium sulphite, and the second tube 10 C.C. of sodium hydroxide, part of which is retained by glass beads in the exit tube. Oxygen is passed through at the rate of a group of bubbles every two or three seconds, and the reaction is assisted by heat. At the end, the contents of the absorption tubes are washed into a flask, boiled, treated with 5 C.C.of hot saturated barium nitrate solution, and filtered. The filtrate is titrated with $ nitric acid in presence of phenolphthalein, the result being corrected by a blank determination. The solution is further acidified with nitric acid, 10 C.C. of FG silver solution are added, and the excess is titrated with & thiocyanate.The silver solution is previously stmdardised by pure potassium bromide treated with the reagents in the. same apparatus, errors being thus compensated. The value for the bromine, reduced t o :, is subtracted from the total alkalimetric result to obtain the value for carbon dioxide. The estimation of carbon and chlorine is carried out in a similar manner, except that between the small U-tube containing chromic and sulphuric acids and the first absorption tube there must be inserted a second short length of silica tubing, which is heated to decompose any chrornyl chloride carried over, chromic oxide being deposited.This is necessary because chromyl chloride is capable of neutralising twice as much alkali as the corresponding amount of chlorine.The results, as in the case of bromine, are sufficientiy satisfactory and concordant. Attempts to estimate nitrogen in the residual acid in the reaction flask, on the principle of the Kjeldahl method, did not yield satisfactory results. J. F. B. Direct Estimation of Carbon Monoxide in Mixtures containing Un- saturated Hydroearbons. A. Piva. (Amali Chi?iz. Applic., 1916, 5, 82-93.)- The gaseous mixture is repeatedly passed through a tube containing soda-lime in which the proportion of sodium hydroxide is about 20 per cent., heated in an air- bath to 230" C.Under these conditions carbon monoxide is absorbed and retained by the soda-lime. The tube should be connected with pipette tubes charged with alkaline pyrogallol and sodium hydrosulphite, to remove carbon dioxide and oxygen before absorption of the carbon monoxide. As a typical example, an analysis of the illuminating gas of Palermo is quoted. By Henipel's method this contained -Carbon dioxide, 1.6 ; unsaturated hydrocarbons, 2.9 ; oxygen, 0.6 ; hydrogen, 47.3 ; methane, 33.2 ; nitrogen, 6-5 ; and carbon monoxide, 7.9 per cent.By direct absorption with soda-lime at 230' C.the amount of carbon monoxide found was 7.8 per cent. C. A. M. Detection of Glycerides by the Maganta - Sulphurous Acid Reagent. M. Francois. ( J . Pharm. Chinz., 1916, 13, 65-77.)--S test for glycerol described recently by Francois and Boismenu (ANALYST, 1915, 40, 238) is modified so that it may be used for the detection of glycerides in the presence of paraffins, waxes, resins, etc. About 1 grm.of the substance is mixed in a test-tube with 10 grms. of sand, and then heated over a flame, the vapours produced being conducted142 ABSTRACTS O F CHEMICAL PAPERS into another test-tube containing 3 C.C. of Schiffs reagent (sulphurous acid, 220 C.C. ; 0.1 per cent. magenta solution, 30 C.C. ; and concentrated sulphuric acid, 3 c.c.). White fumes produced at the commencement of the heating should be prevented from passing into the second test-tube by removing the flame for a few moments at a time.The heating should be continued (nearly to redness) for about ten minutes ; the test-tube containing the reagent is then beated in a boiling water- bath for fifteen minutes. If the substance under examination contained a glyceride, the reagent is coloured red while cold, and the colour changes to blue on heating.All glycerides give this reaction ; paraffins, fatty acids, waxes, resins, spermaceti, tolu balsam, and caoutchouc, yield vapours which give a red coloration with the reagent while cold, but the colour disappears on heating. The mechanism of the decoloration of magenta by sulphurous acid and the formation of the red and blue colorations is discussed.w. P. s. Solubility of' Naphthalene in Ammonia. S. Hilpert. (Zeitsch. angem C?wnm., 1916, 29, 57-59.)-The solubility of naphthalene in ammonia solution and in liquid ammonia is given in the following table, the figures expressing grm. per 1,000 grms. of liquid : ~ _.____________ - I Per Cent. NH,, : 0. I 5. 25. 100. __ - __ .- 1 - _ _ -~ ~ I At 0" C. ~ 0.018 1 0.030 0.042 0.064 I 33.0 At 25°C. 1 0030 1 0,044 1 0074 1 0.162 ~ 120.0 I 1 The presence of 2 per cent. of pyridine in the liquid ammonia increases the solu- bility of the naphthalene by 0.082 grm. at 0" C., and by 0,245 grm. a t 25' C.; the solubility is not influenced by phenol. If carbon dioxide is passed into 25 per cent. ammonia containing dissolved naphthalene, the greater part of then latter is pre- cipitated.When ammonia containing naphthalene is distilled, the condenser tubes are liable to become blocked with naphtha,lene unless they are maintained at a temperature not lower than 30" C. The picrate method may be used for the estimation of naphthalene in ammonia. A definite quantity of the sample is cooled in ice-water, neutralised by the gradual addition of sulphuric acid, care being taken to keep the mixture quite cold, and then distilled until about 30 C.C.of distillate have been collected. The distillate is extracted with ether, the condenser also being rinsed out with this solvent, and the separated ethereal solution is then treated with an excess of gG picric acid solution. After the lapse of two minutes the ether is evaporated under reduced pressure, the residual aqueous solution is cooled in ice- water, the precipitate collected on a filter, washed with 5 C.C.of ice-water, and titrated with FG sodium hydroxide solution, using litmus as indicator (ste also ANALYST, 1915, 40, 64). w. P. s. Estimation of Prussian Blue in Spent Oxide. E. L. Randall. ( J .Gas Lighting, 1916, 1.33, 158; through J. Xoc. Chem. Ind., 1916, 35, 42.)-The process is based on the fact that when titanous chloride is gradually added to a dilute solutionOltGANIC ANALYSIS 143 of an alkali ferricyanide a light green coloration is produced which deepens to an intense green and then slowly becomes reddish-brown, the reaction being quantita- tive in the presence of a large excess of ammonium or alkali thiocyanate.Not less than 0.1 grm. of the ferricyanide should be present, a satisfactory degree of dilution being about 0.6 grm. per litre for the potassium and about 2 grnis. per litre for the sodium salt. The titanous chloride (I per cent. solution) is standardised against ferrous ammonium sulphate or potassium ferricyanide. Results within 003 per cent.of the theoretical were obtained in estimations made on solutions of the latter salt. Alkali ferrocyanides alone and mixed with ferricyanides were also satisfactorily estimated, the former being oxidised by a dilute solution of potassium permanga- nate the slight excess of which was removed by the addition of thiocyanate before titrwtion. For the estimation of Prussian blue in spent oxide the following procedure is suggested : After the removal of free sulphur by extraction with carbon bisulphide the material is treated with sodium hydroxide and a little ferrous sulphate (to convert any cyanide into ferrocyanide), the mixture filtered, and the filtrate agitated with lead carbonate to remove sulphides.The solution is then neutralised and diluted to a definite volume, of which an aliquot portion is treated with perman- ganste and titrated after the addition of sodium thiocyanate, the potassium or ammonium salt not being permissible.Estimation of Mixtures of Paracetaldehyde and Acetal. K. J. P. Orton and P. V. MeKie. ( J . Chem. SOC., 1916, 109, 184-186.)-The method for the separate estimation of paracetaldehyde and acetal is based on the different rates of decomposition of these substances, with formation of acetaldehyde, on heating with dilute aqueous acids.The acetaldehyde is distilled off and estimated by one of the etandard methods. Acetal is stable in alkaline or neutral solutions, but is completely hydrolysed in a few minutes on boiling in 1 per cent. solution with a trace of a strong acid at a concentration of B$& or with a weaker acid at equivalent hydrion concen- tration.Paracetaldehyde shows no signs of decomposition until the hydrion con- centration is increased to && and 2x1 rapid action is only realised when the concentration of the strong acid amounts to & or +YZ. I n the estimation of acetal, the inconvenience of measuring accurately the minute quantity of strong acid necessary for the separation is best avoided by the use of acetic acid.Seven C.C. of acetic acid are diluted to 30 C.C. in a small flask of the Kjeldahl pattern, and 0.2 to 0.4 grm. of the mixture to be analysed is added. The flask is attached to a con- denser connected by means of an adapter with a receiver (e.g., a, small measuring cylinder) containing 10 C.C.of water or alcohol for the better absorption of the acetaldehyde. The receiver is well cooled and the liquid gently heated until 7 C.C. have been distilled in the estimation of acetal. For the estimation of paracetaldehyde (or of the total acetal and paracetaldehyde in a mixture), 30 C.C. of hydrochloric acid are used and 10 C.C. of distillate collected.The acetaldehyde in the distillate may be determined either by the hydrogen sulphite method (Ripper’s modification), or by the neutral sulphite method of Seyewetz; both give concordant results. If the mixture contain acetaldehyde in addition to acetal and paracetaldehyde, it may be estimated by a separate distillation. J. F. B.144 ABSTRACTS OF CHEMICAL PAPERS New Reaction of Picric Acid, and its Applications.J. Castets. ( J . Pharm. C h i m , 1916, 13, 46-49.)-When picric acid is boiled with bromine solution 2-bromo-4 : 6-dinitrophenol is formed ; the test proposed depends on the formation of this substance and the red coloration which it gives with ammonia vapour or potas- sium cyanide. Ten C.C. of the picric acid solution ere treated with 10 drops of saturated bromine-water, the mixture is heated to boiling, cooled, and extracted with ether.The ethereal solution is separated and divided into two portions, one of which is evaporated in a porcelain basin and the residue treated with ammonia vapour ; a red coloration is obtained which is discharged by an excess of ammonia or by the addition of water. The other portion of the ethereal solution is evaporated, drop by drop, on a piece of filter-paper ; a red coloration is obtained if the paper is then treated with ammonia vapour or with a drop of potassium cyanide solution, and again dried.For the estimation of picric acid in urine, 100 C.C. of the sample are mixed with 2 C.C. of hydrochloric acid and 20 C.C. of chloroform, heated on a water bath, cooled, the chloroform layer is separated, and evaporated.The residue thus obtained is dissolved in water, treated with bromine, and the test proceeded with as described. The test will detect the presence of 5 mgrms. of picric acid per litre of urine ; it must be remembered that a positive reaction for picric acid in urine can be obtained only when a quantity of the acid has been taken sufficient to insure that a portion of it shall pass into the urine before it has been converted into picramic acid during its passage through the body.w. P. s. Estimation of Toluene: Application of the Method to Benzene and Xylene. H. W. James. ( J . Xoc. Chem. hid., 1916, 35, 236-240.) Naphthas.-The sample is distilled with the aid of a fractionating column at the rate of one drop per secoad, SO as to obtain two fractions-I.and 11.-consisting respectively of substantially a mixture of toluene and benzene, and of a mixture of benzene, toluene, and xylenes. A . 250 C.C. are distilled up to 140" C. If at least 95 per cent, distils below 140" C., this distillation may be omitted. B. The distillate from A (or the original sample) is redistilled, and the following fractions collected: (1) Below 90" C.; (2) from 90" to 100" C.; (3) from 100" to 109" C.; (4) 109" to 112" C. ; (5) 112" to 120" C. ; (6) residue at 120" C. C. The fractions 2, 3, 4, and 5, are separately redistilled, all fractions distilling below 110*6" C. being collected in receiver No. 1, and all residues left at 1 1 0 - 6 O c. being added to that in receiver No. 6. Should the amount of liquid in any of the flasks become too small for distillation before 110.6" C.is reached, the distillation is stopped, the flask cooled, the contents of the next fraction added to the residue, and the distillation continued. I n the case of crude unwashed naphtha, the sample is distilled to 150" C. in stage A. The distillate is washed with sulphuric acid in the usual way, and distilled as in 13, the following fractions being collected : (1) Below 90" C.; (2) 90° to 100" C. ; (3) 100" to 109" C.; (4) 109" to 112" C. ; (5) 112" to 120" C. ; (6) 120" to 140" C. In stage C all the fractions are separately distilled to 110.6" C. into No. 1, and all residues put together.Toluen per Cen .___ 1 2 3 4 5 6 7 8 9 10 11 12 1 3 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 I 49 5 0 1 1) 90" c ORGANIC ANALYSIS TABLE I.Mixtures of Toluene and Be.lzxene. 95" c 100" ( 108' C - I - - I - Toluen per Cen 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 100 90" ( 95" c. 145 108" C.146 Toluenf per Cenl 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 115"( ABSTRACTS OF CHEMICAL PAPERS TABLE 11.Mixtures of Toluene and Xylenes. 130" C 137" ( Toluene per Ceni 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 115" ( 120" c - - - - 17-0 20.0 23.5 26.5 29.5 32.7 35.7 38.7 41-5 44.0 46.7 49.5 52.1 54.9 57.2 59% 62.0 64.5 67.0 694 71.6 73.7 75.7 77.5 79.2 80.9 - - - - __ - - - - - - - - - - - - - - - 125" ( 130" CORGANIC ANALYSIS 147 The three fractions-viz., No.I., distillate below 110.6" C, ; No. II., from 110*6° to 140" C. ; No.III., residue at 140" C.-are measured, and No. 111. is dis- carded. The others are distilled as follows-No. I . : The fraction, or 100 C.C. thereof, is distilled in a flask with a side-tubulure at the rate of 2 drops per second, and measured fractions are collected at 85" C., 90" C., 95" C., 100" C., and 108' C. Reference to Table I. will then give the percentage of toluene. The distillation is continued until the total percentage of distillate falls within the range of the corresponding column in the table.No. 11. fraction is similarly distilled, fractions being collected at 115" C., 120" C., 125" C., 130" C., and 137" C., and the total per- centage distilling at each temperature noted as before. Reference is then made to Table 11. I n the case of naphthas containing less than 30 per cent.of toluene the error should not reach 1 per cent. When only a very small amount of toluene is present it is advisable to add a known quantity of toluene, and to make the necessary correction in the results. A correction is made for parafinoid hydrocarbons, when present. The specific gravity of fraction 11. should lie between 0,865 and 0.870 at 15.5" C., and it may be assumed that the pure " toluene separated from a naphtha would contain 1 per cent. of paraffinoid hydrocarbons for every 09002 by which the sp. gr. falls short of 0.867. CoaZ Tar.-The crude naphtha is separated by fractional distillation, washed with sulphuric acid, and 250 C.C. distilled as described above. The fractions separated from the tar are (a) light oii and water up to 220" C.; (b) creosote oil 220' to 260" C. The distillate obtained from the light oil (up to 190" C.) is mixed with the similar distillate from b, and the mixture shaken. Any water separating is withdrawn and measured. The residue constitutes the crude naphtha of the tar. I t is purified by shaking with 10 per cent. of its volume of strong sulphuric acid, the flask being kept cool in water. After thirty minutes the brown "acid tar " is drawn off and the residue washed, without agitation, with successive portions of 50 C.C. of water, until the washings cease to be opalescent. The naphtha is then gently shaken with more portions of water, then with 10 C.C. of caustic soda solution (sp. gr. 1*075), and finally twice more with 50 C.C. of water. C. A. M.