摘要:

JUNE, 1912. Vol. XXXVII., NO. 436, THE SEPARATION OF ARSENIC FROM ANTIMONY AND OTHER METALS, WITH SOME APPLICATIONS TO TOXICOLOGICAL WORK.* BY STANLEY W, COLLINS, B.Sc., F.I.C. (Read at the Meeting, April 3, 1912.) FRIEDHEIM AND MICHAELIS (Ber., 1895,28,1414) have described a method for quantita- tively separating arsenic from vanadium, molybdenum, and tungsten, by distillation with methyl alcohol in a current of hydrogen chloride.Under these conditions it is stated that arsenic trioxide forms a volatile methyl ester. In order to estimate the arsenic the distillate was treated with nitric acid and evaporated to dryness. The residue was. dissolved in water, the solution made alkaline with ammonia, and the * This work was carried out under the provisions of the Analytical Investigation Scheme.230 COLLINS: SEPARATION OF ARSENIC FROM ANTIMONY AND OTHER arsenic precipitated by magnesia mixture and weighed as magnesium pyroarsenate .In estimating arsenic pentoxide no reducing agent need be added, as it is reduced i n L presence of methyl alcohol and hydrogen chloride, with the formation of the volatile methyl arsenious ester. separate arsenic from antimony by passing a current of dry air at the ordinary temperature through a solution of arsenic trioxide in hydrochloric acid to which methyl alcohol had been added.They state that under these conditions the arsenic can be completely distilled, leaving the antimony in the residue. I t appeared desirable to determine the possible application of these methods to the separation of arsenic from antimony, particularly when these elements are present with organic matter. The following form of apparatus (see illustration, p.231) was devised and found to be satisfactory for the experiments involving distillations : A pear-shaped distilling flask of about 300 C.C. capacity, with a wide neck about 4 crns. diameter and 8 crns. long, drawn down at the top for a small cork or ground joint through which passes the tube for delivery of hydrogen chloride.The flask is provided with a wide side-tube, A , about 1 cm. internal diameter, at about 4 crns. from the lower end of the neck. This tube is bent upwards at an angle of about 60 degrees with the neck, so that during distillation it is inclined upwards from the flask, which makes an angle of about 30 degrees with the vertical.This arrange- ment allows of rapid distillation, whilst it prevents the solution being carried over by spurting. The inlet tube B, of about 1 cm. diameter and 2 crns. length at the lower end of the neck, serves for the introduction of substances to be analysed, and into it, before the distillation is commenced, a tap-funnel is fitted by means of a cork or a ground joint, to allow of the addition of more methyl alcohol.In a later form of apparatus the inlet tube B was enlarged to about 1& crns. diameter, thus making it more convenient for the introduction of substances to be analysed. The tap funnel was also ground in. The side-tube was ground into a Liebig’s condenser, and bent so that, during an experiment, the condenser is in a position slightly sloping away from the distilling flask.The inside tube of the condenser is about 1 cm. in diameter and 40 crns. in length, passing through a cork into a 30-ounce conical flask. Through another hole in the cork a tube filled with glass beads is fitted. The jacket of the condenser is about 17 crns. long. Several experiments were carried out on the lines proposed by Cantoni and Chautems (Zoc. cit.), but the method proved to be of no value ; for under the conditions described by these chemists only a small proportion of arsenic trioxide could be dis- tilled, even when a current of dry air was passed for four hours.A more successful experiment was carried out by the method of Friedheim and Michaelis (Zoc. cit.). 0.2 grm.of arsenic trioxide was washed into the distillation flask with 50 C.C. of methyl alcohol, which was then saturated with dry hydrogen chloride. The flask was then heated 011 a water-bath; the distillation was repeated twice with 50 C.C. of methyl alcohol, a slow stream of hydrogen chloride being passed during the distillation. On diluting the residue in the flask and passing in sulphuretted hydrogen, no arsenious sulphide was precipitated. The method used in this and many subsequent experi- Cantoni and Chautems (Annul. de Chim.anal., 1905, 10, 213-214) proposed to 1METALS, WITE SOME APPLICATIONS TO TOXICOLOGICAL WORK 231 ments for the estimation of arsenic in the distillate, consisted of saturating with sul- phuretted hydrogen, after diluting with an equal volume of water, washing the precipi- tated arsenious sulphide with hot water, alcohol, a mixture of carbon disulphide and alcohol, alcohol, and then hot water.The precipitate was then washed from the filter-paper [into a weighed dish, dried over a water-bath and then at 105' to l l O o C., and weighed. The precipitate was tested for antimony by digesting with strong hydrochloric acid, filtering, diluting and saturating with sulphuretted hydrogen. The antimony sulphide was dissolved in hydrochloric acid with a little potassium chlorate, and antimony obtained from the solution by means of pure iron wire.Arsenic in the distillate was also estimated satisfactorily by titration with standard iodine solution. An experiment was made on 0.2 grm. of arsenic trioxide232 COLLINS: SEPARATION OF ARSENIC FT~OM ANTIMONY AND OTHER with 0.2 grm.of antimony trioxide, with the result that about 0.02 grm. of antimony trioxide was found with the distilled arsenious oxide. I t will thus be seen that the formation of antimony chloride and its subsequent volatility render the method inap- plicable to the quantitative separation of arsenic from antimony. Many experiments were made in order to overcome this difficulty, in the course of which it was found that the volatility of arsenic trioxide in the vapour of methyl alcohol was considerably increased by the addition of strong sulphuric acid or phosphorus pentoxide to the arsenic trioxide and methyl alcohol before distillation ; neither arsenic pentoxide nor antimony trioxide could be distilled under these conditions.Attempts were made to devise a method based on these facts for the quantitative separation of arsenic from antimony, but it was found that the removal of the last traces of arsenic was a tedious process, in which it was necessary to use a, comparatively large volume of methyl alcohol. A complete separation was, however, effected by taking a mixture of methyl alcohol and the oxides and a sufficient quantity of sulphuric acid, and passing a stream of hydrogen chloride through for some time before commencing the heating.The passage of the gas was continued throughout the distillation. Under these conditions it was found that the arsenic was completely distilled, whilst the more basic antimony was retained by the sulphuric acid.When phosphorus pentoxide was substituted for sulphuric acid in these experiments, small quantities of antimony were carried over with the arsenic: The following account of typical experiments illustrates the application of this method : 0.08 grm. of arsenic trioxide and 0.08 grm. of antimony trioxide were washed into the distilling flask with 30 C.C.of methyl alcohol and a mixture of 6 C.C. of sulphuric acid (sp. gr. 1-84) with 20 C.C. of methyl alcohol. Hydrogen chloride was passed in for ten minutes and the flask heated on a water-bath, the distil- late being collected in a 30-ounce conical flask into which a small quantity of water was occasionally poured over the glass beads. The distillation was repeated twice with 50 C.C.of methyl alcohol. The arsenic in the distillate was precipitated by sulphuretted hydrogen, washed, and weighed; 0.0796 grm. was found, No antimony sulphide could be detected in the precipitate, This analysis was repeated, using 0.2 grm. of arsenious oxide and 0.2 grm. of antimony oxide with 8 C.C. of sul- phuric acid (sp. gr. 1.84). I n the next experiment the conditions were the same, with the exception that the arsenic in the distillate was titrated with iodine, after adding an excess of sodium bicarbonate. 0.200 grm.of arsenic trioxide was found. 2.0 grms. of antimony oxide with 0.1 grm. of arsenic trioxide were mixed with 15 C.C. of sulphuric acid (sp. gr. 1.84) and 50 C.C. of methyl alcohol. The distillation was repeated with the following volumes of methyl alcohol, 50 c.c., 50 c.c., 20 c.c., 20 c.c., when, on testing the distillate with sulphuretted hydrogen, no precipitate was obtained. The precipitate was free from antimony.3.0 grms. of antimony oxide with 0.1 grm. of arsenic trioxide were mixed with 15 C.C. of sulphuric acid (sp. gr. 1.84) and 50 C.C. methyl alcohol. Distillation was repeated as in the last experiment. In this 0-200 grm.of arsenic trioxide was found. 0.1002 grm. of arsenic trioxide was found. 0*101 grm. of arsenia trioxide was found.METALS, WITH SOME APPLICATIONS TO TOXICOLOGICAL WORK 233 case the precipitate contained antimony. The experiment was repeated, using 3.0 grms. of antimony oxide with 0.06 grm. of arsenic trioxide and a smaller quantity of sulphuric acid-ie., 10 C.C.A precipitate equivalent to 0.064 grm. of arsenic frioxide was found. Antimony was present in the precipitate. The application of this method to the separation of arsenic pentoxide from antimony was next investigated. It was found that the arsenic from arsenic pentoxide could be distilled completely only in the presence of a reducing agent such as ferrous chloride. The following experiments show the results obtained for the estimation of arsenic pentoxide; the first one gave a satisfactory result €or the separation of arsenic from other metals of the sulphuretted hydrogen group as well as from antimony : 0.06 grm.of arsenic pentoxide with 0.05 grm. of antimony oxide, 0.04 grm. of bismuth oxide, 0.03 grm. of mercurous chloride, 0.03 grm. of mercuric chloride, and 5 grms.of ferrous chloride, were introduced into the distilling flask with 50 C.C. of methyl alcohol. To these were added 10 C.C. of sulphuric acid (sp. gr. 1.84) ; hydrogen chloride was passed through the contents of the flask before and during the distillation, which was repeated twice with 50 C.C. and twice with 20 C.C. of methyl alcohol. The distillate was saturated with sulphuretted hydrogen and the precipitated arsenious sulphide, washed, dried, and weighed, as before.The weight of arsenic pentoxide found was 0.0596 grm. The precipitated arsenious sulphide was found to be free from antimony, bismuth, and mercury. 0.1 grm. of arsenic pentoxide with 2 grms. of antimony oxide was mixed in the distillation flask with 15 C.C. of sulphuric acid (sp.gr. 1-84), 50 C.C. of methyl alcohol and 5 grms. of ferrous chloride. Hydrogen chloride was passed before and during distillation, which was repeated twice with 50 C.C. and twice with 20 C.C. of methyl alcohol. The weight of arsenic pentoxide found was 0.0998 grm., and there was no antimony in the precipitate. These results indicate that this method is applicable, under suitable conditions, to the quantitative separation of arsenic from antimony and from other metals.Before pointing out the conditions necessary for this method, it is essential to give an account of an experiment which shows the influence of water on the volatilisation of antimony trichloride: 1 grm. of antimony oxide was mixed with 10 C.C. of sulphuric acid (sp. gr. 1.84) and 50 C.C.of water. Hydrogen chloride was passed in for about fifteen minutes as well as during the distillation. Sulphuretted hydrogen was passed into the distillate, which yielded an appreciable precipitate of antimony sulphide. To about 20 C.C. of the residual solution an equal volume of methyl alcohol was added, and hydrogen chloride passed through as before. Separate small volumes of the distillate were collected and saturated with sulphuretted hydrogen.A barely perceptible coloration was obtained in the first portions, but, as the alcohol was distilled off, the quantity of antimony sulphide precipitated became appreciable. It is evident, therefore, that the material used in this method should be sufficiently free from water, since, in presence of water, sulphuric acid fails to retain the antimony completely. With regard to the amount of sulphuric acid necessary to retain the antimony, 10 C.C.of acid (sp. gr. 1.84) is a convenient quantity for not more than 1 grm. of antimony oxide; whilst 15 C.C. of acid234 COLLINS: SEPARATION OF ARSENIC FROM ANTIMONY AND OTHER will retain up to 2 grms. of antimony oxide, and is the maximum quantity convenient to use with 50 C.C.of methyl alcohol, since with more than this amount the distillation becomes too slow. An experiment was next made in order to gain some idea of the maximum amount of arsenic trioxide which could be volatilised, using 50 C.C. of methyl alcohol. Arsenic trioxide 5 grms., with 10 C.C. of sulphuric acid, and 50 C.C. of methyl alcohol, were distilled in a stream of hydrogen chloride in the usual manner, and 0.96 grm.of arsenic trioxide was found to have been volatilised, an amount largely in excess of that used in any estimation. In the quantitative separation of arsenic from antimony, when not more than 0.2 grm. of arsenic trioxide was used, about 60 per cent. of the arsenic was volatilised in the first 50 C.C. of alcohol; a further 50 C.C.generally carried over all but a, small amount, which was volatilised in the last 50 C.C. I n presence of a large excess of antimony, it is advisable to add a further 10 C.C. of methyl alcohol and to test the distillate with sulphuretted hydrogen, as an additional 10 or 20 C.C. of methyl alcohol may be necessary to remove the last traces of arsenic. Pure methyl alcohol, not specially dehydrated, was used in all these experiments.Commercial methyl alcohol is unsatisfactory, as the acids employed induce the formation of compounds of high boiling-point which retard the volatilisation of the arsenic. In an experiment in which ethyl alcohol was used, instead of methyl alcohol, the quantity of arsenio carried over was about half as great as would have been volatilised by the same volume of methyl alcohol.I t has been stated in this paper that the quantity of arsenic trioxide which could be distilled over was increased by mixing sulphuric acid with the alcohol and the oxide in the distillation flask, and that neither arsenic pentoxide nor antimony trioxide could be distilled in this manner. This suggests an easy test for arsenic trioxide in the presence of antimony as well as other metals.This method could only be applied as a test for arsenic trioxide in the presence of arsenic pentoxide when certain organic and other reducing agents were excluded. Equal weights of the following substances were mixed-mercurous chloride, mercuric oxide, stannous chloride, cadmium oxide, antimony oxide.About 1 grm. of this mixture was placed in a small distilling flask, and 20 C.C. of methyl alcohol and 3 C.C. of sulphuric acid were added. The flask was attached to a condenser and heated by a flame, the distillate being collected in a small flask containing about 5 C.C. of water. On passing sulphuretted hydrogen through the distillate no precipitate was obtained. This experiment was repeated, using 1 grm.of the mixed substances with 0.3 grm. of arsenic pentoxide. No precipitate was obtained on passing sulphuretted hydrogen through the distillate. The experiment was again repeated, using 1 grm. of the mixed substances with 2 mgrms, of arsenic trioxide. A precipitate of arsenious sulphide was obtained on passing sulphuretted hydrogen through the distillate.The following experiments were made to test this method.METALS, WITH SOME APPLICATIONS TO TOXICOLOGICAL WORK 235 APPLICATION OF THIS METHOD TO THE SEPARATION OF ARSENIC FROM ANTIMONY AND OTHER METALS IN THE PRESENCE OF ORUANIC MATTER. The character of the material for analysis will determine the mode of application. The separation can be effected without previous destruction of the organic matter if the arsenic be not intimately combined with the organic matter.The material should also be fairly dry. The following experiment is an example of these conditions : 20 grms. of powdered biscuit material were placed in the distillation flask; 0-1 grm. of arsenic trioxide and 0.1 grm. of antimony oxide were washed into the flask with a mixture of 50 C.C.of methyl alcohol and 8 C.C. of sulphuric acid. Hydrogen chloride was then passed through the contents of the flask for about ten minutes. The flask was heated on the water-bath, a stream of hydrogen chloride being continued during the distillation, which was repeated twice with 50 C.C. and twice with 20 C.C. of methyl alcohol. The arsenic was precipitated from the distillate by means of sulphuretted hydrogen, washed, dried, and weighed, and an amount equivalent to 0-0986 grm.of arsenic trioxide was found. For the next experiments I obtained specimens of viscera from a case which had been treated with salvarsan. In the first test 50 grms. of the kidney were heated in a Jena distillation flask with sulphuric acid, and the water distilled off as completely as possible.Then methyl alcohol was added and hydrogen chloride passed through the viscid contents of the flask, and the distillation effected as before. This method was not satisfactory. Ninety-eight grms. of the liver were treated with 10 C.C. of nitric acid (sp. gr. 1.4), and charred in the usual manner with sulphuric acid. The charred mass was washed into the distilling flask with 50 C.C.of methyl alcohol, and about 5 grms. of ferrous chloride were added. Hydrogen chloride was then passed in as usual, and the flask heated on the water-bath. The distillation was repeated twice with 50 C.C. and twice with 20 C.C. of methyl alcohol. The distillate was saturated with sulphuretted hydrogen, and the precipitated arsenious sulphide was purified by dissolving in ammonia and reprecipitating.The arsenious sulphide was then dissolved in nitric acid, the solution made alkaline with ammonia, and the arsenic precipitated as magnesium ammonium arsenate, and finally weighed as the pyroarsenate. Weight of arsenic trioxide per 100 grms. of the liver : From the sulphide, 3.6 mgrms. ; from the pyroarsenate, 3.1 mgrms. I am indebted to Mr. John Webster for the result of an estimation of arsenic in another portion of this liver. This estimation was carried out, in Dr.Willcox’s laboratory, by the usual extraction method, the arsenic being finally weighed as magnesium pyroarsenate.. Weight of arsenic trioxide found per 100 grms. of viscera was 3.3 mgrms. This result was confirmed by an estimation made by the Marsh- Berzelius method.Attempts to confirm the statement of Friedheim and Michaelis (Zoc. cit.), that the volatility of arsenic trioxide in methyl alcohol and hydrogen chloride is due to the formation of a methyl arsenious ester, were unsuccessful. An excess of arsenic trioxide was heated with dehydrated methyl alcohol and sulphuric acid for some hours, also methyl alcohol with arsenic trioxide was treated with hydrogen chloride, and also The residue was found to be free from arsenic.236 COLLINS: SEPARATION OF AIlSENIC FROM ANTIMONY AND OTHER with hydrogen chloride and sulphuric acid.I n each experiment the distillate was fractionally redistilled, with the result that the boiling-point did not rise appreciably above that of methyl alcohol. The compounds of arsenic with alcohol radicles have been investigated by the following: J.M. Crafts (Bull. SOC. Chim., 1870, 14, 99) obtained arsenates and arsenites of the alkyl radicles of the general formulae R,AsO, and R,AsO,. For the arsenites he used three methods involving the inter- action in a sealed tube of (1) ethyl iodide and silver arsenite at 150" C. ; (2) arsenic bromide and sodium ethoxide ; (3) arsenic trioxide and ethyl silicate at 200" C.The boiling-point of methyl arsenite is 128" C. Auger (Compt. rend., 1902, 134, 238, and 1906, 143, 907) states that he obtained small yields of esters by heating arsenic trioxide and alcohol together in a sealed tube for some hours, although he did not succeed in isolating the esters. The maximum amount of esterification in the case of methyl alcohol is given as 6.5 per cent.The method of estimation of the amount of esterification appears not to be very satisfactory (cf. J. Chenz. SOC., 1908, 93, 1367). In experimenting with propyl, normal butyl, and isobutyl alcohol with arsenic trioxide, he states that good results were obtained by removing the water formed either by fractional distillation or by passing the alcohol and water vapour over calcium carbide in an adapter, when the water is removed and the alcohol returns to the flask in which the reaction is taking place.W. R. Lang, J. T. Mackey, and R. A. Gortner (J. Chem. SOC., 1908, 93, 1364) obtained a yield of 33.8 per cent. of methyl arsenite by heating methyl alcohol and arsenic trioxide together in the presence of anhydrous copper sulphate under a reflex condenser, and fractionating the resulting liquid under diminished pressure.They state that they have prepared methyl arsenite by a similar method without the anhydrous copper sulphate. J. T. Mackey (J. Chem. SOC., 1909, 95, 605) prepared esters of the aliphatic alcohols with antimony trioxide, using anhydrous copper sulphate as a dehydrating agent.Some esterification appears therefore to take place when arsenic trioxide and methyl alcohol are heated together, the amount of ester formed being increased in the presence of a desiccating substance such as anhydrous copper sulphate. How- ever, under the conditions which prevailed during these estimations no evidence has been obtained for the formation of an arsenious ester.I propose to further examine the possibility of ester formation involved in the separation described in this com- munication, In conclusion, I desire to express my thanks to Dr. W. H. Willcox and Mr. John Webster for their practical interest and help, and to Mr. A. Chaston Chapman for the help he has afforded me throughout the course of this work. No statement is made as to yield obtained.The boiling-point of methyl antimonite is given as 65" C. CHEMICAL DEPARTMENT, KING'S COLLEGE, LONDON.METALS, WITH SOME APPLICATIONS TO TOXICOLOGICAL WORK 237 DISCUSSION. The PRESIDENT, having invited discussion, Mr. CHASTON CHAPMAN suggested that the function of the sulphuric acid might be not only to retain the more basic antimony, but also to assist esterification. He did not quite know upon what evidence Mr.Collins based his opinion that no forma- tion of ester occurred, and he (Mr. Chapman) could not help feeling that the separation was in reality dependent on the formation of a methyl arsenious ester. With the comparatively small quantities of arsenic used in Mr. Collins' experiments, the detection in the distillate of such an ester would not be easy; but it was quite clear that its formation would be favourably influenced by the presence of the sulphuric acid.Mr. C. REVIS said that this process seemed to be a great advance on that of Clark. He had had occasion to use Clark's process a good deal some years ago, but had not been able to obtain very satisfactory results with it, finding that antimony was carried over during the distillation, especially when the antimony was large in quantity as compared with the arsenic.When, however, no other metal but arsenic was present, the results were fairly good. Dr. WILLCOX said the process was itself an old one, and he (Dr. Willcox) believed that Taylor had claimed to be the originator of a modification of it for toxicological purposes.The process in that form, however, involved considerable risk of explosion, and it was not reliable for the separation of arsenic from small quantities of antimony and tin. Mr. Collins had tested his form of the process very thoroughly, and he (Dr. Willcox) believed that, if carried out under the conditions which Mr. Collins laid down, it would be found to be reliable for the separation of arsenic from antimony, mercury, bismuth, etc.I t would be very useful for toxi- cological work, especially in criminal cases of arsenical poisoning, because in such' cases doses of bismuth were frequently given, and there was considerable difficulty in the analysis of viscera containing arsenic accompanied by large quantities of bismuth when the usual method of precipitation as sulphides was employed.Apart from the question of the separation of arsenic from antimony, etc., this process mould be valuable in dealing with arsenic alone, as was shown by the results that had been obtained in the case of salvarsan-poisoning referred to. Mr. T. MACARA said that the process devised by the late Dr. Clark was intended €or use in connection with white metals, bronzes, and similar materials, and for such purposes it gave very good results when properly applied.It was essential to use strong hydrochloric acid, and not to concentrate the solution too much and so unduly raise its boiling-point. The one fault of the process was that it necessitated two or three distillations, but that difficulty might be overcome by passing hydrochloric acid gas into the distilling flask, as suggested by Mr. Collins, which would enable good results to be obtained at a comparatively low temperature. The PRESIDENT said that it seemed to him that Mr. Collins' process was essentially different from that of Dr. Clark. Of course, in a toxicological investigation i t would not be proper to assume that a volatile product was arsenic without weighing it as a definite arsenical compound; but for many purposes it would probably be238 AUSTIN AND KEANE: sufficient to boil the sulphide of arsenic with water, and titrate the arsenious oxide with iodine. Mr. COLLINS, in reply, said that the statement of Frankheim and Michaelis was that the methoddepended on the formation of a methyl arsenious ester, but so far as his own work had gone he had not been able to obtain any evidence of the formation of such an ester. He had, however, only as yet been able to make one experiment as to this, which showed that the boiling-point of the distillate was not higher than that of methyl alcohol, whereas the boiling-point of pure methyl arsenite mas 128O C. In this experiment dehydrated methyl alcohol and sulphuric acid were heated with arsenic trioxide, and hydrochloric acid gas was passed through ; on fractional dis- tillation it was found that the boiling-point did not rise appreciably above that of methyl alcohol. He hoped later on to make further experiments by carrying out the fractional distillation under reduced pressure, under which conditions some esterifica- tion might possibly take place.

ISSN:0003-2654    

DOI:10.1039/AN912370229b

出版商:RSC    

年代:1912

数据来源: RSC

摘要:

238 AUSTIN AND KEANE: THE ANALYSIS OF LITHOPONE." BY W. L. AUSTIN AND CHARLES A. KEANE, D.Sc., PH.D., F.I.C. (Read at the Meeting, May 1, 1912.) THE chief estimations required for the technical valuation of lithopone com- prise the determination of the zinc sulphide, zinc oxide, and barium sulphate. Several other zinc compounds have been shown to be present, in varying amounts, in the mixture, such as the carbonate, normal sulphate, and basic sulphate. Coffignier (Bull. SOC. Chirn., 1902, 27, 829, 943) has also drawn attention to the /OH in cases occurrence of soluble salts of zinc and of the hydroxy-aulphide Zn where the product has been imperfectly calcined ; but such constituents are to be regarded as abnormal, and should not be present in lithopones that have been properly prepared and washed.In addition, lithopone sometimes contains a small proportion of barium carbonate, and traces of impurities, including ferric oxide, alumina, lime, and magnesia. Whilst the determination of these secondary products is seldom required, their presence is of importance in selecting a method of analysis. This is especially the case in regard to the influence of the iron and calcium on the estimation of the zinc, and also in regard to the possible presence of zinc compounds, other than the sulphide and oxide, as any soluble salts of zinc are, of course, of no value as a pigment. The presence of barium carbonate can be readily determined by a separate estimation if required. The numerous methods that have been proposed for the analysis of lithopone consist in the determination of the total zinc, the total sulphide, and the barium sulphate.By calculating the total sulphide t o zinc sulphide, and subtracting its equivalent in zinc from the total zinc found, the proportion of zinc other than \SH * This work was carried out under the provisions of the Analytical Investigation Scheme,THE ANALYSIS OF LITHOPONE 239 sulphide is calculated and generally expressed as oxide. As a check on this last factor, the zinc oxide and carbonate can be extracted either with dilute acetic acid or with a mixture of sodium acetate and acetic acid, and estimated separately. These determinations have involved simply a combination of ordinary methods of analysis, of which the following may serve as examples : zinc by precipitation as carbonate and the zinc present as sulphide by treating the sample with 1 per cent.acetic acid, to remove the oxide and carbonate, and precipi- tating the remaining zinc again as carbonate. Kochs and Seyfert (Zeitsch. angew. Chem., 1902, 15, 802) precipitate the total zinc as sulphide, and estimate the oxide and carbonate by solution in 5 per cent. acetic acid and precipitation as sulphide. Amsel (Zeitsch. angew. Chem., 1902, 15, 174) also extracts the oxide with 5 per cent. acetic acid, and estimates both it and the total zinc volumetrically by titration with calcium ferrocyanide. For the estimation of the sulphide-sulphur as an indirect means of determining the content of zinc sulphide, J. F. Sacher (Furbenzeitung, 1910, 15, 1810) recommends treatment with metallic tin and hydrochloric acid, passing the evolved sulphuretted hydrogen into an ammoniacal solution of hydrogen peroxide, and precipitating the ammonium sulphate formed with barium chloride, whilst H.Wolff (Farbenzeitung, 1910, 15, 1859) estimates the sulphuretted hydrogen volu- metrically by absorbing it in iodine solution, and titrating back the excess of the latter with thiosulphate. J. S. Remington and C. Smith (Caoutchouc and Gutta- Percha, 1909, 6, 3259) estimate the sulphide-sulphur by oxidation with bromine and precipitation as sulphate, and the total zinc as carbonate, or, if salts of iron, calcium, or barium (soluble in hydrochloric acid) are present, as sulphide. The alternative to these many proposals that we recommend is to estimate the total zinc volumetrically and the zinc sulphide indirectly by oxidising the sulphide- sulphur to sulphate. The gravimetric estimation of the zinc as carbonate is inaccurate in presence of calcium and iron, both of which are usually present, and precipitation as sulphide is a tedious determination.A direct estimation of the sulphide-sulphur is especially desirable, as the value of the pigment is chiefly dependent on the proportion of zinc sulphide present, and of the various methods available oxidation followed by precipitation is simpler and easier to carry out than the determination of the sulphuretted hydrogen evolved on decomposition. For the volumetric estimation of the zinc we have employed a method devised by A. Voigt (Zeitsch. angew. Chem., 1889, 2, 307) for the analysis of zinc ores and products which is sufficiently accurate for the estimation, and which has the advan- tage that the results are not affected by the presence of iron or calcium.For the oxidation of the sulphide-sulphur either bromine or nitric acid may be used; we have found the former preferable, and carry out the estimation on the same lines as proposed by Remington and Smith (Zoc. cit.). Whilst, therefore, there is no claim to originality in the methods employed, we have found the combination adopted more rapid and adaptable than previous proposals, and sufficiently accurate for technical purposes. The method of analysis is as follows: 5 grms. of tbe finely-ground sample are treated with concentrated hydrochloric acid and a little potassium chlorate, the P.Drawe (Zeitsch. angeui. Chem., 1902, 15, 174, 229, 297) estimates the total240 AUSTIN AND KEANE: whole evaporated to a small bulk on the water-bath, taken up with boiling water, and the residual barium sulphate filtered off and thoroughly washed. The filtrate is collected in a 250 C.C. graduated flask, made up to the mark, and aliquot portions taken for the estimation of the total zinc by Voigt's method. The potassium ferrocyanide solution is made up by dissolving 46 grms. of the crystallised salt in a litre of water, and is standardised by a solution of zinc chloride containing 12.4476 grms. of pure zinc oxide per litre ; 1 C.C. of this solution is equivalent to 0.01 grm. of zinc. For the estimation, 50 C.C. of the above filtrate are taken, 10 C.C.of a solution of tartaric acid (200 grms. per litre) and 10 C.C. of a solution of ferric chloride (60 grms. per litre) added, the whole diluted to about 200 c.c., made just alkaline with ammonia, and titrated ; the completion of the reaction is determined by adding a drop of the solution to dilute acetic acid (1 : 3) on a pitted porcelain plate until a permanent blue colour is produced. After a preliminary estimation a sharp end-reaction is easily secured, and it is not affected by the presence of an excess of ammonia. I n a series of tests made, in which the quantity of ammonia added was increased up to twice the amount necessary for neutraiisation, the maximum difference in the titrations did not exceed 0.15 C.C. For the estimation of the sulphide-sulphur, 0.5 grm.of the sample is oxidised with bromine, after the addition of water, hydrochloric acid added, the residual barium sulphate filtered off, and the sulphuric acid in the filtrate precipitated by barium chloride, By calculating the sulphur thus found to zinc sulphide, and sub- tracting its equivalent of zinc from the total zinc estimated volumetrically, the content of zinc, other than sulphide, is obtained by difference, as stated above. The following results were obtained by this method with three different samples of lithopone ; the zinc, other than that present as sulphide, is calculated as oxide : I. 11. 111. BaSO, ... ... ... 64.36 67.59 72.80 ZnS ... ... ... .._ ... 30.98 27.82 24.14 ZnO ... ... ... ... 0.55 1-82 0.88 CaO, Fe203, BaCO,, Moisture, etc.4-11 2.77 2-10 100~00 100.00 100~00 -- -- As a check upon the volumetric estimation of the zinc, that in Sample No. 1 was also estimated by precipitation as sulphide, the result obtained being 21.44 per cent. zinc. as against 21.20 per cent. by titration. The indirect determination of the zinc oxide was also checked by treating 2 grms. of each sample with 5 per cent. acetic acid at the ordinary temperature for one hour, with occasional agitation, the dissolved zinc filtered off, precipitated as sulphide, and finally estimated as oxide in the usual manner. The results obtained were as follows : I. 11. 111. ZnO (calculated) ... ... ... 0.55 1.82 0.88 ZnO (estimated by precipitation) ... 0.75 1.70 0.90 These data confirm the reliability of the indirect method of estimation, and also show that the zinc, other than sulphide, in the samples analysed is practically all present as oxide, CHEMISTRP DEPARTMENT, THE SIR JOHN CASS TECHNICAL INSTITUTE, ALDGATE.THE ANALYSIS OF LITHOPONE 241 DISCUSSION.The PRESIDENT said that the specimens of lithopone that passed through his hands were usually ground in oil. Those who were experienced in work of this kind would know that it was never possible to extract the oil completely; some oxidised oil always remained unextracted, and this caused difficulty in the estimation of the sulphide sulphur by cxidation with bromine in an acid solution. His practice was first of all to add some caustic soda, to the extracted lithopone, and then pour in the liquid bromine and acidify. Under those conditions no globules of sulphur or sulphur bromide escaped oxidation. With regard to the estimation of the zinc, he could understand that, in laboratories where such estimations were made very frequently, it would be worth while to use a volumetric process ; but, when it was only necessary to make such estimations occasionally, a gravimetric process was, perhaps, equally expeditious. He should like to ask whether the authors had tried precipitating the zinc in a solution containing a minute trace of free mineral acid.Under those conditions the zinc was completely precipitated, and in a granular form. Of course, during the precipitation more acid was set free, and one must be careful to neutralise all but the very faintest trace. I n the analyses given by the authors the proportions of what might be called impurities were larger than he had been accustomed to find.Mr. T. MACARA asked whether the authors had tried the phosphate method of estimating zinc in the case of such a compound as lithopone. He thought that this method, which was proposed by the late Dr. Clark for the estimation of zinc in alloys, etc., would probably work well in such a case as this. Dr. H. I?. STEVENS asked whether the lime shown as CaO in the authors' analyses referred to caustic lime, Lithopone was used in rubber mixings, in which the presence of caustic lime was very undesirable. The PRESIDENT added that a certain amount of soluble SO, was always present, and it was his practice to determine this, calculating it as calcium sulphate; and perhaps if that were done in these cases, some, at any rate, of the lime would appear as calcium sulphate.In many cases the quantities of CaO and of soluble SO, found were just equivalent to one another, showing that all the lime was present as calcium sulphate. Mr. J. H. B. JENKINS asked whether the oxide of iron referred to in the analyses amounted to anything appreciable, One of the advantages of lithopone paint was its dead whiteness, and one would not expect this if any sensible quantity of iron were present. The majority of the samples of lithopone that he had examined contained more barium sulphate relatively to the sulphide of zinc than was shown in the authors' analyses. He believed that in the trade the proportions were supposed to be about 3 parts of barium sulphate to 1 part of zinc sulphide.Dr. DYER incidentally observed-although, of course, the remark did not apply to genuine (' lithopone "-that mixed pigments containing barium sulphate in the form of barytes sometimes contained considerable quantities of calcium fluoride, owing to admixture of fluor spar with the heavy spar from which the barytes was ground. He mentioned this because he had seen no published reference to it, and242 BANNISTER AND MCNAMARA: THE EFFECT OF LIME it was at first puzzling to find a considerable quantity of calcium present in a form other than the familiar carbonate or sulphate. Dr. KEANE, in reply, agreed that the method of estimating zinc as phosphate was an excellent one, but they wished to eniploy a volumetric rather than a gravimetric method, in order that the process might be as useful as possible for works purposes. He also agreed with the President as to the advantages of precipitating the zinc in the presence of a trace of mineral acid.Up to the present they had not had occasion to deal with small quantities of organic matter, such as would be derived from oil, but, in another connection, he had found that organic matter could be effectively got rid of by electrolysing the solution in presence of dilute nitric acid. In that way, even in the case of animal products containing a good deal of organic matter, a crystal-clear solution could be obtained containing no trace of organic matter. With regard to the lime, this undoubtedly was not present as such, but probably to some extent as sulphate, and as sulphide and carbonate. Of course, by estimating the soluble constituents one could always, if it were desired to do so, calculate the lime to calcium sulphate, as suggested by the President. The quantity of iron present was small, usually no more than a trace. The moisture was occasionally rather high, sometimes as much as 1 per cent. Of the samples analysed, the first was made in Germany, the other two being English products ; there are, of course, considerable variations in the relative proportions of barium sulphate and zinc sulphide in lithopones.

ISSN:0003-2654    

DOI:10.1039/AN9123700238

出版商:RSC    

年代:1912

数据来源: RSC

摘要:

242 BANNISTER AND MCNAMARA: THE EFFECT OF LIME ON THE EFFECT OF LIME ON THE AMMONIUM MOLYBDATE METHOD OF LEAD ASSAY.* BY C. 0. BANNISTER, A.R.S.M., AND W. McNAMARA. (Read at the Meeting, May 1, 1912,) DURING the analytical portion of the experimental work in connection with a recent research on the ‘( Theory of Blast Roasting of Galena ” (Bull. Inst. Min. and Met., 1912, No. 89), the necessity arose for ascertainiug the effect of lime on the ammonium molybdate method of lead assay. On referring to published results on this point, 8 great divergence of opinion was found to exist, some writers stating that the presence of lime had no influence, and others that the results obtained for lead in the presence of lime were too high. I. C. Bull (School of Mines Quart., 1902, 23, 348), who examined the effect of calcium, barium, strontium, antimony, and bismuth on the assay, came to the conclusion that calcium, bismuth, and antimony had no effect on the result, and that barium and strontium caused low results. In this work, however, Bull added the impurities to the solution of lead sulphate in a form in which they could not ordinarily exist in the course of assays, for the calcium, barium, strontium, and antimony were added as chlorides and the bismuth as nitrate.* This work m s carried out under the provisioiis of the Analytical Investigation Scheme.ON THE AMMONIUM MOLYBDATE METHOD OF LEAD ASSAY 243 On following out the method as originally devised by H. H. Alexander (Eng. and Min. J., 1893, 55, 298), it is obvious that only insoluble sulphates can be left with the sulphate of lead, and also that the impurities most likely to interfere are calcium, barium, and strontium.H. A. Guess (Amer. Inst. Min. Eng., 1904, 35, 359) points out that the molybdate method cannot be used with safety in dealing with poor lead ores containing large quantities of lime. W. H. Seamon (“ A Manual for Assayers and Chemists,” Chapman and Hall, 1910, p. 91) also states that the molybdate method is not reliable on low-grade ores containing much lime, and that it frequently fails when there is iron in the ores, all the sulphate of iron failing to go into solution, and obscuring the end-point with tannin. To overcome these difficulties, Seamon separates the lead by precipitation on aluminium. A. H. Low (“ Technical Methods of Ore Analysis,” Chapman and Hall, fifth edition, p.152) states that Alexander’s original method did not sufficiently provide for the presence of calcium, which is a frequent constituent of lead ores. Calcium forms a molybdate which is more or less insoluble under the conditions of titration, and tends to raise the results in a rather irregular manner. To overcome this, Low recommends a longer method, involving the separation of the lead from the ammonium acetate solution by means of ammonium sulphide. I n a recent paper on (‘ Quick Combination Methods in Smelter Assays ” (Bull. Inst. Hin, and Met., 1912, No. 89), A. F. French gives results showing that lime does not interfere with the molybdate assay as ordinarily carried out, and these results are confirmed by F, C.Robinson (ibid., 1912, No. go), who, in a written con- tribution to the discussion of French’s paper, states that when lime is known to be present the assay, after evaporation to the point at which SO, fumes are liberated, is not diluted, cooled, and allowed to stand overnight, but is diluted only when it can be proceeded with at once after cooling, as calcium sulphate crystallises out and is difficult to redissolve. By washing by decantation twice with 30 C.C. of cold dilute sulphuric acid (1 per cent. by volume) and once with water, calcium sulphate is sufficiently well removed. From the foregoing it is evident that the effect of lime on the assay is by no means clearly understood, and it is also evident that in order to correctly interpret the action of lime on the titration a solution of calcium sulphate in ammonium acetate should be used.Preliminary experiments showed that a solution of ctllcium sulphate in ammonium acetate does not form an insoluble molybdate with ammonium molybdate in the absence of lead, neither does it react with the lead molybdate formed during an ordinary lead assay; for if the assay be carried out to a satis- factory end-point, and then a quantity of solution of calcium sulphate in ammonium acetate be added, the end-point is found to be permanent, no more ammonium molybdate being required to react with the lime. If, however, a solution containing sulphates of lead and lime in ammonium acetate be titrated with ammonium molybdate solution, then the lime causes a larger amount to be required, thus giving high results, the excess necessary increasing with the lime added to a certain point, after which additional lime makes no difference.244 BANNISTER AND MCNAMARA: THE EFFECT OF LIME The following tables indicate the effect of the addition of increasing amounts of calcium sulphate : TABLE I.-Efect of Small Quantities of Calcium Sulphate on the Ammonium Molubdate Titration for Lead.Weight of Pb taken. ... ... ... ... ... ... ... ... ... ... ... ... ... ... Weight of CaSOJ added. nil 0.0036 0-0073 0.0109 0.0146 0-0183 0.0256 0-0292 0.0327 0.0366 0 -0402 0.0439 0.0475 0,0512 c. c. Molybdate Solution required. 29-0 29.1 29.2 29.5 29.8 29.8 29.8 29.9 29.9 30.0 30.2 30.7 30.7 30.7 Lead calculated, per Cent. 100.0 100.3 100.7 101 -7 102.7 103.7 102.7 103 *1 103.1 103 -4 104.1 105.8 105-8 105 -8 On slightly altering the conditions of the experiments-as, for example, by using more dilute solutions or by varying the amounts of reagents used-the results obtained were found to vary considerably.The following table will illuotrate this point; the determinations were carried out in exactly the same manner as the previous set, but a smaller quantity of lead was taken. TABLE 11.-Efect of Larger Quantities of CaSO, on the Ammonium Molybdate Titration for Lead. Weight of Pb taken. ... 1 . . ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... Weight of Ca80, added. nil 0.0549 0.0612 0.0658 0.0695 0.0732 0.0805 0.0842 0.0915 0.0951 0.1134 0.1317 0.1500 0.1685 0.1866 0.2049 0.2232 0.2415 C.C. Molybdste Solution required.18-7 19.4 19.6 19.6 19-6 19.6 19.6 19.7 19.7 19.8 19.9 19.9 20.0 20.0 20.0 20.2 20.2 20.4 Lead calculated, per Cent. 100.0 103.7 104.8 104-8 104-8 104-8 104.8 105.3 105-3 105.9 1064 1064 106.9 106.9 106.9 108.0 108.0 .109*1ON THE AMMONIUM MOLYBDATE NETHOD OF LEAD ASSAY 245 c. c . Ammoniiim Molybdate required. It will be seen, on comparing these tables, that mass action plays a considerable part in these reactions; for in the case of Table II., in which the lead solution is more dilute, 0.0915 grm. calcium sulphate (45.6 per cent.) interferes to a less extent than 0.0512 grm. calcium sulphate (15 per cent.) does in the results given in Table I. The table given below shows the effect of larger quantities of calcium sulphate on the assay.I n this case separate weights of the sulphate and of the calcium sulphate were taken, the same amount of ammonium acetate was used in each case, and the whole conditions of the experiment were kept identical. Lead calculated, per Cent. TABLE 111. Weight of Pb CaSO, taken. taken. Grms. CaSO, per Cent. of PbS04 present. 0.25 grm. ... ... 9 , $ 9 9, 2 , 9 9 > ? 7, 9 , 9 , 9 9 ... 9 9 9 , ... 9 9 9 , ... ... *.. ... - 0.055 0.110 0.183 0.256 0-329 0.366 0.549 - 15 30 50 70 90 100 150 23.7 25.1 26-1 26.7 27.1 27.1 27.1 27.1 100.0 105.9 110.1 112.6 114.3 114.3 114.3 114.3 From the results given it will be seen that calcium sulphate, even in small quantities, causes high results to be obtained for lead; but from Table 111. it will also be seen that the range of this interference is limited to a certain amount, above which no further increase in the lead results is obtained.When the conditions of the experiments are not kept absolutely constant, the presence of calcium sulphate raises the results in an irregular manner, as stated by Low (Zoc. cit.); but when the conditions are kept the same, then additions of calcium sulphate cause systematic increases in the calculated amount of lead present. There are indications that double molybdates are precipitated when calcium sulphate is present; for on calculatini from the molybdate used the composition of the precipitate formed in the case of the results given in Table 111.) the following results are found : Calculated Formula of Precipitate. CaSOl present. 15 per cent. of the PbSO, present.30 9 9 9 , 9 , 10 9 9 9 9 50 $ 9 9 9 9 ) 8 ) 7 7 , 70 to 150 ,, 2 9 > ? 7 9 2 ), 17 PbMoO,, CaMoO,. The results of similar experiments in which a solution of lime in ammonium acetate was used in place of the calcium sulphate solution are given in Table IV., and in this case a more dilute solution of ammonium molybdate was used, to insure greater accuracy, the first 100 C.C. in each case being added by means of a Stas pipette.246 BANNISTER AND MCNAMARA: THE EFFECT OF LIME TABLE IV.-E$ect of L i m e on the Arnmo.ltium Molybdate Titration for Lead. Weight of Pb taken. CaO added. Grms. - 0.0225 0.0450 0.0675 0.0900 0.1350 0.2250 04500 0.6750 CaO per Ceiit. of PbSO, present. ~ - 5 10 15 20 30 50 100 150 c. c. Ammonium bfolybdate required 106.5 108.3 109.1 109.9 110.7 113.0 114.6 115.4 111.8 Lcad calculated, per Cent. 100.0 101.7 102.4 103.2 103.9 105-0 106.1 107% 108.3 From these figures it is seen that lime acts in a, similar manner t o calcium sulphate in causing high results to be obtained for lead when the lime is present in the ammonium acetate solution of the lead sulphate.On carrying out a, large number of experiments with lime, under somewhat varying conditions as to strength of solution, etc., the same irregularities as to the amount of interference introduced were noticed. Similar experiments carried out with barium and strontium sulphates show that these act in quite a different manner; for by boiling mixtures of these sulphates with sulphate of lead in ammonium acetate solution somewhat low results were obtained in every case, owing to these insoluble sulphates holding small quantities of lead sulphate, and thus preventing the complete solution of the lead.From this work the following conclusions may be drawn: (1) That lime or calcium sulphate in solution in ammonium acetate does not form an insoluble molybdate when boiled with ammonium molybdate solution. (2) That lime or calcium sulpbate in ammonium acetate solution does not react with precipitated lead molybdate. (3) That when lime or calcium sulphate is present with lead sulphate in ammonium acetate solution, then, on boiling with ammonium molybdate solution, the insoluble molybdate formed contains both lead and lime. (4) From conclusion (3) it follows that it is impossible to determine lead sulphate direct by solution in ammonium acetate when lime or calcium sulphate is present ; and (5) in the ordinary lead assay for materials containing lime, etc., it is necessary to separate the lead, or else to take special precautions to wash the lead sulphate free from calcium sulphate before proceeding with the solution of the lead sulphate.(6) Barium and strontium sulphates do not act in the same manner as calcium sulphate, and tend to give low results instead of high results. METALLUKGY DEPAKTMENT, SIX JOHN CASS TECHNICAL INSTITUTE, ALDGATE.ON THE AMMONIUM MOLYBDATE METHOD OF LEAD ASSAY 247 DISCUSSION. Mr. RICHMOND asked whether the authors had any further data as to the composition of these double molybdates beyond the excess of molybdate used. Had the double molybdate been actually separated, and its composition determined ? Looking at the figures, it seemed to him that by no means all the calcium had been precipitated, and that the proportion that was precipitated was not the same in every case. He should like also to ask whether the authors could give the reason why there was no further increase in the percentage of lead found when the calcium sulphate exceeded 70 per cent.Was it due to the fact that the solution was saturated, and that the excess of calcium sulphate was not in solution? Or was it due to the fact that under the conditions of mass action an equilibrium was approached at the point where the maximum figure was obtained? Mr. RAYMOND Ross asked whether any of the impurities referred to were likely to be found in the finished lead-Le., in ordinary commercial lead. Mr. BANNISTER, in reply, said that the actual composition of the double molybdates had not been ascertained; it would be seen from the paper that the readings obtained were merely considered to be an indication that double molybdates were formed. I t was, however, curious that the results of a whole set of experiments should calculate out almost exactly in molecular proportions. They could not explain why the increase in the figures for lead stopped when 70 per cent. of calcium sulphate was reached. The saturation-point, however, was certainly not reached there, for even with as much as 150 per cent. of calcium sulphate the whole of this went into solution before the titration was commenced, These impurities were present in many lead ores, and also in lead slags and other furnace products, and it was important that the possibility of interference should be recognised in commercial work.

ISSN:0003-2654    

DOI:10.1039/AN9123700242

出版商:RSC    

年代:1912

数据来源: RSC

摘要:

ON THE AMMONIUM MOLYBDATE METHOD OF LEAD ASSAY 247 THE DETECTION OF HEAVY PETROLEUM IN PAINTS AND VEGETABLE OILS. BY W. B. POLLARD. (Read at the Meeting, iMay 1, 1912.) OWING to the present extensive use of turpentine substitutes, petroleum is now frequently found in mixed paints. Whatever views are held with regard to the value of light petroleum as a turpentine substitute, there is no doubt that heavy petroleum should never be present in paints. As the drying property of such paints can never be satisfactory, it is important to be able to detect the presence of this substance. Light petroleum can be readily separated byisteam distillation; and by the action of sulphuric acid, this may further be separated from any turpentine present. Heavy petroleum is not so readily detected, though it can be found by the method described248 POLLARD : DETECTION OF HEAVY PETROLEVM in Allen’s ‘‘ Commercial Organic Analysis.” This method, though satisfactory for quantitative work, requires far too much time and manipulation to be of much use as a qualitative test for petroleum, Experiments were therefore made to find in what way it could be modified, in order to give rapid results.It was noticed that when linseed oil is poured into an excess of molten caustic soda and stirred with an iron rod, complete saponification takes place in a few seconds ; the soap thus obtained, after cooling, can be extracted with petroleum ether without any passing into the solution and without any emulsion being formed. The petroleum ether extract, which can easily be fiitered, will then contain the unsaponifiable oil.Should heavy petroleum be present, it will often betray its presence by the strong fluorescence imparted to the solution. On evaporating the petroleum ether, petroleum is left behind, and a further examination can be made. In the case of normal linseed oils, a small residue of unsaponifiable oil, rarely exceeding 2 per cent., is always left after extraction ; it was noticed that this unsaponifiable oil could further be removed by boiling with a good excess of strong nitric acid, only the minutest trace, if any, remaining after this treatment. Should, however, heavy petroleum be present in the linseed oil, this will reveal itself by the large amount of unsaponifiable matter, and will remain unattacked by the treatment with nitric acid.It was found that as little as 0-5 per cent. of heavy petroleum, when present in linseed oil, could be detected by this means. It is most conspicuous when the nitric acid is warmed, as the oil tends to creep up the side of the tube. Light petroleum or turpentine is not detected in this way, as they volatilise a t the temperature at which the saponification takes place ; as previously mentioned, they would be detected by steam distillation. The actual method of conducting the test is as follows:-About 10 grms. of caustic soda are placed in a nickel crucible of 60 to 100 C.C. capacity, and fused over a Bunsen burner. When all the soda has just melted, the crucible is removed from the flame, and about 5 C.C. of the oil or paint quickly poured in.The contents are well stirred with an iron rod, and the crucible allowed to stand in a little water to cool. When cool, about 50 C.C. of petroleum ether are poured into the crucible, and the contents well stirred with the iron rod. By this method the soap forms a powder which settles at the bottom of the crucible, and an emulsion is never produced. The petroleum ether solution is decanted through a filter-paper and received in a test-tube. I t is then boiled off, the vapour being conducted away through a glass tube fixed by means of a cork in the mouth of the test-tube. The residue can now be further examined; this is done by boiling with about 3 C.C. of nitric acid, when any residue beyond a minute trace will be petroleum. Should a paint be found by this method to contain petroleum, it must be absolutely condemned, as its drying properties can never b5 satisfactory.Should this occur, it would be detected by the large amount of unsaponifiable matter, but would be distinguished from petroleum by the fact that it is oxidised by boiling with nitric acid. It might be thought that the Zeiss butyro-refractometer reading of the unsaponifiable matter would afford an indication of the presence of petroleum ; this, I t is sometimes stated that resin oil is added to paints.IN PAINTS AND VEGETABLE OILS 249 however, is not the case, as both the heavy petroleum and the unsaponifiable matter normally present in linseed oil give readings which are outside the scale of the instrument. This method may be used in examining other kinds of saponifiable oils for petroleum adulteration.A small quantity of unsaponifiable oil is left in nearly every case, but on boiling this with an excess of nitric acid, it is invariably destroyed. The method can also be applied in the examination of stearine candles, which, when they contain paraffin, soften at a temperature considerably below their melting- point. No claim is made that the method is quantitative, but iC affords a ready means for the detection of heavy petroleums when mixed with saponifiable oil, and should not take more than about ten minutes to perform. SURVEY DEPARTMENT LABORATORIES, CAIEO. DISCUSSION. P - 1 Mr. RAYMOND Ross said that a determination of the factor 7 in the residue would nearly always indicate whether it consisted of petroleum, and would in many cases show to what class of petroleum it belonged.In the case of nearly all petroleum oils this value was quite definite, varying from 0.557 to 0.543, while the figure for fatty oils was inuch different. The PRESIDENT agreed that it seemed improbable that so little as 1 per cent. of mineral oil would be used as an actual adulterant, unless it were added to mask some other addition. Small quantities of mineral oil were, however, sometimes met with in fatty oils obtained by extraction with a solvent, owing to incomplete evaporation of the solvent. This had the effect of considerably lowering the flash-point of the oil, and also, to some extent, the viscosity. Mr. E. RICHARDS BOLTON said that it was now unusual to find any trace of solvent in extracted oil, because in most modern extraction plants arrangements were made for blowing steam through the oil in order to remove the last traces of solvent, and manufacturers, for reasons of economy, were usually careful to carry out this operation as thoroughly as possible. The PRESIDENT said that, although it might occur but rarely, it certainly was occasionally the case that the manufacturer did not succeed in removing the lest traces of volatile solvent. A sample of extracted rape oil that he had lately examined lost 0.36 per cent. when 1 grm. was heated for an hour in the water-oven, and its flash-point (close test) was only 286" F., whereas the flash-point of ram oil obtained by pressure from the seed would not be below 400" F.

ISSN:0003-2654    

DOI:10.1039/AN9123700247

出版商:RSC    

年代:1912

数据来源: RSC

摘要:

250 SINNATT: APPARATUS FOR OBTAINING AN AVERAGE SAMPLE A CONVENIENT APPARATUS FOR OBTAINING AN AVERAGE SAMPLE OF GAS, AND FOR REGULATING THE FLOW OF A GAS INTO AN EVACUATED VESSEL. BY FRANK STURDY SINNATT. (Read at the Meeting, April 3, 1912.) THE apparatus was devised to permit of the accurate regulation of the flow of gas into an evacuated vessel. The rate at which the gas is collected in the exhausted D --ft'---- B €-- G - I I .:-.'. : I c P vessel is uniform within certain limits, and the pressure at which the gas is removed from any source is constant, but may be altered if required. After a preliminary adjustment little attention is necessary. The figure shows a form of the apparatus constructed of glass, and the dimensions mentioned are those used for an apparatus for small volumes of gas (about 1 litre).A long burette makes a I I I I I I I I I I I I I I I B Cg 4 I I I I I I I I I I I I I I I 1 I I I I I I I I I l a ;jy k-J convenient main tube G, but the diameter may be less than that of the usual burette, if the volume of gas to be obtained is small. By this means an economy in the mercury neceseary for the apparatus may be effected. The capillary tube E should be about 1 to 1.5 mm. in bore, and the walls should be of stout glass, in order to withstand the vibration which takes place in the tube. The volume of the capillary tube depends upon the diameter of the main tube. The end of the capillary tube is brought as close to the bottom of the main tube as possible, pro- viding the flow of mercury from the latter is not interfered with.The exit of the main tube should be short, or bent into a seal, as indicated by the broken line. The reservoir F may be replaced by a levelling bottle during the evacuation of the apparatus, but this must be removed before the actual working of the apparatus is started, The vessel in which the gas is to be collected-e.g., an ordinary gas sampler-is attached at A by pressure tubing, and by a second opening connected to a vacuum pump. The apparatus may be used in the following manner : The tap D is closed, and the whole exhausted, mercury boing allowed to enter the main tube from the reservoir F. When the mercury level has reached the required point in the main tube, the sampling vessel is closed to the vacuum pump, and then disconnected, The tap H is then closed. The sample of gas is obtained by con- necting the gas supply to 23, and opening the tap D slowly; owing to the difference in pressure between the gas vessel and the capillary tube, gas is drawn down the capillary tube, and passes through the column of mercuryinto the gas vessel.If the distance between a and b be great, the tap shouldOF GAS, AND FOR REGULATING THE FLOW OF A GAS, ETC. 251 be opened cautiously; the less this distance the better, so long as the diminished pressure is sufficient to overcome that of the gas to be sampled, When the diminished pressure in the gas vessel has reached that represented by the height of the mercury column from a to the surface of the mercury in the main tube, gas no longer enters the gas vessel. Mercury is then allowed to flow from the main tube, drop by drop, into the reservoir.As the mercury leaves the tube, gas enters the gas vessel, The rate at which the mercury is allowed to flow determines the rate of entry of the gas into the gas-sampling vessel. When the tap is set, the rate at which the mercury falls into the reservoir is constant, as it depends upon the height of the column of mercury a to b ; this flow is constant until the end of the capillary tube no longer dips into mercury. The distance a to b may be altered by moving the capillary tube, and consequently the gas may be drawn from a source at any diminished pressure required. By arranging for the mercury to run from the tube in a certain time, a, representative sample of the gas throughout that period is collected in the gas vessel.I t will be seen that the volume of the sample of gas collected is independent of the size of the apparatus itself. The con- dition of the gas is comparable with that of gas which has been collected in vessels previously filled with dry mercury. The use of mercury in aspirators of more than 1 litre capacity is generally con- sidered to be expensive, and too cumbersome for practical purposes. In this respect it is hoped the apparatus may be of some practical use. An efficient water-pump serves for many purposes for exhausting the apparatus, and by washing the whole apparatus through with the gas to be sampled, the error due to the presence of air may be eliminated. If the pressure in the apparatus be noted, and the volume of the sampling vessel is known, the volume of gas entering the vessel may be calculated.I t is essential that the temperature of the vessel in which the gas is being collected should be as constant as possible throughout the period of sampling, in order to pre- vent any variation in pressure from that cause. It will be found advantageous to arrange the capillary tube to touch the wall of the main tube throughout its whole length. Instead of the gas passing through the column of mercury in irregular bubbles, and splashing at the surface, it passes up the canal formed at the two glass surfaces in a continuous and regular stream. The passage of the gas through the column of mercury is then scarcely perceptible, being accompanied by a slight vibra- tion in the tube and mercury column. An alternative way of using the apparatus is to fill the main tube, connections, and capillary tube with mercury, by means of a levelling tube, and then attach a gas vessel which has been previously evacuated at A , and proceed in the manner described for admitting the gas. The tap D may now be opened fully. CHEMISTKY DEPARTMENT, SCHOOL OF TECHXOLOGY, MANCHESTER.

ISSN:0003-2654    

DOI:10.1039/AN9123700250

出版商:RSC    

年代:1912

数据来源: RSC

摘要:

252 SINNATT: THE USE OF METHYLENE BLUE THE USE OF METHYLENE BLUE AS AN INDICATOR IN IODIMETRIC TITRATI ONS. BY FRANK STURDY SINNATT. (Read at the Meeting, June 5 , 1912.) IN a previous publication (ANALYST, 1910, 35, 309) it was shown that a dilute aqueous solution of methylene blue may be used in place of starch as an indicator in iodi- metric titrations. The determinations quoted in that publication were made with a solution of methylene blue prepared by dissolving pure methylene blue in water. I t has since been found that the usual commercial forms of methylene blue may also be used without the delicacy of the colour change at the completion of the titration being interfered with. The solutions should be of a similar concentration to that suggested for the pure methylene blue-viz., 0.05 grm.per litre, and about 1 C.C. be taken for a titration. The following observations may be added to the previous note : I t will be found that the compound of iodine and methylene blue frequently separates from the solution as a precipitate. This occurs when an excessive amount of methylene blue has been added, but it will be found not to interfere with t-he observance of the colour change. There are no objections, such as hold in the case of starch, to the introduction of the methylene blue at the beginning of the titration, and no retardation of the change in colour has been observed. Titrations have been made in order to determine the influence of inorganic com- pounds upon the behaviour of the indicator, and neutral salts and the following acids --acetic, hydrochloric, or sulphuric-have been found not to affect the colour change.When a solution of iodine in alcohol or in aqueous potassium iodide solution is added to methylene blue dissolved in alcohol, the formation of the iodine compound of methylene blue is accompanied by the production of a green-coloured solution. The change in colour is delicate, and, although not very pronounced, is more easily observed than when no indicator at all is used. The addition of a reducing agent such as sodium thiosulphate to the green solution restores the colour of the methyl- ene blue. It is suggested that methylene blue may be of use for the determination of iodine in the presence of alcohol, where it would be impossible to use another indicator. It has been shown by Harrison (KoZZ.Zeit., 1911, 9, 5) that the formation of the blue coloration when iodine solution is added to starch solution is interfered with by the introduction of ethyl alcohol. The volume of ethyl alcohol which prevents the formation of the blue coloration varies with the concentrations of the solutions, and in the experiments quoted the blue coloration was not definitely produced when 30 per cent. of alcohol by volume was present in the liquid. Before this percentage of alcohol is reached the formation of the blue compound is too indefinite to be of value as an indicator. I n order to test the value of methylene blue as an indicator in the presence of alcohol, the following determinations were carried out. The solutions used were sodium thiosulphate solution (1 C.C.= 0.01309 grrn. iodine) made according to theAS AN INDICATOR IN IODTMETRIC TITRATIONS 253 & Iodine Solution. Indicator. usual directions, and starch solution made from pure potato starch. The ethyl alcohol contained 96 per cent. of alcohol by volume, I n carrying out the check titrations with starch solution, sufficient water was added to lower the percentage of alcohol present to about 2 per cent, Generally three titrations were made for each experiment, iodine solution (potassium iodide) was mixed with alcohol, 1 C.C. of methylene blue in alcoholic solution added, and the liquid titrated with sodium thiosulphate solution. +?! Sodium Thiosulphate Alcohol. required. Methylene blue ... Starc; ... ... 9 9 9 , ... 9 , C.C. C.C. C.C.10 20 9.7 10 50 9.7 10 100 9.7 10 - 9.7 The introduction of chloroform or carbon tetrachloride does not affect the titration if the percentage of alcohol present is sufficient to prevent the organic solvent separating from the solution. c. c. 10 10 - - Indicator. c. c. - 10 10 - Methylene blue ... 9 , * . * starci’ ... ... 9 , . . . ... c. c. 10 10 10 10 soform. I Alcohol. C.C. 100 100 - & Sodium Thio- sulphate required. c. c. 9.7 9.7 9.7 9.7 A solution of iodine in alcohol was prepared, having a concentration approxi- mately decinormal ; no potassium iodide was used. Indicator. Methylene blue ... 9 9 Y ? * * 59 1 ) ..- 9 , 9 , -.* ,, Starc: ... . .. i! Iodine. C.C. 10 10 10 10 10 10 Alcohol. C.C. - 50 100 100 100 - Na Sodium Thio- sulphate required. C.C. 12.15 12.20 12-20 12.15 12.20 12.15254 ABSTRACTS OF CHEMICAL PAPERS Indicator. %[ethylene blue Y 9 9 , 9 , 9 9 Starch ... Before the check titration, in which starch was used as the indicator, was per- formed, potassium iodide solution was added to the alcoholic solution of iodine, and then a large volume of water, and the titration carried on in the usual manner. For Wijs’ solution an aqueous solution of potassium iodide was added, then alcohol until the whole formed a clear solution. ?a Sodium Wijs’ Carbon Tetra- Chloro- Potassium Iodide. Alcohol. Thiosulphate Solution. chloride. form. required. C.C. c. c. C.C. C.C. C.C. I c. c. 10 - - 20 100 19.6 10 20 100 19.6 10 - 10 10 20 100 19.6 10 - - 20 - 19.6 - CHEMISTRY DEPARTMENT, SCHOOL OF TECENOLOGY, MANCHESTER.

ISSN:0003-2654    

DOI:10.1039/AN9123700252

出版商:RSC    

年代:1912

数据来源: RSC

摘要:

254 ABSTRACTS OF CHEMICAL PAPERS ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. ANALYSIS OF FOOD AND DRUGS. Amounts of Arsenic in Vegetable Foods. F. Jadin and A. Astrue. (Comptes. rend. , 1912, 154, 893-896.)-Estimations were made, by a modification of Marsh’s method, of the amounts of arsenic in thirty-six samples of fresh and dried vegetables and fruits, typical of all those of most common occurrence. In the fresh vegetables the proportion of arsenic ranged from 0.004 mgrm. (peas) to 0.023 mgrm. (lettuce) per 100 grms. The edible portion of apples contained 0.005, and that of oranges 0.011 mgrm. par 100 grms. ; whilst in the case of dried fruits and vegetables the amounts varied from 0.007 (rice) to 0.025 (almonds and red haricot beans) mgrm. C. A. M. Detection of Small Quantities of Chloral in Presence of Chloroform.T. Jona. (Giorm. Farm. Chim., 1912, 61, 57; through Chem. Zentralbl., 1912, I., 1148.)-Chloral is usually detected by the reducing action of the formic acid into which it is first converted. This method is inapplioable in presence of compounds con- taining the group CC13, such as chloroform and trichloracetic acid, whilst bromoform and iodoform may also give rise to illusive results. Chloral cannot be separated from chloroform by distillation. The following method is recommended : In a flask closed by a plug of cotton-wool, the substance, dissolved in water and slightly acidulated with sulphuric acid, is allowed to stand, without warming, in cohtact with zinc untilFOOD AND DRUGS ANALYSIS 255 the evolution of hydrogen has ceased; a strip of filter-paper, soaked in a fresh solution of sodium nitroprusside and 5 per cent.piperidine solution, is hung in the neck of the flask, and the liquid warmed, but not allowed to reach the paper. The acetaldehyde formed by reduction of the chloral colours the paper blue. In this manner 1 part of chloral hydrate in 20,000 can be detected. 0. E. M. Method for the Determination of the '' Gasoline Colour Value " of Flour. A. L. Winton. (Bull. 137, Bureau of Chem., U.S. Dept. of Agriculture.)-Winton and Shanley (U.S. Dept. of Agriculture, Bull. 122, p. 217) have suggested a qualitative gasoline test for bleaching, which corroborates the nitrite test, but it; does not give quan- titative results. The method described is of value in the comparison of patent or other high-grade flours which owe their colour to the kind of wheat, not to mechanical impurities, as a means of detecting bleaching.Only the colour soluble in oil is measured, and not the total colour, or the colour due to mechanical im-purities. Twenty grms. of the flour are placed in a wide-mouthed glass-stoppered bottle, and 100 C.C. of colourless gasoline are added. The bottle is shaken vigorously for five minutes, and is then allowed to stand for sixteen hours, when it is again shaken, and the contents filtered through a dry paper into an Erlenmeyer flask, keeping the funnel covered with a watch-glass. The colour value of the clear solution is determined in a Schreiner colorimeter, using for comparison a, 0.005 per cent. aqueous solution of potassium chromate; this solution corresponds to a gasoline number of 1.The colorimeter tube containing the gasoline solution should be adjusted so as to read 50 mm., and the tube containing the standard chromate solution is raised or lowered until the shades of yellow in both tubes match. The reading of the chromate solution, divided by that of the gasoline solution, gives the gasoline colour value. If desired, the colour value may be determined in Nessler tubes, using for comparison chromate solutions of various dilutions, and filling the tubes in all cases to the height of 50 mm. A number of determinations are given, showing the effect of variety, grade, ageing, and bleaching, on the gasoline value. Bleaching with nitrogen peroxide (2 parts of nitrous nitrogen per million) immediately produced about the same degree of whitening as ageing for thirty weeks.Ageing of the bleached flour brought about further bleaching, so that after thirty weeks hardly any colour remained. If the flour is known to have been milled within ten weeks, the gasoline colour value may be depended upon to corroborate the determination of nitrous nitrogen as a means of detecting bleaching. H. F. E. H. Composition of Hare Fat. J. Klimont. (Monatsh. Chem., 1912, 33, 441- 446.)--A specimen of this fat yielded the following chemical and physical constants : Sp. gr. at 50" C. ... ... ... ... ... 0.9134 Melting-point ... ... ... ... ... 35" c. Solidifying-point ... ... ... ... ... 29" C. Acid value ... ... ... ... ... 15.3 Saponification value ...... ... ... 204.9 Iodine value ... ... ... ... ... 107.0256 ABSTRACTS OF CHEMICAL PAPERS The fat was found to consist principally of tripalmitin mixed with glycerides of unsaturated, fluid fatty acids ; the latter, however, did not contain linoleic acid to any great extent. w. P. s. Saponificaton of Fats by Sulphuric Acid. A. Grun and 0. Corelli. (Zeitsch. artgew. Chem., 1912, 25, 665-670.)-A11 the experiments carried out by the authors on the action of sulphuric acid on pure triglycerides of the higher fatty acids yielded the same result-namely, the formation of diglycerides. On treating tripalmitin for several hours at 70" C. with a tenfold molecular quantity of sulphuric acid of sp. gr. 1.845, the resulting mixture contained 56.5 per cent. of free palmitic acid and 43.5 per cent.of neutral fat, the latter consisting of 30 per cent. of unaltered tripalmitin and 70 per cent. of dipalmitin. On doubling the quantity of sulphuric acid, the product consisted of 65 per cent. of free acid and 35 per cent. of neutral glyceride, the latter being 25 per cent. unaltered glyceride and 75 per cent. dipalmitin. Tristearin yielded approximately similar results. The reaction appears to proceed beyond the diglyceride stage, as is shown by the proportion of free fatty acid to neutral fat, and, although monoglycerides could not be detected at any stage of the reaction, it is pos- sible that monoglycerides are formed and immediately split up into fatty acids and glycerol or glycerol esters of sulphuric acid. w. P. s. Mkonga Oil.W. Arnold. (Zeitsch. Untersuch. Nahr. Genussm., 1912, 23, 391-392.)-The seeds of the Mkonga-tree (Balamites cagyptica), which grows in German East Africa, yield a yellow oil having a nutty taste and odour. A specimen of the oil yielded the following figures on analysis : Sp. gr. at 15" C. ... Saponification value Reichert-Meissl value Polenske value ... Iodine value ... Acid value ... Solidifying-point . . , Stearic acid ... Unsaponifiable matter [N]D 40: C. * - - ... Non-Volatile Fatty Acids. Oil. - ... ... 0.9173 ... ... ... 55.9 ... ... ... 195.6 ... 206 *1 ... ... 0.56 ... ... 0.40 ... 0-50 ... ... 77.2 ... 82.9 .. 8.5 ... ... ... Turbid at 8.8" C. ... - ... ... 2-4 per cent. ... 2-6 per cent. ... ... 0.07 ,, ,, ... - - - - The oil yielded negative reactions when tested for cottonseed and sesame oils.w. P. s. Sources of Error in the Estimation of the Polenske Value of Fats. W. Arnold. (Zeitsch. Untersuch. Nahr. Genussm., 1912, 23, 389-391.)-From the results of analyses of fats which have been published from time to time, it is evident that incorrect results are sometimes obtained in the estimation of the Polenske value, and the author draws attention to the errors in manipulation which tend to cause these discrepancies. Many of these points have been mentioned alreadyFOOD AND DRUGS ANALYSIS 257 (cf. ANALYST, 1904, 29, 154 ; 1909, 34, 223, etc.). The distillation flask should be of 300 C.C. capacity, and all the dimensions of the apparatus described originally by Polenske should be adhered to; the use of wire gauze under the flask should be avoided, but the flask should rest on an asbestos ring, the hole in which is 6.5 cm.in diameter. The flame of the burner should be sufficiently large to cover the bottom of the flask, but should not make the edges of the asbestos ring incandescent. Powdered pumice must be employed instead of pieces of pumice to prevent (‘ bump- ing,” and the distillate should measure exactly 110 C.C. As the Polenske value varies with the quantity of glycerol used in the saponification of the fat, care should be taken that exactly 20 grms. of glycerol be used for each estimation. In the author’s opinion, the correct working of any apparatus is ascertained by making an estimation of the Polenske value of a specimen of pure, fresh lard; the value found should be approximately 0*5, and should never exceed 0.65. w.P. s. Influence of the Concentration of Hydrogen Ions on the Baking Qualities of Flour. H. Jessen-Hansen. (Comptes rend, Trav. Lab. de Carlsberg, 1911, 10, 170-206.) -Determinations of the concentration of hydrogen ions were made with the object of ascertaining the effect of the addition of acid substances to flour, the methods employed for the measurement being the conductivity and colorimetric processes described previously by the author (ANALYST, 1909, 34, 498). It was found that for dough prepared from wheaten flour there is a certain definite concentration of hydrogen ions at which the best baking results are obtained, and that this concentration is higher than in dough prepared from freshly milled flour and distilled water or fresh milk.The optimum concentration corresponds approxi- mately with the exponent p $ = 5 ; in the case of the best qualities of flour it may be a little higher, and for ordinary flours slightly lower. The flour from the central portion of the wheat grains has a greater acidity-that is, a higher concentration- than the outer portion, the bran having the least. The addition of (( improvers ” to flour has scarcely any other effect than to increase the concentration of the hydrogen ions. Processes for the estimation of the gluten, which depend on the separation of the gluten by washing the flour, do not afford trustworthy data as to the baking qualities of flours. w. P. s. Incineration of Food-Stuffs and other Organic Substances prior to the Determination of their Phosphorus-Content.A. Vozarik. (Zeitsch. physiol. Chem., 1912, 76, 426-432; through J. SOC. Chem. Id., 1912, 31, 250.)-The phos- phorus in food-stuffs is in part present in the form of phosphatides, which are more or less volatile below 100’ C. For the incineration of such products, as a first step in the determination of the phosphorus, the author proposes the following modification of Eschka’s method : One grm. of the finely-ground sample if of animal origin, or 3 grms. if vegetable, is well mixed with 0.2 grm. of magnesia in a platinum crucible. The magnesia employed should be as light and voluminous as possible, and free from lumps. The open crucible is laid obliquely, and at first warmed carefully near the edge until all volatile matters have been expelled.The burner is then moved towards the base of the crucible, the cover is placed obliquely in the mouth,258 ABSTRACTS OF CHEMICAL PAPERS and the heating is continued until incineration is complete. The whole operation may require half an hour to two hours in the case of animal matters, and two to three and a half hours with vegetable matters. The ash is obtained as a powder, which does not adhere to the inside of the crucible. Determination of Phosphorus in Food-Stuffs and other Organic Sub- stances by Titration with Uranyl Solutions. A. Vozarik. (Zeitsch. physiol- Chem., 1912, 76, 433-456; through J. SOC. Chenz. Ind., 1912, 31, 250.)--In the determination of phosphorus in incinerated organic products (see preceding abstract) by titration with uranyl nitrate solutions, errors may arise from the presence of mineral acids, large quantities of acetic acid, various dissolved salts, especially the acetates of the alkali and alkaline earth-metals, and the phosphates of iron and aluminium.The liquid to be titrated should contain no phosphates acid towards phenolphthalein and no aluminium or iron phosphate. The disturbing influence of free acetic acid begins to be appreciable at a concentration of 0.36 per cent. when cochineal is used as indicator, and 0.6 per cent. when potassium ferro- cyanide is used. Errors are introduced by sodium, ammonium, and magnesium acetates at concentrations exceeding 4, 2, and 0-75 per cent. respectively. The presence of strong acids leads to high results with cochineal as indicator, but to low results with ferrocyanide.If both are used at the same time, the divergence of the results gives an indication of the magnitude of the error. When the liquid titrated contains a large quantity of foreign salts, the influence of the latter on the result must be determined and a corresponding correction made. A certain small volume of uranyl solution is required to produce the indicator reaction. If allowance is made for this, there is strict proportionality between the volume of uranyl solution required and the quantity of phosphorus present, irrespective of the concentration of the latter. The author gives full details for carrying out the determination of phosphorus in incinerated organic substances by this method. The results obtained agree well with those found by the grltvimetric citrate method.Formic Acid in Honey. H. Fincke. (Zeitsch. Untersuch. Nab. Genussm., 1912, 23, 255-267.)-That formic acid is not invariably present in honey is shown by the results of analyses of numerous samples of honey recorded by the author. Many of the samples were free from formic acid, but the majority contained traces of a reducing volatile acid (as estimated by the mercuric chloride method; cf. ANALYST, 1911, 36, 103), the quantity of which did not generally exceed 0-003 per cent. ; in a few cases the quantity amounted to 0-02 per cent., calculated as formic acid. The volatile acid was present partly in the free state and partly in the form of salts ; esters of the acid could not be detected.In the course of the investigation it was found that invert sugar, prepared by heating cane sugar with acids, yields small quantities of formic acid when submitted to steam distillation ; invert sugar prepared with invertase does not yield formic acid. w. P. s. Kola Nuts from British West Africa. (Bull. Inzp. Inst., 1912, 10, 34-37.)- The samples of ‘( kola nuts ” received during 1909-10 from the Gold Coast were found on examination to have the following composition :FOOD AND DRUGS ANALYSIS 259 Cola acuminata (genuine kola nuts) ... ... ... ... C. Johnsoni (false kola nuts). .. C. verticillata ,, ,, ... Moisture, per Cent. 1. 2. 1. 2. 1. 2. 50.16 15.2 0.73 1-7 1.5 2.1 49.32 14.8 0.67 0-7 1.3 0.8 52.60 14.2 0.55 0.7 1.1 0.8 I Caffeine. In Nuts after I In received, Nuts I Dryim:i$? C., per Cent.Estimation of Dry Matter in Potatoes and the Applicability of the Ordinary Tables. H. Bjorn-Andersen. (Zeitsch. anal. Chem., 1912, 51,341-359.) -From experiments of his own, as well as irom consideration of the data on which Maercker and others have founded their tables, the author concludes that the sp. gr. of potatoes, however carefully determined, is a most uncertain guide to their content of dry matter. In at least one experiment in ten the error will exceed k 1 per cent., which is far too large an error where the number to be estimated is 22.5 It 3 per cent., which limits for dry substance would include most potatoes grown for their starch. If a table is to be used, it should rather be one connecting dry substance as deter- mined by the most exact methods with the number obtained by some empirical, and not too slow or complicated, method of drying. The author disintegrates 4 or 5 kgrms.of potatoes in a mincing-machine with holes 0.9 cm. in diameter, and 2 kgrms. of the resulting paste is put through the machine twice more, using plates with holes 0.45 and 0.3 cm. in diameter. A 10-grm. sample is then dried for eight days at 99' to 100' C. If the drying be discontinued after twenty-four hours, the results are 0.4 * 0.04 per cent. too high, and it is suggested that drying in a vigorously boiling, well-ventilated steam oven for twenty-four hours, with a deduction of 0.4 per cent. from the value found, would afford a good empirical method. G. C. J. Deteetion of Adulteration of Cayenne Pepper.A. von Sigmond and M. Vuk. (Zeitsch. Untersuch. Nahr. Genussm., 1912, 23, 387-388.)-The authors have endeavoured to find a method for the detection of added oil in this condiment, but without success. As they have mentioned previously (cj. ANALYST, 1912, 56), the added foreign oil would remain on the surface of the particles, but attempts to separate this '' surface " oil by pressing the pepper betweeri filter-paper and extract- ing the oil absorbed by the latter by shaking the pepper with water and separating the oil by centrifugal action, etc., did not yield results which were of any use. The iodine value of the separated oil afforded no evidence of the presence of a foreign oil. w. P. s.260 ABSTRACTS OF CHEMICAL PAPERS Estimation of the Total Solids of Vinegar.P. Lehmann and J . Gerum. (Zeitsch. Untersuch. Nahr. Geruussm,, 1912, 23, 267-274.)-Windisch and others have shown that the total solids of vinegar may be estimated correctly by evaporating the vinegar, treating the residue three times with weter, and weighing the final residue ; the latter may then be dissolved in a definite volume of water and the sp. gr. of the solution determined, a reference to tables giving the amount of total solids corresponding with the sp. gr. found. The authors now give an indirect method of estimating the total solids, which consists in estimating the quantity of acetic acid in the vinegar, multiplying the percentage quantity of this constituent by 0.0015, and subtracting the product from the sp. gr. of the vinegar.The difference is the sp. gr. of the solution of the total solids, and the quantity of the latter is then found by reference to tables. In case traces of alcohol or other substances lighter than water are present, 50 C.C. of the vinegar should be evaporated to 15 c.c., diluted to 50 c.c., and the sp. gr. determined. The solution is then titrated with B potassium hydroxide solution, and the number of C.C. required for neutralis- ation is multiplied by 0.00018. The result, subtracted from the sp. gr., gives the sp. gr. of the solution of the total solids. w. P. s. Estimation of Glycerol in Wine. S. Rothenfusser. (Zeitsch. Urttersuch. Nahr. Genussm., 1912, 23, 332-337.)-1n the method proposed the glycerol is converted into oxalic acid after the wine has been treated with stannous chloride to precipitate organic acids, and with ammoniacal basic lead acetate to remove sugars (cf.ANALYST, 1910, 35, 208). When only a small quantity of sucrose is present in the wine, the addition of a somewhat large amount of the lead reagent effects its removal, provided that the casein solution is not employed; but if more than 1 per cent. of sucrose is present, the wine should be subjected to a preliminary heating with an acid in order to invert this sugar. The estimation is carried out as follows : A quantity of the wine is treated with an excess of sodium carbonate, and crystal- lised stannous chloride is then added until the mixture exhibits an acid reaction. The precipitated lactic and other organic acids are now removed by filtration, and the filtrate is diluted to a definite volume after the addition of ammoniacal basic lead acetate solution. The mixture is filtered, and an aliquot portion of the filtrate is treated with a small quantity of sodium phosphate and potassium carbonate, and evaporated to about one-third its volume. Powdered potassium permanganate is then added to the cold solution in excess, and after the lapse of thirty minutes the excess is destroyed by the addition of hydrogen peroxide. The oxslic acid formed by the oxidation of the glycerol is now precipitated as its calcium salt, which is collected, dissolved in diluted sulphuric acid, and titrated with permanganate solution in the usual way. Estimations of glycerol in artificial wines containing from 0-26 to 0-64 per cent. of glycerol yielded results which did not differ by more than 0.02 per cent. from the quantity present. w. P, s.

ISSN:0003-2654    

DOI:10.1039/AN9123700254

出版商:RSC    

年代:1912

数据来源: RSC

摘要:

EACTERIOLOGICAL, PHYSIOLOGICAL, ETC. 261 BACTERIOLOGICAL, PHYSIOLOGICAL, ETC. Measurement of the Oxidase Content of Plant Juices. H. H. Bunzel. ( J . Amer. Chem. Soc., 1912, 34, 303.)-The oxidising enzymes in plants are of importance in connection with respiration and in many diseases; and in special case8 they are concerned with the darkening of tea and of bread during its making, as well as with the hardening of the black Japanese lacquer prepared from the tree Rhus venzicqera. The author has elaborated a quantitative method involving the measurement of the amount of oxidation of pyrogallol by oxygen in the presence of the plant juice. The apparatus (see Fig.) is Only fresh unfiltered juice was used. clamped on the carriage of a shaking machine in a thermostat. Eight C.C.of fresh 1 per cent. pyro- gallol solution are measured into the compartment C by means of the bulb G. Two C.C. of plant juice are measured into the com- partment A from the burette F. The basket H is charged with 1 C.C. caustic soda solution. Only the stopcock I is left closed, and the interior of the thermostat is heated to the desired tempera- ture. About 30 minutes later the windows of the thermostat are opened enough to allow the infro- duction of the arm, and the stop- cock end closed. The shaking n machine is then brought into action at an approximate rate of 5 complete movements in 3.3 seconds. At 15-minute intervals the shaking is interrupted long enough to read the manometers. The experiment is considered com- G - o pleted when successive readings are identical. The carbonic acid liberated can be approximately determined by lifting off the basket H at the ground joint E, adding a drop of phenolphthaleln, and placing the whole in a special titration flask.In the experiments described, the pressure readings on the manometer are given, and the values reduced to the arbitrary volume of 150 c.c., this being about the capacity of the apparatus used. The absorption of oxygen comes to an end in about 2 hours, and the concen- tration of pyrogallol solution has no appreciable effect on the end result, provided it is over 0.2 and under 16 per cent. A certain quantity of potato juice is capable of bringing about the oxidation of a definite quantity of pyrogallol, any excess of which262 ABSTRACTS OF CHEMICAL PAPERS remains unoxidised. The total oxygen absorption is at least approximately pro- portional to the quantity of potato juice present.Excess of caustic soda is necessary in the basket H to insure complete removal of the carbonic acid formed during the experiment. Care must be taken not to splash any pyrogallol into the alkali, or no satisfactory end result is obtained. Previous workers have shown that many enzymes h e their activity on vigorous shaking, and experiment showed that if the potato juice was shaken for 15 to 30 minutes before the addition of the oxidisable substance, its oxidising power was reduced to about half the original value ; further shaking pro- duces no effect. This indicates that two phases of the process are concerned, each of which may be measured separately, and to obtain tho total oxidising effect of the plant juice measurements must be taken from the moment that shaking begins.The author has applied his method to the investigation of the curly-top disease of sugar beets, and finds that the oxidase content is markedly greater in the diseased leaves, and shows wide variations, while normal leaves appear to be fairly constant. The deviation in oxidase content of the pathological leaves runs parallel with the appearance of the leaves. The main source of error in using the apparatus lies in the rise of pressure within it as soon as shaking is begun; but by allowing the apparatus and its contents to stay at the chosen temperature for at least half an hour, this may be almost entirely overcome, Within the limits of the experiments the amount of the chemical change is directly proportional to the concentration of the oxidase present, all other factors remaining the same.It thus appears that the behaviour of plant oxidase is contrary to the usual conception of enzyme action in general, since it would have been expected that small quantities of juice would bring about the oxidation of relatively large quantities of pyrogallol, and that the action would continue so long as pyrogallol and free oxygen are present ; whereas the process comes to completion when only a small definite portion of the pyrogallol is oxidised, and while there is still an abundance of oxygen. I t therefore seems that the oxidasc in potato juice which accelerates the absorption of oxygen by pyrogallol is not an enzyme in the customary sense of the word, but is a substance entering directly into the reaction, and destroyed in the course of the same.As it is desirable to express the strength of the juice in terms of some standard, the author proposes as a unit for future experiments an oxidase solution of such strength that 1 litre of it will be capable of bringing about the consumption by pyrogallol of 8 grms. of oxygen. This unit of strength may not have any relation to the rate of the absorption, as it refers only to the total amount absorbed. It is left to future investigation to determine whether the strength of an oxidase solution, as expressed by this standard, is pro- portional to the rate at which the absorption takes place. The apparatus may be obtained from Machlett and Son, 143, East Twenty-third Street, New York City.H. F. E. H. Copper Fungicides. S. U. Pickering. (J. A g k Sci., 1912, 4, 273-281.)- Gimingham and Barker (ANALYST, 1911, 36, 351) questioned the validity of the principle that solubility of the copper compound is essential in order that it may exert its fungicidal actign, and doubt that such solubility is brought about by the action of carbon dioxide, suggesting that mere contact between the fungus and anBACTERIOLOGICAL, PHYSIOLOGICAL, ETC. 263 insoluble copper compound is sufficient to account for its action. I t is doubtful whether the carbon dioxide in air can render soluble more than mere traces of copper in ordinary Bordeaux mixture. The author controverts these opinions, and cites further experimental work in support of his view that the efficacy of such fungicides depends upon the proportion of copper in them which is rendered soluble by the carbon dioxide of the air. The amount of copper rendered soluble may be estimated by immersing clean iron rods in the solution, removing them at intervals, wiping them free from copper, and noting their loss in weight.Solutions too dilute to show copper by the ferrocyanide method (unless previously concentrated) can thus be shown to contain copper. The greater the excess of lime present, the longer is the interval before the solution of copper commences; but in practical spraying this time is shortened, since the deposit dries up and isolates the particles of basic sulphate from those of lime.I n spite of the small solubility of lOCuO,SO,, the greater part of the copper in it will pass into solution in a short time if any body be present, such as iron, to react with it and remove it from the solution ; and this will occur to a much greater degree in the presence of such a large reacting surface as fungus spores or plant leaves. If fungicidal action depended, as Gimingham and Barker hold, on a solvent material exuded from the fungus, all copper compounds, or, ah any rate, all the basic sulphates and carbonates, would be equally efficacious for a given weight of copper present, a proposition which is contrary to all experience. Ordinary Bordeaux mixture should be more effective than other compounds of the same copper content, for the lime itself appears to have a conspicuous fungicidal action, which action, the author claims, accounts for the results obtained by Giming- ham and Barker (Zoc.cit.), when they found that ordinary Bordeaux mixture was affected in an atmosphere free from carbonic acid-a result which they accepted as proving that carbonic acid does not bring the copper of such a mixture into solution. H. 2’. E. H. Estimation of Sugar in Blood. E. Herzfeld. (Zeitsch. physiol. Chein., 1912, 77, 420-424.)-The blood, or blood-serum, is treated with metaphosphoric acid to precipitate proteins; these are removed by filtration, and the filtrate, after the addition of potassium hydroxide, is titrated with a standardised methylene-blue solutiou. The latter is decolorised By aldoses, ketoses, and dextrins, in alkaline solution, and also by proteins, Cholesterol, and bilirubin-hence the necessity of removing these by treatment with metaphosphoric acid ; urea does not decolorise the methylene-blue solution.The standardisation of the methylene-blue solution is carried out by adding 0.5 C.C. of a 20 per cent. potassium hydroxide solution to an exactly measured quantity of a 0.1 per cent. dextrose solution, and heating the mixture nearly to boiling. When the hot solution begins to turn yellow in colour, a 0.001 per cent. aqueous methylene-blue solution is run in, drop by drop, until the colour is no longer discharged. Experiments carried out by the author show that 1 C.C. of the methylene-blue solution is equivalent to 0.00062 grm. of dextrose. I n the case of blood, 5 C.C.of the sample are mixed with 15 C.C. of a 10 per cent, meta- phosphoric acid solution, filtered after the lapse of ten minutes, and the precipitate washed with a small quantity of the metaphosphoric acid solution. The filtrate is264 ABSTRACTS OF CHEMICAL PAPERS then neutralised with potassium hydroxide, an excess of 0-5 C.C. of 20 per cent. potassium hydroxide solution is added, and the titration carried out as described above. The method yields concordant results, w. P. s. Copper-Reducing Power of Normal Urine. N. Schulz. (Zeitsch. p hysiol. Chern., 1912, 67, 121-128.)-When normal urine, free from sugar, is treated with copper sulphate solution, and then rendered alkaline with an excess of sodium hydroxide solution, a clear blue solution is obtained, provided that too much copper has not been added; on heating the solution for some minutes, reduction of the copper takes place.If, however, the urine be rendered alkaline before the addition of the copper solution, a turbid solution is produced, and the copper is not reduced (Trommer's test for sugar). Creatinine solutions behave in a similar manner, the solution of the copper hydroxide or otherwise, depending on the order in which the reagents are added ; as creatinine is converted into creatine by the action of alkalis, it is probable that this is the explanation of the reaction, w. P. s. Method of Testing Water. Determination of the Power of Bacteria to Decompose Proteins. W. Spat. (Arch. Hyg., 1912, 74, 237-288; through Chem. Zentralbl., 1912, I., 1332-1333.)-The method proposed consists in mixing quantities of the water under examination with varying amounts of a 2 per cent.sterilised peptone solution, maintaining the mixtures at 37" C., and estimating the ammonia formed. The activity of the bacteria as regards the production of ammonia is expressed in C.C. of i& solution. In the case of sterile waters, ammonia is not formed ; with waters containing but few bacteria the production of ammonia is very small, but impure waters loaded with bacteria produce considerable quantities. The amount of ammonia formed is not always proportional to the number of bacteria present, as the true water bacteria are capable of producing but very small traces of ammonia, and practically the same may be said of pathogenic organisms and B. coli; but soil bacteria produce large quantities of ammonia, and the method mainly serves the purpose of detecting the presence of surface water in a supply, the introduction of which water indicates that other contaminating matters may find an entrance. At ordinary temperatures the production of ammonia is negligible, even in the case of contaminated waters, and samples may therefore be transported without being packed in ice. When a water-supply is examined periodically, any sudden increase in the quantity of ammonia obtained almost certainly points to the introduction of surface water. Sewage produces large quantities of ammonia, the decomposition of the peptone being accompanied by the formation of an extremely offensive odour, and, although there is a certain relation between the amount of ammonia produced and the quality of the sewage, some of the ammonia is lost during the incubation of the mixtures. The method will readily detect as little as 0*0005 C.C. sewage. w. P. s.

ISSN:0003-2654    

DOI:10.1039/AN9123700261

出版商:RSC    

年代:1912

数据来源: RSC

摘要:

ORGANIC ANALYSIS 265 ORGANIC ANALYSIS. Part 111. (Bw,ll. Imp. Inst., 1912, 10, 27-34-cf. ANALYST, 1912, 61 ; 197.)-New essential oils from species of Cymbopogon and from vetiver roots were found to give the following analytical results on examination : Aromatic Grass Oils. Sp. gr. 15'/15' C. ... Opt. rotation, 100 mm. at 20' C. Aldehydes by sodium bisulphite, per cent. Citronellal, ,, Geraniol, 2, Solubility in- 80 per cent. alcohol 70 9 7 9 , Odour . . . ... Sp. gr. 15"/15' C. ... Opt. rotation, 100 Acid value . . . Saponification value Solubility in 80 per rnm. at 203C. ... cent. alcohol. Cymbopogon Coloratus (Fiji). 3 Samples. Cymbopogon Polynewos (Ceylon). 4 Samples. 0.915 to 0.920 to -8.7' 42.0 to 43.5 - 7.7" (at 22' C.) 45.7; 49% 15.6; 15.6 1 : 1 or more vols. 1 : 3; more alcohol gives cloudy solution.Varying between citronella and lemon-grass oils. 0.936 to 0.951 +30*9' to +55.25' - Total alcohols, 38.7 to 51.8 1 : l Opalescence to turbidity with 1 : 10 Unlike citronella or lemon-grass oil. Vetiver Oil (Fiji). 1.0298 - - 35.3 1 : 24; more alcohol gives cloudy solution. Oil from Seychelles Vetiver Roots. 1.0282 + 27.0' 55.9 67.3 1 : 1 or more. Cymbopogon Senna- arensis (Sondan). 1 Sample. 0.9383 + 34-25" (or ketones) 27 - Total alcohols, 17.3 Pennyroyal oil (pulegone present). Oil from Indian Vetiver Roots. 1.015 to 1.030 + 27.7' - 60 to 80 1 : l+ or 2 ; more alcohol gives cloudy solution. A. R. T. Estimation of Acetaldehyde in Paraldehyde. R. Richter. (Pharm. Zeit., 1912, 57, 125 ; through Chem. Zentralbl., 1912, .I, 1148.)-The sulphite method given in the German Pharmacopaeia for the estimation of formaldehyde was used.A solution of 10 grms. paraldehyde in 100 C.C. water is neutralised with 4 potassium hydroxide solution, 2 drops of phenolphthalein solution being used as indicator, 20 C.C. of an aquoous solution containing 25 grms. crystallised sodium sulphite solution in 100 C.C. added, and From the quantity of hydrochloric acid used is subtracted the quantity required to decolorise hydrochloric acid to decoloration.266 ABSTRACTS OF CHEMICAL PAPERS a mixture of 100 C.C. water, 20 C.C. of the sulphite solution, and 2 drops of phenol- phthalein solution. The difference must not exceed 9.1 (or 1-15, as the case may be) C.C. hydrochloric acid, corresponding to a maximum acetaldehyde content of 4 (or 0.5) per cent.acetaldehyde; 1 C.C. hydrochloric acid corresponds to 0.044 grm. acetaldehyde. Estimations of a number of commercial samples showed that paraldehyde with only a few tenths per cent. is easily obtainable. The author considers that paraldehyde of sp. gr. 0.998 to 1.00 may be required to show a, solidification-point of 10" to 11" C., and not more than 0-3 per cent. acid, 0-5 per cent. acetaldehyde, and 0.1 to 0.2 per cent, metaldehyde, determined by evaporating 10 grms. of paraldehyde at 50" C. 0. E. M. Method of Distinguishing between Aldoses and Ketoses. M. Betti. (Gazx. Chim. Ital., 1912, 42, 288-294.)-The method is based upon the fact that P-naphthol-benzylamine forms crystalline compounds of amino-aldehydic character with aldoses, but does not react with ketoses.With d-mannose the reagent forms a compound, C17Hl,0N : C,H1,O,, which crystallises in white needles melting (with decomposition) at 207" to 208" C.; with d-galactose it yields small prisms melting at 206" C. (with decomposition) ; with d-glucose, white silky needles melting (with decomposition) a t 192" C., and being more soluble in alcohol than the preceding compounds ; and with rhamnose, white crystals melting (with decomposition) at 192" C. In using the reagent for the separation of dextrose and lamdose, solution of 0.90 grm. of each of the sugars in dilute alcohol was added to a warm alcoholic solution of 2.5 grms. of the base, and the mixture left for twenty-four hours, and then allowed to evaporate spontaneously.The crystalline mass was triturated with a little water and filtered, and then washed on the filter. The filtrate and washings contained the laevulose. The residue, after being washed with petroleum spirit and crystallised from a large volume of alcohol, melted at 192' C., and by treatment with dilute hydrochloric acid yielded the dextrose again. C. A. M. Examination of Bituminous Materials. F. C. Garrett. (J. Sot. C'hem. Ind., 1912, 31, 314-317.)-The following methods of analysis were adopted : Water is determined by heating for five to six hours at 40" to 50" C., or by desiccation in wucuo over sulphuric acid for several days. The two methods give similar results, and a slight loss of organic matter takes place, but this is not important except in the case of pitches containing much water.Sulphur is determined by a modification of Eschka's method (ANALYST, 1906, 31, 52), the following results being obtained (the figures in brackets show the percentage amount of sulphur calculated on the organic matter only of the sample) : Sulphur per cent. in petroleum, 1.39 (1.39); iiquid pitch, 2.66 (2.66) ; soft pitch from Hatley's hole, Lot C, 2.88 (3.13) ; soft lake pitch, 3-62 (5.54) ; lake pitch, 3.81 (5.98) ; lake pitch dried by steam, 4.12 (6.52) ; land pitch, 3.55 (6.17); cheese pitch, 3.84 (6.38); iron pitch, 3.86 (6.27); impure pitch, seashore, 3.20 (7.67) ; manjak (Grahamite), 2.96 to 3-57 (3.13 to 3.74) in three samples ; glance pitch, 5-12 (5.17); Barbados manjak, 0.86 (0.89) ; Cuban hard pitch (Grahamite), 8.47 (8.59) ; elaterite (eIastic), 1-14 (1.26).ORGANIC ANALYSIS 267 Proximate Analysis by Heating.-About 2 grme.of the sample are placed in a platinum crucible and heated with the lid on by the tip of a Bunsen-flame for half a minute after the disappearance of flames round the crucible lid. The time required is two and a half minutes, the bottom of the crucible being faintly red during the last half minute. The crucible is then heated to bright redness for two minutes in the flame of a blast Bunsen, cooled without removing the lid, weighed, and the loss of weight (less moisture) noted as (( volatile hydrocarbons.” The coke is next heated to redness till burnt off, the loss of weight being “fixed carbon” and the residue “ash.” I n the case of pitches which melt and froth over, smaller quantities are heated in a larger platinum crucible, and the first heating is spread over five minutes.The method allows of distinctions being made between, say, lake pitch and land pitch. The amount of volatile matter, divided by the fixed carbon, gives the “volatility factor,” which in the lake pitches is above 5, and in the land pitches between 4 and 5. The following are typical of the results obtained : PETROLEUM AND THE ‘‘ PITCHES.’~ Moisture ... ... Volatile matter . . . Fixed carbon ... Ash ... ... ... Volatility factor . . . Petroleum Sp.Gr.,0.9226; Flash-point, 27” C.). Per Cent. nil 97.94 2-01 0.05 48.90 Liquid Pitch (Sp. Gr., 0’926). Per Cent. nil 93.00 7.00 0.02 13-28 Lake Pitch. Per Cent. 1-20 53.30 10.25 35.25 5 -20 THE GRAHAMITES, MANJAK, ETC.Land Pitch. Per Cent. 8.60 47.70 9.85 33.85 4.85 Iron Pitch. Per Cent. 0.60 50-60 11 -20 37.60 4.49 Cuban Hard Pitch (Grahamite). Trinidad Manjttk (Grahamite). Barbados Mrtnjak. Glance Pitch. Ela teri te (Elas tic). Moisture ... ... Volatile matter . .. Fixed carbon ... Ash ... ... ... Volatility factor .. . Per Cent. nil 53.55 45.00 1-45 1.19 Per Cent. 0.80 59.05 36.25 3.90 1-62 Per Cent. 0.50 64-60 31.90 3 a00 2.00 Per Cent. nil 74-80 24-20 1.00 3 -09 Per Cent. nil 96.70 1.10 2.20 87 -90 Proximate Analysis by Extraction with Solvents.-A piece of filter-paper, 9 by 6 inches, is dried at 100” C. and weighed, folded lengthwise, about 2 grms. of the pitch, accurately weighed, placed on the central portion of the paper and spread in a very thin268 ABSTRACTS OF CHEMICAL PAPERS Per Cent.- 35.23 10.42 19.64 34.71 layer, the sides of the paper turned over, and the whole rolled from each end towards the centre. The cylinder thus prepared is tied with a weighed platinum wire and introduced into a Soxhlet tube, and extracted with acetone till the pitch, on two successive weighings, after draining and drying for two hours at 100” C., loses less than 10 mgrms. during twelve hours extraction. The loss is recorded as “ petro- lene,” and the extraction then continued with chloroform, which similarly extracts the “ asphaltene.” The residue is ignited and the mineral matter weighed, the diiference between 100, and the sum of the petrolene, asphaltene, and the mineral matter being the organic matter not bitumen.” This usually consists of plant remains.The following are some of the results obtained : ANALYSES BY SOLVENTS. Per Cent. 0.67 35-82 17.03 9-13 37.33 Moisture ... ... Petrolene bitumen ( Organic matter not Asphalt en e} 1 bitumen ... ... Ash ... ... ... Liquid Pitch. Per Cent. - 90.95 5.10 - - Lake Pitch. Per Cent. 39.22 16.69 8.95 35.12 - Pitch. Pitch. Land I Iron Manjak (Grahamite). Per Cent. 17.06 80.70 - - 2.24 Barbados Man j ak . Per Cent. - 38.72 56.27 4.04 0.97 Glance Pitch. Per Cent. 34-64 64.84 L - 0.83 Estimation of Cellulose in Woods and Textile Fibres. A. R. T. J. Konig and F. Hiihn. (Zeitsch. Farb. Ind., 1911-1912. Series of ten articles.)-The authors have reinvestigated the various analytical methods proposed or adopted for cellulose estimation, and give their preference to acid methods of digestion, especially for elimination of unsaturated and oxidisable groups (lignone).They have particularly studied the methods of Konig (digestion with glycerolin sulphuric acid at 137” C., followed by alkaline oxidation) and Tollens-Dmochowski (digestion of d L crude fibre ” with strong nitric acid), The weights obtained in either case show a, considerable deficiency (10 to 12 per cent.) from those of the generally adopted methods. The discrepancy in the former case (Konig) is explained as due to removal of ‘‘ hemi- hexosanes ” from celluloses reputed as “ normal.” In the latter case Tollens ‘‘ corrects ” the figures obtained by multiplying by the factor 1.1, and the present authors see no ground of criticism of such a “ correction.” These two methods are selected and commended, and the methods of ordinary practice are adversely criticised.NorrE.-Either process leads to a product of degradation of doubtful relationship to the original cellulose. The process of chlorination (following alkaline hydrolysis) depends upon specific constitutional (and quantitative) relationships, has been there- fore generally adopted, and must be still commended. C. F. C.ORGANIC ANALYSIS 269 Modification of the Sulphonation Test for Creosote. E. Bateman. (U.S. Dept. Agriczdt., Forest Service, Circular 191, December 9, 1911; through J. SOC. Chem. Ind., 1912, 31, 425.)-The following modification of the sulphonation test is claimed to give more accurate results in a shorter time than the old method: Ten C.C. of the fraction of creosote under examination are treated with four successive portions of 10 C.C.each of $! sulphuric acid in a Babcock's milk-bottle, which is shaken for two minutes afterneach addition. It is then kept for an hour at 98' to 100" C., being meanwhile thoroughly shaken every ten minutes, and is subsequently cooled, filled to the mark with ordinary sulphuric acid, and whirled for five minutes in a Babcock's separator, after which the amount of unsulphonated residue is read upon the graduated scale. The reading, multiplied by two, gives directly the per- centage of unsulphonated oil by volume. The size of the steam-bath used to maintain a constant high temperature during the sulphonation is immaterial, provided it bO large enough to contain completely the bottle beneath its cover.Estimation of Formaldehyde. E. Rimini and T. Jona. (Giom. Farm. Chim., 1912, 61, 49 ; through Chem. Ze?ttraZbl., 1912, I., 1147.)-The usual methods are examined, especially that of Riegler (Zeitsch. anal. Chem., 1901, 40, 92), depending on the conversion of formaldehyde into formalazine by hydrazine : 2CH20 + H2N.NH, = C2H,N2 + 2H20. The excess of hydrazine is determined gasometrically in the sense of the equation : 15NpH,.H2SO4 + 10HIO,= 15N2 + 3H2O + 15H2S04 + 10HI lOHI + 2HI03= 6H2O + 121. Formalazine is slowly attacked by iodic acid, and low values may result. Riminin has simplified the method by working in an alkaline solution, when the reduction, does not go beyond the formation of iodide : 3N2H,.H,S04 + 2KI0, + 6KOH= 3Nz + 2KI + 3K2S04 + 24H20.Formalazine is stable in alkaline solution, and nitrogen from the excess of hydrazine only is measured. The method has been further simplified by the substitution of, mercuric chloride for hydriodic acid : N2H4.H,S0, + 6K0 E + 2HgCI2 = K2SO4 + 4KCl+ 2Hg + N, + 6Hz0. The formalin found by the Riegler method is shown to decrease with the duration of reaction-from 39.00 per cent. after two minutes to 28.40 per cent. after thirty-nine. Tables showing the satisfactory agreement of results by the standard methods with those by the authors' method are given. The estimation is carried out as follows : The volume of nitrogen evolved from 20 C.C. of 1-1 per cent. hydrazine sulphate solution by mercuric chloride is first determined in a Lunge nitrometer; 20 C.C.of the solution are then added to 5 C.C. of the formaldehyde solution (about 1.1 per cent.) and the excess of nitrogen with mercuric chloride measured. The results of the examination of commercial disinfectants containing formaldehyde are given. 0. E. M.270 ABSTRACTS OF CHEMICAL PAPERS Estimation of Formic Acid in the Presence of other Acids by Means of Potassium Permanganate. A. Fouchet. (BzdZ. SOC. Chinz., 1912,11,325-328.)- Small quantities of formic acid (1 to 5 mgrms.) may be estimated in the presence of large amounts of acetic, propionic, butyric acids, etc., by the following method : A solution of 0.05 grm. of the sample in a little water, is added to a mixture of 40 C.C. of crystalline sodium carbonate (50 grms. per litre) and 20 C.C.of potassium perman- ganate (5 grms. per litre), whilst simultaneously a blank test is made with water instead of the solution. In the case of very dilute solutions of formic acid, the permanganate solution should be diluted (1 : lo), and the other reagents also diluted. The two flasks are immersed for exactly three minutes in boiling water, after which 20 C.C. of dilute (1 : 1) sulphuric acid and 50 C.C. of ferrous ammonium sulphate solution (20 grms. of the salt and 30 grms. of sulphuric acid per litre) are introduced, and the excess of the iron salt estimated by titration with 0.5 per cent. permanganate solu- tion. The difference between the results obtained with each flask gives the amount of oxygen corresponding to the oxidised formic acid, each C.C.of the permanganate solution being equivalent to 1.25 mgrms. of oxygen and 3.51 mgrms. of formic acid. The test analyses quoted in detail show that the method gives accurate results. C. A. M. Estimation of Glucose in Leather. J. G. Parker and J. R. Blockey. ( J , SOC. Chem. Ind., 1912, 31, 268-269.)-The authors show that in the estimation of glucose in leather, higher results are obtained when normal lead acetate is used for precipitating the tanning matters, etc., than when basic lead acetate is employed. For instance, in one sample 8.7 per cent. of glucose was found with normal lead acetate as against 7.8 per cent. with basic lead acetate. The discrepancy is due to the fact that eertain cupric reducing matters--e.g., gallic acid, are not completely precipitated in the presence of acid.The aqueous extract from leather contains some free organic acid, acetic or lactic ; consequently the precipitation of the gallic acid would be incomplete whennormal lead acetate is used to an extent depending on the free acidity; on the other hand, basic lead acetate would neutralise the free acid, giving a more complete precipitation of the gallic acid, and a lower but more accurate value for the glucose. Lower results are also obtained if the acidity of the aqueous extract be neutralised with potassium hydroxide, using phenolphthaleln as an external indicator, care being taken not to allow the neutralised solution to stand long before removing the tannins with the lead solution. Previous neutralisation lowers the results for glucose not only with normal, but also with basic lead acetate.It has also been shown that the acidity of the solution affects the precipitation not only of the gallic acid, but also of certain of the tannins-vix., the group of catechol tannins. This is an additional argument against the use of normal lead acetate. J. F. B. Notes on the Gum Resins. H. Ingle. (J. SOC. Chem. Ind., 1912, 31, 272- 273.)-On the question of the b 4 weathering” of kauri and copal gums, the author is inclined to think that the “weathered” gum is probably quite as useful to the varnish-maker as the “bold” gum, and that the varnish prepared from it is lessORGANIC ANALYSIS 271 likely to deteriorate by further oxidation. The effect of 6‘ weathering ” on the gums is shown by a decrease in the iodine value and an increase in the acid value.Estimation of Kauri, M a d a Copal, and Dammar, in a Mixture.-As the acid values of these three gums are so different (kauri=50 to 70, manila=107 to 156, and dammar herrol=30 to 32), it was thought that a separation might be effected by means of sodium carbonate solution. Ten grms. of gum were finely ground, and treated with 300 to 400 C.C. of a 5 per cent. solution of sodium carbonate, boiled, filtered, and washed with the carbonate solution ; the filtrate was acidified, and the precipitate collected on a tared filter-paper. The results showed : Kauri gum, 6.1 per cent. ; manila, 92.9 per cent. ; dammar, 2.5 per cent. Estimation of Dammar.-The author describes a method based on solubility, which may be used in combination with the above method.The sample is ground to powder, and one or more grams are treated in a beaker or flask with a mixture of 3 parts of alcohol and 1 part of benzene; the residue is filtered off, washed with the same mixture, and then dissolved in carbon bisulphide, the filter being washed three times with this solvent, and the solution collected in a tared flask; the solvent is evaporated, and the extract dried and weighed. After the extraction with alcohol and benzene, kauri and manila gums leave no residue soluble in carbon bisulphide. Dammar herrol gives from 36 to 44 per cent. of matter insoluble in alcohol-benzene, but soluble in carbon bisulphide. If the mixture of gums is in lumps, s m e of the dammar may be picked out, and the actual solubility value of the particular sample employed may be determined.I n a mixture of the three gums the dammar may be estimated by this carbon bisulphide method, and the kauri and manila by the acid value and matter soluble in sodium carbonate; very satisfactory results are obtainable. SOLUBILITY TABLE OF GUM RESINS. Solvent. Acetone ... ... Alcohol ... ... Alcohol and Ameri- can turpentine Benzene ... ... American turpen- tine Amy1 acetate ... Ethyl acetate . . . Petroleum ether ... Kauri. Manila. Insoluble Partially soluble Soluble Insoluble 95.5 per cent. in- soluble 5.3 per cent. insoluble Soluble 18.6 per cent. soluble Partially soluble Soluble Soluble Insoluble 104.0 per cent. in- soluble * 11.5 per cent. in- soluble Nearly soluble 12.1 per cent. soluble Dammar. Insoluble Insoluble Partially soluble Partially soluble I 26.7 per cent.in- soluble Insoluble - J. F. B. * Result attributable to the oxidation of the solvent.272 ABSTRACTS OF CHEMICAL PAPERS Separation of Liquids, the Boiling-Points of which lie close together, or which form Mixtures of Minimum Boiling-Point. A. Golodetz. (Chern. Zeit., 1912, 36, 273-274, 297-298, 302-304.)-Young ( J . Chem. Soc., 1902, 81, 709) effected the dehydration of ethyl alcohol by taking advantage of the fact that ethyl alcohol, benzene (or hexane), and water formed a ternary mixture, which boiled constantly at a, temperature lower than the boiling-points of the single constituents, or of the mixtures of minimum boiling-point which alcohol forms with benzene and with water separately.No similar ternary mixture of minimum boiling- point has since been described, but many separations, otherwise difficult or impossible, may be effected by adding a suitable third liquid to the mixture it is desired to resolve, and then distilling. For example, toluene and acetic acid form a mixture of minimum b.-pt. (104O C.), the mixture containing 30 per cent. of acetic acid. Since benzene also forms with acetic acid a mixture of minimum b.-pt. (80" C.), pure toluene remains as the residue in the still when the toluene-acetic acid mixture is distilled with a small excess of benzene. Since a mixture of toluene and water in the propor- tions 4 : 1 boils constantly at 84' C., the toluene-acetic acid mixture may also be resolved by distilling with water; In this case it is the acid which is separated in a pure condition, and if only a, small excess of water be employed, as much as half the acid may be recovered of 99.9 per cent.strength. Similar methods serve for the separation of liquids which do not form mixtures of minimum b.-pt., but which boil at temperatures so nearly the same that separation by ordinary fractional distillation is impossible. For example, the last case may be reversed? and toluene made use of to prepare acetic acid of high strength from more dilute aqueous solutions. Another example of the resolution of mixtures of minimum b.-pt. is that of benzene and methyl alcohol by the agency of carbon bisulphide, which forms with methyl alcohol a, mixture of still lower b.-pt.; whilst among mixtures of liquids with b.-pts.lying very close together, the most striking separation is that of amyl bromide (b.-pt. 1184' to 119" C.) and ethyl butyrate (b.-pt. 119.9' C.) by distillation with propyl alcohol, which forms a mixture of minimum b.-pt. with amyl bromide. G. C. J. Ester Method for the Detection of Mono-Amino-Acids together with Polypeptides. E. Abderhalden and R. Hanslian. (Zeitsch. physiol. Chem., 1912, 77, 283-288.)-The ester method of separating amino-acids resulting from the cleavage of polypeptides, etc., the esters being formed by saturating the alcoholic solution of the acids with dry hydrochloric acid gas, yields trustworthy results, and there is no risk of producing further quantities of amino-acids during the esterification. I t is usual to cool the mixture during the latter operation, but experiments with poly- peptides of high molecular weight showed that further decomposition does not take place when the mixture is not cooled.Silk fibroin yields no soluble nitrogenous product when esterified, but when the alcoholic solution is saturated with hydro- chloric acid and boiled for six hours, considerable quantities of amino-compounds are formed. I n the case of casein, boiling the esterified mixture results in only a very small increase of the amino-compounds. w. P. s.ORGANIC ANALYSIS 273 Estimation of Mercury in Smokeless Powder. J. D. Berkhout. (Zeitsch. yes. Schiess- u. Sprengstofwesen, 1912, 7, 67.)-In a 50 C.C. wide-mouthed stoppered bottle 2.5 grins. nitrocellulose or nitroglycerine powder are covered with 20 C.C.acetone and kneaded with a glass rod. The sample is sufficiently softened in eighteen hours, and is then transferred to a layer of fine copper oxide in a mortar ; the bottle is rinsed out with copper oxide into the mortar. The powder and copper oxide are mixed with a spatula until they form a fine-grained dry mass, most of the acetone having evaporated; this is transferred to a soft-glass tube of 50 cm. length and 14 mm. bore, closed at one end. Ignited asbestos fibre, rinsings of copper oxide from the mortar, asbestos, 15 cm. of coarse copper oxide, and asbestos, are succes- sively introduced, and the tube drawn out into five constrictions, 2 mm. in diameter and 5 cm. long, and four bulbs. The tube is closed with a rubber stopper and bent tube dipping into water.The tube and one bulb are encased in asbestos-lined sheet- iron; the three outer bulbs, projecting through an asbestos card, are water cooled. Heat is applied first to the coarse copper oxide, and gradually extended to the rest of the tube at such a rate that not more than two bubbles per second are given off. When gas evolution has nearly ceased, the capillary is sealed off between the tube and the first bulb, filled up with concentrated hydrochloric alcohol and alcohol, heated in the water-bath, and inverted upon a watchglass containing water. The globule of mercury is measured under the microscope or weighed. 0. E. M. Estimation of Sulphur in Nitrocellulose. C. Kullgren. (Zeitsch. yes. Schiess- ZL. Sprengstofzoesen, 1912, 7, $9.)-The sulphur is present, as a sulphuric acid ester, in combination with nitrocellulose.After several of the known methods for the estimation of sulphur in nitrocellulose had been investigated and rejected, the sample was denitrated by heating it with 10 C.C. of concentrated hydrochloric acid for thirty minutes on the water bath, dried in a large porcelain boat, and burnt in a current of oxygen. The sulphuric acid was arrested by sodium carbonate on granulated quartz, A small quantity of sodium peroxide was added to the solution before the precipitation with barium chloride, to insure that all sulphur is present as sulphuric acid. Nitrocellulose boiled for one-quarter of the usual time showed 27.8 mgrms. of barium sulphate per grm. of substance, unboiled about the same, boiled for one-half the usual time, 17.2, and completely boiled, 14.4. 0.E. M. Influence of Lueidol (Benzoyl Peroxide) on Certain Qualitative Tests for Oils. Atz. (Chem. Rev. Fett- w. Ham. Ind., 1912, 19, 70-72.)-1f excessive quantities of the new bleaching agent, “lucidol,” have beon applied to olive or cottonseed oils, colorations may be obtained with Baudouin’s or Soltsien’s reagents, which have some resemblance to those given by sesame oil. On the other hand, in certain tests for other oils-e.g., Becchi’s reaction for cottonseed oil-the coloration is produced with the same intensity in the presence of ‘( lucidol,” but the occurrence of the reaction is retarded. I n the case of some other colour reactions, the intensity of the coloration is sometimes reduced, as, for instance, in the reactions for cotton- seed and sesame oils.I t is, therefore, necessary to take the possibility of the presence of “ lucidol ” into account in testing oils, such as olive oil, for the presence of small quantities of other oils. C. A. M.274 ABSTRACTS OF CHEMICAL PAPERS Estimation of Petroleum Spirit and Benzene Hydrocarbons in Turpentine oil. J. Marcusson. (Chem. Zeit., 1912, 36, 413-414 ; 421-422.)-1n Burton's method of estimating petroleum spirit in turpentine oil the sample is treated under specified conditions with fuming nitric acid, in which turpentine oil is completely soluble, whereas in the presence of petroleum spirit one portion of the latter (naphthenes and paraffin hydrocarbons) is undissolved, whilst another part (aromatic hydrocarbons) is dissolved, but may be separated from the nitro-derivatives of the turpentine oil by means of its insolubility in water.Criticisms have been brought against this method by Herzfeld (Chem. Zeit., 1910, 34, 885) and others, but the author shows that it gives good results, which are independent of the nature of the petroleum spirit, which may contain but little benzene hydrocarbons (as in the case of Russian and Pennsylvanian products) or be rich in those hydrocarbons (Indian and Galician petroleum spirits). The method is, therefore, to be preferred to that of Eibner and Hue (ANALYST, 1910, 35, 358), which, as its originators showed, is not applicable to the heavy Indian petroleum spirits, such as Sanga oil, which at the present time are extensively used in the preparation of substitutes for turpentine oil.I n testing a mixture 10 C.C. of the sample are introduced, little by little, during the course of thirty minutes, into 30 C.C. of fuming nitric acid (sp. gr., 1.52) chilled to - 10' C. The mixture is left for fifteen minutes, and then treated with 75 to 80 C.C. of concentrated nitric acid (sp. gr., 1-4) ; and should no oily lager appear upon the surface of the liquid, the absence of petroleum spirit is indicated. For the detection of benzene hydrocarbons, the solution is poured into 150 C.C. of water in a 500 C.C. flask with a neck graduated in &ths c.c., and heated for fifteen minutes on the water-bath, after which the flask is allowed to stand for some hours. The separation of heavy reddish-brown oily drops (not merely slight resinous masses) denotes the presence of benzene hydrocarbons. For an approximate estimation of these hydro- carbons, the product of the reaction is treated with sulphuric acid (sp.gr., 1.6) until the oily nitro-derivatives are driven into the graduated neck of the flask, where their volume may be read. I n test estimations with mixtures containing from 10 to 50 per cent. of benzene, toluene, or xylene, the results agreed within 4.5 per cent. of the theoretical amount, whilst with technical benzene preparations (" solvent naphthas ") the maximum difference was 3 per cent. If the presence of more than 50 per cent. of benzene hydrocarbons be suspected from the bromine value, etc., the heating of the diluted nitric acid solution should be omitted, since, in the absence of suficient turpentine oil, the hydrocarbons in these solvent naphthas are attacked under these conditions.On the other hand, omission of the heating causes the proportion of nitro-derivatives to be too high, and to correct this the result should be divided by the factor 1.15. The figures will then correspond within about 4 per cent. with the theoretical amounts. C. A. M. Application of the Formolite Reaction to the Analysis of Paraffin Wax. F. Sommer. (Petroleum, 1912, 7, 409 ; through Chem. Zerrztralbl., 1912, I., 1151.)- The oils contained in commercial paraffin wax determine its tendency to turn yellow and its transparency, both important factors in a valuation. Drainings from the paraffin sweating process were repeatedly sweated, cooled to below 0' C., and pressedORGANIC ANALYSIS 275 in filter-cloths.The resulting oil contained only small quantities of saturated hydro- carbons; its iodine value was 444; three-quarters of it dissolved in concentrated sulphuric acid, and it gave 63 per cent. of formolite. To represent the transparency in figures, the liquid hydrocarboos must be determined ; the yellowing tendency is, however, a function of the formolite number, as the table of results shows. This is determined as follows : To 20 grms. of the molten paraffin in 20 C.C. of concentrated sulphuric acid are gradually added 20 C.C. of formaldehyde. The dark red rnixture is heated on the water-bath for twenty minutes, poured into a porcelain dish, left on the water-bath until paraffin and reaction-mixture separate, and cooled ; the cake of paraffin is removed, and the liquid beneath diluted and shaken out with chloroform, The chloroform extract is evaporated, the residue dried at 105' C., and the formolite weighed.The formolite number is suggested as a criterion of the quality of paraffins; factories having a uniform supply of paraffin oil can compare colorimetrically the chloroform solution with a normal colour. The authors state that a paraffin wax with more than 0.75 per cent. oil is always opaque. All samples examined were exposed to diffused daylight. 0. E. R.I. Estimation of Tar in Producer Gas. A. Gwiggner. (Chem. Zeit., 1912, 36, 461-462.)-The apparatus of Tieftrunk is useless for the estimation of tar in hot-producer gas. A wrought iron tube, S, is screwed into the gas main, G, to support the tar-filter, R, which is formed from combustion tube 30 to 35 mm.in diameter, packed withslagwool, and drawn out a t one end and constricted a t the other. Its wide por- tion is about 40 cm. long. The condenser tube, r , carries a removable con- denser, Ka, and leads into the measuring tube, P. Ku and Kb, the beaker in which P stands, are filled with ice water. C, and C, are tubes containing calcium chloride, and A is an aspirator of at least 10 litre content. Before uBe R is weighed separately; r and P, disconnected from the condenser, Ka, are weighed together; and C, and Cz are weighed separately. Even with a very brisk stream of gas, no visible fog reached C,, and the increase in the weight of this tube is due to moisture alone.The increase in weight of the system r + P is due to light oil and moisture, and the latter can be separately estimated by washing down r with benzene or light petroleum, diluting the contents of P somewhat with the same solvent, and reading off the water by means of the graduations on P. The increase in the weight of R is due to soot, tar, and some moisture. To determine these separately, R is The apparatus now figured serves well for this purpose.276 ABSTRACTS OF CHEMICAL PAPERS connected to apparatus similar to the system r +- P + C,, and a current of hot dry air is led through the whole train of apparatus. The moisture finds its way to the graduated tube and calcium chloride tube. Usually a little light oil finds its way into the graduated tube, and is separated from the water in the manner already described.The tube R is next weighed, its contents washed with benzene so long as the escaping benzene is discoloured, and is finally dried and weighed. The loss of weight on treatment with benzene gives the content of tar, and the increase of weight as compared with the weight at the commencement of the test gives the content of soot. Some cen weighings in all are necessary, but it is said that no simpler method will give results of any value where the gas to be sampled is producer gas at temperatures in the neighbourhood of 600" C. G. C. J. New Uric Acid Reaction. D. Vitali. (Boll. Chim. Farm., 1911, 50, 799; through Chem. Zentralbl., 1912, I., 1252.)-Ganassini (Boll. Chim. Farm., 1908, 47, 715) has stated that uric acid in alcoholic solution gives, with various oxidisiag agents an oxidation product which forms a characteristic blue basic salt with zinc salts. The author finds that the oxidation may be effected with sodium peroxide, and that as little as $T C.C. of freshly prepared alkaline uric acid solution then gives a positive reaction. A reaction is not obtained with other oxidising agents, such as hydrogen peroxide, turpentine, turpentine + blood, pus, and copper sulphate + hydrocyanic acid. Permanganate solution produces a yellow coloration, and, on the addition of zinc sulphate, a white precipitate insoluble on warming. If copper sulphate is substituted for zinc sulphate, a green precipitate, or with a larger quantity, a blue precipitate, is formed; cobalt chloride, and a nickel salt give respectively blue-violet and yellow-green precipitates, reactions which are very sensitive, and can be used to distinguish between nickel and cobalt salts. The precipitate with neutral antimony acetate is brown-red ; with mercuric chloride white, or on warming, vermilion ; with silver nitrate metallic silver is obtained. The blue colour of the basic zinc urate is destroyed by the addition of sulphur or of hydrocyanic acid. The reducing action of zinc or of sodium hydrogen sulphite in the alkaline reaction mixture does not cause decoloration. 0. E. M.

ISSN:0003-2654    

DOI:10.1039/AN9123700265

出版商:RSC    

年代:1912

数据来源: RSC

摘要:

276 ABSTRACTS OF CHEMICAL PAPERS INORGANIC ANALYSIS. Volumetric Estimation of Beryllium. B. Bleyer and A. Moormann. (Zeitsch. anal. Chem., 1912, 50, 360-367.)-Normal beryllium salts are so far hydro- lysed in aqueous solution that their solutions have a fairly strong acid reaction, and will set free iodine from a solution of potassium iodide and iodate. If sufficient of this solution be added, the hydrolysis is quantitative and the beryllium may be estimated from the iodine set free-BeC1, = I,. If the iodine be distilled over into potassium iodide solution, using a curent of hydrogen to drive the last of the iodine vapour out of the apparatus, the results, as determined by titration of the iodine with thiosulphate, will be constantly 1 per cent. below those indicated by theory.Omission of the current of hydrogen doubles the error, which in such circumstances ie also less constant.INORGANIC ANALYSIS 277 In pure aqueous solution, salts of beryllium may be estimated by titration with standard caustic alkali, using phenolpthalein as indicator. Beryllium sulphate must be converted into chloride by addition of barium chloride before it can be titrated in this way. I n the ordinary course of analysis the beryllium salt most often presents itself in solutions containing excess of acid. I n such cases concentrated caustic sod&, free from carbonate, is added to the solution drop by drop until the precipitate of beryllium hydroxide first formed has re-dissolved as beryllate, and the alkaline solution is made up to known bulk. Aliquot portions are then titrated with standard acid, using phenolpthalein and methyl orange as indicators.The difference between the two titrations is the measure of the beryllium-Be = SHCI. The sokiltion should not contain much more than 0-03 per cent. of beryllium, as otherwise the end point of the phenolphthalein titration is a little difficult to determine. The titration should be made at a temperature of 30" to 40" C., and it has to be remembered that towards the end of the methyl orange titration the reaction proceeds somewhat slowly. The test numbers which accompany the paper are all almost exactly 1 per cent. low. G. C. J. Estimation of Boric Acid in Nickel Baths. A. Wogrinz and J. Kittel. (Chem. Zed., 1912, 36, 433-434.)-Experiments are described which show that the amount of boric acid in nickel-plating baths cannot be estimated with any approach to accuracy by the method of Rose, or by that of Marignac.Methods depending on the titration of the boric acid in presence of a large quantity of some polyhydric alcohol give sufficiently accurate results, but only after removal of the ammonium salts, the concentration of which in nickel-baths may be considerable. The sample (100 c.c.) is first electrolysed to throw out nickel and decompose citric acid, which is nearly always present in such baths. The solution is then made strongly alkaline by addition of 10 grms. of caustic soda and evaporated in a nickel dish to decompose ammonium salts. The residue is made acid with hydrochloric acid and the solution diluted to 25 or 30 c.c., and heated to expel carbon dioxide.After allowing to cool, the solution is exactly neutralised to methyl orange, 100 C.C. of a 10 per cent. solution of mannitol is added, and the boric acid titrated in the usual manner with caustic alkali, using a considerable quantity of phenolphthalein as indicator. The results are from 0-5 to 1 per cent. high, calculated on the boric acid. G. C. J. Sources of Error and the Electrolytic Standardisation of the Con- ditions of the Iodide Method of Copper Analysis. A. W. Peters. (J. Amer. Chem. Soc., 1912, 34, 422-454.)-The various methods for the standardisation of sodium thiosulphate may be classified by groups, and the differences between the values found by the different groups far exceed the limits of error of the methods of the same group. Thus, for the estimation of copper, the thiosulphate must be standardised against a copper solution, and accurate results cannot be expected by employing a factor based on an iodine standard.The author has worked out a rapid electrolytic method for standardising a copper solution for this purpose. An alkaline electrolyte is employed, the reagent containing 35 grms. of Rochelle salt and 25 grms. of purified potassium hydroxide per 100 C.C. ; 10 C.C. of the copper solution are mixed278 ABSTRACTS OF CHEMICAL PAPERS with 10 to 20 C.C. of this reagent, and a quantity of a saturated solution of potassium cyanide, ranging from 0.4 to 1.0 C.C. is added. The cyanide controls the rate of deposition and the hardness of the deposit, but if too large a quantity be employed, the time required for complete precipitation is lengthened.Rapid deposition is obtained by using a high voltage, high current density, and keeping the temperature near the boiling-point. For preparing solutions of copper for titration by the iodide method, the copper is heated in an Erlenmeyer flask with the smallest necessary quantity of slightly diluted nitric acid. The solution is diluted, powdered talc is added, and the mixture is boiled vigorously for five to ten minutes. The rapid ebullition quickly removes nitrous acid, and the liquid is then cooled and diluted. Sulphuric acid may be added to the nitric acid solution when a solution of the sulphate is desired. The accuracy of the results obtained by the iodide method is influenced by varying concentrations of mineral acids, and by the presence of large amounts of salts.In the latter case, as, for instance, when the alkaline copper solu- tions from sugar analyses have to be titrated, sulphuric acid must be used instead of acetic acid in order to obtain good results. For instance, when the total volume of reduction-liquor was 60 c.c., consisting of 20 C.C. of Ahlin’s alkaline tartrate solution, 20 C.C. of Fehling’s copper sulphate, and 20 C.C. of water, the precipitated cuprous oxide could be filtered off and washed, giving a total volume of 70 to 75 C.C. of filtrate. The addition of 3.45 to 5.0 C.C. (best 4 c.c.) of concentrated sulphuric acid to this volume of liquid gave good results on titration, but smaller amounts were unsatis- factory.I n working with ordinary solutions of copper (i.e. where no alkali has to be neutralised), normal results are obtained with 1 C.C. of concentrated sulphuric acid in 60 of the final liquid, or with 1 C.C. of strong acetic acid in 6 of the solution, A further essential factor for success is the presence of a sufficiently large quantity of potassium iodide ; this is best employed in the form of a saturated solution containing about 1.1 grm. of the salt per 1 C.C. In titrations completed at a total volume of 60 c.c., 5 C.C. of the iodide solution are required ; for titrations at a final volume of 120 c.c., 6 to 10 C.C. of iodide are used. Lasbly, the question of the end-point requires careful attention, since it is not readily observed in presence of the cuprous iodide.So long as a drop of thiosulphate falling into the solution produces a white area, the end-point is not reached. I t is best to have a burette containing standard iodine always at hand, and to verify the end-point by titrating back again immediately. Provided the above precautions be taken, the results of the iodide method are quite unaffected by tha concentration of the copper. J. F. B. Relation of the Refractive Index of Soda Lime Glasses to their Chemical Composition, E. W. Tillotson, Jun. (J. Ind. Eng. Chem., 1912, 4, 246-249.)-A number of batches of glass of known chemical composition were pre- pared, and the density and refractive index of each was determined, and the specific refractivities calculated by the formula of Lorenz and Lorentz (Wied.Ann., 1880, 9, 641 ; 1882, 11, 70) On plotting the values of K thus obtained as abscissae in a system in which the ordinates represented the chemical Composition of the glasses, which rangedfrom Na,0.3 SiOz to CaO.SiO,, K was found to be a linear n2-1 1 x d=KINORGANIC ANALYSIS 279 function of the chemical composition, at any rate until the molecular ratio of lime to soda reaches 3 : 2. At this point there is a slight change in the direction of the curve, but in very few glasses does the ratio of lime to soda exceed this limit. Since the specific refractivity is linear, it may be computed by means of the following equation IX = %E1+g2+p3s etc., in which P,, P,, P,, etc., are the percentages of the 100 100 100’ oxides and, K,, K,, K,, etc., the empirically determined specific refractivities of the oxides.SiO, . . . 0.1220, CaO . . . 0.1210, Na,O . . . 0.1302. For soda lime glasses these are as follows : G. C. J. Use of the Mercury Cathode, particularly in the Electro-Analytical Separation of Metals. P. Baumann. (Zeitsch. anorg. Qhem., 1912, 74, 315-350.) The author has applied his apparatus (ANALYST, 1911, 36, 529) to the separation of mercury from bismuth, mercury from copper, and silver from bismuth, all of which separations may be mad.e rapidly at ordinary temperatures. For the reasonably rapid separation of copper from antimony, copper from cadmium, and bismuth from cadmium, a temperature of about 70’ C. is necessary. It is probable that cadmium could be separated from zinc by means of the mercury cathode; but the separation is so slow, and requires so much of the analyst’s attention, that the method could serve no useful purpose.One difficulty in the use of a mercury cathode is the need of considerable quantities of ether which shall be without action on mercury. Simple distillation of the ether does not get over the difficulty, as, within a few days, it re-acquires the faculty of giving rise to a black film on the surface of mercury with which it is brought in contact. If the ether be stored in a large distilling flask over caustic potash, and distilled in small quantities as wanted, the trouble is overcome, and the distillate remains for three or four weeks without action on mercury. G. C. J. Perchloric Acid in Electrochemical Analysis. W. S. Hendrixson. (J.Anzer. Chem. Soc., 1912, 34, 389-392.)-1n a recent paper by H. Golblum (Zeitsch. anal, Chem., 1911, 50, 741) on the electrolytic decomposition of nickel and cobalt per- chlorates, it was shown that perchloric acid is remarkably stable, and therefore suitable for general purposes in electrolytic work. The only reducing agents which decompose perchloric acid in solution appear to be freshly precipitated ferrous oxide, sodium thiosulphate, and, to a small extent, some oxides of nitrogen; in fact, in aqueous solutions perchloric acid exhibits only the properties due to its primary ions, and there are no side reactions due to the decomposition of its negative ion, Besides being stable, it is also a “ strong ” acid-i.e., highly dissociated, and metals may easily be converted into their perchlorates for electrolysis.Perchloric acid readily dissolves many oxides of the heavy metals, and in other cases the nitrates may be converted almost quantitatively into perchlorates by evaporation with an excess of perchloric acid. From potassium nitrate 94 per cent. of the nitric acid was expelled, and from copper nitrate 98 per cent. The author describes the electrolytic deposition of copper, silver, and cadmium from perchloric acid solutions, Pure280 ABSTRACTS OF CHEMICAL PAPERS copper oxide was dissolved in a small excess of perchloric acid, and subjected to electrolysis in a platinum dish with a rotating anode; the results were perfectly satisfactory. Silver was dissolved in nitric acid, and the resulting silver nitrate evaporated twice on the water-bath with water and perchloric acid, a trace only of nitric acid being left in the perchlorate.In the subsequent electrolysis silver oxide tended to collect on the anode, but was redissolved, especially on warming, long before the current was stopped. With cadmium also a satisfactory deposit was obtained from a solution of the oxide in perchloric acid. The acidity in all these experiments ranged from 0.03 to 0.4 normal, mostly round about 0.1 normal. J. F. B. Standardisation of Potassium Permanganate Solution by Sodium Oxalate. R. S. MeBride. ( J . Amer. Chem. SOC., 1912, 34, 393-416.)-0n behalf of the Bureau of Standards, the author has subjected the process of standardising permanganate solution by means of sodium oxalate to a complete revision, studying the effect of every conceivable variation and source of error separately-vie., temperature, acidity, volume of solution, rate of addition of the permanganate, access of air, and presence of added manganous sulphate.In connection with the question of the excess of permanganate present when the end-point is determined, the author has confirmed the accuracy of Bray's method, according to which the titrated solution is cooled, treated with potassium iodide, and the iodine liberated is titrated back with 0.02 N thiosulphate. He found that the residual oxidising power of the small excess of permanganate required to establish an end-point remains unchanged for at least half an hour, and consequently the lapse of five to ten minutes required for cooling the solution is quite negligible. Using this control, the author found that if the end-point be approached slowly in a solution at a temperature above 60" C., the depth of colour is proportional to the excess of permanganate present.The chief possible source of error is the continued existence of manganese in the higher stages of oxidation, resulting in an apparent loss of oxygen, due to low temperature, large bulk of the solution, too rapid addition of permanganate (especially at the start or just before the end-point), and insufficient stirring. On the other hand, the addition of manganous sulphate decreases the consumption of permanganate. The author finally recommends the following procedure : In a 400 C.C. beaker dissolve 0.25 to 0.3 grm. of sodium oxalate in 200 to 250 C.C.of hot water (80" to 90" C.), and add 10 C.C. of dilute (1 : 1) sulphuric acid. Titrate at once with & potassium permanganate solution, being very careful to stir the liquid vigorously and continuously. The permanganate must not be added more rapidly than at the rate of 10 to 15 C.C. per minute, and the last 1 to 0-5 C.C. must be added drop-wise, with particular care to rtllow each drop to be fully decolorised before the next is introduced. The excess of permanganate used to cause an end-point colour must be estimated by matching the colour in another beaker containing the same bulk of acid and hot water. The solution should not be below 60" C. by the time the end-point is reached; more rapid cooling may be prevented by standing the beaker on a hot plate during titration.The use of a small thermometer as a atirring-rod is recommended. The author employed a weight burette delivering a drop of 0.03 C.C. for his titrations; the influence of temperature changes due to the steam rising from the hot liquid is thereby avoided.INORGANIC ANA4LYSIS 281 The accuracy of the above method is claimed to be within 0.1 per cent.; probably the error does not exceed 0.05 per cent. J. F. B. Direct Determination of Small Amounts of Platinum in Ores and Bullion. F. P. Dewey. (J. Ind. Eng. Chem., 1912, 4, 257-258.)-The old method of parting by means of sulphuric acid, re-alloying with silver, and parting with nitric acid, is quite useless for the determination of small amounts of platinum. The second weighiug may possibly exceed the first, even when traces of platinum are present, and other members of the platinum group may go into solution in nitiic acid and be estimated as platinum.Small amounts of platinum (and gold) may be estimated accurately and conveniently as follows : The silver alloy first obtained in the regular course of assay is parted with nitric acid in the usual way, platinum, if present in small amount, passing into solution with the silver. The solution is evaporated, if strongly acid, and then diluted, and about 1 or 2 C.C. of strong hydrogen sulphide solution diluted to 15 to 30 C.C. is added. After three or four hours, or, better, after standing overnight, the precipitate of silver sulphide containing all the platinum metals, if these were present in small amount, is filtered off, dried, and ignited.The resulting metallic sponge is wrapped in lead-foil, cupelled, and parted by means of strong sulphuric acid. If the amount of fine metal is considerable, the actual platinum may be determined by the double chloride method. I n the examina- tion of fine silver for traces of platinum and gold, 100 grms. or more of the silver is dissolved in nitric acid, the bulk of the acid evaporated, and the gold and platinum, with some silver, precipitated in the diluted solution (250 c.c.) by 5 C.C. of strong hydrogen sulphide diluted to 50 C.C. The sulphide precipitate is filtered off, ignited, and cupelled, the lead parted with nitric acid, and the gold weighed. The platinum in solution is then precipitated with hydrogen sulphide in the manner already described, and parted from silver by means of sulphuric acid.In this way two purchases of fine silver were found to contain, in each 100 grms., 0.23 and 0.10 mgrm. of gold, and 0.55 and 0-14 mgrm. of platinum, respectively. In the case of material containing a considerable amount of platinum, the well- known fact that platinum alloyed with silver is not entirely soluble in nitric acid must be considered. In such a case the gold must be realloyed with silver, and reparted two or three times. The possibility of the application of the method to the estimation of noble metals in copper and similar materials is indicated, but no experiments in this direction are described. G. C. J. Estimation of Gold, Silver, and Platinum. Trenkner. (Metallurgie, 1912, 9, 103.)-The assay of platinum by cupellation with lead requires an exces- sive temperature, and the bead formed contains lead and is brittle.Gold and platinum may be estimated to 0.05 per cent. by the following method. For the preliminary assay 500 parts by weight (500 mgrms. = 1,000 parts by weight for alloys with more than 25 per cent. gold + platinum ; 1,000 mgrms. = 1,000 parts by weight for poorer alloys) are cupelled at 900Oto 950° C. with 3 to 15 grms. lead, according to the amount of base metal present ; from the more or less grey, flat, and rough appear-282 ABSTRACTS OF CHEMICAL PAPERS ance of the bead, the content of gold and platinum is judged, or is roughly determined by weighing the residue after dissolving in concentrated sulphuric acid, For the assay itself 500 parts by weight, with ten times as much silver (cf.Sharwood, J. SOC. Chem. Ind., 1904, 23, 412) as the platinum and gold found by the preliminary determina- tion, are cupelled with lead; the bead, less the silver added, gives the weight of gold, platinum, and silver. The amount of silver lost is found by cupelling a known similar alloy. The bead is dissolved in 25 to 30 minutes in 25 C.C. ol sulphuric acidof sp. gr. 1.84, which is heated, but not boiled ; decanted and washed three times with water ; the residue gently ignited and dissolved in a, minimum quantity of aqua regia, most of which is then driven off; the liquid is diluted with water, and the silver chloride removed by filtration; 15 C.C. of hydrochloric acid of sp.gr. 1.19 are added to the filtrate, the volume of which is about 150 c.c., and the gold precipitated with 1 grm. of hydrazine hydrochloride, allowed to stand with occasional stirring for one hour, filtered, ignited, and weighed. To remove any platinum it is fused on a cupel with a small quantity of lead and five to eight times its weight of silver, and the platinum and silver dissolved out by boiling with nitric acid. The platinum is precipitated from the gold filtrate by ammonia (Jannasch, Bey. 1903, 37, 1980) or with potassium hydroxide. 0. E. 11. Estimation of Sulphates in Solution by the Physico- Chemical Volumetric Method. A. Bruno and P. T. d'Auzay. (Comptes rend., 1912, 154,984-986.)- Dutoit and Duboux (ANALYST, 1908, 33, 90) proposed a method for the estima- tion of sulphates, notably in wines, by adding to the liquid baryta water, and measuring the electrical conductivity after each addition, plotting the values in the form of a curve.Theoretically, such a curve should show a minimum point corresponding to the total precipitation of the sulphates. On applying this method to a number of French wines, however, the authors found that the results were inaccurate in about three-quarters of the cases. They have therefore tested the method on aqueous solutions containing sulphuric acid, potassium sulphate, potassium bitartrate, tartaric acid, and malic acid, alone or in the form of mixtures. The results in the case of solutions containing sulphates were classified in the following manner : 1. The curve shows no minimum point.This is the case with a pure solution of neutral potassium sulphate, where the end of the precipitation is shown by the intersection of two ascending straight lines. 2. The curve shows a minimum point. This is the case with all the other solutions in which sulphates were present, and the results are subdivided as follows: (1) The minimum corre- sponds actually and exclu$ively to the total precipitation of the barium sulphate with (a) sulphuric acid alone; (b) sulphuric acid in presence of potassium bitartrate, in which case the curve consists of three straight lines, one of them descending. (2) The minimum point never corresponds to the end of the precipitation with (c) potassium bisulphate alone; (d) sulphuric acid in presence of tartaric or malic acid.I n the case of (c) the minimum point coincides with the precipitation of half of the total sulphuric acid. In the case of ( d ) the minimum point always occurs after the total precipitation of the sulphate, and the neutralisation of the organic acid only takes place after that of the sulphuric acid. (3) Lastly, the minimumINORGANIC ANALYSIS 283 point may coincide with the total precipitation in the case of potassium sulphate, neutral or acid, in presence of bitartrate or organic acids ; but the phenomenon is quite fortuitous and unreliable, depending on complex equilibria which may become established. Similar conclusions were arrived at in the case of dilute alcoholic solutions, and the proposed method, therefore, cannot be used for the estimation of sulphates in wines.This case corresponds to wines. J. F. B. Modification of the Method of Arnold and Hardy for the Rapid Estima- tion of Sulphur in Siderurgical Products. F. Giolitti and M. Marantonio. (Rassegna Mineraria, Metallurgica e Claimica, 1911, 35 ; through Chm. Zentralbl., 1912, 1, 1250.)-The method of Arnold and Hardy (Chem. News, 1888, 58, 41) consists in converting the sulphur into sulphuretted hydrogen, and observing how far the precipitation of lead sulphide progresses in a series of vessels containing lead acetate solution of known strength. The authors suggest (1) an improved form of vessel, more easily cleaned than the original one; and (2) the passage of the gas through a heated tube, to decompose organic sulphur compounds incapable of reaction with lead acetate, before it reaches the vessels. Sample analyses, with controls in which the sulphur is determined independently, show the utility of the method.0. E. M. Estimation of Sulphur in Insoluble Sulphides. T, S. Warunis. (Ber., 1912, 45, 869-870.)-The method previously described for the estimation of sulphur i n coal (ANALYST, 1911, 36, 375) has been extended, with slight modification, to the estimation of sulphur in insoluble sulphides. The substance (0.5 grrn.) is heated in a porcelain crucible with a mixture of 4 parts sodium carbonite and 3 parts copper oxide, at first at a low heat, but finally with the full flame of a Bunsen burner. The mixture is frequently stirred during the heating, which should be continued for two hours. The contents of the crucible, when cool, are boiled out successively with water, sodium carbonate solution, and again with water, and the solution is filtered and acidified with hydrochloric acid.I n absence of metals which form insoluble sulphates the contents of the crucible may be dissolved at once in hydrochloric acid. After separation of silica, i f present, by evaporation, the hydrochloric acid solution of the sulphate is precipitated hot with barium chloride, and the precipitate treated in the usual manner. An analysis of pyrites gave 53.16 per cent. sulphur against 53.04 by Fresenius’s method ; and an analysis of mercuric sulphide gave 13.75 per cent. sulphur, as compared with 13-65 per cent. obtained after oxidation by means of potassium chlorate and concentrated nitric acid.G. C. J. Rapid Determination of Vanadium in Steel. F. Garratt. ( J . I d . Eng. Chem., 1912,4,256-257.)-The following modification of Johnson’s method (“ Chemical Analysis of Special Steels,” p. 8) is found by the author to be preferable to any other for routine work. The steel (2 grms.) is dissolved in 50 C.C. of dilute sulphuric acid, and 5 C.C. of nitric acid is then added to oxidise the iron and any tungsten which may be present. The solution is boiled until nitrous fumes are no longer evolved, or, in the case of tungsten steels, until the separated tungstic acid is of a pure yellow284 ABSTRACTS OF CHEMICAL PAPERS colour. The solution is next diluted to about 150 c.c., filtered from tungstic acid, if this is present, and cooled to room temperature, I t is then further diluted to about 350 c.c., ferrous sulphate is added, and then dilute permanganate from a burette until a permanent pink tint is just reached, Potassium ferricyanide is added as indicator, and the vanadium titrated with standard ferrous ammonium sulphate.Blanks must be run on steels of similar composition, and consistent endapoints adhered to in titrations. The author agrees that for umpire work, and in any case where a sample concerning which nothing is known has to be analysed, a, method of more general application, such as that of Cain (ANALYST, 1911, 36, 475), must be used ; but he considers that in routine work these more exact methods give no more useful information, whilst they consume more time. G. C. J. Rapid Method for Determination of Vanadium in Steels, Ores, etc., based on its Quantitative Inclusion by the Phosphomolybdate Precipitate.J. Re Cain and J. C. Hostetter. ( J . Ind. Eng. Chem., 1912, 4, 250.)-For steels containing vanadium, chromium, nickel, titanium, manganese, molybdenum, singly or in combination, an amount of drillings estimated to contain from 2 to 10 mgrms. of vanadium is dissolved in nitric acid (sp. gr. 1.135), of which about 20 to 25 C.C. is taken for each grm. of drillings. The solution is boiled until free from fumes, oxidised with permanganate solution (14 per cent.), the manganese peroxide is dissolved by addition of sulphurous acid, and the solution again boiled until free from fumes. I t is then nearly neutralised with ammonia (1 : 2), and a solution of sodium phosphate is added in such amount that there is a t least ten times as much phosphorus as there is vanadium present.The solution is heated to boiling, and, after removal from the hot plate, ordinary molybdate reagent (0.05 to 0.06 grm. MOO, per c.c.) is added at the rate of at least 2 C.C. for each mgrm. of phosphorus present, and in no case less than 50 C.C. The mixture is agitated for a minute or SO, the supernatant liquid decanted through an asbestos filter, and the precipitate washed three times by decantation with a hot solution of acid ammonium sulphate, made by adding 15 C.C. of ammonia (sp. gr. 0.9) and 25 C.C. of strong sulphuric acid to 1,000 C.C. of water. The last wash liquor is decanted off as completely as possible from the precipitate in the flask, and the filter is sucked dry.The filter is now connected to a small dry pump flask, and the small amount of precipitate on it is dissolved in hot concentrated sulphuric acid, which is then sucked through the filter and transferred to the flask containing the bulk of the precipitate, and the pump flask is rinsed out once with sulphuric acid. The total amount of acid used should b0 from 5 to 8 C.C. for every 10 mgrms. of phosphorus present. The contents of the flask are heated until solution takes place, a few drops of nitric acid ( 1 : 25) are added, and the solution is heated until it fumes strongly. I t is then removed from the hot plate, and successive small additions of hydrogen peroxide solution are made, with vigorous shaking after each addition, until the solution assumes a deep brown colour, due to action on the molybdate.The flask is replaced on the hot plate and kept fuming for four or five minutes, when the brown colour gives place to a green or blue. The flask is covered and its contents cooled, and then poured into four times its volume of cold water and titrated at 70" to 80" C. against & permanganate.JNORGANIC ANALYSIS 285 The function of the hydrogen peroxide is to reduce vanadium to the tetravalent state. With steels containing tungsten, the only change necessary is to dissolve in aqua regia, dilute with hot water, filter off the tungstic acid, nearly neutralise with ammonia, and add 10 grms. of ammonium nitrate for every 100 C.C. of the final volume before precipitating as above described. The behaviour towards solvents of a phosphomolybdate precipitate containing vanadium is very different from that of pure ammonium phosphomolybdate, and the choice of an acid solution of ammonium sulphate as wash liquor is based on experiments which are described in the paper, and which have some bearing on the estimation of phosphorus in steels containing vanadium.The method can be extended to the estimation of as much as 30 mgrms. of vanadium, but the manipulation in such cases becomes inconvenient. From 2 to 10 mgrms. can be estimated with an error not exceeding 2 per cent., and the method permits of the concentration of the vanadium from 50 or 100 grms. of steel of low vanadium content, so that extremely small percentages can be determined with ease. A method is described whereby the vanadium may be separated as pentoxide in cases where there could be the slightest doubt of its being present at all.G. C. J. New Method for Determination of Vanadium in Steel. D. J. Demorest. (J. Ind. Erzg. Chtem., 1912, 4, 249-250.)-The method depends on the selective oxidation of ferrous sulphate in the presence of vanadyl sulphate by means of manganese dioxide. The vanadyl sulphate is then titrated by adding excess of per- manganate, the excess permanganate being determined by means of sodium arsenite. The manganese dioxide should be sufficiently fine to pass through a, 200-mesh sieve, and yet should settle in a beaker of water in thirty seconds (see next abstract). The borings (2 grms.) are dissolved in a mixture of 30 C.C. water and 12 C.C.sulphuric acid contained in a 500 C.C. flask. The iron is oxidised by addition of 1 C.C. of nitric acid, and the solution is boiled for a few minutes to remove nitrous fumes. I t is then diluted with 30 C.C. of water, and excess of strong permanganate solution is added, and the solution boiled to completely oxidise all carbon. Ferrous sulphate is next added to reduce manganese dioxide, permanganate, chromate, and vanadate, and the solution again boiled to expel any possible nitrous fumes. The solution is diluted to about 250 c.c., and, after addition of Tn permanganate until a pink colour persists, it is cooled. Ferrous sulphate is now added in amount at least sufficient to reduce all the chromic and vanadic acid, and any other reducible compound which may be present, but a large excess should be avoided.The solution is next shaken vigorously with about 1 grm. of manganese dioxide, and after two minutes a drop is tested with ferricyanide. The test is repeated at intervals of a minute, shaking all the while, until, usually in about five minutes, ferrous iron is found to be absent, when the shaking is continued for about another half minute, and the solution is then filtered through asbestos, using suction. The reduction of the iron is complete when the drop withdrawn does not instantly develop a blue colour when added to a drop of ferricyanide. A blue colour always develops in the course of a few seconds, due to the reduction of ferrioyanide to ferrocyanide by the vanadyl sulphate. To the286 ABSTRACTS OF CHEMICAL PAPERS filtered solution standard permanganate is added from a burette until a pink tinge persists; 1 C.C. more is then added, and the excess of permanganate is titrated with sodium arsenite solution. The standard solutions employed correspond to 0.001 grm. vanadium per C.C. A blank determination must be run on a vanadium-free steel, the blank being generally about 0.7 mgrrn. vanadium. Quantities of vanadium up to 7 mgrms. in 2 grms. of steel may be determined with an error not exceeding 0.1 mgrm., and the presence of ten times as much chromium is without influence on the results. G. C. J. New Method for Determination of Vanadium. J. R. Cain and D. J. Demorest. (J. Ind. Eng. Chem., 1912, 4, 256.)-J. R. Cain has found (J. Ind. Eng. Chem., 1911, 3, 476) that manganese dioxide oxidises vanadyl compounds as well as ferrous compounds. This appears to conflict with the differential oxidation of iron by manganese dioxide on which Demorest’s method of estimating vanadium (see preceding abstract) is based. Working together, the authors find that the difference in their results is solelydue to the difference in grain of the dioxide each used. Very finely divided manganese dioxide oxidises vanadyl compounds completely, but when only of the fineness indicated in Demorest’s paper, the velocity of this reaction is so small that its effect is inappreciable in the short time required for the quantitative oxidation of the iron. G. C. J.

ISSN:0003-2654    

DOI:10.1039/AN9123700276

出版商:RSC    

年代:1912

数据来源: RSC