摘要:

JUNE, 1905. Vol. XXX., No. 351 THE ANALYST. OBITUARIES. PROFESSOR TICHBORNE. PROFESSOR CHARLES ROBERT TICHBORNB, whoso death we briefly announced last month, was born in 1839 in Birmingham, where he received his early education. At the age of fourteen he was apprenticed as chemist at a chemical manufactory, where he stayed six years, and at the end of this time became a student under Hofmann at the Royal College of Science. In 1859 he was appointed head of the laboratories of the Irish Apothecaries’ Hall, which position he held up to the time of his death. Here he at once commenced a series of investigations, which led to many valuable suggestions being made as to the preparation of a number of pharmaceutical prepara- tions. He also did good work in analytical chemistry, and, amongst other things, was the inventor of a process for collecting and liquefying carbonic acid gas in breweries.He became Lecturer on Chemistry at the Carmichael School of Medicine in 1872 ; was one of the original members of the Pharmaceutical Society of Ireland, which commenced its career in 1875, succeeding Sir D. Corrigan, Bart., as president in 1878. In 1887 Professor Tichborne qualified as a medical practitioner, and in 1896 was appointed representative of the Irish Apothecaries’ Hall on the Medical Council, He was a Fellow of the Chemical Society, a Fellow of the Institute of Chemistry from its foundation, also an original member of the Society of Public Analysts, and held the appointment of Public Analyst for the county of Longford. Dr. Tichborne was a cultured, kindly, and good-hearted gentleman, whose memory will long survive in the minds of his colleagues, and especially in the hearts of those who were fortunate enough to be his students, for to them he was a most painstaking teacher .MR. WILLIAM ACKROYD. WITH regret we have to announce the death at Halifax, on May 9, of Mr. W. Ackroyd, at the age of fifty years, who had been Public Analyst for that borough for the last nineteen years, and was also teacher of Chemistry at the Halifax Technical School. Mr. Ackroyd was a Fellow of the Institute of Chemistry, also of the Chemical Society, and a member of the Society of Public Analysts. He was a keen student in chemistry and physics, and published a number of original papers in the Chemical Society’s Transactions, the British Association Reports, and the Chemical News .- ‘( Researches on Moorland Waters on the Origin of Combined Chlorine,” J. C. S. Trans., 1901, p. 673; papers relating to the circulation and182 THE ANALYST+ distribution of chlorides in natural waters ; to the distribution of the elements in relation to their atomic weights; and to the colour effects and the action of radium rays on the haloid salts of the alkali metals, etc.

ISSN:0003-2654    

DOI:10.1039/AN9053000181

出版商:RSC    

年代:1905

数据来源: RSC

摘要:

182 THE ANALYST+ THE VOLUMETRIC DETERMINATION OF REDUCING SUGARS. BY ARTHUR R. LING, F.I.C., AND THEODORE RENDLE. (Read at the Meeting, February 1, 1905.) THERE are but few subjects in analytical chemistry about which more has been written than the determination of sugars by the cupric reduction method. Every year a number of papers appear in the scientific journals describing certain fresh modifications in the carrying out of the process. The majority of these, however, add little to our knowledge ; whilst in not a few of the more recent contributions to the subject it would seem that the authors have either an insufficient acquaintance with the literature, or that they do not properly appreciate the degree of accuracy requisite in those branches of chemical industry in which sugar determinations play a prominent part.It is, therefore, not without some reluctance that we venture to enter once more upon this well-beaten track. I t is a fact worthy of note that, although numerous other alkaline copper solutions have been proposed from time to time for the determination of reducing sugars, that of Fehling has been found by the majority of workers to be the most satisfactory; and to this day the prescription of Fehling, with but few modifications, is adhered to by most chemists for the preparation of the alkaline copper solution. The methods of determination of sugars gravimetricslly by the cupric reduction method leave little to be desired on the wore of accuracy when applied to com- paratively pure sugars. If a paper filter be employed for collecting the cuprous oxide, it is not possible to remove the whole of the unreduced copper (from the excess of Fehling’s solution employed) from the filter-paper, and, unfortunately, the quantity of copper so retained is far from constant.Again, if a Soxhlet asbestos filter or a Gooch crucible be employed, it is exceedingly difficult to wash out the last traces of alkali from the precipitated cuprous oxide. In any case it is necessary in exact work to make a blank experiment with every set of determinations. Commercial sugars and malt worts, however, contain salts of alkali earths, nitrogenous organic matters, and other substances capable of vitiating the accuracy of the results. Certain alkali earth salts may, under the conditions of the experiment, be precipitated together with the cuprous oxide, whilst nitrogenous organic matters may dissolve a portion of the cuprous oxide.Errors of the first kind may be avoided by dissolving the moist cuprous oxide after it has been washed, and determining the copper in the solution volumetrically. One of the best methods of effecting this object is to dissolve the cuprous oxide in an acidified solution containing an excess of iron alum, and then titrate the resulting ferrous salt with permanganate. This Even here, however, there are possible sources of error.THE ANALYST. 183 and similar methods do not, however, eliminate the error due to the solvent action of nitrogenous organic matter. The determination of sugars by direct titration with Fehling’s solution is, on the other hand, but a rough-and-ready process, except when pure sugars are being dealt with. With commercial products giving dark-coloured solutions and containing nitrogenous organic matters, it is not possible to obtain accurate results, even when potassium ferrocyanide is used to indicate the point at which the precipitation of the cuprous oxide is complete.It was this fact which led F. W. Pavy several years ago to devise an ammoniacal copper solution for use with animal products. By means of this solution cupric reduction takes place without the precipitation of cuprous oxide, the latter being dissolved by the ammonia ; the final point is deter- mined by the disappearance of the blue colour. Those who have used Pavy’s solution know, however, its many disadvantages-the inconvenience of working with an ammoniacal solution, and the much greater dilution of the liquid as compared with the ordinary Fehling’s solution.Pavy’s method is, in fact, only applicable for the determination of amounts of sugars from, say, 1 to 10 per cent., the errors involved being too great to enable it to be used for accurately determining larger amounts of sugars. The titration of raw sugars, malt worts, and other commercial products with Fehling’s solution, employing potassium ferrocyanide as indicator, is, as already stated, anything but an accurate method. It is also a very tedious one; and when certain amino compounds are present in the solution being titrated, so much cuprous oxide may be dissolved that it is impossible to obtain an acidified filtrate which gives no colour with potassium ferrocyanide.This arises from the fact that, almost imme- diately a solution of cuprous oxide is acidified, it is oxidized to cupric salt, What, therefore, is needed to increase the accuracy, and at the same time expedite the method, is an indicator which responds to a minute trace of cupric salt, and can be employed without filtering off a portion of the assay liquid, which is necessary when potassium ferrocyanide is used. Several indicators which can be used in the manner just mentioned for determining the end-point in titrating with Fehling’s solution have been proposed, and among these may be mentioned that recently suggested by E. F. Harrison (Pharm. Jozwn., 1903, lxxi., 170), which consists of a solution of starch and potassium iodide.When a small quantity of this solution is acidified with acetic acid and brought in contact with a cupric salt, liberation of iodine occurs, and the well-known blue colour which iodine gives with starch is developed. The indicator we have devised, and which is certainly the most satisfactory of any we are acquainted with for titrating sugars with Fehling’s solution, is a solution of ferrous thiocyanate. When a drop of this indicator is placed on a slab, and it drop of a solution containing cupric salt brought in contact with it, oxidation of the ferrous salt occurs, with the immediate production of the well-known red colour of ferric thiocyanate. I n order that the indicator may possess the greatest possible delicacy, it appears to be necessary to follow certain conditions in preparing it, and our experience shows that the following prescription is the most satisfactory.One gram of ferrous ammonium sulphate and the same quantity of ammonium thiocyanate are dissolved181 THE ANALYST. in 10 C.C. of water at a moderate temperature-say between 4 5 O and 50" C.-and immediately cooled; 50 C.C. of concentrated hydrochloric acid are then added. The solution so obtained has invariably a brownish-red colour, due to the presence of ferric salt, which latter must therefore be reduced. For this purpose we have found zinc dust the most satisfactory reagent to employ, and as a rule a mere trace suffices to decolorize the solution, The indicator when kept for some hours develops the red coloration by atmospheric oxidation.It may, however, be decolorized by the addition of a further quantity of zinc dust, but its delicacy is impaired after it has been decolorized several times. Prepared according to these instructions, the limit of sensitiveness of the indicator is, at least, 2 C.C. of Fehling's solution diluted to 1 litre. It should be stated that less copper than is contained in 2 C.C. of Fehling's solution diluted to 1 litre can be detected by potassium ferrocyanide, not, however, in presence of the organic matters which occur in the Fehling's titration liquid ; and, again, the potassium ferrocyanide indicator requires the filtration of the titration liquid. It is usual in sugar titrations to dilute the Fehling's solution with an equal volume of water.When working with our new indicator, it is preferable, however, not to dilute the Fehling's solution, and all the experiments herein quoted have been carried out in this manner. This procedure has the obvious advantage, other things being equal, that the maximum sharpness of the end-point in the titration is insured. We prepare the Fehling's solution in the following inanner : Solution No. 1.-69.278 grams of crystallized copper sdphate are dissolved in water, and the solution made up to 1 litre. Solutioiz No. 2.-346 grams of crystallized Rochelle salt are dissolved in hot water, mixed with 142 grams of caustic soda, also dissolved in water, and after cooling made up to 1 litre. Equal volumes of these two solutions are accurately measured out at 15.5" C. and mixed in a dry flask.The mixture constitutes the solution from which measured quantities are taken for titration. One great advantage which the volumetric method possesses over the gravimetric is that the operator is independent of tables and results obtained under conditions not necessarily identical with those under which the analysis is being conducted. I n the volumetric method, each separate preparation of Fehling's solution may be standardized under the exact conditions employed in any given analysis. The method of titration is as follows : Freshly mixed Fehling's solution (10 c.c.)':' is accurately measured into a 200 C.C. boiling flask and raised to boiling. The sugar solution, which should be adjusted to such a strength that 20 to 30 C.C. of it are required to reduce 10 C.C.of Fehling's solution, is then run into the boiling liquid in small amounts, commencing with 5 C.C. After each addition of sugar solution the mixture is boiled, the liquid being kept rotated. About a dozen drops of the indicator are placed on a porcelain or opal glass slab, and when it is judged that the pre- cipitation of cuprous oxide is complete, a drop of the liquid is withdrawn by a clean glass rod or by a capillary tube, and brought in contact with a drop of the indicator on the slab. It is also essential to perform the titration as rapidly as possible, as an atmosphere of steam is then kept in the neck of the flask, and the influence of atmospheric oxygen avoided. At the final point the * In some cases it is advisable t o eniploy 20 C.C.of Fehling's solution in the titration. The mixture should be prepared daily. The test must be carried out rapidly.THE ANALYST. 185 liquid is boiled for about ten seconds. As in the ordinary volumetric method, the first titration may only give approximate results, and a second or third will then be necea- sary to establish the end-point accurately. However, when the operator has gained experience, thq first titration is as much to be relied on as succeeding ones, and this point is clearly brought out in the results we shall cite. One titration takes from two and a half to three minutes. The following are a few examples of results obtained with the method : Invert Sugar, Series A.-Pure sucrose (0.95 gram) was dissolved in water (150 c.c.), and boiled with f hydrochloric acid (30 c.c.), the mixture being maintained in ebullition for one minute, cooled, neutralized by the addition of f sodium hydroxide (30 c.c.), and made up with water to 500 C.C.This solution, which contained 0.2 gram of invert sugar per 100 c.c., was titrated with 10 C.C. portions of Fehling's solution prepared as above described, with the following results : Fehling's Solution. Invert Sugar Solution. Weight of Invert Sugar corresponding with 1 C.C. of Fehling's Solution. ... 1. 10 C.C. ... ... 45-3 C.C. ... 0.00506 gram. 2. 9 , ... ... 25.4 ,, ... 0.00508 ,, 3. 1 , ... ... 23.5 ,, ... ... 0*00510 ,, 4. 9 , ... ... 25.6 ,, ... ... 0.00512 ,, 5. 9 , ... ... 25.2 ,, ... .., 0.00504 ,, 6. 9 , ... ... 25.1 ,, ... ... 0.00502 ,, ... Mean of six titrations: 1 C.C.of the Fehling's solution corresponds with 0.005066 gram of invert sugar. I m ~ t Sugar, Series 23.-The invert sugar solution employed in these titrations was prepared in the same manner as that used in Series A ; another preparation of Fehling's solution was used. Fehling's Solution. Invert Sugar Solution. Weight of Invert Sugar corresponding with 1 C.C. of Fehling's Solution. 1. 10 C.C. ... ... 25.4 C.C. ... ... 0*00508 gram. 2. , I ... ... 25.6 ,, ... ... 0-00512 ,, 3. ' 9 ... ... 25.6 ,, ... 0.00512 ,, 4- ? I ... ... 25-4 ,, ... ... 0.00508 ,, 5. 1 , ... ... 25.4 ,, ... ... 0*00508 ,, ... Mean of five titrations: 1 C.C. of the Fehling's solution corresponds with 0.005096 gram of invert sugar. Dextyose.-The specimen of this sugar was prepared from invert sugar and repeatedly recrystallized from purified wood-spirit.':' As received, it contained a trace of spirit.An aqueous solution was prepared and evaporated, the residue taken up with water and again evaporated. This was repeated until the whole of the spirit was removed. The final solution had a specific gravity at 15.5'/15*5' of 1018-93. Applying the solution divisor of Brown, Morris and Millar, the concentration would therefore be 4.929 grams per 100 C.C. (fluid grams). The specific rotatory power was therefore [ u ] ~ 52-76', It read 5.20' in the polarimeter. * We are indebted to Mr. Lewis Eynon for this specimen of dextrose. Mr. Eynon determined the specific rotatory power of the sample, and found it to be [ a ] . 52-45" in a solution containing 7-88 grams in 100 C.C.186 THE ANALYST.A portion of 25 grams of this solution was carefully weighed out and diluted This 25 grams is equal to 24.535 fluid grams, and contained The dilute solution employed for the titrations contained, therefore, 0.2419 gram The Fehling's solution was the same as used in the invert sugar experiments with water to 500 C.C. 1.2093 grams of the dextrose. of dextrose per 100 C.C. (Series B). Fehling's Solution. Dex hose Solution. 1. 10 C.C. ... ... 20.4 C.C. 2. ?, ... ... 20.3 ,, 4- , I ... ... 20.4 ,, 6. , Y ... ... 20-5 ,, 7. Y , ... ... 20.3 ,, 8. 9 , ... ... 20.3 ,, 3- ?, ... ... 20.1 ,, 5- 9 , ... ... 20.1 ,, Weight of Dextrose corresponding with 1 C.C. of Fehling's Solution. ... ... 0.004934 gram. ... ... 0~004901 ,, ... ...0.004862 ,, ... ... 0.004934 ,, * . I ... 0.004862 ,, ... ... 0.004959 ,, ... ... 0.004901 ,, ... ... 0.004901 ,, Nean of eight titrations: 1 C.C. of the Fehling's solution corresponds with 0.004907 gram of dextrose. Maltose.-The specimen of this sugar was prepared by the action of diastase on potato starch paste. Although it was recrystallized repeatedly from alcohol, it appears not to have been quite pure. The alcohol adhering to the crystals was eliminated in the manner described under dextrose, and the specific gravity of the aqueous solution having been taken, its concentration was determined by the aid of a factor. The specific gravity of the solution at 15.5' compared with water at the same temperature was 1036.94, which corresponds with a concentration of c3.93 = 9.399.The solution gave a reading in the polarimeter of '26.4" in a 200 millimetre tube, whence [aID = 140.4'. This somewhat high result was confirmed by a second determination ; it points to the presence of a trace of one of the malto-dextrins in the maltose. I t is exceedingly difficult to purify maltose which has been prepared by the action of malt-diastase on starch paste, especially if the reaction has been allowed to proceed near its final point. Malto-dextrins are invariably formed which are soluble in alcohol, and to remove them from the maltose, by crystallization from alcohol, is by no means easy. A much simpler and more satisfactory method of preparing maltose is that devised by J. L. Baker, namely, by the action of ungerminated barley on starch paste.This author has shown that the dextrins formed in this way are practically insoluble in alcohol, and the crude maltose is readily separated from the small amount of dextrose it contains by crystallization from alcohol. Ten C.C. of the solution of maltose ( ~ ~ . ~ ~ = 9 . 3 9 9 ) were diluted with water to 250 c.c., which dilute solution, containing 0.3760 gram of anhydrous maltose per 100 c.c., was titrated with Fehling's solution. The Fehling's solution was that The specimen may also have contained a trace of dextrose. employed in the invert sugar experiment (Series A). Weight of Maltose corresponding t o 1 C.C. of Fehling's Solution. Fehling's Maltose Solution. Solution. ... 0-008121 1. 10 C.C. ... ... 21-6 C.C. ... 2. I ? ... ... 21.8 ,, ... ... 0*008196THE ANALYST.187 Mean of two titrations: 1 C.C. of the Fehling’s solution corresponds with 0-008158 gram of anhydrous maltose. As a result of considerable experience, we have found that under the conditions above described the cupric reducing power of invert sugar to that of anhydrous maltose stands in the ratio of 100 : 62. The maltose value of the Fehling’s solution calculated from the results obtained by titrating with invert sugar is, therefore: 0*005066 x 100 62---- -0.008170, a number agreeing fairly well with that found by directly titrating the Fehling’s solution with maltose. In the following examples of determinations with commercial products, 10 C.C. of Fehling’s solution were used in each instance. CANE MOLASSES, ONE PER CENT. SOLUTION. Molasses Solution.Sugar alculated as Invert Sugar in the Sample. 1. 23.6 C.C. ... ... ... ... ... 21.48 per cent. 2. 23.4 ,, ... ... ... ... ... 21.67 ,, CANE MOLASSES, ONE PER CENT. SOLUTION. 1. 24.9 C.C. ... ... ..- ... _.. 20.37 per cent. 2. 25.0 ,, ... ... ... ... ... 20.08 ,, 1. 33.2 C.C. ... ... ... ... 61-08 per cent. 2. 33.3 ,, . .- ... ... ... ... 60.88 ,, SAME MOLASSES AFTER INVERSION, 0.25 PER CENT. SOLUTION. ... RAW CANE SUGAR, 0.5 PER CENT. SOLUTION. ... ... ... 1. 27.4 C.C. ... ... 37.00 per cent. 2. 27.5 ,, ... ... ... ... ... 36.86 ,, 3. 27.3 ,, ... ... ... ... ... 37-14 ,, 4. 27.5 ,, ... ... ... . . ... 36.86 ,, SAMPLE OF STOUT. ORIGINAL GRAVITY 1075. 50 C.C. evaporated to remove alcohol, and made up io 200 C.C. Reducing Power expressed as Percentage of Maltose on Original Wort Solids.1. 30.8 C.C. ... ... ... ... ... 5.64 per cent. 2. 30.7 ,, ... ... ... 1,. ... 5.66 ,, 3. 30.6 ,, ... ... ... ... ... 5.68 ,, 4. 30.6 ,, ... ... 1.. ... ... 5.68 ,, 5. 30.7 ,, ... ... ... ... ... 5-66 ,, 6. 30-5 ,, ... ... ... ... ... 5.70 ,, Many other examples might have been given-e.g., determinations of reducing sugars in malt wort and in caramels. Just as concordant results can be obtained with these as with molasses and stout. The volumetric determination of reducing sugars by the method we have described requires some practice in order to insure satisfactory results, but, as the above examples show, the method is oapable of a high188 THE ANALYSTo degree of accuracy, and we believe it will be found superior in all respects to the gravimetric process.We are indebted to Mr. J. J. Belton for carrying out some of the experimental work contained in this paper. DISCUSSION. The CHAIRMAN (Mr. Blount), in inviting discussion, remarked that the fact that the titration could be carried out by artificial light was a very practical advantage. He should like to ask whether the authors had found any particular form of artificial light to be specially suitable. Dr. THORNE said that he had not yet tried this process, but, from what he had just heard, he certainly thought that it would be of great assistance to those interested in the estimation of sugar. Up to the present volumetric processes had not been satisfactory, and those who had to do with sugar estimations had been compelled reluctantly to fall back on the gravimetric method.He said reluctantly,” because, much as that method had been improved by the work of O’Sullivan, Brown and Iforris, and others, it still was it decidedly unsatisfactory method. So much depended upon minute details of manipulation, and upon the other constituents of more or less unknown character that might be present in commercial products. Mr. Ling had been kind enough to examine for him a sample of raw cane-sugar which had yielded somewhat uncertain results by the gravimetric method, and in that case the volumetric results came out very well, indicating that the discrepancies which occurred in some of the other volumetric processes had been-at any rate to a great extent-obviated. Mr. BAKER said that a short time ago Mr.Ling had outlined this process to him, and it had been tried in his laboratory with results the satisfactory character of which Mr. Dick, who had carried out the determinations, would be able to confirm. The new indicator had been tried side by side with the ordinary ferrocyanide indicator with great success. I t had a very suitable application in the determination of the diastatic power of malt, according to Mr. Ling’s modification of Lintner’s directions. Following that modified method, and using this indicator, it was possible to save, on a dozen determinations of diastatic power, at least ten minutes-a very important saving when the number of analyses was large, He had found the process to work very well by electric light. Mr. W. D. DICK said that he had used the authors’ process in Mr.Baker’s laboratory side by side with the filtration process for the last two months, and had found it to be considerably quicker and more satisfactory in every way. Mr. E. H. JEFFERS congratulated the authors on their historical sketch of the volumetric methods of estimating sugar. I n the process now described there were one or two points in manipulation which seemed to him rather risky. In the first place, the quantity of Fehling’s solution used-namely, 10 c.c.-seemed very small. I n working gravimetrically it was usual to use 50 c.c., and, even under conditions conducing to the greatest accuracy, it was almost impossible to get duplicate results to agree within less than a milligram, which on 50 C.C. of Fehling’s solution meant an experimental error of about a per cent.If smaller quantities were worked upon, the error was correspondingly multiplied. He quite agreed that the gravimetric method was not free from liability to error. In the first place, a blank experimentTHE ANALYST. 189 was necessary in every determination, but it was quite uncertain whether the result of the blank experiment could be applied as it stood. For instance, the filter-paper could never be washed perfectly free from the blue salt which it absorbed, and which had been shown to be an alkaline salt. He had found, however, that, while the filter-paper could be washed until the washings were perfectly free from alkali, the paper itself remained alkaline, which he thought must be due to the blue salt. I n making an estimation on 25 C.C.of a 1 per cent. sugar solution, all, or very nearly all, of the blue salt was decomposed, and consequently the quantity absorbed by the filter-paper was much less than in the blank experiment. With a view to getting over this difficulty, he had tried the method mentioned in the paper, of dissolving off the copper oxide with iron alum and titrating with permanganate, and had found that to give very accurate results. Mr. CHAPMAN said that the great difficulty in getting accurate results with Fehling’s solution was that of keeping the conditions as to dilution, proportion of mgar to copper, etc., perfectly constant. It seemed to him that this might be a matter of greater difficulty in a volumetric than in a gravimetric method, though, of course, it could be to a great extent got over by adjusting the strength of the solutions employed.He noticed that the relationship between the reducing power of invert sugar and that of dextrose appeared to be very different in the volumetric and in the ordinary gravimetric method. In the gravimetric method, with the usual dilution, it was about 97 for invert sugar (dextrose = loo), whereas according to the figures which the authors had given for the volumetric method it was distinctly less, so that the reducing conditions must be quite different in the two cases. He would like to ask whether, in the case of very dark molasses, the colour of the solution after filtering interfered in any way with the delicacy of the production of the red coloration with the indicator.Mr. E. R. BOLTON asked whether the authors had ever compared their process with another gravimetric process, which he thought was due to Soxhlet, in which the cuprous oxide was filtered in an asbestos filter-tube, washed, and reduced in hydrogen. That method in his hands had given very good results, but his experience with it was not large. Dr. SCHIDROWITZ said that he should like to hear again how the Fehling’s solution was standardized, as he had not quite followed this. He, in comnion with many others, invariably used the Soshlet method, filtering through asbestos. In that method all trouble with blank experiments was overcome. The CHAIRMAN said that he had occasion, in another branch of analytical work, frequently to measure small quantities of liquids with great accuracy, and he should never think of making those measurements by the simple observation of a pipette or any calibrated vessel-he weighed the liquid; and he would suggest that possibly some of the objections which had been urged against the authors’process might be based upon this fundamental source of error, which was common to all volumetric determinations. Mr.LING, in reply, said very good results could be obtained by working with ordinary incandescent gas light. The results given in the paper showed the accuracy of the volumetric process, and the use of 10 C.C. of Fehling’s solution instead of 50 C.C. as in the gravimetric method had no bearing on the relative accuracy of the two190 THE ANALYST. methods. He had had a large experience of the gravimetric method, and had hitherto used it for research work, because there were with pure solutions no disturbing factors, and concordance of results might safely be taken as indicating accuracy. Dr. DYER asked whether, in the case of the black molasses, a solution of the original material was worked upon, or whether it was first clarified with lead acetate. Mr. LING said that the original material was worked upon, without clarification, all reducing bodies being taken as sugar, which, strictly speaking, was not the case. He had for many years used the method referred to by Mr. Bolton. The Fehling’s solution was standardized, as described in the paper, by invert sugar or by any other pure reducing sugar, The objection attaching to the use of calibrated vessels applied, of course, to this as to other volumetric processes. One of the difficulties of this process was that very dilute solutions must be used, so that any errors were largely multiplied. The errors, however, could be reduced to a minimum by careful working, especially if the dilutions were made by weight.

ISSN:0003-2654    

DOI:10.1039/AN9053000182

出版商:RSC    

年代:1905

数据来源: RSC

摘要:

190 THE ANALYST. THE DETERMINATION OF HIGHER ALCOHOLS IN SPIRITS.-I. BY PHILIP SCHIDROWITZ PH.D. AND FREDERICK KAYE A.R.C.Sc. (Read at the Meeting April 5 1905.) Introductory. THE present experiments were undertaken with a view to examining the relative accuracy as far as possible of the various processes devised for the determination of the higher alcohols in spirits and not primarily with the object of putting forward any new method of our own. The necessity for some revision of this kind is so obvious to all chemists working in this branch of applied analytical chemistry that an apology for its publication is scarcely necessary; but we may say in explanation of the apparently scanty nature of the present communication that the difficulties of the work have exceeded our anticipations.The experiments were commenced some eighteen months ago but owing to unavoidable interruptions and to the difficulties encountered we are not yet in a position to publish the whole of the results obtained so far as many of these require further confirmation. We nevertheless think it desirable to make this preliminary communication now stating the general scope of the work the methods of purificat'ion of the substance8 employed and giving one short but we believe important series of results namely those concerning the Beckmann process, of which we in common with many others had hoped much but which unfortunately, was found to give entirely unsatisfactory results. As far as we are aware this is the first criticism of this process that has appeared and this €act-as it may save other workers much time and trouble-we venture to put forward in extenuation of what might otherwise be regarded as a somewhat premature paper.Scope of WoTk.-We had as our object the examination of such processes onl THE ANALYST. 191 as seemed apriori to afford a reasonable probability of success. We selected the Beckmann nitrite process (E. Beckmann Zeit. Unters. N a b . u. Genussm. ii. 709 and ditto iv. 1057) the Allen-Marquardt method (ANALYST l891) as modified by one of us (Schidrowitz Journ. SOC. Chem. Ind. June 1902) and finally the French colori-metric process (Girard et Cuniasse Manuel pratique de l’analyse des alcools et des spiritueux,” Parig 1899 Masson). We purposely refrained from further experiments with the German official process (Rose-Stutzer-Windisch Arbeit des Kaiserl.Gesund-heitsa. 1889 v. 391) inasmuch as one of us (Schidrowitz Zoc. cit.) had shown that for one class of spirit namely whisky this process was at any rate quite out of the question ; it gave in fact negative results. As it appeared likely that this might also apply to other spirits and as one of us had had private communications from distin-guished colleagues in several parts of the world confirming these observations we did not think it worth while to devote our time to the process. Puri3cation of Materials ; Ethylic AZcohoL-The (‘ pure absolute alcohol ” of commerce was subjected to fractional distillation in a flask provided with a Yohng’s ‘‘ rod and disc ” head until the resultant product showed no aldehyde reaction was entirely neutral and gave satisfactory blank tests with both the Allen-Marquardt and the colorimetric processes.Carbon Tetrachloride.-This was first shaken with water then boiled for several hours with chromic acid mixture subsequently washed first with water then with very dilute sodium bicarbonate and finally with distilled water until neutral. The product so obtained was then further purified by distillation. Amylic Alcohol.-Several methods of purification of the ‘( pure ” amylic alcohol of commerce (B.P. 128-132) were tried but the following was finally adopted as being most satisfactory. The alcohol was successively shaken with dilute sulphuric acid water warm milk of lime then filtered and allowed to stand overnight over solid quicklime distilled over the latter treated with freshly ignited sodium sulphate distilled over the same and finally fractionated by means of a Young’s ‘( rod and disc ” apparatus.By this means we were able to obtain a product which distilled entirely within half a degree namely at 129.5 to 130” C. Apparatus.-We found it a matter of very considerable difficulty to obtain an apparatus for the prolonged boiling in the Allen-Marquardt process of the carbon tetrachloride solution with the chromic acid mixture which was absolutely air-tight, and would permit of prolonged boiling without loss. Rubber cannot be used and ordinary corks were found very unsatisfactory. In the same way neither an ordinary (24-inch) Liebig nor a five-bulb condenser was capable of preventing all loss.Finally we were compelled to resort to an all-glass apparatus consisting of a 300 C.C. Jena glass Erlenmeyer flask very carefully ground in to a 24-inch Liebig condenser the central tube of the latter being fitted throughout its length with a Young’s ‘‘ rod and disc.” This effectually prevents the slightest loss and our experience is that an apparatus of this kind if properly made lasts for a very considerable time without cracking and is useful for many other purposes in the laboratory notably for the saponification of the ethers in spirits. Otherwise the apparatus used needs no special description 192 THE ANALYST. A. The Beckmann Process. This process consists substantially in a separation of the higher alcohols from the spirit (to which a certain amount of calcium chloride has been previously added) by means of carbon tetrachloride converting the higher alcohols so isolated into their nitrites by means of nitrous acid set free by the interaction of sodium nitrite and sodium bisulphate removing the excew of nitrous acid with sodium bicarbonate, decomposing the nitrites with sulphuric acid and finally titrating the nitrous acid so obtained with permanganate.Our first experiment was made (following Beckmann’s instructions minutely in every detail) with a blank solution of 48.7 per cent. alcohol (by volume). The final titration in this case required some 300 C.C. & permanganate corresponding roughly to 0.8 per cent. of amylic alcohol. This was obviously an impossible result but blank tests made with the materials (carbon tetrachloride ice sulphuric acid etc.) employed failed to yield any explanation.It became therefore necessary in the first place to ascertain whether the fault lay in the actual process of extraction or in the treatment of the carbon tetrachloride solution with nitrous acid etc. In order to test this point 150 C.C. of pure carbon tetrachloride were treated with sodium bisulphate and sodium nitrite as described by Beckmann were then filtered the residue being washed in accordance with the instructions the filtrate treated with bicarbonate water added until the excess of salt was dissolved and the whole of the carbon tetrachloride then divided into two portions one con-sisting of two-thirds of the whole the other of the remaining third. The former was finished off according to Beckmann and required 1.0 to 1.5 C.C.& per-manganate the end point not being sharply marked. The latter was washed with small quantities of water filtered through a dry filter the filter washed with a little tetrachloride and the filtrate then treated in the manner described. In this case 0.5 c.c.‘ permanganate was required. From these results it was quite clear that the process of nitration and subsequent removal of excess of nitrous acid, etc. is not at fault. A second blank experiment with ethylic alcohol only was made, with results practically identical with the first one. It was fairly obvious therefore, that the extraction was the weak point and that under the conditions employed by us a certain amount of ethylic alcohol was extracted by the tetrachloride.As we followed Beckmann’s instructions minutely we could only surmise that Beckmann in his experiments employed some peculiarly favourable alcoholic strength ; for in this respect his instructions are very vague saying as he does that the strength of the spirit to be extracted is not to be more than 50 per cent. but indicating no particular strength as one might have expected. We therefore attempted to modify the method of extraction finishing with the usual nitration. It struck us that in this connection the Allen-Marquardt method of extraction which requires an approximately (according to Allen and Chattaway Zoc. c i t . ) exact specific gravity of 1.1 for the solution of brine and alcohol from which the higher alcohols are extracted might prove useful and some experiments in this direction were therefore made ; these were as follows : 1.One hundred C.C. of spirit (strength as above) were diluted with brine until they showed a specific gravity of 1.1 (by the hydrometer) and were then extracte THE ANALYST. 193 according to Allen. The final tetrachloride solution was then treated as in the Beck-mann process. Forty C.C. permanganate were used corresponding in terms of amylic alcohol to 1.1 grams per litre. 2. Two hundred C.C. were extracted with tetrachloride according to the Allen method at 15" C. exactly and at an exact specific gravity of 1.10. After extraction the carbon tetrachloride was divided into two equal portions the one being treated according to Beckmann the other according to Allen.The first required 32 C.C. permanganate (=to roughly 0.7 gramme per litre) ; the second 2.8 C.C. & baryta of which a part was ascribable to mineral acid. If, however in this particular case the whole of the 2.8 C.C. baryta employed were calculated in terms of amylic alcohol it would amount to 0.24 gramme per litre or about one-third of the blank found in the corresponding Beckmann experiment. As stated however a part of the Allen blank was due to mineral acid. In this case the Beckmann portion required 31.5 C.C. permanganate (=to roughly 0.7 gram per litre); the Allen portion required 2.8 C.C. & baryta of which one half was due to mineral acid. The blank in this case was therefore 0.12 gram per litre. 4. As Experiment 3 except that the specific gravity of the spirit extracted was purposely raised to 1.1235.In this case the Beckmann portion required 25 C.C. permanganate ( = t o roughly 0.4 gram per litre) ; the Allen portion as before an amount of baryta equivalent to roughly 0.1 gram per litre. I n this case the Beckmann portion required 35 C.C. permanganate ( =to roughly 0.95 gram per litre) ; the Allen portion required (excluding the mineral acid) 3.8 C.C. baryta-that is an amount equivalent to roughly 0.34 gram per litre. The latter experiment indicates that in the Allen-Marquardt process it is desirable not to allow the specific gravity of the alcoholic brine to fall below 1.10. From these experiments we conclude : 1. That the Beckmann process as published is unworkable. 2. That the fault lies not in the latter (nitration) part of the process but in the actual extraction.3. That the Allen-Marquardt process of extraction undoubtedly does ultimately leave a certain quantity of ethylic alcohol in the carbon tetrachloride extract but that for some reason unexplained the greater part of this disappears during the oxidation with chromic acid mixture but is not converted into acetic acid. For this curious phenomenon which may perhaps account for the somewhat anomalous results occasionally obtained by this process we suggest the following in explanation : (a) The alcohol is split up into carbonic acid and water; or ( b ) the alcohol forms by the action of mineral acid present or produced an organic compound such as ethyl chloride for example which does not possess acid properties.Whether this curious action affects the higher alcohols also we are not yet in a position to say but we hope in our second paper to be able to throw some light on this point in the course of our consideration of the Allen-Marquardt process as a whole. We are still inclined to believe that if carefully worked this process is still the most reliable of all those published I t appears to be very little known in Borne Continental quarters particularly where the Rose process is most in favour and in 3. Repeated Experiment 2. 5. As before but specific gravity of the fluid extracted was 1.0938 194 THE ANALYST. this connection we think it necessary to draw attention to some recent remarks by Karl Windisch on the subject. I n a paper on Brandy contributed to the Zed.Unte~s. Nahr. u. Genussm. (No. 8 October 15 1904) he says (p. 488) '' Ph. Schidro-witz examined twelve commercial whiskies and determined the fusel-oil by the Allen-Marquardt method I n this process the fusel-oil is shaken out with chZoroform,* etc. . . . This process gives r e s d t s which are much too low " (cf. Karl Windisch, Arbeit des Kaiserl. Gesundheitsa. 1889 v. 373). Now this statement clearly indicates that Windisch is not familiar with the Allen modification of the Marquardt process, for one of the improvements due to Allen was the substitution of carbon tetra-chloride for chloroform. I t is also somewhat unfortunate for the contention of Windisch that the work to which he refers as having disproved the value of the Allen-Marquardt process was published some two years before the Allen-Marquardt process saw the light.His criticism therefore of the work of one of us in this respect is to say the least somewhat premature. DISCUSSION. The PRESIDENT (Mr. Bevan) in inviting discussion said that it seemed obviously very important not only that the alcoholic strength should be well maintained but also that the strength of the brine solution should be kept as constant as possible. It would probably be satisfactory if a rule could be laid down for this and he hoped that the authors might find it possible to do so in the second contribution to the knowledge of this subject which he understood they would make later on. Mr. DIBDIN inquired whether the authors had found the temperature to vary as the strength of the solution varied and if so whether they had made any attempt to keep it constant.Dr. SCHIDROWITZ said that they had all through tried to keep the temperature of the various solutions during shaking out at as nearly as possible 15" C. Dr. J. T. HEWITT said that the authors seemed to have experimented with every available process for the determination of higher alcohols and also to have judged the results in such a way that they knew pretty well what was to be expected from any given mode of procedure. For instance in the official German method-the Riise-Hertzeld method-which con-sisted in shaking out and merely measuring an increase in volume it was difficult to see how any exact idea could be obtained of what it was that was being measured in any particular case.With regard to the reflux apparatus used by the authors he had recently had the privilege of seeing this in use in Dr. Schidrowitz's laboratory and there could be no doubt that it worked most admirably. The efficiency of condensa-tion due to the use of the rod and disc column was surprising. It was interesting to know that a combination of the Beckmann and Allen-Marquardt methods gave good results for both these methods seemed to be based on fairly sound principles. He should have thought that the colorimetric method would have proved more satis-factory than appeared to have been the case. Dr. Schidrowitz however he believed, considered that there were a good many other substances that might affect the coloration obtained with strong sulphuric acid.He (Dr. Hewitt) had been at one time inclined to think that not merely the higher alcohols but also the higher acids, might have some influence and he had tried removing the ethers in addition to other * The italics aIe ours. That could not be said in every case THE ANALYST. 195 by-products by prolonged boiling with caustic potash and then determining the higher alcohols colorimetrically; and he had been surprised to find that the result was not very different from that obtained when merely the aldehydes and furfural were removed before distillation. Dr. Schidrowitz he thought held the view that the other substances in question were not of the nature of 'ethers or acids but were of an entirely different type ; and he (Dr. Hewitt) was now inclined to agree in this.Mr. HEEBEBT E. BURGESS thought that it would be of much assistance to other workers if full details were given of the apparatus which the authors had used and particularly of the still-head. The still-head of Dr. Young was of course familiar but he had never seen one with a condenser outside it. He could confirm the authors' views as to rubber connections. He had had considerable experience in distillation and had found that where rubber could not be used a perfectly tight joint could be made with a wooden stopper painted round with ordinary mucilage and covered with a piece of paper as recently described by Mr. Page in the Chemical News. He had been wondering whether a satisfactory estimation of the higher alcohols could be made on the lines of the method described by Dr.Thorpe for the estimation of ethers in alcoholic solution, namely by distilling a definite quantity with a still-head of the Young-Thomas type, shaking out with petroleum ether acetylating and saponifying. The PRESIDENT said that he had rather expected some mention to be made in the paper of a new apparatus made by Messrs. Priem for the purpose of getting over the difficulty due to corks. The flask was made with a glass trough round the outside of the neck and a mercury seal was formed by slipping over the neck of the flask another tube which dipped into mercury contained in the trough. Its only dis-advantage was its liability to break and with it view of obviating this he (the President) had suggested that the mercury should be retained by means of an india-rubber stopper fixed on the neck of the flask between it and the outer tube which fitted closely over the outside of the stopper.This modification was simple to make, and had been found to work very satisfactorily. Mr. CHAPMAN said thab owing to the difficulties which had been experienced in connection with the estimation of higher alcohols in spirituous liquids and to the large differences which had in many instances been recorded between the results obtained by different chemists employing different methods it was highly important that some definite information should be obtained on this subject and to this end it was suggested to Dr. Schidrowitz and his colleague that they should direct their attention to the matter under the provisions of the Analytical Investigation Scheme which the Council of the Society had recently initiated.These gentlemen had taken the matter up very heartily and he (Mr. Chapman) was sure that all would agree with him that the investigation could not have been placed in better hands. He thought that this was the first occasion on which the various methods for the estima-tion of the higher alcohols had been submitted to a critical and comparative study, and in view of the smallness of the quantities that had to be estimated and of the very great chemical difficulties in the way it would easily be realized how trouble-some the investigation must have been. The difficulty of obtaining perfectly pure ethylic alcohol and of the preparation of solutions containing known quantities of the higher alcohols mustj at the outset have been very considerable.He (Mr. Chapman) had personally had experience of three methods for the estimation of the higher This was different from any he had seen before 196 THE ANALYST. alcohols in spirituous liquids. The earliest was the physical method of Traube in which an instrument known as a stalagmometer was employed. This method was only capable of being used in certain cases and possessed a big inherent error and he was under the impression that it was now rarely if ever employed. In addition to this he had worked with Rose’s process which so far as his experience went, was useless and also with that of Marquardt as modified by Allen. This last-mentioned method he had used for some years and his experience had been on the whole that of the authors of the present paper-viz.that it was a useful method, and one which if properly carried out was capable of giving good results. A great advantage which this method possessed moreover was that the actual oxidation products of the higher alcohols were obtained and could be further used for purposes of examination or identification. Next to the specific gravity of the brine liquid he thought the factor which had the greatest effect on the results was the washing of the carbon tetrachloride extract. I n his own experience he had occasionally found that abnormal and impossible results were obtained by this method even when the greatest care was employed in its execution and he would like to ask the authors whether they had ever had a similar experience.In the cases to which he referred the organic acids produced would require perhaps three or four times as much standard alkali for neutralization as they ought to and he had usually observed at the same time that the mineral acid instead of requiring the customary 1 C.C. or 2 c.c. would require perhaps 6 c.c. or even more of alkali for neutralization. He would also like to ask the authors whether they had yet made any experiments with the colorimetric method which was so largely employed in France. Dr. SCHIDROWITZ in reply said that they had a number of figures which were not included in this paper because they considered that they required amplification. The present paper was merely a preliminary communication recording the revision and control of the Beckmann method and the establishment of certain facts with regard to the Allen-Marquardt method which would be productive of one or two slight changes in the apparatus at present in use and so forth.Their further results would be published later on and would include observations on the colorimetric method with which they had made a considerable number of experiments already. With regard to the colorimetric method although it gave fairly good results with pure amylic alcohol or iso-butylic alcohol it obviously could not give a satisfactory result in a spirit containing a mixture of various alcohols. I t was known for instance that while iso-butylic alcohol gave a very strong reaction with sulphuric acid normal propylic alcohol gave practically no reaction and the amylic alcohol of fermentation gave a coloration which was not as strong as that given by iso-butylic alcohol.In a fermented liquid like whisky the higher alcohols consisted according to Bell-whose figures were the only ones available-of a mixture of about 35 per cent. of butylic alcohol 35 per cent. of amylic alcohol and 30 per cent. of propylic alcohol. The proportions however varied in every spirit and the results obtained by acting on such a mixture with sulphuric acid and referring the colour to a standard for one particular alcohol must necessarily be misleading. They had lately made a good many parallel estimations by the Allen-Marqusrdt method and by the colori-metric method and had worked out a new curve for amylic alcohol. The results were certainly interesting.He did not however think that they indicated higher alcohols pure and simple but probably a variety of substances; and he agreed wit THE ANALYST. 197 Dr. Hewitt that these probably included terpenes and substances which were not removed by saponification with potash. They had made the experiment which Dr. Hewitt mentioned of boiling for a long time with potash and then comparing the colorimetric results with those obtained after removing merely the aldehydes and furfural and they had found a difference but scarcely a radical one. The idea of putting the disc and rod column inside the condenser was their own. He thought that Dr. Thorpe’s process for estimating esters would not be suitable in this case. It was only intended to apply to spirits or liquids containing relatively much larger quantities of those bodies.A process on the same lines was used for liqueurs but they had tried it with whisky and brandy and had found it impracticable because the quantities were so small. With regard to cork certain substances were extracted from it by carbon tetrachloride and ether-a fatal objection to its use when the con-stituents to be estimated were so very minute in quantity. Ground-glass connections seemed by far the simplest and if properly made seldom cracked. They had had flasks with such connections in constant use for over a year, The PRESIDENT inquired what was used to make the joint tight. Dr. SCHIDROWITZ said that in the Allen-Marquardt method they used sulphuric acid but as a matter of fact a well-ground connection did not need anything at all. Continuing he said that they had found that abnormal results such as Mr. Chapman mentioned were occasionally obtained the mineral acid in the blank determination being in such cases also too high but that in the course of a very large number of determinations recently made they had not had trouble in this direction. They could not account for this. The washing of the carbon tetrachloride solution was always carried out in the same way and he thought it must be due to some other cause-possibly to decomposition. The mercury seal which the President had mentioned was originally due to Dr. Wiley of Washington who however used a differently-shaped flask which though suitable for fat extraction did not readily permit of the residue being removed from the flask for quantitative purposes They had not found them liable to crack

ISSN:0003-2654    

DOI:10.1039/AN9053000190

出版商:RSC    

年代:1905

数据来源: RSC

摘要:

THE ANALYST. 197 THE ANALYSIS OF SAMPLES OF MILK REFERRED TO THE GOVERN-MENT LABORATORY IN CONNECTION WITH THE SALE OF FOOD AND DRUGS ACT. BY DR. THOMAS E. THORPE C.B. F.R.S. (Reprodwed in somewhat abridged form from the Journal of the Chenzical Society 1905 lxxxvii. 206.) THE samples of milk submitted to the Government Laboratory under the provisions of the Acts are invariably more or less sour when received and hence it is of im-portance to determine whether this fact in any way prevents a true inference aB to the character of the fresh milk or interferes with the determination of the degree of sophistication to which the milk may have been subjected. It would appear that bacteria which produce steatolytic enzymes do not develop in sour milk to any considerable extent since the proportion of fat suffers little if any alteration during the souring of the milk.Thus of thirteen samples of genuin 198 THE ANALYST. milk the fat in which was determined when fresh and after keeping for periods varying from two to fourteen and a half weeks the largest decrease in fat percentage observed was 0.15 (after thirteen weeks) while the average decrease was only 0.06 per cent. The presence of added water does not affect these results which show that if the fat in milk suffers any alteration in amount as the milk becomes sour it is too inconsiderable to affect to any substantial extent the experimental proof of the validity of any charge based on an alleged deficiency of fat. The proportion of non-fatty solids in milk is however undoubtedly altered to a greater or less extent by the fermentative changes associated with its souring.As is well known the principal constituent of milk suffering such change in the process of souring is the lactose. Lactic acid is produced at an early stage the rapidity of its formation depending within limits upon the temperature. This reaction is pro-duced by micro-organisms ol which there are over 100 kinds capable of effecting the change. C12H22011,H20 = CGH1206 + CGHI206 and CGHiZOG = 2CH,.CH(OH).COaH I t should be noted that in the usual way of expressing the reaction-(dextrose) (galactose) (galactose) no loss of weight is involved. Moreover the lactose is never wholly transformed under ordinary conditions into lactic acid and dextrose generally more than half remaining unaffected after a lapse of about eight weeks although in some exceptional cases about 60 per cent.has suffered conversion. The activity of the organisms concerned in this change appears to be inhibited when the lactic acid reaches a certain limit. If this were the only change taking place in the souring of milk the neutraliza-tion of the lactic acid by a known weight of strontia and the determination of the total weight of milk solids less the corresponding weight of strontia added would give the equivalent of the solids in the fresh milk. As a fact however other changes actually do take place and these involve a slight loss of weight. Thus a series of fifteen experiments showed that the loss in non-fatty solids during two to fourteen and a half weeks ranged from 0.24 to 0.87 per cent.for genuine milks while in the case of watered milks the loss amounted to 0.23 to 0.68 per cent. Similar results were obtained by the examination of separated and watered separated milks the loss in the latter cases amounting after a lapse of fifty-five and fifty-seven weeks to about 1-45 per cent. of non-fatty solids. Thus it is clear that concurrently with the formation of lactic acid there are produced substances which are either gaseous at ordinary temperatures or are volatilized during the operation of determining the solids of the milk. Although not large in amount this loss of weight is sufficient to affect any estimation of the degree of sophistication to which the milk may have been subjected. Butyric acid is a product of the early stages of the souring of milk although the proportion in which it may be formed is probably very small.It is certain that milk in which the butyric stage of fermentation has become very pronounced is of relatively infrequent occurrence in the samples received for analysis at the Government Labora-tory. Acetic acid is however almost invariably present in these samples but the proportion rarely exceeds 0.2 per cent. at the end of one month although occasional samples show almost double this amount. At the end of two months the proportion of volatile acids calculated as acetic acid may exceed 0.5 per cent. and in some cases the presence of butyric acid now becomes marked at this period. The precis THE ANALYST. 199 mechanism of the production of the volatile acids in sour milk is not very clear.Butyric acid is generally assumed to be derived from lactic acid as shown in the equation :-while acetic acid may arise either from the oxidation of lactic acid with the simul-taneous evolution of carbon dioxide or from the oxidation of ethyl alcohol which is also an almost invariable constituent of sour milk. For each molecule of acetic acid produced there is formed in the one case a molecule each of carbon dioxide and of water and in the other case only a molecuke of water. I n this latter instance, however a molecule of carbon dioxide has already been lost in the production of the alcohol so that the resultant loss is the Rame in either case. I n the determination of the total solids of the sour milk any acetic and butyric acids together with any dissolved carbonic acid are neutralized by the strontia used prior to evaporation and are weighed with the non-fatty solids.The carbon dioxide which has escaped from the soured milk before neutralization and the water and free hydrogen formed indirectly from the lactose are however not weighed. Ethyl alcohol is almost always formed in quantities which in the above series of experiments ranged from 0.09 to 0.35 per cent. after eight weeks during the fermentative changes which occur in milk under ordinary conditions. Although the conversion of lactose into alcohol is not brought about by ordinary yeast yet certain of the Schixomycetes species effect the modification by the production first of dextrose and then of alcohol with elimination of carbon dioxide.If however sufficient food for the organisms is present galactose is also completely fermentable so that the reaction in the case of lactose in milk may be expressed thus : 2CH,+CH(OH).CO2H=CH,(CH~)2.COzH + 2COz + 2Hz ; Again the yeast ferments occasionally present in the market milk of large towns may also directly attack the dextrose produced in the sourse of the change from lactose t o lactic acid. I n any case the production of one molecule of alcohol corresponds to the elimination of one molecule of carbon dioxide so that the decrease in weight in the lactose due to the production of the volatile alcohol and carbon dioxide is ::- times the weight of the alcohol produced. Besides ethyl alcohol there may also be found very small amounts of higher alcohols glycerol and succinic acid in a manner analogous to the ordinary alcoholic fermentation.The amounts of these products can only be very small and are negligible. Moreover the higher alcohols produced would be taken account of in the analysis being added to the ethyl alcohol and any succinic acid would be neutralized and weighed with the solids of the milk. The possible changes in the proteids of milk during fermentation are profound, but do not involve any considerable loss of weight. The ordinary “curdling” of milk in the early stages is simply a precipitation of the casein due to the interaction of the lactic or acetic acid with the calcium proteid compound. Proteolytic changes, consequent on the development of enzyme-producing bacteria result in the forma-tion of a certain proportion of proteoses and peptones and later of amino-compounds such as leucine lysine and tyrosine; while small quantities of ammonia and amine bases may also be produced The whole of these products however with th 200 THE ANALYST* possible exceptions of traces of ammonia and amines are weighed with the unchanged casein and other non-fatty solids in the analysis.On the one hand there may be a gain in weight due to the absorption of the elements of water in the hydrolysis of the casein molecule and on the other hand there may be a loss due to the ultimate conversion of some of the casein into carbon dioxide ammonia and water. Experiments in this direction show that as regards the effect upon the weight of the solids the net result of these changes is usually very small.The total loss of solid matter from all causes whatever ranges as a rule only from 0.2 to 0-5 per cent., and nearly the whole of this is accounted for by the production of alcohol and volatile acid from the lactose. Thus in the case of a series of watered and unwatered inilks the deficiency in solids after allowing for that accounted for by the alcohol and volatile acid present ranged from 0.00 to 0.07 per cent. after six weeks' keeping. During eight weeks a series of experiments showed a production of ammonia and amines of 0.001 to 0.012 per cent. equal to a loss of at most 0.006 to 0-075 per cent. of proteids. An investigation with about 20 gallons of milk was carried out by the author and his collaborators with a view of obtaining further knowledge as to the nature and amount of the products formed during the souring of milk under the usual conditions.The milk was diluted with one-fourth of its volume of water placed in stoppered bottles and kept at a mean temperature of 18" C. during a period of ten to twelve weeks At the end of that time the acid liquid was distilled the distillate made alkaline and re-distilled. The distillation was again repeated the liquid being rendered alternately acid and alkaline and concentrated by means of a fractionating column. After further rectification and distillation 76.4 grams of practically pure ethyl alcohol were recovered and identified by the usual methods. The solution of the sodium salts of the volatile acids was examined and found to contain a large quantity (about 390 grams) of normal butyric acid.Formic acid was absent but acetic acid was identified (116 grams) and traces of propionic acid were probably also present. Less than 2 grams of an acid of higher molecular weight than butyric acid (possibly a nonoic acid) were obtained and homologues of this acid were probably also present. A reserved portion of the fermented milk freed from proteids and fat gave on distillation with magnesia a small quantity of volatile bases which indubitably consisted very largely of ammonia. Hence therefore it may be con-sidered proved that by far the greatest portion of the volatile products of the fermentation of milk other than water and carbon dioxide are ethyl alcohol acetic and butyric acids and a small quantity of ammonia.A small proportion of higher acids, possibly traces of propionic acid and of volatile organic bases may also be produced. The amount of these by-products in comparison with the main quantities is so small that they are of no practical significance in this connection. The great majority of samples dealt with by the referees under the Sale of Food and Drugs Acts are only fermented to a relatively small extent being between three and six weeks old and having undergone a loss of weight in non-fatty solids of from 0.2 to 0.5 per cent. Moreover it should be borne in mind that in ordinary careful determinations of the non-fatty solids in the same sample of fresh milk by different persons the results may frequently differ by 0.1 to 0.2 per cent. Thus it may be assumed that a determination of this figure to within these limits ia a satisfactor THE ANALYST.201 result especially since an error of 0.1 per cent. only corresponds to about 1 per cent. in calculating the amount of added water in a sample. The following methods of analysis are employed in the Government Laboratory. The “ maceration ” process is used and the weight of the non-fatty solids and fat is independently ascertained in duplicate experiments while as a control a direct determination of the total solids is made on a third portion of the milk. ‘The contents of the sample bottle are transferred to a suitable vessel and thoroughly mixed with a wire whisk. Portions of the sample about 10 grams in each case are weighed out into flat-bottomed platinum capsules each of which has been tared along with a short glass rod having a flattened end.All the weighings and measurements are independerhly checked by two analysts. The weighed quantities are next neutralized with cn solution of strontia using phenolphthalein as indicator and the measure of strontia used noted. The milk is then evaporated over the water-bath until the residue which towards the end should be dried at a very gentle heat and with constant stirring attains the consistency of dry cheese. About 20 C.C. of dehydrated ether are next poured over the milk solids which are then carefully triturated with the glass rod. The ethereal solution of the fat is passed through a filter which has previously been dried and weighed in a weighing-bottle, and the maceration of the milk is continued with eight successive quantities of ether.At the conclusion of the process the non-fatty solids should be in a fine state of division resembling the precipitated chalk of pharmacy. Before becoming quite dry the solids are transferred as far as practicable to the weighing-bottle; the filter-paper washed free from fat is replaced in the bottle and the whole with the platinum capsule containing the small adherent quantity of solids is dried at 100” C. for three hours and then weighed. The weight is again taken after drying for a further two hours and a final confirmatory weighing after another hour. The last two weights should not differ by more than a mgm. Deducting 0.00428 gram for each 1 C.C. of decinormal strontia used in the neutralization the result gives the amount of non-fatty solids actually present in the quantity of milk taken €or the analysis.The ethereal solution of the milk-fat is received in small tared flasks and after distillation of the ether the residue of dried fat is weighed. From the judicial standpoint it is obviously desirable that the quantity of any constituent on which a legal charge may be based should be determined by direct weighing rather than by difference. The ( 6 maceration ” process is the only one applicable to sour milks which leaves the residual non-fatty solids in a convenient form for accurate weighing. The trustworthiness of the determination of fat by this method has been completely established in the Government Laboratory. When conducted as described above practically the whole of the fat is extracted and is obtained in a form admitting of accurate determination.In the two following series of the results of comparison of the Bell maceration method with other processes the first series was obtained by two analysts working independently and are typical of forty analyses made to compare the (‘ Werner-Schmid ” and “ maceration ” methods ; while the second series shows the results of H. D. Richmond’s analyses of fresh milk by the ‘( Adams’ coil ” and the (( maceration ” method respectively 202 THE ANALYST. Percentages of Fat. Werner-Schmid . 2.59 2-88 1-91 2.33 1-71 2-73 2.52 2-77 2.86 Maceration (Bell) 2.59 2.78 1.91 2-15 1-64 2-74 2.53 2.76 2-94 Werner-Schmid . 2.02 2-48 3.73 2-73 3.14 8.78 2.55 5.22 2.41 Maceration (Bell) 2-04 2-48 3.76 2.79 3.18 8.64 2.64 5.18 2.39 Adams .. 4.29 4.59 0.30 2.61 3.09 3.42 3.05 8.21 4.21 Maceration (Bell) 4.28 4-61 0.19 2.69 3.13 3.45 3.00 8.03 4.16 The following results (means of fifty-four analyses) carried out by three analysts working independently were obtained on nine samples of fresh milk with a view of indicating the variations between the results given by the several methods (Minutes of Evidence Departmental Committee on Milk and Cream Regulations Blue Book. Cd. 484 p. 391) : Method. Fat Per Cent. Method. Fat Per Cent. Adams (dry ether) . . . . 3.76 . Maceration (Bell) . 3.77 Adams (commercial ether). . . 3.78 . . . Centrifugal . 3-72 Werner-Schmid . . . . 3-88 . Centrifugal . 3.72 Thus the figure yielded by the maceration method is exactly the average of all the others and practically the same as that given by the Adams method.For the determination of the alcohol 50 75 or 100 grams of the milk are dis-tilled and the distillate re-distilled after being neutralized with decinormal caustic soda solution litmus-paper being used as the indicator. The specific gravity of the distillate made up to the original or other convenient bulk is determined in a 50-gram pycnometer and the quantity of alcohol corresponding to this specific gravity is deduced from a table. The percentage by weight of alcohol multiplied by :$, gives the percentage of lactose which has disappeared in the production of the alcohol. The experimental error in the determination of these small quantities of alcohol is negligible since the experience gained in the Government Laboratory shows that the differences obtained in successive tests of the same liquid would rarely or never exceed 0.00002 in terms of specific gravity and usually would not be more thaii 0.00001.The higher of these figures would in the case of milk correspond to about 0.02 per cent. of non-fatty solids. The proportion of volatile acid is ascertained by neutralizing 10 grams of the milk contained in a platinum capsule to the extent of one-half the total acidity (previously determined on another portion) with decinormal caustic soda and a little phenolphthalein added. The mixture is then evaporated to dryness on a water-bath with frequent stirring and after treatment with about 20 C.C. of boiling distilled water so as to break up and thoroughly detach the milk-solids from the capsule a further addition of decinormal alkali is made until the neutral point is reached.The difference between the original acidity of the milk and that of the evaporated portion is regarded as acetic acid. The production of 60 parts of acetic acid denotes a loss of 62 parts of the original lactose. When it is desired to take account of any butyric acid that may be present the volatile acids are separated by distillation from the quantity of milk that has been taken for determining the alcohol. A portion of the mixed aqueous solution of the acids is neutralized with baryta-water evaporated and dried until the weight i THE ANALYST. 203 constant. From the percentage of barium contained in the mixed salts the propor-tions of the two acids are calculated.Each molecule of butyric acid (88 parts) denotes a loss of 92 parts of lactose. The loss of lactose due to the formation of acetic acid is calculated as before. To estimate the small quantity of ammonia formed 2 grams of the milk are made up to a volume of 100 C.C. with distilled water (ammonia-free) and filtered through a carefully-washed filter. In 10 C.C. of the clear filtrate increased to 50 C.C. by the addition of ammonia-free distilled water the ammonia is determined by Nessler’s method using a standard solution of ammonium chloride containing 0.01 mgm. of ammonia per 1 C.C. The folIowing results are typical of those given by the author in the original paper. The milks in question are those in which the butyric fermentation did not develop to any extent : Sample No.Weeks kept. r I. Actual. Per Cent. ~~ 11. Calculated from Proportions of Alcohol Acid, and Ammonia. Per Cent. Difference between Actual and Calculated Loss. Per Cent. 1. 9 9 $ 9 $ 9 5. Y9 9 9 6. 7. 8. 9. 14. 9 9 ) 9 1 9 1. 2. 9 9 1 9 1. 9 9 9 9 9 ) TABLE ~.-UNWATERED WHOLE MILKS. 3 4 7 12 11 13 8 11 13 2 4 6 8 7+ lo& 0-26 0-23 0.36 0.48 0.26 0.49 0.67 0.25 0.27 0.36 0.87 0.19 0-23 0.20 0.30 0.20 0.21 0.34 0.49 0.20 0.49 0.61 0.23 0.35 0-32 0.76 0.16 0.18 0.17 0.20 TABLE II.-WATERED WHOLE MILKS. 1. Approximately 10 Per Cent. of Water added.2 4 0.18 0.33 0.36 0.51 0.27 0.42 0.33 0.50 2. Approximately 25 Per Cent. of Water added. 2 4 6 8 0.13 0.18 0-29 0.23 0.13 0.14 0 22 0.19 - 0’06 - 0.02 - 0-02 + 0.01 - 0.06 0.00 - 0.06 - 0.02 + 0.08 - 0.04 - 0.11 - 0.03 - 0.05 - 0.03 - 0.10 + 0.09 + 0.09 - 0.03 - 0.01 0.00 - 0.04 - 0.07 - 0.0 204 Weeks kept. THE ANALYST. Loss OF NON-FATTY SOLIDS. Butyric Acid. r A \ Per Cent. I. Actual. 11. Calculated. Per Cent. Per Cent. TABLE III.-UNWATERED SEPARATED MILKS. 4 6 8 6 0.41 0.32 0.31 0.29 0.31 0-33 0.66 0.67 0.59 1.03 0.95 0.55 0.86 0.90 0.23 0.65 0.41 0.22 0.13 0.26 0.28 0.20 3 4 0-17 9 ) 6 0.28 , 8 0.66 - 0.19 - 0.19 - 0.05 - 0.01 - 0.11 0.34 0.22 0.69 0.53 1-39 1 -10 TABLE IV.-WATERED SEPARATED MILKS (Approximately 10 Per Cent.Water). 3 55 55 57 0.32 1.47 1.54 1.47 0.24 1.50 1-56 1.50 - 0.08 + 0.03 + 0.02 + 0.03 Thus the actual amount of original non-fatty solids may be very closely ascertained from the methods described even although the keeping be very prolonged. The difference between the calculated and actual amounts of non-fatty solids is usually not greater than that obtained in duplicate determinations made by the same operator on fresh milk and the proportion of added water may be correctly deter-mined in fermented milk to within about 1 per cent. To simulate the worst conditions that might be found in practice the 25 per cent. of water added in one series of experiments was a much-polluted sample from the old Fleet River.The following figures show the results obtained on watered and unwatered milks in which butyric fermentation had set in : TABLE V. Sample No. 1 9 2 2 Difference. Per Cent. + 0.01 - 0.08 + 0.04 - 0.24 - 0-12 - 0.16 - 0.29 The occurrence of butyric fermentation to any marked extent is infrequent amongst the samples received but where it has occurred it will be seen from the above figures that the original non-fatty solids may be fairly accurately determined. Where the decomposition is considerable and much acid has been formed it is some-times difficult on account of the separated casein to get the sample into a uniform condition for analysis in which case the examination is not carried out. So far as the butyric acid is concerned however this constituent may be readily determine THE ANALYST 205 on the principle already laid down namely by a determination of the proportion of barium in the mixed acids. A. R. T. Note by Abstractor.-Standard strontia solution is preferred for the neutralization of the acidity of milk because of the tendency of the precipitate to be less flocculent and to settle better. Thus the maceration method can be more rapidly carried out, and less precipitate finds its way on to the weighed filter-paper. The lower molecular weight of strontium also causes less error to be introduced since any inaccuracy in the measurement of the solution has less influence on the amount of base to be deducted from the weight of the solids

ISSN:0003-2654    

DOI:10.1039/AN9053000197

出版商:RSC    

年代:1905

数据来源: RSC

摘要:

THE ANALYST, 205 ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOODS AND DRUGS ANALYSIS. Detection of Saccharin in Wine. Ed. MacKay Chace. (Joz~rn. Anzer. Chem. SOC., 1904, xxvi., 1627.)-In order to eliminate the substance present in many wines which responds to the salicylic acid test for saccharin, the author uses the following method in testing for the latter substance : 50 C.C. of the wine are extracted as usual, the ether is evaporated off, and the extracted matter re-extracted with light petroleum. Whether or not the presence of salicylic acid is indicated, this extract is returned to the dish containing the residue from the extraction with light petroleum, the whole made up to about 10 c.c., and heated to boiling after adding 1 C.C. of sulphuric acid (1 : 3). If salicylic acid is present, an excess of a 5 per cent.solution of potassium permanganate is slowly added, and the boiling continued for one minute, To the hot solution a small piece of caustic soda is added, and the iron and manganese hydroxides are filtered off. The filtrate, which should be strongly alkaline, is transferred to a silver crucible lid, evaporated to dryness, and the residue heated to from 210" to 215" C. for twenty minutes. The melt is dissolved in a small quantity of water, acidulated with sulphuric acid (1 part in 3), and extracted with ether. The ether extract is tested for salicylic acid with a 0-5 per cent. solution of ferric alum. The above method is capable of detecting 5 mgms. of saccharin per litre in the absence of salicylic acid. Salicylic acid is destroyed by it up to 200 mgms.per litre, and 10 mgms. of saccharin per litre can still be detected. A. G. L. Determination of Butter-Fat and Cocoanut Oil in Margarine. A. Kirschner. (Zeit. Untersuch. Nahr. Geizussmittel, 1905, ix., 65-70.)-A modification of a method first published by K. Jensen in the Danish Pharmaceutical Journal (Farma. Tidende, 1903, 385) is described. The method is a continuation of the Reichert- Meissl process, and is based on the precipitation of caprylic acid as insoluble silver caprylate, whilst silver butyrate is soluble. Five grammes of the sample are saponified and distilled in the usual way; 100 C.C. of the filtered distillate are then titrated with206 THE ANALYST. barium hydroxide solution to obtain the Reichert-Meissl value. To the neutral solution 0.5 gramme of silver sulphate is added, and the mixture shaken from time to time for one hour.The solution is then passed through a filter, and 100 C.C. of the filtrate transferred to a distillation flask. After adding 35 C.C. of water, 10 C.C. of dilute sulphuric acid (1 : 40), and a few pieces of pumice-stone, the whole is distilled ; 110 C.C. of distillate are collected, filtered, and 100 C.C. of the filtrate titrated with +; barium hydroxide solution. The number of C.C. of the latter required, calculated back bo 5 grams of fat, is termed the “second titration” value. The results of analyses of mixtures of margarine with butter or cocoanut oil, or both, are given, from which results the author has worked out two formulte for calculating the amounts of butter-fat and cocoanut oil present in any sample from the Reichert- Meissl and ‘‘ second titration ” vahes of the sample.The percentage of butter-fat = 4.319 X-0.456 R-2.15, and the percentage of cocoanut oil = 7-42 R - 8.116 S - 3.57 ; where S = the “ second titration ” value and R = the Reichert-Meissl value of the sample. w. P. s. The Use of Talc as a Coating or Polishing Material for Pearl Barley, Rice, Millet, and Dried Peas. H. Matthes and F. Muller. (Zeit. ofe?ztZ. Chem.. 1905, xi., 76-82.)-0f 25 samples of pearl barley examined by the authors, 16 were free from talc, 5 contained less than 0-2 per cent., and the remainder up to 0.63 per cent. Experiments carried out in the laboratory showed that by simply shaking the grains with talc it was quite possible to cause well over 2 per cent.of the latter to adhere t o the barley. Samples containing talc may be detected by shaking the grains with water, allowing the mixture to settle, and examining the sediment under the microscope. The quantity of talc present may be ascertained by igniting 25 grams of the sample, extracting the ash with hydrochloric acid to remove magnesium compounds, etc., natural to the barley, and then analysing the residue. From 0.24 to 1.00 per cent. of talc was found in 15 samples of rice, whilst 8 samples were free from this substance. Samples of millet contained from 0.04 to 0.1 per cent. of talc, and peas from 0.082 to 0.188 per cent. w. P. s. The Natural Alkalinity of Cocoa Ash. A. Froehner and H. Liihrig.(Zeit. Untersuch. Nahr. Genussmittel, 1905, ix., 257-263.)-The results of the analyses of cocoa beans of varying origin are given, from which it is seen that the water- soluble ash of 100 grams of dry cocoa powder requires from 10.0 to 30.0 C.C. of normal acid for its neutralization, and the insoluble ash from 30.0 to 42.7 C.C. By (‘ cocoa powder ” is meant ordinary commercial cocoa, containing less fat than cocoa beans. With regard to the question of added alkali in cocoa, it appears to be the general custom of manufacturers to treat “alkalized ” cocoas with a quantity of alkali equivalent to less than 3 per cent. of potassium carbonate. The authors considered that the alkalinity of a cocoa which requires less than 74 C.C. of normal acid to neutralize the soluble ash and 114 C.C.for the insoluble ash from 100 grams of cocoa does not exceed this limit. It is important to determine the alkalinity of the insoluble ash, as some makers use magnesium carbonate instead of, or in conjunction with, potassium and sodium carbonates. w. P. s.THE ANALYST. 207 Sugar as a Natural Constituent of Mace. W. Ludwig and H. Haupt. (Zeit. Untersuch. Nahr. GeRussmitteZ, 1905, ix., 200-204.) - Having recently had occasion to determine the quantity of added sugar in samples of adulterated mace, the authors first proceeded to ascertain whether the arillus or "mace" of the Myristicaceca naturally contained sugar. The maces examined were of various kinds-Banda, Menado, Papua, and Bombay-and were thoroughly extracted with petroleum spirit before dissolving out the sugary substances with water. The results obtained are given in the following table, the polariscopic readings being made on a solution containing 26.048 grams of the original substance per 100 C.C.: Polarization in a Sugar (as Dextrose) 200-millimetre Tube. by Fehling's Method. Kind of Mace. Banda ... _.. ... + 8.0 2.80 per cent. Banda ... ... ... + 8.4 4.28 ,, Menado ... ... + 6.8 2.19 ,, Papua ... ... ... + 6.4 1.65 ,, Bombay ... ... + 0.8 2.34 ,, The aqueous sugar solution underwent scarcely any change when inverted with hydrochloric acid at a temperature of 100" C. Two adulterated samples gave polariscopic readings of + 15.2" and + 36.4" respectively. The first, as the result of further investigation, was found to contain 9.5 per cent.of lactose and the second a considerable quantity of dextrin. The copper reductions of this latter sample, calculated as dextrose, were : Direct, 4.26 per cent. ; inverted at 60" C., 4.61 per cent. ; and inverted at 100" C., 17.54 per cent. w. P. s. The Adulteration of Saffron. A. Nestler. (Zeit. Untersuch. Nahr. Genuss- mitteZ, 1905, ix., 337-344.)-The peculiar crystals noticed in some samples of saffron, and previously mentioned by the author (ANALYST, 1904, 94), have now been found to consist of sugar, probably added as an adulterant. Two samples are also men- tioned : one being adulterated with 12.9 per cent. of barium sulphate and the other with 33.4 per cent. of a mixture of potassium nitrate and borax. w. P. s. The Determination of Morphine in Opium. Arthur and Albert Petit.(Journ. Pharm. Chim., 1905, xxi., 107-lll.)-The following method is that adopted as the official one in the new edition of the French Codex: 15 grams of an average sample of the opium are triturated in a mortar with 6 grams of slaked lime, and then mixed with 150 C.C. of water and allowed to stand for two hours with occasional agitation. After this the mixture is transferred to a filter, and 106 C.C. of the filtrate (= 10 grams of opium) collected. This is shaken with 30 C.C. of ether (specific gravity 0.725) so as to saturate the aqueous layer. Two grams of ammonium chloride (free from carbonate) are next dissolved in the liquid, which is shaken until a distinct precipitate appears, and then left for twenty-four hours in a covered beaker. The ethereal layer is then decanted on to two small counterpoised filters, and the aqueous portion again shaken with 30 C.C.of ether, which is decanted in turn. Finally the aqueous layer is poured on to the filters, so that the whole of the precipitate is oollected. The filters are washed with 25 to 30 C.C. of water208 THE ANALYST. previously saturated with morphine, and dried for two hours at 100" %. They are then washed with 30 C.C. of chloroform (deprived of alcohol by shaking with water) until the filtrate is colourless, after which the residue of morphine is again dried and weighed. The morphine thus purified ought to be completely soluble in & sodium hydroxide solution. Opium analysed by this method should yield not less than 10 per cent. of morphine. The authors confirm the statement of Dote and Hesse that morphine retains one molecule of water of crystallization even after prolonged drying at 100" C.C. A. M. Determination of the Purity of lodine. Hennecke. (Pharnz. Zeit., xlix., 957; through Pharnz. Journ., 1904, lxxiii., 958.)-The German and British Pharma- copceias direct that the purity of iodine be determined by dissolving it in potassium iodide solution and titrating with standard sodium thiosulphate solution in the usual manner. Weinland has recently pointed out that any iodine chloride present in com- mercial iodine will react with the potassium iodide as follows : IC1+ K I = KC1 + I,, the liberated iodine being also titrated. The author modifies the official method by dissolving the iodine in dry chloroform, and titrating the liquid by means of sodium thiosulphate, the end-reaction being quite definite without the aid of starch solution.I n the absence of water and potassium iodide, the iodine chloride is unacted on and takes no part in the reaction. A sample of commercial iodine which showed 100.9 per cent. of iodine by the pharmacopceial process gave 99.0 per cent. by the chloroform met hod. A. R. T. Cod-liver Oil Standards. E. J. Parry. (Chemist and Druggist, 1905, h i . , 491, 492.)-The tabulated results of the analyses of forty samples of cod-liver oil are given. The samples include the best brands known to the market, and no samples of doubtful origin have been included. The specific gravities (60' F.) of the samples were all within the limits 0.924 and 0.931. Most of the samples contained less than 1 per cent.of free acid, those containing more than 1 per cent. usually having a very rancid odour. Using Hiibl's solution in excess for eighteen hours, the iodine values found varied from 159 to 169. The Reichert values were generally below 0-7, although some undoubtedly pure oils gave values as high as 1.0 (for 2.5 grams). All the oils of good quality contained less than 1.5 per cent. of unsaponifiable matter. A few samples exceeded this figure, but they were rancid, and could not be accepted as of medicinal quality. The author considers that the high percentages of unsaponifiable matter found by Mann (ANALYST, 1904, 93) are incorrect. w. P. s. The Occurrence of Arsenic in Hydrogen Peroxide. L. Grimbert. (Joun~. Pharm: Chim., 1905, xxi., 385, 386.)-According to the author, arsenic is frequently present in commercial solutions of hydrogen peroxide. One sample recently examined gave a red precipitate on the addition of silver nitrate. When heated on the water-bath with a solution of calcium hypophosphite in hydrochloric acid it rapidly turned brown, and yielded a deposit of arsenic, the amount of which corre- sponded to 0-202 gram per litre. The residue left on evaporation of the hydrogen peroxide was 2.85 grams per litre, of which 2.10 grams consisted of sodium chloride, and the remainder of arsenates and sulphates. C. A. M.

ISSN:0003-2654    

DOI:10.1039/AN9053000205

出版商:RSC    

年代:1905

数据来源: RSC

摘要:

THE ANALYST. 209 ORGANIC ANALYSIS. Potassium Oxalate as a Lead Precipitant in Sugar Analysis. Harris E. Sawyer. (Journ. Amer. Chem. SOC., 1904, xxvi., 1631.)-The author has found that potassium oxalate precipitates lead completely and easily from sugar solutions, without altering the reducing power towards Fehling’s solution in any way. He uses 6 C.C. of a -$? normal potassium oxalate solution for every 50 C.C. of a solution obtained by dissolving 26 grams of sugar in 500 C.C. of water, adding 8 C.C. of a solution of basic lead acetate, and filtering after a few minutes’ shaking. Sodium oxalate may also be used, but the more soluble potassium salt is to be preferred. A. G. L. The Determination of Phenol. S. J. Lloyd. (Jozwn. Amer. Chem. SOC., 1905, xxvii., 16.)-As a result of a study of the sources of error in the determination of phenol by means of bromine, the author recommends the following procedure : The phenol solution is placed in a glass-stoppered flask, and to it is added a volume of hydrochloric acid (specific gravity 1.2) equal to about one-third or one-fourth of the combined volumes of the phenol solution and the hypobromite solution sub- sequently added.(This hypobromite solution is prepared by dissolving 9 C.C. of bromine in 2 litres of a solution containing 28 grams of potassium hydrate ; it is standardized by adding acid and potassium iodide and titrating with TG thiosulphate solution.) The hypobromite solution is added to the acidified phenol solution until the solution becomes permanently yellow ; an additional 10 or 20 per cent.of the hypobromite is then added, and the whole well shaken. An excess of potassium iodide is added, followed by about 10 C.C. of water for every 1 C.C. of acid present ; after adding 10 C.C. of chloroform, the iodine liberated is titrated with thiosulphate. The object of diluting with water is to prevent the acid from acting on the iodide or thiosulphate ; the presence of chloroform ensures more uniform results. Sulphuric acid may be substituted for the hydrochloric acid without impairing the accuracy of the determinati,on. From the test analyses it appears that the method yields results accurate to about 0.2 per cent. A. G. L. Tribromphenolbromide : its Detection, Estimation, Rate of Formation, and Reaction with Hydriodic Acid. S. J. Lloyd.(Joz~rrz. Amel.. Chm. Soc., xxvii., 7.)-The author finds that tribromphenolbromide ( C6H,Br,0Br) is not the primary product of the action of bromine on phenol, but is formed by the gradual action of bromine water on the first formed tribromphenol. As the rate of this reaction is decreased by adding acid or potassium bromide, and increased by adding water or bromine, or by raising the temperature, it is probably due to the presence of hypobromous acid in the bromine water. Aniline and benzidine in chloroform solution form convenient reagents for the detection of tribromphenolbromide. Aniline gives a deep-red colour, turning muddy; benzidine, an intense green, or purple if concentrated ; with tribromphenol, tetrabromphenol, and hexabromphenoquinone, neither aniline nor benzidine gives a colour ; bromine water bleaches aniline and gives a seal-brown colour with benzidine.The following colour reactions of tribrom- phenolbromide are less distinctive : Ammonia gives a brown colour ; dimethylaniline, slight darkening ; paratoluidine, clear red ; diphenylamine, light red ; u-naphthy-210 THE ANALYSTr. lamine, dark blue, turning purple ; P-naphthylamine, light-rose pink, which appears very slowly ; aminoazobenzene, brown-red ; pyridine, light yellow ; azoxybenzene, m-nitraniline, and acetanilide produce no change. The presence of a little sodium thiosulphate does not interfere with the reactions. Tribromphenolbromide is not quantitatively reduced to tribrornphenol by hydriodic acid unless the solution is exceedingly dilute.Powdered zinc and sulphuric acid, on the other hand, reduce it quantitatively. The reaction is conveniently carried out by placing about 0-5 gram of the substance in a 150 C.C. flask, and heating on a water- bath with 20 C.C. of 96 per cent. alcohol, 5 C.C. f sulphuric acid, and 0-25 gram powdered zinc, until the solution has lost its characteristic colour. Excess of calcium carbonate is then added; the alcohol is distilled off on a water-bath, and the residue extracted three times with boiling water, using 25 c.c., 25 c.c., and 10 c.c., respectively. I n the filtrate the bromide is titrated with silver nitrate and potassium chromate. Three tests made in this way gave 99.92, 100.61, and 99.07 per cent. of tribromphenolbromide recovered. A. G. L. A New Colour Reagent of the Polyphenols, their Isomers, and Higher Organic Compounds.E. P. Alvarez. (Chem. News, 1905, xci., 125.)-In a small porcelain basin are placed 0.2 gram of perfectly dry sodium dioxide, immediately afterwards 0.04 gram of the polyphenol to be tested, and then 5 C.C. of pure anhydrous ethyl alcohol. The contents of the basin are gently agitated for about five minutes, after which 15 C.C. of cold water are added. If the water be added at once the mixture may ignite. Colorations are obtained before and after adding the water, and in some cases an " edge coloration " is obtained by breathing on the edge of the liquid before diluting. The following colorations have been observed : Pyrocatechol. Resorcinol. H ydroquinone. Pyrogallol, Oxy-hydroqui- none. Phloroglucinol.Orcinol. Homopyrocate- chol. Thymohydro- quinone. Sodium Peroxide +Alcohol. Transitory pale pink, be- coming green, then brown. Very pale yellow, becoming green. Intense reddish-yellow. Dull red. Reddish-violet, changing Blue-violet. to dark brown. Intense pink. Blue-violet, changing at once to red. Intense orange. After breathing on Edge of Mixture. Blue-green. Transitory blue. - After adding the Water. Permanent red-brown. Permanent deep Persistent orange. Intense red. Yellowish. green. Intense blue - violet, which slowly fades. Rose red. Reddish-brown. Wine red, slowly fading. - Greenish-yellow. Deep yellow. w. P. s. The Rotatory Power of Oil of Turpentine. L. Raby. ( A m . de Chim. anal., 1905, x., 146-147.)-Riban has shown that on distilling neutral anhydrous oil ofTHE ANALYST.211 turpentine the sum of the rotatory powers of the different fractions is exactly that of the original oil. This is not the case, however, if the turpentine be acid or contain moisture. Thus, a commercial sample examined by the author yielded four fractions and a residue, each of which had a rotatory power lower than that of the original oil. After four months the rotatory powers of the four distillates had risen considerably, whilst that of the residue had not materially altered. By diluting one of the freshly-distilled fractions with its own volume of inactive olive oil the rotatory power immediately rose to the value that it only slowly attained in the case of the pure oil. The original oil had the specific gravity of 0-873 at 12" C., and its rotatory power (a,) was -34.19". It was moist and acid, and did not show any trace of resinification.The results obtained on distillation were as follows : First Fraction .., ... ditto +olive oil .., Second Fraction ... ... Third ,, ... ... Fourth ), ... ... Fifth (residue) . . . ... 3oiling-point, O C 153 -154.5 - 154-5-155 155 -156 156 -158 above 158 Specific Gravity at 12OC. 0.8693 0-8697 0.8702 0.8709 0-8951 - Rotatory Power (a,). Immediately after Distillation. - 28.93" - 37.65" - 28.63" - 28.03" - 26.98" - 25-29' Four Months after Distillation. - 37.66" - 37.65' - 34.90" - 34.62" - 35.85" - 23.16' C . A. MI. A New Method of Determining the Viscosity of Light-coloured Mineral Oils. (Chenz. Zeit., 1905, xxix., 385.)-The method depends on noting the time it takes a drop of water to fall through a definite height of the oil whose viscosity is to be determined.The time required is a function of the viscosity and the specific gravity of the oil ; if the latter be known, then by comparison with oils of known specific gravity and viscosity, the unknown viscosity may be calculated. I n carrying out the method, the time taken for a, drop of water to fall through a definite height in two oils, with known viscosities, which should be as widely apart as possible, is first noted, Let FA and F a represent the quotient Time of falling Viscosity according to Engle; Rudolf Nettel. of the two oils, A and a, respectively. oil of unknown viscosity-is given by the equation : Then Fz-Le., the corresponding value for an FZ =FA-[- FA - Fa - x (sp.g r . ~ - sp. gr. z)]. Sp. gr. A - Sp. gr. a And if the time of falling through the same height in the oil 2 be determined, the viscosity in Engler degrees is obtained by dividing this time by Fz. The determination of the time of falling may be conveniently made in a burette of about 12 millimetres diameter ; the bore must be very even throughout the tube.212 THE ANALYST. The height may be 400 millimetres, or, for very thick oils, one quarter of this. standard oils end the height are chosen so that the fraction The equals 100, which simplifies the calculation. The '' Vereinigten Fabriken fur Laboratoriumsbedarf " in Berlin make an apparatus in which a number of tubes are mounted in a water-jacket, the use of which appears to be essential, since an altera- tion of 1" in the temperature causes a large difference in the time of falling.In seven test results quoted the maximum difference between the viscosity as directly determined and as calculated according to this method is 0.2. Of course, care must be taken that the drops of water used are always of the same size. A. G. L. Composition of Grape-seed Oil. F. Ulzer and K. Zumpfe. (Oester. Chem. Zed., 1905, viii., 121-123.)-The following chemical and physical constants were yielded by a sample of this oil : Specific gravity at 15" C. ... ... ... 0.9215 Saponification value ... ... ... .- 190.0 Iodine value ... . . ... ... ... 142.8 Mailmen6 value ... ... 81" to 83" C. Refractive index ... ... ... 1.4623 Refractometer (Zeiss)'reading'at 50" C.... 54.5 ... Acetyl value of the fatty acids ... ... 43.7 The authors also fractionated the solid fatty acids-separated from the fluid fatty acids by the lead method-and from the further examination of these came to the conclusion that the oil consists chiefly of the glyceride of linolic acid, about 10 per cent. of solid glycerides and smaller quantities of the glycerides of oleic, ricinoleic, and linolenic being also present. Erucic acid could not be detected, and the oil certainly contained only the smallest traces, if any, of this acid. W. P. S. On the Alleged Natural Occurrence of Heptadecylic Acid. D. Holde. (Berzchte, 1905, xxxviii., 1247-1248.)-02ive Oil.-The author previously came to the couclusion that a mixed glyceride in olive oil contained a heptadecylic acid, CI7H,,O2 (ANALYST, xxvi., 324).He has since examined a larger amount of this substance, frac- tionally precipitating it with magnesium acetate, and after 15 fractionations succeeded in isolating palmitic acid and other fatty acids, some of which had a molecular equivalent above 290. An artificial mixture of arachidic acid (0-3 gram), stearic acid (0.7 gram), and palmitic acid (1 gram), behaved similarly to the acids from the mixed glyceride, yielding when fractionated four successive fractions, whose melting- points and molecular equivalents agreed closely with those of the supposed heptade- cylic acid. Hence it is not safe to conclude that a substance is a chemical individual even when four or five successive fractions are nearly constant in melting-point and molecular equivalent.The solid fatty acids of the olive oil itself were also fractionated, and after ten crystallizations an acid of molecular equivalent 368.7 and melting-point 72" to 72.8' C. was separated (lignoceric acid melts at 80.5" C., and has a molecular equivalent of 368). Kreis and Hafner's Heptadecylic Acid-The actual preparation of Kreis andTHE ANALYST. 213 Hafner (ANALYST, xxviii., 359), which melted at 5 5 ~ 5 ~ to 56.6" C., was examined in the same way, and eventually, after a large number of fractionations, it yielded stearic acid melting at 68" to 69" C., and having a molecular equivalent of 280.1. Gdrard's '( Daturic '' Acid.-The acid isolated by GBrard from datura oil (Comptes Rend., 1890, iii., 305), and the existence of which was confirmed by the author (Mitt aus dem konigl.techiz. Versuchsamt, 1902, xx., 66), has also been fractionated in large quantities. When fractionally distilled iiz vacuo, and the products of the distillation crystallized repeatedly from alcohol, this supposed acid now yielded a fatty acid with a molecular equivalent of 329 and melting-points of 57"-58-5" C. to 6 3 O to 64" C. Nordlinger's Heptndecylic Acid.-The substance prepared by Nordlinger (Zeit. angew. Chenz., 1892, 110) from palm oil was examined in a similar way, large quantities being used for the fractionation. Several succesgive fractions showed an apparently constant melting-point of 56" C., and the molecular equivalent of heptade- cylic acid, but eventually, after repeated fractionation, an acid melting at 68" to 68.5" C., and with a molecular equivalent of 288, was isolated.The apparent constancy of the earlier fractions was attributed by the author to there being present an intimate admixture of palmitic acid with acids of very high molecular weight. C. A. M. A Note on the Detection of Arehil, Cudbear, and Other Liehen Colours. L. M. Tolman. (Journ. Anzer. Chenz. SOL, xxvii., 25.)-In testing foods and medicines for added colours by dyeing a piece of wool in an acid-bath, extracting the colour with ammonia, acidifying this solution, and dyeing a second piece of wool, the lichen colours, as well as coal-tar colours, will be collected on the second piece of wool. They may be differentiated, however, by separating them in the first place from the natural colouring matters of any fruits or wines which may be present by extracting the ammoniacal solution with amyl alcohol, the alcoholic extract being evaporated and the purified colour tested. An aqueous aolution of the lichen colours is readily reduced by tin and hydrochloric acid, and reoxidized by ferric chloride, being thus distinguished from the azo-dyes and magenta, which are used as substitutes.For further identification, if thought necessary, the colour can be treated according to Allen (Vol. III., Part I., pp. 525 to 541). Sulphonated orceine, as well as the other forms in which lichen colours are used, gives the above reactions. A. G. L. The Nitrosites of Caoutchouc and their Applieation in the Analysis of Crude Caoutehoue and of Caoutchouc Goods.Paul Alexander. (Zeits. angew. Chenz., 1905, xviii., 164.)-The author has analysed the so-called dinitro-compound produced according to Weber's method (Ber. ? xxxvi., 3103) from twenty-eight different samples of caoutchouc, and finds its mean composition to be, for South American samples, C 45.98, H 5.45, N 11.63 per cent. ; and for African samples, C 44.33, H 5.77, N 12.06 per cent.; whilst Weber's formula C10H16N204 requires C 52.63, H 7.02, N 12-28 per cent. The author believes the dinitro-compound to consist of a mixture of the nitrosites of Harries with oxidized derivatives of caoutchouc. A. G. L.214 THE ANALYST. On the Analysis of Caoutchouc and Gutta-percha. W. Esch and A. Chwolles. (Chem. Zeit., 1904, xxviii., 1195.)-From a consideration of the methods employed and the values obtained by C.Harries (Berichte, 1903, xxxvi., 1937), the authors believe his "nitrosite c." method to be unreliable, if not worthless, especially as regards his failure to take into account the resinous and slimy substances always present in Para caoutchouc. They also believe that at present a, chemical analysis is of little value in judging of the properties of a sample of rubber. A. G. L. Detection of Adulteration in Natural Asphalt. B. Walen kovic. (Oester. Chem. Zeit., 1905, viii., 123-126.)-The following qualitative tests are described for distinguishing between asphalt and such substances as petroleum pitch, coal-tar pitch, and wood pitch. By determining the bromine value, the author also attempted to approximately determine the amounts of these substances in mixtures of the same.I n all the tests the asphalt was extracted with carbon disulphide, and the experiment performed on the extracted bitumen after drying the latter for thirty minutes at 105" C. Solubility in Sodium Hydroxide Xolution.--One gram of each substance was boiled for fifteen minutes with 100 C.C. of 10 per cent. sodium hydroxide solution, and the colour of the solution observed. Asphalt (Syrian, Dalmatian, and Trinidad) gave a nut-brown coloration ; petroleum pitch, none ; coal-tar pitch, light yellow ; and wood pitch, reddish-brown. Solubility in FormaZdehyde.-When boiled for fifteen minutes with a 40 per cent. formaldehyde solution asphalt yielded no soluble coloured matter, whilst all the surrogates gave coloured solutions.Solubility in Alcohol.-One gram of the substance was dissolved in 10 C.C. of carbon disulphide, placed in a 100 C.C. cylinder, and treated with-90c.o. of petroleum spirit. After filtration, a portion of the filtrate was mixed with 85 per cent. alcohol and gently shaken, but not sufficiently so to form an emulsion. With asphalt the alcoholic layer remained colourless, whilst in every other case it was coloured yellow From 2 to 5 per cent. of adulteration could readily be detected by means of this test. Bromine Vu1ue.-About 0.5 gram of the substance was placed in a small flask two-thirds filled with carbon tetrachloride, and treated with 25 C.C. of anhydrous bromine. After heating under a reflux condenser for about twenty-five hours, the contents of the flask were poured into a large porcelain basin and evaporated.The residue was dissolved in carbon tetrachloride, transferred to a tared flask, again evaporated, dried for thirty minutes at 105" C., and weighed. The following quanti- ties of bromine were found to be absorbed (by addition and substitution) : Trinidad asphalt, 93 per cent. ; Dalmatian asphalt, 89 per cent. ; petroleum pitch, from 62 to 72.5 per cent. ; coal-tar pitch, from 173 to 191 per cent. w. P. s. The Determination of Organic Carbon in Soils. J. H. Pettit and T. 0. Schaub. (Journ. Amer. Chem. Soc., 1904, xxvi., 1640.)-The author has obtained exceedingly good and rapid results by the use of Parr's method (Journ. Amer. Chem. Soc., xxvi., 294) of combustion with sodium peroxide, with subsequent determination of the carbonate formed. I n four determinations on pure sugar, from 99-46 to 99.89THE ANALYST. 215 per cent. of the carbon present was found. I n dealing with soils, the author used magnesium to start the reaction, a mixture of 1 gram of finely-ground magnesium powder with 2 grams of soil and 10 grams of sodium peroxide being used. The objection to the use of sulphur is that sulphur dioxide was always present in the gases evolved by treating the residue in the bomb with acid, even when potassium nitrate was added to the charge. As the magnesium is not always completely oxidized, and any residual metallic magnesium liberates hydrogen on treatment with acid, the carbon dioxide, after measuring, is absorbed by caustic potash and any residual hydrogen measured and taken into account. Results obtained in this way were practically identical with those given by combustion with copper oxide. A. G. L.

ISSN:0003-2654    

DOI:10.1039/AN9053000209

出版商:RSC    

年代:1905

数据来源: RSC

摘要:

THE ANALYST. 215 INORGANIC ANALYSIS. The Analysis of Tin Samples. Ernst Victor. (Chem. Zeit., 1904, xxix., 179.)-A description of a very complete method for the evaluation of commercial tin samples: 10 to 20 grams of the sample are dissolved in an Erlenmeyer flask in 100 C.C. of hydrochloric acid (specific gravity 1.124). After cooling, small quantities of potassium chlorate are added until the metals which had remained undissolved (Cu, Sb, As) are entirely in solution, and the solution boiled until all chlorine is eliminated. Ammonium tartrate, prepared by dissolving 30 grams of pure tartaric acid in a little warm water, and, after cooling, adding excess of ammonia, is added to the hydrochloric solution, and this then rendered just alkaline with ammonia. Cu, Pb, Sb, and Fe are then precipitated by the addition of small quantities of H,S water, boiled between each addition, filtered off and washed with H,S water.The lead is then determined electrolytically in the following manner : The mixed sulphides are dissolved in warm nitric acid (specific gravity 1*2), heated until the precipitated sulphur just turns yellow, filtered into a platinum dish, and the lead separated electrolytically as the peroxide, using a current of 0.2 to 2 amperes. For the determination of copper, a separate portion of the sample is treated in a similar manner, rendering the nitric acid solution of the sulphides strongly ammoniacal. The precipitated hydroxides are filtered off, and the filtrate, acidified with nitric acid, is electrolysed in the usual manner.Iron is separated as the hydroxide from the solution from which the lead was electrolytically separated. For the determination of antimony, a further portion is dissolved and oxidized as above. The chlorine is boiled off, the solution considerably diluted, and a few clean iron nails and a very little reduced iron are added. The flask, closed with a Bunsen valve, is heated on a sand-bath until the Sb, Cu, and As are separated in metallic form. The flask, avoiding the ingress of air, is cooled and the solution filtered through a paper on which a little reduced iron has been placed, the filter washed very thoroughly with water acid with HCl, and the deposit washed back into the flask. To obtain complete solution of this deposit, a little chlorate is placed on the filter, and this then washed with hot hydrochloric acid.The flask is gently warmed until solution is complete, when small fragments of solid sodium hydrate are added until a distinct excess is present. The solution is filtered through asbestos, 50 C.C. of saturated sodium sulphide solution added, and, after boiling, it is again filtered.216 THE ANALYST. Antimony is separated from this filtrate by electrolysis, using a current of 0.2 to 1 amp&re. The iron solution is prepared by dissolving 57.5 grams of ferric chloride with 25 C.C. of concentrated hydrochloric acid and making up to 1 litre; each C.C. of this solution corresponds t o 0.01 gram of Sn. As indicator, starch and a special iodine solution are used ; this is prepared by adding 10 grams of potassium iodide, dissolved in 10 C.C.water, to 10 grams of hydriodic acid (specific gravity 1.5) containing 3.3 grams of iodide of copper. This solution should be kept in the dark for some days before use. For the titration, 5 grams of the sample are dissolved in 500 C.C. of hydrochloric acid, adding chlorate to complete the oxidation. Fifty C.C. are filtered through asbestos into a flask, and, by means of a doubly perforated stopper, CO, is passed into the air-space in the flask, and some coarse aluminium powder is added. After complete solution of the aluminium and precipitation of the tin, further 50 C.C. of hydrochloric acid are added, and the whole heated until solution is complete. The solution is cooled in a current of CO, and titrated with the iron solution, using 10 drops of the iodine indicator and a little starch solution.This method is correct to 0.2 per cent. with samples containing less than 97 per cent. of Sn. NoTE.-Th. Goldschmidt, in a letter to the Chemiker Zeitung (1905, xxix., 276), draws attention to the fact that the above titration method has been employed in his laboratory (Staatshiitten Laboratory, Essen a.d. Ruhr) for the past fifteen years, and that the author of this paper, E. Victor, had worked in this laboratory in 1900-1901. H. A. T. A Charaeteristie Reaetion of Cobalt. E. Pozzi-Escot. (Ann. de Chim. annZ., 1905, x., 147.)-Solutions of cobalt salts yield insoluble reddish-brown precipi- tates on treatment with mono- substituted thiohydantoic acids or their derivatives. Thus, on adding a few drops of an alcoholic solution of phenyl- or P-naphthyl-thio- hydantoic acid, and then 1 drop of aninionium hydroxide, to a very dilute solution of a cobalt salt, there is an immediate scarlet coloration.Salts of nickel, under the same conditions, give an ochre yellow coloration or a grey precipitate. This coloration completely masks that of the cobalt if also present, but on adding an excess of ammonium hydroxide the nickel salt is dissolved instant aneou~ly, yielding a colour- less liquid, whereas the cobalt salt is much less soluble, and sufficient will be left undissolved to show tbe red coloration. C. A. M. The Use of a Rotating Anode in the Electrolytic Estimation of Zinc. Leslie H. Ingham. (Jozcrn. Amer. Chein. SOC., 1904, xxvi., 1269.)-The author obtained good results in the electrolytic deposition of zinc from various solutions, using a platinum spiral revolving at the rate of 230 or 560 revolutions per minute as anode.The cathode consisted of a silvered dish, from which the zinc was removed at the end of each experiment by dilute sulphuric acid (1 : 50). I t was found that, using an electrolyte containing sodium acetate, 0.5 gram zinc would be Completely deposited with the higher speed in fifteen minutes, using a current of 4 amperes (for about 100 square centimetres) and 11 volts. With sodium hydroxide solutions217 THE ANALYST. 0.25 gram zinc could be deposited in fifteen minutes, using 5 amperes and 6 volts. From sodium formate solutions slightly acid with formic acid, 0.25 gram zinc were deposited in twenty minutes, using 5 amperes and 8 volts. Potassium cyanide solutions gave unsatisfactory results.Both the sodium acetate and formate methods were found to be especially applicable to the determination of zinc in zinc blende, the time required to complete the analysis on 0.5 gram of ore being onlyabout two and a half hours. In each case the iron was removed by two precipitations as basic acetate or formate. The best conditions for the composition of the final mixed filtrate, which was used 8s electrolyte, were found to be : (a) Dilution, 125 C.C. ; NaC2H302, 3.5 to 4.5 grams ; current, 5 amperes and 13 volts; time, thirty minutes. ( b ) Dilution, 125 C.C. ; Na2C03, 5.5 grams ; formic acid (specific gravity 1*22), 4.8 C.C. ; current, 5 amperes and 9 volts; time, twenty-five minutes.(The Na2C03 in ( b ) represents the salt used to neutralize the acid in the solution before the basic formate precipitation.) Working on a zinc blende containing 65-70 per cent. of zinc, the results of ten experiments by both methods varied between 65-62 and 65.96 per cent. I t was not found possible to use the sodium hydroxide method for the deter- mination of zinc in the blende, as the separated ferric hydroxide always contained zinc ; but good results were obtained by substituting ammonia for the sodium hydroxide, the iron being separated by two precipitations, and the mixed filtrates electrolysed under the foliowing conditions : Dilution, 125 C.C. ; hydrochloric acid (specific gravity 1*21), 10 C.C. ; ammonia (specific gravity 0.95), 2 C.C.in excess ; NH,Cl, 0.5 gram; current, 5 amperes and 6 volts; time, twenty minutes. The results varied between 65.63 and 65.75 per cent. The anode was not attacked during these experiments. A. G. L. The Preparation of Pure Earths of the Cerium Group by Means of their Double Alkali Carbonates. R. J. Meyer. (Zoits. anorg. Chenz., 1904, xli., 97.)- The preparation of double carbonates of cerium, lanthanum, praseodymium, and neodymium with the alkalis is described. These are generally of the types KJ~a,(C0,),.12R,0 ; (NH,),La,(C0,),.4H2O ; and Na,La4(CO,),.22R,O ; but ammonium cerium carbonate has 6H,O, and all the sodium compounds are so easily decomposed that their exact composition, especially as regards the amount of water they contain, is doubtful. All the carbonates are obtained by adding a solution of a salt of the rare earth to a concentrated solution of an alkali carbonate; the sodium and ammonium compounds are practically insoluble, whilst the potassium compounds are easily soluble in strong potassium carbonate solution.On diluting this solution with water, the double carbonates are precipitated in the following order : Lanthanum, praseodymium, cerium, neodymium. This behaviour affords a ready means of purifying the rare earths. Job's method (Compt. rend., 1898, cxxvi., 246) is recommended as a delicate test for cerium, 0.5 per cent. of this earth in mixtures with other rare earths being still recognisable. The test consists in dissolving the earths in potassium carbonate and adding a little hydrogen peroxide, a yellow colour indicating cerium.The praseodymium peroxide obtained by ignition of the oxalate corresponded most nearly with the formula Pr,O,,. The oxide Proz was obtained, following218 THE ANALYST. Brauner’s directions,$by heating the nitrate with potassium nitrate to 400’ to 450” C. It could not be obtained by heating Pr,Oll with CeO,. Baskerville and Turrentine’s citric acid method (Jourrt. Amer. Chenz. SOC., 1904, xxvi., 46) for the purification of praseodymium was found to be impracticable. A. G. L. A New Method of Detecting and Determining Ammonia. A. Trillat and Turchet. (BUZZ. SOC. Chiin., 1905, xxxiii., 304-310.) -Nessler’s reagent gives erroneous results in the presence of hydrogen sulphide, carbonic acid, lime, and certain albuminoid substances, owing to appreciable quantities of ammonia being thereby uasked.The author’s new method is claimed to obviate these sources of error, while being as sensitive and rapid as Nessler’s process. It is based upon the formation of nitrogen iodide, which gives an intense black coloration to water, easily perceptible in a solution containing 1 part of ammonia in 500,000. Nitrogen iodide is not produced in very dilute solutions by the direct action of iodine on ammonia, but is instantly formed by the action of iodine chloride on ammonia, and the com- pound is stable in the presence of an alkali : 3C1I + NH, + 3NaOH = 3NaCl+ NI, + 3H,O. Thus, if a water containing ammonia be treated with a solution of potassium iodide, and a few drops of a dilute solution of an alkali hypochlorite, the iodide is decom- posed, with the formation of iodine chloride, which then reacts with the ammonia.The nitrogen iodide is soluble in an excess of either of the reagents, and it is specially necessary to avoid adding too much potassium iodide. No black coloration that could be confused with that given by ammonia is produced by amines, amides, ureides, pyridine derivatives, nitrates or nitrites. Methylamine gives a red coloration, appearing blue by transmitted light, and aniline a reddish-brown colour, both of which can readily be distinguished from the colour produced by ammonia. On the other hand, the nitrogen iodide reaction is given by all ammonium salts, including the cyanide and sulphide, the presence of which interfere with Nessler’s reaction.Drinkzng-WCLter.--From 20 to 30 C.C. of the undistilled water are treated with 3 drops of a 10 par cent. solution of potassium iodide, and 2 drops of a concentrated solution of alkali hypochlorite (commercial e m , de JaveZZe), and there is an immediate turbidity or brownish-black precipitate when the quantity of ammonia exceeds 2 mgms. per litre of water. In the case of smaller quantities the water is evaporated in the presence of a very small amount of sulphuric acid, which is roughly neutralized before applying the test. The small quantity of iodine liberated in the reaction gives a slight yellow tint to the water, which does not interfere with the test. I n doubtful cases, however, a comparative blank test with pure water may be made. A very slight excess of the hypochlorite solution destroys the iodine coloration without affecting that of the nitrogen iodide.The iodine may also be separated from the iodide by means of chloroform, but as a rule these precautions are unnecessary. A colorimetric determination can be made in the same way as in Nessler’s process. A table of Comparative results obtained with different drinking-waters shows that the new method gives figures in close agreement with those obtained by Schloesing’s method. C. A. M.THE ANALYST. 219 On the Quantitative Determination of Iodine in Soluble Iodides and in Mixtures of Iodides with Bromides and Chlorides. Hugo Ditz and B. M. Margosches. (Chem. Zeit., 1904, xxviii., 1191.)-The authors determine iodine in soluble iodides by decomposing them with potassium iodate in the presence of a small excess of dilute sulphuric acid.The liberated iodine may be shaken out with toluene, the iodine being either titrated directly in the toluene solution, or the excess of iodate in the aqueous solution may be determined by adding an excess of iodide and titrating the liberated iodine. Or, again, the iodine may be removed from the solution by boiling, and the excess of iodate determined as above. Any one of these modifications may be used to determine iodides in the presence of chlorides. When bromides are also present, however, the boiling-off method may not be used, as bromides are attacked in the heat by iodate in acid solution, even when acetic acid is substituted for sul- phuric acid; and even working in the cold, the excess of sulphuric acid should be ‘t(s small as possible, the amount necessary being best determined by a preliminary experiment.The results obtained in a large number of test analyses by the above methods were good, the error generally being less than 1 mgm. of iodine on quantities of iodine up to 0.5 gram. A. G. L. Acetic acid, however, may be used in fair excess in the cold. Determination of Small Quantities of Bromine and Chlorine in Iodine. R. R. Tatlock and R. T. Thomson. (Journ. SOC. Chem. Id., 1905, xxiv. 187.)- Five to 10 grams of commercial iodine are treated with 50 to 100 C.C. of water, and finely-granulated zinc or zinc dust carefully added in small quantities at a time with frequent agitation, until all the iodine is converted into zinc iodide.The temperature should not sensibly rise during the operation. The solution is filtered, and 3.5 to 7 grams of sodium nitrite (according to the amount of sample originally taken) added to the filtrate. The liquid is acidified (not strongly) by the cautious addition of diZuted sulphuric acid, when all the iodine is liberated, but no bromine is set free if diluted sulphuric acid is used. The precipitated iodine is filtered off, washed with a little cold water, and the filtrate shaken with benzene, chloroform, or carbon disulphide, to remove any free iodine in solution. The aqueous liquid is separated and again treated with sodium nitrite and acid, and shaken out with benze n e . Any traces of the solvent may next be expelled by heating the liquid, and the bromine and chlorine then precipitated by the addition of excess of silver nitrate and some nitric acid, and the liquid boiled.The silver salts are collected on a tared filter, and thoroughly washed with hot water. The silver chloride is then dissolved out of the mixture by repeatedlypouring through the filter about 60 C.C. of a solution prepared by dissolving 2 grams of silver nitrate in 90 C.C. of water, and adding 10 C.C. of ammonia (specific gravity O.SSO), and the residue finally washed with the remaining 40 C.C. of the ammoniacal silver nitrate. The chloride is completely dis- solved in this way, leaving the insoluble silver bromide, which is washed with warm dilute nitric acid, and then with hot water, dried, and weighed. The ammoniacal solution containing the silver chloride is acidified with nitric acid, heated, and the silver chloride collected as usual.To test for iodide in the bromide precipitate, as should generally be done, it is220 THE ANALYST. treated with dilute sulphuric acid and zinc-dust until decomposition is complete. The liquid is then filtered and a limited quantity of chlorine-water added, and extracted with chloroform. In the presence of iodine a, violet tint will be obtained, but when bromine only is present the chloroform has a reddish-brown colour. A. R. T. The Detection of Bromine in the Presence of a Large Quantityof Iodine. H. Cormimbauf. (Ann. de Chinz. anal., 1905, x., 145-146.)-A solution of the iodides (or of hydriodic acid neutralized with sodium hydroxide), which must be neutral or very slightly acid, is treated with excess of a solution of ferric chloride (specific gravity 1-45), which precipitates the iodine in the form of a black powder.This is separated by filtration through cotton-wool, and the filtrate boiled until violet vapours no longer appear, in order to eliminate the last traces of iodine. The iron is then precipitated by means of sodium hydroxide, and the colourless filtrate from this precipitate is treated with a crystal of potassium chlorate and chloroform, followed by several drops of concentrated sulphuric acid. Any bromine present is liberated and taken up by the chloroform, which becomes more or less yellow. In this way 0.1 per cent. of bromine can easily be detected, For the detection of bromine in commercial iodine several grams of the sample are treated with an excess of reduced iron in the presence of 50 C.C.of water. The ferrous iodide formed under these conditions will contain ferrous bromide if bromine is present. The liquid is filtered from the undissolved iron, and the filtrate treated with ferric chloride, etc., as above described. C. A. M. The Precipitation of Barium Bromide by Hydrogen Bromide. Normann C. Thorne. (Zeits. anorg. Chem., 1905, xliii., 308.)-The author shows that barium is completely precipitated as bromide when barium bromide is dissolved in a small quantity of water and a mixture of ether and of a saturated aqueous solution of hydrogen bromide is added, especially readily if the solution be then saturated with gaseous hydrogen bromide. The presence of ether is not absolutely necessary.Barium may be separated in this way from considerable quantities of calcium and magnesium. On drying the precipitate, a small quantity of oxybromide is formed; this may be obviated by previously washing the precipitate with a solution of ammonium bromide, and drying first at a low temperature and then at 250" C. until the ammonium bromide is completely volatilized. Hydrogen bromide will also Precipitate barium completely as bromide from concentrated solutions of the chloride. Hydrogen chloride throws down the chloride quantitatively from these solutions. When both hydrogen chloride and hydrogen bromide in considerable quantity are added, the Precipitate consists of a mixture of barium chloride and bromide. A. G. L. Diphenylamine as a Reagent for Nitrites, Nitrates, and Chlorates, and its Use when mixed with Resorcinol and $Naphthol.E. P. Alvapez. (Chem. News, 1905, xci., 155.)-The author employs EL sulphuric acid solution of mixtures of diphenylamine with resorcinol or $-naphthol, as the colorations obtained are more persistent and are more easily distinguished from one another than whenTHE ANALYST. 221 diphenylamine is used alone. The reagent is prepared by dissolving 0.1 gram each of diphenylamine and re-sublimed resorcinol in 10 C.O. of sulphuric acid. Five or six drops of this solution are poured on about 1 mgm. of the salt to be tested. With nitrates, a very permanent yellowish green coloration is obtained, the edges of the spot becoming blue on spreading the liquid out.On adding alcohol, an orange- coloured solution is obtained. With nitrites, a deep blue-violet coiour is yielded. On moving the liquid, the edges of the spot become red, and a red solution is obtained on the addition of alcohol. With chlorates, the result is not satisfactory, but by substituting @naphthol for the resorcinol, a dull-green coloration is obtained, which, after a time, changes to a deep grey, almost black. The solution resulting from the addition of alcohol is greyish or blackish. w. P. s. Some Reactions of Hydrofluosilicic Acid. A. Gawalowski. (Zeit. mal. Chenz., 1905, xliv., 191-194.)-An aqueous solution of hydrofluosilicic acid of 1.06 specific gravity gives characteristic reactions with 10 per cent. solutions of the following reagents : Sulphuric acid yields a white crystalline precipitate, which is insoluble in water, dilute nitric acid (1 : lo), and alcohol, but dissolves in potassium hydroxide solution, rapidly on heating, whilst large white flakes simultaneously separate.Potassium cliromate gives a compact yellow precipitate, whilst the super- natant liquid is yellow at first and then colourless. The precipitate darkens in hot water. It is insoluble in cold nitric acid and alcohol, but dissolves in hot or cold potassium hydroxide solution and in sulphuric acid, with the simultaneous separation of flakes. Potassium bichromate gives an orange - yellow crystalline precipitate insoluble in water, cold nitric acid, cold potassium hydroxide solution, and alcohol, but soluble in traces in hot nitric acid and potassium hydroxide solution. Hydro- chloric acid (specific gravity 1-10> gives a white compact crystalline precipitate, while the supernatant liquid is a faint bluish-green.I t is slowly soluble in hot water and in cold nitric acid and potassium hydroxide solution (no separation of flakes). Cbro??&ic acid gives a yellow precipitate which is insoluble in the different reagents, with the exception of hot potassium hydroxide (separation of flakes). Potassium bichrornochloride gives an orange precipitate insoluble in cold water, cold nitric acid, alcohol, and a cold mixture of alcohol and sulphuric acid, but very slightly soluble in hot nitric acid. Cold potassium hydroxide solution turns the precipitate yellow and then dissolves it (separation of flakes).A hot mixture of alcohol and sulphuric acid decomposes it, with the formation of a green colour. C. A. M. Estimation of Phosphoric Acid. F. Raschig. (Zeits. f. aqew. Chem., 1905, xviii., 374.)-One of the chief desiderata in technical analytical chemistry is to replace tedious gravimetric methods by quick practicable volumetric processes. Quite recently G. v. Knorre has shown that it is possible to estimate sulphur in pyrites with equal accuracy and great gain in time, aompared to the barium chloride method, by precipitating as benzidine sulphate, and estimating volumetrically its content of acid by qG soda. The principle underlying this process is that, by washing a precipitate not absolutely insoluble with quantities of water so small as not to make the error due222 THE ANALYST.to the solubility appreciable, it is yet possible to free the precipitate from mother- liquor sufficiently to make error from that source negligible. The author set himself to determine whether this principle could not be applied to other cases, and selected the estimation of phosphoric acid as being even more frequently required than that of sulphuric acid. At present the method in general use is the gravimetric estimation as Mg,P,07, where the phosphoric acid is first precipitated as MgNH,PO,, a substance somewhat soluble in water, and therefore usually washed with dilute ammonia solution. A. Hebebrand suggested the titration of this precipitate with carminic acid as indicator, but he washed with ammonia, and eliminated the ammonia with alcohol.This was somewhat tedious and inconvenient, and the method has not obtained general recognition. The author, who has used the method of washing with small quantities of water, made some trial experiments in the following manner : Twelve lots of 20 C.C. each of a & solution of sodium phosphate (= 3-1 gram P per litre) were precipitated with 10 C.C. magnesia mixture, and the precipitate collected on a suction filter designed by the author (Zeits. f. nngew. Chern., 1903, xvi., 818). The filtrate was returned into the vessel used for the precipitation, shaken well round, and again poured on to the filter, so that the whole precipitate was transferred to the filter without using any water. The first was not washed at all, the other eleven with increasing amounts of water (see table below).Finally the filter and precipitate were placed in a wide-necked flask of about 200 C.C. capacity, a drop of methyl orange added, and titrated with TG hydrochloric acid. The reaction proceeded very rapidly, the precipitate dissolving quite easily off the paper. The colour change occurred when the salt NH,H,PO, was formed. The precipitate from 20 C.C. of the original sodium phosphate solution should require 40 C.C. ;k HCI. The results obtained were : Experiment. 1 ... ... 2 ... ... 3 ... ... 4 ... ... 5 ,.. ... 6 ... ... 7 ... ... 8 ... ... 9 ... ... 10 ... ... 11 ... ... 1 2 ... ... C.C. & HC1. Volume of Wash- water in C.C. 0 ... ... 45.1 1 ... ... 42.6 2 ... ... 41.7 3 ... ... 40.6 4 ... ... 40.0 5 ... ... 40.0 10 ... ... 40.0 20 ... ... 39.8 40 ... ... 39.6 80 ... ... 39.2 160 ... ... 38.7 320 ... ... 37.2 From these figures it appears that 4 C.C. sufficed to get rid of the alkaline mother- liquor sufficiently to avoid error, whilst the volume of the wash-water could be in- creased to 40 C.C. without introducing an error of more than 1 per cent. The author suggests the following mode of procedure for the estimation: An amount of substance should be weighed out which will not contain more than 0.15 gram P,O,, requiring, therefore, less than 50 C.C. of TG acid. The solution is then precipitated in an Erlenmeyer flask with magnesia mixture, and filtered off withTHE ANALYST. 223 the aid of the suction-pump, using a filter-paper of 40 mm. diameter. The clear filtrate is used to rinse the rest of the precipitate on to the filter, and the mother- liquor is sucked off as completely as possible, 10 C.C. of water poured on to the filter and sucked through, then another 5 c.c., and the precipitate once more dried as far as possible by suction. The filter and precipitate are then transferred, with the precipitate upwards, by means of tl, pair of forceps to a beaker. The forceps and funnel are washed with the least possible amount of water, a drop of methyl orange added, excess of & acid added, and the excess estimated by & caustic soda. E. K. H.

ISSN:0003-2654    

DOI:10.1039/AN9053000215

出版商:RSC    

年代:1905

数据来源: RSC

摘要:

THE ANALYST. 223 APPARATUS. A Simple Apparatus for the Determination of Volatile Substances by Difference. L. Lehn Kreider. (Zeds. anorg. Chem., 1905, xliv., 154.)-The apparatus shown consists of three tubes, of which A is an qrdinary test-tube; B is a test-tube in the bottom of which a hole is made; and C is a test-tube drawn out to a long narrow tube a t the lower end. The three tubes fit closely into each other as shown, the capillary of C going through the hole in B, in which it is packed with cotton-wool, the space above in B being filled with granular calcium chloride to dry the escaping gas. The space between A and R is made tight by means of a ring of melted parafin wax. The substance to be acted on is placed in A ; the liquid reagent (acid, hypobromite solution, etc.) is sucked up into C through the narrow tube and kept in the upper part by inserting a stopper carrying a glass tube closed by a piece of rubber and a glass rod.The whole apparatus is then put together and weighed. On removing the piece of glass rod from the top of C the liquid descends into A , acts on the substance, and the gas evolved, after being dried by the calcium chloride, escapes through the narrow annular space between B and C. A current of air is then led in at the top of C and the apparatus again weighed. A number of determinations made in this way of the carbon dioxide in calcium, barium, and strontium carbonates, of the hydrogen generated by magnesium and zinc, and of nitrogen in urea, ammonium oxalate, and ammonium chloride, show errors of, at the most, a few tenths of a mgm.A. 0. L. On the Use of Vessels of Quartz Glass in the Laboratory. F. Mylius and A. Meusser. (Zeds. aizorg. Chem., 1905, xliv., 221.)-A number of tests made in the Reichsanstalt show that quartz vessels are not attacked by pure water, or by neutral or acid solutions. Concentrated sulphuric acid has a very slight action at 100" C.; phosphoric acid corrodes the quartz strongly above 400" C. Alkalies take up considerable quantities of silica. Saturated solutions of barium hydroxide also act on the quartz, but very slowly. From a 30 per cent. solution of caustic potash the vessels take up a small amount of potash, which cannot be washed out with cold water, but can be recovered by boiling with water. Soda does not appear to be absorbed in this way.Certain organic dyes-q., methylene blue, congo red, rhodamine, aniline blue, and iodeosin, are also absorbed in very small quantities by vessels of quartz glass. A. G. L.224 THE ANALYST. On Density Determinations by Means of a Pipette and the Preparation of Volumetric Solutions aecording to Specific Gravity. F. W. Kuster and Siegmar Munch. (Zeits. anorg. Chem., 1905, xliii., 373.)-The authors show that & with proper care the density, as determined by weighing the liquid delivered by a 100 C.C. pipette, is accurate to about 0.00005. The degree of accuracy may be still further increased by making use of the overflow pipette shown. This is filled by suction until the liquid overflows into the upper bulb, when the stopcock is closed : on opening the stopcock the same volume of liquid is always delivered.I n filling and emptying the pipette the point should always dip into the liquid to the same depth (about 0-5 centimetre); the point should be finely drawn out ; the temperature must be kept constant within I" C . ; and the pipette should be very frequently cleansed with a warm solution of potassium bichromate in sulphuric acid. Using this pipette, the error in the density should not exceed 0~~00026. The pipette may be used in conjunction with Table XIII. of Kuster's Logarithmische Rechentafeln fiir Cherniker " for the preparation of standard volumetric solutions, but the authors find that it is more convenient to find the weight of water delivered by the pipette at 18' C., and then always to work at this temperature. They give a table showing the number of C.C. of hydrochloric acid of various densities which diluted to 1,000 C.C. in order to prepare normal acid. The error in the strength of acid prepared in this way does not exceed 0-001 per cent. A. G. L. @%+I+&@

ISSN:0003-2654    

DOI:10.1039/AN9053000223

出版商:RSC    

年代:1905

数据来源: RSC

摘要:

224 THE ANALYST. PROCEEDINGS OF THE SOCIETY OF PUBLIC ANALYSTS. THE monthly meeting of the Society was held on Wednesday evening, May 3, in the Chemical Society’s Rooms, Burlington House. The President, Mr. E. J. Bevan, occupied the chair. The minutes of the previous meeting were read and confirmed. Certificates of proposal for election to membership in favour of Messrs. G. J. Alderton, B.Sc., Oscar Guttmann, J. T. Hewitt, M.A., D.Sc., Ph.D., A.R.C.Sc., and G. D. Lander, D.Sc., were read for the second time. Messrs. E. XI. Bolton, J. T. Dunn, D.Sc., E. K. Hanson, M.A., F. T. Harry, and F. R. Henley, M.A., were elected members of the Society. The following papers were read : ‘‘ Motes on Preservatives,” by E. G. Clayton; ‘( Notes on the History of Distilled Liquors-especially Whisky and Brandy ” (with lantern illustrations), by Thomas Fairley ; and a ‘‘ Note on a,ln Objectionable Method of Fining Wines,” by R. Bodmer.

ISSN:0003-2654    

DOI:10.1039/AN9053000224

出版商:RSC    

年代:1905

数据来源: RSC