摘要:

THE ANALYST. NOVEMBER, 1891. PROCEEDINGS OF THE SOCIETY OF PUBLIC ANALYSTS. THE opening meeting of the Session 1891-2 was held on Wednesday, the '7th ult., Dr. P. Vieth (in the absence of the President) being in the chair. The minutes of the last meeting having been read and confirmed, and Mr. Edgar B. Kenrick having been formally proposed as a member of the Society, Mr. A, H, Allen read the following paper :- E'URTHER NOTES ON ACETINI BY A. H. ALLEN AND D. HOMFRAY, IN a paper published in the September number of the ANALYST, we showed that when acetin was dissolved in alcohol together with a very small proportion of caustic alkali, the mixture gave, on distillation, a quantity of ethyl acetate which constituted a large pro- portion of the acetin employed, and which was largely in excess of the chemical equivalent of the caustic alkali used.Since the publication of the paper containing an account of these researches, further experiments have shown us that a larger proportion of alkali diminished the yield of acetic ether. Thus, with sufficient caustic soda fo saponify 12.7 per cent. of the acetin used, 85.2 per cent. of the acetic acid passed into the distillate as ethyl acetate, This, With the 1237 remaining in the retort as sodium acetate, leaves only 2-1 per cent, unsaponified or otherwise to be accounted fors In an~ther experhent ixi ~,-Ecfi the aka& used was equivalent to 39.8 per cent. of the acetin, the amount converted into acetic ether was 52.4 per cent., leaving 7.8 per cant. unaccounted for. In all cases the distillates were found to be neutral to phenolphthalein, showing that no free acetic acid had distilled over.The residues in the distilling flasks were always found to be slightly, but distinctly, mid to phenolphthalein, a fact which proves that the whole of the alkali employed invariably entered into reaction, and that some reaction occurs resulting in the production of free acetic acid. Hence it appears probable that the first action of the alkali on the acetin is to affect its hydrolysis, with the formation of glycerin and free acetic acid, A portion of the latter then reacts with the alkali to form sodium acetate, while the greater part reacts with the alcohol to form ethyl acetate; but a small proportion of the acetic202 THE ANALYST. acid fails to enter into this latter reaction, and hence remains in the free state in the distilling flask.Why the soda ehould effect the hydrolysis of a much larger amount of acetin than it can chemically react with is not clear; but the behaviour of soda is evidently similar to that of lime, magnesia, and oxide of zinc, very small amounts of which (2 or 3 per cent.) are found to very greatly facilitate the hydrolysis of fats by super-heated steam, and are largely employed with this object. A few supplementary experiments deserve mention. One gramme of anhydrous sodium acetate was distilled with 50 C.C. of alcohol, the distillate boiled with alkali and titrated back with standard acid, when it was found that no acetic ether had been formed. A mixture of acetin, alcohol, and anhydrous sodium acetate gave a small yield of acetic ether.It seemed possible that this result might be due to the presence of alkali in the sodium acetate. Hence the experiment was repeated by dissolving sodium acetate in alcohol, dropping in acetic acid till the alkaline reaction to phenolphthalein disappeared, and then adding acetin and distilling as before. Acetic ether, equivalent to 1-36 and 1.50 per cent. of the acetin taken, was found in the distillate This is no more than was obtained by distilling acetin and alcohol together. DISCUSSION. The Chairman having invited discussion, M i . Bertram Blount wiBhed to call attention to‘a phenomenon which seemed to him to be analogous to that described by the authors, namely, the fact that the complete saponification of a fat, and the determination of the saponification-equivalent was not possible unless a considerable excess of potash was used.If, for instance, one attempted to saponify cocoa-nut oil as in the Koettstorfer process, with only a slight excess of potash, the liquid might have a perfectly definite alkaline reaction to phenolphthalein, and use an appreciable quantity of acid on titrating back; nevertheless, when the experiment came to be figured out, it would be found to be entirely wrong, for the simple reason that saponification had never been completed. The Chairman regarded the President’s absence as being especially regrettable on this occasion, because he had done more work than any member present, with the exception of &ir. Aiien, in connexion with the analysis ~f fats.Xeferrhg to BIr. AIsn’a remark that the reaction described by him was similar to that which took place in the decomposition of fats by superheated steam, and considering that the reaction now before them must take place a t a rather low temperature, seeing that an alcoholic solution of potash was employed, it seemed to him that some further light might be thrown on the subject if the action were tried at higher temperatures with a view of ascertaining whether or no6 it was more energetic under such circumstances, Mr. Allen, in reply, said that, dealing with the Chairman’s remarks h t , he would aimply say that it was hardly possible that the reaction should be much more energetic under any circumstances, seeing that they already had about 85 per cent. of the original acetin converted into acetic ether. He was very much inclined t~ agree withTHE ANALYST. 203 Mr. Blount’s suggestion, Taking cocoa-nut oil as being laurate of glycyl, one would expect the %me reaction between that and alcoholic potash as occurred with the butyrate, acetate, stearate, and palmitate. The matter was of considerable interest, because it threw light on several phenomena, d s c u l t to explain, yet frequently met with in practice; and further, because all the work m connection with it had arisen out of the investigation of the curious reaction between butter and alcoholic potash, described by .Meesrs. Wanklyn and Fox, which had excited considerable controversy, but which had been shown by him (Mr. Allen) to be only one of a large series of similar reactions.

ISSN:0003-2654    

DOI:10.1039/AN8911600201

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

THE ANALYST. 203 The reading of Mr. T. P. Blunt’s paper on Tabarie’s Process for the Indirect Deter- Mr. A. H. Allen then took the Chair, and Dr. P. Vieth read a paper on :- mination of Alcohol was unavoidably postponed until next meeting. FAT-EXTRACTION AND FAT-CALCULATION I N MILK ANALYSIS. BY DR. P. VIETH. THE practical value of a mathematical formula expressing the relation between specific gravity, total solids, and fat in milk, is now very generally acknowledged by those analytical chemists who pay more than a superficial attention to milk analysis. The more frequently such a formula is proved to be in accordance with the actual facts, the more confidently will it be applied, and the more useful will it become. There has, perhaps, in no other place been made such an extensive practical use of zt formula of this kind as in the laboratory which is under my charge.In support of this assumption, I may remind you that in the bulk of the milk samples examined in my laboratory-about 18,000 annually-the percentage of fat is calculated from specific gravity and total solids. There are, however, gravimetric fat determinations also con- tinuously being made, which allow the results of fat-extraction to be compared with those of fat-calculation. As, until a few months ago, Soxhlet’s plaster process was the one employed, the formula constructed by Fleischmann was used for the calculation. At the elid of the year 1887 I extracted from my labomtory journals 628 analyses of whole and skim milks, in which the amount of fat varied from ten down to one-tenth of a per cent., and in which fat calculated differed from fat found from -0.2 to +0*2 per cent.; the average difference was +0*020, These results I communicated to the Society at the February meeting, 1888.I have now extracted some further figures, which allow a comparison of the results of fat-extraction to be made with those of fat-calculation, and thought it not undesirable to put them on record. For a reawn which will be apparent later on, the analysw were divided into two classes; the first class comprising samples of ordinary, the (THE ANALYST, 1888, vol. xiii,, p. 50).204 THE ANALYST. second samples of skim milk, as follows :- The calculated figures compare with those actually found, Maximum Average Description. Differences. Difference.pear, Number of Samples. 1888 .,. 73 ... Milk ... - 0-2 to + 0-2 ... + 0-001 ,, *,. 143 ... Skim Milk - 0*2 ,, + 0.2 ... + 0.019 1889 ... 55 ... Milk ... - 0.2 ,, + 0.2 ... - 0.004 9 , ... 119 .*. Skim Milk - 0.2 ,, + 0.2 ... + 0.087 1890-91 ... 143 ... Milk ... - 0.4 9s + 0.1 ... - 0.143 ,, ... 207 ... Skim Milk - 0.3 ,, + 0.1 ... - 0.048 The same maximum differences, and an equally satisfactory close average agreement, as observed during the years from 1881 to 1887, were again found in 1888 and 1889 ; with regard, however, to analyses made during the period from January, 1890, to end of April, 1891, the fat calculated falls considerably behind the fat extracted. The explana- tion for this variation is easily found, I n a very able paper read before this Society in the year 1889 (THE ANALYST, 1889, vol.xiv., p. 121), Richmond showed “that the difference between Adams’ method and Soxhlet’s method is the result of, on the one hand, extraction of something, not milk-fat, and on the other, incomplete extraction,” and recommended, as remedies in the one case, using paper previously extracted with acidified alcohol; in the other, fine grinding and re-grinding of the dry mass consisting of plaster and milk solids, and prolonged extraction. The circumstance that an extracted paper was not to be had in the market deterred me from abandoning Soxhlet’s and adopting Adams’ method. In order, however, to ensure a more complete extraction, I altered my way of working the plaster process, in so far that, instead of 10 grammes of milk I dried 5 grammes only on the usual quantity of plaster of Paris, thereby doubling the surface over which the milk solids were spread.The result was that-looking at the average differences-the practically entire agreement disappeared, and the fat calculated fell con- siderably short of the fat actually found. This fact i g more particularly noticeable in the case of ordinary milk samples, less so with regard to skim milks, proving that the differ- ences between the plaster process, as worked in former days, and when worked with ad&timml pwai-~tims~ m e increasing with the increase of fat. The =me observation has repeatedly been made with regard to the plaster and paper processes. Why that should be 50 is still an unexplained mystery to me, for I cannot accept Richmmd’s statement that the difference is only an apparent one, as of the total fat present, a higher percentage is left behind in the case of skim, a lower in the case of ordinary milk.The question is, in my opinion, not what proportion, but what actual quantity of fat is left behind, and it appears strange to me that this quantity should be lmger where the conditions for extraction undoubtedly seem more favournble. Since a fat-free paper has been introduced I have discarded the plaster and adopted the paper process for fat extractions. Results obtained by using the last-named process mustil of course, be compared with figures calculated from Heber’s formula. Such aTHE ANALYST. 205 comparison has been possible in a number of analyses made from beginning of May to end of September of the present year, with the following results :- Number Maximum Difterencen.Year. of Description, Average Difference. 1891 ... 27 ... Milk ... - 0.1 to + 0.2 ... + 0-030 Y Y ... 70 ... Skim Milk - 0-1 ,, + 0.2 .,. + 0.060 I may remark that in a considerable number of analyses the agreement is absolute, or, speaking more correctly, the differences are below 0.05 per cent. When the paper process was first brought out, it was stated that the greater facility which it offers for the complete extraction of the fat is due to the fact that the solid residue of the 5 grs. of milk used for the fat determination is spread over a surface of upwards of 5,000 square centimeters, This calculation is based on the assumption that the solids are uni- formly distributed over all the fibres constituting the paper.But, already in the early days of the process, several observers, and later on, Richmond, in his paper to which I have referred, have drawn attention to the fact that the blotting-paper exercises a certain selective or separating action. Milk serum, i.e., milk minus fat, must not be considered a simple solution of various bodies in water. Part of the salts, the milk-sugar, and part of the proteids are certainly dissolved in the water in the usual sense of the term, but far the greater part of the proteids-the casein- in connection with the rest of the mineral matter is present in a kind of swollen state, resembling, but not identical with, solution. As far back as 1877 Professor Lehmann, of Munich, has made use of this for the quantitative determination of casein and fat.He proposed to pour a weighed quantity of a mixture of equal parts of milk and water on a porous clay plate, which would soon suck in the water with everything dissolved in it, leaving behind a pasty layer of casein and, enclosed in the latter, the fat. This mass was to be carefully removed while still damp, dried, and extracted, and fat and casein weighed. In the paper process, the blotting-paper takes the place of the clay plate ; the watery solution sinks into it, and the casein encasing the fat is left on and near the surface. according to this theory, the fat is left in contact with only about one-third of the non- fatty solids, while about two-thirds, including the whole of the milk-sugar, are removed from it.That? under such conditions, extraction of the fat is made easier can be readiIy imagined. The practice of fat-extraction has received such a large share of attention lately, that I thocght a few words on its theory Eight be found of some interest. Whether, with the introduction of the paper process, or some other method giving similar results, the question is definitely closed, I am not prepared to say. In my paper ‘‘ On the relation between specific gravity, fat, and total solids in milk,” read in February, 1888, I stated that formula like Fleischmann’s or Hehner’s can be applied with satisfactory results, if ash, proteids, and milk-sugar are present in their normal relative quantities, namely, roughly speaking, in the proportion of 1 : 5 : 6.AS these figures have been repeatedly quoted, I should like to take this opportunity for stating that, according to my experience, and not roughly, but speaking as correctly as This action is, I believe, only imperfectly understood.206 THE ANALYST. the case admits, normal milk contains ash, proteids, and milk-sugar in the proportion of 2 : 9 : 13. DISCUSSION, The Chairman (Mr. A. H, Allen) said that Dr. Vieth had had so extensive and varied an experience in the analysis of milk and milk-products, that anything he might say thereon must command the utmost attention of the members. The figures brought forward on this occasion were of value, as showing the great variations in the estimat,ion of fat by the processes referred to, and the extent to which such estimations were to be trusted.It seemed to him, however, that the processes dealt with in the paper before them had lost the importance they once had in daily practice, and at the present day possessed only a historical interest, Much time and ingenuity had been spent in devising methods for extracting fat perfectly from milk and other bodies. They had endeavoured to attain that end by the use of plaster of Paris, sand, blotting paper, cotton-wool, etc. ; but the Gordian knot had been a t length cut, when Werner Schmid brought forward his method of estimating the fat by boiling the milk with hydrochloric acid and shaking out with ether, For his own part, he did not believe that anybody who had once fairly tried that method would ever use paper coils or plaster of Paris again for that purpose.The experience of those who had adopted Schmid’s method was so satisfactory that all other fat extrac- tion processes had been abandoned by them ; and he could not but think that if Dr. Vieth would make a comparison between the two processes dealt with in his Paper, and this third method, he would agree with others that the former, after having done excellent duty in the past, were now either obsolete, or would soon become so, owing to the advan- tages possessed by the Schmid method. Mr. Bodmer agreed with Mr. Allen’s view, that when once an analyst had got into the way of working,the Schmid process he would never again use a paper coil, or anything else, for fat estimation. I n this connection, he would like to mention that he had recently applied the Schmid method to a sample of milk which struck him as looking somewhat yellow.On adding the hydrochloric acid, the milk turned a beautiful rose-pink colour. But for the Schmid process he would probably have had no evidence of this peculiarity. On attempting to extract the colouring matter, it became clear that the rose-colour was developed by a mineral acid only, and was obstinately retained by the casein; for, on adding a few drops of acetic acid, and thus curdling the milk, it remained colourless, and the sertim did not re-act with hydrochloric acid. On adding hydrochloric acid, however, to the separated casein, the rose-colour was produced ; the subsequent addition of an aikali discharged the red colour. No doubt the colouring matter was methyl-orange or some similar body.Ms. Stokes agreed with Mr. Allen that Dr. ’Vieth’s Papers were always interesting and valuable, the reason being that they were always based on actual facts. He had been specially interested by the figures resulting from the comparison of the two methods which the author had dealt with. He had made thousands of such comparisons himself, and his experience was that if the samples ran in an ordinary series, showing say, fromTHE ANALYST. 207 5-0 to 0.5 per cent. of fat, then any particular method of calculation would agree with the results obtained. But when higher or lower results were obtained, no process that he knew of would always agree with them closely. With regard to paper coils, etc., they might still be retained and sometimes used, but only, he thought, as curiosities, or as a means of checking the results got by the Schmid process.There could be no doubt as to the advantage of the latter over all other methods for the estimation of fats, Mr. Stokes then drew upon the blackboard and described an apparatus (a full description of which, he said, was to be found in the current number of the JoumaZ ofthe Chemical Xociety"), by means of which the inventor claimed that the Schmid process could be actually performed in 30 minutes. He also exhibited some specimens of flat glazed porcelain dishes, which he had found especially suitable for milk analyses in place of platinum, These differ from the ordinary flat porcelain dishes in having every part, including the bottom, glazed; so that no dirt can stick to them.They were specially made for Messrs. Townson & Mercer. Mr. Harvey admitted that he had not yet tried the Schmid process, chiefly because he was averse to pipetting an ethereal solution at the ordinary temperature of the laboratory. It seemed to him that such a practice was likely to result in not getting accurate results due to the loss of ether. He would be glad to know whether it was practically possible to work the Schmid process so that the whole of the ether employed could be evaporated, and on weighing, the whole of the fat obtained. Mr. Stokes reminded the last speaker that Mr. Allen had, at a previous meeting, described a wash-bottle by means of which the ether could be drawn off and used over and over again without loss, it being necessary, however, at each operation to wait for the ether to settle, With regard to the difficulty of pipetting ether referred to, he would guarantee to carry ether in a pipette the length of the Lecture Room without loss, pro- vided a little fat were present.The Chairman, in proposing a vote of thanks to Dr. Vieth for his Paper, took the opportunity of saying what he had omitted from his former remarks, namely, that although the method in question had originated with Werner Schmid, they were indebted to Mr. Stokes for bringing it before the notice of that Society, and advocating it on many occasions. He in no way intended to impugn the paper-coil and plaster methods, both of which he believed were excellent in their way, and in certain iaboratories might be found more convenient than the Werner Schmid process; but in his experience the last was quite as accurate, more rapid, and less troublesome.Mr. Harvey had probably not been present at a previous meeting when he (Mr. Allen) described an arrangement for avoiding the necessity of taking an aliquot part of the ether. Mr. Allen then made a sketch, and gave a description of the apparatus constructed by Mr. Chattaway for the purpose of drawing off the ethereal liquid instead of pipetting it in the usual way, With regard to the dishes exhibited by Mr. Stokes, he might say that he had often used dishes of the same shape for many years past for evaporation in preference to platinum. He * (See abstract in this Journal, page 209.)208 THE ANALYST.found that no sensible alteration of weight occurred; the basins always dried back to their original weight within a milligramme, and could be ignited with perfect safety, They had this advantage, the weight and other characters could be written on them with black-lead pencil. The dishes he had used, however, were unglazed externally, and he was glad, therefore, t o learn that they could now be obtained enamelled both inside and outside. Mr. Bernard Dyer enquired whether such dishes were suitable for ash determinations ? Mr. Allen replied that they were. Mr. Stokes preferred to use platinum for that purpose, as he found that with porcelain dishes the operation took too long a time. Dr. Vieth, in reply, said that there was only one point to which he need refer.Mr. Stokes has said that in the case of samples containing more that 5 per cent. of fat he would not expect any formula to give results agreeing with the figures actually found. While he admitted that that was true to some extent, he also believed that in such cases duplicate analyses would not be found to agree so well as Mr. Stokes would wish. The following description of two forms of apparatus for the removal of an etherial layer, without loss, has been kindly forwarded by Mr. Allen, and will be read with interest in connection with the preceding paper, They are from an advance sheet of the third volume of the new edition of Mr. Allen’s Commercial Organic Analysis, which is now in the press :- b B r/ A I Fig. 1. Fig. 2. When the volume of fluid treated with immiscible solvent is small, the syringe pipette shown in fig. 1 may be conveniently employed.It is readily con- structed by drawing out a test-tube, so as to form a narrow prolongation, the orifice of which should be turned up so as not to disturb the liquid in which it is immersed. A narrow test-tube, fashioned into a handle a t the upper part, serves as a piston, a short length of india-rubber tubing uniting it to the outer tube, while allowing easy movement up and down, 6 b Another convenient form of separator, devised by W. CEattaway, is showri in fig. 2, It is practically a small wash-bottle fitting, which is adjusted to the tube or cylinder containing the layers of liquid it is desired to separate. It is so arranged that the exit- tube (B) can be adjusted in height by sliding it through the india-rubber collar C, so as to bring the turned-up end just above the junction of the two layers.On then blowing through the side-tube (A), the upper stratum is forced up the inner tube, and can be removed, almost to the last drop, without disturbing the lower layer.’’Apparatus for the Estimation of Fat in Milk, E. Molinari (Berichte Deutsch. Chew. GeseZZ. 24, p. 2204). The author, after criticising the various methods for estimating fat in milk which have been proposed from time to time, agrees with Stokes (ANALYST, 1885, p. 48), Eustace Hill (ANALYST, 1891, p. 67), and Bondzynsky (Land- wirth Jahrb. der Xchweiz, 1889), that the method of Werner Schmid is the simplest, most rapid, and convenient hitherto introduced. The conditions tending to inaccuracy are :-the employment of ether containing alcohol ; boiling the mixture of milk and acid too long, when a caramel-like body is formed, soluble in ether ; the difficulty of reading off the volume of ether left in the tube, owing to the gradations of the instrument being obscured by the flocculent layer of casein ; when onIy a portion of the ether is used, fat may be left behind in the acid mixture, as shown by Allen (Chew.Zeit., 1891, p. 331). The author believes that by the invention of the simple apparatus represented in the accom- panying figure, he has rendered the process both accurate and convenient, This consists of a flask B of about 75 C.C. capacity, which has a glass tap fused on, with two :capillary tubes attached, the one passing upward the other downward.The neck of flask B is ground into the neck of flask A, which holds about 90 C.C. Either of the flasks can be placed in communication with the external air by the opening a, The ether must be pre- viously washed with one or two tenths of its volume of water, to remove traces of alcohol. The operation is performed as follows:-10 C.C. of well-mixed milk are weighed in (or measured into) flask A, 10 C.C. of hydrochloric acid added, and the mixture heated to boiling on an asbestos sheet. The boiling must not exceed a minute and a half, the fluid being shaken from time to time, and not allowed to become of a deeper colour than a dark-brown [not black]. The flask is cooled, and 25 C.C. of ether added. The two flasks are connected as shown in the figure, the tap dosed, and the whole shaken for a few minutes, the flask being vented two or three times by the opening a.The apparatus is now inverted, allowed to stand five or six minutes, the tap turned, and the dark acid liquid drawn off into flask B. By a little shaking of the ether the whole of the acid liquid may be easily got into the lower flask, The apparatus is again inverted, then separated, 10 C.C. of ether are intro- duced into flask B, the tap closed, and the fluids well shaken. When the ether layer is distinct, the acid liquor is run off, and the ether solution transferred to A. The whole of the ether solution is washed in the apparatus two or three times with a little water, the flask A removed to the water-bath, the ether driven off, the last traces of ether and water being removed by placing the flask in a drying oven heated from 107 to llOo.C., where it must remain at least 20 minutes, The usual cooling in the exsiccator and weighing concludes the operation, Examples are given showing its con- coxdance with the Adams' and other recognised processes. Sour milk, which must be210 THE ANALYST.weighed in the flask, can be conveniently analysed; also cream, using 5 grms. cream and 10 C.C. hydrochloric acid. W. J. S. Determination of Citric Acid in Wine. A. Klinger and A. Bujard (Zeitsch. f. angew. Chem. 1891, p . 514).- I n testing wine in accordance with the Nessler- Barth method, a sample of wine yielded a precipitate of the calcium salt which, on being further tested, was found to be the malate, and not citrate.On investigating the matter, the authors found that malate of calcium is precipitated by boiling from its nearly neutral solution in the same manner as citrate. To distinguish between the two salts, the following method is recommended : At least 250 C.C. of the wine t3 be tested are evaporated to about one-third of its bulk. Potassium acetate is then added, the solution strongly acidulated with acetic acid, and a double bulk of strong alcohol added. This throws down the tartrate, After twenty-four hours the precipitate is filtered oE, washed with a few C.C. dilute alcohol, The filtrate is precipitated with basic lead acetate, the precipitate being also washed with alcohol, it is then decomposed with hydrogen sulphide. The filtrate from the lead sulphide is concentrated and made alkaline with lime-water.The precipitate, which consists of phosphate and a trace of tartrate, filtered off after a few hours. The solution is rendered acid with acetic acid, and evaporated to dryness. The residue is dissolved in a little hot dilute hydrochloric acid, and after addition of some ammonium chloride, is made faintly ammoniacal, and boiled for some length of time, when only citrate will be precipitated, and no malate. 0. H. The use of the principle of the Separator in Analytical and Micro- scopical Work. W. Thorner (Chem. Zed. 1891, 15, 1201-1203$.-The author has experimented with the separator as a means for the analysis of meal and butter; an extension of its familiar use for milk in the case of the lactocrite.W e instrument chosen was that known as the ‘‘ Victoria,” by Watson, Laidlaw & Co., of Glasgow. It is arranged to give a speed of 6,000-7,000 revolutions per minute easily. (1.) Experiments with various meccls and starches. One gramme of the material to be tested is shaken with a suitable liquid-e.g., water, alcohol, or ether, in a tube, the lower part of which is very narrow, and divided into & C.C. The capacity of the whole tube is about 15 c.c., aad that of the narrow part 3 C.C. The tube is filled with liquid to within about 2 cm. of the upper edge, and well shaken, to mix the substance to be examined with the liquid. It is then placed in the cenirifugal machine, a plate 350 mm. in diameter being used, and driven at 2,000-2,500 revolutions per minute for five minutes, TheTHE ANALYST. 21 1 following numbers are the mean of several concordant results obtained by this method :- Liquid used.Water, Alcohol. Ether. Percentage of Water. Material. Wheat flour I. ?, 11. ,, 111. 9 9 11. Rye flour I. Oat flour Barley flour Pea flour Buckwheat flour Rice flour Rice starch Potato starch Mondamin 14.6 14.6 14.6 14.0 14.0 11-9 12.5 17.6 14.8 13.8 15.0 11.1 14.1 14.3 14.5 15.3 30.0 30-0 30.0 22.0 22.4 19.B 19.8 12.6 13.3 13.6 16.6 16.5 16.0 16.4 16 0 17.8 16.5 20.6 15.8 14.2 15.8 11.0 12.0 15.2 15.6 15.0 16.0 15.7 17.0 14.0 18.0 15.5 13.9 15.8 12.0 12-4 The numbers in the last three columns are the tenths of a centimeter occupied by the flour after being subjected to centrifugal action. The results got by the use of alcohol and ether are too similar for all the various materials tried to make the use of these liquids valuable, except when it is desired to make several tests under different conditions upon the same sample.With water, on the other hand, the Werences are sufficiently considerable to allow of the different flours and starches being readily distinguished, Especially is this the case with rye flour (see table) -a fortunate circumstance, as it is frequently used as an adulterant of the more costly wheat flour. The following table shows the results obtained with known mixtures of wheat and rye flour :- Pure wheat flour 9Q 'lo wheat flour 10 o/o rye flour 80 9 9 20 9 ) 70 9 , 6 0' 9 9 40 9 9 50 9 ) 50 9, 40 7 ) 60 9 , 30 2 9 70 9 9 20 Y 9 80 ?> 10 7, 90 9 9 30 t' ...... ... 0 . . ... ..* me. ... ..a ... .I. ... .*. ... ... ... 0 . . I.. ... ... Height in & C.C. ... 14.3 ... 15.2 .. 16.3 ... 17.7 ... 18.6 ... 20*4 ... 21.2 ... 23.2 ... 26.0 ... 28.5 The wheat flour contained 14.6 per cent. of water, and the rye flour 14.0 per cent, I& order to avoid fluctuations, due to cWerent amounts of water, it is preferable to conduct212 THE ANALYST. the experiments with samples dried at 100-102° c. to constant weight. The results do not, however, differ greatly from those obtained with the undried samples, as is shown by the following figures :- Liquid used. Material. Water. Alcohol. Ether. Wheat flour I. ... Rye 9 9 I. ... Barley ,, ... Pea ?? Buckwheat flour Ground rice 0 . . Rice starch ... Potato ,, 0..Mondamin ... Oat ,, bmb ... a * * ... ... b.. ..I *be 0.. ... 0.0 15.4 33.0 30.0 22.8 25.3 22.8 23.8 15.0 16.8 15.3 18*0 19.0 18.5 17.5 18.3 17.0 17.6 17.0 12.0 13.7 15.2 17.0 16.2 16.2 16.5 16.0 14.0 16.0 12.0 13.7 The results, as before, are given in tenths of a C.C. Mixtures of wheat and rye flour gave the following results :-- Wheat flour. 90 80 70 60 50 40 30 20 10 100 O l o Rye flour. - O/O 10 20 30 40 50 60 70 80 90 100 Water. 15.4 17.5 18.0 20.0 21-5 22.9 240 25.5 28.0 30-5 33.0 Liquid wed. 90 o/o water and 10 "I, alcohol, 15.2 16.0 17.0 18.0 19-6 22*0 23.2 24.0 25*0 27.0 30.0 These figures confirm those given above, and indicate that a determination of the percentage of rye flour in wheat flour containing this adulterant is quite practicable, The process would probably give equally definite indications with oat-flour, Instead of weighing a portion of the sample, sufficient exactitude could probably be obtained by measuring it in a little g l w vessel with a ground cover.This is a suggestion which the author has not yet verified. The presence of mineral adulterants, such as chalk and barytes, can be detected by the same process, the volume occupied by the flour diminishing as the perceatage of mineral matter increases. A neater method consists in using chloroform instead of water as a separating liquid, and after shaking, adding 4 - 1 C.C. of water, shaking again and revolving in the separator, almost complete separation takes place, the flour floating onTHE ANALYST. 213 the chloroform layer, and the heavy mineral adulterant sinking to the lower graduated part of the tube, where its bulk can be read off'.The microscopical examination of meal and flour, and the isolation of the husk residue from them, can be aided by the use of the separator. 5 grms. of the sample are treated with 500 C.C. of water on the water-bath. 100 C.C. of the mixture are then heated for ten minutes on the water-bath with 5 C.C. of glacial acetic acid, and another portion of 100 C.C. similarly treated with 1 C.C. of caustic potash solution (1 : 2). Both are then revolved for five minutes, and the residue at the bottom of the tube separated and examined under the microscope. (2). Application OJ the method to the analysie of butter. A quantity of the butter to be tested is measured by means of a glass cylinder, closed at each end by a glass plate ground to fit.The cylinder is just small enough to enter the wider and upper part of the separator tube, so that its contents (10 C.C. ~ 9 . 5 grms. of normal butter) can be trans- ferred thereto. The separator tube is revolved on a plate 360 mm. in diameter for 2-3 minutes a t a speed of about 2,000 revolutions per minute. The matter, other than butter fat, collects in the narrow, graduated, lower part of the tube, generally in two layers, the upper consisting of water in which some casein is suspended, the lower also of water, but almost clear, save for such insoluble mineral matter, flour, and starch that may be present. Crystals of salt may also occur. The total watery layer represents the amount of butter milk in the sample.The whole operation can be completed in ten minutes. The separator tube differs from that previously described, only in having an enlargement on its upper part, so as to support it in the water-bath during the introduc- tion of the fused butter from the measuring cylinder. The insoluble fatty acids can be determined by measuring 1 C.C. of butter-fat a t 100 C.C. into a separator tube with the middle part contracted, and graduated to & C.C. saponi- fying therein, driving off the alcohol, acidulating, revolving in the centrifugal machine, and reading off the volume at a temperature of 100 C.C. The data thus collected, together with the refractive index determined in the usual way, suffice to arrive at n fair judgment of the quality of the sample with great expedition. B.B. Method for the determination of the Fat in Sour Milk by means or' the Lactocrite of De Laval. M. Ekenberg (Chemiche. Zeit., 1891, 15, 1239).-In carrying out the examination of milk it often happens, in summer time, that the samples become sour during transport, in which case it is a matter of extreme difficulty to obtain a portion for analysis representative of the whole. When decompo- sition has proceeded so far that the milk is already coagulated, it is utterly impossible to mix tbe layer of cream (the fat) uniformly with the whey. I n the laboratory of the Actiengesellschaft Separator, a t Stockholm, various investigations have been made far some time since to apply the lactocrite to the examination of sour milk, A reliable method for analysing sour milk has hitherto been lacking.214 THE ANALYST.whereby it was proved that sour milk could be made to give results as certain as those. got with fresh milk, in that thoroughly uniform samples could be obtained and accurate measurement was possible. Seeing that, as has been already noted, it was impossible to mix the fat with the rest of a sample of sour milk merely by shaking, a substance was sought which should be without action on the fat, but capable of restoring the milk t o its normal consistency. "he result of the experiments in this direction was, that if ordinary ammonia (5 per cent, by volume) was added to the sour milk and the mixture shaken, it became quite fluid, and the fat could be determined in the same manner as in an ordinary milk.On exami- ning a sample immediately after treatment under the microscope, it was seen that the fat globules were equally disseminated throughout it. This was further substantiated by a series of experiments with the lactocrite, using the new '( lactocrite acid," i.e., lactic acid mixed with hydrochloric acid. The method of conducting the experiments and the results given by them were as follows :- 100 C.C. of whole milk were measured out into each of four clean dry flasks. The fat, as determined by 6 tests with the lactocrite, amounted to 3.22 per cent. The flasks were numbered, well closed, and kept completely free from disturbance at a temperature of 18" to 23OC. After a further period of 24 hours 5 C.C. of ammonia were added to the contents of flask No.1, and the mixture well shaken, a thin fluid product, from which portions for testing by the lactocrite could be easily measured off, being obtained. Six tests were made and gave the following results :-3.00 per cent., 3.00 per cent 3.01. per cent., 3-02 per cent., 2.99 per cent., and 2.99 per cent., the average being 3 00 per cent., corresponding to an amount of fat in the milk of 3.15 per cent.-that is ++8 of the amount of fat given by the sample to which ammonia had been added. After 5 days flask No. 2 was examined in the mme way as No. 1. The mixture in this gave by the lactocrite test a content of fat of 2.90 per cent., 2-90 per cent., 2.90 per cent., 2.90 per cent., 2.92 per cent., corresponding to an amount of fat in the milk of 3.05 per cent.Flask No. 3 gave, after the elapse of 7 days, 2.84 per cent., 2.84 per cent,, 2.85 per cent., 2-85 per cent., 2.83 per cent., 2-84 per cent,, an average of 2-84 per cent. corresponding to a percentage of 2.98 per cent. of fat in the milk. Flask No. 4 was examined after 9 days, and gave 2.82 per cent., 2-83 per cent., 2.84 per cent., 2.83 per cent., 2.82 per cent., an average of 2-83 per cent., corresponding to 2-97 per cent. of fat in the milk. On comparing the results of 5 or 6 determinations of the fat in the mixture of milk and ammonia in each flask among themselves, it is seen that the layer of cream which had separated previous to and during the souring of the milk, had become thoroughly mixed by shaking, after the addition of 5 per cent. of ammonia. On com- paring the results of the different flasks with each other, it is evident that the longer the sour milk stood, the lower was the fat as determined by the lactocrite.That the fat decomposes on keeping the milk is well known, and this fact explains why the result of a (See THE ANALYST, 1891, p. 135.) After 24 hours the milk in all the flasks was sour and coagulated.THE ANALYST. 21 6 determination with the lactocrite is lower as the age of the sample increases. The m n e r the sour milk is treated with ammonia, and tested with the lactocrite, the better the result. When the samples are only 1 to 2 days old-and older samples should not often have to be examined-this tendency is of negligible importance, as is shown by the foregoing figures. Further experiments have proved that the diminution in the result of the determi- nation of the fat in sour milk which was two days old, by means of the lactocrite, averaged 0.05 per cent., wherefore this number should be added to the percentage found.Thus the percentage of fat found in the foregoing investigation in flask No. 1 was 3-15 per cent., which, when increased by the addition of 0.05 per cent., becomes 3.20 per cent., while the fresh milk had, as already noted, 3.22 per cent, of fat. As no perceptible ill effect is to be feared from the addition of 5 per cent. of ammonia, especially when the sample is examined by the lactocrite immediately after the addition, the plan of adding ammonia to sour milk is much to be recommended. I n order to obtain a reliable result, the volume of the sample to be tested must be ascertained before the addition of the ammonia, because after the milk has become sour, the cream separated, and the milk coagulated, no average sample for admixture with 5 per cent.of ammonia can be measured off. The ammonia must be added to the sample in the vessel in which it has become sour. B. B. The Analytical Use of Barium Dioxide and Hydrogen Dioxide. E. Donath (Chem. Zed., 1891, 15, 1085, 1086).-The author, after claiming priority in the use of barium dioxide in attacking chrome iron ore, which he considers has been erroneously ascribed by Baumann to Kinnicut and Patterson, enumerates some of the advantagesit possesses. The ore is completely attacked if about ten times its weight of the dioxide be used, even when heated only over a bunsen burner, provided the dioxide be of good quality.The mass does not fuse, and is therefore more easily removed from the crucible than when sodium carbonate is added, as by Kinnicut and Patterson ; moreover, the use of the alkali causes a portion of the chromium to remain as chromic oxide, which has to be oxidised to chromate by hydrogen peroxide. The author has also used the dioxide with sucoess ir?. attaching slags, lead-glaes, and clays, as well as minerals containing chromium. The usual method of determining the chromate formed from chrome iron ore by any process of oxidising attack with ammonio-ferrous sulphate, and estimating the excess of the latter by permanganate, is, in the author’s opinion, inferior to the plan of determining it by titration of the iodine liberated by its action upon potassium iodide in hydrochloric acid solution with sodium thiosulphate.Hydrogen dioxide is an excellenf oxidant for ferrous salts in acid solution, and excess is more easily removed than is the case with such substances as potassium chlorate, nitric Commercial samplw leave much to be desired in this respect,216 THE ANALYST. acid, or bromine. It can be used with good efl’ect for getting rid of ammonium sulphide without separation of sulphur, and for oxidising an ammoniacal solution of sulphide of arsenic (such as is obtained in the separation of antimony, arsenic, tin, etc.) to ammonium arseniate, prior to precipitation with magnesia mixture. The oxidising effect of hydrogen dioxide in alkaline solution is comparatively weak, being less energetic than that of bromine or potassium permanganate. Thus, pickel hydrate is quickly oxidised by bromine to the peroxide, but is unaffected by hydrogen dioxide ; in this respect the latter reagent is similar to iodine.B. B. On the influence of the so-called Nitrogen-free Extractive Matter on the Result of the Estimation of Starch in Cereals. C. J . Lintner and G. Dull. (Zeit. f. angew. Chem. 1891, p. 537)-The authors show in a paper (1. c. p. 538) that the gum-like body found in barley some time since by Lintner, yields upon hydrolysation Galactose and Xylose, the former being fermentable with great difficulty, the latter unfermentable. Both re-act with Fehling’s solution, and therefore interfere with the accuracy of an ordinary starch determination. To ascertain the amount of error caused by the presence of this body, a number of experiments were made; 3 grammes of the finely powdered substance were heated for 3 hours with 75 C.C.of water in a digester at a pressure of 3.5 atmospheres, filtered through glass wool, 15 C.C. hydrochloric acid s.g. 1.125 added, and the whole made up t o 200 C.C. with water. This was inverted in the water-bath for 3 hours, neutralised, made up to 500 c.c., and a sugar estimation made on 25 C.C. by Allihn’s method. For the calculation dextrose to starch, the factor 0.9 was used. The remainder of the solution, after concentration, was fermented with a little pressed yeast, made up to 100 c.c., and the reducing matters determined in an aliquot part by Allihn’s method.The fermentation removes the whole of the dextrose, a small portion of the galactose, and leaves the xylose intact. In this way the following figures were obtained :- I. 11. Starch Unfermentable reducing matters reckoned as Starch. Barley 56-86 p.c. 2.7 p.c. Wheat 56.27 p.c. 4.0 p*c* Potato Flour 65.03 p.c. 2.0 p.c. Maize 62-50 p.c. 1*6 p.c. If the figures in column [I. are deducted from those in column I., the difference will be the amount of starch actually contained in each sample, i.e., provided the factor employed (0.9) is the correct one. Sachsse (Chem. Cent. 1878, p . 732) states that the relation of dextrose to starch is not 100 : 90 but 108 : 99, giving the factor 0.917; Soxlet (V. Brawr, 1886, p. 193) after much investigation (and probably working under slightly different conditions), gives 0.94.The authors working in the following manner, come toTHE ANALYST. 217 the conclusion that 0.941 is the correct factor; 3 grms. of starch, 15 to 20 c.c, hydrochloric acid 9.8. 1.125, water 200 c.c., time of inversion 3 hours. Special strew is laid upon the point, that the mixture should be well shaken to ensure thorough diffusion of the starch through the fluid, otherwise a yellow colour appears, showing partial destruction of sugar. Consequently, to arrive at the truth as near as possible, the difference between mlumns I. and 11. should be multiplied 0.94. The figures then respectively stand 66.67, 54.59, 63.18, and 66.32, and do not materially diEer from those of column I. Hence, by neglecting the influence of the gum, and employing a slightly inaccurate factor, the two errors nearly counterbalance one another; and for the estimation of starch in cereals this method may be employed.I n the case of bran, brewers’ grains, &c., which contain large amounts of nitrogen-free extractive matter, the result would be very inaccurate. If the estimation be one of a pure starch, then the factor 0.94 must be employed, W. J. S. The Determination of Lead in Tin Plate Vessels used for Canned Goods. J. Pinette (Chem. Zed. 15,1891, 1109).-!I!he exact limit for lead in the covering of tin plate used for vessels containing preserved food, and in their soldered mams, being prescribed by law in Germany, its determination is of some moment. Although no such regulation exists in this country, the intrinsic importance of the question is as great.The plan suggested consists in avoiding the troublesome process of scraping off the tin from the sheet iron, and in simply acting upon fragments of the vessel with dilute nitric acid until the iron is laid bare and determining the tin and lead in the solution, the latter being dissolved as nitrate, the former chiefly remaining as meta- stannic acid, As it is only of importance to ascertain the ratio of tin to lead this plan suffices ; it assumes, however, that the tin plate is coated with the same alloy on both sides. The fact that some iron is also dissolved is of no moment. The examination of the solder presents no di%culty, but it must be remembered that its composition may vary at different parts of the same vessel, that used for finally closing the ‘6 tin ” after sterilization being usually richest in lead on account of its greater fusibility1 and conse- quenteasein working.Acanof American fruit was tinned with an alloy containing 0.55 per cent. of lead ; its solder under the lid contained 50’84 per cent. Another of corned beef gave imponderable traces of lead in the tinning, and 60.06 per cent. of lead under the cover. B B. The Origin of Citric Acid in Milk, A. Scheibe (Landw. Vermchatat, 1891, 153, through Chem. Z&.).-With ordinary fodder citric acid occurs in goat’s milk as in cow’s milk to the extent of 1-105 grammes per litre, but the amount is liable to considerable fluctuations, being sometimes double as great as a t others (calculating it upon the total mlids).It doea not appear to be derived fmm the citric acid present218 THE ANALYST. ready formed in the fodder, as it also occurs, though in smaller quantity, in human milk. Moreover, citric acid added to the fodder does not increase the amount found in the milk, and feeding exclusively upon materials free from citric acid, has no effect upon the quantity excreted. Milk secreted in a state of hunger contains the normal amount of citric acid. Citric acid in milk is apparently not derived from the cellulose undergoing digestion in the intestines of herbivorz as excretion goes on when food is withheld or meal substituted for ordinary fodder; this is consistent with the fact that in the human subject (where no digestion of cellulose takes place) 094-0.57 grammes of citric acid per litre of milk is formed. B.B. Citric Acid as a Normal Constituent of Milk. T. Henkel (Aandw. Versuchsstat, 189 1, 134, through Chem. Zed.).-The author’s investigations as to whether citric acid occurs normally in cows’ milk, or is formed from other constituents during the process of isolation, prove that its presence is not adventitious or occasional, but normal and constant. 13, B. Testing Metallic Iron for Arsenic. 0. Sautermeister (Chern. Ze& 15 1891, 1021, 1022.)-The author examining various samples of iron for arsenic by the method prescribed in the German Pharmacopoeia, which consists in dissolving 1 gramme of the metal in dilute hydrochloric acid and testing the hydrogen evolved by kindling it and allowing the flame to play upon a porcelain dish, as in Marsh’s test, found that no reaction was given even when as much as 0.1 gramme of As40, was purposely added. As a copious deposit was obtained when zinc was substituted for iron, it was evident that the latter had some specific action preventing the evolution of arseniuretted hydrogen.An examination of the residue unattacked by the acid showed the arsenic to be present there, having been reduced to the metallic state. The phenomenon was noticed by Wohler and recorded by him in the AnmaZen, der Pharmacie in 1839, but has apparently been since overlooked. The test can be amended and made useful by the addition of zinc (about 5 grammes of zinc to 2 grammes of iron), 1 milligrammeof arsenic being then detectable. The arsenic does not come off at first, being only evolved when the liberation of hydrogen is taking place freely.An attempt to devise a method for the separation of arsenic and antimony, based upon the tendency of iron to precipitate the former metal in an acid solution, was frustrated by the fact that antimony behaves in exactly the same way. That the author’s observation is not superfluous is proved by the detection of amenic in samples of iron, which would have been wholly overlooked by the stereotyped method, B. B,THE ANALYST. 219 Preparation of Hydrogen Peroxide. L. Crismer (Bull. XOC. Chim., 1892, 3, 24, through Chent. Zeit.).-The process depends on the well-known property of hydrogen peroxide of dissolving in ether. Commercial barium dioxide containing 85- 90 per cent.BaO, is dissolved in a slight excess of hydrochloric acid of sp. gr. 1.10, and the solution shaken with an equal bulk of ether. After separation the ether is shaken with a little water, which removes the hydrogen peroxide which it had taken into solution. After separating the aqueous layer, the ether is again shaken with the same solution of barium dioxide, and once more extracted with water. After 5-6 operations the extraction is complete. A solution of hydrogen peroxide containing 8-9 per cent. H202 perfectly neutral, and free from solid matter may be thus obtained. B. B. HubI’s Iodine Absorption Method. Holde. (Mitt. Konig. tech. Persuch. 1891, 9, 81, through Chm. Zeit.).-The author endeavoured to ascertain the causes of, and remedies for, the discrepancies sometimes observed in determining the iodine absorp- tion numbers of oils.As the result of his experiments he recommends the following procedure :-0*3 gramme of a non-drying oil or 0.2 gramme of a drying oil is weighed out and dissolved in 18-20 C.C. of chloroform in a flask (presumably stoppered) of about 300 C.C. capacity. 50 C.C. of Hubl’s reagent, not more than fourteen days old, for a non-drying oil, and 60 c.c., not more than eight days old, for a drying oil are then added, (Two tests are made of each oil). The Hub1 solution is allowed to act for two hours; 50 C.C. of the solution are titrated with sodium thiosulphate after the addition of 40 C.C. of 10 per cent. solution of potassium iodide, at the beginning of the determination, and 50 C.C. are kept in a well-closed flask until the determination is completed, and then titrated in the same manner, the mean of the two estimations being taken as the true value. The iodine unabsorbed by the oil is similarly determined after the addition of 40 C.C. of 10 per cent. solution of KI for a non-drying, and 50 C.C. for a drying oil, and 120 C.C. of water. Should the chloroformic solution of the oil become turbid while standing, more of the solvent must be added. The foilowing are some of the r e s u h obtained :-Linseed oil, 172-180 ; hempseed oil, 175-176 ; walnut oil, 139-143 ; sesame oil, 106-109; cotton-seed oil, 110-115; crude rape oil, 100-108; refined rape oil, 100-107 ; arachis oil, 91.2-101*5 ; olive oil, 79-84 ; bone oil, 59.1-81.7, B. B. Note by Abstractor.-It is known to most analysts dealing with oils that a period longer than two hours is desirabIe to emure full absorption, and that the alcoholic solu- tions of iodine and mercuric chloride which constitute when mixed Hubl’s solution, are best kept separate and mixed only when wanted. The great loss of strength observed in Hiibl’s solution on keeping is thus avoided.220 THE ANALYST, Determination of €he Ash in Sugar. Albertiand Hempel (Dez~hd. ZmkeriruE., 1891, 16, 1069, through Chem. Z&t.).-6--7 grammes of coarsely-ground quartz sand are placed in a platinum dish holding about 36 c.c., ignited, weighed, and mixed with 5 grammes of the sugar by means of a thin platinum wire. The mixture is ignited in a platinum muffle for a time, varying from In the case of molasses 3 grammes are mixed with 12-15 grammes of sand. The ash is free from carbonates, and contains the quantity of sulphates and chlorides naturally present in the original sample, as is proved by analyses of commercial products and by experiments on known mixtures, A determination can be made as quickly as by the sulphate method, and involves no assumption as does that process. A disadvantage is that the platinum dish is perceptibly attacked ; gold-plated vessels would probably do better. The moisture can previously be determined on the same portion; the diflerence between the weight of the dried sample and that of its ash gives the weight of the organic constituents direct, without the error caused by the presence of a portion of the carbon being left as carbonate in the ash and reckoned with it. to 14 hours. B. B.

ISSN:0003-2654    

DOI:10.1039/AN8911600203

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

220 THE ANALYST, ERRATA in Mr. Allen’s Paper on the hsay of Aconite, p. 188 :- Line 20, for “grammes ” read “milligrammes.” Line 25, for Line 27, for (‘ 12‘14 milligmmmes of japaconitine ’I read 12.44 milligrammen of j%paconibine saponified." Line 31, for ‘‘ 21’8 against 30 milligrammes taken” read ‘‘ 29’8 against 30 milligrammea taken.” In foot-note, for “767 mgrs. taken ” read (‘ 7-67 mgrms. taken, baryta” read “gbaryta”

ISSN:0003-2654    

DOI:10.1039/AN891160220b

出版商:RSC    

年代:1891

数据来源: RSC