摘要:

THE ANALYST. JUNE, 1891. TO OUR READERS. THE issue of this number marks an epoch in the existence of the ANALYST, because it is the last that will appear under the direction of the present Editor, who now, after nearly six- teen years of office, bids his readem farewell. Those who have followed the Journal from its commencement will be aware that it was originally started in the year 1877, by the Society of Public Analysts, and was at first conducted by the late Mr. Wigner and the present Editor, assisted by a Committee of Publication. After some time, the losses upon its production became too great for the funds of a Society then only limited in member- ship, and consequently the Editors took the paper off its hands and conducted it at their own risk. At Mr. Wigner’a death the property passed entirely into the hands of the present Editor, who has conducted it ever since, and has succeeded in converting it from a commercial loss into a self-supporting publication.This being so, and the end for which he has always worked being attained, he decided to ask the Society t o become once more thecustodians of their Journal and to relieve him of the responsibility both of Editorship and Proprietorship. The Society having thus taken up their own burden he breathes a sigh of relief, but yet cannot help expressing his regret in parting both with his subscribers and with the confreres who have aided him in his self-appointed task of establishing a Journal devoted to the Improvement of Food Analysis. The days of chemical politics are over, but the Editor can only look back with satisfac- tionon the time when the establishment of the Institute of Chemistry on its present basis was actively opposed in these columns.It was then argued by this Journal alone, that the basis of the Institute wm bad, and that if it were to be a real benefit to analytical chemists, its foundation should have been on a nucleus of practising analysts (such as the Society of Public Analysts), and the professorial and trade elements it now containp should have been eliminated. As time goes on the views then expressed have come to be recognised by nearly all analysts, and an effort is being made to bring the Institute more within the lines of their needs, but with what effect remains to be seen, because so many of its present members are in no sense pl‘actising analytical chemists, but simply teachers, and even, in some cases, manufacturers or persons practically engaged in trade who have no real interests in common with the very men for whose supposed benefit it was projected.It is an old Americanism to say that ‘‘ every citizen thinks he can run a journal,” and we hop0 that those who are now about to try their hands on this one will succeed in producing a much more ideal print than it has hitherto been. If they wish to do so, they will have to resist the blandishments on the one side of those who want to alter what they said imprudently, and on the other the threats of libel actions from those who object to the publicatios of what mme one elae aaid at the meetings about them. Of102 THE ANALYST. - - _ _ - -- . both of these sensations the present Editor has had ample experience, even, on one occasion, to the extent of such threats from the then President of the Society himself. A committee has no body to assault and no soul to condemn, so that it will be able possibly to take a more independent ground than can ever be occupied by one man in such a posit ion. With these remarks, the Editor bids his readers farewell, and introduces the new Editorial Committee of the Society of Public Analysts,

ISSN:0003-2654    

DOI:10.1039/AN891160101b

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

102 THE ANALYST. ON THE EXAMINATION OF SPIRITUOUS LIQUIDS FOR SECONDARY CONSTITUENTS. BY ALFRED H. ALLEN AND WILLIAM CHATTAWAY. FOR many months past we have had in progress a series of experiments connected with the chemistry of whisky, having mainly for their object the examination of whisky and other spirituous liquids for secondary constituents, The secondary constituents of spirits are by no means to be regarded in the light of impurities, as they have wrongly been called and considered by some. They are the associated bodies which give the alcohol its special and valued characters, and to their production, modification, or elimination by age we owe the changes which spirits undergo during the process of maturing. The secondary or bye products of spirits are naturally most abundant in iihose spirituous liquids manufactured in apparatus where no, or but little, fractionation occurs.This is the case with the spirit distilled in Bcotland in pot-stills, and made wholly from fermented malt, and in Ireland from a mixture of malted and unmalted barley, with, in some cases, a small addition of other grain. I n the manufacture of whisky by the pot-still, the fermented wash is simply distilled over a naked fire, when ‘( low wines ” is obtained as a first product, the (( pot ale ” remaining in the retort being run to waste. On redistilling the “low wines” the first fraction which passes over is called ‘6 foreshots,” the second (( clean spirit,” and the third feints,” while the residue in the retort is called ‘( spent lees.” The foreshots and feints become milky when mixed with an equal measure of water, and are mixed with the low wines of the next period to (Bead at Meeting, May, 6th) I891 .)THE ANALYST.103 ~ be redistilled. It follows that whisky and spent lees (which is practically free from alcohol) are the only final products of the manufacture of spirit by the pot-still process. No fuse1 oil or other special product is obtained. In the manufacture of spirit by the Coffey and other patent stills, in which frac- tionation is effected, the materials used for the production of the spirit are of a much more varied character. I n the United Kingdom, a mixture is employed of malted and un- malted grain, maize, and rice, with, in some instances, sugar and molasses, only enough malt being used to effect the conversion of the starch to maltose.No potatoes, turnips, beets, or other roots are used in this country for the production of spirits, though such materials are largely employed on the Continent. The source of the spirit is evident to the expert only if imperfectly rectified, the best ‘‘ silent spirit ” affording no indication of its origin. But there is as much, or even greater, difference between the finest silent spirit, by which nearly chemically pure alcohol is to be understood, and spirit made from grain in the Coffey still, as there is between the latter and real pot-still whisky. Pot-still spirit containing the largest proportions of those secondary constituents which give to properly-matured spirits their special value, it follows that the process of ageing and maturing is specially applicable to such products, which are :-Scotch and Irish whiskies made in pot-stills, wine-brandy, and real rum from sugar products.Factitious rum concocted from alcohol and flavouring agents would not come under this category ; nor does patent-still whisky undergo anything like the same improvement by maturing as pot-still whisky. Brandy is a term now very much misused. It was formerly applied almost exclusively to cognac or French brandy, a product obtained by the distillation of wine or grape-skins. The German word Branntowein is now, unfortunately, com- monly translated I‘ brandy,” although in the original language it has a far wider meaning than that we give to the term brandy. The table on pages 110 and 111 shows the names and leading properties of the chief constituents of spirituous liquids.Many of the bodies in the foregoing table exist in spirituous liquids in very minute amount, and their presence has never been detected in the spirit itself, but only in the fuse1 oil obtained by fractionating a large quantity of the alcoholic liquid. According to L. Lindet (Compt. rend., cxii., lOZ), the production of higher alcohols is very slow at first, but increases with the progress of the fermentation, and ccntinues with still greater rapidity after fermentation proper has ceased. The prc- portions of higher a!zohc!s ” (gnery, tntal oily matters) per litre were found in one case to be as follow : 14 hours, 3.64 C.C. ; 20 hours, 4.45 ; 38 hours (fermentation complete), 6-44 C.C.; 62 hours, 9.2 C.C. The formation of the higher alcohols appears, therefore, to depend chiefly upon those abnormal conditions of yeast-life consequent upon the disappearance of the sugar. They may also be produced by some special organism which remains almost inactive in presence of the rapidly-developing and vigorous yeast, but becomes active when the work of the yeast is finished. Hence, the longer the time between fermentation and distillation, the larger the proportion of higher alcohols is likely to be. It is a curious fact that the species of ferment affects the character of the higher alcohols produced. Thus the sccccharomyces oereviskz of ordinary brewers’ yeast produces104 THE ANALYST. subsidiary products differing in certain important respects from those of saccharomyces ellipsoides, which is the ferment of grape-skins.In fact, it has been found that by adding the latter ferment to molasses and other saccharine liquids distinct from grape- juice, the product of the fermentation, after distillation, has all the characters of Cognac brandy. The most marked distinction between the subsidiary products of the two fer- ments is that whereas in the case of the grape-juice ferment normal butyl alcohol results, in other cases this is replaced by iso-butyl alcohol. By the fractional distillation of Cognac brandy, twenty-five years old, Ordonneau obtained the following substances (Compt. rend., cii., 217) :- Normal propgl alcohol . . .. .. Normal butyl alcohol . . .. .. Amyl alcohol .. .. o a .. Hexyl alcohol .. . I .. .. Heptyl alcohol . . .. .. .. Ethyl acetate .. .. .. .. Ethyl propionate, butyrate, and caproate . . Cfhanthic ether (about) . . .. .. Aldehyde . . .. .. .. .. Acetal .. .. .. .. .. Amines . . .. .. .. .. Grammes per 100 litres. .. 40.0 . . 218-6 .. 8 3.8 .. 0.6 .. 1.5 .. 35.0 * . 3.0 .. 4.0 .. 3 *O . . traces. . . traces. Ordonneau's results have been substantially confirmed by Clandon and Morin (Compt. rend., civ., llS7), who found the percentage composition of the same fusel oil to be as follows :--- Clandon and Morin. Ordonneau. Propyl alcohol . . .. 11.9 11.7 Normal butyl alcohol . . 49-3 63.8 Amyl alcohol . . .. 34 $4 24.5 Isobutyl alcohol - . . .. 4.5 0.0 The following proportions of various alcohols, etc., were obtained by Rabuteau (Compt.rend,, Ixxxvii., 501) from 1 litre of potato fusel oil :- Iso-propyl alcohol. . .. . . .. . . 150 C.C. .. 30 9 , .. 50 ) 9 .. .. 65 ,? .. fio 1, Jao-amyl alcohol , , .. .. .. .. 275 ,, Prcldocts boiling above 132Qj and retaining amyl alcohol . . .. .. .. .. 170 %, Water . . .. .. .. .. .. 125 ,, Trimethyl-carbinol also appears to have been present. The fusel oil produced in Chicago, from a spirit derived from maize with smaller quantities of other grains, has been examined by Long and Linebarger (American Jourm. Anal. Chem., January, lS90). The specific gravity of the water-saturated oil was 0°810 a t 20* C. It was found impossible to dry it completely by anhydrous sul- phate of copper, but a subsequent treatment for two hours af 40 to 50' C. with dry potassium carbonate removed the remainder of the water.Only a very inconsiderable portion of the oil boiled at a higher temperature than 1339 C. It consisted in part of Normal propyl alcohol . . a . .. Iso-butyl alcohol . . .. .. .. Normal butyl alcohol . . .. Methyl-propyl-carbiuol . . I . .. Ethyl alcohol, aldehyde, and ethyl acetate . . 75 9 ,THE ANALYST. 105 ~~ alcohols and in part of bodies of an ethereal nature, the amounts of which were too small for their identification. About three-fourths of the sample consisted of a mixture of inactive and active amylic alcohols, with possibly some of the isomeric methyl-propyl carbinol. Iso-butyl alcohol appeared to be present in next largest amount, and after that iso-propyl and ethyl alcohols, with traces of normal propyl and normal butyl alcohols.It seemed t o us not improbable that spirituous liquids contained more than mere traces of propylic alcohol. This constituent, if present, would be very difficult to detect, and hence was likely to have been missed by previous observers, though, as a matter of fact, it is known to be produced by the alcoholic fermentation, and is stated to be isolated in Germany, as a commercial product, by fractionation of the feints from ordinary crude spirit. Any ordinary amount of fractionation, however, would be quite inadequate to isolate propylic alcohol or even to demonstrate its presence. E. T. Chapman describes propylic alcohol as the most hygroscopic substance in his experience j so that the last traces of water present in an alcoholic liquid would adhere to it.Its boiling-point is only 1 9 O above that of ordinary alcohol, and its solubility in water would cause it t o be washed out from any fuse1 oil with great facility in which it existed. Iso-propyl alcohol boils at a still lower temperature (83-84O), and forms a hydrate, 2C3H,0 + H,O, boiling constantly at SOQ, and having the same percentage composition as ethyl alcohol. According to Linnemann, other hydrates also exist, boiIing respec- tively at 7s' to SOo and 81" to S2'. I n any attempt to separate the constituents of spirituous liquids by fractional distillation, it is necessary first of all t o get rid of the water. Potassium carbonate removes the water very incompletely, even after protracted digestion, and the action of anhydrous copper sulphate is admittedly imperfect.Quick-lime and anhydrous baryta are more perfect dehydrating agents, and when the latter is used the completion of the process is said to be rendered evident by the solution of the baryta in the absolute alcohol, with yellow colour. We found baryta unavailable at the time we were making the experiments in question, owing to the impossibility of preparing it with the appliances at hand free from any trace of barium dioxide, which impurity rendered it unsuitable for our purpose. With quick-lime we obtained a spirit containing between 9s and 99 per cent. of absolute alcohol, as estimated from the specific gravity, and which boiled at an absolutely constant temperature from first to last. The spirit, after treatment with lime, appeared to be pure alcohol, and was wholly free from the characteristic flavour of the whisky with which we started.As this result was fatal to certain or" the aims we had in view, the experiments in this direction were discontinued. A process appears to be still wanting which will effect a complete separation of water from alcohol without at the same time affecting any aldehydes or ethers which may be present. Our idea that propyl alcohol might perhaps exist, to a notable extent, in spirituous liquids appeared to receive some confirmation from certain experiments of the President. Mr. Hehner, in a paper read before this Society (ANALYST xii., 25) described a process of estimating alcohol by treating the spirit with a known amount of standard chromic acid mixture, determining the excess of tho latter by ferrous sulphate, and from ths difference calculating the alcohol oxidised.The process possesses the advantage that the106 K2Cr,0, gramme. THE ANALYST. HzSOd gramme. homologues of alcohol require for their oxidation very different amounts of chromic acid from ethyl alcohol itself. Thus 100 parts methyl alcohol react with 922.4 parts of K,Cr,O, (to form CO, + 2 H20), while 100 of ethyl alcohol require but 427.8 parts (for the reaction C,H,O + 0, = C2H402 + HzO) and 100 of propyl alcohol only 328 parts (to f o m C K,O,). Mr. Hehner found several samples of whisky and brandy to consume an amount of bichromate considerably (5 to 7 per cent.) less than corresponded to the alcohol present, as ascertained from the specific gravity, a result which pointed to the presence of a notable proportion of propylic alcohol or other higher homologue.A limited number of experiments in this direction have not led us to the same conclusion. On the other hand, we find Hehner’s process, when conducted under favourable conditions, an exceedingly accurate method of estimating ethyl alcohol; and as our method of working differs in certain respects from his, and we consider the modifications of importance in obtaining accurate results, we give the mode of manipulation in detail :- The alcoholic liquid is distilled to separate sugar, tannin, and extractive matter. The specific gravity of the distilled spirit is then determined very carefully with the bottle, and a, weighed portion of it is diluted with a weighed quantity of water, so that 50 C.C.of the dilute liquid will contain approximately 0.2 gram. of absolute alcohol. The strength of the dilute spirit is confirmed by observing its specific gravity, and in our experiments the contained alcohol corresponding to this figure has coincided absolutely with the alcohol calculated from the known quantities of strong spirit and water employed. Exactly 2.400 grammes of solid potassium bichromate was in- troduced next into a stoppered bottle. The reagent had been carefully recrystallised and fused in porcelain at a gentle heat before use. 10 C.C. of dilute sulphuric acid (contain- ing 3 grams. of strong sulphuric acid per 10 c.c.) was next added, and then exactly 50 C.C. of the dilute alcoholic liquid of known strength run in from a pipette.The bottle was then securely closed and kept at looo C. for four hours. After cooling, an approxi- mately solution of ammonio-ferrous sulphate (78.4 grammes of FeSO, + (NH,), SO4 + 6 H,O per litre) was added in quantity just sufficient to reduce the potassium bichromate originally used. Of coiirse, the exact deoxidising power of the iron solution must have been previously ascertained by reaction with the bichromate. In practice, we find that more accurate results are obtained by weighing the iron solution than by measuring it. More dilute sulphuric acid was then added, and the excess of reducing agent determined by titration with potassium bichromate (9.837 grammes of K,Cr2O7 per litre). Each C.C. of bichromate corresponds to 0.0023 gram.of ethyl alcohol. bichromate and acid, we obtained the following results :- Working in this manner on pure ethylic alcohol, and varying the proport.ions of 2.4 2.4 3.6 4-8 3-0 5.5 4.5 6.0 Alcohol. Found. gramme. 1 =per cent. Taken, gramme. I----- I *2000 -2000 *2000 *2000 ~2006 *2002 -2004 -2002 100°30 100.10 100-20 100.1 0THE ANALYST. 107 Two experiments were made on certain fractions obtained by the repeated distillation of whisky, which fractions were considered most likely to contain propylic alcohol- By the oxidation-process they showed 99.93 and 100*12 of ethylic alcohol, for 100*00 as determined by the specific gravity; so that, if propylic alcohol were preaent, the amount was too minute to be distinctly indicated by the process in question, We have also tried the extent of accuracy of the bichromate titration process when applied to the estimation of amylic alcohol.A sample of fusel oil from beet-root spirit was dehydrated with lime and fractionated four or five times, with results showing that it consisted chiefly of iso-amyl alcohol, with lower but not higher homologues. After five very careful fractionations, no fraction of any size was obtained of a boiling-point higher than 1 2 8 O C. When oxidised by chromic acid mixture, as above deecribed, we obtained the following results :- Amylic Alcohol Taken. gramme. *2590 02600 Amylic Alcohol Found. gramme. = per cent, *2618 101-08 -2630 101.15 These results show the probable presence of traces of butylic alcohol, as was also indicated by the boiling-point of the fraction.The estimation of the amylic alcohol in commercial spirituous liquids has consider- able interest and importance, and hence we have devoted much time and attention to the subject, with a view of finding or devising a thoroughly satisfactory process. In thie connection, it is worthy of notice that, with the exception of a sample of Scotch whisky referred to by Dr. Dupr4 in a paper read before this Society in 1877 (ANALYST, i, 4), and a sample of potheen analysed by Sir Chas. Cameron, there was not, up to June of last year, a single pubhhed figure showing the proportion of fusel oil or higher alcohols in whisky. Various figures for brandy and potato-spirit, etc., have been published by Con- tinental chemists, but whisky has been left severely alone.* One of the earliest chemical processes for the laboratory determination of amylic alcohol in spirits was that of Dr.A. DuprB, described before this Society in 1877 (ANALYST i. 4). It consisted in oxidising an amount of spirit containing from 1 to 2 grammes of alcohol with chromic acid mixture, neutralising nine-tenths of the volatile acids produced, and distilling, whereby the greater part of the acetic acid was got ri6 of as fixed sodium acetate, and the valeric acid concentrated in the distillate, which on * Early in iast summer I was ieci to look carefuiiy into the pubiished ststemeiits ieiipe&iing the proportion of fusel oil and amylic alcohol in whisky, and it was then that I discovered the paucity of information on the subject. A well-known firm of Irish distillers supplied a cask of whisky to a customer.When he had drunk the greater part of the whisky the customer refused to pay for it, alleging that it had made him ill; and when sued for the amount due brought a counter-action for injury to health, owing to his having been supplied with whisky containing “ a large percentage’’ of fusel oil. This statement was scarcely borne out by his analyst, who had found 0.22 per cent. of amylic alcohol, while I, who had analysed the spirit on behalf of the distillers, had found but 0.07 per cent. In con- sequence of this discrepancy, the judge, at my suggestion, instructed us to make a joint analysis, the result of which was that we agreed that 0.07 per cent. was the correct figure. As the amylic alcohol in spirits rarely exceeds 0.1 per cent., or 70grains per proof gallon, it seems highly improbable that it can produce the local effects sometimes attributed to it. Its effect on the general system has probably been greatly exaggerated.A pupil of mine informs me that some years ago he took a teaspoonful of fusel oil, mixed with water, without any ill-effect. Recently, for three weeks I took every evening, with a few exceptions, a wine-glass full of whisky to which crude fuse1 oil had been added to the extent 4, 1, and ultimately 2 per cent, The spirit was extremely nauseous, but roduced no headache or other ill-effects,-A. H, ALLEN.10s THE ANALYST. treatment with barium carbonate yielded the barium salts of the volatile acids contained in the fraction. The weight of these salts and the percentage of barium contained in them was then ascertained.As barium acetate contains 53.73 per cent. of Ba, while the valerate contains 40.41 per cent., it follows that the proportions of acetic and valeric acid in a mixture of the two can be determined by this means. Like all similar indirect methods the results are only reliable when neither constituent is in very large excess. Where the proportion of one ingredient is small, the method becomes untrustworthy. Thus Dupr6 obtained from the fraction of the volatile acids resulting from the oxida- tion of a sample of Scotch whisky a mixed barium salt containing 53.49 per cent, of Ba. It is evident that this result; indicates a very small proportion of valeric acid in the mixture, and hence a grave chance of error.But, it is worthy of notice that the amylic alcohol calculated from the above result comes to 0-108 per cent. in the whisky, or 66.5 grains per proof gallon, a result which more modern methods confirms as a perfectly probable figure. DuprB’s process might, of course, be improved by a further concentration of the valeric acid by fractional neutralisation followed by distillation, but the process is even then objectionable on account of the small quantity of alcohol worked on, and wbich cannot be conveniently increased. If a sample of spirit of proof strength contain 0.1 per cent. of amylic alcohol, it is evident that by working on 3 grammes of the sample the valeric acid obtained will correspond only to 0.003 gramme of amylic alcohol, an amount altogether too small t o give safe quantitative results.A great improvement on DuprB’s process has been made by L. Marquardt (Berichte xv. 1661), who operates on 150 grammes of the sample. This he dilutes with water till it has a specific gravity of -980 (corresponding to about 28; per cent. of proof spirit, or 13 per cent. of alcohol by weight), and then repeatedly agitates with chloroform. This dissolves the fuse1 oil, while the traces of ethyl alcohol also taken up are removed by repeated agitation with water. The chloroform is then treated in a strong bottle or eask with 2 grammes of sulphuric acid and 5 grammes of potassium bichromate in 30 C.C. of water. The bottle is then securely closed and heated under pressure at 85’C., with frequent agitation, for six hours.It is then distilled to a small bulk, water added, and the distillation repeated. The mixed chloroform and aqueous distillate are boiled with barium carbonate under a reflux condenser, the chloroform distilled of€‘, and the aqueous liquid concentrated and filtered from the excess of barium carbonate. The filtrate is evaporated to dryness and the residue weighed, after which it is redissolved, and the barium and chlorine determined in separate portions of the solution. From these data the weight of the valeric radical and the corresponding amyiic alcohol can be deduced. With care and good fortune we have found Marquardt’s process capable of yielding very fair results, but the practical difficulties in properly carrying out the method are very considerable.In the first place it is extremely difficult t o obtain chloroform of such purity that it will not yield a sensible quantity of volatile organic acid on treat- ment with chromic acid mixture. Although the chloroform used has been several times purified by heating with chromic acid mixture, it is still liable to give troublesome traces of volatile acids on retreatment, And even if such a purification be effected as to get rid of a11 but negligible traces of such impurities, a sensible amount of hydrochloric acidTHE ANALYST. 109 is formed on oxidation, and this contaminates the barium salts, necessitates a determina- tion of the chlorine, and a correction for the amount found. The necessity of treating the cbloroformic solution with chromic acid mixture under pressure is also a serious objection to the process.The preesure at 8 5 O is very considerable, and bursting of the bottle containing the mixture has happened not unfrequently. At any notably lower temperature the oxidation appears t G be incomplete. The whole process is tedious, owing to the frequent agitations requisite, for the densities of the chloroform and chromic acid mixture are so different that the liquids separate almost instantly after agitation, and hence the reaction occurs but slowly. We have attempted to overcome the last difficulty by increasing the density of the chromic acid mixture till it is equal to that of ahloroform, so as to obtain an emulsion on shaking, but even this device was only partially successful. Owing to the difficulties attendant on the use of chloroform we have now definitely abandoned its use in favour of carbon tetrachloride.This liquid boils at 7 6 V C . , and hence the oxidation can be conducted under a reflux condenser at the ordinary atmos- pheric pressure ; and as the vapour of the heavy chloride of carbon with its contained amylic alcohol is continually passing up through the layer of chromic acid mixture, proper contact is obtained and complete oxidation ensured without much personal attention being requisite. The carbon tetrachloride of commerce is obtained by a single treat- ment with chromic acid mixture in such a condition of purity that it yields no volatile organic acid on further oxidation and a mere trace of hydrochloric acid. It is a more expensive reagent than chloroform, but owing to its higher boiling point there is much less wasted in its use, and it is recovered practically without loss.One curious difficulty has arisen in connection with the use of carbon tetrachloride, and that is the tendency of amylic alcohol to be partially removed from its solution therein by agitation with water, and hence to be only imperfectly extracted from the diluted spirit. This behaviour, which at one time appeared very formidable, we have completely overcome by the use of brine instead of water for diluting the spirit, and of brine or sodium sulphate solution for washing the carbon tetrachloride after the extrac- tion. We have proved the competency of this mode of working in various ways. The following are the details of the process for the determination of amylic alcohol in spirits as now worked by us.As a preliminary step, any fixed matters must be removed, and any ethers and furfurol present destroyed in the following manner. The omission of this step is liable to cause the results obtained to be materially in excess of the truth, and invalidates the figures obtained by Dr. Bell and ourselves before the necessity of the precaution was pointed out by one of us (A. H. Allen). caustic soda added, and the whole heated under a reflux condenser for an hour. The contents of flask are then distilled in the following manner :-A volume of 90 C.C. is allowed to pass over, the flame is then removed, and 30 C.C. of water is introduced into distilling flask. The distillation is con- tinued until 20 C.C.more has been collected. The distilling flask is allowed to cool, and 10 grams. sodium sulphate washed into the distilling flask with 20 C.C. more water. The distillation is again continued until 20 C.C. has passed over, thus bringing the total volume of the distillates to 130 C.C. By this time, all the amylic alcohol will have been volatilised. 100 C.C. measure of the spirit is taken, 20.c.c.110 THE ANALYST. Empirica formula. C,H60 C,H,O C4H100 - CBH,,O I I C6H140 .. C7H160 .* C2H40, .. C,H,O, .* C,H1402 * C,OH*OO,* C7'H1402 ' C,H,O . . C,H,O .. C,H,O, . . C5H5N . . C,H1,02 CHIEF CONSTITUENTS OF SPIRITS AND FUSEL OIL. Name. Ethyl alcohol . . . . . . Normal propyl alcohol . . . . Iso-propyl alcohol . , . . a-Normal butyl alcohol .. P-Iso-primary butyl alcohol , , Tertiary butyl alcohol .. . . a-Normal primary amyl alcohol P-Iso-primary amyl alcohol . . y-Iso-primary amyl alcohol . Methyl-propyl carbinol . . . . [so-primary hexyl alcohol . . [so-primary heptyl alcohol . , Acetic acid . . ,. . Ethyl acetate . . . . . . Ethyl valerate . . . . . . imyl valerate . . . . . . 4myl acetate . . . . . . lldehyde . . . . . . . . ketone (occurrence doubtful). . lcetal (diethyl-aldehydato) . i'urfurol (furfuraldehyde) . . 'vridine . . . . . . . . Constitutional formula. CH,.CH,.OH . . . . . . CH,.CH,.CH2.0H . . . . . . (CH,),:CH.OH . . . . . . CH,.GH,.CH,.CH2.0H . , . . (CH,), : CH.CH2.0H . , .. (CH,), : C.OH . . . . . . CH,.CH,.CH2.CH2.CH,.0H . . (CH,), : CH.CH,.CH,.OH , , . . (CR,)(C,H,) : CH.CH2.0H . , (CH3)(C3H7) : CH.OH .. . . (CH,), : CH.CH,.CHpCH,.OH . (CH,), : CH.CH,.CH,.CH,.CH,.OH CH,.CO.OH.. . . . . .. C,H,. C,H,O, . . . . . . . . C,H,.C,H,O,. . . . . . .. C5H11.C2H302 . . . . . * S,Hll.C,H,O, * . . I .. ?H,.CO.H . . . . . . .. >H,.CO.CH,. . . . . . . . >H,.CH : (0,C,H5),. . . . . . !,H3O.COH . . . . . . . . LH,N . . . . . . .. - Boiling point."C. 78.4 98 83-84 117 108-109 - 137-138 181.4 128 119-120 152-1 53 163-165 118 74.3 134.5 137 188 21-28 56.5 104-106 I61 16.7 The mixed distillates are now diluted with aaturated brine until the density of the liquid is about 1.10, when it is shaken four times successively with carbon tetrachloride, using 40 C.C. the first time, next 30 c.c., then 20 c.c., and lastly 10 C.C. The tetrachloride of carbon now contains the amylic alcohol and probably some ethylic alcohol, and with a view of removing the latter, it is shaken twice with 50 C.C.brine; or preferably, once with brine and once with a saturated solution of sodium sulphate, so as to avoid any traces of chlorides being present after separation. The carbon tefkachloride is finally run off from the separator and filtered. The oxidation may be conducted in a closed bottle, or under a reflux condenser. The oxidising mixture consists of 5 grammes of potassium bichromate, 2 grammes concentrated sulphuric acid and 10 grammes of water, In using a, closed bottle the oxidation is conducted in the water-oven, and it is necessary that the bottle be frequently agitated, Under these conditions the oxidation will be m p l e t e in four hours.When a flask is used, under a reflux condenser, the carbon tetrachloride should be kept in active ebullition for eight hours, this being best effected over a water-bath. After oxidising, the product is diluted with 30 C.C. water and distilled over a nakedTHE ANALYST. 111 CHIEF CONSTITUENTS OF SPIRITS AND FUSEL OIL (continued). Action of H,SO, on dilute alcoholic solution. Not affected . , Not affected . . Not affected . . Not affected . . Strong coloration. , Coloration . . . , Strong coloration. . Strong coloration. . Not affected . , Not affected . . Not affected . . Coloration . . . . Coloration . . . . Coloration . . . . Not affected . . .. .. .. .. * . .. .. .. .. .. .. .. .. .. .. Forms alcohol and aldehyde Strongly blackened . , Forms pyridine sdphate , .Products of treatment with caustic alkali. I Not affacted Acetate . . . . . , Acetate and ethyl alcohol. . Valerate and et.hyl alcohol Acetate and amyl alcohol. . Valerate and amyl alcohol Resin, acetate, and alcohol Not readily affected . . Not affected .. .. .. .. .. .. .. .. .. .. .. Pyromucate and furfuryl alcohol Not affected ,. .. .. ?roducts of oxidation with dilute chromic acid mixture. Acetic acid. Propionic acid. Acetone ; then acetic and car- bonic acids. Normal butyric acid. Iso-butyric acid ; then acetic and carbonic acids. Acetic and carbonic acids. Normal valeric or pentoic acid. Iso-valeric or pentoic acid Dextro-rotatory valeric acid. Methyl-propyl-ketone ; then acetic and propionic acids. Iso-caproic acid. Iso-cenanthylic acid. (inactive).Unchanged (acetic acid). Acetic acid. Acetic and valeric acids. Valerie and acetic acids. Valerie acid. Acetic acid. Acetic and carbonfc acids. Acetic acid. Pyromucic acid,C,H,O.CO.OH Not affected. flame until only 20 C.C. remains in the distilling flask. 80 C.C. of water are now added, and the distillation continued until only 5 C.C. remains in the flask. The distillate wiJl now contain the whole of the valeric acid, a portion being in the aqueous distillate, and a portion in solution in the carbon tetrachloride. The entire distillate is now titrated with decinormal baryta-water, using methyl orange as an indicator, and shaking thoroughly a€ter each addition. Not more than 2 C.C. of $ alkali, and generally much less, ought to suffice to produce neutrality.Phenolphthalein is then added, and the titration continued with frequent shaking until the neutral point is again reached. alkali used in the second stage of the titration corresponds to -0102 gramme of valeric acid or -0088 gramme of amylic alcohol. The alkali consumed in the titration with methyl orange represents the minerd acid formed and is not, of course, taken into account. The foregoing method of dealing with the products of the oxidation is a great im- provement on that prescribed by Marquardt. It substitutes a rapid, easy, and delicate titration for a treatment with barium carbonate (and chance of imperfect neutralisa- tion), followed by filtration, evaporation, drying, weighing, and supplementary determina- Each C.C. of112 THE ANALYST. tions of barium and chlorine.Nevertheless, the observation of the weight of the barium salts is often very valuable, as it enables the combining weight of the organic acid to be calculated, and its identity with valeric acid inferred. For this purpose, the neutralised aqueous liquid is separated from the carbon tetrachloride (which after a precautionary treatment with chromic acid mixture and distillation from barium carbonate can be used again) evaporated to dryness and the residual barium salt dried a t looo (or, preferably, at 130') and weighed. In the event of any notable quantity of mineral acid having been indicated, this weight must be corrected by the weight of barium chloride deduced from the result of the titration with methyl orange. I n the absence of hydrochloric acid, or in its presence after making this correction, the mean combining weight of the organic acids can be found as follows :- Corrected weight of barium salt in milligrammes.- 67.5 = comb, weight of organic acid. Volume of normal baryta in C.C. In experiments where we have added pure amylic alcohol to spirit with a view of testing the process, the orgznic acid obtained has had the characteristic odour of valeric acid and a combining weight closely approximating to 102. I n conducting the determination of amylic alcohol by the foregoing process, the following points must be observed :- 1. The carbon tetrachloride must be previously purified by treatment with chromic acid mixture and distillation over barium carbonate, and must be free from chloroform. 2.All corks used in distilling the spirit must be kept distinct from those employed in distilling the product of oxidation. I n all cases, the corks must be carefully covered with tinfoil, or they will absorb amylic alcohol and valeric acid in quantity sufficient to vitiate the results. 3. A few small fragments of pumice-stone should be added, in each case, to the contents of the distilling flasks. The pumice employed with the oxidised liquid should be treated with chromic acid mixture before use. 4. The brine should be made from clean table-salt, rendered distinctly acid with sulphuric acid, and filtered before use. Of course, the so-called estimation of amylic alcohol in spirits is in reality the esti- mation, in terms of amyl alcohol, of such higher alcohols and other bodies as may be extrwted by chloroform or carbon tetrachloride, and converted into volatile organic acids on oxidation, But it is noticeable that the higher alcohols other than amylic alcohol wiIi give products which do not materiaiiy aEect the remits.Thus isobutyi aicohoi on oxidation yields isobutyric acid, which body undergoes further change into acetic and carbonic acids. But the acetic acid formed will neutralise just the same amount of alkali as the isobutyric acid would have done. Dr. James Bell has modified Marquardt's process by using potassium permanganate in place of bichromate, and continuing the oxidation for a very long period. The change appears a very objectionable one. Permanganate of potassium is very liable to contain traces of perchlorate, which, being isomorphous, cannot be removed by any process of recrystallisation. On distillation with acid such impure permanganate yields distinct traces of perchloric acid (or other oxide of chlorine) which when boiled with barium carbonate yields a soluble salt, the acid in which has nearly the same combining weight a,a valeric acid (HC10, = 100.5 ; HC5H9O2 = 102).Even pure permanganate appears toTHE ANALYST, 113 act on chloroform far more readily than bichromate does, and on subsequently distilling the liquid a distillate is obtained having a yellow colour and strong chlorous odour. Besides the various modifications of the oxidation process of estimating higher alcohols in spirits, certain physical methods have been suggested. We have tried the capillary method of Trauber, as modified by Ekworthy (Jour.Chem. Xoc., liii. 102), but have not found it suitable for our purpose. The Rose-Nerzfeld method, depending on the increase in the volume of chloroform when shaken with the spirit reduced to a con- stant strength, appeared more promising. But the absolute necessity of adjusting the strength of spirit accurately within the limits of 29-96 and 30.04 per cent. of absolute alcohol is a serious bar to the use of the process, which in the end gives rather an esti- mation of the total oily bodies present than of the higher alcohols, and fails even with those when tho proportion is as low as commonly occurs in practice. We have met with more encouraging results by using carbon tetrachloride instead of chloroform, while the employment of brine for dilution renders the strength of the spirit immaterial within wide limits.But the process thus modified has not yet baen perfected, and hsnoe we postpone its description. As already stated, the presence of ethers or furfurol will invalidate the determina- tion of amylic alcohol by oxidation, as these bodies are extracted both by chloroform and carbon tetrachloride, and on oxidation will yield organic acids. Thus ethyl acetate will yield two equivalents of acetic acid, and will falsify the amylic alcohol result to an extent equaI to double its own weight. The ethers of spirits can be determined (in terms of a typical compound such as ethyl acetate) by a process apparently originating with Berthelot, and applied by Dupr6 to the ethers of wine, It is substantially the same as was subsequently used by Koettdtorfer for the examination of butter and other fats, and is based on the amount of alkali required for the saponification of the ether.But in the examination of spirits we are met with the difficulty that bodies of the type of aldehyde and furfurol are present, and these also react with alkali. Aldehyde reacts with alkali with formation of aldehyde-resin and production of a formate and acetate, but the reaction does not appear to have been examined in its quantitative relationships, or to correspond to any simple formula. Furfurol, however, has been found by Dr. A. Colefax, who has been working with us on the subject, to react with alkali almost strictly according to the following equation : - BC4R3O.CO.H +- HOK = C4H36), CO.OK + C4H30.CH,.0H Furf urol. Pyromucate. Eurfuryl alcohol. It is probable that a determination of furfurol might be based on this reaction. Fortunately, the error introduced into the determination of the ethers by the presence of aldehyde and furfurol can be obviated by means recently described by E. Mohler, who has found that on digestion with a solution of aniline in syrupy phosphoric acid the aldehyde and furfurol are converted into non-volatile compounds, while the ethers can be distilled off unchanged. 7;1; e have fully confirmed Mohler’s observation, provided that the treatment with the aniline reagent be conducted at the boiling point of the spirit, under a reflux condenser, for at least two hours. The following is the mode of operation :- 200 C.C.is distilled to about 20 c.c., when 50 C.C. measure of water is added, and the distillation continued till all but about 10 C.C. has passed over. The distillates are mixed and divided into two equal parts (A and B). One portion (A) is titrated with deci- normal caustic alkali and phenolphthalein and the free acid thus found calculated to acetic acid. The neutral liquid is treated with 20 C.C. of 5 caustic alkali, and boiled under a reflux condenser for one hour, when the excess of alkali is ascertained by titra-114 THE ANALYST. tion with decinormal acid. The other (B) portion of the distilled spirit is treated with 1 C.C. of aniline and 1 C.C. of phosphoric acid of 1-442 sp. gr., and boiled under a reflux condenser for at least two hours.It is then distilled to a low bulk, and the distillate neutralised and treated with 20 C.C. of 2 soda, exactly as was done with the other portion. The difference between the alkali added and that found by titration represents that consumed by the saponification of the ethers in 100 C.C. of the spirit; each C.C. of the decinormal alkali representing *0088 gramme of ethyl acetate. The difference between the amounts of alkali required for the saponification of portions A and B represents the alkali which has reacted with furfurol, aldehyde, etc., and, where only the first substance is present, 1 C.C. of alkali represents 0.0192 gramme of furfurol. The presence of furfurol in spirits can be detected, and the proportion roughly guessed at, by the reaction of the sample with a solution of aniline in glacial acetic acid.Ten drops of aniline should be dissolved in 2 C.C. of glacial acetic acid, and the mixture added to 10 C.C. of the spirit to be tested. A red coloration is produced, which increases in intensity on standing. The reaction is peculiar to furfurol and extremely delicate, one part per million giving a distinct coloration. Aldehyde is best detected in spirits by Gayon’s reagent consisting of 30 C.C. of a solution of magenta (rosaniline hydrochloride) in 1,000 parts of water; 20 C.C. of bisulphite of sodium solution (of 1.31 sp. gr.) ; 3 C.C. of sulphuric acid ; and 200 C.C. of water. 4 C.C. of this mixture should be added to 10 C.C. of the spirit to be tested, when a crimson coloration is produced, increasing in intensity on standing; According to Mohler, no satisfactory colorimetric determination can be based on this reaction, but the following proportions of aldehydes can be detected :-Acetic and cenanthic aldehydes, 0.01 gramme per litre ; vaIeric aldehyde, 0.02 ; propionic and iso- butyric, 0.05 ; normal butyric aldehyde, furfurol, and acetone, 0.5 gramme per litre.Alcohols and ethers give no coloration with the rosaniline reagent, but it is difficult to meet with commercia1 alcohol so pure as to give a wholly negative reaction. It has been stated by Borntrager that the reagent is untrustworthy, as it merely indicates the presence of an oxidising agent ; but this is evidently not the case, as the proportion of sulphite present is many times the amount requisite to prevent any oxidising action of the aldehyde. At present there exists no satisfactory means of determining aldehyde in the minute quantities in which it exists in spirits, Its behaviour with alkalies to produce aldehyde resin and the accompanying odour is the most characteristic reaction.Acetal is a body having the constitution of a diethyl-aldehydate. It has an agree- able odour, and is produced by the prolonged contact of aldehyde with alcohol, and hence has been recognised as a constituent of old wine and matured spirits. We have not been able to recognise its presence with certainty in th8 moderate quantities of spirits we have worked on. Acetal is unaffected by alkalies if air be excluded, but on treat- ment with dilute acid is at once split up into alcohol and aldehyde.Its most characteristic reaction is the formation of a colourless liquid with caustic soda and iodine solution, which yields a dense precipitate of iodoform when acidified. This reaction does not occur in very dilute solutions of acetal. Acetal is extracted by chloroform and carbon tetrachloride from its solution in dilute alcohol, and hence will affect the determination of the amylic alcohol if not previously got rid of, which may be done by first heating the spirit with an acid, and subsequently distilling with alkali: Various methods of stating the results of analysis of spirituous liquids have been adopted. The statement in ‘‘ parts per 100 of absolute alcohol ” has certain merits ; but in practice is less convenient than ‘‘ parts per 100 of proof spirit.” “ Grammes per 100 c.c ” and $6 grammes per litre ” have the advantage of ready calculation, but statementsTHE ANALYST. 115 so made are apt to be misleading if the strength of the spirit is not also borne in mind.The statement in ‘ I grains per proof-gallon ” has the advantage of defining the strength of the spirit, and avoids the long decimals necessary in other forms of statement. Much still remains to be done before the methods of analysing spirituous liquids can be regarded as perfect, and still more before the results obtained can be properly interpreted. At present we can only claim to have fairly accurate processess for esti- mating the water, alcohol, higher alcohols, ethers, and free acid, with very delicate coloration tests for furfurol and aldehyde, But there are other bodies in spirit, the nature of which is only guessed at, and the presence of which is only indicated by the smell and flavour they impart to the liquid.The minute proportion in which these constituents exist, their unknown nature, the very large amount of spirit necessary for their isolation, and even for their concentration, will render their investigation extremely difficult. DISCUSSION. DR. DUPRE said, first, that he was perfectly willing to let his old process for determining amplic alcohol die in favour of a better one, but would like to know how it compared with the new. And, secondly, he must prohest against the method of determin- ing ethers based on their saponification being called Koettstorfer’s method.It was brought out many years ago by Berthelot, and he (the speaker) was the first in this country to use it extensively; in fact, Koettstorfer’s process of butter analysis was only a slight variation of a method which he (Dr. Dupr6) brought before the Society at one of its earliest meetings. The PRESIDENT pointed out that the usefulness of the Society had again been demonstrated by the authors, who, in their very excellent paper, had drawn upon the Society’s work in several directions. He suggested that they might combine the process which they had adopted,viz., the determination of the volatile acids formed by oxidation by means of bichromate, with the modification of the bichromate process in- troduced by himself (Mr. Hehner), and estimate a t the same time the amount of bichrornate consumed. The measuring of the bichromate solution required great care, as had been pointed out by the authors ; but if the temperature of the bichrornate solution was taken into consideration, and care was taken to let the solution run out of the burette very slowly, perfectly accurate results could be obtained, although he himself had also, from time to time, employed weighed quantities of bichromate instead of a standard solution. He was glad that Mr. Allen was gradually helping to place the question of fusel oil in spirits on a scientific basis. Hitherto, all evil effects of some spirits had been attributed to fusel oil, although nothing definite was known about the real nature of the injurious constituents. It was a great step in advance, to come to a recognition of our ignorance in the matter, and to get rid of glib talk about fusel oil. In conclusion, he asked the members to accord their best thanks to the authors for their admirable paper, which was evidently based upon careful research and analysis. MR. ALLEN, in reply, said they had undoubtly been led to make the experiments on the estimation of alcohol by titration with bichromate, in consequence of Mr. Hehner’s results. He thought Mr. Hehner’s suggestion to combine the determination of the amount of oxygen taken up by the alcohol with a determination of the acid formed decidedly valuable in certain cases. Dr, DuprB’s figures for amylic alcohol, published in 1887, were perfectly consistent with results more recently obtained by improved methods. With regard to the saponification method of determining ethers, one, of course, associated the process with the beautiful work done by Thudichum and Dupr4 in the analysis of wines. (Conclzcsiom of the Society’s Proceedings. )

ISSN:0003-2654    

DOI:10.1039/AN891160102b

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

116 THE ANALYST. THE DETECTION OF ANTISEPTICS IN BEERS," BY DR. H. ELION. ONE of the less agreeable duties which EL brewer's chemist has now and then to perform is the analysis of samples of beer from other breweries, to see whether they use unfair means to make their inferior product saleable. I f anything goes wrong with their beer, they are almost sure to use antiseptics. A few years ago I published a process for the detection and approximate estimation of minute traces of salicylic acid, but as the use of this substance has been prohibited in several countries, it is now but very seldom used, other bodies being, however, substi- tuted for it. As 1 was lately again called upon to undertake the investigation of a similar product, I was thinking whether it would not be possible to find out a method to ascertain the presence of any antiseptic in a sample of beer, without pretending to say what preservative it really was. I n the present state of our science it must not, of course, be expected to detect the most minute quantity; but as the object of using antiseptics is really to be able to sell a beer which would otherwise be condemned, the amount used will generally be fairly large.The antiseptics plainly serve to make the beer keep better; but in my opinion such simple means are in reality very unfair, as I wish to make plain, as follows :-A good beer must be right as regards odour, taste, colour, clearness, &c.; if a sample is supposed to keep well, this naturally means that it will retain these properties for a long time, even when kept under less favourabls conditions.Pasteur has proved that, after a while, t>here form various micro-organisms, so-called disease-feernzents, which make the beer unfit for consumption. While these are considered to be mostly bacteria, it must be remembered that beer may contain a peculiar yeastf, which, under suitable conditions, may so multiply as t o make the beer very turbid, and affects the taste of the article. It now mostlydepends what treatment the beer undergoes after leaving the brewery. I n England, for instance, it is quite customary to keep the beer in the barrel until used up. I n this case a slight increase of the yeast, which goes to the bottom, is not only harmless, but even desirable to keep the beer well impregnated with carbonic acid.If beer is kept in bottles, the forxation of yeast ought to be only very slight, otherwise it will be almost impossible to pour out a glass of clear beer. I n properly conducted breweries, it is nowadays not so particularly difficult to prevent the growth of bacteria; but if great care is not taken ths brewer will be obliged to stop their multiplication by the use of some antiseptic, The process which I have now worked out to detect such adulteration is as follows : On the fifth of November last year, I received some bottles of beer, which were said to keep particularly well, and this proved indeed to be the case in a high degree. Not only was the beer still quite clear on 21 December, but even last March, after standing for four months in the laboratory, it did not Rhow any alteration.The deposit was so small as not to sensibly render the beer turbid on shaking. On the 21st of December I started the investigation. Under the microscope, the deposit showed an organism like sarcina, also a few bacilli, but scarcely any yeast cells. A portion of this beer was now introduced, with the usual precautions, in previously sterilised Pasteur's flasks, which were only half filled. "Zeitschr. f. Angew Chernie, April, 1891 (slightly abridged).THE ANALYST. 117 In some of these flasks a trace of the saccharomyces cerevisia was introduced, and the whole kept at a temperature of 25O C. In none of them any yeast was formed, pointing to three possible causes :- 1. The absence of fermentable sugar. 2. An insufficient amount of nourishment for the yeast. 3.The admixture of antiseptics. The possibility of the increase in yeast being stopped by a large amount of alcohol was put out of the question by the analysis of the beer : Sp. gr. d $+ = 1.01454. Apparent extract (see my tables) ... Real extract.. . ... ... 4.54 ,, 3-57 per cent. Which of the three causes was the true one was decided by adding to the contents of some of the flasks a previously aterilised solution of maltose, and also some yeast-food. But even after adding more of the saccharomyces, no formation of yeast cells were noticed, which plainly proved that the great durability of this beer was caused by the use of an antiseptic. The beer was now shaken out several times with an equal volume of ether, sterilised in Pasteur’s flasks, and again treated with the saccharomyces cerevisia.By the ether treatment the antiseptic (which was not salicylic acid) was removed, and a very active formation of yeast now set in, An estimation of the fermentable sugar still present could not now be carried out according to my plan, on account of the treatment with the ether. The reducting power of the extract in the original beer was 29.8’7 ; but after the removal of the antiseptic and subsequent fermentation, only 22.27. If one calls the true maltose in 100 parts of the extract z, 100 extract after fermentation gives (100 - x) extract with (29*87 - x) apparent maltose. Expressed in per cent. of the extract left after fermentation, this value is 22.27, consequently : 100 (29.87 X ) -- 22.27. 100 - % The beer extract therefore contained 9.78 per cent. of real maltose, and the beer itself a 4 4 per cent. Now unless the most scrupulous care had been observed during the manufacturing, this beer could not be supposed to keep well. I think it as well not to mention the nature of the preservative, as this would only advertise it ; but I hope I have given an easy process to distinguish genuine beers from adulterated ones.

ISSN:0003-2654    

DOI:10.1039/AN8911600116

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

THE ANALYST. 117 QUANTITATIVE ESTIMATION OF NITRIC-NITROG:EN.* BY DR. E. FRIUICE. SINCE analytical and manufacturing chemists have agreed t o estimate the nitrogen in saltpetre directly, and no longer from the loss, a host of processes have been proposed which are said to answer the purpose, besides being expeditious, In the meeting of the directors of the German Agricultural Stations, held in September, 1888, in Bonn, Jodlhur’s process was recommended. This process, however, gives according to our, and other * L3itschr. f. Angem 0hem.ta. April, 1891. (Slightly abridged.)118 THE ANALYST. analysts’ experiences, only good results when the greatest care is exercised, and takes up rather too much time, so a more easy method is preferred. Fijrster’s process is based on the same principle, and gives very good results indeed.Both methods, however, suffer from the drawback that the analyst is supposed to work on about -5 gram. of the sample, which makes it exceedingly difficult to get a good average portion. A procesa which we used some years with the best success was originally proposed by Lievert. It is as follows :-20 grams. of the sample are dissolved in a litre of water, and of this solution 50 C.C. are put into a 600 C.C. flask, diluted with an equal bulk of water, and 20 grams. of caustic potash are added. After this has dissolved, 75 C.C. of spirits of wine are mixed with it, and further 15 grams. of zinc and iron powder. To prevent frothing a few granules of animal charcoal may be added. The flask is now closed, and connected with a 200 C.C.Peligot’s tube, which contains 10 C.C. of normal sulphuric acid, and is partially immersed in cold water. After three or four hours the spirit, which of course carries over the ammonia, is distilled off by heating with a very small flame, the operation lasting about two hours, when all the alcohol will have distilled over. This method is really an excellent one providing the potash is free from nitrates, and the zinc and iron dust are not too much oxidised. The reduction by means of aluminium wire, proposed by Xtutzer, is not to be recommended, as everything depends on the nature of the metal. The pure metal now-a-days prepared by electrolysis, is far less active than the metal prepared according to tho old process, which gives a product containing sodium and silicon.Schmitt has lately proposed a method based on the reduction of nitric acid by nascent hydrogen in acid solutions. 40 C.C. of glacial acetic acid are put into a 600 C.C. flask containing 15 grams. of the zinc iron mixture. 50 C.C. of the saltpetre solution are now added, and when the evolution of hydrogen has somewhat ceased, another 15 grams. of the metal are added, and if the mass should get pasty, about 30 C.C. of water. The reduction is completed in about forty minutes; soda ley is added until the zinc hydrate has redissolved, and the ammonia is then distilled off. This method gives thoroughly good results if one succeeds in preventing frothing, and the spurting over of any alkali. All these processes (which give excellent results in skilful hands) must, however, make room for the method lately proposed by Ulsch, which also is based on the reduction of the nitric in acid solution.This process is characterised by simplicity, celerity, and economy, and according to my experiments with pure nitre, it yields remarkably good and concordant results. According to Ulsch’s directions, 25 c.c of a solution containing about *5 gram. of nitre or -4 gram. of Chilian nitre, are put into a 600 C.C. flask, and reduced by means of 5 grams. of reduced iron and 10 C.C. of dilute sulphuric acid (1 vol. acid + 2 vols. of water). The flask must be covered with a pear-shaped piece of glass to prevent loss by spurting, owing to the violent evolution of hydrogen. When the action somewhat ceases, the reduction must be accelerated by the gradual application of heat until finally the liquid has boiled for about six minutes. After diluting with about 150 C.C. of water, 30 C.C. of soda ley of 1.25 sp. gr. are added, and after introducing a few pieces of granulated zinc, the ammonia is distilled off as usual. The acid is then titrated back with 2 soda.THE ANALYST. 119 On trying this process on a sample of ohemically pure nitrate of soda, we found 16.44 instead of 16-47 per cent. of nitrogen, When nitric acid has to be estimated in potable waters, we have always used the reduction prooess in a alkaline fluid; but we felt anxious to know whether Ulsch’s process would answer in this case. Wo found that if one litre of water is evaporated down to about 50 c.c,, best after addition of a few drops of soda ley, the resulting fluid may be very accurately treated acoording to Ulsch’s directions.

ISSN:0003-2654    

DOI:10.1039/AN8911600117

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

THE ANALYST. 119 LAW NOTES. IMPORTANT DECISION AS TO LIABILITY OF MILK CARRIERS. HOTCHIN, APPELLANT-HINDMARCH, RESPONDENT. THIS case, an appeal from justices at South Shields, raised the important question whether a milkman or other servant can be convicted under the Adulteration Act of selling milk or any other article adulterated which he sells on behalf of his employer. Hotchin, a milkman, had been convicted under section 6 of the Adulteration Act, 1875, for that he had sold to the respondent Hindmarch three gills of milk “ not of the nature, substance, and quality of the article demanded by the purchaser.” It was proved and found as a fact that Hindmarch, who is sanitary inspector of the urban sanitary authority of the borough of South Shields, met Hotchin with a cart, vending milk, The cart had upon it the name of ‘‘ The Farmers’ and Cleveland Dairy Co.(Limited).” Hindmarch asked of him three gills of milk, which the man sold to him for 3d. The proper steps were then taken to have it analysed (Hindmarch telling him he had bought it for the purpose), and the analyst certified that it contained 12 per cent, of added water. Hotchin told the inspector at the time that he was selling the milk for the company, in whose service he was foreman. It was objected that the company was the sellers, but this the justices overruled, The defendant relied on section 25 of 38 and 39 Vict., c. 63, on the ground that the milk was purchased by the company with a written warranty that it was genuine milk, and that neither he nor the company had any reason to believe it was otherwise, and evidence was given for the defence thus suggested.A railway clerk stated that two cans of milk, on the morning of the sale, came to the station consigned to the company by some one at Dumfries, each can haviDg a label on it stating the date and quantity of milk it contained (which was found to be correct), and they were re- oeived for the company. The witness, however, stated that there was no protection against the milk being tampered with en poute in a distance of akout ninety miles, Each can was labelled--“34 imperial gallons of warranted genuine new milk with all its cream on, from R. Thompson, Dumfries, to the Farmers’ and Cleveland Dairy Company, High Shields,” One of the company’s men stated that he had received the two cans, the contents of which were apportioned between Hotchin and another man for sale.He did not, however, test the milk when received or did he ever test it, and that though Hotchin often tested the milk with a lactometer, he did not do so on the morning in question, for what reason the witness did not know, Hotchin himself was called as a witness, and stated that he was a local foreman of the company, and had a shop where he sold the milk ; that he received the milk on the morning in question and sold it exactly as he received it, and he said he sold it for the company. He admitted that he had the means of testing the milk, but did not do so, and could give no reason why he did not do so on this occasion. The written contract between Thompson and the company was produced, and he agreed to supply them for a period from October, 1890, to March, 1891, “ with the whole of his dairy of genuine good new milk, of the best quality, with all its cream on,” and he thereby 6‘ warranted each and every supply of milk delivered under the contract to be pure genuine new milk, unadulterated, with all its cream on.” But evidence was given that the district manager had stated at the cffice of the inspector that he had known five days before that there was “ something wrong with the milk,” and that it was supposed to be watered, which, however, he denied.On the whole of the evidence the magistrates convicted the accused, but stated a case, in which they stated that it was contended-(1) that he was not the seller ; and (2) that section 26 of the Act exempted him from liabilityon the ground that he believed tbat the milk sold was that sent, But the magistrates found that he had no reason to believe this under the circumstances.Mr. Mansfield appeared for the milkman, the appellant, and urged that he ought not to have been convicted. Mr. Chitty appeared for the prosecutor in support of the conviction. Lord Coleridge, in giving judgment yesterday, said the case had been very ingeniously argued and, in consequence of another decision, not quite easy of determination. But he had come clearly to the conclusion that the conviction ought to be upheld. The Adulteration Act was directed against adulteration, and pointed to particular acts done by particular persons, one of which was selling the adulterated article.The intention or quality of the act pointed at was for the magistrate to ascertain and determine. But the act itself in every case was at all events a physical act which the party must know he is doing ; and if it was outside the proviso for protection he was liable to be convicted, even though he was a servant. The 6th section was that no person shall sell an article which is adulterated. If a person did so he was liable to the penalty, unless he became within any exception of the proviso120 THE ANALYST. z If he, in fact, sold an article adulterated then he was liable to be convicted even though he was a servant. The question had been raised what was “selling ” within the Act. In several seotions it appeared that selling meant the physical act of transfer of the article, and that the person was liable who did such act.Under the 26th section, however, there would be a defence if the seller showad that he purchased the article as pure and sold it as he purchased it ; and it was urged that the seller must mean the purchaser of the commodity afterwards to be sold. That was an ingenious argument, but i E it were admitted the Act might just as well not have been passed, and a fraudulent dealer had only to make such a contract, and neither he nor the actual seller would be liable, however much the article might be adulterated. He could not, therefore, yield to the argument (though it had at one time been accepted valid in cases of contracts), for if it were admitted it would make the Act entirely nugatory.It was urged, however, that in the case cited the Court so held,. but in truth it had not SQ held, for it was held on a dserent state of facts, In that case there was a finding of fact that the contract was bona$de, and that the milk had been sold under it, without being altered, And upon that finding the judgment was quite right. It was not so in the present case, for the magistrates found that there was nothing to show that the milk sold was the milk sent under the contract, and, in fact, there was a transit of ninety miles during which the milk might have been tampered with ; and, further, it appeared that the milkman might have tested the milk and had not done so. Even, therefore, if the 25th section applied-and he thought it did not apply-the case did not raise a defence under that section, and the case cited did not apply, The conviction, therefore, was right and must be confirmed. Mr. Justice Mathew concurred. It was clear, he said, that in point of fact the milk sold was adulterated, and the defendant, the milkman, sold the milk so adulterated. The offence was, in the terms of the Act, that he did sell milk which was adulterated, and that offence he had committed. The Act did not except aervants ; and if they had been excepted it would have been of little avail. The Legislature it was clear meant to deal with the person who actually sold, and i f he were not liable then, though he had himself adulterated the article, he could not be convicted and his innocent employer would be liable. He agreed that section 25 raised in this case no defence for the accused, The conviction, therefore, was right. Appeal dismissed.-Tiimee.

ISSN:0003-2654    

DOI:10.1039/AN8911600119

出版商:RSC    

年代:1891

数据来源: RSC

摘要:

120 THE ANALYST. ERRATA.-h May number, page 98, line 24, for 16.3 read 18.3 ; and page 99, line 10, for Cooled read cooling.

ISSN:0003-2654    

DOI:10.1039/AN891160120b

出版商:RSC    

年代:1891

数据来源: RSC