摘要:

STENHOUSE AND GROVES’S COSTRIBUTIONS, ETC. 415 L.--Cont&butions to the History of the Naphthalene Series. No. 11. P-Naphthaqzdnone. By JOHN STENHOUSE, LL.D., F.R.S., and CHARLES E. GROVES. @-Nap thnq winone. IN a paper communicated to the Chemical Society last year (1877, ii, 47), we mentioned that nitroso-6-naph thol, when converted into the corresponding amido-naphthol, and then submitted to oxidation, yielded a crystalline compound of the nature of a qninone ; this we designated 6-naphthaquinone, in order to distinguish it from the a-naphthaquinone previously described by one of us. Further experi- ence has shown us that in preparing this quinone it is more advan- ta,geous to employ a considerably larger quantity of potaossium dichromate than that formerly used. The proportions which we recommend are 6 parts of a saturated aqueous solution of sulphurous acid, 6 of dilute sulphuric acid (1 vol.acid to 2 of water), and 3 of potassium dichromnte to every 2 parts of the pure nitroso-@-naphthol, the reduction 2nd subsequent oxidation being carried out in the manner previously described (Zoc. cit., 52). In this way the nitroso- naphthol yields from 62-65 per cent. of its weight of the pure quinone, the theoretical quantity being 91.9. We have not ascer- tained the cause ?f this deficiency, but as the sulphur precipitated during the reduction of the nitroso- to the amido-compounds contains but very little organic matter, it is probable that the loss is occasioned chiefly by the solution of a portion of the amido-P-naphthol in the dilute solution of ammonium sulphide.This solution, however, after it has been aciditied with sulphuric acid, and the hydrogen sulphide removed by heating, does no* yield any quinone when mixed with a solution of potassium dichromate, as it should do if there were any quantity of the amido-naphthol present, 2 H 2416 STENHOUSE AND GROVES'S CONTRIBUTIOXS Nit?.o-P-)Lnpht7~a~~inone. It was noticed that P-naphthaquinone dissolved readily in hot dilute nitric acid, and the solution on cooling deposited a crFstalline substance of a magnificent crimson colour. Since our last communi- cation we have prepared this substance in a state of purity. The pure 8-quinone (10 parts) was added to nitric acid of density 1.2 (60 parts by measure), and after the mixture had been agitated until it formed a homogeneous paste, the flask was plunged into boiling water. The quinone dissolved, and in the course of a few minutes a mass of red crystals of the new substance separated, a small quantity of nitrous fumes being at the same time evolved; this was no doubt due to secondary action.When the whole of the quinone was converted into the nitro-derivative, which only required a few minutes, the flask was removed from the water-bath and plunged into cold water, so as to stop the oxidising action of the hot nitric acid on the nitroquinone. After allowing it to stand for a couple of hours, the product was collected and washed, first with a little dilute nitric acid (density 1*2), and then with water. The crude product amounted to 70-75 per cent.of the naphthaquinone originally taken ; it was puri- fied by dissolving it in 15 tinies i t weight of boiling benzene, and allowing the solution to stand for 18 to 24 hours, as it separated but slowly. The hard crimson nodules which formed on the sides of the flask were finally crystallised once or twice from twice their weight of boiling glacial acetic acid, which on cooling deposited the nitro-@ naphthaquinone in magnificent crimson plates, bearing a striking re- semblance in point of colonr and general aspect to chromic anhydride. It gave the following results on analysis :- I. a301 gram of substance gave *654 gram carbonic anhydride and -075 gram water. 11. 0197 gram of substance gave *429 gram carbonic anhydride and -046 gram water. 111. 0427 gram of substance gave 25.4 cub. cent.of nitrogen at 18.5" c., and under a barometric pressure of 764.8 mm. (corr. at 0') equivalent to 23.443 cub. cent. dry nitrogen at 0", or 0029408 gram of nitrogen. IT. ~364 gram of substance gave 21.5 cub. cent. of nitrogen at 17*5", and under a barometric pressure of 745.9" mm. (corr. at Oo) equivalent to 19.830 cub. oent. dry nitrogen at O", or 0024877 gram of nitrogen :-TO THE EISTORY OF THE NAPHTHALENE SERIES. 417 I. 11. 111. IT. Mean. Clo = 120 59.11 59.26 39-39 - 59-32 H5 = 5 2-46 2.77 2.59 - 2-68 N = 14 6.90 - - 6.89 6.83 6-86 0 4 = 64 31.53 - - - - - -- 203 100.00 These numbers correspond with the formula for mononitro-/3- mphthaquinone, Cl,,H5( NOz) 0,. It has been already shown (Zoc. cit.) that this substance yields phthalic acid on oxidation, so that both the NO, group and the two oxygens of the quinone must be in one and the same benzene ring (adopting Graebe's graphic formula for naphthalene), but nothing is known of their relative positions.It is not unlikely, however, if the p-quinone should really prove to be a meta-derivative, that the nitro-quinone will be found to be 0 : 0 : NO2 = 1 : 3 : 4; but at present so little is known of the order in which the hydrogen-atoms are displaced in naphthalene, that this must be re- garded as purely hypothetical. Nitro-P-naphthaquinone is insoluble in light petroleum, almost inso- luble in carbon bisulphide, slightly soluble in ether, more so in benzene and in boiling alcohol, and very readily soluble in hot glacial acetic acid ; boiling with alcohol, however, appears to decompose it, as the substance does not crystallise out again as the solution cools, and on evaporation it merely deposits yellow oily drops.I t dissolves pretty freely also in boiling acetic acid of 30 per cent. (density 1*04), and to a less extent in water; the orange-coloured solutions be- come opalescent on cooling, but do not yield crystals. It is soluble in concentrated sulphuric acid in the cold, but is at the same time decom- posed, the dark-coloured liquid giving a brown precipitate when diluted with water. The nitroquinone melts at 158", and when strongly heated on platinum foil burns with feeble deflagration. It undergoes reduction when treated with hydriodic acid and phosphorus, with formation of two compounds, probably the corresponding nitro- hydroquinone and the amidohydroquinone, crystallising in large crimson plates and in red-brown needles ; these new substances have not as yet been further examined. Aqueous solution of sulphurous acid also reduces the nitroquinone.Dinaph thy ldiquinhy drone. It had been noticed in the preparation of P-naphthaquiaono that if the freshly precipitated quinone was allowed fo stand for any length of time in contact with the acid solution, it lost its bright orange colour, and became darker; this at the time was attributed to oxi-418 STENHOUSE AND GROVES'S CONTRIBUTIONS dation by the excess of potassic dichromate employed for its preci- pitation, but we have since ascertained that it is due to another cause. When the quinone is heated with mineral acids, or even when it is left in contact with them in the cold for a sufficient length of time, it is converted into a compound of a purplish-black colour analogous to quinhydrone : it is to the formation of this substance by the action of the free sulphuric acid, that the darkening of the naphthaquinone just mentioned is due, and not to oxidation. The most convenient agent for the transformation of the &naphtha- quinone was found to be dilute sulphuric acid (1 vol.of acid to 2 of water). When the pure quinone (1 part) was intimately mixed with the acid (10 parts by measure), so as not to leave any portion un- wetted, and allowed to stand for 24 hours, it was completely converted into the black substance, or the change could be more rapidly and conveniently effected by heating the mixture i u a water-bath at about 55" ; the quinone rapidly became dark-coloured, and at the end of ten minutes the reaction was complete.Tf the temperature was allowed to rise much above 60" soft pasty masses were formed inclosing some of the unaltered P-quinone, so that it became somewhat difficult to convert the quinone entirely into the new compound. About twice its bulk of wafer was added to the mixture when cold, and the product collected and thoroughly washed with cold water to free it from adhering sulphuric acid. The diquinhydrone thus prepared forms an indigo-black powder when dry, which is insoluble in water, carbon bisulphicle, and light petroleum, almost insoluble in benzene, but soluble in glacial acetic acid ; the reddish-brown solution is immediately decolorised on addition of sulphurous acid solution.The diquinhydrone dissolves in concentrated sulphuric acid with production of a deep green colour ; the addition of water to this solution produces a precipitate. Treated with reducing agents, such as hydriodic acid and phos- phorus, or an aqueous solution of sulphurous acid, the diquinhydrone is converted into a colourless substance crystallising in needles, whilst with oxidising agents, as nitric acid or potassium dichromate and sulphuric acid, it yields orange-coloured prisms, so that these three substances stand in the same relation to one another as do green quinhydrozle, quinol or hydroquinone, and quinone, in the benzene series. Dinap hth y 1 diq uinone .The quinhydrone just described is easily converted into dinaphthyl- cliquinone by boiling it with nitric acid, or with a solution of potassium dichromate and sulphuric acid, but it is more conveniently prepared by pouring a solution of pure dinaphthyldiquinol in boiling 30 per cent,.TO THE HISTORY OF THE NAPHTHALENE SERIES. 419 acetic acid, into nitric acid (density 1.45) in quantity sufficient to oxidise the hydroquinone. In this way the diquinone is precipitated in small lustrous prisms of a brilliant orange colour. If, however, the nitric acid be added to the acetic acid solution, instead of pouring the latter on to the nitric acid, the precipitate is of a dingy colour, probably owing to the presence of a small quantity of the diquin- hy drone. The following results were obtained on analysis of this substance dried at 100" :- I. *196 gram substance gave ,549 gram carbonic anhydride and *061 gram water.11. -104 gram substance gave 52915 gram carbonic anhydride and 0033 gram water. 111. *261 gram substance gave ,732 gram carbonic anhydride and *082 gram water. I. 11. 111. Mean. Clo = 120 76.43 76.39 76-49 76.49 76-44! H5 = 5 3-18 3.46 3.52 3.49 3.49 0 2 = 32 20.39 -. - - - - 157 100.00 The analytical numbers agree with the empirical formula, C,oH,Oz ; but from a consideration of the mode of formation this should be doubled, making the formula of the dinaphthyldiquinone C,oH,oO*. Dinaphthyldiquinone is insoluble in water, and but very slightly soluble, or quite insoluble, in all ordinary solvents; even boiling glacial acetic acid dissolves it but very sparingly.The diquinone is a very stable body, dissolving in boiling concentrated nitric acid, and crystallising out unchanged on cooling ; by long continued boiling with the acid, however, i t appears to be decomposed, and it is not improbable that a study of the products of oxidation formed in this way will yield interesting results. It is also easily soluble in warm concentrated sulphuric acid, and is precipitated again on dilution. Reducing agents convert it into the corresponding hydroquinone. Dinaphth y ldiguinol. This compound is formed by the reduction of the diquinone or diquinbpdrone, as when they are boiled with phosphorus and a solu- tion of hydriodic acid in acetic acid of 30 per cent., or treated with aqueous solution of sulphurous acid ; the quinone, however, appears to be more easily reduced than the quinhydrone.The diquinone, when allowed to stand for some time in contact with sulphurous acid soh- tion, becomes converted into a mass of colourless crystals of the420 STENHOUSE AND GROVES'S CONTRIBUTIONS diquinol; the qninhydrone does not appear to be produced as an inter- mediate stage i n this reaction, for at no time could the black crystals of this compound be distinguished in the mixture. A very convenient method of preparation is to convert the P-naphthaquinone into the diquinhydrone by treatment with dilute sulphuric acid in the manner previously described, and when the mixture is cold, to add about twice its bulk of a saturated aqueous solution of sulphurous acid, and allow the whole to stand for 24 hours, when the black diquinhydivne will be fonnd to be entirely converted into a mass of snow-white needles of dinaphthyldiquinol, provided the 6-naphthaquinone originally em- ployed was quite pure: these merely require to be thoroughly washed to remove sulphuric acid, and crystallised once from dilute acetic acid, to which a few drops of sulphurous acid solution have been added, to be quite pure.If impure @-quinone has been used for the preparation of the diquinhydrone, the diquinol will be coloured and somewhat difficult to purify ; repeated conversion into the diquinone and recon- version info the diquinol will, however, usually suffice to render it c dourless. Dinaphthyldiquinol, C20H1404, or CzoHlo( OH)d, forms colourless needles, which, on drying, become darkened from partial oxidation.It melts at 176-178", is almost insoluble in water, moderately soluble in 30 per cent. acetic acid, easily in the glacial acid. It is only slightly soluble in benzene, bisulphide of carbon, and ether. The analyt.ica1 results were as follows:-I. ,244 gram substance, gave -676 gram carbonic anhydride and -0905 gram water. Czo = 240 75.47 75.56 Hid = 14 4-40 4-12 0, = 64 20.13 - - 318 100*00 These numbers show a deficiency in hydrogen for the formula, C20H140,, which is what might be expected from the comparative ease with which the substance becomes oxidised, Mode of Formation of Di.nap7LthylaiipzcinhzJcne. The conversion of @-naphthaquinone into the black diquinhy drone, under the influence of acids, would seem to take place by the con- densation of 2 molecules of the quinone, with elimination of two atoms of hydrogen which unite with two out of the four oxygen atoms in the double quinone-molecule to form a quinhydrone- CioHdh + C10HeO2 = GoHioOz( OH)z,TO THE HISTORY OF THE NAPHTHALENE SERIES.421 or regarding the quinhydrone as formed by the union of a molecule of the quinone with one of the quinol: This view is strongly supported by the fact that 6-naphthaquinone yields its own weight of the diquinbydrone. Two experiments were made, in each of which -500 gram of pure 0-naphthaquinone was con- verted into the diquinhydrone by treatment with dilute sulphuric acid : the products collected, washed thoroughly, and dried, weighed 0498 and *495 gram respectively.It was evident that the results of an analysis of the diquinhydrone fur hydrogen (taking into consideration its high molecular weight, and that in all probability it readily alters by oxidation) would not yield satisfactory evidence a8 to whether this is a correct explanation of the change which takes place. It seemed probable, however, that if the hydrogen-atoms external to the nucleus (that is, those not directly attached to carbon-atoms) could be determined, it would at once show whether the reaction was of the nature above indicated: a means of doing this is afforded by the action of bromine in aqueous dution, which readily oxidises the diquinhydrone, but does not attl:ick the diquinone. A weighed quantity of the pure @-nnphthaquinoue was accordingly converted into the black diquinhydrone, and agitated with excess of bromine-water of known strength for a few minutes, whereby i t was converted into the diquinone ; the excess of bromine was then estimated by potassic iodide and hyposulphnte solution in the usual manner. I n this way it was found in two experiments that -200 gram of pure 6-naphthaquinone, after conversion iuto the diquinhydrone, required 0100 and ,098 of bromine respectively to oxidise it to the new quinone j the theoretical amount for being -101 gram bromine.There can be little doubt, therefore, that the black substance is a diquinhydrone analogous to the green quin- hydrone of the benzene series. Although it was almost certain-the nature of the diquinone and diquinhydrone being established-that the formula of the diquinol was C,Hl0( OH),, notwithstanding the analytical results showed a defi- ciency in the hydrogen; yet it was thought advisable to make expe- riments by oxidation with bromine-water in a manner similar to those detailed for the diquinhydrone, with a view to ascertain whether it actually contained four OH-groups.From the readiness with which the substance oxidises, however, the results were not very accordant. A weighed quantity of the dinaphthyldiquinone, C20H,,,04, was reduced by sulphurous acid solution, washed thoroughly, collected on a, vacuum422 SCHTJNCK AND ROEMER ON filter, and the product, which had alrea.dy become slightly coloured, was dissolved in hot glacial acetic acid ; the rather dark-coloured solu- tion was then poured into an excess of bromine-water, and the excess of free bromine determined as before. Three experiments gave 98 : 88.5 and 85 of bromine per 100 of the diquinone, whilst theory requires 101.9. The diquinone formed in these oxidation experiments was examined, to ascertain if any bromine substitution-derivative had been formed, but it was found to be free from bromine. The compound, C2oHlo(OH)4, may be regarded as a tetrahydric alcohol of the hydroquinone or quinol species, derived from a dinaph- thyl, CloH7.CloH7, by the displacement of four hydrogen-atoms by four OH-groups, and is a member of a new class of bodies, the &iquinols. We have, therefore, called it ~inaphth,~idiquinol) and the corresponding quinone-compound, dina~htl~?lEcliqzlinone. We have ascerttlined that benzoquinone and a-nrtphthaquinone also yield diquinones and diquinols by condensation, and have no doubt that other quinones will yield corresponding compounds. We are at present engaged in studying this interesting group. Our thanks are due to Mr. El. Sear for the care with which he has done much of the analytical work in connection with this paper.

ISSN:0368-1645    

DOI:10.1039/CT8783300415

出版商:RSC    

年代:1878

数据来源: RSC

摘要:

422 SCHUECK AND ROEMER ON L1.-Notes on 3ldder Golowing Matters. (Continuation,) By EDWARD SCHUNCK, Ph.D., F.R.S., and HERNANN ROEMER, Ph.D. 6. IDENTITY OF PURPUROXANTHIC ACID AND MUNJISTIN. IN order to remove any remaining doubt as to the identity of the colouring matter obtained by us from commercial purpurin with the munjistin of Stenhouse, we prepared a quantity of the latter sub- stance, employing, with merely slight modifications, the process given by the discoverer. One pmt of ground mnnjeet was treated with a boiling solution of one part of aluminium sulphate in fifteen parts of water, the treatment being repeated four or five times. The red decoction, after straining through calico, was mixed with hydro- chloric acid, which gave a yellow precipitate. This precipitate was treated with boiling concentrated hydrochloric acid, in order to remove the alumina still contained in it ; and after being filtered off and well washed with water, it was exhausted with boiling alcohol.The latter removed all the colouring matter of the precipitate, leaving n, quantityUDDER COLOURING MATTERS. 423 of pectic acid and other impurities undissolved. The liquid left after the alcohol had been distilled off a bulky reddish-yellow mass, con- sisting for the most part of munjistin and purpurin. The two sub- stances were separated by treatment with boiling water, slightly acidified with acetic acid, in which the munjistin dissolved. The fil- tered liquid gave, on addition of hydrochloric acid, an orange- coloured precipitate, which was filtered off and dissolved in boiling glacial acetic acid.This solution deposited, on cooling, golden-yellow scales of munjistin, small quantities of purpurin and other substances remaining in the mother-liquor. The miinjistin was recrystallised from glacial acetic acid until the product melted exactly at 231". It had the properties of purpnroxanthic acid, and also the same composi- tion, as the following analysis shows :- 0.3454 gram of substance gave 0.8002 gram of COz, and 0.0980 gram In 100 parts: HZO. C ........ 63-18 H ........ 3.15 The formula, Cl,H,(COOH)O, requires- C ........ 63.38 H ........ 2.82 The residue obtained, on evaporating the mother-liquor of the mun- j istin, contained, beside purpurin, small quantities of alizarin and other bodies. The alizarin was easily detected by employing the method described by us in Note 1.Although we obtained a much larger quantity of munjistin from munjeet than from commercial purpurin, still in the former, as well as in the latter product, purpurin largely preponderates. The yield from 1OOIbs. of munjeet was so considerable that we were able to subject the munjistin obtained to more minute examination. Its composition, being that of a dioxyanthraquinone-carbonic acid, rendered a study of its derivatives of some interest, more especially as we hoped thereby to arrive at some means of preparing this and similar compounds synthetically. The question arose, whether the decomposition into purpuroxanthin and CO?, which, as %ye have shown, is effected quantitatively at 233", would take place even at a lower temperature, when the substance is exposed to the action of bromine and other reagents.It seemed also of importance to determine whether, by the action of acetic anhydride, the one hydrogen of the carboxyl-group, as well as the two atoms of the hydroxyl-groups, could be replaced by acetyl.424 SCmTCK AND ROENER ON MADDER COLOURING MATTERS. 7. EXPERIMENTS ON MUNJISTIN. Action of Acetic Anhydride on Munjistin. At the temperature of ebullition, acetic anhydride acts on the substance, yielding a product very easily soluble in the liquid. On the addition of water, a pale yellow mass is precipitated, which melts at about 160". The compound is, however, unstable. By the action of boiling alcohol or glacial acetic acid, it is reconverted into mun- jistin.The same compound is formed on heating with acetic anhy- dride from 160 to 200". It crystallises from ether in small needles. Its uncertain melting point and its general character show, however, that it is not a pure substance, but a mixture of an acetyl-componnd with uncombined munjistin. Action of Bromine on Micnjktin. On adding to a solution of munjistin in glacial acetic acid, heated to near the boiling point, an excess of bromine, the solution deposits, on cooling, large lustrous prismatic needles of an orange- yellow colour. The product, after crystallisation from glacial acetic acid, in which it dissolves with some difficulty, has a constant melting point of 231". The same product is obtained when alcohol is used as a solvent instead of glacial acetic acid, but in this case the crystals are much smaller, less lustrous, and of a pure yellow colour. The compound dissolves in alkali, with a reddish-yellow colour.A bromine determination led to the following result :- 0.2542 gram substance yielded 0.2400 gram AgBr, equivalent to 40.17 per cent. of bromine. Hence i t follows that the compound was dibromopurpuroxanthin. The latter contains 40.20 per cent. of bromine, whereas dibromo- munjistin muid contain 36.19. According to P1 a t h (Deut. Chew,. G'es. Ber., ix, l204), dibromopurpuroxanthin melts at 227-230"). There can be no doubt, therefore, that b s the action of bromine on munjistin, the carboxyl-group had been eliminated, and that the bro- mine-compound of purpuroxanthin had been formed. Action of Potassic Hydrate on Munjistin.At the ordinary temperature, a concentrated solution of potassic hydrate does not act on munjistin, but, on boiling, purpurin is formed, as we have already observed. Action of Ammonia on Munjistin. Ammonia converts munjistin at the ordinary temperature, though The change is effected easily and corn- slowly, into purpurinamide.HARROW ON PTROTRITARTARIC ACID, ETC. 425 pletely by heating to 100". The product was recognised as pur- purinamide by the fine purple colour of its solutions in alcohol, alkalis, and baryta-water, by the peculiar green fluorescence of the alcoholic solution, and by the absorption-bands shown by the barytic solution. Actiolz of Nitric Acid om Mzcnjisth. The solu- tion is red, and deposits on further standing long orange-colonred needles. These needles melt at 251", at the same time giving off mnch gas. They dissolve in boiling water with a yellow colour, and yield an insoluble crimson barium compound. As the substance very closely resembled the dinitropurpuroxanthin described by P l a t h, both as regards the melting point, that of dinitropurpuroxanthin being a t 249-250", and other properties, we prepared the nitro-compound of purpuroxanthin and compared i t with the substance obtained from munjistin. They are readily distinguished, however, by their behaviour to ammonium acetate, a boiling solution of which dissolves dinitropurpuroxanthin without much difficulty, yielding on cooling an ammonium salt in long, silky needles ; whereas the nitro-compound of munjistin dissolves with great difficulty in ammonium acetate, giving a salt in microscopic needles, Fuming nitric acid dissolves munjistin after some time. We found them to be very similar.

ISSN:0368-1645    

DOI:10.1039/CT8783300422

出版商:RSC    

年代:1878

数据来源: RSC

摘要:

HARROW ON PTROTRITARTARIC ACID, ETC. 425 LII.-On Pyrotritartaric and Carbopyrotritartario Acids, Products of the Action of Dilute Szcl@uric Acid on Diacetosuccinic Ether. By GEORGE H. U. HARROW, Ph.D. IT has been shown that in the action of alkaline hydrates on ethyl acetoacetate, or on substituted derivatives of the same, the decom- position consists in the formation, on the one hand, of a ketone, alcohol and carbon dioxide, and on the other of a substituted acetic acid by the splibting off of the acetyl group, thus:- CO.CH3 I C0.CH.q + OH2 = I + co2 + C2H60. CHXY426 HARROW ON PTROTRITARTARIC AND C0.CH3 CHXP 7% + I I CXY + OH2 = I COOH coOc2k5 COOC2H, At the suggestion of Professor W i s li c enu s, I undertook the inves- tigation of the action of dehydrating agents on ethyl diacetosuccinate, a substance which Rugheimer (Rer., vii, 892) prepared by the action of iodine on ethyl acetosodacetate :- CO.CH, CO.CH, CO.CH3 + 2NaI.I I I CH I I I 2 CENa 4- It = CH COOC2H5 COOCaHs COOC2H5 If the reaction took place in the normal manner, the production of a double ketone was to be expected, thus :- CO.CRs CO.CH3 CH, CH, I I I CH I I I I + 20H2 = CO CO + 2C2HeO + 2CO2. I CH COOC2H6 i=00c2H5 CH2-CH2 On boiling ethyl diacetosuccinate with dilute sulphuric acid, how- ever, it was found that the above reaction did not take place, but that the products were an acid of the empirical formula, C8H,0,, and the ethyl salt of an acid C7H803, first prepared by Wislicenus and Stadnicki (Ann. C I L R ~ . P/iaim., cxlvi, 306) by the dry distillation of tartaric acid, and called by them yyrotritartaric acid, and afterwards obtained by Bottinger (Deut.Chem. Ges. Ber., vi, 893) by heating pgruvic acid in sealed tubes with water alone, or with an aqueous solution of barium hydroxide (ibid., viii, 1583) for three days i n a flask connected with an inverted condenser. B o t t i n g e r called his acid uvic acid. The acid C8H8O5, which, at Professor Wislicenus’s suggestion, I bave named ‘‘ carbopyrotritartwic acid,” does not appear to have been made previously. Preparation of Solid Ethyl Acetosodacetate. Conrad and Limpach (Anm. Cliem., cxcii, 153) have recently described an easy and rapid method of making tbis compound by simply adding ethyl acetoacetate to an alcoholic solution of sodium ethylate. In preparing ethyl diacetosuccinate, however, it is necessary to have the sodium compound in the solid state, and as free as possible from alcohol.The method I employed to obtain it in this state is as follows :-CARBOPYROTRITARTARIC ACIDS, ETC. 427 Ten grams of eodium were dissolved in 100 grams of absolute alcohol; when cold the crystalline mass was mixed with an equal volume of dry ether, and the calculated quantity, 56.5 grams, of ethyl- acetoacetate, also diluted with its own volume of ether, added gradually. On adding a small quantity of water (about 2 c.c.) and stirring well, ethyl acetosodacetate is Precipitated in the crystalline form, and finally the whole becomes almost solid. The precipitated substance was pressed to remove alcohol and ether, and placed under the air-pump over sulphuric acid until quite dry.The pressed-out liquors furnish a further quantity of sodium com- pound, soon becoming solid. CNa ( C2RJ (CO. CH,) ? by means of water has also been tried, and succeeds perfectly, the precipitate consisting of fine white silky needles. It may be purified by washing with dry ether and finally dried under the air-pump. Much more sodium-compound is obtained by precipitation with water than by the simple addition of ether, as formerly practised ; indeed, from the ethyl acetosodacetate prepared by Conrad and Lim- pach's method nothing is precipitated by ether alone. The precipitation of ethyl acetosodethacetate, 1 COOCZH, Prej? arat ion of Et hy 1 D iacetosu ccinnte. Findy powdered ethyl acetosodacetate was mixed with sufficient ether to form a thin paste, and an ethereal solution of iodine added, with continual shaking, until sodium iodide ceased to be precipitated, and the liquid became slightly coloured.A small excess of iodine- solution was then added and the whole left to itself for some hours. The ethereal solution was filtered from the sodium iodide and shaken with mercury to remoTe iodine, filtered, most of the ether distilled off, and the residue placed in dishes to crystallise. The substance, freed from the adhering brown mother-liquor on a vacuum-filter and washed with a little ether, was almost pure ethyl diacetosuccinate. The mother- liquor, on evaporation of the ether, furnished another crop of crystals. Dry ether was used and the amount o€ decomposition resulting from the heating during distillation was comparatively small.Mercury was employed iiistead of soda to remove iodine, in order to avoid the introduction of water, which might have proved detrimental in the subsequent distillation. The substance so obtained was perfectly white and crystalline, and melted at 78-79'. a 1 5 5 1 gram gave on combustion *3150 (30% and 00943 OH2.428 HARROW ON PYROTRITARTARIC AND Theory C12H1806. Found. C = 55.81 55.39 H = 6-98 6-75 0 = 37.21 - Acfiom of Dilwte Sdphuric Acid on E'thyl Diacetosuccinate. Portions of 20 grams of ethyl diacetosuccinate were boiled with 150 grams dilute sulptiuric acid (1 to 10) in a flask connected with an inverted condenser. On passing the evolved gas through barytn- water, a precipitate of barium carbonate was formed, showing that carbon dioxide is a product of the reaction.A quantitative experiment, made bypassing the dried gas over soda-lime, showed that the amount formed is equal to about 5 per cent. of the ethyl diacetosuccinate employed. The boiling was continued until gas ceased to be evolved, which usually happened with the above quantities in about six hours. On cooling, a white crystalline body separated out, mixed with an oil having a strong anise-like odour. On submitting the whole to dis- tillation in steam, the oil passed over with the watery distillate; the latter was re-distilled, and the first portions treated with potassium carbonate ; a layer then separated which was dried with potassium carbonate and distilled. It boiled on re-distillation at 78-79', burnt with a pale-blue flame, and had the burning taste and characteristic odonr of alcohol.The Oily DistilZate.-Dried with calcium chloride, and fractionally distilled, by far the largest portion of the distillate boiled, with little or no decomposition, at 207-210", and this, on re-distillation, boiled at 208-209'. The analyses of the substance point to the formula C9H1203. I. 02067 gram gave *4830 gram COz and 01304 OH,. XI. *1144 gram gave *2657 gram COz and ,0752 OHz. Theory CSH1203. I. 11. C = 6428 63.72 63.34 H = 7.1 4 7.01 7.30 o = 28-58 - - Pyrotritartaric or Uvic Acid.-On boiling the above-mentioned oil with sodium hydroxide solution for two hours, it was completely con- verted into alcohol and an acid having the empirical formula C7He03, identical with the pyrotritartaric acid of Wislicenus and S t a d n i c k i and with Bottinger's uvic acid.On adding dilute sulphuric acid to the alkaline solution, the pyro- tritartaric acid was precipitated in the crystalline form and of a slightly yellow colour.CARBOPYROTRITARTARIC ACIDS, ETC. 42 9 It was found to be soluble in a considerable quantity of boiling water, almost insoluble in cold water, but soluble in ether, from which solvent it crystallises in prismatic needles. It was repeatedly crys- tallised from ether in order to purify it. When pure it is perfectly colourleRs and melts at 135". Wislicenus and Stadnicki give the melting point of pyrotritartaric acid as 134.5", and Bottinger that of resublimed uvic acid as 134". It is volatile in steam, and sub- limable without decomposition at temperatures considerably below its melting point, beginning to sublime between 100" and 110".The crystals obtained by sublimation are long thin prisms upon which smaller crystals group themselves. On treating a very small quantity of the acid on a watch-glass with two drops of fuming hydrochloric acid and warming slightly, then adding about six drops of strong sulphuric acid, and again gently warming, a fine cherry-red colour makes its appearance. This test was tried with pyrotritartaric acid kindly given me by Professor Wislicenus, and the same red colour was obtained. In applying the test, it is important to add sufficient sulphuric acid, and not to heat too strongly. The reaction does not take phce with either sulphuric or hydrochloric acid alone.Recrystallised from ether the acid gave the following numbers on combustion :- I. -1227 gram acid gave *2675 gram CQ, and -0634 gram OH,. IT. 01057 gram acid gave 02323 gram CO, and -0578 gram OHp. 111. *1276 gram acid gave -2791 gram CO, and *0711 gram OH2. Theory C,H,O,. I. 11. 111. C = 60.00 59.45 59.93 59.65 H = 5.71 5-74 6.08 619 0 = 34-29 - - - Silver Pyrotritartrate.-This salt was prepared by adding silver nitrate to a solution of the sodium salt formed by neutralising the free acid with sodium carbonate. It is a bulky precipitate which, when heated in the water-bath, or on standing, becomes crystalline. It is soluble in a considerable quantity of boiling water with only very slight decomposition, and was in fact purified by crystallisation from water ; from very dilute solutions it crystallises in rosette-shaped groups of needles.It becomes yeJlowish when exposed to light in the dry state. On analysis the following numbers were obtained :- I. -102 gram salt gave *1247 CO,, ,0312 OH,, and *0432 Ag. 11. -0655 gram salt gave -081 C02, 9298 OH,, and a0276 Ag. VOL. XXXlII. 2 1430 HARROW ON PPROTRITARTARIC AND Theory C7HrAg0,. I. 11. C = 34.01 33.34 33.73 H = 2.83 3.39 3.53 Ag = 43.73 42.35 42.13 0 = 19.43 - - "he low percentage of silver in the combustions is due to a, portion of the silver being volatilised, the combustion tube being coloured yellow by the silver. The silver was determined by heating the salt very gradually in a porcelain crucible. Theory. 43.7 - - I.-1491 gram salt gave -0651 gram Ag = 43.66 11. ,1656 gram salt gave *0714 gram Ag = 43.12 111. -1124 gram salt gave -0484gram Ag = 43.06 On heating the silver salt, the sides and lid of the crucible became covered with a crystalline sublimate, which was shown by its melting point, 135-136", to be pyrotritartaric acid. Sodium Pyrotritnrtrate.-Pyrotritartaric acid was neutmlised with sodium carbonate, the solution evaporated to dryness on the water- bath, and the organic salt extracted with absolute alcohol, the alcohol evaporated 0% and the white crystalline residue dissolved in water, in which it is easily soluble, and evaporated. -1914 gram of the air-dried salt lost -0348 gram on heating to 100" = 18-18 per cent. OH2 ; theory, C7H,Na03 + 2Aq requires 18.18 per cent.OH2. -1566 of salt dried at 100" gave ~0685 gram Na$Or, corresponding to *0222 gram Na = 14.16 per cent. Na; theory, C7H7Na0, requires 1419 per cent. Na. Ethyl Pyrotir'tmtrute, C7H7( CzH5) 03.-On adding ethyl iodide to dry silver pyrotritartrate, reaction took place in the cold, so much heat being developed that the ethyl iodide was caused to boil violently. The reaction was completed by heating in the water-bath for an hour. The ethyl pyrotritartrate was extracted from the silver iodide formed by means of ether, and on distilling off the ether, a slightly yellow- coloured oil was left which had the same anise-like smell as the original oil from which the acid was obtained by saponification with soda. On distillation it boiled at 205-208", but a small quantity of crystals of pyrotritartaric acid separated (melting point 135-136"), the formation of which was probably due ko a trace of water remain- ing in the silver salt employed. To remove the acid the substance was mixed with a little ether and shaken with dilute potash, the ethereal layer dried with potassium carbonate, the ether distilled off, and the residual oil fractionated ; the boiling point was again 205- 208", and now no crystals separated, and it was perfectly neutral toCARBOPYROTRITARTARIC ACIDS, ETC.431 test-pper. Owing to the small quantity of substance operated on, the boiling point appears a little lower than that of the original oil *1889 gram gave on combustion 4429 gram C02, and *1265 gram (208-209"). OHz. Theory C;Er(C2H5) 0,. Found. C = 6428 63.94 H = 7.14 7.44 0 = 28.58 - Biittinger (Ann.Chem. Pharrn., clxxii, 245) did not succeed in preparing the ether of uvic acid. He, however, found that reaction takes place between silver uvate and ethyl iodide in the cold, but he used ethyl iodide diluted with alcohol, which alcohol probably con- tained enough water to re-produce nvic acid. That a trace of water would have this effect during the formation of the ether is shown above. Carhop y ro trit artaric Acid. The white crystalline substance, not volatile in steam, from the action of dilute sulphuric acid on ethyl diacetosuccinate, was recrystal- lised first from the mot her-liquors containing sulphuric acid and then several times from boiling water. Dried at 100' and analysed it gave the following results :- I.,1549 gram gave -2950 gram COz, and -0723 gram OH,. 11. *1328 gram gave -2539 gram CO2, and *0526 gram OH,. Theory C8H905. I. 11. C = 52.17 51-93 52-14 H = 4.34 4.46 4.39 0 = 43.49 - - In another preparation the residue, after distillation of the oil above described (ethyl pyrotritartrate), was not recrystallised from the mother-liquor, but at once from water. It was here noticed that the more difficultly soluble portions had a much lower melting point than the others, and fused to an oil before dissolving, so that the body appeared to be a mixture. An attempt was therefore made to separate the constituents by fractional decomposition of the sodium salts by means of dilute sulphuric acid. The sodium salt was prepared by neutralising with sodium carbonate.The first portions precipitated by sulphuric acid, when recrystalliaed from water, consisted of flat needles having a satiny lustre; they melted at 81-82", and were capable of neutralising a solution of sodium carbonate when warmed with it. On analysis the following results were obtained :- 2 1 2432 HARROW ON PYROTRITARTARIC AND I. $096 gram gave -1943 gram CO,, and *0484 gram OH,. 11. -1104 gram gave -2231 gram C02, and -056 gram OH2. Theory meov CsH7(C2Hj)Oj. Z(C,Ht(C,H,)O,).C,KsO,. I. 11. C = 56.60 55.26 55-21 55.11 H = 5.66 5.26 5.60 5.63 0 = 37.74 39.48 - - On boiling this substance for some time with dilute sulphuric acid long slender needles were obtained on cooling, which when recrystal- lised from water melted at 230-2.31". The latter precipitated fractions of the acid, also melted at 230-231" when treated in the same manner.The numbers obtained on analysis are- I. *1611 gram acid gave *3045 gram C02, and -0675 gram OH,. 11. -182 parnoacid gave -3435 gram CO2, and ,0714 gram OHZ. Theory C8Hs0,. I. 11. C = 52.17 51.55 51.47 H = 4.34 4.66 4-36 0 = 43.49 - - The acid crystallises from boiling water, in which it is pretty soluble, in very fine long needles, especially if a little sulphuric acid be present. It is nearly insoluble in cold water. The sodium and ammonium salts have a persist,ent bitter taste, and both are easily soluble in water, crystallising out in long needles. The sodium salt is but slightly soluble in alcohol. Analysis of Sodim Salt. I. Specimen prepared from pure acid by neutralisation with sodium *2107 gram dried at 100" lost -0409 gram = 19.41 per *1698 dried at 100' gave -0592 gram NhSOo = -0191 11.Crystallised out of mixture of sodium salts prepared in order to 0181 gram dried ,1460 dry salt gave carbonate. cent. OE2. gram Na = 11.25 per cent. Na. make the fractional precipitation mentioned above. at 100" lost -335 gram = 19.33 per cent. OH,. $0315 gram N&SOa = -01668 gram Na = 11.42 per cent. Na. Theory-CeH7NaOa requires 11.16 per cent. Na. ,, C8H7Ne05 + 23Aq requires 19.37 OH,; CsH7NaOs + 3Aq requires 20.77 per cent. OH,. Silver carbopyroti-itartrate is thrown down from a solution of the sodium salt as a curdy precipitate, which is insoluble in cold, but soluble in much hot water with but slight decomposition, crystallising out in fine needles on cooling.It becomes grey on exposure to light,CARBOPYROTRITARTARIC ACIDS, ETC. 433 but is a tolerably stable salt. were obtained :- gram Ag. -0448 gram Ag. On cornbustion the following numbers I. 0184.2 gram salt gave 02217 gram COZY *0421 gram OH2, and -067 11. -1229 gram salt gave -1473 gram C02, 00278 gram OH2, and 111. *1632 gram salt gave *0603 gram Ag. Theory C8H,Ag05. I. 11. 111. C = 32.99 32-82 32.68 - H = 2.40 2.54 2.31 - Ag = 37.11 36.37 36.44 36.94 0 = 27-50 - - - There was a well-crystallised sublimate on the sides of the crucible, which melted at 230°, and was therefore carbopyrotritaric acid; the sublimate on the lid melted much lower (130"), and was probably pyrotritartaric acid formed by the decomposition of carbopyrotritar- taric acid by heat.Ethyl carbopyrotritartrate.- Silver carbopyrotritartrate was sealed up in combustion tubes with excess of ethyl iodide and heated in the water- bath for three hours. The contents of the tube were treated with ether, and the extracts on evaporation left the carbopyrotritartaric ether in the crystalline sbate and slightly brown-coloured. By recrys- talliBation from ether it was obtained perfectly white. It is eoluble in ether, benzene, and alcohol, and when pure melts at 80-81"; it crys- tallises in needles, and when heated sublimes with partial decompo- sition. A combustion gave the following results :- -09 gram substance gave *188 gram CO, and 00464 gram OH2. Theory C8H;(C2H5) 05. Found. C = 56-60 56.9 7 H = 5.66 5.73 0 = 37.74 - In order to see if the original acid would be reproduced on saponifi- cation, the ether was boiled with dilute soda; on precipitation of the acid with sulphuric acid the melting point waa 217O, but on further boiling with dilute snlphuric acid the melting point was 230", showing it to be carbopyrotritartaric acid.It crystallises out in the character- istic long fine needles of the acid ; the substance therefore is a true ethyl salt, the ethyl-group not being directly combined with carbon. Behaviozcr of Carbopyrotritartaric Acid when Heated.-A small quan- tity was heated carefully in a test-tube with about double its weight of powdered glass; a sublimate was formed which waa crystallised from ether and re-sublimed. It melted at 135-136".434 HARROW ON PYROTRITARTARIC AND The experiment was repeated with a gram of acid aud the gas evolved passed through baryta-water, in which a precipitation of barium carbonate took place.A small quantity of carbonaceous matter remained mixed with the glass. The glass tube containing the sublimate was cut off, the latter dissolved out with ether, the ether evaporated, and the residue re-sublimed. The melting point was 135-136", and the following numbers were obtained on corn- bustion :- 01608 gram acid gave -3.546 gram C02 and *0818 gram OE,. Theory C7HSOp Found. C = 60.00 60 14 H = 5.71 5.65 0 = 34.29 - So that the reaction taking place on heating carbopyrotitartaric acid is From the fact that carbopyrotritartaric acid splits up on heating into pyrotritartaric acid, which undoubtedly contains a carboxyl group and carbon dioxide, it appeared probable that it belonged to the class of efheric anhydrides, and that it would furnish a dimetallic salt when submitted to the action of alkalis.In order to test this, -6555 gram carbopyrotartaric acid was boiled for an hour with 7.5 C.C. of normal soda (*4 C.C. excess of the quantity required to form the dimetallic salt), but on titrating back with normal hydrochloric acid, 3.75 C.C. were added before the solution became neutral to litmus, so that the carbopyrotartaric acid had appa- rently used up only 3.75 C.C. sodium hydride solution; the mono- metallic salt would require 3.56 c.c., the dimetallic salt 7-12 C.C. It was, however, possible that the dimetallic salt might have an alkaline reaction. Experiment 2. 06704 gram acid was therefore treated with 1.94 C.C.of a soda-solution containing ,08623 gram Na per c.c., equal to -1673 Na (the calculated amount to form this dimetallic salt is -1676 gram). After a quarter of an hour the solution had shill an alkaline reaction, and a small quantity gave with silver nitrate a brown precipitate con- taining silver oxide. The rest of the solution was heated for a short time, and after cooling silver nitrate was added to it, whereby a per- fkctly white precipitate was produced, showing that no free sodium hydroxide existed in solution, although the liquid still had an alkaline reaction. The silver precipitate was filtered off, washed, dried in a vacunm, and analysed, with the following results :- I. -2416 gram salt gave -214.3 gram COZ, 934 gram OH,, and -1279 gram Ag.CaH80, = CTHSO, + C02.CARBOPY ROTRITARTARIC ACIDS, ETC. 435 11. *214Agram salt gave *1924 gram C02, -0318 gram OH,, and 111'. -2602 gram salt heated carefully in a porcelain crucible left -1135 gram Ag. *138 gram Ag. Found. Calcd&d C8H8 &Og. I. I. 111. C = 23.08 24.19 24.47 - H = 1.92 1.56 1.64 - Ag = 51.92 52-93 52.94 53.04 0 = 23.08 - - - The analytical numbers would agree somewhat better with the formula CL~H14Ag4011 = 2C8H8AgZo6 - H20, which requires :- Cia = 23.59 Hi4 = 1-72 Ag4 = 53.07 011 = 21.62 The acid reprecipitated from this salt by dilute nitric acid was un- altered carbopyrotritnrtaric acid (m. p. 230'). Carbopyrotritartaric acid boiled for some hours with strong potavh was not decomposed, the reprecipit.ated acid melting at 230". Carbopyrotritartaric Acid fused with Potassiztrn Hydroaide.-l- 75 grams acid were fused with about 5 grams of potassium hydroxide until a small portion treated with sulphuric acid no longer gave a precipitate.The slightly brown fused mass was acidified with sulphuric acid and distilled. The distillate had a strong acid reaction, and smelt like acetic acid. A small portion, treated with H2SOa and a drop of alcohol, gave the characteristic pleasant odour of ethyl acetate. The distillate was neutralised with sodium carbonate and evaporated to dryness, the residue exhausted with alcohol, and the alcohol driven off on the water bath. The white salt left was dissolved in water and precipitated with silver nitrate. The precipitate, on recrystallisation from a little hot water, gave the flat silky needles of silver acetate.0.0738 gram left on heating 0.0476 gram, or 64.50 per cent. Ag. C2H,Ag02 requires 64.6 7 per cent. The residual liquid from which the acetic acid had been distilled was extracted several times with ether, which on distillation left a slightly yellow-coloured crystalline acid. It was dissolved in a little water, treated with animal charcoal, and the filtered liquid evaporated to the crystallising point ; after recrystallisation from ether it was perfectly white, and melted at 180". A small quantity on heating ghve very irritating vapours, and a pale, red-brown precipitate was produced in a solution of the neutral ammonium salt by ferric chloride ; a white insoluble precipitate was obtained with plumbic acetake.436 HARROW ON PYROTRITARTARIC AND This crystalline acid was, therefore, succinic acid, and the reaction taking place on fusion of carbonic acid with potash may be represented by the equation :- C~H805 I- 3OH2 = C4&04 + 2C2H402.The foregoing observations show that on boiling ethyl diacetosuc- cinate with dilute snlphnric acid, two reactions take place simul- taneously. On the one hand, the elements of ethyl alcohol are split off, ethyl carbopyrotritartrate being formed :- Cl2HI806 =C2H60 + C1&05 ; The latter is, however, soon resolved by the action of water into a second molecule of alcohol and carbopyrotritartaric acid,- cmHizo5 + H20 = CzH6O + C8HtiOa- On the other hand, carbon dioxide and alcohol are both split off, pyrotritartaric acid being formed :- CizHiaOs + OH, = COz + 2C2H,O + G H - 8 0 3 , The fact that carbopyrotritartaric acid is at the same time monobasic acid and etheric anhydride, that by heating it is resolved into pyro- tritartaric acid and carbon dioxide, and that it gives acetic and succinic acids on fusion with potash,-together with the known com- position of ethyl diacetosuccinate,-furnishes the key to the composition of the two acids obtained from the latter substance.Thus the forma- tion of ethyl carbopyrotritartrate probably takes place in the following manner :- CH3 I I CH--CH.CO + H2SO4 = C2HaHSOa + HZO + CH-CH. I CO.O.CZH5 ................ CH, CH2 CH$'H 1 ; I II 1 co co I ]0.C2R" co c-0 I I b ......................... 0 CO.O.C,Ha SO that the formula of carbopyrotritartaric acid would be- CH, CH, I II co c-0 I ! I CH- CH-CO I CO.OH The production of succinic and acetic acids from the dimetallic salt formed by the action of alkalis may be represented by the equation :-CARBOPPROTRITARTARIC ACIDS, ETC.437 CHs CH2 CH3 CH, I II I I I I I CO.0K CO.OK CO C.OH --- CH-GH.CO.OK + BHOK = CH,--CH,.COOK I CO.0K CO.0lK Carbopyrotritartaric acid when heated gives off carbon dioxide, forming the monobasic pyrotritartaric acid, thus :- CHS CH2 CH3 CHZ I II co c I II co c-0 I 1 1 I II I CH-CH.CO = CH-CH + COZ, I COOH CO.0H the production of which from pyruvic acid may be represented in the following manner :- cH ........... CH, 2iH i CH3 CHZ I . . . . . . . . . I II I II I co ClO I ......... I . co c I I . : . : ................... ICO,OiH id’iIjoH. = 2C02 + 2H20 + CH-CH .................... .......................... : CO.0H CO.0H The formation of benzoic acid, which Bottinger got by adding uvic acid to fused potash, precipitating with acid, and resubliming, would be comprehensible from the above formula, although rearrange. ment of the hydrogen atoms would certainly have to take place :- CH3 CH2 CH=CH co c CH-CH = HZ0 + C-CH I t I !I I II II II I I CH CH CO.0H CO.OH BO t tinger’s uvic acid formula- CO.OH I CHs-C=CH-CH=CH-CH.OH, would not explain the formation of the acid from ethyl diacetoeuccinate,438 SENIER AND LOmE ON A and again, it cannot be the correct one, as Wislicenus and Stad- nicki have proved that the acid does not contain a second hydroxyl group. I have to thank Professor Wislicenus for the very kind help which The research was conducted with a grant of 101. obtained from the he has given me in carrying out this investigation. Chemical Society’s Research Fund in July, 1877.

ISSN:0368-1645    

DOI:10.1039/CT8783300425

出版商:RSC    

年代:1878

数据来源: RSC

摘要:

438 SENIER AND LOmE ON A LII1.-A New Test for Glyceriw. By A. SENIER, M.D., and A. J. G. LOWE, Assistants in the Laboratories of the Pharmaceutical Society. I N a previous paper (Pharrn. Joum., viii, 819), we have shown that when certain acid borates, as, for example, ordinary sodium biborate, Stre dissolved by glycerin, the glycerin does not act simply as a solvent. we found that acid was evolved, which was probably free boracic acid, and that there remained behind more basic borates, having a constitu- tion between mono- and bi-borates. Klein (Compt. rend., lxxxvi, 826) says that with acid borates mannite forms conjugated acids. He describes briefly some compounds obtained, and then says that similar reactions take place in the case of glycerin, erythrite, and certain other polyatomic alcohols.It is true that mannite, erythrite, &c., evolve acid with sodium biborate in the same manner as glycerin, but the reactions are probably not the same. The test which we are about to describe is not given by mannite, glucose, &c., although it is readily obtained from glycerin, erythrite, and apparently from glycol. We are at present investigating this reaction more fully. It is interesting, however, to note here, that while we find the action of glycerin and erythrite to differ from the action of mannite and glucose, Klein also finds the action upon polarized light different in the case of the two former, from what it is in t h a t of the latter. The well known green colour which free boracic acid imparts to flame is usually obtained in the case of borates by the use of a stronger acid, but l l e s (Chem.News, xxxv, 204) suggests that glycerin may advantageously be substituted for the acid. This glycerin test for boracic acid involves, of course, the reaction which we have just discussed. It occurred to us that this reaction might also be employed as a test for glycerin, but before we describe the method of applying it to thatNEW TEST FOR GLYCERIN. 439 end, it will be convenient to discuss briefly its limits as regards other bodies, As might be anticipated from what we have said, monoborates do not give this reaction, at least that is the case with sodium mono- borate. A considerable quantity of alkali interferes with the test, doubtless by neutralising and fixing the acid set free.Again, a large excess of borax itself prevents the appearance of the green colour in the flame. This action of a large excess of borax is interesting. Now as to bodies which may be mistaken for glycerin. Acids give the same result to flame as glycerin, and hence the solution to be tested must be neutral or faintly alkaline. Erythrite, and apparently glycol, also give the same green flame, but could hardly occur so as to be mistaken for glycerin. The same may be said of copper salts and certain other substances, which of themselves might impayt a green colour to flame. Ocher substances, such as sugms, fats, &c., were tried, but all failed to give this test. It appears, then, in the absence of erythrite, and probably glycol, to be a property fairly characteristic of glycerin. The test is applied as follows :-The aqueous solution to be exa- mined is first made slightly alkaline with dilute soda-solution, and a borax bead is dipped into it.The bead is allowed to rest for a minute or so to allow solution to take place on its surface. Then it is held in the Bunsen flame in the usual way. Another method, and this we find more delicate, is to place a litkle powdered borax on a watch glass, and pour on to it some of the solution made faintly alkaline. Stir with a glass rod (the mixture becomes acid of course), and by means of a looped platinum wire introduce some of the mixture into the flame. I n either case the deep green colour appears more or less distinctly if glycerin be present in sufficient quantitly. The following experiments were made with the view of determining the extent to which the test could be applied to the detection of glycerin in various solutions.A 20 per cent. solution gave the colour very decidedly. Solutions con- taining 15, 10, and 5 percentages gave the colour distinctly. 2& per cent. gave a faint colour, and the limit seemed to be reached in the case of a 1 per cent. solution, which gave scarcely any colour. 25 and 20 per cent. solutions gave the colour decidedly. A 10 per. cent, solution gave if distinctly, but in the case of a 5 per cent. solution it could scarcely be seen. It was evident from these experiments that, in order to detect less than 10 per cent. in beer, the solution of glyceriii would hare to be concentrated, and some of the foreign matter re- moved.To try this we prepared two solutions of glycerin, one con- taining 5 per cent. and the other 1 per cent. Both were treated in the same manner. 50 C.C. were evaporated to dryness over a water-bath, The percentages are by volume, GZycemh i n Water.-The solutions were made faintly alkaline. Glycei-in in Beer.-The solutions were made faintly alkaline.440 SENIER AND LOWE ON A and the residue was treated with 50 C.C. of absolute alcohol. Solution of the glycerin in the alcohol was insured by well stirring the mixture and heating to boiling. The dishcontaining the mixture was then set aside to cool and subside, after which the clear alcoholic solution con- taining the glycerin was decanted into another dish. The solution was evaporated to dryness as before, a3d again exhausted with absolute alcohol.The solution from this second treatment with alcohol was evaporated nearly to dryness, and a few drops of water added. The solution thus obtained was made faintly alkaline to delicate litmus paper with dilute soda solution, and tested in the manner already described. In each case the colour was distinct, of course most decided in the case of the residue from the 5 per cent. solution. A solution containing one quarter of 1 per cent. was examined in the same way, and gave the colour distinctly. Beer, to which was added only one- benth of 1 per cent. of glycerin was examined by first concentrating and shaking wi6h petroleum ether, to remove hop resin, and then evaporating to dryness and treating as above. In this case the green colour, though less intense than before, was distinctly seen.Several experiments were made with beers to which glycerin was not added ; but treating them in the way described no green colour could be obtained. It was noticed that the consistence of the residues from beers to which glycerin had been added, compared with the residues of beers containing no glycerin, was very different. The former were, of course, moist and syrupy, while the latter were much harder and even brittle. GZycerigb ia Wine.-To some sherry wine was added 1 per cent. of glycerin. Of this solution 50 C.C. were evaporated to dryness over a water-bath. The residue was treated, as previously described, with 25 C.C. of absolute alcohol. The alcoholic solution was then evaporated over a water-bath nearly to dryness, and a few drops of water added.This solution, made faintly alkaline and te,sted, gave the green flame distinctly. The same wine without glycerin gave no green colour to flame. Glycerih ir, Mi'Zk.-l00 C.C. of milk, to which 1 per cent. of glycerin was added, were evaporated to dryness over a water-bath. The residue was treated with 50 C.C. absolute alcohol, and the alcoholic solution so obtained evaporated to dryness as before. A few drops of water were added, and the mixture filtered through a moistened filter to separate fat. The filtrate was then rendered alkaline and tested. 100 C.C. of milk without glycerin similarly treated gave no green colour. The fat of milk likewise gave no colour rettction. GZycerin. in Z'rencZe.-50 C.C.of treacle, in which was dissolved by aid of heat and agitation 5 per cent. of glycerin, were, while warm, well The green colour was distinctly observed.NEW TEST FOR GLYCERIN. 441 ayitated with absolute alcohol. The mixture was set aside, and allowed to cool and separate. The alcoholic solution was then decanted, and evaporated over a water-bath to dryness. This residue was treated with 50 C.C. of absolute alcohol in the usual way, and the alcoholic solution so obtained evaporated over a water-bath to dryness. This second residue was treated in the same way with 25 C.C. of absolute alcohol, and the solution obtained was evaporated as before. The residue was diluted with a few drops of water, and the solution made faintly alkaline. A 1 per cent. solution failed to give the reaction, thus showing at the same time that treacle itself treated in this way is incapable of pro- ducing the green colour. It may be said then of the test for grycerin hero proposed that it is practically characteristic, and is delicate.It can be easily applied, and is not interfered with by the presence of a considerable quantity of foreign matter. We therefore commend i t to trial by chemists as a substitute for the known tests for glycerin, which are not only few but objectionable. In conclusion we have to express our indebtedness to Mr. H. G. Greenish for valuable assistance, and to Professor Attfield for kindly permitting these experiments to be conducted in the Labora- tories of the Pharmaceutical Society. Tested, it gave the green colour distinctly.Authors’ Note.-In the discussion which followed the reading of the above paper our attention was directed to the fact investigated by Woodcock (Jour. Ghem. Soc., xxiv, 785), that ammoniacal salts have the power of decomposing borax and liberating boracic acid. It, was suggested that these salts might be mistaken for glycerin unless further precautions were taken. We find this to be the case. Substances to be examined for glycerin by the test we have sug- gested should be first tested for ammonia, and should this be present the above methods must be SO modified as to secure its complete exclusion. To determine the presence of ammonium salts, sodium carbonate, and in some cases the Nessler reagent, should be em- ployed. The following two experiments will suggest how ammonium salts may be separated. When fats occur they must, of course, be removed prior to treatment with alkali. 50 C.C. of an aqueous solution, containing 1 per cent. each of ammonium chloride and glycerin, were evaporated in a platinum dish over a water-bath to dryness. Saturated solution of sodium carbonate was then added, and the heat of the water-bath continued until no more ammoniacal gas was evolved, and the mixture was distinctly alka- line. The mixture was then extracted with absolute alcohol, and the442 WATTS AND BELL ON THE solution evaporated, neutralised, and tested. The green colour was distinctly seen. 50 C.C. of beer, containing 1 per cent. each of ammonium chloride and glycerin, were evaporated to dryness over a water-bath. The ammoniacal salt was separated from the residue as before. The mixture thus obtained was treated with absolute alcohol, as described in the paper, and tested, the final solution being neutral, or only faintly alkaline. The green flame was very distinct. 50 C.C. of beer containing ammonium chloride only were examined in the same way, but gave no green colour to flame.

ISSN:0368-1645    

DOI:10.1039/CT8783300438

出版商:RSC    

年代:1878

数据来源: RSC

摘要:

442 WATTS AND BELL ON THE LIV.-On. the Preparation of Metallic Chlorides. By CHARLES W. WATTS, and CHICIIESTER A. BELL, B.A., M.B., of University College, London. THE two principal processes €or the preparation of chlorides in the dry way consist in (I), passing chlorine over the heated metal; and (2), passing chlorine over a mixture of the oxide with charcoal. The first is usually troublesome, on account of the difficulty attending the preparation of many of the metals ; and the second, on account of the fine state of division to which it is necessary to reduce the oxide, if it is desired to obtain any considerable portion of the metal in the state of chloride. This being the case, it appeared advisable to find some more convenient process, and carbon tetrachloride seemed likely to afford a nieans for the preparation of metallic chlorides, and more especially of the higher chlorides.The anhydrous oxide was placed in a long porcelain tube, connectgd at one end with a small retort, in which the carbon tetrachloride was kept in a state of gentle ebullition, and at the other with a cooled receiver. When the tube was heated to bright redness, and the vapour of carbon tetrschlor- ide passed slowly over the oxide of tin, a liquid soon began to collect in the receiver, accompanied by a small quantity of a yellowish solid, either oxychloride or hydrated chloride of tin. Carbonic acid and chlorine escaped from the receiver (the chlorine being derived from the decomposition of some of the carbon tetrachloride), and a small quan- tity of stannous chloride was deposited in the cooler parts of the tube.When the contents of the receiver were fractionally distilled, a small The first experiment tried was with stannic oxide.PREPARATION OF METALLIC CHLORIDES. 443 portion came over a t about 80°, consisting for the most part of carbon tetrachloride, and the principal portion of the liquid distilled over at 118-120", the boiling-point of stannic chloride ; it fumed strongly in the air, and united violently with water, forming a solution of stannic chloride. The product amounted to about 40 per cent. of the tetra- chloride used. A small quantity of solid product was left in the dis- tillation flask, subliming in white needles, and consisting of carbon monochloride, C2Cl,. The next oxide used wm titanic acid, the experiment being conducted in precisely the same way ~ E I with stannic oxide. The titanic acid was introduced into the tube in lumps, about the size of buck-shot.The product amounted to about 50 per cent. of the tetrachloride used; it was accompanied by a small quantity of ferric chloride. The residue in the tube consisted of unaltered titanic acid, and of skeletons of the lumps of titanic acid, consisting probably of silica. Similar experi- ments were tried with silica, both precipitated (from SiFJ and well- burnt black flint. The tube, about 30 inches long, was heated nearly to whiteness, but in both cases only the merest traces of silicon chlo- ride were obtained, even with the most efficient condensing arrange- ment. Tungstic acid and carbon tetrachloride gave a mixture of WCl,O, and WC1202, the principal reaction being probably expressed by the equation- WOS + 2ccl4 = 2w02c12 + WOCI, + 2c02.A similar experiment with chloroform and titanic acid gave a quan- tity of titanic chloride, equal to about 30 per cent. of the chloroform used. A large quantity of secondary product was obtained, consisting of titanium oxychloride, chlorides of carbon, &c. ; a notable proportion of the titanic chloride was also carried off by the gases, hydrochloric and carbonic acids, produced in the reaction. The formation of titanous chloride was not noticed. Carbon tetrachloride being somewhat expensive, and a notable pro- portion being always lost, a mixture, in equal volumes, of carbonic oxide and chlorine was tried as a means of producing these chlorides.With titanic acid a very good result was obtained, although some quantity of the titanic chloride was carried off by the carbonic acid produced, which has double the volume produced with carbon tetra- chloride: TiOz + 2CO + 2CI2 = TIC& + 2COZ. Silica again was hardly acted on. Tungstic acid gave a mixture of the two oxychlo- rides, but when a larger proportion of carbonic oxide was used, a small quantity of nearly black crystalline chloride (WC16 or WCl,) was obtained. 5 grams of tungstic acid placed in a small porcelain tmbe and heated to full redness, was entirely volatilised as oxychloride after444 WATTS AND BELL ON THE PREPARATION, ETC. about 3 litres of gas had been passed over it. Amorph.ons chromic oxide was acted on with great readiness, the crystalline oxide not quite so easily; in both cases, however, an abundant production of chromic chloride resulted, which, when the tube was heated a few inches beyond the oxide, was non-deliquescent, and completely inso- luble in water, showing the absence of any chromous chloride.Red oxide of manganese, Mn30a, fres from iron, was acted on very easily, but the chloride being non-volatile, it could not be perfectly separated from the unaltered oxide. A very small quantity of appa- rently colourless crystals, containing manganese and chlorine, sub- limed to the mouth of the tube at the beginning of the experiment,, but the amount was too small for f nrther examination. A similar experiment was tried with bromine, carbonic oxide, and stannic oxide. The dry carbonic oxide was bubbled through bromine kept warm in a small retort, and covered with strong sulphuric acid to prevent surface evaporation. Stannic bromide volatilised into the receiver, and some stannous bromide was left in the tube. In all these experiments the carbonic oxide was previously prepared, and stored in a gasometer. Experiments were made in which a mix- ture of 1 vol. carbonic acid with 2 vols. chlorine, was passed through a tube, three parts filled with carbon, and the rest with oxide, sepa- rated from one another by x plug of asbestos. Both wood-charcoal and coke were found very wasteful of chlorine, on account of the hydrogen they contain ; and with gas-carbon, although carbonic acid alone was easily converted into carbonic oxide, the mixture of chlorine and carbonic acid passed through almost unchanged. These experiments were made in the Laboratory of University College. The authors intend pursuing the subject, more especially with regard to the higher chlorides.

ISSN:0368-1645    

DOI:10.1039/CT8783300442

出版商:RSC    

年代:1878

数据来源: RSC

摘要:

445 LV.-Laboratory Experiences on board the ChaEle.iLger.” By J. Y. BUCEANAN. [A Lecture delivered before the Chemical Societpj Febmq 21st 1878.1 IN commencing my lectuye npoa laboratory experiences on board the ‘‘ Challenger,” it will be fitting that I should describe the laboratory in wbieh they were made. The laboratory was situated on the main deck jnst abreast of the mainmast and therefore almost exactly in the middle of the length of the ship on the right-hand side looking forward. It was lighted and ventilated by a port which under ordinary cimumstances would ac-comniodate a gun. Besides this window or port there were windows into the deck capable of supplying a certain amount of borrowed light. Against the foremost partition was fitted a locker the top of which formed a seat and indeed occasionally a bed and the interior formed a storehouse for instruments and spare apparatus.Above it and fixed to the same partition were two book-shelves which accommodated the greater part of our chemical literature. In front of the seat was a writing-table attached at one end by hooks to the one sash of the window and supported on the other end by folding iron legs. It was capable of easy removal and indeed was only “shipped” when the laboratory became a study. On the othcr side in front of the seat, was zb blow-pipe table of common deal with a double action circular bellows of eight inches diameter underneath. A tallow lamp was supplied to me for use with the blowpipe as being the kind of lamp used by glass-blowers by trade.No doubt for such persons it is suitable enough but in a laboratory a blowpipe is wanted often for a few minutes at odd times when it would be impos-sible to spend the time necessary for the preliminary melting of the tallow and I finally adopted a spirit-lamp made out of a wide-mouthed four-ounce bottle with zb piece of metal made to take a wick just as broad as the mouth would accommodate and with it it was possible to do every sort of glass blowing that was necessary. The blowing of lead glass was especially easy with it. The wind from the bellows was directed across the burning wick by means of a glass jet in the ordinary way. For delicate glass blowing an arrangement of this sort is pre-ferable to any of the forms of gas blowpipe because there is not the glare of the smoky flame andmoreover lead glass can be worked with VOL.XYXIIT. 2 446 BUCHANAN’S EXPERIENCES OF THE LABORATORY the utmost facility without blackening and those who have conquered their first repugnance to lead glass and have learned to work it know how much pleasanter it is to handle than even the softest German glass. On the sternward side of the window was the working bench which was constructed of teak. The height was 3 feet 10 inches which was found extremely convenient for working at as there was no necessity for bending. The re-agents also were accommodated in drawers. Each bottle was par-titioned off so that in the rolling of the ship they could not roll against each other. For smaller bottles these drawers were divided in the middle so as to form two trays one behind the other.This arrangement was convenient because a drawer filled with reagent bottles could thus be taken out and put upon the bench for use. It was not liable to be disturbed by the rolling because it was kept in its place by the battens which surrounded the table. In order to support flasks or apparatus of any kind it was of course impossible to take the ordinary retort stand of the laboratory because it would be difficult to fix it in such a way as would not be absolutely permanent and still secure. I therefore adopted this stand taking the metal rod half an inch in diameter which is used for retort stands having a longitudinal eye forged upon it like the eye of a needle which was then fastened into the beam here by an eye-bolt and had therefore a certain amount of play up and down.The rod was then shoved into a hole in the thick teak board on the top of the working bench the retort stand rings and lamp such as a spirit-lamp being slipped up from below before having the rod put into the hole in the top of the bench. This was an exceedingly convenient apparatus because when not in use it could be slipped up and the rings taken off from below and the rod put up against the roof with its end in a hook overhead and therefore out of the way. This was used then for setting up apparatus such as that for determining the carbonic acid to which I shall refer presently. For drying and evaporating a cast-iron gimbal support was fixed to the ship’s side. It was capable of carrying a sand-bath or water-bath or drying chamber always in a level position.In the corner there was a sink that is to saya pipe which led away to a scupper draining water off the main deck. Distilled water sup-plied by the boiler of the ship was kept in a bottle and always ready for supply. The glass apparatus was arranged in the ordinary way on shelves with double perforation to keep each article in its place. Up above the beams which support the deck were of‘ considerable dimen-sions. They projected about 9 inches down from the deck above so 8s to afford considerable room between them. These were all taken It was fitted in the usual way with drawers ON BOARD THE ‘‘ CHALLENGER.” 447 advantage of for putting away apparatus and articles which were not in constant use.They resembled graduated pipettes the lower end being fitted with a pinchcock. To the upper one an india-rubber tube could be fixed with a glass tube for sucking. When any reagent was going to be used in the burettes a glass tube of such length as to reach well down into the bottom of the bottle was attached to the point of the burette and plunged into the liquid which was sucked up into the burette. One of the reagents in constant use was baryta-water for the determination of the carbonic acid in the sea-water. By using this method of filling the burette so that none was spilt on the neck of the bottle no more liquid was used than was absolutely necessary, and no more air got into the bottle than there was liquid taken out of it. A bottle holding rather more than a litre would last two or three months; and taking a portion each day its standard did not altes sensibly.The remaining arrangements are not of any special interest and need not therefore be further referred to. The principal defect of the laboratory was the want of means of removing noxious vapours. From the position in the ship it was found impossible to provide any kind of draught chamber the only means of ventilation being by the door or window; hence mineral analysis of any extent was rendered impossible. The dimensions of the laboratory were ten feet long and five feet eight inches wide the height being exactly six feet at the highest part. The space therefore was somewhat confined but for one worker it was quite sufficient and after getting accustomed to the motion of the ship there was no great difficulty in working in all ordinary weather.In fact as long as the weather was not so boisterous as to make it necessary to close the outside port the work could always be done. With the port closed light and ventilation were shut out and work became impossible on that account alone even supposing the ship had been steady enough. Having thus described the nature of the laboratory I shall go on now to notice the experiments made with these instruments. Before starting on the expedition it appeared to me to be of great import-ance to have some means of controlling the depth as ascertained by the sounding line. The records of deep sea soundings contained depths of as much as 8,000 fathoms which had been obtained by various ob-servers without bottom having been reached.This was explained by say-ing that in the ocean if there is no current at the surface there is such a multitude of under-currents going in ways which cannot be seen OP The burettes were fixed to the shelves. 2 K 448 BUCHANAN’S EXPERIENCES OF THE LABORATORY estimated that all sounding in this way must be fallacious. Of course having no practical experience of it myself I could form no opinion, but I thought it was desirable to endeavour to have some means of checking the depths as given by the sounding line and the first and most ready means which occurred to me was to adopt the barometric method. After studying and experimenting in various directions I decided to adopt as a measure of the pressure the apparent contraction produced by it in a mass of water enclosed in a glass envelope.During the spring of 1872 I made experiments on an instrument of special dimensions in the deepest water whioh was to be had near home namely in Loch Lomond where khe depth in some parts is as much as 105 fathoms. I thus found that the method was capable of practical application. In Fig. 1 hhe piezometer is depicted in It consists essentially of a thermometer-shaped section, FIG. 1. instrument open at the end. A cylindrical bulb A, contained in the one that was chiefly used about 9 C.C. The stem which was rather more than a foot long bad a diameter of almost exactly one millimeter. The end of the stem dips into the bulb D which was filled so far with mercury and the instrument was set by heating it to suoh a temperature that when it cooled down to the atmospheric temperature the mercury would rise to a con-venient height so as to be visible and to be able to be read at any moment at a given temperature by plunging it in water.The arrangement for protecting the open end of the instrument is somewhat peculiar. It is necessary to allow the water on the outside to get access to the mer-cury in the bulb in order that the pressure may exert itself in the interior of the instrument in the same way as ail- must have access to the mercury in the reser-voir of the barometer. At the same time it is of impor-tanoe that the mercury should not be able to come out of the bulb. The means therefore that have been taken for this purpose was to have a bulb D blown into the neck of which the stem of the instrument fitted with some accuracy.This was connected with the stem by means of a piece of india-rubber, which waa prevented from fastening hermetically on the stem by having a small piece of glass rod H shoved in between the india-rubber and the stem. In this way communication was constantly kept open between the outer water and the mercury in the bulb. The stem of the instrument was divided into millimeters and was carefully calibrated the weight of the water filling the instrument, and also the coefficient of expansion of the glass were at the same time determined ON BOARD THE ‘( CHALLENGER.” 449 If the position of the water-mercury meniscus in the stem be noted under observed conditions of temperature and pressure and the instru-ment be then observed under different conditions of temperature and pressure the apparent volume occupied by the water and therefore the position of the meniscus will depend on the difference of the com-bined effects of temperature and pressure on the water and on the glass.This resultant effect is measured by the position of a magnetic index similar to or in fact exactly the same as that used in Six’s ther-mometer. The deep-sea thermometer which was used was after Six’s pattern with a protected bulb. When the instrument is subjected to increased pressure or diminished temperature or both together, the index is pushed up by the mercury which enters owing to the decrease of temperature and the increase of pressure and its posi-tion thus gkes us the sum of the effects of change of pressure and of temperature on the apparent volume of the water, If now along with this instrument a sufficiently protected thermo-meter has been attached to the line and its readings are taken at; the same time we have a measure of the temperature to which the instru-meut has been subjected.Knowing the dimensions of the instrument. in every particular and its behaviour under varying conditions of temperature we can subtract from the whole reading of the instrument that which is due to temperature and the remainder is that due to pressure. I f we know the coefficient of apparent compressibility of the liquid the depth is given at once. Now before starting I thought that the pressure would be more accurately measured by means of a gauge attached to a hydmulic pump, such as that used by Mr.Casella for testing the strength of the pro-tected bulbs of the deep-sea thermometers. Mr. Casella was good enough to put his pump at my disposal for experiments on my original instruments and their performances were satisfactory. After we had been at sea for some time and had made some experi-ments on the sounding line it became evident from the readings that the depths so found were not fo be relied upon-that in point of fact the gauge which had been used to determine the pressure was not a trustworthy standard and on reflecting upon the matter it became evident that in cases where one has a satisfactory sounding there is no better gauge of pressure than the length of the sounding line.I consequently then directed my attention to determine the apparent compressibility of distilled water and some other liquids by means of the sounding line that is to say usiug the sounding line as the gauge of pressure and taking particular care to observe that these experiments were made when the sounding was not vitiated by per-turbing causes. When currents are present they are always very evident from the behaviour of the sounding line. If the sounding lin 450 BUCHANAX’S EXPERTENCES OF THE LABORATORY remains vertical during the whole of the sounding then we may be perfectly certain that there is no disturbance from currents either at the surface or below. If there is a current of any appreciable force, the sounding line begins to wander about and has to be followed by the ship.This is an operation of considerable delicacy even in good weather and in bad weather when the winds and currents cross and complicatle each other it is one which calls for the highest skill on the part of the officer in charge. There is therefore no difficulty in deter-mining if a sounding has been good and only such soundings free from vitiation by any of the above-mentioned perturbing causes were used for our purpose. It was only after we had been for a considerable time at s e e i n fact when we were really on the way home that I was able to make a sufficient series of experiments with this instrument to determine satisfactorily the apparent compressibility of water. The instruments occasionally got out of order and other work prevented my imme-diately putting them right.Tben it was necessary to get a certain amount of skill in working because repairing instruments of this kind especially at sea was not learnt in a day. The indexes worked very well in the deep-sea thermometer containing a solution of carbolic acid, which appears to preserve the elasticity of the hair (the spring of the index is a human hair which clings to the side of the tube and keeps it .in position) but not so well on being immersed in distilled water or saline solution which seems to have the effect of impairing the elas-ticity in a certain length of time. The consequence was that the indexes began to get loose and in the unavoidable jars of sounding they were shaken from the position which they had taken up at the deepest part and became untrustworthy.And here is where the useful-ness of the removable mercury bulb into which the stem .dipped became apparent. Supposing the end had been permanently fitted with any of the glass blower devices such as a double safety bulb, there would have been no means of getting at the insideof the instru-ment which would thus have become useless as soon as the index had become loose. By the very simple operation of removing the instru-ment from the ebonite backing slipping the bulb off and warming the instrument to expel the mercury from the stem the index could be taken out with a magnet and a fresh hair put on and the instrument set again when it was as good as new. In this way then a number of experiments were made when the ship was on the way home between Tahiti and Valparaiso in a long stretch of water of moderate depth that is to say between 2,000 and 8,500 fathoms where the soundings were always very satisfactory ; the weather was tolerably good and there were no appareut currents at all.Fifteen very satisfactory soundings were thus obtained wit ON BOARD THE " CHALLENQER." 451 this instrument and the result was to give the coefficient of mean apparent expansibility of distilled water in glass as 0.0009 per 100 fathoms of average sea-water. This corresponds to 0.0000491 per atmosphere which is not very different from what Ltegnault obtained at very low pressures and Aim Q experimenting in the Mediterranean got 0*0000486 per atmosphere. I have used however as the unit always 100 fathoms because for our purposes it is more convenient.From experiments made with another instrument at an earlier date, the mean apparent compressibility of water at pressures between 0 and 3,000 fathoms was found to be 0*000878 per 100 fathoms. In an instrument precisely similar the compressibility of sea-water was determined. The results obtained were very concordant. Up to 800 fathoms the rtpparenf compressibility was 0.000770 up to 1,000, 0.000787 and 1,500 0.000753 the temperature being between 2" and 3" C. The mean value is 0*000783 and the ratio to the apparent com-pressibility of distilled water is 0.87. The dimensions of the instru-ment did not allow of its being used a t greater depths. These experiments were all made on the sounding line and at tem-peratures which were comparatively low.At all depths above 800 or 1,000 fathoms the temperature of the ocean varies but little. It is always somewhat under 4" and generally under 3" C. and usually above 1" C. so that it may be said that these compressibilities are good for an interval of temperature between 1' and 3" C. In order to find out the compressibility at higher temperatures there was on board n hydraulic instrument similar to what Mr. C as e 11 a used for testing his thermometers. It was somewhat larger than Mr. Casella's made for the purpose of being taken away. In it a number of experiments were made. at different temperatures. But here we had no gauge, except an artificial one which we found already was not to be relied on.Consequently the only determination that could be made was the relative apparent compressibility of various liquids referred to that of distilled water as unity. Adopting one of the instruments filled with distilled water which had been used frequently on the sounding line as a gauge we knew approximately what the pressure was but we could not of course say that a compression of a certain amount at 20" C. re-presented the same pressure as the same compression at 2" C. and consequently in the result of these experimenis I have only put down the relative appai-ent compressibility of these substances. The apparent compressibility of distilled water at the temperature of the water of the pump being unity the relative apparent compressibility of sea-water and of a 4 per cent.chloride of sodium solution at different tem-peratures are given in the following table : 452 BUCHANAN’S EXPERIENCES OF THE LABORATORY 4 P.0. NaCl Relative apparent compressibility of I Mercurg* Distilled Temperature C. Sea-water. solution. , 26 ‘5’ 25 -8 22 *5 22 -1 13 -7 13 ‘3 2 5 water. 0 -9360 0 a9191 0 *9214 0 *9283 0 *9264 0 -9203 0 -8950 ‘ -0 *0260 0 -0247 0 *0325 0 -9051 0 ‘8813 0 -8661 0 *8684 --1 1 1 1 1 1 -The relative apparent compressibilifies appear to diminish with the temperature. This probably is owing to the fact that the compressi-bility of distilled water was found by Regnault at low pressures at least to diminish with the temperature. Now the compressibility of sea-water probably does not diminish so much and that of chloride of so-dium probably still less-possibly it even increases with the temperat*ure.These results at any rate would tend to show that the compressibility differs in that direction. In the fourth column of the table the values of the relative apparent compressibility of mercury are given. To this I shall now refer. The compressibility of mercury was determined in an instrument (Fig. 2) the same as the water-piezometer ; the stem, FIG. 3. horn-ever was bent round so that the mercury should not have a ten-dency to flow down the stem. Tbc bulb A contained aboat 250 grams of mercury about 19 C.C. The stem was filled with water through ON BOARD THE (( CHALL~NGER." 453 considerable portion of its length BC in which a magnetic index worked.This instrument was finished off in exactly the same way as tbe water-piezometer that is by dipping the open end into a cup of mercury (see Fig. 2). Here again the instrument registered the sum of the effects of temperature and pressure. On the sounding line ex-periments were made with this instrument. A number of observa-tions were obtained at 800 fathoms and at 1,000 fathoms with very concordanf results. There were a number at other and greater depths, and those were plotted in the ordinary way on paper and through them an average line was drawn so as to give the compressibility of the mercury. The result was that the apparent compressibility of mercury for 100 fathoms was 0*0000271 being per atmosphere 0*0000015.The results are given in fhe annexed table. Similar observations were then made in the pump. Temp. Comp. 22.3 0.0301 13.3 0.0261 11.5 00247 There is a remarkable difference in the result at 11.5" and 13.3" from those obtained at lower temperatures than at 22.3" in the pump. The results at 11.5" and 13.3" C. were obtained by two different instm-ments and are each the mean of about 15 different observations SO that they are probably correct. On the sounding line at temperatures between 1" and 3" C. the relative apparent compressibility of mercury was found to be 0.0822 0.0327 and 0.0301 in three different series of observations. It is impossible t o say to what the low relative appa-rent compressibility at these temperatures is due and the subject is worthy of further investigation.Having then determined in this way 'the apparent cornpressibilities of these liquids it was easy to apply it to the practical purposes of sounding. One of these instruments was attached to the line and the apparent contraction cleared for temperature gave the depth. It was however found that in many places the temperature remained apparently constant for a considerable depth ; that is to say after a certain depth we will say for instance 1,000 fathoms was reached, then if the depth of the sea was perhaps 2,500 the readings of the Casella thermometer showed constantly the same temperature all the way from 1,000 fathoms down to the bottom. This afforded evidence that either the temperature was the same all the way from 1,000 fathoms down to the bottom or it was higher than at 1,000 fathoms; but there were no means of ascertaining thia unless the temperature really rose to be higher than that a t the Burface where of course th 454 BUCHANAN'S EXPERIENCES OF THE LABORATORY maximum index of the Millar-Casella thermometer showed it.FOY this purpose the mercury-piezometer came into use and it was for that purpose principally it was made. As in the case of the water-piezo-meter its readings show the combined effect of change of temperature and of pressure either of which being given the other can be deduced. But it will be evident that if after reaching a certain depth the temperature remains constant the apparent volume of the liquid will diminish in simple proportion to the increase of pressure and if the temperature increases from a certain depth downwards the apparent volume of the liquid will diminish in a less proportion than the increase of pressure.If this increase of temperature is such as to expand the liquid so as just to compensate the compressing influence o€ increase of pressure the index will remain stationary. The method of using the two instruments combined will be evident from a reference to the figure. The change of volume of water with change of temperature at the low temperatures found in the deep seas is very slight. The change of volume of mercury however for all ordinary temperatures is very considerable. On the other hand the compressibility of water or its sensibility to change of volume with change of pressure is very great, whereas that of mercury is very small.Consequently by sending a pair of these instruments down upon the sounding-line and reading them when they come up we have two independent values of the sum of the effects of change of temperature and of pressure. Taking as the first approximation to the depth the length of the sounding-line, applying it to the reading of the mercury instrument and so correct-ing it for pressure we have a first approximation to the temperature ; applying this temperature to the reading of the water-piezometer we obtain a second approximation to the depth indeed the true depth. The reading of the mercury-piezometer now being corrected for pres-sure by this value of the true depth we have a second approximation to the temperature.In fact we have now practically the true depth and the true temperature. In the figures the thick lines represent the apparent changes of volume for change of pressure and the dotted lines the apparent changes of volume for changes of temperature. Fig. 4 a refers to the water-piezometer and Fig. 4 b to the mercury-piezometer. Distances measured along the horizontal line of abscissae represent depths in the scale of 0.01 inch to a fathom and tempera-tures on the scale of 0.1 inch to a degree Centigrade. Distances measured along the line of ordinates represent scale divisions (milli-meters) on the scale of 0.1 inch to a division. For 100 fathoms of depth the apparent contraction of the mercury instrument was 0.7 millimeter on the stem; in the water instrument the apparent con-traction for 100 fathoms was somewhat over 7.8 millimeters.Con-sidering that the effect of a change of temperature of 1" C. causes a ON BOARD THE ‘ 6 C ~ A L ~ ~ ~ G E R . ” 455 apparent change of volume represented by about 2.5 millimeters while in the water-piezometer at the low temperature always found in deep FIG. 4 a. seas the temperature may be anything between 0” and 10’ C. without altering the apparent volume of the water by more than 2 millimeters on the stem we see that an error in determination of the depth of 100 fathoms would only make a difference in the reading of the mercury instrument of about 0.6 millimeter equivalent to a difference of tempemture of about 0.25” C. Therefore applying the possibly erroneous depth given by the sounding-line to ‘‘ clear ” the reading of the mercury-piezometer for effect of pressure we obtain a first qproxi-mation to the temperature which would almost always be within half a degree of the truth but which might occasionally differ more than a degree from it.Using the temperature thus found to clear the read-ing of the water-piezometer for the effect of temperature we obtain a second approximation to the depth which cannot differ appreciably from the true depth. Applying the depth so found to clear the read-ing of the mercury instrument for effect of pressure we obtain a second approximation to the temperature which cannot differ appre-ciably from the truth. This process of gradual approximation may of course be carried as much further as we please but the results obtaiued in the second approximation may under all circumstances be taken as representing the truth.In Fig. 1 (page 448) the stem of the water-piezometer is repre-sented as being swelled into a small bulb at F. The purpose of this bulb is to enable the instrument to be used at depths so great that with a uniform stem the contraction produced would be equal to the whole volume of the stem. The capacity of F is equal to the contraction 456 BUCHANAN'S EXPERIENCES OF TRE LABORATORY prodnced by the fall of temperature and the first 1,000 or 1,500 fathoms of depth so that the instrument would only register depths greater than 1,000 or 1,500 fathoms but it would do so with as much precision as can be obtained for less depths. I also made an attempt to determine the compressibility of glass in order to get the true compressibility of these liquids.I shall only allude to it in order to indicate the method-the result not being sufficiently accurate to be applied yet. A tube such as that used for the stems of the piezometers of about 1 millimeter bore and graduated into millimeters was sealed at one end. A glass rod was then drawn out to such a size as to pass easily into the tube and about 2 inches shorter than the tube which was made of the extreme length of the receiver of the pump. The rod having been iutroduced into the tube and a magnetic index on the top of it the end of the tube was sealed up. When now the tube was subjected to high pressure the material of the tube was compressed and diminished in volume in every direction and of course therefore in its length.The rod however in the inside not being exposed to extra pressure maintained its original length. The index having been brought down home on the top of the rod before exposing the system to pressure was kept in its place by the rod and the tube on the ont-side was therefore pieessed down past it to a certain extent. When the pressure was relieved the tube recovered its length and slightly lif'ted the index. The extent to which the index was free above the rod when that pressure was relieved measured the linear compression of the glass ; and 'multiplying that by three we had the cubical compres-sion. The results obtained by these means were too rough to be quoted as of any particular value because measurement of this dis-tance was effected simply by the eye on the scale of millimeters which was etched on the tube but with an apparatus I have since had con-structed I expect to obtain results of great accuracy not only for glass, but for other solids.I shall now direct your attention to a subject of a somewhat more chemical nature namely the atmospberic contents of the sea-water. The carrying out of these investigations was one of the principal ob-jects of having a chemical laboratory on board the ship and that we were able to do so successfully I owe in great measure to the fact of having met Professor Jacobsen the chemist who was in the Prus-sian expedition to the North Sea in the year 1872 who very kindly showed me all the apparatus which he was using and especially his adaptation of Bunsen's metlod of boiling nitrogen and oxygen out of the water which was of the greatest value; in fact it made the opera-tion possible on board ship which otherwise it would not have been.In order to boil atmospheric gases that is to say oxygen and nitroge ON BOARD THX “CHALLENGER.” 457 (because in sea-water carbonic acid may almost be said to belong to the fixed constituents as it is quite impossible to eliminate it in the ordinary way in vaeuo) the general arrangements of Bunsen’s appa-ratus was used. The apparatus has been fully described by Professor Jacobsen in the AnlnuJem der Ckemie ulnd Pharmacie for May 1873. Its speciality consists in the means of making and breaking connec-tion between the flask and the glass tube dest,ined to receive the air eliminated from the water.The water from the bottom was usually colleoted in the so-called ‘‘ slip” water-bottle which has been de-scribed by Professor Jacobsen. Water from intermediate depths is obtained in an instrument represented in section in Fig. 5. It is made entirely of brass which however might ad-vantagedusly be nickel-plated. It consists of a cylin-der A terminated at both ends by similar stopcocks, B B which are connected by the rod C. This rod carries near its upper extremity a piece of stout sheet brass (D) 10 centimeters long by 15 broad, soldered to the casting E which is movable about the axis e. The function of this part of the appara-tus will be more easily explained by describing the manipulations necessary when collecting water.When intermediabo water is to be obtained the water-bottle is firmly attached to the sounding line, which carries at its end usually a 56113. or 1 cwt. lead; the stopcocks are then opened giving them, with the rod C the position represented in the figure. The line is then lowered carefully by hand until the water-bottle is close to the surface when it is let go and the line allowed to run out without a check. During its passage downwards the water courses freely through it being considerably assisted by the conical end-pieoes K K. When the requisite depth has been reached the line is checked hanledin a few fathoms then let go checked again at the same mark and finally hauled in altogether by the donkey-engine. When the line is hauled in at first, FIG.5. K fi the flap D falls down into a horizontal position when it is caught by the movable piece of brass F which moves round an axis f, and is supported on the side opposite to E by the rod G which rests on the spiral spring H. The wa,ter rushing past D when thus in a horizontal position exercises a sufficient pressure upon the rod to close the stopcocks B B. When the speed with which the bottle is hauled through the water is increased the pressure on D becomes so great, that it overcomes the tension of the spring H and E pawes the catc 458 BUCHANAN’S EXPERIENCES OE’ THE LABORATORY F when the rest of the journey upwards is performed with the flay D hanging down aid therefore offering the least possible resistance to the water.The object of at fir& hauling in only a couple of fathoms or so and letting the line go again is to ensure the cocks being closed. For supposing after the first hauling in they were not quite closed, by letting the instrument descend through the water the flap D sets itself again and on hearing in it shuts down the stopcocks which were before but partially closed ; or if they were closed before it only shuts them the tighter. When the water-bottle has been brought up, it is only necessary to substitute for the lowermost brass funnel it small nozzle when the water may be tapped into any vessel destined to receive it. This done the bottle may be at once lowered to any other required depth much time being spared by not having to detach it each time.At the upper end of the bottle a small spring safety valve L is introduced in order that the considerably denser water from below may be able to make room for itself as the surfaceis approached. In order that the instrument may properly do its work, it is evident that first the stopcocks should be so stiff that tbe weight attached to their levers be not sufEcient to close them and secondly the spring H should be so strong as to ensure the shutting of the cocks before it itself gives way. These conditions are secured by the follow-ing means of adjustment. The stopcocks can be made stiffer in the usual way by tightening the screws which secure the “keys” in the barrels ; the tension of the spring H can be increased or diminished by means of a screw at the lower end of the tube containing it; and the mobility of the stopcocks can be further regulated by means of the screws M M.Although from this description the operation of adjust-ment may appear complicated it is in fact practically very simple. After the first time of use it is rare that any further adjustment is required than a turn of the screws M M. The diameter of the apertures at either end is necessarily smaller than that of the cylinder it is therefore impossible for the water in it to be entirely changed while descending through a distance equal to its own length. It became a question therefore for experiment to decide what actually was the rate of change of water. To this end a few experiments were made in a fresh-water lake. The bottle being filled with water containing some yellow prussiate of potash was sunk in the lake until the surface of the waker was on a level with the upper stopcock when the stopcocks were opened and the line let go.On being brought up again the contents were tested with solution of perchloride of iron. It was found that when the bottle had been sunk to a depth of a fathom and a half the water had been entirely changed, the iron solution being wholly without action on it. We may be cer-tain then that the water which we obtain by this means is a ON BOARD THE “CHALLENGER.” 459 avera.ge of the last two fathoms through mhich the bottle has passed. The weight used as a sinker should be chosen so as to impart suffi-cient velocity not to lose time unnecessarily over the operation and at the same time not to give an excessive velocity at the depth where the water is to be collected because the rate of change of water depends on the friction of the water inside the bottle and so on the velocity of descent.In practice for depths over 100 fathoms a weight of 112 lbs. was used and for depths from 25 up to 100 fathoms a weight of 56 lbs. was used. For less depths the weight of the bottle itself was sufficient. The velocity of descent at the depth where the water is to be collected should not exceed twelve feet per second. The mean velocity of descent for the interval between 75 and 100 fathoms from the surface was with 56 lbs. nine feet and with 112 lbs. eleven and a half feet per second. When once let go it is essential that the line should run out to the required depth without a check ; it is then however immaterial as far as the water-bottle is concerned what interruptions occur in heav-ing in.The fulfilment of the condition of running out without a check never presented any difficulty on board the “ Challenger,” depending as it does on the care of those who tend the line. When however by accident a check does occur the line is stopped and the water-bottle brought up again reset and sent down again. In order to utilise any such accidents it is usual to take the water from the greatest depth first ; then if a check does occur it may occur at one of the desired intermediate depths and so no time would be lost. In designing the water-bottle it had been my intention to use it not only for collecting water but also as a flask so that the atmospheric gases could be boiled out of it without transvasingthe water.In practice however I have not been able to get air-tight stopcocks beside8 which it would make an inconveniently large apparatus in a very small laboratory. I have spoken o€ this water-bottle as being only used for intermediate waters, but there is no reason why it should not be used for bottom water; indeed where the sounding lead does not weigh over 1 cwt. it is fre-quently used for this purpose. In the case of deep soundings how-ever where a weight of three and sometimes four hundred-weight is used the “ slip ” water-bottle is always preferred. When the water-bottle has been brought up full of water the lower funnel is unscrewed and a nozzle put in its place with a long glass tube attached to it.This tube reached to the bottom of the flask and by it the water was run into the flask as carefully as possible without allowing any bubbling or nnnecessary commotion in the water. The necessity for transvasing the water before boiling is a defect in the method. When however it is carried out expeditiously and carefully 460 BUCHANAN'S EXPERIENCES OF THE LABORATORY it does not appear to introduce any serious error into the results. As a proof that this is so I may mention that one sample of water brought up by this instrnmeut from a depth of 2,800 fathoms contained only 3 per cent. of oxygen ; and a sample of rain-water which I collected on one occasion had less than 1 per cent. of oxygen. In transvasing, the flask was filled up to the lip and usually allowed to run over for a little so that the upper layer of water which had been most in contact with the air might be eliminated.The connections with the receiving tubes were then made. The water in the small bulb was kept boiling vigorously for from 10 to 12 minutes before the upper end of the gas tube was closed with a glass rod and sealed up. When this was done, connection was established between the flask and collecting apparatus, the gas in the water immediately beginning to be evolved. The lamp was naw lighted below the water-bath which contained the flask and the temperature of the water thus raised until the water in the bath boiled; this was continued always for an hour usually for an hour and a-half and often longer the exact duration of the boiling depend.ing on the nature of the other work going on. Although disadvantageous as regarded comfort and ventilation the lowness of the roof in the laboratory presented some advantages especially for fixing apparatus. It will be evident that the apparatus with which we are at present dealing possesses a good deal of spring. In order to give it more rigidity the upper extremity of the brass rod supporting the gas tube was supported by strings from the roof. As the amount of oxygen and nitrogen contained in a litre of sea-water may amount to 22 or 23 c.c. it is advisable that the gas-tubes used for collecting it should not have less than 50 C.C. capacity in order that the water may continue to boil vigorously when the flask is immersed in the water-bath.A s long &s this is the case the water in the small bulb has no opportunity of cooling and so choking the tube. The gas-tube itself however should not be unnecessarily heated. The rolling of the ship (otherwise an annoyance) was here not altogether disadvantageous. It produced a species of pumping in the apparatus so that when +he boiling was getting sluggish 8 heavy roll was enough to start it again. The apparatus for the determination of the carbonic mid is described by Professor Jacobsen in the memoir above cited. The alterations which I introduced into it were not of an essential character and were suggested by exigencies of locality. In principle the method con-sisted in distilling to dryness a measured volume of the water in a current of air free from carbonic acid.In the receiver and in the tubes interposed between it and the aspirator a measured quantity of baryta-water of known strength was placed. The baryta remove ON BOARD THE “ CHALLENGER.” 461 from this solution in the course of the experiment was a measure of the carbonic acid eliminated from the water. It is well known that sen-water retains its carbonic acid with great tenacity mere boiling having but little effect on it. Amongst the experiments which occu-pied me during nine months previous to my receiving the appointment of chemist in the expedition with a view to settle on a trustworthy basis the programme for the chemical work during the expedition was a thorough investigation of the bebaviour of sea-water to carbonic acid.These investigations showed that the dissolved sulphates were amongst the most energetic agents in the retention of the carbonic acid. From this I concluded that if the sulphates were removed the carbonic acid would come away more easilF. Consequently before beginning to distil the water for determination of the carbonic acid, about 15 C.C. of a saturated solution of chloride of barium were added to it. The flask having been closed was shaken SO its to mix the solntion. This no doubt had to a certain extent the desired effect of assisting the disengagement of the carbonic acid from the water boiling at ordinary pressure although when half of the water was distilled there was still a notable amount (about one-tenth of the whole) in the residue; but it had no effect in rendering it more amenable to the action of reduced pressure.It had a very useful effect in eliminating “ bumping ” altogether even when the distillation was carried to complete dryness. A liquid residue of about 10 C.C. was usually left in the flask. The baryta remaining in the solution was determined by means of hydrochloric acid rosolic acid being used to determine the point of neutralisation. Of the gas samples which were boiled out in the Bunsen apparatus I was enabled to analyse about a third before my connection with the expedition ceased ; the remainder as far as I know have not yet been analysed. The results are therefore subject to correction from future experiments. The mean value of the oxygen per cent. that is oxygen and nitrogen being put as 100 is 33.67.Jacobsen experimenting in the North Sea found a mean of 33.9 and observed a variation between the limits of about half per cent.; that is to say he observed from 33.7 to 34.2 per cent of oxygen iu the mixture of oxygen and nitrogen and he came to the conclusion that this was a real difference and not due to any errors of observation. My observations which extend over a much greater area give a lower average 33.67 and a very much greater variation in the individual results in round numbers from 33 to 35 per cent. The amount of oxygen obtained in samples in the Antarctic regions is usually very much higher than in the moderately warm regions VOL. XXXIIT. 2 L The results obtained are put shortly in the table (p. 462) 462 11-26 11-71 13.00 13.10 13-86 14.37 BUOHANAN'S EXPERIENCES OF THE LABORATORY "(3.- - - - - -14.6 13.0 6.9 5.1 2.5 1.5 - -Depth. Fathoms D. Surface. 25 50 100 200 300 400 800 over 800 Bottom 73 2 10 5 8 4 10 7 7 54 Carbonic acid. Grams in 1 litre. C. 0 *0426 0 -0337 0 -0488 0 *0436 0 *OM6 0 * o w 0 *04ll 0 -0422 0 9446 0 -0474 33.67 33.36 32-33 30.21 2340 11'37 15.46 22-62 23.45 -- -0.49 0.21 0.45 0.90 1.14 1-03 --4-24 3-59 1-67 2'41 4.06 4-40 -- --11 -75 11 *92 13 -45 14 '00 15 -00 15 -40 -about the trade winds; and also in all the warm regions. In the Antarctic regions I found as much as 35 per cent. and as low as 32.9 per cent.in the trade wind regions. The number of observations is, however as yet too small to admit of general conclusions being drawn from them. I have not taken a mean of the bottom ones because the depth and other conditions were too various to make it of value. Looking at the oxygen percentages dissolved in water at different depths we see that they diminish with considerable rapidity as the depth increases until a minimum is reached at about 300 fathoms, after which it slowly increases. The percentage in very deep water appears to depend a great deal on the geographical position being the greater the nearer i t is to a position where the surface-water at some period of the year has the same temperature. This appears certainly to hold in the southern hemisphere.In the northern my observations on bottom water are from the Pacific where there is withouf doubt a considerable admix-ture of southern cold water which is therefore very remote from its source and which therefore mixed with the cold northern water gives a result below what we find in similar southern latitudes. I believe the prevalent opinion attributing the low temperature in the deep waters of the North Pacific entireZy to st southern source to be quite erroneous. In the northern parts of the Pacific the winter tempera-The results obtained at other depths are given in the table ON BOARD THE “CHALLENGER.” 463 ture is quite low enough and the differences of temperature at the same place at different seasons and in different places at the same season are quite sufficient to cool down a large area of the northern Pacific and when the remainder (the larger portion) of the air samples which I boiled out have been analysed much light will no doubt be thrown on the question.Whatever view we may adopt as to the cause of ocean circulation, there can be no doubt that the actual effect is to cause on the whole a rise of cold water towards the surface near the equator. This water has been impoverished since it was last exposed to the atmosphere and would probably not contain more than 15 or 20 per cent. of oxygen in its absorbed atmospheric air. This percentage being so remarkably reduced at 300 fathoms shows the existence of a powerful reducing agent at that depth and as we know from Mr. Mu r ray’s observations that life generally is remarkably abundant at these depths it is a fair inference at least provisionally to assume that the disappearance of oxygen is due to the abundance of life.It must be remembered how-ever that our observations were made in tropical equatorial waters, where surface aihation is hindered by the constant high temperature of the atmosphere. In colder regions we have every reason to believe that the aeration is much more energetic. Unfortunately in dealing with sea-water the amount of carbonic acid present affords no means of controlling the amount of oxygen which is deficient. In distilled water carbonic acid is absorbed according to well-known laws but this is not the case with sea-water for which the laws have yet to be investigated.The amount of carbonic acid per litre in water at &he surface, according to an average of a great number of observations was 42.6 milligrams. The highest average amount 48.8 was at 50 fathoms, falling to 41.1 at 400 fathoms and rising to 47.4 at the bottom. It will be seen from the table that the absolute amount of cdrbonic acid dissolved is very great and 6ha.t the variations with depth are insignificant. Observations at the surface showed that on the whole, less carbonic acid was present in warm than in cold water but no law was apparent. We know from other investigations that the saline contents exercise a very strong influence on the amount absorbed and there are other causes which introduce seeming irregularity into obser-vations of this kind.There is a remarkable difference in the results which I have obtained in the open ocean and those obtained by Dr. Jacobsen in the North Sea. I have found approximately 45 milligrams per litre as a mean amount. J a c o bsen found 88.6 milligrams or about exactly double the quantity. It is difficult to imagine to what cause so great a dis-crepancy is to be attributed if it does not depend on the nature of the 2 L 464 BUCHANAN'S EXPERIENCES OF THE LABORATORY oxygen per mt. 0 + N = 100. water. The water in which nearly all my observations were made was the deep clear ultramarine-blue water of the ocean. The North Sea water in which J a c o b s e n experimented is comparatively opaque and green. In the A.ntarctic Ocean where such water was met with occa-sionally though very sparingly the carbonic acid was always present in marked excess.The green colour of shoal water is generally attri-buted to the influence of solid matter which may also tend to retain carbonic acid as we know is the case with dissolved saline matter. The subject is worthy of investigation ; and as sea-water collected well out at sea coutains as far as I have been able to observe no appreciable amount of putrescible matter and as the carbonic acid dissolved has no tendency to escape at ordinary temperatures samples if carefully collected and preserved in clean stoppered bottles would be in a per-fectly fik state for examination after a long Toyage. In the Table at page 462 we have seen the distribution of gaseous contents at different depths.In the Table below we have the mean results from different zones of boftom water. It will be seen that the highest percentage of oxygen as well as of nitrogen and carbonic acid, is met with in Antarctic regions and that the amount of all these bodies present in the water diminish as we approach equatorial regions. Bottom Water. ca:$t$d Gmms. Latitude. 66's. to 35OS. 35O s. to lo" s. 10" S. t,o 10' N. 10" N to 35' N. 35' N. to 38' N. --0 -0518 0 -0466 0 *04.09 0 *0459 0 *0426 Nitrogen . C.C. in 1 litre. 13 "73 ---11 -71 12 -91 The ahsolut,e maximum of carbonic acid in bottom water was ob-served at the highest swthern latitude where observations were made, namely 0.0829 gram per litre in lat. 65" 42b S. long. 79" 49' E.and 1,675 fathoms. The minimum amount in this zone was 0.0385 gram per litre in bt. 48" 37 S. long. 55" 17' W. and 1,035 fathoms. Be-tween 35" and 10" S. latitude I have analysed only one air-sample from bottom water. It was collected 1st. 12" 45' S. and long. 147" 1' E., in a depth of 1,700 fathoms and contained 17.4 C.C. of oxygen and nitrogen in one litre 20.97 per cent. of which was oxygen and 0.0609 gram CO in a litre. In this part of the ocean which has been called the Melanesian Sea, I observed some anomalous results. Amongst them may be reckoned the above very high percentage of carbonic acid. The a.mounf varie ON BOARD THE (( CHALLENGER." 465 greatly at different stations. In lat. 14" 7' S. long. 153" 43' E. and 2,450 fathoms it was 0.0434 gram and in 1,MO fathoms close to Raine Island it was 0.0361.Both the maximum and the minimum amounts were found over ' I globigerina ooze," while the medium (0.0434) was found over "red clay." At the surface too in lat. 17' 25' S. long. 169" 5' E. there was found in one litre 0.096 gram carbonic acid 14 C.C. of oxygen and nitrogen of which 32.35 per cent. was oxygen. The temperature of the water was 25.8" C. and not-withstanding it contained the greatest amount of carbonic acid observed in any water during the cruise and also tlhe least oxygen percentage of surface water. At other stations in the same sea 0.0317 and 0.0357 gram carbonic acid were observed. Only one air-sample however has as yet been analysed. Again in the same zone but to the north of New Guinea a very large amount of carbonic acid was met with in the bottom water, though the surface water was not abnormal in this respect.In lat. 4' 19' N. long. 130" 15' E. and 2,500 fathoms 0.0524 gram carbonic acid was found and in lat. 5" 31' N. long. 145" 1 3 E. and 2,325 fathoms 0.0619 gram carbonic acid was found in one litre. In the Celebes Sea 0,0524 gram carbonic acid and 21-78 per cent. oxygen were found. In the Sulu Sea the carbonic acid is 0.0243 gram and the oxygen percentage 29.63. The deep water in this sea has a tem-perature not below 10.2" C. so that it is in many respects peculiar; under these circumstances however it is remarkable that there should be so little carbonic acid and so much oxygen. We see then that in this zone there is very considerable fluctuation in the gaseous contents of the sea-water especially at the bottom.In the zone from the parallel of 10" to that of 35" N. latitude there is an analysis of only one air-sample from the bottom. It is from the deepest sounding from which bottom has been obtained namely 4,475 fathoms in lat. 11" 24' N. and long. 143" 16' E. Here again we have a very small quantity of carbonic acid (0.0312 gram) and of nitrogen and oxygen (13.9 c.c.) with a high oxygen percentage (29.11). As the depth reached on this occasion was quite unexpected the sounding was repeated with substantially the same result as regards depth but the water-bottle came up the second time unclosed so that the attempt to verify or control the somewhat abnormal results obtained from water from such exceptional depth was frustrated in the most annoying way possible.The water-bottle was that commonly used for bottom water which is constructed to close on touching bottom. The '' tumbler," which by falling forward disengages the cylinder which encloses the water between suitable valves had in this case fallen backwards thus as it were choking itself. In this zone the maximum carbonic acid was 0.057 grams in lat. 19" 4' N. long. 65" 7' W. and 466 BUCHANAN'S EXPERIENCES OF THE LABORlTORT Latitude . . . . . . . . . . Longitude . . . . . . . . . . . Nitrogen C.C. per litre . . Oxygen C.C. per litre . . . . 32" 36' 8. 137' 43' W. 22 .G5 13.72 3 -88 Oxygen percentage . . . . . 3,875 fathoms being the deepest water found in the Atlantic. The minimum carbonic acid was that above referred to as having been found in the Pacific in the deepest water attained during the cruise, with the exception of this last.All the observations in this zone are from the North Atlantic where the conditions are in many respects exceptional. The salt-ness is much above the average and it i R there-fore not astonishing that the carbonic acid should also be compara-tively high. To the north of 35" N. latitude the observations are principally in the Pacific the mean amount of carbonic acid observed being 0.0426 gram per litre. The two highest results (0,064 and 0.0536 gram) were observed in the North Atlantic. If these are eliminated the mean amount in this zone of the Pacific is 0.0361 gram. The absolute amount of oxygen and nitrogen is higher in this zone than in the equatorial one.The amount of nitrogen is 12.91 C.C. per litre. In the eame latitude in the South At'lantic two observations give 1401 and 14-76 C.C. per litre. If the bottom water in this zone of north latitude were made up of equal parts of water prepared in the southern and in the northern hemispheres the mixture would probably contain nitro-gen in the quantity above quoted. Amongst the air samples collected from intermediate depths those from 800 fathoms present a remarkable uniformity in the oxygen per-centage. The results are given in the following table :-37" 47' S. I 35" 45' S. 3" 10' N. 30" 20' W. 18" 31' W. 14" 51' W. 22 -95 23 *25 22 -22 14 -10 15 *81 11 -82 4 -20 4.79 I 3.38 The first three samples are from southern latitudes between the parallels of 32" and 38" and the fourth is from a position close to the equator in the Atlantic.It will be seen that the oxygen percentages are much more uniform than the absolute amounts of that gas per litre of water. Indeed if we compare the first and the last columns we see that the equatorial Atlantic water possesses absolutely much less oxygen than the North Pacific water though its percentage is higher. The amount of nitrogen present is also very variable. In the South Atlantic we have as much as 15.81 C.C. per litre while at the equator in the same ocean there is only 11.82 C.C. The temperature of the water in all the cases was practically the same being from 2.3" to 3" C., and the specific gravity was similar in the two extreme cases identical.This disappearance of nitrogen is very remarkable. Hum b o 1 d t and Provengal (M&n. d'Arc. ii 377) in their investigations on the respi ON BOARD THE CHALLENOER." 467 ration of fishes noticed a similar phenomenon nitrogen always disap-pearing side by side with the oxygen though in no constant ratio to it. They experimented on tenches letting them remain in the water until they were uneasy then boiling out the water. They found that a fench will exhaust the oxygen in a cubic meter of river-water in 21 months. They observed the oxygen percentage (N + 0 = 100) re-duced as low as 1.3 and observed that want of oxygen affected the fish more than tbe presence of excess of carbonic acid. The lowest oxygen percentage observed was 3.84 in water fetched 75 fathoms above the bottom in lat.30' 22' N. long. 154" 56' W. the depth of the sea being 2,950 fathoms. The absolute amount of oxygen was 0.6 c.c. and of nitrogen 15 c.c. so that there was no deficiency of nitrogen. While becalmed off the north coast of New Guinea I boiledout the air from rain-water collected as it ran off the awning. The tempera-ture of the rain so collected was nsually from 23" to 24" C. and in order not to interfere with the work of the day it was kept in perfectly full bottles over night and boiled the next morning. Three rain-waters were so treated and the oxygen percentages found were 18.36, 25.42 and 0.62 and in the last two were found 11-27 and 11-86 C.C. nitrcgen per litre. There was therefore rather more nitrogen than would have been expected from the temperature.The amount of oxygen present however varied in the most extraordinary manner, the last sample being almost completely deprived of oxygen. If the carbonic acid be reckoned as oxygen we obtain the following per-centages in the same order 28-45 30.83 and 25.68. The rain-water when freshly collected remained perfectly clear on addition of baryta-water ; the carbonic acid had therefore been entirely formed during the night at the expense of the oxygen dissolved in the water. There must therefore have been also a considerable amount of easily oxidisable carbon in the water and in fact I collected a considerable quantity to be used for the preparation of standard solutions but could make no use of it as it speedily produced a fungus-like deposit.Some also which I tested for ammonia continued to give it off on distillation almost to dryness ; and the water remaining in the retort reduced per-manganate energetically. At the time I did not know of the deficiency of oxygen and I attributed the organic matter to contamination from the awning although the explanation never fully satisfied me, because the amount of rain which fell was sufficient to tho-roughly wash the awning and as it did not answer the purpose for which I had particularly collected it I reverted to the ship's distilled water which was always of excellent quality. It will be seen that the carbonic acid does not completely cover the loss of oxygen; no doubt however some water wm formed and it is no 468 BUCHANAN'~ EXPERIENCES OF THE LABORATORY ETC.impossible that sulphurous acid from the furnace gases escaping by the funnel may have had some reducing effect. The determinations which were made of the absolute amount of these permanent gases as we are accustomed to call them in the litre, have been plotted in Fig. 6. The curve represents Bunsen's curve of the absorption of nitrogen in distilled water. The ordinates represent cubic centimeters of nitrogen absorbed by distilled water of different temperatures when exposed to the atmospheric mixture of nitrogen and oxygen. The results obtained by boiling the gases out of the surface water have been plotted on the same figure and also those obtained by Jacobsen in his experiments in the North Sea in order to see how they would fall in with the curve already found for distilled water. I have found it was impossible either to get my own results or Jacobsen's to agree with this curve. With hardly an exception all the observations fall below the curve. About 10" to 12" C. they nearly coincide with it but both at lower and higher temperatures and espe-cially at higher temperatures-and my own experiments were made to a great extent in water of higher temperatures-fall very considerably below it. The shading covers the whole of the experiments so that it shows at a glance how the experiments agree with the curve. We see at once that above 20" C. they depart very markedly from the law indicated by the line A B. The strong convexity of Bunsen's curve does not appear to be probable knowing as we do how rapidly all but the very last traces of air are eliminated from water by heat. The curve as indicated by our experiments on sea-water would very much more nearly approach a straight line ; it would however have a slight convexityin the same sense as A B. On this point the analysis of the remaining air-samples which were collected will throw much light, especially as a large number of them are from surface water in the warmest regions of the earth a I-W s 15 W a !l -I 14 w z a Z -13 z w U 0 U l-z -12 cn tl [I I-d z I I I I- z W 0 -10 0 I m 3 0 9 A - Q I O0 I O Q DECREES 20 CENTIGRADE 3

ISSN:0368-1645    

DOI:10.1039/CT8783300445

出版商:RSC    

年代:1878

数据来源: RSC

摘要:

470 LVI. The Action of Alkaline Hypobronzite on Ammonium Salts, Urea, and OzawLde. Part I. By W I L L I A M F o s T E R, B .A., Lecturer on Chemistry, Middleses Hospital Medical College. ABOUT two years ago I was called upon to determine the amount of urea in solutions containing other nitrogenous bodies. The hypobro- mite method being inapplicable in consequence of the solution of nitrogen from nitrogenous bodies other than urea which were present, I was led to study the action of alkaline hypobromite on several well- defined compounds containing nitrogen. Some of the results obtained I now beg to lay before the Society. Action of Alkaline Hypobromite on Ammonium Sults. When these bodies in aqueous solution are subjected to the action of the hypobromite, the mixture effervesces briskly, evolving gaseous nitrogen.Knop and Wolf (Chem. Centralblatt, 1860, 257) have studied this reaction. In their experiments they obtained the whole of the nitrogen in the gaseous form. They recommend the process, and describe an apparatus for the quantitative estimation of ammo- nium salts. Abnormal results were obtained with ammonium 0x8- late. Dietrich (Zeit. Anal. Chem., v, 40) states that there is a constant loss of 2.50 per cent. of nitrogen in the reaction, and he considers that this proportion of the nitrogen remains dissolved in the solution. Wagner (ibid., xiii, 383) confirms Dietrich as to the percentage of loss of nitrogen. Knop (ibid., ix, 226) states that there is no loss if the solution is not very weak and the nitrogen is not allowed to escape from contact, with the hypobromite until the reaction is completed ; if this precaution be omitted, the nitrogen will contain undecomposed ammonia, and the results will be too low.Prom the ready way in which the nitrogen can be evolved, I was led to hope that Knop’s process would be a good and expeditious one. With excessively strong soda-solution freshly charged with bromine. the amount of gas obtained is equal to the theoretical amount of nitrogen in the quantity of ammonium salt operated on. The mean of two experiments on 5 C.C. of a 2 per cent. solution of ammonium sul- phate gave 99.7 per cent. of the total nitrogen. Such viscid solutions are objectionable on account of their tendency to envelop air on agita- tion which is liberated on dilution during au experiment. The mean of four experiments with the same ammonium sulphateFOSTER ON THE ACTION OF ALKALINE HTPOBROJIITE.471 solution and a slightly weaker alkaline hypobromite solution, gave 95.6 per cent. of the total nitrogen. With an alkaline hypobromite solution of the strength recommended by Knop, the mean of four experiments gave 93.3 per cent. of the total nitrogen of the ammonium salt. With hypobromite of Knop’s strength and an apparatus for collect- ing the gas over the hypobromite without dilution, the mean of three experiments gave 92.8 per cent. of the total nitrogen. In these cases no increase in the volume of the gas was obtained by allowing it to remain in contact with excess of the alkaline hypobromite. A large number of experiments have been made in this direction which lead me to conclude thtit, with Knop’s strength of alkaline hypobromite and the ordinary forms of apparatus, there is a deficiency of about 7 per cent.of the total nitrogen of the ammonium salt. When the solution of Knop’s strength is diluted with an equal volume, or even two volumes, of water, I find that the amount of nitrogen suppressed is practically the same as when the undiluted solu- tion is employed. The Action of Alkaline Hypobromite o r b Urea. This action has been studied by Knop, Russell and West, DuprB, and several other chemists. When the mean solution is treated with the hypobromite, nitrogen is briskly evolved. When excessively strong soda-solution is employed, nearly the whole of the nitrogen is readily obtained as gas.The mean of two experiments gave 98.0 per cent. of the total nitrogen of the urea, no heat being applied to the vessel, and the gas being measured at the end of I5 minutes. The objection to the use of such viscid solutions has already been noticed. When the soda-solution is of the strength recommended by Knop (10 grams in 25 C.C. of liquid), there is a deficiency of nitrogen.. In their experiments on this subject, Drs. R u s s e l 1 and We s t employed Knop’s strength of solution, and found, from a large number of ex- periments, that there was a deficiency of 8 per cent. of the total nitrogen of the urea. Their results have been confirmed by several other chemists. The mean of three experiments with soda-solution of about the normal strength gave me 92.3 per cent.of the total nitrogen at the end of 15 minutes without the application of heat. A large number of experiments have been made with Farious strengths of hypobromite solution, which show that with the stronger solutions there is a greater yield of nitrogen than with the weaker ones,4-72 FOSTER ON THE ACTION OF ALKALINE HYPOBROMITE and that with normalsolutions there is a deficiency of about 8 per cent. of the total nitrogen. Action of Alkaline Hypobrontite o n Oxumide. In some unpublished experiments of Dr. R u s s e 1 l's, this reaction has been briefly studied. When oxamide is added to the alkaline hypobromite, decomposition at once commences, is attended with ft brisk evolution of nitrogen gas. In consequence of the light and in- soluble character of the oxamide, it is advisable to employ a vessel which admits of being easily agitated.If the agitation during decom- position be neglected, some time is required before the last portions of oxamide disappear. The strength of the alkaline hypobromite is a matter of some importance. When the hypobromite is of the normal strength, the decomposition progresses slowly, even on the application of heat, and at the end of some minutes there are generally small quantities of white powder still remaining, which is no doubt unaltered oxamide. When the freshly prepared normal alkaline hypobromite is diluted with about an equal volume of water and then subjected to the action of the oxarnide, the mixture effervesces readily on agitation, giving off gaseous nitrogen, whilst the temperature of the vessel rises in a very marked degree.This elevation of temperature facilitates the decomposition of the oxamide, and generally yesults in a perfectly bright solution at the end of four or five minutes. I n some cases,it is advisable to apply heat to the containing vessel in order to complete the reaction. I n the presence of excess of the hypobromite no odour of ammonia can be detected. The experiments for determining the amount of nitrogen evolved as gas, were made in the form of urea apparatus, such as has been suggested by Dupr6 and Apjohn. This consists of n small test-tube attached to the lower side of an indiarubber plug which accurately fits the neck of a small flask. Through the plug passes a piece of glass tubing attached t o a graduated receiver for collecting and measuring the gas evolved in an experiment.The oxamide having been weighed, is placed in the test-tube, and a small quantity of water added. The freshly prepared hypobrornite solution is then placed in the flask and the plug with attached tube containing the oxamide is then inserted, and the whole apparatus left for a few minutes to assume the temperature of the apartment, the level of the water in the measuring-tube being afterwards carefully adjusted. All being now ready, the flask is gently tilted and shaken alternately, 80 as to cause the admixture of the oxamide and hypobromite, the receiver at the same time being gradually raised so as to keep the level of the water inside and outside the measuring tube as nearly equal as possible.ON AMMONIUM SALTS, UREA, AND OXAMIDE. 473 When the decomposition is complete, which is known by the solution being perfectly bright, and still having a distinctly yellow colour, the whole apparatus is cooled down to the original temperature, and the increased volume of gas in the tube is determined.This number is then corrected in the ordinary way for temperature and pressure. The following four experiments are illustrations of the experimental numbers which have been obtained :- Oxamide taken in grams. C.C. of gas e-rolred uncorrected. Percentage of nitrogen obtained from oxamide as gas. Percentage of total nitrogen of oxamide evolved as gas. ~~~ -245 ‘222 ‘247 ‘201 51 -00 44 -75 52 *oo 40 *25 24 *3 23 05 24 -0 23 5 76 -0 74 -0 75 *5 74 *o The mean of the four experiments gives 74.87 as the percentage of the total nitrogen which can be obtained in the gaseous form in this way. After satisfying myself of the perfectly regular yield of nitrogen under these conditions, attempts were made to ascertain the state of the suppressed nitrogen, amounting to 25 per cent.of the total quantity present in the original oxamide. The perfectly bright solu- tion at the close of an experiment on the amount of gas evolved was then transferred to a large clean flask containing a considerable quan- tity of the “ copper-couple,” and distilled. Ammonia appeared in the condensed liquid very slowly, so that on distilling for three or four hours and collecting about 300 C.C. of distillate, not more than one quarter of the suppressed nitrogen could be accounted for.On allow- ing the flask and its contents to stand over night and commencing the distillation in the morning, the first portions coming over were rela- tively richer in ammonia than those of the previous day. Ammonia continued to come over after distilling for three or four hours on the second day of the experiment, and owing to the fracture of the flask towards the close of each of several experiments, I hare not yet pushed the distillation far enough to yield me a distillate free from ammonia. However, I have recovered more than two-thirds of the suppressed nitrogen as ammonia, and I have no doubt that the whole of it can be recovered in this way. Operating on -223 gram of oxamide, I have obtained -0124 of nitrogen as ammonia, the amount to be accounted for being -0177.The slow reduction of the nitrogen compound by the copper-couple in a boiling alkaline solution led to the belief that it existed as a Oxamide contains 31.8 per cent. of nitrogen.474 FOSTER ON THE ACTION OF ALKALINE HYPOBROMITE, ETC. nitrate. Some of the sodium hypobromite was then treated with an insufficiency of oxamide, evaporated to dryness, fused and kept in that condition for some minutes. The fused mass was then dissolved in water and starch-solution, iodide of potassium, and tartaric acid added in excess. Under these circumstances an intense blue colonr was always produced. The soda employed in these investigations was purchased as that made from metallic sodium, but experiments made to test its purity show that it contained traces of something which yields excessively minute quantities of ammonia when boiled with the copper-couple.I n e n converted into hypobromite, fused, and afterwards treated with starch-solution as above, the results have been generally negative, but not uniformly so. I am therefore at present unable to state the pre- cise condition of the suppressed nitrogen. The experimental evidence is in favour of the notion that it is converted into nitrate, whilst, from general considerations of the decomposition-products of osamide, one would rather expect to find it in the condition of cyanate. This latter view finds further support in the experiments recently made by Mr. F e n t on, who has found that a portion of the nitrogen of urea is con- verted into cyanate, under somewhat similar circumstances. I am not inclined to regard the evolution of three-fourths of the nitrogen of the oxamide as accidental, the result of a particular strength of alkaline hypobromite only. The results obtained would rather favour the view that all the nitrogen in the oxamide molecule is not similarly situated, and in depicting the behaviour of oxamide in this reaction, I would suggest the formula- Cp NHHO. L o ) , It is my intention to have some pure sodium hydrate prepared, and then settle the most interesting part of this inquiry-the condition of the suppressed nitrogen in the experiments described. I also propose investigating the action of hypobromite on acetamide mid other bodies.

ISSN:0368-1645    

DOI:10.1039/CT8783300470

出版商:RSC    

年代:1878

数据来源: RSC

摘要:

475 LVI1.-On the Part played by Carbon in reducing the Szclphates of the Alkalis. By JAMES MACTEAR, of Sh. Rollox Works, Glasgow. THE action when an alkaline sulphate, such as the sulphate of soda, is reduced by carbon, although investigated by such men as Liebig, Gay-Lussac, Berthier, Berthollet, and Vauquelin, has been a much misunderstood subject. The general idea that the decomposition was a simple change of elements resulting in the formation of sodium sulphide (Na2S), though seductive, has been shown by experiments made from time to time to be insufficient. In a series of documents which have been entrusted to me by Mr. John Tennant, of St. Rollox, the late Professor Liebig describes at length a process which he had invented for the production of carbonate of soda. His letter announcing his discovery is dated 11th July, 1839, and his process was founded, as he himself says, on a discovery of Gay-Lussac’s, that “caustic potash heated with wood sawdust, gave off p u r e hydrogen, and produced oxalate and acetate of potash.These two salts, if heated to redness, are converted into carbonates, the formation of oxalic and acetic acid taking place from the decomposition of water, the oxygen of which combines with the carbon while the hydrogen goes off ”:- “ Proceeding from this, and considering the great alkalinity of sulphuret of sodium, I thought it ought to have the same action upon wood fibre, that is, by heating a concentrated solution of this sulphuret with sawdust, there ought to be, at a certain temperature, the formation of oxalate and acetate of soda, and the evolution of sul- phuretted hydrogen.“This is exactly what takes place at 200°C. Heated to 300°C. these two salts are converted into carbonates.” Liebig says that he “had submitted this process to the judgment of the greatest chemists, and Gay-Lussac thinks it the most ra- tional, and at the same time most simple process for producing car- bonate of soda, and for using the sulphur.” And again, in the same communication, he says, “ In the opinion of Gay-Lussac and B erz elius this process will entirely supersede the present method.” The late Professor Graham had the process submitted to him, and reported to Liebig, “Your new experiment I have carefully repeated, and I do hope great things from it.” Gay-Lussac tried the process at Channy on a very extended scale.476 MACTEAR ON THE PART PLAYED BY CARBON These experiments, and also a series conducted in St.Rollox Works, proved the process to be a failure as a method of manufacturing carbonate of soda, and it was abandoned; but Liebig, in reasoning on the causes of failure, seems to have been led to believe that it was owing to the decomposition of sulphate of soda by carbon being much more complex than had been thought, and he formulated the action as being- 2Na2S04 + 4C = NazC03 + Na2Sz + 2C02 + CO. (This formula has been altered into new notation for more con- venient comparison.) Unger, however, further examined the subject at Liebig’s request, and Reems to have decided that the decomposition of sulphate of soda by carbon resulted in- Na$ + CO.Thus:-NazSO, + 4C = NazS + 4CO. But he again, in tphe second part of his memoir, examines the subject, and modifies his views so as to assume that it was- NazSO, + 2c = Na,S -!- 2c02. My own investigations made at various t.imes had led me to be dis- satisfied with the explanation of what at first sight seems a very simple decomposition, and with a view of elucidating the question, I have made a series of experimenbs, the results of which I have now the honour of laying before the Chemical Society. The method of conducting the experiments was as follows :- The sulphate of soda thoroughly dried, and (a Iamp-black which on examination proved to be remarkably pure) carbon thoroughly ignited and cooled out of contact with air under a desiccator, were mixed as rapidly but thoroughly as possible, and placed in a platinum crucible covered with a tight-fitting lid; this was then heated by gas in a jacketed support : after heating it was cooled nnder a desiccator and weighed.The crucible, with its lid and contents, was then placed in a flask fitted with a stopcock funnel and evolution tube, connected with a series of absorption vessels :- A, evolution flask: B, trap flask containing a small quantity of dilute HC1; C, C, flasks containing a solution of A%03 in HCl ; D, D,IN REDUCING THE SULPHATES OF THE ALKALIS. 477 are absorbing tubes containing the same solution of As203 in HC1; E, E, E, E, are absorbing tubes containing solution of barium and sodium hydrates ; F is an exit-tube leading to the Bunsen pump.Tbe crucible and contents having been placed in the evolution flask A, water was added, and the apparatus connected together. The reduced sulphate was then decomposed with dilute hydrochloric acid, the sulphuretted hydrogen given off was absorbed in the arsenic solution in C, C, and D, D, and carbonic acid by the barium and sodium hydrates in E, E, E, E. The decomposition was completed by boiling the flask A, and finally the flasks B and C, C, are also boiled fo expel the carbonic acid that may have been retained in them. The steam from C, C, raises the tubes D, D, to the boiling point, and all the carbonic acid was thereby expelled and collected in E, E, E, E. The last tube of the series was not found at any time to contain any trace of carbonic acid, thus proving the complete absorp- tion by the previous tubes of the series E.The contents of the tubes E, E, E, E, were emptied into a beaker, and enough hydrate of baryta added to precipitate all the carbonic acid as carbonate of baryts; this is then filtered off, the filter and its contents placed in a carbonic acid apparatus, and the carbonic acid determined by loss of weight. The contents of C, C, and D, D, were emptied into a beaker, the filtrate from the carbonate of baryta, from E, E, E, E, if it was found to contain any sulphide, was added, and the whole filtered through a, tared filter, washed, dried, and weighed. The contents of flask A were filtered through a tared filter, and the filtrate made up to 500 c.c., of which 100 C.C. were evaporated to dryness for total soda, and 200 C.C.were precipitated with barium chloride for sulphate undecomposed. The platinum crucible, after being thoroughly cleansed, was dried, weighed, ignited, and again weighed, the loss in weight giving sulphur in combination with platinum. The insoluble matter on the filter, washed and dried, is treated with bisulphide of carbon, dried, and weighed; it was then oxidised with nitric acid, filtered, and any metallic platinum estimated. The snlphuric acid thus formed is determined by precipitation with barium chloride in the usual way; the sulphur thus found is calcu- lated into PtS, ; and this amount subtracted from the total weight of residue on the tared filter, gives by difference the amount of uncon- sumed carbon. The bisulphide of carbon solution was then evaporated to dryness, and the sulphur weighed.The first five experiments were made merely as a test as to whether carbonate of soda, was formed in the decomposition :- VOL. XXXIII. 2 BI478 MACTEAR ON THE PART PLAYED BY CARBON - 53.531 20.61 0.084 -- 20.694 - 32.837 A. A mixtureof- Sulphate of soda .... 7.1 grams. Carbon .............. 1.2 ,, Equal to 2N%s04 + 4C, submitkd to a bright red heat for 45 minutes, yielded for 100 Na&lOo employed- 19.3 per cent. as undecomposed Na2S04. 11.91 ,, ,, converted into Na,C03. 14.75 per cent. of the NazSO4 decomposed being converted into B. A mixture of- Na2C03. Sulphate of soda, .... '7.1 grams. Carbon .............. 1.2 ,, submitted to a bright red heat for 60 minutes, yielded for 100 Na&04 employed- 10.8 per cent.as converted into Na&?03. Equal to SNaaSO4 + 4C, C. The crucible was lined with carbon and a, mixture of- Sulphate of soda.. .. 10.65 grams. Equal to 2NnzS04 + 4c, Carbon ............ 1.8 ,, was placed in the cavity, and covered with a layer of carbon, the excess carbon amounting to- 3.901 grams. The crucible was exposed for 45 minutes to a red heat, and then to Yielded for 100 NkS04 employed- a high red heat for 15 minutes more. 050 per cent. as undecomposed N%S04. 1.00 ,, ,, converted into Na&03. 1.05 per cent. of NaS04 decomposed is converted into Na&03. We can here see the effect of the higher temperature and excess of carbon, the amount of Na$O, being much less than in the cases A and B. 22 -535 - -- 0.112 - 0'112 22 -423 - -- 45'07 32.395 0-225 0'162 - - I - - 0.335 j 0'321 ------ 0.560 : 0'483 - ! 31'912 44-51 i - i N d 0 4 employed ..............with ........................ Obtained as undecomposed carbon > f 9 NazS04 ,; carbonate N+CO~. .. Total obtained ................ As undetermined .............. Loss ......................... 100*00 - - 0 5 0 1-00 1-50 98-50 - C. I 5. I 0. I Na. 1 Na.$04. I- I-i-I- IUJ REDUCING THE SULPHATES OF THE ALKALIS. 479 The carbon and oxygen have in this case been given off in such proportions that it is evident the carbon must have been evolved nearly altogether in the form of carbonic oxide. 32.837 carbon require for conversion into CO 43.782 oxygen ; whilst 44.51 oxygen has actually been lost. Thus 0738 oxygen apparently has been evolved as C02, and of 100 parts carbon employed- 98.34 per cent.hm gone to form CO, and 1-66 ,, 7 9 99 coz. D. Amixtureof- Sulphate of soda ...... 10.65 grams Carbon .............. 1.8 ,, Equal to 2N%S04 + 4C, submitted to a low red heat over a Bunsen lamp for 30 minutes, then to white heat over a blowpipe for 20 minutes more7 yielded for 100 NazSO, employed- 20.377 per cent. as undecomposed Na&04 1.340 ,, ,, converted into Nkc03. 1.68 per cent. of the Na$304 decomposed being converted into N%CO3. E. A mixture of- Sulphate of soda ...... 10.65 grams Carbon .............. 1.8 .. Equal to 2Na2S04 + 4C, submitted to a low red heat over a Bunsen lamp for 60 minutes, yielded for 100 N%SO4 employed- 19.1 per ceut. as converted into Na2C03. A number of subsidiary experiments were made to ascertain the temperature at which the decomposition took place, and it appears from these that below a dull red heat the decompovition of sulphate of soda by carbon does not take place to any appreciable extent, but that at a red heat it takes place readily- F.A mixture of- Sulphate of soda ...... 10.65 grams Carbon .............. 1.8 .. Equal to 2NazSOa + 4C scbmitted to a dull red heat for 60 minutes over a Bunsen lamp. Ap- proximate temperature 1270" F. Note. -The temperature in this and succeeding experiments was taken by heating the copper ball of a pyrometer in the same position as the crucible, and taking the temperature thus determined as ap- proximately that of the crucible and contents.480 MACTEAR ON THE PART PLAYED BY CARBOK Yielded for 100 Nan,SO, employed- 21.73 per cent.as nndecomposed Na$04 23.20 ,, converted into Na2C03. 29.64 per cent. of the NhS04 decomposed being converted into NhCO,. I Employed N%S04 .... ......... With ........................ Obtaiued as undecomposed carbon y y carbonate N+C03. ... ,, sulphide NH&I ...... ¶ ¶ Y > NazS04 .. sulphur, free S ...... ? Y ,, with Pt = S Total obtained.. ............... Undetermined ................. Loss.. ........................ 1 -961 14 *94Q - S. 22 -535 - - 4 -897 11 -412 3 -953 2 '320 22 *582 - - - - 0. 17 '637 27 *433 - In this experiment the Na not accounted for as undecomposed Na&04, or as N%S and Na&03, amounted to Na 1.44, and this calcu- lated into N*S2 = 2.004 S = 3.444 Na,S2; the total sulphur, free and combined with Pt, amounted to 6.273.Deducting 2.004, we have 4.269 calculated into N&S2 = 10.406 NhS = 14.675 Na-32. This added to 3 - 4 4 Na& = 18.119 N%S2, and as the NhS as found = H2S evolved = 27.812 - 10.406 = 17.406 N&S existing as such, there has been decomposed 100 - 21.73 = 78.27 per cent. of the NazS04 employed, with a loss of oxygen equal to 27.433. As it is shown by experiment C that the excess carbon acts by reducing CO, to CO, we must, in calculating the formulae of the actual decomposition of the NhS04, take the carbon in the same ratio as the Na2SOd, inthis case 78.27 per cent. 78.27 per cent. of 16.901 carbon employed = 13.28, corresponding to the equation 2Na2S0, + 4C. Deducting from this the C equiva- lent of the NhCO, = 19.61, there is left 11.267 as evolved in combi- nation with 27.433 oxygen.Calculating the carbon in the first instance into CO- 11.267 C + 15.022 O = 26,289 CO. Then- This into CO, = and- 27.433 0 - 15.022 0 = 12.411 0. 12.411 0 -t 21.719 CO = 34.130 C02, 26.289 CO - 21.719 CO = 4-37 GO.IN REDUCING THE SULPHATES OF THE ALKALIS. These figures when reduced correspond closely with the formula- 10Na2S04 + 20C = 3Na2C03 + 3N%S2 + 4Na2S + l4C0, + 3CO. G. A mixture of- 481 Snlphate of soda. ..... 7.10 grams Carbon .............. 1.20 ,, Equal to 2NazSO4 t 4C, submitted to a dull red heat for 60 minutes over a Bunsen lamp, approximate temperature 1270" F., yielded for 100 Na&Oa em- ployed- 24.412 per cent. undecomposed Na2S04 26.867 ,, converted into Na$03 35.40 per cent.of the Na2SO* decomposed being converted into NhCO3. 22 '465 0.07 - - --- 19 *Y62 25'108 - - Employed Na&304.. ............ With ........................ Obtained as undecomposed carbon 9 9 9, Nn2S04 ,, carbonate Na2C03 . . .. sulphide Na2S ...... ,, sulphur, free S . . .... Y? ,, with Pt = S Tottl.1 obtained ................ Undetermined ...... i .......... Loss ......................... Total Na determined.. ........... C. 3 '379 13 -522 - - s. I 0. 22'535 45.07 - - - - 5*490 10.880 - 9.082 9'124 - 3.155 - 4'746 - Na. 32 -395 - - 7 *822 8 -704 13 -116 - - 29 -642 32 *38 - - NazSO,. loo 000 -- - - '142 26 -867 4.0 *a5 - - 91 -494 8 '506 - - The Na accounted for by the Na2C03, NhS, and undecomposed Na2S04 amounts to 29.642 only, while the amount actually determined is 32.38, the difference 2.738 calculated into Na& = 3.809 S = 6.547 Na&.The total sulphur, free and combined with platinum, is 7.901, this - 3.809 = 4.092 combined with Na,S to form N+S2 = 9.974 NhS = 14.066 Na&; this added to 6.547 Na& = 20.613 N&S2, and as the NhS as found from the H2S evolved = 22.24 - 9.974 = 12.266 as Na&. The loss of carbon amounts to 13.522, and of oxygen to 25.108, while the NhsOp has been decomposed to the extent of 75.858 per cent. Calculating as in case F on the same proportion of carbon as 75.858 there has been of Na2S04 decomposed, we have 16.901 x ~ 100 = 12.820 - 2.370 (= the carbon aa Na.$03) equals 10.550 carbon combined with the 25.108 oxygen. Then-482 MACTEAR ON THE PART PLATED BY CARBON - 5.629 - 9.466 4.183 3.268 22.536 - - - 10.550 C + 14.0660 0 = 24616 CO, 25.108 0 - 14.066 0 = 11.042 0.This into C02, 11.042 + 19.323 CO = 30.365 C02, 24616 CO - 19.323 CO = 5.293 CO, which lead to the approximate formula for the decomposition of- llNs2SOd + 22 C = 4Na2C03 + 4Na& + 3NazS + 14c02 + 4CO. H. A mixture of- Sulphate of soda.. .... 7.1 grams. Carbon.. ............ 1.2 ,, Equal to 2NazSO4 +- 4C, submitted to a dull red heat over a Bunsen lamp for 75 minutes, approximate temperature 1260" F., yielded for 100 N%SO, employed- 24.98 per cent. as undecomposed NhSO,. 25.68 ,, ,, converted into N*CO,. 3423 per cent. of the N%SOp decomposed being converted into N&$O3. - - 11-259 8.092 8.681 8.319 - 13 *591 - - - - --- 19.94Ll 30.002 - - 25.130 - - 32.387 Employed N%SOI.............. With ........................ Obtained as undecomposed C .... 77 > 7 Na2S0, ,, sulphur, free 8 ..... 3, ,) with Pt = S Total obtained ................ Undetermined ................. rota1 Na determined.. .......... ,) carbonate N%C03 .. ,, sulphide Na,8 ...... p33s ......................... ~ c. I - ' 16 -901 1 0.930 2 *170 - 3 -100 13 -801 - - I I I S. I 0. I Na. I NszSO,. I- 1-1- - 24 -98 25 *68 41 -95 92 *616 7 '385 ~~~~~ ~~ The Na not, accounted for as undecomposed NhSO,, or as Na2C03, and Na& amounted to 2.393, and this calculated into NhS2 = 3,329 S = 5.722 Na2S2. The total sulphur, free or combined with platinum, amounted to 7.541 ; deducting 3.329 as above, we have 4.122 ; this calculated into N%S2 = 10.047 NazS = 14-169 Na2S2; this added to 5.722 N*S2 = 19.891 NazS2, and as the N&S as found from H2S evolved = 23.047 - 10-047 = 13.00 NaaS existing as such.Then there has been decomposed 100 - 24.98 = 75.020 per cent. of the N&SO4 employed, and calculating the carbon in the same propor- tion-IN REDUCING TEE SULPHATES OF THE ALKALIS. 483 - 16.901 2.211 2'297 - - - 4.508 12'333 - - 16.901 x - = 12.679 - 2.17 (= the carbon as carbonate) = 10.509 carbon combined with 25.1 30 oxygen. Then calculating the carbon in the first instance into CO- This into C02, 10.509 C X 14.012 O = 24.521 CO, 25.130 0 - 14.012 0 = 11.118 0. 11.118 0 + 19.456 CO = 30.574 CO2, 24'321 CO - 19.456 CO = 5.065 GO, leading to the approximate formula- 3N&s04 + 6C = Na&03 + NazSz + Na?S + CO + 4co2.I. A mixture oE- Siilphate of soda.. .... 7.1 grams. Carbon .............. 1.2 ,, Equal to 2N&S04 + 4C, submitted to a dull red heat over a Bunsen lamp for 60 minutes, approximate temperature 1235" F., yielded for 100 NaS04 employed- 19.781 per cent. as nndecomposed NhS04. 27.181 ,, ,, converted into Na2C0,. 330883 per cent. of the N&so4 decomposed being converted into NhcO3. 22 -535 - 4 '458 - 9.543 4 *070 4 -454 22.525 - -- - - - Employed Na2S04. ............. With ........................ Obtained as undecomposed C .... 9, Y Y N 4 0 4 .. carbonate Na&03. .. .. sulphide NezS ...... .. sulphur, free S ..... Y, ,, with Pt. s.. Total ohtained ................ Undetermined ................ Loss.. ....................... Total Na determined............ c. I s. 0. 32 -395 6 -408 8,805 13 -718 - - - - 28 '931 32 -45 - - The Na not accounted for as N%SO, nndecomposed, or as NhCO3, and NazS amounted to 3.464, and this calculated into N&Sz = 4.819 S The total sulphur free or combined with platinum amounted to 8.524 ; deducting 4.829 as above, we have 3.705 ; this calculated into Na& = 9.031 NhS = 12.736 Na& ; this added to 8.283 N*SZ = = 8.283 NhS2.484 MACTEAR ON THE PART PLAYED BY CARBON 21.019 Na2Sz, and as the Na2S as found from HzS evolved = 23.261 - 9.031 = 14.23 N%S existing as such. Then there has been decomposed 100 - 19.781 = 80.219 per cent. of the Na2S04 employed, and calculating the carbon in the same pro- 80.219 portion, 16.901 x - = 13.557 - 2.297 (= carbon as NazCOs) = 100 11.260 carbon combined with 26.967 oxygen.Then calculating the carbon in the first instance to CO- 11.260 C = 15.013 O = 26.273 CO, 26.967 0 - 15.013 0 = 11.954 0. 11.954 0 + 20.929 GO = 32.873 CO,, This into Con, 26.273 CO - 20.929 CO = 5.354 CO, leading to the approximate foi-mula- 3Na2SOp + 6C = Na2C03 + Na2S2 + Na8 + CO + 4c02. Surnrnary of the Experiments. A and R. These two experiments were merely preliminary, and to prove whether carbonate of soda was or was not formed in the decom- position of sulphate of soda by carbon at a red heat. C. This experiment shows the effect of excess carbon and higher temperature than red heat, the decomposition giving only a very small proportion of carbonate of soda, and approximating closely to the formula- NazS04 + 4C = N@S + 4CO.D. This experiment gave results closely agreeing with C, the carbo- nate of soda formed being equal only to 1.68 per cent. of the sulphate of soda actually decomposed. E. This experiment shows that with the proportions used, submitted to a dull red heat., 19 per cent. of the sulphatc of soda employed was converted into carbonate of soda. Note.-A series of subsidiary experiments showed that under a dull red heat little or no decomposition took place. F. This experiment shows that a mixture of 2NaJ304 + 4C, sub- mitted to a dull red heat for 60 minutes, has given 29.24 per cent. of the sulphate of soda decomposed, as carbonate of soda, the reaction approximating closely to the formula- 10Na&304 + 20C = 3Nn2C03 + 3NazS2 + 4NarS + 3CO + 14COa, the temperature being about 1270" F.G. This experiment, with the same mixture as in F, submitted for 60 minutes to about the same degree of heat = 1270"F., gave 335.4 perIN REDUCING THE SULPHATES OF THE ALKALIS. 485 U. 11 22 4 4 3 4 14 cent. of the sulphate of soda decomposed as converted into carbonate of soda, the reaction approximating closely to the formula- llNa2S04 + 22C = 4Na2C03 + 4Na2S2 + 3Na2S -+ 4CO + 14c02. H. This experiment with the same mixture as in F and G, was sub- mitted to a dull red heat for 75 minutes. Temperature about 1260" F. Gave 34.23 per cent. of the sulphate of soda decomposed as converted into carbonate of soda, the reaction approximating closely to the formula- 3Na804 + 6C = Na?.C03 + Na2S2 + Na2S + 4C0, + CO. I, This experiment was similar to G, except that the temperature was about 1233" F.It gave 33.883 per cent. of the sulphate of soda decomposed as being converted into carbonate of soda, the reaction approximating closely to the formula- 3Na2SOa + 6C = Na&03 + NhS2 + Na2S + 4c02 + GO. Assembling these formulae we have- H. --- 3 6 1 1 1 1 4 I F* Na2SOd.. ............. c .................... Na&O,. .............. Na292 ................ Na2S ................. co .................. C02 .................. 10 20 3 3 4 3 14 r. Total. -- 27 54 9 9 9 9 36 Mean. -- 3 6 1 1 1 1 4 These experiments show that :- 1st. When sulphate of soda is mixed with a quantity of carbon in excess of two eqnivalents, and exposed to a temperature considerably above red heat, out of contact with air, the reaction which takes place is, that the sulphate is ultimately converted into monosulphide of sodium, while the oxygen is evolved as carbon monoxide- Wa&04 + 4C = Na2S + 4CO.2nd. When sulphate of soda and carbon, finely divided and inti- mately mixed, are exposed to a temperature of dull redness (say from 1150" to 1300"F.), out of contact with the air, there is formed a cer- tain quantity of carbonate of soda, with an equivalent proportion of sodium bisulphide, as well as sodium monosulphide, while the carbon combines with the oxygen of the sulphate of soda to form carbon mon- oxide and carbon dioxide. The amount of carbonate and bisulphide formed varies inversely as VOL. XXXIIJ. 2 N486 MACTEAR ON THE PART PLAYED BY CARBON, ETC. the temperature, while the amount of sulphate decomposed varies directly as the temperature.The reaction which is most in accordance with these experiments being- 3Na&304 + 6C = Na&O, + Na2S2 + Na2S + 4co2 + CO. Any excess of carbon acts apparently in the first instance on the C02, reducing it to CO, and finally on the Na2COJ + N%S2, producing Na2S + CO. The decomposition of sulphate of potash by carbon takes place at lower temperature, but as shown by a number of experiments, the decomposition is of the same character as in the case of sulphate of soda, carbonate of potash and bisulphide of potassium being pro- duced. [Although Liebig was clearly the first to explain the reaction of wood-fibre on Na2S, such a method of producing alkaline carbonates had been used practically many years previously by soap-boilers, and is thus described in a note book before me dated 1800 :- " To make Hepar of Soda or Potass from the Sulphates.(' It is necessary to dry these salts in a reverberatory furnace, then mix them with their weight of fir sawdust, and fuse them in a rever- bemtory furnace. When the surface becomes calm, the operation is complete-let the sulphure run out. If it is a sulphure of potasa, break it up with a hammer, dissolve in water, evaporate it, and when the salt begins to form, put in sawdust till it is thick, '( Then put it into the calcining furnace, roast it for an hour, two- thirds of the sulphate will be decomposed. Separate the undecom- posed part by solution. " If you operate on the sulphate of soda, after the first sulphure is produced, add more sawdust, melt a second time to a sulphure, then operate as with the potass."] It is not too great an assumption that this is the general type of decomposition for all the alkaline and earthy sulphntes, the well known slow decomposition of soluble sulphates by organic matter showing that the formation of bisulphides such as FeSz is constantly occurring in the operations of nature. Even when the materials used in above experiments are moistened with as much as 40 per cent. of water and then heated, the reaction st,ill remains apparently the same, the water no doubt being all driven off before the decomposition of the sulphate by the carbon takes place ta any extent.SORBY ON THE DETERMINATION OF THE INDEX, ETC. 487 Having established the formation of Na&O3 and Na&&, it seemed not unlikely that. in the usual method of manufacturing alkali by Leblanc’s process, that the second atom of sulphur in the Na& would be driven 06 while the NazS then remaining was converted into Na&03 and CaS, and a series of experiments made to elucidate this point showed most conclusively that no appreciable quantity of sulphur was lost in this way working in closed crucibles. As, however, the N*S, oxidises very rapidly when exposed to the air while hot, it is probable that a small proportion may be thus lost, when the operation is conducted in large furnaces, where an oxidising flame is employed. I trust to be able shortly to complete an investigation upon this and kindred points in the decomposition of sulphate of soda by carbon and carbonate of lime.

ISSN:0368-1645    

DOI:10.1039/CT8783300475

出版商:RSC    

年代:1878

数据来源: RSC

摘要:

SORBY ON THE DETERMINATION OF THE INDEX, ETC. 487 LVIIL-On the Determination of the I d e a : of Refraction of Lipids by mearw of the Microswpe. By H. C. SORBY, F.R.S., President of the Geological Society. SO long ago as the year 1767, the Duke de Chaulnes pointed out how the index of refraction of a transparent parallel plate could be deter- mined by measuring its thickness, and the amount; of the displacement of the focal length when an object is viewed through it with a micro- scope. The method does not, however, appear to have been practically utilised, either in the identification of minerals or in the examination of liquids, and to have become all but lost sight of, until again brought forward by Dr. Royston-Pigott. In the Molzthly Microscopical Journal, vol. xvi, 1876, p.294, he described his refractometer for the study of liquids, in which the displacement of the focal length was measured by means of a micrometer screw, placed under the stage of a specially constructed instrument. My aim has been to make such additions to an ordinary microscope that the thickness of the plate and the amount of the displacement of the focal length could be measured with sufficient accuracy, without in any way interfering with the general use of the instrument, especially in the study of minerals. This I have accomplished in two ways, one being most suitable for somewhat thick objects, and the other for those which are thin. I propose in the first case to describe the apparatus, as applied in the case of any transparent substances bounded by parallel planes, a3d afterwards to explain in some detail its application to such determi- 2 N 2488 SORBY ON THE DETERMINATION OF nations as are likely to be of much interest to the Fellows of the Chemical Society.Thick Objects. When the object is somewhat thick, it is far the best to measure its thickness and the displacement of the focus by means of a gra- duated scale and vernier. A microscope of the Jackson model is admirably suited for this purpose. The silver scale, graduated to 2 . t h of an inch, is attached to the body of the microscope, and a rernier is fixed to the supporting limb. If made and used with care, there is no difficulty in reading off to &&h of an inch ; but since it is difficult to avoid small errors in the graduation, it is best to take more than one series of observations, in different parts of the scale and vernier, which is easily done by altering the position of the object- glass by means of the fine adjustment screw.It is also well to make several measurements at each adjustment, and take the mean of the whole. In this manner very accurate results may be obtained, and the extreme errors can be reduced to probably less than &$h of an inch. The thickness of the object is ascertained in trhe following manner : -Supposing it to be a section of a mineral or other transparent solid, it is fixed down to a supporting glass with Canada balsam, care being taken to squeeze out as much as possible, so that the subjacent layer may be very thin. The glass is then placed on the stage of the micro- scope, and so arranged that particles of dust on its upper surface, near to the object, are in focus.The exact position on the scale of this focal point is observed, and then the object pushed forwards, so that the position of the focal point for particles of dust or minute scratches on its surface may be similarly determined. Subtracting one from the other, we thus ascertain the true thickness of the object. Several precautions are, however, necessary, in order to avoid small possible errors. I n the first place the supporting glass may not be of absolutely uniform thickness. I find that the plate-glass generally used for such a purpose is almost always slightly wedge-shaped, but this is of little consequence if the object is so mounted and placed that the glass may be slidden along the line in which it is equally thick.This is easily ascertained by previous examination with the microscope. The section of the mineral or other object may also itself be nneqnally thick, and error might arise if the displacement of the focal point were observed in a different place to that of which the thickness was measured. This source of error is, however, readily avoided by having a stop on the stage, up to which the glass is pushed, and the object arranged in the best position by means of the sliding motion of the stage. Another possible source of error is the variation in the focal length of the eyeTHE INDEX OF REFRACTION OF LIQUIDS, ETC. 489 itself. This may vary very distinctly, according as the focus is adjusted by moving the body of the microscope up or down; in the one case the eye being strained to a short, and the other to a long, focal adjustment.This error may be removed by fixing at the focal length of the upper lens of the eye-piece a small thin glass plate, on which two fine lines are ruled, crossing in the centre. The focus is then so adjusted that particles of dust or small scratches are distinctly visible along with these cross lines. If the instrument is properly constructed and carefully used, I am not aware of any other material source of error. In all cases the very smallest particles of dust should be ob- served, since there is then the least range in focal adjustment, Thin Objects. If the thickness of the object be not too great to be measured by means of the fine adjustment, and especially if it be very thin, it is much better to make use of this method than to employ the graduated scale.I n doing so we, however, encounter certain special difficulties. If the milled head of the fine adjustment turns on the screw, and the small arm which moves the lenses rests on a washer, small changes in the points of bearing may give rise to very appreciable dxerences in what should be uniform readings. In order to overcome this source of error, I have had constructed a fine adjustment, in which the milled graduated head is fixed to the screw, which terminates in a point, working against a small flat steel plate, so as to give an uniform bearing ; care has also been taken to ensure the accurate working of the other bearings, and the result is that consecutive measurements, made with moderately high powers, do not differ more than fa&aath of an inch.I have not, however, succeeded in so constructing the apparatus that the actual values of the measurements made by the revolution of the graduated head are perfectly uniform for the upper and lower part of the range of movement; but, as explained below, this does not sensibly vitiate the final result, Displacement of the Focus. After trying various arrangements, I found that the best object to use to determine the amount of the displacement of the focus caused by looking through a transparent plate is the image of a grating of fine cross lines, ruled on a piece of glass. This is placed as far as the build of the microscope will permit below the lenses of an achromatic condenser, corrected so as to give a sharply defined image through superimposed glass about &th of an inch thick.When using only a moderate magnifying power, I find that lines ruled &th of an inch apart give a good result, but with higher powers finer lines d b t h of490 SORBY ON THE DETERMINATION OF 1.30 ........ -0063 1-40 ........ -0082 1.50 ........ .0100 1-60 ........ *0117 an inch distant are better. The amount of the displacement is easily observed. If the object is of some size, and perfectly transparent and free from flaws, no great care is needed in arranging it on the stage of the microscope ; but, if only certain parts are clear and free from irregularities, it is necessary to move the stage backwards and forwards until a place has been found which shows the lines sharply defined, and not distorted or broken up by irregular refraction or reflection.Having found such a place, the position of the focus is determined, either by the scale or by the graduated head of the fine adjustment, and then the object is pushed sufficiently aside to enable the observer to similarly determine the position of the focus, as seen through the supporting glass alone. The difference between the two sets of measurements gives the amount of displacement due to the transparent plate under examination. 1.70 ........ *0133 1-80 ........ -0149 1-90 ........ ,0165 2.00 ........ *0180 Calculation of the Index. Having determined in the manner above described the thickness of *he parallel plate (T), and the amount of the displacement of the focus (d), the value of the index of refraction (p) is approximately given by the equation,p = - This has been adopted as strictly true by previous authors, but when I came to measure carefully the index of various minerals and other substances, I found that it invariably gave too high a result, and on fully examining the question I found that, as suggested to me by Professor Stokes, it may be very confidently attributed to the spherical aberration of the object-glass, due to the introduction in front of it of the parallel plate.If the object-glass has been so perfectly corrected that the focal length for the central and marginal rays is the same, I find that the excess in the value of the index is independent of the thickness of the plate, and very closely T T--d* aperture 2 ' expressed by the formula, 4 sin2, - The co-efficient 4THE INDEX OF REFRACTLON OF LIQUIDS, ETC.491 These quantities to be deducted from the observed index would not necessarily apply to any other object-glass. If, however, the general kind of correction were the same, and only the aperture different, a similar table for its errors could easily be calculated from one accurate observation. Supposing that the apparent value in the case of a sub- stance having an index of about 1.50 were *0090 too great, it would merely be requisite to multiply the numbers in the table by '9, that is to -0090 say, by - .0100' Maya@jing Poum used. In determining the index of objects varying from *1 to -4 inch thick, I have found a +inch object-glass the most convenient, accompanied by an eye-piece the upper lens of which has a +-inch focal length.Such a magnifying power answers admirably when we wish to deter- mine the index with great accuracy, and have at disposal an object *3 or -4 inch thick. I n such a case the measurements are of course made by means of the graduated scale. When only thin objects me at com- mand, I have found that a -;th or +th object-glass, with the same eye- piece, gives good results, but in this case the measurements are made with the fine adjustment screw. As previoiisly named, the value of the upper and the lower range of movement is not exactly the same, but this source of error may be overcome by measuring the clisplace- ment of focus at both the upper and lower range of the screw required to measure the thickness of the object.By a very careful series of observations, I have found that, in the case of my own instrument, the mean of such measurements bears an accurate ratio to ths observed thickness, as measured by the entire range, so that the calculated index is sufficiently correct for all practical purposes. Light employed f o r Illumirmtiolz. In some cases with doubly refracting crystals, if white light be used, t-he lines of the grating are not sharply defined, but are seen as coloured bands. By 'employing for illumination the light transmitted by a moderately dark ruby glass, tinted by suboxide of copper, the lines are well defined, even in the above-named special cases. On the whole, I find that blight gas-light transmitted through such red glass is the best illumination, and that the mean index for such light corre- sponds closely with that for the solar line C.I therefore usuallymake use of it, and look upon the results as equivalent to the index for that line.492 SORBY ON THE DETERMINATION OF Applications of the Method. I have already treated at length* on the determination of the indices of minerals, and on the somewhat remarkable phenomena which are seen in the case of uniaxial and biaxial crystals. Since I thought it desirable to describe more particularly the construction and use of the instrument, and am anxious to direct attention to the measurement of the indices of liquids, it appears to me desirable not to enter into the consideration of the special characters of doubly refracting crystals.It may, however, be well to give a few examples in the case of minerals by way of showing how far the measurements of the indices may be relied on. As an illustration, I will refer to a number of observations made with an excellent section of quartz, -372 inch thick, cut parallel to the axis. The observed index of the ordinary ray was, in different sets of measurements, made as described above, 1*5513,1*5531,1*5524, 1.5531, and 1.5513. The mean of these is 1.55224, and, as will be seen, none of the separate observations differ from this by more than a single unit in'the third place of decimals, which must be looked upon as a very satisfactory proof of the accuracy with which the measure- ments may be made.Though one would scarcely expect it in such a mineral as quartz, yet my observations certainly indicate that there is a decided diflerence in the index for different specimens. The observed values were 1.5322, 1.5582, 1.5478, 1.5552, and 1.5593, the mean of all being 1.5545. In accordance with the principles desciibed above, we must deduct from this -0108, which leaves 1.5437 as the mean index freed from the errors due to spherical aberration. Rudberg gives as the index for the line C 1.5418, which s.grees with my mean result as closely as could be expected, when we consider that it varies in diffe- rent crystals. In a similar manner I found that the index for the ordinary ray of Iceland spar varies slightly in different specimens, but the mean was 1.6537.According to Rudberg, that for the line C is 1.6545. I also found that the index for green fluor was 1.4352, and according to Des Cloizesuz it is 1.434. It will thus be seen that when proper correction is made for the effects of spherical aberration of the object- glass, the errors in the observed indices may be reduced to only one or two -imit,s in the third place of decimals, which must, I think, be looked upon as t D perfectly satisfactory result. It differs only ,0004 from that of my best object. Correction for thin Glass Covers. In many cases it is very desirable, or even necessary, to cover the upper surface of an object with thin glass, cemented down with Canada * Anniversary Address, Mineralogical Society, 1877. Anniversary Address, Royal Microscopical Society, 1878.THE INDEX OF REFRACTION OF IJQUIDS, ETC.493 balsam. In calculating out the results, due allowance must of course be made for the effects of this cover. The index of the glass com- monly used, uncorrected for aberration, is about 1.53. In making our observations, we must determine the position on the scale of minute particles of dust or scratches on the upper surface of the mineral, and also that of particles of dust on the upper surface of the glass. The difference between these is the apparent thickness of the cover, along with a small layer of balsam. The real thickness is of course this apparent thickness multiplied by 1.53, that is to say, for every 1-00 of apparent thickness of the cover, the focus of any object seen throngh it is as it were raised -53.This applies to both the upper surface of the mineral and to the lines of the grating, so that from the apparent thickness, as measured from the surface of the supporting glass to the scratches, &c., on the top of the specimen, and also from the displace- ment of the focus, we must deduct an amount equal to the apparent thickness of the covering glass multiplied by -53. Determinutioft of the Index of Liquids. If the liquid be not very volatile, and the amount at command ample, it is on the whole best to use a glass cell made from stout tubing, with an internal bore of a f or +th of an inch. The length may be + inch, or somewhat less, according to the focal length of the object- glass employed. This cell must be fixed down to a suitable piece of plate-glass, and when filled with the liquid must be closed at the top with a thin glass cover.If the liquid is water, or an aqueous solution, this cover may be cemented down with a little gold size. In calcu- lating out the results, allowance must of course be made for the effects of the thin glass cover, but if it be sufficiently large to project some- what beyond the outside of the cell, the focal point, of the grating may be observed through it in such a, manner that the value of d is true without need of any such correction. If the cell could be cut so that the opposite faces were absolutely parallel, the real thickness of the liquid might be determined by measuring the distance between tho supporting glass and the thin cover ; but, since we cannot rely on the result to z$5ath of an inch, it is better to measure the thickness of the cover, and make the necessary allowance in the manner described above, or else to adopt another method.In no case is it desirable to trust to the depth of the cell itself, since the thickness of the film of liquid between it and the cover may vary. We may, however, measure the apparent thickness ( t ) of the liquid from the upper surface of the supporting glass to the under side of the cover, which, of course, is equal to the real thickness (T), less the494 SORBY ON THE DETERMINATION OF displacement of focus (a). The formula for t + a T then becomes -. If the amount of liquid at disposal is only T--d’ t THE INDEX, ETC. the index of refraction, small, it is not desirable to use a cell, but we Gay very conveniently employ a so-called live box. This has also tho advantage that it enables us to determine the index of very volatile liquids without any difficulty.The apparatus to which I refer has a bottom glass raised up in a short tube, over which slides a cap holding the thin glass cover. A small drop of the liquid may be placed in the centre of the lower glass, and the upper glass brought down until it touches the liquid, and, if necessary, somewhat drawn back again, or so arranged that the space between the glasses incloses a suitable thickness of liquid. It is then easy tlo measure the true distance between the glasses close to the included drop, and the dis- placement of the focus due to it. We thus obtain at once the true values of T and d, independent of the effect of the thin glass cover. Being so completely inclosed, volatile liquids evaporate very slowly, and, if they do, the only effect is to reduce the width of the drop, without in any way changing the thickness. As an example, I may give the results obtained in the case of water, using a cell nearly & inch in depth. Corrected for the effects of spherical aberration, I found that the index was 1.329. That for the line C gitJen in Watts’ Dictionary of Chemistry is 1.330, which shows that with a cell & an inch deep, and measuring with the graduated scale, very accurate results may be obtained. ,By using the fine ad- justment fixed to my microscope, an equally accurate determination might be made with m$mth part of a cubic inch of a liquid inclosed between the glasses of a live box. In order to insure so small an error as only one or two units in the third place of decimals, great care is required; but, if it were unnecessary to determine the index true beyond the second place, the method I have described will be found very convenient, when only small quantities of liquids are at disposal. This fact;, coupled with the great simplicity of the measurements, is, I think, a strong recommendation to the method here described.

ISSN:0368-1645    

DOI:10.1039/CT8783300487

出版商:RSC    

年代:1878

数据来源: RSC

摘要:

495 LIX.-O% the Action of Sodium om the Bemyl Ether of Isobzctyric Acid. By W. R. HODGKINSON, Ph.D. I SHOWED, in 1877 (BerZin Berichte, x, 65), in conjunction with Max Conrad, that the action of sodium on the benzyl ether of acetic acid is not analogous to that which it has on the acetic ethers of the lower numbers of the CnH2, + 10H series of alcohols. Whilst the latter form according to one or other of the following equations, the CnH2, + (or alkyl) ether of sodium acetoacetic acid and sodium ethylate :- CO.CH3 + H2 + Na0.C2H5. CH3 I’ * CO.OC2H5 + Na2 = I CHNa. CO. 0 C2H5 CO.CH3 CH3 I CO.OC2Hs I 11. 3 I + 2N% = CHNa + 3Nao.c2H6. C 0. 0CzH5. the primary products of the action of sodium on benzyl acetate are sodium acetate and the benzyl ether of hydrocinnamic acid (hydro- cinnamein) :- CH3 CH3 CH2.CH2.C6H, CO.OCH2.C6H:, COONa GO.O CH2. C6H5 111. 4 I +Naa=21 + 21 + H2. and, secondarily, hyclrocinnsmic and cinnamic acids as sodium salts and toluene :- CH2.CH2.CsHS CH2.CH2. CsH, CHZCH.CeH5 IV. 2 I +h=I + I + C 0. OCH,. C6H5 CO.ONa CO.ONa Hydrocinname’in. Sodium Sodium phenyl-propionate. phenyl-acrylate. 2CrjHs,CH3. CHs. C HpC6H5 CH=CH.C~HEI, v. 2 I + Na2 = 21 + H, + 2 C J L CO.OCH2. C6H5 CO.0Na In an extension of our work we found that the ethers of propionic and normal butyric acids behave in quite an analogous manner : the ether of the latter acid giving, according to reaction 111, the benzyl ether of a-benzyl-butyric acid or benzyl-ethyl hydrocinnamate :-496 HODGKIXSON ON THE ACTION OF SODIUN . ___.____._.____.____------.--- which also, following equations IV and V, further yield the sodium salts of ethyl hydrocinnamic and ethyl-cinnamic acids and toluene.Although the benzyl ethers of the normal fatty acids, or more gene- rally of the (C,Hz, + 1.CH2.COOH) series of acids, comport themselves similarly towards sodium, the same, for want of experimental evidence, could not be said of ethers of acids containing 2(CnH2, + In the case of isobutyric acid, the action could only be predicted to proceed so far as the formation of a benzylated acid, according to the reasoning of equation I11 :- CH(CH3)2 C(CH,)2.CH2.C6H, CH(CH3)Z VI. 41 +Na2=21 +21 + HZ. CO.OCH,.CsH, C 0. O CH,. C6H5 COONa This benzyl-dimethacetic acid could evidently not be further affected after the manner of equations IV and V.I have therefore studied, to some extent, the action of sodium on benzyl isobutyrate. Benzyl Isobutyrate, CH. (CH3)&0.0CH2.CsH5. This substance was prepared similarly to the benzyl ethers of the normal fatty acids by heating potassic isobutyrate with an equivalent quantity of C6H5.CH2C1, dissolved in alcohol, for four or five days, in a retort attached to a condenser ; the mass was then treated with water, and the oil which separated fractionally distilled. The oil came over completely between 225" and 230°, and a second distillation yielded the pure substance boiling at 228" (normal thermometer entirely io vapour). Two combustions gave the following results :- I. el475 gram = *lo30 OHz, and *3960 GO,. 11. -0995 ,, = -0712 ,, and ,2690 ,, I. 11. CI1 ....132 74.16 73.22 73.73 Hla .... 14 7.86 7-76 7.95 - - O2 .. .. 32 17-98 178 100.00 -- - It is a powerfully refracting liquid, of sp. gr. 1*0160 a t 2 2 O com- pared with water at 17.5", and has a very pleasant odour.ON THE BENZYL ETHER OF ISOBUTYRIC ACID. 497 Actiolz of Sodium on Benzyk Isobutyrale. 90 grams of the ether were treated with 8 grams of finely cut sodium in a retort with condenser. Scarcely any action took place at the ordinary temperature, but on warming, towards the melting point of sodium, a violent action with evolution of hydrogen set in, the sodium being almost entirely dissolved without further external heating. On adding water to the cooled product-a solid yellowish mass-an oil rose to the surface, a sodium salt being dissolved. (1.) The oil began to boil about 120°, but the temperature rose quickly to 250°, between which and 300" the whole passed over.By repeated distillahion of the fraction 120-200", a mobile liquid, smelling like toluene and boiling at 110-112°, was finally obtained; this gave the following figures on analysis :- I. 01658 gram substance = -1347 OH, and -5491 CO,. 11. *1727 gram (again distilled) = -1310 OH2 and ~5785 C02. I. 11. C, ...... 84 91.30 90.32 91.36 H g . . . . . . 8 8.70 9-03 8.48 - - 92 100*00 The fraction 250-300" came over on distillation mostly between 290" and 300", leaving each time a small residue of still higher boiling point. As it was possible that this high-boiling substance might be the result of decomposition at each distillation, I made combustions of different preparations of the 290-300" fraction with the annexed results :- (I.) 01420 gram = *0985 OH, and -4253 COz.(11.) -1242 ,, = .0900 ,, and -3769 ,, (111.) -1701 ,, = *1230 ,, and -5145 ,, @.I (11.) (111.1 C ...... 81.68 82-76 82.49 H ...... 7-71 8.05 8.05 0 ...... 10.61 9.19 9.48 These values of the three elements not only do not agree with that of the sought-for ether, C,aH2,02, but also do not agree amongst them- selves: the first of the above figures leading to the improbable for- mula C20H2202, the two latter giying the formula CZ3H2,02. (cEi%!Oo2-) (C2oHZ02.) (c23H26oP) C ...... 80.60 81.64 82-64! 11 ...... 7.46 7-48 7.78 0 ...... 11.94 10.88 9.58 100-00 100~00 100~00498 HODGKINSON ON THE ACTION OF SODIUM To prevent decomposition by possible over-heating, fresh portions of the fraction 250-300" were repeatedly distilled under a pressure of 40 mm., whereby the greater portion was easily obtained boiling between 200-210°, and a smaller residue at about 280".The fraction 200-210" boils under the ordinary pressure at 280" to 285", and is a somewhat mobile, colourless, highly refractive liquid of pleasant aromatic odour. The numbers obtained on analysis agree with the formula of benzyl- dimethyl-benxyl acetate :- I. -1428 gram = -1006 OH2 and -4200 GO2. 11. *lo84 ,, = -0765 ,, and *3180 ,, Refractioned at 285" (normal thermometer in vapour). 111. *1530 gram = -4530 GO2, water lost. C18H2002. I. 11. 111. C .... 80.60 80.21 80.01 80.75 H.... 7.46 7.82 7.84 - 0 .... 11.94 - - - 100~00 Its sp. gr. at 18" = 1.0285 compared with water at 17.5".The portion boiling at 280" under a pressure of 40 mm. distilled under ordinary pressure at 340" to 350" without decomposition, and on combustion gave tbe eollowing results :- *lo31 gram substance = -0736 OH2 and *3180 C02, approximating to the formula CIkHleO. Calculated. Found. Cia ...... 8400 84.12 His.. .... 8.00 7.93 0 ...... 8.00 - 100~00 This substance, which may be regarded as the principal product of the action of sodium on benzyl-dimethyl-benzyl acetate, will be again referred to later on. (2.) The watery solution of the product of the rea.ction of sodium on benzyl isobutyrate was strongly acidified with sulphuric acid and extracted with ether, which on evaporation left a residue of oily organic acids of strong isobutyric odonr separable by distillation into a portion boiling between 152-137", and remaining liquid on cooling, and a portion between 240" and 250" solidifying in the neck of the retort.The fraction 1.52-157" gave numbers agreeing with those required by isobutyric acid :- ,2365 gram substance = 01925 OH, and *4717 CO,.ON TEIE BENZYL ETHER OF ISOBUTYRIC ACID. 499 Calculated. Found. c1 ...... 54.55 54.40 H8 ...... 9.00 9.04 Or ...... 36.35 - 100~00 -1680 gram substance = ,0335 OHz, -1482 C02 and 00925 Ag. The d v e r salt gave :- Cq ...... 24.62 24.06 H7 ...... 3.59 3.54 Ag ...... 55.38 55.06 O2 - ...... 16.41 100~00 The portion boiling art 240" to 250°, and solidifying on cooling, was On distillation with steam and re- to all appearance benzoic acid.crystallisation it melted it 119-120" ; the analysis gave- I. -1700 gram = -0673 OHz and 04268 COZ. 11. 00950 ,, = *0445 ,, and .2395 ,, I. 11. C7 ...... 68.85 68.47 68.76 H, ...... 4.92 441 5-20 O2 ...... 26.23 - - 100~00 The products of the action of sodium on benzyl iaobutyrate are, therefore, hydrogen, benzyl benzyl-dimethacetate, sodic isobutyrate and benzoate, toluene, and the oil (Cl*H160),, the production of the first three of which is expressed in equation VI, sodic benzoate, toluene, and the oil, (C14H160)n, being probably the result of secondary action. Action. of Sodium 0% Benxyl Be%zyl-dimethacetnte. To test the coprectness of the above statement of the reaction, pure benzyl benzyl-dimethacetate was treated with sodium, the liquid being heated to the melting point of the latter to induce reaction, which then took place without further heating.The same solid product was treated with water as before. From the oily layer a small quantity was separated by fractional distillation, boiling between 110-112", and identified as toluene. On distillation of the hi.gher-boiling portion under a pressure of 40 mm., with the exception of a little unaltered ether boiling at 200- 210°, all came over as a thick oil at about 280°, and this latter portion500 HODGKINSON ON THE ACTION OF SODITJM distilled completely and without the least decomposition under ordi- nary pressure at 350-355". On combustion it gave the following numbers :- I. -1260 gram gave -0929 OH2 and -3890 CO,. 11. *0865 ,, -0625 ,, and ,2665 ,, These results proclaim the identity of this product and the one obtained in small quantity by the action of sodium on benzyl dimethacetate for the formula C,,H,,O requires- I.11. C ...... 84.00 84.20 84.03 II ...... 8.00 8.19 8.03 0 ...... 8.00 - - 100*00 The watery solution, treated with sulphuric acid and extracted with ether, furnished an oily mass from which crystals of benzoic acid separated on cooling. From the liquid portion a small quantity was obtained boiling below 200°, and by its physical properties was identified with isobutyric acid. The solid part, after distillation with steam and crystallisation, melted at 119-120", and gave the numbers required by benzoic acid. -2050 gram = *0970 OH2 and *5133 GO2 = H 5.26 per cent., C 68-29 per cent. ; theory requires H 4.92 per cent., C 68.85 per cent.(1.) Xaponij%ation of Benzyl Benxyl-dimetl~acetate. This ether is very slowly attacked by aqueous or alcoholic solutions of alkaline hydrates, and does not yield benzyl-dimethacetic acid as might have been expected. From 30 grams of ether boiled with a hot concentrated solution of baric hydrate for 14 days, only 1.5 gram of a mixture of barium salts, soluble in alcohol, was obtained. On addition of hydrochloric acid an acid oil was precipitated which partly solidified ; the solid pressed from the liquid portion, and dis- tilled with steam, gave crystals melting finally at 119-120", yielding on analysis :- *0957 gram substance = *0418 OH2 and 02410 CO, = H 41.85 per cent., C 68.68 per cent. ; theory for benzoic acid, The watery solution acidulated with hydrochloric acid, was ex- hausted with ether, and the portion of oil extracted boiling between 150-158", soluble in water, made into silver salt. -0640 gram substance = 00213 OH2, *C585 COZ, and -0350 Ag.H 41.92 per cent., C 68.85 per cent.ON THE BENZYL ETHER OF ISOBUTYRIC ACID. 501 Found. Ca ...... 2462 2493 H7 ...... 3.59 3.70 Ag ...... 55.38 5469 0, - ...... 16-41 Benzyl separated ether. 100-00 alcohol if formed during the saponification could not be from the comparatively very large volume of unchanged Concentrated alcoholic pofash boiled with the ether for nearly three weeks failed to saponify so much as the baric hydrate. A further proof of the non-production of benzyl-dimethacetic acid was afforded by the bebaviour of the ether on heating with soda- lime in a distillation flask.In the first experiment the temperature was raised to about 200°, when a somewhat considerable quantity of thin oily distilIate was obtained, smelling strongly of toluene, and coming over on first fractionation at 205-11 5". I. -0797 gram = :0650 OH, and 2593 CO,. 11. Q780 gram lost ,2585 COZ. Refractioned 109-110" gave 111. -1245 gram = .lo10 OH, and ,4176 CO,. C7H,. I. 11. 111. C ...... 91-30 88-73 90.38 91.48 H ...... 8.70 9.06 - 9.01 The residue in the flask washed out with water, decomposed with hydrcjchloric acid and extracted with ether, gave on evaporation of the latter a mixture of oily acids partly solidifying on cooling. Frac- tionation yielded much between 150-160", further identified as isobn- tyric acid.Its watery solution was made into calcium salt, which on concentration and heating to 70" did not precipitate. On crystallisa- tion it formed shining prisms of (C4H702)&a + SOH2, as the analysis shows :- I. *5705 gram, air dried, lost at 100" *1760 OHz. 11. .3203 ,, 9 , 9 9 *0985 OH,. I. 11. Per cent. Per cent. Per cent. Calculatcd, OH, ...... 29.61 30.85 30.75 Theory = CaO 26.17 per cent. 111. -4075 gram dry sdt, strongly ignited, gave -1077 CaO. III. = 26.43 per cent. The fraction above 200", solidifying on cooling, melted after purifica- VOL. XXXTIl. 2 0302 HODGKINSON ON THE ACTION OF SODIUM tion in the usual manner at 120°, and showed itself to be quite part, benzoic acid :- I. -0957 gram substance gave *0418 OH, and 52410 CO,. 11. -1592 gram (fused) gave -0722 OH, and -4010 CO,.I. 11. Cs ...... 68.85 68.68 68-70 H g ...... 4.92 4.85 5.04 0, ...... 26-23 - - 100~00 Its barium salt dried at 100" yielded- I. -1974 gram = -1018 BaC03. 11. 01250 gram = -0640 BaC03. (C7H502)2Ba = Ba 36.15 per cent. I = 35-86 per cent. I1 = 35.61 per cent. As other products than those described could not be found, the action of alkalis on benzyl-benzyl-dimethacetate may perhaps be re- presented thus :- C (C H,),. CH,. CgH, CH( CH3)2 VIII. I + 2NaOH = I + CH3.CsHe + GO. OCH,.CsH, COONa CsH5.COONa + H,. The conversion of benzyl compounds by alkalis into salts of benzoic acid and toluene was observed by Canni zzaro (AnnaZe?z, xc, 252), and it is only here remarkable that in the saponification the benzyl- group directly combined with the carbon of the acid radicle, as well as the ether-forming benzyl-group, should be split 0% an oxidation to benzoic acid and reduction to toluene simultaneously taking place.To get a t some idea of the structure of the oil CllH160, I attempted its oxidation with chromic acid, but after about ten days' heating, the amount of benzoic acid produced was so small that it only sufficed for qualitative recognition. Soda-lime at 200" had very little more effect, whilst baric hydrate solution and alcoholic potash have no action. As the simple production of benzoic acid does not throw any light on the constitution of the oil C,,H,,O, an equation fully expressing the action of sodium on benzyl-benzyl-dimethacetate, in which, together with hydrogen, toluene, sodic benzoate, and ieobutyrate, it is khe prin- cipal product, cannot as yet be formed.Conclzcsiorts from the above reszclts. (1.) Benzyl isobutyrate is exactly analogous in its behaviour with sodium to the bemzjl ethers of acetic, propionic, and normal butyric itcids :-ON TI3;E BENZYL ETHER OF TSOBUTPRIC ACID. 503 CH,.CH,.CH, CO.OCH2C6H5 41 The replaceability of hydrogen a t the carbon atom immediately in combination with carboxpl by the benzyl-group under the influence of sodium is probably general with all acids having such hydrogen com- binations. (2.) The benzylated benzylic isobutyrate does not, however, comport itself towards sodium like the before-mentioned ethers of the fatty normal series. In the latter case there is formed, together with toluene and salts of benzylated fatty acids, acids of the cinnamic series with two atoms less hydrogen. But, with the iso-body, C18H2002, such an action evidently cannot take place, the carbon directly connected with carboxyl having no hydrogen atom. (3.) In the saponification of ethers of normal benzylated fatty acids the benzyl-group of the acid radicle is not affected by alkalis, benzylated fatty acids being produced. CH,.CH,.C6H5 CH2. CH2. CSH5 C O.OCH2. C6H5 + a.OKH = I + (a-I)HOK + CO.0K I HO.CH2.CGHh. In the isobutyric group, however, the benzyl-group is evidently not so firmly combined, since alkalis reproduce isobutyric acid by saponifica- tion, the acid as well as the ether benzyl-group being liberated. Whether the saponification reaction takes place according to a previous equation, or after this one, c( cH3)2c6HD CH(CH,)2 2(C&COO??a) CO.0. CHzCsHs + 5NaOH = 3 I + + OH2, COONa 4CsH,.CHs 3 1 I do not at present venture to decide. This research was carried on in the University Laboratory, Wurzburg, under the direction of Prof. Wislicenus, to whom I am much indebted for kind assistance and advice. 2 0 2

ISSN:0368-1645    

DOI:10.1039/CT8783300495

出版商:RSC    

年代:1878

数据来源: RSC