摘要:

THE PUNGENT PRINCIPLE OF GINGER. PART I. 777 LXV.-The Pungent Principle of Ginger. Part I. The Chemical Characters and Decomposition Products of Thresh’s ( ( Gi.lzgero1.” By ARTHUR LAPWORTH (MRs.) LEONORE KLETZ PEARSON and FRANK ALBERT ROYLE. THE pungent principles of ginger the rhizome of Ziizgiber oficinale, were studied by Thresh ((( Year-Book of Pharmacy,” 1879 426 ; 1884 516; Pharm. J. 1882 [iii] 12 721). The pungency was attributed to an active principle ( ( gingerol,” an inodorous sub-stance having the general characters of the class of compounds now known as ('oleo- resin^," which he isolated by means of a somewhat complicated fractionation of the alcoholic extract of the drug. He attributed to it the approximate formula sC,H80 and found that on oxidation with chromic acid i t yielded ‘( apparently ” acetic and caproic acids (‘( Year-Book of Pharmacy,” 1884 520) together with a volatile oil.Thresh seems to have abandoned the work a t an early stage. Garnett and Grim (Pharm. J. 1907 [iv] 25 118; ( ( Year Bookof Pharmacy,” 1909 344 ; compare also ibid. 1907 443) re-examined the pungent oleo-resin gingerol and simplified and improved the method of isolation and purification used by Thresh. They frac-tionated the crude mixture of oleo-resins with the aid of petroleum, and ultimately obtained a (‘clear viscous oil of a pale straw colour distilling within a range of 235O C. t o 250° C.,” the pressure referred t o being 18 mm. They add “It could not be assumed without further proof that this body” (that is, the distillate) (‘was not either a product of destructive dis-tillation or that it was not contaminated with such products ” ; they showed however that the distillate had the pungency and many other characteristics of the original gingerol.I n the course of their experiments they noted that gingerol was possessed of phenolic characters gave a green colour in alcoholic solution with ferric chloride a precipitate with bromine and that the solution in alkalis especially when warm undergoes a change which results in a loss of the pungent character in the dissolved oil. The latter observation was made use of in devising a simple means of distin-guishing between extracts of ginger and of capsicum respectively, as the latter is not affected in pungency by alkalis and is often used t o adulterate or fraudulently to replace the former.Ths present authors were invited by Messrs. Garnett and Grier to continue the work on gingerol and carried out the experiments described in this paper with a large sample of an alcoholic extrac 78 LAPWORTH PEARSON AND ROYLE : of African ginger the most pungent variety on the market. This extract was kindly made fmor them by James Woolley and Sons, Manc he st er. Method Used in Isolatilbg Gingerol. The residue from the alcoholic extract was treated repeatedly with aqueous alcohol of 35 per cent. strength which is considerably weaker than was used by Thresh o r by Garnett and Grier; it was found that this dissolved the inactive resins much less freely and gave a more rapid separation. Alcohol was removed from the clear extract by evaporation in a vacuum a t the lowest possible tempera-ture the oil which separated being freed from as much water as possible and re-extracted with alcohol of about 50 per cent.strength. Milk of lime was then added as suggested by Thresh, until only a very slight deposition of calcium compounds took place, when the whole was allowed t o settle and the clear liquid made neutral to litmus by addition of dilute hydrochloric acid and again freeld from alcohol by distillation under diminished pressure. The resulting oil was separated from the bulk of water by decantation, dissolved in a solvent such as chloroform which was subsequently dried and removed by evaporation. The dry crude gingerol was then extracted fractionally with hot petroleum and those portions retained which dissolved freely in the hot petroleum but were redeposited on cooling.After removing all traces of petroleum, the “refined gingerol” was obtained as a viscous faintly yellow oil entirely soluble in dilute aqueous alkali and generally having all the properties assigned to the material by Garnett and Grier, Many various methods were tried for the further purification of the refined gingerol but even the best samples obtained were almost certainly not homogeneous materials and could be frac-tionate’d by means of light petroleum into portions having some-what different solubilities. As many of the less soluble fractions of the oils shared the characters of the most soluble portion it was suspected that gingerol either tends to polymerise or to decompose into simpler products such as those mentioned later, with subsequent recombination of these t o products more complex than gingerol proper.The most highly purified samples obtained were distilled in a cathode-ray vacuum ; they passed over almost completely when the temperature of the bath was 135-140° and the distillate was in such cases a clear faintly yellow oil. Even under these) conditions there was evidence of slight decomposition (odour of fatty alde-hydes) and only traces of camphoraceous solid separated from the oil after many months THE PUNGENT PRINCIPLE OE‘ GINGER. PART I. 779 Messrs. Gariiett and Grier directed the authors’ attention a t the commeiiceinent of the work t o the fact that some samples of gingerol give a crystalline product when they are shaken with a solution of sodium hydrogen sulphite; the present authors found that the1 most highly purified samples do not give any such product, but do so after they have been distilled or otherwise somewhat strongly heated or treated with dilute alkalis under certain con-ditions.Analysis of the1 best samples of gingerol obtained gave: C=71*4; H=9.2; C=71*6; H=9*4. C17H2e04 requires C = 69.4 ; H = 8.8 per ce’nt. C18H2804 , C=70.1; H=9.1 , ,, Thresh found C=71*27 71.39 71.43 and H=9*61 9.82 9-52. Determination of methoxyl in gingerol : Found Me = 7.3 7.7. lMeO in C17H2oO4 requires Me0 = 10.5 ; in C,8H,804 MeO=10*05 per cent. (This for reasons which appear later would indicate a purity of only about 75 per cent.) Other P w p e r t i e s of ( ( G’iiigerol.” The presence of one hydroxyl group a t least is indicatesd by the phenolic character of gingerol.The alkaline solution when treated with benzoyl chloridel benzenesulphonyl chloride and similar com-pounds or wi€h chloroformic esters deposited neutral non-pungent oils wliich did not show any s i p s of assuming crystalline form. The oil yielded 110 crystalline oxime or seniicarbazone although theire was some evidence that a nitrogelnous compound was formed on treating it with hydroxylamine. Slight heat was evolved on mixing phenylhydrazine with gingerol ; but neither with that sub-stance nor with substituted phenylhydrazinee benzyl- o r naphthyl-hydrazines did ginger01 give any crystalline derivatives. The sole derivative of gingerol which was obtained in definitely crystalline form was its monomethyl derivative.As this was easily purified whilst the purity of gingerol could not be guaranteed all conclusions as to the true composition and constitution of ‘’ gin-gerol ” have been based on an examination of ‘‘ methylgingerol.” Preparation and Properties of ( ( illethylgitzgerol.” Refined gingerol is dissolved in methyl alcohol and treated with methyl sulphate and potassium hydroxide successively. From the product ether extracts a neutral oil which deposits crystals and these can b drained and recrystallised from light peltroleum. About 15 grams of methylgingerol melting a t 6 4 O can be obtained from 24 grama of the best gingerol which the authors have pre 780 LAPWORTH PEARSON AJXD ROYLE : pared; this is again 'indicative of the mixed charactefrs of gingerol.Analyses of different samples of methylgingerol gave the follow-ing results: C = 70.21 70.44 70.17 70.15 70.05 70.2. H=8.64 9.08 8.92 9.46 9.19 9.0. Me0 = 22.6 19.8 19.8. Qualitative tests and analysis by the Dumas method showed that no nitrogen was present : C,,H,,04 requires C = 70.1 ; H = 9.2 ; 2Me0 = 20.1 per cent. ClgH,,O , C=71.2; H=8-75; 2Me0=20.0 , ,, C,gH,,O4 , C=70.75; H=9.4; 2Me0=19.9 , ,, The molecular weight of I' methylgingerol " was determined by the cryoscopic method in benzene. The numbers obtained were 271 309 308 316 323 323 the numbers required for C,,H,,O and clgHa04 being 308 and 322 respectively. I' Methylgingerol " crystallises in slender needles melting a t 64O.It is insoluble in cold alkalis or acids but is altered by these if hot (compare later). It is also slmowly decompose'd when it is heated above 150° and rapidly near its boiling poinb-the odour of fatty aldehydes becoming perceptible. It is optically active in 2 per cent. solution in chloroform 'having [a]$ + 27.3'. Ketonic Properties of '' Met Jtylgingerol." Methylgingerol appears to be attacked slowly by phenylhydr-azine etc. but the products were not obtained in a crystalline form. When it is warmed in alcoholic solution with hydroxylamine hydro-chloride and sodium acetate however it gives a crystalline deriv-ative which was obtained in slender needles from light petroleum. Found c=64-9 64 0 64.1 64.5; H=9*1 9.4 9.0 9.2; N=4*3.C,,H,gO,N,H,O requires C= 63.3 ; H= 9.1 ; N = 4.1 per cent. ClgH3,04N,H20 , C=64.2; EL'=9*3; N=3*9 , ,, The analyses indicate that the substance is methylgingerol oxime hydrate. When heated a t 110-115° for four hours it lost 4.09 per cent. in weight whilst the theoretical loss for 1H,O is about 5.1. The treatment was not pressed as there were signs of more profound decomposition. The oxime dissolves in cold hydrochloric acid and is reprecipi-t a t 4 by alkali. When it is hydrolysed by acids it is reconverted into methylgingerol and hydroxylamine the former after recrystal-lisation being unchanged in melting point. When this oxime is subjected to Piloty's test f o r ketoximes it gives a definite positiv THE PUNGENT PRINCIPLE OF GINGER. PART I.781 reaction a yellowish-green coloration being obtained in the ethereal layer. The whole of the phenomena observed during the various stages i f the test are not to be distinguished from those observed when the related methylzingeroneoxinw (p. 786) is treated in the ~ a m e way. Methylgingerol is a t once attacked by chromic acid giving a black compound (chromate?) much as many alcohols (for example men-thol) do. It is instantly attacked by phosphorus pentachloride o r thionyl chloride and hydrogen chloride is evolve'd. A cold solution of methylgingerol in chloroform does not a t once discharge the colour of bromine in the same solvent but an action takes place only on heating when hydrogen bromide is evolved. Pure methyl-gingerol is also stable towards cold permanganate in acetone solu-tion; even after long heating with excess of this reagent the bulk of the compound is recovered unchanged and this fact may be utilised t o purify the crude substance from more easily oxidisable materials.Prolonged action of hot aqueous perinanganate destroys methyl-gingerol fatty acids veratric acid and carbon dioxide being formed. It is not reduced in the cold by sodium amalgam or by hydrogen in the presence of oolloidal platinum or palladium. Tests far the Hydroxyl Group in Methylgingerol. I n order to ascertain whether methylgingerol contained free hydroxyl the compound in amyl ether was treated with mag-nesium methyl iodide in the same solvent in an apparatus similar to that used by Sudborough for the determination of hydroxyl by this method.On mixing the two solutions a white precipitate of an additive compound was instantly formed but no trace of methanel or other permanent gas was detected even when excess of magnesium methyl iodide was used and the temperature was raised nearly t o looo. The1 experiment was repeated seveaal times and always gave the same result. On the other hand methylgingerol is quickly attacked by cold acetyl chloride thionyl chloride or phos-phorus chlorides and hydrogen chloride is evolved ; this change takes place rapidly on heating. Again when methylgingerol is sealed up in a tube with rather less than one molecular proportion of phenylcarblamide it does not a t first dissolve in it but after some weeks in the cold the solid disappears and a clear viscid liquid is formed in which the odour of phenylcarbamide cannot be detelcted.No crystalline delrivatives were isolate'd in either in-stance but the observations more especially the last one are very difficult to explain escept on the assumption that methylgingerol contains a free hydroxyl group and the same remark applies to it 782 LAPWORTH PEARSON AND ROYLE : behaviour with chromic acid. The failure of the substance to give methane with magnesium niethyl iodide is possibly due to the insolubility of the additive product which the reagent seems to form by uniting with the compound at the ketonic carbonyl group. Oxidation of Ginger01 with Chromic Acid. Pormntiou of n-Heptoac and (probably) n-Hexoic Acid. Thresh oxidised gingem1 with chromic acid and obtained what he considered to be probably acetic and caproic (hexoic) acids together with a volatile oil (Zoc.cit. 1884 520). The present authors dis-solved about 50 grams of refined gingerol in warm acetic acid of about 90 per cent. Btrength and added solid chromic acid in small quantities a t a time until further addition caused no immediate effervescence. The liquid was then subjected to distilla-tion in a current of steam the latter portions of the dist'illats being collected extracted with ether and the1 latter evaporated. The re'sidue was rendered alkaline with sodium hydroxide and once more treated with a current of steam when a small quantity of a neutral oil probably identical with that rr,entioned by Thresh passed over. This oil was not examined more closely but the alkaline residues in the flask were acidified extracted with ether and the latter dried and fractionated.After some water and acetic acid had distilled, the main bulk of the residue amounting t o about 7 grams passed over a t 210-225O and was clearly a mixture of saturated fatty acids ; on refractionation the range was not appreciably altered (a-hexoic acid boils a t 205O; 11-heptoic acid a t 223O). The fractions, except the highest and lowest were mixed an& a sample wr?s titrated with standard alkali in the presence of phenolphthalein as indicator. The equivalent found was 125 which corresponds with about one part of hexoic acid and two parts of heptoic acid The portion of the oxidation product which was not volatile in steam was small and nothing definite could ble isolated from it.Oxidatiom of Methylgingerol. As the presence of phenolic hydroxyl in gingerol evidently led to destruction of the aromatic portion of the molecule by the oxidising agent the oxidation of 8 grams of methylgingerol by means of chromic acid in acetic acid solution was carried out and the products were worked up as in the immediately preceding descrip-tion. A mixture of fatty acids was obtained as before but from that part of the product which was not volatile in steam about 0.9 gram of acid was obtained on extraction with ether. This when Forrnatioq% of Fatty Acids and Veratric Acid THE PUNGENT PRJNCIPLE OF GINGER. PART I. 783 purified by crystallisation from hot water formed flat needles melt-ing a t 180-181° and its equivalent fo’und by titration with standard alkali using phenolphthalein as indicator was 184.It was readily identified as veratric acid (equivalent = 182). Fusion of Gingerol with Potnssiztnz Hydroxide. Formation of Stenhouse and Groves (T.> 1877 31 i 533) found protocatechuic acid in the product obtained by fusing cruds ginger resins with potassium hydroxide. The present authors tried the same experi-ment with refined gingerol and although the great bulk of the material carbonised a small quantity of protocatechuic acid was isolated from the product and identified. Prototechzcic ,4 cid. L4ction of Heat a12d of Hydrolytic Agents o n Gingerol. Formntioii of Aliphatic Aldehydes (maidy n-Heptaldehydr) and a Ketone ‘( Zingerone.” It has been nientioned that Messrs.Garnett and Grier had noticed that some specimens; of gingerol gave a quantity of mixed crystalline solids when shaken with aqueous sodium hydrogen sul-phitel. The present authors find that this is only the case with old specimens or material which have been heated alone or with acids or alkalis and such specimens have a peculiar’odour whilst pure ones are practically odourless. If a sample of gingerol which has been distilled under diminished pressure is subjected to a current of steam a certain quantity of a volatile oil with an odour resembling cenantliol (heptaldehyde), passes over. This if extracted from the distillate by ether is obtained as a colourless oil. If the ethereal solution is shaken with freshly prepared sodium hydrogen sulphite the bulk of the con-tained oil is converted into a mixture of crystalline additive com-pounds with the reagentl and theset after washing and draining, can be decomposed with aqueous potassium carbonate.The result-ing oil consists almost wholly of addehydes of the fatty series, mainly heptaldehyde ; when shaken with aqueous hydroxylamine acetate it loses its characteristic odour and yields more than one-half its weight of crystals which after being drained and recrystallised once from methyl alcohol form plates melting a t about 5 3 O . (Found C = 65.4 ; H= 12.0. C,H,,:NOH requires C = 65.1 ; H = 11.7 per cent.) The compound was compared with heptaldoxime (m. p. 53-55O) prepared from commercial enanthol ; the substances were identical in all respects 784 LAPWORTH PEARSON AND ROYLE : For the further proof that the compound forming the bulk of the above oil is n-heptaldehyde its cyanohydrin was prepared from the crystalline hydrogen sulphite compound and hydrolysed.The resulting hydroxy-acid (m. p. 66-67O) on titration gave the equiva-lent 162 (C,H,,O requires 160) and direct comparison of the corre-sponding substance from commercial cenanthol afforded proof of its identity with a-hydroxy-n-octoic acid. The oily residue left after passing steam through distilled gingerol if dissolved in ether and then shaken with aqueous sodium hydrogen sulphite frequently furnishes a small quantity of a solid hydrogen sulphite compound of a ketone; the bulk of this may remain dissolved in the aqueous layer from which as well as from the solid the ketone can be recovered by adding a slight excess of hydrochloric acid boiling off sulphur dioxide and extracting with ether.I n this way there is usually obtained a sweet-smelling oil with a very pungent taste. zingerone ” is proposed, was not obtained in solid form until a highly purified synthetic specimen solidified ; various specimens of the compound obtained from gingerol subsequently set t o solid masses by infection. I n order to obtain supplies of the ketone from extract of ginger, the authors Save since utilised the fact that gingerol is readily in part decomposed by hot baryta water into aldehydes mainly volatile in steam and the new ketone. By boiling crudel extracts of the drug with baryta water so long as t.he d o u r of cenanthol is perceptible in the distillate acidifying the residue extracting with ether and subsequently dealing with the ethereal extract in the manner above indicated the new ketone is conveniently and quickly obtained.The yield of zingerone from gingerol is very much below that theoretically possible ; this is doubtless due to the occurrence of polymerisation or secondary re-condsenlsation of the sensitive products of hydrolysis. The new ketone f o r which the name Decomposition of it! ethglgingerol. Formation of Fatty Aldehydes and Methylzingerone. Purified methylgingerol behaves towards hydrolytic agents and when heated alone in much the same way as does gingerol itself. The yields of the simpler products were in this case however more satisfactory especially when boiling baryta water was used and the fatty aldehydes were removed with the steam as soon as formed.Even from recrystallised methylgingerol the fatty aldehydes were of mixed character and this is in agreement with the mixed nature of the fatty acids obtained on oxidation (p. 782). The ket.onic product is in this instance not soluble in alkalis ha,s no perceptibl THE PUNGENT PRINCIPLE OF GINGER. PART I. 785 odour and as i t is but very slightly volatile in steam most of it remains in the flask after removal of fatty aldehydes. It is recovered by extraction with ether and is readily obtained in crystalline form. The properties of this compound which is the monomethyl ether of zingerone are described later. Properties of t h e New Ketone CllHI4O3.The compound when pure is a colourless solid which dissolves somewhat freely in most of the' usual organic media with the sxcep tion of petroleum and crystallises from ether in needles rhombo-hedra or large lustrous platels melting at 31-34". It has a difr tinct sweet odour reminiscent of salicylaldehyde and t o a less extent of vanillin. It has an e'xtremely pungent taste like that of ginger itself but quantitative comparisons have not yet been carried out. When warmed with concentrated mineral acids best with hydro-bromic acid it gives a striking colour reaction. The liquid a t first faintly yellow passes through brownish-yellow reddish-brown to brown tints then becomes opaque purple and blue in thin layers, and ultimately deelp purple; on careful addition of alkali the colour becomes blue then faintly green or nearly colourless.Zingerone is but slightly volatile in steam. As it was not obtained in crystalline form until the research was otherwise com-pleted no attempt was ever made t o determine the composition of the product from gingerol by direct analysis. The formula was deduced from those of its solid and more easily purified derivatives. The substance dissolves only very sparingly in water but freely in dilute aqueous sodium or potassium hydroxides being reprecipi-tated by carbon dioxide. I n alcoholic solutions i t gives a green coloration with ferric chloride. I t s alkaline solutions give neutral, non-pungent insoluble products when treated with benzoyl or sulphonyl chlorides and with chlorocarbonic esters.These observa-tions show that zingerone has a phenolic character. It is optically inactive in alcohol or benzene. Zingerone also1 has the character of a ketone and readily yields a crystalline phelnylhydrazone (plates m. p. about 1 4 3 O ) and semi-carbazone (needles m. p. about 1 3 3 O ) ; but these as well as other hydrazones and oximes were found very difficult t o purify as they quickly decomposed in solution. The e t hy lcarb ona t o-deriva tive C,,H ,,O,*O*CO,Et was prepared by cautiously adding ethyl chloroformate t o an ice-cold solution of the ketone in aqueous sodium hydroxide. The resulting solid was collected dried and crystallised from ether 786 LAPWORTH PEARSON AND ROYLE : Found C=63.1 63.2; H-6-7 6.7. It formed large flat calcite-like prisms melting a t 45-47O.The methyl derivative (rnet<hylzingerone) C,,H,,O,*OMe was pre-pared by shaking an ice-cold solution of zingerone in aqueous sodium hydroxide with methyl sulphate. It is also obt’ained as has already been mentioned by heating o r hydrolysing ‘‘ methylgin-gerol.” It dissolves readily in most of the usual organic media with the exception of petroleum; it is insoluble in water or alkalis, and crystallises from alcohol in colourlees neledles melting a t 5 5 -5-5 6*Z0. CI4Hl8O requires C = 63.2 ; H = 6.7 per cent. Found C = 68.9 ; H = 7.7. C,,H,,O requires C = 69.2 ; H = 7.7 per cent. The methyl derivative has no phenolic properties but it gives the same colour reactions with hydrochloric acid as does zingerone itself.It displays ketonic characters gives crystalline derivatives with semicarbazide phenylhydrazines and hydroxylamine. It is fully saturated and does not reduce cold permanganate in acetone solution or decolorise bromine in chloroform. When oxidised with permanganate in dilute aqueous sulphuric acid i t gives veratric acid (m. p. 1 7 9 O ; proof by mixed melting point method). Jf et7~yZziii~ero~aeox~me C‘,,H,,O,*NOH was prepared by warming the foregoing compound in alcoholic solution with hydroxylamino hydrochloride and excess of sodium acetate. It was isolated by diluting with water and extracting with ether. It crystallises from light petroleum or ether in slender needles melting a t 91-92O. Found C=65*0; H=7.8; N=6*7. C,,H,,O,N requires C = 64.6 ; H = 7.6 ; N = 6.3 per cant.When dissolved in a little pyridine and ether treated with bromine water in excess and then with hydrogen peroxide (Piloty’s test f o r ketoximes) it gives a definite although not very intense yellowish-green colour (bromonitroso-compound) which passee into thel ethereal layer. Cons ti t u t i o n of Met h ylzingero n e and Zingerone. As the sensitive phenolic hydroxyl group of zingerone is absent in the methyl derivative the latter was considered to be the most suitable compound with which to begin experiments with t’he view of throwing light on the structure of these substances. Attempts t o bring about the Beckmann change in the oximel were not fruitful but the oxidation of methylzingerone itself gave the necessary clue. When the methyl ether was warmed with aqueous sodium hypo THE PUNGENT PRINCIPLE OF GINGER.PART I. 787 bromite the odour of broinoform won became perceptible and the solid passed into solution. A t the end of some hours’ interaction, sodium sulphite in excess was added to’ destroy unchanged hypo-bromite steam was passed through the solution t o remove carbon tetrachloride and other impurities then the acid products were liberated by addition of mineral aqid and extracted with ether. The latter extracted a mixture which on repeated recrystallisatioii from ether was resolved into two main portions one containing halogen and the other free from it. The latter formed plates and when recrystallised from water was found t o be hydrated but after exposure in a vacuum became anhydrous and melted a t 95-97O: 0.6573 Gram required 31.5 C.C.N / 10-NaOH for neutralisation with phenolphthalein as indicator whence the equivalent = 208 ; that required for a monobasic acid C,,H,,O,*CO,H being 210. The acid agreed in general characters with B-3 4-dimethoxy-OM6 OMe phenylpropionic acid (I) and this inference appeared to be more probable as gingerol had yielded protocatechuic acid and methyl-gingerol veratric acid (11). It was clear that methyl zingerone must be the methyl ketone of the acid or C,,H,,O,*CO~CH,, and therefore probably OMe Me0 /-\C ET;CH,-CO*C K3. Synthetic experiments described in Part II. have established the truth of this inference. Moreover a synthesis of zingerone itself, which is described in the same communication has shown that of the two alternative formula deducible from that of the methyl ether the formula \-/ OMo HO/ ‘ClI,*CH2*CO*CH,.must be assigned t o zingerone; the latter is therefore 4-liydroxy-3-met hox yphenylethy 1 me thy1 ketone. \-/ The Coiistitution of Ginyerol. There is no reason t o doubt that except for the numbers of carbon atoms contained in the residues which furnish the fatty aldehydes on hydrolysis the structure of the two or more pungent constituents of gingerol are essentially similar and therefore 788 LAPWORTH PEARSON AND ROmE : mutatis mzczandis the following remarks are applicable t o all such constituents. The molecule of that component of gingerol which furnishes heptaldehyde clearly has the composition of a compound of that aldehyde with zingerone that is C,H,,O + C,,H,,(OMe) *OH, methylgingerol being C,H,O + C,,,H,,(OMe),.The ketonic charac-ter of these two compounds is not a matter reasonably admitting of doubt but the function of the fourth oxygen atom is ce'rtainly not quite clear. It is eatre8mely difficult if not impossible t o depict a condensa-tion product of one molecule of heptaldehyde with one of methyl-zingerone which accords with the above conclusions without assum-ing that thO fourth oxygen atom has hydroxylic functions. The only reasonable alternative would appear t o involve one of the following assumptions (1) that the carbonyl groups in the ketone and the aldehyde are jointply engaged in the grouping but this is not in accordance with the ready hydrolysis of gingerol and methylgingerol with baryta water as such complexes are normally stable t o alkalis; or (2) that the molecule contains a peroxide-like grouping such as : 0-0 O- 0 .I t or I I , but the two substances have none of the characters of peroxides. Again neither of these assumptions is in harmony with th0 fact that methylgingerol reacts with hydroxylamine giving a ketoxime which yields unchanged methylgingerol once more by mild hydro-lysis. The authors have been forced to infer therefore that the fourth oxygen atom has hydroxylic functions and that the failure of methylgingerol to give methane with magnesium methyl iodide is due to the primary reaction of the reagent with the carbonyl group and the insolubility of the additive product; the reaction of methyl-gingerol with acetyl chloride and more especially with phenyl-carbamide are fully in agree'ment with this view.I f the presence of the hydroxyl group in methylgingerol is assumed then this compound and gingerol itself appear as " aldols," and tho whole of their properties including the optical activity with one signal difficulty are readily understood and the pungent constituents of gingerol can be represented by one of the two general formulze : -C c! H,* C- -(j-c- 9 I I I THE PUNGENT PRINCIPLE OF GINGER. PART r. 789 CH,*CH,*CO*CH,*CH(OH)*(CK~]~*CH~ /\ ()OMe OH or where the residue CzHm+l is a saturated alkyl radicle probably normal in all cases and n=5 and either 4 or 3 (or both). Having regard to the very general occurrence in natural products of open, straight-chain aliphatic residues the firsti of the two general formulze appears the more probable.The optical activity of the compounds is explicable on the basis of either of the formulz suggested. Summary. The oleo-resin gingerol the pungent principle of Zingib er oficinaie which was first investigated by Thresh and more recently by Garnett and Grier is essentially a mixture of optically active saturated phenolic compounds derived from a residue of zingerone, or 4-hydroxy-3-methoxyphenylethyl methyl ketone in association with a molecular proportion of the residue of a saturated aliphatic aldehyde which in the main constituent is n-heptaldehyde. The constituents are probiably '' aldols " (P-hydroxy-ketones) of the general type : C,H,(OH)(OMe)*CH,*CH,-CO*CH,-CH(OH)*[CH2],*CH,, where n in the principal constituent is 5 and 4 or possibly 3 in the main secondary constituent. Only traces of solid matter of camphoraceous appearance have been observed in gingerol but methylgingerol a mixture of the monomethyl ethers of the gingerol constituents and methylgingerol oxime has been obtained in crystalline form. The work described in this paper and in Part 11. was begun early in 1914 and carried to its present stage in July 1915. We desire to express our very cordial thanks t o Messrs. Garnett and Grier for entrusting us with the continuance of their investiga-tions and for the considerable trouble they took t o facilitate its progress. Grateful acknowledgments are also due t o James Woolley and Sons Mancbester for carrying out the large-scal 790 LAPWORTH AND WYKES: extraction of th0 drug and to tlie British Pharmaceutical Confer-ence for the! use of a grant from which part of the expenses of the research was defrayed. ORGANIC CHENZICAL LABORATORIES, THE UNIVERSITY MANCHESIER. [Recsived April 27th 1917.

ISSN:0368-1645    

DOI:10.1039/CT9171100777

出版商:RSC    

年代:1917

数据来源: RSC

摘要:

790 LAPWORTH AND WYKES: LXVI.-Tiie Pungent Pyinciples of Ginger. Part Synthetic Prepcwatz’ons of Zingeq-one Methyl- II. zingerone and Some Related Acids. By ARTHUR LAPWORTH and FREDERICK HENRY WYKES. IN Part I. evidence was adduced that the phenolic ketone, (‘ zingerone,” obtained from gingerol the pungent oleo-resin of ginger was 4-hydroxy-3-methoxyphenylethyl methyl ketone (I) or a position isomeride and that ‘( methylzingerone,” the correspond-ing ketone from “methylgingerol,” is its methyl ether (11) 3 4-dimethoxyphenylethyl methyl ketone. CH2* OH,* CO-CH CH,*CH,*CO*CH, /\ /\ (1.1 (11.1 The authors have prepared these and also several important acids‘ that were required for purposes of comparison by synthetic methods which in certain cases represent the first direct syntheses of the compounds in question and in other cases are simpler or give better yields than previous methods.The synthesis of methylzingerone presented no difficulty as 3 4-dimethoxystyryl methyl ketone (11) from veratraldehyde and acetone is easily niade in good yield and is readily reduced by sodium amalgam giving 3 4-dimethoxyphenylethyl methyl ketone, which is identical with methylzingerone (11). CH:CH*CO*CH ciio C H:CH* CO-CH, /\ /\ /\ I ‘ ! ,!OMe ( / 0 1 \ l e OH \/oale OH OMe (111.) (IV. 1 (V.) A similar process for preparing 4-hydroxy-3-methoxyphenylethyl methyl ketone by condensing vanillin (IV) with acetone and reducing the resulting 4-hydroxy-3-methoxystyryl methyl ketone (V) gave very poor yields a t each stage but the final product wa THE PUNGENT PRINUIPLES O F GINGER.PART 11. 791 identical with zingerone. On the other hand good yields were quickly obtained by the following process. Vanillin (V) gives an excellent yield of ethyl vanillylideneaceto-acetate (VI) by Knoevenagel and Albert’s method (Ber. 1904, 37 4476); this unsaturated ester was easily reduced by means of sodium amalgam and the product with excess of alkali was con-verted into an acid doubtless vanillylacetoacetic acid (VII) which when heated lost the elements cf carbon dioxide and was con-verted into zingerone (V). CH :CAc:CO,Et CH,*CHAc*CO,H CH:CAc, /\ 1 , /\ /’\ <)OM0 ( ) 0 Mi3 \/io&le OH OH OH (VI. 1 (VII.) (VIII.) Synthetic zingerone was obtained in crystalline form and had the characteristic sweet odour and pungent taste of the ketone from the decomposition of gingerol.The solid when used to infect samples of ketone obtained from the drug caused these to solidify f o r the first time. The properties of the synthetic ketone were identical in every respect with those of the compound from ginger. An attempt to imitate this synthesis through vanillylideue-acetylacetone (VIII) was not successful and the reduction pro-ducts of the latter compound were not of the type expected. The authors’ work included some simple direct syntheses of hydroferulic acid (XI) and of hydrocaffeic acid (XIII). These have previously been obtained by the reduction of synthetic ferulic and caffeic acids (Tiemann and Nagai Ber. 1878 11 650 672) prepared by the Perkin “ cinnamic acid synthesis.” As the author&’ methods although obvious enough are new if not in principle and are very easily carried out the steps may be indicated.Vanillin (V) was condensed with diethyl malonate or ethyl cyanoacetate yielding diethyl vanillylidenemalonate and ethyl v anill ylidenec y ano ace t a t e (IX) respectively. These were readily reduced a t the double bond and on subsequent hydrolysis with excess of alkali were converted into acids doubtless vanillylmaloaic acid (X) in both instances. The product when heated yielded hydroferulic acid (XI). CH C( CN) *CO,Eb CH2*CH(C0,H) CH,*CH,* CO,H /\ /\ /\ (S.) I ‘ \,/OM‘ OH (XT.) I 792 LAPWORTH AND WYKES A precisely analogous process applied to protocatechualdehyde instead of vanillin gave hydrocaffeic acid (XIII) through the inter-mediate condensation product (XII).CH:C(CN)*CO,Et CH ,*CH,*CO,H /\ /\ !,)OH OH (XII.) OH (XIII. ) It is worthy of remark that all the unsaturated phenolic ketones and esters (I V VIII IX and XII) give solutions in alkali which in the thinnest layers exhibit an intense yellow colour that dis-appears on reduction a t the double bond. The effect is perhaps most pronounced with (XII) which exhibits with boric acid a reaction very like that associated with curcumin the colouring matter of turmeric. The constitution which Kostanecki suggests for curcumin is (IH:CH*CO* CH,-CO*CH CH /\ in which case it is obviously very closely relat-ed to the similar compounds dealt with in the present paper.The question of the groupings essential t o the pungency of gingerol zingerone and similar compounds is one which one of us is hopeful of reserving for a short time. So far it would appear certain that the presence of the free phenolic hydroxyl group is essential and also not. improbably the ketonic carbonyl suitably disposed in a saturated chain att*ached t o the phenolic residue. E x P E R I M E N T A L . I . S p t h e s i s of Zingerone (4-Eydroxy-3-methoxyp~~enylethyl Methyl Ketone). Preparation of Ethylcarborntovanillin, OM0 -CO,Et*O/ \CHO \-/ As vanillin and its acetyl and benzoyl derivatives did not readily condense with acetone the ethylcarbonato-derivative which is new, was prepared by adding one molecular proportion of ethyl chloro-formate to vanillin dissolved in the requisite quantity of AT-sodiur THE PUNGENT PRINCIPLES OP GINGER.PART 11. $93 hydroxide. The derivative separated in the cold in small white needles and after an hour was collected and crystallised from hot alcohol. Found C = 63.3 ; H = 5.9. CI1Hl2O4 requires C = 63.5 ; H = 5.8 per cent. The substance crystallises from alcohol in slender needles and melts a t 65O. When it was dissolved in excess of acetone and the mixture treated with a little dilute sodium hydroxide an intense yellow colour was pro-duced and apparently a small quantity of the desired condensa-tion product was formed as the neutral product of the reaction, when reduced with sodium amalgam gave an oily mixture which had a pronounced pungent taste. The authors have not yet had opportunities to follow up these observations.It is slowly hydrolysed by cold dilute alkali. R eductisni of Vanillylidene Derivatives of A cet ylacet one a nd E t h y l Acetoacetate. Formatioit of Zingerone. Vanillylideneacetylacetone which was prepared by Knoevenagel and Albert's method (Ber. 1904 37 4480) forms an intensely yellow solution in sodium hydroxide; this colour is discharged by shaking the solutioii with sodium amalgam and on saturating the resulting liquid with carbon dioxide an oil doubtless vanillyl-acetylacetone HO*C,H3(0Me)*CH2*CH(CO*CII,), is deposited and ultimately tends to crystallise; as the oil when heated with acid or alkaline hydrolytic agents gave no product with the characters of vanillylacetone its further investigation was not undertaken.Ethyl vanillylideneacetoacetate which was also obtained in nearly theoretical yield by Knoevenagel and Albert's method (luc cit. p. 4476) crystallised in pale yellow needles melting a t 112*5-113*5° and as Knoevenagel and Albert give the melting point as 120-121° the author? analysed their product. (Found, C = 63.3 ; H = 6.0. The ester was dissolved in 10 per cent. aqueous sodium hydroxide and the intensely yellow solution shaken violently with washed, fluid sodium amalgam the whole being kept very cool. When the solution no longer displayed a yellow colour in thin layers it was separated from mercury mixed with 35 per cent. of its weight of solid sodium hydroxide and heated for about eight hours on the water-bath when it was coslsd saturated with carbon dioxide, and extracted with ether to remove unhydrolysed ethyl vanillyl-acetoacetate OMe*C,H3(OH)*CEf,~CHAc*C0,Et an oily com-pound which represents the first stage in the reduction process.CI4Hl6O5 requires C =.63*5 ; H = 6.1 per cent.) 11 794 LAPWORTH AND WYKES: The aqueous residue was next acidified and extracted with ether, the latter then being dried and evaporated. The oily extract was belated in a vacuum when a t first carbon dioxide was evolved, and the residual material distilled a t 175-210°/15 mm. I n order to separate the products the distillate was dissolved in aqueous sodium hydroxide which was then saturated with carbon dioxide and extracted with ether (“ phenolic extract ”), the aqueous residue being subsequently acidified with hydrochloric acid and again extracted with ether (“ acidic extract ”).The “phenolic extract” when dried and evaporated left a brown oil. This was dissolved in N-sodium hydroxide solution and treated in the cold with ethyl chloroformate when the bulk of the material in solution was converted into an oil which solidified on scratching the vessel with a glass rod. The solid was dissolved in ether thO solution dried and allowed to evaporate spontaneously. Large tabular crystals were deposited which on recrystallisation from light petroleum formed colourless hex-agonal plates melting a t 47.50. Found C = 62.7 ; H = 6.7. CI4H,,O requires C = 63.1 ; H = 6.8 per cent. The substance was identical in all respects with the ethyl-carbonate-derivative of the ketone (“ zingerone ”) obtained from gingerol.I n order to obtain the free phenolic ketone the foregoing com-pound was heated on the water-bath with dilute aqueous sodium hydroxide until a homogeneous liquid resulted. Excess of hydro-chloric acid was then added and the cooled product extracted with ether. After drying and evaporating the ethereal extract left a residue which was distilled in a vacuum when the dis-tillate set to a crystalline mass. The solid material obtained as above was purified by dissolv-ing it in dry ether adding enough petroleum t o cause a turbidity, allowing the latter t o settle and then infecting the clear solution with a trace of solid 4-hydroxy-3-methoxyphenylethyl methyl ketone. On spontaneous evaporation the liquid deposited lustrous, flat colourless crystals which had the odour of the above ketone and melted a t 36-37O.Found C = 67.7 ; H = 7.2. C,,H,,O requires C = 68.0 ; H,= 7.3 per cent. 4-Hydroxy-3-methoxyphenylethyl methyl ketone obtained in this way had an extremely pungent taste and was in every resEect identical in properties with “ zingerone.” As has already been mentioned the ‘* zingerone ” from the natural ~ource was no THE PUNGENT PRINCIPLES OF GINGER. PART 11. 795 obtained in crystalline condition until infected with a trace of the synthetic ketone which had solidified spontaneously after dis-tillation in a vacuum. 4-Hydroxy-3-methoxyphenylethyl methyl ketone is not the only product which is formed by reduction of ethyl vanillylideneaceto-acetate and subsequent hydrolysis.The " acidic extract " (com-pare p. 794) gave on evaporation a small quantity of an acid which was moderately soluble in cold water and readily so in hot; this formed colourless leaflets melting a t 133-134O and was a t first believed t o be hydrocaffeic acid which also crystallises in leaflets and melts a t 137-139O. The substance obtained as above, however unlike hydrocaff eic acid gives no coloration with ferric chloride and when it is mixed with hydrocaffeic acid the mixture melts a t 127-132O. A titration with alkali using phenol-phthalein as indicator gave an equivalent' for thi4 acid of about 240; the quantity of this material obtained in the pure state was, however too small to permit of further investigation. Oxidation and Reduction of 3 4-Dimethoxystyryl Methyl Retone.Formation of Dimethylcaffeic Acid and of Methylzingerone. 3 4-Dimethoxystyryl methyl ketone was prepared by condensing veratraldehyde with acetone (compare Francesconi and Cusmano, Gazzetta 1908 38 ii 70 et sep.) and purifying it by recrystallisa-tion from light petroleum. C,,H,,O, requires C=69*9; H=6*8 per cent.) The compound crystallises from carbon tetrachloride in micro-scopic leaflets. When it is warmed with concentrated hydro-chloric acid it gives a deep red coloration doubtless due to the intermediate formation of veratraldehyde which gives a similar reaction. 3 4-Dimethoxystyryl methyl ketone is readily oxidised when shaken with aqueous sodium hypobromite being converted into dimethylcaffeic acid C,H,(OMe),*CH:CH-CO,H which was isolated in small flat needles (from water) rnelting a t 180-181O.Reduction of 3 ddimethoxystyryl methyl ketone with the aid of Paal and Skita's or Willstatter's methods did not proceed in a very satisfactory manner but when an alcoholic solution of the compound was shaken with liquid sodium amalgam the bright yellow colour rapidly lost its intensity and when this process was carried out in presence of excess of potassium hydrogen carbonate, good yields of the desired reduction product were obtained. It was isolated by diluting the aqueousalcoholic solution with wat'er, extracting with ether and shaking the ethereal extract with (Found C = 70.0 ; H = 6.8. I I* 796 LAPWORTH AND WYKES: freshly prepared sodium hydrogen sulphite solution.The solid hydrogen sulphite compound was collected washed with ether, dried and then decomposed by warming it with excess of aqueous sodium carbonate. By extracting the resulting liquid with ether, 3 4-dimethoxyphenylethyl methyl ketone was obtained in quantity corresponding with about 80 per cent. of that theoretically possible. It was purified by crystallisation from methyl alcohol. Found C = 69.4 ; H = 7.8. C,,H,,O requires C = 69.2 ; H = 7.7 per cent. The substance formed colourless odourless needles melting at 55-56O and its identity with “ methylzingerone,” obtained by methylating “ zingerone ” or by the decomposition of ‘‘ methyl-gingerol,” was established by the usual methods. Its colour reactions with hydrochloric acid and the properties of its oxirne (long white needles m.p. 920) were indistinguishable from those of the corresponding compounds obtained from ‘‘ gingerol.” 111. Synthesis of Hydroferulic Acid. E t h y 1 Vanill ylidenec yanoac e t at e, HO( )CH:C(C N)-C0,Et . Vanillin and ethyl cyanoacetate were mixed in molecdar pro-portions and heated on the water-bath until a homogeneous liquid was obtained when a few drops of piperidine were added and the heating was continued until a test portion solidified com-pletely. Alcohol (twice the weight of vanillin present) was added, and the whole allowed to cool. The crystals obtained were washed with dilute hydrochloric acid dried and crystallised from alcohol. With material obtained from mother liquors the total yield of condensation product approached that theoretically possible.OMe .~ ___ Found C = 62*2 ; H,= 5.3. Ethyl vanillylidenecyanoacetate separates from alcohol in yellow needles melting a t 107O. It dissolves in aqueous sodium hydroxide giving a solution which shows an intensely yellow colour even in thin layers; it is reprecipitated unchanged from this solution on the additioa of acids. C,H,,O,N requires C = 63.1 ; H = 5.3 per cent. Formation of Hydroferulic Acid from Ethyl Vanillylideize-cyanoacetate. The reduction of ethyl vanillylidenecyanoacetate was accom-plished by means of sodium amalgam in precisely the same manne THE PUNGENT PRINCIPLES OE GINGER. PART 11. 797 as was used for the reduction of ethyl vanillylideneacetoacetate (p. 793). The reduction product [mainly no doubt a mixture of vanillylcyanoacetic acid HO*C,H,(OMe)*CH,*CH (CN) *CO,H , and its ethyl ester] was heated with excess of potassium hydr-oxide until the ester present had been completely hydrolysed when excess of acid was added and the liquid extracted with ether.The latter was evaporated and the residue heated in a vacuum to expel carbon dioxide; the hydroferulic acid left was purified by converting it into its lead salt which is very sparingly soluble, in water and decomposing the latter in aqueous suspension with hydrogen sulphide. Hydroferulic acid obtained in the above way crystallises from hot water in stout white needles melting ah 89-90°. It dis-solves readily in hot water less readily in cold and only spar-ingly in concentrated hydrochloric acid.Its aqueous solution gives no coloration with ferric chloride. (Found C = 60.6 ; H=6*2. C,,H120 requires C=60*6; H=6*1 per cent.) The acid agreed very closely in properties with those assigned by Tiemann and Nagai (Ber. 1878 11 650) to the acid obtained by the reduction of ferulic acid. Diethyl vanillylidenemalonate HO*C,H,(OMe)*@H:C(CO,Et),, was prepared from vanillin and ethyl malonate by Knoevenagel and Albert’s method (Zoc. cit. p. 4481). It had the properties ascribed to i t by these authors. On reduction with sodium amalgam and subsequent treatment in the manner described in the case of ethyl vanillylidenecyane acetate i t yields hydroferulic acid. Iv. Synthesis of Hydrocaffeic A c i d . Hydrocaffeic acid was prepared by Tiemann and Nagai (Ber., The 1878 11 672) by the reduction of synthetic caffeic acid.following is a Con,densation simple alternative synt,hesis. of Protocatechualdehyde with Ethyl Cyanoacetate. Formation of Ethyl a-CyanocafJeat e, OH HO’ \CH:C(CN)*CO,Et. \-/ This condensation was effected by means very similar to those used in similar condensations in previous sections but it was found desirable to dilute the mixture of aldehyde and ester with a little absolute alcohol. The product which was contaminate 798 THE PUNGENT PRINCIPLES OF GINGER. PART 11. with a brown impurity was purified by extraction with benzene and recrystallisation therefrom. Found C= 61.6 ; H =4*9. CI2H,,0,N requires C = 61.8 ; H = 4.7 per cent. Ethyl a-cyanocaffeate is a yellow microcrystalline solid which melte somewhat indefinitely a t 162-166O and dissolves in alkali to give an intensely orange solution.A test-paper made by dipping paper in an alcoholic solution of this ester behaves towards boric acid in much the same way as does turmeric paper, that is if moistened with boric acid solution and then heated it turns brown but the temperature required is somewhat higher than with turmeric. Formation of Hyd.rocaffeic Acid from Eth.yZ a-Cyanocafleate. A solution of ethyl a-cyanocaffeate in alkali was reduced with sodium amalgam until colourless. The resulting liquid which very quickly turned brown on exposure to air was made strongly alkaline by the addition of solid sodium hydroxide boiled for two days under a reflux condenser cooled acidified and extracted with ether. The ethereal extract gave an oil which soon deposited crystals; these were not isolated but the whole was heated t o expel carbon dioxide then dissolved in water treated with animal charcoal and allowed to crystallise. The crystals of hydrocaffeic acid obtained in this way were hexagonal leaflets melting a t 138-139O were moderately soluble in cold water readily so in hot and their aqueous solution gave a green colour with ferric chloride changing t o a rich purple on the addition of ammonia. These properties correspond closely with those assigned to hydro-caffeic acid by Tiemann and Nagai (Zoc. cit.). The authors have also found that hydrocaffeic acid is readily obtained from hydroferulic acid by heating i t with dilute hydro-chloric acid (about 5 per cent.) a t 200° for six hours in a closed tube. Acknowlsdgmente are due to Mrs. L. Kletz Pearson who kindly carried out some preliminary experiments on the preparation of vanillin and verat.raldehyde derivatives. ORGANIC CHEMICAL LABORATORIES, THE UNIVERSITY MANCHESTER. [Received April 2 7th 19 1 7.

ISSN:0368-1645    

DOI:10.1039/CT9171100790

出版商:RSC    

年代:1917

数据来源: RSC

摘要:

USHER AND RAO THE DETERMINATION OF OZONE ETC. 799 LXVI1.-The Determination of Ozone and Oxides of Nitrogen in the Atmosphere. By FRANCIS LAWRY USHER and BASRUR SANJIVA RAO. THE investigation recorded in this paper arose from the observation, common to residents in tropical climates that rubber articles and cotton and silk fabrics pei-ish and certain colouring matters are bleached far more rapidly than in temperate latitudes. The explanation which a t first seems natural namely that increased temperature or increased intensity of light o r both of these may be chiefly responsible for the observed effects appears to be inadequate because although the factors named undoubtedly contribute to the deterioration of certain articles exposed to their influence this deterioration is well marked in the case of articles kept in compara-tive darkness.An alternative hypothesis is that the atmosphere in the tropics contains some chemically destructive substance which is either absent from or present in far smaller quantity in the atmo-sphere of higher latitudes. This hypothesis obviously suggests ozone as the substance postulated and it receives support from the well-known observations that tropical sunlight exceeds both in quantity and in actinic power that of temperate climates and that ozone is a product of the action of ultra-violet light on oxygen. We have been unable to discover any satisfactory record of experiments on the amount of ozone in the atmcrjphere in the tropics and since the question ia one of considerable practical importance as well as of scientific interest the problem has been attacked experimentally in Bangalore.The experiments so far carried out do not support the hypothesis but a t present they are too few to permit any definite conclusion being drawn. The object of the present paper is t o show that the methods of estimation hitherto practised are probably untrustworthy and t o describe a new method which we believe t o be free from their defects. This method is applicable to the esti-mation of ozone nitrogen peroxide and in certain cases of hydro-gen peroxide. Previous Work. Much work has been done on the methods of estimating ozone and on the application of those methods to the analysis of the atmo-sphere; a useful summary of it is given in a paper on the subject by Hayhurst and Pring (T.1910 97 868). Besides the work cited in that paper we may mention that of Keiser and McMaster (Amer. Chem. J. 1908 39 96) who used a solution of permanganate to distinguish between ozone hydrogen peroxide and nitrogen per 800 USHER AND RAO THE DETERMINATION OF OZONE oxide. According t o these authors air when passe'd through per-manganate solution is deprived of hydrogen peroxide' and nitrogen peroxide and any ozone which i t may contain being unaffected by the permanganate can be estimated by one of the usual methods. On the other hand if the air is passed over powdered manganese dioxide both ozone and hydrogen peroxide are destroyed and the nitrogen peroxide may be estimated. Finally passage through chromic acid serves t o distinguish betwe'en ozone and hydrogen peroxide of which i t destroys only the latter.Rothmund and Burgstaller (Monntsh. 1913 34 7 5 ) showed thatq, in the estimation of ozone and hydrogen peroxide potassium iodide is untrustworthy on account of secondary reactions but pointed out that satisfactory results could be obtained by substituting the bromide' in acid solution f o r the iodide and by subsequently adding iodide and titrating the liberated iodine. This process is however, unsuited to the estimation of ozone in exceedingly small concen-trations. Value of Previous Determiiiatioiis of Ozone. Before describing a new method i t is desirable! to state the grounds on which the older methods are held to be objectionable. I n the absence of a generally accepted standard method the objec-tions are necemarily incapable! of direct proof but are none the less valid.I n some instances the results obtained by different methods under the same conditions of time and place are mutually contra-dictory but usually no such check is available and in that case the experiments may be open to criticism on physical and chemical grounds. Objections on physical grounds are best exemplified in the method adopted by Hayhurst and Pring (Zoc. cit.) and subse-quently used in a slightly modified form by Pring (Proc. Roy. SOC., 1914 [ A ] 90 204). Here the air to be examined was blown across the surface of a liquid reagent contained in a shallow vessel the efficiency of the arrangement having been tested by passing the gases from an ozoniser mixed with excess of air through two such vessels a t the rate of 6 litres per minute and considered satisfactory because "all the ozone was taken up in the first vessel and no appreciable liberation of iodine occurred in the second." "'he only conclusion that can justifiably be drawn from this experiment is that the greater part of the ozone in a mixture which was presum-ably far richer in that gas than is atmospheric air was absorbed in the first vessel whereas the only important question how much ozone escaped absorption is left unanswereld.Chemists who have been concerned with the removal of minute traces of a gas by a liquid absorbent will have difficulty in assuming that ozone presen AND OXIDES OF NITROGEN IN THE ATMOSPHERE. 801 to the maximum extent of say I part in 200,000 parts of air can have been quantitatively absorbed under such conditions for i t is the last trace6 of the absorbable gas that’ are so difficult to remove from a mixture.The foregoing criticism is probably applicable, although not in so marked a degree to all estimations of ozone where the current of air has not been finely subdivided either by passagel over a closely packed solid reagent or mechanically before passage through a liquid reagent. Unless the error introduced by inefficient absorption is large compared with the quantity to be measured i t is important only in those experiments which claim to have a quantitative significance ; whereas if they are objectionable on chemical grounds even their qualitative value may be questioned.We venture to suggest that all the recorded determinations based on the direct use of potassium iodide are untrustworthy for the following reasons in addition to those stated by Rothrnund and Burgstaller (Zoc. cit.). (1) The liberated iodine is appreciably volatile in the current of air even when fairly concentrated potassium iodide is used; (2) when iodate is formed the reaction between the iodic and hydriodic acids which occurs when the liquid is acidified is very slow a t great dilutions; (3) any nitrous acid present is reduced to nitric oxide which then combines with dissolved oxygen and furnishes more iodine; (4) solu-ti’ons of potassium iodide are unstable in the presence of air even in the dark. It is true that the difficulties under the first three heads might be surmounted but the oxidation of a neutral iodide solution by air in the dark is bound to occur unless some suitable negative catalyst can be found and must vitiate the results of any experiment in which a long time elapses between its commencement and the final titration.Most previous workers who have examined the stability of potassium iodide solutions agree that iodine is liberated by the combined action of air and light and that in the presence of acidsnven of 50 weak an acid as carbonic-oxidation may take! place in darkness but it has been found that carefully purified iodide * also furnish- iodine in the dark as is seen from the following figures (p. 802) which relate t o solutions kept in darkness for two days in the presence of air freed from oxidising impurities.The error introduced in this way is certainly small if an experi-ment is of short duration but in some of those camed out by Hayhurst and Pring nearly three weeks elapsed before the solution was titrated. Even when determinations are carried through * Ordinary “pure” potassium iodide appears always to contain free alkali as well as iodate and it is possible that the presence of the former accounts for the negative results obtained by some. The iodide used in our own experiments was treated with aluminium amalgam to reduce the iodate, and was recrystallised from alcohol to remove the alkali 802 USHER AND RAO THE DETERMINATION OF OZONE C.C. N/lOOO-thiosulphate equivalent to iodine liberated - Experi- as free &S I.Purified air containing carbon dioxide ......... 1.55 nil. ment. Solution in contact with iodine. iodate. 11. Purified air free from carbon dioxide ......... 1-30 0.4 111. Same as in 11. ; more concentrated solution.. . 1-40 0- 7 rapidly however the unavoidable errors due to secondary reactions, demonstrated by Rothmund and Burgstaller remain. The substi-tution of cadmium potassium iodide for potassium iodide recom-mended by Baskerville and Crozier (J. ,4mer. C‘hem. Soc. 1912, 34 1332) does not seem likely to eliminate those errors the only advantage claimed for the cadmium salt being that i t is more stable to light and to certain impurities. The method proposed by Keiser and McMaster (Zoc. cit.) is open to several objections. In the first place ozone will not pass un-changed through permanganate s’olution unless nitrogen peroxide and hydrogen peroxide are absent for both these substances reduce the permanganate to a manganous salt which is known to be attacked by ozone.Moreover the speed of the reaction between nitrous acid and 02~0110 in aqueous solution is comparable with that of the reaction between nitrous acid and pemanganate so that a conside8rablel proportion of the ozone must be destroyed during its passage through the solution and the amount of nitrous acid found must be too low. Finally the actual estimation of ozone was carried out by the potassium iodide method. E x P E R I M E N T A L . The initial difficulty lay in securing the quantitative removal of the very minute1 traces of ozone and oxides of nitrogen present in air f o r which purpose probab’ly the only effective apparatus is that devised by Reiset and described in Hempel’s “Methods of Gas Analysis.” By using this contrivance it would have been possible to deal with velry large quantities of air and to attain a corre-sponding accuracy; but there is one serious objection to such a procedure namely the risk of decomposing the ozone catalytically during its passage through the fine perforations in the platinum disks.This risk would probably be serious even if glass disks were used in place of platinum f o r it is known that ozone is decomposed by contact with broke? glass. The alternative method is to shake the reagent with a known volume of air in a closed vessel but if this is done the maximum quantity of air that can be convenientl AND OXIDES OF NITROGEN IN THE ATMOSPHERE.SO3 dealt with is about 7 litres which may be expected to contain not more than 0-02 inilligram of ozone. This is just ten times the quantity that can be detected by any process involving a titration of iodine and since no other tit'ration process is so sensitive i t was necessary either to sacrifice accuracy o r to employ a more sensitive method. After prolonged trial of various processes-chiefly modi-fications of Rothmund and Burgstaller's-we1 finally adopted the one which will now be described. Its trustworthiness and accuracy will be discussed later. I n principle the method is extremely simple and depends on the reaction between ozone and alkali nitrite in aqueous solution a reaction which we have found to take place quantitatively accord-ing to the equation: 0 +- NaNO = 0 + NaNO,.Two samples of air are taken and collected in 7-litre stoppered bottles. One sample is admitted through two tubes containing respectively chromic acid and powdered manganese dioxide and the other through a tube containing chromic acid only. The samples thus oollected are shaken with a dilute standard solution of sodium nitrite made slightly alkaline and the nitrite content of the bottles is subsequently determined colorimetrically by the Griess-Iloavay method (production of red dye with a-naphthylamine and sulphanilic acid ; see Sutton's " Volumetric Analysis," 9th ed., p. 449). The first sample of air contains only nitrogen peroxide, the ozone and hydrogen peroxide having been destroyed and the increase in the quantity of nitrite in the bottle is equivalent t o the nitrogen peroxide absorb'ed.The second sample contains o'mne and nitrogen peroxide and the difference between the quantities of nitrite in the two bottles after shaking is equivalent to the ozone present. The particulars of the procedure a t present adopted are as follows and since the SUCCMS of the method depends on the care observed in attention to small details these will be dealt with rather The vessels used for collecting the air are 5-kilogram " ammonia " bottles with glass stoppers and have a capacity of about 7 litres. They are cleaned thoroughly with chromic acid mixture washed with purified water and afterwards left for several days full of ozonised air to remove all traces of oxidisable matter.The stoppers must fit accurateJy and be free from grease. The samples of air are collected over wster and it is necessary that the latter should be quite free from dissolved impurities particularly nitrous acid. We prepare a large quantity of distilled water of suitable purity and keep i t in a stock bottle protected from the) atmosphere, the same water being used repeatedly and tested a t frequent fully 804 USHER AND RAO THE DETERMINATION OF OZONE intervals. The manner of filling the bottles with air is important, the following method having been found the most satisfactory. The bottle (3) is first filled completely with purified water and a rubber stopper is then pushed into the neck. This stopper carries two tubes of which one ( R ) admits air whilst the other (C) serves as an e'xit for the water.The tube (C) is about 30 cm. long and of 3 mm. bore and projects only 1 or 2 mm. inside the stopper the other end being provided with a tap. ( B ) is of capillary bore and A B passes to the bottom of the bottle. Outside the stopper i t is sealed to the tube or tubes containing the chromic acid and manganese dioxide only sealed glass junctions are permissible in this system of tubes. On inverting the whole apparatus and opening the tap, the bottle fills with air the volume of which is equal to that of the water that escapes. The object of this arrangement is of course to avoid bubbling the air through water which might deprive it of part of the ozone and oxides of nitrogen.The tap is regulated so that air enters a t the rate of about 1 litre per minute AND OXIDES OF NITROGEN IN THE ATMOSPHERE. 805 The chromic acid tube has a bore of 1 cm. and may be either straight or U-shaped. It is filled with glass beads coated with purified solid chromic acid and for safety these beads should occupy 30 cm. of the tube. Chromic acid purchased as ‘‘ pure ” has been found to contain traces of nitric acid and contaminates air passed over i t with nitrogen peroxide. The reagent is therefore prepared by precipitating barium chromate washing this thor-oughly with water and then digesting i t with moderately dilute sulphuric acid free from nitric acid. The solution of chromic acid thus obtained is treated with sufficient baryta water t o remove any sulphuric acid and is ihen concentrated on a water-bath until i t becomes pasty.At this stage the glass beads are stirred into it, and are then introduced into the tube that is to be used and dried in a current of air a t looo. The manganese dioxide is mixed with asbestos wool that has been well washed and ignited and the mixture is packed into a tube which may safely be one-half the length of the chromic acid tube. It is necessary t o test the manganese dioxide since some specimens contain traces of alkali or of manganous oxide and may take up small amounts of nitrogen peroxide. When the samples of air have been collected the rubber stoppers are withdrawn and immediately replaced by the glass ones. The latter are then lifted just sufficiently to admit the point of a pipette and the following liquids are introduced (1) 25 C.C.of S /40,000-sodium nitrite made up with AT/ 1000-sodium hydroxide in place of water; (2) 100 C.C. of pure water. The water used in these experiments must of course; be proved free from nitrous acid or any other impurity that could affect the result and the sodium hydroxide must be prepared from metallic sodium and purified water since the solid as purchased always contains traces of nitrite. After the introduction of the liquid the stoppers are secured and the bottles shaken for a t least half an hour on a shaking machine. The contents of each separately are then made up to 250 C.C. These solutions are examined separately in the fol-lowing way. Fifty C.C.are mixed with 5 C.C. of the Griess-Ilosvay reagent in a small stoppered flask which is then kept in a water-bath a t 75O for ten minutes. A t the same time 50 C.C. of a standard sodium nitrite solution (N/400,000) is treated in exactly the same way. The solutions are then placed in the observation tubes of a Duboscq colorimeter and thedr strengths compared. Readings are taken for eight different positions of the tubes. A simple formula can be obtained for calculating the results: let V=c.c. of air collected through chromic acid, r=ratio of length of column of standard solution t o that of column of experimental solution 806 USHER AND RAO THE DETERMINATION OF OZONE Then the' volume concentration of nitrogen peroxide will be VI=c.c. of air collected through chromic acid and manganese rC=ratio of length of column of standard soluti'on to that of the volume' concentration of ozone will be 1 in 1/0.007(1./V-r ' / V ' ) of air assuming that nitrite solutions of the strengths given above are used.It is of course possible t o estimate hydrogen peroxide in addi-tion to ozone and nitrogen pearoxide' by collecting a third sample of air which has not been passed over any reagent and treating i t in exactly the same way as the other samples,'K but probably this determination would be unsatisfactory in any place where the air was not exceptionally " pure," especially if traces of sulphur dioxids or hydrogen sulphide were present. 1 in V/O*014(r-l) of air. dioxide, column of experimental solution, And if Critical Elm?niizntion of the Foreyoiug Method.Since it is claimed that the process just described is more trust-worthy than those previously used for the determination of ozone and oxides of nitrogen in the1 atmosphere and should replace them, i t will be well to state the grounds on which that claim is based. To give full experimental details would make this paper inordinately long; we shall therefore mention only the results of a number of experiments which have been made to test various possible sources of elrror. I n the first place the reaction on which the method is based is complete even at very great dilutions. This was proved by shaking a small quantity of exceedingly dilute nitrite solution for half an hour with air containing in one case a slight excess of ozone and in the other a slight deficiency; it was impossible1 afterwards to detect nitrite in the first or ozone in the second.The nature of the reaction could not be investigated a t very great dilution owing to the absence of a standard method by which to check the estima-tion of ozone but it was elucidated in the case of ozone concentra-tions between 0.01 and 0.08 per cent. by the application of the potassium bromide method. Two bottles we're filled simultaneously, through a T-tube with feeQly ozonised air and the percentage1 of * It has been found that hydrogen peroxide does not oxidise nitrite in alkaline or neutral solution but does so rapidly in acid solution. In esti-mating hydrogen peroxide i t is therefore necessary to acidify the liquid. This must be done just before addition of the Griess-Ilosvay reagent AND OXIDES OF NITROGEN IN THE ATMOSPHERE.807 ozone in it was determined by the nitrite and bromide methods respectively. Four experiments gave the following results : I. 11. 111. IV. Nitrite method ...... . 0.0263 0.0136 0.0124 0.0802 per cent. Bromide method ... 0.0254 0.0127 0,0124 0.0800 ,, The calculations were made on the assumption that one molecule of ozone reacted with one) of sodium nitrite and the satisfactory agreement shown by the figures proves that assumption to be justi-fied. (1) When purified air is shaken with water or an aqueous solu-tion for one hour no detectable quantity of ozone is produced. (2) Ozone a t a great dilution is not appreciably destroyed when shaken with water for one hour.(3) Hydro’gen peroxide present as vapour in air is completely destroyed * by passage of the air a t a ratel not exceeding 1 litre per minute through a chromic acid tube prelpared as described above. (4) Ozone is completely destroyed by passage through a tube con-taining asbestos-wool mixed with manganese dioxidel. (5) Ozone is not affected by passage over chromic acid. (6) Nitrogen peroxide is not absorbed fromm air containing it by solid chromic acid -1 or by manganese dioxide provided the latter contains no alkali or mangaiious oxide. (7) N / 400,000-Sodium nitrite solution undergoes no alteration in strength by shaking for two hours with purified air. (8) I f air containing nitrogen peroxide is passed through a tube of manganese dioxide (a) she and ( b ) with ozone the amount of nitrogen peroxide recovered is in each case the same as that which was introduced.EfJect of Impurities. The presence in air of traces of ammonia sulphur dioxide and hydrogen sulphide does not interfere with the estimation of ozone and nitrogen peroxide since all three gases are completely absorbed during passage through the chromic acid tube. Air charged with small quantities of these impurities was passed through the chromic acid tube and afterwards shaken with a sensitive reagent and in no case was the latter affected. Nessler’s reagent was used for detecting ammsonia dilute permanganate for sulphur dioxide and an alkaline lead solution for hydrogen sulphide. It is difficult to say whether traces of ammonia would affect the esti-* By “completely destroyed” we mean that the quantity of substance remaining was undetectable by the most delicate tests.t Even chromic acid solution absorbs nitrogen peroxide with great diffi-culty. In one experiment several hours’ shaking was necessary before it became impossible to detect nitrogen peroxide in the residual air. The following facts were established by further experiments 508 USHER AND RAO THE DETERMINATION OF OZONE ETC. mation of hydrogen peroxide but the other two gases would cer-tainly invalidate it. For this reason we attach no importance to the nitrite process for the determination of hydrogen peroxide, although it might be used in places far removed from towns and railways where sulphur compounds can be shown to be absent.In this connexion we would point out that the estimation and even the detection of hydrogen peroxide in the atmosphere presents great difficulties. Hitherto titanium sulphate has been used for the purpose but since nitrous acid immediately destroys the yellow colour produced in that reagent by hydrogen per~xide and ozone does so less readily it is clear that the absence of a coloration does not prove the absence of hydrogen peroxide although the produc-tion of tRe colour would be a qualitative proof of its presence. Other impurities likely to be present in air are probably without any effect on the process. Chlorine or hydrochloric acid would make the estimation of nitrogen peroxide impossible but would not affect that of ozone. A ccuracy. The accuracy of the determination is limited by that of the final comparison of colours provided the experimental solutions are1 " developed " by the Griess-Ilosvay reagent always under the same conditions.I n order t o find the degre'e of accuracy obtainable two equal quantities of the same nitrite solution were " developed " and compared. The mean ratio of the lengths of the columns of liquid given by eight readings was 1.012 instead of the correct value, 1.000. This error corresponds with 1 part of ozone (or nitrogen peroxide) in 25 millions and is therefore unimportant for a differ-ence of this order of magnitude po~sesses no interest. A source of error was certainly present in our own experiments in the chang-ing of the stoppers of the bottles containing the samples of air and in lifting them for the purpose of introducing liquid.The amount of diffusion that could take place in two seconds through the neck of a bottle of 7 litres' capacity must however be very small and the remedy is obvious although we were unfortunately not able to apply it. Results of Det erminutions. Hitherto only fourteen compleb determinations have been made covering the period between July 1916 and January 1917. The results are interesting chiefly because on no occasion was any ozone found. Indeed with two exceptions no one of the three substances looked for was found in a quantity exceeding 1 in 20 millions. On November 18th and again on the 22nd nitrogen peroxide was present t o the extent of 1 part in 5 millions and 1 i COMPOUNDS OF FERRIC CHLORIDE WITH ETHER ETC.800 4 millions respectively the weather having been thundery with very little rain during the preceding week. The twelve negative results show more clearly than argument' the trustworthiness of the process adopted in the sense that in spite of its delicacy (the Griess-Ilosvay reagent used in the way we have described is sensi-tive to 1 part of nitrogen peroxide in 56 millions of air) it is possible with care t o obtain concordant results. The interpretation of the results may well be deferred until a much greater number covering a longer period are available We niay remark however that there is some ground for the opinion that ozone and nitrogen peroxide never occur together in the atmo-sphere. It has been shown that at moderate dilutions ozone rapidly oxidises nitrogen peroxide so that the latter substance cannot be detected for example in the air from an ozoniser whereae nitric acid is found in water through which such air is passed (compare Chapman and Jones T. 1911 99 1813). If this oxida-tion takes place a t the great dilutions in which ozone must be supposed to exist in the atmosphere it not only helps t o explain the surprisingly small quantities both of ozone and of nitrogen peroxide found but may also be considered an important factor in the production of the nitric acid which is a normal constituent of air. CENTRAL COLLEGE, BANCIALORE S. INDIA. [Received May 18th 1917.

ISSN:0368-1645    

DOI:10.1039/CT9171100799

出版商:RSC    

年代:1917

数据来源: RSC

摘要:

COMPOUNDS OF FERRIC CHLORIDE WITH ETHER ETC. 800 LXVI1I.- Compounds of Ferric Chloride zuitlz Ether and with Dibenxyl Sulphide. By AQUILA FORSTER CHRISTOPHER COOPER and GEORGE YARROW. THE work described in this paper was brought t o a close in August, 1914 and as an early return to these investigations does not seem probable it is thought advisable t o place on record such results as were obtained. As examples of double compounds of metallic chlorides with ether may be quoted GlC12,2(C2H5),0 (Atterberg Ber. 1876 9 856). AlBr,,( C,H,),O ; HgBr,,3 (C,H,),O ; SnBr2,( C2H,),0 (Nicklks, TiCl,,(C,H,),O (Bedson Joztrn. Chenz. Soc. 1876 i 311). I n the following work a double compound of ferric chloride and ether hae been isolated and its properties have been examined. VOL.CXI. K K Jnhresber. 1861 200) 810 FORSTER COOPER AND YARROW COMPOUNDS OF E'ERRIC It has been found t o combine with aniirionia and with organic amines forming insoluble unstable complexes. A double compouiid of ferric chloride with benzyl sulphide has also been isolated. With benzyl chloride it' forms tribenzyl-sulphinium chloride f errichloride a compound prepared by Hof-mann and Ott (Ber. 1907 40 430) by the condensation of benzyl sulphide benzyl chloride and ferric chloride. By means of these reactions tribenzylsulphinium cyaiiide f errichloride has been prepared from which the free salt tribeiizylsulphinium cyanide has been isolated. EX P E R I M E N T A L. Ethel* Ferric Chloride. Ten grams of anhydrous ferric chloride were added gradually to an excess of pure dry ether.The ferric chloride dissolved with the evolution of much heat and during addition t'he mixture was cooled. The solution was placed in a desiccator the excess of ether distilled off under diminished pressure and the residue dried on a porous plate in a vacuum f o r several days. The product was a dark red highly deliquescent solid soluble in benzene alcohol, or water: 0.6390 gave 0-2305 Fe,O and 1.1293 AgCl. Fe=25*2; C1=44'02. C,H,,O,FeCl requires Fe = 23.7 ; C1= 45.0 per cent. In boiling benzene solution no chlorination was observed. Alcohol and water immediately decompose the compound into ferric chloride and ether. Action of Heat on Ether Ferric Chloride. A t looo ether ferric chloride slowly decomposes with the evolu-tion of ethyl chloride.This decomposition takes place quantita-tively a t higher temperatures over the free flame pure ethyl chloride being obtained according to the equation : (C,H,),O,FeCl == 2C;H,Cl+ FeOC1. The ethyl chloride burnt with the characteristic green flame and gave a precipitate of silver chloride with a solution of silver nitrate in nitric acid. The gas was condensed in tubes in a freez-ing mixture and the weight obtained corresponded closely with the loss in weight of the ether ferric chloride in agreement with the above equation. The liquid ethyl chloride boiled at 12-13O and its vapour density was found to be 35.5 (calc. 32.2). I n one experiment 19.5 grams of the compound lost on heating 10.1 grams or 52 per cent. whilst the theoretical loss according to th above equation is 54 per cent.The ethyl chloride collected weighed 9.5 grams. l’he Action of Ammouia on Ether Ferric Chloride. On passing dry ammonia gas into a solution of ether ferric chloride in cold ether or benzene a brick-red amorphous powder separated. This was non-hygroscopic insoluble in organic solvents, and was immediately decomposed with water or alcohol. With water some samples dissolved whereas others deposited ferric hydroxide with solution of ammonium chloride. An analysis showed that the ammonia absorbed varied indefinitely from one to five molecules per molecule of ether ferric chloride and a pro-duct of definite composition was not prepared although no doubt this could be effected by the use of solutions of ammonia in place of ammonia gas.On destructive distillation these substances gave ammonia and ethylamines etc. Aniline forms a double compound with ether ferric chloride which separates in dark brown plates decomposing a t 90° before melting. Compoztds of Ferric Chloride and Beitzyl Sulpkide. Benzyl Sulphide Ferric Chloride.-Molecular proportions of benzyl sulphide and ferric chloride were dissolved in the minimum amounts of ether and the well-cooled solutions were mixed. The mixture was allowed to remain for several hours during which an evolution of heat was evident and a brownish-yellow mass separated. The product after filtration and washing with dry ether was obtained as lemon-yellow minute crystals rapidly becoming brown on contact with the moist atmosphere. It is soluble in chloroform sparingly so in alcohol and insoluble in ether or acetone.It was purified by crystallisation from hot anhydrous alcohol or chloroform and melted a t 94O : 0.3371 gave 0.0711 Fe20,. (C7H7),X,FeCI3 requires Fe = 14.87 per cent. Benzyl sulphide ferric chloride is distinguished from tribenzyl-sulphinium chloride ferrichloride which i t resembles by only a slight colour change when brought into contact with water. On relnaining for some hours under water it completely dissociates into benzyl sulphide and ferric chloride. Alkalis a t once dis-sociate benzyl sulphide ferric chloride forming benzyl sulphide and ferric hydroxide. Fe = 14.90 812 FORSTER COOPER AND YARROW COMPOUNDS OF FERRIC Condensation of Bensyl Sulpkide Ferric Chloride with Beizzyl Chloride.The readiness with which benzyl sulphide and benzyl chloride combine to form tribenzylsulphinium chloride in the presence of ferric chloride is probably due to the intermediate formation of benzyl sulphide ferric chloride. This view is supported by the ready condensation of benzyl sulphide ferric chloride with benzyl chloride. Benzyl sulphide ferric chloride (10.3 grams 1 mol.) dissolved in 50 C.C. of chloroform was added to a solution of 4 grams (1 mol.) of benzyl chloride in chloroform. The mixture remained in a closed vessel overnight and was then allowed to evaporate. Yellow needles were obtained and identified as tribenzyl-sulphinium chloride ferrichloride (m. p. 98.5O). Benzyl sulphide ferric chloride does not condense with ethylene dibromide with bromoacetic acid or wit% benzyl alcohol.I n agreement with this is the singular fact that benzyl sulphide does not form sulphinium derivatives with these compounds when brought together in ethereal d u t i o n with ferric chloride; in each case only benzyl sulphide ferric chloride was obtained. Decomposition of Benzyl Sulphide with Ferric Chloride. When benzyl sulphide or dibenzyl disulphide and ferric chloride in molecular proportions were heated together a t looo for two to three hours tribenzylsulphinium chloride f errichloride (m. p. 98'5O) in approximately one-half molecular equivalent was formed. The filtrate after washing the product with dry ether was found to contain much ferrous iron but no iron sulphide and its con-tents proving t o be of a complicated nature were not further identified.Similarly dibenzyl disulphide benzyl chloride and ferric chloride in molecular proportions in ether do not give a hexa-benzyldisulphinium chloride derivative but tribenzylsulphinium chloride ferrichloride together with the decomposition products mentioned above. This decomposition of dibenzyl disulphide is in agreement with its well-known hydrolysis with alkalis into benzyl mercaphn and the complex decomposition products of benzyl sulphy droxide CHLORIDE WITH ETHER AND WITH DIBENZYL SULPHIDE. 813 A ction of A rmmonia o n Trib enzylsulphinizcm Chloride Ferric hloride. Ammonia gas is absorbed by a solution of tribenzylsulphinium chloride f errichloride and an insoluble brown amorphous solid is formed melting a t 80-85O.It was unstable towards water and analyses indicated three equivalents of ammonia per molecule but they were not completed. Aniline and tribenzylsulphinium chloride ferrichloride combine in chloroform solution to form in-soluble yellowish-green plates which decompose before melting. Trib enz ylsulphinium Chloride. No means were found for the preparation of tribenzylsulphinium chloride in a state of purity. It was obtained from tribenzyl-sulphinium chloride ferrichloride by removal of the ferric chloride with alkalis as a viscous oil which contrary to former statements, was found but sparingly soluble in cold water. It appeared in-different to solutions of mineral salts and no double decomposi-tion reactions were obtained. The rapid dissociation of tribenzyl-sulphinium chloride into benzyl sulphide and benzyl chloride was very apparent; it takes place in contact with water and is com-plete in three to four hours and benzyl sulphide is generally t o be found as a by-product of the reactions of its derivatives.The oil obtained as tribenzylsulphinium chloride doubtless contained benzyl sulphide and other products of decomposition. Trib enzylsulphiniurn Cyanide Ferrichloride, (C,H,*CH,)3SCN,FeC13. Molecular proportions of benzyl sulphide and benzyl cyanide (phenylacetonitrile) were dissolved in dry ether and added t o a well-cooled solution of ferric chloride in ether. The mixture was cooled and kept in a desiccator overnight when a mass of crystals separated. They were collected washed with ether and obtained in lemon-yellow crystals melting at 76O insoluble in ether but soluble in alcohol or chloroform: 0.2091 gave 0.4103 CO and 0.0812 H,O.0.5301 , 0.01596 (gram) N2. N=3.01. 0.3163 , 0.0504 Fe,O,. Fe 11.10. 0.2800 , 0.1329 BaSO S=6*52. C=53.50; H=4*31. C',,H,,NS,FeCl requires C = 53.5 ; H = 4.29 ; N = 2.84 ; Fe = 11.32 ; S := 6.50 per cent. Tribenzylsulphinium cyanide ferrichloride was also prepared i 814 COMPOUNDS OF FERRIC CHLORIDE WITH ETHER ETC. good yield by the condensation of benzyl sulphide ferric chloride and benzyl cyanide in molecular proportions in ethereal solution, and identified by the mixed melting-point test with the product described above. On the addition of tribenzylsulphinium cyanide ferrichloride to water only a slight colour change takes place and no compound corresponding with the red bistribenzylsulphinium chloride ferri-chloride [(C,H,*CR,),SCl,],,Fe~l~ was obtained.Tri b ens ylsulphiniu rn Cyanide (C,R,* CH,) ,SCN. To a warm solution of tribenzylsulphinium cyanide ferrichloride in alcohol was added an excess of ammonia solution (D 0.88). The precipitated ferric hydroxide was separated by filtration and tribenzylsulphinium cyanide cryst allised from the filtrate. It forms large white prisms melting a t 41° and is only sparingly soluble in water but readily so in organic solvents: 0.1355 gave 0.3932 CO and 0.0782 H,O. 0.2105 , 0.1492 BaS04. S=9*74. 0.6506 , 0.0288 (gram) N,. N=4.50. CzzHzlNS requires C=79.76; fE=6.34; S=9*67; N=4.23 per cent. Tribenzylsulphinium cyanide gave no double decomposition with mineral salts in solution.It' condensed readily with ferric chloride in anhydrous solution forming tribenzylsulphinium cyanide f erri-chloride. A double compound of tribenzylsulphinium cyanide and platinum chloride was precipitated from an aqueous alcoholic solution in minute sparingly soluble red crystals melt,ing a t 162O : 0.6890 gave 0.1378 Pt. Pt=20'0. C44H,,S,N,PtC14 requires Pt = 19.0 per cent. The dissociation of tribenzylsulphinium cyanide in contact with water was complete in a few hours benzyl sulphide and benzyl cyanide being formed. The reverse reaction the direct combina-tion of benzyl sulphide and benzyl cyanide was not found to take place in twelve hours' heating on the water-bath. Hydrolytic agents invariably dissociated tribenzylsulphinium cyanide into benzyl sulphide and benzyl cyanide and its products and no deriv-ative of the original material could be prepared. Preliminary experiments indicated that the reactions described also take place when mercuric chloride is used instead of ferric c>hloride and i t was proposed t o examine the decomposition of benzyl sulphide and dibenzyl disnlphide wit'h this condensation agent. C=79*89; H=6*41. ARnmTRoNU COLLEGE, NEWCASTLE - ON-TY NE. [Received May 24th 1917.

ISSN:0368-1645    

DOI:10.1039/CT9171100809

出版商:RSC    

年代:1917

数据来源: RSC

摘要:

EFFECT OF ADDITIONAL AUXOCHROMES ON COLOUR OF DYES. 816 LXIX-The Effect of Additional Auxochromes on the Colour of Dyes. Part II. Triphen ylmethane and Azo-dyes. By PRAPHULLA CHANDRA GHOSH and EDWIN ROY WATSON. IN two previous papers from this laboratory (Medhi and Watson, T. 1915 107 1581; Meek and Watson T. 1916 109 544) the effect of multiplying the auxochrome groups has been studied spectroscopically in the xanthene and anthraquinone groups of dyes. I n the present paper the investigation has been extended to some azo- and triphenylnlethane dyes and some phenylfluorone dyes have been prepared but have not yet been examined spectro-scopically The results here recorded have been obtained by observing with an ordinary spectroscope the positions of the edges of the absorption bands in solutions of different concentrations.Experience gained with the anthraquinone group has shown that it is much preferable t o investigate the absorption spectra by means of a spectrophotometer and it is hoped that the present work may subsequently be checked by the use of that instrument; but the results already obtained are now published as it is un-likely that the authors will have further opportunity to study the subject together. Little insight has yet been obtained into the mechanism by which additional auxochronies affect the absorption spectra. I n fact the effects produced are very different in different cases and little explanation is as yet forthcoming as to why such varying effects should be produced. I n some cases the multiplication of auxochromes produces the comparatively small effect of strengthen-ing and broadening the absorption bands without much effect on their positions as f o r example in the azobenzene series.I n other cases the bands are shifted to a considerable extent but remain of about the same breadth. I n some cases the breadth is much affected ; it may be either increased o r diminished whilst yet again in other cases additional bands appear. Even in the 8ame series very diverse effects may be produced f o r no apparent reason. Apparently much more material is required before any explana-tion of the effects can be offered 816 GHOSH AND WATSON THE EFFECT OF ADDITIONAL Absorption Spectra Exnrnincd. A 20-group. In dilute .................. hydrochloric acid solution.In dilute potassium Aminoazobenzene C6H;N,'C,H;NH, Chrysoidine CGH5*NP[ 1].C,H3(NH2),[ 2 41 4 4'-Diaminoazobenzene NH,.CsH;N;C6H4'NH, Benzeneazopyrogallol ... CGHB*N2[ l]*c6H2( OH),[2 3 41 solution' Benzeneazophenol C6H5*N,*C,H,'OH Benzeneazoresorcinol CGH;N2[ l]*C6H3( OH),[2 41 Benzeneazoquinol C,H;N,[ 1]'C,H3( OH),[2 51 Benzeneazocatechol C8H,'N,[11'C,H,(OH),[3 41 Absorption spectrum recorded by Tuck T. 1909,95 1809.) (Benzeneazo-a-naphthol.. C6H5. N,[ 1]'Cl,H,'OH[4] p-Hydroxybenzeneazo - u-Benzeneazo-B-naphthol.. C,H;N,[ l]C,,H,*OH[2] p-Hydroxybenzeneazo- 8-Benzeneazo- B - naphthyl-pHydroxybenzeneazo- B - acid ......... ......... ... ......... hydroxide ...... potassium In dilute solution. 1 naphthol ..................OH*C,H,*N,[ 1]'C1,H;OH[4] hydroxide 1 ! . .................. In dilute naphthol OH'C,H,'N,[ l]*C,,H;OH[ 21 amine ..................... C,H,'N,[ 11 'C,,H,'NH,[ 21 1 hydrochloric naphthylamine ......... OH'C,H,*N,[ 1]*C1,H,*NH,[2] I solution. B e n z e n e a z o - 1 6-di-hydroxynaphthalene ... p-Hydroxy be n z e ne a z o-1 Ei-dihydroxynaphth-alene ........................ OH*C,H4*N,[1]*C,,H,(OH),[4 . 81 p-Hydrox y benzeneazo-1 3-dihydroxynaph t h -alene ........................ OH'C,H,*N,[ 13 *Cl,H,( OH),[ 2 . 41 C,H;N,[ 1]*Cl,H,(OH),[4 . 81 I n dilute potassium ' hydroxide solution. The results are recorded in Figs. 1-6." It will be observed that the effect of adding a second amino-group t o aminoazobenzene is slight. When the second amino-group is added to the same benzene nucleus the front edge of the band (the edge nearer the red end of the spectrum) is shifted slightly towards the red; when the second amino-group is added to the other benzene nucleus the edge of the band is shifted slightly towards the blue.The effect of adding a second hydroxyl group to bcnzeneazo-phenol is similar. When the second hydroxyl group is added t o the same benzene nucleus the front edge of the band is shifted towards the red. Two additional hydroxyl groups in the same benzene nucleus produce the same general effect namely a shift of the front edge of the band towards the red. The shift is greatest in the case of benzeneazocatechol then come benzeneaze pyrogallol benzeneazoquinol and benzeneazoresorcinol in order, * Compare Tuck (T.1907 91 450) for absorption spectrum of benzene-azophenol; Hewitt and Thole (T. 1910,97 513) for aminoazobenzene and Rartley (T. 1887,51 166) for chrysoidine AUXOCHROMES ON THE COLOUR OF DYES. PART 11. 817 according t o the displacement produced. These resulte are in agreement with the generalisation already pointed out (Meek and Watson h c . cit. SSS) that. two hydroxyl groups in the o- or p-position with respect to one another give a deeper colour than FIG. 1. - Arninoazobewzene With 20 times the theoretical ‘i quantity of hydrochlorio -- Chrysoidine . . . . . . . . . 4 4’-Diaminoazobenxene I acid. 0 0 0 0 0 0 3 0 0 0 0 0 0 3 0 0 0 0 0 ( 0 ~ ~ 0 0 0 ( 0 w u5 1 0 u3 m u5 d( * two hydroxyl groups Wave. lengths. BIG.2. Benzeneazophenol Benzeneazopyrogallol With times the of potassium hydr -Benzeneaxoresorcinol I Oxide‘ Benzeneazoquinol BenzeneaZocatechoJ ‘heoreticaE quantity in the ni-position. The effect of two hydroxyl groups in the o-position with respect to one another is greater than that of three adjacenta. (The blue colour of benzeneazo-quinol in potassium hydroxide is fugitive and the absorptioli spectrum of this blsle solutioii could not ba observed.) K K 818 GHOSH AND WATSON THE EFFECT OF ADDITIONAL Hewitt (T.? 1909 95 1295) has recorded the absorption spec-trum of aminobenzeneazophenol in dilute hydrochloric acid. Com-paring i t with that of aminoazobenzene we see that in this case also the additional auxochrome has shifted the front edge of the band towards the red.Hewith's observations extend into the ultra-violet so that the effect of the additional auxochronie can be more completely described. Aminolazobenzene in dilute hydro-chIoric acid aolution has two bands one partly in the visible spectrum and the other completely in the ultra-violet. The addi-tional auxochrome shifts both bands a little towards the red and increases their persistency. FIG. 3. Wave-lengths. - Benzeneazo-a-naphthoE with sodium allyloxide. (Tuck.) - - - - - p -Hydroxybenzeneazo- a-naphthol with 20 times the theoretical quantity of potassium hydroxide. The addition of a hydroxyl group t o the benzene nucleus of benmneazo-a-napht,hol produces only a slight increase in the per-sistency of the band but if the extra group is added t o the naphthalene nucleus in the am-position with respect to the original hydroxyl group the barid is not only strengthened but is shifted a long way towards the red.Two additional hydroxyl groups one in the benzene nucleus and the other in the naphtha-lene nucleus in t.he ma-position with respect to the original hydroxyl group have an additive effect so that the band is shifted a little further than in the case of benzeneazo-1 5-dihydroxy-naphthalene and is made still a little stronger. Two additiona AUXOCHROMES ON THE COLOUR OF DYES. PART 11. 819 hydroxyl groups one in the benzene nucleus mid the other in the 112-position with respect to the original hydroxyl group in the naphthalene nucleus produce scarcely any sliif t of the original absorption band but only a strengthening.This result is in agreement with our previous experience that an auxochrome ac’ded in the m-position with respect t o one idready present has very little effect on the absorption. Very different’ is the effect of the addition of a hydroxyl group to the benzene nucleus of benzeneazo-B-naphthol which not only FIG. 4. Wave-lengths. 7000 66 62 58 54 5000 46 42 3 -_- Benxeneazo-a-naphthol with sodium allcyloxide. (Tuck.) - - - - - Benzeneazo- 1 5-dihydroxynaphthalene W i t h 20 times the theoretical p-Hydroxybenzeneazo- 1 ; 5-dihydroxynaphthalene quantity of p-Hydroxybenzeneazo- 1 3-dihyd~oxynaphthalene p o t a s s i u in i hydroxide. . - . - . - . . . . . . . . . shifts the existing band further towards the red but calls into existence an entirely new band still further towards the red.In the case of benzeneazo-P-naphthylamine a similar addition also seems to call into existence a new band but apparently the already existing band is shifted backwards that is towards the ultra-violet. This may be because the additional auxochrome that is, the hydroxyl group cannot exert its full effect in the acid solu-tion. (It is however to be regretted that the absorption spec-trum of benzeneazo-P-naphthylamine was not more completely observed .) K K* 820 GHOSH AND WATSON THE EPFECT OE' ADDTTIONAL Malaahite-green ... 4-Hydr o x y m a1 a-chite-green ...... 3 4-Dihy droxy -malachite -green 2 3 4-Trihydroxy-malachite -green Aurin ............... Resaurin ............Trihydroxy au r in Trihydr o x yau r i n (a) (b) .................. The results obtained by introducing hydroxyl groups into the The band phenyl group of malachite-green are very interesting. of least frequency is shifted backwards that is towards the ultra-violet whilst the next band is moved forwards from the ultra-violet into the visible part of the spectrum. Two hydroxyl groups in the o-position with respect to one another shift this second band further forwards than one hydroxyl group alone is able to do, whilst three adjacent hydroxyl groups produce an intermediat AUXOCHROMES ON THE COLOUR OF DYES. PART 11. 821 effect (compare the effect of additional hydrosyi groups added to benzeneazophenol). These results fall into lins with the effect produced by the addition of an amino-group to the unsubstituted benzene nucleus in Doebner’s violet when we pass to pararosaniline, or the similar addition of a dimethylamino-group to malachite-green when we pass to hexaniethylpararosaniline.In all these cases the addition of auxochromes t o the unsubstituted third benzene nucleus shifts the first band (the band of least frequency) backwards so that the dye becomes lighter instead of deeper in colour. We see the same effect if we compare benzaurin (Meyer a X 4: FIG. 6. 1 Vase- letzgtks. 7000 66 62 58 64 5000 46 48 20 18 16 14 12 10 8 6 With 20 times the theoretical quantity of hydrochloric acid. p - Hydroxy benzeneazo - B - naphth y lamine Benxeneazo - B -nap hthy lamine i _ _ - - -and Fischer Ber.1913 46 74) and aurin so that it appears t o be quite a general rule in the triphenylmethane group. It seems possible t o explain this effect by means of the theory put forward by one of us (Watson and Meek T. 1915 10‘7 1567) as to the nature of the vibrations causiilg the colour of dyes. In Doebner’s violet malachite-green and benzaurin the vibration passes back-wards and forwards through two of the benzene nuclei. I n pararosaniline hexaniethylpararosaniline and aurin the vibratory pulse after passing through one benzene nucleus finds two paths open to it as each of the remaining nuclei is now capable of vibration. We may find a mechanical analogy by comparing th 822 CHOSH AND WATSON THE EFFECT OF ADDITIONAL transverse vibration of a string with that of a system of three strings each of half its length tied together at one point and all under the same tension so that they make equal angles with one another.Such a system will vibrate quicker than the simple string. I n the same way when all three nuclei of the triphenyl-methane moIecule are capable of vibrating the period is less than when the vibration is confined to two of the nuclei. I n the latter case the unsubstituted benzene nucleus niay be coinpared to a dead weight a t the centre of the simple string which would make FIG. 7 . Wave-lengths. X 5 2 3 With 20 times the theoretical quantity of hydrochloric acid. -- Malachite -green - . . _ . . 4 - Hydroxymalachite-green - . __ . __ . . . . . . . . 3 4-Dihydroxymalachite-green 2 3 4-Trihydroxymalachite-grcr,~ the vibration slower.By comparing Michler's hydro1 and nialachite-green we see this effect Nothing can be said as to the cause of the second band in malachitegreen and the hydroxymalachite-greens. It may be suggested that the first ban6 of magenta which is obviously com-pound is produced by the fusion of the first and second bands of Doebner's violet the second band having been brought further forward by the additional amino-group which can exert its maxi-mum effect in dilute acid solution than by the hydroxyl groups in the hydroxymalachite-greens which could not exert their ful AUXOCHROMES ON THE COLOUR OF DYES. PART 11. 823 effect in acid solution. The same suggestion may be made as to the nature of the first7 band of aurin which is also obviously compound.With regard to the auiin derivatives examined the series is not so completel as was desired. Pyrogallaurin could n'ot be obtained by Caro's method (Ber. 1892 25 2678) and the alkaline solution of catecholaurin €ades too rapidly to permit of the ccnvenient examination of its absorption spectrum. Quinolaurin could not be prepared either by the condensation of formic acid and quinol or by Caro's method (Ber. 1892 25, 1941) by the condensation of 3 6 3/ 6/-tetrahydroxydiphenyl-methane and quinol. I n trihydroxyaurins (a) and (b) the band is shifted towards the red and becomes much narrower. As might have been expected where there are hydroxyl groups in the o-position with respect to one another the shift is great,er than where they are para with respect to one another.I n resaurin the band is shifted backwards. It may perhaps be suggested that in this compound fluorone condensation has occurred. The Colour of Some Pluorone Dyes. The following new compounds of this series have been prepared and their colours noted: 3 - Ihni t t Iiyla 111 isz n-g-p-hydl.oxyphe~zyl-6-dirne t h yl fluorime - dyes dull rad shades. 3- Dime t Jz,ylccnziii o-9-o-pdi 71y~roxypl~enyl-6-dime t hylfluorime-dyes red shades on chrome. 2 3 7-Trih,ydroxy-9-o-pdiJi ydroxyplz,enyl-6-fluorone - reddish-violet in potassium hydroxide solution ; dyes reddish-violet shades on chrome. in potassium hydroxide solution ; dyes orange shades on chrome. Discussion of the colour is reserved until the absorption spectra have been examined.1 3 8-T~hydroxy-9~p-~hydroxyphc~l-6-fEzcoro~e-orange EX P E R I M E N T A L. p-Hydcroxyb enaeneazo-1 5dihydroxynaphthalene, OH* C,jH,-N,*C,0H5(OH)2. An alcoholic solution of 1 5-dihydroxynsphthalene was mixed with a concentrated aqueous solution of p-hydroxybenzene-diazonium chloride and then excess of sodium acetate was added. The precipitate was collected and washed thoroughly with water. The substance could not be crystallised but was extracted with a mixture of benzene and alcohol. On concentrating the extract 824 QHOSH AND WATSON THE EPBECT OF ADDITIONAL the pure dyestuff was deposited. raddish-violet shades on alum- or chrome-mordanted wool. dyeings are obtained with 2-3 per cent. of the dyestuff. yield is almost theoretical : It melts a t 213-215O and dyes Full The 0.1038 gave 8.8 C.C.N a t 2 5 O and 763 mm. N=9*7. Ci6HI2O3N2 requires N = 10.0 per cent, pHydro x y b e n z e n ea z 0-1 3-di h y dr o x y n n p h t h n 1 e n e , OH*C6H4*N2*Cl,H,( OH),. pAminopheno1 hydrochloride was diazotised in the ordinary way and combined with an alkaline solution of 1 :8dihydroxy-naphthalene. The mixture was acidified after keeping overnight, and the precipitate was collected washed with water and dried. It could not be crystallised. It was purified by dissolving in alkali precipitating with acid and again dissolving in alcohol and precipitating with water. It does not melt below 300O: 0.1107 gave 10.2 C.C. N a t 29.2O and 763.5 mm. N=10*4. @16H@,N requires N = 10.0 per cent.p-zydroxy b enz enea-zo-P-mph t h y 1 ani,ine 0 H C,H,-N,* C loH6*NH,. An alcoholic solution of P-naphthylamine was added to a con-centrated aqueous solution of phydroxybenzenediazonium chloride. The precipitate was crystallised from alcohol and melted a t 156-157': 0.0825 gave 11.4 C.C. N a t 27O and 762 mm. C16H130N requires N = 15.97 per cent. The hydrochloride was deposited in crystalline form on adding hydrochloric acid to a hot alcoholic solution of the base and allow-ing the solut,ion t o cool. It melts a t 170O. N=15*8. o-A mino b enz eneazo-a-n aph t h oZ NH,-C6H4*N,*CloH6*OH. o-Nitrobenzeneazo-a-naphthol was obtained in a finely divided state by dissolving in cold concentrated sulphuric acid and pre-cipitating with water. The precipitate was thoroughly washed free from acid and was then warmed with a large volume (100 C.C.for 1 gram of the substance) of freshly prepared aqueous ammonium sulphide on the water-bath. After a few minutes the whole of the azo-compound passed into solution. On remaining overnight the solution deposited a fine brown shining precipitate, which was collected well washed with water and dried on a porous tile. The substance is very readily tsoluble in all ordinary solvents and could not be crystallised. It was purified by extract AUXOCHROMES ON THE COLOUR OF DYES. PART 11. 825 ing with alcohol and precipitating with water. It melts a t 19!%196O : 0.0915 gave 13.2 C.C. N a t 3 2 O and 761 mm. N=16*18. C,,H,,ON requires N = 15-97 per cent. 3 6 3l 6f-TetrahydroxycFi;phenyEmethanu CH,[C,H3(OH),]2.According t o Schorigin ( J . Russ. Phys. C‘he,n. SOC. 1907 39, It was however obtained as a 1094) this substance is brown, FIG. 8. Tl’ave. lengths. X 0 l-l 20 18 16 14 12 10 8 6 With 20 times the theoretical quantity of p o t a s s i u rn Trihydroxyaurin (a) Trihpdroxyaurin (b) i hydroxide. Aurin Resaurin __ . . - . . _ . _ . _ . . . . . . . . white amorphous substance under the following conditions. Two and a-half grams of quinol were dissolved on gently warming with 25 C.C. of hydrochloric acid (1 10); 1 gram of 40 per cent. form-aldehyde solution was then added and the mixture heated on the water-bath. The white precipitate of 3 6 3 l 6’-tetrahydroxy-diphenylmethane began to appear in a minute and the heat-ing was continued for ten minutes.The compound could not be crystallised. It decomposes above 275O. (Found C = 66.68 ; H=5*44. Cl3HI2O4 requires C=67*2; H=5.17 per cent. 826 GHOSH AND WATSON THE EFFECT OF ADDITIONAL Tr&ydroxyaurb ( b ) , C([1]C6H3(oH),[2 :5]),:[1]C6H3(oH)[2]’0[4], was prepared from 3 6 3’ 6’-tetrahydroxydiphenylrnethane and resorcinol by Caro’s method (Ber. 1892 25 941). It could not be crystallised. It was purified by dissolving in alkali precipita-ting with acid and again dissolving in alcohol and precipitating with water. The operation was repeated twice t o obtain the sub-stance in a pure state. The yield was very low: 0*1020 gave 0.2633 CO and 0.0413 H20. Trihydroxyaurin ( ( I ) , C=70*41; H=4*50.ClQHI4O5 requires C = 70.8 ; H = 4.34 per cent. c( [1]C6H,(0H)2[3 4])2:[ 1]C6H3( OH)[2]:0[4], was prepared and purified in the same way as the above-mentioned trihydroxyaurin ( b ) : 0*1100 gave 0.2836 CO and 0.0436 H20. C= 70.3 ; H=4*4. CIQH,,O requires C = 70.8 ; H = 4.34 per cent. 4-Hydro ~ y - 4 ~ 4 1;-t e tra m e t h y Zdiacmiii o triphenylcarbinol Anhydride, or 4-Hydroxymalachit e-green 0:C6H4 C(C,H,*NMe& This dyestuff had already been obtained in a qualitative way by 0. Fischer (Ber. 1881 14 2523)) but not analysed. It was pre-pared by oxidising the leuco-compound in acetic acid solution with freshly precipitated manganese dioxide. The paste of manganese dioxide obtained from 0.52 gram of potassium permanganate was added to a solution of 1 gram of the leuco-compound in 35 C.C.of 30 per cent. acetic acid. The colour became dark green and the mixture was warmed on the water-bath for an hour and filtered. The filtrate was treated with sodium acetate solution whereby a violet precipitate of the dyestuff was obtained which was crystal-lised from a mixture of benzene and toluene. The yield was good. It dyes mordanted wool in the following shades blue on alum, bluish-violet on chrome bluish-black on iron. Tannin-mordanted cotton is dyed to a dark violet shade: 0.1240 gave 8.8 C.C. N at 33O and 763 mm. N=7*98. C,,H,,ON requires N‘= 8.13 per cent. 2 4-Dihydroxy-4’ 4 11-t etra me t h,yldiami~otripheilcar b i n o l A nh ydride or 2 4-Dih ydrox y malachit e-gre en, 0 C6H3( OH) :C(C,H,*NMe,),. This dyestuff had already been obtained in a qualitative way by VotoZek and Krauz (Bey.1909 42 1605)) but not analysed. It was prepared by oxidising a 30 per cent. acetic acid solutio AUXOCHROMES ON THE COLOUR OF DYES. PART 11. 827 of the leuco-compound with freshly precipitated manganese dioxide, and obtained as a fine blue solid which could not be crystallised. It dyes chrome-mordanted wool to a very brilliant blue shade: 0.1000 gave 7.1 C.C. N2 a t 3 3 O and 762 mm. N=7*93. C,,H,O,N requires N = 7.77 per cent. 3 6-Tetra tiw t Ayldk n r i n o - 9 - p - l ~ ? / ~ ~ o ~ ~ ~ h e ~ ~ ~ l ~ ~ ~ i t k e i i e. One gram of p-hydroxybenzaldehyde and 2 grams of m-dimethyl-aininophenol were dissolved in 50 per cent. aqueous alcohol and 4 grams of concentrated sulphuric acid were then slowly added.The mixture was boiled under a reflux condenser for ten hours, and then the alcohol was expelled by boiling with water. The leuco-base was fractionally precipitated by sodium acetate and the pure substance was obtained by crystallising the second frac-tion from dilute acetone. It melts a t 220-221O: 0.0995 gave 6.8 C.C. N at. 325O and 762 nun. The benzoyl derivative was prepared by the Schotten-Baumann reaction. It could not be crystallised because of its ready solu-bility in all common solvents. It melts and decomposes a t N= 7.68. C,,H,,O,N requires N = 7-77 per cent. 100-1 05O : 0.1500 gave 7.9 C.C. N a t 3 2 O and 760 mm. C30H2903N2 requires N = 6.02 per cent. 3-L)ime thylamino-g-p-hy~?-o~ yphenyl-6-dim e Ihylflu orim e was pre-pared by oxidising the leuco-compound in acetic acid solution with freshly precipitated manganese dioxide.The dyestuff was pre-cipitated from the acetic acid solution by aqueous sodium acetate solut*ion as it could not be crystallised. It dyes dull red shades: N=7-83. N=5.89. 0*1000 gave 7.0 C.C. N a t 3 3 O and 762 mm. C,,H,O,N requires N= 7.77 per cent. 3-Dim e t hylu nzino-9-o-p-dz hyclroxyyhe )i yl-6-dim et hylfEuori m.e. Two grams of 2:4-dihydroxybenzaIdehyde and 4 grams of m-dimethylaminophenol were dissolved in 50 per cent. aqueous alcohol and 8 grams of sulphuric acid (D 1.84) were then gradu-ally added. The mixture was heated under a reflux condenser for twenty-four hours and filtered. The filtrate was boiled with water to expel alcohol and then sodium acetate was added to precipitate the dyestuff.It was partly purified by dissolving in sulphuric acid and precipitating with sodium acetate and finally obtained in a pure state by debenzoylating the crystalline benzoyl deriv-ative by boiling its alcoholic solution with sulphuric acid. Th 828 EFPECT OF ADDITIONAL AUXOCHROMES ON COLOUR OF DYES. dyestuff could not be crystallised. to a red shade which is not brilliant: It dyes chrome-mordanted wool 0.1055 gave 7.6 C.C. N a t 33O and 760 mm. C,,H,O,N requires N = 7.6 per cent. The leuco-compound can be obtained by reducing the dyestuff with zinc dust and acetic acid but is oxidised again t o the dye-stuff on coming into contact with air. The monobenzoyl deriv-ative of the dyestuff was prepared by the Schotten-Baumann reac-tion in the ordinary way.It is soluble in all the common solvents. By the spontaneous evaporation of a cold chloroform solution it crystallises in prismatic needles melting a t 125O : 0.1665 gave 8.0 C.C. N at 33O and 763.2 mm. N=8*03. N=5.42. C3,H2704Nz requires N=5.8 per cent. 2 3 7-Trihydroxy-9-o-p-dihydroxypheny~-6-fZuorone. One gram of 2 4-dihydroxybenzaldehyde and 2 grams of hydroxyquinol were dissolved in 10 C.C. of 50 per cent. alcohol;. 1.5 C.C. of sulphuric acid (D 1-84} were then added and the mix-ture was heated on the water-bath for fifteen minutes when the fluorone sulphate was precipitated. It was collected and washed with 50 per cent. alcohol: 0.5475 gave 0*1800 BaSO,. The dyestuff was obtained by boiling the sulphate with water for one and a-half hours.It dis-solves in potassium hydroxide solution with a reddish-violet colour and dyes chrome-mordanted wool in fine reddish-violet shades. Acetylation resulted in the production of a mixture of derivatives which could not be separated by fractional crystal-lisation : R,SO,= 12-5. (ClgH,407),H2S04 requires H,SO = 12.15 per cent. It could not be crystallised. 0.1200 gave 0.2805 CO and 0.0405 H,O. C= 63.75 ; H = 3.75. C19Hl4O7 requires C= 64.4 ; H = 3-95 per cent. 1 3 8-Tr~hydroxy-9-o-pdihydroxyphenyl-6-~uorone. One gram of 2 4-dihydroxybenzaldehyde and 2 grams of phloro-glucinol were dissolved in 10 C.C. of 50 per cent. alcohol and 1.5 C.C. of sulphuric acid (D 1-84) were then added. After shaking for two minutes orange-coloured needles of the fluorone suZphute began to form. They were collected and washed with 50 per cent. alcohol : 0*5000 gave 0.1515 BaSO,. The dyestuff was obtained by boiling the sulphate for one and H,SO = 12.4. (ClgHl,07)2R2S0 requires HzS04 = 12.15 per cent THE ABSORPTION SPECTRA OF SUBSTANCES ETC. 829 a-half hours with water. It could not be crystallised. It dis-solves with a11 orange coloiir in potassium hydroxide solution and dyes chrome-mordanted wool in orange shades : 0*1000 gave 0.2350 CO and 0.0330 H,O. c'=64*09; H=3*66. CI9Hl40 requires C=64 4; H=3.95 per cent. The experimental work described in this paper has been done entirely by P. C. Ghosh. DACCA COLLEGE LABORATORY, DACOA BENOAL INDIA. [Received July 6th 1917.

ISSN:0368-1645    

DOI:10.1039/CT9171100815

出版商:RSC    

年代:1917

数据来源: RSC

摘要:

THE ABSORPTION SPECTRA OF SUBSTANCES ETC. 829 LXX.-T?ze Absorption Spectra of Substances con-taining Conjugated and Unconjugated Systems o f Triple Bonds. By ALEXANDER KILLEN MACBETH and ALFRED WALTER STEWART. IT has long been known that saturated substances possess an absorptive power much lower than that shown by the correspond-ing ethylenic derivatives ; and further investigations have proved that of two isomeric diethylene derivatives that which contains a conjugated system of double linkings has a greater absorptive power than the isomeride in which the ethylenic bonds are isolated from each other (Crymble Stewart Wright and Glendinning T., 1911 99 451; Crymble Stewart Wright and Miss Rea ibid., 1262). The present investigation wm undertaken with the idea of testing whether or not a similar rule could be established in the case oS acetylenic derivatives but the results show that no such generalisatlion can be looked for in this series.The data however, are of some interest from other points of view. It was decided to examine acetylenic compounds belonging to both aliphatic and aromatic series. Some of the latter class had already been investigated by Stobbe (Bey. 1911 44 l289) but 110 general rule could be deduced from his results as the phenyl nucleus might be expected t o interfere with the influence of the acetylenic linking to some extent. It will probably render the data clearer i f the discussion of each group of compounds be takeu separately final conclusions being reserved until the whole facts have been reviewed.Behenolic Acid and Erucic Acid (Fig. 1 A).-Here Series I 830 MACBETH AND STEWART : the ethyleiiic derivative is much more absorbent than the acetylenic analogue. In this case the effect of the double and triple link-ings may be taken as being unaffecteld by the preisence of the carboxyl radicle from which they are separated by eleven methylene groups. Series 11. Stearolic Acid Elaidic Acid and Stearic Acid (Fig. 1 B).-In this case the acetylenic derivative is the most strongly absorbent the ethylenic next and the saturated sub-stance least otf all. The only factor that might account for this marked absorptive power of the acetylenic substance is to be found in the fact that the carbonyl radicle of the carboxyl group is FIU. 1. Oscillation frequencies.situated a t the ninth carbon atom from both the triple and double linkings which would imply under the usual assumptions as to the arrangements of carbon chains in space that the unsaturated centres were spatially conjugated with two turns of the spiral between them. It is sufficient t o indicate this and leave the matter without further elaboration. Acetylenedicarb oxylic A cid Fumaric Acid Maleic Acid and Succinic Acid (Fig. 1 C).-As can be seen from the graph the two1 ethylenic isomerides are more absorptive than the acetylenic compound-fumaric acid being markedly so-and succinic acid is least absorptive of all. Series IZI THE ABSORPTION SPECTRA OF SUBSTANCES ETC. 831 Series Z I f . Di-iodoctcetylene D i i o d o e t k y l e n e ccnd .Ethylene Iodide (Fig.2).-Thie case presents a fresh problem. The saturated compound ethylene iodide possesses a distinct absorp-tion band having its head at' a frequency of 3900 and this band is directly attributable tot the presence of the two iodine atoms FIa. 2. Oscillation frequencies. GO 68 56 g 54 '$ 52 50 6 48 2 46 -0 0 5 44 2 42 3 s 3 2 34 5 32 3 30 40 % 38 ' 36 -* c Q 28 26 24 22 20 ______ Di-iodoacet ylene. __ . __ . __ Di-iodoethylene. - - - - - - - - Ethylene iodide. nm. mm. mm. mm. mna. N/1 N/10 N/ 100 N/lOOO N/10,000 in the molecule. Now the introduction of fresh unsaturated centres in t,he molecule might have either of two effects since there may be an enhancement of the band due t o stimulation of the process which produces it or conversely a cancellation of affinity leading to a decrease in absorptive power.An inspection of th 832 MACBETH AND STEWART : curves shows that the second process is in operation to Some extent, for the acetylenic compound is much less absorptive than the saturated substance. The absorption of di-iodoacetylene is limited to a frequency of 4200 at a logarithmic thickness of 36 whereas the corresponding ethylenic compound continues to absorb to the same extent in a logarithmic thickness of 32. Further the absorptive power of the acetylenic compound is except for a por-tion in the visible region of the curve actuaJly less than that of the aliphatic analogue from which it must be deduced that the introduction of the triple bond into the molecule has considerably influenced the absorptive power of the iodine atoms.On the other hand by the same reasoning the ethylenic linking has an effect opposite to that of the triple bond whether this be attribu-table to a stimulation of the iodine vibration by the affinity of the residual affinity of the carbon atoms or merely to the increase in absorptive power usually noted when passing from the paraffin to the olefine derivatives. Series V. Dimethyldiacetylene €€exatriene,+ Benzene f (Fig. 3). -This case is one of the most interesting of all those which have yet been examined. A comparison of the structural formulz of the three compounds must first be made: ‘ / I ’ C6H6 Dimethyldiacetylene CH3-<5C-C&-OH3 C$3 Hexatriene C,H Benzene \ / ‘“CH.It will be noticed that dimethyldiacetylene and benzene are isomeric whilst hexatriene contains ail extra pair of hydrogen atoms and from this it might be expected that the first two sub-stances would show greater absorptive power than hexatriene. An examination of the curves in Fig. 3 shows that this deduction is unwarranted; contrary to the usual rule the more saturated sub-stance is the most strongly absorbent except just at the very t o p of the curve. I n order to explain this apparent anomaly i t is necessary t o bear in mind what has already been said with respect to the influence of conjugation. Take first the case of benzene and hexatriene. I n the benzene nucleus there are three double linkings but> these * This curve is taken from Baly and Tuck T.1908,93 1909. t This curve is taken from Baly and Collie ibid. 1905,87 1332 the banded region being indicated by shading THE ABSORPTION SPECTRA OF SUBSTANCES ETC. 833 three linkings saturate each other and form a completely linked system with no free partial valencies. I n the case of hexatriene, on the other hand although in its molecule there are also three conjugated double linkings these double linkings do not com-pletely saturate each other for there is a considerable amount of FIG. 3. Oscillation frequencieo. 3 0 0 m G 0 CJ m 0 0 00 CI?. 0 3 0 0 0 g s z _ _ ~ . Dinzethyltliacet ylene. _ . _ . _ Hexatriene. - - - - - - - - Benzene with banded region shown by shadiw. residual affinity left untouched a t each end of the chain.From this it may be deduced that the presence of free affinity (or partial valencies) tends to increase absorptive power which is borne out by the well-known fact that the introductioii of any unsaturated centre into a molecule enhances the power of absorption 834 MACBETH AND STEWART : A further step in advance is found when the case of dimethyl-diacetylene is considered in conjunction with the foregoing. Dimethyldiacetylene contains two conjugated double linkings and it is reasonable t o assume that t o some extent these will behave like two conjugated double bonds saturating each other to a certain degree but leaving a considerable amount of residual affinity a t either end of the chain. I n the case of the acetylene derivative however there is not the same long chain of conjuga-tion which is present in the case of the hexatriene molecule.The whole residual affinity of the acetylenic compound is concentrated in two centres in close proximity to one another. Thus in hexatriene we find two active centres far removed from each other ; dimethyldiacetylene contains two centres of residual affinity sufficiently near one another to interfere with each other and hence reduce the amount of free affinity present whilst benzene contains no independent centres of partial valency under normal conditions. Further the absorptive powers of the three com-pounds stand in this order also. It may reasonably be suggested that absorptive power is greatest when a molecule contains more than one centre of residual affinity and when such centres are so situated as t o be incapalcle of mutual interference.Under such an assumption it would be expected that of two stereoisomerides for example the fumaroid form would exhibit most absorptive power a conclusion which is borne out by facts. For instance an examination of Fig. 1 C shows that maleic acid is much less absorptive than fumaric acid. An ex-amination of their spatial formulze indicates that in the case of fumaric acid the two accumulations of residual affinity at the ends of the chain are sufficiently far removed from one another to pre-clude mutual interference whereas in the case of maleic acid they are near together. OH OH 1- -1 .....O=C-p-~T H-C-C=O.., I1 I I II H-c-G=O.. ... -1 OH Fumaric acid.Further support of this view will be found in the facts described in Series IX. Series V l . Phe?i,ylacetyleue Styrene (PA e?iylethyle?ze) aud Xthylbenzene (Fig. 4).-In this case the ethylenic derivative exhibits the greatesta power of absorption ; the acetylenic compound stands next in order and the ethane derivative has least absorptiv THE ABSORPTION SPECTRA OF SUBSTANCES ETC. 835 power. These data hold good whether the absorptive potwer be judged by the wave-length absorbed by a given thickness of solu-tion or by the dilution t o which absorption persists. As regards the manner of absorption in the case of ethylbeczene the region of banded absorption in benzene itself is occupied by a broader FIU. 4. Oscillation frequencies. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 O h l d ( C D ~ 0 h l ~ 10 m ? ~ N/10 10 mm.N/100 10 mm. N/1000 10 mm. N/10,000 Phenylacet y lene. _ . _ . - Styrene (phenylethylene). - - - - . - - - Ethylbenxe ne. single baiid and it appears that the introduction of the ethyl radicle into the benzene iiucleus has no marked effect on the character of the absorptive power. There is no increase in the persistence of the band. With styrene (phenylethylene) and phenylacetylene on the other hand there is an increase in general absorptive power as compared with benzene itself which increas 836 MACBETH AND STEWART : can be attributed only to the effect of introducing the new un-saturated centre into the molecule. Tolsne Stilbene and s-Biphenylethane (Fig. 5).-The introduction of a second phenyl nucleus into the molecules mentioned in the last section leads to somewhat Deculiar results.Series VII. FIG. 5. Oscillation frequencies. 48 __ 46 44 42 40 -38 36 34 32 30 28 26 24 22 20 18 16 14 12 10 L -I 10 mm. NjlO 10 mwz. N/100 LO mm. N/1000 LO mm. N/10,000 Tolane. Stilbene. _ . _ . _ _ _ - _ _ _ s - Diphenylet han.e. I n the case of s-diphenylethane the graph shows that banded absorption still persists but that i t is completely modified in character. In the first place the absorptive power is increased as the band appears a t a greater dilution in the s-diphenylethane spectrum than was found t o be the case in phenylethylene. Further instead of a single narrow band s-diphenylethane show THE ABSORPTION SPECTRA OF SUBSTANCES ETC.837 a band extending roughly from 3100 to 4100 although it is much shallower than the band of ethylbenzene and shows a tendency to split up into two smaller bands. One of these narrow bands extends from 3600 t o 4100 and thus corresponds roughly with the characteristic band of ethylbenzene; the other band has its head at 3400 and does not occur a t all in the spectrum of ethylbenzene. Its appearance must therefore be attributed to the introduction of the second phenyl radicle into Lhe molecule. Turning t o the ethylene derivative stilbene it will be observed that i t possesses an absorptive power markedly greater than that exhibited by the ethane analogue. Further there is no sign what-ever of the band characteristic of ethylbenzene; in fact a t this point of the spectrum there is especially good transmission.On the other hand the band atl 3400 traceable in the spectrum of s-diphenylethane is here broadened and deepened to a very remark-able extent. It seems probable that the two bands arise from a common cause but that in the case of stilbene the absorptive power is reinforced by the presence of the long chain of conju-gated linkings seven in number which exists in the molecule of the ethylenic compound. I n their banded absorption the ethylenic and acetylenic analogues are approximately similar but it will be seen from the graph that the band in stilbene extends over a much greater range of dilution than that exhibited by tolane. This fact reinforces the idea that the band of stilbene is in some way influenced by the conjugated chain for in the case of tolane the conjugation is imperfect as compared with that of stilbene.I n stilbene the con-jugated chain is made up completely of double and single bonds, whereas in tolane it contains single bonds alternating with double and triple linkings. Thus instead of having the whole of the free affinity of the molecule concentrated at' two points far removed from each other as in stilbene the tolane molecule has two main concentrations plus some excesses of affinity at the ends of the triple bonds. c ,-CH-CH=CH CH-CH-CH I I I I1 I' CH-CH=C-CH=CH-CC'==CH-CH Stilbene. 'v' - - '-' /- n CH-CH=CH CHLCH-CH 'I i I I CH-CH=C-CHEC-C=-CH-CH u W i ;u \-/ Tolane 838 MACBETH AND STEWART : I n this way the contentions put forward in connexion with Series V.receive further support. Series VZZZ. Phenylpropiolic Acid Ciitrtamic Acid and P-Phertylp-opionic Acid (Fig. 6).-In this series the saturated member of the group shows very much less absorptive power than FIG. 6. Oscillation frequencies. 10 mm. N/100 10 mm. N/1000 10 nwn. N/10,000 Phenylpropiolic acid. - . _ . _ Cinnamic acid. - _ - - - - - - B-Phenylpropionic acid. either of the other two. Traces of a modified benzenoid vibration are to be found in the one broad and two narrow bands which extend from 3700 t o 4200; the dilution a t which these bands occur is approximately the same as that a t which the benzene bands make their appearance THE ABSORPTION SPECTRA OF SUBSTANCES ETC.839 Cinnamic acid and phenylpropiolic acid show a somewhat similar pqwer of absorption through a certain range of the graph, although cinnamic acid absorbs slightly longer wave-lengths whilst phenylpropiolic acid shows im absorptive power a t a dilution much higher than that a t which cinnamic acid ceases to absorb markedly. Further cinnamic acid shows a well-defined band of coiisiderable FIG. 7. Oscillation frequencies. 10 mm. N/100 10 mm. N/lOOO 10 mm. N/10,000 -__- Ethyl phenylpropiolate. _ . _ . - Cinnamylideneacetic acid. persistence in the region 3300-4200 whereas phenylpropiolic acid exhibits a ‘‘ step-out” a t a higher dilution. Testing the absory-tive power by the length of the light-waves absorbed the order is: ethylenic acetylenic and saturated but as regards the power of absorbing a t greatest dilution they stand in the order of satura-tion the acetylenic derivative having most absorptive capacity 840 THE ABSORPTION SPECTRA OF SUBSTANCES ETC.Series I X . Ethzy 1 Y h e n y lproln’ola t e u n’d Cirmamylide?wa c e t ic Acid (Fig. ‘I).-This case furnishes another example of the pheno-menon which was discussed in Series V. The two compounds ester and acid are isomerides so that as far as additive factors go they should exhibit the same absorptive power. An examination of their structural formuke shows that in the one case there is a long chain of conjugated double bonds in the molecule so that the residual affinity is concentrated at two distant points whereas in the other case the affinity centres are nearer together and are more likely to disturb one another: CH-CH=CH.....OH CH-CH=C-CH=CH- CH=CH-C-O-,--l Ii 1- 7 1 ,-.. I _ u Cinnamylideneacetic acid. n CH-CH=CH.-... O*C,H, CH-CH=C-C=C-C=O Ethyl phenylpropiolate. I1 l--l . -. -. ’. ,/‘ u From what has already been said on this matter it may be deduced that cinnam ylideneacetic acid will show the greatest absorption whilst the absorptive power of ethyl phenylpropiolate will be considerably less. An examination of the graphs shows that this prophecy is justified. The absorptive power of the ester is less than that of its isomeride both from the point of view of the wave-length absorbed by a fixed concentration and also from the point of view of the dilution a t which the substance continues to show absorption.Thus although the examination of acetylenic derivatives has not brought to light any close parallel between them and the ethylenic analogues so far as light-absorbing power is concerned, it has elicited new facts with regard to the influence exerted by the distribution of residual affinity in the molecule on the absorp-tive power of the substance. At the present time our knowledge of the subject in its broadest outlines may be summarised ass follows The introduction of a centre of residual affinity into a saturated molecule increases that molecule’s power of absorbing light. The introduction of a second centre of residual affinity into the molecule normally increases the absorptive power although in mme cases (such a’s the iodine compounds) it may act in a con-verse manner. If the two centres of residual affinity me placed so far apart in the molecule either spatially or structurally tha UNIFORM MOVEMENT OF FLAME ETC. 841 they fail to exert any mutual influence on one another then the absorptive power of the molecule is greater than is found to be the case when the same amount of residual affinity is concentrated in two adjacent points of the molecule. THE SIR DONALD CURRIE LABORATORIES, THE QUEEN’S UNIVERSITY OF BELFAST. PHPSICAL CHEMISTRY DEPARTMENT, [Received July 18th 19 17.1 THE UNIVERSITY OF GLASGOW

ISSN:0368-1645    

DOI:10.1039/CT9171100829

出版商:RSC    

年代:1917

数据来源: RSC

摘要:

UNIFORM MOVEMENT OF FLAME ETC. 841 LXX1.-The Uniform Movement of Flame in Mixtures of Acetylene and Air. By WILLIAM ARTHUR HAWARD and SOSALE GARALAPURY SASTRY. IN continuation of the revision of previous work on the propaga-tion of flame t h a t is b’eing carried out a t the Home Office Experi-mental Station under the direction of Dr. R. V. Wheeler we have determined the speeds of the “ uniform movement ” in mixtures of acetylene and air. It may be recalled that the “uniform movement” occurs when an inflammable mixture of gases is ignited a t tho open end of a horizontal tube closed a t the other end and is usually regarded as the normal speed of propagation of flame from layer t o layer of the mixture by conduction of heat (Mallard and Le Chatelier, Ann. des Mines 1883 [viii] 4 374).Previous determinations of the speed of the uniform movement in mixtures of acetylene and air have been made by L0 Chatelier (Compt. rend. 1895 121 1144) who used a glass tube 4 cm. in diameter. The speeds recorded were : Acetylene in mixture. movement. Speed of uniform Per cent. Cm. per second. 2-9 10 8.0 500 9.0 and 10.0 600 22.0 40 64.0 5 I n a later publication namely “ Le Carbone ” (Paris 1908) Le Chatelier gives intermediate values : VOL. CXI. Acetylene in mixture. Per cent. 5.0 7.0 15.0 40.0 60.0 Speed of uniform movement. Cm. per second. 200 400 300 22 7 L 842 HAWARD AND SASTRY THE UNIFORM MOVEMENT OF but it is not clear whether these arc3 actual determinations or numbers interpolated on a curve constructed from the &x results recorded in the Comptes rendus.Le Chatelier thus describes the curve he obtained (Zoc. cit., p. 280): “La courbe . . . prdsente une forme toute sp5ciale; elle se compose de trois droites une droite montamte et une descendante se coupant pour la vitesse maxima vers 10 p. 100 d’acdtylhne puis ensuite une droite tres peu inclinee coupant la Fecond Q la teneur de 20 p. 100 e t se prolongeant jusqu’i la limite d’inflammabilitd supdrieure. Ce troisihme segment de la courbe correspond Q la combustion avec flamme fuligineuse et dBp6t de charbon. Au-dessous de 20 p. 100 il ne se forme par la combustion que des produits gazeux acide carbonique oxyde de carbone et hydrogbne.” Our own experiments were made in a glass tube 12 mm.in diameter and are not therefore directly comparable with Le Chatelier’s so far as the absolute measurements of speeds are con-cerned. It is permissible however to compare the shapes and characters of the two curves. We can confirm the statement that mixtures containing more than about 20 per cent. of acetylene deposit soot owing t o decomposition of excess of acetylene and that the speed of propagation of flame in such mixtures is slow; but we cannot agree that the curve can be represented by gtraight lines. As with mixtures of other inflammable gases and air there is a gradual flattening of the curve towards the limits and the maximum speed is obtained over a range of mixtures containing between 8 and 10 per cent. of acetylene so that the crest of the curve also is flattened.The speed of the flame in mixtures containing more than 20 per cent. of acetylene decreases gradually as the percentage of acetylene is increased. The mixture of acetylene and air for complete com-bustion contains 7.75 per cent. of acetylene; the fact that the fastest speed of flame is found in mixtures containing rather more than this can be explained on the assumption that the thermal conductivity of acetylene is higher than that of air (compare Haward and Otagawa T. 1916 109 83). Our results are shown graphically in the diagram. EXPERIMENTAL. The method of experiment and the electrical means of measuring the speeds of the flames were the same as employed by Wheeler (T. 1914 105 2606). The acetylene was obtained compressed in cylinders without acetone as solvent and was of a high degree of purity (98 to 9 FLAME IN MIXTURES OF ACETYLENE AND AIR.843 per cent. C2H2). The mixtures were made over brine in metal gas-holders of 70 litres capacity and were analysed before use; the explosion tube was filled with the required mixture by dis-placement of air six times the volume of the tube being used. The initial temperature of the mixtures was that of the room ( 1 5 O to 20°) and they were saturated with water-vapour at that temperature. Ignition was by means of a lighted taper which was drawn rapidly past the open end of the tube. The speeds were meamred 2 4 6 8 10 12 14 16 18 20 Acetylene per cent. between two points 40 cm. apart the first point being 10 cm. from the open end of the tube. I n addition to the experiments recorded in the diagram for which a tube 12 mm. in internal diameter was used the speeds in tubes of 9 mm. and 25 mm. diameter were also determined. The speeds in the 9 mm. tube w0re rather slower than those obtained in the 12 mm. tube but the shape of the curve was the same. With the larger tube 25 mm. the duration of the uniform movement was too short to admit of accurate measurement by the means employed. IMPERIAL COLLEGE OF SCIENOE AND TECHNOLOGY, SOUTH KENSINGTON S.W. [Received August 9th 1917.1 L L

ISSN:0368-1645    

DOI:10.1039/CT9171100841

出版商:RSC    

年代:1917

数据来源: RSC

摘要:

844 WERNER METHYLATION BY MEANS OF LXXI1.--Methylation by Means of Formaldehyde. Part I. The Mechanism of the Interaction of Formaldehyde and Ammonium Chloride ; The Preparation of Methylamine and of Dimethyl-amine. By EMIL ALPHONSE WERNER. THE preparation of methylamine from formaldehyde and ammonium chloride is perhaps not so well known as it certainly deserves to be. Having occasion to make use of this reaction from time to time during the last few years the author has studied this interesting and important change the mechanism of which it must be admitted has up to the present remained decidedly obscure. Several new facts have been brought to light which have greatly helped to disclose what is now believed t o be the true mechanism of the progressive changes involved in this rather complex reaction.The interaction of formaldehyde and ammonium chloride was first examined by Plochl (Bey. 1888 21 2117) who from a rather superficial study of the cha.nge observed the formation of tri-methylamine and the evolution of carbon dioxide. No attempt was made to give an explanation of the reaction whereby the amine was produced. Some years later Brochet and Cambier (Cornpt. rend. 1895, 120 449 557; Bull. SOC. chim. 1895 [iii] 13 392) reinvestigated this reaction and showed what were the best conditions f o r the economic preparation of methyinmine hydrochloride by its means. The explanation which they have given of what they considered to be the main reaction is far from satisfactory and they have been decidedly at fault in their observations concerning certain experimental facts connected with the reaction.As a result of their investigations they have drawn the con-clusion that whilst methylamine is almost the sole ,product when ammonium chloride is used in excess trimethylamine is the main product when formaldehyde is in excess. The latter conclusion has not been confirmed. These investigators have evidently mistaken impure dimethyl-amine hydrochloride for the salt of the tertiary base only a small quantity of which has been detected when formaldehyde was used in excess and even then only when a relatively high temperature was attained. As regards their explanation of the reaction Brochet and Cambier have drawn the conclusion that " formaldehyde condense FORMALDEEYDE.PART I. 845 with ammonium chloride in a very complex manner ” ; it has been assumed that trimethylenetriamine hydrochloride (C‘H,:NR,HCl),, is a t once formed as a result probably of the polymerisation of methyleneimine CH,:NH produced in the first instance ; the polymeride then condenses further with formaldehyde whether the latter is present in excess or not to give methylamine hydro-chloride according to the equation 2(CH2:NH,HC1) + 3CH,O + 3H20 = 6CH,*NH2,HC1 + 3COp When the aldehyde is present in large excess they state “on arrive finalement au chlorhydrate du trimethylanline par une serie de rQactions identiques,” that is the compound, (CH, N-CH,,HCl),, is formed which in its turn condenses with kormaldehyde to form carbon dioxide methylamine and trimethylamine.It must be admitted that the above equation still leaves the ques-tion of the origin of methylamine rather obscure let alone the case of t rimet hylamine. More recently Knudsen (Ber. 1914 47 2694) has re-examined this reaction and whilst he has forestalled the author in showing that dimethylamine was produced he has not added any aseful information towards the elucidation of the mechanism by which the amines take origin. Thus it has been assumed that dimethyl-amine and trimethylamine are produced from the decomposition of a complex not t o say doubtful dimethylpentamethylene-tetramine formed in the first instance as a result of the union of methylamine with excess of formaldehyde. I n both cases then obscure condensation reactions have been regarded as explaining the formation of all three amines.This, i t will be shown is quite unnecessary in order to explain the different changes and moreover such a view is in contradiction to the observed facts. Mechanism of the Formation of Methylamiiie and of Dimethylamine. When a solution of ammonium chloride and formaldehyde (com-mercial formalin) in the proportions recommended by Brochet and Cambier was gradually heated a volakile liquid began to distil a t 50°; after heating a t 104O until distillation had prac-tically ceased the amount of distillate obtained was equal to 22 per cent. of the weight of the formaldehyde solut’ion taken. This distillate has been considered by Brochet and Cambier t o consist of methylal and water only; thus-they state that the pro-duct may contain from 60-79 per cent.of the former. Whilst methylal formed a considerable portion of the distillate 846 WERNER METHYLATION BY MEAN8 OF the chief constituent was methyl formate an important fact which Brochet and Cambier have completely overlooked. The composi-tion of thk distillate was fairly constant whether ammonium chloride or formaldehyde was present in an excess a t the outset. When methylamine hydrochloride and formaldehyde were heated in solution 90° was reached before distillation was observed ; the distillate contained methylal and methyl formate in nearly equal proportions; on the other hand from dimethyl-ammonium chloride and formaldehyde after heating under similar conditions a distillate was obtained which contained practically no methyl formate.The following results represenb the average composition of the distillates in the three respective cases : Percentage composition NH,CI + CH,O. of distillate. t = 104". Methyl formate.. .......... = 39.0 Methylal ..................... =34.6 Free formic acid ......... = 1.6 Water ........................ =24*8 CO freely evolved. (11.) NH,MeCl + CH,O. t = 104". 13.10 13.32 3.64 69-94 C02 freely evolved. (111.) N&Me,Cl+ CH,O. t = 110". 0.13 8.70 2-80 88.37 a traoe of CO evolved. Whilst the formation of methylal is a normal result of a reaction between formaldehyde and methyl alcohol (present in commercial formalin) promoted by hydrochloric acid set free during the process the production of methyl formate (and carbon dioxide) in considerable quantity furnishes an important clue to the mechanism of the reaction.The oxidation of formaldehyde to formic acid and. of the latter t o carbon dioxide must in the present circumstances be provoked by the presence of another substance having an equal tendency to undergo reduction. Through the decomposition of water the desired result is attained, a simultaneous oxidation and reduction is brought about and as a result of the latter methylamine and dimethylamine are pro-duced in accordance with the following scheme. The first phase of the change gives rise to methyleneimine, thus : (1) H-COH + NH,(HCl)* -+ H-CH<gEa I CH,:NH(HCl) + H,O It is the great tendency of formaldehyde to react with ammonia that no doubt determines a rapid dissociation of the haloid aalt; * When ammonium chloride and formaldehyde are mixed in solution, the liquid quickly becomes strongly acid from liberation of hydrogen chloride ; the brackets are used here to indicate the dissociated salt FORMALDEHYDE.PART I. 847 the feeble base CH,:NH remains largely dissociated since titra-tion using phenolphthalein as indicator shows all the hydro-chloric acid to be in the free state. Its presence however prevents polymerisation of the base and equilibrium (that is neutralisa-tion) is rapidly established by reduction and oxidation thus : (2) CH,:NH( HCI) + H,iO + H-COH = CH,*NH,,HCl + H*CO,H, and whilst part of the formic acid is neutralised by esterification, the larger portion is oxidised to carbon dioxide and water.As the temperature rises the main reaction (2) is soon accom-panied by a change similar to (l) in which methylammonium chloride takes part whereby dimethylammonium chloride is ultimately formed thus : (3) CH,O + NH2*CH,(HC1) = CH,:N*CH,(HCl) + H,O. (4) CH,:N*CH,(HCI) + H j0 + H*COH = (CH,),NH,HCl + H*CO,H. The next and final phase in the change is the result of a reaction between formaldehyde and dimethylammonium chloride thus : ( 5 ) CH,O + 2NH(CH3),(HCl) = C H 2 < ~ [ ~ $ 3 3 2 v ~ ~ ~ 0 + H,O. and since a stable saturated base is produced i t will show no tend-ency to suffer reduction and consequently there will be no oxida-tion of formaldehyde during this phase as shown by the results obtained under 111. I n accordance with this scheme theref ore trimethylamine cannot be directly formed during the progress of the changes which give rise to the primary and secondary bases and this explains why i t has not been detected in the relaction product when the temperature was not allowed to rise above say l l O o .If on the other hand the temperature be carried too high as was the case in Knudsen's experiments (Zoc. cit.). or the heating be unduly prolonged a condition which can scarcely be avoided in dealing with the final mother liquors after the separation of the bulk of the chlorides of methylamm~onium and of dimethylammon-ium then some trimethylamine is undoubtedly produced. The evidence goes t o show that the tertiary base very probably arises from the decomposition of the above methylene base thus : (6) CH,<N(CH3)2,2HCl = N(CH,),,ECI + CH,:N*CH,(HCI).N(CH,), The unsaturated base in the absence of excess of formaldehyde is polymerised to the compound (CH2:N*CH&,* the presence of which * This base has been prepared by Henry (Bull. Acad. TOY. Belg, 1893, [iii] 26,200) and later by Brochet and Cambier (Zoc. c i t . ) who determined its molecular weight 848 WERNER METHYLATION BY MEANS OF can be shown by the formation of a copious precipitate on addition of picric acid. The picrate (m. p. 1 2 7 O Duden and Scharff Ber., 1895 28 936) cannot be crystallised from water on account of the eaw with which it dissociates. Since the changes represenbd by equations (2) and (4) overlap during the progress of the reaction the production of dimethyl-ammonium chloride in moderate quantity cannot be avoided.Its separation from a considerable amount of methylammonium chloride is however a very simple matter as dewribed later on. By adopting such conditions as were indicated by the above scheme a very large yield of dimethylammonium chloride has been obtained with the use of much less formaldehyde than was found necessary by Knudsen (Zoc. cit.). The absence of condensation pro-ducts in the early stages of the reaction that is both before and after large quantities of methylammonium and dimethylammonium chlorides have been produced has been proved by the fact that no precipitation was produced on the addition of picric acid. This reagent forms sparingly soluble compounds with all the condensa-tion products or polymerides which have hitherto been supposed to play a past in the formation ,of the1 primary and secondary bases produced in this interesting and important reaction.Preliminary experiments have proved the wide scope of tlie use of formaldehyde for inethylating amino-compounds of various types, o n the lines of the scheme just recorded. These it is to be hoped, will be described in the near future. E x P E R I M E N T A L. The P r e p r a t i o n of Methylammoiiiuni Chloride. The proportions of ammonium chloride and formaldehyde (40 per cent. formalin) * recommended by Brochet and Cambier (Zoc. cit,), namely one part by weight of the1 former and two parts by weight of the latter were found after several trials to give the best results.Since about 35 per cent. of ammonium chloride has always been recovered unchanged the1 molecular ratios NH4C1 2CH,O required by theory are very closely represented by the above proportions. Expt. I.-Two hundred and fifty grams of ammonium chloride and 500 grams of formaldehyde solution were gradually heated in a distillation flask which carried a thermometer with the bulb well below the surface of the liquid. The temperature was slowly raised to 1 0 4 O and was not allowed to rise above this point a t which it * Analyses of seven different samples of commercial formalin gave as a mean result 35 per cent. of formaldehyde and in no case was a sample found to contain 40 per cent. The highest value was 37.4 per cent. the lowest 33-2 per cent FORMALDEHYDE.PART I. 849 was maintained until no more volatile liquid distilled ; this required about four and a-half hours. The distillate weighed 110 grams. The product was allowed t o cool and after filtration from 62 grams of ammonium chloride which had separated was concentrated by evaporation a t looo to about one-half of the original volume. After removal of 19 grams of ammonium chloride,* the liquid was again concentrated by evaporation until a crystalline Scum had formed on thel surface of the hot solution. After cooling 96 grams of methylammonium chloride (Found, C1= 52.46. Calc. C1= 52.59 per cent.) were separated; after further concentration a second crop (18 grams C1= 52.39 per cent.) was obtained. The filtrate was now concentrated as far as possible a t looo and was left for twenty-four hours in a vacuum oveir sodium hydroxide after which the semi-solid residue was digested with chloroform when 20 grams of methylammonium chloride (C1= 52.63 per cent.) which had been washed with chloroform to remove dimetliylammonium chloride were obtained.The total yield was 128 grams. From the chloroform solution after removal of much of the solvent by distillation 27.5 grams of dimetliylanimonium chloride1 were obtaine'd (Found C1= 44.38. Calc. C1= 43.5 per cent.). A visoous residue (76 grams) which did not crystallise after remaining for a week in a vacuum over sulphuric acid was finally obtained; it contained C1=40*37 pelr cent. It was distilled after the addition of an excess of a 40 per cent. solution of sodium hydr-oxide and the alkaline vapours evolved were absorbed in an alco-holic solution of hydrochloric acid when a small quantity of methyl-ammonium chloride and a relatively large quantity of dimethyl-ammonium chloride were! obtained but no trimethylammonium chloride could be detected.Formaldehyde was also regenerated by the action oi sodium hydroxide on the viscous material which no doubt contained much tetramethylmethylenedianiine hydrochloride, C13,(NMe2),,2HCl which requireis C1= 40.57 per cent. Examination of Volatile Liquid Distillate.-This had D15 0.927. Five C.C. after digestion with 50 C.C. of IV-dium hydroxide required 18.15 C.C. of N-sulpliuric acid f o r neutralisation. Five c.c. after removal of methyl formate and methylal by heat a t 70° left an acid liquid which required 1.65 C.C.of N-sodium hydroxide for neutralisation. Hence H*CO,Me = 39.09 ; CH,O = 1.62 per cent. The proportion of methylal was determined by differ-ence after removal of water by anhydrous calcium chloride. The * Ammonium chloride is very sparingly soluble in a concentrated solution of methylammonium chloride and consequently its separation from the latter salt is very sharp. L L 860 WERNER METHYLATION BY MEANS OF separation o i methyl formate from methylal by fractional distilla-tion was found t o be an extre8mely tedious process and was aban-doned as useless from an economic point of view. The value of the distillate is of iriiportance since all the formic acid can be easily recovered as sodium formate after shaking with a solution of sodium hydroxide in the cold and thus separated from methylal.The following results illustrate the value of the whole process from an economic point of view. From an experiment with 4000 grams of formalin and 2000 grams (of ammonium chloride and without working up the final viscous residue therei were obtained 1037 gram of pure methylammonium chlocride 218 grams of nearly pure dimethylammoniuin chloride 408 grams of anhydrous sodium formate and 264 grams of pure methylal (b. p. 42-43O) whilst 698 grams of ammonium chloride were recovered. The' yield of methylammonium chloride was equal to 79.6 per cent. of the weight of ammonium chloride which had entered into' reaction. When the value of the by-products is taken into consideration it will be seen that the methylammonium salt is obtained for a very small outlay.Pr epura ti o n of Dime thy la mmo 7 1 iu m C h lo ride. The formation oi the above salt in this reaction has been recently pointed out by Knudsen (Zoc. c i t . ) but the method adopted for its preparation distinctly shows the absence of a reasonable apprecia-tion of the probable mechanism of the changes. Thus in an experi-ment designed with the object of obtaining the best yield of the secondary amine a useless not to say a wasteful excess of form-aldehyde was employed without any particular advantage. The following experiment carried out on the lines of t.he present theory, gave a very good result. E'xpt. 11.-Two hundred grams of ammonium chloride and 400 grams of formalin were heated t o 104O as in Expt.I and 65 grams of ammonium chloride were recovered. To the filtrate 300 grams of formalin were now added and the solution was again heated at this stage to 1 1 5 O and maintained as nearly as possible a t this hi-p r a t u r e until no more liquid distilled. This required about three and a-half hours. Since methylammonium chloride produced during the first stage is less easily dissociated than ammonium chloride a higher temperature was required to bring about reac-tion (3). It was noticed that whilst a volatile liquid commenced to distil a t about 52O in the first stage 92O was reached in the second stage before any liquid distilled which is quite in agreement with theory. The product was concentrated by evaporation a t looo until a scum appeared on the surface of the hot liquid; 7 grams of ammon FORMALDEHYDE.PART I. 851 ium chloride and 27 grams of pure methylammonium chloride were recovered from the material which had separated after cooling. The product was now heated t o 120° until a portion when cooled became a semi-solid crystalline mass after which it was allowed t o remain for two days in a partial vacuum over sodium hydroxide. It was then treated with chloroform as described under Expt. I and 122 grams of nearly pure dimethylammonium chloride (Found, C1= 43.14. Calc. C1= 43.5 pel* cent.) were ultimately obtained. The final residue contained some trimethylammonium chloride but was not further dealt; with. The yield of dimethylammonium chloride calculated on $he weight of ammonium chloride which had entered into reaction (that is, 200-72=128 grams) was therefore 95.3 per cent.with the use of 700 grams of formaldehyde solution. Knudsen obtained a yield of 70 per cent. from 100 grams of ammonium chloride and 1000 grams of formalin. Production of Tm'methylaminoizium C'hloride from the Iuteraction of For maldeh yde and Di in e t lip lnm m o 11 i u ni C'?i lo ride. I n order to prove the origin of trimetliylamine in accordance with the present theory the following experiment was made. Expt. ZI1.-Dimethylammonium chloride (20.5 grams) and 45 grains of a solution of formaldehyde (molecular ratio 1 :2) were heated in a distillation flask t o l l O o f o r four hours. The composi-tion of the distillate which weighed 18 grams is given under I11 (p. 846).The product after concentration as far as possible by evaporation a t 100° was heated to 120° after which i t was allowed to remain over sodium hydroxide as in Expt. 11. The residue was dissolved in chloroform and on addition of ether (well dried) 4.5 grams of crystals were precipitated which contained C1= 36-93 (C,HgN,HC1 requires C1= 37.17 per cent.). A platinichloride was prepared which contained Pt = 37.03 (C,HgN,H2PtC1 requires Pt = 36.95 per cent.). The residue after removal of the solvent was distilled with a 40 per cent. pot,assium hydrotxide solution ; t h s alkaline distillate, which possessed a strong odour of formaldehyde was easily proved t o consist chiefly of dimethylamine with only a small proportion of the tertiary base. The original reactio'n product tetramethyl-methylenediamine hydrochloride was readily hydrolysed when heated with a solution of potassium hydroxide thus, CH2(NM%),,2HC1+ 2KOH = CH20 + ZNHMe + 2KC1+ H20.The small yield of trimethylamine was due to the relatively low temperature attained during the process. L L* 852 METHYLATION BY MEANS OF FORMALDEHYDE. PART I. E s p t . IV.-The above was repeated but t h e temperature was raised to 160° after evaporation a t looo; a yield of 14 grams of trimethylarnmonium chloride was obtained thus proving the 'origin of the tertiary base as shown in the' schemel equation (6). Bekaviour of :~~et~i'ylanzmoniz~m Chloride and Formaldehyde i n t h e Presence of Ethyl Alcohol. According to theory no inethylammonium chloride should result from the interaction of ammonium chloride and formaldehyde in the abmnce of water a point which is not suggested by such an equation as 2CH,O + NH = CH,*NH + H*CO,H which is given by Knudsen t o show the formation of methylamine.The insolubility of amnionium chloride in pure alcohol preseats a difficulty in the use of this salt; however the test has been made with the methyl derivative . Expt. V.-Seventeen grams of inethylammonium chloride and 15 grams of paraformaldehyde (2 niols.) were heated with 50 C.C. of alcohol under reflux. The aldehyde was rapidly depolymerised and a clear homogeneous solution was obtained as soon as the boil-ing point of aicohol was reached. After a short time the liquid gradually separated into two layers and the change was completed after one hour.The lower layer when cold was a semi-solid crystal-line mass from which 10.5 grams of methylammoniuni chloride were recovered. The supernatant liquid was shaken with a saturated aqueous solu-tion of calcium Chloride dried and distilled; 46 grams of ethylal (b. p. S8-SS0) were obtained which was equal to 88.4 per cent. of the theoretical from 15 grains of formaldehyde. When molecular proportions of methylammonium chloride (1 7 grams) and paraform-aldehyde (7.5 grams) were used 13.5 grams of the amine salt were recovered and 20 grams of ethylal were obtained. No dimethylammonium chloride was formed which bears out the part played by water in the general reaction. Ititeractign of Formaldehyde and Animonizcm Chloride in t h e Presence of W a t e r alone.It was pointed out in the results given under Ex@. I t h a t the yield of mehhylammonium chloride was equal to 79.6 per cent. of t h e weight of ammonium chloside which had entered into reaction. Theoretically from equations (I) and (2) one molecular proportion of ammonium chloride should yield one of methylammonium chloride t h a t is 126.1 parts of the latter salt from 100 parts of the former. With the use of commercial formalin there is inevitabl DRAKELEY LIBERATION OP HYDROGEN SULPHIDE ETC. 853 loss of formaldehyde as methylal which for obvious reasons cannot be profitably counteracted by using an exoess of the aldehyde solu-tion. This loss of aldehyde) is undoubtedly one of the factors t h a t affects the ultimate yield of methylamine; when it was eliminated bv the use of paraformaldehyde a larger yield of the amine was obtained.Expt. VZ.-Twenty-seven grams of ammonium chlo,ride 30 grams of paraformaldehyde (molecular ratio 1 :2) and 80 C.C. of water were gradually heated. A t SOo a clear solution was obtained and the temperature was maintained for four hours a t 1 0 4 O . Slightly more than one-third (9.06 grams) of the ammonium chloride was recovered whilst 18-96 grams of pure methylammonium chloride were obtained. This equals 105.6 parts from 100 parts of am-monium chloride. The amount af dimethylammonium chloride produced was not estimated. It is not suggested from the results of this experiment that paraforinaldehyde could be economically used on a large scale with advantagel since quite apart from its relatively high cost neither formic acid nor part of the unchanged aldethyde can be recovered as by-products. The experiment has served to support the views put forward and perhaps on a small scale may have some advantage. Many 9ther points dealing chiefly with the identification of the intermediate products which have not been touched upon in the present paper will be elaborated in a future communication. UNIVERSITY CHEMICAL LABORATORY, TRINITY COLLEGE DUBLIN. [Received July 3 lst 19 17.

ISSN:0368-1645    

DOI:10.1039/CT9171100844

出版商:RSC    

年代:1917

数据来源: RSC

摘要:

DRAKELEY LIBERATION OP HYDROGEN SULPHIDE ETC. 853 LXXIK-The Liberation of Hydrogen Subhide from Gob Fires in Coal Mines. By THOMAS JAMES DBAHELEY. EXPERIMENTS described in an earlier paper (T. 1916 109 723) have shown that iron pyrites favourably influences to a small extent the oxidation and bherefore spontaneous ignition of coal. Lewes (Second Report of the Royal Commission on Coal Sup-plies 1904 [Cd. 19911 Vol. 2 232) however expressed the view that iron pyrites has no connexion whatever with the spontaneous heating of coal. This inference is based on the fact that in an inspection of a case of spontaneous combustion where the coal was heated the sulphur was found to be evolved as hydrogen sulphid 854 DRAKELEY THE LIBERATION OF HYDROGEN SULPHIDE and not as sulphur dioxide as would have been the case “ i f the pyrites had anything to do with” the fire.The fact that hydrogen sulphide is evolved during t’he initial stages of a gob fire is indisputable; but the assumption that if iron pyrites favoured the heating the sulphur would be given off necessarily as sulphur dioxide appeared to need verification. There is the reasonable possibility that the suiphur dioxide may be reduced to hydrogen sulphide by contact with the heated coal. Experiments were conducted to investigate this question and a detailed account is given in the following pages. The investdgation has shown that hydrogen sulphids may be formed from heated coal and iron pyrites in a number of ways, as for example by heating coal by passing sulphur dioxide over heated coal by heating mixtures of coal with sulphur or iron pyrites by passing water vapour or hydrogen over heated iron pyrites etc.Therefore at the seat of a gob fire it would appear to be quite possible for sulphur dioxide to be formed but previous to an external outbreak of the fire there would be very little likelihood of this sulphur dioxide being liberated into the mine atmosphere. If the sulphur dioxide were not reduced completely to hydrogen sulphide by contact with heated coal in the immediate vicinity of the fire it would encounter the larger volumes of hydrogen sulphide that would be issuing from the gradually heated material in the locality and thereby would be decomposed to give sulphur. I n consequence of such reactions no sulphur dioxide would reach the external air.I n view of this Lewes’s statement that if iron pyrites assisted in the ignition of the coal the sulphur would be liberated as sulphur dioxide seems t o be erroneous. The Formation of Hydrogen Sulphide b y Passing Sulphur Dioxide over Heated Coal. I n order to reproduce to a certain extent the conditions of the liberation of the gas from the gob a moist mixture of four volumes of carbon dioxide and one volume of sulphur dioxide was passed over 20 grams of powdered coal contained in a glass tube. The tube (45 cm. long and 2 cm. in diameter) was gradually heated in a 28 cm. platinum-wound electric furnace until the temperature reached 500O. The temperature was maintained a t 500° until no further evolution of gas from the coal could be detected.During this time the mixture of carbon dioxide and sulphur dioxide was passed over the coal at the rate of 2 litres per hour FROM GOB FIRES IN COAL MINES. 855 The gases evolved were delivered first into a Woulfe’s bottle, where tarry oils were deposited through a spiral tube immersed in water a t 100 and then through pure concentrated sulphuric acid. After this treatment the hydrogen sulphide in the gases was absorbed by passing them through a train of four bottles each containing a solution of copper sulphate acidified with hydrochloric acid. A duplicate absorption train was arranged so that by means of a three-way tap the gases could be diverted from one train to the other. At the commencement of an experiment the air in the glass tube in the furnace was displaced by a stream of carbon dioxide.The heating was started and the mixture of carbon dioxide and sulphur dioxide was passed through the tube. At first no pre-cipitate appeared in the copper sulphate solution but as Boon as the slightest trace was discernible the tap was turned so that the gases were washed with fresh copper sulphate solution in the second absorption train. This was t o eliminate any possible errors that might arise from a reaction between the hydrogen sulphide and any soluble gases with which the copper sulphate solution had been contaminated during the initial stage of the experiment. After the temperature of the furnace had been 500° for about two hours the mixture of carbon dioxide and sulphur dioxide was turned off to test whether the evolution of volatile products was continuing.When this ceased the precipitated copper sulphide was collected washed dried and detached from the filter paper. The filter paper was ignited and the ash was added to the main part of the precipitate which was mixed with sulphur heated t o redness in an atmosphere of coal gas cooled and weighed. From the weight of cuprous sulphide the quantity of sulphur evolved as hydrogen sulphide was calculated. Blank experiments (in which only carbon dioxide was passed over the coal) were made for each sample of coal. Three samples of different coals were used in the experiments. Coal “ A ” was a selected sample of Arley coal from a local The coal was black possessing considerable lustre and a It fractured along lines chiefly defined by “mother The seam is not subject to spontaneous com-The sample was picked carefully and all visible impurity Coal “B ’’ was a selected sample of “Five Feet ” coal from a The coal was moderately dull black and The fracture was mainly From its properties it would appear to be a canneloid colliery.brown streak. of coal ’’ (charcoal). bustion. excluded. colliery near Chester. was exceptionally close and compact. splinty 856 DRAKELEY THE LIBERATION OF HYDROGEN SULPHIDE coal. The sample was freed carefully from all visible impurity. Coal “ C ” was from a local colliery and was a deliberately chosen inferior sample from the Ravine Seam which is subject’ to spontaneous combustion. The coal consisted of alternate lustrous black and comparatively dull black layers; it fractured easily along planes which contained considerable deposits of calcium carbonate.Iron pyrites was present in the finely disseminated form. The sample was obtained from the neighbourhood of a gob fire. The seam is not subject t o gob fires. Analyses of the coals gave the following results: Estimation. Specific Gravity. Ultimate Analysis :-Carbon .............................. Hydrogen ........................... Nitrogen ........................... Sulphur .............................. Ash ................................. Oxygen etc. (by diff .) ............ Coal “ A.” 1.248. Per cent. 80.937 4.815 1.365 1.412 1.660 9.81 1 Coal c c B.” 1.247. Per cent. 75-761 5.783 1.487 0.953 3.646 12.350 Coal “ C.” 1.341.Per cent. 67.730 2.992 1.015 4-017 12.084 12.162 Proximate Analysis :-Fixed Carbon ..................... Ash .................................... Moisture ........................... Volatile matter ..................... Yield of Volatile Matter on heating to 500” .................. 100~000 66.475 1.660 0-828 31-037 100~000 14.421 100~000 47.606 3-646 0-566 48-182 100~000 25.830 100*000 51.637 12.084 3.738 31.541 100~000 20.516 The results of the experiments are given in the accompanying table and are calculated for a weight of 100 grams.: Weight of sulphur in Result. Sample. Time. coke residue. Mins. Grams! 1.352 1.351 1.373 1.379 I : Coal “ A.” f l 0.618 0.609 0-711 11 1; ) Coal C.” 2.878 2.879 4.186 Weight of as H2S.Grams. 0-032 0-035 0.172 0.194 sulphur evolved 0.049 0.049 0.316 0.351 1.096 1.098 2.065 2-619 In the above table each result represents the average of five experiments FROM GOB FIRES IN COAL MINES. 857 The time mentioned in the second column gives the number of minutes taken to raise the temperature of the furnace from 15O t o 500O. Subsequently the temperature of the furnace was main-tained a t 500° until the end of the experiment which occupied from two to two hours and a-half. Results 1-4 were obtained with coal '' A," 5-8 with coal " B," and 9-12 with coal '((2.'' Results 1 2 5 6 9 10 were blank determinations and give the weight of sulphur that was eliminated as hydrogen sulphide when 100 grams of the coal were heated in a current of moist carbon dioxide.Results 3 4 7 8 11 12 were obtained by passing the mixture of carbon dioxide and sulphur dioxide over the heated coal. I n the blank experiments the temperature of the furnace reached the following values before any appreciable quantity of hydrogen sulphide was evolved from the samples of coal. Coal '' A." Coal " B." Coal " C." Temperature ............... 290" 270" 275" It was observed however that when the mixture of carbon dioxide and sulphur dioxide was passed over the coal no hydrogen sulphide was evolved until a much higher temperature had been reached. The values are given below. Coal '' A." Coal " B." Coal " C." Temperature ...............440" 435" 445" During this time the coal in the tube appeared t o have become coated with a whitish-yellow powder. No doubt this was a slight deposit of sulphur which probably had been formed by the reac-tion between the sulphur dioxide passing over and the hydrogen sulphide evolved from the coal. Possibly the organic tarry liquids from the coal may have influenced the interaction of the hydrogen sulphide and the sulphur dioxide (compare Klein J . Physical Chem. 1910 15 1 who also showed that water was a catalyst). When the temperature of the furnace reached about 450° the evolution of hydrogen sulphide became so rapid that the stream of sulphur dioxide was insufficient to react with it completely and a rapid precipitation of copper sulphide occurred.It may be noted that a deposit of sulphur is observed frequently in the initial stages of an actual occurrence of a gob fire (compare Henshaw Departmental Committee on Spontaneous Combustion in Coal Mines Minutes of Evidence 12th Feb.-16th July 1913, p. 26); but this slight yellow deposit in the cracks of the coal, where the heating is evidenced may be due in addition t o other causes than the one stated above. However the similarity between the actual gob fire and the experiment was striking 858 DRAKELEY THE LIBERATION OF HYDROGEN SULPHIDE The percentage of sulphur in the coke residue which was slightly more friable when sulphur dioxide was passed over the coal would indicate that a quantity of t-he sulphur becomes fixed in the coke.This may be due t'o an interaction between the sulphur dioxide and the ash constituents. It is possible that if the mineral matter of the coal contained lime at 500° a little sulphur may have been formed by the action of the sulphur dioxide on the lime (Veley T. 1893 63 821; com-pare Hammick this vol. 379). The results indicate that the evolution of the hydrogen sulphide is increased where sulphur dioxide is passed over heating coal. The observed fact that the sulphur is evolved as hydrogen sulphide and not as sulphur dioxide from a gob fir0 does not therefore decide the question whether or not iron pyrites assists in the spontaneous ignition of coal. The Formation of Hydrogen Sulphide by Heating Mixtures of Coal and Sulphur. From the results obtained in the previous section it would appear that the production of hydrogen sulphide resulted from heating the coal with the sulphur which previously had been deposited upon it.To investigate this question mixtures of 20 grams of coal with various proportions of sulphur were heated in a glass tube in an electric furnace to 500O. I n the first series of experiments a stream of moist carbon dioxide was passed through the tube and was interrupted only to test whether all the volatile gases had been liberated from the coal. I n the second series of experiments the air was displaced from the glass tube containing the mixturs of coal and sulphur by means of a stream of dry carbon dioxide. The carbon dioxide was turned off until all volatile matter had been expelled from the mixture, then the residual gases were swept into the absorption train by means of the current of dry carbon dioxide.Dry carbon dioxide was driven regularly over the heated mix-tures in the experiments comprising the third series. The current of carbon dioxide was interrupted only to test for the cessation of the evolution of gas from the mixtures. The apparatus and the method of collecting the hydrogen sulphide have been described in the previous part of the paper. The following table gives the weight (in grams) of sulphur that was eliminated as hydrogen sulphide and the weight of sulphur in the coke from mixtures of 100 grams of coal with different quaiiti-ties of sulphur FROM GOB FIRES IN COAL MINES. 859 Result. ’i 5 6 7 8 9 10 11 12 13 14 Weight of sulphur Series 1.mixed with 100 grams S in S as Sample. of coal. coke. S S . 0 1.352 0.032 1 2-373 0.801 2 2.374 1.652 3 1.371 2.302 4 1.379 3.109 5 1.370 4.036 6 1.373 4.783 Grams. Grams. Grams. Series 2. Series 3. - - S in S as S in S as coke. H,S. coke. H,S. Grams. Grams. Grams. Grams. 1.361 0.031 1.363 0-031 1.394 0.605 1.394 0.515 1.401 1.151 1.391 0.997 1.393 1.572 1.404 1.354 1.402 2.095 1.395 1.736 1-401 2.097 1.397 1.831 1.395 2.103 1.403 1-987 Coal “ B.” 19 20 21 0 0.608 0.049 0.618 0.049 0.619 0.049 1 0.613 0.872 0.617 0.693 0.623 0.604 2 0-618 1.801 0.835 1.502 0.871 1.025 3 0.617 2.673 1.201 1.712 1-114 1.513 4 0.615 3.397 1.439 2.383 1.503 2.007 5 0.612 4.413 1.423 2.524 1-501 2.188 6 0.617 5.216 1.457 2.591 1.512 2-436 0 2.877 1.098 2.878 1.097 2.905 1.097 1 3.054 1,897 3.082 1.803 3.085 1.802 2 3.236 2-633 3.298 2-516 3.284 2.517 3 3.509 3.189 3.611 3.167 3.627 3-091 4 3.605 4.314 3.828 3.591 3.971 3.388 5 3.892 4.846 3.909 3.807 4.426 3.645 .6 3.955 5-684 4.136 4.070 4.518 3.923 It is well known that sulphur below its boiling point combines with hydrogen and when heated to 200° reacts with paraffin or vaselin forming hydrogen sulphide. Even if little free hydrogen is evolved from coal a t 500° the action of sulphur on bituminous coal would explain the production of hydrogen sulphide. Evidence that the coal behaved differently when heated with sulphur was noted by the change in the character of the coke. When coal “ A ” was heated alone to 500° a firm hard coke with submetallic lustre was produced but as the quantity of sulphur admixed with the coal increased the coherence of the coke residue diminished.Indeed in the experiments of series 1, when the coal was heated with 6 grams of sulphur the resulting coke was quite pulverulent. The cokes were less powdery in series 2 than in series 1 and in series 3 than series 2. The above state-ments were true also for coals (‘ B ’’ and (‘ C.” The influence of moisture on the production of hydrogen sulphide was pafticularly noticeable (compare Jones Mem. Munchester Phil. SOC. 1904 48 No. XVI.). I n the experiments in series 1 in which m,oisture was admitted the sulphur evolved as hydrogen sulphide exceeds the quantity collected in the corresponding ex-periments in series 2 and 3.In series 3 dry carbon dioxide was passed continually over the mixtures and the production of hydrogen sulphide was the srnalIest 860 DRAKELEY THE LIBERATION OF HYDROGEN SULPHIDE It is impossible however to state definitely how far moisture is responsible for the formation of hydrogen sulphide as water is one of the decomposition products on heating coal and therefore its elimination is incapable of being accomplished. That it is a very important factor is demonstrated by the greater production of hydrogen sulphide in series 1 than in either series 2 or 3. Furthermore with coal ‘(C,” which contained more than 3 per cent. of moisture the difference between the results in series 2 and 3 and those in series 1 is not so marked as with the compaxatively dry coals ‘‘ A ” and ‘‘ B.” That moisture is probably not the sole factor is substantiated by the production of hydrogen sulphide on heating such a mixture as sulphur and paraffin.The decomposition of the moisture by the heated sulphur would produce oxygen and hydrogen sulphide and the loss of the caking power would appear to be explained satisfactorily as the result of the oxidation to inert substances of those particular compounds which endow the coal with this property. This was upheld by the fact that the coke increased in hardness from series 1 t o 3 that is, as the moisture admitted t o the reaction was diminished. No attempt was made to determine1 whether any carbon dioxide was reduced by the! heated coal t o carbon monoxide (see Meyer and Schuster Ber.1911 44 1931). The weight of sulphur fixed in the ooke showed no marked varia-tion with the different mixtures containing coal “ A ”; but the cokes from coals “ B ” and ‘‘ C ” exhibited a gradually increasing fixation of sulphur. The presence of moisture tends to reduce the quantity of sulphur left in the coke. During each experiment the’re accumulated a precipitate of sulphur in the sulphuric acid wash-bottle. This was noticeable particularly with the oolal mixtures containing the larger quanti-ties of sulphur. A t the conclusion of an experiment thel sulphuric acid was found to emit a pungent and disagreeable garlic-like odour. The experiments indicate that where mixtures of sulphur and coal are heated the sulphur is evolved largely as hydrogen sulphide and that$ its formation is increased by the presence of moisture.Tlmse conditions are fulfilled admirably during the initial stage of a gob1 fire’; for sulphur becomes mixed with the coal moisture is present and the temperature gradually rises. Therefore it is n o t surprising that hydrogen sulphide is liberated from the gob and that sulphur dioxide is not observed until the fire breaks externally. Incidentally the author may add that he has been assureid by one wh’o! has had experience of gob fires that the odour produced in the room by heating coal with sulphur is almost identical with the peculiar and characteristic ‘‘ gob stink. FROM GOB ITIRES IN COAL MINES. 861 Hence it may b s argued t h a t such a series of reactions as the following take place in the gob.First iron pyrites assists the ignition arid liberates sulphur dioxide. The materials i n the locality are heated and evolve hydrogen sulphide which reacts with tlie sulphur dioxide t o form sulphur. The sulphur is deposited on the coal which is being heated gradually and froin this mixture the sulphur is disengaged as hydrogen sulphide. It cerkainly cannot be inferred t h a t iron pyrites has no influence on the spontaneous ignition merely because hydrogen sulphide and i7ot sulphur dioxide is liberated from the gob. The Formation o f Hydroyen Sulphide hy Hpntiiig L7liztures of Coal ccnd IroiL Pyrites. The experiments in this section were niade in the same manner as those just described. The mixtures of coal and iron pyrites were heated in a glass tube in an electric tube furnace to 500O.Previouy t o the heating the air was displaced from tlie tube by a stream of carbon dioxide whilst a t the conclusion the residual gases were swept into the1 absorptioa train contzining acidified copper sulphate solution by again turning on the current of carbon dioxide. The analysis of the iron pyrite’s gave the following results: Silica ................................. 1.41G per cent. Iron .................................... 45.902 y y y y Sulphur .............................. 5 1.97 8 , Copper .............................. trace The weight of sulphur evol-”-ed as hydrogen sulphide from various inixtlures of coal and iron t a b.1 el. Result. Sample. 1 ‘1 5 6 Coal ,‘ A ” 9 Coal “ B ” 12 13 Coal “ C ” 17 18 pyrites is given in the accompanying Weight of iron pyrites mixed with 100 grams of coal.Grams. 0 2 4 6 8 10 0 2 4 6 8 I 10 0 2 4 6 8 i 10 Weight of sulphur evolved as hydrogen sulphide. Grams. 0.031 0.208 0.387 0.533 0.666 0.876 0-049 0.604 0.685 0.789 0.901 1.013 1.089 1.620 1.953 2- 194 2.576 3.05 862 DRAKELEY LIBERATION OF HYDROGEN SULPHIDE ETC. Hence the liberation of hydrogen sulphide from a gob fire may be due to a certain extent to the effect of the coal being heated in the presence of iron pyrites. The Formation of Hydrogen Sulyhide by Passing Moist C'arbon Dioxide over Heated Zron Pyrites. A stream of dried carbon dioxide was passed over iron pyrites heated as in the previous experiments to 500° in a n electric tube furnace and no hydrogen sulphide could be detected in the issuing gases.When however a current of moist carbon dioxide was passed ovelr the heated iron pyrites it resulted in the immediate prolduc-tion of hydrogen sulphide (compare Jones Zoc. cit.). A consider-able accumulation of sulphur collected in the cooler parts of the glass tube. This experiment indicates that some of the hydrogen sulphide liberated from a gob fire may owe its origin to the action of mois-ture on heated iron pyrites. I n the gob there would be oxidisable niaterial which would acquire with avidity the oxygen produced by the decomposition of the moisture and would thereby tend to promote the formation of the hydrogen sulphide.The Formation of Hydrogen Sulphide b y Passing Hydrogen over Heated Iron Pyrites. For the experimefnts in this part 2 grams of iron pyrites were placed in each of two tubes (48 cm. long and I cm. in diameter), which were introduced simultaneously into an electric tube furnace a t 500O. Through one tube was passed a stream of dry hydrogen a t the same rate as moist hydrogen was driven through the other. The gases leaving the tube were delivered direct into acidified copper sulphate solution. The precipitated copper sulphide was collected every fifteen minutes and the sulphur estimated as previ-ously described. The quantity of sulphur emitted as hydrogen sulphide is given in the following table [under (aj when dry hydrogen was passed, Time. Sulphur evolved as hydrogen sulphide Minutes. 15 30 46 60 75 90 105 120 (a) Grams. 0.436 0.608 0.705 0.753 0.773 0.792 0.796 0.801 ( b ) Grams. 0.779 1-140 1.388 1.518 1.590 1.628 1.632 1.63 WERNER THE CONSTITUTION OF CARBAMIDES. PART IV. 863 under ( b ) when wet hydrogen was used. The results are calculated for a weight of 10 grams of iron pyrites.] The values clearly demonstrate the paramount importance of moisture in the reduction of iron pyrites by hydrogen. I n the coal mine the gob may reach a high temperature when hydrogen will form a considerable proportion of the gases emitted from the coal (Burgess and Wheeler T. 1910 97 1917). I n these circumstances should it pass over heated iron pyrites, hydrogen sulphide would constituhe one of the gases emitted from the gob. The author's thanks are due t o Mr. George William Farmer for assistance in the last series of expmiments. THE CHEMISTRY DEPARTMENT, MINING AND TECHNICAL COLLEGE, WIGAN. [Received July 14th 191'7.

ISSN:0368-1645    

DOI:10.1039/CT9171100853

出版商:RSC    

年代:1917

数据来源: RSC

摘要:

WERNER THE CONSTITUTION OF CARBAMIDES. PART IV. 863 LXX1V.-The Constitution of Carbumides. Part IV. The Mechanism of the Interactioiz o f Urea and Nitrous Acid. By EMIL ALPHONSE WERNER. THE decomposition of urea by nitrous acid is generally considered to be properly expressed by the following simple equation: CON,H + 2HN0 = CO + 2N + 3H,o, and this reaction is frequently cited in text-books as additional evidence iu support of the " carbamids " structure of urea. Theoretically this reaction should be available for the estimation of urea as is commonly suggested in the literature; it is never used for this purpose and i t is doubtful whether it ever has been since experiment has proved i t to be quite valueless. On the1 other hand, i t constitutes a well-known metho'd f o r the estimation 'of nitrous acid with a very fair degree of accuracy on the supposition that ther above equation is true.No doubt for this reason and on account of the employment of other methods for the estimation of urea this reaction has been considered a11 along as a normal change, scarcely deserving of any further inve'stigation. I n continuation of the author's work on the constitution and pro-perties of urea its behaviour towards nitrous acid has been sub-mitted to a careful quantitative study. The following are some of the more important facts which have been observed and whilst the 864 WERNER THE CONSTITUTION OF CARBAMIDES. PART IV. show what an erroneous ooaception has been generally entertained regarding the nature of this reaction they fully justify the neces-sity f o r a proper investigation into the true mechanism of the change.1. Usea and pure nitrous acid in aqueous solution did not inter-act. (Expta. IX X.) 2. The presence of a strong acid (hydrochloric or nitric) quickly promoted a brisk interaction even in dilute solutions and the reac-tion was coImp_letad in a relative’ly short time. 3. The presence of a weak acid such as acetic acid did not pro-mote an interaction unless the concentration was abnormally high, and qven then the velocity of the reaction was extremely slow. 4. The1 volume of nitrogen evolved was not a direct measure of the amount of urea decomposed calculabd on the basis -of the above equation ; the quantity decomposed was much greater * than that indicated by the evolved nitrogen.5. Only when urea was present in considerable excess was the volume of nitrogen evolved an approximately true estimate of the amount of nitrous acid decomposed. 6. The volume ratio of carbon dioxide t o nitrogen (1 2) required by the equation has never been obtained; thel proportion of carbon dioxide was always much higher; moreover the composition of the gas was liable to much variation with small changes in concen-tration. It is obvious that so far as the usual explanation of this reaction is concerneld all thesel facts stand out as anomalies f o r which the ordinary equation offelrs no explanation. Now anomalies in such a reaction can have no reality; their apparent existeace is the natural consequence of an erroneous con-ception of the change and when the true constitution of urea is considelred they appear as normal phenomena which reveal the true mechanism of the interaction.Mechanism of the Interaction of Urea and A’itrous Acid. I n the course of a recent investigation on the properties of pure nitrous acid RBy Dey and Ghoah (this vol. p. 414) noticed that a solution of the acid ( N / 32-HNO,) was practically without action on urea “ no matter how much urea was added.” They found that the addition of gulphuric acid was necessary to promohe and complete a reaction. This anomaly they remark “was without any appa-rent rea-son,” a just comment when urea is believed to be carb-amide. Pure nitrous acid in aqueous solution does not react with * See remarks on page 87 WERNER THE CONSTITUTION OF CARBAMIDES.PART IV. 865 urea until an amino-group is presented for attack a condition brought about by the production of a salt of urea on the addition of a sufficiently strong acid thus: NH,,HX HN:C<xH3+HX = HN:C< OH The first stage of the reaction then takes place in accordance (a) HX:C<Z29HX + HNO = N + HNCO + 2H,O + HX. the equation * : with The cyanic acid is decomposed in two ways as fast as i t is generated. It is hydrolysed,? thus: and directly attacked by nitrous acid according to the equation : ( b ) HNCO + H20 = NH3 + Cog, ( c ) HNCO + HNO = CO + N + H,O (see Expts. V I I and VIII). Both these decompositiom proceeld simultaneonsly with the primary reaction (a) but the relative proportions in which they take place can be varied a t will within certain limits by adopting suitable conditions which will be presently shown.The production of cyanic acid has been easily demonstrated by its isolation in the form. of the silver salt; thus when urea was attacked by nitrous acid in the presence of silver nitrate and a small excess of nitric acid a yield of pure silver cyanate was obtained equal to 42 per cent. of the theoretical calculated on the equation : N + AgOCN + 2HN0 + 2H,O. Considering the favourable oonditions for hydrolysis and the very sensible solubility of silver cyarrate in dilute nitric acid such a result was even more successful than could re’asonably have been ex-pected. It will be seen nolw that when urea (in the form of an salt) and nitrous acid interact a certain proportion of nitrogen from the urea is always fixed as an ammonium salt and herein liw the fallacy of the reaction so far as the estimation of urea is concerned.The variations observed in the ratios of carbon dioxide to nitrogen are thus easily explained since the volume of nitrogen evolved is lowered in proportion to the amount of cyanic acid hydrolysed. The latter change can be only partly suppressed even under the * No doubt this decomposition originates through the medium of diazotisation. t Cyanic acid in water alone is hydrolysed to urea 2HNCO+H20= CON2H4+C0 (Normand and Cumming T. 1912 101 1859); in the presence of mine ral acid of course the change is aa above 866 WERNER THE CONSTITUTION OF CARBAMIDES. PART IV. most f avourable conditions (that is high concentration and nitrous acid in excelss) with the result that the ratio of carbon dioxide to nitrogen evolved is never that which has been erroneously assumed.Now acoording to the above explanation the interaction of urea and nitrous acid is theuretically clearly divisible into two stages, during the first of which one molecule of urea is completely decom-posed by one molecule of nitrous acid instead of by two molecules, as has been commonly but falsely supposeld. This has been elasily proved experimentally by adopting the exact conditions which the theory rigorously demands namely (1) the presence of urea in excess a t the outset (2) a low concentration of nitrous acid (3) the presence of mineral acid in excess of that required to neutralise ammoiiia generated from the hydrolysis of cyanic acid and so to maintain the proper configuration of the urea molecule.Under these conditions the decomposition of cyanic acid by nitrous acid can be almost completely suppreased in favour of its decomposition by hydrolysis. A knowledge of the amount of cyanic acid hydrolysed compared with the volume of nitrogeln evolved is an all-important factor by means of which a very clear insight into the mechanism of the reaction has been obtained. The following results illustrate the degrele of success which has been realised in experimentally proving the problem which is indi-cated by the theory of the change now put forward. TABLE I. I. 11. 111. CON$34+HN0,. CON,H,+HNO,. CON,H,+HNO,, Molecular ratios 1 l 1.5 1 2 1 Nitrogen evolved, calculated on the theoretical ......92.5 per cent. 95.73 per cent. 99.34 per cent. HNCO hydrolysed 57.0 , 96.0 , 99.5 ,) posed by HNO 13.0 , 4.0 Y 9 0.5 ,? HNCO decom-Proportion of urea actually de-composed by one molecule of HNO ... .... .. . .. 79.5 , 91.73 , 98.84 ,) CO,=43*1 , CO2=43-2 , co,=44.2 ,, N =54*5 , N =54.3 ), NO = 3.2 , NO = 2.2 , NO = 1.3 ,, Ratio CO t o N.. . 1 1.24 1 1-26 1 1-22 It will be seen on viewing the results of the above experiments (the full details of which are given under Expts. V. VII. an WERNER THE CONSTITUTION OF CARBAMIDES. PART IV. 867 VIII.) that the amount of urea decomposed by one molecular proportion of nitrous acid according t o the equation N2 + HNCO + 2H,O + HX was less than that indicated by the volume of nitrogen evolved.The difference was most marked when the exact proportions (equal molecules) of urea and nitrous acid required by the equation were used since the conditions were less favourable for a quantitative realisation of the second change namely, HNCO + H20 + HX=NH,X + CO,, than when a considerable excess of urea was present. In the latter case the desired object was almost fully attained (111) and the true nature of the primary stage of the reaction thereby established. As regards the composition of the evolved gases the ratio of carbon dioxide t o nitrogen was in each case approximately 1 1.20; this of course was not the true value since a very sensible amount of carbon dioxide was held in solution in the residual liquid; when corrected in the case of result 111 for example the true ratio was CO,= 1 N= 1.02 or 1.1 as required by the combination of the two equations. It' may be well t o direct attention here t o the constant presence of a small amount of nitric oxide in the evolved gas; whilst this was no doubt due to the decomposition of a corresponding pro-portion of nitrous acid thus 3HN0,=HN03+ H,O + 2N0 i t was not found possible to eliminate it completely even when urea was in excess and the concentration of nitrous acid a t t'he outset of the reaction was as low as N / 2 0 . Under such conditions as are commonly adopted in the estimation of nitrous acid by the aid of urea the proportions of nitric oxide may easily amount to between 6 and 8 per cent. of the evolved gases according to the particular concentration of the solution used.This fact appears to have been generally overlooked. It is obvious when the ratios HN02:N and 3HN02:2N0 are compared that the presence of nitric oxide must lead to a result in excess of the true value; for example in the case of result 111, table I if the nitric oxide found was included as nitrogen. the yield of the latter would appear as 101.8 per cent. of the theoretical. RBy and his co-workers (Zoc. cit.) using a solution of nitrous acid of concentration N/32 obtained a result which they found was in excess of the true value in nearly the same proportion as above and no doubt for the same reason. This decomposition o 868 WERNER THE CONSTITUTION OF CARBAMIDES. PART IV. nitrous acid readily explains why a solution of comparatively high concentration say HNO = N / 6 can slowly attack urea ; the generation of nitric acid gradually brings about the required condition.The results under table I were obtained by adding the theoretical proportion of nitrous acid slowly and a t intervals to the acid solution of urea. A very low concentration of nitrous acid was thus ensured throughout the progress of the reaction. When the molecular proportion of nityous acid was added all at once the concentration a t the outset being HNO,=N/6 the results as was to be expected were very different as shown below (Expts. IV. and V.). TABLE 11. I. 11. 111. CON,Hd + HNO,. CON,H + HNO CON,H + HNO, Molecular ratios 1 l 1.5 1 2 l Nitrogen evolved 91-46 per cent.94.40 per cent. 96.48 per cent. HNCO hydro-HNCO decom-lysed.. ............. 7 1.5 .. 74.5 , 76-0 ,, posed by HNO 28.5 ) 25.5 )) 24.0 ), Urea actually de-composed by HNO ............ 62.96 .. 68-90 ,) 72.48 ,) Whilst the volume of nitrogen evolved was only slightly below that previously observed the amount of urea decomposed was in each case much less than before This was the natural result of the much greater facility offered for the decomposition of cyanic acid by nitrous acid a t the higher concentration. The latter was also responsible for the slight increase in the proportions of nitric oxide. The constancy* to be observed in the proportions of carbon dioxide and nitrogen in the evolved gases as shown in both tables in spite of the fairly wide differences in the propor-tions of urea decomposed is easily explained when the ratios of cyanic hydrolysed to cyanic decomposed by nitrous acid are con-sidered.As regards the very slow reaction which was noticed to take place between urea and nitrous acid in the presence of acetic acid (when HNO,=N/4) this was entirely due to the gradual decom-* Within the limits of experimental error the rate of mixing for instance, which affects the velocity of the reaction has a deoided-influence on the above ratios WERNER THE CONSTITUTION OF CARBAMIDES. PART IV. 869 position of the former acid. Urea acetate has been described by Matignon (Compt. rend. 1891 112 1369) as a compound which was completely dissociated in aqueous solution hence acetic acid could not establish the essential condition required t o promote the desired interaction (Expts.IX. and X.). Decomposition of UreG in the Presence of Two Molecular Proportions of Nitrous Acid. The results recorded in table I have conclusively proved that urea can be completely decomposed by one molecular proportion of nitrous acid; that this is not accomplished under the conditions which are commonly employed is solely due to the disturbing effect of the secondary reaction ( c ) . Now according t o the usual interpretation of the change two molecular proportions of nitrous acid should be required t o decompose one of urea. The effect after treating urea directly with nitrous acid in these proportions and at different concentrations,. in the presence of hydrochloric acid to promote the change are given below (for details me Expts.I. II. and 111.). TABLE 111. Nitrogen evolved. Composition of evolved gases. Per cent. Per cent. 72-02 HNO N/3 I. Urea N/6 (2 C.C. N-HC1) 71.99 C0,=32.1 N2=68.2 NO=9.G 69.19 C 0 2 ~ 32.6 N2-57*3 NO = 10.0 HNO N/6 11. Urea N/12 HNO N / 8 111. Urea N/16 HNO N/10 IV. Urea N/20 (3 C.C. N-HC1) (2 C.C. N-HC1) 72.07 CO,=31.1 N2=58*1 NO=10.4 1 I (2 C.C. N-HC1 I n each case it was readily proved that all the urea had been decomposed whilst an excess of nitrous acid remained yet in round numbers only about' 70 per cent. of the theoretical proportion of nitrogen was evolved. The remainder of the nitrogen was of course present as ammonium chloride* in the residual solution * It is interesting t o note that Clam (Ber.1872 4 140) long ago noticed the formation of ammonia when nitrous acid reacts with urea ; thus iie gave the following equation for the reaction in the cold; BCON,H,+N,O, 870 WERNER THE CONSTITUTION OF CARBAMIDES. PART IV. It will be noticed also that even without allowing for carbon dioxide held in solution the ratio of carbon dioxide t o nitrogen was still well below that of 1 2 as required. An equally marked divergence from the theoretical results was obtained by carrying out the reaction in two separate stages. It will be obvious that according to the usual but false explanation of the change such a procedure should give similar results for each stage. Whilst the first stage was repre-sented by the values given under I table 11 the volume of nitrogen evolved in the second stage was equal to only 26-38 per cent.of the original amount of urea present hence 62.96 + 26*38= 89-34 per cent.; nitrogen from the remainder of the urea (10.66 per cent.) was fixed as ammonium chloride in the second stage (see Expt. IV.). The composition of the evolved gases namely Co2=22.4 per cent. N2=50*7 per cent. NO=26*8 per cent. was very different from that of the gases set free in the first stage. A rather remarkable paradox makes its appearance when the results are compared an the basis of the false and of the truer equations; thus according to t,he usual interpretation of the reac-tion the amount of urea decomposed was roughly 30 per cent. greater (table 111) than indicated by the volume of nitrogen set free whereas in reality the amount of urea decomposed was much less than required by the volume of nitrogen evolved.The para-dox of course is but a phantom; its existence is just as unreal as the usual explanation of the change is incorrect. A contemplation of the results just recorded and so easily demonstrated makes i t almost impossible to believe that the behaviour of urea towards nitrous acid has ever been seriously studied with the object of obtaining evidence to support the sup-posed (' carbamide " formula." Whilst the present study of the reaction has supplied further proof of the cl'yclic formula it has also brought to light yet another This was not so. (NH4),C0 + 2N2 + C 0 2 . It was assumed however that urea was hydrolysed to ammonium carbonate during the process apparently in-dependent of the reaction with nitrous acid since the proportion of carbon dioxide to nitrogen evolved is shown to be the same as in the usual equation ; probably for this reason the observation is never mentioned in the text-books.* Emmerling (A. 1886 50 747; the original paper in Landw. Ver-suchs-Stat. 1886 440 was not available) studied the decomposition of urea by nitrous acid in the presence of nitric acid and acetic acid respectively both in cold and in hot solutions. The volume of nitrogen evolved was found never to be equal to the theoretical but no apparent attempt was made to offer any explanation of the results WERNER THE CONSTITUTION OF CARBAMIDES. PART IV. 871 of the many fallacies which abound throughout the chemistry of urea.The origin of these is not far to seek. Instead of a careful study of the properties and reactions of urea being made the groundwork for solving the problem of its constitution an almost infallible belief in the truth of the carb-amide formula has all along been the predominant factor in deter-mining what these properties and reactions should be. Secondary changes seemingly unimportant by-products apparent abnormalities in certain reactions and so forth have been pushed aside as of little consequence so long as the end result could be made to fit in with the “carbamide” structure. E X P E R I M E N T A L. With one exception all the experiments were made with the aid of a Lunge nitrometer. The specimen of sodium nitrite used for the generation of nitrous acid contained 97.18 per cent.of NaNO,; * a proportionate weight (71 69) corresponding with the theoretical required was used in each case. Action of Nitrous Acid o n Urea in Molecular Proportions of Two t o One. Expt. 1.-0.03 Gram of pure urea and 0.071 gram of sodium nitrite dissolved in 1 C.C. of water were introduced into the nitro-meter over mercury and 2 C.C. of N-hydrochloric acid directly added. Concentration a t outset HNO =N/3 CON,H,=N/6. The reaction was apparently coiripleted within thirty minutes whilst more than four-fifths of the gas had been evolved after five minutes. I n this and all other experiments not less than one hour was allowed to elapse before the gas was measured and analysed.Gas evolved=31.4 C.C. a t 18O and 766.5 mm. CO,=11 c.c., NO = 3 c.c. N,= 17.4 C.C. Volume of nitrogen a t N.T.P.=16.134 c.c. =72*02 per cent. of the theoretical. (Theory = 22.4 C.C. a t N.T.P.) Expt. 11.-As above but HNO = N / 6 CON,H = N / 12. Gas evolved=29*7 C.C. at’ 16O and 763.5 mm. C0,=9*55 c.c., NO=2*85 c.c. N,= 17.3 C.C. * Estimated by the thiourea method (T. 1912,101 2190 and Coade and Werner T. 1913 103 1221) 872 WERNER THE CONSTITUTION OF CARBAMIDES. PART IV. Volume of nitrogen a t N.T.P.=16.126 c.c. =71.99 per cent. of the theoretical. Expt. irZI.-The last experiment was repeated to prove the cause f o r the deficiency of the evolved nitrogen. After the evolution of gas had ceased the residual liquid was well washed out of the nitrometer.It required for neutralisation, using methyl-orange as indicator 6.1 C.C. of N/lO-sodium hydr-oxide. Since 1 C.C. of N-hydrochloric acid was directly neutralised in liberating nitrous acid there should have remained according to the usual equation free acid equivalent to 10 C.C. of N / l O -sodium hydroxide. Hence 10 - 6-1 =3*9 C.C. of N/lO-hydrochloric acid were neutralised by ammonia from the hydrolysis of cyanic acid. Now according to equations ( a ) and (b) (p. 865) the maxi-mum amount of acid that could be thereby neutralised would be 5 C.C. of N/10-hydrochloric acid therefore 78 per cent. of the theoretical proportion of cyanic acid was hydrolysed or so much of its nitrogen was fixed as ammonium chloride. The remainder of its nitrogen namely 22 per cent.was set free (equation c ) , together with that from urea in accordance with equation (a). Since all the urea was decomposed with the liberation of half of its nitrogen we have 50+22=72 per cent. of the total nitrogen set free which was in complete agreement with the result obtained from Expt.. 11. As the results with urea and nitrous acid (1 2) a t lower con-centrations (table 111) were obtained in a manner similar t o the above further details are unnecessary. Action of Nitrows Acid on Urea (2 1) in Two Stages. Expt. ZV. First Stage.-O*O6 Gram of urea and 0.071 gram of sodium nitrite were dissolved in 4 C.C. of water and 2 C.C. of N-hydrochloric acid directly added. Concentration CON,H = N / 6 HNO,= N / 6. Gas evolved=37.5 C.C. a t 1 8 O and 763 mm.CO,=13*7 c.c., Volume of nitrogen a t N.T.P.=20.49 c.c. =91*46 per cent. of the theoretical. Second Stage.-The gas having been expelled from the nitro-meter (from a repeated experiment) 0.071 gram of sodium nitrite dissolved in 1 C.C. of water was added and then 1 C.C. of N-hydro-chloric acid. The evolution of gas was very much slower than in the first stage and five hours were allowed for the completion of the reaction. NO = 1.6 c.c. N2=22*2 C.C WERNER THE CONSTITUTION OF CARBAMIDES. PART IV. 873 Gas evolved=25*3 C.C. a t 1 8 O and 769.3 mm. C0,=10*25 c.c., NO=2.35 c.c. N2=12*7 C.C. Volume of nitrogen a t N.T.P. = 11.82 C.C. =52*76 per cent. of the theoretical. An analysis of the residual solution after the first stage showed t h a t 7.15 C.C.of 2\7/10-hydrochloric acid had been neutralised, equivalent t o 71.5 per cent. of cyanic acid hydrolysed therefore 28.5 per cent. of the evolved nitrogen was t h e result of the reaction between nitrous acid and cyanic acid (equation c ) ; hence 91.46 - 28.5 = 62.96 per cent. of the urea present was decomposed in this stage. Therefore only 37.04 per cent. of urea remained t o be attacked by nitrous acid i n the second stage and since 52.76 per cent. of nitrogen was evolved it follows t h a t 52-76 - 37.04 = 15-72 per cent. of the nitrogen set free in this stage waL due to the above reaction (equation c ) . A comparison of the results from the two stages is not without interest. First stage. Second stage. Per cent. Per cent. Urea decomposed ..................= 62.96 37-04 HNCO hydrolysed .................. ~ 7 1 . 5 0 57.60 HNCO decomposed by HNO ... = 28.50 42-40 CO =36.5 40.5 50.1 9.3 Composition of evolved gas.. . . Ratio CO N 1 1.62 1 1.23 Since nitrous acid was in considerable excess i n the second stage (which should not be the case according t o the usual equation), the proportion of cyanic acid attacked by it to cyanic acid hydro-lysed was much greater than in the first stage. Decomposition of Urea by Oibe Molecdar Proportion of Nitrous Acid. In order to illustrate how the re'sults given under table I were obtained it will be sufficient to state1 the details of the most succw-f ul e.xperiment. Expt. 1.'.-0*12 Gram of urea was dissolved in 3 C.C. of N-hydro-chloric acid and the solution introduce'd into the nitrometer; 0,071 gram of sodium nitrite dissolved in 2 C.C.of water was placed in the cup (previously rinsed) of the nitrometer and added gradually in four separate portions to the urea solution. The1 reaction which was hastened by shaking t o ensure rapid mixing was allowed to completle itself before each addition of the sodium nitrite. VOL. (3x1. M 874 WERNER THE CONSTITUTION O F CARBAMIDES. PART IV. Gas evolved (after one hour) =44*2 C.C. a t 1 8 O and 766.4 mm. ; Volume of nitrogen a t N.T.P.=22*252 c.c.=99-34 per cent. of the theoretical. The residual solution from a similar experiment required 10.05 C.C. 'of N / 10-sodium hydroxide for neutralisation. Hence 20 (2 C.C. of 1";-hydrochloric acid originally free) - 10.05 = 9.95 C.C.of N/lO-liydrochloric acid were neutraliseld as the result of cyaiiic acid hydrolysis which was therefore alniost complete. There-fore the amount of urea actually decomposed was 99*34-0.5= 98.84 per cent. 'of the theoretical only 0.5 per cent. of the evolved nitrogen being derived from cyanic acid. Therefore one molecule of urea was decomposed by one molecule1 of nitrous acid. CO,=19*6 C.C. ; NO =0*6 C.C. ; N,=24*0 C.C. Isolation of Cynnic Acid as the Silver Salt from the Interaction of Urea and Nitrous Acid. Ex@. VI.-0*6 Gram of urea and 0.71 gram of sodium nitrite were dissolved in 40 C.C. of ice-cold water and t o the solution 1.7 grams of silver nitrate previously dissolved in 5 C.C. of water and 5 C.C. of iY-nitric acid welre added. As t h e pale cre'am-coloured pre-cipitate 'of silver nitrite which was immediately formed gradually disappeared it was replaced by a snow-white precipitate of silver cyana te.During the progress of the reaction further 5 C.C. of N-nitric acid were added. After an hour the precipitate was collected washed, and dried. It gavel none of the reactions for nitrous acid and contained Ag = 71-84 per cent. (AgOCN requires Ag = 72 per cent.) ; on adding a few drops of sulphuric acid to the dry salt the character-istic pungent odour of cyanic acid was evolved. The w-eight of silver cyanate obtained was 0.63 gram which was equal t o 42 per cent. of the theoretical for equation (d). The Interaction uf Cyanic Acid and Nitrous Acid. As this change does not appear to have been hitherto examined, the following experiments were made in order to prove the validity of equation ( c ) already given.Expt. VII.-0*081 Gram of pure potassium cyanate and 0.071 gram of sodium nitrite werel dissolved in 2 C.C. of water and intro-duced into the nitro8meter; 3 C.C. of AT-hydrocliloric acid were added, t h a t is 1 C.C. of acid in excess t o counteract the neutralising effects of hydrolysis. Concentration a t outset LINO and HNCO = A ' / 5 WERNER THE CONSTITTJTION OP CARBAMIDES. PART IV. 875 The evolution of gas was very rapid and the reaction was practi-Gas evolved after ‘one hour=34’3 C.C. a t 16O and 757.8 mm.; C02=19.1 c.c.; NO-3.6 c.c.; N2=11*6 C.C. Volume of nitrogen a t N.T.P.=10*72 c.c.=47*8 per cent. of the theoretical. Therefore 52.2 per cent.of cyanic acid had been hydro-lysecl. The residual solution required for neutralisation 8.5 C.C. of N / 10-sodium hydr-oxide inste’ad of 4.78 C.C. as required by the gasometric analysis. The apparent discrepancy was easily elxplained when the above results were considered. Tho volume of nitric oxide evolved (3.33 C.C. a t N.T.P.) represents a decomposition of 22.3 per cent. of nitrous acid with the generation of nitric acid equivalent to 0.74 C.C. of N / 10-sodium hydroxide whilst the proportion of cyanic hydrolysed showed that free nitrous acid remained equivalent t o 3 C.C. of NI10-sodium hydroxide. Hence 4.78 + 0.74 + 3.0 = 8.52 C.C. of AT/ 10-sodium hydroxide were required which is in complete agree-ment with the value actually found. The presence of unchanged nitrous acid in the residual liquid was easily proved.Therefore the reaction between cyanic acid and nitrous acid takes place theoretically between equal molecular proportions but a t a concentration of N / 5 the velocity of hydrolysis of cyanic acid is slightly higher than that of the primary change. cally completed witshin five minutes E,rpt. Vlll.-The above experiment was repeated. The Behaviour of Urea towards Pure Nitrous Acid alone 04 in the Preseitce of Acetic Acid. Expt. ZX.-O*O3 Gram of urea and 0.071 gram of sodium nitrite were dissolved in 2 C.C. of water and 2 C.C. of N-acetic acid were added. There was no preceptible evolution of gas until after a consider-able time; thus after twenty-four hours 7.4 C.C. had been evolved and a t the end of ninety-six hours when the experiment was stopped the volume of the evolved gas was=12*6 C.C.a t 1 3 O and 758 mm. The volume of nitrogeln at N.T.P. was=6.56 c.c. and the original gas contained 9.5 per cent. of nitric oxide. The slow action was primarily brought about as a result of the gradual decomposition of nitrous acid whereby urea nitrate was slowly generafed. RSy and his co-workers (Zoc. cit.) have shown that even a t Oo the most concentrated solution of nitrous acid, stable for only a short time was approximately N/5.5. Expt. X.-The same proportions of urea and sodium nitrite as before were dissolved in 29 C.C. of water and 1 c.c of N-acetic acid was added. Concentration of HNO = N / 4. Concentration of HNO =A’/ 30. &I M 876 ROBINSON A THEORY OF THE MECHANISM OF THE After remaining for three days iii the nitrometer the voluine of gas evolved was 0.8 C.C.Yet when 2 C.C. of N-hydrochloric acid were added a fairly brisk reaction was quickly promoted and even a t this low concentration was almost completed at the end of half an hour. &urn mary . (1) Urea is not attacked by pure nitrous acid alone or even when a second very weak acid is present. (2) When a salt of urea is produced by the presence of a sufficiently strong acid i t is immediately attacked by nitrous acid, because an amino-group is thereby presented for such attack. (3) One molecule of urea (as a salt) requires but one molecule of nitrous acid for its deconipositioii into nitrogen cyanic acid, and water since only one amino-group is present. (4) Cyanic acid and nitrous acid react in equal molecular pro-portions with the production of nitrogen carbon dioxide and water. (5) The usual interpretation of the reaction between urea and nitrous acid which has been hitherto accepted is incorrect; first, because it is in contradiction to the experimental facts and, secondly because it is based on an erroneous conception of the constit'ution of urea. UNIVERSITY CHEMICAL LABORATORY, TRINITY COLLEQE DUBLIN. (Received July Zlst 1917.

ISSN:0368-1645    

DOI:10.1039/CT9171100863

出版商:RSC    

年代:1917

数据来源: RSC