摘要:

CHATTAWAY Ah'D LEWIS : 6-CHLORONAPHTHALENE. 875 LXIX.-Prepuratioik of p- Chlowna phthulene. Bp F. D. CHATTAWAT, B.A., Christ Church, and W. H. LEWIS, Jesna College, Oxford. As P-chloronaphthalene mas required in very large quantity in the preceding investigation, and as no satisfactory method of preparing it has been described, the various ways of producing the compound were investigated, and it may perhaps be of service t o record the results obtained. The action of phosphorus pentachloride or phosphorus pentabromide on &naphthol is not at all well adapted for the prep:tration of p-chloro- naphthalene or /3-bromonaphthalene in any quantity. Not only is the process unpleasant and tedious, but the yield is very small, reach- ing usually t o on17 about 10 per cent. of the theoretical.On distilling the product of the action orer a naked flame, the @-halogen derivative comes over first together with the liquid product mentioned by Rimarenko (Ber., 9, 663), probably a dihalogen naphthalene deriva- tire ; the P-dcrivative can easily be separated from this by a good filter-pump and subsequent recrystallisation from alcohol. After these products have passed over, on raising the temperature to incipient redness, a thick brown tar distils, and finally a large amount of a coked residue is left in the distilling flask. This tar dissolves in benzene, yielding a dark-brown fluorescent solution, and it is also somewhat soluble in alcohol, giving a pale yellow, strongly fluorea- cent solution. On allowing a not too concentrated alcoholic solution to cool, a sulphur-yellow substance is deposited in small feathery tufts ; if the solution be too concentrated it falls out in small resinous drops, as it melts somewhat below the boiling point of ordinary alcohol.p-Chloronaphthalene is best prepared from p-naphthylamine b,- diazotising it and converting the diazochloride into the chloro-deriva- tive by means of cuprous chloride (Lellmann and Remy, Ber., 19, 610). I f proper precautions be taken, a very satisfactory yield can be obtained, and considerable quantities of /3-chloronaphthalene can be very easily prepared. I t is essential throughout the operation876 CHATTAWAY AND LEWIS : &CHLORONAPHTEALENE. that as little water as possible should be present, Concentrated hydro- chloric acid being used to suspend the /3-naphthylamine in, and t o hold in solution the cuprous chloride.The solution of sodium nitrite used for diazotising should be as strong as can be added without much loss of nitrous acid. After the addition of the nitrite, thc liquid should be left for several hours to allow the diazotisation to complete itself. The solution of the diazochloridc should be added to the cnprous chloride solution a t a temperature of 70°, and the copper solution should be maintained at this temperature during the addition. It is better, after evolution of nitrogen has enti~ely ceased, to allow the mixture to cool, and to stand 8 t o 12 hours before distilling with superheated steam, which then may be blown directly through the acid liquid. Not more than about 30 grams of 6-naphthylamine should be used in one operation ; if larger qnan- tities are used, or if the solutions are not allowed to stand between the operations, the results are not so good.Very little decomposition of the /3-diazonaphthalene chloride solu- tion takes place on standing in a cool place during 8 to 12 hours, but if it is allowed to stand 24 hours, or more, a gradually increasing amount of decomposition takes place. An excess of hydrochloric acid greater than that required to keep the various substances in solution, or in the state of a thiu paste, seems to be of no advantage, the point being that the per- centage of hydrogen chloride in the solutions should always be as large as possible. By strongly cooling the solution of the diazochloride, and of the cnprous chloride, and pouring them together in a thin stream into a large flask, the double compound, CloH7N,CI*Cu2Cl2 (Lellmann and Remy, Zoc.cit.), can be obtained; it is of a bright orange-yellow colonr, and there is no darkening or apparent decomposition; 011 allowing the pasty mass to stand, bubbles of nitrogen begin slowly to be given off as the temperature rises, but the decomposition, even after 12 hours' standing at the ordinary temperature, is small ; on gently heating, more nitrogen is slowly evolved, but no consider- able evolution oE gas takes place till abont 65-70", when gas is rapidly evolved, the action completing itself st this temperature. The double compound entirely decomposes, and the liquid becomes clear, the crude P-chloronaphthalene floating as a black oily layer on the surface of the hot liquid.Tbis method, however, was not found to give such a good result as adding the diazochloride to the solution of cuprous chloride heated to 70". In the prepara- tion, the p-naphthylamke, after being very finely powdered and sifted, is thoroughly rubbed down to a thin paste with about 12 times its weight of concentrated hydrochloric acid ; the mixtureON .B-MERC CRYDINAPHTHYL AND ~~-DIXAPHTHYL. 8 77 is then cooled by means of ice and salt, and diazotised with the calculated quantity of sodium nitrite dissolved in about 10 times its weight of water. The nitrite may be added fairly rapidly, as no great rise of t'emperature takes place. The diazotised solution is allowed to stand in ice during about six hours, after which it is slowly poured down an invertsd condenser into a slight excess of rz solution of cuprous chloride dissolved in about 12 times its weight of concentrated hydrochloric acid previously hezted to about 70°, and maintained as far as possible at that temperature during the addition.After the addition, the whole is allowed t o stand overnight and then distilled in a current of superheated steam. The slightly reddish soiid product is collected, ground up with a little potash solution, and redistilled from a slightly alkaline solution by means of super- heated steam. The @-chloronaphthalene is thus obtained practically pure in bea,utiful, white, crysialline masses, and, if necessary, may be crystallised from alcohol. /3-Naphthylamine yields thus, as a rule, about, its own weight of impure product in the first distillation, the final yield of the pure substance being about 75 to 80 pel. cent. of the theoretical. In this process, a certain amount of p-naphthol is always formed, and also a considerable quantity of a black pitch-very probably a condensation product of the P-naphthol, as the amount formed is increased if the diazochloride be allowed to stand too long before adding it to t,he cuprous chloride solution. After the first steam distillation is completed, on cooling the residual liqnid, a small quantity of bulky feathery crystals are noticed floating i n the green solution. These can be recrgstallised from glacial acetic acid when they form long, silky, needle-shaped crystah, which dry to a flesh-coloured felted mass, and melt at about 58".

ISSN:0368-1645    

DOI:10.1039/CT8946500875

出版商:RSC    

年代:1894

数据来源: RSC

摘要:

ON .B-MERC CRYDINAPHTHYL AND ~~-DIXAPHTHYL. 8 77 LX X. --iVote o ?z /I- Me1 YY wy7 i 11 c@ thy 1 and PP-Di?ZUpkt hyl. By F. D. CHATTAWAY, B.A., Christ Church, Oxford. U-h~ERCURYDIS4PHTBYL was first obtained Otto and Jloi-ies, in 1868 (AnnuZen, 147, 164), while endeavouring to synthesise dinaphthyl as Fittig had synthesised diphenyl, by the action of sodiuni on bromo- naphthalene. As so large an amount of products other thsu di- naphtyl was obtained, they tried t o moderate the action by using sodium amalgam in place of sodinm, and found that the action then took place in quite a different direction, sc-mercurydinaphthyi being produced. They heatec! sc-bromonaphthalene dissolved in manyti78 ON /3-MERCURYDINAPHTEPL AND flfl-DIKAPHTHYL. times its volume of coal-tar oil of high-boiling point with an excess OP pasty sodium amalgam during about 18 hours, and, on cooling, a-mercurydinaphthyl separated out in small white plates. I n a later communication in 1870, after investigating mercury diphenyl (Annaleu, 154, 93), Otto and Dreher mention that the action is much furthered by the presence of ethylic acetate in the proportion of one-tenth of the volume of the bromonapthalene used.At that time, it was not possible to synthesise the corresponding p-clerivative, as /i?-bromonaphthalene was not obtained till 1876 (Annuh, 183, 225). I n the course of an investigation of the di- naphthyls, it was desired to attempt the synthesis of P/3-dinaphthyl by distilling the p-mercurydinaphthyl with soda lime. It also seemed desirable t o try the action of thionyl chloride on this compound, as Henmann and Kochlin, in 1883 (Be,.., 16,1625), obtained p- and not a-chloronapthalene, b)- the action of the chloride on a, mei-curydi- naphthyl which must have been the a-compound.The author, there- fore, successfully attempted to obt,ain /3-mercurydinapthjI, adopting the method of Otto and Dreher. After i t had been prepared and some of its derivatives partially studied, Behrens, in the course of an investigation of aromatic boron compounds (Michaelis. Ber., 27, 244) published a short account of it, he having prepared it in order to study its beha-i-iour with boron chloride. As the results obtained by the author confirm Behren’s work, it seems desirable briefly to record them, especially as /3-mercury- dinaphthyl is another exception to the rule that a-naphthalene de- rivatives have a lon-er melting point than the corresponding p-com- pounds.P - ~ e r c u . r y d i n n ~ h ~ ~ ~ ~ Z . - T ~ i s compound was prepared by dissolving P-bromonaphthalene in about three times its weight of dry xylene, boiling slightly oTey 140°, adding 5 per cent. of its weight of pure dry ethylic acetate, and a large excess of sodium amalgam, which was just fluid ; the liquid was kept gently boiling in an oil bath during about 30 hours. A brownish-gray deposit formed on the amalgam and, on cooling, a white solid crjstallised out in crusts on the sides of the flask. A large excess of benzene was then added and the liquid having been heated t c boiling was filtered while hot from the excess of sodium amalgam, and from a gray residue of very unpleasant smell.An orange coloured filtrate was obtzined which deposited white crptals of /3-rncrcui~ydinaphth~l on cooling. These were recrystallised from benzene. ~-~lercurSdinaphthSl crystalliees in 11-hite glitteriog scales melting at 23S0, and haring exactly the appearance of pearl wheu pressed together. It is insoluble in water, almost insoluble in ether and alcohol, but moderately soluble in hot benzene and its homologues,REDUCTION OF PARATOLUENEAZODIMETHYLASILXNE. 8 7 9 xrom which, however, it crystallises almost entirely on cooling. Concentrated mineral acids completely decompose i t on warming, Porming the corresponding mercury salts, and apparently substituted derivatives of naphthalene. When distilled over red hot soda lime, it yields among other products PL?-dinaphthyl.The yield obtained was about 30 to 35 per cent. of the theoretical. p-chloronaphthalene also gives the compound under similar condi- tions, but the yield is not so good, a considerable reduction of the chloro-derivakive to naphthalene t'aking place. An inres tigation of the action of heat and of thionyl chloride on p-mercurydinapthyl is proceeding. /3p-DinuphthyZ.-This compound can be obtained in small quan- tity by distilling p-mercui.ydinaphthy1 over red-hot pumice or soda lime in a combustion tube. It c m also be easily obtained by the action of sodium at 140-150" on /3-chloronaphtbalene dissolved in xylene. A s in the case of /jr-phenylnaphthalene, the action is much facilitated by the presence of dry ethylic acetate in the proportion of 5 per cent. of the weight of' the p-chloronaphthalene. The yield of pp-dinaphthyl mas by this addition raised from 3-4 per cent. to 12-15 per cent. of the theoretical. pp-Dinaphthyl crystallises in glittering plates, melting at about 157'. It is moderately soluble in hot alcohol, rery readily in benzene and xylene. The author is engaged in in-i-estigating the dinaphthyls and their derivatives, and hopes later to communicate to tke Society a paper on the subject, and t o refer also to the remarkable effect of ethylic ace- tate in promoting the interactions of sodium 01- sodium amalgam and the halegen derivatives of naphthalene dissolved in dry xylene. St. Bartholnnzezc's Hmpitnl.

ISSN:0368-1645    

DOI:10.1039/CT8946500877

出版商:RSC    

年代:1894

数据来源: RSC

摘要:

REDUCTION OF PARATOLUENEAZODIMETHYLASILXNE. 8 7 9 LXXI.-Red.zcct ion of PcLs.utol.zceiieu~o~~~eth y7unilint. By D. R. BOYD, B.Sc., late Donaldson Scholar of Glasgow University. Demomtrator of Chemistry in the Mason College, Birmingham. HIS inTestigation is connected with the recent researches of P. Jacobson and his pupils (Ber-., 25, 992 ; 26, 681, 688, 699), of 0. N. Witt and Schmidt (Ber., 25, l O l Y ) , and of Tauber (Ber., 25, 1019), which have led to the discorerg of the molecular transformation of hxdrazo-compounds into semidine bases. In the course of their wol-k, Jacobson and Knnz reduced benzeneazodimetbylaniline with stannous chloride and hydrochloric acid, and in addition to the products of decomposition-aniline and nmidodimethplaniline-they880 BOYT): REDUCTION OF isolated a base isomeric with benzenchydrazodimethylaniline, whicli w&s found to have the constitution o€ a dimethyl triamidodiphenyl.C,Hs-fi' * p d N H 2 (CH,),N*C,jH,*N (CH,)2NaC,H3.NH, So far, only a very brief account of this investigation has appeared in the journals (Eel-., 26, 704) ; a more exhaustive account is to be found in the dissertation of Kinnx (Inavg. Diss., Eeidelbcrg, 1893). Although, in the case investigated by Jacobson and Kunz, trans- formation into a semidine base was not observed, notwithstanding the substituting group was in the para-position, yet it appeared possible that the action might take a different course if a substituting group were present in the second para-position also. For this reason I was requested b7 Professor Jacobson to investigate the behaviour of pa,-atolzieneazodimethylaniline, c,H,/\s:s/-'s(cH,),, when acted on by reducing agents.This investigation, which was carried out in the laboratory of Heidelberg University, has showii that paratolueneazodimethyl- aniline, on reduction, yields n semicline base of the constitution CH,/-\NH/-\S(CH,),, in very considerable quantity. \-/ \-/ \--/ \---/ XH, Before giving a detailed description of the experiments, I desire to express my indebtedness to Professor Jacobson for t,he invaluable help received from him a t all points of the work. Preparation qf P a mtolrieueazod irnrth ylnizil in P . This compound has already been described by Mohlau (Be].., 17, 1492), but the method I employed was somewhat different from his. 50 grams of paratoluidine mere dissolved in 500 C.C.of water and 112 C.C. of concentrated hydrochloric acid (i38 per cent.), and the solution was diazotised with 32 grams of sodium nitrite. After a short time, the solution of diazotoluene chloride was poured into a well-cooled solution of 60 grams of dimethylaniline in 300 C.C. of water and 45 C.C. of Concentrated hydrochloric acid, and the whole left f o r about an hour; the azo-compound was then precipitated by adding 250 grams of sodium acetate, and allowing the mixture to stand overnight. It crjstalliseq well fi*om benzene, and melts at 16g3 ; Mohlsu found 168-168.3". Red uct iou o j Pa /-a to 1 I I e i 1 e c 1 zod i me t h ~ i 1 a i ~ i 1 i ne , The reducing agent nsed was a solution of stannous chlo;.ide inPAHATOLUENEAZODIMETEIPLA?\'ZINE. 88 1 concentrated hydrochloric acid (4-0 grams in 100 c.c.). In the first place, experiments were made to find out the most favonrable condi- tions for transforming the hydrazo-compound into an isomeric base, ii-ith the smallest amount of decomposition into its constituents.1. Ren'wtion in presence qf Akohol.-5 grams of the nzo-compound were ground up with 25 C.C. alcohol, and gradually added to 30 C.C. o€ the warm stannous chloride solution. The alcohol was partially eraporated, and the yellow liquid thus obtained, which had deposited cr@als of the double salts, mas treated when cold with caustic soda and extracted i d l ; ether. The ethereal solution, after being dried with potassium carbonate, was evaporated to drpess, when a dark- coloured, oily residue, about 4.5 grams, was obtained ; t h i s was dis- tilled, under diminished pressure (25-30 mm.), in a current of carbcin dioxide, the temperature of the surrounding air bath being gradually raised to 200".The distillate amounted to about 3.5 grams, and in tlie distilling flask there remained 0.6 gram. The distillate was after- wards fractionated under the ordinary pressure, and separated into two fractions, the boiling points of which, namely, 198" and 257", showing that it mas a mixture of paratoluidine and amidodimethyl- aniline. The experiment accordingly jielded the following results. 70 per cent. hydrazo-compound split up into its constituents. 12 77 9 7 ,. transformed into isomeric base. 2. Reduction at ix Gentle. Heat L r i t h i t Alcnhol.--1L? grams of the azo-compound were added gradually to 60 C.C.of warm stannous chloride solution, the mixture heated gently for a short time, and then treated as above described. In this case, the mixed bases from the ethereal solution amounted t o 9 grams, and of these 3.7 grams were found to be non-volatile at- R temperature of 200", under diminished pressure. The result was therefore 53 per cent. h-j-drazo-compound decomposed. 37 .. .. transformed. 3. Eedirction a t the O r d i u a y Terjzpemtiire ulithozit d1cohcl.-Rise of temperature during the process was prevented by means of ice. From 20 grams of azo-ccmpound, 19 grams of tlie mixed bases were obtained, and the residue: after distillation under diminished pressure, weighed 8.7 grams, or 52 per cent.hpdrazo-compound decomposed. 43 ) ) 7 ,, transformed. From these results, it is evident that the presence G f alcohol farours the splitting up of the hydrazo-compound, and is therefore to883 BOYD: REDUCTION OF be avoided. The conditions most favourable to the molecular trans- formation of the hydrazo-compound are reduction in the cold with- out alcohol. This, however, is somewhat tedious, and the method given under 2 is preferable, the yield being only slightly less. It was found that be.1i=er2eazodimeth~la?~il~ne, when reduced under conditions 1 and 2, gave the following results. Decomposed. Transformed. 80 per cent. 10 per cent. Isolation of the Base.-After the paratohidine and amidodimethyl- aiiiliiie have been distilled off under diminished pressure, a brown, resinous mass is left, from which.by extraction with light petroleum, a base can be obtained in the crystalline form. For this purpose a veyy considerable quantity of light petzoleum is required, and it is there- fore advisable to transfer the resinous mass to a larger flask by dis- solving it in benzene and afterwards evaporating the solvent on the water bath. From the petroleurn solution, the base separates out partly in masses of almost colourless needles and partly in oily drops ; the latter, from which the crystals can easily be separated, are again dis- solved in light petroleum, when a further crop of crystals is obtained. The crystalline base dissolves in light petroleum much more easily than the original residue from the vacuum distillation, and can easily be obtained pure by repeated crystallisation out of this solvent.The yield on an average was 1 G 1 5 per cent. of the original am- compound. The substance is very easily soluble in alcohol, and even more so in benzene ; it crystallises from light petroleum in beautiful, white needles melting at 69-70". With ferric chloride, a solution of the base in dilute hydrochloric acid gives a violet coloration, which rapidly becomes red, and, 011 addition of concentrated hydrochloric acid, pale yellow. The analysis shows that the base is isomeric with pamtoluenehydr- axodimethylaniline. 0.2014 gave 30.2 C.C. moist, nitrogen at 15" and 761 mm. N = 17.59. 0-1807 ,, 0.4958 CO, and 0.1254 H,O. C = 74.83; H = 7.71. C15H19N3 requires C = 74.69 ; II: = 7.88 ; N = 17.43 per cent.Behaviour of the Ease towards Formic Acid. 1 gram of the base and 10 grams of formic acid were boiled for five hours i n a flask with reflux air condenser, the flask being kept filled with carbon dioxide to prevent oxidation as far as possible. The liquid was then poured into water, filtered, and treated with carbonate of soda. The precipitate thus obtaified is easily soluble in .tlcohol, and, on carefully adding water to the solution, is deposited 1. With alcohol.. . . 2. Without alcohol. 42 ,, 50 7 9 4PARATOLUENEAZODIMETHY LAEXLINE. 88'3 in lnstrons prisms melting at 110-111". It is easily soluble in dilute hydrochloric acid, and, on adding nitric acid to this solution, a preci- pitate of the nitrate is deposited after a time in bealitiful, white needles.The analysis showed that the action had taken place according to the following equation. ClJH19N3 + CH202 = CsHi7N3 + 2H2O. 0.1961 gave 28.6 C.C. moist nitrogen a t 20" and 761 min. N = 16-71. 0,1906 ,, 0.5361 CO, and 0.1190 H20. C = 76-71 ; H = 6.94. C,H,,N, requires C: = 76.51; H = 6-77; N = 16.73 per cent. The formic acid derivative is therefore free from oxygen, and consequcntly must be a methenyl and not, a formyl compound. From this it followed that the base was probably an orthosemidine, either 2-amido-5-dimethylamido-4'-methyldiphenylamine, or 2-amido- 5 -m ethyl-4' - dimethy lamidodip heny lamine. XH2 CH3 \-/ \--/ \-/ \-/ CH3, /-\ .NH./-'\ or /--' .N H.,' --\ X (C H3) 2, N(CH3), NH, in which citsc such a reaction is easily understood.Still the formula of a diphenyl base, could not be regarded as impossible, inasmuch as anhydro-condensa- tions, as is well known, sometimes take place in the peri-position also. Behaviour of the Base towards Xalicylaldehyde.-An alcoholic solu- tion of the base (1 mol.) and salicylaldehyde (2 mols.) was boiled for about two hours on a water bath : the action, however, begins even in the cold. The compound forms magnificent, red prisms which melt at 131", with the exception of a small residue which disappears at 134". It dissolves in dilute hydrochloric acid forming a golden-yellow liquid, which, on boiling, becomes coloui-less and has the smell of salicylaldehyde. The analysis showed that in spite of the presence of an excess of salicylaldehyde, only one molecule of the latter had reacted with each molecule of the base.C,,H19N3 + CiH60, = C22H3N30 + H20. 0.2015 gave 20.9 C.C. moist nitrogen at 16" and 765 mm. N = 12.17. 0.1965 ,, 0.5527 CO, and 0.1189 H,O. C = 76.71 ; H = 6.72. C22H,N30 requires C = 76.52; H = 6-67; N = 12.17 per cent.584 BOYD: REDUCTION OF As already mentioned, the salicylaldehyde derivative is easily soluble in dilute mineral acids, and on passing a current of steam through this solution, the aldehyde can be driven off; if it is now rendered alkaline with sodium carbonate, the original base is obtained as an oily precipitate, which can be extracted by means of ether, and crystallised out of light petroleum. This fact, and the remarkable tendency which the compound has to crystallise, suggested t'hat it might with advmtage be used in the preparation of the new base.It seemed possible that bF dissolving in alcohol the crude product obtained by distillation under diminished pressure, treating this directly with salicylaldehyde, and breaking up the salicylic compound thus obtained, a better jield of base might be got. Accordingly an alcoholic solution of 1 part of the distillation residue and 1.1 parts of salicylaldehyde was heated for several hours in the water bath. To my great astonishment, a yellow substance, entirely different from the compound already described, was deposited from the hot solution, whilst the mother liquor on cooling yielded a mixture of red and y e l l o ~ crystals, which could be separated by further crystallisat ion. The red substance, cry~t~allised from alcohol, melted a t 131-1341", and proved to be the salicylaldehyde derivative already known ; this method of preparing it is, however, by no means to be recom- mended, as, owing to the iiumerons ci*ystallisations, the yield is but moderate.The point of interest was now to determixe the nature of the yeZEow substance, which might possibly be derived from a second isomeric base. This presents a striking contrast to the red compound, in being remarkably insoluble, even i n boiling alcohol ; it dissolves, however, easily in benzene and can be obtained from this solvent in pretty, yellow crptals. The melting point is much higher than that of the red substance, but, in spite of repeated crystallisation, i t was never very sharp. It lies between 215" and 220". Analysis showed that it had approximately the same percentage composition as the red compound.0.1817 gave 18.2 C.C. moist nitrogen a t 15" and 751 mm. N = 11.70. 0.2165 ., 22.6 .) ., 16" ,, 747 ,, N = 11.96. 0.2006 ., 21.2 7 , ,, 19" ), 754 ,, N = 12.05. 0.2357 ,) 0.6682 CO, and 0.1312 H,O. C = 77.31 ; H = 6.5. 0.1698 ,, 0.4813 GO, ,, 0.0962 H20. C = 77.30 ; I3 = 6.30. In order to test the above indicated hypothesis, that this substance was derived from a second isomeric base, it was of importance to study its behayiour towards acids. In dilute sulphuric acid, it dis-PARATOLUENEAZODIMETHP LANILISE. 88 5 solved only on heating, forming a colourless solution, smelling strongly of salicylaldehyde. On adding sodium carbonate to this solution, after the aldehyde had been driven off with steam, a white, solid precipitate formed, which was collected by means of the pump, and dried on the water bath ; it was insoluble in light petroleum, but dissolved easily in benzene, and crystallised from the latter in colour- less cubes melting at 234-235'.It is easily soluble in dilute acids but insoluble in alkalis. From these properties, namely, the insolubility and remarkably high melting point, it appeared almost impossible that it could be a simple base isomeric with paratoluenehydrazodimethylaniliiie, and in fact the analysis pointed to an entirely different composition, one remarkably similar to those of the red and yellow substances. 0.1722 gave 18.7 C.C. moist nitrogen at 18.5Oaiid 757 mm. N = 12.G.0.1901 ,, 0.539 CO, and 0.10'78 H,O. C = 77.34; H = 6.30. These observations v-ere ultimately explained in the light of the investigations which 0 Fischer and his pupils (Ber., 26n, 197) have carried out on the derivatives of orthodiamines. According to these authors, such aldehyde derivatives easily undergo oxidation to the corresponding acid auhydro-derir~tives. Consequently the yellozu and white Eubstances might be oxidation products of the red one, and thus all three be derived from one and the same base. As a matter of fact, the analysis of the white substance agrees csactly with a salicylic acid derivative of the new base. C15H19X3 + CiH603 = C2,H,,N30 + 2H20. Calculated. Found. C ........... 76.97 i 7 *34 H ........... 6-12 6.:30 N ........... 12-24 12-45 This supposition became a certainty when it was found possible to convert the red compound into the white one by oxidation. An alco- holic solution of the red substance was boiled with mercuric oxide, whereby the latter rapidly turned grey, and the solution became colourless.After cooling, the insoIuble residue was extracted with hot benzene, and this solution now gave the white substance in cha- racteristic crystals, melting at 230-234'. It follows, therefore, that the white compound is a salicylic acid derivative, corresponding t o the red aldehyde derivative. The nature of the yellow substance is less clear ; the fact that by decomposition it yields abundantly, on the one hand, salicylaldehyde, and, on the other hand, the salicylic acid derivative of the new base, makes it most probable that it is a complex product intermediate between the VOL.LXV. 3 E88 6 BOYD: REDUCTION OF aldehyde and acid compounds. Biit this view can only be considered as a supposition. On the other hand, it must be regarded as certain that the yellow substance is derived fi-om the same base as the other two, and therefore there is no ground for assuming that by the reduction of paratolueneazodimethylaniline more than one base iso- meric with the hydrazo-compound is produced. Behatiour of the Base towawls Benzile. It has already been stated that the formation of a methenyl com- pound makes the conception that the new base is an orthosemidine very probable, but a t the same time does not exclude the possibility of its being a peridiamine.I t s behaviour towards salicylic aldehyde is also no absolute proof of its semidine nature, since peridiamines might possibly react with the aldehyde group in the following way. \ I I \/*NHz (,l.XH ,/\-NH, (\$I1 I + CHO-R = 1 /CR*R. A decision on this point was to be expected from the behaviour of An orthosemidine must unite with benzile the base towards benzile. to form an azonium base contailling oxygen, whereas a peridiamine, according to the researches of Tauber (Ber., 25, 3287), must yield a condensation product free from oxygen. One gram of the base and 1 gram of benzile were dissolved in 25 C.C. of alcohol, to which 1 gram of concentrated hydrochloric acid was added, and the solution heated on the water bath for some four hours. The liquid was then poured into water, and, after being filtered from the excess of benzile thus precipitated, was made alka- line with ammonia ; in this xay, a voluminous precipitate is obtained.On dissolving this in chloroform and adding alcohol to the solution, the compound crystallises out in yellow prisms, melting and decom- posing at 173". The substance dissolves in concentrated sulphuric acid forming a cherry- red liquid, which becomes light yellow on dilution. I n concentrated hydrochloric and nitric acids, it dissolves, giving orange-yellow solu - tions, which, on dilution, become paler. The dilute alcoholic solution is not fluorescent.PBRATOLUENEAZODIMETHYLAXJLINE. 887 It is obtained with difficulty perfectly free from water and alcohol ; before making the carbon and hydrogen estimation, it was allowed to stand a week in a vacuum over phosphoric anhydride.0.2093 gave 17.3 C.C. moist nitrogen at 17' and 748 mm. N = 9.44. 0.1790 ,, 0.5268 CO, and 0.1072 H,O. C = 80.26 ; H = 6.65. @,H,7N30 requires C = 80.37; H = 6-23 ; N = 9-70 per cent. These figures speak decisively for the orthosemidine conception. Behariour of the Base towards A-itrous Acid. As was to be expected from its orthosemidine nature, the new base reacts with nitrous acid, giving an azimide, according to the eqna- tion, C15H19N3 + HNO, = CJ316XTJ + 2HZO. If, to a dilute sollltion of the base in excess of hydrochloric acid, the equivalent quantity of sodium nitrite is added, a clear reddish solu- tion is obtained, from which a resinous precipitate separates on adding sodium carbonate.It crystallises out of light petroleum in needles melting at 88-89". It is t-erj easily soluble in alcohol aiid benzene. The amount of nitrogen found agrees with that required for an iizimide. 0.1650 gave 33.0 C.C. moist nitrogen at 21" and 7-52 mm. N = 22.52. C15H,,N, requires N = 22-28 per cent. This azimide i n acid solution is further acted on by uitrous acid. On adding sodium nitrite to its solution in hydrochloric acid, a yellow, flocculent precipitate is formed, which can also be obtained directly from the base by treating the acid solution of the latter with more than the equivalent quantity of sodium nitrite. This substance crystallised well, both from alcohol and benzene, in fine, hair-like ciytals, but, in spite of this, was very difficult to purify, the melting point of the different cry stallisations remaining always ill-defined 0.1524 gave 31.6 C.C.moist nitrogen a t 18" and 753 mm. N = 23.72. 0.2554 ,, 0.5763 CO, and 0.1149 E20. C = 62-08 ; H = 5.04. A nitroso-derivatit-e of the azimide requires C = 64.06 ; H = 5-34 ; 1 made no further investigation of the substance, as it appeared to (106-123"). 0.1510 ,, 32.3 ,, 9 7 ,, 17" ,, 752 ,, N = 24-56. N = 24.91 per cent. be of little value for the main purpose of nij- work.888 BOYD: REDUCTION OF Decomposition o j the Base ,with Hydrochloric acid. The orthosemidine nature of the base being established, there remained fo:. it two possible formulae, To decide which of the two is coryect, ;L clecomposition experiraent was tried. A mixture of 4 grams of the base, 10 C.C.of 1iydi.o- chloric acid (38 per cent.), and 10 C.C. of n-ater n-as heated set-en hoiirs in a closed tube a t a temperature of 1~0-160", the tube being filled with carbon dioxide, to prevent oxidation. The acid contents of the tube were then extracted repeatedly with sniall quantities of ether (A) ; a large excess of caustic soda was then added to the aqueous liquid, making it smell strongly of fatty amines, and it was again extracted x<th ether (B). The two ethereal solutions were evaporated to dryness separately. From solution (A), a very sniall quantity of colourless crystalliiie residue was obtained, sparingly soluble in benzene, from which i t separated in pretty crystals melting at 167-169". It dissolved easilj in water ; the aqueon s solution reduced silver nitrate, and on warming with chromic acid mixture evolved the pungent smell of quinone.From these observations the body is with certainty characterised as h ydroguiizoiLe. The residue from the ethereal solution (B), which had been dried with potassium carbonate, was distilled under diminished pressure (:30 mm.), the flask being heated to 220". St about 140", a colourless substance distilled over in considerable quantity (about 1.5 grain), and solidiEed a t once in the receiver ; it crystallised from light petro- leum, and melted a t 88.5-89.5. Both melting and boiling poiiit indicated tli e presence of 3 : 4-toZz~yZeiiediai~2i.i!e, which melts a t 88.5'. In order t o identify t h e substance with certainty, a portion was clis- solved in alcohol, and treated with a concentrated solution or' phenan- thrayuinone in glacial acetic acid. A thick, greyish-white crystalline precipitate formed a t once, which, after being twice crystallised from benzene melted a t 215.5-216.5". Hinsberg (dnnaZe7i, 237, :3:39) gives 212-213" as the melting point of toluphenanthrazine. In accordance with this author's description, the compound, when treated with concentrated hydrochloric acid, becomes red, and again coiourless on the addition of water. 0.2025. gave lie5 C.C. moist nitrogen a t 20" and 7.58 mu. N = 9.813. C21H11N2 requires 9 - 5 3 per cent.PARATOLUENEAZODI?dETHP LANILIAW. 889 Thus all the properties of this decomposition prodnct agree exactly with those of 3 : 4-toluylenediamine. From the nature of the products into which my base is resolved- hSdroquinone and 3 : 4-to1nylenedi:tmine-there can be no question as to its constitution. It is a t once apparent that the second of the two formulae given above (p. SSS) easilr explains a decomposition of t h i s kind, SH, OH A t I I + SH(CH)?, OH whereas the first is quite irrecoiicilsble with it.

ISSN:0368-1645    

DOI:10.1039/CT8946500879

出版商:RSC    

年代:1894

数据来源: RSC

摘要:

PARATOLUENEAZODI?dETHP LANILIAW. 889 L XX 1 I. - - Pwj, Q 7*c1 t i o t 2 of f-' : 3 - D iph en yll')2 ~ 7 0 7 0 ~ fi *om Beiizoin CI H d 1'1 inzn 1-y R e i i reiioid d tli i i r es. By Fi:.isc~s 1%. JAPP, F.R.S.. and T. S . MCREAP, D.Sc. XEYCXI and Berlinerblau Fere tlie first t o show (Ber., 20, R. 753) that indoles could be obtained by the interaction of a-lialogen- aldehydes, or x-halogen-ketones, with primarc benzenoid amines. Thus, from chloraldehyde and aniline, they prepared indole itself, :tiid from c1,loracetone and aniline, 2'-methylindole ('' rnethylketole "), tlie latter interaction occurring according to the equation Kurnerous other substituted indoles hare since been prepared in an a-1 a 1 o qo u s manner . It occurred to us t h a t it might be possible to replace the a-halogen- ketones in the foregoing class of interactions by z-hydrox)--ketones. W e therefore heated a mixture G f benzoin, aniline, and zinc chloride, and fonud that E.Fi>cher's 2' : 3'-diphenjlindo!e nas furnied according t o tlie equation I n like manner, by employing orthotolnidine, paratoluidine, a-naph- thylamine, and /3-naphthyClaiiiine, in place of aniline, we obtained respectively the corresponding diphenylorthotolnindolc, diphenyl- VOL. LXV. 3 s890 JAPP AXD MURRAY : 3' : 3'-DIPHESTLINDOLES FROM paratoluindole, diphenyl-a-nsphthindole, and diphenyl-p-naphth- indole. The investigation had proceeded thus far when a paper appeared by Bischler and Fireman ( B e y . , 26, 1336), in which the preparation of thwe indoles (with the exception of diphenyl-x-naphthindole) by n different method was described.This method consisted in first acting OD desylbrornide with a primary benzenoid amine in the colcl, so as t o obtain a desylanilide, thus C,H,*CO*C H(NH.C,H,)*C,H,. and tben boilinz tbis compound. with an excecs of amine, when, according to Bischler and Fireman, itt is converted into an indole. We were a t once struck by the close agreement of the melting points of Bischler and Fireman's slipposed new series of desjlanilides with those of an already known and obviously very similar series, which had been prepared by Voigt ( J . p r . Chem., [a], 31, 544: 34, 2), hy the action of primary benzmoid amities on benzoin. Thus, by heatkg benzo'in with aniline, Voigt obtained x compound which he at first described as " benzoynanidil '' (desFlanilide), but which lie afterwarils, and in our opinion on insuf€icient< evidence, formulated as anilhenzoyn, C,H,-CH( OH)*C(NsC,Hj)-C,H5. By substituting para- toluidine and R-naphthylamine for aniline, he obtained nnalogous con-pounds.We subjoin the two series with the meltiiig points assigned to them :- Voigt. 31. p. 99' 153 144 130 Anilbenzoin . . . . . . . . . . . Acetylanilbenzoin . . . . . . Parat olil ben zo'in . . . . . . . p-Naphthilbenzoin . . . . . Bischler and Firernav. Desylanilide ....... 97-9e" AcetyldeFj-lanilide. . 155 Desylpnratoluide . . . 143 Desyl-p- naphtlalide 131-132 31. p. We, therefore, pepared anilbenzoiin by Voigt's method and d e s j l - aniiide hy that of Bischler a n d Firmian, and found that the two preparations were absolutely indistinguishable.Simultaneous deter- niinations of the melting points in the same bath gave 9%--99" f o r both. Judginq, however, from the mode of formation from desjlbromide and amines, the compounds are in all probability desylaniiides, as assumed b-j- Bischler and Fireman. A s Voigt's method OF preparation is by far the simpler of the two, we resorted to it to obtain the substance required for our subsequent experiments. Pure desylanilide, thus prepared, was boiled for two hours with four times its weight of aniline, as prescribed by Bischler and Fireman, in order t o convert it into diphenylindole. To our surprise, not only was the greater part oE the desylanilide unchanged, but not The two series fire therefore identical.BESZOIN AND PRIMARY BESZENOID BMIYES.891 a trace of the indole had been formed, as was shown by the fact tbat the product gave no coloration when a trace of solid sodium nitrite was added to its solution in coricentrated sulphuric acid, whereas we have found that, diphenylindole, when thus treated, gives an intense bluish -,oreen colorattion. As, however, in another experiment, we were able to confirm Bischler and Fireman’s statement that desjlbromide wheii hoiled with excess of orthotoluidine yields diphen~lorthotoluindole, i t occurred to us that in the latter case the liberated l~ydrobromic acid might, hare effected the cocdensntion of the desylortbotoluide, which wonld be formed in the first instance, to the corresponding indole. We therefore repeated our experiment with desyrlanilide and aniline, bnt, before heating the mixture, added a little aniline hjdrochloride, the hydrochloric acid of which might be expected to act like the hpdrobromic acid in the precedincr case.Onder these altered cortdi- tions the condensation readily took place with elimination of water, and an excell~nt jield of diphen-j-lindole was obtained. We are, therefore, forced to the conchision that Bischler and Fireman must have employed in thSr experiments a desylanilide which had not been entire17 freed from t h e amine hydrobromide that is formed along Kith it in their method of preparation, and that t’he hydrobromic acid was responsible for the condensatiou which they attributed to the amine alone. Guided by these considerations, we next tried the effect of heating henzoi’n with a mixture of aniliue and aniline hydrochloride.Here Voigt’s desylanilide would be first formed, and would then undergo condensation to diphenylindole. As a fact, we foiind that an ex- cellent yield of diphenylindole could be thus obtained. The yield is better than by the zinc chloride method, as there is far less resinifica- tion. Substituting other benzenoid amines for aniline, we prepared by this new method the yarious other diphecSlindoles already referred to. The yield of diphenyl-p-napht’nindole, however, is very poor, since, as Bischler and Fireman also found, /3-dinnphthylamine is formed in large quantity. But a good yield of this indole may be obtained by the zinc chloride method. We further found that a11 these diphenylindoles are deposited from an acetone solution with 1 mol.of “ acetone of crystdlisation.” Most of these double compounds part very readily with their acetone; that of diphenyl-x-naphthindole, however, is permanent in air at the ordinary temperature, and this inclole also yields stable crystalline compounds with 1 mol. each of methylethylketone and diethplketone. The unstable acetone compounds cannot always be obtained with certainty ; sometimes the indole separates from the solution un- changed. 3 s 2In the following account of our experimental results, we describe, under the heading of each indole, only t h a t method of preparation which gave the best yield, except in tbe case of the first member of t h e series, diphenylindole, for the preparation of which both t h e zinc chloride method and t h e amitie hydrochloride method are given.E s P E P, I JI E STA L PA E T. 3'-Diphenylindole, Prepamtioil f?.o1n Reuzoih, Aiziline, and ZL'm Chloriile.-The pro- portions eiiiployed were : benzoin, 21 grams ; aniline, 15 grams ; ziiic chloride, 30 grams. The benzoin and zinc chloride were powdered together, mixed with the aniiine, and the whole was heated in a retort orer a free flame. The mass fused, water was given off, and as the temperature rose, tlie mixture became dark-coloured, whilst aniline distilled eyer along with tlie m-ater. As soon as the evolution 9f water had ceased, tlie fused mass was poured into water. boiled, first writ11 dilute hydrock loi.ic acid and tlieii with water, and extracted with boiling benzene.l'he tarrj mass which yecuained on distilliug of-f t h e benzene icas dist illecl under reduced pressure. The distillate, which was solid when cold and of a clear brown colour, was re- distilled; it came over about 273-273" under 17 mm. pressure, and was now of a clenr yellow colour. It was recrjstallised twice from benzene, from which it was depdsitecl in rosettes of oblique prisms, and aiterwards fi-oni light ptroleiim, froin which i t separated in large, snowball-like masses consisting of slender, colourless needles melting a t 123-124". The solutions diyplayed a blue fluorescence, as described by E. Fischer and b j Bischier anti Fireman. Calculated for GoH ,5S. Found. N ...... ..... 5% 5-16 per cent. Preparatio?L $-om Benzoin, AnI'linc7, and 14nilitre HydroclJo,-itle.-A mixture of 10.6 grams of benzoin, 16 grams of aniline, and 6.5 gyains of aniline hydrochloride ~ 1 s heated to i t s boiling point in a small flask fitted with a short upright air condenser. By regulating tlie neating, the operation could be so conducted that the water which \\-as formed during the reaction was driven off from the top of t h e con- condenser whilst the aniline flowed back itlto the flask.After heating for two hours no more water was given off. The fused mass was poured iuto water, and then shaken in a sepamting globe with ether ancL excess of dilute hydrochloric acid siniultaneou~ly, uiltii eve1-y-BEXZOIX ASD PRIMART BENZEXOID AJITSES. 893 t h i n g dissolved. The ethereal layer, which contained the diphenyl- indole, was separated ; the ether was expelled by warming, and the substance which remained was purified, first by distillation under reduced pressure, and subsequently by recrystallisation from benzerle and light petroleiim, as in the previous experiment.The diphenyl- indole thus obtained, agreed in all its properties with that already described. nark- coloured bye-products are not formed in this process, so that the substance is easily purified. d cetone Conzpoumi.-A solution of diphenylindole in acetone yielded, by spontaneous evaporation, crystals melting, when rapidily heated, between 80" and 90". 0.3373 gram lost, on heating a t loo", 0.0592 gram = 17.55 per cent. Calculated for C,oH15X,C,Hs0 : acetone = 17-74 per ceut. The yield was about 75 per cent.of the theory. They effloresced on exposure to air. 10.6 grams of benzoin, 16.5 grams of orthotoluidine, and 7.5 grams of orthotoiuidine liydrochloride were heated to the boiling point of the mixture for 2-3 hours. The crude product was extracted with boiling fairly strong h j drochloric acid to remove basic substances, and the residue distilled under reduced pressure. It was then further purified by recrystallisation. The different crystalline forms and different melting points which we have observed with this substance lead to the conclusion that it is trimorphous. Owing to the complexity of the problem we venture t o give the experimental details somewhat fully. Thus, when t8he crude redistilled substance was dissolved in benzene and light petroleum then added, it was deposited in rosettes of f l n ~ prisms with bevelled ends, melting a t 100-101". When recrjstalliscd from light petroleum, i t melted a t 101" ; again recrystallised from a mixture of absolute alcohol and light petroleum, a t 102'; again re- crystalltsed from the same solvent a t 102".A little substance from the previous crop was used t o start each crystallisation. This specimen, melting at 102," was analysed and gave figures agreeing with the expected formula C,,H,,K. calculated . . . . C, 89.05 H, 6.01 N, 4.94 per cent. Found .... .. . . C, 89.10 H, 6-22 N, 5-03 ,, After standing in a stopperrci bottle for t w o months the same specimen had become opaque, and was found to melt a t 136", pre- viously softening at 128". I t had thus spontaneuusly chaiiged intu a89.1 JAPP AXD 3lURRAY 2' : 3'-DIPHENYLINDOLES FROM mixture of the two modifications melting respectively a t 118" and 146", which we describe later on.Three specimens of diphenylorthotoluindole, which we prepared by three different methods-by the zinc chloride method, by the hydro- chloride method, and by Biscliler and Fireman's method of boiling desylbromide with orthotoluidine-all exhibited this behaviour ; when the freshly distilled prcduct was recrjstallised, no matter from which of the foregoing sol\-ents, crjstals melting a t temperatures bet weeu 99" and 102" were obtained; and these, on keeping, changed their melting point to 128-1336", as already described. This modification of low melting point was obtained only from the freshly-distilled crude substance." When the substance which had undergone the change of melting point was dissolved in a little boiling absolute alcohol, taking care that nothing was left undissolved, and then light petroleum added, the solution on standing depobited thick plates melting sharply a t 12SO.t Solutions in light petroleum, or in a mixture of benzene and light petroleum, yield the same forms, frequently, however, grouped together into rosettes.Recrjstallisation does not change the form or the melting pgint. This is the modification described by Bischler and Fireman. A third form, melting a t 136", may be obtained as follows :- A small quantity of pure snbstauce melting at 128' is introduced into a widish melting-point tube, which is tlien sealed up to prevent oxidation.The upper part of the tube is gently warmed, so a s to melt all adhering particles. These solidify when cold, but in an amorphous form, xhich does not cozplicate tbe szbsequent resuit. The tube is then heated in a sulphuric acid bath to 128", so as to melt nearly the whole of the cr~stalline substance, after which the temperatcre is allowed to fall to about 126--12i0, and kept there from 10 to 15 minutes. (If, through inadvertence, the whole of the substance has been melted, ii fresh tube of substance must be taken and the experiment started anew.) At this slightly lower temperature * Since writing the abore, we have made two further attrmpts to obtain thi. unstable modification, but without success. Two frcsh specimens of orthotolu- indolt! mere prepared, one by the hjdrochloride method, the other by the zinc chloride method ; but in both cases the solutions of the crude distillate deposited only the stable form melting a t 125".We are not sure that in all three prepara- tions above described the low-melting modification was obtained inclepende:rtlj-, i.e., without starting the crystallisation. It was first obtained from a preparation made by the zinc chloride method. f Very likely the unchanged pure substance melting at 102' would yield b- re- crystalljsation this modification melting at 1W, if it were allowed to crjstallise spontaneously, instead of starting it Hith its own crjstals. Unfortunately He had uo more of it to try the experiment.BENZOIN AND PRIMARY BENZENOID AMINES. 895 tbe substance begins to resolidify.Needle-shaped crystals are seen to form in the melted substance, which is thus ultimately converted into a mass of needles. Finally the temperature may be raised to 130°, to make sure that, none of the modification melting a t 128" is left. If now some pure substance melting a t 128" be dissolved in boiling light petroleum, and the crgstallisation be started with a few frag- inents of the foregoing resolidified substance, the solution will deposit large t u f t s of slender, colourless, silky needles, melting a t 1%', and these are in like manner capable of inducing their own form of crys- tallisation in other solutions. When a little absolute alcohol is mixed with the light petroleum in this experiment, well-developed, concen- trically-grouped, thin prisms are obtained, also melting a t 136".I n order to obtain the modi6cation melting a t 136": pure substance must be employed, for when the needles are introduced into a yellowifih soliition of impure substance in light petroleum, they do not grow, and tlie liquid deposits the tabular form melting a t 128". A solution of the pure acicular substance (m. p. 136') in light petroleum, when care is taken to leave nothing undissolved, and the crgstallisation is allowed to start spontaneousij, deposits the tabular form melting a t 123'." The following crystalline modifications of dipherylorthotoluindole are therefore to be distinguished :- z. Flat prisms with bevelled ends, melting at. 102". Gradually clianges at ordinary temperatures into a mixture of /3 and '1./3. Thick tabular crystals, melting a t 128". Stable a t ordinary te~iiperat~ures, but changed just below its melting point into el. -1. Acicular form, melting a t 136". Stable a t all temperatures up to its melting point. Transformed by recrjstallisation into /3. Aceto9ze - Compound.-A solution of diphenylorthotoluindole in metone yielded, by spontaneous evaporation, efflorescent, flat prisms, containiDg acetone of cr~-stallisatioii. When rapidly heated, they softened a t about 50", and melted at 90". CPlC02 gram, heated a t lllOo, lost 0.0267 gram = 16.66 per cent. Calculated for CZ1HliXI,C3H60 : acetone = l t * O l per cent. * The acicrilar form melsing at 136' was first accidentally obtained by us under c*onditions which we unfortunatrlg- did not note a t the time.I t was used to start subsequent crptallisations, and consequently, sc. long as u-e adhered to this practice, our specimeus, when pure, alwap had this melting point. Indeed, we assumed that this was the true melting point of the substance, and in a preliminary note (Be).., 26, 2641) called attention to the fact that Bischler and Fireman had not obtained a melling point higher than 128". W e were not a t that time aware that tlris *as the melting point G f a distinct fcrm. dfterw-ards, when we had recogniwd that all the supposed discrepancies were due to the existence of different modifica- tioil$, it, was some time before m e succeeded in discovering the conditions under n I1ic.h the a( icular form melting a t 136" could be independentlj obtained.896 .JAPP AND MURRAY: 2' : 3'-DIPHENYLINDOLES RROM 10.6 ,Tams of benzoin, 16.5 granis of paratoluidine, and 7m.7 graws of pamtoluidiue hydrochloride were heated together as in the p r c ~ vious experiment.The product was shaken with ether and hydro- chloric acid ; the ethereal solution was dried with calcium chloride and allowed t o evaporate spontaneously, when i t deposited needle- shaped crystals melting constantly a t 152-15.3". The substance re- maining after expulsion of the ether frcjm the mother liquor was distilled under reduced pressure, and further purified by recryst>al- lisation from a mixtizre of acetone and light petroleum. This product was identical with the foregoing. The total yield of pure substance was close upon 10 grams. Analysis gave results agreeing with the formula C,,H,;N.Calculated . . . . C, 89-05 H, 6.01 K, 4-94 per cent. Found ........ C, 88-81 H, 6-20 N, 4-91 ,, -4cetone Conzi;ound.-This substance mas obtained by the spon- taneous evaporation of a solution of diphenylparatoluindole in acetone. The melting point of the crystals was very indefinite, and they rapidly effloresced on exposure to air. 0.2788 gram, heatrd a t loo", lost 0.0484 gram = 1i.36 per cent. Calculated for C,,H,,N,C,K.,O : acetoue = 17.01 per cent. This compound was not prepared by Bischler and Firernail. 106 grams of benzoin, 21 grams of a-naphthylamine, and -1.5 grams of a-naphthjlamine hpdrochloride were heated to the boiling point of the mixture for 2+ hours. The proportion of hydrochloride is only half of t h a t employed in the previous experiments, as it was Found that the addition of a larger quantity prevented the fusion of the mixture.The product was treated exactly as in t h e preparation of diphenylindole by the hydrochloride method. The substance dis- tilled under a pressure of 10 mm. a t 315-330". The crude distil- late, which weighed 12.5 grams, was purified by recrystallisation from a mixture of acetone and light petroleum; this furnishes a practically pnre product. For analysis, it was subsequently recrys- tallised from a mixture of ether and light petroleum, and from aBENZOIX AXD PRIMARY BENZESOID AMlXES. 897 mixture of benzene and light petroleum. It formed tufts of slender needles, melting constantly a t 140-141". It is readily soluble in benzene, ether, and acetone, b n t only sparingly so i n light petroleum.Analysis gave figures agreeing with the formula CY4Hl7N. 0.1 566 gave 0,5162 CO, mid 0.07i2 H1O. 0.2021 ,. 7.6 C.C. moist nitrogen a t 1io and 762 mm. N = 4.37. C,,H,,N requires C = 90 88 ; H = 5-33 ; X = 4-39 per cent. We have not overlooked the possibility of formulating the fare- going compound as a peri-derivative of naphthalene. I n this caqe, however, the nitrogen ring mould consist oi six atoms, and the com- pound would not be aIi inclole, a class to which, bearing in mind its close resemblance to the other indubitable diphenylindoles here de- scribed, we have preferred to assign it. I t crgstallises from a solution of diphenyl-z-naphthindole in acetone, or even from the solution in a mixture of acetone and light petroleum, unless the petrolenm greatly predominates.It forms oblique plates, generally grouped together in large rosettes, melting between 90" and 100". The crystals do not effloresce when exposed to air. 06199 gmm heated a t 110" lost 0*0930 gram = 15.32 per cent. Calculated for C,,H,,N,C,H,O : acetone = 15.38 per cent. Compound with illethyl Ethyl Ketone.-Obtained, like the foregoing, using methyl ethyl ketone as a solvent. Large rosettes of flat prisms. Non-efflorescent . C = S9.89 ; H = 5-4s. Acetone Compound.--This is very easily obtained. 0-330.3 gram lost, on heating, 0.0616 gram = 14.46 per cent. Calculated for C,,HliN,C,H,O : methyl ethyl ketone = 18.39 per cent. Compound with Diethyl Ketone -Obtained like the acetone com- pound, using diethyl ketoue.Short, flat prisms or plates. Non- efflorescent. 0.1450 gram lost, on heating, 0.0309 gram = 21.31 per cent. Calculated for C,H,,N,C,H,,O : diethyl ketone = 61-21 per cent. /\A 2' : 3'-Diphe?iy1-/3-naphthindole, I 1 i ? \/\,, -y 9 s already mentioned, the hydrochloride method is not, any more than Bischler and Fireman's method, suited for the preparation of this compound. On performing the experiment in the manner de- scribed in t h e case of the other indoles, hardly any water is elimi-898 nated, and the product consists mainly of /i?-dinaphthylamine. The zinc chloride method, however, gives a satisfactory yield of the indole. 21 grams of benzoyn, 21 grams of P-naphthylamine, and 50 grams of zinc chloride were heated together in a retort, the operation being conducted, and the product treated, as in the preparation of diphenyl- indole by the zinc chloride method.I t was twice distilled under rsduced pressure. On the second occasion it passed over a t 330-340" under a pressure of 1.5 mm. It was recrystallised first from benzene, from which i t separated in minute, granular crystals ; then from a mixture of acetone and light petroleum, which deposited it in small rosettes of prisms of a greenish colour. The colour was removed by further recrystallisation from a mixture of alcohol and light petroleum and finally from alcohol alone. It melted constantly at 166-167". (Rischer and Fireman give the melting point 153-158", but state that, owing to the difficulty o f getting rid of the P-dinaphthylamine which is formed in their process, their product was only approsi- mately pure.) 2' : 3'-DIPHEXTLISDOLES FROM BENZOIS, ETC. Analysis gave figures agreeing with the formula CZ4HliN.Calculated . . . . C, 90.28 H, 5.33 N, 4.39 per cent. Found.. .. .... C, 89-90 H, 5.46 N, 4-35 ,, Of course this compound might be a /3/3-naphthalene deriratire r.2 : 31, instead of as here represented, an ap-derivati-re [l : 2j. Analogy, however, is strongly in faiwur of the assuniption which we have made. Acetone Con7poztnd.-By spontaneous evaporation of a solution of diphenyl-p-napbthindole in acetone, large prismatic cr_vstals contain- ing acetone of cry stallisation were obtained. They parted with their acetone so rapidly, especially on heating, that they gave only tbe melting point of tiie indole itself. For analpis they were dried between folds of filter paper. 0.7660 gram, heated a t loo", lost 0.1170 gram = 15.27 per cent. Calculated for C2rH-11,N,C3H60 : acetone = 15.33 per cent. Colour Reactioiis of the 2' : 3'-Diphenylindoles. With iVTL'Zrom czciJ.-X'hen any O F these indoles is dissolved in concentrated sulpliuric acid and a few small crystals of sodium nitrite are added, an intense bluish-green coloration is produced. With Renzotrichlor-ide aicd Ziuc Ch loride.-When the diphenT-1- indoles are heated a t 100" with beuzotrichloride and a little zinc chloride, the following cnlour reactions are obuerved. Diphenjlinclole. T' iolet.OXIDATION OF TARTARIC ACID IN PRXSEXCE OF IRON. $99 Diphen~'orthotoluindole Violet, changing to intense red. Diphenyl-a-naphthindole. Red, changing to dark brown. Diphenyl-P-naphthindole. Reddiah-violet. These colouring matt,ers dye silk without a mordant, but the shades Diphenglparatoluindole I have no brilliancy. Chemical Department, C'lbiversity of AberdetJn.

ISSN:0368-1645    

DOI:10.1039/CT8946500889

出版商:RSC    

年代:1894

数据来源: RSC

摘要:

OXIDATION OF TARTARIC ACID IN PRESEXCE OF IRON. 899 LXXZIl.-Oxictutioiz of Tcxriaric Acid iia prcseizce of Jrolz. By H. J. H. FESTOS, 51.8. WHEX tartaric acid in aqiieous solution interacts with certain oxidis- ing agents in presence of a trace of a ferrous salt, a solution is obtaiued which gives a beautiful violet colour on the addition of caustic alkali. I observed this change some years ago, and proposed it as a distinguishing test for tartaric acid. Tartaric and racemic acids are the only substances yet examined which behave in this way, citric, malio, succiuic, oxdic, and a variety of other acids having been tried with negative results. The soiution to be examined is mixed with a drop of ferrous sulphate solutiou, follon-ed by a drop of hydrogen pei~oxide, and I hen made alkaline with caustic soda or potash.Acids destroy the colour, but alkalis restore it. Excess of ferrous salt or of oxidising agent prevents or destroys the effect. It was observed subsequently that the iron may be removed f r o m the acid solution by precipitation with a ferrocyanide ; the filtered liquid has very powerful reducing properties, and, with ferric chloride and alkali, gives a violet colour, which is changed to a transient emerald-green bj- dilute miueral acids. The eifects are rery similar to those giveu by ferric chloride with pyrocatechin o r with phloroglucin. Chlorine water, hypochlorites, bayium peroxide, sodium peroxide, or potassium permanganate may be employed as oxidising agents in place of hydrogen dioxide, but the resiilts are not so good. Xitric and nitrous acids are inactive in this respect, as also, appa- rently, is ozone.If, however, ozonihed air is passed into ether, i t becomes strongly active ; this is, perhaps, due to the formation of ethyl peroxide. If moist ferrous tartrate, prepared by precipitation, is exposed to the air for about 10 minutes and then treated with excess of caustic900 FENTON : OStDATIOK OF TARTARIC ACID alkali, a faint violet coloration is produced, and from comparative experiments it would appear that fresh external air is more active than the air of a room. I hope, shortlF, to make further experiments in t h i s direction. If a strongly alkaline solution of a tartrate, containing a trace of ferrous salt, be electrolysed, using platinum electrodes, beautiful violet, strim will appear round the anode, and a strong solution of the substance may be produced in tllis manner if the temperature is kept low.Or, if a strong solution of tartaric acid is electrolysed, using an iron anode, the liquid turns yellow round the anode, and on the addi- tion of caustic alkali this changes to violet. In order to throw some light ou the nature of this reaction, experi- ments were made with standard solutions of tartaric acid, ferrous snlphate, and hydrogen peroxide. Measured volumes of these solutions were mixed in the order named, made alkaline with soda, diluted to the same volume in tall glasses, and the depths of violet colour com- pared. With equal quantities of tartaric acid and a tixed proportion of k o n , it mas o1)served t h a t the depth of colour increased with i n - creasing quantities of hydrogen peroxide up to a certain limit, beyond which the colour diminished as the peroxide increased, and beyond a certain point disappeared altoget>her.The mzximum was always produced when the ratio was 1 mol. of tartaric acid t v 1 atom, or‘ slightly more, of available oxygen ; and the complete disappearance of colour occurred when the ratio was 1 mol. of tartaric acid to 5 atoms of ava ilabl e ox jgen . I n Ihe next series of experiments, the proportion of iron was varied from 1 atom to -& atom for 1 mol. of tartaric acid, and it was found that the ratio of oxygen t o tartaric acid required to produce the maximum colour, and the discharge of colour, remained practically unchanged.From these results, it seems probiihle that the iron acts in a manner usually termed ‘‘ c2talytic,” a wry small quantity of iron being sufficient to determine the oxidation i n this direction of an almost unlimited amount of tartaric acid. Ferric salts are quite inoperative in bringing aLout the change, but if, in the first instance, the quantity of ferrous salt is very small, the colour produced by the alkali is greatly intensified by adding a few drops of ferric chloride. It would appear that ferrous salt is essential in producing the compound, and that ferric salt gives the violet coloration with the compound produced. Many attempts were made with a view of isolating this colour- giving substance. but, owing to the unstable character of the solution, the problem was an exceedingly difficult one.Exposure to the air, heating, or even evaporation in a vacuum, scein to destroy It.IS PRESENCE OF IROX. 901 Dialysis, in an inert atmosphere, fractional precipitation by variona metallic salts, and extraction by various solvents under different con- ditions, were also tried without success. SubsequenCly, whilst experimenting with rery strong solntions of ferrous tartrate and hydrogen peroxide, I observed that it was possible to extract a minute quantity of a substance which gave a violet coloration witb ferric chloride and caustic alkali, but the quantity was too small for further examination. I t seemed probable, from this result, that the substance divides itself very unequally between water and ether, and that the addition of a dehydrating agent moiild enable the ether to extract a larger quantity.This was tried with considerable success-strong sulphuric acid, dry hydrogen chloride, phosphorus pentoxide, alcohol, niany dehydrated salts, and even finely divided silica, very greatly increased the quantity of substance mhich ether was able to extract. On evaporating the ether, a white powder was left, generally xrnorphous, but sometimes semi-crystalline : i t mas sparingly soluble in cold imter, acted as a pomerfnl reducing agent, and, with ferric chloride, gave the characteristic reaction mentioned Fbbore. The siibslance prepared in this way gave somewhat variable results on analysis, and the yield was so small as to make the method almost prohibitive owing t o the cost of production.Whilst engaged in these experiments, however, it was observed that, occasionally, R minute quantity of brilliant, white crystals appeared in the liquid on standing after addition of the dehpdmtiug agent, and these proved, rJn examina- tion, to he identical in reactions with the substance extracted by ether. Their production was a t first very uncertain, but, recently, I have worked out the conditions by which theF ma)- be obtained with certainty aiid in considerable quantitj. The method Tvhich gives the best result is as follows:-Tartaric acid is dissolved in the least possible quantity of hot watei- ; finely-divided iron (ferrum redacturn) is added, and the liquid boiled until all the iron has disappeared. The quantity of iron must b e insufEcieiit to cause a separation of ferrous tartrate when tbe action is finished ; about part of the weight of tartaric acid employed amwers well, but the final result docs not appear to be much influenced by the proportion oE iron in solutiou, a t auy rate, witbin considerable limits.The solution, filtered, i f necessary, through cotton wool, is carefully cooled, sur- rounded by ice, and hydrogen peroxide (20 volume) added in small quantities a t a time, allowing a few minutes to elapse between eech addition. The first portions of the peroxide merelJ produce a yellowish colour, but, as the action proceeds, each addition produces a dark green or nearly black appearance, transient a t firat, but bFcoming more and more persistent. When this dark colour remains for two or three minutes, i t is a rough guide that sufficient peroxide has been902 FESTON : OXIDATION OF TARTARIC ACID added, although it is, perhaps, safer to palculate the quantity neces- sary from the proportiom idicated in the colour experiments men- tioned above.Great care must be taken not to add an excess of the peroxide, or the whole of the material mill be wasted. Nordhausen sulphuric acid is now added by means of a thistle funnel, drawn out to a fine point, in very small qnantities a t a time, cooling carefully between each addition, preferably by ice and salt. The quantity added is a matter of importance, too much o r too little giving an indifferent yield of the substance ; the best proportion is found by ex- perience to be about +h OF the total volume of the liquid operated on.The niixhre, still surrounded by ice, is put aside in a, cold place, and after a few hours crystals begin to form ; the first deposit is often dis- coloured 2nd the crystals small, but the subsequent crops are beauti- fuj1-y white and pure. If the experiment is properly conducted, and the liquid kept sufficiently cool, crystals continue to form for several days, but the greater part is deposited within about 24 hours. The crystals are collected with the aid of a pump, carefully drained, and washed repehtedly with small quantities of cold water. After again thorouglily draining, they art: spread on filter-paper and air-(?ried. They appear to undergo no change in the air, even after several weeks' exposnrz Propwfies.--The substance obtained in this manner has the appem- ance of a shining crystalline mass with a pearly lustre.The crystals are described by Jlr. Solly as follows :-" Thin, orthorhombk, diamnnd-shaped p;ates, 100 A 110 = 5 2 O 35'. Clearage perfect parallel to 010. The plane of the optic axes probably parallel to 001, the bwal plane. The direction of greatest optic elasticity parallel The substance dissolres very sparingly in cold water, ether, o r acetic acid, but more readily in etbylic or methylic alchol, or i n warm water, or hot acetic acid. It may be recrystallised from hot water. but with considerable loss, since decomposition, w i t h evolution of carbon dioxide, takes place rapidly at about 50--f;0°, and slowly even at ordinary temperatul-es. The crystals give off water when heated, or when placed in a vacuum over sulphuric acid, a, white, amorphous powder being left.If the dehydrated substance is dis- eolved in absolute alcohol, and a few drops of water added, a crystal- line precipitate of the hydrated substance is thrown down. The dehydrated sabstance begins t o decompose, without melting, a t abcut 155". The aqueous solution has a strongly acid taste, and gives an acid reaction with indicators. It rapidly reduces silver, cupric. and mer- curic salts, potassium permanganate, Pzc. Ferric clrlokde gives a blackish colour, changing to a beantiful violet on the addition of to 100,001."IN PRESEXCE OF IROX. 903 caustic alkali, and to a transient emerald-green when acidified with dilate snlphuric acid. Lead acetate gives a yellow precipitate, barium chloride a white crystalline precipitate, and silver nitrate a white precipitate quickly turning black.These precipitates all give the original substance on treatment with acids. Carbamide gives a white crystalline precipitat,e with the alcoholic solution. EIydroxylamine hydrochloride gives a crjstalline precipitate con- sisting of long, narrow, obliquely modified prisms ; but its formation is somewhat uncertain, and the conditions have yet to be investi- gated. Phenylhrdrszine appears to give three distinct compounds, perhaps more. The Erst is obtained as a yellow, amorphous powder on add- ing a limited quantity of alcoholic phenjlhydrazine to an excess of the dehydrated substance in alcohol. The second comes dowu slowly in beautiful silver-white, flat, oblique plates, m*heu excess of phenyl- hy&azine i n alcoholic or acetic acid solution is added to a cold dilute alcoholic solution of the hydrated substance ; when collected by the aid of a filter-pump it has a very brilliant appearance, almost like that of metallic silver.The third Compound is obtained in beautiful golden plates i f the solutions are heated after mixing. The second silver-lilre compound seems to change spontaneously into this golden substance after a short time, but whether this is clue to dehydration or to the action of light, &c., has not yet been made out. In aqueous solution the substance slowly decGmposes, giving off bubbles of gas, and after a few hours a weak solntion bas lost j t g original properties. When heated to about GO", the decomposition is very rapid, carbon dioxide is evolved with effervescence, and the solution ceases to give the ferric cliloride reaction after a few minutes.The solution which has been heated, leaves a gummy mass when exraporated in a vacuum over sulphuric acid ; this acts as a powerful reducing azent, giving a silver " mirror " with ammoniacal silver nitrate, and reducing Fehling's solution in the cold ; it also gives the '' aldelij-de reaction " with magenta decolorised by sulphurous acid, whereas the original acid does not. If the solution is boiled, neutralised with chalk, filtered, aniline oxalate added, and the liquid, after filtering, heated, an orange colour and precipitate is produced (Perkin's reaction). With dimethylaniline and mercuric chloride a blue colour is produced after heating for some time (Bottinger).From these results, gZyoxyZic acid seems to be indicated, but later experiments show that the quantity of acid is very small, the neutralising power being insigni6cant as compared with the reducing power. The main product appears to be an alde- hyde ; the latter is now being iirvestigated. If the original substance is heated with a concentrated solution of904 FESTOS : OXIDATION OF TARTARIC -4CID hydrogen iodide, a copious liberation of iodine takes place after a. few minutes. Three or four grams of t h e original acid were heated with excess of fuming hydrogen iodide at about 100" in a stoppered bottle, and small pieces of phosphorus were added from time t o time until the liquid became colourless.After heating on a water-bath until n~ost of the hjdrogen iodide was expelled, the liquid was evapo- rated t o d r p e s s , and, o n heating the solid residue, white fumes were evolved which condensed upon a cold surface as a crptalline mass ; this, on re-sublimation, yielded white, featberF crystals which nieited at abont 116". These, after cl-jstallisation from water, melted at 180-182", and hy optical examination and by t h e characteristic reaction with ferric chloride, were found t o be sziccinic acid. 0.1806 p r e 0.2666 CO? and 0.0782 H20. C = 40-25 ; R = 4.81. Succinic acid requires C = 40.67 per cent. ; H = 5.08 per cent. In ot+her experiments, it W ~ S obseryed that if the heating wit11 hydrogen iodide were not continued too long, a crystalline substance separated as the liquid cooled : on recrjstallisation from water, t!his was found to melt at about 191". It gires no reaction with ferric chloride, but if treated with ferrous sulphate and hjdrogen dioxide, in t h e manner described above, i t gires the tartaric acid reactiov.From its high melting point, racemic ucid was suspected, a n d this has since been confirmed by a careful optical examination by Mr. Solly. Analysis of the Origiriul CrystaZG.-Thoroughly air-dried specimens gave the following results on combustion. I. 0.2262 gare 0 21.58 COz and 0.0878 H1O ; C = 26-01 ; H = 4.31. 11. 0.16i3 gave 0.16313 GO, and O.cIG42 H,O ; C = 26.13 ; H = 4.26. III. 0.2201 g a r e 0.2094 GO, and 0.0829 H,O ; C = 2;-94; H = 4.18. The specimens ana12-sed in I and I1 were obtained as t h e third crop of crystals in t h e original preparation, and that in 111 was a specimen which had been recrystallised from hot water.E ~ t i m a t i ~ n of TJ'ufu.-The substance was placed in a vacuum o r e r sulphuric acid until the weight was constank. I. 0.5743 lost 0.1097 H,O = 19.!0 per cent. 11. 0.4933 lost 0.0941 H,O = 19.22 per cent. The weight became constant after two days or lees, and no further Heated in a current of d r y hjdrogen until t h e weight was con- 111. 0.,7477 lost n.lOZ3 H,O a t $0" = 19-59 per cent. IV. 0.1 G 1 1 I n s t 0.0314 F20 at 60" = 3 9.413 per cent. V. 0.2041 lost 0.0399 H20 at 90' = 19.54 per cent. loss took place eren after f o u r weeks. s t n u t .IN PRESENCE OF IRON. 905 No further loss occurred at 100".It appears from these results that the substance cannot be com- pletely dehydrated at ordinary temperatures in a vacuum over snlphuric acid. Analysis of the Dehydrated Sz~bstance.-The residue from Experi- ment I V (0.1297 gram), on combustion, gave 0.1546 CO, and 0.0308 H20 ; C = 82.40 ; H = 2-63 per cent. The residne from Experiment V (0.1642 gram), on combustion, gave 0.1966 CO, and 0.0393 H,O ; C = 32#65 ; H = 2.61 per cent. The specimens dried in a raciium, which were analysed in the first instance, gave an averag3 of 30.8 per cent. C and 2.8 per cent. H. Methylie Salt.-To prepare this compcund. the origiual substance is dissolved in the least possible quantity of methylic alcohol (about 1 i n 5 ) and dry hydrogen chloride passed into the solution.A white, amorphous snbstance begins to deposit as soon as the liquid is satu- i-nted with the gas, and continues to come down for some hours, after which it is deposited in a beautifully crystalline form. The liquid is then drained off b3 means of a pump, the solid washed with small quantities of cold methylic alcohol, and dried in a steam cupboard. It may then be recrystallised from boiling methylic alcohol, glacial acetic acid, or benzene. The crgstals originally produced, and often on first recrystallising from methylic alcohol ( a ) , consist of long, slender, oblique prisms. a makes with the long edges of the prism an angle of 37i". In convergent light gives an unsymmetric optic picture. When recrystnllised a second time ( b ) , the crystals are very different in habit; they consist of a number of very nearly square plates twined together, and built up in rows nearly at 90" apart.The first mean line is apparently perpendicular t o the basal plane. The second mean line makes aC angle of 17" with a prism edge. Both these modifications give identical results on analysis, as will be seen below. The difference may possibly be caused by a trace of some impurity, but if so the quantity is too small to be detected. When heated to about lso", the substance may be easily vaporised, and condensed upon n cold surface in crystals which are identical with those obtained by repeated recrystallisation ( b ) . Heated in a capillary tube, these modifications all melt at about 151a, but the exact point is difficult to observe owing to the rolatility of the substance.I t is very sparingly soluble in cold acetic acid and ether, rather more so in cold methylic or ethylic alcohol, and fairly easily in these liquids when hot. When treated with caustic alkalis, a lemon-yellom coloration is produced. It^ ferric chloride is added to the alkaline mixture no effect occurs at first, but after a fern minutes the characteristic vioiet Optically positive. Optically megatire. VOL. LXV. 3 T906 FENTON : 0XIL)ATION OF TARTARIC: ACID colonr is produced. ' ' saponification " of the ethereal salt. This effect is evidently due to the gradnd I. 0.1619 gave 0.2386 COz and 0.0644 H20; C = 40-19 ; H = 4-41. 11. 0-1506 gave 0.2249 CO, and 0.0568 HzO ; C = 40.72 ; H = 4.18. 111. 0.1824 gave 0.2733 GO, and 04729 HzO ; C = 40.86 ; H = 4-44.IV. 0.2224 gave 0,3331 COz and 0.0897 HzO ; C = 40.84 ; H = 4-48. I was sublimed on a watch glass, I1 and I T 1 mere the result of a first recrystallization, and IV of a second recrystallisation. Ammonium Salt.-Aqueous ammonia is added to a solution of the acidin dilute alcohol until it is just alkaline to litmus paper. The salt begins to separate after a few minutes. The crystals consist, of flat, nearly square, plates. Optically positive. The first mean line inclined t o the basal plane, as only one optic axis with rings is seen. The salt dried in a vacuum does not lose weight when heated in dry hydrogen at 90". 0.2108 gave 0.2023 GO, and 0.1086 H20 ; C = 26-1 '7 ; H = 5-72. 0.1751 gave 22.6 C.C. of nitrogen at 17" and 757.3 mm.; H = 15-19. 0-3495 gave 45 C.C. of nitrogen at 19" and 755.6 mm. ; N = 15-01. Sodiuln XaZt.-This may be prepared either by adding alcoholic soda t o a solution of the acid i n dilute alcohol until alkaline, washing with a little water and then with alcohol, or by neutralizing the acid with sodium carbonate, when crystals separate after an hour or two if the solution is sufficiently strong. They are long, narrow, obliquely developed prisms. It does not appear to be altered in a vacuum over sulphuric acid, and the salt dried in a vaciium does not lose weight even at 140" in a current of dry hydrogen. The second meau line is parallel to one of the prism edges. a parallel to long edge of the prism. The salt is somewhat sparingly soluble in cold water. Analysis gave the following results. I.0.2269 gave 0.2031 C02 and 0.0260 HzO ; C = 24.41 ; H = 1.27. 11. 0.1985 gave 0.1441 Na2S04 ; Na = 23.51 per cent. $11. 0.14.53 gave 0.1043 Na.SO4 ; Na = 113.25 per cent. IV. 0.2976 gave 0.2142 NazS04 ; Na = 23.31 per cent. Barium Salt.-The crystalline precipitate, obtained by adding barium chloride to an aqueous solution of the acid, mas dried under diminished pressure at 50 -60" until the weight, was constant. 0.8941 gave 0.5520 BaC03 ; Ba = 42.93 per cent. Fornazda for the Acid. The results of the above analysis agree closely with the emppical formula, C&Os, for the dehydrated acid, and C2H203,Hz0 for tbeIN PRESENCE OF IRON. 907 crystals. It is evident, from the composition of the salts, that the number of carbon atoms in the molecule must be even, since two atoms of carbon are present for each atom of “replaceable ” hydrogen. The direct production of the acid from tartaric acid, and the forma- (ion of succinic acid by interaction with hydrogen iodide, suggest the formula C1H406, whereas some of the properties of the acid might lead one to expect a higher multiple.It was important, therefore, to obtain conclusive evidence as to its molecular weight. Attempts were made, in the first instance, to determine the vapour density of the methylic salt. Hofmann’s apparatus was employed, and the heating liquids tried were amylic alcohol, crude xylene, oil of turpentine, and aniline. With the two first named, the vaporiza- tion of the substance was so slow as to make the determination practically impossible, and with both turpentine and aniline there were signs of decomposition.The experiment had, therefore, to be abandoned for the time, since no liquid boiling a t the required temperature was available in sufficient quantity. Dumas’ method was also tried, using an oil bath at 150-155”, but owing to the slow rate of vaporization it was impossible to displace more than about 4 of the air. The next experiments were made by a vapour pressure method first suggested, I believe, by Ostwald, in the apparatus and after the manner recommended by Will and Bredig (Bey., 1889, 1084). A current of dry air is drawn first through the solotion of the substance in absolute alcohol, and then through alcohol alone, the loss of weight of each being determined. The air after being thoroughly dried is led through (u) a leaden spiral, ( b ) a set of three potash bulbs sealed together containing the solution, ( c ) a similar set containing the solvent, ( d ) an empty U-tube, and lastly through a drying apparatus to the aspirator.The parts of the apparatus a, b, c, and d, are immersed together in a large bath through which a very gentle stream of water flows from a cistern. Will and Bredig use a motor to keep the temperature uniform, but the arrangement men- tioned seemed to answer almost as well, and the temperature of the bath did not alter more then 2” during a determination. About 18-25 litres of air were drawn through the apparatus in each experiment at the rate of 6 to 1 litre per hour. In some experiments the air was passed through alkaline ferrous tartrate and alkaline p~rogallol in t’he first instance, in order to avoid any risk of oxidation, but this precaution appeared to be unnecessary.If S, = the loss experienced by the solution containing W1 grams uf substance in Wz grams of solvent, and S, = the loss experienced by the solvent, i t follows, from the relation -- - that P N + n’ P-PP, - n 37’2998 FENTON: OXIDATION OF TARTARIC ACID the molecular weight = -- MS1wl, where M = the molecular weight of the solvent. absolute alcohol as solvent. S*W2 The following results were obtained using the dried acid, with [M = 461. S,. IT-,. IT.,. 3101. wt. 8,. I. 0.8106 0.0215 3.8256 36-08 1% 11. 1.9974 0.0339 5.3359 62.25 232 111. 1.2922 04313 4.2572 52.68 15.3 IV. 2.0130 0.0743 4.4163 41.70 131 It is not surprising that the results should differ considerably, since an error of a few milligrams in weighing the bulbs containing the solvent mould make a very great difference in the result, and such an error is probable considering tbe bulky nature of the apparatus to be weighed, and the conditions of the experiments.Still the average number pointed clearly to the formula C4H40,. Freezing point determinations were hardly applicable to the acid itself owing to its verg sparing solubility in cold acetic acid, water, benzene, melted phenol, naphthalene, stearin, &c. The boiling point method was, therefore, next tried, in the ap- paratus recommended by Sakurai (Trans., 1892,61, p. 994). Absolute alcohol, dehydrated by copper sulpliate and boiling a t ‘78.3”.mas used as the solvent, and the dried acid was examined. I n order to estimate the contents of the solution after the experi- ments I, 11, and 111, thc weighed quantity of solution was diluted with water and titrated with soda, which had beeu standardized by the same dry acid. In experiment LV, the quantity was estimated by weighing in a vacuum the residue left on evaporation. (Mol. wt. = 148.) The following results were obtained. Boiling point. r-”- --, Grams of Grams of SolTent. Solution. substance. solvent. ,%ol. u t . I. 3.510 3.390 0.1.50:3 lW4EXi 2c5 11. 3.555 3.80t5 0.5177 17.0695 139 111. 3.330 3.500 0.3047 20.5181 195 IV. 3-260 3.430 0.2311 12.3892 126 The alcoholic solution of the acid had rather a tendency to froth u p on boiling, which made the experiments somewhat difficult.The most satisfactory results, however, were obtained with the methglic salt ; this substance, although very sparingly soluble in cold acetic acid, or benzene, dissolres fairly well in melted phenol 01‘IN PRESENCE OF IRON. 909 melted naphthalene, so that it is possible to determine its molecular weight by the freezing point method. The following were the results with phenol as solvent. Freezing point. r- -, Grams of Grams of Solvent. Solution. substance. phenol. Mol. wt. I. 3.200 3.015 0.0593 11-62 209 11. 3.200 2.560 0.1820 11.62 185 111. 4-640 4.335 0.1 904 27-36 188 IV. 4.672 3.900 0-2 758 1214 179 and with naphthalene as solvent, V. 4.075 3.735 0.1257 13-48 189 The " constants " for phenol and for naphthalene are taken as 76 and 69 respectively.The calculated value of the molecular weight for C4H2(CH3),0, is 176. The dry acid must, therefore, be represented as a bibasic acid having the formula C1H406, and the original crystals as C,H40,,2H20. I n the following table, the results calculated from this formula arc compared with those obtained by analysis. It will be observed that the agreement is everywhere very close except in the case of the sodium salt, where the results are somewhat low ; it would seem, therefore, that it retains water, even at 140". The barium salt agrees exactly with the calculated result if it be considered as the hydmted normal salt. Original Crystals. Found . T-i- -3 Calculated for I. 11. 111. C4H40,j,2 HZO. C .......... 26.01 26-13 25.94 26.08 per cent.H.. ........ 4-31 4-26 4-18 4-02 ,, H20 ........ 19-59 19-49 19-54 19-56 ,* D r y Acid. Found. f--J--- Calculated for I. 11. G H 4 0 6 . C . . .......... 32-40 32-65 32.43 per cent. H ............ 2.63 2.61 2.70 ,, Jfethylic Salt. Found. (.--A- 7 Calculated fm I. 11. 111. IV. C,HZ(CH,) 2 0 6 . C . . ........ 40.19 40-72 40.86 40.84 40.99 p. c. H.. ........ 4.41 4-18 4-44 4-48 454 ,,910 OXIDATION OF WTARIC ACID IN PRESENCE OF IROX. Ammonium Salt, Found. r-- 7 I. 11. 111. C .......... 26.17 - - H .......... 5-72 - N .......... - 15-19 15.01 Sodium Salt. Found. I. 11. 111. I+. C .......... 24.41 - - - H .......... 1.27 - - - Na ......... - 23.51 23-25 23-31 Calculated for C4WNH.4) 2 0 6 . 26-37 per cent. 5.48 ,, 15.38 ,, Calculated for c* El,h'Eb,OG. 25.00 p. c. 1.04 ,, 23.95 ,, Ba&m Salt. Calculated for Found. C4H2Ba06, 2H,O. Ea ............ 42.93 42.94 per cent. Titration of the Acid with Sodn.-0.6431 gram of the original crystals dissolved in water, and titrated with soda (prepared from metallic sodium) containing 0.007268 gram Na per c.c., required 22.8 C.C. for neutralization, phenolphthaleh being used as indicator. = 0.1657 gram Na. Theory for a bibasic acid of the formula C,H,O, = 0.1620 gram Na = 22.29 C.C. The properties of these two acids are very different ; dibydroxy- tartaric acid (Miller, Ber., 22, 2015) crystallises in prisms, melts and decomposes at 98", is very easily soluble in cold water, and gives no reaction with ferric chloride. The " constitution " of the acid might, of course, be easily con- jectured from the results above recorded, but I prefer t o leave the consideratiou of this point to a future communication, in which I hope to give an account of its phenylhydrazine and hydroxylamine compounds and of the interaction of the acid with acetyl chloride and other agents. In conclusion, I wish to express my best thanks to Mr. It. H. Solly, M.A., Demonstrator of lbfineralogy, for his kindness in examining and describing the various specimens of crystals. University Chemical L a born tory, Cambridge.

ISSN:0368-1645    

DOI:10.1039/CT8946500899

出版商:RSC    

年代:1894

数据来源: RSC

摘要:

911 LXXIV.-me SpeciJic Character of the Ferrnentatiue Functions of Yeast Cells. By ADEISN J. BROWK, THE usually accepted view regarding the cause of the exhibition of the fermentative functions of yeast cells is that it is the result of cell-life in a fermentable medium when deprived of free oxygen. This well-known theory concerning fermentation is due to M. Pasteur. In a paper on the influence of oxygen on alcoholic fermentation (Trans., 1898, 61, 369), I described some experiments the results of which appeared to disagree with M. Pastenr’s theory, and others, carried out more recently, have emphasised this disagreement. Before enki-ing on this question, however, i t is as well to recall briefly Pasteur’s experiments and his interpretation of their results, the most recent and complete account of which is to be found in his 2hudes sur la biire, Chap.VI, “Thborie physiologique de la fermentation.” In this recapitnlation, no reference will be made to Pasteur’s experiments on the growth of moulds, fungi, and bacteria, in fermentable media, as it is by these he seeks to make his theory universal for all cell-life; this enlarged view of his theory is not under consideration, it being proposed to discuss merely fermentation caused by yeast cells. If the truth of the theory as regards yeast is not proved by experiments with yeast, it cannot be proved by experi- ments with perfectly distinct forms of cell-life, a1 though the prob- ability of the theory being correct may possibly be somewhat strengthened by analogy. The experiments brought forward by Pasteur in support of his theory may be divided into two classes, namely, one, those which appear to suggest the probability of his theory being correct, the other, and most important, those made with the object of proving its truth.This arrangement is not Pasteur’s, but is adopted in order to bring out briefly the salient points in his argument, which might otherwise appear somewhat involved. The first class, just referred to, includes Pasteur’s experiment, by means of which he showed that a considerable weight of oxygen is absorbed by yeast cells growing in a thin stratum of fermentable liquid exposed to the air (&tucks, p. 248) ; this fact is established by it, but no knom1ed.g.e is gained concerning the fermentative power of yeast cells under the given circumstances.The first class also includes an experiment made by gowing a trace of yertst in a solution o€ milk-sugar exposed to the influence of air (Zoc. cit., p. 257), whereby912 BROWN: THE SPECIFIC CHARACTER OF THE he shows that yeast grows at the expense of milk-sugar under these circumstances, and absorbs oxygen and exhales carbon dioxide during its growth. The agrobic or ‘‘ fungoid ” mode of yeast growth shown in the previous experiment is thereby confirmed, but as milk- sugar is not fermentable by ordinary yeast forms under any circum- stances, nothing is ascertained regarding the fermentative functions of yeast cells. On p. 232, an experiment belonging t o the first class is described, showing that yeast cells grdw freely in a fermentable solution out of contact with air, and that a brisk fermentation is carried on at the same time.The experiments included in the first class, and just described, conclusively demonstrate that yeast cells are capable of a dual mode of existence; namely, one carried on in the presence and by the aid of free oxygen, and the other carried on without the influence of free oxygen. So far, however, the onIy knowledge we have gained is that fermentation is carried on by yeast during its anaerobic state. It is most noticeable that living organisms very generally require a supply of oxygen to enable them to carry on their life functions. The knowledge gained from the experiments just mentioned, that yeast can live without free oxygen, and that at the same time it carries on fermentation, is therefore suggestive that fermentation may be intimately connected with the requirement of oxygen by the yeast.Pasteur’s experiments described so far suggest this, and indicate a working hypothesis that might lead to an explanation of the cause of fermentation, but they carry us no further than this. The experiments by which Pasteur establishes his theory of the cause of fermentation will now be described. Pasteur claims that fhey show that the fermentative power of yeast cells is at a maximum in the absence of free oxygen, and diminishes to zero when fully exposed fo itR influence. The proof of his theory that fermentation is “life without air ” (loc. n’t., p. 261) rests entirely upon these conclusions. But i n order that his experiments may bear out the conclusions he maintains, two points are essential : first, that the ratios of the weight of yeast formed to sugar fermented should be numerical expressions of the fermentative powers of the yeast ; and secondly, that these ratios should be comparable with each other.For by comparisons between these ratios, Pasteur determines the relative fermentative powers of the yeast cells under varying conditions of experiment, and established his theory upon the conclusions thus arrived at. If, then, the rat,ios of yeast to fermented sugar do not express the true fer- mentative powers of the yeasts, or if’ these ratios are not comparable, Pasteur’s deductions fail, and his theory remains without experi- mental proof.FERMENTATIVE FUNCTIONS OF YEAST CELLS.913 In the first of Pasteur's experiments (Eoc. cit., p. 234) two large Basks, called A and B, containing equal volumes of a 5 psr cent. Bola- tion of cane sugar in yeast water were used. A was arranged so that air could not gain access to the contained liquid, and the liquid was boiled to drive off all air held in solution. The liquid in B, on the contrary, was exposed to the influence of the air. The solutions in both A and B were inoculated at the same time with a cultivation of yeast, aud allowed to ferment at 25". Both showed signs of fer- mentation in 24 hours, but in B, the flask with air, the fermentation proceeded the more rapidly, and by the 10th day W:LS finished; whilst in 8, a languid fermentation wm still apparent nine days later. Both flasks were opened at this time, and their contents examined.In A, it was found that 4.6 grams of sugar were still left unfermented ; but in B not a trdce of the 150 grams of sugar it rigi in ally contained was left updecomposed. The weight of dried yeast found in A was 1,638 grams, and in B 1.970 grams. From these figures, a proportion of 1 of yeast to 89 fermented sugar was found for A, the experiment without air; and 1 of yeast to 76 fer- mented sugar for B, the experiment made in the presence of air. If these ratios, expressing the fermentative powers of the yeasts, are com- pared, the yeast under the influence of air appears to haye exhibited less power than the other. Pasteur next describes an experiment made under somewhat similar conditions to those of A, but in which el-en greater precautions were taken to exclude air.The fermentation was started with an inocula- fion from an old cultivation of yeast having little vigour. The feeble fermentation apparent i n this experiment continued for three months, and when the flask was opened a slight fermentation was still noticeable. In this experiment, 0.255 gram of dry yeast was found, and 45 grams of sugar were fermented out of the 150 grams origin- ally present. The proport,ion of yeast to sugar fermented under the anasrobic conditions of this experiment was therefore 1 : 176. Another experiment. C, in which the conditions of asration were increased beyond those i n B (Zoc. cit., p. 242), is also described. 200 C.C. of a 5 per cent. solution of cane sugar in yeast water were fermented in a large flask so that the solution was exposed as a thin stmtum to the influence of the air.Under these conditions, the fermentation pro- ceeded rapidly, and all the sugar was speedily deconiposed. 0.44 gram of yeast was found, and 10 grams of sugar were fermented, which amounts are in the proportion of 1 : 23. The experiments just referred to show that if fermentative powers expressed by the ratios of yeast formed to sugar fermented are com- parable, then the more ,?Grated the environment during fermentation the less the fermentative power of the yeast becomes.914 BROWN: THE SPECIFIC CHARACTER OF THE Pastenr describes two more experiments beazing on this point, in which the conditions of asration were increased beyond those of Experimect C.In the last of these he succeeded in obtaining a ratio 85 high as 1 of yeast to 4 of sugar fermented. These experiments will be referred to in detail later on. Concerning them, however, he makes the following observations (Zoc. cit., p. 244) : ’‘ Dans ces conditions CaLiration], la fermentation du sucre est des plus faibles ; c’est presque le rapport auquel donnent lieu les moisissures akrobies vnlgaires ; l’acide carbonique qui se d6gage est form6 en giande partie par les combustions qui rksultent de l’assimilation du gaz oxyghe de l’sii.. La l e d r e v i t alors et agit B la manihre des moisissures. Elle n’est plus ferment pour ainsi dire, et il est sensible qu’elle ne le serait plus dn tout, si l’on pouvait entourer chaque cellule isol6ment de tout l’air qui lni est nkcessaire.” These words indicate the conclusions Pasteur draws from the experiments.In considering the experiments just described, it will be noticed that the ratios of yeast to fermented sugar are considered to be numbers expressing fermentative powers which are comparable with each other; it is obvious, therefore, if this is correct, that these ratio- unmbers must either represent the whole power, or s3me known fraction of the power of the yeasts, for otherwise no true comparison could be made. No suggestion is made by Pasteur that the numbers he makes use of are fractional representations of the fermentative powers of yeast, and there is no doubt that they must represent, according to his arguments, the whole fermentative power of which the yeasts concerned are capable.In the experiments A and 13, just described, in A, that without air, the fermentation, after exhibiting moderate activity, gradually de- clined until the 19th day from its commencement, when it was very languid indeed. The experiment was terminated a t this stage, and on opening the flask it was found that although the greater part of the sugar originally present had been fermented, 4.6 grams remained in the solution undecomposed. In t h i s experiment, the yeast found was in the presence of fermentable sugar, on which it could exercise its fernentative powers during its whole existence. In B, carried on nader the influence of air, the fermentation proceeded briskly for a few days, and was quite over on the 11th day. When the contents of the flask was examined, it was found that no sugar remained un- fermented in the solution. Fermentation must end under such circumstances.In Experiment A, the yeast exhibited a fermentative power which slowly declined in a fermentable solution ; whilst in Experiment R, exhibition of the fermentative power of the yeast ceased when there was no more supm to ferment. If. at the termination of Exoeri-FERMENTATIVE FUNCTIONS OF YEAST CELLS. 915 ment A, fermentation had quite ceased, it might be assumed that the yeast had exerted all the fermentative power of which it was capable, because there was still sugar in the bolution on which it could act; but Pasteur states that the fern?entation was still proceeding in a very feeble manner. In such case, however, the ratio of the yeast to sugar fermented may perhaps be considered an approximate repre- sentation of the total fermentatii-e power of the yeast in A.But in B there is nothing to show that the ratio of yeast to sugar fermented is even an approximate representation of the total fermentative power of the yeast formed. It is only by means of the decomposition of sugar that fermentative power can be measured, and if, as in Experi- ment B, all the sugar is fermented, there is no means of ascertaining if the yeast present still retains fermentative power. Yet, according to Pasteur, the ratio of yeast to sugar fermented in B is comparable with the ratio found for A. If the weight of yeast formed during fermentation increased in direct ratio to the weight of sugar fermented, the proportion of yeast to sugar fermented would remain constant no matter what qiiantity of sugar was present.Under these conditions, the relative fermenta- tive powers of yeasts, as determined by Pastenr, would be com- parable. But the weight of yeast formed during fermentation is but slightly influenced by the amount of sugar it ferments ; the weight of sugar fermented is only one factor governing yeast production in a fermentable liquid, and comparatively a minor one. I have called attention to this fact before, but describe some experiments on this point which have been made more recently, and which are i n closer agreement with Pasceur's method of experimen t. Five similarly shaped flasks of 400 C.C. capacity were prepared, each containing 100 C.C. of yeast water, and varying weights of cane sugar in the following amounts.Yeast water. Grams of cane sugar. No. 1 flask .......... 100 C.C. 2.5 , , 2 ,, .......... 3, 5.0 , , 3 ............ 7 7 10.0 , , 4 ............ 9 7 20-0 , , 5 ,, ,? 30.0 .......... The flasks were closed with plugs of cotton-wool and sterilised in the usual manner. On February 23, each flask was inoculated with a trace of a pure culture of high-fermentation yeast. On February 28, fermentation was orer in S o s . 1 and 2 flasks, the whole of the sugar originally present i n each being decomposed. The number of yeast cells present in the solutions was ascertained by counting, and the total weight of the cells bg filtration through a weighed filter, drying at lOS", a i d weighing.81 6 BROWN: THE SPECIFIC CHARACTER OF THE No.of enpt. 1 2 3 4 5 On March 2, fermentation was over in No.3 flask, and no sugar remained in the solution. Yeast dcterminatioris were made as i n Nos. Z and 2 flasks. On March 7, fermentation was over in No. 4 flask, no sugar remaining undecomposed. Yeast determinations were made as above. On March 15, No. 5 flask was opened, after fermentation had ceased for several days. On examination, the solution was found to contain unfermented sugar. An estimation of it* was made, which showed that 4-83 grams remained undecomposed out of the original 30 grams present a t the commencement of the experi- ment. Yeast determinations were made as in the previous experi- men ts. The results of the foregoing experiments are tabulated below. Grams of sugar in solution a t Grams commence- of sugar experiment.ment of fermented. ------ 2 -5 30 -0 10 -0 20 -0 20 *@ 30.0 1 25'17 I Weight of Feast found, in grams. --- 0 'I 2-10 0 *1550 0 *17i5 0 '1 400 0 -1380 No. of cells in standard voiume of qGlGG cb. mm. 8 *51 9 -94 10 -44 l I . 1 i 12 26 Ratio of yeast to suear fermented. --- 1: 20'2 1: 32.3 1: 56.3 1 : 142 -9 1 : 182 -4 These experiments show how little the amount of sugar fermented influences the weight of yeast formed. The weight of yeast in- creases a liktle as the strengths of the sugar solutions increase as far as the 10 per cent. strength, but there is a falling away in weight as the solutions become stronger.? The increase of weight observed in any experiment is very small, and distinctly disproves the idea that weights of yeast formed during fermentation increase in direct ratio to the weights of sugar fermented.In the last column of the above table, the ratios of yeast t o fermented sugar for the different experiments are given. These ratios, according to Pasteur, represent the fermentative * The sugar prored to be nearly pure levulose. t The numbers of cell8 determined increase slightly up to the 10 per cent. strength of solution, iu a similar manner to the weights of jeast found; but, curiously, beyond this point the numbers still increase, whilst the weight dimin- ishes. Microscopic measurement showed that tlie cells in Experiments 4 and 5 were distinctly smaller than those in tlie other experiments, a result which appears probable from a comparison of the numbers of cells and the weights of yeast found.FERblENThTIm FUNCTIONS OF TEAST CELLS.Yl? powers of the yeasts, and should be comparable with each other. From this point of view, if the ratio determined for Experiment 2, -,vhich is 1-32, is compared with that of Experiment 4, which is 1-143, i t will be found that the yeast in Experiment 4 had more than four times the fermentative power of the yeast in Experi- ment 2. NearlF equal weights of yeast were formed in both these experiments, and in both the whole of the sugar present was fermented. Now as the conditions of the two experiments only differed in so far that the sugar solution in Experiment 4 con- tained four times more sugar than the solution iu Experiment 2, it appears much more probable that the yeast in Experiment 4 exhibited four times the fermentative power of the yeast in Experi- ment 2, because i t had four times the amount of sugar on which t o exhibit its power, rather than that there was any real difference between the two yeasts in their ability to ferment sugar.I n the. results of either experiment there is nothing tc, show that the yeasts concerned have exerted more than an unknown fraction of t h eir f ermen t at i v e power. It will he noticed in Experiment 5 that the fermentation was more prolonged than that of the other experiments, and when it ceased there was ferrnent,able sugar still present in the solution. There- fore in this experiment the yeast formed was under circumstances in which it was able to decompose all the sugar of which it was capable, and the very low ratio of 1 of yeast to 183 of sugar fermented was found.This ratio is lower than the lowest determined by Pasteur when working under the most perfect anazrobic conditions, j e t Experiment 5 was conducted in a flask to which air had free access. So in Experiment 5 , in which the >-east present has had, so far as its fermentative powers are concerned, an unlimited amount of sugar to ferment, the yeast has exhibited even greater fermznt‘itive power than Pasteur’s anaikobic yeast iinder similar conditions. It seems hardly possible under these circumstances to come to any other conclusion than that it is the amount of sugar a t the disposal of the Feast which is the main factor in influencing the ratios oE yeast to fermented sugar in experiments like Pasteur’s and mine, and not relatite conditions of agration and non-aiktion.Doubtless, during 2u fermentation experinlent, there are many other factors, besides the amount of sugar present, which come into play and must affect the ratio of jeast to s u g a r fermented, but the main factor under ordinary condiiions is the amount of sugar. In order t o show more clearly that the ratio of yeast to fermented sugar determined i n a manner siniilar to Pasteur’s Experiment B, in which all the sugar present was fermented, does not express the true fermentative power of the yesst, another experinlent ivas made of a918 BROWN: THE SPECIFIC CHARACTER OF THE dXerent nature from that just described. Three flasks, Nos. 1, 2, and 3, were prepared, each containing 300 C.C.of a 5 per cent. solu- tion of cane sngar in yeast water ; they were closed with cotton-wool plugs and sterilised. The flasks were similar in every respect, excepting that No. 1 was furnished with two glass Cubes penetrating the cotton- wool plug, and arranged so that any quantity of the liquid contents of the flask could be withdrawn through one of them by meaus of a syphon, and fresh liquid introduced through the other. The contents of the three flasks were inoculated with a tracc of yeast on February 13, and kept at a constant temperature of 26". On February 17, fermen- tation in all the flasks had ceased ; the yeast had also subsided, leaving the fermented solutions quite clear. 10 grams of sugar had been fermented in each experiment, and its the conditions in all three were precisely similar during fermentation, t'he same weight and number of yeast cells must have been fbrmed in each.Experiment 2 was now terminated in order to ascertain the ratio of yeast to sugar fermented in all. It represents, therefore, ac- cording to Pasteur, the total fermentative power of the yeast. On the same day that No. 2 experiment was terminated, 120 C.C. of the clear and completely fermented liquid in No. 1 was withdrawn by means of one of the tubes with which the flask was furnished, without disturbing the deposit of yeast at the bottom of the flask. Microscopic examination of the liquid withdrawn showed that no yeast had passed over with it. 50 C.C. of a solutiori containing 10 grams of cane sugar were now passed into the flask t o partially replace the fermented liquid withdrawn.The fresh sngar was added in a strong solution in this manner, in order to keep the total volume of the liquid contents of the flask at a distinctly lower point than the original volume of the solution at the commencement of fermentation. This was done in order to check any tendency of the yeast cells to increase by budding, when treated with the fresh fermentable solu- fion. The reason for it turns on a point to which I have called attention before (Trans., 1892, 61, 369), namely, that yeast cells have a limit to their power of multipiication when in a given rolume of a, fermentable solution, and if this limit is intentionally exceeded by over-crowding, a vigorous fermentation of sugar can be carried on without the ye& cells increasing in number.Shortly after the introduction of the fresh sugar into No. 1, rapid fermentation took place. On February 20, this fermentation was over, and the liquid quite clear. A further portion of the solution was now withdrawn from the flask in the manner de- scribed before. Examination of the solution removed showed that it contained no unfermented sugar, and no yeast cells had been This ratio found was 1 : 20.FERMENTATIVE FUNCTIONS OF YEAST CELLS. 919 Termination of expt. carried over in it." A solution containing 10 grams of cane sugar was again introduced in place of the fermented solution removed. This sugar was completely fermented by February 26. The flask WEB then opened, and the yeast cells counted and weighed in the usual manner. No.3 flask, which had remained at rest during the whole time that the successive amounts of sugar were beicg fermented in No. 1, was also opened, and the yeast cells counted and weighed. This experiment. was carried on to act as a check on the others with regard to the number of yeast cells formed, and also to a, ncertain to what extent yeast cells lose weight when resting iuert in a completely fermented medium. The results of the three experiments are tabulated below. I Ratio of sugar of S o . of cells Total weight Total sugar expt. ~ I I I I-- ---- -- 1- I February 26.. . 9 '38 0 -2990 30 1 : 100.4 ,, 17. .. 2 0.3435 1 10 1 : 29.1 ,, 26...1 1 "9g 1 : 41.0 The numbers given show that, within experimental error, the number of yeast cells found in each experiment was the same, although, in No.1, three times more sugar has been fermented by the yeast than in the other experiments. The weights of yeast cells found differ a little. On February 17, when No. 2 experiment was terminated, there is no doubt the weights of yeast in all the experi- ments were the same, for the conditions were in all cases precisely the same up to this point, and we know also that the numbers of the cells were the same; but during the next nine days the weights of yeast in Nos. 1 and 3 must hare diminished. This me should quite expect with the yeast in No. 3, for it had rested inert in an exhausted solution during that time, and it is well known yeast loses weight under these conditions. But the yeast in No.1 was differently placed, and during most of this time had been actively fermenting sugar. The loss of weight is cornparatirely small, but the fact that the yeast lost slightly in weight instead of gaining during the fermenta- tion of a considerable amount of sugar, indicates that the exercise of the fermentation functions of yeast cells is not connected with the addition of any permanent matter t o the structure of the cells. * It is necessary in experiments like this that the solutions used should bc sterilised and allowed to come in contact with filtered air only, €or otherwise foreign organisms are apt to s ~ o i l the results.$120 BROWN: THE SPECIFIC CHARACTER OF THE The point in view in describing the above experiments was, howA erer, t o demonstrate that during fermentation when yeast cells hare only a limited amount of sugar to ferment, the weight of yeast formed, compared with the amount of sugar fermented, does not give a true expression of the total fermentative power of the yeast.In Experiments 2 and 3, thc ratios found for yeast and fermented sugar were 1 : 29 and 1 : 41 respectively (the latter being the higher only because the weight of Feast formed had diminished on standing), yet the same amount of yeast formed under similar conditions in Experi- ment 1 has given a ratio of 1 of yeast to 100 of fermented sugar. when fed with more sugar than was present in Experinients 2 and 3. This difference in the ratios turns, therefore, on the amount of sugar a t the disposal of the yeast in Experiment 1, and not on any real differ- ence between the capabilities of this yeast to exert fermentative power, and those of the yeasts in Experiments 2 and 3.The results of the three experiments just described agree with those previously given, in showing that the ratios of yeast to sugar fermented, when the yeasts have only a limited amount of sugar on which to exert their power, do not represent the true fermentative powers of the yeasts or any known fractions of them, and are, therefore, not comparable. Reference will now be made to those experiments by which Pasteur determined such high ratios of yeast to sugar decomposed, that he considered they approached those given by ordinary 11011- fermentative fungoid growths in the presence of air. A brief reference has already been made to them.These experiments (Zoc. pit.., p. 243) were conducted in flat, very shallow dishes, in order that the fermenting solutions, and the yeast they contained, might be exposed to the maximum &ration attainable. The strength of the cane sugar solutions used was reduced to the very low point of 0.86 gram in 100 C.C. Two experiments are described by Pasteur, in each of which 200 C . C . of the solution were piaced in one of the flat dishes, and inoculated with a trace of j-east. As the method in botli experiments was the same, the laht experiment only will be described. I n this, about 24 hours after inoculation, as soon a s the deposit of yeast folmed appeared to be weighable, the experiment was stopped by the addition of a little sulphuric acid to prevent any further yeast increase or fermentation.The yeast was collected, dried, and weighed, in the usual manner, and the sugar left unfer- mented in the filtrate was de'ermined. I t was found that 0.024 gram of yeast had been formed, and out of the original 1-72 grams of s u p . - present only 0.033 gram had been fermented, showing a ratio of 1 of yeast to 4 of fermented sugar. This veyy high ratio has, according to Pasteur, been brought about by excessire exposure of the yeast toFERMENTATIVE FUP;CTIONS OF YEAST CELLS. 921 the influeme of the air, which bas rednced its fermentative power to a point closely resembling that of ordinary fungi ; in fact, that the yeast had pi,actically ceased to be a. ferment a t all. Pastenr also considers that the ratio of yegst’ t o fermented sugar determined i n this way, is comparable as au expression of fermeutative power with the ratios determined in Iiis other expeiiments. But the experiments just described are open to great objection if the results are used i n this manner.Attention has already been called to the fact that in most of Pssteur’s experiluerts the true fermentative powers of the yeast were not determined, because the amounts of sngar present were limited ; but i n these last experiments, although excess of fermentable sugar was present when they were terminated, yet a free exercise of the fermentative powers of the yeasts was limited by t i m e . Pasteur him- self most particularly insists that time has nothing to do with an esti- mate of fernient,ative power-that it’ is independent of time altogether.I n his criticism of Schutzenberger’s views regarding fermentation (Zoc. c;t., p. 245), this point is brought forward very strongly. Pasteur considers it is onl. the ‘. energy,” or “ activity,” of a ferment that can be estimated in an interval of time. Yet, in the experiment under consideration, time has been introduced as a factor in the estimate of the trork of fermentation. The experiment was terminated so soon as enough )-east> was foimed as to bc weighable, so that the sugar fermented by this yeast was only the amount it had tinae to ferment under these circumstances. 1 he yeast would have fermented more sugar if time h i d been allowed for it to do so, but its action was stopped with sulphuric acid. It is well known that yeast cells, once formed, carry on their fermentative functions for a long time if placed under favourable circumstances ; if, therefore, tbe time is limited during which they are decomposing sugar, the ratio of yeast to siigar in this rcstricted ihterval of time must, according to Pnstenr, represent the “ activity ” of the yeast cells in this time, not the fer- meiitative power.Yet Pasteur uses the ratio determined in this way as an expression of fernientative power also. There is a further objection to Pasteur’s experiment, of equal importance. I n making tlhis experiment, he used cane sugar as the fermentable medium. Cane sugar, as such, is not fermentable by yeast ; i t must be inverted before it can be fermented. Inversion of the cane sugar is brought about by a function of the yeast cell, aiid i t has recent1,y been shown by James O’Sullivan (Trans., 1892, 61, 593), that this function of the yeast cell is distinct from its functions of fermentation.In Pastenr’s experiment, therefore, as the s u g a r fermented by the yeast cells in a limited time had first to be inverted by the cells, the ratio of yeast to fermented sugar expresses the “ activity,” or pomer in an interval of YOL. LHV. 3r.932 FERMENTATXVE FUNCTIOW OF YEAST CELLS. Cane sugar experiments. ~ e a u t S u p r fer- f o l d , in mented, grams. in grams. pu~IBr fer- ------- 0.OgQfi 0.531 1: 5 . 4 0.1785 4 026 1 : 2 2 . 6 0.1900 6.812 I: 35.8 time, of the combined action8 of the invertfive and fermentative func- tions of the yeast cells.I n order to show experimentally that the time occupied by yeast in inverting cme sugar has a very distinct influence on its apparent " nctivitv " of fermentation, as determined by the ratio of yeasf tn suqar fermented i n an interval of time, six flasks were prepared, three of each contsininp 100 c c. of a 10 per cent solut,ion of cane sugar in peast water. and the other three each containinp an equal quantity of a 10 per cent solution of dextrose in yeast water. All six solutions were inocnlnted with a trace of yeast and allowed to ferment, iinder esactlv the same conditions. In 42 hours, a distinctly weiphable amoiinc of yeast was noticeable in the flasks. At, this staqe. one of the flaskc! conthining cane snpar and one containin? dextrose were opened, and the amount of yeast formed and of mgar fermented in each were determined.24 hours after the first flasks were opened, two others, one containing dextrose and one cane finpar, were opened, and determinations of t h a yeast and sugar fer- mented made. Tbe last two flasks were opened and examined 24 honrs later still. The table below gives the results of these experi- ments. Dextrose experiments. Yeast Sugar fer- of found, in mented, to sugar fer- mented. grams. in grams. - ---- 0'1040 2.334 1 :22'4 @*16Q2 4.364 1 : 2 5 ' 8 0*2090 6 9 2 0 1 :33'1 Time frow in- oculation. -- 42 houra 66 9 9 93 7 9 I On comparing the ratios fonnd for the cane sugar and dextrose fermentations in the first interval of time, i t will be noticed that the ratio when peast acted on A directly fermentable s'igar like dextrose, is very different, from the ratio when Feast acted on cane snpar which mnst be inverted prior to fermentatinn.111 the first 42 honrs, some of the time must have been taken up in the cane sugar becoming inverted so a$ to be in a fit condition for fermentation ; under these circumstniices i t was found that, in the same interval of time, 1 part of yeast was able to ferment only 5 parts of cane s u g a r ; whilst in the case of dextrose, a suoar t h a t is directly fermentable, 1 part of yeast was able to ferment 22 parts of this sugar. I n the second a9.d third intmrals of time during. which the experiments were conducted, the " activity " of the yeast (estimated by the ratio of yeast to sugarCOLOURING PRINCIPLES OF VENTILAGO NADRASPATAKA. 923 fermented) in the cane sngar experiments rapidly developed, until it even slightly exceeded the ‘‘ activity ” of the yeast in the dextrose experiments. It was found by direct experiment that a considerable amount of invert mgar had accumulated in the cane sugar solutions during these intervals, hence the yeast present in them had plenty of fermentable sugar a t its disposal, and its “activity” as a ferment it icrea sed. The above experiments show sufficiently well how it was tohat Pasteur observed such a low fermentative “ activity ’’ in his experi- menta, without taking into consideration that he employed very dilute solutions of sugar, which also tends to increase somewhat the ratio of yeast to fermented sugar. There is no doubt that the proof of the truth of Pasteur’s theory turns entirely on the correctness of the method he adopted for determining and cnmparing the fermentative powers of yeast cells under vaq-ing conditions of aiiration. If it has been shown i n this paper that the fermentative powers of the yeasts have not been cor- rectly determined, and that they are not comparable one with another, then M. Pasteur’s theory remains without esperimeiital proof. On the other hand, there is experimental evidence directly opposed to this theory. I refer to my experiments on alcoholic fermentatim (Trans., lr392, 61, SSS), by which it is shown that’ aeration does not prevent, or hinder, the exhibition of the fermentative functions of non-increasing yeast cells. If, in the place of M. Pastenr’s theory that fermentation is “ life without air,” is substituted the consideration that yeast cells can use oxvgen in the manner of ordinary aerobic fungi, and probably re- quire it for the full completion of their life history, but that the exhibition of their fermentative functions is independent of their en- vironment with regard to free oxvgen, it will be found that there is nothing in the results of any of 31. Pasteur’s experiments contradic- tory to such an hypothesis.

ISSN:0368-1645    

DOI:10.1039/CT8946500911

出版商:RSC    

年代:1894

数据来源: RSC

摘要:

COLOURING PRINCIPLES OF VENTILAGO NADRASPATAKA. 923 LXXV.- The Colouring Priizciides of Ventilago 31aclras- pataiza. By A. G. PEHKIN and J. J. HTMMEL. Verztilago madrasptuna is a large, climbing shrub, belonging to the order Rhamnacee, the root-bark of which furnishes a dye-stuif much valued in Southern India. It is very common in the Western Penin- sula from the Konkan southwards, as well as in Ceylon and in Burma, and, according to Liotard, it is largely collected in Mysore at certain 3 u f l924 PERKW AND HUMMEL: THE COLOURING PRINCIPLES periods of the year and exported to other districts of India. The annual production is given as 1-3 tans, its price being rariously stated as 1+-7&L per lb. The following 81-3 a few of its Indian vernacular names : pitti (Hindi) : raktapita (Bengali) ; pappili-chakkn, suralpattai (Tamil) ; poplichukai (Run.) ; lokandi, kanwail (Bomb.).Oiir attention was first attracted to i t under the name of souronl- pnttay, in Gonfreville's " L'art de la Teinture des Lakes," 1F49, p. 539, where it is mentioned as a dye-stuff jielding fast colours, nxnielg violet, dull red, and black. The identity of this product with TTwztilago nzadraspatamz was established by correspondence with the authorities of the Imperial Institute, to whom we are also indebted for a supply of 10 Ibs. of the root-bark with which the experimental work here recorded has been carried out. The root-bark appears as dark, purplish-brown scales, ribbons, or filaments, the dust from which when it is ground to powder, irritates the throat in a marked manner.When treated with boiling water or alcohol, it gives a red solution, which, on the addition of caustic alkali, changes to a deep crimson. Many writers on tbe economic products of India, as Bidie, Lio- tard, Wardle, and others, refer to this dye-stnff, but we find no record that any previous ehernical rxamination of it has been made ; at most, the actions of various chemical agento, metallic salt solutions, &c., 011 its aqueous decoction, ace referred to, e.g., by Gonfreville (Zoc. cit.), and also by Dgmock, Waden, and Hooper in the Pharmacogmphia Indica, i, 355, the last named authors hazarding tbe suggestion that the colouring matter in probably one of the deriratires of anthracen e. The present examination is of interest as showing that Fentilago P?zadraspntana contains a colouring matter differing from all others a t prese n t known.Summary of Re.su Et s. On treating this dye-stuff with car-bon bisnlphide, five crFstalline substances are extracted, together with a wax and a resinous colour- ing matter. The separation of these from one another by fractional crystallisation and other means proved to be a niatter of considerable difficulty, the details of which will be found i n the experimental portion of this paper. 1. A Substance qf the Formula Cl6Hl2O5.-This crjstallises in long, orange-red needles melting a t 200"; it sublimes a t higher tempera- tures, partially carbonising. Its alkaline solutions bave a purple tint, and the corresponding salts can be obtained in the form of' F-iolet- colonred needles sparingly soluble in alcohol.It yields a diacetglOF VEXTILAGO MADRASPATANA. 92 5 compound, Cl,Hl,05(C,H,0),, in long, yellow needles melting at 185-186", and a mononitro-compound, C16H1105*N03, melting at 215--%17". When distilled with zinc dust, a hgdrocarbon is ob- tained me1 ting at 203", evidently a-methylanthracene, and its be- liaviour with zinc dust in alkaline solution shows i t to be an x-rnethyl- aiithraquinone drrivative. Treatment with hydriodic acid showed that it contained one methoxy-group, a pale ello ow, crystalline reduction product, probably an ant,hranol derivative, being formed a t the same time ; the quantity thus obtained was insufficient for analysis. The substance C,,H,,O, is therefore a wzommeth!/Z ether of a t r i h y d r o ~ y - z - ?rzelhyZarzthrapIci,io~Le. I t does not dFe mordanted fibre ; when heated with sulphuric acid to 160" i t jields a substacce of the formula CljHl,05, crystallising in orange-red needles melting a t 254" Although only a small qiianlity of this compound has as yet been obhined, owing to lack of material.it is evidently a trihydr-0x9-x- ?nethylanthra~zii)io?~e, the reactions of :Irhich, so f a r as they could be studied, beit\g almost identical with those of emodin. The substance C16H,,05 yields a nitro-compound, C,,H,O,(NO,),, which in appearance and properties closely resembles tetranitro- chrysophanic acid, CljH601(X02)1 ; when heated with sulphuric acid, i t yields a new product soluble in alkalis with a violet coloration, 5t jwoperty possessed by numerous nitroantlirsqninone derivatives.If the substance Cl,H,,O, be an emadin methyl ether, it would be ex- pected that emodin itself, when treated in a similar way, should yield a tetranitro-compound. Experiment proved this to be the case, the product obtained, evidently a t e t l anitroemodin, closely resembling t h e above compound C1GHB05(X02),, which shouId be its methyl ether. This. together with the general similarity iu appearance, properties, mid melting point between emodin from Rhamuzis fr-ci?igtiZa, and the substance C15Hlo05, both also derivatives of a-methylanthracene, seems to point almost conclusively to their identity. Until, however, a comparison can be made of the properties and melting points of their acetyl derivatives, this point cannot be absolutely decided, although in this paper we shall regard the substance C,H,,05 as errtodiit, inoiiometli y l ether." 2.A Substance cf the formula Cl6H,,O4 (A).-This forms long, coloudess needles, which decompose at about 260" before melting; it is soluble in alkaline solutions with a yeliowish-brown coloration. It yields a-rnetliylanthracene, and what is probably 5t triacetyl com- pound, C,,Hl104( C,H,O),. On gentle oxidation with chromic acid, it is converted into the previously described emodin methyl ether, and t h e same reaction takes place when its alkaline solution is oxidised * It is probable that several months will elapse before m-e receiPe a fresh supply of the l ' e n t i l a p madraspatana, so as to enable us t o confb: our opinion.926 PERKIX AND HUNMEL: THE COLOURING PRINCIPLES with hydrogen peroxide.Boiling uitric acid (sp. gr. 1.5) converts it into the tetranitro-compound, C16H,0,(NOJd. The substance C16H,,00 (A) appears, therefore, to be a trihydroxy-cL-methylanthrancrle monomethyl ether. As an alternative constitution for this substance, it might be considered as a hydranthrone derivative, thus : Anthranole. Hydranthrone. But tbe analysis of its acetyl derivative shows i t to be a trihydroxy- rather than a dihydroxy-derivative. 3. A ,Substance of the FurnauZa C,,H1,O, (B).-This crystaliises in pale yellow needles melting a t 1%". With acetic anhydride, it yields what is probably a triacetyl compound melting a t 227--2;29", the alcoholic solution of which has a smong blue fluorescence.It dissolves i n alkalis forming yellowish-brown solutions which on long exposure t o air become red, and on treatment with acid yield a precipi- tate of emodin methyl ether. Oxidation with chromic acid gives a similar result, whilst with nitric acid the tetranitro-derivative, C16H805(NOJ4, is formed, identical with that obtained from the ernodin methyl ether by similar means. The substance C16H1404 should there- fore represent the secona trihydroxy-a .)tLetiiyl~nthranole mouometk y l ether capable of yielding the emodin methyl ether by oxidation. The only compound of the anthrncene group previously isolated from natural sources, and which can be considered as nearly related to the substances C,,R,,O, (A and B), is the chrysarobin obtained from Goa powder, and examined by Liebcrmann and Seidler (Annulen, 212, 291).This, when oxidised in alkaline solution, yields chrpophanic acid, dihydroxymethylanthraquinone, the constitution assigned to it > CH*OH],. C,H,(OH) -- being o [ C H < ~ a ~ , ( CH,)(OH) If the trihydroxy-a-methylanthraquinone methyl ether here described is an emodili methyl ether, these results are of considerable interest, for, as is well known, emcdin is considered t o be a hydroxychryso- phanic acid. It, is remarkable that the reduction products of these t w o similarly constituted compounds should be the only ones of the snkhraquinone group known to exist in a natural condition. Thrtt the substaiices C,,H,,O, (A and B), however, have not the complicatedOF VEXTILAGO MADRASPATANA. 927 constitution assigned to chrysarobin is evident, for if formula ex- plaining their reactions be constructed on similar lines, they should contain S per cent.less carbon than is shown by the analytical results. The effect of the presence and absence of the quinone group on the colour of these substances is here well illustrated, for whilst C,6H,,0, (A) is colourless and C,H,aOa (B) nearly so, the emodin methyl ether has a marked orange-red tint. 4. d substance of the Formula C16H,0,.-This is an orange-red, crystalline powder, which, when heated, begins to darken at 260", slid melts aiid carbonises a t L?i&--28Uo. It is distinguished from the preceding substances by its sparing solubility in most solvents. Solu- tions of the alkalis dissolve it wich an orange-red coloration, and it yields an acetyl derivative, CJ&G~(CJLO), which crystallises in yellow needles melting and &composing at 216-2dO".When heated with zinc dust, i t yields a hydrocarbon, wliick is probably a-rrJethyl- anthracene. Treated with hydriodic acid, it loses one methoxy-group, and at the same time a pale yellow, crystalline reduction product is formed, probably a hydranthrone derivative. Sulphuric acid a t 1 ~ " converts it into what is probably a new colouring matter, evidectlp by elimination of methyl from the methoxy-group, and from this result it also appears that the hydroxy- and methoxy- groups are in the ortho position relatively to one another. The action of alkaline reducing ageuts is very interesting, for they change the orange-red solutions to yellow, but on exposure to air they become, first orange-red, and finally blue- violet, when t h e j contain the alkali salt of a new colouring matter ; this, although still most probably a methoxy-derivative, must, from its reactions, be distinct from the colouring matter produced by sulphuric acid at 160". From its deep orange-red colour and other reactions, it appears to be an a-methyl- anthraquinone derivative, aud if i t has the formula C,H,O,, and not a multiple of the same, i t may be represehted ay C H ~ ~ C , , H O , ( C , ) " ( O H > ~ ~ ~ ~ ~ , .The production also of a colouring nlatter from it by means of reducing agents, and subsequent oxida- tion, points t o the presence of more than one quinone group in it. It might therefore belong to a class of compounds suggested by the following formula, .O P C.CH3 OC/\C-CO-CH\CH I ' 1 1 OC'QC- C O - C V C - O C H , \ c C-OH ' 0 As no more than 1 gram of pure substance has been obtaiaed from938 PERKIN -4ND HOMMEL: THE COLOURXNG PRINCTPLES 10 lbs.of the root-bark, it will be extremedy difficult to thoroughly investigate the nature of this most interesting substance. 5 . A suEstame of the Iformula C1,H,,05.-This is a, chocolate- coloured, crystalline powder. When treated with dilute alkali, it dissolves with a yellow coloration, but on exposure to air the solution deposits a blue, amorphous precipitate, and it therefore appears to contain in its molecule a reduced quinone group. The study of this substance will be continued when snfficient material is to hand.6. Tlie Wuz, (C,H,,O),,, consists of nearly co1ourless, minute needles melting a t i 2 " . 7. The Coloziring Matter is a, reddish-brown, brittle resin of the formula Cl5HI4O6, and, up to the present, has resisted all attempts to obtain i t in a crystalline condition. It softens a t about loo", and melts a t 100--110". Dilute alkalis dissolve it with a purple-violet coloration, and the corresponding salts are obtained as violet, amor- phous precipitates on adding common salt to these solutions. When distilled with zinc dust, it yields r-methylant hracene, and when treated with zinc dust ili alkaline solution, behaves as a derivative of methyl- anthraquinone. Nitric acid converts i t into oxalic acid. On mor- danted calico, it dyes shades which remind one somewhat of those given by brazilein, although the reds are much more purple in hue, and faster to boiling soap solution.From its nature and properties, it appears possible that i t is allied t o alkannin, CISHl4O1, the colouring matter of the roots of the Anchusa tinctoria, inves tignted, among others, by Liebermann and Komer (Ber., 20, 24%). Alkannin is also of a resinous nature, is a derivative of a-methylanthracene, and appears to contain two hydroxyl groups. I t is possible, therefore, that the colonring matter of the Ventilago rnadraspatuwa, for which we propose the name Veiltilagin, is represented by alkannin con- taining two additional hydroxjl groups. The investigation of this colonring matter will be continued when a further supply of raw material has been obtained.EXPERIMLNTAL PART. 250 grams of the coarsely-powdered root-bark was treated in the cold with 2 litree of carbon bisulphide for 48 hours, with occasional shaking, and, after removal of the extract, the residual root was twice treated in a similar manner, and the carbon bisulphide was distilled [ ~ f f from the deep brown liquid until there remained 200 C.C. of extract, which was then mixed with 50 C.C. of alcohol. The crystals which separated on cooling (Cryst. prod. A) were collected, and then washed with about 190 C.C. of alcohol ; the washings, on being added t o the mother liquor, caused a further crystalliue precipitate aft'er someOF VENTILAOO MADRASPATANA. 929 hours (Cryst. prod. B). It was evident that some separation of the coir- stitnents of the product had been effected in this way, for whereas the crystals first deposited were reddish-brown, the second crop was pale yellow.The filtrate, after standing for some days, frequently de- posited a third fraction, which was usually added to the second. A quantitative estimation in this way gave 2.834 grams of crystalline product from 250 grams of root-bark or 1.13 per ceut., and, as subse- quent inrestigation showed, This represents approximately the amount present in the root. As stated in the introductiori, this crude product consisted principally of four substances, the separation of which was most readilr effected by the methods described in the following pages. Subsfance of the Fornaula CI6H8O8. The finely-powdered, " cr-stnlline product, A," was extracted two or three times 1%-ith small quantities of boiling benzene or carboil bisulphide, which left a small amount of a sparingly soluble residue ; this was dissolved in a large quantity of boiling benzene, treated with animal charcoal, 2nd filtered.On cooling, a finely divided, crystal- line rrecipitate separated, and was further purified bj- a second treat- ment in a similar manner. C = 58.45 ; H = 3.08. 11. 0.1138 ,, 0.2443 CO, and 0.0310 H,O. C = 58-47 ; H = 3-02. 111. 0.0825 ,, 0.17i5 CO, and 0.0235 H20. C = 58-66 ; H = 3.16. C,,H,O, requires C = 58.5 ; H = 2.4 per cent. The amount of this substance present in the dye-stuff is exceed- ingly sniall, not more than 1 gram of i t having been hitherto obtained in the pure condition. It; forms an orange-red, glistening powder, consisting of microscopic prisms, and when heated i t begins t o darken at 260", and melts and carbonises at 875-280".If heated rapidly between match glasses, it yields a small quantity of a sub- limate, which melts in a similar manner. It is almost insoluble in boiling alcohol, and but sparingly soluble in benzene, carbon bisnlph- ide, and curnene, crjstaliising from the latter in orange-yellow needles. It was a t first thought that dissolving it a t the Ligh boiling point of this liquid had effected decomposition, but this was negatived by the results of analjsis 111. Dilute alkalis, if cold, dissolve it very sparingly, more readily, however, i f boiling, with formation of crimson-red solutions, from which the corresponding salts are thrown out as brownish-violet precipitates on adding excess of alkali.In baryta water it is insoluble. On boiling the alkaline solutions with zinc dust, pale orange-coloured liquids are obtained, which, on exposure to air, first become red, and then blue-violet ; if acid is now added, claret-coloured flocks of a new colonring matter are precipitated I. 0.1062 gave 0.227'7 CO, and 0.0295 H@.930 PEREIN AND HUMMEL: THE COLOURING PRINCIPLES On fusing the substance Cl6H8O8 with potash, a blue-violet melt is obtained, from which, on neutralisation, a similar, though probably not identical, colouring matter is obtained. Disti!lation with zinc dust yielded a hydrocarbon crystnlliaing in greenish leaflets ; the quantit#y mas too small for purification, but on oxidation it yielded, first, a quinone, crystatlising f r o u dilute acetic acid in colourless needles, aud, finally, an acid soluble in dilute ammonia. This substance is believed to be a-rLletliylanthrncene, not only on account of its be- haviour, but also because this hydrocarbon appears to be the basis of the other principal substances described in t h i s paper.It is b u t little attacked by nitric acid (sp. gr. 1.42) in the cold, but on heating i t gives a beautiful, violet-red solution, which gradually passes to brown, and then to yellow. By t h i s reaction, the conipound, ClGH809, can be readily dist'iuguished irom all other substances found in the root- bark. ActiolL of Hydricidic Acid.-O.l@l gram, heated with hydriodic acid iu Zeisel's apparatus, gave 0.081 gram of AgI = 4.57 per cent.methyl groups. C15H,0i*OCH, requires 5.15 per cent. methyl groups. The substance C;,,H,O, theretore conta.ins one methoxy-group. The residual hydriodic acid contained a nearly co!ourless, crystal- line product in suspension, which was collected, washed, dried, and recrystallised from toluene. It dissolved in a1 kalis, forming brownish- ~ e l l o w solutions which gradually oxidised on exposure to air, be- coming of a dull, reddish-violet t i n t ; it appeared therefore to be a b ydran throne or d i e d corn pou nd. Cold suipii~u~-ic acid dissolves the substance C&&,08, forming a purple liquid resembling aqueous sodium anthrapurpurate. On heating to 160" this solution becomes deep blue-violet, and on adding water a new substance, evidently a colouring matter, is precipitated ; this forms, with alkalis, an aimost iusoluble, purple-red compound.The above reaction is no doubt caused b y the decomposition of the methoxy-group, and this result is interest,ing, showiug, as it does, t h a t the rnethoxy-group is in the ortho-position relatively to a hydroxyl group. This colouring matter ( b ) must not be coulhnded lvith (u) that produced by the reduction and oxidation of the sub- stance c16H:sOs, as it difiers from it in its sparing solubility in alkaline solutions. It is most probable that the colouring matter ( c ) , pro- duced by the action of fused alkali, is distinct, and forms a third, which should consist of (u) from which the methoxy-group has been removed. Action of Acetic Auhydm.de.-When digested witlh acetic anhy- dride, the subdance, C,,H,O,, dissolved only after prolonged boiling, alld the solution thus formed was digested for four huuis longer.OF TENTILAGO MADRASPATANA.931 On cooling, a yellow, crystalline powder separated, which was washed with methylic alcohol and recrystallised from acetic acid. O*ll5i3 gave 0.2447 CO, and 0.025 H,O. C = 57.8 ; H = 3.31. c,6Hjo8(c,H30) requires c = 58.37 ; H = '2.70 per cent. C,6H6O,(C,H,jO), ,, c = 58.25; H = 2-91 ,, C,6H,O,(C,H,O), ,, C = 58-15; H = 3-08 ,, The analytical numbers, therefore, are insufficient to decide whether it is a mono-, di-, or triacetyl compound, but from a con- sideration of the results obtained above it is evident that the snb- stance C,,H,O, is most probably CH,*ClaH,0j*OCH3, and accordingly it can, st most, form no higher derivative than a diacetyl compoond. On the other hand, if this substance is represented by the formula c,6H& and not a multiple of the same, i t is highly possible that a hydrogen in one of the two side rings is not displaced, in which case a monacetyl derivative alone can exist.That this substance is a monacetyl derivative appears to us most probable, in which case the formula of the substance C16H808 might be written CH,-C,,HO,(O,jff (OH)*OCH,. The acetyl compound forms lemon-yellow needles melting and decomposing a t 216-220", almost insoluble in boiling alcohol, but more soluble in acetic acid. Cold solutions of tbe alkali hydroxides do not dissolve it, but, on boiling, orange-red solutions are gradually formed.The benzene extracts and mother liquors from the preparation of the previously described substance (p. 929) were boiled with animal charcoal and concentrated ; on cooling, a crystalline product was deposited, which, on examination under the microscope, was seen to consist of thin leaflets interspersed with a few needles. On examina- tion, however, it was found that the mixture was by no means SO simple as appeared a t first sight, crystallisation from carbon bi- sulphide giTing four fractions having the respective melting points 150*5", 167-172", 163", and 172" ; similar results were obtained with benzene, although under the microscope each fraction usually ap- peared homogeneous. Further crystallisations from the various solvents gradually raised the melting points of these fractions, but tlie labour and loss of product involved rendered it necessary to adopt some better means of separation.It was clear that a t least two substances were present, as the portions having the higher melting point dissolved in cold sulphuric acid with a reddish-violet colora- tion, whilst those of the lower melting point gave a desp brown. Ultimately, by the use of t w o or three solvents, and after several months' labour, three principal products were isolated, two melting respectirely at 200" and 173", and a third decomposing at about 2 GU O , be fore me I t iiig .932 PERKIN AND HUMMEL: THE COLOURLVG PRIKCIPLES Substance of the Formula ClsH1205, nz. p. 200". The mixed product obtained from the benzene extracts and mother liquors (p.9:31) was crystallised from a large quantity of carbon bi- sulphide, and the product which melted at l&3" was recrystallised from benzene, and t h e product twice crystallised from boiliug acetone, and finally from isobutylic alcohol. There could be no doubt that this final product consisting of slender needles was a homo- geneous substance, and this was further corroborated by the analjsis of three distinct preparations. 0-119s gave 0 2958 CO, and 0.0454 H,O. C = 67-34 ; H = 4.21. 0.1139 ,, 0.2811 ,, ,, 0.U43.5 ,, C = 67-29 ; H = 4-24. 0.1180 ,, 0.29%2 ,, ,, 0.0480 ,, C = 67.53 ; H = 4.52. C,,H,,O, reqnires C = 67.6 ; H = 4.22 per cent. It crystallises in long, glistenin?, orange-yellow, hair like needles melting a t 200°, and at higher temperatures partially carbonising and yielding a sublimate of leaflets.It is sparingly soliible in alcohol, acetone, and isobutylic alcohol, more reacdily in benzene and carbon bisulphide. Solutions of the alkali hydroxides dissolve it, yielding red liquids, 2nd the corresponding salts map be obtained in Tiolet, micro- scopic needles by treating hot alcoholic solutions of the substance with alkdi, and cooling. It is not aifected by boiling barium hydr- oxide solution. When distilled with zinc dust, a small quantity of greenish leaflets were obtained, which were proved to be a-methj-lan tlimwne. Treated with zinc dust in alkaline solution, the original b l o c d - r d colour changes t o orange, but on exposure to air t h e original t i n t relums, and it i s probable, therefore, that the substance is a deriva- tive of a-methplanthraquinone.It dissolves in sulphuric acid with a violet-red coloration. this reaction distingui3hing it from the sub- stances C,,H,,O, A and B described later. Action of Acetic A)ihydrirle.-The sabstance C,,H,,O,, when boiled with a small quantity of acetic anhydride for six hours and t h e solution cooled, gave a mass of needles, which were washed btith alcohol and recrjstallised from t h e same so'ivent. On analysis, the folloa~iug numbers were obtained. 0.1078 gave 0.237 CO, apd 0.0435 H,O. C16t11,,05(C2~30)~ requires C = 65-21 ; H = 4 34 per cent. I t was, therefore, a diacetyl derivative. It crystallises in long, Fellow needles melting a t 185-186", sparingly soluble in alcohol, readily in acetic acid.In cold alkaline solutions it is insoluble, but, on boiling, it is decomposed with formation of red sulutions. C = 65.01 ; H = 4.48.OF FENTILAGO MADRASPATANA. 933 Action of Hydriodic acid.-From the formula of this substance, and t h e fact that it yielded o n l ~ a diacetyl derivative, i t appeared possible t,hat i t might contain a methoxy-group. It was, therefore, submitted t o the action of hydriodic acid quantitatively by Zeisel's method. 0.184s gave 0.1676 AgI. CH,O = 5.78 per cent. C,,H,O,*OCH, contains 5.28 per cent. The substance C16H,20, is, therefore, a monomethylic ether of a tri- h s d r o s y - 3: - m e t h -j-1 an*r a q uino 11 e . As the residual hydriodic acid contained some crystalline matter in suspension, itJ was diluted with water, and the solid collected, washed, dried, and pnrified by cr-j-stallisation from toluene, i n which it was but sparingly soluble.I t appeared as a pale jellow, crystalline powder, soluble in caustic soda with a brown coloration, which on exposure t o air changed to purple. The quantity of t h i s substance obtained was insufficient for analysis, but, j n d g i n g by its pyoperties, it is most prohably a trihSdrosy-sl-niethvlanthranole or hydranthrone. A some- what similar product is obtained from t h e substarice Cl6Hl2O, by digestion iu acetic acid solution with fuming hydrochloric acid and tin. These products will be further investigated when more material is to hand. Acfion of Szclphzuic acid. --In order t o decompose the methoxy- group of the substance CI6Hl2O5, i t was heated F i t 1 1 sulphuric acid a t 160" for about a quarter of an hour, by which means the violet- red solution became considerably browner.I t was then diluted, fir>t with twice i t s volume of acetic acid, then with alcohol, and t h e small amount of black precipitate so formed was filtered off. To the hot filtrate, boiling water was then carefully added, so that on cooling a brownish, crystalline product was deposited ; this was purified by crjstallisation from benzene with the aid of animal charcoal. 0.0531 gRve 0.1421 CO, and 0.022 H,O. C,,H1605 requires C = 66.66 ; H = 3.70 per cent. It forms long, orange-red needles somewhat resembling alizarin in appearance, melting a t 2 5 4 O , and at higher tempeyatures yielding a sublimate of needles.The antall amount of substance available for analysis might perhaps throw some doubt on the trustworthiness of this result. On consideration, however, it is evident t h a t the action of sulphuric acid on the substance C16H,205 at this temperature wonld most probably result either in the decomposition of the meth- oxp-group, or in tbe oxidation of the methyl group of the n-methyl- anthraquinone with formation of a carboxylic acid. T h a t the snb- stance obtained ( annot be the produrt of the latter action is evident, as in t b a t mse it, would only contain 60 per cent. of carbon. It dis- solves i n alkalis with a crimson-red coloration, which, ofi treatment C = 66.7 ; H = 4-20.934 PERKIX ASD EUMMEL : THE COLOURING PRIXCIPLES with zinc dust, becomes pale yellow, and, on exposure to a i 1 3 .rtigains its original tint. Its sulphuric acid solution is orange-red, changing to pale yellow when treated with a crystal of potassium nitrate. The melting point and general actions of this trihydroxy-a-methpl- anthraquinone appeared exceedingly similar to those described for emodin, and in order to make a comparison a sample of crude emodin was obtaiiied from Merck, of Darmstadt. This was purified by extracting with boiling tolnene, treating the latter with dilute caustic soda solution, and precipitating the red aqueous liquid with acid. The orange-coloured precipit'ate was crystnllised two or three times from acetic acid, and finally from benzene. As stated in the intro- duction, in general reactions, appearance, and melting point, no difference could be detected between emodin and the tribydroxy- methylanthraquiuone above described.Moreover, an examination of their behaviour with boiling nitric acid (sp. gr. l s . 5 ) resulted in both cases in the formation of a yellow, crj-stalline substance, probably a tetranitro-derivative. It is believed that when more material can be obtained, a comparison of the acetpl derivatives of these two sabstances will further confirm this identity. Action of Xitric acid on the substance C,,H,,O,.-It is not attacked by nitric acid (sp. gr. 1-42) in the cold, but on boiling it dissolres, forming a brownish-red solution, which, if rapidly cooled, deposits a crystalline, orange powder, consisting of microscopic, prismatic needles. These were collected, washed with nitric acid, then with water, and dried.0.1410 gave 4.9 C.C. moist nitrogen a t 22" and 756 mm. N = 3-91. C.1132 ,, 0.243 CO, and 0.0383 H,O. C = 58.54; H = 5-75. C,6Hl105(N0,) requires C = 58-36; H = 3-34; N = 4.25 per cent. Mononitroewzoclin methyl efher forms a bright 'yellow mass, melting a t 'L15--'L1i0, which, when heated strongly, decomposes suddealy with emission of a clond of black vapour. If suspended in water and treated with sodium amalgam, the original scarlet colour of the solution first formed passes to blue-violet, resembling that of sodium alizarate, and on adding acid brick-red flocks are precipitated, evidently an amido-compound. It dissolves in sulphuric acid with a bright red coloration, which, on heating, becomes dirty violet ; the addition of water then throws down dark brown flocks, dissolving i n alkali with a beautiful violet coloration, but possessing only feeble dyeing power.The preceding nitro-compound of the substance C16H,,0, was added to nitric zcid of sp. gr. 1.54, the solut,ion boiled, evaporated to a small bulk and cooled, when a semi-solid, crystalline mass was obtained ; this as treated with a little glacial acetic acid, collected,OF VEXTILAGO MADRASPATAXA. 935 and purified by cryatallisation from the same liquid. gave the following result. An analysis 0.1284 gave 13.7 C.C. moist nitrogen at 23" and 753 mm. N = 11.90. CI6HB,0,(~O2), = 12-07 per cent. S. CI~H,O~(??O~)* = 12.40 ,, K. This substance is, tberefore, either a tetranih-oemodin metA yl ether nr a tetrrcnifroewodin, a point which cannot he definitely decided until more matterial is obtainable.It forms a beautiful, glistening mass of long, thin, orange-yellow needles which melt and decompose a t about 275"; when more sti-onalyheated there i 3 R slight explosion. It dissolves iu solutions of the alkali hydroxides or ammonia with a red coloration, and if the crystals he exposed in an atmosphere which contsins but a minute trace of ammonia. the? are at) once reddened. When treated with alkaline redlicing aoents, a heautiful, blue-riolet liquid is obtained, from which a similar colonred precipitate grndnnllv sepa- rates, whilst on nentralisinc the solution maroon-colonred flocks are deposited. T t is insolnble in cold snlphuric acid, but, on heating, it, dissolves with an oranpe-red coloration ; the solution, when strongly heated, become5 of a brownish-violet tint.ans being erolred ; on now pouring it into water, dirty brown-violet flocks are precipitated, soluble in alkalis with a riolet coloration. The action of nitric acid on emodin itself is undergoing investiga- tion. Suhsfance of the Formula C,,H,,O1 (A). The ncefone mother ISquors from the purificaiion of the suhstance C,,Hl,05 (p. 9.32) were erapornted to a small bulk and cooled, when an almost colourless crptalline product separated ; this was collected and extracted two or three times with cold, and finally with a little hoiling benzene, and the sparingly soluble residue puri6ed by two or three crystallisations from much toluene with the aid of animal charcml.0.1143 gave 0.2962 CO, and 0.0512 H,O. C = 70.69; H = 4.98. C,6R,404 requires C = 71.11 ; H = 5.18 per cent. It formR a mass of beantiful, colourless Elistenin7 needles, and decomposes a t ahont 260" before melting. It is almost infolnble in henzene and alcohol, more soluble in acetone and acetic a d , crystal- lising readily from the latter. It dissolves in sulphuric acid with an orange-red coloration. which. on heaticp, becomes green ; on add- ing water, a greenish-black precipitate is produced, which i s pnrtiadlv soluble in a1 kalis with a pnrple, red-violet coloration. When distilled with zinc dust, a sublimate of greenish-yellow leaflets is obtained,936 PERKIN AND HUMMEL: THE COLOURING PRINCIPLES consisting of metbylanthracene, identical with that obtained in a similar manner from the emodin methyl ether. The substance (A) i s insoluble in cold alkaline solutions, but, on warming, gradually dissolves, forming a brownish-yellow solutioii ; on continued boiling, this darkens, aud if i t be now treated with acid a brownish-green precipitate is deposited.Action of Acetic Anhydride.-The subst'ance C16H1404 was heated with acetic anhydride until completely dissolved, about two hours usually being required to effect t h i s ; it was then boded for four hours longer. The resulting red solution, which had a strong green fluorescence, was mixed with a little methylic alcohol, and the greenish semi-cr~statlline mass which was precipitated, was collected, washed with metbylic alcohol, and purified by redissolving it in acetic anhydride and precipitating with methj-lic alcohol. 0.1177 gave 0.2842 CO, and 0.0524 H,O.Thus obtained it forms a pale yellgwish-green powder which, when heated, darkens a t 200-210", and decomposes before melting. It is almost insoluble i n alcohol, moderately soluble in benzene or acetic acid, and may be obtained from the latter in an indistinct ci*~stRlline form. I t s solution in benzene is red, with R strong, greenish flucres- cence. It is insoluble in cold alkali, but on continued boiling it is decomposed with formation of a brownish-red solution. Oxidation of the substaiice C,,H,,O, (A).-In studying this action, it was dissolved in excess of boiling glacial acetic acid, and a few drops of a dilute solution of chromic acid in the same solvent added ; the solution, when mixed with an equal volume of water and allowed to cool, deposited glistening orange-red needles, which were collected and analgsed with the following result.C = 65-SS ; H = 4.95, C,GH,lO,(C,K,O), requires C = 66.66. C and 5.05 H. 0.1293 gave 0.3193 CO, and 0.05 H,O. The product melted a t 200°, and was identical with the emodin methyl ether pi.eriously described. As stated in the introducrion, this substance appears to be a trilzydroxy-x-inethyla.llthl.ar~ole mono- nzetlayl ether. Action of A-itrL'c acid on C16H,404 (A).-It is not attacked by cold nitric acid (sp. gr. 1-42}, b u t on heating a violent action takes place ; on now adding water to the deep brown solution, orange-yellow flocks are deposited. Cold nitric acid (sp. gr. 1.5) gives a similar result, but on boiling, followed by evaporation and the addition of a trace of acetic acid, a crystalline product is obtained ; tlhis, after purification, was found to be identical with the tetranitroemodin or tetranitro- emodin methyl ether previously described (p.9%). C = 67.35 ; H = 4.29. ClaH,,05 requires C = 6i-6 ; H = 4-22 per cent.OF VENTILAGO MADRASPATANA. 937 8i'ubstance of the Formula C1sH,404 (B), m. p . 173. The extraction of this fourth substance from the residual crystalline products was at first a matter of considerable difficulty, for though these usually melted at temperatnresvarying between 163'and 176", they still contained about 30 per cent. or more of the two previously described substances melting at 200" and above 260" respectively. This diffi- culty mas further increased by the fact that this fourth substance can be obtained in two very distinct crystalline forms.The crystallising media found most useful for the separation were benzene, and some- times acetic acid, and by their use a definite product was obtained rnelting at 1'73". 0.1021 gave 0.2653 CO, and 0.0470 H,O. C = 70.86 ; H = 5.11. 0.1197 ,, 0.3108 ,, ,, 0.0540 ,, C = 70.82 ; H = 5.01. C16H1404 requires C = 71.11 ; H = 5.18 per cent. It crystallises in pale yellow or almost colourless needles, melting at 173", and at higher temperatures gives a sublimate and at the same time carbonises strongly. Frequently it crj-stallises from benzene in long needles of a more orange tint, which appear to contain benzene of crystallisation, since they become opaque and devoid of lustre on drying.One form was frequently obtained from the other during recrystallisation from benzene, though it was not possible to find any d.efinite means by which this change could with certainty be brought about. Both forms melt at 173", have identical properties, and on analysis give similar numbers. The sabstance C,,H,,O, (B) is sparingly soluble in alcohol, more readily in acetone, acetic acid, or benzene. When distilled with zinc dust, it yields methylanthracene, identical with that obtained from the previous products. I t dissolves in cold sul- phuric acid with an orange-brown coloration, which, on heating: be- comes green, and deposits greenish flocks on adding water (compare snbstance Cl,H,,O, A, p. 935).-4cfiol.i cf Acefic AdaycE,.ide.--It was boiled for six hours with acetic anhydride, and the hot solution was then mixed with a little alcohol. On cooling, a crystalline powder separated, which under the micro- scope was seen t o consist of minute prisms of a light yellow colour. It melts at 227 -229", and is moderately soluble in acetic acid, from which it can be crystallised. Its solution in alcohol has a strong blue fluorescence. Though chemically dist'inct from one another, the resemblance between the substance C,,H,,O, B a:id chrysarobin, obtained from Goa powder, is remarkable. They melt respectively at 173" and 171-178", their acetyl derivatires each at 227--229", and the eolu- %ions of the latter in alcohol both have a sti-olig blue fluorescence. VOL.LXV. 3 x938 PERKIN AND HUMMEL: THE COLOURING PRINCIPLES Oxidation of the szcbstance C16H1404 (B).-This was readily carlaied out by adding chromic acid to an acetic acid solution of the substance in a manner similar to that described under the head of '' ClsH1,O, (A) ,' (p. 936). The glistening mass of orange-yellow needles thus obtained melted at ZOO", and was identical with the emodin methyl ether previously described. 0.1080 gave 0-266 GO, and 0.043 H,O. C = 67.16 ; H = 4.4. C16H,,05 requires C = 67.6 ; H = 4-62 per cent,. The substance C16Hl,0, (B) is almost insoluble in cold, dilute alkalis, but, on warming, it dissolves, forming an orange-brown solution ; this, on exposure t o the air, slowiy acquires a blood-red colour, and on now adding an acid t o it, a brown precipitate is thrown down, which, after washing and drying, was extracted with much light petroleum. On concentrating and cooling the petroleum solu- tion, yellow needles were obtained, melting at 200", and identical with emodin methyl ether.Action of X i f r i c acid.-Ktric acid (sy. gr. 1.5) attacked it violentlj, and, on boiling and concentrating, a crystalline mass was obtained, which after purification gave yellow needles, identical with the tetra- nitroemodin or tetranitroemodin methyl ether obtained from emodiir methyl ether in a similar way. A comparison of the behaviour of the substances C16H,,0, (A) and (B) shows that their properties are very similar, the only difference of note appearing to be the slightly readier oxidation of the latter in alkaline solution.The substailce C16H14O4 (B) appears to be the secoucl trihydroxg-a-methylanthranolemonomethyl ether. The second main crystalline product, B, referred to on p. 929, affords perhaps the most ready material for the preparation of the substance CI6Hl4O, (B), consisting. as it does, chiefly of a mixture of this with the sitbstance (A), only a trace of emodin methyl ether being present. Its isolation is effected in the mauner previously described. In the various fractions and mother liquors obtained during the purification and isolation of the four previously described crystal- line compounds, no other substance has been detected ; should any- thing of the kind be present, it can only exist in the root-bark i n minute quantity. Further search will be instituted when largela quantities of material have accumulated.Substance of the Tormula CliHl,05. The main carbon bisulphide filtrate, which contained some quantity of alcohol (pp. 928--929), was evaporated to about one-third of its bulk ; on cooling, it became almost semi-solid,owing to the deposition ofOF VEXTILAGO MADRASPA4TAhT..A. 939 a somewhat gelatinous product. This was collected, and, after being repeatedly washed with small quantities of methylated spirit, the dark brownish residue, consisting of an amorphous substance mixed with a wax, was freed from the latter by repeated extractions with large quantities of methylated spirit, the treatment being continued as long as the filtrates deposited waxy matter on evaporation and cooling. The insoluble residue was then purified by crystallisatioii from benzene, and analysed, with the following result.0.104 gal-e 0.262 CO, and 0.0448 H,O. C = 68.7; H = 4.78. C,jHl,O, requires C = 68.92 ; H = 4.05 per cent. As thus obtained, it appears as a chocolate-coloured, friable mass, having an indistinct, crystalline appearance when examined under the. microscope. Its most interesting reaction is that which takes place on boiling it with alkali solutions, for it dissolves, forming an orange- brown solution, which on exposure to air deposits a blue precipitate, probably the alkali salt of a new compound ; from this behaviour, it appears to be an anthranole or allied compound. The above pre- paration most probably represents a pnre specimen of this substance ; as, however, only a small quantity has been obtained in this condition up to the present time, it is desirable to obtain more before extend- ing the study of its reactions.T h e Wax of the Formula C9Hl60,,, mt. p . 72". The hot, alcoholic extracts obtained during the purification of the substance mentioned in the preceding paragraph were evapo- rated to a small bulk, and the wax which sepamted on cooling was collected, rinsed with alcohol, pressed, dissolved in a large quantity of boiling alcohol, animal charcoal being added, and filtered. Whi!e still warm, a considerable quantity of a nearly colourless product separated, and this was filtered off before the whole had cooled down. The filtrate contained a small quantity of a crystalline pro- duct, which adhered to t'he wax with much tenacity, and from which it could only be freed by warm filtration.The wax was further purified by crystallisation from alcohol with use of animal charcoal. 0.1078 gave 0.333 CO, and 0.111 H,O. C = 76.66 ; H = 11.44. (C,H,,O), requires C = 77.14 ; H = 21.42 per cent. It forms a light, friable mass, having a faint flesh-coloured tint, and under the microscope is seen to consist of minute tangled needles. It melts at 72", and at higher temperatures gives off a vapour having the acrid smell of heated Tax. Purz3cation of the Crude C'oloitn'ng Natter. The main alcoholic filtrate, on standing, frequently deposited a 3 x 2940 PERKIX AND HUJIMEL: THE COLOURING PRINCIPLES small quantity of the two substances previously described ; the dark red filtrate from these, when evaporated, left a thick, treacly mass of the colonring matter still contaminated wit,h wax and other impnrit8ies ; this was dissolved in a large quantity of boiling benzene, and treated with an excess of animal charcoal ; the filtrate was then agitated with a boiling solution of dilute caustic soda, and the blue- violet alkaline solution containing all the colouring matter was separated from the benzene; this was a matter of some difficulty, as the latter now held in suspension a semi-solid substance in some quantity, very often necessitating the previous removal of the latter by filtration of the whole mixture through calico.This semi-solid substance was ciytallised from alcohol, and proved to be a wax-like substance, containing, no doubt, a certain quantity o€ tbe wax previously described.It will, however, be more thoroughly in- vestigated when larger quantities can be obtained. The orange-coloured benzene solution, from which the colonring matter had been extracted by treatment with alkali, was evaporated t o dryness ; the viscid residue after a time deposited crj-stnls, mhicli were thrown on to calico, washed with a lit'tle alcohol, and finally purified by crystallisation from the same solvent with the aid of animal charcoal. 0.11 ga.ve 0.2722 COz and 0.044 H,O. C = 67.48 ; H = 4.4. Cl6Hl2O5 requires C = 67.6 ; H = 4-22 per cent. This substance crystallises i n orange-coloured needles melting at ZOO', and is evidently identical with the emodin methyl ether pre- viously described. Its appearance a t this point is singular, as, owing t o its somewhat sparing solubility, one would naturally expect it t o be separated at a mucli earlier stage.It is highly probable, there- fore, that as alkali was used for the extraction of t h e colouring matter from the benzene solution, it is here present merely as a de- composition product, either of a small quantity of the compound CL6Hl4O4 (B), or of some other substance which has not yet been iso- lated. I n furt>her support of this suggestion, i t is worthy of note t h a t not one of the other crystalline products here described could be detected in this benzene solution. The Coloiiriiig X a t f e r . Vmtilagili. The red-violet alkaline solution mentioned above, when nentralised with acid, deposited a maroon-coloured resinous precipitate, which became semi-liquid when the mixture was warmed.It was dissolved on agitating with benzene, and again removed from the latter by means of dilute alkali; this treatment being continued with fresh quantities of benzene, until the final benzene solution, after extraction nith alkali,OF VEXTILAGO MADRASPATANA. 941 was only faintly yellow. The alkaline solution was then saturated with common salt, and the amorphous precipitate of the alkali salt of the colonring matter was washed with a strong solution of common salt containing a trace of alkali, until a nearly colourless filtrate was obtained. By this means, a small quantity of substance, soluble in alkali, with an orange-red coloration, was removed. The alkali salt was then suspended in water, neutralised with acid, and the free colonring ma.tter dissolved in ether, and freed from a small quantity of dark- coloured impurity by filtration.The resinous residue left on evapo- rating the ethereal solution to dryness, was now extracted with a little warm methylic alcohol ; by this means a further quantity of the sparingly soiuble impurity remaining undissolved was removed, and the alcoholic extract was then poured into a large bulk of dilute potas- sium carbonate solution. On adding common salt to the alkaline liquid, the potassium salt could be readily precipitated in two frac- tions, the first of which carried down with it the last traces of the less soluble substances, the second and largest representing a very pure product. The latter was suspended in water, decomposed with acid, the colonring matter dissolved in ether, the solution evapoiated, and the residue dried at 110" until the weight mas constant.Ana- lyses of different preparations gave the following numbers. C = 61.72 ; H = 5.02. 11. 0.1181 ,, 0.2700 CO, and 0.0565 H,O. C = 62.34; H = 5.31. Cl,H1406 requires C = 62.06; H = 4-83 per cent. I. 0.1182 gave 0.2675 CO, and O-OG35 H,O. As thus obtained, ventilagin is a reddish-brown, brittle, resinous mass, possessing a slight metallic lustre ; it softens a t about lOO", and becomes liquid a t 110". Although the niimerous attempts made t o obtain it in a crystallice condition have been unsnccessful, rotwith- standing the modes of purification adopted have been very varied, it is highly probable that the preparations described above consist of the substance in a pure condition.It is Tery readily soluble in alcohol, ether, and bznzene, very strong solutions of m-hich deposit it, on cool- ing, as a gelatinous mass, and it is only sparingly soluble in light petroleum and in water. When strongly heated, it is carbonised with evolution of a vapour having a phenolic odour, which is best collected by passing it through ether or alcohol ; on evaporating these soh- tions, a pale brown-coloured oil is obtained, which does not solidify on standing; it is soluble in alkali, b u t gives no characteristic reaction with ferric chloride either in aqueous or in alcoholic solu- tion. On distillation with zinc dust, ventilagin yields a small quantity of greenish leaflets consisting of x-methylanthracene.VeDtilagin dissolves in cold sulphuric acid, forming a magenta-coloured solution, which, on heating to 150", evolves sulphurous acid ; a carbonaceous942 PERKIN AXD HUMMEL: THE COLOURING PRIPU'CIPLES crystalline product from benzene fractionally crystallised product is produced, only a trace of which is soluble in alkdis. It dissolves in solutions of the alkali hydroxides and carbonates, forming colonred solutions slightly redder than those obtained from anthra- pnrpurin ; on adding common salt, the alkali salts are precipitated in an amorphous condition, and may be purified by taking advantage of their solnbiiity in alcohol. The barium salt, which is precipitated on adding barium chloride to aqueous solutions of the sodium or potas- sium salt, is a violet-red, amorphous powder.When a boiling alka- line solution of the colouring matter is treated with zinc dust, an orange-yellow solution is obtained, which, on exposure to air, re- sumes its original tint; from this it appears that ventilagin is a derivative of a-methylanthraqninone. It dissolves in nitric acid (sp. gr. 1.4 or 1.5), but no precipitate is produced on adding water to the orange-brown solution. If, however, the nitric acid solution is evaporated to dryness, a colourless, crystalline residue is obtained, consisting entirely of oxalic acid. As stated in the introduction, t h i s colonring matter is possibly allied to alkannin, the colouring matter substance C,iH,?O, Ventilago ntadrnspatana is extracted with carbon bisulphide, the extract evaporated, and mixed with alcohol. subvtance ~16H1200, m.p. 800" substance C16H140py, filtrate XI. p. l r 3 containsOF VESTILAGO XADRASPATANA. 943 from Anchiisa tinctoria, and we propose for it the name omtilagin; further experiments will be carried out with the object of determin- ing its constitution, and an investigation of alkannin from the L4~tchusa fimtoria will be carried out at the same time. The details of the separation of the various substances here described being somewhat difficult of explanation, the accompanying scheme (pre- ceding page) has been drawn out in which are embodied the main methods used for this purpose. DyeiiLg Pruprties. Experiments in dyeing with the iaoot-bark of TTPntilago madraspatantt have been previously made by Gonfreville (L'B1.t de la Teinture des Laines, 1849, p. 542) and by T. Wardle (Report ou the Dyes and Tam cf India, 2887, pp. 3, 21). Our experiments were made on ordinary stripe mordanted calico printed with aluminium and iron mordmts, also on wool and silk with chromium and tin mordants in addition. Its application presents no difficulty, although, owing to the slight solubility of the colouring matter in water, dyeing does not begin till the temperature reaches 70-80". No additions t o the dye-bath are necessary, except in the case of wool and silk, with which it is advantageous to use a small percentage of calcium acetate to correct, the acidity of the mordanted fibre. On cotton, it gives, with aluminium mordant, a purpIish-red, much bluer in shade and duller than that given by Lima-wood ; with iron mordant, it gives grey to black according to the concentration of the mordant, the grey being a more neutral tint than the purplish-grey given by Lima-wood ; w i t h a mixture of aluminium and iron mor- dants, the colour is purplish-chocolate. On oil-prepared calico with aluminium, chromium, and iron mor- dants, one obtains rich claret-red, purplish-brown, and purplish-black colours repectively. Colours similar to the foregoing are obtained on wool and silk, and with tin mordant a red, very similar to that given by alizarin with aluminium mordant. Generally speaking, the Ventilago colours ai-3 somewhat similar to those obtainable frem Camwood or from Lima-wood ; like the latter, they are fugitive to light, but are decidedly faster to soap ; they do not, however, equal in this respect the alizarin colours. Although the chemical part of this paper seems to indicate that the colonring matters of Ventilago madraspatana and of Anchusa finctol-ia (alkannin) are possibly related to each other, the colours they yield in dyeing are very different, the latter giving, with aluminium mordant, a, bluish- grey, the former, as already stated, a purplish-red. We have been unable, indeed, to find any other colouring matter possessing exactly944 MACDONALD AND MASSOS: A PRODUCT OF THE the same dyeing properties, and, therefore, taking the latter in con- junction with the results of the chemical examination, we have come to t h e conclusion that this dye-stuff contains a new colouring mattel-. Clothworked Research Laborntor.y, Dyeing Depcr rtnzeiz f, Y o rkshire Coll eg P .

ISSN:0368-1645    

DOI:10.1039/CT8946500923

出版商:RSC    

年代:1894

数据来源: RSC

摘要:

944 MACDONALD AND MASSOS: A PRODUCT OF THE LXXVL-A Prodztct of the Actiou of Nit& Oxiclcj o?a Soc7 i 2 I na Eth y 1 u t e. By G. W. &~ACDOSALD, B.Sc., and ORNE DLissos, KA., n.Sc. THE perusal of W. Traube’s paper, “ Ueber Isonitramine,”* has led US to decide on the publication of such results as we have already obtained in the study of the above interaction, though they are in- complete, and we should otherwise have waited till we had obtained fuller and more satisfactory analytical data, and had succeeded in clearing up certain doubtful points. We have found that nitric oxide is absorbed in large quantity, though but slom-Is, by an alcohoiic solution of sodium ethoside, and that the most characteristic product of the change is the sodium salt of a new acid, which, being insoluble in alcohol, separates as a crystalline precipitate.From the sodium compound various other salts have been made, and the unstable acid has been cbtained in aqueous solution. Analyses of the copper salt, point to CH,:N,OdH, as the formula of the acid, half the hydrogen of which is replaceable by metal. This formula is identical with that which we now find accorded by Traube (who bases it on the analysis of a barium salt) to a compound which he designates wiefhylene-dii,coi~if,.cLnzil2e, CH,(N,O,H),. He obtained the sodium salt, together with sodium acetate, by the action of nitric oxide on acetone in the presence of alcoholic soda, and he states that the action occurs according to t h e equation CHs-CO.CH, + 4x0 + 3KaOH = CH3*COONa + CH:,(NZOZKa)2 + 2H,O.As the compounds obtained by us and those described by Traube agree not o n l ~ in composition but also, so far as we can judge from his account, in their characteristic properties, there can be no doubt that they are identical. In that case, it would appear probable. This number. issued on the 23h June, W‘BS received The resolts of our work, now described, were * Berichte, 1894, 11. 1507. in Melbourne on the 2nd August. all obteined before that date.-O.M., 6th August, 1891.ACTIOX OF NITRIC OXTDE ON SODIUX ETHYLATE. 945 either that the acetone employed by Traube is not, an essential parti- cipator in the action, or that in our apparently simpler mode of preparation acetone or some similar compound is produced as a, preliminary step. We have, however, given but little attention as yet to the products of the action other than the new sodium salt, this being an aspect of the question which we had set aside in the meantime.Bode of Action.-The first experiments were undertaken in order- to ascertain whether nitric oxide would react with sodium amalgam in presence of absolute alcohol to produce sodium hypo- nitrite, this salt having recently been shown by Jackson (Proc., 1893-94, No. 128, p. 210) to be insoluble in, and stable in the presence of, alcohol, though it is decomposed by water. Hyponitrite was not, obtained except in certain cases where the cQmplete exclusion. of air was somewhat uncertain; but in all other cases a white o r yellowish crystalline precipitate was formed, which had peculiar pro- perties, and could not be identified with any knom-n substance.This led to the trial of an alcoholic solution of sodium ethoxide, instead of sodium amalgam and alcohol, and resulted in the production of the same precipitate as before. If the nitric oxide and excess of the ethoxide be enclosed in a eudio- meter tube oyer mercury, the absorption of the gas can be observed, and is found to go on slowly, and with great regularity, till at the end of about a fortnight it is complete. During this time, the crys- talline precipitate, which is at first 11-hite, increases in quantity and acquires a decided yellow tint, as does also the alcohol. The develop- ment of colour is similar to, but not so pronounced as, that seen in the case of sodium ethoxide solution exposed to air.In one such ex- periment, 6.2 C.C. of a solution of the ethoxide containing 0.237 gram of sodium was put up with 61.7 C.C. (at normal pressure and tempera- ture) of nitric oxide, that is, the ratio of Na to NO was about 4 : 1 ; and in another experiment the ratio taken was 6 : 1. In both these cases, the appearances were those described above, and the absorption became complete in about 13 days, only a very small volume of what appeared to be nitrogen remaining, equal to about 3 per cent. of the oyiginal gas. This, however, was not impurity derived from the nitric oxide used, which had been prepared with Epecial care, and proved t o be pure to the extent of 99.8 per cent. In another experi- ment, in which the ratio Na : NO was made unity, the absorption is not yet complete, nor has it altogether stopped, although almost three months hare elapsed since it was started; in this case, both the crjs- talline 7recipitate and the alcohol are quite devoid of colour.Iu order to obtain larger quantities of the salt, a glass gas-holder was used, capabie of holding about 3 litres. A known qnantityof t h e946 MACDONALD AKD MASSON: A PRODUCT OF TRE ethoxide solution was poured in so as to form a layer at the bottom, and the air was then displaced by dry hydrogen, and this in turn by pure nitric oxide. This gas-holder was put in connection, through a drying tube, with another and similar gas-holder full of nitric oxide, on which a constant water pressure could be maintained. Special care was taken t o exclude air at the outset, and to guard against leakage through the taps.After two or three weeks, when several litres of the gas had been absorbed, the residual nitric oxide was dis- placed by hydrogen, and the gas-holder then opened. The crystalline precipitate was then separated from the excess of ethoxide and from any soluble products, by filtration and thorough washing with alcohol, but it was notl found possilile to entirely free i t from colouring matter in this way. It can, however, be dissolved in water and reprecipitated by alcohol; and if the alcohol is allowed to diffuse slowly into the aqueous solution large plates and long delicate needles are obtained. From the sodinm salt, certain insoluble metallic salts have been prepared,which may be described before proceeding further.The copper salt is the most characteristic, and its formation is a delicate test for the presence of the sodium, or other soluble, salt in solution. It separates as a sky-blue powder when copper sulphate is added, not in excess, to a solution of the sodium salt slightly acidi- fied with acetic or sulphuric acid. It is completely insoluble in water, cold or hot, and is but slightly affected by cold dilute acids, except nit8ric acid, which attacks i t readily ; it dissolves in excess of copper sulphate, however, forming a green solution, whilst with ammonia i t gives a dark blue solution, from which it may be repre- cipitated by neutralisation. The dry salt, when heated, decomposes with so much energy as to produce glowing and to scattsr the resulf- ing cupric oxide, Four different preparations were used for analysis, of which I, 11, and 111 were got from the earlier experiments with sodium amalgam, and IV by the sodium ethoxide method.The salt was in all cases dried over sulphuric acid. The discrepancies i n the resnlts obtained are somewhat greater than could be desired, but they leave little room for doubt as to the correctness of the formula CH2N40aCu for the pure salt. Found. -7 Cdculated for /------i-- C'H,N,04C~i. I. 11. 111. IV. C ....... 6-07 - 6-19 6.23 6-89 H ........ 1-01 - 1.35 1.42 1.56 x ........ 28-55 27-43 - 26-00 0 ........ 32.42 C u ........ :32- 1 .i 31-46 31-21 :31*9:3 31.07 - - - - - The cadn&itc?n .w7f falls rather slowly as a white, crystalline precipi-ACTION OF NITRIC OXIDE ON SODIUN ETHYLATE.947 tate of characteristic appearance, when cadmium snlphate is added to an aqueous solution of the sodium salt. In its behaviour towards acids, and on heating, it resembles the copper salt. T t is a hydrated compound, and estimations of the cadmium in samples obtained from two different preparations of the sodium salt and dried over snlphuric acid are in accordance wit11 the formula CH2N40aCd,2H20. This formula requires 39.72 per cent. of cadmium; found, 40.03, 39.50, and 40.22 per cent. Further analyses of this salt have not yet been made. The silver salt is a white, curdy precipitate, somewhat soluble in cold water, and readily dissolved by dilute acids or ammonia, but can be re-obtained from such solutions by exact neutralisation. It darkens slowly in the cold, and when boiled with water i t decomposes rapidly with evolution of gas and deposition of a silver mirror.It explodes when heated in the dry state. A single estimation of silver in a sample partially decomposed by exposure gave 63.22 per cent.; calculated for CH2N40aAg2, 61.71 per cent. T h e barium salt, obtained by precipitation with barium chloride, has properties agreeing, so far as we have obserred them, with those described by Traube (Zoc. /-it.). He gives the formula CH2K,04Ba,2H20 for the salt. dried oyer sulphuric acid. His description of the silver salt (the only other salt examined by him) also agrees with our observations. Other characteristic insoluble salts ale obtained by precipitating a solution of the sodiuni salt by calcium chloride, lead acetate, mer- curous nitrate, and ferrous sulphate.As stated abore, it is obtained in the first instance, after washing the product with alcohol and drying over sulphuric acid, as a crystalline powder, more or less yellow from the presence of impurity. It is freely soluble in water, but does not deliquesce or alter on exposure to air. It explodes violently when heated. Estimations of the sodium and nitrogen contained in different preparations, not quite pure, point to the ratio Na : 2N, and to the formula CH?",O,N~,,H,O. This requires 23.23 per cent, of sodium and 38-28 per cent. of. nitrogen ; and the results oh tained were The sodizm salt itself has not, as yet, been fully examined. Sodium, Nitrogen, ,, 26-84, ,, 26.77, ,, 26.36 ,, (I) 23.22, (11) 23.37, (111) 2351 per cent.The large crystals already mentioned as obtained by diffusion of alcohol into the aqueous solution appear to be a different hydrate, but have not been analysed. They lose water at abont 150", leaving948 NACDONALD AND MASSON: A PRODUCT OF THE an anhydrous salt with otherwise unaltered chemical properties, Explosion occurs above 200". The aqueous solution of the salt is. faintly alkaline to litmus. It is decomposed by sulphuric acid with evolution of gas which contains nitrous and nitric oxides, and nitrogen ; this decomposition is rapid if strong acid is empIoyed or if heat is applied, but with cold diliite sulphuric acid the action is slower and a peculiar odour is noticeable. The salt gives the Liebermann reaction with phenol and strong sulphuric acid.It readily reduces acid potassium permanganate, but not Fehling's solu- tion. It rednces gold chloride even in the cold, and platinum chloride on prolonged boiling. T h e amnzoizium salt has been obtained by decomposing the copper salt, suspended in water, with a slight excess of ammonium sulphide, filtering, and evaporating on the water-bath. Partial decomposition occurs, but a crystalline residue is obtained which contains ammonia and gives all the precipitations already mentioned as characteristic of the sodium salt. I t does not explode when lieated, but decomposes. with evolution cf gas. T h e f r e e acid has also been obtained, though only in solution, from the copper salt by suspending it in water and passing in a stream of hydrogen sulphide.The excess of the latter is then removed by a current of hydrogen, which also serves to coagulate the copper sulphide and make filtration possible ; the solution obtained in this way is strongly acid to litmus, and reduces potassium per- manganate solution. When fresh, i t can be reconverted into the sky- blue copper salt by precipitation with copper sulphate, or into t h e sodium salt by neutralising with soda and precipitating with alcohol, There can thus be no doubt that the solution, when first prepared, contains the acid CH2:N404H,. But it is rapidly decomposed whelm heated; and, even in the cold, it slowl>- evolves gas, apparently nitrous oxide and a small amount of nitric oxide, and a yellow solution remains which contains a different acid and gives different reactions. After some days, it no longer gives the characteristic: copper precipitate; it yields a solid residue on evaporation; with alk;clis, it gives deep omnge-coloured solutions ; and with silver nitrate, it gives a salt which separates from boiling q-ater in the form of beautiful, red crystals, insoluble in cold water, and highly ex- plosive when heated.Further examination of the free acid and it5 decomposition products has not yet been possible. Constitzctiow of the Acid.-Among compounds already known, those which approach nearest to the new acid and its salts in mode of formation and in properties are, as Traube also notices, Frankland's so-called dinitromethylates ai;d dinitroethylates ( Phil. Tram., 147, 59). These salts differ from oiirs in being all soluble in water, but It decolorises iodine solution.ACTION OF SITRIC OXIDE ON SODIUM ETHTLATE. 949 they resemble them in their explosive characters and in other respects, and especially in their mode of decomposition by acids.It is therefore probable tbat their constitution is similar, the new com- pound differing from Frankland's only in so far as it contains the dyad methylene group in place of monad alkyl. CH,*N,OJEI: c H? (N,O,H), '' Dinitromethylic " acid. The constitution of the --N,O,H-group requires further investiga- tion, but some evidence is already available. Franchimont and van Erp (Abstr., 1894, i, 273) have quite recently shown (contrary to the view suggested in the new edition of Watts' Dictionary) that this group in Frankland's compounds must be isomeric, and not identical with that contained in the nitramines, which do not decompose under the influence of dilute snlphuric acid.It is on this account that Tranbe applies the name .I-nethyleilediLo?~~fra~~?ae to the new com- pound, implying thereby that it contains a group isomeric with the iiitramine group -NH-NO2, but leaving the question of t>he exact Constitution unanswered. The choice seems to lie between the alter- S e w acid. NO natives -N*N*OHand Of these, the former 1%-as suggested V 0 by Frankland himself in his account of the dinitroethylates, but no sufficient reasons mere given in support of it. Either formula is capable of expressing the production of nitrous oxide by decomposi- tion of the salts with acids and the formation, in the case of the new methylene compound, of a stroDg reducing agent, namely, formaldehyde, capable of reducing gold and silver salts, and potas- sium permanganate. Eut the second formula is better in accord v i t h the fact now observed that the new sodium salt gives the Liebermann reaction, which is generally accepted as evidence oE the presence of a nitrosyl group, and is here strongly suggestive of a nitrosamine. According to this view, which seems the most likely one in the present state of our knowledge, the new compounci ?,zt.thyZe7iedihydroey.lzif rosanzine, CH2 [ X(S O)*OH],. The Chemical Laboratcrry, Uiz ir-ersity of Xd boil me.

ISSN:0368-1645    

DOI:10.1039/CT8946500944

出版商:RSC    

年代:1894

数据来源: RSC

摘要:

LXXVII. -De?-iziatives of Tefimmeth.ylen e. By W. H. PERKIN, jun. THE close relation ship which exists be tween te tramet h ylenecarboxy lic acid I and the fatty acids, especially the valeric acids, CH, yH2 CH2*CH-COOH not only in physical properties, but also in general chemical behavionr, has been freqnently pointed out; and, in view of this close agree- ment, the statement of Freund and Gudeman (Ber., 21, 2695), that the amide of tetramethylenecarboxylic acid on treatment with bromine and potash does not yield more than traces of tetramethyleneamine, Q H ~ --QH? , seemed remarkable, especially when it is remembered CH2*CH-NH2 t h a t under similar conditions 90 per cent. of the theoretical yield of isobutylamine is obtained from the amide of isoraleric acid (Hofmann, Ber., 15, 769).In order to investigate the cause of this difference in behavionr, numerous experiments on this reaction were made, and it was fouucl that when the pzwe amide of tetrametliylenecarboxylic acid (melting at 152-153") is employed, an excellent yield of tetramethyleneamine is obtained ; on the other hand, an impure amide, such as was used by Freund and Gudeman (they give the melting point of 138"), gires only traces of the base under the same conditions. SH?*YH2 is a colourless liquid which CH*- CH NH~' Tetrurn eth y leneamine, boils at Sl", and closely resembles isoamylamine in general proper- ties; it has a strong, basic odonr, mixes with water with develop- ment of heat, and gives beautifully crystalline salts, of which the hydrochloride, CaH9N, HCl, and the p Zntin ochlorid el ( C,H9N) 2, H2P t C Is, were analysed : the former salt, when treated with silver nitrite ill aqueous solution, is conrerted almost quantitatively into hydroxj- tetramethylene.This secondary alcohol boils at 133" and shows the closest resemblance to the fatty alcohols containing the same number of carbon atoms ; it might, indeed, be readily mistaken for normal butylic alcohol. When digested with fuming hydrobromic acid, hydroxytetra- methylene behaves in a remarkable manner, as although i t is in part converted into bromoteh-amef?) ylene, a colourless oil boiling at 104", and closely resembling butylic bromide in properties,DERITATI17ES OF TETRAMETHTLESE. 951 + HZO, p 8 - p + HBr = p*p CH2*CH*OH CII,*CHBr the principal product is a dibromobntnne, C1H8Br2, boiling at 173- 174", which must have been foymed by the disruption of the fonr- carbon ying, a behaviour which, so far, has only been observed in the case of trimethylene and some of its derivatives.As none of the dibromobntanes," which had, so far, been prepared, agreed in their propert'ies with the product obtained from hydroxy- tetramethylene, experiments were instituted with the object of synthesising the missing mernbers of the series, and in this way determining with which of them this dibromobutane was identical. Ultimately, the required isomeride was obtained as follows. Aldol was reduced, in neutral solution, with sodium amalgam and thus converted into 1 : 3-dihydroxybutane, CH,.CH(OH)*CH,*CHO + H? = CH,-CH(OHj*CH2*CH,.0H, and this, when treated with hydrobromic acid, yielded 1 : 3-dibromo- butane, CHS*CHBr*CH2* CH2B~*, which boils a t 174.", and was found to be identical in all respects with thc dibromide obtained from hydroxytetrameth ylene.I n the formation of this dibromobut'nne from hydroxytetramethyl- ene, it is probable that bromotetramethylene is first produced, which then reacts with a fnrther quantity of hydrobromic acid with dis- ruption of the tetramethylene ring, thus: yH2i * $I H, BrCH CE& + HBr = CH3.CHBr*CH,*CHzBr. That the tetramethylene ring may be split in this way is a very interesting fact when taken in connection with Baeyer's view of t h e stability of rings as developed by him in his well-known " Spannung's Theorie " (Ber., 18, 2277), which is, briefly stated, as follows.I f it be supposed that the four affinities of the carbon-atom act in the direction of lines drawn from the centre of a tetrahedron to the * This will be readily seen from the following table, which gives the know11 isomerides with their boiling points :- 1 : 2 Dibromobutane, CH, - CH?. CHBr - CH,Br B. p. 166" 1 : 4 9 9 CH,Br - CH2 * CH2 - CH2Br ,, 190 2 : 2 1 9 CH,-CH,- CBr,-CH3 ,, 145 2 : 3 1 9 CEII.CHBr -CHBr-CH3 ,) 158 1 : 3 Dibromobutane, CH, . CHBr- CH,. CH,Br obtained as explained above boils at 174".952 R. H. PERKIN, JUN.: corners, then these directions are inclined to one another at an angle of 109" 28'. When such carbon-atoms combine together to form a closed chain the direct,ion of these affinities must be altered, and to do this a certain strain (" Spannung ") must be applied, which may be measured by the angle through which the line of affinity is supposed to be deviated.CH, I n the case of trimethylene H2C---CH2, /\ for example, in which the carbon-atoms are assumed t o be situated at the corners of an equi- lateral triangle, the angles between two affinities of each carbon-atom forming the ring is 60", that is to say, the direction of each aflinity has been displaced through an angle of i(109" 28'-60°) = 24" 4A' ; in HZC, -, C H Z I - the formation of tho tetramethylene ring, : ' , the angle of H,C~-~CR, displacement is much less, e.g., &(109" 2Sf-900) = 9" 44. The following table, taken from Baeyer's paper, shows the devia- tion of the direction of' each affinity of the carbon-atom necessary in combining with other carbon-atoms to form rings.CH, CH? /\ /\ 1 1 /\ 1 1 I I I I \/ CH, CH, CH,.CH, E,C C'HZ HZC CH: CH, HZC-CHZ CHZ-CHZ H,C---CH, HZC CHZ CH, + 54" 44' + 24' 44' + 9" 44' + 0" 44' - 5" 16' In ethylene, the deviation is greatest, and the tendency of the ring t o open out with formation of additive products vould, therefore, be assumed ts be yery pronouiiced ; and this is found experimentally to be the case, since ethylene coinbines with the greatest ease with hydro- gen bromide, bromine, chlorine, and even iodine ; in the case of tri- methylene, the deviation is very much less, and this gas, although it combines with hydrogen bromide and with bromine with disruption of the ring, does so with much greater difficulty than ethylene; chlorine does not act on trimethylene at all except in the presence of sunlight, and then the principal products of the action are the substitu- tion products, nionochlorotrimethylene, C3H5C1, and dichlorotrimethyl- ene, C3H1C1?, some trimethylene chloride, CH2C1.CH2* CH,Cl, and other products being, however, also produced (Gustavson, J. pr. Chem., 1894, 50, 380). In accordance with Baeyer's theory, the tetramel hylene ring should be much more stable than the trimethylene ricg, and this is clearly proved to be the case from a comparison of the properties of some of the derivatives of these hydrocarbons.DERIVATIVES OF TETR-UIETHTLENE. 953 Tri~ethyleneca;l.boxylic acid, for example, is readily acted on by hydrogen bromide with formation of -l-bromobutyric acid, whereas tetramethylenecarboxylic acid is not decomposed even when heated with fuming hydrobromic acid at 150", and many other similar cases might be cited ; on the other hand, bromotetramethyvl- ene is decomposed by boiling concentrated hydrobromic acid, but comparatively slowly. This difference in the behavioar of bromotetramethylene and tetramethylenecarboxylic acid is very remarkable, the nature of the substituting groups obviously having a very pronounced effect on the stability of the ring; this same difference of stability is also very marked in the case of the trimethTlene compounds, as will be shown in a subsequent paper.It would be exceedingly interesting to studj- the behaviour of pentamethylene and hexamethylene and their derivatives towards hydrobromic acid, as, according to Baeyer's theory, the former, owing to the slight strain necessary for its formation, would probably not be attacked ; this,moreover,seems to be true, since pentamethylene-1 : 2-di- carboxylic acid, for example, may be heated with hydrobromic acid at 150" without change.Hexamethylene and its derivatives might under similar conditions be decomposed, but probably not so readily as tetramethylene com- pounds, and the behaviour of these substances would be an important test of the value of the " Spannung's Theorie." In connection with this point, it should be noted that benzene, when heated at 260--270", with hydriodic acid containing iodine, yields hexane (comp. Baeyer, Banalen, 278,89) with disruption of the ring. Possibly, in this case, hexamethylene iodide, C6HI3I, or di- iodide, C6HIZ11, may first be produced, and then redwed t o hexane by the further action of the hydriodic acid.yH2'yH2 may be readily obtained by the CH,*CHCl' Clilorotet ramethylem, action of phosphorus pentachloride on hydroxytetrnmethylene. It is a colourless oil which boils at Sjo, and when treated with potas- sium iodide in alcoholic solution yields iodotef).amethyZei2.e, C4HiI (h. p. 138"), although the action does not take place readily. The last-named substance was prepared in considerable quantity (30 grams), in the hope that tetramethylene might be obtained from it by reduction under suitable conditions, and that by treatment with quinoline (Baeyer, AnnuZen, 278, 107), dihydrotetrene" might be formed, thus. * With regard to this nomenclature, see Trans., 1890, 57, 214.VOL. LXV. 3 Y?54 TT. H. PEREXS, JUX.: The preliminary experiments gave results which seemed to bear out these suppositions ; buc, on the other hand, it mas soon evident that a detailed examination of such hydrocarbons, which would probably boil at about 20" and 35O respectively, could only be carried out with such quantities of material as it would be almost impossible t80 obtain by the methods described in this paper, and therefore this part of the snbject was not further investigated. The close resemblance of chloro-, bromo-, and iodo-tetramcth- ylene to the corresponding butyl compounds is very marked, as d l be seen from a comparison of their properties. The tetm- methylene compounds all boil soiiiewhat higher than t h e butyl coni- pounds, and this is also noticeable in the case of other tetramethylene derivatives, which all, without) esception, boil higher than the corre- sponding saturated fatty compounds, as is seen from the following CH,*FH1 table, in which the s)-mbol [7 is nsed for I CHz*CH-' 1 B.p. 1 Diff. Butane i 1 derivatives. I B. p. Tetramethylene derivatives. i I--- I-------- --- o-COOH.. ....... 195' ' C,H,. COOH ....... 0 .NH2, .......... i 81 1 C,Hg*NH,. . . . . . . .......... .......... 0.033.. 12? 1 C4HS.OH 0.C1 ............. 1 85 1 C,H,Cl ............. 0 - B r 1 10-4 C4H,Br 0.1 ............. ; 135 CIH,I ............ ............ ............. I i 186" 76 116 77 100 131 Other examples illustrating the same point might also be given.I n the second part of this paper, the results of a detailed in- vestigation of the action of bromine on tetramethylenedicarboxylic are described, an inrestigation which as acid, FH,*YH*COoH CH,.CH* COOH' undertaken in the first instance in the hope clf obtaining tetrene- the study of which, in considem- dicarboxylic acid, YH:Y*COOH CH: C *COOH' tion of the similarity of its constitution to that of phthalic acid, must have given very interesting results ; but this acid could not be. isolated. Nevertheless, mauy other substances were obtained which are of considerable interest. When tetramet hjleiiedicarboxylic anhydride is treated R T i & excess of bromine in presence of phosphorus, substitution takes.DPRIVATIVES OF TETRAMETHYLEKE. 953 place, and the crystallised product of the reaction, after treatment with water, is found to consist principally of cis-dibromotetm me th ylenedicarboxy lic acid, ?H2- yBr* COOH CH2*CBr* COOH + 2HBr.+ 2Br, = QH,*$IH*COOH CH, CH * COOH This acid melts at, 202-205", and when treated with acetic anhy- dride yields an anhydride (m. p. 104") ; as this is reconverted into the original acid when dissolved in water, the latter must be the cis- dicarboxylic acid. T?-am-dibromotetramethylenedicarboxylic acid could not be ob- tained by heating the cis-acid in sealed tubes with concentrated hydrochloric or hydrohromic acid at 180" ; the product from these experiments melted at 200--203", and could not be distinguished from the cis-acid. The trans-acid is therefore either not formed in this way: or is very similar to the cis-acid in physical properties; the latter alternative is by no means improbable, as the cis- and trans- modifications of ktramethylenedicarboxylic acid are themselves so very similar that a careful examination of their chemical properties is necessary before they can be distinguished (Trans., this TO]., 572).It would be very remarkable if dibromotetramet hylenedicarboxylic acid were found to exist in only one form, not only from the fact that the parent substance tetramethylenedicarboxylic acid itself exists in the cis- arid fraizs-modifications, but also because succinic acid, which is similar to these acids in constitution, when treated with bromine, yields a mixture of stereoisomeric dibromo-acids. Br Br H+COOH H+COOH H* C - COOH COOH*C*H Br Br Cis-acid. Trails-acid.The study of the action of alkalis on dibromotetramethylenedi- carboxylic acid has yielded very unexpected results, as, when this acid is treated with either strong or weak bases, such as potash, baryta, silver oxide, quizloline, or dimethylaniline, under various con- ditions, the principal product of the action, and indeed, in most cases the only product which could be isolated, was bromodihydrotetrene- carboxylic acid, the formation of which may be represented thus I 1 QH2* YBr- COOH yH2*$Br CH2* CBr-COOH - - i HBr + CO,. CH,. C COOH That the constitution of this new acid is represented as above, and not is rendered very probable from an CH : YH CH2* CBr COOH' by the formula I956 W. H. PERBIS, JUN.: examination of the properties of the acid.Its most remarkable pro- perty is its stability ; when, for example, it is boiled for some time with strong potash (sp. gr. 1*3), or treated with silver oxide, it is not decomposed to any extent, and, on acidifying, the unchanged acid separates in crystals ; this fact can only be explained on the assump- tion that the bromine-atom is attached to doubly bound carbon: in bromomaleic acid, COOH* CH 1 CBr- COOH, the byornine-atom is not removed by boiling with baryta water (Carius, Annalen, 149, 264)) whereas in saturated bromo-acids, as, for example, in the case of bromosuccinic acid, COOH- CH,. CHEr- COOH, the bromine-atom is readily eliminated by boiling with alkalis. The tendency of dibromo- tetramethylenedicarboxylic acid to be converted into bromodi hydro- tetrenecarboxylic acid is verl- remarkable, this cbange taking place, with separation of iodine, when the aqueous solution of the dibromo- acid is warmed with potassium iodide, a change which it is difficult to represent by an equation.Bromodihydrotetrenecarboxylic mid melts at 121--122", and shows all the properties of an unsaturated acid ; thus, it readily decoiorises pennanganate solution, and combines with bromine to form tribromo- tetramethylenecarboxylic acid. When silver oxide acts on an aqueous solution of dibromotetra- metbylenedicarboxylic acid, broniodihydrotetrenecarboxylic acid is formed in considerable quantities, but the principal product of the action is a thick, syrupy acid, which, alt,hough it could not be ob- tained quite pure, is obviously dihydroxytetramethylenedicarboxylic acid, YH,*V(OH)*COOH CH2 C(0H) * COOH + 2AgBr.+ Ag,O + H,O = QHz* C;Br*COOH CH2* CBr*COOH The reaction is in fact very similar to the formation of bromofumazic acid and tartaric acid from isodibromosuccinic acid under the same conditions. YHBr. COOH gives CBr* , , COOH and $.!H(OH) *COOEI CHBr*COOH CH * COOH CH(OH)*COOR' The behaviour of methplic dibi*omotetramethylenedicarboxylatc towards alkalis is, on the other hand, quite different from that of the free acid ; when treated with alcoholic potash, for example, a curious action takes place ; since bromine, and not,, as was to be expected, hydrogen bromide is eliminated, dihydrotetrenedicarboxylic acid being formed thus.DERIVATIVES OF TETRAMETHYLENE.957 + Br, + 2CH3*OH. YH,*g*COOH CHz* C COOH + 2H20 = QH2* QBr*COOCH3 CH,-CBrv COOCH, I hare not been able to find a case exactly analogous to this, and illustrating the unstable nature of the bromine-atoms in this position, but i t may be mentioned that dimethylsuccinic acid %hi&: in constitution, closely resembles tetramethylenedicarboxylic acid, when treated with bromine and phosphorus, yields the anhj-- COOH. CK(CH3). CH(CH,)*C'OOH, cH3*E 'O>O ; the intermediate di- dride of pyrocinchonic acid CHs*C.CO bromodimethylsuccinic acid which is probably formed, in the first instance: losing the two bromine atoms (Zelinsby, Krapiwin Berichte, 22, 653). Dihydrotetrenedicarboxylic acid is sparingly soluble in cold water, melts a t I78", and when heated at 200" loses 1 mol.H20, and is converted into an anhydride sH2'$ 'co>O, a resinous substaiice CH,*C*CO which could not be obtained in a crystalline form, and which dissolves in water with formation of a crysta.lline acid of the same empirical formula as dih ydrotetrenedicarboxylic acid, from whicl I , however, it differs in a ver? marked manner, notably in being es- cessively soluble in water. Whether these two acids are stereoiso- meric or whether a change in the position of the double bond has taken place during the above treatment, has not yet been satisfac- torily determined, owing to the small amount of material available for investigation. The constitution of dihydrotetrenedicarboxrlic acid is proved in t h e following may. The methylic salt of this acid is formed quantitatively when methylic dibromotetramethylenedicarboxylate is digested in alcoholic solution with potassium iodide, iodine being liberated. + 1, + 2KB13.CH2 $ CO OCH, CH2.C .COOCRs + 2KI = 1 YH,*~Br*COOCH, CH,* CBr COOCH, It is a beautifully crystalline substance which melts at 46", and when exposed to bromine vaponr is reconverted into methylic dibromotetramethylenedicarbox~-late ; it must therefore have the constitution represented above ; dihydrotetrenedicarboxylic acid itself, on the other hand, may be exposed to bromine vapour for some days wit,hout change. QH2*yH2 CH,*CH*CO*NH, Am ide of Tetrame t?) y Zeir eca doxy 7 ic ctcic7, This has already been prepared b - Freund and Gudeman (Be)..,958 W. H. PERKIX, JUN.: 21, 2694), but not in a pure condition, as these chemists give the melting point 138", whereas tlie pure amide melts at 153".In preparing large quantities of this amide, the following method was generally employed. Pure tetramethylenecarboxylic chloride (10 grams) prepared as described in a previous paper (Trans., 1891, 59, 41) is added slowly through a dropping funnel to the strongest aqueous ammonia, (100 c.c.), the whole being well cooled during the operation. As each drop of the chloride comes in contact with the ammonia solution, a very vigorous action takes place, accompanied by volumes of white fumes, and, after about half has been added, the amide cfimmences to separate in glistening plates which ultimately quite fill the liquid. The crystals are collected by means of the pump, washed once or twice with strong ammonia solution, dried on a porous plate in the air, and, if necessary, recrystallised from ether, in which the substance is sparingly soluble.The amide is thus obtained in the form of magnificent, silky plates, which, on analysis, gave the following results.* Tlieorp. C,H,.CO.NH,. Found. C .......... 60.60 per cent. 60.75 per cent. N .......... 14.1.5 ,, 14.30 ,, H .......... 9-09 .. 9-30 ), The amide of tetramethylenecarboxylis acid melts at 152-153", and is very readily soluble in water and alcohol, but only sparingly i n ether; when heated, it sublimes in iridescent plates, which are very like sublimed benzoic acid in appearance ; i t is volatile even at loo", and caunot be dried in a water oven without loss. * Considerable difficulty mas experienced in determining the nitrogen in this sub- stance, as, when burnt in a stream of carbon dioxide in the inanner usually adopted, the result was alwajs much too high, in two cases the analysis indicating 20 per cent.of nitrogen. Obviously, during the coinbustion of this aniide, Tery stable gases (ethylene ?) are evolved, which pass away with the nitrogen into tlie eudiometer, and, in order to obviate this, an experiment was made, in which the tube --as kept a t a bright red lieat and the combustion conducted slowly, but eren then the result was 2 per cent, too high ; the above nitrogen determination, which gare correct numbers, was cai-ried out as f ollov J . The substance was digested n itli pire concentrated hydrochloric acid f C i * fire minutes, the solution then iiiiscd with water and boiled until the odour of tetramethylenecarbosylic acid was imperceptible ; the amount of aminoi iium chloride in the solution was then determined as ammonium platinochloride in the usual manner.In several othei. cases it has been noticed that closed chain compounds art: exceedingly J i f f i i ~ l t to anal>-se, and accurate results have only been obtained when tlie tube was kept Teryv hot, m c l tile coinbustion conducted throughout in a stream of oxygen.DERI T'XTIVES OF TETRAMETHPLENE. 959 It is very stable, as is illustrated by the fact that it was not entirely decomposed even after boiling for half an hour with strong potash, some of the unchanged amide separating in crystals on cooling ; it is, however, very readily hydrolysed by boiling concen- trated hydrochloric acid.I n preparing this amide in the manner described above, a consider- able quantity remains in the ammoniacal mother liquors, but most of this may be recovered by repeated extraction with ether ; any loss of valuable product is avoided by passing air through the aqueous solu- tion until the bulk of the ammonia has been expelled, and then hydrolysing the dissolved amidc hg boiling with hydrochloric acid. The regenerated tetramethylenecarboxylic acid can be extracted with ether in the usual manner, and used for a subsequent) operation. yH2* $lHz CH, CH NH2- Te f ramet Ii y 1 e I 1 ea I I 2 i.i 2 e, This interesting substance is formed when the pure amide of tetramethylenecarboxylic acid is treated with bromine and potash, according to the well-known method devised by Hofmann (Ber., 15, 762) ; the details of the preparation are as follows.Bromine (16.2 grams) is weighed o u t into a flask of about 500 C.C. capacity, and then the pure, finely-powdered amide (10 grams) is added in two or three portions with constant shaking, the whole gradually dissolving with evolution of very little heat.* At the end of two hours, potash (10 per cent.) is slowly added, the whole being well cooled during the operation; after a time, a reddish-brown, crystalline precipitate separates. On adding more alkali, very little rise of temperature takes place, but the preci- pitate becomes colourless, and doubtless consists of the bromamide of tetramethylenecarboxylic acid, C4Hi CO NHBr.On continuing to add strong potash (about 30-35 per cent.) in small quantities a t a time with constant shaking, the bromamide gradually passes into solution; as now much heat is developed, the flask must be cooled from time to time by plunging it into matel-; when all the crystals haTe disappeared, the whole is distilled in a curreiit of steam until the con- densed water is no longer alkaline, and the distillate is then rendered sliyhtly acid by the addition of hydrochloric acid, and evaporated to dryness. The residue, which consists of a mixture of tetramethylenearnine hydrochloride and ammonium chloride, is extracted with absolute alcohol, the extract filtered from undissolved ammonium chloride, eva- Freund and Gudeman (loc. cit.), using the impure nmicle, noticed the develop- ment of a considerable amoint of heat.960 w.H. PERKIN, JU?CT.: porated to dryness, and the extraction with alcohol repeated. The hydrochloride obtained on eraporating the second extract is sufficiently pure for use in the preparation of hydroxytetramethylene, &C. In order to isolate the base itself, the hydrochloride is transferred to a small retort, mixed with powdered potash, and distilled, the aqueous distillate is dehydrated with potash. and the oily base separated and fractioned ; almost the whole of it distils between 81" and 83", and, on refractionation, boils constantly at 82". On analysis the following numbers were obtained. Found. Theor-. rd--- 7 C'JLJ. I. 11. C .......... 6i.tiO pel. cent. 67-25 67.13 per cent.N 19-72 ,, 19.58 - H .......... 12-67 ,, 12.60 12-58 ,, .......... ?? Tetrametliyleneamine boils at &do, and possesses a very pungent basic odour somewhat similar to that of isoam;-lamine ; it absorbs carbon dioxide from the air fumes strongly in contact with hydrogen chloride, and mixes with water with development of heat. The hydrochloride, C,H,N,HCl, was prepared by neu tralising an aqueous solution of the base with hydrochloric acid, and evaporating to dry- ness. It crj-stallises from alcohol, in which it is readily soluble, in long, striated, prismatic needles, closely resembling crystals of ammonium chloride in appearance, and very readily soluble in water. The results of nnalJ-sis gave Theory. C4H9K,HEZC1. Found. Cl.. ...... 33.02 per cent. 33.24 per cent.On adding platinic chloride to a strong solution of the hydro- chloride, the plcztinochloricle, (C4H,N),,H2PtC16, is obtained as a deep yellom-, crystalline precipitate, which dissolves readily in hot water, and separates on cooling in groups of deep orange octahedra. For analysis, the salt mas dried at 100". Found. Theory. r--- -7 (C,H9K).:H2PtCI6. I. 11. Pt .......... 34-87 per cent. 34.79 34.82 per cent. This salt is rather sparingly soluble in cold water ; when heated in a, capillary tube, i t begins to darken at 200°, and becomes qnite black at 210-215". The hydrochloride of tet ramethyleneamine reacts readily with silver nitrite, evolving nitrogen and forming bydroxytetramethylenePERIVATIVES OF TETRAIIETHTLESE. '361 fortu~ately the reaction takes place with the formation of traces only of bye-products, and the yield of the hydroxy-compound obtained is, therefore, very good.The experiment was conducted as fol- lows :-Crude tetramethyle~eamine hydrochloride (20 grams), w hi& had been freed from ammonium chloride by treatment with alcohol, as explained on p. 959, was disvolved in water (80 c.c.), and to the nrell-cooled solution a slight excess of freshly precipitated silver nitrite (prepared from 40 grams of silver nitrate), added in small quantities at a time. Effervescence soon began, and after remaining for one hour at the ordinary temperature the whole was heated on a water bath for a few minutes, until the t-igorous evolution of gas had almost ceased. The product was then Jvell cooled, mixed with anhydrous potassium carbonate, and extracted fire times with pure- ether ; the ethereal solution was dried over anhydrous potassiuni carbonate, filtered, and the ether t-ery slowl>- distilled off ; a rectifying column being used so as to obviate, as far as possible, loss of the h y droxy -derivative by evaporation. The residual, almost colourless, oil (12 grams), after twice fractioning, boiled constaxitly at 123", and gave the following results on analj-sis. Found.Theory. r---J-- -7 C,H;.OH. I. 11. C . . . . . . 66-66 per cent. 66-52 66.41 per cent. H...... 11.11 ,, 11.10 11-23 ,, ITydroxytetramethylene is a colonrless oil, which smells like butylic alcohol ; it is readily soluble in water, and dissolves sodium with evolution of hydrogen and formation of a white sodium com- pound.Action of Hyd~obromic acid on Hydr.on.ytetranzethylel2e. Fotmatiou of (?H2'?H2 , a d of 1 : 3-D&brcnnobuta?ie, CH, C HBr Brom ot e tram ethylene, CH3*CHBr*CH2* CH&r. Having thus obtained bTdroxgtetramethylene. the next point was to investigate its behaviour with the halogen acids, in order, if possible, to prepare halogen derivatives of the hydrocarbon ; the first experiments instituted were with hydrobromic acid. Hydroxytetramethylene dissolves in fuming hydrobromic acid (satu- rated at about 10"),with slight development of heat ; as, however, little action appeared to take place in the cold, a large excess of the acid was added, and the whole heated to boiling in a flask, into the neck oE which a reflns condenser bad been ground.During the operation, :t heavy oil separated on the surface of the liquid ; hut as the action continued and the hjdrobi-omic acid became weak, the oil sank ; after962 W. IT. PERKIN, JUN.: heating for an hour, the product was poured into water the heavy, brownish oil extracted with ether, the ethereal solution well washed with water, dried over calcium chloride, and the ether evaporated. The residual oil was then repeatedly fractioned, and thus readily separated into two fractions, boiling at 102-104" and 17.3-174" (737 mm.) respectively; of these, the latter was present in by far the larger quantity. The substance boiling at 102-104" is evidently bromotetramethylene, as is shown by the following analyses. Found. Theory. r--- -7 C .I H,B r . I. 11. C ..........3355 per cent. 55-02 - per cent. H 5.06 - .......... 5-19 ,, 7, R r . ........ 59-26 ,, . 3 7 7 58-85 ,, Brornotetrarnethylcue is a coloudess oil, which boils at about 104" (760 mm.); it is specifically heavier than water, and has an odour which can scarcely be distinguished from that of isobntylic 01' iso- amplic bromide. The substance boiling at 1'73-174" (737 mm.) is a very heavy oil, possessing in a iiiarked degree the odour of trimethylene bromide ; z t gave, on analysis, the following numbers. Fonnd. Theory. r--- -7 C',H,Br,. I. 11. C .......... 22-22 per cent. -19.00 _- - per cent. Br ......... 74.07 .. 74.32 74-21 ,, - H .......... 3.71 .. 3-70 7, These analpes show that this substance must be a dibromobutane, and its formation proves that, under the conditions of the experiment, the tetramethylene ring in bromotetmmethylene has been split, addition of hydrogen bromide simultaneously taking place.As ex- plained in the introduction, this dibromohutane is different from any of the known isomerides ; and experiment proved that it is identical with the 1 : 3-dibromobutane, which is obtained when 1 : 3-dihj-drosy- butane, CH,* CH(0H) CH2*CH2- OH, is treated with Eydrobromic acid (see next section). Prepamtion, of 1 : 3-Dibi.un,c,bicta71e7 CH3*CHBr* CH,. CH2Br. This new snbstance was prepared in order to ascertain whether i t mas identical with the dibromobutane obtained by the action of hydrobromic acid on hydroxytetramethyleue, as described in the last section, and the method employed was briefly as follows.* * Compare Orndorff and Newbury, -3foitatshefte fur Chemie, 13,516.DERIVATIVES OF TETRAMETHYLENE.963 Aldehyde (200 grams) mas mixed with an equal volume of water, aiid then a strong solution of potassium carbonate (12 dgrams) added drop by drop, the whole being kept at 0" during the addition, which extended over about two hours ; after standing for two days at the ordinary temperature, the mixture was neutralised with hydrochloric acid, extracted three times with ether, and the ethereal solution evaporated. In order to convert the residue, which consisted of crude aldol (98 grams), into the correspondiug 1 : 3-dihydroxybutane, it was dissolved in water (1 litre), and redncecl in flat, porcelain basins with nearly twice the calculated quantity of 2$ per cent.sodium amalgam, the solution being kept neutral by the constant addition of small quanti- ties of dilute hydrochloric acid, a rapid stream of carbon dioxide being also passed through the liquid, to prevent the accidental accu- mulation of any large quantitp of alkali arising from rapid decom- position of the amalgam. During the reduction, a sticky substance separated, and the soln- tioii also acquired a very penetrating and irritating odour, due pos- sibly to the formation of crotonaldeliyde ; in order to remove these bye-products, the solution was filtered, and extracted with a small quantity cf ether. The filtrate was then slowly concentrated to about 250 C.C. in a flask connected with a long, rectifying column, and the residue saturated with hydrogen bromide, without cooling, loss being prevented by using a reflus apparatus.After 24 hours, the product was diluted with twice its volume of water, extracted five times with ether, the ethereal solution washed with dilute sodium carbonate, dried orer calcium chloride, and the ether distilled off; the residual dark brown oil was then twice frac- tionated, and yielded $0 grams of 1 : 3-dibromobutane boiling at 172-17.5", of which the greater portion passed over between 173" and 174". Analysis- I Found. Theory. r--- C4HsBr,. I. 11. P i ) . 4.3 52 73-67 per cent. BP . . . . . . . . . 74.07 per cent. The 1 : :3-dibromobutaiie obtained in this way boiled at the same temperature, and showed all the properties of the substance obtained by treating hydroxytetramethjlene with hydrobromic acid, so that there can be 30 doubt the two pi-oducts are identical.I have to thank JIessrs. G. H. Cross and J. N. Goldsmith, students in the Owens College, for help in carrying out these experiments.964 W. H. PERKIS, JUN.: YHa'YH2 CH2*CHC1' Chlorotetramethylene, This substance was prepared by the action of phosphorous penta- chloride at moderate temperatures on hydroxptetraniethylene, and not hy heating this hydroxy-derirative with hydrochloric acid, as i t appeared possible that, in the larter case, the chlorotetramethylene formed might be further acted on by the halogen acid with formation cf dichlorobutane (see previous section). Pure hydroxytetrametbylene (11 grams) was placed in a flask fitted with a reflux condenser, and pure phosphorous pentachloride (35 grams) gradually introduced in small portions a t n time ; each successive quantity being allowed to entirely disappear before any more was added; the action, which is very vigorous and accom- panied by the evolution of torrents of hydrogen chloride, was mod- erated by cooling the flask with ice-cold water ; at the close, however, the temperature was allowed to rise to 60'.The product was slowly fractioned, everything which passed over below 105" being collected,* the distillate was agitated with water t o decompose the phosphorous oxychloride which i t contained, and the supernatant oily layer separated, dried over calcium chloride and fractionated; almost the whole of it (11.5 grams) passed over between 83" and 87". After refractionation, a colourless oil boiling constantly a t S5" and specifically lighter than water, was obtsined.On analysis it gare t h e following results. Theory. C,HiC1. Found. C ........ 53.10 per cent. 53.07 per cent. c 1 ........ 39.16 ,, 38-73 ,, H ........ 7-74 ,, 7.53 ,, Chlorotetramethylene possecses a not disagreeable odour. very similar to that of isonmylic chloride. Tn order to prepare this substance, chlorotetramethylene (8 grams) was heated with finely powdered potassium iodide (18 grams) and metliylic alcohol (20 c.c.) in a sealed tube a t 120-125" for five hours ; the product, which was discoloured by iodine, was mixed with water, and the heavy oil which separated extracted with ether ; the ethereal solution, washed with dilute sodium hydrogen sulphite, was dried orer calcium chloride, evaporated, and the residue carefully fractioned.* On decomposing the residue xith mt-tter. a ver- small quantitp of oil q x w d e s ; this was neglected.DERIVATIVES OF TETRANETHYLENE. 965 T t was theu found that the oil contained about 50 per cent. of nn- chmged chloroteti-amethylene ; after this had passed over, the ther- mometer rose rapidly to 1;30°, bet*ween which temperature and '140" t h e remainder distilled as a colourless oil; on refractionat.ion, it boiled almost constantly at IS", and gave the following results on a ~ a l y si s . Theory. C,HTT. Found. C ........ 26.37 per cent. 26-85 per cent. H ........ 3-65 ,, 4.05 ,, I ......... 69.78 .. 69-22 ,, Todotetramethylene is a colourless, licavy oil, which i n its proper- ties shows the greatest resemblance t o isobutylic or isoamylic iodide ; its odour can scarcely be distinguished from that of the latter, and, when exposed to light, it quickly t u r n s brown owing to separa- tion of iodine.It is readily reduced by zinc dust and hydrochloric acid with fmnzation of a very volatile hydrocarbon, which is prob- ably tetramethylene ; unfortunately the amount of material at my disposal was too small to allow of the systematic examination of this react ion. ActiolL of Q2iinoli)l.e on Iodotetranzethy 1eue.--This action (compare Eaeyer, Aszizaleii, 278, 107) was investigated with the view of ob- tainin g the unsaturated hydrocarbon clihydro te trene, explained in the introduction, but the results were not very satisfactory. Iodotetramethylene (8 grams) was mixed with pure quinoline (15 grams) and the whole heated iu a sealed tube at 180" for two hours ; the product was then distilled from the tube, every precaution being taken to avoid loss of the volatile hydrocarbon by cooling the condenser and the receiver with a freezing mixture.In this way, a small quantity of a colourless oil was obtained which boiled between i30" and 45", and contained halogen ; after- heating in a small sealed tube with sodium and refractioning, the greater portion distilled between 35" and 35": but it was not quite free from halogen. 'This substance is instantly oxidised by potassium permanganate and decolorises bromine ; on analysis it gax-e numbers agreeing only ap- proximately with the formula C1H6, owing not only to the presence of traces of halogen, but' especially to the great difficulties attending the analysis of a hydrocarbon of this low boiling point.pp as CH2 CH' This acid is produced wlien the anhydride of tetramethylene- dicarboxylic acid (this vol., 581) is treated with a large exces966 W. H. PERKIN, Ju?\'.: of bromine and amorphous phosphorus, according to the Hell- Volh ard - Zelinsky method. Tetramethy lenedicarboxylic anhydride (10 grams) is ground up in a mortar with amorphous phosphorus ( 3 grams), transferred to a flask, into the neck of which a condensing tube has been ground, 2nd dry bromine (70 grams) then gi*adually added, the very vigorous action which takes place being moderated from time to time by cool- ing with water. On heating the mixture on a water bath, torrents of hydrogen bromide are erol-ced during the first hour; the action then slackens considerably, and a t the end of six hours, a second quantity of phosphorus ( 3 grams) and bromine (70 grams) is added, and the beating ccntbned for about 15 hours; air is then blown through the mass in order to remove escess of bromine, and the dark- coloured liquid residue, which on cooling frequently deposits crystals of phosphorus pentabromide, is decomposed cautiously by means of ice- cold water.After standing some time, the thick, dark brown product is extracted with ether, the ethereal solution, vell washed with water containing a little sodium hSdrogen sulphite, dried over calcium chloride, the ether distilled OB, and the residual oil fractionated under reduced pressure (70 mm ).The mass froths a good deal a t first, and R considerable quaiitit? of water distils orer ; the thermometer then rises rapidly, the greater portion distilling between 185 and 190" as an almost. coloui-less oil, which solidifies in the receirer ; a good deal of a dark brown residue is, however, left behind in the distilling flask. The distillate is ground up and digested with sufficient hydrobromic acid (sp. gr. 1-49) to dissolve it, the solution filtered through glass wool and allowed to cool slowly; the mass of glistening crptals which separates is collected, washed with hydrobromic acid, drained on a porous pinte, and recrystallised from the same solvent. The purified substance, after drying first over sticks of potash in a vacuum, and then at loo", gave the following results on analysis.Found. 7- -7 Theory. I. 11. CBHgBr20,. C ...... 24.10 24.05 per cent. 23.84 per cent. 33.. .... 2.10 2.20 .. 1.99 ,, > 7 52.98 ,, Br.. 52.71 - .... When rapidly heated in a capillary tube, dibromotetramethylece- dicarboxylic acid melts at about 202-205" with evolution of gas. probably due t o the formation of its anhydride ; it dissolves readily in alcohol or ether, and also in boiling hydrobromic acid, but is only sparingly soluble in the latter solvent in the cold. It is remarkableDERIVATIVES OF TETRAMETHTLENE. 961 that it is readily soluble in meter ; a strong, hot aqueous solution, on cooling, deposits the substance in beautiful glistening plates. A moderately concentrated aqueous solution of this acid gives, with silver nitrate, a copious crystalhe precipitate of a silver salt, which contains no silver bromide, as it is completely dissolved by dilute nitric acid.On boiling, however, decomposition sets in with separation of sil-rer bromide in abundance. $!€€,*~Br*COOCH,. CH2*CBi** COOCH, N c t h ylic dibromotetrameth~~enedicarboxylntr, Considerable quantities of this methylic salt were necessary for many of the experiments described in this paper, and in the first instance i t wzs prepared from the pure dibromo-acid by treating it with methylic alcohol and hydrogen chloride in the usual way ; ulti- mately, however, it was found that the best method of preparation mas the follotviag. The crude product of the action of bromine and phosphorus on tetramethylenedicarboxylic anhydride (from which the excess of bromine had been removed by blowing in air, as described on p.966), was poured in a thin stream, and without cooling, into a large excess of methylic alcohol ; after the vigorous action had subsided, and the product cooled down, water was added, and the methylic salt extracted twice with ether. The ethereal solution was well mashed with dilute sodium carbonate, dried over calcium chloride, the ether evaporated, and the residual dark brown oil fractionally distilled under reduced pressure (50 mm.). With the exception of a small quantity of a substance of low boiling point, nearly the ~-1iole of the oil distlilled between 180" and 205", undergoing slight decomposition and e~olving some hydrogen bromide ; on redistilling, two principal fractions were obtained boiling at 185-195" and 1Y5-!2OUo (50 mm.) respectirely.Both of these, after 14 days, deposited hard, prismatic crystals, which were separated from the oil by filtration through glass wool, left in contact with porous porcelain until quite d r y and colourless, ground up, and then recrystallised twice from boiling light petroleum (b. p. 60-70") ; in this way the methylic salt was obtained in magni- ficent, colonrless prisms, which on analysis gave the following result. Theor)-. 29.09 per cent. Found. C,H,B1*, (COOCHJ,. C .. . . . . H ...... 3.30 ,, 3-03 ,, Br . . . . . 48.48 ,, 48.48 ,, 29-21 per cent. Methylic dibromotetramethylenedicarboxylate melts at 88-&9"968 W. E. PERKIN, JUN.: and is readily soluble in alcohol, chloroform, benzene, and hot, light petroleum, but only sparingly so in the latter solvent in the cold.It separat,es from dilute methylic alcohol i n colourless groups of glistening, feathery crystals. The oily mother liquor which was separated from the crystals, as explained above, was refractioned, and in this way a further, though small, quantity of crystalline methylic salt was obtained. After sepa- i%ting this. and again refractioning the mother liquor, no fiirther crystallisation took place, even after the oil had been lelt for months in contact with a crystal of the methylic salt ; analyses of two samples of this oil boiling at 190-195" and 295--200" (50 mm.) respectively, gave the following results. Found. F- Theory. I. 11. C4H4Br2 (COOCH,),. Br... . . . 46-50 47.10 per cent. 48.48 per cent. It is possible that this product contains some impurity which retards crystallisation. As the analysis, howerer, seems to show that it can contain, at the most, only traces of such impurity, and as this would hardly prevent the separation of a substance charac terised, like the methylic salt,, by great facility for crystallising, it is possible that the liquid methylic salt is ti stereoisomeride of the crystalline salt. That there is a considerable difference between the solid and liquid methylic salt is showu clearly in their behaviour with potash (see p. 973). Anhydride of Uibrornotetl.amethylenedical.bosylic acid, YH,*YBr*CO CH,*CBr*CO >o- I t has been shown in a previous commiinication (t,his vol., 572) that tetrarnethylenedicarboxylic acid [ 1 : 21 exists in two well-defined cis- and trans-modifications, and it might be supposed that its dibromo- derivative would show the same behaviour.In order to investigate this point, it was necessary, in the first place, to prepare the anhy- dride of this acid, which was ultimately nccomFlished i n the following way. Nearly pure di bromotetramethylenedicarboxylic acid (m. p. 185-19O"j was distilled under a pressure of 40 mm., when almost the whole of it passed over at lS0--183", aud solidified on cooling to a hard crystalline cake ; this was dissolved in freshly distilled acetic anhydride, the solution boiled for five minutes i n a reflux apparatus, and then placed over potash in a vacuum dessicatw. As the excess of acetic anhydride evztporated, beautiful, colourless, prismaticDERIVATIVES OF TETRAMETHYLENE. 969 crystals separated ; these were collected, washed with acetic anhydride, and dried over potash aad sulphuric acid in a vacuum; they gave the following result on analysis.Theory. 25.31 per cent. Found. C,H,Br,O,. C . . . . . . H .... .. 1-72 ,, 1.41 ,, Br ... .. 56.43 ,, 56-33 ,, 25.21 per cent. Dibromotetrametbylenedicarboxylic anhydride melts at 103 -104", and is readily soluble in alcohol, ether, benzene, and boiling light petroleum (b. p. ZOO'), but only sparingIy in the latter in the cold. I n contact with moisture, it is readily converted into the acid from which it was derived ; some of the powdered substance which was left exposed to the air for a few days had almost entirely changed, and then melted at about 280".The anhydride dissolves readily in warm water, and the solution on cooling deposits pure dibrornotetramebhylenedicarboxylic acid in glistening crystals (m. p. 204-205"). This acid is therefore the cis-modification. In order to determine whether the corresponding trans-modi fication existed, some of the pure anhydride was dissolved in warm hydro- bromic acid solution (sp. gr. 1.49) and heated in a sealed tnbe at 190-200" for two hours. On cooling, the tube was found to be filled with glistening cryvstals, very little discoloration and no charring having taken place ; these crystals were collected and re- crystailised from hydrobromic acid ; they melted then at about 200", and were exactly similar in appearance to the original &-acid. It appears, therefore, that the ti-ans-modification of dibromotetra- methylenedicarboxylic acid does not exist: unless indeed it is so similar in properties to the cis form that it could only be identified on very prolonged investigation.Act io,i. of Alkalis on Dibrom otetrai,zethyl~zedicarbox yb ic acid. YH2* GBr CH,*C COOH' B~omodihydmtet~renecarboxylic acid, As explained in the introdnction, these experiments were instituted C H : 7 C OOH with the object of obtaining tetrenedicarboxplic acid, I CH: C-COOH; but, although the conditions were varied its far as possible, this acid could never be isolated; the reaction proceeded in all cases in the same way, carbon dioxide and hydrogen bromide being eliminated, and bromodihgdrotetrenecarboxylic acid formed, thus : $lH,*yBr*COOH yH2*GBr CH,*CBr*COOH - CH,* C'COOH + CO, f HBr.- VOL. LXV. 3 2970 W. H. PEREIN, JUN. : 1. Experiments with Barium Hydroxide. - Pure dibromotetra- methylenedicarboxylic acid (about 5 grams) was dissolved in a little warm water and the solution mixed with a large excess of hot, strong barium hydroxide ; this caused the separation of a white, glistening, crystalline precipitate which was very sparingly soluble even in boiling water, and probably consisted of the barium salt of the dibromo-acid ; on continued boiling, however, this was gradually de- composed and passed into solution, much hydrogen bromide being eliminated, as was proved by testing the solution with silver nitrate. After filtering from a small quantity of insohble matter, carbon dioxide was passed through the boiling liquid to precipitate the excess of baryta, and the whole filtered hot; the filtrate, on being evaporated to a small bulk and allowed to coo1, deposited colourless crystals which, after recrystallisation from water and drSing a t lUO", gave the following results on analysis.Found. --A- -7 Theory. I. 11. C,,Hg Br?OlBai. Ba.. .... 28.03 28-02 per cent. 28.02 per cent. ..... 9 , 32-72 ,, Br* 33.01 - On dr+g at loo", the air-dry glistening crystals become opaque and lose 9.77 per cent. in weight, which appears to show that the freshly prepared salt has the composition, C,,H,Br,O,Ba + 3H20 (containing 9.96 per cent. H20). This water of crystallisation is gradually given off, when the crystals are lefl.exposed over sulphuric acid in a dessicator for a long time. The barium salt, is sparingly soluble in cold, but readily in hot water : on adding hydrochloi-ic acid to the hot solution, and allowing it to cool slowly, bromodihydrotetrenecarboxylic acid separates in colourless needles which, after reci=ystallising from water, gave the following results on analysis. Theory. Found. C,H,Br.COOH. C ........ 34.20 per cent. 33.90 per cent. H ........ 2-97 ,. 2.83 ,, Br.... .... 45.23 ,, 45.18 ,, The acid melts at 121-122", and is sparingly soluble in cold water, readily in hot water, chloroform, alcohol, ether, benzene, and hot light petroleum; it is best crystallised from hot water or light petroleum. That if is an unsaturated acid is shown firstly by the behaviour of the cold solution of its sodium salt with a drop of permanganate, which it instantaneously decolorises, and, secondlyr, by * I n order to determine the bromine in this salt, it was found necessary to heat the substance with fuming niLric acid and silver nitrate for 5 hours at 200-220'.DERIVATIVES OF TETRAMETHYLENE.971 its conversion into tribromotetramethylenecnsboxylic acid when treated with bromine. The bromine-atom in bromodihydrotetrenecar boxylic acid is very firmly bound, and is only very slowly eliminated by boiling with silver hydroxide, potash, quinoline, and other alkalies ; the acid dis- solves also in fuming nitric acid, and the solution may be warmed almost to the boiling point without elimination of bromine ; ultimately, however, vigorous oxidation sets in, and the acid is rapidly decom- posed. Action of Potash on Dibromotetramethylenedicarbox~lic acid.--Tf this acid be digested with excess of potash instead of with baryta, and the solution acidified and extracted with ether, a yellow, semi-solid acid is obtained on evaporating the ethereal solution.If left in con- tact with porous porcelain, this becomes qnite hard and almost colonr- less, and, after recrystallisation from water with the aid of animal charcoal, yields colourless crystals of bromodihpdrotetrenecarboxylic acid. It melted at 122", and on analysis gave the following result. Ponnd. Theory. CjHHjBr02. Br.. .. . . . . 45.32 per cent. 45.18 per cent. Many experiment,s were then made with the object of eliminating hydrogen bromide from this monobromo-acid by boiling with aqueous and alcoholic solutions of potash of various degrees of concentration, $?H:$?H CH: C-COOH' but in no case could the desired tetrenecarboxylic acid, be isolated.Action of Quinoline and of Dimethylaniline on Dibromoteframethylene- dicarboxylic Anhydride.-When this anhydride is gently heated with quinoline, a vigorous action sets in, and a good deal of charring takes place ; on examining the product, the only crystalline substance which could be isolated was found to be bromodihydrotetrenecarb- oxylic acid melting at 122", showing that the quinoline had acted much in the same way as the baryta water and the potash, 1 mol. of hydrogen bromide only being eliminated. A similar result was obtained with dimethylaniline, but the action proceeded much more smoothly, and with little charring ; 0.5 gram of the pure anhydride was dissolved in 10 grams of pure dimethyl- aniline, and the solution heated gently until the vigorous action which soon sets in had subsided ; the temperature was then raised to the boiling point of the liquid for a few seconds.The product was then digested with alcoholic potash for 10 minutes, the dimethylaniline removed by distillation in steam, the alkaline solution filtered, acidifled and extracted with ether ; the residue left on distilling the ether gave colourless needles, which when crystallieed 3 Z 2972 m. H. PERKIY, JUS.: from water melted at 122", and showed all the properties of bromo- dihydrotetrenecarboxylic acid. Actwn of Silver Oxide on Dibromotetramethylenedicarboxylic acid.- As it was not found possible to remove all the bromine from this acid by means of baryta or potash, experiments on the action of silver oxide were instituted.About 5 grams of the pure dibromo-acid was dissolved in half a litre of water, and freshly prepared silver oxide was added in small quantities to the solution heated nearly to boiling, until, after stand- ing for some time and filtering, the solution was found to contain a considerable quantity of silver on testing it with hydrochloric acid ; when this point was reached it mas found to be disadvantageous to warm any further, otherwise the organic silver salt present decom- poses with deposition of silver. A stream of hydrogen sulphide was then passed through the warm filbered liquid until the whole of the silver had been precipitated, and the clear solution was evaporated to a small bulk and allowed to stand over sulphuric acid in a vacuum.After two days, the crystalline precipitate which had separated was collected and recrystallised from water ; it then melted at 122" and consisted of bromodihydrotetrenecrboxylic acid, as the following analysis shows. Theory. 45.18 per cent. Fomd. C,H,Br.C 0 OH. Br . .. . . . .. 45.41 per cent. The dark coloured filtrate from these crystals was digested with carefully purified animal charcoal and allowed to evaporate t o dryness over sulphuric acid in a vacuum, the resulting syrupy mass was then rubbed up with a little water, a few crystals which remained undis- solved removed by filtration, and the filtrate again allowed to evapo- rate, the treatment with cold water being repeated until the solution contained only traces of bromine.The pale yellow syrup ultimately obtained on evaporation showed no signs of crystallising even after standing in a dessicator for a fortnight ; when heated at loo", it lost water and yielded a brittle, transparelit resin which on analysis gave the following results. Found. * I. 11. Theory. CH,.C(OH) *COOH ~H,-~(OH)-COOH C . . . . . . . . 40.41 40.46 per cent. 40.91 per cent. H.. .. .. .. 4-34 4.51 ,, 4.55 ,) As these analyses were carried out with the product from two dis- tinct experiments, it seems probable that this substance is slightly impure dihydroxytetramethylenedicnrboxylic acid, the formation of which was to be expected from the conditions of the experimeut :DERIVATIVES OF TETRAMETHYLENE.973 and the behaviour of the acid favours this view. When heated, it readily chars, giving off a strong smell of burnt sugar; sulphuric acid aIso quickly chars the acid when the mixture is gently wazmed ; the acid is very readily soluble in water and reduces ammoniacal silver solution rapidly at 80-100", i n these respects behaving very like tartaric acid. On the other hand, it does not appear to form a sparingly soluble potassium salt. Action qf Potash on liquid Methylic Dibronaoteti.arnethylenecEica,.- boxy1afe.-A quantity of this substance which boiled constantly a t 190-195" (50 mm.) was digested with alcoholic potash, the product diluted with water, evaporated till free from alcohol, acidified and extracted with ether.The residue left on distilling of€ the ether, when repeatedly recrptallised from water with the aid of animal charcoal, yielded considerable quantities of bromodihydrotetrenecar- boxylic acid, also a mixture of acids melting about 150°, which mere not further investigated, and traces only of dihydrotetrendicarboxylic acid. This difference in the behaviour of the liquid and solid methylic salt towards potash is remarkable, and shows that there must be a considerable difference in their constitutions. CH, QBrz CH2*CBr*COOH' a&- Trib romoiet ramet h yleneznrboxy lie acid, I A solution of bromodihydrotetrenecarboxylic acid in chloroform is only very slowly attached by bromine. Addition, howerer, readily takes place if the fhely divided acid is left under a bell-jar in contact with dry* bromine vapour, the reaction proceeding quantitatively as is shown by the following experiment.0.1074 gram of substance left in contact with dry bromine vapour daring 24 hours yielded a red liquid, which, after standing over potash in a vacuum dessicator until the excess of bromine had evaporated, became quite hard and almost colonrless, and weighed 0.2044 gram, an increase of 04970 gram. On the assumption that addition takes place according to the equation the increase should have been 0.0971 gram. The product vi-as analysed by heating it with fuming nitric acid and silver nitrate a t 210" for 5 hours, and gave the following results agreeing with the formula of tribromotetramethylenecarboxylic acid * If the bromine vapour be moist, oxidation sets in, and an uninviting sticky substance is formed.974 W.H. PERKIN, JON.: Theory. Found. C4H4Br3- COOH. 71.21 per cent. Br.. . . . . .. 70.96 per cent. Tribromotetramethylenecarboxylic acid is readily soluble in benzene and alcohol, sparingly in water and light petroleum ; but attempts to recrystallise the substance were unsuccessful, as it readily decom- poses when warmed with solvents. YHz- #*COOCH3 CH, C GO 0 C €I3' Meth ylic dihydyo tetrenedicarboxylate, This beautiful substance is formed when methylic dibromotetrs- methylenedicarboxylate is digested in alcoholic solution with potas- sium iodide, thus VH2 FBr*GOOCH, yHz*g*COOCH3 CHz*CBr*COOCH3 CHz.C.COOCH3 + 2KI = + 2KBr + I,. Pure methylic dibromotetramethylenedicarboxylate (2 grams) is dissolved in absolnte alcohol (20 c.c.) finely powdered potassium iodide (4 grams) adcieci, and the mixture heated to boiling in a flask connected with a reflux apparatus ; the solution very soon begins to darken in colour owing to the separation oE iodine, and after boiling for 2 hours the decomposition is complete.In order to isolate the product of the action, water is added, the dark oily mass extracted four times with ether, the ethereal solution washed with dilute sodium hydrogen sulphite until colourless, and then with water; after drying over calcium chloride, and evaporating, a thick colonr- less oil is obtained, which, on cooling, solidifies to a mass of crystals. The crystalline mass after being left in contact with porous porcelain until quite free from oily mother liquor, was recrystallised twice from light petroleum (b.p. 50-60"), when magnificent glisten- ing crystals were obtained. On analysis, these ga.ve the following results. Found. Theory. I. Ir: CaH,(COOCH3)2. C . . .. . . . . 56.14 56.21 per cent. 56.47 per cent. H.. .. .. .. 5-93 5.95 ,, 5-88 ,, Methylic dihydrotetrenedicurbo.rylate melts at 44-46" and is readily soluble in methylic or ethylic alcohols, ether, benzene, and hot light petroleum, but only sparingly in the latter solvent in the cold; it crystallises from most solvents with great facility. A small quantity of the finely powdered substance was exposed under a bell glass to the vapour of dry bromine for 15 hours ; aFter removal of the excess of bromine over potash in a vacuiim, a colour- less mass was left which crystallised in needles melting at 84-86",DERITATrCTES OF TETRAMETHTLEhJIZ. 975 and showed all the properties of methylic dibromotetramethylene- dicarboxylate ; i t was doubtless identical with the latter, but no analysis of the product could be made owing to the small amount of substance available.?Ha. fi COOH CH2* C COOH' Dih ydrotetrenedicai-boxyl ic acid, h order to prepare this acid, the pure methylic salt was digested for 10 minutes with an excess of alcoholic potash, the solution of the potassium salt mixed with water, evaporated until free from alcohol, acidified and extracted at least 10 times with ether; the ethereal solution was then dried by calcium chloride, evaporated, and the residual almost colourless acid purified by recrystallisation from water. Andy sis Found.Theory. I. I I. C,H,(COOH)p C ...... 50.49 -50.60 per cent. 50.70 per cent. H...... 4.15 4-25 7, 4.23 7 , D ihydrotetrenedicarboxylic acid crystallises from wat,er in colour- less needles which melt and decompose at about 178" forming the anhydride. It is readily soluble in hot water and alcohol, sparingly in cold water, benzene, or ether, and almost insoluble in light petro- leum. It dissolves readily in dilute sodium carbonate, and the solution decolorises permanganate very rapidly, a proof that the acid is un- saturated. When exposed to dry bromine vapour, i t is either not acted on at all or action takes place with extreme slowness, which is mther remarkable as the methylic salt under the same conditions is readily converted into methylic dibromotetramethylenedicarboxylate. I f a little of the acid be carefully heated in a test tube, water is given off at first, and the residue then rapidly decomposes, a carbonaceous mass being left: in this respect the acid differs widely from com- pounds of analogous constitution, such as fumnric acid or pyro- cinchonic anhydride which distil without decomposition.Silver Ralt C,H,(COOAg),.-This salt is obtained as a white amorphous precipitate on adding a large excess of silver nitrate to a warm, slightly alkaline, solution of the ammonium salt; it was collected, washed with warm water, dried over sulphuric acid in a vacuum and then at, 100": and analysed with Found. C ........ 20.22 per cent. H . ....... 0.95 ,, Ag ...... 60.32 ,, the following result. Theory. Cb-H,~%%O,. 20.22 per cent. 60.67 ,, 1.12 ,,9?6 W. H. PERKLV, JUN.: This siher salt is very sparingly soluble in water ; when heated it decomposes all at once, yielding a very voluminous mass of silver, for which reason the above silver determination had to be carried out in the wet way. €Qdrogeiz Silver Salt, COOH C4H4* CO0Ag.-The warm mother liquors from the above salt, after two days, deposited magnificent,, glistening needles of the hydrogen silver salt ; t'hese mere collected, washed with water, dried over sulphuric acid in a vacuum and then at loo", and analysecl. Found. Y - - 3 I. 11. Theory. C,H&O,. Ag . . . . 42-94 43.12 per cent. 43.35 per cent. 1 t is remarkable that the corresponding dihydropentenedicarb- oxylic acid (p.9&:3) should give a similarly constituted hydrogen silver salt, and that both salts, when they have once crystallised, should be so sparingly soluble in water. Dihydrotetrenedicarboxylic acid is also produced when met hylic dibromotetramethylenedicarboxylate (p. 967) is treated with alcoholic potash, the action in this case being quite different from that which takes place when dibromotet,ramethylenedicarboxylic acid itself is heated with alkalis. The pure methFlic salt (5 grams) was dissolved in a little boiling alcohol, and a strong solution of potash (7 grams) in alcohol gradually added ; a vigorous action took place, and in a short time the whole selidified to a crystalline mass. On subsequently warming on a water bath, most of the crystals gradually disappeared, and the colour of the liquid changed to dark brown ; after two hours' digestion in a reflux apparatus, water was added, the solution evaporated uotil free from alcohol, filtered, acidified, and extracted 10 times with ether ; the ethereal solution was then dried over calcium chloride, filtered, and the ether slowly distilled off.During the distillation, and after the bulk of the ether had distilled over, crystals separated ; these were collected, washed with a litt,le ether to remove the dark-coloured mother liquor, and purified by recrystallisation from water with the aid of animal char- coal. The colourless crystals thus obtained melted at 178", and con- sisted of pure dihydrotetrenedicarboxylic acid, as the following analysis shows.Theor?. 59-70 per cent. Found. C,H4 (COOH),. C . . . . . . . . 50.35 per cent. H . . . . . . . . 4-31 ,, 4.23 ,,DERLVATrVES OF TETRAYETHTLENE. 977 This reaction does not afford a convenient means of preparing this acid, as the yield is not good, and the product is much more difficult to purify than that obtained from the pure metbylic salt, as previously explained (p. 975). >o. yH2 $ CO CH, C CO Dih y drotet renedica d o x y lie h h y d r i d e , In the first experiments made with the object of preparing this anhydride, the pure acid was digested with acetyl chloride and with acetic anhydride for one hour endeavours to purify the dark-coloured product by distillation under reduced pressure (20 mm.) mere fruit- less, because as soon as the excess of acetic anhydride or acetyl chloride had passed over and the temperature rose above 150", rapid decomposition set in, and the whole mass became quite black.When, however, the pure acid was heat,ed in a test tube at 190-200", 6 g 0 - rous effervescence took place, water was eliminated, and the action was complete in five minutes; the product, which on cooling set to a hard, transparent resin, was analysed, two distinct preparations giving the following numbers. Found. r-- 7 Th eoq- I. 11. C,H,CZ03. C ...... 57.47 57-55. per cent. 58.06 per cent. H...... 3-29 l3.27 ,, 3.23 ,, This substance is obviously the anhydride of dihydrotetrenedicarb- oxylic acid. It is a resinons substance, which melts in the steam oven, and could not be obtained in a crystalline €orm; when boiled with water, it melts and gradually dissolves, forming a clear solution, and this, if allowed to evaporate over sulphuric acid in a vacuum leaves a syrupy residue which, on long standing, deposits crystals. In contact with porous porcelain, these crystals lose the adherent, syrupy mother liquor, and become quite colourlees, but all attempts to recrystallise the substance failed ; two distinct preparations were dried at loo", and analjsed with the following results. Found. 7-- 7 Theory. I. 11. C , H, (COOH),. C ...... 50.63 50.59 per cent. 50-iO per cent. H.. .... 4-10 4.25 ,, 4.23 ,, This acid is, therefore, isomeric with dihydroietrenedicarboxylic acid (m. p. l i 8 O ) , and it differs from this in n most marked manner; it is, for example, excessively soluble in water and ether, and cannot be recrystallised, whereas the acid of melting poiut 178" is sparingly solub!e in these solrents, and is remarkable for the great978 E. EAWORTH AND W. H. FERKIN, JUN.: facility with which it crystallises. Unfortunately the small quantity of material at my disposal made it impossible for me to further investigate this substance. Action of Potassitim Iodide on Dibromotet,.amethytenedicarboxylic acid. Formation of Bromodihydrotetrenecarhoxylic acid. The formation of methylic dihydrotetrenedicarboxylate by the action of potassium iodide on methylic di bromotetramethylenedicarb- oxylate made i t interesting to determine whether dihydro tetrenedi- carboxylic acid itself might be produced by treating the dibromo- acid with potassium iodide. The pure dibromo-acid (2 grams) was mixed with a strong solu- tion of potassium iodide (10 grams) and allowed to stand, when it was noticed t h a t even in the cold the solution gradually acquired a yellow tint, due to liberation of iodine; at loo", i n a sealed tube, decomposition took place rapidly, and after two hours the liquid had become dark brown. On allowing the tube to remain over night in a cool place, crystals were deposited, and when the tube was opened a slight pressure was noticeable, due to carbon dioxide. The crystals were collected, washed with water, and recrys tallised several times from this solvent, when they became colourless. The substance melted at 122", and, ou analysis, was found to consist of bromodi- hydrotetrenecarboxylic acid. Found. Theory. Br . . . . . . . . 44.95 per cent. 45-18 per cent. It is remarkable that there should be such a difference between the beh aviour of di bromote t rame th y lenedicarboxylic acid and that of its methylic salt, when treated with potassium iodide. The Owens College, Nmchester.

ISSN:0368-1645    

DOI:10.1039/CT8946500950

出版商:RSC    

年代:1894

数据来源: RSC

摘要:

978 E. EAWORTH AND W. H. FERKIN, JUN.: LXXVIII. -1 : 2- Pentamethylenedicarboz~lic Acid. By E. HAWORTH, B.Sc., and W, H. PERKIS, jun. THE anhydride of &s-pentametbylenedicarboxylic acid behaves towards bromine, in the presence of phosphorus, in a manner similar to the anhydride of cis-tetramethylenedicarboxylic acid (see preceding paper), the principal product of the action being dibromopentamethyl- enedicarbosy 1 ic acid, CH,* FBr*COOH + 2Br2 = CHs<CH2*CBr.COOH + C Hz. C OOH CHz< C H2 CH COOH 'LHBr.1 : !%PENTBMETHPLENEDICARBOXYLIC ACID. 979 but the yield of this acid is comparatively small, as much carbon- aceous matter is produced during the bromination and during the subsequent purification of the substance by recrystallisation from hy d robromic acid. This tendency to decompose into carbon and hydrogen bromide, when heated with bromine, has been previously observed in the crtse of pentamethylenedicarboxylic acid (Trans., 1887, 51, 247), and may possibly be characteristic of penlamethylene derivatives ; in the ex- periment referred t o here, the pure bibasic acid was heated with bromine water at 180' for four hours, at the end of which time it had completely decomposed into carbon and hydrogen bromide.Dibromopentamethylenedicarboxylic acid is a colourless, crystalline substance, which melts at 183-184" ; i t cannot be distilled, and all attempts to obtain an anhydride with the aid of acetyl cbloride, $c., failed, owing to the instability of the acid ; when treated with alkalis, it is readily decomposed with formation of bromodihydropentenecarb- oxylic acid, thus- CH,*RBr + HBr + GO, C H, * YBr C OOH - CHz<CH,*CBr*COOH - CHz<CHz-C*COOH just as under similar conditions dibromotetramethylenedicmboxylic acid yields bromodihydrotetrenecarboxylic acid.This new acid crystallises from water in needles, melts at about 130", and, i n contact with bromine vapour, is rapidly converted into tribromupentamethylenecarboxylic acid, Dihydropentenedicarboxylic acid is produced when methylic di- bromopentamethylenedicarboxylate is digested with potassium iodide in alcoholic solution, and the product hydrolgsed with potash. + I, + 2KBr. CHz g COOE c H ~ < ~ ~ , - ~ + + 2Hz0 = CH2 2 COOCH, CH2<CH2-C*COOCH3 2CH3*OE. I t crystallises in colourless prisms, melts at 178", and is much less readily converted into an anhydride than the corresponding dihydro- tetrenedicarboxylic acid, from which it also differs in that it is converted by contact with bromine vapour ink0 dibromopenb methylenedicarboxylic acid, whereas under these circumstances the tetrene acid remains unchanged.980 E.HARORTH AND W. H. PERKIN, JUN.: Experiments are also described in this paper, made with the object of preparing the anhydride of trans-pentamethylenedicarboxylic acid ; as this anhydride should, according to Baeyer's views (Annalen, 257, 179), be capable of existence, and should, indeed, be as readily produced as the anhydride of ti-am-hexahydrophthalic acid. As. however, trans-pent am~thylenedicarboxylic acid is not appreciab1:- acted on by acetyl chloride at 100" (this TO]., 587), and at 140" yields the anhydride of the cis-acid, results which Baeyer was kind enough to confirm, it seemed unlikely that the trans-anhydride could be isolated even if it existed.However, a t Baeyer's suggestion, we systematically studied the action of acetic anhydride on the pure trans-acid, and as the result of a large number of experiments we find that the almost pure trans- anhydride is formed when the acid is digested with acetic anhydride for 25 minutes and the excess of acetic anhydride removed by passing a current of dry air through the product heated at 120" under 20 mm. pressure; i t is an oil which contains traces of the anhydride of the cis-acid, into which it is completely converted b~ distillation. The analogy between the dimethylsuccinic acids, the hexahydro- phthalic acids, and the pentameth ylenedicarboxylic acids is now corn- plete, the only remarkable point being that whereas the anhydrides of the trans-modifications of the two first-named acids are so easily formed, tmns-pentamethjlenedicarboxylic acid can only be converted into its anhydride with considerable difficulty.CH, 7 Br C OOH CH,*CB~*COOH' Dibromopentamethy lenedicarboxylic acid, CH,< I n order to prcpare this substance, cis-pentamethylenedicarboxylic anhydride (10 grams) was mixed with amorphous phosphorus ( 2 grams), dry bromine (60 gmms) gradually added, with constant cooling, and the whole heated in a reflus apparatus for six hours to boiling ; torrents of hydrogen bromide were given off, and the action appeared to proceed in much the same way as in the case of the formation of dibromotetramethylenedicarboxy lic acid (p.966). A further quantity of bromine (30 grams) and amorphous phosphorous (1 gram) was then added, and the heating continued for six hours longer ; the product, after being freed from excess of bromine by blowing in air, was poured into a large quantity of water, care being taken to keep the temperature below 30" by coding with water. After about an hour, the heavy oil was extracted with ether, the ethereal solution filtered from a considerable quantity of carbonaceous matter, washed with a, little dilute sodium hydrogen sulphite solu- tion to remove the last traces of bromine, dried by calcium chloride,1 : %PENTA4NETHYLENEDICARBOX~IC ACID.981 evaporated, aud the dark-colonrod, oily residue allowed to stand over sulphnric acid in a vacuum for about a week. At the end of that time, the thick, prismatic crystals which had separated were freed as far as possible from the dark, oily, mother liquor by decantation, allowed to remain in contact with porous porcelain nntil quite colonr- less, ground up, and recrystallised twice from fuming hydrobromic acid. In this way, colourless, glistening crystals were obtained, which, after drying at looo, gave the following resu1t8s on analysis. Found. -----7 TheorF. I. 11. C,H6Br2(COOH),. H.. ...... 2-54 2-95 ,, 2.53 ,, C ........ 26.71 27-24 per cent. 26.58 per cent. Br ....... 50.21 4969 ,, 50.63 ,, I?ibromopentamethylenedica.rboxylic acid melts at 183-189" with evolution of gas; i t is readily soluble i n warm water, alcobol, and et,her, and in hot concentrated hydrochloric or hydrobromic acids, but only sparingly in the cold acids.In general poperties, it resembles the corresponding dibromotetramethylenedicarboxylic acid, except that it is less stable and very readily carbonises when treated with reagents or when heated ; thus, for example, all attempts to pmpare its anhydride by distillation under diminished pressure failed, owing fo the acid decomposing into a carbonaceous mass with evolution of bjdrogen bromide in abundance. When heated with acetic anhydride in a reflux apparatus, the solution rapidly darkened and ultimately became almost black, hydrogen bromide being evolved ; ou distilling the product under redaced pressure (20 mm.) decomposition set i n as soon as the acetic anhydride had passed over ; many other experi- ments were made with the object of preparing the anhydride, but i n no case could a crystalline compound be isolated.CH,*gBr Bromod ihydl.oiventeiiecarboxy lic acid, C H, < CH~*C-COOH' This is formed when alcoholic potash acts on dibromopenta- methylenedicarboxylic acid. The pure dibromo-acid ( 5 grams) was d d e d in small quantities at a time to a strong boiling solution of pure potash (10 grams) in ethylic alcohol ; a vigorous action set in, potassium bromide separated, and atl the sama time the mass became dark coloured. After boiling for a few minutes, using a reflux con- denser, water was added, the dark brown solution, filtered from small quantities of carbonaceous matter, was acidified, and extracteG five times wihh ether ; the ethereal solution, after being mashed well wit,h982 E.HAWORTH AXD W. H. PERKIN, JUN.: water, and dried oTer cdcinm chloride, was evaporated, and the cry5 talline residue left in contact with porous porcelain for some hours nntil free from oily mother liquor. The crude residual substance was then purified by repeated recrystalli~ation from water with the aid of animal charcoal, and the colourless crystals analysed. Found. rL 7 Theory. I. 11. C~HGBPCOOH. C . . ...... 38.15 38.02 per cent. 37.69 per cent. H ........ 4-0e 3.94 ,) 3-66 ,, Br.. ..... 42-15 42.01 .. 41.88 ,, Bromodihydropentenecarboxglic acid melts at about 130", but softens a few degrees lower, and, when strongly heated, distils without leaving any residue and apparently without undergoing decomposi- tion; it is readily soluble in alcohol, ether, benzene, chloroform, formic acid, and hot water or light petroleum, but only sparingly in the two latter solvents in the cold; i t crystallises best from hot water. The solution of the acid in dilute sodium carbonate decolorises permanganate rapidly.Bromodihydropentenecarboxy lic acid is apparently not the only product of the action of alcoholic potash on dibromopentamethylene- dicarboxylic acid, but the other substance or substances present in the mother liquor of this acid could not be isolated in quantity suffi- cient for analysis. CH,*YBr, CH, CBr* COOH. z&- Tribromopentameth yler,ecarboxyl ic acid, CH, < This acid is formed quantitatively when bromodihydropentene- carbosylic acid is exposed to the action of bromine vapour.CH,*gBr CHz* YBr, CH2<CH2*CBr COOH' + Brz = C H % ~ , * ~ -COOH 0.5546 gram of pure bromodihydropentenecarboxylic acid left in contact with dry bromine vapour for 24 hours, and then freed from excess of bromine by exposure over potash in a vacuum, had gained 0.481 gram in weight, whereas, in the formation of the tribromo- acid, according t o the above equation, theoretically 0.464 gram of bromine should have been taken up. The product is highlyunstable, but, with care, it may be recrystallised from formic acid, and is thus obtained in colourless nodular masses, about half of the substance remaining in the mother liquor. The pure substance, on analysis, gave the following result.1 : 2-PENTAMETHYLENEDICARBOXYLIC ACID.983 Theorp. 68.37 per cent. Found. C,HiBr,02. Br.. . . . . . . 68.04 per cent. Tribromopenhmethylenecarboxylic acid is rea,dily soluble in ether and alcohol, but only sparingly iE water, formic acid, or light petro- leum ; it is very readily decomposed when boiled with water, hydrogen bromide being eliminated. The freshly-prepared solution of the acid in cold sodium carbonate does not decolorise permanganate instantaneously, but oxidation takes place slowly on standing, and very rapidlF if the solution is warmed, due doubtless to the gradual decomposition of the tribromo- acid into unsaturated compounds. CH,-#*COOH CH2* C . C OOH Dih y dropeutenedicar box y lie acid, CH,< In order to prepare this beautiful crystalline substance, the following method was employed.The crude product of the action of bromine and phosphorus OD pentamethylenedicarboxylic acid, after removal of excess of bromine by blowing in air (p. 980), was treated with methylic alcohol ; and as soon as the mixture had cooled down it was poured into water; and the crude methylic salt which separated as a dark-coloured heavy oil extracted with ether. The ethereal solution was washed with dilute sodium carbonate solution, dried over anhydrous potassium car- bonate, filtered, and evaporated. The dark-coloured, oily residue, con- sisting of crude methylic dibromopentamethylenedicarbosylate, did not deposit crystals, even after standing for some d a p over sulphnric acid in a vacuum, and therefore no analysis was made.This crude methylic salt was dissolved in alcohol, a large excess of powdered potassium iodide added, and the whole heated in a reflux apparatus for two hours ; iodine was liberated in abundance, and the action appeared to proceed exactly as in the case of the decomposi- tion of methylic dibromotetramethylenedicarboxylate under similar ecnditions (p. 974). To the product, mixed with water, sufficient dilute sodium hydrogen sulphite solution was added to remove the free iodine, and the oily methylic sslt extracted twice with ether; the ethereal solution was then dried over calcium chloride, evaporated, and the residual, oily, ethereal salt at once converted into the corre- sponding acid by hydrolysis with potash. After boiling with excess of alcoholic potash for about five minutes, the solution was diluted with water, evaporated on a water bath till free from alcohol, acidified, and extracted 1.5 times with ether ; the dark, crystalline mass which remained after distilling off the ether was purified by recrystallisation,984 E.HAWORTH AND W. H. PERKIN, JCIN.: first from water with the aid of animal charcoal, azd then from ether. Analysis, Found. r---A--7 Theory. I. 11. C5HB(COOH)Z. C . . . . . . . . 53.90 53.77 per cent. 53.84 per cent. H .. .. . . . . 5-46 5-33 ,, 5.13 ,, Dihydropentenedicarboxjlic acid crystallises in hard, glistening, transparent crystals, and melts unaltered at about 178" ; it is readily soluble in hot water, alcohol, acetic acid, and acetic anhydride, moderately in ether, but only sparingly in benzene, light petroleum, and cold water.It is a strong acid ; its aqueous solution is strongly acid to litmus, and even thick crystals effervesce vigorously when brought in contact with sodium cai-bonnte solution ; a dilute solution of its sodium salt decolorises perrnauganate, although not so instan- tmeously as the sodium salts oE many other unsaturated acids. In its behaviour towards bromine vapour, this acid appears to differ in a very marked manner from the corresponding dihydrotetrenedicar- boxylic acid; a small quantity oE the former was exposed to dry bromine vapour under a bell-jar during two days, the reddish- colonl=ed product was freed from bromine over potash in a vacuum dessicator, and the residue recrystallised from hydrobromic acid, The crystals obtained mdted and decomposed at 180°, and as they con- tained bromine, they probably consisted of dibromopentamethylene- dicarboxylic acid ; mfortunately the identity could not be proved by analysis, owing to the small quantity of material at our disposal; under similar conditions, dihydrotetrenedicarboxylic acid remains unchanged.I n its behaviour when heated, dihydropentenedicarboxylic acid .differs also in a marked manner from the Corresponding tetrene derivative ; the latter, on distillation, carbonises and is entirely decomposed, whereas the former, if rapidly heated, distils without decomposition, the solid distillate melting at 175-17i". I f , how- ever, the pentene acid be heated for 10 minutes in a narrow test- tube in sucb a way that the distillate constantly runs back, and the podnct be then distilled, an oily distillate will be obtained which, on long standing, becomes opaque from separation of a small quantity of a crystalline substance, and, on warming with water, it first melts and then gradually passes into solution.On standing over sulphuric acid in a vacuum, the aqueous sducion deposits a small quantity of dihydropentenedicaPb~~~lic acid: and the filtrate from these crystals, on evaporation, dries up to a gummy residue, which resembles the product obtained on dissolving the anhydride of dihydrotetrenedi- carboxylic acid in water; it could not be analysed, the quantity being insufficient .Salts of Dihydropentenedicar~ox~~iG acid. Hydrogen Silcer Salt, COOH*C,H6* CoOhg.-one of the most characteristic properties of dibydrotetrenedicarboxylic acid is the tendency which it shows to form a hydrogen silver salt, and this property is even more pronoanced in the case of dihydropentenedi- carboxylic acid.The pure acid was dissolved in water, the solution rendered slightly alkaline with ani11z?onia, warmed to about 40°, and excess of silver nitrate added, wben a white precipitate separated, which was collected, washed with water, and dried over sulphuric acid in a vacuum. When heated in a crucible, this salt decomposes suddenly with formation of voluminous threads of silT-er, and the amount of silver contained in the salt was found to be only 44.3 per cent., whereas the neutral salt, CiH6O4Ag?, would contain 58.38 per ceut.; indeed, this analysis agrees very much better with the composition of the hydrogen silver salt, CiHi04A,a, which contains 41.08 per cent. of silver. The mother liquors from this salt, on standing, deposited long, colonrless needles vl.hich in appearance closely resemble the crystals of the hydrogen silver salt of dihydroteirenedicarboxFlic acid. They =ere collected, washed with water, and aiialysed with the following result. Theory. 41.08 per cent. Found. CjHj0,Ag. Ag . . .. . . . . 41-41 per cent. This salt is, therefore, evidently the hydrogen silver salt. A neutral solution of the ammonium salt of dihydropentenedi- carboxylic acid shows the following behaviour with reagents. Lend acetate ; a white, gelatinous precipitate. Copper sulphate and barium nit1 ate ; no precipitate.C1a1ci7im chloride; no precipitate at first, but, on standing, t h e calcium salt separates slowly in magnificent four-sided, glistening, tabular crystals. A&ydride of Trans-peii~ametl~y7e~zecEicnrboxylic acid. As explained in t'he introduction to this paDer, considerable in- terest attaches to the preparation of an anbydride of trans-penta- methylenedicarboxylic acid, and many experiments were made with the object of obtaining this substance, of wliich only one or two can be described here. The trans-acid used in these experiments was very carefully puri- fied by repeated recrystallisation ; it was then heated in sealed tubes VOL. LXV. 4 A986 I : ~-PENTA~ETEBLE~EDICARBOSYLIC ACID. at 200" with concentrated hydrochloric acid, in order to free it from traces of tho cis-acid which might possibly be present (compare this vol., 590), and then quantities of about 1 gram were heated with pure acetic anhydride to boiling on a.reflux apparatus for 5, 10, 15, 20, and 2.5 minutes respectively. After standing for several days over potash in a vacuum, and until free from acetic anhydride, ft very thick, oily residue was in all cases, obtained, which, even when the heating had been continued for 25 minutes, gradually deposited varying quantities of unchanged acid, and all the samples, ou stand- ing, exposed to the air completely solidified to a cake of the pure acid in a few days. Ultimately, a substance consisting evidently of nearly pure trans- pentamethylenedicarboxylic anhydride was obtained as follows. The pure acid was heated with 10.times its weight of pure acetic anbydride on a reflux apparatus for 25 minutes, YO that the liquid just boiled ; the product mas then transferred to a small Wiirtz flask, heated at 120" under a pressure of 20 mm., and a slow streani of carefully dried air allowed to pass through the liquid until the excess of acetic anhydride had been completely removed. The sliqhtly brownish residue gave, on analysis, the following num- bers, which agree with those required for the anhydride of h a m - pentstm&hylenedicarboxglic acid. Found. 7- Theory. I. 11. CSW%. C . . . . . . .. 59-75 59-81 per cent. 60.00 per cent. H . . . . . . . . 5-90 5.87 ,, 5.71 ,, This substance is, however, not quite pure, and contains traces of the cis-anhydride, as was shown by the fact that when it was dissolved in hot water, and the solution allowed to cool, crystals were obtained which did not melt quite sharply at 160" (the m.p. of the trnns- acid), and the mother liquor was found to contain traces of the cis-acid; nevertheless, there can be no doubt that the greater por- tion of the substance must have consisted of the anhydride of the trans-acid. In its properties, this anhydride diffem in many ways from the beantifully crystalline anhydride of the cis-acid ; it dissolves, for emmple, in sodium carbonate very rapidly with effervescence, and when exposed to the air on a watch-glass, it gradually solidifies to a solid cake consisting principally of the trans-acid ; the cis-anhydride, on the other hand, is stable in the air, and dissolves only very slowly in sodium carbonate. Possibly this tendency to combine with water may account f o r the gradual deposition of trans-acid from the products of the time ex-SUBSTITUTED PIBIELIC ACIDS. 98 7 periments described above, as it is difficult to be certain that potash is dry, and, moreover, water would be produced by the absorption of the acetic acid. The remainder of the trms-anhydride was distilled under reduced pressure, when a colourless distillate was obtained, which solidified in the manner so highly characteristic of tlhe cis-anhydride ; the dis- tiIlate was dissolved in acetic anhydride and allowed to slowly era- porate over pot'ash in a vacuum, when crystals gradually separated which melted at 70-7'1" (the cis-anhydride melts at 73"), ard gave on analysis the following result. Theoyi. 60.00 per cent. Found. CjHgCd03. C . . . . . . . . 59.68 per cent. H .... .... 5.89 ,, 5.71 ,, It is, therefore, obvious that trans-pentamethylenedicarboxylic anhydride, on distillation, is converted into the anhydride of the cis- acid. Chemical Laboratory, Owens College, HancAester.

ISSN:0368-1645    

DOI:10.1039/CT8946500978

出版商:RSC    

年代:1894

数据来源: RSC