[CONTRIBUTION FROM T H E
CHEMICALLABORATORY OF TEXAS].
T H E U N I V E R S I T Y OF
I. DIAZOAMINO COMPOUNDS FROM SEMICARBAZINO ACIDS. 11. CHEMICAL BEHAVIOR OF THE DERIVATIVES OF CARBONAMIDOAZO-i-BUTYRICACID. BY J. R . BAILEY A N D LOCIS K N O X .
Introduction I t has been shown by one of the authors in investigations, conducted in part conjointly with students, that the semicarbazino fatty acids show a close resemblance in chemical behavior to secondary amines. The a-hydrogen atom in these semicarbazino compounds is readily replaced by acyl radicals1 and the a-imino group condenses with mustard oils.? -4s yet no work has been published on the behavior of these substances toward nitrous acid, although from the work of Thiele and Heuser on hydrazo-i-butyric acid3 and of 0.Widman on phenyl-semicarbazine4, it is probable that the analogy in the chemical behavior of secondary amines and semicarbazino fatty acids will also hold here. I n this paper the behavior of the semicarbazino fatty acids to\vard the acid salts of diazonium compounds, more especially Iit-nitrophenyleiiediazonium chloride is discussed. Probably all semicarbazine and hydrazine bodies react with diazonium bodies although the reactions are not always similar, and in many instances the instability of the substances formed make the isolation and purification of the reaction products difficult. T h e action of diazonium salts on hydrazine and phenylhydrazine has been shown by other investigators to result in the formation of azoi1nides.j Ey working with the acetate of phenylhydrazine and diazobenzene acetate, \170hlc succeeded in isolating the primary reaction product of diazobenzene on phenylhydrazine, a substance with the empirical formula C12H12K4. Furthermore, Curtius obtained from hippurylhydrazine and diazobenzene sulphate hippurqlpheii!lbuzylene, to which he ascribed the constitution, C , H , C O K H C H , C O N H S H S = SC,,H-,.’ Diazonium salts in solution with semicarbazine pass into aromatic diazoamino bodies, in which the semicarbazine radical is not found, and in some instances these products ere found to result when diazonium salts react with semicarbszino acid derivatives, owing primarily to the strongly reducing action of the latter compounds. However, as concerns the semicarbazino acids and their derivatives, it is in most cases possible to prepare substances, which appear to be diazoamino bodies, in which the diazonium radical is coupled directly Ber., 33, 1520. This Journal, 26, 1006. Ann. 290, 23. Ber., 28, 1925. j Ibid., 26, 1263; 23, 3025. Ibid., 26, 1587. Ibid., 26, 1268.
882
J . R . BAILEY A N D LOUIS K X O X
with the a-nitrogen of the seinicarbazino compounds. Experiments instituted on another class of hydrazo bodies closely related to the semicarbazino acids, the hydrazo-i-butyric acid derivatives described by 'l'hiele and Heuserl, did not result in the isolation of diazoamiiio coinpounds, corresponding to those obtained in the semicarbazino series. As typical for the coiistitution of the substances 1;ere described, the reaction product of phen\.ldiazonium chloride on semicarbazinopropionic acid ethyl ester. may be selected. This substance is probably constituted as follows : (SH,CONHS (S,C,H,) CH (CH,) COOC,IS,). T h e formula C
I (S,C,H,)COOC,H, is excluded, i l l vie\v of t h e CH,, fact that semicarbazino-i-butyric acid derivatives also couple with diazonium bodies. Here the a-carbon atom is not in union with hydrogen and, therefore, it is improbable that the diazo groun is joined directly to carbon. Furthermore, phenylsemicarbazine exhibits a behavior toivard diazonium salts, similar to that of the ,other semi-carbazino bodies. These €acts indicate strongly that the seniicarbaziiio compouiitls react with diazonium salts through the labile hjdrogen of the a-ilnitio group. ;is a negai ive proof that the hydrogen atom replaced in these reactions is on t h e a-nitrogen, may be cited the fact that the acyl tleri\xtivcs ut' wiiicarbaziiiopropionic acid' do not react with diazoniutii salts. The tlinzoaniino bodies here tlcscribed do riot shon. tl!e characteristic rractioii of diazoamino bodies in general, e. g.,the resorcinol reaction suggested by Caeyer: is not shown by the diazoaiiiiiio compounds described in this article. I n their behavior ton-ards alcoliolic potash, these diazoaiiiino bodies slio\v interesting differences. \\lien o.-iu-nitropheiiylcIiediazoseiiiicarbazinopropionic acid ethyl ester is treated ivith alcoholic potash, two substances result : on the one hand. 6 - m e t l i ~ l - ~ . ~ - c l i l i ~ ~ l r o ~ I .2,3-triazitie. an osidatioti product of semicarbazinopropionic acid, is formed and, on the other hand, nitrobenzene, a reduction product of the original diazonium salt is formed. This reaction may be interpreted as follows : N H , C O N H S C H ( CH,,)COOC,H, -:= C,H,?JO, -: X , t
NH,CONHNH
S
S,C,H,SO,
?J
/ \
OH--C
C ----e H,
C---OH
'
+ C,H,OH.
\ / N
* Ann.,
290, I .
loc. cit. J3er. 8, 148. Here and in similar c3:np)un 1s r-fcrrcl t o in this artic;e, to distinguish between the several tautomxic possibi!ities.
11)
attempt is made
DIAZOAMINO COMPOUNDS
853
It was to be expected from the above that a-Vreido-a-m-nitrophenylenediazoamino-i-butyric acid ethyl ester would similarly yield with alcoholic potash, 5,6-dihydro-6-dinieth;vl-3-hydroxy-5-keto-1,2,4-triazine.
x KO-C
I
j
\ /
c=o
N However, ring condensation results in this case without the elimination of the diazo radical I ,6-dihydro-6-dimethyl-3,j-dihydroxy-1-m-nitrophenylenediazo-1,2,4-triazinebeing formed.
HO-C
C-OH
Attempts to obtain 5,6-dihydro-6-dfmethyl-3-hydroxy-5-keto-1,2,4-triazine in the above reaction and in other ways were without avail. Carbenamidoazo-i-butyric acid ethyl ester was prepared, in the hope that it, similarly to carbonamidehydrazo-i-butyric acid esters would condense with sodium alcoholate under ring formation' and yield the desired triazine. A further attempt to prepare it by oxidation of 1,6-dihydro-B-dinaethyl-3,5-dihydroxy-1,2,4-triazinewas without the desired result?. In the experimental part of this article, it is shown that carbonamidoazo-ibutyric acid, as well as the corresponding triazine, breaks down spontaneously at the moment of its formation, giving off two atoms of nitrogen, a decomposition similar to that of azo-i-butyric acid3. Furthermore, carbonamidoazo-i-butyric acid ethyl ester decomposes at a few degrees above its melting point. These experiments go to confirm the observation of Thiele and Heuser that the azo compounds in the aliphatic series, in contrast to the aromatic azo compounds are not as a rule very .stable and in many instances are incapable of existence4. Experimental Part. P a r t I. Diazoamino Compounds from Semicarbazino Acids Acctala'eliydeseinicarba=oiie-In the preparation of acetaldehydesemicarbazone, a slight nmdification of the method of Thiele and Bailey was 'Am. Ch. J. 28, 40.2. Compare Ann., 303, 8 2 Ibid. 290, 8. * Ibid., 290, 12.
'3'34
J , K . BAILEY A N D LOUIS K S O S
eiiiplo!wl'. I t \\-as found that acetaldeli!-tle coLiltl be substituted for alcleh>~cleaiiimoiiiain iiiakiiig the seiiiicarl~azoneant1 in this \va]. an aliiiost pure pro(luct is tlirectl!. obtaiiictl, \\-hicli !.ieltls \\.it!i prushic acitl scmicarl~azinnpropioiiitrile of slifticicilt ptirit!. for direct c~iiversi~iii iiito other tlerii-at i \ w . 130 grams of hytlraziiie sulpliate antl 5 j graiiis of sodium cariwnate are iiiatle into a thin paste ivitli jo cc. of \\.atel- a i i t l to this a soliitioii of 8 j gra~iisof potassium cyanate in i o cc. of \\-ater is atltlctl. .litctire iii ;i >-ieltl of ahout 55% of tlic t I1eore t ica I. ~ ~ ' , / 1 i ~ ~ .A ~ c7i di ,L;'stcrs.-Thc ~ ~ ~ ~ ~ ~ i ~preparation o o f thc scmicarlnzino aciti esters has 11eeii studied by Tliiele ant1 Lkiiie!.' antl by hilt!, aiitl \\~essoii'. 111 this \vork the folloiving modification of tlie iiietliotls ciiiplo!~etl 11y these iiivcstigators \\-as usetl. I t \\-asf o m d that the !.ieltl of tlie ester is not materiall!, aii'ectetl by the piirit!. of the semicar1)aziiiopropionitrile r.mplo!.eti, aiitl. therefore. it is not necessar!- CIT even tlesirable t o Inirii!. the latter substance. 1;urtliermore. in saturating the alcoholic solution of this Iiitrilc Ivitli hytlrochloric acid gas. the rcpulation of t!ie teiiiperature d o c s iiot iiiaterially affect the yield of ester. To prepare the etliyl or propyl ester of ..emicarl:azinoproi)ioiiic acid, the nitrile is dissol\-etl i i i four times its \\-eight of alcohol antl. \\-ithout cooling, h!drochloric acitl gas is let1 in to saturation. -After stantling tlvelve hours, the alcoholic solution is neutrali z u l \vith sodium rarhoiic7fr, tlie alcohol is then tlistilletl iii ':'nriio. tile residue is tlissolvetl i n \ v a t u , aii(1 tlie ester is cztractetl i r w the acliicoiis solution lvitli chloroform. The semicarbaziiiopro1)ioiiic acid iiieth>.l cster cannot be prepare(1 according to this method. Sumeroiis attempts \\-err nlso iiiatlc to prepare semicarbaziiio-i-l~~ityric acid ethyl cstrr. through the formation of t h e corresponding iniiiio ether, but, although t h e ester, tiiisetl \\-it11a larger amouiit of tliiiieth!.l triazine, \\-as oi)tainetl in this \\'a!., this iiiethotl of preparing it was finally abantloiietl for tlie mcthotl suggested by Bailey and IYesson'. u-Crciiio-a-Piicii!'In'ia-ociiiiii!o~ropioiiic arid ethyl cstcr, SH,CC)SHS(S,C,,H,) C H (CH,)COOC,H,.-A diazo solution is prepared by dissolving 9.3 grams of aniline in 1,000 cc. of \ v a t u , containing 3.6 grains of hydrochloric acid, and adding to this solution cooled to 0'. under constant stirring, a previously cooletl solution of 6.9grams o f sotliiuii nitrite in 2 5 0 ?
Ann., 303, 79. Ibid., 303, S j .
":\in. C h . J.
28, 396.
' I l d . 28, 402.
DIAZOAMINO C03IPOUNDS
885
of water. T h e diazo solution thus prepared is immediately added under stirring to a solution of 17.j grams of seniicarbazinopropionic acid ethyl ester in 1,500 cc. of water, also previously cooled to 0 ' . The substance begills to separate out a t once and in half an hour the compound is filtered off. T h e yield obtained was 9 grams. This diazoamino coinpound is soluble in alcohol, acetic ester, and chloroform, very sparingly soluble in ether, and insoluble in benzene and petrolic ether. For purification, the substance is dissolved in alcohol and precipitated with m-ater as a yellowwhite crystalline mass. Recrystallized twice in this manner, the substance is colorless, melts under explosion to a dark-red liquid at 12j01, and under the microscope it is seen to consist of clusters of long interlacing needles. On exposure to light, it rapidly turns yellow. Calculated for C,,H,,O,N,: C, j1.61 ; H , 6.09; N, 2j.09. Found: C, j1.71 ; H, 6.0j ; N, 24.92. JF7edekind has Zhown that by employing a nitrophenylenediazoniuni salt, instead of the simple phenyldiazonium salt, in effecting condensation< of an analogous character to those described here, the formation of smeary substances is considerably avoided and the resulting compounds, as a rule, admit of ready c r j stallization'. Similar observations have been made in this work. a - U r r i d o - a - i , z - i i r t r o ~ l ~ e ~ z ~ l e i i e d i a ~ o a i i i i i i o ~ acid ~ o ~ i ethyl o i i ~ c ester, SH,COS"( N 2 C R H 4 S 0 3C) H (CH,) COOC,H;.-The preparation of this substance mal- be taken as typical for the substances described in this article, I\ hich were obtained by the use of iic-nitrophenj lenediazoniuni chloride. X diazo solution is prepared by dissolving 8 grains of fit-nitraniline in 200 cc. of dilute hydrochloric acid, containing 6.4 grains of HC1. To this solution, cooled to o " , is added 4 grams of sodium nitrite, dissolved in 100 cc. of mater, also cooled to 0 ' . The diazo solution thus prepared is added iinder constant stirring to a solution of I O grams of semicarbazinopropionic acid ethyl ester in 300 cc. of water a t 0'. T h e new diazoainino body separates out immediately as a bulky, cream-white precipitate and in about 30 minutes the reaction is complete. T h e yield is almost quantitative. The subtance is soluble in alcohol and acetic ester, very soluble in cold chloroform, insoluble in ether, petrolic ether and water. -Icetic ester is best adapted for purification of the compound. Crystallized once from this medium, the substance decomposes under explosion at 146' to a darkred liquid. CC.
I The melting points and explosion points given here are corrected. However, the explosion points depend in large measure on the rate of heating. I t is characteristic of all the diazoamino compounds described in this article that they decompose under explosion a t elevated temperatures without previously melting. Furthermore, they are all sensitive to light. Ann., 307, 283.
856
J . R. BAILEY A S D LOUIS KNOM
Calculated for Cl2Hl6OZK6: C, 34.44; H j 4.94;N,33.03. FOLIII~ : C, 44.57 ; H,5.06: S,25.74. The substance dissolves readily
it1
the cold in bromine water, from which
on the addition of ammonia there is precipitated a brick-red substance. Trcattnent of this tliazoaniino body in absolute alcohol so:ution with bromine yields a substance ireely soluble i n d i e r antl alcohol and volatile
wit!i steam, which, after recrystallization froin tlilute alcohol, tnelts at ~~j-i;o". This suhstance is undoubtedl>- ~ir-iiitrc,pIieri!.lc.nc.azoimide. The tliazoatnino bod! is readily decomposed I)!, :tcitis, nl1;alies. ant1 clwi boiling u,a.tt'r. Eehavior of a - i i i c i d o - a - ! i i - i i i j ~ o ~ l i c i ~ ~ ~ c i z c ~ i ~ ~arid o u , irihyE ~j/~o~~~~ rstcr to;iai,tr' alcoholic pofassiiiiiz 1iydro.ridc. potcrssiu,ii rtlid.1-iJe U7ld alcohoiic airzi~~o7~ia.-Ifto this cliazoaniiiio bod!. i n concentrated alcoholic SOIrrtion, therc be atldetl alcoholic potassitini h!,-tlroriilt.. containing oiie molecui? of potassiutii h!.droside, and tlie wisture then be lieated on the Ivatcrbath, the solution clianges from a light y e l h v t o a cleep red color and a cnl)iotis evolution of nitrogen takes place. .\t the m i i t . time a solid separates out of the solution in bright yellon- fialics. ?lie substaim \vas ohtained in silver-n.liite flakes l)?. IioiIing i t in ivater with anittinl charcoal and niter filtration precipitating the salt from its aqtieow m!uti,)n n.ith a n alcohol-ether mixture. a \ ~ ianalysis of this substance pro? .tirosy- I .? .+-t riaz ii! c . Calculated for C 4 H ~ 0 2 S , , :, I ~I ,z,A-triaziiie. The csperinients instituted in this connection are tlescribetl above. In the second place an attempt was made to condense carbonaiiiidoazo-i-butyric acid ethyl ester ivith sotlium alcoholate to the desired triazine. This experiment \vas m successful as the substances reacted violently ivith gas evolution, foriniiig a colorless solution from which nothing \vas isolatetl. ( FROXTHE LABORATORY O F BIOLOGICAL CHEMISTRY OH COLVliI31.1 COLLEGEOF PHYSICIAKS A N D SCRGEOSS, NEW YORK.)
UNIVERSITY,
A T THE
A PRELIMINARY COMMUNICATION ON THE TOXICITY OF SOME ANILINE DYESTUFFS. BY GYSTAYRM. M E Y E R Received April
20,
1907.
Introduction. Recent discussions coucerniiig t h e use of various artificial organic dyes a s food colorants made it seem very desirable to study the toxic effects of 'loc. cit. ?Presented before the Xew York Section of the Am. Cheni. SOC.,Science, 2 5 , 343 (1907). See also, Proc. sect. Biol. Chem. of Am. Chem. SOC.with sect. C., A. .4. A . S., Science, 25, 457 (1907).
