Caffeine Derivatives. I. The 8-ESthet;s of Caffeine

curic chloride yielded 2-chloro-mercurithiophene which did flot depress the melting point of authentic material. Carbon bisulfide was again noted as a...
0 downloads 0 Views 359KB Size
1356

RALPHC. HUSTOMAND WINSTUNF. ALLEN

The temperature a t which the sulfur i s kept affeets the yields: a t 300" the reaction was slow (5% conversion per passage) while a t 400" the ultimate yield was 41% and an inconvedeiltly lafge amoufit 6f sulfur distiIIed over. B h t i o n of the isopt-ene with carbrrn bisdfide (1:l) increased the yield 6f crude 3aiethylthiaphene at 350' to 51%. Thiophene.-Crude 50% butadiene6 yielded 6% of thiophene when bubbled through molten sulfur at temperatures ranging from 320 to 420'. The same yield was obtained when pure liqttid btrtadiene (from the tetrabromide) was dropped into the zpparatus from a chilled separatmy funnel The thiophene boiled a t 82-84",had a density d f g 105, gave a blue indophenin reaction, and with mercuric chloride yielded 2-chloro-mercurithiophene which did flot depress the melting point of authentic material. Carbon bisulfide was again noted as a by-product. 2,3-Dimethylthiophene.-A distillate fraction boiling a t 45-160" was obtained from the pyrolysis products of a mixture of crepe r u b b r and zinc oxide. This fractiun had previously been shown to contain about 4y0 of 3methyi-l,3-pe*tz&&' mixed witk mono-oiefins and hydromatic compounds. One hundred grams of this ma(6) We are indebted to Dr. B. T. Brooks for a generous supply of this material. (7) Midgley and Henne, Trns JOWRNAL. I S , 204 (1931); 51, 1215 (1929).

[COl'?TRtB~TION FROM KEDZik

Vol. 56

terial Was dripped into boiing stdfitr d d gave 48 g. of crude product. The thiophene homolags were isolated therefrom in a manner similar to that previously described8 and there were obtained 0.3 g. of 3-methyithiophene and 0.15 g. of 2,3-dimethyltfriophene. The 2,3;&methylthiopfiene was identified as the mefcurkhforide, melting!cit 216' (abne, or mixed With authentic material) and the mercmithiocyanate which blackened but did not melt at 2 0 5 O . 6 3,4-Dimethylthiophene.-From 50 cc. of pure dimethylbutadiene which had reacted wi€h sulfur at 40(1-420° were recovered 10 cc. of unconverted material a d 12.5 g. (31 yield) of 3,4-dimethyfthiaphene, which boiled st 146.5-148.6', had a density dl: 5.994,a refractive index na: 1.5187 and gave a blue indophenin reaction. For identification, its mercurichloride (melting at 139-140.5"), mercuriiodide (melting at 140-141 ") and bis-2-(3,4dimethylthienyl) mercury (melting a t 153-154 ") derivatives were prepared 6

summary A procedure is k i b e d for the conversion of diolefins into thiophene homologs by interaction with inuiten sulfur. (8) hlidgley, Henne and Shepard, ibid., 54, 2957 (1932).

TWEM&G%rtrF O I J ~ A T E ~ COLUMBUS, OHIO RECEIVED FEBRUARY 19, 1934

h W d A L LABORPiTORY, MfCWtoAN &ATE &LLBGE ]

Caffeine Derivatives. I. The 8-ESthet;s of Caffeine BY RALPHc. HUSTONAND The extensive investigations and syntheses of compounds in the purige g o u p by Em2 Fischer and his associa& include the preparation of only two 8-mmaethers af cdeine, nameIy, methoxyand efhmrycaffe.ine. Esrher's methoff ,* laftr repeated by W. Wislicenus3 and H. BiItz,4 consisted simply in heating 8-chloro- or &bromocaffeine in a rrreshyl or ethyl lcohol gOlutiOft cuntainilzg an excess of sodium or put&m alcoholate (ssdimn or potasdurn added in the form of the metal or hydroxide). Isoamyloxydeine was later prepared by the same general procedure.5 The prepaxa6tiam of fifteen 8-phenyl ethers of cdeiae Was wcmplished by A. Baurnann.6 The (I) Abstract from a thesis presented in partial fulfilment of therequiremefits for fke PED. degree by df. 1. Aben, W. IC. Itellogg FcRor at Mlctdgsn &&e Chileg@,1982. (2) E Fischer, Bw., 17,17%6(18%1); ibcd., SO, 569 (1897); Ann., '215, 281 (18821, and "Untersuehttngen in der Puringruppe," pp. 38, 96-96, 162. (3) W W i s l i m s add #. K&irbCr, Bey., s6,isgz (1902) (4) H Biltz rhad M. Wgh, Alae., &4, S(Y (1017). (5) Magnanini, private communication. (6) A Baumann, Arck. P h m . Inst. Unrv. Barlrn, 10, 127-M7 (1913). and Chem. Zcnlr , 11, 2036 (1913).

WfNSTON

F. ALLEN'

methods involved hatifig 8-cYrfclrocaffeirie wlth a phenol in the ptesence af m equivalent momt of putassinm hydroxide. The reactim took p k in aa aquedtts sdutim v&h r&ziag, or tlder presure at a higher tentperatwe. Xyfene was used as the solvent instead of water some eases. Gelleral Methods o€ Preparation Both 8-cMoro- itnd 8abronmcaffaine wete used in the preparation of the sixteen ethers described in this paper. The former was used in most cases because of the ease and ecofiomy of p r e p a r a t i ~ n . ~P&ir& good yields d 8'metiiony- err& SeMK+xFcaffefne e r e &dined by wing M excess a f metallic wditfm ut So&* kydrmide, as k Fischer's method. Attempts to prepare the higher alkyl and aryl ethers by the same procedure met with little or no success. A more successful procedure was adopted whereby an exactly equivalent amount of sodium dtohoiate or pkenyfate was used. In pepsrfrig tRe alkyl ethers, clean metalle sodium waa qnichlg weighed mt to Q.0P g. m d cut into s n d pkccs.8 (7) E.Fischer and Reese, Ann., M l , a39 (.1&33>. (8) Sodium in the lorm af "bird shet" was used sometimes to an advantage when kept in a toluene solutian and weighed out n s needed.

THEÐERSOF CAFFEINE

June, 1934

The sodium was added to ennough absolute alcohol to effect Sdrjti~nof the resulting caffeine ether while hot. The proportion was generally 50 g. of the chlorocaffeine to 300 cc. of alcohol. After disappearance of the sodium, a n equivalent amount of 8-chlorocaffeine (dried at 120") was added and shaken well. The mixture was refluxed, with occasional shaking, on a steam or oil bath for one-half to five hours, depending upon the speed of the reaction. The hot solution was filtered immediately to remove the sodium chloride, the flask and filter being washed with a small quantity of the absolute alcohol. Methoxy-, ethoxy-, normal propyloxy-, secondary propyloxy- and isoamyloxycaffeine crystallized out more or less completely when cooled. With the higher alkyl ethers, such as n-hexyloxyand n-heptyloxycafieine, it was necessary to concentrate the solution to a smaller volume by vacuum distillation. In order to secure iz maximum yield, the mother liquor was diluted with about an equal volume of water. In case of

1357

the lower members of the series, merely concentrating the solution to a smaller volume w w found suilicient. Dilute (40-50%) ethyl alcohol was found very suitable

for recrystallizing the alkyl ethers. The amounts and proportions varied because of the decreased solubility of successive members of the series in water. To secure a more nearly anhydrous compound, the first four members were also successfully recrystallized from carbon tetrachloride. The higher members were recrystallized from petroleum ether for the same reason. The normal alkyl ethers were obtained in nearly quantitative yields and the reaction was usually completed within one-half an hour. The secondary caffeine ethers were formed rather slowly and gave from 40-50% yields. Only about 12% yield was obtained of t-butyloxycaffeine after several hours. Benzyloxy- and phenylethoxycaffeine were prepared in the same manner as the alkyl derivatives. However, be-

TABLE I PREPARATION AND PROPERTIES OF THE-ÐERSOF CAFFEINE -Materials 8-Cle&., Alc., g. cc.

Name

Methoxycaffeine2*sj4 EthoxycaiTeine2~3~4 n-Propyloxycaffeine s-Propyloxycaffeine n-Butyloxycaffeine s-Butyloxycaffeine t-Butyloxycaffeine n-Amyloxycaffeine Isoamyloxycaffeinr~ n-Hexyloxycaffeine n-Heptyloxycaffeine Allyloxycaffeine Phenyloxycaffeine p-Hydroxyphenyloxycaffeine Benzyloxycaff eine Fhenylethoxycaff eine

200 100 60 50 30 30 30 20

C-OR

CH-N-C-N Caffeine ether

heat

+

O=C

1

I I

1

C-N-CHs

I

>&O CHa-N-C-N-R Substituted uric acid

(1) Abstract from a thesis presented in partial fulfilment of the requirements for the Ph.D. degree by W. P. Allen, W. K. Kellogg Fellow at Michigan State College, 1932. (2) Wheeler and Johnson, Am. Chcm. J., 11,185 (1899). (3) Wislicenus and K6rher, Bcr., 86, 164, 1991 (1903). (4) Wislicenus and Goldschmidt, ibdd., 88, 1467 (1900); Lander, J. Chcm. Soc., 88, 406 (1903).

In repeating the work of Biltz and his co-workers6on the molecular rearrangement of methoxy and ethoxycaffeine to tetramethyluric acid and trimethyl-9-ethyluric acid, respectively, it was found that the rearrangement took place fully as well, if not better, when the ethers were heated in an open tube contained in a paraffin bath. Tetramethyluric acid was obtained in yields as high as 95% (Anal. Calcd. for COHIZN~O~: N, 25.00. Found: N, 24.42.), while trimethyl-9-ethyluric acid was formed in about 50% yields (Anal. Calcd. for C&I~N,O~: N, 23.53. Found: N, 22.77). The trimethyl-9-alkyluric acids were conveniently separated from hydroxycaffeine by forming the insoluble barium salts of the latter and extracting with chloroform. All attempts to cause the molecular rearrangement of n-propyloxycaffeine to trimethyl-9-n-propyluric acid by heating in a closed tube were unsuccessful. However, by heating this ether in an open tube a t 250-270' for eight hours, a yield of trimethyl-9-n-propyluric acid was produced, melting at 138.8-140.6" Anal. Calcd. for CllH~sN40a.Hz0:N, 20.74; C,48.88; H, 6.66. Found: N, 20.87; C, 49.20, 49.09; H, 6.40, 6.34. The properties of this compound and the method of its preparation indicate its structure. However, further proof of its configuration would be desirable, as well as the explanation of the formation of intermediate products during the pyrolysis. ( 5 ) Biltz and Strufe, Ann.. 418, 199-200 (1018-1917); Biltz and Berguis, ibid., 414, 54-57 (1917).