15151'
i'
1,.
~$VTLI.:R AND
ALICEC,. KENFREW
Linc sulfide from sodium mercuric sulfide solu1 ions which are uiidersaturatcd wit11 rntm-uric sulfide. Such a postprecipitatioti occurs onlv when the mercuric' sulfide is v)lublc. i n the pri mary precipitate, arid the speed 01 separatioii of the latter from solution is greater than the speed of dissociatiori of the HgS? io11 mto HgS ,lllri
s
2. The distribution coeficient of niercuric
wlfide between aqueous solutioii and solid has 1,rcii determiiied a t 1'3 aiid hOo. The solid solution does not behave as an ideal solution, but the values of the distributioii coefficient are of the xuiic ordpr of magnitude as tlic x-aluc calculatcd
Yol. 60
upon the basis of formation of ideal solutions. 'This calculated value is iridepmdent of the nature of the solid, as long as the latter acts as a solvent frir mercuric sulfide. The speed of attainment of distribution cyilibrium betrveeii sodium mercuric sulfide solution and zinc sulfide is a good indicator of the degree of perfection ant! the progress of aging of the latter. I t is shown that zinc sulfide precipitated at a pH of 1.1.5 IS more perfect and ages more rapidly a t this pH than the sulfide formed and aged a t a PH of 3 and the latter more so than the sulfide formed and aqed in alkaline medium. A f t VXI.
AIWLIS,
hlr\
\
RECEIVED APRIL14, l % B
[CONTRIBrTION I.'iZON fHL DBPARTMEN r OF REbSARCH IN P U R E CHEMISTRY, MELLON INSTITUTE OF INDUSTRIAL RESEARCH]
Hydroxyalkyl Ethers of Basic Phenols. The Antipneumococcic Activity of Some 8-Quholyl Ethers BY C. I,. RUTLERAKD AIXE C . KENFREW I t appeared to be of considerable general interest to extend the application of the hydroxyalkylation method used in the cinchona alkaloid field' to other types of basic phenolic substances. More specifically, it was hoped that studies carried out on simpler basic pheiiols would gi\-e a partial explanation, a t least, of the low yields obtained iii attempts to hydroxyethylate phenolic cinchona alkaloids with ethylene chlorohydrin.' Further, the preparation of simpler substaticcs, related structurally to these alkaloids, was of interest ill the chemotherapeutic study of piicumonia. Esperiments undertaken with these considerations in view are described in the present report. Phenol, aniline, p-aminophenol, p-acetaminophenol, nr-diethylamiiiophenol arid S-hydrosyquinc)line were chosen as suitable starting materials for the iiivestigatioiis The compounds were alkylated in the usual way with benzyloxyalkyl p-toluenesulfonates" and the resulting aminoaryl beiizyloxyalkyl ethers were hydrolyLed in dilute hydrochloric acid to hydroxyalkyl derivatives. Yields were high in iiearly all cases. The reactioii with p-aminophenol was complicated by iiitrogeri alkylatioii. p-Acetaminophenol, however, gave with benzyloxyethyl p-tolueiiesul, I ) Butler aiid Kenfrm, 'IHIYJOURNAL, 60, 1473 (1933) ( 2 ) Butler, Rcnfxew, Cretcher and Souther, h i d , 69, 227 (1937) ( 3 ) Butler Renfrew arid Ciapp, d i d , 60, 1472 (19
fotiatr, a high yield of the benzyloxyalkyl ether, which was readily converted to P-hydroxyphenetidine. Hydroxyalkylation of nz-diethylaminophenol and 8-hydroxyquinoline was accomplished without difficulty. I t seemed possible that the failure of ethylene chlorohydrin to hydroxyethylate phenolic cinchona alkaloids2 might be due t o the presence 111 the cinchona structure of basic groups, with which this substance reacted with greater ease than with the phenolic group; or by which it was destroyed before it could react. Several alkylatioils with ethylene chlorohydriii were carried out 111 attempts to get further information on this point According to Rindfusz4 a 50% yield CJf hydroxyethyl phenyl ether is obtained on alkylation of sodium phenolate with this reagent. In the present work the reaction was carried out in the presence oi one molecular equivalent of triethylamine. Only 13c/, of the theoretical quaiitity of hydroxyethyl phenyl ether was obtained in this experiment. The presence of triethylamine, however, did riot decrease the yield of alkylation product when bciizyloxyethyl P-tolueiiesulfonate was used as alkylating reagent. %Hydroxyquinoline, on alkylation with ethyletie chlorohydrin, gave only 19yoof the theoretical amount of (4) liindfusz, i b t d , 41, 669 (IOIQ) see d5v Dentley, Haworth and I'erkin, . IChmn i o < , 69, 164 (1896)
ANTIPNEUMOCOCCIC ACTIVITY OF 8-QUINOLYL ETHERS
July, 1938
TABLE I ANTIPNEUMONIC ACTIVITY AND TOXICITY OF ~-QUINOLYL ETHERS 7-
Substance
--Bactericidal Growth
power in Ditroa-
Dilution
1 mg.
1 oxicityo 20 g . mice; deaths a t dosages of 2 mg. 3 mg. 7 mg. 8 mg.
1583
10 mg.
Complete inhibition 1:400000 8/30 30/30 30/30 8-Quinolyl Ether Hydrochlorides 1:50000 2/10 13/30 Ethyl Slight inhibition 1: 100000 0/30 0/10 0/30 P-Hydroxyethyl Slight inhibition y-Hydroxypropyl No inhibition 1:50000 0/10 a-Methyl P-hydrdxyethyl No inhibition 1:50000 0/30 0/10 0/10 2/30 ‘Detailed results will be published by the medical staff associated with the Mellon Institute in the chemotherapeutic study of pneumonia. 8-Hydroxyquinoline sulfate
hydroxyethyl 8-quinolyl ether, as compared with the 50% yield obtained by Rindfusz4 in the hydroxyethylation of phenol, where there could be no question of complication with basic nitrogen groups. Hydroxyethyl 8-quinolyl ether was obtained in 67% yield, using benzyloxyethyl p toluenesulfonate. m-Diethylaminophenol gave practically identical quantities of hydroxyethyl ether with the two reagents. It is believed that these results confirm the opinion that the unfavorable yields reported in the earlier work on hydroxyalkylation of cinchona alkaloids with ethylene chlorohydrin2 were due to complications involving the two nitrogen atoms of the alkaloids. The difficulty can be avoided to a very large extent by the use of the benzyloxyalkyl aromatic sulfonates for alkylations of this type. The preparation of the quinoline derivatives had an added interest because of the structural relationship of these compounds to the cinchona alkaloids, and because of the marked germicidal activity of many substances of this type.5 Hirschfelder and co-workerss found that 8-hydroxyquinoline sulfate was highly active in vitro against the pneumococcus. In spite of this observation the possibility of the usefulness of simplei quinoline derivatives as antipneumococcic agents has received very little attention. More recently, a series of eighteen quinoline derivatives was investigated by BUhrmann7 and several substances derived from 2-phenyl-4-amino- and 4-amino-6hydroxyquiiiolines, were found which inhibited the growth of pneumococci. In view of Hirschfelder’s results,6 it seemed advisable to have further tests run 011 S-hydroxyquinoline aiid the S-quinolyl ethers prepared ill the course of this nnrk. Interest i i i these compounds was enhanced ( 5 ) Von nettingen, “7 herapeutiL Agent> of the Quinoline Group,” rhe Chemical Catalog Company, Inc , New York, 1933 (6) Hirsehfelder, Jensen and Swanson, P70c Soc. Exfill B i d M i d , 20, 402 (1923) ( 7 ) Buhrmann, Z Immunrta(sfurschuirg. 84, 300 (1935)
by the suggestion that 8-hydroxyquinoline may stimulate the defense mechanism of the body5 (p. 33). In Table I, both the mouse toxicity and the pneumococcicidal activity are seen to be reduced sharply by,alkylation of the phenolic hydroxyl group. Some pneumococcicidal activity is maintained in the ethyl and hydroxyethyl derivatives. Further biological examination of these substances would appear to be desirable, especially in view of their very low toxicities. Experimental Half molecular quantities of the compounds to be alkylated were converted to potassium salts with the calculated amount of potassium hydroxide in about 300 cc. of absolute alcoholic solution. The alkylating reagent was then added and the mixture was refluxed for two to two and onehalf hours on a water-bath. When ethylene chlorohydrin was used as alkylating reagent, the reaction mixtures were worked up directly in the usual way for hydroxyethyl ether. The crude reaction products from the benzyloxyalkyl p-toluenesulfonate alkylations were isolated by evaporating the solvent from the filtered alcoholic solution and separating from alkali-soluble material. The benzyl derivatives were hydrolyzed in dilute hydrochloric acid to hydroxyalkyl ethers as previously described1 and the products were worked up and purified by the usual methods. Phenoxyethyl Benzyl Ether.-The ether was obtained as an oil on alkylation of phenol with benzyloxyethyl p-toluenesulfonate, The product was purified by fractional distillation; yield 67%; b. p. 175’ a t 3 mm. I n a second experiment, conditions were altered by adding to the alkylation mixture one equivalent of triethylamine. The presence of this base did not interfere with the desired reaction, as was evidenced by the high yield (78% of the theoretical) of ether obtained. Anal. Calcd. for ClsHl8Oz: C, 78.9; H, 7.1. Found: C, 79.05; H, 6.9. 8-Hydroxyethyl Phenyl Ether.eThe alkylation of phenol with cthylenc chlorohydrin was carried out in the usual way rxcrpt that one cquivalenl. of triethylamine was added to the reaction inisturc. The yield of fihydroxyethyl phenyl ether was only 12.5%, as compared with the 5Oy0yield obtained by Rindfusz.‘ p-Hydroxyethy1aniline.-Aniline was alkylated with benzyloxyethyl p-toluenesulfonate and the crude benzyl-
o\qethylanilinc wa5 Iiytlrol ed in 11 9" l r 1 #jhytho\vc t I I r l a n l l l n t eld jOTo 1
b
ii) I:
drochloric acidi 11 157 1.58' at
illill
i\poi tion
8-Hydroxyethyl &Quinolyl Ether. -8-Hydroxyquinoline wa? alkylated with bmiyloxychthyl p-toluenesulfonate, dntl 9 he CI iidc beriLylohyethyl 8-quinolyl ether was hydrolvzed to @-hydrouyethyl8-quinolyl ether in dilute hydrochloric acid The subhtance was purified by crystallizaitoil of it5 hydrochloride from absolute alcohol, yield 70%,
of thc I~ydroxqethylanillii~ was converted to N-benLoxjethy1 htnzanilidt~with benzoyl chloridc, in 80% yield The inelting po:n: wac 91 92', whtch agrecs with in p 189-500" the figure found b y Schorlgin and Below ' Alkylation of p-Aminophenol with Benzyloxyethyl d4nal Calcd for C:IH:IO~NHCl C1, 16 T Found p-Tolueneaulbonate. I h t onlv product which could be i ' l . 1.78 iu,latcd in this e.ipttntitrrt wa4 a crude trialkylated subThe base was recovered froin the hydrochloride and ivaiice 01 probable ,trtti turt C,H ~ C H ~ ~ ) C € ~ * C ~ ~ Pcry\talhzed ( ~ C * H ~from ~ - alcohol, ni p 23-84" (CHpCH20CH,C6H,), A small amount of this material Anal Calcd for Cl:HtlOPN C, 69 8 , H, 5.9; N, 7 4 H, ti 1, N, 7 .i . I d G l c d fu: Cz 1-1 ,N*') ;i H,SO, so,t crvstdli/atioii of it.; hydrochloride from alcohol; yield stalh/atioii from aliwlutc alcohol yicltl %%,, in 1)
Anal Calcd for Cli&&3h C, 71 t), H, 0.7 Fou~id C, 71 3 , H, 7 0 B-Hy~~'oxgp~netiBine.-BeIizyloxyethyl p-acetaminophenyl ether was hydrolyzed 111 11% hydrochloric acid in thc usual way1 to j-liydrouqpheii~tidmc, yield 80% h 4 1 Anal Calcd i i ) i LJIL,U-\ \ , ( I i C ' , (189 , 11, 9 I l'rrund. li, 7 0 , L , $19.i.ET, 9 o An acetyl derivativr was obtained as a V I I C O U ~ liquid 11 ith acetic anhydrId{~ and \odium acetate
Ill
-hl 7" Airid
Calcd for G2Hj3O2NHCI C1, 14 8 . N, 5 8 I'ound C1, 14B, N , 6 3 The haw was liberated from the hydrochlarih and crystallized from ether, 11) p 65". With acetyl chloride it gave a zoltd acetyl derivative, which was crystallized from acetone, ni p 99" 4 n d Calcd lor CL,HI,O~N. CHsCO, 17 5 Found CI-I,CO, 17 0, 18 1 *!-Hydroxy-n-propyl 8-quinolyl ethcr was siniilarly purziied. Field SU%, 111 p 129" . I d Hydrochloride Calcd for CI~HI,&N.HCI C1, Rase Calcd for C1iHisOaN
-
Acetory-n propyl 3-quinoiyl ether was obtained a? a reddish oil on acetylation with acetic anhydride and SOdium itcetate .Ind Calcd for C I ~ H I ~ OCH,CO, ~N 17 5 Pound CH,CO, I7 5, 17 6 8-Ethoxyquinoline.--8-BydroxyquinoLme was alkylated in t h e uiual way with ethyl p-toluenesulfonate, b. p 178' d t 28 tnm 1" The base was converted to hydrochloride, w hich M a i crystallized from alcohol Atiril Calrd for CIIHIIONHCI C1, 1 6 9 Found
c1,
lbcl
Summary The preparation of several hydroxyalkyl ethers (81 basic phenols has been described. Surne 1l'i
I rxlier and Renotif
R,I
11
;ill
11884)
July, 193X
KINETICS OF' THE ISOMERIZATION OF CAMPHENE HYDROCHLORIDE
physiological properties of hydroxyalkyl 8-quinolyl ethers having a bearing on the chemotherapy
[CONTRIBUTION FROM
THE
of
1585
pneumonia have been presented briefly.
PITTSBURGH,
PENNA.
RECEIVEB MAY19, 1938
CONVERSE MEMORIAL LABORATORY OF HARVARD UNIVERSITY]
The Wagner-Meerwein Rearrangement. A Kinetic Reinvestigation of the Isomerization of Camphene Hydrochloride' BY PAUL D. BARTLETT AND IRVING POCKEL About a year ago we reviewed the evidence2 that the rearrangement of camphene hydrochloride (I) into isobornyl chloride (11), and of pinene hydrochloride (111) into bornyl chloride (IV) involves complete Walden inversion a t carbon atom no. 2 of the camphane ring system (the point of attachment of the chlorine in bornyl and isobornyl chlorides). Analogy with well-known cases of CH3
CH3
CCI
C
I
-
H
H
1
I1
H I11
H IV
Walden inversion led us to propose that this rearrangement was a collision process, involving donors ur acceptors of chloride ions, or both. Such a potential donor and acceptor is always present in solutions of camphene hydrochloride, since this compound dissociates rapidly and reversibly into camphene and hydrogen chloride, and this e q d i b rium is attained more rapidly than the rearrangement occurs. If hydrogen chloride played an essential part in the rearrangement, the reaction could not be a spontaneous, monomolecular one as reported by Meerwein and van E m ~ t e r . Since ~ these authurs had not considered higher orders of readiun, we (1) Most of this material was included in a paper presented at t h e Organic Symposium a t Richmond, Va., on December 30, 1937 (2) Bartlett and PBckel, THISJOURNAL, 69, 820 (1937) (3) Meerwein and van Emster, Brr , S I B , 2300 (1922)
recalculated their data in terms of first, second and three-halves order equations and found that in six of the nine solvents studied the data were best fitted by the formulation of the second order with respect to camphene hydrochloride, while in the remaining three solvents the three-halves order formulation was best. This left it quite uncertain in what manner hydrogen chloride entered into the rearrangement process. We have now carried out a number of kinetic experiments designed to provide clear evidence of the order of the reaction. Nitrobenzene was chosen as a solvent because it allows the rearrangement to proceed a t the most convenient rate for measurement. The results show clearly that the rearrangement involves one molecule of camphene hydrochloride and one of hydrogen chloride, and they provide a complete explanation for the apparent variation of the order of the reaction with change of solvent under the conditions of measurement used by Meenvein and van Emster and, initially, by ourselves. The dissociation equilibrium Camphene hydrochloride J_ Camphene HCl (1) lies, in the case of most of our solutions, more than 94% to the left. With pure camphene hydrochloride, the hydrogen chloride concentration in the solution will be proportional to the square root of the camphene hydrochloride concentration, and hence the rate of reaction is proportional to the 3//z power of the camphene hydrochloride concentration. However, when a large excess of camphene is present, the amount of hydragen chloride a t equilibrium is diminished and its concentration becomes proportional to that of the camphene hydrochloride. The reaction under these conditions is slowed down and becomes apparently bimolecular with respect to camphene hydrochloride. The explanation of the fact that the rearrangement seems t o be sometimes second and sometimes 3/2 order is that it is almost im-
+
