h g . 20, 1937
~ ~ - ( ~ - ~ I ~ ~ E T H Y L A V I N O P R O P Y L ) - ~ - ( ~ R I F L U O R O 1 L I E T H Y L ) - P H E N O T H I A Z I r * ; E H4375 CI
m.p. 149-151°, and 0.28 g. (15%) of crude tetraphenylsilane, m.p. 193-205', which after recrystallization from benzene gave 0.18 g. (10%) of pure tetraphenylsilane, m.p. 230-2320, all products identified by mixed melting point.
Achowledgment.-The author is indebted to N . Wolfish and M. Gold for technical assistance, t o
Dr. Henry Gilman and the Institute for Atomic State college, for the infrared Research, spectrum and to the National Research Council of Canada for a grant supporting this research. TOROXTO, CANADA
[CONTRIBUTION FROM THE SQUIBB IXSTITUTE FOR LfEDICAL RESEARCH]
10-(3-Dimethylaminopropyl)-2-(Trifluoromethyl)-phenothiazine Hydrochloride (VESPRIN') and Related Compounds. I B Y HARRY L. I ~ A LFRANCIS E, SOWIh3KI AXD
JACK
BERNSTEIN
RECEIVED MARCH 18, 1957 The thiotiation of 3-(trifluoromethyl)-diphenylamine has yielded 2- and 4-(trifluoromethyl) -phenothiazine, Thesc two nuclei have been allowed t o react with various dialkylaminoalkyl chlorides in the presence of sodamide to give 10-dialkylaminoalkyl-2- and 4-(trifluoromethyl) -phenothiazine. Hydrogen pcroside has converted these compounds t o various oxygenated derivatives. An improved procedure is described for the preparation of 3-(trifluoromethyl)-phenothiazine. Since this nucleus did not react with dimethylaminopropyl chloride in the presence of sodamide, an indirect procedure was used to prepare the 10-(3-dimethylaminopropyl)derivative.
stituted phenothiazines ; the less soluble, higher melting isomer (m.p. 187-188") when heated with copper bronze gave 2-methylcarbazole, while the more soluble, lower melting isomer (m.p. 114-1 15') gave 4-methylcarbazole. Lye have re-examined the thionation of 3-(trifluoromethyl)-diphenylamine, and have isolated both 2-(trifluoromethy1)-phenothiazine,1n.p. 1hh1 8 9 O , in 45yo yield, and the hitherto unreported 4-(trifluoromethyl) -phenothiazine, n1.p. 2-73' in 32% yield. It is of interest that the latter isomer showed a strong band a t 12.7 p , as anticipated, since it corresponds t o a viciizal substituted benzene. 3-(Trifluoromethy1)-phenothiazinewas prepared in one step in 52y0yield by Roe and Little' via the Smiles rearrangement of 2-formamido-2'-1iitro-4'(trifluoromethy1)-diphenyl sulfide. Based on our own experience, we prefer the two step modification of the Smiles rearrangement, which employed the 2-ucetumido derivative and which gave a 91qoyield over-all. The reactions of 2- and 4-(trifluoroniethyl) phenothiazine with various dialkylaminoethyl chlorides and dialkylarninopropyl chlorides in toluene or xylene, in the presence of sodarnide, gave the desired 10-substituted derivatives in good yields. The reaction of 3-(trifluoromethyl)-phenothiazine with dimethylaminopropyl chloride under the same conditions was unsuccessful. Instead of the usual dark brown colored reaction mixtures, deep violet colored solutions were obtained, which, even after prolonged reflux periods, when cooled and hydrolyzed with water became colorless and when (1) T ' E S P R I S is a trademark of the Olin Slathieson Chemical Corgoratinn. worked up in the usual manner yielded only un( 2 ) 11. I,. Yale. THISJ O U R N A L , 77, 2270 (I%%). 3-(trifluoromethy1)-phenothiazine. It bereacted (3) T h e tranquilizing activity 10~(X-dimethylaminoprop~l)-2came necessary t o use the indirect procedure out(trifluoromethy1)-phenothiazine hydrochloride was reported by J. C. lined below which gave the desired product. Burke, H. L. Yale, 0 . Id. Haseert and J. P.High and b y J. J. Piala, 1. P. High. K. Greenspan and I. C. Burke at t h e 19.5ii Meeting of T h e 2 - (Trifluoromethvl) - 10 - phenothiazinepropioniAmerican Society lor Pharmaculugy and ICxperimental Therapeutics a t trile' and lithium aluminum hydride gave 10French Lick Spriugs, Indiana S o l e r n h e r 8-10, 1956. (3-arninopropyl) -2-(trifluoromethyl) -phenothiazine. (1) S . L. Smith, J . Ova. Chem., 15, 1125 (1950). The preparation of 10-(4-dimethylarninobutyl)-2( 5 ) A. Roe and W. F. Little, ibid , 2 0 , 1577 (19551, have discussed the infrared spectra of this compound as well as a number of other (trifluoromethy1)-phenothiazine followed the sefluorine substituted phenothiazines. quence of reactions shown below, employing 4( G ) P. Charpentier. P. Gailliot, R. Jacob. J Oaudechon and J. We have for several years been interested in variously substituted phenothiazine derivatives2 and wish now to report on a series of 10-dialkylaminoalkyl-2-, -3- and -4-(trifluoromethy1)-phenothiazines and their oxygenated derivatives. Several of these compounds have shown good specificity and high potency as ataractic agents in laboratory animals3 and are now undergoing clinical evaluation. The thionation of 3-(trifluoromethy1)-diphenylamine should lead to the formation of both 2-(trifluoromethy1)-phenothiazine and 4-(trifluoromethyl)-phenothiazine. From such a reaction, however, Smith4 isolated only one (trifluoromethy1)phenothiazine. This compound showed a deep band in the infrared a t 12.17 p , an absorption band not seen in the spectrum of phenothiazine itself. Since the infrared spectra of asymmetrical trisubstituted benzenes show a characteristic deep band in the region 12.0-12.5 p while vicinal trisubstituted benzenes show a characteristic deep band in the region 12.5-13.15 p , Smith concluded that his product was 2-(trifluoromethyl) -phenothiazine. This method of differentiating isomers in monosubstituted phenothiazines received additional support in the observation of Roe and Little5 t h a t 3-(trifluoromethyl) -phenothiazine, prepared by an unambiguous synthesis, showed a deep band in the infrared a t 12.2 p . The work of Charpentier'j has clenionstrated that the thionation of 3-substituted diphenylamines, e.g., 3-rnethyldiphenylainine, leads to 2- and 4-sub-
(if
Buisso, C o m p f . r e n d . , 235, 59 (1952)
(7) A'. L. Smith, J OUE C h ~ i n, 1 6 , 415 (1951)
€I A
/s\
chlorobutyl-"tetrahydropyranyl ether as the side chain precursor; several of the intermediates 71-ere high boiling viscous oils which were difficult to obtain analytically pure, although the final product,
phenothiazine some cleavage to 2-(trifluoromethyl)phenothiazine occurred. This is not surprising since Charpentiergfound t h a t 1O-('-dimethylatiiinopropyl) -phenothiazine and boiling 48YG hydro-
TABLE
I
ln-DIALKYL.4~INO.4LKYL-2-( TKIFLUORO.\IET~IYL)-PHESO7.HI.4ZINESA S D TIiEIR DERIVATIVES Analyses, c1 Yield, Roiling point Calcd. Found Side chain Mol. formula c; oc. MIU. c H s H
c
A-
12
81 93 20
82 50 73 -(CH2)3---S' ' \CH?AH?
100
Salts 1101.formula
hf.p, " C .
I S
Ci7Hi7F3NzS.HCle
c~~H~~F~s~s.Hc~~J
70 ti1 11 48 74
CisH1yF3XzS.HClB C1yH2iF3SzS.HClh CiyH2iF3SzS~HCli CmH23F3S2S.HC1" 79 C ~ H ~ ~ F ~ K ~ S . C Z H Z O ~ ~1i)o CZOHZIFIYZS~HCI' 22
I . 17 1 32 ,?, 18 $5 18 5 5fl
Itjl-lti2 245-2iG 173-174 221-226 tiec. 133- 134
5;i. 25 54.76 5 j 39 55.39 ,56.61
1fjiJ-103
;iii.iil L 5 . 5 0 ,j7,t i % 3 . ,5B
163-165 192-194 tlec. 17X - 1 i 1
.?;.XR
5:;4
1 70
i .82
4.08
?.:-l+
nitrilc.
and 10 rnl. of 207, aqueous sodium hydroxide. The ctlier solution ~ r a sseparated, dried, concentrated and distillet1 to give 9.3 g. (3:3[;6 yield) of 10-(3-aminopropyl)-3-(trifluoromethy1)-phenothiazine. T o the 3-aminopropyl derivative, 9.2 g. (0.028 mole) in 14.7 g. (0.28 mole) of formic acid, was added 7 g. (0.086 mole) of 377'formalin and the mixture heated in an oil-bath a t YO-lOOo for 8 hr. T h e mixture W;Lcooled, filtered arid the filtrate treated with 30.4 nil. o f 0.955 LVaqueous hydrochloric acid. T h e mixture was concentrated t o dryness in uacuo, the residue was dissolvcd ili 100 ml. of water and then made alkaline with an excess of 20", aqueous sodium hydroxide. The base was isolated, distilled arid converted to the hydrochloride in the manner described above. lo-( 4-Dimethylamino butyl)-2-( trifluoromethy1)-phenothiazine Hydrochloride. 10- [4'-(2"-Tetrahydropyrano1oxy)1'-butyl .--I mixture of - -1-2-(trifluoromethvl)-~henothiazine . 66.8 g. (0,.25 mole) of -2-(frifluorotnethyl)-phenothiazinc, 11.7 g. (0.30 mole) of sudamide a n d 750 ml. of dry syleiic were refluxed for 2 hr. The reaction misture was c i ~ o l e d , and a solution of 58.6 g. (0.30 mole) of 2-tetrahydropyranyl4-chlorobutL-1 ether8 in 250 ml. of dry xylene was added tiropiriie. T h e mixture was then reflused for 3.5 hr., filtered hot aiid the xylene removed in V Q C I L O . T h e residue !vas extracted with 500 i d . of ether, t h e filtered ether estracts \\-ere concentrated and the residue distilled t o give 97.8 g. (06'/; yield) of crude product, b.p. 200-203° (0.2 nun.). . I m 1 Calcd. for C ? Z I H ~ ~ F ~ S O C,S 64.84; : H, 5.93. Found: C , 62.44; 13, 5.67. 10-(4-Hydroxybutyl)-2-(trifluoromethyl)phenothiazine.--.i solution of 97.8 g. (0.24 mole) of the pyranyl derivative, 2
