Dianions of 3-oxodithioic acids: Preparation and conversion to 3H-1,2-dithiole-3-thiones

Article abstract of DOI:10.1016/S0040-4039(00)01191-6

Reaction of ketones with CS₂ and 2 equivalents of KH in THF-N,N'-dimethylpropyleneurea solution produces the dianions of 3-oxodithioic acids. These dianions are converted in good yield to 3H-1,2-dithiole-3-thiones by the sequential action of hexamethyldisilathiane and an oxidizing agent such as hexachloroethane. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01191-6

Tetrahedron Letters 41 (2000) 6977±6980
Dianions of 3-oxodithioic acids: preparation and conversion
to 3H-1,2-dithiole-3-thiones
Thomas J. Curpheya,* and Adam H. Libbyb
aDepartment of Pathology, Dartmouth Medical School, Hanover, NH 03755, USA
bDepartment of Chemistry, Dartmouth College, Hanover, NH 03755, USA
Received 6 June 2000; revised 6 July 2000; accepted 7 July 2000
Abstract
Reaction of ketones with CS2 and 2 equivalents of KH in THF±N,N0-dimethylpropyleneurea solution
produces the dianions of 3-oxodithioic acids. These dianions are converted in good yield to 3H-1,2-
dithiole-3-thiones by the sequential action of hexamethyldisilathiane and an oxidizing agent such as hexa-
chloroethane. # 2000 Elsevier Science Ltd. All rights reserved.
Keywords: dithioles; sulfur compounds; sulfur heterocycles; thioacids.
The dithiolethiones (3H-1,2-dithiole-3-thiones, 1) continue to be of interest as potent chemo-
protective agents against a variety of animal models of cancer.1 One representative of this class of
heterocyclic sulfur compounds (Oltipraz, 1, R1=pyrazinyl, R2=methyl) is currently undergoing
human trials in an area of China which has a high incidence of liver cancer.2 Unfortunately,
®nding a general and high-yielding method for synthesis of the dithiolethione ring system remains
a challenging problem. In this connection, some time ago we published a method for
dithiolethione synthesis in which the 3-oxodithioic acids 2 resulting from condensation of CS2
with ketones were converted to dithiolethiones by the action of NCS and hexamethyldisilathiane
(HMDT) in the presence of a catalytic amount of imidazole.3 We presented evidence that this
novel cyclization proceeded via oxidative dimerization of the 3-oxodithioic acids to tetrathianes 3,
which were then converted to dithiolethiones by reaction with HMDT.3 However, in a number of
cases the yields of dithiolethiones obtained were low, in part because isolation of the
corresponding 3-oxodithioic acids was accompanied by severe decomposition and losses.4 In a
continuing e€ort to improve the yields obtained by this sequence, we have investigated the
formation and reactions of dianions 4 of the 3-oxodithioic acids. This work has now led to an
improved synthesis of dithiolethiones 1 by a route which does not appear to involve the
intermediacy of tetrathianes 3 and which is the subject of this communication.
* Corresponding author.
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01191-6

6978
In a model experiment, reaction of acetophenone with CS2 in the presence of 2 equivalents of
KH led to the evolution of 2 equivalents of hydrogen as measured gasometrically. Addition of
excess methyl iodide then produced ketenedithioacetal 5 in 99% chromatographic yield without
any additional gas evolution. This experiment is clearly consistent only with the formation of
dianion 4 prior to addition of the electrophile. In order to obtain complete conversion to dianion
4 in a THF solution it was found necessary to use a dipolar aprotic cosolvent, either HMPA or
N,N0-dimethylpropyleneurea (DMPU). This is most likely due to the low solubility of mono-
potassium salts such as 6 which hinders further reaction with an insoluble base like KH. The need
for a dipolar aprotic solvent to achieve complete conversion to the dianion 4 is likely not due to
the weak acidity of the ®rst formed monoanion. Sulfur acids in general show a greater acidity
than their oxygen analogues, and the pKa2 of acetoacetic acid (in water) is reported to be 14.1.5
Initial attempts to oxidize the dianions 4 to tetrathianes, which could then be converted to
dithiolethiones with HMDT, gave results no better than those obtained previously.3 However,
yields improved dramatically when the dianion 4 was ®rst treated with HMDT and then with
oxidant. The most generally satisfactory oxidant was found to be hexachloroethane, although
yields using bromine or iodine were similar or only slightly lower. The overall transformation is
shown by Eq. (1) and the results obtained with a range of ketone substrates are given in Table 1.6
In general, the yields ranged from good to essentially quantitative, with aralkyl ketones (entries
1±4) tending to give higher yields than dialkyl ketones (entries 6±10). In all cases the yields were
superior to those obtained by the previously published two-step procedure involving isolation
and oxidative cyclization of the free 3-oxodithioic acid 2.3
ꢀ1†
While certain details remain obscure, we believe that the transformation of dianion 4 to
dithiolethione 1 proceeds by the sequence of steps shown in Scheme 1. Direct con®rmation of the
®rst step in this scheme was possible using pinacolone as the ketone, since the corresponding
dianion 4, R1=t-Bu, R2=H, was soluble enough in THF±DMPU to allow NMR spectra to be

6979
Table 1
Conversion of ketones to 3H-1,2-dithiolethiones by Eq. (1)a
Scheme 1.
taken and the course of the reaction with HMDT to be observed directly. The dianion 4 from
pinacolone showed the expected two singlets in its 1H NMR spectrum at ꢀ 1.0 (9H, t-Bu) and 7.0
(1H, vinyl hydrogen). When 1 equivalent of HMDT was added to the dianion solution, the vinyl
hydrogen resonance shifted to ꢀ 6.1 and two new resonances appeared in the spectrum at ꢀ 0.20
(9H) and 0.05 (9H). We attribute these to the TMS groups of enol ether 7 and thiolate 8,
respectively, since authentic thiolate 8 (generated from HMDT, KH, and water) gave a peak at ꢀ
0.05 in this solvent mixture. Furthermore, the same set of peaks minus the one at ꢀ 0.05 was
generated when the dianion was quenched with 1 equivalent of TMS chloride in place of
HMDT.8 Addition of hexachloroethane or iodine to the dianion plus HMDT mixture brought
about a rapid conversion to dithiolethione 1, R1=t-Bu, R2=H, without any evidence for the
formation of tetrathiane 3 either by NMR or, in a separate experiment, by HPLC. In the absence
of evidence for any intermediates, we believe that the further transformation of the mixture of 7
and 8 into dithiolethione 1 upon addition of oxidant is most simply rationalized as proceeding via
the intermediacy of the mixed disul®de 9. Formation of the disul®de bond in 9 probably involves
initial attack of the oxidizing agent on the more nucleophilic thiolate 8 to form a sulfenylating

6980
species such as Me3SiSCl, which then reacts with 7 to form 9. We propose that 9 is then converted
to dithiolethione 1 by migration of a TMS group from oxygen to sulfur giving the bis-S-silylated
intermediate 10, whose ring-closure by addition of an Si±S bond across the carbonyl group leads
to intermediate 11. Elimination of the elements of hexamethyldisiloxane from 11 then gives the
dithiolethione 1.
In conclusion, we have developed a simple one-pot procedure for conversion in good yield of
ketones to dithiolethiones via the intermediacy of the dianions 4 derived from 3-oxodithioic acids.
The further synthetic utility of dianions 4 is under investigation and will be reported in due
course.
Acknowledgements
This work was supported by a grant from the National Institutes of Health (CA 39416).
References
1. Kensler, T. W.; Groopman, J. D.; Roebuck, B. D.; Curphey, T. J. ACS Symp. Ser. 1994, 546, 154±163.
2. Wang, J.-S.; Shen, X.; He, X.; Zhu, Y.-R.; Zhang, B.-C.; Wang, J.-B.; Qian, G.-S.; Kuang, S.-Y.; Zarba, A.;
Egner, P. A.; Jacobson, L. P.; Munoz, A.; Helzlsouer, K. J.; Groopman, J. D.; Kensler, T. W. J. Natl. Cancer Inst.
1999, 91, 347±354.
3. Curphey, T. J.; Joyner, H. H. Tetrahedron Lett. 1993, 34, 7231±7234.
4. Curphey, T. J.; Joyner, H. H.; Libby, A. H., unpublished observations.
5. Olander, A. Z. Phys. Chem. 1929, 144, 73±117; Olander, A. Z. Phys. Chem. 1930, 146, 406.
6. CAUTION: Because of the noxious odor and probable toxicity of HMDT and because ¯ammable hydrogen gas is
evolved, this procedure should be conducted only in a well-ventilated fume hood. Representative procedure: To a
well-stirred suspension of KH (2.06 g, 51.3 mmol) in dry THF (50 mL) and dry DMPU (25 mL) was added under
an atmosphere of dry argon a solution of 40-methoxyacetophenone (3.75 g, 25 mmol) in dry THF (7 mL) at a rate
sucient to maintain rapid evolution of hydrogen. The enolate suspension was stirred for an additional 15 min
after gas evolution had ceased, then a solution of carbon disul®de (2.09 g, 27.5 mmol) in dry THF (12 mL) plus
dry DMPU (6 mL) was added at a rate such that hydrogen was continually evolved. The resulting red solution was
stirred for 10 min and HMDT (6.69 g, 37.5 mmol) was then added. After stirring an additional 20 min, the mixture
was cooled to 0ꢁC, treated with a solution of hexachloroethane (5.92 g, 25 mmol) in dry THF (15 mL), and stirred
for 30 min more. MeOH (10 mL) was added cautiously to destroy any unreacted hydride and the mixture allowed
to stand for 15 min. THF was removed in vacuo and the crude dithiolethione was precipitated with an excess of
water. TLC (silica gel, 10% ethyl acetate in hexane) of this material showed only minor amounts of impurities.
Recrystallization of the dried product from CCl4 gave 5-(4-methoxyphenyl)-3H-1,2-dithiole-3-thione (5.06 g, 84%)
1, R1=4-MeOPh, R2=H, as a crystalline orange solid, mp 107.5±109ꢁC (lit.7 mp 109ꢁC).
7. Thuillier, A.; Vialle, J. Bull. Soc. Chim. Fr. 1959, 1398±1401.
8. We cannot rule out an alternative structure for silylated monoanion 7 in which the TMS group resides on sulfur
rather than oxygen. Indeed, these two forms of the silylated monoanion may well be in equilibrium with each
other via O to S silyl group migration. However, the well-known anity of silicon for oxygen, as well as the fact
that the alternative S-silylated monoanion would be the conjugate base of a weaker acid, argues in favor of the O-
silylated structure 7. The exact position of the TMS group is not crucial to the proposed mechanism, since either
the O-silylated or the S-silylated monoanion may serve as a precursor to 10.