Mendeleev Communications Electronic Version, Issue 5, 2001 1 New reactions of Hünig’s base with S2Cl2: formation of monocyclic 1,2-dithioles Lidia S. Konstantinova,a Oleg A. Rakitin*a and Charles W. Rees*b a N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 119991 Moscow, Russian Federation. Fax: +7 095 135 5328; e-mail: orakitin@ioc.ac.ru b Department of Chemistry, Imperial College of Science, Technology and Medicine, London SW7 2AY, UK. E-mail: c.rees@ic.ac.uk 10.1070/MC2001v011n05ABEH001493 Systematic variation of the ratio of Hünig’s base 1 to S2Cl2 in their reactions in chloroform shows that, in addition to known tricyclic bisdithiolothiazine thiones 4 and 5, monocyclic dithiole-3-thiones 6 and 7 can be isolated; when the inert base DABCO is added the extent of sulfuration of the products is reduced and thiones are replaced by their oxo analogues, and diisopropylamines 15 are converted into monocyclic dithiol-3-ones 16; an overall mechanistic rationalisation of the results, extending earlier work, is presented.We have shown that N-ethyldiisopropylamine (Hünig’s base) 1 and other N-alkyldiisopropylamines are converted in one-pot reactions by disulfur dichloride, S2Cl2, and 1,4-diazabicyclooctane (DABCO) into fully unsaturated tricyclic bis[1,2]dithiolo- [3,4-b][4',3'-e][1,4]thiazines such as 3, 4 and 5.1–3 We proposed a mechanism for this transformation, which involves the conversion of isopropyl groups into 3-chlorodithiolium salts followed by formation of the central 1,4-thiazine ring and final nucleophilic displacement of the reactive 3-chloro groups.1 If the S2Cl2 is in molar excess over DABCO the dithiole rings are chlorinated further to give tricyclic N,N-bis(chlorodithiolyl)- amines, such as 2, as well as tricyclic products, such as 3 (Scheme 1).4 In all of these reactions, both isopropyl groups have been transformed into 1,2-dithiole rings.If only one isopropyl could be so transformed, we would have a new synthesis of monocyclic 3H-1,2-dithioles under unusually mild conditions.5 We therefore systematically investigated the reaction of Hünig’s base 1 with S2Cl2, varying the ratio of the reactants, initially in the absence of another base such as DABCO.The two reactants were mixed in chloroform for 3 days at 0 °C and then quenched by brief heating with formic acid, which gives cleaner reactions by converting the intermediate 1,2-dithiolium salts into 1,2-dithiol-3-ones.1 The four products shown in Scheme 2 were isolated in the yields given in Table 1.When the molar quantities of 1 and S2Cl2 are in ratio 0.9 to 1.2 (Table 1, entries 1–4), the main products are tricyclic bisdithiolothiazines 41 and 5.1 As the ratio of 1 to S2Cl2 increases, there is insufficient S2Cl2 to react with the second isopropyl group in 1 and monocyclic 1,2-dithiole-3-thiones 6† and 7 are indeed formed.In 6 the dithiole has been sulfurated at the 3- and 5-positions but as the relative amount of S2Cl2 decreases further, product 7 of mono-sulfuration appears. Separate experiments showed that 6 and 7 are converted into 4 and 5 in high yield by S2Cl2 (in the presence of DABCO or triethylamine to neutralise the hydrogen chloride liberated). It is striking that all the products in Scheme 2 have thiono groups, in spite of the final quench with formic acid, which usually converts chlorodithiolium intermediates into the keto compounds (see, for example, Scheme 1).This suggests that in the DABCO-free reactions the 3-chlorodithiolium ions are further chlorinated by S2Cl2 to give 3,5-dichloro salts and that both the mono- and dichlorodithiolium salts are attacked by some reactive sulfur nucleophile,‡ which is present in the Hünig’s base reactions without added DABCO, but not in those 1 L 6&O '$%&2 LL +&2+ 1 6 6 6 6 &O &O 2 2 1 6 6 6 6 2 2 6 6FKHPH N S S S S O S 4 S 1 i, S2Cl2, ii, HCO2H N S S S S S S 5 S N S S S 6 HS N S S S 7 Scheme 2 † General procedure for the reactions of tertiary amines with S2Cl2.Disulfur dichloride was added dropwise to a stirred solution of a corresponding amine and DABCO (in the case of N-ethyldiisopropylamine without DABCO) in chloroform at –15–20 °C. The mixture was stirred at 0 °C for 72 h.Formic acid (3.75 ml, 100 mmol) was added; the mixture was refluxed for 1.5 h, and filtered, and the solvents were evaporated. The residue was separated by column chromatography (Merck 60 Silica Gel, light petroleum and then light petroleum–CH2Cl2 mixtures). All new compounds were fully characterised by elemental analysis, 1H and 13C NMR, IR and mass spectra and HMRS. 15a: obtained from the commercial hydrochloride (Aldrich). 15b: an oil prepared from 15a and sodium azide in DMSO at room temperature in 88% yield. 6: yellow crystals, mp 128–130 °C. 1H NMR (CDCl3) d: 1.24 (t, 6H, 2Me, J 6.5 Hz), 1.61 (d, 3H, Me, J 6.5 Hz), 3.55 and 4.78 (2m, 2H, CH2), 4.93 (septet, 1H, CH, J 6.5 Hz), 8.70 (br. s, H, SH). 13C NMR (CDCl3) d: 198.73 (C=S), 193.71 (=C–SH), 133.83 (sp2 tertiary C), 57.75 (CH), 46.32 (CH2), 20.27, 18.97 and 10.76 (3Me).MS, m/z: 251 (M+, 100%), 236 (15), 218 (19), 208 (39). 7: yellow oil. 1H NMR (CDCl3) d: 0.99 (t, 3H, Me, J 7.2 Hz), 1.11 (d, 6H, 2Me, J 6.5 Hz), 3.06 (q, 2H, CH2, J 7.2 Hz), 3.96 (septet, 1H, CH, J 6.5 Hz), 7.71 (s, 1H, CH). 13C NMR (CDCl3) d: 211.89 (C=S), 154.69 (sp2 tertiary C), 141.07 (CH), 51.29 (CH), 40.35 (CH2), 20.15 and 13.43 (2Me).MS, m/z (%): 219 (M+, 69%), 204(36), 176 (38), 142 (31), 128 (28). 16a: yellow oil. 1H NMR (CDCl3) d: 1.11 (d, 6H, 2Me, J 6.5 Hz), 3.31 (m, 1H, CH), 3.39 (s, 4H, 2CH2). 13C NMR (CDCl3) d: 187.96 (C=O), 155.37 and 137.92 (2sp2 tertiary C), 54.65 (CH), 48.89 and 43.71 (2CH2), 22.45 (Me). IR, n/cm–1: 2980 (CH), 1660 (C=O). MS, m/z (%): 275 (M+, 4%), 273 (M+, 16), 271 (M+, 23), 236 (20), 222 (22), 182 (23), 180 (64). 16b: yellow oil. 1H NMR (CDCl3) d: 1.12 (d, 6H, 2Me, J 6.5 Hz), 3.17 and 3.29 (2m, 4H, 2CH2), 3.35 (m, 1H, CH). 13C NMR (CDCl3) d: 187.43 (C=O), 154.87 and 137.02 (2sp2 tertiary C), 54.33 (CH), 50.80 and 44.98 (2CH2), 21.50 (Me). IR, n/cm–1: 2970 (CH), 2120 (N3), 1660 (C=O). MS, m/z (%): 278 (M+, 10%), 222 (69), 180 (100). 16c: yellow oil. 1H NMR (CDCl3) d: 1.15 (d, 6H, 2Me, J 6.5 Hz), 3.51 (septet, 1H, CH, J 6.5 Hz), 4.01 (s, 2H, CH2). 13C NMR (CDCl3) d: 187.17 (C=O), 155.97 and 136.12 (2sp2 tertiary C), 117.15 (CN), 53.59 (CH), 35.92 (CH2), 21.17 (Me). IR, n/cm–1: 2980 (CH), 2140 (CN), 1660 (C=O). MS, m/z (%): 248 (M+, 57%), 233 (43), 212 (32), 206 (55), 179 (36). ‡ This nucleophile is not S8 formed in the reaction since addition of S8 to the initial reaction mixture made no significant difference to the product distribution.Mendeleev Communications Electronic Version, Issue 5, 2001 2 with DABCO.This difference could arise from the greater nucleophilicity of DABCO over Hünig’s base. DABCO could attack S2Cl2 at sulfur to give salt 8, whilst the much more sterically demanding Hünig’s base is more likely to attack S2Cl2 at a terminal chlorine atom to give 9.The latter generates the Cl–S–S– anion, which could rapidly convert a 3-chlorodithiolium salt into the corresponding 3-thione and SCl2. An outline mechanism for the formation of 4 to 7 is suggested in Scheme 3. Some of the early steps have been given in more detail before.1 In the DABCO-containing reaction, electrophilic species 8 could act as a chlorinating agent to give ultimately tetrachloro compound 13, which, in the absence of a strong sulfur nucleophile like Cl–S–S–, would be converted by formic acid into bis-keto compound 2.The formation of monocyclic products 6 and 7 and their conversion into tricyclic products 4 and 5 lends support to our overall mechanism1 in which the dithiole rings are formed first, followed by completion of the central 1,4-thiazine ring to give tricyclic bisdithiolothiazines 3–5.Having established (Scheme 2) that treatment of Hünig’s base with S2Cl2 could provide a route to monocyclic 1,2-dithiole-3- thiones 6 and 7, we tried to enhance the utility of the reactions by replacing that part of starting amine 1 which is acting simply as a base by another more inert tertiary amine, DABCO.Furthermore, the reactions were conducted at 0 °C to minimise conversion of the second isopropyl group. Under these conditions, a new reaction, which involved an attack at the ethyl rather than isopropyl group of Hünig’s base, was observed; however, other substituted diisopropylamines gave monocyclic 1,2-dithioles in low to modest yields among other products.Thus, the treatment of amines 15a, 15b and 15c6 with S2Cl2 (7 equiv.) and DABCO (7 equiv.) in chloroform at 0 °C for 3 days followed by addition of formic acid and heating under reflux for 1.5 h gave 5-chloro- 1,2-dithiol-3-ones 16a (32%), 16b (12%) and 16c (20%), respectively (Scheme 4). In agreement with the proposed mechanism (Scheme 3), the combination of an excess of S2Cl2 over tertiary amine 15 and the presence of DABCO is expected to yield dichlorodithiolium salt (cf. 11 in Scheme 3), which, in the absence of a strong sulfur nucleophile, gives 5-chlorodithiol-3-one 16 rather than dithiole-3-thione. In summary, these experiments throw further light on the likely mechanism of the complex reactions between Hünig’s base and S2Cl2, which have now produced no less than 13 different products.Furthermore, they show that monocyclic 1,2-dithiol- 3-ones and 3-thiones can be added to the bicyclic and tricyclic products already reported for these simple one-pot reactions.1–4 This work was supported by the Russian Foundation for Basic Research (grant no. 99-03-32984a), the Royal Society, MDL Information Systems (UK) Ltd and an RSC Journals Grant to O.A.R., and we thank the Wolfson Foundation for establishing the Wolfson Centre for Organic Chemistry in Medical Science at Imperial College. References 1 (a) C.F. Marcos, C. Polo, O. A. Rakitin, C.W. Rees and T. Torroba, Angew. Chem., Int. Ed. Engl., 1997, 36, 281; (b) C. W. Rees, A. J. P. White, D. J. Williams, O.A. Rakitin, C. F. Marcos, C. Polo and T. Torroba, J. Org. Chem., 1998, 63, 2189. 2 (a) C. F. Marcos, O. A. Rakitin, C. W. Rees, L. I. Souvorova, T. Torroba, A. J. P. White and D. J. Williams, Chem. Commun., 1998, 453; (b) C. W. Rees, A. J. P. White, D. J. Williams, O. A. Rakitin, L. S. Konstantinova, C. F. Marcos and T. Torroba, J. Org. Chem., 1999, 64, 5010. 3 (a) C. F.Marcos, O. A. Rakitin, C. W. Rees, T. Torroba, A. J. P. White and D. J.Williams, Chem. Commun., 1999, 29; (b) L. S. Konstantinova, N. V. Obruchnikova, O. A. Rakitin, C. W. Rees and T. Torroba, J. Chem. Soc., Perkin Trans. 1, 2000, 3421. 4 S. Barriga, L. S. Konstantinova, C. F. Marcos, O. A. Rakitin, C. W. Rees, T. Torroba, A. J. P. White and D. J. Williams, J. Chem. Soc., Perkin Trans. 1, 1999, 2237. 5 C. Th. Pedersen, Adv. Heterocycl. Chem., 1982, 31, 63. 6 D. B. Luten, J. Org. Chem., 1939, 3, 588. Table 1 Reaction of Hünig’s base 1 (x mmol) with S2Cl2 (10 mmol) in CHCl3 (25 ml). Entry Quantity of Hünig’s base 1 (x mmol) Product yield, mmol (%)a aYields are calculated on the basis that 15 mol of Hünig’s base 1 should give 1 mol of 4 or 51 together with 14 mol of the Hünig’s base hydrochloride, and analogously 7 mol of the Hünig’s base should give 1 mol of 6 or 7. 4 5 6 7 1 9 0.20 (33) 0 0 0 2 10 0.14 (21) 0.01 (1.5) 0 0 3 11 0.11 (15) 0.03 (4) 0.04 (2.5) 0 4 12 0.08 (10) 0.06 (7.5) 0.06 (3.5) 0 5 13 0 0.13 (15) 0.12 (6.5) 0 6 15 0 0.12 (12) 0.12 (5.5) 0 7 18 0 0 0.38 (15) 0.08 (3) 8 20 0 0 0.72 (25) 0.23 (8) 9 22 0 0 0.68 (22) 0.32 (10) 1 1 6 6 &O &O 6 6 &O &O1(W3U L N S S Cl 10 1 S2Cl2 N S S S S Cl Cl N S S S 6 HS N S S S 7 Scheme 3 N S2Cl2 12 SSCl N S S Cl 11 Cl S2Cl2 SSCl S2Cl2 N S S S S Cl Cl 13 Cl Cl SSCl N S S S S Cl S 14 Cl Cl SSCl 5 4 SSCl HCOOH 5 1 5 1 6 6 &O 2 L 6&O '$%&2 LL +&2+ D 5 &+&+&O E 5 &+&+1 F 5 &+&1 6FKHPH Received: 4th July 2001; Com. 01/1819