Aminosilanetrithiol RSi(SH)₃: an experimental and quantum-chemical study

Abstract

An interesting aminosilanetrithiol RSi(SH)₃ (R = N(SiMe₃)-2,6-iPr₂C₆H₃) has been prepared by the reaction of lithium aminosilanetrithiolate {RSi[SLi(THF)]₃}₂ with MeCOOH. Theoretical calculations indicate that the LP(N) → σ*(Si–S) and LP(S) → σ*(Si–S) electron donations remarkably contribute to the stabilization of the Si(SH)₃ part of the molecule. RSi(SH)₃ is the first example of a stable molecule containing three SH groups attached to one element.

---

Orthoformic acid (HC(OH)₃) and its sulfur congener (HC(SH)₃) are hypothetical molecules. Aqueous formic acid possibly contains HC(OH)₃ which is considered to be extremely unstable.¹ The analogous silicon species HSi(OH)₃ and HSi(SH)₃ have also been proposed and theoretically studied.² RSi(OH)₃ compounds with the bulky R substituents (R = alkyl, aryl, aryloxy, or aryl-substituted amide group or a metal cluster) have already been reported since the 1950s.³ They are extensively employed as building blocks for assembling lipophilic three-dimensional Si–O metal clusters.^3b–d However, the sulfur analogue RSi(SH)₃ has not been prepared so far.

It has been documented that some of the triorganosilanethiols (R₃SiSH) and diorganosilanedithiols (R₂Si(SH)₂) can be obtained by a variety of methods including alcoholysis of SiS₂,⁴ protonation of potassium silanethiolate,⁵ insertion of sulfur into triorganosilane,⁶ treatment of silylsulfide with hydrogen halide,⁷ and LiAlH₄ reduction of a silanepolysulfide.⁸ More recently, a transition metal-trapped silylenylthiol LSi[Ni(CO)₃]SH (L = HC[C(Me)CN-2,6-iPr₂C₆H₃]₂) has been achieved by 1,4-addition of H₂S to L′Si[Ni(CO)₃] (L′ = HC[C(Me)N-2,6-iPr₂C₆H₃][C(CH₂)N-2,6-iPr₂C₆H₃]).⁹ In comparison with the preparation of the organosilanols,³ the synthesis of the organosilanethiols appears to be more complex. The approach to prepare organosilanetrithiol by reacting the R′SiCl₃ (R′ = Me₂iPrC or Me) precursor with Li₂S¹⁰ or H₂S/NEt₃ in the presence of MeCl₂SiSiCl₂Me¹¹ was not successful, and instead the silylsulfide clusters ((Me₂iPrCSi)₄S₆ and (MeSi)₄S₅) were produced. An intermediate in these reactions might be the Si–S–M (M = Li, H) moiety, which further reacted to yield the Si–S–Si unit observed in the clusters.

RSi(OH)₃ has been prepared by using RSiCl₃ with a bulky R group which prevents the condensation shown in the previous work.^3c Following the strategy of synthesizing RSi(OH)₃ by controlled hydrolysis of RSiCl₃ (1, R = N(SiMe₃)-2,6-iPr₂C₆H₃) in the presence of an amine as the HCl-acceptor,¹² we used H₂S instead of H₂O. The experiments were carried out with different amines (NEt₃, 2,6-iPr₂C₆H₃NH₂, or pyridine) at various temperatures.¹³ However, no reaction was observed.

Subsequently, we employed a salt metathesis reaction by treating 1 with Na₂S or K₂S, however, according to NMR analysis no transformation occurred. When we used a freshly prepared Li₂S⁶ obtained from the reaction of sulfur with LiBEt₃H, the reaction proceeded and several products were formed depending on the amount of Li₂S and the reaction conditions. As illustrated in Scheme 1, the reaction of 1 with 5.3 equivalents of Li₂S in THF was carried out at 25 °C and stirred for 10 days, affording {RSi[SLi(THF)]₃}₂ (2) as colorless crystals in high yield (85%). However, upon changing the temperature, time, and ratio of the precursors, the reaction only resulted in products RSiCl₂SLi(THF)₃ (3) and RSi[SLi(THF)](μ-S)₂Si[SLi(THF)₂]R (4). Compound 3 is an incompletely sulfurized species which can be isolated as colorless crystals in 80% yield, when the reaction was conducted under stirring at 10 °C for 24 h using three equivalents of Li₂S. Increasing the reaction temperature to 35 °C and extending the reaction time to 72 h yielded a mixture of 3 and 4. The latter shows the formation of a Si(μ-S)₂Si moiety, which is generated at a little higher temperature than 25 °C and is not observed during the formation of 2. Compared with the results reported in the literature,^10,11 it is worth noting that a combination of the right RSiCl₃,¹⁴ restricted temperature, long reaction time, and excess of Li₂S is important to the successful and high yield production of 2.

Scheme 1 Reaction of 1 with freshly prepared Li₂S under different reaction conditions to produce compounds 2–4.

Compounds 2–4 are air and moisture sensitive and have been characterized by NMR spectroscopy and X-ray crystallography. Compound 2 shows symmetric patterns in ¹H, ¹³C and ²⁹Si NMR spectra in solution, indicating that all the R groups in 2 are equivalent. The CHMe₂ resonance (4.05 ppm) of R in 2 is lowfield shifted when compared with those of 3 (3.60 ppm) and 4 (3.43 ppm). The ²⁹Si NMR spectra of 2 display the respective resonances at δ 1.2 (SiMe₃) and −4.2 ppm (SiS). The X-ray single-crystal structural analysis of 2 (Fig. 1) reveals that two Si(SLi)₃ are assembled into a metal cage cluster which is comparable to those of the organosilanetriol-derived metallosiloxanes formed.^3b–d The rhomboid-based dodecahedral Si₂S₆Li₆ core of structure 2 contains six tetra-coordinate Li atoms each linked to one O atom from the THF molecule and three S atoms. Each of the six S atoms is located at the apex of a tetragonal pyramid and a SiLi₃ four-membered ring forms the base of the pyramid. Similar structures containing Si₂N₆Li₆,¹⁵ Si₂P₆Li₆,¹⁶ and Ge₂As₆Li₆¹⁷ frameworks have been reported. The congener RSi(OX)₃ (X = alkali metal) has not been prepared so far. The Si–S (2.1098(10)–2.1383(10) Å) and S–Li (2.398(5)–2.528(5) Å) bond lengths in 2 are comparable with those in 3 and 4, respectively. The X-ray structures and detailed descriptions of 3 and 4 are given in the ESI.†

Fig. 1 X-ray crystal structure of 2 with H atoms omitted for clarity.

Compound 2 is a potential precursor for preparing aminosilanetrithiol. As expected, treatment of 2 with MeCOOH easily produced RSi(SH)₃ (5) as colorless crystals in 40% yield (Scheme 2). The alternative use of CF₃COOH, C₆H₅COOH, or p-MeC₆H₄SO₃H is also possible but does not improve the yield.¹⁸ Compound 5 is air and moisture sensitive. It has a melting point of 154 °C, indicative of good thermal stability. In addition, 5 exhibits good solubility in common organic solvents such as n-hexane, toluene, diethyl ether, THF, and chlorinated hydrocarbons. The ¹H NMR spectrum of 5 clearly shows the SH proton resonance at δ 1.18 ppm, which is comparable to those found in Tbt(Mes)Si(SH)₂ and Tbt(Mes)Si(OH)SH (0.87–1.18 ppm).⁸ The SiS silicon resonance (2.7 ppm) is remarkably lowfield shifted even to a positive value in contrast to those of the SiO ones of the organosilanetriols (−40 to −90 ppm).^3d,13 A strong SH absorption band is observed at [small nu, Greek, tilde] 2543 cm⁻¹ in the IR spectrum.

Scheme 2 Synthesis and the crystal structure of 5.

The structural analysis unambiguously reveals 5 to be a monomeric molecule (Scheme 2). However, the refinements indicate a severe disorder for the N-bonded SiMe₃ and Si(SH)₃ groups which are pseudo-symmetrically arranged along the N–C_aryl axis, although the Me and SH groups show different appearances (see Fig. S4 in the ESI†). Treatment with a splitting mode results in a setting of the related bond lengths (Si–S, 2.10–2.25 ± 0.01; Si–C, 1.86 ± 0.01; Si_SiMe3–N, 1.76 ± 0.01; Si_SiS–N, 1.72 ± 0.01; S–H, 1.35 ± 0.03 Å) on the basis of the reported coordinate data. The final refinement gives convergence with bond parameters (Si–S, 2.101(8)–2.110(7) Å; S–H, 1.31–1.34(3) Å; ∠Si–S–H, 94(4)–109(2)°) comparable to those found in Tbt(Mes)Si(SH)₂, Tbt(Mes)Si(OH)SH,⁸ and LSi[Ni(CO)₃]SH⁹ as well as in the predicted [Si(SH)₃]⁺.^2a

To further understand the bonding matrix of compound 5, we carried out quantum chemical calculations on the basis of density functional theory (DFT). By optimizing the geometry of the molecule 5 at the M06-2X/def2-TZVPP level,¹⁹ the general structure (Fig. S6, ESI†) calculated is basically in agreement with the results obtained by the X-ray structural analysis (Fig. S4, ESI†). The Si_Si(SH)3–N distance (1.712 Å) is computed to be ca. 0.020 Å longer than the experimental value while the Si_SiMe3–N one (1.774 Å) is 0.014 Å shorter. Both the experimental and theoretical values are significantly shorter than the Si–N standard value (1.87 Å).²⁰ The experimental Si–S bond lengths range from 2.101(8) to 2.110(7) Å, which are a little shorter than the calculated distances (2.139–2.147 Å).

We also performed the calculation on the ¹H and ²⁹Si NMR spectroscopic data by means of DFT-GIAO calculations.²¹ Since the NMR measurements were carried out in solution (CDCl₃), we optimized the geometry of 5 using the PCM (Polarized Continuum Model)²² method in which the effect by solvent was considered. Therefore, the geometry of 5 optimized at the PCM(CHCl₃)-M06-2X/def2-TZVPP level²² is only slightly different from the one at the M06-2X/def2-TZVPP level. It is computed that the chemical shifts for the Si(SH)₃ group are at δ 2.02 ppm for the proton and 5.97 ppm for the silicon atom. The data for the hydrogen and silicon in the SiMe₃ moiety are at δ 0.35 and 11.25 ppm, respectively. These data generally fit to the experimental^5,9 and the other related computational data.²³ Furthermore, the computed IR spectrum exhibits a band for the SH groups at [small nu, Greek, tilde] 2753 cm⁻¹, which is a little higher than the experimental value.

A natural bond orbital (NBO) analysis²⁴ was accomplished to investigate the bonding situation of 5. The NBO results identify two N–Si bonds which are strongly polarized toward nitrogen (84% for the N atom and 16% for the Si atoms). The partial charges calculated suggest that the nitrogen atom carries a negative charge (−1.38 e) in a large extent while the silicon atoms have strong positive charges, namely, +1.91 e (for SiMe₃) and +1.46 e (for Si(SH)₃). The Wiberg bond order values P for Si_SiS–N and Si_SiMe3–N bonds have been calculated at different levels of theory (Table S5, ESI†). The data point out that the former (P = 0.61–0.63) has a higher bond order than the latter (P = 0.56–0.58). The former one can be explained by hyperconjugation of the lone pair orbital of the nitrogen, which is calculated by using a second-order perturbation theory included in the NBO method. The LP(N) → σ*(Si–S) donation amounts to 17.8 kcal mol⁻¹ while the LP(N) → σ*(Si–C) hyperconjugation is only 11.2 kcal mol⁻¹ (Table S3, ESI†). Furthermore, application of the second-order perturbation theory revealed delocalization by three stabilizing two-electrons from the sulfur (LP(S)) to the anti-bonding orbital σ*(Si–S), giving hints of a high conjugation of the Si(SH)₃ fragment. The computed associated energies (ΔE(2)) are 13.54, 12.81 and 11.67 kcal mol⁻¹ for each of the LP(S) → σ*(Si–S) interactions.

In summary, we have successfully synthesized the aminosilanetrithiol RSi(SH)₃ (5) through protonation of its precursor {RSi[SLi(THF)]₃}₂ (2) by MeCOOH. Precise control of the reaction conditions is crucial for the synthesis of 2, while varying the reaction conditions led to the products 3 and 4. DFT calculations of 5 confirm the experimental data. The corresponding NBO analysis shows that the LP(N) → σ*(Si–S) and LP(S) → σ*(Si–S) donations remarkably contribute to the stabilization of the Si(SH)₃ fragment. Compound 5 shows a structure containing three SH groups attached to one element. The reactivity studies of 5 are now in progress.²⁵

This work was supported by the 973 Program (2012CB821704), the National Nature Science Foundation of China (91027014), and the Innovative Research Team Program (IRT1036 and 20923004). Support of the Deutsche Forschungsgemeinschaft (DFG) is highly acknowledged.

Notes and references

1. (a) S. Böhm, I. Senf, H.-D. Schädler and J. Kuthan, THEOCHEM, 1992, 253, 73–82; (b) S. Böhm, D. Antipova and J. Kuthan, Int. J. Quantum Chem., 1996, 60, 649–655; (c) R. Glaser, G. S.-C. Choy, G. S. Chen and H. Grützmacher, J. Am. Chem. Soc., 1996, 118, 11617–11628.
2. (a) C. M. Marchand, U. Pidun, G. Frenking and H. Grützmacher, J. Am. Chem. Soc., 1997, 119, 11078–11085; (b) A. Herman and W. Wojnowski, Struct. Chem., 1992, 3, 239–244.
3. See selected examples: (a) L. J. Tyler, J. Am. Chem. Soc., 1955, 77, 770–771; (b) R. Murugavel, V. Chandrasekhar and H. W. Roesky, Acc. Chem. Res., 1996, 29, 183–189 and references therein; (c) R. Murugavel, A. Voigt, M. G. Walawalkar and H. W. Roesky, Chem. Rev., 1996, 96, 2205–2236; (d) V. Chandrasekhar, R. Boomishankar and S. Nagendran, Chem. Rev., 2004, 104, 5847–5910.
4. R. Piekos and W. Wojnowski, Z. Anorg. Allg. Chem., 1962, 318, 212–216.
5. (a) J. E. Drake, B. M. Glavinčevski and R. T. Hemmings, Can. J. Chem., 1980, 58, 2161–2166; (b) J. Pikies and W. Wojnowski, J. Organomet. Chem., 1989, 373, 317–326.
6. H. Lange and U. Herzog, J. Organomet. Chem., 2002, 660, 36–42.
7. H. Beckers and H. Bürger, Z. Anorg. Allg. Chem., 2001, 627, 1217–1224.
8. (a) T. Tanabe, N. Takeda and N. Tokitoh, Eur. J. Inorg. Chem., 2007, 1225–1228; (b) T. Tanabe, Y. Mizuhata, N. Takeda and N. Tokitoh, J. Organomet. Chem., 2009, 694, 353–365.
9. A. Meltzer, S. Inoue, C. Präsang and M. Driess, J. Am. Chem. Soc., 2010, 132, 3038–3046.
10. M. Unno, Y. Kawai, H. Shioyama and H. Matsumoto, Organometallics, 1997, 16, 4428–4434.
11. U. Herzog and G. Rheinwald, J. Organomet. Chem., 2001, 628, 133–143.
12. R. Murugavel, V. Chandrasekhar, A. Voigt, H. W. Roesky, H.-G. Schmidt and M. Noltemeyer, Organometallics, 1996, 14, 5298–5301.
13. We have tried H₂S in the liquid form at ca. −100 °C and allowed it to slowly warm to room temperature in the course of the reaction.
14. We have also tested the group tolerance in the syntheses of compounds 2 and 5. Related results are shown in our recent patent which is still under authorization (CN10222598; Patent Application No. CN20111115911 20110505). However, compound 5 is selected as a typical example in this manuscript for profound experimental and theoretical studies.
15. D. J. Brauer, H. Bürger, G. R. Liewald and J. Wilke, J. Organomet. Chem., 1985, 287, 305–320.
16. (a) M. Driess, G. Huttner, N. Knopf, H. Pritzkow and L. Zsolnai, Angew. Chem., Int. Ed. Engl., 1995, 34, 316–318 ( Angew. Chem. , 1995 , 107 , 354–356 ); (b) M. Driess, C. Monsé and K. Merz, Z. Anorg. Allg. Chem., 2001, 627, 1225–1230; (c) M. Driess, Adv. Inorg. Chem., 2000, 50, 235–284.
17. L. Zsolnai, G. Huttner and M. Driess, Angew. Chem., Int. Ed. Engl., 1993, 32, 1439–1440 ( Angew. Chem. , 1993 , 105 , 1549–1551 ).
18. We tried HCl or HCl·NEt₃ as the proton source. As a result, the chlorination happened and ionic (RCl₂SiS)⁻(HNEt₃)⁺ was isolated.
19. (a) Y. Zhao and D. G. Truhlar, Theor. Chem. Acc., 2008, 120, 215–241; (b) F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005, 7, 3297–3305.
20. P. Pyykkö and M. Atsumi, Chem. – Eur. J., 2009, 15, 12770–12779.
21. (a) K. Wolinski, J. F. Hinton and P. Pulay, J. Am. Chem. Soc., 1990, 112, 8251–8260; (b) M. W. Lodewyk, M. R. Siebert and D. J. Tantillo, Chem. Rev., 2012, 112, 1839–1862.
22. (a) S. Miertuš, E. Scrocco and J. Tomasi, Chem. Phys., 1981, 55, 117–129; (b) V. Barone, M. Cossi and J. Tomasi, J. Comput. Chem., 1998, 19, 404–417; (c) F. Jensen, J. Chem. Theory Comput., 2008, 4, 719–727; (d) N. C. Handy and A. J. Cohen, Mol. Phys., 2001, 99, 403–412; (e) J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
23. (a) D. Cremer, L. Olsson and H. Ottosson, THEOCHEM, 1994, 313, 91–109; (b) L. Olsson, C.-H. Ottosson and D. Cremer, J. Am. Chem. Soc., 1995, 117, 7460–7479.
24. (a) A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev., 1988, 88, 899–926; (b) A. E. Reed and F. Weinhold, J. Chem. Phys., 1985, 83, 1736–1740; (c) A. E. Reed, R. B. Weinstock and F. Weinhold, J. Chem. Phys., 1985, 83, 735–746.
25. Currently, we have obtained some preliminary results by isolating Ag₄ clusters of composition [R(HS)₂SiSAg]₄ (6) and [R(HS)₂SiSAg]₂(AgMes)₂ (7) from the reaction of 5 with (AgMes)₄, which indicate a monofunctional reaction of the trithiol (5). Due to poor quality of the crystals of 7, a preliminary structure of this compound was determined. The synthesis of these two compounds and crystallographic data of 6 are included in ESI†.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental procedures and X-ray crystallographic and computational information. CCDC 973535 (2), 973536 (3), 973538 (4), 979368 (5), and 987309 (6). For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c4cc00912f

---