Reactivity of hypervalent silicon and tin compounds toward electrophilic sulfur

Article abstract of DOI:10.1016/S0040-4039(00)00890-X

Sulfuration of hypervalent complexes of silicon and tin was attempted with several electrophilic sulfur reagents for making carbon - sulfur bonds under almost neutral conditions. We provided a qualitative scale of reactivity between various hypervalent salts. Simple and practical methodologies involving non-odorous elemental sulfur were developed. It avoided the use of H₂S or standardized organolithium and Grignard solutions with sulfur, under highly basic conditions. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00890-X

Tetrahedron Letters 41 (2000) 5841±5845
Reactivity of hypervalent silicon and tin compounds toward
electrophilic sulfur
Sebastien Kerverdo, Xavier Fernandez, Sophie Poulain and Marc Gingras*
Â
Department of Chemistry, Faculty of Sciences, University of Nice-Sophia Antipolis, 28 Avenue Parc Valrose,
6108 Nice Cedex 2, France
0
Received 17 March 2000; accepted 1 June 2000
Abstract
Sulfuration of hypervalent complexes of silicon and tin was attempted with several electrophilic sulfur
reagents for making carbon� sulfur bonds under almost neutral conditions. We provided a qualitative scale
of reactivity between various hypervalent salts. Simple and practical methodologies involving non-odorous
elemental sulfur were developed. It avoided the use of H S or standardized organolithium and Grignard
2
solutions with sulfur, under highly basic conditions. # 2000 Elsevier Science Ltd. All rights reserved.
Keywords: sulfuration; hypervalent elements; tin and compounds; silicon and compounds; disul®des.
Although several methods have been created for introducing sulfur into organic molecules,
1
there is a general lack of studies involving hypervalent complexes. Sulfuration of these species is
the purpose of this study. Hypervalent compounds have lead to a great interest in recent organic
methodologies, due to a larger electronic density and particularly, to the weakening of some
2
3
carbon� heteroatom bonds; it was exploited in Pd-catalyzed procedures involving boron, silicon
4
and tin compounds. Furthermore, hypervalent species are thought to be the reactive inter-
mediates in the sequence hydroboration±oxidation, or the Tamao's oxidation with organosilicons.
Because commercial hypervalent ¯uorinating agent 1 was originally produced by one of us, as an
anhydrous synthetic equivalent to TBAF, it served as a starting point in this study.5
,6
The question raised here concerns electron-rich hypervalent species and a scale of reactivity
towards some electrophilic sulfur reagents. Another curiosity is the relative reactivity of hyper-
valent silicon and tin compounds. A goal pursued is the creation of simple sulfuration procedures.
The abundance of sulfur, its low cost and its almost non-odorous nature, is highly desirable in
industrial or academic processes. It sharply contrasts to the regulated use of poisonous, smelly
and gaseous H S.
2
*
Corresponding author. Fax: 33 (0)4 93 44 04 25; e-mail: marc.gingras@wanadoo.fr
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00890-X

5
842
Equation 1 (in Scheme 1) shows classic methods for making Ar� S bonds, when using strong
bases. Our procedures avoid the use of pyrophoric materials, highly basic organolithiums, or
Grignard reagents.7
Scheme 1. Reaction of nucleophiles with elemental sulfur
Manipulation of a large amount of highly reactive, air- and moisture-sensitive carbanions is
undesirable. Some synthetic equivalents to these carbanions are commercial, crystalline and non-
hygroscopic salts, which are operating under almost neutral conditions. Importantly, they can be
weighed in open air and do not need a standardization with some anhydrous and cumbersome
titrations. Alternatively, they can be prepared in situ (Fig. 1). The most stable structures proposed
for 3 and 4 agree with the `electronegativity rules' for the stereochemistry of TBP complexes (the
most electronegative groups are in the apical positions).
Figure 1. List of hypervalent complexes used or generated
We then investigated their reactivities under various experimental conditions: sulfurating
agents, variation of the elements (Si, Sn), solvent, temperatures, time, molar ratios and ligand
transferred (Table 1). A constant observation is the delivery of only one ligand and the highly
selective formation of alkyl or aryl disul®des.
The electrophilic sulfur species utilized are presented in Fig. 2. From this list, only sulfur
dichloride (SCl ) is less convenient, because of its easy decomposition to sulfur monochloride
2
(S Cl ). Powdered sulfur and 8 are convenient stable solids, and S Cl is cheap and produced in
bulk amount.
2 2 2 2

5
843
Table 1
Sulfuration of hypervalent silicon and tin compoundsa

5
844
Figure 2. List of electrophilic sulfur reagents used
As shown in Table 1, all of the sulfur reagents react with hypervalent organotins. In contrast,
8
organosilicon complexes seem to decompose into triphenylsilyl ¯uoride. On no occasion, have
they led to a signi®cant amount of sulfurated organics. The optimized conditions with 1 could not
be reproduced with organosilicon 2 and sulfur. Sulfur dichloride 5 also gave negative results.
Apparently, the reactivity of the C� Si bond does not surpass its decomposition. These ®ndings
represent a case where the chemical properties between 1 and 2 are very di€erent, when thay are
compared as nucleophilic ¯uorinating agents. The C� Sn bond is longer, more polarizable and
has a richer electronic density than the C� Si bond. As a consequence, organotin complexes
behave as better nucleophiles towards electrophilic sulfur species.
Several experimental parameters were optimized with organotins and elemental sulfur (Table 1).
ꢀ
ꢀ
We found slow kinetics at 50 C with 1, but the reaction proceeded well at 120±130 C. Completion
ꢀ
of the reaction was estimated between at 2±5 h at 130 C in DMF (entries 2, 3). A variation in the
sulfur/organotin molar ratio was veri®ed and a 3:1 ratio gave the best results. A lower ratio has a
negative e€ect on the yields. These results are highly dependent on temperature and solvent:
DMSO and DMF provide the best results, but THF was inadequate. Interestingly, the cyclohexyl
group can be transferred from tin, if the hypervalent complex is generated in situ with CsF (1.25
ꢀ
equiv.) in DMF at 130 C for 48 h (sulfur/organotin: 3:1). However, reagent 8 did not lead to a
signi®cant amount of disul®de. Starting with nBu F (1.0 equiv.), some reactions were attempted
3
with CsF (3.0 equiv.) and sulfur (1.0 equiv.). Complex 4 and its isomer are thought to be gener-
ated in situ, because of the GC/MS detection of some n-butyl disul®de. However, due to its
volatility, the isolation and quanti®cation were dicult.
Freshly distilled sulfur dichloride gave poor yields of disul®de, in spite of many assays at various
temperatures and molar ratios. Sulfur monochloride 6 could be useful as an alternative reagent to
sulfur powder (entry 13, 80% yield). Even if reagent 8 usually supplies monosul®de in a reaction
with Grignard reagents, it gave us unexpected disul®de formation. The best conditions used a 8/
ꢀ
organotin molar ratio of 2:1, at 130 C, in less than 16 h (80% yield). Surprisingly, DMSO cannot
be utilized with 8 (entry 15, 0% yield).
Overall, the reactions of hypervalent compounds produced organic disul®des in a selective
manner. Many precautions were taken for detecting organic thiols by GC/MS or TLC but it has
always failed, even when using inert gas techniques. Entry 11 shows that forcing conditions, by
the addition of CsF or an excess of sulfur powder under prolonged heating, do not lead to more
than one ligand transfer. This fact is distinctive of hypervalent compounds, where one apical
group is the most reactive in trigonal bipyramidal (TBP) complexes.
At this stage, we cannot delineate an exact mechanism for those sulfurations. However,
because organotin compounds can accommodate some C� Sn bond insertion reactions (for
instance with SO ), one could hypothesize that sulfur could be introduced in this way, rather than
2
going through an ionic mechanism. If it were so, one would expect the detection of thiols, from a
release of carbanions, as in Scheme 1. Moreover, organomagnesiums reacting with 8 usually
produce organic monosul®des, in contrast to our results. Because most sulfur reagents lead to a

5
845
selective production of disul®des, we are questioning the possibility of generating a common
radical intermediate, leading to a dimerisation of RS.
Through a Berry pseudorotation (Scheme 2), the dynamic stereochemistry of the hypervalent
species might exist as an equilibrium between cis/trans TBP complexes; the cis complex being the
most reactive one, because of apical alkyl or aryl groups.
Scheme 2.
Acknowledgements
We thank UNSA for start-up funds. S.K. is grateful to the government of France (MENRT),
for a doctoral scholarship. Professor Roland FELLOUS is acknowledged for allowing us to share
the laboratory and his deep knowledge in chemistry.
References
1
. (a) Cremlyn, R. J. An Introduction to Organosulfur Chemistry; John Wiley & Sons: New York, 1996. (b) Metzner,
P.; Thuillier, A. Sulfur Reagents in Organic Synthesis; Academic Press: London, 1994. (c) Organosulfur Chemistry;
Page, P., Ed.; Academic Press: London, 1995.
2
3
. Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457±2483.
. (a) Mowery, M. E.; DeShong, P. Org. Lett. 1999, 1, 2137±2140 and references therein. (b) Hiyama, T. In Metal-
Catalysed Cross-Coupling Reactions; Diederich, F.; Stang, P. J., Eds. Organosilicon Compounds in Cross-
Coupling Reactions; Wiley-VCH: New York, 1998; pp. 421±453.
4. Vedejs, E.; Haight, A. R.; Moss, W. O. J. Am. Chem. Soc. 1992, 114, 6556.
5. Gingras, M. Tetrahedron Lett. 1991, 32, 7381±7384.
6. Uses of 1: (a) Brigaud, T.; Lefebvre, O.; Plantier-Royon, R.; Portella, C. Tetrahedron Lett. 1996, 37, 6115±6116.
(
b) Bu€noir, S.; Mestdagh, H.; Rolando, C. Electron. Conf. Trends Org. Chem. 1995, Royal Society of Chemistry:
Cambridge, UK. (c) Brigaud, T.; Doussot, P.; Portella, C. J. Chem. Soc., Chem. Comm. 1994, 18, 2117±2118. (d)
Honda, H.; Miwako, M. Chem. Lett. 1994, 6, 1013±1016. (e) Jang, M.; Janzen, A. F. J. Fluorine Chem. 1994, 66,
1
29±135. (f) Garcõ
Martõnez, A.; Osio Barcina, J.; Rys, A. Z.; Subramanian, L. R. Synlett 1993, 8, 587±588. (h) Wang, C.; Mo, Y.;
Jang, M.; Janzen, A. F. Can. J. Chem. 1993, 71, 525±528. (i) Garcõa Martõnez, A.; Osio Barcina, J.; Rys, A. Z.;
Subramanian, L. R. Tetrahedron Lett. 1992, 33, 7787±7788. (j) Daktemieks, D.; Zhu, H. Inorg. Chim. Acta 1992,
96, 19±25.
a Martõnez, A.; Osio Barcina, J.; De Frensno Cerezo, A. Synlett 1994, 12, 1047±1048. (g) Garcõa
  Â
Â
Â
Â
1
7
8
. March, J. Advanced Organic Chemistry, 4th Ed.; John Wiley & Sons: New York, 1992; pp. 613±614.
. Analyzed by GC/MS.