A catalytic synthesis of thiosilanes and silthianes: Palladium nanoparticle-mediated cross-coupling of silanes with thio phenyl and thio vinyl ethers through selective carbon-sulfur bond activation

Article abstract of DOI:10.1021/ja049386u

Palladium nanoparticles generated in situ from N, N-dimethyl-acetamide (DMA) solutions of PdX₂ (X = Cl^-, OAc^-, OCOCF₃^-) or Pd₂(dba)₃ by reduction with alkyl silanes R₃SiH (R = Me, Et, i-Pr, t-Bu) are selective catalysts for the cross-coupling of the silanes R₃SiH with phenyl and vinyl thioethers forming the corresponding thiosilanes and silthianes in high yields and under mild conditions. The method is applicable to phenyl thioglycosides, giving access to thiosilyl glycosides a new class of sugar derivatives.

Full text of DOI:10.1021/ja049386u

Published on Web 05/20/2004
A Catalytic Synthesis of Thiosilanes and Silthianes:
Palladium Nanoparticle-Mediated Cross-Coupling of Silanes
with Thio Phenyl and Thio Vinyl Ethers through Selective
Carbon-Sulfur Bond Activation
Mee-Kyung Chung and Marcel Schlaf*
Contribution from the Guelph-Waterloo Centre for Graduate Work in Chemistry (GWC)2,
Department of Chemistry and Biochemistry, UniVersity of Guelph, Guelph,
Ontario, Canada, N1G 2W1
Received February 3, 2004; E-mail: mschlaf@uoguelph.ca
Abstract: Palladium nanoparticles generated in situ from N,N-dimethyl-acetamide (DMA) solutions of PdX2
(X ) Cl-, OAc-, OCOCF3-) or Pd2(dba)3 by reduction with alkyl silanes R3SiH (R ) Me, Et, i-Pr, t-Bu) are
selective catalysts for the cross-coupling of the silanes R3SiH with phenyl and vinyl thioethers forming the
corresponding thiosilanes and silthianes in high yields and under mild conditions. The method is applicable
to phenyl thioglycosides, giving access to thiosilyl glycosides a new class of sugar derivatives.
Introduction
chlorosilane generates the corresponding alkylthio trimethylsi-
lane. With dimethyl disulfides and dibenzyl disulfides, hexa-
Thiosilanes are a distinct class of organosulfur compounds
characterized by the high reactivity of the relatively weak sili-
con-sulfur bond (estimated at ∼300 kJ/mol)1 and the oxophi-
licity of silicon, which result in interesting applications in
organic synthesis. For example, they have been used as selective
masking reagents of carbonyl groups to generate O-silylhemith-
ioacetals and ketals, which are precursors for the preparation
of unsymmetrical sulfides.2,3 They also react with acid halides
to produce thio carboxylic esters (thiol esters),4 which upon
sulfur activation act as synthons for carboxyl in esterification
reactions.5 In addition, thiosilanes have been employed for
oxirane ring-opening through C-O bond cleavage, which is a
useful starting point for the synthesis of multifunctionalized
organic compounds6 and in the synthesis of vinyl sulfides.7
methyldisilthiane is produced.10 Hexamethyldisilthiane, widely
used as a sulfur transfer agent, can also be prepared by a reaction
of chlorosilanes and Na/S/naphthalene or Li2S in THF11,12 and
by the silylation of hydrogen sulfide with 1-(trimethylsilyl)-
imidazole.13 An interesting class of cyclic silthianes is formed
by reaction of organochlorosilanes with hydrogen sulfide and
base.14,15 Here we wish to report on the serendipitous discovery
of a new chemoselective palladium nanoparticle-catalyzed route
to thio silanes and silthianes starting from phenyl or vinyl
sulfides.
Discovery and Context. Recently, we reported a new method
for the regioselective silylation of simple alkyl and phenyl
glycosides using a palladium nanoparticle colloid-catalyzed
silane alcoholysis reaction in which the catalyst is formed in
situ by reduction of a Pd2+ precursor.16 Palladium nanoparticles
formed in situ from Pd(OAc)2 have also been shown to be the
active catalyst in the silaesterification reaction of polyhydrosi-
loxanes with carboxylic acids.17 Analogous to the role of the
silane in the silane alcoholysis reaction described by us, in this
reaction the polyhydrosiloxanes also act as the reductant,
reactant, and nanocolloid stabilizer at the same time.
Thiosilanes are normally prepared from the chlorosilane and
metal thiolate.1 The silylation of thiols with trialkyl/aryl silanes
or alkyl/aryl disilane catalyzed by transition-metal complexes
such as Wilkinson’s catalyst also provides a convenient route
to thiosilanes.8 At elevated temperatures, the reaction can also
be carried out without catalyst.9 The reductive cleavage of
disulfides with metallic sodium in the presence of trimethyl-
To further test the scope and limitations of the silane alco-
holysis method for the synthesis of sugar sily ethers, we sub-
(1) Armitage, D. A. F. Chemistry of Compounds with Silicon-Sulphur, Silicon-
Selenium and Silicon-Tellurium Bonds; Patai, S., Rappoport, Z., Eds.; John
Wiley & Sons: Toronto, 1991; pp 213-243 and references therein.
(2) Glass, R. S. Synth. Commun. 1976, 6, 47-51.
(3) Evans, D. A.; Truesdale, L. K.; Grimm, K. G.; Nesbitt, S. L. J. Am. Chem.
Soc. 1977, 99, 5009-5017.
(10) Kuwajima, I.; Abe, T. Bull. Chem. Soc. Jpn. 1978, 51, 2183-2184.
(11) So, J.-H.; Boudjouk, P. Synthesis 1989, 306-307.
(12) Curphey, T. J. Phosphorus, Sulfur Silicon Relat. Elem. 2001, 173, 123-
142.
(4) Talley, J. J. Synthesis 1981, 549.
(5) Masamune, S.; Yamamoto, H.; Kamata, S.; Fukuzawa, A. J. Am. Chem.
Soc. 1975, 97, 3513-3515.
(13) Louis, E.; Urry, G. Inorg. Chem. 1968, 7, 1253-1254.
(14) Reikhsfel’d, V. O.; Lebedev, E. P. Zh. Obshch. Khim. (Engl. Transl.) 1970,
40, 586-590.
(6) Tanabe, Y.; Mori, K.; Yoshida, Y. J. Chem. Soc., Perkin Trans. 1 1997,
671-675.
(15) Herzog, U.; Bo¨hme, U.; Roewer, G.; Rheinwald, G.; Lang, H. J. Organomet.
Chem. 2000, 602, 193-207.
(7) Degl’Innocenti, A.; Ulivi, P.; Capperucci, A.; Mordini, A.; Reginato, G.;
Ricci, A. Synlett 1992, 499-500.
(8) Ojima, I.; Nihonyanagi, M. J. Organomet. Chem. 1973, 50, C26-C28.
(9) Becker, B.; Wojnoski, W. Synth. React. Inorg. Met.-Org. Chem. 1984, 14,
537-556.
(16) Chung, M.-K.; Orlova, G.; Goddard, J. D.; Schlaf, M.; Harris, R.; Beveridge,
T. J.; White, G.; Hallett, F. R. J. Am. Chem. Soc. 2002, 124, 10508-10518.
(17) Chauhan, B. P. S.; Rathore, J. S.; Chaihan, M.; Krawicz, A. J. Am. Chem.
Soc. 2003, 125, 2876-2877.
9
7386
J. AM. CHEM. SOC. 2004, 126, 7386-7392
10.1021/ja049386u CCC: $27.50 © 2004 American Chemical Society

Pd-Catalyzed Silane Thio Ether Cross-Coupling
A R T I C L E S
Table 1. Survey of the Cross-Coupling Reaction of Various Alkyl and Aryl Sulfides with Excess Silane t-BuMe2SiH (TBDMSiH) and Various
Palladium Catalyst Precursors at 298 K in DMA
yields (%) of products with various catalyst precursors by quantitative
GC (reaction time in hours)a
entry
sulfide
products
PdCl2b
quant.
(3)
quant.
(4)
Pd(OAc)2b
Pd(OCOCF ) b
Pd2(dba)3b
Pd0
Pd(PPh3)4
3 2
1
HOCH2CH2SPhc
TBDMSi-OCH2CH2S-SiTBDM
quant.
(2)
quant.
(5)
quant.
(2)
quant.
(3)
quant.
(2)
quant.
(4)
14
(19)
0
0
(2)
0
(1)
0/ 0
(24)
0/ 0
(12)
0/ 0
(24)
0/0
2
3
4
5
6
EtSPhd
EtS-SiTBDM
(5)
EtSEte
EtS-SiTBDM/(TBDMSi)2S
PhS-SiTBDM/(TBDMSi)2S
PhS-SiTBDM/(TBDMSi)2S2
0/0
0/0
0/0
0/0
0/ 0
(24)
19/ 2
(9)
0/ 0
(24)
0/0
(24)
68/32
(4)
0/0
(24)
0/quant.
(24)
71/29
(4)
1/0
(24)
0/quant.
(24)
73/27
(4)
0/0
(24)
0/quant.
(24)
57/43
(7)
3/0
(24)
0/quant.
PhSPhe
PhSSPhe
PhSCH2SPhe
PhSCH2S-SiTBDM/
TBDMSi-SCH2S-SiTBDM
(11)
(11)
(11)
(11)
(11)
(11)
d
7
EtSCHdCH2
EtS-SiTBDM/EtSEt
22/71
(24)
37/6/30/7
45/54
(24)
48/21/6/21
38/62
(24)
47/12/18/8
35/65
(24)
46/23/6/10
0/0
(24)
0/0/2/0
0/0
(24)
0/0/0/0
e
8
PhSCHdCH2
PhS-SiTBDM/
TBDMSi-SEt/PhSEt/
(TBDMSi)2S
(12)
quant.
(12)
quant.
(12)
(12)
quant.
(12)
quant.
(12)
(12)
quant.
(12)
quant.
(12)
(12)
quant.
(12)
quant.
(12)
(12)
0
(12)
0
(12)
0
(12)
0
9
MeSPh-p-OMed
MeSPh-p-Cld,f
MeS-SiTBDM
MeS-SiTBDM
10
(12)
(12)
a
b
c
Catalyts load 3 mol % with respect to thioether in all reactions. TEM analysis shows that Pd0 nanoparticles are generated in situ. Amount of
d
e
f
TBDMSiH ) 3.5 equiv. Amount of TBDMSiH ) 1.3 equiv. Amount of TBDMSiH ) 2.3 equiv. Byproducts are chlorobenzene and benzene
(∼21-30%).
sequently extended our study to sugars containing other func-
tional groups. With phenyl 1-thio-â-D-glucopyranoside as the
substrate, we observed the following surprising reaction (eq 1)
confirmed by elemental analysis of the product for sulfur and
to date a silicon-sulfur bond formation by the corresponding
reaction via C(sp2)-S bond cleavage is, to our knowledge,
unknown.
Results
1
loss of the signals for phenyl in the H and 13C NMR spectra.
To establish the scope and limitation of the generalized
reaction (eq 2), we investigated a series of simple sulfides with
t-BuMe2SiH (TBDMSiH) as the silane and various palladium
catalyst precursors in N,N-dimethyl-acetamide (DMA). The
results of this survey with individual yields as determined by
quantitative GC analysis calibrated against authentic samples
are summarized in Table 1.
In addition to the anticipated normal silane alcoholysis of
the hydroxyl functions at C-2 and C-6,16 the reaction produces
the sugar thiosilane, t-BuMe2Si-1-thio-2,6-O-bis(t-BuMe2Si)-
â-D-glucopyranoside, as the major product through phenyl
C(sp2)-S bond cleavage and subsequent Si-S bond formation
at the anomeric position (C-1) with the concomitant formation
of benzene, whose presence was confirmed by GC-MS. This
result is unique both from the point of view of C(sp2)-S bond
activation as well as S-Si bond formation to produce a sugar
thiosilane, an entirely new class of anomerically functionalized
sugars.
The transition-metal-mediated activation of C(sp3)-S and
C(sp2)-S bonds for both synthetic purposes18-29 and petro-
chemical hydrodesulfurization30-34 as well as in bioinorganic
related systems35 has been extensively studied. In many cases,
transition-metal-catalyzed C-S bond scissions are followed by
carbon-carbon bond formation through a cross-coupling reac-
tion with Grignard or other organometallic reagents. However,
Pd nano colloid
R-S-R′ + H-Sit-BuMe2
R and R′ ) alkyl, aryl
8
DMA, 298 K
R-S-Sit-BuMe2 + R′-H (2)
Comparison of the reactions of entries 2-4 and 6 shows that
C(sp2)-S bonds are cleaved selectively by the palladium
(23) Wong, K.-T.; Luh, T.-Y. J. Am. Chem. Soc. 1992, 114, 7308-7310.
(24) Wong, K.-T.; Yuan, T.-M.; Wang, M. C.; Tung, H.-H.; Luh, T.-Y. J. Am.
Chem. Soc. 1994, 116, 8920-8929.
(25) Luh, T.-L.; Leung, M.-K.; Wong, K.-T. Chem. ReV. 2000, 100, 3187-
3204.
(26) Hua, R.; Takeda, H.; Onozawa, S.-Y.; Abe, Y.; Tanaka, M. J. Am. Chem.
Soc. 2001, 123, 2899-2900.
(27) Planas, J. G.; Marumo, T.; Ichikawa, Y.; Hirano, M.; Komiya, S. J. Chem.
Soc., Dalton Trans. 2000, 2613-2625.
(28) Srogl, J.; Allred, G. D.; Liebeskind, L. S. J. Am. Chem. Soc. 1997, 119, 9,
12376-12377.
(29) Liebeskind, L. S.; Srogl, J. J. Am. Chem. Soc. 2000, 122, 11260-11261.
(30) Angelici, R. J. Acc. Chem. Res. 1988, 21, 387-394.
(31) Dong, L.; Duckett, S. B.; Ohman, K. F.; Jones, W. D. J. Am. Chem. Soc.
1992, 114, 151-160.
(18) Tiecco, M.; Testaferri, L.; Tingoli, M.; Chianelli, D.; Wenkert, E.
Tetrahedron Lett. 1982, 23, 4629-4632.
(32) Luo, S.; Rauchfuss, T. B.; Gan, Z. J. Am. Chem. Soc. 1993, 115, 4343-
4944.
(19) Wenkert, E.; Fernandes, J. B.; Michelotti, E. L.; Swindell, C. S. Synth.
Commun. 1983, 701-703.
(33) Janak, K. E.; Tanski, J. M.; Churchill, D. G.; Parkin, G. J. Am. Chem. Soc.
2002, 124, 4182-4183.
(20) Wenkert, E.; Shepard, M.; McPhail, A. T. J. Chem. Soc., Chem. Commun.
1986, 1390-1391.
(34) Planas, J. G.; Prim, D.; Rose-Munch, F.; Rose, E. Organometallics 2002,
21, 4385-4389.
(21) Wenkert, E.; Chianelli, D. J. Chem. Soc., Chem. Commun. 1991, 627-628.
(22) Luh, T.-Y. Acc. Chem. Res. 1991, 24, 257-263.
(35) Srogl, J.; Liu, W.; Marshall, D.; Liebeskind, L. S. J. Am. Chem. Soc. 1999,
121, 9449-9450.
9
J. AM. CHEM. SOC. VOL. 126, NO. 23, 2004 7387

A R T I C L E S
Chung and Schlaf
Table 2. Isolated Yields of Silylated Products Obtained from Representative Sulfides with PdCl2 and Excess Silane TBDMSiH (2.3 equiv) at
298 K
isolated yields (%)
entry
sulfides
silylated products
vacuum distillation
column chromatographya
1
2
3
4
5
HOCH2CH2SPh
AcOCH2CH2SPh
TBDMSi-OCH2CH2SPh
PhSCH2Ph
TBDMSi-OCH2CH2S-SiTBDM
AcOCH2CH2S-SiTBDM
95
60
89
56
85
42
45
33
57
54
TBDMSi-OCH2CH2S-SiTBDM
TBDMSi-SCH2Ph
PhSCH2SPh
TBDMSi-SCH2S-SiTBDM
a
Balance of products decomposed during column chromatography.
nanoparticles generated from a variety of PdX2 salts or
Pd2(dba)3, while C(sp3)-S bonds do not react under the reaction
conditions. In all reactions, the nano structure of the catalyst
obtained by reaction of the precursor with the silane before
addition of the thioether is a necessary prerequisite for catalytic
activity (vide infra). Bulk Pd0 (palladium black) and the
homogeneous molecularly dispersed Pd0 precursor Pd(PPh3)4,
which does not result in nanoparticle formation, give only
marginal or no catalytic activity at all. The reaction of entry 4
also shows that even with the sterically demanding silane
TBDMSiH repeated activation of C(sp2)-S bonds on the same
sulfur atom are possible, leading to a mixture of the thiosilane
and silthiane.
of silane present), and both C(sp2)-S bonds present are activated
quantitatively, leading to TBDMSi-SCH2S-SiSiTBDM as the
sole product. Compared to PhSPh (entry 4), this is likely the
result of the better steric accessibility of the second sulfur once
the first has been functionalized by the bulky silane.
The reactions of entries 7 and 8 in Table 1 demonstrate that
the reaction can be extended to vinyl rather than aryl sulfides,
but the hydrogenation of the carbon-carbon double bond
becomes the dominant reaction with ethyl vinyl sulfide. With
vinyl phenyl sulfide (entry 8), the displacement of the vinyl
group is the more favorable process; however, activation of the
phenyl-sulfur bond as well as double activation of both the
phenyl and vinyl carbon sulfur bond also take place. The
theoretically possible product TBDMSi-S-CHdCH2 is not
observed, i.e., in all cases the vinyl group is either cleaved off
or hydrogenated. The overall product distributions depend
somewhat on the nature of the counterion in the catalyst
precursor, but they elude obvious interpretation. The hydrogen
required for the hydrogenation of the vinyl group can only
originate from the silane, either by activation of the silicon-
hydrogen bond by the nanoparticles or through the reaction of
the silane with trace amounts of water in the highly polar DMA
solvent leading to the formation of molecular hydrogen and
silanols. In either case, active hydrogen must be present on the
surface of the nanoparticles, effecting a heterogeneous hydro-
genation of the carbon-carbon double bond as previously
described for this type of system.39,40 Comparison of the
reactions of entries 2, 9, and 10 in Table 1 also shows that the
presence of an electron withdrawing or donating substituent in
the para position of the phenyl sulfide has no influence on the
reaction yield.
In contrast, PhSSPh (entry 5 in Table 1) either reacts not at
all or at best stoichiometrically with the silane (using Pd2(dba)3
as the nanoparticle precursor). Poisoning of heterogeneous
precious metal catalysts by sulfur or thiols is a well-established
phenomenon36 and probably operates here as well, i.e., the
catalysts are poisoned by interaction between the surface of
Pd(0) nanoparticles and the sulfur atoms of S-Ph fragments
formed through the cleavage of relatively weak S-S bonds in
the disulfides. Thermodynamically S-S bond cleavage in
PhSSPh (∼230 kJ/mol) is substantially easier than that of
C(sp2)-S bonds (∼370 kJ/mol), and it has been reported that
the disulfide bond can be cleaved by a nucleophililc substitution
reaction through the combined catalytic action of both an
electrophile and a nucleophile.37,38 To promote electrophilic S-S
bond cleavage, metal ions with a high affinity for sulfur, such
as palladium, silver, and mercury, are commonly used.
To test the hypothesis of catalyst deactivation by Ph-S
fragments resulting from the cleavage of PhSSPh, we carried
out a series of control reactions, in which various phenyl sulfides
including HOCH2CH2SPh (cf. entry 1 of Table 1) were added
to DMA solutions containing PhSSPh, excess amounts of
TBDMSiH, and Pd(0) nanoparticles pregenerated from PdCl2.
Except for a trace amount of silane alcoholysis product of
HOCH2CH2SPh, only the starting sulfides and PhSSPh were
detected in all cases. The control reactions thus assert that an
irreversible reaction occurs between PhSSPh and the palladium
surface, which poisons the catalyst as subsequently added phenyl
sulfides were not converted to the corresponding thiosilanes and
only marginal catalytic activity for silane alcoholysis is observed.
When the two sulfur atoms are separated by a methylene unit
as in PhSCH2SPh (entry 6, Table 1) the selective activation of
the C(sp2)-S bond is again achieved, but under conditions
identical to those of the reaction of diphenyl sulfide (2.3 equiv
To evaluate the actual synthetic utility of the new method,
some thiosilanes were isolated by solvent removal in vacuo
followed by flash column chromatography or vacuum distillation
(Table 2). Thiosilanes are thermally stable, but the Si-S bond
is readily cleaved by many protic reagents, particularly those
containing the O-H group, in part because the formation of
relatively stronger Si-O single bonds (∼800 kJ/mol)41 com-
pared to the Si-S single bonds (300-330 kJ/mol).1 Silica gel
adsorbent for column chromatography contains Si(OSi≡)4,
Si(OSi≡)3(OH), and Si(OSi≡)2(OH)2 groups, which explains
the substantially lower yields obtained by chromatography and
makes either distillation or crystallization the isolation method
of choice for this class of compounds.
The effect of the nature of the silane on the C(sp2)-S bond
cleavage and silicon-sulfur cross-coupling reaction was ex-
(36) Wilke, S. Appl. Phys. A 1996, A63, 583-593.
(39) Aiken, J. D., III; Finke, R. G. J. Mol. Catal. A 1999, 145, 1-44.
(40) Bo¨nnemann, H.; Richards, R. M. Eur. J. Inorg. Chem. 2001, 2455-2480.
(41) Weast, R. C. CRC Handbook of Chemistry and Physics, 64th ed.; CRC
Press: Boca Raton. FL, 1984.
(37) Bach, R. D.; Rajan, S. J. J. Am. Chem. Soc. 1979, 101, 3112-3114.
(38) Kice, J. L. In The Sulfur-Sulfur Bond; Senning, A., Ed.; Marcel Dekker
Inc.: New York, 1971; Vol. 1, pp 153-207.
9
7388 J. AM. CHEM. SOC. VOL. 126, NO. 23, 2004

Pd-Catalyzed Silane Thio Ether Cross-Coupling
A R T I C L E S
Table 3. Effect of Various Silanes on the Palladium
Nanoparticle-Mediated Cross-Coupling Reaction with Ethyl Phenyl
Sulfide at 298 Ka
Table 4. Results of the Characterization of the Palladium Colloids
by TEM
catalyst precursor
entry
silane (2.3 equiv)
yields(%) of EtS−SiR a
3
PdCl2
100
25
2.04
0.57
1.03
3.40
Pd(OAc)2
Pd(OCOCF )
Pd2(dba)3
3 2
1
2
3
4
5
Et3SiH
quant.
quant.
quant.
42
no. of particles sampled
temp (°C)
mean size (nm)
ESD
min. size (nm)
max. size (nm)
213
25
62.21
33.74
21.83
207.71
100
25
1.52
0.30
0.91
2.61
100
25
1.49
0.31
0.86
2.15
t-BuMe2SiH
i-Pr3SiH
PhMe2SiH
Ph3SiH
0
a
Compound identity determined by comparative GC-MS with authentic
samples. Yields are determined by quantitative GC.
in the rapid formation of nanoparticles, while with the Pd(0)
precursor Pd2(dba)3 and silane alone, clear purple homogeneous
solutionssthe original color of Pd2(dba)3sinitially result.
However, upon addition of sulfides or methanol, the solution
turns dark black within ∼30 min, and TEM reveals the presence
of nanoparticles, which are catalytically active. Pd(PPh3)4 does
not give nanoparticles under any of the reaction conditions
investigated. The Pd(0) nanoparticles obtained from all usable
precursors have been characterized by transmission electron
microscopy (TEM) with size distributions of the nanoparticles
determined over an ensemble of at least 100 individually
measured particles. The particle size distribution was reproduc-
ible. Table 4 summarizes the results of the TEM analyses of
all four nanoparticle colloids obtained in DMA from PdCl2, Pd-
(OAc)2, Pd2(dba)3, and Pd(OCOCF3)2 at 25 °C in the presence
of a 40-fold excess of t-BuMe2SiH. The nanoparticle nature of
the catalysts is a necessary condition for the sulfur-silicon
cross-coupling reaction to occur because neither bulk Pd(0)
black nor Pd(PPh3)4 are catalytically active. From the combined
results in Table 1 and Table 4, the actual particle size is,
however, not critical for the catalytic activity since relatively
smaller particles derived from PdCl2 (∼2 nm) show similar
catalytic activity as much larger Pd(OAc)2 (∼60 nm) derived
particles.
amined with ethyl phenyl sulfide as the substrate and PdCl2 as
the palladium nanoparticle precursor. Reactions were moni-
tored by GC and analyzed for compound identity by GC-MS
(Table 3).
From the results in Table 3, it is apparent that the C(sp2)-S
bond cleavage is not influenced by the steric properties of the
silane employed, as i-Pr3SiH having the most sterically demand-
ing environment activates C(sp2)-S bonds quantitatively, while
Ph3SiH bearing the less bulky phenyl substituents does not.
Instead, the yields of the cross-coupling reaction appear to be
dependent on the electronic properties of the substituents present
on silicon: silanes bearing only alkyl substituents (entries 1-3
in Table 3) lead to a quantitative conversion, while the exchange
of one alkyl for a phenyl substituent leads to the much lower
yield of 42% or no conversion at all for triphenyl silane. This
is in contrast to the high reactivity of all these silanes in
nanoparticle-catalyzed silane alcoholysis reactions. For example,
both Et3SiH and Ph3SiH are quantitatively consumed in reactions
with ethylene glycol42 and sugars,16 respectively, giving the
corresponding silyl ethers in very high yields. Quantum me-
chanical calculations on the free silanes (see Supporting Infor-
mation for details) show that the HOMO energies of the si-
lanes monotonously increase in the order Et3SiH < i-Pr3SiH <
TBDMSiH < PhMe2SiH < Ph3SiH and that the LUMO ener-
gies as well as ∆ELUMO-HOMO monotonously decrease in the
same order. While the absolute orbital energies of any silyl
species involved will change relative to those of the free silanes
during the reaction (see Discussion section for a possible
mechanism), their relative energetic ordering should remain the
same. Extrapolating from the relative orbital energies of the free
silanes, the absolute hardness43,44 of a surface-adsorbed Ph3Si
fragment should then be lower compared to that of any R3Si
(R ) alkyl) fragments. This results in more synergistic bonding,
i.e., a larger amount of back-bonding from the palladium into
an energetically more accessible LUMO and hence stronger
adsorption of Ph3Si to the surface, potentially explaining its low
reactivity.45
Application to Sugar Model System. Returning to our
original objective of expanding the scope of our previously
described method for the silylation of sugars,16 we further
investigated the influence of the nanoparticle precursor and the
amount of silane on the reaction and product distribution with
phenyl-1-thio-â-D-glucopyranoside as the model substrate.
Entries 1-7 of Table 5 list the results of these reactions, and
Figure 1 shows the structures and silylation patterns (1,2; 1;
1,2,6; 6; or 3,6) of some of the products. Mixtures of
chromatographically separable products originating from cross-
coupling and silane alcoholysis or silane alcoholysis are always
obtained, but in no case was exclusive thiosilyl glycoside
formation without silane alcoholysis observed.
Silane alcoholysis is precluded when the fully protected sugars
phenyl 2,3,4,6-tetra-O-benzyl-1-thio-â-D-glucopyranoside and
phenyl 2,3,4,6-tetra-O-acetyl-1-thio-â-D-glucopyranoside are
reacted with PdCl2 and 1.3 equiv of silane (entries 9 and 10,
Table 5); however, the cross-coupling product is only generated
for the perbenzylated phenyl thio-â-D-glucopyranoside, but not
with the peracetylated phenyl thio-â-D-glucopyranoside. This
is in contrast to the result of a simple acetylated sulfide (Entry
2, Table 2), which is formed in at least 60% yield. This renders
improbable a direct reactivity of the silane with the acetyl group
Nature of the Catalysts. We and others have previously
shown that PdCl2, Pd(OAc)2, and Pd(hfacac)2 give nanoparticles
when treated with a silane in DMA or less polar organic
solvents.16,46 The nanoparticle formation is complete within
seconds of silane addition to a solution of the palladium
precursor. Under the same conditions, Pd(OCOCF3)2 also results
(42) Chung, M.-K.; Ferguson, G.; Robertson, V.; Schlaf, M. Can. J. Chem. 2001,
79, 949-957.
(43) Parr, R. G.; Pearson, R. G. J. Am. Chem. Soc. 1983, 105, 7512-7516.
(44) Pearson, R. G. Inorg. Chem. 1988, 27, 734.
(45) A more complete computational study modelling the relative orbital energies
of surface-adsorbed silyl species on actual and well-defined palladium nano
clusters may be able to explain the observed reactivity differences more
rigorously but is at present beyond our capabilities.
(46) Fowley, L. A.; Michos, D.; Luo, X.-L.; Crabtree, R. H. Tetrahedron Lett.
1993, 34, 3075.
9
J. AM. CHEM. SOC. VOL. 126, NO. 23, 2004 7389

A R T I C L E S
Chung and Schlaf
Table 5. Results of Silylation Reactions of Phenyl Thio Glucopyranosides with in Situ Generated Pd0 Nanoparticles and t-BuMe2SiH at 298
K in DMA
yields (%) of cross-coupling
products
yields (%) of O-silylated
derivatives
nanoparticle
precursor
amount of
silane (equiv)
total yields (%) of
silylated derivatives
entry
thioglycosides
1,2 or 1
1,2,6
6
3,6
a
1
2
3
4
5
6
7
8
phenyl-1-thio-â-D-glucopyranoside
PdCl2
1.3
2.5
3.5
1.3
2.3
3.5
1.3
1.3
-
17
45
61
7
18
35
21
-
47
-
8
72b
64b,c
80b
50
65
71
-
19
19
-
-
-
6
-
-
Pd(OAc)2
31
31
22
-
12
16
14
61
-
Pd(OCOCF3)2
PdCl2
87
70
phenyl-2,3,4,6-tetra-O-benzyl-1-thio-
â-D-glucopyranoside
phenyl-2,3,4,6-tetra-O-acetyl-1-thio-
â-D-glucopyranosidef
70d,e
-
9
PdCl2
1.3
n.r.
n.r.
n.r.
n.r.
0
a
b
Traces of other cross-coupling products (<3%) were produced, but could not be identified because of decomposition during workup. 4,6-Disilylated
c
d
e
derivatives (<2%) were generated. Trisilylated derivatives (<10%) were also produced. Cross-coupling product at anomeric position (C-1). Reaction
carried out at 1/3 experimental scale. No reaction.
f
Figure 1. Structures and silylation patterns of the reaction products listed in Table 5.
as the origin of this effect. The adsorption of one or several
acetyl groups on the surface of the palladium catalyst may
explain the lack of reactivity of the peracetylated sugar.
Scheme 1
Discussion
Possible Nanoparticle Surface Composition and Mecha-
nistic Considerations. The fact that the nanoparticle catalyst
is only stable against agglomeration and precipitation of bulk
palladium as long as excess silane is present in solution16 has
to date prevented us from gaining further insights into the actual
surface structure of the nanoparticles, a common problem with
this type of system.47 In the absence of such structural
information, any statements about a possible mechanism of the
C(sp2)-S bond activation are speculative. However, the neces-
sary presence of excess silane for colloid stability along with
several other pieces of ancillary information derived from the
macroscopically observed reaction outcomes and GC-MS
analysis of the reaction solution give some clues about the
surface structure of the nanoparticles and also allow the
exclusion of certain reaction pathways.
GC-MS analysis of nanoparticle colloids obtained from the
reaction of PdCl2 and excess TBDMSi-H reveals the presence
of (TBDMSi)2, whose formation is a possible consequence of
the reduction of Pd2+ to Pd0 by the silane. Starting from PdCl2,
Scheme 1 suggests plausible pathways for the initial nucleation
and subsequent Pd0s possibly autocatalyticsparticle growth,
generating (TBDMSi)2 and HCl. Previous experiments with
added base,16 as well as the fact that Pd(OAc)2, Pd(OCOCF3)2,
and Pd2(dba)3 result in active catalysts for silane alcoholysis
and the cross-coupling reaction (Table 1), preclude any role of
HCl or other acids in the catalytic reaction and also make a
role for the corresponding anion in the stabilization of the colloid
unlikely.
The fact that excess silane is necessary to stabilize the
nanoparticles against agglomeration and precipitation demands
that, once all palladium has been reduced, the resulting particles
are stabilized through silane adsorbed on the surface, whose
steric bulk then prevents agglomeration as has also been
proposed for other systems with bulky polymeric or surfactant-
type stabilizers.40,47 Since the observed reactivity requires a
breaking of the silicon hydrogen bond, we propose a dissociative
adsorption of the silane on the palladium nanoparticle surface
as illustrated in Figure 2 as the first reaction step. Reductive
elimination of two adjacent trialkyl silyl fragments then
constitutes a plausible pathway for the formation of (TBDMSi)2
and the presence of hydrogen adsorbed on the particle surface
accounting for the observed hydrogenation of vinyl sulfides.
The proposed dissociative adsorption of the silane mirrors
the behavior of the molecularly disperse Pd(0) complex [(µ-
dcpe)Pd]2, dcpe ) 1,2-bis(dicyclohexylphosphino)ethane, that
has recently been shown to activate silanes R3SiH (R ) phenyl,
(47) Finke, R. G. In Transition-Metal Nanoclusters: Solution-Phase Synthesis,
Then Characterization and Mechanism of Formation, of Polyoxoanion- and
Tetrabutylammonium-Stabilized Nanoclusters; Feldheim, D. L., Foss, C.
A., Jr., Eds.; Marcel Dekker: New York, 2002; pp 17-54.
9
7390 J. AM. CHEM. SOC. VOL. 126, NO. 23, 2004

Pd-Catalyzed Silane Thio Ether Cross-Coupling
A R T I C L E S
Figure 2. Proposed silane activation pathway and nanoparticle surface structure.
Figure 3. Proposed C(sp2)-S bond activation pathway. (Only the reacting adsorbed silane is shown.)
Electron microscopy was carried out on a LEO 912AB operating at
100 kV with a liquid nitrogen anticontaminator in place. Digital
images were collected using a 1 K × 1 K PROSCAN CCD camera
and processed using the measurement software in the SIS EsiVision
program. Both negatively stained (using 2% w/v uranyl acetate) and
unstained samples were imaged for measurement comparison. GC
quantitative analyses were carried out using naphthalene as the inter-
nal standard. GC-MS spectra were obtained in EI mode. Isomer
assignments of the silylated sugar species were made on the basis of
2D COSY and HSQC spectra (see Supporting Information for de-
tails). Sugar substrates, solvents, silanes, and metal salts were purchased
from commercial sources and used as received. DMA was dried by
vacuum distillation from BaO and subsequently stored over activated
4 Å molecular sieves. Flash column chromatography was performed
on wet-packed silica gel (60 µm) at 1 psi static pressure set by a break-
through valve at the column head. Details for the chromatographic
separations of the individual sugars are given in the Supporting
Information.
alkyl) through oxidative addition, yielding the mononuclear silyl
palladium hydrides (dcpe)Pd(H)SiR3.48 From a comparison of
the activation barrier for this process, the relative reactivity of
the silanes in this reaction directly parallels that in the
nanoparticle-catalyzed reaction (Table 3): the greater the
number of phenyl groups on the silane, the greater its tendency
to bind (and remain bound) to the palladium and the lower its
subsequent reactivity. The dissociative adsorption pathway is
also congruent with the mechanism of silane alcoholysis on bulk
palladium metal derived by Sommer and Lyons on the basis of
stereochemical studies that showed an inversion of stereochem-
istry of chiral silanes in this reaction, which is best explained
by a “backside attack” of an alcohol onto a SiR3 fragment
adsorbed on the metal surface.49
In analogy, we postulate a “backside attack” of a thioether
on a surface-adsorbed silicon center as the second reaction step
in the nanoparticle-catalyzed formation of the thiosilanes. This
scenario is also consistent with the poisoning of the catalyst by
PhS fragments generated from the disulfide PhSSPh, as it
excludes an actual adsorption of sulfur to the nanoparticle
surface, which should equally lead to catalyst poisoning and
therefore seems unlikely. Instead, an activation of the C(sp2)-S
bond must occur, in which the carbon rather than the sulfur
atom is adsorbed to the nanoparticle surface. A reductive
elimination of the resulting surface-bound phenyl or ethyl groups
(resulting from the hydrogenation of vinyl by insertion into a
palladium hydrogen bond) and surface bound hydrogen then
leads to the formation of benzene or ethane. Figure 3 sum-
marizes this proposed pathway.
General Procedure for the Silylation of Sulfide or Sugar
Substrates Using Pd Catalyst Precursors. A DMA (3 mL) solution
of the substrate (1.0 mmol) was prepared in a one-necked round-bottom
flask fitted with a Schlenk stopcock. t-BuMe2SiH (1.3-3.5 equiv
depending on the experiment performed) was added to a second Schlenk
flask containing the catalyst precursors (3 mol % with respect to
substrate) in 3 mL DMA and stirred for 1 min to give a black colloidal
solution. Subsequently the substrate solution was added, the apparatus
connected to a gas bubbler, and the reaction stirred until hydrogen gas
evolution stopped (2-24 h) and TLC indicated completion of the
reaction. DMA was removed by rotary evaporation at 45 °C under full
oil-pump vacuum. Flash column chromatography of the residue with
ethyl acetate and hexane in various ratios removes the catalyst and
yields the silylated substrates and also separated isomers. For simple
sulfides, Kugelrohr vacuum distillations using a vacuum controller were
Experimental Section
All synthetic experiments were performed under a dry argon
atmosphere by standard Schlenk-tube techniques. Sample solutions for
TEM were prepared inside an inert-gas drybox under argon atmosphere.
(48) Boyle, R. C.; Mague, J. T.; Fink, M. J. J. Am. Chem. Soc. 2003, 125, 3228-
3229.
(49) Sommer, L. H.; Lyons, J. E. J. Am. Chem. Soc. 1969, 91, 7061-7066.
9
J. AM. CHEM. SOC. VOL. 126, NO. 23, 2004 7391

A R T I C L E S
Chung and Schlaf
carried out for compound isolation. See Supporting Information for
detailed spectral and other data on individual compounds.
Supporting Information Available: Comprehensive collec-
1
tion of 400 MHz H and 13C NMR data (COSY, HSQC) for
sugar and (new) thiosilane products with images of all spectra
and peak assignments; conditions for chromatographic or dis-
tillation isolation; images of GC-MS spectra for thiosilanes
and silthianes; and HOMO/LUMO energies of free silanes as
determined by HF 6-311g(d,p) calculations (PDF). This material
is available free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. Funding was provided by the Natural
Science and Engineering Research Council of Canada (NSERC),
the Canadian Foundation for Innovation (CFI), the Ontario
Innovation Trust Fund (OIT), and the University of Guelph.
Electron microscopy was performed by Robert Harris in the
NSERC Guelph Regional STEM Facility, which is partially
funded by an NSERC Major Facilities Access Grant.
JA049386U
9
7392 J. AM. CHEM. SOC. VOL. 126, NO. 23, 2004