2
Y. Chen et al. / Tetrahedron Letters 61 (2020) 152235
reacted with chlorosilane, so the redistribution reaction between
hydrosilane and chlorosilane may happen. Herein we demon-
strated our effort on realization of a simple and efficient redistribu-
tion system of hydrosilane and chlorosilane, by using the most
readily available and cheapest alkali metal halide as catalyst. This
We then selected DMI as the solvent to optimize the catalyst. It
was found that compared with LiCl, the redistribution reaction
proceeded more slowly using NaCl or KCl as catalyst (see in sup-
porting information Figs. 36–38). Considering all of them can be
completely dissolved in DMI, the difference in reaction rate of
the redistribution may be due to the different solvation effect of
metal cation with the dipolar aprotic solvent, i.e., Li+ has a larger
solvation sheath than Na+ and K+ for its smaller cation size [9a],
which may be more conducive to the dissociation of metal chloride
ion pairs [9b]. The similar catalytic effect of these different metal
chlorides also supports our previous assumption that the dissoci-
ated chloride ion of the catalyst played a key role in this redistribu-
tion process. Besides, it was found that apart from the chlorides,
other common halides, such as KF, KBr, and KI, had similar catalytic
effect on the redistribution (Table 1, entries 14–16). These results
may reflect the similar Si-H bond activation effect of all the halo-
gen ions produced from the dissociation of halides in dipolar apro-
tic solvent. On the other hand, we also found that for the model
substrate system, ClCH2SiH3/ClCH2SiCl3, with each of these halides
salts as catalyst, the redistribution conversion efficiencies of the
reaction were close (Table 1, entries 2, 3, 11–16). As is known,
18-crown-6 was usually used to encapsulate K+ of KF to produce
naked F- for preparation of pentacoordinate silicates [10]. We spec-
ulated that the interaction of 18-crown-6 with K+ may also pro-
mote the redistribution of KF catalytic system and elevate the
redistribution conversion efficiency. Thus, the redistribution reac-
tion of ClCH2SiH3/ClCH2SiCl3 catalyzed by KF/18-crown-6 was car-
ried out to verify the point (Table 1, entries 17–18). Compared with
that of using KF alone as catalyst, the addition of 18-crown-6 as
cocatalyst increased the redistribution conversion of 0.5 h from
52% to 81% (Table 1, entries 13, 17). But no further increase of con-
version efficiency was observed even extending the reaction time
to 3 h. Considering the total yield of hydrochlorosilanes in the
same reaction time, as well as the cost, LiCl and KF/18-crown-6
are preferred catalysts in DMI.
On the basis of the above optimized reaction condition, we then
investigated the substrate scope of the reaction in DMI, with
KF/18-crown-6 or LiCl as the catalyst (Table 2). Various substituted
silane substrates, including chloroalkyl, alkyl, and phenyl silane,
were studied. Except for the systems of Et2SiH2/Et2SiCl2 and Ph2-
SiH2/Ph2SiCl2 (Table 2, entries 7, 8, 12, 13), all the redistribution
reactions proceeded efficiently with the yield mainly ranging from
44% to 71%. The large steric hindrance of double phenyl and ethyl
groups may influence the interaction between substrate and F- and
inhibited the Si-H bond activation effect, while the relatively better
conversion efficiency of Ph2SiH2/Ph2SiCl2 system may be attributed
to the higher electron-withdrawing ability of phenyl than ethyl
(Table 2, entries 7, 8, 12, 13). Compared with that of using LiCl as
catalyst alone, the redistribution conversion catalyzed by KF/18-
crown-6 is 5% higher for the substrates ClCH2SiMeH2/ClCH2-
SiMeCl2 and Cl2CHSiMeH2/Cl2CHSiMeCl2 (Table 2, entries 3–6),
but much lower for the substrates Et2SiH2/Et2SiCl2 and Ph2SiH2/
Ph2SiCl2 (Table 2, entries 7, 8, 12, 13). The results reflected that
KF/18-crown-6 catalytic system may be more easily influenced
by steric hindrance effect. For substrates with less crowded sub-
stituents, the introduction of 18-crown-6 may improve the redis-
tribution conversion slightly. Compared with previously reported
redistribution system [6], the very simple and mild reaction condi-
tion is one of the great advantages of this alkali metal halides cat-
alytic system. To further demonstrate the utility of the new system,
we then carried out a larger scale redistribution reaction to prepare
ClCH2SiMeHCl and Cl2CHSiMeHCl. The pure hydrochlorosilane pro-
duct ClCH2SiMeHCl or Cl2CHSiMeHCl were easily separated
through packed fractional distillation.
maybe will provide
hydrochlorosilanes.
a
more facile way to prepare
We first investigated the possible redistribution reaction of
ClCH2SiH3/ClCH2SiCl3 with AlCl3 as the catalyst. Although AlCl3
was reported can catalyze the redistribution reaction between
hydrosilane and chlorosilane at high temperature without solvent
[6a], no reaction was observed for the AlCl3 catalyzed ClCH2SiH3/
ClCH2SiCl3 system in tetrahydrofuran (THF) at room temperature
(Table 1, entry 1). The result may be due to that in THF, AlCl3 can-
not be well dissociated to produce enough chloride ion. To verify
the assumption, THF was replaced with 1,3-dimethyl-2-imidazo-
lidinone (DMI), which has a better ability to separate charges for
its higher dielectric constant [7] according to Coulomb’s law. As
expected, the redistribution reaction occurred quickly with two
hydrochlorosilanes formed in a total yield of 84% (Table 1, entry
2). The result verified presence of the possible Si-H/Si-Cl redistri-
bution reaction we assumed. Considering the better solubility of
lithium salts, we then used LiCl as the catalyst to investigate the
workable solvent for this redistribution system. It was found that
many solvents, including N, N’-1, 3-dimethylpropyleneurea
(DMPU), N-Methyl pyrrolidone (NMP), hexamethyl phosphoric tri-
amide (HMPA), tetrahydrofuran (THF), and diethylene glycol
dimethyl ether (diglyme) worked well for the redistribution reac-
tion between hydrosilane and chlorosilane (Table 1, entries 3–8).
Besides, the redistribution reaction did not take place in Bu2O or
Et2O (Table 1, entries 9 and 10), which may be due to the weaker
solvent polarity of them and the low solubility of LiCl in these sol-
vents [8a]. We also noticed that the reaction has lower conversion
efficiency in CH3CN, although it has higher polarity than diglyme.
The relatively low solubility of LiCl in CH3CN may be the reason
[8b]. These results indicated the appropriate polarity of solvent
and the solubility of catalyst in it are keys for the redistribution
reaction.
Table 1
Optimization of the reaction.a
Entry
Catalyst
Solvent
Time (h)
Products (yield)b
1
2
3
4
5
6
7
8
AlCl3
AlCl3
LiCl
LiCl
LiCl
LiCl
LiCl
LiCl
LiCl
LiCl
LiCl
NaCl
KCl
KF
KF
KBr
THF
DMI
DMI
20
3
3
3
3
3
3
3
3
20
20
10
10
0.5
10
10
10
0.5
3
No reaction
(1): 16%, (2): 68%
(1): 15%, (2): 70%
(1): 15%, (2): 68%
(1): 16%, (2): 61%
(1): 11%, (2): 70%
(1): 10%, (2): 69%
(1): 21%, (2): 18%
(1): 5%, (2): 8%
No reaction
DMPU
NMP
HMPA
diglyme
THF
CH3CN
Bu2O
Et2O
DMI
DMI
DMI
DMI
DMI
9
10
11
12
13
14
15
16
17
18
19
No reaction
(1): 16%, (2): 69%
(1): 16%, (2): 69%
(1): 28%, (2): 24%
(1): 15%, (2): 70%
(1): 15%, (2): 70%
(1): 14%, (2): 71%
(1): 16%, (2): 65%
(1): 14%, (2): 69%
KI
DMI
DMI
DMI
KF/18-crown-6
KF/18-crown-6
a
Reaction conditions: ClCH2SiH3 (0.005 mol), ClCH2SiCl3 (0.01 mol), catalyst
(3.0 mol/%), solvent (1 mL), room temperature.
To have an insight into the catalytic mechanism, we further
studied Si-H/Si-Cl redistribution reaction with PhSiH3/PhSiCl3 as
b
Yields were determined by 1H NMR.