Organic Letters
Letter
opening by intermolecular attack of the borohydride13 (rather
than the intramolecular processes outlined before6,7) could
explain the diastereoselective formation of the trans-3,4-
disubstituted silolane. We present here a B(C6F5)3-catalyzed
formal (4 + 1) cycloaddition of VCPs and Et2SiH2.
extended to 14 days. Because of the minor quantities, the
silolane could not be obtained in analytically pure form.
We then subjected several homologues of VCP 1a having
longer and shorter alkyl chains to the general procedure (1b−e
with n = 4 to 0; Scheme 2). VCP 1b (n = 4) with a 4-
On the basis of our previous work,12a we started the
investigation with the reaction of VCP 1a and Et2SiH2 in
toluene with double the amount of B(C6F5)3 (1.0 versus 0.50
mol %). The conventional hydrosilylation product 3a was
obtained as the major product along with the cycloadduct 2a
(Table 1, entry 1). We examined the effect of different
Scheme 2. Substrate Scope I: Variation of the Internal
a
Substituent
a
Table 1. Optimization of Reaction Conditions
yield
b
b
entry B(C6F5)3 (mol %)
solvent
c (M) 2a (%)
3a (%)
1
2
3
4
5
6
7
8
1.0
3.0
5.0
7.0
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
PhCl
1
1
1
1
26
34
82
78
73
37
61
70
80
75
52
58
<5
47
67
52
13
<5
<5
45
25
8
<5
<5
<5
24
43
46
a
Reactions were performed on a 0.2 mmol scale according to the
optimized procedure (Table 1, entry 9). Unless otherwise noted, the
formation of side products was not observed. Yields are of analytically
pure material obtained after flash chromatography on silica gel. The
relative configuration was confirmed by NMR spectroscopic analysis
10
1
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
1.0
0.1
0.5
2
1
1
1
1
1
1
b
9
<5% of the hydrosilylation product obtained at a 1 mmol scale.
c
d
Formation of a tricyclic byproduct in 13% yield.7 Formation of the
10
11
12
13
14
PhF
ring-opened hydrosilylation product in 9% yield.12a Formation of the
e
CH2Cl2
benzene
n-hexane
toluene
conventional hydrosilylation product 3e in 73% yield.6,12a
c
phenylbutyl group furnished the silolane 2b in 61% yield. The
model substrate 1a (n = 3) afforded 2a in 69% yield on a 1
mmol scale. Conversely, VCPs 1c (n = 2) and 1d (n = 1) with
less methylene groups in their carbon chains reacted in
significantly lower yields of 40% for 2c and 30% for 2d.
Byproducts were a previously observed tricylic compound7 in
opened hydrosilylation product12a in the case of 1d (9% yield;
see Table S3). The phenyl-substituted VCP 1e (n = 0) gave
the conventional hydrosilylation product 3e in 73% yield (not
shown).6,12a
To assess the functional-group tolerance, a variety of VCPs
1f−n (n = 3) decorated with electronically modified aryl
groups were converted into the corresponding silolanes 2f−n
(Scheme 2). As for the parent VCP 1a, isolated yields were
consistently good, ranging from 60 to 75%. Larger aryl groups
such as naphthyl or biphenyl-4-yl groups as in 1o−q were not
detrimental, except for a biphenyl-2-yl group as in 1r.
The effect of an additional substituent on the cyclopropane
ring was probed in the reaction of several trans-disubstituted
VCPs 1s−v derived from 1a (Scheme 3). Substrates trans-1s−
u with different linear alkyl chains as the R group did undergo
the formal (4 + 1) cycloaddition in good yields yet with
complete loss of the stereochemical information. Silolanes 2s−
u were obtained as diastereomeric mixtures, and the reactions
were generally less clean. VCP trans-1v with R = Ph converted
into the ring-opened hydrosilylation product (not shown).12a
As outlined above, the proportion of the silolane 2a
increased with longer reaction times (Table 1, entry 14).
a
Unless otherwise noted, all reactions were performed on a 0.1 mmol
b
1
scale for 12 h. Yields were determined by H NMR spectroscopy
using CH2Br2 as an internal standard. The reaction time was 90 h.
c
amounts of the catalyst on the product distribution (entries 2−
5). The ratio of the two products shifted toward the silolane 2a
at higher catalyst loadings. While 2a did form exclusively with
more than 5.0 mol % of B(C6F5)3, the global yield slightly
decreased. The concentration was also crucial, and the
formation of the silolane 2a was found to be favored over 3a
with less solvent (entries 6−8). Solvents other than toluene
were tested with 5.0 mol % of B(C6F5)3 at a 1 M substrate
concentration (entries 9−13). The selectivity in favor of the
formal (4 + 1) cycloaddition further improved in polar aprotic
solvents, such as halogenated arenes and CH2Cl2 (entries 9−
11); the proportion of the hydrosilylation product was
substantial in benzene (entry 12). Conversely, almost none
of the cycloadduct formed in hydrocarbon solvents, such as n-
hexane (entry 13).
We note here that the product distribution was also
dependent on the reaction time (entry 1 versus entry 14);
longer reaction times led to more of 2a at the expense of 3a,
indicating that the conventional hydrosilylation product 3a can
be further converted into the silolane 2a (see the mechanistic
analysis). We also tested other dihydrosilanes, such as Ph2SiH2
and Me(Ph)SiH2, under the optimized reaction conditions of
entry 9. However, predominant formation of the hydro-
silylation products was seen, even when the reaction time was
4835
Org. Lett. 2021, 23, 4834−4837