2 . . . . . . . . . . . . . . . . . . . . . . . . . . . .Wang et al. Sci China Chem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Scheme 1 Synthetic methods for silanols (color online).
radiation at room temperature, an array of silanes was con-
verted into the corresponding silanols in exceptionally high
yields (92%–99%) within 12 h. Moreover, our studies un-
covered a reaction mechanism that is very different than the
ones reported in metal catalysis including Fu’s work.
were tested but none of them is comparable with THF (Table
1, entries 11–17). By note, the reaction in methanol only
produced the alcoholysis product Ph3SiOMe but not silanol
A number of controlled experiments were carried out to
dissect the roles of oxidants. Under nitrogen atmosphere,
water alone could not oxidize the silane (Table 1, entry 18).
Consistent with this observation, when an oxygen balloon
was used instead of open air, 90% conversion was achieved
in dry THF (i.e., without appreciable amount of water) after
72 h (Table 1, entry 19) and complete conversion could be
achieved within 12 h when water is added which gave silanol
2a (Table 1, entry 20). These experiments collectively sug-
gested that molecular oxygen but not water was likely the
oxidant. This is remarkably different from metal catalysis
[11c] as well as Rh(III) mediated photochemistry [12], where
water itself was a competent oxidant. Rose Bengal, visible
light and oxygen were all necessary for the success of this
transformation. The optimal conditions were thus identified
as listed in entry 2 (Table 1). The reaction time could be
shorten to 3 h when heating at 60 °C, which might be more
desirable for industrial application.
2 Experimental
To a solution of Rose Bengal (0.008 mmol, 2 mol%) in THF
2 mL was added silane 1 (0.4 mmol, 1 equiv.) and H2O (50
μL). The reaction mixture was open to the air (with a syringe
needle inserted into the rubber cover to intoduce the air into
the system and avoid the solvent evaporation heavily, the
same below), and stirred under the irradiation of a household
200 W white LED at room temperature for 12 h. After
completion of the reaction (monitored by GC-MS or TLC),
the solvent was removed under reduced pressure, the crude
product was purified by column chromatography with pet-
roleum ether/ethyl acetate (10:1) as eluent to give the desired
silanol 2.
With the optimal conditions in hand, we successfully ex-
tended the reaction to a scope of silane substrates. In all the
reactions, silanols were obtained as the only isolable product
and the formation of troubling siloxanes was not observed. In
the case of the hindered silane 1b, the reaction was more
sluggish: the desired silanol 2b was obtained in 78% yield
after 12 h irradiation (Table 2). However, extending the re-
action time to 24 h could achieve complete conversion and
an excellent yield (95%). Replacing one of the phenyl group
with either bulky tert-butyl group (1c) or smaller methyl
group (1d) did not affect the reaction, giving the product in
97% and 99% yield, respectively. Further introducing of a
second alkyl group was also well tolerated as shown in the
case of 1e (97% yield). Trialkylsilanes are also amenable to
this reaction despite of the alkyl group sizes, giving silanols
2f, 2g and 2h in excellent yields. Unsaturated carbon-carbon
bonds were compatible with this method: alkynyl silane 1i
and alkenyl silane 1j furnished the desired products in almost
quantitative yields, albeit the former required longer reaction
time. It is worthy to point out that dihydrosilane 1k, as well
as silane bearing an electron withdrawing group 1m, reacted
smoothly under the current reaction conditions. Importantly,
without tuning of reaction conditions except for reaction
3 Results and discussion
We started our investigation using the less reactive triphe-
nylsilane 1a as the model substrate. In the presence of
2 mol% of Rose Bengal in tetrahydrofuran (THF) at room
temperature in open air, triphenylsilanol 2a was isolated as
the single product in 67% yield (70% conversion as mon-
itored by crude 1 H NMR) after 12 h irradiation of white LED
(Table 1, entry 1). Gratifyingly, complete conversion of 1a
was obtained when 7 equiv. of water was added to the re-
action (Table 1, entry 2). Three other photocatalysts were
also examined but none of them appeared to be effective
(Table 1, entries 3–5). Importantly, very minimal conversion
was observed when the reaction was carried out in the ab-
sence of a photocatalyst (Table 1, entry 6) or without light
(Table 1, entry 7). The catalyst loading of the Rose Bengal
was examined, showing that increased loading from 2 mol%
to 5 mol% to 10 mol% slightly diminished the yields (cf.
Table 1, entry 2, 8, 9). When 1 equiv. of Rose Bengal was
added, the yield drastically dropped to 43% (Table 1, entry
10), probably due to the shield of light resulting from the
incomplete resolution of catalysts. A number of solvents