ACS Catalysis
with Ti(O-i-Pr)4 (Table 1, entry 1), the use of NHC ligands pro-
Page 2 of 5
entry X
ligand
Ti(O-i-Pr)4 Et3SiH
1
2
3
4
5
6
7
8
vided good turnover at 120 °C. The NHC ligand IMes led to
poor reactivity in all cases, but the bulkier NHC ligands IPr and
IPrMe provided good yields of reduction product 2a derived
from reductive replacement of the silyloxy group using a tita-
nium reductant (Table 1, entries 2-4). The IPrMe ligand,7a in par-
ticular, provided a highly efficient process, with direct reductive
removal of the t-butyldimethylsilyloxy group being observed in
94% yield. In the above experiments using Et3SiH, aryl silane
2b was commonly seen in low yield along with product 2a.
However, in an experiment with the exceptionally bulky ligand
IPr*OMe,7b the production of aryl silane 2b was observed in
97% yield when Et3SiH was employed (Table 1, entry 5; more
complete Table S2 is available in the supporting information).
While not commonly employed in cross-coupling procedures,
triethylsilylarenes were recently demonstrated to be useful sub-
strates in palladium-catalyzed cross couplings.8 This outcome
is highly surprising considering the vast number of silane-me-
diated reductions of other aromatics, where hydrogen installa-
tion is uniformly observed. 1d,f,4b,5 For example, studies from
Martin and Chatani have described methods where aryl methyl
ethers were reduced to simple Ar-H products using silane re-
ductants. Furthermore, studies from Nakao described reduction
of silyloxyarenes (forming Ar-H products) using trialkylsilanes
with less hindered nickel NHC complexes rather than the for-
mation of arylsilanes described herein. The direct conversion of
phenol derivatives to aryl silanes using simple trialkylsilane re-
ductants thus complements alternate methods for installation of
C(sp2)-Si functionality and represents an unusual reactivity pat-
tern for silane-based reductions with nickel catalysts.9
% yield
of 2a
% yield
of 2b, 2a
1
Si(t-Bu)Me2 PCy3
Si(t-Bu)Me2 IMes
Si(t-Bu)Me2 IPr
--
--
2
16%
72%
94%
80%
10%, 14%
20%, 2%
18%, 5%
97% 1%
32%, 2%
90%, 1%
1%, 37%
3%, 4%
3
4
Si(t-Bu)Me2 IPrMe
5
Si(t-Bu)Me2 IPr*OMe
9
6
Si(i-Pr)3
SiEt3
Me
IPrMe / IPr*OMe 90%
IPrMe / IPr*OMe 49%
IPrMe / IPr*OMe 16%
IPrMe / IPr*OMe --
IPrMe / IPr*OMe 2%
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7
8
9
Piv
10
Tf
25%, 25%
R1
R1
R2
R2
R1 = H; R2 = Me; R3 = Me
R1 = H; R2 = i-Pr; R3 = H
IMes
IPr
N
N
IPrMe
IPr*OMe
R3
R1 = Me; R2 = i-Pr; R3 = H
R1 = H; R2 = CHPh2; R3 = OMe
R3
R2 R2
Experiments with Ti(O-i-Pr)4 use Ni(acac)2 as precatalyst, and experiments
with Et3SiH use NiCOD)2 as precatalyst. In entries 6-10, IPrMe·HCl was
used with Ti(O-i-Pr)4, IPr*OMe was used with Et3SiH. Piv = pivalate, Tf =
trifluoromethane sulfonate. Conditions for entries 6-10, using Ti(O-i-Pr)4:
Ni(acac)2 (5 mol%), IPrMe·HCl (10 mol%), Ti(O-i-Pr)4 (1.1 equiv), NaO-t-
Bu (2.5 equiv), toluene (0.5 M) at 120 °C for 6 h.. Conditions for entries 6-
10, using Et3SiH: Ni(COD)2 (10 mol%), IPr*OMe (10 mol%), triethylsilane
(6 equiv), NaO-t-Bu (2.5 equiv), toluene (0.5 M) at 120 °C for 6 h. Major
product for entries 9 and 10 was 4-phenylphenol. All yields determined by
GCFID with tridecane as internal standard.
The above experiments thus uncovered efficient methods for
the conversion of substrate 1 to either arene product 2a using
Ti(O-i-Pr)4 as reductant (Table 1, entry 4), or to aryl silane prod-
uct 2b using a trialkylsilane as reductant (Table 1, entry 5). The
above outcome with TBS ethers was then compared with other
types of silyl ethers as well as more conventional substrate clas-
ses. Using the bulkier Si(i-Pr)3 (TIPS) ether of the starting phe-
nol led to the formation of reduction product 2a in high yield,
whereas aryl silane 2b was obtained with poorer conversion us-
ing Et3SiH (Table 1, entry 6). In contrast, the smaller SiEt3
(TES) ether of the starting phenol was an efficient substrate for
formation of the aryl silane 2b, while formation of reduction
product 2a using Ti(O-i-Pr)4 was less efficient than the corre-
sponding reactions with bulkier silyl ethers (Table 1, entry 7).
Reactions of widely used methyl ethers, aryl pivalates, and aryl
triflates in either reaction with Et3SiH or Ti(O-i-Pr)4 as reduct-
ant were markedly less efficient and less selective in the for-
mation of 2a and 2b than the corresponding reactions of silylox-
yarene substrates, thus demonstrating synthetic utility of the si-
loxyarene substrate class (Table 1, entries 8-10). Based on the
promising outcomes of entries 4 and 5, TBS ethers were rou-
tinely used in the remainder of our studies.
With promising procedures in hand for the reductive conver-
sion of silyloxyarene substrates to either Ar-H or Ar-SiEt3 prod-
ucts, the scope and chemoselectivity of both procedures were
explored (Table 2). The use of Ti(O-i-Pr)4 to produce reduced
arenes was examined in toluene at 120 °C using Ni(acac)2 as the
precatalyst in combination with IPrMe·HCl as ligand since this
air-stable pre-catalyst generally provided excellent results. Sim-
ilar conditions with IPr*OMe as ligand were employed in silyl
transfer using silane reagents, but the more sensitive pre-cata-
lyst Ni(COD)2 provided superior results than air stable Ni(II)
sources. Under these conditions, simple naphthol derivatives
were cleanly reduced or silylated (3a-b and 4a-b) and an ar-
ylsilane functionality was tolerated in the process (5a-b). Addi-
tionally, benzyl silyl ethers, free phenols, and methoxyarenes
were similarly unaffected in substrates in which silyloxyarenes
were cleanly reduced or silylated (6a-b, 7a-b, 8a-b). Alterna-
tively, a substrate possessing two silyloxy groups underwent
bis-reduction or bis-silylation (9a-b). A variety of nitrogen-con-
taining substrates were tolerated, as evidenced by examples
with a quinoline, carbazole, or pyridyl functionality on the si-
lyloxy arene (10a-b, 11a-b, 12a-b). In contrast, morpholino-
and N-acylamino derivatives were sluggish substrates in the
Ti(O-i-Pr)4-mediated reduction, leading to modest conversion
after extended reaction times (13a, 14a). However, with this
substrate class, silylation with Et3SiH remained efficient and
high yielding (13b, 14b). A variety of frameworks that possess
alkyl substituents on the silyloxyarene were also tolerated in the
reductive transformation (15a-b, 16a-b, 17a-b). In the produc-
tion of aryl silane products, some variation in silane structure is
tolerated, as evidenced by the production of dimethylethylsilane
and dimethylisopropylsilane derivatives (18a-b). However,
benzyldimethylsilane underwent addition only in modest yields
Table 1. Optimization of nickel-catalyzed reductions of phe-
nol derivatives.
Ni(COD)2 or Ni(acac)2
(5-10 mol%)
Ligand (10 mol%)
H
SiEt3
OX
+
Ph
Ph
Ph
reductant
NaO-t-Bu (2.5 equiv)
toluene, 120 °C, 6-16h
2a
2b
1
ACS Paragon Plus Environment