Journal of the American Chemical Society
b,7
Article
6
Mechanistic Studies. This interesting transformation
displayed a remarkable selectivity that encouraged us to
elucidate the reaction mechanism. Usually, hydrosilylation of
alkenes with dihydrosilanes is facile under appropriate
of hydrosilane, which undergoes oxidative addition into one
of the Si−H bonds of dihydrosilane, affording the key
intermediate B. Then reductive elimination of dihydrogen
from intermediate B generates a Rh(I) silyl complex C, which
subsequently undergoes enantioselective C−H bond activa-
tion, affording intermediate D. Then reductive elimination of
intermediate D occurs to form the C−Si bond and Rh(I)
hydride species, which instantly proceeds oxidative addition
into the second Si−H bond, furnishing intermediate E. Finally,
insertion of alkene into intermediate E, followed by reductive
elimination from intermediate F, completes the catalytic cycle,
and it regenerates the catalytically active Rh(I) species. To
section 7.6). We were able to locate the C−H activation
transition states TS-major and TS-minor, which lead to the
two enantiomeric products. Various coordination types and
conformations were considered to ensure that the most
favorable transition states are located. TS-major is 5.0 kcal/
mol more favorable than TS-minor in terms of free energy at
room temperature, which is consistent with the observed
enantioselectivity. Nonetheless, these DFT results are very
preliminary and further detailed studies on the whole reaction
pathway is in progress, which will be disclosed in due course.
Substrate Scope. On the basis of the optimized conditions
and the preliminary mechanistic studies, we next explored the
substrate scope of this process. As the created silicon-
stereogenic centers are decorated with four groups, here we
used four colors to assess the scope of this reaction (Table 2).
First, biaryl dihydrosilane substrates bearing a wide range of
substituents on either the reacting C−H bond side of the
aromatic ring (light blue) or the silyl group tethered aromatic
ring (yellow) at different positions all reacted smoothly with
5f,m,n,9
transition-metal-catalyzed conditions,
and the generated
hydrosilylation monohydrosilanes can reasonably undergo the
C−H activation/silylation reaction. Therefore, a control
experiment in the absence of alkene was reinvestigated.
Under the optimized conditions only without the alkene
partner, the monohydrosilane product 5 could be obtained in
1
7% yield via C−H activation/silylation (Scheme 2b). Despite
low yield, the comparable 95% ee of 5 clearly indicates that the
stereodetermining step is the SiH -steered C−H activation/
2
silylation process. Further treatment of the resulting
enantioenriched monohydrosilane 5 with 3,3-dimethylbut-1-
ene under the Rh-catalyzed conditions using either (R,Sp)-
Josiphos L6 or (S,Rp)-Josiphos L6 or (racemic)-Josiphos L6
furnished the identical tetrasubstituted silane product 4 with
5
was subjected to the reaction conditions using (R,Sp)-
Josiphos L6, a racemic product 4 was obtained (Scheme 2c).
Moreover, the racemic monohydrosilane 6 was prepared and
also subjected to the standard reaction conditions. In this case,
no reaction occurred (Scheme 2d). These results further
suggest that monohydrosilane 5 is the key intermediate in the
reaction, and the chirality of the silicon-stereogenic center of 5
is induced in the first SiH -steered C−H activation/silylation
2
process, which leads to the formation of the final product 4. It
should be noted that the key intermediate monohydrosilane 5
is not stable under the Rh-catalyzed conditions, and the one-
pot trap of it with alkenes in a stereospecific fashion stabilized
the silicon-stereogenic center, providing a perfect strategy for
5m,10
the construction of asymmetrically tetrasubstituted silanes.
3
,3-dimethylbut-1-ene to afford the desired asymmetrically
To the best of our knowledge, this is also the first example of
rhodium-catalyzed stereospecific intermolecular hydrosilyla-
tion of alkene with retention of the silicon-stereogenic center.
In addition, this tandem C−H silylation/alkene insertion
tetrasubstituted silane products in high yields (55−88%) with
good to excellent enantioselectivities (81−99% ee). Electron-
donating groups, such as methyl, tert-butyl, methoxy groups
7−9, 19), and electron-withdrawing groups, such as
trifluoromethyl group (10) and halogen substituents (11, 17,
8), were well tolerated. Interestingly, compound 11 and
(
process also suggests that the SiH -steered intramolecular C−
2
H activation of dihydrosilane is more favored than the
intermolecular alkene hydrosilylation.
1
compound 17 are also a pair of enantiomers in this
transformation. Extended aromatic groups and heterocycles
such as furan, thiophene, and indole (12−16, 20) could also
be transformed into the 9-silafluorene scaffold successfully.
Next, changing the third substituents (green) on the
dihydrosilane substrates into functionalized aromatic rings,
heterocycles, and aliphatic methyl group also produced the
corresponding products in good yields without the loss of
enantioselectivities (21−26). For the scope of alkenes (blue),
we find that besides 3,3-dimethylbut-1-ene, other types of
alkenes containing adamantyl group, aromatic rings, amino
group, styrene, vinylsilane, and vinylgermane were competent
substrates, providing the desired highly functionalized silane
products in good yields with excellent enantioselectivities
(27−30, 32, 33). However, because of the unselective
hydrosilylation of carbonyl groups, ester or ketone functional
groups gave poor results in the transformation (31). In
addition, linear aliphatic alkenes (34) and vinyl ethers are not
very compatible with biaryl dihydrosilane substrates in the
tandem process. In these cases, direct hydrosilylation of the
Next, two parallel kinetic isotope effect (KIE) experiments
were carried out using 1-H and 1-D , as shown in Scheme 3a.
5
5
The reactions gave rise to a k /k value of 1.08. The lack of a
H
D
kinetic isotope effect rules out C−H bond cleavage occurring
11
during the turnover-limiting step in this tandem process.
Moreover, the reaction of dihydrosilane 3 with [Rh(cod)Cl]
2
31
1
and Josiphos L6 in toluene-d was monitored by P and H
8
NMR at room temperature (Scheme 3b). Two doublets of
doublets at 101.3 ppm (dd, JRh−P = 140.9 Hz, JP−P = 32.6 Hz,
P(tBu) ) and 28.9 ppm (dd, J
= 136.5 Hz, JP−P = 32.6 Hz,
2
Rh−P
3
1
P(Ph) ) were observerd in the P NMR spectrum. And two
2
broad rhodium hydride species (−6.28 and −3.32 ppm) were
1
formation of the rhodium silyl dihydride intermediate B,
which was proposed to be the catalyst resting state in the
7
a
catalytic cycle. On the basis of the results described above,
we propose that this SiH -steered tandem enantioselective C−
2
H silylation/alkene hydrosilylation process occurs by the
mechanism as shown in Scheme 3c. First, Rh(I) hydride
phosphine active catalyst species A is generated from precursor
Rh(cod)Cl dimer and chiral phosphine ligand in the presence
E
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX