Hidden Enantioselective Hydrogenation of N-Silyl Enamines and Silyl Enol Ethers in Net CN and CO Hydrosilylations Catalyzed by Ru-S Complexes with One Monodentate Chiral Phosphine Ligand

Article abstract of DOI:10.1021/acs.organomet.7b00030

Ruthenium thiolate complexes with one chiral monodentate phosphine ligand are applied to enantioselective hydrosilylation of enolizable imines and ketones. The structural features of the catalyst exclude the presence of more than one phosphine ligand at t

Full text of DOI:10.1021/acs.organomet.7b00030

Article
pubs.acs.org/Organometallics
Hidden Enantioselective Hydrogenation of N‑Silyl Enamines and Silyl
Enol Ethers in Net CN and CO Hydrosilylations Catalyzed by Ru−
S Complexes with One Monodentate Chiral Phosphine Ligand
Susanne Bahr and Martin Oestreich*
̈
Institut fur Chemie, Technische Universitat Berlin, Strasse des 17. Juni 115, 10623 Berlin, Germany
̈
̈
S
* Supporting Information
ABSTRACT: Ruthenium thiolate complexes with one chiral monodentate phosphine
ligand are applied to enantioselective hydrosilylation of enolizable imines and ketones.
The structural features of the catalyst exclude the presence of more than one phosphine
ligand at the ruthenium center in the enantioselectivity-determining step. The
enantiomeric excesses obtained in these reduction reactions are moderate (up to 66%
ee), but the stereochemical outcome enables an experimental analysis of the reaction
pathways operative in this catalysis. A two-step sequence consisting of successive N−
Si/O−Si dehydrogenative coupling and enamine/enol ether hydrogenation is the
prevailing mechanism of action. Both steps involve cooperative bond activation at the
Ru−S bond of the coordinatively unsaturated ruthenium complex: Si−H bond activation in the dehydrogenative coupling and
heterolytic H−H splitting in the hydrogenation. Previously documented side reactions such as deprotonation/protonation
equilibria as well as competing direct CN or CO hydrogenation have been excluded.
detail, revealing the formation of a ruthenium hydride together
with a sulfur-stabilized silylium ion ([1]+ → [2]+; Scheme 1,
top right).2 In collaboration with Ohki and Tatsumi, our group
was also able to catalytically access the electrophilic silicon
species [2]+, and several transformations including electrophilic
C−H silylations3 or hydrodefluorinations4 were accomplished.
A few of those catalytic reactions result in the formation of
chiral compounds such as benzosiloles5 and 4-substituted 1,4-
dihydropyridines6 (Scheme 1, bottom). These latter examples
are of particular interest, as enantioenriched silicon-stereogenic
benzosiloles are rare,7 and the enantioselective synthesis of N-
silylated 1,4-dihydropyridines is even unprecedented. Con-
versely, the asymmetric hydrosilylation of ketones8 and
ketimines9 can be regarded as a solved problem.10 Especially
the former of these two transformations serves as a benchmark
reaction for the purpose of evaluating the performance of new
INTRODUCTION
■
A few years ago, Ohki and Tatsumi reported the synthesis of
the ruthenium(II) thiolate complexes [1]+[BArF ]− (Scheme 1,
4
top left) and their application in the catalytic hydrogenation of
carbonyl compounds.1 The cooperative bond activation of
hydrosilanes with [1]+[BArF ]− was later investigated in great
4
Scheme 1. Potential Applications of [1]+[BArF ]− in
4
a
Asymmetric Catalysis
chiral catalysts. To make chiral congeners of [1]+[BArF ]−, we
4
considered several approaches to introduce chirality, one being
the replacement of the ligand L in [1]+[BArF ]− by a chiral
4
monodentate phosphorus ligand L*.11
However, several structural and electronic requirements must
be fulfilled to obtain catalytically active complexes. (1) The
vacant coordination site at ruthenium must be retained. This a
priori excludes bidentate ligands, e.g., binap, or ligands with
additional Lewis basic groups. (2) Steric bulk around and
electronic properties of the phosphorus atom are restricted.
Sterically demanding as well as electron-deficient ligands have
been shown not to be compatible with the electron-deficient
ruthenium center in Ohki−Tatsumi complexes.12 To us,
a
ArF = 3,5-bis(trifluoromethyl)phenyl.
Received: January 12, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.7b00030
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
monodentate phosphines appeared appropriate for the
Table 1. Synthesis of Chiral Ruthenium Chloride Complexes
preparation of chiral variants of [1]+[BArF ]−.13 Chiral
4
phosphines 3 with an asymmetrically substituted phosphorus
atom represent one class of such potentially suitable ligands
(Chart 1, top). 3 was initially applied in asymmetric
Chart 1. Monodentate Chiral Phosphine Ligands
a
b
entry
L*
T (°C)
complex
yield (%)
dr
1
2
3
(S)-6
(S)-7
(S)-8
65
80
65
(R,RuRS)-9
(S,RuRS)-10
(S,RuRS)-11
45
71
52:48
82:18
0
a
b
Isolated yields after flash column chromatography. Diastereomeric
1
ratio determined by H NMR spectroscopy.
complex 12 in toluene. (R,RuRS)-9 and (S,RuRS)-10 were
isolated in moderate and good yields, respectively, as a mixture
of two diastereomers (entries 1 and 2). However, coordination
of (S)-8 to the ruthenium center in 12 was not observed, and
the corresponding chloride complex (S,RuRS)-11 was not
obtained (entry 3). We ascribe this to the steric demand of
ligand (S)-8.
For the other two complexes, we noticed the predominant
formation of one diastereomer, especially in the case of
phosphepine (S)-7 as the ligand (forming (S,RuRS)-10). The
major diastereomer (S,RuR)-10 is obtained exclusively by
crystallization, as revealed by 1H NMR analysis of the
crystalline solid (Scheme 2, bottom). Another 1H NMR
hydrogenation14 and hydrosilylation15 with success, and further
applications with different metals are known today.13a,c,16 As
another ligand class, binol-derived phosphepines 4 introduced
by Gladiali17 and then further elaborated by Beller18 are
attractive. Apart from hydrogenation reactions,18a,b compounds
4 were also applied in enantioselective carbonyl hydro-
silylation.19 It is important to note though that, for application
of both 3 and 4, two ligand molecules are coordinating the
metal center during the catalysis, and the use of only 1 equiv of
the chiral phosphine on the basis of catalyst loading was shown
to be detrimental for both conversion and enantioinduction.20
The situation changes with Hayashi’s phosphines 5.13b,21 These
were especially designed for those cases where bidentate,
chelating phosphine ligands such as binap lead to less active
catalysts.22
Scheme 2. Configurational Instability of the Ruthenium
Center
We hence envisioned the synthesis of chiral ruthenium
thiolate complexes with just one monodentate chiral phosphine
as the origin of enantioinduction. One congener each of the
above ligands ((S)-6−(S)-8; Chart 1, bottom) was chosen to
replace the achiral phosphine ligand in [1]+[BArF ]−. The new
4
complexes have been applied as catalysts in the enantioselective
net hydrosilylation of CN and CO bonds involving
cooperative Si−H bond activation.23 The reaction is found to
follow a sequence of dehydrogenative N−Si or O−Si coupling
and hydrogenation of the intermediate N-silyl enamines and
silyl enol ethers, respectively.
RESULTS AND DISCUSSION
■
Our investigations on chiral ruthenium(II) thiolate complexes
started with the synthesis of monodentate phosphine ligands.
We decided to use one typical representative of each class
(Chart 1, bottom). (S)-6 described by Genet
̂ ́
and Juge,24
phosphepine (S)-7,18a,b and (S)-821 were synthesized according
to reported procedures. The targeted catalyst precursors
(R,RuRS)-9, (S,RuRS)-10, and (S,RuRS)-11 (Table 1) were
tackled following the established protocol for the preparation of
the achiral ruthenium chlorides. The indicated chiral phosphine
ligand was added to in situ prepared4,10b or isolated1 ruthenium
measurement of the same sample after 1 day showed that
epimerization at ruthenium of (S,RuR)-10 occurred in solution.
The originally obtained dr of 82:18 was reestablished (for a full
1H NMR spectrum, see the Supporting Information).
It was further possible to secure the molecular structure of
the major diastereomer (S,RuR)-10 by X-ray diffraction analysis
B
DOI: 10.1021/acs.organomet.7b00030
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(Scheme 3, left).25 The configurational lability of the ruthenium
center in the chloride complexes is irrelevant in subsequent
15), but in the case of (R,RuRS)-9 it was obtained in a low yield
of 20% and as a racemate (entry 1). (S,RuRS)-10 together with
NaBArF4 resulted not only in near-quantitative yield but also in
promising 53% ee, with (S)-15 being the major enantiomer
Scheme 3. Structure of (S,RuR)-10a and Synthesis of [(S)-
13]+[BArF ]−
(entry 2). Preformed chloride-free [(S)-13]+[BArF ]− furnished
4
4
the same enantioselectivity (entry 3),26 thereby making
interference by chloride-bridged dimers4 unlikely.27 It turned
out that the level of enantiocontrol did not vary at different
conversions; (S)-15 was obtained after 0.5 and 1 h in
diminished yields of 48% and 61% with the enantiomeric
excess unchanged (entries 3 and 4). Furthermore, considerable
amounts of enamine 17 were detected by NMR spectroscopy
when the reaction was stopped after 0.5 h (see the Supporting
Information), whereas 17 was almost completely reduced to
the amine (S)-16 after 3 h (entry 5). N- and C2-disilylated 18
which would result from 2-fold dehydrogenative coupling was
not observed.28 The substitution pattern on the silicon atom
had little effect on the enantioinduction: MePh2SiH, Et3SiH,
and Me2EtSiH reacted with selectivities similar to that of
Me2PhSiH (entries 6−8). However, a high yield was obtained
with MePh2SiH. With Ph3SiH (entry 9), both the enantiomeric
excess and yield dropped significantly. PhSiH3 resulted in full
conversion but hardly any enantioinduction (entry 10). In
addition, surrogate 1929 was suitable for the catalysis with [(S)-
a
Thermal ellipsoids represent the 50% probability level. Hydrogen
atoms and solvent molecules have been removed for the sake of clarity.
steps, as this stereoinformation is lost in the following chloride
abstraction. Treatment of (S,RuRS)-10 with NaBArF results in
4
13]+[BArF ]− under the same setup and resulted again in 53%
the formation of cationic complex [(S)-13]+[BArF ]−
4
4
((S,RuRS)-10 → [(S)-13]+[BArF ]−; Scheme 3, right). [(S)-
ee (see the box in the graphic of Table 2). This result shows
4
13]+[BArF ]− was isolated in 70% yield.
that cyclohexa-1,4-diene-based surrogate 19 can be activated by
4
[(S)-13]+[BArF ]−,30 thereby engaging in enantioselective
With the ruthenium complexes in hand, we began testing
their potential in enantioselective hydrosilylation. We used 1
mol % of either chloride (S,RuRS)-9 and (R,RuRS)-10 together
4
transfer hydrosilylation using chiral Ohki−Tatsumi complexes.
A similar screening was performed with acetophenone (20)
in combination with catalyst [(S)-13]+[BArF ]− (Table 3).
with NaBArF for the in situ generation of the active catalysts.
4
4
With hydrosilane Me2PhSiH, an even better 65% ee was
obtained for silyl ether (S)-21a; the corresponding silyl enol
ether 2210a was not isolated in this particular reaction (entry 1).
These were then combined with Me2PhSiH and phenyl-
protected imine 14 (Table 2). The corresponding amine 15
was isolated after hydrolysis of the N−Si bond (14 → 16 →
a
Table 2. Hydrosilylation of Imine 14
b
c
entry
catalyst system
Si−H
time (h)
yield of (S)-15 (%)
ee (%)
1
2
3
4
5
6
7
8
(R,RuRS)-9/NaBArF
Me2PhSiH
Me2PhSiH
Me2PhSiH
Me2PhSiH
Me2PhSiH
MePh2SiH
Et3SiH
22
22
0.5
1
20
99
48
61
96
93
61
71
35
99
0
53
54
53
54
53
54
55
36
4
(S,RuRS)-10/NaBArF
4
[(S)-13]+[BArF ]−
4
[(S)-13]+[BArF ]−
4
[(S)-13]+[BArF ]−
3
4
[(S)-13]+[BArF ]−
3
4
[(S)-13]+[BArF ]−
3
4
[(S)-13]+[BArF ]−
Me2EtSiH
Ph3SiH
3
4
d
9
[(S)-13]+[BArF ]−
3
4
e
10
[(S)-13]+[BArF ]−
PhSiH3
3
<5
4
a
b
All reactions were performed according to GP1 (see the Experimental Section) on a 0.2 mmol scale. Isolated yield after flash column
c
d
e
chromatography. Determined with HPLC on a chiral stationary phase. Performed with C6H6 (0.1 mL) as solvent. (R)-15 was obtained as major
enantiomer.
C
DOI: 10.1021/acs.organomet.7b00030
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a b
,
Table 3. Hydrosilylation of Ketone 20
Scheme 4. Mechanism of the Net CN and CO
Hydrosilylation
time
(h)
yield
ee
(%)
c
d
e
entry
Si−H
silyl ether
(%)
(S)-21:22
1
2
3
4
5
Me2PhSiH
Me2PhSiH
Me2PhSiH
MePh2SiH
Et3SiH
2
1.5
1
(S)-21a
(S)-21a
(S)-21a
(S)-21b
(S)-21c
93
nd
nd
65
66
66
66
51
100:0
91:9
68:32
100:0
100:0
f
3
82
82
3
a
All reactions were performed according to GP2 (see the Experimental
b
c
Section) on a 0.2 mmol scale. nd = not determined. Isolated yield
after flash column chromatography on silica gel. Determined with
HPLC on a chiral stationary phase. Ratio determined by H NMR
spectroscopy. Obtained along with (MePh2Si)2O (12%).
d
e
1
f
However, at shorter reaction times of 1 or 1.5 h, NMR
spectroscopic analysis indeed showed substantial amounts of 22
(entries 2 and 3; see the Supporting Information for further
details). Similar to the case for imine 14, the enantiomeric
excess of (S)-21a turned out to be independent of the reaction
time. Other hydrosilanes such as MePh2SiH and Et3SiH also
resulted in the formation of silyl ethers (S)-21b and (S)-21c,
respectively (entries 4 and 5). Again, the influence of the
hydrosilane on enantioinduction was rather minor. The
corresponding silyl ethers were not isolated in these cases.
With regard to the results previously obtained in the CN
and CO hydrosilylation, the high yields of amine (S)-15 and
silyl ethers (S)-21 were unexpected. For similar but achiral
concerted mechanism is conceivable for this hydrogenation,32 it
is assumed to occur stepwise. It commences with proton
transfer from VI to VII (VI + VII → IV + V). Hydride V then
serves as a reducing agent for IV, resulting in the products VIII.
This two-step pathway is a dehydrogenative coupling−
hydrogenation sequence resulting in net hydrosilylation of C
N and CO groups. The sequence involves both cooperative
Si−H and H−H bond activation. However, as the N-silyl
enamine and the silyl enol ether VII have never been obtained
exclusively, the direct hydrosilylation pathway is likely to
compete to a certain extent; VIII is then directly obtained
from IV through hydride transfer from V. Intermediates V and
IV (gray box in Scheme 4) in the enantioselectivity-
determining reduction step are formally the same for both
scenarios: i.e., the direct reduction pathway and the enamine or
enol reduction with dihydrogen. However, the enantiomeric
excess obtained in the irreversible hydride transfer could still be
significantly different for the following reason. The hydrosilane
addition to I occurs syn2 but with the chiral phosphine as ligand
L* diastereomers (S,RuR)-II and (S,RuS)-II are obtained. These
will form and react at different rates to yield diastereomeric
ruthenium(II) hydrides (S,RuS)-V and (S,RuR)-V. V formed
through dihydrogen activation with I and subsequent
protonation of VII will most likely differ in the diastereomeric
ratio. The more bulky silyl group (in comparison to the
hydrogen atom) will likely allow for better differentiation of the
diastereotopic sides in I.33 The discrete diastereomeric ratios
obtained for V, originating from either II or VI, will most
probably lead to different enantiomeric excesses of VIII.
Furthermore, the experimental data obtained with chiral
[1]+[BArF ]− the dehydrogenative N−Si coupling to the
4
corresponding N-silyl enamine 17 is the major reaction
pathway,10b and (depending on the hydrosilane) almost
exclusive formation of the silyl enol ether 22 by dehydrogen-
ative O−Si coupling is observed.10a These results are usually
obtained after short reaction times or with removal of
dihydrogen gas.31 In the present case, high yields of the
amine 15 (after hydrolysis) and silyl ethers 21 are formed after
prolonged reaction times in closed vessels: i.e., without release
of dihydrogen from the reaction mixture. For both 14 and 20,
the intermediate formation of dehydrogenative coupling
products, that is, silyl enamine 17 and silyl enol ether 22,
was verified. On this basis, a preliminary mechanistic picture
can be formulated (Scheme 4): the preferred reaction pathway
starts with hydrosilane activation by I, resulting in the
formation of hydrosilane adduct II (I → II). The silyl unit of
II is then transferred onto the Lewis basic nitrogen or oxygen
atom of starting material III; this step leads to silyliminium or
carboxonium ion IV (II + III → IV + V, gray box). On the
basis of the results obtained for short reaction times, the
catalytic cycle proceeds now with the deprotonation of IV by
the Lewis basic sulfur atom of the neutral ruthenium hydride
complex V. The CC bond in VII is formed along with the
dihydrogen adduct VI of the initial cationic complex (IV + V →
VI + VII). This adduct is in equilibrium with free dihydrogen
gas and the initial catalyst I (VI → I + H2).30 The backward
reaction, i.e., dihydrogen activation with I,1 eventually results in
hydrogenation of the double bond in VII. Even though a
catalyst [(S)-13]+[BArF ]− allow for the exclusion of potential
4
side-reaction pathways. Inconsistent ee values in the borane-
catalyzed imine hydrosilylation had been investigated in great
detail by us,34 revealing a competing deprotonation of
silyliminium ion IV (X = NR; Scheme 5, gray box) with
unreacted starting material III (III + IV → XI + VII, clockwise
pathway top left). The resulting iminium ion IX can also accept
a hydride from V to form X with unknown enantioinduction (V
D
DOI: 10.1021/acs.organomet.7b00030
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Scheme 5. Side Reactions in the Net CN and CO Hydrosilylation
a
For the sake of clarity, stereodescriptors have been omitted, and the diastereomers of V and VI are not shown.
+ IX → X + I, counterclockwise bottom right). X would
undergo dehydrogenative N−Si or O−Si coupling35 with I and
hydrosilane to form VIII (X + I + Si−H → VIII + I + H2). The
global ee would hence be determined by the reduction of two
different ions IV and IX. As III is consumed over the course of
the reaction, the amount of IX decreases with increasing
conversion and, consequently, the overall enantioselectivity
would depend on the reaction time. Consistent ee values for
both 14 and 20 were obtained in the reduction catalyzed by
substantial quantities. Furthermore, hydride transfer onto
iminium or carboxonium ion IX is independent of the
hydrosilane used in the net hydrosilylation. However, we did
observe different levels of enantioinduction in reduction of 14
and 20 with different hydrosilanes (Tables 2 and 3)
CONCLUSION
■
We disclosed here an enantioselective net hydrosilylation of
enolizable imines and ketones catalyzed by the chiral ruthenium
[(S)-13]+[BArF ]−, and this clearly excludes such a deprotona-
complex [(S)-13]+[BArF ]−. These reductions are a rare
4
4
tion−reduction sequence.
example of asymmetric catalysis where just one chiral
monodentate phosphine ligand at the transition metal is
responsible for the enantioinduction. The obtained enantio-
meric excesses are moderate, but this work still represents the
first example of an enantioselective net CX hydrosilylation
involving cooperative Si−H bond activation. It also includes the
first example of an enantioselective transfer hydrosilylation
employing a hydrosilane surrogate. An experimental mecha-
nistic analysis revealed that a two-step mechanism involving
successive dehydrogenative X−Si coupling and enamine/enol
ether hydrogenation is mainly operative. Both steps require
bond activation at Ru−S bond of the catalyst, that is, Si−H
bond activation in the first step and heterolytic H−H cleavage
in the second step.
However, dihydrogen activation with I (Scheme 5, gray oval)
and subsequent proton transfer of VI onto III also leads to IX
(VI + III → V + IX, counterclockwise top right). In addition to
the direct reduction with V (IX → X → VIII, counterclockwise
bottom right), the formation of XI through deprotonation by V
must be considered as another possibility (V + IX → XI,
clockwise bottom left).36 XI could then react in a dehydrogen-
ative coupling to VII (XI + I + Si−H → VII + I + H2).
Protonation by dihydrogen adduct VI would then result in the
two intermediates V and IV (VII + VI → V + IV). In principle,
the formation of V and IX from I is possible as the
hydrogenation of imines with [1]+[BArF ]− was recently
4
reported by us,30 and the hydrogenation of 20 with
[1]+[BArF ]− is also known.1 Transformations of hydrosilanes
4
and catalysts [1]+[BArF ]− are however generally more facile
EXPERIMENTAL SECTION
4
■
than those involving the splitting of dihydrogen. Hydrogenation
reactions require pressures between 5.0 and 10 bar for full
conversions of III.37,38 The formation of V and IX is hence less
favored compared to V and IV, and the following 3-step
sequence (Scheme 5, clockwise bottom left) seems very
unlikely. In addition, the direct reduction through hydride
transfer from V onto IX (V + IX → I + X, Scheme 5,
counterclockwise bottom right) and subsequent dehydrogen-
ative N−Si and O−Si coupling (X + I + Si−H → VIII + I +
H2) can be largely dismissed because N-silyl enamines and silyl
enol ethers are not formed by this pathway but are observed in
General Remarks. All reactions were performed in flame-dried
glassware using an MBraun glovebox (O2 <10 ppm, H2O <2 ppm) or
conventional Schlenk techniques under a static pressure of argon
(glovebox) or nitrogen. CH2Cl2, C6H6, and n-hexane were purified and
dried using an MBraun solvent system. Toluene was distilled over
sodium, degassed, and stored in the glovebox over 4 Å molecular
sieves. CD2Cl2 and C6D6 were degassed and stored in a glovebox over
4 Å molecular sieves. CDCl3 was stored over Cs2CO3. Technical grade
solvents for extraction and chromatography (cyclohexane, CH2Cl2,
ethyl acetate, and tert-butyl methyl ether) were distilled prior to use.
(S)-2,2′-Dimethyl-1,1′-binaphthalene,39 phosphine (S)-6,40 phosphine
(S)-8,21 2,6-dimesitylphenylthiole (HSdmp),41 dichloro(p-cymene)-
E
DOI: 10.1021/acs.organomet.7b00030
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
ruthenium(II) dimer,42 ruthenium complex 12,1 NaBArF ,43 imine
cyclohexane/tert-butyl methyl ether 100/0 → 20/1) to afford the
corresponding silyl ether (S)-21.
4
14,44 and surrogate 1929a were synthesized according to reported
procedures. TMEDA and dichlorophenylphosphine were distilled
before use. Hydrosilanes and acetophenone (20) were degassed and
stored in a glovebox over 4 Å molecular sieves. All other commercially
available reagents were used as received. Analytical thin-layer
chromatography (TLC) was performed on silica gel SIL G-25 glass
plates from Macherey-Nagel. Flash column chromatography was
Ruthenium Chloride Complex (R,RuRS)-9. 2,6-Dimesitylphe-
nylthiole (67 mg, 0.19 mmol, 2.0 equiv) was placed in a 25 mL
Schlenk flask, dissolved in THF (5 mL), and the resulting solution was
cooled to 0 °C. n-BuLi (2 M in hexanes, 0.09 mL, 0.2 mmol, 2 equiv)
was added, and the resulting mixture was stirred for 30 min at room
temperature. The mixture was then added via syringe to a suspension
of dichloro(p-cymene)ruthenium(II) dimer (60 mg, 0.097 mmol, 1.0
equiv) in THF (4 mL) at 0 °C. The resulting dark green suspension
was stirred at room temperature for 1 h, the solvent was removed
under reduced pressure, and toluene (7 mL) was added. The mixture
was filtered over a Schlenk frit into a solution of phosphine (S)-6 (35
mg, 0.19 mmol, 2.0 equiv) in toluene (4 mL). The reaction mixture
was stirred at 65 °C for 21 h. The solvent was removed under reduced
pressure, and the residue was subjected to flash column chromatog-
raphy on silica gel (eluent cyclohexane/tert-butyl methyl ether 1/2).
Ruthenium chloride complex (R,RuRS)-9 (56 mg, 0.085 mmol, 45%, dr
= 52:48) was obtained as a red-brown solid. Mp: 108−110 °C
(cyclohexane/tert-butyl methyl ether). Rf = 0.24 (cyclohexane/tert-
performed on silica gel 60 (40−63 μm, 230−400 mesh, ASTM) by
1
Merck using the indicated solvents. H, 7Li, 11B, 13C, 19F, 29Si, and 31
P
NMR spectra were recorded in CDCl3, C6D6, or CD2Cl2 on Bruker
AV 400, Bruker AV 500, and Bruker AV 700 instruments. Chemical
shifts are reported in parts per million (ppm) downfield from
tetramethylsilane and are referenced to the residual solvent resonance
1
as the internal standard (CHCl3, δ 7.26 ppm for H NMR; CDCl3, δ
1
77.16 ppm for 13C NMR; C6D5H, δ 7.16 ppm for H NMR; C6D6, δ
1
128.06 ppm for 13C NMR; CHDCl2, δ 5.32 ppm for H NMR; and
7
CD2Cl2, δ 53.84 ppm for 13C NMR). Li, 11B, 19F, 29Si, and 31P NMR
spectra were calibrated according to the IUPAC recommendation
using a unified chemical shift scale based on the proton resonance of
trimethylsilane as primary reference. Data are reported as follows:
chemical shift, multiplicity (sbr = broad singlet, s = singlet, d = doublet,
t = triplet, q = quartet, m = multiplet, mc = centrosymmetric
multiplet), coupling constant (Hz), integration. Signals labeled with
asterisks overlapped with the residual solvent signal and were detected
̃
butyl methyl ether 3/2). IR (ATR): ν 2914 (m), 2861 (m), 2725 (w),
2332 (w), 2116 (w), 1610 (w), 1574 (w), 1434 (m), 1382 (m), 1282
(m), 1215 (w), 1175 (w), 1107 (m), 1037 (m), 893 (m), 848 (m),
789 (m), 739 (s), 695 (s) cm−1. HRMS (ESI): calculated for
C35H42PRuS+ [M − Cl]+, 627.1783; found, 627.1794. NMR
spectroscopic data for the major diastereomer are as follows. 1H
1
1
by 2D measurements (1H,13C HMQC, H,13C HSQC, and H,13C
3
NMR (500 MHz, C6D6): δ/ppm 0.49 (t, JH,H = 6.6 Hz, 3H), 0.68−
HMBC). High-resolution mass spectrometry (HRMS) and elemental
2
0.82 (m, 3H), 1.17 (s, 3H), 0.89−0.97 (m, 1H), 1.42 (d, JH,P = 10.1
analysis were performed by the Analytical Facility of the Institut fur
̈
2
Hz, 3H), 1.44 (s, 3H), 1.97 (mc, 1H), 2.12 (d, JH,P = 3.0 Hz, 3H),
Chemie, Technische Universitat Berlin. Infrared (IR) spectra were
̈
2.26 (s, 3H), 2.36 (s, 3H), 2.43 (s, 3H), 2.40−2.49 (m, 1H), 3.99 (d,
recorded on a Agilent Technologies Cary 630 FTIR spectropho-
tometer equipped with an ATR unit and are reported as wavenumbers
(cm−1). Melting points (mp) were determined with a Leica Galen III
apparatus from Wagner & Munz and are not corrected. Enantiomeric
excesses were determined by high-performance liquid chromatography
(HPLC) analysis on an Agilent Technologies 1290 Infinity instrument
with a chiral stationary phase using Daicel Chiralcel OJ-H or OD-H
columns (n-heptane/isopropyl alcohol mixtures as solvent) or on an
Agilent Technologies 1200 Infinity instrument with a stationary phase
using a Daicel Chiralcel OJ-RH column (acetonitrile/water mixtures as
solvent). Absolute configurations were assigned by comparison of the
retention times of the enantiomers of 1544 and 2145 with literature
data. Data for the single-crystal structure determination were collected
with an Agilent SuperNova diffractometer equipped with a CCD area
Atlas detector and a mirror monochromator by utilizing Cu Kα
radiation (λ = 1.5418 Å). Software packages used: CrysAlis PRO for
data collection, cell refinement, and data reduction,46 SHELXS-97 for
structure solution,47 SHELXL-97 for structure refinement,48 and
Mercury 3.1.149 for graphics.
3
4
JH,P = 5.5 Hz, 1H), 5.14 (s, 1H), 6.77 (dd, JH,H = 6.9 Hz, JH,H = 1.7
3
Hz, 1H), 6.95 (t, JH,H = 6.9 Hz, 1H), 6.97−7.04 (m, 4H), 7.08 (mc,
2H), 7.61 (mc, 2H). 13C{1H} NMR (126 MHz, C6D6): δ/ppm 8.3 (d,
1JC,P = 34.4 Hz), 13.7 (s), 17.0 (s), 17.1 (s), 17.4 (s), 20.9 (s), 21.0 (s),
3
2
21.4 (s), 23.9 (d, JC,P = 10.7 Hz), 25.5 (d, JC,P = 19.5 Hz), 25.5 (d,
1JC,P = 30.9 Hz), 79.9 (s), 85.7 (s), 93.1 (s), 93.9 (d, JC,P = 11.4 Hz),
100.6 (d, JC,P = 5.4 Hz), 111.3 (s), 121.7 (s), 126.0 (s), 128.3 (mc,
2C),* 128.4 (s),* 129.2 (2s), 129.5 (d, 4JC,P = 1.9 Hz), 130.7 (d, 3JC,P
1
= 7.5 Hz, 2C), 135.4 (s), 136.3 (s), 136.6 (d, JC,P = 36.8 Hz), 136.8
(s), 137.3 (s), 138.4 (s), 142.8 (s), 159.4 (d, JC,P = 2.6 Hz). 31P{1H}
NMR (202 MHz, C6D6): δ/ppm 12.6. Selected NMR spectroscopic
1
data for the minor diastereomer are as follows. H NMR (500 MHz,
3
C6D6): δ/ppm 0.49 (t, JH,H = 6.6 Hz, 3H), 0.68−0.82 (m, 3H), 0.89
2
(s, 3H), 0.89−0.97 (m, 1H), 1.75 (d, JH,P = 10.1 Hz, 3H), 1.60 (s,
2
3H), 2.11 (d, JH,P = 3.2 Hz, 3H), 2.29 (s, 3H), 2.09−2.24 (m, 2H),
2.39 (s, 3H), 2.43 (s, 3H), 4.08 (d, JH,P = 5.2 Hz, 1H), 5.19 (s, 1H),
3
4
3
6.73 (dd, JH,H = 7.3 Hz, JH,H = 1.2 Hz, 1H), 6.82 (t, JH,H = 7.3 Hz,
1H), 6.97−7.04 (m, 4H), 7.08 (mc, 2H), 7.54 (mc, 2H). 13C{1H}
1
NMR (126 MHz, C6D6): δ/ppm 9.4 (d, JC,P = 31.7 Hz), 13.7 (s),
General Procedure for the Reduction of Imine 14 (GP1).
3
16.5 (s), 17.3 (2s), 20.9 (s), 21.0 (s), 21.4 (s), 23.8 (d, JC,P = 11.5
Hz), 24.5 (d, 1JC,P = 32.5 Hz), 25.5 (d, 2JC,P = 19.5 Hz), 80.5 (s), 85.1
(s), 91.7 (s), 92.8 (d, JC,P = 11.0 Hz), 100.9 (d, JC,P = 5.6 Hz), 111.4
(s), 121.9 (s), 126.0 (s), 128.3 (mc, 2C), 128.4 (s), 129.2 (2s), 129.4
(d, 4JC,P = 2.2 Hz), 130.7 (d, 3JC,P = 7.5 Hz, 2C), 135.4 (s), 136.3 (s),
Phenyl-protected imine 14 (39 mg, 0.20 mmol, 1.0 equiv) was added
to a mixture of NaBArF (1.8 mg, 2.0 μmol, 1.0 mol %) and the
4
indicated ruthenium chloride complex (R,RuRS)-9 (1.3 mg, 2.0 μmol,
1.0 mol %) or (S,RuRS)-10 (1.7 mg, 2.0 μmol, 1.0 mol %) or to the
preformed catalyst [(S)-13]+[BArF ]− (3 mg, 2 μmol, 1 mol %) in a
1
4
136.6 (d, JC,P = 30.0 Hz), 136.8 (s), 137.4 (s), 138.5 (s), 142.8 (s),
159.4 (d, JC,P = 3.2 Hz). 31P{1H} NMR (202 MHz, C6D6): δ/ppm
13.4. Elemental analysis results were outside the tolerance range.
Ruthenium Chloride Complex (S,RuRS)-10. Phosphepine (S)-7
(0.12 g, 0.30 mmol, 1.0 equiv) was added to ruthenium complex 12
(0.19 mg, 0.30 mmol, 1.0 equiv), and the solids were dissolved in
toluene (5 mL). The mixture was heated to 80 °C for 21 h. The
solvent was removed under reduced pressure, and the resulting residue
was directly subjected to flash column chromatography on silica gel
(eluent: cyclohexane/tert-butyl methyl ether 3:2). Ruthenium chloride
complex (S,RuRS)-10 (0.18 g, 0.21 mmol, 71%, dr = 82:18) was
obtained as a red-brown solid. Crystallization from a CH2Cl2/
cyclohexane solution afforded single crystals from the major
diastereomer suitable for X-ray analysis (see the Supporting
Information for further details). Mp: 170−173 °C (cyclohexane/tert-
butyl methyl ether). Rf = 0.2 (cyclohexane/tert-butyl methyl ether 3/
GLC vial. The indicated hydrosilane or surrogate (0.20 mmol, 1.0
equiv) and, in the case of Ph3SiH, C6H6 (0.1 mL) were added. The
mixture was stirred at room temperature for the indicated time,
followed by the addition of cyclohexane (0.5 mL). Direct submission
to flash column chromatography on silica gel (eluent cyclohexane/
ethyl acetate 100/0 → 99/1) afforded amine (S)-15 as a clear liquid.
General Procedure for the Reduction of Ketone 20 (GP2).
Acetophenone (20; 24 mg, 0.20 mmol, 1.0 equiv) was placed in a GLC
vial, and the ruthenium catalyst [(S)-13]+[BArF ]− (3 mg, 2 μmol, 1
4
mol %) was added. The GLC vial was closed tightly, and the
corresponding hydrosilane (0.20 mmol, 1.0 equiv) was rapidly added
through the septum of the GLC cap with a gastight syringe. The
mixture was stirred at room temperature for the indicated time,
followed by the addition of cyclohexane (0.5 mL). The mixture was
directly subjected to flash column chromatography on silica gel (eluent
F
DOI: 10.1021/acs.organomet.7b00030
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
4
2). HRMS (ESI): calculated for C52H46PRuS+ [M − Cl]+, 835.2096,
2C), 131.8 (s), 132.2 (s), 132.2 (s), 132.5 (d, JC,P = 2.1 Hz), 132.8
(s), 133.4 (s), 133.4 (s), 133.9 (s), 134.8 (s), 134.9 (s), 135.3 (s),
135.4 (s, 8C), 135.8 (s), 137.9 (s), 142.9 (s), 162.4 (m), 162.7 (q, 1JC,B
= 49.9 Hz, 4C). 11B{1H} NMR (160 MHz, CD2Cl2): δ/ppm −6.6.
19F{1H} NMR (471 MHz, C6D6): δ/ppm −62.1. 31P{1H} NMR (203
MHz, C6D6): δ/ppm 40.4. Elemental analysis results were outside the
tolerance range.
̃
found, 835.2084. IR (ATR): ν 3046 (w), 2915 (m), 2847 (m), 2341
(w), 2114 (w), 1899 (w), 1718 (w), 1507 (m), 1432 (m), 1376 (m),
1250 (m), 1210 (m), 1104 (m), 1028 (m), 933 (m), 830 (s), 738 (s),
695 (s) cm−1. NMR spectroscopic data for the major diastereomer
(S,RuR)-10 are as follows. 1H NMR (500 MHz, CD2Cl2): δ/ppm 1.40
(s, 3H), 1.54 (s, 3H), 1.95 (s, 3H), 1.97 (s, 3H), 2.01 (d, JH,P = 3.6 Hz,
3H), 2.39 (s, 3H), 3.01 (dd, 2JH,P = 16.9 Hz, 2JH,H = 12.6 Hz, 1H), 3.38
(S)-N-(1-Phenylethyl)aniline ((S)-15). Prepared according to
2
2
2
(dd, JH,H = 14.8 Hz, JH,P = 3.6 Hz, 1H), 3.73 (dd, JH,H = 14.8 Hz,
2JH,P = 9.5 Hz, 1H), 4.30 (dd, 2JH,H = 12.6 Hz, 2JH,P = 2.9 Hz, 1H), 4.48
(d, JH,P = 5.1 Hz, 1H), 5.53 (s, 1H), 6.43 (s, 1H), 6.72 (dd, 3JH,H = 6.5
Hz, 4JH,H = 2.2 Hz, 1H), 6.90 (s, 1H), 6.90−6.93 (m, 2H), 6.96−6.97
(m, 2H), 7.08−7.11 (m, 1H), 7.17 (mc, 1H), 7.32−7.35 (m, 3H), 7.39
GP1 from imine 14 (39 mg, 0.20 mmol, 1.0 equiv), catalyst [(S)-
13]+[BArF ]− (3 mg, 2 μmol, 1 mol %), and Me2PhSiH (40 μL, 0.20
4
mmol, 1.0 equiv). After it was stirred for 3 h at room temperature, the
reaction mixture was directly subjected to flash column chromatog-
raphy on silica gel (eluent cyclohexane/ethyl acetate 100/0 → 99/1).
Amine (S)-15 (38 mg, 0.19 mmol, 96%, 54% ee) was obtained as a
clear liquid. HRMS (ESI): calculated for C14H16N+ [M + H]+,
198.1277; found, 198.1280. 1H NMR (500 MHz, CDCl3): δ/ppm 1.55
3
(mc, 2H), 7.48 (t, JH,H = 7.4 Hz, 1H), 7.62 (mc, 2H), 7.65−7.66 (m,
3
1H), 7.73 (mc, 2H), 7.77 (mc, 1H), 7.86 (d, JH,H = 8.2 Hz, 1H).
13C{1H} NMR (126 MHz, CD2Cl2): δ/ppm 17.1 (s), 17.3 (s), 17.7
(s), 19.8 (s), 20.7 (s), 21.4 (s), 28.0 (d, 1JC,P = 25.0 Hz), 31.2 (d, 1JC,P
= 24.7 Hz), 82.3 (s), 89.2 (s), 94.0 (s), 94.2 (JC,P = 11.3 Hz), 102.1 (d,
JC,P = 6.2 Hz), 110.1 (s), 121.8 (s), 125.2 (s), 125.6 (s), 125.7 (s),
126.1 (s), 126.2 (s), 127.0 (s), 127.4 (s), 127.6 (s), 128.0 (s), 128.1
3
3
3
(d, JH,H = 6.7 Hz, 3H), 4.49 (q, JH,H = 6.7 Hz, 1H), 6.57 (d, JH,H
=
3
8.1 Hz, 2H), 6.69 (mc, 1H), 7.10 (mc, 2H), 7.23 (tt, JH,H = 7.2 Hz,
4JH,H = 1.5 Hz, 1H), 7.32 (mc, 2H), 7.36−7.39 (m, 2H), the NH signal
was not detected. 13C{1H} NMR (126 MHz, CDCl3): δ/ppm 24.7 (s),
54.4 (s), 114.2 (s, 2C), 118.2 (s), 126.2 (s, 2C), 127.2 (s), 128.8 (s,
2C), 129.2 (s, 2C), 144.6 (s), 146.4 (s). Elemental analysis results
were outside the tolerance range. The enantiomeric excess was
determined by HPLC analysis on a chiral stationary phase (Daicel
Chiracel OD-H column, column temperature 20 °C, solvent n-
heptane/isopropyl alcohol 90/10, flow rate 0.7 mL/min, λ 250 nm): tR
= 11.6 min for (S)-15, tR = 14.4 min for (R)-15. If C6D6 (0.4 mL) is
added to the reaction mixture after 0.5 h and the mixture is directly
submitted to NMR spectroscopic analysis, the corresponding N-silyl
enamine 17 is detected together with the N-silyl amine (S)-16 (see the
Supporting Information for a 1H NMR spectrum). Selected NMR
spectroscopic data for (S)-16 (Si = SiMe2Ph) are as follows. 1H NMR
(s), 128.3 (s), 128.3 (s), 128.8 (d, 3JC,P = 8.4 Hz, 2C), 129.1 (d, 3JC,P
=
1.8 Hz), 129.2 (s), 129.3 (s), 130.0 (s), 130.4 (s), 130.4 (d, 2JC,P = 6.7
Hz, 2C), 132.5 (s), 132.7 (d, 2JC,P = 6.7 Hz), 132.7 (2s), 132.9 (d, 2JC,P
3
3
= 11.3 Hz), 133.3 (s), 133.7 (d, JC,P = 3.2 Hz), 134.7 (d, JC,P = 2.5
Hz), 135.7 (s), 136.0 (s), 136.2 (s), 136.5 (s), 137.7 (s), 138.5 (d, 1JC,P
= 32.8 Hz), 142.2 (s), 157.7 (s). 31P{1H} NMR (202 MHz, CD2Cl2):
δ/ppm 52.8. Selected NMR spectroscopic data for the minor
diastereomer (S,RuR)-10 are as follows. 1H NMR (500 MHz,
CD2Cl2): δ/ppm 1.06 (s, 3H), 1.81 (s, 3H), 2.05 (d, JH,P = 3.8 Hz,
1H), 2.08 (s, 3H), 2.14 (s, 3H), 2.30 (s, 3H), 3.06 (dd, 2JH,P = 16.9 Hz,
2
2
2JH,H = 12.9 Hz, 1H), 3.54 (dd, JH,H = 14.7 Hz, JH,P = 8.4 Hz, 1H),
2
2
2
3.63 (dd, JH,H = 14.7 Hz, JH,P = 3.5 Hz, 1H), 3.91 (dd, JH,H = 12.9
(500 MHz, C6D6): δ/ppm 0.21 (s, 3H), 0.22 (s, 3H), 1.39 (d, 3JH,H
7.0 Hz, 3H), 4.66 (q, JH,H = 7.0 Hz, 1H). Selected NMR
spectroscopic data for 17 (Si = SiMe2Ph) are as follows. H NMR
(500 MHz, C6D6): δ/ppm 0.33 (s, 3H), 0.35 (s, 3H), 5.07 (s, 1H),
5.45 (s, 1H). The analytical data for 15,44 16,34 and 1734 are in
accordance with those reported.
=
Hz, 2JH,P = 2.1 Hz, 1H), 4.87 (d, JH,P = 5.6 Hz, 1H), 5.28 (s, 1H), 6.88
3
(dd, JH,H = 7.0 Hz, 4JH,H = 1.5 Hz, 1H), 7.00 (t, JH,H = 7.0 Hz, 1H).
13C{1H} NMR (126 MHz, CD2Cl2): δ/ppm 16.4 (s), 17.1 (s), 17.6
(s), 20.6 (s), 20.9 (s), 21.3 (s), 30.6 (d, 1JC,P = 21.2 Hz), 31.6 (d, 1JC,P
= 26.9 Hz), 80.9 (s), 85.2 (s), 93.4 (d, JC,P = 11.2 Hz), 94.9 (s), 100.3
(d, JC,P = 3.0 Hz), 115.0 (s), 121.9 (s), 126.1 (s), 128.1 (s), 128.2 (s),
129.0 (s), 136.0 (s), 136.1 (s), 136.5 (s), 136.7 (s), 138.0 (s), 142.0
3
3
1
(S)-Dimethylphenyl(1-phenylethoxy)silane ((S)-21a). Pre-
pared according to GP2 from acetophenone (20; 24 mg, 0.20
(s), 158.4 (d, JC,P = 1.7 Hz). 31P{1H} NMR (202 MHz, C6D6): δ/
3
mmol, 1.0 equiv), catalyst [(S)-13]+[BArF ]− (3 mg, 2 μmol, 1 mol
4
ppm 51.2. Elemental analysis results were outside the tolerance range.
The crystallographic data are available online in the CCDC database
under number CCDC 1521692.
%), and Me2PhSiH (31 μL, 0.20 mmol, 1.0 equiv). After 2 h at room
temperature, the reaction mixture was directly subjected to flash
column chromatography on silica gel (eluent cyclohexane/tert-butyl
methyl ether 100/0 → 20/1). Silyl ether (S)-21a (48 mg, 0.19 mmol,
93%, 65% ee) was obtained as a clear liquid. HRMS (ESI): calculated
for C16H19OSi+ [M − H]+, 255.1200; found, 255.1196. 1H NMR (500
Cationic Ruthenium Complex [(S)-13]+[BArF ]−. NaBArF4 (0.10
4
g, 0.11 mmol, 1.0 equiv) was added to a solution of ruthenium
chloride complex (S)-10 (0.10 g, 0.11 mmol, 1.0 equiv) in CH2Cl2 (3
mL), and the mixture was stirred at room temperature for 3 h. The
resulting green suspension was filtered over a PTFE syringe plug and
rinsed with CH2Cl2 (3 × 2 mL). The solvent of the filtrate was
evaporated under reduced pressure to afford the cationic ruthenium
3
MHz, CDCl3): δ/ppm 0.31 (s, 3H), 0.36 (s, 3H), 1.44 (d, JH,H = 6.4
Hz, 3H), 4.84 (q, 3JH,H = 6.4 Hz, 1H), 7.22−7.26 (m, 1H), 7.29−7.33
(m, 4H), 7.35−7.42 (m, 3H), 7.56−7.58 (m, 2H). 13C{1H} NMR
(126 MHz, CDCl3): δ/ppm −1.2 (s), − 0.7 (s), 27.0 (s), 71.2 (s),
125.6 (s, 2C), 127.0 (s), 127.9 (s, 2C), 128.3 (s, 2C), 129.7 (s), 133.7
(s, 2C), 138.3 (s), 146.4 (s). 29Si{1H} DEPT NMR (99 MHz, CDCl3,
optimized for J = 7 Hz): δ/ppm 6.6. Elemental analysis results were
outside the tolerance range. The enantiomeric excess was determined
by HPLC analysis on a chiral stationary phase (Daicel Chiracel OD-
RH column, column temperature 20 °C, solvent acetonitrile/water
60/40, flow rate 0.3 mL/min, λ 210 nm): tR = 35.2 min for (S)-21a, tR
= 37.5 min for (R)-21a. In cases of reaction times between 0.5 and 1.5
h, silyl ether 22 is obtained as a side product. Selected NMR
complex [(S)-13]+[BArF ]− (0.13 g, 77 μmol, 70%) as a green solid.
4
HRMS (ESI): calculated for C52H46PRuS+ [M − BArF ]+, 835.2096;
4
1
found, 835.2103. H NMR (500 MHz, C6D6): δ/ppm 1.05 (s, 3H),
1.20 (s, 3H), 1.35 (s, 3H), 1.74 (s, 3H), 1.99 (s, 3H), 1.99 (s, 3H),
2.43 (mc, 1H), 2.52 (dd, 2JH,H = 14.3 Hz, 2JH,P = 4.3 Hz, 1H), 2.66 (d,
2JH,H = 12.6 Hz, 1H), 3.04 (dd, 2JH,P = 16.3 Hz, 2JH,H = 12.6 Hz, 1H),
3
3.50 (s, 1H), 3.96 (s, 1H), 6.42 (d, JH,H = 8.3 Hz, 1H), 6.52 (s, 1H),
6.70 (s, 1H), 6.77−6.80 (m, 3H), 6.94−7.00 (m, 5H), 7.04−7.06 (m,
1H), 7.10−7.15 (m, 2H), 7.18−7.26 (m, 4H), 7.36 (d, 3JH,H = 8.4 Hz,
3
3
1
1H), 7.52 (d, JH,H = 8.3 Hz, 1H), 7.55 (d, JH,H = 8.1 Hz, 1H), 7.65
spectroscopic data for 22 (Si = SiMe2Ph) are as follows. H NMR
(sbr, 4H), 7.69 (d, JH,H = 8.2 Hz, 1H), 8.38 (sbr, 8H). 13C{1H} NMR
3
(500 MHz, CDCl3): δ/ppm 0.55 (s, 6H), 4.35 (d, 2JH,H = 1.9 Hz, 1H),
4.87 (d, JH,H = 1.9 Hz, 1H). The analytical data for 21a45 and 2210a
2
(126 MHz, C6D6): δ/ppm 17.4 (s), 17.7 (s), 18.0 (s), 20.2 (s), 20.5
1
1
(s), 21.0 (s), 31.5 (d, JC,P = 27.4 Hz), 33.0 (d, JC,P = 23.9 Hz), 73.5
(s), 73.6 (s), 103.1 (s), 105.0 (s), 105.5 (s), 107.9 (s), 118.1 (mc, 4C),
125.3 (q, 1JC,F = 272 Hz, 8C), 126.6 (s), 126.7 (s), 126.7 (s), 127.1 (s),
127.1 (s), 127.1 (s), 127.2 (s), 127.9 (s), 128.0 (s), 128.3 (s), 128.5
(s), 128.6 (s), 128.6 (s), 128.7 (s), 128.8 (s), 128.9 (s), 129.3 (d, 3JC,P
are in accordance with those reported.
(S)-Methyldiphenyl(1-phenylethoxy)silane ((S)-21b). Pre-
pared according to GP2 from acetophenone (20; 24 mg, 0.20
mmol, 1.0 equiv), catalyst [(S)-13]+[BArF ]− (3 mg, 2 μmol, 1 mol
4
%), and MePh2SiH (40 μL, 0.20 mmol, 1.0 equiv). After 3 h at room
temperature, the reaction mixture was directly subjected to flash
column chromatography on silica gel (eluent cyclohexane/tert-butyl
2
4
= 9.7 Hz, 2C), 129.9 (qq, JC,F = 31.2 Hz, JC,F = 2.8 Hz, 8C), 129.9
1
2
(s), 130.0 (s), 130.8 (d, JC,P = 35.7 Hz), 131.2 (d, JC,P = 10.7 Hz,
G
DOI: 10.1021/acs.organomet.7b00030
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
methyl ether 100/0 → 20/1). Silyl ether (S)-21b (52 mg, 0.16 mmol,
82%, 66% ee) was obtained as a clear liquid along with small amounts
of (Ph2MeSi)2O (12%). HRMS (ESI): calculated for C21H21OSi+ [M
REFERENCES
■
(1) Ohki, Y.; Takikawa, Y.; Sadohara, H.; Kesenheimer, C.;
Engendahl, B.; Kapatina, E.; Tatsumi, K. Chem. - Asian J. 2008, 3,
1625−1635.
1
− H]+, 317.1356; found, 317.1362. H NMR (500 MHz, C6D6): δ/
ppm 0.51 (s, 3H), 1.40 (d, 3JH,H = 6.4 Hz, 3H), 4.91 (q, 3JH,H = 6.4 Hz,
1H), 7.06 (tt, 3JH,H = 7.3 Hz, 4JH,H = 1.3 Hz, 1H), 7.13−7.19 (m, 8H),
7.28−7.30 (m, 2H), 7.60−7.66 (m, 4H). 13C{1H} NMR (126 MHz,
CDCl3): δ/ppm −2.2 (s), 27.2 (s), 72.0 (s), 125.0 (s), 127.3 (s), 128.2
(s, 2C),* 128.2 (s, 2C),* 128.2 (2C),* 130.0 (s), 130.1 (s), 134.8 (s,
2C), 134.9 (s, 2C), 136.9 (s), 137.0 (s), 146.6 (s). 29Si{1H} DEPT
NMR (99 MHz, CDCl3, optimized for J = 7 Hz): δ/ppm −3.9.
Elemental analysis results were outside the tolerance range. The
enantiomeric excess was determined by HPLC analysis on a chiral
stationary phase (Daicel Chiracel OJ-RH column, column temperature
20 °C, solvent acetonitrile/water 70/30, flow rate 0.4 mL/min, λ 254
nm): tR = 19.8 min for (R)-21b, tR = 23.9 min for (S)-21b. The
analytical data are in accordance with those reported.45
́
(2) Stahl, T.; Hrobarik, P.; Konigs, C. D. F.; Ohki, Y.; Tatsumi, K.;
̈
Kemper, S.; Kaupp, M.; Klare, H. F. T.; Oestreich, M. Chem. Sci. 2015,
6, 4324−4334.
(3) For the first application of [1]+[BArF ]− in catalytic Si−H bond
4
activation, see: (a) Klare, H. F. T.; Oestreich, M.; Ito, J.-i.; Nishiyama,
H.; Ohki, Y.; Tatsumi, K. J. Am. Chem. Soc. 2011, 133, 3312−3315.
For a recent summary of electrophilic C−H silylation including the use
of cationic ruthenium(II) complexes, see: (b) Bahr, S.; Oestreich, M.
̈
Angew. Chem., Int. Ed. 2017, 56, 52−59.
(4) Hydrodefluorination with [1]+[BArF ]− (L = (4-FC6H4)3P) is
4
described in: Stahl, T.; Klare, H. F. T.; Oestreich, M. J. Am. Chem. Soc.
2013, 135, 1248−1251.
(5) The synthesis of dibenzosiloles with [1]+[BArF ]− (L = (4-
(S)-Triethyl(1-phenylethoxy)silane ((S)-21c). Prepared accord-
4
ing to GP2 from acetophenone (20; 24 mg, 0.20 mmol, 1.0 equiv),
FC6H4)3P) is reported in: (a) Omann, L.; Oestreich, M. Angew. Chem.,
Int. Ed. 2015, 54, 10276−10279. For the synthesis of indole-fused
benzosiloles, see: (b) Omann, L.; Oestreich, M. Organometallics 2017,
36, 10.1021/acs.organomet.6b00801.
(6) The 1,4-hydrosilylation of pyridines is described in: Konigs, C. D.
F.; Klare, H. F. T.; Oestreich, M. Angew. Chem., Int. Ed. 2013, 52,
10076−10079.
catalyst [(S)-13]+[BArF ]− (3 mg, 2 μmol, 1 mol %), and Et3SiH (32
4
μL, 0.20 mmol, 1.0 equiv). After 3 h at room temperature, the reaction
mixture was directly subjected to flash column chromatography on
silica gel (eluent cyclohexane/tert-butyl methyl ether 100/0 → 20/1).
Silyl ether (S)-21c (39 mg, 0.16 mmol, 82%, 50% ee) was obtained as
̈
1
a clear liquid. H NMR (500 MHz, CDCl3): δ/ppm 0.57 (mc, 6H),
0.92 (t, 3JH,H = 7.9 Hz, 9H), 1.43 (d, 3JH,H = 6.4 Hz, 3H), 4.87 (q, 3JH,H
= 6.4 Hz, 1H), 7.22 (mc, 1H), 7.29−7.35 (m, 4H). 13C{1H} NMR
(126 MHz, CDCl3): δ/ppm 5.0 (s, 3C), 6.9 (s, 3C), 27.4 (s), 70.7 (s),
125.4 (s, 2C), 126.9 (s), 128.2 (s, 2C), 147.1 (s). 29Si{1H} DEPT
NMR (99 MHz, CDCl3, optimized for J = 7 Hz): δ/ppm 18.4.
Elemental analysis results were outside the tolerance range. The
enantiomeric excess was determined by HPLC analysis on a chiral
stationary phase (Daicel Chiracel OD-RH column, column temper-
ature 20 °C, solvent acetonitrile/water 70/30, flow rate 0.3 mL/min, λ
210 nm): tR = 21.8 min for (S)-21c, tR = 24.4 min for (R)-21c. The
analytical data are in accordance with those reported.45
(7) For the rhodium-catalyzed synthesis of chiral dibenzosiloles by
cycloadditions, see: (a) Shintani, R.; Otomo, H.; Ota, K.; Hayashi, T. J.
Am. Chem. Soc. 2012, 134, 7305−7308. (b) Shintani, R.; Takagi, C.;
Ito, T.; Naito, M.; Nozaki, K. Angew. Chem., Int. Ed. 2015, 54, 1616−
1620. For an example involving rhodium-catalyzed Si−C bond
activation, see: (c) Onoe, M.; Baba, K.; Kim, Y.; Kita, Y.; Tobisu, M.;
Chatani, N. J. Am. Chem. Soc. 2012, 134, 19477−19488. For a recent
summary, see: (d) Shintani, R. Asian J. Org. Chem. 2015, 4, 510−514.
(8) Zaidlewicz, M.; Pakulski, M. M. In Science of Synthesis
Stereoselective Synthesis 2; Molander, G. A., Ed.; Thieme: Stuttgart,
2011; pp 59−131.
(9) Xu, L.; Wu, X.; Xiao, J. In Science of Synthesis Stereoselective
Synthesis 2; Molander, G. A., Ed.; Thieme: Stuttgart, 2011; pp 251−
309.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.7b00030.
(10) The dehydrogenative coupling as the major reaction pathway
and minor hydrosilylation of ketones with [1]+[BArF ]− (L = Et3P) is
4
reported in: (a) Konigs, C. D. F.; Klare, H. F. T.; Ohki, Y.; Tatsumi,
̈
K.; Oestreich, M. Org. Lett. 2012, 14, 2842−2845. A similar
NMR spectra of the compounds synthesized in this
paper and crystallographic data (PDF)
Crystallographic data (CIF)
transformation of imines is documented with [1]+[BArF ]− (L =
4
iPr3P): (b) Hermeke, J.; Klare, H. F. T.; Oestreich, M. Chem. - Eur. J.
2014, 20, 9250−9254.
(11) For the synthesis and application of a cationic ruthenium(II)
AUTHOR INFORMATION
Corresponding Author
*E-mail for M.O.: martin.oestreich@tu-berlin.de.
■
complex where the thiolate ligand and the η6-mesityl ring are tethered
by a chiral binaphthyl backbone, see: Wubbolt, S.; Maji, M. S.; Irran,
E.; Oestreich, M. Chem. - Eur. J. 2017, 23, DOI: 10.1002/
chem.201700304.
(12) Sterically demanding phosphines such as tBu3P and 2-
dicyclohexylphosphino-2′-methylbiphenyl (MePhos) or the electron-
poor phosphine (C6F5)3P could not be applied to the synthesis of the
̈
ORCID
Susanne Bahr: 0000-0001-5113-7646
̈
Martin Oestreich: 0000-0002-1487-9218
Notes
catalyst precursors: Stahl, T. Ph.D. Thesis, Technische Universitat
Berlin, Berlin, Germany, 2014.
̈
The authors declare no competing financial interest.
(13) For a review of the synthesis and application of monodentate
chiral phosphines, see: (a) Lagasse, F.; Kagan, H. B. Chem. Pharm. Bull.
2000, 48, 315−324. (b) Hayashi, T. Acc. Chem. Res. 2000, 33, 354−
362. Asymmetric hydrogenations are discussed in, for example:
(c) Erre, G.; Enthaler, S.; Junge, K.; Gladiali, S.; Beller, M. Coord.
Chem. Rev. 2008, 252, 471−491.
ACKNOWLEDGMENTS
■
This research was supported in part by the Deutsche
Forschungsgemeinschaft (Oe 249/10-1). S.B. thanks the
Studienstiftung des deutschen Volkes for a predoctoral
fellowship (2015−2017), and M.O. is indebted to the Einstein
Foundation (Berlin) for an endowed professorship. Dr. Timo
Stahl (TU Berlin) is acknowledged for preliminary mechanistic
investigations, and we thank Dr. Elisabeth Irran (TU Berlin) for
the X-ray analysis as well as Franz-Lukas Haut (FU Berlin) for
experimental support.
(14) (a) Knowles, W. S.; Sabacky, M. J. Chem. Commun. 1968,
1445−1446. (b) Knowles, W. S.; Sabacky, M. J.; Vineyard, B. D. J.
Chem. Soc., Chem. Commun. 1972, 10−11.
(15) For the hydrosilylation of ketones with platinum complexes, see:
Yamamoto, K.; Hayashi, T.; Kumada, M. J. Organomet. Chem. 1972,
46, C65−C67. For a summary, see also ref 13b.
H
DOI: 10.1021/acs.organomet.7b00030
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(16) Several routes toward phosphorus-chiral phosphines have been
documented: Grabulosa, A.; Granell, J.; Muller, G. Coord. Chem. Rev.
2007, 251, 25−90.
(30) For the use of [1]+[BArF ]− with L = Et3P or (4-FC6H4)3P in
imine reductions, see: Lefranc, A.; Qu, Z.-W.; Grimme, S.; Oestreich,
M. Chem. - Eur. J. 2016, 22, 10009−10016.
4
(31) The reactions presented in this work are performed in closed,
small vials after fast addition of the hydrosilanes through septa (see the
Experimental Section for further details). Otherwise, dihydrogen gas is
immediately released from the reaction mixture. Reactions with 20
were usually stopped after 5 min,10a whereas those with imines were
performed in open vessels.10b
(32) The question of a two-step or one-step hydrogenation of the
double bond remains open. In the related imine hydrogenation,
hydride V and iminium ion IX do not exist as separated molecules,30
which could also hold for IV and V. On the other hand, a two-step
mechanism seems likely, in analogy to the corresponding hydro-
silylation.
(33) An NMR spectroscopic analysis of the hydrosilane adducts II is
in principle possible,2,11 whereas the spectroscopic evidence for the
dihydrogen adduct VI is elusive.1,30 We are hence unable to either
compare diastereoselectivities in the two activation scenarios or
correlate the preferred formation of one diastereomer with the
enantioselectivity obtained in the hydride transfer step.
(17) The first asymmetric synthesis was accomplished by Gladiali:
Gladiali, S.; Dore, A.; Fabbri, D.; de Lucchi, O.; Manassero, M.
Tetrahedron: Asymmetry 1994, 5, 511−514.
(18) The derivatization and a potential application of 4 was later
shown by Gladiali and Beller: (a) Junge, K.; Oehme, G.; Monsees, A.;
Riermeier, T.; Dingerdissen, U.; Beller, M. Tetrahedron Lett. 2002, 43,
4977−4980. (b) Junge, K.; Hagemann, B.; Enthaler, S.; Spannenberg,
A.; Michalik, M.; Oehme, G.; Monsees, A.; Riermeier, T.; Beller, M.
Tetrahedron: Asymmetry 2004, 15, 2621−2631. For a summary, see:
(c) Gladiali, S.; Alberico, E.; Junge, K.; Beller, M. Chem. Soc. Rev. 2011,
40, 3744−3763. See also ref 13c.
(19) For a copper-catalyzed enantioselective hydrosilylation of
ketones with 4 as ligands, see: Junge, K.; Wendt, B.; Addis, D.;
Zhou, S.; Das, S.; Beller, M. Chem. - Eur. J. 2010, 16, 68−73. For
various applications including for example hydroboration or C−C
bond formation reactions, see also ref 18c.
(20) The mechanism of the rhodium-catalyzed hydrosilylation with
two phosphine ligands of type 3 on the rhodium center is presented
in: Ojima, I.; Kogure, T.; Kumagai, M.; Horiuchi, S.; Sato, T. J.
Organomet. Chem. 1976, 122, 83−97. For a detailed discussion and
various examples of hydrogenations, see ref 13a. The use of a 2:1
ligand:metal ratio for the in situ generation of metal catalysts has also
been established for 4; see for example ref 19.
(34) For an in-depth investigation of the borane-catalyzed hydro-
silylation of imines, see: Hermeke, J.; Mewald, M.; Oestreich, M. J. Am.
Chem. Soc. 2013, 135, 17537−17546.
(35) The dehydrogenative N−Si coupling with [1]+[BArF ]− (L =
4
Et P) has been documented: Konigs, C. D. F.; Muller, M. F.;
̈
̈
3
Aiguabella, N.; Klare, H. F. T.; Oestreich, M. Chem. Commun. 2013,
49, 1506−1508.
(21) Uozumi, Y.; Tanahashi, A.; Lee, S.-Y.; Hayashi, T. J. Org. Chem.
1993, 58, 1945−1949.
(36) This transformation was shown to be a possible pathway in the
stepwise hydrogenation of N,N-dialkyl iminium ions with ruthenium
hydrides: Guan, H.; Iimura, M.; Magee, M. P.; Norton, J. R.; Zhu, G. J.
Am. Chem. Soc. 2005, 127, 7805−7814.
(22) Hayashi describes such a trend for a palladium-catalyzed
hydrosilylation of double bonds: (a) Uozumi, Y.; Hayashi, T. J. Am.
Chem. Soc. 1991, 113, 9887−9888. An allyl−monophosphine−
palladium complex was later isolated during their studies of the
reduction of allylic esters with formic acid: (b) Hayashi, T.; Iwamura,
H.; Naito, M.; Matsumoto, Y.; Uozumi, Y.; Miki, M.; Yanagi, K. J. Am.
Chem. Soc. 1994, 116, 775−776. For more examples employing 5 in
asymmetric catalysis, see ref 13b.
(37) DFT calculations also support that the formation of II is more
favorable than the formation of VI. The relative free energies for II2
and VI30 differ significantly, with II being the more stable
intermediate.
(38) The reduction of silyl enol ethers with [1]+[BArF ]− (L = Et3P)
4
readily occurs with 1 atm of H2.12
(23) For a recent review of metal−ligand cooperation, see:
Khusnutdinova, J. R.; Milstein, D. Angew. Chem., Int. Ed. 2015, 54,
12236−12273.
(39) Prepared by slightly modified procedures starting from (S)-
binol: (a) Sengupta, S.; Leite, M.; Raslan, D. S.; Quesnelle, C.;
Snieckus, V. J. Org. Chem. 1992, 57, 4066−4068. (b) Ooi, T.; Kameda,
M.; Maruoka, K. J. Am. Chem. Soc. 2003, 125, 5139−5151.
(40) The BH3-protected precursor for (S)-6 was synthesized
(24) (a) Juge,
(b) Juge, S.; Stephan, M.; Laffitte, J. A.; Genet
1990, 31, 6357−6360.
́
S.; Genet
̂
, J.-P. Tetrahedron Lett. 1989, 30, 2783−2786.
́
̂
, J.-P. Tetrahedron Lett.
according to a procedure from Genet and Juge.24
́
Decomplexation
was performed with DABCO as described for similar phosphines:
Bayardon, J.; Bernard, J.; Remond, E.; Rousselin, Y.; Malacea-Kabbara,
R.; Juge, S. Org. Lett. 2015, 17, 1216−1219.
(25) The phosphorus atom in (S)-7 is not a chiral center, as it lies in
the C2 axis of the binaphthyl backbone. The generation of two
diastereomers is explained by the formation of another chiral center at
the ruthenium atom.
́
́
(41) Ellison, J. J.; Ruhlandt-Senge, K.; Power, P. P. Angew. Chem., Int.
Ed. Engl. 1994, 33, 1178−1180.
(26) As we did not notice a difference in the performance of in situ
generated and isolated catalyst [(S)-13]+[BArF ]−, we did not expect
4
an enhancement of the enantiomeric excess for [1]+[BArF ]− with
(42) Prepared according to a modified procedure from: Sun, Y.;
Machala, M. L.; Castellano, F. N. Inorg. Chim. Acta 2010, 363, 283−
287.
4
phosphine (S)-6 as ligand. Its cationic congener was hence not
isolated.
(27) For [1]+[BArF ]− with L = (4-FC6H4)3P, the intermediate
(43) Prepared according to a modified procedure from: Brookhart,
M.; Grant, B.; Volpe, A. F., Jr. Organometallics 1992, 11, 3920−3922.
(44) Imamoto, T.; Iwadate, N.; Yoshida, K. Org. Lett. 2006, 8, 2289−
2292.
4
[Ru(Sdmp)L2]+ was observed during the chloride abstraction.12 If
such a complex is obtained in the chloride abstraction from (S,RuRS)-
10, it would unlikely be an active catalyst, as this would affect the
enantiomeric excess in comparison to the use of isolated [(S)-
(45) Mewald, M. Ph.D. Thesis, Westfalische Wilhelms-Universitat
̈
̈
13]+[BArF ]−. We hence conclude that a complex with two phosphine
Munster, Munster, Germany, 2012.
̈
̈
4
(46) Agilent CrysAlis PRO; Agilent Technologies, Yarnton, U.K., 2012
(47) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.
1990, 46, 467−473.
ligands at the ruthenium center either is not formed or is not
catalytically active.
(28) Products of such a transformation were previously obtained in
case of large protective groups at the nitrogen atom in substantial
amounts.10b
(29) 19 was introduced for transfer hydrosilylations by our
laboratory; see: (a) Simonneau, A.; Oestreich, M. Angew. Chem., Int.
Ed. 2013, 52, 11905−11907. (b) Keess, S.; Simonneau, A.; Oestreich,
M. Organometallics 2015, 34, 790−799. For a recent summary of
transfer hydrosilylation, see: (c) Oestreich, M. Angew. Chem., Int. Ed.
2016, 55, 494−499.
(48) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.
2008, 64, 112−122.
(49) Cambridge Crystallographic Data Centre: http://www.ccdc.
cam.ac.uk/Solutions/CSDSystem/Pages/mercury.aspx.
I
DOI: 10.1021/acs.organomet.7b00030
Organometallics XXXX, XXX, XXX−XXX