Benzimidazoles as ligands in the ruthenium-catalyzed enantioselective bifunctional hydrogenation of ketones

Article abstract of DOI:10.1021/om900897b

A series of Cl₂Ru(diphosphane)L₂ (II) complexes in which L = N1-alkylated benzimidazoles, bonding to the metal through nitrogen, have been synthesized and characterized. In the case of 1-methylbenzimidazole, the resulting complexes exist as statistical mixtures of all possible conformational isomers. When the size of the substituent on the benzimidazole was increased to complexes could be prepared that exist as a single diastereomer. All complexes possessing benzimidazole ligands bound to the ruthenium center are active for the mild and chemoselective hydrogenation of ketones in the presence of alkenes. Catalysts that exist as a single diastereomer, prepared with enantiomerically pure diphosphanes, catalyze the hydrogenation of prochiral ketones with moderate levels of enantioselectivity that are significantly improved relative to catalysts existing in several conformations.

Full text of DOI:10.1021/om900897b

554 Organometallics 2010, 29, 554–561
DOI: 10.1021/om900897b
Benzimidazoles as Ligands in the Ruthenium-Catalyzed Enantioselective
Bifunctional Hydrogenation of Ketones
Jeremy M. Praetorius, Ruiyao Wang, and Cathleen M. Crudden*
Department of Chemistry, Queen’s University, 90 Bader Lane, Kingston, Ontario, Canada K7L 3N6
Received October 13, 2009
A series of Cl2Ru(diphosphane)L2 (II) complexes in which L = N1-alkylated benzimidazoles,
bonding to the metal through nitrogen, have been synthesized and characterized. In the case of
1-methylbenzimidazole, the resulting complexes exist as statistical mixtures of all possible conforma-
tional isomers. When the size of the substituent on the benzimidazole was increased to complexes
could be prepared that exist as a single diastereomer. All complexes possessing benzimidazole ligands
bound to the ruthenium center are active for the mild and chemoselective hydrogenation of ketones
in the presence of alkenes. Catalysts that exist as a single diastereomer, prepared with enantio-
merically pure diphosphanes, catalyze the hydrogenation of prochiral ketones with moderate
levels of enantioselectivity that are significantly improved relative to catalysts existing in several
conformations.
Introduction
diphosphane (e.g., Cl2Ru((R)-Binap); Binap=2,20-bis(diphenyl-
phosphino)-1,10-binapthyl) produces an exceptionally enan-
tioselective precatalyst, with high turnover numbers and
frequencies8-12 (Scheme 1). Noyori’s research group,13
among others,14-17 has investigated the mechanism of this
reaction, with the aim of explaining its remarkable reactivity
and selectivity.
It is clear from numerous studies that the presence of a
diamine ligand bearing at least one N-H bond is critical for
high activity and chemoselectivity;6,18 however, the precise
explanation of this effect has been the subject of recent
debate. The most commonly accepted mechanism has the
reaction occurring through a six-membered, concerted tran-
sition state, in which the metal is coordinatively saturated,
preventing coordination of substrates directly to the metal
and instead promoting their interaction with the coordinat-
ing ligands.13,15,19 The protic amine is believed to act as a
Brønsted acid, activating the carbonyl compound to reduc-
tion by the metal hydride, which adds to the carbonyl carbon
concurrently with the proton of the diamine ligand, simulta-
neously generating a Ru-N double bond. This rationale also
The hydrogenation of carbonyl compounds is an impor-
tant reaction on both laboratory1 and industrial2 scales.
Considering the complexity and functional group density
of common synthetic targets, efficient and chemoselective
methods to affect this transformation are necessary.3
Furthermore, the development of enantioselective methods
to affect this transformation is highly desirable.4
The greatest breakthrough in this field has come from the
Noyori laboratories,5 where it was discovered that the addi-
tion of a protic diamine to ruthenium diphosphane species
results in catalysts displaying extraordinary chemoselectivity
for carbonyl groups over olefins under relatively mild hydro-
genation conditions.6,7 Furthermore, proper matching of a
chiral diamine, for example (R,R)-diphenylethylenediamine
(dpen), with the correct enantiomer of a chiral ruthenium
*To whom correspondence should be addressed. E-mail: cathleen.
crudden@chem.queensu.ca.
(1) Andersson, P. G.; Munslow, I. J. Modern Reduction Methods;
Wiley: New York, 2008.
(2) Akutagawa, S. In Chirality in Industry: The Commercial Manu-
facture and Applications of Optically Active Compounds; Collins, A. N.,
Sheldrake, G. N., Crosby, J., Eds.; Wiley: New York, 1997.
(3) Ohkuma, T.; Noyori, R. In Comprehensive Asymmetric Catalysis.
Supplement; Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: New
York, 2004.
(11) Ohkuma, T.; Koizumi, M.; Ikehira, H.; Yokozawa, T.; Noyori,
R. Org. Lett. 2000, 2, 659–662.
(12) Doucet, H.; Ohkuma, T.; Murata, K.; Yokozawa, T.; Kozawa,
M.; Katayama, E.; England, A. F.; Ikariya, T.; Noyori, R. Angew.
Chem., Int. Ed. 1998, 37, 1703–1707.
(13) Sandoval, C. A.; Ohkuma, T.; Muniz, K.; Noyori, R. J. Am.
Chem. Soc. 2003, 125, 13490–13503.
(4) Noyori, R. Asymmetric Catalysis in Organic Synthesis; Wiley: New
York, 1994.
(5) Noyori, R.; Ohkuma, T. Angew. Chem., Int. Ed. 2001, 40, 40–73.
(6) Ohkuma, T.; Ooka, H.; Ikariya, T.; Noyori, R. J. Am. Chem. Soc.
1995, 117, 10417–10418.
(7) Ohkuma, T.; Ikehira, H.; Ikariya, T.; Noyori, R. Synlett 1997,
1997, 467–468.
(8) Ohkuma, T.; Ooka, H.; Hashiguchi, S.; Ikariya, T.; Noyori, R. J.
Am. Chem. Soc. 1995, 117, 2675–2676.
(9) Ohkuma, T.; Ooka, H.; Yamakawa, M.; Ikariya, T.; Noyori, R. J.
Org. Chem. 1996, 61, 4872–4873.
(14) Hartmann, R.; Chen, P. Angew. Chem., Int. Ed. 2001, 40, 3581–
3585.
(15) Abdur-Rashid, K.; Faatz, M.; Lough, A. J.; Morris, R. H. J. Am.
Chem. Soc. 2001, 123, 7473–7474.
(16) Abdur-Rashid, K.; Clapham, S. E.; Hadzovic, A.; Harvey, J. N.;
Lough, A. J.; Morris, R. H. J. Am. Chem. Soc. 2002, 124, 15104–15118.
(17) Zimmer-, M.; Morris, R. H. J. Am. Chem. Soc. 2009, 131, 11263–
11269.
(18) Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. Rev.
(10) Ohkuma, T.; Koizumi, M.; Doucet, H.; Pham, T.; Kozawa, M.;
Murata, K.; Katayama, E.; Yokozawa, T.; Ikariya, T.; Noyori, R. J.
Am. Chem. Soc. 1998, 120, 13529–13530.
2004, 248, 2201–2237.
(19) Sandoval, C. A.; Yamaguchi, Y.; Ohkuma, T.; Kato, K.; Noyori,
R. Magn. Reson. Chem. 2006, 44, 66–75.
r
pubs.acs.org/Organometallics
Published on Web 01/07/2010
2010 American Chemical Society

Article
Organometallics, Vol. 29, No. 3, 2010 555
Scheme 1. Representative Hydrogenation of a Simple Aromatic Ketone under Noyori Conditions
explains the chemoselectivity of this system, since the less
basic, electronically symmetrical π system of the alkene
interacts poorly with the weakly acidic bound amine. The
orthogonal electrostatics of the diamine and hydride ligands
and the unique reactivity they impart to the system lead to
the use of the phrase “bifunctional catalysis” to describe this
process.5,20,21
Stoichiometric studies by Bergens on this system have
shown that while the protic diamine may act as a binding
point for the nucleophilic oxygen atom, activating the
π-bond toward attack of the hydride, its addition to the
carbonyl does not necessarily occur concurrently with re-
formation of the Ru-N double bond.22-24 As evidence for
this alternative mechanism, they were able to isolate a
ruthenium alkoxide resulting from a 1,2-addition of the
carbonyl. This suggests that, after reduction of the bound
carbonyl compound, the amine ligand is deprotonated by the
alkoxide base present in solution, generating a lone pair on
nitrogen that ejects the alkoxide and forms the ruthe-
nium-amide complex proposed by Noyori to be the catalyst
resting state.
Interested by these observations, we endeavored to dis-
cover whether alternative Lewis acidic functional groups
might be capable of promoting the hydrogenation of carbo-
nyl compounds. Bergens has previously demonstrated that
Ru bisphosphane catalysts modified by pyridyl substituents
instead of protic diamines are capable of affecting the
hydrogenation of ketones with enantioselectivities up to
49% ee.25 It is well-known from the preparation of N-
heterocyclic carbene chemistry that the C2 position of N,
N0-disubstituted imidazolium26a and benzimidazolium26b
salts are sufficiently acidic to be deprotonated by relatively
mild alkoxide bases (Figure 1). Furthermore, in studies of
Figure 1. Relative acidities of imidazolium and benzimidazo-
lium cations in DMSO and representative polarization of the
C2-H bond by coordination of benzimidazole to a metal.
ionic liquids, R values of Kamlet-Taft parameters for
imidazoliums have been determined and these solvents are
about as protic as tert-butyl alcohol (0.62727 versus 0.68,28
respectively). A recent example in the literature has also
shown that imidazolium-based ionic liquids are capable of
acting as protic catalysts in the BOC protection of phenols.29
In analogy with benzimidazolium cations, coordination of
the basic nitrogen of 1-substituted benzimidazoles results in
polarization of the C2-H bond in a manner similar to
alkylation at this position (Figure 1). Coordinative satura-
tion of the metal is important (vide supra); therefore, in order
to generate a chemoselective hydrogenation catalyst, addition
of 2 equivalents of the benzimidazole to a Ru(diphosphane)
precursor would be required. Herein we report the synthesis
of a number of ruthenium-diphosphane precatalysts disub-
stituted with benzimidazole ligands and their application in
the chemoselective hydrogenation of aromatic ketones under
relatively mild conditions.
Results and Discussion
In order to generate the desired benzimidazole-substituted
Ru complex, Cl2Ru(rac-Binap) was reacted overnight with 2
equivalents of methylbenzimidazole (BIMe) to give Cl2Ru(rac-
Binap)(BIMe)2 (1) in 46% yield after recrystallization. 31P
NMR of a pure sample of the orange solid in C6D6 showed
signals attributable to a mixture of three different diastere-
omeric conformers: two singlets (one sharp and one broad)
and a pair of doublets. The two conformers represented in
the 31P NMR by singlets, stem from two C2-symmetric
systems in which the phosphorus atoms of the Binap ligand
are magnetically equivalent. The other conformer, which
(20) Haack, K.-J.; Hashiguchi, S.; Fujii, A.; Ikariya, T.; Noyori, R.
Angew. Chem., Int. Ed. 1997, 36, 285–288.
(21) Noyori, R.; Yamakawa, M.; Hashiguchi, S. J. Org. Chem. 2001,
66, 7931–7944.
(22) Hamilton, R. J.; Leong, C. G.; Bigam, G.; Miskolzie, M.;
Bergens, S. H. J. Am. Chem. Soc. 2005, 127, 4152–4153.
(23) Hamilton, R. J.; Bergens, S. H. J. Am. Chem. Soc. 2006, 128,
13700–13701.
(24) Hamilton, R. J.; Bergens, S. H. J. Am. Chem. Soc. 2008, 130,
11979–11987.
(25) (a) Akotshi, O. M.; Metera, K.; Reid, R. D.; McDonald, R.;
Bergens, S. H. Chirality 2000, 12, 514–522(48% ee for hydrogenation of
acetophenone). (b) Leong, C. G.; Akotshi, O. M.; Ferguson, M. J.; Bergens,
S. H. Chem. Commun. 2003, 750–751(enantioselectivities in corrected
version of manuscript at 14-49% ee).
(26) (a) Alder, R. W.; Allen, P. R.; Williams, S. J. J. Chem. Soc.,
Chem. Commun. 1995, 1267–1268. (b) Bordwell, F. G. Acc. Chem. Res.
2002, 21, 456–463.
(27) Crowhurst, L.; Mawdsley, P. R.; Perez-Arlandis, J. M.; Salter, P.
A.; Welton, T. Phys. Chem. Chem. Phys. 2003, 5, 2790–2794.
(28) Kamlet, M. J.; Abboud, J. L. M.; Abraham, M. H.; Taft, R. W. J.
Org. Chem. 2002, 48, 2877–2887.
(29) Chakraborti, A. K.; Roy, S. R. J. Am. Chem. Soc. 2009, 131,
6902–6903.

556 Organometallics, Vol. 29, No. 3, 2010
Praetorius et al.
Figure 2. Possible conformations of the benzimidazole ligands bound to Cl2Ru(diphosphane).
Table 1. Catalytic Results in the Hydrogenation of Acetophenone with Cl2Ru((rac)-Binap)(BIMe)2 (1)
entry
amt of Ru (mol %)
amt of KOtBu (mol %)
time (h)
amt of H2 (atm)
yield (%)a
1
2
3
4
1
10
1
1
3
3
3
16
10
5
1
>98
50
14
0.1
0.1
0.1
1
0
9
a Yields determined by 1H NMR with an internal standard (1,4-dimethoxybenzene).
gives a pair of doublets in the 31P NMR, indicates that the
two phosphorus atoms of the Binap ligand are inequivalent
in this conformer. NMR evidence also indicates that two
benzimidazoles are bound to a single ruthenium center,
resulting in coordinatively saturated and octahedral com-
plexes. Interestingly, addition of 1 equivalent of methyl-
benzimidazole per mole of ruthenium results only in the
formation of complex 1 and Cl2Ru(rac-Binap); no other
species could be detected by 31P NMR, illustrating the strong
preference in this system for coordinative saturation.
Once bound to Ru, the benzimidazole ligands can exist in
two possible conformations, syn and anti, relative to each
other with the N1 methyl substituents on the same or
opposite sides of the plane defined by P-Ru-P, respectively.
In addition to this, the presence of the C2-symmetric Binap
ligand further differentiates these planes into two inequiva-
lent quadrants, allowing for two possible anti conforma-
tions, both of which have overall C2 symmetry (Figure 2b,c).
On the basis of the 31P NMR spectra, 1 appears to exist as a
mixture of all three of these conformations. Analysis of a
related compound by X-ray crystallography confirms this
hypothesis (vide infra).
benzimidazole. To further test the ability of 1 as a hydro-
genation catalyst, its loading was decreased to 0.1% and the
pressure of hydrogen gas in the reactions reduced. The
turnover frequency was greatly reduced upon lowering the
pressure ofhydrogen from5 to1 atm (Table 1, entries 2 and3),
and very little conversion was observed when the reaction was
performed under an atmosphere of argon (<10% after 16 h).
The dependence of the reaction rate on hydrogen pressure
suggests that the primary mechanism of ketone reduction is
hydrogenation and not transfer hydrogenation as might
be expected from use of a solvent such as 2-propanol. In
the latter case, hydrogen gas would not be required, and
the metal hydride would be regenerated by oxidation of the
solvent to acetone.
On the basis of our initial hypothesis, after formation of a
ruthenium hydride, coordination of the carbonyl oxygen to
the C2-H of one of the benzimidazoles may occur, orienting
and activating the carbonyl toward insertion, forming a
ruthenium alkoxide. Unlike Noyori’s catalysts, however,
deprotonation of the assisting ligand (in our case the benzi-
midazole, in Noyori’s the diamine) likely does not occur, and
the protic solvent likely facilitates release of the product
molecule. It is also possible that an alternative mechanism
for hydrogenation is responsible for the observed reactivity,
since there have been some reports of hydrogenations with
non-N-protic complexes.25
In order to test the catalytic activity of this benzimidazole-
containing Ru complex, acetophenone was exposed to cat-
alyst 1 in basic 2-propanol at room temperature. The reduc-
tion of acetophenone was complete in 3 h under mild
conditions (100:10:1 substrate/KOtBu/Ru, 0.5 M in 2-pro-
panol, 293 K, 10 atm of H2). This is in stark contrast to
Cl2Ru(Binap) itself, which is essentially unreactive under
identical conditions, without the addition of a diamine or
Although complex 1 obviously displayed the desired cat-
alytic activity, the presence of multiple conformations did
not bode well for eventual enantiodiscrimination. In fact, C2-
symmetric metal-BINAP complexes are known to react

Article
Organometallics, Vol. 29, No. 3, 2010 557
with high enantioselectivity in a variety of processes, since
the number of possible diastereomeric transition states is
reduced by a factor of 2. In the case of Binap-derived and
related ligands, the remaining binding sites on the metal are
differentiated into pairs of sterically hindered and open
quadrants by the edge-on and face-on phenyl groups of the
chelating diphosphane.30 Thus, we hypothesized that if a
large substituent were employed on the benzimidazole, steric
interactions between this group and the phenyl substituents
of BINAP might enforce only one binding mode of the
benzimidazole, resulting in a compound with overall C2
symmetry (Figure 2c). If steric congestion in the quadrant
is not great enough, the two substituents can face the same
way (Figure 2, syn conformation a) or even adopt an anti
conformation with the substituents of both benzimidazoles
in unfavorable quadrants (Figure 2b). As mentioned above,
this is believed to be the case for 1-methylbenzimidazole-
substituted compound 1, whose 31P NMR spectrum is con-
sistent with a near-equal mixture of all possible conforma-
tions. On the basis of this hypothesis, we prepared complexes
with larger substituents at this position.
Thus, Cl2Ru((R)-TolBinap) (TolBinap=2,20-bis(di-4-tolyl-
phosphino)-1,10-binaphthyl) was reacted with a number of
different benzimidazoles substituted at N1 with alkyl groups
of increasing size (BIR, where R=Me, Et, Ph, Bz, CPh3) and
the resulting solids were analyzed by 31P NMR in C6D6.
Compound 2a (R=Me) yielded a 31P NMR spectrum nearly
identical with that of its racemic counterpart 1, again con-
sisting of a mixture of diastereomers (vide supra). Increasing
the size of the substituent at N1 on the benzimidazole had
little effect on the distribution of the mixture of diastereo-
mers for compounds BIEt, BIPh, and BIBz as observed by
31P NMR; however, 2b (R = CPh3) exhibits only a single
resonance in the 31P NMR with no evidence of any other
diastereomer. This indicates that the very large R substituent
at N1 has induced only the favorable C2-symmetric conforma-
tion in which the R substituents on the two cis benzimidazoles
are oriented anti to one another. The 1H NMR spectra are also
consistent with the existence of a single species in solution. A
similar synthesis was performed with Cl2Ru((R)-XylBinap)
(XylBinap=2,20-bis(3,5-xylylphosphino)-1,10-binaphthyl) to
prepare Cl2Ru((R)-Xylbinap)(BIMe)2 (3a) and Cl2Ru((R)-
XylBinap)(BICPh3)2 (3b) in 23 and 28% yields, respectively,
after recrystallization. Apart from the additional methyl
signals attributable to the xylyl groups, the spectra were
analogous to those of the TolBinap-based complexes. The
31P NMR spectra of Cl2Ru(XylBinap) after reaction with
2 equiv of BIMe (3a) and BICPh3 (3b) are shown in Figure 3.
X-ray-quality crystals of 3a were obtained by slow diffu-
sion of hexanes into a dichloromethane solution of 3a, which
further confirmed our hypothesis of its molecular structure.
The solid-state structure is a distorted octahedron about Ru,
with the two benzimidazoles positioned cis to one another
and the chlorines trans. Consistent with our interpretation of
the NMR spectra of this and related compounds, the crystal
is a statistical mixture of three conformational diastereo-
mers, one syn and two anti. The two anti conformations are
nonidentical C2-symmetric structures. The benzimidazoles
are canted slightly and appear nearly parallel to the axial
xylyl rings of the XylBinap ligand, possibly as a result of π
stacking between the two aromatic groups. Two of the
Figure 3. Relevant areas of the 31P{1H} spectra of compounds
3a and 3b.
possible conformers are displayed in Figure 4, and the
crystallographic data are presented in Table 2.
The catalytic activity of these new complexes, synthesized
from chiral enantiomerically pure diphosphane ligands, was
then tested in the hydrogenation of acetophenone under our
standard conditions (1000:10:1 ketone/KOtBu/Ru, 0.5 M in
2-propanol, 5 bar of H2, 293 K) (Table 3). The methylbenzi-
midazole-modified catalyst (2a) gave the S enantiomer of
the product alcohol in 27% ee (Table 3, entry 1). Increasing
the size of the group on the benzimidazole from methyl to
ethyl and eventually phenyl and benzyl resulted in only
modest increases in the observed enantioselectivity. As
expected, however, a large increase in enantioselectivity
was observed for the catalyst substituted with 2 equivalents
of N-(triphenylmethyl)benzimidazole, which gave sec-phe-
nethanol in 67% ee. Thus, it appears that the presence of only
one diastereomeric form in solution does lead to a more
enantioselective catalyst. By decreasing the temperature of
the reaction to 273 K and increasing the pressure of hydrogen
gas pressure to 20 bar, a slight increase to 71% ee was
achieved (Table 3, entry 6). The use of the slightly bulkier
(R)-XylBinap as the chiral diphosphane gave marginal in-
creases in enantioselectivity relative to the TolBinap cata-
lysts.
Interestingly, the hydrogenation of acetophenone with 2b
does proceed under a nitrogen atmosphere (i.e., in the
absence of hydrogen gas), however, at a significantly dimin-
ished rate. If the reaction is carried out under 5 atm H2, it
reaches completion in 6 h; however, under a nitrogen atmo-
sphere, only 7% yield is achieved after 6 h and only 60% after
72 h. The presence of base in the catalytic reaction is also
essential; in its absence no conversion is observed.
Using catalyst 2b, we then examined the effect of solvent
on the reaction (Table 4). Neither CH2Cl2 nor THF gave any
(30) Ohta, T.; Takaya, H.; Noyori, R. Inorg. Chem. 1988, 27, 566–
569.

558 Organometallics, Vol. 29, No. 3, 2010
Praetorius et al.
Figure 4. Crystallographically determined structure of 3a with selected atoms displayed as ellipsoids at the 50% confidence level.
Hydrogen atoms have been omitted for clarity. Two of the three different diastereomeric conformations present in the unit cell are
˚
shown. Selected interatomic distances (A) and angles (deg): Ru(1)-N(1A), 2.187(6); Ru(1)-N(1B), 2.20(2); Ru(1)-P(1), 2.3042(11);
Ru(1)-Cl(1), 2.4206(12); N(1A)-Ru(1)-N(1A)#1, 81.9(3); N(1B)-Ru(1)-N(1B)#1, 104.3(10); N(1B)-Ru(1)-N(1A)#1, 91.5(5);
N(1A)#1-Ru(1)-P(1), 94.28(14); N(1B)#1-Ru(1)-P(1), 84.8(5); N(1A)-Ru(1)-P(1)#1, 94.28(14); P(1)-Ru(1)-P(1)#1, 89.52(5);
Cl(1)-Ru(1)-Cl(1)#1, 166.76(6).
different chiral diphosphanes on the catalyst precursor. In
the recent literature, a number of atropisomeric chiral dipho-
sphanes have been reported which have shown exceptional
results when employed as ligands in asymmetric hydro-
genations.31-35 We chose a number of these diphosphanes
(Chart 1), and, using a similar procedure for 2b, prepared the
analogous ruthenium precatalysts (4-8) in moderate to
good yields after recrystallization from THF/hexanes solu-
tions (Scheme 2). While all of these new ruthenium species
were able to efficiently hydrogenate acetophenone under our
standard conditions, giving quantitative conversion in less
than 6 h, none of them gave better enantioselectivity than the
BINAP-based catalysts 2b and 3b (Table 5). Interestingly,
catalyst 8, prepared using (R)-XylPhanePhos, gives a nearly
racemic product mixture, with a slight excess of the opposite
enantiomer (Table 5, entry 7); this is consistent with previous
reports in which (R)-PhanePhos matches with (S,S)-dpen
(the opposite match to (R)-Binap) to give the R enantio-
mer.35
With 2-propanol as the solvent, and 3b as our best catalyst,
the hydrogenation of a number of other aromatic ketones
was examined (Table 6). Under these conditions, benzophe-
none was readily reduced (Table 6, entry 1), negating the
possibility of the catalyst hydrogenating an enolate inter-
mediate. While aryl methyl ketones with mildly electron
withdrawing groups p-bromo and p-fluoro or electron-do-
nating p-methoxy groups were all hydrogenated at faster
rates than was acetophenone, p-nitroacetophenone (Table 6,
entry 2) was completely unreactive under our standard
conditions. Of the compounds that were reactive, the highest
Table 2. Crystallographic Information for Compound 3a
formula
fw
T (K)
C68H64Cl2N4P2Ru•C4H8O
1243.25
180(2)
color/habit
cryst dimens (mm3)
cryst syst
space group
orange/prism shaped
0.25 ꢀ 0.16 ꢀ 0.08
orthorhombic
C222(1)
˚
a (A)
12.5121(16)
22.639(3)
21.746(3)
˚
b (A)
˚
c (A)
β (deg)
90
3
V (A )
˚
6162.6(13)
4
1.34
0.441
2592
Z
Dcalcd (Mg/m3)
μ (mm-1
F(000)
)
θ range (deg)
1.80 - 25.99
(15, ( 27, ( 26
30803
6045/0.0783
6045/12/436
0.0446/0.0898
0.0671/0.1000
1.022
0.653/-0.692
P
index ranges (h,k,l)
no. of rflns collected
no. of indep rflns/Rint
no. of data/restraints/params
R1/wR2 (I > 2σ(I))a
R1/wR2 (all data)a
GOF (on F2)a
-3
˚
largest diff peak/hole (e A
P
)
P
P
a R1= ||Fo| - |Fc||/ |Fo|; wR2={ [w(Fo2 - Fc2)2]/ [w(Fo2)2]}1/2
;
P
GOF={ [w(Fo2 - Fc2)2]/(n - p)}1/2
.
reaction, but aromatic solvents such as benzene and toluene
were effective, giving comparable enantioselectivities. Inter-
estingly, the use of tert-butyl alcohol as the solvent gave a
much less reactive system than 2-propanol under the same
conditions (entries 6 and 7, Table 4, the reaction was
performed at 303 K instead of 293 K; mp(tert-butyl alcohol)
298 K). Given the slow rate of background transfer hydro-
genation operative in this system, this is quite surprising but
could be attributable to the larger size of size of tert-butyl
alcohol making approach into the coordination sphere of the
metal to protonate and release a possible ruthenium-alk-
oxide intermediate more difficult.
(31) Duprat de Paule, S.; Jeulin, S.; Ratovelomanana-Vidal, V.;
^
Genet, J.-P.; Champion, N.; Dellis, P. Tetrahedron Lett. 2003, 44,
823–826.
(32) Zhang, Z.; Qian, H.; Longmire, J.; Zhang, X. J. Org. Chem. 2000,
65, 6223–6226.
(33) Maligres, P. E.; Krska, S. W.; Humphrey, G. R. Org. Lett. 2004,
6, 3147–3150.
(34) Jeulin, S.; Duprat de Paule, S.; Ratovelomanana-Vidal, V.;
^
Genet, J.-P.; Champion, N.; Dellis, P. Angew. Chem., Int. Ed. 2004,
Having chosen 2-propanol as our optimal solvent and
encouraged by the moderate enantioselectivities achieved
with this catalyst system, we next examined the effect of
43, 320–325.
(35) Burk, M. J.; Hems, W.; Herzberg, D.; Malan, C.; Zanotti-
Gerosa, A. Org. Lett. 2000, 2, 4173–4176.

Article
Organometallics, Vol. 29, No. 3, 2010 559
Table 3. Hydrogenation of Acetophenone with Cl2Ru(diphosphane) Precatalysts Bis-Substituted with Different Benzimidazolesa
entry
catalyst
amt of H2 (atm)
temp (K)
TOF (h-1 b
)
ee (%)c
1
2
3
4
5
6
Ru((R)-TolBinap)Cl2
5
5
5
5
5
20
293
293
293
293
293
273
NRd
320
162
293
210
67
NDe
27
67
34
72
71
2a
2b
3a
3b
2a
a Reaction conditions: 1000:10:1 ketone/KOtBu/Ru, 0.5 M in 2-propanol, H2 gas. b Turnover frequencies determined at low conversion (tR = 1.5 h)
by 1H NMR. c % ee was determined using supercritical fluid chromatography. d No reaction was observed after 1.5 h. e % ee not determined.
Table 4. Optimization of the Solvent in Hydrogenation of Acet-
ophenone by Catalyst 2ba
Scheme 2. Synthesis of 1-(Triphenylmethyl)benzimidazole-
Modified Ruthenium-Diphosphane Precatalysts
entry
solvent
TOF (h-1 b
)
ee (%)
1
2
3
4
2-propanol
THF
CH2Cl2
benzene
toluene
162
NRc
NRc
40
80
88
67
ND
ND
70d
58
31
30
5
6e
7e
tert-butyl alcohol
2-propanol
Table 5. Turnover Frequencies and Enantioselectivities of
1-(Triphenylmethyl)benzimidazole-Modified Ruthe-
nium-Diphosphane Precatalystsa
282
a Reaction conditions: 1000:10:1 ketone/KOtBu/Ru, [substrate] =
0.5 M, 5 bar of H2, 293 K. b Turnover frequencies determined at low
conversion (tR = 1.5 h) by 1H NMR. c % ee was determined using
supercritical fluid chromatography. d Catalyst 3b used instead of 2b.
e T = 303 K.
Chart 1. Chiral Atropisomeric Diphosphane Ligands Used in
This Study31-35
entry
diphosphane
TOF (h-1 b
)
ee (%)c
1
2
3
4
5
6
7
(R)-TolBinap (2b)
(R)-XylBinap (3b)
(R)-SynPhos (4)
162
210
302
337
261
388
306
67
72
57
59
4
65
-8
(R)-C3-TunaPhos (5)
(R)-Cl-OMe-BIPHEP (6)
(R)-FluorPhos (7)
(R)-PhanePhos (8)
a Reaction conditions: 1000:10:1 ketone/KOtBu/Ru, 0.5
M in
2-propanol, 5 atm of H2, 293 K. b Turnover frequencies determined at
low conversion (tR = 1.5 h) by 1H NMR. c % ee was determined by using
supercritical fluid chromatography.
conditions. While not nearly as selective as Noyori’s dia-
mine-modified catalyst, the observed preference for carbonyl
hydrogenation is consistent with the catalyst acting similarly
to those prepared with NH diamines. The decrease in
selectivity relative to the Noyori catalyst may be related
directly to the decreased rate of turnover of our catalyst
relative to Noyori’s, making the alkene hydrogenation more
competitive. Alternatively, dissociation of one of the mono-
dentate benzimidazole ligands to give an open coordination
site capable of binding the alkene may also occur.
enantioselectivity was observed for the methylene dioxy
substrate shown in entry 6, at 70% ee; all others gave
enantioselectivities in the 50% range.
Finally, we examined the chemoselectivity of catalyst 3b
under our optimized conditions for the hydrogenation of
acetophenone versus the isoelectronic alkene R-methylstyr-
ene (Scheme 3). At complete conversion of the ketone to
its corresponding alcohol, only a 6% yield of the correspond-
ing styrene was detected by GC, indicating a greater than
15:1 selectivity for the carbonyl functionality under these
Conclusions
A number of ruthenium-diphosphane bis-benzimidazole
complexes have been prepared and characterized. Through
crystallographic and NMR analysis, it has been demonstrated

560 Organometallics, Vol. 29, No. 3, 2010
Praetorius et al.
Scheme 3. Intermolecular Competition Experiment between Acetophenone and r-Methylstyrene
added in a single portion. After the mixture was stirred for 30
min at 0 °C, trityl chloride (3.06 g, 11.0 mmol) and a catalytic
amount of tetrabutylammonium iodide were added, the flask
was fitted with a condenser, and the reaction mixture was
refluxed for 24 h. The reaction mixture was then cooled to room
temperature, quenched by the addition of water, and extracted
three times with 100 mL of CHCl3. The combined organic
fractions were dried over MgSO4, the solvent was evaporated,
and the residue was purified by column chromatography (2:1
hexanes/EtOAc) to give 2.53 g of a yellow solid (82% yield). 1H
NMR (CDCl3, 400 MHz): δ 7.82 (s, 1H, NCHN); 7.70 (d, 1H,
J = 8.1 Hz, Ar-CH); 7.26-7.05 (m, 16H, Ar-CH); 6.81 (t, 1H,
J = 8.1 Hz); 6.39 (d, 1H, J = 8.1 Hz) ppm. 13C NMR (CDCl3,
100 MHz): δ 144.6, 144.2, 141.4, 134.8, 130.0, 128.2, 128.0,
122.3, 122.0, 120.3, 115.4, 75.4. HRMS (EI-TOF): calcd for
[M]þ (C26H20N2) m/z 360.1626, found 360.1620.
Table 6. Hydrogenation of Various Aromatic Ketones with
a
Cl2Ru(XylBinap)2(BICPh3)2
entry
R1
Ph
CH3
CH3
CH3
CH3
CH3
R2
TOF (h-1 b
)
ee (%)c
1
2
3
4
H
375
NR
332
286
351
126
N/A
ND
52
55
56
p-NO2
p-F
p-Br
p-OCH3
m,p-OCH2O
5
6d
70
a Reaction conditions: 1000:10:1 ketone/KOtBu/Ru, 0.5
M in
Synthesis of Catalysts (Cl2Ru(diphosphane)(K-N3-(1-R)-
benzimidazole)2. General Procedure. In a nitrogen atmosphere
glovebox, [Ru(C6H6)Cl2]2 (0.050 mmol) and a chiral dipho-
sphane (0.100 mmol) were added to a 50 mL Schlenk flask with
a magnetic stirrer and the solids were dissolved in 5 mL of DMF.
The reaction flask was then put on the Schlenk line and the
solution degassed three times via freeze-pump-thaw cycles
before being heated to 100 °C. After cooling to 35 °C, the
volatiles were removed in vacuo overnight to give a brown,
solid residue. The reaction flask was then brought back into the
glovebox, the residue was dissolved in 10 mL of THF, benzimi-
dazole was added (0.220 mmol), and the reaction mixture was
stirred for 16 h. The reaction mixture was then concentrated to
about 1 mL on the glovebox vacuum pump and the concentrated
mixture filtered through a plug of Celite. This was then reduced
to a minimal amount of THF, layered with hexanes, and
recrystallized at -20 °C and the solid collected by filtration.
Cl2Ru(rac-Binap)(KN3-1-methylbenzimidazole)2 (1). Orange
solid, 46% yield. Anal. Found (calcd) for C60H48Cl2N4-
2-propanol, 5 atm of H2, 293 K. b Turnover frequencies determined at
low conversion (tR = 1.5 h) by 1H NMR. c % ee was determined using by
supercritical fluid chromatography. d Solvent 4:1 2-propanol/toluene.
that the relative conformations of the benzimidazoles within
the complexes are dependent on the N1 substituent present
on the benzimidazole; small substituents such as methyl give
mixtures of all possible conformations, while the very large
triphenylmethyl results in a single C2-symmetric conforma-
tion. Benzimidazole modification of chiral ruthenium dipho-
sphanes resulted in complexes capable of hydrogenating
carbonyls under mild conditions both enantioselectively
and chemoselectively. While these studies are far from
providing conclusive evidence for one mechanism in prefer-
ence to the other, they do demonstrate that Ru complexes
ligated by species other than monoprotic diamines are cap-
able of catalyzing the chemo- and enantioselective hydro-
genation of ketones.
P2Ru H2O: C, 67.11 (66.91); H, 5.09 (4.68); N, 5.14 (5.20). 31
P
3
NMR (C6D6, 161 MHz): δ 43.74 (d, J = 28 Hz), 43.03, 42.21
(br), 41.78 (d, J = 28 Hz). 1H NMR (C6D6, 400 MHz): δ 9.45
(m, 2H, NCHN, three conformers); 8.83-8.17 (m, 12H, Ar-
CH); 7.72-7.54 (m, 2H, Ar-CHBinap, three conformers); 7.45-
7.31 (m, 2H, Ar-CHBI, three conformers); 7.01-6.39 (m, 24H,
Ar-CH); 2.34, 2.30, 2.25 (6H, NCH3, three conformers) ppm.
Cl2Ru((R)-TolBinap)(KN3-1-methylbenzimidazole)2 (2a). Or-
ange solid, 56% yield. Anal. Found (calcd) for C64H56Cl2N4-
P2Ru: C, 68.60 (68.94); H, 5.20 (5.06); N, 5.03 (5.02). 31P NMR
(C6D6, 161 MHz): δ 41.40 (d, J = 35 Hz), 40.81, 40.20, 39.45 (d,
J = 35 Hz). 1H NMR (C6D6, 400 MHz): δ 9.48-9.27 (m, 2H,
NCHN, three conformers); 8.87-8.10 (m, 12H, Ar-CH);
7.74-7.62 (m, 2H, Ar-CHBinap, three conformers); 7.45-7.34
(m, 2H, Ar-CHBI, three conformers); 7.08-6.30 (m, 20H, Ar-
CH); 2.37, 2.32, 2.31 (6H, NCH3, three conformers); 1.77, 1.74,
1.71 (12 H, PC6H4CH3, three conformers).
Experimental Section
General Considerations. [Ru(C6H6)Cl2]2, chiral dipho-
sphanes, methylbenzimidazole, ketone substrates, and potas-
sium tert-butoxide were purchased from commercial sources
and used as received. Isopropyl alcohol was dried and purified
by distillation from magnesium activated with iodine. Benzene,
d6-benzene, and tert-butyl alcohol were dried over CaH2 and
purified by distillation. All other solvents (THF, CH2Cl2,
hexanes, toluene) were purified using an MBraun SPS solvent
system. All solvents were purged of oxygen using a minimum of
three freeze-pump-thaw cycles before being brought into the
glovebox. 1H and 13C{1H} NMR spectra were performed using
Bruker Avance 400 and 500 MHz spectrometers. Elemental
analysis was performed by Canadian Microanalytical Systems
Ltd. X-ray data collection was performed on a Bruker SMART
CCD 1000 X-ray diffractometer. Analytical supercritical fluid
chromatography (SFC) was performed on a Berger SFC HPLC
using the specified Chiracel Berger Silica column and specified
conditions of coeluent, flow rate, and pressure in order to
determine enantiomeric purity.
Cl2Ru((R)-TolBinap)(KN3-1-(triphenylmethyl)benzimidazole)2 (2b).
Yellow solid, 74% yield. Anal. Found (calcd) for C100H80Cl2N4P2Ru:
C, 76.96 (76.42); H, 5.13 (5.56); N, 3.56 (3.62). 31P NMR (C6D6, 161
1
MHz): δ 40.43 ppm. H NMR (C6D6, 400 MHz): δ 9.03 (s, 2H,
NCHN), 8.37 (m, 4H, Ar-CHP-Ar),8.15(m,2H,Ar-CHBinap),8.09(d,
2H, J=8.0Hz,Ar-CHBI),8.01(m,4H,Ar-CHP-Ar), 7.38 (d, 2H, J=
8.7 Hz, Ar-CHBinap), 7.30 (d, 2H, J = 8.0 Hz, Ar-CHBI), 7.01-6.87
(m,30H,Ar-CHTr),6.67(d,4H,J=8.0Hz, Ar-CHP-Ar),6.61(t, 2H,
J = 8.0, Ar-CHBI), 6.48 (m, 6H, Ar-CH), 6.15 (d, 2H, J = 8.7 Hz),
1.93 (s, 6H, C6H4CH3), 1.83 (s, 6H, PC6H4CH3) ppm.
1-(Triphenylmethyl)benzimidazole. In a 100 mL flame-dried
round-bottom flask was added benzimidazole (1.00 g, 8.46
mmol) and 20 mL of THF. Once dissolved, the mixture was
cooled in an ice bath and sodium hydride (264 mg, 11.0 mmol)

Article
Organometallics, Vol. 29, No. 3, 2010 561
Cl2Ru((R)-XylBinap)(KN3-1-methylbenzimidazole)2 (3a). Or-
F4N4O4P2Ru 2H2O: C, 67.42 (67.08); H, 4.14 (4.25); N, 3.48
3
ange solid, 23% yield. 31P NMR (C6D6, 161 MHz): δ 42.79 ppm
(d, J = 36 Hz), 41.69, 41.43, 40.79 (d, J = 36 Hz). H NMR
(3.48). 31P NMR (C6D6, 161 MHz): δ 41.96 ppm. 1H NMR
(C6D6, 400 MHz): δ 9.06 (s, 2H, NCHN); 8.31 (d, 2H, J = 8.5 Hz,
Ar-CHBI); 8.11 (m, 8H, Ar-CHP-Ar); 7.09-6.65 (m, 46H, Ar-
CH); 6.47 (t, 2H, J = 8.1 Hz, Ar-CHBI); 6.40 (d, 2H, J = 8.1 ppm,
Ar-CHBI); 6.05 (d, 2H, J = 8.5 Hz, Ar-CHbiaryl) ppm.
1
(C6D6, 400 MHz): δ 9.34 (m, 2H, NCHN, three conformers);
8.90-8.69 (m, 2H, Ar-CHBinap, three conformers); 8.45-7.98
(m, 6H, Ar-CH); 7.70-7.58 (m, 2H, Ar-CHBinap, three con-
formers); 77.41-7.31 (m, 2H, Ar-CHBI, three conformers);
Cl2Ru((R)-XylPhanePhos)(KN3-1-(triphenylmethyl)benzimida-
7.06-6.20 (m, 16H, Ar-CH); 6.20, 5.96, 5.91 (4H, Ar-CHPAr
,
zole)2 (8). Anal. Found (calcd) for C100H90Cl2N4P2Ru H2O: C,
3
three conformers); 2.32, 2.26, 2.21 (NCH3, three conformers);
1.86, 1.72 (m, PC6H3(CH3)2, three conformers).
75.46 (75.08); H, 6.21 (5.80); N, 3.64 (3.54). 31P NMR (C6D6, 161
MHz): δ 33.69 ppm. 1H NMR (C6D6, 400 MHz): δ 10.02 (m, 2H,
Ar-CHPhane); 9.68 (s, 2H, NCHN); 9.48 (d, 2H, J = 8.5 Hz, Ar-
CHBI); 9.23 (m, 2H, Ar-CHPhane); 8.17 (m, 2H, Ar-CHPhane); 7.24
(m, 12H, Ar-CHTr); 7.06-6.49 (m, 32H, Ar-CH); 6.41 (d, 2H,
J = 8.5 Hz, ArCHPhane); 6.19 (d, 2H, J = 8.5 Hz, ArCHbiaryl);
3.18 (m, 2H); 2.76 (m, 2H); 2.34 (m, 2H); 1.98, 1.94, 1.83 (s, 24H,
PC6H3(CH3)2); 1.75 (m, 2H) ppm.
Cl2Ru((R)-XylBinap)(KN3-1-(triphenylmethyl)benzimidazole)2
(3b). Yellow solid, 28% yield. Anal. Found (calcd) for
C
104H88Cl2N4P2Ru 2H2O: C, 75.45 (75.08); H, 5.59 (5.57);
3
3.47 (3.37). 31P NMR (C6D6, 161 MHz): δ 38.86 ppm. 1H
NMR (C6D6, 400 MHz): δ 8.93 (s, 2H, NCHN), 8.36 (d, 2H,
J = 8.4 Hz, Ar-CHBI), 8.31 (m, 2H, Ar-CHBinap), 7.86 (br, 4H,
Ar-CHP-Ar), 7.37 (d, 2H, J = 8.5 Hz, Ar-CHBinap), 7.31 (d, 2H,
J = 8.0 Hz, Ar-CHBinap), 7.03-6.87 (m, 30H, Ar-CHTr), 6.78-
6.57 (m, 14H, Ar-CH), 6.41 (m, 4H, Ar-CHBI), 5.92 (s, 2H, Ar-
CHP-Ar), 1.90 (s, 12H, PC6H3(CH3)2), 1.79 (s, 12H, PC6H3-
(CH3)2).
Representative Hydrogenation Procedure. In the glovebox, the
glass liner of a steel autoclave was charged with acetophenone
(5.0 mmol), potassium tert-butoxide (5.6 mg, 0.05 mmol),
Cl2Ru((R)-TolBinap)(κN3-1-(triphenylmethyl)benzimidazole)2
(2b; 7.9 mg, 0.005 mmol), and 1,4-dimethoxybenzene (69.1 mg,
0.5 mmol, internal standard). The solid mixture was then
dissolved in 10 mL of 2-propanol and placed in the auto-
clave, which already contained 1 mL of solvent to prevent
unwanted movement of the liner. The autoclave was removed
from the glovebox, and the gauge block assembly was attached
and pressurized with 5 atm of hydrogen gas. The reaction
mixture was then stirred at room temperature for 6 h, after
which time the pressure was released, and the crude reaction
mixture was passed through a plug of silica to remove the
ruthenium catalyst. The crude mixture was then analyzed by
chiral supercritical fluid chromatography to determine the
enantiopurity and analyzed by 1H NMR in CDCl3 to determine
the yield.
Cl2Ru((R)-SynPhos)(KN3-1-(triphenylmethyl)benzimidazole)2
(4). Yellow solid, 53% yield. Anal. Found (calcd) for C92H72-
Cl2N4O4P2Ru H2O: C, 71.05 (71.31); H, 4.89 (4.81); N, 3.32
3
1
(3.62). 31P NMR (C6D6, 161 MHz): δ 38.25 ppm. H NMR
(C6D6, 400 MHz): δ 9.09 (s, 2H, NCHN); 8.41 (m, 6H, Ar-
CHBinap, PAr); 8.21 (m, 4H, Ar-CHPAr); 7.27 (m, 2H, Ar-
CHbiaryl); 7.08-6.88 (m, 28H, Ar-CH); 6.73 (t, 4H, J = 7.3
Hz, Ar-CHPAr); 6.64 (t, 2H, J = 7.7 Hz, ArCHBI); 6.52 (d, 2H,
J = 8.4 Hz, Ar-CHbiaryl); 6.46 (t, 2H, J = 7.7 Hz, Ar-CHBI);
6.37 (d, 2H, J = 8.4 Hz, Ar-CHBI); 3.48 (m, 8H, OCH2) ppm.
Cl2Ru((R)-C3-TunaPhos)(KN3-1-(triphenylmethyl)benzimidazole)2
(5).Brown solid, 77% yield. Anal. Found (calcd) for C91H72Cl2N4O2-
P2Ru: C, 73.52 (73.48); H, 4.88 (4.88); N, 3.77 (3.78). 31P NMR
(C6D6, 161 MHz): δ 42.48 ppm. 1H NMR (C6D6, 400 MHz): δ 9.02
(s, 2H, NCHN); 8.44 (d, 2H, J = 8.5 Hz, ArCHBI); 8.25-8.07 (m,
8H, Ar-CH); 7.23 (m, 2H, Ar-CHbiaryl); 7.10-6.43 (m, 40H, aromatic
H); 6.38 (d, 2H, J = 8.5 Hz, Ar-CHBI); 3.80-3.57 (m, 4H, OCH2);
1.31 (m, 2H, (OCH2)2CH2) ppm.
Acknowledgment. The Natural Sciences and Engineer-
ing Research Council of Canada (NSERC) is acknowl-
edged for support of this research in terms of a discovery
grant to C.M.C. and a Canada Graduate Scholarship
to J.M.P. The Canada Foundation for Innovation is
also thanked for providing funds for instrumentation.
Dr. Franc-oise Sauriol is thanked for assistance with
NMR spectroscopy.
Cl2Ru((R)-Cl-OMe-BIPHEP)(KN3-1-(triphenylmethyl)benzimida-
zole)2 (6). Yellow solid, 58% yield. Anal. Found (calcd) for C90H70-
Cl4N4O2P2Ru H2O: C, 68.92 (69.19); H, 4.83 (4.64); N, 3.49 (3.59).
3
1
31P NMR (C6D6, 161 MHz): δ 40.43 ppm. H NMR (C6D6, 400
MHz): δ 9.00 (s, 2H, NCHN); 8.37 (d, 2H, J = 8.3 Hz, Ar-CHBI);
8.14 (m, 8H); 7.28 (m, 2H, Ar-CHbiaryl); 7.01-6.72 (m, 44H); 6.67 (t,
2H, J=7.3Hz,Ar-CHBI);6.47(t,2H,J=8.3Hz,Ar-CHBI);6.42(d,
2H, J = 8.1 Hz, Ar-CHBI); 3.47 (s, 6H, OCH3) ppm.
Supporting Information Available: A CIF file giving crystal-
lographic data for 3a and text giving the experimental procedure
for acquisition of X-ray data. This material is available free of
charge via the Internet at http://pubs.acs.org.
Cl2Ru((R)-FluorPhos)(KN3-1-(triphenylmethyl)benzimidazole)2
(7). Yellow solid, 79% yield. Anal. Found (calcd) for C90H64Cl2-