H.-y. Jiang et al. / Catalysis Communications 11 (2010) 584–587
585
chiral modifier 1 obtained 39% conversion and 68% ee (40 °C,
40 h). These indicate that phosphine not only stabilized the metal
particles in the preparation of catalyst but also served to addition-
ally modify the supported metal with diamine in the asymmetric
hydrogenation. Because of the famous Ru/phosphine/diamine cat-
alyst in homogeneous catalysis [8,9], the homogeneous
RuCl2[P(C6H5)3]3/diamine 1 catalyst was also investigated accord-
ing to the literature [9] for the hydrogenation of acetophenone
but poor result was given (50% ee). Mercury-poisoning experiment
[20–23], which can selectively poison metal nanoparticles, by
forming an amalgam with mercury, is now used to clarify whether
2. Results and discussion
Phosphine stabilized 1%Ru/c-Al2O3 catalysts were prepared
according to our previously reported method [11]. This catalyst
system was well elucidated to be a heterogeneous catalysis by con-
ventional filtering test [17,18], etc. in our previous report [11]. The
solution and catalyst were separated at the end of the hydrogena-
tion or at low conversion (15%), by high speed centrifugation under
argon atmosphere. It was found that the reacted solution had no
catalytic activity after some fresh substrate was introduced in it.
However, the catalyst obviously maintained the hydrogenation
activity and enantioselectivity. After the first hydrogenation of ace-
tophenone, the filtrate was determined by ICP-AES, only 0.03% Ru
and 0.36% PPh3 were leached to the organic solvent. The molar ra-
tio of PPh3/Ru leached into solution is large (12:1), which might
poisoned metal Ru leached into the solution according to literature
[10,28] and experiment phenomena (Fig. 2 in Supplementary
Noyori-type Ru(II) catalyst confined in the porous c-Al2O3 solids or
ligands modified Ru(0) catalyst plays the catalytic role in our sys-
tem. In such an experiment for our system, the conversion in the
presence of modifier 1 was 24% after the initial 60 min; mercury
was then added under argon atmosphere and the reaction was
completely terminated as evidenced by no change in the conver-
sion (24.5%) after an additional reaction time of 2 h (Fig. 3 in Sup-
plementary material). So, the possibility of Noyori-type Ru(II)
material). Furthermore, the content of Cl in 1%Ru/c-Al2O3/2tpp
was measured by X-ray photoelectron spectroscopy (XPS) [19],
and no Cl was detected. The catalysts are remarkably stable in
the solid state and can be stored open to air.
catalyst confined in the porous c-Al2O3 solids is denied. Since the
properties of the stabilizers greatly influence the particle size and
dispersion of the reduced metal in the catalyst, phosphines, such
as tris(4-methoxyphenyl)phosphine (motpp), tris(4-trifluorometh-
ylphenyl)phosphine (tftpp), tris(sodium-m-sulfonatophenyl) phos-
phine (tppts), 1,2-bis(diphenylphosphinomethyl)benzene (bdpx),
2,20-bis(diphenylphosphinomethyl)-1,10-biphenyl (bisbi), were
tested (Table 2 in Supplementary material), however, none of them
were better than the simple PPh3. Furthermore, the influence of
molar ratio of PPh3 to Ru on the catalytic performance was inves-
tigated. When the initial molar ratio of PPh3 to Ru is 2:1 in the
preparation of catalysts, the highest activity and enantioselectivity
was obtained. Increasing or decreasing the ratio of PPh3 to Ru was
unfavorable to both the activity and the enantioselectivity (Fig. 2 in
Supplementary material). Since the choice of solvent has a great
impact on the enantioselectivity of the hydrogenation, different
solvents were tested (Table 1 in Supplementary material). iPrOH
is the best choice for a solvent. The reaction in methanol or ethanol
is much slower, while H2O, THF and toluene are not usable. Similar
to some previous work [11,17], the results during solvent choice
indicated that solvent polarity obviously influence hydrogenation
activity and enantioselectivity. In the absence of H2, no conversion
was observed with or without modifier. So, the absence of transfer
hydrogenation was confirmed according to literature [9]. Ligand
acceleration, which is commonly perceived in heterogeneous cata-
lytic systems [10,11,24], is also observed in this system. Cinchona
modifiers not only render the catalyst enantioselective but
strongly accelerate the hydrogenation. However, if in the absence
of modifier 1, the conversion of acetophenone hydrogenation was
only 4% and the racemic product was obtained (Fig. 1 in Supple-
mentary material). The addition of both modifier 1 and KOH
greatly enhanced the activity (Table 1, entry 1). Similar to some re-
cent reports [17,25], a high modifier concentration is needed in the
simple aromatic ketone asymmetric hydrogenation. By using
1 mol% (compared to supported metal Ru) of chiral diamine 1,
the enantioselectivity was maintained during the asymmetric
hydrogenation at a prolonged time (40 °C, 40 h, 100% conversion
and 72% ee). Similar to homogeneous [8,9] and some heteroge-
neous [11,17] reactions, a base is necessary to accelerate the reac-
tion. The addition of different bases, such as tBuOK (17%
conversion and 80% ee), LiOH (2% conversion and 40% ee), NaOH
(96% conversion and 78% ee), KOH (99% conversion and 83% ee),
can enhance both the activity and the enantioselectivity more or
less. For the alkali metal hydroxides both activity and enantioselec-
tivity decrease in the order of K+ > Na+ > Li+. However, if there is no
base in the reaction solution, the conversion was only 2% with 32%
ee. The results during base choice indicated that the positive effect
IR is effective for the investigation of the mutual ligand–metal–
support interactions of solid catalysts [10,29]. Information on the
interaction between the metal and the phosphine was confirmed
by FTIR methods in which the red shift of the P-C6H5 deformation
around 1434 cmꢀ1 in 1%Ru/
c-Al2O3/2tpp is an indication of the
coordination of PPh3 to the metal. Interaction between the cin-
chona alkaloid and the metal was also confirmed by FTIR analysis
in which the blue shift of the N–H deformation around 3366 cmꢀ1
in 1%Ru/c-Al2O3/2tpp/diamine 1 is an indication of the coordina-
tion of the amine with the metal. It is worth noting that similar
phenomena were observed in our previous report [10].
Asymmetric hydrogenation of aromatic ketones was performed
in a 20 ml stainless autoclave with a magnetic stirrer bar, by using
PPh3 stabilized 1%Ru/c-Al2O3 as a catalyst, in the presence of chiral
diamine 1 (9-amino(9-deoxy)epicinchonine) (Scheme 1) derived
from natural cinchona alkaloid as chiral modifier. Up to 99% con-
version and 83% ee were obtained for the asymmetric hydrogena-
tion of acetophenone. However, the hydrogenation of
acetophenone catalyzed by 1%Ru/c-Al2O3 just showed a conver-
sion of as low as 4% and an ee of 10% in the presence of chiral dia-
mine 1. Classically modified 1%Ru/c-Al2O3 catalyst with tpp and
N
H
H
H2N
N
1
OH
O
1%Ru/phosphine/ -Al2O3
diamine
R2
*
R2
+
H2
R1
R1
iPrOH
a: R1 = H, R2 = CH3
b: R1 =
d: R1 =
f: R1 =
h: R1 =
o
-F, R2 = CH3
-Br, R2 = CH3
-CF3, R2 = CH3
m-Br, R2 = CH3
c: R1 =
e: R1 =
g: R1 =
i: R1 =
o
-Cl, R2 = CH3
-OMe, R2 = CH3
-Cl, R2 = CH3
-OMe, R2 = CH3
o
o
o
m
p
j: R1 = H, R2 = CH2CH3
k: R1 = H, R2 = CH(CH3)2
Scheme 1. Modifier used and the general reaction.