Catalytic enantiodifferentiating hydrogenation with commercial nickel powders chirally modified by tartaric acid and sodium bromide

Article abstract of DOI:10.1002/cctc.201300727

The chirally modified nickel catalysts for the enantiodifferentiating hydrogenation of β-ketoesters are prepared conventionally by immersing hydrogen-activated metallic nickel into an aqueous solution of enantiopure tartaric acid, in which the preactivation of nickel is essential. Herein, we revealed that even commercially available nickel powders without any pretreatment can catalyze the enantiodifferentiating hydrogenation of β-ketoesters to give the corresponding β-hydroxyesters in quantitative yield and high enantioselectivity (up to 91 %) under optimized conditions. The immediate use of commercially available nickel powders and the reproducible high chemical and optical yields not only expand the scope of heterogeneous asymmetric catalysis but also pave the way for the practical application and industrial use of chirally modified nickel catalysts. Copyright

Full text of DOI:10.1002/cctc.201300727

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201300727
Catalytic Enantiodifferentiating Hydrogenation with
Commercial Nickel Powders Chirally Modified by Tartaric
Acid and Sodium Bromide
[
a]
[b]
[a]
[b]
[b]
Tsutomu Osawa,* Tomoko Kizawa, Fumika Takano, Shinji Ikeda, Takayuki Kitamura,
[
c]
[b, c]
Yoshihisa Inoue, and Victor Borovkov*
The chirally modified nickel catalysts for the enantiodifferenti-
ating hydrogenation of b-ketoesters are prepared convention-
ally by immersing hydrogen-activated metallic nickel into an
aqueous solution of enantiopure tartaric acid, in which the pre-
activation of nickel is essential. Herein, we revealed that even
commercially available nickel powders without any pretreat-
ment can catalyze the enantiodifferentiating hydrogenation of
b-ketoesters to give the corresponding b-hydroxyesters in
quantitative yield and high enantioselectivity (up to 91%)
under optimized conditions. The immediate use of commer-
cially available nickel powders and the reproducible high
chemical and optical yields not only expand the scope of het-
erogeneous asymmetric catalysis but also pave the way for the
practical application and industrial use of chirally modified
nickel catalysts.
Introduction
[6–9]
Optically active compounds are the key elements of natural
acid (TA)-modified nickel catalysts.
Both catalysts give high
[
1]
systems such as living organisms. Therefore, the production
of such compounds is of prime importance, especially in the
fields of pharmaceuticals, agrochemicals, and so on. Further-
more, novel and emerging approaches to supramolecular
chemistry, nanoscience, biomimetics, and sensing also require
optically active molecules of high enantiopurity. Several meth-
ods, such as optical resolution and catalytic and enzymatic
asymmetric syntheses, have been developed thus far to obtain
enantiomeric compounds. Of these approaches, heterogene-
ous chiral catalysis is one of the most promising techniques for
the large-scale production of enantiopure compounds, featur-
ing facile preparation, simple separation, and easy recovery
and reuse of the catalyst, as well as the time- and cost-saving,
enantioselectivities for specific prochiral substrates. The plati-
num-based chiral catalysts hydrogenate activated ketones,
such as a-ketoesters, ketopantolactones, pyrrolidinetriones, a-
ketoacetals, a-ketoethers, a-diketones, and other related com-
[10–18]
pounds, in 95–98% ee,
catalysts reduce a,b-unsaturated carboxylic acids and alkene
whereas the modified palladium
[19–22]
derivatives in 90–94% ee.
In contrast, the TA-modified
nickel catalysts prepared in the presence of NaBr reduce b-ke-
toesters and 2-alkanones and give the corresponding alcohols
[23]
[24]
in up to 98 and 85% ee, respectively.
In the preparation of TA-modified nickel catalyst, activated
[
25–28]
metallic nickel powders, such as Raney nickel,
reduced
nickel (prepared by the reduction of nickel oxide), supported
[29]
[
2]
[30–32]
[33,34]
environmentally benign methods. Practically, there are two
most effective heterogeneous catalytic systems for enantiodif-
ferentiating hydrogenation: (1) cinchona alkaloid-modified
nickel,
and activated commercial nickel powder,
are
commonly used. The commercially available nickel powder is
usually activated through treatment with hydrogen stream at
an elevated temperature, which is followed by the chiral modi-
fication by immersing the activated nickel powder into an
[
3–5]
metallic platinum and palladium catalysts
and (2) tartaric
[33,34]
aqueous solution containing TA and NaBr.
pretreatment is not necessary if the commercial nickel powder
The hydrogen
[a] Dr. T. Osawa, F. Takano
Graduate School of Science and Engineering for Research
University of Toyama
[35]
is used as a nickel base, because the modification solution
(containing TA and NaBr) is adjusted to pH 3.2 and thus it can
remove the oxidized material from the nickel surface. Thus, the
TA-NaBr solution plays dual roles of cleaning and modifying
the nickel surface to afford a smooth surface structure appro-
3
190 Gofuku, Toyama 930-8555 (Japan)
Fax: (+81)76-445-6549
E-mail: osawa@sci.u-toyama.ac.jp
[b] T. Kizawa, S. Ikeda, T. Kitamura, Dr. V. Borovkov
Metek Co., Ltd.
[9]
priate for the enantiodifferentiation by removing defects.
1
Warada-cho, Kamitoba
Minami-ku, Kyoto 601-8133 (Japan)
This risk-free surface activation without using the hydrogen
pretreatment is an important step toward the large-scale pro-
duction and application of chirally modified nickel catalysts in
industry.
[
c] Prof. Y. Inoue, Dr. V. Borovkov
Department of Applied Chemistry
Osaka University
Yamada-oka, Suita 565-0871 (Japan)
Fax: (+81) 6-6879-7923
Herein, to establish the protocol for preparing highly effi-
cient TA/NaBr-modified nickel catalysts without preactivation,
E-mail: victrb@chem.eng.osaka-u.ac.jp
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 170 – 178 170

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Scheme 1. Enantiodifferentiating hydrogenation of MAA.
the parameters for chiral modification and subsequent hydro-
genation were optimized by using commercially available
nickel powders as a nickel source and methyl acetoacetate
(MAA; Scheme 1) as a benchmark substrate for assessing the
hydrogenation and enantiodifferentiation abilities.
Results and Discussion
The overall effectiveness and enantioselectivity of the catalytic
hydrogenation are generally affected by the conditions used in
the following three major stages: (1) the preparation and acti-
vation of metallic nickel, (2) the chiral modification, and (3) the
hydrogenation reaction. The previous comprehensive studies
to optimize the conditions enabled us to establish the best
protocol for each step of the catalytic hydrogenation with the
Figure 1. SEM images of commercial nickel powders of different sizes:
a) 3 mm nickel (magnification: ꢁ10000); b) 5 mm nickel (ꢁ10000); c) 100 nm
nickel (ꢁ30000); d) submicrometer nickel (ꢁ10000); e) 150 mm nickel
[
25,27,28]
chirally modified Raney nickel catalyst.
In contrast, the
(
ꢁ10000). Scale bars: 1 mm (a,b,d,e), 100 nm (c).
direct use of commercially available nickel powders in chiral
hydrogenation does not appear to have been explored inten-
[
33–35]
sively,
although they are more suitable for the industrial
use from the economic and environmental points of view. Our
experience with the chirally modified Raney nickel catalyst was
proven to be useful in optimizing the conditions used in the
three major steps of heterogeneous hydrogenation with chiral-
ly modified nickel powders.
of similar morphology whereas the other nickel powders were
appreciably different in shapes and surface structures. Thus,
the 100 nm and submicrometer nickel powders, which were
also aggregates, had smoother surfaces and more spherical
shapes. The 150 mm nickel powder appeared to be positioned
morphologically in between these two extrema, which had
smoother surfaces than those of the 3 mm and 5 mm nickel
powders but less round shape than that of the 100 nm and
submicrometer nickel powders. Such a distinction is apparently
due to the use of different preparation methods.
Choice and pretreatment of nickel powders
The effects of the source and type of nickel powders as well as
the activation of the nickel surface on the overall hydrogena-
tion efficiency and enantioselectivity were examined with com-
mercially available nickel powders of different sizes.
In the chiral modification, the optimum amount of NaBr in
the modification solution varies with the type of nickel source
[36]
used. Thus, the effect of the amount of NaBr added on enan-
tioselectivity was examined for each type of nickel powder
Source and type of nickel powders
(
with a fixed amount of TA as 0.5 g). As shown in Figure 2, all
The performance of TA/NaBr-modified nickel catalyst signifi-
cantly depends on the nature of metallic nickel used, for exam-
ple, the particle size and morphology of the commercial nickel
types of nickel powders, except the submicrometer nickel
powder, afforded the hydrogenation product in high enantio-
selectivities (ranging from 84 to 91% ee) at the optimized NaBr
concentration; the best ee value was obtained with the 5 mm
nickel powder after the addition of 2 g of NaBr. The poorer ee
values obtained with the submicrometer nickel powder may
be related to its different shape and surface structure
(Figure 1), which apparently discourage the adsorption of the
TA–MAA complex on the nickel surface in right conforma-
[
33]
powder to be activated and the source of nickel oxide to be
[
36]
reduced. Therefore, the commercially available nickel pow-
ders of different sizes were first characterized morphologically
by using SEM and then were examined for the effects of the
source and type of nickel powders on the catalytic activity and
enantioselectivity.
[9]
The analysis of SEM images of the commercial nickel pow-
ders of different sizes (Figure 1) revealed that the 3 mm and
tion—an essential condition for achieving high ee.
The conversion of MAA in the same reaction was also a criti-
cal function of the amount of NaBr added as well as the type
5
mm nickel powders were aggregates of sharp-edged particles
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 170 – 178 171

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
from the aforementioned results, we selected 5 mm and
100 nm nickel powders for further experiments to establish the
optimized hydrogenation conditions.
Nonessentiality of hydrogen activation before chiral
modification
The pretreatment of nickel powder with hydrogen at an elevat-
[33,34]
ed temperature (to reduce nickel oxides on the surface
) is
a common, apparently indispensable, method in the standard
protocol for preparing a highly efficient TA-modified nickel cat-
alyst. Hence, the effects of pretreatment temperature on con-
version and enantioselectivity were examined for the 100 nm
nickel powder at 473–633 K, and the results are illustrated in
Figure 4. The conversion was consistently quantitative
Figure 2. Effects of the amount of NaBr added to the modification solution
containing (R,R)-TA on the enantioselectivity of methyl 3-hydroxybutyrate
obtained upon hydrogenation with chirally modified nickel powders of dif-
ferent sizes: 5 mm (^); 3 mm (
&
); 100 nm (~); 150 mm (!); submicrometer
(
*). Modification: TA (0.5 g); pH 3.2; temperature 373 K. Hydrogenation: ad-
ditive acetic acid (0.1 g); hydrogen pressure 9 MPa; temperature 373 K.
Figure 4. Effects of the temperature for hydrogen pretreatment on the con-
version (~) and enantioselectivity (*). Nickel powder: 100 nm. Modification:
TA (0.5 g); NaBr (3.0 g); pH 3.2; temperature 373 K. Hydrogenation: additive
acetic acid (0.1 g); hydrogen pressure 9 MPa; temperature 373 K.
throughout the temperature range examined. In contrast, the
enantioselectivity increased appreciably from 84 to 89% ee
with the increase in the temperature from 473 to 573 K owing
to accelerated reduction by hydrogen; however, heating above
this temperature gradually decreased the ee to 86% at 633 K,
probably owing to the sintering of the nickel powder
Figure 3. Effects of the amount of NaBr added to the modification solution
containing (R,R)-TA on the conversion of MAA upon hydrogenation with
chirally modified nickel powders of different sizes: 5 mm (^); 3 mm (
00 nm (~); 150 mm (!); submicrometer (*). Modification: TA (0.5 g);
&
);
1
pH 3.2; temperature 373 K. Hydrogenation: additive acetic acid (0.1 g); hy-
drogen pressure 9 MPa; temperature 373 K.
(
vide infra).
The catalyst prepared with the 5 mm nickel powder was
of nickel powder used (Figure 3). The 100 nm and 5 mm nickel
powders achieved nearly quantitative conversions over the
entire range of NaBr concentration used, except the case of
much more sensitive to temperature, and the conversion and
enantioselectivity decreased steadily with temperature, which
reached 48% conversion and 15% ee at 573 K (Figure 5) owing
to more pronounced sintering (vide infra).
0
8
g of NaBr for the latter catalyst, which nevertheless gave
3% conversion. The 3 mm nickel powder was sensitive to the
The analysis of SEM images taken before and after the hy-
drogen activation at 573 K (Figures 6 and 7) revealed that the
pretreatment caused sintering with the corresponding mor-
phological changes in nickel particles, which were more signifi-
cant for the 5 mm nickel powder than for the 100 nm nickel
powder. In addition, the different degree of sintering may
affect the crystal structure. The XRD analyses of the two sam-
ples indicated that the pretreatment at 573 K augmented the
mean crystallite size of the 100 nm nickel powder from 85 to
109 nm and that of 5 mm nickel powder from 103 to >200 nm.
overdose of NaBr, which led to a significant decrease in con-
version upon the addition of 3 and 5 g of NaBr. This decrease
is due to the excessive adsorption of NaBr on the nickel sur-
face that blocks the access of the substrate to the catalyst. The
100 nm nickel powder showed a much weaker but a similar
tendency upon the addition of 5 g of NaBr. The 150 mm and
submicrometer nickel powders gave modest and low conver-
sions, respectively, regardless of the amount of NaBr added,
which indicates their low hydrogenation activities. Judging
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 170 – 178 172

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
nickel catalyst, for example, by skipping the potentially hazard-
ous hydrogen activation process. The nickel catalysts prepared
by directly immersing the commercial 100 nm and 5 mm nickel
powders into the modification solution possessed equally high
hydrogenation and enantiodifferentiation abilities as did those
with hydrogen preactivation, which afforded 86% ee (100%
conversion) and 90% ee (99% conversion), respectively. These
results confirm our observation with the 3 mm nickel powder
that the hydrogen activation of nickel before chiral modifica-
tion is not necessarily indispensable to achieve high enantiose-
lectivity and conversion, as the modification solution at pH 3.2
[35]
can activate the nickel surface by removing nickel oxides.
This finding is crucial from a practical application point of view
because other nickel sources require hydrogen activation. The
direct use of commercially available nickel powders without
any pretreatment eliminates the potential risk of using hydro-
gen at elevated temperatures and facilitates the large-scale
production of optically active compounds.
Figure 5. Effects of the temperature for hydrogen pretreatment on the con-
version (~) and enantioselectivity (*). Nickel powder: 5 mm. Modification: TA
(0.5 g); NaBr (2.0 g); pH 3.2; temperature 373 K. Hydrogenation: additive
acetic acid (0.1 g); hydrogen pressure 9 MPa; temperature 373 K.
Optimization of the modification conditions
As the chiral modification of metallic nickel is an essential part
of the enantiodifferentiating hydrogenation, a considerable
portion of our study has thus far been devoted to the optimi-
zation of major modification parameters, such as pH, tempera-
[25,27,28,37]
ture, immersion period, and solvent.
Of these factors,
the pH of the modification solution and the amounts of TA
and NaBr are the two most important parameters to be
optimized.
Figure 6. SEM images of the 100 nm nickel powder a) before and b) after hy-
drogen pretreatment at 573 K (magnification: ꢁ50000). Scale bars: 100 nm
Optimization of the solution pH
(a,b).
The above experiments have shown that both the 100 nm and
5
mm nickel powders are equally suitable for the preparation of
TA/NaBr-modified nickel catalysts without hydrogen activation,
but the latter is much less expensive. Thus, we used the 5 mm
nickel powder in our further endeavor to optimize the pH of
the solution for higher conversion and enantioselectivity.
The chiral modification of the 5 mm nickel powder was per-
formed at pH values varying from 2.0 to 6.5 without changing
other parameters or methods to show modest pH dependen-
ces. The conversion increased quickly with the increase in pH
and reached approximately 100% at pH>3.0, whereas the
enantioselectivity maximized at pH 3.0–3.2 and attained the
highest ee value (91%), which decreased to 80% at higher pH
values (Figure 8). This trend and the ee values are identical to
those reported for Raney nickel, reduced nickel, and fine nickel
Figure 7. SEM images of the 5 mm nickel powder a) before and b) after hy-
drogen pretreatment at 573 K (magnification: ꢁ40000). Scale bars: 100 nm
(a,b).
[25,38,39]
powders with hydrogen pretreatment.
This finding indi-
The significant increase in the crystallite size of the 5 mm nickel
powder could be one of the reasons for the observed decrease
in hydrogenation activity, although the detailed mechanism is
yet to be clarified.
cates the same chiral modification mechanism for all types of
nickel catalysts, which is controlled by two major counterbal-
[9]
ancing factors: surface corrosion and TA adsorption. At pH<
4.5, TA exists as a monoacid and/or a diacid (pKa2 of TA=4.25)
and corrodes the nickel surface to provide a fresh surface suit-
able for chiral modification, whereas at pH>3.0 (pKa1 of TA=
2.95), TA is adsorbed on the nickel surface as a monosodium
salt, which is the effective chiral species that induces high
Although the chirally modified nickel with preactivation with
hydrogen can give the hydrogenation product in high enantio-
selectivities of up to 89% ee and quantitative yield under opti-
mized conditions, we wanted to avoid the effects of sintering
and simplify the protocol for preparing the chirally modified
[9]
enantioselectivity. Counterbalancing these two conflicting
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 170 – 178 173

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Figure 8. Effect of the pH of the modification solution on the conversion (~)
and enantioselectivity (*). Nickel powder: 5 mm. Modification: TA (0.5 g);
NaBr (2.0 g); temperature 373 K. Hydrogenation: additive acetic acid (0.1 g);
hydrogen 9 MPa; temperature 373 K.
factors is essential to maximize the enantioselectivity, which
was achieved at pH 3.0–4.0, or more specifically at 3.0–3.2, in
the present case. In this context, it is reasonable that all the
chirally modified nickel catalysts attain the best enantioselec-
tivity and 100% conversion in this particular pH region, irre-
spective of the nickel source used.
To better understand the origin of the pH effect, we per-
formed the X-ray photoelectron spectroscopy (XPS) analyses of
the chirally modified nickel catalysts prepared at different pH
values to characterize the surface composition. Preliminary re-
sults of the pH dependence were reported in Ref. [40]. The
content of metallic versus oxidized nickel is important to evalu-
ate the catalyst properties, and the 2p region of nickel is the
most suitable for this purpose, as the composition of the met-
allic and oxidized forms of nickel can be assessed quantitative-
Figure 9. Effects of the pH of the modification solution on the XPS spectra
of the 5 mm nickel powder: a) pH 2.0; b) pH 3.2; c) pH 6.5.
Optimization of the amounts of TA and NaBr
0
2+
3+
ly (Figure 9). Thus, the Ni /(Ni +Ni ) ratios of 17:83, 43:57,
and 35:65 were obtained for the modified nickel catalysts pre-
pared at pH 2.0, 3.2, and 6.5, respectively. These results are
consistent with our earlier report that the modification at
pH 3.2 activated the nickel surface by removing the oxidized
The amounts of TA and NaBr in the modification solution are
also crucial parameters that need to be optimized to attain
[25,28,41]
high enantioselectivity and conversion.
Although the ab-
solute configuration of TA is directly responsible for the chiral
nature of the hydrogenation product obtained, NaBr is an aux-
iliary modifier to increase the enantioselectivity by blocking
the non-enantio-differentiating sites (a nickel domain in which
[
35]
material from the surface. Although the TA solution has the
highest ability to remove the oxidized nickel species by corrod-
ing the surface at pH 2.0 of all the pH values examined, it also
has the highest ability to produce nickel tartrate on the sur-
face. Therefore, the surface modified at pH 2.0 had the lowest
[42]
the racemic products are produced).
The effects of the
amounts of TA and NaBr on enantioselectivity and conversion,
with the use of the 5 mm nickel powder without hydrogen pre-
treatment as a catalyst base, are summarized in Table 1.
0
proportion of Ni . In contrast, the surface modified at pH 6.5
0
had a moderate amount of Ni , as the TA solution at pH 6.5
has the lowest activity of corrosion and negligible ability to
produce nickel tartrate because the pKa2 of TA is 4.25, which in-
dicates that disodium tartrate is the dominant species at
pH 6.5. Because of these two incompatible processes, the
Table 1. Effects of the amounts of TA and NaBr on enantioselectivity and
conversion (with the use of 5 mm nickel powder).
0
modification at pH 3.2 afforded the highest Ni content. This
TA
g]
Amount of NaBr, Enantioselectivity [%] (Conversion [%])
[
0 g
0.5 g
1 g
2 g
3 g
5 g
observation is reasonable because the TA solution at pH 3.0–
3
.2, which contains both diacid and monosodium salts of TA at
0.25
0.5
86 (100)
87 (96)
90 (87)
90 (82)
86 (100)
90 (99)
90 (69)
89 (66)
89 (100)
87 (100)
90 (77)
90 (73)
91 (93)
89 (99)
90 (74)
89 (58)
pH close to the pKa1 value (2.95), is active enough to remove
the oxidized species from the surface of the nickel particles
and modify the surface with the monosodium salt of TA.
77 (83)
77 (89)
87 (99)
89 (83)
1
2
.0
.5
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 170 – 178 174

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
As can be observed from Table 1, both the enantioselectivity
and conversion were moderately affected by the amounts of
TA and NaBr. Thus, the addition of NaBr (0.5–5 g) increased the
product’s ee from 77 to 86–91%, whereas the highest conver-
sions of 96% or more were achieved in a narrower region of
the TA–NaBr matrix, that is, the region with 0.25–0.5 g of TA
and 0.5 g or more of NaBr. A closer examination of the data re-
veals a sort of competition between TA and NaBr. Thus, the ee
value of 89% or more was achieved upon addition of 3 g of
NaBr in the presence of 0.25 g of TA whereas only 0.5 g of
NaBr was needed to obtain the same level of ee in the pres-
ence of 1 g of TA. In contrast, the use of smaller amounts of TA
and NaBr led to higher conversions and the overdose de-
creased the conversion and it reached 58% at the largest
amounts of TA (2.5 g) and NaBr (5 g). This finding seems rea-
sonable because TA and NaBr competitively block the active
sites of the nickel surface. Considering the two conflicting fac-
tors, we used the modification solution containing 0.5 g of TA
and 2 g of NaBr to obtain high enantioselectivity and nearly
quantitative conversion in the case of the 5 mm nickel powder.
Unexpectedly, the 100 nm nickel powder (without hydrogen
pretreatment) modified with TA and NaBr under identical con-
ditions behaved differently upon hydrogenation. As shown in
Table 2, the conversion was consistently quantitative in most
ee) in each case. However, the 100 nm nickel powder is consid-
erably more expensive than the 5 mm nickel powder, which
prompted us to use the modified 5 mm nickel powder (without
hydrogen pretreatment) to optimize the conditions for the
final hydrogenation step.
Optimization of the hydrogenation conditions
The factors affecting the hydrogenation step include solvent,
substrate concentration, additive and its concentration, tem-
[9,43–45]
perature, and hydrogen pressure.
However, the solvent
and the substrate concentration have already been optimized
for the catalytic hydrogenation of MAA in the previous stud-
[43,44]
ies,
and thus the remaining influential factors, that is, tem-
perature and pressure, were studied in detail here.
Effect of the hydrogenation temperature
The examinations of the results of foregoing studies on the
catalytic hydrogenation of MAA with chirally modified nickel
[44,46–48]
catalysts
revealed that the temperature-dependent be-
havior varies with the catalyst and/or the phase used. For ex-
ample, an optimum temperature to obtain the highest ee exist-
[46]
[44]
ed in the liquid phase hydrogenation at 0.1 or 10 MPa
whereas the enantioselectivity was independent of the tem-
[47]
perature in the liquid phase hydrogenation at 9 MPa, but it
decreased with increasing temperature in the gas phase hydro-
Table 2. Effects of the amounts of TA and NaBr on enantioselectivity and
conversion (with the use of 100 nm nickel powder).
[48]
genation at 0.1 MPa. Although some of the results are puz-
zling and not readily explainable, it is clear that the hydrogena-
tion temperature has to be optimized for each catalyst.
TA
g]
Amount of NaBr, Enantioselectivity [%] (Conversion [%])
[
0 g
0.5 g
1 g
2 g
3 g
5 g
0
1
2
5
.5
.0
.5
.0
81 (100)
80 (100)
81 (100)
83 (100)
82 (100)
84 (100)
88 (100)
85 (100)
89 (100)
91 (100)
80 (96)
87 (100)
89 (100)
87 (100)
The catalytic hydrogenation of MAA with the 5 mm nickel
powder (without hydrogen pretreatment) modified with 0.5 g
of TA and 2.0 g of NaBr at pH 3.2 was performed at tempera-
tures ranging from 353 to 393 K and at a hydrogen pressure of
72 (100)
59 (100)
87 (100)
9
MPa. The enantioselectivity increased gradually with temper-
ature and reached the highest value of 89% ee at 373 K and
decreased quickly thereafter and reached 77% ee at 393 K
examined cases, except the low TA/high NaBr case, whereas
the enantioselectivity was modest (59–81% ee) at low amounts
of TA and NaBr but increased gradually with an increase in the
amounts of TA and NaBr and attained the highest ee value
(
Figure 10), which indicated structural deformation and/or de-
(91%) upon modification with a solution containing 2.5 g of TA
and 3 g of NaBr. Notably, the two nickel powders of different
sizes afford the same optimized enantioselectivity of 91%,
which indicates that they share the same coordination geome-
try around the active hydrogenation site on the nickel surface
despite significant differences in size and morphology (see Fig-
ure 1b and c). Notably, the 100 nm nickel particles, possessing
larger surface areas, show higher hydrogenation activity and
require larger amounts of TA and NaBr to achieve the full cov-
erage of the surface.
From a practical application point of view, the cost of the
nickel source is another important factor to be taken into ac-
count. In the catalytic asymmetric hydrogenation of MAA, the
hydrogenation ability assessed by the conversion in a given re-
action period is appreciably higher for the 100 nm nickel
powder (100%) than for the 5 mm nickel powder (93%) under
the condition that affords the highest enantioselectivity (91%
Figure 10. Effects of the hydrogenation temperature on the conversion (~)
and enantioselectivity (*). Nickel powder: 5 mm. Modification: TA (0.5 g);
NaBr (2.0 g); temperature 373 K. Hydrogenation: additive acetic acid (0.1 g);
hydrogen pressure 9 MPa.
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 170 – 178 175

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
tachment of the adsorbed TA at higher temperatures. In con-
trast, the conversion was rather low (<40%) at 353 K or lower
temperatures but it suddenly increased to 98–100% at 373 K
and higher temperatures, which reflected the increased hydro-
genation ability of the nickel surface. This result indicates that
the favorable adsorption of MAA on the nickel surface produc-
ing (R)-alcohol is followed by the rate-limiting hydrogenation
enantioselectivity at hydrogen pressures varying from 6 to
11 MPa.
Effect of the amount of acetic acid added to the reaction
medium
The addition of acetic acid to the reaction medium is necessary
to achieve the high enantioselectivity for which the accelerat-
ed hydrogenation on the enantiodifferentiating sites is respon-
[
49]
step.
[52]
sible. Therefore, in all the aforementioned experiments, we
performed the catalytic hydrogenation of MAA in the presence
of 0.1 g of acetic acid as a default value, the amount of which
was finally optimized to give the results shown in Figure 12.
Effect of the hydrogen pressure
Some conflicting results have been reported for the effect of
hydrogen pressure on the hydrogenation of MAA with sup-
ported and Raney nickel catalysts. Nitta et al. reported that the
enantioselectivity decreased with increasing hydrogen pressure
(0.1–13 MPa) upon hydrogenation with the chirally modified
Ni/SiO catalyst under conditions in which the rate was con-
2
[
50]
trolled by the surface reaction. Kukula and Cerveny reported
that the enantioselectivity increased slightly with increasing
pressure (1–12 MPa) upon hydrogenation with the chirally
[
44]
modified Raney nickel catalyst. Some of us reported that the
enantioselectivity was constant at 1–9 MPa but decreased sig-
nificantly below 0.5 MPa upon hydrogenation with the chirally
[
51]
modified Raney nickel catalyst. These apparent discrepancies
may arise from the different nature of the chiral catalysts pre-
pared with various nickel sources under various conditions and
presumably from the use of different types of reactors.
Hence, we performed the catalytic hydrogenation of MAA
with our own nickel catalyst at hydrogen pressures varying
from 6 to 11 MPa to obtain the pressure-dependence profiles
of the ee and conversion (Figure 11). The enantioselectivity in-
creased appreciably with pressure and reached the highest ee
value (91%) at 10 MPa. Similar to the temperature-dependence
profile (Figure 10), the conversion increased from 36% at
Figure 12. Effects of the amount of acetic acid in the reaction mixture on
the conversion (~) and enantioselectivity (*). Nickel powder: 5 mm. Modifica-
tion: TA (0.5 g); NaBr (2.0 g); temperature 373 K. Hydrogenation: hydrogen
pressure 9 MPa; temperature 373 K.
The enantioselectivity was not much affected by the addition
of acetic acid at least up to 0.2 g, which afforded 88–90% ee,
but it decreased appreciably to 76% ee upon addition of 0.3 g
of acetic acid. The conversion was a critical function of the
amount of acetic acid, which increased appreciably upon the
addition of 0.05–0.2 g of acetic acid but decreased significantly
at a higher dose of 0.3 g. This behavior is a result of the deli-
cate role played by acetic acid in stabilizing the MAA–TA com-
plex formed on the chirally modified nickel surface through hy-
drogen bonding–electrostatic interactions, which are essential
to give the hydrogenation product in high ee. However, the
excess amount of acetic acid weakens the MAA–TA complex
on the surface because of the competition between TA and
acetic acid.
7
MPa to a plateau of 94–99% at 9–11 MPa in the pressure-de-
pendence profile. The contrasting behaviors of conversion and
ee indicate that the hydrogenation rate does not affect the
Conclusions
The chiral nickel catalysts modified with tartaric acid and NaBr
were prepared from commercially available nickel powders of
different particle sizes without any pretreatment (such as acti-
vation with hydrogen) under various conditions. Their hydro-
genation and enantiodifferentiation abilities were assessed to
optimize the parameters for chiral modification and subse-
Figure 11. Effects of the hydrogen pressure on the conversion (~) and enan-
tioselectivity (*). Nickel powder: 5 mm. Modification: TA (0.5 g); NaBr (2.0 g);
temperature 373 K. Hydrogenation: additive acetic acid (0.1 g); temperature
373 K.
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 170 – 178 176

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
quent hydrogenation by using the enantiodifferentiating hy-
drogenation of methyl acetoacetate as a benchmark reaction.
Of the nickel powders examined, 5 mm and 100 nm nickel
powders (both from Aldrich) gave the best results in terms of
the enantioselectivity and conversion, which are comparable
to those obtained with the conventional chirally modified cata-
lysts prepared from the activated nickel powder. The highest
enantioselectivity (91%) was attained with quantitative conver-
sion with the 5 mm nickel powder without hydrogen pretreat-
ment under optimized conditions. The direct use of commer-
cially available nickel powders, as well as the high enantiose-
lectivity and conversion values, will promote the practical ap-
plication of the heterogeneous chiral nickel catalysts as a prom-
ising route to enantiopure compounds in industry.
chiral GLC analysis on a CP-Chirasil-Dex CB column (0.25 mmꢁ
2
5 m) operated at 363 K. The ee value was calculated from the
peak integration of the corresponding enantiomer peaks. The re-
producibility of the ee value was found to be within Æ2%.
Characterization of the catalyst
The powder XRD spectra were recorded with a Bruker D8 DISCOV-
ER high-resolution X-ray diffractometer with CuKa radiation. The
nickel crystallite size was calculated with the Scherrer equation.
SEM was performed at a beam voltage of 5 kV with a JEOL JSM-
6
700F scanning electron microscope.
XPS was performed on an ULVAC-PHI ESCA 5800 with a MgKa
source. The samples for the measurements were prepared under
air at ambient temperature and then placed into the chamber. The
XPS spectra obtained were processed with the CasaXPS software
(
2
version 2.3.1.15). The carbon (1s) line was used for calibration at
84.8 eV. For the XPS measurement, the washed sample after chiral
Experimental Section
modification was dried under vacuum (4.0 kPa) at 323 K for 18.5 h
and stored under argon in a sealed glass ampoule. The stored dry
catalyst was used for the measurement immediately after opening
the ampoule.
Materials
All the chemicals used were available commercially and used as re-
ceived. The nickel powders of different diameters (<100 nm, sub-
micrometer, 3 mm, 5 mm, and 150 mm) were purchased from
Aldrich.
Acknowledgements
Chiral modification of metallic nickel
This work was supported by Adaptable & Seamless Technology
Transfer Program through Target-driven R&D by Japan Science
and Technology Agency (grant no. AS2311578D).
Nickel powders with and without hydrogen activation were used
for comparison purpose. For hydrogen activation before chiral
modification, nickel powder (0.5 g) was exposed to a stream of hy-
drogen at 473–573 K for 0.5 h. Chiral modifications were performed
at 373 K by immersing the activated and nonactivated nickel pow-
Keywords: asymmetric catalysis · heterogeneous catalysis ·
hydrogen activation · nickel · tartaric acid
3
ders in an aqueous solution (50 cm ) containing varying amounts
of (R,R)-TA and NaBr, pH of which was preadjusted to 3.2 with an
aqueous NaOH solution (1m), unless noted otherwise. After 1 h of
immersion, the modification solution was removed through de-
cantation and the catalyst was washed successively once with de-
nd
[1] D. Voet, J. G. Voet, Biochemistry, 2 ed., Wiley, New York, 1995.
[2] H. U. Blaser, E. Schmidt, Asymmetric Catalysis on Industrial Scale, Wiley-
VCH, Weinheim, 2004.
[3] T. Mallat, E. Orglmeister, A. Baiker, Chem. Rev. 2007, 107, 4863–4890,
and references therein.
4] H. U. Blaser, M. Studer, Acc. Chem. Res. 2007, 40, 1348–1356, and refer-
ences therein.
3
3
ionized water (10 cm ), twice with methanol (25 cm ), and twice
3
with THF (10 cm ).
[
[
[
[
5] M. Bartꢂk, Chem. Rev. 2010, 110, 1663–1705, and references therein.
6] Y. Izumi, Adv. Catal. 1983, 32, 215–271, and references therein.
7] A. Tai, T. Sugimura in Chiral Catalyst Immobilization and Recycling (Eds.:
D. E. De Vos, I. F. J. Vankelecom, P. A. Jacobs), Wiley-VCH, 2000, pp. 173–
Enantiodifferentiating hydrogenation with the chirally modi-
fied nickel catalyst
A typical method for the hydrogenation reaction used was as fol-
lows. The modified catalyst was added to a mixture of MAA (5 g),
2
09, and references therein.
8] D. Y. Murzin, P. Mꢃki-Arvela, E. Toukoniitty, T. Salmi, Catal. Rev. 2005, 47,
75–256, and references therein.
[
3
acetic acid of varying amounts, and THF (10 cm ) placed in an au-
1
toclave equipped with a magnetically coupled mechanical stirrer.
The hydrogenation was performed for 20 h at 373 K and at a vary-
ing initial pressure of hydrogen. The hydrogenation product, a mix-
ture of methyl (R)-3-hydroxybutyrate and methyl (S)-3-hydroxybu-
tyrate, was isolated from the reaction mixture through distillation.
The conversion was determined from gas–liquid chromatography
[9] T. Osawa, T. Harada, O. Takayasu, Curr. Org. Chem. 2006, 10, 1513–1531,
and references therein.
[10] K. Szçri, M. Sutyinszki, K. Felfçldi, M. Bartꢂk, Appl. Catal. A 2002, 237,
2
75–280.
11] C. Exner, A. Pfaltz, M. Studer, H. U. Blaser, Adv. Synth. Catal. 2003, 345,
253–1260.
[
1
[
[
12] S. Diezi, T. Mallat, A. Szabo, A. Baiker, J. Catal. 2004, 228, 162–173.
13] K. Balꢄzsik, I. Bucsi, S. Cserꢅnyi, G. Szçllçsi, M. Bartꢂk, J. Mol. Catal. A
(
GLC) analysis with
a packed column of 5% Thermon 100 on Chromosorb W (2 m) at
63 K or with a GL Sciences Model GC4000 equipped with a CP-
a Hitachi Model 263-30 equipped with
2
008, 280, 87–95.
3
[
14] C. Mondelli, C. Bucher, A. Baiker, R. Gilmour, J. Mol. Catal. A 2010, 327,
87–91.
Chirasil-Dex CB capillary column (0.25 mmꢁ25 m) at 363 K.
[
[
15] B. Li, X. Li, H. Wang, P. Wu, J. Mol. Catal. A 2011, 345, 81–89.
16] Z. Chen, Z. Guan, M. Li, Q. Yang, C. Li, Angew. Chem. 2011, 123, 5015–
5019; Angew. Chem. Int. Ed. 2011, 50, 4913–4917.
Determination of the enantioselectivity
[
17] F. Hoxha, B. Schimmoeller, Z. Cakl, A. Urakawa, T. Mallat, S. E. Pratsinis,
A. Baiker, J. Catal. 2010, 271, 115–124.
The enantioselectivity was determined from chiral GLC analysis
after the acetylation of the reaction product with acetyl chloride
and pyridine. A portion of the acetylated sample was subjected to
[18] K. Balꢄzsik, K. Szçri, G. Szçllçsi, M. Bartꢂk, Catal. Commun. 2011, 12,
1410–1414.
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 170 – 178 177

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[19] Y. Nitta, T. Kubota, Y. Okamoto, Bull. Chem. Soc. Jpn. 2001, 74, 2161–
[37] M. A. Keane, Catal. Lett. 1993, 19, 197–209.
2
165.
[38] T. Harada, Y. Imachi, A. Tai, Y. Izumi, Metal-Support and Metal-additive Ef-
fects in Catalysis, Lyon, 1982, pp. 377–384.
[
20] W. R. Huck, T. Mallat, A. Baiker, New J. Chem. 2002, 26, 6–8.
[
21] G. Szçllçsi, T. Varga, K. Felfçldi, S. Cserꢅnyi, M. Bartꢂk, Catal. Commun.
[39] T. Osawa, A. Ozawa, T. Harada, O. Takayasu, J. Mol. Catal. A 2000, 154,
271–275.
2008, 9, 421–424.
[22] T. Sugimura, T. Uchida, J. Watanabe, Y. Kubota, T. Okamoto, T. Misaki, T.
Okuyama, J. Catal. 2009, 262, 57–64.
[40] T. Osawa, T. Kizawa, S. Ikeda, T. Kitamura, Y. Inoue, V. Borovkov, Chem.
Lett. 2013, 42, 1225–1226.
[
[
23] S. Nakagawa, T. Sugimura, A. Tai, Chem. Lett. 1997, 26, 859–860.
24] T. Osawa, T. Harada, A. Tai, Abstracts for the 7th International Symposium
on Relations Between Homogeneous and Heterogeneous Catalysis, Tokyo,
[41] L. J. Bostelaar, W. M. H. Sachtler, J. Mol. Catal. 1984, 27, 387–395.
[42] T. Harada, A. Tai, M. Yamamoto, H. Ozaki, Y. Izumi, Proc. 7th Intl. Congr.
Catal. Tokyo, 1980, pp. 364–375.
1992, pp. 230–231.
[43] M. A. Keane, Langmuir 1997, 13, 41–50.
[
25] T. Harada, M. Yamamoto, S. Onaka, M. Imaida, H. Ozaki, A. Tai, Y. Izumi,
[44] P. Kukula, L. Cerveny, J. Mol. Catal. A 2002, 185, 195–202.
[45] H. Chen, R. Li, H. Wang, L. Yin, F. Wang, J. Ma, Chem. Lett. 2006, 35,
910–911.
[46] H. Ozaki, Bull. Chem. Soc. Jpn. 1978, 51, 257–259.
[47] T. Osawa, T. Harada, A. Tai, J. Catal. 1990, 121, 7–17.
[48] I. Yasumori, M. Yokozeki, Y. Inoue, Faraday Discuss. Chem. Soc. 1981, 72,
385–396.
[49] H. Ozaki, A. Tai, S. Kobatake, H. Watanabe, Y. Izumi, Bull. Chem. Soc. Jpn.
1978, 51, 3559–3563.
[50] Y. Nitta, T. Imanama, S. Teranishi, J. Catal. 1983, 80, 31–39.
[51] T. Osawa, M. Onogi, I-Yin. S. Lee, T. Harada, React. Kinet. Mech. Catal.
2011, 103, 443–449.
Bull. Chem. Soc. Jpn. 1981, 54, 2323–2329.
[26] J. Masson, P. Cividino, J. Court, J. Mol. Catal. A 1996, 111, 289–295.
[27] P. Kukula, L. Cerveny, Appl. Catal. A 2001, 210, 237–246.
[28] H. Chen, R. Li, H. Wang, J. Liu, F. Wang, J. Ma, J. Mol. Catal. A 2007, 269,
125–132.
[
[
[
[
[
29] T. Harada, S. Onaka, A. Tai, Y. Izumi, Chem. Lett. 1977, 6, 1131–1132.
30] L. Fu, H. H. Kung, W. M. H. Sachtler, J. Mol. Catal. 1987, 42, 29–36.
31] M. Keane, Can. J. Chem. 1994, 72, 372–381.
32] D. Jo, J. S. Lee, K. H. Lee, J. Mol. Catal. A 2004, 222, 199–205.
33] H. Brunner, M. Muschiol, T. Wishert, Tetrahedron: Asymmetry 1990, 1,
1
59–162.
34] T. Osawa, S. Sakai, T. Harada, O. Takayasu, Chem. Lett. 2001, 30, 392–
93.
35] T. Osawa, I. S. Lee, S. Ikeda, T. Kitamura, Y. Inoue, V. Borovkov, Appl.
Catal. A 2012, 445–446, 269–273.
[
[
[
[52] T. Osawa, M. Ando, S. Sakai, T. Harada, O. Takayasu, Catal. Lett. 2005,
105, 41–45.
3
36] T. Osawa, E. Mieno, T. Harada, O. Takayasu, J. Mol. Catal. A 2003, 200,
Received: August 28, 2013
315–321.
Published online on October 23, 2013
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 170 – 178 178