Heterogeneous enantioselective hydrogenation: PH dependence and interplay between catalytic efficacy and surface composition

Article abstract of DOI:10.1246/cl.130553

The performance of a catalytic system consisting of metallic Ni powder, tartaric acid (TA), and NaBr in the enantioselective hydrogenation of methyl acetoacetate was strongly influenced by the pH of TA solution upon chiral modification, which is attributable to the pH-induced change in the surface composition of Ni catalyst as unambiguously confirmed by X-ray photoelectron spectroscopy for the first time.

Full text of DOI:10.1246/cl.130553

doi:10.1246/cl.130553
Published on the web July 3, 2013
1225
Heterogeneous Enantioselective Hydrogenation: pH Dependence and Interplay
between Catalytic Efficacy and Surface Composition
1
2
2
2
3
2,3
Tsutomu Osawa,* Tomoko Kizawa, Shinji Ikeda, Takayuki Kitamura, Yoshihisa Inoue, and Victor Borovkov*
1
Graduate School of Science and Engineering for Research, University of Toyama, Gofuku, Toyama 930-8555
2
Metek Kitamura Co., Ltd., 1 Warada-cho, Kamitoba, Minami-ku, Kyoto 601-8133
3
Department of Applied Chemistry, Osaka University, 2-1 Yamada-oka, Suita, Osaka 565-0871
(
Received June 14, 2013; CL-130553; E-mail: osawa@sci.u-toyama.ac.jp, victrb@chem.eng.osaka-u.ac.jp)
The performance of a catalytic system consisting of metallic
Ni powder, tartaric acid (TA), and NaBr in the enantioselective
hydrogenation of methyl acetoacetate was strongly influenced
by the pH of TA solution upon chiral modification, which is
attributable to the pH-induced change in the surface composition
of Ni catalyst as unambiguously confirmed by X-ray photo-
electron spectroscopy for the first time.
Scheme 1. Enantiodifferentiating hydrogenation of MAA with
the TA-modified Ni catalyst.
Table 1. The pH effects on the enantiodifferentiating hydro-
genation of MAA with the TA-modified Ni catalyst
Optically active compounds are one of the most fundamen-
tal elements of all living organisms, also playing a key role as in
the traditional fields of pharmaceuticals, agrochemicals, perfum-
ery, bioadditives, and in the newly emerging technologies
related to nanoscience, supramolecular chemistry, biomimetics,
chirality, and enantioseparation. Amongst various methods for
manufacturing enantiopure molecules, the heterogeneous cata-
lytic approach is of particular importance owing to facile
handling, simple separation, easy recovery, effective reuse, time-
and cost-saving, and environmentally benign procedures. In
the case of hydrogenation reactions, catalytic systems based on
tartaric acid (TA)-modified metallic nickel are known to be
highly efficient in reducing activated prochiral ketones, such as
Modification pH
Conversion/%
ee/%
2
3
3.2
4
5
50
92
99
100
100
100
89
91
90
88
80
82
6.5
1­5
at pH >3.2. However, the ee was consistently high (88­91%) at
the modification pH ranging from 2 to 4 follow by considerable
deterioration down to 80­82% at the modification pH >4, thus
exhibiting the contrasting behavior to the conversion. Consid-
ering the combination of these two factors, the ee and
conversion, pH 3.2 was established to be the best value for this
particular catalytic system as well. Apparently, this optimization
was achieved as a result of the trade-off of two oppositely
working effects of pH: the removal of nickel oxide and other
oxidation products from the surface of catalyst, which accel-
erates the hydrogenation process, is more effective at low pHs,
while the formation of sodium tartrate, which is essential for
¢
-ketoesters and 2-alkanones, to yield the corresponding chiral
alcohols. There are several influencing factors controlling the
efficacy of this process including the pH of TA solution used for
the modification of the Ni surface.4
,6­10
Previously, it was
experimentally shown that pH 3.2 is optimal for the catalytic
performance of different types of metallic Ni in terms of
enantiomeric excess (ee).4
,8,10
Apparently, this is due to the
structure and composition of catalytic surface optimized at this
specific pH for the enantioselective hydrogenation. However, the
detailed investigation of surface properties related to the overall
catalytic outcomes (i.e., ee and conversion) has yet to be done.
In the present study, we revealed the structural and composi-
tional origins of the strong pH dependences of the ee and
conversion values by using X-ray photoelectron spectroscopy
4
obtaining high ee, is favorable at high pHs, as reported earlier.
These results corroborate the previous data obtained with
4
,8,10,12
different types of metallic nickel,
and hence, unequiv-
ocally indicating that the catalyst surface composition is a more
important factor for controlling the catalytic performance than
the Ni source itself. In order to prove this hypothesis and also to
elucidate the surface composition, the Ni powders modified with
TA at three representative pHs of 2, 3.2, and 6.5 were subjected
to the corresponding XPS analysis (see, Supporting Information
(
XPS).
As a base of metallic nickel, commercially available 5 ¯m
Ni powder was chosen because this kind of catalyst is especially
1
1,12
promising for industrial applications as shown recently.
13
After standard treatment with (R,R)-TA /NaBr solution at
different pH values, the chirally modified Ni powder was used
for the hydrogenation of methyl acetoacetate (MAA), a bench-
mark substrate, to yield a pair of enantiomeric alcohol
1
4
for experimental details). Figure 1 shows the deconvoluted
areas of nickel species of different valencies, relative abundance
of which is of particular significance in discussing the catalytic
performance since the distribution of metallic versus cationic
form of Ni can be quantitatively evaluated. Hence, the ratios
(
Scheme 1 and Supporting Information for experimental de-
1
2,14
tails).
The resulting ee and conversion values are shown in
0
2+
3+
Table 1. As one can see, the conversion value was rather low for
the modification at pH 2 but was gradually enhanced with
increasing the modification pH and reached the quantitative value
of Ni to Ni + Ni were found to be 17/83 at pH 2.0, 43/57
at pH 3.2, and 35/65 at pH 6.5. These results are in good
agreement with the obtained hydrogenation data. In particular,
Chem. Lett. 2013, 42, 1225­1226
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

1
226
on the surface also enhances the ee value up to 90­91% at
pH 3.0­3.2. Further increasing the pH up to 6.5 produces
disodium tartrate as the main form of TA (see, Supporting
1
4
Information), which possesses poor abilities to remove the
0
oxidized Ni, resulting in the reduced Ni peak in comparison to
0
that at pH 3.2. However, this amount of Ni is fairly sufficient to
maintain the quantitative conversion, but the ee value is
considerably decreased to 80­82% apparently as a result of
the inferior enatiodifferentiating ability of disodium tartrate.
Other energy regions of XP spectra were also in agreement
with the above-mentioned explanation of the surface composi-
tion of modified Ni. Thus, the corresponding signals associated
with TA, the O­C=O and H­C­O peaks of C 1s region and the
O­C and O=C peaks of O 1s region, are almost of the same
relative intensity. This indicates that the amount of TA adsorbed
on the Ni surface is essentially comparable in the whole pH
range studied, while the relative composition of free acid, nickel
tartrate, mono-, and disodium tartrate differs at each pH (see,
1
4
Supporting Information). However, the reliable quantitative
analysis is not feasible for these peaks due to possible
carbonaceous and oxygen contamination of the surface. It is
also to note that the Na 1s peak of sodium tartrate was detected
in the catalysts modified at pH 3.2 and 6.5, but was absent in
1
4
that modified at pH 2 (see, Supporting Information).
In conclusion, we have revealed unambiguously that the
optimal surface composition can be obtained by the TA
modification at pH 3.2 for the efficient hydrogenation of MAA
as a representative ¢-ketoester substrate. Hence, the quantitative
conversion and high ee value of up to 91% are simultaneously
achieved by removing the oxidized Ni species and having TA
adsorbed as monosodium salt, yet preventing the formation of
nickel tartrate.
Figure 1. Ni 2p3/2 and Ni 2p1/2 areas of XP spectra of Ni
powder modified with the TA­NaBr system at different pHs.
the calculated population of each TA form (free acid, mono-
sodium tartrate, and disodium tartrate) at different pHs reveals
that at pH 2.0 the corresponding free acid as the dominant
species of TA (see, Supporting Information).14 Therefore, on the
one hand the free acid form effectively removes the oxidized
nickel products from the surface, while on the other hand it
corrodes the metallic nickel to yield the corresponding tartrates.
References and Notes
1
2
3
4
M. Bartók, Chem. Rev. 2010, 110, 1663.
T. Mallat, E. Orglmeister, A. Baiker, Chem. Rev. 2007, 107, 4863.
H.-U. Blaser, M. Studer, Acc. Chem. Res. 2007, 40, 1348.
T. Osawa, T. Harada, O. Takayasu, Curr. Org. Chem. 2006, 10,
1513.
5
6
D. Y. Murzin, P. Mäki-Arvela, E. Toukoniitty, T. Salmi, Catal.
Rev. 2005, 47, 175.
H. Chen, R. Li, H. Wang, J. Liu, F. Wang, J. Ma, J. Mol. Catal. A:
Chem. 2007, 269, 125.
0
Subsequently, this tendency of reducing Ni area nicely accounts
for the lower hydrogenation ability and hence the conversion
reduced to 50%. Nevertheless, the enantiodifferentiating ability
is rather high owing to the adsorption of free acid and a modest
amount of monosodium tartrate on the Ni surface. While the
substantial amount of nickel tartrate is obviously formed under
these conditions as evidenced by the saturated green color of
modification solution and the corresponding XPS analysis, the
major contributors to the enhanced ee value are apparently the
free acid form and monosodium salt of TA as was previously
established in the case of Raney Ni10 and also observed in the
hydrogenation experiments at the concentrated (1 M) TA con-
7
8
P. Kukula, L. Červený, Appl. Catal., A 2002, 223, 43.
T. Osawa, A. Ozawa, T. Harada, O. Takayasu, J. Mol. Catal. A:
Chem. 2000, 154, 271.
9
1
M. A. Keane, Catal. Lett. 1993, 19, 197.
0 T. Harada, M. Yamamoto, S. Onaka, M. Imaida, H. Ozaki, A. Tai,
Y. Izumi, Bull. Chem. Soc. Jpn. 1981, 54, 2323.
11 T. Osawa, T. Kizawa, I.-Y. S. Lee, S. Ikeda, T. Kitamura, Y. Inoue,
V. Borovkov, Catal. Commun. 2011, 15, 15.
12 T. Osawa, I.-Y. S. Lee, S. Ikeda, T. Kitamura, Y. Inoue, V.
Borovkov, Appl. Catal., A 2012, 445­446, 269.
13 Although the opposite enantiomer of TA has not been tested for
this particular catalytic system, previously it was shown by using
Raney Ni as a base for metallic Ni that (S,S)-TA yields the
antipodal alcohol as the major product, see: M. Nakahata, M.
Imaida, H. Ozaki, T. Harada, A. Tai, Bull. Chem. Soc. Jpn. 1982,
1
5
ditions. At pH 3.2 the less corrosive monosodium tartrate
1
4
becomes the main form of TA (see, Supporting Information),
which in combination with a portion of free acid can effectively
remove the oxidized nickel species from the surface. This leads
to the considerable increase of the Ni0 contribution and
consequently the conversion value up to 99%. The superior
enantiodifferentiating ability of monosodium tartrate adsorbed
55, 2186.
14 Supporting Information is available electronically on the CSJ-
Journal Web site, http://www.csj.jp/journals/chem-lett/index.html.
15 T. Osawa, T. Kizawa, V. Borovkov, unpublished data.
Chem. Lett. 2013, 42, 1225­1226
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/