Insight into the mechanism of hydrogenation of amino acids to amino alcohols catalyzed by a heterogeneous MoOx-modified Rh catalyst

Article abstract of DOI:10.1002/chem.201405769

Hydrogenation of amino acids to amino alcohols is a promising utilization of natural amino acids. We found that MoO_x-modified Rh/SiO₂ (Rh-MoO_x/SiO₂) is an efficient heterogeneous catalyst for the reaction at low temperature (323 K) and the addition of a small amount of MoO_x drastically increases the activity and selectivity. Here, we report the catalytic potential of Rh-MoO_x/SiO₂ and the results of kinetic and spectroscopic studies to elucidate the reaction mechanism of Rh-MoO_x/SiO₂ catalyzed hydrogenation of amino acids to amino alcohols. Rh-MoO_x/SiO₂ is superior to previously reported catalysts in terms of activity and substrate scope. This reaction proceeds by direct formation of an aldehyde intermediate from the carboxylic acid moiety, which is different from the reported reaction mechanism. This mechanism can be attributed to the reactive hydride species and substrate adsorption caused by MoO_x modification of Rh metal, which results in high activity, selectivity, and enantioselectivity.

Full text of DOI:10.1002/chem.201405769

DOI: 10.1002/chem.201405769
Full Paper
&
Heterogeneous Catalysis
Insight into the Mechanism of Hydrogenation of Amino Acids to
Amino Alcohols Catalyzed by a Heterogeneous MoO -Modified Rh
x
Catalyst
Masazumi Tamura, Riku Tamura, Yasuyuki Takeda, Yoshinao Nakagawa, and
[a]
Keiichi Tomishige*
Abstract: Hydrogenation of amino acids to amino alcohols
is a promising utilization of natural amino acids. We found
that MoO -modified Rh/SiO (Rh–MoO /SiO ) is an efficient
amino acids to amino alcohols. Rh–MoO /SiO is superior to
x 2
previously reported catalysts in terms of activity and sub-
strate scope. This reaction proceeds by direct formation of
an aldehyde intermediate from the carboxylic acid moiety,
which is different from the reported reaction mechanism.
This mechanism can be attributed to the reactive hydride
x
2
x
2
heterogeneous catalyst for the reaction at low temperature
323 K) and the addition of a small amount of MoO drasti-
(
x
cally increases the activity and selectivity. Here, we report
the catalytic potential of Rh–MoO /SiO and the results of ki-
species and substrate adsorption caused by MoO modifica-
x
2
x
netic and spectroscopic studies to elucidate the reaction
tion of Rh metal, which results in high activity, selectivity,
and enantioselectivity.
mechanism of Rh–MoO /SiO2 catalyzed hydrogenation of
x
Introduction
the basicity, although the selection of a suitable acid is of
great importance. In contrast, it is known that hydrogenation
of carboxylic acids is more difficult than hydrogenation of car-
bonyl compounds such as aldehydes, ketones, or esters; this
can be explained by the low electrophilicity of the carbonyl
Amino acids are important organic compounds because they
are essential intermediates or building blocks in vivo and are
also utilized as various chemicals for animal-feed additives, cat-
alysts, flavor enhancers, and medicines. Amino acids can be
readily prepared by fermentation, enzymatic transformation,
[6]
carbon atom. Ru, Ir, and Rh complexes have been reported
[7]
to be effective for the hydrogenation of carboxylic acids.
[
1]
chemical transformation, or extraction. Recently, synthesis of
However, serious drawbacks of the catalyst systems are severe
[
2]
amino acids from green biomass (grass or leaf) or microor-
reaction conditions, such as high temperature (>373 K) or
[
3]
ganisms (microalgae or cyanobacteria) has attracted much at-
tention due to their high potential as renewable biomass re-
sources. Therefore, amino acids can be regarded as important
key chemicals for biorefinery processes, thus the development
of methods for the transformation of amino acids will become
high H pressure (>10 MPa). For heterogeneous catalysis vari-
2
ous metal species (including Ru, Rh, Cu, Pt, Pd, Ni, Au, and Re)
[8]
have been used for hydrogenation of carboxylic acids.
Among these metal species, supported Ru catalysts have been
used for the hydrogenation of various carboxylic acids (e.g.,
alkyl carboxylic acids, lactic acid, and amino acids) and, gener-
ally, the activity of Ru catalysts is the highest among simple
[
4]
more significant.
Hydrogenation of amino acids to amino alcohols is one of
the most promising transformations from the viewpoint that
amino alcohols are important components in many synthetic
and natural products, for example, as intermediates for the
synthesis of pharmaceutical agents, insecticidal reagents, and
[9]
supported metal catalysts. However, to obtain high yields (ꢀ
90%), supported Ru catalysts require high temperatures (ꢀ
[9l,10]
373 K) because of their insufficient activity.
Therefore, the
development of new catalyst systems to surpass the Ru cata-
lysts is required. There are several reports on the mechanism
of hydrogenation of carboxylic acids, investigated by DFT cal-
[
5]
chiral auxiliaries. However, hydrogenation of amino acids has
two crucial problems to overcome: 1) hydrogenation of the
carboxylic acid and 2) deactivation by the basic amino group.
The addition of acids can avoid the unfavorable influence of
[9e–f,11]
culations, kinetic studies, or spectroscopic methods,
and
the reaction pathway by generation of an acyl intermediate
from the carboxylic acid moiety has been proposed for various
metal catalysts. However, the reaction mechanism for hydroge-
nation of carboxylic acids is quite complex and various inter-
mediates can be expected, which means that the reaction
pathway can be easily changed by the catalyst systems. There-
fore, clarification of the reaction mechanism of the new cata-
lyst system is essential for improvement of the catalyst, which
[
a] Dr. M. Tamura, R. Tamura, Y. Takeda, Dr. Y. Nakagawa, Prof. K. Tomishige
Graduate School of Engineering, Tohoku University
Aoba 6-6-07, Aramaki, Aoba-ku
Sendai 980-8579 (Japan)
E-mail: tomi@erec.che.tohoku.ac.jp
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201405769.
Chem. Eur. J. 2014, 20, 1 – 12
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
leads to a clue for the new design strategy for the carboxylic
acid hydrogenation catalyst.
species of the catalyst or autoclave reactor into the solution.
Reactions in the presence or absence of additives were per-
formed with Rh–MoO /SiO (Mo/Rh=1:4) to assess the reactivi-
To date, we have reported that the reducible metal oxide
x
2
(
M’O ) modified noble metal (M) catalysts (M–M’O catalysts)
ty and final pH (Table 1). First, the reaction without any addi-
tive was performed to confirm whether an additive was neces-
x
x
are effective for hydrogenolysis and hydrogenation reactions,
for example the selective hydrogenolysis of polyols and cyclic
[
12]
[13]
ethers
and hydrogenation of a,b-unsaturated aldehydes
[
14]
and so on. Other research groups have also reported M–
x 2
Table 1. Hydrogenation of l-alanine to l-alaninol over Rh–MoO /SiO
[
a]
with various additives.
M’O catalysts for various hydrogenolysis and hydrogenation
x
[
15]
reactions.
The performance of M–M’Ox catalysts is much
[b]
[c]
[d]
Entry Additive
Catalyst
[mg]
t
Conv.
Sel.
[%]
ee
[%]
pH
better than monometallic noble-metal catalysts in terms of ac-
[h] [%]
tivity and selectivity, which implies that M–M’O catalysts have
x
1
2
3
4
5
6
7
8
9
1
none
none
none
H-ZSM-5
H-ZSM-5
H-Mordenite
H-Beta
H-USY
50
200
200
50
200
50
50
50
50
50
4
4
15.4
29.9
80.0
62.3
63.6
79.3
75.2
81.3
78.9
76.5
88
87.2
80.3
80.1 9.5
84.6
87.4 9.5
97.1
95.2
88.9
–
–
great potential to replace conventional noble-metal catalysts
in the future. Recently, we reported that a MoO -modified Rh
24 30.4
15.4
24 29.6
x
4
–
catalyst (Rh–MoO ) is an effective heterogeneous catalyst for
x
the hydrogenation of amino acids to amino alcohols and dem-
4
4
4
4
4
15.1
15.6
16.0
1.3
–
–
–
–
[
16]
onstrated the preliminary results. The addition of MoO to
x
Rh/SiO2 enhanced the reaction rate (54-fold rate enhance-
ment); the selectivity and enantiomeric excess (ee) were also
drastically improved. Herein, we report our continuous experi-
CeO
PO
2
>99
98.3 1.6
0
H
3
4
81.4
87.1
[
4
(
a] Reaction conditions: Rh–MoO
.5 mmol), additive (solid acids=50 mg, H
20.0 g), H
2
x
/SiO
2
(Mo/Rh=1:4), l-alanine (0.4 g,
PO =0.57 g; 5.8 mmol), H
(8 MPa), 353 K. [b] Conversion. [c] Selectivity. [d] pH was mea-
sured after the reaction.
mental research to elucidate the potential of Rh–MoO /SiO2
x
3
4
2
O
catalysis and the reaction mechanism of hydrogenation of
amino acids over Rh–MoO /SiO . We have studied the effect of
x
2
MoO on the catalytic activity for the hydrogenation of amino
x
acids by determining the catalyst structure, kinetic parameters,
and adsorption state of the substrate. The intermediate species
was ascertained by FTIR studies and the rate-determining step
was clarified by kinetic studies, FTIR studies, and substrate re-
activity trends.
sary for the Rh–MoO /SiO -catalyzed hydrogenation reaction
x
2
(Table 1, entries 1–3). The reaction provided l-alaninol with
80% selectivity at 15.4% conversion after 4 h by using Rh–
MoO /SiO (50 mg) as the catalyst (Table 1, entry 1). However,
x
2
the reactions were limited to about 30% conversion despite
a larger catalyst amount and longer reaction time (Table 1, en-
tries 2 and 3). The pH of the resultant solution was 9.5. This
result indicates that the basic conditions, caused by the amino
alcohol product, inhibit the reaction progress. Zeolites, such as
H-ZSM-5, H-mordenite, H-beta, and H-USY were also applied to
the reaction as solid acid additives and resulted in almost the
same outcome as in the case of no additive (Table 1, entries 1,
Results and Discussion
Catalyst performance
The amino acid moiety is a well-known zwitterion and the car-
boxy group is in the carboxylate anion state under neutral
conditions (Scheme 1).
4, and 6–8). Moreover, although a longer reaction time and
a larger catalyst amount were adopted for H-ZSM-5, high con-
version was not obtained (Table 1, entry 5), rather the resulting
conversion was similar to the case of no additive (ꢁ30%).
These results suggest that solid acid additives are not effective
Scheme 1. Ionization state of an amino acid.
for the reaction. On the other hand, CeO was tested as a solid
2
base additive in the reaction (Table 1, entry 9). The conversion
was very low (1.3%), although the selectivity slightly increased
(88%). It was ascertained that a base additive is unfavorable
for the reaction. Finally, H PO , the most commonly used acid
The state of the carboxylate anion (especially under basic
conditions) is unfavorable for the hydrogenation of carboxy
groups owing to decreased electrophilicity of the carbonyl
carbon atom. Moreover, in hydrogenation of amino acids the
product is an amino alcohol, a basic compound, which increas-
es the pH of the solution (the equilibrium is shifted to the left
side). For these reasons, the reactivity of the carboxy group de-
creases as the reaction proceeds. Therefore, addition of acid is
of great importance to maintain the protonated state of the
carboxy group and avoid the increase of pH. From academic
and industrial viewpoints, the effect of acids on the reaction is
quite intriguing because acids often dissolve the active metal
3
4
[9]
additive for hydrogenation of amino acids, was utilized as
a liquid inorganic acid (Table 1, entry 10). High conversion and
selectivity for l-alaninol were obtained and the conversion
(81.4%) is much higher than the results above. Considering
the pH of the reaction (pH 1.6), acidic conditions are quite im-
portant for the reaction progress, which highlights the signifi-
cance of the state of the carboxy group in amino acids. There-
fore, a liquid acid, such as H PO , that can maintain low pH will
3
4
be an effective additive for the reaction.
&
&
Chem. Eur. J. 2014, 20, 1 – 12
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Table 2. The amount of dissolved Rh and Mo metal in the reusability test
[
a]
of the Rh–MoO
x
/SiO
2
catalyst.
Entry Acid Used
time
Dissolved amount of Rh Dissolved amount of Mo
(wt%)
(wt%)
1
2
3
4
5
6
H
H
H
H
H
H
3
3
3
2
2
2
PO
PO
PO
4
4
4
1
2
3
1
2
3
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
7.2
0.4
0.2
1.1
0.3
0.3
SO
SO
SO
4
4
4
[
(
a] Reaction conditions: Rh–MoO
2.0 g, 22.5 mmol), H
x
/SiO
2
(Mo/Rh=1:8; 250 mg), l-alanine
2
2
O (19.3 g), acid (40 mmol), H (8 MPa), 353 K, 4 h.
Figure 1. The effect of the support in Rh–MoO
conditions: Rh–MoO /SiO (Mo/Rh=1:8; 50 mg), l-alanine (0.4 g, 4.5 mmol),
PO (0.58 g, 5 mmol), H O (19.3 g), H (8 MPa), 353 K, 4 h.
x
supported catalysts. Reaction
was estimated to be 7.2 and 1.1% for H PO and H SO , re-
3 4 2 4
x
2
H
3
4
2
2
spectively (Table 2, entries 1 and 4). The large dissolved
amount of Mo in the presence of H PO may be explained by
3
4
The catalyst support will also influence the state of the
the formation of phosphomolybdic acid. Moreover, the reusa-
bility of the catalyst was also tested with both acids (Figure 2).
The conversion with H PO decreased as the usage increased,
amino acid by the acid/base properties of the support. The
effect of the support for Rh–MoO catalysts was investigated in
x
3
4
the hydrogenation of l-alanine with H PO4 (Figure 1). SiO2–
whereas with H SO4 the conversion was almost unchanged,
3
2
Al O and SiO2 provided higher conversion than other sup-
which could be due to the difference in the leached amount
of Mo between acids H PO4 and H SO . Therefore, H SO is
2
3
ports, whereas the selectivity was almost the same for all sup-
ports, which suggests that nonbasic supports are effective for
the reaction. This is in good agreement with the above result
that a base additive is unfavorable for the reaction (Table 1,
3
2
4
2
4
a more preferable additive than H PO from a practical view-
3
4
point.
The effect of the amount of Mo on the reaction rate was in-
entry 9). A SiO support was used for the following studies.
vestigated with H PO or H SO in the hydrogenation of l-ala-
2
3
4
2
4
For the liquid acid, we compared H PO with H SO in the
nine catalyzed by Rh–MoO /SiO (Figures S1 and S2 in the Sup-
x 2
3
4
2
4
hydrogenation of l-alanine over a Rh–MoO /SiO2 catalyst
porting Information). The catalytic activity in the presence of
H PO showed a volcano pattern with respect to the amount
x
(Figure 2 and Table 2). For the first use, the pH values before
3
4
the reaction were 1.3 and À0.7 in the presence of H PO and
of Mo; a similar activity trend was obtained in the case of
H SO . A 1:8 molar ratio of Mo/Rh in the Rh–MoO /SiO catalyst
3
4
H SO , respectively, and the pH values after the reaction were
2
4
2
4
x
2
1
.5 and À0.5, respectively. H SO provided higher activity than
afforded the highest conversion with both acids.
The effect of the amount of H SO was examined for the hy-
2
4
H PO , although the selectivities are similar (Figure 2). The
3
4
2
4
higher activity in solution with H SO may be explained by the
drogenation of l-alanine over Rh–MoO /SiO (Table 3). In the
x 2
2
4
lower pH of the solution (see below). The amount of dissolved
Rh and Mo metal was analyzed by inductively coupled plasma
atomic emission spectroscopy (ICP-AES) and the results are
shown in Table 2. The leached amount of Rh was below the
detection level for either acid but the leached amount of Mo
case of H SO (5 mmol; ꢁ22 mol% relative to l-alanine), the
2
4
conversion hardly altered with an increased reaction time from
4 to 24 h, and the pH of the resulting solution (pH 8.7) indicat-
ed that the amino alcohol product increased the pH of the so-
lution and suppressed the reaction progress (Table 3, entries 1
and 2). Although addition of H SO (11 mmol) increased the
2
4
conversion at the longer reaction time of 24 h, the reaction
2 4
Table 3. Effect of the concentration of H SO on the hydrogenation of l-
[
a]
x 2
alanine to l-alaninol over Rh–MoO /SiO .
[
b]
Entry
H
2
SO
4
[mmol]
t [h]
Conv. [%]
Sel. [%]
ee [%]
pH
1
2
3
4
5
5
4
24
4
24
4
8
12
16
35.6
47.9
40.8
94.5
47.4
79.8
99.8
99.9
91.8
91.2
90.8
90.8
88.9
88.8
87.7
87.8
99.6
98.8
99.9
3.3
8.7
1.8
8.2
0.4
0.5
0.7
0.7
11
11
20
20
20
20
99.5
[
[
[
[
c]
c]
c]
c]
5
6
7
8
>99.9
>99.9
>99.9
>99.9
Figure 2. The reusability of Rh–MoO
hydrogenation of l-alanine to l-alaninol. Reaction conditions: Rh–MoO
x
/SiO
2
with a) H
3
PO
4
and b) H
2
SO
4
in the
/SiO
[
a] Reaction conditions: Rh–MoO
x
/SiO
2
(Mo/Rh=1:8; 100 mg), l-alanine
x
2
(
2.0 g, 22.5 mmol), H O (19–20 g), H
2
2
(8 MPa), 353 K. [b] pH was measured
(
(
Mo/Rh=1:8; 250 mg), l-alanine (2.0 g, 22.5 mmol), H
40 mmol), H (8 MPa), 353 K, 4 h.
2
O (19.3 g), acid
after the reaction. [c] Data from reference [16].
2
Chem. Eur. J. 2014, 20, 1 – 12
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
was dramatically suppressed at higher conversion due to the
increased pH of the solution (pH 8.2; Table 3, entries 3 and 4).
With a sufficient amount of H SO (20 mmol) as additive
x 2
Table 4. Hydrogenation of l-alanine to l-alaninol over Rh–MoO /SiO at
various temperatures.
[a]
2
4
À1
Entry T [K] Cat. [mg] t [h] Conv. [%] Sel. [%] ee [%] TOF [h
]
(Table 3, entries 5–8), the reaction proceeded smoothly to
afford the corresponding amino alcohol in high conversion
1
2
3
4
5
6
7
8
303 100
313 50
323 50
333 20
353 20
373 20
393 20
413 10
16
16
8
17
8
4
2
2
8.7 93.9
7.9 94.1
8.1 93.0
14.3 92.0
19.2 88.9
23.6 85.7
26.9 80.6
24.2 72.3
>99.9
>99.9
>99.9
>99.9
>99.9
99.6
3.1
5.7
(
99.8%), ee (>99.9%), and selectivity (ꢁ90%) after 12 h
11.7
24.6
67.1
169
388
712
(
Table 3, entry 7). The initial pH before the reaction was pH 0.4.
The pH of the solution was maintained below pH 0.7 through-
out the reaction, which led to almost complete conversion.
After a prolonged reaction time (16 h), the selectivity and ee
were unchanged, which suggested that racemization and over-
reactions such as hydrogenolysis and hydration hardly oc-
curred. This pH dependence of the conversion was in good
98.5
97.0
[a] Reaction conditions: Rh–MoO
x
/SiO
2
(Mo/Rh=1:8), l-alanine (2.0 g,
(8 MPa).
22.5 mmol), H SO (2.0 g, 20 mmol), H
2
4
2
O (19.3 g), H
2
[
9j–k]
agreement with previously reported data.
In addition, the
activity seems to increase at lower pH (Table 3, entries 3 and 5;
Figure 2, cycle (number of times used)=1). Focusing on the re-
lationship between the pH and selectivity (Table 3, entries 1, 3,
and 5; Figure 2, cycle=1), the selectivity gradually decreased
at lower pH, which may be caused by the hydrogenolysis of
the CÀN bond of the substrate.
tivity, and substrate scope (Table 5). For the hydrogenation of
l-alanine, the TOF over Rh–MoO /SiO was more than 22-times
x
2
higher than that over Ru/C at 353 K, but the selectivity over
Rh–MoO /SiO was lower than over Ru/C (Table 5, entries 1–3).
x
2
The TOF for the hydrogenation of l-alanine was compared at
a similar level of selectivity by lowering the reaction tempera-
ture to 323 K for catalysis over Rh–MoO /SiO (Table 5, entry 4).
Another important factor in the hydrogenation of amino
acids is the reaction temperature. According to the previous
reports in which Ru/C, the most common catalyst, was used in-
creased reaction temperature tends to cause racemization of
amino acids and decreases the selectivity for the amino alco-
x
2
Despite the lower reaction temperature of 323 K, the TOF for
À1
catalysis over Rh–MoO /SiO (4.5 h ) was higher than that over
x
2
À1
Ru/C at 353 K (3.1 and 1.8 h ; Table 5, entries 1 and 2 versus
4). These results indicate that Rh–MoO /SiO was an intrinsically
[
9h–l]
hol.
Table 4 shows the results of the Rh–MoO /SiO -cata-
x 2
x
2
lyzed hydrogenation of l-alanine at different reaction tempera-
superior catalyst relative to Ru/C. Similarly, the performance of
Rh–MoO /SiO is much better than previously reported cata-
tures. The turnover frequency (TOF) was calculated as follows:
x
2
À1
TOF [h ]=(l-alaninol [mmol]/total Rh [mmol])/time [h]. As the
lysts in terms of reaction conditions (temperature and H pres-
2
reaction temperature decreased, the TOF decreased and, in
sure), yield, and TOF (see Table S1 in the Supporting Informa-
tion). Furthermore, the substrate scope was compared under
optimized reaction conditions by using various amino acids
(glycine, l-valine, and l-serine; Table 5, entries 5–13). We pre-
liminarily reported that Rh–MoO /SiO catalysts are applicable
contrast, the selectivity and ee increased. This tendency agreed
[
9h–l]
with previous reports.
To achieve high selectivity and ee,
temperatures less than 353 K were desirable for catalysis over
Rh–MoO /SiO .
x
2
x
2
The catalytic performance of Rh–MoO /SiO and Ru/C under
to various substrates at low temperature (323 K) to provide the
corresponding amino alcohols in high yield (90–94%), with
x
2
optimized conditions was compared in terms of activity, selec-
[
a]
Table 5. Comparison between the catalytic performance of Ru/C and Rh–MoO
x
/SiO
2
(Mo/Rh=1:8).
T [K]
À1
Entry
Substrate
Catalyst [mg]
Acid
t [h]
Conv. [%]
Sel. [%]
TOF [h ]
1
2
3
Ru/C
Ru/C
Rh–MoO
Rh–MoO
200
200
20
H
H
H
H
3
3
2
2
PO
PO
4
4
353
353
353
323
4
24
8
28.7
99.9
19.2
96.7
96.0
88.9
92.8
3.1
1.8
67.1
4.5
x
/SiO
/SiO
2
SO
SO
4
4
[
b]
b]
[
4
x
2
100
24
>99.9
5
6
7
Ru/C
Rh–MoO
Rh–MoO
100
100
100
H
3
H
2
H
2
PO
4
353
353
323
24
2
24
24.5
99.9
>99.9
91.5
86.4
92.3
0.9
49.5
4.5
x
/SiO
/SiO
2
SO
SO
4
4
[
x
2
8
9
1
Ru/C
Rh–MoO
Rh–MoO
100
100
500
H
3
H
2
H
2
PO
4
353
353
323
24
8
24
7.3
96.6
>99.9
99.9
89.6
92.5
0.3
12.5
0.9
x
/SiO
/SiO
2
SO
SO
4
4
b]
0
x
2
1
1
Ru/C
Rh–MoO
Rh–MoO
100
100
200
H
3
H
2
H
2
PO
4
353
353
323
24
4
24
37.1
>99.9
>99.9
80.2
87.1
92.8
1.1
25.2
2.2
12
3
x
/SiO
/SiO
2
SO
SO
4
4
[
b]
1
x
2
[a] Reaction conditions: substrate (4.5 mmol), H
3
PO
4
(3.9 g, 40 mmol; for Ru/C), H
2
SO
4
(0.5 g, 5 mmol; for Rh–MoO
x
/SiO
2
), H
2
O (19.3 g), H
2
(8 MPa). [b] Data
from reference [16].
&
&
Chem. Eur. J. 2014, 20, 1 – 12
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[
16]
high selectivity and ee. The tendency of activity and selectiv-
ity for the amino acids over both catalysts is similar to those
for l-alanine, but in the case of the hydrogenation of l-serine
to l-serinol, the TOF and selectivity over Rh–MoO /SiO were
which indicates that the amino group at the a-position of the
carboxy group in a-amino acids plays an important role in in-
creasing the reactivity of the substrates. Generally, it is well-
known that a-amino acids form intramolecular hydrogen
bonds between the carboxy and amino groups (three types of
hydrogen bond can be formed) and that the configuration
with a hydrogen bond between the hydrogen atom of the
amino group and the carbonyl oxygen atom is the most stable
x
2
exceptionally higher than over Ru/C at 353 K (Table 5, en-
tries 11–13). Considering that l-serinol is used as an intermedi-
ate for medicines and X-ray contrast agents, Rh–MoO /SiO2
[
17]
x
is practically useful. Therefore, Rh–MoO /SiO is a superior cata-
x
2
[18]
lyst relative to previously reported Ru catalysts, including Ru/C,
from the viewpoint of activity and substrate scope.
in the gas phase. Taking this into consideration, the similar
configuration—with hydrogen
a
bond between the hydrogen atom
of the ammonium group and the
carbonyl oxygen atom—will be the
most stable (Scheme 2).
Catalyst structure
We have reported that the Rh–MoO /SiO catalyst showed the
x
2
highest activity at a 1:8 Mo/Rh ratio in the hydrogenolysis of
This hydrogen bonding is known
to cause the redshift of the n˜ (C=O)
Scheme 2. Intramolecular hy-
drogen bonding of a-amino
[
12a,13a–c]
glycerol or cyclic ethers.
The structure of Rh–MoO /SiO2
x
under the reductive conditions was proposed as follows: Rh in
the metallic state with a particle size of about 3 nm; Mo par-
tially reduced (valence of Mo about +4) and highly dispersed;
band of the carboxylic acid in acids under acidic conditions.
[19]
amino acids,
which means that
the carbonyl bond is weakened by
the MoO species partially covering the Rh metal. Considering
hydrogen bonding and the electrophilicity of the carbonyl
carbon atom increases. This effect induced by hydrogen bond-
ing leads to enhancement of the reactivity of a-amino acids
relative to other carboxylic acids. On the other hand, focusing
on the difference between the activity of Rh/SiO2 and Rh–
MoO /SiO , Rh–MoO /SiO showed higher selectivity and ee
x
that the hydrogenation of amino acids was conducted under
similar reaction conditions to those used for the hydrogenoly-
sis of glycerol (aqueous solvent, acidic additive (H SO ), high
2
4
H2 pressure), the structure of Rh–MoO /SiO2 in the reaction
x
should be the same as described above. To confirm the struc-
x
2
x
2
ture of the optimized Rh–MoO /SiO2 catalyst (Mo/Rh=1:8),
than Rh/SiO2 and the TOF was about 50-times higher than
over Rh/SiO2 (Table 6, entries 1 and 2). a-, b-, and g-Amino
acids and acetic acid were also transformed to the correspond-
ing alcohols (Table 6, entries 3–10) and the similar tendency for
selectivity and activity to increase (50–80 fold) with Rh–MoOx/
SiO versus Rh/SiO was observed. These results indicate that
x
XRD and EXAFS analyses (see Figures S3–S5 in the Supporting
Information) were performed. The results showed that Rh was
in the metallic state, the particle size of Rh was about 3.0 nm,
and MoO was in the partially reduced state. The XRD and
x
EXAFS data, including the coordination number and bond dis-
tance, are consistent with previous data for Rh–MoO /SiO ,
2
2
Rh–MoO /SiO is an effective catalyst for hydrogenation of the
x
2
x
2
which suggests that the structure of Rh–MoO /SiO in the hy-
carboxy group regardless of the presence or absence of an
amino group. It should be noted that the difference in the ac-
tivity of isobutyric acid between the Rh/SiO and Rh–MoO /
x
2
drogenation of amino acids is the same as in hydrogenolysis of
glycerol. Furthermore, the electronic state of the Rh surface
was also investigated with Rh–MoO /SiO and Rh/SiO catalysts
2
x
SiO2 catalysts is significantly large (>200 fold; Table 6, en-
tries 11 and 12). In comparison with the activity of acetic acid,
the reactivity of isobutyric acid was drastically decreased over
Rh/SiO2 but was only slightly decreased over Rh–MoO /SiO
2
x
2
2
by CO adsorption on FTIR (see Figure S6 in the Supporting In-
À1
formation). The bands at n˜ =2048 and 1905 cm were ob-
served for Rh–MoO /SiO , which can be assigned to on-top
x
2
x
and bridged n˜ (CO), respectively. The band area due to bridged
(Table 6, entries 9–12). It can be easily assumed that the de-
n˜ (CO) on Rh–MoO /SiO2 is slightly smaller than on Rh/SiO2.
crease in reactivity for isobutyric acid catalyzed over Rh/SiO is
x
2
These band positions agreed well with those observed on Rh/
due to steric hindrance of the substrate. One explanation for
the reactivity of isobutyric acid catalyzed over Rh–MoO /SiO is
SiO , which indicated that the electronic states of Rh metal are
2
x
2
similar in both catalysts.
that the adsorption state of the carboxy group over Rh–MoOx/
SiO is different from that over Rh/SiO , supported by our re-
2
2
sults of the effect of substrate concentration and FTIR spec-
troscopy (see below).
Kinetic and spectroscopic studies
To clarify the effect of the MoO species on the reaction, kinetic
Next, the effect of H pressure on the reaction rate was in-
x
2
and FTIR spectroscopic studies were carried out to compare
the performance between the Rh/SiO and Rh–MoO /SiO cata-
vestigated with Rh–MoO /SiO2 and Rh/SiO2 catalysts (Fig-
x
À1 À1
ure 3a). The formation rates of l-alaninol (V [mmolg h ]) in-
2
x
2
lysts.
creased with increasing H pressure for both catalysts. The se-
2
First, to determine the effect of the substrate structure and
lectivity gradually increased (84% at 1 MPa H2 to 89% at
8 MPa H ) with Rh–MoO /SiO , although the selectivity was un-
MoO species, the reactivity (TOF) of various substrates cata-
x
2
x
2
lyzed by Rh/SiO and Rh–MoO /SiO was compared (Table 6).
changed with Rh/SiO . From the slopes of the approximately
2
x
2
2
Rh–MoO /SiO showed about five-times higher activity for a-
straight lines (Figure 3a), the reaction orders with respect to H2
pressure were calculated to be +1.1 and +0.6 for Rh–MoOx/
SiO and Rh/SiO , respectively, which indicates that hydrogen is
x
2
amino acids than other acids including b- and g-amino acids
and alkyl carboxylic acids (Table 6, entries 1, 3, 5, 7, 9, and 11),
2
2
Chem. Eur. J. 2014, 20, 1 – 12
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
[
a]
x 2
Table 6. Comparison of the reaction rate of various amino acids and alkyl carboxylic acids over Rh–MoO /SiO .
À1
Entry Substrate
Cat.
t [h] Conv. [%] Product and selectivity [%]
ee [%] TOF [h ]
[mg]
1
2
Rh–MoO
x
/SiO
2
50
2
78.0
88.4
78.8
11.6
20.6
0.0 >99.9
80.3
1.5
Rh/SiO
2
100
24 40.1
0.6 98.0
3
4
Rh–MoO
Rh/SiO
x
x
/SiO
/SiO
2
2
50
2
84.7
87.8
70.3
12.2
29.7
–
–
87.8
1.5
2
100
24 45.9
5
6
Rh–MoO
Rh/SiO
100
100
2
32.5
86.9
80.9
11.5
12.3
–
–
16.3
0.24
2
24 6.3
7
8
Rh–MoO
Rh/SiO
x
x
/SiO
/SiO
2
2
100
100
2
32.7
88.5
87.7
11.5
12.3
–
–
16.6
0.19
2
24 4.5
9
Rh–MoO
Rh/SiO
100
100
2
32.5
83.4
78.3
C
2
H
6
16.6
21.7
–
–
15.9
0.25
10
2
24 6.8
1
1
Rh–MoO
x
/SiO
2
100
100
2
25.3
63.4
56.3
36.6
43.7
–
–
10.6
0.04
1
2
Rh/SiO
2
24 1.4
[a] Reaction conditions: Rh–MoO
x
/SiO
2
(Mo/Rh=1:8) or Rh/SiO
2
, substrate (4.5 mmol), H
2
SO
4
(0.5 g, 5 mmol), H
2
O (20.0 g), H
2
(8 MPa), 353 K.
nine concentration was also investigated with both catalysts
Figure 3b). In the case of the Rh/SiO catalyst, the reaction
(
2
rate increased with increasing the concentration of l-alanine
and provided reaction order of +0.4 with respect to the l-ala-
nine concentration. In contrast, the reaction rate hardly
changed for the Rh–MoO /SiO catalyst and the reaction order
x
2
was almost zero (À0.1). The selectivities were almost the same
for both catalysts under any reaction conditions. These results
suggest that l-alanine is strongly adsorbed on the surface of
Rh–MoO /SiO due to the addition of the MoO species. More-
x
2
x
Figure 3. a) Effect of hydrogen pressure (conditions: catalyst (50 mg), l-ala-
over, a similar reaction order (ꢁ0.0) was observed for the hy-
nine (2.0 g, 22.5 mmol), H
2
O (19.3 g), H
2
SO
4
(2.0 g, 20 mmol), H
2
(0.5–8 MPa),
drogenation of acetic acid over the Rh–MoO /SiO catalyst (see
x
2
353 K, 2–32 h) and b) effect of concentration of l-alanine (conditions: cata-
Figure S7 in the Supporting Information), which suggests that
the amino group of the amino acid has no influence on the
adsorption state. Taking the results of substrate scope into ac-
lyst (10–300 mg), l-alanine (0.1–2.0 g), H
2
SO
4
/l-alanine=1:1, H
2
O (20 g), H
2
(8 MPa), 353 K, 2 or 8 h) in the hydrogenation of l-alanine over Rh/SiO
2
(*)
and Rh–MoO /SiO
x
2
(Mo/Rh=1:8) [*].
count, the MoOx species of the Rh–MoO /SiO2 catalyst will
x
change the adsorption state of the carboxy group.
involved in the rate-determining step. Furthermore, these re-
sults mean that the addition of a MoO species to Rh/SiO in-
The activation energy (E ) for Rh–MoO /SiO was calculated
act
x
2
by using the results of the temperature effect (Table 4) and
x
2
creased the reaction order. According to our previous reports
was compared with that for Rh/SiO . The formation rates were
2
[
12c,h]
for hydrogenolysis of polyols over M+M’O catalysts,
a re-
measured under the conditions that gave conversion below
30%. Figure 4 shows the Arrhenius plots for catalysts Rh–
MoO /SiO and Rh/SiO (detailed data are shown in Table 4, en-
x
action order of +1 indicates that hydrogen is heterolytically
dissociated on the interface between the Rh metal and MoO
x
x
2
2
+
À
species to afford a proton (H ) and a hydride (H ) species, and
that the hydride species produced on the Rh–MoO /SiO cata-
tries 4–8 and Table S2 in the Supporting Information). The fit is
quite good in both cases, and the activation energies obtained
x
2
À1
lyst is highly reactive. Hydride attack on l-alanine will be the
rate-determining step and the enhanced reaction rate can be
attributed to the reactive hydride species. The effect of l-ala-
from the gradients are Eact =45 and 51 kJmol for Rh–MoOx/
SiO and Rh/SiO , respectively. These results indicate that addi-
2
2
tion of MoO to Rh decreases the activation energy. Consider-
x
&
&
Chem. Eur. J. 2014, 20, 1 – 12
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
MoO and n˜ (C(=O)O) of the carboxylate adspecies on MoO /
x
x
[
21]
SiO2. On the other hand, the Rh/SiO and Rh–Mo/SiO cata-
2
2
À1
lysts showed no band around n˜ =1548 cm and a new band
À1
around n˜ =1420 cm , which suggested that acetic acid is ad-
sorbed mainly on Rh metal rather than on the MoO species. In
x
À1
addition, a shoulder band around n˜ =1720 cm was also ob-
served on Rh/SiO and Rh–MoO /SiO . To estimate the position
2
x
2
of the n˜ (C=O) band of acetic acid on Rh metal of Rh/SiO and
2
Rh–Mo/SiO the difference spectra were obtained by subtract-
2
ing the spectrum of SiO from the spectra of Rh/SiO or Rh–
2
2
À1
Mo/SiO2 (Figure 5b). New bands at n˜ =1743 cm
for Rh–
À1
MoO /SiO and n˜ =1736 cm for Rh/SiO appeared and can be
x
2
2
assigned to the n˜ (C=O) band of the molecular acetic acid ad-
[9e,11a–b,22c]
species.
These bands are shifted to lower wavenum-
ber than that of CH COOH adsorbed on SiO and that for gas-
Figure 4. Arrhenius plots for Rh–MoO
x
/SiO
*) catalysts in the hydrogenation of l-alanine. Reaction conditions: catalyst
10–100 mg), l-alanine (2.0 g, 22.5 mmol), H SO (2.0 g, 20 mmol), H
19.3 g), H (8 MPa), 333–413 K.
2
(Mo/Rh=1:8) [*] and Rh/SiO
2
3
2
À1 [20c]
(
(
(
eous CH COOH ( n˜ =1788 cm ) , which indicates that the
3
2
4
2
O
carbonyl-stretching vibration is weakened by adsorption onto
the Rh/SiO and Rh–Mo/SiO catalysts. These results mean that
2
2
2
the carbonyl group of acetic acid is adsorbed on Rh metal and
À1
ing the results of the effect of H pressure and substrate con-
is activated. The band at n˜ =1743 cm for Rh–MoO /SiO is lo-
2
x
2
À1
centration, the reactive hydride species produced on Rh–
MoO /SiO will reduce the activation energy. Moreover, the ac-
cated at higher wavenumber than for Rh/SiO ( n˜ =1736 cm ),
2
which can be explained by the increase of the electron density
of the C=O moiety of acetic acid through hydrogen bonding
between the MoO species of the Rh–MoO /SiO catalyst and
x
2
tivation energy for Rh–MoO /SiO is much smaller than that of
x
2
À1 [10c]
Ru/C (Eact =88.5 kJmol ),
which is one reason for the much
x
x
2
higher activity of Rh–MoO /SiO relative to Ru/C.
hydrogen atom of the carboxylic acid.
x
2
From the above results, the adsorption state of the carboxy
group will largely influence the activity of the catalyst and se-
lectivity of the products (Table 6 and Figure 3b; Figure S7 in
the Supporting Information). To clarify the adsorption state of
the carboxy group and determine the intermediate, adsorption
To confirm the adsorption strength of acetic acid on Rh–
MoO /SiO , Rh/SiO , and SiO , desorption of acetic acid was
x
2
2
2
monitored by FTIR spectroscopy with increasing temperature
from 353 to 473 K. Figure 6a shows the spectra of acetic acid
desorption on Rh–MoO /SiO . The bands at n˜ =1747, 1423, and
x
2
À1
of CH COOH as a probe molecule for carboxylic acids on Rh–
1381 cm gradually decreased with increasing temperature.
3
À1
MoO /SiO was monitored at 353 K by FTIR, followed by intro-
Because the band at n˜ =1423 cm is characteristic for Rh–
x
2
duction of H . First, CH COOH was introduced at 353 K to SiO ,
MoO /SiO and Rh/SiO , the changes of the band areas at n˜ =
2
3
2
x
2
À1
2
À1
MoO /SiO , Rh/SiO , and Rh–MoO /SiO . The FTIR spectra of the
1423 cm for Rh–MoO /SiO and Rh/SiO and n˜ =1381 cm
x
2
2
x
2
x
2
2
acetic acid adspecies are shown in Figure 5a. The main bands
for SiO were monitored (Figure 6b). The change of the band
2
À1
at n˜ =1751 and 1379 cm were observed on SiO2 and as-
area is depicted as a function of temperature versus area% of
the corresponding band at 353 K. The strength of the band for
acetic acid over SiO decreased more readily than over Rh/SiO ,
signed to n˜ (C=O) and n˜ (CÀH) of the carboxy ester on the sur-
[
20]
face silanol of SiO , respectively. The MoO /SiO catalyst pro-
2
x
2
À1
2
2
vided new bands at n˜ =1684, 1548, and 1441 cm in addition
which suggested that acetic acid is adsorbed more strongly
to the bands due to SiO . These new bands can be assigned to
over the Rh species than on the SiO support. The strength of
2
2
n˜ (C=O) of on-top adsorption species on the Lewis acid sites of
the acetic acid band on Rh–MoO /SiO decreased more slowly
x
2
Figure 5. a) FTIR spectra of the acetic acid adspecies on SiO
SiO , and Rh–MoO /SiO (Mo/Rh=1:8) at 353 K. b) The difference spectra of
(Mo/Rh=1:8) and Rh/SiO after subtraction of the SiO spec-
2
, MoO
x
/SiO
2
, Rh/
x
Figure 6. a) Change of the spectra of the acetic acid adspecies on Rh–MoO /
2
x
2
SiO
acid on SiO
SiO
2
(Mo/Rh=1:8) at 353–473 K under He. b) Band-area changes of acetic
À1
À1
Rh–MoO
trum.
x
/SiO
2
2
2
2
(
&
; n˜ =1379 cm ), Rh/SiO
2
(*; n˜ =1423 cm ), and Rh–MoO
x
/
À1
(*; n˜ =1423 cm ) as a function of temperature.
2
Chem. Eur. J. 2014, 20, 1 – 12
www.chemeurj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
than on Rh/SiO , which indicated that acetic acid is more
ties at n˜ =1718 (Figure 7bii), 1660 (Figure 7biii), and
2
À1
strongly adsorbed on Rh–MoO /SiO . This result agreed with
1423 cm (Figure 7bi) were plotted as a function of reaction
x
2
À1
the above results for the substrate concentration effect. There-
fore, addition of MoO species to Rh/SiO increases the adsorp-
time (Figure 7c). The band intensity at n˜ =1423 cm was im-
mediately decreased by introduction of H and simultaneously
x
2
2
À1
tion strength of acetic acid on the catalyst. Taking into account
the bands at n˜ =1660 and 1718 cm increased. These results
that MoO species have no influence on the electronic state of
indicate that the acetic acid adspecies is converted to an alde-
hyde adspecies on Rh–MoO /SiO . With increasing reaction
x
Rh metal (from the results of CO adsorption monitored by FTIR
spectroscopy; see Figure S6 in the Supporting Information),
x
2
À1
time (>30 s), the band intensities at n˜ =1660 and 1718 cm
this result supports that MoO species directly interact with the
were decreased until these intensities reached zero, although
x
À1
carboxy group of the substrate (Figure 5). This may be related
to the difference in reactivity between isobutyric acid and
acetic acid (Table 6, entries 9–12).
the band at n˜ =1423 cm was not largely decreased. The pro-
duction of an ethanol adspecies on the catalyst was also ob-
served in the n˜ (CÀH) region (see Figure S9 in the Supporting
Information). In addition, the FTIR spectrum of the acetalde-
hyde adspecies on Rh–MoO /SiO was obtained to confirm the
Moreover, H was introduced to the adspecies of acetic acid
2
on Rh–MoO /SiO and the change of the difference spectra
x
2
x
2
with respect to the reaction time is shown in Figure 7a. The
difference spectra were obtained by using the spectrum of the
acetic acid adspecies (t=0, spectrum at 353 K shown in Fig-
ure 6a) as a background (the representative spectra are in Fig-
ure S8 in the Supporting Information). Introduction of H2
assignment to an aldehyde species (see Figure S5 in the Sup-
porting Information). The main adspecies was a p-bonded al-
dehyde adspecies (Figure 7bii), which supports the FTIR spec-
troscopy results for the hydrogenation of acetic acid. Taking
into consideration that the decrease of the band at n˜ =
À1
À1
À1
showed negative bands at n˜ =1745, 1423, and 1379 cm ,
1379 cm is much less than that at n˜ =1423 cm , the acetic
acid adspecies on Rh metal is mainly converted to ethanol (see
Figure S8 in the Supporting Information). Therefore, hydroge-
nation of a carboxylic acid to an alcohol proceeds by formation
of an aldehyde adspecies (Scheme 3). Because the acetic acid
which indicates that acetic acid reacted with H . At the same
2
À1
time, positive bands at n˜ =1718 and 1660 cm were observed.
According to the previous work on adsorption of aldehydes
[
22]
À1
over Rh/SiO2, these bands at n˜ =1718 and 1660 cm can be
assigned to n˜ (C=O) of p-bonded aldehyde adspecies (Fig-
1
ure 7bii) and n˜ (C=O) of donating-on-top type h aldehyde spe-
cies (Figure 7biii) on Rh–MoO /SiO . The adsorption models of
x
2
these adspecies are also shown in Figure 7b. The band intensi-
x 2
Scheme 3. Reaction pathway over Rh–MoO /SiO .
adspecies is stable at 353 K without H , first the carboxylic acid
2
is directly hydrogenated to the aldehyde species. According to
[9e–f,11]
previous reports,
hydrogenation of the carboxylic acid
proceeds by dissociation of the carboxylic acid or carboxylate
to an acetyl adspecies on the catalyst with subsequent hydro-
genation of the acetyl species. This reaction mechanism is dif-
ferent from that over Rh–MoO /SiO ; in fact, acetyl adspecies
x
2
À1
(
n˜ ꢁ1620 cm ) were not observed over Rh–MoO /SiO (Fig-
x
2
ure 7a). This route, over Rh–MoO /SiO by direct aldehyde for-
x
2
mation, may be derived from the generation of a reactive hy-
dride species on Rh–MoO /SiO , which leads to decreased acti-
x
2
vation energy for the hydrogenation of acetic acid to acetalde-
hyde. Moreover, stabilization of the carboxylic acid adspecies
by MoO may influence this unusual reaction route.
x
Finally, the reactivity of the aldehyde was compared with
that of the carboxylic acid over Rh–MoO /SiO (Mo/Rh=1:8) by
x
2
using acetaldehyde and acetic acid as model compounds
Scheme 4). The TOF for acetaldehyde was about 40-times
(
higher than for the carboxylic acid, which supports the hydro-
genation of acetic acid to acetaldehyde as the rate-determin-
ing step.
Figure 7. a) Change of difference spectra of the acetic acid adspecies on Rh–
MoO
x
/SiO
2
(Mo/Rh=1:8) at 353 K by introduction of H
2
. b) Models of the ad-
Reaction mechanism
species. c) Changes of the band intensity as a function of time: p-bonded al-
dehyde adspecies (*), donating-on-top h aldehyde adspecies (*), carboxylic
1
From the above results, the reaction mechanism over Rh–
acid adspecies (
&
).
MoO /SiO for the hydrogenation of l-alanine as a typical com-
x
2
&
&
Chem. Eur. J. 2014, 20, 1 – 12
www.chemeurj.org
8
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
tance between the a-carbon atom of the OH group and the re-
active hydrogen species will control over-hydrogenolysis of
amino alcohols to amines. Considering that MoO species are
x
[12a,13a–c]
known to have high oxophilicity,
the OH group of the
amino alcohol is preferentially adsorbed on the MoO species,
x
which distances the OH group of the amino alcohol from the
reactive hydrogen species. This leads to a decrease of over-re-
actions of amino alcohols. As for the enantioselectivity, a de-
crease of the ee is generally caused by dehydrogenation of the
amino group to an imine and hydrogenation of the imine to
an amino group. Adsorption of an amino (or ammonium)
group on the catalyst surface will be essential. Taking into con-
sideration that the configuration of l-alaninol was unchanged
over Rh/SiO and Rh–MoO /SiO (Table 2), the configuration of
Scheme 4. Comparison between the reactivity of acetic acid and acetalde-
hyde.
2
x
2
the substrate will be changed during the hydrogenation of l-
alanine. The Rh–MoO /SiO2 catalyst controls the adsorption
x
state of the amino acid by hydrogen bonding between the
carboxylic acid group of l-alanine and the MoO species, which
x
leads to decreased contact of the ammonium group with the
catalyst surface. This adsorption control by Rh–MoO /SiO re-
x
2
sults in suppression of the configurational change of the sub-
strate.
Conclusion
MoO -modified Rh catalyst (Rh–MoO ) is an effective heteroge-
x
x
neous catalyst for the hydrogenation of amino acids to amino
alcohols. From detailed studies of the performance of Rh–
MoO /SiO , we found that acidic reaction conditions are essen-
x
2
tial for the reaction progress and H SO is the preferred acid
2
4
(
selected among various solid acids and inorganic liquid acids)
from the viewpoints of activity and reusability. The Rh–MoOx/
SiO catalyst is efficient, not only for amino acids, but also for
Scheme 5. Proposed reaction mechanism for hydrogenation of l-alanine to
2
l-alaninol over a Rh–MoO
x
/SiO
2
catalyst.
alkyl carboxylic acids, which means that this catalyst is effective
for the hydrogenation of various carboxylic acids. Comparison
of the catalytic performance between Rh–MoO /SiO and Ru/C,
x
2
pound is proposed. The reaction proceeds as follows
Scheme 5): I) Adsorption of l-alanine onto the catalyst as a car-
the most common heterogeneous catalyst, demonstrated that
Rh–MoO /SiO is inherently superior to Ru/C in terms of the
(
x
2
boxylic acid species and stabilization of the adsorption state
catalytic activity and substrate scope. Based on the kinetics
and various spectroscopic studies, such as FTIR, XRD, EXAFS,
and CO adsorption, the structure of Rh–MoO /SiO and the re-
by the MoO species through hydrogen bonding between the
x
MoO species and hydrogen atom of the carboxy group. II) He-
x
x
2
terolytic dissociation of H on Rh–MoO /SiO to produce hy-
action mechanism over the Rh–MoO /SiO catalyst were inves-
x 2
2
+
x
2
À
dride (H ) and proton (H ) species. III) Hydride attack on the
carboxylic acid adspecies to afford the aldehyde adspecies. IV)
tigated. In the structure of Rh–MoO /SiO Rh is in the metallic
x 2
state and the particle size is about 3 nm, Mo is partially re-
À
+
Heterolytic dissociation of H to H and H . V) Hydride attack
duced (the valence of Mo is about +4) and highly dispersed,
2
at the aldehyde adspecies to produce the corresponding
amino alcohol adspecies. VI) Desorption of the amino alcohol
and the MoO species partially covers the Rh metal. From a re-
x
activity comparison of the substrates, the reactivity of a-amino
acids is about five-times higher than other carboxylic acids,
which can be explained by intramolecular hydrogen bonding
between the ammonium group and carboxy group of a-amino
acids. FTIR analyses of acetic acid adsorption and introduction
of H to Rh–MoO /SiO showed that the carboxylic acid is ad-
and regeneration of the catalyst. Intramolecular hydrogen
+
bonding between H N and the carbonyl oxygen atom of the
3
carboxy group increases the reactivity of the amino acids. Ste-
p III is the rate-determining step for the hydrogenation of l-
alanine over Rh–MoO /SiO . From the reaction mechanism, the
x
2
2
x
2
increased selectivity and ee after Mo modification of the cata-
sorbed on Rh–MoO /SiO as a molecular carboxylic acid, and
x 2
lyst can be explained by the adsorption state of the amino
hydrogenation proceeds by direct formation of an aldehyde in-
acid or amino alcohol. As mentioned in Table 6, Rh–MoO /SiO2
termediate from reaction of the carboxylic acid with H , which
x
2
suppressed the over-hydrogenolysis of amino alcohols to
amines. The adsorption state of the amino alcohol and the dis-
is different from the previously reported reaction mechanism.
Hydrogenation of the carboxylic acid to an aldehyde by the
Chem. Eur. J. 2014, 20, 1 – 12
www.chemeurj.org
9
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
hydride species is the rate-determining step. A drastic increase
pounds. Detailed HPLC conditions are described in the Supporting
Information.
of activity, selectivity, and ee were derived from the reactive
À
hydride (H ) species on the interface between the Rh metal
and MoO species, and promotion and control of the substrate
In situ FTIR analysis
x
adsorption by the MoO species, which generates a new route
x
In situ FTIR spectra were recorded with a NICOLET 6700 spectrom-
eter (Thermo Scientific) equipped with a liquid nitrogen-cooled
mercury cadmium telluride (MCT) detector (resolution=4 cm ), by
for hydrogenation of carboxylic acids. Hence Rh–MoO /SiO2
x
will provide new insight into the selective hydrogenation of
various carbonyl compounds.
À1
using an in situ IR cell with CaF windows, which was connected to
2
a conventional gas-flow system. The Rh–MoO /SiO2 sample (30–
x
5
0 mg) was pressed into a self-supporting wafer (20 mm diameter)
Experimental Section
and mounted into the IR cell. Adsorption of acetic acid and hydro-
genation of acetic acid were carried out by using the following
General
method: the catalyst was reduced at 473 K under a H [H2
2
À1
À1
À1
(
3 mLmin /He (60 mLmin )] (3:60 mLmin ) flow for 10 min. The
HPLC analysis was performed with a Shimadzu Prominence instru-
ment fitted with a Daicel Crownpak CR(+) 4.0ꢁ150 mm or Waters
Nova-Pak C-18 3.9ꢁ150 mm column. GC analysis was performed
with a Shimadzu GC-2025 instrument fitted with an Agilent Tech-
nologies CP-Sil-5 capillary column (0.25 mmꢁ50 m) with nitrogen
as the carrier gas. All chemicals for the organic reactions were
commercially available and were used without further purification.
catalyst was cooled to 353 K under a He flow. Acetic acid (3 mL)
was injected into the gas line heated at 423 K under a He flow,
which was fed to the in situ IR cell. The IR measurement was car-
À1
ried out at 353 K. After the spectrum stablized, H [H (3 mLmin /
2
2
À1
He (60 mLmin )] was introduced into the IR cell and the spectrum
change was monitored. Spectra were obtained by subtraction of
the reference spectrum of Rh–MoO /SiO measured at 353 K under
x
2
a He flow.
Catalyst preparation
Rh/SiO was prepared by impregnating SiO (Fuji Silysia Chemical
2
2
2
À1
Ltd., G-6, BET surface area=535 m g ) with an aqueous solution Acknowledgements
of RhCl ·3H O (Wako Pure Chemical Industries, Ltd.). SiO -support-
3
2
2
ed Rh–MoOx catalysts were prepared by impregnating Rh/SiO2
after the drying procedure with an aqueous solution of
This study was supported by JSPS KAKENHI, grant number
6249121.
2
[
(NH ) Mo O ·4H O] (Wako Pure Chemical Industries, Ltd.). After
4 6 7 24 2
evaporation of the solvent, the catalyst was dried at 383 K for 12 h,
then calcined in air at 773 K for 3 h. The loading amount of Rh in
Rh–MoO /SiO was 4%w/w. The molar ratio of Mo/Rh metal was
Keywords: amino acids · amino alcohols · heterogeneous
catalysis · hydrogenation · reaction mechanisms
x
2
1
3
:8, unless otherwise noted. The catalyst was reduced at 773 K for
h in a H flow, then passivated in a 10% O /N flow before the
2
2
2
[1] a) M. Ikeda, in Advances in Biochemical Engineering/Biotechnology,
Volume 79 (Eds.: R. Faurie, J. Thommel), Springer, Berlin, Heidelberg,
reaction.
2
003, pp. 1–35; b) J. Becker, C. Wittmann, Curr. Opin. Biotechnol. 2012,
2
3, 718–726; c) A. Imaizumi, C. Koseki, Y. Usuda, H. Yasueda, H. Kojima,
Typical procedure for Rh–MoO /SiO catalyzed hydrogena-
tion of amino acids to amino alcohols
x
2
K. Matsui, S.-i. Sugimoto, J. Biotechnol. 2005, 117, 111–118; d) H.-Y.
Hsiao, J. F. Walter, D. M. Anderson, B. K. Hamilton, Biotechnol. Genet. Eng.
Rev. 1988, 6, 179–219.
2] a) M. Andersen, P. Kiel, Ind. Crops Prod. 2000, 11, 129–137; b) B. Kamm,
M. Kamm, in Advances in Biochemical Engineering/Biotechnology, Volume
Activity tests were performed in a 190 mL stainless-steel autoclave
[
with an inserted glass vessel. Catalyst (50 mg), H PO4 (3.9 g,
3
4
0 mmol) or H SO (2.0 g, 20 mmol), water (19–20 g), and l-alanine
2 4
1
2
05 (Eds.: R. Ulber, D. Sell), Springer, Berlin, Heidelberg, 2007, pp. 175–
04; c) B. Kamm, C. Hille, P. Schonicke, Biofuels Bioprod. Bioref. 2010, 4,
(
0.4 g, 4.5 mmol or 2.0 g, 22.5 mmol) were put into the autoclave,
together with a stirrer bar. The reactor was sealed, then the air
content was quickly purged by flushing with H2 (ꢁ3; 1 MPa;
253–262; d) M. Mandl, Biofuels Bioprod. Bioref. 2010, 4, 268–274.
[3] a) J. M. Romero Garcꢂa, F. G. A. Fernꢃndez, J. M. F. Sevilla, Bioresour. Tech-
nol. 2012, 112, 164–170; b) M. M. Rebolloso-Fuentes, A. Navarro-Pꢄrez,
F. Garcꢂa-Camacho, J. J. Ramos-Miras, J. L. Guil-Guerrero, J. Agric. Food
Chem. 2001, 49, 2966–2972.
4] N. Kalutharage, C. S. Yi, Angew. Chem. Int. Ed. 2013, 52, 13651–13655;
Angew. Chem. 2013, 125, 13896–13900.
5] a) K. Soai, S. Niwa, Chem. Rev. 1992, 92, 833–856; b) E. J. Corey, R. K.
Bakshi, S. Shibata, J. Am. Chem. Soc. 1987, 109, 5551–5553; c) G. A.
Rogers, S. M. Parsons, D. C. Anderson, L. M. Nilsson, B. A. Bahr, W. D.
Kornreich, R. Kaufman, R. S. Jacobs, B. Kirtman, J. Med. Chem. 1989, 32,
1217–1230; d) E. J. Corey, F.-Y. Zhang, Angew. Chem. Int. Ed. 1999, 38,
1931–1934; Angew. Chem. 1999, 111, 2057–2059; e) M. Kitamura, S.
Suga M. Niwa, R. Noyori, J. Am. Chem. Soc. 1995, 117, 4832–4842;
f) D. J. Ager, I. Prakash, D. R. Schaad, Chem. Rev. 1996, 96, 835–875.
9
3
9.99%; Nippon Peroxide Co., Ltd.). The autoclave was heated to
53 K and the temperature was monitored by using a thermocou-
ple inserted in the autoclave. Once the temperature reached 353 K,
[
the H pressure was increased from 1.0 to 8.0 MPa. During the ex-
2
periment, the stirring rate was fixed at 500 rpm (magnetic stirring).
After 4 h, the gases were collected in a gas bag. The catalyst was
[
separated by filtration. The filtrate was diluted with aqueous HClO4
(
180 g, 0.16m, pH 1.1) and transferred to a vial. The reaction condi-
tions were changed appropriately in accordance with the purpose.
Details of the reaction conditions are described for each result.
À1
TOF was calculated by the following equation: TOF [h ]=(l-alani-
nol [mmol]/total Rh [mmol])/time [h]. The formation rate was cal-
À1 À1
[6] P. A. Dub, T. Ikariya, ACS Catal. 2012, 2, 1718–1741.
culated by the following equation: V [mmolg h ]=(l-alaninol
[
7] a) T. P. Brewster, A. J. M. Miller, D. M. Heinekey, K. I. Goldberg, J. Am.
Chem. Soc. 2013, 135, 16022–16025; b) F. M. A. Geilen, B. Engendahl, M.
Holscher, J. Klankermayer, W. Leitner, J. Am. Chem. Soc. 2011, 133,
[
mmol]/catalyst [g])/time [h]. The products in the gas phase were
analyzed by GC (CP-Sil-5 capillary column). The products in the
liquid phase were analyzed by HPLC (Crownpak CR(+) or Nova-Pak
14349–14358; c) F. M. A. Geilen, B. Engendahl, A. Harwardt, W. Mar-
C-18 column) with aqueous HClO (0.16m, pH 1.1) as the eluent.
Products were also identified by comparison with standard com-
quardt, J. Klankermayer, W. Leitner, Angew. Chem. Int. Ed. 2010, 49,
5510–5514; Angew. Chem. 2010, 122, 5642–5646; d) S. Wesselbaum, T.
4
&
&
Chem. Eur. J. 2014, 20, 1 – 12
www.chemeurj.org
10
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
von Stein, J. Klankermayer, W. Leitner, Angew. Chem. Int. Ed. 2012, 51,
499–7502; Angew. Chem. 2012, 124, 7617–7620; e) H. Mehdi, V.
Fabos, R. Tuba, A. Bordor, L. T. Mika, I. T. Horvath, Top. Catal. 2008, 48,
Chem. C 2012, 116, 23503–23514; n) K. Chen, M. Tamura, Z. Yuan, Y. Na-
kagawa, K. Tomishige, ChemSusChem 2013, 6, 613–621; o) Y. Nakagawa,
K. Mori, K. Chen, Y. amada, M. Tamura, K. Tomishige, Appl. Catal. A 2013,
468, 418–425; p) S. Liu, Y. Amada, M. Tamura, Y. Nakagawa, K. Tomish-
ige, Green Chem. 2014, 16, 617–626; q) M. Tamura, Y. amada, S. Liu, Z.
Yuan, Y. Nakagawa, K. Tomishige, J. Mol. Catal. A 2014, 388–389, 177–
187; r) S. Liu, Y. Amada, M. Tamura, Y. Nakagawa, K. Tomishige, Catal. Sci.
Technol. 2014, 4, 2535–2549; s) S. Liu, M. Tamura, Y. Nakagawa, K. To-
mishige, ACS Sustainable Chem. Eng. 2014, 2, 1819–1827.
7
49–54; f) L. Rosi, M. Frediani, P. Frediani, J. Organomet. Chem. 2010,
695, 1314–1322; g) A. Salvini, P. Frediani, C. Giannelli, L. Rosi, J. Organo-
met. Chem. 2005, 690, 371–382; h) M. Bianchi, G. Menchi, F. Francalanci,
F. Placenti, J. Organomet. Chem. 1980, 188, 109–119.
8] a) R. D. Cortright, M. Sanchez-Castillo, J. A. Dumesic, Appl. Catal. B 2002,
[
39, 353–359; b) L. Yu, X.-L. Du, J. Yuan, Y.-M. Liu, Y. Cao, H.-Y. He, K.-N.
Fan, ChemSusChem 2013, 6, 42–46; c) Z. He, X. Wang, J. Energy Chem.
[13] M. Tamura, K. Tokonami, Y. Nakagawa, K. Tomishige, Chem. Commun.
2013, 49, 7034–7036.
[14] a) Y. Takeda, Y. Nakagawa, K. Tomishige, Catal. Sci. Technol. 2012, 2,
2221–2223; b) Y. Amada, N. Ota, M. Tamura, Y. Nakagawa, K. Tomishige,
ChemSusChem 2014, 7, 2185–2192.
2
2
013, 22, 883–894; d) G. Onyestyꢃk, S. Harnos, D. Kallo, Catal. Commun.
011, 16, 184–188; e) U. G. Hong, H. W. Park, J. Lee, S. Hwang, J. Yi, I. K.
Song, Appl. Catal. A 2012, 415–416, 141–148; f) R. Luque, J. H. Clark, K.
Yoshida, P. L. Gai, Chem. Commun. 2009, 5305–5307; g) P. Mꢅki-Arvela,
I. L. Simakova, T. Salmi, D. T. Murzin, Chem. Rev. 2014, 114, 1909–1971;
and references therein.
[15] a) Z. Wang, B. Pholjaroen, M. Li, W. Dong, N. Li, A. Wang, X. Wang, Y.
Cong, T. Zhang, J. Energy Chem. 2014, 23, 427–434; b) M. Stein, B. Breit,
Angew. Chem. 2013, 125, 2287–2290; Angew. Chem. Int. Ed. 2013, 52,
2231–2234; c) T. Buntara, S. Noel, P. H. Phua, I. Melian-Cabrera, J. G.
de Vries, H. J. Heeres, Angew. Chem. Int. Ed. 2011, 50, 7083–7087;
Angew. Chem. 2011, 123, 7221–7225; d) M. Li, G. Li, N. Li, A. Wang, W.
Dong, X. Wang, Y. Cong, Chem. Commun. 2014, 50, 1414–1416; e) Z.
Wang, G. Li, X. Liu, Y. Huang, A. Wang, W. Chu, X. Wang, N. Li, Catal.
Commun. 2014, 43, 38–41; f) H. G. Manyar, C. Paun, R. Pilus, D. W.
Rooney, J. M. Thompson, C. Hardacre, Chem. Commun. 2010, 46, 6279–
6281; g) B. K. Ly, D. P. Minh, C. Pinel, M. Besson, B. Tapin, F. Epron, C. Es-
pecel, Top. Catal. 2012, 55, 466–473; h) D. P. Minh, M. Besson, C. Pinel,
P. Fuertes, C. Petitjean, Top. Catal. 2010, 53, 1270–1273; i) K. Kon, W.
Onodera, S. Takakusagi, K.-i. Shimizu, Catal. Sci. Technol. 2014, 4, 3705–
3712; j) S. Antos, A. S. Tilling, E. Wolters, US Patent 6310254, 2002; k) H.
Urtel, M. Roesch, A. Haunert, WO 2005077870A1, 2005; l) H. Urtel, M.
Roesch, A. Haunert, M. Schubert, WO 2005077870A1, 2005.
[
9] a) Y. K. Lugo-Josꢄ, J. R. Monnier, C. T. Williams, Appl. Catal. A 2014, 469,
410–416; b) H. Wan, A. Vitter, R. V. Chaudhari, B. Subramaniam, J. Catal.
2014, 309, 174–184; c) H. Wan, R. V. Chaudhari, B. Subramaniam, Energy
Fuels 2013, 27, 487–493; d) L. Chen, Y. Li, X. Zhang, Q. Zhang, T. Wang,
L. Ma, Appl. Catal. A 2014, 478, 117–126; e) H. Olcay, L. Xu, Y. Xu, G. W.
Huber, ChemCatChem 2010, 2, 1420–1424; f) H. Olcay, Y. Xu, G. W.
Huber, Green Chem. 2014, 16, 911–924; g) Y. K. Lugo-Josꢄ, J. R. Monnier,
C. T. Williams, Appl. Catal. A 2014, 469, 410–416; h) W. Mꢅgerlein, C.
Dreisbach, H. Hugl, M. K. Tse, M. Klawonn, S. Bhor, M. Beller, Catal. Today
2
007, 121, 140–150; i) F. T. Jere, J. E. Jackson, D. J. Miller, Ind. Eng. Chem.
Res. 2004, 43, 3297–3303; j) F. T. Jere, D. J. Miller, J. E. Jackson, Org. Lett.
003, 5, 527–530; k) P. S. Metkar, M. A. Scialdone, K. G. Moloy, Green
2
Chem. 2014, 16, 4575–4586; l) K. P. Pimparkar, D. J. Miller, J. E. Jackson,
Ind. Eng. Chem. Res. 2008, 47, 7648–7653.
[
[
10] a) A. Primo, P. Consepcion, A. Corma, Chem. Commun. 2011, 47, 3613–
615; b) B.-W. Mao, Z.-Z. Cai, M.-Y. Huang, Y.-Y. Jiang, Polym. Adv. Tech-
3
[16] M. Tamura, R. Tamura, Y. Takeda, Y. Nakagawa, K. Tomishige, Chem.
Commun. 2014, 50, 6656–6659.
nol. 2003, 14, 278–281.
11] a) S. M. Augustine, J. P. Blitz, J. Catal. 1993, 142, 312–324; b) V. Pallassa-
na, M. Neurock, J. Catal. 2002, 209, 289–305; c) M. A. N. Santiago, M. A.
Sanchz-Castillo, R. D. Cortright, J. A. Dumesic, J. Catal. 2000, 193, 16–
[17] a) E. Bieberich, T. Kawaguchi, R. K. Yu, J. Biol. Chem. 2000, 275, 177–181;
b) F. Pinkerton, G. Strobel, Proc. Natl. Acad. Sci. USA 1976, 73, 4007–
4011; c) B. Andreeben, A. Steinbꢆchel, AMB Express 2011, 1, Article no.
12; d) M. V. M. Paiocchi, US Patent 6506938, 2003.
28; d) R. Alcala, J. W. Shabaker, G. W. Huber, M. A. Sanchez-Castillo, J. A.
Dumesic, J. Phys. Chem. B 2005, 109, 2074–2085; e) L. Chen, Y. Zhu, H.
Zheng, C. Zhang, Y. Li, Appl. Catal. A 2012, 411–412, 95–104; f) L. Chen,
Y. Zhu, H. Zheng, C. Zhang, B. Zhang, Y. Li, J. Mol. Catal. A 2011, 351,
[18] A. Lesarri, E. J. Cocinero, J. C. Lopez, J. L. Alonso, Angew. Chem. Int. Ed.
2004, 43, 605–610; Angew. Chem. 2004, 116, 615–620.
[19] a) R. Linder, M. Nispel, T. Haber, K. Kleinermanns, Chem. Phys. Lett. 2005,
409, 260–264; b) J. M. Bakker, C. Plꢆtzer, I. Hꢆnig, T. Hꢅber, I. Compag-
non, G. Helden, G. Meijer, K. Kleinermanns, ChemPhysChem 2005, 6,
120–128.
2
17–227.
12] a) Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa, K. Tomishige, Appl. Catal. B
010, 94, 318–326; b) S. Koso, N. Ueda, Y. Shinmi, K. Okumura, T.
[
2
Kizuka, K. Tomishige, J. Catal. 2009, 267, 89–92; c) S. Koso, Y. Nakagawa,
K. Tomishige, J. Catal. 2011, 280, 221–229; d) S. Koso, H. Watanabe, K.
Okumura, Y. Nakagawa, K. Tomishige, Appl. Catal. B 2012, 111 – 112, 27–
[20] a) A. G. Robietti, J. C. Thompson, Spectrochim. Acta 1965, 21, 2023–
2030; b) R. P. Eischens, Science 1964, 146, 486–493; c) W. Rachmady,
M. A. Vannice, J. Catal. 2002, 207, 317–330; d) I. X. Green, W. Tang, M.
Neurock, J. T. Yates, Jr., Faraday Discuss. 2013, 162, 247–265.
[21] a) L.-F. Liao, C.-F. Lien, J.-L. Lin, Phys. Chem. Chem. Phys. 2001, 3, 3831–
3837; b) J. M. Guil, N. Homs, J. Llorca, P. R. de La Piscina, J. Phys. Chem. B
2005, 109, 10813–10819; c) D. A. Edwards, R. N. Hayward, Can. J. Chem.
1968, 46, 3443–3446.
3
1
7; e) Y. Nakagawa, Y. Shinmi, S. Koso, K. Tomishige, J. Catal. 2010, 272,
91–194; f) K. Chen, S. Koso, T. Kubota, Y. Nakagawa, K. Tomishige,
ChemCatChem 2010, 2, 547–555; g) Y. Amada, S. Koso, Y. Nakagawa, K.
Tomishige, ChemSusChem 2010, 3, 728–736; h) Y. Amada, Y. Shinmi, S.
Koso, T. Kubota, Y. Nakagawa, K. Tomishige, Appl. Catal. B 2011, 105,
1
17–127; i) S. Koso, H. Watanabe, K. Okumura, Y. Nakagawa, K. Tomish-
[22] S. Nishiyama, T. Hara, S. Tsuruya, M. Masai, J. Phys. Chem. B 1999, 103,
4431–4439.
ige, J. Phys. Chem. C 2012, 116, 3079–3090; j) Y. Nakagawa, X. Ning, Y.
Amada, K. Tomishige, Appl. Catal. A 2012, 433–434, 128–134; k) K.
Chen, K. Mori, H. Watanabe, Y. Nakagawa, K. Tomishige, J. Catal. 2012,
294, 171–183; l) Y. Amada, H. Watanabe, Y. Hirai, Y. Kajikawa, Y. Nakaga-
wa, K. Tomisige, ChemSusChem 2012, 5, 1991–1999; m) Y. Amada, H.
Watanabe, M. Tamura, Y. Nakagawa, K. Okumura, K. Tomishige, J. Phys.
Received: October 21, 2014
Published online on && &&, 0000
Chem. Eur. J. 2014, 20, 1 – 12
www.chemeurj.org
11
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
&
Heterogeneous Catalysis
M. Tamura, R. Tamura, Y. Takeda,
Y. Nakagawa, K. Tomishige*
&& – &&
Insight into the Mechanism of
Hydrogenation of Amino Acids to
Amino Alcohols Catalyzed by
a Heterogeneous MoO -Modified Rh
x
Super modifier: Rh–MoO /SiO is a pow-
ty and substrate scope. For the hydro-
x
2
Catalyst
erful catalyst for hydrogenation of
amino acids to amino alcohols (see
scheme) and is superior to reported hy-
drogenation catalysts in terms of activi-
genation of carboxylic acids, MoO spe-
x
cies play important roles in the genera-
tion of reactive hydride species and ad-
sorption of the substrate.
&
&
Chem. Eur. J. 2014, 20, 1 – 12
www.chemeurj.org
12
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!