Enantio-differentiating hydrogenation of methyl acetoacetate over asymmetrically modified reduced nickel catalysts: The effects of the nickel sources on the enantio-differentiating ability

Article abstract of DOI:10.1016/j.molcata.2003.09.022

The enantio-differentiating hydrogenation of methyl acetoacetate was carried out over a tartaric acid-NaBr-modified reduced nickel catalyst. Nickel hydroxide and nickel carbonate were used as the precursors of the nickel oxide. They were calcined to nickel oxides, then reduced to nickel metal (reduced nickel catalyst). It was revealed that the calcination temperature of the precursors was an important variable for attaining the high enantio- differentiating ability (e.d.a.) of the catalyst. The nickel surface with the least lattice defects would be appropriate for attaining a high e.d.a.

Full text of DOI:10.1016/j.molcata.2003.09.022

Journal of Molecular Catalysis A: Chemical 211 (2004) 93–96
Enantio-differentiating hydrogenation of methyl acetoacetate over
asymmetrically modified reduced nickel catalysts
The effects of the nickel sources on the enantio-differentiating ability
Tsutomu Osawaa,∗, Yuki Amayaa, Tadao Haradab, Osamu Takayasua
a
Faculty of Science, Toyama University, Toyama, Gofuku 930-8555, Japan
b
Faculty of Science and Technology, Ryukoku University, Otsu 520-2194, Japan
Received 17 June 2003; received in revised form 28 August 2003; accepted 9 September 2003
Abstract
The enantio-differentiating hydrogenation of methyl acetoacetate was carried out over a tartaric acid–NaBr-modified reduced nickel catalyst.
Nickel hydroxide and nickel carbonate were used as the precursors of the nickel oxide. They were calcined to nickel oxides, then reduced
to nickel metal (reduced nickel catalyst). It was revealed that the calcination temperature of the precursors was an important variable for
attaining the high enantio-differentiating ability (e.d.a.) of the catalyst. The nickel surface with the least lattice defects would be appropriate
for attaining a high e.d.a.
© 2003 Elsevier B.V. All rights reserved.
Keywords: Modified nickel catalyst; Nickel hydroxide; Nickel oxide; Enantio-differentiating; Hydrogenation
1. Introduction
were directly added to the reaction media) was applicable
for the tartaric acid–NaBr-modified nickel catalysts with a
high enantio-differentiating ability (e.d.a.) [25]. The in situ
modification is environmentally benign, because it saves en-
ergy for the modification process and generates no waste
solution. When the in situ modification was applied to the
preparation of a tartaric acid–NaBr-modified nickel catalyst,
the reduced nickel prepared by the reduction of nickel ox-
ide or fine nickel powder was needed for attaining a high
e.d.a. [26,27]. However, it was demonstrated that the e.d.a.
of the modified reduced nickel catalyst was influenced by
the nickel oxide manufacturers [18–20]. As the preparation
details of the commercial nickel oxides are scarcely known,
nickel oxides were prepared from nickel hydroxide or nickel
carbonate and used for the preparation of a modified nickel
catalyst in this study. The effects of the preparation variables
of the nickel oxide and nickel on e.d.a. were investigated.
It is well known that the tartaric acid–NaBr-modified
nickel is a solid catalyst, which hydrogenates ␤-ketoesters
and 2-alkanones with high enantio-selectivity. Many types
of the base nickel catalysts for preparing modified catalysts
have been investigated, that is, Raney nickel [1–11], sup-
ported nickel [1–5,12–16], nickel obtained by the reduc-
tion of nickel oxide (reduced nickel) [7,17–20], nickel ob-
tained by the decomposition of nickel formate [7,17,21],
nickel powder [19,22], and intermetallic catalysts [23,24].
Among them, Raney nickel is widely used for preparing
enantio-differentiating catalysts, because of its high hydro-
genation activity and high enantio-differentiating ability. The
tartaric acid–NaBr-modified Raney nickel catalyst has been
prepared in an aqueous solution of tartaric acid and NaBr at
pH 3.2 and 373 K (pre-modification). This pre-modification
process generated a large amount of waste solution con-
taining nickel ions and bromide ions [17]. We recently re-
ported that an in situ modification (tartaric acid and NaBr
2. Experimental
∗
The GLC measurements of the hydrogenated products
were carried out using a Hitachi 263-30 gas chromato-
Corresponding author.
E-mail address: osawa@sci.toyama-u.ac.jp (T. Osawa).
1381-1169/$ – see front matter © 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2003.09.022

94
T. Osawa et al. / Journal of Molecular Catalysis A: Chemical 211 (2004) 93–96
graph. The optical rotations were measured with a JASCO
DIP-1000 polarimeter. The X-ray diffraction patterns of the
nickel oxide and nickel were measured using a Shimadzu
XD-3A diffractometer.
2.1. Nickel oxide
Nickel oxide was prepared by the calcination of nickel hy-
droxide (Wako Pure Chemical Industries Ltd., lot TPG6939)
or nickel carbonate (approximately NiCO3·2Ni(OH)2·4H2O,
Wako Pure Chemical Industries Ltd., lot SKG2014) under
a stream of nitrogen (40 ml/min) and oxygen (10 ml/min)
for 3 h at the specified temperatures.
2.2. Reduced nickel catalyst
Nickel oxide (1 g) was reduced at the specified tempera-
tures in a hydrogen stream for 1 h.
Fig. 1. Effect of the calcination temperature of nickel hydroxide on e.d.a.
and crystallite size of nickel. Reduction temperature of nickel oxide was
623 K.
2.3. Enantio-differentiating hydrogenation of methyl
acetoacetate
3. Results and discussion
Modification of the reduced nickel catalyst was carried out
by an in situ modification method [27]. The reaction mixture
of methyl acetoacetate (3.3 g), THF (6.7 ml) and acetic acid
(0.067 g) containing (R,R)-tartaric acid (3.3 mg) and NaBr
(0.3 mg in 50 mm3 of H2O) was subjected to hydrogenation
over the reduced nickel catalyst. The hydrogenation was
carried out at 373 K and the initial hydrogen pressure was
9 MPa. Simple distillation gave a product of more than 98%
in chemical yield (GLC analyses: 5% Thermon 1000 on
Chromosorb W at 383 K).
3.1. Effect of the calcination temperature of nickel
hydroxide
The nickel oxide was prepared by the calcination of nickel
hydroxide from 773 to 1373 K. The nickel oxides were then
reduced at 623 K for 1 h under a hydrogen stream for prepar-
ing a reduced nickel catalyst. Fig. 1 shows the effects of
the calcination temperature on the e.d.a. of the modified
reduced nickel catalysts. When the calcination temperature
was raised, the e.d.a. of the modified reduced nickel in-
creased and the maximal e.d.a. was attained for the 1373 K
calcination of nickel oxide.
2.4. Determination of e.d.a.
It was demonstrated that the e.d.a. of the modified nickel
catalyst depended on the nickel crystallite size and that a
larger sized nickel crystallite was preferable for attaining
a high e.d.a. [7,15,16]. Therefore, the relation between the
mean crystallite size of nickel and the calcination temper-
ature was investigated. The values of the crystallite size of
nickel are also shown in Fig. 1. The higher calcination tem-
perature of nickel hydroxide gave the reduced nickel with
the larger nickel crystallite size after the reduction of nickel
oxide.
The e.d.a. of the catalyst was evaluated using the optical
purity of the hydrogenated product determined by polarime-
try.
[α]2D0 of methyl 3-hydroxybutyrate
e.d.a. (%) =
× 100
[α]2D0 of pure enantiomer
The specific optical rotations [α]D of the optically pure
(R)-methyl 3-hydroxybutyrate is [α]2D0 = −22.95◦ (neat)
[17].
3.2. Effect of the reduction temperature of nickel oxide
2.5. Measurement of crystallite size
Fig. 2 shows the effect of the reduction temperature of
nickel oxide on the e.d.a. and the crystallite size of nickel.
The nickel oxide was prepared by the calcination of nickel
hydroxide at 1373 K. The e.d.a. significantly decreased from
83 to 57% with an increase in the reduction temperature of
the nickel oxide from 598 to 673 K, while the nickel crys-
tallite size was almost constant at ca. 60 nm. These results
show that the e.d.a. is significantly related to the reduction
The mean crystallite sizes of nickel oxide and nickel
metal were obtained from the half-width of the NiO(1 1 1)
or Ni(1 1 1) peak using Scherrer’s method. Nickel oxide pre-
pared by the calcination of nickel carbonate at 1373 K and
nickel metal prepared by the reduction of nickel oxide at
873 K for 5 h in a hydrogen stream were used as the refer-
ences, respectively.

T. Osawa et al. / Journal of Molecular Catalysis A: Chemical 211 (2004) 93–96
95
Fig. 2. Effect of the reduction temperature of nickel oxide on e.d.a. and crystallite size of nickel. Nickel oxide was prepared from nickel hydroxide at
1373 K calcination.
temperature but that nickel crystallite size is not affected by
the reduction temperature of 598–673 K. The reduction at
673 K could result in enhanced sintering of the nickel surface
and the surface of nickel would become unsuitable for the
effective enantio-differentiation. The results of Fig. 2 could
be explained by the crystallite size distribution of nickel.
Nitta et al. demonstrated that the reduction at a high temper-
ature would cause the broad crystallite size distribution of
nickel and decrease the e.d.a. At high reduction temperature,
nickel with small crystallite size with low e.d.a. and high
hydrogenation activity would governed the reaction [28,29].
the modified reduced nickel catalysts with different nickel
crystallite size when nickel oxides were prepared from the
different precursors.
The relation between the nickel oxide crystallite size pre-
pared from both precursors (nickel hydroxide and nickel car-
bonate) and e.d.a. were also investigated. These results are
shown in Table 1. For both precursors, the crystallite size of
the nickel oxides increased with an increase in the calcina-
tion temperatures. The e.d.a. values also increased with the
increase in the crystallite size of the nickel oxides. At the low
calcination temperatures of 773 and 973 K, the e.d.a. values
significantly differed between the two precursors. However,
3.3. Effect of the types of the precursors of nickel oxide
In order to study the relation between the crystallite size
of nickel and e.d.a. in more detail, nickel oxide was also pre-
pared from nickel carbonate and used for the preparation of
the modified reduced nickel. Fig. 3 shows these results. For
both precursors (nickel hydroxide and nickel carbonate), the
e.d.a. of the catalysts increased with the nickel crystallite
sizes. However, the nickel crystallite size giving an e.d.a.
of 80% depended on the type of the precursor. That is, as
for the catalyst prepared from nickel hydroxide, the e.d.a.
of 80% was attained with the nickel crystallite size of ca.
60 nm, but as for the catalyst prepared from nickel carbon-
ate, the catalyst with a nickel crystallite size of ca. 40 nm
gave an 80% e.d.a. These findings reveal that there could ex-
ist factor(s) determining the e.d.a. other than the crystallite
size of the nickel, because the appropriate nickel crystallite
size for attaining a high e.d.a. depended on the type of pre-
cursor. It was demonstrated that the different types of the
modified nickel (Raney nickel, reduced nickel, and nickel
obtained by the decomposition of nickel formate) showed
the same e.d.a. with different nickel crystallite size [7]. Our
present study revealed that the same e.d.a. was attained over
Fig. 3. Effect of the crystallite size of nickel on e.d.a. (᭹) Precursor: nickel
hydroxide; (ᮀ) precursor: nickel carbonate. Temperature in the figure
indicates calcination temperature of the precursors. Reduction temperature
of nickel oxide was 623 K.

96
T. Osawa et al. / Journal of Molecular Catalysis A: Chemical 211 (2004) 93–96
Table 1
reduced nickel catalyst. Nickel hydroxide and nickel car-
bonate were calcined and the resulting nickel oxides were
reduced to nickel catalysts. The value of the nickel crystal-
lite size was not directly correlated with the e.d.a. The cal-
cination temperature of the nickel oxide precursor was the
important issue for attaining a high e.d.a. A smooth nickel
surface without the lattice defects would be appropriate for
attaining a high e.d.a.
Relation between the nickel oxide crystallite size and e.d.a.
Calcination
temperature
(K)
Precursor
Nickel hydroxide
Nickel carbonate
Crystallite size of
nickel oxide (nm)
e.d.a. Crystallite size of
e.d.a.
(%)
(%)
nickel oxide (nm)
773
973
1173
1373
54
121
141
50
70
81
83
50
132
>200
>200
8
55
74
79
>200
References
Reduction temperature of nickel oxide was 623 K.
at the higher calcination temperature of 1373 K, a greater
than 80% e.d.a. was attained for both precursors.
[1] W.M.H. Sachtler, Chem. Ind. (Dekker) 22 (1985) 189, and references
therein.
[2] A. Tai, T. Harada, in: Y. Iwasawa (Ed.), Tailored Metal Catalysts,
Reidel, Dordrecht, The Netherlands, 1986, p. 265, and references
therein.
[3] T. Sugimura, Catal. Surv. Jpn. 3 (1999) 37, and references therein.
[4] T. Osawa, T. Harada, O. Takayasu, Top. Catal. 13 (2000) 155, and
references therein.
[5] L.H. Gross, P. Rys, J. Org. Chem. 39 (1974) 2429.
[6] G.V. Smith, M. Musoiu, J. Catal. 60 (1979) 184.
[7] Y. Nitta, F. Sekine, T. Imanaka, S. Teranishi, Bull. Chem. Soc. Jpn.
54 (1981) 980.
[8] A. Bennett, S. Christie, M.A. Keane, R.D. Peacock, G. Webb, Catal.
Today 10 (1991) 363.
[9] G.V. Smith, R. Song, J.M. Delich, M. Bartók, Stud. Surf. Sci. Catal.
78 (1993) 67.
[10] J. Masson, P. Cividino, J. Court, J. Mol. Catal. A: Chem. 111 (1996)
289.
[11] P. Kukula, L. Cerveny, J. Mol. Catal. A: Chem. 185 (2002) 195.
[12] Y. Orito, S. Niwa, S. Imai, Yuuki Gosei Kagaku 34 (1976) 236.
[13] Y. Nitta, M. Kawabe, T. Imanaka, Appl. Catal. 30 (1987) 141.
[14] M.A. Keane, Langmuir 13 (1997) 41.
[15] T. Osawa, S. Mita, A. Iwai, O. Takayasu, H. Hashiba, S. Hashimoto,
T. Harada, I. Matsuura, J. Mol. Catal. A: Chem. 157 (2000)
207.
[16] A. Wolfson, S. Geresh, M.V. Landau, M. Herskowitz, Appl. Catal.
A: Gen. 208 (2001) 91.
[17] T. Harada, M. Yamamoto, S. Onaka, M. Imaida, H. Ozaki, A. Tai,
Y. Izumi, Bull. Chem. Soc. Jpn. 54 (1981) 2323.
[18] T. Harada, Y. Imachi, A. Tai, Y. Izumi, Metal-Support
and Metal-Additive Effects in Catalysis, Elsevier, Lyon, 1982,
p. 377.
[19] T. Osawa, T. Tai, A. Harada, O. Takayasu, I. Matsuura, Stud. Surf.
Sci. Catal. 108 (1997) 199.
[20] T. Osawa, E. Mieno, T. Harada, O. Takayasu, J. Mol. Catal. A:
Chem. 200 (2003) 315.
[21] I. Yasumori, M. Yokozeki, Y. Inoue, Faraday Discuss. Chem. Soc.
72 (1982) 385.
[22] H. Brunner, M. Muschiol, T. Wischert, Tertahedron: Asymm. 1 (1990)
159.
Nickel oxide prepared by the calcination at 1373 K has
a olive green color, but the color depended on the calci-
nation temperatures. A lower calcination temperature pro-
duced a black color. It is known that black nickel oxide is a
non-stoichiometric compound with excess oxygen and has
lattice defects including vacancies [30]. On the other hand,
the olive green nickel oxide has a lower amount of excess
oxygen and a lower amount of lattice defects compared with
the black nickel oxide. As the reduction of nickel oxide at
623 K for 1 h would not be enough for the rapid bulk ag-
glomeration of the nickel particles and the rearrangement of
the lattice atoms, nickel produced by the reduction of nickel
oxide would inherit the characteristics of the parent nickel
oxide such as the lattice defects. It has been reported that a
nickel surface with less disorder would be favorable for the
regularly arranged adsorption of tartaric acid thus giving a
high e.d.a. [7]. For the two precursors, the higher calcina-
tion temperature would be favorable for forming a lattice of
nickel oxide with less defects. This would result in the for-
mation of a smooth nickel surface with less defects after the
reduction of nickel oxide. A smooth nickel surface would
be appropriate for the effective enantio-differentiation.
From the results of Figs. 1–3 and Table 1, the e.d.a. of
the modified reduced nickel catalyst is not determined only
by the crystallite size of nickel. The e.d.a. also depended
on the characteristics of the parent nickel oxide, that is, the
type of the precursor of nickel oxide and the calcination
temperature of the precursor. The types of precursors could
contribute to the surface structure, adsorption species on the
surface, and/or contamination of nickel oxide and nickel. It
is not clear at present that what factor affects the e.d.a. of
modified reduced nickel catalyst most significantly. How-
ever, for attaining a high e.d.a., the calcination temperature
of the precursor of nickel oxide was the very important issue
for obtaining the appropriate nickel surface for the effective
enantio-differentiation.
[23] E. I Klabunovskii, Russ. Chem. Rev. 60 (1991) 980.
[24] A. Rives, E. Leclercq, R. Hubaut, in: R.E. Malz Jr. (Ed.), Catalysis
of Organic Reactions, Marcel Dekker, 1996, p. 241.
[25] T. Osawa, A. Ozawa, T. Harada, O. Takayasu, J. Mol. Catal. A:
Chem. 154 (2000) 271.
[26] T. Osawa, Y. Hayashi, A. Ozawa, T. Harada, O. Takayasu, J. Mol.
Catal. A: Chem. 169 (2001) 289.
[27] T. Osawa, S. Sakai, K. Deguchi, T. Harada, O. Takayasu, J. Mol.
Catal. A: Chem. 185 (2002) 65.
4. Conclusion
[28] Y. Nitta, F. Sekine, T. Imanaka, S. Teranishi, J. Catal. 74 (1982) 382.
[29] Y. Nitta, T. Imanaka, S. Teranishi, J. Catal. 96 (1985) 429.
[30] J. Deren, J. Stoch, J. Catal. 18 (1970) 249.
The enantio-differentiating hydrogenation of methyl ace-
toacetate was carried out over a tartaric acid–NaBr-modified