Preparation and characterization of the chirally modified rapidly quenched skeletal Ni catalyst for enantioselective hydrogenation of butanone to R-(-)-2-butanol

Article abstract of DOI:10.1016/j.molcata.2010.04.018

A rapidly quenched skeletal Ni (RQ Ni) prepared by the melt-spinning method followed by alkali leaching was used as a new category of Ni precursor to fabricate effective chiral Ni catalyst. A series of tartaric acid (TA)-NaBr-modified RQ Ni catalysts (MRQNi) were derived, and the textural and structural properties of the MRQNi catalysts were systematically investigated. It is found that the modification conditions influence profoundly the composition, texture, and Ni crystallite size of the MRQNi catalysts. In liquid phase hydrogenation of butanone to R-(-)-2-butanol, the optimized MRQNi catalyst exhibits much better activity and enantioselectivity than the similarly chirally modified Raney Ni catalyst. Based on the characterizations, the superior catalytic performance of the MRQNi catalyst is mainly attributed to the higher surface content of the sodium-nickel(II) tartrate complex than that on the chirally modified Raney Ni catalyst.

Full text of DOI:10.1016/j.molcata.2010.04.018

Journal of Molecular Catalysis A: Chemical 326 (2010) 113–120
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Preparation and characterization of the chirally modified rapidly quenched
skeletal Ni catalyst for enantioselective hydrogenation of butanone to
R-(−)-2-butanol
Zhiying Loua, Xueying Chena, Li Tiana, Minghua Qiaoa,∗, Kangnian Fana,
Heyong Hea,∗, Xiaoxin Zhangb, Baoning Zongb
a Department of Chemistry, Shanghai Key Laboratory of Molecular Catalysis and Innovative Materials, Fudan University, No. 220 Handan Road, Shanghai 200433, PR China
b State Key Laboratory of Catalytic Material and Chemical Engineering, Research Institute of Petroleum Processing, SINOPEC, Beijing 100083, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
A rapidly quenched skeletal Ni (RQ Ni) prepared by the melt-spinning method followed by alkali leaching
was used as a new category of Ni precursor to fabricate effective chiral Ni catalyst. A series of tartaric
acid (TA)–NaBr–modified RQ Ni catalysts (MRQNi) were derived, and the textural and structural proper-
ties of the MRQNi catalysts were systematically investigated. It is found that the modification conditions
influence profoundly the composition, texture, and Ni crystallite size of the MRQNi catalysts. In liquid
phase hydrogenation of butanone to R-(−)-2-butanol, the optimized MRQNi catalyst exhibits much bet-
ter activity and enantioselectivity than the similarly chirally modified Raney Ni catalyst. Based on the
characterizations, the superior catalytic performance of the MRQNi catalyst is mainly attributed to the
higher surface content of the sodium–nickel(II) tartrate complex than that on the chirally modified Raney
Ni catalyst.
Received 12 February 2010
Received in revised form 26 April 2010
Accepted 28 April 2010
Available online 6 May 2010
Keywords:
Rapid quenching
Skeletal Ni
Tartaric acid
Butanone
© 2010 Elsevier B.V. All rights reserved.
Enantioselective hydrogenation
1. Introduction
2-alkanones, such as 2-octanone, chirally modified Raney Ni (M-
Raney Ni) is more effective for achieving higher hydrogenation
There is a growing interest in the synthesis of optically pure
chiral substances especially in the pharmaceutical, agrochemical,
and perfume industries [1,2]. Currently, the sales of single iso-
mer drugs have already accounted for about half of the total drug
sales worldwide [3]. Among many strategies established to this
end, enantioselective hydrogenation over a heterogeneous catalyst
is one of the promising routes owing to its inherent operational
and economical advantages [4]. Up to now, one of the most inten-
sively studied chiral catalytic systems is the TA–NaBr–modified Ni
catalyst (MNi) [5–10], which has been demonstrated effective for
the enantioselective hydrogenation of pre-chiral ␤-ketoesters, ␤-
diketones, and 2-alkanones to chiral alcohols [5–18]. It was found
that nickel materials suitable as precursors for the preparation of
the MNi catalysts mainly fall in the following three categories:
Raney Ni leached from commercial Ni–Al alloy [11–13], Ni black
reduced from pure nickel oxide or nickel carbonate [14–16], and
commercial Ni powder [17] or fine Ni powder condensed from
Ni vapor in vacuum [10]. For enantioselective hydrogenation of
activity and enantioselectivity than pure Ni powder [7,10,18].
Recently, the introduction of the rapid quenching technique to
the preparation of the Ni–Al alloy has exerted a substantial impact
on catalytic investigations [19–26]. Explorations demonstrated
that after alkali leaching, the resulting RQ Ni catalyst exhibited
higher activity and selectivity for the hydrogenation of olefinic,
carbonyl, and nitryl groups than the commercial Raney Ni cata-
lyst [22,24,25]. As compared to Raney Ni that is leached from the
naturally solidified Ni–Al alloy, more defects are present in the
rapidly quenched Ni–Al alloy (RQ Ni–Al), leading to coordinatively
more unsaturated Ni crystallites in RQ Ni. Such crystallites are more
beneficial for the adsorption and activation of the reactants [21].
Moreover, RQ Ni has larger Ni crystallite size than Raney Ni [24]. It
has been proposed that larger Ni crystallites tend to afford MNi cat-
alysts better enantioselective hydrogenation ability [16,27]. Thus,
it will be very attractive to use RQ Ni as a new category of Ni
precursor to formulate a novel MNi catalyst for enantioselective
hydrogenation.
In this paper, we prepared a series of TA–NaBr–modified RQ
Ni catalysts (MRQNi) by pre-modification of RQ Ni in an aqueous
solution containing (2S,3S)-(−)-TA and NaBr, and examined their
catalytic performances in the enantioselective hydrogenation of
butanone. The M-Raney Ni catalyst was prepared according to the
∗
Corresponding authors. Tel.: +86 21 65643916; fax: +86 21 6562978.
E-mail addresses: mhqiao@fudan.edu.cn (M. Qiao), heyonghe@fudan.edu.cn
(H. He).
1381-1169/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2010.04.018

114
Z. Lou et al. / Journal of Molecular Catalysis A: Chemical 326 (2010) 113–120
typical modification condition of Raney Ni reported in the litera-
ture for comparison [5,28]. Based on systematic characterizations,
it is found that the modification conditions influenced significantly
the composition, texture, and crystallite size of the MRQNi cata-
lysts. The activity and enantioselectivity of the MRQNi catalysts in
the liquid phase hydrogenation of butanone with respect to the
modification conditions and reaction variables are reported.
The mean Ni crystallite size was obtained from the integral width of
the Ni(1 1 1) reflection using the Scherrer relation after correction
for instrumental broadening with the Warren procedure [30,31].
In order to reveal the nature of the tartrate species formed on
the MRQNi catalyst, the thermal decomposition behavior of the
surface species on the MRQNi catalyst was compared with those
of the nickel(II) tartrate and sodium–nickel(II) tartrate complexes
ˇ
synthesized according to Kukula and Cerveny´ [32] by thermogravi-
metric analysis (TGA; Perkin Elmer TGA7). The heating rate was
2. Experimental
10 K min−1, and the N2 flow rate was 10 ml min−1
.
The surface composition and chemical state were detected by
X-ray photoelectron spectroscopy (XPS; Perkin Elmer PHI5000C)
using Mg K␣ line (hv = 1253.6 eV) as the excitation source. The sam-
ple was pressed into a self-supported disc, mounted on the sample
stage, degassed in the pretreatment chamber at 383 K for 2 h in
vacuo, and then transferred into the analyzing chamber where the
background pressure was better than 2 × 10−9 Torr. All the binding
energy (BE) values were referenced to the C 1s peak of contam-
inant carbon at 284.6 eV. For quantitative purpose, the intensity
of the photoelectron peak was integrated after being subtracted a
Shirley-shaped background.
2.1. Catalyst preparation
The RQ Ni–Al alloy was prepared by a single roller melt-spinning
technique. The details are described in our previous work [24]. The
MRQNi catalysts were prepared by the following procedure. Exactly
1 g of the RQ Ni–Al alloy was treated with 10 ml of 6.0 M NaOH
solution at 363 K for 1 h under stirring to leach out Al. The result-
ing black solid was washed with distilled water until neutrality.
Then, the as-prepared RQ Ni was soaked in an aqueous solution
with different volume but the same amount of (2S,3S)-(−)-TA of
0.4 g and a desired amount of NaBr, and stirred gently for 1 h at a
predetermined temperature in a three-necked flask fitted with a
condenser. Prior to modification, the pH of the modification solu-
tion was adjusted to 3.2 using a NaOH aqueous solution. The MRQNi
catalysts were labeled MRQNi-x-y-z, where x refers to the weight
of NaBr in the modification solution; y, the volume of the mod-
ification solution; and z, the modification temperature in K. For
example, MRQNi-9-10-413 denotes the RQ Ni catalyst modified in
10 ml aqueous solution containing 9.0 g NaBr at 413 K. After mod-
ification, the supernatant was decanted, and the resulting MRQNi
catalyst was washed with distilled water three times, followed with
methanol three times to replace water. The catalyst was stored in
methanol for characterization and activity test.
2.3. Catalytic testing and product analysis
The hydrogenation reaction was carried out in
a 100 ml
mechanically stirred stainless steel autoclave at desired reaction
temperature and H2 pressure. Other reaction conditions were as
follows: 1.0 ml of butanone, 20 ml of methanol, 0.5 g of the cata-
lyst, 1.8 g of pivalic acid, and a stirring rate of 800 rpm to exclude
diffusion limitations. It is known that pivalic acid can effectively
facilitate tartaric acid to recognize the structures of 2-alkanone by
expelling the bulky alkyl group of the substrate [7]. Conversion
and enantiomeric excess (ee) were determined on a gas chro-
matograph equipped with a CP-Chirasil-Dex CB capillary column
(25 m × 0.25 mm × 0.25 ␮m) and a flame ionization detector (FID)
at 318 K. N2 was used as the carrier gas. The ee value was expressed
as:
The M-Raney Ni catalyst was leached from a commercially avail-
able Ni–Al alloy (Ni/Al, 50/50, w/w, Shanghai Chemical Corp.) and
modified according to the reported optimal condition [5,28]. In
brief, after alkali leaching, Raney Ni was soaked in 40 ml aqueous
solution containing 0.4 g (2S,3S)-(−)-TA and 6.0 g NaBr at 373 K
for 1 h. It should be cautioned that since RQ Ni and Raney Ni are
pyrophoric, care must be taken to preclude air oxidation during
sample handling and disposal.
[R-(−)-2-butanol] − [S-(+)-2-butanol]
ee (%) =
× 100.
[R-(−)-2-butanol] + [S-(+)-2-butanol]
The assignments of the absolute configurations of the products
were based on the retention time of authentic R-(−)-2-butanol
(Aldrich) and racemic 2-butanol (Shanghai Chemical Corp.)
reagents. It should be noted that in this work that we did not express
the activity in turnover frequencies per site or so on, because it is
infeasible to determine the number of active sites on these chirally
modified catalysts.
2.2. Characterization
The Ni and Al compositions of the catalyst were determined
by inductively coupled plasma-atomic emission spectroscopy
(ICP-AES; Thermo Elemental IRIS Intrepid). The multipoint
Brunauer–Emmett–Teller surface area (SBET) and pore size distribu-
tion (PSD) were acquired on a Micromeritics TriStar3000 apparatus
using N2 physisorption at 77 K. Prior to the measurement, the cat-
alyst was transferred to a glass adsorption tube and degassed at
383 K under N2 flow for 2 h. The pore volume was calculated from
the amount of N2 adsorbed at a relative pressure of 0.995. The PSD
curve was calculated from the desorption branch of the isotherms
using the Barrett–Joyner–Halenda (BJH) algorithm [29]. The surface
morphology was observed by scanning electron microscopy (SEM;
Philips XL30). Before being transferred into the SEM chamber, the
catalyst with methanol was dispersed on the sample holder and
quickly moved into the vacuum evaporator, in which a thin gold
layer was deposited after being dried in vacuo. The powder X-ray
diffraction (XRD) pattern was collected on a Bruker AXS D8 Advance
X-ray diffractometer using Cu K␣ radiation (ꢀ = 0.15418 nm). The
tube voltage was 40 kV, and the current was 40 mA. The cata-
lyst with methanol was loaded in the in situ cell, with Ar flow
(99.9995%) purging the cell during the detection to avoid oxidation.
3. Results and discussion
3.1. Catalyst composition
The bulk Ni and Al contents in the RQ Ni and Raney Ni catalysts
before and after chiral modification are given in Table 1. Since it
has been proposed that the enantioselectivity of the MNi catalyst
is related to the surface Ni/Al ratio [14], we also calculated the sur-
face Ni and Al contents by means of XPS. According to Table 1, the
bulk and surface Ni and Al contents of RQ Ni are similar, with the
amount of Al being ca. 21 wt% both in the bulk and on the surface.
Raney Ni has a lower bulk Al content than RQ Ni, while its surface
Al content is higher. After modification, irrespective of the modi-
fication conditions, all the MRQNi catalysts retained ca. 2/3 of the
initial amount of Al in the bulk. However, the surface Al contents of
the MRQNi catalysts deviated remarkably from the bulk ones, with
much less Al on the surface. The reason that Al was preferentially
removed from the surface is possibly due to the fact that the resid-

Z. Lou et al. / Journal of Molecular Catalysis A: Chemical 326 (2010) 113–120
115
Table 1
Physicochemical properties of the chirally modified skeletal Ni catalysts.
Catalyst
Bulk comp. (wt%)
Surf. comp. (wt%)
SBET (m2 g−1
)
Vpore (cm3 g−1
)
dpore (nm)
dcryst. (nm)
RQ Ni
Raney Ni
M-Raney Ni
Ni78.8Al21.2
Ni81.7Al18.3
Ni89.1Al10.9
Ni85.8Al14.2
Ni85.7Al14.3
Ni85.6Al14.4
Ni85.5Al14.5
Ni85.4Al14.6
Ni85.6Al14.4
Ni85.8Al14.2
Ni86.0Al14.0
Ni79.3Al21.7
Ni75.3Al24.7
Ni99.6Al0.4
Ni92.5Al7.5
Ni94.5Al5.5
Ni96.4Al3.6
Ni97.6Al2.4
Ni98.0Al2.0
Ni98.6Al1.4
Ni99.4Al0.6
Ni99.8Al0.2
90
78
29
47
40
35
30
26
21
18
18
0.10
0.10
0.21
0.15
0.13
0.12
0.10
0.09
0.09
0.08
0.08
4.6
3.9
6.8
5.6
14.2
8.7
21.7
12.4
14.3
16.5
18.1
19.4
14.2
12.0
10.9
MRQNi-9-40-373
MRQNi-9-40-383
MRQNi-9-40-393
MRQNi-9-40-403
MRQNi-9-40-413
MRQNi-9-30-413
MRQNi-9-20-413
MRQNi-9-10-413
9.0
10.2
12.5
14.2
14.3
14.5
14.6
40-413 to 18 m2 g−1 for MRQNi-9-10-413 when the modification
solution became more concentrated.
ual non-stoichiometric Ni2Al3 phase in the bulk of RQ Ni is very
resistant to corrosion [20,24,33,34]. For Raney Ni, after modifica-
tion more Al was leached away from the bulk, attributable to the
more facile removal of hydrated alumina adsorbed in the pores of
Raney Ni [11].
In Table 1, it is also found that the higher the modification tem-
perature, the lower is the surface Al content in the MRQNi catalysts.
When the modification temperature was elevated to 413 K, the sur-
face Al content was decreased to 0.2 wt% (entry 11, Table 1). On
the other hand, higher concentration of the modification solution
resulted in lower surface Al content. For example, the surface Al
content decreased monotonically from 2 wt% for MRQNi-9-40-413
to 0.2 wt% for MRQNi-9-10-413.
The modification temperature and the concentration of the
modification solution influenced the pore volumes of the MRQNi
catalysts as well. Higher modification temperature (entries 4–8,
Table 1) and concentration of the modification solution (entries
8–11, Table 1) led to smaller pore volume. And when the modi-
fication temperatures were below 403 K, the pore volumes of the
MRQNi catalysts were larger than that of RQ Ni, but further elevat-
ing the modification temperature led to pore volumes smaller than
that of RQ Ni. For Raney Ni, after modification, the pore volume
increased from 0.10 to 0.21 cm3 g−1, attributable again to the more
facile removal of the hydrated alumina blocking the pores of Raney
Ni [11].
3.2. Catalyst texture
Fig. 1a illustrates the N2 physisorption isotherms of Raney Ni,
M-Raney Ni, RQ Ni, and MRQNi-9-10-413 as a representative of the
MRQNi catalysts, since the isotherms of the MRQNi catalysts are
similar. Raney Ni and RQ Ni exhibited a type E isotherm according
to de Boer’s classification [35], indicating the ink-bottle pore shape
of both the unmodified catalysts. After modification, the isotherms
changed to type A, indicating that the pore shape of the chirally
modified Ni catalysts changed to the open-ended cylindrical pore
shape [35]. The corresponding PSD curves of the above catalysts
are plotted in Fig. 1b. Raney Ni exhibited only one maximum at
3.6 nm, while RQ Ni exhibited a bimodal distribution with a small
maximum at 2.4 nm and a large one at 3.6 nm. After modification,
the features of both Raney Ni and RQ Ni almost diminished, and
a new maximum at much larger pore diameter emerged. Detailed
analysis of the PSD curves of the MRQNi catalysts revealed that the
new maximum depended on the modification condition and varied
in the range of 8–15 nm. As compared to MRQNi, the maximum on
the PSD curve of M-Raney Ni shifted to a much larger diameter of
∼23 nm, presumably due to the more facile removal of the hydrated
alumina from Raney Ni.
The textural properties of Raney Ni and RQ Ni before and after
chiral modification are summarized in Table 1. RQ Ni exhibited
a BET surface area of 90 m2 g−1 and an average pore diameter
of 4.6 nm, which are larger than those of Raney Ni (78 m2 g−1
,
3.9 nm), while their pore volumes are similar. After modification,
the BET surface areas of the MRQNi and M-Raney Ni catalysts were
decreased as compared to those of RQ Ni and Raney Ni. The modifi-
cation temperature has a distinct effect on the BET surface areas of
the MRQNi catalysts. As shown in Table 1, the BET surface areas of
the MRQNi-9-40-z-series catalysts decreased from 47 to 26 m2 g−1
when the modification temperature was raised from 373 to 413 K.
The BET surface areas of the MRQNi catalysts also varied with the
concentration of the modification solution. Table 1 shows that the
BET surface areas decreased steadily from 26 m2 g−1 for MRQNi-9-
Fig. 1. (a) N2 adsorption–desorption isotherms and (b) pore size distribution curves
of the RQ Ni, Raney Ni, M-Raney Ni, and MRQNi-9-10-413 catalysts.

116
Z. Lou et al. / Journal of Molecular Catalysis A: Chemical 326 (2010) 113–120
Fig. 2. SEM images of the (a) RQ Ni, (b) Raney Ni, (c) MRQNi-9-10-413, and (d) M-Raney Ni catalysts.
3.3. Morphology and structure
(2 0 2), and (2 1 2) planes of Ni2Al3 were visible [36]. The existence
of the residual Ni2Al3 phase has been reported for leaching sim-
ilarly as-quenched Ni–Al [14] and Ni–P–Al alloys [13], which is
ascribed to the unique microstructure of the Ni–Al alloy prepared
by the rapid quenching technique [24]. The reflections of the Ni2Al3
phase still can be discerned in MRQNi-9-10-413, indicating that the
modification did not completely eliminate the Ni2Al3 phase.
The Ni mean crystallite sizes of the catalysts calculated by X-ray
line broadening are summarized in Table 1. It is apparent that the
Ni crystallite size in RQ Ni (6.8 nm) is larger than that in Raney Ni
(5.6 nm). It has been proposed that because of the rapid quench-
ing technique, more defects are generated in the RQ Ni–Al alloy,
thus Ni crystallites in the resulting RQ Ni are more coordinatively
unsaturated and inclined to aggregate into larger crystallites [24].
The Ni crystallite sizes in the MRQNi catalysts are larger than that
in RQ Ni. It is found in Table 1 that the Ni crystallite sizes in the
MRQNi catalysts were mainly dependent on the modification tem-
perature. The higher the modification temperature, the larger the
Ni crystallite size is. When the modification temperature was raised
to 413 K, the Ni crystallite size in the MRQNi catalysts was increased
to around 14 nm. On the contrary, the concentration of the modifi-
cation solution has less distinct effect on the Ni crystallite size in the
MRQNi catalysts, as evidenced by the last four entries in Table 1. It is
also noted that for M-Raney Ni the Ni crystallite size was increased
to ∼14 nm after modification.
Figure 2 shows the SEM images of RQ Ni and MRQNi-9-10-413
as a representative of the MRQNi catalysts. For comparison, the
SEM images of Raney Ni and M-Raney Ni are presented. It is found
that the particles of RQ Ni and MRQNi-9-10-413 displayed round
contour rather than the angular morphology of Raney Ni and M-
Raney Ni. After modification, the particles of MRQNi-9-10-413 and
M-Raney Ni became smaller and less compact than those of RQ
Ni and Raney Ni, which is consistent with the morphology change
between TA–NaBr–modified Raney Ni and Raney Ni observed pre-
viously [32].
The XRD patterns of the Raney Ni, M-Raney Ni, RQ Ni, and
MRQNi-9-10-413 catalysts are presented in Fig. 3. Both Raney Ni
and M-Raney Ni only showed reflections at 2ꢁ of 44.0◦, 51.2◦, and
75.5◦, corresponding to the (1 1 1), (2 0 0), and (2 2 0) planes of fcc Ni
[36]. For RQ Ni, besides the reflections of fcc Ni, additional reflec-
tions at 2ꢁ of 44.2◦, 65.6◦, and 83.0◦ characteristic of the (1 1 0),
3.4. Surface species and chemical states
To reveal the nature of the tartrate species formed on the
MRQNi catalysts, the thermal decomposition behaviors of MRQNi-
9-10-413, sodium–nickel(II) tartrate, and nickel(II) tartrate were
analyzed. The latter two are possible tartrate species on the MNi
catalysts [8,32,37]. For comparison, the thermal decomposition
behavior of M-Raney Ni catalyst is also analyzed. As shown in
Fig. 4, aside from a distinct weight loss at ca. 373 K due to dehy-
dration, decomposition of nickel(II) tartrate occurred at ca. 623 K.
The total weight loss of nickel(II) tartrate was ca. 73.8%, which is in
accordance with the weight loss of 76.7% reported before [32]. For
sodium–nickel(II) tartrate, following the continuous weight loss
from the beginning of the heating process, there was a distinct
Fig. 3. XRD patterns of the RQ Ni, Raney Ni, M-Raney Ni, and MRQNi-9-10-413
catalysts.

Z. Lou et al. / Journal of Molecular Catalysis A: Chemical 326 (2010) 113–120
117
Fig. 4. TGA curves of Ni(II) tartrate, sodium–nickel(II) tartrate, M-Raney Ni and the
MRQNi-9-10-413 catalyst.
weight loss at ca. 591 K, resulting in a total weight loss of 43.9%
also close to the literature value [32]. For MRQNi-9-10-413 and M-
Raney Ni, besides the weight loss at ca. 323 K attributable to the
evaporation of methanol, another weight loss process took place
at the temperature identical to that of sodium–nickel(II) tartrate,
inferring the formation of this tartrate species on the MRQNi and
the M-Raney Ni catalysts. As estimated by the weight loss in TGA
measurements, the amounts of sodium–nickel(II) tartrate species
adsorbed on the MRQNi-9-10-413 catalyst are ca. 10.5 wt.%, which
is higher than that on the M-Raney Ni catalysts (8.4 wt.%). On the
other hand, because the complex of sodium–nickel(II) tartrate is
amorphous [32], it is not detectable in Fig. 3 by XRD.
The surface states of the RQ Ni and MRQNi catalysts were ana-
lyzed by XPS. As the Ni 2p spectra of the MRQNi catalysts are similar,
the spectrum of MRQNi-9-10-413 is presented in Fig. 5a as a rep-
resentative. The sodium–nickel(II) tartrate complex was used for
comparison. RQ Ni only showed one sharp Ni 2p3/2 peak at 852.7 eV,
indicating that nickel is predominant in the metallic state [38]. The
sodium–nickel(II) tartrate complex is represented by the Ni 2p3/2
peak at 856.2 eV and a broad satellite at higher BE value, which
are characteristic of the divalent paramagnetic Ni2+. For the M-
Raney Ni and MRQNi-9-10-413 catalysts, after curve fitting, there
are three components assignable to metallic Ni with Ni 2p3/2 BE
of 852.7 eV, sodium–nickel(II) tartrate with Ni 2p3/2 BE of 856.1 eV
and its satellite peak with BE of 861.7 eV (Fig. 5b) [38]. Based on the
curve fitting results, the surface Ni2+/Ni0 ratio on the MRQNi-9-10-
413 catalyst was calculated to be 1.19, while the ratio was 0.98 on
the M-Raney Ni catalyst, showing that the surface density of the
Fig. 5. (a) Ni 2p spectra of RQ Ni, sodium–nickel(II) tartrate, MRQNi-9-10-413, and
M-Raney Ni, and (b) curve fitting results of the Ni 2p level of the MRQNi-9-10-413
and M-Raney Ni catalysts.
sodium–nickel(II) tartrate species on the former is higher than that
on the latter, which is in accordance with the TGA observations.
3.5. Enantioselective hydrogenation of butanone
3.5.1. Effect of the modification parameters
The catalytic results of the RQ Ni and MRQNi catalysts in liquid
phase enantioselective hydrogenation of butanone are compiled in
Table 2. The modification conditions are also presented in the table
Table 2
Effect of the modification conditions on the conversion and enantioselectivity over the MRQNi catalysts.a.
Catalyst
Modification condition
Conversion (%)
eemax (%)
WTA (g)
WNaBr (g)
VH (ml)
T (K)
O
2
RQ Ni
TA-MRQNi
M-Raney Ni
–
–
0
6
6
9
9
9
9
9
9
9
9
–
–
373
373
373
373
383
393
403
413
413
413
413
99.9 (1 h)
17.1
18.0
45.1
50.0
48.2
45.6
43.4
40.1
39.2
38.6
36.5
0 (1 h)
5.6
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
40
40
40
40
40
40
40
40
30
20
10
19.4
9.8
MRQNi-6-40-373
MRQNi-9-40-373
MRQNi-9-40-383
MRQNi-9-40-393
MRQNi-9-40-403
MRQNi-9-40-413
MRQNi-9-30-413
MRQNi-9-20-413
MRQNi-9-10-413
12.4
15.0
17.5
20.3
22.7
24.3
27.5
29.8
a
Reaction conditions: T = 333 K, PH = 9 MPa, and t = 6 h. Other reaction conditions are described in Section 2.3.
2

118
Z. Lou et al. / Journal of Molecular Catalysis A: Chemical 326 (2010) 113–120
surface Al content and the larger Ni crystallite size may account
for the higher enantioselectivity of the MRQNi catalyst modified at
higher temperature.
As to the influence of the concentration of the modification
solution on the catalytic performance of the MRQNi catalyst,
Table 2 shows that higher concentration led to lower conversion
and higher enantioselectivity. Over the MRQNi-9-40-413 catalyst,
the butanone conversion was 40.1%, and the ee value was 22.7%.
Over the MRQNi-9-10-413 catalyst, the butanone conversion was
decreased to 36.5%, while the ee value was increased to 29.8%. It is
expected that with the increment of the concentration of TA, more
tartrate species were formed and adsorbed on the catalyst sur-
face, resulting in lower activity but higher enantioselectivity [39].
Moreover, more concentrated modification solution contributed to
lower surface Al content and larger Ni crystallite size, which are also
beneficial to the enantioselectivity as demonstrated by the last four
entries in Table 2.
Fig. 6. Dependence of the conversion and enantioselectivity on the modification
temperature over the MRQNi-9-40-z-series catalysts.
The effect of the amount of NaBr in the modification solution on
the catalytic performance of the MRQNi catalyst was also investi-
gated. As shown by entry 4 in Table 2, when there were 6.0 g of NaBr
in the modification solution, the butanone conversion of 45.1% and
the ee value of 9.8% were obtained. When the amount of NaBr was
increased to 9.0 g, the conversion and the ee value were increased
to 50.0% and 12.4%, respectively (entry 5, Table 2). However, exces-
sive amount of NaBr such as 12.0 g reduced both the conversion
for clarity. Over RQ Ni, the reactant was nearly completely con-
verted to 2-butanol within 1 h, but with no enantioselectivity being
observed. When being modified with TA only, the resulting TA–RQ
Ni catalyst exhibited a much lower activity, giving a conversion of
only 17.1% after a reaction time as long as 6 h with an ee value of
5.6%. It has been reported that the addition of a co-modifier NaBr
to the modification solution can effectively improve the enantios-
electivity of the MNi catalysts in reactions such as hydrogenation
of 2-octanone and methyl acetoacetate [5,39], so we modified RQ
Ni with a combination of TA and NaBr.
We first attempted the typical modification condition for Raney
Ni [5,28] to prepare the MRQNi catalyst; the resulting catalyst is
MRQNi-6-40-373. As shown in Table 2, the butanone conversion
over MRQNi-6-40-373 at a reaction time of 6 h was 45.1%, which is
substantially higher than that over TA–RQNi modified by TA only.
However, the ee value was still low (9.8%), although it has been
improved somewhat relative to TA–RQNi. For the control M-Raney
Ni catalyst, a much higher ee value of 19.4% was obtained, showing
that the typical modification condition for Raney Ni is not suitable
for RQ Ni.
Fig. 6 shows the effect of the modification temperature on the
conversion of butanone and enantioselectivity over the MRQNi-
9-40-z-series catalysts. When the modification temperature was
raised from 373 to 413 K, the butanone conversion decreased from
50.0% to 40.1%, while the ee value increased almost linearly from
12.4% to 22.7%. The decrease in the activity with the modification
temperature may be attributed to the fact that the MRQNi catalyst
prepared at higher modification temperature possessed lower BET
surface area accessible by the reactants (Table 1). As to the factors
affecting the enantioselectivity, Harada et al. suggested that the Al
species on the catalyst surface is adverse to the enantioselectivity
over the chirally modified Raney Ni catalyst, and the removal of
which is important for obtaining a highly enantioselective catalyst
for the hydrogenation of methyl acetoacetate [14]. Tai and Harada
also demonstrated that the surface Al content affected the enan-
tioselectivity [6]. In order to enhance the enantioselectivity, they
introduced ultrasound irradiation to the preparation of Raney Ni
based on the idea that most Al species could be removed in this way
[40]. As verified by elemental analysis, the higher the modification
temperature is, the lower the Al content on the catalyst surface is
(Table 1). On the other hand, Table 1 shows that higher modification
temperature led to larger Ni crystallite size in the MRQNi catalyst. It
has been revealed by Nitta et al. that the catalyst made of larger Ni
crystallite tended to afford the higher enantioselective ability [16].
Similarly, Osawa et al. observed that larger Ni crystallite size was
favorable for attaining higher optical yields [27]. Thus, the lower
ˇ
and enantioselectivity (not shown). Kukula and Cerveny´ suggested
that NaBr changed the ion strength of the modification solution and
decreased the concentration of H+ ions, resulting in lower degree of
Ni leaching to the modification solution, which thus preserves the
activity of the chirally modified Ni catalyst in the hydrogenation
of methyl acetoacetate [39]. As reported in the literature, the opti-
mal amount of NaBr to maximize the enantioselectivity depends
on the type of the Ni catalysts [41]. We found that M-Raney Ni
exhibited the best catalytic performance when modified with 6.0 g
of NaBr, which is less than that required for RQ Ni. The role of
NaBr in enhancing the enantioselectivity has been widely studied
on the chirally modified Raney Ni catalysts [5–7,39]. It is generally
accepted that the Na+ and/or Br− ions can block the unmodified
active sites for the racemic reaction, thus increasing the enantiose-
lectivity [5,39]. Excessive NaBr will also block the chiral sites, which
is detrimental to both the activity and the enantioselectivity of the
MRQNi catalyst.
3.5.2. Effect of the reaction variables
Table 3 summarizes the butanone conversion and the enantios-
electivity over the MRQNi-9-10-413 catalyst at different reaction
temperature and H2 pressure. The effect of the reaction temper-
ature on the catalytic performance was investigated under the
constant H2 pressure of 9 MPa. It can be seen that the butanone con-
version was improved when the reaction temperature was raised
from 323 to 343 K, while the ee value first increased and then
declined dramatically at 343 K. The occurrence of the maximum in
the enantioselectivity with the reaction temperature was observed
on the chirally modified Raney Ni and Ni/SiO2 catalysts in the
Table 3
Effect of the reaction variables on the conversion and enantioselectivity over the
MRQNi-9-10-413 catalyst at a reaction time of 6 h.a.
T (K)
PH (MPa)
Conversion (%)
eemax (%)
2
333
333
333
323
343
3.0
6.0
9.0
9.0
9.0
16.2
26.4
36.5
24.3
50.7
14.5
17.3
29.8
27.1
13.4
a
Other reaction conditions are described in Section 2.3.

Z. Lou et al. / Journal of Molecular Catalysis A: Chemical 326 (2010) 113–120
119
hydrogenation of methyl acetoacetate [42,43]. Similarly, Osawa et
al. found that the hydrogenations of 2-alkanones over the chirally
modified Raney Ni catalyst gave higher optical yields at 333 K than
those at 373 K [10].
The evolution of the enantioselectivity with the reaction time
was monitored and illustrated in Fig. 7b. Over both catalysts, the ee
value first increased with the reaction time, reached a maximum at
the reaction time of 2 h, and then declined at prolonged time. More-
over, the ee value over the MRQNi-9-10-413 catalyst during the
whole hydrogenation process was substantially higher than that
over M-Raney Ni. The maximum ee value of 29.8% was obtained
over the MRQNi-9-10-413 catalyst, while it was only 19.4% over
M-Raney Ni.
With the reaction temperature being fixed at 333 K, it is found
that when the H2 pressure was increased from 3 to 9 MPa, the
butanone conversion and the ee value over the MRQNi catalysts
increased respectively from 16.2% to 36.5% and 14.5% to 29.8%. This
ˇ
result is similar to that of Kukula and Cerveny´, where the opti-
cal yield increased steadily with the increment of the H2 pressure
from 1 to 10 MPa in the enantioselective hydrogenation of methyl
acetoacetate over the chirally modified Raney Ni catalyst [43].
As shown in Table 1, the MRQNi-9-10-413 catalyst has a slightly
lower surface Al content and a slightly larger mean Ni crystallite
size than M-Raney Ni. It has been proposed that lower surface
Al content [6,14,40] and larger Ni crystallite size [16,27] are both
favorable for attaining higher enantioselectivity. However, the rel-
atively small differences in the surface Al content and the Ni
crystallite size, although reproducible, may not suffice to justify
the much larger difference between the ee values of the two cat-
alysts. On the other hand, it has been proposed that the complex
formed by TA is important to achieve the high enantioselectivity
[8,32,37]. As revealed by the XPS measurements, the percentage of
the sodium–nickel(II) tartrate complex with respect to the total
amount of the surface Ni species on the MRQNi-9-10-413 cata-
lyst is ca. 10% higher than that on M-Raney Ni, which may be one
important reason underlying the higher enantioselectivity over the
MRQNi catalysts.
3.5.3. Comparison of the catalytic behaviors of M-Raney Ni and
MRQNi
In order to have a better insight into the difference in the cat-
alytic performances of the chirally modified Raney Ni and RQ Ni
catalysts, the reaction courses of two typical catalysts, M-Raney
Ni and MRQNi-9-10-413, were compared under the optimized
reaction conditions of these catalysts. Fig. 7a shows that over
the MRQNi-9-10-413 catalyst, the butanone conversion increased
much faster than that over M-Raney Ni: after a reaction time of
6 h, the butanone conversion over the former was 36.5%, which is
twice as that over the latter. Fouilloux has observed that the resid-
ual Ni2Al3 phase in Raney Ni is responsible for the enhancement of
the hydrogenation activity [44], which may be the reason for the
higher activity of the MRQNi-9-10-413 catalyst.
4. Conclusions
The MRQNi catalyst derived from the rapidly quenched Ni–Al
alloy can be both more active and more enantioselective than the
M-Raney Ni catalyst in the liquid phase hydrogenation of butanone
to R-(−)-2-butanol. As compared to M-Raney Ni, higher modifica-
tion temperature and higher concentration of NaBr were needed for
MRQNi to achieve the optimal enantioselectivity. The superior cat-
alytic performance of the MRQNi catalyst to M-Raney Ni is mainly
attributed to the higher surface population of the sodium–nickel(II)
tartrate complex on the former, although the lower surface Al
content and the larger Ni crystallite size may also have some con-
tribution. This work, although preliminary, clearly demonstrates
the prospective of the rapidly quenched Ni–Al alloy as a new
category of Ni precursor for the preparation of efficient MNi cat-
alyst for enantioselective hydrogenation of pre-chiral ketones or
other carbonyl group-containing pre-chiral molecules to chiral
alcohols.
Acknowledgments
This work was supported by the National Basic Research Pro-
gram of China (2006CB202502), the NSF of China (20203004,
20721063, 20803009, and 20873027), the Program of New Century
Excellent Talents in Universities (NCET-08-0126), and the Science &
Technology Commission of Shanghai Municipality (08DZ2270500
and 09DZ2271500).
References
[1] A. Baiker, H.U. Blaser, in: G. Ertl, H. Knözinger, J. Weitkamp (Eds.), Handbook of
Heterogeneous Catalysis, Wiley-VCH, Weinheim, 1997, p. 2422.
[2] G. Jannes, V. Dubois, Chiral Reactions in Heterogeneous Catalysis, Plenum Press,
New York, 1995, pp. 1–2.
[3] F. Zaera, J. Phys. Chem. C 112 (2008) 16196–16203.
[4] G. Webb, P.B. Wells, Catal. Today 12 (1992) 319–337.
[5] Y. Izumi, Adv. Catal. 32 (1983) 215–271.
[6] A. Tai, T. Harada, in: Y. Iwasawa (Ed.), Tailored Metal Catalysts, Reidel, Dor-
drecht, 1986, p. 265.
[7] T. Osawa, T. Harada, A. Tai, Catal. Today 37 (1997) 465–480.
[8] T. Osawa, T. Harada, O. Takayasu, Top. Catal. 13 (2000) 155–168.
[9] T. Sugimura, Catal. Surv. Jpn. 3 (1999) 37–42.
Fig. 7. The evolutions of (a) the conversion of butanone, and (b) the enantioselec-
tivity over the M-Raney Ni and MRQNi-9-10-413 catalysts. The reaction conditions
are the same as those described in Table 2.
[10] T. Osawa, T. Harada, A. Tai, J. Catal. 121 (1990) 7–17.

120
Z. Lou et al. / Journal of Molecular Catalysis A: Chemical 326 (2010) 113–120
[11] J. Freel, W.J.M. Pieters, R.B. Anderson, J. Catal. 14 (1969) 247–256.
[12] T. Harada, Y. Izumi, Chem. Lett. (1978) 1195–1196.
[13] L.H. Gross, R. Rys, J. Org. Chem. 39 (1974) 2429–2430.
[14] T. Harada, S. Onaka, A. Tai, Y. Izumi, Chem. Lett. (1977) 1131–1132.
[15] T. Harada, Y. Imachi, A. Tai, Y. Izumi, Metal–Support and Metal–Additive Effects
in Catalysis, Elsevier Science, New York, 1982, p. 377.
[16] Y. Nitta, F. Sekine, T. Imanaka, S. Teranishi, Bull. Chem. Soc. Jpn. 54 (1981)
980–984.
[27] T. Osawa, S. Mita, A. Iwai, O. Takayasu, H. Hashiba, S. Hashimoto, T. Harada, I.
Matsuura, J. Mol. Catal. A 157 (2000) 207–216.
[28] T. Harada, M. Yamamoto, S. Onaka, M. Imaida, H. Ozaki, A. Tai, Y. Izumi, Bull.
Chem. Soc. Jpn. 54 (1981) 2323–2329.
[29] E.P. Barrett, L.G. Joyner, P.P. Halenda, J. Am. Chem. Soc. 73 (1951) 373–380.
[30] B.E. Warren, J. Appl. Phys. 12 (1941) 375–383.
[31] S.D. Robertson, R.B. Anderson, J. Catal. 23 (1971) 286–294.
ˇ
[32] P. Kukula, L. Cerveny´, Appl. Catal. A: Gen. 223 (2002) 43–55.
[17] H. Brunner, M. Muschiol, T. Wischert, Tetrahedron: Asymmetry 1 (1990)
159–162.
[18] Y. Izumi, M. Imaida, T. Harada, T. Tanabe, S. Yajima, T. Ninomiya, Bull. Chem.
Soc. Jpn. 42 (1969) 241–243.
[33] M.L. Bakker, D.J. Young, M.S. Wainwright, J. Mater. Sci. 23 (1988) 3921–3926.
[34] J. Masson, P. Cividino, J. Court, J. Mol. Catal. A 111 (1996) 289–295.
[35] J.H. de Boer, in: D.H. Everett, F.S. Stone (Eds.), The Structure and Properties of
Porous Materials, Butterworth, London, 1958, p. 68.
[19] K. Klement, R.H. Willens, P. Duwez, Nature 187 (1960) 869–870.
[20] Á. Molnár, G.V. Smith, M. Bartók, Adv. Catal. 36 (1989) 329–383.
[21] G.A. Somorjai, Catal. Rev. Sci. Eng. 18 (1978) 173–181.
[22] B. Liu, M.H. Qiao, J.F. Deng, K.N. Fan, X.X. Zhang, B.N. Zong, J. Catal. 204 (2001)
512–515.
[23] H.R. Hu, M.H. Qiao, Y. Pei, K.N. Fan, H.X. Li, B.N. Zong, X.X. Zhang, Appl. Catal. A:
Gen. 252 (2003) 173–183.
[24] H.R. Hu, M.H. Qiao, S. Wang, K.N. Fan, H.X. Li, B.N. Zong, X.X. Zhang, J. Catal. 221
(2004) 612–618.
[25] H.R. Hu, F.Z. Xie, Y. Pei, M.H. Qiao, S.R. Yan, H.Y. He, K.N. Fan, H.X. Li, B.N. Zong,
X.X. Zhang, J. Catal. 237 (2006) 143–151.
[26] F.Z. Xie, X.W. Chu, H.R. Hu, M.H. Qiao, S.R. Yan, Y.L. Zhu, H.Y. He, K.N. Fan, H.X.
Li, B.N. Zong, X.X. Zhang, J. Catal. 241 (2006) 211–220.
[36] PDFMaint Version 3.0, Powder Diffraction Database, Bruker Analytical X-ray
Systems GmbH, 1997.
[37] A. Hoek, W.M.H. Sachtler, J. Catal. 58 (1979) 276–286.
[38] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, in: J. Chastain (Ed.), Handbook
of X-ray Photoelectron Spectroscopy, Perkin-Elmer, Eden Prairie, MN, 1992.
ˇ
[39] P. Kukula, L. Cerveny´, Appl. Catal. A: Gen 210 (2001) 237–246.
[40] A. Tai, T. Kikukawa, T. Sugimura, Y. Inoue, S. Abe, T. Osawa, T. Harada, Bull.
Chem. Soc. Jpn. 67 (1994) 2473–2477.
[41] T. Harada, A. Tai, M. Yamamoto, H. Ozaki, Y. Izumi, Stud. Surf. Sci. Catal. 7 (1981)
364–375.
[42] M.A. Keane, Langmuir 13 (1997) 41–50.
ˇ
[43] P. Kukula, L. Cerveny´, J. Mol. Catal. A 185 (2002) 195–202.
[44] P. Fouilloux, Appl. Catal. 8 (1983) 1–42.