Platinum-encapsulated zeolitically microcapsular catalyst for one-pot dynamic kinetic resolution of phenylethylamine

Article abstract of DOI:10.1016/j.jcat.2012.04.008

The platinum-encapsulated zeolitically microcapsular catalyst, associated with the immobilized Candida antartica lipase B (Novozyme435), is successfully employed in the dynamic kinetic resolution of phenylethylamine. A conversion of 80% and a selectivity of 95% are achieved, and negligible loss of activity is detected even after reaction of 5 runs. It is found that the existence of the silicalite-1 shell not only effectively prevents the deactivation of both enzyme and Pt by isolating them in different regions of reaction system, but also significantly reduces the formation of by-products on the Pt nanoparticles within the protected space of zeolitic microcapsule. Such features of zeolitic shell should further promote the designing of various catalysts for multistep reaction network.

Full text of DOI:10.1016/j.jcat.2012.04.008

Journal of Catalysis 291 (2012) 87–94
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Platinum-encapsulated zeolitically microcapsular catalyst for one-pot dynamic
kinetic resolution of phenylethylamine
⇑
⇑
Jing Shi, Xiang Li, Quanrui Wang, Yahong Zhang , Yi Tang
Department of Chemistry, Shanghai Key Laboratory of Molecular Catalysis and Innovative Materials, Institute of Catalysis, and Laboratory of Advanced Material,
Fudan University, Shanghai 200433, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
The platinum-encapsulated zeolitically microcapsular catalyst, associated with the immobilized Candida
antartica lipase B (Novozyme 435), is successfully employed in the dynamic kinetic resolution of phen-
Received 17 November 2011
Revised 12 March 2012
Accepted 15 April 2012
Available online 16 May 2012
Ò
ylethylamine. A conversion of 80% and a selectivity of 95% are achieved, and negligible loss of activity is
detected even after reaction of 5 runs. It is found that the existence of the silicalite-1 shell not only effec-
tively prevents the deactivation of both enzyme and Pt by isolating them in different regions of reaction
system, but also significantly reduces the formation of by-products on the Pt nanoparticles within the
protected space of zeolitic microcapsule. Such features of zeolitic shell should further promote the
designing of various catalysts for multistep reaction network.
Keywords:
Zeolitic microcapsule
Multifunction catalyst
Isolation of active sites
Phenylethylamine
Ó 2012 Elsevier Inc. All rights reserved.
Dynamic kinetic resolution
1
. Introduction
During the last decades, microcapsule has been developed to be
coupling reaction, direct synthesis reactions of middle isopraffins
and dimethyl ether (DME) from syngas and so on [21–24]. And it
has been proved that the zeolitic shell can not only protect the ac-
tive species from poisoning and leaching outside during the reac-
tion but also contribute a high spatial selectivity and shape
selectivity for the products with different molecular sizes.
an attractive material used for various applications in pharmaceu-
tical, cosmetic, food, textile, adhesive, agricultural industries and
catalytic domains [1–7]. Up to now, the structure of the capsule
has been evolved from the zero-dimensional solid spheres and
one-dimensional wires, rods or belts (identified as the first
generation of one-level structure) to the second generation with
core–shell construction as the two-level structural materials and
then the third generation with multilevel interiors including di-
verse complex architectures such as the multishell spheres, the
multichamper spheres, the multiwalled tubes and the multichan-
nel tubes [8–11]. In terms of the composition of the microcapsule,
it mainly covers the inorganic compounds (carbon, silicon and cal-
cium), organic compounds or inorganic/organic hybrid substances
as well as the organometallic species [12–17]. Among them, the
inorganically zeolitic microcapsule (ZMC) has attracted much
attention due to the unique characteristics provided by its special
configuration, such as the confined space effect, the high-thermal
stability and the molecular sieving effect [18–20]. The encapsula-
tion of catalytic active species into ZMC has been highlighted by
using them in the practical reactions of alcohol oxidation, Heck
Nowadays, the optically pure enantiomers are of great signifi-
cance in chiral chemistry, agricultural chemistry and pharmaceuti-
cal chemistry [25–27]. A variety of methods have been used to
obtain pure enantiomers, such as the kinetic resolution (KR), the
asymmetric synthesis and the chiral chromatography separation
[28–30]. The KR process using enzyme as an acylation catalyst
has grown to be an important and appealing method owing to its
high efficiency and moderate condition. However, the theoretical
yield limit of 50% remains a main drawback for this process [31].
An effective strategy to overcome this problem is dynamic kinetic
resolution (DKR) by combining the KR with the racemization of
enantiomers in a one-pot system [31]. By surpassing the limitation
of KR, DKR is theoretically possible to obtain 100% yield of the pure
enantiomers from a racemic mixture. Various catalysts, such as the
metal nanoparticles, the solid acidic materials and the organome-
tallic complexes, have been explored to catalyze the racemization
reaction under a defined reaction condition [32,33]. The metal cat-
alyzed racemization mainly relies on the reversible hydrogen-
transfer process and/or the hydrogenation and dehydrogenation
course [34,35], but its development in this field is still a challenge.
Platinum, known as an efficient catalyst for hydrogenation and
dehydrogenation procedure [36], has been used for racemization
⇑
Corresponding authors. Address: Department of Chemistry, Fudan University,
No. 220, Handan Road, Shanghai 200433, PR China. Fax: +86 21 65641740.
E-mail addresses: zhangyh@fudan.edu.cn (Y. Zhang), yitang@fudan.edu.cn
(
Y. Tang).
0
021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2012.04.008

8
8
J. Shi et al. / Journal of Catalysis 291 (2012) 87–94
0
of 1,1 -binaphthyl ever since 1982 by Pincock [37]. However, only a
few reports were published for further application of platinum in
DKR process [38], probably due to its interaction with the enzyme
D/MAX-IIA diffractometer over a 2 theta range of 5–80° with Cu Ka
radiation. The X-ray photoelectron spectroscopy (XPS) analysis
was carried out with a Perkin-Elmer PHI 5000C ESCA system using
[
39,40]. The metal could act as an inhibitor for enzyme in the acyl-
Al Ka radiation (1486.6 eV) at a power of 250 W. The pass energy
ation reaction, and the interference of enzyme with metal may also
lead to poor racemization on the latter, as was pointed out by Mar-
tín-Matute and Backväll [40].
Herein, the ZMC with encapsulated platinum nanoparticles and
an intact thin shell of silicalite-1 (denoted as intact-Pt@S1) is
employed as racemization catalyst in the DKR of phenylethylamine
was set at 93.9 eV, and the binding energies (BE) were calibrated
by using contaminant carbon at a BE of 284.6 eV. All samples were
heavily grinded to expose the Pt species before XPS measurement.
The textural properties of the samples were determined from the
nitrogen sorption isotherms at ꢁ196 °C using a Micromeritics ASAP
2010 system. The elemental analyses were performed by induc-
tively coupled plasma atomic emission spectroscopy (ICP-AES)
with the limit of detection 0.001 mg/mL. The solid-state near UV
Circular Dichroism (CD) spectra were measured using a MOS-450
(Biologic, France) instrument, with finely grounded KBr powder
and a detection range of 200–350 nm.
(
PEA) for the first time. Thanks to the protection of the zeolitic
shell, this emerged catalyst associated with enzyme displays high
selectivity and activity for DKR of PEA. Through the comparison
with the data obtained from the heavily grinded broken capsules
(
denoted as broken-Pt@S1), the positive effects of the intact zeo-
litic shell on the one-pot DKR of PEA are well confirmed.
2.4. Racemization catalytic reaction
2
. Experimental
The racemization reactions were performed at 70 °C under a
2
.1. Materials
hydrogen atmosphere, using 0.1 mmol of S-PEA, 2 mL of toluene
and 100 mg of Pt@S1 catalyst for 35 h. The Pt@S1 was isolated by
centrifugation and washed with toluene after reaction for further
characterizations. The processes were monitored by gas chroma-
tography (GC, HP 5890) with a CP-CHIRASIL-DEX CB chiral column
Polydiallyldimethylammonium chloride (PDDA, Mw = 200,000,
0 wt% in water), poly (sodium 4-styrene sulfonate) (PSS, Mw =
0,000), 3-amino-propyltriethoxysilane (APS) and Novozyme 435
1
7
Ò
(Candida antartica lipase B immobilized onto acrylic resin) were
(
25 m ꢂ 0.25
lm, CP7502) and FID detection using decane as an
internal standard. The enantiomeric excess (ee) of substrate and
purchased from Aldrich. ±-1-PEA (rac-PEA, 98%), S-1-PEA (98%)
and R-1-PEA (98%) were obtained from Alfa Aesar. Decane (>99%)
was achieved from Acros. Vinyl octanoate (98%) was purchased
from ABCR. Acetaldehyde (99.5%) was obtained from Adamas.
the selectivity of reaction (Selec.) were calculated as follows:
½
S-PEAꢃ ꢁ ½R-PEAꢃ
t
t
t
ee ¼
ꢂ 100%
ð1Þ
ð2Þ
½
S-PEAꢃ þ ½R-PEAꢃ
H
PtCl ꢀ6H
2 6 2
O and octanoic acid were supplied by Sinopharm
(94 wt%), tetrapropylammoni-
t
Chemical Reagent Company. KBH
4
½
R-PEAꢃ þ ½S-PEAꢃ
um hydroxide (TPAOH, 25 wt% in water), tetrapropyl ammonium
bromide (TPABr) and tetraethoxy-orthosilane (TEOS) were ob-
tained from Shanghai Chemical Reagent Company. Toluene
t
t
Selec: ¼
ꢂ 100%
½S-PEAꢃ
0
(
>99%) and sodium hydroxide (NaOH) were achieved from Shang-
2
.5. DKR catalytic reaction
hai Dahe Chemical Reagent Company. All chemicals were used
without further purification.
Typical DKR reactions were performed under the same condi-
tion as above but with 0.1 mmol of rac-PEA, R-PEA or S-PEA as
reactants, 50 mg of Novozyme 435 as acylation catalyst and
2.2. Catalyst preparation
Ò
0
.2 mmol of vinyl octanoate as acyl donor. After cooling to room
The Pt@S1 catalyst was synthesized according to the hydrother-
temperature, the catalysts were isolated by centrifugation and
washed with toluene for further characterizations and applica-
tions. The products were analyzed by the GC as mentioned above,
and the by-products were characterized by high-resolution mass
spectrometer (HRMS) with a Finnigan MAT 95 (San Jose, CA;
USA) double-focusing magnetic sector mass spectrometer.
The conversion of substrate (Conv.) as well as ee(product), selectiv-
ity (Selec.(product)) and yield of product R-amide were calculated by
the Eqs. (3)–(6), respectively.
mal transformation as described in our previous reports [21]. The
mesoporous silica spheres (MSSs, prepared according to Grun
et al. [41]) were used as templates and modified with APS at room
temperature for 8 h. After the impregnation in the aqueous solu-
tion of 0.4 M H
MSSs) were in situ reduced with 0.01 M KBH
2
PtCl
6
for 30 min, the platinum-loaded MSSs (Pt-
aqueous solution
4
and calcined in air at 550 °C for 3 h. Afterward, the particles were
seeded with 70 nm silicalite-1 through a layer-by-layer procedure
and treated with hydrothermal transformation in the sol with the
composition of TPABr:H
2
O:NaOH = 18:2000:0.2 for 8 h [42]. The
½Productꢃ
ꢁ ½ProductꢃðS;tÞ
ðR;tÞ
eeðproductÞ
¼
ꢂ 100%
ð3Þ
ð4Þ
ZMC with Pt-encapsulated was separated by centrifugation and
washed with distilled water for three times. To remove the organic
ingredients, the obtained samples were calcined in air at 550 °C for
½
ProductꢃðR;tÞ þ ½Productꢃ
ðS;tÞ
½
Substrateꢃt
Conv
: ¼ 1 ꢁ
ꢂ 100%
0
½
Substrateꢃ
6
h. Finally, a reduction process was adopted in a continuous
½
Productꢃ
hydrogen flow at 200 °C to obtain the metallic Pt-encapsulated
Pt@S1 catalyst. Some of the as-synthesized Pt@S1 was heavily
grinded and used as the reference sample.
t
Yield ¼
ꢂ 100%
ð5Þ
ð6Þ
½Substrateꢃ
0
Selec:ðproductÞ ¼ Yield=Con
v
:
2.3. Catalyst characterization
3
. Results
Scanning electronic microscopic (SEM) and energy dispersive
spectroscopy (EDS) of the catalysts were recorded on Philips
XL30 scanning electron microscope. Transmission electronic
microscopic (TEM) images were obtained on JEOL JEM-2010 instru-
ment. The X-ray diffraction (XRD) patterns were tested on a Rigaku
3.1. Characterizations of catalysts
Fig. 1 depicts the typical SEM and TEM images of intact-Pt@S1.
Obviously, the catalyst shows a uniform spherical morphology

J. Shi et al. / Journal of Catalysis 291 (2012) 87–94
89
Fig. 1. (A) SEM, (B) TEM and (C) high-resolution TEM images of intact-Pt@S1 and (D) size distribution for the encapsulated Pt nanoparticles. The insets in (A) are the images of
zeolitic shell lattice fringes.
(
1.6 lm in diameter) with a closely packed silicalite-1 nanocrystal
on the surface (Fig. 1A). The lattice of silicalite-1 indicates good
crystallinity of the shell, as seen in the inset of Fig. 1A. The hollow
interior as well as the dense shell with the thickness of ca. 150 nm
can be clearly observed in the TEM image of Fig. 1B. Fig. 1C pre-
sents the highly dispersed nanoparticles of Pt inside the capsules.
The size distribution of Pt nanoparticles (Fig. 1D, counted with
more than one hundred nanoparticles from five pictures of HRTEM
of Pt@S1) indicates that they possess a relatively narrow size range
with the most probable size of about 10 nm. The platinum content
in microcapsule is tested as 3.4 wt% by ICP-AES analysis. The XRD
pattern further proves the composition of the intact-Pt@S1, and the
sets of peaks attributed to platinum and silicalite-1 are indicated in
Fig. 2A-a. To further determine the chemical state of the encapsu-
lated platinum, the XPS operation is conducted after heavily grind-
ing intact-Pt@S1 to expose the Pt component. The binding energies
of the Pt 4f band are 71.2 and 74.2 eV (Fig. 2B), respectively, sug-
gesting that platinum in the microcapsule is existed as metallic
state [43]. A representative N
tact-Pt@S1 are shown in Fig. 3. The specific surface area calculated
by Brunaue–Emmett–Teller (BET) method at P/P of 0.06–0.2 and
2
sorption isotherm and t-plot of in-
0
2
the micropore volume calculated by t-plot method are 343 m /g
and 0.1 cm /g, respectively, manifesting the good crystallinity
3
and open microporosity of zeolite shell [44].
3.2. The catalytic behavior of Pt@S1 in racemization reaction
Fig. 2. (A) XRD patterns for intact-Pt@S1 (a) before and (b) after racemization
ꢄ
The racemization performance of Pt@S1 was first tested in the
reaction. The characteristic diffractions attributed to zeolite are indicated by . (B)
XPS Pt 4f spectra of grinded Pt@S1.
racemization of S-PEA. As shown in Fig. 4b, the racemization reac-
tion could carry out successfully on Pt@S1 with the selectivity al-
ways higher than 99% (Fig. 4d), indicating almost no side
reaction occurred in this process. Due to the high selectivity, the
ee values of substrate are used directly to express the catalytic
activity of Pt@S1 for racemization. For comparison, the catalytic
performance of broken-Pt@S1 catalyst prepared by heavily

9
0
J. Shi et al. / Journal of Catalysis 291 (2012) 87–94
Fig. 5. The (a) yield, (b) conversion and (c) selectivity vs reaction time by using
intact-Pt@S1/Novozyme 435 as catalysts. The DKR reaction was performed at 70 °C
Fig. 3. Representative N
2
sorption isotherm and t-plot (inset) of intact-Pt@S1.
Ò
under hydrogen atmosphere for 35 h, with 0.1 mmol of rac-PEA, 50 mg of
Ò
Novozyme 435, 0.2 mmol of vinyl octanoate, 2 mL of toluene and 100 mg of
grinding the intact ones is measured under the same conditions,
and it is observed that the broken-Pt@S1 catalysts exhibit a similar
behavior to intact ones (intact-Pt@S1), illustrating the negligible
influence of the zeolitic shell on the diffusion of the reactant
intact-Pt@S1.
faster rate for R-PEA than that for S-PEA during the same period,
which may be attributed to their different reaction paths as dis-
cussed later.
(Fig. 4a and b). Besides, compared to that of the fresh intact-
Pt@S1 (Fig. 2A-a), the complete maintenance of its XRD diffraction
peaks after reaction suggests the high stability of the microcapsul-
ar catalyst during the reaction (Fig. 2A-b).
Ò
The stability of the intact-Pt@S1/Novozyme 435 catalysts was
measured for DKR reaction as well (Table 1). Both catalytic activity
and selectivity remain almost unchanged for at least five runs
without detectable leaching of Pt, suggesting the good reusability
of the ZMC and the enzyme. According to our previous reports
3.3. The catalytic behavior of intact-Pt@S1 in one-pot DKR
[
21,22], this promising recycle ability may be attributed to the
protection of the zeolitic shell and will be further discussed in
Section 4.
The platinum-encapsulated zeolitically microcapsular catalysts
(
intact-Pt@S1), associated with commercially available Novo-
Ò
Ò
zyme 435 (intact-Pt@S1/Novozyme 435), were then evaluated in
the one-pot DKR of rac-PEA by using vinyl octanoate as acyl donor,
and the results were displayed in Fig. 5. It can be clearly found that
the whole DKR process for the rac-PEA is able to proceed smoothly
with prolonging reaction time. After 35 h, the conversion of rac-
PEA and the yield of R-amide could reach up to 80% (Fig. 5b) and
3.4. The necessity of the zeolitically microcapsular structure for DKR
process
To confirm the necessity of the zeolitic shell on DKR process,
the broken-Pt@S1 prepared by heavily grinding the intact-Pt@S1
was also adopted in this procedure with the association of
7
7% (Fig. 5a), respectively. It is worth mentioning that the DKR pro-
Ò
cedure driven by intact-Pt@S1/Novozyme 435 displays an out-
standing selectivity as high as 95% (Fig. 5c) of the target product,
and an ee(product) always higher than 99% throughout the whole
process.
Ò
Ò
Novozyme 435 (donated as broken-Pt@S1/Novozyme 435).
Fig. 7 presents the reaction result of rac-PEA in the system of
Ò
broken-Pt@S1/Novozyme 435. Its selectivity (Fig. 7c, ꢅ60%) is
Ò
much lower than that of the intact-Pt@S1/Novozyme 435 catalysts
Fig. 6 further presents the DKR performances of intact-Pt@S1/
(
Fig. 5c, ꢅ95%), although its initial conversion is almost similar to
Ò
Novozyme 435 with pure S-PEA or R-PEA as reactants under the
that of the latter. Even worse, after reaction of 10 h, both the con-
version and the selectivity for the broken system keep unchanged
and remain at only 38% and 61%, respectively, indicating no further
reactions occurred after that time. The same behaviors are also
observed on both R-PEA and S-PEA. The reactions terminate after
same conditions. Both R-PEA and S-PEA exhibit similar tendency
to rac-PEA, that is, they also show a high product selectivity (c.a.,
5%, Fig. 6c and d) and an increasing conversion with reaction time
Fig. 6a and b). However, the conversion of R-PEA reaches 100% at
5 h, while that of S-PEA just reaches 65% at 35 h, indicating a
9
(
2
Fig. 6. The DKR (a and b) conversion and (c and d) selectivity of (a and c) S-PEA and
Ò
Fig. 4. The (a and b) ee values of substrate and (c and d) reaction selectivity vs
reaction time on (a and c) broken-Pt@S1 and (b and d) intact-Pt@S1. The
racemization reactions were performed at 70 °C under a hydrogen atmosphere
for 35 h, using 0.1 mmol of S-PEA, 2 mL of toluene and 100 mg of catalyst.
(b and d) R-PEA with reaction time by using intact-Pt@S1/Novozyme 435 as
catalysts. The DKR reaction was performed at 70 °C under hydrogen atmosphere for
35 h, with 0.1 mmol of S-PEA or R-PEA, 50 mg of Novozyme 435, 0.2 mmol of vinyl
Ò
octanoate, 2 mL of toluene and 100 mg of intact-Pt@S1.

J. Shi et al. / Journal of Catalysis 291 (2012) 87–94
91
Table 1
The reusability of intact-Pt@S1/NovozymeÒ435 in DKR process.
Recycling numbera
Conv. (%)
Selec. (%)
ee(product) (%)
1
2
3
4
5
80
77
78
77
76
93
89
90
89
87
>99
>99
>99
>99
>99
a
DKR reaction was performed at 70 °C under hydrogen atmosphere, with
Ò
0
.1 mmol of rac-PEA, 50 mg of Novozyme 435, 0.2 mmol of vinyl octanoate, 2 mL of
toluene and 100 mg of intact-Pt@S1.
1
0 h (Fig. 8b and c), and the final results of S-PEA (Conv. = 22% and
Fig. 8. The DKR (b and c) conversion and (a and d) selectivity of (a and b) S-PEA and
Selec.(product) = 7%, Fig. 8a and b) and R-PEA (Conv. = 42% and
Selec.(product) = 70%, Fig. 8c and d) are all much lower than those
of intact-Pt@S1/Novozyme 435 (Fig. 6).
Ò
(
c and d) R-PEA with reaction time by using broken-Pt@S1/Novozyme 435 as
catalysts. The DKR reaction was performed at 70 °C under hydrogen atmosphere for
Ò
Ò
35 h, with 0.1 mmol of S-PEA or R-PEA, 50 mg of Novozyme 435, 0.2 mmol of vinyl
octanoate, 2 mL of toluene and 100 mg of broken-Pt@S1.
4
. Discussion
As found in the experimental results, the microcapsular
structure is crucial for the good performance of the DKR. The in-
tact-Pt@S1 system with the protection of the intact silicalite-1
shell displays high activity, high selectivity and outstanding reus-
ability, whereas the broken one without the shell protection pre-
sents poor results. The scheme 1 illustrates the possible reaction
network in our systems. During the process, the rac-PEA would re-
act with an acyl donor through a catalytic acylation process on the
Scheme 1. The reaction network for the whole DKR process.
Ò
enzyme of Novozyme 435. However, the rate of this transforma-
the broken system may imply the deactivation of enzyme
in this case. The similar conclusion can also be drawn from
the experimental results of R-PEA. Its final conversion on
tion for R-PEA is much greater than that for S-PEA [45] (proved
by the ee(product) > 99%, that is, negligible acylation of S-PEA), and
this sharp difference will lead to a concentration gradient of these
two enantiomers. Consequently, there is a high tendency for S-PEA
to be converted to R-PEA with the assistance of racemization cata-
lyst (Pt@S1 herein), and finally, the optically pure enantiomer can
be obtained with an ideal yield of 100%, supposing no by-products
generated.
Ò
intact-Pt@S1/Novozyme 435 can reach 100% after 25 h,
Ò
while this parameter on broken-Pt@S1/Novozyme 435 is
only 42%.
Ò
(
(
2) The intact-Pt@S1/Novozyme 435 with intact shell has a
much higher selectivity than the broken one. For example,
the selectivity is as high as 95% for rac-PEA on the intact-
By comparing the proposed scheme with the experimental data,
some information could be obtained:
Ò
Pt@S1/Novozyme 435 but that of the broken system is at
most 61%. The decrease in selectivity might indicate the
occurrence of some side reactions (Scheme 1).
Ò
(
1) The highest conversion of intact-Pt@S1/Novozyme 435 for
3) Theoretically, if the platinum could keep its catalytic activity
during the whole reaction process, the conversion of PEA
will continuously increase with the reaction time through
either racemization reaction or side reactions. However,
both the conversion and the selectivity on the microcapsular
catalyst with broken shell structure keep unchanged after
reaction of 10 h, as shown in Figs. 7 and 8. Such phenome-
non means that platinum may also lose its activity for race-
mization reaction and side reaction when it is exposed in the
system.
Ò
rac-PEA is 80%, while that of broken-Pt@S1/Novozyme 435
is only 38%. According to the mechanism shown in scheme
Ò
1
, if the Novozyme 435 performs regularly, the conversion
of rac-PEA should be 50% in the DKR process even if the race-
mization catalyst does not work, that is, the R-PEA compo-
nent is completely acylated. The much lower conversion of
4.1. Inhibition of zeolitically microcapsular structure for side reactions
To explain the poor selectivity on catalyst without the protec-
Ò
tion of the zeolite shell, that is, broken-Pt@S1/Novozyme 435, a
HRMS experiment was conducted to analyze the products, and
one by-product of N-ethyl-1-phenylethanamine was identified,
which might be generated through the reductive amination pro-
cess [46]. Referred to the nearly 100% racemization selectivity of
broken-Pt@S1 in Fig. 4c, it should be the addition of the vinyl octa-
noate that causes series of side reactions.
Along with the DKR process, the acyl donor of vinyl octanoate
would produce enol-like intermediate and octanoic acid during
the reaction. It is known that the vinyl alcohol is unstable in the
solution and prefers to undergo the keto-enol tautomerization,
leading to the formation of acetaldehyde [46]. Then, the amine in
Fig. 7. The (a) yield, (b) conversion and (c) selectivity vs reaction time by using
broken-Pt@S1/Novozyme 435 as catalysts. The DKR reaction was performed at
Ò
7
0 °C under hydrogen atmosphere for 35 h, with 0.1 mmol of rac-PEA, 50 mg of
Ò
Novozyme 435, 0.2 mmol of vinyl octanoate, 2 mL of toluene and 100 mg of
broken-Pt@S1.

9
2
J. Shi et al. / Journal of Catalysis 291 (2012) 87–94
the system could condense with the acetaldehyde to form the
imine via a nucleophilic addition process in the presence of proper
acid [46]. Under the atmosphere of hydrogen, reduction in the
imine on Pt would give the by-product of N-ethyl-1-phenylethan-
amine. However, the formation process of imine is greatly depen-
dent on the acidity of the system, since an appropriate acidity has
to be provided to give a reasonable concentration of the protonated
aldehyde or ketone for the nucleophilic addition reaction [46]. In
this system, octanoic acid should be the sole acid source that can
promote the side reaction. For the intact-Pt@S1, the more sterically
demanding octanoic acid molecules may be difficult to diffuse into
the microcapsule due to the space confinement of the micropore
on zeolitic shell. As a result, a relatively neutral medium could be
formed inside the intact-Pt@S1, limiting the formation of imine.
Then, the successive reduction in the imine on Pt in the capsules
would become meaningless, ensuring a high selectivity of the
DKR. On the other hand, once the microcapsule is crushed, the con-
fined effect of octanoic acid is excluded and the by-product could
be easily formed in the open system with the help of octanoic acid,
rendering the catalyst with a low selectivity of product.
The substrate has to go through a tandem reaction of racemization
on Pt firstly and then acylation on Novozyme 435 to form the final
product of R-amide. This process greatly increases the possibility to
form the by-products on Pt surface. As a result, the selectivity for
S-PEA is lower than that for R-PEA for the broken-Pt@S1.
Ò
Therefore, the protective zeolitic shell of the intact-Pt@S1/
Ò
Novozyme 435 functions just as a ‘fence’ that can successfully lim-
it the generation of by-products and ensure the high selectivity of
the DKR process for all kinds of enantiomers (Figs. 5 and 6).
4.2. Inhibition of zeolitically microcapsular structure for the
deactivation of the catalysts in DKR process
As shown from the experiment results, without the protection
of the zeolitic shell, both main and side reactions in DKR process
will cease within 10 h and the final conversion and selectivity of
Ò
substrates on the broken-Pt@S1/Novozyme 435 are all much low-
er than that with intact ones. According to the mechanism in
Scheme 1, they could be attributed to the deactivation of both en-
zyme and Pt catalysts.
This explanation of the differences in product selectivity can be
validated by three controlled experiments. When octanoic acid or
acetaldehyde is added into the DKR system individually instead
of vinyl octanoate with the catalyst of only broken-Pt@S1, no by-
product could be detected in the reaction system by GC, and the
solution is colorless after reaction of 35 h. However, when both
octanoic acid and acetaldehyde are synchronously added instead
of vinyl octanoate, the solution color changes to light yellow after
reaction and the GC analysis shows the formation of by-product.
Scheme 2 illustrates these differences of the side reaction paths be-
tween intact- and broken-Pt@S1 systems.
The solid-state near UV circular dichroism (CD) was applied to
Ò
Novozyme 435 before and after reaction (Fig. 9). The fresh Novo-
Ò
zyme 435 presents a CD band at around 270 nm, which is consis-
tent with the literature [47]. This band is almost unchanged in the
system involving intact-Pt@S1 after the reaction. However, when
the broken-Pt@S1 is applied, the band shifts from 270 nm to
2
25 nm, illustrating the alteration in the tertiary structure of en-
zyme [48]. Referring to previous reports, the Pt species in the bro-
ken-Pt@S1 can easily leach out from the microcapsule [21,22], and
Ò
then may migrate to the Novozyme 435 and interact with the N or
S atoms in it, leading to the deactivity of the enzyme [49]. Such
interaction of Pt with N and S, in turn, will also decrease the activ-
ity of Pt species. This migration of Pt species could be experimen-
tally proved by EDS measurement on the Pt@S1 and
The occurrence of side reaction on Pt can also be applied to ex-
plain the even worse selectivity for S-PEA than that for R-PEA in
the broken-Pt@S1 systems (Fig. 8a and d). As indicated in Scheme
1
, when R-PEA is used as reactant, the racemization reaction on Pt
Ò
Novozyme 435. It is found that the intact-Pt@S1 can keep 91% of
Ò
and the acylation reaction on Novozyme 435 would occur in par-
allel. Thus, considerable part of R-PEA would be directly acylated to
its corresponding amide, leading to the relative lower possibility of
side reactions on Pt. On the contrary, the direct acylation of S-PEA
is negligible because of the exclusive stereoselectivity of enzyme.
the original Pt content after reaction, while the broken sample
can only keep 18% of Pt comparing to the original catalyst. Corre-
Ò
spondingly, the mass ratio of Pt/(C + N + S + Pt) on Novozyme 435
Ò
in the intact-Pt@S1/Novozyme 435 system is only 0.1% after reac-
tion, whereas the datum increases to 1.4% when the broken micro-
capsules are used.
A two-step simulative experiment was further designed to con-
Ò
firm the Pt migrating from broken-Pt@S1 to Novozyme 435 during
the DKR process. In step I, only Pt@S1 was put into the reaction
system, and in step II, after filtrating out the ZMC, the Novo-
Ò
zyme 435 was then added into the solution to trigger the reaction.
In this two-step experiment, the racemization and acylation
Ò
Scheme 2. The schematic illustration of the inhibition of zeolitically microcapsular
Fig. 9. The solid-state near UV CD spectrum of Novozyme 435 (a) before and after
structure for the side reactions.
reaction with (b) intact-Pt@S1 and (c) broken-Pt@S1 catalysts.

J. Shi et al. / Journal of Catalysis 291 (2012) 87–94
93
processes are separated each other, and the transferring of Pt spe-
cies could be easily monitored by chemical composition analysis.
For the convenience of the ICP-AES analysis for solution, an aque-
ous system is used instead. The Pt concentration in the solution
after filtrating intact-Pt@S1 is lower than the limit of detection,
the active Pt species but also prohibits the direct contacting of en-
zyme with Pt species encapsulated in the ZMC, which ensures the
activity and reusability of the catalysts. More importantly, the zeo-
litic shell provides a protected space that greatly hinders the for-
mation of by-product and gives rise to the higher DKR selectivity.
With these advantages, this functional zeolitcally microcapsular
catalyst may further find applications in the relevant field.
while that for broken-Pt@S1 is 1.340 mg/mL, corresponding to
Ò
7
9% of initial Pt content. However, when the fresh Novozyme 435
is added in the second step, the Pt concentration in the solution for
broken-Pt@S1 dramatically decreases to 0.062 mg/mL, meaning
Acknowledgments
Ò
near 95% of Pt has transferred to the Novozyme 435. The strong
Ò
adsorption of Pt on Novozyme 435 could also be certified by EDS
This work is supported by NSFC (21073041, 30828010,
measurement for broken-Pt@S1, in which the mass ratio of Pt/
2
0873025 and 20890122), STCSM (10QH1400300, 08DZ2270500
and 09DZ2271400), and 973 and 863 Programs (2009CB930400,
009CB623506 and 2009AA033701).
Ò
(
C + N + S + Pt) on Novozyme 435 isolated after reaction is about
1
.3%.
Taking the changes in the CD band of Novozyme 435 on differ-
2
Ò
ent catalysts and the migration of Pt from the broken-Pt@S1 to
References
Ò
Novozyme 435 into account, it can be concluded that there is a
Ò
strong interaction between the Pt species and Novozyme 435,
[
1] C.S. Peyratout, L. Dähne, Angew. Chem. Int. Ed. 43 (2004) 3762.
and such interaction leads to the denaturation of enzyme and the
deactivation of Pt. Thus, the reactions terminate in 10 h when the
[2] A.N. Zelikin, Q. Li, F. Caruso, Angew. Chem. Int. Ed. 45 (2006) 7743.
[
[
3] A.F. Faria, R.A. Mignone, M.A. Montenegro, A.Z. Mercadante, C.D. Borsarelli, J.
Agric. Food Chem. 58 (2010) 8004.
4] M. Zandi, S.A. Hashemi, P. Aminayi, J. Appl. Polym. Sci. 1 (2011) 586.
Ò
broken-Pt@S1/Novozyme 435 is used. For the reaction with in-
Ò
tact-Pt@S1/Novozyme 435, it is the protection of the zeolitic shell
[5] A. Alexeev, R. Verberg, A.C. Balazs, Langmuir 23 (2007) 983.
[6] F. Liu, L.X. Wen, Z.Z. Li, W. Yu, H.Y. Sun, J.F. Chen, Mater. Res. Bull. 41 (2006)
Ò
that isolates the Pt and Novozyme 435 separately at the inner and
2268.
outer regions of microcapsule and makes the racemization and
acylation reactions get along well with each other. Such protection
effect also ensures the high reusability of intact-Pt@S1/Novo-
[
[
7] B. Samanta, X.C. Yang, Y. Ofir, M.H. Park, D. Patra, S.S. Agasti, O.R. Miranda, Z.H.
Mo, V.M. Rotello, Angew. Chem. Int. Ed. 48 (2009) 5341.
8] Y. Zhao, L. Jiang, Adv. Mater. 21 (2009) 3621.
Ò
[9] A. Zabet-Khosousi, A.A. Dhirani, Chem. Rev. 108 (2008) 4072.
10] H.F. Zhang, I. Hussain, M. Brust, M.F. Butler, S.P. Rannard, A.I. Cooper, Nat.
Mater. 4 (2005) 787.
11] A. Corma, Chem. Rev. 97 (1997) 2373.
12] R.H. Baughman, A.A. Zakhidov, W.A. de Heer, Science 297 (2002) 787.
13] O. Kreft, A.G. Skirtach, G.B. Sukhorukov, H. Möhwald, Adv. Mater. 19 (2007)
zyme 435 catalyst by preventing the leakage of the Pt and the
[
deactivation of both catalytic components.
[
[
[
4.3. Conversions of different enantiomers
3142.
[
[
14] X.W. Lou, C. Yuan, L.A. Archer, Adv. Mater. 19 (2007) 3328.
15] J. Wang, H. Zhang, X. Yang, S. Jiang, W. Lv, Z. Jiang, S.Z. Qiao, Adv. Funct. Mater.
Because of the well synergic catalysis of platinum inside and
Ò
Novozyme 435 outside of the zeolitic shell, the intact-Pt@S1/
21 (2011) 971.
Ò
Novozyme 435 system presents promising performance during
[16] S. Ikeda, S. Ishino, T. Harada, N. Okamoto, T. Sakata, H. Mori, S. Kuwabata, T.
Torimoto, M. Matsumura, Angew. Chem. Int. Ed. 45 (2006) 7063.
[
[
[19] D.G. Shchukin, G.B. Sukhorukov, H. Möhwald, Angew. Chem. Int. Ed. 42 (2003)
4471.
[
the DKR of PEA. The catalytic behaviors of various reactants on
Pt@S1 can also be well illustrated on the basis of the Scheme 1
and the discussion above.
17] J. Choi, H.Y. Yang, H.J. Kim, U.S. Son, Angew. Chem. Int. Ed. 49 (2010) 7718.
18] K. Kamata, Y. Lu, Y. Xia, J. Am. Chem. Soc. 125 (2003) 2384.
Ò
For intact-Pt@S1, both the Novozyme 435 and the platinum
20] G.H. Yang, J.J. He, Y. Zhang, Y. Yoneyama, Y.S. Tan, Y.Z. Han, T. Vitidsant, N.
Tsubaki, Energy Fuels 22 (2008) 1463.
could keep their activity well and the one-pot DKR process could
proceed successfully. Therefore, the conversion is continuously
increasing with high selectivity as the reaction proceeding (Figs.
[21] N. Ren, Y.H. Yang, J. Shen, Y.H. Zhang, H.L. Xu, Z. Gao, Y. Tang, J. Catal. 251
2007) 182.
[
[
(
22] N. Ren, Y.H. Yang, Y.H. Zhang, Q.R. Wang, Y. Tang, J. Catal. 246 (2007) 215.
23] X. Li, J. He, M. Meng, Y. Yoneyama, N. Tsubaki, J. Catal. 265 (2009) 26.
5
and 6). More interestingly, the increasing rates of conversion
with time are different for various amines, that is, R-PEA > rac-
PEA > S-PEA. This phenomenon can be well explained by Scheme
[24] G. Yang, N. Tsubaki, J. Shamoto, Y. Yoneyama, Y. Zhang, J. Am. Chem. Soc. 132
(2010) 8129.
[
[
[
25] T.C. Nugent, M. El-Shazlya, Adv. Synth. Catal. 352 (2010) 753.
26] V. Farina, J.T. Reeves, C.H. Senanayake, J.J. Song, Chem. Rev. 106 (2006) 2734.
27] Y. Kim, J. Park, M.J. Kim, Tetrahedron Lett. 51 (2010) 5581.
1
. The R-PEA can be acylated to its corresponding amide directly
Ò
on Novozyme 435 but S-PEA cannot. When the initial contents
of total PEA in the three above systems are the same, the initial
concentration of the R-PEA is actually different, thus acylation pro-
[28] D. Koszelewski, B. Grischek, S.M. Glueck, W. Kroutil, K. Faber, Chem. Eur. J. 17
2011) 378.
(
[
29] M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Keßeler, R. Stürmer, T. Zelinski,
Angew. Chem. Int. Ed. 43 (2004) 788.
Ò
cess on the Novozyme 435 becomes different. Thereafter, the rates
of DKR process are varied during the same period and give rise to
various conversions. This difference for the rate of the conversion
also appeared for the broken systems during the first 10 h (Figs.
[30] C. Ma, X.L. Xu, P. Ai, S.M. Xie, Y.C. Lv, H.Q. Shan, L.M. Yuan, Chirality 23 (2011)
79.
31] A. Kamal, M.A. Azhar, T. Krishnaji, M.S. Malik, S. Azeeza, Coord. Chem. Rev. 252
2008) 569.
[32] A. Parvulescu, J. Janssens, J. Vanderleyden, D. De Vos, Top. Catal. 53 (2010) 931.
3
[
(
7
and 8), although their conversion and selectivity were much low-
[
[
33] A. Parvulescu, D. De Vos, P. Jacobs, Chem. Commun. 42 (2005) 5307.
34] J.S.M. Samec, J.E. Bäckvall, P.G. Andersson, P. Brandt, Chem. Soc. Rev. 35 (2006)
er than the intact ones. However, since the deactivation of the cat-
alysts took place as indicated above, the conversions for all the
systems were remained unchanged finally.
237.
[35] A.N. Parvulescu, P.A. Jacobs, D.E. De Vos, Chem. Eur. J. 13 (2007) 2034.
[
36] H.P. Liang, H.M. Zhang, J.S. Hu, Y.G. Guo, L.J. Wan, C.L. Bai, Angew. Chem. Int.
Ed. 43 (2004) 1540.
5
. Conclusions
[37] L.G. Hutchins, R.E. Pincock, J. Org. Chem. 47 (1982) 607.
[
38] E.J. Ebbers, G.J.A. Ariaans, J.P.M. Houbier, A. Bruggink, B. Zwanenburg,
Tetrahedron 53 (1997) 9417.
In this work, we have explored the feasibility of the ZMC in the
[
39] Y. Govender, T.L. Riddin, M. Gericke, C.G. Whiteley, J. Nanopart. Res. 12 (2010)
261.
Ò
DKR process. The intact-Pt@S1/Novozyme 435 has been found as a
promising catalyst for DKR of PEA, and they display a high selectiv-
ity of product with the conversion of 80% as well as a high stability
and reusability. The zeolitic shell not only prevents the leakage of
[
[
40] B. Martín-Matute, J.-E. Backväll, Curr. Opin. Chem. Biol. 11 (2007) 226.
41] M. Grun, C. Buchel, D. Kumar, K. Schumacher, B. Bidlingmaier, K.K. Unger, Stud.
Surf. Sci. Catal. 128 (2000) 155.
[42] J. Shi, N. Ren, Y.H. Zhang, Y. Tang, Micropor. Mesopor. Mater. 132 (2010) 181.

9
4
J. Shi et al. / Journal of Catalysis 291 (2012) 87–94
[
[
43] R. Li, W. Chen, H. Kobayashi, C. Ma, Green Chem. 12 (2010) 212.
44] A.G. Dong, Y.J. Wang, D.J. Wang, W.L. Yang, Y.H. Zhang, N. Ren, Z. Gao, Y. Tang,
Micropor. Mesopor. Mater. 64 (2003) 69.
45] R. Stürmer, Angew. Chem. Int. Ed. 36 (1997) 1173.
46] F.A. Carey, Organic Chemistry, fourth ed., The McGraw-Hill Companies, United
States of America, 2000.
[47] A. Ganesan, B.D. Moore, S.M. Kelly, N.C. Price, O.J. Rolinski, D.J.S. Birch, I.R.
Dunkin, P.J. Halling, ChemPhysChem 10 (2009) 1492.
[48] M.A. Andrade, P. Chacón, J.J. Merelo, F. Morán, Protein Eng. 6 (1993) 383.
[49] J.R. Chang, S.L. Chang, T.B. Lin, J. Catal. 169 (1997) 338.
[
[