Activity, recyclability, and stability of lipases immobilized on oil-filled spherical silica nanoparticles with different silica shell structures

Article abstract of DOI:10.1002/cctc.201300072

Candida antarctica lipaseA was immobilized on spherical silica nanoparticles with oil-filled core and oil-induced mesoporous silica shell with different silica shell structures. The immobilization of enzymes was achieved by directly adding enzymes to the oil-in-water emulsion system under ambient synthesis conditions, and the silica shell structure was controlled by the addition of the cosolvent ethanol to the initial synthesis medium. Detailed structural analysis revealed the formation of oil-filled spherical silica nanoparticles with 3.4-4.2nm mesopores randomly arranged in the silica shell; the thickness and pore characteristics of these pores markedly changed with the addition of ethanol. The retention of the enzyme activity during biocatalysis was significantly affected by the structural properties of the silica shells, and it was found that a thick and dense silica shell is essential to afford an active, recyclable, and stable biocatalyst. Furthermore, the oil encapsulated within the core cavity was found to play an important role in achieving a high catalytic efficiency. Trapped oil: Candida antarctica lipaseA is immobilized on oil-filled spherical silica nanoparticles with different silica shell structures through an anionic surfactant-induced self-assembly approach (see scheme) with ethanol as a cosolvent. The entrapped enzymes mostly retain their activities and exhibit recyclability and thermal and chemical stability, depending on the thickness and pore characteristics of the silica shells. TEOS=Tetraethoxyorthosilicate, APTES=3-aminopropyl triethoxysilane.

Full text of DOI:10.1002/cctc.201300072

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201300072
Activity, Recyclability, and Stability of Lipases Immobilized
on Oil-Filled Spherical Silica Nanoparticles with Different
Silica Shell Structures
Yasutaka Kuwahara,[a] Takato Yamanishi,[a] Takashi Kamegawa,[a] Kohsuke Mori,[a, b] and
Hiromi Yamashita*[a, b]
Candida antarctica lipase A was immobilized on spherical silica
nanoparticles with oil-filled core and oil-induced mesoporous
silica shell with different silica shell structures. The immobiliza-
tion of enzymes was achieved by directly adding enzymes to
the oil-in-water emulsion system under ambient synthesis con-
ditions, and the silica shell structure was controlled by the ad-
dition of the cosolvent ethanol to the initial synthesis medium.
Detailed structural analysis revealed the formation of oil-filled
spherical silica nanoparticles with 3.4–4.2 nm mesopores ran-
domly arranged in the silica shell; the thickness and pore char-
acteristics of these pores markedly changed with the addition
of ethanol. The retention of the enzyme activity during bioca-
talysis was significantly affected by the structural properties of
the silica shells, and it was found that a thick and dense silica
shell is essential to afford an active, recyclable, and stable bio-
catalyst. Furthermore, the oil encapsulated within the core
cavity was found to play an important role in achieving a high
catalytic efficiency.
Introduction
Since the initial discovery of synthetic routes to ordered meso-
structured silicate materials, tremendous efforts have been di-
rected toward the development of synthetic approaches for
fabricating inorganic materials with controlled composition, hi-
erarchical structure, and multifunctionalities.[1–7] The use of
mesostructured silicate materials produced by using the surfac-
tant-induced self-assembly approach as hosts or carriers for
biocatalysts has been of great interest in a wide range of appli-
cations in biochemistry, such as drug-delivery systems,[8] bio-
sensors,[8,9] and enzymatic catalysis.[10,11] A fundamental motiva-
tion for the use of mesostructured silicate materials is to stabi-
lize biocatalysts by maintaining their fragile structural confor-
mations and inherent enzymatic activities and improve their
reusability, objectives to which they are ideally suited owing to
their tunable structures (e.g., through morphology, porosity,
and surface area) and high chemical stability.[10–12]
To date, various synthetic strategies have emerged for host-
ing enzymes in cavities or onto silicate materials. The over-
whelming majority for enzyme immobilization is chemical or
physical fixation within the pore channels of functionalized
mesoporous silica[9–26] or onto the void cavities of silica-based
macrocellular foam,[27,28] the pore diameters of which are ap-
proximate to, or more than, the molecular diameter of en-
zymes because they offer rigid, uniform open pore structures
and large pore volumes; however, the leaching of enzymes is
suspicious, and the robust confinement of enzymes is not
guaranteed in these open-ended silicate supports. Although
the uptake of enzymes and their stabilities can be improved
by enforcing the interactions between enzymes and the sili-
cate surfaces by using the chemical modification tech-
niques,[12–20,27–30] the complicated synthetic methods, coupled
with the high cost of these materials, limit their applicability
on the industrial scale. The design of high-performance hetero-
geneous biocatalysts through a facile and scalable immobiliza-
tion approach, by keeping the manufacturing cost low, still re-
mains a challenge.
[a] Dr. Y. Kuwahara,+ T. Yamanishi, Dr. T. Kamegawa, Dr. K. Mori,
Prof. Dr. H. Yamashita
Division of Materials and Manufacturing Science
Graduate School of Engineering
Of various silica-based host materials, silica hollow spheres
have been envisaged as promising carriers for biomolecules
owing to the presence of silica shell that functions as a physical
barrier to protect them from the surrounding environment and
a void volume to accommodate various guests.[31–33] For exam-
ple, Fujiwara et al. succeeded in directly encapsulating biomol-
ecules (bovine serum albumin and duplex DNA) within the
silica hollow spheres adopting an interfacial reaction by using
a multiple (W1/O/W2) emulsion system.[34] However, the integri-
ty of biomolecule properties upon the encapsulation has been
unclear. The direct addition of enzymes during the synthesis of
silicate materials seems to be an inexpensive, operationally
Osaka University
2-1, Yamada-oka, Suita, Osaka 565-0871 (Japan)
Fax: (+81)6-6879-7457
E-mail: yamashita@mat.eng.osaka-u.ac.jp
[b] Dr. K. Mori, Prof. Dr. H. Yamashita
Unit of Elements Strategy Initiative for Catalysts & Batteries
Kyoto University
Katsura, Kyoto 615-8510 (Japan)
[+] Present address:
Research Institute for Innovation in Sustainable Chemistry
National Institute of Advanced Industrial Science and Technology (AIST)
16-1 Onogawa, Tsukuba, Ibaraki 305-8569 (Japan)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201300072.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2527 – 2536 2527

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
simple, and extremely versatile
route for fabricating enzyme–
silica composite biocatalysts;
however, the synthetic method-
ology sometimes requires intri-
cate multiple preparation steps
and careful control of the syn-
thetic conditions to circumvent
the
denaturation
of
en-
zymes.[35,36]
The alternative strategy is
their encapsulation inside sol–
gel-derived silica matrices. These
materials (designated as silica-
coated enzymes) improve stabili-
Scheme 1. A formation mechanism of enzyme-immobilized oil-filled silica nanoparticles.
ties on enzymes because silica
matrices function to efficiently
protect the enzymes from the surrounding environment (e.g.,
pH and temperature) and against the denaturation/leaching of
enzymes during catalytic operations owing to the robust con-
finement effect.[37–42] However, most studies showed lower spe-
cific activity than that of the free enzymes owing to the prob-
lems associated with low diffusion rates and slow reaction ki-
netics derived from its nonporous structure. Thus, the silica
coating of enzymes to improve their stabilities is not commen-
surate with good accessibility. The key to design high-perfor-
mance enzyme–silica composite biocatalysts is likely an optimi-
zation of the accessibility between the reactive surface and the
enzymes via a precise control over the location of enzymes.
We recently reported a novel synthetic protocol for immobi-
lizing enzymes within spherical silica nanoparticles via an
anionic surfactant-induced self-assembly approach and demon-
strated that the resultant enzyme–silica composites act as effi-
cient heterogeneous biocatalysts in both aqueous and organic
solvents.[43] In comparison with the prototypical composite bio-
catalysts without a hierarchical core–shell structure, such as en-
zymes impregnated on mesoporous silicas and silica-coated
enzymes, the oil-filled silica nanoparticles (OSNs) with a unique
core–shell structure that we have developed are promising
supports for enzymes owing to the following fascinating fea-
tures: 1) ambient synthetic conditions that preserve the inher-
ent activity of enzymes, 2) a direct entrapment of enzymes via
a facile preparation route, which largely simplifies the manu-
facturing process, 3) a stabilizing effect endowed by the silica
shell, which enables increased stability and recyclability of en-
zymes, and 4) an optimized configuration of enzymes in the vi-
cinity of the reactive silica shell surface, which facilitates the
access of reactants.[43] The overall synthesis process of enzyme-
immobilized OSNs is illustrated in Scheme 1. In this synthetic
approach, oleic acid (OA), an anionic surfactant with a carboxyl-
ic acid group, immersed in water is emulsified initially by an
amino group of 3-aminopropyl triethoxysilane (APTES) with op-
posite charges, which self-assembles afterward to form meso-
porous silica shell through a localized polymerization of silicon
precursors (APTES and tetraethoxyorthosilicate [TEOS]) at the
water–oil interface, and the enzymes are sequestered selective-
ly within the silica matrix of the shell.[44] Although the as-syn-
thesized enzyme–silica composite was found to act as an effi-
cient, recyclable, and stable biocatalyst, the optimization of its
structure to maximize the catalytic activity of enzymes had not
been conducted. Some key questions related to the structural
optimization are as follows: 1) How is the silica shell structure
most suitable for efficiently immobilizing enzymes? 2) Is the
presence of oil in the core truly favorable for catalysis? 3) Is
the immobilized enzyme active and stable toward an applied
process? For example, the addition of specific amount of etha-
nol (EtOH) in the sol–gel process of mesoporous silica has a sig-
nificant effect on the silicate structures (e.g., particle size and
porosity) owing to the ability of EtOH to suppress the hydroly-
sis and condensation rate of TEOS[45] and to reduce the surfac-
tant packing parameter g.[44] An extensive exploration of suita-
ble structures of this composite biocatalyst may drastically
change the activity, recyclability, and stability of enzymes and
may provide constructive insights into the strategy for design-
ing high-performance enzyme–silica composite biocatalysts.
We have examined activity, recyclability, and stability of the
enzymes immobilized on OSN materials with different silica
shell structures and investigated the relationships between the
structures of the silicate support and the performances of the
immobilized enzymes. Similar to our previous report,[43] Candi-
da antarctica lipase A (Cal-A) was used as a model enzyme
throughout this work because it is one of the most recognized
biocatalysts in a broad range of synthetic applications of indus-
trial importance, such as kinetic resolutions, aminolysis, esterifi-
cation, and transesterification reactions.[28,46–48] The amount of
EtOH added during the synthesis was varied to give the final
products with different silica shell structures. The effect of
EtOH on enzyme properties was investigated by monitoring
change in pH during the syntheses and by comparing infrared
spectra. The structures such as morphology, particle size, and
porosity of the Cal-A-immobilized OSNs (Cal-A@OSN) were
carefully investigated by using SEM, TEM, XRD, and nitrogen
adsorption–desorption analysis. The residual enzyme activity,
recyclability, and thermal/chemical stability of the composite
biocatalysts were assessed by using the hydrolysis of n-butyl
acetate in water as a test reaction. Furthermore, the effect of
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2527 – 2536 2528

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
the presence of oil (OA) on the enzyme activity was
investigated.
Results and Discussion
Effect of EtOH on enzymes
As described in the Experimental Section, the immobilization
of Cal-A was performed by directly adding enzymes to the ini-
tial emulsion system composed of OA, water, and EtOH. After
the addition of silicon sources (TEOS and APTES), the resulting
solution was aged for 24 h at 508C to yield silica networks. A
higher hydrolysis and condensation rates of silicon sources can
undoubtedly be obtained at a higher temperature; however,
more moderate synthetic temperature, 508C, was adopted
here to circumvent denaturation of the enzymes by heat.
Another important factor that potentially affects enzyme
properties is change in pH during synthesis. The pH profiles
obtained during the synthesis of Cal-A@OSN(0) and Cal-
A@OSN(110) are shown in Figure 1. The solution containing
Figure 2. FTIR spectra of a) native Cal-A, b) Cal-A@OSN(0), c) Cal-A@OSN(110),
and d) Cal-A@OSN(0) calcined at 7008C in air.
Cal-A@OSN(110) were recorded and compared with that of
native Cal-A. As shown in Figure 2, native Cal-A demonstrated
several characteristic absorption bands; the bands seen at
1546 cmꢁ1 is assigned to the bending and stretching modes of
the NꢁH and CꢁN bonding of amide groups (amide II) and the
band at 1652 cmꢁ1 is assigned to the CꢁO stretching of the a-
helical conformation of Cal-A molecules (amide I).[10,49] The
peak intensities and positions of these amide bands are typi-
cally used for the determination of the conformation and un-
folding of the proteins.[50] The bands seen in the range of
1350–1490 cmꢁ1 are assigned to the ꢁCH2ꢁ and ꢁCH3 stretch-
ing modes of the aliphatic moieties of amino acid side chains
of enzymes.[10,49] The FTIR spectra of Cal-A@OSN(0) and Cal-
A@OSN(110) also demonstrated such bands characteristic of
the protein structure of Cal-A (the bands seen at 1406 and
1463 cmꢁ1 may also include those corresponding to the ali-
phatic chains of OA and APTES) as well as additional bands at
1712 cmꢁ1, which are assignable to the C=O bonding of the
carboxylic acid groups of OA. The presence of the amide I and
amide II bands confirms the immobilization of enzymes in
their structures. Nevertheless, the spectra of Cal-A@OSN(0)
showed distinct shifts of the amide I band toward 4 cmꢁ1
lower wave numbers (from 1652 to 1648 cmꢁ1) with a de-
creased intensity (Figure 2b). Our precedent study has con-
firmed that enzymes are selectively sequestered within the
silica matrix of the shell, not in the oil phase.[43] Thus, the pro-
tein structures of Cal-A are found to be perturbed by the con-
finement inside the silica matrix and do not have space for
changing their structural conformations. This shift was most
distinctly appeared in the spectrum of Cal-A@OSN(110) with
a 15 cmꢁ1 shift (from 1652 to 1637 cmꢁ1), which suggests an
extensive conformational change of enzymes upon the immo-
bilization because of the addition of EtOH (Figure 2c). Notably,
no specific absorption bands were observed for Cal-A@OSN
materials calcined at 7008C in air, which confirms that the
above specific bands are all attributable to the organic con-
tents, not to the silica matrix (Figure 2d). These results confirm
the immobilization of Cal-A molecules inside the silica matrix
Figure 1. pH Profiles of the synthesis of Cal-A@OSN(0) (*) and Cal-A@
OSN(110) (&).
Cal-A was initially at pH 5.3 for Cal-A@OSN(0), which increased
instantaneously to pH 9.7 after the addition of silicon sources,
followed by a gradual decrease along with aging. The lowest
and highest pH values were slightly out of the permissible pH
range of Cal-A (the optimum pH range for Cal-A is ꢀ6–9 at
room temperature)[48]; however, it was found to decrease
below pH 9 within 60 min and finally approached neutral, in
which Cal-A can remain enzymatically active. Such ambient
synthetic conditions would have little damage on the enzyme
structures, which enables for an efficient entrapment of en-
zymes into their silica networks without affecting enzyme ac-
tivities. On the other hand, Cal-A@OSN(110), an enzyme–silica
composite synthesized under the highest EtOH concentration,
showed a pH profile with relatively higher pH values compared
to Cal-A@OSN(0), in which the pH approached 10.5 at the max-
imum, which is indicative of a significant effect of EtOH on
enzyme structures during the synthesis.
To check the structural stability of Cal-A after the immobiliza-
tion within the OSNs, the FTIR spectra of Cal-A@OSN(0) and
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2527 – 2536 2529

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
of the OSN and its strong confinement effect whereas the mo-
lecular configuration of the immobilized enzymes is likely de-
formed upon the immobilization, depending on the amount of
EtOH added.
Effect of EtOH on OSN structures
The morphological features of the Cal-A@OSN materials were
observed by using the combined SEM and TEM analysis. The
left panel of Figure 3 shows the SEM images of Cal-A@OSN ma-
terials synthesized with varied EtOH concentrations. The SEM
image of Cal-A@OSN(0) confirms that it is composed of highly
dispersed spherical nanoparticles with an average particle size
(determined based on the sizes of 200 particles for each data)
of approximately 230 nm (Figure 3a). Although Cal-A@OSN(50)
provided an average particle size similar to that of Cal-
A@OSN(0), the samples prepared with higher EtOH concentra-
tions—Cal-A@OSN(70) and Cal-A@OSN(110)—showed in-
creased average particle sizes, 241 and 263 nm, respectively
(Figure 3c and d). As an unusual case, some nanoparticles
with a wrinkled surface morphology were observed for Cal-
A@OSN(110) (Figure 3d) whereas other materials showed per-
fect spherical morphologies. The final size/shape of the OSN
materials are, in principle, determined by the ones of the oil
phase (OA); hence, it can be assumed that EtOH affects the
size/shape of the oils by reducing the surface tension and
changing the curvature of OA micelles.[44]
The TEM images of Cal-A@OSN materials are shown in the
right panel of Figure 3, which indicate that Cal-A@OSN materi-
als include not only monodispersed spherical particles but also
some polydispersed/aggregated particles and some particles
with a size smaller than 100 nm. All samples showed clear con-
trasts between cores and shell layers in the TEM images. Ther-
mogravimetric data of Cal-A@OSN(0) measured in air flow
showed 29.5 % weight loss at ꢂ2108C, which is ascribed to
the combustion of organic contents, such as OA, enzymes, and
aminopropyl groups of APTES (Figure S1). Considering the
composition of the initial reaction solution, this weight loss is
mostly due to the combustion of OA. Furthermore, such core–
shell structures had been retained even after calcination at
7008C in air (Figure S2). These results confirm that the core
cavities of these materials are occupied by OA, which is cov-
ered with shell layers made of silica. Along with the addition
of increased amount of EtOH, the thickness of the silica shell
layer decreased unambiguously; the shell thickness of Cal-
A@OSN(0) was estimated to be 16–20 nm and the one of Cal-
A@OSN(110) was estimated to be 8–12 nm from the TEM
images (cf. Figure 3a and d). This is most likely due to the in-
verse relationship between the particle size and the silica shell
thickness; that is, the larger the OSN particle, the thinner the
silica shell. Considering the fact that the molecular dimensions
of Cal-A calculated from crystallographic data are approximate-
ly 6.3ꢁ5.6ꢁ4.2 nm (although the dimensions in the solution
and the immobilized state may differ from the crystallographic
dimensions),[28,51] it is suggested that the Cal-A molecules em-
bedded in the thinner silica shell of Cal-A@OSN(110) are most
likely surface exposed, whereas the thick silica shell of Cal-
Figure 3. SEM (left) and TEM (right) micrographs of Cal-A@OSN synthesized
with varied EtOH/OA ratios: a) 0, b) 50, c) 70, and d) 110.
A@OSN(0) can sufficiently accommodate Cal-A molecules in its
silica matrix. This difference in geometric dimensions would
affect the stability and recyclability of the embedded enzyme
species, as discussed later.
The existence of mesopores in the silica shell was not ob-
served in the TEM images probably owing to a relatively disor-
dered, random orientation of the pores[43] but was instead con-
firmed by using XRD analysis. The XRD patterns of the as-syn-
thesized Cal-A@OSN materials and those calcined at 7008C in
air are shown in Figure 4a and b, respectively. The as-synthe-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2527 – 2536 2530

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Figure 5. N2 adsorption–desorption isotherms of A) as-synthesized Cal-
A@OSN materials synthesized with varied EtOH/OA ratios—a) 0, b) 50, c) 70,
and d) 110—and B) those calcined at 7008C in air. Filled and empty symbols
represent adsorption and desorption branches, respectively.
Figure 4. XRD patterns of A) as-synthesized Cal-A@OSN materials synthe-
sized with varied EtOH/OA ratios—a) 0, b) 50, c) 70, and d) 110—and
B) those calcined at 7008C in air.
sized Cal-A@OSN materials showed no diffraction peaks in the
low-angle region of XRD patterns because of the filling of the
pores with OA (Figure 4a). On the other hand, the materials
calcined at 7008C in air showed clear diffraction peaks at ap-
proximately 0.92–1.238, which are indexed as the (100) reflec-
tion typical of wormhole-like mesopore structures (Figure 4b).
These are the mesopores created through the self-assembly of
OA as an anionic surfactant.[44] With an increase in the amount
of EtOH added, these peaks appeared more clearly and d-spac-
ing values increased correspondingly from 7.15 to 9.63 nm,
which suggests a substantial change in the mesoporosity of
the silica shell.
The steep increases in the isotherms at P/P0 >0.90 are due to
the filling of the hollow cavities and the interparticle voids
caused by particle aggregation. This type of isotherm is in
agreement with the presence of a hierarchical porosity organi-
zation of the mesoporous silica hollow spheres, as reported
previously.[44] The textural parameters obtained from nitrogen
adsorption isotherms are summarized in Table 1. The calcined
Cal-A@OSN(0) was found to have a BET surface area of
325 m2 gꢁ1 and a total pore volume of 0.472 cm3 gꢁ1, whereas
the as-synthesized Cal-A@OSN(0) demonstrated quite low
structural parameters (Table 1). These structural parameters ap-
parently increased along with the addition of EtOH in the syn-
thesis medium; the specific surface area of the calcined materi-
als increased linearly from 325 to 540 m2 gꢁ1. The increase in
surface area is attributable to the densification of the pores in
the silica shells, which can be confirmed from the increased
peak intensities of the (100) reflections in XRD patterns as well
(Figure 4b). The total pore volume increased from 0.472 to
0.933 cm3 gꢁ1 and thus the micropore/mesopore volume in-
creased from 0.29 to 0.44 cm3 gꢁ1 with the increase in the
EtOH/OA ratio up to 110. Considering that the total pore
volume includes micropore/mesopore volumes and cavity vol-
umes within the hollow spheres (as well as interparticle voids
caused by particle aggregation), it is suggested that the Cal-
A@OSN materials synthesized with higher EtOH concentrations
The pore characteristics of the samples were further charac-
terized by using nitrogen physisorption analysis. The nitrogen
adsorption–desorption isotherms of the Cal-A@OSN materials
are shown in Figure 5. The nitrogen adsorption isotherms of
the as-synthesized Cal-A@OSN materials demonstrated typical
type III isotherms according to the IUPAC classification, and the
sum of micropore and mesopore volumes (Vpore) was negligible
(<0.06) in all cases, which indicates that they are nonporous
solids. After the calcination treatment at 7008C in air, the ad-
sorbed nitrogen quantities increased appreciably owing to the
loss of organic contents, and the nitrogen sorption isotherms
showed type II isotherms with broad hysteresis loops at P/P0 =
0.5–1.0, which correspond to the mesopores in the silica shells.
Table 1. Textural properties of Cal-A@OSN materials synthesized with 1OA/1APTES/6.7TEOS/XEtOH/1600H2O.
Sample
XEtOH Enzyme
Average
Silica shell As-synthesized
After calcination
[d]
[e]
[f]
[g]
[d]
[e]
[f]
[g]
loading[a]
particle size[b] thickness[c] SBET
Vtotal
[cm3 gꢁ1
Vpore
Dp
SBET
Vtotal
[cm3 gꢁ1
Vpore
Dp
ꢁ1
[mgenzyme gcat
]
[nm]
[nm]
[m2 gꢁ1
]
]
[cm3 gꢁ1
]
[nm] [m2 gꢁ1
]
]
[cm3 gꢁ1
]
[nm]
Cal-A@OSN(0)
Cal-A@OSN(50)
Cal-A@OSN(70)
0
50
70
18.9
231
224
241
263
16–20
11–16
10–15
8–12
26
58
57
26
0.284
0.455
0.411
0.236
<0.01
0.05
0.05
n.d.
n.d.
n.d.
4.0
325
391
452
540
0.472
0.866
0.926
0.933
0.29
0.32
0.35
0.44
4.2
3.7
3.4
3.4
18.8
18.4
18.4
Cal-A@OSN(110) 110
0.06
[a] Determined by subtracting the amount of enzyme remained in the supernatant solution from the initial amount of enzyme added; [b] Determined by
SEM observations; [c] Determined by TEM observations; [d] Specific surface area calculated by using the BET method; [e] Total pore volume at P/P0 =0.99;
[f] Volume of pores including micropores and mesopores; [g] Pore diameter estimated by using the BJH method and the N2 desorption branch.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2527 – 2536 2531

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
possess silica shells with higher porosities and cores with
larger cavity spaces.
Enzyme activity assay
The addition of EtOH also caused substantial changes in
pore sizes and their distributions. Pore size distribution curves
calculated by using the BJH method with desorption branches
of the N2 isotherms are shown in Figure 6. As expected, the as-
synthesized Cal-A@OSN materials demonstrated no clear pore
distributions owing to their nonporous nature (Figure 6a). On
The enzyme activity of the Cal-A@OSN materials was assessed
by using the hydrolysis of n-butyl acetate in the phosphate
buffer (pH 7.0) as a test reaction (Scheme 2) and was compared
with the activity of the free enzyme.
Scheme 2. Hydrolysis of n-butyl acetate.
The reaction profiles of the Cal-A@OSN materials as well as
that of the free enzyme are shown in Figure 7a, in which the
enzyme dosage was kept constant (1.0 mg of Cal-A) and the
enzyme activity was compared on the basis of the yields of n-
butanol. As shown in Figure 7a, Cal-A@OSN(0) showed as high
a reaction rate as that of free Cal-A, with a >99% stoichiomet-
ric yield of n-butanol after 6 h of the reaction time, which indi-
Figure 6. Pore size distributions of A) as-synthesized Cal-A@OSN materials
synthesized with varied EtOH/OA ratios—a) 0, b) 50, c) 70, and d) 110—and
B) those calcined at 7008C in air calculated by using the BJH method with
the desorption branch of the N2 isotherm.
the other hand, the calcined Cal-A@OSN materials showed dis-
tinct pore distributions centered at approximately 3.4–4.2 nm
(Figure 6b), which again confirms that these samples have de-
fined mesoporous structures. The pore size of the calcined Cal-
A@OSN(0) was determined to be 4.2 nm and decreased to 3.7,
3.4, and 3.4 nm with the increase in the EtOH/OA ratio to 50,
70, and 110, respectively. This is due to the fact that EtOH re-
duces the surface tension of OA dramatically and affects the
packing of OA in micelles.[44] More importantly, the samples
prepared by adding EtOH demonstrated secondary pores in
the pore size range of 5–20 nm (Figure 6B, curve b and curve
d). The origin of the creation of these secondary pores, that is,
whether they are caused by different micelle formations in-
duced by adding EtOH or they are the voids caused by the re-
moval of enzyme molecules through calcination,[52] still remains
unclear; however, the former may be a valid reason because
such pores were not observed in the sample prepared without
adding EtOH (Figure 6B, curve a).
These combined analyses revealed that the addition of EtOH
to the initial synthesis medium yields Cal-A@OSN with larger
particle sizes and results in the formation of thinner, highly
porous silica shells with both defined mesopores and secon-
dary mesopores compared to the material synthesized without
adding EtOH because EtOH can reduce the surfactant packing
parameter g.[44] As an exceptional case, the silica shell morphol-
ogy changed from a perfect spherical shape to a wrinkled sur-
face morphology upon the addition of an excess of EtOH
(EtOH/OA>110).
Figure 7. A) Reaction profiles obtained during the hydrolysis of n-butyl ace-
tate catalyzed by free Cal-A (ꢁ), Cal-A@OSN(0) (*), Cal-A@OSN(50) (~), Cal-
^
A@OSN(70) ( ), and Cal-A@OSN(110) (&). B) Comparison of reaction profiles
obtained during the hydrolysis of n-butyl acetate catalyzed by Cal-A@OSN(0)
(*, *) and Cal-A@OSN(110) (&, &). Filled symbols: without removal of bio-
catalysts; empty symbols: with removal of biocatalysts through filtration of
the reaction mixtures (the reaction had previously been allowed to proceed
for 1 h). Reaction conditions: Biocatalyst (including 1.0 mg of Cal-A), n-butyl
acetate aqueous solution (50 mm, 1 mL), sodium phosphate buffer (50 mm,
9 mL), pH 7.0, 408C, and reaction time 6 h.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2527 – 2536 2532

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
cates an excellent retention of its enzyme activity upon immo-
bilization. This is probably due to the combination of the am-
bient synthetic conditions that preserve the inherent activity
of enzymes and an optimized configuration of enzymes in the
vicinity of the reactive surface that maximizes the contact be-
tween enzymes and reactants. In contrast, the Cal-A@OSN ma-
terials synthesized by adding EtOH demonstrated much slower
reaction rates than Cal-A@OSN(0), and the initial reaction rate
decreased in the following order: Cal-A@OSN(0)>Cal-
A@OSN(50)>Cal-A@OSN(70)>Cal-A@OSN(110), yet they re-
tained most of their activities. Considering the fact that the
enzyme loadings are the same in all samples, this decreased
activity can be associated with the denaturation of enzymes
during the synthesis, as confirmed by the pH profiles and the
FTIR spectra (Figures 1 and 2). One could hypothesize that the
reaction kinetics may be affected by the diffusion effect of the
reactants; however, as confirmed above, the addition of
a larger amount of EtOH affords Cal-A@OSN materials with
thinner, highly porous silica shells, which should, in principle,
contribute to an efficient access of the reactants to enzymes
and thus discounts this possibility.
Figure 8. Recyclability of Cal-A@OSN biocatalysts. Reaction conditions: bio-
catalyst (50 mg), n-butyl acetate aqueous solution (50 mm, 1 mL), sodium
phosphate buffer (50 mm, 9 mL), pH 7.0, 408C, and reaction time 6 h.
cellent recyclability. This may be the consequence of the
robust confinement effect of the thick, dense silica matrix of
Cal-A@OSN(0), which enables the prevention of the undesired
leaching of enzymes. In contrast, Cal-A@OSN(50), Cal-A@
OSN(70), and Cal-A@OSN(110) lost large parts of their enzyme
activities during their first five to six cycles, followed by
modest decreases; although Cal-A@OSN(50) still retained ap-
proximately 60% of its enzyme activity after 10 cycles of the
reaction, Cal-A@OSN(70) and Cal-A@OSN(110) lost more than
a half of their initial activities during the same number of
cycles, which show poorer recyclabilities. Considering the over-
all trend seen in Figures 7 and 8, the reduction in enzyme ac-
tivity and recyclability can be associated with the subpar struc-
tural properties of the silica shells (i.e., the thinner the silica
shell and the higher the silica shell porosity, the higher the
degree of enzyme leaching), which thereby leads to a continual
leaching of enzymes into reaction solutions during their first
several cycles. Enzymes could possibly be leached through the
secondary pores, the diameters of which are sufficiently wider
than the molecular dimensions of the enzymes, as confirmed
by the structural analysis (see Figure 6); such a leaching was
hardly observed for Cal-A@OSN(0), which has no secondary
pores.
A strong relationship between the pore characteristics of the
silica shells and the residual enzyme activities exists. The reac-
tion profiles obtained during the hydrolysis of n-butyl acetate
over Cal-A@OSN(0) and Cal-A@OSN(110) with and without the
removal of biocatalysts through filtration of the reaction mix-
tures are compared in Figure 7b. When Cal-A@OSN(0) was
used as a biocatalyst, the reaction was quenched immediately
after the recovery of the particles from the reaction mixture
through simple filtration, which demonstrated a robust immo-
bilization of enzymes within the OSN support. On the other
hand, Cal-A@OSN(110) showed a modest increase in the n-bu-
tanol yield even after the recovery of the particles [n-butanol
yield: 11.8% (1 h)!23.6% (6 h)], which indicates a leaching of
some fractions of the enzymes into the reaction solution
during its use. Similar reaction kinetics were observed for Cal-
A@OSN(50) and Cal-A@OSN(70) (data not shown). On the basis
of the detailed structural characterizations disclosed above, the
leaching of enzymes can be associated with the thin silica shell
and/or the high porosity of the silica shell; that is, the enzymes
immobilized within such permeable silica shells are considered
to be partly surface exposed and loosely bound, which thus re-
sults in an easy elution of enzymes. Such a hypothesis is also
confirmed by using a recycling test, as demonstrated in the
following section.
Stability assay
In addition to the recyclability of enzymes, the thermal/chemi-
cal stability is of key importance for their practical biotechno-
logical applications because the ability to retain enzyme activi-
ty in wide pH ranges and at high temperatures provides
a number of processing advantages such as improved reaction
rate and substrate solubility, which enables the efficient cataly-
sis of enzymes.
Recycling assay
Relative enzyme activities of the Cal-A@OSN biocatalysts over
10 reaction cycles in the hydrolysis of n-butyl acetate, in which
the same amount of the biocatalyst (50 mg) was used for each
catalytic run to show the degree of enzyme leaching, are
shown in Figure 8. The Cal-A@OSN(0) biocatalyst, which was
proven to be the most efficient biocatalyst that we examined,
could be repeatedly used at least 10 times without appreciable
loss of its activity (>98% relative to the initial activity had
been retained even after 10 cycles), which demonstrates its ex-
To assess the thermal/chemical stability of the Cal-A@OSN
biocatalysts, some selected biocatalysts were subjected to the
hydrolysis of n-butyl acetate at different solution temperatures
and pH values, in which the activities were normalized by that
of free Cal-A at pH 7.0 and 408C. As shown in Figure 9A and B,
the free enzyme in the buffer solution was readily denatured
and lost most of its activity at ꢂ608C or under extreme pH
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2527 – 2536 2533

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Effect of OA on enzyme activity
Another important factor that needs to be examined is the
effect of the presence of oil (OA) on the enzyme activity. In our
precedent study, it was demonstrated that this enzyme–silica
composite could act as an efficient, reusable, and stable bio-
catalyst in the catalysis in an organic solvent (in n-heptane)
whereas native enzyme alone was found to be inactive under
the same reaction conditions,[43] which implies a productive
role of OA in the catalysis. However, the question of whether
the presence of oil in the core is truly favorable for catalysis re-
quires elucidation. A pertinent result is provided by a compara-
tive testing with Cal-A@OSN(0) and OA-extracted Cal-
A@OSN(0). The reaction profiles of Cal-A@OSN(0) and OA-ex-
tracted Cal-A@OSN(0) in the hydrolysis of n-butyl acetate in
the phosphate buffer (pH 7.0) are shown in Figure 10. Most of
Figure 9. Enzyme activity of Cal-A@OSN(0) (*), Cal-A@OSN(110) (&), and
native Cal-A (ꢁ) at different a) temperature and b) pH values. The activities
were normalized by that of free Cal-A at pH 7.0 and 408C. Reaction condi-
tions: biocatalyst (including 1.0 mg of Cal-A), n-butyl acetate aqueous solu-
tion (50 mm, 1 mL), buffer solution (50 mm, 9 mL), and reaction time 2 h.
Figure 10. Effect of the presence of oil (OA) on the enzyme activity of Cal-
A@OSN(0) biocatalyst. Reaction conditions: biocatalyst (including 1.0 mg of
Cal-A), n-butyl acetate aqueous solution (50 mm, 1 mL), sodium phosphate
buffer (50 mm, 9 mL), pH 7.0, 408C, and reaction time 6 h.
conditions. In contrast, Cal-A@OSN(0) retained most of its activ-
ity over wide temperature and pH ranges, especially at ꢂ608C,
which proves its improved thermal/chemical stability. This im-
proved stability can be associated with the robust stabilizing
effect of the thick and rigid silica matrix of the shell (see
above). Compared to such a tightly immobilized enzyme–silica
composite, the Cal-A@OSN(110) biocatalyst with a thinner and
more permeable silica shell showed a limited thermal/chemical
stability of the enzymes under most of the conditions exam-
ined. This is primarily because of the harsh synthesis conditions
applied, which reduce the residual activity of enzymes (Fig-
ures 1 and 2) and secondarily because of the poor stabilizing
effect of the silica shell of Cal-A@OSN(110); the latter factor
leads to a substantial leaching of enzymes during the reaction
and extensive conformational changes typical of thermal/
chemical denaturation, which therefore results in markedly low
relative activities under the harsh reaction conditions. These
findings elucidate again that a silica shell with optimized shell
thickness and porosity offers a robust confinement effect of
enzymes, which thereby limits the undesired denaturation/
leaching of enzymes, and accordingly providing thermal/chem-
ical stability necessary for enzymes to retain their activities
even under harsh conditions.
the oil (OA) was removed from the particles by treating with n-
heptane at 408C while retaining the enzyme structure, and
enzyme dosage in the reaction mixture was kept constant
(1.0 mg of Cal-A). As shown in Figure 10, OA-extracted Cal-
A@OSN(0) apparently showed a slower reaction rate than pris-
tine Cal-A@OSN(0). A similar result was observed in the hydrol-
ysis of tricaprylin (C27H50O6) in water-saturated n-heptane (Fig-
ure S3), which confirms that the OA in the core plays an impor-
tant role in achieving a high catalytic efficiency in both organic
and water phase reactions. A plausible reason for this is an am-
phiphilic environment on the reactive surface induced by oil-
filled core–shell structure, which facilitates the access of organ-
ic substrates dissolved in water to the active sites of enzymes
in the water phase reaction and increases the affinity with the
organic media in the organic phase reaction, and thus leads to
the effective catalytic efficiencies.
Conclusions
Candida antarctica lipase A (Cal-A) was immobilized on oil-
filled spherical silica nanoparticles (OSNs) with different silica
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2527 – 2536 2534

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
afford Cal-A@OSN(X), in which X represents the molar ratio of
EtOH/OA. The prepared samples were stored in the refrigerator at
48C until use to circumvent denaturation. The loading amount of
the enzyme in the sample was determined by subtracting the
amount of the enzyme remained in the supernatant solution,
which was spectroscopically determined with a Shimadzu UV-2450
spectrophotometer, from the initial amount of the enzyme added.
shell structures through an anionic surfactant-induced self-as-
sembly approach and using EtOH as a cosolvent. The addition
of EtOH to the initial synthesis medium yielded Cal-A@OSN
with larger particle sizes and resulted in the formation of thin-
ner, highly porous silica shells with both defined mesopores
(3.4–4.2 nm) and secondary mesopores (5–20 nm) compared to
the material synthesized without adding EtOH because EtOH
can reduce the surfactant packing parameter g. The retention
of the enzyme activity during biocatalysis was affected signifi-
cantly by the structural properties of the silica shells (thickness
and porosity), and it was found that a dense silica shell with
a thickness sufficiently larger than the molecular dimensions of
enzymes is essential to yield an active, recyclable, and stable
biocatalyst. Further optimization of the silica shell thickness,
for example, through the addition of extra silicon sources such
as TEOS, has not been developed here; however, additional
studies focused on this aspect would be needed in the future
to obtain high-performance heterogeneous biocatalyst com-
bining a maximized catalytic efficiency of enzymes and an en-
hanced recyclability preferably over hundreds to thousands of
cycles. Furthermore, oleic acid encapsulated within the core
cavity of Cal-A@OSN was found to play an important role in
achieving a high catalytic efficiency probably by inducing an
amphiphilic environment on the reactive surface. This study
demonstrated several close relationships between the struc-
tures of the support and the performances of the immobilized
enzymes, which would provide important insights into the
strategy for designing high-performance enzyme–silica compo-
site biocatalysts, which combines high catalytic efficiency, recy-
clability, and stability.
Characterization
The FTIR spectra were recorded on a JASCO FT/IR-6100 instrument
in the spectral range of 2000–400 cmꢁ1 under vacuum with a reso-
lution of 4 cmꢁ1. Field-emission SEM images were recorded on
JEOL JSM-6500F. TEM images were obtained with a Hitachi HF-
2000 FE-TEM equipped with a Kevex energy-dispersive X-ray detec-
tor operated at 200 kV. The sample was suspended in EtOH using
ultrasound, and then a droplet of the suspension was dried on
a carbon grid. The powder XRD patterns were recorded on
a
Rigaku Ultima IV diffractometer with CuKa radiation (l=
1.54056 ꢂ) at 2q=0.6–10.08. Nitrogen adsorption–desorption iso-
therms were measured at ꢁ1968C with the BELSORP-max system
(BEL Japan). Enzyme-loaded samples were outgassed at 808C for
at least 12 h to vaporize physisorbed water, whereas calcined sam-
ples were outgassed at 3508C for 3 h before the measurements.
Specific surface area was calculated by using the BET method with
nitrogen adsorption data at P/P0 =0.05–0.35. The pore size distribu-
tion was obtained by using the BJH method and the desorption
branch of the N2 isotherm. Thermogravimetric analysis was per-
formed by using TG-DTA 2000S (MAC Science Co. Ltd.) from RT to
8008C at
50 cm3 minꢁ1
a
heating rate of 108Cminꢁ1 in the air flow of
.
Enzyme activity assay
Experimental Section
The enzyme activity of the enzyme–silica composites was assessed
by using the hydrolysis reaction of n-butyl acetate. In a typical
method, the reaction was initiated by adding n-butyl acetate aque-
ous solution (1 mL, 50 mm) into the mixture of the buffer solution
(9 mL, 50 mm; sodium acetate for pH 3.0–5.0, sodium phosphate
for pH 6.0–8.0, and Gly-NaOH for pH 9.0–11.0) and the biocatalyst
containing Cal-A (1.0 mg), which had been preincubated at 408C.
The reaction mixture was stirred magnetically at 408C for 6 h. A
portion of the reaction mixture was taken at appropriate intervals
through filtration and then analyzed by using GC (Shimadzu GC-
14B) with a flame ionization detector equipped with a Porapak Q
column. The solid catalysts were recovered through centrifugation
(5 min at 20000g) after the reaction, washed thrice with deionized
water, and subjected to the next catalytic run.
Materials
Cal-A (EC 3.1.1.3; ꢂ1.0 Umgꢁ1) and APTES were purchased from
Sigma–Aldrich and used without any further purification. OA and
TEOS were purchased from Nacalai Tesque, Inc., and used as such.
All commercially available organic compounds for catalytic reac-
tions were purified by using standard methods.
Synthesis of Cal-A@OSN
The synthesis of Cal-A@OSN materials with different silica shell
structures was performed according to the method reported previ-
ously but with a few minor modifications.[43] In a typical synthesis,
OA (0.282 g, 99%, 1 mmol) was dispersed in the aqueous solution
(60 mL) containing the desired amount of EtOH (99.5%) with vigo-
rous stirring. The mixture was then sonicated for 1 min. Cal-A
(14 mg) was dispersed into this solution, and then a mixture con-
taining TEOS (1.40 g, 98%, 6.7 mmol) and APTES (0.221 g, 99%,
1 mmol) was added dropwise with stirring. The solution was stirred
magnetically for approximately 10 min at RT, left to age for 2 h at
the same temperature under static conditions, and aged for anoth-
er 24 h at 508C to form the silica network. The initial molar compo-
sition was adjusted to 1 OA/1 APTES/6.7 TEOS/X EtOH/1600 H2O, in
which the molar ratio of EtOH/OA (X) was varied from 0 to 110.
The resultant suspension was centrifuged at 20000g, washed twice
with distilled water, filtered, and dried overnight under vacuum to
Acknowledgements
This work was financially supported by the Grant-in-Aid for Sci-
entific Research (KAKENHI) in Priority Area “Molecular Nano Dy-
namics” from the Ministry of Education, Culture, Sports, Science
and Technology of Japan (no. A233603560). The authors are
grateful to Dr. Eiji Taguchi and Prof. Hirotaro Mori at the Re-
search Center for Ultra-High Voltage Electron Microscopy, Osaka
University, for assistance with TEM measurements. Y.K. thanks the
JSPS Research Fellowships for Young Scientists.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2527 – 2536 2535

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[26] D. Jung, C. Streb, M. Hartmann, Microporous Mesoporous Mater. 2008,
113, 523–529.
Keywords: biocatalysis · enzymes · heterogeneous catalysis ·
silicates · structure–activity relationships
[27] a) N. Brun, A. Babeau Garcia, H. Deleuze, M. F. Achard, C. Sanchez, F.
Durand, V. Oestreicher, R. Backov, Chem. Mater. 2010, 22, 4555–4562;
b) N. Brun, A. Babeau Garcia, M. F. Achard, C. Sanchez, F. Durand, G. Lau-
rent, M. Birot, H. Deleuze, R. Backov, Energy Environ. Sci. 2011, 4, 2840–
2844.
[1] a) C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli, J. S. Beck, Nature
1992, 359, 710–712; b) J. S. Beck, J. C. Vartuli, W. J. Roth, M. E. Leono-
wicz, C. T. Kresge, K. D. Schmitt, C. T.-W. Chu, D. H. Olson, E. W. Shep-
pard, S. B. McCullen, J. B. Higgins, J. L. Schlenker, J. Am. Chem. Soc.
1992, 114, 10834–10843.
[28] M. Shakeri, K. Engstrçm, A. G. Sandstrçm, J.-E. Bꢆckvall, ChemCatChem
2010, 2, 534–538.
[29] A. Petri, T. Gambicorti, P. Salvadori, J. Mol. Catal. B 2004, 27, 103–106.
[30] R. S. Prakasham, P. R. Likhar, K. Rajyalaxmi, C. Subba Rao, B. Sreedhar, J.
Mol. Catal. B 2008, 55, 43–48.
[31] Z. Cao, J. Zhang, J. Zeng, L. Sun, F. Xu, Z. Cao, L. Zhang, D. Yang, Talanta
2009, 77, 943–947.
[2] a) P. T. Tanev, T. J. Pinnavaia, Science 1995, 267, 865–867; b) S. A. Bag-
shaw, E. Prouzet, T. J. Pinnavaia, Science 1995, 269, 1242–1244.
[3] D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka,
G. D. Stucky, Science 1998, 279, 548–552.
[4] A. Corma, Chem. Rev. 1997, 97, 2373–2420.
[32] N. Hao, H. Wang, P. A. Webley, D. Zhao, Microporous Mesoporous Mater.
2010, 132, 543–551.
[5] J. Y. Ying, C. P. Mehnert, M. S. Wong, Angew. Chem. 1999, 111, 58–82;
Angew. Chem. Int. Ed. 1999, 38, 56–77.
[33] J. Liu, C. Li, Q. Yang, J. Yang, C. Li, Langmuir 2007, 23, 7255–7262.
[34] M. Fujiwara, K. Shiokawa, K. Hayashi, K. Morigaki, Y. Nakahara, J. Biomed.
Mater. Res. Part A 2007, 81, 103–112.
[35] a) M. Mureseanu, A. Galarneau, G. Renard, F. Fajula, Langmuir 2005, 21,
4648–4655; b) A. Galarneau, M. Mureseanu, S. Atger, G. Renard, F.
Fajula, New J. Chem. 2006, 30, 562–571; c) A. Galarneau, G. Renard, M.
Mureseanu, A. Tourrette, C. Biolley, M. Choi, R. Ryoo, F. Di Renzo, F.
Fajula, Microporous Mesoporous Mater. 2007, 104, 103–114.
[36] A. Macario, M. Moliner, A. Corma, G. Giordano, Microporous Mesoporous
Mater. 2009, 118, 334–340.
[37] P. Audebert, C. Demaille, C. Sanchez, Chem. Mater. 1993, 5, 911–913.
[38] D. Avnir, S. Braun, O. Lev, M. Ottolenghi, Chem. Mater. 1994, 6, 1605–
1614.
[39] a) K. Kawakami, S. Yoshida, Biotechnol. Tech. 1994, 8, 441–446; b) K. Ka-
wakami, S. Yoshida, Biotechnol. Tech. 1995, 9, 701–704; c) K. Kawakami,
Biotechnol. Tech. 1996, 10, 491–494.
[6] S. Okada, K. Mori, T. Kamegawa, M. Che, H. Yamashita, Chem. Eur. J.
2011, 17, 9047–9051.
[7] a) Y. Kuwahara, K. Nishizawa, T. Nakajima, T. Kamegawa, K. Mori, H. Ya-
mashita, J. Am. Chem. Soc. 2011, 133, 12462–12465; b) Y. Kuwahara, D.-
Y. Kang, J. Copeland, N. A. Brunelli, S. A. Didas, P. Bollini, C. Sievers, T. Ka-
megawa, H. Yamashita, C. W. Jones, J. Am. Chem. Soc. 2012, 134,
10757–10760.
[8] E. Katz, I. Willner, Angew. Chem. 2004, 116, 6166–6235; Angew. Chem.
Int. Ed. 2004, 43, 6042–6108.
[9] C. Ispas, I. Sokolov, S. Andreescu, Anal. Bioanal. Chem. 2009, 393, 543–
554.
[10] M. Hartmann, Chem. Mater. 2005, 17, 4577–4593.
[11] S. Hudson, J. Cooney, E. Magner, Angew. Chem. 2008, 120, 8710–8723;
Angew. Chem. Int. Ed. 2008, 47, 8582–8594.
[12] H. H. P. Yiu, P. A. Wright, J. Mater. Chem. 2005, 15, 3690–3700.
[13] H. H. P. Yiu, C. H. Botting, N. P. Botting, P. A. Wright, Phys. Chem. Chem.
Phys. 2001, 3, 2983–2985.
[40] I. Gill, A. Ballesteros, J. Am. Chem. Soc. 1998, 120, 8587–8598.
[41] H. R. Luckarift, J. C. Spain, R. R. Naik, M. O. Stone, Nat. Biotechnol. 2004,
22, 211–213.
[14] C. Lei, Y. Shin, J. Liu, E. J. Ackerman, J. Am. Chem. Soc. 2002, 124,
11242–11243.
[42] a) S. A. Miller, E. D. Hong, D. Wright, Macromol. Biosci. 2006, 6, 839–845;
b) J. D. Swartz, S. A. Miller, D. Wright, Org. Process Res. Dev. 2009, 13,
584–589.
[43] Y. Kuwahara, T. Yamanishi, T. Kamegawa, K. Mori, M. Che, H. Yamashita,
Chem. Commun. 2012, 48, 2882–2884.
[44] L. Han, C. Gao, X. Wu, Q. Chen, P. Shu, Z. Ding, S. Che, Solid State Sci.
2011, 13, 721–728.
[45] Z.-A. Qiao, L. Zhang, M. Guo, Y. Liu, Q. Huo, Chem. Mater. 2009, 21,
3823–3829.
[15] a) A. Salis, D. Meloni, S. Ligas, M. F. Casula, M. Monduzzi, V. Solinas, E.
Dumitriu, Langmuir 2005, 21, 5511–5516; b) A. Salis, M. F. Casula, M. S.
Bhattacharyya, M. Pinna, V. Solinas, M. Monduzzi, ChemCatChem 2010,
2, 322–329.
[16] P. H. Pandya, R. V. Jasra, B. L. Newalkar, P. N. Bhatt, Microporous Mesopo-
rous Mater. 2005, 77, 67–77.
[17] Y. Kuwahara, T. Yamanishi, T. Kamegawa, K. Mori, H. Yamashita, J. Phys.
Chem. B 2011, 115, 10335–10345.
[18] C. Li, J. Liu, X. Shi, J. Yang, Q. Yang, J. Phys. Chem. C 2007, 111, 10948–
10954.
[19] E. Serra, E. Dꢃez, I. Dꢃaz, R. M. Blanco, Microporous Mesoporous Mater.
2010, 132, 487–493.
[20] F. Secundo, G. Roda, M. Vittorini, A. Ungureanu, B. Dragoi, E. Dumitriu,
J. Mater. Chem. 2011, 21, 15619–15628.
[21] a) H. Takahashi, B. Li, T. Sasaki, C. Miyazaki, T. Kajino, S. Inagaki, Chem.
Mater. 2000, 12, 3301–3305; b) H. Takahashi, B. Li, T. Sasaki, C. Miyazaki,
T. Kajino, S. Inagaki, Microporous Mesoporous Mater. 2001, 44–45, 755–
762.
[22] M. Tortajada, D. Ramꢄn, D. Beltrꢅn, P. Amorꢄs, J. Mater. Chem. 2005, 15,
3859–3868.
[46] R. D. Schmid, R. Verger, Angew. Chem. 1998, 110, 1694–1720; Angew.
Chem. Int. Ed. 1998, 37, 1608–1633.
[47] Y. Watanabe, P. Pinsirodom, T. Nagao, A. Yamauchi, T. Kobayashi, Y. Nishi-
da, Y. Takagi, Y. Shimada, J. Mol. Catal. B 2007, 44, 99–105.
[48] O. Kirk, M. W. Christensen, Org. Process Res. Dev. 2002, 6, 446–451.
[49] S. Adams, A. M. Higgins, R. A. L. Jones, Langmuir 2002, 18, 4854–4861.
[50] K. Griebenow, Y. D. Laureano, A. M. Santos, I. M. Clemente, L. Rodrꢃguez,
M. W. Vidal, G. Barletta, J. Am. Chem. Soc. 1999, 121, 8157–8163.
[51] D. J. Ericsson, A. Kasrayan, P. Johansson, T. Bergfors, A. G. Sandstrçm, J.-
E. Bꢆckvall, S. L. Mowbray, J. Mol. Biol. 2008, 376, 109–119.
[52] T. Shiomi, T. Tsunoda, A. Kawai, S. Matsuura, F. Mizukami, K. Sakaguchi,
Small 2009, 5, 67–71.
[23] Y. Wang, F. Caruso, Chem. Mater. 2005, 17, 953–961.
[24] Y. Urabe, T. Shiomi, T. Itoh, A. Kawai, T. Tsunoda, F. Mizukami, K. Sakagu-
chi, ChemBioChem 2007, 8, 668–674.
[25] a) A. Vinu, V. Murugesan, M. Hartmann, J. Phys. Chem. B 2004, 108,
7323–7330; b) A. Vinu, V. Murugesan, O. Tangermann, M. Hartmann,
Chem. Mater. 2004, 16, 3056–3065; c) A. Vinu, N. Gokulakrishnan, V. V.
Balasubramanian, S. Alam, M. P. Kapoor, K. Ariga, T. Mori, Chem. Eur. J.
2008, 14, 11529–11538.
Received: January 28, 2013
Revised: March 6, 2013
Published online on April 29, 2013
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2527 – 2536 2536