Synthesis and characterization of a magnetic hybrid catalyst containing lipase and palladium and its application on the dynamic kinetic resolution of amines

Article abstract of DOI:10.1016/j.mcat.2020.111106

Recent papers estimates that about 40 % of drugs present chiral amines in their structure and their synthesis in a sustainable and cost-competitive way is still a challenge for the industry. Kinetic resolution is one of the most applied method to produce these desired compounds where the association with lipase as a catalyst is a good alternative. However, the use of separate racemization catalyst and enzymes in the reaction medium still limits recovery, recycling and can occasionally be responsible for decreasing in selectivity for the desired product. In this work we proposed the synthesis and characterization of a hybrid magnetic catalyst composed containing lipase CaL B and Pd immobilized on the same recovered nanometric magnetic support for the application on Dynamic Kinetic Resolution of (rac)-1-phenylethylamine both in batch and continuous flow conditions. As results it was possible to achieve 99 % of conversion, with 95 % of selectivity and 93 % of enantiomeric excess after 12 h in batch. For a continuous flow system, it was possible to achieve 95 % of conversion with 71 % of selectivity and ee > 99 % after 60 min of reaction. The hybrid catalyst had around 50?100 nm with nanoparticulated Pd (5?10 nm) on its surface, presented a superparamagnetic behavior without remaining magnetization and 22 emu/g of saturation magnetization.

Full text of DOI:10.1016/j.mcat.2020.111106

MolecularCatalysis493(2020)111106
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Synthesis and characterization of a magnetic hybrid catalyst containing
lipase and palladium and its application on the dynamic kinetic resolution of
amines
Clara A. Ferraza, Marcelo A. do Nascimentob, Rhudson F.O. Almeidab, Gabriella G. Sergioa,
Aldo A.T. Juniora, Gisele Dalmônicoc, Richard Caraballoc, Priscilla V. Finotellic,
Raquel A.C. Leãob,d, Robert Wojcieszake, Rodrigo O.M.A. de Souzad, Ivaldo Itabaiana Jra,e,
*
a Department of Biochemical Engineering, School of Chemistry, Federal University of Rio de Janeiro, 21941-910, Brazil
b BOSSGroup—Biocatalysis and Organic Synthesis Group, Chemistry Institute, Federal University of Rio de Janeiro, 21941-909, Brazil
c CBPF, Brazilian Center of Physical Research, Rio de Janeiro, RJ, 22290-180, Brazil
d Pharmacy School, Federal University of Rio de Janeiro, Rio de Janeiro, 21941-170, Brazil
e Univ. Lille, CNRS, Centrale Lille, ENSCL, Univ. Artois, UMR 8181-UCCS-Unité de Catalyse et Chimie Du Solide, F-59000, Lille, France
A R T I C L E I N F O
A B S T R A C T
Keywords:
Lipases
Hybrid catalysts
Chiral amines
Dynamic kinetic resolution
Nanoparticles
Recent papers estimates that about 40 % of drugs present chiral amines in their structure and their synthesis in a
sustainable and cost-competitive way is still a challenge for the industry. Kinetic resolution is one of the most
applied method to produce these desired compounds where the association with lipase as a catalyst is a good
alternative. However, the use of separate racemization catalyst and enzymes in the reaction medium still limits
recovery, recycling and can occasionally be responsible for decreasing in selectivity for the desired product. In
this work we proposed the synthesis and characterization of a hybrid magnetic catalyst composed containing
lipase CaL B and Pd immobilized on the same recovered nanometric magnetic support for the application on
Dynamic Kinetic Resolution of (rac)-1-phenylethylamine both in batch and continuous flow conditions. As re-
sults it was possible to achieve 99 % of conversion, with 95 % of selectivity and 93 % of enantiomeric excess
after 12 h in batch. For a continuous flow system, it was possible to achieve 95 % of conversion with 71 % of
selectivity and ee > 99 % after 60 min of reaction. The hybrid catalyst had around 50−100 nm with nano-
particulated Pd (5−10 nm) on its surface, presented a superparamagnetic behavior without remaining mag-
netization and 22 emu/g of saturation magnetization.
Introduction
the use of cofactors, it has high specificity to substrates, good stability
in organic media and it is able to catalyze a number of reactions with
Enantiomerically pure amines have been playing an important role
in the synthesis of building blocks for the pharmaceutical and agro-
chemical industries, being an important segment of the chiral molecule
market [1]. Among the different production routes for these molecules,
the kinetic resolution is still the most widely used method industrially
[2].
In the context of green chemistry and new environmental restric-
tions being applied in several countries [3], the use of enzyme catalysts
has many advantages, like the mild reaction conditions, and their good
chemo, regio- and stereoselectivity [4]. Among the enzymes used in
organic synthesis, lipase (triacylglycerol acylhydrolases, (E.C. 3.1.1.3)
has gained great importance in the last years, since it does not require
great selectivity [5].
In an enzymatic kinetic resolution, lipase reacts with both en-
antiomers at different rates arriving at optimal conditions in a max-
imum theoretical yield of 50 % [6]. In order increase this yield and
convert the unreacted enantiomer, a racemization is performed, usually
through a metal catalyst, generating a dynamic kinetic resolution
(DKR). The combination of enzymatic resolution and metal racemiza-
tion in a one-pot reaction increases the efficiency of the process, re-
duces costs, time and labor effort [7].
Even though some lipase-catalyzed DKR may be considered well
stablished, the development of catalytic systems for the DKR of amines
continues to attract great attention, and several papers were published
⁎ Corresponding author.
E-mail address: ivaldo@eq.ufrj.br (I. Itabaiana).
https://doi.org/10.1016/j.mcat.2020.111106
Received 25 March 2020; Received in revised form 22 June 2020; Accepted 28 June 2020
2468-8231/©2020ElsevierB.V.Allrightsreserved.

C.A. Ferraz, et al.
MolecularCatalysis493(2020)111106
recently on this topic [2,6,8]. However, the construction of a robust
DKR system is not easy, once some criteria should be taken into ac-
count: (i) the KR must display a sufficient enantioselectivity (E-
value = kfast/kslow ≥ 20), (ii) the enzyme and the racemization catalyst
must be compatible with one another, (iii) the rate of racemization
(krac) must be at least ten times faster than the enzyme catalyzed re-
action of the slow reacting enantiomers (kslow), and (iv) the racemiza-
tion catalyst must not react with the product formed from the resolu-
tion. Among these requirements, the compatibility between the enzyme
and racemization catalyst is generally the critical issue, since these
catalysts often operate optimally under very different condition [9].
A new catalytic approach developed by our group [8] and by a few
others [9,10] involves the co-immobilization of a metal and an enzyme
in one single support. These hybrid catalysts allow cascade reactions to
occur with an increased efficiency, since enzyme and metal are able to
work in cooperation.
Once the catalytic reaction is conduced, the possibility to recycle
biocatalysts is also determinant for the success of enzyme applications
in industry [11]. Furthermore, in the pharmaceutical industry for ex-
ample, it is essential to remove any traces of catalyst which may in-
terfere in subsequent reactions and contaminate the final product [12].
The application of a magnetic field is an efficient way to separate and
recover a catalyst and it has several advantages compared to other
traditional techniques, since it is highly selective, it reduces the use of
consumables and solvents, and it minimizes mass loss, since the se-
paration can be done without removing the material from the reaction
medium [13,14]. Several magnetic supports have already been devel-
oped for the immobilization of enzymes [15,16], however, to the best of
our knowledge, this is the first time that a magnetic nanoparticle is used
as support for a hybrid catalyst.
not improve the catalytic efficiency in the racemization of 1-pheny-
lethylamine [2]. In this work, the best racemization results (46 %
conversion, 76 % selectivity and 15 % enantiomeric excess) where
achieved after 18 h. According to Shakeri et al., the relation between
size and amount of Pd is not properly a role and should be evaluated in
each system [17]. In this work total racemization of 1-phenylethyla-
mine was achieved with 96 % of selectivity after 4 h with 3 % palladium
supported on mesocellular foam. Our group also tested different com-
mercial palladium catalysts and obtained the best results after 2.5 h
with commercially available palladium supported on barium sulphate
(96 % selectivity and 3 % enantiomeric excess) [18]. The previous work
of Bäckvall’s group reported a production of an hybrid catalyst in which
CALB and nanopalladium species were co‐immobilized into the com-
partments of mesoporous silica, where 1 % of enantiomeric excess after
4 h of reaction was achieved on racemization of 1-phenylethylamine.
Recently, the group of Li et al. applied protein–polymer nanoconjugates
as confined nanoreactors for the in situ synthesis of lipase–palladium
(Pd) nanohybrids. The 0.8 nm Pd nanoparticles exhibited increased
activity in racemization of (S)-1-phenylethylamin at 55 °C, that was
more than 50 times that of commercial Pd/C. In the dynamic kinetic
resolutions of pharmaceutical intermediates ( )-1-phenylethylamine,
the lipase–Pd nanohybrids displayed 7.6 times higher efficiencies than
the combination of N435 and Pd/C [10]. These results reaffirm the
robustness of the data obtained by our catalyst.
The next steps in the construction of the hybrid catalyst starting
from MN@4.5%Pd were accompanied by evaluation of its structure,
morphology and magnetic properties. Through transmission electron
microscopy (Fig. 2), the MN synthesized showed around 10–20 nm and
non-uniform rounded shape (Fig. 2a). After coating with TEOS
(Fig. 2b), it was still possible to confirm silica recovery of magnetic
nuclei, achieving around 50 nm of diameter. Once Pd nanoparticles
were produced in situ and adsorbed into surface, the deposition of
medium diameter metallic particles ranging from 5−10 nm could be
confirmed as shown in Fig. 2c that is in agreement with previous re-
ported palladium catalysts supported on silica [2,19,20].
The presence of Pd on surface and the magnetic nuclei were con-
firmed by energy-dispersive X-ray spectroscopy as shown in Fig. 3. Iron
and silicon (red and blue, respectively) were evenly distributed in the
nanoparticles, and Pd (green) remained present only at some spots.
X-ray diffraction assays were also performed to evaluate iron oxide
crystal’s phase, the crystallinity and verify nanoparticles sizes (Fig. 4).
The peaks obtained in the MN diffractogram are in accordance to
those already expressed in previous works and are characteristic of
magnetite and maghemite crystals being a transitory phase between
each other [21,22]. This transitory phase between crystal was already
reported by other research groups on co-precipitation method [23,24]
(Table 1).
The amorphous phase in the MN with TEOS and MN@4.5%Pd dif-
fractograms are typical of amorphous silica coating. Besides the new
wide peak in 40° in the MN@4.5%Pd is characteristic of non-valent
palladium in a cubic face centered structure [25]. Through Scherrer’s
equation, the estimated crystallite size for iron oxide nanoparticles was
9 nm and for palladium nanoparticles was 3 nm.
In the present work, we synthesized and characterized a hybrid
super magnetic biocatalyst containing the lipase B from Candida ant-
arctica (CALB) immobilized and reduced in situ palladium adsorved
and its application in the DKR of (rac)-1-phenylethylamine both under
batch and continuous flow reactors as well as their comparisons with
the commercial lipase Novozym 435 (N435) and palladium im-
mobilized on barium sulfate.
Results and discussion
Synthesis and characterization of the magnetic support with palladium
The magnetic nanoparticles (MN) were synthesized using the Fe2+
and Fe3+ ions co-precipitation method. In order to obtain small and
well dispersed nanoparticles, sonication steps and surface coating with
oleic acid were also necessary. After covering and functionalization
steps, it was investigated different concentrations of adsorbed palla-
dium (MN@3%Pd, MN@4.5%Pd and MN@6%Pd, respectively) in the
efficiency of the racemization of (S)-phenylethylamine. Aiming to
evaluate conversion, enantiomeric excess and selectivity, a kinetic
study was performed until 3 h of reaction. The results are shown in
Fig. 1.
It is important to note that a robust hybrid catalyst is the one that
converts 50 % of the starting enantiomer with 100 % of selectivity with
the lowest enantiomeric excess value possible. This step is important,
since some amine oxidation products in addition to the corresponding
imine can be formed, reducing the selectivity of the reaction. On ra-
cemization assay, the MN@4.5%Pd showed 56 % of conversion of (S)-
phenylethylamine with 96 % of selectivity and 4% of enantiomeric
excess after 60 min. Although the selectivity of this catalyst was slightly
less than that presented by MN@6% Pd, it was chosen for the next steps
since the concentration of 3% Pd was not as efficient. The 4.5 % of Pd
concentration may also have shown a good distribution of metal on the
catalyst surface, generating a contact surface favorable to the reaction.
Xu and colleagues recently reported the increasing on Pd nano-
particles concentration from 1.5 % to 3 % inside cages of MIL-101 did
The relation between the magnetization (M) and the applied field
(H) on MN and MN@4.5%Pd showed in Fig. 5 confirmed the super-
paramagnetic properties of the nanoparticles, since the curves does not
present hysteresis loops or magnetization remaining after field removal
[26]. The saturation magnetization was around 60 emu/g and 22 emu/
g for MN and MN@4.5%Pd, respectively [27]. These finding are even
more interesting when it is take into account that catalyst MN@4.5%
Pd, even coated with mesoporous silica maintains its paramagnetic
potential.
Immobilization of lipase CALB in the MN@4.5%Pd support
The immobilization efficiency of CALB on MN@4.5%Pd support
2

C.A. Ferraz, et al.
MolecularCatalysis493(2020)111106
Fig. 1. Conversion, selectivity and ee of MN with palladium in the racemization reaction. * MN@3% Pd, MN@4.5% Pd and MN@6% Pd, = Magnetic nanoparticles
with 3, 4.5 and 6 % of palladium, respectively.
with both glutaraldehyde (GLU) and epichlorohydrin (EPI) links was
evaluated by Bradford method [28]. The FT-IR spectra (Fig. 6) of both
immobilized biocatalysts were similar and evidences the presence of
peptide bonds when compared to the support without (MN@4.5%Pd).
The band at 1640 cm−1 corresponds to the axial vibration typical of
secondary amide binding (CO]NeH), evidencing enzyme im-
mobilization [27,29].
The new hybrid biocatalysts were investigated for hydrolytic ac-
tivity, esterification potential as well as their application on in the DKR
of (rac)-1-phenylethylamine. Table 2 compares the results for hydro-
lytic activity and esterification conversion after 1 h, 3 h and 6 h in
comparison to commercial N435.
immobilization reaction. Comparing the esterification conversions
through the time, the hybrid catalyst with GLU achieved superior
higher results when compared with EPI and commercial N435 after 3 h
and 6 h of reaction. Enzymatic immobilization can slightly modify en-
zymes’ structures, which changes its activity depending on the reaction
[30]. Rodrigues et al. compared the hydrolytic activity of chitosan
support activated with both GLU and EPI in the immobilization of
CALB. It was found that enzyme immobilized with GLU showed higher
activity [31]. It was also verified that chitosan derivatives showed
higher hydrophobicity, wich is important for immobilization. The hy-
brid catalyst with GLU (MN@4.5%PdCALB) was chosen for the next
steps.
As observed, the hydrolytic activities showed by the new hybrid
catalyst were substantially lower than MN@ catalyst. Lipase CALB is
known to have limited hydrolysis activity of triacylglycerols. In addi-
tion, hybrid catalysts have a larger area filled with adsorbed Pd, which
may have limited the functionalization step and subsequent
Dynamic kinetic resolution of (rac)-1-phenylethylamine in batch conditions
The selected hybrid catalyst was submitted to optimization of the
amount of immobilized enzyme, in order to evaluate the influence of
Fig. 2. Transmission electron microscopy of the MN@4.5%Pd. (a) MN synthesized through the co-precipitation method. (b) coating with TEOS and (c) palladium.
3

C.A. Ferraz, et al.
MolecularCatalysis493(2020)111106
Fig. 3. Energy-dispersive X-ray spectroscopy of the MN@4.5%Pd.
the concentration of the biological catalyst on the conversion, se-
lectivity and ee on the DKR of 1-phenylethylamine. It was performed by
increasing the enzymatic solution during the immobilization. Results
are show on Table 3.
These optimizations could enhance considerably the amount of
enzyme in the support and consequently resulted in an increase in the
hydrolytic activity. The optimization 2 resulted in a hybrid catalyst
with hydrolytic activity similar that found in N435, with a much lower
enzyme loading. However an increasing on enzyme activity could be
noted in both optimizations, it was not proportional. Nevertheless, the
immobilization efficiency decreases with the successive increase of
volumes, which probably shows a saturation in the number of available
sites for lipase immobilization [31,32].
For palmitic acid esterification conversion, we decided to in-
vestigate the kinetic of the reaction with these catalysts, in order to
obtain their initial rates, as shown in Fig. 7.
As seen before, N435 showed faster conversion an initial rates,
achieving almost 77 % in the first 15 min of reaction and maximum
conversion (around 80 %), after only 30 min. Compared to N435, the
optimized hybrid catalysts reached a better conversion, 85 %, after
90 min, denoting lower initial rates. The esterification results did not
correlate with the hydrolysis activity and with the amount of enzyme
load in the support.
After obtaining and evaluating the three hybrid catalysts in the
MN@4.5%Pd support, we performed the DKR of 1-phenylethylamine
following reactional conditions previous published by our group [18].
Results are shown in Table 4.
Although the conversion was high for most immobilizations, the
selectivity was lower and the enantiomeric excess did not improve.
According to [18], the low selectivity in DKR is linked to the hydrogen
source used, since in most cases the racemization is the limiting step. In
the case of our hybrid catalyst, Palladium nanoparticles showed very
high racemization potential. The most challenging step was to equili-
brate enzyme and metal reaction times. As seen in the racemization
study, once the racemic amide is obtained, its prolonged exposure to
the catalyst generates by-products, decreasing the selectivity. As the
enzyme is also in the reaction medium, it may react with these by-
products generating more by-products. To overcome these issues and
improve selectivity, ammonium formate was replaced for hydrogen gas
as hydrogen donor in a new series of DKR experiments. Results are
expressed on Table 5. However this modification improved the se-
lectivity of the reaction indeed, it significantly changed the conversion
results for most the immobilized biocatalysts. Comparing the results
with the enzymatic load in the supports, we conclude that the best
conversions were obtained by hybrid catalyst with Optimization 2.
Aiming to increase lipase CALB activity, the DKR reaction was in-
vestigated at 60 °C, since this is the best temperature applied in pre-
vious DKR reactions [33]. Results are expressed on Table 6).
Fig. 4. XRD of MN, MN coated with TEOS and MN@4.5%Pd.
Table 1
Magnetite, Maghemite and MN XRD peaks.
Magnetite*
Maghemite*
MN
2Ө (°)
30.10
35.43
43.06
53.41
56.96
62.53
73.97
2Ө (°)
30.28
35.69
43.35
53.87
57.42
63.03
74.56
2Ө (°)
30.35
35.80
43.50
53.90
57.35
63.10
74.25
MN synthesized in this work. *Magnetite and Maghemite peaks published by
references [21,22].
4

C.A. Ferraz, et al.
MolecularCatalysis493(2020)111106
Fig. 5. Magnetization as funtion of the applied field.
Table 2
Results of Immobilization of lipase CALB using glutaraldehyde and epi-
chlorohydrin as activating agents.
Support*
Activating
agent
Enzyme
Loading
(mg/g of
support)
Hydrolytic
activity (U/g) conversion (%)
Esterification
1 h
3 h
6 h
MN@4.5%Pd GLU
MN@4.5%Pd EPI
2.96
3.07
13
5
8
161
80
63
13
7
86
33
22
80
84
64
81
79
MN@
GLU
N435**
–
30
80
*MN@4.5%Pd = hybrid catalyst, MN@ = Hybrid catalyst without Pd **N435
= Novozym® 435 (commercial imobilized Lipase B de Candida antarctica).
of conversion and 99 % enantiomeric excess using commercial N435
and Pd\BaSO4 at 70 °C after 12 h [18]. Xu et al. developed an palladium
catalyst in MIL-101 and applied it in the reaction with N435. After 18 h,
the (R)-amide was obtained with 99 % conversion, 93 % selectivity
and > 99 % enantiomeric excess [2].
De Souza et al. develop a hybrid non-magnetic catalytic for DKR of
1-pehnylethylamine and achieved 84 % conversion and > 99 % en-
antiomeric excess after 17 h. The constructed hybrid catalyst contained
1% palladium and 1.9 mg/g of CALB supported on silica nanoparticles
[8]. Engström et al. obtained 99 % conversion with 99 % enantiomeric
excess after 16 h using a hybrid catalyst containing 15.6 % palladium
and 4.8 % of CALB in mesoporous silica [10].
The good results can be explained by the fact that palladium,
through the applied protocol could be regularly arranged on the surface
of mesoporous silica, offering a large contact surface for the racemi-
zation reaction, as well as proximity to the immobilized lipase, favoring
the final DKR reaction.
The hybrid catalyst of Optimization 2 was submitted to a new series
of DKR experiments under batch conditions in order to compare the
metal efficiency with commercial commercial (Pd/BaSO4) and home-
made (MN@4.5 %Pd) sources of palladium, in addition to their com-
bination with lipase N435 and CALB immobilized on magnetic particles
(MN@CALB). All results were also compared with the final hybrid
biocatalyst in the Table 7. As ammonium formate is a greener source of
Hydrogen and easily adapted to different kinds of reactor, we decided
to use the organic source of Pd.
Fig. 6. Fourier transform infrared of the MN, CALB and hybrid catalysts using
glutaraldehyde and epichlorohydrin.
The lower temperature contributed to the improvement of the
conversion, selectivity and enantiomeric excess, and the hybrid cata-
lysts succeeded in the DKR of (rac)-1-phenylethylamine. This result
corroborates with the hypothesis that the enzymatic resolution was the
limiting step of the DKR and that the metal catalyst is super activated.
These results were similar to the obtained by Miranda et al. com-
mercial or non-commercial catalysts. In this work it was achieved 94 %
Comparing the obtained values, it can be observed that MN@4.5 %
5

C.A. Ferraz, et al.
MolecularCatalysis493(2020)111106
Table 3
Influence of the lipase amount on immobilization efficiency.
Optimization
Support mass/enzymatic solution (g/mL)
Immobilization effciency (%)
Enzyme load (mgCALB/g)
Hydrolytic activy (U/g)
Initial
1
2
1/6
1/12
1/24
–
76.0
50.0
35.5
–
2.96
3.89
5.51
30
5
10
63
80
N435
Pd as a palladium source showed lower conversions than Pd/BaSO4 for
all combinations performed. However, despite the low conversion when
compared to the systems containing commercial catalysts, the values
for enantiomeric excess and selectivity remained practically constant.
Among the home-made catalysts, MN@4,5% Pd_CALB showed better
results, with 74 % conversion, 90 % selectivity and enantiomeric ex-
cess > 99 % in 17 h. This result supports the hypothesis that ammo-
nium formate can be used as a hydrogen source for chiral amine DKR
[18].
methylbenzylamine with the hybrid catalyst using ammonium formate
as hydrogen source, the reaction was also investigated in a packed bed
reactor as a way to evaluate the catalyst performance in this system, as
well as reduce the reaction times.
For the supports containing only immobilized CALB, the kinetic
resolution step was performed in a different compartment from the
racemization reaction, allowing it to be performed at room tempera-
ture. Conversion and enantiomeric excess were monitored for up to 9 h
of reaction in the looping system (figure on Table 8) and the best results
obtained are presented in Table 8. For others to see supporting in-
formation.
Recycling
Despite the low selectivity for MN@4.5%Pd + MN@CALB com-
pared to MN@4.5%Pd + N435 (Entries 1 and 2) there was a decrease in
reaction time from 4 to 1 h. In addition, the conversion values were
higher than those obtained using the commercial catalyst with the
palladium source, with a small difference in selectivity and without
significantly affecting the enantiomeric excess of the product (Entries 3-
4).
For catalyst MN@4.5%Pd_CALB (Entry 5) no increase in total re-
action time conversion was observed, but selectivity and enantiomeric
excess did not change significantly over time (see supporting informa-
tion). The low conversion observed may be related to the conditions of
the reaction medium, which favors the palladium racemization stage
that has a high activity, however, the high temperature required may
end up compromising the lipase resolution stage, reflecting in the lower
conversion results. Presented [33]. Unlike the batch reaction, in the
continuous flow system the product formed is less exposed to the cat-
alyst, which may justify the small changes in selectivity.
Once the best reaction conditions were determined, the recycling
efficiency of the hybrid catalyst was investigated. After each cycle of
DKR performed, the hybrid catalyst was washed with toluene, dried
under vacuum and submitted to a new essay. In this work it was pos-
sible to perform 7 cycles of reaction as showed on Graph 1 .
As observed, the hybrid catalyst was able to maintain the conver-
sion, selectivity and enantiomeric excess values for 4 whole DKR cycles,
demonstrating robustness of both adsorbed palladium and lipase, which
remain active. From the fifth cycle onwards, there was a dramatic re-
duction in conversion extending to the seventh cycle. Because it is a
covalently bound lipase-containing biocatalyst, the risks of lipase
leaching are low, as there was a strong binding between the enzyme
and the support. Even so, one sample of the hybrid catalyst before the
reaction and another after the reaction was submitted to Triton X-100
enzymatic desorption assay, and the residual protein was quantified by
the Bradford method. As expected, no significant amounts of enzyme
were detected (data not shown), demonstrating that conversion re-
ductions are likely due to enzymatic inactivation, as the reaction takes
place at 70 °C. ICP analysis (data not shown) showed that no significant
amounts of palladium were found in the reaction medium, showing that
it was not leached with the evolution of the recycles.
Conclusion
The magnetic hybrid catalyst showed a very good performance in
the DKR of (rac)-1-phenylethylamine with four regular recycles without
significant loss in conversion, selectivity and enantiomeric excess.
Similar results in conversion and selectivity were found if compared to
commercial catalytic systems (N435 and Pd\BaSO4), and higher to the
Dynamic kinetic resolution under continuous flow conditions
Considering the optimized conditions for DKR of (+/−)-α-
ones obtained with previous biocatalysts in a lower reaction
Fig. 7. Conversion and initial rates of the esterification reaction.
6

C.A. Ferraz, et al.
MolecularCatalysis493(2020)111106
Table 4
DKR of optimized hybrid ctalysts MN@4.5%Pd using ammonium formate and at 70 °C [18].
Immob
Conversion (%)
Selectivity (%)
eep(%)
Initial
Optimization 1
Optimization 2
88
89
94
75
55
63
86
87
86
*MN@4.5%PdGLU = hybrid catalyst with 4.5 % of palladium using glutaraldehyde.
temperature.
Racemization reactions
The characterization showed that the support used in the hybrid
catalyst is composed by magnetite or maghemite, is nanometric, su-
perparamagnetic and that the palladium nanoparticles are supported in
the silica-coating surface. The saturation magnetization was reduced
when the silica was coated on the surface, however, it was still possible
to separate it using a magnetic field.
Racemization of (S)-1-phenylethylamine was conducted accordin to
[18]. In 4 mL vial, 3 mL of dried toluene, 0.1 M of (S)-1-phenylethyla-
mine, 1.5 equivalents of ammonium formate, 12 mg Na2CO3 and
375 mg of molecular sieves were added. The reaction was kept at 70 °C
under inert atmosphere and the amount of MN added varied according
to the Pd concentration in each, in order to have 30 mol% of palladium.
Samples were derivatized with 5 μL of trifluoracetic anhydride and 5 μL
of triethylamine and analyzed on a Shimadzu GC-2010 gas chromato-
graph with flame ionization detector and CP-Chirasil-Dex CB (25 m
X0.25 mm ID) chiral column using hydrogen as carrier gas. Injector and
detector temperatures were set at 220 °C. The column started at 90 °C
and held for 15 min before it was heated to 190 °C at 40 °C/min and
held at that temperature for 5 min. Biphenyl was used at internal
standard for racemization reactions and conversion, selectivity and
enantiomeric excess were calculated according to the equations bellow.
Experimental section
Pd- nanomagnetic particles(@MNPd) preparation
Iron oxide nanoparticles were synthesized according to [22].
0.02 mol of FeSO4.7H2O and 0.04 mol FeCl3 were dissolved in 40 mL of
distilled water and heated to 80 °C for 1 h with magnetic stirring. 5 mL
of an acetone solution containing 0.1 mL of oleic acid was added. A
stoichiometric amount of NH4OH was added and the sample was so-
nicated for 15 min, followed by washing with acetone and methanol
(1:1) solution at room temperature. 2 g of iron oxides were sonicated in
300 mL of ethanol and 25 mL of water, 9 mL of NH4OH were added and
stirred at 700 rpm for 10 min. Then, 45 mL of TEOS were added and the
medium was stirred for 3 h at 40 °C and 21 h at room temperature. The
silica coated nanoparticles were recovered with a magnet and washed
with ethanol. Then, 2.25 mLmL of APTES were added and the medium
was stirred at 700 rpm for 16 h. The nanoparticles were recovered with
a magnet and washed with ethanol.
In a different reactor, 0.15 M solution of H2PdCl4 was prepared with
HCl and PdCl2 in a aqueous solution at 50 °C for 30 min, followed by
addition of 500 mg of the nanoparticles previous prepared. Then,
400 μL of a 10 % NaBH4 solution was added and the medium was
homogenized for 10 min. 2 mL of NH4OH was added and the medium
was stirred for 1 h. The nanoparticles were recovered with a magnet
and washed with ethanol and dried under vacuum.
A
A
S
n
A
A
B
Conversion (%) = 100 −
n *100
S
0
B
0
Asub
n
Selectivity(%) = 100 −
× 100
AS + AR + ASub
n
n
n
AS
AR
n
n
⎛
⎞
⎟
ee (%) =
−
× 100
⎜
AS + AR
AS + AR
n
n
n
n
⎝
⎠
Where, is the area corresponding to (S)-1-phenylethylamine before the
reaction, is the area of biphenyl before the reaction and AS
, , ,
AR
AB
n n
n
Asub are the areas of (S)-1-phenylethylamine, (R)-1-phenylethylamine,
n
biphenyl and byproducts at the appropriate time.
Lipase immobilization
Lipase immobilization by covalent bond were performed according
a to [34]: 500 mg of covered @MNPd were dispersed in 10 mL of a
phosphate buffer (pH7 and 50 mM) containing 2 mL of glutaraldehyde
Table 5
DKR of MN@4.5%Pd using H2 and at 70 °C [18].
Immobilization
Conversion (%)
Selectivity (%)
eep(%)
Initial
1
2
61
87
> 99
72
94
94
85
86
91
*MN@4.5%PdGLU = Magnetic nanoparticles with 4.5 % of palladium using glutaraldehyde.
7

C.A. Ferraz, et al.
MolecularCatalysis493(2020)111106
Table 6
DKR of MN@4.5%Pd using H2 and at 60 °C.
Immobilization
Conversion (%)
Selectivity (%)
eep(%)
Initial
1
2
78
90
> 99
96
97
95
95
95
92
*MN@4.5%PdGLU = Magnetic nanoparticles with 4.5 % of palladium using glutaraldehyde.
during 24 h. After washing with distilled water, 1 mL of CALB diluted in
5 mL of phosphate buffer (pH 7 and 25 mM) were added to 1 g of
support and the immobilization was done in shaker at 40 °C and
150 rpm for 24 h.
Esterification potential
The esterification potential was done with palmitic acid and ethanol
according to [34]. In a 4 mL vial, a solution containing 100 mM of
palmitic acid, 100 mM ethanol in heptane was added together with
30 mg of the biocatalyst. The reaction was conducted at 50 °C and the
aliquots were derivatized with M-methyl-N-9-trimethylsilyl) tri-
fluoroacetamide and analyzed on a Shimadzu GC-2010 gas chromato-
graph with flame ionization detector and HP-5MS column using hy-
drogen as carrier gas. Injector and detector temperatures were set at
250 °C. The column started at 100 °C and held for 1 min before it was
heated to 180 °C at 15 °C/min and held at that temperature for 1 min.
The conversion was calculated according to the equation bellow, where
Aaci is the area of the palmitic acid and Aest is the area of the produced
ester. A calibration curve was done with the palmitic acid.
Immobilization using epichlorohydrin was done according to [35].
1 g of the support was activated in a shaker for 1 h at 120 rpm and 25 °C
with 10 mL solution of epichlorohydrin (2.5 %) in phosphate buffer (pH
7.5 and 0.1 M). The support was recovered with magnetic separation
and washed with distilled water. Then, the support was soaked with
10 mL of hexane and stirred at 120 rpm for 2 h at 25 °C. Excess of
hexane was removed and 1 mL of CALB diluted in 5 mL of phosphate
buffer (pH 7 and 25 mM) was added to the support. The immobilization
was conducted at 4 °C for 16 h followed by washing with hexane. Im-
mobilization efficiency in both biocatalysts was evaluated by Bradford
method [28].
Aaci
Conversion (%) = 100 −
*100
Aaci + Aest
Hydrolytic activity
The hydrolytic activity was made by the titration method with olive
oil, according to the protocol of [36], being an unit of enzyme activity
(U) defined as the amount of enzyme in grams used to release 1 μmol of
fatty acid per minute under the experimental conditions. An emulsion
containing 5 % of extra virgin olive oil and 5 % of arabic gum was
prepared in a phosphate buffer (pH 7 and 100 mM) and 20 mL of this
emulsion was added to 10 mg of the biocatalyst. The reaction was in-
cubated at 37 °C for 30 min and the reaction was stopped with 20 mL of
an acetone and ethanol solution (1:1). The titration of the fatty acids
produced was done using a 0.04 M NaOH until pH 11 using an Metler
Toledo automatic titrator. Control samples were made under the same
conditions without the incubation of the catalyst.
Dynamic kinetic resolution
The dynamic kinetic resolution of (rac)-1-phenylethylamine was
conducted according to [18]. In 4 mL vial, 3 mL of previously dried and
distilled toluene, 200 mg of the hybrid catalyst, 0.1 M of (S)-1-pheny-
lethylamine, 0.2 M methyl 2-methoxyacetate, 12 mg Na2CO3 and
375 mg of molecular sieves were added. As hydrogen donor, H2gaz or
1.5 equivalents of ammonium formate were used (in this case, a bubbler
was coupled to the vial to reduce the pressure of the medium). 100 μL
samples were analyzed on a Shimadzu GC-2010 gas chromatograph
with flame ionization detector and CP-Chirasil-Dex CB (25 m X0.25 mm
ID) chiral column using the same methodology as the one described for
racemization. Conversion, selectivity and enantiomeric excess were
Table 7
DKR of catalyst using ammonium formate and at 70 °C in batch condition.
Catalyst*
Conversion (%)
Selectivity (%)
eep(%)
MN@4.5%Pd + MN@CALB
MN@4.5%Pd + N435
Pd/BaSO4 + MN@CALB
Pd/BaSO4 + N435
56
60
71
> 99
74
89
87
89
94
90
98
98
> 99
> 99
> 99
MN@4.5%Pd_CALB
*MN@4.5%Pd = Magnetic nanoparticle with 4.5 % palladium, MN@ = hybrid catalyst without enzyme; **N435 = Novozym® 435 (Lipase B de Candida antarc-
ticaimobilized) commercial; Pd/BaSO4 = Palladium immobilized on Barium Sulphate (Commercial)e MN@4.5%Pd_CALB = optimized hybrid catalyst.
8

C.A. Ferraz, et al.
MolecularCatalysis493(2020)111106
Graph 1. Recycles of hybrid catalyst on DKR of (+/−)-α-methylbenzylamine in batch conditions using ammonium formate under batch conditions.
Table 8
DKR of ctalyst using ammonium formate in continuous flow condition.
Entry
Catalyst
Reaction Time (h)
Conversion (%)
Selectivity (%)
eep
(Pd + CALB-immob.)
(%)
1
2
3
4
5
MN@4.5%Pd + MN@CALB
MN@4.5%Pd + N435
Pd/BaSO4 + MN@CALB
Pd/BaSO4 + N435
1
4
1
1
9
95
> 99
65
72
65
71
96
84
94
88
> 99
> 99
> 99
97
MN@4.5%Pd_CALB*
> 99
*MN@4.5%Pd = Magnetic nanoparticle with 4.5 % palladium, MN@ = hybrid catalyst without enzyme; **N435 = Novozym® 435 (Lipase B de Candida antarc-
ticaimobilized) commercial; Pd/BaSO4 = Palladium immobilized on Barium Sulphate (Commercial)e MN@4.5%Pd_CALB = optimized hybrid catalyst.
calculated according to the equations bellow.
Dynamic resolution under continuous flow conditions
AP + AP + ABy
R
S
Conversion (%) =
A stock solution containing 3.2 mmol of rac-(α-Methylbenzylamine)
and methyl 2-methoxyacetate (6.4 mmol) in toluene was transferred
into an Erlenmeyer flask containing Pd/BaSO4 (0.48 mmol of Pd),
Na2CO3 (60 mg) and molecular sieves at room temperature and con-
nected to a packed-bed reactor (2.35 cm3) filled with immobilized
CALB (900 mg). The solution was pumped through the reactor at a rate
of 250 μL.min−1 (residence time of 9.4 min) at room temperature.
Ammonium formate (150 mg, 1.6 mmol) was added to flask which was
heated to 70 °C while the reactor was filled with immobilized CALB at
room temperature. When hybrid catalyst was applied, sodium carbo-
nate and ammonium formate were packaged together with the catalyst
and kept at 70 °C during reaction. The solution was circulated through
system until 9 h. Samples were collected and Enantiomeric excesses
were determined by GC equipped with a chiral column. A GC–MS was
used to identify byproducts and as well as the relative amount of
AS + AP + AP + ABy
R
S
ABy
Selectivity (%) = 100 −
*100
AP + AP + ABy
R
S
AP
AP
R
s
⎛
⎜
⎞
eep
=
−
× 100
⎟
AP + AP
AP + AP
R
S
R
S
⎝
⎠
Where, AP is the area of the (R)-2-methoxy-N-(1-phenylethyl) acet-
R
amide, AP is the area of the (S)-2-methoxy-N-(1-phenylethyl) acet-
S
amide, AS is the area of the substrate, and ABy is the area of the by-
products.
Hybrid catalyst recycles
(+/−)-(α-Methylbenzylamine,
amide and reaction byproducts.
2-methoxy-N-(1-phenylethyl)acet-
The hybrid catalysts were separated with a magnet after the race-
mization and the kinetic resolution and washed 4 times with 2 mL of
toluene. They were dried overnight and applied again in the same re-
action conditions.
9

C.A. Ferraz, et al.
MolecularCatalysis493(2020)111106
Characterizations
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2020.111106.
Fourier-transform infrared (FTIR) spectroscopy: KBr pellets were
•
•
prepared for the evaluation in a Nicolet Magna FTIR-760 equip-
ment in the range of 4000−400 cm−1
.
References
X-ray diffraction: The analysis were performed in the Laboratory
of Hydrogen Technology of the School of Chemistry, Federal
University of Rio de Janeiro in a Rigaku Mminiflex II drif-
fractometer with copper anode (Cu Kα, 30 kV and 15 mA) with
0.05° step and 1 s per step in the range of 5–90 °. The diameter of
the crystallites were estimated using Scherrer’s equation as a
function of the highest intensity peak [37], where Dc is the dia-
meter in nm, k the proportionality constant (0.94 for spherical
particles), λ the wavelength of the radiation (0.1542 nm for Cu
Kα), θ the angle of the highest peak, and β the width at half height
of the peak.
[1] D. Ghislieri, N.J. Turner, Biocatalytic approaches to the synthesis of en-
antiomerically pure chiral amines, Top. Catal. 57 (2014) 284–300, https://doi.org/
10.1007/s11244-013-0184-1.
[2] S. Xu, M. Wang, B. Feng, X. Han, Z. Lan, H. Gu, H. Li, H. Li, Dynamic kinetic
resolution of amines by using palladium nanoparticles confined inside the cages of
amine-modified MIL-101 and lipase, J. Catal. 363 (2018) 9–17, https://doi.org/10.
1016/j.jcat.2018.04.006.
[3] A. Schmid, J.S. Dordick, B. Hauer, A. Kiener, M. Wubbolt, B. Witholt, Industrial
biocatalysis and tomorrow, Nature 409 (2001) 258–268.
[4] D. Muñoz Solano, P. Hoyos, M.J. Hernáiz, A.R. Alcántara, J.M. Sánchez-Montero,
Industrial biotransformations in the synthesis of building blocks leading to en-
antiopure drugs, Bioresour. Technol. 115 (2012) 196–207, https://doi.org/10.
1016/j.biortech.2011.11.131.
[5] A.S. De Miranda, L.S.M. Miranda, R.O.M.A. De Souza, Lipases: valuable catalysts for
dynamic kinetic resolutions, Biotechnol. Adv. (2015), https://doi.org/10.1016/j.
biotechadv.2015.02.015.
[6] Z.S. Seddigi, M.S. Malik, S.A. Ahmed, A.O. Babalghith, A. Kamal, Lipases in
asymmetric transformations: recent advances in classical kinetic resolution and li-
pase–metal combinations for dynamic processes, Coord. Chem. Rev. 348 (2017)
54–70, https://doi.org/10.1016/j.ccr.2017.08.008.
[7] O. Långvik, T. Saloranta, D.Y. Murzin, R. Leino, Heterogeneous chemoenzymatic
catalyst combinations for one-pot dynamic kinetic resolution applications,
ChemCatChem 7 (2015) 4004–4015, https://doi.org/10.1002/cctc.201500459.
[8] S.P. de Souza, R.A.C. Leão, J.F. Bassut, I.C.R. Leal, S. Wang, Q. Ding, Y. Li,
F.L.Y. Lam, R.O.M.A. de Souza, I. Itabaiana, New Biosilified Pd-lipase hybrid bio-
catalysts for dynamic resolution of amines, Tetrahedron Lett. 58 (2017) 4849–4854,
https://doi.org/10.1016/j.tetlet.2017.11.031.
[9] (a) A.K. Ganai, P. Shinde, B.B. Dhar, S. Sen Gupta, B.L.V. Prasad, Development of a
multifunctional catalyst for a ‘“relay”’ reaction, RSC Adv. 3 (2013) 2186–2191,
https://doi.org/10.1039/c2ra22829g;
k λ
Dc =
βcos(θ)
Transmission electron microscopy: 10 mg of the samples were
sonicated in 10 mL of distilled water for 5 min. The images were
made in a JEOL 2100 F 200 kV microscope and a Noran Seven
energy-dispersive X-ray spectroscopysystem.
Magnetometry: The essay was conducted in a Qauntum Design
PPMS Dynacool equipment equipped with a vibrating sample
magnetometer at 300 K in the range of −50,000 to 50,000 Oe.
•
•
(b) O. Verho, J.-E. Bäckvall, Am. Chem. Soc. 137 (12) (2015) 3996–4009, https://
doi.org/10.1021/jacs.5b01031.
Declaration of Competing Interest
[10] (a) K. Engström, E.V. Johnston, O. Verho, K.P.J. Gustafson, M. Shakeri, C.W. Tai,
J.E. Bäckvall, Co-immobilization of an enzyme and a metal into the compartments
of mesoporous silica for cooperative tandem catalysis: an artificial metalloenzyme,
Angew. Chemie - Int. Ed. 52 (2013) 14006–14010, https://doi.org/10.1002/anie.
201306487;
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
(b) X. Li, Y. Cao, K. Luo, Y. Sun, J. Xiong, L. Wang, Z. Liu, J. Li, J. Ma, J. Ge, H.
Xiao, R.N. Zare, Highly active enzyme–metal nanohybrids synthesized in
protein–polymer conjugates, Nat. Catal. 2 (2019) 718–725.
CRediT authorship contribution statement
[11] A. Pellis, S. Cantone, C. Ebert, L. Gardossi, Evolving biocatalysis to meet bioec-
onomy challenges and opportunities, N. Biotechnol. 40 (2018) 154–169, https://
doi.org/10.1016/j.nbt.2017.07.005.
[12] C.W. Lim, I.S. Lee, Magnetically recyclable nanocatalyst systems for the organic
reactions, Nano Today 5 (2010) 412–434, https://doi.org/10.1016/j.nantod.2010.
08.008.
[13] Z.Y. Ma, X.Q. Liu, Y.P. Guan, H.Z. Liu, Synthesis of magnetic silica nanospheres
with metal ligands and application in affinity separation of proteins, Colloids
Surfaces A Physicochem. Eng. Asp. 275 (2006) 87–91, https://doi.org/10.1016/j.
colsurfa.2005.04.045.
[14] L.M. Rossi, Recent Advances in the Development of Magnetically Recoverable Metal
Nanoparticle Catalysts, (2012), https://doi.org/10.1590/S0103-
50532012001100002.
[15] S.A. Ansari, Q. Husain, Potential applications of enzymes immobilized on/in nano
materials: a review, Biotechnol. Adv. 30 (2012) 512–523, https://doi.org/10.1016/
j.biotechadv.2011.09.005.
[16] M. Bilal, Y. Zhao, T. Rasheed, H.M.N. Iqbal, Magnetic nanoparticles as versatile
carriers for enzymes immobilization: a review, Int. J. Biol. Macromol. 120 (2018)
2530–2544, https://doi.org/10.1016/j.ijbiomac.2018.09.025.
[17] M. Shakeri, C. Tai, E. Gçthelid, S. Oscarsson, Small Pd Nanoparticles Supported in
Large Pores of Mesocellular Foam: An Excellent Catalyst for Racemization of
Amines, (2011), pp. 13269–13273, https://doi.org/10.1002/chem.201101265.
[18] A.S. De Miranda, R.O.M.A. De Souza, L.S.M. Miranda, Ammonium formate as a
green hydrogen source for clean semi-continuous enzymatic dynamic kinetic re-
solution of (+/-)-a-methylbenzylamine, RSC Adv. 4 (2014) 13620–13625, https://
doi.org/10.1039/c4ra00462k.
[19] L.L. Chng, N. Erathodiyil, J.Y. Ying, Nanostructured Catalysts for Organic
Transformations, XXX (2012).
[20] D.K. Yi, S.S. Lee, J.Y. Ying, Synthesis and Applications of Magnetic Nanocomposite
Catalysts 18 (2006), pp. 2004–2006.
[21] M.Z. Kassaee, H. Masrouri, F. Movahedi, Applied Catalysis A : general sulfamic acid-
functionalized magnetic Fe 3 O 4 nanoparticles as an efficient and reusable catalyst
for one-pot synthesis of α-amino nitriles in water, Appl. Catal. A, Gen. 395 (2011)
28–33, https://doi.org/10.1016/j.apcata.2011.01.018.
[22] D. Maity, D.C. Agrawal, Synthesis of iron oxide nanoparticles under oxidizing en-
vironment and their stabilization in aqueous and non-aqueous media, J. Magn.
Magn. Mater. 308 (2007) 46–55, https://doi.org/10.1016/j.jmmm.2006.05.001.
Clara A. Ferraz: Conceptualization, Formal analysis, Investigation.
Marcelo A. do Nascimento: Data curation, Formal analysis, Writing -
original draft. Rhudson F.O. Almeida: Methodology, Formal analysis.
Gabriella G. Sergio: Methodology. Aldo A.T. Junior: Validation,
Formal analysis, Methodology. Gisele Dalmônico: Formal analysis,
Data curation. Richard Caraballo: Data curation. Priscilla V.
Finotelli: Validation, Data curation, Writing - original draft. Raquel
A.C. Leão: Data curation, Writing
- review & editing. Robert
Wojcieszak: Funding acquisition, Writing - original draft, Resources.
Rodrigo O.M.A. de Souza: Conceptualization, Funding acquisition,
Project administration, Resources, Supervision, Writing - original draft,
Writing - review & editing. Ivaldo Itabaiana: Conceptualization,
Funding acquisition, Project administration, Resources, Supervision,
Writing - original draft, Writing - review & editing.
Acknowledgements
Authors thank CAPES, CNPq and FAPERJ for financial support. LIA
CNRS France-Brazil “Energy & Environment” and CatBioInnov FEDER
(LS197801) are also acknowledged.
Chevreul Institute (FR 2638), Ministère de l’Enseignement
Supérieur, de la Recherche et de l’Innovation, Région Hauts-de-France,
Métropole Européenne de Lille, and FEDER are acknowledged for
supporting and funding partially this work. This study was supported by
the French government through the Programme Investissement
d’Avenir (I-SITE ULNE / ANR-16-IDEX-0004 ULNE) managed by the
Agence Nationale de la Recherche.
10

C.A. Ferraz, et al.
MolecularCatalysis493(2020)111106
[23] E.S.D.T. de Mendonça, A.C.B. de Faria, S.C.L. Dias, F.F.H. Aragón, J.C. Mantilla,
J.A.H. Coaquira, J.A. Dias, Effects of Silica Coating on the Magnetic Properties of
Magnetite Nanoparticles, (2018), https://doi.org/10.1016/J.SURFIN.2018.11.005.
[24] Y. Hammiche-Bellal, A. Benadda-Kordjani, R. Benrabaa, A. Djadoun, L. Meddour-
Boukhobza, Efficiency of magnetite as nanocrystalline bulk catalyst for the reduc-
tion of 4-nitrophenol, J. Environ. Chem. Eng. 6 (2018) 2355–2362, https://doi.org/
10.1016/j.jece.2018.03.042.
[31] D.S. Rodrigues, A.A. Mendes, W.S. Adriano, L.R.B. Gonc, R.L.C. Giordano,
Multipoint covalent immobilization of microbial lipase on chitosan and agarose
activated by different methods, J. Mol. Catal., B Enzym. 51 (2008) 100–109,
https://doi.org/10.1016/j.molcatb.2007.11.016.
[32] D. Zhang, L. Peng, Y. Wang, Y. Li, Lipase immobilization on epoxy-activated poly
(vinyl acetate-acrylamide) microspheres, Colloids Surf. B Biointerfaces 129 (2015)
206–210, https://doi.org/10.1016/j.colsurfb.2015.03.056.
[25] M. Ganesan, R.G. Freemantle, S.O. Obare, Monodisperse thioether-stabilized pal-
ladium nanoparticles: synthesis, characterization, and reactivity, Chem. Mater. 19
(2007) 3464–3471, https://doi.org/10.1021/CM062655Q.
[33] U. Jonas, T.H. Mogens, P. Shamkant, J. Alwyn, The sequence, crystal structure
determination and refinement of two crystal forms of lipase B from Candida ant-
arctica, Structure 2 (1994) 293–308, https://doi.org/10.1016/S0969-2126(00)
00031-9.
[26] S.J. Schön, Magnetic Properties 65 (2011), pp. 373–391, https://doi.org/10.1016/
S1567-8032(11)08010-4.
[34] S.P. de Souza, I.I. Junior, G.M.A. Silva, L.S.M. Miranda, M.F. Santiago, F.L. Lam,
A. Dawood, U.T. Bornscheuer, R.O.M.A. De Souza, Cellulose as an efficient matrix
for lipase and transaminase immobilization, RSC Adv. 6 (2016) 6665–6671,
https://doi.org/10.1039/c5ra24976g.
[35] L.S. Martin, A. Ceron, P.C. Oliveira, G.M. Zanin, H.F. de Castro, Different organic
components on silica hybrid matrices modulate the lipase inhibition by the glycerol
formed in continuous transesterification reactions, J. Ind. Eng. Chem. 62 (2018)
462–470, https://doi.org/10.1016/j.jiec.2018.01.029.
[27] B. Thangaraj, Z. Jia, L. Dai, D. Liu, W. Du, Effect of silica coating on Fe 3 O 4
magnetic nanoparticles for lipase immobilization and their application for biodiesel
production, Arab. J. Chem. (2016), https://doi.org/10.1016/j.arabjc.2016.09.004.
[28] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem.
254 (1976) 248–254, https://doi.org/10.1016/0003-2697(76)90527-3.
[29] G. Chen, C. Kuo, C. Chen, C. Yu, C. Shieh, Y. Liu, Effect of membranes with various
hydrophobic / hydrophilic properties on lipase immobilized activity and stability,
JBIOSC 113 (2012) 166–172, https://doi.org/10.1016/j.jbiosc.2011.09.023.
[30] R.C. Rodrigues, C. Ortiz, A. Berenguer-Murcia, R. Torres, R. Fernández-Lafuente,
Modifying enzyme activity and selectivity by immobilization, Chem. Soc. Rev. 42
(2013) 6290–6307, https://doi.org/10.1039/c2cs35231a.
[36] D.M. Freire, E.M.F. Teles, E.P.S. Bon, G.L.S. Anna, Lipase production by Penicillium
restrictum in a bench-scale fermenter, Appl. Biochem. Biotechnol. 63–65 (1997)
409–421, https://doi.org/10.1007/BF02920442.
[37] A.L. Patterson, The scherrer formula for X-ray particle size determination, Phys.
Rev. 56 (1939) 978–982, https://doi.org/10.1103/PhysRev.56.978.
11