Medium and reaction engineering for the establishment of a chemo-enzymatic dynamic kinetic resolution of rac-benzoin in batch and continuous mode

Article abstract of DOI:10.1016/j.molcatb.2014.10.011

The effect of the reaction parameters water activity and reaction solvent was investigated for the dynamic kinetic resolution (DKR) of rac-benzoin with immobilized Lipase TL as biocatalyst for transesterification and the heterogeneous chemo-catalyst Zr-TUD-1 (Si/Zr = 25) for in situ racemization. Overall dry reaction conditions led to the best results for both catalysts. The immobilized lipase in a more environmentally benign solvent like cyclopentyl methyl ether (CPME) exhibited a 1.6-fold higher activity and an up to 1.5-fold higher half-life time than in the standard solvents such as toluene and 2-methyltetrahydrofuran (2-MeTHF). Among a variety of deep eutectic solvents (DESs) choline chloride:isosorbide (ChCl:Iso) was found to be suitable for the reaction system. The activity was lower than in the aforementioned solvents, but the very low solubility of the product (S)-benzoin butyrate in ChCl:Iso compared to the investigated organic solvents possesses great potential with respect to downstream processing. Optimized reaction parameters (dry CPME) were applied for DKR in batch and continuous mode yielding comparable or slightly better results than in toluene or 2-MeTHF.

Full text of DOI:10.1016/j.molcatb.2014.10.011

Accepted Manuscript
Title: Medium and reaction engineering for the establishment
of a chemo-enzymatic dynamic kinetic resolution of
rac-benzoin in batch and continuous mode
Author: Annika Petrenz Pablo Dom ´ı nguez de Mar ´ı a Anand
Ramanathan Ulf Hanefeld Marion B. Ansorge-Schumacher
Selin Kara
PII:
S1381-1177(14)00286-0
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcatb.2014.10.011
MOLCAB 3049
To appear in:
Journal of Molecular Catalysis B: Enzymatic
Received date:
Revised date:
Accepted date:
1-9-2014
2-10-2014
20-10-2014
´
Please cite this article as: A. Petrenz, P.D. Maria, A. Ramanathan, U. Hanefeld,
M.B. Ansorge-Schumacher, S. Kara, Medium and reaction engineering for the
establishment of a chemo-enzymatic dynamic kinetic resolution of rac-benzoin in
batch and continuous mode, Journal of Molecular Catalysis B: Enzymatic (2014),
http://dx.doi.org/10.1016/j.molcatb.2014.10.011
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

*Graphical Abstract (for review)
Zr-TUD-1
immoLipTL
Page 1 of 34

HIGHLIGHTS


Water activity plays an important role for the dynamic kinetic resolution of rac-benzoin
in chemo-enzymatic cascade. Catalytic performances of both the enzyme (Lipase TL) and
the chemo-catalyst (Zr-TUD-1) were higher in dry solvents.
Specific activity and half-life time of Lipase TL for kinetic resolution of rac-benzoin was
higher in cyclopentyl methyl ether (CPME) in comparison to toluene, 2-
methyltetrahydrofuran (2-MeTHF) and 1,3-dioxolane.


Racemization activity of Zr-TUD-1 was the best in CPME and in toluene compared to 2-
MeTHF and 1,3-dioxolane.
Specific activity of Lipase TL in deep eutectic solvents was significantly lower than in
toluene, 2-MeTHF, CPME, and 1,3-dioxolane, but product solubility in these alternative
media hold promises for efficient downstream processing.

Continuous dynamic kinetic resolution for synthesis of enantiopure (S)-benzoin ester was
successfully established in stirred tank under optimized conditions (dry CPME).
Page 2 of 34

Medium and reaction engineering for the establishment of a chemo-
enzymatic dynamic kinetic resolution of rac-benzoin in batch and
continuous mode
a
b
c
Annika Petrenz , Pablo Domínguez de María , Anand Ramanathan , Ulf
d
a,
a,
Hanefeld , Marion B. Ansorge-Schumacher *, Selin Kara *
a
Institute of Microbiology, Chair of Molecular Biotechnology, Technische Universität
Dresden, 01062 Dresden, Germany
b
Sustainable Momentum, SL Ap. Correos 3517. 35004, Las Palmas de Gran Canaria,
Canary Islands, Spain
c
Center for Environmentally Beneficial Catalysis, The University of Kansas, Lawrence,
KS 66047, United States
d
Biocatalysis Group, Department of Biotechnology, Delft University of Technology,
Julianalaan 136, 2628 BL Delft, The Netherlands
*
Corresponding Authors:
Prof. Dr. Marion B. Ansorge-Schumacher
Tel.: +49 (0)351 463 39518
Fax: +49 (0)351 463 39520
E-mail address: marion.ansorge@tu-dresden.de
Dr. Selin Kara
Tel.: +49 (0)351 463 39517
Fax: +49 (0)351 463 39520
E-mail address: selin.kara@tu-dresden.de
Page 3 of 34

1
. Introduction
Enzyme catalysis represents an increasingly important research field, where many
practical applications for (enantioselective) organic synthesis can be derived, and have
been already implemented at industrial scale [1]. The use of non-aqueous media for
enzymatic reactions offers several advantages such as high solubility of water-insoluble
substrates, less waste water formation and more straightforward downstream processing.
In the field of biocatalysis in non-aqueous media lipases have proven to be extremely
useful biocatalysts, based on their outstanding resilience to these media, together with
the fact that no cofactors are needed for their catalytic performance [2].
To further provide stronger arguments for (industrial) biocatalysis, the use of more
environmentally-benign solvents as well as the set-up of continuous processes is
presently an area of high interest. Thus, solvents like biomass-derived 2-
methyltetrahydrofuran (2-MeTHF) have been reported as a promising alternative to
conventional solvents and hence attracted great attention in organic synthesis [3].
Likewise, cyclopentyl methyl ether (CPME) [4] and 1,3-dioxolane [5] have been
assessed as environmentally benign alternatives to conventional organic solvents. CPME
has been commercially available since 2005 [4b] and has preferable characteristics, e.g.
a high boiling point, relative stability under acidic and basic conditions, low peroxide
formation, high hydrophobicity (ease of drying), and a narrow explosion range [4b].
Similarly, 1,3-dioxolane is considered as a solvent as well as a reagent [5a] owing to its
preferable physical, chemical and toxicological properties [5b,c]. Furthermore, deep
eutectic solvents (DESs) have been evaluated, which are also environmentally friendly
and cost-effective alternatives to many conventional solvents [6]. DESs – eutectic
mixtures of an ammonium salt (e.g. choline chloride, ethylammonium chloride) and a
hydrogen bond donor (HBD) (e.g. alcohols, amines, amides and carboxylic acids) – have
melting points below room temperature, low volatility, high thermal stability, and
tailored characteristics similar to ionic liquids, but largely offer diminished toxicity,
better biodegradability, large availability at acceptable costs, and simplified preparation
[
6]. Applicability of DESs in biocatalyzed reactions has been demonstrated for
hydrolases [7a–h], and recently for whole-cell redox biocatalysis [7i], for lyases [7j].
2
Page 4 of 34

Recently, also tandem catalysis of a hydrolase and organocatalysts was demonstrated
7k]. Thus, DESs are currently also receiving great attention as eco-friendly solvents [8].
[
Following these important emerging fields – biocatalytic continuous processes and the
employment of environmentally-benign non-conventional solvents – the present work
focuses on the set-up of a chemo-enzymatic dynamic kinetic resolution process (DKR)
for racemic benzoin (aromatic α-hydroxyketone) in a continuous fashion, while
assessing different environmentally-friendly non-aqueous solvents at the same time.
Chiral α-hydroxyketones are important building blocks with a broad range of
applications in the synthesis of biologically active compounds such as pharmaceuticals,
agrochemicals, and pheromones [9]. On this basis, various chemical and biocatalytic
approaches for their synthesis have been developed [10]. Among them, enzymatic routes
show great potential delivering lower amounts of waste and higher stereoselectivities
and overall yields compared to the chemical methods [11]. Biocatalytic syntheses of
chiral α-hydroxyketones are possible either by (i) reduction of α-diketones using
oxidoreductases (EC 1) [12], (ii) acyloin condensation of aldehydes [13], (iii) kinetic
resolution of racemic mixtures via C–C bond cleavage both catalyzed by thiamine
diphosphate (ThDP) dependent lyases (EC 4) [14], or by (iv) dynamic kinetic resolution
(DKR) catalyzed by hydrolases (EC 3) combined with a chemo-catalyst [15]. With
regard to lipases, as stated above, their high stability in and good compatibility with
organic solvents provide special synthetic advantages.
In this study, DKR of racemic benzoin with a lipase from Pseudomonas stutzeri (under
the trade name Lipase TL) and the chemo-catalyst Zr-TUD-1 (Si/Zr = 25) [16] was
investigated (Scheme 1) for solvent needs and continuous operation. Several TUD-1
catalysts with aluminum (Al), zirconium (Zr) and tungsten (W) as incorporated metals
had been evaluated for this reaction in a previous study [17]. Among these, Zr-TUD-1
(Si/Zr = 25) excelled by its high racemization rate and hence was chosen for further
study. The combination of both catalysts in a heterogeneous formulation leads to a yield
and enantioselectivity of >99%, and offers high recyclability in batch reaction systems
[
17].
[
Please insert Scheme 1]
3
Page 5 of 34

For achieving high productivities process parameters such as water activity (a ) [18] and
w
solvent [19] are crucial. Hence, the effect of these parameters on the activity of the
catalysts and the stability of the enzyme (i.e. half-life time) were in-depth evaluated. The
optimized reaction conditions identified for the DKR of rac-benzoin were combined and
then demonstrated in a continuous stirred tank reactor (CSTR, see SI) under temperature
and water activity controlled conditions.
2
. Experimental
2
.1 Materials
All chemicals were purchased from Sigma-Aldrich (Schnelldorf, Germany), Carl Roth
GmbH (Karlsruhe, Germany) or VWR (Dresden, Germany) and were used as received
with a purity ≥98%. Lipase from P. stutzeri (trade name Lipase TL) was kindly provided
by Meito Sangyo Co., Ltd. (Tokyo, Japan). Accurel MP1001 carrier was obtained from
Membrana GmbH (Obernburg, Germany). The humidity sensor HMT337 was purchased
from Vaisala GmbH (Bonn, Germany).
2
.2 Immobilization of Lipase TL
Lipase TL solution at 30 g/L concentration was prepared in a potassium phosphate buffer
200 mM, pH 6.2) and stirred (at 1000 rpm) for at least 1 h at room temperature.
(
Insoluble components were removed by centrifugation (2600 g) for 10 minutes. Accurel
MP1001 carrier was mixed with absolute ethanol (7.6 mLethanol/gcarrier) and incubated for
3
0 minutes. The lipase solution was added to the pre-treated carrier (20 mL lipase
solution/gcarrier) and the suspension was shaken (at 340 rpm) for 4 h at 40 °C. After
incubation, the particles were washed two times with buffer (20 mL buffer/gcarrier) and
filtered afterwards. The immobilized enzyme preparation was dried over silica gel under
reduced pressure for at least five days before further application. For the fragmented
immobilized lipase the formulation was then crushed in a mortar.
4
Page 6 of 34

2
.3 Analysis of protein loading
Protein loading was determined by measurement of the protein concentration before and
after immobilization in the supernatant fraction using Bradford reagent (Roti®-
Nanoquant, Carl Roth GmbH & Co. KG.) according to the manufacturer’s instructions
and bovine serum albumin (BSA) as a standard.
2
.4 Synthesis of deep eutectic solvents (DESs)
DES forming components (i.e. ammonium salts and hydrogen bond donors) were
weighed in a glass vessel at a defined molar ratio (e.g. 1:2 or 1:1.5). The mixture was
stirred at 100 °C until a clear liquid was obtained (at least 1 h up to 24 hours). Water was
removed using a rotary evaporator and the DESs were stored for 72 hours over silica gel
under reduced pressure.
2
.5 Adjustment of water activity
Saturated salt solutions were prepared by mixing a large excess of each salt and distilled
water and stored at 25°C in a desiccator for several days before using. Nine reagent grade
salts with defined water activities were used: LiCl , KAc, MgCl , K CO , Mg(NO ) ,
2
2
2
3
3 2
NaBr, NaCl, KCl, K SO (Table 1).
2
4
For initial experiments, all reaction components (i.e. substrates, solvent, and chemo-
catalyst) were stored separately in a desiccator over saturated salt solutions for three days
at 50 °C. Only in case of the enzyme, the a values were adjusted at 25 °C as a decrease
w
in the enzyme specific activity (51% (in toluene) and 54% (in 2-MeTHF) within 72 h)
was observed when incubation was performed at 50 °C. Later, all reaction components
were incubated with saturated salt solutions at 25 °C as the a values at 25 °C and at
w
5
0 °C were not significantly different (Table 1). The time for reaching equilibrium aw
(72 hours) was previously determined using a Humidity Sensor HMT337.
2
.6 Analysis of specific activity for kinetic resolution of rac-benzoin
Benzoin was prepared in the organic solvent typically at a concentration of 20 g/L
94 mM). The immobilized lipase was pre-incubated with the benzoin solution for
(
1
0 minutes at 50 °C at 20 g/L. The reaction was started by the addition of vinyl butyrate
5
Page 7 of 34

(6 molar equiv.) and the reaction mixture was shaken at 1100 rpm and 50 °C. Aliquots
were taken (30 µL) at definite time intervals and mixed with 1000 µL isopropyl alcohol
followed by HPLC analysis (see 2.11). Experiments were performed at least in
duplicates, all shown results are averages.
2
.7 Set-up for continuous operation and analysis of process stability
Continuous experiments were performed in a 6 mL glass vessel using a HPLC pump (K-
01, Knauer GmbH). The reaction vessel was equipped with two tubes made of Teflon.
5
Substrate solution was pumped continuously through one tube and samples were taken at
the outlet of the second tube (see SI). The reaction mixture was rigorously mixed via a
magnetic stirrer at 500 rpm. The flow rate was chosen according to the specific activity of
the immobilized lipase in the respective solvent (typically 0.5–1.0 mL/min). For stability
measurements, the reaction vessel was loaded with 50 mg of immobilized Lipase TL and
the substrate solution (47 mM benzoin was chosen to avoid crystallization of benzoin in
toluene or CPME within the pump or pipes due to the low solubility in these solvents)
with 3 equiv. vinyl butyrate was pumped through the reactor with the chosen flow rate at
5
0 °C. Aliquots were taken (30 µL) at definite time intervals, mixed with 500 µL
isopropyl alcohol and conversion was analyzed via HPLC. Process stability was
determined from the decrease of conversion over time (i.e. half-life time).
2
.8 Analysis of (R)-benzoin racemization
The chemo-catalyst Zr-TUD-1 (Si/Zr = 25) at 40 g/L was loaded in a Schlenk flask under
nitrogen atmosphere. The solution of (R)-benzoin (typically 20 g/L, 94 mM) was added
and the flask was placed in an oil bath at 50 °C. The reaction mixture was magnetically
stirred at 500 rpm and samples were taken at definite time intervals. The ee values were
measured via HPLC.
2
.9 Analysis of dynamic kinetic resolution (DKR)
The immobilized Lipase TL at 20 g/L and the chemo-catalyst Zr-TUD-1 (Si/Zr = 25) at
0 g/L were loaded in a Schlenk flask under nitrogen atmosphere. Substrate solution
4
(typically 20 g/L benzoin, 94 mM) was added and the mixture was incubated for 10 min.
6
Page 8 of 34

The reaction was started by the addition of 6 equiv. vinyl butyrate and the flask was
placed in an oil bath at 50 °C. Samples were taken at definite time intervals, the
conversion and enantiomeric excess (ee) values were followed by HPLC analysis.
2
.10 Continuous synthesis of (S)-benzoin butyrate
For continuous synthesis of (S)-benzoin butyrate the previously described CSTR set-up
see 2.7) was used. The reaction vessel was loaded with 100 mg of immobilized Lipase
(
TL and 200 mg Zr-TUD-1 (Si/Zr = 25). The substrate solution containing 10 g/L rac-
benzoin (47 mM) and 3 equiv. vinyl butyrate in dry CPME was pumped through the
reactor with a flow rate of 0.05 mL/min. The reaction mixture was rigorously mixed via a
magnetic stirrer at 900 rpm (for better dispersion of the chemo-catalyst in the reactor
volume) at 50 °C. Samples were taken at definite time intervals and conversion and ee
values were measured via HPLC.
2
.11 HPLC analytics
HPLC measurements were performed using PLATINblue uHPLC (Knauer GmbH)
equipped with a Daicel Chiralpak IA column. The flow rate of n-hexane/isopropyl
alcohol (90:10) was 2 mL/min and the column temperature was 40 °C. Typical retention
times were: (R)-benzoin butyrate 2.9 min, (S)-benzoin butyrate 3.3 min, (R)-benzoin
5
.8 min, (S)-benzoin 6.6 min.
3
. Results and discussion
3
.1 Effect of water activity on enzyme activity and stability and on racemization rate
In a first set of experiments, the solvents toluene and 2-MeTHF were assessed with a
focus on water activity, to evaluate the activity and stability of the biocatalyst as well as
the racemization rate of the chemo-catalyst. For a proper understanding of the effects of
reaction parameters (a and solvent), the catalytic performance of the enzyme and
w
chemo-catalyst were investigated separately.
First, kinetic resolution of rac-benzoin catalyzed by Lipase TL was investigated in
toluene and in 2-MeTHF at definite a values at 50 °C (Table 1). These were adjusted as
w
described in the experimental section (see 2.5). For dry reaction conditions all
7
Page 9 of 34

compounds were dried over molecular sieve or in a desiccator over silica gel for several
days resulting in an a of ≤0.02.
w
[
Please insert Table 1 here]
The effect of a on the specific activity of immobilized Lipase TL for transesterification
w
of benzoin in toluene and in 2-MeTHF is shown in Figure 1. In case of toluene a sharp
decrease in the enzyme activity was observed which was significant up to an a value of
w
0
.43. At a of 0.96 only 6.5% residual activity was measured. A similar decrease in
w
Lipase TL activity was also observed in 2-MeTHF between the a values of ≤0.02 and
w
0
.74 and no activity was detected at an a value of 0.96. The results indicate that Lipase
w
TL requires only a very small amount of water for optimal transesterification activity.
This was advantageous as it has been reported in literature that free water can hydrolyse
the ester substrate [21a] or be in a competition with the alcohol [21b]. Hence the
reaction system could be as dry as required for optimal performance.
[
Please insert Figure 1 here]
Half-life time (t ) of the immobilized Lipase TL in toluene and in 2-MeTHF at definite
1
/2
aw was evaluated (Figure 2). Herein, values between ≤0.02 and 0.43 for toluene and
between ≤0.02 and 0.33 for 2-MeTHF were chosen, as the highest residual enzymatic
activities (>70%) were detected in these ranges (see Figure 2). It was observed that t1/2 of
Lipase TL in toluene decreased initially with increasing a , but slightly increased at an aw
w
value of 0.33. In 2-MeTHF the half-life time is lower at 0.11, but increased by 16% at
0
.23. The reason for the high t1/2 at 0.23 in case of 2-MeTHF is not clear yet. We
attributed this observation to the different hydration levels of Lipase TL in the
investigated solvents and a values, which can also influence the activity and/or stability
w
of an enzyme [21b]. Nevertheless, for highest stability of the immobilized Lipase TL the
a value in toluene and in 2-MeTHF should be ≤0.02. In case of 2-MeTHF, the a could
w
w
be increased to 0.23, if required.
8
Page 10 of 34

[
Please insert Figure 2 here]
The catalytic performance of the chemo-catalyst Zr-TUD-1 for racemization of (R)-
benzoin (ee ≥99%) at similar a values was evaluated in toluene and 2-MeTHF (Figure
w
3
). The results showed no significant change in the racemization rate in toluene with
increasing a values which means the dynamic kinetic resolution could be run, if needed,
w
in toluene up to a water activity of 0.33 without any change in the activity of Zr-TUD-1.
In 2-MeTHF the racemization rate was identical at the a values ≤0.02 and 0.11. At 0.23
w
a sharp decrease in the racemization rate was observed. Based on these results 2-MeTHF
can be used as dry form (a of ≤0.02) or at an a value of 0.11 without a loss in the Zr-
w
w
TUD-1 activity. For deeper understanding the exact effect of the water on the Zr-TUD-1
has to be further investigated at catalytic level, and surely novel options for (further)
catalyst design would appear.
[
Please insert Figure 3 here]
3
.2 Effect of alternative organic solvents on enzyme and chemo-catalyst
performance
Apart from (petroleum-based) toluene and the biomass-derived 2-MeTHF, alternative,
environmentally more benign organic solvents were evaluated, namely CPME and 1,3-
dioxolane. As enzymatic activity had been demonstrated to decrease with increasing aw
values (Figure 2), the solvents were used only in dry form. It was observed that in CPME
1
0 g/L of benzoin could be dissolved at 25 °C and 20 g/L at 50 °C. In case of 1,3-
dioxolane, 20 g/L benzoin could already be dissolved at 25 °C. Analysis of the activity of
immobilized Lipase TL showed that the lipase activity in CPME at equal starting
conditions is 1.6-fold and 6-fold higher than in toluene (or 2-MeTHF) and in 1,3-
dioxolane, respectively (Figure 4). Furthermore, in stability analyses Lipase TL showed
the best performance in CPME. Half-life time was determined to be 1.5-fold higher than
in 2-MeTHF and 1.3-fold higher than in toluene. Due to the very low activity of Lipase
TL in 1,3-dioxolane, no stability was determined. Thus, considering activity of
immobilized Lipase TL, CPME seems to be the most suitable solvent among the screened
9
Page 11 of 34

ones, and would hold promising potential for establishing continuous biocatalytic
processes. Interestingly, this observation is consistent with improved enzyme activity and
enantioselectivity reported recently for the transesterification of rac-solketal in CPME
with lipase from Pseudomonas cepacia [4a].
[
Please insert Figure 4 here]
Motivated from such promising activity and stability of Lipase TL in CPME, the catalytic
performance of Zr-TUD-1 in dry CPME was evaluated for comparison with toluene and
2
-MeTHF, and with 1,3-dioxolane despite the low activity of Lipase TL in this solvent
(Figure 5). The highest Zr-TUD-1 activity was observed in toluene, but the racemization
rate in CPME was only slightly lower. (R)-Benzoin (ee ≥99%) was completely racemized
after about 10–12 hours both in CPME and in 2-MeTHF; however, with a lower
racemization rate in 2-MeTHF than in CPME or in toluene. Conversely, a much slower
racemization rate was observed in 1,3-dioxolane. Hence, 1,3-dioxolane appeared to be
not suitable for Zr-TUD-1. As this was consistent with the effect on Lipase TL, this
solvent was excluded for further investigation.
[
Please insert Figure 5 here]
3
.3 Applicability of deep eutectic solvents
Apart from the above-assessed conventional organic solvents, the applicability of deep
eutectic solvents (DESs) for DKR of benzoin – and eventually continuous processes –,
was evaluated. So far several studies have been devoted to the application of DESs for
different transesterification reactions catalyzed by various lipases [7]. Recently, novel
DESs from renewable resources (e.g. choline chloride:isosorbide (ChCl:Iso) and choline
chloride:levulinic acid (ChCl:LA), etc.) were introduced [6f]. Hence, in this study a wide
variety of DESs such as ChCl:Gly, ChCl:U, EAC:Gly, ChCl:Iso, choline chloride:oxalic
acid (ChCl:Ox) and ChCl:LA were applied. At limited solubility of benzoin, isopropyl
alcohol (iPrOH) was investigated as cosolvent in some of these DESs.
10
Page 12 of 34

The solubility of benzoin at chosen concentrations (standard 20 g/L, minimum 5 g/L) in
the aforementioned DESs (with or without iPrOH (10% (v/v) as cosolvent) was
investigated (Table 2). For the DESs ChCl:Gly and EAC:Gly with iPrOH, 5 g/L of
benzoin were not soluble. When the suspension was heated up to 50 °C (by shaking at
1
200 rpm for 24 hours) the benzoin concentration in the soluble fraction was detected to
be at least 2.5 g/L, which was confirmed by HPLC analysis. The DES ChCl:U formed a
two-phase system with iPrOH. Due to these given limitations these DESs were excluded
from further investigation.
[
Please insert Table 2 here]
Subsequently, the activity of Lipase TL (free and immobilized form) in the above
described DESs for kinetic resolution of rac-benzoin was evaluated (Table 3). Despite
the highest benzoin solubility observed, even in ChCl:LA no product formation could be
detected with free enzyme. In the case of the novel DES ChCl:Iso (without iPrOH),
however, an activity of 28.5 U/gprotein was found. The use of immobilized enzyme
exhibited 2.5-fold lower activity in the same DES. For increasing enzyme activity,
immobilized enzyme was fragmented since Durand et al. [7c] had demonstrated that this
can improve lipase activity in DESs. Accordingly, with this treatment the highest
activity of 77.3 U/gprotein (1.94 U/gimmoLipTL) was obtained. However, fragmented
immobilized enzymes showed no activity in other DESs (Table 3). By adding iPrOH the
viscosity could be reduced and the benzoin solubility could be increased, but no activity
improvement was observed.
It is worth mentioning that free enzymes applied in organic media are suspensions as they
are insoluble in non-aqueous solvents [22]. Hence, the reason why free Lipase TL
exhibited low activities in DESs could be explained by the diffusion limitation resulting
from aggregation of the enzyme. When Lipase TL was immobilized, enzyme activity
increased 3-fold in ChCl:Iso, presumably due to two effects: reduced mass transfer
limitation (i.e. increased accessibility of enzymes) by spreading on a solid support and
hyper-activation of lipase by hydrophobic interactions with the Accurel carrier [17].
Overall, lipase activity was significantly lower in DES ChCl:Iso in comparison to the
11
Page 13 of 34

previously investigated organic solvents (see Figure 4). One problem could probably be
the high viscosity (even with iPrOH) of DESs which may result in mass transfer
limitations [7b].
[
Please insert Table 3 here]
Finally, the solubility of the product (S)-benzoin butyrate in DES ChCl:Iso without
iPrOH (which showed the highest activity) and the kinetic resolution of benzoin were
investigated. The maximum solubility of (S)-benzoin butyrate was determined to be
0
.3 g/L. Due to the large difference to benzoin solubility (5 g/L), this offers a great
potential for downstream processing by simplifying the product separation. For
transesterification a conversion of 10% could be achieved after 48 h (see Figure SI 4),
but no further increase occurred. Nevertheless, dry ChCl:Iso without iPrOH is certainly
a candidate for further investigation. One possibility to improve the lipase activity could
be combining two different hydrogen bond donors (e.g. isosorbide and levulinic acid)
for an increased benzoin solubility.
3
.4 Continuous synthesis of (S)-benzoin butyrate via DKR
Among all tested solvents CPME (dry form) turned out to be the best candidate for DKR
of rac-benzoin. DKR was first established in batch mode. Data given in Table 4
compares the results in CPME with the published conversion and ee values [17] in
toluene and in 2-MeTHF. With regard to conversion, DKR in CPME yielded similar
result as in toluene, but achieved a slightly higher ee value. Compared to the results in 2-
MeTHF, conversion and ee value in CPME was improved. Considering the lower toxicity
of CPME compared to toluene (see SI) and its lower boiling point (see SI), which could
simplify downstream processing of products, exchange of toluene for CPME in the
process thus appears beneficial.
[
Please insert Table 4 here]
12
Page 14 of 34

DKR in CPME (dry form) was finally investigated in a continuous process (Figure 6)
using the same reactor system as used for stability measurements. In the simple set-up, a
maximum conversion of 40% was achieved after 2.5 h. Conversion then decreased
steadily to 11% in 76 h. The ee value was ≥98% (S) during the whole run-time. Thus, it
was demonstrated that continuous synthesis of (S)-benzoin butyrate can successfully be
performed, but requires further optimization.
[
Please insert Figure 6 here]
4
. Conclusion
In this study, the effect of water activity and reaction solvent on the dynamic kinetic
resolution of rac-benzoin via a chemo-enzymatic cascade was investigated. Herein
special attention was devoted to environmentally-benign solvents. Activities of both the
enzyme and chemo-catalyst were higher under dry conditions. Among the solvents
evaluated, CPME was determined to be a promising solvent for the enzyme as well as
chemo-catalyst.
The application of DESs for the model chemo-enzymatic cascade was explored for the
first time. The enzyme performs to a lesser extent in DESs than in CPME; nevertheless,
further improvements (alternative hydrogen bond donors by combining such as
isosorbide and levulinic acid) could result in higher lipase activities. The very limited
solubility of the ester product in DESs on the other hand offers great potential for
downstream processing.
Dry CPME was chosen as solvent for the DKR based on the high performance of enzyme
and chemo-catalyst. DKR was demonstrated in a continuous fashion and enantiopure (S)-
benzoin ester was synthesized over 76 h with conversions between 40 to 11%.
Optimization will be performed in future research.
5
. Acknowledgements
The authors thank Shinnosuke Hashikawa (Tokyo Research Laboratory, Meito Sangyo
Co., Ltd) for the generous supply of Lipase TL. A. P. and M.B. A.-S. thank AiF-ZIM for
financial support within project KF3068202AJ4 (SynKat-alpha). S. K. thanks
13
Page 15 of 34

DECHEMA e.V. (Frankfurt am Main, Germany) for financial support in the form of a
Max Buchner” fellowship. Julius Kramer is gratefully acknowledged for technical
“
assistance.
References
[
1] a) A. Schmid, F. Hollmann, J.B. Park, B. Bühler, Curr. Opin. Biotech. 13 (2002) 359–
66; b) A. Liese, K. Seelbach, C. Wandrey. Industrial Biotransformations second ed.,
Wiley-VCH, Weinheim, 2006; c) R. Wohlgemuth, Curr. Opin. Biotech. 21 (2010) 713–
24; d) R. Wohlgemuth, Curr. Opin. Microbiol. 13 (2010) 283–292; e) G.W. Huisman,
3
7
Chim. Oggi. 29 (2011) 44–46; f) U.T. Bornscheuer, G.W. Huisman, R.J. Kazlauskas, S.
Lutz, J.C. Moore, K. Robins, Nature 485 (2012) 185–194.
[
2] a) P. Adlercreutz, Chemical Society Reviews, 42 (2013) 6406–6436; b) M.T. Reetz,
Current opinion in chemical biology, 6 (2002) 145–150; c) K.E. Jaeger and T. Eggert,
Curr Opin Biotechnol. 13 (2002) 390–397.
[
3] a) C. Capello, U. Fischer, K. Hungerbühler, Green Chem. 9 (2007) 927–934; b) P.G.
Jessop, Green Chem. 13 (2011) 1391–1398; c) S. Shanmuganathan, D. Natalia, A. van
den Wittenboer, C. Kohlmann, L. Greiner, P. Domínguez de María, Green Chem. 12
(2010) 2240–2245; d) V. Pace, P. Hoyos, L. Castoldi, P. Domínguez de María, A.R.
Alcántara, ChemSusChem. 5 (2012) 1369–1379.
[
4] a) Y. Mine, L. Zhang, K. Fukunaga, Y. Sugimura. Biotechnol. Lett. 27 (2005) 383–
3
2
88; b) K. Watanabe, N. Yamagiwa, Y. Torisawa, Org. Process Res. Dev. 11 (2007)
51–258; c) K. Beydoun and H. Doucet, ChemSusChem 4 (2011) 526–534; d) S.
Sakamoto. Chim. Oggi. 31 (2013) 24–27.
[
5] a) N.P. Cheremisinoff. Industrial Solvent Handbook, Marcel Dekker Inc., New York,
003; b) R. Moscoso, J. Carbajo, J.A. Squella. Electrochem. Commun. 48 (2014) 69–72;
2
c) Material Safety Data Sheet, 1,3-Dioxolane CAS = 646-06-0, Carlo Erba Reagenti SpA,
Rodano, Italy.
14
Page 16 of 34

(www.carloerbareagents.com/ChemicalsSearchMSDSPP.aspx?msds=3058).
[
6] a) A.P. Abbott, G. Capper, D.L. Davies, R.K. Rasheed, V. Tambyrajah, Chem.
Commun. 2003 (2003) 70–71; b) A.P. Abbott, D. Boothby, G. Capper, D.L. Davies, R.K.
Rasheed, J. Am. Chem. Soc. 126 (2004) 9142–9147; c) K.D. Weaver, H.J. Kim, J. Sun,
D.R. MacFarlane, G.D. Elliott, Green Chem. 12 (2010) 507–513; d) P. Domínguez de
María, Z. Maugeri, Curr. Opin. Chem. Biol. 15 (2011) 220–225; e) Y.H. Choi, J. van
Spronsen, Y. Dai, M. Verberne, F. Hollmann, I.W.C.E. Arends, G.J. Witkamp, R.
Verpoorte, Plant. Physiol. 156 (2011) 1701–1705; f) Z. Maugeri, P. Domínguez de
María, RSC Adv. 2 (2012) 421–425.
[
7] a) J.T. Gorke, F. Srienc, R.J. Kazlauskas, Chem. Commun. (2008) 1235–1237; b) H.
Zhao, G.A. Baker, S. Holmes, Org. Biomol. Chem. 9 (2011) 1908–1916; c) E. Durand, J.
Lecomte, B. Baréa, G. Piombo, E. Dubreucq, P. Villeneuve, Process Biochem. 47 (2012)
2
081–2089; d) D. Lindberg, M. de la Fuente Revenga, M. Widersten, J. Biotechnol. 147
2010) 169–171; e) H. Zhao, G.A. Baker, S. Holmes, J. Mol. Catal. B: Enzym. 72 (2011)
63–167; f) C. Lehmann, F. Sibilla, Z. Maugeri, W.R. Streit, P. Domínguez de María, R.
(
1
Martinez, U. Schwaneberg, Green Chem. 14 (2012) 2719–2726; g) Z. Maugeri, W.
Leitner, P. Domínguez de María, Eur. J. Org. Chem. (2013) 4223–4228; h) Z. Maugeri,
W. Leitner, P. Domínguez de María, Eur. J. Org. Chem. (2013) 4223–4228; i) Z.
Maugeri, P. Domínguez de María, ChemCatChem. (2014) DOI: 10.1002/cctc.201400077;
j) Z. Maugeri, P. Domínguez de María, J. Mol. Catal. B: Enzym. (2014) 107, 120–123; k)
C.R. Müller, I. Meiners, P. Domínguez de María. RSC Adv. (2014) DOI:
1
0.1039/c4ra09307k.
[
8] a) Z. Chen, W. Zhu, Z. Zheng, X. Zhou, J. Fluorine Chem. 131 (2010) 340–344; b)
Y.A. Sonawane, S.B. Phadtare, B.B. Borse, A.R. Jagtap, G.S. Shankarling, Org. Lett. 12
(2010) 1456–1459.
[
9] a) M. Pohl and A. Liese in: R.N. Patel (Eds.), Biocatalysis in the Pharmaceutical and
Biotechnology Industries. CRC Press Taylor & Francis Group, Florida, 2007, pp. 661–
76; b) S. Kara and A. Liese, Enzymatic C–C bond formation, asymmetric, in: M.C.
6
15
Page 17 of 34

Flickinger (Eds.), Encyclopedia of Industrial Biotechnology. Bioprocess, Bioseparation,
and Cell Technology, Wiley-VCH, 2010, pp. 2034–2049; c) M. Müller, G.A. Sprenger,
M. Pohl, Curr. Opin. Chem. Biol. 17 (2013) 261–270.
[
10] P. Hoyos, J.-V. Sinisterra, F. Molinari, A.R. Alcántara, P. Domínguez de María,
Acc. Chem. Res. 43 (2010) 288–299.
[
11] a) D. Enders, O Niemeier, A. Henseler, Chem. Rev. 107 (2007) 5606–5655; b) S.E.
O'Toole, C.A. Rose, S. Gundala, K. Zeitler, S.J. Connon, J. Org. Chem. 76 (2011) 347–
57.
3
[
12] a) K. Edegger, W. Stampfer, B. Seisser, K. Faber, S.F. Mayer, R. Oehrlein, A.
Hafner, W. Kroutil, Eur. J. Org. Chem. 8 (2006) 1904–1909; b) T. Johanson, M.
Carlquist, C. Olsson, A. Rudolf, T. Frejd, M.F. Gorwa-Grauslund, Appl. Microbiol.
Biotechnol. 77 (2008) 1111–1118; c) L.N. Monsalve, P. Cerrutti, M.A. Galvagno, A.
Baldessari, Biocatal. Biotransform. 28 (2010) 137–143.
[
13] a) P. Domínguez de María, T. Stillger, M. Pohl, M. Kiesel, A. Liese, H. Gröger, H.
Trauthwein, Adv. Synth. Catal. 350 (2008) 165–173; b) S. Kara, W.S. Long, M.
Berheide, S. Peper, B. Niemeyer, A. Liese, J. Biotechnol. 152 (2011) 87–92; c) A. van
den Wittenboer, B. Niemeijer, S. Kumar Karmee, M.B. Ansorge-Schumacher, J. Mol.
Catal. B: Enzym. 67 (2010) 208–213; d) P. Ayhan, I. Simsek, B. Ciftci, A.S. Demir, Org.
Biomol. Chem. 9 (2011) 2602–2605; e) T. Gerhards, U. Mackfeld, M. Bocola, E. von
Lieres, W. Wiechert, M. Pohl, D. Rother, Adv. Synth. Catal. 354 (2012) 2805–2820; f)
M. Müller, G.A. Sprenger, M. Pohl, Curr. Opin. Chem. Biol. 17 (2013) 261–270; g) M.
Pérez-Sánchez, C.R. Müller, P. Domínguez de María, ChemCatChem. 5 (2013) 2512–
2
516.
[
14] a) A.S. Demir, M. Pohl, E. Janzen, M. Müller, J. Chem. Soc. Perkin Trans. 1 (2001)
33–635; b) E. Janzen, M. Müller, D. Kolter-Jung, M.M. Kneen, M.J. McLeish, M. Pohl,
Bioorg. Chem. 34 (2006) 345–361; c) Ş.B. Sopaci, I. Şimşek, B. Tural, M. Volkan, A.S.
6
16
Page 18 of 34

Demir, Org. Biomol. Chem. 7 (2009) 1658–1664; d) C.R. Müller, M. Pérez-Sánchez, P.
Domínguez de María, Org. Biomol. Chem. 11 (2013) 2000–2004.
[
(
15] a) P. Hoyos, M. Fernandez, J. V. Sinisterra, A. R. Alcántara, J. Org. Chem. 71
2006) 7632–7637; b) P. Hoyos, M. A. Quezada, J. V. Sinisterra, A. R. Alcántara, J. Mol.
Catal. B: Enzym. 72 (2011) 20–24; c) P. Hoyos, A. Buthe, M.B. Ansorge-Schumacher,
J.V. Sinisterra, A.R. Alcántara, J. Mol. Catal. B: Enzym. 52–53 (2008) 133–139.
[
16] a) A. Ramanathan, M. C. Castro Villalobos, C. Kwakernaak, S. Telalovic, U.
Hanefeld, Chem. Eur. J., 14 (2008) 961-972; b) S. Telalović, A. Ramanathan, G. Mul, U.
Hanefeld, J. Mater. Chem. 20 (2010) 642–658; c) A. Ramanathan, D. Klomp, J. A.
Peters, U Hanefeld, J. Mol. Catal. A: Chem. 260 (2006) 62–69.
[
17] R. Nieguth, J. ten Dam, A. Petrenz, A. Ramanathan, U. Hanefeld, M.B. Ansorge-
Schumacher. RSC Adv. 4 (2014) 45495–45503.
[
18] a) A. Jakoblinnert, R. Mladenov, A. Paul, F. Sibilla, U. Schwaneberg, M.B.
Ansorge-Schumacher, P. Domínguez de María, Chem. Commun. 47 (2011) 12230–
2232; b) R.J. Mikolajek, A.C. Spiess, M. Pohl, S. Lamare, J. Büchs, ChemBioChem. 8
1
(2007) 1063–1070.
[
19] a) K. Faber, Biotransformations in Organic Chemistry, sixth ed., Springer, Berlin,
2
011; b) P. Adlercreutz, in: G. Carrea, S. Riva (Eds.), Organic Synthesis with Enzymes in
Non-Aqueous Media, Wiley-VCH, Weinheim, 2008, pp. 3–34.
[
20] L. Greenspan, Journal of Research of the National Bureau of Standards – A. Physics
and Chemistry. 81A (1977) 89–96.
[
21] a) G. Chowdary and S. Prapulla, Process Biochem. 38 (2002) 393–397; b) R.
Bovara, G. Carrea, G. Ottolina, S. Riva, Biotechnol. Lett. 15 (1993) 937–942; c) Y.-X.
Li, A. J. J. Straathof, U. Hanefeld, Tetrahedron: Asymmetry 13 (2002) 739–743.
17
Page 19 of 34

[
22] S.J. Singer, Adv. Protein. Chem. 17 (1962) 1–68.
18
Page 20 of 34

Table Captions
Table 1. Water activity values of selected saturated salt solutions at 25 °C and 50 °C.
Table 2. Evaluation of the solvents for the solubility of rac-benzoin at 50 °C (by shaking
at 1100 rpm for 24 hours).
Table 3. Activity of immobilized Lipase TL in different dry DESs.
Table 4. Result of conversion and ee value of the dynamic kinetic resolution of rac-
benzoin in CPME compared to toluene and 2-MeTHF after 5 h.
19
Page 21 of 34

Table 1
Salt
aw
aw
[50°C]
0.11
a
a
[
25°C]
0.11
0.23
0.33
0.43
0.53
0.58
0.75
0.84
0.97
LiCl
KAc
b
0.19
MgCl2
KCO3
Mg(NO3)2
NaBr
0.31
0.43
0.45
0.51
0.74
0.81
0.96
NaCl
KCl
K SO4
2
a
b
L. Greenspan (1977) [20], Data measured with a humidity sensor.
20
Page 22 of 34

Table 2
Solvent
iPrOH
10% (v/v)]
-
Benzoin
[g/L]
20
Solubility
Remarks/Observations
[
Toluene
+
+
+
+
-
Completely soluble
Completely soluble
Completely soluble
Completely soluble
Many crystals visible
Many crystals visible
Many crystals visible
Very few crystals visible
Many crystals visible
Completely soluble
Many crystals visible
Many crystals visible
Highly-viscous mixture
Completely soluble
2
-MeTHF
CPME
-
-
20
20
20
5
1
,3-Dioxolane
-
a
EAC:Gly (1:1.5)
ChCl:Gly (1:2)
ChCl:U (1:2)
+
+
+
-
5
-
5
-
ChCl:Iso (1:2)
ChCl:Iso (1:2)
ChCl:Iso (1:2)
ChCl:Iso (1:2)
ChCl:Iso (1:2)
ChCl:Ox (1:2)
ChCl:LA (1:2)
5
+
-
-
10
10
20
20
5
+
-
+
-
+
-
-
-
-
20
+
a
Molar ratio (MR) of ammonium salt : hydrogen bond donor.
21
Page 23 of 34

Table 3
DES
Benzoin [g/L]
Enzyme form
Activity
Activity
[
U/gimmoLipTL
]
[U/gprotein
]
ChCl:Gly:iPrOH
EAC:Gly:iPrOH
ChCl:Iso
2.5
2.5
5
free
free
-
-
n.d.
n.d.
free
-
28.5
14.0
n.d.
ChCl:Iso:iPrOH
ChCl:LA
10
20
5
free
-
free
-
ChCl:Iso
immobilized
f. immobilized
f. immobilized
f. immobilized
f. immobilized
f. immobilized
0.30
n.d.
n.d
1.94
0.30
n.d.
12.0
n.d.
ChCl:Gly:iPrOH
EAC:Gly:iPrOH
ChCl:Iso
2.5
2.5
5
n.d.
77.3
12.0
n.d.
ChCl:Iso:iPrOH
ChCl:LA
10
20
Reaction conditions: various rac-benzoin concentrations, c(vinyl butyrate) = 6 equiv., c(immoLipTL) = 20 g/L, at
1
2
100 rpm and 50 °C. Activity of immoLipTL in U/gprotein was determined with a carrier loading of
5.1 mgprotein/gimmoLipTL. f. = fragmented.
22
Page 24 of 34

Table 4
a
a
Solvent
CPME
Conversion [%]
98.2
ee [%]
99.0
[
17]
[17]
Toluene
98.4
96.9
98.5
97.0
[17]
[17]
2
-MeTHF
a
after 5 h. Reaction: c(benzoin) = 94 mM, c(vinyl butyrate) = 6 equiv., c(immoLipTL) =
0 g/L, c(Zr-TUD-1) = 40 g/L at 1100 rpm and 50 °C.
2
23
Page 25 of 34

Scheme Caption
Scheme 1. Lipase TL-catalyzed dynamic kinetic resolution of racemic benzoin using
vinyl butyrate as an acyl donor and the chemo-catalyst Zr-TUD-1 for in situ racemization
of (R)-benzoin.
Scheme 1
24
Page 26 of 34

Figure Captions
Figure 1. Effect of water activity on the Lipase TL-catalyzed kinetic resolution of rac-
benzoin running in (A) toluene and (B) 2-MeTHF. Reaction conditions: c(benzoin) =
2
5
0 g/L (94 mM), c(vinyl butyrate) = 6 equiv., c(immoLipTL) = 20 g/L, at 1100 rpm and
0 °C.
Figure 2 Effect of water activity on process stability of immobilized Lipase TL for
kinetic resolution of rac-benzoin running in toluene and 2-MeTHF. Reaction conditions:
c(benzoin) = 10 g/L (47 mM), c(vinyl butyrate) = 3 equiv., c(immoLipTL) = 8.3 g/L
(50 mg immoLipTL), at 500 rpm, 0.5–1 mL/min flow rate and 50 °C.
Figure 3. Effect of water activity (a ) on catalytic performance of Zr-TUD-1 for
w
racemization of (R)-benzoin in (a) toluene and (b) 2-MeTHF. Reaction conditions: c((R)-
benzoin) = 20 g/L (94 mM), c(TUD-1) = 40 g/L, at 1100 rpm and 50 °C.
Figure 4. Activity and half-life time of immobilized Lipase TL in different dry solvents.
Reaction conditions: for activity measurement: c(benzoin) = 20 g/L (94 mM), c(vinyl
butyrate) = 6 equiv., c(immoLipTL) = 20 g/L, at 1100 rpm and 50 °C; for process
stability measurement: c(benzoin) = 10 g/L (47 mM), c(vinyl butyrate) = 3 equiv.,
c(immoLipTL) = 8.3 g/L (50 mg immoLipTL), at 500 rpm, 0.5–1 mL/min flow rate and
5
0 °C.
Figure 5. Catalytical performance of Zr-TUD-1 for racemization of (R)-benzoin in dry
solvents: CPME and 1,3-dioxolane compared to toluene and 2-MeTHF. Reaction
conditions: c((R)-benzoin) = 20 g/L (94 mM), c(Zr-TUD-1) = 40 g/L, at 1100 rpm and
5
0 °C.
Figure 6. Continuous DKR of rac-benzoin catalyzed by immobilized Lipase TL and Zr-
TUD-1 in dry CPME. Reaction conditions: c(benzoin) = 10 g/L (47 mM), c(vinyl
butyrate) = 3 equiv., c(immoLipTL) = 16.7 g/L (100 mg immoLipTL), c(Zr-TUD-1) =
3
3.3 g/L (200 mg Zr-TUD-1) at 500 rpm, 0.5–1 mL/min flow rate and 50 °C.
25
Page 27 of 34

Figure 1
26
Page 28 of 34

Figure 2
27
Page 29 of 34