Lipase-mediated synthesis of six-membered cyclic carbonates from trimethylolpropane and dialkyl carbonates: Influence of medium engineering on reaction selectivity

Article abstract of DOI:10.1016/j.molcatb.2011.07.019

Six-membered cyclic carbonates are potential monomers for aliphatic polycarbonates and polyurethanes in a process without using toxic phosgene and isocyanate. Lipase catalyzed transesterification of the polyol, trimethylolpropane (TMP) with dimethyl carbonate (DMC) or diethyl carbonate (DEC) followed by thermal cyclization was used for synthesis of six-membered cyclic carbonates with pendant hydroxyl and alkoxycarbonyloxyl groups. Immobilized lipase B from Candida antarctica (Novozym435) was used as the catalyst. Mixture of a hydrophilic solvent such as THF for high solubility of TMP, and a hydrophobic solvent such as toluene, were selected as the best solvent system for achieving high substrate conversion and selectivity. A relationship between polyol conversion and solvent hydrophobicity (log P) and solvent type, respectively, was established. THF:toluene system at a ratio of 0.5:1.0 (v/v) provided high degree of TMP conversion to product with high proportion of cyclic carbonates (>80%). The cyclic carbonate with pendant hydroxyl group was obtained with almost 85% selectivity at TMP conversion of 68.6% using 10% (w/w) Novozym435 at TMP:DMC ratio of 1:1. However, at TMP:DMC ratio of 1:5 and the same biocatalyst concentration, the TMP conversion was 100% with 72% selectivity for the cyclic carbonate with pendant alkoxycarbonyloxyl group. The product formed was without or with less content of linear carbonates, bis and tris(methoxycarbonyloxy)-TMP, as compared to that in a solvent-free system. The reactivity of DEC was lower than that of DMC. The reaction pathway leading to the formation of cyclic carbonate in this process comprised enzymatic carbonation of TMP with alkylcarbonates and thermal cyclization of linear carbonates. The process affords high degree of conversion of polyol to cyclic carbonates and provides a potentially attractive synthetic route for monomers of polycarbonates and polyurethanes.

Full text of DOI:10.1016/j.molcatb.2011.07.019

Journal of Molecular Catalysis B: Enzymatic 73 (2011) 67–73
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Lipase-mediated synthesis of six-membered cyclic carbonates from
trimethylolpropane and dialkyl carbonates: Influence of medium
engineering on reaction selectivity
Sang-Hyun Pyoa,∗, Katarzyna Nuszkiewicza, Per Perssonb, Stefan Lundmarkb, Rajni Hatti-Kaula
a Department of Biotechnology, Center for Chemistry and Chemical Engineering, Lund University, Box 124, SE-221 00 Lund, Sweden
b Perstorp AB, 284 80 Perstorp, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 May 2011
Received in revised form 26 July 2011
Accepted 26 July 2011
Available online 3 August 2011
Six-membered cyclic carbonates are potential monomers for aliphatic polycarbonates and polyurethanes
in a process without using toxic phosgene and isocyanate. Lipase catalyzed transesterification of the
polyol, trimethylolpropane (TMP) with dimethyl carbonate (DMC) or diethyl carbonate (DEC) followed
by thermal cyclization was used for synthesis of six-membered cyclic carbonates with pendant hydroxyl
and alkoxycarbonyloxyl groups. Immobilized lipase B from Candida antarctica (Novozym®435) was used
as the catalyst. Mixture of a hydrophilic solvent such as THF for high solubility of TMP, and a hydrophobic
solvent such as toluene, were selected as the best solvent system for achieving high substrate conversion
and selectivity. A relationship between polyol conversion and solvent hydrophobicity (log P) and solvent
type, respectively, was established. THF:toluene system at a ratio of 0.5:1.0 (v/v) provided high degree of
TMP conversion to product with high proportion of cyclic carbonates (>80%). The cyclic carbonate with
pendant hydroxyl group was obtained with almost 85% selectivity at TMP conversion of 68.6% using 10%
(w/w) Novozym®435 at TMP:DMC ratio of 1:1. However, at TMP:DMC ratio of 1:5 and the same biocatalyst
concentration, the TMP conversion was 100% with 72% selectivity for the cyclic carbonate with pendant
alkoxycarbonyloxyl group. The product formed was without or with less content of linear carbonates, bis
and tris(methoxycarbonyloxy)-TMP, as compared to that in a solvent-free system. The reactivity of DEC
was lower than that of DMC. The reaction pathway leading to the formation of cyclic carbonate in this
process comprised enzymatic carbonation of TMP with alkylcarbonates and thermal cyclization of linear
carbonates. The process affords high degree of conversion of polyol to cyclic carbonates and provides a
potentially attractive synthetic route for monomers of polycarbonates and polyurethanes.
Keywords:
Lipase-catalyzed transesterification
Medium engineering
Six-membered cyclic carbonate
Trimethylolpropane
Thermal cyclization
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
bonates [6]. Polyurethanes are produced using polyols, such as
alkanediols and glycerol, and isocyanate, which is derived from
Cyclic carbonates have attracted attention in recent years to
provide an alternative route for the production of aliphatic polycar-
bonates and polyurethanes by ring-opening polymerization [1–6].
These aliphatic polymers, besides having traditional applications in
engineering, optical devices, seatings, seals, coatings and high per-
formance adhesives, are expected to find applications in the field
of biomedicine because of their biocompatibility and low toxicity
[7–10]. Furthermore, they are considered to be environmentally
benign due to their promising features of biodegradability and
recyclability [11,12]. The process for their manufacture however
involves toxic chemicals.
the reaction between an amine and phosgene [13,14]. Since phos-
gene and low-molecular weight isocyanates have disadvantageous
toxicological profiles, attempts have been made to develop routes
to make polyurethanes from other sources however none of these
have yet been commercially established [14]. Recent studies have
reported on the syntheses of five- and six-membered cyclic carbon-
ates by different procedures, including addition of carbon dioxide to
epoxides using metal containing catalysts under pressure [1–4,15],
and by transesterification of polyols with dialkylcarbonate using
metal- [16], or enzymatic (lipase) catalysis [17,18]. Six-membered
cyclic carbonate is the preferred monomer for the ring-opening
polymerization process, since in contrast to the five-membered
monomer it is thermodynamically less stable than the ring-opened
polymer, and thus retains CO2 during the polymerization reaction
[16,19,20].
Aliphatic polycarbonates are currently produced industrially by
reacting an alkanediol with phosgene, triphosgene or dialkylcar-
In an earlier report, enzymatic synthesis of the six-membered
cyclic trimethylene carbonate from 1,3-diol and dimethyl or diethyl
∗
Corresponding author. Tel.: +46 46 222 4838; fax: +46 46 222 4713.
E-mail address: Sang-Hyun.Pyo@biotek.lu.se (S.-H. Pyo).
1381-1177/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2011.07.019

68
S.-H. Pyo et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 67–73
Table 1
Summary of the reactions between TMP and dialkylcarbonates (DMC or DEC) catalyzed by Novozym®435 under different conditions.
Run
N435 (%, w/w)
Solvent
Ratio of TMP:DMC
Reaction
Reaction
time (h)
TMP
conversion
(%)
Products (%)
Cyclic
carbonates
(4 + 6)
temp. (◦C)
3
4
5
6
7
1
2
3
4
5
6
7
8
9
10
11
12
13
50
50
50
20
20
10
10
10
5
10
10
10
10
ACN:toluene (0.5:1)
t-BuOH:n-hexane (0.5:1)
THF (1.5)
1:1.5
1:1.5
1:1.5
1:1.5
1:1.5
1:1
1:2
1:5
1:1
1:2
60
60
60
60
60
60
60
60
60
50
60
60
60
93
93
72
72
72
115
115
115
72
121
115
115
115
47.7
68.4
64.5
85.6
95.7
68.6
96.3
100
20.3
93.5
40.0
86.7
99.4
10.3
28.5
50.9
10.4
16.9
10.8
9.1
0.2
1.4
13.0
33.2
26.1
7.3
89.7
33.2
30.2
73.0
41.3
84.7
53.5
14.0
98.6
46.2
62.0
47.9
39.2
0
0
12.3
7.3
13.2
41.8
4.6
31.0
72.1
0
30.8
0
18.0
47.0
0
89.7
20.2
11.6
3.4
0
5.8 45.5
0
0
0
0
0
37.5
86.2
83.1
89.2
84.5
THF:toluene (1:0.5)
THF:toluene (0.5:1)
THF:toluene (0.5:1)
THF:toluene (0.5:1)
THF:toluene (0.5:1)
THF:toluene (0.5:1)
THF:toluene (0.5:1)
THF:toluene (0.5:1)
THF:toluene (0.5:1)
THF:toluene (0.5:1)
0
6.4
4.0
0
9.3
4.8
8.0
6.5
9.7 86.1
98.6
0.7 76.9
0
1:1 (DEC)
1:2 (DEC)
1:5 (DEC)
0
0
0
62.0
65.9
86.2
carbonate has been achieved in a solvent system of acetonitrile
and toluene (4:1, v/v) using very high concentration (600–900%,
w/w of the diol) of the immobilized Candida antarctica lipase B,
Novozym®435, however with moderate yield (53%) and low pro-
ductivity [18]. Synthesis of the cyclic carbonates with functional
groups from poly-functional alcohols such as trimethylolpropane
(TMP) and pentaerythritol (PE) have required more complicated
reaction steps and harsh conditions such as thermal dispropor-
tionation of the transesterification products followed by distillative
depolymerization both at >200 ◦C under reduced pressure, result-
ing in low yields (30%) [21].
We have recently investigated the synthesis of six-membered
cyclic carbonates with functional groups using lipase-mediated
reaction between TMP and dialkylcarbonate under solvent free
conditions [22]. The product formed was a mixture of cyclic and
linear (mono-, di- and tri-) carbonates, the proportions of which
depended on the reaction conditions used. Subjecting the prod-
uct mixture to thermal disproportionation at 60–80 ◦C without the
biocatalyst converted the linear carbonates to the cyclic ones. The
present study investigates the influence of different solvent types
and their hydrophobicity, and other reaction parameters on the
reaction efficiency and selectivity of cyclic carbonate production
during the reaction between the polyol and dialkylcarbonate cat-
alyzed by Novozym®435.
2. Experimental
2.1. Materials
Trimethylolpropane (TMP) was
a
product of Perstorp AB
(Sweden). Immobilized lipase
B from C. antarctica (CALB,
Novozym®435) was kindly provided by Novozymes A/S (Bagsvaerd,
Denmark). Dimethyl carbonate (97%), diethyl carbonate (97%),
˚
and molecular sieves (4 A beads, 8–12 mesh) were purchased
from Sigma–Aldrich (St. Louis, MO, USA). Tetrahydrofuran (THF),
toluene, n-hexane, tert-butanol and acetonitrile of HPLC grade
(<0.1% water) were purchased from MERCK (Germany). All chemi-
cals were used without further treatment.
2.2. Reaction procedures
The reactions were performed in 5 mL capped vials on a
thermomixer (MKR 13, HLC Biotech, Germany). Typically, 50 mg
(0.37 mM) TMP was dissolved in 1.5 mL solvent by mild shaking at
700 rpm at the defined reaction temperature, prior to addition of
0.2 g molecular sieves and certain amounts of dialkylcarbonate and
Novozym®435 (N435) to start the reaction. The amount of biocat-
alyst (w/w) was calculated based on the TMP amount used in the
reaction. Small aliquots were withdrawn from the reaction media at
O
O
O
O
OR
-ROH
OH
OH
O
2
OH OH
O
O
O
O
O
2, -ROH
RO
OR
Lipase, -ROH
Thermal
Enzymatic
OH
OH
Enzymatic
RO
4
6
3
TMP, 1
2,
O
Thermal
-ROH
Enzymatic
-ROH
O
O
O
OR
OR
RO
O
OH
O
O
-ROH
2,
Enzymatic
RO
RO
O
O
O
O
5
7
Scheme 1. Possible reaction pathway of synthesis of carbonates from TMP and DMC or DEC in solvent system by lipase-catalyzed and thermal reactions (R = methyl from
DMC or ethyl from DEC).

S.-H. Pyo et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 67–73
69
100
80
60
40
20
0
different time intervals for analysis of the polyol and the products.
Effects of solvent type, biocatalyst amount, molar ratio of substrates
and reaction temperature on the reaction were investigated.
A
2.3. Quantitative analyses and elucidation using GC, GC–MS and
NMR
Quantitative analyses were performed by gas chromatography
(GC, 430-GC, Varian, Palo Alto, USA) equipped with FactorFour
Capillary column, VF-1 ms (Varian, 15 M × 0.25 mm) and a flame
ionization detector. The initial column oven temperature was
increased from 50 to 250 ◦C at a rate of 20 ◦C/min. The samples
diluted at concentration of 0.1–0.5 mg/mL in acetonitrile were
injected in a split injection mode of 10% at 275 ◦C. The degree of
TMP conversion and concentration of products formed were calcu-
lated from the peak areas on the chromatograms. The identity of
the products was confirmed by measuring the molecular masses of
products by GC–MS (Varian 431-GC, Varian 210-MS) equipped with
FactorFour Capillary column, VF-5 ms (Varian, 30 M × 0.25 mm),
and structure elucidation by 1H and 13C NMR using 400 MHz NMR
(Bruker, UltraShield Plus 400, Germany) as described earlier [22].
0
20
40
60
80
100
Reaction time (h)
100
80
60
40
20
0
B
3. Results and discussion
3.1. Selection of solvent system for lipase-catalyzed reaction
between TMP and DMC
0
20
40
60
80
100
Although the most desirable strategy from a green chemistry
perspective is to carry out the reaction in a solvent free medium,
different solvents can influence both the selectivity and yield of
the reaction [23,24]. It is also known that the stability of an
enzyme is usually higher in a more hydrophophic solvent, i.e. with
higher log P (logarithm of partition coefficient between water and
octanol). In contrast, solvents with lower log P value such as pyri-
dine, dimethylformamide (DMF) and acetonitrile can solubilise
many polar molecules like TMP, but often inactivate the enzyme
by their ability to remove water of hydration, and promote accu-
mulation of water in the reaction media [25–28].
To investigate the effect of the solvents, THF, t-butanol and
acetonitrile were chosen as hydrophilic solvents, and toluene and
n-hexane as the hydrophobic ones. These solvents were used indi-
vidually and in mixture systems for reactions at 60 ◦C. In pure
solvents, the highest TMP conversion of 60% was obtained in THF
after 72 h reaction using 40% (w/w) N435, while in t-butanol and
acetonitrile the highest conversion values obtained were 44% and
12%, respectively, after 92 h of reaction with the same concen-
tration of the biocatalyst. TMP was generally not soluble in the
hydrophobic solvents, n-hexane and toluene, and there was practi-
cally no reaction with DMC in these solvents. The conversion of TMP
was significantly increased by using mixtures of hydrophilic and
hydrophobic solvents (Fig. 1). THF is freely miscible with toluene,
and varying the ratio of these solvents showed that the reac-
tivity was improved with decreasing THF:toluene ratio (Fig. 1A).
At THF:toluene ratio of 0.5:1, conversion of 94% was achieved
already after 24 h using 20% (w/w) biocatalyst, while at a ratio
of 1.0:0.5 similar degree of conversion was reached after 72 h of
reaction. Other solvent mixtures, t-butanol/n-hexane and acetoni-
trile/toluene, gave significantly lower conversions (Fig. 1B and C).
Fig. 2 shows TMP conversion obtained after 92 h of reactions
with respect to the log P of different solvent mixture systems. The
log P of the solvent mixtures was calculated from the log P values of
the component solvents and the ratio of the solvents used. It is clear
from the figure that for each solvent mixture the TMP conversion
was increased with increase in log P but the extent of conversion
in the THF/toluene system was significantly higher than that in the
Reaction time (h)
100
80
60
40
20
0
C
0
20
40
60
80
100
Reaction time (h)
Fig. 1. Selection of solvent system for synthesis of cyclic carbonates by lipase medi-
ated reaction between TMP and DMC at a ratio of 1:1.5, 60 ◦C. Profiles of TMP
conversion (A) using 20% (w/w) N435 in THF:toluene system, at ratios of 0.5:1.0
(ꢀ), 1.0:0.5 (ꢁ), and 1.5:0 (ꢂ); (B) using 50% (w/w) N435 in t-butanol:n-hexane sys-
tem at ratios of 0.15:1.35 (ꢀ), 0.5:1.0 (ꢁ) and 1.5:0 (ꢂ); (C) using 50% (w/w) N435 in
acetonitrile:toluene at ratios of 0.5:1.0 (ꢀ), 1.0:0.5 (ꢁ) and 1.5:0 (ꢂ).
other solvent mixtures under conditions of similar log P. An earlier
study by Tasaki et al. reported the synthesis of cyclic trimethy-
lene carbonate in a solvent system of acetonitrile:toluene (4:1, v/v),
which has a log P value of 0.236 that is in the lower end of the curve
for the solvent mixture in Fig. 2. This was expected to give low con-
version yield, which was actually 53% in the report [18]. However,
besides log P even other features of the solvent systems affecting
the reactants and products (such as solubility, miscibility, and sta-
bilization) and the biocatalyst (e.g. stability and activity) influence
the reaction rate and yield.
Besides the substrate conversion, the product profile for the dif-
ferent systems was also determined. Table 1 shows the degree of
conversion and ratio of different products formed in some selected

70
S.-H. Pyo et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 67–73
100
80
60
40
20
0
product formed was 4 but the TMP conversion was as low as 48%
(w/w) using 50% (w/w) N435 (Run 1). The system of t-butanol and
hexane gave low product selectivity besides incomplete conver-
sion (Run 2). THF:toluene system (Runs 4–10) provided high yields
(except at equimolar ratio of TMP:DMC) as well as high proportion
of cyclic carbonate products 4 and 6 (>80%). Based on these obser-
vations, mixture of THF:toluene at a ratio of 0.5:1 was selected as
the best solvent system, and employed for further optimization of
the processes.
3.2. Analysing the influence of temperature on cyclic carbonate
formation
Reducing the reaction temperature from 60 ◦C to 50 ◦C reduced
both the reaction rate as well as the proportion of cyclic carbonates
formed (for comparison see Runs 7 and 10 in Table 1). This is in
line with our earlier observations in the solvent free system, which
indicated that cyclization of the linear carbonate, formed by lipase
catalysis was promoted at higher temperature [22]. Since THF has
a boiling point of 66 ◦C and N435 is not very stable above 60 ◦C,
higher temperatures were not investigated in the present study.
In order to distinguish between the enzymatic and tempera-
ture effects, the reaction mixture obtained after a certain time of
biocatalyst treatment was further processed with or without the
biocatalyst. For this, 50 mg (0.37 mM) TMP was reacted with 75 mg
(0.83 mM) DMC in THF:toluene (0.5:1.0 mL) using 5 mg N435 and
0.2 g molecular sieves at 60 ◦C for 22 h (Fig. 3A). Subsequently, pro-
cessing of one part of the reaction mixture was continued under the
same conditions, while the other part was centrifuged to remove
-0.5
0.5
1.5
2.5
3.5
log P
Fig. 2. Relationship between TMP conversion (at 92 h) and hydrophobicity (log P)
of THF:toluene system (ꢀ), t-butanol:n-hexane system (ꢁ) and acetonitrile:toluene
(ꢂ). Data obtained in Fig. 1 is used.
systems. In majority of the reactions (except in t-butanol:n-hexane
and THF systems, Runs 2 and 3), cyclic carbonates of TMP, product
4 (with pendant hydroxyl group) and/or product 6 (with pendant
methoxycarbonyloxyl group) were the predominant products with
either none or low proportions of the linear carbonate derivatives
of TMP (mono-(3), bis (5) and tris(methoxycarbonyloxy) (7)-TMP)
(Scheme 1). In the acetonitrile:toluene system, nearly 90% of the
Fig. 3. Gas chromatograms of reaction products prepared by N435-catalyzed reaction between TMP and DMC, and thermal reaction. Peaks were assigned by GC–MS and 1
H
NMR in previous report [22]. Peaks 1, 3, 4, 5, 6, and 7 indicate the structures shown in Scheme 1. Reaction solution A, prepared by N435-catalyzed reaction at 60 ◦C for 22 h,
was subjected to further reaction under initial conditions and analysed after 4 h (B) and 40 h (C). Solution A was incubated at 60 ◦C after removal of N435 and analysed after
4 h (D) and 40 h (E).

S.-H. Pyo et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 67–73
71
100
100
80
60
40
20
0
A
80
60
40
20
0
0
20
40
60
80
100
120
Reaction time (h)
0
20
40
60
80
100
120
100
80
60
40
20
0
Reaction time (h)
B
Fig. 4. Effect of the amount of Novozym®435 on synthesis of cyclic carbonates at
TMP:DMC ratio of 1:1 (0.37 mM) at 60 ◦C in THF:toluene (0.5:1.0) system. N435 was
used at concentrations of 2.5% (x), 5% (ꢂ),10% (ꢁ), and 20% (ꢀ) (w/w), respectively.
solids including the immobilized lipase preparation, and then sup-
plemented with 0.2 g molecular sieves prior to incubation at 60 ◦C.
In the reaction with the biocatalyst, the remaining TMP was further
converted to different products (Fig. 3B and C). On the other hand,
the TMP level remained unchanged in the reaction without the bio-
catalyst, and product 3 was decreased with corresponding increase
in the cyclic carbonate product 4, and product 5 was decreased with
increase in 6 (Fig. 3D and E). Further conversion of 4 to 6 did not take
place. Therefore the cyclizations from 3 to 4, and from 5 to 6 seem to
be caused by thermal reaction at 60 ◦C. This would also imply that
in the sample containing the lipase, TMP is sequentially converted
to 3, 5 and 7, while 4 is converted to 6 (Fig. 3B and C, Scheme 1).
This indicates that only dialkylcarbonate served as the acyl-donor
for the lipase active site, while mono- and di-carbonated TMP (3,
5), and cyclic carbonate (4) were acyl-acceptors in the lipase cat-
alyzed reaction. The reaction conditions providing higher rate of
thermal cyclization led to a higher proportion of the desired cyclic
carbonate products.
0
20
40
60
80
100
120
Reaction time (h)
100
80
60
40
20
0
C
3.3. Effect of biocatalyst concentration and TMP–DMC ratio in
THF/toluene system
0
20
40
60
80
100
120
Reaction time (h)
The amount of biocatalyst used is one of the important param-
eters for determining reaction yield, selectivity of products and
process cost. N435 was used at 2.5, 5, 10, and 20% (w/w) of
TMP in THF:toluene (0.5:1.0) system at 60 ◦C. As shown in Fig. 4,
the reactions employing higher biocatalyst amounts (10 and 20%,
w/w) showed increasing TMP conversion with time, reaching 78%
with 10% (w/w) N435 after 120 h of reaction, and 86% with 20%
(w/w) biocatalyst after 72 h. The product yield per gram of bio-
catalyst was however higher, and showed higher proportion (85%)
of product 4 with 10% (w/w) N435. The reaction yields achieved
are also much higher with significantly lower biocatalyst concen-
tration than those reported earlier for the six-membered cyclic
trimethylene carbonate (1,3-dioxane-2-one) monomer [18] and
five membered cyclic glycerol carbonate [17].
100
80
60
40
20
0
D
0
20
40
60
80
100
120
The ratio of two reactants also influenced the reaction and
proportion of products as shown in Fig. 5. At TMP:DMC ratios of
1:5, 1:2 and 1:1, TMP conversion reached 100%, 96.3% and 68.6%,
respectively, after 115 h of reaction (Fig. 5A). At TMP:DEC of 1:5,
the conversion had reached 95% already at 48 h. The product pro-
portions were very different with reaction time for the different
substrate ratios. For the TMP:DMC ratio of 1:1, the proportion of
the cyclic carbonate 4 increased with time reaching 84.7% at 115 h
(Table 1 (Run 6), Fig. 5B). With increasing concentration of DMC,
Reaction time (h)
Fig. 5. Effect of TMP:DMC ratio on (A) TMP conversion, (B) formation of cyclic car-
bonate 4, (C) formation of cyclic carbonate 6, and (D) 4 and 6, using 10% (w/w) N435
at 60 ◦C in THF:toluene (0.5:1.0). The ratios of TMP:DMC used were 1:5 (ꢀ), 1:2 (ꢂ)
and 1:1 (ꢁ), respectively.

72
S.-H. Pyo et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 67–73
100
80
60
40
20
0
the profiles of TMP conversion and products in the reaction with
A
TMP:DEC ratio of 1:5 using 10% N435 at 60 ◦C. The highest pro-
portion of the cyclic carbonate 4 in the reaction was 40% at 30 h
and subsequently remained quite stable while the cyclic carbonate
with pendant ethoxycarbonyloxyl group, 6ꢀ level increased grad-
ually to just over 40%. This implies also a slower reaction rate in
conversion of 4 to 6ꢀ in comparison with the reaction with DMC at
a similar reactant ratio (see Fig. 4B). At 115 h, 99% of TMP was con-
sumed and the product mixture contained predominantly (>86%)
4 and 6 (in almost equal proportions) with minor contents of 3ꢀ
and 5ꢀ (Fig. 5B). The slower reaction rate with DEC could be due to
the lower leaving rate of the acyl group for DEC as compared with
DMC at 60 ◦C, and perhaps lower adsorption (mass transfer) rate of
0
20
40
60
80
100
120
˚
resulting ethanol into pore (4 A) of molecular sieves.
Reaction time (h)
4. Conclusion
100
80
60
40
20
0
B
Synthesis of functional six-membered cyclic carbonates by
lipase-mediated reaction between TMP and DMC or DEC was
achieved with high yield in a single step by using a mixture of
hydrophilic and hydrophobic solvents. A useful relationship of sol-
vent hydrophobicity (log P) and solvent type with conversion yield
was established to choose a suitable solvent system. The propor-
tion of the cyclic carbonate products, 4 and 6 were particularly
influenced by the ratio of TMP and dialkylcarbonate used. The
reaction starts with the formation of a linear mono-carbonated
derivative of TMP, which undergoes further conversion to other
linear carbonates with higher degree of substitution and cyclic
carbonates depending on the reaction conditions. The reaction con-
ditions providing highest TMP conversion rate and cyclic carbonate
yield were the solvent mixture comprising THF:toluene at a ratio
of 0.5:1, TMP:dialkylcarbonate at 1:5 ratio and 10% Novozym®435.
In order to further enhance the rate of cyclic carbonate forma-
tion, especially product 4, a useful strategy would be to obtain
rapid conversion to 3 and subjecting the product to high temper-
ature. Further development of this synthetic route is in progress
for production of polyurethane without using toxic phosgene and
isocyanate.
0
20
40
60
80
100
120
Reaction time (h)
Fig. 6. Effect of TMP:DEC ratio on synthesis of cyclic carbonates using 10% (w/w)
N435 at 60 ◦C in THF:toluene (0.5:1.0) system. (A) TMP conversion (%) at TMP:DEC
ratios of 1:5 (ꢀ), 1:2 (ꢂ) and 1:1 (ꢁ), respectively. (B) Conversion of TMP (ꢀ), and
production of 3 (ꢂ), 4 (x), 5 ( ) and 6 (ꢁ) in the reaction at TMP:DEC ratio of 1:5.
the content of product 4 decreased after an initial increase. At
TMP:DMC ratio of 1:5, product 4 constituted the highest propor-
tion of 50% after 24 h reaction (at TMP conversion of 80%), and
subsequently decreased to 14%. This was attributed to its further
conversion to the other cyclic carbonate derivative, product 6 by
the lipase catalyzed transesterification at the remaining hydroxyl
group with excess DMC. This was seen in increase in the content
of product 6 to 72.1% (Table 1 (Run 8), Fig. 5C). Regardless of the
TMP:DMC ratio, the total amount of cyclic carbonates, 4 and 6 was
maintained over 80% (Table 1, Fig. 5D).
Acknowledgement
The work was performed within an interdisciplinary program
“Greenchem” supported financially by Mistra (The Swedish Foun-
dation for Strategic Environmental Research).
References
3.4. Comparison of reactivity and selectivity between DMC and
DEC
[1] B. Barkakaty, K. Morino, A. Sudo, T. Endo, Green Chem. 12 (2010) 42–44.
[2] K. Motokura, S. Itagaki, Y. Iwasawa, A. Miyaji, T. Baba, Green Chem. 11 (2009)
1876–1880.
[3] J. Song, Z. Zhang, S. Hu, T. Wu, T. Jiang, B. Han, Green Chem. 11 (2009)
1031–1036.
Although DMC and DEC undergo the same type of reaction
with TMP, differences in their physicochemical properties such
as molecular size and hydrophobicity influence their interaction
with the enzyme. As observed in the solvent-free system [22],
the reactivity of DEC with TMP was lower than that of DMC
(Fig. 6A, Table 1). For example, at substrate ratio (TMP:DEC or
DMC) of 1:1 and 1:2 at 60 ◦C, the conversion using 10% N435 after
115 h of reaction with DEC reached 40% and 86.7%, respectively
(Fig. 6), while the corresponding conversions with DMC were 68.9%
and 96.3%, respectively (Fig. 5). Additionally, further conversion
of the initially formed product 3ꢀ (mono-carbonated TMP) to the
other products progressed slowly with DEC, and 33 and 26% of 3ꢀ
remained in the product at TMP:DEC ratio of 1:1 and 1:2, respec-
tively (Table 1). Highest degree of TMP conversion (99.4%) was
achieved at TMP:DEC ratio of 1:5 which was similar to that with
DMC but at a slower rate (Table 1, Figs. 5A and 6A). Fig. 6B shows
[4] M.M. Dharman, J.-I. Yu, J.-Y. Ahn, D.-W. Park, Green Chem. 11 (2009)
1754–1757.
[5] L. Liao, C. Zhang, S. Gong, Eur. Polym. J. 43 (2007) 4289–4296.
[6] J. Fahlén, R. Martinsson, B. Midelf, Eur. Coat. J. 6 (2007) 28–32.
[7] J. Xu, Z.L. Liu, R.X. Zuho, J. Appl. Polym. Sci. 101 (2006) 1988–1994.
[8] Z. Zhang, R. Kuijer, S.K. Bulstra, D.W. Grijpma, J. Feijen, Biomaterials 27 (2006)
1741–1748.
[9] Y. Yamamoto, S. Kaihara, K. Toshima, S. Matsumura, Macromol. Biosci. 9 (2009)
968–978.
[10] S.A. Guelcher, Tissue Eng. B 14 (2008) 3–17.
[11] T. Suyama, Y. Tokiwa, Enzyme Microb. Technol. 20 (1997) 122–126.
[12] T. Suyama, H. Hosoya, Y. Tokiwa, FEMS Microbiol. Lett. 161 (1998) 255–261.
[13] G.T. Howard, Int. Biodeterior. Biodegrad. 49 (2002) 245–252.
[14] S. Matsumura, Y. Soeda, K. Toshima, Appl. Microbiol. Biotechnol. 70 (2006)
12–20.
[15] D.J. Darensbourg, A. Horn Jr., A.I. Moncada, Green Chem. 12 (2010) 1376–1379.
[16] A.-C. Albertsson, M.J. Sjoeling, Macromol. Sci. Pure Appl. Chem. A29 (1992)
43–54.

S.-H. Pyo et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 67–73
73
[17] S.C. Kim, Y.H. Kim, H. Lee, D.Y. Yoon, B.K. Song, J. Mol. Catal. B: Enzym. 49 (2007)
75–78.
[24] P.L.A. Overbeeke, J.A. Jongejan, J.J. Heijnen, Biotechnol. Bioeng. 70 (2000)
278–290.
[18] H. Tasaki, K. Toshima, S. Matsumura, Macromol. Biosci. 3 (2003) 436–441.
[19] S. Kéki, J. Török, G. Deák, M. Zsuga, Macromolecules 34 (2001) 6850–6857.
[20] H.R. Kricheldorf, B. Weegen-Schulz, Polymer 36 (1995) 4997–5003.
[21] G. Rokicki, T. Kowalczyk, M. Glinski, Polym. J. 32 (2000) 381–390.
[22] S.-H. Pyo, P. Persson, S. Lundmark, R. Hatti-Kaul, Green Chem. 13 (2010)
976–982.
[25] S.H. Pyo, D.G. Hayes, Biobased Surfactants and Detergents: Synthe-
sis, Properties, and Applications, AOCS Press, Champaign, IL, 2009, pp.
323–350.
[26] P. Adlercreutz, in: G. Carrea, S. Riva (Eds.), Organic Synthesis with Enzymes in
Non-aqueous media, Wiley-VCH, Weinheim, 2008, pp. 3–22.
[27] D.G. Hayes, J. Am. Oil Chem. Soc. 81 (2004) 1077–1103.
[28] P. Degn, W. Zimmermann, Biotechnol. Bioeng. 74 (2001) 483–491.
[23] Z.-Q. Duan, W. Du, D.-H. Liu, Bioresour. Technol. 101 (2010) 2568–2571.