Unique role of water content in enzymatic synthesis of ethyl lactate using ionic liquid as solvent

Article abstract of DOI:10.2478/s11696-009-0117-4

Enzymatic synthesis of ethyl lactate was studied in organic solvents and in ionic liquids to determine optimal media for the reaction, and to investigate the effect of water content on the ester yield. Experiments proved that Cyphos 202 ionic liquid is the best solvent affording the highest ethyl lactate yields. Furthermore, 20 times less enzyme sufficed to carry out the reaction in this ionic liquid compared to organic solvents; ionic liquid could be applied as solvent. Using water removal, the ester yield decreased since a side reaction, dimerization of lactic acid, occurred as well. In contrast to these results, without water removal, the produced water was partly consumed by the decomposition of lactoyllactic acid to free lactic acid, increasing thus the substrate concentration of the mixture and enhancing the ester yield.

Full text of DOI:10.2478/s11696-009-0117-4

Chemical Papers 64 (2) 261–264 (2010)
DOI: 10.2478/s11696-009-0117-4
SHORT COMMUNICATION
Unique role of water content in enzymatic synthesis
of ethyl lactate using ionic liquid as solvent‡
Brigitta Major, Gergely Németh, Katalin Bélafi-Bakó, László Gubicza*
University of Pannonia, Research Institute of Chemical and Process Engineering, Egyetem u. 10, 8200 Veszprém, Hungary
Received 21 July 2009; Revised 21 October 2009; Accepted 23 November 2009
Enzymatic synthesis of ethyl lactate was studied in organic solvents and in ionic liquids to de-
termine optimal media for the reaction, and to investigate the effect of water content on the ester
yield. Experiments proved that Cyphos 202 ionic liquid is the best solvent affording the highest
ethyl lactate yields. Furthermore, 20 times less enzyme sufficed to carry out the reaction in this
ionic liquid compared to organic solvents; ionic liquid could be applied as solvent. Using water re-
moval, the ester yield decreased since a side reaction, dimerization of lactic acid, occurred as well.
In contrast to these results, without water removal, the produced water was partly consumed by the
decomposition of lactoyllactic acid to free lactic acid, increasing thus the substrate concentration
of the mixture and enhancing the ester yield.
c
ꢀ 2009 Institute of Chemistry, Slovak Academy of Sciences
Keywords: ionic liquid, esterification, enzyme, water removal, lactic acid
Water content has a particular role in ester synthe-
sis from organic acids (e.g. lactic acid (LA)) and short
alkyl alcohols (e.g. ethanol) using non-conventional
media. Esterification is an equilibrium reaction, hence
the maximal ester yield can be influenced either by
using an excess of one of the substrates (usually alco-
hol), or by removing the product, ester, or the side-
product, water. In case of enzymatic synthesis, water
content has also another role, since a monomolecular
water layer is necessary for the enzyme to maintain
its native conformation (Yang & Pan, 2005). Even so,
water removal can increase the ester yield effectively.
In our earlier works, the effectiveness of enzymatic
synthesis of ethyl acetate (Bélafi-Bakó et al. 2003; Gu-
bicza et al., 2008) and isoamyl acetate (Fehér et al.,
2009) was successfully enhanced by water removal us-
ing pervaporation or zeolite adsorption, respectively.
However, using LA as substrate, a new role of wa-
ter content emerges. In lower water content, LA un-
dergoes self esterification producing its open chain
dimer, lactoyllactic acid (LA2), and other oligomers
(Vu et al., 2005). Therefore, the commercially avail-
able 90 % LA solution contains a significant amount of
LA2 beside the monomeric form. Furthermore, dimer-
ization/decomposition of LA (Eq. (1)) has to be con-
sidered as a side reaction in the reaction mixture be-
side ethyl lactate (EL) synthesis (Eq. (2))
LA + LA ↔ LA2 + H2O
LA + EtOH ↔ EL + H2O
(1)
(2)
The reverse reaction (decomposition of LA2 to LA)
is very slow as highlighted by Eugin et al. (2003).
Therefore, it is the rate limiting step in the ester syn-
thesis, especially at higher conversions. In our recent
publication (Major et al., 2009), we have pointed out
that microwave irradiation enhances the decomposi-
tion of LA2 to LA, increasing thus the rate of ester
synthesis and achieving higher yield than that calcu-
lated from the monomeric acid content of the sub-
strate.
*Corresponding author, e-mail: gubicza@mukki.richem.hu
‡
Presented at the 36th International Conference of the Slovak Society of Chemical Engineering, Tatranské Matliare,
25–29 May 2009.

262
B. Major et al./Chemical Papers 64 (2) 261–264 (2010)
On the other hand, removal of the side-product,
Table 1. Comparison of EL yields using different media
water, during the esterification reaction may influence
the monomer LA content, and may also affect the ester
yield. To study this effect, pervaporation (Delgado et
al., 2008) and zeolite adsorption (Roenne et al., 2005)
have been investigated and they seem to be suitable
techniques for water removal.
Our aim was to compare the ester yield in two
different types of media (organic solvents, ionic liquids
(ILs)) and to determine the complex effect of water
content on the ethyl lactate production using the best
solvent.
Immobilized Candida antarctica lipase B (Novo-
zymꢀ435) was a generous gift from Novo Nordisk,
Denmark. All the used ILs (95 % purity), trihexyl-
tetradecyl-phosphonium-bis(2,4,4-trimethylpentyl)-
phosphinate (Cyphos 104), trihexyl-tetradecyl-phos-
phonium-bromide (Cyphos 102), trihexyl-tetradecyl-
Yield (%)a
Medium
With enzyme
Without enzyme
Hexane
Toluene
14
82
80
95
104b
90
74
60
36
0
1
2
Cyphos 104
Cyphos 202
Cyphos 163
Cyphos 166
Cyphos 106
Cyphos 102
Cyphos 110
2
102b
87
72
59
32
a) After 24 h; b) calculated on basis of the monomer acid con-
tent.
phosphonium-dodecylbenzene-sulfonate (Cyphos-202), dry ice/acetone mixture as the cooling agent.
trihexyl-tetradecyl-phosphonium-hexafluorophospha-
te (Cyphos 110), tetrabutyl-phosphonium-bromide
(Cyphos 163), tetraoktyl-phosphonium-bromide (Cy-
phos 166), and triisobutyl-methyl-phosphonium-tosy-
late (Cyphos-106), were bought from IoLiTec GmbH,
Germany. LA (90 mass % water solution), ethanol
(absolute), hexane, and toluene were purchased from
Spektrum 3D, Hungary. All other chemicals used were
of analytical grade. Composition of the LA solution
was determined both by acid-base titration and by
HPLC (53 mass % LA, 26 mass % LA2, 7 mass %
other acids, 14 mass % water). Esterification yields
were calculated on the basis of the monomer acid con-
tent.
Reactions without water removal were carried out
in a GFL 3031 type shaking incubator (150 min−1) at
40◦C for 24 h. Reactions using microwave conditions
(without water removal) were performed in a com-
mercial microwave equipment (Discover series, Bench-
Mate model, CEM Corporation, USA). The reaction
mixture was stirred using a magnetic stirrer, and a
fibre-optic sensor was used to monitor the tempera-
ture which was set by varying the microwave power.
For the esterification reactions of LA, 10 W energy
was used to keep the reaction temperature at 40◦C.
The volume and composition of the reaction mixture
were the same as under conventional conditions.
To study the effect of water content, an integrated
system was developed applying a pervaporation mem-
brane (active area: 12 cm2) or zeolite adsorption (40
g) for water removal. Reactions were carried out in a
bioreactor (volume of the reaction mixture of 50 cm3)
equipped with a magnetic stirrer (500 min−1), thermo-
stat (T = 40◦C), and cooling condenser. Reaction time
was 13 h. Reaction mixture was circulated between the
reactor and the membrane module/adsorber column
by a peristaltic pump (flow rate of 30 cm3 min−1). In
case of membrane separation, permeate was collected
under vacuum (pressure of 1 kPa) in a cold trap using
The applied STANDARD type flat sheet hy-
drophilic membrane was purchased from Beroplan
GmbH, Germany, and the 4 A molecular sieve from
˚
Zeolyst International, USA.
The reaction was monitored using a HP 5890 A
gas chromatograph equipped with a HP-FFAP column
(Macherey-Nagel, Germany) (head pressure of 80 kPa)
and a FID detector. Water content was measured us-
ing a Mettler DL 35 Karl-Fisher titrator. To analyze
the acid content of the reaction mixtures, HPLC was
applied using a MERC type equipment with a Zor-
bax SB-Aq 76 column, and a L-7450 detector, at the
monitoring wavelength of 215 nm.
Reaction conditions using organic solvents were as
follows: 0.2 mmol cm−3 LA, 1 mmol cm−3 ethanol,
4.5 mass % water in organic solvent; total volume:
5 cm3; enzyme: 250 mg per mmol LA. Using ILs, the
reaction conditions were the following: 1.3 mmol cm−3
LA, 9.3 mmol cm−3 ethanol, 3 mass % water in IL;
total volume: 1.5 cm3; enzyme: 12.5 mg per mmol LA.
From the reaction mixture, 50 µL samples were
taken. In case of IL media, EL was extracted by hex-
ane (4 × 80 µL) before the gas chromatography anal-
ysis. In case of organic solvents, no preparation was
used. Before the HPLC analysis, the samples were di-
luted with a phosphate buffer (5 mL, 6 % acetonitrile
content, pH 2.3). All reactions were carried out in trip-
licate, and the average error was less than 6 %.
Enzymatic EL synthesis was carried out in two
different organic solvents (hexane and toluene) and
in seven ILs (Table 1). The self-catalytic activity of
Cyphos 104, Cyphos 202, and of the organic solvents
was negligible since the final EL yield after 24 h of re-
action time was lower than 2 %, without enzyme load-
ing. The other ILs (Cyphos 163, Cyphos 166, Cyphos
106, Cyphos 102, and Cyphos 110) showed high cat-
alytic activity without enzyme loading.
The substrates and products were completely mis-
cible with the applied solvents, except for hexane and

B. Major et al./Chemical Papers 64 (2) 261–264 (2010)
263
Cyphos 110 therefore resulting in the lowest ester
yields of 14 % and 36 %, respectively, in 24 h of the
reaction time.
According to these preliminary investigations,
toluene and Cyphos 202 IL were found to be the most
suitable media for enzymatic esterification affording
the highest yields of esters (82 % and 95 %, respec-
tively). After comparing the optimal parameters of
the reaction in these two solvents, the following con-
clusions were drawn: i) in an IL medium, 20 times
less enzyme sufficed to complete the reaction and to
achieve the same EL yield than in toluene, since 250
mg per mmol LA of the enzyme were necessary in
toluene compared to only 12.5 mg per mmol LA in IL;
ii) Cyphos 202 and Cyphos 104 can be applied as sol-
vents reducing the amount of an expensive component
since the reaction can be carried out in a more concen-
trated solution (LA concentration was 0.2 mmol cm−3
in case of toluene and 1.3 mmol cm−3 in case of ILs).
According to these results, Cyphos 202 was proved to
be the best medium for EL production; further exper-
iments were carried out using this IL.
Fig. 1. Effect of water removal on ethyl lactate yield without
water removal (ꢀ), with pervaporation water removal
( ), and with water removal by adsorption (zeolite) (ꢁ)
(solvent: Cyphos 202; T = 40◦C).
To study the effect of water content, water removal
was performed using two different methods, which re-
sulted in different water removal rates. Pervaporation
represents the slower process removing 0.26 mol dm−3
of water in 13 h and zeolite adsorption the faster pro-
cess removing 0.61 mol dm−3 of water in the same
time (water concentration data are referred to the re-
action mixture). Using zeolite, the amount of adsorbed
water had to be calculated from a preliminary exper-
iment since the water content of the reaction mixture
was influenced by the water removal and other factors
(e.g. esterification reaction, dimerization of LA). So,
to estimate the adsorption capacity of the zeolite, a
mixture of ethanol and water (ϕr = 95 : 5) was de-
hydrated under the same conditions as those of the
reaction mixtures.
Reactions were started at optimized conditions,
and the obtained EL yields were compared (Fig. 1).
As can be seen from Fig. 1, in this case, water
removal decreased the ester yield unlike the water re-
moval aided by the ethyl acetate (Gubicza et al., 2008)
and the isoamyl acetate synthesis (Fehér et al., 2009)
using on-line water removal processes.
Fig. 2 shows that water content of the reaction
mixture was successfully decreased by the two pro-
cesses. However, using zeolite adsorption, the water
content was higher than in case of pervaporation.
These unique water content results are in accordance
with our hypothesis of water removal influencing the
LA monomer content of the reaction mixture by shift-
ing equilibrium (Eq. (1)) towards dimerization, where
water is formed, too.
Esterification reaction was also carried out using
microwave irradiation instead of conventional heating.
The mass balance of the reaction was calculated from
the water content and yield data (Table 2). Data in
Fig. 2. Water content of the reaction mixtures without water
removal (black columns), using pervaporation (white
columns), and using zeolite adsorption (gray columns)
(solvent: Cyphos 202; T = 40◦C).
Table 2 confirm two facts: i) without water control,
the amount of produced ester exceeded the amount of
produced water although the stoichiometry of the es-
terification reaction (Eq. (2)) predicts equal product
formation. This is in accordance with literature data
(Eugin et al., 2003) and confirms that the hydroly-
sis of LA2 (Eq. (1)) proceeds at a low rate under the
reaction conditions considered. Using microwave irra-
diation, this reaction is enhanced as it was described
earlier (Major et al., 2009); ii) in contrast to these
data, in both cases when water removal was used,
the amount of produced water was higher than the
amount of produced ester although the stoichiometry
of the reaction (Eq. (2)) predicts equal production.
The explanation of this phenomenon is as follows:
using water removal, the higher total water content
proved that, beside the EL synthesis, dimerization of
LA (Eq. (1)) occurred and water was formed resulting
in higher water than ester production. The higher wa-
ter removal rate was applied (by adsorption) the more
LA was dimerized and formed, respectively. Thus,
higher water production was observed in case of higher
water removal rates (zeolite adsorption). Furthermore,
since LA was consumed by this side-reaction, a lower
EL yield was observed than without water removal.
To verify our hypothesis, acid concentration of the
reaction mixture was analyzed by HPLC and the re-
sults for water adsorption by zeolite and ester produc-
tion were compared (Fig. 3). Fig. 3 shows that the con-

264
B. Major et al./Chemical Papers 64 (2) 261–264 (2010)
Table 2. Amount of total produced water and EL after 13 h of reaction time
Concentration (mol dm−3
Water
)
Reaction type
Ester
Total
In the reaction mixture
Removed
Total
Microwave irradiation
Shaking flask reaction
1.00
0.65
0
0
1.00
0.65
1.20
0.70
Pervaporation
Zeolite adsorption
0.24
0.46
0.26
0.61a
0.5
1.07
0.39
0.36
a) Estimated from the water control of the ethanol/water solution.
References
Bélafi-Bakó, K., Badr, A. K., Nemestóthy, N., Ehrenstein, U.,
& Gubicza, L. (2003). Kinetics of ethyl acetate formation by
lipase in organic solvent and solvent-free system. Chemical
Papers, 57, 278–281.
Delgado, P., Sanz, M. T., & Beltrán, S. (2008). Pervaporation
study for different binary mixtures in the esterification sys-
tem of lactic acid with ethanol. Separation and Purification
Technology, 64, 78–87. DOI: 10.1016/j.seppur.2008.08.002.
Eugin, A., Haluk, H., & Gurkan, K. (2003). Production of
lactic acid esters catalysed by heteropoly acid supported
over ion-exchange resins. Green Chemistry, 5, 460–466. DOI:
10.1039/b303327a.
Fehér, E., Major, B., Bélafi-Bakó, K., & Gubicza, L. (2009).
Semi-continuous enzymatic production and membrane as-
sisted separation of isoamyl acetate in alcohol – ionic liq-
uid biphasic system. Desalination, 241, 8–13. DOI: 10.1016/j.
desal.2007.11.080.
Gubicza, L., Bélafi-Bakó, K., Fehér, E., & Fráter, T. (2008).
Waste-free process for continuous flow enzymatic esterifica-
tion in ionic liquid medium. Green Chemistry, 10, 1284–1287.
DOI: 10.1039/b810009h.
Major, B., Kelemen-Horváth, I., Csanádi, Z., Bélafi-Bakó, K.,
& Gubicza, L. (2009). Microwave assisted enzymatic esterifi-
cation of lactic acid and ethanol in phosphonium type ionic
liquids as co-solvents. Green Chemistry, 11, 614–616. DOI:
10.1039/b820575b.
Fig. 3. Concentrations of lactic acids and ethyl lactate in the
reaction mixture using water adsorption by zeolite: LA
content (gray columns), LA2 content (white columns),
and ester content (black columns); solvent: Cyphos 202;
T = 40◦C.
centration of LA decreases during the reaction, while
the LA2 content increases. After 13 h, 0.66 mol dm−3
of LA was reacted but only 0.36 mol dm−3 of ester was
produced. The remaining part of LA was consumed by
the dimerization reaction increasing the LA2 content
of the reaction mixture as it was predicted by the wa-
ter content data (Table 2).
In conclusion, based on the experiments it was
found that the water content has a characteristic role
in the esterification of lactic acid with ethanol in
an ionic liquid solvent. While the removal of water,
formed in similar esterification reactions, e.g. synthe-
sis of ethyl acetate (Gubicza et al., 2008) or isoamyl
acetate (Fehér et al., 2009), enhanced the ester yield
significantly; in case of the ethyl lactate synthesis, the
effect was just the opposite. The presence of open
chain dimer lactoyllactic acid (beside the lactic acid
monomer) in the initial lactic acid and of water formed
during the esterification facilitated the hydrolysis of
dimers. It was proven that the removal of water de-
creased the ester yield in this reaction; therefore, water
should not be removed in this particular case.
Roenne, T. H., Xu, X., & Tan, T. (2005). Lipase-catalysed ester-
ification of lactic acid with straight-chain alcohols. Journal
of the American Oil Chemists’ Society, 82, 881–884. DOI:
10.1007/s11746-005-1159-1.
Vu, D. T., Kolah, A. K., Asthana, N. S., Peereboom, L., Lira,
C. T., & Miller, D. J. (2005). Oligomer distribution in con-
centrated lactic acid solutions. Fluid Phase Equilibria, 236,
125–135. DOI: 10.1016/j.fluid.2005.06.021.
Yang, Z., & Pan, W. (2005). Ionic liquids: Green solvents for
nonaqueous biocatalysis. Enzyme and Microbial Technology,
37, 19–28. DOI: 10.1016/j.enzmictec.2005.02.014.