1 s2.0 S1381117704001791 Main
1 s2.0 S1381117704001791 Main
1 s2.0 S1381117704001791 Main
Received 16 March 2004; received in revised form 6 May 2004; accepted 7 May 2004
Abstract
The use of ionic liquids as reaction media for lipase-catalyzed enantioselective acylation of 1-phenylethylamine (1) and 2-phenyl-1-
propylamine (2) with 4-pentenoic acid was investigated. The best performing ionic liquid for each of these amines as well as its solvent
properties were very different. Preparative scale kinetic resolution of 1 was performed efficiently in 1-butyl-2,3-dimethylimidazolium triflu-
oromethanesulphonate.
© 2004 Elsevier B.V. All rights reserved.
Keywords: Enantioselective amine acylation; Ionic liquids; Lipase; 4-Pentenoic acid; Solvent properties
1381-1177/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2004.05.003
190 R. Irimescu, K. Kato / Journal of Molecular Catalysis B: Enzymatic 30 (2004) 189–194
Several excellent reviews on this topic were published re- ture (40 ◦ C for 1 or 30 ◦ C for 2) and connecting the system
cently [9–11]. Unlike organic solvents, ionic liquids have to a vacuum pump through a cooling trap. The pressure was
virtually no vapor pressure and therefore vacuum can be maintained at 5 mmHg.
used to remove volatile products: methanol in [12], water The preparative scale kinetic resolution of amine 1 was
in our system. Ionic liquids are polar solvents that do not performed by adding 0.200 g CALB catalyst to a mixture
inactivate enzymes like the organic solvents with similar 1 (0.484 g, 4 mmol) and 4-pentenoic acid (300 mg, 3 mmol)
polarities and therefore enzymatic reactions in more polar in [bdmim]tfms ionic liquid (2 mL) and kept under stirring
solvents became possible. The more polar ionic liquids pro- at 40 ◦ C and 5 mmHg for 72 h. The reaction mixture work
moted lipase-catalyzed reactions with polar substrates [11]. up was carried out as described below for sample prepara-
In the present work, we attempted to improve the ef- tion. The enantiomeric excess was determined to be 91%
ficiency of the amine acylation reactions in ionic liquids for amine 3 (0.203 g, 84% yield) and >99% for amide 4
taking into consideration the solvent properties of the ionic (0.328 g, 81% yield) by HPLC analysis.
liquids used as reaction media. The effect of the alteration
of the cation and anion correlated with the polarity and 2.3. Analysis
nucleophilicity of a range of ionic liquids was investigated.
At intervals, samples (0.15 mL) were withdrawn from the
reaction mixture and extracted twice with diethyl ether. The
2. Materials and methods combined extract was treated with HCl solution (2 N) that
dissolved the amine as a chloride salt. The organic layer
2.1. Materials containing the amide and unreacted acid was removed and
the water layer was washed with diethyl ether. The ether
The immobilized CALB catalyst (Chirazyme L-2, c.-f, solutions were united and the acid was removed by washing
C2, Lyo.) (EC 3.1.1.1.) was purchased from Roche Diagnos- with NaOH solution (2 N). The organic layer containing
tics (Mannheim, Germany). R-, S-, and rac-1-phenylethyla- only the amide was collected, dried over MgSO4 and the
mine (all >98%) (1), and 4-pentenoic acid (>98%) were solvent evaporated. The water extract with the amine salt
bought from Wako Pure Chemical Industries Ltd. (Os- was neutralized with NaOH solution (2 N) and extracted
aka, Japan). Rac-2-phenyl-1-propylamine (>98%) (2) twice with diethyl ether. The combined extract containing
was from Avocado Research Chemicals Ltd. (Heysham, the free amine was dried over dried over MgSO4 and the
UK). R- and S-2-phenyl-1-propylamine (both >99%), solvent evaporated.
1-butyl-4-methylpyridinium tetrafluoroborate ([bmp]BF4 ) The enantiomeric compositions of the residual amine
(>97%), 1-butyl-3-methylimidazolium hexafluorophosphate 5 was determined by HPLC on a Chiralpak AD-H (Dicel
([bmim]PF6 ) (>97%) and 1-octyl-3-methylimidazolium Chemical Industries, Tokyo, Japan) column eluted with
tetrafluoroborate ([omim]BF4 ) (>97%) were purchased from a mixture of hexane/2-propanol/diethylamine (95:5:0.05,
Fluka Chemie GmbH (Buchs, Switzerland). 1-Ethyl-3-meth- v/v/v), at a flow rate of 0.5 mL/min at 20 ◦ C with UV de-
ylimidazolium tetrafluoroborate ([emim]BF4 ) and 1-ethyl- tection at 254 nm. The enantiomeric excess of amine 3 was
3-methylimidazolium trifluoromethanesulphonate ([emim]- determined after its transformation into the corresponding
tfms) were bought from Tokyo Kasei Kogyo (Tokyo, acetamide (acetic anhydride, pyridine). The enantiomeric
Japan). 1-Butyl-2,3-dimethylimidazolium trifluoromethane- compositions of the amide products 4 and 6 as well as
sulphonate ([bdmim]tfms), 1-butyl-3-methylimidazolium the acetamide derivatives of amine 1 was determined on
tetrafluoroborate ([bmim]BF4 ), 1-hexyl-3-methylimidazol- the above mentioned column eluted with a mixture of
ium tetrafluoroborate ([hmim]BF4 ), 1-butyl-2,3-dimethyli- hexane/2-propanol (95:5, v/v) under the same conditions.
midazolium tetrafluoroborate ([bdmim]BF4 ), 1-hexyl-2,3- The (R)-enantiomer was preferentially acylated for amine
dimethylimidazolium tetrafluoroborate ([hdmim]BF4 ), 1- 1 while for amine 2, it was the (S)-enantiomer. The abso-
hexyl-3-methylimidazolium hexafluorophosphate ([hmim]- lute configuration of amine 5 was determined by comparison
PF6 ) and 1-octyl-3-methylimidazolium hexafluorophos- with the retention times of the pure enantiomer standards
phate ([omim]PF6 ) were from Acros Organics (New Jersey, on chiral HPLC. The (S)-configuration of amine 3 isolated
USA). 4-Pentenoic anhydride (>99%) was purchased from from the reaction mixture and transformed in acetamide was
Aldrich (Milwaukee, USA). Other chemicals used were of determined after comparison of the retention times with the
analytical grade. enantiomerically pure (R)- and (S)-acetamides of 1 prepared
chemically from pure enantiomers of 1. The absolute con-
2.2. Lipase-catalyzed acylation of amines in ionic liquids figuration of the formed amides 4 and 6 was determined
similarly using chemically prepared amide enantiomer stan-
The reaction was started by adding the biocatalyst dards as described below.
(50 mg) to a mixture of amine (1 mmol) and 4-pentenoic The amine acylation was performed by adding 10 L
acid (0.75 mmol for 1 or 1.5 mmol for 2) dissolved in an amine to a mixture of 0.3 mL anhydride (acetic or 4-pente-
ionic liquid (1.5 mL) under stirring at a specified tempera- noic anhydride) and 0.1 mL pyridine in a 15 mL tube with
R. Irimescu, K. Kato / Journal of Molecular Catalysis B: Enzymatic 30 (2004) 189–194 191
Scheme 1.
a screw cap. The tube was heated at 80 ◦ C for 30 min. Two available data describing their solvent properties found in the
milliliters of water was then added and the tube heated at literature on this topic were summarized in Table 2 [14–16].
the same temperature for 10 min. The reaction mixture was Solvatochromic probes were used to measure specific
extracted twice with diethyl ether. The organic extract was solvent–solute interactions in ionic liquids [14]. The normal-
washed with NaOH solution (2 N) to remove the acid, then ized polarity scale, ETN ,based on solvatochromic interactions
with HCl solution (2 N) to remove the pyridine traces, dried of Reichardt’s betaine dye (Table 2, column 3) is largely a
over MgSO4 , and finally, the solvent was evaporated. The measure of the hydrogen bond donor strength or hydrogen
concentrate was dissolved in the solvent used for HPLC bond acidity of the system [16], and its values range from
analysis. 0 for tetramethylsilane to 1.0 for water. The polarity scale,
The conversion, c, and the enantioselectivity, E, were cal- ENR ,obtained using the solvatochromic dye Nile red [17]
culated from the enantiomeric excess of the product, eeP , measures solvent dipolarity/polarizability (Table 2, column
and the substrate, eeS , according to Rackels et al. [13] using 4). The hydrogen bond acceptor ability, or nucleophilicity
the equations below. of the ionic liquids was correlated with the wave length
eeS value, λmax , for the lowest energy d–d absorption band at
c= (1) solvation of acetylacetonatotetramethylethyldiaminecop-
eeS + eeP
per(II) tetraphenylborate: [Cu(acac)(tmen)[BPh4 ] (Table 2,
ln[(1 − eeS )/(1 + eeS /eeP )] column 6) [16]. The water miscibility information was ob-
E= (2)
ln[(1 + eeS )/(1 + eeS /eeP )] tained from the manufacturer’s product description (Table 2,
column 6).
The polarity data show that the ionic liquids employed in
3. Results and discussion this study are displaying hydrogen bond acidity and dipo-
larity/polarizability comparable to lower alcohols (Table 2,
It is well known that the role of the solvent can be crucial entries 16–18), while their nucleophilicity varies in a larger
for heterolytic reactions in organic synthesis and also that it range and seems to be entirely anion dependent (Table 2).
can influence drastically the activity of lipases. Our previous The available polarity data are very limited at the present
study on enantioselective acylation of 1-phenylethylamine and there were no available data for several ionic liquids
(1) and 2-phenyl-1-propylamine (2) catalyzed by CALB employed in this work (Table 2, entries 1, 3, 5). To help
(Scheme 1) showed that the reaction rates improved dra- with the discussion and data interpretation, the polarity data
matically when the reaction was performed in [bmim]PF6 of two ionic liquids not tried in the experiments here were
compared to the solvent-free system [8]. added to Table 2 (entries 2, 6). The water miscibility of ionic
However, the reaction of (1) with 4-pentenoic acid re- liquids is different from that of lower alcohols and depends
mained still sluggish. In this study, we intended to further on the substituents on the cation.
improve the performance of the ionic liquid system by mod- The present extended study showed that [bmim]PF6 , used
ifying the properties of the ionic liquids by changes of the in our previous work [8], is not the best reaction media for
anion and cation. Some of the ionic liquids that gave good re- CALB catalyzed acylation of 1 (Table 1, entry 9). The best
sults for lipase-catalyzed reactions in previous studies were reaction rate, more than double of that for [bmim]PF6 , was
investigated here (Table 1). The reactions were performed given by [bdmim]tfms (Table 1, entry 2). Unfortunately, no
at an optimized amine/acid molar ratio for each amine. polarity data for [bdmim]tfms were found and therefore only
The ionic liquids chosen are based on 1-alkyl-3-methyl- some considerations based by interpolation are possible.
imidazolium and 1-alkyl-2,3-dimethylimidazolium cations Taking into consideration that the hydrogen bond acidity
combined with the [BF4 ]− , [PF6 ]− and [CF3 SO3 ]− decreases when the 2-position of the imidazolium cation is
= [tfms]− anions. In addition, 1-butyl-4-methylpyridinium replaced by a methyl group (see the values for [bdmim]BF4
tetrafluoroborate ([bmp]BF4 ) was also investigated. The and [bmim]BF4 , entries 4 and 8, respectively, of Table 2)
192 R. Irimescu, K. Kato / Journal of Molecular Catalysis B: Enzymatic 30 (2004) 189–194
Table 1
Lipase-catalyzed enantioselective acylation of primary amines with 4-pentenoic acid
Entry Amine Ionic solvent Initial rate Reaction Conversion eeS b (%) eeP c (%) E
(mol/h mg)a time (h) amine (%)
1 1 [emim]tfms 0.153 24 18.4 22.6 >99.0 >500
2 1 [bdmim]tfms 0.177 24 21.2 27.0 >99.0 >500
3 1 [bdmim]BF4 0.113 24 13.6 15.8 >99.0 >500
4 1 [hdmim]BF4 0.027 24 3.2 3.3 >99.0 >500
5 1 [emim]BF4 0.128 24 15.3 18.0 >99.0 >500
6 1 [bmim]BF4 0.123 24 14.8 17.4 >99.0 >500
7 1 [hmim]BF4 0.058 24 7.0 7.6 >99.0 >500
8 1 [omim]BF4 0.102 24 12.3 14.0 >99.0 >500
9 1 [bmim]PF6 0.077 24 9.2 10.2 >99.0 >500
10 1 [hmim]PF6 0.122 24 14.6 23.2 >99.0 >500
11 1 [omim]PF6 0.050 24 6.0 6.4 >99.0 >500
12 1 [bmp]BF4 0.045 24 5.4 5.7 >99.0 >500
13 2 [emim]tfms 0.68 5 16.6 7.5 37.9 2.4
14 2 [bdmim]tfms 0.58 10 21.7 7.4 26.7 1.9
15 2 [bdmim]BF4 0.16 24 19.6 8.4 34.4 2.2
16 2 [hdmim]BF4 0.08 24 9.2 3.8 37.3 2.3
17 2 [emim]BF4 0.67 10 18.4 11.8 52.3 3.6
18 2 [bmim]BF4 0.59 5 14.8 8.3 47.9 3.1
19 2 [hmim]BF4 0.16 24 18.9 9.2 39.4 2.5
20 2 [omim]BF4 0.40 24 27.9 12.3 31.7 2.2
21 2 [bmim]PF6 0.76 5 21.6 14.4 52.1 3.6
22 2 [hmim]PF6 0.40 24 25.7 13.1 37.8 2.5
23 2 [omim]PF6 0.08 24 9.5 5.3 50.1 3.2
24 2 [bmp]BF4 0.33 24 23.3 9.2 30.4 2.0
a Amount of amine (mol) esterified per hour and milligram of immobilized enzyme.
b Enantiomeric excess of unreacted amine.
c Enantiomeric excess of resulting amide.
Table 2
Solvent properties of ionic liquids and some non-ionic solvents determined by solvatochromic compounds and their water solubility
Entry Solvent Reichardt’s Nile red ENR b λmax c (nm) Water Ref.
dye ETN a (kcal/mol) miscibilityd
1 [emim]tfms M
2 [bmim]tfms 0.67 601.5 M [14]
3 [bdmim]tfms I
4 [bdmim]BF4 0.58 I [14]
5 [hdmim]BF4 I
6 [odmim]BF4 0.54 [16]
7 [emim]BF4 0.71 M [15]
8 [bmim]BF4 0.67 51.9 M [14]
9 [hmim]BF4 51.8 I [14]
10 [omim]BF4 0.54 52.0 I [14]
11 [bmim]PF6 0.67 52.2 516.5 I [14]
12 [hmim]PF6 51.8 I [14]
13 [omim]PF6 0.63 52.2 516.5 I [14]
14 [bmp]BF4 0.63 [15]
15 Water 1.00 48.2 [14]
16 Methanol 0.77 52.0 [14]
17 Ethanol 0.65 52.2 585 [14,16]
18 2-Propanol 0.55 591 [16]
19 Hexane 59.0 [14]
20 N,N-Dimethylformamide 0.39 602 [16]
21 1,2-Dichloroethane 0.33 500 [16]
22 Tetramethylsilane 0.00 [15]
a
Normalized polarity values.
b
Electronic transition energy.
c Wave length value for the lowest energy d–d absorption band at solvation of acetylacetonatotetramethylethyldiaminecopper(II) tetraphenylborate:
[Cu(acac)(tmen)[BPh4 ].
d M: miscible, I: immiscible.
R. Irimescu, K. Kato / Journal of Molecular Catalysis B: Enzymatic 30 (2004) 189–194 193
and also its value for [bmim]tfms (Table 2, entry 2) which is the (R)-enantiomer acylated. The value for enantioselectivity
the same with that for [bmim]BF4 (Table 2, entry 8), it might in ionic liquids was actually higher than the previously re-
be concluded that [bdmim]tfms has probably a hydrogen ported results on CALB catalyzed acylation of 1 with simple
bond acidity between ethanol and 2-propanol. Regarding esters or carbonates in classical organic solvents [3]. The an-
its nucleophilicity, [bdmim]tfms is probably very close to alytical scale kinetic resolution of amine 1 produced amide
N,N-dimethylformamide (Table 2, entry 20). This estimation 4 (>99% purity) with 81% yield after purification in 72 h.
was based on the value for [bmim]tfms (Table 2, entry 2) For the acylation of amine 2, [bmim]PF6 was the best
and the fact that, as mentioned above, the nucleophilicity is reaction media (Table 1, entry 21), but [emim]tfms and
dependent almost exclusively on the anion and that a methyl [bmim]BF4 afforded comparable reaction rates also (Table 1,
group instead of hydrogen on the 2-position of imidazolium entries 13, 17). [bdmim]tfms gave a slightly lower rate than
[14] produced only a very slight increase of nucleophilicity. [emim]tfms. The same minimum of the reaction rate for
[Bdmim]tfms is immiscible with water and therefore, hy- [hmim]BF4 (Table 1, entry 19) in this series (1-alkyl chain
drophobic. A quite close value for the reaction rate was ob- varied from ethyl to octyl) was observed. These results sug-
tained in [emim]tfms (Table 1, entry 1). Applying the same gest that in the acylation of amine 2, the difference in nucle-
rationale at as above, its nucleophilicity is almost the same as ophilicity of the ionic liquids affects very little the reaction
for [bmim]tfms and very similar to N,N-dimethylformamide rates and that the hydrogen bond acidity of the solvent corre-
(Table 2, entry 2). Its hydrogen bond activity places it lates with the reaction rates. The enantioselectivity for acy-
probably between methanol and ethanol, as its ETN value is lation of amine 2 was very low and did not vary much with
expected to be lower than for [bmim]tfms. [Emim]tfms is the nature of the ionic liquid employed as reaction media.
water miscible and therefore hydrophilic (Table 2, entry 1). The water miscibility of the ionic liquids does not ap-
The change of the anion countering [bdmim]+ from pear to influence the reaction rates for the acylation of both
[tfms]− to [BF4 ]− produced a decrease in the reaction amines.
rate (Table 1, entry 3). The decrease was very sharp when The two amines used in this study have similar basicity
[hdmim]BF4 was employed. (1 slightly less polar than 2), but the amine 1 has the chi-
The ionic liquids based on 1-alkyl-3-methylimidazolium ral center at the a carbon atom (directly bound to the amine
cation combined with [BF4 ]− anion gave better rates group) while amine 2 was a more remote chiral center at
(Table 1, entries 5, 6, 8) than those based on the same the  carbon atom. This structural difference accounts for
cation and [PF6 ]− (Table 1, entries 9, 11). The decrease in the outcome of their enzymatic acylation. Due to the steric
the reaction rate seemed to be correlated with the increase hindrance around the amine group of 1 that acts as the nu-
in the length of the alkyl chain on the 1-position of the cleophile attacking the acyl enzyme intermediate to form the
cation. The [hmim]+ cation combined with either [BF4 ]− amide, only the (R)-enantiomer can access the active site of
or [PF6 ]− produced strikingly different results than the the lipase. Moreover, a very strict orientation is necessary
next higher and lower chain alkyl cation, i.e. [bmim]+ and by entering the active site of the lipase. This step determines
[omim]+ (Table 1, entries 7, 10). A similar behavior of the high enantioselectivity of CALB for 1 and limits the rate
these two liquids was observed in the polarity studies for of amide formation. The equilibrium of the whole process
the same series using Nile red (Table 2, entries 9, 12). The would be driven towards amide formation by a higher con-
dipolarity/polarizability of both kinds of ionic liquids rises centration of amine 1 in the unionized form in the reaction
upon changing from butyl to hexyl and then falls for octyl system. This rationale was proved in the process of optimiz-
chain [17]. The same study mentions that the melting point ing the molar ratio of the reactants: an excess of amine 1
of these two series of ionic liquids (1-alkyl chain varied produced higher reaction rate. The ionic liquid used as re-
from ethyl to octadecyl) reaches a minimum for hexyl chain action medium has a crucial effect on the reaction process.
also [17]. The nucleophilicity or donor strength of [PF6 ]− It provides a better solvation of the ionic reactants (when
based ionic liquids (Table 2, entries 11, 13) is very low and put in contact, the amine and acid spontaneously form a salt
very close to that of 1,2-dichloroethane (Table 2, entry 21) that has a low solubility in organic solvents and also is re-
which is non-polar and has a donor number equal to zero sponsible for the very high viscosity of non-solvent system
[18]. No nucleophilicity data are available for the [BF4 ]− [8]). The best reaction medium for the acylation of 1, [bd-
based ionic liquids, but a study on the donor numbers of mim]tfms, solves the substrates and does not further ionize
anions in 1,2-dichloroethane using [Cu(acac)(tmen)]+ es- the amine as it would be expected in the case of ionic liq-
tablished the donor number 6.03 for [BF4 ]− and 16.9 for uids based on 1-alkyl-3-methylimidazolium cation that have
[tfms]− [18]. This indicates that the nucleophilicity order is a high hydrogen bond donor ability at the 2-position of the
probably: [PF6 ]− < [BF4 ]− < [tfms]− . The reaction rates imidazolium cation [16,19]. The concentration of the union-
for the acylation of amine 1 follow the same order and ized amine in the reaction is therefore higher and conse-
therefore a correlation with the nucleophilicity of the ionic quently, raising the reaction rate. There is no doubt that the
liquid might be implied. ionic liquid properties affect the ionization state of the en-
The enantioselectivity of the lipase in acylation of amine 1 zyme too with further effects on the enantioselectivity and
was almost perfect in all ionic liquids investigated with only reaction rate.
194 R. Irimescu, K. Kato / Journal of Molecular Catalysis B: Enzymatic 30 (2004) 189–194
In contrast, amine 2 can access easily the active site of the Acknowledgements
enzyme due to its more remote stereogenic center and less
steric hindrance around the amine group. As a result, the acy- A postdoctoral fellowship from the Japanese Society for
lation reaction is faster than for amine 1, but the enantiose- Promotion of Science to Roxana Irimescu is gratefully ac-
lectivity of the enzyme is lower. For CALB-catalyzed acyla- knowledged.
tion of amine 2, the rate-limiting step of the process seems to
be the formation of the acyl-enzyme intermediate. A higher
concentration of free acid in the reaction media would favor References
the formation of the acyl-enzyme intermediate and speed up
the whole process. To support this assumption, the optimum [1] U.T. Bornscheuer, J.T. Kazlauskas, Hydrolases in Organic Synthesis,
reactant molar ratio found for acylation of 2 has the acid Wiley-VCH, Weinheim, 1999, Chapter 5, p. 103; Chapter 6, p. 151.
in excess. An ionic liquid with high hydrogen bond acidity [2] V. Gotor, Bioorg. Med. Chem. 7 (1999) 2189.
would modify the ionization state of the acid by substituting [3] F. Van Rantwijk, M.A.P.J. Hacking, R.A. Sheldon, Monatsh. Chem.
it in the formation of the salt with the amine, and therefore 131 (2000) 549.
[4] R. Madsen, C. Roberts, B. Fraser-Reid, J. Org. Chem. 60 (1995)
providing a higher concentration of acyl donor available to 7920.
lipase in the system and raising the reaction rate as a result. [5] S. Takayama, W.J. Moree, C.-H. Wong, Tetrahedron Lett. 37 (1996)
The results presented in this study demonstrated that 6287.
finding the best reaction media for a specific enzymatic re- [6] S. Takayama, S.T. Lee, S.-C. Hung, C.-H. Wong, Chem. Commun.
action is quite a tedious task and that it can not be assumed (1999) 127.
[7] R. Irimescu, T. Saito, K. Kato, J. Mol. Catal. B: Enzym. 27 (2004)
that the most performing reaction media for one substrate 69.
would perform the same even for a very similar compound. [8] R. Irimescu, K. Kato, Tetrahedron Lett. 45 (2004) 523.
The interactions involved in ionic liquid systems are very [9] U. Kragl, M. Eckstein, N. Kaftzik, Curr. Opin. Biotechnol. 13 (2002)
complex and beside the effects on the reaction produced by 565.
solvent–solute interactions, the microenvironment of the en- [10] F. Van Rantwijk, R.M. Lau, R.A. Sheldon, Trends Biotechnol. 21
(2003) 131.
zyme is also affected. Although there is a wide range of ionic [11] S. Park, R.J. Kazlauskas, Curr. Opin. Biotechnol. 14 (2003)
liquids readily available currently, the data on the solvent 432.
properties of the ionic liquids is quite limited and completely [12] T. Itoh, E. Akasaki, Y. Nishimura, Chem. Lett. 31 (2002) 154.
insufficient to allow a reliable prediction for selection of [13] J.L.L. Rackels, A.J.J. Straathof, J.J. Heijnen, Enzym. Microb. Tech-
the ‘best’ reaction media. However, ionic liquids have great nol. 15 (1993) 1051.
[14] C.F. Poole, J. Chromatogr. A 1037 (2004) 49.
potential for biotransformations of highly polar substrates [15] S. Park, R.J. Kazlauskas, J. Org. Com. 66 (2001) 8395.
such as amines and carboxylic acids in our case. Although [16] M.J. Muldoon, C.M. Gordon, I.R. Dunkin, J. Chem. Soc., Perkin
they are very polar, the ionic liquids do no inactivate the Trans. 2 (2001) 433.
enzymes as would organic solvents of similar polarity or nu- [17] A.J. Carmichael, K.R. Seddon, J. Phys. Org. Chem. 13 (2000)
cleophilicity (lower alcohols and N,N-dimethylformamide 591.
[18] W. Linert, R.F. Jameston, A. Taha, J. Chem. Soc., Dalton Trans.
for our system). Unlike organic solvents specified above, (1993) 3181.
ionic liquids systems can be run under low pressure to [19] T. Itoh, Y. Nishimura, N. Ouchi, S. Hayase, J. Mol. Catal. B: Enzym.
remove volatile products such as water here, driving the 26 (2003) 41.
reaction equilibrium towards product formation.