1134
D. Dennewald et al. / Process Biochemistry 46 (2011) 1132–1137
Table 1
mixed to the ionic liquid containing the substrate. The cosubstrate (sodium for-
mate) necessary for the cofactor regeneration was contained in the aqueous phase
and was thus supplied simultaneously with the latter. Samples were taken before
starting the reaction, during the biotransformation, as well as before and after the
distillation.
Initial substrate concentrations in the ionic liquid resulting from the variation of the
phase ratio while keeping the substrate quantity constant.
Ionic liquid
volume fraction
Resulting substrate
concentration, mM
10%
20%
30%
40%
50%
60%
70%
1200
600
400
300
240
200
171
2.6. Purity of the product recovered at each cycle
The purity of the product recovered at the end of each cycle was determined
by gas chromatography. The values indicated were calculated by comparison of the
concentration of (R)-2-octanol in the distillate to the concentration of (R)-2-octanol
determined in purchased samples of known analytical degree.
2.7. Analytics
To determine the amounts of substrate and product(s) in the aqueous phase
and in the ionic liquid phase, the samples taken during the biotransformation were
extracted in one step with ethyl acetate (volume ratio 1:1) and hexane (volume
ratio 1:4), respectively. The concentrations of 2-octanone, (R)-2-octanol and (S)-2-
octanol in the extract were then determined by chiral gas chromatography (CP-3800,
Varian, Palo Alto, USA), using a flame ionization detector (FID) and helium as carrier
gas. The separation and quantification of the different compounds were performed
using a BGB-175 column (BGB Analytik, Schlossboeckelheim, Germany), a flow rate
of 2.5 mL min−1 helium, and a temperature of 50 ◦C. Typical retention times were
31.1 min for 2-octanone, 34.0 min for (S)-2-octanol and 35.3 min for (R)-2-octanol.
Samples of the ionic liquid phase were analysed by 1H NMR to detect compounds
resulting from the possible degradation of the cation due to the repeated use of the
solvent. Accumulation of potentially harmful substances in the ionic liquid phase
through contamination during the recycling procedure or through degradation of
either the anion or the cation was also verified by ion chromatography. These anal-
yses were carried out on a Metrohm system (Deutsche METROHM GmbH & Co. KG,
Filderstadt, Germany), equipped with a model 820 IC separation center, a 819 IC
detector, and a Metrosepp A Supp 5 column. The eluent was composed of an aque-
ous mixture containing 3.2 × 10−3 mol L−1 Na2CO3, 1.0 × 10−3 mol L−1 NaHCO3 and
5% acetonitrile.
initial substrate concentrations in the aqueous phases of the vari-
ous systems (Table 1), and consequently, the initial reaction rates
conversions reached in the different setups.
As mentioned above, the initial reaction rates observed were
indeed larger for systems with lower ionic liquid volume frac-
tions, due to the larger substrate concentrations found within them
(Fig. 2). However, increased concentrations of substrate – and later
also of product – are also detrimental for the process, as both sub-
stances show severe toxicity towards the biocatalyst [20–22]. In
fact, as was observed for the reaction system containing only 10%
[HMPL][NTF], reaction systems with low ionic liquid fractions –
and thus large substrate and product concentrations in the aqueous
phase (up to ∼1 mM) – reach only relatively low final conversions.
Systems with larger ionic liquid volume fractions show lower
initial reaction rates, because the initial substrate concentrations
in the aqueous phase are lower. However, they also reach larger
final conversions, because the exposure of the biocatalyst to toxic
substrate and product is reduced. This was observed for reaction
systems with ionic liquid volume fractions of 20–40%, with final
conversions of 95–98% (Fig. 2). These three systems reach very simi-
lar final conversions, indicating that a further increase of the phase
ratio above 20% does not lead to a considerable improvement in
terms of productivity.
On the contrary, when a given phase ratio is exceeded, a second
effect is observed due to the interaction between the ionic liquid
the advantage created by decreased substrate and product concen-
trations in systems with larger ionic liquid volume fractions. Even
if some ionic liquids show better biocompatibility than commonly
used organic solvents [12], their large presence does influence the
biocatalyst and decrease its activity. Here, this limit is reached at
an ionic liquid volume fraction between 40% and 50%. Above this
value, the final conversion decreases with increasing phase ratio.
The enantiomeric excesses observed did not differ significantly
in the different reaction setups. They were ≥99.5% (R) over the
whole duration of the reaction for each biotransformation per-
formed.
3. Results and discussion
Guidelines for selecting ionic liquids that are suitable for
applications in whole-cell biotransformations in biphasic ionic liq-
uid/water systems have been presented previously [9,19]. The
methodology is based on physicochemical properties of the ionic
liquids, such as melting point (≤30 ◦C) or density (≥1.2 g cm−3),
but it also includes biochemical and industrial rating criteria, e.g.
biocompatibility, chemical yield, toxicity, stability and water sol-
ubility of the ionic liquid, and costs. On the basis of these criteria,
([HMPL][NTF]) was chosen as the second phase for the production
tems [17]. Using this solvent, good conversions of 95% were reached
after 6 h. However, the remaining substrate present after the bio-
transformation constitutes a contamination of the end product
(Fig. 1) and it should thus be reduced to a minimum to avoid costly
product purification steps after the biotransformation. To this end,
a parameter study varying the ionic liquid volume fraction and the
initial substrate concentration was performed to further increase
the conversion up to values >99%.
3.2. Varying initial substrate concentrations
3.1. Varying ionic liquid volume fractions
Another important parameter in the process design is the initial
substrate concentration. The initial substrate concentrations in the
ionic liquid phase were varied in the range of 150–1200 mM 2-
octanone. This corresponds to a range of initially 0.12–0.97 mM 2-
octanone in the aqueous phase after equilibration with the loaded
As expected, reaction systems initially containing larger con-
centrations of substrate in the ionic liquid phase show larger initial
reaction rates than those with lower initial substrate concentra-
tions (Fig. 3). Due to the constant equilibrium between both phases,
larger substrate concentrations in the ionic liquid phase also mean
Different phase ratios ranging from 10% to 70% ionic liquid were
analysed, while keeping the initial substrate quantity added into
the ionic liquid phase, as well as the biocatalyst quantity in the
aqueous phase constant. The biocatalyst and substrate quantity
added to each system, regardless of the ionic liquid volume frac-
tion used, were 8 gCDW and 3.83 mL 2-octanone. This corresponds
to a biocatalyst concentration of 50 gCDW L−1 in the aqueous phase
and a substrate concentration of 600 mM in the ionic liquid phase
for a reaction system with an ionic liquid volume fraction of 20%.
Due to the defined partitioning of the substrate between the ionic
liquid and the aqueous phase, this procedure results in different