Productive asymmetric styrene epoxidation based on a next generation electroenzymatic methodology

Article abstract of DOI:10.1002/adsc.200900291

We have established a novel and scalable methodology for the productive coupling of redox enzymes to reductive electrochemical cofactor regeneration relying on efficient mass transfer of the cofactor to the electron-delivering cathode. Proof of concept is

Full text of DOI:10.1002/adsc.200900291

FULL PAPERS
DOI: 10.1002/adsc.200900291
Productive Asymmetric Styrene Epoxidation Based on a Next
Generation Electroenzymatic Methodology
Reto Ruinatscha,a Christian Dusny,a Katja Buehler,a and Andreas Schmida,b,*
a
Laboratory of Chemical Biotechnology, Department of Biochemical and Chemical Engineering, TU Dortmund
University, Emil-Figge-Strasse 66, 44227 Dortmund, Germany
Fax : (+49)-231-755-7382; phone: (+49)-231-755-7380; e-mail: andreas.schmid@bci.uni-dortmund.de
ISAS – Institute for Analytical Sciences, Bunsen-Kirchhoff-Strasse 11, 44139 Dortmund, Germany
b
Received: April 24, 2009; Revised: July 27, 2009; Published online: September 23, 2009
Abstract: We have established a novel and scalable 150 mLminꢀ1, an average space-time yield of
methodology for the productive coupling of redox 0.35 gLꢀ1 hꢀ1 could be achieved during 2 h with a
enzymes to reductive electrochemical cofactor regen- final (S)-styrene oxide yield of 75.2%. At two-fold
eration relying on efficient mass transfer of the co- lower aeration rates, the electroenzymatic reaction
factor to the electron-delivering cathode. Proof of could be sustained for 12 h, albeit at the expense of
concept is provided by styrene monooxygenase lower (59%) overall space-time yields. Under these
(StyA) catalyzing the asymmetric (S)-epoxidation of conditions, as much as 20.5% of the utilized current
styrene with high enantiomeric excess, space-time could be channeled into (S)-styrene oxide formation.
yields, and current efficiencies. Highly porous reticu- In comparison with state-of-the-art electroenzymatic
lated vitreous carbon electrodes, maximized in volu- methodologies for the same conversion, (S)-styrene
metric surface area, were employed in a flow- oxide synthesis could be improved up to 150-fold
through mode to rapidly regenerate the consumed with respect to both reaction time and space-time
FADH2 cofactor required for StyA activity. A sys- yield. These productivities constitute the most effi-
tematic investigation of the parameters determining cient reaction reported for asymmetric in vitro epoxi-
cofactor mass transfer revealed that low FAD con- dations of styrene.
centrations and high flow rates enabled the continu-
ous synthesis of the product (S)-styrene oxide at high
rates, while at the same time the accumulation of the Keywords: asymmetric synthesis; cofactor regenera-
side-products acetophenone and phenylacetaldehyde tion; electrochemistry; enzyme catalysis; styrene
was minimized. At 10 mm FAD and a flow rate of monooxygenase; three-dimensional electrodes
Introduction
for enantioselective epoxidations based on iron cata-
lysts and hydrogen peroxide as oxidant.[6] In addition,
Asymmetric epoxidations are key transformations en chiral ketones instead of transition metals are increas-
route to biologically active pharmaceuticals and fine ingly recognized as valuable organocatalysts for asym-
chemicals. The introduction of two C O bonds in one metric epoxidations,[7] as well as the use of molecular
ꢀ
reaction not only leads to the formation of up to two oxygen instead of hazardous, waste producing oxi-
chiral centers, but also provides access to a diverse dants.[8] Although many of these future-oriented
array of key intermediates due to the possibility of methodologies show already promising enantioselec-
facile opening of the epoxide ring.[1] Much research tivities, low catalyst efficiencies and low productivities
effort has thus been dedicated to the development of often restrict their practical value.[5,7]
efficient asymmetric epoxidations, and many practical
A potential solution for these challenges is the uti-
and proficient methodologies based on transition lization of enzymatic epoxidation catalysts, that is,
metals have been reported[2] since the breakthroughs monooxygenases. Monooxygenases frequently exhibit
of Katsuki and Sharpless in the 1980s with titanium high conversion rates at excellent enantioselectivities,
tartrate catalysts.[3] Current interests towards more are active at ambient reaction conditions, use molecu-
economical and sustainable synthetic methodologies lar oxygen as oxidant, and are produced from natural,
led to the development of less expensive, atom effi- renewable raw materials.[9] In the case of flavin-de-
cient and less toxic catalytic systems.[4,5] Such a system pendent monooxygenases, the selective incorporation
has recently been reported by Beller and colleagues of one oxygen atom into the substrates, and the re-
Adv. Synth. Catal. 2009, 351, 2505 – 2515
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2505

Reto Ruinatscha et al.
FULL PAPERS
duction of the other oxygen atom to water, is accom- ric (S)-epoxidation of styrene with respect to produc-
plished by peroxo-FAD, generated from reductive ac- tivities and current efficiencies.
tivation of molecular oxygen with FADH2. As this co-
factor is consumed during catalysis, it needs to be
continuously supplied for the epoxidation reaction. Results
Since FADH2 is prone to rapid re-oxidation under
aerobic conditions its in situ regeneration is essential We aimed at coupling the enzymatic styrene epoxida-
to sustain the catalytic cycle. The most efficient enzy- tion reaction (R1) to the electrochemical regeneration
matic methodologies in terms of productivities (R2) of the necessary cofactor FADH2.
employ living microbial cells as biocatalysts, perform-
ing intracellular FADH2 regeneration at the expense
of energy sources such as glucose,[10,11] or isolated re-
generation enzymes utilizing chemical sources of re-
duction equivalents such as formate.[12] Alternatively,
electrical power, the cheapest source of reduction
equivalents for redox enzymes,[13] can be used for the
reagent-free regeneration of FADH2 at cathodes in
Space-time yields of the StyA-catalyzed styrene
cell-free epoxidations, omitting the cellular cofactor epoxidation reaction in the electrochemical flow-
regeneration machinery and additional regeneration through reactor were maximized with respect to
enzymes from the reaction.[14] In general, such electro- FADH2 regeneration rates (STYFADH ) to meet the re-
enzymatic approaches might not only be more practi- quirements of the flavin-dependent2 monooxygenase,
cal because of their simplicity, but they would also considering:[20]
combine two environmentally friendly methods for
the selective synthesis of chiral synthons.[15] Despite
the potential advantages of electroenzymology, there
have been some drawbacks limiting the implementa-
tion of this methodology in synthetic applications.
Most of the so far developed electroenzymatic pro-
Here AV denotes the cathode surface to volume
cesses display insufficient reaction stabilities and too ratio, kM the mass transfer coefficient, and cFAD the
low electrochemical cofactor regeneration rates with FAD concentration. The electrochemical cell was
respect to the redox enzyme used as catalyst.[16]
equipped with highly porous RVC electrodes having a
The driving force for high electrochemical cofactor large surface to volume ratio (AV =19685 m2 mꢀ3), cor-
regeneration rates is the mass transfer from the bulk responding to nearly 1 m2 of surface area at a cathode
solution to the depleted region near the electrode. As volume of 50 cm3 only, while at the same time provid-
redox enzymes typically require very low cofactor ing a high reaction volume of 45.5 cm3. This volumet-
concentrations in the micromolar to millimolar range, ric surface area not only enabled high regeneration
high cofactor amounts adjacent to electrode surfaces rates of consumed FADH2 (E1), but resulted also in
can only be sustained by increasing both cofactor an optimal mass transfer due to its high porosity, serv-
transport rates from the bulk solution to the near- ing as turbulence promoter,[21] thus shortening diffu-
electrode volume and high volumetric electrode sur- sion distances in dependence of applied flow rates.[17]
face areas for increasing the volume of the near-elec-
Based on these reactor characteristics, we systemat-
trode layer. By taking these considerations into ac- ically investigated the influence of flow rates (up to
count, we employed microporous, three-dimensional 50 mLminꢀ1) and FAD concentrations (up to 200 mM)
reticulated vitreous carbon (RVC) electrodes for im- at an aeration of 45 mLminꢀ1 (corresponding to
proved mass transfer in a flow-through plate and 22.8 mmol molecular oxygen hꢀ1) on the average elec-
frame electrochemical filter press cell, enabling high troenzymatic space-time yield during 105 min. StyA
FADH2 regeneration rates.[17] This system was evalu- coupled to electrochemical regeneration of FADH2
ated using the FADH2 dependent two-component sty- catalyzed the formation of (S)-styrene oxide from sty-
rene monooxygenase StyAB from Pseudomonas sp. rene with an enantiomeric excess of 99.5%.
VLB120 that catalyzes (S)-selective epoxidations of a
Average (S)-styrene oxide space-time yields at a
broad range of styrene derivatives at high rates, under flow rate of 10 mLminꢀ1 moderately increased with
ambient reaction conditions with molecular oxygen as FAD concentrations to 0.26 mMhꢀ1. Interestingly, the
oxidant.[12,18,19] Coupling this enzyme system to the electroenzymatic reaction also led to the formation of
electrochemical regeneration allows us to omit the re- acetophenone and phenylacetaldehyde. The highest
ductase component StyB from the reaction, thus sim- accumulation
rate
of
these
side-products
plifying the set-up.[14] We systematically evaluated this (0.09 mMhꢀ1) was observed at 200 mM FAD
new electroenzymatic methodology for the asymmet- (Figure 1A). Up to three-fold higher space-time
2506
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2505 – 2515

Productive Asymmetric Styrene Epoxidation
FULL PAPERS
Figure 1. Electroenzymatic reaction performances as a function of FAD concentration and flow rate during 105 min. (A–C)
&
*
Average space-time yields (STY) for (S)-styrene oxide ( ), acetophenone and phenylacetaldehyde ( ), and total products
~
formed ( ). (D) Corresponding (S)-styrene oxide yields over the totally formed products. (E) Average current efficiencies
for (S)-styrene oxide (shaded bars) and the side-products (unshaded bars).
yields regarding overall product formation were ob- 82.2% (S)-styrene oxide yields were achieved at
served at a flow rate of 30 mLminꢀ1. However, under 10 mM FAD and 10 mLminꢀ1 (Figure 1D). Highest
this condition average (S)-styrene oxide space-time current efficiencies (100% efficiency corresponds to
yields were maximal at already 50 mM FAD the utilization of 2 electrons per molecule of product,
(0.49 mMhꢀ1), while acetophenone and phenylacetal- based on the reaction stoichiometry) were generally
dehyde space-time yields still increased in conjunction observed between 50 mM and 100 mM FAD. Average
with the FAD concentrations, finally accounting for current efficiencies with respect to all products
70.6% of the totally formed products at 200 mM FAD formed were mainly determined by the flow rates,
(Figure 1B). Increasing the flow rate to 50 mLminꢀ1 and were in the range of 5.0%ꢁ0.5% at 10 mLminꢀ1,
not only further enhanced overall electroenzymatic 7.5%ꢁ0.8% at 30 mLminꢀ1, and 9.4%ꢁ1.0% at
space-time yields, but also led to a more distinctive 50 mLminꢀ1. In contrast, average (S)-styrene oxide
(S)-styrene oxide space-time yield maximum at 50 mM current efficiencies were strongly affected by FAD
FAD (0.61 mMhꢀ1) (Figure 1C).
concentrations and flow rates above 10 mLminꢀ1.
Overall, side-product formation rates were strongly Highest current utilization (7.5%) for the epoxidation
dependent on both increasing FAD concentrations reaction was observed at 50 mLminꢀ1 and 50 mM
and flow rates above 10 mLminꢀ1. Accordingly, up to FAD (Figure 1E).
Adv. Synth. Catal. 2009, 351, 2505 – 2515
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2507

Reto Ruinatscha et al.
FULL PAPERS
Figure 2. (A) Average electroenzymatic space-time yields (STY) at flow rates from 10 mLminꢀ1 to 200 mLminꢀ1 (10 mM
FAD) during 105 min. (B) Corresponding (S)-styrene oxide yields over the totally formed products. (C) Average current effi-
ciencies for (S)-styrene oxide, the side-products acetophenone and phenylacetaldehyde, and the totally formed products.
The performance of the electroenzymatic system avoidance or with a mixture of antifoam 204, BSA
was investigated in more detail at 10 mM FAD since and sucrose resulted in a constant (S)-styrene oxide
highest (S)-styrene oxide yields were achieved at this formation rate over at least 120 min (10 mM FAD,
cofactor concentration. Increasing the flow rates to 150 mLminꢀ1) and a significantly lowered enzyme
200 mLminꢀ1 had a significant impact on space-time precipitation (Figure 3A). In comparison to the non-
yields, (S)-styrene oxide yields, and current efficien- stabilized reaction system, addition of antifoam 204
cies. The maximum (S)-styrene oxide yields were de- resulted in 34% lower initial (S)-styrene oxide forma-
termined at 50 mLminꢀ1 (82.2%) (Figure 2B). The tion rates, whereas additional supplementation with
overall productivity was highest at 150 mLminꢀ1 BSA and sucrose could sustain the initial styrene ep-
(1.7 mMhꢀ1) (Figure 2A), indicating that the enzyme oxidation rate (Figure 3B). Current efficiencies with
stability was significantly impaired at elevated flow respect to (S)-styrene oxide were around 12% for the
rates. Average current efficiencies reached a maxi- non-stabilized and the antifoam 204 supplemented
mum at 150 mLminꢀ1 of 13.3% for (S)-styrene oxide electroenzymatic reaction system. A 40% higher cur-
and 22.4% with respect to the totally formed products rent efficiency was determined for the reaction
(Figure 2C). Apparently the transfer efficiency of system containing all three stabilizers. Apparently the
electrochemically reduced FAD to StyA could not efficient utilization of electrochemically generated
further be improved above flow rates of FADH2 correlates with the amount of active StyA.
150 mLminꢀ1.
The electroenzymatic reaction system stabilized
During all electroenzymatic synthesis, white foam with antifoam 204, BSA, and sucrose was evaluated
accumulated at the aqueous/organic interphase of the for maximal productivities at 10 mM FAD and
cathodic styrene feed/product extraction reservoir in 150 mLminꢀ1 regarding synthesis times and aeration
the course of aeration. According to SDS-PAGE anal- rates. At the initially applied aeration with
ysis (Figure 3A), this precipitate accumulation could 22.8 mmolhꢀ1 molecular oxygen, up to 6.9 mM (S)-
be attributed to StyA and catalase denaturation in the styrene oxide could be synthesized within 12 h, which
catholyte, resulting in nearly complete termination of was equivalent to an average space-time yield of
product formation after 120 min. Supplementation of 0.1 gLꢀ1 hꢀ1 (Synthesis 1, Table 1). Under these condi-
the catholyte with either antifoam 204 for foam tions, the average current efficiency was 20.5%. Con-
2508
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2505 – 2515

Productive Asymmetric Styrene Epoxidation
FULL PAPERS
Figure 3. (A) Influence of stabilizing agents in the cathodic storage reservoir on StyA and catalase precipitation at the aque-
ous/organic interphase, visualized by SDS-PAGE. (B) Influence of the stabilizing agents on (S)-styrene oxide formation
rates. Experimental conditions: 10 mM FAD, 150 mLminꢀ1.
sidering the concurrent side-product formation, the
average overall electroenzymatic space-time yield was
0.15 gLꢀ1 hꢀ1 with a current utilization efficiency of
30.7%. Instead of forming white precipitates at the
aqueous/organic interphase of the cathodic reservoir,
the stabilized reaction mixture became slightly turbid
over time. This observation was much more pro-
nounced at the higher aeration rate, pointing to
enzyme denaturation by shear stress (Synthesis 2,
Table 1). Changing the aeration mode from a capillary
to an HPLC frit for finer dispersion and distribution
of the supplied air, while at the same time increasing
the molecular oxygen input to 46.8 mmolhꢀ1, more
than doubled average (S)-styrene oxide and overall
space-time yields to 0.26 gLꢀ1 hꢀ1 and 0.35 gLꢀ1 hꢀ1,
Figure 4. Electroenzymatic syntheses of (S)-styrene oxide at
a molecular oxygen input of 45 mLminꢀ1 (Synthesis 1) and
92 mLminꢀ1 (Synthesis 2), stabilized with a mixture of anti-
foam 204, BSA, and sucrose. Flow rate: 150 mLminꢀ1,
FAD : 10 mM.
respectively. However, product formation stopped
after 120 min, thereby leading to a final (S)-styrene
oxide concentration of 3.5 mM (Figure 4). Average
current efficiencies were almost four times lower than
at the lower aeration rate, presumably due to in-
creased cathodic reduction of molecular oxygen to
hydrogen peroxide. Electroenzymatic syntheses per-
formances are summarized in Table 1.
Table 1. Overall electroenzymatic synthesis performance at a molecular oxygen input of 22.8 mmolhꢀ1 (45 mLminꢀ1 air) and
46.8 mmolhꢀ1 (92 mLminꢀ1 air), stabilized with a mixture of antifoam 204, BSA, and sucrose. Flow rate: 150 mLminꢀ1,
FAD : 10 mM.
Synthesis 1
(S)-styrene oxide
Synthesis 2
(S)-styrene oxide
Parameter
Total[a]
Total[a]
STY,[b] mMhꢀ1 (gLꢀ1 hꢀ1)
0.8 (0.10)
20.5
6.9 (0.83)
45 (22.8)
12
1.2 (0.14)
30.7
10.5 (1.26)
2.2 (0.26)
5.7
3.5 (0.42)
92 (46.8)
2
2.9 (0.35)
7.4
4.6 (0.55)
Current efficiency,[b]
%
Final concentration, mM (gLꢀ1)
Aeration, mLminꢀ1 (mmol O2 hꢀ1)
Duration, h
[a]
Including acetophenone and phenylacetaldehyde.
Average value during synthesis.
[b]
Adv. Synth. Catal. 2009, 351, 2505 – 2515
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2509

Reto Ruinatscha et al.
FULL PAPERS
Discussion
As a result of the nearly diffusion controlled chemi-
cal uncoupling reactions, a reactive near-cathode
The simultaneous maximization of volumetric elec- layer consisting of FAD, FADH2, semiquinone radi-
trode surface areas (AV) and mass transfer coefficients cals, and superoxide formed during electrochemical
(kM) is a common practice in electrochemical engi- FAD reduction, while hydrogen peroxide accumula-
neering to optimize electrochemical process perform- tion was minimized due to the presence of catalase.
ances with respect to space-time yields (E1). Based FADH2 was therefore only available for StyA cataly-
on these considerations, we employed highly porous sis in close proximity to the cathode surfaces at a cer-
and volumetric surface area maximized RVC elec- tain steady state concentration, determining electro-
ACHTUNGTRENNUNGtrodes to accelerate volumetric FADH2 regeneration enzymatic space-time yields. Since the accumulation
rates, as they are stoichiometrically coupled to StyA of the side-products acetophenone and phenylacetal-
space-time yields (R2+R1). Since mass transfer can dehyde was dependent on StyA activity (Figure 1A–
be influenced by the flow rates through the porous C, Figure 2A), and hence on local (S)-styrene oxide
foam electrodes and the concentration of the cofactor, concentrations within the reactive near-cathode
the influence of both parameters on electroenzymatic region, it is very likely that the observed side-products
space-time yields was systematically evaluated.
originated from radicalic semiquinone attack of in situ
At a constant FAD concentration, space-time yields generated (S)-styrene oxide (Figure 5). Similar rear-
and current efficiencies steadily increased up to a rangements of epoxides to the corresponding carbonyl
flow rate of 150 mLminꢀ1. Above this flow rate elec- compounds in presence of radicals are described in
troenzymatic space-time yields decreased, indicating literature.[25]
StyA denaturation (Figure 2). We assume that the
Accordingly, we observed at improved FAD mass
porous structure of the electrode may cause high transfer conditions (raising flow rates, elevated FAD
shear stress. Moreover, the foam formation observed concentrations), and hence faster semiquinone forma-
in the cathodic storage reservoir independent of the tion rates, an increasing accumulation of these side-
flow rate points to additional shear stress due to the products at the expense of (S)-styrene oxide (Fig-
aeration of the system. The loss of active enzyme ure 1D), as well as lowered (S)-styrene oxide current
could be minimized by the addition of a mixture of efficiencies (Figure 1E). Based on these observations
antifoam 204, BSA, and sucrose, all well known pro- we assume that the theoretical electroenzymatic reac-
tein stabilizers.[12,22] Inactivation of StyA by hydrogen tion performance is reflected by the overall space-
peroxide, generated in the course of FADH2 uncou- time yields including side-products. Overall space-
pling (R3–R5), was counteracted using catalase. As time yields slightly decreased at FAD concentrations
FADH2 uncoupling reactions also generate superox- above 50 mM, suggesting that the steady state FADH2
ide, addition of superoxide dismutase could be benefi- concentration within the reactive near-cathode layer
cial to increase the stability of StyA, and hence the was also lowered. Apparently, under these conditions,
productivity of the electroenzymatic reaction.
reduced FAD was increasingly channelled into uncou-
Increasing FAD concentrations negatively influ- pling reactions (R3–R5) than the enzymatic product
enced (S)-styrene oxide space-time yields and (S)-sty-
rene oxide current efficiencies due to the enhanced
accumulation of the side-products acetophenone and
phenylacetaldehyde (Figure 1). These observations
may be attributed to the chemical behaviour of FAD
and FADH2 in solution. Whereas FADH2 is continu-
ously regenerated at the cathodes, the reduced
FADHꢀ anion, which is in equilibrium with FADH2
(pKa ~6.5)[23], and FAD symproportionate nearly dif-
fusion-controlled to semiquinone radicals (R3) that
are rapidly re-oxidized to FAD by one-electron ac-
ceptors such as molecular oxygen (R4). The transient-
ly formed superoxide finally leads to the accumula-
tion of hydrogen peroxide (R5):[24]
Figure 5. Proposed reaction scheme for the side-products
(acetophenone and phenylacetaldehyde) formed during
electroenzymatic styrene oxide synthesis.
2510
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2505 – 2515

Productive Asymmetric Styrene Epoxidation
FULL PAPERS
formation (R1). This FAD concentration-dependent need to be optimized in order to achieve electroenzy-
re-oxidation may also explain why elevated flow rates matic productivities close to the possible electrochem-
at constant FAD concentrations could steadily in- ical cofactor regeneration rate.
crease overall space-time yields (Figure 2A).
State-of-the-art electrochemical regeneration sys-
In a related study we have evaluated the influence tems for the reductive regeneration of cofactors com-
of initial FAD concentration on FAD reduction rates monly employ cylindrical electrodes in batch
for the electrochemical sub-reaction using the same cells.[16,26] Since mixing needs to be carried out by a
electrochemical cofactor regeneration module.[17] stirrer, these reactor systems are generally character-
Anaerobic conditions were applied to minimize un- ized by low electrode surface areas per reaction
coupling reactions of reduced FAD with molecular volume, as well as impaired mass transport from the
oxygen, and thus to determine the maximal FAD re- stirred bulk solution to distant electrode surfaces.
duction rates possible with this system. In contrast to Volumetric cofactor regeneration rates are thus very
the electroenzymatic reaction system described here, likely to be limiting the enzymatic synthesis reaction.
where FAD concentrations higher than 50 mM did not In contrast, the presented electrochemical reactor
further improve productivities, electrochemical reduc- design enables high volumetric regeneration rates of
tion rates steadily increased up to 360 mM FAD to FADH2 in close proximity to StyA due to the em-
93 mMhꢀ1. Comparison of the two reaction systems ployed three-dimensional microporous cathodes pro-
under similar reaction conditions (e.g., 50 mM FAD, viding high cofactor mass transfer. This is of utmost
30 mLminꢀ1 flow rate) indicates that electroenzymat- importance in the case of oxygen-dependent redox
ic (S)-styrene oxide space-time yields are by a factor enzymes. These enzymes are dependent on molecular
of 60 lower than the possible electrochemical FAD re- oxygen as a co-substrate, and their reduced cofactors
duction rate. Several aspects have to be taken into ac- (e.g., flavins) or mediators (commonly cobalt sepulch-
count when comparing this seemingly huge difference. rate, or methyl viologens)[26,27] are often capable of
Firstly, at 50 mM FAD and a flow rate of 30 mLminꢀ1, performing radicalic one-electron chemistry. Since
only 4% of the consumed current was channeled via molecular oxygen naturally exists as a diradical in its
FADH2 into (S)-styrene oxide formation. Since ground state, fast uncoupling reactions with reduced
FADH2 rapidly uncouples to the enzymatically inac- cofactors and/or mediators are always very likely to
tive FAD cofactor (R3–R5) prior to reaching StyA, it occur.[24,28,29] These uncoupling reactions not only
can be assumed that the net concentration of reduced impair the net cofactor regeneration rates, but they
FAD available for the enzymatic conversion was may also limit the concentration of enzymatically
much lower than the initial 50 mM. As electrochemical active cofactors. As a result, space-time yields of co-
reduction rates were shown to increase with the ap- factor mass transfer non-optimized systems are usual-
plied FAD concentration, electrochemical FAD re- ly rather low with respect to the maximal activity of
duction rates can only be compared with the electro- the production enzyme[16] because of fast competing
enzymatic productivities on the basis of identical ini- side-reactions. For example, cytochrome P450cam
tial and net cofactor concentrations, respectively. Sec- coupled to the reductive electrochemical regeneration
ondly, according to Michaelis–Menten, enzymatic pro- of putidaredoxin exhibited around 3% of the native
ductivities are a function of enzyme concentration. activity with respect to 5-exo-hydroxy camphor for-
Application of a higher StyA concentration might mation from camphor,[30] cytochrom P450 BM3 regen-
therefore be useful to channel more of the reduced erated by 1,10-dicarboxycobaltocene was able to hy-
FAD cofactor into the epoxidation reaction (R1) than droxylate lauric acid up to 3.6% of the theoretical
in non-productive uncoupling reactions (R3–R5), and rate,[29,31] and StyA coupled to the reductive electro-
to overcome enzymatic reaction rate limitations. For chemical regeneration of FADH2 exhibited only 0.6%
instance, in the previously mentioned study using the of the native (S)-epoxidation activity towards sty-
same
electrochemical
cofactor
regeneration rene.[14]
In comparison to the previously described electro-
module,[17] StyA was added to the reaction to demon-
strate electroenzymatic coupling. Application of a 2.5 enzymatic batch reaction system employing StyA,[14]
times higher StyA concentration than in the electro- we were able to exploit up to 8.7% of the native ini-
enzymatic reactions described here resulted in corre- tial enzyme activity (2.1 Umgꢀ1)[19] with respect to
spondingly higher (S)-styrene oxide space-time yields. (S)-styrene oxide space-time yields, and up to 11.5%
Furthermore, in case of the electroenzymatic process regarding overall space-time yields (Synthesis 2). Fur-
using StyA, productivities were strongly depending on thermore, the presented electroenzymatic approach
the molecular oxygen input. Increasing the aeration did not only permit higher enzyme activities than the
by a factor of two more than doubled (S)-styrene corresponding batch approach,[14] but also allowed for
oxide and overall space-time yields (Figure 4, longer synthesis times. Moreover, productivity in
Table 1), which points to substrate limitation by mo- terms specific enzyme activity and synthesis time was
lecular oxygen. Therefore, enzymatic sub-reactions 150-fold higher regarding (S)-styrene oxide, and 215-
Adv. Synth. Catal. 2009, 351, 2505 – 2515
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2511

Reto Ruinatscha et al.
FULL PAPERS
fold higher with respect to the overall electroenzy- Conclusions
matic performance. In comparisons of electroenzy-
matic and native activities it has to be noted that in The present study shows that isolated monooxygenas-
the natural system, StyA activities are dependent on es can efficiently and productively be coupled to re-
the FAD reductase component StyB,[19] which is not ductive electrochemical cofactor regeneration for the
required in electroenzymatic reactions using StyA. synthesis of high value products. With this engineer-
StyA and StyB are assumed to form transient, equi- ing approach it could be demonstrated that even a
molar complexes during catalysis,[32] possibly altering fast re-oxidizing FADH2 cofactor can be regenerated
catalytic properties of StyA, thereby making a direct at high overall rates, allowing for the productive cou-
comparison of native and electroenzymatic StyA ac- pling of styrene monooxygenase (StyA) for the syn-
tivities difficult. Surprisingly however, when compar- thesis of (S)-styrene oxide. Next to the achieved syn-
ing electroenzymatic StyA activities achieved in this thesis performances, which are already in the same
study with native StyA activity under process condi- range as for whole-cell methodologies, the employed
tions, the electroenzymatic approach was at least 40% electrochemical flow reactor design might also be
more productive with respect to the average specific suitable for scale-up, as has been demonstrated in in-
enzyme productivity.[12]
dustrial electro organic processes.[20] We therefore
Cellular and enzymatic cofactor regeneration meth- expect that further developments of electroenzymatic
odologies are generally stated to be the most efficient methodologies relying on mass transfer improved re-
and productive for the application of redox en- generation of diffusible cofactors/mediators may help
zymes.[33] However, whereas for whole cell styrene ep- establishing sustainable redox enzyme catalysis (e.g.,
oxidation an average specific (S)-styrene oxide pro- using flavin-fusion enzymes like cytochrome P450
duction rate of 2.6 g gStyAꢀ1 hꢀ1 during 10 h at an elec- BM-3 monooxygenases) beyond lab-scale.
tron efficiency of 10% is reported,[10,34] the here pre-
sented electroenzymatic methodology could channel
more than 20% of the consumed electrons into (S)- Experimental Section
styrene oxide for at least 12 h at an average specific
(S)-styrene oxide production rate of 0.5 g gStyAꢀ1 hꢀ1 Materials
(Table 2). This exceeds the corresponding in vitro en-
zymatic FADH2 regeneration approach, where an
Nitrogen 5.0 was supplied by Air-Liquide (Dꢁsseldorf, Ger-
many). Chemicals, catalase, and bovine serum albumin were
average specific styrene oxide production rate of
purchased from Sigma–Aldrich (Steinheim, Germany) in
0.35 g gStyAꢀ1 hꢀ1 (based on the epoxidation activity to-
the highest quality available and were used without further
wards 3-chlorostyrene) was reached during 10.5 h.[12]
purification. Buffers were prepared according to Beynon
and Easterby[36] using ultra-pure water from a Seralpur PRO
However, a drawback of the developed electroenzy-
90 CN system (Seral, Germany). Solvents were obtained
from Fischer Scientific GmbH (Schwerte, Germany). For
protein quantification according to Bradford,[37] the Quick
Start kit from Bio-Rad Laboratories GmbH (Munich, Ger-
many) was used with bovine serum albumin as standard.
matic styrene epoxidation constitutes the side-product
formation, which was under synthesis conditions (Syn-
thesis 1) up to 2.5 times higher than for the corre-
sponding whole-cell process (13%).[35] Interestingly,
for this whole-cell process only the formation of 2-
phenylethanol was reported, whereas neither aceto-
phenone nor phenylacetaldehyde could be observed.
Table 2. Comparison of different reaction concepts for the biocatalytic synthesis of (S)-styrene oxide utilizing StyA as biocat-
alyst.
Reaction concept
Specific (S)-styrene oxide
STY [mMhꢀ1] Total running time [h] Electron
efficiency [%]
Ref.
[10, 34]
[12]
this study
formation rate [g gStyA hꢀ1]
Whole cell
2.6[a]
0.35
30.6
10
10[b]
ACHTUNGTRENNUNG
Natural system in vitro[c]
6.2
10.5
n.d.
Electroenzymatic concept 0.5[d], 1.3[e]
0.8[d], 2.2[e]
12[d], 2[e]
20.5[d], 5.7[e]
[a]
Assuming that proteins account for 50% of dry cell mass, and StyA overexpression is 25%.[19]
Based on glucose as electron source.
Based on the epoxidation activity towards 3-chlorostyrene.
Synthesis 1.
Synthesis 2.
[b]
[c]
[d]
[e]
2512
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2505 – 2515

Productive Asymmetric Styrene Epoxidation
FULL PAPERS
units (0.8 cmꢂ5 cmꢂ10 cm) and were separated by a Nafion
N324 membrane (DuPont de Nemours, Bad Homburg, Ger-
many). Leak-proof construction was ensured by placing sili-
con gaskets on both frame sides. Cover plates were made
out of glass. The cell assembly was compressed by means of
screw-couplings. External electrical contacting was accom-
plished by screwable, PTFE sealed platinum plugs. An RE-
5B AgjAgCl reference electrode (Bioanalytical Systems
Inc., Warwickshir, UK) complemented the three-electrode
set-up. Electrolyte circulation was maintained in longitudi-
nal direction, vertically upwards, by means of triple distribu-
tion devices placed on both sides for efficient electrode
flow-through. Separate anolyte and catholyte circulation
was carried out with PrepStar SD-1 Solvent Delivery Mod-
ules (Varian Deutschland GmbH, Darmstadt, Germany),
connected to the electrolysis cell and the anodic or cathodic
storage reservoirs by means of PTFE tubings (CS-Chroma-
tographie GmbH, Langerwehe, Germany) and VITON
Fluran HCA F-5500 A (ISMATEC GmbH, Wertheim-Mon-
dfeld, Germany) connectors.
Production of Styrene Monooxygenase Component
StyA
The styA gene was overexpressed in E. coli JM101
(pSPZ10) and purified as described in the literature.[19] The
purification method was modified as follows: clarified super-
natants were loaded onto an XK16/20 column (Amersham
Biosciences, Dꢁbendorf, Switzerland) filled with 28 mL of
Sepabeads EB-QA405 at a flow rate of 2 mLminꢀ1 in
20 mM Tris pH 6.8. Elution was performed at the same flow
rate by applying a linear gradient of 2 mM NaCl minꢀ1 in
20 mM Tris pH 6.8. StyA was fractionated from 40 mM to
240 mM NaCl. During hydrophobic interaction chromatog-
raphy, elution of StyA was achieved by applying a linear
gradient from 1.6M to 0.8M (NH4)2SO4 using 20 mM Tris
pH 7.15, at a rate of 11.8 mMminꢀ1. Fractions were collected
from 1.4M to 1.1M (NH4)2SO4. Desalting was carried out
using Sephadex G25 medium packed in a 16/20 column
(Amersham Biosciences, Dꢁbendorf, Switzerland) at a flow
rate of 5 mLminꢀ1 in 200 mM potassium phosphate buffer
pH 7.25 (buffer that was used later for electroenzymatic syn-
theses). Fractions were concentrated 200-fold using
30.0 kDA MWCO Amicon Ultra-15 Centrifugal Filters (Mil-
lipore Corporation, Schwalbach, Germany) at 3990ꢂg
(48C), aliquoted to a concentration of 5 mgmLꢀ1 and stored
at ꢀ208C. StyA stocks were stored no longer than two
weeks prior to use.
Electrochemical Maintenance Procedures
Nafion N324 membranes were cleaned four times sequen-
tially with 3% H2O2, 1 N H2SO4, and water (808C, 1 h) prior
to first use.[39] RVC electrodes were treated with ultrasound
in methanol, water, and 200 mM potassium phosphate
buffer pH 7.25 (30 min each) before first use. Regular elec-
trochemical electrode conditioning was carried out in
200 mM potassium phosphate buffer (pH 7.25, 308C) by ap-
plying cycling potentials between ꢀ0.9 V and +0.9 V versus
Ag/AgCl for 3 min and 5 times each,[40] at a flow-rate of
15 mLminꢀ1 (room temperature).
Electrochemical Cell
The electrochemical cell was developed for continuous elec-
troenzymatic syntheses and was constructed as a lab-scale
plate and frame filter press cell.[38] Anodes and cathodes
consisted of three-dimensional reticulated vitreous carbon
(RVC) flow-through electrodes (ERG Materials and Aero-
space Corporation, Oakland, CA) having a pore grade of
100 nominal pores per inch (ppi) and being threefold com-
pressed (Figure 6A). Both electrodes (1 cmꢂ5 cmꢂ10 cm)
were moulded into poly(tetrafluoroethene) (PTFE) frame
Electroenzymatic Syntheses
Anolyte (200 mL) and catholyte (120 mL) solutions consist-
ed both of 200 mM potassium phosphate buffer pH 7.25
(308C). Throughout the experiments, anodic and cathodic
Figure 6. (A) Micrograph of the employed three-fold compressed 100 ppi reticulated vitreous carbon (RVC) electrodes
(adapted from http://www.ergaerospace.com). (B) Electroenzymatic reaction set-up. ADC: Analog-to-digital converter chan-
nel, CE: counter electrode (anode), RE: reference electrode, WE: working electrode (cathode).
Adv. Synth. Catal. 2009, 351, 2505 – 2515
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2513

Reto Ruinatscha et al.
FULL PAPERS
storage reservoirs were kept at 378C in order to maintain
constant electrolyte temperatures of 308C inside the electro-
chemical reactor. Before each synthesis reaction, FAD
(e450nm =11.3 mMꢀ1 cmꢀ1)[41] was added to the catholyte and
concentrations were adjusted using a Libra S11 photometer
(Biochrom LTD, Cambridge, UK) at equilibration, and
under continuous pumping (30 mLminꢀ1). Subsequently,
StyA was added to a final catholyte concentration of
0.2 mgmLꢀ1, as well as 2000 U mLꢀ1 catalase. Enzyme stabi-
lizing agents were added optionally to the catholyte in the
following aqueous concentrations: 0.5 mLmLꢀ1 antifoam
204, 2.5 mgmLꢀ1 BSA (bovine serum albumin), and
20 mgmLꢀ1 sucrose. Pre-equilibration of the reaction system
was done as follows: anolyte and catholyte were both circu-
lated for 10 min at 30 mLminꢀ1. Afterwards, the catholyte
was covered with 50 mL dodecane and 6.5 mL styrene
(1M), following electrolyte circulation for another 30 min at
the same flow rate. Subsequently, defined reaction condi-
tions were applied by adjusting flow-rates (10 mLminꢀ1 to
200 mLꢀ1) and aeration (45 mL air minꢀ1 by means of a ca-
pillary, or 92 mL air minꢀ1 using an HPLC frit) into the
cathodic storage reservoir. Syntheses were started 3 min
later by applying a cathodic potential of ꢀ0.75 V versus
AgjAgCl. Reaction control and data recording was per-
formed with an Autolab PGSTAT 302 instrument unit
(Deutsche Metrohm GmbH & Co. KG, Filderstadt, Germa-
ny), equipped with General Purpose Electrochemical
System (GPES) software (version 4.9.006 for Windows)
(Figure 6B).
For determination of enantiomeric excesses (ee), reaction
mixtures were extracted with diethyl ether and dried over
Na2SO4 prior to analysis. Enantiomeric excesses were calcu-
lated using the equation ee=j(SꢀR)j/
ACHTUNGTREN(NGNU S+R), with S and R
representing the concentrations of the two stereoisomers.
Analysis was performed by gas chromatography on a Finni-
gan Focus GC (Thermo Electron Corporation, Langensel-
bold, Germany) equipped with a FI detector and a RT-Beta-
DEXsm (0.25 mmꢂ0.25 mmꢂ30 m, Restek GmbH, Bad
Homburg, Germany) column, using a temperature ramp
from 608C to 1008C (18C minꢀ1) with nitrogen as carrier
gas.
Acknowledgements
The authors express their gratitude to Dr. Klaus-Michael
Mangold (Karl-Winnacker-Institut der DECHEMA e.V.,
Frankfurt) and Prof. Dr. Ing. Jakob Jçrissen (Department of
Biochemical and Chemical Engineering, TU Dortmund) for
helpful discussions and their support in designing the electro-
chemical cell. Furthermore, we thank Mr. Klaus Hirschfeld
and Mr. Peter Schmitz (Department of Biochemical and
Chemical Engineering, TU Dortmund) for expert fabrication
of the electrochemical cell, and D. Ream (ERG Materials
and Aerospace Corporation, Oakland, CA) for excellent
technical advice. This work was financially supported by the
Deutsche Bundesstiftung Umwelt (ICBio, AZ 13123-32), by
the Zentrum fꢀr Angewandte Chemische Genomik, the Euro-
pean Union (EFRE) and by the Ministry of Innovation, Sci-
ence, Research and Technology of North Rhine-Westphalia.
Sample Preparation and Analysis
Aqueous phase samples were diluted with an equal volume
of ice cold acetonitrile. Organic phase samples were extract- References
ed with an equal volume of water. Both samples were mixed
in a vertically positioned Eppendorf thermo mixer (Eppen-
dorf, Hamburg, Germany) for 2 min (1400 rpm, 108C), fol-
lowed by centrifugation for 5 min (48C, 16200ꢂg) in a Her-
aeus Fresco 17 Microcentrifuge (Thermo Electron Corpora-
tion, Langenselbold, Germany). Aqueous phases of the or-
ganic samples were directly analyzed by reversed phase
HPLC. In contrast, samples diluted in acetonitrile were
mixed by gentle inversion prior to reversed phase HPLC
analysis. Concentrations of styrene, (S)-styrene oxide, and
the side-products acetophenone and phenylacetaldehyde
were determined by HPLC on a LaChrom Elite Merck-Hi-
tachi system (Darmstadt, Germany) equipped with a diode
array detector and a reverse phase CC Nucleosil 100–5 C18
HD (Machery-Nagel, Oensingen, Switzerland) column. In-
jected sample volumes were 60 mL, mobile phase consisted
of water and acetonitrile (ratio 60:40), elution was isocratic
(1 mLminꢀ1), and isotherm (258C). The substances were
identified by comparing the retention times to commercially
available standards. Quantifications were carried out by
means of standard curves, recorded under identical condi-
tions as aqueous phase samples, and organic phase samples,
respectively. Space-time yields were normalized to the cath-
ode void volume of the electrochemical cell (45.5 mL), in
order to determine catalyst productivities per reaction
volume. Unless indicated otherwise, average space-time
yields were calculated based on the space-time yields ob-
tained from regular sampling intervals of 15 min.
[1] a) E. N. Jacobsen, A. Pfaltz, H. Yamamoto, (Eds.),
Comprehensive Asymmetric Catalysis, Springer, Berlin,
2004; b) U. Sundermeier, C. Dçbler, M. Beller, Modern
Oxidation Methods, Wiley-VCH, Weinheim, 2004;
c) M. Christmann, S. Brꢃse, Asymmetric Synthesis: The
Essentials, 2nd edn., Wiley-VCH, Weinheim, 2007.
[2] a) W. Zhang, J. L. Loebach, S. R. Wilson, E. N. Jacob-
sen, J. Am. Chem. Soc. 1990, 112, 2801–2803; b) A. U.
Barlan, A. Basak, H. Yamamoto, Angew. Chem. 2006,
118, 5981–5984; Angew. Chem. Int. Ed. 2006, 45, 5849–
5852; c) W. Zhang, H. Yamamoto, J. Am. Chem. Soc.
2007, 129, 286–287.
[3] T. Katsuki, K. B. Sharpless, J. Am. Chem. Soc. 1980,
102, 5974–5976.
[4] D. Lenoir, Angew. Chem. 2006, 118, 3280–3284;
Angew. Chem. Int. Ed. 2006, 45, 3206–3210.
[5] S. Enthaler, K. Junge, M. Beller, Angew. Chem. 2008,
120, 3363–3367; Angew. Chem. Int. Ed. 2008, 47, 3317–
3321.
[6] F. G. Gelalcha, B. Bitterlich, G. Anilkumar, M. K. Tse,
M. Beller, Angew. Chem. 2007, 119, 7431–7435;
Angew. Chem. Int. Ed. 2007, 46, 7293–7296.
[7] Y. Shi, Acc. Chem. Res. 2004, 37, 488–496.
[8] C. J. Chang, J. A. Labinger, H. B. Gray, Inorg. Chem.
1997, 36, 5927–5930.
[9] W. J. H. van Berkel, N. M. Kamerbeek, M. W. Fraaije, J.
Biotechnol. 2006, 124, 670–689.
2514
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2505 – 2515

Productive Asymmetric Styrene Epoxidation
FULL PAPERS
[10] J.-B. Park, B. Bꢁhler, T. Habicher, B. Hauer, S. Panke,
B. Witholt, A. Schmid, Biotechnol. Bioeng. 2006, 95,
501–512.
[24] V. Massey, J. Biol. Chem. 1994, 269, 22459–22462.
[25] A. M. d. A. Rocha Gonsalves, A. C. Serra, J. Chem.
Soc. Perkin Trans. 2 2002, 715–719.
[11] R. Gross, B. Hauer, K. Otto, A. Schmid, Biotechnol.
Bioeng. 2007, 98, 1123–1134.
[26] C. Kohlmann, W. Mꢃrkle, S. Lꢁtz, J. Mol. Catal. B:
Enzym. 2008, 51, 57–72.
[12] K. Hofstetter, J. Lutz, I. Lang, B. Witholt, A. Schmid,
Angew. Chem. 2004, 116, 2215–2218; Angew. Chem.
Int. Ed. 2004, 43, 2163–2166.
[27] F. Hollmann, K. Hofstetter, A. Schmid, Trends Biotech-
nol. 2006, 24, 163–171.
[28] V. L. Vilker, V. Reipa, M. Mayhew, M. J. Holden, J.
[13] C. Wandrey, Chem. Rec. 2004, 4, 254–265.
[14] F. Hollmann, K. Hofstetter, T. Habicher, B. Hauer, A.
Schmid, J. Am. Chem. Soc. 2005, 127, 6540–6541.
[15] E. Steckhan, Topics in Current Chemistry, Vol. 170,
Springer, Berlin, 1994.
Am. Oil Chem. Soc. 1999, 76, 1283–1289.
[29] A. K. Udit, F. H. Arnold, H. B. Gray, J. Inorg. Biochem.
2004, 98, 1547–1550.
[30] V. Reipa, M. P. Mayhew, V. L. Vilker, Proc. Natl. Acad.
Sci. USA 1997, 94, 13554–13558.
[31] N. Shirane, Z. Sui, J. A. Peterson, P. R. Ortiz de Mon-
tellano, Biochemistry 1993, 32, 13732–13741.
[16] R. Ruinatscha, V. Hçllrigl, K. Otto, A. Schmid, Adv.
Synth. Catal. 2006, 348, 2015–2026.
[17] R. Ruinatscha, K. Buehler, A. Schmid, unpublished re-
sults.
[32] A. Kantz, F. Chin, N. Nallamothu, T. Nguyen, G. T.
Gassner, Arch. Biochem. Biophys. 2005, 442, 102–116.
[33] a) W. A. Duetz, J. B. van Beilen, B. Withold, Curr.
Opin. Biotechnol. 2001, 12, 419–425; b) B. Bꢁhler, A.
Schmid, J. Biotechnol. 2004, 113, 183–210.
[34] J.-B. Park, Dissertation, ETH No. 15754, 2004, Zꢁrich.
[35] S. Panke, M. G. Wubbolts, A. Schmid, B. Witholt, Bio-
technol. Bioeng. 2000, 69, 91–100.
[36] R. J. Beynon, J. S. Easterby, Buffer Solutions: The
Basics, BIOS Scientific Publishers, Oxford, UK, 1996.
[37] M. M. Bradford, Anal. Biochem. 1976, 72, 248–254.
[38] J. D. Genders, D. Pletcher, Electrosynthesis – from Lab-
oratory to Pilot, to Production, Electrosynthesis Com-
pany, Lancaster, NY, 1990.
[18] A. Schmid, K. Hofstetter, H.-J. Feiten, F. Hollmann, B.
Witholt, Adv. Synth. Catal. 2001, 343, 732–737.
[19] K. Otto, K. Hofstetter, M. Rçthlisberger, B. Witholt,
A. Schmid, J. Bacteriol. 2004, 186, 5292–5302.
[20] a) D. Pletcher, F. C. Walsh, Industrial Electrochemistry,
2nd edn., Chapman and Hall, London, 1990; b) H.
Wendt, G. Kreysa, Electrochemical Engineering: Sci-
ence and Technology in Chemical and Other Industries,
Springer, Berlin, 1999; c) V. M. Schmidt, Elektrochemi-
sche Verfahrenstechnik: Grundlagen, Reaktionstechnik,
Prozessoptimierung, Wiley-VCH, Weinheim, 2003.
[21] D. Pletcher, I. Whyte, J. Appl. Electrochem. 1991, 21,
659–666.
[39] H. Wang, G. A. Capuano, J. Electrochem. Soc. 1998,
145, 780–784.
[40] J. Wang, Electrochim. Acta 1981, 26, 1721–1726.
[41] L. G. Whitby, Biochem. J. 1953, 54, 437–442.
[22] a) L. Gianfreda, M. R. Scarfi, Mol. Cell. Biochem.
1991, 100, 97–128; b) F. Schulz, F. Leca, F. Hollmann,
M. T. Reetz, Beilstein J. Org. Chem. 2005, 1.
[23] K. H. Dudley, A. Ehrenberg, P. Hemmerich, F. Mꢁller,
Helv. Chim. Acta 1964, 47, 1354–1383.
Adv. Synth. Catal. 2009, 351, 2505 – 2515
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2515