Electrocatalytic Biosynthesis using a Bucky Paper Functionalized by [Cp*Rh(bpy)Cl]+ and a Renewable Enzymatic Layer

Article abstract of DOI:10.1002/cctc.201800681

A bioelectrode for electroenzymatic synthesis was prepared, combining a layer for NADH regeneration and a renewable layer for enzymatic substrate reduction. The covalent immobilization of a rhodium complex mediator ([Cp*Rh(bpy)Cl]⁺) on the surface of a bucky paper electrode was achieved by following an original protocol in two steps. A bipyridine ligand was first grafted on the electrode by electro-reduction of bipyridyl diazonium cations generated from 4-amino-2,2′-bipyridine, and the complex was then formed by reaction with [RhCp*Cl₂]₂. A turnover frequency of 1.3 s^?1 was estimated for the electrocatalytic regeneration of NADH by this immobilized complex, with a Faraday efficiency of 83 %. The bucky paper electrode was then overcoated by a bio-doped porous layer made of glassy fibers with immobilized D-sorbitol dehydrogenase. This assembly allowed for the efficient separation of the enzyme and the rhodium catalyst, which is a prerequisite for effective bioelectrocatalysis with such bioelectrochemical system, while allowing effective mass transport of NAD⁺/NADH cofactor from one layer to the other. Thereby, it was possible to reuse the same mediator-functionalized bucky paper with three different enzyme layers. The bioelectrode was applied to the electroenzymatic reduction of D-fructose to D-sorbitol. A turnover frequency of 0.19 s^?1 for the rhodium complex was observed in the presence of 3 mM D-fructose and a total turnover number higher than 12000 was estimated.

Full text of DOI:10.1002/cctc.201800681

Accepted Article
Title: Electrocatalytic biosynthesis using a bucky paper functionalized
by [Cp*Rh(bpy)Cl]+ and a renewable enzymatic layer
Authors: Lin Zhang, Mathieu ETIENNE, Neus Vila, Thi Xuan Huong Le,
Gert Kohring, and Alain Walcarius
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201800681
Link to VoR: http://dx.doi.org/10.1002/cctc.201800681
A Journal of
www.chemcatchem.org

10.1002/cctc.201800681
ChemCatChem
ARTICLE
Electrocatalytic biosynthesis using a bucky paper functionalized
by [Cp*Rh(bpy)Cl]+ and a renewable enzymatic layer
Lin Zhang,[a] Mathieu Etienne,*[a] Neus Vilà,[a] Thi Xuan Huong Le,[a] Gert-Wieland Kohring,[b] Alain
Walcarius[a]
Abstract:
A
bioelectrode for electroenzymatic synthesis was
potential in enzymatic synthesis as it avoids the use of additional
reagents in the medium.[5] However, for NAD(P)H regeneration, a
catalyst is required to decrease the high overpotential of the
reaction. The redox mediator [Cp*Rh(bpy)Cl]+ has been proved to
be the most efficient non-enzymatic catalyst for the reduction of
NAD(P)+ to NAD(P)H.[6,7] The presence of [Cp*Rh(bpy)Cl]+ not
only reduced the overpotential needed for this reaction, but also
avoided the unfavorable formation of inactive NAD(P)2 dimer.[8–10]
Its immobilization on electrodes in a bioelectrocatalytically active
form is remaining challenging, but highly recommended when
intended to be applied in bioconversion. Indeed, in large scale
synthesis, the purification of reaction products can be facilitated if
the catalyst is not present in the final mixture solution. Moreover,
the immobilization of rhodium complex can increase the
reusability of the catalyst, which is economically desirable.[11]
Finally, some authors observed that this complex interacts with
some functional groups of the proteins, leading to a degradation
of its catalytic properties. Thus, the separation in space of the
rhodium complex from the enzymes will significantly increase the
operational stability of this mediator as well as the lifetime of the
system.[12–14] In order to overcome this challenge, the separated
immobilization of rhodium complex and enzymes can be
considered as a wise choice.[14] However, it is also desirable to
prepared, combining a layer for NADH regeneration and a renewable
layer for enzymatic substrate reduction. The covalent immobilization
of a rhodium complex mediator ([Cp*Rh(bpy)Cl]+) on the surface of a
bucky paper electrode was achieved by following an original protocol
in two steps. A bipyridine ligand was first grafted on the electrode by
electro-reduction of bipyridyl diazonium cations generated from 4-
amino-2,2’-bipyridine, and the complex was then formed by reaction
with [RhCp*Cl2]2. A turnover frequency of 1.3 s-1 was estimated for the
electrocatalytic regeneration of NADH by this immobilized complex,
with a Faraday efficiency of 83 %. The bucky paper electrode was
then overcoated by a bio-doped porous layer made of glassy fibers
with immobilized D-sorbitol dehydrogenase. This assembly allowed
for the efficient separation of the enzyme and the rhodium catalyst,
which is a prerequisite for effective bioelectrocatalysis with such
bioelectrochemical system, while allowing effective mass transport of
NAD+/NADH cofactor from one layer to the other. Thereby, it was
possible to reuse the same mediator-functionalized bucky paper with
three different enzyme layers. The bioelectrode was applied to the
electroenzymatic reduction of D-fructose to D-sorbitol. A turnover
frequency of 0.19 s-1 for the rhodium complex was observed in the
presence of 3 mM D-fructose and a total turnover number higher than
12000 was estimated.
keep
a close contact between the catalyst for NAD(P)H
regeneration and the enzymes. Moreover, this separated
immobilization would give the possibility to renew independently
either the molecular catalyst or the enzymatic catalyst.
Introduction
Regarding the immobilization of the rhodium complex on
electrode surfaces, some strategies based on weak interactions
have been proposed, such as its accommodation into conductive
polymer matrices,[15–19] or its soft binding to carbon nanomaterials
(e.g. carbon nanotube[20] or graphene[21]) via non-covalent π-π
interactions. More durable immobilization could be expected in
case of covalent binding. We have recently proposed an approach
for the functionalization of a porous carbon electrode (carbon felt,
CF), with covalently immobilized [Cp*Rh(bpy)Cl]+ moieties,
involving four steps:[22] (1) modification of CF with multi-walled
carbon nanotube (CF-MWCNT), (2) functionalization of the CF-
MWCNT electrode by azide groups via diazonium grafting, one of
The β-nicotinamide adenine dinucleotide NAD(P)H is an
important cofactor in redox reactions involving oxidoreductase
enzymes. During the electroenzymatic synthesis, NAD(P)H acts
as an electron donor/acceptor that allows hydride transfer.[1] The
release of a hydride from the reduced nicotinamide moiety leads
to the production of NAD(P)+. To avoid the continuous addition of
this expensive cofactor, it is necessary to find an efficient way to
recycle it in long-term experiments. Various approaches have
been developed for such cofactor regeneration, including
chemical, electrochemical, photochemical, microbial and
enzymatic reactions.[2–4] Electrochemical regeneration via
continuous electron feeding provided by the electrode has great
the
most
convenient
method
for
carbon
surface
functionalization,[23] (3) click chemistry between the azido-
functionalized electrode and an alkynyl-bipyridine derivative, and
(4) complexation with (RhCp*Cl2)2. The obtained electrode was
then successfully applied to the electrochemical regeneration of
NADH (87% Faraday efficiency, turnover of 3790 and turnover
frequency of 0.05 s−1). This electrode was even modified with
redox enzymes and the resulting electroenzymatic system
showed good catalytic response, but its long-term stability was
rather poor (40% of the catalytic current for NAD+ reduction was
lost over 90h of continuous electrochemical operation).
[a]
[b]
Dr. L. Zhang, Dr. N. Vilà, Dr. A. Walcarius, and Dr. M. Etienne
Laboratoire de Chimie Physique et Microbiologie pour les Matériaux
et l’Environnement, UMR 7564
CNRS-Université de Lorraine,
405 rue de Vandoeuvre, 54600 Villers-lès-Nancy, France
E-mail: mathieu.etienne@univ-lorraine.fr
Dr. G. Kohring
Microbiology, Saarland University,
Campus, Geb. A1.5, D-66123 Saarbruecken, Germany.
In order to improve NADH regeneration and to increase the
stability of the electroenzymatic synthesis system, two major
Supporting information for this article is given via a link at the end of
the document
contributions are described here, (1)
a new protocol for
This article is protected by copyright. All rights reserved.

10.1002/cctc.201800681
ChemCatChem
ARTICLE
immobilization of the rhodium catalyst and (2) a new concept of
renewable enzymatic layer. Firstly, the three initial steps in the
above study can be combined in single one by directly
electrografting the bipyridine ligand modified with an amino group
with covalently immobilized [Cp*Rh(bpy)Cl]+ was obtained (BP-
bpy-Rh). Compared to previous report on covalent attachment of
this rhodium complex by combining diazonium electrografting and
click chemistry,[22] not only the preparation process was simplified,
but the distance between the rhodium center and the electrode
surface has been reduced and thus, the electron transfer reaction
is expected to be improved.
on
a bucky paper (BP) electrode. This route was used
successfully beforehand for the immobilization of some other
transition metal complexes (e.g. Mn, Re, Fe, Co, Ni, or Cu) on
carbon-based electrodes.[24,25] Compared to the CF-MWCNT
composite, a BP electrode is prepared by filtering the MWCNT
solution without using CF as substrate.[26–28] This approach
circumvents the low adhesion of MWCNT on CF, so enhancing
the longevity of the prepared electrode. Secondly, the physical
separation between the mediator and the protein is mandatory to
avoid inhibition effects, and the possibility of reversible assembly
of enzymes and catalysts for NADH regeneration is desired. This
last point is motivated by the fact that the enzymes and the
rhodium complex have different lifetimes. Enzymes stability is
limited (e.g. DSDH has a half-life period of 48 h at 28°C, 5 h at
35°C and 0.2 h at 40°C in solution),[29] while immobilized
[Cp*Rh(bpy)Cl]+ was proved to be stable at least for 2 weeks in
solution.[22] As soon as the enzymes lose their activity, the
electrode with co-immobilized enzymes and rhodium catalyst has
to be discarded, justifying the development of a selective and
reversible immobilization procedure.
Figure 1
Diazonium electrografting was performed on BP electrode by
cyclic voltammetry from 0.4 V to -0.8 V in an acidic aqueous
solution containing 1mM of 4-amino-2,2’-bipyridine and 2 mM of
sodium nitrite (see the two first scans in Figure 2A). During the
first scan, an irreversible wave located around -0.3 V vs. Ag/AgCl
was observed (curve a), that corresponds to the grafting of
bipyridine accompanied by the release of N2. During the second
scan (curve b), the cathodic reduction signal was lower (almost
disappeared) than during the first scan because of the partial
blocking of the BP electrode surface. From that observation, it can
be considered that the electrografting step is successful.
Figure 2
In this study, a novel method for separated immobilization of the
rhodium catalyst and enzymes is proposed and applied to
electroenzymatic synthesis. We built a co-immobilized electrode
by the assembly of a bucky paper electrode functionalized with
[Cp*Rh(bpy)Cl]+ (BP-bpy-Rh) with different enzyme layers. An
individual layer with enzymes was prepared by depositing D-
sorbitol dehydrogenase (DSDH) in silica gel on a piece of
microfiber filter, then it was overlaid on the top of BP-bpy-Rh
electrode. The distance between enzymes and rhodium
complexes was close enough to enable the electroenzymatic
reaction, but physically restricted in two separated layers. In this
case, the interactions between the rhodium complex and the
functional groups of the enzymes were prevented completely, in
contrary to sol-gel encapsulated enzyme layers directly deposited
on an electrode surface grafted with the catalyst.[22] The
To further validate this hypothesis, the BP electrode has been
analyzed by XPS after electrografting of the diazonium cation
generated in situ from the amino-bipyridine. The N1s high
resolution spectrum obtained after modification shows a main
component located at 398.8 eV attributed to the presence of
bipyridine functions immobilized on the electrode surface (Figure
3A).[31] Then, the bipyridine functionalized electrode (BP-bpy) was
allowed to react with ((RhCp*Cl2)2 to form the rhodium complex
on the BP electrode surface (BP-bpy-Rh). The Rh signal is
observed around 312 eV in the XPS survey spectrum of BP-bpy-
Rh, in addition to the C1s (285 eV), O1s (532 eV) and N1s (400
eV) peaks (Figure S2). The main envelope of the Rh3d core level
spectrum (Figure 3C) has been curve-fitted with Rh3d5/2 and
Rh3d3/2 peaks, respectively centered at 308.1 and 312.6 eV. The
weaker peaks at higher binding energies could correspond to the
reusability of the rhodium functionalized BP electrode was studied, presence of Rh in higher oxidized state. These signals are not due
and the new system was characterized and applied to the
electroenzymatic conversion of D-fructose to D-sorbitol.
to the presence of satellites peaks since they are usually
observed when there are unpaired electrons. Changes have been
also observed for the N1s core level spectrum, with a shift of the
main component up to 400.5 eV, as a result of the complexation
with Rh (Figure 3B), consistent with previous observations made
for bipyridine ligands complexed with osmium.[32]
Results and Discussion
Covalent immobilization of [Cp*Rh(bpy)Cl]+ on BP electrode
Figure 1 illustrates the immobilization route for rhodium complex,
involving a first electrografting process and a subsequent
complexation reaction. 4-amino-2,2’-bipyridine (compound A)
was reacting quantitatively in the solution to produce bipyridyl
diazonium cations[30] (compound B). A negative potential was
applied to promote the generation of the reactive aryl radical and
the release of N2 in a concerted process, resulting in the covalent
functionalization of the bucky paper electrode surface with 2,2’-
bipyridine moieties (BP-bpy). After a final complexation step with
rhodium dimmer ((RhCp*Cl2)2 (compound C), the BP electrode
Figure 3
The catalytic response of BP-bpy-Rh electrode for NADH
regeneration was evaluated by cyclic voltammetry (Figure 2B). In
the absence of NAD+, a cathodic peak was observed around -
0.65V, which was higher than the potential observed with the
complex covalently attached by click chemistry.[22] This potential
shift indicating a facilitated electron transfer can be ascribed to
the influence of the linker (i.e., smaller length) or the electrode
material on the electron density of the rhodium complex center.[33]
This article is protected by copyright. All rights reserved.

10.1002/cctc.201800681
ChemCatChem
ARTICLE
The addition of NAD+ in the solution, from 0.5 to 3 mM, led to
increasing cathodic currents and somewhat negatively-shifted
cathodic peaks, in agreement to what was reported for the
electrocatalytic reduction of NAD+ using this Rh complex in
solution.[34] A maximum cathodic current of 363 µA cm-2 was
observed at the concentration of 3 mM NAD+. In addition, the
catalytic activity toward NADH regeneration of BP-bpy-Rh
electrode was also confirmed by amperometry at constant
potential (-0.78 V vs. Ag/AgCl).
Mediator and enzyme co-immobilization with a replaceable
enzyme layer
Figure 4A depicts the system that has been developed. DSDH
was encapsulated inside a silica gel, deposited in a 260 µm thick
glassy fiber layer. The rhodium complex was immobilized at the
surface of the bucky paper electrode, as described in section 3.1,
and the bio-doped glassy fiber membrane was placed on the top
of the BP-bpy-Rh electrode. The combination of enzymatic
reaction to produce D-Sorbitol from D-fructose and
electrochemical cofactor regeneration was realized by diffusion of
NAD+/NADH between/through the two layers. Optical images of
the assembly and SEM of the bucky paper are presented in
Figure S5. The bioelectrocatalytic response of this electrode was
tested in phosphate buffer, in the presence of 1 mM NADH, by
chronoamperometry at -0.72 V vs. Ag/AgCl. As shown in Figure
4B, the current increased after additions of D-fructose from 0.5 to
2.5 mM. An apparent Km of 0.25 mM was calculated, i.e.
comparable to the Km estimated without complete separation of
enzymes and rhodium catalyst (0.4 mM, data reported in Figure
S4). This result shows that the bioelectrode prepared with
separated layer is really efficient for D-fructose reduction in
combination with NADH regeneration.
As shown in Figure S3A, the catalytic current increased after
additions of NAD+ from 0.5 to 2.5 mM and reached saturation.
After 20 min of electrolysis in that conditions, 0.217 mM NADH
was detected by UV spectroscopy (Figure S3B). By measuring
with CV the quantity of electroactive rhodium complex on the
electrode, 1.27x10-9 mole, a turnover frequency of 1.3 s-1 was
calculated, which is 30 times higher than rate constant estimated
with the same complex immobilized by click chemistry.[22] This
value is also very close to the highest value ever reported in the
literature for an immobilized rhodium complex, 3.6 s-1, achieved
with Rh complex linked to multiwalled carbon nanotubes via π−π
stacking.[20] By measuring the charge involved in the
electrocatalytic process, 0.506 C, a Faraday efficiency of 83 %
was calculated. After confirming the catalytic response of the thin
BP-bpy-Rh electrode, its application in electroenzymatic
synthesis was further explored.
Figure 4
The reusability of BP-bpy-Rh electrode was then investigated.
After each experiment, the DSDH layer is easily
separated/removed from the BP-bpy-Rh electrode surface, and a
new active enzymatic layer can be packed onto the same BP-bpy-
Rh electrode to reinitiate the electroenzymatic synthesis. The
reusability of such BP-bpy-Rh/bio-doped glassy fiber electrode
was proved by measuring the amperometric response to 3 mM D-
fructose with three different DSDH gel layers, each experiment
lasting 25 min. The catalytic current of the three bioelectrodes
stabilized at a current of 37 ± 1 µA cm-2 for the same BP-bpy-Rh
electrode, showing good repeatability and stability (see Figure
S6).
Mediator and enzyme co-immobilization with bio-doped gel
on the top
The electroenzymatic catalytic response was first tested by
directly depositing the D-sorbitol dehydrogenase (DSDH) gel on
the top of the BP-bpy-Rh electrode (Figure S4A). In this
bioelectrochemical system, D-fructose was reduced to D-sorbitol
by DSDH in the presence of NADH. This enzymatic reaction led
to the production of NAD+ that was reduced back to NADH by the
rhodium complex at the BP-bpy-Rh electrode surface. This
electrode assembly was named BP-bpy-Rh-DSDH-gel
The bioelectrocatalytic response of BP-bpy-Rh-DSDH-gel
electrode was demonstrated by gradually adding 1 mM of D-
fructose substrate into 50 mM PBS buffer in the presence of 0.5
mM NADH as cofactor. All the experiments were performed in the
absence of oxygen. As shown in Figure S4B, the cathodic current
increased with the concentration of D-fructose and reached a
maximum current, following the Michaelis–Menten kinetics, an
apparent Km was estimated to be 0.4 mM, much less than the Km
obtained from free DSDH in the solution (Km is 49.4 mM).[29]
However, this bio-electrode suffered from a limited long-term
stability, as shown by the dramatic decrease of the catalytic
current after 6h of electrolysis (see Figure S4C). This decrease
could be attributed to the degradation of enzymes or Rh complex
caused by the interaction between Rh metal centre and some
surface functional groups of enzymes.[14] We could improve the
long-term stability of the electroenzymatic system with a real
physical separation between the protein and the mediator, which
will be realized hereafter by immobilizing enzymes and rhodium
complex into separate layers.
Figure 5
The electrode was finally applied for the full bioconversion of a D-
fructose solution, generating D-sorbitol when used in combination
to DSDH (Figure 5). A 4 cm2 BP-bpy-Rh electrode was packed
with a 4 cm2 enzymatic layer. The electrochemical reactor
contained 30 mL of phosphate buffer (50 mM, pH 6.5) with 1 mM
NADH as cofactor. A potential of -0.72 V vs. Ag/AgCl was applied
to the working electrode. After addition of 3 mM D-fructose, the
catalytic current reached rapidly -48 µA cm-2 and decreased then
slowly to -5 µA cm-2 while D-fructose was converted to D-sorbitol
(Figure 5A) over the 95 h of the experiment. The evolution of
substrate and product concentrations during the bioconversion is
shown in Figure 5B. 87% of D-fructose was transformed to D-
sorbitol after 95 h and a high Faraday efficiency of 83% was
obtained. This value is the same as the one calculated from the
electrocatalytic oxidation of NAD+ into NADH (i.e., without
enzymatic reaction, Figure S3), showing that the catalytic activity
This article is protected by copyright. All rights reserved.

10.1002/cctc.201800681
ChemCatChem
ARTICLE
of the BP-bpy-Rh electrode was not affected by the association
with the enzymes. The amount of rhodium complex on the BP
electrode, 6.5×10-9 mole, was estimated from cyclic voltammetry
by integrating the cathodic peak. From that amount, and
considering the conversion of 0.44 ± 0.07 mol L-1 of D-fructose to
D-sorbitol after 3 hours of the electroenzymatic synthesis, a
turnover frequency of 0.19 ± 0.03 s-1 was calculated. A total
turnover of more than 12000 was obtained for the rhodium
complex after 95h, which is the highest value ever obtained with
rhodium complex applied in electroenzymatic synthesis.[21,35]
Moreover, the total turnover of NAD+ cofactor was 2.61, which
further proves that NADH produced with this electrode was
enzymatically active. The turnover frequency of the enzyme, as it
can be estimated from the amount of protein introduced in the
electrode, is 0.33 s-1, very close to the turnover frequency of the
molecular catalyst on the electrode surface (1.3 s-1 in the
regeneration of NADH, 0.19 s-1 in the electroenzymatic synthesis)
and more than 2 order of magnitude lower than the turnover
frequency of the protein as determined in solution (138.4 s-1). The
electroenzymatic synthesis is thus essentially limited by the
electrocatalytic regeneration of NADH. The performance of the
system still needs to be improved, nevertheless, the results that
we are presenting are among the best reported in the literature for
such immobilized rhodium complex and a rare example of
electrosynthesis with enzyme and rhodium catalyst immobilized
on electrode. This electrode architecture is thus promising to be
applied in larger scale bioconversion experiments considering the
simplicity of the protocol and the possibilities for increasing the
size of the bioelectrode, bucky paper electrodes are indeed
already commercialized.
optimization of the bioreactor and its application to redox proteins
showing an interesting substrate spectrum, such as
acenaphthenol-dehydrogenase and P450 cytochromes.
Experimental Section
Chemicals
β-Nicotinamide adenine dinucleotide hydrate (NAD+, ≥ 96.5%, Sigma-
Aldrich), β-nicotinamide adenine dinucleotide, reduced disodium salt
hydrate (NADH, ≥97%, Sigma-Aldrich), 4-azidoaniline hydrochloride (97%,
Sigma-Aldrich), sodium nitrite (≥97.0%, Sigma-Aldrich), D-fructose (99
wt%, Sigma), sodium dihydrogen phosphate (99.5%, Merck),
pentamethylcyclo-pentadienylrhodium (III) chloride dimer ((RhCp*Cl2)2,
Sigma-Aldrich), dichloromethane (DCM, Carlo Erba), 1 M hydrochloric acid
solution (Sigma-Aldrich), tetraethoxysilane (TEOS, 98%, Alfa Aesar), 3-
glycidoxypropyltrimethoxysilane
(GPS,
98%,
Sigma-Aldrich),
poly(ethylene imine) (PEI, 50% w/v in water, Mn = 60000, Fluka), multi-
walled carbon nanotubes (MWCNT, > 95%, Φ 6-9nm, L 5µm, Sigma-
Aldrich), glassy fiber layer (0.26 mm thickness, Whatman®), PVDF
membrane filter (0.45 µm pore size, Durapore®), were used as received.
Aqueous solutions were prepared with high purity water (18 MΩ cm) from
a Purelab Option water purification system (Elga LabWater, Veolia Water
STI, Antony, France). Overproduction of N-His(6) D-sorbitol
dehydrogenase (800 U mL-1 ) was done with Escherichia coli BL21GOLD
(DE3) containing the corresponding expression-vectors pET-24a(+)
(Novagen) and purification of the enzyme was performed with Histrap
columns (GE Healthcare).[36]
Immobilization of [Cp*Rh(bpy)Cl]+ on a bucky paper electrode
Preparation of bucky paper electrode (BP): The bucky paper electrode
was prepared following the reported literature.[28] 10 mg MWCNT was
dispersed in 50 mL ethanol by ultrasonication for 5 h. Afterwards, the
suspension was decanted and vacuum filtered using the PVDF membrane
filter. Then, the material was dried at 50 ℃ overnight. Finally, the electrode
was cut at the suitable size before use.
Conclusions
A bioelectrode for electroenzymatic synthesis was prepared,
combining a layer for NADH regeneration and a renewable layer
for enzymatic substrate reduction. [Cp*Rh(bpy)Cl]+ was
Synthesis of 4-amino-2,2’-bipyridine: 4-amino-2,2’-bipyridine was
synthesized following the reported protocol,[25,37,38] the scheme of synthetic
route is shown in Figure S1.
covalently immobilized onto
a bucky paper electrode via
diazonium coupling and complexation processes. The
immobilized [Cp*Rh(bpy)Cl]+ showed good catalytic activity
toward NADH regeneration, reaching a turnover frequency of 1.3
s-1. The reusability of the [Cp*Rh(bpy)Cl]+ functionalized BP
electrode was demonstrated by packing replaceable glassy fiber
layers with immobilized DSDH, leading to comparable current
responses. Finally, this bioelectrode was applied to the
bioconversion of D-fructose to D-sorbitol in the presence of the
cofactor NADH, leading to a conversion rate of 87% after 95 h and
a Faraday efficiency of 83%. In these conditions, the total turnover
of the rhodium catalyst was more than 12000 and the turnover
frequency was 0.19 s-1. Moreover, the total turnover of NAD+
cofactor was 2.6, which further proves that NADH produced with
this electrode was enzymatically active. This constitutes
significant advances in the field of electroenzymatic synthesis
with NAD(P)-dependent redox proteins, both in terms of
enhanced stability and improved efficiency of the catalyst for
NADH regeneration, and flexibility of the bioelectrode
construction/regeneration. Future researches could see the
2,2’-bipyridine-N-oxide: 5 g of 2,2’-bipyridine was dissolved in 25 mL of
trifluoroacetic acid, then 5.32 mL of 30% H2O2 was slowly added into the
solution under stirring. The solution was stirred at room temperature for 4
h, its pH was adjusted to 9 by addition of a small volume of 25% NaOH
aqueous solution, before to be extracted with CHCl3 (30 mL×5). The
organic phase was gathered and dried over Na2SO4, filtered and vacuum
evaporated. The product was a pale yellow oil.
4-nitro-2, 2’-bipyridine-1-oxide: 2.1 g of 2,2’-bipyridine-N-oxide and 6.7 g
of potassium nitrate were dissolved in 16.3 mL of concentrated (97%)
sulfuric acid and stirred for 30 h at 95 ℃. Then the resulting solution was
poured into 50 g of ice and its pH was adjusted to 9.0 by adding a small
volume of 25% NaOH aqueous solution. A yellow precipitate was formed.
It was collected by vacuum filtration and washed with cold water. After
drying the solid in air, it was dissolved in DCM, filtered off and then, DCM
was removed by vacuum evaporation. The product was a yellow solid.
4-amino-2,2’-bipyridine: 500 mg of 4-nitro-2,2’-bipyridine-1-oxide was
dissolved in 100 mL of methanol, 115 mg of 10% Pd/C catalyst was added
This article is protected by copyright. All rights reserved.

10.1002/cctc.201800681
ChemCatChem
ARTICLE
under nitrogen atmosphere with vigorous stirring. The solution was
incubated in an ice bath and 1.25 g of sodium borohydride was slowly
added. The solution was stirred at 4 ℃ overnight, until the gas evolution
ceased. The catalyst was removed by gravity filtration, and methanol was
removed from the filtrate by vacuum evaporation. 30 mL of water was
added to dissolve the solid that was then extracted with DCM (50 mL×5).
The organic phase was dried with NaSO4 and then, the solvent was
removed. The product was a white solid.
This work was supported partly by the French PIA project «
Lorraine Université d’Excellence », reference ANR-15-IDEX-04-
LUE. L.Z. gratefully acknowledges CNRS and Region Grand Est
for Ph.D. funding. We also thank Aurelien Renard (LCPME) for
XPS measurements.
Keywords: [Cp*Rh(bpy)Cl]+ mediator • dehydrogenase • NADH
• co-immobilization • bioelectrocatalytic reactor
2, 2’-bipyridine functionalized BP electrode
[1]
[2]
[3]
[4]
C. E. Paul, I. W. C. E. Arends, F. Hollmann, ACS Catal. 2014, 4,
788–797.
1 mM of 4-amino-2,2’-bipyridine was mixed with 2 mM of sodium nitrite in
0.5 M HCl aqueous solution under stirring for 5 min to generate 2,2’-
bipyridyl diazonium cations. The electrografting on the BP electrode was
achieved by running two cyclic voltammograms from 0.4 V to -0.8 V in this
solution. These parameters were adapted from two previous work
involving diazonium electrografting.[22,39] After electrografting, the
electrode was rinsed carefully with water and left to dry before use.
A. Weckbecker, H. Gröger, W. Hummel, Adv. Biochem. Eng.
Biotechnol. 2010, 120, 195–242.
R. Wichmann, D. Vasic-Racki, Adv. Biochem. Eng. Biotechnol.
2005, 92, 225–260.
V. Urbanova, G. W. Kohring, T. Klein, Z. Wang, O. Mert, M.
Emrullahoglu, K. Buran, A. S. Demir, M. Etienne, A. Walcarius, Z.
Phys. Chem. 2013, 227, 667–689.
[Cp*Rh(bpy)Cl]+ functionalized BP electrode
The 2,2’-bipyridine functionalized electrode was immersed for 4 h into a
DCM solution containing 0.15 mM ((RhCp*Cl2)2. A gentle stirring was
applied. Then, the electrode was rinsed with DCM before to be immersed
inside a second DCM volume for 5 min to remove the unreacted residues.
[5]
L. Zhang, M. Etienne, N. Vilà, A. Walcarius, in Funct. Electrodes
Enzym. Microb. Electrochem. Syst. (Eds.: N. Brun, V. Flexer), World
Scientific, 2017, pp. 215–271.
[6]
[7]
[8]
[9]
R. Ruppert, S. Herrmann, E. Steckhan, Tetrahedron Lett. 1987, 28,
6583–6586.
Immobilization of DSDH inside a glassy fiber layer
F. Hollmann, B. Witholt, A. Schmid, J. Mol. Catal. B Enzym. 2002,
19–20, 167–176.
A sol containing DSDH was prepared following the reported protocol.[40]
0.18 g of TEOS and 0.13 g of GPS were mixed together with 0.5 mL water
and 0.625 mL 0.01 M HCl and stirred overnight. Then, this sol was diluted
3 times and a 40 µL aliquot was mixed with 20 µL of PEI solution (20%),
20 µL of water and 30 µL of DSDH stock solutions. A piece of glassy fiber
layer that had the same shape as the bucky paper electrode was prepared.
It was then modified by the sol containing DSDH (50 µL cm-2). Before use,
the modified layer was dried overnight at 4 ℃. The glassy fiber layer
modified with DSDH (1 cm2 or 4 cm2) was put on the top of the
[Cp*Rh(bpy)Cl]+ functionalized bucky paper electrode (same geometric
surface area) and the electroenzymatic experiment was performed.
F. Hildebrand, C. Kohlmann, A. Franz, S. Lütz, Adv. Synth. Catal.
2008, 350, 909–918.
J. Gajdzik, J. Lenz, H. Natter, a. Walcarius, G. W. Kohring, F.
Giffhorn, M. Gꢀllꢁ, a. S. Demir, R. Hempelmann, J. Electrochem.
Soc. 2012, 159, F10.
[10]
[11]
[12]
[13]
L. Gorton, E. Dominguez, in Encycl. Electrochem., Wiley-VCH
Verlag GmbH & Co. KGaA, Weinheim, Germany, 2007, pp. 67–143.
S. Kochius, A. O. Magnusson, F. Hollmann, J. Schrader, D.
Holtmann, Appl. Microbiol. Biotechnol. 2012, 93, 2251–2264.
M. Poizat, I. W. C. E. Arends, F. Hollmann, J. Mol. Catal. B Enzym.
2010, 63, 149–156.
Apparatus
J. Lutz, F. Hollmann, T. V. Ho, A. Schnyder, R. H. Fish, A. Schmid,
J. Organomet. Chem. 2004, 689, 4783–4790.
F. Hildebrand, S. Lütz, Chem. - A Eur. J. 2009, 15, 4998–5001.
S. Chardon-Noblat, S. Cosnier, A. Deronzier, N. Vlachopoulos, J.
Electroanal. Chem. 1993, 352, 213–228.
All electrochemical experiments were conducted with an Autolab
PGSTAT-12 potentiostat. The three-electrode configuration cell included
an Ag/AgCl reference electrode (3 M KCl), a stainless steel or a titanium
auxiliary electrode, and a piece of BP as working electrode. Glassy carbon
plates and a stainless steel clip were used for connecting the BP electrode.
The NADH regeneration was evaluated by measuring the absorbance of
the electrolyte between 300 and 400 nm with a Cary 60 Scan UV-Vis
spectrophotometer. X-Ray Photoelectron Spectroscopy (XPS) analyses
[14]
[15]
[16]
[17]
M. Beley, J.-P. Collin, J. Mol. Catal. 1993, 79, 133–140.
J. H. van Esch, M. A. M. Hoffmann, R. J. M. Nolte, J. Org. Chem.
1995, 60, 1599–1610.
were performed using
a KRATOS Axis Ultra X-ray photoelectron
spectrometer (Kratos Analytical, Manchester, UK) equipped with
a
[18]
E. Höfer, E. Steckhan, B. Ramos, W. R. Heineman, J. Electroanal.
Chem. 1996, 402, 115–122.
monochromated AlKα X-ray source (hν = 1486.6 eV) operated at 150 W.
The base pressure in the analytical chamber was 10-9 mbar during XPS
measurements. Wide scans were recorded using a pass energy of 160 eV
and narrow scans using a pass energy of 20 eV (instrumental resolution
better than 0.5 eV). Charge correction was carried out using the C(1s) core
line, setting adventitious carbon signal (H/C signal) to 284.6 eV.
[19]
[20]
E. Steckhan, in Top. Curr. Chem., 1994, pp. 83–111.
B. Tan, D. P. Hickey, R. D. Milton, F. Giroud, S. D. Minteer, J.
Electrochem. Soc. 2015, 162, H102–H107.
[21]
[22]
[23]
J. S. Lee, S. H. Lee, J. Kim, C. B. Park, J. Mater. Chem. A 2013, 1,
1040–1044.
L. Zhang, N. Vilà, G. W. Kohring, A. Walcarius, M. Etienne, ACS
Catal. 2017, 7, 4386–4394.
Acknowledgements
M. R. Axet, O. Dechy-Cabaret, J. Durand, M. Gouygou, P. Serp,
This article is protected by copyright. All rights reserved.

10.1002/cctc.201800681
ChemCatChem
ARTICLE
Coord. Chem. Rev. 2016, 308, 236–345.
200
0
A
B
5
0
[24]
[25]
[26]
[27]
X. Zhou, D. Micheroni, Z. Lin, C. Poon, Z. Li, W. Lin, ACS Appl.
Mater. Interfaces 2016, 8, 4192–4198.
a
X. Zhou, T. Zhang, C. W. Abney, Z. Li, W. Lin, ACS Appl. Mater.
Interfaces 2014, 6, 18475–18479.
-200
-400
-600
400
b
b
300
-5
c
200
I. Mustafa, I. Lopez, H. Younes, R. A. Susantyoko, R. A. Al-Rub, S.
Almheiri, Electrochim. Acta 2017, 230, 222–235.
C. W. Narváez Villarrubia, R. a. Rincón, V. K. Radhakrishnan, V.
Davis, P. Atanassov, ACS Appl. Mater. Interfaces 2011, 3, 2402–
2409.
100
d
e
a
0
0
1
2
3
-10
[NAD+]
f
-0.8
-0.4
0.0
0.4
-0.8
-0.6 -0.4
E vs. Ag/AgCl / V
E vs. Ag/AgCl / V
[28]
[29]
[30]
L. Hussein, G. Urban, M. Krüger, Phys. Chem. Chem. Phys. 2011,
13, 5831–9.
Figure 2. (A) (a) first and (b) second cyclic voltammogram for the reduction of
diazonium cations generated ‘in situ’ from 1mM 4-amino-2, 2’-bipyridine in 0.5
M HCl, as recorded on BP electrode at 100 mV s-1. (B) Cyclic voltammograms
S. Schauder, K. H. Schneider, F. Giffhorn, Microbiology 1995, 141,
1857–1863.
recorded at
a potential scan rate of 5 a
mV s-1 using [Cp*Rh(bpy)Cl]+
functionalized bucky paper electrode (BP-bpy-Rh) in 50 mM PBS buffer at pH
6.5 under nitrogen to the addition of NAD+ (a) 0 mM (b) 0.5 mM (c) 1 mM (d) 1.5
mM (e) 2 mM (f) 3 mM. The geometric surface area of the electrode was 1 cm2.
S. Baranton, D. Bélanger, J. Phys. Chem. B 2005, 109, 24401–
24410.
[31]
[32]
J. Agullo, M. Morin, D. Bélanger, in ECS Trans., 2011, pp. 19–26.
E. De La Llave, S. E. Herrera, C. Adam, L. P. Méndez De Leo, E. J.
Calvo, F. J. Williams, J. Chem. Phys. 2015, 143, DOI
10.1063/1.4935364.
A
[33]
[34]
E. Steckhan, S. Herrmann, R. Ruppert, E. Dietz, M. Frede,
Organometallics 1991, 10, 1568–1577.
A. Walcarius, R. Nasraoui, Z. Wang, F. Qu, V. Urbanova, M.
Etienne, M. Göllü, A. S. Demir, J. Gajdzik, R. Hempelmann,
Bioelectrochem. 2011, 82, 46–54.
406
404
402
400
398
396
[35]
[36]
F. Hildebrand, S. Lütz, Tetrahedron: Asymmetry 2007, 18, 1187–
1193.
Binding Energy / eV
Z. Wang, M. Etienne, G. W. Kohring, Y. Bon Saint Côme, A. Kuhn,
A. Walcarius, Y. Bon-Saint-Côme, A. Kuhn, A. Walcarius,
Electrochim. Acta 2011, 56, 9032–9040.
B
[37]
[38]
[39]
[40]
M. Zalas, M. Ceglowski, G. Schroeder, Chem. Pap. 2012, 66, 733–
740.
M. Sprecher, R. Breslow, O. Uziel, T. M. Link, Org. Prep. Proced.
Int. 1994, 26, 696–701.
406
404
402
400
398
396
L. Zhang, N. Vilà, A. Walcarius, M. Etienne, ChemElectroChem
2018, 1–11.
Binding energy / eV
I. Mazurenko, M. Etienne, G. W. Kohring, F. Lapicque, A. Walcarius,
Electrochim. Acta 2016, 199, 342–348.
C
Figures
318 316 314 312 310 308 306 304
Binding Energy / eV
C
Cl
Rh
Cl
Cl
Cl Rh-
Cp*
Rh
A
B
Cl
N
N
N
N
Figure 3. XPS characterization: N1s core level spectra for (A) BP-bpy and (B)
BP-bpy-Rh electrodes; (C) Rh3d core level spectrum for the BP-bpy-Rh
electrode.
N
N
N
N
+
NH
2
N2
BP-bpy
BP-bpy-Rh
Figure 1. Synthetic route followed for the functionalization of the bucky paper
with [Cp*Rh(bpy)Cl]+ by combination of diazonium salt electrografting and
complexation process.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201800681
ChemCatChem
ARTICLE
D-sorbitol
D-fructose
A
Glassy fiber
B
gel layer
-20
-40
-60
-80
NAD+
260 µm
NADH
NADH
b
Buckypaper
layer
NAD+
a
40
30
+
+
50 µm
0
1 2 3
Rh
H
Rh
H+
C / mM
Rh
Cl
N
N
N
N
N
N
2e-
0
10 20 30 40 50
t / min
Figure 4. (A) Configuration and electroenzymatic process involved in the
bioelectrochemical system with D-sorbitol dehydrogenase immobilized in a
glassy fiber layer (260 µm thick) and rhodium functionalized BP layer (50 µm
thick). (B) Amperometric response to additions of 0.5 mM D-fructose recorded
using a 1 cm2 electrode (a) in the presence of DSDH and (b) in the absence of
DSDH in the glassy fiber layer. The experiments were performed under nitrogen
at -0.72 V vs. Ag/AgCl in 50 mM PBS (pH 6.5) with 1 mM NADH in solution.
0
-10
-20
-30
-40
-50
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
c
B
A
a
b
0
20 40 60 80 100
0
20 40 60 80 100
t / h
t / h
Figure 5. (A) Electroenzymatic synthesis of D-sorbitol from D-fructose in 30 mL
solution containing 1 mM NADH (and 50 mM PBS, pH 6.5) with a combination
of a BP-bpy-Rh electrode and a glassy fiber layer containing immobilized DSDH
(2 cm  2 cm) at −0.72 V vs. Ag/AgCl under nitrogen. (B) Evolution with time of
the concentrations of (a) D-fructose, (b) D-sorbitol, and (c) the sum of D-fructose
and D-sorbitol, as determined by HPLC.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201800681
ChemCatChem
ARTICLE
Entry for the Table of Contents (Please choose one layout)
Layout 1:
ARTICLE
D-sorbitol
D-fructose
A bioelectrode for
Lin Zhang, Mathieu
electroenzymatic synthesis
was prepared, combining a
layer for NADH regeneration
Etienne,* Neus Vilà, Thi
Xuan Huong Le, Gert-
Wieland Kohring, Alain
Walcarius
Glassy fiber
gel layer
NAD+
NAD+
260 µm
NADH
NADH
with immobilized
Buckypaper
layer
[Cp*Rh(bpy)Cl]+ complex and
a renewable layer for
Page No. – Page No.
enzymatic substrate reduction
with immobilized
dehydrogenase.
+
+
Title
50 µm
Rh
Rh
H+
H
Rh
Cl
N
N
N
N
N
N
2e-
Layout 2:
ARTICLE
Author(s), Corresponding Author(s)*
((Insert TOC Graphic here))
Page No. – Page No.
Title
Text for Table of Contents
This article is protected by copyright. All rights reserved.