Bioinspired Electro-Organocatalytic Material Efficient for Hydrogen Production

Article abstract of DOI:10.1002/chem.201705655

Commercial carbon fibers can be used as electrodes with high conductive surfaces in reduced devices. Oxidative treatment of such electrodes results in a chemically robust material with high catalytic activity for electrochemical proton reduction, enabling the measurement of quantitative faradaic yields (>95 %) and high current densities. Combination of experiments and DFT calculations reveals that the presence of carboxylic groups triggers such electrocatalytic activity in a bioinspired manner. Analogously to the known Hantzsch esters, the oxidized carbon fiber material is able to transfer hydrides, which can react with protons, generating H₂, or with organic substrates resulting in their hydrogenation. A plausible mechanism is proposed based on DFT calculations on model systems.

Full text of DOI:10.1002/chem.201705655

A Journal of
Accepted Article
Title: Bioinspired Electro-Organocatalytic Material Efficient for
Hydrogen Production
Authors: Ruben Mas-Balleste, Octavio Gonzalez-del Moral, Arnau
Call, Federico Franco, Alicia Moya, Jose Antonio Nieto, María
Frias, Jose L. G. Fierro, Miquel Costas, Julio Lloret-Fillol, and
José Alemán
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: Chem. Eur. J. 10.1002/chem.201705655
Link to VoR: http://dx.doi.org/10.1002/chem.201705655
Supported by

10.1002/chem.201705655
Chemistry - A European Journal
FULL PAPER
Bioinspired Electro-Organocatalytic Material Efficient for
Hydrogen Production
Octavio González-del Moral,[a,b] Arnau Call,[c] Federico Franco,[d] Alicia Moya,[a] Jose Antonio Nieto-
Rodríguez,[a] María Frias,[e] Jose L. G. Fierro,[g] Miquel Costas,[c] Julio Lloret-Fillol,[d,h] José Alemán,[e,f]
and Rubén Mas-Ballesté*[a,f]
Abstract: Commercial carbon fibers can be used as electrodes with
high conductive surfaces in reduced devices. Oxidative treatment of
such electrodes results in a chemically robust material with high
catalytic activity for electrochemical proton reduction, enabling the
measurement of quantitative faradaic yields (>95%) and high current
densities. Combination of experiments and theoretic calculations
reveals that the presence of carboxylic groups triggers such electro-
catalytic activity in a bioinspired manner. Analogously to the known
Hantzsch esters, the oxidized carbon fiber material is able to transfer
hydrides, which can react with protons generating H2 or with organic
substrates resulting in their hydrogenation. A plausible mechanism is
proposed based on DFT calculations on model systems.
form), is a coenzyme that is involved in a large number of
biological processes (A and B, Figure 1).[1,2] NADPH provides one
of the best reducing biological agents for a large variety of bio-
reduction reactions (left, Figure 1), which nature uses in a catalytic
manner. Interestingly, inspired by NADPH, the asymmetric
transfer hydrogenation from Hantzsch esters (C) has appeared as
a green alternative for the reduction of C=C, C=N and C=O bonds
(middle, Figure 1).[3–10] In this case, the reductive form (C) is easily
oxidized to the pyridine derivative (C´) through a hydride-transfer
process. Some important drawbacks of these synthetic reagents,
such as their difficult recovery due to its challenging purification,
the need of super-stoichiometric amounts (>2 equiv.) and the high
cost, all difficult the implementation of these reagents in industry.
Consequently, it would be highly desirable to find a new
technology that can include this type of species into a material (D,
Figure 1) and thus, to overcome the problems associated with
Hantzsch ester derivatives.
Introduction
Nicotinamide adenine dinucleotide phosphate, which is
abbreviated as NADP+ (oxidized form) and NADPH (reduced
[a]
Dr. O. González-del Moral, Dr. A. Moya, I. J.A. Nieto-Rodríguez, Dr.
R. Mas-Ballesté
Department of Inorganic Chemistry (module 07)
Facultad de Ciencias, Universidad Autónoma de Madrid
28049 Madrid, Spain.
E-mail: ruben.mas@uam.es
[b]
[c]
Dr. O. González-del Moral
Department of Chemistry (module 13)
Facultad de Ciencias, Universidad Autónoma de Madrid
28049 Madrid, Spain.
Dr. Arnau Call, Dr. Miquel Costas
Departament de Química and Institute of Computational Chemistry
and Catalysis (IQCC)
Universitat de Girona,
Campus Montilivi, 17071 Girona, Catalonia, Spain.
Dr. Federico Franco, Dr. Julio Lloret-Fillol
Institute of Chemical Research of Catalonia (ICIQ),
The Barcelona Institute of Science and Technology, 43007
Tarragona, Catalonia, Spain.
I. María Frias, Dr. José Alemán
Department of Organic Chemistry (module 01)
Universidad Autónoma de Madrid,
[d]
[e]
[f]
28049 Madrid, Spain.
Figure 1. Inspiration and initial goals of this work.
Dr. José Alemán, Dr. R. Mas-Ballesté
Institute for Advanced Research in Chemical Sciences (IAdChem),
Universidad Autónoma de Madrid,
28049 Madrid, Spain.
Dr. Jose L. G. Fierro
One of the processes that could be facilitated by a material
including structures acting as hydride-transfer mediators is the
hydrogen evolution reaction (HER). In this context, hydrogen
evolution reaction is a main target in modern chemistry because
it can represent a change in the energy production paradigm. In
fact, hydrogen is intended to be the abundant, renewable and
zero-emission fuel that has been tapped to replace today’s fossil
fuels.[11] Water electrolysis is one of the simplest ways to produce
[g]
[h]
Instituto de Catálisis y Petroleoquímica, CSIC
Marie Curie, 2, Cantoblanco, 28049 Madrid, Spain.
Dr. Julio Lloret-Fillol
Catalan Institution for Research and Advanced Studies (ICREA)
Passeig Lluïs Companys, 23, 08010 Barcelona, Spain
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201705655
Chemistry - A European Journal
FULL PAPER
hydrogen of high purity and thus, it has attracted considerable Results and Discussion
attention.[12] This process requires new electrocatalysts to
improve the production rates and decrease overpotentials. HER
electrocatalysts can be homogeneous or heterogeneous.[13,14] For
homogeneous molecular catalysts, application requires their
stable grafting onto electrode materials. Such a problem could be
solved by using heterogeneous systems, which are generally
more stable and easier to implement into devices. Pt-group
metals are excellent heterogeneous HER catalysts, but their wide
use is limited by its terrestrial scarcity and its high price.[15,16] Thus,
investigations in precious-metal-free catalysis[17] or even metal-
free catalysis[18] are being developed with the aim to obtain acid-
stable (since conditions to generate H2 usually require acidic
conditions) and highly active catalysts for the HER based on
inexpensive earth-abundant materials (such as carbon-based
materials).
Preparation and characterization of the Catalytic Material
In this work, carbon fibers are considered as a good material for
the design of electrocatalysts because they allow the fabrication
of high surface area and covalently stable functionalized
electrodes. This approach avoids drawbacks such as the
inconvenience of working with reduced amounts of material at a
nanometric scale.[18,38–40] According to the manufacturer, the
carbon fiber material consists in bundles of 3000 fibers[34] where
each fiber has an average diameter of 8 μm. We designed
electrodes in a brush disposition (main details on the electrode
design are provided in Figure S1 of supplementary information).
In a typical configuration, we used a brush composed by 42000
fibers with an exposed length of 2 cm. This design implies that, if
aggregation does not take place at all and every fiber behaves
independently from each other, the theoretical maximum
available surface area of the carbon fiber is approximately 2100
cm2 (for evaluation of catalytic active surface area see below).
This is a remarkable feature for a reduced size device. Moreover,
the high conductivity of such carbon fibers effectively enables
their use as efficient electrodes. A main advantage of this new
design is the extremely low prize of the carbon material. However,
not only the prize, but also the special chemical features of this
material make carbon fibers very attractive.[39,41]
Carbon-based materials can be produced on a large scale with
lower manufacturing cost compared to metal-based materials.[19–
21]
In addition, carbonaceous materials often offer synergy
between the material and attached active centres due to the
electrical conductivity and the electron acceptor/donor properties
of graphitic structure such as graphene or carbon nanotubes.[22]
Therefore, carbon-based materials are very attractive alternatives
to conventional heterogeneous electrocatalysts used for
hydrogen evolution and other electrochemical reactions.[23,24]
They are attractive not only from the economic point of view, but
also from a scientific side. In fact, while HER pathways catalyzed
by metals have been widely studied,[25,26] much less is known
about the catalytic mechanism for H2 formation in metal-free
materials. Following these ideas, in recent work, carbon
nanotubes and other modified graphene/graphitic materials have
been explored as electrocatalysts in HER.[22,27–32] Some of these
contributions consisted on deposition of carbon nanotubes, or
functionalized graphene on small surface area glassy carbon
electrodes.[33] Good overpotential and current intensity values
have been reported. However, in some cases, the results were
limited by common problems, such as adhesion of deposited
phases into the electrode surface or the small area of such
electrodes, which compromised absolute yields of produced
hydrogen. In general, the price, the manipulation of these carbon
nanostructures, the deposition onto a substrate material (e.g.
glass), and the small scale of these nanostructures limit the
production of hydrogen. Therefore, it would be very convenient to
find another appropriate carbon-based structure for large scale
electrocatalytic HER.
Inspired by these findings, we propose in this report a cheap
and easy-to-handle alternative that consists in the use of
commercially available carbon fibers from pyrolysis of
polyacrylonitrile (PAN).[34] This material has a well-established
structure,[35] which includes pyridyl fragments (right and bottom of
Figure 1), and can be oxidized to form acid derivatives.[36,37] If
such carboxylic groups are in meta position in the pyridine ring
the resulting structure is a nicotinic fragment. In that case, it can
act like NADP+ (oxidized form) and NADPH (reduced form) in
electrochemical systems, this is, as hydride-transfer agents in
HER processes. Here we report that this material can be easily
handled and oxidized to be used for HER in large quantities.
Moreover, a novel plausible mechanism (via hydride-transfer) has
been proposed based on the combination of experimental and
theoretical observations.
Oxidative procedures are expected to affect only the surface of
the material. Accordingly, although superficial properties are
changed (interaction with polymeric matrices or, as show below,
electrocatalytic behavior), bulk features remain unaltered
(mechanical
and
electrical
properties).
Therefore,
characterization of the chemical modification resulting from
oxidative procedure should be performed using a technique
specifically sensitive to surface chemical features. Thus,
comparative spectroscopic study of pristine and oxidized fibers
was performed by means of XPS (Figure 2). Comparison of XPS
spectra in the C1s region between pristine carbon fibers and
Highly Oriented Pyrolytic Graphite (HOPG, top, Figure 2)
indicated that carbon fibers are a less structured material, as can
be deduced from the increased broadness of C-C signal (284-285
eV) and the lack of signal in the region characteristic of stacked
materials with large π-π interactions (291-293 eV). An increase of
the C-O (286.2 eV) and C=O (287.7 eV) signals are also observed
in the carbon fiber, especially in the oxidized one. Although these
signals can be due to a certain extent to adhesion of adventitious
organic molecules, they are mostly indicative of the degree of
chemical functionalization. Such structural imperfections and the
polarization of the material by the pyridine moieties facilitates a
chemical functionalization by oxidizing over H2SO4/H2O2 (1:1).
Such reactivity has been previously reported for modification of
carbon fibers to increase their interaction with polymeric
matrices.[42–44]
Although harsh, such conditions are known to be innocuous to
graphite, which needs harsher conditions to be transformed to
graphene oxide.[45] After the oxidation process, the electric
conductivity was not lost as measured with a macroscopic
multimeter. This is a good indication that the bulk properties of
fibers are not altered. Accordingly, the morphology of the carbon
fiber has not been considerably modified during the oxidation
treatment (Figure S2). However, an alteration of the chemical
This article is protected by copyright. All rights reserved.

10.1002/chem.201705655
Chemistry - A European Journal
FULL PAPER
composition of carbon fibers is determined by XPS. This analysis
revealed the presence of highly oxidized carbon atoms (O-C=O)
at 289.1 eV, which were assigned to carboxylic groups resulting
from the oxidative process. On average, 5% of the carbon atoms
in the surface correspond to carboxylic groups. Therefore,
considering that the maximum number of catalytic centers is close
to the number of carboxylic groups (vide infra), the approximate
active area can be, at most, 5% of the available electrode surface.
In order to discard the possible role of adventitious metal centers
in the catalytic activity (vide infra), we measured a XPS survey
spectra, which does not show the presence of metal atoms
(Figure S10). In addition, presence of nitrogen atoms from pristine
material is also observed in oxidized fibres by means of the XPS
signal at binding energy of 401.9 eV, which can be assigned to
pyridinic nitrogen atoms. Furthermore, TXRF was performed to
the oxidized fiber (Figure S11). Only small content of Si and Cu
(below the ppm level) were found, which are not significant to
account for catalytic activity.
not result in significant changes in bulk properties and morphology,
it produces limited changes in the chemical features of the
surfaces consisting in generation of carboxylic groups, which can
result on transformation of pyridinic fragments into nicotinic
structures.
Electro-Catalytic Behaviour
The effect of the oxidative process on the electrochemical
behaviour of carbon fiber brushes was analyzed towards proton
reduction in CH3CN, using diluted solutions of CF3COOH (TFA)
as proton donor and 0.1 M N(Bu)4PF6 as inert electrolyte. This
electrolyte system allowed to evaluate the catalytic performance
of oxidized brushes in low proton content solutions, avoiding
saturation of signal in a wide range of electrochemical potentials.
In addition, experiments using controlled variable amounts of
proton source have also enabled to determine at what point all
catalytic sites arrive at the saturation point. Figure 3 shows the
typical linear sweep voltammetry measured for a 20 mM solution
of TFA using both pristine (grey line) and oxidized electrodes
(black line) between 0.0 and -0.7 V vs Ag/AgCl at a scan rate of
10 mV/sec. The observed cathodic current was significantly
increased as a consequence of the oxidative functionalization
process on the carbon fiber electrodes (black line). The measured
current intensity depends on the TFA concentration with a higher
proportionality factor at lower acid amounts, indicating the
saturation of all catalytic sites at higher TFA content (inset Figure
3). Overall, oxidized carbon fibers show a remarkable catalytic
activity towards proton reduction, while pristine electrodes do not
show such activity.
Figure 3. Typical response of pristine (grey line) and oxidized (black line)
electrodes in CH3CN (0.1 M N(Bu)4PF6) with TFA 20 mM. Inset: electrochemical
response of oxidized electrode with variable amounts of TFA (Scan rate:
10mV/sec).
Figure 2. C1s core-level spectra corresponding to pristine HOPG, pristine
carbon fibers and oxidized carbon fibers.
Formation of H2 was tested using an airtight cell and
quantitatively detected by Gas Chromatography (GC) (see Figure
4). Controlled-potential electrolysis was performed at E= -0.9 V vs.
Ag/AgCl in a 0.1 M TBAPF6/MeCN solution by using the oxidized
carbon fiber as the working electrode in the presence of 20 mM
TFA as the proton source. The electrocatalytic reaction provided
an average Faradaic efficiency >95% for H2 during 1 h of
Thus, overall, C1s and N1s XPS signals, together with TXRF
analytical data, SEM microscopy images and electric
characterization are all consistent with our previous assumption.
Therefore, experimental data suggest that while oxidation does
This article is protected by copyright. All rights reserved.

10.1002/chem.201705655
Chemistry - A European Journal
FULL PAPER
experiment. After that time, hydrogen production showed a slow
reduction. However, a freshly added aliquot of TFA (identical to
the starting one, corresponding to ca. 20 mM) restored the
electrocatalytic H2 production, providing analogous results in
terms of both faradaic efficiency and mmol of H2 produced for
another hour (Figure 4). This clearly demonstrates the capability
of the oxidized carbon fiber in the catalysis of H2 evolution reaction
under acidic conditions, while similar experiments using a pristine
electrode did not show any evidence of H2 formation.
Although data obtained in CH3CN has been very insightful, it is
neither the ideal working medium nor the standard conditions
used in comparable studies reported in the literature. Therefore,
as a benchmarking experiment and in order to compare our
system (pristine and oxidized electrodes) with previously reported
data, we performed the proton reduction experiments in 0.5 M
H2SO4 aqueous solution. Figure 5 shows the current densities
(considering active area = 2 cm2) measured for linear sweep at
potentials between -0.224 and -0.724
V
vs Ag/AgCl
(corresponding to the range between 0.0 and -0.5 V vs NHE) at a
scan rate of 10 mV/sec. In this experiment, the major activity of
oxidized electrode compared with pristine electrode is clear. An
onset overpotential of 250 mV with a Tafel slope of 178 mV dec-1
is observed. Although the value of the slope in the Tafel plot is
indicative of a process quite hampered kinetically, the current
density of 20 mA/cm2 was achieved at an overpotential lower than
400 mV. Such relative intensity value is comparable to other
electrocatalytic HER using modified carbon based electrodes
(see figure S4 in supplementary information).[24] However,
considering the large active area found, our absolute currents
overcome those reported on a lab scale device. In fact, the
hydrogen production at overpotential of 0.5 V was massive as can
be observed in the inset picture of Figure 5.
Figure 4. a) Electrocatalytic H2 evolution over time at E= -0.9 V vs. Ag/AgCl in
a 0.1 M TBAPF6/MeCN solution of carbon fiber in the presence of added 20 mM
TFA. The same volume of fresh TFA was injected after ca. 1 h of experiment.
b) Charge passed in the course of the electrocatalytic H+ reduction at E= -0.9 V
vs. Ag/AgCl in a 0.1 M TBAPF6/MeCN solution of carbon fiber upon addition of
two distinct aliquots of TFA (20 mM).
Figure 5. Typical response of pristine (grey line) and oxidized (black line)
electrodes in aqueous 0.5 M solution of H2SO4 (scan rate: 10 mV/sec). As insets
are shown the corresponding Tafel plot and a photograph of the massive
hydrogen production observed at 0.5 V vs NHE.
As explained above, considering the design applied to brush
electrodes, the maximum area available considering all the
microfibers independent from each other is 2100 cm2. However,
aggregation of fibers would reduce the available surface. The
effect of aggregation was evaluated by comparing the
performance of an oxidized electrode composed by few fibers with
a regular one. In particular, the experiment was carried out with
54 fibers and 47 fibers (see S.I., Figure S3). Assuming that with
these electrodes the effect of aggregation is negligible and
considering that the electrochemical signal is directly proportional
to the available electrode area, the available area in the complete
brush was determined to be approximately 40 cm2 (see details in
Supplementary information). Considering that only 5% of the
available area is in fact active (see XPS data on carboxylic
content), the overall active area would be approximately 2 cm2,
which is the electrode area used to calculate current densities in
the following experiment of this work.
The durability of these electrodes was tested by applying 1000
cycles from 0.0 to -1.0 V vs Ag/AgCl at 100 mV/sec on a 100 mM
solution of TFA in CH3CN (0.1 M of NBu4PF6). After such test, the
activity of electrodes remained unaltered (see S.I., Figure S5),
confirming the attractive potential of the oxidized carbon fiber
brush electrode for large-scale HER.
Experimental Mechanistic Evidences
All observations point out that the oxidation process is critical for
the electrocatalytic HER. In particular, according to XPS data, it
seems that the presence of carboxylic groups is associated to the
catalytic activity of the material. In our effort to provide further
evidences to this concept, we demonstrated that the effect of
oxidation in catalytic activity could be reversed when oxidized
electrodes were reduced by treating them with LiAlH4 (see Figure
This article is protected by copyright. All rights reserved.

10.1002/chem.201705655
Chemistry - A European Journal
FULL PAPER
6 and more details in Scheme S1 and Figure S6 in supplementary
information).
groups have been previously proposed as the catalytic centers for
Hydrogen Evolving Reaction[23] in oxidized carbon nanotubes.
Nevertheless, up to now, the role of carboxylic groups has not
been elucidated. To test the direct involvement of carboxylic
groups in electro-catalytic activity, we analyzed the reactivity of
oxidized carbon fiber electrodes containing carboxylic groups that
have been esterified by means of the reaction with
trimethylsilyldiazomethane or CH3I. Both reactions should result
in conversion of the carboxylic groups to their methyl esters (see
Supplementary information for details).[17,48] In any case, no
significant decrease on HER catalytic activity was observed.
(Figure S9) Consequently, although formation of carboxylic
groups in the oxidation process is crucial to promote catalytic
activity, such activity remains identical regardless of the fact that
the carboxylic groups (-COOH) were converted into esters (–
COOR), which could seem to indicate that carboxylic groups are
necessary but not directly involved in the catalytic process.
In order to discard some other functional groups as possible
catalytic centers, we performed selective derivatizations that
should affect specifically hydroxyl or carbonyl (aldehydes or
ketones) moieties (Figure 6). Due to the difficulty to follow the
possible reactions considering the C1s XPS, we introduced
functionalities with heteroatoms, and considered presence level
of such heteroatoms as an evidence of the effective
transformation of the material (see Table S1-S5 for more details).
In particular, reaction with trifluoroacetic anhydride is expected to
quantitatively react with hydroxyl groups since it has been widely
reported.[18,46] Effectively, the relative atomic abundance found for
C, O and F were 83.4%, 11.0% and 5.6%, respectively. Therefore,
considering the stoichiometry of the reactions from hydroxyl
groups to trifluoroacetate esters, approximately 33% of the
oxygen atoms present in the sample reacted in this
functionalization process (right-bottom, Figure 6). The activity of
the oxidized electrode before and after reaction with trifluoroacetic
anhydride was evaluated by linear sweep voltammetry, showing
that such treatment did not alter the activity of oxidized electrode.
(Figure S7). Therefore, the null effect of the trifluoroacetylation
process suggested that hydroxyl groups were not responsible for
the catalytic activity of oxidized electrodes.
Mechanism Proposal and DFT Calculations
In order to find a qualitative hint on the mechanism of the HER
process, DFT calculations were carried out at the M06-2X/6-
311G** level.[22,49–52] A plausible scenario is shown in Scheme 1.
As an approximation, we modeled the oxidized carbon fiber as
shown in Scheme 1 for species (I). Graphitic surface was
modelled using a fragment composed by four fused aromatic rings
(one pyridine and three benzenes) as a compromise between
performance and computational cost.[53,54] We introduced a
carboxylic group at the meta position of the pyridinic nitrogen in
order to investigate the effect of oxidation, generating a similar
structure to the hydride transfer centers found in nature (such as
NADPH).
Figure 6. Derivatization reactions performed in the experimental mechanistic
study.
A similar strategy was followed to determine the possible role of
carbonyl groups. Reaction of dithioethane results in the
transformation of ketones or aldehydes to dithioacetals (left-
bottom, Figure 6).[47] The amount of S atoms found after this
reaction corresponds to the functionalization of only 5% of the O
atoms detected. Therefore, the amount of carbonyl groups in
oxidized electrodes is low compared with hydroxyl fragments
(33%). The remaining 60% of oxygen content corresponds to
carboxylic functionalities, deduced from the fact that 5% of the
carbon atoms on the surface are assigned to -COOH. As
expected, reaction with dithioethane had no effect on the catalytic
activity of this material (Figure S8).
The overall results seem to indicate that oxidation somehow
activated the material as electrocatalysts in HER, but the active
sites were neither hydroxyl nor carbonyl groups. Carboxylic
Scheme 1. Mechanistic proposal calculated for hydrogen evolution.
This article is protected by copyright. All rights reserved.

10.1002/chem.201705655
Chemistry - A European Journal
FULL PAPER
The overall process (see Figure 7) should start with the injection
of electrons into the graphitic system. It is known that the value
for the work function of HOPG is about 4.8 eV,[55] which is
expected to be close to the value of the carbon fiber material.
Stabilization of injected electrons through the graphitic π-electron
density is the key factor that allows carbon based oxidized
materials to act efficiently as an electric conductor and as an
electro-catalyst. Otherwise, in simple carboxylic molecular
systems the first reduction step would be energetically prohibitive.
However, considering the reduced size of the structure used in
our calculations to simulate the graphenic region, the stabilization
energy after injection of two electrons should be underestimated.
Indeed, injection of two electrons results in a stabilization of 3.13
eV.
After injection of two electrons in our molecular model, the
negative charge increases the basicity of the pyridinic nitrogen
atom. Therefore, calculations suggest that, after two electron
injection, the encounter of a TFA molecule with (I) results in a
barrier-less protonation process. Consequently, the route from (I)
to (II) could be described as a proton-coupled electron transfer. In
addition, injection of two electrons also enhances the basicity of
the carbon in para position to the pyridinic nitrogen in species (II).
Indeed, protonation of (II) proceeds via an activation energy of 4.5
kcal/mol (TS1) and results in an extra stabilization of the system
by approximately 12 kcal/mol in compound (III).
Figure 7. Energetic profile corresponding to the mechanism shown in scheme
1. Structures of TS1 and TS2 are shown.
Furthermore, the notion that oxidation is needed in order to
favor the catalytic activity is reinforced by the calculations
performed in the system where the carboxylic group is substituted
by a hydrogen atom (mechanism b in supplementary information).
In this case, intermediate species show higher energies than their
oxidized counterparts, which is due to the extra conjugation added
to the system by the carboxylic groups. In addition, the possible
role of carboxylic groups as catalytic centers have also been
According to the calculations carried out in our system, the
intermediate (III) should exchange a trifluoroacetate molecule by
trifluoroacetic to generate (III´). Such exchange is slightly
exergonic by around 3 kcal/mol. Species (III´) evolves towards
hydride scission assisted by an acid molecule, resulting in the
formation of a hydrogen molecule and trifluoroacetate fragment in
species (IV). Such species is almost iso-energetic with (IV´), only
0.4 kcal more stable than (IV) and therefore, in this system the
pyridine protonation is in equilibrium with the protonation of a
trifluoroacetate. Furthermore, in the most stable final structure
(IV’’) the trifluoroacetate fragment is stabilized by hydrogen
bonding with the carboxylic group of the nicotinic moiety.
Formation of (IV)/(IV’) proceeds through a significant kinetic
barrier (approximately 30 kcal/mol), which is similar to values
found for reactions mediated by the biomimetic Hantzsch ester.[56]
Owing to the fact that this is the only step with a significant kinetic
barrier, it is clearly the rate determining step of the overall process.
The finding of a relatively high barrier is in agreement with the
high Tafel slope experimentally measured that suggested a
kinetically hampered process. In addition, oversimplification of the
molecular system considered to simulate the material can
contribute to barrier overestimation. The studied mechanism
considers the assistance of a TFA molecule in the hydride
scission, decreasing the value of the energetic barrier by 17
kcal/mol with respect to the unassisted hydride formation (see
supplementary information). However, more complicated
possibilities where several TFA and/or water molecules that could
assist hydride formation and stabilize the transition state have not
been computed, yet they are plausible scenarios in the
experimental system. In conclusion, theoretical and experimental
results are consistent with a mechanism in which the key step
consists of the TFA-assisted formation of a hydride from species
(III) that is concertedly protonated to generate hydrogen.
considered (mechanism
c in supplementary information).
Interestingly, when a carboxylic group acts as catalytic center,
scission of hydride to form H2 by protonation with a TFA molecule
is kinetically prohibitive.
Overall, although we cannot rule out other possible
mechanisms that may explain the observed reactivity, all data
together are fully consistent with the proposal that oxidized carbon
fibers assist hydride transfer in a very close manner to how nature
does in pathways mediated by NADH or NADPH.
Reactivity of Carbon Fibers as Hydride Donor versus
Benzaldehyde as Organic Hydride Acceptor
As final evidence for the formation of a hydride donor species on
the surface of carbon fibers at negative potentials, we investigated
the reduction of benzaldehyde. This organic substrate typically
evolves towards formation of benzylic alcohol by a reduction
process initiated by a hydride transfer.[57]
As a first set of experiments, we performed the linear sweep
voltammetry measured for solutions containing different
concentrations of TFA (20 mM and 2 mM) in acetonitrile in the
presence or absence of an excess of benzaldehyde (160 mM). All
experiments were carried out using oxidized carbon fiber brushes
as working electrodes at potentials between 0.0 and -0.7 V vs
Ag/AgCl at a scan rate of 10 mV/sec (Figure 8). Interestingly, at a
higher concentration of TFA (20 mM), intensity is independent on
the presence of benzaldehyde in the reaction medium. However,
at a lower concentration of TFA (2 mM), the presence of
benzaldehyde results in an increase of the measured intensity,
suggesting the involvement of benzaldehyde in the electron
transfer process. These observations indicate a competition
between hydride transfer to the protons (from TFA) or to the
organic substrate (benzaldehyde).
This article is protected by copyright. All rights reserved.

10.1002/chem.201705655
Chemistry - A European Journal
FULL PAPER
amount of benzyl alcohol was detected (Figure S13). Therefore,
these data are consistent with our mechanistic proposal in which
a species able to transfer hydrides is generated at the correct
electrochemical potential. In addition, our experiments confirm the
notion that the oxidation of the carbon fiber brushes is necessary
to generate such active intermediate able to transfer hydrides to
either protons (hydrogen evolution) or organic substrates
(hydrogenation). Furthermore, the balance between these two
possible processes, can be modulated controlling the acidity of
proton source. Thus, the more acidic additive (TFA) promotes
hydrogen evolution, while the less acidic molecule (water) makes
possible hydrogenation of organic substrates.
Conclusions
The combination of good electrical conductivity and the existence
of chemical reactive sites, as a result of the presence of pyridinic
nitrogen atoms on the surface of commercial carbon fibers, allows
the design of new chemically modified electrodes with
predetermined properties. Indeed, the oxidative functionalization
of carbon fiber electrodes results in the formation of carboxylic
groups, which are responsible for the efficient electro-catalytic
generation of H2 with quantitative faradaic yields and high current
densities. The combination of experiments and calculations
suggests that oxidized carbon fibers act as electrocatalysts in a
bioinspired manner. The collected data led to the proposal that
some of the carboxylic groups generated from oxidation process
can be located in meta position versus the pyridinic nitrogen atom
generating a nicotinic fragment. Such fragment assists the
formation and scission of hydride that can easily evolve to
molecular hydrogen through protonation. Moreover, hydride
transfer mediated by oxidized carbon fibers can also be
performed towards the reduction of organic substrates such as
benzaldehyde.
Figure 8. Linear voltammetry measured using oxidized carbon fiber electrodes
in the absence (black) and in the presence (grey) of 160 mM benzaldehyde in
CH3CN (0.1 M N(Bu)4PF6) with TFA 2 mM (a) or 20 mM (b).
Overall, this work reports a very cheap carbon-based material
alternative, which can be used as a modified electrode by means
of easy manipulations, that avoids coverage and/or deposition of
materials into the electrodes, allowing the design of a reduced
device with a high effective area able to produce H2 in large
amounts.
Experimental Section
Scheme 2. Electrolytic reduction of benzaldehyde using oxidized carbon fiber
brush electrode as working electrode in CH3CN (0.2 M NaClO4) with the
presence of H2O (1 M) and benzaldehyde 16 mM.
Materials: All reagents, solvents and materials were purchased from
commercial sources and used without further purification. Specifically,
fabric of carbon fiber (Twill 2x2 3K weight 200gr/m2 width 1200 mm, Model
HA2301) was purchased from ClipCarbono.
Considering that hydride transfer to proton from TFA (i.e.
Electrode preparation: the commercial carbon fiber is turned 7 times
around a 3 cm plastic strip to make a 6 cm long bunch of fibers. Then, it is
joined to a copper wire and finally everything is tight with a Teflon tape.
The electrodes are hand-made, but we try to make then as
homogeneously as we could keep the same design. The ready-made
electrodes are treated 30 minutes into commercial sulfuric acid (98%) at
room temperature with stirring and later they are put into a mixture 1:1 of
H2SO4/H2O2. The H2O2 should be fresh and the mixture with H2SO4 has to
be prepared few minutes before use. If the mixture is not ready-made the
oxidative potential of the reactive mixture it will not be enough to oxidize
the carbon fiber. Later, the electrodes are washed and sonicated in distilled
water to remove all the acid traces among fibers. It requires several
hydrogen evolution) is
a
competitive reaction versus
hydrogenation of benzaldehyde, we considered water as an
alternative as proton source necessary to generate the
corresponding hydride. Thus, we tested the oxidized brush
electrode in a chrono-amperometry in acetonitrile at -0.5V vs
Ag/AgCl in a system containing water (1 M) and benzaldehyde
(16 mM) and NaClO4 0.2M as inert electrolyte. After 12 hours, the
benzaldehyde has been completely reduced to benzyl alcohol
(Figure S12). In contrast, under the same experimental conditions,
after 24 hours, using pristine carbon fiber as electrode, any
This article is protected by copyright. All rights reserved.

10.1002/chem.201705655
Chemistry - A European Journal
FULL PAPER
washing cycles and once pH of the water keeps almost constant, the
electrodes are sonicated 10-15 minutes in isopropanol or ethanol and they
are dried with a heat pistol.
Acknowledgements
This work was supported by Spanish Ministerio de Economia y
Competitividad (MINECO) (CTQ2016-76061-P and CTQ2015-
64561-R) European Research Council (StG 239910), and
Generalitat de Catalunya (2014 SGR 862 and ICREA Academia
award to MC an J. LL-F.). J.Ll.-F. thanks the CELLEX foundation
for the starting career program for financial support. Financial
support from the MINECO through the “Maria de Maeztu”
Program for Units of Excellence in R&D (MDM-2014-0377) is also
acknowledged. We are very grateful to Dr. Sergio Diaz-Tendero
for many fruitful discussions. We acknowledge the generous
Photoelectron spectroscopy (XPS): Spectra were obtained with a VG
Escalab 200R spectrometer equipped with a hemispherical electron
analyzer (pass energy of 50 eV) and a Mg Kα (hν = 1254.6 eV, 1 eV =
1.6302 x 10-19 J) X-ray source, powered at 120 W. The kinetic energies of
photoelectrons were measured using a hemispherical electron analyzer
working in the constant pass energy mode. The background pressure in
the analysis chamber was kept below 6 x 10-9 mbar during data acquisition.
The XPS data signals were taken in increments of 0.1 eV with dwell times
of 50 ms. Binding energies were calibrated relative to the C 1s peak at
284.8 eV. High-resolution spectra envelopes were obtained by curve fitting
synthetic peak components using the software XPS peak. The raw data
were used with no preliminary smoothing. Symmetric Gaussian-Lorentzian
product functions were used to approximate the line shapes of the fitting
components. Morphological characterizations of the carbon fiber was
carried out using a scanning electron microscope (Philips XL30 S-FEG).
Measurements of the fiber diameter were done using ImageJ software.
Identification of the chemical elements that contains the carbon fiber have
been measured in a Total Reflection X-Ray Fluorescence (TXRF) 8030c-
FEI Spectrometer. A Mo Kα X-ray source was used as energy source at
50kV and 0.6mA.
Electrochemistry: Electrochemical experiments were performed under
an argon atmosphere at room temperature in CH3CN o water solutions.
Tetrabutylammonium hexafluorophosphate (Bu4NPF6) or sodium
perchlorate (NaClO4) were used as supporting electrolytes.
Measurements were carried out by using an Ivium CompaqStat
potentiostat interfaced with a computer. A standard three-electrode
electrochemical cell was used. Potentials were referred to an Ag/AgCl,
Et4NBr 0.4 M reference electrode in ethylene glycol, and measured
potentials were calibrated using an internal Fc/Fc+ standard. The working
electrode was a carbon fiber brush. The auxiliary electrode was a pristine
carbon fiber brush.
allocation of computer time at the Centro de Computacion
́
Cientifica at the Universidad Autonoma de Madrid (CCC-UAM).
́
́
Keywords: electrocatalysis • hydrogen evolution • hydride
transfer • bioinspired materials
References
[1]
Biol. Cell, Garland, New York & London, 3rd Edn., 2002.
[2] J. M. Berg, J. L. Tymoczko, L. Stryer, Biochemistry, W. H. Freeman, New
York, 5th Edn, 2002.
B. Alberts, D. Bray, J. Lewis, M. Raff, K. Roberts, J. D. Watson, in Mol.
[3]
[4]
[5]
[6]
H. Adolfsson, Angew. Chem. Int. Ed. 2005, 44, 3340–3342.
J. Rosen, Chemtracts 2005, 18, 65–71.
S.-L. You, Chem. – Asian J. 2007, 2, 820–827.
S. G. Ouellet, A. M. Walji, D. W. C. Macmillan, Acc. Chem. Res. 2007,
40, 1327–1339.
[7]
[8]
[9]
S. J. Connon, Org Biomol Chem 2007, 5, 3407–3417.
Gas Chromatography: The evolved amount of H2 during the
electrocatalytic reaction was monitored by gas chromatography,
C. Wang, X. Wu, J. Xiao, Chem. – Asian J. 2008, 3, 1750–1770.
M. Rueping, E. Sugiono, F. R. Schoepke, Synlett 2010, 2010, 852–865.
approximately each
4 minutes by injecting samples of the intern
[10] M. Rueping, J. Dufour, F. R. Schoepke, Green Chem 2011, 13, 1084–
1105.
atmosphere to an Agilent 490 micro gas chromatograph equipped with a
thermal conductivity detector and a 1/8” Molesieve 100/120 column,
calibrated with different H2/He/CO/CH4 mixtures of known composition.
For gas chromatography quantification of alcohol benzaldehyde, 0,1 mL of
the solution is mixed with 0,2 mL of methyl-naphthalene as standard and
0,7 mL chloroform. A calibrate of benzyl alcohol was carried out in the
range of (1·10-6-1·10-5 moles) using the same amount of the standard,
enabling the quantification of the benzyl alcohol amount. The analysis was
performed in an Agilent 7820A GC system equipped with a mass
spectroscopy detector (5977B MSD).
Nuclear Magnetic Resonance: Quantification of benzylic alcohol
generated was analyzed during the chrono-amperommetry experiment by
extracting samples of 1 mL of reaction mixture. CH3CN solvent was
removed by vacuum. Then, 1 mL of chloroform was added, which dissolve
the organic compounds but not the inorganic salt. Finally, the solution is
filtrated (0,45 µm pore size) and analyzed by 1H-NMR and gas
chromatography. Spectra were acquired on a Bruker 300 spectrometer
running at 300 MHz for 1H. Chemical shifts (δ) are reported in ppm relative
to residual solvent signal (CDCl3: 7.26 ppm).
[11] D. G. Nocera, Inorg. Chem. 2009, 48, 10001–10017.
[12] M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A.
Santori, N. S. Lewis, Chem. Rev. 2010, 110, 6446–6473.
[13] M. Zeng, Y. Li, J Mater Chem A 2015, 3, 14942–14962.
[14] Y. Zheng, Y. Jiao, M. Jaroniec, S. Z. Qiao, Angew. Chem. Int. Ed. 2015,
54, 52–65.
[15] W. Sheng, Z. Zhuang, M. Gao, J. Zheng, J. G. Chen, Y. Yan, Nat
Commun 2015, 6, 5848.
[16] Y. Shao-Horn, H. A. Gasteiger, W. Sheng, J Electrochem Soc 2010, 157,
B1529–B1536.
[17] X. Zou, Y. Zhang, Chem Soc Rev 2015, 44, 5148–5180.
[18] Y. Hou, B. L. Abrams, P. C. K. Vesborg, M. E. Björketun, K. Herbst, L.
Bech, A. M. Setti, C. D. Damsgaard, T. Pedersen, O. Hansen, et al., Nat Mater
2011, 10, 434–438.
[19] K. S. Novoselov, V. I. Fal[prime]ko, L. Colombo, P. R. Gellert, M. G.
Schwab, K. Kim, Nature 2012, 490, 192–200.
[20] K. R. Paton, E. Varrla, C. Backes, R. J. Smith, U. Khan, A. O’Neill, C.
Boland, M. Lotya, O. M. Istrate, P. King, et al., Nat Mater 2014, 13, 624–630.
[21] M.-R. Azani, A. Hassanpour, V. Carcelén, C. Gibaja, D. Granados, R.
Mas-Ballesté, F. Zamora, Appl. Mater. Today 2016, 2, 17–23.
[22] S. Navalón, J. R. Herance, M. Álvaro, H. García, Chem. – Eur. J. 2017,
Accepted Manuscript.
Scanning
Electron
Microscopy
(SEM):
Morphological
characterizations of the carbon fiber was carried out using a scanning
electron microscope (Philips XL30 S-FEG). Measurements of the fiber
diameter were done using ImageJ software.
Total Reflection X-Ray Fluorescence (TXRF): Identification of the
chemical elements that contains the carbon fiber have been measured in
a TXRF 8030c-FEI Spectrometer. A Mo Kα X-ray source was used as
energy source at 50kV and 0.6mA.
[23] Y. Xu, M. Kraft, R. Xu, Chem Soc Rev 2016, 45, 3039–3052.
[24] W. Zhou, J. Jia, J. Lu, L. Yang, D. Hou, G. Li, S. Chen, Nano Energy
2016, 28, 29–43.
[25] J. Hu, C. Zhang, X. Meng, H. Lin, C. Hu, X. Long, S. Yang, J Mater Chem
A 2017, 5, 5995–6012.
[26] V. R. Stamenkovic, D. Strmcnik, P. P. Lopes, N. M. Markovic, Nat Mater
2017, 16, 57–69.
This article is protected by copyright. All rights reserved.

10.1002/chem.201705655
Chemistry - A European Journal
FULL PAPER
[27] W. Cui, Q. Liu, N. Cheng, A. M. Asiri, X. Sun, Chem Commun 2014, 50,
9340–9342.
[28] Q. Liu, J. Tian, W. Cui, P. Jiang, N. Cheng, A. M. Asiri, X. Sun, Angew.
Chem. Int. Ed. 2014, 53, 6710–6714.
[29] R. K. Das, Y. Wang, S. V. Vasilyeva, E. Donoghue, I. Pucher, G.
Kamenov, H.-P. Cheng, A. G. Rinzler, ACS Nano 2014, 8, 8447–8456.
[30] J. K. Kim, G. D. Park, J. H. Kim, S.-K. Park, Y. C. Kang, Small 2017, 13,
1700068–n/a.
[31] A. Xie, N. Xuan, K. Ba, Z. Sun, ACS Appl. Mater. Interfaces 2017, 9,
4643–4648.
[32] G. Valenti, A. Boni, M. Melchionna, M. Cargnello, L. Nasi, G. Bertoni, R.
J. Gorte, M. Marcaccio, S. Rapino, M. Bonchio, et al., Nat. Commun. 2016, 7,
13549.
[33] J. D. Blakemore, A. Gupta, J. J. Warren, B. S. Brunschwig, H. B. Gray,
J. Am. Chem. Soc. 2013, 135, 18288–18291.
[34] Commercial carbon fiber, http://www.clipcarbono.com/es/home/266-
tejido-de-fibra-de-carbono-sarga-2x2-3k-peso-200gr-m2-ancho-1200-mm.html
n.d.
[35] M. S. A. Rahaman, A. F. Ismail, A. Mustafa, Polym. Degrad. Stab. 2007,
92, 1421–1432.
[36] L.-G. Tang, J. L. Kardos, Polym. Compos. 1997, 18, 100–113.
[37] E. Pamula, P. G. Rouxhet, Carbon 2003, 41, 1905–1915.
[38] D. D. Edie, Carbon 1998, 36, 345–362.
[39] X. Huang, Materials 2009, 2, 2369–2403.
[40] R. Sengupta, M. Bhattacharya, S. Bandyopadhyay, A. K. Bhowmick,
Spec. Issue Conduct. Polym. 2011, 36, 638–670.
[41] Y. Liu, S. Kumar, Polym. Rev. 2012, 52, 234–258.
[42] X. Zhang, L. Liu, M. Li, Y. Chang, L. Shang, J. Dong, L. Xiao, Y. Ao, RSC
Adv 2016, 6, 29428–29436.
[43] B. Gao, W. Du, R. Zhang, F. Lu, J. Zhang, Mater. Lett. 2016, 179, 16–
19.
[44] G. Zhang, S. Sun, D. Yang, J.-P. Dodelet, E. Sacher, Carbon 2008, 46,
196–205.
[45] W. S. Hummers, R. E. Offeman, J. Am. Chem. Soc. 1958, 80, 1339–
1339.
[46] A. B. Charette, Handbook of Reagents for Organic Synthesis: Reagents
for Heteroarene Functionalization, John Wiley & Son, 2005.
[47] J. R. Hanson, in Funct. Group Chem., Royal Society Of Chemistry, 2001.
[48] F. A. Carey, Sundberg, in Adv. Org. Chem., Springer, 5th Edn., 2008.
[49] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, et al.,
Gaussian 09 Revision E.01, Wallingford CT: Gaussian, Inc., 2009.
[50] T. Ziegler, Chem. Rev. 1991, 91, 651–667.
[51] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2007, 120, 215–241.
[52] Y. Zhao, D. G. Truhlar, Acc. Chem. Res. 2008, 41, 157–167.
[53] W. R. Fawcett, Langmuir 2008, 24, 9868–9875.
[54] A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B 2009, 113,
6378–6396.
[55] M. Shiraishi, M. Ata, Carbon 2001, 39, 1913–1917.
[56] L. Simón, J. M. Goodman, J. Am. Chem. Soc. 2008, 130, 8741–8747.
[57] A. Proto, R. Cucciniello, A. Genga, C. Capacchione, Catal. Commun.
2015, 68, 41–45.
This article is protected by copyright. All rights reserved.

10.1002/chem.201705655
Chemistry - A European Journal
FULL PAPER
FULL PAPER
Life was doing it. Biological systems contain molecular fragments able to transfer
hydrides in a catalytic manner. The same kind of fragments incorporated to a
conductive carbon-based material acts as electro-catalytic site for Hydrogen
evolution.
Octavio Gonzalez-del Moral, Arnau Call,
Federico Franco, Alicia Moya, Jose
Antonio Nieto-Rodríguez, María Frias,
Jose L. G. Fierro, Miquel Costas, Julio
Lloret-Fillol, José Alemán and Rubén
Mas-Ballesté*
Page No. – Page No.
Bioinspired Electro-Organocatalytic
Material Efficient for Hydrogen
Production
This article is protected by copyright. All rights reserved.