ARTICLES

PUBLISHED ONLINE: 5 AUGUST 2012 | DOI: 10.1038/NMAT3385

A Janus cobalt-based catalytic material for

electro-splitting of water
Saioa Cobo1 , Jonathan Heidkamp2 , Pierre-Andr Jacques1 , Jennifer Fize1 , Vincent Fourmond1 ,
Laure Guetaz3 , Bruno Jousselme4 , Valentina Ivanova5 , Holger Dau2 , Serge Palacin4 ,
Marc Fontecave1,6 and Vincent Artero1 *
The future of energy supply depends on innovative breakthroughs regarding the design of cheap, sustainable and efficient
systems for the conversion and storage of renewable energy sources. The production of hydrogen through water splitting
seems a promising and appealing solution. We found that a robust nanoparticulate electrocatalytic material, H2 CoCat , can
be electrochemically prepared from cobalt salts in a phosphate buffer. This material consists of metallic cobalt coated with
a cobalt-oxo/hydroxo-phosphate layer in contact with the electrolyte and mediates H2 evolution from neutral aqueous buffer
at modest overpotentials. Remarkably, it can be converted on anodic equilibration into the previously described amorphous
cobalt oxide film (O2 CoCat or CoPi) catalysing O2 evolution. The switch between the two catalytic forms is fully reversible
and corresponds to a local interconversion between two morphologies and compositions at the surface of the electrode. After
deposition, the noble-metal-free coating thus functions as a robust, bifunctional and switchable catalyst.

xploitation of abundant but intermittent and diluted renewable energy sources, such as sunlight or wind1 , requires efficient energy storage technologies. The cost-effective conversion of these energy sources into a chemical form via the production
of a fuel is one of the greatest challenges of the century. Hydrogen
is such a fuel which can be simply produced via water electrolysis,
when the electricity is derived from renewable energy sources
(photovoltaics, turbines and so on)2 . Promising devices, based on
proton exchange membrane (PEM) technology, rely on the use of
noble metal (Pt, Ir and so on)-based electrocatalysts but, owing
to the low abundance and consequent high cost of such precious
metals, the long-term viability of these systems is questionable.
Instead the transition to a hydrogen economy requires the design
and construction of efficient systems that use earth-abundant
elements3 . Although important reports have been made recently
regarding the use of cobalt411 , nickel12 and manganese8,13 oxide
materials as robust electrocatalysts for water oxidation (OER), few
such materials exist for hydrogen evolution1417 (HER). Innovative
breakthroughs based on cobalt compounds have appeared in the
past decade18 for electrocatalytic4,5,7,8 OER. We1921 and others2224
reported that a series of cobaloxime and diimine-dioxime20 compounds exhibit remarkable properties for proton reduction with
low overpotential requirements25 . However, practical utilization of
molecular catalysts requires their grafting with retention of the
catalytic activity onto an electrode material15 , which in the specific
case of Co catalysts has so far been prevented by synthetic issues18 .
We thus turned towards a more direct and smoother method
based on the reductive electrodeposition of a cobalt-based material,
because this methodology recently proved successful for the preparation of MoS2 -based16,26 or NiMo-based27,28 HER catalysts. We
here establish that reduction of Co(NO3 )2 6H2 O from an aqueous
phosphate buffer (KPi, 0.5 mol l1 , pH 7) at a F-doped SnO2

(FTO) electrode is a straightforward and highly convenient preparation of very stable electrocatalytic materials for H2 evolution.
Similar results are obtained if the water-soluble diimine dioxime
cobalt(iii) complex [Co(DO)(DOH)pnCl2 ] ((DOH)(DOH)pn =
0
N2 , N2 -propanediylbis(2,3-butandione 2-imine 3-oxime); Supplementary Fig. S1) or the cobaloxime [Co(dmgBF2 )2 (H2 O)2 ]
(dmgH2 = dimethylglyoxime) are used as cobalt precursors under
the same conditions.
Figure 1 reports the current densities obtained during a linear
sweep voltammetry experiment (plain black trace, 0.05 mV s1 ) of
a Co(NO3 )2 6H2 O solution (0.5 mmol l1 ) in KPi (0.5 mol l1 ,
pH 7) at a FTO electrode. A reductive process is observed with
onset at 0.9 V versus Ag/AgCl. Simultaneous chromatographic
monitoring of H2 production (dotted black trace) indicates that
no hydrogen is produced at this point. If the electrode potential is
switched to more negative values (below 0.95 V versus Ag/AgCl)
H2 is produced. To provide more insights into the reductive process
at work at the onset of the wave, we carried out an electrolysis
experiment at 0.9 V versus Ag/AgCl for 1 h (Q = 0.1 C cm2
geometric ),
which resulted in a grey coating of the electrode. The scanning
electronic micrograph (Fig. 2a) shows isolated nanoparticles with
an average 10 nm in diameter. Performing the same electrolysis
but at 1.0 V versus Ag/AgCl for 3 h yields a film of 2 mm
thickness made from larger particles (100 nm), as shown in Fig. 2b.
During this experiment the current density stabilizes to a value
of 2 mA cm2
geometric . Neither the use of a longer electrolysis time
nor a second electrolysis experiment in a new Co2+ solution
could increase the current density. The modified electrode was
then transferred to a Co-free KPi electrolyte and studied for its
electrocatalytic properties under the conditions described above.
Figure 1 (red traces) reports current density (plain) and specific rate
of H2 evolution (dotted) traces as a function of the electrochemical

1 Laboratoire de Chimie et Biologie des Mtaux (CEA/Universit Grenoble 1/CNRS), 17 rue des Martyrs, 38054 Grenoble cedex 09, France, 2 FB Physik, Free

University Berlin, Arnimallee 14, D-14195 Berlin, Germany, 3 Institut LITEN, CEA LITEN/DEHT/LCPEM, 17 rue des Martyrs, F-38054 Grenoble cedex 9,
France, 4 CEA, IRAMIS, SPCSI, Chemistry of Surfaces and Interfaces group, F-91191 Gif sur Yvette Cedex, France, 5 CEA-Leti, MINATEC Campus, 17 rue des
Martyrs, F-38054 Grenoble cedex 9, France, 6 Collge de France, 11 place Marcelin-Berthelot, 75231 Paris cedex 05, France. *e-mail: vincent.artero@cea.fr.
802

NATURE MATERIALS | VOL 11 | SEPTEMBER 2012 | www.nature.com/naturematerials

2012 Macmillan Publishers Limited. All rights reserved

NATURE MATERIALS DOI: 10.1038/NMAT3385

ARTICLES

E (V) versus Ag/AgCl

0.9

0.8

0.7

0.6
0
2

0.5

j (mA cm2
geometric)

4
1.0
1.5
2.0
2.5

1.0 0.9 0.8 0.7 0.6

0.0

0.1

0.5

10

0.2

1.0

12

0.3

1.5

14

3.0
0.4 0.3 0.2 0.1
3.5

0.4

0.0

0.3

0.2
E (V) versus RHE

0.1

2.0
0.0

16
18

2
Spec. H2 evolution rate (nmol s1 cmgeometric
)

0.0

1.0

0.0

Figure 1 | Linear voltammetry experiments. Black: current densities (solid

line) recorded at a FTO electrode (1 cm2 ) in phosphate buffer (KPi,
0.5 mol l1 , pH 7) containing Co(NO3 )2 6H2 O (0.5 mmol l1 ) at a low
scan rate (0.05 mV s1 ) with simultaneous GC monitoring of H2 evolution
(dotted line): N2 was continuously bubbled through the electrolyte at a
constant flow (5 ml min1 ) during the experiment and the concentration of
H2 in the output gas was determined every 2 min by gas chromatography.
This allowed the determination of the specific rate of H2 evolution, which is
reported in the right scale. With the right and left scales used,
superimposition of both current density and H2 evolution rate traces
indicates 100% faradaic yield. Red: current densities (solid line) and
specific rate of H2 evolution (dotted line) recorded at a FTO electrode
previously modified by controlled potential electrolysis for 3 h at 385 mV
versus RHE in the Co-containing solution described above and then
transferred into a cobalt-free buffer. Inset: blow-up of the low overpotential
region showing evolution of H2 with 50 mV onset overpotential.

potential values scaled versus the reversible hydrogen electrode

(RHE), from which overpotential values can be directly obtained.
Hydrogen evolution could be detected for overpotential values as
low as 50 mV (inset in Fig. 1) and overpotential values of 270 mV
and 385 mV are required to reach current density values of 0.5 and
2 mA cm2
geometric respectively. With such an electrode, the faradaic
yield is found to be quantitative within the experimental accuracy.
It should be noted that these overpotential values are much lower
than those reported with cobalt-based molecular catalysts assayed
under aqueous solutions. For instance, Co macrocycles generally
require overpotentials of 500 mV700 mV (ref. 18) and a recently
described pentadentate polypyridyl cobalt complex catalyses H2
evolution with onset of the catalytic current occurring for an
overpotential of 660 mV (ref. 29). Our catalytic material, hereafter
named H2 CoCat, seems to be stable as long as the electrode
is kept at a potential more negative than about -0.6 V versus
Ag/AgCl. When poised at a more positive potential, or left at
open-circuit potential, the catalytic film readily dissolves in the
electrolyte, yielding Co(ii) species. By ICP-MS titration of Co(ii)
ions in the electrolyte after full redissolution, we determined that
1.0 106 mol of cobalt are deposited per geometric cm2 , from
which we derive an H2 -evolution turnover frequency of 80 h1 per
cobalt centre at 385 mV overpotential.
Figure 3 also shows that H2 CoCat exhibits much higher
performances than a bulk cobalt foil. Tafel analysis (inset in Fig. 3)
gives a Tafel slope of 140 mV decade1 and exchange current density
of 105.5 A cm2
geometric . This value is significantly higher than that
found for metallic cobalt (106 A cm2
geometric ) or electrodeposited
MoS2 catalysts (106.5 A cm2
geometric ; ref. 26). NiMo-based materials
exhibit current densities between 106 and 104 A cm2
geometric
(ref. 28) and a recent NiMo nitride material was reported with
2
3.5
an exchange current density of 10
A cmgeometric (ref. 30). We

also compared H2 CoCat with a similar cobalt coating described

by Soto et al.31 but deposited from NH4 Cl solution (pH 5.6),
and with another nanoparticulate metallic Co coating reductively
obtained from a Co2+ -containing LiClO4 solution (pH 7; ref. 32).
Surface morphologies of the three samples were characterized
using atomic force microscopy (Supplementary Fig. S6) and root
mean squared roughness values of 87, 60 and 160 nm were
determined for H2 CoCat and the coatings deposited from NH4 Cl
and LiClO4 respectively. Cobalt quantification after oxidative
dissolution of the films gave surface concentrations of 1.0 and
2.9106 molCo cm2
geometric for the latter two samples respectively, in
the range of that of H2 CoCat. Polarization curves (Supplementary
Fig. S4) show that H2 CoCat is superior to the other two samples.
This definitively speaks for H2 CoCat being distinct in composition
and intrinsically more active than previously described Co-based
coatings. We note that molecular cobalt complexes combined with
carbon-based materials also yield active electrocatalytic electrodes
for H2 evolution33,34 . This activity has been recently assigned
to the formation of metallic cobalt nanoparticles at the surface
of the electrode35 .
The structure of H2 CoCat was characterized by X-ray photoelectron spectroscopy (XPS), energy dispersive X-ray (EDX)
spectroscopy, X-ray absorption spectroscopy (XAS), in particular
near-edge structure (XANES) and extended X-ray absorption
fine-structure (EXAFS).
The XPS spectrum of freshly electrodeposited H2 CoCat,
recorded under limited air exposure conditions, shows the presence
of cobalt, phosphorus and oxygen (Fig. 4) and was comparable to
that of commercially available Co3 (PO4 )2 xH2 O. No significant
signals are found in the N1s region when [Co(DO)(DOH)pnCl2 ]
or [Co(dmgBF2 )2 (H2 O)2 ] are used as cobalt precursors, confirming
that the organic ligand is completely split up from the cobalt
ion during the cathodic deposition process. The P2p region of
both samples exhibits two sharp peaks with 133.4 eV and 134.4 eV
binding energies (ratio of 2:1) corresponding to the 2p3/2 and 2p1/2
core levels of central phosphorus atoms in phosphate species36 . In
the Co region, two broad sets of signals corresponding to 2p3/2
(782 eV) and 2p1/2 (798 eV) core levels are observed, excluding
the presence of metallic cobalt (778.0 eV) at the surface of the
H2 CoCat coating. O1s signals are centred for both materials at
531.7 eV. P/Co/O ratios are however significantly different, with
a slight excess of cobalt and oxygen for H2 CoCat (2:3.9:11.6)
as compared with Co3 (PO4 )2 xH2 O (2:3:11.1). As the Co2p
and O1s core levels binding energies37 of cobalt oxides and
hydroxides are in the same range as those of cobalt phosphate, we
tentatively describe the surface of electrodeposited H2 CoCat as
a combination of cobalt(ii) phosphate with a cobalt oxo/hydroxo
species Cox Oy (OH)z , probably in the Co(ii) state, as observed
for native cobalt oxide/hydroxide that forms at the surface
of metallic cobalt.
EDX spectra (Supplementary Fig. S2a) confirm the presence of
Co, P and O together, and further signals arise from silicon and
tin from the FTO-coated glass substrate. As EDX spectroscopy,
which probes deeply into the cathodically deposited film, indicates
a Co/P ratio of 5:1, whereas XPS, which probes only a few
nanometres below the surface, indicates a 1:2 Co/P ratio, we
conclude that H2 CoCat is not homogeneous in composition
between bulk and surface.
The absence of crystalline features in the powder low-angle
X-ray diffraction pattern recorded at low angles and in the electron
diffraction patterns recorded in a transmission electron microscope
indicates the amorphous nature of H2 CoCat. For insight into the
atomic structure, X-ray absorption spectra were collected at the Co
K -edge after H2 CoCat formation and fast freezing of the still-wet
electrode in liquid nitrogen (quasi-in situ measurements)6 . When
deposition (paralleled by catalytic H2 evolution) was achieved at

NATURE MATERIALS | VOL 11 | SEPTEMBER 2012 | www.nature.com/naturematerials

2012 Macmillan Publishers Limited. All rights reserved

803

NATURE MATERIALS DOI: 10.1038/NMAT3385

ARTICLES
a

2 m

0.5 m

10 m

3 m

Figure 2 | Scanning electrochemical micrographs. a,b, SEM images of electrodes modified by electrolysis at 0.9 V versus Ag/AgCl for 1 h
([Co(DO)(DOH)pnCl2 ], 0.5 mmol l1 , ITO, 0.1 C cm2
geometric ) (a) and 1.0 V versus Ag/AgCl for 3 h (Co(NO3 )2 6H2 O or [Co(DO)(DOH)pnCl2 ],
1
0.5 mmol l1 , FTO, 6.5 C cm2
geometric ) versus RHE in phosphate buffer (KPi, 0.5 mol l , pH 7) (b). c, SEM image of an H2 CoCat film on a FTO electrode
1
equilibrated at 1.16 V versus Ag/AgCl for 90 min in a cobalt-free 0.5 mol l KPi, pH 7 electrolyte. d, SEM image of an H2 CoCat film on a FTO electrode
initially equilibrated at 1.0 V versus Ag/AgCl and taken out of the solution just after a potential switch to 1.16 V versus Ag/AgCl before equilibration
of the current.
0

log(j) (A cm2
geometric)

j (mA cm2
geometric)

0.5

0.4

3
4
5
6
0.6

0.4
0.2
E(V) versus RHE

0.3
0.2
E(V) versus RHE

0.1

0.0
0.0

Figure 3 | Compared HER properties of H2 CoCat and metallic cobalt.

Linear voltammetry experiments recorded at a FTO electrode (1 cm2 ) in
phosphate buffer (KPi, 0.5 mol l1 , pH 7) at a low scan rate (0.05 mV s1 )
for a FTO electrode modified with H2 CoCat (filled triangles) and at a Co
foil electrode (open squares). Potentials have been converted to RHE. The
corresponding Tafel plots are shown in the inset.

1.0 V versus Ag/AgCl, XANES and EXAFS measurements suggest

a dominating contribution of the hexagonal close-packed phase of
metallic cobalt (Fig. 5 and Supplementary Fig. S3). However, the
magnitude of EXAFS oscillation is about 40% smaller than that
observed for a Co metal foil and furthermore the edge spectra
(inset in Fig. 5) suggest a sizeable non-metallic contribution. An
appropriate subtraction of the metallic contribution results in an
804

EXAFS spectrum that suggests the presence of light atoms (O, N,

C) in the first Co coordination sphere (Supplementary Fig. S3d),
but the determination of bond lengths by EXAFS simulations
is prevented by noise in conjunction with uncertainties in the
approach used to correct for the dominating contributions of
the metallic cobalt.
All these data indicate that the new material is made of
nanoparticles with a cobalt oxo/hydroxo phosphate component
mostly located at the surface and metallic cobalt in the bulk.
Equilibration of the H2 CoCat electrode at a fairly positive
potential (+1.16 V versus Ag/AgCl) results in a stable anodic
current density of 1 mA cm2 (Supplementary Fig. S7) and in a
concomitant oxygen evolution with quantitative faradaic yield.
SEM observations of the electrode after 90 min (Fig. 2c) show a
homogeneous thin film, with cracks originated from drying, very
similar to those obtained for the O2 CoCat material (also named
CoPi) reported in 2008 (refs 4,6).
To characterize the electrocatalytic material after this redox shift
a detailed analysis by EDX spectroscopy, XPS and XAS has been
carried out. According to EDX spectra (Supplementary Fig. S2b)
the Co:P ratio is 1:2.5, which differs significantly from the 2:1 ratio
previously reported for O2 CoCat4 . XPS analysis confirms the large
phosphorus accumulation in the oxidized film (Supplementary
Fig. S8). On the other hand, the XANES and EXAFS spectra (Fig. 4
and Supplementary Fig. S9) of H2 CoCat films equilibrated at
+1.16 V versus Ag/AgCl, indicate that about 50% of the Co film
has undergone a transformation, resulting in a Co oxide material,
consisting of clusters of edge-sharing CoIII O6 octahedra6,7,38 , similar
to that found in O2 CoCat.
After cathodic deposition (black trace in Fig. 6), alternate
switches between oxidative (+1.775 V versus RHE, blue traces
in Fig. 6) and reductive conditions (385 mV versus RHE, red
NATURE MATERIALS | VOL 11 | SEPTEMBER 2012 | www.nature.com/naturematerials

2012 Macmillan Publishers Limited. All rights reserved

NATURE MATERIALS DOI: 10.1038/NMAT3385

a

1,400,000

1,000,000

OKLL

800,000

Co2p

5,000 cps

O1s

1,200,000

Intensity (cps)

ARTICLES
c

O1s

20,000 cps

Co2p

Co2s
538

600,000
CoLMM

400,000

N1s

200,000
0
1,200

1,000

250 cps

P2p
C1s P2s Co3s
Co3p

800
600
400
Binding energy (eV)

200

536 534 532 530

Binding energy (eV)

810

800
790
780
Binding energy (eV)

138

528

P2p

136
134
132
Binding energy (eV)

130

Figure 4 | XPS characterization. ac, XPS survey (a), Co2p (b), O1s (c, top) and P2p (c, bottom) core levels spectra of H2 CoCat deposited on FTO
substrates (red) and commercial Co3 (PO4 )2 xH2 O (black).

1.0

15
10
Q (C)

0.5
40
0.0
20

5
0
5

7,710

7,720
7,730
Energy (eV)

3
4
Reduced distance ()

Figure 5 | Fourier-transformed EXAFS spectra collected at the Co K edge.

Red trace: H2 CoCat equilibrated at 1.0 V versus Ag/AgCl. Blue trace:
H2 CoCat equilibrated at +1.16 V versus Ag/AgCl for 4 min. Black trace: Co
metal foil (hexagonal close-packed state). The XANES spectra of the film
formed at 1.0 V versus Ag/AgCl and of metallic cobalt are shown in the
inset. The blue arrows mark features that are assignable to the contribution
(50%, Supplementary Fig. S6) of a phase of edge-sharing CoO6
octahedra to the spectrum of the film equilibrated at +1.16 V versus
Ag/AgCl (blue line).

traces in Fig. 6) show that the deposited material catalyses both

water oxidation and H2 evolution respectively. Importantly we
could not find evidence of any decrease in activity for both
H2 and O2 evolution after several switchings. From the current
densities and chromatographic measurements, we derive turnover
frequencies of 10 and 80 h1 per Co centre at 545 and 385 mV
overpotential for O2 and H2 evolution respectively. We conclude
that the material reported here enjoys a fast redox-dependent
interconversion between the H2 CoCat form and the O2 CoCat
form, catalytically competent for H2 and O2 evolution respectively.
To determine whether this reversible transformation proceeds
through the complete dissolution of one form of the CoCat film
followed by the electrodeposition of the other form, we stopped
the experiment before current stabilization after a switch from
reductive to oxidative conditions. SEM analysis then revealed the
coexistence of different domains at the surface of the electrode
(Fig. 2d) corresponding to both morphologies. This clearly stands
for a progressive and local transformation of the material.

Spec. gas evol. rate

(nmol s1cm2geometric)

FT of k3 weighted EXAFS

60

10,000

20,000
30,000
Time (s)

40,000

15

H2
O2

12
9
6
3
0
0

10,000

20,000
30,000
Time (s)

40,000

Figure 6 | Electrocatalytic H2 and O2 evolution catalysed by the CoCat

film. a, Charge passed through a FTO electrode (1 cm2 ) during controlled
potential coulometry initially at 1.0 V versus Ag/AgCl (3 h, H2 CoCat
deposition, black curve) in 0.5 mol l1 KPi, pH 7 electrolyte containing
Co(NO3 )2 6H2 O (0.5 mmol l1 ) and after transfer to a cobalt-free
0.5 mol l1 KPi, pH 7 electrolyte, with potential switching between
oxidative (blue, 1.16 V versus Ag/AgCl) and reductive conditions (red, 1 V
versus Ag/AgCl). b, Specific rates of hydrogen (red) and oxygen (blue)
evolution quantified through gas chromatography measurements during
the same experiment. N2 was continuously bubbled through the electrolyte
at a constant flow (5 ml min1 ) during the experiment and the
concentration of O2 and H2 in the output gas was determined every 2 min
by gas chromatography.

First-row transition metals, namely Ni and Co, still exhibit exchange current densities (106 A cm2
geometric ) lower than
those of platinum (103 A cm2
geometric ) for catalytic H2 evolution (HER; refs 39,40). However photo-electro-chemical (PEC)
water-splitting devices require only low current densities (few
mA cm2
geometric ; refs 27,4143) as their activity is limited by the
photon flux from the sun, but they must be made from abundant
and cheaply processed materials both for light harvesting17,44,45
and catalysis (both HER; refs 17,28 and OER; refs 46,47)
to be economically viable. Oxidatively (O2 CoCat4 ) or reductively (MoS2 -; refs 16,26, NiMo-based materials27,28 ) deposited
OER and HER catalysts are fulfilling such requirements and

NATURE MATERIALS | VOL 11 | SEPTEMBER 2012 | www.nature.com/naturematerials

2012 Macmillan Publishers Limited. All rights reserved

805

NATURE MATERIALS DOI: 10.1038/NMAT3385

ARTICLES
are thus potential alternatives to platinum. Further investigation of new compositions and new nanostructures based on
abundant metals should thus provide improved electrocatalytic
materials. We report here a new HER electrocatalytic material
for such applications, namely H2 CoCat consisting of metallic
cobalt coated with a cobalt oxo/hydroxo phosphate layer, as
shown from detailed spectroscopic and microscopic characterizations with catalytic performances significantly improved with
respect to metallic cobalt.
H2 CoCat is remarkable for the following reasons. First, it
is easily prepared by electroreduction of simple divalent cobalt
salts. Second, H2 CoCat is a robust and active electrocatalytic
material for hydrogen evolution working at low overpotentials
(>50 mV) and under strictly neutral conditions. Third, it can
be converted on anodic equilibration into the O2 CoCat cobaltoxide film4 catalysing O2 evolution. The switch between the two
catalytic forms is fully reversible and corresponds to a local
interconversion between two morphologies at the surface of the
electrode and two types of non-metallic atomic structures. After
deposition, the coating thus functions as a robust, bifunctional
and switchable catalyst. To the best of our knowledge, comparable
properties have never previously been observed for any nonnoble metal catalyst.
This behaviour is possible because such electrodeposited
materials are in equilibrium with solution metal ions, depending
on the applied electrochemical potential. This has the advantage of
allowing self-repair to take place, as demonstrated for O2 CoCat4 .
In that specific case, Co2+ ions are released in the solution at mild
potential and can be redeposited under strong anodic conditions.
We show here that these Co2+ ions can also be redeposited
under cathodic conditions, but now forming H2 CoCat. This
process is reversible and O2 CoCat can be prepared either from
H2 CoCat or by anodic oxidation of metallic cobalt48 . These
findings thus open up new possibilities with respect to simple and
simultaneous electrodeposition of OER and HER catalysts from
the same phosphate-buffered electrolyte, as well as self-repair at
both electrodes. This could work in an artificial-leaf device27,4143 ,
even in the absence of a proton-conducting membrane separating
(photo)anode and (photo)cathode (for example, substitution of
expensive Nafion membranes by an inexpensive mesh for separation of gas bubbles). Such an ingeniously simple approach may also
be applicable towards photo-electrodeposition on heterogeneous
semiconductor nanoparticles49,50 .

Methods
Before starting the deposition, the FTO substrate is cycled hundreds of times
between 1 V and +1 V in 0.5 mol l1 potassium phosphate buffer (KPi), pH 7.0,
to ensure the stability and reproducibility of experiments. Catalyst films were grown
by controlled potential electrolysis of freshly prepared 0.5 mmol l1 Co(NO3 )2
solution and in 0.5 mol l1 KPi, pH 7.0. Performing the electrolysis at 1 V (versus
Ag/AgCl) gives rise to a catalyst film of several micrometres thickness formed by
nanoparticles after a course of around three hours. During this time, a film is
formed on the working electrode surface. After the film formation is completed,
the substrate is transferred to a cobalt-free 0.5 mol l1 KPi, pH 7.0, with potential
switching between oxidative (+1.16 V versus Ag/AgCl) and reductive conditions
(1 V versus Ag/AgCl).

Received 22 December 2011; accepted 20 June 2012;

published online 5 August 2012

References
1. Lewis, N. S. & Nocera, D. G. Powering the planet: Chemical
challenges in solar energy utilization. Proc. Natl Acad. Sci. USA 103,
1572915735 (2007); 104, 20142 (2007).
2. Armaroli, N. & Balzani, V. The hydrogen issue. ChemSusChem 4, 2136 (2011).
3. Gordon, R. B., Bertram, M. & Graedel, T. E. Metal stocks and sustainability.
Proc. Natl Acad. Sci. USA 103, 12091214 (2006).
4. Kanan, M. W. & Nocera, D. G. In situ formation of an oxygen-evolving
catalyst in neutral water containing phosphate and Co2+ . Science 321,
10721075 (2008).
806

5. Yin, Q. S. et al. A fast soluble carbon-free molecular water oxidation catalyst

based on abundant metals. Science 328, 342345 (2010).
6. Risch, M. et al. Cobalt-oxo core of a water-oxidizing catalyst film. J. Am.
Chem. Soc. 131, 69366937 (2009).
7. Dau, H. et al. The mechanism of water oxidation: From electrolysis via
homogeneous to biological catalysis. ChemCatChem 2, 724761 (2010).
8. Jiao, F. & Frei, H. Nanostructured cobalt and manganese oxide clusters as
efficient water oxidation catalysts. Energy Environ. Sci. 3, 10181027 (2010).
9. Shevchenko, D., Anderlund, M. F., Thapper, A. & Styring, S. Photochemical
water oxidation with visible light using a cobalt containing catalyst. Energy
Environ. Sci. 4, 12841287 (2011).
10. Wee, T-L. et al. Photochemical synthesis of a water oxidation catalyst based on
cobalt nanostructures. J. Am. Chem. Soc. 133, 1674216745 (2011).
11. Chou, N. H., Ross, P. N., Bell, A. T. & Don Tilley, T. Comparison
of cobalt-based nanoparticles as electrocatalysts for water oxidation.
ChemSusChem 11, 15661569 (2011).
12. Dinca, M., Surendranath, Y. & Nocera, D. G. Nickel-borate oxygen-evolving
catalyst that functions under benign conditions. Proc. Natl Acad. Sci. USA 107,
1033710341 (2010).
13. Zaharieva, I. et al. Synthetic manganese-calcium oxides mimic the
water-oxidizing complex of photosynthesis functionally and structurally.
Energy Environ. Sci. 4, 24002408 (2011).
14. Tran, P. D. et al. Noncovalent modification of carbon nanotubes with
pyrene-functionalized nickel complexes: Carbon monoxide tolerant
catalysts for hydrogen evolution and uptake. Angew. Chem. Int. Ed. 50,
13711374 (2011).
15. Le Goff, A. et al. From hydrogenases to noble metal-free catalytic nanomaterials
for H2 production and uptake. Science 326, 13841387 (2009).
16. Merki, D. & Hu, X. Recent developments of molybdenum and
tungsten sulfides as hydrogen evolution catalysts. Energy Environ. Sci. 4,
38783888 (2011).
17. Hou, Y. D. et al. Bioinspired molecular co-catalysts bonded to a silicon
photocathode for solar hydrogen evolution. Nature Mater. 10, 434438 (2011).
18. Artero, V., Chavarot-Kerlidou, M. & Fontecave, M. Splitting water with cobalt.
Angew. Chem. Int. Ed. 50, 72387266 (2011).
19. Baffert, C., Artero, V. & Fontecave, M. Cobaloximes as functional
models for hydrogenases. 2. proton electroreduction catalyzed by
difluoroborylbis(dimethylglyoximato)cobalt(ii) complexes in organic
media. Inorg. Chem. 46, 18171824 (2007).
20. Jacques, P-A., Artero, V., Pcaut, J. & Fontecave, M. Cobalt and
nickel diimine-dioxime complexes as molecular electrocatalysts for
hydrogen evolution with low overvoltages. Proc. Natl Acad. Sci. USA 106,
2062720632 (2009).
21. Razavet, M., Artero, V. & Fontecave, M. Proton electroreduction catalyzed
by cobaloximes: Functional models for hydrogenases. Inorg. Chem. 44,
47864795 (2005).
22. Dempsey, J. L., Winkler, J. R. & Gray, H. B. Mechanism of H2 evolution
by a photogenerated hydridocobaloxime. J. Am. Chem. Soc. 132,
1677416776 (2010).
23. Hu, X. L., Cossairt, B. M., Brunschwig, B. S., Lewis, N. S. & Peters, J. C.
Electrocatalytic hydrogen evolution by cobalt difluoroboryl-diglyoximate
complexes. Chem. Commun. 47234725 (2005).
24. Hu, X., Brunschwig, B. S. & Peters, J. C. Electrocatalytic hydrogen evolution at
low overpotentials by cobalt macrocyclic glyoxime and tetraimine complexes.
J. Am. Chem. Soc. 129, 89888998 (2007).
25. Fourmond, V., Jacques, P. A., Fontecave, M. & Artero, V. H2 evolution and
molecular electrocatalysts: Determination of overpotentials and effect of
homoconjugation. Inorg. Chem. 49, 1033810347 (2010).
26. Jaramillo, T. F. et al. Identification of active edge sites for electrochemical H2
evolution from MoS2 nanocatalysts. Science 317, 100102 (2007).
27. Reece, S. Y. et al. Wireless solar water splitting using silicon-based
semiconductors and earth-abundant catalysts. Science 334, 645648 (2011).
28. McKone, J. R. et al. Evaluation of Pt, Ni, and NiMo electrocatalysts for
hydrogen evolution on crystalline Si electrodes. Energy Environ. Sci. 4,
35733583 (2011).
29. Sun, Y. et al. Molecular cobalt pentapyridine catalysts for generating hydrogen
from water. J. Am. Chem. Soc. 133, 92129215 (2011).
30. Chen, W-F. et al. Hydrogen-evolution catalysts based on non-noble
metal nickelmolybdenum nitride nanosheets. Angew. Chem. Int. Ed. 51,
61316135 (2012).
31. Soto, A. B., Arce, E. M., Palomar-Pardave, M. & Gonzalez, I. Electrochemical
nucleation of cobalt onto glassy carbon electrode from ammonium chloride
solutions. Electrochim. Acta 41, 26472655 (1996).
32. Cui, C. Q., Jiang, S. P. & Tseung, A. C. C. Electrodeposition of cobalt from
aqueous chloride solutions. J. Electrochem. Soc. 137, 34183423 (1990).
33. Pantani, O., Anxolabehere-Mallart, E., Aukauloo, A. & Millet, P. Electroactivity
of cobalt and nickel glyoximes with regard to the electro-reduction of
protons into molecular hydrogen in acidic media. Electrochem. Commun. 9,
5458 (2007).
NATURE MATERIALS | VOL 11 | SEPTEMBER 2012 | www.nature.com/naturematerials

2012 Macmillan Publishers Limited. All rights reserved

NATURE MATERIALS DOI: 10.1038/NMAT3385

34. Berben, L. A. & Peters, J. C. Hydrogen evolution by cobalt tetraimine catalysts
adsorbed on electrode surfaces. Chem. Commun. 46, 398400 (2010).
35. Anxolabehere-Mallart, E. et al. Boron-capped tris(glyoximato) cobalt
clathrochelate as a precursor for the electrodeposition of nanoparticles
catalyzing H2 evolution in water. J. Am. Chem. Soc. 134, 61046107 (2012).
36. Hu, G-R., Deng, X-R., Peng, Z-D. & Du, K. Comparison of AlPO
4 and
Co3 (PO4 )2 coated LiNi0.8 Co0.2 O2 cathode materials for Li-ion battery.
Electrochim. Acta 53, 25672573 (2008).
37. Yang, J., Liu, H., Martens, W. N. & Frost, R. L. Synthesis and characterization
of cobalt hydroxide, cobalt oxyhydroxide, and cobalt oxide nanodiscs.
J. Phys. Chem. C 114, 111119 (2009).
38. Kanan, M. W. et al. Structure and valency of a cobalt-phosphate water
oxidation catalyst determined by in situ X-ray spectroscopy. J. Am. Chem. Soc.
132, 1369213701 (2010).
39. Trasatti, S. Work function, electronegativity, and electrochemical behavior
of metals. 3. Electrolytic hydrogen evolution in acid solutions. J. Electroanal.
Chem. 39, 163184 (1972).
40. Miles, M. H. Evaluation of electrocatalysts for water electrolysis in
alkaline-solutions. J. Electroanal. Chem. 60, 8996 (1975).
41. Khaselev, O. & Turner, J. A. A monolithic photovoltaic-photoelectrochemical
device for hydrogen production via water splitting. Science 280, 425427 (1998).
42. Rocheleau, R. E., Miller, E. L. & Misra, A. High-efficiency photoelectrochemical
hydrogen production using multijunction amorphous silicon photoelectrodes.
Energy Fuels 12, 310 (1998).
43. Yamada, Y. et al. One chip photovoltaic water electrolysis device. Int. J.
Hydrogen Energy 28, 11671169 (2003).
44. Sivula, K., Le Formal, F. & Gratzel, M. Solar water splitting: Progress using
hematite (-Fe2 O3 ) photoelectrodes. ChemSusChem 4, 432449 (2011).
45. Paracchino, A., Laporte, V., Sivula, K., Gratzel, M. & Thimsen, E. Highly active
oxide photocathode for photoelectrochemical water reduction. Nature Mater.
10, 456461 (2011).
46. Pijpers, J. J. H., Winkler, M. T., Surendranath, Y., Buonassisi, T. & Nocera,
D. G. Light-induced water oxidation at silicon electrodes functionalized
with a cobalt oxygen-evolving catalyst. Proc. Natl Acad. Sci. USA 108,
1005610061 (2011).
47. Zhong, D. K., Cornuz, M., Sivula, K., Graetzel, M. & Gamelin, D. R.
Photo-assisted electrodeposition of cobalt-phosphate (Co-Pi) catalyst on
hematite photoanodes for solar water oxidation. Energy Environ. Sci. 4,
17591764 (2011).

ARTICLES
48. Young, E. R., Nocera, D. G. & Bulovic, V. Direct formation of a water
oxidation catalyst from thin-film cobalt. Energy Environ. Sci. 3,
17261728 (2010).
49. Maeda, K., Higashi, M., Lu, D. L., Abe, R. & Domen, K. Efficient nonsacrificial
water splitting through two-step photoexcitation by visible light using a
modified oxynitride as a hydrogen evolution photocatalyst. J. Am. Chem. Soc.
132, 58585868 (2010).
50. Maeda, K. et al. Photocatalyst releasing hydrogen from waterenhancing
catalytic performance holds promise for hydrogen production by water
splitting in sunlight. Nature 440, 295295 (2006).

Acknowledgements
The authors thank P. Jegou for XPS measurements, B. Sartor for the design and
construction of a specific electrochemical cell allowing working with FTO-coated glass
electrodes, P. Chernev, K. Klingan, M. Risch and I. Zaharieva (FU Berlin) as co-workers
during the XAS measurements at the KMC-1 beamline of the BESSY synchrotron
(Helmholtz Zentrum Berlin, HZB) which were technically supported by F. Schfers
and M. Mertin (HZB). Financial support by the Nanosciences Program of CEA (Grant
Nanocat O2 ), the UniCat cluster of excellence (Unifying Concepts in Catalysis, Berlin)
and the European Commission (7th Framework Programme, SOLAR-H2, grant #
212508) is gratefully acknowledged.

Author contributions
V.A. and M.F. designed research; S.C., V.F., P-A.J., J.F., V.I. and V.A. performed research;
J.H. and H.D. performed XAS studies; B.J. and S.P. performed XPS studies; L.G. and S.C.
performed SEM measurements and surface EDX analysis; V.A. wrote the paper.

Additional information
Supplementary information is available in the online version of the paper. Reprints and
permissions information is available online at www.nature.com/reprints. Correspondence
and requests for materials should be addressed to V.A.

Competing financial interests

A European patent application (EP-12352001) has been filed for the preparation,
characterization and properties of H2 -CoCat.

NATURE MATERIALS | VOL 11 | SEPTEMBER 2012 | www.nature.com/naturematerials

2012 Macmillan Publishers Limited. All rights reserved

807