ing power that causes most chemical functional
groups to degrade. Accordingly, the generation of
oxygen from water presents a substantial challenge toward realizing artificial photosynthesis (8).
The fine-tuned molecular machinery of the
OEC oxidizes water at a low overpotential using
a Mn4O4Ca cluster (9–12). Outside the OEC, examples of water oxidation catalysts include firstrow spinel and perovskite metal oxides, which
require concentrated basic solutions (pH > 13)
and moderate overpotentials (<400 mV), and
precious metals and precious metal oxides, which
operate with similar efficiencies under acidic conditions (pH < 1) (13–15). However, few catalysts
operate in neutral water under ambient conditions. Neutral water is oxidized at Pt electrodes,
and some precious metal oxides have been reported to operate electrocatalytically in neutral or
weakly acidic solutions (16). The development of
an earth-abundant, first-row catalyst that operates
at pH 7 at low overpotential remains a fundamental chemical challenge. Here, we report an
oxygen-evolving catalyst that forms in situ upon
anodic polarization of an inert electrode in neutral
aqueous phosphate solutions containing Co2+.
Oxygen generation occurs under benign conditions: pH = 7, 1 atm, and room temperature.
Cobalt ions in the presence of chemical oxidants such as Ru(bpy)33+ (bpy, bipyridine; E o =
1.26, where E o is the standard potential) catalyze
the oxidation of water to O2 in neutral phosphate
solutions (17, 18). Oxygen yields drop in these
reactions when oxidized Co species precipitate
from solution because the catalytically active
species is removed from the solution-phase reaction. However, an oxidation-induced precipita-
In Situ Formation of an
Oxygen-Evolving Catalyst in Neutral
Water Containing Phosphate and Co2+
Matthew W. Kanan and Daniel G. Nocera*
The utilization of solar energy on a large scale requires its storage. In natural photosynthesis,
energy from sunlight is used to rearrange the bonds of water to oxygen and hydrogen
equivalents. The realization of artificial systems that perform “water splitting” requires catalysts
that produce oxygen from water without the need for excessive driving potentials.
Here we report such a catalyst that forms upon the oxidative polarization of an inert indium
tin oxide electrode in phosphate-buffered water containing cobalt (II) ions. A variety of analytical
techniques indicates the presence of phosphate in an approximate 1:2 ratio with cobalt in this
material. The pH dependence of the catalytic activity also implicates the hydrogen phosphate
ion as the proton acceptor in the oxygen-producing reaction. This catalyst not only forms in situ
from earth-abundant materials but also operates in neutral water under ambient conditions.
Department of Chemistry, 6-335, Massachusetts Institute
of Technology, Cambridge, MA 02139–4307, USA.
*To whom correspondence should be addressed. E-mail:
[email protected]
1072
O2 + 4H+ + 4e–
4H+ + 4e–
2H2O
Eanodic = 1.23 V – 0.059 (pH) V vs NHE
2H2
Ecathodic = 0 V – 0.059 (pH) V vs NHE
2H2 + O2
2H2O
Erxn
= –1.23 V
Scheme 1.
B
A
1.25
50 µA
1.4
25 µA
1.2
1.0
V (vs NHE)
1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
V (vs NHE)
1.00
mA/10–5 cm2
S
Scheme 1 (half-cell potentials given in the convention of reduction potentials).
The voltage in addition to E that is required
to attain a given catalytic activity—referred to as
overpotential—limits the efficiency of converting
light into catalytic current. Of the two reactions,
the H2O/O2 reaction is considerably more complex (5). This reaction requires a four-electron
oxidation of two water molecules coupled to the
removal of four protons to form a relatively weak
oxygen-oxygen bond. In addition to controlling
this proton-coupled electron transfer (PCET)
(6, 7), a catalyst must tolerate prolonged exposure
to oxidizing conditions. Even at the thermodynamic limit, water oxidation requires an oxidiz-
mA/cm2
unlight is the only renewable and carbonneutral energy source of sufficient scale to
replace fossil fuels and meet rising global
energy demand (1). The diurnal variation in local
insolation, however, demands a cost-effective
storage of solar energy for its large-scale utilization. Of the possible storage methods, nature
provides the blueprint for storing sunlight in the
form of chemical fuels (1, 2). The primary steps
of natural photosynthesis involve the absorption
of sunlight and its conversion into spatially separated electron/hole pairs. The holes of this wireless
current are then captured by the oxygen-evolving
complex (OEC) to oxidize water to oxygen and
the electrons are captured by photosystem I to
reduce NADP+ (nicotinamide adenine dinucleotide phosphate) to NADPH (the reduced form of
NADP+), nature’s form of hydrogen (3). Thus,
the overall primary events of photosynthesis store
solar energy in a fuel by rearranging the chemical bonds of water to form H2 (i.e., NADPH)
and O2.
An approach to duplicating photosynthesis
outside of a photosynthetic membrane is to convert sunlight into spatially separated electron/
hole pairs within a photovoltaic cell and then
capture the charges with catalysts that mediate
“water splitting” (1, 4). The four holes are
captured by a catalyst at the anode to produce
oxygen, and the four electrons are captured by a
separate catalyst at the cathode to produce
hydrogen. The net result is the storage of solar
energy in the chemical bonds of H2 and O2.
A key determinant of energy storage in artificial photosynthesis is the efficiency of the
water-splitting catalysts. Electrocatalysts that are
efficient for solar-to-fuels conversion must operate close to the Nernstian potentials (E) for the
H2O/O2 and H2O/H2 half-cell reactions shown in
0.75
0.50
6.0
4.0
2.0
0
0.25
0
1
2
3
4 5
time / h
1 2 3 4 5
time / h
6
7
8
Fig. 1. (A) Cyclic voltammagram in 0.1 M KPi electrolyte at pH 7.0 with no Co2+ ion present (black line)
and with 0.5 mM Co2+ present (red line). The potential was measured against a Ag/AgCl reference and
converted to NHE potentials by using E(NHE) = E(Ag/AgCl) + 0.197 V. (B) Current density profile for bulk
electrolysis at 1.29 V (versus NHE) in 0.1 M KPi electrolyte at pH 7.0 containing 0.5 mM Co2+. (Inset)
Profile in the absence of Co2+.
22 AUGUST 2008
VOL 321
SCIENCE
www.sciencemag.org
Downloaded from www.sciencemag.org on September 3, 2008
REPORTS
tion may be exploited to prepare electrocatalysts
in situ if the precipitated material remains catalytically active and can be oxidized at an electrode
surface. To explore this possibility for Co-catalyzed
water oxidation, we examined electrochemical oxidations of aqueous solutions containing phosphate
and Co2+. Cyclic voltammetry of a 0.5 mM solution of Co(NO3)2 in 0.1 M potassium phosphate
pH 7.0 (KPi electrolyte) exhibits an oxidation wave
at Ep = 1.13 V (where Ep is the peak potential) versus
the normal hydrogen electrode (NHE), followed by
the onset of a strong catalytic wave at 1.23 V (Fig.
1A). A broad, relatively weak reduction wave is observed on the cathodic scan. The presence of a catalytic wave prompted us to examine the electrode
activity during controlled-potential electrolysis.
Indium tin oxide (ITO) was used as the electrode for bulk electrolysis to ensure a minimal
background activity for O2 production. An electrolysis at 1.29 V without stirring in neutral KPi
electrolyte containing 0.5 mM Co2+ exhibits a
rising current density that reaches a peak value
>1 mA/cm2 after 7 to 8 hours (Fig. 1B). During
this time, a dark coating forms on the ITO
surface, and effervescence from this coating be-
A
through cracks in the film that form upon drying,
as evidenced by particles that are split into complementary pieces. The film thickness gradually
increases over the course of the electrodeposition
(see fig. S4 for additional images). At maximum
activity under these electrolysis conditions, the
film is >2 mm thick. The x-ray powder diffraction
pattern of an electrodeposited catalyst shows
broad amorphous features and no peaks indicative of crystalline phases other than the peaks associated with the ITO layer (fig. S1).
In the absence of detectable crystallites, the
composition of the electrodeposited material was
analyzed by three complementary techniques.
Energy-dispersive x-ray analysis (EDX) spectra
were obtained from multiple 100-to-300–m2
regions of several independently prepared samples. These spectra identify Co, P, K, and O as the
principal elemental components of the material
(Fig. 2B). Although the material’s morphology is
not ideally suited for quantitative EDX, the analyses consistently indicate a Co:P:K ratio between
~2:1:1 and 3:1:1. To obtain an independent determination of elemental composition, electrolysis
was performed with several larger ITO electrodes;
the deposited material was scraped off and combined for a total yield of ~3 mg. Microanalytical
elemental analysis of the combined material indicates 31.1% Co, 7.70% P, and 7.71% K, corresponding to a 2.1:1.0:0.8 Co:P:K ratio. Finally,
the surface of an electrodeposited catalyst on the
ITO substrate was analyzed by x-ray photoelectron
spectroscopy (XPS). All peaks in the XPS spectra
are accounted for by the elements detected above,
in addition to In and Sn from the ITO substrate.
The high-resolution P 2p peak at 133.1 eV is
consistent with phosphate. The Co 2p peaks at
780.7 and 795.7 eVare in a range typical of Co2+
or Co3+ bound to oxygen (fig. S2) (21). Together,
the x-ray diffraction and analytical results indicate that electrolysis of a Co2+ solution in neutral
KPi electrolyte results in the electrodeposition of
an amorphous Co oxide or hydroxide incorporating a substantial amount of phosphate anion at a
stoichiometric ratio of roughly 2:1:1 for Co:P:K.
comes increasingly vigorous (19).The same results
are observed with either CoSO4, Co(NO3) 2, or
Co(OTf )2 (where OTf = triflate) as the Co2+
source, which indicates that the original Co2+
counterion is unimportant and that this activity
does not depend on an impurity found in a specific
source. The amount of charge passed during the
course of an 8-hour electrolysis far exceeds what
could be accounted for by stoichiometric oxidation
of the Co2+ in solution (20). These observations are
indicative of the in situ formation of an oxygenevolving catalyst. Catalyst formation also proceeds on a fluorine tin oxide electrode and if KPi
is replaced by NaPi electrolyte. In a control experiment, the current density during bulk electrolysis under identical conditions in the absence of
Co2+ rapidly drops to a baseline level of ~25 nA/cm2
(inset in Fig. 1B).
The morphology of the electrode coating
formed during electrolysis in the presence of Co2+
was examined by scanning electron microscopy
(SEM). The electrodeposited material consists of
particles that have coalesced into a thin film and
individual micrometer-sized particles on top of
the film (Fig. 2A). The ITO substrate can be seen
B
1.2
cps / eV
1.0
0.8
K
P
O
0.6
Co
K
Co
0.4
5µm
0.2
1 2 3
4 5 6 7 8 9 10
E / keV
Fig. 2. (A) SEM image (30° tilt) of the electrodeposited catalyst after 30 C/cm2 were passed in 0.1 M KPi
electrolyte at pH 7.0, containing 0.5 mM Co2+. The ITO substrate can be seen through cracks in the dried
film. (B) Typical EDX histogram acquired at 12 kV. cps, counts per second.
A
4000
B
120
C
80
73.4 ± 0.6%
Downloaded from www.sciencemag.org on September 3, 2008
REPORTS
100
40
0
2000
1.0
2.0 3.0
time (h)
1000
80
60
µmol O2
counts
3000
% Abundance
80
40
24.5 ± 0.6%
20
0
0
0.5 1.0 1.5 2.0 2.5 3.0 3.5
time / h
40
20
2.1 ± 0.2%
1.0
Fig. 3. (A) Mass spectrometric detection of isotopically labeled 16,16O2 (black
line), 16,18O2 (blue line), and 18,18O2 (red line) during electrolysis of a catalyst
film on ITO in a KPi electrolyte containing 14.6% 18OH2. The green arrow
indicates the initiation of electrolysis at 1.29 V (NHE), and the red arrow
indicates the termination of electrolysis. (Inset) Expansion of the 18,18O2
signal. (B) Percent abundance of each isotope over the course of the
www.sciencemag.org
60
SCIENCE
1.5
2.0
time / h
2.5
0
0
2
4
6
8 10 12 14 16
time / h
experiment. Average observed abundance of T2s is indicated above each line.
Statistical abundances are 72.9, 24.9, and 2.1%. (C) O2 production measured
by fluorescent sensor (red line) and theoretical amount of O2 produced (blue
line), assuming a Faradic efficiency of 100%. The green arrow indicates the
initiation of electrolysis at 1.29 V, and the red arrow indicates the termination
of electrolysis.
VOL 321
22 AUGUST 2008
1073
1074
profile may be attainable without changing the
catalyst composition by depositing on alternative
substrates or improving ohmic contact to the ITO.
The catalyst used to obtain the Tafel plot at
pH 7 was subsequently transferred to KPi
electrolyte at pH 4.6, and the current density
was measured at a constant applied potential
(1.24 V) while the pH was increased incrementally to 9.4 by adding aliquots of concentrated
KOH. A plot of the log of current density versus
pH exhibits a steep initial rise that levels off in the
high-pH range such that increasing the pH from 8
to 9.4 at this applied potential has little effect
(Fig. 4B). These data can be converted to a Tafel
plot by using Eq. 1 (Scheme 1) and accounting
for iR drop (see Fig. 4 legend). A comparison to
the Tafel plot obtained at pH 7 indicates that the
catalyst exhibits approximately Nernstian behavior from pH 5 to 8: Increasing the pH by one unit
at constant applied potential (1.24 V) has nearly
the same effect as increasing the overpotential by
0.059 V at pH 7 (red circles in Fig. 4A). This
result implicates a reversible ne–, nH+ removal
before the rate-determining step for O2 evolution
in this pH range (here, n is the number of equivalents). Thus, an important component of the
activity at pH 7 with this catalyst is the existence
of one or more intermediates preceding O2 formation that are deprotonated reversibly by HPO42– in
a PCET event (22). The pH-independent behavior
above pH 8 at the applied potential may indicate a
change in mechanism, most likely involving a
deprotonated intermediate.
In addition to mediating the deprotonation required for catalysis, the KPi electrolyte provides
a medium for in situ catalyst formation. Given
that phosphate is a structural element and that the
catalyst forms under oxidizing conditions, it is
plausible that deposition is driven by the interaction of phosphate and Co3+. By judicious choice
of other metal-anion pairs or combinations of
multiple metals and anions, it may be possible to
access other oxygen-evolving catalysts that form
in situ and operate in neutral solutions. In situ
formation is advantageous because, in principle,
it enables catalyst deposition on a variety of substrates, including those that are too delicate to
tolerate traditional catalyst preparation techniques.
This attribute is important for interfacing a catalyst with a variety of electrochemical or photoelectrochemical cell designs.
In situ formation also implies a self-repair mechanism. Proposed molecular mechanisms involving
O2/H2O cycles at Co centers suggest that catalytic
reactions cycle among Co2+-, Co3+-, and Co4+oxo oxidation states (18, 23). The propensity of
metal ion dissolution has been shown to correlate
with ligand substitution (24). Given that Co3+ is
substitutionally inert relative to Co2+, a dynamic
equilibrium between Co2+-HPO42– in solution
and Co3+-HPO42– on the anodically poised electrode may be established. More generally, if a
catalytic cycle involves an oxidation state that is
prone to dissolution, this process can be countered
by continual catalyst formation by establishing an
equilibrium with the judicious choice of an anion.
The results reported herein highlight a new
area of exploration for the development of easily
prepared, earth-abundant catalysts that oxidize
water. If artificial photosynthesis is to enable the
storage of solar energy commensurate with global
demand, water-splitting chemistry will need to be
performed at a daunting scale. Storing the equivalent of the current energy demand would require
splitting more than 1015 mol/year of water, which
is roughly 100 times the scale of nitrogen fixation
by the Haber-Bosch process. The conditions under
which water splitting is performed will determine
how solar energy is deployed. The catalyst reported
here has many elements of natural photosynthesis,
including (i) its formation from earth-abundant
metal ions in aqueous solution, (ii) a plausible
pathway for self-repair, (iii) a carrier for protons
in neutral water, and (iv) the generation of O2 at
low overpotential, neutral pH, 1 atm, and room
temperature.
A
B
–2.0
–2.5
–2.5
–3.0
–3.0
–3.5
pH = 5.0
pH = 8.0
–4.0
log(A/cm2)
log(A/cm2)
Three experiments were performed to establish that the catalytic activity observed with this
material corresponds to authentic water oxidation. Each of these experiments was performed in
neutral KPi electrolyte in the absence of Co2+.
Catalyst coatings (~1.3 cm2) were prepared in a
preliminary step as described above and stored
under ambient laboratory conditions until they
were used.
To confirm that water is the source of the O2
produced, electrolysis was performed in heliumsaturated electrolyte containing 14.6% 18OH2 in
a gas-tight electrochemical cell in line with a mass
spectrometer. The helium carrier gas was continuously flowed through the headspace of the
anodic compartment into the mass spectrometer,
and the relative abundances of 32O2, 34O2, and
36
O2 were monitored at 2-s intervals. Within
minutes of initiating electrolysis at 1.29 V, the
signals for the three isotopes began to rise above
their background levels as the O2 produced by
the catalyst escaped into the headspace. Upon
terminating the electrolysis 1 hour later, these signals slowly returned to their background levels
(Fig. 3A). The 32O2, 34O2, and 36O2 isotopes were
detected in the statistical ratio (72.9, 24.9, and
2.1% relative abundances, respectively) (Fig. 3B).
The Faradaic efficiency of the catalyst was
measured with a fluorescence-based O2 sensor.
Electrolysis was performed in KPi electrolyte in a
gas-tight electrochemical cell under an N2 atmosphere with the sensor placed in the headspace.
After initiating electrolysis at 1.29 V, the percentage of O2 detected in the headspace rose in accord with what was predicted by assuming that
all of the current was caused by 4e– oxidation of
water to produce O2 (Fig. 3C). The amount of O2
produced (95 mmol, 3.0 mg) greatly exceeds the
amount of catalyst (~0.2 mg), which shows no
perceptible decomposition during the course of
the experiment.
The stability of phosphate under catalytic conditions was assayed by 31P nuclear magnetic resonance (NMR). Electrolysis in a two-compartment
cell with 10 mL of KPi electrolyte (1 mmol phosphate) on each side was allowed to proceed until
45 C had been passed through the cell (0.46 mmol
electrons). Electrolysis solutions from both chambers show single, clean 31P resonances, which
indicate that the electrolyte is robust under these
conditions (fig. S3). Together, the mass spectrometry, Faradaic efficiency, and 31P NMR results
demonstrate that the electrodeposited catalyst
cleanly oxidizes H2O to O2 in neutral KPi solutions.
The current density of a catalyst on ITO was
measured as a function of the overpotential (h) in
KPi electrolyte without Co2+ (black circles in
Fig. 4A). At pH 7.0, appreciable catalytic current
is observed beginning at h = 0.28 V, and a current
density of 1 mA/cm2 (corresponding to 9 mmol
O2 cm–2 h–1) requires h = 0.41 V. The Tafel plot
deviates slightly from linearity, possibly reflecting an uncompensated iR drop caused by the
surface resistivity of the ITO (8 to 12 ohms per
square). Substantial improvements in the activity
–3.5
–4.0
–4.5
–4.5
0.25 0.30 0.35 0.40 0.45 0.50
η/V
–5.0
4
5
6
7
pH
8
9
10
Fig. 4. (A) Tafel plot (black circles), h = (Vappl – iR) – E(pH 7) (where Vappl is the applied potential),
of a catalyst film on ITO in 0.1 M KPi electrolyte pH 7.0, corrected for the iR drop of the solution.
pH data were converted into a Tafel plot (red circles), h = (Vappl + 0.059∆pH – iR) – E(pH 7),
assuming Nernstian behavior and correcting for the iR drop of the solution. The pH = 5 and pH = 8
data points are indicated by arrows. (B) Current density dependence on pH in 0.1 M KPi electrolyte.
The potential was set at 1.24 V (versus NHE) with no iR compensation.
22 AUGUST 2008
VOL 321
SCIENCE
www.sciencemag.org
Downloaded from www.sciencemag.org on September 3, 2008
REPORTS
REPORTS
14. J. O. Bockris, T. J. Otagawa, J. Electrochem. Soc. 131,
290 (1984).
15. M. R. Tarasevich, B. N. Efremov, in Electrodes of
Conductive Metal Oxides, S. Trasatti, Ed. (Elsevier,
Amsterdam, 1980), chap. 5.
16. M. Yagi, E. Tomita, S. Sakita, T. Kuwabara, K. Nagai,
J. Phys. Chem. B 109, 21489 (2005).
17. V. Y. Shafirovich, N. K. Khannanov, V. V. Strelets, Nouv.
J. Chim. 4, 81 (1980).
18. B. S. Brunschwig, M. H. Chou, C. Creutz, P. Ghosh,
N. Sutin, J. Am. Chem. Soc. 105, 4832 (1983).
19. Materials and methods, videos of an active electrode, and
figs. S1 to S4 are available as supporting material on
Science Online.
20. In a typical experiment, >40 C are passed over 8 hours,
whereas oxidation of all the Co2+ in solution requires
1.9 C per oxidation-state change.
21. K. D. Bomben, J. F. Moulder, P. E. Sobol, W. F. Stickel, in
Handbook of X-Ray Photoelectron Spectra: A Reference
Book of Standard Spectra for Identification, J. Chastain,
Ed. (Perkin Elmer, Eden Prairie, MN, 1992).
22. T. Irebo, S. Y. Reece, M. Sjödin, D. G. Nocera,
L. Hammarström, J. Am. Chem. Soc. 129, 15462 (2007).
The Global Atmospheric Circulation
on Moist Isentropes
The global atmospheric circulation transports energy from the equatorial regions to higher
latitudes through a poleward flow of high-energy and -entropy parcels and an equatorward
flow of air with lower energy and entropy content. Because of its turbulent nature, this
circulation can only be described in some averaged sense. Here, we show that the total mass
transport by the circulation is twice as large when averaged on moist isentropes than when
averaged on dry isentropes. The additional mass transport on moist isentropes corresponds
to a poleward flow of warm moist air near Earth's surface that rises into the upper
troposphere within mid-latitudes and accounts for up to half of the air in the upper
troposphere in polar regions.
E
1
Courant Institute of Mathematical Sciences, New York
University, 251 Mercer Street, New York, NY 10012, USA.
Space and Atmospheric Physics Group, Department of
Physics, Imperial College, Huxley Building, Room 726, London
SW7 2AZ, UK. 3Department of Atmospheric Sciences, Texas
A&M University, 3150 TAMU, College Station, TX 77843–
3150, USA.
2
*To whom correspondence should be addressed. E-mail:
[email protected]
circulation (1), obtained by averaging the flow
at constant pressure or geopotential height. The
Eulerian mean stream function Yp is defined as
Yp ðp; fÞ ¼
1 t 2p psurf
dp
va cos f dldt
t ∫0 ∫0 ∫p
g
ð1Þ
Here, p is pressure, f is latitude, t is the time
period over which the average is computed, psurf
is surface pressure, l is longitude, a is Earth's
radius, v is the meridional velocity, and g is the
gravitational acceleration. Figure 1A shows the
annual mean stream function on pressure surfaces based on the National Centers for Environmental Prediction–National Center for Atmospheric
Research (NCEP-NCAR) Reanalysis monthly
data (3) from January 1970 to December 2004.
The Eulerian-mean circulation exhibits a three-cell
structure in each hemisphere: the Hadley cell in
the tropics, the Ferrel cell in mid-latitudes, and a
polar cell at high latitudes. The Hadley and polar
cells, with air parcels moving poleward at high
altitude and equatorward at low altitude, are direct
circulations that transport energy toward the poles.
In the Ferrel cell, the flow is poleward near the
surface and equatorward at high altitude. This corresponds to an energy transport toward the equa-
www.sciencemag.org
SCIENCE
VOL 321
Supporting Online Material
www.sciencemag.org/cgi/content/full/1162018/DC1
Materials and Methods
Figs. S1 to S4
Movies S1 and S2
19 June 2008; accepted 18 July 2008
Published online 31 July 2008;
10.1126/science.1162018
Include this information when citing this paper.
tor. Nonetheless, in mid-latitudes, synoptic-scale
(~1000 km) eddies transport more energy toward
the poles than is brought equatorward by the Ferrel
cell, so that the total energy transport in the atmosphere remains poleward.
An alternative to the Eulerian mean circulation is to average the circulation on isentropic
surfaces (4–6). In atmospheric sciences, it is common to use the potential temperature q instead of
entropy. The potential temperature is given by
CR
p
q ¼ pp0 T , with p the pressure, R the ideal gas
Olivier Pauluis,1* Arnaud Czaja,2 Robert Korty3
arth absorbs shortwave radiation from the
Sun and emits back longwave radiation to
space. Although the total amounts of energy received and emitted are about equal, Earth
absorbs more energy than it emits in the equatorial regions and emits more energy than it absorbs at high latitudes (1). Such imbalance requires
an energy transport by the atmosphere and the
oceans, with the former responsible for the bulk
of the transport in mid-latitudes (2). Determining
the relationship between the atmospheric energy
transport and the global distribution of temperature and humidity is a central question for our
understanding of the Earth's climate.
Averaging the global atmospheric circulation
usually implies computing a zonal and temporal
mean over a sufficiently long period. One of the
most common descriptions is the Eulerian mean
23. C. J. Chang, Z.-H. Loh, C. Shi, F. C. Anson, D. G. Nocera,
J. Am. Chem. Soc. 126, 10013 (2004).
24. W. H. Casey, J. Colloid Interface Sci. 146, 586
(1991).
25. This work was supported by a grant from the NSF
Chemical Bonding Center (CHE-0802907). M.W.K. is
supported by a Ruth L. Kirchenstein National Research
Service Award postdoctoral fellowship provided by NIH
(F32GM07782903). We thank E. Shaw for obtaining XPS
spectra, G. Henoch for providing the videos in the
supporting online material, and Y. Surendranath for
many productive discussions.
constant, Cp the specific heat, T the temperature,
and p0 = 1000 mbar an arbitrary reference pressure. Potential temperature is conserved for reversible adiabatic transformations in the absence
of a phase transition, and a surface of constant
potential temperature corresponds to isentropic
surfaces. The stream function Yq(q,f) on potential temperature surfaces is defined by
Yq ðq0 ; fÞ
¼
1 t 2p psurf
dp
Hðq0 − qÞva cos f dldt
t ∫0 ∫0 ∫0
g
Downloaded from www.sciencemag.org on September 3, 2008
References and Notes
1. N. S. Lewis, D. G. Nocera, Proc. Natl. Acad. Sci. U.S.A.
103, 15729 (2006).
2. N. Nelson, A. Ben-Shem, Nat. Rev. Mol. Cell Biol. 5, 971
(2004).
3. J. Barber, Philos. Trans. R. Soc. London Ser. A 365, 1007
(2007).
4. A. J. Bard, M. A. Fox, Acc. Chem. Res. 28, 141 (1995).
5. T. A. Betley, Q. Wu, T. Van Voorhis, D. G. Nocera, Inorg.
Chem. 47, 1849 (2008).
6. R. I. Cukier, D. G. Nocera, Annu. Rev. Phys. Chem. 49,
337 (1998).
7. M. H. V. Huynh, T. J. Meyer, Chem. Rev. 107, 5004 (2007).
8. R. Eisenberg, H. B. Gray, Inorg. Chem. 47, 1697 (2008).
9. K. N. Ferreira, T. M. Iverson, K. Maghlaoui, J. Barber,
S. Iwata, Science 303, 1831 (2004), published online
5 February 2004; 10.1126/science.1093087.
10. S. Iwata, J. Barber, Curr. Opin. Struct. Biol. 14, 447 (2004).
11. J. Yano et al., Science 314, 821 (2006).
12. B. Loll, J. Kern, W. Saenger, A. Zouni, J. Biesiadka,
Nature 438, 1040 (2005).
13. S. Trassati, in Electrochemistry of Novel Materials,
J. Lipkowski, P. N. Ross, Eds. (VCH, New York, 1994), chap. 5.
ð2Þ
Here, H(x) is the Heavyside function, with H(x) =
1 for x ≥ 0 and H(x) = 0 for x < 0. Figure 1B
shows the annual mean stream function on potential temperature surfaces based on the NCEPNCAR Reanalysis daily data from January 1970
to December 2004.
In contrast to the Eulerian mean circulation,
the circulation in isentropic coordinates exhibits a
single overturning cell in each hemisphere. Because the atmosphere is stratified in potential
temperature (∂zq > 0), the isentropic circulation
corresponds to a poleward flow at upper levels
balanced by a return flow near Earth's surface—
in the direction opposite to the Eulerian mean
circulation (4). The meridional mass transport on
an isentrope can be written as
22 AUGUST 2008
rq v ¼ rq v þ rq′ v′
(3)
1075
Electrolyte-Dependent Electrosynthesis and Activity of Cobalt
Based Water Oxidation Catalysts
Yogesh Surendranath, Mircea Dincă, Daniel G. Nocera
Department of Chemistry, 6-335, Massachusetts Institute of Technology, Cambridge,
Massachusetts 02139-4307
Abstract: Electrolysis of Co2+ in phosphate, methylphosphonate and borate electrolytes effects
the electrodeposition of an amorphous highly-active water oxidation catalyst as a thin-film on an
inert anode. Electrodeposition of a catalytically competent species immediately follows
oxidation of Co2+ to Co3+ in solution. Methylphosphonate and borate electrolytes support
catalyst activity comparable to that observed for phosphate. Catalytic activity for O2 generation
in aqueous solutions containing 0.5 M NaCl is retained for catalysts grown from phosphate
electrolyte.
Introduction
The solar-driven electrochemical splitting of water to produce hydrogen and oxygen
provides an effective means of energy storage.1,2 This water-to-solar fuels conversion requires
proton-coupled multielectron oxidation of water to O2 with the release of four protons and their
subsequent reduction to H2. Of these two processes, the oxidation reaction is particularly
demanding3,4 because it requires the removal of four protons and four electrons, and the
formation of two oxygen-oxygen bonds. Commercial electrolyzers perform the water splitting
reaction at high efficiency and current density, but this technology is not well matched to
conditions needed for many envisioned non-concentrated (e.g. distributed) solar applications.5 In
these electrolyzers, the water-splitting reaction is performed under harsh physical and chemical
conditions and the systems are expensive. We have turned our attention to developing watersplitting catalysts for inexpensive and highly manufacturable systems for non-concentrated solar
applications. A recent report from our laboratory shows that the electrolysis of Co2+ salts in pH 7
phosphate electrolyte (Pi) effects the electrodeposition of a highly active water oxidation catalyst
as an amorphous thin film on an inert indium-tin-oxide (ITO) or fluorine-tin-oxide (FTO)
electrode.6 In contrast to spinel and perovskite metal oxides that oxidize water under highly
alkaline conditions,7 the electrodeposited cobalt-phosphate compound is among the few catalysts
that oxidize water at neutral pH,8,9 and among these, is unique because it does not involve a
precious metal constituent.
We now elaborate the design of the system in two ways: (1) catalyst formation and
function is not restricted to Pi; a fully functional catalyst is formed from a methylphosphonate
(MePi) or borate (Bi) electrolyte; and (2) oxygen production from the electrodeposited catalyst is
not impeded by the presence of chloride, allowing efficient oxygen evolution from salt water.
The activity of the catalyst is evaluated in comparison to materials deposited from electrolytes
that are poor proton acceptors at neutral pH. In the absence of proton-accepting electrolytes, (1)
catalyst formation is significantly impeded at a given potential and is not observed under
2
conditions similar to those employed for Co-Pi, Co-MePi or Co-Bi catalysts; (2) catalyst activity
is severely diminished and deteriorates over time; and (3) chloride oxidation out-competes water
oxidation from salt solutions.
These results together establish the imperative for proton-
accepting electrolytes, which enable catalyst formation, sustained activity and functional
stability.
Results
Cyclic Voltammetry. Cyclic voltammetry (CV) scans of a glassy carbon working
electrode in an aqueous 0.5 mM Co2+ solution in 0.1 M potassium phosphate electrolyte at pH
7.0 (Pi), 0.1 M sodium methylphosphonate electrolyte at pH 8.0 (MePi), and 0.1 M potassium
borate electrolyte at pH 9.2 (Bi) are shown in Figure 1 along with the respective background
traces taken in pure electrolyte with no Co2+. As reported previously,6 in Pi electrolyte a sharp
anodic wave is observed at Ep,a = 1.13 V vs. NHE, followed by a strong catalytic wave at 1.23 V.
The corresponding sharp anodic wave in MePi is observed at Ep,a = 1.04 V, followed by the
onset of a catalytic wave at 1.14 V. In Bi electrolyte, the anodic wave is observed at Ep,a = 0.77
V and is well separated from the catalytic wave at 1.10 V. A catalytic current of 100 μA is
observed at 1.34, 1.27, and 1.20 V for Pi, MePi, and Bi electrolytes, respectively. The 70 mV
shift between MePi and Bi reflects the 72 mV shift in the thermodynamic potential for water
oxidation between pH 8.0 and 9.2. A broad cathodic wave at Ep,c = 0.93, 0.81, and 0.55 is
observed in Pi, MePi, and Bi, respectively; for the latter electrolyte, the cathodic wave is also
followed by a broad cathodic shoulder. On subsequent scans, the sharp anodic pre-feature of all
electrolyte solutions is replaced by a broad anodic wave that grows upon repetitive scanning
suggesting adsorption of an electroactive species (Figure S1, inset).
The phenomenon of electrodeposition was probed further in MePi and Bi electrolytes
where the anodic pre-feature is well separated from the catalytic wave. A single CV scan of a
fresh polished glassy carbon electrode was performed in solutions of 0.5 mM Co2+ in MePi and
Bi electrolytes and the scan was reversed beyond the anodic pre-feature wave but prior to the
3
catalytic wave. The electrode was removed from solution, rinsed with water, and placed in fresh
electrolyte solution containing no Co2+. The subsequent CV scans of the electrode in MePi and
Bi are shown in Figure 2. For both electrolytes, a strong catalytic wave is observed.
Film Preparation and Characterization. To investigate the nature of the catalytic
wave, controlled potential electrolysis was performed at 1.3 V in a conventional two
compartment cell. In each case, the working compartment was charged with either a 1 mM Co2+
solution in MePi electrolyte, or a 0.5 mM Co2+ solution in Bi electrolyte, whereas the auxiliary
compartment was charged with pure electrolyte. ITO coated glass slides were used as working
electrodes in each case. In MePi, the current density reaches an asymptotic limit of 1.5 mA/cm2
over the course of 2 hours (Figure 3a). In Bi, the current density reaches an asymptotic limit of
2.3 mA/cm2 over the course of 10 minutes (Figure 3b). In both cases, the rise in current is
accompanied by the formation of a dark green film on the ITO electrode and O2 effervescence
(vide infra).
The morphology of films from Pi, MePi and Bi electrolytes (Co-Pi, Co-MePi and Co-Bi,
respectively) has been analyzed by scanning electron microscopy. Depositions from MePi
electrolyte were conducted from quiescent solutions. Progressively thicker films are observed at
longer deposition times. Early in the course of electrolysis, a film of uniform ~1 μm thickness is
observed upon passage of 6 C/cm2 (Figure S2). Prolonged electrolysis (passage of 40 C/cm2)
produces a film ~3 μm thick with the concomitant formation of spherical nodules of 1 to 5 μm in
diameter on the surface of the film (Figure 3a, inset). These morphological features are similar to
those of films deposited from Pi electrolyte.6 Depositions from Bi electrolyte under quiescent
conditions lead to a rapid decrease of current arising from local pH gradients and associated
resistive losses due to the formation of neutral H3BO3 species (Figure S8). As such, bulk
electrolyses in Bi electrolyte were conducted with stirring, whereupon stable currents were
observed for hours, as shown in Figure 3b. Unlike Co-Pi or Co-MePi, Co-Bi displays a
somewhat different surface morphology. Spherical nodules appear early in the course of
deposition (upon passage of 2 C/cm2, Figure S3) and merge into larger aggregates upon
4
prolonged electrolysis, as shown in the inset of Figure 3b. SEM images of Co-Bi films grown
from quiescent solutions also reveal similar morphological features.
Powder x-ray diffraction patterns of Co-MePi and Co-Bi exhibit only broad amorphous
features and no detectable crystallites besides those corresponding to the ITO substrate (Figure
S4). In line with this observation, transmission electron microscopy does not reveal crystalline
domains nor are electron diffraction spots observed on a length scale of 5 nm (Figure S5). The
chemical compositions of the films were determined by elemental analysis and energy dispersive
x-ray analysis (EDX). Catalyst films were electrodeposited on large surface area electrodes (as
large as 20 × 20 cm) and the catalyst was removed from the surface to furnish up to 100 mg of
black powder whose composition was determined by microanalysis. The mole ratios of the
species present in the film for all deposition conditions attempted are shown in Table 1.
Regardless of Co2+ concentration, Co-MePi films exhibit a Co:P ratio of 4.5:1, while lowering
the pH to 7.0 caused a slight increase in the Co:P ratio. The elemental analysis data is
corroborated by EDX analysis, which reveals Co:P ratios of 4:1 to 6:1 (Figure S6) for films that
ranged in thickness from ~100 nm to >3 μm as well as for those prepared using Co2+
concentrations ranging from 0.1 mM to 10 mM. For Co-Bi films, a Co:B ratio of ~10:1 is
observed by elemental analysis.
Water Oxidation Catalysis and Activity. Mass spectrometry establishes that gas
effervescence from the electrode is a result of O2 production from water. The provenance of O2
was determined by using 18O-labeled water. MePi enriched with 18.9%
18
OH2 was chosen as a
representative electrolyte. Evolved gases were detected in real-time by mass spectrometry
(Figure 4). Signals for all three isotopes of O2 rise from their baseline levels minutes after the
onset of electrolysis and then slowly decay after electrolysis is terminated and O2 is purged from
the head space. Notwithstanding, the same ratio of isotopes is preserved throughout. The
observed isotopic ratio of 66.0:30.4:3.6 =
16,16
predicted statistical ratio of 65.8:30.6:3.6 =
O2:18,16O2:18,18O2 is in good agreement with the
16,16
O2:18,16O2:18,18O2. A small amount of CO2
(~0.5%) is also observed, suggesting the oxidation of MePi to Pi within the catalyst film. In line
5
with this contention, a
31
P NMR spectrum of dissolved films of the catalyst shows a
phosphate:methylphosphonate ratio of ~3:1 (Figure S9). The oxidation of MePi within the film
is also reflected by a P:C ratio of ~2:1 as determined by microanalysis. The lower carbon content
observed by 31P NMR relative to elemental analysis is attributed to the longer electrolysis time
that was used for the NMR study. The issue of CO2 evolution is obviated upon the replacement
of MePi electrolyte with Bi or Pi electrolytes.
Whereas MePi is partially degraded within the film, NMR of the MePi electrolyte
solution does not reveal decomposition of the electrolyte under prolonged electrolysis. Bulk
electrolysis was conducted using an ITO anode and a 1 mM Co2+ solution. A charge of 87 C (1.8
eq vs. MePi; 180 eq vs. Co2+) was passed and the solution from the working compartment of the
electrochemical cell showed a single 31P signal at 24.86 ppm (pH = 6.3) and an 1H signal at 1.22
ppm (J = 16.5 Hz). These values are similar to that of fresh electrolyte, which exhibits a
31
P
signal at 21.76 ppm (pH = 8.0) and a 1H signal at 1.05 ppm (J = 15.5 Hz). No other major signals
are observed in the NMR spectrum of solutions from either the working or auxiliary
compartment (see Figure S10).
The Faradaic efficiency of catalysts was determined by fluorescence based O2 sensing of
the evolved gases. In a bulk electrolysis using MePi, the amount of O2 produced (145 μmol)
accounted for 98(±5)% of the current passed (57 C; 148 μmol). For a Bi electrolyte, the amount
of O2 produced (135 μmol) accounted for 104(±5)% of the current passed (50 C; 130 μmol).
The log of current versus overpotential relationship (Tafel plot) was used to evaluate the
activity of catalysts grown from MePi and Bi electrolytes. Tafel plots data for Co-Pi, Co-MePi
and Co-Bi catalysts in their native electrolytes (Figure 5) are similar in slope indicating that each
of the electrolyte environments is equally competent at shuttling protons during catalytic
turnover to sustain high activity.
Catalyst Electrodeposition and Activity in Non-buffering Electrolytes. To assess the
role of the electrolyte in catalyst formation and activity, experiments were performed in solutions
containing Co2+ and electrolytes that are poor proton acceptors (e.g., SO42–, NO3–, ClO4–). CVs
6
of a glassy carbon working electrode in 0.1 M K2SO4 at pH 7.0, containing varying
concentrations of Co2+ are shown in Figure 6. The first and fifth CV scans, taken without pause,
are displayed along with the correspond traces of a 0.5 mM Co2+ solution in Pi electrolyte. The
CV traces of 0.5 mM Co2+ in the 0.1 M
K2SO4 solution are indistinguishable from the
background scan in the absence of Co2+ whereas a slight current enhancement over background
is observed at 1.56 V from 5 mM Co2+ solutions. At 50 mM Co2+, a pronounced anodic wave,
with an onset of 1.40 V, is observed. At this concentration, the return scan exhibits a small
cathodic wave at Ep,c = 1.15 V. CVs recorded on Co2+ in K2SO4 solution exhibit slightly
diminished currents on subsequent scans, contrasting those recorded in Pi electrolyte solution
from which pronounced current enhancements are observed upon subsequent scanning. The
same behavior is observed when 0.1 M NaClO4, pH 7.0, is substituted for K2SO4 as the
electrolyte (Figure S11). Hence, in electrolytes that are poor proton acceptors, catalyst formation
does not occur for Co2+ ion at modest concentrations.
Co-based films electrodeposit from unbuffered electrolyte (SO42–, NO3–, ClO4–) solutions
containing high concentrations of Co2+ ion (Co-X films).10 A film forms on a nickel foil
substrate11 upon controlled current electrolysis (ia = 8 mA/cm2) of 500 mM Co(SO4) in reagent
grade water in a three electrode single compartment cell.12 Upon conclusion of electrolysis, the
working electrode was placed in fresh electrolyte solution (0.1 M K2SO4, pH 7.0) containing no
Co2+. Electrolysis was initiated with stirring for 1 hr at 1.3 V vs NHE using the standard two
compartment cell separated by a glass frit (as used for all previously described experiments). The
current density traces obtained over this time are displayed in Figure 7. The current rapidly
declines to 70 μA/cm2 after one minute and continues to diminish over the course of electrolysis
to 36 μA/cm2 after 1 hour. For side-by-side comparison, a catalyst film was prepared on a nickel
foil substrate by controlled potential electrolysis (1.40 V) of a 0.5 mM Co2+ in Pi electrolyte
solution. Upon conclusion of electrolysis, the electrode was placed in fresh Pi electrolyte
solution containing no Co2+. Electrolysis was initiated for 1 hr at 1.3 V vs NHE and the same
electrode geometry and stir rate was used as chosen for electrolysis in unbuffered solution. The
7
current density trace is shown in Figure 7. Unlike Co-X systems, the current of the Co-Pi system
remains stable at ~1 mA/cm2 over the entire course of the electrolysis.
Electrolytes possessing poor buffering capacity lead to diminished activity (vide supra)
and to large pH gradients across a two-compartment cell. The pH drop may be circumvented by
utilizing a single compartment cell for water oxidation. To assess the Faradaic efficiency of a
single compartment setup, a Co-X film prepared from 500 mM CoSO4 solutions as described
above was electrolyzed using a three electrode configuration in a single compartment cell
containing 0.1 M K2SO4 at pH 7.0. Evolved O2 was detected by direct fluorescence-based
sensing (Figure 8). Throughout the course of electrolysis, the amount of O2 evolved is
significantly attenuated relative to the amount of O2 expected on the basis of 100% Faradaic
efficiency.
Water Oxidation from Salt Water. Catalyst function does not require pure water.
Controlled potential electrolysis of a Co-Pi film at 1.3 V in Pi electrolyte containing 0.5 M NaCl
reveals sustained current densities greater than 0.9 mA/cm2 (Figure S12). These current densities
are comparable to those observed in the absence of NaCl, suggesting that chloride anions do not
inhibit O2 evolving catalysis (vide infra). EDX analysis of a film used for prolonged (16 h, 76.5
C passed) electrolysis in the presence of 0.5 M NaCl reveals that Co and P are retained in a ratio
similar to that of the parent film.6 In addition, EDX analysis also indicates significant
incorporation of Na+ ion, but only minimal incorporation of Cl– (Na:Cl = ~6:1), suggesting
significant exchange of Na+ ion for K+ ion (Figure 9).
Noting the stability of the film in chloride-containing electrolyte, we quantified the
Faradaic efficiency of water oxidation in this medium using fluorescence-based sensing of
evolved O2. Figure 10 shows the amount of oxygen produced at 1.30 V vs. that expected for O2
production with 100% Faradaic efficiency. The observed O2 signal rises shortly after initiation of
electrolysis as oxygen saturates the solution and fills the headspace, and hence the offset. The
observed O2 signal rises throughout the electrolysis (15 h) and plateaus upon termination of
electrolysis at a value in accordance with the net current passed in the experiment (35.3 C, 91.4
8
μmol O2). These results show that water oxidation to O2 predominates (100±5%) from salt
solutions. This property of the system is further corroborated by direct quantification of oxidized
chloride species (HOCl and OCl–). A Co-Pi film was operated in the presence of 0.5 M NaCl for
16 h (76.5 C passed) at 1.30 V and then the solution was analyzed for hypochlorite using a
standard N,N-diethyl-p-phenylenediamine titrimetric assay.13 We observe 9.3 μmol of oxidized
chloride species, which accounts for 1.80 C or 2.4% of the total current passed in the experiment.
To exclude the possibility of Cl2 production in this medium, the evolved gases were analyzed in
real time by an in-line mass spectrometer. The only gas detected was O2 and no isotopes of Cl2
rose above the baseline level during the course of the experiment (6 h) (Figure S13).
Discussion
Electrolyte is a crucial determinant in the formation, activity and selectivity of selfassembled cobalt-based electrocatalysts for water oxidation. In the absence of suitable
electrolytes, the generation of oxygen at appreciable activities from neutral water under ambient
conditions cannot be achieved.
Large catalytic waves for water oxidation are observed from CVs of low concentrations
of Co2+ (0.5 mM Co2+) in solutions of Pi, MePi or Bi electrolytes. Prior to the onset of catalytic
current (Figure 1), an anodic wave is observed in the CV that is consistent with a Co3+/2+ couple.
The observed potential for this couple is well below that of Co(OH2)63+/2+ (1.86 V) but is similar
to the 1.1 V potential estimated for the Co(OH)2+/0 couple.14 As Figure 2 shows, the catalytic
wave is preserved upon the placement of the once anodically scanned electrode in a fresh
electrolyte solution containing no Co2+ cation. Polishing the electrode restores a clean
background in the CV indicating that a catalytically competent species electrodeposits
immediately following oxidation of Co2+ to Co3+ at modest potentials. This behavior is in sharp
contrast to CV traces obtained from Co2+ in electrolytes of poor proton-accepting abilities. In
electrolytes such as SO42– and ClO4– (Figures 6 and S11), no electrochemical features of
significance are observed above background for solutions containing 0.5 mM Co2+. Only when
9
the Co2+ ion concentration is increased by 2 orders of magnitude is a slight enhancement in
current observed near the solvent window at 1.56 V. This current enhancement is anodically
shifted >150 mV relative to the corresponding wave in Pi at drastically lower Co2+
concentration. Electrolyte promotes catalyst formation; in the absence of an effective proton
acceptor, at a given pH, the formation of a catalyst film is significantly inhibited.
Whereas an active catalyst can be generated on an anodic single scan, films of desired
thickness may be prepared on conducting electrodes (metal or semiconductor) by controlled
potential electrolysis of 0.5 mM Co2+ solutions of Pi, MePi and Bi. The bulk composition of the
films, as determined by elemental analysis and corroborated by EDX, reveals a lower Co:P ratio
for Co-MePi (~5:1) relative to the Co:P ratio of ~2:1 observed for Co-Pi. An even lower anion
content is observed for Co-Bi which exhibits a Co:B ratio of 10:1. We note that the anion
composition is balanced by a monovalent cation in all cases, regardless of the Co to anion ratio.
The disparate anion incorporation into the bulk material is not reflected in altered activity,
suggesting that a common Co-oxide unit effects catalysis in all films. The active unit is <5nm in
dimension as evidenced by the absence of crystalline features in the power X-ray diffraction
pattern (Figures S4) and diffraction patterns in the TEM (Figure S5). This is in contrast to the
structural properties of Co-X materials, which are asserted10 to exhibit long range ordering
corresponding to CoOx crystallites.
An electrolyte environment that has good proton-accepting properties is required for
sustained catalyst activity at appreciable current density. The Pi electrolyte is an efficient proton
carrier and it preserves a stable local pH environment required for high catalytic activity. In
addition, it functions as an acceptor of the protons furnished from water oxidation, and
participates in the PCET activation of oxygen.15 Alternative electrolytes are able to support
catalysis as long they have sufficient proton-accepting capacity in the pH regime of interest and
are stable under the conditions of catalysis. MePi and Bi electrolytes meet these criteria at pH
9.2 and 8.0, respectively. Co-MePi and Co-Bi films support catalytic activity comparable to that
observed for Co-Pi, as demonstrated by their associated Tafel behaviors (Figure 5). We envision
10
other oxidation resistant buffers would also function in a similar capacity enabling robust water
oxidation catalysis over a large pH range. MePi offers an advantage in being able to sustain
higher concentrations of Co2+ relative to either Pi or Bi. Despite the low concentration of PO43–
expected in Pi electrolytes at pH 7.0, the low solubility of Co3(PO4)2 (Ksp = 2.05 × 10–35) elicits
the slow precipitation of Co2+ from solution.16 Precipitation is instead averted in MePi
electrolyte, where Co2+ is indefinitely soluble at 2 mM.
In the absence of efficient proton-accepting electrolytes (e.g., SO42–, NO3–, ClO4–) proton
buildup results in dramatically reduced current densities that decay over time. In solutions of
these counteranions, the Co-oxide catalyst is the best base and consequently the catalyst is
subject to corrosion induced by the protons produced from water splitting. In addition, catalyst
activity is inhibited by changes in local and bulk pH. Bulk pH can reach a steady-state value in a
single compartment cell. However, as expected, and verified by the data in Figure 8, significant
short circuit current from redox cycling between the anode and cathode stifles oxygen
production. For these reasons, a sustainable water-splitting reaction cannot be achieved by Cobased catalysts in the absence of proton-accepting electrolytes.
The ability of the electrolyte to maintain the pH during water oxidation is manifested in a
robust and functional catalyst in the presence of 0.5 M NaCl. Direct measurement of Faradaic
efficiency and titrimetry of chloride oxidation products establishes that Co-Pi is able to produce
oxygen from salt water at current efficiencies commensurate with those observed for pure water.
At pH 7.0, the HOCl/Cl– redox process has a thermodynamic potential of 1.28 V, 0.46 V beyond
the thermodynamic potential for water oxidation to O2. With decreasing pH, the oxidation of Cl–
becomes more thermodynamically competitive with water oxidation. As such, in the absence of
proton-accepting electrolytes (such as Co-X), chloride oxidation will interfere with water
oxidation. The ability of the Pi electrolyte to preserve the pH of the solution allows O2
production to out-compete Cl– oxidation.
11
Conclusion
The observations reported herein highlight several attractive properties of this new
oxygen evolving catalyst for water oxidation under benign conditions. High activities for water
oxidation demand the involvement of a proton-accepting electrolyte. The electrolyte facilitates
catalyst formation, allows for high activity for water oxidation, and promotes catalyst
preservation during turnover. By maintaining pH with the electrolyte, the catalyst is able to
produce oxygen from high concentrations of salt water at current efficiencies commensurate with
those observed for pure water.
Water splitting is a solar energy storage mechanism of sufficient scale to address future
global energy needs.17 As we have emphasized, the conditions under which water splitting is
performed will determine how solar energy is deployed.6 Commercial electrolyzers are
extremely efficient and operate at current density as high as 1 A/cm2. However, commercial
electrolyzers operate under harsh conditions and accordingly they are difficult to maintain and
costly to engineer. Moreover, the current density of commercial electrolyzers is at variance with
many applications of non-concentrated solar energy. The cobalt catalyst described herein is
better matched to the 10-20 mA/cm2 output of a conventional photovoltaic device that provides
the required voltage for water-splitting.18 With an improved cell design, these current densities
will be achieved with the cobalt catalyst. Therefore, the CoPi catalyst is well adapted for the
design of inexpensive and highly manufacturable water-splitting systems. These systems extend
beyond an electrolyzer. We demonstrate here that an active catalyst forms immediately following
oxidation of Co2+ in solution suggesting that the cobalt catalysts are amenable to rapid, ultra-thin
film electrodeposition on a wide array of substrates with complicated geometries and large
surface areas, such as those involving nanostructured semiconducting materials.19-21 Of added
potential, since water splitting is not performed in highly acidic or basic conditions, the catalyst
is amenable to integration with charge-separating networks comprising protein,22,23 organic and
inorganic24-26 constituents. In these systems, the one-photon, one-electron charge separation can
12
be accumulated by the catalyst to attain the four equivalents needed for water splitting. The ease
of implementation of the catalyst with a diverse array of substrates suggests that the catalyst will
be of interest to many in their endeavors to store solar energy by water splitting.
Acknowledgements. We thank Drs. Bruce Brunschwig, Harry Gray, Nathan Lewis and Bruce
Parkinson for their thoughtful comments. This research was supported by a Center for Chemical
Innovation of the National Science Foundation (Grant CHE-0533150) and a grant from the
Chesonis Family Foundation. Grants from the NSF also provided instrument support to the DCIF
at MIT (CHE-9808061, DBI-9729592). Y.S. gratefully acknowledges the Department of
Defense for a predoctoral fellowship. We thank M. W. Kanan for collecting powder X-ray
diffraction spectra and for many productive discussions.
Supporting Information Available. Full experimental details, additional SEM and TEM
images, EDX and 31P spectra, and MS traces. This information is available free of charge via the
Internet at http://pubs.acs.org.
13
References
(1)
Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 15729.
(2)
Dempsey, J. L.; Esswein, A. J.; Manke, D. R.; Rosenthal, J.; Soper, J. D.; Nocera, D. G.
Inorg. Chem. 2005, 44, 6879.
(3)
Eisenberg R.; Gray H. B. Inorg. Chem. 2008, 47, 1697.
(4)
Betley, T. A.; Wu, Q.; Van Voorhis, T.; Nocera, D. G. Inorg. Chem. 2008, 47, 1849.
(5)
Turner, J. A. Science 1999, 285, 5428
(6)
Kanan, M. W.; Nocera, D. G. Science 2008, 321, 1072.
(7)
Tarasevich, M. R.; Efremov B. N. In Electrodes of Conductive Metal Oxides; Trasatti, S.,
Ed.; Elsevier: Amsterdam, 1980; Chapter 5.
(8)
Yagi, M.; Tomita, E.; Sakita, S.; Kuwabara, T.; Nagai, K. J. Phys. Chem. B. 2005, 109,
21489.
(9)
Hara, M.; Waraksa, C. C.; Lean, J. T.; Lewis, B. A.; Mallouk, T. E. J. Phys. Chem. A
2000, 104, 5275.
(10) Suzuki, O.; Takahashi, M.; Fukunaga, T.; Kuboyama, J. U.S. Patent 3,399,966, Sep. 3,
1968.
(11) Nickel was chosen as the substrate because Co-X films exhibited more robust adhesion to
the Ni foil substrate relative to ITO.
(12) Deposition from a standard two compartment cell using 500 mM Co2+ resulted in
significant Co2+ precipitation in the auxiliary chamber as a result of pH increases
associated with H2 evolution at the cathode (see Figure S14).
(13) Eaton, A. D.; Clesceri, L. S.; Rice, W. E.; Greenberg, A. E. Standard Methods for the
Examination of Water and Wastewater, 21st ed.; American Public Health Association,
14
American Water Works Association, Water Pollution Control Federation: Washington,
DC, 2005; Ch. 4.
(14) Brunschwig, B. S.; Chou, M. H.; Creutz, C.; Ghosh, P.; Sutin, N. J. Am. Chem. Soc. 1983,
105, 4832.
(15) Irebo, T.; Reece, S. Y.; Sjödin, M.; Nocera, D. G.; Hammarström, L. J. Am. Chem. Soc.
2007, 129, 154622.
(16) Lide, D.R. CRC Handbook of Chemistry and Physics, 77th ed.; CRC Press: Boca Raton,
1996; pp 8-91.
(17) Nocera, D. G. ChemSusChem, 2009, accepted for publication.
(18) Green, M. A.; Emery, K.; Hishikawa, Y.; Warta, W. Prog. Photovolt. Res. Appl. 2009, 17,
85.
(19) Kay, A.; Cesar, I.; Grätzel, M. J. Am. Chem. Soc. 2006, 128, 15714.
(20) Maiolo, J. R. III; Kayes, B. M.; Filler, M. A.; Putnam, M. C.; Kelzenberg, M. D.; Atwater,
H. A.; Lewis, N. S. J. Am. Chem. Soc. 2007, 129, 12346.
(21) Yang, F.; Forrest, S. R. ACS Nano 2008, 2, 1022.
(22) Shih, C.; Museth, A. K.; Abrahamsson, M.; Blanco-Rodriguez, A. M.; Di Bilio, A. J.;
Sudhamsu, J.; Crane, B. R.; Ronayne, K. L.; Towrie, M.; Vlcek, A., Jr.; Richards, J. H.;
Winkler, J. R.; Gray, H. B. Science 2008, 320, 1760.
(23) Stubbe, J.; Nocera, D. G.; Yee, C. S.; Chang, M. C. Y. Chem. Rev. 2003, 103, 2167.
(24) Fukuzumi, S. Phys. Chem. Chem. Phys. 2008, 10, 2283.
(25) Hambourger, M.; Moore, G. F.; Kramer, D. M.; Gust, D.; Moore, A. L.; Moore, T. A.
Chem. Soc. Rev. 2009, 38, 25.
(26) Flamigni, L.; Collin, J.-P.; Sauvage, J.-P. Acc. Chem. Res. 2008, 41, 857.
15
Table. 1. Elemental composition of catalyst films.
Deposition Conditions
Co
P
Na
C
MePi, pH 8.0, 1 mM Co2+
4.5
1
1.2
0.6
MePi, pH 8.0, 10 mM Co2+
4.5
1
0.9
0.8
MePi, pH 7.0, 10 mM Co2+
5.6
1
0.6
0.7
Bi, pH 9.2, 0.5 mM Co2+
9.5
Pi, pH 7.0, 0.5 mM Co2+
2.7
1
16
B
K
1
1.0
1.0
Figure Captions
Figure 1. Cyclic voltammogram using a glassy carbon working electrode, 50 mV/sec scan rate,
of aqueous 0.5 mM Co2+ in 0.1 M Pi electrolyte, pH 7.0 (▬▬▬), 0.1 M MePi electrolyte, pH
8.0 (▬ ▬ ▬), and 0.1 M Bi electrolyte, pH 9.2 (▪▪▪▪▪▪▪). Background traces in each electrolyte
medium in the absence of Co2+ are overlaid. Inset shows CVs in the presence of Co2+ on an
expanded current scale.
Figure 2. Cyclic voltammogram using a glassy carbon working electrode, 50 mV/sec scan rate,
in 0.1 M MePi electrolyte, pH 8.0 (▬ ▬ ▬), and 0.1 M Bi electrolyte, pH 9.2 (▪▪▪▪▪▪▪), with no
Co2+ present after 1 scan in the presence of 0.5 mM Co2+. Background scan of a freshly polished
electrode is overlaid. Inset shows CVs on an expanded current scale.
Figure 3. Current density trace for bulk electrolysis at 1.30 V in 0.1 M MePi electrolyte, pH 8.0,
1 mM Co2+ on an ITO anode (3a) and for bulk electrolysis at 1.30 V in 0.1M Bi electrolyte, pH
9.2, 0.5 mM Co2+ (3b). Insets: SEM images (30° tilt) of catalyst films electrodeposited from
MePi electrolyte (3a) and Bi electrolyte (3b).
Figure 4. Mass spectrometric detection of isotopically-labeled
16,16
O2 (▬),
16,18
O2 (▬),
18,18
O2
(▬), and CO2 (▬) during electrolysis of a catalyst film on ITO in 0.1 M MePi electrolyte, pH
8.0, containing 18.9%
18
OH2. Green and red arrows indicate start and end of electrolysis,
respectively. Inset: Percent abundance of each isotope over the course of the experiment.
Average observed abundance ±2σ indicated above each line. Statistical abundances: 65.8%,
30.6%, and 3.6%.
Figure 5. Tafel plot, η = (Vappl – IR – E°), of a catalyst film deposited from and operated in 0.1
M Pi electrolyte, pH 7.0 (●), 0.1 M MePi electrolyte, pH 8.0 (■), and 0.1 M Bi electrolyte, pH
9.2 (▲).
17
Figure 6. Cyclic voltammogram using a glassy carbon working electrode, 50 mV/sec scan rate,
of 0.1 M K2SO4 electrolyte, pH 7.0, containing 0 mM Co2+ (▬▬▬), 0.5 mM Co2+ (▬ ▬ ▬), 5
mM Co2+ (▪▪▪▪▪▪▪), and 50 mM Co2+ (▬ ▪ ▬ ▪). CV traces of glassy carbon working electrode,
50 mV/sec scan rate, of 0.5 mM Co2+ in Pi electrolyte, pH 7.0 (▬ ▪ ▪ ▬) are shown for
comparison. Vertical arrows indicated progression between the first and fifth scans taken without
pause. Inset shows all CVs on an expanded current and potential scale.
Figure 7. Controlled potential electrolysis at 1.3 V of a catalyst film operated in 0.1 M Pi
electrolyte, pH 7.0 (▬▬▬), and 0.1 M K2SO4, pH 7.0 (▪▪▪▪▪▪▪). Catalysts deposited from Pi
electrolyte and K2SO4 electrolyte respectively (see SI).
Figure 8. Faradaic efficiency of a single compartment bulk electrolysis at 0.5 mA/cm2 constant
current. O2 detected by fluorescence sensor (▬▬▬) and theoretical O2 trace assuming 100%
Faradaic efficiency (▪▪▪▪▪▪▪). Green and red arrows indicated start and end of electrolysis,
respectively. Conditions: 0.1 M K2SO4 electrolyte, pH 7.0. Catalyst prepared from 0.1 M K2SO4
electrolyte, pH 7.0, 500 mM Co2+.
Figure 9. EDX histogram of a catalyst film after 16 hrs (76.5 C passed) of electrolysis in 0.5 M
NaCl, 0.1 M KPi electrolyte, pH 7.0. Vappl= 1.30 V; cps = counts per second. Catalyst prepared
from 0.1 M KPi electrolyte, pH 7.0, 0.5 mM Co2+.
Figure 10. O2 detected by fluorescence sensor (▬▬▬) and theoretical O2 trace assuming 100%
Faradaic efficiency (▪▪▪▪▪▪▪). Green and red arrows indicated start and end of electrolysis,
respectively. Conditions: 0.1 M KPi electrolyte, pH 7.0, 0.5 M NaCl, Vappl = 1.30 V, no
compensation for IR drop. Catalyst prepared from 0.1 M KPi electrolyte, pH 7.0, 0.5 mM Co2+.
18
Figure 1
19
Figure 2
20
Figure 3
21
Figure 4
22
Figure 5
23
Figure 6
24
Figure 7
25
Figure 8
26
Figure 9
27
Figure 10
28
TOC
29
A Self-Healing Oxygen-Evolving Catalyst
Daniel A. Lutterman, Yogesh Surendranath, and Daniel G. Nocera*
Department of Chemistry, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA, 02139
RECEIVED DATE (automatically inserted by publisher); E-mail: [email protected]
Metal centers of catalysts activate many small molecules,
especially those pertaining to renewable energy, by mediating
multielectron transformations of the substrate. To accommodate
these reactions without imposing prohibitively high-energy
barriers, the primary coordination environment of the metal center
is typically reorganized and ligand exchange is common.
Accordingly, catalysts that effect multielectron transformations
are prone to structural rearrangement, and hence instability,
during turnover. For this reason, the design of catalysts that repair
themselves has been identified as a primary challenge for energy
science. 1,2 The need for repair is particularly germane to water
oxidation catalysts, which must mediate a four-electron, fourproton exchange under highly oxidizing conditions. 3 Recently,
we reported the electrodeposition of a cobalt-based oxygen
evolving catalyst from phosphate electrolyte and other protonaccepting electrolytes. 4,5 Molecular mechanisms involving
O2/H2O cycles at cobalt centers suggest the involvement of Co2+,
Co3+ and likely Co4+ oxidation states during catalysis. 6,7 Co2+ is a
high spin ion and is substitutionally labile whereas Co3+ and
higher oxidation states are low spin and substitutionally inert in
an oxygen-atom ligand field. 8 As the propensity of metal ion
dissolution from solid oxides has been shown to correlate with
ligand substitution rates, 9 the cobalt oxygen-evolving catalyst is
expected to be structurally unstable during turnover. To probe the
dynamics of catalyst during water-splitting, we now report the
electrosynthesis of the catalyst using radioactive 57Co and 32P
isotopes. By monitoring these radioactive isotopes during watersplitting catalysis, we show that the catalyst is self-healing and
that phosphate is responsible for repair.
The cobalt-phosphate water oxidation catalyst (Co-Pi) forms in
situ upon the application of a potential of 1.3 V vs NHE to an ITO
or FTO electrode immersed in a 0.1 M phosphate (pH = 7.0)
electrolyte (Pi) containing 0.5 mM Co2+.4 At this potential, Co2+
is oxidized to Co3+ and an amorphous catalyst deposits on the
electrode that incorporates phosphate as a major constituent. The
characteristics of the deposition and the properties of the catalyst
prepared in this manner have been described previously.4,5
For the studies described here, a Pi solution containing 0.5 mM
Co(NO3)2 was enriched with 10 mCi of 57Co(NO3)2. Details of the
sample preparation and handling are provided in the Supporting
Information. After deposition, catalyst films were washed with Pi
to remove adventitious 57Co2+ ion (see Supporting Information).
Two separate electrodes coated with the catalyst were placed in
the working compartment of two different electrochemical Hcells containing Co-free Pi electrolyte. A potential of 1.3 V was
applied to one electrode and no potential bias was applied to the
other; the catalyst is active on the potential biased electrode, and
water-oxidation catalysis proceeds as previously described.4,5
Aliquots of the electrolyte were removed from the H-cell at
different time points and the radioactivity was quantified for each
aliquot at the conclusion of the experiment. The total available
57
Co was determined by acidifying the electrolyte with
Figure 1. Percentage of 57Co leached from films of the Co-Pi catalyst on
an electrode: with a potential bias of 1.3 V (NHE) (■) turned on and off at
the times designated; and without an applied potential bias (●). Lines
added to figure simply as a guide to the eye.
concentrated HCl to dissolve the catalyst completely and
accounting for the radioactivity of removed aliquots (see
Supporting Information). Figure 1 plots the amount of 57Co that
leached from the catalyst film as a percentage of the total
available 57Co. Cobalt was continually released from the catalyst
film on the electrode with no potential bias; after 39 hrs, ~1.5% of
the cobalt ion is detected in solution. Conversely, no cobalt is
observed in the electrolyte solution when the electrode is held at
1.3 V. After the potential bias is removed from the electrode at 4
hrs, cobalt promptly dissolves from the catalyst. Re-absorption of
the cobalt is observed upon the re-application of the potential to
the electrode at 15 and 25 h at which time the cobalt ion
concentration in solution is 1.8 μM and 1.0 μM, respectively.
Cobalt uptake is complete with continuous application of a
potential bias; after 14 hours, only 0.002% cobalt remains in solution. The results of Figure 1 are consistent with (i) the slow
liberation of Co2+ from the catalyst in the absence of an applied
potential and (ii) re-oxidation of the liberated Co2+ to re-form the
catalyst when the 1.3 V potential is re-applied.
Given the dynamic behavior of cobalt in the catalyst, the other
major constituent of the catalyst, phosphate, was monitored by
means of a 32P-phosphate label. Simultaneous electrodepositions
of the catalyst were performed on two electrodes immersed in a Pi
solution of 0.5 mM Co(NO3)2 that was enriched with 1.5 mCi of
32
P-orthophosphoric acid. Catalyst films were washed and then
placed in two different electrochemical H-cells containing Pi.
Figure 2a shows that 32P-phosphate leaches from a catalyst film
with no applied potential at double the rate for a film held at 1.3
V. The same trend is observed for phosphate incorporation into
the catalyst film. Eight ITO electrodes were arranged in a
concentric arrangement within the working electrode compartment of the H-cell (see Figure S1) and the catalyst was electrodeposited from non-isotopically enriched Pi solution. After
deposition, the electrodes were separated into two groups of four,
and arranged in a concentric array. The two sets of electrodes
Figure 2. Plots monitoring: (a) 32P leaching from Co-Pi catalyst; and
(b)32P uptake by the Co-Pi catalyst on an electrode with an applied
potential of 1.3 V (NHE) (■, ) and without an applied potential (●, ).
were immersed in individual H-cells containing Pi electrolyte that
was enriched with ~1.5 mCi of 32P-phosphate. One group of
electrodes was held at a potential of 1.3 V and the other had no
applied potential. Every hour one electrode was removed from
each H-cell, washed, and the catalyst was dissolved with
concentrated HCl. Figure 2b plots the total 32P activity obtained at
each time point. Consistent with the results of Figure 2a, more
phosphate exchange is observed for the electrodes under no
applied potential. Elemental analysis of catalyst films establishes
that the phosphate anion composition is balanced by an alkali
cation (Na+ or K+).5 In contrast to the slow exchange of
phosphate, >90% exchange of Na for K (or K+ for Na+) is
observed after 10 min of catalyst operation in the alternate
electrolyte medium (Table S2). These data together suggest that
the phosphate is coordinated to cobalt since a slower ligand
exchange would be expected for Co3+, which predominates on the
electrode held against a potential bias. In addition, the much
higher exchange of phosphate as compared to cobalt suggests that
the metal ion is a constituent of a more robust metal-oxygen
framework.
In the absence of proton accepting electrolytes (X = SO42–,
NO3–, ClO4–), catalyst dissolution is rapid and irreversible. Cobased films (Co-X) electrodeposit from unbuffered electrolyte
solutions containing high concentrations of Co2+ ion. 10 A film
was electrodeposited onto an ITO electrode from a solution of 25
mM Co(NO3)2 containing 2 mCi of 57Co(NO3)2 in 0.1 M K2SO4
(pH = 7.0) at 1.65 V. 57Co dissolution measurements and assays
were performed with a procedure analogous to that employed for
Figure 1 (see Supporting Information). At 1.3 V, the initial
sustained current densities were <0.1 mA/cm2. A potential of 1.5
V was applied to Co-X films to achieve current densities (~1
mA/cm2) comparable to those of Co-Pi operated at 1.3 V. The
data in Figure 3 deviates significantly from that in Figure 1.
Whereas an applied potential led to cobalt uptake for Co-Pi, the
same potential applied to the Co-X system leads to enhanced
cobalt release relative to an unbiased electrode. Moreover, cobalt
dissolution increases with increased applied potential. These
results are consistent corrosion of the Co-X system. In the
absence of a proton accepting electrolyte, the best proton acceptor
is the electrodeposited Co-X film itself. With increased potential,
increased production of protons engenders hastened corrosion of
these films.
A repair mechanism is not established in the absence of
phosphate or other proton-accepting electrolyte (e.g., borate,
methylphosphonate).5 This contention is demonstrated by adding
phosphate to the corroding film of Figure 3. Addition of KPi
electrolyte (1 M, pH = 7.0) to attain a final concentration of 0.1 M
Pi leads to a rapid re-deposition of cobalt into the catalyst film
(no precipitation of cobalt is observed, see Supporting
Information).
Figure 3. Percentage of 57Co leached from Co-X films on an electrode
under a potential bias of 1.3 V (●) and 1.5 V (■) (NHE) and an unbiased
electrode (▲). Pi was added at the time points indicated by the arrows.
The results reported here establish that phosphate is the crucial
component in the self-healing of the Co-Pi catalyst. As originally
surmised,4 in situ formation of the catalyst implies a pathway for
catalyst self repair. Any Co2+ formed and released into solution
during water-splitting catalysis will be re-deposited upon
oxidation to Co3+ in the presence of phosphate. Moreover,
catalyst degradation, in the absence of an applied potential, is
repaired when the potential is re-applied and phosphate is present
in solution. Thus phosphate ensures long-term stability of the
catalyst system. More generally, redox reactions of small
molecules such as H2O, O2, N2, and CO2, entail that the metal
centers of catalysts occupy multiple oxidation states. Because the
most stable ground state of a metal in widely varying oxidation
states often possesses very different coordination environments,
the same ligand field cannot stabilize the system across the entire
multielectron transformation. If a ligand field is imposed about
the metal, then excess overpotential will be introduced into the
redox cycle. By introducing a repair mechanism, the constraint of
a structurally stable catalytic center may be relaxed while
retaining functional stability at a lower overpotential.
Acknowledgment. We thank Matt Kanan and Mircea Dinca for
helpful discussions and Bill McCarthy for assistance in radiolabeling
experiments. This research was supported by a Center for Chemical
Innovation grant from the NSF (Grant CHE-0533150) and a grant
from the Chesonis Family Foundation. D.A.L. thankfully
acknowledges the Jane Coffin Childs Memorial Fund for Medicinal
Research for a postdoctoral fellowship. Y.S. gratefully acknowledges
the Department of Defense for a pre-doctoral fellowship (NDSEG).
Supporting Information Available: Full experimental details,
photographs of electrode arrays, and elemental composition data for
Na/K exchange experiments. This information is available free of
charge via the Internet at http://pubs.acs.org.
References
(1)
“New Science for a Secure and Sustainable Energy Future”; Basic
Energy Sciences Advisory Committee, U. S. Department of Energy:
Washington, DC, 2008.
(2) “Directing Matter and Energy: Five Challenges for Science and the
Imagination”; Basic Energy Sciences Advisory Committee, U. S.
Department of Energy: Washington, DC, 2007, Ch. 3.
(3) Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U.S.A. 2006, 103,
15729.
(4) Kanan, M. W.; Nocera, D. G. Science 2008, 321, 1072.
(5) Surendranath, Y.; Dinča, M.; Nocera, D. G. J. Am. Chem. Soc. 2009, in
press.
(6) Chang, C. J.; Loh, Z.-H.; Shi, C.; Anson, F. C.; Nocera, D. G. J. Am.
Chem. Soc. 2004, 126, 10013.
(7) Shafirovich, V. Y.; Khannanov, N. K.; Strelets, V. V. Nouv. J. Chim.
1980, 4, 81.
(8) Basolo, F.; Pearson, R. G. “Mechanisms of Inorganic Reactions”; Wiley
and Sons: New York, 1967.
(9) Casey, W. H. J. Coll. Inter. Sci. 1991, 146, 586.
(10) Suzuki, O.; Takahashi, M.; Fukunaga, T.; Kuboyama, J. U.S. Patent
3,399,966, Sept. 3, 1968.
A cobalt-phosphate water-oxidizing catalyst forms from the oxidation of Co2+ to Co3+ in the presence of phosphate. We have
employed radioactive 57Co and 32P isotopes to probe the dynamics of this catalyst during water-oxidation catalysis. We show
that the catalyst is self-healing and that phosphate is the crucial factor responsible for repair.
TUTORIAL REVIEW
www.rsc.org/csr | Chemical Society Reviews
Cobalt–phosphate oxygen-evolving compoundw
Matthew W. Kanan, Yogesh Surendranath and Daniel G. Nocera*
Received 3rd November 2008
First published as an Advance Article on the web 28th November 2008
DOI: 10.1039/b802885k
The utilization of solar energy on a large scale requires efficient storage. Solar-to-fuels has the
capacity to meet large scale storage needs as demonstrated by natural photosynthesis. This
process uses sunlight to rearrange the bonds of water to furnish O2 and an H2-equivalent. We
present a tutorial review of our efforts to develop an amorphous cobalt–phosphate catalyst that
oxidizes water to O2. The use of earth-abundant materials, operation in water at neutral pH, and
the formation of the catalyst in situ captures functional elements of the oxygen evolving complex
of Photosystem II.
Introduction
Modern day society relies on a continuous energy supply that
must be available day and night. Although solar energy is of
sufficient scale to meet future energy needs, it is diurnal.1
Consequently, solar energy will not be used as a large
scale energy supply for society unless it can be stored.
Unfortunately, most current methods of solar storage are
characterized by low energy densities and therefore present
formidable challenges for large scale solar implementation.
For instance, consider the energy densities by mass of
the following storage methods: compressed air (300 atm)
B0.5 MJ kg 1, batteries B0.1–0.5 MJ kg 1, flywheels
B0.5 MJ kg 1, supercapacitors B0.01 MJ kg 1, and water
pumped uphill (100 m) B0.001 MJ kg 1. Conversely, the
energy density of liquid fuels (B50 MJ kg 1) is 102 larger
than the best of the foregoing storage methods and H2
(700 atm) possesses an even greater energy density at
140 MJ kg 1. Indeed, society has intuitively understood this
disparity in energy density as it has developed over the last
century as all large scale energy storage in our society is in the
Department of Chemistry, 6-335, Massachusetts Institute of
Technology, 77 Massachusetts Avenue, Cambridge, MA 02139-4307,
USA. E-mail: [email protected]
w Part of the renewable energy theme issue.
Yogesh Surendranath, Daniel Nocera and Matthew Kanan
This journal is
c
The Royal Society of Chemistry 2009
form of fuels. But these fuels are carbon-based. The imperative
for the discipline of chemistry, and more generally science, is
to develop fuel storage methods that are easily scalable,
carbon-neutral and sustainable.
A fuel-forming reaction that meets this imperative is:
2H2O + light = 2H2 + O2
(1)
Obviously, light does not directly act on water to engender its
splitting into its elemental components. Hence, catalysts are
needed to effect the overall transformation. In nature, the
water-splitting reaction is accomplished by photosynthesis.2
Outside of the leaf, solar fuels other than hydrogen may be
produced with the protons and electrons extracted from water,
including the reduction of carbon dioxide to methanol. However, all water-splitting schemes require oxygen production
and the efficiency of this step is a primary impediment toward
realizing artificial photosynthesis.3
A key design element of photosynthesis is the separation of
the functions of light collection and conversion from catalysis.
Light is collected and converted by Photosystem II (PSII) into
a wireless current. The holes of this current are fed to the
oxygen-evolving complex (OEC) where water is oxidized to O2
and the electrons are fed to Photosystem I where additional
light capture occurs to provide sufficient reducing power for
the reduction of NADP+ to NADPH by ferredoxin:NADP+
Matthew Kanan (right) received his BA in Chemistry from Rice
University in 2000. He then moved on to Harvard where he
completed his PhD in 2005 under the mentorship of Professor
David Liu. Currently, he studies water oxidation at MIT as a
NIH Ruth Kirchenstein Postdoctoral Fellow.
Yogesh Surendranath (left) received his BS in Chemistry from
the University of Virginia in 2006. As a graduate student at
MIT, he examines water oxidation catalysis as a DoD NDSEG
Fellow.
Daniel Nocera (middle) is the Henry Dreyfus Professor of
Energy at the Massachusetts Institute of Technology. He
received his BS degree from Rutgers University in 1979 and
his PhD degree from Caltech in 1984. He studies the mechanisms of biological and chemical energy conversion.
Chem. Soc. Rev., 2009, 38, 109–114 | 109
oxidoreductase. The separation of collection/conversion from
catalysis is dictated by the thermodynamics of the water
splitting reaction. To match the solar spectrum and at the
same time deliver oxidizing and reducing equivalents of sufficient potential to split water, PSII is confined to generating an
electron/hole pair one photon at a time. However, water
splitting is a four-electron/hole process.4 Hence, multielectron
catalysts at the terminus of the charge-separating network
are compulsory so that the one photon-one electron/hole
equivalency can be bridged to the four-electron/hole chemistry
of water splitting. Additionally, the catalysts must couple
protons to the multielectron transformation in order to avoid
high-energy intermediates.
In addition to the separation of light collection/conversion
from catalysis:
OEC is an all-inorganic metal oxide core. The recent X-ray
diffraction and spectroscopic studies reveal PSII to be very
complex5–7 However, most of this complexity is associated
with the generation of a wireless current at high efficiency.
OEC is postulated to be a distorted Mn3Ca cube with oxygen
atoms at alternating corners of the cube and a dangling Mn
atom.8,9
The OEC core self-assembles upon oxidation of incoming
metal ions. Light drives the oxidation of Mn2+ to higher
oxidation states, which then leads to OEC self-assembly.10
The OEC–protein complex is not structurally stable and hence
a repair mechanism is required. Water oxidation at OEC
produces reactive oxygen species, which damage the associated
proteins of the PSII complex. Oxygenic photosynthetic organisms have evolved to replace the D1 protein in which the OEC
resides with a newly synthesized copy every B30 minutes.11
Thus functional stability is maintained despite structural
instability.
Proton management is required for water oxidation catalysis
in aqueous solutions at pH = 7. The protons released upon
water oxidation cannot be captured and transported by H2O
because it is a very weak base. Moreover, the [OH ] = 10 7 M
at pH = 7. Proton transport from OEC is accomplished along
water channels lined by Lewis basic amino acid side
chains.12,13
Within the foregoing framework, an OECmpd is presented.
The catalyst self-assembles from aqueous solution upon oxidation of Co2+ to Co3+. Phosphate anion manages the
protons released from water oxidation and also provides a
mechanism for repair. The catalyst is extremely versatile and it
can form on diverse conducting surfaces of varying geometry.
Thus the catalyst can be easily interfaced with a variety of light
absorbing and charge separating materials.
Catalyst formation and characterization
Controlled potential electrolysis of Co2+ salts in pH 7 phosphate (Pi) buffer at 1.3 V (vs. NHE) results in observed
currents that asymptotically approach 1.5 mA cm 2 over
several hours. During this time, a dark green-black film forms
on the surface of the electrode surface. Similar behavior is
observed if methyl phosphonate (MePi) is used as the supporting electrolyte instead of phosphate.14 Fig. 1 shows SEMs of
the electrode surface at various times during the electrolysis.
110 | Chem. Soc. Rev., 2009, 38, 109–114
Fig. 1 Current density profile for bulk electrolysis at 1.29 V
(vs. NHE) in 0.1 M KPi electrolyte at pH 7.0 containing 0.5 mM
Co2+. SEM images of the electrode surface taken at indicated time
points during the electrodeposition of the catalyst film. The ITO
substrate can be seen through cracks in the dried film.
The ITO substrate can be seen through cracks in the film that
form upon drying, as evidenced by particles that are split into
complementary pieces. To date, indium-tin-oxide (ITO) and
fluorine-tin-oxide (FTO) have been the electrodes of choice
because these materials exhibit high overpotentials for O2
production from water and thus ensure minimal background
activity. Nevertheless, the thin film forms on many conducting
surfaces including glassy carbon, carbon felt, Ni and other
metals. The thickness of the electrodeposited catalyst is determined by the length of the electrodeposition and the concentration of Co2+ in the deposition solution. Prolonged
electrolysis (passage of 40 C cm 2) produces films with limiting
thicknesses of B3 mm with the concomitant formation of
spherical nodules of 1–5 mm in diameter. The nodules are of
similar composition to the film. Prepared films do not need to
be used immediately. They can be stored under ambient
conditions and subsequently used as an anode in Co-free
solutions.
Cyclic voltammetry (CV) of solutions of 1 mM Co2+ in
Pi electrolyte suggests that a catalytically active film forms
immediately following oxidation of Co2+. A quasi-reversible
wave in the CV is observed at 1.14 V vs. NHE (pH 7.0) which
is more similar to the Co3+/2+ couple for cobalt ion with
hydroxo ligands (E[Co(OH)2+/0] = 1.1 V vs. NHE).15 A large
catalytic wave is observed immediately beyond the Co2+/3+
couple and catalyst appears to deposit immediately following
oxidation of Co2+ to Co3+.
Powder X-ray diffraction patterns of the electrodeposited
catalysts exhibit broad amorphous features; no peaks indicative of crystalline phases are observed other than the peaks
associated with the ITO sublayer. Films prepared from Pi
exhibit a 2 : 1 Co : P ratio as determined by energy-dispersive
X-ray (EDX) and elemental analyses. Elemental analysis of
films prepared from MePi permits the carbon content of the
film to be ascertained. A P : C ratio of B2 : 1 indicates partial
decomposition of methylphosphonate to phosphate within the
This journal is
c
The Royal Society of Chemistry 2009
film; phosphate signals in the {1H}31P NMR spectrum of films
isolated from the electrode support this observation. In contrast, MePi and Pi buffer solutions remain intact under
prolonged electrolysis. No other major signals excepting those
from the buffer are observed in the 31P NMR spectra of
solutions taken from the working or auxiliary compartments.
Together, the XRD and analytical results indicate that
electrolysis of a Co2+ solution in aqueous phosphate buffer
results in the electrodeposition of an amorphous Co oxide or
hydroxide incorporating a substantial amount of phosphate
anion. X-Ray absorption spectroscopy (XAS) studies are
underway to provide additional structural information.
Oxygen catalysis
Electrodeposition of the film is accompanied by vigorous
effervescence of O2, as confirmed by mass spectrometric
analysis. Mass spectrometric detection of O2 in real-time from
18
OH2 enriched Pi and MePi indicate an isotopic ratio of
16,16
O2, 18,16O2 and 18,18O2 in agreement with the predicted
statistical ratio, indicating that water is the source of the
O-atoms in the evolved O2. The data shown in Fig. 2a for
films prepared in Co2+/Pi solutions is exemplary. Signals for
all three isotopes of O2 rise from their baseline levels minutes
after the onset of electrolysis and then they slowly decay after
electrolysis is terminated and O2 is purged from the head
space. The O2 isotopic ratios are invariant over hours
(Fig. 2b). The Faradaic efficiency of the catalyst is most
conveniently measured by a fluorescence-based O2 sensor.
Fig. 2c shows the current passed during an electrolysis performed at 1.3 V (blue line) is completely accounted for by the
quantity of O2 produced (red line). Moreover, the amount of
O2 produced (95 mmoles, 3.0 mg) greatly exceeds the amount
of catalyst (B0.2 mg), which shows no perceptible decomposition over the course of the experiment. Thus, all current
passed through the catalyst is used for O2 production.
Fig. 3 shows the Tafel plot for a CoPi catalyst run in Pi; a
similar Tafel plot is obtained for the catalyst run in MePi.
Appreciable current densities are obtained beginning at 0.28 V
overpotential. The red circles on Fig. 3 show the current
density measured at a constant applied potential (Vappl =
1.24 V) at various pH values. The overpotential (Z) was
determined by correcting Vappl for iR drop and assuming
Nernstian behavior for the potential at various pH values
according to the following:
Z = (Vappl + 0.059DpH
iR)
E(pH 7)
(2)
The consistency between the Tafel plot (black circles) obtained
at pH = 7 and the Tafel plot obtained from the pH data
(red circles) using eqn (2) indicates that increasing the pH by
one unit at constant applied potential (1.05 V) has approximately the same effect as increasing the overpotential by
0.059 V at pH 7. This result implicates a reversible ne , nH+
removal in a PCET event prior to the rate determining step
for O2 evolution when there is a significant concentration of
both H2PO4 and HPO42 in solution. Based on our knowledge of O2 reduction at cobalt centers16 and our measurements
of oxygen atom activation by PCET,17–19 we tentatively
propose a Co2+–OH2/Co3+–OH or Co3+–OH/Co4+–oxo
This journal is
c
The Royal Society of Chemistry 2009
Fig. 2 (a) Mass spectrometric detection of isotopically-labeled 16,16O2
(
), 16,18O2 (
) and 18,18O2 (
) during electrolysis of a
catalyst film on ITO in KPi electrolyte containing 14.6% 18OH2.
Green and red arrows indicate initiation and termination of electrolysis at 1.29 V (NHE). Inset: expansion of the 18,18O2 signal.
(b) Percent abundance of each isotope over the course of the experiment. Average observed abundance 2s indicated above each line and
calculated statistical abundances are indicated in the parenthesis.
(c) O2 production measured by fluorescent sensor (
) and the
theoretical amount of O2 produced (
) assuming a Faradaic
efficiency of 100%. Green arrow indicates initiation of electrolysis at
1.29 V and red arrow indicates termination of electrolysis. Reproduced
with permission from Science 2008, 321, 1072. Copyright 2008
American Association for the Advancement of Science.
for this PCET equilibrium (Fig. 4, bottom). The nuclearity
of the catalytic active site is not known. Accordingly, the oxo
does not necessarily have to be terminal.
These results are reminiscent of the oxidation of tyrosine in
model PCET systems shown in Fig. 4 (top). A metal-to-ligand
charge transfer excited state of a Re polypyridyl is capable of
oxidizing an appended tyrosine, but only if HPO42 is present.20 The reaction kinetics are pH dependent and consistent
with a PCET mechanism in which ET from Y to the oxidant is
accompanied by PT from Y to HPO42 . Accompanying
Chem. Soc. Rev., 2009, 38, 109–114 | 111
(1.30 V, pH 7.0) near the formal HOCl/Cl potential (1.28 V
at pH 7), it is oxygen that is produced at high Faradaic
efficiency from 0.5 M NaCl solutions.
A robust catalyst
Fig. 3 Tafel plot (K), Z = (Vappl iR) E(pH 7), of a catalyst film
on ITO in 0.1 M KPi electrolyte pH = 7.0, corrected for the iR drop
of the solution. pH data converted into a Tafel plot ( ) using eqn (2).
The pH = 5 and pH = 8 data points are indicated by arrows.
Reproduced with permission from Science 2008, 321, 1072. Copyright
2008 American Association for the Advancement of Science.
theoretical work on the model system supports such a concerted PCET mechanism with HPO42 acting as the proton
acceptor.21
Catalyst function does not require pure water. Catalyst
prepared from Pi or MePi solutions retain high activity for
oxygen production from Co-free buffer solutions containing
high salt concentration.14 Although the catalyst operates
The design of a cobalt-based catalyst presents significant
challenges to the inorganic chemist. Co2+ in an oxygen-atom
ligand field is a high spin, d7, ion. Hence, the eg(M–Ls*)
orbital is populated and cobalt ion in the 2+ state is substitutionally labile. Conversely, Co3+ and higher oxidation
states are low spin and substitutionally inert in an oxygenatom ligand field. This long known reactivity22 of Co2+ and
Co3+ (and higher oxidation states) explains the 109 difference
in the substitution rates of aqua ligands on cobalt ions in
solution. Substitution rates on metal ions in solution have
been correlated with metal oxide dissolution rates.23 The
Co2+, Co3+ and likely Co4+ oxidation states will be accessed
in any 4e water–oxygen redox cycle involving cobalt. Thus a
quandary is presented. How can a ‘‘stable’’ catalyst be prepared? The typical approach of an inorganic chemist is to
prepare a ligand that enforces, usually by chelation, a stable
ligand environment. But in such a case, a ligand field that is
appropriate for Co2+ will not be so for Co3+ and excess
potential will be required to oxidize Co2+ to Co3+. Thus by
pursuing the strategy of a stable ligand environment about the
cobalt ion, excess overpotential will be introduced into the
redox cycle.
Fig. 4 (top) The PCET activation of tyrosine upon excitation of a Re(I) complex. The pH dependence of the PCET rate constant arises from the
H2PO4 /HPO42 equilibrium. The increase in rate constant with increasing pH establishes HPO42 as the proton acceptor. (bottom) Oxidation to
the active catalysts from which O2 generation occurs may proceed by the PCET reaction of a Co3+–hydroxide intermediate in which HPO42 is the
proton acceptor.
112 | Chem. Soc. Rev., 2009, 38, 109–114
This journal is
c
The Royal Society of Chemistry 2009
Though solubility products (Ksp) for cobalt ions and
HPO42 are not easily found, the Ksp for Ca(HPO4) is
10 7.24 This value offers an ‘‘electrostatic’’ baseline for the
Ksp of HPO42 with a 2+ ion. Overlaying ligand field considerations on this Ksp, an HPO42 salt of high spin Co2+
would be expected to be more soluble than Ca(HPO4) whereas
one of low spin Co3+ would be expected to be less soluble than
Ca(HPO4). The window provided by the surmised differences
in the Ksp of HPO42 with Co2+ and Co3+ explains the
electrodeposition process since the film only forms upon
oxidation of Co2+ to Co3+. In light of in situ formation, a
mechanism for reformation of the catalyst during cycling is
viable. If Co2+ is produced during the cycle and released from
the film prior to re-oxidation, a dynamic equilibrium between
Co2+–HPO42 in solution and Co3+–HPO42 on the electrode may be established. Isotope labeling studies are underway to observe the exchange of ions directly between the
electrodeposited film and solution. In addition, if Co2+ is part
of a cluster of higher oxidation state cobalt centers, the 2+
oxidation state may be shared over the entire cluster core thus
retarding the release of cobalt ions from the cluster prior to
re-oxidization.
Table 1 Comparison of functional properties of OEC and the CoPi
OECmpd
Photosystem II OEC
CoPi OECmpd
Self-assembly
Earth-abundant
metal (Mn)
All oxo core
Self-assembled from
water upon metal
oxidation
Earth-abundant
metal (Co)
All oxo framework
Self-assembled from
water upon metal
oxidation
Repair
D1 protein
HPO42 /Co3+
equilibrium
O2 generation
From neutral water
At 1 atm and RT
At low overpotential
Proton carrier
(amino acid)
From neutral water
At 1 atm and RT
At low overpotential
Proton carrier
(HPO42 )
mechanism involving chloride does not effectively compete
with the inner sphere redox processes involving oxygen. In
light of the foregoing considerations, Table 1 proposes several
parallels between OEC and the CoPi OECmpd.
Future prospects
A working model for the cycle
Several useful concepts of catalysis are embodied by the CoPi
OECmpd. A working model for operation of the catalyst is
shown in Fig. 5. The Co2+ is oxidized to Co3+ and then is
deposited on the electrode in the presence of HPO42 . The pH
dependence of the current density is consistent with the
oxidation of Co2+–OH2 to Co3+–OH and/or the oxidation
of Co3+–OH to Co4+–oxo; in Fig. 5 we emphasize the latter
PCET process to produce a Co4+–oxo from which O2 is
produced and cobalt is returned to the 2+ oxidation state.
In situ EXAFS experiments on an active electrode are currently underway to establish the nature of the catalytically
active state of cobalt ion. The overall cycle of catalysis need
not proceed strictly in the heterogeneous or homogeneous
phase. The catalyst may cycle through both phases, which
communicate with each other via equilibria processes. Importantly, we believe that catalysis is occurring at molecular
centers with discrete electronic structure. Unlike most solid
state catalysts, an electronic structure of the bulk or a
nanodomain does not prevail. We suspect that this is one
reason why the catalyst works well in salt solution at modest
overpotentials. Presumably, outer sphere electron transfer
The CoPi OECmpd advances the viability of water-splitting as
a solar storage mechanism by enabling the solar-to-fuels
conversion to be performed in neutral water under benign
conditions. Mechanistic studies of CoPi OECmpd may provide
insight into the requirements for O2 evolution under the
conditions of natural photosynthesis. Additionally, CoPi
OECmpd offers inroads to artificial systems aimed at mimicking photosynthesis. Although many ingenious synthetic charge
separating networks comprising protein, inorganic and organic dyads and triads have been developed, the realization of
artificial photosynthesis using these constructs has been limited by the difficulty of connecting them to water-splitting
catalysts. Because most of these constructs are not stable in
highly acidic or basic environments, CoPi OECmpd is a good
candidate as a catalyst interface for these charge-separating
networks. The catalyst can also be deposited on conductive or
semi-conductive substrates with complicated geometries and
large surface areas. As in the charge-separating networks,
one-photon, one-electron charge separation of a photoanode
can be accumulated by the catalyst to attain the four equivalents needed for water splitting. The ease of implementation
of the CoPi OECmpd with such a diverse array of photoactive
materials suggests that the catalyst will be of interest to many
in their endeavors to store solar energy by water splitting.
Acknowledgements
Fig. 5 A working model for the Co–phosphate OECmpd.
This journal is
c
The Royal Society of Chemistry 2009
This research was supported by a Center for Chemical
Innovation of the National Science Foundation (Grant
CHE-0533150) and a grant from the Chesonis Family Foundation. Grants from the NSF also provided instrument
support to the DCIF at MIT (CHE-9808061, DBI-9729592).
M.W.K. is a Ruth Kirchenstein NIH Postdoctoral Fellow.
Y.S. gratefully acknowledges the Department of Defense for a
pre-doctoral fellowship.
Chem. Soc. Rev., 2009, 38, 109–114 | 113
References
1 N. S. Lewis and D. G. Nocera, Proc. Natl. Acad. Sci. U. S. A.,
2006, 103, 15729.
2 J. Barber, Philos. Trans. R. Soc. London, Ser. A, 2007, 365, 1007.
3 R. Eisenberg and H. B. Gray, Inorg. Chem., 2008, 47, 1697.
4 T. A. Betley, Q. Wu, T. Van Voorhis and D. G. Nocera, Inorg.
Chem., 2008, 47, 1849.
5 K. N. Ferreira, T. M. Iverson, K. Maghlaoui, J. Barber and
S. Iwata, Science, 2004, 303, 1831.
6 B. Loll, J. Kern, W. Saenger, A. Zoun and J. Biesiadka, Nature,
2005, 438, 1040.
7 J. Yano, J. Kern, K. Sauer, M. J. Latimer, Y. Pushkar,
J. Biesiadka, B. Loll, W. Saenger, J. Messinger, A. Zouni and
V. K. Yachandra, Science, 2006, 314, 821.
8 J. Barber, Inorg. Chem., 2008, 47, 1700.
9 J. M. Peloquin, K. A. Campbell, D. W. Randall, M. A. Evanchik,
V. L. Pecoraro, W. H. Armstrong and R. D. Britt, J. Am. Chem.
Soc., 2000, 122, 10926.
10 R. L. Burnap, Phys. Chem. Chem. Phys., 2004, 6, 4803.
11 E.-M. Aro, M. Suorsa, A. Rokka, Y. Allahverdiyeva,
V. Paakkarinen, A. Saleem, N. Battchikova and E. Rintamaki,
J. Exp. Bot. E, 2005, 56, 347.
114 | Chem. Soc. Rev., 2009, 38, 109–114
12 J. W. Murray and J. Barber, J. Struct. Biol., 2007, 159, 228.
13 H. Ishikita, W. Saenger, B. Loll, J. Biesiadka and E.-W. Knapp,
Biochemistry, 2006, 45, 2063.
14 Y. Surendranath and D. G. Nocera, J. Am. Chem. Soc., submitted
for publication.
15 B. S. Brunschwig, M. H. Chou, C. Creutz, P. Ghosh and N. Sutin,
J. Am. Chem. Soc., 1983, 105, 4832.
16 C. J. Chang, Z.-H. Loh, C. Shi, F. C. Anson and D. G. Nocera,
J. Am. Chem. Soc., 2004, 126, 10013.
17 S. Y. Reece, J. M. Hodgkiss, J. Stubbe and D. G. Nocera, Philos.
Trans. R. Soc. London, Ser. B, 2006, 361, 1351.
18 J. Rosenthal and D. G. Nocera, Acc. Chem. Res., 2007, 40,
543.
19 J. D. Soper, S. V. Kryatov, E. V. Rybak-Akimova and
D. G. Nocera, J. Am. Chem. Soc., 2007, 129, 5069.
20 T. Irebo, S. Y. Reece, M. Sjödin, D. G. Nocera and
L. Hammarström, J. Am. Chem. Soc., 2007, 129, 154622.
21 H. Ishikita, A. V. Soudackov and S. Hammes-Schiffer, J. Am.
Chem. Soc., 2007, 129, 11146.
22 F. Basolo and R. G. Pearson, Mechanisms of Inorganic Reactions,
John Wiley & Sons, London, 1958.
23 W. H. Casey, J. Colloid Interface Sci., 1991, 146, 586.
24 A. C. Bennett and F. Adams, Soil Sci. Soc. Am. J., 1976, 40, 39.
This journal is
c
The Royal Society of Chemistry 2009
This article was published as part of the
2009 Renewable Energy issue
Reviewing the latest developments in renewable
energy research
Guest Editors Professor Daniel Nocera and Professor Dirk Guldi
Please take a look at the issue 1 table of contents to access
the other reviews.
EDITORIAL
www.rsc.org/csr | Chemical Society Reviews
Living healthy on a dying planetw
Daniel G. Nocera
DOI: 10.1039/b820660k
Living healthy on a dying planet—we
are a world out of balance. We seek
immortality at the individual level and
are oblivious to the health of our humanity at a global level. The average life
expectancy in the world is 66 years and
in the developed world it is 478 years.1
But this is not enough. We stridently seek
to extend our existence. One of the great
discoveries of recent science is the mapping of the human genome. And what is
the hopeful outcome of this human genome project? Most of all practitioners
say it is to cure disease at a genetic level.
While this is a noble cause for the young,
at what cost does an octogenarian
society aspire to eternal life when the
environment on which humanity finds
sustenance is in peril?
A clear picture is emerging about the
change that confronts our planet in the
near future. Professor James Anderson
of Harvard has determined that the current level of CO2 in our atmosphere now
ensures sufficient heating to melt the
Arctic summer ice cap within the next
decade. The ice in the Arctic Ocean acts
as a heat shield for our planet by preventing the flow of warm ocean currents
to the furthest reaches of the northern
hemisphere. With the polar ice cap gone,
significant global changes to our environment are assured the most concerning
of which is the collapse of the Greenland
ice cap and the loss of permafrost. Much
of the world’s industrial centers and
population are situated near coastline,
which will be submerged under the water
accompanying the 7 metre rise in ocean
levels with the collapse of the Greenland
ice cap. Even more troublesome is the
inability of living organisms to adapt to
our quickly changing environment. Environmental biologists have documented
the disappearance of entire ecosystems
Henry Dreyfus Professor of Energy,
Massachusetts Institute of Technology,
Cambridge, MA, USA.
E-mail: [email protected]
w Part of the renewable energy theme issue.
This journal is
c
from our planet in a single human generation. The worrisome concern is
whether humans can adapt quickly enough to our changing world.
With our humanity blindly staggering
toward an ominous future, we are
swayed by the folly of futurists who
promise immortality. We allocate resources as if we believe that dying is an
option. In the US, which is no different
from most societies in the developed
world, GDP is roughly divided equally
between health and energy. However, the
emphasis we place on immortality is
plainly evident in the 30 : 1 ratio of
R&D funding for health : energy
science.2 Despite this imbalance in
resources, a global cohort of scientists,
undeterred by societal indifference,
confronts the energy challenge head on.
These scientists, some of which are
collected in this issue, provide hope for
our global future. The problem they
tackle is daunting. In the past several
years, it has been well documented that
our global energy need will roughly double by mid-century and triple by 2100.3,4
Holding atmospheric CO2 levels to even
twice their preanthropogenic values by
mid-century will require invention,
development, and deployment of
schemes for carbon-neutral energy production on a scale commensurate with,
or larger than, the entire present-day
energy supply from all sources combined. One pre-eminent solution to the
energy challenge is offered by the sun.
More energy from the sun strikes the
Earth’s surface than humans currently
use in a year. However, current options
to harness and store this energy are too
expensive to be implemented on a large
scale. Hence the objective to science is to
develop new materials, reactions and
processes to enable solar energy to be
sufficiently inexpensive to penetrate
global energy markets.
This themed issue of Chemical Society
Reviews on Renewable Energy collects
the work of scientists that seek to transform the dream of a solar-powered
The Royal Society of Chemistry 2009
society into reality. To do this, they
undertake science to (i) uncover the
secrets of bioenergy conversion and biocatalysis, (ii) design novel solar capture
and conversion materials and (iii) create
catalysts that can store energy in hierarchical materials or in the form of the
chemical bonds of fuels.
James Barber (DOI: 10.1039/b802262n)
presents a review on the solar energy
blueprint offered by nature. The primary
steps of natural photosynthesis involve
the capture of sunlight and its conversion
into a wireless current. The anodic
charge of the wireless current is used at
the oxygen-evolving complex (OEC) in
Photosystem II to oxidize water to oxygen, with the concomitant release of four
protons. The cathodic charge of the
wireless current is channeled to Photosystem I where protons are reduced to
‘‘hydrogen’’ by ferredoxin NADP reductase—with the reduced hydrogen equivalents stored via the conversion of NADP
to NADPH. Maria Ghirardi (DOI:
10.1039/b718939g) exploits the photosynthetic properties of oxygenic and nonoxygenic microbes in combination with
the H2-producing capabilities of hydrogenases and nitrogenases. By uniting
these different enzymatic systems, she
provides a photobiological path for storing sunlight by the production of hydrogen. Biological production of H2 on a
large scale brings certain practical requirements that need to be addressed at
the molecular as well as the microbial
level. The synthetic model compounds
described
by
Frédéric
Gloaguen
and
Thomas
Rauchfuss
(DOI:
10.1039/b801796b) permit detailed mechanistic studies by which hydrogenases
operate. Fraser Armstrong’s (DOI:
10.1039/b801144n) tutorial review shows
how the H2-producing reactivity of hydrogenases may be deciphered using
specialized electrochemical techniques.
Especially crucial is to understand how
hydrogenases can sustain high H2 fluxes
in the presence of the O2 that the photosynthetic enzymes produce. Devens
Chem. Soc. Rev., 2009, 38, 13–15 | 13
Gust, Ana Moore and Thomas Moore
(DOI: 10.1039/b800582f) transition the
themated issue from the world of biology
to the world of technology. In their fascinating tutorial review, they show the parallels between energy conversion in
biology, accomplished with protonmotive-force (pmf), and energy conversion
in technology, accomplished by electromotive-force (emf). They show aspects of
biology that might be advantageously
incorporated into emerging energy technologies, as well as ways in which technology might improve upon the design of
biological energy-based systems.
The photosynthetic process may be
realized artificially by spatially separating energy storage from the light capture
and conversion system. For each component, the material must be earth-abundant
and easily manufactured.
Charles Lieber (DOI: 10.1039/b718703n),
Dirk Guldi (DOI: 10.1039/b802652c) and
Gerald Meyer (DOI: 10.1039/b804321n)
look to nanoscience for the construction
of high efficiency and low cost solar
energy capture and conversion devices.
The effort from the Lieber group seeks to
elucidate the properties and potential of
semiconductor nanowires as building
blocks for photovoltaic devices based
on investigations at the single nanowire
level. Si-nanowires with reproducible
and carefully tunable PV properties
are presented. The tutorial review progresses to show the rational design of
more complex architectures based on
nanowire tandem cell and quantum well
structures. The use of carbon nanotubes
as electron and photon carriers is
detailed in the review of Dirk Guldi. He
shows how superior transport properties
may be achieved by tuning the properties
of the carbon nanotubes. Gerald Meyer
provides a review on Grätzel’s original
discovery that interfaced nanoscience to
energy conversion—dye-sensitized TiO2
solar cells. A comprehensive treatment
of the current understanding of the
charge-transfer processes at sensitized
TiO2 interfaces is offered. Against this
backdrop of knowledge, new avenues of
exploration are identified that have the
potential for yielding dye-sensitized solar
cells with high light-to-electrical power
conversion efficiencies.
If solar energy is to be a major primary
energy source for society, then it must be
stored owing to the diurnal variation in
14 | Chem. Soc. Rev., 2009, 38, 13–15
local insolation. Debra Rolison (DOI:
10.1039/b801151f) seeks to store the electricity from photovoltaics directly in novel
3D nanoarchitectures. Void space and disorder within the 3D architecture allow for
rapid molecular fluxes that lead to amplification within the electrical interface.
Alternatively, solar energy may be stored
in the form of chemical bonds of fuels. The
storage of solar energy in the bond is
attractive in view of its high energy density.
A particularly attractive fuel forming
reaction is the aforementioned watersplitting reaction of photosynthesis. In
this process, solar energy is stored in the
bond rearrangement of H2O to H2 and O2.
Oleg Ozerov (DOI: 10.1039/b802420k)
explores the little studied activation of
water by oxidative addition to a transition metal center. A popular design for
achieving water splitting is to interface
water reduction and oxidation catalysts
to a photovoltaic membrane.5 Daniel
DuBois and Mary Rakowski DuBois
(DOI: 10.1039/b801197b) describe the
construction of a tool set for the development of highly active catalysts for the
production of H2, whereas developments
from my group have led to the discovery
of a highly active cobalt phosphate
catalyst for O2 production (DOI:
10.1039/b802885k). This latter catalyst
produces O2 from neutral water at room
temperature and low pressure and
captures many of the functional elements
of the OEC of Photosystem II. Akihiko
Kudo (DOI: 10.1039/b800489g) combines photochemistry and catalysis in
the creation of metal (oxy)sulfide and
metal
(oxy)nitride
water-splitting
photocatalysts. Bruce Parkinson (DOI:
10.1039/b719545c) seeks to improve the
overall efficiency of heterogeneous photocatalysts by developing new methods for
the combinatorial production and high
throughput screening of metal oxides.
The use of hydrogen as a fuel requires
its storage by materials with high volumetric and gravimetric hydrogen densities.
Jason Graetz (DOI: 10.1039/b718842k)
emphasizes the potential of metal hydrides as effective hydrogen storage materials, whereas Tom Baker (DOI:
10.1039/b800312m) explores the hydrogen-storing properties of BN compounds
owing to their light weight and propensity for bearing multiple protic (NH) and
hydridic (BH) hydrogens. Stephen Shevlin
and Xiao Guo (DOI: 10.1039/b815553b)
complement these experimental approaches
by demonstrating the power of density
functional theory (DFT) simulations in
evaluating, developing and discovering
hydrogen storage materials. Hydrogen
storage may be circumvented if production of oxygen from water is accompanied
by reduction of CO2 to a liquid fuel.
Cliff Kubiak (DOI: 10.1039/b804323j)
reviews electrocatalytic and homogenous
approaches to CO2 reduction and then
defines benchmarks for accomplishing
this reaction with high efficiencies.
I close with Kurt Vonnegut’s paradoxical
words of comfort shortly before his
death. In a PBS NOW interview with
David Brancaccio, Vonnegut describes
the planet as a living organism. He
reminds us that the immunological
response of sophisticated life forms will
eliminate irksome intruders when the
organism is sufficiently compromised.
Vonnegut sees humans as the irksome
intruder of our planet. As we carelessly
choose a path to suffocate the planet in
CO2, Vonnegut assures Brancaccio that
he need not worry: the planet’s immunological system will respond and eliminate
humans by not sustaining them in the
dramatically altered environment that
they created. Elaborating on Vonnegut’s
perspective, the ‘‘dying planet’’ in the
title of this preface has little to do with
the Earth, which will continue to exist
and flourish at high CO2 levels, though
not as we know it. Rather, it is the humans on the Earth that are in a precarious state. When Brancaccio confronts
Vonnegut about a solution to the ‘‘dying
planet’’, Vonnegut responds, ‘‘join a
gang and do something about it’’. In this
themed issue on Renewable Energy, it is
a pleasure to collect a ‘‘gang’’ of scientists who have taken up Vonnegut’s callto-arms by providing the discovery
needed to answer the greatest challenge
confronting humanity—the large scale
supply of carbon-neutral energy.
Daniel G. Nocera
Henry Dreyfus Professor of Energy
Massachusetts Institute of Technology
Cambridge, MA USA
November 2008
References
1 CIA: The World Fact Book, https://www.
cia.gov/library/publications/the-world-fact
book/rankorder/2102rank.html.
This journal is
c
The Royal Society of Chemistry 2009
2 K. Koizumi, AAAS Report XXXII,
Research and Development FY 2008,
American Association for the Advancement
of Science, Washington DC, ch. 2, 2008.
3 M. I Hoffert, K. Caldeira, A. K. Jain,
E. F. Haites, L. D. Harvey, S. D. Potter,
This journal is
c
M. E. Schlesinger, T. M. Wigley
and D. J. Wuebbles, Nature, 1998, 395,
881.
4 Climate Change and Water. Technical
Paper of the Intergovernmental Panel on
Climate Change, ed. B. C. Bates,
The Royal Society of Chemistry 2009
Z. W. Kundzewicz, S. Wu and
J. P. Palutikof, IPCC Secretariat, Geneva,
2008, 210 pp.
5 N. S. Lewis and D. G. Nocera, Proc.
Natl. Acad. Sci. U. S. A., 2006, 103,
15729.
Chem. Soc. Rev., 2009, 38, 13–15 | 15
Personalized Energy:
The Home as a Solar Power Station and Solar Gas Station
Daniel Nocera
Henry Dreyfus Professor of Energy
Massachusetts Institute of Technology
Cambridge, MA USA
December 2008
No more gas stations, no more towers and transmission lines… just the sun and your home,
which will generate all the energy you need to live. This is not a storyline for a futuristic sci-fi
novel. A committed science and engineering research effort could make the home into a solar
power station and solar gas station in the near future. In pursuing the goal of personalized energy,
science and engineering drives inextricably to the heart of the energy challenge by addressing the
triumvirate: secure, carbon neutral and plentiful low-cost energy. Because energy use scales
directly with wealth, 1 point-of-use solar energy will put individuals, in the smallest village in the
developing world and in the largest city of the developed world, on a more level playing field.
And the individual will be energy secure as they will control the energy on which they live. More
powerfully, the possibility of generating terawatts of carbon-free energy may be realized by
making available personalized energy to the 3 billion low-energy users and 3 billion people to
inhabit our planet over the next half century.
Personalized energy will be possible only if solar energy is a 24/7 available supply. Hence, the
key enabler for personalized energy is inexpensive storage. There are many methods of solar
storage, however, low energy densities of most of these methods present formidable challenges
for the implementation of solar energy at the individual level. For instance, the energy densities
by mass of compressed air (300 atm, ~0.5 MJ/kg), flywheels (~0.5 MJ/kg), supercapacitors
(~0.01 MJ/kg) and water pumped uphill (100 m, ~0.001 MJ/kg) are too low for sufficient storage
of solar energy. Batteries are often discussed as an effective energy storage medium for solar
energy and considerable efforts are currently being devoted to their continued development. 2
Most advances in battery technologies, however, have little to do with energy density but rather
they are concerned with power density (i.e. the rate at which charge can flow in and out of the
battery) and lifetime. Energy densities of batteries are also low (~0.1 - 0.5 MJ/kg) with little
room for improvement since the densities of battery materials (electrolytes and inorganic redox
materials) cannot be dramatically altered. Conversely, the energy density of liquid fuels (~50
MJ/kg) is ≥102 greater than that of the best of the foregoing storage methods and H2 possesses an
even larger energy density of 140 MJ/kg. Indeed, society has intuitively understood this disparity
in energy density as it has developed over the last century. Energy storage in our society is based
on fuels. Moreover, despite the prevalence of biology to translocate charge across a membrane, 3
proton-motive forces are not directly stored but rather are transmuted into the bonds of chemical
fuels.
1
2
3
M. I. Hoffert K. Caldeira, A. K., Jain, E. F. Haites, L. D. Harvey, S. D. Potter, M. E. Schlesinger, T. M. L.
Wigley, D. J. Wuebbles, Nature 1998, 395, 881.
B. B. Owens, W. H. Smyrl, J. J. Xu, J. Power Sources 1999, 81-82, 150.
M. Hambourger, G. F. Moore, D. M. Kramer, D. Gust, A. L. Moore, T. A. Moore, Chem.Soc. Rev. 2009, DOI:
10.1039/b800582f.
A fuel-forming reaction that provides a carbon-neutral and sustainable method of solar storage
is:
2H2O + light = 2H2 + O2
(1)
Since light cannot directly act on water to engender its splitting into its elemental components,
catalysts are needed to effect the overall transformation. In Nature, this reaction is accomplished
by photosynthesis. 4,5 The oxygen evolving complex (OEC) splits water by first releasing oxygen
to leave four electrons and four protons, which are then combined with NADP+ (to NADPH) at
the spatially remote site of ferredoxin-NADP+ reductase (FNR) residing in Photosystem I (PS I).3
A key design element of photosynthesis is the separation of the functions of light collection and
conversion from catalysis. Light is collected and converted by Photosystem II (PS II) into a
wireless current, which is fed to OEC and FNR so that they can perform water-splitting. The
separation of collection/conversion from catalysis is dictated by the thermodynamics of the water
splitting reaction. To match the solar spectrum and at the same time deliver oxidizing and
reducing equivalents of sufficient potential to split water, PS II is confined to generating an
electron/hole pair one photon at a time. However, water splitting is a four-electron/hole process
that is coupled to proton transfer. 6- 8 Hence, multielectron catalysts at the terminus of the chargeseparating network are compulsory so that the one photon-one electron/hole equivalency can be
bridged to the four-electron/hole chemistry of water splitting. Additionally, the catalysts must
couple protons to the multielectron transformation in order to avoid high-energy intermediates.
An artificial system that
captures the essential
functional elements of
photosynthesis is shown
in Figure 1. 9 Water
oxidation and reduction
catalysts are connected
to a photovoltaic assembly based on a solid state
junction, either on the
macroscale or on the
Figure 1. An photoconversion cell that stores solar energy by the solar-driven
conversion of water to H2 and O2. This cell forms the basis for a society sustained
nanoscale. As in photoon personalized energy.
synthesis, light is captured and converted into
a spatially-separated single electron-hole pair. The electron and hole are captured by the catalyst
until the necessary four electron-hole equivalents are attained to drive the bond rearrangement of
water to H2 and O2. Recombination of the reduced fuel with released O2 would then regenerate
the original species, closing the cycle in a carbon-neutral fashion.
4
5
6
7
8
9
J. Barber, Phil. Trans. Roy. Soc. A, 2007, 365, 1007.
J. Barber, Inorg. Chem. 2008, 47, 1700.
J. L. Dempsey, A. J. Esswein, D. R. Manke, J. Rosenthal, J. D. Soper, D. G. Nocera, Inorg. Chem. 2005, 44,
6879.
T. A. Betley, Y. Surendranath, M. V. Childress, G. E. Alliger, R. Fu, C. C. Cummins, D. G. Nocera, Phil. Trans.
Royal Soc. B 2008, 363, 1293.
T. A. Betley, Q. Wu, T. Van Voorhis, D. G. Nocera, Inorg. Chem. 2008, 47, 1849.
N. S. Lewis, D. G. Nocera, Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 15729.
2
As depicted in Figure 1, the key first step of water
splitting is the production of oxygen with the
concomitant release of four protons and four
electrons. Hydrogen can be produced with the
protons and electrons extracted from water; fuels
other than hydrogen are possible as well including
the reduction of carbon dioxide to methanol.
Notwithstanding, it is the oxygen-producing step
where the efficiencies of most water splitting
schemes suffer and accordingly is the primary
impediment toward realizing the artificial
photosynthetic system of Figure 1. 10
Figure 2. A new cobalt–phosphate catalyst
that generates oxygen with high activity from
pure water, brine or sea water at neutral pH,
atmospheric pressure and room temperature.
We have responded to the challenge of water
splitting with the design of an O2-generating catalyst
that captures many of the functional elements of
OEC. The catalyst self-assembles from aqueous solution (pure water, brine, sea water are all
acceptable) 11 upon the oxidation of Co2+ to Co3+ in the presence of phosphate or other
electrolytes that maintain neutral pH. 12,13 Phosphate anion manages the protons released from
water oxidation and also provides a mechanism for catalyst repair during turnover. 14 The cobaltphosphate (CoPi) catalyst is extremely active; without any cell engineering, currents as high as
10 mA/cm2 have been obtained from water at room temperature and pressure. Figure 2 shows
oxygen generation from a catalyst film on ITO.
The CoPi catalyst is unique because, unlike most crystalline metal oxides,15 it is functional under
neutral aqueous solutions and not subject to corrosion at non-basic pH owing to the presence of
the repair mechanism. Moreover, the catalyst is extremely versatile. It will form on diverse
conducting surfaces of varying geometry. Thus the catalyst can be easily interfaced with
semiconducting materials to enable the realization of a device such as that shown in Figure 1.
How will personalized energy based on water-splitting chemistry come to fruition? In the
simplest configuration, a photovoltaic on your roof will generate the power you need to live
when the sun shines. In an integrated module, the surplus electricity from the photovoltaic can be
fed to the water-splitting catalysts to generate hydrogen and oxygen, which are stored locally (if
you don’t like hydrogen, then another research imperative arises to use hydrogen to reduce
carbon dioxide to liquid fuels in small scale reactors). At night (or when the sun isn’t out), the
stored hydrogen and oxygen can be recombined in a fuel cell to give the electricity needed to
power your home at night. And I forgot to mention that electric car in your garage; its battery
will be recharged while you sleep. It’s not a pipe dream. There is no fundamental basic science
show-stopper to derail this vision. Photovoltaics generate electricity efficiently. New catalysts,
such as the aforementioned CoPi catalyst, exist to split water to hydrogen and oxygen, which can
10
11
12
13
14
15.
R. Eisenberg , H. B. Gray, Inorg. Chem. 2008, 47, 1697.
Y. Surendranath, M. Dinca, D. G. Nocera, J. Am. Chem. Soc. 2009, in press.
M. W. Kanan, D. G. Nocera, Science 2008, 321, 1072.
Y. Surendranath, M. W. Kanan, D. G. Nocera, Chem. Soc. Rev. 2009, DOI: 10.1039/b802885k.
D. Lutterman Y. Surendranath, D. G. Nocera, J. Am. Chem. Soc. 2009, submitted for publication.
M. R. Tarasevich, B. N. Efremov, Electrodes of Conductive Metal Oxides, S. Trasatti, Ed., Elsevier, Amsterdam,
1980, Ch. 5.
3
be stored and combined in a fuel cell at a later time to generate electricity. If all the basic science
for personalized energy is known, then why hasn’t the dream of a highly distributed energy
system catered to the individual not been enacted? It’s all about cost.
A market penetrable energy system for the individual must be inexpensive. However, science
discovery generally targets high efficiencies with cost as a secondary target. As an example,
consider the CoPi catalyst within the context of commercial electrolyzers. Commercial
electrolyzers are extremely efficient, operating at 1000 mA/cm2. But this current density comes
with a price. To achieve high current densities, commercial electrolyzers operate under harsh
conditions, which are costly to engineer. Indeed, within a PV-electrolyzer-storage-fuel cell
system, the electrolyzer is the highest capex cost. This high cost, however, is not necessary
because high current densities are incommensurate with personalized energy. The current density
of the CoPi catalyst at 80% efficiency is 10 mA/cm2 (in an un-engineered cell), which is better
matched to the ~20 mA/cm2 output of a standard photovoltaic. By sacrificing current density, an
electrolyzer using the CoPi catalyst can be designed to operate under benign conditions, thus
driving the electrolyzer down to a negligible capex cost.
To provide a framework for personalized energy, consider that the average American home uses
20 kW-hr of electricity per day. The storage of 20 kW-hr can be achieved by splitting only 5.5 L
of water to hydrogen and oxygen. A 3 × 2.5 m2 solar panel (operating at 20 mA/cm2) provides
the current needed to split 5.5 L of water in 2.5 hrs. Thus, a catalyst such as CoPi, operating at
the same current efficiency and same surface area of the photovoltaic, can produce 20 kW-hr
equivalents of hydrogen in 3 hrs (80% efficiency for water splitting). Of course, compression
efficiencies for H2 and the efficiency of the fuel cell must also be factored into the cycle, which
will increase amount of hydrogen needed to be stored. The point here is that energy storage
needed for personalized energy is currently within reach at minimal electrolyzer costs. But this is
only ¼ of the personalized energy equation. Discovery is still needed minimize the costs of the
photovoltaic, hydrogen storage and the fuel cell.
Personalized energy at low cost presents new basic research targets. Many of these targets are
squarely centered in the endeavors of chemical and materials research. Here are three:
The fuel cell operates at 50% efficiency largely because Pt is inefficient at O2 reduction
(as it is for the reverse reaction of water splitting).16 Moreover, the cost of Pt accounts for
38% of the fuel cell. Thus, the discovery of new catalysts for the cathode of the fuel cell
will not only deliver higher cell efficiency but it will remove one of the most costly
elements of the fuel cell. And if the cathode can be made to operate under benign
conditions (such as the CoPi water-splitting catalyst), more inexpensive membranes that
have longer lifetimes may be employed. Note that the chemistry of fuel cells is the
reverse of water splitting. Thus discovery on the water-splitting front can directly impact
the advances needed for fuel cells and vice versa.
The photovoltaic module can be eliminated if water splitting catalysts can be directly
interfaced to a semiconductor. The one-photon, one-electron charge separation of a
semiconductor can be accumulated directly by the catalyst to attain the four equivalents
needed for water splitting. Many of the semiconducting materials are not stable in highly
16
M. Winter, R. J. Brodd, Chem. Rev., 2004, 104, 4245.
4
acidic or basic environments. Thus catalysts such as CoPi are good candidates for these
materials. In realizing an integrated catalyst-semiconductor system, the PV and
electrolyzer are effectively combined for the direct generation of a solar fuel. By
removing the photovoltaic module, the other high capex cost of a personalized energy
system is eliminated.
The use of hydrogen as a fuel requires its storage by materials with high volumetric and
gravimetric hydrogen densities. 17 For stationary applications such as personalized energy,
storage of pressurized hydrogen in tanks is a viable option. Carbon-fiber reinforced
composite tanks are light and capable of storing hydrogen at pressures up to 700 bar, and
continued advances in composite materials should lead to further improvements.
However, there is an energy penalty for gas compression (15–20% of the lower heating
value for hydrogen). 18 Compression an be circumvented with new sorbent, including
carbon-based nanostructures, solid foams, metal-organic frameworks, and a host of other
nanoporous structures. The challenge here is that hydrogen is immobilized by weak
physisorption forces and hence low temperatures (77 K) are needed for reasonable
hydrogen uptake. Storage of hydrogen at higher temperatures is possible with hydrides
owing to stronger chemical interactions. Metal hydrides have good volumetric storage
capacities but poor volumetric storage capacities owing to heavy metals. A promising
line of research is to form the hydride with lighter main-group elements such as nitrogen
and boron. 19
Personalized energy should be a major goal of national and global energy policies. It is
transformative in its scope to attain a secure energy future, to provide economic equity to people
of the developing world and to stem the flow of non-anthropogenic sources of CO2 into our
environment. The one issue that personalized energy does not currently address is a cheap energy
supply. For this reason, the realization personalized energy rests on the shoulders of science and
engineering. Discovery of new materials, new reactions and new processes are needed to permit
personalized energy to be an attractive economic alternative. If science and engineering can
decrease the cost of personalized energy through discovery, then the development of the nonlegacy world can occur within an energy infrastructure that is of the future and not the past.
Considering that it is the 6 billion non-legacy users that are driving the enormous increase in
energy demand by mid-century,1,9 a research target of personalized energy provides science and
engineering with its most direct path to providing a solution to the energy challenge. By
developing an inexpensive 24/7 solar energy system for the individual, science and engineering
will make available a carbon-neutral energy supply for 1 × 6 billion.
17
18
19
S. I. Orimo, Y. Nakamori, J. R. Eliseo A. Züttel, C. M. Jensen, Chem. Rev. 2007, 107, 4111.
S. Saylapal, J. Petrovic, C. Read, G. Thomas, G. Ordaz, Catalysis Today 2000, 120, 246.
F. H. Stephens, V. Pons, R. T. Baker, Dalton Trans. 2007, 2613.
5