Article

Cite This: Acc. Chem. Res. XXXX, XXX, XXX−XXX pubs.acs.org/accounts

Innovative Strategies for Electrocatalytic Water Splitting

Bo You* and Yujie Sun*
Department of Chemistry and Biochemistry, Utah State University, Logan, Utah 84322, United States

CONSPECTUS: Electrocatalytic water splitting driven by renewable energy input to

produce clean H2 has been widely viewed as a promising strategy of the future energy
portfolio. Currently, the state-of-the-art electrocatalysts for water splitting in acidic
solutions are IrO2 or RuO2 for the O2 evolution reaction (OER) and Pt for the H2
evolution reaction (HER). Realization of large-scale H2 production from water splitting
requires competent nonprecious electrocatalysts. Despite the advances of decades in this
field, several challenges still exist and need to be overcome: (1) Most efforts in the design
of nonprecious electrocatalysts have focused on developing HER catalysts for acidic
conditions but OER catalysts for alkaline conditions owing to their thermodynamic
convenience, potentially resulting in incompatible integration of the two types of catalysts
and thus inferior overall performance. (2) In conventional water electrolysis, HER and
OER are strictly coupled and therefore H2 and O2 are produced simultaneously, which may
lead to explosive H2/O2 mixing due to gas crossover. Meanwhile, the coexistence of H2, O2,
and electrocatalysts could produce reactive oxygen species that might shorten the lifetime of an electrolyzer. (3) The HER rate is
often limited by that of OER due to the more sluggish kinetics of the latter, which lowers the overall energy conversion efficiency.
Moreover, the product of OER, O2, is not highly valuable. (4) It remains challenging to develop efficient and low-cost H2 storage
and transport systems for the future H2 economy.
In this Account, we describe recent progress in innovative strategies to address the aforementioned four challenges in
conventional water electrolysis. These novel strategies include (1) overall water electrolysis based on bifunctional nonprecious
electrocatalysts (or precursors) to drive both HER and OER under the same conditions, (2) decoupled water electrolysis achieved
by redox mediators for temporally and spatially separating HER from OER, (3) hybrid water electrolysis by integrating
thermodynamically more favorable organic upgrading reactions to replace OER, and (4) tandem water electrolysis by utilizing
biocatalysts for converting the in situ produced H2 with foreign compounds (e.g., CO2 and N2) to more valuable products.
Finally, the remaining challenges and future perspectives are also presented. We hope this Account will function as a momentum
call for more endeavors into the development of advanced electrocatalytic systems and novel strategies for practicable H2
production from water as well as the electrocatalytic upgrading of diverse organic compounds.

1. INTRODUCTION
Along with the rise of population and the development of
economy, global energy demand will increase continuously in
the foreseeable future.1,2 The vast majority of contemporary
energy sources are derived from fossil fuels such as coal, oil, and
natural gas. However, the depletion of nonrenewable fossil fuels
and the detrimental effects of fossil fuel utilization on the
environment have stimulated intense research in innovative
technologies for the conversion and storage of sustainable and
clean energy sources, like solar and wind.1−3 To address the Figure 1. (a) Scheme of conventional water electrolyzers. (b) Water
intermittency issue of renewable energy, electrocatalytic splitting reactions under acidic and alkaline conditions.
generation of H2 from water has been considered as an
attractive approach.4−6 In order to alleviate charge transport
losses during electrochemical processes, traditional water Unfortunately, the sluggish kinetics of both H2 evolution
electrolysis is usually carried out under either acidic conditions reaction (HER) and O2 evolution reaction (OER) require high
with a proton exchange membrane (PEM) or in alkaline media overpotentials to reach appreciable current density, resulting in
with a diaphragm (Figure 1a).4,5 Depending on the reaction relatively low energy conversion efficiencies. Thus, the input
conditions, the two half-reactions for water splitting can be potential of water splitting in practical electrolyzers is much
expressed in different ways (Figure 1b). Under standard larger than 1.23 V. In addition, electrocatalysts are usually
conditions, a thermodynamic potential of 1.23 V is required to
drive electrochemical water splitting (Figure 1b), which Received: January 1, 2018
corresponds to an energy input of ΔG = 237.1 kJ mol−1.

© XXXX American Chemical Society A DOI: 10.1021/acs.accounts.8b00002

Acc. Chem. Res. XXXX, XXX, XXX−XXX
Accounts of Chemical Research Article

needed such as noble metal-based catalysts (e.g., Pt for HER

and IrO2 or RuO2 for OER) for acidic media and transition
metal-based catalysts (e.g., Ni electrodes for HER and stainless
steel composites for OER) for alkaline electrolytes (Figure 1).7
However, the prohibitive cost, scarce reserve, and unsatisfactory
stability of noble metal-based electrocatalysts, as well as the
mediocre activity of noble metal-free electrocatalysts, hinder
large-scale H2 production through water electrolysis.4−6
Indeed, enormous efforts have been made to develop
nonprecious electrocatalysts with improved HER and OER
performance.4−6 For example, metal alloys, borides, carbides,
nitrides, phosphides, and chalcogenides have been explored as
competitive HER electrocatalysts in acidic solutions,5,6 while
many nonprecious OER catalysts of metal oxides and
(oxy)hydroxides have been reported with promising activities
in alkaline media.2 Nevertheless, numerous challenges still exist:
(1) The prevalent strategies for nonprecious catalyst develop-
ment often lead to incompatible or complicated integration of
HER and OER electrocatalysts as they are usually developed
Figure 2. Four innovative strategies for nonconventional water
under different conditions.8−13 (2) HER and OER in splitting: (a) overall water electrolysis, (b) decoupled water
conventional water electrolysis are strictly coupled, which electrolysis, (c) hybrid water electrolysis, and (d) tandem water
may result in the formation of explosive H2 /O2 gas electrolysis.
mixtures.14−19 The gas crossover not only poses safety issues,
but also reduces the energy efficiency in that O2 would be water splitting in alkaline media would be better than that in
reduced back to water on the cathode side.7 Meanwhile, the acidic media. Moreover, most earth-abundant OER catalysts
coexistence of gas mixtures and electrocatalysts may generate cannot survive strong acidic electrolytes.
reactive oxygen species, which can degrade the membrane and Following our previous success in preparation of CoSx29 and
cause premature device failure.14 (3) Since OER is a four- NiSx films30 via electrodeposition, a similar potentiodynamic
electron process, it requires much higher overpotential than deposition was conducted to obtain Co−P film by consecutive
HER to afford the same current density. Ironically, the product linear scanning between −0.3 and −1.0 V vs Ag/AgCl at a scan
of OER, O2, is not highly valuable.20−25 (4) It is still challenging rate of 5 mV s−1 and a rotation speed of 500 rpm under stirring
to obtain cost-effective H2 storage and transport systems for its in the stock solution (50 mM CoSO4, 500 mM NaH2PO2, and
use on demand.26,27 100 mM CH3COONa in water).8 Scanning electron micros-
In this Account, we summarize recent advances in non- copy (SEM) image and the cross-section SEM image of Co−P
conventional water electrolysis to tackle the aforementioned reveal nearly complete coverage of Co−P on a copper substrate
four challenges. Four innovative strategies termed as overall and thickness of 1−3 μm (Figure 3a). Linear sweep
water electrolysis, decoupled water electrolysis, hybrid water voltammetry (LSV) was conducted to investigate the electro-
electrolysis, and tandem water electrolysis will be presented catalytic HER, OER, and overall water splitting performance of
(Figure 2), with an emphasis on rationally designing earth- Co−P in 1.0 M KOH. As depicted in Figure 3b, inset, the HER
abundant electrocatalysts, constructing novel electrolyzer polarization curve of Co−P demonstrates an onset potential of
modules, and tailoring the reaction interfaces. approximately −50 mV vs the reversible hydrogen electrode
(RHE), slightly more negative than that of Pt/C (∼0 mV vs
2. NONCONVENTIONAL WATER ELECTROLYSIS RHE). However, the catalytic current density of Co−P quickly
surpasses that of Pt/C beyond −167 mV vs RHE (Figure 3b),
2.1. Overall Water Electrolysis highlighting its remarkable HER activity. Co−P requires an
Based on the Nernst equation, acidic conditions are favorable overpotential of only 94 mV to reach 10 mA cm−2 and features
toward HER, while OER proceeds more easily in alkaline a small Tafel slope of 42 mV dec−1, representing an impressive
media. Therefore, most efforts in developing HER and OER performance for HER under alkaline conditions. The SEM
electrocatalysts are conducted in acidic and alkaline electrolytes, image of Co−P after 24 h electrolysis (post-HER Co−P)
respectively. However, in order to accomplish water electrol- confirms the retention of its morphology (Figure 3c), implying
ysis, both HER and OER electrocatalysts are preferred to its robust stability for long-term H2 evolution. Subsequently, its
function in the same electrolyte. The disparity of conditions OER performance was assessed in the same electrolyte (1.0 M
during the development stage of HER and OER catalysts will KOH). Co−P achieves 10 mA cm−2 at an overpotential of 345
likely result in complicated integration and mediocre perform- mV (Figure 3d), lower than those of IrO2 and many other
ance for full water splitting.28 With these considerations in nonprecious OER catalysts. In contrast to the uniform
mind, our group is interested in developing nonprecious morphology of the fresh and post-HER Co−P, the SEM
bifunctional electrocatalysts for both HER and OER in the image of the post-OER Co−P indicates particle aggregates
same electrolyte to achieve efficient overall water electrolysis (Figure 3e). Further characterizations reveal the oxidation of
(Figure 2a), due to the advantages of simplifying the both Co and P during OER.8 Therefore, the real active sites for
electrolyzer configuration and reducing the overall cost. OER are attributed to the in situ generated cobalt oxides and
Owing to the nature of a four-electron process, the over- (oxy)hydroxides on the bulk Co−P film. Finally, a water
potential of OER is typically much higher than that of HER to electrolyzer using Co−P as the electrocatalyst (or catalyst
afford same current density, therefore the overall efficiency of precursor) on both anode and cathode was assembled and
B DOI: 10.1021/acs.accounts.8b00002
Acc. Chem. Res. XXXX, XXX, XXX−XXX
Accounts of Chemical Research Article

also explored nickel sulfides as bifunctional electrocatalysts for

overall water splitting. High-temperature sulfurization of Ni/
NF can readily achieve 3D hierarchically porous NiS x
superstructures (hp-NiSx, Figure 4c). The hp-NiSx possesses
both macropores and mesopores, in contrast to Ni2P/Ni/NF
with singular macropores. The hp-NiSx electrocatalyst couple
only requires a voltage of 1.47 V to reach 10 mA cm−2 for
overall water electrolysis (Figure 4f).
Benefiting from high specific surface area, controllable pore
texture, and modular organic and inorganic components,
metal−organic frameworks (MOFs) represent promising
candidates to obtain competent electrocatalysts.12,33,34 We
first reported a MOF-derived route to produce mesoporous
Co−P/NC nanopolyhedrons composed of CoPx nanoparticles
embedded in N-doped carbon matrices as bifunctional
electrocatalysts for both HER and OER.12 The Co−P/NC
was prepared by direct carbonization of Co-based zeolitic
imidazolate frameworks (ZIF-67) followed by phosphorization
(Figure 4d). SEM images of Co−P/NC suggest its inheritance
of the polyhedron-like morphology of ZIF-67 (Figure 4e).
Elemental mapping results (Figure 4e) confirm the presence of
Co, P, N, and C in Co−P/NC, and P is highly localized with
Co, consistent with the formation of CoPx. The specific surface
area of Co−P/NC is 183 m2 g−1, larger than that of hp-NiSx
(11 m2 g−1).11 As expected, Co−P/NC exhibits remarkable
HER and OER activities and affords a current density of 165
mA cm−2 at 2.0 V for overall water splitting.12 Figure 4f
compares the cell voltages of overall water electrolysis based on
our reported bifunctional electrocatalysts at different current
densities, including the Pt/C and RuO2 couple as a reference.
Such a bifunctional electrocatalyst strategy for overall water
Figure 3. SEM images of (a) fresh Co−P, (c) post-HER Co−P, and
splitting can be readily extended to other transition metal and
(e) post-OER Co−P. LSV curves for (b) HER, (d) OER, and (f)
overall water splitting. Reproduced from ref 8. Copyright 2015 John noble metal derivatives and even metal-free nanocarbons.35
Wiley & Sons, Inc. 2.2. Decoupled Water Electrolysis
In overall water electrolysis, the production rate of H2 is still
tested in 1.0 M KOH. The Co−P catalyst couple demonstrates strictly limited by the rate of O2 evolution, as HER and OER
a cell voltage of only 1.744 V to afford 100 mA cm−2, even take place simultaneously. However, sometimes it is more
lower than that of the Pt/C and IrO2 couple (Figure 3f). desirable to produce H2 at a much higher rate with a moderate
Analogous bifunctional performance for overall water splitting voltage input. Decoupled water electrolysis achieved by redox
can be obtained from electrodeposited Ni−P film as well.9 mediators can temporally and spatially separate HER and OER
In order to accelerate electrocatalytic kinetics and lower the (Figure 2b), eliminating H2/O2 mixing and the formation of
required overpotential, an electrocatalyst with optimal reaction reactive oxygen species. Conceptually, when HER takes place
interfaces is critical for both triphase HER and OER (solid, on the cathode, the counter reaction is the oxidation of the
liquid, and gas).10−12 Within this scenario, catalysts featuring redox mediator instead of OER. Subsequently, the oxidized
large specific surface area, open porous structure, and high redox mediator will be reduced back to its original state at the
accessibility of active sites are anticipated to be beneficial to same electrode, which couples O2 evolution on another
electrocatalytic performance.10−12 Therefore, we synthesized working electrode. Overall, the redox mediator functions as a
three-dimensional (3D) hierarchically porous Ni2P super- reversible electron (and proton) donor/acceptor.
structures on Ni foam (Ni2P/Ni/NF) as nonprecious bifunc- Cronin’s group reported decoupled water electrolysis in
tional electrocatalysts for overall water splitting.10 Ni2P/Ni/NF acidic electrolytes using a series of electron-coupled proton
was prepared through a template-free electrodeposition of buffers such as silicotungstic acid,14 phosphomolybdic acid,15
porous metallic Ni microsphere arrays on a Ni foam (Ni/NF) and quinone derivatives.17 However, acidic medium is highly
followed by phosphorization (Figure 4a). SEM images corrosive and also prohibits the utilization of nonprecious OER
corroborate its hierarchical macroporosity and urchin-like catalysts. Our group recently reported that a ferrocene
morphology (Figure 4b). The cell voltage of a Ni2P/Ni/NF- derivative, (ferrocenylmethyl)trimethylammonium chloride
based electrolyzer at 10 mA cm−2 is 1.49 V (Figure 4f), much (FcNCl, Figure 5b), which is highly soluble in water, can act
lower than those of Co−P (1.64 V)8 and Ni−P (1.67 V).9 The as a proton-independent electron reservoir to realize decoupled
exceptional activity of Ni2P/Ni/NF is ascribed to its unique water electrolysis in neutral electrolyte (0.5 M Na2SO4). Due to
superstructure. The 3D hierarchically macroporous structure the benign conditions, nonprecious electrocatalysts like Ni2P,
with interconnected configuration not only provides high Co−P, and metallic Ni could be adopted to drive both HER
accessibility of active sites but also buffers electrolyte to and OER. As illustrated in Figure 5a,16 in step 1, the HER
facilitate ion transport and gas diffusion.31,32 In addition, we electrode is connected and a negative voltage bias is applied to
C DOI: 10.1021/acs.accounts.8b00002
Acc. Chem. Res. XXXX, XXX, XXX−XXX
Accounts of Chemical Research Article

Figure 4. (a) Preparation scheme of Ni2P/Ni/NF. SEM images of (b) Ni2P/Ni/NF and (c) hp-NiSx. (d) Preparation scheme of Co−P/NC. (e)
SEM and element mapping images of Co−P/NC. (f) Comparison of cell voltages to achieve benchmark current densities for diverse bifunctional
electrocatalysts of overall water splitting.8−13,21,23 Panel b reproduced from ref 10. Copyright 2016 American Chemical Society. Panel c reproduced
from ref 11. Copyright 2016 John Wiley & Sons, Inc. Panels d and e reproduced from ref 12. Copyright 2015 American Chemical Society.

initiate H2 generation, and simultaneously FcNCl in the also discovered that Na4[Fe(CN)6] can function as an electron
counter chamber is oxidized because of its less positive reservoir in both neutral and alkaline electrolytes.16 As alluded
oxidation potential than that of OER (Figure 5c). Figure 5d in the above discussion, an ideal molecular redox mediator
shows the LSV curves of HER on the Ni2P/Ni/NF cathode should be low cost and highly soluble in water and features a
with and without FcNCl in the counter compartment. In the fast and reversible redox couple that is positioned between the
absence of FcNCl, Ni2P/Ni/NF exhibits an onset voltage of onset potentials of HER and OER. Besides molecular
−2.2 V. Upon the addition of 50 mM FcNCl in the counter complexes, Ni(OH)2 can also act as a solid-state redox
chamber, the onset voltage shifts to −1.2 V, indicating that the mediator for decoupled water electrolysis, even though the
oxidation of FcNCl is significantly easier than OER and nearly operation is rather complicated and its redox capacity is largely
1 V voltage input is saved. Subsequently, in step 2, the OER determined by the electrode area and mass loading.18,19
electrode (a Ni foam) is connected, and a positive voltage bias 2.3. Hybrid Water Electrolysis
is applied to drive O2 evolution on the Ni foam, while the Following the aforementioned strategy of decoupled water
oxidized FcNCl+ is reduced back to FcNCl in the counter electrolysis, OER still takes place as the oxidation reaction to
chamber. Because the reduction potential of FcNCl+ is more complete the electrocatalysis cycle, which requires large
positive than that of HER, the required voltage for OER with overpotential, and its product O2 does not bear a high price
FcNCl in the counter chamber is reduced about 1.8 V relative tag. The primary purpose of water oxidation is to extract
to the case without FcNCl (Figure 5e). Gas chromatography electrons and transport them to the cathode side for the
(GC) analysis confirmed 100% Faradaic efficiencies for both production of H2, which is the real desirable product of water
steps, and only one gas product was produced each time. The splitting. O2 is merely a side product of the whole process, even
similarity of 20 electrolysis cycles (50 h) indicates the great though it is responsible for the major energy loss of water
robustness and negligible crossover of FcNCl as an electron electrolysis. In response, we proposed a concept of hybrid water
reservoir for long-term decoupled water electrolysis (Figure 5f). electrolysis by replacing OER with thermodynamically more
When integrated with photovoltaic cells, decoupled water favorable organic oxidation reactions, wherein biomass-derived
electrolysis can also be driven by sunlight irradiation without intermediates can serve as attractive organic substrates (Figure
external voltage bias (Figure 5g).16 In addition to FcNCl, we 2c).21−24 Such a hybrid water electrolysis strategy possesses at
D DOI: 10.1021/acs.accounts.8b00002
Acc. Chem. Res. XXXX, XXX, XXX−XXX
Accounts of Chemical Research Article

Figure 5. (a) Scheme of FcNCl-mediated decoupled water electrolysis. (b) Chemical structure of FcNCl. (c) Cyclic voltammograms of 50 mM
FcNCl (red), HER on glassy carbon (black) and Ni2P/Ni/NF (green), and OER on glassy carbon (black) and Ni foam (blue) in 0.5 M Na2SO4.
LSV curves of (d) HER on Ni2P/Ni/NF and (e) OER on Ni foam in a two-electrode configuration in 0.5 M Na2SO4 with and without 50 mM
FcNCl in the counter chamber. (f) Charge evolution for repeated decoupled water electrolysis in 0.5 M Na2SO4 with 10 mM FcNCl in the counter
compartment. Voltage bias between the working and counter electrodes was alternated at −1.6 V for HER and 1.8 V for OER. (g) HER current
density on Ni2P/Ni/NF with and without 10 mM FcNCl in the counter chamber driven by a solar cell under chopped sunlight irradiation.
Reproduced from ref 16. Copyright 2018 Elsevier.

least four advantages: (1) Owing to the more favorable that of OER (Figure 6c), saving nearly 200 mV voltage input to
thermodynamics of selected organic oxidation reactions, such a reach the same current density of 100 mA cm−2. The HMF
hybrid electrolyzer can deliver higher current density with conversion and product evolution over time are plotted in
lower voltage input, hence increasing the energy conversion Figure 6d, confirming the nearly 100% yield of FDCA and its
efficiency. (2) The anode side will yield value-added organic unity Faradaic efficiency. The goal of simultaneously producing
products, maximizing the return of energy investment. (3) H2 and FDCA necessitates great electrocatalytic activity and
Since no O2 will be produced, no H2/O2 mixing and reactive selectivity of Ni2P NPA/NF toward H2 production even in the
oxygen species issues exist. (4) The high selectivity of H2 presence of HMF. Indeed, as demonstrated in Figure 6e, the
production on the cathode side in the presence of organic cathodic LSV curves of Ni2P NPA/NF display negligible
substrates enables single-compartment membrane-free electrol- difference with regard to the addition of 10 mM HMF. The
ysis, substantially reducing the cost of electrolyzers. In order to robust stability and preference of Ni2P NPA/NF for HER
achieve these advantages, an ideal organic substrate should versus HMF reduction is further corroborated by a 12 h
satisfy the following requirements: high solubility in water at chronopotentiometry experiment conducted at −10 mA cm−2
room temperature, production of value-added nongaseous with 10 mM HMF, showing 100% Faradaic efficiency of H2
product(s) through oxidation, less positive oxidation potential
production.21 The foregoing results were obtained in a three-
at the anode relative to the onset of OER, and negligible
electrode configuration. When a two-electrode electrolyzer
competition with HER at the cathode. Among many biomass-
employs Ni2P NPA/NF as the electrocatalysts for both cathode
derived intermediates, 5-hydroxymethylfurfural (HMF) stands
out as a promising candidate because it can be transformed to and anode, simultaneous production of FDCA and H2 can be
various valuable products, including 2,5-furandicarboxylic acid realized. As compared in Figure 6f, the LSV curve of HMF
(FDCA) via oxidation (Figure 6a).20−25 In fact, FDCA has oxidation integrated with HER takes off at much smaller voltage
been advocated as a replacement of terephthalic acid for the input than that of pure water splitting. Nearly 200 mV voltage
production of polyamides, polyesters, and polyurethanes.25 can be universally saved to reach benchmark current densities
To our delight, a reported nonprecious bifunctional electro- of 10, 20, and 50 mA cm−2 (Figure 6g). Long-term controlled
catalyst for overall water splitting is able to readily catalyze the potential electrolysis conducted at 1.5 V produces H2 with a
oxidation of HMF to FDCA in 1.0 M KOH.21 SEM and yield close to the theoretical predicted amount (Figure 6h),
elemental mapping images in Figure 6b present the indicative of a unity Faradaic efficiency for HER. In fact,
morphology and elemental distribution of the model electro- continuously running three repetitive electrolysis cycles using
catalyst, 3D Ni2P nanoparticle arrays coated on nickel foam the same Ni2P NPA/NF catalyst couple shows no degradation
(Ni2P NPA/NF). Linear sweep voltammetry study clearly of the electrocatalysts, and they retain great activity, selectivity,
demonstrates that in the presence of 10 mM HMF, the anodic and robustness for both HER and HMF oxidation to FDCA
current takes off at a potential substantially less positive than (Figure 6i).
E DOI: 10.1021/acs.accounts.8b00002
Acc. Chem. Res. XXXX, XXX, XXX−XXX
Accounts of Chemical Research Article

Figure 6. (a) Two possible pathways of HMF oxidation to FDCA. (b) SEM and elemental mapping images of Ni2P NPA/NF. (c) LSV curves of
Ni2P NPA/NF in 1.0 M KOH with and without 10 mM HMF. (d) HMF conversion and product evolution during electrolysis at 1.423 V vs RHE in
1.0 M KOH with 10 mM HMF. (e) LSV curves of Ni2P NPA/NF in 1.0 M KOH with and without 10 mM HMF. (f) LSV curves and (g)
comparison of the cell voltages to achieve benchmark current densities using a Ni2P NPA/NF catalyst couple in 1.0 M KOH with and without 10
mM HMF. (h) GC-measured H2 quantity compared with that theoretically calculated assuming a 100% Faradaic efficiency. (i) Faradaic efficiencies
of the Ni2P NPA/NF catalyst couple for simultaneous H2 and FDCA generation in 1.0 M KOH with 10 mM HMF for three successive electrolysis
cycles. Reproduced from ref 21. Copyright 2016 John Wiley & Sons, Inc.

Encouraged by the above results and thanks to the modular would expect an electron-donating moiety such as a methyl
nature of hybrid water electrolysis, we were able to explore the group should enable the oxidation to occur at a more negative
electrocatalytic upgrading of a diverse array of biomass-derived potential while an electron-withdrawing group like a −NO2
intermediate compounds (e.g., ethanol, benzyl alcohol, furfural, group should shift the oxidation potential more positively.
and furfuryl alcohol) and the utilization of nonprecious However, regardless of the different substituents at the para-
bifunctional electrocatalysts (e.g., transition metals, sulfides, position of the phenyl ring in benzyl alcohol, these three
phosphides, etc.).23,24 In fact, the large anodic potential substrates exhibit almost identical LSV curves for oxidation on
requirement of OER leaves us a wide potential window to
hp-Ni (Figure 7i). These preliminary results prompt us to
carry out desirable organic oxidation reactions without the
formation of O2 in aqueous media. For instance, Figure 7a−e hypothesize that these alcohol oxidation reactions on solid-state
presents five representative organic oxidation reactions together electrocatalysts like hp-Ni is primarily determined by the
with their corresponding LSV curves on 3D Ni3S2/NF (Figure potential to achieve the desirable oxidation state of the catalyst,
7f). All of these reactions take place at potentials less positive rather than the intrinsic thermodynamics of each organic
than that of OER, consistent with their more favorable oxidation, in contrast to typical homogeneous oxidation
thermodynamics under alkaline conditions. One interesting reactions catalyzed by molecular catalysts.
observation is that their catalytic currents take off at very similar 2.4. Tandem Water Electrolysis
potentials (∼1.36 V vs RHE), regardless of the intrinsic
thermodynamic difference of each organic reaction. In order to One challenge of the wide deployment of H2 utilization is its
avoid the potential influence of nonmetal elements in the cost-effective storage and transport.36,37 A potential strategy to
catalyst composition, we purposely fabricated a highly porous bypass this challenge is directly consuming H2 during water
Ni foam (hp-Ni) as the working electrode (Figure 7g). electrolysis to produce value-added chemical fuels (e.g., CH4,
Subsequently, benzyl alcohol with different electronic sub- NH3, etc.) via biological catalysts (i.e., enzymes); within this
stituents was selected as the organic substrate (Figure 7h). One context tandem water electrolysis emerges (Figure 2d).
F DOI: 10.1021/acs.accounts.8b00002
Acc. Chem. Res. XXXX, XXX, XXX−XXX
Accounts of Chemical Research Article

Figure 7. (a−e) Oxidation of selected organics to value-added products and their corresponding LSV curves on Ni3S2/NF in 1.0 M KOH (BA,
benzyl alcohol; FFA, furfuryl alcohol; FF, furfural; HMF, 5-hydroxymethylfurfural). SEM images of (f) Ni3S2/NF and (g) hp-Ni. (h) Chemical
structures of MBA, BA, and NBA. (i) LSV curves on hp-Ni in 1.0 M KOH with 10 mM MBA, BA, and NBA. Panels a−f reproduced from ref 23.
Copyright 2016 American Chemical Society. Panels g−i reproduced from ref 24. Copyright 2017 American Chemical Society.

Figure 8. (a) Scheme of tandem water electrolysis. (b, c) CH4 production and Faradaic efficiency with varying applied current. (d) High-resolution
mass spectrometry analysis of gas products. (e) Optimized structure of water adsorption on N−Ni(111). (f) Adsorption energy of water on Ni(111)
and N−Ni(111). (g) Free energy barrier profiles of water dissociation on Ni(111) and N−Ni(111). (h) Comparison of the current densities at
different overpotentials for N−Ni, Ni framework, and Pt/C for HER under neutral conditions. Panels a−d reproduced from ref 36. Copyright 2015
National Academy of Sciences of the United States of America. Panels e−h reproduced from ref 39. Copyright 2017 American Chemical Society.

Figure 8a schematically depicts the concept of tandem water currents (1.0−7.5 mA) with an average Faradaic efficiency
electrolysis, wherein an airtight two-compartment electro- above 74% (Figure 8b). The proportion of CH4 production to
chemical cell is charged with Methanosarcina barkeri (M. the applied current suggests that electrocatalytic H2 production
barkeri) and CO2 in the cathodic chamber.36 Sustainable is the rate-limiting step of the entire system. As M. barkeri
electricity or solar input or both drives water splitting using operates at a nearly thermodynamic potential, the only
inorganic catalysts to produce H2, which is consumed by M. overpotential involved is due to HER in neutral aqueous
barkeri in the presence of CO2 to produce CH4. After medium (1.0 M phosphate buffer). Such a hybrid bioinorganic
galvanostatic electrolysis for 11.5 h and sampling for 0.5 h, system for tandem water electrolysis shows robust durability,
the GC analysis of gas products revealed that CH4 production producing CH4 at a constant rate over 7 days with a Faradaic
linearly and cumulatively increased under all the applied efficiency close to 75% (Figure 8c). The carbon source for CH4
G DOI: 10.1021/acs.accounts.8b00002
Acc. Chem. Res. XXXX, XXX, XXX−XXX
Accounts of Chemical Research Article

is indeed the original CO2 as confirmed by an isotope labeling most nonprecious bifunctional electrocatalysts for overall water
experiment. When 13CO2 was utilized during electrocatalysis, splitting only function well in alkaline (and neutral) electro-
only 13CH4 was detected via high-resolution mass spectrometry lytes. Nearly no low-cost electrocatalyst couples have been
(Figure 8d), whereas only 12CH4 was observed if 12CO2 used reported for water electrolysis at low pH. Along with the
instead. In terms of electrocatalysts for the in situ H2 success in proton-exchange membrane and high conductivity of
production, this hybrid system possesses high flexibility. A acidic electrolytes, it is very appealing to conduct water
variety of earth-abundant and biocompatible inorganic electro- electrolysis under low pH conditions with earth-abundant
catalysts can be utilized, such as NiS and NiMo. Once electrocatalysts. In addition, most nonprecious electrocatalysts
integrated with photoelectrodes, direct solar-to-chemical will undergo surface (if not bulk) oxidation during OER; hence
conversion is also feasible based upon this tandem water the real active sites are usually different from those on the
electrolysis strategy.36 original catalysts. How to rationally control and understand
Recently, Nocera et al. reported a similar hybrid water such in situ oxidation to yield active sites while still retaining
splitting−biosynthetic system utilizing Co−P as the inorganic highly conductive bulk composite necessitates further research.
HER catalyst and Ralstonia eutropha as the biocatalyst.37 The The synthetic cost of redox mediators should also be further
overall energy conversion efficiency was reported to be higher reduced to be economically attractive for large-scale decoupled
than that of photosynthesis. An alternative tandem water water electrolysis. In particular, solubility, stability, and fast
electrolysis system was also reported by the same group, electron transfer kinetics of redox mediators are critical factors
wherein H2-oxidizing autotrophic bacterium Xanthobacter determining their potential for practical applications. For hybrid
autotrophicus was employed and NH3 was produced from N2 water electrolysis, high selectivity for desirable products from
and H2O.38 those selected organic oxidations should be the primary goal,
In order to be fully compatible with biological systems for and the subsequent separation and purification should also be
tandem water electrolysis, an ideal reaction medium is neutral taken into account. As organic electrocatalysis and electrosyn-
water. However, most earth-abundant HER electrocatalysts thesis is about to witness a renaissance, we are optimistic that
exhibit mediocre performance under neutral condi- integrating many promising organic oxidation reactions (e.g.,
tions.11,29,30,39 Our group recently reported a general surface oxidative C−C coupling) will burgeon in this field.40 It is
nitrogen modification strategy to considerably enhance the exciting to see recent successes in tandem water electrolysis
activity of first-row transition metals (e.g., Fe, Co, Ni, Cu, and integrating biological enzymes to yield highly valuable products
NiCo alloy) for electrocatalytic H2 evolution in neutral aqueous from inert starting molecules like CO2 and N2. Nevertheless,
buffer.39 In particular, the resulting N−Ni electrocatalyst rivals the long-term stability, scalability, and overall cost of these
commercially available Pt/C for HER at pH 7. Detailed physical bioinorganic electrocatalytic systems should be thoroughly
characterizations reveal that the bulk composite of N−Ni analyzed prior to practical implementation. We hope this
retains metallic nickel, whereas nitrogen only exists on the Account will encourage more efforts into the development of
catalyst surface in the formation of “surface nitrides”. Density advanced electrocatalysts and the design of novel strategies for
functional theory computation suggests that the surface H2 production and more importantly organic upgrading under
nitrogen is likely positioned at the hollow fcc site of nickel benign conditions.
and bound to three surface nickel atoms (Figure 8e). It is
widely acknowledged that water adsorption and dissociation are
critical steps for HER under neutral and alkaline conditions. An
■ AUTHOR INFORMATION
Corresponding Authors
ideal HER electrocatalyst should have sufficient affinity to bind
water to accelerate the initial electron-transfer process but also *E-mail: [email protected].
possess enough repellency to facilitate the subsequent water *E-mail: [email protected].
dissociation. Figure 8f compares the adsorption energy (Eads) of ORCID
water on Ni (−0.30 eV) vs N−Ni (−0.41 eV). It is apparent Bo You: 0000-0003-1849-0418
that the presence of N in N−Ni facilitates water adsorption Yujie Sun: 0000-0002-4122-6255
wherein the hydrogen bond is naturally invoked. Furthermore,
Notes
the free energy barrier for water dissociation on N−Ni is
predicted theoretically at 0.42 eV (Figure 8g), also substantially The authors declare no competing financial interest.
lower than that on pure nickel (0.84 eV). Taken together, the Biographies
surface nitrogen modification has been experimentally and
theoretically proven as an effective approach to boost HER Bo You received his B.S. degree in Chemistry in 2008 from
under neutral conditions. As highlighted in Figure 8h, N−Ni Heilongjiang University and a Ph.D. degree under the supervision of
even exhibits superior HER activity at pH 7 to that of Pt/C, not Prof. Zhaoxiang Deng at University of Science and Technology of
even to mention pure nickel and other nonprecious electro- China (USTC) in 2014. After the postdoctoral positions in Prof. Yujie
catalysts reported so far, rendering N−Ni an excellent HER Sun’s group at Utah State University and Prof. Hong Li’s group at
electrocatalyst to integrate with many biological systems for Nanyang Technological University, he joined Prof. Shi-Zhang Qiao’s
tandem water electrolysis and beyond. group as an ARC Research Associate at The University of Adelaide.
His research interests include inorganic catalysts for advanced
3. SUMMARY AND OUTLOOK renewable energy (ICARE).
This Account has summarized recent advances in four Yujie Sun obtained his B.S. degree in Chemistry from Fudan
innovative strategies for nonconventional water splitting, University in 2005. He then pursued graduate studies with Prof.
circumventing many challenges in traditional water electrolysis Claudia Turro at The Ohio State University and was awarded a Ph.D.
and showcasing various advantages. In this burgeoning field, degree in 2010. Subsequently, he conducted a postdoctoral stint with
many opportunities await further investigation. For instance, Prof. Christopher J. Chang at the University of California, Berkeley,

H DOI: 10.1021/acs.accounts.8b00002
Acc. Chem. Res. XXXX, XXX, XXX−XXX
Accounts of Chemical Research Article

and Lawrence Berkeley National Laboratory. He started his Reactions of Electrolytic Water Splitting. J. Am. Chem. Soc. 2013, 135,
independent academic career at Utah State University in 2013. His 13656−13659.
group is interested in developing and understanding inexpensive (18) Chen, L.; Dong, X.; Wang, Y.; Xia, Y. Separating Hydrogen and
materials and complexes with applications related to energy and Oxygen Evolution in Alkaline Water Electrolysis Using Nickel
Hydroxide. Nat. Commun. 2016, 7, 11741.
health.
(19) Landman, A.; Dotan, H.; Shter, G. E.; Wullenkord, M.; Houaijia,

■ ACKNOWLEDGMENTS
Y.S. acknowledges the financial support of Utah State
A.; Maljusch, A.; Grader, G. S.; Rothschild, A. Photoelectrochemical
Water Splitting in Separate Oxygen and Hydrogen Cells. Nat. Mater.
2017, 16, 646−651.
(20) Han, G.; Jin, Y. H.; Burgess, R. A.; Dickenson, N. E.; Cao, X. M.;
University, USTAR of the State of Utah, Ralph E. Powe Junior Sun, Y. Visible-Light-Driven Valorization of Biomass Intermediates
Faculty Enhancement Award from Oak Ridge Associated Integrated with H2 Production Catalyzed by Ultrathin Ni/CdS
Universities, Inc., and National Science Foundation (CAREER Nanosheets. J. Am. Chem. Soc. 2017, 139, 15584−15587.
Award, CHE-1653978). (21) You, B.; Jiang, N.; Liu, X.; Sun, Y. Simultaneous H2 Generation

■
and Biomass Upgrading in Water by An Efficient Noble-Metal-Free
Bifunctional Electrocatalyst. Angew. Chem., Int. Ed. 2016, 55, 9913−
REFERENCES 9917.
(1) Lewis, N. S.; Nocera, D. G. Powering the Planet: Chemical (22) Jiang, N.; You, B.; Boonstra, R.; Terrero Rodriguez, I. M.; Sun,
Challenges in Solar Energy Utilization. Proc. Natl. Acad. Sci. U. S. A. Y. Integrating Electrocatalytic 5-Hydroxymethylfurfural Oxidation and
2006, 103, 15729−15735. Hydrogen Production via Co-P-Derived Electrocatalysts. ACS Energy
(2) Hunter, B. M.; Gray, H. B.; Müller, A. M. Earth-Abundant Lett. 2016, 1, 386−390.
Heterogeneous Water Oxidation Catalysts. Chem. Rev. 2016, 116, (23) You, B.; Liu, X.; Jiang, N.; Sun, Y. A General Strategy for
14120−14136. Decoupled Hydrogen Production from Water Splitting by Integrating
(3) Chu, S.; Majumdar, A. Opportunities and Challenges for A Oxidative Biomass Valorization. J. Am. Chem. Soc. 2016, 138, 13639−
Sustainable Energy Future. Nature 2012, 488, 294−303. 13646.
(4) Jiao, Y.; Zheng, Y.; Jaroniec, M.; Qiao, S. Z. Design of (24) You, B.; Liu, X.; Liu, X.; Sun, Y. Efficient H2 Evolution Coupled
Electrocatalysts for Oxygen- and Hydrogen-Involving Energy Con- with Oxidative Refining of Alcohols via A Hierarchically Porous Nickel
version Reactions. Chem. Soc. Rev. 2015, 44, 2060−2086. Bifunctional Electrocatalyst. ACS Catal. 2017, 7, 4564−4570.
(5) You, B.; Sun, Y. Chalcogenide and Phosphide Solid-State (25) Cha, H. G.; Choi, K. S. Combined Biomass Valorization and
Catalysts for H2 Generation. ChemPlusChem 2016, 81, 1045−1055. Hydrogen Production in A Photoelectrochemical Cell. Nat. Chem.
(6) Zhu, Y. P.; Guo, C.; Zheng, Y.; Qiao, S. Z. Surface and Interface 2015, 7, 328−333.
Engineering of Noble-Metal-Free Electrocatalysts for Efficient Energy (26) Murray, L. J.; Dincǎ, M.; Long, R. R. Hydrogen Storage in
Conversion Processes. Acc. Chem. Res. 2017, 50, 915−923. Metal-Organic Frameworks. Chem. Soc. Rev. 2009, 38, 1294−1314.
(7) Carmo, M.; Fritz, D. L.; Mergel, J.; Stolten, D. A Comprehensive (27) Struzhkin, V. V.; Militzer, B.; Mao, W. L.; Mao, H.; Hemley, R.
Review on PEM Water Electrolysis. Int. J. Hydrogen Energy 2013, 38, J. Hydrogen Storage in Molecular Clathrates. Chem. Rev. 2007, 107,
4901−4934. 4133−4151.
(8) Jiang, N.; You, B.; Sheng, M.; Sun, Y. Electrodeposited Cobalt- (28) Luo, J.; Vermaas, D. A.; Bi, D.; Hagfeldt, A.; Smith, W. A.;
Phosphorous-Derived Films as Competent Bifunctional Catalysts for Grätzel, M. Bipolar Membrane-Assisted Solar Water Splitting in
Overall Water Splitting. Angew. Chem., Int. Ed. 2015, 54, 6251−6254. Optimal pH. Adv. Energy Mater. 2016, 6, 1600100.
(9) Jiang, N.; You, B.; Sheng, M.; Sun, Y. Bifunctionality and (29) Sun, Y.; Liu, C.; Grauer, D. C.; Yano, J.; Long, J. R.; Yang, P.;
Mechanism of Electrodeposited Nickel-Phosphorous Films for Chang, C. J. Electrodeposited Cobalt-Sulfide Catalyst for Electro-
Efficient Overall Water Splitting. ChemCatChem 2016, 8, 106−112. chemical and Photoelectrochemical Hydrogen Generation from Water.
(10) You, B.; Jiang, N.; Sheng, M.; Bhushan, M. W.; Sun, Y. J. Am. Chem. Soc. 2013, 135, 17699−17702.
Hierarchically Porous Urchin-Like Ni2P Superstructures Supported on (30) Jiang, N.; Bogoev, L.; Popova, M.; Gul, S.; Yano, J.; Sun, Y.
Nickel Foam as Efficient Bifunctional Electrocatalysts for Overall Electrodeposited Nickel-Sulfide Films as Competent Hydrogen
Water Splitting. ACS Catal. 2016, 6, 714−721. Evolution Catalysts in Neutral Water. J. Mater. Chem. A 2014, 2,
(11) You, B.; Sun, Y. Hierarchically Porous Nickel Sulfide 19407−19414.
Multifunctional superstructures. Adv. Energy Mater. 2016, 6, 1502333. (31) Li, W.; Liu, J.; Zhao, D. Mesoporous Materials for Energy
(12) You, B.; Jiang, N.; Sheng, M.; Gul, S.; Yano, J.; Sun, Y. High- Conversion and Storage Devices. Nat. Rev. Mater. 2016, 1, 16023.
Performance Overall Water Splitting Electrocatalysts Derived from (32) You, B.; Jiang, N.; Sheng, M.; Sun, Y. Microwave vs.
Cobalt-Based Metal-Organic Frameworks. Chem. Mater. 2015, 27, Solvothermal Synthesis of Hollow Cobalt Sulfide Nanoprisms for
7636−7642. Electrocatalytic Hydrogen Evolution and Supercapacitors. Chem.
(13) Liu, X.; You, B.; Sun, Y. Facile Surface Modification of Commun. 2015, 51, 4252−4255.
Ubiquitous Stainless Steel Led to Competent Electrocatalysts for (33) You, B.; Jiang, N.; Sheng, M.; Drisdell, W. S.; Yano, J.; Sun, Y.
Overall Water Splitting. ACS Sustainable Chem. Eng. 2017, 5, 4778− Bimetal-Organic Framework Self-Adjusted Synthesis of Support-Free
4784. Nonprecious Electrocatalysts for Efficient Oxygen Reduction. ACS
(14) Rausch, B.; Symes, M. D.; Chisholm, G.; Cronin, L. Decoupled Catal. 2015, 5, 7068−7076.
Catalytic Hydrogen Evolution from A Molecular Metal Oxide Redox (34) Lee, K. J.; Lee, J. H.; Jeoung, S.; Moon, H. R. Transformation of
Mediator in Water Splitting. Science 2014, 345, 1326−1330. Metal-Organic Frameworks/Coordination Polymers into Functional
(15) Symes, M. D.; Cronin, L. Decoupling Hydrogen and Oxygen Nanostructured Materials: Experimental Approaches Based on
Evolution during Electrolytic Water Splitting Using An Electron- Mechanistic Insights. Acc. Chem. Res. 2017, 50, 2684−2692.
Coupled-Proton Buffer. Nat. Chem. 2013, 5, 403−409. (35) Jamesh, M. I. Recent Progress on Earth Abundant Hydrogen
(16) Li, W.; Jiang, N.; Hu, B.; Liu, X.; Song, F.; Han, G.; Jordan, T. J.; Evolution Reaction and Oxygen Evolution Reaction Bifunctional
Hanson, T. B.; Liu, T. L.; Sun, Y. Electrolyzer Design for Flexible Electrocatalyst for Overall Water Splitting in Alkaline Media. J. Power
Decoupled Water Splitting and Organic Upgrading with Electron Sources 2016, 333, 213−236.
Reservoirs. Chem. 2018, 4, 637−649, DOI: 10.1016/ (36) Nichols, E. M.; Gallagher, J. J.; Liu, C.; Su, Y.; Resasco, J.; Yu, Y.;
j.chempr.2017.12.019. Sun, Y.; Yang, P.; Chang, M. C. Y.; Chang, C. J. Hybrid Bioinorganic
(17) Rausch, B.; Symes, M. D.; Cronin, L. A Bio-Inspired, Small Approach to Solar-to-Chemical Conversion. Proc. Natl. Acad. Sci. U. S.
Molecule Electron-Coupled-Proton Buffer for Decoupling the Half- A. 2015, 112, 11461−11466.

I DOI: 10.1021/acs.accounts.8b00002
Acc. Chem. Res. XXXX, XXX, XXX−XXX
Accounts of Chemical Research Article

(37) Liu, C.; Colón, B. C.; Ziesack, M.; Silver, P. A.; Nocera, D. G.
Water Splitting-Biosynthetic System with CO2 Reduction Efficiencies
Exceeding Photosynthesis. Science 2016, 352, 1210−1213.
(38) Liu, C.; Sakimoto, K. K.; Colón, B. C.; Silver, P. A.; Nocera, D.
G. Ambient Nitrogen Reduction Cycle Using A Hybrid Inorganic-
Biological System. Proc. Natl. Acad. Sci. U. S. A. 2017, 114, 6450−6455.
(39) You, B.; Liu, X.; Hu, G.; Gul, S.; Yano, J.; Jiang, D. E.; Sun, Y.
Universal Surface Engineering of Transition Metals for Superior
Electrocatalytic Hydrogen Evolution in Neutral Water. J. Am. Chem.
Soc. 2017, 139, 12283−12290.
(40) Shi, W.; Liu, C.; Lei, A. Transition-Metal Catalyzed Oxidative
Cross-Coupling Reaction to Form C-C Bonds Involving Organo-
metallic Reagents as Nucleophiles. Chem. Soc. Rev. 2011, 40, 2761−
2776.

J DOI: 10.1021/acs.accounts.8b00002
Acc. Chem. Res. XXXX, XXX, XXX−XXX