Fuel Production from Seawater and Fuel Cells UsingSeawaterShunichi Fukuzumi,*[a, b] Yong-Min Lee,*[a] and Wonwoo Nam*[a]

ChemSusChem 2017, 10, 4264 – 4276 T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4264

ReviewsDOI: 10.1002/cssc.201701381

1. Introduction

Although fresh water is essential to maintain human life, the

world is facing a global and domestic challenge to reliablysupply its population with fresh and safe water, owing to

shortages resulting from global population growth,[1–3] climatechange,[4–7] contamination of clean water supplies,[8] and public

policy.[9, 10] Water is also needed to produce hydrogen as a

clean and sustainable alternative energy source to fossil fuelsby water splitting to prevent global warming stemming from

large-scale CO2 emissions from the combustion of fossilfuels.[11–20] Therefore, water and clean energy are inextricably

linked with each other. About 97 % of the water on our planetis seawater (3.0–5.0 % salts) with 1.0 % being brackish ground

water (0.05–3.0 % salts), indicating that the vast amounts of

seawater would provide a nearly unlimited water supply ifsaline water could be used directly as a clean and nearly infin-

ite energy source. Microorganisms living in seawater can alsobe used in sediment microbial fuel cells to generate electric

power from various substrates (electron sources) containing inseawater, which reduce dioxygen to water.[21, 22] Another impor-tant energy source is hydrogen sulfide (H2S), which is abun-

dant in Black Sea deep water, where the total amount of H2S isabout 4600 Tg.[23]

This Review focuses on production of hydrogen as a cleanand sustainable energy source by electrolysis of seawater in-

stead of pure water [Eq. (1)] , by hydrolysis of Mg in seawater[Eq. (2)] ,[24] by decomposition of H2S in Black Sea deep water

[Eq. (3)] , and electric power generation by fuel cells using H2S

[Eq. (4)] and also microbial fuel cells using seawater. The solar

light-driven production of hydrogen peroxide (H2O2) from sea-water and dioxygen (O2) in the air [Eq. (5)] and its use as a

liquid fuel in seawater in H2O2 fuel cells [the reverse reaction inEq. (5)][24–26] is demonstrated as a proof of concept for the

onsite production and use of renewable and sustainableenergy.

2 H2O! 2 H2 þ O2 ð1ÞMgþ 2 H2O! H2 þMgðOHÞ2 ð2Þ2 H2S! 2 H2 þ S2 ð3Þ2 H2Sþ 3 O2 ! 2 H2Oþ 2 SO2 ð4Þ2 H2Oþ O2 ! 2 H2O2 ð5Þ

2. Electrolysis of Seawater

2.1. Production of hydrogen and dioxygen from seawater

Direct electrolysis of seawater with robust electrocatalysts re-

sults in the cathodic evolution of hydrogen at high current effi-ciency [Eq. (1)] .[27–31] However, large volumes of chlorine and

hypochlorite are normally evolved at the anode by the oxida-tion of Cl@ under acidic and basic conditions [Eqs. (6) and (7),respectively] instead of O2 evolution by the oxidation of water[Eq. (8)] , presenting a major environmental problem.[32, 33]

2Cl@ ! Cl2 þ 2 e@ E0 ¼ 1:36 V ð6ÞCl@ þ 2 OH@ ! OCl@ þ H2Oþ 2 e@ E0 ðpH 14Þ ¼ 0:89 V ð7Þ4 OH@ ! 2 H2Oþ O2 þ 4 e@ E0 ðpH 14Þ ¼ 0:40 V ð8Þ

Values of E0 in Equations (6)–(8) are given vs. SHE. The rela-

tive ratios of Cl2, HOCl, OCl@ , and O2 with hydrogen (H2) are

changed in seawater depending on pH, catalysts, and appliedpotentials.[34–39]

A Pourbaix diagram of electrochemical OER (oxygen evolu-tion reactions) from an aqueous NaCl solution at room temper-

ature (total mass of chlorine species: 0.5 m) is shown inFigure 1, where the OER is thermodynamically more favorable

Seawater is the most abundant resource on our planet andfuel production from seawater has the notable advantage that

it would not compete with growing demands for pure water.This Review focuses on the production of fuels from seawater

and their direct use in fuel cells. Electrolysis of seawater underappropriate conditions affords hydrogen and dioxygen with100 % faradaic efficiency without oxidation of chloride. Photo-

electrocatalytic production of hydrogen from seawater pro-vides a promising way to produce hydrogen with low cost and

high efficiency. Microbial solar cells (MSCs) that use biofilmsproduced in seawater can generate electricity from sunlightwithout additional fuel because the products of photosynthe-

sis can be utilized as electrode reactants, whereas the elec-trode products can be utilized as photosynthetic reactants. An-

other important source for hydrogen is hydrogen sulfide,which is abundantly found in Black Sea deep water. Hydrogen

produced by electrolysis of Black Sea deep water can also beused in hydrogen fuel cells. Production of a fuel and its direct

use in a fuel cell has been made possible for the first time by a

combination of photocatalytic production of hydrogen perox-ide from seawater and dioxygen in the air and its direct use in

one-compartment hydrogen peroxide fuel cells to obtain elec-tric power.

[a] Prof. Dr. S. Fukuzumi, Prof. Dr. Y.-M. Lee, Prof. Dr. W. NamDepartment of Chemistry and Nano Science, Ewha Womans UniversitySeoul 03760 (Republic of Korea)E-mail : fukuzumi@chem.eng.osaka-u.ac.jp

yomlee@ewha.ac.krwwnam@ewha.ac.kr

[b] Prof. Dr. S. FukuzumiGraduate School of Science and Engineering, Meijo UniversityNagoya, Aichi 468–8502 (Japan)

The ORCID identification number(s) for the author(s) of this article canbe found under https://doi.org/10.1002/cssc.201701381.

This publication is part of a Special Issue on the topic of Artificial Photo-synthesis for Sustainable Fuels. To view the complete issue, visit :http://dx.doi.org/10.1002/cssc.v10.22.

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4265

Reviews

than the chlorine evolution reaction (ClER), irrespective ofpH.[40] However, the ClER, which is a two-electron oxidation re-

action with only a single intermediate, proceeds with a muchfaster evolution rate than the OER and therefore the ClER

[Eq. (6)] is the predominant anodic reaction in acidic electro-lytes on a variety of metal oxide catalysts.[33, 34]

Under basic conditions, formation of hypochlorite also has akinetic advantage over OER. However, thermodynamics is

highly favored for OER over hypochlorite formation.[40] In addi-tion, the standard redox potential of formation of hypochlorite

is pH dependent in contrast with ClER, and the pH depend-ence slope is the same as the OER potential in the Pourbaix di-

agram (Figure 1).[40] In such a case, the electrode potential dif-ference to OER remained roughly 0.48 V, which is larger than

that under acidic conditions. If the electrocatalytic oxidation is

performed at an overpotential (h) lower than 0.48 V under al-kaline conditions, no hypochlorite is formed and OER occurs

exclusively with faster kinetics.[40] In addition, non-noble metalelectrocatalysts that may degrade under acidic conditions, can

be used under alkaline conditions. Thus, a general design crite-rion for the selective OER operation at pH>7.5 in seawater is

given by Equation (9):

hOER , 0:48 V at pH > 7:5 ð9Þ

where the pKa of hypochlorous acid is 7.5, below which hypo-

chlorous acid is dominantly formed as compared to the hypo-chlorite ion.[40]

The faradaic efficiencies for the OER using NiFe layereddouble hydroxide (LDH) catalyst and the applied current densi-ties over their corresponding time-averaged electrode poten-

tials are shown in Figure 2, where the OER faradaic efficienciesare 100:5 % in both fresh water and seawater, as predicted

by the overpotential design criterion [Eq. (9)] , with no ClER inseawater electrolysis under alkaline conditions.[40]

High-temperature solid oxide electrolyzer cells (SOECs),

which are the reverse of the solid oxide fuel cells (SOFCs), haveattracted much interest because of their advantages over low-

temperature electrolysis for high efficiency H2 production,given that an increase in operating temperature causes a de-

crease in the electrical energy demand for steam electrolysisand thereby a reduction in the H2 production cost.[41] SOECs

Shunichi Fukuzumi received his bache-

lor’s degree and Ph.D. degree at Tokyo

Institute of Technology in 1973 and

1978, respectively. After working as a

postdoctoral fellow at Indiana Universi-

ty in the USA from 1978 to 1981, he

joined the Department of Applied

Chemistry at Osaka University as an

Assistant Professor in 1981 and was

promoted to Full Professor in 1994. His

research has focused on electron

transfer chemistry, particularly artificial

photosynthesis. He is currently a Distinguished Professor at Ewha

Womans University, a Designated Professor at Meijo University, and

a Professor Emeritus at Osaka University.

Yong-Min Lee obtained his M.S. and

Ph.D. degrees in Inorganic Chemistry

at Pusan National University, Republic

of Korea, under the supervision of Pro-

fessor Sung-Nak Choi in 1999. After

working in the Magnetic Resonance

Center (CERM) at the University of Flor-

ence, Italy, as a postdoctoral fellow

and researcher under the direction of

Professors Ivano Bertini and Claudio

Luchinat from 1999 to 2005, he joined

the Center for Biomimetic Systems at

Ewha Womans University as a Research Professor in 2006. He has

been a Special Appointment Professor at Ewha Womans University

since 2009.

Wonwoo Nam earned his B.S. (Honors)

degree in Chemistry from California

State University, Los Angeles, and his

Ph.D. degree in Inorganic Chemistry

from the University of California, Los

Angeles (UCLA) under the supervision

of Professor Joan S. Valentine in 1990.

After working as a postdoctoral fellow

at UCLA for one year, he became an

Assistant Professor at Hong Ik Universi-

ty in 1991. In 1994, he moved to Ewha

Womans University, where he is cur-

rently a Distinguished Professor. His current research interests are

dioxygen activation, water oxidation, and the important roles of

metal ions in bioinorganic chemistry.

Figure 1. Pourbaix diagram for an aqueous NaCl solution with total chlorineconcentration cT,Cl = 0.5 mol L@1, with plots of electrode potentials for OER atan oxygen partial pressure of 0.21 bar (= 0.021 MPa) vs. pH. Red squares in-dicate the operating potentials (vs. SHE) after constant current electrolysis(10 mA cm@2) for 1 h with an NiFe LDH catalyst in 0.1 m KOH and 0.5 m NaCl(pH 13) or 0.3 m borate buffer and 0.5 m NaCl (pH 9.2). The light blue shadedarea shows the proposed general criterion for selective OER according toEquation (9). Reprinted with permission from Ref. [40] . Copyright 2016,Wiley-VCH Verlag GmbH.

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4266

Reviews

can electrolyze not only H2O but also a mixture of H2O andCO2 to produce synthesis gas (a mixture of H2 and CO),[41, 42]

which can be further converted into various types of syntheticfuels.[43–47]

An SOEC for simulated seawater electrolysis is shown inFigure 3, where a button cell is sealed with a glass sealant toseparate the gas environment between the anode and the

cathode.[48] The NiO/YSZ electrode (cathode; YSZ = yttria-stabi-lized zirconia) was obtained by mixing 60 wt % nickel oxide

power and 40 wt % YSZ powder together with a small amountof carbon black.[48] The 50:50 w/w La0.8Sr0.2Co0.2Fe0.8O3@d (LSCF)–

Ce0.8Gd0.2O2@d (GDC) composite anode (oxygen electrode) was

obtained by mixing LSCF-GDC50 powder together with an ap-propriate amount of poly(ethylene glycol).[48] The water bath

was heated up, providing a high moisture content (70 % partialpressure of H2O). Dry H2 was flowed into the water bath while

the anode (oxygen electrode) side was exposed to air. TheSOEC was left for around 3 h at operating temperature (800 8C)

and H2 was kept flowing to make sure that the nickel oxidewas completely reduced prior to the electrochemical measure-

ments. The two electrodes are in contact with two Pt wires,which serve as the voltage and current probes.[48]

The energy efficiency (helec) of steam electrolysis of simulatedseawater is expressed by Equations (10) and (11):

helec ¼H2 production rate> LHV of H2

electrical energy input> 100 % ð10Þ

helec ¼ ðMDH=nFVÞ> 100 % ¼ 1:291 V=V ð11Þ

where DH is the enthalpy change of H2 (124.6 MJ kg@1 at800 8C) and V is the applied voltage. The electrical energy effi-

ciency, which is the amount of electrical energy input per unitof H2 produced, at a current density of @1.0 A cm@2 from sea-

water was obtained as 109.4 %, being nearly the same as that

from pure water (110.3 %).[48] The electrical energy efficienciesof higher than 100 % result from the high heat supplied by the

furnace, leading to a reduction in electrical energy needed forthe high-temperature electrolysis.[48] The simulated seawater

and pure water electrolysis showed similar degradation ratesof about 15 % per 1000 h in terms of increases in the operatingvoltage with constant current.[48] The degradation rate in theelectrolysis of steam produced from simulated seawater was

virtually the same as that from pure water.[48]

2.2. Extraction of carbon dioxide and hydrogen from sea-water

The world’s oceans contain approximately 100 mg L@1 totalcarbon dioxide, of which 2–3 % is dissolved CO2 gas in the

form of carbonic acid (H2CO3), 1 % is carbonate (CO32@), and

the remaining 96–97 % is bicarbonate (HCO3@).[49] When the pH

of seawater is decreased to 6 or less, carbonate and bicarbon-ate in the seawater are re-equilibrated to release CO2 gas

[Eq. (12)] .[50]

HCO3@ þ Hþ Ð H2CO3 Ð H2Oþ CO2ðgÞ " ð12Þ

Thus, electrochemical approaches based on continuous elec-trodeionization (CEDI) principles are being developed to usepH to exploit seawater as a means to recover CO2 from seawa-

ter.[50–54] The electrolytic cation exchange module (E-CEM) con-figuration (Figure 4) is composed of a center compartment andelectrode compartments (cathode and anode capable of re-

versing polarities), and the three compartments are separatedby two cation-permeable membranes.[50, 54] Extruded cation-

permeable heterogeneous monolithic polyethylene mem-branes incorporate gel polystyrene cross-linked with divinyl-

benzene and functionalized with sulfonic acid groups, which

provide discrete channels for cations to migrate through thepolymer matrix, whereas blocking the passage of anions.[54]

Electrons traveled from the cathode to the anode in externalcircuit. Seawater is passed through the center compartment of

the E-CEM (Figure 4), whereas seawater through a reverse os-mosis (RO) unit is passed through the anode and cathode

Figure 2. Plots of faradaic efficiencies (*, !, &, ~) and current densities (*,!, &, ~) for the OER with NiFe LDH on carbon support vs. averaged poten-tials measured during constant-current potentiometric steps for 15 mineach. The h value is determined as 0.48 V, which is inside the design criterialimit [Eq. (9)] , as marked by a dotted vertical line. Electrolytes : a) 0.1 m KOH(*, *) and borate buffer (!, !) ; b) 0.1 m KOH + NaCl (&, &) and borate buf-fer + NaCl (~, ~). Measurement conditions: 1600 rpm, N2 bubbling at roomtemperature. The total metal loading on the working electrode was7.9 mg cm@2. Reprinted with permission from Ref. [40] . Copyright 2016,Wiley-VCH Verlag GmbH.

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4267

Reviews

compartments.[54] When direct current is applied to themodule, H+ ions and O2 gas are produced at the anode during

the oxidation of the anolyte RO water [Eq. (13)] .[54]

2 H2O! 4 Hþ þ O2 þ 4 e@ ð13Þ

The O2 gas is flushed from the anode compartment with theflow of the anolyte RO water.[54] The protons are driven from

the anode surface, through the cation-permeable membrane,and into the center compartment where protons are replaced

by Na+ in the flowing seawater, resulting in the effluent sea-water, which can be acidified without the need for any addi-

tional chemicals.[54] At a seawater pH,6, the bicarbonate andcarbonate in the seawater are re-equilibrated to carbonic acid

[Eq. (12)] .[50, 54] The CO2 from the carbonic acid in the effluent

acidified seawater is vacuum stripped by a gas-permeablemembrane contactor [Eq. (14)] .[54]

H2CO3 ! H2Oþ CO2 ð14Þ

The Na+ ions from seawater in the center compartment arepassed through the cation-permeable membrane, which is the

closest to the cathode.[54] Water is reduced at the cathode toH2 gas and OH@ [Eq. (15)] .[54]

2 H2Oþ 2 e@ ! 2 OH@ þ H2 ð15Þ

The Na+ ions react with the OH@ ions to produce NaOH inthe cathode compartment.[54] The NaOH and H2 gases are con-

tinuously flushed from the cathode compartment with theflow of the catholyte RO water.[54] The acidified seawater is re-

combined with the solutions from the cathode and anodecompartments and the overall reaction is given by Equa-

tion (16).[54]

2 H2Oþ 2 HCO3@ ! 2 OH@ þ O2 þ 2 H2 þ 2 CO2 ð16Þ

Figure 3. An SOEC setup fed with simulated seawater. Reprinted with permission from Ref. [48] . Copyright 2017, Elsevier.

Figure 4. The E-CEM configuration for extraction of CO2 and H2 from seawa-ter. Reprinted with permission from Ref. [50] . Copyright 2011, AmericanChemical Society.

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4268

Reviews

Seawater containing HCO3@ (142 ppm, 0.0023 m) and a flow

rate of 1900 mL min@1 require a theoretical applied minimum

current of 7.0 A to lower the pH to less than 6.0 and convertHCO3

@ to H2CO3.[54] The theoretical amount of H2 gas produce

at 7.0 A is 0.0022 mol min@1.[54] The ultimate goal is to produceCO2 and H2 in quantities needed to make fuel at energy effi-ciencies similar to commercial electrolysis.[54]

2.3. Solar hydrogen production from seawater

Hydrogen is produced photoelectrocatalytically from seawaterunder solar irradiation at ambient temperature without any ex-

ternally applied bias potential by using a materially optimizedthin-film transparent titania photocatalyst.[55] The overall solarenergy conversion efficiency, which is defined as the heat

energy of reaction attained by combustion of H2 with O2 perthe total solar energy supplied, was determined to be

0.388 %.[55] The maximum quantum efficiency of 54 % was ob-tained at the irradiation wavelength of 320 nm.[55] However,

the quantum efficiency became zero at irradiation wavelengths

longer than 410 nm.[55] A number of photocatalysts, which canabsorb visible light (>400 nm), have been utilized for photoca-

talytic H2 production from seawater, which, however, requiredthe assistance of an externally applied bias potential.[56–60]

Electrolysis in seawater is often inhibited by the presence ofsalts and impurities contained in seawater, which fouled the

catalytic systems to decrease the required voltage and electrol-

ysis efficiency.[61] In contrast to liquid seawater, the atmosphericwater vapor over the sea is virtually free of contaminants, with

high humidity near the ocean surface all year round.[61] Thus,stable solar H2 production has been made possible by using

the ambient humidity over the seawater at efficiencies compa-rable to land-based H2 production by liquid pure water split-

ting.[61] The near-surface relative humidity over the oceans is

fairly constant between 75 and 90 % with only minor variationby season or time of day.[62, 63] Marine solar water vapor elec-

trolysis can indirectly split seawater without any water purifica-tion process.[63] The electrolyzer consists of two stainless steel

grids for the anode and the cathode, two carbon Toray papersheets as gas-diffusion layers, and one membrane electrode as-

sembly (MEA) with catalyst particles (Pt/C) on either side of asulfonated polytetrafluoroethylene (Nafion) proton-exchange

membrane (Figure 5).[61] The photovoltage required for solar-driven water electrolysis was obtained by employing a com-mercially available triple-junction amorphous Si (a-Si) photovol-

taic from SolarFocus.[61] The solar-to-hydrogen conversion effi-ciency before and after 50 h of operation was increased from

6.0 % to 6.3 % by using an ambient seawater humidity feed-stock as compared to a drop from 6.6 % to 0.5 % using liquid

seawater feedstock.[61] The solar-driven seawater vapor electrol-

ysis system could create a new market in offshore solar utilitieswith built-in energy storage to solve the intermittency issue,

minimizing land usage and environmental impacts of solararrays. A recent analysis of a potential solar-driven seawater

vapor electrolysis facility estimated that a 1 GW annual averageoutput would require around 180 km2 of land usage.[64]

3. Hydrolysis of Metals in Seawater

Hydrogen production from the hydrolysis of active metals,

such as Mg and Al, has attracted much attention because oftheir abundance and nontoxicity.[65–70] Although Mg has a low

H2 storage density (3.3 wt %) in water compared to that of Al

(3.7 wt %), it is capable of generating hydrogen in a neutralaqueous solution instead of the alkaline solution that is re-

quired for the hydrolysis of Al.[71–74] The Mg alloy can be fullyrecycled by electrical system with the efficiency of 41 %.[75, 76]

Hydrolysis of Mg in an aqueous solution produces Mg(OH)2

and H2 as shown in Equation (2), where a passive Mg(OH)2

layer is formed on the Mg surface to hamper the further hy-

drolysis of Mg.[77–82] However, the Mg(OH)2 layer is readilybroken by the penetration of Cl@ ions, which results in pitting

corrosion to maintain the continuous hydrolysis of Mg in sea-water.[72, 83, 84]

The effects of NH4Cl on the rate of magnesium hydride(MgH2) hydrolysis are shown in Figure 6, where the rate of

MgH2 hydrolysis in deionized water is too slow to be used in

controlled H2 generation.[85] In contrast, the hydrolysis of MgH2

in a 0.5 wt % NH4Cl solution produces H2 efficiently (501 mL g@1

in 5 min and 980 mL g@1 H2 in 30 min) at 60 8C (Figure 6 b).[85]

The 4.5 wt % NH4Cl solution afforded the best hydrolysis per-formance (Figure 6 c), where H2 is produced more efficiently(1310 mL g@1 in 5 min and 1660 mL g@1 H2 in 30 min).[85] The

rate of hydrolysis of MgH2 is enhanced by addition of NH4Clbecause Cl@ may effectively decrease the compactness ofMg(OH)2.[85]

Hydrogen was also produced efficiently by the hydrolysis ofmilled waste magnesium scraps (MWMS) in nickel(II) chloride

(NiCl2) solutions (0.5–2.5 m) and by the hydrolysis of NiCl2-added seawater (Aegean Sea and Marmara Sea water).[83]

MWMS was prepared by mechanical grinding with use of a

planetary type ball-milling apparatus at a speed of 300 rpm for15 h grinding time at ball-to-powder ratio of 70:1 (Figure 7).[83]

The MWMS sample in Marmara Sea Water, which contained ap-proximately 23 wt % NaCl, produced 6.6 mL H2 in 260 s

(ca. 70 % conversion), whereas 100 % conversion was attainedin 233 s in Aegean Sea water containing roughly 37 wt %

Figure 5. Electrolysis cell containing matched projected areas of triple-junc-tion amorphous Si (tj-a-Si) and MEA. Reprinted with permission fromRef. [61] . Copyright 2016, Royal Society of Chemistry.

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4269

Reviews

NaCl.[83] An increase in the concentration of NaCl in seawater

affected its conductivity, resulting in an increase in the rate ofH2 production.[83]

Economical and safe Mg alloys have been developed forfaster H2 production from seawater.[86] The Mg-2.7Ni-1Sn alloy

afforded an excellent H2 production rate of 28.71 mL min@1 g@1,which is 1700 times more rapid than that of pure magnesium

(Mg), owing to galvanic and intergranular corrosion, as well aspitting corrosion in seawater (Figure 8).[86] As the solution tem-perature was increased from 30 to 70 8C, the H2 productionrate in hydrolysis by the Mg-2.7Ni-1Sn alloy was dramaticallyincreased from 34 to 257.3 mL min@1 g@1.[86]

4. Microbial Fuel Cells Using Seawater

Sediment microbial fuel cells (SMFCs) generate electric power

from the organic matter content of sediments using bacterialmetabolism.[87–92] The anode electrode was embedded within

anoxic marine sediments and connected through electronic cir-cuits to a similar electrode in the overlying aerobic seawater(the cathode) to make an SMFC.[93] The SMFC with an unmodi-

fied graphite electrode afforded an electric power generationof 0.01 W m@2, which can power marine-deployed electronic in-

strumentation.[93] SMFCs with graphite anodes in the anoxicmarine sediment and graphite cathodes were also made in

overlying aerobic seawater.[94] A specific enrichment of micro-organisms of the family Geobacteraceae on graphite anodes al-

lowed these microorganisms to conserve energy, supportingtheir growth by oxidation of organic compounds with an elec-trode that serves as the sole electron acceptor.[94]

SMFCs with carbon fabric cathodes were acclimated withsediment and seawater from San Diego Bay, whereas anodes

were buried in a flat orientation, which is placed about 7.6 cmbelow the sediment-water interface and about 7.6 cm above

the bottom of the tank as shown in Figure 9, where seawater

is introduced up to a height of 50.8 cm.[21] Figure 10 A showsthe average power generation of two identical sediment mi-

crobial fuel cells (SMFC-1 and SMFC-2) operating in the sameseawater tank, where the power increased to about 1.8 mW

within roughly 40 days.[21] The long time power generation re-sulted in the power decrease to around 1 mW after

Figure 6. Time courses of H2 evolution in the hydrolysis of MgH2 in (a) deion-ized water and in (b) 0.5 wt % and (c) 4.5 wt % NH4Cl solutions at differenttemperatures. Reprinted with permission from Ref. [85] . Copyright 2015,Elsevier.

Figure 7. Waste magnesium scraps (WMS): a) From a gold manufacturingfactory; b) after 15 h mechanical grinding. Reprinted with permission fromRef. [83] . Copyright 2015, Elsevier.

Figure 8. Time profiles of H2 generation from pure Mg and Mg-2.7Ni-xSnalloys (x = 0, 0.5, 1, 2) in an aqueous NaCl solution (3.5 wt %) at 25 8C. Re-printed with permission from Ref. [86] . Copyright 2017, Elsevier.

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4270

Reviews

160 days.[21] The corresponding electrode potentials during theoperation of SMFC-1 are shown in Figure 10 B, where both the

cathode and the anode potentials are initially around 75 mVvs. Ag/AgCl (black arrow), the cathode potential increased to

around 375 mV vs. Ag/AgCl within a few days, and then slowlyincreased to 395 mV vs. Ag/AgCl while the anode potential de-

creased initially to about @400 mV vs. Ag/AgCl and then in-creased slowly to a value slightly below 0 mV vs. Ag/AgCl.[21]

The acclimated cathodes of SMGCs have distinct performanceadvantages as compared to unacclimated cathodes with possi-

ble implications for cathode scaling and size.[21]

5. Hydrogen from Black Sea Deep Water

Hydrogen sulfide (H2S) is found naturally in many gas wellsand also in gas hydrates and gas-saturated sediments especial-ly at the bottom of the Black Sea, where 90 % of the sea wateris anaerobic.[95] The total H2 energy potential is estimated to beabout 270 million tons, produced from 4.587 billion tons of

H2S in Black Sea deep water.[23, 96] This amount of H2 can pro-vide 38.3 million TJ of thermal energy or 8.97 million GWh ofelectrical energy.[23] Moreover, the total H2 potential in Black

Sea deep water corresponds to 808 million tons of gasoline,766 million tons of natural gas, 841 million tons of fuel oil or

851 million tons of petroleum.[23] These values suggest that H2

production from H2S in Black Sea deep water may play an im-

portant role in satisfying the energy demands of the Black

Sea’s surrounding countries.[23]

The simplest and direct method that can be applied to uti-

lize Black Sea deep water is catalytic or non-catalytic thermaldecomposition of H2S [Eq. (3)] , which requires high tempera-

ture to break down H2S into H2 and S2.[97] This reaction is large-ly endothermic, becoming more favorable at higher tempera-

tures. The pyrolysis of H2S occurs in a perfectly stirred reactor

for residence times of 0.4–1.6 s at 1073–1373 K under 1 barpressure.[97] Figure 11 shows the H2S conversion and yields of

H2 and S2 with a feed stream composed of 5 mol % H2S and95 mol % Ar under 1 bar pressure at different temperatures,

which are fitted by the computed time profiles based on theproposed mechanisms.[97, 98] Because higher temperatures

(above 1200 8C) are required to obtain an H2 yield above

30 %,[99] La2SrOx was used as an effective catalyst to decreasethe temperature to 950 8C to obtain the highest conversion of

35.7 % for the decomposition of H2S.[100] The results from thelaboratory-scale extraction pilot plant unit for the separation ofH2S from Black Sea deep water enabled the building of an in-dustrial extraction pilot plant to concentrate H2S from 10 ppmto above 10 000 ppm.[101] The processing of 109 m3 of watercontaining 10 ppm H2S would produce only 0.833 tons of H2.

Thus, a new technology for extraction and concentration ofH2S is essential for the practical application.[101]

The feasibility of H2 production from H2S contained in Black

Sea deep water has been examined by employing a micro-structured electrochemical membrane reactor, which is assem-

bled with optimized cell materials, to produce H2 in the elec-trolyzer mode and to generate electrical energy in the fuel cell

mode.[102] Any process developed for H2S decomposition to

produce H2 in Black Sea deep water, which is energy intensive,will have substantial equipment and operating costs, heavily

depending on technology and the commercial price ofenergy.[102] The cost of H2 from H2S in Black Sea deep water

may gradually shift closer to that of H2 from solar-driven waterelectrolysis in the long term.[103]

Figure 9. Experimental setup for orientation of the SMFC with seawater in-troduced inside the tank. Reprinted with permission from Ref. [21] . Copy-right 2014, Wiley-VCH Verlag GmbH.

Figure 10. a) Average power generated by SMFC-1 and SMFC-2 over time.b) Anode and cathode potentials of SMFC-1 over time. Shaded areas repre-sent time periods during electrochemical characterization. Reprinted withpermission from Ref. [21] .

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4271

Reviews

6. Production of Hydrogen Peroxide from Sea-water and Its Direct Use in Fuel Cells

Hydrogen peroxide (H2O2) is a clean and sustainable energy

source, because the decomposition products are only waterand oxygen, and H2O2 is produced from water and dioxygenusing solar energy.[104–112] H2O2 is superior to H2 or natural gas,

owing to its safe storage and convenient transportation inliquid form.[104–112] Furthermore, a one-compartment configura-

tion is possible for H2O2 fuel cells, which is structurally muchsimpler and cheaper than two-compartment hydrogen fuel

cells with expensive membranes.[24–26, 113]

Efficient photocatalytic production of H2O2 from seawater

and O2 in the air was performed in a two-compartment photo-electrochemical cell, composed of mesoporous WO3 supportedon a fluorine doped tin oxide (FTO) glass substrate (m-WO3/

FTO) as a photocatalyst for H2O oxidation and a cobalt chlorincomplex supported on a carbon paper ([CoII(Ch)]/CP) as an

electrocatalyst for the selective two-electron reduction ofO2,[114, 115] as shown in Figure 12.[116] The photoirradiation of m-

WO3/FTO with a solar simulator (1 sun = AM 1.5G) in the anode

cell resulted in formation of H2O2 in the cathode cell of thetwo-compartment photoelectrochemical configuration with no

external bias potential being applied.[116] The time courses ofthe photocatalytic formation of H2O2 are shown in

Figure 13.[116] It was confirmed that no H2O2 was producedwithout [CoII(Ch)] on the carbon paper electrode. [CoII(Ch)] was

reported to catalyze the selective two-electron/two-proton re-duction of O2 to yield H2O2 efficiently.[114, 115] After illumination

for 24 h, the concentration of H2O2 produced in seawater was48 mm (blue circles in Figure 13), which is high enough to op-

erate an H2O2 fuel cell (see below).[116] However, the productionof H2O2 is much decelerated when pure water was used in-

stead of seawater for the photocatalytic H2O2 production(Figure 13).[116]

The similar enhancement of the photocatalytic H2O2 produc-

tion is observed when seawater is replaced by an NaCl solutionwith the same concentration as that of seawater (blue squares

in Figure 13). The effects of Cl@ on photocatalytic H2O oxida-tion with m-WO3/FTO and the two-electron reduction of O2

with [CoII(Ch)]/CP were studied by using a photoelectrochemi-

cal cell with a three-electrode configuration (see below).[116]

The current-potential curves of m-WO3/FTO under simulated

solar illumination (1 sun = AM 1.5G) are compared to those indark in Figure 14 a, where the onset of photocurrent for the

H2O oxidation is observed at 0.2 V vs. SCE at pH 1.3 in purewater, which corresponds to the 110 mV overpotential with re-

Figure 11. Time profiles for (a) conversion of H2S, (b) H2 mole fraction, and(c) S2 mole fraction for feed containing 5 mol % H2S in Ar under 1 bar pres-sure at different temperatures. The dotted and solid lines were computedbased on the proposed mechanisms reported in Refs. [97] and [98], respec-tively. Reprinted with permission from Ref. [97] . Copyright 2016, Elsevier.

Figure 12. A two-compartment cell for the photocatalytic production ofH2O2 from water and air with m-WO3/FTO photoanode and [CoII(Ch)]/CPcathode performed in pure water and seawater under simulated solar illumi-nation (1 sun = AM 1.5G). Reprinted with permission from Ref. [116]. Copy-right 2016, Nature Publishing Group.

Figure 13. Time courses of generation of H2O2 with m-WO3/FTO photoanodeand [CoII(Ch)]/CP cathode in water (pH 1.3; red circles) in seawater (pH 1.3;blue circles) and in an aqueous NaCl solution (pH 1.3; blue squares) under si-mulated 1 sun (AM 1.5G) illumination. No H2O2 was produced without[CoII(Ch)] at carbon paper cathode under simulated 1 sun (AM 1.5G) illumina-tion in water (pH 1.3; black circles). Reprinted with permission fromRef. [116]. Copyright 2016, Nature Publishing Group.

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4272

Reviews

spect to the value of flat band potential of WO3 (0.09 V vs. SCEat pH 1.3).[116] When pure water was replaced by seawater

under otherwise the same reaction conditions, roughly fourtimes larger a photocurrent was detected at 0.3 V vs. SCE andthe onset potential was negatively shifted from 0.2 V vs. SCEto 0.1 V vs. SCE as compared to those in pure water.[116] Thecatalytic current for the two-electron reduction of O2 with[CoII(Ch)]/CP at pH 1.3 in seawater appears at around 0.25 V vs.

SCE, which is nearly the same as the value measured at pH 1.3in pure water (Figure 14 b).[116] Thus, the acceleration of therate of H2O2 production in seawater results from the enhance-

ment of the photocatalytic oxidation of seawater at the photo-anode.

The enhanced water oxidation in seawater may result fromthe preferred oxidation of Cl@ over that of water (see

below).[117–119] Upon photoexcitation of WO3, photogenerated

holes oxidize Cl@ more easily than H2O to produce chlorine(Cl2) [Eq. (17)] , because the two-electron oxidation of Cl@ is ki-

netically more favorable than the four-electron oxidation ofH2O although the oxidation potential of Cl@ [E(Cl@/Cl2) = 1.28 V

vs. SCE at pH 1.3] is thermodynamically less favorable thanthat of water (0.91 V vs. SCE at pH 1.3).[117] Thus, the Cl2 pro-

duced reacts with H2O to form HClO, depending on the pH ofthe reaction solution [Eq. (18)] .[118] Decomposition of HClO

yields O2 and Cl@ under sunlight irradiation [Eq. (19)] .[119] Thus,Cl@ can accelerate the water oxidation as a catalyst [Eq. (20)] .

The formation of HClO/Cl2 was confirmed during the photoca-talytic H2O oxidation in the presence of Cl@ .[116] The amount of

O2 evolved in the anode cell containing seawater (12.7 mmol)after photoirradiation for 1 h was more than three times largerthan that evolved in the anode cell containing pure water

(3.7 mmol).[116] Thus, the enhancement of photocatalytic pro-duction of H2O2 in seawater (Figure 13) results from the Cl@-catalyzed photooxidation of water.[116]

2 Cl@ þ 2 hþ hvK! ! Cl2 ð17Þ

Cl2 þ H2OÐ HClOþ Hþ þ Cl@ ð18Þ

2 HClO hvK! O2 þ 2 Hþ þ 2 Cl@ ð19Þ

2 H2Oþ 4 hþ hvK! O2 þ 4 Hþ ð20Þ

A solar energy conversion efficiency of 0.55 % has been re-

ported for the photocatalytic production of H2O2 in seawaterunder simulated solar illumination with a solar simulator

(1 sun = AM 1.5G).[116] A higher solar energy conversion efficien-cy of 0.94 % was attained under the simulated solar illumina-

tion,[116] being significantly higher than that of switchgrass(0.2 %) producing biofuel.[120]

When the WO3 photoanode was replaced by surface modi-

fied BiVO4 with iron(III) oxyhydroxide [FeO(OH)] , the highestsolar energy conversion efficiency of H2O2 production in pure

water was achieved as 6.6 % with a decreased simulated solarlight intensity (0.05 sun) after 1 h photocatalytic reaction,

whereas the efficiency was decreased to 0.89 % under 1 sun il-lumination.[121] In this case, however, seawater deactivated the

photocatalyst, owing to the instability of BiVO4, which started

to dissolve in the presence of Cl@ under solar illumination.[121]

The chemical energy of H2O2 produced in the photocatalytic

oxidation of seawater by O2 in air can be converted into elec-trical energy by using a one-compartment H2O2 fuel cell, com-posed of a carbon cloth cathode modified with FeII

3[CoIII(CN)6]2

and a nickel mesh anode.[116] A seawater solution containing

H2O2 (48 mm) produced by photocatalytic reduction of O2 wastransferred to a hydrogen peroxide (H2O2) fuel cell. TheH2O2 fuel cell exhibited an open circuit voltage (OCV) of 0.78 V

and a maximum power density (MPD) of 1.6 mW cm@2

(Figure 15).[116] The energy conversion efficiency of the H2O2

fuel cell was evaluated as roughly 50 % by measurement of theoutput energy as electrical energy vs. consumed chemical

energy of H2O2, which is comparable to the efficiency of an H2

fuel cell.[116]

A direct H2S/H2O2 fuel cell was constructed with artificial

Black Sea deep water containing H2S as a fuel and H2O2 as anoxidant instead of O2.[122–124] The anode, cathode and overall re-

actions are given by Equations (21), (22), and (23), respective-ly.[24, 122]

Figure 14. Photoelectrochemical performance of m-WO3/FTO electrode forthe four-electron/four-proton oxidation of water and that of [CoII(Ch)]/CPelectrode for the two-electron/two-proton reduction of O2. a) Current–volt-age (I–V) curves for the WO3/FTO photoanode in water (pH 1.3; black solidline) and seawater (pH 1.3; red solid line) under simulated solar illumination(1 sun = AM 1.5G). Dashed lines with the same color definition are I–V curvesin the dark at a sweep rate of 10 mV s@1. b) Cyclic voltammograms (CVs) of[CoII(Ch)]/CP in N2-saturated water (pH 1.3; black solid line) and O2-saturatedwater (pH 1.3; red solid line). The dashed lines are the CVs of [CoII(Ch)]/CP inseawater (pH 1.3) at a sweep rate of 20 mV s@1. Reprinted with permissionfrom Ref. [116]. Copyright 2016, Nature Publishing Group.

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4273

Reviews

Anode : H2S! Sþ 2 Hþ þ 2 e@ E0 ¼ @0:14 V vs: SHE

ð21Þ

Cathode : H2O2 þ 2 Hþ þ 2 e@ ! 2 H2O E0 ¼ 1:77 V vs: SHE

ð22Þ

Overall : H2Sþ H2O2 ! Sþ 2 H2O Ecell ¼ 1:91 V ð23Þ

The maximum cell voltage of an H2S/H2O2 fuel cell is ob-

tained as 1.91 V,[122] which is much higher than that of an H2O2

fuel cell (1.09 V).[24] The effects of anolyte and the catholyte

concentrations and the pH of the anolyte were examined foran H2S/H2O2 fuel cell.[124] An increase in the solution pH result-

ed in an increase in HS@ concentration of the anolyte, leading

to generation of the higher cell power.[124] The H2O2 concentra-tion has a greater impact on the maximum power density as

shown in Figure 16.[124] The H2S/H2O2 fuel cell yielded a power

density of 23 mW cm@2 at a cell voltage of 300 mV and currentdensity of 75 mA cm@2 (Figure 16), which are much higher than

those in a direct H2O2 fuel cell (Figure 15).[124]

7. Conclusions

Direct electrolysis of seawater to evolve H2 and O2 selectivelywithout formation of chlorine or hypochlorite has been made

possible by a general design criterion at pH>7.5 in seawaterwith an overpotential (h)<0.48 V. The electrical energy effi-

ciency, that is the amount of electrical energy input per unit of

hydrogen produced, was obtained as 109.4 % for a high-tem-perature solid oxide electrolyzer cell with the NiO/YSZ elec-

trode, which exhibited a slow degradation rate of 15 % in1000 h under steam produced from simulated seawater. Sea-

water is also utilized for electrodeionization principles to recov-er CO2 from seawater, accompanied by production of H2. H2

can also be produced from seawater under solar illumination

at ambient temperature without the assistance of externallyapplied bias potential by using a materially optimized thin film

transparent titania photocatalyst. Economical and safe Mgalloys have been developed for faster hydrogen production

from seawater. Anoxic marine sediments were used for sedi-ment microbial fuel cells (SMFCs) for electric power generation

sufficient to power marine-deployed electronic instrumenta-

tion.Hydrogen sulfide contained in Black Sea deep water has

been used for H2 production by pyrolysis of H2S. A direct H2S/H2O2 fuel cell was developed and exhibited high cell per-

formance compared with H2O2 fuel cells. H2O2 can be pro-duced efficiently from seawater and air by using a two-com-

partment photoelectrochemical cell, composed of mesoporous

WO3 supported on a FTO glass substrate (m-WO3/FTO) for thephotocatalytic seawater oxidation and a cobalt chlorin com-

plex adsorbed on a carbon paper ([CoII(Ch)]/CP) for the selec-tive two-electron reduction of O2 (Figure 12). The chloride

anion contained in seawater acts as a catalyst to accelerateseawater oxidation. H2O2 produced from seawater and air by

using solar energy has been used directly as a fuel in an H2O2

fuel cell. Thus, H2O2 can be an alternative sustainable andgreen fuel for portable electric items, because oxygen, seawa-

ter, and sunlight are abundant resources, and power genera-tion is possible with a fuel cell with a simple one-compartment

structure. Further improvement of the photocatalytic activityfor production of H2O2 from Black Sea deep water, together

with more efficient one-compartment H2S/H2O2 fuel cells with-

out membranes may enable portable sustainable energy con-version systems in the future.

Acknowledgements

The authors gratefully acknowledge the contributions of their col-

laborators and co-workers mentioned in the cited references, andsupport from a JST SENTAN project (to S.F.), JSPS KAKENHI (No.

16H02268 to S.F.), the NRF of Korea through CRI (NRF-2012R1A3A2048842 to W.N.), GRL (NRF-2010-00353 to W.N.), and

Figure 15. I–V (blue squares) and current–power density (I–P ; red circles)curves of the one-compartment H2O2 fuel cell using a FeII

3[CoIII(CN)6]2/carboncloth cathode and a Ni mesh anode in seawater containing H2O2 (48 mm),which was produced by photocatalytic reduction of O2 with seawater, asshown in Figure 13 (blue circles). Reprinted with permission from Ref. [116].Copyright 2016, Nature Publishing Group.

Figure 16. I–V and I–P curves for the one-compartment H2O2 fuel cell with aPt plate cathode and a Ni mesh anode. Effects of H2O2 concentration (green:6 m ; brown: 3 m ; blue: 1 m) in artifical Black Sea water on the cell per-formance (anolyte: 12 mmol Na2S, T = 25 8C). Reprinted with permission fromRef. [124] . Copyright 2015, Elsevier.

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4274

Reviews

the Basic Science Research Program (2017R1D1A1B03029982 toY.M.L. and 2017R1D1A1B03032615 to S.F.).

Conflict of interest

The authors declare no conflict of interest.

Keywords: electrolysis · fuel cells · photocatalysis · solar cells ·water splitting

[1] C. J. Vçrçsmarty, P. Green, J. Salisbury, R. B. Lammers, Science 2000,289, 284 – 288.

[2] M. Kummu, J. H. A. Guillaume, H. de Moel, S. Eisner, M. Flçrke, M.Porkka, S. Siebert, T. I. E. Veldkamp, P. J. Ward, Sci. Rep. 2016, 6, 38595.

[3] C. M. Chini, K. L. Schreiber, Z. A. Barker, A. S. Stillwell, Environ. Sci. Tech-nol. 2016, 50, 9003 – 9012.

[4] V. Ramanathan, P. J. Crutzen, J. T. Kiehl, D. Rosenfeld, Science 2001,294, 2119 – 2124.

[5] M. Alamaro, Nature 2014, 514, 7.[6] O. Schiermeier, Nature 2014, 514, 10 – 11.[7] A. Hartmann, T. Gleeson, Y. Wada, T. Wagener, Proc. Natl. Acad. Sci. USA

2017, 114, 2842 – 2847.[8] R. P. Schwarzenbach, B. I. Escher, K. Fenner, T. B. Hofstetter, C. A. John-

son, U. von Gunten, B. Wehrli, Science 2006, 313, 1072 – 1077.[9] S. L. Postel, G. C. Daily, P. R. Ehrlich, Science 1996, 271, 785 – 788.

[10] K. Pakalniete, J. Aigars, M. Czajkowski, S. Strake, E. Zawojska, N. Hanley,Sci. Total Environ. 2017, 584 – 585, 29 – 40.

[11] P. V. Kamat, Acc. Chem. Res. 2017, 50, 527 – 531.[12] Y.-J. Yuan, Z.-T. Yu, D.-Q. Chen, Z.-G. Zou, Chem. Soc. Rev. 2017, 46,

603 – 631.[13] C. Ding, J. Shi, Z. Wang, C. Li, ACS Catal. 2017, 7, 675 – 688.[14] K. C. Christoforidis, P. Fornasiero, ChemCatChem 2017, 9, 1523 – 1544.[15] D. Bae, B. Seger, P. C. K. Vesborg, O. Hansen, I. Chorkendorff, Chem.

Soc. Rev. 2017, 46, 1933 – 1954.[16] N. S. Lewis, Nat. Nanotechnol. 2016, 11, 1010 – 1019.[17] D. M. Fabian, S. Hu, N. Singh, F. A. Houle, T. Hisatomi, K. Domen, F. E.

Osterlohf, S. Ardo, Energy Environ. Sci. 2015, 8, 2825 – 2850.[18] Z. Yu, F. Li, L. Sun, Energy Environ. Sci. 2015, 8, 760 – 775.[19] M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A.

Santori, N. S. Lewis, Chem. Rev. 2010, 110, 6446 – 6473.[20] A. Kudo, Y. Miseki, Chem. Soc. Rev. 2009, 38, 253 – 278.[21] J. T. Babauta, L. Hsu, E. Atci, J. Kagan, B. Chadwick, H. Beyenal, ChemSu-

sChem 2014, 7, 2898 – 2906.[22] B. Erable, R. Lacroix, L. Etcheverry, D. F8ron, M. L. Delia, A. Bergel, Biore-

sour. Technol. 2013, 142, 510 – 516.[23] A. Midilli, M. Ay, A. Kale, T. N. Veziroglu, Int. J. Hydrogen Energy 2007,

32, 117 – 224.[24] S. Fukuzumi, Y. Yamada, ChemElectroChem 2016, 3, 1978 – 1989.[25] S. Fukuzumi, Y. Yamada, Aust. J. Chem. 2014, 67, 354 – 364.[26] S. Fukuzumi, Y. Yamada, K. D. Karlin, Electrochim. Acta 2012, 82, 493 –

511.[27] J. E. Bennett, Int. J. Hydrogen Energy 1980, 5, 401 – 408.[28] Y. Zhao, Q. Tang, B. He, P. Yang, RSC Adv. 2016, 6, 93267 – 93274.[29] X. Niu, Q. Tang, B. He, P. Yang, Electrochim. Acta 2016, 208, 180 – 187.[30] H. Li, Q. Tang, B. He, P. Yang, J. Mater. Chem. A 2016, 4, 6513 – 6520.[31] J. Zheng, Appl. Surf. Sci. 2017, 413, 360 – 365.[32] A. F. Adamson, B. G. Lever, W. F. Stones, J. Appl. Chem. 1963, 13, 483 –

495.[33] S. Trasatti, Electrochim. Acta 1984, 29, 1503 – 1512.[34] C. Belmont, R. Ferrigno, O. Leclerc, H. H. Girault, Electrochim. Acta

1998, 44, 597 – 603.[35] K. Fujimura, K. Izumiya, A. Kawashima, E. Akiyama, H. Habazaki, N. Ku-

magai, K. Hashimoto, J. Appl. Electrochem. 1999, 29, 769 – 775.[36] E. Baniasadi, I. Dincer, G. F. Naterer, Int. J. Hydrogen Energy 2013, 38,

2589 – 2595.

[37] R. Venkatkarthick, S. Elamathi, D. Sangeetha, R. Balaji, B. S. Kannan, S.Vasudevan, D. J. Davidson, G. Sozhan, S. Ravichandran, J. Electroanal.Chem. 2013, 697, 1 – 4.

[38] Y. Xue, J. Zhao, R. Qiu, J. Zheng, C. Lin, B. Ma, P. Wang, Appl. Surf. Sci.2015, 357, 60 – 68.

[39] Z. Kato, R. Kashima, K. Tatsumi, S. Fukuyama, K. Izumiya, N. Kumagai,K. Hashimoto, Appl. Surf. Sci. 2016, 388, 640 – 644.

[40] F. Dionigi, T. Reier, Z. Pawolek, M. Gliech, P. Strasser, ChemSusChem2016, 9, 962 – 972.

[41] S. D. Ebbesen, S. H. Jensen, A. Hauch, M. B. Mogensen, Chem. Rev.2014, 114, 10697 – 10734.

[42] A. Odukoya, G. F. Naterer, M. Roeb, C. Mansilla, J. Mougin, B. Yu, J. Ku-pecki, I. Iordache, J. Milewski, Int. J. Hydrogen Energy 2016, 41, 7878 –7891.

[43] V. Menon, Q. Fu, V. M. Janardhanan, O. Deutschmann, J. Power Sources2015, 274, 768 – 781.

[44] W. L. Becker, R. J. Braun, M. Penev, M. Melaina, Energy 2012, 47, 99 –115.

[45] Y. Zheng, J. Wang, B. Yu, W. Zhang, J. Chen, J. Qiao, J. Zhang, Chem.Soc. Rev. 2017, 46, 1427 – 1463.

[46] Y. Wang, T. Liu, L. Lei, F. Chen, Fuel Process. Technol. 2017, 161, 248 –258.

[47] L. Lei, T. Liu, S. Fang, J. P. Lemmon, F. Chen, J. Mater. Chem. A 2017, 5,2904 – 2910.

[48] C. K. Lim, Q. Liu, J. Zhou, Q. Sun, S. H. Chan, J. Power Sources 2017,342, 79 – 87.

[49] T. Takahashi, W. S. Broecker, A. E. Bainbridge in Carbon Cycle Modelling,Vol. 16 (Ed. : B. Bolin), SCOPE, New York, 1981, pp. 271 – 286.

[50] H. D. Willauer, F. DiMascio, D. R. Hardy, M. K. Lewis, F. W. Williams, Ind.Eng. Chem. Res. 2011, 50, 9876 – 9882.

[51] H. D. Willauer, F. DiMascio, D. R. Hardy, M. K. Lewis, F. W. Williams, Ind.Eng. Chem. Res. 2012, 51, 11254 – 11260.

[52] H. D. Willauer, F. DiMascio, D. R. Hardy, F. W. Williams, Ind. Eng. Chem.Res. 2014, 53, 12192 – 12200.

[53] M. D. Eisaman, K. Parajuly, A. Tuganov, C. Eldershaw, N. Chang, K. A.Littau, Energy Environ. Sci. 2012, 5, 7346 – 7352.

[54] H. D. Willauer, F. DiMascio, D. R. Hardy, F. W. Williams, Energy Fuels2017, 31, 1723 – 1730.

[55] S. Ichikawa, Int. J. Hydrogen Energy 1997, 22, 675 – 678.[56] W. Luo, Z. Yang, Z. Li, J. Zhang, J. Liu, Z. Zhao, Z. Wang, S. Yan, T. Yu, Z.

Zou, Energy Environ. Sci. 2011, 4, 4046 – 4051.[57] Z. Li, W. Luo, M. Zhang, J. Feng, Z. Zou, Energy Environ. Sci. 2013, 6,

347 – 370.[58] Y. Li, R. Wang, H. Li, X. Wei, J. Feng, K. Liu, Y. Dang, A. Zhou, J. Phys.

Chem. C 2015, 119, 20283 – 20292.[59] K.-W. Cheng, W.-T. Tsai, Y.-H. Wu, J. Power Sources 2016, 317, 81 – 92.[60] Y. Li, X. Wei, X. Yan, J. Cai, A. Zhou, M. Yanga, K. Liu, Phys. Chem. Chem.

Phys. 2016, 18, 10255 – 10261.[61] S. Kumari, R. T. White, B. Kumar, J. M. Spurgeon, Energy Environ. Sci.

2016, 9, 1725 – 1733.[62] M. Benetti, G. Reverdin, C. Pierre, L. Merlivat, C. Risi, H. C. Steen-Larsen,

F. Vimeux, J. Geophys. Res. Atmos. 2014, 119, 584 – 593.[63] J. Jouzel, L. Merlivat, C. Lorius, Nature 1982, 299, 688 – 691.[64] R. Sathre, C. D. Scown, W. R. Morrow III, J. C. Stevens, I. D. Sharp, J. W.

Ager, III, K. Walczak, F. A. Houle, J. B. Greenblatt, Energy Environ. Sci.2014, 7, 3264 – 3278.

[65] L. Z. Ouyang, J. M. Huang, H. Wang, Y. J. Wen, Q. A. Zhang, D. L. Sun,M. Zhu, Int. J. Hydrogen Energy 2013, 38, 2973 – 2978.

[66] L. Z. Ouyang, X. S. Yang, M. Zhu, J. W. Liu, H. W. Dong, D. L. Sun, J. Zou,X. D. Yao, J. Phys. Chem. C 2014, 118, 7808 – 7820.

[67] J. M. Huang, L. Z. Ouyang, Y. J. Wen, H. Wang, J. W. Liu, Z. L. Chen, M.Zhu, Int. J. Hydrogen Energy 2014, 39, 6813 – 6818.

[68] L. Ouyang, M, Ma, M. Huang, R. Duan, H. Wang, L. Sun, M. Zhu, Ener-gies 2015, 8, 4237 – 4252.

[69] G. N. Ambaryan, M. S. Vlaskin, A. O. Dudoladov, E. A. Meshkov, A. Z.Zhuk, E. I. Shkolnikov, Int. J. Hydrogen Energy 2016, 41, 17216 – 17224.

[70] Y. Liu, X. Liu, X. Chen, S. Yang, C. Wang, Int. J. Hydrogen Energy 2017,42, 10943 – 10951.

[71] J.-Y. Uan, M.-C. Lin, C.-Y. Cho, K.-T. Liu, H.-I. Lin, Int. J. Hydrogen Energy2009, 34, 1677 – 1687.

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4275

Reviews

[72] M.-S. Zou, R.-J. Yang, X.-Y. Guo, H.-T. Huang, J.-Y. He, P. Zhang, Int. J. Hy-drogen Energy 2011, 36, 6478 – 6483.

[73] G. S. Frankel, A. Samaniego, N. Birbilis, Corros. Sci. 2013, 70, 104 – 111.[74] S. Fajardo, G. S. Frankel, Electrochim. Acta 2015, 165, 255 – 267.[75] H. Zhong, H. Wang, J. W. Liu, D. L. Sun, F. Fang, Q. A. Zhang, L. Z.

Quyang, M. Zhu, J. Alloys Compd. 2016, 680, 419 – 426.[76] L. Z. Ouyang, H. Zhong, Z. M. Li, Z. J. Cao, H. Wang, J. W. Liu, X. K. Zhu,

M. Zhu, J. Power Sources 2014, 269, 768 – 772.[77] A. Atrens, G. L. Song, Adv. Eng. Mater. 1999, 1, 11 – 33.[78] J. Kruger, G. L. Makar, Int. Mater. Rev. 1993, 38, 138 – 153.[79] H. H. R. Tunold, M. B. H. Berge, A. Lasson, R. S. Hansen, Corros. Sci.

1977, 17, 353 – 365.[80] G. Williams, H. N. McMurray, R. Grace, Electrochim. Acta 2010, 55,

7824 – 7833.[81] S.-H. Yu, J.-Y. Uan, T.-L. Hsu, Int. J. Hydrogen Energy 2012, 37, 3033 –

3040.[82] G. Williams, N. Birbilis, H. N. McMurray, Faraday Discuss. 2015, 180,

313 – 330.[83] A. Kanterk Figen, B. C. Filiz, Int. J. Hydrogen Energy 2015, 40, 16169 –

16177.[84] S. Oh, M. Kim, K. Eom, J. Kyung, D. Kim, W. Cho, H. Kwon, Int. J. Hydro-

gen Energy 2016, 41, 5296 – 5303.[85] M. Huang, L. Ouyang, H. Wang, J. Liu, M. Zhu, Int. J. Hydrogen Energy

2015, 40, 6145 – 6150.[86] S. Oh, T. Cho, M. Kim, J. Lim, K. Eom, D. Kim, E. Cho, H. Kwon, Int. J. Hy-

drogen Energy 2017, 42, 7761 – 7769.[87] A. Zabihallahpoor, M. Rahimnejad, F. Talebnia, RSC Adv. 2015, 5,

94171 – 94183.[88] M. Sharma, S. Bajracharya, S. Gildemyn, S. A. Patil, Y. Alvarez-Gallego,

D. Pant, K. Rabaey, X. Dominguez-Benetton, Electrochim. Acta 2014,140, 191 – 208.

[89] P. Pandey, V. N. Shinde, R. L. Deopurkar, S. P. Kale, S. A. Patil, D. Pant,Appl. Energy 2016, 168, 706 – 723.

[90] T. Ewing, P. T. Ha, H. Beyenal, Appl. Energy 2017, 192, 490 – 497.[91] A. Schievano, A. Colombo, M. Grattieri, S. P. Trasatti, A. Liberale, P.

Tremolada, C. Pino, P. Cristiani, J. Power Sources 2017, 340, 80 – 88.[92] J. Wu, W. Chen, Y. Yan, K. Gao, C. Liao, Q. Li, X. Wang, Bioelectrochemis-

try 2017, 117, 9 – 14.[93] C. E. Reimers, L. M. Tender, S. Fertigg, W. Wang, Environ. Sci. Technol.

2001, 35, 192 – 195.[94] D. R. Bond, D. E. Holmes, L. M. Tender, D. R. Lovley, Science 2002, 295,

483 – 485.[95] S. Z. Baykara, E. H. Figen, A. Kale, T. N. Veziroglu, Int. J. Hydrogen Energy

2007, 32, 1246 – 1250.[96] A. Demirbas, Energy Sources Part A 2009, 31, 1866 – 1872.[97] T. Y. Cong, A. Raj, J. Chanaphet, S. Mohammed, S. Ibrahim, A. Al Shoai-

bi, Int. J. Hydrogen Energy 2016, 41, 6662 – 6675.[98] M. Binoist, B. Lab8gorre, F. Monnet, P. D. Clark, N. I. Dowling, M.

Huang, D. Archambault, E. Plasari, P.-M. Marquaire, Ind. Eng. Chem. Res.2003, 42, 3943 – 3951.

[99] A. Ravikumar, A. Raj, S. Ibrahim, R. K. Rahman, A. A. Shoaibi, EnergyFuels 2016, 30, 10823 – 10834.

[100] N. O. Guldal, H. E. Figen, S. Z. Baykara, Chem. Eng. J. 2017, 313, 1354 –1363.

[101] S. A. Namana, I. E. Ture, T. N. Veziroglu, Int. J. Hydrogen Energy 2008,33, 6577 – 6585.

[102] D. Ipsakis, T. Kraia, G. E. Marnellos, M. Ouzounidou, S. Voutetakis, R.Dittmeyer, A. Dubbe, K. Haas-Santo, M. Konsolakis, H. E. Figen, N. O.Geldal, S. Z. Baykara, Int. J. Hydrogen Energy 2015, 40, 7530 – 7538.

[103] K. Petrov, S. Z. Baykara, D. Ebrasu, M. Gulin, A. Veziroglu, Int. J. Hydro-gen Energy 2011, 36, 8936 – 8942.

[104] S. Fukuzumi, Biochim. Biophys. Acta Bioenerg. 2016, 1857, 604 – 611.[105] S. Kato, J. Jung, T. Suenobu, S. Fukuzumi, Energy Environ. Sci. 2013, 6,

3756 – 3764.[106] F. Zhou, C. McDonnell-Worth, H. Li, J. Li, L. Spiccia, D. R. Macfarlane, J.

Mater. Chem. A 2015, 3, 16642 – 16652.[107] Y. Isaka, S. Kato, D. Hong, T. Suenobu, Y. Yamada, S. Fukuzumi, J. Mater.

Chem. A 2015, 3, 12404 – 12412.[108] Y. Kofuji, Y. Isobe, Y. Shiraishi, H. Sakamoto, S. Tanaka, S. Ichikawa, T.

Hirai, J. Am. Chem. Soc. 2016, 138, 10019 – 10025.[109] Y. Isaka, K. Oyama, Y. Yamada, T. Suenobu, S. Fukuzumi, Catal. Sci. Tech-

nol. 2016, 6, 681 – 684.[110] Y. Isaka, Y. Yamada, K. Ohkubo, T. Suenobu, S. Fukuzumi, RSC Adv.

2016, 6, 42041 – 42044.[111] S. Thakur, T. Kshetri, N. H. Kim, J. H. Lee, J. Catal. 2017, 345, 78 – 86.[112] Y. Aratani, T. Suenobu, K. Ohkubo, Y. Yamada, S. Fukuzumi, Chem.

Commun. 2017, 53, 3473 – 3476.[113] Y. Yamada, M. Yoneda, S. Fukuzumi, Energy Environ. Sci. 2015, 8, 1698 –

1701.[114] K. Mase, K. Ohkubo, S. Fukuzumi, J. Am. Chem. Soc. 2013, 135, 2800 –

2808.[115] K. Mase, K. Ohkubo, S. Fukuzumi, Inorg. Chem. 2015, 54, 1808 – 1815.[116] K. Mase, M. Yoneda, Y. Yamada, S. Fukuzumi, Nat. Commun. 2016, 7,

11470.[117] Z. Chen, J. J. Concepcion, N. Song, T. J. Meyer, Chem. Commun. 2014,

50, 8053 – 8056.[118] S. Nakagawara, T. Goto, M. Nara, Y. Ozawa, K. Hotta, Y. Arata, Anal. Sci.

1998, 14, 691 – 698.[119] L. Huang, R. Li, R. Chong, G. Liu, J. Han, C. Li, Catal. Sci. Technol. 2014,

4, 2913 – 2918.[120] A. Melis, Plant Sci. 2009, 177, 272 – 280.[121] K. Mase, M. Yoneda, Y. Yamada, S. Fukuzumi, ACS Energy Lett. 2016, 1,

913 – 919.[122] A. E. Sanli, B. Canan, A. AytaÅ, ECS Trans. 2015, 65, 51 – 58.[123] A. E. Sanli, O. Yilmaz, A. AytaÅ, Int. J. Energy Res. 2013, 37, 1205 – 1212.[124] A. E. Sanlı, M. D. Mat, Int. J. Hydrogen Energy 2015, 40, 6440 – 6448.

Manuscript received: July 28, 2017

Accepted manuscript online: September 15, 2017Version of record online: October 24, 2017

ChemSusChem 2017, 10, 4264 – 4276 www.chemsuschem.org T 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4276

Reviews