Hh

Sa

b

c

a

ARRAA

KHOHCC

1

cbqtiotriuttb

S(

k

0h

Electrochimica Acta 82 (2012) 493– 511

Contents lists available at SciVerse ScienceDirect

Electrochimica Acta

jou rn al hom epa ge: www.elsev ier .com/ locate /e lec tac ta

ydrogen peroxide as a sustainable energy carrier: Electrocatalytic production ofydrogen peroxide and the fuel cell

hunichi Fukuzumia,b,∗, Yusuke Yamadaa, Kenneth D. Karlinb,c,∗∗

Department of Material and Life Science, Graduate School of Engineering, Osaka University, ALCA, Japan Science and Technology Agency (JST), Suita, Osaka 565-0871, JapanDepartment of Bioinspired Science, Ewha Womans University, Seoul 120-750, Republic of KoreaDepartment of Chemistry, The Johns Hopkins University, Baltimore, MD 21218, USA

r t i c l e i n f o

rticle history:eceived 29 December 2011eceived in revised form 24 March 2012ccepted 27 March 2012vailable online 4 April 2012

eywords:ydrogen peroxidexygen reduction

a b s t r a c t

This review describes homogeneous and heterogeneous catalytic reduction of dioxygen with metalcomplexes focusing on the catalytic two-electron reduction of dioxygen to produce hydrogen per-oxide. Whether two-electron reduction of dioxygen to produce hydrogen peroxide or four-electronO2-reduction to produce water occurs depends on the types of metals and ligands that are utilized. Thosefactors controlling the two processes are discussed in terms of metal–oxygen intermediates involvedin the catalysis. Metal complexes acting as catalysts for selective two-electron reduction of oxygen canbe utilized as metal complex-modified electrodes in the electrocatalytic reduction to produce hydrogenperoxide. Hydrogen peroxide thus produced can be used as a fuel in a hydrogen peroxide fuel cell. A hydro-

ydrogen peroxide fuel cello complexu complex

gen peroxide fuel cell can be operated with a one-compartment structure without a membrane, whichis certainly more promising for the development of low-cost fuel cells as compared with two compart-ment hydrogen fuel cells that require membranes. Hydrogen peroxide is regarded as an environmentallybenign energy carrier because it can be produced by the electrocatalytic two-electron reduction of O2,which is abundant in air, using solar cells; the hydrogen peroxide thus produced could then be readilystored and then used as needed to generate electricity through the use of hydrogen peroxide fuel cells.

. Introduction

The rapid consumption of fossil fuel is expected to cause unac-eptable environmental problems such as the greenhouse effecty CO2 emission, which may lead to disastrous climatic conse-uences in the near future [1]. Even if climate change, such ashat due to global warming, turns out to be a less than expectedmportant problem, we are certainly on the verge of running outf fossil fuels by the end of 21st century, because the consump-ion rate of fossil fuel is expected to increase further by worldwideapid population and economic growth, particularly in the develop-ng countries [2]. Thus, renewable and clean energy resources arergently required in order to solve global energy and environmen-

al issues. Among renewable energy resources, solar energy is by farhe largest exploitable resource [3–8]. Of course, solar energy haseen utilized for ages in photosynthesis, leading to accumulated

∗ Corresponding author at: Department of Material and Life Science, Graduatechool of Engineering, Osaka University, ALCA, Japan Science and Technology AgencyJST), Suita, Osaka 565-0871, Japan.∗∗ Corresponding author.

E-mail addresses: fukuzumi@chem.eng.osaka-u.ac.jp (S. Fukuzumi),arlin@jhu.edu (K.D. Karlin).

013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.electacta.2012.03.132

© 2012 Elsevier Ltd. All rights reserved.

fossil fuel which we have been using so rapidly. It is therefore quiteimportant for us to obtain sustainable solar fuels such as hydrogenor others [3–10].

Hydrogen is a clean energy source for the future and it canbe used to reduce the dependence on fossil fuels and the emis-sions of greenhouse gases in the long-term [11–14]. The importantadvantage of hydrogen is that carbon dioxide is not produced whenhydrogen is burned to produce only water. Hydrogen should beideally produced by splitting water using solar energy. However,the storage of hydrogen has been a difficult issue, because hydro-gen is a gas having a low volumetric energy density. Tank systemshave been employed, either for gaseous pressurized hydrogen orliquid hydrogen. However, high-pressure equipment and a largedemand for energy for cryogenic purposes are involved. Otherapproaches, such as in the use of metal hydrides, carbon nano-tubes, and metal–organic frameworks can store or liberate onlylow amounts of hydrogen and unfavorable high temperatures arerequired to release the stored hydrogen [15–19]. Thus, none of theexisting processes for storage and carriage of hydrogen are envi-ronmentally benign.

On the other hand, hydrogen peroxide has merited significantattention, because H2O2 can oxidize various chemicals selectivelyto produce no waste chemicals but water [20–22]. Hydrogen per-oxide can be an ideal energy carrier alternative to oil or hydrogen,

4 himica Acta 82 (2012) 493– 511

btbpw[cbfTgawdrh

2

2

aa(ei(Oacpsaoobcm2e(tsao

2

w

dei

Fig. 1. Time profiles of formation of Fe(C5H4Me)2+ monitored at 650 nm in electron

transfer oxidation of Fe(C5H4Me)2 (1.0 × 10−1 mol L−1) by O2 (1.7 × 10−3 mol L−1),catalyzed by Co2(DPX) (2.0 × 10−5 mol L−1), Co2(DPA) (2.0 × 10−5 mol L−1),

94 S. Fukuzumi et al. / Electroc

ecause it can be used in a fuel cell leading to the generation of elec-ricity [23]. Thus, a combination of hydrogen peroxide productiony the electrocatalytic reduction of dioxygen in air with electricalower generated by a photovoltaic solar cell and power generationith a hydrogen peroxide fuel cell provides a sustainable solar fuel

24]. Currently H2O2 is mainly produced by the anthraquinone pro-ess, in which the hydroquinone in an organic solvent is oxidizedy molecular oxygen to produce H2O2 and quinone. The quinoneormed can then be reduced by hydrogen using Ni or Pd catalysts.hus, H2O2 is produced by the reduction of oxygen with hydro-en. In recent years, more than 3.5 million metric tons of H2O2re produced all over the world annually [25]. In this review, firste describe homogeneous vs. heterogeneous catalytic reduction ofioxygen with a variety of metal complexes and then we introduceecent development in the electrocatalytic production of H2O2 andydrogen peroxide fuel cells.

. Catalytic reduction of dioxygen with metal complexes

.1. Cobalt porphyrins

No reduction of O2 occurred by ferrocene derivatives suchs 1,1′-dimethylferrocene [Fe(C5H4Me)2] in acetonitrile (MeCN)t 298 K [26], because electron transfer from Fe(C5H4Me)2Eox = 0.26 V vs. SCE) [27] to O2 (Ered = –0.86 V vs. SCE) [28] is highlyndergonic. In the presence of HClO4 in MeCN solvent, however, O2s slowly reduced by Fe(C5H4Me)2 to produce hydrogen peroxideH2O2) via proton-coupled electron transfer from Fe(C5H4Me)2 to2 [27]. Mononuclear cobalt complexes with macrobicyclic hex-mine cage ligands have been reported to act as oxygen reductionatalysts for H2O2 production [29,30]. The addition of cobalt por-hyrin monomers or dimers and HClO4 to air-saturated MeCNolutions containing ferrocene derivatives results in significantlyccelerated O2-reduction (Scheme 1) [31,32]. The stoichiometryf the oxidation of ferrocene derivatives by O2 in the presencef HClO4 (two- vs. four-electron reduction of O2) was determinedy the concentration of ferrocenium cations compared with theoncentration of O2, when the concentration of O2 made to beuch smaller [32]. When a monomer cobalt porphyrin such as

,3,7,8,12,13,17,18-octaethyl-porphinato cobalt(II) [Co(OEP)] wasmployed as a catalyst, two equiv. of the ferrocenium cation[Fe(C5H4Me)2]+) were produced from O2, using Fe(C5H4Me)2 inhe presence of excess HClO4 in benzonitrile (PhCN) at 298 K ashown in Fig. 1 [32]. Thus, only two-electron reduction of O2 occursnd no further reduction occurs to produce more than two equiv.f Fe(C5H4Me)2

+ [Eq (1)].

Fe(C5H4Me)2 + O2 + 2H+ −→Co(OEP)

2Fe(C5H4Me)+2 + H2O2 (1)

Confirmation that a stoichiometric amount of H2O2 was formedas carried out using iodometric measurements [31].

In contrast to the monomeric cobalt porphyrin, when a cofacialicobalt porphyrin (Co2(DPX) in right column in Scheme 1) wasmployed as a catalyst, four equiv. of [Fe(C5H4Me)2]+ was producedn the catalytic reduction of O2 by Fe(C5H4Me)2 in the presence of

Scheme

Co2(DPB) (2.0 × 10−5 mol L−1), Co2(DPD) (2.0 × 10−5 mol L−1), and Co(OEP)(3.0 × 10−5 mol L−1) in the presence of HClO4 (2.0 × 10−2 mol L−1) in PhCN at 298 K.

HClO4 in PhCN at 298 K (Fig. 1) [Eq. (2)] [32]. Thus, the four-electronreduction of O2 occurs here efficiently. It was separately confirmedthat no H2O2 was formed in during this process [32].

4Fe(C5H4Me)2 + O2 + 2H+ −→Co2(DPX)

4Fe(C5H4Me)+2 + 2H2O (2)

The other cofacial dicobalt porphyrins [Co2(DPA), Co2(DPB) andCo2(DPD) in the left column of Scheme 1] also catalyze the reduc-tion of O2 by Fe(C5H4Me)2, but the amount of Fe(C5H4Me)2

+ formedis less than four equiv. based on the amount of O2 (Fig. 1) [32].This indicates that the clean four-electron reduction of O2 byFe(C5H4Me)2 occurs only in the case of Co2(DPX) and other dicobaltporphyrins as catalysts lead to a mixture of processes, i.e., involvingboth two- and four-electron stoichiometries.

The cyclic voltammograms of Co(OEP) and [Co2(DPX)] arecompared in Fig. 2, where the one-electron reduction poten-tial of Co(OEP) is determined to be 0.31 V corresponding tothe Co(II)/Co(III) couple, whereas the Co(II)/Co(III) couple for[Co2(DPX)] is split into two one-electron redox waves at 0.53, 0.39 V(vs. SCE) [32]. This indicates that two Co ions are interacting witheach other in [Co2(DPX)].

The catalytic cycle of the two-electron reduction of O2 byFe(C5HMe)2 in the presence of an acid in PhCN is shown in Scheme 2[32]. The initial electron transfer from Fe(C5H4Me)2 to Co(III)OEP+ isthermodynamically feasible judging from the one-electron oxida-tion potentials known for the compounds in question: Fe(C5H4Me)2(Eox = 0.26 V vs. SCE) and Co(OEP)+ (Ered = 0.31 V vs. SCE in Fig. 2).Indeed, electron transfer from Fe(C5HMe)2 to Co(OEP)+ occursand this is followed by the subsequent fast electron transferfrom Co(II)OEP to O2 in the presence of an acid to produce thehydroperoxyl species Co(III)(OEP)O2H+, which is further oxidized

by Fe(C5H5)2 in the presence of an acid to produce H2O2, accompa-nied by regeneration of Co(III)OEP+. The initial electron transfer isthe rate-determining step in the catalytic cycle, when the catalytic

1.

S. Fukuzumi et al. / Electrochimic

Fi

r

dtcC

tmMsdtcfge

ig. 2. Cyclic voltammograms of (a) Co2(DPX) and (b) Co(OEP) (1.0 × 10−3 mol L−1)n PhCN containing 0.1 mol L−1 TBAP; scan rate 100 mV s−1.

ate is given by Eq. (3) [32]. In such a case the catalytic rate

d[Fe(C5H4Me)+2 ]

dt= 2ket[Fe(C5H4Me)2][Co(OEP)+]

in the presence of O2 (3)

oes not depend on the concentration of O2 or an acid. In addition,he observed second-order rate constant is twice that of the rateonstant (ket) for the initial electron transfer from Fe(C5HMe)2 too(III)OEP+ (kobs = 2ket) [32].

The same catalytic scheme can be applied for the selectivewo-electron reduction of O2 by ferrocene derivatives with other

onomeric metalloporphyrin complexes: CoTPP+, FeTPP+ andnTPP+ (TPP2– = tetraphenylporphyrin dianion) [31]. The rate con-

tants of the rate-determining electron transfer from ferroceneerivatives to CoTPP+, FeTPP+ and MnTPP+ can be evaluated usinghe Marcus theory of outer-sphere electron transfer [33]. The Mar-

us relation provides that the rate constant for electron transferrom an electron donor (1) to an electron acceptor (2), k12, isiven by Eq. (4), where k11 and k22 are the rate constants for self-xchange or each component, 1 and 2, K12 is the electron-transfer

Scheme 2

a Acta 82 (2012) 493– 511 495

equilibrium constant, which is obtained from the one-electron oxi-dation potential of 1 and the one-electron reduction potential of2. The parameter f in Eq. (4) is given by Eq. (5), where Z is the fre-quency factor (1 × 1011 mol−1 L s−1) [33]. The k11 value of ferroceneis reported to be 5.3 × 106 mol−1 L s−1 [34], and the k22 values ofCo, Fe and Mn porphyrins are reported to be 20 [35], 1 × 109 [36],3.2 × 103 mol−1 L s−1 [37], respectively. The results are that basedon Eqs. (4) and (5) and using the k11 and k22 values, the rate con-stants for electron transfer from ferrocene derivatives to CoTPP+,FeTPP+ and MnTPP+ agree well with the observed rate constants[31]. Such agreement strongly indicates electron transfer from fer-rocene derivatives to metalloporphyrins occurs via an outer-spherepathway.

k12 = (k11k22K12f )1/2 (4)

log f = (log K12)2

[4 log(k11k22/Z2)](5)

The reason why only Co2(DPX) can act as a catalyst for theselective four-electron reduction of O2 (Fig. 2) can be understoodby comparing the distances between porphyrin moieties in thecofacial porphyrins complexes, based on data from reportedcrystal structures of Co2(DPB) [38], Co2(DPA) [39], Co2(DPX)[40,41], and Co2(DPD)(2MeOH) [40,41] (Fig. 3). The metal–metalseparations in Co2(DPA) (0.453 nm) and Co2(DPX) (0.458 nm) arevirtually the same. However, the xanthene spacer of Co2(DPX)is more flexible than the anthracene spacer of Co2(DPA) andthereby more suitable for the strong binding between two cobaltnuclei and O2. The metal–metal separation in Co2(DPB) (0.373 nm)may be too short, whereas the separation in Co2(DPD)(2MeOH)(0.862 nm) is too long to bind O2 between two cobalt nuclei.Thus, the interaction of two cobalt nuclei with an active form ofoxygen seems essential for the four-electron reduction of O2. It isinteresting to note that the Co–Co distance of Co2(DPX) (0.458 nm)is nearly the same as the Co–Co distance of a dicobalt(III) �2-�1:�1-peroxo complex with the tetrapodal pentaamine ligand2,6-bis(1′,3′-diamino-2′-methylprop-2′yl)pyridine (0.450 nm),which was structurally characterized by X-ray crystallographicanalysis [42]. The Co–Co distances in dicobalt bis-�-oxo com-plexes, {[Me2NN]Co}2(�-O)2 (Me2NN = ˇ-diketiminato) [43]and {[TpMe3]Co}2(�-O)2 (TpMe3 = hydrotris(3,5-dimethyl-4-methylpyrazolyl)borate), were reported to be 0.3067 nm and0.2724 nm, respectively [44]. The Co–Co distance of a bis-

cobalt peroxo complex, [Co3+(�,�1:�2-O2)(oxapyme)Co3+]2+

(oxapyme(H)2 = 2-(bis-pyridin-2-ylmethyl-amino)-N-[2-(5-{2-[2-(methyl-pyridin-2-ylmethyl-amino)]-phenyl}-[1,3,4]-oxadiazol-2-yl)-phenyl]-acetamide), in which one of the oxygen atoms

.

496 S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511

Fa

bmptb

fidpmspOase(CiyadbetHntshitt(

mottt

Fig. 4. EPR spectra of the �-superoxo complex (∼10−3 mol L−1) produced by addingiodine (∼10−3 mol L−1) to an air-saturated PhCN solution and ESR simulation of

ig. 3. Selected distance (nm) in Co2(DPB) [38], Co2(DPA) [39], Co2(DPX) [40,41],nd Co2(DPD)(2MeOH) [40,41].

ridges the two metals and is sideways bonded to one of theetals, was reported to be 0.3339 nm [45]. Thus, the biscobalt(III)

eroxo complex responsible for the catalytic four-electron reduc-ion of O2 may have the �2-�1:�1 bonding mode rather thanis-�-oxo or �,�1:�2 bonding modes.

The proposed mechanism of four-electron reduction of O2 byerrocene derivatives is summarized as shown in Scheme 3. Thenitial two-electron reduction of the Co(III)2 complex by ferroceneerivatives gives the Co(II)2 complex, which reacts with O2 toroduce the �-peroxo Co(III)-O2-Co(III) complex via the inter-ediacy of a �-superoxo complex (not shown). The �-superoxo

pecies in these cofacial dicobalt porphyrins could be separatelyroduced by the reactions of cofacial dicobalt(II) porphyrins with2 in the presence of a bulky base (1-tert-butyl-5-phenylimidazole)nd the subsequent one-electron oxidation of the resulting peroxopecies by iodine [32]. The EPR spectra of the �-superoxo speciesxhibit superhyperfine structure due to two equiv. cobalt nucleiFig. 4) [32]. The heterolytic O O-bond cleavage of the Co(III)-O2-o(III) complex affords the high valent Co(IV)-oxo species which

s reduced by ferrocene derivatives in the presence of protons toield H2O (Scheme 3). Alternatively the homolytic O O bond cleav-ge affords two Co(III)-oxyl species which are reduced by ferroceneerivatives to H2O in the presence of an acid. In each case, the O Oond cleavage of the Co(III)-O2-Co(III) complex leads to the four-lectron reduction of O2 (Scheme 3) whereas if protonation firstakes place then overall two-electron reduction of O2 to produce2O2 would occur. The stronger the binding between two cobaltuclei and oxygen in the Co(III)-O2-Co(III) complex, the weaker ishe O O bond, and the faster is the rate of O O bond cleavage. Thiseems to be the case for Co2(DPX), which has the largest super-yperfine coupling constant of the �-superoxo species (Fig. 4),

ndicating greater interaction of the superoxide moiety spin withhe cobalt(III) centers, and this acts as the most efficient catalyst forhe selective four-electron reduction of O2 by ferrocene derivativesFig. 1).

Detailed kinetic investigations and analyses on the rate of for-ation of Fe(C5H5)2

+ in the Co2(DPX)-catalyzed electron transferxidation of Fe(C5H5)2 by O2 in the presence of HClO4 revealed that

he rate-determining step (Scheme 3) is a proton-coupled electronransfer from Co(III)Co(II)(DPX)+ to O2 following initial electronransfer from Fe(C5H5)2 to Co(III)2(DPX)2+ [32]. In this case the

(a) Co2(DPB), (b) Co2(DPA) and (c) Co2(DPX) in the presence of 1-tert-butyl-5-phenylimidazole (5 × 10−3 mol L−1) at 298 K [32].

rate of formation of Fe(C5H5)+ is proportional to the concentra-tions of Co(III)2(DPX)2+, O2 and HClO4. When Fe(C5H5)2 is replacedby a much stronger reductant, that is Fe(C5Me5)2 (Fc*) however,the kinetics of formation of Fc*+ change drastically from first-orderkinetics in the case of Fe(C5H5)2 to zero-order kinetics, because therate-determining step is changed to an O O bond cleavage stepin the Co(III)-O2-Co(III) complex [32]. In this situation, the rateremains constant irrespective of change in concentrations of O2and HClO4. The O O bond cleavage rate has been determined as320 s−1 [32].

The electrocatalytic reduction of O2 was examined usingCo2(DPX) and Co2(DPD), which were adsorbed onto an electrodesurface by means of a dip-coating procedure [46]. Rotating Pt ring-

disk voltammograms for reduction of O2 at pyrolytic graphite diskscoated with Co2(DPX) and Co2(DPD) revealed the catalytic four-electron reduction of O2 with 72% and 80% selectivity, respectively

S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511 497

eme 3.

[raifs

2

eocc1Fwua(pr

becpeHvta

Fig. 5. Visible absorption spectra changes in the catalytic reduction of O2

(1.7 × 10−3 mol L−1) by Fe(C5H4Me)2 (2.0 × 10−2 mol L−1) in the presence of HClO4

(2.0 × 10−2 mol L−1) and 1–6 (2.0 × 10−5 mol L−1) in PhCN at 298 K; 1 (red solid line),2 (green solid line), 3 (blue solid line), 4 (red broken line), 5 (green broken line) and6 (blue broken line). Black: in the absence of cobalt complex. (For interpretation ofthe references to color in this figure legend, the reader is referred to the web versionof the article.)

Table 1Electroreduction of dioxygen by adsorbed biscobalt porphyrin–corrole dyads in air-saturated 1.0 mol L−1 HClO4.

Bridge (Y) (PMes2CY)Co2 (1–3) (PCY)Co2 (4–6)

Epa E1/2

b nc Epa E1/2

b nc

O 0.25 0.32 2.4 0.34 0.41 3.547

Ox 0.27 0.33 2.5 0.35 0.40 3.1X 0.25 0.32 2.5 0.38 0.45 3.747

Sch

46]. The partial formation of H2O2 was clearly detected by theotating ring current [46]. Thus, the selectivity for the electrocat-lytic four-electron reduction of O2 is lower than the selectivityn the homogeneous system (Fig. 1). Such differences in selectivityor homogeneous versus heterogeneous catalysis of O2-reducingystems will be discussed further in the next section.

.2. Biscobalt porphyrin–corrole complexes

The same selectivity with regard to two-electron vs. four-lectron reduction of O2 by Fe(C5H4Me)2 depending on the typef linkage (Y) of biscobalt complexes (Fig. 1) was observed for theatalytic reduction of O2 with cofacial biscobalt porphyrin–corroleomplexes (Chart 1) in the presence of HClO4 in PhCN [47]. When, 2, 4 or 5 is used as a catalyst, two-electron reduction of O2 bye(C5H4Me)2 occurred efficiently in the presence of HClO4 in PhCN,hereas the four-electron reduction of O2 occurred when 3 or 6 wassed as a catalyst as shown in Fig. 5 [47]. Thus, in this case as well,

suitable metal–metal separation with Y = 9,9-dimethylxantheneX) is required to produce the �-peroxo Co(III)-O2

2–-Co(III) com-lex that is the key intermediate for the catalytic four-electroneduction of O2.

The catalysts 1–6 were also adsorbed onto a graphite disky transferring aliquots of a solution in CHCl3 directly to thelectrode surface followed by evaporation of the solvent as thease of Co2(DPX) and Co2(DPD) (vide supra) [46]. The biscobaltorphyrin–corrole complex-modified electrodes were used toxamine the electrocatalytic properties in the presence of 1 mol L−1

ClO4 in PhCN [47]. The catalytic activity was determined by cyclicoltammetry as well as by rotating disk electrode voltammetry andhe results are summarized in Table 1 [47]. All six complexes cat-lyze the electrocatalytic reduction of O2 at potentials close to their

a Peak potential of the dioxygen reduction wave (V vs. SCE).b Half-wave potential (V vs. SCE) for dioxygen reduction at rotating disk electrode

(ω = 100 rpm).c Electrons consumed in the reduction of O2 as estimated from the slope of the

Koutecky−Levich plots.

498 S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511

alt po

Eo

wrwFti[

I

ws

F(tS

Chart 1. Structures of biscob

1/2 values (Table 1). The average E1/2 value for the electroreductionf O2 at a rotating disk electrode coated with 1–3 is 0.32 V.

The number of electrons transferred during oxygen reductionas calculated from the magnitude of the steady-state limiting cur-

ent values, which were taken at a fixed potential on the catalyticave plateaus of the different current-voltage curves, as shown in

ig. 6a [47]. If mass transport alone controls the reduction of O2 athe 2-modified electrode, then the relationship between the lim-ting current and rotation rate obeys the Levich equation [Eq. (6)]48,49]:

Lev = 0.62nFAv−1/6D2/3[O2]ω1/2 (6)

here n is the number of electrons transferred, F is the Faraday con-tant, A is the electrode area (cm2), v is the kinematic viscosity of

ig. 6. Electrocatalytic reduction of O2 in 1 mol L−1 HClO4 at a rotating graphite disk elecw) are indicated on each curve. The disk potential was scanned at 5 mV s−1. (b) Levich plheoretical curve expected for the diffusion–convection limited reduction of O2 by 2e–. (upporting electrolyte: 1 mol L−1 HClO4 saturated with air.

rphyrin–corrole complexes.

the solution (cm2 s−1), D is the O2 diffusion constant (cm2 s−1), [O2]is bulk concentration of O2 (mol L−1), and ω is the rate of rotation(rad s−1). According to Eq. (6), the plot of the limiting current den-sity I/A vs. ω1/2 gives a straight line intersecting the origin as shownin Fig. 6b. The deviation from the initial linearity suggests that thecatalytic reaction is limited by kinetics, not by mass transport alone.In that case, the number of electrons transferred can be determinedby using the Koutecky–Levich equation [Eq. (7)] [50,51]:

(I

A

)−1= (nFk[O2])−1 + (0.62nFv−1/6D2/3[O2]w1/2)

−1(7)

where k is the second-order rate constant of electron transfer(mol−1 L s−1), which limits the plateau current. The number ofelectrons transferred in the O2 electroreduction process involving

trode coated with (PMes2CO)Co2 1. (a) Values of the rotation rates of the electrodeots of the plateau currents of (a) vs. (rotation rate)1/2. The dashed line refers to thec) Koutecky–Levich plots of the reciprocal plateau currents vs. (rotation rates)−1/2.

S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511 499

eme 4

cipvTcm[

aFiep(cstrpiscp

5eotc

F(

Sch

omplexes 1–3 ranges from 2.4 to 2.5 (Table 1 and Fig. 6c) [47]. Thisndicates that the electrocatalytic reduction of O2 by the biscobaltorphyrin–corrole complexes results in formation of H2O2 mainlyia the two-electron reduction of O2 in the presence of HClO4.hese values are different from the biscobalt porphyrin–corroleomplexes 4–6 where n = 3.1–3.7 (Table 1). In the case of 6, O2 wasainly reduced to H2O via a four-electron and four-proton process

47].The number of electrons transferred in the O2-electroreduction

t the electrode process involving complexes 4–6 (Table 1 andig. 6) is different from that observed in the homogeneous phasen Fig. 4 except for the case of complex 6 (n = 3.7). This differ-nce (Table 1), and similar variations observed for the biscobaltorphyrin complexes in homogeneous vs. heterogeneous solutionFig. 1) may result from different coordination geometries of bis-obalt complexes which may occur of their interactions with theurface material when they are adsorbed onto the graphite elec-rode. As already described, the selectivity for the four-electroneduction of O2 is quite sensitive to the Co–Co distance of biscobaltorphyrins (Fig. 3), thus an electrode surface induced small change

n the nature of the proximity of each porphyrin moiety within aynthetic dyad, i.e., the biscobalt complexes, may result in signifi-ant changes in terms of the selectivity toward differing reductionrocesses.

When a monomeric cobalt corrole ([10-pentafluorophenyl-,15-dimesityl-corrole]cobalt complex, Co(F5PhMes2Cor), was

mployed as an electrocatalyst for the reduction of O2, the slopef the Koutecky–Levich plot shows that the catalytic electroreduc-ion of O2 is a pure two-electron process to produce H2O2 [52]. Theatalytic process employing Co(F5PhMes2Cor) was confirmed for

ig. 7. (a) X-ray structures of the fully reduced bimetallic heme a3/CuB center in CcO fromc) Cu-free version of synthetic model for CcO (6LFe).

.

the homogeneous phase using Fe(C5H4Me)2 as a reductant in PhCNsolvent [52]. Electron transfer from Fe(C5H4Me)2 (Eox = 0.26 V vs.SCE) to [Co(F5PhMes2Cor)]+ (Ered = 0.38 V) [53] occurs efficiently toproduce [Fe(C5H4Me)2]+ and Co(F5PhMes2Cor) [52]. The cobalt(III)corrole complex, Co(F5PhMes2Cor), can reduce O2 in the pres-ence of HClO4. The site of electron transfer has been confirmed tobe the corrole ligand, based on the finding by EPR spectroscopyof an observed g-value of 2.0032 for the singly oxidized cobaltcorrole, that separately obtained by the chemical oxidation ofCo(F5PhMes2Cor) with one equivalent of [Fe(bpy)3]3+ (bpy = 2,2′-bipyridine). This signal is characteristic of an organic radical; itis quite different from the large g-value (2.037) observed forcobalt(IV) porphyrin complexes. In contrast to the case of cobaltporphyrins (Scheme 2), the cobalt corrole complex acts as an effec-tive catalyst in the two-electron reduction of O2 with HClO4 viathe redox couple between the cobalt(III) corrole and the cobalt(III)corrole radical cation (Scheme 4).

2.3. Cytochrome c oxidase models

In the final step in the biological respiratory chain, the four-electron and four-proton reduction of O2 to H2O is efficientlycatalyzed by the heme/copper (heme a3/CuB) heterodinuclear cen-ter in cytochrome c oxidases (CcO) (Fig. 7a) [54–56]. Biomimeticchemical modeling of the CcO active site has extensively beenstudied to provide not only the mechanistic insights into the four-

electron and four-proton reduction of O2 but also as a blueprint forbioinspired fuel cells [56–62]. A number of heme a3/CuB syntheticanalogues have been developed to mimic the coordination envi-ronment of the heme a3/CuB bimetallic center in CcO [56–61]. The

bovine heart (FeII· · ·CuI = 0.519 nm), (b) heme/Cu synthetic model for CcO (6LFeCu),

5 himic

eamre(smwtitdi[

ripro[rOr[zirOdFOfb5atv[

t

Fra

00 S. Fukuzumi et al. / Electroc

lectrocatalytic function of heme a3/CuB synthetic analogues haslso been examined by using electrodes modified with the syntheticodels to perform the catalytic four-electron and four-proton

eduction of O2 [59–63]. However, the structure of synthetic mod-ls may be changed when they are adsorbed on an electrode surfacevide supra). In addition, the solid supported state employed foruch studies has precluded any spectroscopic monitoring or inter-ediates detection. Thus, the catalytic reduction of O2 to wateras examined using ferrocene derivatives as one-electron reduc-

ants and a heme/Cu functional model of CcO (6LFeCu, Fig. 7b) andts Cu-free version (6LFe, Fig. 7c) as catalysts in homogeneous solu-ions [64]. The detailed kinetic analysis together with spectroscopicetection of reactive intermediates has provided new mechanistic

nsights into the O–O reductive cleavage process as described here64].

The catalytic mechanism for the four-electron and four-protoneduction of O2 by decamethylferrocene (Fc*) with 6LFeCu and 6LFen acetone is summarized in Scheme 5a and b, respectively. In theresence of acid, [6LFeIII-O-CuII]+ is converted to [6LFeIIICuII]3+ byeleasing water and the catalytic cycle starts via a fast reductionf the heme and then the Cu to generate the reduced complex

6LFeIICuI]+. Then O2 binds to [6LFeIICuI]+, and this is followed byapid protonation affording the Fe-hydroperoxo complex {6LFeIII-OH CuII}2+. Reductive O O bond cleavage is followed by further

apid reduction to produce H2O accompanied by regeneration of6LFeIIICuII]3+. The rate of formation of Fc* was zero-order and theero-order rate constant increased proportionally with increas-ng the catalyst concentration, but the zero-order rate constantemained constant with variation of concentrations of TFA and2 at 213 K [64]. This unusual kinetics indicates that the rate-etermining step is a process which does not involve reactions withc*, H+, O2. Such a reaction is the O O bond cleavage in {6LFeIII-OH CuII}2+, followed by rapid electron transfer to complete the

our-electron and four-proton reduction of O2. This is confirmedy the steady-state observation of {6LFeIII-OOH CuII}2+ (�max = 415,38 nm) during the catalytic reduction of O2 by Fc* with 6LFeCut 213 K. {6LFeIII-OOH CuII}2+ was independently generated at lowemperature (193 K) by the addition of an excess of TFA to the pre-

iously well characterized peroxo complex [6LFeIII-(O2

2−)-CuII]+

65].In the case of the Cu-free version 6LFe (Scheme 5b) as well,

he rate of formation of Fc* was zero-order and the zero-order

ig. 8. (a) Levich plot from the plateau currents in plot of current vs. (rotation rate)1/2. (bate)1/2]. The dashed lines in (a) and (b) were obtained from the calculated diffusion–convs two and four.

a Acta 82 (2012) 493– 511

rate constant increased proportionally with increasing the catalystconcentration, but the zero-order rate constant remained constantwith variation of concentrations of TFA and O2 at 213 K [64]. Thisagain indicates that the rate-determining step in the catalytic cycleis the O O bond cleavage of 6LFeIII-OOH. Surprisingly the bondcleavage rate of 6LFeIII-OOH is the same as that of {6LFeIII-OOHCuII}2+. This suggests that the Cu is not bound to the FeIII-OOHmoiety in {6LFeIII-OOH CuII}2+.

This rate-determining step found to occur at 213 K is changedto be the process of O2-binding to [6LFeIICuI]+ at 298 K when thezero-order rate constant increases proportionally with increasingconcentration of O2 [64]. The change in the rate-determining stepat 298 K was confirmed by the change in the steady-state speciesidentified to be {6LFeIII-OOH CuII}2+ at 213 K instead to [6LFeIICuI]+

(�max = 422 nm) at 298 K. In contrast to the case at 213 K, however,the O2-binding rate at 298 K to [6LFeIICuI]+ is significantly fasterthan that to 6LFeII [64]. This result suggests that the role of the Cuin 6LFeCu, at ambient temperature, is to assist the heme and leadto faster O2-binding during the catalytic cycle.

The electrocatalytic reduction of O2 was examined using an edgeplane pyrolytic graphite (EPG) disk shape electrode, which wasmodified with 6LFeCu [66]. The modified EPG electrode was pre-pared by transferring an MeCN solution of [6LFeIICuI]+ on the EPGsurface and allowing the solvent to evaporate in an argon atmo-sphere giving a dried surface [66]. Levich and Koutecky–Levichplots [Eqs. (6) and (7)] are shown in Fig. 8a and b, respec-tively [66]. The dashed lines were obtained from the calculateddiffusion–convection controlled currents based on the Levich equa-tion assuming the number of electrons for O2 reduction as two orfour. The open circles were from the measured plateau currentsand they were higher than those with the two-electron reduction.The deviation from linearity of the four-electron transfer plot sug-gests that the catalytic reaction is limited by kinetics in additionto the mass-transfer process. In the Koutecky–Levich plot in Fig. 8bobtained from the data in Fig. 8a, the measured values (open circles)were parallel to the line of four-electron reduction. This indicatesthat the adsorbed complex [6LFeIICuI]+ catalyzed the four-electronreduction of O2 to H2O as is the case in the homogeneous solu-

tion (vide supra). The four-electron reduction of O2 to H2O wasconfirmed by using rotating EPG disk-platinum ring electrodes tomeasure the fraction of O2, which was reduced to H2O, rather thanto H2O2. With adsorbed [6LFeIICuI]+, a small anodic ring plateau

) Koutecky–Levich plot from the plateau currents in Fig. 2 [(current)−1 vs. (rotationection controlled currents for the reduction of O2 assuming the number of electrons

S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511 501

eme 5

ce[

2

ipfiaroa

Sch

urve was observed during the reduction of O2 at the EPG disklectrode and it was found that most O2 (85%) was reduced to H2O66].

.4. Cu complexes

There has been considerable interest in the use of coppern O2-reduction chemistry. One reason is the existence of cop-er containing enzymes which efficiently effect the four-electronour-proton reduction to water as part of their function. Thesenclude so-called multi-copper oxidases (MCO’s) [67–70] wherein

n array of copper ions arranged in a tricopper cluster plus a sepa-ate but electronically linked copper ion, the latter which effectsne-electron oxidations (at a time) of biological substrates, e.g.,scorbate, phenols, diamines, Fe(II) and Cu(I). Such studies have

.

lead to the use of such enzymes, supported on electrodes, to carryout electrocatalytic dioxygen reductions in fact having very highactivity [71–73].

However, the existence of MCO’s and particularly inspired bythe existence of dicopper enzymes which are blood oxygen carriersin mollusks and arthropods (i.e., hemocyanins) [71,73], or bio-logically ubiquitous tyrosinases which ortho-hydroxylate phenolsgiving catechols and/or quinone products which are converted tomelanin pigments [70,72], has led synthetic bioinorganic chemiststo vigorously pursue the study of dioxygen binding and reactivitywith discrete copper ion complexes [74–78]. Such investigations

have revealed many new fundamental aspects in the field, includingthe uncovering of a number of now very well defined copper(I)-O2derived complexes, existing in different structural forms, althoughall possessing the Cu2-O2 core. Four structural types are known

502 S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511

F educti .

[opcSp

pftvo[gtmfortl

ct

ig. 9. Complexes used as catalysts for the solution four-electron four-proton rntermediates known to form during the course of reaction. See text for discussions

75,77,78], and the generation of one or another depends primarilyn the nature of the nitrogenous ligand employed, and the com-ounds formed can arise from differing mononuclear copper-ligandomplexes or those derived from the use of binucleating ligands.ome of these are shown in Fig. 9, along with diagrams of the coreeroxo-dicopper(II) or bis-�-oxo-dicopper(III) structures.

While the activity of supporting discrete ligand–metal com-lexes on electrode surfaces is an active area of research, includingor copper (see below), homogeneous catalytic O2-reduction usinghe complexes shown in Fig. 9 in water and/or organic sol-ents comes about when employing ferrocene derivatives asne-electron outer-sphere reductants and acids as proton sources79,80]. Electrocatalysis certainly has advantages, including thereater likelihood for practical application, however solution inves-igations enable kinetic and spectroscopic (e.g., UV–vis, EPR)

onitoring of key steps during catalysis. This can and does provideor insights into mechanism, via the determination of the identityf intermediates and the order of reaction of a particular step withespect to electron and/or proton concentrations. Such informa-ion can provide for elaboration or altering of the catalyst complex

igand, so as to improve future performance.

In fact, all of Fig. 9 complexes can catalyze O2-reductionhemistry [79,80]. First using Fc* and perchloric acid in ace-one solutions, the catalytic 4e–/4H+ reduction of O2 to water

Scheme

ion of O2 by ferrocene derivatives and the copper(I)-dioxygen derived complex

with [(tmpa)CuII(H2O)]2+ was examined by following detailedspectroscopic and kinetic monitoring [79]. The mechanism derivedis shown in Scheme 6 [79]. The addition of a catalytic amount of[(tmpa)CuII(H2O)]2+ to an O2-saturated acetone solution of Fc* andHClO4 resulted in the efficient oxidation of Fc* by O2 to afford fer-rocenium cation (Fc*+). Fig. 10 shows the spectral changes observedfor stepwise addition perchloric acid. For each time period ofacid addition (see Fig. 10 Inset), the concentration of Fc*+ whichwas formed immediately (absorption increases at �max = 380 and780 nm) was the same as the mole-amount of added HClO4.

The production of water was confirmed by mass spectrom-etry when running reactions using 18O water. Two-electronO2-reduction chemistry could be ruled out since iodometric titra-tion experiments yielded no detectable hydrogen peroxide. In thepresence of limiting [O2], and excess Fc*, still only 4 equiv. Fc*+

was formed in the presence of 4 equiv. of HClO4. Thus, theseexperiments (and others) demonstrated the overall reaction sto-ichiometry, as catalyzed by [(tmpa)CuII(H2O)]2+, to be:

4Fc∗ + O2 + 4H+ → 4Fc∗+ + 2H2O.

It has been demonstrated that the rate limiting step inthe reaction cycle was reduction of [(tmpa)CuII(H2O)]2+ by Fc*[79]. Then, under conditions where Fc* and acid were depleted,

6.

S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511 503

Fig. 10. UV–vis spectral changes in four-electron reduction of O2 by Fc*(1.5 × 10−3 mol L−1) with [(tmpa)CuII(H2O)]2+ (9.0 × 10−5 mol L−1) in the pres-ence of HClO4 in acetone at 298 K. The inset shows the changes inat(

sg[tsttoiw[

aeff[[taeatieTo1

i4EcobKmke

Fig. 11. Depiction of the structure of the �-�2:�2-peroxo dicopper(II) complex[CuII

2(N3)(O2)]2+ (b) Formation of the �2:�2-peroxo complex (�max = 490 nm) inthe reaction of [CuI

2(N3)]2+ (1.0 × 10−4 mol L−1) with O2 in the presence of Fc*(8.0 × 10−2 mol L−1) in acetone at 193 K. The Inset shows the time profiles of theabsorbance at 490 nm (black line) and 780 nm (red line) due to [CuII

2(N3)(O2)]2+

and Fc*+, respectively. (c) UV/Vis spectral changes observed in the four-electronreduction of O2 (0.22 × 10−3 mol L−1) by Fc* (3.0 × 10−3 mol L−1) at 298 K and with

−2 −1 II 2+ −4 −1

bsorbance at 380 and 780 nm due to Fc*+ produced by stepwise addi-ion of HClO4 (0.18–1.44 × 10−3 mol L−1) to an O2-saturated acetone solution[O2] = 11 × 10−3 mol L−1) of Fc* and [(tmpa)CuII(H2O)]2+.

topped-flow kinetic measurements revealed the very rapideneration of the well known dioxygen adduct, peroxo complex(tmpa)CuII(O2)CuII(tmpa)]2+ [75,77–79] (Scheme 6). Separately,his complex could be cleanly generated at 193 K in solution andubjected to Fc* and/or acid; such experiments led to the findinghat when both are present, electron transfer reductive cleavage ofhe peroxidic O–O bond in [(tmpa)CuII(O2)CuII(tmpa)]2+ occurs, butnly if acid is present. Thus, this key peroxo-dicopper(II) complexntermediate is reduced faster than it is simply protonated (which

ould yield H2O2), leading to overall 4e–/4H+ O2 reduction to water79].

Other complexes shown in Fig. 9 also have been shown to cat-lyze the O2-reduction to water, even though they form very differ-nt copper-dioxygen adduct structures. [CuII

2(N3)(H2O)2](ClO4)4,ollowing reduction, reacts rapidly with molecular oxygen toorm a side-on bound �-�2:�2-peroxo dicopper(II) complexCuII

2(N3)(O2)]2+ (Figs. 9 and 11a), as previously demonstrated81–83]. When a catalytic amount of [CuII

2(N3)(H2O)2]2+ is addedo and O2-saturated acetone solution with Fc* and trifluoroaceticcid, O2 is reduced to water and four equiv. Fc*+ are formed,ven if excess Fc* relative to O2 (i.e., limiting [O2]) and acidre present (also see the inset for Fig. 11b) [80]. Along withhe finding by iodometric titration that no hydrogen peroxides produced, it was concluded that [CuII

2(N3)(H2O)2]2+ is anfficient catalyst for the four-electron reduction of O2 by Fc*.he same results were obtained using the weaker reductant,ctamethylferrocene (Me8Fc), however no reaction occurred with,1′-dimethylferrocene [80].

To further break down the individual reaction steps, exper-ments were carried out on [CuII

2(N3)(O2)]2+ (�max = 365 and90 nm, Fig. 11c) which was separately generated at 193 K [80].lectron transfer reduction of this complex by Fc* occurs with con-omitant production of two equiv. Fc*+ in acetone, see the decayf absorbances due to [CuII

2(N3)(O2)]2+ which are accompaniedy the rise of absorbance at 780 nm due to Fc*+ (Fig. 11c Inset).

inetic interrogation on the rate of electron transfer led to deter-ination that Fc* reduction of [CuII

2(N3)(O2)]2+ occurred withet = 18 mol−1 L s−1 (193 K), a value that was not altered in the pres-nce of one equiv. acid. Me2Fc was not able to effect this reduction.

TFA (1.0 × 10 mol L ) catalyzed by [Cu 2(N3)(H2O)2] (1.0 × 10 mol L ). (Forinterpretation of the references to color in this figure legend, the reader is referredto the web version of the article.)

The latter finding indicates that the electron transfer is not coupledwith the protonation of [CuII

2(N3)(O2)]2+ [80].Overall, the data supported a reaction mechanism where the

dicopper(II) catalyst was reduced by Fc* to a dicopper(I) com-plex, previously established to be [CuI

2(N3)]2+, which reactsvery rapidly with O2, giving [CuII

2(N3)(O2)]2+. In a rate-limitingstep, this is reduced likely to a transient bis-�-oxo-dicopper(II)species which is very rapidly protonated to give back the catalystand two mol-equiv H2O, thus completing the 4e–/4H+ reduc-tion of dioxygen, Scheme 7 [80]. Consistent with the proposed

mechanism and a piece of valuable new information derivedfrom the studies was a very good estimate of the potential forreduction of [CuII

2(N3)(O2)]2+. Based on the known one-electronoxidation potentials for ferrocene derivatives and that of N,N,N′,

504 S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511

eme

N[pp(EtlcTd�q([n[twfree[iic[

Sch

′-tetramethylphenylenediamine (TMPD) (Eox = 0.12 V vs. SCE)80], which also effected the reduction of [CuII

2(N3)(O2)]2+ accom-anied by the formation of TMPD•+, the one-electron reductionotential of [CuII

2(N3)(O2)]2+ could be estimated to be Ered = 0.19±0.07) V vs. SCE [80]. That value is significantly lower than thered value of (0.37 V vs. SCE) for the catalyst [CuII

2(N3)(H2O)2]2+,hese results being consistent with rapid reduction of the latter fol-owed by rate limiting reduction of the �-�2:�2-peroxo dicopper(II)omplex [CuII

2(N3)(O2)]2+ formed during the catalytic cycle [80].here now exists a huge literature describing how �-�2:�2-peroxoicopper(II) complexes [Cu· · ·Cu ∼ 0.36 nm; O–O ∼ 0.14–0.15 nm;O–O < 760 cm−1, resonance Raman spectroscopy (rR)] not infre-uently exist in equilibrium with a bis-�-oxo-dicopper(III) isomerCu· · ·Cu ∼ 0.28 nm, O· · ·O ∼ 0.23 nm, �Cu–O ∼ 600 cm−1 (intense)75,77,83–85]. While rR measurements on [CuII

2(N3)(O2)]2+ doot show any hint of the presence of a species [CuIII

2(N3)(O)2]2+

83], it still had to be considered that it might be formed here inhis catalytic process, and that the bis-�-oxo-dicopper(III) speciesas actually the species which was being reduced by Fc*, thus

ollowing and not prior to O–O cleavage. However, this could beuled out as follows. The temperature dependence of the variouslectron-transfer steps were evaluated, leading to the finding thatlectron transfer from Fc* or Me8Fc to both [CuII

2(N3)(H2O)2]2+ andCuII

2(N3)(O2)]2+ virtually the same values for �S /= (∼0, whichs expected and typical for outer-sphere electron transfer chem-stry) indicating a direct process, not what one would find if a

onversion of [CuII

2(N3)(O2)]2+ to bis-�-oxo-dicopper(III) speciesCuII

2(N3)(O)2]2++ occurred [80].

Scheme

7.

With the value Ered = 0.19 (±0.07) V vs. SCE determined forperoxo complex [CuII

2(N3)(O2)]2+, �Get values for Fc* and Me8Fccould be determined, (–0.27 ± 0.07) and (–0.23 ± 0.07) eV, respec-tively. The � value for electron transfer from Fc* and Me8Fc to[CuII

2(N3)(O2)]2+ could then be estimated [80] and found to be2.2 (±0.1) eV, which was significantly larger than the correspond-ing value for electron transfer from ferrocene derivatives to thedicopper(II) catalyst [CuII

2(N3)(H2O)2]2+. This finding is also con-sistent with direct electron transfer from ferrocene derivativesto the peroxo-dicopper(II) complex intermediate, as large struc-tural changes would of course be a part of an O–O cleavageprocess, but not so for the case where a bis-�-oxo-dicopper(III)complex with already cleaved O–O bond were undergoing simpleelectron-transfer reduction to give the corresponding bis-�-oxo-dicopper(II) complex [CuII

2(N3)(O)2]2+ (that in Scheme 8) [80].Investigations on O2-reduction catalysis with [CuII(BzPY1)]2+

(Fig. 9) were carried out in a similar manner to those discussedfor [CuII

2(N3)(H2O)2]2+ (vide supra) [80]. The findings can be sum-marized as follows (Scheme 9) [80]: [CuII(BzPY1)]2+ is reducedby Fc* giving copper(I) species [CuI(BzPY1)]2+ whose reactionwith O2 is known to afford the bis-�-oxo dicopper(III) complex[{CuIII(BzPY1)}2(O)2]2+ (�max = 390 nm). Electron transfer reduc-tion by Fc* occurs rapidly upon mixing in acetone even at 193 Kto produce two equiv. of Fc*+ and the reaction rate was not affectedby TFA. Fast protonation gives two mole-equiv. H2O and the cat-alyst. Reaction of [{CuII(BzPY1)}2(O)2]2+ also occurs with weaker

electron donors, Me2Fc and even Fc itself. In the latter case how-ever, no catalysis occurs because these latter reductants can’t even

8.

S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511 505

eme 9

rt

irrp[rttaitraacc

alCOo

Sch

educe [CuII(BzPY1)]2+ to [CuI(BzPY1)]+ as �Get > 0, and thereforehe catalytic cycle cannot begin [80].

To summarize (Scheme 10), different copper(II) complexes giv-ng rise to completely very different Cu2O2 structures followingeduction and reaction with molecular oxygen, all are readilyeduced by ferrocene one-electron reductants, leading to the cop-er ion catalyzed overall 4e–/4H+ reduction of dioxygen to water80]. The key step leading to this reaction stoichiometry is fer-ocenyl reductive O–O bond cleavage (by two-electron equiv.) athe peroxo-dicopper(II) level, at least for two of the three cases. Inhe other, with catalyst [CuII(BzPY1)]2+, the copper(I)/O2 chemistrylready gives rise to O–O cleavage and a bis-�-oxo-dicopper(III)ntermediate which is then reduced (Scheme 10). To finish (or start)he catalytic cycle, the copper(II) complex(es) then formed areeduced to copper(I) for further O2-reaction. With the great vari-tions in ligand design for copper complexes which are possible,nd can lead to varying patterns of Cu(II)/Cu(I) or Cu(III)/Cu(I) redoxycles, the possibility of developing systems which may effect theatalytic 2e–/2H+ O2-reduction to hydrogen peroxide seems real.

The electrocatalytic four-electron reduction of O2 has also beenchieved using electrodes on which Cu complexes are immobi-

ized [86–96]. Whether a mononuclear Cu complex or a dinuclearu complex is responsible for the four-electron reduction of2 was examined by determining the dependence of the raten the Cu coverage. Anson and coworkers determined that the

Scheme 1

.

electrocatalytic rate of the O2-reduction was first-order in Cu cov-erage, suggestive of a mononuclear Cu site as the active catalyst[86,87]. On the other hand, the rate of O2 reduction with a CuI com-plex of 3-ethynyl-phenanthroline covalently immobilized onto anazide-modified glassy carbon surface is second-order in the Cu cov-erage at moderate overpotential, suggesting that two CuI speciesare necessary for the efficient four-electron four-proton reductionof O2 [95,96]. In this case, a dinuclear peroxo-Cu complex is pro-posed to be a key intermediate for the four-electron reduction ofO2, in a reaction cycle which is similar to that shown in Scheme 6.In the context of the present review, it should be noted that forrecent cases of electrocatalytic systems, and depending on condi-tions of applied potential or pH, increases in the relative amount oftwo-electron O2-reduction to H2O2 may be observed [96,97].

In the context of the chemical reduction of molecular oxygenand production of hydrogen peroxide which may come about withcopper complexes, there is in fact a pertinent literature concerningthe (di)copper complex mediated catalytic oxidation of catechols toquinones [98–100]. This research is bioinspired as enzyme catecholoxidases are known to effect this reaction at an active site compris-ing two adjacent (0.3–0.4 nm) copper ions. The stoichiometry of

reaction for the enzyme is thought to be:

2Ar-(OH)2 + O2 → 2Ar-( O)2 + 2H2O.

0.

506 S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511

Fig. 12. Cobalt porphyrins employed for electrocatalytic reduction of O2 to H2O2 [24].

F (– · –)3 erformt

ai3p

D

ivtricpiaiwscs

s

cssdp

ig. 13. (a) Electrocatalytic current at oxygen reduction. ([Co(DPP)] (– –), [Co(OEP)] × 10−3 mol L−1 H2O2 (– –), with oxygen bubbling: (—)). The measurements were phe coordinating ion or ligands, ultrapure water was used in the experiments.

However, for a number of chemical model system studiesimed at exploring mechanistic details and typically employ-ng 3,5-di-tert-butylcatechol (DTBC) as a convenient substrate;,5-di-tert-butyl-1,2-benzoquinone (DTBQ) along with hydrogeneroxide are produced stoichiometrically, according to

TBC + O2 → DTBQ + H2O2.

Mechanisms proposed by several groups [101–106] involve thenteraction of DTBC at a one ligand-copper(II) site, whereuponalence tautomerism to give a Cu(I)-semiquinone species leadso O2-reactivity giving a copper(II)-superoxo species which in aate limiting step abstracts a substrate hydrogen atom, eliminat-ng DTBQ and hydrogen peroxide while regenerating the catalyticopper(II) containing complex. Another possible mechanism wouldroceed via acid–base chemistry where a Cu(II)-peroxo-Cu(II)

ntermediate reacts with a catechol, releasing H2O2 and leaving bound catecholate dicopper(II) center; the latter would undergonternal redox to give quinone product and a dicopper(I) moiety

hich reacts with dioxygen to repeat the cycle [99,104,107]. Withuch known chemistries, it may well be that investigation of copper

omplex electrocatalytic or chemically selective reduction to H2O2hould be pursued.1

In general, copper complexes, as mentioned here, haveeen increasing use and success in O2-reduction electrocatalytic

1 In a recent study we have shown that a different kind of binuclear copperomplex, that known to form a differing peroxodicopper(II) end-on peroxo-tructure and a �-1,1-hydroperoxodicopper(II) complex, selectively effects theolution two-electron two-proton reduction of dioxygen to hydrogen peroxide usingecamethylferrocene and trifluoroacetic acid, rather than the four-electron four-roton reduction to water seen with other copper complexes (see [132]).

, [Co(TPP)] (- - -), [Co(TCPP)] (—)). (b) CV of [Co(TCPP)] at scan rate = 20 mV s−1 (withed in 1.0 × 10−1 mol L−1 hydrosulphuric acid at 298 K. To prevent contamination of

applications [73,94–97,108]. But, also as additives or otherenhancements of various types, copper in various forms hasbeen shown to boost O2-reduction electrocatalytic behavior. Somerecent examples include (i) a system in which a multicopper oxi-dase is linked covalently to a multiwall carbon nanotube [109],(ii) a fuel cell which utilizes a Cu-Cu2O redox cycle [110], (iii)A Cu(II) grafted TiO2 catalyst, wherein reduced copper(I) formedfrom a photolytic process facilitates O2-reduction [111], (iv) cop-per nanoclusters deposited onto a glassy-carbon electrode showedvery favorable O2 reductive electrocatalytic activity [112] and (v)a modified platinum electrode, where sub-monolayer quantitiesof copper were incorporated into Pt(111) resulted in an eight-foldenhancement in O2-reduction electrocatalytic activity [113].

3. H2O2 formation by solar cell

In order to utilize H2O2 as an energy carrier, hydrogen peroxideshould be produced by a convenient method without any specialequipment. One such method is electroreduction of O2 with a solarcell in an acidic aqueous solution under air [24]. The solar cell isa convenient power source usable anywhere during the sun-shinydays. H2O2 can be produced anywhere by plugging electrodes withan O2 reduction catalyst into solar cells in an acidic solution. H2O2production by the electrocatalytic reduction of O2 under air withelectrical power generated by a photovoltaic solar cell has been per-formed using Co-porphyrin compounds. This is depicted in Fig. 12for catalysts in an acidic solution at room temperature [24], becausecobalt porphyrins act as efficient selective two-electron reduction

of O2 as described in a previous section [31,32].

Fig. 13 shows the O2 reduction current recorded against volt-age vs. saturated calomel electrode (SCE) using a glassy carbonelectrode modified by the cobalt porphyrins shown in Fig. 12 in

S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511 507

b) Ag

a[[pwfopti

gpiscT1

sspmnia[rtr

4

aauno[

Fig. 14. TEM images of Ag or Ag–Pb alloy nanoparticles. (a) Ag nanoparticles, (

n aqueous solution containing hydrosulphuric acid (1 mol L−1)24]. Dioxygen reduction currents were observed when usingCo(TCPP)], [Co(TPP)] and [Co(OEP)], the details depending on thearticular porphyrin structure. The highest on-set potential of 0.3 Vas observed with [Co(TCPP)] [24]. The thermodynamic potential

or the two-electron reduction of O2 at a given pH is 0.65 V, thus, theverpotential for O2 reduction was 0.35 V for [Co(TCPP)]. The onsetotential observed on [Co(TPP)] was around 0.25 V comparable tohat of [Co(TCPP)]. The onset potential of [Co(OEP)] was 0.1 V, thusndicating a larger overpotential is required for O2 reduction.

A cyclic voltammogram of [Co(TCPP)] in the presence of oxy-en gas and H2O2 (3 × 10−3 mol L−1) (Fig. 13b) exhibits a reductioneak which appears around 0 V in the presence of O2, correspond-

ng to two-electron reduction of O2 to produce H2O2 [24]. Theelective production of H2O2 can be achieved with the [Co(TCPP)]atalyst when the applied potential is controlled to around 0 V.he current efficiency for H2O2 production was nearly 100% in.0 × 10−1 mol L−1 hydrosulphuric acid [24].

H2O2 production by the reduction of O2 in an acidic aqueousolution was also performed using a conventional Si photovoltaicolar cell, which behaves as an electric power source, with an out-ut of 0.5 V at 2.5 mA [24]. A glassy carbon electrode (∼1 cm2)ounted with a Co-porphyrin and Pt wire electrode were con-

ected to the negative and positive electrodes. After 11 h runningn a 1.0 × 10−1 mol L−1 hydrosulphuric acid (∼7 mL), solution, themount of H2O2 produced reached 1.46 × 10−5 mol with [Co(TCPP)]24]. Thus, H2O2 was conveniently produced by the electrocatalyticeduction of O2 with a conventional photovoltaic solar cell althoughhe catalytic behavior as well as cell and electrode structures willequire improvement for practical application.

. Direct H2O2 fuel cells

Fuel cells require oxygen as the oxidant and therefore the use inir-free environments such as outer space and underwater require

compressed oxygen tank. Hydrogen peroxide (H2O2) has been

tilized as an alternative liquid oxidant in place of gaseous O2. Aumber of fuel cells with H2O2 as the oxidant have been devel-ped using borohydride [114–116], metals [117–119], methanol120,121], hydrazine [122] and biofuels [123] as reductants. H2O2

–Pb alloy (Ag:Pb = 9:1), (c) Ag–Pb alloy (7:3) and (d) Ag–Pb alloy (Ag:Pb = 6:4).

can also be used as a reductant and direct H2O2 cells have recentlybeen developed [124–127]. Direct H2O2 fuel cells have number ofmerits as compared with other fuel cells: (1) H2O2 is liquid andsoluble in water and thereby easy to store and carry: (2) H2O2 hashigher standard reduction potential than O2 (e.g., 1.776 V vs. SHEfor H2O2 and 1.229 V vs. SHE for O2 in acidic medium [124]; (3)there is no need to use membranes, because H2O2 can act both oxi-dant and reductant. Thus, fuel cells with H2O2 used as both oxidantand reductant can have much simpler cell structure (i.e., one com-partment cell without membrane), theoretically providing higherpower output than those with oxygen used as an oxidant [124]. Thereactions occurring at anode and cathode of the H2O2 fuel cell aregiven by Eqs. (8)–(10) [24,125–127].

Anode : H2O2 → O2 + 2H+ + 2e– 0.68 V (8)

Cathode : H2O2 + 2H+ + 2e– → 2H2O 1.77 V (9)

Total : 2H2O2 → O2 + 2H2O (10)

Thus, H2O2 fuel cells emit only oxygen after electrical powergeneration. The theoretical maximum of the output potential of theH2O2 fuel cell is 1.09 V [125–127], which is comparable to those of ahydrogen fuel cells (1.23 V) and a direct methanol fuel cell (1.21 V)[127]. A one compartment H2O2 fuel cell has been constructedby using an Au plate as an anode and an Ag plate as a cathode,because these metal plates act as selective oxidation and reductioncatalyst toward H2O2, respectively [126]. Such a one-compartmentstructure without membrane is certainly more promising for devel-opment of low-cost fuel cells as compared with two compartmentfuel cells with membranes. However, the voltage achieved fromthe fuel cell described above was 0.12 V, which is lower than thetheoretically achievable voltage of 1.09 V; this was due to a largeover-potential at the Ag cathode [125].

Increasing the specific surface area is one of the easiest meansto achieve high catalytic activity per weight. In order to increasethe surface area of Ag plate, nanoparticles of Ag with high specific

surface area were examined as a cathode of the one-compartmentH2O2 fuel cell [24]. Also addition of foreign elements of Pb to modifythe electronic structure of Ag nanoparticles was examined [24].Fig. 14 shows the TEM images of Ag and Ag–Pb alloy nanoparticles

508 S. Fukuzumi et al. / Electrochimic

Fig. 15. I–V and I–P curves of a one-compartment H2O2 fuel cell with Ag or Ag–Pballoy cathode (Au anode. 1.0 mol L−1 NaOH, 3.0 × 10−1 mol L−1 H2O2. black: Ag,green: Ag:Pb = 6:4, red: Ag:Pb = 7:3 and blue: Ag:Pb = 9:1). (For interpretation of thert

ua

eHu(fmobnh7ictwc

tgcbin

tion was investigated by an acid titration of [FeIII(Pc)Cl] with

Ff

eferences to color in this figure legend, the reader is referred to the web version ofhe article.)

sed for the construction of one-compartment H2O2 fuel cells. Theverage Ag nanoparticles diameter was 41 nm.

The Ag based nanoparticles were mounted on a glassy carbonlectrode by a drop-casting method [24] and the one-compartment2O2 fuel cells constructed were operated under basic conditionssing an aqueous solution containing NaOH (1.0 mol L−1) and H2O23.0 × 10−1 mol L−1). Fig. 15 shows I–V and I–P curves for such H2O2uel cells where the current density was normalized by the geo-

etric surface area of the glassy carbon electrode [24]. The resultsbtained with Ag nanoparticles without Pb addition are plotted inlack color in Fig. 15. The performance of the H2O2 fuel cell using Aganoparticles was comparable to that using Ag plate. On the otherand, when Ag–Pb alloys were used as cathodes [Ag:Pb = 9:1 (blue),:3 (red), 6:4 (green) in Fig. 15], cell performance was improved,

.e., higher power densities, open circuit voltages and short-circuiturrents were obtained compared to those using the Ag nanopar-icles as the cathode. Large values were achieved on Ag-Pb alloysith the ratios of 9:1 and 7:3 although the open-circuit potential of

a. 150 mV was still far from the theoretical potential (1.09 V) [24].In order to achieve a more efficient system, the medium condi-

ions should be the same or similar for H2O2 production and powereneration. Hydrogen peroxide fuel cells using an Ag-based cathodean generate power only in basic media, however, H2O2 production

y O2 reduction was performed under acidic conditions [24]. H2O2

s not stable under basic conditions, thus, there is a tremendouseed to develop an H2O2 fuel cell which can be operated under

ig. 16. Chemical structures of porphyrin and phthalocyanine iron(III) complexes similaruel cell. (a) [FeIII(Pc)Cl], (b) [FeIII(OEP)Cl] and (c) [FeIII(TPP)Cl].

a Acta 82 (2012) 493– 511

acidic conditions for the ideal combination with the production ofH2O2 via the two-electron reduction of O2 with a solar cell.

A one-compartment H2O2 fuel cell using an iron phthalocyan-inato complex can be operated under acidic conditions [126]. Thechoice of iron porphyrin and analogous compounds for H2O2 reduc-tion is quite reasonable because in natural systems, the reductionof hydrogen peroxide is owing to hydroperoxidases, which con-tain iron(III)-porphyrins in their active sites [128–130]. The activityof iron complexes depicted in Fig. 16 was examined for H2O2reduction [126]. Fig. 17 shows the cyclic voltammograms (CV) ofH2O2 with an iron complex mounted glassy carbon as a workingelectrode in an aqueous solution of acetate buffer (pH 4) containing3.0 × 10−1 mol L−1 H2O2 [126]. The black curves in Fig. 17 are CVsof each iron complex in an acetate buffer solution without H2O2. Acatalytic current of the H2O2 reduction was observed in a cathodicsweep with all the Fe complexes, indicating they act as catalystsfor the H2O2 reduction in acidic media. The onset potentials forH2O2 reduction on electrodes with [FeIII(OEP)Cl], [FeIII(TPP)Cl] and[FeIII(Pc)Cl] were ca. 0.2 V, 0.2 V and 0.5 V (vs. SCE), respectively.Thus, [FeIII(Pc)Cl] showed the smallest overpotential in this acidicsolution, in comparison to the FeIII porphyrin complexes.

A one-compartment H2O2 fuel cell working in an acidicmedia was constructed with a glassy carbon electrode mounting[FeIII(Pc)Cl] as a cathode and Ni metal as an anode [126]. The cellperformance was evaluated by dipping the anode and cathode in abuffer solution containing 3.0 × 10−1 mol L−1 H2O2. Fig. 18 showsthe cell performances of the H2O2 fuel cells in acidic solutions con-taining 3.0 × 10−1 mol L−1 H2O2 at 293 K [126]. Fig. 18a displays theI–P and I–V curves obtained by the operation of the fuel cell underthe conditions of pH 4 (black) and pH 5 (blue). In both cases, theopen circuit potentials were ca. 0.24 V, which is significantly higherthan that (0.15 V) of the H2O2 fuel cell using a Ag cathode work-ing under basic conditions [24,125]. The maximum power densitywas improved from 0.39 �W cm−2 at pH 5 to 0.44 �W cm−2 at pH4 by decreasing the pH. The I–V and I–P curves obtained for theH2O2 fuel cell operated at pH 3 were much higher in both opencircuit potential and power density as displayed in Fig. 18b. Thepower density reached to 10 �W cm−2, which is more than 20 timeshigher than that obtained under pH 4 and pH 5 conditions. Theopen-circuit potential of 0.5 V is more than three times higher thanthat (0.15 V) of the one-compartment H2O2 fuel cell operated underbasic conditions [24,125]. A further decrease in pH resulted in thedecomposition of H2O2 on the surface of the Ni anode. Thus, theconditions of pH 3 are the optimum for operating the H2O2 fuel cellusing [FeIII(Pc)Cl] as the cathode.

The chemical structure of [FeIII(Pc)Cl] under acidic condi-

trifluoroacetic acid (TFA) using UV–vis spectroscopy as shownin Fig. 19 [126]. To the benzonitrile solution of [FeIII(Pc)Cl](4.0 × 10−5 mol L−1), a known amount of TFA (2–16 × 10−3 mol L−1)

to active site structures of hydroperoxidases as candidates of cathodes for an H2O2

S. Fukuzumi et al. / Electrochimica Acta 82 (2012) 493– 511 509

Fig. 17. Cyclic voltammograms of H2O2 on glassy carbon electrodes modified with FeIII complexes. (a) [FeIII(Pc)Cl], (b) [FeIII(OEP)Cl] and (c) [FeIII(TPP)Cl]. The measurementswere performed in an acetate buffer solution (pH 4) in the presence and absence of 3.0 × 10−3 mol L−1 H2O2.

Fig. 18. I–V and I–P curves of a one-compartment H2O2 fuel cell with Ni anode and [FeIII(Pc)Cl] cathode. Performance tests were conducted in an acetate buffer containing3.0 × 10−1 mol L−1 H2O2. The pH of the solutions was fixed to 5 (a, blue), 4 (a, black) or 3 (b, red). Currents and powers were normalized by a geometric surface area ofelectrode. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

Fig. 19. (a) UV–vis absorption change of a benzonitrile solution of [FeIII(Pc)Cl] (4.0 × 10−5 mol L−1) by adding trifluoroacetic acid (2–16 × 10−3 mol L−1). (b) The Hill plotobtained from monitoring of the absorption changes at 520 nm.

510 S. Fukuzumi et al. / Electrochimic

Fig. 20. Potential changes by repetitive measurements of H2O2 fuel cells with theNafion® coated [FeIII(Pc)Cl] cathode (red) and without Nafion® coating (black).P −2

it

watatucaiK[po

o[if(o[tiNgwftificp

5

pic

otentials required to achieve the power density of 20 �A cm were recorded. (Fornterpretation of the references to color in this figure legend, the reader is referredo the web version of the article.)

as added and UV–vis spectral changes monitored. By the acidddition TFA, an intense absorption band around 656 nm assignedo [FeIII(Pc)Cl] became weaker and new absorption bands appearedround 520 and 720 nm (Fig. 19a), which were assigned to the pro-onated iron phthalocyanate complex [131]. The slope of Hill plot bysing absorption change at 520 nm shown in Fig. 19b was 1.2, indi-ating that one proton is associated with in the change of UV–visbsorbance. The change was owing to the protonation of one ofmine group of the phthalocyanine ligand. The protonation constant

was determined from the titration in Fig. 19 to be 160 mol−1 L126]. The protonation to the phthalocyanine ligand provided aositive shift of a redox potential of Fe(II)/Fe(III) and stabilizationf the reduced state of Fe(II) then suitable for H2O2 reduction.

The protonation of [FeIII(Pc)Cl] improves the catalytic activityf [FeIII(Pc)Cl] for H2O2 reduction, however, the cationic nature ofFeIII(PcH)Cl]+ decreases the stability because its solubility in waterncreases [126]. As a result, a serious deterioration in the cell per-ormance was observed during the course of repetitive tests. Fig. 20black) shows the resulting potential decrease of the power densityf 20 �A cm−2 (not open circuit potential) by the repetitive tests126]. A high potential of 0.28 V observed in the first test decreasedo ca. 0.1 V after repetition. The robustness of [FeIII(PcH)Cl]+ wasmproved by coating of the complex with a protic membrane,afion®, which is a highly acidic polymer containing sulphonateroups at the end of side chains [126]. A glassy carbon electrodeas coated with Nafion® after mounting [FeIII(Pc)Cl], and an H2O2

uel cell was constructed with the electrode. Subsequent repetitiveests on this H2O2 fuel cell were performed at pH 3. As indicatedn Fig. 20 (red), the potential of more than 0.25 V observed in therst cycle was maintained in the repetitive tests [126]. Thus, theoating with Nafion® effectively improved the stability of the cellerformance.

. Conclusions

In this review, it has been shown that a variety of metal com-lexes can catalyze reduction of O2 by one-electron reductants

n the presence of an acid in homogeneous solution. Electro-atalytic reduction of O2 can also be achieved using electrodes

a Acta 82 (2012) 493– 511

modified with metal complexes. The number of electrons trans-ferred during O2 reduction is two or four depending of type ofmetals and ligands. In the case of biscobalt porphyrins and bis-cobalt porphyrin–corrole complexes, the Co–Co distance is crucialto attain the four-electron reduction of O2, because the formationof the dinuclear Co(III) �2-�1:�1-peroxo complex with suitableCo–Co distance and the cleavage of the O O bond are requiredfor the four-electron reduction of O2. In the case of cytochromec oxidase model compounds, however, a Fe porphyrin withouta Cu unit can catalyze the four-electron reduction of O2. In thiscase, an Fe(III)-hydroperoxo complex can be further reduced byone-electron reductants in the presence of an acid to produce water.Electrocatalytic four-electron reduction of O2 can also be attainedusing electrodes modified with cytochrome c oxidase model com-plexes. Cu complexes alone can also catalyze the four-electronreduction of O2 by one-electron reductants in the presence of anacid in homogeneous and heterogeneous systems. Hydrogen per-oxide produced by electrocatalytic reduction of O2 using electrodesmodified with metal complexes acting as catalysts for selectivetwo-electron reduction of O2 can be used as a fuel in a hydrogenperoxide fuel cell. A hydrogen peroxide fuel cell has great advan-tages in comparison with other fuel cells which require O2 as theoxidant, because it can be operated using a one-compartment cellwithout a membrane and in air-free environments such as in outerspace and underwater. Future scrutiny is desired to improve thecatalytic activity for the selective two-electron reduction of O2 toH2O2, the latter being promising candidates as renewable and cleanenergy sources.

Acknowledgments

The authors gratefully acknowledge the contributions of theircollaborators and coworkers mentioned in the cited references,and support by a Grant-in-Aid (Nos. 20108010 and 237500141),a Global COE program, “the Global Education and Research Centerfor Bio-Environmental Chemistry” from the Ministry of Education,Culture, Sports, Science and Technology, Japan and KOSEF/MESTthrough WCU project (R31-2008-000-10010-0), Korea. K.D.K. alsothanks the USA National Institutes of Health for support.

References

[1] J.H. Seinfeld, AIChE J. 57 (2011) 3259.[2] D.L. Royer, R.A. Berner, J. Park, Nature 446 (2007) 530.[3] N.S. Lewis, D.G. Nocera, Proc. Natl. Acad. Sci. U.S.A. 103 (2006) 15729.[4] D.G. Nocera, Chem. Soc. Rev. 38 (2009) 13.[5] H.B. Gray, Nat. Chem. 1 (2009) 7.[6] F. Jiao, H. Frei, Energy Environ. Sci. 3 (2010) 1018.[7] V. Balzani, A. Credi, M. Venturi, ChemSusChem 1 (2008) 26.[8] S. Fukuzumi, Eur. J. Inorg. Chem. (2008) 1351.[9] H. Dau, I. Zaharieva, Acc. Chem. Res. 42 (2009) 1861.

[10] D. Gust, T.A. Moore, A.L. Moore, Acc. Chem. Res. 42 (2009) 1890.[11] S. Fukuzumi, Y. Yamada, T. Suenobu, K. Ohkubo, H. Kotani, Energy Environ.

Sci. 4 (2011) 2754.[12] M. Momirlan, T.N. Veziroglub, Int. J. Hydrogen Energy 30 (2005) 795.[13] S. Dunn, Encyclopedia of Energy, vol. 3, Elsevier Inc., 2004, p. 241.[14] S. Fukuzumi, T. Kobayashi, T. Suenobu, ChemSusChem 1 (2008) 827.[15] A. Baldi, B. Dam, J. Mater. Chem. 21 (2011) 4021.[16] L. Schlapbach, A. Züttel, Nature 414 (2001) 353.[17] J.L.C. Rowsell, O.M. Yaghi, Angew. Chem. Int. Ed. 44 (2005) 4670.[18] M.E. Bluhm, M.G. Bradley, R. Butterick III, U. Kusari, L.G. Sneddon, J. Am. Chem.

Soc. 128 (2006) 7748.[19] R. Orinakova, A. Orinak, Fuel 90 (2011) 3123.[20] I. Yamanaka, T. Murayama, Angew. Chem. Int. Ed. 47 (2008) 1900.[21] R.S. Disselkamp, Int. J. Hydrogen Energy 35 (2010) 1049.[22] R.S. Disselkamp, Energy Fuels 22 (2008) 277.[23] A.E. Sanli, A. Aytac, Int. J. Hydrogen Energy 36 (2011) 869.

[24] Y. Yamada, Y. Fukunishi, S. Yamazaki, S. Fukuzumi, Chem. Commun. 46 (2010)

7334.[25] T. Nishimi, T. Kamachi, K. Kato, T. Kato, K. Yoshizawa, Eur. J. Org. Chem. (2011)

4113.[26] S. Fukuzumi, K. Ishikawa, T. Tanaka, Chem. Lett. (1986) 1.

himic

S. Fukuzumi et al. / Electroc

[27] S. Fukuzumi, M. Chiba, M. Ishikawa, K. Ishikawa, T. Tanaka, J. Chem. Soc.,Perkin Trans. 2 (1989) 1417.

[28] D.T. Sawyer, T.S. Calderwood, K. Yamaguchi, C.T. Angelis, Inorg. Chem. 22(1983) 2577.

[29] A.M. Sargeson, P.A. Lay, Aust. J. Chem. 62 (2009) 1280.[30] I.I. Creaser, R.J. Geue, J.MacB. Harrowfield, A.J. Herlt, A.M. Sargeson, M.R. Snow,

J. Springborg, J. Am. Chem. Soc. 104 (1982) 6016.[31] S. Fukuzumi, S. Mochizuki, T. Tanaka, Inorg. Chem. 28 (1989) 2459.[32] S. Fukuzumi, K. Okamoto, C.P. Gros, R. Guilard, J. Am. Chem. Soc. 126 (2004)

10441.[33] R.A. Marcus, Ann. Rev. Phys. Chem. 15 (1964) 155.[34] E.S. Yang, M.-S. Chan, A.C. Wahl, J. Phys. Chem. 84 (1988) 3094.[35] D.F. Rohrbach, E. Deutsch, W.R. Heineman, R.F. Pasternack, Inorg. Chem. 16

(1977) 2650.[36] R.F. Pasternack, E.G. Spiro, J. Am. Chem. Soc. 100 (1978) 968.[37] R. Langley, P. Hambright, Inorg. Chem. 24 (1985) 1267.[38] J.P. Collman, J.E. Hutchison, M.A. Lopez, A. Tabard, R. Guilard, W.K. Seok, J.A.

Ibers, M. L’Her, J. Am. Chem. Soc. 114 (1992) 9869.[39] F. Bolze, C.P. Gros, M. Drouin, E. Espinosa, P.D. Harvey, R. Guilard, J. Organomet.

Chem. 643–644 (2002) 89.[40] C.J. Chang, Y. Deng, C. Shi, C.K. Chang, F.C. Anson, D.G. Nocera, Chem. Commun.

(2000) 1355.[41] C.J. Chang, E.A. Baker, B.J. Pistorio, Y. Deng, Z.-H. Loh, S.E. Miller, S.D. Carpenter,

D.G. Nocera, Inorg. Chem. 41 (2002) 3102.[42] S. Schmidt, F.W. Heinemann, A. Grohmann, Eur. J. Inorg. Chem. (2000) 1657.[43] X. Dai, P. Kapoor, T.H. Warren, J. Am. Chem. Soc. 126 (2004) 4798.[44] S. Hikichi, M. Yoshizawa, Y. Sasakura, H. Komatsuzaki, Y. Moro-oka, M. Akita,

Chem. Eur. J. 7 (2001) 5011.[45] A.L. Gavrilova, C.J. Qin, R.D. Sommer, A.L. Rheingold, B. Bosnich, J. Am. Chem.

Soc. 124 (2002) 1714.[46] C.J. Chang, Z.-H. Loh, C. Shi, F.C. Anson, D.G. Nocera, J. Am. Chem. Soc. 126

(2004) 10013.[47] K.M. Kadish, L. Frémond, J. Shen, P. Chen, K. Ohkubo, S. Fukuzumi, M.E. Ojaimi,

C.P. Gros, J.-M. Barbe, R. Guilard, Inorg. Chem. 48 (2009) 2571.[48] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and Appli-

cations, 2nd ed., John Wiley & Sons, Inc., New York, 2001.[49] V.G. Levich, Physicochemical Hydrodynamics, Prentice-Hall, Inc., Englewood

Cliffs, N. J., 1962.[50] J. Koutecky, V.G. Levich, Zh. Fiz. Khim. 32 (1958) 1565.[51] N. Oyama, F.C. Anson, Anal. Chem. 52 (1980) 1192.[52] K.M. Kadish, J. Shen, L. Frémond, P. Chen, M.E. Ojaimi, M. Chkounda, C.P. Gros,

J.-M. Barbe, K. Ohkubo, S. Fukuzumi, R. Guilard, Inorg. Chem. 47 (2008) 6726.[53] K.M. Kadish, L. Frémond, Z. Ou, J. Shao, C. Shi, F.C. Anson, F. Burdet, C.P. Gros,

J.-M. Barbe, R. Guilard, J. Am. Chem. Soc. 127 (2005) 5625.[54] S. Ferguson-Miller, G.T. Babcock, Chem. Rev. 96 (1996) 2889.[55] M.M. Pereira, M. Santana, M. Teixeira, Biochim. Biophys. Acta 1505 (2001)

185.[56] E. Kim, E.E. Chufán, K. Kamaraj, K.D. Karlin, Chem. Rev. 104 (2004) 1077.[57] E.E. Chufán, S.C. Puiu, K.D. Karlin, Acc. Chem. Res. 40 (2007) 563.[58] T. Chishiro, Y. Shimazaki, F. Tani, Y. Tachi, Y. Naruta, S. Karasawa, S. Hayami,

Y. Maeda, Angew. Chem. Int. Ed. 42 (2003) 2788.[59] J.P. Collman, R. Boulatov, C.J. Sunderlan, L. Fu, Chem. Rev. 104 (2004) 561.[60] J.P. Collman, R. Boulatov, C.J. Sunderland, in: M. Kadish, K.M. Smith, R. Guilard

(Eds.), The Porphyrin Handbook, vol. 11, Academic Press, San Diego, CA, 2003,p. 1.

[61] J.P. Collman, S. Ghosh, A. Dey, R.A. Decreau, Y. Yang, J. Am. Chem. Soc. 131(2009) 5034.

[62] F.C. Anson, A. Shi, B. Steiger, Acc. Chem. Res. 30 (1997) 437.[63] C.H. Kjaergaard, J. Rossmeisl, J.K. Nørskov, Inorg. Chem. 49 (2010) 3567.[64] Z. Halime, H. Kotani, S. Fukuzumi, K.D. Karlin, Proc. Natl. Acad. Sci. U.S.A. 108

(2011) 13990.[65] R.A. Ghiladi, H.W. Huang, P. Moënne-Loccoz, J. Stasser, N.J. Blackburn, A.S.

Woods, R.J. Cotter, C.D. Incarvito, A.L. Rheingold, K.D. Karlin, J. Biol. Inorg.Chem. 10 (2005) 63.

[66] H. Shin, D.-H. Lee, C. Kang, K.D. Karlin, Electrochim. Acta 48 (2003) 4077.[67] A. Messerschmidt, Adv. Inorg. Chem. 40 (1993) 121.[68] K.Y. Djoko, L.X. Chong, A.G. Wedd, Z. Xiao, J. Am. Chem. Soc. 132 (2010) 2005.[69] D. Kosman, J. Biol. Inorg. Chem. 15 (2010) 15.[70] E.I. Solomon, J.W. Ginsbach, D.E. Heppner, M.T. Kieber-Emmons, C.H. Kjaer-

gaard, P.J. Smeets, L. Tian, J.S. Woertink, Faraday Discuss. 148 (2011) 11.[71] J.A. Cracknell, K.A. Vincent, F.A. Armstrong, Chem. Rev. 108 (2008) 2439.[72] E.I. Solomon, U.M. Sundaram, T.E. Machonkin, Chem. Rev. 96 (1996) 2563.[73] A.A. Gewirth, M.S. Thorum, Inorg. Chem. 49 (2010) 3557.[74] N. Fujieda, A. Yakiyama, S. Itoh, Dalton Trans. 39 (2010) 3083.[75] L.M. Mirica, X. Ottenwaelder, T.D.P. Stack, Chem. Rev. 104 (2004) 1013.[76] E.A. Lewis, W.B. Tolman, Chem. Rev. 104 (2004) 1047.[77] L.Q. Hatcher, K.D. Karlin, J. Biol. Inorg. Chem. 9 (2004) 669.[78] L.Q. Hatcher, K.D. Karlin, Adv. Inorg. Chem. 58 (2006) 131.[79] S. Fukuzumi, H. Kotani, H.R. Lucas, K. Doi, T. Suenobu, R.L. Peterson, K.D. Karlin,

J. Am. Chem. Soc. 132 (2010) 6874.[80] L. Tahsini, H. Kotani, Y.-M. Lee, J. Cho, W. Nam, K.D. Karlin, S. Fukuzumi, Chem.

Eur. J. 18 (2012) 1084.[81] K.D. Karlin, M.S. Haka, R.W. Cruse, Y. Gultneh, J. Am. Chem. Soc. 107 (1985)

5828.

a Acta 82 (2012) 493– 511 511

[82] K.D. Karlin, Z. Tyeklár, A. Farooq, M.S. Haka, P. Ghosh, R.W. Cruse, Y. Gultneh,J.C. Hayes, P.J. Toscano, J. Zubieta, Inorg. Chem. 31 (1992) 1436;H.-C. Liang, K.D. Karlin, R. Dyson, S. Kaderli, B. Jung, A.D. Zuberbühler, Inorg.Chem. 39 (2000) 5884.

[83] E. Pidcock, H.V. Obias, M. Abe, H.-C. Liang, K.D. Karlin, E.I. Solomon, J. Am.Chem. Soc. 121 (1999) 1299.

[84] M.J. Henson, P. Mukherjee, D.E. Root, T.D.P. Stack, E.I. Solomon, J. Am. Chem.Soc. 121 (1999) 10332.

[85] L. Que Jr., W.B. Tolman, Angew. Chem. Int. Ed. 41 (2002) 1114.[86] J. Zhang, F.C. Anson, Electrochim. Acta 38 (1993) 2423.[87] Y. Lei, F.C. Anson, Inorg. Chem. 33 (1994) 5003.[88] A.L.B. Marques, J. Zhang, A.B.P. Lever, W.J. Pietro, J. Electroanal. Chem. 392

(1995) 43.[89] J. Losada, I. del Peso, L. Beyer, Inorg. Chim. Acta 321 (2001) 107.[90] V.L.N. Dias, E.N. Fernandes, L.M.S. da Silva, E.P. Marques, J. Zhang, A.L.B. Mar-

ques, J. Power Sources 142 (2005) 10.[91] Y.C. Weng, F.-R. Fan, A.J. Bard, J. Am. Chem. Soc. 127 (2005) 17576.[92] M. Wang, X. Xu, J. Gao, N. Jia, Y. Cheng, Russ. J. Electrochem. 42 (2006) 878.[93] C. Pichon, P. Mialane, A. Dolbecq, J. Marrot, E. Riviere, B. Keita, L. Nadjo, F.

Secheresse, Inorg. Chem. 46 (2007) 5292.[94] M.S. Thorum, J. Yadav, A.A. Gewirth, Angew. Chem. Int. Ed. 48 (2009) 165.[95] C.C.L. McCrory, X. Ottenwaelder, T.D.P. Stack, C.E.D. Chidsey, J. Phys. Chem. A

111 (2007) 12641.[96] C.C.L. McCrory, A. Devadoss, X. Ottenwaelder, R.D. Lowe, T.D.P. Stack, C.E.D.

Chidsey, J. Am. Chem. Soc. 133 (2011) 3696.[97] M.A. Thorseth, C.S. Letko, T.B. Rauchfuss, A.A. Gewirth, Inorg. Chem. 50 (2011)

6158.[98] K. Selmeczi, M. Reglier, M. Giorgi, G. Speier, Coord. Chem. Rev. 245 (2003)

191.[99] I.A. Koval, P. Gamez, C. Belle, K. Selmeczi, J. Reedijk, Chem. Soc. Rev. 35 (2006)

814.[100] I.A. Koval, K. Selmeczi, C. Belle, C. Philouze, E. Saint-Aman, I. Gautier-Luneau,

A.M. Schuitema, M. van Vliet, P. Gamez, O. Roubeau, M. Lüken, B. Krebs, M.Lutz, A.L. Spek, J.-L. Pierre, J. Reedijk, Chem.-Eur. J. 12 (2006) 6138.

[101] M. Kodera, T. Kawata, K. Kano, Y. Tachi, S. Itoh, S. Kojo, Bull. Chem. Soc. Jpn.76 (2003) 1957;J. Ackermann, F. Meyer, E. Kaifer, H. Pritzkow, Chem. Eur. J. 8 (2002) 247.

[102] A. Neves, L.M. Rossi, A.J. Bortoluzzi, B. Szpoganicz, C. Wiezbicki, E. Schwingel,W. Haase, S. Ostrovsky, Inorg. Chem. 41 (2002) 1788.

[103] J. Balla, T. Kiss, R.F. Jameson, Inorg. Chem. 31 (1992) 58.[104] J.-P. Chyn, F.L. Urbach, Inorg. Chim. Acta 189 (1991) 157.[105] J. Kaizer, T. Csay, G. Speier, M. Giorgi, J. Mol. Catal. A-Chem. 329 (2010) 71.[106] A. Kupán, J. Kaizer, G. Speier, M. Giorgi, M. Réglier, F. Pollreisz, J. Inorg.

Biochem. 103 (2009) 389.[107] K. Born, P. Comba, A. Daubinet, A. Fuchs, H. Wadepohl, J. Biol. Inorg. Chem. 12

(2007) 36.[108] F.R. Brushett, M.S. Thorum, N.S. Lioutas, M.S. Naughton, C. Tornow, H.-R. Jhong,

A.A. Gewirth, P.J.A. Kenis, J. Am. Chem. Soc. 132 (2010) 12185.[109] R.P. Ramasamy, H.R. Luckarift, D.M. Ivnitski, P.B. Atanassov, G.R. Johnson,

Chem. Commun. 46 (2010) 6045.[110] Y. Wang, H. Zhou, Chem. Commun. 46 (2010) 6305.[111] H. Yu, H. Irie, K. Hashimoto, J. Am. Chem. Soc. 132 (2010) 6898.[112] W. Wei, Y. Lu, W. Chen, S. Chen, J. Am. Chem. Soc. 133 (2011) 2060.[113] I.E.L. Stephens, A.S. Bondarenko, F.J. Perez-Alonso, F. Calle-Vallejo, L. Bech,

T.P. Johansson, A.K. Jepsen, R. Frydendal, B.P. Knudsen, J. Rossmeisl, I. Chork-endorff, J. Am. Chem. Soc. 133 (2011) 5485.

[114] L. Yi, Y. Song, X. Liu, X. Wang, G. Zou, P. He, W. Yi, Int. J. Hydrogen Energy 36(2011) 15775.

[115] D. Cao, Y. Gao, G. Wang, R. Miao, Y. Liu, Int. J. Hydrogen Energy 35 (2010) 807.[116] H. Wu, C. Wang, Z. Liu, Z. Mao, Int. J. Hydrogen Energy 35 (2010) 2648.[117] J.P. Charles, R.B. Russell, K.K. Yong, R.S. J. Electrochem. Soc. 155 (2008) B558.[118] O. Hasvold, N.J. Storkersen, S. Forseth, T. Lian, J. Power Sources 162 (2006)

935.[119] W. Yang, S. Yang, W. Sun, G. Sun, Q. Xin, Electrochim. Acta 52 (2006) 9.[120] T. Bewer, T. Beckmann, H. Dohle, J. Mergel, D. Stolten, J. Power Sources 125

(2004) 1.[121] D.N. Prater, J.J. Rusek, Appl. Energy 74 (2003) 135.[122] S.J. Lao, H.Y. Qin, L.Q. Ye, B.H. Liu, Z.P. Li, J. Power Sources 195 (2010) 4135.[123] B. Tartakovsky, S.R. Guiot, Biotechnol. Prog. 22 (2006) 241.[124] X. Jing, D. Cao, Y. Liu, G. Wang, J. Yin, Q. Wen, Y. Gao, J. Electroanal. Chem. 658

(2011) 46.[125] S. Yamazaki, Z. Siroma, H. Senoh, T. Ioroi, N. Fujiwara, K. Yasuda, J. Power

Sources 178 (2008) 20.[126] Y. Yamada, S. Yoshida, T. Honda, S. Fukuzumi, Energy Environ. Sci. 4 (2011)

2822.[127] A.E. Sanli, A. Aytac , Int. J. Hydrogen Energy 36 (2011) 869.[128] T. Ruzgas, L. Gorton, J. Emnéus, G. Marko-Varga, J. Electroanal. Chem. 391

(1995) 41.[129] C. Obinger, Arch. Biochem. Biophys. 500 (2010) 1.

[130] P. Vidossich, M. Alfonso-Prieto, X. Carpena, I. Fita, P.C. Loewen, C. Rovira, Arch.

Biochem. Biophys. 500 (2010) 37.[131] M.J. Lin, X. Fang, M.B. Xu, J.D. Wang, Spectrochim. Acta Part A 71 (2008) 1188.[132] S. Fukuzumi, L. Tahsini, Y.-M. Lee, K. Ohkubo, W. Nam, K.D. Karlin, J. Am. Chem.

Soc. (2012), http://dx.doi.org/10.1021/ja211656g, in press.