Solar Fuels: Approaches to Catalytic Hydrogen Evolution

8.15

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

JL Dempsey, JR Winkler, and HB Gray, California Institute of Technology, Pasadena, CA, USA ã 2013 Elsevier Ltd. All rights reserved.

8.15.1 Solar Fuels 8.15.1.1 Background 8.15.1.2 Devices for Artificial Photosynthesis 8.15.2 Catalytic H2 Production 8.15.2.1 Electrocatalysis 8.15.2.2 Biomimetic Systems 8.15.2.3 Molecular Hydrogen-Evolving Complexes 8.15.2.4 Cobaloximes 8.15.2.4.1 Electrochemical systems 8.15.2.4.2 Photochemical systems 8.15.2.4.3 Reaction pathways 8.15.2.4.4 Mechanistic analysis 8.15.3 Conclusion Acknowledgments References

8.15.1 8.15.1.1

Solar Fuels Background

Developing renewable energy resources is widely viewed as one of the most important challenges that must be addressed by this generation. With both political and environmental motivations to advance alternative sources of energy, a great deal of research is focused on developing renewable resources capable of meeting projected energy demands.1,2 Solar energy conversion is an area of enormous promise, yet current technologies convert the sun’s energy to electricity which is not efficiently stored by conventional methods. Using sunlight to make ‘solar fuels’ such as H2 (from H2O splitting, Figure 1) and CH3OH (by reducing CO2) is one of the holy grails of twenty-first century chemistry, a highly desirable but challenging process.3–5 A potential of 4.92 eV is stored when two H2O molecules are split into 2H2 þ O2, but the reaction requires two separate multielectron redox processes, a 4-electron oxidation and a two-electron reduction.6 Further, sequential electron or hole transfers directly to H2O produce extremely high-energy intermediates.7,8 Often, though, these intermediates can be stabilized via coordination to metal complexes, thereby lowering the barriers for solar fuel generation.9

8.15.1.2

Devices for Artificial Photosynthesis

In green plants, water splitting reactions are carried out by Photosystem II, and the reducing equivalents produced are ultimately stored in energy-rich sugars. Many solar-driven water splitting devices aiming to mimic nature’s ability to use sunlight to convert abundant energy-poor molecules to energyrich molecules have been proposed.1,3 Notably, an integrated, monolithic, photoelectrochemical photovoltaic device that operates with a hydrogen production efficiency of 12.4% was reported by Turner in 1998.10 While this device demonstrates that water splitting can be carried out in synthetic devices at efficiencies far greater than those achieved in nature, it is based

Comprehensive Inorganic Chemistry II

553 553 553 554 554 554 555 557 557 559 560 561 564 564 564

on a GaInP2/GaAs tandem-junction electrode and uses a platinum catalyst, materials far too rare to be considered in a viable renewable energy device. Inspired by the great success of Turner and the urgent need to address the global energy crisis, significant efforts are being put forth by researchers around the world with the goal of developing an efficient and economical solar-driven water splitting device based on earth-abundant materials (Figure 2). Several device designs have been proposed, and most involve common key elements: materials or molecules that absorb sunlight and separate holes and electrons across large distances coupled with robust and efficient catalysts for the oxidation of water to dioxygen and for reduction of protons to dihydrogen. One noteworthy concept is a device based on a dual array of microstructured semiconductors embedded in a membrane with both electron- and proton-conducting capabilities.3 Upon absorption of blue light, the photoanode will interact with catalysts on its surface that lower barriers for water oxidation. The remaining red light will be absorbed by the photocathode material, where catalysts decorating the semiconductor surface facilitate reduction of protons to dihydrogen. The microstructured rod-like architecture is central to the economics of this device. Planar crystalline Si is the basis for most modern photovoltaic devices and a promising photocathode material for this proposed water splitting device, but extremely high purity (and thus expensive) materials are necessary to obtain minority-carrier diffusion lengths of similar magnitude to the optical absorption depth ( 200 mm) needed to excite silicon’s indirect band gap.11,12 However, a rod-based geometry allows light absorption and minority-carrier diffusion along orthogonal directions; optical absorption occurs along the length of the rod, while minority carriers are collected at radial junctions.13,14 With short diffusion lengths engendered by radial geometry, silicon purity requirements are relaxed. While significant progress has been made toward the goal of developing individual components of many of the proposed systems, optimization of the materials and catalysts necessary

http://dx.doi.org/10.1016/B978-0-08-097774-4.00806-8

553

554

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

D+ + ½H2O A- + H+

*DA

Catalyst Catalyst

D+ + A+ ½H2O

hydrogen-evolving catalysts, including synthetic advances that have led to new complexes capable of evolving H2 at low overpotentials. Also reviewed are experimental efforts aimed at elucidating mechanisms of H2 evolution. Reports of notable hydrogen evolving catalysts are discussed, including those on extensive chemical, electrochemical, and photochemical studies of cobaloximes that have appeared over the past decade.

D + H+ + ¼O2 A + ½H2

DA + ¼O2 + ½H2

D+ AA-

knr

G⬚

DA + ¼O2 + ½H2

hV hνsun sun

8.15.2 8.15.2.1

kcr

ΔG⬚

DA + ½H2O Reaction coordinate Figure 1 Solar energy conversion is initiated by photoexcitation (hnsun) of DA to *DA, which undergoes excited-state electron transfer to yield a charge-separated state (DþA). Dþ and A drive water oxidation and proton reduction via catalysts. Reproduced from Dempsey, J. L.; Brunschwig, B. S.; Winkler, J. R.; Gray, H. B. Acc. Chem. Res. 2009, 42, 1995–2004.

H+ H+

HH

H+

+

H

+

H+

H

H

H-

+

H+

H+

Figure 2 A proposed solar-driven water splitting device based on microrod-structured semiconductors (that act as light absorbers and charge separators) embedded in a proton- and electron-conducting membrane. Substances anchored to the photocathode and the photoanode catalyze water splitting to hydrogen and oxygen. Reproduced from Gray, H. B. Nat. Chem. 2009, 1, 7.

for efficient solar fuel generation devices is an ongoing effort, as is integration of different device components. Still, a great deal of fundamental chemistry must be developed before a solardriven water splitting device based on earth-abundant materials can be put into operation. This chapter focuses on recent progress in developing and understanding molecular-based

Catalytic H2 Production Electrocatalysis

The best catalyst for hydrogen evolution is undeniably platinum, which operates at the thermodynamic potential for Hþ/H2 conversion (0 V vs. standard hydrogen electrode (SHE), pH 0).15 The relatively low abundance of this element, however, limits its use in scalable devices. In response, attention has turned to highly active [FeFe], [NiFe], and [Fe] hydrogenases that catalyze hydrogen oxidation and reduction reactions at metal-complex active sites embedded in folded polypeptides.16–18 Significant effort has been expended to develop molecular systems that mimic both the structures and the functions of these naturally occurring organometallic active sites (vide infra).5,9,19,20 Although the ultimate objective is to use water as a feedstock,21–27 many potential H2-evolving catalysts are tested for catalytic activity in organic solvents, as they often are not soluble or stable in aqueous environments. Proton sources have markedly different electrocatalytic behavior in organic solvents and aqueous environments. In nonaqueous solvents, the standard potential for hydrogen evolution, E HA=H2 , is directly related to acid strength, and can be described by   2:303 RT E HA=H2 ¼ E Hþ =H2  pKa;HA F where E Hþ =H2 is the solvated proton/dihydrogen standard potential in the given solvent and Ka,HA is the acid dissociation constant for the proton source, determined from a selfconsistent scale of acidity in a given solvent.28–30 The standard potentials for hydrogen evolution from a variety of commonly used acids are given in Table 1. The overpotential for proton reduction is the difference between the potential of catalytic activity and E HA=H2 ; efficient catalysis at potentials more positive than E HA=H2 is not expected. Lastly, it must be noted that the standard potentials for hydrogen evolution in organic solvents are very sensitive to water. Trace water has been shown to influence the relative strengths of acids via selective solvation processes.31 Further, water can act as a proton relay or form hydrogen bonds, and hydronium ions may have easier access to interior sites of catalysts.32

8.15.2.2

Biomimetic Systems

The active site structures of the [NiFe], [FeFe], and [Fe] hydrogenases have inspired a great deal of work on synthetic analogs designed to mimic hydrogen evolution and oxidation functionalities (Figure 3).17–20,35,36 Despite the hundreds of structural models developed to date, few have shown electrocatalytic activity related to proton reduction or hydrogen oxidation. The chemistry of these biomimetic systems has been reviewed elsewhere.19,20

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

Table 1 pKa values and standard potentials for reduction of acids in acetonitrile

+

Acid

pKaa

E HA=H2 (V vs. SCE)b

OC

HBF4 Et2O Trifluoromethanesulfonic acid p-Cyanoanilinium TsOHH2O HClEt2O CF3COOH [Et3NH]þ [Et3NH]þ Acetic acid

0.10c 2.60 7.60 8.00c 8.90c 12.70c 18.70 9.20 (DMF) 13.20 (DMF)

0.23 0.09 0.21 0.23 0.29 0.51 0.86 0.84 1.08

OC

S Fe

S Ni PPh 2

H

Ph2P

C O

Figure 4 The iron–nickel dithiolato hydride, [(CO)3Fe(pdt)(m-H)Ni (dppe)]þ, an active electrocatalyst for hydrogen evolution from trifluoroacetic acid and a structural mimic of the active site of [NiFe] hydrogenase (pdt ¼ 1,3-propanedithiolate; dppe ¼ 1,2-C2H4(PPh2)2).47

+

a

Ref. [28] unless noted otherwise. E HA=H2 ¼ E  Hþ =H2  0.059pKa(HA). A potential of 0.24 V versus SCE was chosen as the thermodynamic potential for E  Hþ =H2 in acetonitrile.34 c Ref. [33].

b 

S S

Ni

S S

S Fe

NC

C N

S

S Ni

Fe

Fe S

OC C N

Figure 5 The iron–nickel complex, [Ni(xbsms)FeCp(CO)]þ, an active electrocatalyst for hydrogen evolution from trifluoroacetic acid in DMF and a structural mimic of the active site of [NiFe] hydrogenase (H2xbsms ¼ 1,2-bis(4-mercapto-3,3-dimethyl-2-thiabutyl)benzene).48

S S

S

C O

CN C O

Figure 3 Line drawings of the active sites in [NiFe] hydrogenase and [FeFe] hydrogenase. S ¼ cysteine residue.

While upwards of 50 structural mimics of [NiFe] hydrogenases have been reported, initially only biomimetic complexes that deviated significantly from the enzyme structure displayed any electrocatalytic behavior, including a trinuclear [Ni0 S20 Fe2] system36–38 and a few Ni–Ru complexes.39–46 More recently, two competent Ni–Fe catalysts were reported that mimic both the structure and function of active [NiFe] hydrogenase.47,48 A nickel–iron dithiolato hydride (Figure 4) from the Rauchfuss lab is an active catalyst for the reduction of trifluoroacetic acid in CH2Cl2, with catalytic currents appearing at 1.37 V versus Fc/ Fcþ.47 Catalytic activity of another nickel–iron dithiolato complex with a chelating S4 ligand (Figure 5) was reported by Fontecave and Artero, with catalysis observed at 1.73 V versus Fc/Fcþ upon the addition of trifluoroacetic acid in dimethylformamide (DMF).48 Many more [FeFe] hydrogenase analogs have been reported.49,50 Rauchfuss and coworkers reported in 2001 the first di-iron dithiolate system, [Fe2[m-S2(CH2)3](CN) (CO)4(PMe3)], which evolved hydrogen catalytically (at 1.13 V vs. Ag/AgCl in CH3CN (Figure 6)).51 Although the efficiency of this system is far below that of the natural hydrogenase, it nevertheless is a noteworthy example of biomimetic function. Following this seminal work, many related systems displaying electrocatalytic properties have been reported (most, however, operate with relatively large overpotentials).19,52 On the bright side, the models have shed new light on the roles of bridging thiolate ligands and the hydride-binding modes associated with catalysis (Figure 6).20

8.15.2.3

Fe C O

NH OC

555

Molecular Hydrogen-Evolving Complexes

Several nonbiomimetic systems show remarkable efficiency for H2 evolution (Figure 7). The most noteworthy examples are

HS SH (Me3P)(OC)2Fe

Fe(CO)2(CN)

Figure 6 The iron–iron complex, [Fe2[m-S2(CH2)3](CN)(CO)4(PMe3)], an active catalyst for proton reduction and a structural mimic of the active site of [FeFe] hydrogenase.51

molecular Co-, Ni-, and Mo-based systems that evolve hydrogen efficiently without the use of noble metals. A mononuclear nickel complex with noncoordinating pendant amines incorporated into the backbone of cyclic diphosphane ligands is highly active for catalytic hydrogen evolution. [Ni(PPh2NPh2)2]2þ catalyzes proton reduction at 0.48 V versus the standard calomel electrode (SCE) with trifluoromethanesulfonic acid or protonated DMF at overpotentials of 0.57 and 0.35 V, respectively.53–55 The reported turnover frequency under optimal conditions, (350 s1) is comparable to those of [NiFe] hydrogenases. Extensive mechanistic studies have indicated that the pendant bases act to relay protons to the metal site, stabilize dihydrogen binding, and lower the barrier for heterolytic formation and cleavage of the dihydrogen bond. This functionality in the second coordination sphere leads to much higher turnover rates and lower overpotentials than complexes lacking these bases.55,56 Further, when the substituent in the para position of the N-aryl group is varied, the reduction potentials of the nickel center are affected, and increased catalytic rates are seen in complexes with more positive Ni(II/I) potentials.32 Notably, similar nickel complexes with 0 or 2 pendant nitrogen bases display activity for hydrogen oxidation, emphasizing the proximity of the operating potential to the thermodynamic potential for H2 evolution.57 Extensive spectroscopic and electrochemical studies of the hydrogen oxidation catalyst [Ni(PNP)2]2þ have led to a

556

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

Ph

N

Ph

n+

2+

Ph

N

N

O

Ph Ph P P H Ni H P P Ph Ph

S S N

Ph

N L

R

CpMo

MoCp S

S

N L N

R⬘

R

N Br N

Me

CH3

Ph N

Ph N

C Ph N P Co P

CH3

2+ N

C

N

N

F

N

F

N C

O B O

N CH3

2+ [Co(PPh2NPh2)(CH3CN)3]

N N

N L N Co N N L

R

MeO

Ph

R

2+

2+

R

2+

[LRCo2]

[(PY4)Co(CH3CN)2]

O

F

O

N

B O

Me

Z Z [Co((DoR)2X)Br2]

[Co((Do)2X)pnBr2]

N L Co N L

X = H, BF2 Y= Z= H H OMe H H CH2COOH Ph H

Y

Y

CH3

Co

Me

N Br N

Me

Me

X = H, BF2

R

2+

C

CH3 N C N

[Co(TimR)L2]n+

(CpMom-S)2S2CH2

O

N Br N

Me

Co

R, R¢ = Me, Ph L = CH3CN (n = 3), Br (n = 1) 2+ [Ni(PPh2NPh2)2]

Me

Co

Co

X

O

O

N Br N

Me

R¢

N

X

N

Mo F

N

N N

R = H, Me L = CH3CN

2+

[(PY5Me2)MoO]

2+

Figure 7 Molecular H2-evolving catalysts.

N

N

Ni

P R2

N

H2

N

H

H

R2 P

2+

R2P

R2 P

N

Ni P R2

R2P

P R2

N

2+

Ni

P R2

-BaseH+

H

R2 P

N

N

P R2

N

P R2

+

R2P

R2P Ni

-e-

R2 P

2+

R2 P

P R2

2+

H R2P Ni

P R2

P R2

P R2

H N

-BaseH+

N

R2 P

H

2+ R2P

Ni P R2

N P R2

-e-

N

R2 P

H Ni

P R2

+

R2P N P R2

Scheme 1 Mechanism of [Ni(PNP)2]2þ-catalyzed hydrogen oxidation.57

mechanism (Scheme 1) with key roles for pendant bases that are also thought to facilitate proton reduction in related systems.57,58 Again, the pendant bases are involved in the heterolytic cleavage of H2 (or heterolytic HdH bond formation), and act as relays, exchanging protons between solution (and external bases) and the metal site. The ligand also is thought to play a role in the coupling of successive proton and electron transfer events in pathways that do not require high-energy intermediates. Calculations on a related [Ni (PCy2NBz2)]2þ system suggest that the P2N4 ligand stabilizes a nickel–dihyrogen complex via the positioning of two nitrogen bases prior to H2 activation, thereby leading to an enhancement in turnover frequency over the related [Ni(PNP)2]2þ (Scheme 2).55 Upon HdH bond cleavage, a Ni(0) complex is formed, and subsequent deprotonation of the ligand is

followed by proton transfer from the second protonated ligand to the metal ion. Subsequent reactivity is expected to be similar to the proposed cycle for [Ni(PNP)2]2þ, while H2 production by [Ni(PPh2NPh2)2]2þ is thought to proceed via the reverse of this cycle.57 A molybdenum complex, (CpMo(m-S))2(S2CH2), is an efficient electrocatalyst, operating at a 50 mV overpotential with (p-cyanoanilinium)BF4 as a proton source (at 0.26 V vs. SCE in acetonitrile).59 Notably, hydrogen evolution is proposed to proceed via reductive elimination from a neutral hydrosulfide complex, (CpMo)2(m-SH)2(S2CH2), formed upon net addition of two electrons and two protons to (CpMo(m-S))2(S2CH2). A great many cobalt complexes with nitrogen-based ligands have been reported to be active hydrogen-evolution catalysts,

557

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

Bz

Bz N P P Cy

N

N Bz

P Cy

Bz

P

Cy

H N

Bz

Cy N

P Ni

P

2+

N

H H Cy Cy

H2 P

Ni

Bz

Bz N

Cy Cy N

Bz

2+

N

N Bz

P

P Cy

2+

N

Bz +Base

Cy P N

Ni

Bz

N H

Cy

N

Ni P

Cy

Bz

Bz

+

Cy Cy P H P

N

-BaseH+

P

P

Cy

Bz

N

Bz

P Cy

Bz

Scheme 2 Stabilization of a nickel–dihyrogen complex and HdH bond cleavage in a [Ni(PCy2NBz2)]2þ system.57

the most notable being those with imine, oxime, and glyoxime ligands. Cobalt diglyoxime complexes (also known as cobaloximes), which are the most remarkable catalysts among these systems, are discussed in Section 8.15.2.4. Among several reported cobalt complexes with [14]-tetraene-N4 ligands are the methyl-substituted analogs, [Co(TimMe)Br2]þ and [Co (TimMe)(CH3CN)2]3þ, which display catalytic peak currents at potentials corresponding to CoII reduction (0.35 and 0.38 V vs. SCE, respectively), with near Faradaic yields in bulk electrolysis experiments.34 Hybrid dioxime–diimine complexes containing one propyl linkage and one oxime linkage, [Co((DO)2X)pnBr2] (X ¼ H, BF2), also demonstrate catalytic activity near their respective CoII/I potentials (0.73 V vs. SCE for the proton-linked complex; 0.46 V for the BF2 linkage) with near Faradaic yields.60 Several semi-cyclic complexes supported by N-aryl Schiff base/oxime ligands with either proton or BF2 linkages, [Co((DOR)2X)Br2] (X ¼ H, BF2), also are active catalysts, with reduction potentials occurring between 0.48 and 0.75 V versus SCE, depending on the aryl substituent.61 Notably, the carboxylic acid-substituted complex adsorbed onto an indium tin oxide (ITO) electrode showed some catalytic behavior, though under acidic conditions the catalyst eventually dissociated from the surface. Many cobalt complexes other than those containing imines and oximes catalyze hydrogen evolution.62 One example with a single cyclic diphosphane ligand, [Co(PPh2NPh2)(CH3CN)3]2þ, is an active electrocatalyst, operating with bromoanilinium tetrafluoridoborate at a moderate overpotential (258 mV) with a turnover frequency of 90 s1.63 A cobalt polypyridyl complex, [(PY4)Co(CH3CN)2]2þ (PY4 ¼ 2-bis(2-pyridyl)(methoxy)methyl6-pyridylpyridine), has been shown to catalyze the reduction of trifluoroacetic acid at the CoII/I reduction potential (0.81 V vs. SCE), corresponding to an overpotential of  400 mV.64 A rigid dicobalt macrocycle, [LRCo2]2þ, evolves H2 upon addition of 2,6-dichloroanilinium tetrafluoridoborate at 0.3 V and 0.7 V versus SCE for the H- and CH3-substituted complexes; these potentials are less than the CoIICoI/CoICoI couple and correspond to overpotentials of 0.24 and 0.64 V, respectively. Bimetallic cooperativity in this system is thought to be important for catalytic function. Only a few molecular catalysts have been reported to produce H2 from water: these include Ni(cyclam)2þ,22 Co(bpy)32þ (bpy ¼ 2,20 -bipyridine),24 several water-soluble CoII(porphyrin)21 and other cobalt(II) complexes with nitrogen-based macrocycles,65–67 and a cobaloxime (Co (dmgBF2)2(H2O)2) (dmgBF2 ¼ difluoroboryl-dimethylglyoxime, [CoII], vide infra).5,68 Among recent reports, [(PY5Me2)MoO]2þ (PY5Me2 ¼ 2,6bis(1,1-bis(2-pyridyl)ethyl)pyridine) is notable for its ability to produce H2 from neutral water.69 The molybdenum–oxide

complex operates at an overpotential of 0.52 V in 0.6 M phosphate buffer (pH 7), and under optimal conditions (1.4 V vs. SHE) achieves a turnover frequency of 2.4 mol H2 per mol catalyst per second and a turnover number of 6.1  105 mol of H2 per mol of catalyst. This activity is thought to be the highest among reported molecular catalysts that produce hydrogen electrochemically in neutral water. The catalyst is stable under catalytic conditions, maintaining activity for 71 h. The authors have proposed a catalytic cycle (Scheme 3) for [(PY5Me2)MoO]2þ-mediated water reduction that suggests proton reduction is centered at the ligand, similar to that reported for (CpMo(m-S))2(S2CH2) (vide supra),59 as well as nanostructured MoS2.70,71

8.15.2.4

Cobaloximes

Cobaloximes are a promising class of molecules that evolve hydrogen in both aqueous and organic media at relatively low overpotentials (Figure 7).5 They are particularly attractive for catalytic applications, as they can be prepared readily, allowing detailed investigations of their chemical, electrochemical, and photochemical properties. Much work has focused on their ability to catalyze H2 evolution from solutions, although there is still a need to develop new complexes with modified glyoxime ligands for deployment in a solar water splitting devices (Figure 8).

8.15.2.4.1 Electrochemical systems Connolly and Espenson first reported that Co(dmgBF2)2(L)2 (1, Figure 8) catalyzes the Cr2þaq reduction of protons in aqueous HCl.68 Dissociation of an intermediate chloridobridged complex, [(H2O)5Cr–Cl–Co(dmgBF2)2]þ, formed during inner-sphere electron transfer from CrII to CoII, produces [Co(dmgBF2)2 L] (1). In the presence of acid, the CoI anion is likely protonated rapidly to form a hydride, [HCo (dmgBF2)2L], which then reacts to produce H2. Complex 1 and a related species, Co(dpgBF2)2(L)2 (2, dpg ¼ difluoroboryl-diphenylglyoxime),72 were later shown to catalyze electrochemical H2 evolution in acetonitrile (Figure 8).73,74 The reversible, one-electron reduction of 1 occurs at 0.55 V versus SCE in acetonitrile (Table 2). Upon addition of a sufficiently strong acid, catalytic currents were observed near the CoII/I potential in cyclic voltammograms. At higher acid concentrations the peak current increased, eventually reaching a plateau, and the peak position showed a slight positive shift. Further, since CoII is regenerated during H2 production, a loss of the return oxidation wave was observed. Bulk electrolysis experiments confirmed nearquantitative Faradaic yields of H2. A range of proton sources was examined: catalysis was observed with CF3COOH, HClEt2O, p-toluenesulfonic acid monohydrate (TsOHH2O),34

558

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

2+ O

N

N e-

Mo

H2

N

N N 2+

+ O

HO H N

N

N

Mo

N Mo

N

N

N

N

N

N

e2+

N

OH2

N

Mo N

N

N

N

O

N

N

Mo

N

N

2OH-

2H2O

2þ

Scheme 3 Proposed catalytic cycle for [(PY5Me2)MoO] -catalyzed water oxidation.

F

F

LB F O O N N Co N N O O BL F

LB F O O Ph N N Co N N Ph O O BL F

F

F

1

2

Ph Ph

CI H O O N N Co N N O O H N

R

3a R = H 3b R = NMe2

L H O O N N Co N N O O H L

4

3c R = COOMe N H O O N N Co N N O O H N

5

+

CI H O O N N Co N N O O H H CI

CI H O O N N Co N N O O H P(n-Bu)4

H H O O N N Co N N O O H P(n-Bu)4

6

7

7H

Figure 8 Cobaloxime complexes. L is typically H2O or CH3CN.

(p-cyanoanilinium)BF4, and HBF4 Et2O. Although HBF4Et2O was shown to effect H2 evolution with substantially increased rates, catalyst degradation greatly limited overall reaction efficiencies. Substitution of methyl groups by electron-withdrawing phenyl substituents to form 2 shifts the CoII/I reduction potential positively by 270 mV to 0.28 V versus SCE. Electrocatalysis with currents reaching plateaus at high acid

concentrations was observed with HClEt2O, (p-cyanoanilinium)BF4, TsOHH2O, and HBF4Et2O, but not CF3COOH. Catalyst activity (H2 evolution) and CoII/I potentials are well correlated, as demonstrated from electrochemical studies of cobalt difluoroboryl-bridged diglyoximes as well as related compounds with [14]-tetraene-N4 ligands.34 Complexes with more negative reduction potentials catalyze proton reduction with weaker acids at higher rates than those with more positive

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

Table 2

Electrochemical potentials (V vs. SCE in acetonitrile)

Complex

E (CoIII/II)

E (CoII/I)

Reference

1 2 3a 3b 3c 5 6 7

0.20a 0.30a 0.68 0.84 0.78 0.39 0.49 0.92

0.55 0.28 1.13 1.13 1.09 1.10 0.80 0.99

73 73 75b 75b 75b 75b 75b 75b

a

Irreversible couple. NHE versus SCE ¼  0.25 V in acetonitrile.76

b

Table 3

Electrochemical potentials (V vs. SCE in DMF)a

Complex

E (CoIII/II)

E (CoII/I)

Reference

3a 3b 4 5 7

0.67 0.74

1.06 1.09 1.06 1.06 0.89

78 78 78 78 78

0.32 0.70

Converted to SCE from Ag/AgCl/3 mol l1 NaCl, Fcþ/Fc versus Ag/AgCl ¼ 0.55 V in DMF,78 Fcþ/Fc versus SCE ¼ 0.47 V in DMF.79

a

reduction potentials. Metal hydride pKa values increase as the CoII/I reduction potentials become more negative; thus, the CoII/I potentials are related to CoI basicities.77 The overpotentials for proton reduction catalyzed by cobaloxime complexes can be estimated from the difference between E HA=H2 and the potential where catalytic behavior is  observed, which occurs just negative of E (CoII/I). Catalysts that reduce protons near the thermodynamic potential should also be capable of oxidizing H2 in the presence of a conjugate base. Indeed, in the presence of [NBu4][CF3CO2], slow oxidation of H2 by 1 has been observed. From the reaction equilibrium constant, the H2 evolution overpotential was calculated to be 90 mV.34 The electrocatalytic behavior of similar cobalt glyoximes also has been examined. Co(dmgH)2pyCl (3a, py ¼ pyridine) catalyzes H2 evolution at a CoII/I potential of 1.06 V versus SCE in DMF with [Et3NHþ]Cl (Table 3).78 Upon reduction to CoII, the halide ligand is labilized, while the axial pyridine remains coordinated. Substitution of the axial pyridine with 4-(dimethylamino)pyridine (3b) does not substantially modify the CoII/I potential; however, the catalyst appears to be more electroactive. Co(dmgH)2 L2 (4) has the same CoII/I reduction potential as 3 in DMF.78 This and similar hydrogen-bridged complexes can be reacted with BF3Et2O80 to prepare the cobaloximes with glyoxime-bridging BF2 groups, which leads to a positive shift in the CoII/I potentials by  0.5 V. However, these complexes require stronger acids for electrocatalytic hydrogen evolution. The bridging BF2 groups also increase the stability of the catalysts in acidic solutions.

559

8.15.2.4.2 Photochemical systems Catalytic hydrogen production can be driven directly with light when sacrificial electron donors are present to scavenge oxidized photosensitizers, which act both as light harvesters and as photoreductants.81,82 The electronic excited states of photosensitizers can be quenched via electron transfer, directly to a catalyst or via an electron mediator, to generate a reduced catalyst. A multicomponent photochemical system for homogeneous H2 generation with a cobaloxime catalyst in organic solvents, utilizing [Ru(bpy)3]2þ (bpy ¼ 2,20 -bipyridine) as photosensitizer, triethanolamine (TEOA) as sacrificial electron donor, and 4 as catalyst, was first demonstrated by Ziessel and coworkers.83 In DMF solution at pH 8.8, the turnover number for 1 h of irradiation (based on photosensitizer concentration) was 38. Among other works of note,84 Eisenberg and coworkers have shown that 3a catalyzes H2 evolution with a platinum (II) terpyridyl phenylacetylide photosensitizer, [Pt(ttpy) (C  CPh)]þ (ttpy ¼ 40 -p-tolylterpyridine), and TEOA as a donor in 3:2 (v/v) CH3CN:H2O solutions (pH 7–12).85 At pH 8.5 with 0.27 M TEOA, 103 turnovers were achieved after 10 h of irradiation (l > 410 nm). Other cobaloxime catalysts with varying axial bases (3a–c, 5–7) and Pt(II) acetylide photosensitizers have been investigated.75 With 3a as catalyst, 2150 turnovers of H2 were achieved after 10 h of irradiation (l > 410 nm) in a 24:1 (v/v) CH3CN:H2O mixture at pH 8.5. Fontecave and coworkers have focused on multicomponent photosystems for H2 evolution with 1 and both [Ir(ppy)2(phen)]þ (ppy ¼ 2-phenylpyridine, phen ¼ phenanthroline) and [ReBr(CO)3(phen)] as photosensitizers in acetone. Solutions were buffered using Et3N/Et3NHþ, which provided both a sacrificial donor (Et3N) and a proton source (Et3NHþ).86 A quantum yield of 16  1% and 273 turnovers were achieved with [ReBr(CO)3(phen)] in a 15-h experiment (l > 380 nm) with 600 equivalents each of Et3N and Et3NHþ. Alberto and coworkers studied a similar system with 4, employing [ReBr(CO)3(bpy)] as a photosensitizer, 1 M TEOA as a sacrificial electron donor, and 0.1 M acetic acid as a proton source in DMF.87 A total of 150 turnovers in 9 h of irradiation (l > 400 nm) were observed for the system, with a 26  2% quantum yield. Dimethylglyoxime (7.5 mM) was found to enhance catalytic activity, possibly by inhibiting catalyst degradation. When Br was replaced by [NCS] and acetic acid by [HTEOA][BF4], the long-term system stability increased significantly, and up to 6  103 turnovers were achieved.88 Systems also have been reported in which organic chromophores replace noble metal photosensitizers.89,90 In 12 h of irradiation (l > 450 nm), 900 turnovers were achieved using the Eosin Y photosensitizer with 3a, TEOA, and 3 mM free dimethylglyoxime in 1:1 CH3CN:H2O at pH 7.89 Degradation of the photosensitizer/catalyst system was minimized with added dimethylglyoxime, although it was difficult completely to avoid photodecomposition. When a Se-substituted rhodamine analog was utilized as photosensitizer, turnover numbers reached 9  103 in 8 h of irradiation under similar conditions.90 Notably, the turnover frequency achieved in this system is the highest yet reported for the photoreduction of water (5.5  103 mol H2 per mol photosensitizer per h, F ¼ 32.8%). In a similar noble-metal-free system utilizing rose bengal as photosensitizer with the robust catalyst 1, Sun

560

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

OH2 F B F O O N N Co N N O O F B N F

Cl H

O O N N Co N N O O H N

O

N

N

N

N

N

N

S5

OH2 F B F O O N N Co N N O O F B N F

Cl H O O N N Co N N O O H N R = C(CH3)3 M = Zn, Mg, 2H

O

NH

N

N

N

N

S6

R N

N

N

N

N M

N

Ru

N N

N

N R

N Ru

N Ir

N

N

N H N

O

N N

S4

F OH2 B F O O N N Co N N O O F B N F O

N

N

N

S2 R = Me S3 R = Ph

R R

N Ru

N

F OH2 B F O O N N Co N N O O F B N F

O

N N

N

N

S1

N

N Ru

N

R R

O

N N

Ru

N

R R

F OH2 B F O O N N Co N N O O F B N F

R

S7

S8

Figure 9 Sensitizer–cobaloxime conjugates for photocatalytic hydrogen evolution.

and coworkers noted a turnover number of 327 in 1:2 CH3CN at pH 10 with 10% TEOA in 5 h of irradiation (l > 400 nm).91 Bifunctional systems (Figure 9) have been reported with photosensitizers coordinated directly via axial pyridine ligands to [Co(dmgBF2)2(H2O)] and [Co(dmgH)2Cl] (S1–S5).86,92 Irradiation of these sensitizer–catalyst conjugates in Et3N/ Et3NHþ-buffered acetone solution triggers intramolecular electron transfer from photoexcited ruthenium or iridium sensitizers to the cobalt center, leading to H2 evolution. These conjugates exhibited efficiencies up to 8.5 times greater than analogous multicomponent systems under the same conditions; for example, [(ppy)2Ir(L-pyr)Co(dmgBF2)2(H2O)]þ (S5, L-pyr ¼ (4-pyridine)oxazolo[4,5-f]phenanthroline) managed 210 turnovers after 15 h of irradiation with 600 equivalents of Et3N and Et3NHþ in acetone. Related heterodinuclear Ru–Co systems with (S6) and without (S7) linker conjugation have also been studied93; while both complexes were more active than the corresponding multicomponent systems, the species with a nonconjugated bridge exhibited more turnovers. Catalytic hydrogen evolution also has been seen in a noble-

metal-free system (S8) containing a porphyrin sensitizer with a peripheral pyridyl group coordinated to the axial site of a cobaloxime.91

8.15.2.4.3 Reaction pathways Pathways for proton reduction by cobalt complexes are shown in Scheme 4. Reduction of a CoII species to CoI, followed by protonation to form CoIIIH, initiates catalysis.5 The hydride can react in a bimolecular step with another hydride to eliminate H2 (homolytic or bimetallic route, red pathway), or it can be protonated,94 release H2, and generate a CoIII species that is subsequently reduced back to CoII (heterolytic or monometallic route, blue pathway). Alternatively, CoIIIH can be reduced further to CoIIH,95 which can react via analogous homolytic (orange) or heterolytic (purple) pathways. If the CoI species is not protonated, it can be reduced further to a ‘Co0’ species (potentially a CoI complex with a ligand radical) which, upon protonation to form CoIIH (green pathway), can react as above.

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

561

1/2 H2 1/2 H2

CoIII

kET2, e-

ol

k hom

CoII

kh

Homolytic

kET1, e-

CoI

kHCo, HA

CoIIIH

eter

ol

H2 HA

kET3, e-

CoIIIH

HA

‘Co0’

HA Heterolytic H2

Scheme 4 H2 evolution pathways.

8.15.2.4.4 Mechanistic analysis In order to probe the pathway of H2 formation, Chao and Espenson studied the reactivity of kinetically stabilized [HCo (dmgH)2P(n-C4H9)3] (7H).96 Interpretation of hydride reaction kinetics in the presence of acid (aqueous) required a rate expression that was both first and second order in 7H concentration, suggesting parallel homolytic and heterolytic hydrogen evolution pathways. Detailed mechanistic analysis indicated that the homolytic pathway predominates with a rate constant of 1.7  104 M1 s1. The hydridocobaloxime protonation rate constant was found to be 0.42 M1 s1, indicating that the heterolytic pathway would be competitive only at low catalyst or high acid concentrations. 8.15.2.4.4.1 Mechanistic insights from electrochemical experiments Fontecave and coworkers investigated catalysis by 1 in both DMF and acetonitrile.74 Irreversible cathodic waves near the CoII/I potential were observed in the presence of (pcyanoanilinium)BF4, indicating electrocatalytic proton reduction. When CF3COOH was employed as a proton source, lower current densities were seen near the same potential, and an additional catalytic wave appeared near 1.0 V versus SCE, which was attributed to the reduction of CoIIIH to CoIIH and subsequent reaction. With a weak acid, ([Et3NH]Cl), catalytic waves were observed near the CoI/‘0’ potential (1.47 V vs. SCE). As CoI is not sufficiently basic to be protonated by this acid and the thermodynamic potential for hydrogen evolution from [Et3NH]Cl is more negative than the CoII/I potential, catalysis was not observed until CoI reduction. It is noted that the peak currents at these negative potentials were convoluted with direct [Et3NH]Cl reduction at the glassy carbon electrode. Modeling of the cyclic voltammetry data97 suggested that reactions with (p-cyanoanilinium)BF4 and CF3COOH proceed by heterolytic proton reduction (monometallic pathway) involving both CoIIIH and CoIIH. It was concluded that with acids strong enough to protonate both CoI and CoIIIH (e.g., (p-cyanoanilinium)BF4), hydrogen evolution occurred via a heterolytic pathway. In cases

where the acid is strong enough to protonate CoI, but not the CoIIIH intermediate, the hydride is further reduced to CoIIH before reacting either heterolytically or homolytically. Weak acids such as [Et3NH]Cl that are unable to protonate CoI can protonate ‘Co0,’ and H2 evolution can occur via a pathway in which CoIIH is a reactive intermediate. Cyclic voltammograms of 1 and 2 also showed catalytic waves upon addition of TsOHH2O, as discussed above.34 As indicated by the shape of the catalytic wave of 1, it was determined that the rapid reactivity was limited by proton diffusion to the electrode surface. At higher concentrations of TsOHH2O, the wave approached but did not reach a plateau. At low acid concentrations, a quasi-reversible peak at about 1.0 V versus SCE for 1 was observed and attributed to the CoIIIH/CoIIH couple; the reduction potential is similar to that seen for cobalt(III) alkyl species.98 As catalysis at the CoII/I potential was relatively rapid, the concentration of CoIIIH remained low in the reaction layer near the electrode, and only small amounts of CoIIH were generated under these conditions (the hydride reacts slowly with itself or with acid at low concentrations). At high concentrations of TsOHH2O, catalytic waves for 2 reached a plateau, indicating that catalyst reduction at the electrode was equal to the rate of reoxidation. At lower acid concentrations, a second wave at 0.85 V versus SCE indicated electrocatalytic hydrogen evolution by CoIIH, formed by reduction of CoIIIH. As the catalytic reaction of 2 was slower than that of 1, concentrations of CoIIIH and acid were high enough to yield catalytic behavior upon reduction to CoIIH. On the basis of digital simulations of catalytic waves of 1 in the presence of TsOHH2O, it was concluded that bimolecular reactivity of CoIIIH was responsible for hydrogen evolution, a finding consistent with that of Chao and Espenson.96 However, the catalytic waves of 2 could be simulated equally well assuming either a heterolytic or a homolytic mechanism, and thus neither reaction pathway was identified as predominant. As electrocatalysis occurred near the CoII/I reduction potential, it was concluded that pathways through CoIIH were unlikely, although those routes could open up at more negative potentials. In order to shed more

562

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

light on the mechanism of H2 evolution, Peters and coworkers also examined catalysis by 1 with (p-cyanoanilinium)BF4, employing higher acid:catalyst ratios (>10) than used in earlier work.34 With ratios less than 10, simulations were found to be consistent with both homolytic and heterolytic pathways. However, over a wider range (1:1 to 40:1), the data could be simulated successfully only by assuming a homolytic route for H2 evolution. The overall rate constant for H2 evolution was estimated directly from plateau currents observed in cyclic voltammograms. For complexes 1 and 2, the rate constants are 7000 and 200 M1 s1, respectively. In both cases, the reactions are first order in acid concentration. These data indicate that hydride formation is a key step in catalytic H2 evolution. Attempts to isolate a CoIIIH complex with glyoxime ligands have been unsuccessful, except for the kinetically stabilized phosphane-supported hydride 7H (vide supra).96 Several approaches have been explored, including protonation of isolable CoI species with a variety of proton sources, as well as reaction of CoIII halide species with borohydrides. These reactions consistently yield CoII complexes and H2, suggesting that CoIIIH is extremely reactive.34

8.15.2.4.4.2 Mechanistic insights from bulk photochemical experiments Photocatalytic hydrogen evolving systems are considerably more complex than corresponding electrocatalytic ones. Reactions in photosystems with tertiary amine sacrificial donors (TEOA or Et3N) are set out in Scheme 5. The excited photosensitizer (PS*) can be quenched by cobaloxime, producing PSþ, which reacts with R2N–CH2–R0 to generate R2N–CH2–R0 þ; PS*

PS

hn

PS*

PS* + CoII

PS+ + CoI

¨ PS++ R2N–CH 2–R¢

+ ¨ PS + R2N–CH 2–R¢

¨ PS*+ R2N–CH 2–R¢

+ ¨ PS-+ R2N–CH 2–R¢

PS-+ CoII

PS + CoI

+ ¨ R2N–CH 2–R¢

¨ H++ R2N–CH–R¢

¨ R2N–CH–R¢ + CoII

CoI + H+

R2N+= CH–R¢ + CoI

CoIIIH

PS- + CoIIIH

PS + CoIIH

PS* + CoIIIH

PS++ CoIIH

¨ R2N–CH–R¢ + CoIIIH

R2N+ = CH–R¢

R2N+ = CH–R¢ + CoIIH

glycoaldehyde and di(ethanol)amine

Scheme 5 H2 evolution pathways in photochemical systems. The CoIIIH and CoIIH intermediates may react via homolytic and heterolytic reaction pathways (Scheme 1) to evolve H2. Similar pathways exist with Et3N as a sacrificial electron donor.

also can be reductively quenched by R2N–CH2–R0 , producing PS. Decomposition of R2N–CH2–R0 þ yields a proton and a second reducing equivalent, R2N–CH–R0 . CoII can be reduced by PS*, PS, or R2N–CH–R0 , and is then protonated to give CoIIIH, which can be reduced to CoIIH by any of the powerful reductants. The weakly basic conditions under which these photocatalytic systems operate disfavor protonation of CoIIIH. The extensive studies of photochemical systems using PtII acetylide chromophores75,85 and organic photosensitizers89,90 with several cobaloximes showed that hydrogen evolution was first order with respect to catalyst concentration. A monometallic route via CoIIH to produce hydrogen was favored by the authors. The [Re(CO)3Br(phen)]/1/Et3N/Et3NHþ system also exhibited a first-order dependence on catalyst concentration, although above a certain concentration the yield decreases, likely owing to light-absorption competition between the colored catalyst and the photosensitizer.86 However, in the [Re(CO)3Br (bpy)]/4/TEOA/AcOH/DMF system, a second-order dependence on 4 in the rate of H2 evolution (3.7 M1 s1) was observed, and it was concluded that the reaction occurred primarily via a homolytic route, although other mechanisms were not ruled out.87 The authors noted that a heterolytic process likely would require the generation of CoIIH as a reactive intermediate. Lastly, while few investigations have focused on hydrogen evolution in purely aqueous systems, Szajna-Fuller and Bakac utilized spectrophotometric monitoring to determine the kinetics of hydrogen evolution from solutions containing TiIII citrate and 1.99 At near-neutral pH, protonation of CoI was deemed to be fast on the timescale of the overall reaction, and citrate was suspected to be involved as either a proton donor or a ligand for CoI. Upon protonation, an intermediate, attributed to a citratecoordinated CoIIIH with a maximum absorbance at 770 nm, disappears rapidly at low pH. Decomposition of the catalyst under the reaction conditions was attributed to the decay of a [CoIII(dmgBF2)]þ intermediate. The authors noted that these two observations suggest that a heterolytic route via CoIIIH is primarily responsible for H2 evolution. 8.15.2.4.4.3 Mechanistic insights from time-resolved spectroscopy Photochemical methods can be utilized to trigger electron or proton transfer processes and form reactive catalytic species. These pulsed-laser methods,100–102 when coupled with timeresolved spectroscopy, allow intermediates to be identified and monitored on timescales appreciably shorter than those accessible by conventional stopped-flow spectroscopy.103 Employing these techniques, key steps in H2 evolution reaction cycles have been elucidated. Formation of CoIIIH via proton transfer from the triplet excited state of brominated naphthol to a reduced cobaloxime, [CoI(dmgBF2)2(CH3CN)], was monitored via transient absorption (Scheme 6).104 The rate constant for CoIIIH formation (3.5–4.7  109 M1 s1) suggests that proton transfer is coupled to deactivation of the excited photoacid. Under the reaction conditions explored, the reactive CoIIIH intermediate is subsequently reduced by excess CoI–diglyoxime in solution to produce a reduced hydride species, CoIIH (kred ¼ 9.2  106 M1 s1). The CoIIH is then protonated by the weakly acidic brominated naphthol to yield CoII–diglyoxime and

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

F

F

B F O O N N Co N N O L O B

kp

F

2H+, 1e-

F

[CoI]

F

B F OL O N N Co N N O L O B

+ H2

F [CoII]

+ 3*[BrNaphOH]

+BrNaphOH

kH2

-[Na][BrNaphO]

-[Na][BrNaphO] F

F

F

563

B

B F OH O N N Co N N O L O B

OH

kred + [CoI] - [CoII] F

F [CoIIIH]

F O

N N Co N N O L O B F

[CoIIH]

Scheme 6 Proposed mechanism of hydrogen evolution from a photogenerated hydridocobaloxime.

Ru2+*

Ru2+

k1

Ru2+*

+ MV2+

Ru3+

+ MV•+

k2

Ru3+

+ MV•+

Ru2+

+ MV2+

k3

MV•+

+ Co2+

MV2+

+ Co+

k4

Ru3+

+ Co+

Ru2+

+ Co2+

k5

Ru3+

+ Co2+

Ru2+

+ Co3+

k6

MV•+

+ Co3+

MV2+

+ Co2+

k7

Co3+

+ Co+

2Co2+

k8

Scheme 7 Elementary electron transfer steps following photoexcitation of Ru(bpy)32þ (Ru2þ) in the presence of methyl viologen (MV2þ) and 2 (Co2þ).

H2, indicating that CoIIH is strongly hydridic. This heterolytic reaction pathway is expected to dominate when low concentrations of CoIIIH and an excess of educing equivalents are present. In separate work, laser flash-quench methods100–102 were employed to trigger the reduction of CoII to CoI. Here, it was shown that methyl viologen oxidatively quenches pulsed-laser excited [Ru(bpy)3]2þ (bpy ¼ 2,20 -bipyridine) and the reduced viologen species (MV •þ) acts as an electron relay to deliver an electron to CoII–diglyoxime (CoII(dpgBF2)2(CH3CN)2), producing anionic CoI and [Ru(bpy)3]3þ. In the absence of a proton donor, the reduced CoI species recombines with the oxidized species in solution ([Ru(bpy)3]3þ, CoIII, Scheme 7), as supported

by global analysis of transient absorption spectra recorded at various time delays after excitation. Single-wavelength kinetics traces were simulated with a kinetics model, providing rate constants for the elementary electron transfer steps described by Scheme 7. From this analysis, rate constants and reorganization parameters (l) for the CoIII–CoII and CoII–CoI electron selfexchange reactions were determined (CoIII–CoII self-exchange, k ¼ 9.5 108–2.6 105 M1 s1, l ¼ 3.9 eV; CoII–CoI selfexchange, k ¼ 1.2 105 M1 s1, l ¼ 1.2 eV). Thermodynamic analysis of competing H2 evolution pathways using these data has shed light on the barriers and driving forces of the elementary reaction steps involved in proton reduction by CoI–diglyoximes (Figure 10).103 The barriers to H2 evolution along the heterolytic CoIIIH pathway are, in most cases, substantially greater than those of the homolytic CoIIIH route. In particular, a formidable barrier is associated with CoIII– diglyoxime formation. Mechanistic details from electrochemical investigations of cobaloxime-catalyzed H2 evolution suggest that the rate-limiting step of the thermodynamically preferred homolytic CoIIIH pathway is associated with formation of the hydride.34 These mechanistic investigations suggest that two competing H2-evolution pathways are responsible for catalytic H2 production. Thermodynamic analysis of CoIIIH reactivity indicates that bimolecular elimination of H2 from two hydrides (homolytic CoIIIH pathway) is preferred over protonation (heterolytic CoIIIH pathway). However, in the presence of strong reductants, the heterolytic pathway involving CoIIH is competitive. Understanding the conditions under which these competing H2-evolution pathways operate in cobaloximes will guide the design and construction of electrode-bound catalyst components for solar fuel devices.

564

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

° II I ⎤ ΔG°1 = ℑ ⎡E D°+ D − E Co Co ⎥ ⎢⎣ ⎦ ° II I ⎤ ΔG2° = ΔG1° + ℑ ⎡E D° + D − E Co Co ⎥ ⎢⎣ ⎦ ΔG3° = ΔG2° + 2.303RT pK a (HA ) − pK a CoIII H

[

(

)]

ΔG2°a = ΔG1° + 2.303RT pK a (HA ) − pK a CoIII H ° III ΔG3°a = ℑ ⎡E D° + D + E Co ⎢⎣

II

Co

⎤ ° − 2E HA H2 ⎥ ⎦

ΔG °

° ° H2a = ΔG 3a − ΔG2a

[ ( )] ΔG ° = ΔG ° + 2.303RT [pK (HA ) − pK (Co H)] III

4

3

a

a

‡ ΔGET 2

ΔGH°2 = ΔGR° − ΔG4°

° H ⎤ ΔGR° = 2ℑ ⎡E D° + D − E HA 2 ⎥ ⎢⎣ ⎦

D + D+ + 2A- + H2 + CoII + CoIII ΔG3°a

‡ ΔGET 1

2CoII + 2HA+ 2D

ΔG 1°

‡ ΔGET 1

CoI + CoII + 2HA+ D+ + D

ΔG2°

ΔG3°

ΔG2°a

D + D+ + HA+ A-+CoII+CoIIIH

ΔG4°

2CoI + 2HA+ 2D+

ΔG

‡ ET 1

⎞ ⎛ ℑ ⎡E° − E ° ⎤ ⎜ ⎢ D + D CoII CoI ⎥ + 12 λCoII CoI + λD + D ⎟ ⎣ ⎦ ⎠ =⎝ 2 ( λCoII CoI + λD + D )

(

)

2

ΔGR°

CoIIIH + CoI + A- + HA+ 2D+

ΔG

‡ ET 2

⎛ ℑ ⎡E° − E° ⎤ + 1 λ III II + λ + ⎞ ⎜ ⎢ D+ D CoIII CoII ⎥ D D ⎟ ⎣ ⎦ 2 Co Co ⎠ =⎝ 2 ( λCoIII CoII + λD + D )

(

)

2

2CoIIIH + 2A- + 2D+ 2CoII + H2 + 2A- + 2D+

Figure 10 Thermodynamic analysis of cobaloxime-catalyzed hydrogen evolution.

8.15.3

Conclusion

Recent research on molecular hydrogen-evolving catalysts has revealed promising candidates for a solar-driven water splitting device based on earth abundant materials. Detailed electrochemical and photochemical studies have provided mechanistic insights which are guiding current research in this field. Continued efforts in catalyst design and mechanistic investigations are expected to lead to highly efficient catalysts for electrocatalytic hydrogen generation and help make solar fuels a reality. For a related chapter in this Comprehensive, we refer to Chapter 8.11.

Acknowledgments This work was supported by NSF Center for Chemical Innovation (Powering the Planet, CHE-0947829), the Arnold and Mabel Beckman Foundation, and CCSER (Gordon and Betty Moore Foundation). J.L.D. was supported by an NSF Graduate Research Fellowship.

References 1. Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 15729–15735. 2. Lewis, N. S. MRS Bull. 2007, 32, 808–820. 3. Gray, H. B. Nat. Chem. 2009, 1, 7.

4. Dempsey, J. L.; Esswein, A. J.; Manke, D. R.; Rosenthal, J.; Soper, J. D.; Nocera, D. G. Inorg. Chem. 2005, 44, 6879–6892. 5. Dempsey, J. L.; Brunschwig, B. S.; Winkler, J. R.; Gray, H. B. Acc. Chem. Res. 2009, 42, 1995–2004. 6. Weaste, R. C., Astle, M. J., Beyer, W. H., Eds.; CRC Handbook of Chemistry and Physics, 64th ed.; CRC Press: Boca Raton, FL, 1983. 7. Koelle, U. New J. Chem. 1992, 16, 157–169. 8. Eisenberg, R.; Gray, H. B. Inorg. Chem. 2008, 47, 1697–1699. 9. Artero, V.; Fontecave, M. Coord. Chem. Rev. 2005, 249, 1518–1535. 10. Khaselev, O.; Turner, J. A. Science 1998, 280, 425–427. 11. Dominey, R. N.; Lewis, N. S.; Bruce, J. A.; Bookbinder, D. C.; Wrighton, M. S. J. Am. Chem. Soc. 1982, 104, 467–482. 12. Kayes, B. M.; Atwater, H. A.; Lewis, N. S. J. Appl. Phys. 2005, 97, 114302. 13. Boettcher, S. W.; Spurgeon, J. M.; Putnam, M. C.; Warren, E. L.; TurnerEvans, D. B.; Kelzenberg, M. D.; Maiolo, J. R.; Atwater, H. A.; Lewis, N. S. Science 2010, 327, 185–187. 14. Kelzenberg, M. D.; Boettcher, S. W.; Petykiewicz, J. A.; Turner-Evans, D. B.; Putnam, M. C.; Warren, E. L.; Spurgeon, J. M.; Briggs, R. M.; Lewis, N. S.; Atwater, H. A. Nat. Mater. 2010, 9, 239–244. 15. Britton, H. T. S. In Hydrogen Ions Chapman & Hall: London, 1955; pp 55–59. 16. Koper, M.; Bouwman, E. Angew. Chem. Int. Ed. 2010, 49, 3723–3725. 17. Frey, M. ChemBioChem 2002, 3, 153–160. 18. Adams, M. W. W. Biochim. Biophys. Acta 1990, 1020, 115–145. 19. Tard, C. D.; Pickett, C. J. Chem. Rev. 2009, 109, 2245–2274. 20. Gloaguen, F.; Rauchfuss, T. B. Chem. Soc. Rev. 2009, 38, 100–108. 21. Kellett, R. M.; Spiro, T. G. Inorg. Chem. 1985, 24, 2373–2377. 22. Collin, J. P.; Jouaiti, A.; Sauvage, J. P. Inorg. Chem. 1988, 27, 1986–1990. 23. Efros, L. L.; Thorp, H. H.; Brudvig, G. W.; Crabtree, R. H. Inorg. Chem. 1992, 31, 1722–1724. 24. Krishnan, C. V.; Sutin, N. J. Am. Chem. Soc. 1981, 103, 2141–2142. 25. Rillema, D. P.; Endicott, J. F. Inorg. Chem. 1976, 15, 1459–1461. 26. Abdel-Hamid, R.; El-Sagher, H. M.; Abdel-Mawgoud, A. M.; Nafady, A. Polyhedron 1998, 17, 4535–4541. 27. Bernhardt, P. V.; Jones, L. A. Inorg. Chem. 1999, 38, 5086–5090. 28. Felton, G. A. N.; Glass, R. S.; Lichtenberger, D. L.; Evans, D. H. Inorg. Chem. 2006, 45, 9181–9184.

Solar Fuels: Approaches to Catalytic Hydrogen Evolution

29. Himmel, D.; Goll, S. K.; Leito, I.; Krossing, I. Angew. Chem. Int. Ed. 2010, 49, 6885–6888. 30. Fourmond, V.; Jacques, P.-A.; Fontecave, M.; Artero, V. Inorg. Chem. 2010, 49, 10338–10347. 31. Kaupmees, K.; Kaljurand, I.; Leito, I. J. Phys. Chem. A 2010, 114, 11788–11793. 32. Kilgore, U. J.; Roberts, J. A. S.; Pool, D. H.; Appel, A. M.; Stewart, M. P.; Dubois, M. R.; Dougherty, W. G.; Kassel, W. S.; Bullock, R. M.; Dubois, D. L. J. Am. Chem. Soc. 2011, 133, 5861–5872. 33. Izutsu, K. Acid–Base Dissociation Constants in Dipolar Aprotic Solvents. Blackwell Scientific Publications: Oxford, 1990. 34. Hu, X.; Brunschwig, B. S.; Peters, J. C. J. Am. Chem. Soc. 2007, 129, 8988–8998. 35. Darensbourg, M. Y.; Lyon, E. J.; Smee, J. J. Coord. Chem. Rev. 2000, 206, 533–561. 36. Darensbourg, M. Y.; Lyon, E. J.; Zhao, X.; Georgakaki, I. P. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 3683–3688. 37. Wang, Q.; Barclay, J. E.; Blake, A. J.; Davies, E. S.; Evans, D. J.; Marr, A. C.; Mcinnes, E. J. L.; Mcmaster, J.; Wilson, C.; Schro¨der, M. Chem. Eur. J. 2004, 10, 3384–3396. 38. Perra, A.; Davies, E. S.; Hyde, J. R.; Wang, Q.; Mcmaster, J.; Schroder, M. Chem. Commun. 2006, 1103–1105. 39. Ogo, S.; Kabe, R.; Uehara, K.; Kure, B.; Nishimura, T.; Menon, S. C.; Harada, R.; Fukuzumi, S.; Higuchi, Y.; Ohhara, T.; Tamada, T.; Kuroki, R. Science 2007, 316, 585–587. 40. Kure, B.; Matsumoto, T.; Ichikawa, K.; Fukuzumi, S.; Higuchi, Y.; Yagi, T.; Ogo, S. Dalton Trans. 2008, 4747–4755. 41. Matsumoto, T.; Kure, B.; Ogo, S. Chem. Lett. 2008, 37, 970–971. 42. Canaguier, S.; Vaccaro, L.; Artero, V.; Ostermann, R.; Pe´caut, J.; Field, M.; Fontecave, M. Chem. Eur. J. 2009, 15, 9350–9364. 43. Oudart, Y.; Artero, V.; Norel, L.; Train, C.; Pe´caut, J.; Fontecave, M. J. Organomet. Chem. 2009, 694, 2866–2869. 44. Oudart, Y.; Artero, V.; Pe´caut, J.; Fontecave, M. Inorg. Chem. 2006, 45, 4334–4336. 45. Tran, P. D.; Artero, V.; Fontecave, M. Energy Environ. Sci. 2010, 3, 727–747. 46. Canaguier, S.; Artero, V.; Fontecave, M. Dalton Trans. 2008, 315–325. 47. Barton, B. E.; Whaley, C. M.; Rauchfuss, T. B.; Gray, D. L. J. Am. Chem. Soc. 2009, 131, 6942–6943. 48. Canaguier, S.; Field, M.; Oudart, Y.; Pecaut, J.; Fontecave, M.; Artero, V. Chem. Commun. 2010, 46, 5876–5878. 49. Liu, X. M.; Ibrahim, S. K.; Tard, C.; Pickett, C. J. Coord. Chem. Rev. 2005, 249, 1641–1652. 50. Capon, J. F.; Gloaguen, F.; Schollhammer, P.; Talarmin, J. Coord. Chem. Rev. 2005, 249, 1664–1676. 51. Gloaguen, F.; Lawrence, J. D.; Rauchfuss, T. B. J. Am. Chem. Soc. 2001, 123, 9476–9477. 52. Capon, J.-F.; Gloaguen, F.; Pe´tillon, F. Y.; Schollhammer, P.; Talarmin, J. Coord. Chem. Rev. 2009, 253, 1476–1494. 53. Wilson, A. D.; Newell, R. H.; Mcnevin, M. J.; Muckerman, J. T.; Rakowski Dubois, M.; Dubois, D. L. J. Am. Chem. Soc. 2006, 128, 358–366. 54. Fraze, K.; Wilson, A. D.; Appel, A. M.; Rakowski Dubois, M.; Dubois, D. L. Organometallics 2007, 26, 3918–3924. 55. Wilson, A. D.; Shoemaker, R. K.; Miedaner, A.; Muckerman, J. T.; Dubois, D. L.; Dubois, M. R. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 6951–6956. 56. Yang, J. Y.; Bullock, R. M.; Shaw, W. J.; Twamley, B.; Fraze, K.; Dubois, M. R.; Dubois, D. L. J. Am. Chem. Soc. 2009, 131, 5935–5945. 57. Dubois, M. R.; Dubois, D. L. Chem. Soc. Rev. 2009, 38, 62–72. 58. Curtis, C. J.; Miedaner, A.; Ciancanelli, R.; Ellis, W. W.; Noll, B. C.; Rakowski Dubois, M.; Dubois, D. L. Inorg. Chem. 2002, 42, 216–227. 59. Appel, A. M.; Dubois, D. L.; Rakowski Dubois, M. J. Am. Chem. Soc. 2005, 127, 12717–12726. 60. Jacques, P.-A.; Artero, V.; Pe´caut, J.; Fontecave, M. Proc. Nat. Acad. Sci. U.S.A. 2009, 106, 20627–20632. 61. Berben, L. A.; Peters, J. C. Chem. Commun. 2010, 46, 398–400. 62. Losse, S.; Vos, J. G.; Rau, S. Coord. Chem. Rev. 2010, 254, 2492–2504.

565

63. Jacobsen, G. M.; Yang, J. Y.; Twamley, B.; Wilson, A. D.; Bullock, R. M.; Dubois, M. R.; Dubois, D. L. Energy Environ. Sci. 2008, 1, 167–174. 64. Bigi, J. P.; Hanna, T. E.; Harman, W. H.; Chang, A.; Chang, C. J. Chem. Commun. 2010, 46, 958–960. 65. Endicott, J. F.; Rillema, D. P.; Papaconstantinou, E. Inorg. Chem. 1971, 10, 1739–1746. 66. Tait, A. M.; Hoffman, M. Z.; Hayon, E. J. Am. Chem. Soc. 1976, 98, 86–93. 67. Brown, G. M.; Brunschwig, B. S.; Creutz, C.; Endicott, J. F.; Sutin, N. J. Am. Chem. Soc. 1979, 101, 1298–1300. 68. Connolly, P.; Espenson, J. H. Inorg. Chem. 1986, 25, 2684–2688. 69. Karunadasa, H. I.; Chang, C. J.; Long, J. R. Nature 2010, 464, 1329–1333. 70. Hinnemann, B.; Moses, P. G.; Bonde, J.; Jørgensen, K. P.; Nielsen, J. H.; Horch, S.; Chorkendorff, I.; Nørskov, J. K. J. Am. Chem. Soc. 2005, 127, 5308–5309. 71. Jaramillo, T. F.; Jorgensen, K. P.; Bonde, J.; Nielsen, J. H.; Horch, S.; Chorkendorff, I. Science 2007, 317, 100–102. 72. Tovrog, B. S.; Kitko, D. J.; Drago, R. S. J. Am. Chem. Soc. 1976, 98, 5144–5153. 73. Hu, X. L.; Cossairt, B. M.; Brunschwig, B. S.; Lewis, N. S.; Peters, J. C. Chem. Commun. 2005, 4723–4725. 74. Baffert, C.; Artero, V.; Fontecave, M. Inorg. Chem. 2007, 46, 1817–1824. 75. Du, P.; Schneider, J.; Luo, G.; Brennessel, W. W.; Eisenberg, R. Inorg. Chem. 2009, 48, 4952–4962. 76. Pavlishchuk, V. V.; Addison, A. W. Inorg. Chim. Acta 2000, 298, 97–102. 77. Creutz, C.; Chou, M. H.; Fujita, E.; Szalda, D. J. Coord. Chem. Rev. 2005, 249, 375–390. 78. Razavet, M.; Artero, V.; Fontecave, M. Inorg. Chem. 2005, 44, 4786–4795. 79. Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877–910. 80. Schrauzer, G. N.; Windgassen, R. J. J. Am. Chem. Soc. 1966, 88, 3738–3743. 81. Gray, H. B.; Maverick, A. W. Science 1981, 214, 1201–1205. 82. Esswein, A. J.; Nocera, D. G. Chem. Rev. 2007, 107, 4022–4047. 83. Hawecker, J.; Lehn, J. M.; Ziessel, R. Nouv. J. Chim. 1983, 7, 271–277. 84. Wang, M.; Na, Y.; Gorlov, M.; Sun, L. Dalton Trans. 2009, 6458–6467. 85. Du, P.; Knowles, K.; Eisenberg, R. J. Am. Chem. Soc. 2008, 130, 12576–12577. 86. Fihri, A.; Artero, V.; Pereira, A.; Fontecave, M. Dalton Trans. 2008, 5567–5569. 87. Probst, B.; Kolano, C.; Hamm, P.; Alberto, R. Inorg. Chem. 2009, 48, 1836–1843. 88. Probst, B.; Rodenberg, A.; Guttentag, M.; Hamm, P.; Alberto, R. Inorg. Chem. 2010, 49, 6453–6460. 89. Lazarides, T.; McCormick, T.; Du, P.; Luo, G.; Lindley, B.; Eisenberg, R. J. Am. Chem. Soc. 2009, 131, 9192–9194. 90. McCormick, T. M.; Calitree, B. D.; Orchard, A.; Kraut, N. D.; Bright, F. V.; Detty, M. R.; Eisenberg, R. J. Am. Chem. Soc. 2010, 132, 15480–15483. 91. Zhang, P.; Wang, M.; Dong, J. F.; Li, X. Q.; Wang, F.; Wu, L. Z.; Sun, L. C. J. Phys. Chem. C 2010, 114, 15868–15874. 92. Fihri, A.; Artero, V.; Razavet, M.; Baffert, C.; Leibl, W.; Fontecave, M. Angew. Chem. Int. Ed. 2008, 47, 564–567. 93. Li, C.; Wang, M.; Pan, J. X.; Zhang, P.; Zhang, R.; Sun, L. C. J. Organomet. Chem. 2009, 694, 2814–2819. 94. Besora, M.; Lledos, A.; Maseras, F. Chem. Soc. Rev. 2009, 38, 957–966. 95. Bhugun, I.; Lexa, D.; Save´ant, J.-M. J. Am. Chem. Soc. 1996, 118, 3982–3983. 96. Chao, T.-H.; Espenson, J. H. J. Am. Chem. Soc. 1978, 100, 129–133. 97. Rudolf, M. DigiElch 2.0, http://digielch.de/. 98. Shi, S.; Bakac, A.; Espenson, J. H. Inorg. Chem. 1991, 30, 3410–3414. 99. Szajna-Fuller, E.; Bakac, A. Eur. J. Inorg. Chem. 2010, 2010, 2488–2494. 100. Ogata, T.; Yanagida, S.; Brunschwig, B. S.; Fujita, E. J. Am. Chem. Soc. 1995, 117, 6708–6716. 101. Berglund, J.; Pascher, T.; Winkler, J. R.; Gray, H. B. J. Am. Chem. Soc. 1997, 119, 2464–2469. 102. Bjerrum, M. J.; Casimiro, D. R.; Chang, I. J.; Dibilio, A. J.; Gray, H. B.; Hill, M. G.; Langen, R.; Mines, G. A.; Skov, L. K.; Winkler, J. R.; Wuttke, D. S. J. Bioenerg. Biomembr. 1995, 27, 295–302. 103. Dempsey, J. L.; Winkler, J. R.; Gray, H. B. J. Am. Chem. Soc. 2010, 132, 1060–1065. 104. Dempsey, J. L.; Winkler, J. R.; Gray, H. B. J. Am. Chem. Soc. 2010, 132, 16774–16776.