Energy-Related Small Molecule Activation ... - ACS Publications

Energy-Related Small Molecule Activation...

0 downloads 86 Views 17MB Size

Review pubs.acs.org/CR

Energy-Related Small Molecule Activation Reactions: Oxygen Reduction and Hydrogen and Oxygen Evolution Reactions Catalyzed by Porphyrin- and Corrole-Based Systems Wei Zhang,† Wenzhen Lai,‡ and Rui Cao*,†,‡ †

Key Laboratory of Applied Surface and Colloid Chemistry, Ministry of Education, School of Chemistry and Chemical Engineering, Shaanxi Normal University, Xi’an 710119, China ‡ Department of Chemistry, Renmin University of China, Beijing 100872, China S Supporting Information *

ABSTRACT: Globally increasing energy demands and environmental concerns related to the use of fossil fuels have stimulated extensive research to identify new energy systems and economies that are sustainable, clean, low cost, and environmentally benign. Hydrogen generation from solar-driven water splitting is a promising strategy to store solar energy in chemical bonds. The subsequent combustion of hydrogen in fuel cells produces electric energy, and the only exhaust is water. These two reactions compose an ideal process to provide clean and sustainable energy. In such a process, a hydrogen evolution reaction (HER), an oxygen evolution reaction (OER) during water splitting, and an oxygen reduction reaction (ORR) as a fuel cell cathodic reaction are key steps that aﬀect the eﬃciency of the overall energy conversion. Catalysts play key roles in this process by improving the kinetics of these reactions. Porphyrin-based and corrole-based systems are versatile and can eﬃciently catalyze the ORR, OER, and HER. Because of the signiﬁcance of energy-related small molecule activation, this review covers recent progress in hydrogen evolution, oxygen evolution, and oxygen reduction reactions catalyzed by porphyrins and corroles.

CONTENTS 1. Introduction 1.1. Energy-Related Small Molecule Activation Reactions 1.2. Porphyrins and Corroles 1.2.1. Biological Importance of Porphyrins in O2 Activation 1.2.2. Corrole as a Mimic of Macrocycles in Nature 1.2.3. Advantages of Porphyrins and Corroles for ORR, HER, and OER 1.3. Scope of This Review 2. Oxygen Reduction Reaction 2.1. Oxygen Reduction Reaction in Nature 2.2. Implications from Synthetic Modeling Studies of Cytochrome c Oxidases 2.2.1. Role of the Proximate CuII Ion in O2 Reduction 2.2.2. Role of the Tyr 244 Residue in O 2 Reduction 2.2.3. Role of the Distal Pocket Environment in O2 Reduction 2.2.4. Role of the Trans Ligand of Fe in O2 Reduction 2.2.5. Implications from Synthetic Models

© 2016 American Chemical Society

2.3. Simple Oxygen Reduction Reaction Catalysts 2.3.1. Iron-Based ORR Catalysts 2.3.2. Cobalt-Based ORR Catalysts 2.3.3. ORR Catalysts Based on Other Transition Metals 2.3.4. Metal-Free Porphyrin ORR Catalysts 2.4. ORR Catalysts Based on Porphyrin and Corrole Architectures 2.4.1. Fe-Based Architectures for ORR Catalysis 2.4.2. Co-Based Architectures for ORR Catalysis 3. Hydrogen Evolution Reaction 3.1. Simple HER Catalysts 3.2. HER Catalysts Based on Porphyrin and Corrole Architectures 4. Oxygen Evolution Reaction 4.1. Simple OER Catalysts 4.2. OER Catalysts Based on Porphyrin and Corrole Architectures 5. Porphyrin and Corrole Grafted Materials 5.1. Materials for ORR Catalysis

3718 3718 3719 3719 3720 3720 3720 3721 3721 3723 3723 3725 3727 3729 3731

3731 3731 3737 3742 3743 3744 3744 3746 3753 3753 3757 3759 3759 3764 3765 3765

Special Issue: Expanded, Contracted, and Isomeric Porphyrins Received: May 10, 2016 Published: November 9, 2016 3717

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews 5.2. Materials for HER Catalysis 5.3. Materials for OER Catalysis 6. Other Applications of Porphyrins and Corroles in Small Molecule Activation Reactions 6.1. Porphyrin- and Corrole-Based Chromophores in Catalysis 6.1.1. Porphyrin-Diiron Systems for HER 6.1.2. Porphyrin-Monocobalt Systems for HER 6.1.3. Porphyrin-Pt Nanoparticle Systems for HER 6.1.4. Porphyrin-Based Systems for OER 6.2. Materials Derived from Porphyrin and Corrole Precursors 6.3. Porphyrin and Corrole Analogues in Catalysis 7. Concluding Remarks Associated Content Supporting Information Author Information Corresponding Author Notes Biographies Acknowledgments Abbreviations References

Review

splitting, and they are connected to the electrodes through ohmic contact. Another promising type of solar-driven water splitting system is based on photoelectrodes coated with eﬃcient catalysts.34−43 One photoelectrode is immersed in an electrolyte and is directly connected with another counter electrode through ohmic contact. The counter electrode employed is usually highly eﬃcient for the desired half reaction of water splitting. This system is typically called the photoelectrochemical water splitting cell. These two systems perform similarly in solar to hydrogen conversion, and they have relatively higher eﬃciency than photocatalytic water splitting systems.10−17,42,50,51 To boost solar-to-hydrogen conversion eﬃciency, one research direction seeks to increase light absorption and electron/hole separation in semiconductors (in solar cells or in photoelectrodes); another important research direction involves improving water splitting kinetics on an electrode surface (i.e., ﬁnding eﬃcient and stable catalysts). In photodriven water splitting systems, the kinetics of the hydrogen evolution reaction (HER) and the oxygen evolution reaction (OER), especially the latter, greatly limit solar to chemical energy conversion eﬃciency.52 To lower the energy barriers of the HER and OER, various catalysts have been developed for loading on the surfaces of semiconductors/ electrodes or for being presented as homogeneous complexes in aqueous reactants/electrolytes. To insightfully and conveniently study the performance of these catalysts, electrochemical analysis techniques are widely adopted to survey promising candidates for practical applications in solar-driven water-splitting devices.53−58 Electrocatalytic HER involves two diﬀerent processes for the formation of H2 from the reduction of protons on catalytic sites.59 For both processes, a proton is ﬁrst reduced into hydride species on catalytic centers. For the formation of H−H, in one mechanism, two hydride species react to release one hydrogen molecule. In the other mechanism, one hydride reacts with one proton−electron couple to release one hydrogen molecule. In contrast, the OER is a much more complicated process. Oxidative reactions leading to O2 formation from water are challenging from both thermodynamic (1.23 V vs reversible hydrogen electrode, RHE) and kinetic (signiﬁcant molecular rearrangement) perspectives. 60−64 To generate one O 2 molecule, two water molecules must be involved in the reaction. Each water molecule will lose two proton−electron couples in sequence, and four proton−electron couples in total will be lost from the two water molecules for the formation of an O2 molecule.53,54,56,65,66 The proton-coupled electron transfer (PCET) discharge of water molecules on catalytic sites with the formation of a hydroxyl adduct is believed to be the ﬁrst step in the OER. Two such intermediates with close geometric positions may, in some rare cases, interact with each other to release a hydrogen peroxide molecule, which can be further oxidized to generate O2 molecules. Alternatively, hydroxyl species on catalytic centers may be more likely to be further discharged to form a catalytic center-oxo intermediate (typically metal-oxo). There are two representative pathways for the generation of O2 molecules starting from this metal-oxo species. The more probable mechanism is the nucleophilic attack of the metal-oxo by a water molecule. With two consequent PCET discharge steps from water, O2 adducts on metal centers will be formed to release O2 molecules. Another pathway is metal-oxo-oxo-metal coupling between two proximate metal-oxo intermediates, which more commonly

3770 3770 3771 3771 3771 3773 3773 3776 3777 3780 3782 3785 3785 3785 3785 3785 3785 3785 3785 3786

1. INTRODUCTION With the rapid depletion of fossil fuels dating back to the industrial revolution of the 18th century, humankind faces critical energy and environmental crises. On the basis of current reserve levels and the consumption level of fossil fuels, all the available resources on our planet will be inadequate to supply the global energy demand in a century or two.1−3 To reach the next level of technological advancement for our civilization, we must exploit clean and sustainable energy resources. Solar irradiation energy reaching earth far exceeds the energy needs for all human activities.1,4 The conversion and storage of solar energy in an eﬃcient and cost-eﬀective manner are subjects that attract great attention. Photosynthesis by green plants converts and stores solar energy in chemical form. In this process, water is eﬀectively split into oxygen and hydrogen gases. Oxygen is released from cells, whereas hydrogen is captured in NAD(P)H. Water splitting is a thermodynamically nonspontaneous process with a positive Gibbs free energy change of 237 kJ mol−1.5 Thus, the combustion of hydrogen in fuel cells produces great free energy, and the only exhaust is recyclable water. These two reactions compose an ideal process to provide humankind with clean and sustainable energy. In such a process, hydrogen and oxygen evolution reactions (the redox reactions of water splitting) and an oxygen reduction reaction (the cathodic reaction of the hydrogen fuel cell) are key elements that aﬀect the eﬃciency of the overall energy conversion. These energy-related small molecule activation reactions widely exist in biological processes. Scientists, inspired by nature, are trying to understand these processes to design eﬃcient systems for energy conversion. 1.1. Energy-Related Small Molecule Activation Reactions

Among the various types of established solar-driven water splitting systems,6−49 a water electrolysis cell coupled with a solar cell is a promising setup.44−49 Solar cells must be connected in series to provide suﬃcient potential for water 3718

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

typically intensely colored (the name “porphyrin” comes from the Greek for purple). Deprotonated porphyrins are dianionic macrocycle ligands of proper size that can in principle bind almost all metal ions to produce a four-coordinated structure with both the axial positions unoccupied. In nature, metalloporphyrins are involved in a wide variety of important biological processes. For example, heme, the iron complex of porphyrin, plays a key role in various proteins and enzymes for O 2 binding, transport, and storage (hemoglobin and myoglobin), electron transfer (cytochrome b and c), O2 activation and utilization (cytochrome oxidase and cytochrome P450), and peroxide management and degradation (peroxidase and catalase). In these heme structures, although the central iron porphyrin units are the same, their activities can be considerably ﬁne-tuned by axial ligands and surrounding environments. As shown in Figure 2, for hemoglobin,

occurs in heterogeneous electrocatalysts or binuclear molecular catalysts that can provide metal-oxo species with proper spacing. Due to the strict requirements for the OER, it shows much slower kinetics compared to the HER. The poor performance on the water oxidation side greatly limits the overall eﬃciency of water-splitting systems.53,55,67 Hydrogen provided from water splitting can be oxidized in a hydrogen fuel cell to generate electricity. Compared to hydrogen oxidation on the anode, the oxygen reduction reaction (ORR) on the cathode has a much higher overpotential and thus greatly limits the electromotive force of fuel cells.68−74 Typically, the ORR is accomplished via 2e− or 4e− pathways, depending on the nature of the catalyst. In nature, the most common type of ORR is catalyzed by the a3-CuB site with an iron porphyrin core structure in cytochrome c oxidase (CcO).75−78 To mimic the natural system for eﬃcient ORR, transition metal porphyrins, corroles, and other macrocyclic molecules have been designed and synthesized for electrocatalytic ORR since the discovery of cobalt phthalocyanine in 1964.73,79 For these planar molecules with a metal-N4 coordination structure, an oxygen molecule can bind to the metal center at the axial vacant coordination site. With the acceptance of electrons and protons, O2 is ﬁrst reduced to H2O2 on metal sites. This peroxide species is either released to electrolytes or is further reduced to water. Unlike CcOs, which produce a minimal amount of peroxide from ORR, a mixture of H2O and H2O2 is generated from most of these metal complex catalysts in laboratory electrocatalytic ORR studies. The activity, stability, and composition of the reduction products are strongly dependent on the nature of the macrocycle catalysts, speciﬁcally the metal centers and substituent groups at the rim of macrocycles. 1.2. Porphyrins and Corroles

1.2.1. Biological Importance of Porphyrins in O2 Activation. Porphyrins are a group of cyclic tetrapyrrolic compounds with four pyrrole subunits interconnected at their α carbon atoms via methine bridges (Figure 1). The unsubstituted porphyrin is called porphin, which represents the simplest porphyrin. The heterocyclic porphyrin structure is a largely conjugated aromatic ring; thus, porphyrin complexes are

Figure 2. X-ray structures of heme active sites of (a) cytochrome P450, (b) hemoglobin, (c) catalase, and (d) CcO.

myoglobin, and cytochrome c oxidase, a histidine imidazole group axially binds the Fe ion via the N atom; for cytochrome P450, a cysteine thiolate ligand binds the Fe ion via the S atom; and for catalase, a tyrosine phenolate group binds the Fe ion via the O atom. The sixth position is therefore vacant for the coordination of substrates. These diﬀerences in the axial trans ligands are believed to regulate a variety of heme activities involving diﬀerent substrates through a so-called “push eﬀect”.80−83 A signiﬁcant feature related to both O2 activation and O2 evolution is the potential generation of high-valent metal oxo intermediates in the catalytic cycle. For example, an FeIV-oxo πcation porphyrin radical (compound I) formed upon heterolytic O−O bond scission of FeIII-OOH2, is generally considered to be a key intermediate involved in all cytochrome P450, peroxidase, cytochrome oxidase, and heme oxygenase enzymes.81 The formation of a MnVO unit has been suggested in the photosynthetic water oxidation process.84−87 Nucleophilic water attack at this MnVO unit could be critical for the O−O bond formation. Therefore, metal-oxo species have attracted extensive research interests because of their relevance to O2 chemistry. Porphyrins have been welldocumented in both enzymes and synthetic complexes to stabilize high-valent metal oxo units.81,83,88−94 The electronic nature of porphyrin ligands has a strong inﬂuence on the stability of the incorporated metal-oxo species; electron-

Figure 1. Comparison numbering schemes and the coordination environments of porphyrin and corrole macrocycles. 3719

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 3. Mass and energy circulation of a H2 fuel cell coupled with H2 production from solar-driven water splitting (left) and the ORR pathways on the cathode (right).

catalysts for these energy-related small molecule activation reactions is based primarily on the following considerations. (1) Both porphyrin and corrole ligands can accommodate a variety of transition metals and main group elements and can provide the central metal a rigid and stable coordination environment. The four-coordinated square planar geometry of metal ions is ideal for substrate binding and activation. The resulting metal complexes are usually quite robust in diverse media, ranging from organic to aqueous and from highly acidic to highly basic solutions. (2) Porphyrin and corrole ligands can be readily and systematically modiﬁed with diﬀerent functional moieties, including electron-donating and electron-withdrawing substituents, proximal trans axial ligands, intramolecular proton relay (for ORR and HER), and base (for OER) groups, accessional redox sites, and sterically bulky groups with either hydrophobic or hydrophilic characteristics. Importantly, their basic geometries and coordination features persist after these modiﬁcations. This feature makes porphyrins and corroles an excellent platform to investigate the structure−function relationship. Moreover, porphyrins and corroles can be readily assembled with other molecular architectures and/or grafted onto materials. (3) In addition to their stable and rigid coordination, another fascinating feature of porphyrin and corrole ligands is their ability to stabilize high-valent metal ions (metal-oxo units), which are generally believed to be involved as key intermediates in oxygen reduction and evolution catalytic cycles. Associated with this eﬀect, porphyrin and corrole ligands, particularly corrole ligands, may oﬀer low-valent metal ions with strong reducing powers. Such a feature is favored for the ORR and HER. (4) On the one hand, porphyrin and corrole ligands are redox noninnocent and can greatly enrich the redox chemistry of their metal complexes. On the other hand, they are resistant to both reductive and oxidative decompositions.

deﬁcient porphyrin ligands are more eﬀective than electron-rich porphyrin ligands.92 1.2.2. Corrole as a Mimic of Macrocycles in Nature. Corroles are another group of cyclic tetrapyrrolic aromatic compounds; they have a skeleton structure similar to the corrin ring of vitamin B12. The corrole macrocycle is more closely related to the porphyrin ring (Figure 1), but it consists of 19 carbon atoms (lacking one meso-carbon atom compared to porphyrin) bearing a direct pyrrole−pyrrole connection. This structural disparity makes corrole a trianionic ligand with fundamentally diﬀerent coordination chemistry compared with dianionic porphyrin. For example, corrole is much more eﬀective in stabilizing high-valent metal centers;95 it has a smaller coordination cavity and thus gives shorter metal− nitrogen (M−N) bond distances, and it typically displays ﬁvecoordinated structures with domed conformations.96 Although corroles were ﬁrst synthesized in 1964 by Johnson and Kay,97 the rapid development of corrole chemistry started in 1999,98,99 when Gross and co-workers reported an eﬃcient, one-pot procedure for the synthesis of stable and robust mesosubstituted corroles (the unsubstituted meso-carbon atoms are very reactive and sensitive to oxidation). The discovery of facile methodologies for the synthesis of meso-substituted corroles and corresponding metal complexes allowed the utilization of these macrocycle complexes in various ﬁelds. Several exhaustive reviews summarized the potential use of corroles in catalysis, sensing, dye-sensitized solar cells, photoactive arrays, and medicinal applications.100−102 Corroles have been shown to be versatile trianionic ligands for the coordination of a number of transition metals and main group elements.103−114 In these complexes, the central metals are usually in high-valent states, although the macrocycles are also noninnocent with signiﬁcant radical character in many cases. Therefore, the special ability of corroles to stabilize highvalent transition metal ions forms a basis for better understanding corrole chemistry. Gross and Gray presented an electronic structural explanation for this stabilization after comparing Mn corroles and porphyrins.115 It is thought that the strong σ-donation of trianionic corrole ligands greatly increases the energies of metal d orbitals. Consequently, oxidation of the metal center will occur much more easily in open d-shell metallocorroles. However, the corrole macrocycles are more prone to oxidation in d10 systems. A number of highvalent metal-oxo complexes116−131 and metal-nitrido complexes132,133 have been synthesized using corrole ligands. 1.2.3. Advantages of Porphyrins and Corroles for ORR, HER, and OER. The choice of porphyrins and corroles as

1.3. Scope of This Review

This review covers recent signiﬁcant achievements in the ORR, HER, and OER catalyzed by porphyrins and corroles. These three reactions are involved in energy-related small molecule activation reactions and are the focus of chemistry, material, and environmental science in the past decades. In the following sections, we will ﬁrst address the implications learned from studies of CcOs and synthetic models and summarize recent advancements in ORR catalysis. Considering that several comprehensive reviews on metalloporphyrins as biomimetic models and ORR catalysts have been published,75,78,134,135 the current review will only focus on the progress that has been 3720

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

heterolytic O−O bond cleavage during O2 reduction (Figure 4). For late transition metal elements, electrons in the d orbitals

made in the past decade with ORR catalysts based on porphyrins and corroles. In the next sections, we will ﬁrst concentrate on HER and OER catalysis by porphyrins and corroles with a particular emphasis on structure designs and mechanistic aspects, and then we will focus on porphyrin- and corrole-grafted materials with an emphasis on various methods used for assembling. In the last section, we will brieﬂy present examples of ORR, HER, and OER catalysis using porphyrins and corroles as photosensitizers (rather than catalysts), using them as precursors to make pyrolyzed (carbonized) materials with active M-Nx/C sites, and using structural analogues of porphyrins and corroles that contain a square planar N4 metalbinding environment.

2. OXYGEN REDUCTION REACTION The ORR is a signiﬁcant process in cell metabolism. An eﬃcient ORR is necessary for the realization of a hydrogenbased society. As illustrated in Figure 3, a H2 fuel cell coupled with H2 production from solar-driven water splitting is a promising strategy to solve the energy and environmental problems related to the use of fossil fuels.79,136 At the anode of a fuel cell, H2 is oxidized into protons with the release of electrons to the external circuit. At the cathode, O2 reacts with protons and electrons to produce water. This ideal process based on solar-driven water splitting and a H2 fuel cell is a system with zero exhaust emission. Compared to H2 oxidation, ORR is sluggish, which greatly limits the electromotive force of fuel cells.68−73 The development of ORR catalysts has therefore attracted extensive research interests. In general, ORR is accomplished via either the 2e− or the 4e− pathway as described in Figure 3, depending on the nature of the catalysts. Although the four-electron reduction of O2 is biologically relevant and is more favored from a thermodynamic point of view (eqs 1−3), the two-electron reduction of O2 to H2O2 is also interesting due to an alternative H2O2-based energy system (eqs 4−6). On the basis of these equations, H2O2 can be regarded as an energy carrier alternative to oil and hydrogen.137,138 The maximum output potential is 1.09 V, which is comparable to the 1.23 V potential of a H2 fuel cell. Compared to H2 fuel cells, H2O2 is optimal for storage and transportation. Therefore, the energy system outlined in eq 4−6, which involves selective H2O2 production by electrocatalytic ORR and electric power generation by a H2O2 fuel cell, is another attractive scheme to solve our current energy and environmental problems. anode: 2H 2 → 4H+ + 4e− 0 V

(1)

cathode: O2 + 4H+ + 4e− → 2H 2O 1.23 V

(2)

total: 2H 2 + O2 → 2H 2O

(3)

anode: H 2O2 → O2 + 2H+ + 2e− 0.68 V

(4)

cathode: H 2O2 + 2H+ + 2e− → 2H 2O 1.77 V

(5)

total: 2H 2O2 → 2H 2O + O2

(6)

Figure 4. Possible reaction pathways of O2 reduction catalyzed by metal complexes, giving the two-electron reduction product H2O2 and the four-electron reduction product H2O.

will occupy the antibonding orbitals of the metal-oxo unit. Consequently, terminal oxo complexes of late transition metal elements are unlikely to be generated due to the electrostatic repulsion of electrons between the d orbitals and the oxo ligand. However, late transition metal complexes can still catalyze the four-electron reduction of O2 through a bimetallic mechanism with the formation of dinuclear peroxo species. Subsequent homolytic O−O bond cleavage can lead to the complete reduction of O2 to H2O. In this section, we will ﬁrst brieﬂy introduce the biological ORR process and then discuss the roles of CuB, tyrosine residue, the distal pocket coordination environment, and the trans ligand of heme iron in the catalytic reduction of O2 mediated by synthetic modeling complexes. In the subsequent two subsections, recent advances in porphyrin and corrole ORR catalysts are addressed, including the use of simple macrocycles and architectures based on porphyrins and corroles. 2.1. Oxygen Reduction Reaction in Nature

Respiration is a vital process of hydrocarbon oxidation in organisms that provides the energy required for cell activities.75,76 Molecular O2 functions as the oxidant for terminal electron acceptance in this process. Hydrocarbon oxidation reactions and ORRs occur associatively in an electron balance, but they proceed separately in diﬀerent locations in cells. The electrons donated from the hydrocarbon oxidation are delivered to the ORR by the electron transport chain of respiration. In eukaryotes, respiration occurs in mitochondria. The ORR is catalyzed by the trans-membrane mitochondrial protein cytochrome c oxidase (CcO), which belongs to a superfamily of heme/Cu oxidases. CcO is the destination of electrons from the electron transport chain in respiration.76,139,140 CcO accepts electrons from proximate reduced cytochrome c proteins and pass electrons to molecular O2. In the presence of protons, O2 is thus reduced to water in CcO.75−78 The electron pathway from cytochrome c to O2 and the proposed ORR mechanism in CcO are illustrated in Figure 5.76,77,139 The structure of CcO is fundamentally composed of two heme proteins with iron porphyrin core structures (heme a, heme a3) and two copper centers (CuA, CuB).141−146 Electrons are transferred from reduced cytochrome c proteins to the binuclear CuA center of CcO, and then to cytochrome heme a, and subsequently to their ﬁnal destination of heme a3

For mononuclear metal species, in principle, early transition metal complexes are able to catalyze the four-electron reduction of O2, whereas late transition metal complexes typically catalyze the two-electron reduction of O2. This diﬀerence in ORR selectivity can be explained by the ease of forming a terminal metal-oxo species, which is the key intermediate produced by 3721

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 5. Electron and mass transfer pathways for ORR in CcO (left) and the catalytic cycle for ORR at the FeIICuI site from the heme a3-CuB site in CcO (right).

Figure 6. Synthetic models of 1−6 studied by Karlin and co-workers.

thought to be formed after O2 uptake. The O−O bond cleavage proceeds immediately with the acceptance of two stepwise proton−electron couples to generate FeIII-OH and CuII-OH species. These two hydroxo intermediates accept two proton− electron couples separately to release two water molecules with the reduction of metal centers into the original FeIICuI state to complete the catalytic cycle. The net result of a full catalytic cycle is that four electrons are transferred from reduced cytochrome c to the a3-CuB site along with four protons provided by protein cofactors to reduce one O2 molecule into two water molecules. In this vital process, hemes with core structures of iron porphyrins are involved as both the electron transfer pathway and the ORR site. Notably, the phenol group of the tyrosine residue linked to the CuB-ligating histidine unit has been proposed to provide an additional site for electron/ proton transfer during the catalytic cycle of O2 reduction.

and CuB for O2 reduction. The free energy released from the ORR is utilized to pump protons across the mitochondrial membrane to higher potentials, which creates a proton electrochemical potential driving force for ATP synthesis and other energy-requiring biological reactions. A possible catalytic ORR cycle in the a3-CuB site with a fully reduced FeIICuI binuclear center is simpliﬁed and proposed in Figure 5.76−78,140,147 The space between FeII and CuI is ideal for the incorporation of an O2 molecule. In the initial stages, peroxo/superoxo intermediates may exist. However, they are rapidly reduced into water on a much shorter time scale than the time required for electron transfer from cytochrome c to the a3-CuB site. This ability of CcO minimizes the potential of releasing any peroxo/superoxo ORR intermediates into the organism matrix. The complete reduction of O2 to water not only provides the maximum free energy released from the ORR but also protects organisms from the damage of toxic peroxo/ superoxo chemicals. At the beginning of the catalytic cycle, an FeIII-O-O-CuII peroxo species with a trans conﬁguration is 3722

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

the superoxo unit.151 Instead, the μ-peroxo FeIII porphyrin dimer [FeIII(F8TPP)]2(O22−) was formed in noncoordinating solvents (i.e., dichloromethane, toluene). Both processes are fully reversible; the regeneration of 1 with the evolution of O2 is realized under bubbling argon. Unlike other FeIII-(O22−)-FeIII peroxo-bridged FeIII porphyrins, the release of an O2 molecule from [FeIII(F8TPP)]2(O22−) is most likely due to the presence of strong electron-withdrawing meso-2,6-diﬂuorophenyl groups. With a covalently linked Cu-binding site proximal to the Feporphyrin ring, Karlin and co-workers reported the O 2 reactivity of a bimetallic FeII-CuI complex that is relevant to the heme/Cu site in CcOs.150 This FeII-CuI complex 4 (Figure 6) reacts with O2 to form a μ-peroxo complex FeIII-(O22−)-CuII, which has greatly improved thermal stability at room temperature (t1/2 ∼ 60 min). This result implies that the presence of a properly appended Cu unit can considerably stabilize the μ-peroxo species, highlighting the importance of ligand design in biomimetic studies because the μ-peroxo species formed by Fe porphyrin and untethered Cu-polypyridyl subunit is not stable under these conditions. Nuclear magnetic resonance (NMR) studies revealed that this peroxo intermediate represents the ﬁrst case of a paramagnetic high-spin heme-Cu O2 adduct. The high-spin nature of the Fe porphyrin is most likely attributed to the absence of a strong axial base ligand. Because of the thermal stability of this μ-peroxo intermediate, its reactions with various reagents are of considerable interest. FeIII-(O22−)-CuII reacted with two or more equivalents of cobaltocene to give the μ-oxo species FeIIIO-CuII. This behavior is diﬀerent from the low-spin heme-Cu peroxo species, which is reduced completely to the FeII-CuI form as reported by Collman and co-workers.154 This result demonstrates the reactivity diﬀerence between high-spin and low-spin heme-Cu species; the latter are much more easily reduced. It is interesting to compare the electrocatalytic ORR activities of mononuclear complex 1, binuclear Fe-Cu complex 4, and the Cu-free form of 4 (with an empty tethered chelate).148 The eﬃciency of the four-electron reduction of O2 to water was 74% for 4, 59% for the Cu-free form of 4, and 20% for 1. This result suggests that the Cu ion is not necessary for the four-electron ORR. A possible role of the Cu ion in 4, and the empty pyridyl groups in the Cu-free form of 4 is to stabilize the superoxo unit bound to the FeIII ion (to prevent the release of superoxide ion O2•−). In combination with the mentioned O2 reactivity studies, it may be concluded that the stabilization of FeIIIsuperoxo is critical for the ORR pathway to favor the fourelectron reduction of O2 to water. To better understand the ORR mechanism, Karlin and coworkers further studied the chemical reduction of O2 using complex 4 or its Cu-free form as the catalyst and decamethylferrocene (Me10Fc) and triﬂuoroacetic acid (TFA) as the electron and proton source, respectively.153 For O2 reduction with either 4 or its Cu-free form, the catalytic cycle steady state observed at room temperature is the fully reduced form (i.e., FeII-CuI for 4 and FeII for the Cu-free form), indicating that the rate-determining step (rds) at room temperature is the O2-binding process. However, at −60 °C, the steady state observed is assigned to a new hydroperoxo species FeIII-OOH for both 4 and its Cu-free form. Importantly, this [FeIII-OOH, CuII] (the peroxo unit is not linked to the Cu ion) could also be generated independently at −80 °C by adding TFA to the peroxo complex FeIII-(O22−)-CuII, a species described in the previous paragraph. Further kinetic studies

2.2. Implications from Synthetic Modeling Studies of Cytochrome c Oxidases

2.2.1. Role of the Proximate CuII Ion in O2 Reduction. Two roles of CuB are proposed for O2 reduction, decreasing the activation energy barrier of O2 reduction by forming a transient FeIII-O-O-CuII μ-peroxo intermediate or simply providing an electron to reduce O2 (the latter is conﬁrmed in single-turnover O2 reduction experiments showing CuB is one-electron oxidized). Although it is reasonable that the formation of FeIII-O-O-CuII would stabilize the peroxo moiety to avoid the release of damaging partially reduced oxygen species (PROS) into the medium and might also facilitate the O−O bond cleavage, theoretical studies indicate that a ferric-hydroperoxo FeIII-OOH intermediate was generated instead of FeIII-O-OCuII.75 Compared to enzymatic experiments, studies using synthetic models have a major advantage in illustrating the role of CuB because synthetic analogues of CcOs in both the binuclear Fe-Cu and mononuclear Fe (Cu-free) forms can be readily prepared, and their catalytic performance toward O2 reduction can be systematically investigated and compared. Several excellent review papers have been published to summarize these synthetic analogues.75,78,134,135 Therefore, in the following content, only a few examples will be discussed that may provide valuable implications for catalytic ORR with metal porphyrins and metal corroles. Karlin and co-workers designed and synthesized a number of synthetic model complexes (several are depicted in Figure 6) to mimic CcOs with a particular attempt to shed light on the role of the proximate Cu ion,148−153 including models without a tethered Cu-polypyridyl subunit and an axial ligand (complex 1), models with a tethered Cu-polypyridyl subunit but without an axial ligand (complex 4), and the more relevant models containing both a tethered Cu-polypyridyl subunit and an axial pyridyl ligand (complex 5 and 6). With these heme/Cu synthetic analogues (in the reduced form, FeII-CuI), Karlin and co-workers were able to study their stoichiometric reactions with O2 under nonprotic conditions with beneﬁts, including (1) monitoring the reaction process by a variety of spectroscopic methods and (2) performing the reaction under low temperature to stabilize active intermediates for detection and/or isolation. Reaction of O2 with FeII tetrakis(2,6-diﬂuorophenyl)porphyrin [FeII(F8TPP)] (1) and one equivalent of a CuI complex of tripicolylamine [CuI(TPA)]+ in acetonitrile at −40 °C led to the formation of a heterobinuclear peroxo complex 2 (Figure 6).152 Signiﬁcantly, an FeIII-superoxo intermediate prior to the formation of the FeIII-peroxo-CuII complex 2 was detected in stopped-ﬂow experiments below −75 °C, whereas little or no Cu-only O2-adducts were observed. Peroxo complex 2 is not stable and undergoes thermal degradation to the FeIIIO-CuII μ-oxo complex 3 at room temperature with the concomitant release of a half equivalent of O2. These results indicate that a FeIII-peroxo-CuII intermediate can be generated with insuﬃcient thermal stability. The observation of an O2 adduct of FeII-porphyrin rather than an O2 adduct of [CuI(TPA)]+ in acetonitrile is noteworthy because it suggests that the heme is the position in a CcO where O2 binding happens. Subsequent O2 reactivity studies of 1 (without the Cu moiety) showed that it reacted with O2 at low temperatures to form a stable FeIII-superoxo complex in coordinating solvents, such as tetrahydrofuran, propionitrile, and acetone, with a solvent molecule binding to the axial position of FeIII trans to 3723

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

demonstrated that the rds at −60 °C is the O−O bond cleavage of the steady-state FeIII-OOH species. Moreover, the O2binding rate constant of complex 4 is approximately two times larger than that of its Cu-free form, and the rates of the O−O bond cleavage of FeIII-OOH observed for 4 and its Cu-free form are nearly identical. On the basis of these results, the role of the Cu ion is likely to assist Fe porphyrin in O2 binding, but it has a negligible eﬀect on the O−O bond cleavage of the FeIIIOOH species. The proposed catalytic cycles for 4 and its Cufree form are depicted in Figure 7.

bimetallic FeII-CuI form of complex 4 reacts with O2 to give the hydroperoxo species [FeIII-OOH, CuII] in the presence of TFA, and (2) the addition of TFA to the peroxo complex FeIII(O22−)-CuII at −80 °C also generates this hydroperoxo species. In 2003, Naruta and co-workers reported the O2 reactivity of a binuclear FeII-CuI complex 7 with an FeII porphyrin tethered to a tripicolylamine [CuI(TPA)]+ derivative (Figure 8).155 The

Figure 8. Reaction between synthetic CcO model complex 7 and O2 in acetonitrile to give the μ-peroxo complex 8.

resulting peroxo-bridged FeIII-(O22−)-CuII species 8 exhibited a remarkable stability due to the presence of bulky meso-Mes substituents and methyl groups on the TPA unit that protected the peroxo unit through steric hindrance. Importantly, this peroxo species was unambiguously characterized by singlecrystal X-ray diﬀraction, revealing a μ−η2:η1 bridging peroxo coordination mode with an observed O−O bond length of 1.460(6) Å. This is the ﬁrst crystal structure report of a peroxobridged binuclear heme-copper model complex and is also the ﬁrst structure report of a heterodimetallic peroxo-bridged complex. The O−O bond was further characterized by resonance Raman spectroscopy, exhibiting an isotope sensitive band at 790 (16O2)/746 (18O2) cm−1. Collman and co-workers have also contributed signiﬁcantly to the biomimetic chemistry of CcOs with attempts to elucidate the role of CuB by analyzing the electrochemical behavior of a number of well-deﬁned synthetic analogues. With the use of binuclear FeII-CuI complexes 9 and 10 (Figure 9), Collman and co-workers demonstrated that the state of the bound peroxo was very sensitive to the nature of the CuI-coordinating site. Complex 9 predominately catalyzes the two-electron reduction of O2 to H2O2, whereas 10 catalyzes the four-electron reduction of O2 to H2O under physiological conditions.156 The 4e− ORR with catalysts on the graphite disk electrode can be written as eqs 7−9.

Figure 7. Proposed reaction mechanisms of the four-electron reduction of O2 by Me10Fc catalyzed by (a) synthetic model complex 4 and (b) its Cu-free form in the presence of TFA in acetone. Redrawn with permission from ref 153. Copyright 2011 Proceedings of the National Academy of Sciences of the United States of America.

More relevant synthetic model complexes 5 and 6 (Figure 6) containing both a tethered Cu-polypyridyl subunit and an axial pyridyl ligand were synthesized and studied by Karlin and coworkers.149 Ligands for 5 and 6 possess diﬀerent linkers connecting the tridentate Cu-binding chelate and the axial pyridine ligand to the porphyrin ring. The Cu-free form of 5 binds O2 reversibly, whereas the Cu-free form of 6 reacts with O2 irreversibly, a result that demonstrates very eﬀective reactivity regulation via subtle changes in ligand architecture. However, both 5 and 6 react with O2 irreversibly to produce the peroxo species FeIII-(O22−)-CuII, again highlighting the role of the Cu ion in stabilizing the O2 adduct. It is noteworthy that in Karlin’s work, these stoichiometric reactions using synthetic analogues under nonprotic conditions diﬀer from the enzymatic O2 reduction, largely because protons are available in the latter reaction. This discrepancy may change the O2 reactivity of the Fe-Cu core. Under enzymatic conditions, negatively charged distal oxygen can be protonated to form the putative ferric-hydroperoxo species instead of binding to the Cu ion to form a bridged FeIII-O-O-CuII peroxo intermediate under nonprotic conditions. This hypothesis is corroborated by experimental observations that (1) the

catalyst + O2 → catalyst−O2 ΔG1

(7)

catalyst−O2 + 4e− + 4H+ → 2H 2O + catalyst ΔG2

(8)

O2 + 4e− + 4H+ → 2H 2O ΔGall

(9)

The ﬁrst step (eq 7) occurs in solution without the electron transfer from the electrode, whereas the second step (eq 8) requires the input of electrons from the electrode; the potential value of this step can be determined by electrochemistry. The value of ΔG1 can be calculated by subtracting ΔG2 from ΔGall. For complex 10, ΔG1 is −47.4 kcal mol−1, and for 9, ΔG1 is only −7.4 kcal mol−1. This diﬀerence is because the CuII oxidation state was more favorable in 10 with four Cu-binding N atoms compared to 9 with only three N atoms. 3724

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

1:1 stoichiometry to produce a bridging peroxo species. The Cu ion plays a crucial role in O2 binding because the O2 adduct of 11 remains intact under a continuous purge with pure argon, but the O2 adduct of the Cu-free form of 11 shows reversible O2 binding. This result reinforces the thermodynamic implication mentioned above that the formation of a stable peroxo intermediate (thermodynamically trapped peroxide) leads to the 4e− ORR pathway. Both a Cu ion and an internal axial ligand are required for the 4e− ORR by 11; its Cu-free form and the CoII-CuI complex lacking the appended imidazole ligand catalyze the two-electron reduction of O2 to H2O2. The presence of an internal axial ligand may direct O2 binding to the distal intermetallic face instead of the proximal face. Closer structural analogues of CcOs were ﬁrst described by Collman and co-workers with the use of three imidazole ligands as the Cu coordination site (Figure 9) instead of the TPA, N,Nbis(2-pyridylethyl)amine, triazacyclononane, or N,N,N-tribenzyl-tris(aminomethyl)amine binding sites presented previously.157 Importantly, these new model complexes 12 and 13 show clean four-electron electroreduction of O2 over a wide pH range. By using a series of new synthetic analogues containing both the trisimidazole Cu-binding site and the imidazole proximal ligand of the Fe ion, Collman and co-workers investigated the reaction mechanism of O2 reduction in detail.158 Key ﬁndings include the following. (1) These synthetic analogues catalyze the 4e− ORR at physiologically relevant potentials with catalytic rate constants comparable to those reported for CcO enzymes. The formally ferric-hydroperoxo intermediate is generated in the turnover-determining step, which minimizes the concentration of its steady-state species. (2) Although these synthetic analogues can catalyze the decomposition of H2O2 as CcO does, a two-step ORR (in which H2O2 is generated ﬁrst and is subsequently catalytically dismutated or reduced to H2O by neighboring metalloporphyrin molecules, which is biologically irrelevant) is unlikely to operate in these systems. (3) Cu ion does not provide a lowerenergy pathway for the ORR relative to heme alone and thus does not inﬂuence the rate or the kinetic mechanism. Instead, it can accelerate the O2 binding process and acts mainly as a oneelectron storage site, which suppresses the superoxide-releasing autoxidation of the oxygenated catalyst. 2.2.2. Role of the Tyr244 Residue in O2 Reduction. The other component in addition to CuB that has attracted intensive research interests is the Tyr244 residue (the tyrosine residue in the vicinity). In the aforementioned electrochemical studies, synthetic analogues are adsorbed on graphite electrodes, which are able to rapidly transfer electrons to the catalysts. Consequently, although these model complexes contain only two redox sites, an iron porphyrin and a distal Cu ion, they can selectively catalyze the four-electron reduction of O2 to water at the physiological pH and potential. However, in enzymatic reactions, electrons transfer from external reductants (i.e., cytochrome c) to CuA then to heme a, and ﬁnally to the heme a3-CuB site, a process that is much slower than the electrochemical process on a graphite electrode. It has been proposed that in an enzymatic cycle of CcOs, four electrons, two from the heme a3 Fe, one from CuB, and one from Tyr244, are required to reduce the bound O2 to water. The oxidized sites [heme a3, CuB, Tyr244] are then recharged slowly by ferrous cytochrome c for a subsequent cycle of O2 binding and reduction. To elucidate the function of Tyr244, Collman and coworkers moved further to develop new structural and functional analogues of the CcO active site by mimicking heme a3, CuB,

Figure 9. Synthetic CcO model complexes 9, 10, 12, and 13 studied by Collman and co-workers to elucidate the role of CuB.

Consequently, the CuI ion in 10 has a larger reducing power compared with the CuI ion in 9. The conversion of the stable peroxo intermediate (10-O22−) to H2O2 is endergonic and thermodynamically unfavorable, and its conversion to H2O becomes a predominant pathway (Figure 10). In contrast, the

Figure 10. Proposed energy diagram of O2 reduction and the relative energy changes in the four-electron vs two-electron reduction of O2. Redrawn with permission from ref 156. Copyright 1998 Wiley-VCH.

pathways from the less stable peroxo intermediate (9-O22−) to H2O2 and H2O are both exergonic. This result has valuable implications for CcO: the calculated ΔG1 for the O2-binding step of CcO (eq 7) is −52 kcal mol−1, which may be why CcO releases no dangerous PROS compounds. Therefore, it can be concluded that the Cu ion appears to be the deciding factor as to whether the O2 reduction reaction yields a two-electron or four-electron reduced product. Interestingly, unlike 9, its Co-analogue 11 (not shown), which contains a CoII-porphyrin with a fastened CuI ion, electrocatalyzes the four-electron reduction of O2 at physiological pH.154 This CoII-CuI complex 11 binds O2 strongly in a 3725

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

and Tyr244 (Figure 11).159 Key factors involved in this model include the following. (1) A proximal trans imidazole ligand and

can be predictably tuned by varying the length and the degree of conjugation of the SAM. Two SAMs were chosen for this study, a slow electron ﬂux SAM composed of 1-azidohexadecanethiol and hexadecanethiol with a standard electron-transfer rate constant of 6 ± 0.1 s−1 between the gold electrode and the model complexes and a fast electron ﬂux SAM composed of highly conjugated azidophenylene-ethynylenebenzyl thiol and octanethiol with a standard electron-transfer rate constant >1 × 104 s−1. As the authors expected, electrochemical studies revealed that the catalytic ORR current was limited by electron transfer on slow SAM, whereas the current was limited by O2 diﬀusion on fast SAM because the electron delivery to the catalysts on the fast SAM was suﬃciently rapid. When immobilized on the fast SAM, the complete Fe-Cu-Tyr functional analogue 14, the tyrosinelacking Fe-Cu analogue 15, and the Fe-only analogue 16 (not shown) catalyze the four-electron reduction of O2 to water with selectivities of 95% for 14, 93% for 15, and 89% for 16. When immobilized on the slow SAM, these numbers are 96% for 14 and 87% for 15. For complex 16 on the slow SAM, the catalyst was rapidly degraded due to the generation of a substantial amount of PROS. Some conclusions can be made. (1) The selectivity of 14 is nearly identical when it is immobilized on either fast or slow SAMs, resembling the most important ability of CcOs to catalyze O2 reduction by four electrons to water in both fully reduced and partially reduced states (under limited electron transfer ﬂux). (2) The rather diﬀerent behaviors of both complexes 15 and 16 on fast and slow SAMs suggested that with limited electron transfer ﬂux, both Cu and phenol are

Figure 11. Synthetic CcO model complexes 14 and 15 studied by Collman and co-workers to elucidate the role of the Tyr244 residue.

a distal trisimidazole Cu-binding site are tethered to Fe porphyrin with an Fe···Cu distance of approximately 5 Å. (2) A phenol group is covalently linked to one of the Cu-ligating imidazoles with direction toward the O2-binding site on the Fe porphyrin. (3) A terminal alkyne group is introduced to the proximal trans imidazole ligand with the aim of selectively attaching these analogues onto azide-terminated self-assembled monolayer (SAM) ﬁlms on gold electrodes through azidealkyne cycloaddition (Figure 12). This strategy is crucial to control the electron ﬂux from electrodes to the covalently attached catalysts because the rate of such an electron transfer

Figure 12. Schematic representation showing the slow SAM functionalized with model complex 14 (left) and the fast SAM functionalized with model complex 14 (right). Redrawn with permission from ref 159. Copyright 2007 Science. 3726

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

give a superoxo FeIII-(O2•−) species. (2) Subsequent proton transfer and electron transfer from CuI and PhOH to this superoxo unit leads to heterolytic O−O bond scission to produce one molecule of H2O and a [FeIVO, CuII, PhO•] state. (3) This oxidized intermediate is then reduced by two equivalents of cytochrome c to give a [FeIII-OH, CuII, PhOH] state. (4) Further reduction by two equivalents of cytochrome c regenerates catalyst 17 with the release of a H2O molecule. Kinetic studies were performed, revealing that O2 binding was the rds in the catalytic cycle. The rate of O2 binding to the FeII porphyrin (0.1 s−1) is at least 10 times less than the rate of electron transfer from cytochrome c to the FeIII-CuII-PhOH form of 17 (1.2 s−1). The rate of O−O bond cleavage is ≫0.1 s−1. This work is the ﬁrst report of O2 reduction to H2O using the biologically relevant reductant cytochrome c and synthetic model catalysts. 2.2.3. Role of the Distal Pocket Environment in O2 Reduction. To elucidate the role of the distal pocket in CcOs, Collman and co-workers designed three model complexes (Figure 14) with diﬀerent picket fences (18 and 19) or without

required to improve the four-electron reduction pathway. Therefore, the primary function of these two redox centers is to rapidly provide the additional two electrons to reduce O2 to water. A small but signiﬁcant diﬀerence between model complex 14 and CcO is worth noting: the selectivity of the four-electron reduction of O2 is >99% for CcO, but it is only 96% for 14 under limited electron transfer ﬂux comparable to conditions in the enzymatic cycle. Two plausible reasons are suggested by Collman and co-workers. First, only the fully reduced active site FeII-CuI in CcOs binds O2, whereas 14 may bind O2 at the FeIICuII state. Second, the active site of CcOs is buried in a membrane, whereas the active site of 14 is in direct contact with water. Hydrolysis of the partially reduced intermediate may thus occur to increase the release of PROS. Remarkably, treating the SAM ﬁlms with hydrophobic surfactants, which can act as hydrophobic blocking layers, reduced the formation of PROS. In 2009, Collman and co-workers studied the biologically relevant O2 reduction by using cytochrome c as the reductant and a synthetic CcO model complex as the catalyst.160 This new model complex 17 (Figure 13) is structurally quite similar to

Figure 14. Synthetic CcO model complexes 18−20 studied by Collman and co-workers to elucidate the role of the distal pocket environment.

this component (20) and compared their reaction behaviors for O2 reduction.161 The second-order O2-binding rate constants were 1 M−1 s−1 for 18, 4.3 × 108 M−1 s−1 for 19, and 5 × 107 M−1 s−1 for 20 in dichloromethane (DCM). Although O2 binding involves an electron transfer from FeII to form an FeIIIsuperoxo intermediate, this disparity of O2 binding rates is not due to the diﬀerent redox potentials (the redox potential is 87 mV for 18, 90 mV for 19, and 180 mV for 20 vs normal hydrogen electrode, NHE). It is obvious that the redox potentials of 19 and 20 are quite diﬀerent, but their O2-binding rates are in the same range. Moreover, very diﬀerent O2-binding rate constants are observed for 18 and 19, which have nearly identical redox potentials. Interestingly, a carefully dried sample of 18 displayed extremely fast binding of O2 (the rate constant was too fast to be measured). These results suggest that the slow O2 binding rate of 18 is caused by the presence of a cluster of water molecules hosted in its distal pocket with stabilization via hydrogen-bonding interactions with the N atoms of the imidazole groups. The hydrophobic distal pocket of 19 is unlikely to embrace such a water cluster. Two eﬀects of the water cluster are thought to explain why it inhibits O2 binding. First, the water cluster in the distal pocket around the Fe porphyrin coordinates to the FeII ion and blocks the entrance for the O2 molecule by inducing steric hindrance. A cluster of six water molecules exists in the pocket, suggesting a full occupancy of the distal pocket. Spectroscopic studies conﬁrmed that the Fe ion in 18 is six-coordinated with a Soret band at 426 nm in the UV−visible spectrum, but it is ﬁvecoordinated with the Soret band shifting to 435 nm upon

Figure 13. Synthetic model complex 17 and the proposed reaction mechanism for the catalytic O2 reduction by cytochrome c as reported by Collman and co-workers.

the previous model 14 (Figure 11) except that the terminal acetylene group covalently attached to the proximal trans imidazole ligand (used for the immobilization of the catalyst onto SAM-coated gold electrodes via azide-alkyne cycloaddition) is replaced by a triﬂuoromethyl group. Three redox centers are assembled together to provide the four electrons (two from Fe, one from Cu, and one from PhOH) required for the reduction of O2 to water. Such a Fe-Cu-PhOH functional model has been reported to be able to catalyze the selective four-electron reduction of O2 at physiological pH using an electrode as the electron source.159 By using the one-electron reductant cytochrome c, these authors showed that 17 could also catalyze the selective four-electron reduction of O2 to water in a homogeneous 1:1 aqueous buﬀer/acetonitrile solvent at pH 7. With the use of this homogeneous system, the reaction kinetics can be studied. The obtained pseudo-ﬁrst-order rate constant of 1.3 × 10−3 s−1 is much greater than that for the slow auto-oxidation of cytochrome c (ca. 1 × 10−5 s−1). In total, 3.9 ± 0.1 equiv of cytochrome c are oxidized per equivalent of O2 consumed, which indicates the selective four-electron reduction of O2 to water. A proposed reaction mechanism is displayed in Figure 13. (1) O2 ﬁrst binds to the FeII-CuI-PhOH catalyst to 3727

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

drying. Second, resonance Raman studies indicate that both 19 and 20 contain a mixture of high spin (HS) S = 2 and low spin (LS) S = 0 FeII, but 18 contains only LS FeII because of the strong donating eﬀect of water molecules. Therefore, there is a spin state barrier for 18 to bind O2; an LS FeII porphyrin is diamagnetic, whereas O2 is paramagnetic. Interestingly, if a pH 7 buﬀered-acetonitrile mixture was used to replace DCM as the solvent, an increase of O2-binding was observed for 18 (rate constant 30 M−1 s−1). This increase in O2-binding rates is due to the decreased solvation barrier of the water cluster in hydrophilic environments and due to the decreased free energy barrier for the removal of the Fe-bound water cluster (hydrogen bonding network inside the bound water cluster is extended to the external aqueous environment). These results together imply that eﬃcient water cluster removal in CcOs is critical for high reactivity, which is related to the importance of water pumping channels in the enzyme. Although water molecules in the distal pocket of synthetic analogues inhibit O2 binding, they can increase the selectivity toward the four-electron reduction of O2 by stabilizing the bound FeIII-superoxo unit as described by Collman and coworkers in a following work.162 Complex 20 binds O2 rapidly but produces the most PROS because it does not have a picket fence to stabilize the FeIII-superoxo unit. Complex 19 produces twice the amount of PROS as 18 generates. It is reasonable to expect that the water cluster in the distal pocket can form a stabilizing network of hydrogen bond interactions with the FeIII-superoxo unit. Theoretical calculations showed that water molecules are held above the superoxide and are located on the upper portion of the distal pocket through hydrogen bonding with the N atoms of the three imidazole groups. It is thought that the presence of a distal metal ion (redox active or not) can lead to similar water clusters through metal-H2O coordination because complexes 18, 18-CuI, and 18-ZnII all showed similar selectivity for electrochemical ORR. Boitrel and co-workers reported a series of biomimetic studies of CcO active sites with an emphasis on the distal coordination environment.163−166 Complexes 21 and 22 (Figure 15) contain quinolinoyl picket fences for CuI binding with either an external (for 21) or a tailed (for 22) nitrogen base for FeII binding.163 The three quinolinoyl pickets are thought to provide a ﬁxed separation between two metal centers and to protect the distal side of the porphyrin through steric hindrance. Both 21 and 22 bind O2 irreversibly in aprotic solvents, whereas their CuI-free forms bind O2 reversibly under identical conditions. Electrocatalytic results showed the following. (1) The Cu-free forms of 21 and 22 were more eﬃcient than the Fe-Cu forms in terms of the number of electrons transferred during ORR. In other words, the Fe-only complexes are better four-electron reduction catalysts than the Fe-Cu analogues, which catalyze both the two-electron and four-electron reductions. (2) The anchored imidazole does not improve the eﬃciency of O2 reduction compared with the external imidazole base as the trans axial ligand for the Fe ion. In addition to quinolinoyl picket fences, Boitrel and coworkers also used tripodal tetraamine pickets as the distal Cu coordination site (Figure 15).164−166 The Cu-free forms of complexes 23−25, which diﬀer only on the alkyl groups appended to the amine N atoms of the capped tris(2aminoethyl)amine (tren) motif, were ﬁrst examined to elucidate the electronic eﬀects of the distal cage on the catalyst activity for O2 reduction.164 The reduction overpotential was shown to have the order 25 > 24 > 23, implying that alkylation

Figure 15. Synthetic CcO model complexes 21−27 studied by Boitrel and co-workers to elucidate the role of the distal pocket environment.

of the secondary amine N atoms decreased the activity; in particular, alkylation by a benzyl group caused an ∼0.5 V increase of the overpotential. In addition, the Cu-free form of 23 dominantly catalyzed the four-electron reduction of O2, whereas the Cu-free forms of 24 and 25 catalyzed both the twoelectron and four-electron reduction of O2. These results suggested that, for these Cu-free catalysts, secondary amines were more eﬃcient than tertiary amines for catalytic O2 reduction. In a subsequent work, these authors showed that complex 23 catalyzed O2 reduction through both the twoelectron and four-electron pathways, which was unlike its Cufree form catalyzing the four-electron ORR selectively.165 These studies using Cu-free CcO synthetic models showed that Fe porphyrin is an intrinsically eﬃcient catalyst for O2 reduction. To further elucidate the electronic eﬀects of the distal cage on ORR catalysis, Boitrel and co-workers attached electrondonating and electron-withdrawing groups on the secondary amine N atoms of the capped tren motif.166 Complexes 25−27 and their corresponding Cu-free counterparts were synthesized and examined (Figure 15). The use of these models can reveal the electronic properties of the distal Cu binding site without any modiﬁcation of the basic backbone, steric hindrance, or altering the relative position of the two metal ions. When adsorbed on graphite electrodes, the Cu-free forms of 25 and 3728

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Interestingly, ORR studies by Dey and co-workers using Fe porphyrins 30 and 31 (Figure 16) showed that the PROS generated with 31 increased upon decreasing the electron transfer rate (a trend as expected), but the PROS generated with 30 reduced upon decreasing the electron transfer rate (an unusual trend).168 The results are summarized in Table 1. This

27 are eﬃcient catalysts for the four-electron reduction of O2 in a pH 6.86 aqueous solution, whereas the Cu-free form of 26 catalyzes the two-electron reduction of O2. However, these catalysts are not stable during catalysis, preventing Koutecky− Levich analysis to accurately determine the number of electrons transferred for the ORR. In contrast to the Cu-free forms, complex 26 does not exhibit catalytic activity, 25 exhibits poor catalytic activity and produces almost exclusively H2O2, and 27 has modest catalytic activity and mainly catalyzes the fourelectron reduction of O2 to H2O. These results further conﬁrm that for synthetic models adsorbed on electrodes, Cu does not improve the catalysis compared with Fe-only analogues. 2.2.4. Role of the Trans Ligand of Fe in O2 Reduction. Dey and co-workers investigated the eﬀect of the trans axial ligation of Fe on ORR. Fe porphyrins with a covalently attached thiolate ligand (28) or phenolate ligand (29) have been synthesized and examined as ORR catalysts (Figure 16).167 Complex 28 can be physiadsorbed on graphite

Table 1. Percentage of PROS Formed by 30 and 31 in AirSaturated pH 7 Buﬀers under Fast (EPG), Moderate (C8SH SAM on Au), and Slow (C16SH SAM on Au) Electron Fluxes percentage of PROS formed (%) Fe porphyrin

EPG

C8SH SAM on Au

C16SH SAM on Au

30 31

19 ± 1.5 3.5 ± 1

15.5 ± 0.5 10 ± 0.5

11 ± 1 16 ± 1

diﬀerence suggests that diﬀerent reaction mechanisms are involved for these two complexes. For 31, the hydrolysis of FeIII-O2•− competes with electron transfer; as the electron transfer rate reduces, the hydrolysis rate, which remains constant, dominates and produces more PROS. For 30, its unusual behavior is likely due to the presence of an additional O2 reduction pathway by an LS FeII species in addition to the normal four-electron reduction pathway catalyzed by an HS FeII species (Figure 17). Because the replacement of the aqua

Figure 17. Outer sphere reduction of O2 with a six-coordinated LS FeII porphyrin and the inner sphere reduction of O2 with a ﬁvecoordinated HS FeII porphyrin. Figure 16. Fe porphyrins 28−31 studied as ORR catalysts by Dey and co-workers to elucidate the role of the trans ligand.

ligand of the LS FeII species by an O2 molecule is expected to be thermodynamically unfavorable (a slow step) and because the outer sphere O2 reduction by LS FeII species generates O2•−, the formation of PROS thus reduces as the electron transfer rate decreases because less LS FeII species will accumulate on the electrode surface. The H/D isotope eﬀects on catalytic ORR were reported by Dey and co-workers with the use of Fe porphyrins 28 and 31 (Figure 16).169 When adsorbed on EPG electrodes, both 28 and 31 catalyzed the four-electron reduction of O2 to H2O with kinetic isotope eﬀect (KIE) values of 18 and 1.04, respectively. In combination with experiments using a rotating ring-disk electrode (RRDE) to determine the percentage of formed PROS, it is possible to estimate the KIE of the two-electron ORR with 28 and 31 as 47 and 4.7, respectively. These electrochemical data suggested that the reaction mechanisms for 28 and 31 are diﬀerent. It is proposed that the rds with 28 is the O−O bond heterolysis of FeIII-OOH species, and the rds with 31 is the O−O bond heterolysis of FeII-OOH. This work further shows that diﬀerent axial ligands can lead to distinct rds involved in O2 reduction. The push eﬀect of the trans axial

electrodes or it can be covalently attached to azide-terminated SAMs on Au electrodes for ORR activity studies. Electrochemical studies using edge-plane pyrolytic graphite (EPG) electrodes show that both 28 and 29 can catalyze the fourelectron reduction of O2. The reduced FeII species is the catalytic active species, and the second-order rate constants for O2 reduction by 28 and 29 are (5.6 ± 1) × 106 and (3.8 ± 0.1) × 105 M−1 s−1, respectively. On the basis of this work, it is obvious that Fe porphyrins bearing axial thiolate ligands are much more active than those bearing axial phenolate ligands for O2 reduction. This diﬀerence is most likely caused by the “push eﬀect” of thiolate ligands in accelerating O−O bond cleavage. When covalently attached to the SAM of Au electrodes, 13 ± 1% PROS are generated during O2 reduction with 28. The increased amount of PROS for O2 reduction on the SAM-Au electrode is due to the hydrolysis of FeIII-O2•− under slower electron ﬂux conditions. 3729

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

OOH and thus more PROS; (2) the greater hydrophilicity of the distal pocket increases the hydrolysis of FeIII-O2•− and/or FeIII-OOH to generate more PROS; and (3) additional electron donors decrease the formation of PROS. The previously mentioned eﬀects of the axial ligand of Fe porphyrins on catalytic ORR, as studied by Dey and co-workers, have been reviewed by these authors.171 The diﬀerent electronic structures and reactivities of Fe porphyrins with imidazole, phenolate, and thiolate axial ligands are discussed based on the results from spectroscopy and density functional theory (DFT) calculations. The distinct “push eﬀect” of these axial ligands is the major cause of the diﬀerent kinetics and selectivity of O2 reduction catalyzed by Fe porphyrins. Very recently, Dey, Karlin, and co-workers used surfaceenhanced resonance Raman spectroscopy (SERRS) to investigate the synthetic heme-Cu CcO model complex 4 (Figure 6).172 The catalytic ORR activity of 4 has been thoroughly studied under homogeneous conditions,150 but the combination of SERRS and the rotating disk electrode (RDE) technique allows direct in situ investigations of the reactive intermediates formed under the steady state conditions to better understand the reaction mechanism of electrocatalytic O2 reduction. As revealed by these authors, complex 4 and its Cu-free form can both selectively reduce O2 to H2O under fast electron ﬂux, whereas the imidazole-ligated 4 is less selective for this four-electron reduction of O2 under fast electron ﬂux. Interestingly, the four-electron reduction selectivity of imidazole-ligated 4 increases with decreased electron ﬂux, which is in contrast to 4, its Cu-free form, and most other reported Fe and Fe-Cu systems (an exception is discussed above, see complex 30 for details168). SERRS data show that the Cu-free form of 4 under the resting oxidized conditions contains a ﬁve-coordinated HS FeIII center and becomes an HS FeII upon reduction; complex 4 at resting oxidized conditions contains a six-coordinated LS FeIII center and becomes an HS FeII upon reduction. The imidazole-ligated 4 at resting oxidized conditions contains a six-coordinated LS FeIII center and becomes a major LS FeII upon reduction. The number of electrons transferred during ORR catalysis is almost 4 for both 4 and its Cu-free form and is 3.4 for imidazole-ligated 4 under fast electron ﬂux conditions. The second-order rate constants are (3.65 ± 1.1) × 106, (4.49 ± 0.9) × 106, and (1.28 ± 0.11) × 106 M−1 s−1 for the Cu-free form of 4, 4, and imidazole-ligated 4, respectively. The correlation of ORR selectivity and electron transfer rates for these three catalysts has been evaluated, and the results are summarized in Table 3. As expected, 4 and its Cu-free form show the general trend of increasing PROS generation with a decrease in electron ﬂux, and the “extra” redox center (i.e., Cu ion) compensates for this eﬀect under slow electron ﬂux conditions. However, the unusual trend observed with the imidazole-ligated 4 is unexpected.

thiolate ligand increases the electron density at the Fe center and increases the pKa of the FeIII-OOH species for facile protonation and thus O−O bond cleavage. However, for the trans axial imidazole ligand, the pKa of the FeIII-OOH species is too low to allow the protonation and O−O bond cleavage; consequently, it must be reduced to FeII-OOH for the protonation and O−O bond cleavage. Another example elucidating the eﬀects of the trans axial ligation and the distal environment on ORR was reported by Dey and co-workers with Fe porphyrins 32, 33, and 34 (Figure 18).170 The Au electrode was ﬁrst modiﬁed with SAM bearing

Figure 18. Immobilization of Fe porphyrins 32-34 onto SAMmodiﬁed Au electrodes to elucidate the eﬀects of both the trans axial ligation and the distal environment on the ORR.

either terminal imidazole or thiol groups to bind Fe porphyrins. Cyclic voltammograms (CVs) of these Fe porphyrins revealed that the FeIII/FeII process shifted to the negative potentials for the thiolate ligation relative to the imidazole ligation, which is consistent with the stronger electron-donating character of anionic thiolate ligands. In addition, pH dependence studies revealed that at pH 7, the trans ligand for imidazole ligation is a hydroxyl group (Imd-FeIII-OH), whereas it is a trans aqua ligand for thiolate ligation (S-FeIII-OH2). O2 reduction using catalysts immobilized with thiolates occurred at more negative potentials relative to reduction using catalysts immobilized with imidazoles, a result that is consistent with the more negative reduction potentials of the thiolateligated Fe porphyrins. The amount of PROS produced during O2 reduction is summarized in Table 2. Three main conclusions can be made based on this table: (1) thiolate ligation leads to greater hydrolysis of FeIII-O2•− and/or FeIII-

Table 3. PROS Values (%) Detected in the ORR with Catalyst 4 and Its Derivatives at pH 7

Table 2. PROS Values (%) Detected in the ORR with Catalysts 32, 33, and 34

PROS determined under diﬀerent conditions

catalyst

imidazole ligation

thiolate ligation

catalyst

EPG (fast electron ﬂux)

32 33 34

9±1 10 ± 2 5±1

10.5 ± 1 17 ± 2 9.8 ± 1

Cu-free form of 4 4 imidazole-ligated 4

6±1 6.3 ± 0.1 25 ± 2

3730

C8SH SAM Au (moderate electron ﬂux)

C16SH SAM Au (slow electron ﬂux)

9 ± 0.5 7 ± 0.1 10 ± 1

23 ± 3 13.5 ± 1 4±1

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

aﬀord an additional (the fourth) electron to reduce O2 to water and temporarily store the resulting oxidizing equivalent as a tyrosine radical. Implications from the biomimetic chemistry of CcOs suggest that electron transfer, proton transfer, and water transfer should be concerted to obtain high reactivity and selectivity for the catalytic reduction of O2. The following factors may be valuable for the design and development of new ORR catalysts. (1) The extra redox centers attached to the heme with appropriate distance and orientation can rapidly provide electrons to favor the four-electron reduction of O2 to water. (2) The electron transfer ﬂux from electrodes or reductants (internal or external) to catalysts is the major decisive factor for the two-electron versus four-electron reduction of O2. (3) Reaction media have noteworthy eﬀects on O2 reduction. The water cluster in the distal pocket has dual functions, inhibiting O2 binding to the Fe porphyrin and stabilizing the FeIII-superoxo unit to suppress the release of PROS. (4) The hydrolytic release of superoxide is one of the major pathways leading to PROS formation. Stabilization of this superoxo unit (via picket fences, distal metal ions, or other molecular superstructures) is therefore critical for the selective four-electron reduction of O2. (5) The proximal trans axial ligand increases the electron density on heme Fe (via the push eﬀect), thus assisting O2 binding and subsequent inner sphere electron transfer from FeII to O2, which may be the rds in the O2 reduction catalytic cycle.

On the basis of SERRS data under the steady-state catalytic ORR conditions, the formation of a Fe-peroxo intermediate was suggested, but the FeIVO species cannot be identiﬁed by either SERRS measurements or [Fe(CN) 6]3−/4− assay, indicating that no FeIVO species accumulates during O2 reduction. The stronger O−O bond observed for imidazoleligated 4 compared to 4 indicates the formation of an end-on bridging peroxo unit for imidazole-ligated 4 and the formation of a side-on bridging peroxo unit for 4. The relatively stronger σ-donating imidazole ligand can facilitate the formation of an end-on bridging peroxo. As in SERRS-RDE experiments, species with faster formation and slower decay rates will accumulate and can be detected; it is suggested that the heterolytic O−O bond cleavage is a slow step. Similar to complex 30 (Figure 16),168 a six-coordinated LS FeII was the major form of imidazole-ligated 4 at resting reduced conditions. Because ferrous porphyrins in the LS ground state are known to have sluggish rates of O2 binding, under fast electron ﬂux conditions, the distal Cu reduces O2 to produce PROS, but under slow electron ﬂux conditions, O2 binding occurs on the LS FeII and the 4H+/4e− ORR proceeds. The ferrous center in its LS state may also reduce O2 via outer sphere one-electron reduction under fast electron ﬂux conditions. 2.2.5. Implications from Synthetic Models. On the basis of these aforementioned studies, a reaction mechanism for O2 reduction to water mediated by a fully reduced CcO (Fea3II, CuBI, FeaII, and [CuICuI]A) is proposed (Figure 19). The

2.3. Simple Oxygen Reduction Reaction Catalysts

2.3.1. Iron-Based ORR Catalysts. Mayer and co-workers used a series of Fe porphyrins to explore the factors that determine catalytic ORR activity and selectivity (Figure 20).173−176 With simple Fe tetraphenylporphyrin (35), a direct

Figure 20. Fe porphyrin chloride complexes 35−40 investigated as ORR catalysts by Mayer and co-workers.

comparison is made between electrochemical and spectrochemical catalysis of O2 reduction,173 which are two common but diﬀerent methods used for catalytic ORR studies. In electrochemical studies, “foot-of-the-wave” (FOW) analysis was used. It is worth noting that to apply FOW analysis, data from a “well-behaved” electrocatalysis region should be used, indicating that the catalyst is relatively stable under catalytic processes and that the catalysis is limited only by chemical steps. In spectrochemical studies, a stopped-ﬂow technique was used to examine catalytic O2 reduction in the presence of HClO4, Me10Fc, and 35. This study shows that catalytic ORR under electrochemical and spectrochemical conditions has similar kinetics and selectivity; thus, a direct comparison of the two kinetic methods is possible in this work. The four-electron reduction of O2 with 35 is conﬁrmed in both electrochemical and spectrochemical studies. In both systems, electron transfer steps

Figure 19. A proposed O2 reduction reaction mechanism with a fully reduced CcO, showing the roles of heme a3, CuB, Tyr, heme a, and CuA.

mixed-valence enzyme can reduce O2 to water in a very short time scale of Sr2+ > Ca2+ > none (Na+) > Mg2+. These data suggest that Ba2+ can facilitate the formation of face-to-face Fe porphyrins. The Fe···Fe separation in a face-to-face structure in the presence of bridging Ba2+ ions can be roughly estimated at 4.4 Å using the ionic radii of FeIII (0.7 Å) and Ba2+ (1.5 Å). The resulting cavity is adequate to accommodate one O2 molecule (3.4 Å). An example of O2 reduction using biological reductants and Fe porphyrin catalysts was reported by Gröger and co-workers in 2011.197 The examined Fe porphyrin catalysts have good solubility in aqueous solutions, and the reductant is the twoelectron reducing agent NAD(P)H. The results show that complexes 47−49 (Figure 27) are not active in catalyzing the

Figure 26. Proposed reaction mechanism of O2 reduction in organic solvents catalyzed by 32. Redrawn from ref 192. Copyright 2013 American Chemical Society.

be stabilized by hydrogen-bonding interactions in the triazolebased distal pocket. This O2 adduct is diamagnetic, which indicates the formation of an end-on S = 0 Fe-O2 adduct. Two O−O vibrations at 1004 and 967 cm−1 were also observed, and these vibrations shifted to 951 and 911 cm−1 upon 18O-labeling and to 996 and 961 cm−1 upon D-labeling. The unusually high Fe-O2 bond vibration and low O−O bond vibrations imply strong Fe-O2 bonds and weak O−O bonds, which is consistent with an FeIII-O2•− description instead of FeII-O2. DFT calculations showed that hydrogen bonding interactions polarized the Fe-O2 unit, which shortened the Fe-O2 bond by 0.04 Å and stretched the O−O bond by 0.03 Å. Consequently, the calculated Fe-O2 bond vibration increased by 27 cm−1 and the O−O bond vibration decreased by 70 cm−1. Importantly, with the use of rotating disk electrochemistry coupled to resonance Raman, Dey and co-workers reported in situ spectroscopic detection of Fe porphyrin intermediates formed on the electrode surface during steady-state electrocatalytic ORR.194 SERRS data collected for 33 on the electrode held at −0.5 V (vs Ag/AgCl) in the presence of O2 showed the formation of a high-spin FeII, both high-spin and low-spin FeIII, and a high-valent FeIVO species during steady-state O2 reduction. As more negative potentials were applied, the electrocatalytic O2 reduction current increased and changed from being potential dependent to being mass transfer limited. Accordingly, the high-spin FeII, the low-spin FeIII, and the FeIV species gradually accumulated at the electrode, whereas the amount of high-spin FeIII species decreased. These experiments were also performed with 18O2 to identify the formation of FeOx species derived from O2 reduction. However, because of the fast exchange of the ferryl oxygen with water, the formed FeIV18O will quickly exchange with H216O to produce FeIV16O, which makes the detection of isotopically labeled FeIV18O very challenging. In addition to SERRS data, the formation of high-valent FeIVO was also evaluated using the [Fe(CN)6]3−/4− assay in RRDE. The observation of FeIVO implies that its reduction to FeIII-OH is slow on the electrode.

Figure 27. Water-soluble Fe porphyrin chloride complexes 47−49 studied as catalysts for catalytic O2 reduction by NAD(P)H.

oxidative transformation of NAD(P)H into NAD(P)+ in aqueous media with O2, which may be due to the presence of sterically demanding (charged) substituents anchored on these Fe porphyrins. However, structurally simpliﬁed Fe porphyrin 46 (Figure 24) can catalyze the oxidation of NAD(P)H to NAD(P)+ in the presence of O2 with turnover frequency (TOF) values of 0.11 s−1 for the oxidation of NADH and 0.06 s−1 for the oxidation of NADPH. Subsequent test experiments revealed that no H2O2 was detected in the system, indicating the four-electron reduction of O2 to water. A catalytic cycle for the oxidation of NAD(P)H into NAD(P)+ by O2 was proposed (Figure 28). FeIII porphyrin ﬁrst reacts with an NAD(P)H to form the FeIII hydride complex, which rapidly reacts with an O2 molecule to give the hydroperoxo FeIII(OOH) complex. Subsequent protonation and heterolytic O− 3736

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

reduction of O2, and the second step produced 35% fourelectron reduction of O2. The authors speculated that there were two parallel reaction pathways involved in the ORR with 50: the reduction of one O2 molecule by two CoII centers yielding the four-electron reduction pathway and reduction by a single CoII center yielding the two-electron reduction pathway. This unusual ability of a monomeric Co porphyrin to catalyze the four-electron reduction of O2 is likely a result of the strong tendency of the porphine to form dimeric, cofacially disposed rings through van der Waals interactions. In contrast to this simplest Co porphyrin, Anson and coworkers examined the ORR activity and selectivity of Co porphyrin 51 with a “picket fence” superstructure (Figure 29).199 The original objective of using complex 51 was to stabilize and thus observe the catalytically important CoII-O2 adduct. Although the four “pickets” appended at the porphyrin ring aﬀorded unusually high aﬃnity for O2, they hindered (rather than enhanced) the catalytic electroreduction of O2. By using an imidazole ligand to occupy the axial coordination site on the unencumbered side of the porphyrin ring, the aﬃnity of 51 for O2 binding was enhanced substantially. The O2 adduct of 51 could be CoII-O2 or CoIII-O2•−; the latter is more plausible based on spectral evidence. Electrocatalytic ORR studies using 51 and its all-meso-phenyl analogue 52 (Figure 29) showed that 51 was much less active. One reason is that the rate of O2 coordination to the CoII center is evidently slower for 51 than for 52. Therefore, the addition of “pickets” enhanced the aﬃnity for O2 but did not improve the catalytic ORR. Fukuzumi and co-workers investigated the reduction of O2 catalyzed by four Co porphyrins 52−55 (Figure 29).200 Electrochemical studies revealed that complexes 52 and 54 showed the most positive onset potential of 0.3 V versus SCE, complex 53 showed modest activity with an onset potential of 0.1 V versus SCE, and complex 55 was not active for ORR under the examined conditions. The CV of 54 displayed two well-separated reduction peaks of approximately 0 V and −0.3

Figure 28. Proposed reaction mechanism of the in situ cofactor regeneration of NAD(P)+ catalyzed by Fe porphyrin 46. Redrawn with permission from ref 197. Copyright 2011 Wiley-VCH.

O bond scission (to eliminate a water molecule) generate the FeIVO π-cation porphyrin radical species, a process similar to the O2 activation mediated by cytochrome P450 monooxygenases. The ﬁnal step consists of the regeneration of the FeIII porphyrin by NAD(P)H reduction. Signiﬁcantly, this catalytic in situ generation of NAD(P)+ cofactor was coupled to the biotransformation of D-glucose into D-gluconolactone or the transformation of alcohol into the corresponding ketone in aqueous media. 2.3.2. Cobalt-Based ORR Catalysts. The other family of extensively investigated metal porphyrin ORR catalysts is Co based. In 1997, Anson and co-workers reported the ORR catalytic activity of the simplest Co porphyrin, Co porphine (50, Figure 29).198 In the presence of O2, the reduction occurred in two well-separated steps; the ﬁrst step (0.53 V vs SCE) was near the CoIII/CoII potential, and the second step showed a plateau current. Signiﬁcantly, the current of the second step was considerably larger than the diﬀusion-limited current for the reduction of O2 to H2O2. RRDE experiments showed that the ﬁrst step produced >90% four-electron

Figure 29. Co porphyrin complexes 50−55 studied as ORR catalysts. 3737

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 30. Co porphyrin complexes 56−59 studied by Kadish and co-workers.

V (vs SCE). The ﬁrst reduction peak corresponds to the twoelectron reduction of O2 to H2O2, and the second reduction peak is responsible for the reduction of H2O2 on the catalystcoated electrode. In 2014, Kadish, Ou, and co-workers reported the synthesis, electrochemistry, and catalytic ORR property of two Co porphyrin complexes each bearing one β-pyrrole −NO2 substituent.201 The syntheses of these β-pyrrole nitrosubstituted porphyrin ligands were ﬁrst described by these authors.202 The introduction of one −NO2 group to the βpyrrole position renders the porphyrins more easily reduced and less easily oxidized compared with the corresponding nitrofree counterparts, a feature which is favored for ORR catalysis. The electrochemical behaviors of Co porphyrins 57 and 59 (Figure 30) and their corresponding nitro-free counterparts 56 and 58 were investigated. In electrocatalytic ORR studies, the onset potentials for 57 and 59 shifted in the positive direction by 60 mV compared to their nitro-free counterparts, and the number of electrons transferred during ORR was 3.1 for 57 and 59 and 2.8/2.6 for 56 and 57, respectively. All these results indicate that Co porphyrins with one strong electronwithdrawing −NO2 substituent at the β-pyrrole position are better catalysts than the nitro-free counterparts for the electroreduction of O2. Similar to Fe porphyrins, water molecules also display an inhibiting eﬀect on Co-porphyrin-catalyzed ORR. By using Co meso-tetraphenylporphyrin (52, Figure 29), Samec and coworkers demonstrated this inhibitory eﬀect in the chemical reduction of O2, which is considered to be a result of the competitive coordination of water to the Co center.203 Electrochemical studies of 52 revealed a CoIII/CoII redox couple at 0.40 V (vs ferrocene) in 1,2-dichloroethane (DCE). However, the CoIII/CoII redox couple in acetonitrile was found at −0.02 V (vs ferrocene). This potential shift indicates that

CoIII reduction is more diﬃcult due to the axial coordination of acetonitrile. UV−visible measurements showed that the gradual addition of water to an air-saturated DCE solution containing complex 52 and tetrakis(pentaﬂuorophenyl)boric acid (HTB) caused a red-shift of the Soret band from 401 to 428 nm and an increase of the Q-band at 541 nm. These results suggest that the axial coordination of the Co center will have substantial eﬀects on the electrochemical and spectrochemical behaviors. Subsequent ORR studies revealed that in the presence of acid HTB and external reductant Fc, complex 52 could catalyze the two-electron reduction of O2 to H2O2 in DCE. The initial reaction rate was independent of the Fc concentration but was proportional to the concentrations of 52 and HTB. Signiﬁcantly, the reaction was remarkably slowed in the presence of water. These results together suggest that the rds is the proton-assisted coordination of O2 to the Co center. On the basis of DFT calculations, the coordination complex 52(H2O)4 (the tetramer (H2O)4 is considered as a representative of the general water aggregate) is much more stable than 52O2. It is interesting to compare the kinetic ORR behaviors in DCE and in acetonitrile. Unlike in DCE in which the coordination of O2 is the rds, in acetonitrile, the reduction of CoIII to CoII by Fc is the rds. This diﬀerence is caused by the diﬀerent formal CoIII/CoII redox potentials in DCE (0.40 V vs ferrocene) and in acetonitrile (−0.02 V vs ferrocene); CoIII reduction is more diﬃcult due to the axial coordination of acetonitrile. The estimated acceleration of the reduction of CoIII by Fc has a factor of 2.4 × 103. Consequently, the reduction of CoIII by Fc was identiﬁed as the rds of the catalytic cycle in acetonitrile, but this was not the rds in DCE. Girault and co-workers investigated the electrocatalytic O2 reduction at liquid−liquid interfaces to mimic the biological O2 reduction that occurs at membranes.204−209 To realize such a biomimetic process, an interface between two immiscible 3738

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

electrolyte solutions (ITIES) is used, which can provide a physical separation of the reactants and products and can allow an electrochemical control for diﬀerent charge transfer reactions through the polarization of this soft interface (i.e., ion transfer, assisted ion transfer, and heterogeneous electron transfer reactions). In a typical ITIES system, reactants are physically separated by the liquid−liquid interface with protons in the aqueous phase and catalysts and reductants in the organic phase. Voltammetry at the ITIES for the catalytic reduction of O2 records the current due to the transport of protons from the aqueous phase to the interfacial reaction site, a process in which proton transfer and electron transfer are tightly coupled. It is worth mentioning that in contrast to voltammetry at the ITIES, classical voltammetry using a solid electrode measures a current response associated with electron transfer rather than ion (i.e., proton) transfer. Therefore, voltammetry at the ITIES provides a very eﬃcient method to investigate the PCET process of the ORR by controlling the applied Galvani potential diﬀerence and thus the rates of either proton or electron transfer across the interface. Compared to ORR in bulk systems, this interfacial system allows the rapid delocalization of generated H2O2 from the organic phase to the aqueous phase, thus decreasing the possibility of both the degradation of catalyst by H2O2 and the further reduction of H2O2 by reductants. Electrochemical ORR at the ITIES was studied by using Co porphyrins 50, 52, and 53 (Figure 29) as catalysts and Fc derivatives as the reductants.204,205,207 In the presence of both 50 and Fc,204 a large irreversible current wave was observed at positive potentials, indicating proton transfer across the interface. Importantly, no current due to the back transfer of H+ from organic to aqueous phase was observed on the return scan, but a current wave for Fc+ transfer could be observed on this return scan, indicating the consumption of H+ and the formation of Fc+ at the positively polarized interface. A steady increase of this Fc+ ion transfer wave was observed upon repetitive cycling, which implies the constant formation of more Fc+ ions during this process. In addition to this electrochemical ORR study with an applied Galvani potential diﬀerence, a two-phase ORR experiment was also performed, which conﬁrmed the catalytic reduction of O2 by Fc in the presence of Co porphine 50. However, the amount of H2O2 detected corresponded to a yield of 18% according to a two-electron reduction of O2. Importantly, with the addition of more Fc, the yield of H2O2 further decreased. These results suggest either a direct 4e− ORR to H2O or a 2e− ORR to H2O2, which is further reduced by Fc. The results from comparative tests revealed that H2O2 formed in situ could be reduced by Fc. However, the direct four-electron reduction pathway cannot be completely ruled out at this stage because catalyst 50 may form dimers at the liquid−liquid interface in a face-to-face geometry or may form one-dimensional polymers, which have been shown to catalyze the electrochemical four-electron reduction of O2 to H2O.198 On the basis of these results, a reaction mechanism was proposed by these authors (Figure 31). Kinetic studies showed that in the presence of excess protons and reductants, the reaction rate was limited by the potential-dependent PCET step involving CoIII-O2•−, H+, and Fc. Other steps, such as the reduction of CoIII state by Fc and the formation of the O2 adduct, are bulk reactions and are therefore potential independent.

Figure 31. Proposed O2 reduction mechanism with mononuclear Co porphyrin catalysts in the ITIES system. Redrawn from ref 204. Copyright 2009 American Chemical Society.

Girault and co-workers further investigated the catalytic ORR behaviors of an amphiphilic Co porphyrin catalyst 60 (Figure 32) that can be adsorbed at the ITIES interface.206 The introduction of a hydrophilic meso-4-aminophenyl group to the porphyrin ring provides the resulting catalyst 60 with strong aﬃnity for the water−DCE interface. The CVs of 60 in ITIES showed a large capacitive current compared to the background current, indicating the formation of an adsorbed layer. This adsorption is due to the presence of both the amino group as the hydrophilic moiety and the metalated porphyrin ring as the lipophilic part. Lipophilicity mapping calculations and surface tension measurements also conﬁrmed this adsorbed layer and suggested that the area occupied by each molecule was approximately 161 Å2 and that the adsorbed 60 thus had an orientation normal to the interface with the hydrophilic amino group toward the aqueous phase. The catalytic ORR with 60 and Fc was veriﬁed by both voltammetric and biphasic experiments. H2O2 and Fc+ were produced in water and DCE, respectively. A comparison of the catalytic current of 60 and those of 50, 52, and 53 revealed that the amphiphilic Co porphyrin 60 had much better electrocatalytic activity. Particularly, both 60 and 53 have the same redox potential for the CoIII/CoII couple (0.69 V vs SHE), but 60 is much more eﬃcient than 53 in catalyzing the reduction of O2. The strong interfacial aﬃnity of 60 is believed to account for its excellent catalytic activity because the proton-coupled O2 reduction is an interfacial process. In the above-mentioned ITIES examples, O2 is generally reduced by two electrons to generate H2O2 except with catalyst 50, from which the majority of O2 is reduced by four electrons to form H2O. Catalyst 50 may form dimers at the liquid−liquid interface in a face-to-face geometry, or it may form onedimensional polymers, which are considered to be crucial for the 4e− ORR by using monomeric Co porphyrin catalysts. To study the catalytic ORR behaviors of interfacial self-assembled porphyrins, Girault and co-workers used a water-soluble cationic Co porphyrin catalyst 61 (Figure 32).208 Instead of using Fc derivatives as the reductants, the lipophilic electron donor tetrathiafulvalene (TTF) was used in this study. Unlike Fc derivatives, TTF does not react with H2O2, allowing a more accurate quantiﬁcation of the O2 reduction products. In contrast to the ITIES systems, the catalyst and the reductant are now located in aqueous and organic phases, respectively; 3739

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 32. Co porphyrins 60−62 as ORR catalysts studied in the ITIES system.

superoxo bridge complex between two CoIII centers. (3) Protons are required for the ORR process. Control experiments showed that the reaction was slow in the absence of acids, indicating that the protonation of the μ-superoxo unit was a key step to achieve catalytic eﬃciency. Therefore, these studies from Girault and co-workers demonstrate that it is possible to employ well-designed molecular or supramolecular assemblies to conduct selective reactions in ITIES systems. In addition to DCE, a variety of organic solvents including 1,2-dichlorobenzene could be used to construct an ITIES system with water for catalytic ORR as reported by Murtomäki and co-workers.210 Compared to DCE, the use of 1,2dichlorobenzene as the organic phase is favored by these authors because of its lower vapor pressure. The catalytic ORR properties of Co corroles have also been investigated. In 2012, Kadish, Ou, and co-workers examined ﬁve meso-substituted Co corroles 63−67 (Figure 33) as ORR

consequently, the ORR is expected to occur only at the interface where 61 is adsorbed. Interestingly, the selectivity of 61 for the four-electron reduction of O2 is remarkably enhanced by adding tetrakis(pentaﬂuorophenyl)borate to the initial organic phase. The results from surface second harmonic generation (SSHG) revealed that the presence of tetrakis(pentaﬂuorophenyl)borate in the organic phase signiﬁcantly enhanced the interfacial selfassembly of 61 with an increase in the number density of porphyrins and its ﬁrst hyperpolarizability at the interface, which is most likely due to the electrostatic interaction between positively charged 61 and negatively charged tetrakis(pentaﬂuorophenyl)borate anions. A possible face-to-face stack of molecules of 61 to form a sandwich-type arrangement may occur for interfacial self-assembled porphyrins, in which the distance between two Co atoms is appropriate to drive the four-electron reduction of one O2 molecule to form water. Girault and co-workers also investigated the catalytic ORR properties of interfacial self-assembled oppositely charged water-soluble Co porphyrins in the ITIES system.209 Positively charged Co porphyrin 61 and negatively charged Co porphyrin 62 (Figure 32) are able to form molecular “rafts” at the waterDCE interface, and these rafts can catalyze the interfacial fourelectron reduction of O2 by TTF with very high eﬃciency and a selectivity of 78%. It is worth noting that complex 61 alone has intermediate activity and selectivity, and 62 alone has poor activity and selectivity for the direct four-electron reduction of O2. The interfacial self-assembly of 61 and 62 has been characterized by UV−visible spectroscopy, SSHG, and scanning electron microscopy (SEM). The number of adsorbed molecules at the interface has the following trend: [61-62] > 61 > 62. This trend is parallel to the catalytic activity and selectivity observed for the reduction of O2, indicating that the interfacial ORR performance strongly depends on the adsorption of catalysts at the interface. Several considerations are worth mentioning. (1) Biphasic reactions using the free bases of 61 and 62 showed signiﬁcantly reduced activity and selectivity, indicating that the formation of interfacial self-assembled 61 and 62 with a face-to-face structure and an appropriate Co···Co distance is crucial for the high activity and selectivity toward the direct four-electron reduction of O2. (2) DFT calculations conﬁrmed that the formation of a stable [61−62] assembly is possible. The calculated Co···Co distance for the CoIII−CoIII state of the [61−62] assembly is 3.93 Å, and the Co···Co distance is shorter by ∼0.5 Å in the CoIII−CoII state of the [61−62] assembly. These distances are appropriate for the four-electron reduction of O2 as found in synthetic Pacman cofacial Co porphyrins. Upon O2 binding to the CoIII−CoII state, the Co···Co distance increases to 4.41 Å with an O−O bond distance of 1.29 Å. The calculated spin density located on the central Co2O2 unit corresponds to a μ-

Figure 33. Co corroles 63−72 studied as ORR catalysts by Kadish and co-workers.

electrocatalysts.211 With these complexes, the authors are able to systematically study the eﬀect of electron-donating and electron-withdrawing meso-substituents on the electrocatalytic activity. When adsorbed on an EPG electrode, the CVs of 63− 67 showed well-deﬁned cathodic currents assigned to the reduction of O2. The peak potentials Epc in the range between 0.10 and 0.15 V (vs SCE) are approximately 260 mV more positive compared to that using a bare electrode (Epc = −0.13 V vs SCE). 3740

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

result in a negative shift of the half-wave potential for O2 reduction. The number of electrons transferred during O2 reduction was 2. This value was also conﬁrmed by the homogeneous chemical ORR in benzonitrile containing HClO4 and Me2Fc, revealing a two-electron reduction process of O2 to H2O2. Interestingly, the number of electrons transferred using Co corroles with β-pyrrole substituents is 2.9,216 indicating the presence of both two-electron and four-electron reduction processes. It is generally accepted that the formation of dimers (or higher oligomers) of Co porphyrins and corroles with a face-to-face orientation and appropriate Co···Co separation for O2 binding is crucial for the four-electron reduction pathway. The steric hindrance caused by the bulky meso-substituents of 72 may prevent the dimerization of Co corroles; thus, only the two-electron reduction pathway was observed. Collman and co-workers further investigated the electrocatalytic ORR with Co corrole 72 and its Fe counterpart.217 Two main conclusions can be made based on this work. First, ORR catalysis begins at potentials that are 0.5−0.7 V more positive than the potential of the MIII/MII (M = Co or Fe) couple, conﬁrming that the formal MIII state is the active species for O2 reduction. Second, Co corrole 72 catalyzes O2 reduction to H2O2, whereas its Fe counterpart catalyzes O2 reduction via parallel two-electron and four-electron pathways with the four-electron reduction dominating, a result that is similar to Co and Fe porphyrins. Because MIII is not expected to bind O2, it is unusual that the formal MIII state is the active species for O2 reduction. Inhibition experiments using NaCN revealed that O2 activation occurred at the metal center because the reduction waves were strongly suppressed in the presence of CN−. It is proposed by these authors that an intramolecular corrole-to-metal electron transfer happens (i.e., [(Cor3−)M3+] ⇌ [(Cor•2−)M2+]), and the MII center is able to bind and activate the O2 molecule. In the presence of CN−, this intramolecular corrole-to-metal electron transfer is not favored because of the stabilization of the MIII state by CN−. In addition, the reduction waves at more negative potentials are not sensitive to CN− because the resulting MII center in [(Cor3−)M2+]− has a low aﬃnity for CN−. Gross and co-workers reported the electrocatalytic ORR activity of the Co complex of β-octabromo tpfc (72-Br8).218 Introducing eight β-pyrrole electron-withdrawing bromide groups can shift the redox potential of Co corroles to the much more positive direction. As these authors expected, the ORR with 72-Br8 had an onset potential at 0.56 V (vs Ag/ AgCl) at pH 0, which is more positive than its unbrominated analogue. In addition, the introduction of bromide groups at βpyrrole positions changes the O2 reduction from a two-electron pathway to a four-electron pathway, and at pH 3−4, O2 was reduced completely to H2O via the direct four-electron pathway with no H2O2 formation (the catalytic activity of 72-Br8 toward H2O2 reduction is negligible). The results from pH-dependence and Tafel studies indicated that the ﬁrst one-electron transfer step from the formal CoIII state to O2 is the rds. In a subsequent report, Gross, Elbaz, and co-workers compared the ORR activity of Co corrole 72-Br8 with its Mn, Fe, Ni, and Cu analogues.219 When adsorbed on a high-surface-area carbon powder (BP2000) for electrochemical measurements in acidic aqueous solutions (0.5 M H2SO4), the catalytic activity showed a clear dependence on the metal center with Co corrole 72-Br8 as the most eﬃcient (the onset potential was as high as 0.81 V vs RHE). The activity order is Co > Fe > Ni > Mn > Cu.

Interestingly, complexes 63−67 displayed nearly identical catalytic ORR peak potentials, and their electrochemical behaviors in RDE and RRDE experiments were also similar; the number of electrons transferred during the catalytic reduction of O2 ranged from 2.2 (for 63) to 2.8 (for 67), and the amount of H2O2 formed upon O2 reduction was 87.7%, 75.6%, 72.7%, 69.2%, and 64.9% for 63−67, respectively. These results indicate that electron-donating and electron-withdrawing meso-substituents of Co corroles have a relatively small eﬀect, although electron-withdrawing substituents increased the percentage of O2 to H2O conversion. These authors also studied the catalytic reduction of H2O2 by Co corroles, and the results showed that the catalytic activity toward H2O2 reduction is lower than that toward O2 reduction under the same conditions. Another series of Co corroles with diﬀerent numbers of electron-withdrawing NO2 groups at the para-position of the three meso-phenyl rings were examined as ORR catalysts by Kadish, Ou, and co-workers.212 Complexes 65 and 68−70 contain 0, 1, 2, and 3 p-nitrophenyl groups, respectively (Figure 33). When they were coated on EPG electrodes, well-deﬁned irreversible cathodic currents for O2 reduction were observed in 1.0 M HClO4 solution under air with the peak potential Epc = 0.13 V (vs SCE) for 65 and 0.17 V (vs SCE) for 70. The number of electrons transferred during ORR was determined to be 2.6, 2.7, 2.9, and 3.0 for 65, 68−70, respectively. This result further conﬁrmed that Co corroles with stronger electronwithdrawing substituents will be better electrocatalysts for the reduction of O2 to H2O, most likely because of the increased O2-binding aﬃnity in a similar fashion as the DFT results proposed for transition metal porphyrins.213 CoIII corrole 71 (Figure 33) was also synthesized and examined as an electrocatalyst for the reduction of O2 by Kadish, Ou, and co-workers.214 The CV of complex 71 adsorbed on an EPG electrode in 1.0 M HClO4 under air showed an irreversible but well-deﬁned cathodic reduction peak at Epc = 0.15 V (vs SCE) corresponding to the reduction of O2. RDE and RRDE experiments revealed that O2 was exclusively reduced to H2O2. Therefore, complex 71 is a selective catalyst for the two-electron reduction of O2 to H2O2 in acidic media. These authors propose that although Cl ligands are electronwithdrawing, their substitution at the ortho-positions of the three meso-phenyl rings may prevent the formation of a face-toface structure of two Co corrole molecules on the electrode surface due to steric hindrance. Such a face-to-face structure of two Co corrole molecules is considered to be critical for the four-electron reduction pathway with Co-based ORR catalysts. Co 5,10,15-tris(pentaﬂuorophenyl)corrole (tpfc) 72 (Figure 33) was shown to be active for ORR by Kadish and coworkers.215 UV−visible studies suggest that the ﬁrst and second reduction processes of 72 are metal-centered, involving CoIII/ CoII and CoII/CoI, respectively. The catalytic activity was examined in both heterogeneous and homogeneous systems. When adsorbed on a graphite electrode, the catalytic reduction of O2 occurred at 0.34 V (vs SCE), a potential similar to that of 72 on the electrode under argon. This result indicates that the neutral CoIII form catalyzes the electroreduction of O2 in acid media. The ORR half-wave potential for 72 is 0.30 V (vs SCE), which is more negative than those for Co corroles with βpyrrole substituents (i.e., 0.38 V vs SCE).216 This shift in redox potential is thought to be a result of the bulky mesosubstituents of 72, which could decrease the π−π interactions between the macrocycle and the electrode surface, and it could 3741

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 34. Other transition metal corroles 73−75 studied as ORR catalysts.

DFT calculations reveal that the one-electron reduction of these metal corroles will occur on the metal center for Mn, Fe, and Co, but it will occur on the corrole ligand for Ni and Cu. Because the formation of an O2 adduct and electron transfer from the metal center to O2 is usually the rds in O2 reduction, this result implies that the inner-sphere electron transfer will be faster for the Mn, Fe, and Co complexes than for the Ni and Cu complexes. Compared to Co and Fe, the relatively low reduction potential of the Mn corrole (−0.37 V for Mn, +0.05 V for Fe, and +0.36 V for Co) makes it a much less potent ORR catalyst. The activity and selectivity of O2 reduction catalyzed by Co porphyrins and corroles are diﬀerent from those catalyzed by their corresponding Fe analogues. In general, monomeric Co porphyrins and corroles catalyze the two-electron reduction of O2 to H2O2, but monomeric Fe analogues are intrinsically eﬃcient for the four-electron reduction of O2 to H2O. It is informative to compare the catalytic ORR performance of Febased and Co-based porphyrins and corroles. Xia and coworkers used DFT methods to illustrate this activity diﬀerence.220 The energy level of the highest-occupied 3d orbital of the metal in Fe-based porphyrins is higher than that of the corresponding Co-based porphyrins, thus implying the strong ability of O2 reduction for the Fe-based porphyrins. 2.3.3. ORR Catalysts Based on Other Transition Metals. Several examples of ORR catalysts based on other transition metals have been reported. In 2015, Fukuzumi, AbuOmar, and co-workers reported the catalytic two-electron reduction of O2 by Me8Fc with MnIII tpfc 73 (Figure 34).221 This MnIII corrole was produced by the reduction of the MnV imidocorrole complex with two equivalents of Me8Fc in the presence of acids. For catalytic ORR studies, MnV imidocorrole was initially added as a catalyst precursor. Because the steady state observed by UV−visible spectroscopy in the catalytic ORR was complex 73 with excess O2, Me8Fc, and TFA, it is suggested that the formation of an O2 adduct of 73 is the rds in the catalytic reaction. After the reaction, the Me8Fc+ formed is twice that of H2O2. This result implies that 73 catalyzes the selective two-electron reduction of O2 to H2O2. Kinetic studies revealed that the observed rate constant kobs was proportional to the concentrations of 73 and O2 but remained constant irrespective of the concentrations of Me8Fc and TFA. The ﬁrst-order dependence of kobs on the concentrations of 73 and O2 further conﬁrmed that the coordination and subsequent electron transfer from 73 to O2 to produce [(tpfc)MnIV(O2•−)] is the rds in the catalytic cycle, which is proposed and shown in Figure 35. Importantly, the involvement of [(tpfc)MnIV]+ in this catalytic cycle was suggested by EPR measurements. This result conﬁrmed that this catalytic two-electron reduction of O2 proceeded via a MnIII/MnIV redox couple.

Figure 35. Proposed reaction mechanism of O2 reduction with Fc derivatives catalyzed by Mn corrole 73. Redrawn from ref 221. Copyright 2015 American Chemical Society.

Similarly, Fukuzumi, Abu-Omar, and co-workers also reported the selective two-electron reduction of O2 by Me8Fc catalyzed by a CrIII corrole complex 74 (Figure 34), which is formed in situ from a high-valence CrVO corrole acting as a catalyst precursor.222 Catalytic ORR studies were performed by mixing catalyst precursor and excess amounts of Me8Fc and TFA in an O2-saturated acetonitrile solution. The amount of formed Me8Fc+ is twice the amount of produced H2O2, indicating a two-electron reduction pathway. Similar to Mn catalyst 73, the steady state observed by both UV−visible and EPR spectroscopy in the catalytic reaction was the CrIII complex 74, suggesting that the oxidation of this CrIII catalyst by O2 is the rds in the catalytic reaction. Kinetic studies revealed that the observed rate constant kobs is proportional to the concentrations of 74 and O2 but is independent of the concentrations of Me8Fc and TFA, further conﬁrming that the inner-sphere electron transfer from CrIII corrole to O2 to give [(tpfc)CrIV(O2•−)] is the rds in the catalytic reaction. Importantly, if a weaker reductant Fc (Eox = 0.37 V vs SCE) was used instead of Me8Fc (Eox = −0.04 V vs SCE), the O2 reduction rate became much slower, and the steady state observable in the UV−visible spectra was [(tpfc)CrIV]+. This result indicates that the rds switched from the oxidation of CrIII by O2 to the reduction of CrIV by Fc because the driving force of electron transfer from Fc to CrIV becomes much smaller compared to that of Me8Fc. The oxidation of CrIII by O2 was investigated by using stopped-ﬂow experiments under singleturnover conditions. Interestingly, the reaction of CrIII and O2 3742

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

gave [(tpfc)CrIV]+ in the presence of excess TFA but gave [(tpfc)CrV(O)] in the absence of acids. It is proposed that a binuclear CrIV-μ-peroxo species [(tpfc)CrIV(O22−)CrIV(tpfc)] is produced via the reaction between [(tpfc)CrIV(O2•−)] and CrIII, which subsequently undergoes homolytic O−O bond cleavage to aﬀord [(tpfc)CrV(O)]. This result may explain why CrIII complex 74 catalyzes the selective two-electron reduction of O2 to H2O2 under catalytic conditions because with a large excess of acids, formation of the hydroperoxide complex [(tpfc)CrIV(OOH)] is favored for the subsequent release of H2O2. On the basis of these results, a catalytic cycle is proposed (Figure 36).

porphyrin ligands on ORR catalysis. The nature of porphyrin rings may either deplete or increase the electron density around the central metal ion and thus aﬀect the catalytic activity because the O2 activation involves the binding of O2 to the metal center and subsequent electron transfer from metal to O2. In the case of Mn porphyrins, the Mn-O bonds are intrinsically stronger than the Fe-O and Co-O bonds, which makes Mn porphyrins less attractive as ORR catalysts. However, the Mn-O bonds can be weakened by modifying porphyrin ligands with electron-donating groups. It is generally accepted that electron-donating meso-substituents increase the electron density on the central metal ion, whereas electronwithdrawing meso-substituents decrease this electron density. For example, it is known in the literature that porphyrins modiﬁed with both strong electron-deﬁcient and electron-rich groups are not eﬃcient for intermetal oxygen atom transfer reactions because the interaction between Mn and O is either too strong or too weak.92 Consequently, according to the Sabatier principle, it is thought that a particular porphyrin ligand may optimize the electrocatalytic ORR eﬃciency. The plot of the potential at half-maximum peak currents for ORR as a function of the formal potential of the MnIII/MnII redox couple also gives a volcano-like curve with the Mn complex of tetra(4-pyridyl)porphyrin as the most active. Therefore, based on this work, it is concluded that both the nature of the central metal ion and the type of porphyrin ligand determine the electrocatalytic properties of metalloporphyrins for ORR. 2.3.4. Metal-Free Porphyrin ORR Catalysts. Metal-free porphyrins have also been shown to be active in the catalytic reduction of O2. In 2011, Samec and co-workers reported the catalytic activity of tetraphenylporphyrin (76, Figure 37) in the

Figure 36. Proposed reaction mechanism of O2 reduction with Fc derivatives catalyzed by Cr corrole 74. Redrawn from ref 222. Copyright 2014 American Chemical Society.

Aguirre and co-workers reported electrocatalytic ORR with a CuIII corrole 75 (Figure 34).223 When physiadsorbed on a GC electrode, complex 75 can catalyze the four-electron reduction of O2 to H2O in both pH 3 and 7 aqueous solutions. If it is electropolymerized on a GC electrode, complex 75 catalyzes O2 reduction through two parallel pathways, giving H2O2 and H2O simultaneously at pH 3 and through the two-electron reduction pathway giving H2O2 as the unique product at pH 7. However, the modiﬁed electrodes in both cases are unstable and quickly lose activity during electrocatalysis. Schuhmann and Masa performed a systematic examination of metalloporphyrin-based ORR catalysts.224 A variety of the ﬁrstrow transition metal complexes of tetra(4-aminophenyl)porphyrin are synthesized and are used to investigate the eﬀect of the central metal ion on electrocatalytic O2 reduction. The results show that the electrocatalytic activities of these metal porphyrins have a volcano-shaped curve with Co as the most active (Co > Mn, Fe, Ni > Cr, Cu). These data suggest that the binding of an O2 molecule on the central metal ion is a vital factor in the catalytic cycle, and the optimal catalysis may occur when the interaction of metal ion with substrates (i.e., O2), intermediates, and products is neither too weak nor too strong according to the Sabatier principle.225 In addition to the central metal ions, these authors also investigated the eﬀect of

Figure 37. Free-base porphyrins 76-79 studied as O2 reduction catalysts.

reduction of O2 with external reductant Fc and acid HTB.226 Adding HTB to an air-saturated DCE solution of 76 caused a red shift of the Soret band in UV−visible spectroscopy, implying the protonation of 76. Similar to the study by Rosa and co-workers,227 this protonation process consists of two steps to form monoprotonated and diprotonated forms of 76 in solution (Figure 38). Adding Fc to an air-saturated DCE solution containing both 76 and HTB initiated the ORR 3743

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

porphyrins 76−78 were employed (Figure 37). Because O2 activation by metal-free porphyrins is thought to involve the binding of O2 to the protonated imino N atoms, the diﬀerence in the acidity of these porphyrins should have noteworthy eﬀects on their catalytic ORR. On the basis of pKaDCE values, the introduction of electron-donating meso-substituents renders the porphyrin more easily protonated, whereas the introduction of electron-withdrawing meso-nitrophenyl groups makes the porphyrin protonation more diﬃcult. ORR studies showed that all three complexes were able to catalyze the reduction of O2 to H2O2; 78 was a stronger catalyst than the other two. The trend of catalytic activity 78 > 76 > 77 is consistent with molecule acidities. It is thought that the electron-withdrawing group makes the tetrapyrrole ring electron-deﬁcient, leading to a stronger polarization of the O−O bond. These results support the proposal that the O2 activation pathway involves the binding of O2 to the protonated porphyrins via NH+···O2 hydrogen bond interactions and subsequent polarization of the O−O bond. Girault and co-workers examined the catalytic ORR feature of an amphiphilic porphyrin 79 (Figure 37).231 Complex 79 contains a hydrophilic meso-p-aminophenyl substituent to increase the aﬃnity at the ITIES interface and three mesopentaﬂuorophenyl groups to ensure a strong electron-withdrawing eﬀect to facilitate the polarization of the O−O bond. The results demonstrated catalytic O2 reduction to H2O2 by 79. However, complex 79 is much less active than amphiphilic Co porphyrin 60 (Figure 32) under the same conditions reported by these authors.206 In a short summary, a few conclusions can be made based on the mentioned O2 reduction studies with metal-free porphyrin catalysts. First, upon binding to diprotonated porphyrins via NH+···O2 hydrogen bond interactions, the O−O bond is polarized, and O2 is activated for subsequent reduction by external reductants. The acidities of porphyrins have noteworthy eﬀects on ORR, whereas porphyrin redox properties have insigniﬁcant eﬀects. Second, acid anions and water molecules can bind to protonated porphyrins and thus have an inhibitory eﬀect on O2 activation by blocking the binding and thus the reaction sites for O2. The ability of acid anions to associate with protonated porphyrins decreased with increasing anion size. Third, the reduction of activated O2 with external reductants is the rds in the catalytic cycle. Therefore, increasing the reducing power will increase the rate of O2 reduction. Fourth, metal-free porphyrins are generally less eﬃcient than their corresponding metal porphyrins for catalytic O2 reduction.

Figure 38. Mono- and diprotonation of free-base porphyrins.

process. The reaction rate reached the maximum with the molar ratio of HTB:76 as 2.5. Further increasing this ratio decreased the reaction rate. These data suggest that the increase of HTB will lead to an increase of the protonated binding sites for O2, but at the same time, more TB− anions will ion pair to the protonated porphyrin and thus inhibit the O2 binding and activation. As a consequence, the O2 reduction rate will have a maximum when the molar ratio of HTB:76 is adjusted. This hypothesis is conﬁrmed by ORR experiments in which HTB is replaced by TFA. Because the acid anion of TFA has stronger association with the protonated porphyrin, the O2 reduction rate with TFA decreased to almost zero. Because the maximum reaction rate is observed at the HTB:76 molar ratio of 2.5, O2 is likely to bind to the diprotonated rather than the monoprotonated porphyrin. DFT studies revealed that the binding of O2 on the diprotonated form of 76 polarized the O−O bond, and the association of acid anions on protonated porphyrins decreased considerably upon increasing the anion size (i.e., small acid anions are more likely to block the binding and thus the reaction site for O2). This work is important because it illustrates the eﬀect of counteranions in the reduction of O2 using metal-free porphyrins as catalysts. In a subsequent study investigating the catalytic ORR properties of porphyrin 76, Samec and co-workers demonstrated that water has an inhibitory eﬀect similar to acid anions.228 Catalytic ORR studies in DCE solutions containing Fc, O2, HTB, 76, and various amounts of H2O were performed. The results showed that the reaction rate decreased sharply when the water concentration increased, which was thought to result from the replacement of O2 by H2O molecules in the protonated active site. The competitive binding of various groups, including TB−, H2O, (H2O)n (n = 2, 3, or 4), and H2O2 with the diprotonated form of 76 was studied by DFT calculations. As shown in Table 4, the acid anion TB− exhibits Table 4. DFT Calculated Stabilization Energy ΔE(L) and the Corresponding Free Energy ΔG(L) for the Removal of L from the Diprotonated Form of 76 energy (L)

O2

TB−

H2O

(H2O)4

H2O2

ΔE(L) (eV) ΔG(L) (eV)

0.118 −0.145

0.759 0.241

0.592 0.275

0.510 0.179

0.535 0.084

2.4. ORR Catalysts Based on Porphyrin and Corrole Architectures

2.4.1. Fe-Based Architectures for ORR Catalysis. In the section discussing CcOs and synthetic models, the role of electron ﬂux in the eﬃciency of ORR was investigated by Collman and co-workers.159 In 2011, Hosseini, Collman, and co-workers reported the role of proton ﬂux in O2 reduction with Fe porphyrin catalyst 19 (Figure 14) by controlling the proton ﬂux using an electrode-supported hybrid bilayer membrane system.232 Typically, PCET reactions are studied by changing the pH of the bulk solution. However, this process will cause an accompanying change in the thermodynamic potential of the redox center; thus, it provides limited information about the role of the proton ﬂux in the PCET processes. To solve this problem, Hosseini, Collman, and coworkers constructed a catalyst-embedded hybrid bilayer

the largest stabilization energy; molecules of H2O, (H2O)4, and H2O2 have comparable stabilization energy, and the binding of O2 to the diprotonated form of 76 is much less favored. In addition, Samec and co-workers investigated the reduction of O2 by various Fc derivatives catalyzed by porphyrin 76.229 The ORR rates have the trend of Fc < Me2Fc < Me4Fc < Me6Fc. In the case of Me8Fc and Me10Fc, the reactions were too fast to be reliably analyzed. These results suggest that the reduction of activated O2 with Fc derivatives is the rds in the catalytic cycle. The eﬀects of meso-substituents on metal-free porphyrins for catalytic ORR were investigated by Su and Wu.230 Three 3744

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 39. (A) Schematic diagram representing the hybrid bilayer membrane system. (B) Fe porphyrin catalyst 19 used for this study. Reprinted from ref 232. Copyright 2011 American Chemical Society.

membrane (Figure 39) in which a monolayer of lipid molecules is anchored onto a catalyst-immobilized SAM of alkanethiols covalently attached to an Au electrode. The polar head groups of the lipids are oriented outward toward the aqueous solution, and the hydrophobic tails are oriented inward to the hydrophobic SAM. The proton ﬂux was then controlled by introducing decanoic acid (an aliphatic proton carrier) into the lipid layer. Fast SAM was used to eliminate the eﬀects of electron transfer. In the absence of decanoic acid, protons cannot readily diﬀuse through the lipid layer; consequently, the FeIII/FeII redox couple of 19 is independent of the pH of the solution, which is in contrast to an exposed catalyst showing a 59 mV shift per pH unit. Under this condition, O2 was suggested to be one-electron reduced to superoxide. Decanoic acid was then introduced as a proton carrier, and the amount of decanoic acid was changed to control the rate of proton ﬂux to the catalyst. Electrocatalytic ORR studies showed that increasing the decanoic acid concentration in the lipid layer increased the current, reﬂecting an enhanced proton transport across the lipid layer to the active site. The number of electrons transferred during ORR with the addition of adequate decanoic acid was ∼3.4, which is consistent with the value of ∼3.6 for the exposed Fe porphyrin 19. Signiﬁcantly, this hybrid bilayer membrane system is able to control the proton ﬂux for the study of PCET reactions independent of the pH of the bulk solution. Maruyama and co-workers reported the enhanced catalytic ORR activity of Fe tetra(4-pyridyl)porphyrin 37 (Figure 20) by introducing other transition metal ions (Co2+, Ni2+, and Cu2+) coordinated with the N atoms of the pyridine groups.233 Electrochemical measurements showed that the catalytic current with Fe porphyrin 37 coordinated with the additional metal ion was higher than that with 37 alone. The largest enhancement was observed for the [37-Cu] system. UV−visible and X-ray photoelectron spectroscopy (XPS) revealed the coordination between the added transition metal ion and the N atom of the pyridine group. The enhancement could likely be attributed to the electronic interaction between the additional transition metal and the Fe center through the coordination bonds. Su and co-workers reported the two-electron reduction of O2 to H2O2 catalyzed by microperoxidase-11.234

A series of hangman Fe porphyrins and corroles bearing pendant proton relays were synthesized and studied as catalysts for O−O bond activation by Nocera and co-workers. In addition to their enhanced catalytic activities, the hangman motif with intramolecular proton relays represents an excellent platform for the investigation of PCET processes involved in the bond-making and bond-breaking reactions of small molecule substrates. In 2003, Nocera and co-workers reported the synthesis of hangman Fe porphyrins 80-85 (Figure 40) containing pendant groups with a wide range of protondonating abilities (pKas ranging from ∼2 to 25).235 Complexes 80−85 were evaluated as catalysts for the O−O bond activation of H2O2, which is relevant to the reduction of O2. The results show that the initial rates of H2O2 disproportionation track the acidity of the pendant group; a lower pKa results in a higher initial rate constant. These authors rationalize that pendants with higher pKa values are less eﬃcient in providing protons for heterolytic O−O bond cleavage; in contrast, more acidic pendants (i.e., phosphonic acid) are deprotonated, and their conjugate base forms are not eﬀective in catalysis. The rigid xanthene and dibenzofuran pillars (Figure 40) were compared for the construction of hangman porphyrins by Nocera and co-workers.236 The results showed that complex 81, containing a xanthene bridge and a carboxylic acid group, was the most active for H2O2 disproportionation and produced almost an order of magnitude more O2 than catalysts 86−88. Signiﬁcant decreases in catalytic activity were observed by replacing xanthene with dibenzofuran as the pillar or by replacing carboxylic acid with methyl ester as the “hanging” group. Therefore, it is concluded by these authors that the protic nature of the pendant group and its proper orientation are key factors for the observed high eﬃciency O−O bond activation. In addition to the pillar and the “hanging” group, Nocera and co-workers investigated the eﬀect of diﬀerent meso-substituents of hangman Fe porphyrins on the O−O bond activation of H2O2.237 Complexes 81, 89, and 90 (Figure 40), all containing a xanthene bridge and a carboxylic acid group but bearing diﬀerent aryl groups at the meso-positions of the porphyrin ring, were synthesized. The results show that the disparate electronic properties have an unimportant eﬀect on H2O2 dismutation 3745

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 41. Proposed reaction mechanism of H2O2 activation catalyzed by hangman Fe porphyrins. Sub = substrate; Sub(O) = oxidized substrate. Redrawn from ref 238. Copyright 2007 American Chemical Society.

the initial catalytic rate has the trend of 92 > 93 ≈ 94, the TONs of 92 and 93 are similar and are smaller than that of 94. These authors attributed this observation to the greater stability of 94 compared to 92 and 93. For example, the presence of one more pentaﬂuorophenyl meso-substituent in 94 is thought to decrease electron density in the corrole π-system and thus suppress the oxidative degradation. In addition, FeIII corroles are nearly twice as active in H2O2 dismutation compared with their corresponding formal FeIV corroles. This improvement is likely caused by avoiding the formation of high-energy Fe-oxo intermediates. The oxidative degradation of Fe corroles during catalytic H2O2 disproportionation was investigated by Nocera and co-workers.240 The meso positions of the corrole macrocycle are susceptible to attack by oxygen species, which becomes prevalent under high oxidation states. This result is consistent with the fact that FeIII corrole, not the formal FeIV corrole, is the proper precatalyst state for eﬀective H2O2 dismutation. 2.4.2. Co-Based Architectures for ORR Catalysis. In 2010, Nocera and co-workers reported the reduction of O2 catalyzed by hangman Co porphyrins 95−99 (Figure 42).241 Complex 100 was also studied to assess the hangman eﬀect. The onset of the catalytic ORR wave was at 550−600 mV (vs RHE), and the plateau wave was at 300−400 mV (vs RHE). The number of electrons transferred during the ORR was determined, which indicated that the selectivity for the fourelectron reduction of O2 to H2O was enhanced with both electron-withdrawing meso-substituents and the hanging carboxyl group that is capable of proton transfer. Likewise, Nocera and co-workers reported the reduction of O2 catalyzed by hangman Co corroles 101−103 (Figure 42) and Co tpfc 72 (Figure 33).242 The onset for ORR occurred at 650−700 mV (vs RHE), and the half wave potential for ORR

Figure 40. Hangman Fe porphyrins and corroles studied by Nocera and co-workers.

because the electron-rich (89) and electron-poor (90) hangman systems display similar activities, and both are less eﬃcient compared with 81. The deactivation of Fe porphyrin catalysts may be due to the formation of FeIII-O-FeIII dimers. The sterically encumbered meso-mesityl groups hinder such a dimerization and thus allow 81 to achieve the highest turnover number (TON) for H2O2 dismutation. The reaction mechanism of O−O bond activation catalyzed by hangman Fe porphyrin 81 was investigated by Nocera and co-workers.238 Control experiments using its methyl ester derivative 86 and Fe tetramesitylporphyrin 91 (not shown) were performed to elucidate the “pull eﬀect” of the “hanging” acid group. Key results obtained included the following. (1) The substitution and coordination of peroxyacids on ferric Fe porphyrins is an acid-assisted process with the trend of 81 > 86 > 91. (2) For 81, subsequent heterolytic scission of the O−O bond produces an FeIVO porphyrin π-cation radical [FeIV(O)-Por•+], whereas for 86 and 91, both heterolytic (2e−) and homolytic (1e−) scissions of the O−O bond occur to give [FeIV(O)-Por•+] and [FeIV(O)-Por], respectively. These results suggest that the acid−base residue of the hangman system suppresses the homolytic O−O bond activation. (3) Both [FeIV(O)-Por•+] and [FeIV(O)-Por] are detected by spectroscopic methods in stopped-ﬂow measurements. A proposed catalytic reaction mechanism is depicted in Figure 41. The hangman eﬀect on H2O2 disproportionation was also studied by Nocera and co-workers using Fe corroles.239 Hangman Fe corroles 92 and 93 (Figure 40) and Fe tpfc 94 (not shown) were synthesized. The results show that although 3746

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 42. Hangman Co porphyrins and Co corroles studied by Nocera and co-workers.

was at E1/2 = ∼550 mV (vs RHE). Similar to hangman Co porphyrins, hangman Co corroles catalyzed a parallel twoelectron and four-electron reduction of O2. The presence of both electron-withdrawing meso-substituents, and the hanging carboxyl group improved the four-electron reduction selectivity of O2 to H2O, although this selectivity was higher for hangman Co porphyrins than for hangman Co corroles. For Co-based catalysts, dimeric species are usually required for the four-electron reduction of O2. For example, Fe porphyrin 32 (Figure 18) bearing four Fc moieties as additional electron donors is a selective four-electron ORR catalyst even under very slow electron ﬂux conditions, but the introduction of these electron donors to Co porphyrins does not promote the four-electron reduction.243 Kadish and co-workers synthesized Co porphyrins 104−106 (Figure 43) bearing 1, 3, and 4 electron-donating Fc appendages, respectively, and compared their electrocatalytic ORR activities with that of simple Co

tetra(4-methylphenyl)porphyrin 56 (Figure 30). The number of electrons transferred during O2 reduction was 2.8 for 56 and 2.0−2.1 for 104−106. Similar results were obtained by determining the yield of H2O2 generated using RRDE: 60% for 56 and ∼100% for 104−106. This result indicates that Co porphyrins 104−106 are selective catalysts for the two-electron reduction of O2 to H2O2. As we addressed in the previous section, the formation of a face-to-face structure with an appropriate Co···Co distance is necessary for Co porphyrins to catalyze the four-electron reduction of O2. The selectivity observed with 104−106 is therefore explained by the steric hindrance of the bulky Fc groups at meso positions, which prevent the formation of dimers on the electrode surface and thus prevent the four-electron reduction pathway. Co porphyrin 107 containing four RuII centers (Figure 43) reported by Anson and co-workers catalyzed the two-electron reduction of O2 in homogeneous solutions, but it catalyzed the four-electron reduction of O2 when adsorbed on graphite electrodes.244 Although 107 is a potential ﬁve-electron reductant, it cannot transfer more than one or two electrons to its O2 adduct. Swavey and co-workers investigated the electrocatalytic O2 reduction with a bimetallic CoII/PtII porphyrin 108 (Figure 43), in which a [PtCl2(DMSO)] unit (DMSO = dimethyl sulfoxide) is coordinated to the meso-pyridyl group of the porphyrin macrocycle.245,246 The introduction of three meso-(3methoxy-4-hydroxy)phenyl units is intended to direct the electropolymerization of porphyrins deposited onto electrode surfaces. The ORR studies were conducted in 1.0 M HClO4 aqueous solutions using EPG electrodes coated with 108. The number of electrons transferred during ORR was 2.3 and 3.3 before and after, respectively, the surface oxidation treatment of complex 108. In addition to O2 reduction, this complex was shown to catalyze HER from deoxygenated 1.0 M HClO4 aqueous solutions with the onset at −0.75 and −0.50 V (vs Ag/ AgCl) before and after the surface oxidation treatment. As proposed by these authors, the enhanced catalytic ability upon the surface oxidation of 108 may result from the formation of a more stable catalyst ﬁlm on the electrode surface. It is worth noting that the Co-free form of 108 showed a decrease in catalytic activity, suggesting that the Pt center is not directly involved in catalysis. The free base porphyrin of 108 showed even smaller catalytic activity for the reduction of both H+ and O2.

Figure 43. Co porphyrins 104−109 studied as ORR catalysts. 3747

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 44. Cofacial mononuclear free-base porphyrin-Co corrole dyads 110−112 studied as ORR catalysts by Kadish and co-workers.

Figure 45. Co corroles 113−122 studied as ORR catalysts by Kadish and co-workers.

catalytic wave in the presence of O2 with the E1/2 at 0.35 V (vs SCE), indicating that CoIII was the active species for O2 reduction. The number of electrons transferred in O2 reduction was in the range from 2.5 to 2.9, suggesting the presence of parallel two-electron and four-electron pathways to produce H2O2 and H2O. The free porphyrin unit may increase the selectivity for the four-electron pathway by stabilizing the partially reduced O2 species. When ORR was performed in 1 M HCl aqueous solutions, the catalytic wave shifted to the negative direction by 60 mV (E1/2 = 0.29 V vs SCE), which was consistent with Cl− coordination on the Co center. Kadish and co-workers also investigated the ORR properties of Co complexes of monocorrole, biscorroles, and porphyrincorrole dyads 113−122 (Figure 45).250 The electrocatalytic O2 reduction features of these catalysts, including the peak potential of the O2 reduction wave (Ep), the half-wave potential of the O2 reduction wave (E1/2), and the number of electrons transferred per O2 reduction (n), are summarized in Table 5. Major conclusions obtained are as follows. (1) The CoIII corroles are the catalytic active species for O2 reduction. (2) The CoIII biscorroles are less active and less selective than the CoII-CoIII porphyrin-corrole dyads for O2 reduction, but these bimolecular Co complexes are more selective than the mononuclear CoIII corrole. (3) Catalysts 113−122 are not active in the reduction of H2O2 to H2O. Kadish and co-workers further explored the reaction mechanism of biscobalt porphyrin-corrole dyads of 116, 117, and 123−126 (Figure 46).251 On the basis of electrochemical and spectroscopic comparisons with mononuclear CoIII corroles and CoII porphyrins, the proposed reduction site of these biscobalt porphyrin-corrole dyads is depicted in Figure

Swavey and co-workers also synthesized a highly conjugated Co porphyrin 109 (Figure 43) and demonstrated its enhanced O2 reduction properties relative to Co tetraphenylporphyrin 52 (Figure 29).247 The half-wave potential of catalytic O2 reduction current was at 0.16 and 0.30 V (vs SCE) with complexes 52 and 109, respectively. Complex 52 catalyzed the two-electron reduction of O2 to H2O2, whereas 109 catalyzed the reduction of O2 to both H2O2 and H2O with 3.2 electrons transferred. The ratio of the reaction rate k(109)/k(52) was 1.70. All these results suggest that the highly conjugated Co porphyrin has enhanced ORR activity and selectivity. The face-to-face linked porphyrin-porphyrin and porphyrincorrole dyads have been extensively investigated for Co-based architectures as ORR catalysts. Kadish and co-workers synthesized a series of free-base porphyrin-Co corrole dyads 110−112 (Figure 44) and examined their catalytic ORR properties.248 Complexes 110−112 contain a free porphyrin unit and a Co corrole unit linked by three diﬀerent spacers, anthracenyl, 9,9-dimethylxanthenyl, or biphenylene, to construct the face-to-face arrangement. The anthracenyl spacer has been widely used to create this geometry. For example, Collman and co-workers have used it to link two iridium porphyrins for catalytic ORR studies.249 It is known that the larger the internal interaction between the porphyrin and corrole macrocycles, the harder the reduction. Therefore, the following order of the internal π−π interaction is proposed based on electrochemical data: 110 < 111 ≈ 112 with the CoIII/CoII redox couple at −0.20, −0.39, and −0.38 V (vs SCE) in DCM. Electrocatalytic O2 reduction was ﬁrst performed in a 1 M HClO4 aqueous solution. The CoIV/CoIII wave became a 3748

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Table 5. Electrocatalytic ORR by Catalysts 113−122 Adsorbed on EPG Electrodes in Air-Saturated 1 M HClO4 Aqueous Solutionsa

a

Table 6. Electrocatalytic ORR by Catalysts 116, 117, and 123-126 Adsorbed on EPG Electrodes in Air-Saturated 1 M HClO4 Aqueous Solutionsa

catalyst

Ep

E1/2

n

catalyst

Ep

E1/2

n

113 114 115 116 117 118 119 120 121 122

0.36 0.40 0.38 0.38 0.34 0.36 0.35 0.34 0.33 0.33

0.38 0.47 0.46 0.45 0.41 0.39 0.37 0.37 0.35 0.35

2.9 3.9 3.7 3.7 3.5 3.4 2.4 2.9 3.4 3.1

116 117 123 124 125 126

0.38 0.34 0.35 0.25 0.25 0.27

0.45 0.41 0.40 0.32 0.32 0.33

3.7 3.5 3.1 2.5 2.4 2.5

a

The potentials are vs SCE.

O2 to H2O2. This diﬀerence between homogeneous and heterogeneous ORR catalysis may result from the diﬀerent geometry of biscobalt complexes dissolved in solution compared to their geometry when adsorbed on the electrode surface. The reaction mechanism of O2 reduction with these biscobalt porphyrin-corrole dyads was proposed (Figure 48). It

The potentials are vs SCE.

Figure 46. Cofacial binuclear Co porphyrin-corrole dyads 116, 117, and 123−126 studied as ORR catalysts by Kadish and co-workers.

47; the ﬁrst and third reductions of these dyads are Cocentered in the corrole units, whereas the second reduction is Co-centered in the porphyrin units. The electrocatalytic ORR data by these dyads are summarized in Table 6. As observed from this table, the catalytic activity and selectivity of mesitylcontaining porphyrin-corrole dyads 124−126 are less than those of 116, 117, and 123. One possible explanation is that the bulky mesityl substituents on the corrole cause slippage between the two macrocycles, disfavoring the accommodation of O2 reduction intermediates during catalysis. However, a diﬀerent ORR selectivity trend was observed under homogeneous conditions in benzonitrile using Me2Fc as the reductant and HClO4 as the proton source; 116 and 124 catalyzed the four-electron reduction of O2 to H2O, whereas 117, 123, 125, and 126 catalyzed the two-electron reduction of

Figure 48. Proposed reaction mechanism of O2 reduction catalyzed by cofacial binuclear Co porphyrin-corrole dyads. Redrawn from ref 251. Copyright 2009 American Chemical Society.

is concluded that under heterogeneous conditions, the position of corrole substituents is important, whereas under homogeneous conditions, the eﬀect of the spacer becomes signiﬁcant. Kadish and co-workers also investigated the ORR properties of face-to-face linked heterobimetallic dyads 127−133 (Figure

Figure 47. Proposed reduction sites of cofacial binuclear Co porphyrin-corrole dyads. Redrawn from ref 251. Copyright 2009 American Chemical Society. 3749

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

49) containing Co corroles and Fe or Mn porphyrins.252 The O2 reduction was conducted in 1 M HClO4 aqueous solutions.

Figure 50. Cofacial binuclear Co porphyrin-porphyrin dyads 134−137 studied as ORR catalysts by Fukuzumi and co-workers.

O2. For 134, 136, and 137, the amount of Me2Fc+ was less than four equivalents of O2, indicating the presence of parallel twoelectron and four-electron pathways. The suitable Co···Co separation in 135 may be the reason why it is the most eﬃcient catalyst for the selective four-electron reduction of O2 by Fc derivatives. Kinetic studies reveal that for mononuclear Co octaethylporphyrin 53, the pseudo-ﬁrst-order rate constant (kobs) has a linear dependence on the concentration of the catalyst but remains constant for changing O2 and HClO4 concentrations. This result indicates that the electron-transfer from Me2Fc to the CoIII state of 53 is the rds, which is also consistent with the ﬁnding that the catalytic steady state observed in ORR with 53 is its CoIII state. For 135 with Fc or Me2Fc as the reductant, the pseudo-ﬁrst-order rate constant (kobs) increases linearly with the increase of the concentration of catalyst, HClO4, and O2. This result indicates that the PCET from the CoIIICoII state of 135 to O2 is the rds in the reduction of O2. However, if the much stronger reductant Me10Fc is used, the kinetics change from pseudo-ﬁrst-order to zero-order (i.e., the rate remains constant irrespective of Me10Fc concentration). In addition, this zero-order rate constant in the case of Me10Fc remains constant with changing O2 and HClO4 concentrations. This result indicates that the rds changes to the O−O bond cleavage with the use of the stronger reductant Me10Fc. The O2 reduction mechanism with cofacial dicobalt porphyrin catalysts is proposed by these authors (Figure 51). Interestingly, Fukuzumi and co-workers also investigated O2 activation coupled to the dehydrogenation or oxygenation of 10-methyl-9,10-dihydroacridine (AcrH2) and 9-alkyl-10-methyl-9,10-dihydroacridines (AcrHR; R = Me, Et, CH2COOEt, CH2Ph, CMe2COOMe, and tBu) catalyzed by cofacial dicobalt porphyrins 134−137 (Figure 50).254 AcrH2 is an analogue of the biological reductant NADH. The results show that cofacial dicobalt porphyrins can eﬃciently catalyze the four-electron reduction of O2 with AcrH2 in the presence of HClO4 in benzonitrile, but monomeric Co porphyrins catalyze only the two-electron reduction of O2 under the same experimental conditions. When AcrHR is used as the reductant, cofacial dicobalt porphyrins catalyze the four-electron reduction of O2 with either the dehydration or oxygenation of AcrHR, depending on the nature of R substituents, to yield AcrR+ and H2O or AcrH+ and ROH, respectively (Figure 52). Similar selectivity was observed in the case of monomeric Co porphyrins, although they catalyze the two-electron reduction of O2 to form AcrR+ and H2O2 for the dehydration pathway or AcrH+ and ROOH for the oxygenation pathway. The selectivity of the C−H versus C−C bond cleavage depends on the relative bond strength of the C(9)−H and C(9)−C bonds. Kinetic and KIE studies revealed that the C(9)−H and/or C(9)−C bond cleavage is involved in the rds for the reduction of O2 with AcrH2 or AcrHR as catalyzed by both monomeric Co porphyrins and cofacial dicobalt porphyrins.

Figure 49. Cofacial heterobimetallic Fe/Mn porphyrin-Co corrole dyads 127−133 studied as ORR catalysts by Kadish and co-workers.

In general, two reduction processes were observed, and the data are summarized in Table 7. The ﬁrst reduction process is due to Table 7. Electrocatalytic ORR by Catalysts 127−133 Adsorbed on EPG Electrodes in Air-Saturated 1 M HClO4 Aqueous Solutionsa ﬁrst wave

a

second wave

catalyst

E1/2

n

E1/2

n

127 128 129 130 131 132 133

0.39 0.33 0.35 0.30 0.33 0.36 0.32

2.8 2.6 2.8 2.6 2.5 2.6 2.8

−0.04 −0.08 −0.03 −0.06 −0.16 −0.02 −0.14

3.0 3.2 3.2 3.1 2.5 2.7 2.8

The potentials are vs SCE.

O2 reduction by the CoIII corroles, and the second reduction process is likely due to O2 (or H2O2) reduction by the FeII or MnII porphyrins. Comparison of the four series of Co-based catalysts as mentioned above, including the binuclear Co biscorrole dyads, the free-base porphyrin-Co corrole dyads, the binuclear Co porphyrin-corrole dyads, and the heterobimetallic Fe/Mn porphyrin-Co corrole dyads, revealed that the binuclear Co porphyrin-corrole dyads had the highest activity and selectivity for the four-electron reduction of O2. The binuclear Co porphyrin-corrole dyads catalyzed O2 electroreduction at potentials that were more positive by an average of 110 mV compared to the other series of catalysts. The number of electrons transferred per reduced O2 was 3.5−3.9 for the binuclear Co porphyrin-corrole dyads compared to 2.4−3.4 obtained with other dyads. For the binuclear Co biscorrole dyads, the free-base porphyrin-Co corrole dyads, and the heterobimetallic Fe/Mn porphyrin-Co corrole dyads, CoIII corrole is the active species for O2 reduction, whereas for the binuclear Co porphyrin-corrole dyads, CoII porphyrin is the active site. Fukuzumi and co-workers investigated the ORR catalyzed by cofacial dicobalt porphyrins 134−137 (Figure 50).253 The reaction was performed in air-saturated benzonitrile with HClO4 and Me2Fc. When mononuclear Co octaethylporphyrin 53 (Figure 29) was used as the catalyst, the quantity of Me2Fc+ formed was two times the quantity of O2 consumed, indicating the two-electron reduction of O2. However, when 135 was used as the catalyst, the amount of Me2Fc+ formed was four times the O2 consumption, indicating the four-electron reduction of 3750

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 53. Cofacial binuclear Co porphyrin-porphyrin dyads 138 and 139 studied as ORR catalysts by Nocera and co-workers.

greatly hinder the vertical ﬂexibility of the cofacial platforms. Similarly, electrochemical measurements showed that the redox properties of 138 and 139 were not perturbed relative to 135 and 136 by the introduction of these aryl groups. Table 8 lists Table 8. Electrochemical and Electrocatalytic Data of 135, 136, 138, and 139a electrocatalytic O2 reduction

Figure 51. Proposed reaction mechanism of O2 reduction with Fc derivatives catalyzed by cofacial binuclear Co porphyrin-porphyrin dyads. Redrawn from ref 253. Copyright 2004 American Chemical Society. a

complex

Eoxidation

E1/2

% H2O produced

135 136 138 139

0.28, 0.17 0.33 0.31, 0.14 0.33

0.38 0.37 0.24 0.25

72 80 52 46

The potentials are vs Ag/AgCl.

the electrochemical and electrocatalytic data of these complexes. The catalytic studies show that 138 and 139 have decreased activity and selectivity for the four-electron reduction of O2 to H2O compared to their parent complexes 135 and 136 despite the fact that their structural ﬂexibility and redox behavior are similar. These authors suggested that the selective four-electron reduction of O2 is dependent on the eﬃciency of targeted proton delivery to the bridging superoxo species [CoIII(O2•−)CoIII]. Without protonation, the bridging superoxo species will be reduced by one electron to yield the peroxide, whereas with eﬃcient proton delivery, it will be reduced by two electrons to trigger the O−O bond cleavage and produce CoIII-OH and CoIVO species because the O−O bond is weakened upon protonation. Therefore, for 135 and 136, proton transfer to the [CoIII(O2•−)CoIII] core is eﬃcient, favoring the complete O−O bond cleavage to produce H2O, whereas for 138 and 139, the insuﬃcient proton transfer to [CoIII(O2•−)CoIII] results in a stronger O−O bond and increased H2O2 formation. The catalytic ORR features of a variety of cofacial dicobalt porphyrins, including 135 and 136, were also examined in the ITIES system by Murtomäki and co-workers.257 The results revealed that all examined catalysts showed reasonable selectivity for the four-electron reduction of O2, and complex 135 showed the highest selectivity. DFT studies indicate that O2 reduction can proceed on both the outside (dock-on) and inside (dock-in) of the cofacial porphyrins, leading to the formation of H2O2 and H2O, respectively (Figure 54). These results provide insights to better understand the O2 reduction in CcOs; the proximal side of the porphyrin is occupied by an axial histidine group, which blocks the “dock-on” path and prevents the coordination of O2 outside the heme/Cu distal pocket. This protection is important to favor the complete reduction of O2 to H2O without the release of harmful PROS. Other cofacial dicobalt porphyrin ORR catalysts reported before 1994 have been summarized by Collman and co-

Figure 52. Dehydration or oxygenation of AcrHR.

Nocera and co-workers investigated the electrocatalytic ORR features of cofacial dicobalt porphyrins 135 and 136 (Figure 50).255 Crystallographic studies showed that the Co···Co separation is 4.582 and 8.624 Å in 135 and 136, respectively. The CV of 135 in nitrobenzene gave two reversible oxidation waves at +0.28 and +0.17 V (vs Ag/AgCl), whereas the CV of 136 gave a single reversible oxidative wave at +0.33 V (vs Ag/ AgCl), which is consistent with the expectation that the two Co centers in 135 will interact and that the two Co centers in 136 do not interact. When adsorbed on EPG electrodes, both 135 and 136 can eﬃciently catalyze the four-electron reduction of O2 to H2O at the half-wave potential E1/2 = 0.37 V (vs Ag/ AgCl) in an air-saturated aqueous solution containing 0.5 M HClO4 and 1.5 M CF3COOH. The similar ORR activity and selectivity of these two complexes, which have notable diﬀerences in structure and redox behavior, indicate that the longitudinal open-to-closed conformational change requires only a small energy change to accommodate an O2 molecule. To further investigate the conformational eﬀect of this opento-closed change of cofacial dicobalt porphyrins on ORR, Nocera and co-workers compared the activity of 135 and 136 with 138 and 139 (Figure 53), which bear bulky aryl groups trans to the spacer to modify the structural ﬂexibility of these Pacman derivatives.256 Crystallographic studies showed that the introduction of bulky aryl groups trans to the spacer did not 3751

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 54. Proposed reaction mechanism of O2 and H2O2 reduction catalyzed by cofacial binuclear Co porphyrin-porphyrin dyads. Redrawn from ref 257. Copyright 2012 American Chemical Society.

potentials of 0.10 and −0.05 V (vs Ag/AgCl), respectively. The proportion of H2O2 generated during ORR decreased with the increase of the concentration of 52 in the polypyrrole ﬁlm. This result indicates that although 52 catalyzes the two-electron reduction of O2 when adsorbed on graphite, it predominantly catalyzes four-electron reduction of O2 when trapped in a concentrated form within the polypyrrole ﬁlm. Another example of using simple mononuclear Co porphyrins for the four-electron reduction of O2 to H2O was reported by D’Souza and co-workers.260 Oppositely charged complexes 61 and 62 (Figure 32) formed an ion-pair Co porphyrin dimer when adsorbed on a GC electrode. The peak potential for O2 reduction in O2-saturated 0.05 M NH4Cl aqueous solution (pH 5.3) was located at −0.22 V (vs Ag/ AgCl). The number of transferred electrons was 3.8 ± 0.2, whereas this number was 2.0 ± 0.2 using 61 or 62 alone as the catalyst. In a short summary, to mimic the natural system for eﬃcient ORR, transition metal porphyrins and corroles have been designed and synthesized for electrocatalytic ORR. However, unlike CcO, which produces a minimal amount of peroxide from ORR, a mixture of H2O and H2O2 is generated from most of these metal complex catalysts in laboratory ORR studies. Compared to biological systems, the lack of the secondcoordination sphere in the simple synthetic macrocyclic system is responsible for poor proton transfer capacity and poor intermediate stabilization. These factors are essential to contribute the high selectivity for the full reduction of O2 to H2O. With the second-coordination sphere provided, O2 molecules will be chelated at the pocket of the molecule. However, in the case of simple macrocyclic systems, O2 coordination is exposed to reaction media. The lack of proton relay and chelating/stabilizing capability from cofactors surrounding O 2 molecules increase the probability of intermediates being released, which leads to decreased H2O selectivity. The activity, stability, and composition of the reduction products are strongly correlated with the nature of

workers.135 Zhou and co-workers synthesized and examined a series of binuclear Co bisporphyrins (Figure 55) as ORR

Figure 55. Binuclear Co porphyrin-porphyrin complexes without the cofacial structure.

catalysts.258 All these complexes, when adsorbed on GC electrodes, catalyzed the two-electron reduction of O2 to H2O2 in O2-saturated pH 6.8 aqueous buﬀer solutions. Structural consideration suggests that the two porphyrin rings in these complexes are not arranged in a “face-to-face” but in a largely slipped conformation. Therefore, they cannot act in concert for O2 reduction. To circumvent the diﬃculty in synthesizing cofacial dicobalt porphyrins, Wallace and co-workers reported the immobilization of Co tetraphenylporphyrin 52 (Figure 29) within vaporphase polymerized polypyrrole deposited on indium tin oxide (ITO) electrodes for ORR catalysis.259 It was suggested by these authors that molecules of 52 could be securely immobilized and concentrated in such a vapor-phase polymerized polypyrrole ﬁlm. Electrochemical measurements showed that this ﬁlm catalyzed O2 reduction in 0.5 M H2SO4 aqueous solutions with the half-wave potential of the reduction wave at 0.36 V (vs Ag/AgCl). The number of electrons transferred per O2 reduction was 3.3 and 4.0 at applied 3752

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

protonolysis) to release H2 and give Mn+ (heterolytic route, pathway A) or react with another H−Mn+ to give H2 and two molecules of M(n−1)+ (homolytic route, pathway B). In some cases, the one-electron reduced species M(n−1)+ can undergo oxidative protonation to form H−M(n+1)+, which can react with a proton to release H2 and M(n+1)+ (heterolytic route, pathway C), or it can be further reduced by one electron to form H− Mn+ (pathway D) for subsequent H2 generation through either pathway A or B. Alternatively, the H−M(n+1)+ intermediate can undergo a bimolecular homolytic route to generate H2 and a Mn+ starting complex (pathway E). For these ﬁve possible pathways, pathways A and D have been well-established, whereas pathway C has not been documented because the H− M(n+1)+ unit is thought to be insuﬃciently basic to drive the protonolysis. Unlike mononuclear mechanisms, only a few examples for bimetallic pathways B and E are likely to be involved in HER, but a well-deﬁned bimetallic HER mechanism has not been fully recognized.

the macrocycle catalysts, speciﬁcally the metal centers and the substituent groups at the rim of the macrocycle. In general, the activity and stability of the transition metal macrocycle complexes are far from satisfactory for their practical applications in fuel cells. However, this kind of molecule with structural similarity to the ORR site in CcO provides us an excellent platform to study the detailed ORR mechanism. The modiﬁcation of the ring substituent groups can conveniently alter the electronic structure of the molecules, which can be correlated with catalytic performance. The insightful correlation between the catalyst structure and performance can improve understanding of the ORR process and assist the design of eﬃcient catalysts in future studies. Hangman metal porphyrins and corroles have been demonstrated by Nocera and co-workers to be a useful platform to investigate PCET processes involved in small molecule activation. In particular, hanging acid groups within the secondary coordination sphere can provide hydrogenbonding interactions to participate in acid−base chemistry. This eﬀect provides kinetic control of proton transfer and bond polarization for heterolytic O−O bond cleavage; it is known as the “pull eﬀect”. In contrast, cofacial dinuclear metal porphyrins and corroles have been shown to promote the four-electron reduction of O2 based on the following considerations. First, dinuclear metal species are beneﬁcial for O2 binding and, thus, activation. The formation of bridging μ-peroxo species is indicated and suggested as a key intermediate in ORR catalytic cycles. Second, the inclusion of two redox centers is likely to provide more reducing equivalents during ORR, which has been shown to favor the complete reduction of O2 to water. Third, the proton delivery capacity in the O2-binding cavity of cofacial bisporphyrins is critical for determining O2 reduction pathways. The bridging superoxo species will be reduced by one electron to yield the peroxide without protonation, whereas it will be reduced by two electrons to trigger the O−O bond cleavage with eﬃcient proton delivery. The second metal ion also plays considerable roles in regulating the pKa of the O2 adduct.

3.1. Simple HER Catalysts

Co,262,263 Fe,264 and Rh265 porphyrins were shown to be active for electrocatalytic H2 evolution a few decades ago. CoI, Fe0, and RhI porphyrin species were suggested as the catalytic active species to reduce protons for the production of H2. The metalhydride intermediates CoIII-H, FeII-H, and RhIII-H were postulated to be involved in these catalytic cycles. However, the catalytic eﬃciency and catalyst stability remain critical issues in the design and development of HER catalysts. In 2014, Scandola and co-workers reported photocatalytic H2 generation from 1 M pH 7 phosphate buﬀer by cationic CoII tetrakis(1-methyl-pyridinium-4-yl)porphyrin 61 (Figure 32).266 The reaction was performed in a three-component system with porphyrin 61 as the catalyst, ascorbic acid as the electron donor, and [Ru(bpy)3]2+ (bpy = 2,2′-bipyridine) as the photosensitizer. The CV of 61 in acetonitrile displayed two metal-centered redox processes at −0.67 V and −1.47 V versus SCE, corresponding to the CoII/CoI and CoI/Co0 couples, respectively. Upon the addition of benzoic acid to this acetonitrile solution, the CoII /CoI couple was mostly unaﬀected, whereas the CoI/Co0 redox wave became a catalytic wave with the onset ca. −1.2 V versus SCE, implying that Co0 is the catalytic active species in this system. In photocatalytic experiments, porphyrin 61 was found to be active for generating H2, and the activity was strongly dependent on the concentration of 61. At low catalyst loading, the initial HER rate has a linear correlation with the catalyst concentration. The rate constant of the reductive quenching of [Ru(bpy)3]2+ by ascorbic acid was determined to be kQ = 3.1 × 107 M−1 s−1, and it was kQ = 7.8 × 109 M−1 s−1 for the oxidative quenching of [Ru(bpy)3]2+ by 61. However, the concentrations of ascorbic acid and 61 under the working conditions were 0.1 M and 2.5− 30 μM, respectively. This result suggests that the reductive quenching is the dominant pathway. Kinetic studies revealed that the electron transfer from [Ru(bpy)3]+ (generated by the reductive quenching) to 61 proceeded quickly with a calculated bimolecular rate constant of 2.3 × 109 M−1 s−1. This electron transfer rate is close to the diﬀusion limit and may explain the very high eﬃciency for the HER catalyzed by 61 at low concentrations. By incorporating a molecular Co porphyrin to myoglobin, Ghirlanda and co-workers demonstrated that the protein secondary shell could enhance its photocatalytic HER features.267 Co protoporphyrin IX 140 (Figure 57) was chosen

3. HYDROGEN EVOLUTION REACTION A better understanding of the reaction mechanism for H2 evolution is of fundamental signiﬁcance to provide new insights into the development of eﬃcient catalytic systems. Five possible reaction pathways for the HER catalyzed by metal complexes are summarized in Figure 56.261 H2 evolution is typically triggered upon the formation of a formal M(n−2)+ state by the two-electron reduction of a metal catalyst. Subsequent oxidative protonation of M(n−2)+ leads to a hydride intermediate H−Mn+, which can either react with a proton (known as

Figure 56. Possible HER mechanisms catalyzed by metal complexes showing both the protonolysis and homolysis pathways. Redrawn with permission from ref 261. Copyright 2016 Wiley-VCH. 3753

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

with a GC electrode showed a pronounced catalytic wave with an onset at −0.823 V (vs SHE), corresponding to an onset overpotential of 410 mV. The controlled potential electrolysis (CPE) of 62 at an applied potential of −1.39 V (vs SHE) yielded a quantitative Faradaic eﬃciency with a TOF of 1.83 s−1 in 1 h CPE experiments, suggesting its high activity and durability for HER electrocatalysis. Cao, Lai, and co-workers investigated HER mechanisms with Ni porphyrins 141−143 (Figure 58) and provided exper-

Figure 57. Co protoporphyrin IX 140 and the crystal structure of the Co-myoglobin active site, highlighting H93, H97, and H64. The crystal structure of the Co-myoglobin active site is reprinted with permission from ref 267. Copyright 2014 Royal Society of Chemistry.

and was buried into the myoglobin protein scaﬀold (deﬁned as 140-Myo) primarily through a coordination bond between the Co atom and the N atom of His93, leading to a ﬁvecoordinated geometry (Figure 57). The sixth position of Co is vacant and is available for the coordination of substrates. In an acetonitrile solution, the CV of complex 140 displayed a quasireversible wave at −1.17 V (vs SHE), which was assigned to the CoI/Co0 redox couple. This wave became a catalytic wave upon the addition of para-toluenesulfonic acid with an onset at −1.0 V (vs SHE), indicating that Co0 is the catalytic active species for proton reduction. In a pH 7.5 Tris-HCl buﬀered aqueous solution, the CV of 140-Myo showed a strong catalytic wave with an onset at −0.95 V (vs SHE), which was consistent with Co0 being the active species. Such a catalytic wave was not observed for apomyoglobin alone. Importantly, 140-Myo lost its electrocatalytic activity at pH < 6 due to the protonation of ligating His93 in myoglobin and thus the removal of 140 from the protein scaﬀold. In photocatalytic HER studies, a typical three-component system was used with 140-Myo as the catalyst, ascorbate as the electron donor, and [Ru(bpy)3]2+ as the photosensitizer. The results showed that 140-Myo could catalyze H2 generation under mild aerobic conditions with a TON of 454 over 8 h at pH 7.5, which was a 3-fold increase compared to free 140. Signiﬁcantly, three mutants of myoglobin, H64A, H97A, and H64/97A, were used to examine the possible role of the two distal histidines His64 and His97 in the active site (Figure 57). The catalytic activity of 140-Myo mutants of H64A and H64/ 97A increased, but the activity of the H97A mutant decreased in both photocatalytic and electrocatalytic HER measurements. Closer inspection of the myoglobin active site revealed that His97 could provide additional hydrogen bond interactions with one of the carboxyl groups of porphyrin 140; thus, the removal of this residue destabilized the binding of 140 in the protein scaﬀold. In contrast, His64 was thought to compete for proton binding, and the loss of this residue should be favored for reducing protons. This work is therefore signiﬁcant in demonstrating the capability of protein scaﬀolds to improve the activity of molecular HER catalysts. Another example of molecular Co porphyrin HER catalysts was reported by Hung and co-workers in 2015.268 The watersoluble anionic CoII tetrakis(p-sulfonatophenyl)porphyrin 62 (Figure 32) was examined as an electrocatalyst for H2 generation from neutral phosphate buﬀer solution. Similar to the previous two examples, Co0 is the catalytic active species for proton reduction. In a pH 7.0 phosphate buﬀer, the CV of 62

Figure 58. Ni porphyrins 141−143 studied as HER catalysts by Cao and co-workers.

imental and theoretical evidence for H2 evolution through a bimolecular homolytic route.261 By replacing meso-C6F5 with meso-C6H5, the resulting complexes 142 and 143 were shown to be less eﬃcient than 141 for the HER because of the signiﬁcantly decreased solubility and the considerable cathodic shift of the reduction waves. This result highlights the eﬀect of meso-C6F5 groups in regulating the redox chemistry of Ni porphyrins. Because the formation of reduced metal catalysts triggers the reduction of protons, strong electron-withdrawing meso-substituents are thought to improve H2 evolution by reducing the energy cost of generating H2, despite the fact that they might decrease metal center basicity and make the metal center less reactive to protons. The CV of 141 displayed two reversible one-electron reduction waves at −1.28 and −1.82 V (vs ferrocene) in acetonitrile. The ﬁrst reduction is metal-centered, and the second reduction is ligand-centered. For simplicity, the formulation of [Ni-Por]− and [Ni-Por]2− will be used for one-electron and two-electron reduced species, respectively. The second reduction peak of 141 became a catalytic wave with the addition of acetic acid. In contrast, the ﬁrst reduction wave of 141 exhibited a pronounced catalytic activity when TFA was added. This result suggests that doubly reduced [Ni-Por]2− is involved in H2 production from acetic acid, whereas singly reduced [Ni-Por]− initiates HER with stronger acid TFA. At low TFA concentrations, the catalytic current had a ﬁrst-order dependence on acid concentrations up to 108 mM TFA, above which the current was unaﬀected. The overpotential determined at the half-wave position is ∼420 mV with 0.1 M TFA, and the icat/ip value of ∼80 is remarkable and represents one of the highest among well-established competent HER catalysts. At high TFA concentrations, the catalytic current exhibited linear dependence on the concentration of 141, conﬁrming a molecular nature of this catalysis. The controlled potential electrolysis of TFA with 141 at −1.49 V (vs ferrocene) gave a Faradaic eﬃciency of 97% for H2 generation and a TON of 4.2 in 30 min under an applied overpotential of 540 mV. Catalyst 141 was stable during this process. These results suggest that 141 is a highly eﬃcient HER catalyst from both the activity and stability points of view. Signiﬁcantly, when 141 was chemically reduced to [Ni-Por]− by sodium borohydride (NaBH4), subsequent TFA addition 3754

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 59. Stopped-ﬂow experiments showing (a) the generation of [Ni-Por]− by NaBH4 and (b, c) the reaction of [Ni-Por]− with (b) TFA and (c) acetic acid. Reprinted with permission from ref 261. Copyright 2016 Wiley-VCH.

Figure 60. Energy diagram for the HER catalyzed by Ni porphyrin 141. Free-energy values are given in kcal mol−1 and electrochemical potentials in volts. Reprinted with permission from ref 261. Copyright 2016 Wiley-VCH.

established from stopped-ﬂow measurements. The identiﬁcation of HER cycling through one-electron reduction and homolysis is signiﬁcant in providing valuable information to decrease the energy input for making H2 and will have broad implications for the design of new exquisite cycles for H2 generation. In addition to porphyrins, Cao and co-workers investigated the catalytic HER features of Cu corroles.269 Copper is a signiﬁcant transition metal involved in numerous biological redox processes.139,270 Synthetic Cu complexes with prominent biomimetic and redox chemistry have also been extensively investigated.270,271 However, molecular Cu species that can catalyze H2 production have rarely been recognized, largely because the reduction potential of low-valent Cu species is typically insuﬃcient to reduce protons. Recently, Sun and Wang reported the ﬁrst example of a homogeneous Cu-based HER electrocatalyst.272 A polypyridine ligand was used to provide the two-electron reduced Cu center with adequate reducing power for proton reduction. Because trianionic corrole ligands are very eﬀective in stabilizing high-valent metal centers and thus are able to oﬀer low-valent metal centers with large reducing powers, the examination of Cu corroles as HER catalysts is intriguing. Cu corroles 144−147 (Figure 61) were synthesized, and their catalytic HER features were examined. For all four complexes, a formally d8 CuIII center was suggested based on the short Cu-N bond distances and the diamagnetism observed in NMR spectroscopy and was consistent with previous studies on Cu corroles by Gross and co-workers273 and by Ghosh and co-workers.274 For neutral Cu corroles [(Cor)Cu]0, the [(Cor3−)Cu3+] formulation and the [(Cor•2−)Cu2+] formula-

completely regenerated the initial catalyst, a process that could be monitored by stopped-ﬂow experiments (Figure 59). These data suggest that [Ni-Por]− ﬁrst undergoes an oxidative protonation with TFA to yield a [H-Ni-Por] intermediate, which undergoes bimetallic homolysis to yield H2 and two equivalents of 141. Direct protonolysis of [H-Ni-Por] for H2 evolution could be ruled out in this system because this process would give H2 and [Ni-Por]+, a species that was not observed in the stopped-ﬂow experiments. Although the replacement of TFA by a large excess of acetic acid can considerably slow the reaction (Figure 59), no intermediates are detected. On the basis of these results and those from electrochemical studies, it can be concluded that the oxidative protonation of [Ni-Por]− is the rds for H2 evolution and that this step is followed by a fast bimetallic homolysis reaction. The changes in UV−visible spectroscopy only showed the kinetic behavior of the protonation step in this two-step process, and the detection of the [H-Ni-Por] intermediate is highly challenging, if not impossible, under the experimental conditions. The bimetallic homolysis reaction pathway was further supported by DFT calculations. The calculated ﬁrst and second reduction potentials of 141 are −1.19 and −1.77 V (vs ferrocene), which agree quite well with the experimental values (−1.28 and −1.82 V vs ferrocene). The calculated separation between the ﬁrst and second reduction is 0.58 V, which is almost identical to the value of 0.54 V from experimental measurements. The bimetallic homolysis reaction proceeded through a transition state (TS) with a very small activation energy barrier of 3.7 kcal mol−1 and a large exothermicity of 93.6 kcal mol−1 (pathway I in Figure 60). This calculated small activation energy barrier suggests a fast bimetallic reaction 3755

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

and −1.41 V versus ferrocene. The addition of TFA to an acetonitrile solution of 146 also resulted in the rapid growth of the catalytic current, despite that 146 was less eﬀective compared to 144 as an HER catalyst. The CV of 147 showed two reversible one-electron reduction waves at −0.29 V and −2.14 V versus ferrocene. Because the second reduction potential of 147 is too negative, it is not a competent HER catalyst. These results suggest that electron-withdrawing mesosubstituents on corroles are preferred because they shift the reduction potentials of Cu corroles in the positive direction and thus decrease the overpotential for electrocatalytic H 2 generation. As shown in Figure 62, the catalytic currents for 0.30 mM 144 increase with increased TFA and peak (enter an acidindependent region) at a TFA concentration of 160 mM. Interestingly, the addition of water can further increase the catalytic current to give a remarkable value of icat/ip = 303, which might be attributed to the enhanced proton delivery in the presence of water. Complex 144 has suﬃcient stability during electrocatalytic H2 generation based on results from a variety of experiments. The reaction mechanism has been carefully studied. A twoelectron reduced species [(Cor)Cu]2− is likely the catalytic active species based on CV measurements. To gain more insights into catalytic mechanisms, stopped-ﬂow and spectroelectrochemistry experiments were performed. In stoppedﬂow experiments, the complex [(Cor)Cu]2− was ﬁrst generated by the chemical reduction of 144 using NaBH4. Adding excess TFA to the solution of [(Cor)Cu]2− allowed the observation of an intermediate with absorption maxima at 328 and 423 nm that rapidly decayed by reacting with TFA to release H2 and regenerate the initial catalyst. The intermediate observed in stopped-ﬂow studies was believed to be the hydride [(Cor)CuH]−. In spectroelectrochemistry measurements, [(Cor)Cu]2− was generated by the CPE of 144 at −1.50 V (vs ferrocene), and the injection of TFA to the resulting solution caused an immediate growth of catalytic currents and the regeneration of the initial catalyst. Signiﬁcantly, electronic absorption spectra obtained from the chemical and electrochemical reduction of 144 are identical, conﬁrming the formation of this two-electron reduced species [(Cor)Cu]2− in both cases. A catalytic cycle of 144 for the HER is proposed with [(Cor)Cu]2− as the catalytic active species to reduce protons (Figure 63). Upon two-electron reduction, [(Cor)Cu]2− is generated and reacts with a proton to give [(Cor)CuH]−. The reaction of [(Cor)CuH]− with a second proton releases H2 and regenerates the initial [(Cor)Cu]0 (pathway A), which is suggested from both stopped-ﬂow and spectroelectrochemistry studies. However, the following two other possible reaction pathways from [(Cor)CuH]− to produce H2 cannot be excluded at this stage: (i) two such intermediates undergo bimolecular homolytic Cu-H bond cleavage to give H2 and two equivalents of [(Cor)Cu]− (pathway B) or (ii) [(Cor)CuH]− is further reduced to [(Cor)CuH]2− under electrochemical conditions for subsequent reaction with H+ to yield H2 and [(Cor)Cu]− (pathway C). For all three pathways, the formation of [(Cor)CuH]− is a common step. Signiﬁcantly, DFT studies revealed that the calculated electronic absorption spectrum of [(Cor)CuH]− agrees well with the spectrum obtained in stopped-ﬂow experiments, which conﬁrms the formation of a hydride intermediate. In addition, the calculated one-electron reduction potential of [(Cor)CuH]− is −1.06 V

Figure 61. Cu corroles 144−147 studied as HER electrocatalysts by Cao and co-workers.

tion are nearly degenerate in energy; the neutral [(Cor)Cu]0 formulation will be used for simplicity. The CV of 144 in dry acetonitrile showed two reversible one-electron reduction events at −0.12 V and −1.42 V versus ferrocene, and upon the addition of TFA, the second reduction wave became a large catalytic wave with the onset at −0.95 V versus ferrocene (Figure 62), corresponding to an overpotential of ∼450 mV

Figure 62. Successive CVs of 144 in acetonitrile with increasing concentrations of TFA followed by the addition of small aliquots of water. Conditions: 0.1 M (Bu4N)PF6 as the electrolyte; 3 mm GC working electrode; 100 mV s−1 scan rate; 20 °C. Redrawn from ref 269. Copyright 2015 American Chemical Society.

based on its half-wave potential. This result indicates that the two-electron reduced species of 144 is the catalytic active species for proton reduction. The CV of 145 showed two reversible one-electron reduction events at −0.15 V and −1.28 V versus ferrocene, but it degraded rapidly in acidic solution through an oxidative dimerization mechanism. The CV of 146 displayed two reversible one-electron reduction waves at 0.26 V 3756

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

solution. In addition, by using water from the tap, a local pond (IACS), the Ganges (delta region, high saline content and silt, and industrial waste contamination), the Bay of Bengal (saline and silt bearing), the Dead Sea (most saline containing natural H2O source), and the Sea of Galilee (natural fresh water lake), HER was observed from all water sources, which demonstrated that 72-F8 was an eﬃcient HER catalyst in natural environmental conditions. It is interesting to compare the catalytic H2 evolution performance of Co tpfc 72 with that of its derivatives halogenated at the eight β-pyrrole positions. By using 72 and halogenated analogues 72-F8, 72-Cl8, and 72-Br8, Gross, Dey, and co-workers reported a surprising eﬀect of halide substituents on the reduction potentials and HER activities.276 Electrochemical measurements showed that the ﬁrst reduction process was irreversible due to the dissociation of axial ligands upon the reduction of CoIII to CoII. The second reduction process of CoII to CoI was reversible, and all halogenated complexes were reduced at much more positive potentials (up to 600 mV) compared to 72. However, the second reduction potentials were −1.39, − 0.97, − 0.80, and −0.78 V (vs Ag/ AgCl) for 72, 72-F8, 72-Cl8, and 72-Br8, respectively, a trend which was not fully anticipated. DFT calculations on the ﬂuoro complex 72-F8 and chloro complex 72-Cl8 revealed that the stronger electron-withdrawing eﬀect of F vs Cl is overcome by the more eﬀective π-donation of ﬂuoride into the corrole π system. Consequently, the addition of the second electron to the dz2 orbital of CoII occurs at more negative potentials for βoctaﬂuoro than for β-octachloro Co corroles. In electrocatalytic HER, the CVs of halogenated 72 in acetonitrile in the presence of TFA showed catalytic waves with onset potentials of −0.4, − 0.55, and −0.6 V (vs Ag/AgCl) for 72-Br8, 72-Cl8, and 72-F8, respectively. Although the onset potential is the earliest for 72-Br8, which is followed by 72-Cl8 and 72-F8 (a similar trend for the second reduction event as discussed above), the β-octaﬂuoro Co tpfc 72-F8 is the most active catalyst in terms of the catalytic current. Signiﬁcantly, the two-electron reduced form of 72-Cl8 was generated by chemical reduction with Co(Cp*)2 and was characterized by NMR, which is consistent with a diamagnetic d8 CoI center in a square planar geometry. Subsequent reaction with TFA resulted in the paramagnetic shift and broadening of all resonances in 19F NMR spectra, implying the formation of a CoII complex of oneelectron reduced 72-Cl8. Further supporting evidence for the CoII oxidation state was obtained from UV−visible and EPR spectroscopy. On the basis of these results, a reaction mechanism was proposed by these authors with CoI as the catalytic active species for proton reduction (Figure 64).

Figure 63. Proposed electrocatalytic cycle for H2 evolution with 144. There are three possible reaction pathways from the hydride intermediate [(Cor)CuH]−: via the reaction with a proton to release H2 and regenerate the initial [(Cor)Cu]0 (pathway A, black), via a bimolecular reaction between two such intermediates to form H2 and [(Cor)Cu]− (pathway B, green), or via further one-electron reduction to give [(Cor)CuH]2− followed by reaction with a proton to yield H2 and [(Cor)Cu]− (pathway C, red). Redrawn from ref 269. Copyright 2015 American Chemical Society.

(vs ferrocene), which is 0.33 V more positive than the second reduction potential of 144. Thermodynamic analysis indicates that pathway A is endergonic and is unlikely to be directly involved in electrocatalysis. As a consequence, it is likely that pathway C is dominant in electrochemical experiments, in which [(Cor)CuH]− can undergo further one-electron reduction for subsequent reaction with a proton to form [(Cor)Cu]− and H2. Gross and co-workers reported HER catalyzed by a Co complex of β-octaﬂuoro tpfc 72-F8 under ambient conditions (see Figure 33 for the structure of 72).275 Complex 72-F8 was isolated as the bis-pyridine complex, and its single-crystal X-ray structure showed a perfectly planar macrocycle. The CV of 72F8 in degassed acetonitrile displayed a reversible CoII/CoI redox couple at −0.97 V (vs Ag/AgCl), and this reduction wave became a large electrocatalytic wave upon the addition of TFA with an onset potential of −0.75 V (vs Ag/AgCl). This result implies that the CoI state is the active form for HER catalysis. When physiadsorbed onto an EPG electrode for heterogeneous catalysis, the CV of 72-F8 in a 0.5 M H2SO4 solution showed a large electrocatalytic current that began to increase at a potential of −0.5 V (vs Ag/AgCl). Similar catalytic features were observed when 72-F8 was adsorbed on SAM-covered Au/ Ag electrodes. The TOF for the HER under anaerobic conditions at room temperature was 600 and 1140 s−1 at −0.7 and −0.8 V (vs Ag/AgCl), respectively. CPE at −0.8 V (vs Ag/AgCl) gave a TON ≫ 107 in 16 h. SERRS was used to investigate the in situ electro-active species involved in HER catalysis. The results conﬁrmed that the active form of the catalyst on SAM-covered Au/Ag electrodes has the CoI oxidation state. Signiﬁcantly, these authors showed that Co corrole 72-F8 was able to catalyze the formation of H2 under aerobic conditions. The Faradaic yield for H2 production under aerobic conditions was 52% under −0.8 V (vs Ag/AgCl) in a 0.5 M H2SO4

3.2. HER Catalysts Based on Porphyrin and Corrole Architectures

Cofacial bisorganometallic diporphyrin complexes of Ru and Os were shown by Collman and co-workers to be active for catalytic HER more than 20 years ago.277 Two plausible mechanisms for H2 evolution were suggested by these authors. The ﬁrst is the formation of key dihydride intermediates, which undergo bimetallic homolysis to produce H2. Alternatively, successive protonation and H2 elimination occur at a single metal center, which is described as a unimetallic process. In 2014, Bren and co-workers reported the H2 evolution from neutral water under aerobic conditions catalyzed by Co microperoxidase-11 (148, Figure 65).278 The HER was performed in 2 M pH 7.0 phosphate buﬀer. CPE experiments 3757

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

the introduction of meso-Fc substituents considerably increases the catalytic activity of Cu and Pd porphyrins toward HER. DFT calculations suggested that the two-electron reduced species is active for proton reduction and that it reacted with a proton to give a phlorin anion. Subsequent protonation released H2 and regenerated the catalyst. Nocera and co-workers reported H2 evolution by using hangman Ni porphyrins 153 and 154 (Figure 66) and

Figure 64. Proposed reaction mechanism for H2 evolution catalyzed by Co corroles. Redrawn with permission from ref 276. Copyright 2014 Royal Society of Chemistry.

Figure 66. Hangman metal porphyrins 153−158 studied as HER catalysts by Nocera and co-workers.

Figure 65. Metal porphyrins 148−152 studied as HER catalysts.

demonstrated the role of pendant proton relays and PCET on the HER.280 The CV of 154 in acetonitrile displayed two reversible one-electron waves centered at −1.39 and −1.96 V (vs ferrocene), corresponding to the formal NiII/NiI and NiI/ Ni0 redox couples, respectively. The CV of 153 displayed two similar reversible waves centered at −1.37 and −2.01 V (vs ferrocene) and an additional irreversible wave that peaked at −1.76 V (vs ferrocene). The ﬁrst reversible wave for 153 was assigned to the formal NiII/NiI redox couple. The irreversible wave was suggested to be associated with the H2 catalysis, which increased upon titration with benzoic acid. Moreover, the addition of K2CO3 to the solution of 153 eliminated the irreversible wave due to the deprotonation of the pendant acid group, further conﬁrming the catalytic H2 formation feature of this irreversible peak associated with the hangman group. Spectroscopic and DFT studies indicated that the ﬁrst reduction was metal-centered to give a NiI state and that the second reduction was ligand-centered. On the basis of results from CV modeling and DFT calculations, a reaction mechanism for HER with 153 was proposed. Upon oneelectron reduction, an intramolecular proton transfer from the pendant carboxylic acid to NiI resulted in a formal NiIII-H species. Subsequent reduction generated NiII-H, which was followed by another reduction to give NiI-H. Protonation of this hydride intermediate eventually produced H2 in a facile step. The enhanced catalytic activity with a pendant proton relay conﬁrms the signiﬁcance of PCET in HER. The HER features of hangman Fe porphyrins 155−157 and the corresponding Fe porphyrin 158 (Figure 66) were

at −1.5 V (vs Ag/AgCl) using a mercury pool working electrode showed the increase in current as a function of 148 concentration, yielding a Faradaic eﬃciency of 95 ± 3% and a TOF of 6.7 s−1. Because 148 was not stable during catalysis, only a 10 min CPE was conducted to measure the TOF. Signiﬁcantly, the HER activity of 148 was not aﬀected in the presence of oxygen. CPEs under argon and air gave Faradaic eﬃciencies of 98 ± 2% and 85 ± 5%, respectively. TON values obtained from 4-h CPE at −1.5 V (vs Ag/AgCl) are 2.5 × 104 and 1.9 × 104 under argon and air, respectively. Benniston and co-workers reported the HER features of Cu (149) and Pd (150) porphyrins bearing four meso-Fc substituents (Figure 65).279 The CV of 150 in dried DMF showed the Fc+/Fc redox wave at 0.08 V (vs ferrocene) and two porphyrin-centered quasi-reversible one-electron waves at −1.72 and −2.21 V (vs ferrocene). The CV of 149 in DMF was similar to the Pd analogue with three processes at −2.23, − 1.70, and 0.08 V (vs ferrocene). Adding TFA to the DMF solution of 149 and 150 gave a catalytic wave at −2.0 V (vs ferrocene). Addition of the weaker acid (Et3N)HCl resulted in no catalytic activity for 150, whereas 149 showed similar behavior as for TFA addition. To elucidate the importance of the meso-Fc substituents, the activities of Cu (151) and Pd (152) porphyrin analogues bearing four meso-phenyl substituents were also examined. Compared to 150, the catalytic HER wave with 152 was much smaller and was shifted to a more negative potential by ∼200 mV. For 151, there was no catalytic response with (Et3N)HCl. These results indicate that 3758

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Table 9. E1/2 Values (without Acids) and log k Values (with Diﬀerent Concentrations of Acids) for Fe Porphyrins 155−158 log k (M−1 s−1)

E1/2 (V vs ferrocene) III

II

catalyst

Fe /Fe

155 156 157 158

−0.629 −0.632 −0.567 −0.627

II

Fe /Fe

I

−1.431 −1.466 −1.551 −1.499

I

Fe /Fe

0

−2.046 −2.040 −2.098 −2.055

L/L

−

−2.522 −2.532 −2.569 N/A

investigated by Nocera and Graham.281 The catalytic eﬃciency of these Fe porphyrins was evaluated by FOW analysis, which is very useful for evaluating the activity of a molecular homogeneous electrocatalyst even under conditions of substrate consumption, catalyst decomposition, or product inhibition. The electrochemical and electrocatalytic behaviors of these catalysts are summarized in Table 9. The HER was performed in acetonitrile with or without p-toluenesulfonic acid as the proton source. The CVs of 155−158 under these conditions displayed catalytic waves corresponding to the FeII/ FeI redox couple, indicating that FeI was the catalytic active species for proton reduction. This result is diﬀerent from that reported by Savéant and co-workers in which a Fe0 porphyrin was suggested as the active species.264 This diﬀerence was caused by the diﬀerent acids used in these studies; weak triethylamine hydrochloric acid was used by Savéant and coworkers, whereas strong toluenesulfonic acid was used by Nocera and Graham. A similar observation was reported by Cao and co-workers as discussed in the previous section.261 On the basis of the catalytic HER data, it was concluded that the hangman eﬀect resulted in a high local proton concentration with a nearly 3 orders of magnitude increased rate of catalysis. The hangman eﬀect on catalytic HER was investigated by Nocera and co-workers using hangman Co porphyrins bearing a pendant carboxylic group (95, Figure 42) or a bromide group (159, structure not shown) and Co tetra(pentaﬂuorophenyl)porphyrin (100, Figure 42).282,283 Complexes 95 and 159 exhibited almost identical reversible CoII/CoI reduction potentials in acetonitrile: −1.08 and −1.10 V (vs ferrocene) for 95 and 159, respectively.282 Interestingly, the CV of 159 showed a reversible CoI/Co0 reduction wave at −2.14 V (vs ferrocene), whereas the CV of 95 showed an irreversible CoI/ Co0 wave that was positively shifted by ∼200 mV. These authors proposed that for 95, an immediate proton transfer from the hanging carboxyl group to the Co center produced a CoII-H intermediate upon the reduction to Co0 because the second wave of 159 also became irreversible upon the addition of external benzoic acid (pKa = 20.7 in acetonitrile). In the presence of benzoic acid, the CVs of both 95 and 159 displayed catalytic cathodic waves at the CoI/Co0 redox couple, and the CoII/CoI redox couple was mostly unaﬀected, indicating that Co0 was the active species for the proton reduction with benzoic acid. The overpotential for the HER with 95 is ∼120 mV lower than that of 159 at 3 mM acid concentration. However, in the presence of stronger tosic acid (pKa = 8.3 in acetonitrile), both 95 and 159 exhibited catalytic cathodic waves ca. −1.50 (vs ferrocene), indicating the obviation of the hangman eﬀect with strong acids. Although the CoII/CoI redox couple became irreversible with tosic acids, the electrocatalysis occurred at signiﬁcantly higher potentials than the CoII/CoI potential. This result indicates that CoI is protonated by the stronger tosic acid to give a CoIII-H species. However, this CoIII-H is not suﬃciently basic to drive protonolysis and needs to be further reduced to CoII-H for subsequent H2 generation.

5 mM

10 mM

15 mM

20 mM

3.74 4.25 5.06 4.96

3.72 4.43 5.44 5.02

3.69 4.48 5.81 5.08

3.66 4.53 6.12 5.17

The PCET kinetics for HER catalyzed by hangman Co porphyrins 95 and 159 were further investigated by Nocera and co-workers.283 They determined that the protonation of Co0 to form CoII-H is the rds. The proton transfer rate constant (kPT) from the hanging carboxyl group to Co0 was 8.5 × 106 s−1, suggesting a rapid intramolecular pathway of 95 for the formation of CoII-H to produce H2. However, for the nonhangman Co porphyrin, the intermolecular proton transfer rate constant (kPT) from benzoic acid to the Co0 state of 100 to form CoII-H was estimated to be on the order of 1000 M−1 s−1, and it was ∼2500 M−1 s−1 for the intermolecular proton transfer from benzoic acid to the Co0 center of 159. These results suggest that the presence of a hanging carboxyl group proximate to the Co center enhances the reaction rate to one equivalent to an eﬀective benzoic acid concentration >3000 M. Therefore, this work provides direct evidence for the “hangman eﬀect” in promoting HER. Hammes-Schiﬀer, Nocera, and co-workers further investigated the HER catalyzed by hangman Co and Ni porphyrins by using theoretical methods.284,285 Under strong acid conditions, the hangman Co porphyrin was found to evolve H2 by the protonation of the one-electron reduced species at the Co site to produce the hydride intermediate, as found for its nonhangman analogues. However, in weak acid, the HER mechanism was thought to be initiated after two one-electron reductions. Subsequent intramolecular proton transfer from the carboxylic acid to the closest meso carbon of the porphyrin ring leads to the formation of a phlorin complex. The ﬁnal steps involve the reprotonation of the hangman carboxyl group, spontaneous one-electron reduction, and the elimination of H2. More recently, these authors studied the HER mechanism for the hangman Ni porphyrin and found that a phlorin intermediate was likely to be involved in both weak-acid and strong-acid regimes. However, the phlorin intermediate formed through the protonation of the two-electron-reduced Ni porphyrin is stable and does not react with acids to generate H2.

4. OXYGEN EVOLUTION REACTION 4.1. Simple OER Catalysts

In 2009, Sun and co-workers investigated O−O bond formation using Mn tris(p-nitrophenyl)corrole 160 (Figure 67).286 The oxidation of MnIII corrole 160 by t-BuOOH gave an MnV-oxo species, and the subsequent addition of an aqueous solution of (n-Bu)4N(OH) caused rapid O2 evolution. Intermediates proposed in the catalytic cycle, including MnVO and MnIV-OO−, were all characterized and conﬁrmed by electronic absorption measurements, high-resolution mass spectrometry, and isotopic 18O-labeling experiments. Importantly, experiments using 18O-labeled water unambiguously illustrated that the O atoms in evolved O2 molecules originated from water. Although no catalytic OER was reported with complex 160 in this paper, this work is signiﬁcant because it 3759

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 68. Proposed O−O bond formation pathways with Co porphyrin OER catalysts. Redrawn with permission from ref 288. Copyright 2013 Royal Society of Chemistry.

Figure 67. Mn corrole 160 and a possible mechanism of the O2 evolution cycle. Redrawn from ref 286. Copyright 2009 American Chemical Society.

provides direct experimental evidence that O−O bond formation via the nucleophilic attack of a MnVO species by water (i.e., hydroxide) is feasible; thus, it sheds light on the mechanism of water oxidation to O2 occurring at the oxygen evolving complex in nature. Another Mn-based OER catalyst was reported by D’Eramo and co-workers in 2014.287 The cationic water-soluble MnIII 5,10,15,20-tetra(N-methylpyridyl)porphyrin could catalyze the electrochemical oxidation of water to O2 in both pH 7 and pH 10 phosphate buﬀer solutions, although the chloride counteranions of the MnIII porphyrin were also oxidized during this process. If the chloride-free Mn porphyrin was used, the catalytic current decreased by approximately 40%. In 2013, Sakai and co-workers reported photocatalytic water oxidation with three water-soluble Co porphyrins-CoII tetrakis(4-carboxylphenyl)porphyrin 54 (Figure 29), CoII tetrakis(1methyl-pyridinium-4-yl)porphyrin 61, and CoII tetrakis(4sulfonatophenyl)porphyrin 62 (Figure 32).288 A typical threecomponent system was used with Co porphyrin as the catalyst, [Ru(bpy)3]2+ as the photosensitizer, and Na2S2O8 as the sacriﬁcial oxidant. The photocatalytic activities of these Co porphyrins are pH-dependent and peak at pH 11. The TOF measured for O2 evolution at pH 11 is 0.138 s−1 for 54, 0.118 s−1 for 61, and 0.170 s−1 for 62. Kinetic studies revealed a second-order dependence of the initial reaction rate on the catalyst concentration, indicating that the rds of the catalytic cycle is either a bimolecular radical coupling of two Co-oxyl radicals (Figure 68) or the disproportionation of two CoIV species to form CoV and CoIII species. However, these Co porphyrin OER catalysts decomposed to produce less active (or inactive) species during catalysis; the catalytic activity was decreased in subsequent runs. To improve the stability of Co porphyrin OER catalysts, Sakai and co-workers synthesized a new ﬂuorinated Coporphyrin 161 (Figure 69) and examined its photocatalytic features.289 It is thought that singlet dioxygen (1O2), which is a highly active oxidant capable of rapidly reacting with organic substrates, can be generated during water oxidation. Because steric shielding can protect porphyrin rings from 1O2 attack, these authors introduced two ﬂuorine groups to the 2- and 6-

Figure 69. Co porphyrin OER catalyst 161 and the proposed catalytic cycle for water oxidation. Redrawn with permission from ref 289. Copyright 2015 Wiley-VCH.

positions of each meso-3-phenylsulfonate substituent. Photocatalytic OER studies with [Ru(bpy)3]2+ and Na2S2O8 showed that complex 161 had a greater TON (approximately 2-fold) compared with 62. The improved performance of 161 is due to its increased resistance toward 1O2, the existence of which is conﬁrmed by using 9,10-diphenylanthracene as a chemical probe. The maximum water oxidation rate measured corresponds to a TOF of 1.1 s−1 (TON = 570). Unlike 62 with a second-order dependence of the initial reaction rate on the catalyst concentration, the reaction rate was ﬁrst order with respect to the catalyst concentration of 161. In addition, the reaction rate does not vary with the concentration of [Ru(bpy)3]2+ and is independent of light intensity. These results suggest that the rds of the 161-mediated OER catalytic cycle is the nucleophilic attack of water or hydroxide at the Cooxo species (Figure 69) rather than the formation of [Ru(bpy)3]3+ or the bimolecular oxyl−oxyl coupling. The steric hindrance resulting from the bulkier meso-substituents of 161 is believed to be the reason leading to this reaction mechanism change. Du, Cao, and co-workers reported the electrocatalytic water oxidation with Co tetraphenylporphyrin 52 (Figure 29) and Co tetrakis(4-bromophenyl)porphyrin 162 (not shown).290 When deposited onto ﬂuorine-doped tin oxide (FTO) electrodes, the catalyst ﬁlm showed good activities for electrocatalytic water oxidation in 0.5 M pH 9.2 borate buﬀer with the onset overpotentials of 530 and 580 mV for 52 and 162, respectively. CPE under an applied potential of 1.3 V (vs Ag/AgCl) conﬁrmed the evolution of O2 with a TOF of 0.50 s−1 for 52 and 0.40 s−1 for 162. These results show that 52 is more active than 162 in catalyzing the electrochemical water oxidation. The stability of the catalyst ﬁlms was evaluated, showing the 3760

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Interestingly, the catalytic current increased with the concentration of phosphate up to 60 mM, but it started to decrease at higher phosphate concentrations. A reaction mechanism was proposed (Figure 71) with the two-electron oxidized species Por•+-CoIV-O as the catalytic active species for water oxidation. As shown in this mechanism, buﬀer anions have two opposing roles in catalytic water oxidation, a facilitating role of accepting the proton of the attacking water molecule or an inhibitory role of blocking the Co center via coordination. Similarly, the nature of the buﬀer anion can also aﬀect the activity of 163 for water oxidation; the onset potential at pH 7 decreases with more basic buﬀer anions, and the plot of the onset potential versus the pKa of buﬀer anions gives a linear correlation with a slope of −54 mV pKa−1. This result indicates that the generation of the active species Por•+-CoIV-O requires the removal of one proton from the resting state. In 2015, Cao and co-workers reported the ﬁrst Ni porphyrin OER catalyst.294 Water-soluble cationic NiII tetrakis(1-methylpyridinium-4-yl)porphyrin 166 (Figure 72) was synthesized

molecular nature of these water-insoluble Co porphyrin catalysts. Doctorovich and co-workers reported the immobilization of Co porphyrins on a gold surface via covalent gold-sulfur bonds.291,292 The modiﬁed gold electrode with coated Co porphyrin ﬁlms displayed OER activities in aqueous solutions. In 2013, Groves and Wang reported eﬃcient water oxidation electrocatalyzed by homogeneous cationic CoIII porphyrins 163−165 (Figure 70) in neutral phosphate buﬀer solutions.293

Figure 70. Water-soluble Co porphyrins 163−165 studied as OER catalysts.

The CoIII form of these catalysts was generated by the electrochemical one-electron oxidation from their corresponding CoII counterparts at an applied potential of 300 mV (vs Ag/ AgCl). Titration studies indicated that the resting state of these catalysts at pH 7 is H2O-CoIII-OH. The catalytic activity of 163−165 has the following sequence: 163 > 164 > 165. The CV of 163 displayed a strong catalytic current with an onset potential of ∼1.2 V (vs Ag/AgCl) in 0.2 M pH 7 aqueous phosphate buﬀer, and the catalytic current had a linear dependence on the catalyst concentration. The O2 formation rate constant was 1.4 × 103 s−1. The catalyst stability during sustained electrocatalytic O2 evolution was established by a variety of experimental methods. The normalized catalytic currents decreased with increasing scan rates, indicating a chemical rds, likely O−O bond formation. This hypothesis was further conﬁrmed by KIE experiments. The CV of 163 taken in D2O showed a signiﬁcantly smaller catalytic current compared to that obtained in H2O with a KIE value of 2.8 calculated according to KIE = kcat,H2O/kcat,D2O = (icat,H2O/icat,D2O)2.

Figure 72. Ni porphyrin 166 and the thermal ellipsoid plot (50% probability) of its X-ray structure.

and examined as a homogeneous electrocatalyst for water oxidation. The CV of 166 displayed a pronounced catalytic wave with an onset at ∼1.0 V (vs NHE) in 0.10 M pH 7.0 phosphate buﬀer. The onset overpotential with complex 166 is lower than with the Co porphyrin analogue 164 (Figure 70) by

Figure 71. Proposed mechanism for water oxidation catalyzed by 163. Roles of the buﬀer anion (B) include serving as a base to assist proton transfer and inhibiting the catalyst through coordination. Redrawn with permission from ref 293. Copyright 2013 Proceedings of the National Academy of Sciences of the United States of America. 3761

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

rds in the catalytic cycle. This water nucleophilic attack O−O bond formation step proceeds via a concerted O atom-proton transfer pathway. If a cluster of four water molecules was included in the calculation to accept the proton from the attacking water molecule, an energy barrier of 24.9 kcal mol−1 was found. However, if an acetate anion was used as the proton acceptor, the energy barrier decreased to 18.0 kcal mol−1. The transition state structure shows an increased O−O bond length (2.15 vs 1.96 Å, Figure 74), indicating a much earlier transition state with an acetate anion compared to four water molecules acting as the proton acceptor.

approximately 200 mV and is even lower than that for the biomimetic Cu catalyst recently reported by Lin and coworkers.295 This value is the lowest among the existing molecular OER catalysts in neutral aqueous solutions. The normalized catalytic currents decreased with increasing scan rates, suggesting a catalytic process with a chemical rds, which is likely the O−O bond formation. A KIE value of 1.55 was measured, which was consistent with a rate-limiting O−O bond formation step. The molecular and homogeneous nature of catalysis with 166 was carefully investigated and was conﬁrmed by more than ten lines of evidence. Signiﬁcantly, complex 166 was catalytically active in the pH range of 2.0−8.0, a result that strongly argues against NiOx materials acting as the catalyst for water oxidation because these species are unstable in acidic solutions. The TOF value of 0.67 s−1 at room temperature was also determined. Similar to its Co porphyrin analogue 164, an inhibitory eﬀect was observed with 166 at high phosphate concentrations. The catalytic current increased rapidly and considerably with the increase of phosphate concentration up to 33 mM, increased slowly in the phosphate concentration range of 35−65 mM, and started to decrease with further increased phosphate concentrations. This competitive coordination-based inhibition was veriﬁed by inhibition experiments with acetonitrile. The coordination of acetonitrile at the NiII site could be monitored by the changes observed in UV−visible spectra, and the CV of 166 in the presence of acetonitrile showed obviously inhibited catalysis. These results conﬁrm that the molecular catalysis of water oxidation by 166 is Ni-centered. Mechanistic studies suggest that a two-electron oxidized species of 166 with a formal NiIV oxidation state is the catalytic active species for water oxidation. On the basis of CVs in diﬀerent buﬀer solutions, diﬀerential pulse voltammetry (DPV), and DFT results, a reaction mechanism with relative energy information was proposed (Figure 73). Theoretical calculations demonstrate that water oxidation with 166 proceeding via the nucleophilic attack of Por-NiIII-O• by a water molecule is feasible, and the O−O bond formation step is thought to be the

Figure 74. Transition-state structure of the O−O bond formation step via nucleophilic water attack. (a) [Por-NiIII-O•]4+ with four water molecules and (b) [Por-NiIII-O•]4+ in the presence of one water molecule and one acetate anion. The hydrogen atoms in the porphyrin ring are not shown for clarity. Reprinted from ref 294. Copyright 2015 American Chemical Society.

It is interesting to compare the catalytic mechanism of Ni porphyrin 166 and its Co porphyrin analogues. For Co porphyrin, a formal CoV porphyrin is likely the active catalytic species for water oxidation, but for Ni porphyrin, a formal NiIV porphyrin, Por-NiIII-O•, is thought to have a suﬃcient oxidizing power to react with water. This diﬀerence causes the onset of the catalytic wave for water oxidation with 166 to decrease by approximately 200 mV compared to its Co porphyrin analogues. This improvement was likely because the redox potential of high-valent Ni atoms is considerably higher than that of Co in the same valence state. In 2014, Cao and co-workers reported the OER catalysis with CoIII tpfc 72 (Figure 33).296 The electrochemical, spectroscopic, and structural properties as well as the axial ligand exchange reactions of 72 have been reported by Gross and co-workers.297 The MnIII tpfc 73 (Figure 34) and four other MnIII corroles with diﬀerent numbers of p-nitro groups (0−3) on the three meso-phenyl substituents were synthesized, and their catalytic activities were also examined. The ﬁrst and second oxidation waves for these metal corroles are summarized in Table 10, showing that the introduction of

Figure 73. Proposed catalytic cycle for water oxidation with Ni porphyrin 166. Redrawn from ref 294. Copyright 2015 American Chemical Society. 3762

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

electron-withdrawing meso-substituents increases the oxidizing power of metal corrole complexes.

the proton acceptor. Importantly, this calculated barrier with an acetate ion as the base is in good agreement with the experimental activation barrier (18.4 kcal mol−1 estimated from TOF = 0.20 s−1 using the transition-state theory). A subsequent 1H+/1e− PCET reaction generates [Cor-CoIII-OO•−]− at a potential of −0.56 V (vs NHE), and further one-electron oxidation yields [Cor•+-CoIII-OO•−] at a potential of −0.02 V (vs NHE). The resulting [Cor•+-CoIII-OO•−] species was able to release O2 by displacement with water, a process with a small energy barrier of 3.6 kcal mol−1. It can be observed based on calculations that the O−O formation step is the rds. In addition to OER catalysis, the CV of a GC electrode coated with 72 showed a large electrocatalytic current with an onset at −0.90 V (vs Ag/AgCl) in a 0.50 M sulfuric acid solution (pH = 0.5), which was due to the formation of H2. The TOF for electrocatalytic HER with 72 was then determined to be 1010 s−1 and 6150 s−1 at −1.0 and −1.1 V (vs Ag/AgCl), respectively. A CoI corrole was proposed as the active species that reacted with a proton to give a CoIII-hydride intermediate, which subsequently reacted with a second proton to produce H2 and the initial catalyst. Although Mn tpfc 73 was shown to be ineﬃcient and unstable for electrocatalytic OER in neutral phosphate buﬀer by Cao and co-workers,296 Mn complexes of tpfc ligand derivatives were shown to be active for OER in strong basic aqueous solutions by Dey, Schöfberger, and co-workers.298 Complexes 167 and 168 (Figure 76) were synthesized and characterized. The CV of 167 in acetonitrile showed two oxidations at 0.53 and 0.78 V (vs Ag/AgCl), corresponding to MnIV/MnIII and MnV/MnIV redox processes, respectively. An additional irreversible catalytic wave appeared at 1.38 V (vs Ag/AgCl), which increased upon the addition of a NaOH aqueous solution. This result suggests catalytic O2 evolution. The function (icat/ip)2 has a linear relationship with the NaOH concentration, which is consistent with a pseudo-ﬁrst-order process with respect to the base. This pseudo-ﬁrst-order rate constant was calculated as 11.4 s−1 with 25 mM NaOH and a scan rate of 10 mV s−1. To test Mn corroles as heterogeneous catalysts, their adsorption behaviors and electronic properties on diﬀerent solid surfaces were studied. Individual molecules are assumed to have an approximately parallel orientation at the solid−liquid interface with respect to the surface. The presence of face-toface stacked dimers is also suggested because brighter features are occasionally observed with an apparent height approximately twice that of the majority of the other molecules in the layer. In contrast, Mn corroles were found to form highly ordered monolayer ﬁlms at a solid-vacuum interface on silver with individual molecules orienting approximately parallel to the substrate surface. Surface-immobilized catalysts existed as the MnIII state based on electrochemical studies and were expected to exhibit similar catalytic properties as catalysts dissolved in the solution. Signiﬁcantly, catalytic OER was conﬁrmed by using surfaceimmobilized catalyst 167 on an EPG electrode in strong basic solutions. At pH 7.0, very small OER currents were obtained, which is similar to the results reported by Cao and coworkers.296 However, at pH 11.0, large catalytic OER currents were observed at potentials >1.3 V (vs Ag/AgCl). With an applied potential of 1.4 V (vs Ag/AgCl) in CPE studies, a TON of 1.90 × 104 over 11.1 h was obtained with a Faradaic eﬃciency of 82%, and a TOF of 0.47 s−1 was calculated. In addition to OER, 167 is capable of catalyzing ORR with

Table 10. Electrochemical Oxidation Potentials of Co and Mn Corrole Complexes in an Acetonitrile Solution E1/2 (V vs ferrocene) complex Co tpfc 72 Mn tpfc 73 Mn corrole, Mn corrole, Mn corrole, Mn corrole,

no-NO2 one-NO2 two-NO2 three-NO2

1st oxidation

2nd oxidation

0.38 0.26 −0.21 −0.15 0.13 0.17

0.94 0.80 0.62 0.65 0.70 0.73

When deposited on ITO electrodes, the CV of 72 displayed a strong catalytic wave for water oxidation in 0.1 M pH 7.0 phosphate buﬀer with an onset at ∼1.15 V (vs Ag/AgCl), corresponding to an overpotential of 530 mV. For Mn corroles, although their oxidizing powers were comparable, they were not as eﬃcient and stable as 72 and underwent decomposition during the electrocatalysis in neutral phosphate buﬀer. The molecular nature of the catalysis with 72 was further conﬁrmed, and a TOF of 0.20 s−1 was calculated at an applied potential of 1.4 V (vs Ag/AgCl). On the basis of experimental and theoretical results, a catalytic cycle for 72-mediated water oxidation was proposed (Figure 75). The reaction cycle starts from CoIII corrole [Cor-

Figure 75. Proposed catalytic cycle for water oxidation with 72. Free energy values in kcal mol−1 (red), electrochemical potential in volts (blue), and pKa values in log units (purple). Reprinted with permission from ref 296. Copyright 2014 Royal Society of Chemistry.

CoIII-OH2]. Two sequential 1H+/1e− oxidation steps result in the formation of [Cor•+-CoIII-OH] and [Cor•+-CoIII-O•−], respectively. The two-electron oxidized species [Cor•+-CoIIIO•−] is the catalytic active species for the O−O bond formation. The calculated reaction barrier for nucleophilic water attack to produce [Cor-CoIII-OOH]− is 29.9 kcal mol−1 with a cluster of four water molecules acting as the proton acceptor, or it is 18.1 kcal mol−1 with an acetate anion acting as 3763

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 76. Mn corroles 167 and 168 studied as OER and ORR catalysts by Dey, Schöfberger, and co-workers.

catalytic currents appearing at a potential of −0.35 V (vs Ag/ AgCl). The number of electrons involved in O2 reduction was 2.3, and the second order rate constant was 3.81 × 103 M−1 s−1 for the two-electron reduction of O2 at −0.64 V (vs Ag/AgCl). As an alternative to water splitting, simpler light-driven chemical transformations, such as the oxidation of halide anions, are also attractive for solar energy conversion. Gross and co-workers recently reported the photocatalytic oxidation of bromide to bromine with metallocorroles 169−172 (Figure 77).299 Complex 172 was not competent for the photoFigure 78. Proposed reaction mechanism for the photocatalytic oxidation of bromide to bromine with metallocorroles as both catalysts and photosensitizers (PS). Redrawn with permission from ref 299. Copyright 2015 Wiley-VCH.

Figure 77. Metal complexes of β-octabromo tpfc 169−172 studied as catalysts for the photocatalytic oxidation of bromide to bromine by Gross and co-workers.

oxidation of bromide because it was demetalated by hydrobromic acid. The λmax (Soret band) values of 171, 169, and 170 are 428, 437, and 438 nm, respectively. The redox potentials were determined to be in the order 171 (1.40 V vs Ag/AgCl) > 170 (1.14 V vs Ag/AgCl) > 169 (1.0 V vs Ag/AgCl), implying that their photoexcited states are suﬃciently high to oxidize bromide. With the use of a blue-light-emitting diode (LED, λmax(emission) = 450 nm, power intensity at 150 mW cm−2), the TOFs obtained in aerobic conditions for the production of bromine were 93, 290, and 341 h−1 with 171, 169, and 170 as the catalysts, respectively. Thus, the catalytic eﬃciency for the photooxidation of bromide to bromine is in the order 170 > 169 > 171. On the basis of these results, two conclusions are obtained: (1) the metallocorroles whose absorption is better matched with the wavelength emitted by the LED light perform better and (2) the complexes that have more positive redox potentials are more eﬃcient catalysts. Co β-octabromo-tpfc 72Br8 was not active for this reaction. A plausible reaction mechanism was proposed (Figure 78) with metallocorroles acting as both catalysts and photosensitizers.

Figure 79. Binuclear Mn porphyrins 173−175 studied as OER catalysts by Naruta and co-workers.

subsequent work.301 The addition of four equivalents of oxidants to 174 aﬀorded the [MnVO]2 species (Figure 80), which was stable in basic solutions. This [MnVO]2 species was characterized by resonance Raman, showing two isotopesensitive bands at 791 (16O2)/757 (18O2) cm−1 and 518 (16O2)/491 (18O2) cm−1. The 791 band was not sensitive to H/

4.2. OER Catalysts Based on Porphyrin and Corrole Architectures

Catalytic OER with binuclear Mn porphyrins was reported by Naruta and co-workers in 1994.300 Complexes 173−175 (Figure 79) could catalyze electrocatalytic water oxidation to evolve O2 in an aqueous-acetonitrile solution, whereas corresponding monomeric Mn porphyrins were not active. Complex 175 showed the highest eﬃciency. With the use of 174, these authors investigated the reaction mechanism in a

Figure 80. Formation of the [MnVO]2 species and its subsequent reaction with protons to evolve O2 and regenerate the initial binuclear MnIII catalyst. 3764

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

D labeling, but the latter band shifted to 503 cm−1, using a deuterated sample. On the basis of these results, these two bands were assigned to MnVO (for 791) and MnV-OH (for 518) stretching vibrations. Signiﬁcantly, the addition of triﬂic acids to [MnVO]2 caused the evolution of O2 and the regeneration of 174 in a few seconds. The formation of O2 (yield 92%) was conﬁrmed by mass spectrometry along with 18 O-labeling experiments. The O−O bond could be formed either via the nucleophilic attack of water on the MnVO unit or by a coupling reaction between two oxo groups of the [MnVO]2 species. Sun and co-workers investigated the electrocatalytic OER features of mononuclear and binuclear Mn and Cu corroles 176−179 (Figure 81).302 Electrochemical studies revealed that

Br8 was at 1.25 V (vs Ag/AgCl), corresponding to an overpotential of 600 mV, and a CoIV-Cor•+ was proposed as the catalytic active species. It was suggested by these authors that the hanging carboxyl group was able to preorganize water molecules within the hangman cleft to facilitate the O−O bond formation. The “hangman eﬀect” in the OER and HER was recently summarized by Nocera and co-workers.305 Lai, Cao, and co-workers investigated the O−O bond formation step of a variety of β-octaﬂuoro hangman metal corroles using DFT calculations.306 The hanging carboxyl group was proposed to serve as a proton acceptor during the water nucleophilic attack on the metal-oxo unit. The eﬃciency of catalytic water oxidation was studied by exploring the eﬀect of metal oxidation states on the O−O bond formation process. The calculated barriers for this key step were found to vary in the following order: Co(V) ≪ Fe(V) < Mn(V) < Ir(V) < Co(IV) < Ru(V) < Ir(IV) < Mn(IV). This trend is controlled by the ease of the two-electron reduction and the aﬃnity of the singly reduced metal-oxo species to the HO• radical. In contrast, Cramer and Ertem investigated the OER catalytic cycle of hangman Co β-octaﬂuoro corrole 101-Br8 using quantum chemical modeling.307 An interesting feature of this catalyst is the “early”, noninnocent character of the corrole ligand. The ﬁrst-solvent-shell water was found to act as a catalyst to shuttle a proton from the attacking water to the carboxylate group. Fluorination of the supporting corrole appears to modulate the electrophilicity of the metal-oxo fragment and to mitigate the decomposition reaction. In 2014, Baran, Grönbeck, and Hellman investigated the OER and ORR over various transition metal complexes of hangman porphyrins by DFT calculations.308 The impact of the side groups and the hangman scaﬀold on the reaction landscape were investigated by comparing the Gibbs free energies of the reaction intermediates *OOH, *O, and *OH for metal porphyrin, metal tetraﬂuorophenyl porphyrin, and hangman metal porphyrin. Fe and Co were considered the best metal centers of hangman porphyrin to catalyze OER and ORR. The hangman motif was able to stabilize the *O intermediate, whereas the addition of meso-C6F5 groups can reduce the binding energy of all three intermediates.

Figure 81. Mononuclear and binuclear Mn/Cu porphyrins 176−179 studied as OER catalysts by Sun and co-workers.

both Mn complexes 176 and 178 were able to catalyze the oxidation of water to O2 in DCM/acetonitrile with the addition of an (n-Bu)4N(OH) aqueous solution, whereas both Cu analogues 177 and 179 were inactive. DFT calculations were used to explore the O−O bond formation catalyzed by 176 and 178.303 It was found that the Mn-oxo species had a closed-shell singlet ground state with a d2 MnVO electronic structure, but it became a quintet MnIV-oxyl species in a polar solvent environment. For mononuclear complex 176, two O−O bond formation mechanisms were proposed: the concerted pathway for the formation of MnOOH and the two-step pathway via the coordination of an OH− anion with Mn followed by the reductive elimination to produce a MnOOH species. It was suggested that the concerted pathway was favored based on DFT results. For binuclear complex 178, a third pathway, the direct coupling of the two Mn-oxo groups to form a peroxo intermediate, was found to be energetically improbable. Nocera and co-workers demonstrated that hangman Co corrole 101 (Figure 42) and its β-octaﬂuoro derivative 101-Br8 were also competent OER electrocatalysts.304 When immobilized in Naﬁon ﬁlms and deposited onto FTO electrodes, the TOF of 101-Br8 for the 4H+/4e− water oxidation was 0.81 s−1 in neutral phosphate buﬀer with an applied 780-mV overpotential. Both 101 and 101-Br8 exhibited larger catalytic currents with the earlier onset compared to the nonhangman Co tpfc 72 (Figure 33), and 101-Br8 showed the highest activity in this series. The onset of the catalytic current for 101-

5. PORPHYRIN AND CORROLE GRAFTED MATERIALS 5.1. Materials for ORR Catalysis

Materials with grafted metal porphyrins and corroles have also been widely studied as ORR catalysts. In 2012, Loh and coworkers reported a graphene-porphyrin composite for electrocatalytic O2 reduction.309 As shown in Figure 82, pyridinefunctionalized graphene was used as a template for the assembly of an Fe porphyrin-based metal organic framework (MOF), which was created by the reaction of FeCl3 and tetrakis(4-carboxylphenyl)porphyrin. It was thought that the incorporation into graphene could increase the electroactive surface area and enhance the charge transfer kinetics. The resulting graphene-porphyrin MOF composite 180 was examined as an ORR catalyst in 0.1 M KOH aqueous solution, which showed a well-deﬁned cathodic peak centered at −0.23 V (vs Ag/AgCl). The number of electrons transferred during ORR was always ∼4. The durability of 180 as a competent ORR catalyst was evaluated by CPE at −0.23 V (vs Ag/AgCl) in 0.1 M KOH solution and by methanol crossover experiments, which showed it was more robust during catalysis than Pt and Ni foam catalysts. 3765

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

through Fe-N(imidazole) coordination (Figure 83). ORR studies were performed in O2-saturated 0.1 M HClO4 aqueous

Figure 82. Schematic representation of (a) reduced graphene oxide, (b) pyridine-functionalized graphene, (c) 5,10,15,20-tetrakis(4carboxylphenyl)porphyrin, (d) Fe-Por MOF, (e) graphene modiﬁed with Fe-Por MOF, and (f) a magniﬁed view of layers inside the graphene modiﬁed with Fe-Por MOF showing how graphene sheets intercalated between porphyrin networks. Reprinted from ref 309. Copyright 2012 American Chemical Society.

Figure 83. Schematic representation showing the immobilization of Fe porphyrin 1 on MWCNTs through the axial ligation of the Fe ion.

A heme-functionalized graphene sheet prepared via the noncovalent assembly of heme on nitrogen-doped graphene was reported as an ORR catalyst by Ju and co-workers.310 The immobilization of heme on graphene was demonstrated to occur via axial ligation and π-stacking interactions. The resulting heme-graphene material showed highly eﬃcient electrocatalytic activity in the four-electron reduction of O2 to H2O. Another example using graphene as the support for metal porphyrins to enhance O2 reduction was reported by Tang and co-workers.311 The assembly of Co tetrakis(4-hydroxyphenyl)porphyrin with reduced graphene oxide sheets through the layer-by-layer assembly technique produced composite material 181. The use of reduced graphene oxide was intended to enhance the conductivity in the ORR process, which was performed in 0.1 M KOH aqueous solutions. The CV of 181 showed a well-deﬁned O2 reduction peak centered at −0.22 V (vs Ag/AgCl), which is close to the value of −0.18 V (vs Ag/ AgCl) observed with the Pt/C catalyst. The number of electrons transferred was ca. 3.85 in the −0.30 to −0.50 V (vs Ag/AgCl) range. In addition, catalyst 181 showed great stability in CPE measurements; approximately 80% of the initial catalytic current was maintained after 40000 s, whereas only 50% was maintained for the Pt/C catalyst. In addition, ORR by 181 showed great methanol tolerance. In 2014, Liu and co-workers reported the catalytic ORR features of Fe tetra(2,6-diﬂuorophenyl)porphyrin 1 grafted on multiwalled carbon nanotubes (MWCNTs).312 In previous studies, Liu, Naruta, and co-workers contributed signiﬁcantly to the synthetic modeling of CcOs.313−315 In these studies, the authors showed that the incorporation of a trans axial imidazole ligand on the Fe center is critical to control the O 2 activation.316,317 Therefore, the surface of MWCNTs was ﬁrst functionalized with imidazole groups, and Fe porphyrin 1 was then assembled on the imidazole-functionalized MWCNTs

solutions. The half-wave potential (E1/2) was 0.880 and 0.842 V (vs RHE) for 1-MWCNTs and a commercial Pt/C catalyst, respectively. Of note, the E1/2 for a simple mixture of 1 and MWCNTs was 0.54 V (vs RHE), implying the important role of the trans axial imidazole group. The four-electron reduction of O2 was established for 1-MWCNTs at 0.55−0.70 V (vs RHE). Durability tests at 0.7 V CPE showed that 1-MWCNTs retained approximately 90% of its initial current after 12.5 h, whereas the Pt/C catalyst maintained only approximately 57% of its initial current. In addition, 1-MWCNTs showed an excellent tolerance to the methanol crossover test. MWCNTs covalently grafted with Co porphyrins were prepared by Swager and co-workers and were examined as ORR catalysts.318 The Co porphyrin-MWCNT composite 182 (Figure 84) coated on a carbon electrode showed an O2 reduction peak potential at 0.25 V (vs Ag/AgCl) in 1.0 M H2SO4 solutions. The peak potential for O2 reduction gradually shifted to more positive values, and the catalytic current increased as the thickness of coating was increased up to seven layers. In contrast to MWCNTs, similar composite materials prepared with single-walled carbon nanotubes (SWCNTs) showed very low activity, which was believed to be caused by the covalent modiﬁcation of the sidewalls and thus the reduction of the charge transportation eﬃciency. RDE and RRDE measurements showed that 182 catalyzed a direct fourelectron reduction of O2 to H2O in the pH range from 0.0 to 3.75 at room temperature. In contrast to the study by Swager and co-workers, Jeon and co-workers reported Co porphyrin-SWCNT composites as eﬃcient ORR catalysts.319,320 In addition, Jeon and Kim reported the covalent attachment of Co tetrakis(2aminophenyl)porphyrin onto the surface of three carbon materials, including graphene, SWCNT, and MWCNT, via diazonium salt reactions and compared the ORR activities of 3766

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 84. Schematic representation showing the immobilization of Co porphyrins on MWCNTs through covalent attachments.

Figure 85. Schematic representation showing the MWCNT-templated polymerization of a covalent Co porphyrin 183 network for O2 reduction.

these three composites.321 Electrocatalytic studies revealed that Co porphyrin on graphene showed the highest eﬃciency for the four-electron reduction of O2. Jousselme and co-workers investigated the eﬀects of diﬀerent supporting carbon materials on ORR in either alkaline or acidic solutions using Co octaethylporphyrin.322 Their main conclusions include the following. (1) Carbon-supporting materials played an important role in catalytic ORR. (2) The composite with MWCNTs as the support showed better activity and fourelectron selectivity for O2 reduction. (3) The oxidative chemical treatment of MWCNTs further increased the electrocatalytic performance (i.e., higher current density, smaller overpotential, and larger number of electrons transferred during ORR). Zhang, Cao, and co-workers reported that MWCNTsupported Co tpfc 72 (Figure 33) was an eﬃcient catalyst for O2 reduction in acidic media.323 When 72 was directly adsorbed on GC electrodes, the onset potential of the O2 reduction wave was ca. + 0.42 V (vs Ag/AgCl) with the peak potential at +0.20 V (vs Ag/AgCl) in 0.5 M H2SO4 aqueous solutions. In comparison, these values were improved to +0.60 and +0.42 V (vs Ag/AgCl) for 72-MWCNT-coated GC electrodes. RDE and RRDE measurements revealed that 72MWCNT primarily catalyzed the four-electron reduction of O2. For example, in RRDE, the amount of H2O2 produced was 21.86% for 72-MWCNT, 73.6% for 72, and 81.16% for MWCNTs. In RDE, the number of electrons transferred during ORR was 3.6 for 72-MWCNT, 2.7 for 72, and 2.5 for MWCNTs. The ability of 72-MWCNT to catalyze the reduction of H2O2 was also examined. The result showed that the catalytic H2O2 reduction by 72-MWCNT at pH 0 was negligible; thus, the electrocatalytic reduction of O2 by 72-MWCNT proceeded via a direct 4e− pathway instead of a 2e− plus 2e− pathway. This

result suggests that the spontaneous dimerization or higher aggregation of 72 on the MWCNT surface and synergetic interactions between 72 and MWCNTs are responsible for the observed activity and selectivity. The durability of 72-MWCNT for ORR was examined by CPE at +0.40 V (vs Ag/AgCl). After 20000 s, the current with 72-MWCNT decreased by only 10%, whereas the current with the Pt/C catalyst decreased by 30%. In addition, 72-MWCNT exhibited an unchanged current response after the addition of 3.0 M methanol, showing a much better tolerance to the methanol crossover test relative to the Pt/C catalyst. In 2014, Chlistunof and Sansiñ ena reported diﬀerent methods of Fe meso-tetraphenylporphyrin immobilization on carbon supports.324 Two approaches were used. In the ﬁrst approach, reduced graphene oxide was modiﬁed with 3aminopropyl imidazole groups and was then used as the support (denoted as G-IMD). In the second approach, the carbon material Vulcan XC72 mixed with polyvinyl imidazole (PVI) was used as the support (denoted as Vulcan-PVI). ORR catalysis is signiﬁcantly enhanced when Vulcan-PVI and G-IMD are used as catalyst supports. The half-wave ORR potentials are shifted in the positive direction by approximately 200 mV compared with corresponding carbon-alone materials to support the catalyst, and the number of electrons involved in ORR increases from two to four. This eﬀect is thought to be a consequence of not only increased electron density at the Fe center but also the speciﬁc complex geometry upon axial coordination with imidazole ligands. Campidelli and co-workers reported the MWCNT-templated polymerization of covalent Co porphyrin networks for O2 reduction.325 As shown in Figure 85, the adsorption of Co meso-tetraethynylporphyrin 183 on the sidewalls of MWCNTs and the subsequent dimerization of triple bonds via Haycoupling gave composite material 183-MWCNT, in which 3767

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

pH 5.5 aqueous solutions, and the number of electrons transferred was 3.99. Fuerte and co-workers reported the immobilization of metalloporphyrins on robust inorganic solids such as silica gel, purely siliceous MCM-41, and delaminated zeolites ITQ-2 and ITQ-6.328 Two diﬀerent methods were used for immobilization. Electrochemical data showed that O2 was reduced primarily through a two-electron process for both Co and Fe porphyrins. Nagai and co-workers reported the immobilization of metalloporphyrins on a perovskite-type oxide-carbon (Vulcan XC72) mixture (La0.6Sr0.4Mn0.6Fe0.4O3/ C) for electrocatalytic O2 reduction.329 Polymerization is another frequently used method for the immobilization of metalloporphyrins on electrodes. Instead of using porous carbon materials and metal oxides as the supports, an alternative approach is to incorporate metalloporphyrins into conducting polymers such as polyaniline, polypyrrole, and polythiophene. Li and co-workers reported the synthesis of a Co porphyrin-polypyrrole composite and its electrocatalysis for O2 reduction.330 CoII tetrakis(4-sulfonatophenyl)porphyrin 62 (Figure 32) was chosen because it forms self-assembled molecular J-type aggregates (known as J-aggregates), which could act as templates for the electropolymerization of polypyrrole. Linear J-aggregates of 62 were believed to form in a neutral aqueous medium based on atomic force microscopy (AFM); the polar groups (negative sulfonic groups) of one porphyrin molecule should have strong intermolecular interactions with the positive central Co ions of neighboring porphyrin molecules. Because 62 is soluble in water, it could be incorporated into polypyrrole by electrochemical copolymerization, and J-aggregates of 62 could act as a template to form a 62-polypyrrole composite. Electrochemical data of this 62polypyrrole composite on Au electrodes showed ORR catalysis with a positively shifted O2 reduction wave and largely increased peak current compared with the bare or polypyrrole-modiﬁed Au electrode in an O2-saturated phosphate buﬀer (pH 7.4). It was suggested that the role of polypyrrole in this composite was to host Co porphyrin catalysts and provide suﬃcient conductivity. The number of electrons transferred during ORR was 3.84, indicating that the O2 reduction is mainly conducted through the four-electron pathway. In a subsequent work, Li and co-workers reported the synthesis of Co porphyrin-polyaniline composite and its electrocatalysis for O2 reduction.331 Again, Co porphyrin 62 was used because of its ability to form J-aggregates and function as the template for the polyaniline electropolymerization. The resultng 62-polyaniline composite is more robust than the 62polypyrrole composite. The peak potential of the ORR catalytic

MWCNT acted as a template for the formation of polymeric layers. The catalytic ORR activity of the 183-MWCNT composite adsorbed on GC electrodes was examined in 0.5 M H2SO4 solution. Polarization curves of 183-MWCNT showed a reduction current starting at 0.55 V (vs Ag/AgCl). Importantly, the catalytic activity of 183-MWCNT is approximately 3.6 times higher than that of physiadsorbed hybrid material obtained by simply mixing 183 and MWCNTs. The number of electrons transferred during O2 reduction was 3.82 for 183-MWCNT and 3.35 for physiadsorbed 183 on MWCNTs as measured at −0.05 V (vs Ag/AgCl). The enhanced performance of 183-MWCNT is thought to be a result of the multilayers of Co porphyrins around MWCNTs, providing a conﬁguration similar to the cofacial “face-to-face” bis-porphyrin systems. In addition to carbon materials, other supports have also been used. For example, Mauzeroll and co-workers reported self-assembled thiol-porphyrin monolayers on Au or GC for electrocatalytic O2 reduction.326 Co tetraphenylporphyrin was immobilized on an Au or GC electrode with a self-assembled monolayer of 4-aminothiophenol through Co-N coordination interactions (Figure 86). The catalytic activity of chemisorbed

Figure 86. Schematic representation showing the immobilization of Co porphyrin on Au/GC electrode through the axial ligation of the Co ion.

Co porphyrins on Au or GC electrodes was examined in 0.5 M H2SO4 solutions. The peak potential for catalytic ORR was ca. 150 mV (vs Ag/AgCl), and the two-electron reduction of O2 to H2O2 was established as the main pathway. Gushikem and co-workers reported electrocatalytic O2 reduction by CoII tetrakis(1-methyl-pyridinium-4-yl)porphyrin immobilized on binary oxide SiO2/Sb2O3.327 The peak potential of catalytic O2 reduction was −0.25 V (vs SCE) in

Figure 87. Schematic representation showing the electrochemical polymerization and the growth of a polymerized conducting ﬁlm of 184 on GC electrodes. 3768

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Table 11. Summary for Porphyrin and Corrole Grafted Materials for ORR Catalysis catalysts Fe-Por/graphene Fe-Por/MWCNT Fe-Por/N-graphene Fe-Phth/MWCNT Co-Por/rGO Co-Por/MWCNT Co-Por/SWCNT Co-Por/graphene Co-Por/SWCNT Co-Por/MWCNT Co-Cor/MWCNT Co-Por/Au Co-Por/SiO2−Sb2O3 Co-Por/MCM-41 Fe-Por/ITQ-6 Co-Por/perovskite-C Co-Por/polypyrrole Co-Por/polyaniline polymerized Co-Por adsorbed Co-Por stacked Co-Por Co-Por/MWCNT Co-Por/POM a

Eonset

Ehalf‑wave

Epeak

n

ref

0 Va 1.0 Vb −0.20 Va −0.05 Vc −0.12 Va 0.44 Va 0.39 Va 0.46 Va 0.46 Va 0.40 Va 0.53 Va 0.30 Va −0.05 Vc 0.42 Va 0.6 Va 0.88 Vb 0.31 Vc 0.32 Vc −0.12 Va −0.15 Vc 0.06 Va 0.55 Va 0.05 Va

−0.12 Va 0.88 Vb −0.31 Va −0.09 Vc −0.18 Va 0.31 Va 0.26 Va 0.32 Va 0.32 Va 0.30 Va 0.48 Va 0.22 Va −0.17 Vc 0.28 Va 0.44 Va 0.79 Vb 0.13 Vc 0.24 Vc −0.18 Va −0.20 Vc −0.06 Va 0.43 Va −0.10 Va

−0.23 Va 0.75 Vb −0.38 Va −0.12 Vc −0.22 Va 0.25 Va 0.17 Va 0.14 Va 0.14 Va 0.10 Va 0.41 Va 0.13 Va −0.25 Vc 0.20 Va 0.25 Va 0.72 Vb −0.12 Vc 0.17 Vc −0.22 Va −0.25 Vc −0.15 Va 0.25 Va −0.20 Va

3.82 4 −d 3.81 3.85 3.7 3.5 3.9 3.5 2.8 3.6 −d 3.99 2.1 2.2 2.0 3.84 3.7 3.82 2 4 3.82 2.2

309 312 310 322 311 318 319 321

media 0.1 0.1 0.1 0.1 0.1 1.0 0.1 0.5

M M M M M M M M

KOH HClO4 pH 8.0 phosphate NaOH KOH H2SO4 H2SO4 H2SO4

0.5 0.5 1.0 0.5

M M M M

H2SO4 H2SO4 pH 5.5 KCl H2SO4

0.1 M KOH 0.01 M pH 7.4 phosphate 1.0 M HCl 0.01 M pH 7.0 phosphate 1.0 M NaOH 0.1 M NaOH 0.5 M H2SO4 0.1 M pH 3.78 acetate

323 326 327 328 329 330 331 332 334 335 325 339

Potentials are reported vs Ag/AgCl. bPotentials are reported vs RHE. cPotentials are reported vs SCE. dData not reported.

polymerized ﬁlm on aerogel carbons, showing the reduction of CoIII to CoII in inner layers of the ﬁlm due to the closer vicinity to the aerogel carbon surface. It is therefore suggested that surface quinones serve as the electron-transfer agent to reduce CoIII porphyrin to the CoII state, which in turn reduces O2. Zhang, Jung, and co-workers reported the ORR catalysis of CoII meso-tetrakis(4-methoxyphenyl)porphyrin adsorbed on graphite electrodes.334 The measured value of Co porphyrins on the electrode surface was 3.8 × 10−11 mol cm−2, and the calculated monolayer gave a surface concentration of 3.3 × 10−11 mol cm−2. This result suggests that molecules have a ﬂat orientation on electrodes. In addition, the eﬀect of anion adsorption on electrocatalytic ORR was investigated. The overpotential for O2 reduction has the sequence Br− > Cl− > SO42− > F−, which is consistent with the order of adsorption capacity for these anions on electrodes. Therefore, the potential shift may be attributed to the adsorption competition between O2 molecules and anions on the catalytic active sites. Ramı ́rez and Canales investigated the interactions between Co octaethylporphyrin and GC electrodes or oxidized GC electrodes.335 Oxidized GC electrodes with adsorbed Co porphyrins showed improved electrocatalytic O2 reduction activity compared with oxidized GC electrodes alone or regular GC electrodes with adsorbed Co porphyrins. AFM studies revealed morphological changes on the GC surface, implying that oxidized groups on the GC surface would bind to Co ions at the axial position and thus give a diﬀerent orientation of Co porphyrins on the electrode surface. In addition to these carbon materials, the interactions of metal porphyrins, in particular, Mn porphyrins, on metal surfaces have been investigated.336−338 Key ﬁndings include the following. (1) Diﬀerent oxidation states of individual Mn porphyrins during ORR are detected using scanning tunneling microscopy at solid/liquid interfaces. (2) The adsorption of MnIII porphyrins on Ag(111) causes the reduction of MnIII to

wave was at 0.16 V (vs SCE), and the number of electrons transferred was 3.7. This result indicates that this composite catalyzes the eﬃcient four-electron reduction of O2 to H2O regardless of the “direct” or “indirect” pathway. Another electropolymerized Co porphyrin composite used as an ORR catalyst was reported by Lei and co-workers.332 Repetitive CV scans of Co porphyrin 184 (Figure 87) in DCM showed the gradual increase of currents over 1.25 V (vs Ag/ AgCl), indicating an oxidation process of thienyl groups and the electrochemical polymerization and growth of polymerized conducting ﬁlms of 184 on GC electrodes. The CV of polymerized 184 ﬁlm displayed three redox events ca. −0.35 (CoII/CoI), 0.31 (CoIII/CoII), and 0.80 V (Por•+/Por) (vs Ag/ AgCl) in a pH 7.0 phosphate buﬀer solution. In the presence of O2, a large irreversible cathodic peak appeared at −0.2 V (vs Ag/AgCl), indicating the reduction of O2. The numbers of electrons transferred during ORR were 3.9 in pH 2.0, 3.82 in pH 7.0, and 4.03 in pH 13 buﬀers. The formation of multiple layers containing suitable Co···Co separation for the O2 binding is thought to be responsible for the facile four-electron process. As carbon-based materials/electrodes have been widely used in the examination of metal porphyrin catalysts for ORR, the interactions between metal porphyrins and carbon-based surfaces have been investigated. Bettelheim and co-workers reported evidence for the formation of Co porphyrin-quinone interactions with adsorbed resorcinol on GC electrodes and with surface quinone functionalities on aerogel carbons.333 Reﬂection UV−visible spectrum of an electropolymerized ﬁlm of Co tetrakis(2-aminophenyl)porphyrin on a GC electrode showed Soret and Q bands at 431 and 544 nm, respectively. Adsorption of resorcinol onto this ﬁlm caused an 8 nm red shift for the Soret band, indicating the interaction of Co porphyrin and resorcinol. Such an interaction was further suggested based on XPS measurements. Similar Co porphyrin-quinone interactions were also observed by XPS for the electro3769

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

MnII. (3) The MnII center is able to bind O2 and form O axial ligands, which is accompanied by the oxidation state change from MnII to MnIII. (4) The activation energies for the Cl and O removal were derived from the XPS data and were found to be 0.35 ± 0.02 and 0.26 ± 0.03 eV, respectively. The catalytic properties and conditions reported in these examples are summarized in Table 11. Because diﬀerent experimental conditions are used in these studies, a direct comparison of these catalysts is challenging. Therefore, data shown in this table are from the original published papers.

When coated on ITO electrodes, the CVs of 185-poly(terthiophene) displayed substantially larger currents at 0.68 V (vs Ag/AgCl) with illumination than without illumination in a 0.1 M pH 7.0 Na2SO4 aqueous solution. Because the thermodynamic equilibrium potential for water oxidation at pH 7.0 is 0.62 V (vs Ag/AgCl), the overpotential for the onset of water oxidation catalysis is approximately 430 nm. For comparison, FTO electrodes covered with BiVO4 surface-modiﬁed with Cu(OAc)2 or meso-tetra(4-carboxyphenyl)porphyrin were also examined. The results showed that BiVO4 modiﬁed with Cu porphyrin outperformed BiVO4 modiﬁed with Cu(OAc)2 or free base porphyrin and attained a photocurrent up to 4.5 ±

5.2. Materials for HER Catalysis

Shen, Wang, and co-workers reported the preparation and HER features of hybrid ﬁlms of CoII tetrakis(1-methyl-pyridinium-4yl)porphyrin 61 (Figure 32) and reduced graphene oxide.340 Functional multilayer ﬁlms were prepared by the alternating layer-by-layer assembly of negatively charged graphene oxide and positively charged Co porphyrin 61 in combination with an electrochemical reduction procedure. The electrocatalytic HER activity of 61-graphene thin ﬁlms was evaluated in 0.1 M KOH solution under N2, which revealed a pronounced catalytic current with the onset potential at −0.4 V (vs RHE). Catalytic activity increased with the bilayer number up to 7. For example, the current density increased from 0.222 to 1.729 mA cm−2 at a potential of −0.5 V (vs RHE) as the bilayer number increased from 1 to 7, and the overpotential required to achieve 1 mA cm−2 current density decreased from 572 to 474 mV. A further increase of the bilayer number resulted in minor changes in the activity. EIS measurements revealed that the ohmic resistance decreased upon increasing the bilayer number up to 7 and then increased with additional bilayer numbers. In addition to alkaline solutions, this 61-graphene thin ﬁlm showed similar activity in an N2-saturated 0.5 M H2SO4 solution. Control experiments conﬁrmed the crucial role of Co porphyrin 61 for HER, and the graphene material was believed to provide a high conductive support. Dong and co-workers reported the fabrication, and catalytic features of an organic−inorganic hybrid ﬁlm consisted of positively charged Co porphyrin 61 and negatively charged polyoxometalate anion [SiW12O40]4−.339 Co porphyrin 61 and H4[SiW12O40] were alternately deposited through a layer-bylayer method onto a GC electrode covered with 4-aminobenzoic acid. In 0.1 M H2SO4 solution, the CV of this 61[SiW12O40] hybrid material with [SiW12O40] as the outermost layer displayed remarkable catalytic activity for HER, and the current for the HER increased drastically with the number of layers up to nine. Importantly, this material showed the same HER activity after being kept in air or in water for more than 3 months. In contrast, the CV of 61-[SiW12O40] with 61 as the outermost layer in an air-saturated 0.1 M H2SO4 solution showed a marked catalytic O2 reduction current that increased with the number of layers. The peak potential of O2 reduction was at −0.18 V (vs Ag/AgCl). The number of electrons transferred was 2.3. 5.3. Materials for OER Catalysis

In 2012, Swiegers and co-workers reported the immobilization of MnIII tetra(4-sulfonatophenyl)porphyrin chloride 185 (not shown) in conducting polymer poly(terthiophene) and the photocatalytic water oxidation feature of the resulting composite material.341 This material was fabricated by the electrochemical polymerization of terthiophene in ethanol/ DCM containing MnIII porphyrin 185 to incorporate 185 as anionic counterions into the generated poly(terthiophene) ﬁlm. 3770

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 88. Structure of [Fe2S2] complexes 186−189 and Zn porphyrins 190 and 191. Self-assembled biomimetic porphyrin-[Fe2S2] photocatalyst [190-188-191] for HER.

1.5 μA cm−2. However, this composite decomposed, which gradually decreased activity during prolonged irradiation. Nakanishi, Hashimoto, and co-workers reported a series of transition metal (Mn, Fe, Co, Cu, and Zn) complexes of tetraphenylporphyrin as eﬃcient electrocatalysts for aprotic OER on Li2O2, which is a crucial reaction for the charging of Li−air batteries.345 Co tetraphenylporphyrin showed the best catalytic eﬀect among the examined catalysts. The reduction potential of the formal MIII/MII redox couple increased in the following order: Co < Zn < Mn < Cu < Fe; this order is consistent with the order observed for OER activities. It is worth noting that for Zn porphyrin, the redox is ligandcentered. This result therefore suggests that the central metal ion plays an essential role in catalytic OER processes. It is thought that the electrochemically oxidized metal porphyrin catalyst reacts with Li2O2 to evolve O2, which is associated with Li2O2 decomposition and the reduction of metal porphyrins.

porphyrins 190 and 191 (Figure 88) were used for the preparation of several supramolecular assemblies. Crystallographic studies of 186−189 revealed that the phosphine ligand was located at the apical position trans to the Fe−Fe bond and had a very small eﬀect on Fe−Fe distances. The CVs of 186 and 187 displayed typical irreversible reduction waves at −2.10 and −2.05 V (vs ferrocene), respectively, corresponding to a FeIFeI/Fe0FeI redox event, and these waves became pronounced catalytic waves in the presence of acetic acid, indicating that (1) both 186 and 187 are eﬃcient electrocatalysts for proton reduction, and (2) phosphine ligands do not signiﬁcantly alter the redox behavior and electrocatalytic properties. The formation of self-assembled supramolecular porphyrin-[Fe2S2] species was conﬁrmed. The structure of supramolecular complex [186−190] was further established. The distance between Zn and the diiron core was 7.2 Å, which was well within the range for an electron-hopping process. Steady-state ﬂuorescence measurements on complexes [186−190] and 190 showed signiﬁcant quenching of the singlet excited state of Zn porphyrin in the assembly; the ﬂuorescence lifetime of Zn porphyrin was 1.96 ns for free 190 and 0.3 ns for the assembly [186−190]. This quenching is attributed to the electron transfer from the excited Zn porphyrin sensitizer to the [Fe2S2] core. The photocatalytic H2 evolution of catalysts 186−189 and their assemblies with photosensitizers 190 and 191 were then evaluated in toluene in the presence of acetic acid as the proton source and NiPr2Et as the sacriﬁcial electron donor. The results showed that the assemblies containing a single chromophore did not have any photocatalytic HER activity, but catalyst 188 with both 190 and 191 generated signiﬁcant amounts of H2, which indicated the requirement of two diﬀerent chromophores for active assemblies. Interestingly, catalyst 186, which only has one pyridyl functional group, was also active for photocatalytic HER with 190 and 191. However, the active species formed in the case of 186 was identical to that in the case of 188 based on FTIR studies. It was thus suggested that during photoirradiation, complex 186 disproportionates to 188 in the

6. OTHER APPLICATIONS OF PORPHYRINS AND CORROLES IN SMALL MOLECULE ACTIVATION REACTIONS 6.1. Porphyrin- and Corrole-Based Chromophores in Catalysis

Porphyrin derivatives are involved in photosynthesis for harvesting light energy.84,85,87,346 In biomimetic approaches, synthetic porphyrin complexes have been extensively studied in light-harvesting processes from the viewpoint of photoinduced electron transfer and energy transfer at both molecular and supramolecular levels.347−350 In this section, we will brieﬂy introduce recent advances in the use of porphyrin- and corrolebased systems for photochemical HER and OER. 6.1.1. Porphyrin-Diiron Systems for HER. Synthetic diiron complexes have been widely studied as HER catalysts, mimicking the diiron active sites in [FeFe]-hydrogenases.351−353 In 2009, Reek and co-workers reported selfassembled biomimetic porphyrin−[Fe2S2] photocatalysts for the HER.354 Various diiron complexes 186−189 and Zn 3771

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

was 0.16 based on the amount of [Fe2S2] and 16 based on Zn porphyrin. The noncovalent assembly strategy can facilitate the fast intramolecular electron transfer from the singlet excited state of photosensitizers to the catalytic sites and can also reduce the charge recombination via complex dissociation. In addition to self-assembled porphyrin-[Fe2S2] photocatalysts, Wasielewski and co-workers investigated photoinitiated multistep charge separation in Fc-porphyrin-[Fe2S2] triads with covalent linkages.356 Photochemical HER results show that a long-lived charge-separated state is crucial for eﬃcient photochemical H2 generation. Feng and co-workers incorporated a biomimetic [Fe2S2] moiety into a Zr-[Zn-Por]-based MOF (Figure 90) and veriﬁed its photocatalytic HER properties.357 Fluorescence emission quenching of photoexcited [Zn-Por] was observed in the presence of [Fe2S2]. The reduction potential of 1*[Zn-Por] is −1.45 V (vs SCE). The ﬁrst and second reduction potentials of [Fe2S2] are approximately −1.20 and −1.80 V (vs SCE), respectively. Therefore, the direct electron transfer from 1*[ZnPor] to [Fe2S2] to produce [FeIFe0S2] is thermodynamically feasible, but further reduction to [Fe0Fe0S2] by 1*[Zn-Por] is thermodynamically unfeasible. Consequently, [FeIFe0S 2] should be protonated ﬁrst for subsequent reduction. Photocatalytic HER was conducted in a 1.0 M pH 5 acetate buﬀer solution under visible-light irradiation. Compared to homogeneous photocatalysis using [Fe2S2], heterogeneous photocatalysis with MOF-incorporated [Fe2S2] showed higher eﬃciency in terms of rates and total H2 yields, which was due to the enhanced stability of [Fe2S2] inside MOFs. Other porphyrin diruthenium-based MOF materials were reported by Mori and co-workers as photocatalysts for H2 evolution.358 Because Ru and Fe are in the same group and share similar chemical properties, these porphyrin dirutheniumbased MOF complexes are discussed in this subsection. Photocatalytic studies showed relatively low activity. These authors proposed that the low catalytic activity was due to the low eﬃciency of the intramolecular electron transfer from porphyrins to Ru2 units. Importantly, the addition of methylviologen (MV2+) could increase electron transfer eﬃciency and thus the photocatalytic activity. These data suggested that MV2+ acted as an intermolecular electron mediator between porphyrins and Ru2 sites.

presence of 190 and 191. These results suggest that the assembly of [190-188-191] (Figure 88) is the active species for H2 evolution, which is consistent with recent theoretical studies showing that the asymmetry of the diiron center may be a desirable feature. Sun and co-workers reported photoinduced HER with a noncovalent assembly 192 (Figure 89) consisting of Zn

Figure 89. Self-assembled biomimetic Zn porphyrin-[Fe2S2] photocatalyst 192 for HER.

tetraphenylporphyrin and an [Fe2S2] complex.355 The ﬂuorescence of Zn porphyrin was quenched in the presence of the [Fe2S2] complex. In accordance with the reduction potential of the excited Zn porphyrin (−1.45 V vs SCE) and the ﬁrst reduction potential of [Fe2S2] (ca. −1.20 V vs SCE), the direct electron transfer from the singlet excited state of Zn porphyrin to [Fe2S2] is thermodynamically feasible, giving Zn-Por•+ and Fe0FeI species. This photoinduced electron transfer process in the assembled complex 192 was further conﬁrmed by transient absorption measurements. Photocatalytic studies showed that H2 evolution occurred with a DCM solution containing assembly 192, 2-mercaptobenzoic acid, and TFA under visible light irradiation. The TON

Figure 90. (a) Structural building unit of Zr6O8(CO2)8(H2O)8 and (b) structural building unit of Zn tetrakis(4-carboxyphenyl)porphyrin. (c) Model structure of Zr-[Zn-Por]-based MOF. (d) Complex [Fe2S2]. (e) Model structure of the complex. Color scheme: Zr, green; Zn, dark gray; C, light gray; O, red; N, blue; Fe, light green; S, yellow. Reprinted with permission from ref 357. Copyright 2014 Royal Society of Chemistry. 3772

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

molecule, and the rapid dissociation of TEA•+ from the porphyrin unit is crucial to prevent the nonproductive back electron transfer. Further photoinduced reduction of CoII to CoI was detected by time-dependent UV−visible spectroscopy; CoI acts as the catalytic active species for proton reduction to H2. However, similar work by Pryce and co-workers showed that the three-component system of Zn porphyrin, cobaloxime, and triethylamine was not active in photocatalytic H 2 evolution.361 The lack of photocatalytic activity is likely due to the cleavage of the Co-N(pyridyl) bond between porphyrin and cobaloxime, based on excitation measurements and DFT calculations.362 Guldi and co-workers also investigated the photoinduced charge transfer process in porphyrin-cobaloxime and corrolecobaloxime dyads (Figure 91).363 Their main conclusions are as follows. (1) The selective photoexcitation of porphyrin or corrole to its ﬁrst singlet excited state leads to electron transfer from the chromophore to the Co center of cobaloxime. (2) The charge recombination process is faster than this charge separation process, which is detrimental for the accumulation of reduced Co species for H2 production. (3) The addition of a sacriﬁcial electron donor slows down the charge recombination process. 6.1.3. Porphyrin-Pt Nanoparticle Systems for HER. Pt nanoparticles are known to catalyze H2 evolution with very high eﬃciency, and the combination of Pt nanoparticles with chromophore molecules is potentially useful for photoinduced H2 evolution due to the eﬃcient energy and electron transfer between photosensitizers and Pt nanoparticles. Wang and coworkers reported photocatalytic H2 evolution using porphyrinfunctionalized Pt nanocomposites.364 Two donor-bridgeacceptor conjugates 193 and 194 (Figure 92) were synthesized,

6.1.2. Porphyrin-Monocobalt Systems for HER. Cobaloximes are another family of extensively investigated HER catalysts, and they are combined with metal porphyrins to construct photocatalysts for H2 evolution. Natali and coworkers designed a self-assembled system containing ascorbic acid as the sacriﬁcial donor, Al porphyrin as the photosensitizer, and cobaloxime as the catalyst (Figure 91) and examined its

Figure 91. Self-assembled HER photocatalysts containing cobaloxime as the catalyst and porphyrin or corrole as the photosensitizer.

photocatalytic HER features.359 Photocatalytic experiments showed that the quantum yield for H2 evolution is 0.046, the TONs are 352 relative to Al porphyrin and 117 relative to cobaloxime, and the corresponding TOF values are 10.8 and 3.6 min−1, respectively. However, contrary to the initial design, the association of cobaloxime and ascorbic acid with Al porphyrin does not produce photocatalytic H2 evolution, likely due to the rapid reversible electron transfer between Al porphyrin and the sacriﬁcial donor or catalyst. Sun and co-workers reported photochemical HER using cobaloxime and various porphyrin photosensitizers (Figure 91).360 The interaction between triethylamine and Zn porphyrin was demonstrated through axial coordination of triethylamine to the Zn ion, but negligible interaction was shown between triethylamine and Mg porphyrin or metal-free porphyrin. On the basis of thermodynamic analysis, the electron transfer from the singlet excited state of porphyrin to CoIII and CoII centers is possible for Zn and Mg porphyrin, whereas this electron transfer process is thermodynamically unfavorable for metal-free porphyrin. Photoinduced H2 evolution showed that the system using Zn porphyrin displayed a TON of 22, and as expected, a small amount of H2 was produced in the other two systems. This result suggests that the greater H2-evolving eﬃciency of the Zn porphyrin system can be attributed to the precoordination of triethylamine to the Zn ion, which facilitates inner-sphere electron transfer from triethylamine to the porphyrin unit. On the basis of these results, a reaction mechanism was proposed. Upon light irradiation, intramolecular electron transfer occurs from excited Zn porphyrin to CoIII ion. The resulting [ZnPor•+]+ is then reduced by a precoordinated triethylamine

Figure 92. Porphyrins 193−195 used as the photosensitizer for photocatalytic HER with metallic Pt as the catalyst.

in which triphenylamine and porphyrin moieties were expected to act as energy donors and acceptors, respectively. Timeresolved emission and decay studies revealed that the ethylene linkage between triphenylamine and porphyrin in 194 is more suitable than the ester linkage in 193 for the intramolecular energy transfer because the intramolecular energy transfer rate and quantum yield of 194 are approximately 9 and 1.8 times higher than those of 193. In combination with Pt nanoparticles, the ﬂuorescence emission of porphyrins was quenched by over 90% at an excitation of 420 nm, indicating the binding of porphyrins on Pt nanoparticles and heterogeneous charge transfer at the porphyrin−Pt interface. Electron-transfer rate 3773

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

constants were 7.88 × 107 and 1.68 × 109 s−1 for 193-Pt and 194-Pt, respectively. This result suggests that the photoinduced electron transfer from photoexcited 194 to Pt nanoparticles is more eﬃcient. These porphyrin-Pt nanocomposites were then examined as photocatalysts for the HER in aqueous solutions with ethylenediaminetetraacetate (EDTA) as the sacriﬁcial reductant. In the case of 194, the TON numbers of H2 production are 65.9 (based on Pt) and 6590 (based on 194), and the quantum yield is 2.76%. In the case of 193, the TON numbers are 40.7 (based on Pt) and 4070 (based on 193), and the quantum yield is 1.71%. These results also show that the ethylene linkage is more suitable than the ester linkage for photocatalytic HER. In addition, these porphyrin-Pt nanocomposites functionalized with donor-bridge-acceptor conjugates are more active than analogous simple porphyrin-Pt nanocomposites. In a subsequent work, Wang and co-workers studied photocatalytic H2 production by Pt nanoparticles functionalized with donor−acceptor porphyrin 195 (Figure 92).365 Fluorescence and photoelectrochemical studies indicated that the photoinduced energy and electron transfer occurs from photoexcited anthracene to porphyrin and then to Pt. For 195, the energy transfer rate of 3.91 × 108 s−1 and quantum yield of 90% demonstrate eﬃcient intramolecular energy transfer from anthracene to porphyrin. The emission of 195 excited at 420 nm was quenched more than 95% in 195-Pt composite, producing an electron transfer rate constant of 5.92 × 108 s−1 from porphyrin to Pt. Photocatalytic studies were performed, and the total amount of H2 evolved from 195-Pt was 175.3 μmol, which was much higher than that from pristine porphyrin-Pt (55.6 μmol). A reaction mechanism was proposed: upon light irradiation, anthracene is excited, and the energy is transferred to porphyrin. Subsequent electron transfer occurs from photoexcited anthryl-porphyrin to the metallic Pt particle, where protons are catalytically reduced to produce H2. In addition to triphenylamine and anthracene, fullerene was used to make donor−acceptor conjugates with porphyrins.366,367 Song and co-workers synthesized several porphyrinfullerene dyads (Figure 93) and examined the noncovalently associated dyad as a photosensitizer for H2 evolution with metallic Pt catalyst.367 In the X-ray structure, the Zn atom of tetraphenylporphyrin was axially coordinated to the pyridine N atom of fullerene with a Zn-N(pyridine) bond length of 2.137 Å. The distance from porphyrin to the center of C60 is 12.225 Å. Strong intramolecular electron transfer from photoexcited Zn porphyrin to axially coordinated C60 moiety was suggested. With the ﬁve-component system consisting of electron donor EDTA, photosensitizer dyad, electron mediator methyl viologen, catalyst Pt, and proton source acetic acid, H2 evolution was observed in an aqueous solution under irradiation >400 nm. The highest TON was 73 in 4 h. Importantly, the control experiment using Zn porphyrin instead of the dyad as the photosensitizer produced a TON of 2.3 in 4 h, showing that the C60 moiety plays a crucial role in photoinduced H2 production. A reaction mechanism was proposed, suggesting that a charge-separated [Zn-Por]+-C60− state was involved in the catalytic cycle. Hayashi and co-workers investigated photocatalytic H2 generation by Zn porphyrin bearing anionic patches (Figure 94).368 Two other points are worth noting: they used a reconstituted myoglobin to stabilize porphyrin units, and they used monomethylated bipyridinium instead of methyl viologen

Figure 93. Fullerene-porphyrin dyads investigated as donor−acceptor conjugates for photocatalytic HER with metallic Pt as the catalyst.

Figure 94. Zn porphyrin containing anionic patches.

as the electron mediator. Anionic patches introduced to porphyrin can increase the interaction with cationic electron mediators and facilitate photoinduced electron transfer. The protein matrix of myoglobin can prevent the self-aggregation of porphyrins. Eﬃcient photocatalytic H2 generation was observed with the system of porphyrin-myoglobin, colloid Pt, monomethylated bipyridinium, and triethanolamine. Interestingly, a negligible amount of H2 was produced when methyl viologen was employed as the electron mediator. This result suggested that upon accepting an electron from porphyrin, monomethylated bipyridinium becomes a neutral species, and this reduced form is expected to readily escape from the anionic domain. Consequently, monomethylated bipyridinium can prevent the back electron transfer to the photosensitizer and lead to a smooth electron transfer to the HER catalyst. The pHdependent studies revealed that the maximum eﬃciency occurred at pH 6.0, implying that the protonation of the porphyrin anionic domain changes its binding with the electron mediator. Shelnutt and co-workers reported self-assembled porphyrin composites consisting of positively charged Sn tetra(13774

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 95. Schematic illustration of the procedure of organizing Zn porphyrin-Pt-TiO2 composite 198. The electron micrograph images show (A) Pt/TiO2 (TEM), (B) Zn porphyrin-based nanocylinder (SEM), and (C) Pt/TiO2-Zn porphyrin nanorods (SEM). Reprinted from ref 375. Copyright 2013 American Chemical Society.

hydroxyethyl-pyridinium-4-yl)porphyrin and negatively charged Zn tetra(4-sulfonatophenyl)porphyrin and their uses as photosensitizers for H2 evolution with metallic Pt.369 The photocatalytic H2 generation of self-assembled cloverlike porphyrinPt composites was performed in an aqueous solution containing triethanolamine and methyl viologen. Importantly, porphyrin clovers are generally more active and more durable than the individual porphyrin for photocatalytic H2 evolution, which suggests that the cooperative behavior of binary solids may be crucial for enhanced performance. Yang and co-workers reported photocatalytic H2 evolution using a conjugate of tetrakis(4-hydroxylphenyl)porphyrin and pyrene as the photosensitizer, metallic Pt as the catalyst, and EDTA as the sacriﬁcial reductant.370 Their data suggested that upon photoexcitation, energy transfer occurred from pyrene to porphyrin and that electron transfer occurred subsequently from the excited porphyrin to Pt, where catalytic H2 generation occurred. In a 12 h irradiation, TON numbers based on Pt and porphyrin were 63 and 6311, respectively, and the quantum yield for H2 production was 2.65%. The control experiment using porphyrin-Pt composite (without pyrene) generated much less H2 under identical conditions. Kondo and co-workers compared the photocatalytic H2 generation features of Zn pyropheophorbide a and its dimer as the photosensitizer.371 In this work, colloidal Pt was used as the catalyst with methyl viologen as the electron mediator and NADPH as the sacriﬁcial reductant in pH 7.0 phosphate buﬀer. After 2 h of irradiation, the amount of H2 generated was approximately 0.3 and 0.2 μmol by Zn pyropheophorbide and its dimer, respectively, corresponding to TONs of 20 and 13. This result indicated that the photocatalytic HER activity of the dimer was less than that of the monomer, which is likely due to the intramolecular aggregation. Choi and co-workers reported Sn-porphyrin-sensitized TiO2/Pt for H2 production under visible light irradiation.372

Dye-sensitized semiconductor oxides have been widely studied for light-harvesting processes, and it is generally accepted that eﬃcient visible light sensitization requires strong binding of sensitizer molecules on semiconductor surfaces to maximize the electronic coupling between excited sensitizer orbitals and semiconductor conduction bands. In this work, water-soluble Sn tetrakis(4-pyridyl)porphyrin was studied as the visible light sensitizer of TiO2/Pt nanoparticles without the adsorption of Sn porphyrin on the TiO2 surface, and photocatalytic H2 evolution was accomplished in an aqueous solution using EDTA as the sacriﬁcial electron donor under irradiation >420 nm. Control experiments used two Ru bipyridyl complexes, [RuII(bpy)3]2+ and [RuII(dcbpy)3] (with three pairs of dicarboxylate groups on bpy). As expected, in the case of [RuII(bpy)3]2+, which does not adsorb on the TiO2 surface, a negligible amount of H2 was produced, whereas in the case of [RuII(dcbpy)3], which can be anchored on TiO2 through carboxylate groups, marked H2 production was observed. However, Sn porphyrin has a very weak adsorption on TiO2, but it is as active as [RuII(dcbpy)3] for photocatalytic H2 evolution, giving a TON of 410 in 9 h of irradiation and an apparent photonic eﬃciency of 35% with monochromatic 550 ± 10 nm radiation. Interestingly, the production of H2 with [RuII(dcbpy)3] is negligible at pH > 6 because [RuII(dcbpy)3] is only adsorbed on TiO2 in acidic media. In contrast, for Sn porphyrin, photoactivity is observed over a wide pH range. It is thought that the anchoring of Ru-based dyes onto the TiO2 surface is required for the visible light sensitization because the excited free Ru-based dyes rapidly decay to the ground state if strong coupling between the excited orbitals of Ru-based dyes and Ti3d orbitals is absent. However, because of its strong oxidation power, the excited free Sn porphyrin is ﬁrst reduced by EDTA to give a stable and long-lived π-radical anion [Sn-Por•−]−, which is able to survive slow diﬀusion onto the TiO2 surface for 3775

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

approximately −0.7 V (vs SCE). As a consequence, photoinduced electron transfer from 1*[Zn-Por] to TiO2 in nanorods is thermodynamically possible, which is conﬁrmed by femtosecond transient absorption measurements. The subsequent reduction of [Zn-Por•+]+ by NADH (E1/2 = 0.76 V vs SCE) or ascorbic acid (E1/2 = 0.29 V vs SCE) is thermodynamically possible because the redox potential of [Zn-Por•+]+/[Zn-Por] is 0.98 V (vs SCE). 6.1.4. Porphyrin-Based Systems for OER. Crabtree and co-workers reported a TiO2 photoanode codeposited with Zn porphyrin 199 and an Ir catalyst for photocatalytic water oxidation (Figure 96).376 Zn porphyrin 199 and the Ir catalyst

the electron transfer. In addition, the photocatalytic activity of Sn porphyrin is initiated primarily by the Q-band excitation and not by the Soret band excitation. To facilitate electron transfer from photosensitizers to Pt catalysts, Yang and co-workers combined tetra(4hydroxylphenyl)porphyrin and Pt particles on reduced graphene oxides and investigated photocatalytic H2 evolution with this porphyrin-Pt-functionalized graphene material.373 Porphyrin could be readily immobilized on graphene nanosheets through noncovalent electrostatic and π−π stacking interactions. The rapid electron transfer from photoexcited porphyrin to graphene was indicated by the results of ﬂuorescence quenching and photocurrent enhancement. For a porphyrin-Pt-functionalized graphene composite on ITO electrodes, a photocurrent of 1.2 μA cm−2 was observed under UV−visible light irradiation, whereas control experiments using ITO electrodes covered with porphyrin-Pt or graphene-Pt showed negligible photocurrent responses. The total amount of H2 evolved in 5 h of photocatalysis with graphene-Pt, porphyrin-Pt, and porphyrin-Pt-functionalized graphene was 0.89, 2.74, and 5.29 mmol g−1, respectively. The enhancement observed for porphyrin-Pt-functionalized graphene could be attributed to the eﬃcient electron transfer from photoexcited porphyrin to Pt particles through graphene sheets, which act as a solid-state electron mediator to facilitate the charge separation and to suppress the charge recombination. Importantly, with the addition of cetyltrimethylammonium bromide, the performance of porphyrin-Pt-functionalized graphene was further enhanced. The surfactant cetyltrimethylammonium bromide is thought to prevent the aggregation of composites and thus enhance the photocatalytic performance. In addition to the previously mentioned molecular porphyrin photosensitizers, porphyrin-based MOF materials have also been studied in photocatalytic H2 evolution. For example, Rosseinsky and co-workers reported water-stable porphyrinbased MOFs for visible-light HER photocatalysis.374 The reaction of AlCl3 and meso-tetra(4-carboxylphenyl)porphyrin produced the MOF material 196. The postsynthetic metalation of the porphyrin unit by Zn produced a new MOF material 197. Two systems were studied: MOF/MV2+/EDTA/Pt and MOF/EDTA/Pt. In the ﬁrst system, a small amount of H2 was produced with visible light irradiation in aqueous solutions for both 196 and 197. In the second system, a noticeable increase in H2 generation was observed for both MOFs, and the quantum yield was enhanced by more than 1 order of magnitude. The rate of H2 production in the photocatalytic MOF/EDTA/Pt system was 200 μmol g−1 h−1 for 196 and 100 μmol g −1 h −1 for 197 after an induction period of approximately 3 h. Fukuzumi and co-workers reported photocatalytic H2 evolution using a porphyrin-based MOF encapsulating Ptcolloid-deposited TiO2 nanoparticles.375 This Zn porphyrin-PtTiO2 composite 198 (Figure 95) was synthesized. Photocatalytic H2 evolution was conducted in a pH 4.5 aqueous buﬀer solution containing 198 as the photocatalyst and NADH or ascorbic acid as the sacriﬁcial electron donor under visible light irradiation (λ > 420 nm). Importantly, the amount of generated H2 is ∼60 mol g−1 (based on Pt) for composite 198, which is 2 orders of magnitude greater than ∼0.3 mol g−1 (based on Pt) for the nonencapsulated simple mixture of Pt/ TiO2 and Zn porphyrin MOF. Electrochemical data showed that the redox potential of [Zn-Por•+]+/1*[Zn-Por] was −1.09 V (vs SCE), and the conduction band of TiO2 was

Figure 96. Zn porphyrin 199 and the Ir complex used for the photocatalytic OER.

contain carboxylic acid groups for attachment onto TiO2 particles. The estimated potential for excited-state couple 199•+/199* is approximately −0.77 V (vs NHE), and the conduction band of TiO2 is approximately −0.57 V (vs NHE). Therefore, the couple is energetically capable of transferring an electron from photoexcited 199* into the conduction band of TiO2 to generate 199•+ at the surface. The redox potential of 199•+/199 is 1.35 V (vs NHE), which is suﬃciently high to oxidize water. Photocurrent measurements were made under visible light irradiation (λ > 400 nm). With an applied bias potential of 0.3 V, a signiﬁcant and sustained photocurrent of ∼30 μA cm−2 was observed. Belcher and co-workers reported sustained light-driven water oxidation catalyzed by a coassembly of Zn porphyrin and an Ir oxide hydrosol cluster.377 Zn deuteroporphyrin IX 2,4 bisethylene glycol and Ir oxide (IrO2) were coassembled in close proximity in a genetically engineered M13 virus scaﬀold to produce the photocatalytic composite. The observed photocatalytic activity for water oxidation showed a TON ≈ 1100 and a TOF ≈ 1.68 s−1 (based on Ir), values that were approximately six times higher than those with unconjugated Zn porphyrin and IrO2 catalyst. The quantum yield was 0.86 using monochromic radiation at 550 nm with a light intensity of 200 mW cm−2. In addition, the use of an excess of IrO2 with Zn porphyrin could increase the possibility and the rate of hole transfer from oxidized photosensitizers to adjacent catalysts and could thus improve the durability of this photocatalytic composite. The photocatalytic durability was further enhanced with porous microgels as an immobilization matrix to prevent aggregation, and the resulting material could be reused with 56% activity retention after four cycles. Natali and co-workers reported light-driven water oxidation with Zn tetrakis(1-methyl-pyridinium-4-yl)porphyrin as the photosensitizer, tetraruthenium polyoxometalate as the catalyst, and persulfate as the sacriﬁcial oxidant in neutral aqueous 3776

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

media.378 Brouwer and co-workers reported light-driven water oxidation with water-soluble Pt meso-tetrakis(4carboxylphenyl)porphyrin as the photosensitizer and persulfate as the sacriﬁcial oxidant in neutral phosphate buﬀer.379 A variety of OER catalysts have been used in this study, including two homogeneous catalysts (Ir carbene and Co4O4 complexes) and two heterogeneous catalysts (IrO2 and Co3O4 nanoparticles). Electrochemical data showed that the PtIII/PtII redox potential was at 1.42 V (vs NHE), implying that it was able to oxidize these OER catalysts for water oxidation catalysis (the onset potential for water oxidation was approximately 1.20 V for Ir carbene, 1.32 V for Co4O4, 1.15 V for IrO2, and 1.17 V for Co3O4). Photocatalytic studies showed that Pt porphyrin had signiﬁcantly enhanced activity and stability compared with [Ru(bpy)3]2+. In addition to its improved photochemical stability, Pt porphyrin was considered a better photosensitizer than [Ru(bpy)3]2+ in terms of the light harvesting eﬃciency and the thermodynamic driving force because E1/2(PtIII/PtII) is approximately 200 mV higher than E1/2(RuIII/RuII). The quantum yield of water oxidation was estimated to be 1.1% for Ir carbene and Co4O4 complexes and 0.6% for IrO2 and Co3O4 nanoparticles. Imahori, Sun, and co-workers reported light-driven water oxidation using a TiO2 electrode covalently linked with the sensitizer-catalyst dyad (Figure 97).380 The sensitizer is a Zn

band of TiO2 is thermodynamically feasible because the driving force for such an electron injection was determined to be 0.5 eV. Photoelectrochemical measurements showed that the photocurrent density reached its maximum of 0.13 mA cm−2 at an external bias of −0.2 V (vs NHE) using the dyad/TiO2/ FTO working electrode under irradiation. A TON of 6 for H2 evolution and a TON of 1.3 for O2 evolution were obtained. Initial incident photon-to-current eﬃciencies of 18% at 424 nm and 6.4% at 564 nm were attained under monochromatic illumination with an external bias of −0.2 V (vs NHE). After photoelectrochemical studies, this dyad/TiO2/FTO electrode was analyzed, showing the accumulation of high oxidation state Ru species (RuIII and RuIV) on the electrode. It was proposed that upon irradiation, photoinduced electron transfer would proceed from 1*[Zn-Por] to the conduction band of TiO2, which would be followed by an intramolecular electron transfer from RuII to [Zn-Por•+]+ to aﬀord RuIII-[ZnPor]/TiO2(e−). This state could be regarded as a distant hole− electron pair, and its charge recombination process should be suﬃciently slow to make proton reduction at the counter electrode possible. However, because the oxidation potential of Zn porphyrin in an aqueous environment is approximately 1.0 V (vs NHE), which is insuﬃcient to oxidize high oxidation state Ru centers, catalytic activity shows a gradual drop and the TOF for O2 evolution is low. 6.2. Materials Derived from Porphyrin and Corrole Precursors

Carbon-supported transition metal materials with M-Nx/C (M = Mn, Fe, Co, Ni; x = 2−4) active sties have been intensively investigated as ORR electrocatalysts.79 Because porphyrin ligands are able to establish a stable M-N4 coordination environment for transition metal ions, metal porphyrin complexes have been widely used as precursors to create carbon-supported transition metal ORR catalysts. This topic, however, exceeds the scope of this review, which focuses on the molecular nature of porphyrin- and corrole-based catalytic systems. Therefore, we will only brieﬂy introduce several recent advancements regarding materials for catalytic ORR derived from porphyrin and corrole precursors. In 2014, Dai and co-workers reported two-dimensional covalent metal porphyrin polymers as precursors for making highly eﬃcient ORR electrocatalysts.381 A class of 2D covalent metal porphyrin polymers of Mn, Fe, and Co were synthesized, and subsequent carbonization at 950 °C produced graphenelike materials (Figure 98). Raman spectra of these carbonized materials showed a strong G band and a single-peak 2D band, indicating a multilayered graphitic structure. Porphyrin ring structures in these carbonized materials have been largely preserved. The CV measurements of these materials showed well-deﬁned cathodic ORR peaks in O2-saturated 0.1 M KOH aqueous solutions. Importantly, more positive ORR potentials and higher cathodic currents were observed for materials using metal-incorporated precursors compared with those using metal-free porphyrin precursors. In particular, the ORR peak potential for Fe- and Co-derived materials reached 0.77 V (vs RHE). The number of electrons transferred per O2 molecule was 3.81 and 3.56 at 0.35 V (vs RHE) for Fe- and Co-derived materials, respectively. Fe-derived material outperformed commercial Pt/C; each had similar onset potential, but the catalytic current at 0.75 V (vs RHE) was 13.50 mA cm−2 for Federived material and 9 mA cm−2 for Pt/C. Fe- and Co-derived materials also showed great stability and tolerance to methanol-

Figure 97. Schematic diagram of visible light-driven water oxidation using wide-bandgap TiO2 functionalized with a Ru OER catalystporphyrin-linked dyad as the photoanode. Reprinted with permission from ref 380. Copyright 2016 Royal Society of Chemistry.

porphyrin, and the OER catalyst is a Ru complex. The CV of this dyad in THF showed RuIII/RuII, RuIV/RuIII, and [ZnPor•+]+/[Zn-Por] redox peaks at 0.61, 0.96, and 1.29 V (vs NHE), respectively. The increased solvent polarity achieved by adding water resulted in a shift of the [Zn-Por•+]+/[Zn-Por] redox potential in the negative direction, but the RuIII/RuII and RuIV/RuIII redox potentials were nearly unchanged due to the protective eﬀect of Ru ligands from surrounding solvents. These results indicate that the use of more hydrophobic environment is beneﬁcial in providing the Zn porphyrin sensitizer with a higher oxidation potential. The electron transfer from the photoexcited 1*[Zn-Por] to the conduction 3777

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 98. Two-dimensional covalent metal porphyrin polymers and carbonization to make graphenelike materials for electrocatalytic ORR. Reprinted with permission from ref 381. Copyright 2014 Wiley-VCH.

respectively), and the number of electrons transferred per O2 molecule was 4.1 ± 0.1 in the potential range of 0.6−0.1 V (vs RHE). These numbers were 3.2 for Co-based, 2.6 for Zn-based, and 2.3 for H2-based materials. In addition, these materials showed great stability and tolerance in methanol crossover eﬀect studies and were highly active in O2-saturated 0.1 M HClO4 aqueous solutions with similar activity and selectivity as that observed in alkaline media. In 2014, Müllen, Feng, and co-workers reported the Coporphyrin-based conjugated mesoporous polymer 200 as the precursor for an ORR electrocatalyst.383 Polymer 200 was synthesized via Yamamoto polycondensation of CoII tetrakis(4bromophenyl)porphyrin. Subsequent thermal treatment at 600, 800, and 1,000 °C produced porous carbon materials of 200600, 200-800, and 200-1000, respectively. The CV measurements of these materials in O2-saturated 0.1 M KOH showed well-deﬁned cathodic ORR currents with the onset potential of −0.12 V (vs Ag/AgCl) for 200-800, −0.15 V (vs Ag/AgCl) for 200-600, −0.14 V (vs Ag/AgCl) for 200-1000, and −0.25 V (vs Ag/AgCl) for 200. The half-wave potential was −0.18 V (vs Ag/AgCl) for 200-800, −0.26 V (vs Ag/AgCl) for 200-600, −0.23 V (vs Ag/AgCl) for 200-1000, and −0.48 V (vs Ag/ AgCl) for 200. In addition, 200-800 showed a highly stable diﬀusion-limiting current of ∼4.6 mA cm−2, which was superior to 200-600, 200-1000, 200, and Pt/C (∼4.5 mA cm−2). The number of electrons transferred per O2 molecule was 3.83− 3.86 at −0.40 to −0.30 V (vs Ag/AgCl) for 200-800, and these numbers were 3.55 for 200, 3.63 for 200-600, and 3.60 for 2001000 at −0.35 V (vs Ag/AgCl). Similar activity and selectivity trends were observed in O2-saturated 0.5 M H2SO4. Stability tests conﬁrmed that 200-800 had an excellent catalytic stability for ORR in both alkaline and acidic solutions. Three factors were suggested to account for the outstanding ORR activity: (1) high surface area (∼480 m2 g−1) and unique mesoporous (∼5.4 nm) structure; (2) high content of coordinated Co (∼6 wt %); and (3) low electrolyte resistance. In addition to metal porphyrin-based polymers and MOFs as the precursors for ORR catalysts, simple metal porphyrins have also been heat treated to prepare catalysts for O2 reduction. Schulenburg and co-workers reported the heat treatment of carbon-supported FeIII tetramethoxyphenylporphyrin chloride as a precursor to make an ORR electrocatalyst and investigated

crossover and CO-poisoning eﬀects. Similar catalytic activities were obtained in O2-saturated 0.1 M HClO4 aqueous solutions. In 2015, Feng and co-workers reported heterometallic MOFs consisting of Zr and metal porphyrins of Fe, Co, and Zn (Figure 99) and their use as catalyst precursors for ORR.382

Figure 99. (a) Augmented tetracarboxylic porphyrinic linkers and (b) 12-connected Zr6(O)4(OH)4(O2C)12 cluster. (c) Cubic cage with 2.5 nm edge length. (d) The 3D network with Zr6 clusters shown as polyhedra and 3D cubic-cavity packing in the MOF. Color scheme: Zr (teal); Fe (lime); O (red); N (blue); and C (gray). (e) Photograph of the resulting MOF crystals. Reprinted from ref 382. Copyright 2015 American Chemical Society.

Solvothermal reactions of tetrakis(4-carboxybiphenyl)porphyrin or its metal complexes with ZrOCl2·8H2O yielded black cubic ∼0.20 mm-sized single crystals. Subsequent carbonization at 700 °C converted the crystals to porphyrinic carbons, which retained microcubic morphology and had porphyrinic active sites, hierarchical porosity, and highly conductive networks. The CV measurements in O2-saturated 0.1 M KOH solution showed well-deﬁned cathodic ORR peaks for these carbonized MOF materials with an activity order of Fe > Co > Zn ≈ H2. The Fe-based MOF material displayed the most positive ORR peak potential at 0.836 V (vs RHE, the onset and half-wave potentials were 0.950 and 0.802 V, 3778

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

the structure and stability of the active sites.384 The data suggest that after heat treatment, a six-coordinated FeIII species is the active site, and the main electrocatalytically reduced product of O2 is H2O2, which causes the degradation of active sites. Fe porphyrins on carbon supports have been widely used to make pyrolyzed Fe-Nx/C materials for ORR catalysis, but monolayer coverage of Fe porphyrin precursors on the exterior surface of carbon supports to create large volumetric active site densities is challenging. Shan and co-workers reported the direct construction of a three-dimensional ordered mesoporous Fe-porphyrin-like material with the use of FeIII tetrapyridylporphyrin chloride.385 This Fe porphyrin is soluble in polar solvents and can be readily introduced into channels of mesoporous SBA-15 silica templates by a wet-impregnation procedure. Subsequent heat treatment at 700 °C produced the desired material with Fe-Nx species uniformly distributed on the carbon support formed during the graphitization of frameworks. This material showed great electrocatalytic ORR activity in both alkaline and acidic media. For example, in O2saturated 0.1 M KOH, the cathodic ORR peak current was at −0.15 V (vs Ag/AgCl), and the number of electrons transferred per O2 reduction was 4.1. Control experiments using materials made from tetrapyridylporphyrin as the precursor or from Fe porphyrin but pyrolyzed under higher temperature (i.e., 900 °C) to decompose Fe-Nx active sites showed much lower catalytic activity. These results indicate that Fe-porphyrin-like Fe-Nx species are the critical active sites for catalytic O2 reduction. Metal corroles have also been used as precursors to prepare pyrolyzed materials for ORR catalysis. For example, in 2013, Wang and co-workers reported that carbon-black-supported pyrolyzed (nitrosyl) FeIII triphenylcorrole was a highly eﬃcient electrocatalyst for the four-electron reduction of O2.386 X-ray absorption near edge structure (XANES) and extended X-ray absorption ﬁne structure (EXAFS) measurements provided valuable information to better understand the structural features related to the high performance of this pyrolyzed Fecorrole/C material, which can be attributed to the network structure of polyaromatic hydrocarbons, the quaternary (graphitic)-type N atoms, and the coordination structure of Fe-N4 species. Another example was reported by Chen and coworkers,387 who used carbon-black-supported pyrolyzed (triphenylphosphine) CoIII triphenylcorrole as the ORR electrocatalyst. Pyrolyzed Co-corrole/C at 700 °C exhibited optimized ORR activity in the direct four-electron reduction of O2 to H2O, and its use in the cathode of a H2-O2 polymer electrolyte fuel cell showed a maximum power density of 275 mW cm−2. The results from XPS and XANES measurements indicated that the coordination structure and the oxidation state of Co corroles changed during pyrolysis, leading to increased ORR activity. Chen and co-workers reported carbon-blacksupported pyrolyzed vitamin B12 as a nonprecious ORR catalyst.388 In a H2-O2 polymer electrolyte fuel cell using pyrolyzed vitamin B12/C, a maximum power density of 370 mW cm−2 and current density of 0.720 A cm−2 at 0.5 V applied potential and 70 °C were obtained. Wang and co-workers reported carbon-black-supported pyrolyzed Co corrin, Co triphenylcorrole, and Co tetramethoxyphenylporphyrin for electrocatalytic O2 reduction.389 Their main conclusions are the following points. (1) Pyrolyzed Cocorrin/C has a higher ORR activity than pyrolyzed Co-corrole/ C and Co-porphyrin/C, and their electron-transfer numbers are

3.90, 3.87, and 3.37 at 0.3 V (vs RHE), respectively. (2) The Co-N4 coordination environment is mostly retained in pyrolyzed Co-corrin/C, and this structure is partially destroyed during pyrolysis in the Co corrole and Co porphyrin. These results suggest that the structure and coordination of Co-N4 active sites are key factors aﬀecting ORR activity. Savastenko and co-workers investigated plasma-treated nonprecious catalysts for ORR.390 Compared to the thermal pyrolysis of catalyst precursors, plasma treatment does not require a heat treatment step under elevated temperatures. Six diﬀerent carbon-supported Fe and Co porphyrins and phthalocyanines were treated with Ar-, N2-, or Ar:O2-radio frequency plasma to obtain Fe-Nx/C or Co-Nx/C catalysts. The main conclusions follow. (1) Plasma treatment of these catalyst precursors can improve their electrocatalytic O2 reduction capacity in terms of onset potential and catalytic current density. (2) Catalytic activity increases with the increase of the surface concentration of N and the relative concentrations of N-metal (or N−H) and N-pyrrolic species. Instead of directly using carbon-supported metal porphyrins or metal corroles as precursors to make catalytically active materials for O2 reduction, Su and co-workers synthesized three Fe/N-containing ORR electrocatalysts by pyrolyzing a mixture of polymerized o,m,p-phenylenediamine, ferric chloride, and carbon black, and they elucidated the structure-performance relationship of the resulting catalysts by determining the active site structures of Fe-Nx/C materials with the use of advanced electron microscopy and Mössbauer spectroscopy.391 Electrocatalytic studies in O2-saturated 0.5 M H2SO4 showed that onset potentials for Fe-PoPD/C, Fe-PmPD/C, and Fe-PpPD/C were 0.681, 0.819, and 0.826 V (vs RHE), respectively. Importantly, catalysts pyrolyzed without Fe showed decreased activity with the onset potential at 0.4−0.5 V (vs RHE), indicating that Fe plays a decisive role in Fe-Nx/C catalysts. Although FeS and Fe3C species were found in the pyrolyzed graphitic carbon material, these two species could not be ORR active sites because they were not stable in acidic media. In addition to FeS and Fe3C, other Fe coordination compounds were found in catalysts and were considered the true active sites for ORR. On the basis of the results from XPS and Mössbauer spectroscopy, a six-coordinated FeIII species with a porphyrinlike coordination environment, [Fe III (py)2 -Por] (py = pyridine), was proposed as the active site for ORR. In addition, Fe-PpPD/C mainly catalyzed the four-electron reduction of O2 to H2O with the electron-transfer number of 3.8 and showed excellent stability during catalysis. Du and co-workers reported carbon-supported pyrolyzed Co porphyrins as active OER electrocatalysts.392 The CVs of pyrolyzed catalysts on FTO electrodes in 0.1 M pH 9.2 borate buﬀer showed a catalytic OER current with an onset potential of ∼1.12 V (vs NHE), corresponding to an onset overpotential of 430 mV. The eﬀects of Co content and heating temperature were investigated, revealing 15 wt % Co porphyrin and 1000 °C for pyrolysis as optimal conditions. The maximum TOF number of 0.078 s−1 (based on Co) was achieved under an applied anodic potential of 1.31 V (vs NHE) at pH 9.2. Bell and co-workers used DFT calculations to investigate electrochemical OER and ORR processes catalyzed by metalporphyrin-like centers incorporated into graphene or SWCNTs.393 There were two key ﬁndings. (1) For ﬁrst-row transition metal cations, the minimum overpotential value, ηOER, is dictated by the diﬀerence of Gibbs free energies for the binding of adsorbed OOH and OH species, ΔG*OOH − ΔG*OH. 3779

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

By using porphyrin-coordinated ﬁrst-row transition metal cations, the values of ΔG*OOH − ΔG*OH can be decreased to between 3.4 and 2.8 eV with the lowest values for trivalent metal cations of late transition metal elements (e.g., CoIII, FeIII, and NiIII). (2) The lowest ηOER value will occur when the value of ΔG*O lies midway between the values of ΔG*OOH and ΔG*OH. The axial ligands play a critical role in tuning the overpotential for OER and ORR. For example, if a porphyrin motif is incorporated into graphene or SWCNTs, the lowest ηOER value is ∼0.35 V using N-heterocyclic carbene C as the axial ligand. In the case of the ORR, when the porphyrin motif is incorporated into graphene, the lowest ηORR value of 0.37 V is obtained for the PH3 axial ligand; when the porphyrin motif is incorporated into SWCNTs, the lowest ηORR value of 0.31 V is obtained for the CO axial ligand. This study is therefore important for its prediction that the ηOER and ηORR values of metal porphyrins (or porphyrin-like ligands) can be signiﬁcantly decreased to smaller values than those observed with highly active metal oxides (for OER) and metallic Pt (for ORR) by using the ﬁrst-row late transition metal cations (e.g., CoIII, FeIII, and NiIII ions), incorporating them into graphene and SWCNTs, and employing an appropriate axial ligand.

ligand-centered. Upon the addition of two equivalents of CF3SO3H, the redox potentials of CoIII/CoII and CoII/CoI positively shifted to 0.81 and 0.02 V (vs SCE), whereas the redox potentials of the two ligand-centered oxidation processes were nearly unchanged. This result indicates that the electrondonating ability of the phthalocyanine ligand decreases upon protonation. Upon the ﬁrst oxidation, phthalocyanine is deprotonated due to the decreased basicity of the ligand bound to CoIII. As a result of this deprotonation, the CoIII/CoII oxidation process became irreversible, and the two ligand oxidation processes remained unchanged. The protonation constants of CoI, CoII, and CoIII states of 201 were determined to be log K = 4.3, 2.7, and 0.26, respectively. Catalytic O2 reduction was observed in O2-saturated benzonitrile containing 201, HCOOH, and Me2Fc. Because the amount of generated Me2Fc+ was twice the amount of consumed O2, O2 was selectively reduced by two electrons to produce H2O2. Kinetic studies revealed that the formation rate of Me2Fc+ obeyed zero-order kinetics, and the zero-order rate constant showed a saturation behavior with an increasing concentration of HCOOH but increased linearly with increasing concentrations of 201 and O2. These results indicate that a PCET process from 201 to O2 to produce CoIII and HO2• is the rds in the catalytic cycle (Figure 101). The formed

6.3. Porphyrin and Corrole Analogues in Catalysis

In this section, several representative examples of porphyrin and corrole analogues as ORR catalysts will be discussed. Although many porphyrin and corrole analogues have been used and reported as catalysts for O2 reduction, we will only brieﬂy introduce several examples to address the general potential of using these complexes in catalysis. In 2012, Fukuzumi and co-workers investigated the selective twoelectron reduction of O2 to H2O2 catalyzed by a saddledistorted Co phthalocyanine 201 (Figure 100).394 The CV of 201 in deaerated benzonitrile showed three reversible oxidations and one reversible reduction. The ﬁrst oxidation (E1/2 = 0.50 V vs SCE) and reduction (E1/2 = −0.31 V vs SCE) processes were assigned to CoIII/CoII and CoII/CoI redox couples, respectively, and the second (E1/2 = 0.93 V vs SCE) and third (E1/2 = 1.15 V vs SCE) oxidation processes were

Figure 101. Proposed reaction mechanisms of O2 reduction by diﬀerent Fc derivatives catalyzed by 201. Redrawn from ref 394. Copyright 2012 American Chemical Society.

CoIII state can be rapidly reduced by Me2Fc because the reduction potential of Me2Fc (0.26 V vs SCE) is negative by 240 mV relative to the CoIII/CoII reduction potential (0.50 V vs SCE). The rds rate constant of the PCET from 201 to O2 in the presence of 0.50 M HCOOH was (1.4 ± 0.1) × 102 M−2 s−1 at 298 K, which agreed very well with the value of (1.6 ± 0.3) × 102 M−2 s−1 determined from the reaction of 201 and O2 under single-turnover conditions in the presence of HCOOH but without Me2Fc.

Figure 100. Co complexes 201 and 202 and their corresponding protonated species investigated as ORR catalysts by Fukuzumi and coworkers. 3780

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

When Me10Fc was used as the reductant, the reaction was much faster and was completed in a few seconds. Stopped-ﬂow studies revealed that the formation rate of Me10Fc+ obeyed pseudo-ﬁrst-order kinetics, and the pseudo-ﬁrst-order rate constant remained constant irrespective of O2 concentration but was proportional to the initial concentration of 201. This result indicates that the rds in the case of Me10Fc as the reductant is not the PCET from 201 to O2 but the PCET from Me10Fc to the protonated 201 (Figure 101) because the reduction potential of Me10Fc (−0.08 V vs SCE) is only 100 mV more negative than the CoII/CoI reduction potential (0.02 V vs SCE) of protonated 201. The observed rate constant increased with the increasing concentration of HCOOH without saturation. The second-order rate constant of the PCET from Me10Fc to protonated 201 in the presence of 0.50 M HCOOH was (1.6 ± 0.1) × 105 M−1 s−1 at 298 K, which agreed very well with the value of (1.1 ± 0.2) × 105 M−2 s−1 determined from the reaction of 201 and Me10Fc under singleturnover conditions in the presence of HCOOH but without O2. This work demonstrates that the protonation of 201 signiﬁcantly enhances the catalytic reduction of O2 by Fc derivatives, and the catalytic pathway is switched from a CoIII/ CoII cycle to a CoII/CoI cycle by increasing the reducing power of the reductant. Similarly, Fukuzumi and co-workers investigated the selective two-electron reduction of O2 to H2O2 catalyzed by Co chlorin complex 202 (Figure 100).395 The protonation equilibrium constant of 202 with HClO4 was 2.2 × 104 M−1 in deaerated benzonitrile at 298 K, and this number was 1.1 × 102 M−1 for the one-electron oxidized species. Reversible binding of O2 to 202 was conﬁrmed by UV−visible and EPR spectroscopy. The CV of 202 in deaerated benzonitrile showed two reversible oxidations at E1/2 = 0.37 and 0.82 V (vs SCE), corresponding to the CoIII/CoII and ligand-centered oxidation, respectively. In the presence of HClO4, the second oxidation was nearly unchanged, whereas the CoIII/CoII oxidation potential was positively shifted to E1/2 = 0.48 V (vs SCE). This result indicates the protonation of the chlorin ligand (Figure 100). As a comparison, the CV of Co octaethylporphyrin showed two reversible oxidations at E1/2 = 0.31 and 0.91 V (vs SCE), corresponding to similar CoIII/CoII and ligand-centered oxidation, respectively. However, unlike 202, the demetalation of Co from the porphyrin ligand was observed in the presence of HClO4. The CV of 202 in O2-saturated benzonitrile showed catalytic ORR currents with an onset potential of 0.6 V (vs SCE). Catalytic O2 reduction with 202 was studied in air-saturated benzonitrile solution containing HClO4 and Me2Fc; the quantity of Me2Fc+ generated was twice that of O2 consumed, revealing the selective two-electron reduction of O2 to H2O2. The steady state observed during catalytic O2 reduction was protonated 202, which indicated that the PCET from protonated 202 to O2 was the rds under these catalytic conditions. The observed rate constant obeyed pseudo-ﬁrstorder kinetics; it was proportional to the concentration of 202 without an intercept, increased linearly with an increasing concentration of HClO4 with an intercept, and remained constant irrespective of the Me2Fc concentration. This result further conﬁrmed that the PCET from protonated 202 to O2 was the rds under these catalytic conditions. On the basis of kinetic studies, a reaction mechanism was proposed (Figure 102), and the second-order rate constant was determined to be (1.2 ± 0.2) × 103 M−1 s−1; the third-order rate constant was

Figure 102. Proposed reaction mechanisms of O2 reduction by Fc derivatives catalyzed by complex 202. Redrawn from ref 395. Copyright 2013 American Chemical Society.

(1.9 ± 0.3) × 105 M−2 s−1. These numbers agreed well with those of (1.0 ± 0.3) × 103 M−1 s−1 and (2.1 ± 0.2) × 105 M−2 s−1 determined from single-turnover experiments without Me2Fc. These agreements conﬁrmed that the PCET from protonated 202 to O2 was the rds in the catalytic cycle. Importantly, the reduction of one-electron oxidized 202 (CoIII state) by one equivalent of Me2Fc was studied by a stopped-ﬂow technique, giving the second-order rate constant of (4.0 ± 0.3) × 107 M−1 s−1, a number that is 4 orders of magnitude larger than the second-order rate constant of PCET from 202 to O2. Therefore, the use of the stronger one-electron donor Me8Fc gave the same observed ﬁrst-order rate constant because the electron transfer from Me2Fc or Me8Fc to oneelectron oxidized 202 was not the rds. In addition, the use of the weaker one-electron donor Fc (E1/2 = 0.37 V vs SCE) or BrFc (E1/2 = 0.54 V vs SCE) also gave similar observed ﬁrstorder rate constants, although the electron transfer from BrFc to one-electron oxidized protonated 202 (E1/2 = 0.48 V vs SCE) is slightly endergonic. However, when Br2Fc (E1/2 = 0.72 V vs SCE) was used as the one-electron donor, the steady state observed during catalytic O2 reduction was one-electron oxidized protonated 202, the formation rate of Br2Fc+ became much slower, and the observed ﬁrst-order rate constant increased with increasing concentration of Br2Fc. These results suggest that the rds switches from the PCET reduction of O2 to the reduction of one-electron oxidized protonated 202 with the use of the much weaker one-electron donor Br2Fc. Love and co-workers reported a series of binuclear Co complexes of Schiﬀ base calixpyrrole ligands 203−205 (Figure 103) as catalysts for O2 reduction.396−398 The structure of bis(pyridine) adduct 203-py2 was determined, revealing a Pacman structural motif with two low-spin, square pyramidal CoIICoII ions.396 The reaction of 203 with O2 in THF is spontaneous, leading to the O2 adduct of 203 with two CoIIICoIII ions. Importantly, the structure of this O2 adduct was also determined. The two Co atoms adopt octahedral geometries with equatorial macrocyclic N4 donors and 3781

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Figure 103. Binuclear Pacman Co complexes 203−205 investigated as ORR catalysts by Love and co-workers.

pyridine-N and dioxygen-O donors in the axial sites. The O−O bond length is 1.361(3) Å, which is at the boundary of bridging end-on peroxido (1.34−1.53 Å) and superoxido (1.29−1.36 Å) units. Consequently, it is challenging to ascertain the degree of O2 reduction based on this O−O bond length. The reaction of 203 with O2 was further investigated by UV−visible and EPR spectroscopy, showing that O2 binding on 203 was irreversible and that both peroxido [CoIIICoIII(O2)] and superoxido [CoIIICoIII(O2)]+ species were generated in solution in a 9:1 ratio. In a subsequent work, these authors further investigated the O2 reduction activity of binuclear Co complexes 203 and 204.397 Although 203 and 204 rapidly react with O2, the catalytic reduction of O2 to H2O shows low TON and TOF, largely due to the formation of stable hydroxy-bridged binuclear Co complexes during catalysis. By using complex 205 with a modiﬁed structure, Love and co-workers demonstrated that it was possible to shift the reactivity of this binuclear Co complex toward the catalytic four-electron reduction of O2.398 These authors rationalized that the use of sterically hindering aryl meso-groups such as ﬂuorenyl would obviate the formation of hydroxy-bridged binuclear Co complexes. Importantly, bridging superoxo [CoIIICoIII(O2)]+ was isolated and structurally characterized with a O−O bond distance of 1.389(4) Å and a Co···Co separation of 4.15 Å. Catalytic O2 reduction by 205 was conducted in an air-saturated benzonitrile solution containing Me2Fc and TFA. On the basis of the amount of Me2Fc+ generated, a four-electron reduction process was conﬁrmed. The catalytic activity of 205 was signiﬁcantly enhanced compared to 203 under the same conditions. The formation rate of Me2Fc+ obeyed pseudo-ﬁrst-order kinetics, and the second-order rate constant was 6.1 × 102 M−1 s−1. Baumgarten, Müllen, and co-workers reported triangular trinuclear metal-N4 complexes 206 and 207 (Figure 104) and their high electrocatalytic O2 reduction capacity.399 The CV of carbon-black-supported 206 (20 wt %) in O2-saturated 0.1 M KOH aqueous solution showed a pronounced cathodic peak at −0.26 V (vs Ag/AgCl), indicating electrocatalytic ORR activity for 206/C. The data from RDE measurements and corresponding Koutecky−Levich analyses exhibited good linearity with nearly identical slopes over the potential range from −0.275 to −0.450 V (vs Ag/AgCl). These results suggest similar electron transfer numbers for O2 reduction under diﬀerent electrode potentials, and the linearity and parallelism of the plots indicate ﬁrst-order reaction kinetics with respect to the concentration of dissolved O2. The number of electrons transferred per O2 molecule was then calculated at 3.7 for 206/ C and 2.9 for 207/C. Compared to the benchmarked Pt/C

Figure 104. Triangular trinuclear metal-N4 complexes 206 (for Co) and 207 (for Fe) investigated as ORR electrocatalysts.

catalyst, the onset potentials for catalytic ORR current with 206/C (−0.14 V vs Ag/AgCl), 207/C (−0.14 V vs Ag/AgCl), and Pt/C (−0.11 V vs Ag/AgCl) were similar, whereas the catalytic current with 206/C (9.63 mA cm−2) was more than two times larger than that with Pt/C (4.44 mA cm−2) at −0.35 V (vs Ag/AgCl). Stability tests were performed at a constant potential of −0.26 V (vs Ag/AgCl), revealing that 80.6% of the current was maintained for 206/C, whereas only 52.5% of the current was maintained for Pt/C after 20000 s. These results reveal that the 206/C catalyst has better electrochemical activity and long-term stability than the Pt/C catalyst for ORR in alkaline media.

7. CONCLUDING REMARKS In this review, porphyrin- and corrole-catalyzed small molecule activations that are related to energy topics, including ORRs, HERs, and OERs, have been addressed. The activation/ reduction of O2 is an important reaction in both biological processes and energy conversion systems such as fuel cells and metal-air batteries. Due to its sluggish kinetics, ORR eﬃciency is considered one of the limiting factors of fuel cells; thus, exploring new ORR electrocatalysts with improved activity and stability is a subject of fundamental interest and signiﬁcance for renewable energy applications. Water splitting is an appealing method to convert solar energy to chemical energy with the production of H2 as a clean, carbon-free, and renewable energy source; thus, it is a promising approach for solving energy and environmental problems related to the use of fossil fuels. Extensive eﬀorts have been made to develop eﬃcient and robust catalysts for HER and OER. In particular, water 3782

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

are able to draw the following conclusions related to the design and development of more eﬃcient catalysts. First, Fe- and Cobased porphyrins and corroles have been dominantly studied as ORR catalysts, although other transition metal complexes (i.e., Cr, Mn, Ni, and Cu) and free-base porphyrins have been shown to be active for the catalytic reduction of O2. Co complexes in general display the highest activity in terms of catalytic currents (or rates) and overpotentials. However, mononuclear Co complexes usually catalyze the two-electron reduction of O2 to H2O2, whereas Fe complexes can intrinsically catalyze the fourelectron reduction of O2 to H2O. This activity diﬀerence is most likely due to two reasons: (1) the energy level of the highest-occupied 3d orbital of Fe-based porphyrins is higher than that of corresponding Co-based porphyrins, and (2) the heterolytic O−O bond scission of the Fe-OOH2 unit is relatively easier compared with Co-OOH2 unit because the resulting Fe-oxo unit is more stable. Self-assembled dimeric Co species or Co porphyrin/corrole dyads with face-to-face geometries and appropriate Co···Co distances for O2 binding are able to catalyze the four-electron reduction of O2. Second, in addition to the nature of metal centers, several other factors are critical in determining the reduction level of O2, including the electron transfer rate, the trans axial ligand, and the distal pocket environment. It is established that faster electron transfer from electrodes to catalytic sites generally leads to the four-electron reduction of O2. The electrondonating trans axial ligand will cause the reduction potential to shift in the negative direction, which is not favored for ORR, but it can increase the electron density on the metal center via the so-called “push eﬀect” to assist the O2 binding, subsequent inner sphere electron transfer to O2, and heterolytic O−O bond scission. The eﬀects of distal pocket environments are complicated. To improve the activity and selectivity for the four-electron reduction of O2, a distal pocket is desired that can quickly and strongly bind O2 and provide the correct protons to break O−O bonds. It is known that hydrogen-bonding interactions can stabilize O2 reduction intermediates and prevent the release of PROS, but these hydrogen-bonding interactions can also keep water molecules in the pocket that sterically hinder the binding of O2. In the case of proton delivery, the release of PROS will be increased due to excessively fast and excessively slow proton delivery rates. Consequently, the pKa of the distal pocket needs to be ﬁnetuned. Third, electron-withdrawing substituents on porphyrin and corrole macrocycles are favored. They can decrease the electron density on the macrocycle and thus increase its stability during catalysis, and they can cause the reduction potential to shift in the positive direction, which will decrease the overpotential for ORR and HER catalysts and increase the oxidizing power of OER catalysts. For example, Co porphyrins without strong electron-withdrawing groups are catalysts for O2 reduction but not as eﬃcient for proton reduction because O2 reduction is typically initiated at the CoII state, whereas CoI or Co0 are required for proton reduction. The absence of strong electronwithdrawing substituents makes the equilibrium potentials for CoI or Co0 too negative to reduce protons. It is worth noting that the metal complexes of tpfc ligand bearing three strong electron-withdrawing meso-pentaﬂuorophenyl substituents and its β-octaﬂuoro derivatives, as reported by Gross and coworkers, are versatile catalysts for ORRs, HERs, and OERs. As demonstrated in previous sections, corrole ligands are very eﬀective in stabilizing high-valent metal centers and accordingly

oxidation is the key step involved in natural and artiﬁcial photosynthesis for the solar energy conversion. However, because it is thermodynamically unfavorable and is challenging from a kinetic point of view, water oxidation is the bottleneck for large-scale water splitting. Noble metals and their complexes are very active for these processes. For example, Pt and Pt-based materials are excellent catalysts for both ORR and HER, and they are predominantly used as ORR catalysts in current fuel cells. For OER, Ru and Ir oxides and (oxy)hydroxides have been shown to generally have great electrocatalytic activity. Nevertheless, the scarcity and high cost of these precious metals pose serious limitations to widespread use and large-scale applications. Consequently, it is extremely desirable to ﬁnd low-cost, active, and robust catalysts for ORR, HER, and OER processes. Extensive eﬀorts have been made recently to develop eﬃcient and robust catalysts from cheap and earth-abundant metal elements that can catalyze these reactions at considerable rates at low overpotentials; these eﬀorts lead to the identiﬁcation of homogeneous and heterogeneous systems based on the ﬁrst-row transition metals, such as V, Cr, Mn, Fe, Co, Ni, and Cu, and systems based on mixed-metal materials. Although these catalytic systems have shown considerable activity and durability, substantial improvements are still required. As inspired by nature, metal porphyrins and corroles have been extensively investigated as catalysts for ORR. Deprotonated porphyrins are dianionic macrocycle ligands, whereas deprotonated corroles are trianionic macrocycle ligands. Both porphyrin and corrole ligands contain a proper size that can in principle bind almost all metal ions to give a four-coordinated structure. Such a rigid and stable coordination environment is crucial for metal centers to bind and activate a small molecule substrate. In addition to tight binding, negatively charged porphyrin and corrole ligands are very eﬀective in stabilizing high-valent metal centers, and associated with this eﬀect, they are able to oﬀer low-valent metal ions large reducing powers. These features are crucial in ORR, HER, and OER catalysis. For example, high-valent metal oxo intermediates are believed to be involved in both O2 activation and O2 evolution; the heterolytic O−O bond scission of metal-OOH2 units and its reversible reaction, the nucleophilic attack of water on metal-oxo units, both require the generation and stabilization of high-valent metal oxo intermediates. Another signiﬁcant feature of porphyrin and corrole ligands is that they are redox noninnocent, which greatly enriches the redox chemistry of their metal complexes. In contrast, planar and aromatic porphyrin and corrole rings enable strong adsorption of these molecules on carbon-based materials through noncovalent π−π interactions, which can further increase their catalytic activity and stability. For example, a very recent work by Cao and coworkers demonstrated that increased noncovalent π−π interactions between catalyst molecules and the supports signiﬁcantly improved the catalytic performance by not only increasing the adsorption of catalysts on the supports but also facilitating the electron transfer processes.400 Moreover, the carbon-rich backbones of porphyrin and corrole rings are beneﬁcial and attractive for use as precursors for the construction of pyrolyzed materials with highly active metalN4/C sites. All these features together make porphyrin and corrole complexes fascinating in catalytic small molecule activations. With the review of recent advances in porphyrin- and corrole-based systems for ORR, HER, and OER catalysis, we 3783

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

During the ORR, unstable and toxic peroxo/superoxo species will be generated. Nature systems evolved to completely reduce O2 into water, which ensures the maximum energy release and the minimal release of toxic intermediates into cells. To stabilize the highly active peroxo/superoxo intermediates, proton incorporation is crucial for this 4H+/4e− reduction pathway. In CcOs, the hydroxyl species are formed on both Fe and Cu centers after the O−O band cleavage. The two hydroxyl species individually accept a 1H+/1e− couple via another PCET process to form two water molecules.179,204,236,238,404 For the HER, the discharge of protons on the catalytic center to form a hydride species is a typical PCET process. Two hydride species will couple to release H2, or alternatively, another PCET process occurs with the hydride species for the formation of H2.280,283 Because proton transfer is the key determinant in these energy-related redox reactions, many studies have emphasized the creation of proton relays at the second coordination sphere of the catalytic metal center. These relays help in the proton translocation to increase the eﬃciency of the challenging small molecule activations. The coupling of electron transfer and proton transfer is the key point to understand the mechanism in these energy-related small molecule activations and to design eﬃcient artiﬁcial systems for energy conversion. Despite the achievements from exploring metal porphyrins and corroles in catalytic ORR, HER, and OER summarized in this review, more eﬀort is required to better understand the reaction mechanisms involved in these catalytic processes; to recognize key factors regulating the activity and selectivity; to develop new catalysts that are cost-eﬀective, eﬃcient, and robust; and to integrate these catalysts into energy conversion and storage systems. The development of novel procedures for synthesizing porphyrins and corroles, in particular porphyrinand corrole-based architectures, is important because current methodologies suﬀer from low yields and diﬃcult puriﬁcation procedures. Porphyrins and corroles have been used as photosensitizers in light-induced catalytic processes, but their limited stability under irradiation needs to be improved. In addition, several crucial events involved in photocatalytic water splitting, such as light sensitization, charge separation, water reduction to produce H2, and water oxidation to produce O2, have not been satisfactorily combined for the construction of an “artiﬁcial leaf” that can achieve eﬃcient solar to chemical energy conversion. A better understanding of the structure−activity relationships will provide new insights into the development of more eﬃcient catalysts. Moreover, although the trans eﬀect has been intensively investigated in O2 reduction and oxygen atom transfer reactions,93,94,407−409 such an eﬀect has not been well addressed in catalytic OER with metal porphyrins and corroles. Because the O−O bond formation step is typically the rds in the OER, it is expected that the trans eﬀect will play a signiﬁcant role in this process. As the analogue of porphyrins, corrolazine ligands have also been shown to stabilize highvalent metal ions, and metal corrolazines have been used as catalysts in oxygen atom transfer reactions.410−414 However, few studies have examined their catalytic features for OER. Compared to ORR, carbon-supported pyrolyzed metal porphyrins and corroles have rarely been used as OER catalysts. In summary, globally increasing energy demand and environmental issues caused by the use of fossil fuels have forced people to ﬁnd and use renewable, clean, and environmentally benign energy resources. The use of hydrogen fuel, which is an ideal alternative to fossil fuels, requires eﬃcient

oﬀering low-valent metal ions large reducing powers. The introduction of meso-pentaﬂuorophenyl and β-octaﬂuoro groups will further increase this eﬀect. Fourth, intramolecular acid−base groups are useful in these catalytic reactions, which require the concerted coupling of electron transfer and proton transfer. This process is known as the PCET process to avoid the generation of high-energy intermediates. In the cases of ORR and HER, the incorporation of proton relays in the second coordination sphere of molecular catalysts can signiﬁcantly improve the proton transfer rate. In the case of the OER, the incorporation of base groups can facilitate the O−O bond formation via a concerted O atomproton transfer pathway. Nocera and co-workers have contributed signiﬁcantly to this topic by designing and developing a series of hangman metal porphyrins and corroles that contain pendant acid/base groups hanging above the macrocyclic redox centers. These hangman macrocycle complexes have been examined as catalysts for ORR, HER, and OER, and more importantly, they provide an excellent platform to investigate the mechanistic details of the PCET chemistry and corresponding O−O, O−H, and H−H bondbreaking and bond-making processes involved in these catalytic reactions. The PCET process was ﬁrst reported in 1981 to describe the comproportionation of [Ru(bpy)2(py)OH 2]2+ and [Ru(bpy)2(py)O]2+, in which a 1H+/1e− couple was transferred from the former to the latter for the formation of two [Ru(bpy)2(py)OH]2+ molecules.401 Since then, PCET has been widely used to identify reaction pathways involving electron transfers that are also assisted or aﬀected by protons. The PCET processes broadly exist in energy-conversion reactions in nature systems, including the OER in oxygenevolving complexes, the ORR in CcOs, the HER in hydrogenases, and many others.402−404 The proton and electron transfers may occur synchronously, or they can proceed sequentially with the generation of proton-transferred or electron-transferred intermediates. The accompaniment of protons in these redox reactions can minimize the energy barriers of key electron transfer steps and stabilize highly reactive intermediates and thus allow high-eﬃciency energy conversion. The coupling of protons and electrons is in general thermodynamically favored in these redox reactions; however, the addition of protons to the reactions can always impede the kinetics.401 Thus, nature-based systems evolved sophisticated cofactors in enzymes to assist the proton transfer into the reaction species coordinated to the metal centers in the catalyst complexes.402−405 Inspired by nature, scientists paid extra attention to the pH values of the reaction media and designed molecular architectures with proton acceptor/donor docking near the catalytic centers.179,204,236,238,280,283,305 The OER from water oxidation is thermodynamically challenging with multiple steps of electron transfer. These electron transfer steps are highly unfavorable, and the OER reaction cannot be eﬃciently accomplished without proton translocation. At the ﬁrst step of water oxidation, the strong oxidative power of the catalytic center to accept an electron from water is the prerequisite for OER. Equally importantly, the presence of a base to accept one proton from water is indispensable. With this important PCET process, a hydroxyl adduct is formed on the catalytic center. Typically, another PCET will occur to this hydroxyl species to generate the oxo species before the formation of O2.305,402,403,405,406 In contrast, ORR is an exergonic reaction and is the fuel cell in mammals. 3784

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

China under Grants 21101170, 21203245, 21503126, 21573139, and 21673286, the Fundamental Research Funds for the Central Universities, and the Research Funds of Shaanxi Normal University and Renmin University of China.

generation through water splitting (photocatalytic HER and OER) and eﬃcient utilization through fuel cells (combined with ORR). Therefore, the research on ORR, HER, and OER catalysts is a signiﬁcant scientiﬁc challenge that requires the knowledge of numerous research ﬁelds.

ABBREVIATIONS AFM atomic force microscopy bpy bipyridine CcO cytochrome c oxidase Cor corrole CPE controlled potential electrolysis CV cyclic voltammogram DCE 1,2-dichloroethane DCM dichloromethane DFT density functional theory DMF dimethylformamide EDTA ethylenediaminetetraacetate EDX energy dispersive X-ray spectroscopy EIS electrochemical impedance spectroscopy EPG edge-plane pyrolytic graphite EPR electron paramagnetic resonance EXAFS extended X-ray absorption ﬁne structure η overpotential Fc ferrocene FOW foot-of-the-wave F8TPP 5,10,15,20-tetrakis(2,6-diﬂuorophenyl)porphyrin FTIR Fourier transform infrared spectroscopy FTO ﬂuorine-doped tin oxide GC glassy carbon HER hydrogen evolution reaction HS high spin HTB tetrakis(pentaﬂuorophenyl)boric acid ITIES interface between two immiscible electrolyte solutions ITO indium tin oxide KIE kinetic isotope eﬀect LS low spin MOF metal organic framework MWCNT multiwalled carbon nanotube NHE normal hydrogen electrode NMR nuclear magnetic resonance OER oxygen evolution reaction ORR oxygen reduction reaction PCET proton-coupled electron transfer Por porphyrin PROS partially reduced oxygen species RDE rotating disk electrode rds rate-determining step RHE reversible hydrogen electrode RRDE rotating ring-disk electrode SAM self-assembled monolayer SCE saturated calomel electrode SEM scanning electron microscopy SERRS surface-enhanced resonance Raman spectroscopy SHE standard hydrogen electrode SSHG surface second harmonic generation SWCNT single-walled carbon nanotube TEM transmission electron microscopy TFA triﬂuoroacetic acid THF tetrahydrofuran TOF turnover frequency TON turnover number TPA tripicolylamine

ASSOCIATED CONTENT S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.chemrev.6b00299. Refs with full author list for 1, 61, 86, 144, 152, 215, 231, 298, 363, 387, and 388 (PDF)

AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing ﬁnancial interest. Biographies Wei Zhang is currently an associate professor in the School of Chemistry and Chemical Engineering at Shaanxi Normal University. His current research is focused on the catalytic reactions of water splitting. He received his B.S. (2007) in chemistry from Peking University in China and his Ph.D. degree (2012) from Nanyang Technological University in Singapore with Professor Rong Xu. After photocatalytic CO2 reduction postdoctoral work with Professor Rong Xu, he joined the faculty at Shaanxi Normal University in 2014. He is a recipient of the 2012 Ph.D. Prize in environmental and sustainability research from the World Future Foundation. Wenzhen Lai is currently an associate professor in the Department of Chemistry at Renmin University of China. Her main research interests are the catalytic mechanisms of metal-containing systems. She received her B.S. (2002) in chemistry and her M.S. (2005) in physical chemistry from Sichuan University. In 2008, she obtained her Ph.D. in theoretical and computational chemistry from Nanjing University under the supervision of Prof. Daiqian Xie. She was a visiting student (2007−2008) at the University of New Mexico, working with Prof. Hua Guo in the ﬁeld of reaction dynamics. Between 2008 and 2011, she was a postdoctoral fellow with Prof. Sason Shaik at the Hebrew University of Jerusalem. In 2011, she joined the faculty of Renmin University of China. Rui Cao is currently a professor in the School of Chemistry and Chemical Engineering at Shaanxi Normal University. His main research interests are in bioinorganic chemistry, catalysis including water splitting (i.e., water oxidation and reduction) and the activation of small molecules, and molecular self-assembly. He received his B.S. (2003) in chemistry from Peking University in Beijing, China and his Ph.D. (2008) from Emory University in Atlanta, GA. He was the Dreyfus Postdoctoral Fellow from 2009 to 2011 in the Department of Chemistry at Massachusetts Institute of Technology under the guidance of the Arthur Amos Noyes Professor Stephen J. Lippard. In 2011, he became a professor in the Department of Chemistry at Renmin University of China, and he transferred to Shaanxi Normal University in 2014.

ACKNOWLEDGMENTS We are grateful for the support from the “Thousand Talents Program” of China, the National Natural Science Foundation of 3785

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews tpfc tren TTF vs XANES XPS

Review

(21) Zhou, T. H.; Du, Y. H.; Borgna, A.; Hong, J. D.; Wang, Y. B.; Han, J. Y.; Zhang, W.; Xu, R. Post-Synthesis Modification of a MetalOrganic Framework to Construct a Bifunctional Photocatalyst for Hydrogen Production. Energy Environ. Sci. 2013, 6, 3229−3234. (22) Xi, G. C.; Ye, J. H. Synthesis of Bismuth Vanadate Nanoplates with Exposed {001} Facets and Enhanced Visible-Light Photocatalytic Properties. Chem. Commun. 2010, 46, 1893−1895. (23) Wang, D. E.; Jiang, H. F.; Zong, X.; Xu, Q. A.; Ma, Y.; Li, G. L.; Li, C. Crystal Facet Dependence of Water Oxidation on BiVO4 Sheets under Visible Light Irradiation. Chem. - Eur. J. 2011, 17, 1275−1282. (24) Li, R. G.; Zhang, F. X.; Wang, D. E.; Yang, J. X.; Li, M. R.; Zhu, J.; Zhou, X.; Han, H. X.; Li, C. Spatial Separation of Photogenerated Electrons and Holes among {010} and {110} Crystal Facets of BiVO4. Nat. Commun. 2013, 4, 1432. (25) Dempsey, J. L.; Brunschwig, B. S.; Winkler, J. R.; Gray, H. B. Hydrogen Evolution Catalyzed by Cobaloximes. Acc. Chem. Res. 2009, 42, 1995−2004. (26) Wang, M.; Na, Y.; Gorlov, M.; Sun, L. C. Light-Driven Hydrogen Production Catalysed by Transition Metal Complexes in Homogeneous Systems. Dalton Trans. 2009, 6458−6467. (27) Lazarides, T.; McCormick, T.; Du, P. W.; Luo, G. G.; Lindley, B.; Eisenberg, R. Making Hydrogen from Water Using a Homogeneous System without Noble Metals. J. Am. Chem. Soc. 2009, 131, 9192−9194. (28) Zhang, W.; Hong, J. D.; Zheng, J. W.; Huang, Z. Y.; Zhou, J. R.; Xu, R. Nickel-Thiolate Complex Catalyst Assembled in One Step in Water for Solar H2 Production. J. Am. Chem. Soc. 2011, 133, 20680− 20683. (29) Tran, P. D.; Wong, L. H.; Barber, J.; Loo, J. S. C. Recent Advances in Hybrid Photocatalysts for Solar Fuel Production. Energy Environ. Sci. 2012, 5, 5902−5918. (30) Han, Z. J.; Qiu, F.; Eisenberg, R.; Holland, P. L.; Krauss, T. D. Robust Photogeneration of H2 in Water Using Semiconductor Nanocrystals and a Nickel Catalyst. Science 2012, 338, 1321−1324. (31) Zhou, T. H.; Wang, D. P.; Goh, S. C.-K.; Hong, J. D.; Han, J. Y.; Mao, J. G.; Xu, R. Bio-Inspired Organic Cobalt(II) Phosphonates toward Water Oxidation. Energy Environ. Sci. 2015, 8, 526−534. (32) Zhang, W.; Xu, R. Hybrid Photocatalytic H2 Evolution Systems Containing Xanthene Dyes and Inorganic Nickel Based Catalysts. Int. J. Hydrogen Energy 2012, 37, 17899−17909. (33) Zhang, W.; Zhou, T. H.; Zheng, J. W.; Hong, J. D.; Pan, Y. X.; Xu, R. Water-Soluble MoS3 Nanoparticles for Photocatalytic H2 Evolution. ChemSusChem 2015, 8, 1464−1471. (34) Osterloh, F. E. Inorganic Nanostructures for Photoelectrochemical and Photocatalytic Water Splitting. Chem. Soc. Rev. 2013, 42, 2294−2320. (35) Chen, H. M.; Chen, C. K.; Liu, R. S.; Zhang, L.; Zhang, J. J.; Wilkinson, D. P. Nano-Architecture and Material Designs for Water Splitting Photoelectrodes. Chem. Soc. Rev. 2012, 41, 5654−5671. (36) Li, Z. S.; Luo, W. J.; Zhang, M. L.; Feng, J. Y.; Zou, Z. G. Photoelectrochemical Cells for Solar Hydrogen Production: Current State of Promising Photoelectrodes, Methods to Improve Their Properties, and Outlook. Energy Environ. Sci. 2013, 6, 347−370. (37) Sivula, K.; Le Formal, F.; Grätzel, M. Solar Water Splitting: Progress Using Hematite (α-Fe2O3) Photoelectrodes. ChemSusChem 2011, 4, 432−449. (38) Swierk, J. R.; Mallouk, T. E. Design and Development of Photoanodes for Water-Splitting Dye-Sensitized Photoelectrochemical Cells. Chem. Soc. Rev. 2013, 42, 2357−2387. (39) Khaselev, O.; Turner, J. A. A Monolithic PhotovoltaicPhotoelectrochemical Device for Hydrogen Production via Water Splitting. Science 1998, 280, 425−427. (40) Reece, S. Y.; Hamel, J. A.; Sung, K.; Jarvi, T. D.; Esswein, A. J.; Pijpers, J. J. H.; Nocera, D. G. Wireless Solar Water Splitting Using Silicon-Based Semiconductors and Earth-Abundant Catalysts. Science 2011, 334, 645−648. (41) Jacobsson, T. J.; Fjällström, V.; Sahlberg, M.; Edoff, M.; Edvinsson, T. A Monolithic Device for Solar Water Splitting Based on

5,10,15-tris(pentaﬂuorophenyl)corrole tris(2-aminoethyl)amine tetrathiafulvalene versus X-ray absorption near edge structure X-ray photoelectron spectroscopy

REFERENCES (1) Blankenship, R. E.; Tiede, D. M.; Barber, J.; Brudvig, G. W.; Fleming, G.; Ghirardi, M.; Gunner, M. R.; Junge, W.; Kramer, D. M.; Melis, A.; et al. Comparing Photosynthetic and Photovoltaic Efficiencies and Recognizing the Potential for Improvement. Science 2011, 332, 805−809. (2) Gust, D.; Moore, T. A.; Moore, A. L. Solar Fuels via Artificial Photosynthesis. Acc. Chem. Res. 2009, 42, 1890−1898. (3) Balzani, V.; Credi, A.; Venturi, M. Photochemical Conversion of Solar Energy. ChemSusChem 2008, 1, 26−58. (4) Barber, J. Photosynthetic Energy Conversion: Natural and Artificial. Chem. Soc. Rev. 2009, 38, 185−196. (5) Walter, M. G.; Warren, E. L.; McKone, J. R.; Boettcher, S. W.; Mi, Q. X.; Santori, E. A.; Lewis, N. S. Solar Water Splitting Cells. Chem. Rev. 2010, 110, 6446−6473. (6) Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238, 37−38. (7) Osterloh, F. E. Inorganic Materials as Catalysts for Photochemical Splitting of Water. Chem. Mater. 2008, 20, 35−54. (8) Maeda, K.; Domen, K. Photocatalytic Water Splitting: Recent Progress and Future Challenges. J. Phys. Chem. Lett. 2010, 1, 2655− 2661. (9) Kudo, A.; Miseki, Y. Heterogeneous Photocatalyst Materials for Water Splitting. Chem. Soc. Rev. 2009, 38, 253−278. (10) Hisatomi, T.; Kubota, J.; Domen, K. Recent Advances in Semiconductors for Photocatalytic and Photoelectrochemical Water Splitting. Chem. Soc. Rev. 2014, 43, 7520−7535. (11) Jang, J. S.; Kim, H. G.; Lee, J. S. Heterojunction Semiconductors: A Strategy to Develop Efficient Photocatalytic Materials for Visible Light Water Splitting. Catal. Today 2012, 185, 270−277. (12) Wang, X.; Xu, Q.; Li, M. R.; Shen, S.; Wang, X. L.; Wang, Y. C.; Feng, Z. C.; Shi, J. Y.; Han, H. X.; Li, C. Photocatalytic Overall Water Splitting Promoted by an α-β Phase Junction on Ga2O3. Angew. Chem., Int. Ed. 2012, 51, 13089−13092. (13) Moniz, S. J. A.; Shevlin, S. A.; Martin, D. J.; Guo, Z. X.; Tang, J. W. Visible-Light Driven Heterojunction Photocatalysts for Water Splitting - a Critical Review. Energy Environ. Sci. 2015, 8, 731−759. (14) Maeda, K. Z-Scheme Water Splitting Using Two Different Semiconductor Photocatalysts. ACS Catal. 2013, 3, 1486−1503. (15) Zhou, P.; Yu, J. G.; Jaroniec, M. All-Solid-State Z-Scheme Photocatalytic Systems. Adv. Mater. 2014, 26, 4920−4935. (16) Maeda, K.; Higashi, M.; Lu, D. L.; Abe, R.; Domen, K. Efficient Nonsacrificial Water Splitting through Two-Step Photoexcitation by Visible Light Using a Modified Oxynitride as a Hydrogen Evolution Photocatalyst. J. Am. Chem. Soc. 2010, 132, 5858−5868. (17) Iwase, A.; Ng, Y. H.; Ishiguro, Y.; Kudo, A.; Amal, R. Reduced Graphene Oxide as a Solid-State Electron Mediator in Z-Scheme Photocatalytic Water Splitting under Visible Light. J. Am. Chem. Soc. 2011, 133, 11054−11057. (18) Zong, X.; Yan, H. J.; Wu, G. P.; Ma, G. J.; Wen, F. Y.; Wang, L.; Li, C. Enhancement of Photocatalytic H2 Evolution on CdS by Loading MoS2 as Cocatalyst under Visible Light Irradiation. J. Am. Chem. Soc. 2008, 130, 7176−7177. (19) Yan, H. J.; Yang, J. H.; Ma, G. J.; Wu, G. P.; Zong, X.; Lei, Z. B.; Shi, J. Y.; Li, C. Visible-Light-Driven Hydrogen Production with Extremely High Quantum Efficiency on Pt-PdS/CdS Photocatalyst. J. Catal. 2009, 266, 165−168. (20) Zhang, W.; Wang, Y. B.; Wang, Z.; Zhong, Z. Y.; Xu, R. Highly Efficient and Noble Metal-Free NiS/CdS Photocatalysts for H2 Evolution from Lactic Acid Sacrificial Solution under Visible Light. Chem. Commun. 2010, 46, 7631−7633. 3786

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Series Interconnected Thin Film Absorbers Reaching over 10% Solarto-Hydrogen Efficiency. Energy Environ. Sci. 2013, 6, 3676−3683. (42) Hu, S.; Xiang, C. X.; Haussener, S.; Berger, A. D.; Lewis, N. S. An Analysis of the Optimal Band Gaps of Light Absorbers in Integrated Tandem Photoelectrochemical Water-Splitting Systems. Energy Environ. Sci. 2013, 6, 2984−2993. (43) Haussener, S.; Xiang, C. X.; Spurgeon, J. M.; Ardo, S.; Lewis, N. S.; Weber, A. Z. Modeling, Simulation, and Design Criteria for Photoelectrochemical Water-Splitting Systems. Energy Environ. Sci. 2012, 5, 9922−9935. (44) Brinner, A.; Bussmann, H.; Hug, W.; Seeger, W. Test Results of the Hysolar 10 kW PV-Electrolysis Facility. Int. J. Hydrogen Energy 1992, 17, 187−197. (45) Gibson, T. L.; Kelly, N. A. Optimization of Solar Powered Hydrogen Production Using Photovoltaic Electrolysis Devices. Int. J. Hydrogen Energy 2008, 33, 5931−5940. (46) Khaselev, O.; Bansal, A.; Turner, J. A. High-Efficiency Integrated Multijunction Photovoltaic/Electrolysis Systems for Hydrogen Production. Int. J. Hydrogen Energy 2001, 26, 127−132. (47) Arriaga, L. G.; Martínez, W.; Cano, U.; Blud, H. Direct Coupling of a Solar-Hydrogen System in Mexico. Int. J. Hydrogen Energy 2007, 32, 2247−2252. (48) Peharz, G.; Dimroth, F.; Wittstadt, U. Solar Hydrogen Production by Water Splitting with a Conversion Efficiency of 18%. Int. J. Hydrogen Energy 2007, 32, 3248−3252. (49) Luo, J. S.; Im, J. H.; Mayer, M. T.; Schreier, M.; Nazeeruddin, M. K.; Park, N. G.; Tilley, S. D.; Fan, H. J.; Grätzel, M. Water Photolysis at 12.3% Efficiency via Perovskite Photovoltaics and EarthAbundant Catalysts. Science 2014, 345, 1593−1596. (50) McKone, J. R.; Lewis, N. S.; Gray, H. B. Will Solar-Driven Water-Splitting Devices See the Light of Day? Chem. Mater. 2014, 26, 407−414. (51) Shaner, M. R.; Atwater, H. A.; Lewis, N. S.; McFarland, E. W. A Comparative Technoeconomic Analysis of Renewable Hydrogen Production Using Solar Energy. Energy Environ. Sci. 2016, 9, 2354− 2371. (52) Yang, J. H.; Wang, D. E.; Han, H. X.; Li, C. Roles of Cocatalysts in Photocatalysis and Photoelectrocatalysis. Acc. Chem. Res. 2013, 46, 1900−1909. (53) Jiao, Y.; Zheng, Y.; Jaroniec, M. T.; Qiao, S. Z. Design of Electrocatalysts for Oxygen- and Hydrogen-Involving Energy Conversion Reactions. Chem. Soc. Rev. 2015, 44, 2060−2086. (54) Artero, V.; Chavarot-Kerlidou, M.; Fontecave, M. Splitting Water with Cobalt. Angew. Chem., Int. Ed. 2011, 50, 7238−7266. (55) McCrory, C. C. L.; Jung, S.; Ferrer, I. M.; Chatman, S. M.; Peters, J. C.; Jaramillo, T. F. Benchmarking Hydrogen Evolving Reaction and Oxygen Evolving Reaction Electrocatalysts for Solar Water Splitting Devices. J. Am. Chem. Soc. 2015, 137, 4347−4357. (56) McCrory, C. C. L.; Jung, S.; Peters, J. C.; Jaramillo, T. F. Benchmarking Heterogeneous Electrocatalysts for the Oxygen Evolution Reaction. J. Am. Chem. Soc. 2013, 135, 16977−16987. (57) Costentin, C.; Drouet, S.; Robert, M.; Savéant, J. M. Turnover Numbers, Turnover Frequencies, and Overpotential in Molecular Catalysis of Electrochemical Reactions. Cyclic Voltammetry and Preparative-Scale Electrolysis. J. Am. Chem. Soc. 2012, 134, 11235− 11242. (58) Costentin, C.; Passard, G.; Savéant, J. M. Benchmarking of Homogeneous Electrocatalysts: Overpotential, Turnover Frequency, Limiting Turnover Number. J. Am. Chem. Soc. 2015, 137, 5461−5467. (59) Morales-Guio, C. G.; Stern, L. A.; Hu, X. L. Nanostructured Hydrotreating Catalysts for Electrochemical Hydrogen Evolution. Chem. Soc. Rev. 2014, 43, 6555−6569. (60) Cao, R.; Lai, W. Z.; Du, P. W. Catalytic Water Oxidation at Single Metal Sites. Energy Environ. Sci. 2012, 5, 8134−8157. (61) Wu, Y. Z.; Chen, M. X.; Han, Y. Z.; Luo, H. X.; Su, X. J.; Zhang, M.-T.; Lin, X. H.; Sun, J. L.; Wang, L.; Deng, L.; et al. Fast and Simple Preparation of Iron-Based Thin Films as Highly Efficient WaterOxidation Catalysts in Neutral Aqueous Solution. Angew. Chem., Int. Ed. 2015, 54, 4870−4875.

(62) Chen, M. X.; Wu, Y. Z.; Han, Y. Z.; Lin, X. H.; Sun, J. L.; Zhang, W.; Cao, R. An Iron-Based Film for Highly Efficient Electrocatalytic Oxygen Evolution from Neutral Aqueous Solution. ACS Appl. Mater. Interfaces 2015, 7, 21852−21859. (63) Qi, J.; Zhang, W.; Xiang, R. J.; Liu, K. Q.; Wang, H. Y.; Chen, M. X.; Han, Y. Z.; Cao, R. Porous Nickel-Iron Oxide as a Highly Efficient Electrocatalyst for Oxygen Evolution Reaction. Adv. Sci. 2015, 2, 1500199. (64) Zhang, W.; Qi, J.; Liu, K. Q.; Cao, R. A Nickel-Based Integrated Electrode from an Autologous Growth Strategy for Highly Efficient Water Oxidation. Adv. Energy Mater. 2016, 6, 1502489. (65) Kärkäs, M. D.; Verho, O.; Johnston, E. V.; Åkermark, B. Artificial Photosynthesis: Molecular Systems for Catalytic Water Oxidation. Chem. Rev. 2014, 114, 11863−12001. (66) Singh, A.; Spiccia, L. Water Oxidation Catalysts Based on Abundant 1st Row Transition Metals. Coord. Chem. Rev. 2013, 257, 2607−2622. (67) Dau, H.; Limberg, C.; Reier, T.; Risch, M.; Roggan, S.; Strasser, P. The Mechanism of Water Oxidation: From Electrolysis via Homogeneous to Biological Catalysis. ChemCatChem 2010, 2, 724− 761. (68) Nie, Y.; Li, L.; Wei, Z. D. Recent Advancements in Pt and PtFree Catalysts for Oxygen Reduction Reaction. Chem. Soc. Rev. 2015, 44, 2168−2201. (69) Zhu, C. Z.; Li, H.; Fu, S. F.; Du, D.; Lin, Y. H. Highly Efficient Nonprecious Metal Catalysts Towards Oxygen Reduction Reaction Based on Three-Dimensional Porous Carbon Nanostructures. Chem. Soc. Rev. 2016, 45, 517−531. (70) Jaouen, F.; Proietti, E.; Lefèvre, M.; Chenitz, R.; Dodelet, J. P.; Wu, G.; Chung, H. T.; Johnston, C. M.; Zelenay, P. Recent Advances in Non-Precious Metal Catalysis for Oxygen-Reduction Reaction in Polymer Electrolyte Fuel Cells. Energy Environ. Sci. 2011, 4, 114−130. (71) Wu, G.; Zelenay, P. Nanostructured Nonprecious Metal Catalysts for Oxygen Reduction Reaction. Acc. Chem. Res. 2013, 46, 1878−1889. (72) Guo, S. J.; Zhang, S.; Sun, S. H. Tuning Nanoparticle Catalysis for the Oxygen Reduction Reaction. Angew. Chem., Int. Ed. 2013, 52, 8526−8544. (73) Dai, L. M.; Xue, Y. H.; Qu, L. T.; Choi, H. J.; Baek, J. B. MetalFree Catalysts for Oxygen Reduction Reaction. Chem. Rev. 2015, 115, 4823−4892. (74) Gasteiger, H. A.; Kocha, S. S.; Sompalli, B.; Wagner, F. T. Activity Benchmarks and Requirements for Pt, Pt-Alloy, and Non-Pt Oxygen Reduction Catalysts for PEMFCs. Appl. Catal., B 2005, 56, 9− 35. (75) Collman, J. P.; Boulatov, R.; Sunderland, C. J.; Fu, L. Functional Analogues of Cytochrome c Oxidase, Myoglobin, and Hemoglobin. Chem. Rev. 2004, 104, 561−588. (76) Yoshikawa, S.; Shimada, A. Reaction Mechanism of Cytochrome c Oxidase. Chem. Rev. 2015, 115, 1936−1989. (77) Hematian, S.; Garcia-Bosch, I.; Karlin, K. D. Synthetic Heme/ Copper Assemblies: Toward an Understanding of Cytochrome c Oxidase Interactions with Dioxygen and Nitrogen Oxides. Acc. Chem. Res. 2015, 48, 2462−2474. (78) Chufán, E. E.; Puiu, S. C.; Karlin, K. D. Heme-Copper/ Dioxygen Adduct Formation, Properties, and Reactivity. Acc. Chem. Res. 2007, 40, 563−572. (79) Choi, H. J.; Kumar, N. A.; Baek, J. B. Graphene Supported NonPrecious Metal-Macrocycle Catalysts for Oxygen Reduction Reaction in Fuel Cells. Nanoscale 2015, 7, 6991−6998. (80) Denisov, I. G.; Makris, T. M.; Sligar, S. G.; Schlichting, I. Structure and Chemistry of Cytochrome P450. Chem. Rev. 2005, 105, 2253−2277. (81) Meunier, B.; de Visser, S. P.; Shaik, S. Mechanism of Oxidation Reactions Catalyzed by Cytochrome P450 Enzymes. Chem. Rev. 2004, 104, 3947−3980. (82) Shaik, S.; Kumar, D.; de Visser, S. P.; Altun, A.; Thiel, W. Theoretical Perspective on the Structure and Mechanism of Cytochrome P450 Enzymes. Chem. Rev. 2005, 105, 2279−2328. 3787

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

(83) Green, M. T.; Dawson, J. H.; Gray, H. B. Oxoiron(IV) in Chloroperoxidase Compound II Is Basic: Implications for P450 Chemistry. Science 2004, 304, 1653−1656. (84) Ferreira, K. N.; Iverson, T. M.; Maghlaoui, K.; Barber, J.; Iwata, S. Architecture of the Photosynthetic Oxygen-Evolving Center. Science 2004, 303, 1831−1838. (85) Umena, Y.; Kawakami, K.; Shen, J. R.; Kamiya, N. Crystal Structure of Oxygen-Evolving Photosystem II at a Resolution of 1.9 Å. Nature 2011, 473, 55−60. (86) Suga, M.; Akita, F.; Hirata, K.; Ueno, G.; Murakami, H.; Nakajima, Y.; Shimizu, T.; Yamashita, K.; Yamamoto, M.; Ago, H.; et al. Native Structure of Photosystem II at 1.95 Å Resolution Viewed by Femtosecond X-ray Pulses. Nature 2014, 517, 99−103. (87) Loll, B.; Kern, J.; Saenger, W.; Zouni, A.; Biesiadka, J. Towards Complete Cofactor Arrangement in the 3.0 Å Resolution Structure of Photosystem II. Nature 2005, 438, 1040−1044. (88) Shaik, S.; Hirao, H.; Kumar, D. Reactivity of High-Valent IronOxo Species in Enzymes and Synthetic Reagents: A Tale of Many States. Acc. Chem. Res. 2007, 40, 532−542. (89) Nam, W. High-Valent Iron(IV)-Oxo Complexes of Heme and Non-Heme Ligands in Oxygenation Reactions. Acc. Chem. Res. 2007, 40, 522−531. (90) Jin, N.; Ibrahim, M.; Spiro, T. G.; Groves, J. T. Trans-Dioxo Manganese(V) Porphyrins. J. Am. Chem. Soc. 2007, 129, 12416− 12417. (91) Gross, Z. The Groves-Spiro Dioxomanganese(V) Story. Angew. Chem., Int. Ed. 2008, 47, 2737−2739. (92) Song, W. J.; Seo, M. S.; George, S. D.; Ohta, T.; Song, R.; Kang, M. J.; Tosha, T.; Kitagawa, T.; Solomon, E. I.; Nam, W. Synthesis, Characterization, and Reactivities of Manganese(V)-Oxo Porphyrin Complexes. J. Am. Chem. Soc. 2007, 129, 1268−1277. (93) Kang, Y.; Chen, H.; Jeong, Y. J.; Lai, W. Z.; Bae, E. H.; Shaik, S.; Nam, W. Enhanced Reactivities of Iron(IV)-Oxo Porphyrin π-Cation Radicals in Oxygenation Reactions by Electron-Donating Axial Ligands. Chem. - Eur. J. 2009, 15, 10039−10046. (94) Takahashi, A.; Kurahashi, T.; Fujii, H. Effect of Imidazole and Phenolate Axial Ligands on the Electronic Structure and Reactivity of Oxoiron(IV) Porphyrin π-Cation Radical Complexes: Drastic Increase in Oxo-Transfer and Hydrogen Abstraction Reactivities. Inorg. Chem. 2009, 48, 2614−2625. (95) Abu-Omar, M. M. High-Valent Iron and Manganese Complexes of Corrole and Porphyrin in Atom Transfer and Dioxygen Evolving Catalysis. Dalton Trans. 2011, 40, 3435−3444. (96) Thomas, K. E.; Alemayehu, A. B.; Conradie, J.; Beavers, C. M.; Ghosh, A. The Structural Chemistry of Metallocorroles: Combined Xray Crystallography and Quantum Chemistry Studies Afford Unique Insights. Acc. Chem. Res. 2012, 45, 1203−1214. (97) Johnson, A. W.; Kay, I. T. The Pentadehydrocorrin (Corrole) Ring System. Proc. Chem. Soc. 1964, 89−90. (98) Gross, Z.; Galili, N.; Saltsman, I. The First Direct Synthesis of Corroles from Pyrrole. Angew. Chem., Int. Ed. 1999, 38, 1427−1429. (99) Gross, Z.; Galili, N.; Simkhovich, L.; Saltsman, I.; Botoshansky, M.; Bläser, D.; Boese, R.; Goldberg, I. Solvent-Free Condensation of Pyrrole and Pentafluorobenzaldehyde: A Novel Synthetic Pathway to Corrole and Oligopyrromethenes. Org. Lett. 1999, 1, 599−602. (100) Aviv, I.; Gross, Z. Corrole-Based Applications. Chem. Commun. 2007, 1987−1999. (101) Flamigni, L.; Gryko, D. T. Photoactive Corrole-Based Arrays. Chem. Soc. Rev. 2009, 38, 1635−1646. (102) Haber, A.; Gross, Z. Catalytic Antioxidant Therapy by Metallodrugs: Lessons from Metallocorroles. Chem. Commun. 2015, 51, 5812−5827. (103) Grodkowski, J.; Neta, P.; Fujita, E.; Mahammed, A.; Simkhovich, L.; Gross, Z. Reduction of Cobalt and Iron Corroles and Catalyzed Reduction of CO2. J. Phys. Chem. A 2002, 106, 4772− 4778. (104) Simkhovich, L.; Mahammed, A.; Goldberg, I.; Gross, Z. Synthesis and Characterization of Germanium, Tin, Phosphorus, Iron, and Rhodium Complexes of Tris(Pentafluorophenyl)Corrole, and the

Utilization of the Iron and Rhodium Corroles as Cyclopropanation Catalysts. Chem. - Eur. J. 2001, 7, 1041−1055. (105) Simkhovich, L.; Galili, N.; Saltsman, I.; Goldberg, I.; Gross, Z. Coordination Chemistry of the Novel 5,10,15-Tris(Pentafluorophenyl)Corrole: Synthesis, Spectroscopy, and Structural Characterization of Its Cobalt(III), Rhodium(III), and Iron(IV) Complexes. Inorg. Chem. 2000, 39, 2704−2705. (106) Simkhovich, L.; Gross, Z. Halogeno-Coordinated Iron Corroles. Inorg. Chem. 2004, 43, 6136−6138. (107) Bendix, J.; Dmochowski, I. J.; Gray, H. B.; Mahammed, A.; Simkhovich, L.; Gross, Z. Structural, Electrochemical, and Photophysical Properties of Gallium(III) 5,10,15-Tris(Pentafluorophenyl)Corrole. Angew. Chem., Int. Ed. 2000, 39, 4048−4051. (108) Bendix, J.; Gray, H. B.; Golubkov, G.; Gross, Z. High-Field (High-Frequency) EPR Spectroscopy and Structural Characterization of a Novel Manganese(III) Corrole. Chem. Commun. 2000, 1957− 1958. (109) Simkhovich, L.; Luobeznova, I.; Goldberg, I.; Gross, Z. Monoand Binuclear Ruthenium Corroles: Synthesis, Spectroscopy, Electrochemistry, and Structural Characterization. Chem. - Eur. J. 2003, 9, 201−208. (110) Luobeznova, I.; Raizman, M.; Goldberg, I.; Gross, Z. Synthesis and Full Characterization of Molybdenum and Antimony Corroles and Utilization of the Latter Complexes as Very Efficient Catalysts for Highly Selective Aerobic Oxygenation Reactions. Inorg. Chem. 2006, 45, 386−394. (111) Aviv, I.; Gross, Z. Iron Porphyrins Catalyze the Synthesis of Non-Protected Amino Acid Esters from Ammonia and Diazoacetates. Chem. Commun. 2006, 4477−4479. (112) Okun, Z.; Gross, Z. Fine Tuning the Reactivity of CorroleBased Catalytic Antioxidants. Inorg. Chem. 2012, 51, 8083−8090. (113) Haber, A.; Aviram, M.; Gross, Z. Variables That Influence Cellular Uptake and Cytotoxic/Cytoprotective Effects of Macrocyclic Iron Complexes. Inorg. Chem. 2012, 51, 28−30. (114) Haber, A.; Ali, A. A.-Y.; Aviram, M.; Gross, Z. Allosteric Inhibitors of HMG-CoA Reductase, the Key Enzyme Involved in Cholesterol Biosynthesis. Chem. Commun. 2013, 49, 10917−10919. (115) Gross, Z.; Gray, H. B. How Do Corroles Stabilize High Velent Metals. Comments Inorg. Chem. 2006, 27, 61−72. (116) Simkhovich, L.; Goldberg, I.; Gross, Z. Iron(III) and Iron(IV) Corroles: Synthesis, Spectroscopy, Structures, and No Indications for Corrole Radicals. Inorg. Chem. 2002, 41, 5433−5439. (117) Golubkov, G.; Bendix, J.; Gray, H. B.; Mahammed, A.; Goldberg, I.; DiBilio, A. J.; Gross, Z. High-Valent Manganese Corroles and the First Perhalogenated Metallocorrole Catalyst. Angew. Chem., Int. Ed. 2001, 40, 2132−2134. (118) Liu, S.; Jung, J.; Ohkubo, K.; Hicks, S. D.; Bougher, C. J.; AbuOmar, M. M.; Fukuzumi, S. Activationless Electron Self-Exchange of High-Valent Oxo and Imido Complexes of Chromium Corroles. Inorg. Chem. 2015, 54, 9223−9228. (119) Bougher, C. J.; Liu, S.; Hicks, S. D.; Abu-Omar, M. M. Valence Tautomerization of High-Valent Manganese(V)-Oxo Corrole Induced by Protonation of the Oxo Ligand. J. Am. Chem. Soc. 2015, 137, 14481−14487. (120) Kim, S. H.; Park, H.; Seo, M. S.; Kubo, M.; Ogura, T.; Klajn, J.; Gryko, D. T.; Valentine, J. S.; Nam, W. Reversible O-O Bond Cleavage and Formation between Mn(IV)-Peroxo and Mn(V)-Oxo Corroles. J. Am. Chem. Soc. 2010, 132, 14030−14032. (121) Zhao, H. L.; Pierloot, K.; Langner, E. H. G.; Swarts, J. C.; Conradie, J.; Ghosh, A. Low-Energy States of Manganese-Oxo Corrole and Corrolazine: Multiconfiguration Reference ab Initio Calculations. Inorg. Chem. 2012, 51, 4002−4006. (122) Latifi, R.; Valentine, J. S.; Nam, W.; de Visser, S. P. Predictive Studies of H-Atom Abstraction Reactions by an Iron(IV)-Oxo Corrole Cation Radical Oxidant. Chem. Commun. 2012, 48, 3491−3493. (123) Gross, Z.; Golubkov, G.; Simkhovich, L. Epoxidation Catalysis by a Manganese Corrole and Isolation of an Oxomanganese(V) Corrole. Angew. Chem., Int. Ed. 2000, 39, 4045−4047. 3788

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

(124) Kumar, A.; Goldberg, I.; Botoshansky, M.; Buchman, Y.; Gross, Z. Oxygen Atom Transfer Teactions from Isolated (Oxo)Manganese(V) Corroles to Sulfides. J. Am. Chem. Soc. 2010, 132, 15233−15245. (125) Liu, H. Y.; Yam, F.; Xie, Y. T.; Li, X. Y.; Chang, C. K. A Bulky Bis-Pocket Manganese(V)-Oxo Corrole Complex: Observation of Oxygen Atom Transfer between Triply Bonded MnVO and Alkene. J. Am. Chem. Soc. 2009, 131, 12890−12891. (126) Czernuszewicz, R. S.; Mody, V.; Czader, A.; Gałęzowski, M.; Gryko, D. T. Why the Chromyl Bond Is Stronger Than the Perchromyl Bond in High-Valent Oxochromium(IV,V) Complexes of Tris(Pentafluorophenyl)Corrole. J. Am. Chem. Soc. 2009, 131, 14214−14215. (127) Mahammed, A.; Gray, H. B.; Meier-Callahan, A. E.; Gross, Z. Aerobic Oxidations Catalyzed by Chromium Corroles. J. Am. Chem. Soc. 2003, 125, 1162−1163. (128) Mahammed, A.; Gross, Z. Albumin-Conjugated Corrole Metal Complexes: Extremely Simple yet Very Efficient Biomimetic Oxidation Systems. J. Am. Chem. Soc. 2005, 127, 2883−2887. (129) Meier-Callahan, A. E.; Gray, H. B.; Gross, Z. Stabilization of High-Valent Metals by Corroles: Oxo[Tris(Pentafluorophenyl)Corrolato]Chromium(V). Inorg. Chem. 2000, 39, 3605−3607. (130) Meier-Callahan, A. E.; Di Bilio, A. J.; Simkhovich, L.; Mahammed, A.; Goldberg, I.; Gray, H. B.; Gross, Z. Chromium Corroles in Four Oxidation States. Inorg. Chem. 2001, 40, 6788−6793. (131) Nigel-Etinger, I.; Goldberg, I.; Gross, Z. Intriguing Chemistry of Molybdenum Corroles. Inorg. Chem. 2013, 52, 4139−4141. (132) Golubkov, G.; Gross, Z. Chromium(V) and Chromium(VI) Nitrido Complexes of Tris(Pentafluorophenyl)Corrole. Angew. Chem., Int. Ed. 2003, 42, 4507−4510. (133) Golubkov, G.; Gross, Z. Nitrogen Atom Transfer between Manganese Complexes of Salen, Porphyrin, and Corrole and Characterization of a (Nitrido)Manganese(VI) Corrole. J. Am. Chem. Soc. 2005, 127, 3258−3259. (134) Kim, E.; Chufán, E. E.; Kamaraj, K.; Karlin, K. D. Synthetic Models for Heme-Copper Oxidases. Chem. Rev. 2004, 104, 1077− 1133. (135) Collman, J. P.; Wagenknecht, P. S.; Hutchison, J. E. Molecular Catalysts for Multielectron Redox Reactions of Small Molecules: The ″Cofacial Metallodiporphyrin″ Approach. Angew. Chem., Int. Ed. Engl. 1994, 33, 1537−1554. (136) Chen, Z. W.; Higgins, D.; Yu, A. P.; Zhang, L.; Zhang, J. J. A Review on Non-Precious Metal Electrocatalysts for PEM Fuel Cells. Energy Environ. Sci. 2011, 4, 3167−3192. (137) Fukuzumi, S.; Yamada, Y.; Karlin, K. D. Hydrogen Peroxide as a Sustainable Energy Carrier: Electrocatalytic Production of Hydrogen Peroxide and the Fuel Cell. Electrochim. Acta 2012, 82, 493−511. (138) Yamada, Y.; Yoshida, S.; Honda, T.; Fukuzumi, S. Protonated Iron-Phthalocyanine Complex Used for Cathode Material of a Hydrogen Peroxide Fuel Cell Operated under Acidic Conditions. Energy Environ. Sci. 2011, 4, 2822−2825. (139) Solomon, E. I.; Heppner, D. E.; Johnston, E. M.; Ginsbach, J. W.; Cirera, J.; Qayyum, M.; Kieber-Emmons, M. T.; Kjaergaard, C. H.; Hadt, R. G.; Tian, L. Copper Active Sites in Biology. Chem. Rev. 2014, 114, 3659−3853. (140) Kaila, V. R. I.; Verkhovsky, M. I.; Wikström, M. ProtonCoupled Electron Transfer in Cytochrome Oxidase. Chem. Rev. 2010, 110, 7062−7081. (141) Iwata, S.; Ostermeier, C.; Ludwig, B.; Michel, H. Structure at 2.8 Å Resolution of Cytochrome c Oxidase from Paracoccus Denitrificans. Nature 1995, 376, 660−669. (142) Tsukihara, T.; Aoyama, H.; Yamashita, E.; Tomizaki, T.; Yamaguchi, H.; Shinzawa-Itoh, K.; Nakashima, R.; Yaono, R.; Yoshikawa, S. Structures of Metal Sites of Oxidized Bovine Heart Cytochrome c Oxidase at 2.8 Å. Science 1995, 269, 1069−1074. (143) Tsukihara, T.; Aoyama, H.; Yamashita, E.; Tomizaki, T.; Yamaguchi, H.; Shinzawa-Itoh, K.; Nakashima, R.; Yaono, R.; Yoshikawa, S. The Whole Structure of the 13-Subunit Oxidized Cytochrome c Oxidase at 2.8 Å. Science 1996, 272, 1136−1144.

(144) Yoshikawa, S.; Shinzawa-Itoh, K.; Nakashima, R.; Yaono, R.; Yamashita, E.; Inoue, N.; Yao, M.; Fei, M. J.; Libeu, C. P.; Mizushima, T.; et al. Redox-Coupled Crystal Structural Changes in Bovine Heart Cytochrome c Oxidase. Science 1998, 280, 1723−1729. (145) Ostermeier, C.; Harrenga, A.; Ermler, U.; Michel, H. Structure at 2.7 Å Resolution of the Paracoccus denitrif icans Two-Subunit Cytochrome c Oxidase Complexed with an Antibody FV Fragment. Proc. Natl. Acad. Sci. U. S. A. 1997, 94, 10547−10553. (146) Harrenga, A.; Michel, H. The Cytochrome c Oxidase from Paracoccus denitrif icans Does Not Change the Metal Center Ligation Upon Reduction. J. Biol. Chem. 1999, 274, 33296−33299. (147) Proshlyakov, D. A.; Pressler, M. A.; Babcock, G. T. Dioxygen Activation and Bond Cleavage by Mixed-Valence Cytochrome c Oxidase. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 8020−8025. (148) Shin, H.; Lee, D. H.; Kang, C.; Karlin, K. D. Electrocatalytic Four-Electron Reductions of O2 to H2O with Cytochrome c Oxidase Model Compounds. Electrochim. Acta 2003, 48, 4077−4082. (149) Kopf, M. A.; Karlin, K. D. Dioxygen Reactivity of Reduced Heme and Heme-Copper Complexes Utilizing Tetraarylporphyrinates Tethered with Both a Pyridyl Axial Ligand and N,N-Bis[2-(2pyridyl)ethyl]amine Chelate. Inorg. Chem. 1999, 38, 4922−4923. (150) Ghiladi, R. A.; Ju, T. D.; Lee, D. H.; Moënne-Loccoz, P.; Kaderli, S.; Neuhold, Y. M.; Zuberbühler, A. D.; Woods, A. S.; Cotter, R. J.; Karlin, K. D. Formation and Characterization of a High-Spin Heme-Copper Dioxygen (Peroxo) Complex. J. Am. Chem. Soc. 1999, 121, 9885−9886. (151) Ghiladi, R. A.; Kretzer, R. M.; Guzei, I.; Rheingold, A. L.; Neuhold, Y. M.; Hatwell, K. R.; Zuberbühler, A. D.; Karlin, K. D. (F 8 TPP)Fe II /O 2 Reactivity Studies {F 8 TPP = Tetrakis(2,6Difluorophenyl)Porphyrinate(2−)}: Spectroscopic (UV-visible and NMR) and Kinetic Study of Solvent-Dependent (Fe/O2 = 1:1 or 2:1) Reversible O2-Reduction and Ferryl Formation. Inorg. Chem. 2001, 40, 5754−5767. (152) Ghiladi, R. A.; Hatwell, K. R.; Karlin, K. D.; Huang, H. W.; Moënne-Loccoz, P.; Krebs, C.; Huynh, B. H.; Marzilli, L. A.; Cotter, R. J.; Kaderli, S.; et al. Dioxygen Reactivity of Mononuclear Heme and Copper Components Yielding a High-Spin Heme-Peroxo-Cu Complex. J. Am. Chem. Soc. 2001, 123, 6183−6184. (153) Halime, Z.; Kotani, H.; Li, Y. Q.; Fukuzumi, S.; Karlin, K. D. Homogeneous Catalytic O2 Reduction to Water by a Cytochrome c Oxidase Model with Trapping of Intermediates and Mechanistic Insights. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 13990−13994. (154) Collman, J. P.; Fu, L.; Herrmann, P. C.; Zhang, X. M. A Functional Model Related to Cytochrome c Oxidase and Its Electrocatalytic Four-Electron Reduction of O2. Science 1997, 275, 949−951. (155) Chishiro, T.; Shimazaki, Y.; Tani, F.; Tachi, Y.; Naruta, Y.; Karasawa, S.; Hayami, S.; Maeda, Y. Isolation and Crystal Structure of a Peroxo-Bridged Heme-Copper Complex. Angew. Chem., Int. Ed. 2003, 42, 2788−2791. (156) Collman, J. P.; Fu, L.; Herrmann, P. C.; Wang, Z.; Rapta, M.; Bröring, M.; Schwenninger, R.; Boitrel, B. A Functional Model of Cytochrome c Oxidase: Thermodynamic Implications. Angew. Chem., Int. Ed. 1998, 37, 3397−3400. (157) Collman, J. P.; Rapta, M.; Brö ring, M.; Raptova, L.; Schwenninger, R.; Boitrel, B.; Fu, L.; L’Her, M. Close Structural Analogues of the Cytochrome c Oxidase Fea3/CuB Center Show Clean 4e− Electroreduction of O2 to H2O at Physiological pH. J. Am. Chem. Soc. 1999, 121, 1387−1388. (158) Boulatov, R.; Collman, J. P.; Shiryaeva, I. M.; Sunderland, C. J. Functional Analogues of the Dioxygen Reduction Site in Cytochrome Oxidase: Mechanistic Aspects and Possible Effects of CuB. J. Am. Chem. Soc. 2002, 124, 11923−11935. (159) Collman, J. P.; Devaraj, N. K.; Decréau, R. A.; Yang, Y.; Yan, Y. L.; Ebina, W.; Eberspacher, T. A.; Chidsey, C. E. D. A Cytochrome c Oxidase Model Catalyzes Oxygen to Water Reduction under RateLimiting Electron Flux. Science 2007, 315, 1565−1568. 3789

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

(160) Collman, J. P.; Ghosh, S.; Dey, A.; Decréau, R. A.; Yang, Y. Catalytic Reduction of O2 by Cytochrome c Using a Synthetic Model of Cytochrome c Oxidase. J. Am. Chem. Soc. 2009, 131, 5034−5035. (161) Collman, J. P.; Decréau, R. A.; Dey, A.; Yang, Y. Water May Inhibit Oxygen Binding in Hemoprotein Models. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 4101−4105. (162) Collman, J. P.; Decréau, R. A.; Lin, H. W.; Hosseini, A.; Yang, Y.; Dey, A.; Eberspacher, T. A. Role of a Distal Pocket in the Catalytic O2 Reduction by Cytochrome c Oxidase Models Immobilized on Interdigitated Array Electrodes. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 7320−7323. (163) Ricard, D.; Didier, A.; L’Her, M.; Boitrel, B. Application of 3Quinolinoyl Picket Porphyrins to the Electroreduction of Dioxygen to Water: Mimicking the Active Site of Cytochrome c Oxidase. ChemBioChem 2001, 2, 144−148. (164) Ricard, D.; L’Her, M.; Richard, P.; Boitrel, B. Iron Porphyrins as Models of Cytochrome c Oxidase. Chem. - Eur. J. 2001, 7, 3291− 3297. (165) Ricard, D.; Didier, A.; L’Her, M.; Boitrel, B. The Reduction of Molecular Oxygen by Iron Porphyrins. C. R. Chim. 2002, 5, 33−36. (166) Didier, A.; L’Her, M.; Boitrel, B. Substituted Tren-Capped Porphyrins: Probing the Influence of Copper in Synthetic Models of Cytochrome c Oxidase. Org. Biomol. Chem. 2003, 1, 1274−1276. (167) Samanta, S.; Das, P. K.; Chatterjee, S.; Sengupta, K.; Mondal, B.; Dey, A. O2 Reduction Reaction by Biologically Relevant Anionic Ligand Bound Iron Porphyrin Complexes. Inorg. Chem. 2013, 52, 12963−12971. (168) Chatterjee, S.; Sengupta, K.; Samanta, S.; Das, P. K.; Dey, A. Electrocatalytic O2 Reduction Reaction by Synthetic Analogues of Cytochrome P450 and Myoglobin: In-Situ Resonance Raman and Dynamic Electrochemistry Investigations. Inorg. Chem. 2013, 52, 9897−9907. (169) Chatterjee, S.; Sengupta, K.; Samanta, S.; Das, P. K.; Dey, A. Concerted Proton-Electron Transfer in Electrocatalytic O2 Reduction by Iron Porphyrin Complexes: Axial Ligands Tuning H/D Isotope Effect. Inorg. Chem. 2015, 54, 2383−2392. (170) Sengupta, K.; Chatterjee, S.; Samanta, S.; Bandyopadhyay, S.; Dey, A. Resonance Raman and Electrocatalytic Behavior of Thiolate and Imidazole Bound Iron Porphyrin Complexes on Self Assembled Monolayers: Functional Modeling of Cytochrome P450. Inorg. Chem. 2013, 52, 2000−2014. (171) Samanta, S.; Das, P. K.; Chatterjee, S.; Dey, A. Effect of Axial Ligands on Electronic Structure and O2 Reduction by Iron Porphyrin Complexes: Towards a Quantitative Understanding of The ″Push Effect″. J. Porphyrins Phthalocyanines 2015, 19, 92−108. (172) Chatterjee, S.; Sengupta, K.; Hematian, S.; Karlin, K. D.; Dey, A. Electrocatalytic O2-Reduction by Synthetic Cytochrome c Oxidase Mimics: Identification of A ″Bridging Peroxo″ Intermediate Involved in Facile 4e−/4H+ O2-Reduction. J. Am. Chem. Soc. 2015, 137, 12897− 12905. (173) Wasylenko, D. J.; Rodríguez, C.; Pegis, M. L.; Mayer, J. M. Direct Comparison of Electrochemical and Spectrochemical Kinetics for Catalytic Oxygen Reduction. J. Am. Chem. Soc. 2014, 136, 12544− 12547. (174) Matson, B. D.; Carver, C. T.; Von Ruden, A.; Yang, J. Y.; Raugei, S.; Mayer, J. M. Distant Protonated Pyridine Groups in WaterSoluble Iron Porphyrin Electrocatalysts Promote Selective Oxygen Reduction to Water. Chem. Commun. 2012, 48, 11100−11102. (175) Carver, C. T.; Matson, B. D.; Mayer, J. M. Electrocatalytic Oxygen Reduction by Iron Tetra-Arylporphyrins Bearing Pendant Proton Relays. J. Am. Chem. Soc. 2012, 134, 5444−5447. (176) Rigsby, M. L.; Wasylenko, D. J.; Pegis, M. L.; Mayer, J. M. Medium Effects Are as Important as Catalyst Design for Selectivity in Electrocatalytic Oxygen Reduction by Iron-Porphyrin Complexes. J. Am. Chem. Soc. 2015, 137, 4296−4299. (177) Savéant, J. M. Molecular Catalysis of Electrochemical Reactions. Mechanistic Aspects. Chem. Rev. 2008, 108, 2348−2378.

(178) Bettelheim, A.; Kuwana, T. Rotating-Ring-Disc Analysis of Iron Tetra(N-Methylpyridyl)Porphyrin in Electrocatalysis of Oxygen. Anal. Chem. 1979, 51, 2257−2260. (179) Rosenthal, J.; Nocera, D. G. Role of Proton-Coupled Electron Transfer in O-O Bond Activation. Acc. Chem. Res. 2007, 40, 543−553. (180) He, Q. G.; Mugadza, T.; Kang, X. W.; Zhu, X. B.; Chen, S. W.; Kerr, J.; Nyokong, T. Molecular Catalysis of the Oxygen Reduction Reaction by Iron Porphyrin Catalysts Tethered into Nafion Layers: An Electrochemical Study in Solution and a Membrane-ElectrodeAssembly Study in Fuel Cells. J. Power Sources 2012, 216, 67−75. (181) He, Q. G.; Wu, G.; Liu, K.; Khene, S.; Li, Q.; Mugadza, T.; Deunf, E.; Nyokong, T.; Chen, S. W. W. Effects of Redox Mediators on the Catalytic Activity of Iron Porphyrins Towards Oxygen Reduction in Acidic Media. ChemElectroChem 2014, 1, 1508−1515. (182) Costentin, C.; Dridi, H.; Savéant, J. M. Molecular Catalysis of O2 Reduction by Iron Porphyrins in Water: Heterogeneous versus Homogeneous Pathways. J. Am. Chem. Soc. 2015, 137, 13535−13544. (183) Widegren, J. A.; Finke, R. G. A Review of the Problem of Distinguishing True Homogeneous Catalysis from Soluble or Other Metal-Particle Heterogeneous Catalysis under Reducing Conditions. J. Mol. Catal. A: Chem. 2003, 198, 317−341. (184) Kaeffer, N.; Morozan, A.; Fize, J.; Martinez, E.; Guetaz, L.; Artero, V. The Dark Side of Molecular Catalysis: Diimine-Dioxime Cobalt Complexes are not the Actual Hydrogen Evolution Electrocatalyst in Acidic Aqueous Solutions. ACS Catal. 2016, 6, 3727−3737. (185) Stracke, J. J.; Finke, R. G. Distinguishing Homogeneous from Heterogeneous Water Oxidation Catalysis When Beginning with Polyoxometalates. ACS Catal. 2014, 4, 909−933. (186) Sinha, S.; Aaron, M. S.; Blagojevic, J.; Warren, J. J. Electrocatalytic Dioxygen Reduction by Carbon Electrodes Noncovalently Modified with Iron Porphyrin Complexes: Enhancements from a Single Proton Relay. Chem. - Eur. J. 2015, 21, 18072−18075. (187) Mukherjee, S.; Bandyopadhyay, S.; Chatterjee, S.; Dey, A. Electrocatalytic O2 Reduction by a Monolayer of Hemin: The Role of pKa of Distal and Proximal Oxygen of a FeIII-OOH Species in Determining Reactivity. Chem. Commun. 2014, 50, 12304−12307. (188) Amanullah, S.; Das, P. K.; Samanta, S.; Dey, A. Tuning the Thermodynamic Onset Potential of Electrocatalytic O2 Reduction Reaction by Synthetic Iron-Porphyrin Complexes. Chem. Commun. 2015, 51, 10010−10013. (189) Bhagi-Damodaran, A.; Petrik, I. D.; Marshall, N. M.; Robinson, H.; Lu, Y. Systematic Tuning of Heme Redox Potentials and Its Effects on O2 Reduction Rates in a Designed Oxidase in Myoglobin. J. Am. Chem. Soc. 2014, 136, 11882−11885. (190) Samanta, S.; Sengupta, K.; Mittra, K.; Bandyopadhyay, S.; Dey, A. Selective Four Electron Reduction of O2 by an Iron Porphyrin Electrocatalyst under Fast and Slow Electron Fluxes. Chem. Commun. 2012, 48, 7631−7633. (191) Samanta, S.; Mittra, K.; Sengupta, K.; Chatterjee, S.; Dey, A. Second Sphere Control of Redox Catalysis: Selective Reduction of O2 to O2− or H2O by an Iron Porphyrin Catalyst. Inorg. Chem. 2013, 52, 1443−1453. (192) Mittra, K.; Chatterjee, S.; Samanta, S.; Dey, A. Selective 4e−/ 4H+ O2 Reduction by an Iron(Tetraferrocenyl)Porphyrin Complex: From Proton Transfer Followed by Electron Transfer in Organic Solvent to Proton Coupled Electron Transfer in Aqueous Medium. Inorg. Chem. 2013, 52, 14317−14325. (193) Mittra, K.; Chatterjee, S.; Samanta, S.; Sengupta, K.; Bhattacharjee, H.; Dey, A. A Hydrogen Bond Scaffold Supported Synthetic Heme FeIII-O2− Adduct. Chem. Commun. 2012, 48, 10535− 10537. (194) Sengupta, K.; Chatterjee, S.; Samanta, S.; Dey, A. Direct Observation of Intermediates Formed During Steady-State Electrocatalytic O2 Reduction by Iron Porphyrins. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 8431−8436. (195) Khorasani-Motlagh, M.; Noroozifar, M.; Ghaemi, A.; Safari, N. Iron(III) Octaethylporphyrin Chloride Supported on Glassy Carbon as an Electrocatalyst for Oxygen Reduction. J. Electroanal. Chem. 2004, 565, 115−120. 3790

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

(196) Yamaguchi, T.; Tsukamoto, K.; Ikeda, O.; Tanaka, R.; Kuwabara, T.; Takahashi, K. Oxygen Reduction at Negatively Charged Iron Porphyrins Heat-Treated and Bridged by Alkaline-Earth Metal Ions. Electrochim. Acta 2010, 55, 6042−6048. (197) Maid, H.; Böhm, P.; Huber, S. M.; Bauer, W.; Hummel, W.; Jux, N.; Gröger, H. Iron Catalysis for in Situ Regeneration of Oxidized Cofactors by Activation and Reduction of Molecular Oxygen: A Synthetic Metalloporphyrin as a Biomimetic NAD(P)H Oxidase. Angew. Chem., Int. Ed. 2011, 50, 2397−2400. (198) Shi, C. N.; Steiger, B.; Yuasa, M.; Anson, F. C. Electroreduction of O2 to H2O at Unusually Positive Potentials Catalyzed by the Simplest of the Cobalt Porphyrins. Inorg. Chem. 1997, 36, 4294− 4295. (199) Steiger, B.; Anson, F. C. Examination of Cobalt ″Picket Fence″ Porphyrin and Its Complex with 1-Methylimidazole as Catalysts for the Electroreduction of Dioxygen. Inorg. Chem. 2000, 39, 4579−4585. (200) Yamada, Y.; Fukunishi, Y.; Yamazaki, S.; Fukuzumi, S. Hydrogen Peroxide as Sustainable Fuel: Electrocatalysts for Production with a Solar Cell and Decomposition with a Fuel Cell. Chem. Commun. 2010, 46, 7334−7336. (201) Sun, B.; Ou, Z. P.; Yang, S. B.; Meng, D. Y.; Lu, G. F.; Fang, Y. Y.; Kadish, K. M. Synthesis and Electrochemistry of β-Pyrrole NitroSubstituted Cobalt(II) Porphyrins. The Effect of the NO2 Group on Redox Potentials, the Electron Transfer Mechanism and Catalytic Reduction of Molecular Oxygen in Acidic Media. Dalton Trans. 2014, 43, 10809−10815. (202) Yang, S. B.; Sun, B.; Ou, Z. P.; Meng, D. Y.; Lu, G. F.; Fang, Y. Y.; Kadish, K. M. β-Nitro-Substituted Free-Base, Iron(III) and Manganese(III) Tetraarylporphyrins: Synthesis, Electrochemistry and Effect of the NO2 Substituent on Spectra and Redox Potentials in Non-Aqueous Media. J. Porphyrins Phthalocyanines 2013, 17, 857−869. (203) Trojánek, A.; Langmaier, J.; Kvapilová, H.; Záliš, S.; Samec, Z. Inhibitory Effect of Water on the Oxygen Reduction Catalyzed by Cobalt(II) Tetraphenylporphyrin. J. Phys. Chem. A 2014, 118, 2018− 2028. (204) Hatay, I.; Su, B.; Li, F.; Méndez, M. A.; Khoury, T.; Gros, C. P.; Barbe, J. M.; Ersoz, M.; Samec, Z.; Girault, H. H. Proton-Coupled Oxygen Reduction at Liquid-Liquid Interfaces Catalyzed by Cobalt Porphine. J. Am. Chem. Soc. 2009, 131, 13453−13459. (205) Partovi-Nia, R.; Su, B.; Li, F.; Gros, C. P.; Barbe, J. M.; Samec, Z.; Girault, H. H. Proton Pump for O2 Reduction Catalyzed by 5,10,15,20-Tetraphenylporphyrinatocobalt(II). Chem. - Eur. J. 2009, 15, 2335−2340. (206) Su, B.; Hatay, I.; Trojánek, A.; Samec, Z.; Khoury, T.; Gros, C. P.; Barbe, J. M.; Daina, A.; Carrupt, P. A.; Girault, H. H. Molecular Electrocatalysis for Oxygen Reduction by Cobalt Porphyrins Adsorbed at Liquid/Liquid Interfaces. J. Am. Chem. Soc. 2010, 132, 2655−2662. (207) Partovi-Nia, R.; Su, B.; Méndez, M. A.; Habermeyer, B.; Gros, C. P.; Barbe, J. M.; Samec, Z.; Girault, H. H. Dioxygen Reduction by Cobalt(II) Octaethylporphyrin at Liquid|Liquid Interfaces. ChemPhysChem 2010, 11, 2979−2984. (208) Olaya, A. J.; Schaming, D.; Brevet, P. F.; Nagatani, H.; Xu, H. J.; Meyer, M.; Girault, H. H. Interfacial Self-Assembly of Water-Soluble Cationic Porphyrins for the Reduction of Oxygen to Water. Angew. Chem., Int. Ed. 2012, 51, 6447−6451. (209) Olaya, A. J.; Schaming, D.; Brevet, P. F.; Nagatani, H.; Zimmermann, T.; Vanicek, J.; Xu, H. J.; Gros, C. P.; Barbe, J. M.; Girault, H. H. Self-Assembled Molecular Rafts at Liquid|Liquid Interfaces for Four-Electron Oxygen Reduction. J. Am. Chem. Soc. 2012, 134, 498−506. (210) Peljo, P.; Rauhala, T.; Murtomäki, L.; Kallio, T.; Kontturi, K. Oxygen Reduction at a Water-1,2-Dichlorobenzene Interface Catalyzed by Cobalt Tetraphenyl Porphyrine - a Fuel Cell Approach. Int. J. Hydrogen Energy 2011, 36, 10033−10043. (211) Ou, Z. P.; Lü, A. X.; Meng, D. Y.; Huang, S.; Fang, Y. Y.; Lu, G. F.; Kadish, K. M. Molecular Oxygen Reduction Electrocatalyzed by meso-Substituted Cobalt Corroles Coated on Edge-Plane Pyrolytic Graphite Electrodes in Acidic Media. Inorg. Chem. 2012, 51, 8890− 8896.

(212) Li, B. H.; Ou, Z. P.; Meng, D. Y.; Tang, J. J.; Fang, Y. Y.; Liu, R.; Kadish, K. M. Cobalt Triarylcorroles Containing One, Two or Three Nitro Groups. Effect of NO2 Substitution on Electrochemical Properties and Catalytic Activity for Reduction of Molecular Oxygen in Acid Media. J. Inorg. Biochem. 2014, 136, 130−139. (213) Shi, Z.; Zhang, J. J. Density Functional Theory Study of Transitional Metal Macrocyclic Complexes’ Dioxygen-Binding Abilities and Their Catalytic Activities toward Oxygen Reduction Reaction. J. Phys. Chem. C 2007, 111, 7084−7090. (214) Tang, J. J.; Ou, Z. P.; Ye, L. N.; Yuan, M. Z.; Fang, Y. Y.; Xue, Z. L.; Kadish, K. M. Meso-Dichlorophenyl Substituted Co(III) Corrole: A Selective Electrocatalyst for the Two-Electron Reduction of Dioxygen in Acid Media, X-ray Crystal Structure Analysis and Electrochemistry. J. Porphyrins Phthalocyanines 2014, 18, 891−898. (215) Kadish, K. M.; Shen, J.; Frémond, L.; Chen, P.; Ojaimi, M. E.; Chkounda, M.; Gros, C. P.; Barbe, J. M.; Ohkubo, K.; Fukuzumi, S.; et al. Clarification of the Oxidation State of Cobalt Corroles in Heterogeneous and Homogeneous Catalytic Reduction of Dioxygen. Inorg. Chem. 2008, 47, 6726−6737. (216) Kadish, K. M.; Shao, J. G.; Ou, Z. P.; Gros, C. P.; Bolze, F.; Barbe, J. M.; Guilard, R. Alkyl- and Aryl-Substituted Corroles. 4. Solvent Effects on the Electrochemical and Spectral Properties of Cobalt Corroles. Inorg. Chem. 2003, 42, 4062−4070. (217) Collman, J. P.; Kaplun, M.; Decréau, R. A. Metal Corroles as Electrocatalysts for Oxygen Reduction. Dalton Trans. 2006, 554−559. (218) Schechter, A.; Stanevsky, M.; Mahammed, A.; Gross, Z. FourElectron Oxygen Reduction by Brominated Cobalt Corrole. Inorg. Chem. 2012, 51, 22−24. (219) Levy, N.; Mahammed, A.; Kosa, M.; Major, D. T.; Gross, Z.; Elbaz, L. Metallocorroles as Nonprecious-Metal Catalysts for Oxygen Reduction. Angew. Chem., Int. Ed. 2015, 54, 14080−14084. (220) Sun, S. R.; Jiang, N.; Xia, D. G. Density Functional Theory Study of the Oxygen Reduction Reaction on Metalloporphyrins and Metallophthalocyanines. J. Phys. Chem. C 2011, 115, 9511−9517. (221) Jung, J.; Liu, S.; Ohkubo, K.; Abu-Omar, M. M.; Fukuzumi, S. Catalytic Two-Electron Reduction of Dioxygen by Ferrocene Derivatives with Manganese(V) Corroles. Inorg. Chem. 2015, 54, 4285−4291. (222) Liu, S.; Mase, K.; Bougher, C.; Hicks, S. D.; Abu-Omar, M. M.; Fukuzumi, S. High-Valent Chromium-Oxo Complex Acting as an Efficient Catalyst Precursor for Selective Two-Electron Reduction of Dioxygen by a Ferrocene Derivative. Inorg. Chem. 2014, 53, 7780− 7788. (223) Isaacs, F.; Dehaen, W.; Maes, W.; Ngo, T. H.; Ruiz-León, D.; Herrera, F.; Arce, R.; Arévalo, M. C.; Aguirre, M. J. Direct 4-Electron Reduction of Molecular Oxygen to Water Mediated by Cu-10-(4Aminophenyl)-5,15-Dimesitylcorrole-Modified Electrodes. Int. J. Electrochem. Sci. 2013, 8, 3406−3418. (224) Masa, J.; Schuhmann, W. Systematic Selection of Metalloporphyrin-Based Catalysts for Oxygen Reduction by Modulation of the Donor-Acceptor Intermolecular Hardness. Chem. - Eur. J. 2013, 19, 9644−9654. (225) Rideal, E. K. Presidential Address. Concepts in Catalysis. The Contributions of Paul Sabatier and of Max Bodenstein. J. Chem. Soc. 1951, 1640−1647. (226) Trojánek, A.; Langmaier, J.; Šebera, J.; Záliš, S.; Barbe, J. M.; Girault, H. H.; Samec, Z. Fine Tuning of the Catalytic Effect of a Metal-Free Porphyrin on the Homogeneous Oxygen Reduction. Chem. Commun. 2011, 47, 5446−5448. (227) De Luca, G.; Romeo, A.; Scolaro, L. M.; Ricciardi, G.; Rosa, A. Evidence for Tetraphenylporphyrin Monoacids. Inorg. Chem. 2007, 46, 5979−5988. (228) Trojánek, A.; Langmaier, J.; Záliš, S.; Samec, Z. Competitive Inhibition of a Metal-Free Porphyrin Oxygen-Reduction Catalyst by Water. Chem. Commun. 2012, 48, 4094−4096. (229) Trojánek, A.; Langmaier, J.; Samec, Z. Thermodynamic Driving Force Effects in the Oxygen Reduction Catalyzed by a MetalFree Porphyrin. Electrochim. Acta 2012, 82, 457−462. 3791

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

(230) Wu, S. Z.; Su, B. Metal-Free-Porphyrin-Catalyzed Oxygen Reduction at Liquid-Liquid Interfaces. Chem. - Eur. J. 2012, 18, 3169− 3173. (231) Hatay, I.; Su, B.; Méndez, M. A.; Corminboeuf, C.; Khoury, T.; Gros, C. P.; Bourdillon, M.; Meyer, M.; Barbe, J. M.; Ersoz, M.; et al. Oxygen Reduction Catalyzed by a Fluorinated Tetraphenylporphyrin Free Base at Liquid/Liquid Interfaces. J. Am. Chem. Soc. 2010, 132, 13733−13741. (232) Hosseini, A.; Barile, C. J.; Devadoss, A.; Eberspacher, T. A.; Decreau, R. A.; Collman, J. P. Hybrid Bilayer Membrane: A Platform to Study the Role of Proton Flux on the Efficiency of Oxygen Reduction by a Molecular Electrocatalyst. J. Am. Chem. Soc. 2011, 133, 11100−11102. (233) Maruyama, J.; Baier, C.; Wolfschmidt, H.; Bele, P.; Stimming, U. Enhancement of Oxygen Reduction at Fe Tetrapyridyl Porphyrin by Pyridyl-N Coordination to Transition Metal Ions. Electrochim. Acta 2012, 63, 16−21. (234) Xuan, Y. F.; Huang, X.; Su, B. Biomimetic Oxygen Reduction Reaction Catalyzed by Microperoxidase-11 at Liquid/Liquid Interfaces. J. Phys. Chem. C 2015, 119, 11685−11693. (235) Chng, L. L.; Chang, C. J.; Nocera, D. G. Catalytic O-O Activation Chemistry Mediated by Iron Hangman Porphyrins with a Wide Range of Proton-Donating Abilities. Org. Lett. 2003, 5, 2421− 2424. (236) Chang, C. J.; Chng, L. L.; Nocera, D. G. Proton-Coupled O-O Activation on a Redox Platform Bearing a Hydrogen-Bonding Scaffold. J. Am. Chem. Soc. 2003, 125, 1866−1876. (237) Rosenthal, J.; Chng, L. L.; Fried, S. D.; Nocera, D. G. Stereochemical Control of H2O2 Dismutation by Hangman Porphyrins. Chem. Commun. 2007, 2642−2644. (238) Soper, J. D.; Kryatov, S. V.; Rybak-Akimova, E. V.; Nocera, D. G. Proton-Directed Redox Control of O-O Bond Activation by Heme Hydroperoxidase Models. J. Am. Chem. Soc. 2007, 129, 5069−5075. (239) Graham, D. J.; Dogutan, D. K.; Schwalbe, M.; Nocera, D. G. Hangman Effect on Hydrogen Peroxide Dismutation by Fe(III) Corroles. Chem. Commun. 2012, 48, 4175−4177. (240) Schwalbe, M.; Dogutan, D. K.; Stoian, S. A.; Teets, T. S.; Nocera, D. G. Xanthene-Modified and Hangman Iron Corroles. Inorg. Chem. 2011, 50, 1368−1377. (241) McGuire, R., Jr.; Dogutan, D. K.; Teets, T. S.; Suntivich, J.; Shao-Horn, Y.; Nocera, D. G. Oxygen Reduction Reactivity of Cobalt(II) Hangman Porphyrins. Chem. Sci. 2010, 1, 411−414. (242) Dogutan, D. K.; Stoian, S. A.; McGuire, R., Jr.; Schwalbe, M.; Teets, T. S.; Nocera, D. G. Hangman Corroles: Efficient Synthesis and Oxygen Reaction Chemistry. J. Am. Chem. Soc. 2011, 133, 131−140. (243) Sun, B.; Ou, Z. P.; Meng, D. Y.; Fang, Y. Y.; Song, Y.; Zhu, W. H.; Solntsev, P. V.; Nemykin, V. N.; Kadish, K. M. Electrochemistry and Catalytic Properties for Dioxygen Reduction Using FerroceneSubstituted Cobalt Porphyrins. Inorg. Chem. 2014, 53, 8600−8609. (244) Shi, C. N.; Anson, F. C. Comparison of the Catalytic Reduction of Dioxygen by [5,10,15,20-Tetrakis((Pentaammineruthenio(II))-4-Pyridyl)Porphyinato]Cobalt(II) in Solution and on Graphite Electrode Surfaces. Inorg. Chem. 1995, 34, 4554−4561. (245) Gadamsetti, K.; Swavey, S. Electrocatalytic Reduction of Oxygen at Electrodes Coated with a Bimetallic Cobalt(II)/Platinum(II) Porphyrin. Dalton Trans. 2006, 5530−5535. (246) Knoll, J.; Swavey, S. Electrocatalytic Reduction of Protons and Oxygen at Glassy Carbon Electrodes Coated with a Cobalt(II)/ Platinum(II) Porphyrin. Inorg. Chim. Acta 2009, 362, 2989−2993. (247) Swavey, S.; Eder, A. Enhanced O2 Electrocatalysis by a Highly Conjugated Cobalt(II) Porphyrin. Inorg. Chem. Commun. 2013, 29, 14−17. (248) Kadish, K. M.; Shao, J. G.; Ou, Z. P.; Frémond, L.; Zhan, R. Q.; Burdet, F.; Barbe, J. M.; Gros, C. P.; Guilard, R. Electrochemistry, Spectroelectrochemistry, Chloride Binding, and O2 Catalytic Reactions of Free-Base Porphyrin-Cobalt Corrole Dyads. Inorg. Chem. 2005, 44, 6744−6754.

(249) Collman, J. P.; Chng, L. L.; Tyvoll, D. A. Electrocatalytic Reduction of Dioxygen to Water by Iridium Porphyrins Adsorbed on Edge Plane Graphite Electrodes. Inorg. Chem. 1995, 34, 1311−1324. (250) Kadish, K. M.; Frémond, L.; Ou, Z. P.; Shao, J. G.; Shi, C. N.; Anson, F. C.; Burdet, F.; Gros, C. P.; Barbe, J. M.; Guilard, R. Cobalt(III) Corroles as Electrocatalysts for the Reduction of Dioxygen: Reactivity of a Monocorrole, Biscorroles, and PorphyrinCorrole Dyads. J. Am. Chem. Soc. 2005, 127, 5625−5631. (251) Kadish, K. M.; Frémond, L.; Shen, J.; Chen, P.; Ohkubo, K.; Fukuzumi, S.; El Ojaimi, M.; Gros, C. P.; Barbe, J. M.; Guilard, R. Catalytic Activity of Biscobalt Porphyrin-Corrole Dyads toward the Reduction of Dioxygen. Inorg. Chem. 2009, 48, 2571−2582. (252) Kadish, K. M.; Frémond, L.; Burdet, F.; Barbe, J. M.; Gros, C. P.; Guilard, R. Cobalt(IV) Corroles as Catalysts for the Electroreduction of O2: Reactions of Heterobimetallic Dyads Containing a Face-to-Face Linked Fe(III) or Mn(III) Porphyrin. J. Inorg. Biochem. 2006, 100, 858−868. (253) Fukuzumi, S.; Okamoto, K.; Gros, C. P.; Guilard, R. Mechanism of Four-Electron Reduction of Dioxygen to Water by Ferrocene Derivatives in the Presence of Perchloric Acid in Benzonitrile, Catalyzed by Cofacial Dicobalt Porphyrins. J. Am. Chem. Soc. 2004, 126, 10441−10449. (254) Fukuzumi, S.; Okamoto, K.; Tokuda, Y.; Gros, C. P.; Guilard, R. Dehydrogenation versus Oxygenation in Two-Electron and FourElectron Reduction of Dioxygen by 9-Alkyl-10-Methyl-9,10-Dihydroacridines Catalyzed by Monomeric Cobalt Porphyrins and Cofacial Dicobalt Porphyrins in the Presence of Perchloric Acid. J. Am. Chem. Soc. 2004, 126, 17059−17066. (255) Chang, C. J.; Deng, Y. Q.; Shi, C. N.; Chang, C. K.; Anson, F. C.; Nocera, D. G. Electrocatalytic Four-Electron Reduction of Oxygen to Water by a Highly Flexible Cofacial Cobalt Bisporphyrin. Chem. Commun. 2000, 1355−1356. (256) Chang, C. J.; Loh, Z. H.; Shi, C. N.; Anson, F. C.; Nocera, D. G. Targeted Proton Delivery in the Catalyzed Reduction of Oxygen to Water by Bimetallic Pacman Porphyrins. J. Am. Chem. Soc. 2004, 126, 10013−10020. (257) Peljo, P.; Murtomäki, L.; Kallio, T.; Xu, H. J.; Meyer, M.; Gros, C. P.; Barbe, J. M.; Girault, H. H.; Laasonen, K.; Kontturi, K. Biomimetic Oxygen Reduction by Cofacial Porphyrins at a LiquidLiquid Interface. J. Am. Chem. Soc. 2012, 134, 5974−5984. (258) Lu, W. B.; Wang, C. X.; Lv, Q. Y.; Zhou, X. H. Electrocatalytic Reduction of Dioxygen on the Surface of Glassy Carbon Electrodes Modified with Dicobalt Diporphyrin Complexes. J. Electroanal. Chem. 2003, 558, 59−63. (259) Chen, J.; Zhang, W. M.; Officer, D.; Swiegers, G. F.; Wallace, G. G. A Readily-Prepared, Convergent, Oxygen Reduction Electrocatalyst. Chem. Commun. 2007, 3353−3355. (260) D’Souza, F.; Hsieh, Y. Y.; Deviprasad, G. R. Four-Electron Electrocatalytic Reduction of Dioxygen to Water by an Ion-Pair Cobalt Porphyrin Dimer Adsorbed on a Glassy Carbon Electrode. Chem. Commun. 1998, 1027−1028. (261) Han, Y. Z.; Fang, H. Y.; Jing, H. Z.; Sun, H. L.; Lei, H. T.; Lai, W. Z.; Cao, R. Singly versus Doubly Reduced Nickel Porphyrins for Proton Reduction: Experimental and Theoretical Evidence for a Homolytic Hydrogen-Evolution Reaction. Angew. Chem., Int. Ed. 2016, 55, 5457−5462. (262) Kellett, R. M.; Spiro, T. G. Cobalt(I) Porphyrin Catalysis of Hydrogen Production from Water. Inorg. Chem. 1985, 24, 2373−2377. (263) Kellett, R. M.; Spiro, T. G. Cobalt Porphyrin Electrode Films as H2 Evolution Catalysts. Inorg. Chem. 1985, 24, 2378−2382. (264) Bhugun, I.; Lexa, D.; Savéant, J. M. Homogeneous Catalysis of Electrochemical Hydrogen Evolution by Iron(0) Porphyrins. J. Am. Chem. Soc. 1996, 118, 3982−3983. (265) Grass, V.; Lexa, D.; Savéant, J. M. Electrochemical Generation of Rhodium Porphyrin Hydrides. Catalysis of Hydrogen Evolution. J. Am. Chem. Soc. 1997, 119, 7526−7532. (266) Natali, M.; Luisa, A.; Iengo, E.; Scandola, F. Efficient Photocatalytic Hydrogen Generation from Water by a Cationic Cobalt(II) Porphyrin. Chem. Commun. 2014, 50, 1842−1844. 3792

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

(267) Sommer, D. J.; Vaughn, M. D.; Ghirlanda, G. Protein Secondary-Shell Interactions Enhance the Photoinduced Hydrogen Production of Cobalt Protoporphyrin IX. Chem. Commun. 2014, 50, 15852−15855. (268) Beyene, B. B.; Mane, S. B.; Hung, C. H. Highly Efficient Electrocatalytic Hydrogen Evolution from Neutral Aqueous Solution by a Water-Soluble Anionic Cobalt(II) Porphyrin. Chem. Commun. 2015, 51, 15067−15070. (269) Lei, H. T.; Fang, H. Y.; Han, Y. Z.; Lai, W. Z.; Fu, X. F.; Cao, R. Reactivity and Mechanism Studies of Hydrogen Evolution Catalyzed by Copper Corroles. ACS Catal. 2015, 5, 5145−5153. (270) Mirica, L. M.; Ottenwaelder, X.; Stack, T. D. P. Structure and Spectroscopy of Copper-Dioxygen Complexes. Chem. Rev. 2004, 104, 1013−1045. (271) Lewis, E. A.; Tolman, W. B. Reactivity of Dioxygen-Copper Systems. Chem. Rev. 2004, 104, 1047−1076. (272) Zhang, P. L.; Wang, M.; Yang, Y.; Yao, T. Y.; Sun, L. C. A Molecular Copper Catalyst for Electrochemical Water Reduction with a Large Hydrogen-Generation Rate Constant in Aqueous Solution. Angew. Chem., Int. Ed. 2014, 53, 13803−13807. (273) Luobeznova, I.; Simkhovich, L.; Goldberg, I.; Gross, Z. Electronic Structures and Reactivities of Corrole-Copper Complexes. Eur. J. Inorg. Chem. 2004, 2004, 1724−1732. (274) Wasbotten, I. H.; Wondimagegn, T.; Ghosh, A. Electronic Absorption, Resonance Raman, and Electrochemical Studies of Planar and Saddled Copper(III) meso-Triarylcorroles. Highly SubstituentSensitive Soret Bands as a Distinctive Feature of High-Valent Transition Metal Corroles. J. Am. Chem. Soc. 2002, 124, 8104−8116. (275) Mondal, B.; Sengupta, K.; Rana, A.; Mahammed, A.; Botoshansky, M.; Dey, S. G.; Gross, Z.; Dey, A. Cobalt Corrole Catalyst for Efficient Hydrogen Evolution Reaction from H2O under Ambient Conditions: Reactivity, Spectroscopy, and Density Functional Theory Calculations. Inorg. Chem. 2013, 52, 3381−3387. (276) Mahammed, A.; Mondal, B.; Rana, A.; Dey, A.; Gross, Z. The Cobalt Corrole Catalyzed Hydrogen Evolution Reaction: Surprising Electronic Effects and Characterization of Key Reaction Intermediates. Chem. Commun. 2014, 50, 2725−2727. (277) Collman, J. P.; Ha, Y. Y.; Wagenknecht, P. S.; Lopez, M. A.; Guilard, R. Cofacial Bisorganometallic Diporphyrins: Synthetic Control in Proton Reduction Catalysis. J. Am. Chem. Soc. 1993, 115, 9080−9088. (278) Kleingardner, J. G.; Kandemir, B.; Bren, K. L. Hydrogen Evolution from Neutral Water under Aerobic Conditions Catalyzed by Cobalt Microperoxidase-11. J. Am. Chem. Soc. 2014, 136, 4−7. (279) Sirbu, D.; Turta, C.; Gibson, E. A.; Benniston, A. C. The Ferrocene Effect: Enhanced Electrocatalytic Hydrogen Production Using meso-Tetraferrocenyl Porphyrin Palladium(II) and Copper(II) Complexes. Dalton Trans. 2015, 44, 14646−14655. (280) Bediako, D. K.; Solis, B. H.; Dogutan, D. K.; Roubelakis, M. M.; Maher, A. G.; Lee, C. H.; Chambers, M. B.; Hammes-Schiffer, S.; Nocera, D. G. Role of Pendant Proton Relays and Proton-Coupled Electron Transfer on the Hydrogen Evolution Reaction by Nickel Hangman Porphyrins. Proc. Natl. Acad. Sci. U. S. A. 2014, 111, 15001− 15006. (281) Graham, D. J.; Nocera, D. G. Electrocatalytic H2 Evolution by Proton-Gated Hangman Iron Porphyrins. Organometallics 2014, 33, 4994−5001. (282) Lee, C. H.; Dogutan, D. K.; Nocera, D. G. Hydrogen Generation by Hangman Metalloporphyrins. J. Am. Chem. Soc. 2011, 133, 8775−8777. (283) Roubelakis, M. M.; Bediako, D. K.; Dogutan, D. K.; Nocera, D. G. Proton-Coupled Electron Transfer Kinetics for the Hydrogen Evolution Reaction of Hangman Porphyrins. Energy Environ. Sci. 2012, 5, 7737−7740. (284) Solis, B. H.; Maher, A. G.; Honda, T.; Powers, D. C.; Nocera, D. G.; Hammes-Schiffer, S. Theoretical Analysis of Cobalt Hangman Porphyrins: Ligand Dearomatization and Mechanistic Implications for Hydrogen Evolution. ACS Catal. 2014, 4, 4516−4526.

(285) Solis, B. H.; Maher, A. G.; Dogutan, D. K.; Nocera, D. G.; Hammes-Schiffer, S. Nickel Phlorin Intermediate Formed by ProtonCoupled Electron Transfer in Hydrogen Evolution Mechanism. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 485−492. (286) Gao, Y.; Åkermark, T.; Liu, J. H.; Sun, L. C.; Åkermark, B. Nucleophilic Attack of Hydroxide on a MnV Oxo Complex: A Model of the O-O Bond Formation in the Oxygen Evolving Complex of Photosystem II. J. Am. Chem. Soc. 2009, 131, 8726−8727. (287) Luna, M. A.; Moyano, F.; Sereno, L.; D’Eramo, F. Spectroscopic and Electrochemical Studies of High-Valent Water Soluble Manganese Porphyrine. Electrocatalytic Water Oxidation. Electrochim. Acta 2014, 135, 301−310. (288) Nakazono, T.; Parent, A. R.; Sakai, K. Cobalt Porphyrins as Homogeneous Catalysts for Water Oxidation. Chem. Commun. 2013, 49, 6325−6327. (289) Nakazono, T.; Parent, A. R.; Sakai, K. Improving Singlet Oxygen Resistance During Photochemical Water Oxidation by Cobalt Porphyrin Catalysts. Chem. - Eur. J. 2015, 21, 6723−6726. (290) Han, A.; Jia, H. X.; Ma, H.; Ye, S. F.; Wu, H. T.; Lei, H. T.; Han, Y. Z.; Cao, R.; Du, P. W. Cobalt Porphyrin Electrode Films for Electrocatalytic Water Oxidation. Phys. Chem. Chem. Phys. 2014, 16, 11209−11217. (291) Vásquez, M. A. M.; Suárez, S. A.; Doctorovich, F. Gold and Silver Anchored Cobalt Porphyrins Used for Catalytic Water Splitting. Mater. Chem. Phys. 2015, 159, 159−166. (292) Suárez, S. A.; Fonticelli, M. H.; Rubert, A. A.; de la Llave, E.; Scherlis, D.; Salvarezza, R. C.; Martí, M. A.; Doctorovich, F. A Surface Effect Allows HNO/NO Discrimination by a Cobalt Porphyrin Bound to Gold. Inorg. Chem. 2010, 49, 6955−6966. (293) Wang, D.; Groves, J. T. Efficient Water Oxidation Catalyzed by Homogeneous Cationic Cobalt Porphyrins with Critical Roles for the Buffer Base. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 15579−15584. (294) Han, Y. Z.; Wu, Y. Z.; Lai, W. Z.; Cao, R. Electrocatalytic Water Oxidation by a Water-Soluble Nickel Porphyrin Complex at Neutral pH with Low Overpotential. Inorg. Chem. 2015, 54, 5604− 5613. (295) Zhang, T.; Wang, C.; Liu, S. B.; Wang, J. L.; Lin, W. B. A Biomimetic Copper Water Oxidation Catalyst with Low Overpotential. J. Am. Chem. Soc. 2014, 136, 273−281. (296) Lei, H. T.; Han, A. L.; Li, F. W.; Zhang, M. N.; Han, Y. Z.; Du, P. W.; Lai, W. Z.; Cao, R. Electrochemical, Spectroscopic and Theoretical Studies of a Simple Bifunctional Cobalt Corrole Catalyst for Oxygen Evolution and Hydrogen Production. Phys. Chem. Chem. Phys. 2014, 16, 1883−1893. (297) Mahammed, A.; Giladi, I.; Goldberg, I.; Gross, Z. Synthesis and Structural Characterization of a Novel Covalently-Bound Corrole Dimer. Chem. - Eur. J. 2001, 7, 4259−4265. (298) Schöfberger, W.; Faschinger, F.; Chattopadhyay, S.; Bhakta, S.; Mondal, B.; Elemans, J. A. A. W.; Müllegger, S.; Tebi, S.; Koch, R.; Klappenberger, F.; et al. A Bifunctional Electrocatalyst for Oxygen Evolution and Oxygen Reduction Reactions in Water. Angew. Chem., Int. Ed. 2016, 55, 2350−2355. (299) Mahammed, A.; Gross, Z. Metallocorroles as Photocatalysts for Driving Endergonic Reactions, Exemplified by Bromide to Bromine Conversion. Angew. Chem., Int. Ed. 2015, 54, 12370−12373. (300) Naruta, Y.; Sasayama, M.; Sasaki, T. Oxygen Evolution by Oxidation of Water with Manganese Porphyrin Dimers. Angew. Chem., Int. Ed. Engl. 1994, 33, 1839−1841. (301) Shimazaki, Y.; Nagano, T.; Takesue, H.; Ye, B. H.; Tani, F.; Naruta, Y. Characterization of a Dinuclear MnVO Complex and Its Efficient Evolution of O2 in the Presence of Water. Angew. Chem., Int. Ed. 2004, 43, 98−100. (302) Gao, Y.; Liu, J. H.; Wang, M.; Na, Y.; Åkermark, B.; Sun, L. C. Synthesis and Characterization of Manganese and Copper Corrole Xanthene Complexes as Catalysts for Water Oxidation. Tetrahedron 2007, 63, 1987−1994. (303) Privalov, T.; Sun, L. C.; Åkermark, B.; Liu, J. H.; Gao, Y.; Wang, M. A Computational Study of O-O Bond Formation Catalyzed 3793

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

by Mono- and Bis-MnIV-Corrole Complexes. Inorg. Chem. 2007, 46, 7075−7086. (304) Dogutan, D. K.; McGuire, R., Jr.; Nocera, D. G. Electocatalytic Water Oxidation by Cobalt(III) Hangman β-Octafluoro Corroles. J. Am. Chem. Soc. 2011, 133, 9178−9180. (305) Dogutan, D. K.; Bediako, D. K.; Graham, D. J.; Lemon, C. M.; Nocera, D. G. Proton-Coupled Electron Transfer Chemistry of Hangman Macrocycles: Hydrogen and Oxygen Evolution Reactions. J. Porphyrins Phthalocyanines 2015, 19, 1−8. (306) Lai, W. Z.; Cao, R.; Dong, G.; Shaik, S.; Yao, J. N.; Chen, H. Why Is Cobalt the Best Transition Metal in Transition-Metal Hangman Corroles for O-O Bond Formation During Water Oxidation? J. Phys. Chem. Lett. 2012, 3, 2315−2319. (307) Ertem, M. Z.; Cramer, C. J. Quantum Chemical Characterization of the Mechanism of a Supported Cobalt-Based Water Oxidation Catalyst. Dalton Trans. 2012, 41, 12213−12219. (308) Baran, J. D.; Grö nbeck, H.; Hellman, A. Analysis of Porphyrines as Catalysts for Electrochemical Reduction of O2 and Oxidation of H2O. J. Am. Chem. Soc. 2014, 136, 1320−1326. (309) Jahan, M.; Bao, Q. L.; Loh, K. P. Electrocatalytically Active Graphene-Porphyrin MOF Composite for Oxygen Reduction Reaction. J. Am. Chem. Soc. 2012, 134, 6707−6713. (310) Deng, S. Y.; Lei, J. P.; Huang, Y.; Cheng, Y.; Ju, H. X. Electrochemiluminescent Quenching of Quantum Dots for Ultrasensitive Immunoassay through Oxygen Reduction Catalyzed by Nitrogen-Doped Graphene-Supported Hemin. Anal. Chem. 2013, 85, 5390−5396. (311) Tang, H. J.; Yin, H. J.; Wang, J. Y.; Yang, N. L.; Wang, D.; Tang, Z. Y. Molecular Architecture of Cobalt Porphyrin Multilayers on Reduced Graphene Oxide Sheets for High-Performance Oxygen Reduction Reaction. Angew. Chem., Int. Ed. 2013, 52, 5585−5589. (312) Wei, P. J.; Yu, G. Q.; Naruta, Y.; Liu, J. G. Covalent Grafting of Carbon Nanotubes with a Biomimetic Heme Model Compound to Enhance Oxygen Reduction Reactions. Angew. Chem., Int. Ed. 2014, 53, 6659−6663. (313) Liu, J. G.; Ohta, T.; Yamaguchi, S.; Ogura, T.; Sakamoto, S.; Maeda, Y.; Naruta, Y. Spectroscopic Characterization of a Hydroperoxo-Heme Intermediate: Conversion of a Side-on Peroxo to an End-on Hydroperoxo Complex. Angew. Chem., Int. Ed. 2009, 48, 9262−9267. (314) Liu, J. G.; Naruta, Y.; Tani, F. A Functional Model of the Cytochrome c Oxidase Active Site: Unique Conversion of a Heme-μPeroxo-CuII Intermediate into Heme-Superoxo/CuI. Angew. Chem., Int. Ed. 2005, 44, 1836−1840. (315) Liu, J. G.; Shimizu, Y.; Ohta, T.; Naruta, Y. Formation of an End-on Ferric Peroxo Intermediate upon One-Electron Reduction of a Ferric Superoxo Heme. J. Am. Chem. Soc. 2010, 132, 3672−3673. (316) de Visser, S. P.; Valentine, J. S.; Nam, W. A Biomimetic Ferric Hydroperoxo Porphyrin Intermediate. Angew. Chem., Int. Ed. 2010, 49, 2099−2101. (317) Kieber-Emmons, M. T.; Qayyum, M. F.; Li, Y. Q.; Halime, Z.; Hodgson, K. O.; Hedman, B.; Karlin, K. D.; Solomon, E. I. Spectroscopic Elucidation of a New Heme/Copper Dioxygen Structure Type: Implications for O···O Bond Rupture in Cytochrome c Oxidase. Angew. Chem., Int. Ed. 2012, 51, 168−172. (318) Zhang, W.; Shaikh, A. U.; Tsui, E. Y.; Swager, T. M. Cobalt Porphyrin Functionalized Carbon Nanotubes for Oxygen Reduction. Chem. Mater. 2009, 21, 3234−3241. (319) Choi, A.; Jeong, H.; Kim, S.; Jo, S.; Jeon, S. Electrocatalytic Reduction of Dioxygen by Cobalt Porphyrin-Modified Glassy Carbon Electrode with Single-Walled Carbon Nanotubes and Nafion in Aqueous Solutions. Electrochim. Acta 2008, 53, 2579−2584. (320) Ahmed, M. S.; Jeong, H.; You, J. M.; Jeon, S. Electrocatalytic Reduction of Dioxygen at a Modified Glassy Carbon Electrode Based on Nafion®-Dispersed Single-Walled Carbon Nanotubes and CobaltPorphyrin with Palladium Nanoparticles in Acidic Media. Electrochim. Acta 2011, 56, 4924−4929.

(321) Kim, S. K.; Jeon, S. Improved Electrocatalytic Effect of Carbon Nanomaterials by Covalently Anchoring with CoTAPP via Diazonium Salt Reactions. Electrochem. Commun. 2012, 22, 141−144. (322) Morozan, A.; Campidelli, S.; Filoramo, A.; Jousselme, B.; Palacin, S. Catalytic Activity of Cobalt and Iron Phthalocyanines or Porphyrins Supported on Different Carbon Nanotubes Towards Oxygen Reduction Reaction. Carbon 2011, 49, 4839−4847. (323) Wang, Z. J.; Lei, H. T.; Cao, R.; Zhang, M. N. Cobalt Corrole on Carbon Nanotube as a Synergistic Catalyst for Oxygen Reduction Reaction in Acid Media. Electrochim. Acta 2015, 171, 81−88. (324) Chlistunoff, J.; Sansiñena, J. M. Effects of Axial Coordination of the Metal Center on the Activity of Iron Tetraphenylporphyrin as a Nonprecious Catalyst for Oxygen Reduction. J. Phys. Chem. C 2014, 118, 19139−19149. (325) Hijazi, I.; Bourgeteau, T.; Cornut, R.; Morozan, A.; Filoramo, A.; Leroy, J.; Derycke, V.; Jousselme, B.; Campidelli, S. Carbon Nanotube-Templated Synthesis of Covalent Porphyrin Network for Oxygen Reduction Reaction. J. Am. Chem. Soc. 2014, 136, 6348−6354. (326) Mezour, M. A.; Cornut, R.; Hussien, E. M.; Morin, M.; Mauzeroll, J. Detection of Hydrogen Peroxide Produced During the Oxygen Reduction Reaction at Self-Assembled Thiol-Porphyrin Monolayers on Gold Using SECM and Nanoelectrodes. Langmuir 2010, 26, 13000−13006. (327) Ribeiro, E. S.; Dias, S. L. P.; Gushikem, Y.; Kubota, L. T. Cobalt(II) Porphyrin Complex Immobilized on the Binary Oxide SiO2/Sb2O3: Electrochemical Properties and Dissolved Oxygen Reduction Study. Electrochim. Acta 2004, 49, 829−834. (328) Fuerte, A.; Corma, A.; Iglesias, M.; Morales, E.; Sánchez, F. Approaches to the Synthesis of Heterogenised Metalloporphyrins Application of New Materials as Electrocatalysts for Oxygen Reduction. J. Mol. Catal. A: Chem. 2006, 246, 109−117. (329) Nagai, T.; Yamazaki, S.; Asahi, M.; Siroma, Z.; Fujiwara, N.; Ioroi, T. Metalloporphyrin-Modified Perovskite-Type Oxide for the Electroreduction of Oxygen. J. Power Sources 2015, 293, 760−766. (330) Zhou, Q.; Li, C. M.; Li, J.; Cui, X. Q.; Gervasio, D. TemplateSynthesized Cobalt Porphyrin/Polypyrrole Nanocomposite and Its Electrocatalysis for Oxygen Reduction in Neutral Medium. J. Phys. Chem. C 2007, 111, 11216−11222. (331) Zhou, Q.; Li, C. M.; Li, J.; Lu, J. T. Electrocatalysis of Template-Electrosynthesized Cobalt-Porphyrin/Polyaniline Nanocomposite for Oxygen Reduction. J. Phys. Chem. C 2008, 112, 18578−18583. (332) Chen, W.; Akhigbe, J.; Brückner, C.; Li, C. M.; Lei, Y. Electrocatalytic Four-Electron Reduction of Dioxygen by Electrochemically Deposited Poly{[meso-Tetrakis(2-Thienyl)Porphyrinato]Cobalt(II)}. J. Phys. Chem. C 2010, 114, 8633−8638. (333) Elbaz, L.; Korin, E.; Soifer, L.; Bettelheim, A. Evidence for the Formation of Cobalt Porphyrin-Quinone Complexes Stabilized at Carbon-Based Surfaces toward the Design of Efficient Non-NobleMetal Oxygen Reduction Catalysts. J. Phys. Chem. Lett. 2010, 1, 398− 401. (334) Liu, H. S.; Zhang, L.; Zhang, J. J.; Ghosh, D.; Jung, J.; Downing, B. W.; Whittemore, E. Electrocatalytic Reduction of O2 and H2O2 by Adsorbed Cobalt Tetramethoxyphenyl Porphyrin and Its Application for Fuel Cell Cathodes. J. Power Sources 2006, 161, 743− 752. (335) Canales, C.; Ramírez, G. Glassy Carbon Electrodes Modified with Supramolecular Assemblies Generated by π-Stacking of Cobalt(II) Octaethylporphyrins. A 4 Electrons-Dioxygen Reduction Reaction Occurring at Positive Potentials. Electrochim. Acta 2015, 173, 636− 641. (336) Murphy, B. E.; Krasnikov, S. A.; Sergeeva, N. N.; Cafolla, A. A.; Preobrajenski, A. B.; Chaika, A. N.; Lübben, O.; Shvets, I. V. Homolytic Cleavage of Molecular Oxygen by Manganese Porphyrins Supported on Ag(111). ACS Nano 2014, 8, 5190−5198. (337) den Boer, D.; Li, M.; Habets, T.; Iavicoli, P.; Rowan, A. E.; Nolte, R. J. M.; Speller, S.; Amabilino, D. B.; De Feyter, S.; Elemans, J. A. A. W. Detection of Different Oxidation States of Individual 3794

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

Manganese Porphyrins During Their Reaction with Oxygen at a Solid/ Liquid Interface. Nat. Chem. 2013, 5, 621−627. (338) Hulsken, B.; Van Hameren, R.; Gerritsen, J. W.; Khoury, T.; Thordarson, P.; Crossley, M. J.; Rowan, A. E.; Nolte, R. J. M.; Elemans, J. A. A. W.; Speller, S. Real-Time Single-Molecule Imaging of Oxidation Catalysis at a Liquid-Solid Interface. Nat. Nanotechnol. 2007, 2, 285−289. (339) Shen, Y.; Liu, J. Y.; Jiang, J. G.; Liu, B. F.; Dong, S. J. Fabrication of a Metalloporphyrin-Polyoxometalate Hybrid Film by a Layer-by-Layer Method and Its Catalysis for Hydrogen Evolution and Dioxygen Reduction. J. Phys. Chem. B 2003, 107, 9744−9748. (340) Huang, D. K.; Lu, J. F.; Li, S. H.; Luo, Y. P.; Zhao, C.; Hu, B.; Wang, M. K.; Shen, Y. Fabrication of Cobalt Porphyrin. Electrochemically Reduced Graphene Oxide Hybrid Films for Electrocatalytic Hydrogen Evolution in Aqueous Solution. Langmuir 2014, 30, 6990− 6998. (341) Chen, J.; Wagner, P.; Tong, L.; Wallace, G. G.; Officer, D. L.; Swiegers, G. F. A Porphyrin-Doped Polymer Catalyzes Selective, Light-Assisted Water Oxidation in Seawater. Angew. Chem., Int. Ed. 2012, 51, 1907−1910. (342) Chen, J.; Wagner, P.; Tong, L.; Boskovic, D.; Zhang, W. M.; Officer, D.; Wallace, G. G.; Swiegers, G. F. A Light-Assisted, Polymeric Water Oxidation Catalyst That Selectively Oxidizes Seawater with a Low Onset Potential. Chem. Sci. 2013, 4, 2797−2803. (343) Jia, H. X.; Sun, Z. J.; Jiang, D. C.; Du, P. W. Covalent Cobalt Porphyrin Framework on Multiwalled Carbon Nanotubes for Efficient Water Oxidation at Low Overpotential. Chem. Mater. 2015, 27, 4586− 4593. (344) Nakashima, S.; Negishi, R.; Tada, H. Visible-Light-Induced Water Oxidation by a Hybrid Photocatalyst Consisting of Bismuth Vanadate and Copper(II) meso-Tetra(4-Carboxyphenyl)Porphyrin. Chem. Commun. 2016, 52, 3665−3668. (345) Matsuda, S.; Mori, S.; Hashimoto, K.; Nakanishi, S. Transition Metal Complexes with Macrocyclic Ligands Serve as Efficient Electrocatalysts for Aprotic Oxygen Evolution on Li2O2. J. Phys. Chem. C 2014, 118, 28435−28439. (346) Auwärter, W.; Écija, D.; Klappenberger, F.; Barth, J. V. Porphyrins at Interfaces. Nat. Chem. 2015, 7, 105−120. (347) Imahori, H. Giant Multiporphyrin Arrays as Artificial LightHarvesting Antennas. J. Phys. Chem. B 2004, 108, 6130−6143. (348) Fukuzumi, S.; Ohkubo, K.; Suenobu, T. Long-Lived Charge Separation and Applications in Artificial Photosynthesis. Acc. Chem. Res. 2014, 47, 1455−1464. (349) Ladomenou, K.; Natali, M.; Iengo, E.; Charalampidis, G.; Scandola, F.; Coutsolelos, A. G. Photochemical Hydrogen Generation with Porphyrin-Based Systems. Coord. Chem. Rev. 2015, 304-305, 38− 54. (350) Hasobe, T. Porphyrin-Based Supramolecular Nanoarchitectures for Solar Energy Conversion. J. Phys. Chem. Lett. 2013, 4, 1771− 1780. (351) Fontecilla-Camps, J. C.; Volbeda, A.; Cavazza, C.; Nicolet, Y. Structure/Function Relationships of [NiFe]- and [FeFe]-Hydrogenases. Chem. Rev. 2007, 107, 4273−4303. (352) Tard, C.; Pickett, C. J. Structural and Functional Analogues of the Active Sites of the [Fe]-, [NiFe]-, and [FeFe]-Hydrogenases. Chem. Rev. 2009, 109, 2245−2274. (353) Vincent, K. A.; Parkin, A.; Armstrong, F. A. Investigating and Exploiting the Electrocatalytic Properties of Hydrogenases. Chem. Rev. 2007, 107, 4366−4413. (354) Kluwer, A. M.; Kapre, R.; Hartl, F.; Lutz, M.; Spek, A. L.; Brouwer, A. M.; van Leeuwen, P. W. N. M.; Reek, J. N. H. SelfAssembled Biomimetic [2Fe2S]-Hydrogenase-Based Photocatalyst for Molecular Hydrogen Evolution. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 10460−10465. (355) Li, X. Q.; Wang, M.; Zhang, S. P.; Pan, J. X.; Na, Y.; Liu, J. H.; Åkermark, B.; Sun, L. C. Noncovalent Assembly of a Metalloporphyrin and an Iron Hydrogenase Active-Site Model: Photo-Induced Electron Transfer and Hydrogen Generation. J. Phys. Chem. B 2008, 112, 8198−8202.

(356) Poddutoori, P.; Co, D. T.; Samuel, A. P. S.; Kim, C. H.; Vagnini, M. T.; Wasielewski, M. R. Photoinitiated Multistep Charge Separation in Ferrocene-Zinc Porphyrin-Diiron Hydrogenase Model Complex Triads. Energy Environ. Sci. 2011, 4, 2441−2450. (357) Sasan, K.; Lin, Q. P.; Mao, C. Y.; Feng, P. Y. Incorporation of Iron Hydrogenase Active Sites into a Highly Stable Metal-Organic Framework for Photocatalytic Hydrogen Generation. Chem. Commun. 2014, 50, 10390−10393. (358) Sato, K.; Kataoka, Y.; Mori, W. Photochemical Production of Hydrogen from Water Using Microporous Porphyrin Coordination Lattices. J. Nanosci. Nanotechnol. 2012, 12, 585−590. (359) Natali, M.; Argazzi, R.; Chiorboli, C.; Iengo, E.; Scandola, F. Photocatalytic Hydrogen Evolution with a Self-Assembling ReductantSensitizer-Catalyst System. Chem. - Eur. J. 2013, 19, 9261−9271. (360) Zhang, P.; Wang, M.; Li, C. X.; Li, X. Q.; Dong, J. F.; Sun, L. C. Photochemical H2 Production with Noble-Metal-Free Molecular Devices Comprising a Porphyrin Photosensitizer and a Cobaloxime Catalyst. Chem. Commun. 2009, 46, 8806−8808. (361) Manton, J. C.; Long, C.; Vos, J. G.; Pryce, M. T. PorphyrinCobaloxime Complexes for Hydrogen Production, a Photo- and Electrochemical Study, Coupled with Quantum Chemical Calculations. Dalton Trans. 2014, 43, 3576−3583. (362) Natali, M.; Orlandi, M.; Chiorboli, C.; Iengo, E.; Bertolasi, V.; Scandola, F. Porphyrin-Cobaloxime Dyads for Photoinduced Hydrogen Production: Investigation of the Primary Photochemical Process. Photochem. Photobiol. Sci. 2013, 12, 1749−1753. (363) Peuntinger, K.; Lazarides, T.; Dafnomili, D.; Charalambidis, G.; Landrou, G.; Kahnt, A.; Sabatini, R. P.; McCamant, D. W.; Gryko, D. T.; Coutsolelos, A. G.; et al. Photoinduced Charge Transfer in Porphyrin-Cobaloxime and Corrole-Cobaloxime Hybrids. J. Phys. Chem. C 2013, 117, 1647−1655. (364) Zhu, M. S.; Dong, Y. P.; Du, Y. K.; Mou, Z. G.; Liu, J.; Yang, P.; Wang, X. M. Photocatalytic Hydrogen Evolution Based on Efficient Energy and Electron Transfers in Donor-Bridge-Acceptor Multibranched-Porphyrin-Functionalized Platinum Nanocomposites. Chem. - Eur. J. 2012, 18, 4367−4374. (365) Zhu, M. S.; Du, Y. K.; Yang, P.; Wang, X. M. Donor-Acceptor Porphyrin Functionalized Pt Nano-Assemblies for Artificial Photosynthesis: A Simple and Efficient Homogeneous Photocatalytic Hydrogen Production System. Catal. Sci. Technol. 2013, 3, 2295−2302. (366) El-Khouly, M. E.; Wijesinghe, C. A.; Nesterov, V. N.; Zandler, M. E.; Fukuzumi, S.; D’Souza, F. Ultrafast Photoinduced Energy and Electron Transfer in Multi-Modular Donor-Acceptor Conjugates. Chem. - Eur. J. 2012, 18, 13844−13853. (367) Song, L. C.; Liu, X. F.; Xie, Z. J.; Luo, F. X.; Song, H. B. Synthesis and Structural Characterization of Some New PorphyrinFullerene Dyads and Their Application in Photoinduced H2 Evolution. Inorg. Chem. 2011, 50, 11162−11172. (368) Matsuo, T.; Asano, A.; Ando, T.; Hisaeda, Y.; Hayashi, T. Photocatalytic Hydrogen Generation Using a Protein-Coated Photosensitizer with Anionic Patches and a Monocationic Electron Mediator. Chem. Commun. 2008, 3684−3686. (369) Tian, Y. M.; Martin, K. E.; Shelnutt, J. Y. T.; Evans, L.; Busani, T.; Miller, J. E.; Medforth, C. J.; Shelnutt, J. A. Morphological Families of Self-Assembled Porphyrin Structures and Their Photosensitization of Hydrogen Generation. Chem. Commun. 2011, 47, 6069−6071. (370) Zhu, M. S.; Lu, Y. T.; Du, Y. K.; Li, J. A.; Wang, X. M.; Yang, P. Photocatalytic Hydrogen Evolution without an Electron Mediator Using a Porphyrin-Pyrene Conjugate Functionalized Pt Nanocomposite as a Photocatalyst. Int. J. Hydrogen Energy 2011, 36, 4298−4304. (371) Kondo, M.; Ishigure, S.; Maki, Y.; Dewa, T.; Nango, M.; Amao, Y. Photoinduced Hydrogen Production with Photosensitization of Zn Chlorophyll Analog Dimer as a Photosynthetic Special Pair Model. Int. J. Hydrogen Energy 2015, 40, 5313−5318. (372) Kim, W.; Tachikawa, T.; Majima, T.; Li, C.; Kim, H. J.; Choi, W. Tin-Porphyrin Sensitized TiO2 for the Production of H2 under Visible Light. Energy Environ. Sci. 2010, 3, 1789−1795. 3795

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

(373) Zhu, M. S.; Li, Z.; Xiao, B.; Lu, Y. T.; Du, Y. K.; Yang, P.; Wang, X. M. Surfactant Assistance in Improvement of Photocatalytic Hydrogen Production with the Porphyrin Noncovalently Functionalized Graphene Nanocomposite. ACS Appl. Mater. Interfaces 2013, 5, 1732−1740. (374) Fateeva, A.; Chater, P. A.; Ireland, C. P.; Tahir, A. A.; Khimyak, Y. Z.; Wiper, P. V.; Darwent, J. R.; Rosseinsky, M. J. A Water-Stable Porphyrin-Based Metal-Organic Framework Active for Visible-Light Photocatalysis. Angew. Chem., Int. Ed. 2012, 51, 7440−7444. (375) Hasobe, T.; Sakai, H.; Mase, K.; Ohkubo, K.; Fukuzumi, S. Remarkable Enhancement of Photocatalytic Hydrogen Evolution Efficiency Utilizing an Internal Cavity of Supramolecular Porphyrin Hexagonal Nanocylinders under Visible-Light Irradiation. J. Phys. Chem. C 2013, 117, 4441−4449. (376) Moore, G. F.; Blakemore, J. D.; Milot, R. L.; Hull, J. F.; Song, H. E.; Cai, L.; Schmuttenmaer, C. A.; Crabtree, R. H.; Brudvig, G. W. A Visible Light Water-Splitting Cell with a Photoanode Formed by Codeposition of a High-Potential Porphyrin and an Iridium WaterOxidation Catalyst. Energy Environ. Sci. 2011, 4, 2389−2392. (377) Nam, Y. S.; Magyar, A. P.; Lee, D.; Kim, J. W.; Yun, D. S.; Park, H.; Pollom, J. T. S.; Weitz, D. A.; Belcher, A. M. Biologically Templated Photocatalytic Nanostructures for Sustained Light-Driven Water Oxidation. Nat. Nanotechnol. 2010, 5, 340−344. (378) Natali, M.; Deponti, E.; Vilona, D.; Sartorel, A.; Bonchio, M.; Scandola, F. A Bioinspired System for Light-Driven Water Oxidation with a Porphyrin Sensitizer and a Tetrametallic Molecular Catalyst. Eur. J. Inorg. Chem. 2015, 3467−3477. (379) Chen, H. C.; Hetterscheid, D. G. H.; Williams, R. M.; van der Vlugt, J. I.; Reek, J. N. H.; Brouwer, A. M. Platinum(II)-Porphyrin as a Sensitizer for Visible-Light Driven Water Oxidation in Neutral Phosphate Buffer. Energy Environ. Sci. 2015, 8, 975−982. (380) Yamamoto, M.; Wang, L.; Li, F. S.; Fukushima, T.; Tanaka, K.; Sun, L. C.; Imahori, H. Visible Light-Driven Water Oxidation Using a Covalently-Linked Molecular Catalyst-Sensitizer Dyad Assembled on a TiO2 Electrode. Chem. Sci. 2016, 7, 1430−1439. (381) Xiang, Z. H.; Xue, Y. H.; Cao, D. P.; Huang, L.; Chen, J. F.; Dai, L. M. Highly Efficient Electrocatalysts for Oxygen Reduction Based on 2D Covalent Organic Polymers Complexed with NonPrecious Metals. Angew. Chem., Int. Ed. 2014, 53, 2433−2437. (382) Lin, Q. P.; Bu, X. H.; Kong, A. G.; Mao, C. Y.; Zhao, X.; Bu, F.; Feng, P. Y. New Heterometallic Zirconium Metalloporphyrin Frameworks and Their Heteroatom-Activated High-Surface-Area Carbon Derivatives. J. Am. Chem. Soc. 2015, 137, 2235−2238. (383) Wu, Z. S.; Chen, L.; Liu, J. Z.; Parvez, K.; Liang, H. W.; Shu, J.; Sachdev, H.; Graf, R.; Feng, X. L.; Müllen, K. High-Performance Electrocatalysts for Oxygen Reduction Derived from Cobalt Porphyrin-Based Conjugated Mesoporous Polymers. Adv. Mater. 2014, 26, 1450−1455. (384) Schulenburg, H.; Stankov, S.; Schünemann, V.; Radnik, J.; Dorbandt, I.; Fiechter, S.; Bogdanoff, P.; Tributsch, H. Catalysts for the Oxygen Reduction from Heat-Treated Iron(III) Tetramethoxyphenylporphyrin Chloride: Structure and Stability of Active Sites. J. Phys. Chem. B 2003, 107, 9034−9041. (385) Kong, A. G.; Dong, B.; Zhu, X. F.; Kong, Y. Y.; Zhang, J. L.; Shan, Y. K. Ordered Mesoporous Fe-Porphyrin-Like Architectures as Excellent Cathode Materials for the Oxygen Reduction Reaction in Both Alkaline and Acidic Media. Chem. - Eur. J. 2013, 19, 16170− 16175. (386) Huang, H. C.; Wang, C. H.; Shown, I.; Chang, S. T.; Hsu, H. C.; Du, H. Y.; Chen, L. C.; Chen, K. H. High-Performance Pyrolyzed Iron Corrole as a Potential Non-Precious Metal Catalyst for PEMFCs. J. Mater. Chem. A 2013, 1, 14692−14699. (387) Huang, H. C.; Shown, I.; Chang, S. T.; Hsu, H. C.; Du, H. Y.; Kuo, M. C.; Wong, K. T.; Wang, S. F.; Wang, C. H.; Chen, L. C.; et al. Pyrolyzed Cobalt Corrole as a Potential Non-Precious Catalyst for Fuel Cells. Adv. Funct. Mater. 2012, 22, 3500−3508. (388) Chang, S. T.; Wang, C. H.; Du, H. Y.; Hsu, H. C.; Kang, C. M.; Chen, C. C.; Wu, J. C. S.; Yen, S. C.; Huang, W. F.; Chen, L. C.; et al. Vitalizing Fuel Cells with Vitamins: Pyrolyzed Vitamin B12 as a Non-

Precious Catalyst for Enhanced Oxygen Reduction Reaction of Polymer Electrolyte Fuel Cells. Energy Environ. Sci. 2012, 5, 5305− 5314. (389) Chang, S. T.; Huang, H. C.; Wang, H. C.; Hsu, H. C.; Lee, J. F.; Wang, C. H. Effects of Structures of Pyrolyzed Corrin, Corrole and Porphyrin on Oxygen Reduction Reaction. Int. J. Hydrogen Energy 2014, 39, 934−941. (390) Savastenko, N. A.; Anklam, K.; Quade, A.; Brüser, M.; Schmuhl, A.; Brüser, V. Comparative Study of Plasma-Treated NonPrecious Catalysts for Oxygen and Hydrogen Peroxide Reduction Reactions. Energy Environ. Sci. 2011, 4, 3461−3472. (391) Zhu, Y. S.; Zhang, B. S.; Liu, X.; Wang, D. W.; Su, D. S. Unravelling the Structure of Electrocatalytically Active Fe-N Complexes in Carbon for the Oxygen Reduction Reaction. Angew. Chem., Int. Ed. 2014, 53, 10673−10677. (392) Sun, Z. J.; Li, J. S.; Zheng, H. F.; Liu, X.; Ye, S.; Du, P. W. Pyrolyzed Cobalt Porphyrin-Modified Carbon Nanomaterial as an Active Catalyst for Electrocatalytic Water Oxidation. Int. J. Hydrogen Energy 2015, 40, 6538−6545. (393) Cheng, M. J.; Head-Gordon, M.; Bell, A. T. How to Chemically Tailor Metal-Porphyrin-Like Active Sites on Carbon Nanotubes and Graphene for Minimal Overpotential in the Electrochemical Oxygen Evolution and Oxygen Reduction Reactions. J. Phys. Chem. C 2014, 118, 29482−29491. (394) Honda, T.; Kojima, T.; Fukuzumi, S. Proton-Coupled Electron-Transfer Reduction of Dioxygen Catalyzed by a SaddleDistorted Cobalt Phthalocyanine. J. Am. Chem. Soc. 2012, 134, 4196− 4206. (395) Mase, K.; Ohkubo, K.; Fukuzumi, S. Efficient Two-Electron Reduction of Dioxygen to Hydrogen Peroxide with One-Electron Reductants with a Small Overpotential Catalyzed by a Cobalt Chlorin Complex. J. Am. Chem. Soc. 2013, 135, 2800−2808. (396) Givaja, G.; Volpe, M.; Edwards, M. A.; Blake, A. J.; Wilson, C.; Schröder, M.; Love, J. B. Dioxygen Reduction at Dicobalt Complexes of a Schiff Base Calixpyrrole Ligand. Angew. Chem., Int. Ed. 2007, 46, 584−586. (397) Volpe, M.; Hartnett, H.; Leeland, J. W.; Wills, K.; Ogunshun, M.; Duncombe, B. J.; Wilson, C.; Blake, A. J.; McMaster, J.; Love, J. B. Binuclear Cobalt Complexes of Schiff-Base Calixpyrroles and Their Roles in the Catalytic Reduction of Dioxygen. Inorg. Chem. 2009, 48, 5195−5207. (398) Askarizadeh, E.; Yaghoob, S. B.; Boghaei, D. M.; Slawin, A. M. Z.; Love, J. B. Tailoring Dicobalt Pacman Complexes of Schiff-Base Calixpyrroles Towards Dioxygen Reduction Catalysis. Chem. Commun. 2010, 46, 710−712. (399) Liu, R. L.; von Malotki, C.; Arnold, L.; Koshino, N.; Higashimura, H.; Baumgarten, M.; Müllen, K. Triangular Trinuclear Metal-N4 Complexes with High Electrocatalytic Activity for Oxygen Reduction. J. Am. Chem. Soc. 2011, 133, 10372−10375. (400) Lei, H. T.; Liu, C. Y.; Wang, Z. J.; Zhang, Z. Y.; Zhang, M. N.; Chang, X. M.; Zhang, W.; Cao, R. Noncovalent Immobilization of a Pyrene-Modified Cobalt Corrole on Carbon Supports for Enhanced Electrocatalytic Oxygen Reduction and Oxygen Evolution in Aqueous Solutions. ACS Catal. 2016, 6, 6429−6437. (401) Siewert, I. Proton-Coupled Electron Transfer Reactions Catalysed by 3d Metal Complexes. Chem. - Eur. J. 2015, 21, 15078− 15091. (402) Gagliardi, C. J.; Westlake, B. C.; Kent, C. A.; Paul, J. J.; Papanikolas, J. M.; Meyer, T. J. Integrating Proton Coupled Electron Transfer (PCET) and Excited States. Coord. Chem. Rev. 2010, 254, 2459−2471. (403) Gagliardi, C. J.; Vannucci, A. K.; Concepcion, J. J.; Chen, Z. F.; Meyer, T. J. The Role of Proton Coupled Electron Transfer in Water Oxidation. Energy Environ. Sci. 2012, 5, 7704−7717. (404) Sharma, V.; Enkavi, G.; Vattulainen, I.; Róg, T.; Wikström, M. Proton-Coupled Electron Transfer and the Role of Water Molecules in Proton Pumping by Cytochrome c Oxidase. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 2040−2045. 3796

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Chemical Reviews

Review

(405) Meyer, T. J.; Huynh, M. H. V.; Thorp, H. H. The Possible Role of Proton-Coupled Electron Transfer (PCET) in Water Oxidation by Photosystem II. Angew. Chem., Int. Ed. 2007, 46, 5284−5304. (406) Yamaguchi, A.; Inuzuka, R.; Takashima, T.; Hayashi, T.; Hashimoto, K.; Nakamura, R. Regulating Proton-Coupled Electron Transfer for Efficient Water Splitting by Manganese Oxides at Neutral pH. Nat. Commun. 2014, 5, 4256. (407) Hessenauer-Ilicheva, N.; Franke, A.; Meyer, D.; Woggon, W. D.; van Eldik, R. Low-Temperature Rapid-Scan Detection of Reactive Intermediates in Epoxidation Reactions Catalyzed by a New Enzyme Mimic of Cytochrome P450. J. Am. Chem. Soc. 2007, 129, 12473− 12479. (408) Suzuki, N.; Higuchi, T.; Nagano, T. Multiple Active Intermediates in Oxidation Reaction Catalyzed by Synthetic HemeThiolate Complex Relevant to Cytochrome P450. J. Am. Chem. Soc. 2002, 124, 9622−9628. (409) Ueda, T.; Kitagishi, H.; Kano, K. Intramolecular Direct Oxygen Transfer from Oxoferryl Porphyrin to a Sulfide Bond. Inorg. Chem. 2014, 53, 543−551. (410) Ramdhanie, B.; Telser, J.; Caneschi, A.; Zakharov, L. N.; Rheingold, A. L.; Goldberg, D. P. An Example of O2 Binding in a Cobalt(II) Corrole System and High-Valent Cobalt-Cyano and Cobalt-Alkynyl Complexes. J. Am. Chem. Soc. 2004, 126, 2515−2525. (411) Mandimutsira, B. S.; Ramdhanie, B.; Todd, R. C.; Wang, H. L.; Zareba, A. A.; Czernuszewicz, R. S.; Goldberg, D. P. A Stable Manganese(V)-Oxo Corrolazine Complex. J. Am. Chem. Soc. 2002, 124, 15170−15171. (412) Tangen, E.; Ghosh, A. Electronic Structure of High-Valent Transition Metal Corrolazine Complexes. The Young and Innocent? J. Am. Chem. Soc. 2002, 124, 8117−8121. (413) Wang, S. H. L.; Mandimutsira, B. S.; Todd, R.; Ramdhanie, B.; Fox, J. P.; Goldberg, D. P. Catalytic Sulfoxidation and Epoxidation with a Mn(III) Triazacorrole: Evidence for A ″Third Oxidant″ In High-Valent Porphyrinoid Oxidations. J. Am. Chem. Soc. 2004, 126, 18−19. (414) McGown, A. J.; Kerber, W. D.; Fujii, H.; Goldberg, D. P. Catalytic Reactivity of a Meso-N-Substituted Corrole and Evidence for a High-Valent Iron-Oxo Species. J. Am. Chem. Soc. 2009, 131, 8040− 8048.

3797

DOI: 10.1021/acs.chemrev.6b00299 Chem. Rev. 2017, 117, 3717−3797

Energy-Related Small Molecule Activation ... - ACS Publications

Recommend Documents