WO2012112758A2

WO2012112758A2 - Virtual hydrogen storage processes and related catalysts and systems

Info

Publication number: WO2012112758A2
Application number: PCT/US2012/025415
Authority: WO
Inventors: Robert Crabtree; Oana LUCA
Original assignee: Yale University
Priority date: 2011-02-17
Filing date: 2012-02-16
Publication date: 2012-08-23
Also published as: WO2012112758A3

Abstract

The present invention relates to virtual hydrogen storage processes which electrocatalytically reduce a liquid organic hydrogen carrier (LOHC) in a direct organic fuel cell/flow battery. Novel homogeneous catalysts comprising a tridendate, redox-active pincer ligand complex and related direct organic fuel cell/flow battery systems are also provided.

Description

Virtual Hydrogen Storage Processes and Related Catalysts and Systems

Related Applications/Research Support

This application claims the benefit of priority of U.S. Provisional Application Serial No. 61/443,898, entitled "Nickel pincers for proton and heterocycle reduction", filed

February 17, 2011, the entire contents of which are incorporated by reference herein.

The invention described herein was supported, in whole or in part, United States Department of Energy Grant No. DE-SC0001055. Consequently, the United States government has certain rights in the invention.

Field of the Invention

The invention relates to virtual hydrogen storage processes which electrocatalytically reduce a liquid organic hydrogen carrier (LOHC) in a direct organic fuel cell/flow battery. Novel homogeneous catalysts comprising a tridendate, redox-active pincer ligand complex and related direct organic fuel cell/flow battery systems are also provided.

Background of the Invention

Hydrogen is currently produced by steam reforming of fossil fuels , which is both expensive and detrimental to the environment. If hydrogen is to be a fuel in environmentally friendly alternative energy strategies, ³'⁴ more sustainable sources of hydrogen are required.

"Virtual hydrogen storage", the efficient storage of hydrogen

equivalents within the carbon framework of heterocyclic compounds (known as "liquid organic hydrogen carriers (LOHC)"), provides a potential solution to the dangers and design constraints posed associated with on-board and remote hydrogen storage tanks used in powering fuel cell-powered vehicles. In brief, in a most promising aspect of virtual hydrogen storage, LOHC's are stored on board a vehicle and fed directly to a proton exchange membrane (PEM) fuel cell (sometimes referred to in this context as an "organic fuel cell"). The LOHC is electrochemically dehydrogenated in the fuel cell without generating ¾; electrochemical dehydrogenation is achieved at relatively low temperatures (e.g. 100°C - 200°C) and high reaction rates.

Commercially- viable electrochemical proton generation in LOHC-based fuel cells, however, has been hindered by a number of factors. Elemental Pt is currently the best catalyst for the reduction of protons to H₂ ⁵ but its low abundance and high cost make it unsuitable for global use.³ A range of different transition metal complexes can act either as electrocatalysts or photocatalysts, including systems involving Co, ⁶ Mo^{7, 8} andNi.⁹' ^{10, 11} In a recent report, a tetradentate Co system with a redox active ligand also operates in aqueous conditions.¹² Improved systems that can operate in aqueous conditions with abundant first row transition metals are of current interest.

Known aqueous proton reduction processes have relied primarily on macrocyclic ligand catalysts which have only exhibited moderate activity and which are often not robust under low pH conditions.

Pincer ligands would be attractive aqueous proton reduction catalysts because they are easy to assemble from readily available materials and impart high stability to the resulting complexes. Furthermore, their modular nature facilitates tuning of ligand properties.¹³ To date, however, such catalysts have not been utilized successfully in aqueous proton reduction processes.

Accordingly, the need exists for improved aqueous proton reduction processes that are readily adaptable to proton exchange membrane (PEM) fuel cells, that utilize bi-functional (redox), homogeneous catalysts that are active under aqueous or non-aqueous acidic conditions, and that achieve high proton reduction turnover frequencies. The need also exists for highly-active aqueous proton reduction catalysts that are useful in such processes.

Summary of the Invention

In one embodiment, the invention provides a process comprising electrocatalytically reducing a liquid organic hydrogen carrier (LOHC) in a direct organic fuel cell/flow battery by contacting the LOHC with a homogeneous redox catalyst under aqueous or non-aqueous acidic conditions and at a temperature of between about 100° C to about 300° C. The catalyst employed in this process comprises a tridendate, redox-active pincer ligand complex containing a group 10 transition metal and preferably one or two coordinating groups bound to the transition metal.

In another embodiment, the invention provides supported homogeneous redox catalysts comprising tridendate, redox-active pincer ligand complexes as descried herein.

In one embodiment, a pincer ligand complex of the invention is defined by formula

(I):

wherein M is a group 10 transition metal selected from the group consisting of Ni, Pd, Pt, and Ds;

Xi and X₂ are the same or different and are a halogen, provided that either (but not both) of or X₂ may be absent;

Ri and R₂ are independently selected from the group consisting of a C₅-C₁₄ aryl or a

C₅-C₁₄ heteroaryl, said C5-Ci₄ aryl or C₅-C₁₄ heteroaryl being optionally substituted with one or more substituents selected from the group consisting of C C₆ alkyl, Ci-C₆ alkenyl, and Ci- C₆ alkynyl;

R₃ and R4 are independently selected from the group consisting of optionally substituted Ci- C₆ alkyl, optionally substituted Ci-C₆ alkenyl, and optionally substituted Ci-C^ alkynyl;

or R₁ and R₃ and R₂ and R4, together with the nitrogen to which they are bound, form an optionally substituted, single ring or fused two or three ring C₅-Ci₄ heteroaryl, said single ring heteroaryl optionally containing one or two additional ring heteroatoms selected from the group consisting of N and 0 and said two or three fused ring heteroaryl optionally containing between one, two, or three additional ring heteroatoms selected from the group consisting of N and 0;

R₅ and ¾ are absent or are independently selected from the group consisting of substituted or unsubstituted Ci-C₆ alkyl, Cj-C₆ alkenyl, and Ci-C alkynyl, or R₅ and R₆, together with the carbon atoms to which they are bound, form an optionally substituted C₅-Ci₀ aryl or a C5-C₁0 heteroaryl; and

n is 0 or 1.

In certain embodiments of a pincer ligand complex of formula (I):

M is Ni;

Xi and X₂ are both present and are Br; Ri and R₂ are independently selected from the group consisting of phenyl, naphthalene, anthracene, pyrrole, imidazole, pyrazole, pyridine, pyridazine, cinnoline, pyrimidine, diazine, triazine, indole, indoline, indolizine, quinoline, isoquinoline, quinoxaline, quinazoline, pteridine, quinolizidine, benzopyridine, benzoquinoline, perimidine, phenanthridine, acridine, phenazine, phenanthroline, carbazole, pyrazino pyridazine, pyrido pyrimidine, indazole, purine, and imidazo triazine, each which is optionally substituted with one or more Ci-Ce alkyl groups;

R₃ and R₄ are independently selected from the group consisting of optionally substituted Q- C₆ alkyl;

R₅ and ¾ are absent; and

is 1.

In still other embodiments of a pincer ligand complex of formula (I):

M is Ni;

Xi and X are both present and are Br;

R_\ and R₃ and R₂ and R4, together with the nitrogen to which they are bound, form pyrrole, imidazole, pyrazole, pyridine, pyridazine, cinnoline, pyrimidine, diazine, triazine, indole, indoline, indolizine, quinoline, isoquinoline, quinoxaline, quinazoline, pteridine,

quinolizidine, benzopyridine, benzoquinoline, perimidine, phenanthridine, acridine, phenazine, phenanthroline, naphthyridine, carbazole, pyrazino pyridazine, pyrido pyrimidine, indazole, purine, and imidazo triazine, each which is optionally substituted with one or more d-C₆ alkyl groups;

R₅ and ¾ are absent; and

n is 1.

In still other embodiments of a pincer ligand complex of formula (I):

M is Ni;

X_\ and X₂ are both present and are Br;

Ri and R₂ are independently selected from the group consisting furan, oxazole, isoxazole, oxadiazole, pyran, oxazine, dioxine, xanthine, benzofuran, dibenzofuran, and benzoxazine, each which is optionally substituted with one or more Ci-C₆ alkyl groups;

R₃ and are independently selected from the group consisting of optionally substituted Cr C₆ alkyl;

R₅ and ¾ are absent; and

n is 1.

In still other embodiments of a pincer ligand complex of formula (I): M is Ni;

Xi and X₂ are both present and are Br;

Ri and R₃ and R₂ and R4, together with the nitrogen to which they are bound, form isoxazole, oxazole, oxazine, oxazoline, oxadiazolidine, benzoxazine, and pyrido oxazine,

each which is optionally substituted with one or more C_\-C alkyl groups;

R₅ and ¾ are absent; and

is 1.

The pincer ligand complexes described herein include tautomers, regioisomers, geometric isomers, and where applicable, optical isomers (e.g. enantiomers) of the

complexes. Within its use in context, the term "pincer ligand complex" generally refers to a single compound, but also may include other compounds such as stereoisomers, regioisomers and/or optical isomers (including racemic mixtures) as well as specific enantiomers or enantiomerically enriched mixtures, as well as diastereomers and epimers, where applicable in context.

In one embodiment, where the pincer ligand complexes of the invention have surface tethering capability by click chemistry, one of R₅ or R₆ is preferably absent and the other is p- C₆H₄-alkynyl. See, for example, Meudtner, et ah, Multifunctional "Clickates" as Versatile Extended Heteroaromatic Building Blocks: Efficient Synthesis via Click Chemistry,

Conformational Preferences, and Metal Coordination, Chem. Eur. J. 2007, 13, 9834 - 9840, for representative "click chemistry" techniques.

In another embodiment, where a pincer ligand complex of the invention tethers directly to an oxide surface, one of R₅ or R₆ is preferably absent and the other is preferably p- C₆H4-acetylacetonato, /»-C₆H₄-phosphonate, or ^-CeKi-hydroxamate.

See, for example, Phan, et al., On the Nature of the Active Species in Palladium Catalyzed Mizoroki-Heck and Suzuki-Miyaura Couplings - Homogeneous or Heterogeneous Catalysis, A Critical Review, Advanced Synthesis Catalysis (2006), Volume: 348, Issue: 6,

Wiley Online Library, Pages: 609-679 for representative oxidative tethering techniques.

In another embodiment, a pincer ligand complex is defined generally by the formula M₂(OH)₂(L)₂, as illustrated in the complex Ni₂(OH)₂(N₇)₂ described in Example 2

hereinafter; these complexes may be generically represented by formula (I)(A) as follows:

where each substituent is as defined in formula (I) above, and Y is a Lewis acid (any species that donates an electron, e.g. BF₃, H⁺, Cu²⁺, Cr³⁺), a complex Lewis acid (e.g. B₂F₄), or the ligand of a coordination compound with the metal acting as the Lewis Acid (e.g.

[Zn(NH₃)₄]⁴⁺).

In a preferred embodiment the pincer ligand complex has the formula (II):

In another preferred embodiment, the pincer ligand complex has the formula (III):

In still another embodiment, a pincer ligand complex of the invention is defined by the formula (IV):

wherein M is a group 10 transition metal selected from the group consisting of Ni, Pd, Pt, and Ds; Xi and X₂ are the same or different and are a halogen, provided that either (but not both) of Xi or X₂ may be absent;

R[ and R₂ are independently selected from the group consisting of a C₅-C[₄ aryl or a C₅-Ci₄ heteroaryl, said C₅-C[₄ aryl or C₅-Ci₄ heteroaryl being optionally substituted with one or more substituents selected from the group consisting of Ci-C₆ alkyl, Ci-C₆ alkenyl, and Ci-C₆ alkynyl; and

R₅ and ¾ are absent or are independently selected from the group consisting of substituted or unsubstituted Ci-C₆ alkyl, Ci-C₆ alkenyl, and C[-C alkyny.

In yet another embodiment, the invention provides a process comprising

electrocatalytically reducing a liquid organic hydrogen carrier (LOHC) in a direct organic fuel cell/flow battery by contacting the LOHC with a homogeneous redox catalyst as depicted in Figure 1X3 herein under aqueous or non-aqueous acidic conditions and at a temperature of between about 100° C to about 300° C.

In still another embodiment the pincer ligand complex has the formula (V):

Rio, Rn, and Ri₂ are selected from the group consisting of mono or di-Ci-C alkyl, mono or di-Ci-Ci-C₆ alkenyl, and mono or di-Q-Q-Ce alkynyl (for R₁₀ and Rn, most preferably (tBu)₂, and for Ri₂, most preferably Me); and

Y is a Lewis acid (i.e., any species that donates an electron, e.g. BF₃, H⁺, Cu²⁺, Cr³⁺), a complex Lewis acid (e.g. B₂F₄), or the ligand of a coordination compound with the metal acting as the Lewis Acid (e.g. [Zn(NH₃)₄]⁴⁺). In certain embodiments, subsequent to the reduction of the LOHC, the reduced LOHC is electrohydrogenated in the presence of the redox catalyst using protons generated by electrooxidation of water at a cathode of the direct organic fuel cell/flow battery.

Additionally, in certain embodiments according to the present invention, prior to

electrohydrogenation, the LOHC may be dehydrogenated during an oxidation step in which stored energy is released, the dehydrogenation step occuring in the presence of Nickel oxide nanoparticles produced in solution or Nickle oxide which has solidified on the anode of the direct organic fuel cell/flow battery from the pincer ligand complex (M is Ni).

As used herein, a "LOHC" is a partially or fully hydrogenated, nitrogen-containing aromatic heterocycle. Representative LOHC s are described further hereinafter.

Techniques for supporting catalysts of the invention are described further in the examples below and can also employ known methodologies, such as those described in U.S. Patent Application Document No. 20100081034; Huang, et al., J. Am. Chem.

Soc, 2009, 131 (39), pp 13898-13899; and Fang, et al. J Am. Chem. Soc, 2009, 131 (42), pp 15330-15338.

In still other aspects, the present invention is directed to a composition comprising a partially or fully hydrogenated, nitrogen-containing aromatic heterocycle as described above as a liquid organic hydrogen carrier in combination with a catalyst as otherwise described herein. In this aspect of the invention, the heterocycle is selected from the group consisting of decalin, 2-aminopyridine, 4-methylpyrimidine, dipyrimidinemethane, dimethyltetrazine, dipirimidine, diazacarbazole, alkylcarbazole, 4-aminopyridine, dipyrazinemethane, tripyrazinemethane, tripyrazineamine, dipyrazine, tetrazacarbazole, isoquinoline, di(2- pyridyl)amine, quinazoline, perhydro-N-ethylcarbazole, tetrahydroisoquinoline, perhydro- dibenzofuran, perhydro-indole, N-methyl perhydro-indole, 4, 4'-bipiperidine, 4-amino- methylpyrimidine, aminomethylcyclohexane, 1 ,2,3,4-tetrahydroquinaldine, 2,3- dimethyltetrahydroquinoxaline or a mixture thereof. In certain preferred aspects, the heterocycle is 1,2,3,4-tetrahydroquinaldine, 2,3-dimethyltetrahydroquinoxaline or a mixture thereof. In a further preferred aspect, the present invention is directed to a composition comprising a homogeneous redox catalyst in combination with 1,2,3,4-tetrahydroquinaldine, 2,3-dimethyltetrahydroquinoxaline or a mixture thereof. Brief Description of the Figures

Example 1 Figures:

Figure 1. Cyclic voltammograms of 2 mM catalyst 1 in 0.1 M NBU4BF4 acetonitrile solution: Black: 25 mV/s, Red: 50 mV/s, Blue: 100 mV/s.

Figure 2. Upper panel: Catalytic cycle based on formation of a Niⁿ-hydride intermediate by PCET proposed based on calculations at the DFT B3LYP/cc-pVTZ level of theory. Lower panel: Reaction free energy profile showing that PCET enables transformation of [ (ΟΗ₂)]⁺ into [Γ(Η)]⁺ at low voltage. Transformation of [l'(OH₂)]⁺ into [l'(H)]⁺ via an alternative pathway (shown in grey, lower panel) with sequential protonation and reduction without invoking PCET is thermodynamically unfavorable.

Figure 3. Ligand-centered reduction illustrated by Mulliken spin population analysis of intermediates [Γ(ΟΗ₂)]⁺ (left), traw-[r(OH₂)₂H]²⁺ (center) and cw-[r(OH₂)₂H]²⁺ (right) showing spin derealization (in green) in the reduced ligand for [Γ(ΟΗ₂)]⁺, and localization of the excess spin density mostly on the Ni center of the Ni-hydride complexes cis- [ (ΟΗ₂)2Η]²⁺ traw-[l'(OH₂)₂H]²⁺. No excess spin is observable in the analyses of

[Γ(ΟΗ₂)]²⁺, [Γ(Η)]⁺, and [Γ(Η₂)]²⁺. Color key population: dark red (-0.08) - bright green (+0.27), [l'(OH₂)]⁺; dark red (-0.20) - bright green (0.99), traw-[l'(OH₂)₂H]²⁺; dark red (-0.20) - bright green (1.04), cw-[l'(OH₂)₂H]²⁺.

Figure Sl-1. Cyclic voltammograms of a background 0.1 M NBu₄BF₄ acetonitrile solution (black) in the presence of 10pL (red), 20uL (blue), 30uL (purple) and 40 pL (green) 1 M HC1 (100 mV/s). Smoothed 5 point adjusted averaging was used to remove electrical noise.

Figure Sl-2. Cyclic voltammograms at 100 mV/s (black), 200 mV/s (red), 300 mV/s (blue) of 2 mM 1 in 0.1 M NBu₄BF₄ acetonitrile solution.

Figure Sl-3. Cyclic voltammograms of 2 mM 1 in 0.1 M NBu₄BF₄ acetonitrile solution (black) in the presence of 10 pL (blue), 20 pL (red), 30 uL (purple) and 40 μΙ_ (green) 1 M HC1.

Figure Sl-4. Cyclic voltammograms at 100 mV/s (black), 200 mV/s (red), 300 mV/s (blue) of 2 mM 2 in 0.1 M NBu₄BF₄ acetonitrile solution.

Figure Sl-5. Cyclic voltammograms of 2 mM 2 in 0.1M NBu₄BF₄ acetonitrile solution (black) in the presence of 10 uL (red), 20 uL (blue), 30 uL (purple) and 40 uL (green) at 100 mV/s.

Figure Sl-6. Cyclic voltammograms at 100 mV/s(black), 200 mV/s(red), 300 mV/s (blue) of 2 mM 3 in 0.1 M NBu₄BF₄ in acetonitrile solution.

Figure Sl-7. Cyclic voltammograms of 2 mM 3 in 0.1M NBu₄BF₄ acetonitrile solution (red) in the presence of 10 uL (black), 20 uL (blue), 30 uL (purple) and 40 uL (green) at 100 mV/s. Figure Sl-8. Cyclic voltammograms at 100 mV/s (black), 200 mV/s (red), 300 mV/s (blue) of 2 mM 3-H solution in 0.1 M NBu₄BF₄ in acetonitrile.

Figure Sl-9. Cyclic voltammograms of 2 mM 3-H (red) overlaid with CV of 2 mM solution 3 (black) in 0.1 M NBU BF₄. acetonitrile solution at 100 mV/s. The reduction potential at the metal center is drastically shifted.

Figure Sl-10. Cyclic voltammograms of 2 mM 3-H in 0.1M NBu₄BF₄ acetonitrile solution (black) in the presence of 10 uL (red), 20 uL (blue), 30 uL (purple) and 40 uL (green) 1 M HC1 at 100 mV/s. Bubbles on electrode formed immediately after addition of acid.

Figure S2-1. Current density vs. applied potential: steady state current density at the end of chronoamperometry experiments at progressively more negative potentials - chronoamperograms (dwell time: 60 sec) at a glassy carbon electrode (in a three electrode setup- Pt counter electrode, non-aqueous Ag pseudoreference) in 5.2 mL acetonitrile

NBu₄BF₄ solutions containing 200 iL 1 M aqueous HC1 at catalyst concentrations of 0.2 mM with magnetic stirring.

Figure S3-1. Quantitative Mass Spectrometry calibration and catalytic runs.

Figure S5-1. Plot of i_c currents vs. concentration of catalyst 1 at 100 mV/s.

Figure S5-2. Plots of i_c / i_p ratios vs. acid concentration at 4mM catalyst 1 at 100 mV/s. 400 mV/s, and 500 mV/s.

Figure S5-3. Plots of the slopes of i_c / i_p ratios vs. acid concentration in Figure S5-2 vs. l/(sqrt (D)).

Figure S5-4. Plot of i_c currents vs. concentration of catalyst 2 at 100 mV/s.

Figure S5-5. Plots of i_c / i_p ratios vs. acid concentration at 4mM catalyst 1 at 100 mV/s, 200 mV/s and 300 mV/s.

Figure S5-6. Plots of the slopes of i_c / i_p ratios vs. acid concentration in Figure S5-5 vs. l/(sqrt (D)).

Figure S5-7. Plot of i_c currents vs. concentration of catalyst 3 at 100 mV/s.

Figure S5-8. Plots of i_c / i_p ratios vs. acid concentration at 4mM catalyst 3 at 100 mV/s, 200 mV/s, 300 mV/s, 400 mV/s and 500 mV/s.

Figure S5-9. Plots of the slopes of i_c / i_p ratios vs. acid concentration in Figure S5-8 vs. l/(sqrt (D)).

Example 2 Figures:

Figure 1 A: presents synthesis and spectro graphic data for the N5 ligand of Example 2.

Figure IB: presents synthesis and spectro graphic data for the N7 ligand of Example 2. Figure 1C: presents elemental analysis of the 2,6-di(naphtyridyn-2-yl)pyridine Nickel (II) dibromide (l)-(3) complexes of Example 2.

Example 3 Figures:

Figure 1X3. Nickel complexes screened for proton-reduction electrocatalysis.

Figure 2X3. Solid lines: Non-aqueous electrochemistry of compounds 1-3: 2 mM Ni complex in a 0.1 M NBu₄BF acetonitrile solution at a glassy carbon electrode (100 mV/s) top to bottom: green 1; blue 2; purple 3. See SI for CVs of complexes 1-3 at different scan rates. Dashed: 2 mM solutions of complexes 1-3 in a 0.1 M NBu₄BF₄ acetonitrile solution with 20 uL 1 M HC1 at 100 mV/s.

Figure 3X3. Blue: Cyclic voltammogram of 2 mM 3-H in a 0.1 M NBu₄BF₄ acetonitrile solution under air/water free conditions 100 mV/s. Grey: Cyclic voltammograms of parent compound 3 (top grey CV completely silent in the scanned region) involved in proton reduction with 4*10 uL increments of 0.1 M HC1.

Figure 4X3. Overlay of CVs of acetonitrile (0.1 M NBu₄BF₄) solutions 2 mM 3 (red), 3-H (black), 3-MeCN⁺ (blue) at 100 mV/s.

Example 4 Figures

Figure 1X4. Ni complexes with pincer ligands screened for proton-reduction electrocatalysis.

Figure 2X4. Solid lines: Non-aqueous reduction electrochemistry of the Ni complexes 1-5: 2 mM Ni complex in 0.1 M NBu₄BF₄ acetonitrile solution using a glassy carbon electrode (100 mV/s) left panel: green 1; blue 2; purple 3; right panel: orange 4; red 5. Dashed: with 20 uL 1 M HC1 at 100 mV/s; referenced vs. NHE externally against the Fc/Fc⁺ couple at 690 mV; See SI for wide scan cyclic voltammograms (CVs) of complexes 1-5 at different scan rates.

Figure 3X4. Cyclic voltammograms (25 mV/s, basal plane graphite working electrode,¹ Pt counter electrode, Ag/AgCl reference electrode, Ar purge) for 1 mM 4 in 3 mL 0.1 M aqueous KC1 (black: neutral) with incremental amounts of added acid: 40 uL (red), 90 uL (green), 140 uL (blue).¹¹ Inset: onset region of current increase in the presence of added acid.

Figure 4X4. Cyclic voltammograms (25 mV/s, basal plane graphite working electrode,³⁵ Pt counter electrode, Ag/AgCl reference electrode, Ar purge) for 1 mM catalyst precursor 5 in 3 mL 0.1 M aqueous KC1 and 1 mL acetonitrile (black: neutral) with incremental amounts of added acid: 40 uL (red), 90 uL (blue), 140 uL (purple). Inset: onset of current increase in the presence of added acid. Figure 5X4. Upper panel: Catalytic cycle based on formation of a Niⁿ-hydride intermediate by PCET proposed based on calculations at the DFT B3LYP/cc-pVTZ level of theory. Lower panel: Reaction free energy profile showing that PCET enables transformation of [4'(OH₂)]⁺ into [4'(H)]⁺ at low voltage. Transformation of [4'(OH₂)]⁺ into [4'(H)]⁺ via an alternative pathway (shown in grey, lower panel) with sequential protonation and reduction without invoking PCET is thermodynamically unfavorable.

Figure 6X4. Ligand-centered reduction illustrated by Mulliken spin population analysis of intermediates [4'(OH₂)]⁺ (left), trans-[4'(OH₂)₂H]²⁺ (center) and m-[4'(OH₂)₂H]²⁺ (right) showing spin delocalization in the reduced ligand for [4'(OH₂)]⁺, and localization of the excess spin density mostly on the Ni center of the Ni-hydride complexes ci5-[4'(OH₂)2H]²⁺ trims-[4^*(OH₂)₂H]²⁺. No excess spin is observable in the analyses of [4'(OH₂)]²⁺, [4'(H)]⁺, and [4^*(H₂)]²⁺. Color key population: dark red (-QM)-bright green (+0.27), [4*(OH₂)]⁺; dark red (-0.20)-bright green (0.99), tri s-[4'(OH₂)₂H]²⁺; dark red (-0.20)-b right green (1.04), cw-[4'(OH₂)₂H]²⁺.

Figure 7X4. Proton reduction cycle for 3 based on calculations at the DFT B3LYP/cc-pVTZ level of theory.

Figure 8X4. Mulliken spin population analysis of intermediates in the proton reduction cycle of 3 (Figure 7X4) illustrating that no spin is delocalized onto the ligand framework for the intermediates.

Figure 9X4. Blue: Cyclic voltammogram of 2 mM [3'(H)] in 0.1 M NBu₄BF₄ in acetonitrile under air/water free conditions 100 mV/s. Grey: Cyclic voltammograms of parent compound 3 (top grey CV completely silent in that region) involved in proton reduction with 4* 10 μΕ increments of 0.1 M HCl.

Figure 10X4. Overlay of CVs of 0.1 M NBU₄BF₄ in acetonitrile containing 2 mM 3 (red),

[3'(H)] (black), [3'(MeCN)]⁺ (Z>/we) at 100 mV/s.

Figure 11X4. Overlay of voltammograms of 0.1 M NBu₄BF₄ in acetonitrile solutions containing 2 mM [3^»(H)]- dark blue and [3'(MeCN)]⁺ in the presence of 5* 10 μΕ 1 M HCl at 100 mV/s -grey. Figure 4. Figure 4 depicts a schematic illustration of a direct feed LOHC PEM fuel cell system that operates in accordance with the processes described herein.

Detailed Description of the Invention

The following terms, among others, are used to describe the present invention. It is to be understood that a term which is not specifically defined is to be given a meaning consistent with the use of that term within the context of the present invention as understood by those of ordinary skill.

The symbol is used in chemical compounds according to the present invention to signify that a bond between atoms is a single bond or double bond according

to the context of the bond's use in the compound, which depends on the atoms (and substituents) used in defining the present compounds. Thus, where a carbon (or other) atom is used and the context of the use of the atom calls for a double bond or single bond to link that atom with an adjacent atom in order to maintain the appropriate valence of the atoms used, then that bond is considered a double bond or a single bond.

The term "alkenyl", as used herein, refers to an aliphatic group containing at least one double bond and is intended to include both "unsubstituted alkenyls" and "substituted alkenyls", the latter of which refers to alkenyl moieties having substituents replacing a hydrogen on one or more carbons of the alkenyl group. Such substituents may occur on one or more carbons that are included or not included in one or more double bonds. Moreover, such substituents include all those contemplated for alkyl groups, as discussed herein, except where stability of the moiety is prohibitive. For example, substitution of alkenyl groups by one or more alkyl, carbocyclyl, aryl, heterocyclyl, or heteroaryl groups is contemplated.

The term "alkyl" refers to the radical of saturated aliphatic groups, including straight- chain alkyl groups, branched-chain alkyl groups, cycloalkyl (alicyclic) groups, alkyl- substituted cycloalkyl groups, and cycloalkyl-substituted alkyl groups. In preferred embodiments, a straight chain or branched chain alkyl has 10 or fewer carbon atoms in its backbone (e.g., Q-Cici for straight chains, Q-Qo for branched chains), and more preferably 8 or fewer, and most preferably 6 or fewer. Likewise, preferred cycloalkyls have from 3-10 carbon atoms in their ring structure, and more preferably have 5, 6, 7 or 8 carbons in the ring structure.

Moreover, the term "alkyl" (or "lower alkyl") as used throughout the specification, examples, and claims is intended to include both "unsubstituted alkyls" and "substituted alkyls", the latter of which refers to alkyl moieties having substituents replacing a hydrogen on one or more carbons of the hydrocarbon backbone. Such substituents can include, for example, a halogen, a hydroxyl, a carbonyl (such as a carboxyl, an alkoxycarbonyl, a formyl, or an acyl), a thiocarbonyl (such as a thioester, a thioacetate, or a thioformate), an alkoxyl, a phosphoryl, a phosphate, a phosphonate, a phosphinate, an amino, an amido, an amidine, an imine, a cyano, a nitro, an azido, a sulihydryl, an alkylthio, a sulfate, a sulfonate, a sulfamoyl, a sulfonamido, a sulfonyl, a heterocyclyl, an aralkyl, or an aromatic or heteroaromatic moiety or as otherwise described herein. It will be understood by those skilled in the art that the individual substituent chemical moieties can themselves be substituted. For instance, the substituents of a substituted alkyl may include substituted and unsubstituted forms of amino, azido, imino, amido, phosphoryl (including phosphonate and phosphinate), sulfonyl

(including sulfate, sulfonamido, sulfamoyl and sulfonate), and silyl groups, as well as ethers, alkylthios, carbonyls (including ketones, aldehydes, carboxylates, and esters),— CF₃, --CN and the like. Exemplary, non-limiting substituted alkyls are described herein. Cycloalkyls can be further substituted with alkyls, alkenyls, alkynyls, alkoxys, alkylthios, aminoalkyls, carbonyl-substituted alkyls, ~CF₃, --CN, and the like.

Analogous substitutions can be made to alkenyl and alkynyl groups to produce, for example, without limitation, aminoalkenyls, aminoalkynyls, amidoalkenyls, amidoalkynyls, iminoalkenyls, iminoalkynyls, thioalkenyls, thioalkynyls, carbonyl-substituted alkenyls or alkynyls.

Unless the number of carbons is otherwise specified, "lower alkyl" as used herein means an alkyl group, as defined above, but having from one to eight carbons, more preferably from one to six carbon atoms in its backbone structure. Likewise, "lower alkenyl" and "lower alkynyl" have similar chain lengths. Throughout the application, preferred alkyl groups are lower alkyls. In preferred embodiments, a substituent designated herein as alkyl is a lower alkyl.

The term "alkynyl", as used herein, refers to an aliphatic group containing at least one triple bond and is intended to include both "unsubstituted alkynyls" and "substituted alkynyls", the latter of which refers to alkynyl moieties having substituents replacing a hydrogen on one or more carbons of the alkynyl group. Such substituents may occur on one or more carbons that are included or not included in one or more triple bonds. Moreover, such substituents include all those contemplated for alkyl groups, as discussed above, except where stability is prohibitive. For example, substitution of alkynyl groups by one or more alkyl, carbocyclyl, aryl, heterocyclyl, or heteroaryl groups is contemplated. The term "aryl" as used herein includes 5-, 6-, and 7-membered single-ring or aromatic groups containing from zero to four heteroatoms, for example, benzene, pyrrole, furan, thiophene, imidazole, oxazole, thiazole, triazole, pyrazole, pyridine, pyrazine, pyridazine and pyrimidine, and the like. Those aryl groups having heteroatoms in the ring structure may also be referred to as "aryl heterocycles", "heteroaromatics" or "heteroaryl groups". The aromatic ring can be substituted at one or more ring positions with such substituents as otherwise described herein, for example, halogen, azide, alkyl, aralkyl, alkenyl, alkynyl, cycloalkyl, polycyclyl, hydroxyl, alkoxyl, amino, nitro, sulfhydryl, imino, amido, phosphate, phosphonate, phosphinate, carbonyl, carboxyl, silyl, ether, alkylthio, sulfonyl, sulfonamido, ketone, aldehyde, ester, heterocyclyl, aromatic or heteroaromatic moieties, --CF₃, --CN, or the like. The term "aryl" also includes polycyclic ring systems having two or more cyclic rings in which two or more carbons are common to two adjoining rings (the rings are "fused rings") wherein at least one of the rings is aromatic, e.g., the other cyclic rings can be cycloalkyls, cycloalkenyls, cycloalkynyls, aryls and/or heterocyclyls.

The terms "heteroaryl" includes substituted or unsubstituted aromatic single or fused ring structures, preferably 5- to 14-membered rings whose ring structures include at least one heteroatom, preferably one to four heteroatoms, more preferably one or two heteroatoms. Heteroaryl groups include, for example, pyrrole, furan, thiophene, imidazole, oxazole, thiazole, pyrazole, pyridine, pyrazine, pyridazine, and pyrimidine, and the like.

Pyrrole, imidazole, pyrazole, pyridine, pyridazine, cinnoline, pyrimidine, diazine, triazine, indole, indoline, indolizine, quinoline, isoquinoline, quinoxaline, quinazoline, pteridine, quinolizidine, benzopyridine, benzoquinoline, perimidine, phenanthridine, acridine, phenazine, phenanthroline and carbazole are one group of heteroaryls that may be used in the invention.

The terms "halo" and "halogen" as used herein means halogen and includes chloro, fluoro, bromo, and iodo.

As used herein, the term "substituted" is contemplated to include all permissible substituents of organic compounds. In a broad aspect, the permissible substituents include acyclic and cyclic, branched and unbranched, carbocyclic and heterocyclic, aromatic, non- aromatic and inorganic substituents of organic compounds. The permissible substituents can be one or more and the same or different for appropriate organic compounds. For purposes of this invention, the heteroatoms such as nitrogen may have hydrogen substituents and/or any permissible substituents of organic compounds described herein which satisfy the valences of the heteroatoms. Substituents can include any substituents (groups) as otherwise described herein, for example, a halogen, a hydroxyl, a carbonyl (such as a carboxyl, an alkoxycarbonyl, a formyl, or an acyl), an ether, a thioether, a thiocarbonyl (such as a thioester, a thioacetate, or a thioformate), an alkoxyl, a phosphoryl, a phosphate, a

phosphonate, a phosphinate, an amino, an amido, an amidine, an imine, a cyano, a nitro, an azido, a sulfhydryl, an alkylthio, a sulfate, a sulfonate, a sulfamoyl, a sulfonamido, a sulfonyl, a heterocyclyl, an aralkyl, or an aromatic or heteroaromatic moiety. It will be understood by those skilled in the art that the moieties substituted on a moiety or chemical group can themselves be substituted.

It will be understood that "substitution" or "substituted with" includes the implicit proviso that such substitution is in accordance with permitted valence of the substituted atom and the substituent, and that the substitution results in a stable compound, e.g., which does not spontaneously undergo transformation such as by rearrangement, cyclization, elimination, etc. It is acknowledged that the term "unsubstituted" simply refers to a hydrogen substituent or no substituent within the context of the use of the term.

A liquid organic hydrogen carrier ("LOHC") as used herein means a partially or fully hydrogenated, nitrogen-containing aromatic heterocycle. Non-limiting examples of a partially or fully hydrogenated, nitrogen-containing aromatic heterocycle include decalin, 2- aminopyridine, 4-methylpyrimidine, dipyrimidinemethane, dimethyltetrazine, dipirimidine, diazacarbazole, alkylcarbazole, 4-aminopyridine, dipyrazinemethane, tripyrazinemethane, tripyrazineamine, dipyrazine, tetrazacarbazole, isoquinoline, di(2-pyridyl)amine, quinazoline, perhydro-N-ethylcarbazole, tetrahydroisoquinoline, perhydro-dibenzofuran, perhydro-indole, N-methyl perhydro-indole, 4, 4'-bipiperidine, 4-amino-methylpyrimidine, and

aminomethylcyclohexane, or a combination of two or more of the foregoing. A preferred liquid organic hydrogen carrier for use in the present invention is 1,2,3,4- tetrahydroquinaldine (l,2,3,4-Tetrahydro-2-methylquinoline), 2,3

dimethyltetrahydroquinoxaline or a mixture thereof.

A "direct organic fuel cell/flow battery" uses a hydrogen-rich organic liquid (LOHC) as a reversible fuel in an electrochemical cell. In this concept, in the discharge mode, the organic liquid is fed directly to the anode of a PEM fuel cell,

where it is electrochemically dehydrogenated to form a stable, hydrogen-depleted organic compound to produce power without ever generating gaseous H₂. During this dehydrogenatioin (energy releasing) process, nanoparticles of nickel oxide may be formed in solution and these nickel oxide nanoparticles may actually be performing or facilitating the chemistry observed during the dehydrogenation step. The depleted (dehydrogenated) LOHC can be electrohydrogenated in the presence of the redox catalyst using protons generated by electrooxidation of water at a cathode of the direct organic fuel cell/flow battery. In this charge mode, water is electrolyzed to hydrogenate the spent LOHC back to the starting LOHC compound.

The following is for the oxidative (Dehydrogenation^ Removal of hydrogen equivalents) direction which is in the preliminary stages

These reactions are represented by the following equations (where LQ^*H_n is the LOHC and LQ is the dehydrogenated LOHC):

LQ^*H_n→ LQ + nH⁺ + ne^"

n/2 0₂ + nH⁺→ n/2 H₂0 + ne^"

LQ*H_n + n/20₂→ n/2 H₂0

Thus, the system works as a hybrid of a fuel cell and a flow battery. For example, the LOHC decalin can be oxidatively dehydrogenated to naphthalene to produce 10 protons and 10 electrons that, taking into account decalin' s molecular weight of 128.17, makes it an anode material with specific energy density only

approximately two times less than lithium metal. The theoretical energy density of the organic liquid-oxygen couple may reach 1,350 Wh kg-1 , approximately 20-40 times higher than that of common flow batteries.

Operation of a "direct organic fuel cell/flow battery" is illustrated in Figure 11. A hydrogenated organic liquid carrier (LOHC) is fed to the anode of a direct organic fuel cell/flow battery where it is electrochemically dehydrogenated, generating electricity, while air oxygen is reduced at the cathode to water. During this dehydrogenation of the LOHC, nickel oxide nanoparticles may be formed in solution and/or as a solid on the anode and this nickel oxide (Ni0₂) in solution or as a deposited solid on the anode may partake in or facilitate the dehydrogenation reaction. To recharge the flow battery, the reactions are reversed and the organic liquid is electrochemically re-hydrogenated, or rapidly replaced with the hydrogenated form at a refueling station. System flexibility allows for usage in mobile and stationary energy applications, including renewable energy sources. This novel energy storage solution is safer than hydrogen fuel cells and lithium batteries.

Understanding of the electrodehydrogenation and electrohydrogenation catalysis is critical for realization of the organic fuel cell/flow battery concept. A good

electro(de)hydrogenation catalyst should combine an effective (de)hydrogenation activity with an ability to mediate transport of protons and electrons. Only a few examples of catalysts for electrohydrogenation and electrodehydrogenation of organic heterocycles are known. Those provided by the instant invention exhibit superior performance when used in direct-feed LOHC PEM fuel cell systems as described herein.

A PEM used in an organic fuel cell/flow battery must be substantially water-free, as water is detrimental to anode chemistry. Also, the PEM must exhibit low fuel and products solubility to preserve its mechanical integrity. Further, the PEM must exhibit a proton conductivity of 10^"3 S/cm at 120°C and thermal stability of greater than 150°C.

Examples of useful PEM's include, but are not limited to, polybenzimidazole (PBI) fiber membranes, sulfonated tetrafluoroethylene membranes such as DuPont's Nation®, and phosphoric acid-based membranes.

United States Patent Application Document No. 20100055513 describes generally apparatus that can be modified and adapted for use in the processes and systems described herein. For example, adapting and modifying the systems of United States Patent

Application Document No. 20100055513, an organic fuel cell/flow battery is in fluid communication with a source of a LOHC and an oxidant, typically air, purified oxygen or their mixture, as shown in Figure 1 1. The organic fuel cell/flow battery receives, catalyzes and electrochemically reduces at least a portion of the hydrogen contained in the LOHC to generate electricity, a hydrogen depleted liquid, and water. The hydrogen depleted liquid may include both fully hydrogen depleted liquids and partially hydrogen depleted liquids. While the appropriate organic liquid carrier of hydrogen will vary from system to system, the selection process will typically be based on criteria such as the hydrogen storage capacity of the carrier, the rate and the energy of dehydrogenation of the carrier, the boiling point of the carrier and the overall cost of the carrier. The organic fuel cell/flow battery comprises a Proton Exchange Membrane (PEM) fuel cell that includes a solid electrolyte that separates an anode portion and a cathode portion. The electro(de)hydrogenation catalyst is typically disposed on the anodic side of the solid electrolyte, for accelerating the reduction of the LOHC. The electro(de)hydrogenation catalyst may be affixed to high-surface-area conductive supports like carbon or conductive polymers. The catalyst may be anchored to the anode portion via formation of chemical bonds between the catalyst and the anode portion, for example, by using functionalized silanes or by adsorption on a ligand-modified surface. Also, the system may further comprise a catalyst material on the cathode portion to increase the electrochemical cell potential and improve the oxygen reduction reaction.

Also, in accordance with United States Patent Application Document No.

20100055513, a vessel (storage tank) can receive the hydrogen depleted LOHC; the source of the LOHC and the vessel for receiving the hydrogen depleted LOHC include a liquid storage unit that comprises a first compartment for the LOHC and a second compartment for hydrogen depleted LOHC. A fuel inlet line directs the LOHC to the direct organic fuel cell/flow battery. The hydrogen depleted LOHC is directed via a spent fuel outlet line from the direct organic fuel cell/flow battery to the vessel for receiving the hydrogen depleted LOHC.

A "pincer ligand" is a type of a chelating agent that can bind tightly to three adjacent coplanar sites, usually on a transition metal. Typical tridentate pincer type ligands have the general form D^D², wherein C is a carbon atom that can potentially interact with a metal; and D¹ and D² are groups containing coordinating atoms (also referred to herein as electron donating atoms). In most pincer ligands, the carbon atom forms a part of an aryl ring, typically phenyl. The carbon atom can be replaced by other coordinating atoms such as nitrogen or sulfur, which typically form a part of a heterocyclyl such as a heteroaryl. The reactivity, selectivity and catalytic performance of pincer-based systems greatly rely on the characteristics of the donor groups D in the carefully selected ligand. These characteristics depend on the type of the coordinating atom, and further on the nature of its organic substituents.

A wide variety of tridentate pincer type ligand catalysts, and syntheses for making such catalysts, are known. See e.g. Crabtree, Organometallics, 2011, 30 (1), pp 17-19; Gnanamgari and Crabtree, Organometallics, 2009, 28 (3), pp 922-924.

More specifically, anionic, homogenous tridentate pincer type ligand catalysts may consist of a metal center bonded to an anionic tridentate ligand with the general formula {2, 6-(ECH2)2C₆H₃}^", where E is a neutral two-electron donor (N, P, O, S). These compounds are very stable due to the formation of two five-membered metallocycles that provide additional stabilization of the carbon-metal bond. Although the most widely studied complexes are derivatives of the group 10 metals, Ni, Pt and Pd, containing N or P as donor atoms, a limited number of nickel complexes containing NCN pincer ligands have been reported. Hurtado, et al, J. Braz. Chem. Soc. vol.22 no.9 Sao Paulo Sept. 2011

The invention is described further in the following non-limiting examples.

Examples

Example 1

A tridentate Nickel pincer for aqueous electrocatalytic hydrogen production A Ni¹¹ complex with a redox-active pincer ligand reduces protons at a low overpotential in aqueous acidic conditions. A combined experimental and computational study provides mechanistic insights into a putative catalytic cycle.

Here we report that a Ni pincer gives good activity as an operationally homogeneous electrocatalyst for proton reduction in aqueous conditions. Electrochemistry in acetonitrile was used to pinpoint redox events of the metal complex in the presence of added acid and subsequently aqueous conditions. DFT calculations offer insight into a possible mechanism.

1

Scheme 1. Catalyst 1.

Results and Discussion

One electron reduction of catalyst 1 is known to give a ligand-centered reduction of the NNN pincer ligand,¹⁴ as shown by EPR data. In inital voltammetry in MeCN, we observe a first reduction wave just above 0 V for complex 1 (Figure 1) that we assign a ligand- centered process given the literature precedent. The second reduction at ~ -0.5 V vs. NHE is therefore assigned as a N^/Ni¹ couple.

To the precatalyst in acetonitrile increasing amounts of acid were added and an increase in current response was observed by cyclic voltammetry (ESI: SI -2). This allowed us to apply the voltammetric kinetic treatment of DuBois et a/.¹⁵ (S7) that leads to kinetic parameters for the H₂ evolution reaction. Like the DuBois case, our rate law follows Eq. 1 and an apparent rate constant of 1.05 (±0.21) x 10⁴ has been determined for 0.1 M H⁺, 5 mM 1 at -250 mV vs. NHE. This corresponds to a voltammetric rate of hydrogen formation of 105

rate = k[H⁺ f [cat] (1)

Further information was obtained at variable potential where we find sustained H₂ evolution in a series of chronoamperometry experiments. A plot of current density vs. overpotential (Fig. S3-1) shows that the overpotential for 1 is a very satisfactory 220 mV at a current density of 1 mA cm^" .

The compound can also operate in water. A high surface area reticulated vitreous carbon working electrode was used to determine quantitative H₂ evolution from 1 via mass spectroscopy (see ESI). Specifically, a 50 mL 0.1 M KC1/HC1 solution (pH 1) containing 0.2 mM precatalyst was held at -1.1 V vs. NHE for one hour. After subtraction of the relevant background current, 132 C were consumed by the catalytic chemistry. This is equivalent to a predicted production of 0.68 millimoles of H₂. From quantitative mass spectrometry, 0.65 millimoles of H₂ were detected, also after background subtraction. Thus, our Faradaic efficiency for H₂ production under these conditions was 95±4 %. These data correspond to a minimum of 65 mole H₂ per mole of Ni per hour (1456L H₂ produced per mol of Ni catalyst per h). Only a small amount of the soluble catalyst is electroactive at any one time and thus the observed rate is a lower limit for the absolute activity.

In order to propose a mechanism of H₂ evolution using complex 1 , we performed ab initio calculations starting with the aqua complex of Fig. 2. As in previous studies,¹⁷ we applied DFT B3LYP to characterize the structural and spin/electronic properties of the reaction intermediates and performed free energy calculations. Gas phase free-energy changes were calculated at the B3LYP/cc-pVTZ level, using minimum energy structures obtained at the DFT B3LYP/LANL2DZ level of theory, and then corrected by solvation free- energy calculations with the LANL2DZ basis set, using the Polarizable Continuum Model (ε = 78.4 for water) as implemented in Gaussian 09.¹⁹

Consistent with our proposal, the calculations predict that in aqueous conditions, 1 may readily lose one Br^" to form a square planar complex with a single Br^" ligand (AG for ligand loss is -4.7 kcal mol^"1). Subsequently, the Br^" ligand can easily exchange with water (AG for replacement of Br^" with water is -6.3 kcal mol^"1). This is consistent with experimental results indicating that dissolution of 1 in <¾-DMSO, an even better O-donor ligand than H₂0, results in the formation of a diamagnetic species, most likely a 4-coordinate square planar dication with coordinated DMSO (see the Figure S6 in the ESI for the NMR spectrum). Given that the catalytic experiments are performed in water, the DFT results suggest that the catalytic cycle begins with a square planar Ni¹¹ aqua complex and that 1 is only a precatalyst.

The first step in the catalytic cycle (Figure 2) could involve either protonation of the Ni¹¹ starting material to give a formally Ni^IV hydride or reduction of [l'(OH₂)]²⁺. Both experimental and theoretical data suggest that reduction can be thermodynamically favored over protonation to the unfavorable Ni^IV oxidation state. The lowest energy reduction product is the square planar, water-ligated, complex [l'(OH₂)]⁺,^lu with high-spin density on the pincer ligand (as shown by the spin-distribution in Figure 3). However, our DFT calculations indicate that an isomer of [l'(OH₂)]⁺ with the unpaired electron on the Ni is close in energy to the isomer with the ligand centered radical and within the errors of DFT we cannot differentiate between the two. EPR measurements of the one electron reduced species¹³ provide supporting evidence that the first reduction process is ligand centered.

Once [l'(OH₂)]⁺ has been formed, three main pathways are available: (a) further reduction to give a Ni¹ species with a ligand-centered radical, (b) protonation of [l'(OH₂)]⁺ to give either a Ni^IV hydride with a ligand-centered radical or a Ni¹¹¹ hydride with a neutral ligand, or (c) PCET to give a Ni¹¹ hydride with a neutral ligand. Our DFT and experimental results suggest that the second reduction of 1 to give a Ni¹ species and a ligand-centered radical is energetically disfavored and requires a significantly greater reduction potential than that needed during catalysis (vide supra); pathway (a) has therefore been dismissed.

Protonation of [Γ(ΟΗ₂)]⁺ could result in the formation of two different Ni hydride isomers, cw-[l'(OH₂)₂H]²⁺ or ί/"αη£-[1'(ΟΗ₂)₂Η]²⁺ that differ in the coordination geometry around Ni. In cw-[ (OH₂)₂H]²⁺, the hydride is coplanar with the NNN ligand with two axially-bound water ligands (which stabilize the higher oxidation state), while in /ταπ£-[Γ(ΟΗ₂)₂Η]²⁺ the hydride is perpendicular to the NNN ligand. The spin density analysis indicates that both cis- [l'(OH₂)₂H]²⁺ and traw-[l'(OH₂)₂H]²⁺ are formally Ni^m hydrides (see ESI for more information), suggesting that a Ni^IV species with a ligand-centered radical is not accessible. However, protonation to form w-[l'(OH₂)₂H]²⁺ or trans-[V(OYLi)_TP\²⁺ is significantly uphill energetically, (AG_aq([l'(OH₂)]⁺/ cis-\Y (ORi)₂nf⁺) - 40 kcal mol^"1 (1.73 eV) and AG_aq([l'(OH₂)]⁺/tmw-[l^,(OH₂)₂H]²⁺) - 26 kcal mol^"1 (1.13 eV)), and is not spontaneous. On this basis, we propose that the PCET pathway (c) is the most likely route that converts

[l'(OH₂)]⁺ into the square planar Ni° hydride [l'(H)]⁺. The free energy requirement for

[Γ(ΟΗ₂)]⁺→[Γ(Η)]⁺ conversion thus includes (1.13 + 0.059 pH) eV to protonate [l'(OH₂)]⁺ to give tra?w-[l'(OH₂)₂H] minus the excess free energy (4.60 eV) due to reduction of trans- [l'(OH₂)₂H]²⁺ into [l'(H)]⁺. Therefore, by coupling protonation and reduction in PCET, [l'(OH₂)]⁺ is converted into [l'(H)]⁺ at a potential (1.13 + 0.059 pH - 3.60) V. This very large driving force force²¹ of AG_aq([r(OH₂)]⁺/[r(H)]⁺ = -80 kcal mol^-1 (-3.47 V vs NHE) at pH = 0 is consistent with our measured KIE (kn ko) of 4.2(1).

The catalytic cycle is completed by protonation of [l'(H)]⁺ to give the Ni¹¹ dihydrogen complex [Γ(Η₂)]²⁺) (AG_aq([l'(H)]⁺/ [l'(H₂)]²⁺) = 9 kcal mol^"1) and subsequent H₂ evolution, with the H₂ ligand substituted by water to regenerate [l'(OH₂)]²⁺. Notably, the highly reactive nature of the radical intermediates proposed to intervene in the cycle has been previously documented: a species related to the one-electron reduced catalyst precursor 1 has been shown to undergo oxidative demethylation upon exposure to air.¹³ Our experiments were performed under rigorous Ar purge and we believe that such a degradation pathway is only accessible in the presence of adventitious oxidizing species, such as aerial dioxygen.*

Conclusions

Our Ni¹¹ pincer complex is an excellent water reduction precatalyst in aqueous acid solutions. This seems to be the first report of such a pincer complex being an operationally homogeneous catalyst for H⁺ reduction. Precatalyst 1 incorporates a redox-active ligand, a factor that could be important in facilitating the catalytic process, and shows a low

overpotential for H₂ production with a rate of 105 s^"1. Bulk electrolysis followed by macroscopic determination of the H₂ produced, demonstrated that complex 1 gives at least 65 mole H₂ per mole of catalyst per hour (-1.1 V vs. NHE in 50 mL 0.1 M KCl/HCl solution pH 1 with 0.2 mmol catalyst). However, a reliable comparison of H₂ evolution measurements across different systems is difficult because of the wide array of experimental conditions used in proton reduction.¹¹ Factors that change between systems include experimental setups, electrolysis potentials and choice of solvent and proton source. DFT studies suggest that the mechanism of proton reduction for 1 involves a key PCET step. Further mechanistic work is under way to investigate the proposed intermediates and to tune the properties of the pincer ligand to enhance the kinetics of proton reduction.

Additional experimental details are as follows.

1. Cyclic Voltammetry:

General experimental

1.. Cyclic voltammo grams (CVs) in aqueous solution were collected using a 0.09 cm² basal plane graphite working electrode prepared by the method of Blakemore et ah, ^1SI a platinum wire counter electrode and an aqueous Ag/AgCl, KC1 (sat) reference. In acetonitrile, a glassy carbon electrode was used as the working electrode and a Ag wire reference (referenced vs. NHE with ferrocene as external standard). For the aqueous electrochemistry 3 mL 0.1 M KC1 aqueous solutions were used, with incremental amounts of acid added (1 M HC1, 40, 90, 140 \iL). Basal plane graphite electrodes were preferred over glassy carbon electrodes in aqueous conditions due to low background currents in the absence of catalyst. Background CVs are provided in sections 1 and 2. CVs were recorded after addition of 4* 10 \xL 1 M HC1 via volumetric syringe. All other CVs were recorded after rigorous exclusion of air via Argon purge. Data workup was performed on OriginPro v8.0988 and AfterMath Data Organizer Version 1.2.3383. Kinetic isotope studies studies were carried out with a Pine AFCBP1 Bipotentiostat and a Pine MSR variable-speed rotator. The reference was a Ag/AgCl electrode (Bioanalytical Systems, Inc.) and the counter electrode was a platinum wire. The disc (surface area: 0.07 cm²) material was basal plane graphite (with stabilizing resin around the graphite; from Pine). The disc was assembled in a Pine E6-series

ChangeDisk Setup. The disk voltage was held at 0.5 V vs NHE for 30 min. Data were collected in 12 mL of a 0.1 M KC1 solution containing 1 mM catalyst 1 and 20% HC1 or 20% DC1 (procured from Cambridge Isotopes) at 500 rpm.

Tafel Plots were constructured from short chronoamperometry experiments (dwell time: 60 seconds) at potentials lower than 0 V (vs NHE) at a glassy carbon electrode in a single-chamber, three-electrode configuration. The experiments were performed in 5.2 mL of a 0.1 M NBu₄BF₄ acetonitrile solution containing 0.2 mM catalyst and 200 μΐ of a 1 M aqueous HC1 solution. Magnetic stirring was used to avoid diffusion limitations from concentration gradients at the working electrode.

The order of reaction with respect to acid was determined by analysis of cyclic voltammograms of 2 mM catalyst solutions with different acid concentrations. To a 0.1 M NBU₄BF₄ acetonitrile solution containing 2 mM catalyst (from the CVs of which i_p was determined at different scan rates), 5 μΐ, increments of an aqueous 1 M HC1 solution (also containing 0.1 M NBU4BF₄) were added. The voltammograms thus collected at each acid concentration (for several scan rates) were used to obtain catalytic currents (denoted i_c). The ratio of i_c/i_p was plotted against acid concentration for the different scan rates. From the linear behavior at each scan rate, we conclude that our system obeys a rate law that is second order in acid. The slopes of the scan rate-dependent data were then used to calculate the third order rate constants.

The order in catalyst was determined from voltammograms collected at 100 mV/s with 3.5 mL of an acetonitrile 0.1 M NBu₄BF₄ solution with a concentration of 2.82 mM catalyst and 4 mM acid; Increments of 0.5 mL of a 0.1 M NBU₄BF₄ acetonitrile solution were added and the dilutions were adjusted to maintain a constant acid concentration. We assumed no large variations in the acid concentration throughout the experiment. From each voltammogram, i_c was plotted against catalyst concentration. From the linearity of the resulting graph we conclude that at reasonably low catalyst concentration first order behavior is observed. 6 Controlled potential headspace ¾ detection experiments were performed in a custom built two cylinder 50 mL bulk electrolysis H cell anode/cathode chamber separated by a coarse frit. The working electrode was a BASi RVC electrode referenced vs. Ag/AgCl (KCl_sat). Headspace H₂ detection was performed at the Yale Department of Geology on a calibrated mass spectrometer: dual inlet Thermo Finnegan MDT 253 and an air-tight bulk electrolysis H Cell equipped with a sampling port. 1 mL volumes of gas were compressed in the bellow to 10% then opened to the mass spectrometer.

Cyclic Voltammetry in nonaqueous conditions:

Table S I . Tabulated cyclic voltammetry data of a 2 mM nickel pincer complex (1) in 0.1 M Bu4BF₄ acetonitrile solution at a glassy

Table SI. Tabulated cyclic voltammetry data of 2 mM nickel complexes (1-3) in 0.1 M NBu₄BF₄ acetonitrile solutions at a glassy carbon working electrode.

y

Cyclic voltammograms were recorded using a Teflon coated BASi glassy carbon working electrode, and a Platinum wire counter electrode in an 0.1 M NBu₄BF acetonitrile solution versus a pseudoreference electrode: silver wire (BASi double junction reference electrode setup) referenced externally vs. the Fc/Fc⁺ couple at 690 mV vs. NHE. All CVs were recorded after rigorous Argon purge. See Figures Sl-l-Sl-10.

2. Current density vs. potential:

Figure S2-1 shows current density vs. applied potential: steady state current density at the end of chronoamperometry experiments at progressively more negative potentials - chronoamperograms (dwell time: 60 sec) at a glassy carbon electrode (in a three electrode setup- Pt counter electrode, non-aqueous Ag pseudoreference) in 5.2 mL acetonitrile NBu₄BF₄ solutions containing 200 1 M aqueous HCl at catalyst concentrations of 0.2 mM with magnetic stirring.

3. Bulk electrolyses and H₂ measurement: Controlled potential experiments for headspace H₂ detection were performed at -0.6V vs. NHE (0.5 h) in a custom built two cylinder bulk electrolysis H cell: The cathode chamber had a working volume of 50 mL (0.04 mM in the respective catalyst for the catalytic runs) and 20 mL for the cathode chamber. The two were separated by a coarse frit.

A stock solution was prepared from 900 mL MeCN (0.1 M NBu₄BF₄) and 100 mL 1 M HC1. In the catalytic run the cathode was charged with 50 mL stock solution and catalyst and the anode with 20 mL of the same solution. The cathode and anode chamber solutions were sparged with He for 5 min prior to starting the experiment.

Background runs were performed after the catalytic runs with the cathode containing 50 mL of the stock solution with no catalyst and the anode 20 mL of the stock solution. The working electrode was a BASi RVC electrode referenced vs. Ag/AgCl ( Cl_sat). The counter electrode was a 2.5 cm x 2.5 cm Pt mesh.

Quantitative Mass Spectrometry calibration of voltage response against H₂ detection was performed using 2.05%, 15%, 25% and 50% H₂/He custom prepared mixed gases from Tech Air and AirProducts Inc. The average of duplicate catalytic run analyses are shown in Figure S3-1.

4. Crystal Structure Data for [3(MeCN)]⁺ :

The diffraction experiments were carried out on a Bruker AXS SMART CCD three-circle diffractometer with a sealed tube at 23 °C using graphite-monochromated Mo KR radiation (A) 0.71073 A). The software used were SMART for collecting frames of data, indexing reflections, and determination of lattice parameters; SAINT for integration of intensity of reflections and scaling; SADABS for empirical absorption correction; and SHELXTL for space group determination, structure solution, and least-squares refinements on \F\2. The crystals were mounted at the end of glass fibers and used for the diffraction experiments. Anisotropic thermal parameters were refined for the rest of the non-hydrogen atoms. The hydrogen atoms were placed in their ideal positions. Table S4-1. Crystal data and structure refinement for [(PCP)Ni(NCCH₃)]⁺ [BF₄]^"

Empirical Formula P₂N₂C₂₈H₄₉NiBF₄

Formula Weight 621.16

Crystal Color, Habit yellow, chunk

Crystal Dimensions 0.20 X 0.20 X 0.20 mm

Crystal System monoclinic

Lattice Type Primitive

Detector Position 49.90 mm

Pixel Size 0.146 mm

Lattice Parameters a = 10.4927(18) A

b = 15.606(3) A

c = 10.6505(18) A

β = 1 11.242(4) °

V = 1625.5(5) A³

Space Group P2j (#4)

Z value 2

Dcalc 1.269 g/cm³

FOOO 660.00

μ(ΜοΚα) 7.376 cm^"1

Data Images 462 exposures

ω oscillation Range (χ=54.0, φ=120.0) -120.0 - 60.0°

Exposure Rate 60.0 sec./⁰

Detector Swing Angle -28.40°

52.0°

No. of Reflections Measured Total: 12989

Unique: 6348 (R_mt = 0.0485)

Friedel pairs: 3023

Corrections Lorentz-polarization

Absorption

(trans, factors: 0.697 - 0.863)

Structure Solution Direct Methods (SIR92)

Refinement Full-matrix least-squares on F²

Function Minimized ∑w (Fo² - Fc²)²

Least Squares Weights w = 1/ [a²(Fo²) + (0.0616 · P)²

+ 0.0000 · P ]

where P = (Max(Fo²,0) + 2Fc²)/3

Anomalous Dispersion All non-hydrogen atoms No. Observations (All reflections) 6348

No. Variables 343

Reflection/Parameter Ratio 18.51

Residuals: Rl (Ι>2.00σΙ)) 0.0581

Residuals: R (All reflections) 0.0768

Residuals: wR2 (All reflections) 0.1264

Goodness of Fit Indicator 1.047

Flack Parameter -0.003(18)

Max Shift/Error in Final Cycle 0.001

Maximum peak in Final Diff. Map 0.48 eVA³

Minimum peak in Final Diff. Map -0.34 eVA³

Table S4-2. Atomic coordinates and B;_so/B,

atom x y z B_eq

Ni(l) 0.71359(5) 0.89352(4) 0.28474(5) 2.195(12)

P(l) 0.51687(12) 0.82594(8) 0.19220(1 1) 2.32(2)

P(2) 0.89869(12) 0.96336(8) 0.41141(12) 2.28(2)

F(l) 0.8801(6) 0.6468(3) -0.0275(5) 9.92(16)

F(2) 1.0539(8) 0.5636(5) 0.0823(6) 15.1(3) F(3) 1.0429(9) 0.6890(5) 0.1539(9) 19.0(4)

F(4) 0.9145(8) 0.5935(9) 0.1655(9) 21.5(5)

N(l) 0.7772(4) 0.8710(3) 0.1421(4) 2.96(8)

N(2) 0.6183(7) 1.2028(5) 0.3541(7) 7.03(15)

C(l) 0.6466(5) 0.9221(3) 0.4244(5) 2.60(9)

C(2) 0.7124(5) 0.9841(3) 0.5239(5) 2.66(9)

C(3) 0.6631(6) 1.0033(4) 0.6264(5) 3.85(11)

C(4) 0.5478(6) 0.9641(4) 0.6308(6) 4.19(11)

C(5) 0.4817(5) 0.9029(4) 0.5361(5) 3.44(10)

C(6) 0.5293(4) 0.8822(4) 0.4334(4) 2.75(9)

C(7) 0.8113(5) 0.8633(3) 0.0543(5) 3.31(10)

C(8) 0.8577(7) 0.8559(5) -0.0606(6) 5.12(15)

C(9) 0.4557(5) 0.8152(3) 0.3318(5) 3.01(9)

C(10) 0.8350(5) 1.0285(3) 0.5187(5) 3.21(10)

C(l l) 0.3869(4) 0.8935(4) 0.0641(4) 2.96(8)

C(12) 0.3685(6) 0.9747(4) 0.1368(6) 4.83(13)

C(13) 0.4403(6) 0.9223(4) -0.0449(6) 4.46(13)

C(14) 0.2473(5) 0.8498(4) -0.0017(6) 5.20(15)

C(15) 0.5239(6) 0.7125(3) 0.1363(5) 3.32(10)

C(16) 0.3963(6) 0.6606(4) 0.1282(7) 4.64(13)

C(17) 0.5380(7) 0.7091(4) -0.0019(6) 4.72(13)

C(18) 0.6508(6) 0.6736(4) 0.2418(6) 4.59(13)

C(19) 1.0265(4) 0.8890(4) 0.5252(4) 2.99(8)

C(20) 0.9626(6) 0.8482(4) 0.6198(5) 3.92(11)

C(21) 1.0606(5) 0.8176(3) 0.4414(5) 3.73(11)

C(22) 1.1603(5) 0.9331(4) 0.6154(5) 3.83(11)

C(23) 0.9778(5) 1.0425(3) 0.3276(5) 3.27(10)

C(24) 1.0710(5) 0.9993(4) 0.2641(6) 3.57(11)

C(25) 1.0622(6) 1.1102(4) 0.4264(7) 4.57(13)

C(26) 0.8567(6) 1.0852(4) 0.2171(6) 4.31(12)

C(27) 0.5153(7) 1.1846(5) 0.3509(6) 4.68(13)

C(28) 0.3800(8) 1.1604(6) 0.3473(8) 7.4(2)

B(l) 0.9733(9) 0.6262(6) 0.0886(8) 4.84(16)

B_eq = 8/3 7t²(U„(aa*)² + U₂₂(bb*)² + U₃₃(cc*)² + 2U₁₂(aa*bb*)cos γ + 2U₁₃(aa*cc*)cos β + 2U₂₃(bb*cc*)cos a) Table S4-3. Anisotropic displacement parameters

atom Ul l U22 U33 Ul2 Ul3 U23

Ni(l) 0.0236(3) 0.0343(3) 0.0253(3) -0.0015(3) 0.0086(2) -0.0035(3)

P(l) 0.0255(6) 0.0327(7) 0.0271(6) -0.0024(5) 0.0062(5) 0.0002(5)

P(2) 0.0245(6) 0.0305(7) 0.0305(6) -0.0011(5) 0.0087(5) -0.0036(5)

F(l) 0.153(5) 0.081(3) 0.086(3) 0.030(3) -0.025(3) 0.007(3)

F(2) 0.211(8) 0.204(7) 0.115(5) 0.132(7) 0.004(5) -0.020(5)

F(3) 0.245(9) 0.158(7) 0.200(8) -0.014(7) -0.061(7) -0.056(6)

F(4) 0.150(7) 0.488(20) 0.197(8) 0.017(9) 0.083(7) 0.168(1 1)

N(l) 0.035(2) 0.046(3) 0.029(2) -0.0041(20) 0.0091(18) -0.0078(18)

N(2) 0.071(4) 0.094(5) 0.100(5) -0.010(4) 0.028(4) -0.020(4)

C(l) 0.029(2) 0.042(3) 0.031(2) 0.008(2) 0.013(2) -0.0041(20)

C(2) 0.032(2) 0.035(3) 0.036(3) 0.001(2) 0.015(2) -0.005(2)

C(3) 0.051(3) 0.055(3) 0.043(3) 0.003(3) 0.021(3) -0.014(3)

C(4) 0.059(3) 0.062(4) 0.052(3) -0.006(3) 0.035(3) -0.011(3)

C(5) 0.045(3) 0.050(3) 0.045(3) 0.004(3) 0.027(2) -0.003(3)

C(6) 0.034(2) 0.041(3) 0.030(2) 0.006(3) 0.0128(19) 0.000(2)

C(7) 0.038(3) 0.051(3) 0.041(3) -0.003(2) 0.019(2) -0.004(2)

C(8) 0.082(5) 0.082(5) 0.045(3) -0.002(4) 0.039(3) -0.006(3)

C(9) 0.035(3) 0.045(3) 0.035(3) -0.004(3) 0.014(2) 0.005(2)

C(10) 0.041(3) 0.037(3) 0.041(3) -0.001(2) 0.011(2) -0.012(2)

C(l l) 0.027(2) 0.038(2) 0.041(2) 0.004(3) 0.0058(19) 0.004(3)

C(12) 0.048(3) 0.056(4) 0.075(4) 0.019(3) 0.015(3) 0.016(3) C(13) 0.045(3) 0.068(4) 0.047(3) -0.002(3) 0.005(3) 0.020(3)

C(14) 0.039(3) 0.081(5) 0.057(4) -0.010(3) -0.008(3) 0.018(3)

C(15) 0.054(3) 0.035(3) 0.033(3) -0.005(3) 0.010(2) -0.007(2)

C(16) 0.062(4) 0.043(3) 0.068(4) -0.013(3) 0.020(3) 0.001(3)

C(17) 0.080(5) 0.045(3) 0.062(4) -0.013(3) 0.035(4) -0.009(3)

C(18) 0.054(3) 0.041(3) 0.071(4) 0.008(3) 0.013(3) -0.002(3)

C(19) 0.031(2) 0.041(3) 0.032(2) 0.002(3) 0.0001(18) 0.008(3)

C(20) 0.048(3) 0.050(3) 0.044(3) 0.002(3) 0.009(3) 0.009(3)

C(21) 0.041(3) 0.042(3) 0.054(3) 0.013(3) 0.010(3) 0.000(3)

C(22) 0.030(3) 0.059(3) 0.051(3) -0.001(3) 0.008(2) -0.008(3)

C(23) 0.036(3) 0.035(3) 0.054(3) -0.003(2) 0.017(3) 0.005(2)

C(24) 0.036(3) 0.056(4) 0.053(3) -0.008(3) 0.026(3) 0.002(3)

C(25) 0.051(3) 0.036(3) 0.093(5) -0.012(3) 0.034(4) -0.007(3)

C(26) 0.057(4) 0.045(3) 0.066(4) 0.007(3) 0.028(3) 0.024(3)

C(27) 0.048(4) 0.071(4) 0.058(4) 0.007(3) 0.017(3) -0.009(3)

C(28) 0.078(5) 0.120(7) 0.084(6) -0.019(5) 0.030(5) -0.033(5)

B(l) 0.061(4) 0.070(5) 0.043(4) 0.006(4) 0.007(4) -0.012(4)

The general temperature factor expression: exp(-2ji²(a*²Ui ih² + b*²U22k² + c*²U33l + 2a*b*Ui2hk

2a*c*Ui3hl + 2b*c*U23kI))

Table S4-4. Bond lengths (A)

atom atom distance atom atom distance

Ni(l) P(l) 2.2068(13) Ni(l) P(2) 2.2094(12^'

Ni(l) N(l) 1.901(5) Ni(l) C(l) 1.914(6)

P(l) C(9) 1.828(6) P(l) C(l l) 1.867(5)

P(l) C(15) 1.879(5) P(2) C(10) 1.828(6)

P(2) C(19) 1.853(5) P(2) C(23) 1.884(6)

F(l) B(l) 1.309(8) F(2) B(l) 1.309(12)

F(3) B(l) 1.269(11) F(4) B(l) 1.295(15)

N(l) C(7) 1.122(8) N(2) C(27) 1.106(11)

C(l) C(2) 1.417(6) C(l) C(6) 1.413(7)

C(2) C(3) 1.399(9) C(2) C(10) 1.480(8)

C(3) C(4) 1.370(9) C(4) C(5) 1.380(8)

C(5) C(6) 1.395(8) C(6) C(9) 1.502(7)

C(7) C(8) 1.476(9) C(U) C(12) 1.533(9)

C(l l) C(13) 1.528(9) C(l l) C(14) 1.536(7)

C(15) C(16) 1.540(9) C(15) C(17) 1.532(9)

C(15) C(18) 1.522(7) C(19) C(20) 1.536(8)

C(19) C(21) 1.549(8) C(19) C(22) 1.546(6)

C(23) C(24) 1.532(9) C(23) C(25) 1.527(7)

C(23) C(26) 1.536(7) C(27) C(28) 1.456(11)

Table S4-5. Bond angles (°)

atom atom atom angle atom atom atom angle

P(l) Ni(l) P(2) 168.96(6) P(l) Ni(l) N(l) 95.57(12)

P(l) Ni(l) C(l) 84.88(14) P(2) Ni(l) N(l) 95.39(12)

P(2) Ni(l) C(l) 84.25(14) N(l) Ni(l) C(l) 176.97(18)

Ni(l) P(l) C(9) 103.17(15) Ni(l) P(l) C(l l) 11 1.94(18)

Ni(l) P(l) C(15) 117.02(18) C(9) P(l) C(l l) 105.5(2)

C(9) P(l) C(15) 103.7(3) C(l l) P(l) C(15) 113.8(2)

Ni(l) P(2) C(10) 102.50(16) Ni(l) P(2) C(19) 110.98(18)

Ni(l) P(2) C(23) 118.36(15) C(10) P(2) C(19) 106.3(2)

C(10) P(2) C(23) 104.8(3) C(19) P(2) C(23) 112.4(2)

Ni(l) N(l) C(7) 175.0(4) Ni(l) C(l) C(2) 121.6(4)

Ni(l) C(l) C(6) 121.6(3) C(2) C(l) C(6) 116.8(5)

C(l) C(2) C(3) 120.8(5) C(l) C(2) C(10) 119.1(5)

C(3) C(2) C(10) 120.1(4) C(2) C(3) C(4) 120.7(5)

C(3) C(4) C(5) 120.1(6) C(4) C(5) C(6) 120.3(5)

C(l) C(6) C(5) 121.3(4) C(l) C(6) C(9) 119.4(5)

C(5) C(6) C(9) 119.2(5) N(l) C(7) C(8) 178.3(6)

P(l) C(9) C(6) 106.3(4) P(2) C(10) C(2) 106.4(4)

P(l) C(l l) C(12) 106.9(3) P(l) C(l l) C(13) 110.5(3)

P(l) C(l l) C(14) 113.9(4) C(12) C(l l) C(13) 106.8(5)

C(12) C(l l) C(1 ) 108.9(5) C(13) C(l l) C(14) 109.6(4)

P(l) C(15) C(16) 112.3(4) P(l) C(15) C(17) 11 1.4(4)

P(l) C(15) C(18) 105.4(3) C(16) C(15) C(17) 108.2(4)

C(16) C(15) C(18) 110.0(5) C(17) C(15) C(18) 109.4(5)

P(2) C(19) C(20) 107.7(3) P(2) C(19) C(21) 109.7(3)

P(2) C(19) C(22) 114.0(4) C(20) C(19) C(21) 109.0(5)

C(20) C(19) C(22) 106.9(4) C(21) C(19) C(22) 109.3(4)

P(2) C(23) C(24) 112.4(4) P(2) C(23) C(25) 111.9(4)

P(2) C(23) C(26) 105.2(4) C(24) C(23) C(25) 107.7(5)

C(24) C(23) C(26) 109.6(5) C(25) C(23) C(26) 109.9(4)

N(2) C(27) C(28) 179.7(8) F(l) B(l) F(2) 114.2(7)

F(l) B(l) F(3) 114.3(8) F(l) B(l) F(4) 109.3(8)

F(2) B(l) F(3) 110.5(8) F(2) B(l) F(4) 101.8(9)

F(3) B(l) F(4) 105.6(9)

5. Kinetics for proton reduction:

Method for EC_catrate determination obtained from DuBois et al. a. Pool, D. H.; DuBois, D. L. J. Organomet. Chem., 2009, 694, 2858-2865. b. Wilson, A. D.; Newell, R. H.; McNevin, M. J.; Muckerman, J. T.; Rakowski DuBois, M.; DuBois, D. L. J. Am. Chem. Soc, 2006, 128, 358-366.

For a diffusion limited catalytic process that occurs at high enough [H⁺] that the concentration remains unchanged, the observed current obeys the following equation:

i_c = nFA [cat]^Dk[H⁺]² _(i) For a reversible one e^" wave, the current observed can be expressed as:

FvD

0A43FA [catU

V RT Dividing (1) by (2) the following expression is obtained:

A = area of the electrode, D is the diffusion coefficient of the catalyst, n = 2 for H₂ production , R = 8.314 J/(mol K), F = 96485 C/mol, υ scan rate in V/s, k is the third order rate constant. Linearity of :

1. plots ofiji_p vs acid concentration confirms the electrocatalytic process is second

order in acid

2. plots ofi_c vs [catalyst] confirms the process is first order in catalyst

The rate law for the third order process is derived as:

rate = k[H⁺ ]² [cat]

Where I_p did not correspond with the onset of catalysis ,the metal centered reduction peak current at the respective scan rate was taken as I_p . See Figures S5-1-S5-9 and Table S5-1.

Table S5-1

Tabulated kinetic data for catalysts 1-3.

Numeric

Error % error on TOF Rate Potential at % error

Catalyst k (IffV) Slope error on

value k is^"1) (M s ¹)" 1mA cm^"2 on slope slope

1 1.55*10* ±2.3₅* 10³ ■ 15.84 155 0.775 -0.273V 89.5 7.09 7.92

0.546* 10

2 ±0.327* 10³ 6.05 54.6 0.275 -0.397V 53.23 1.61 3.02

3 2.09* 10" ±1.7* 10³ 8.31 209 1.045 -0.424V 122.74 5.1 4.15 a. calculated for 0.1 M H⁺, 5 mM catalyst b. as determined by plots of Current density vs. Potential constructed from a series of 60 s chronoamperometry experiments at progressively more negative potentials.

References for Background of the Invention and Example 1

¹ Pena, M. A.; Gomez, P. P.; Fierro, J. L. G. Appl. Catal. A: Gen. 1996, 144, 7-57.

² Navarro, R. M; Pena, M. A.; Fierro, J. L. G. Chem. Rev. 2007, 107, 3952-3991.

3Artero, V.; Fontecave, M. Coord. Chem. Rev. 2005, 249, 1518-1535.

⁴ Vincent, K. A.; Parkin, A.; Armstrong, F. A. Chem. Rev. 2007, 107, 4366-4413.

⁵ (a) Bockris, J. O'M.; Conway, B. E. Trans. Faraday Soc, 1949, 45, 989-999. (b) Ezaki, H.; Morinaga, M.; S. Watanabe Electrochimica Acta, 1993, 38, 557-564. ⁶ (a) Koelle, U.; Ohst, S. Inorg. Chem. 1986, 25, 2689. (b) Connolly, P.; Espenson, J. H.

Inorg. Chem. 1986, 25, 2684-2688. (c) Chao, T-H.; Espenson, J.H. J. Am. Chem. Soc , 1978, 100, 129-133. (d) Kellett, R. M.; Spiro, T. G. Inorg. Chem. 1985, 24, 2373-2377. (e) Hu X.; Brunschwig B. S.; Peters J. C. J. Am. Chem. Soc. 2007, 129, 8988-8998. (f) Szymczak, N. K.; Berben, L. A.; Peters, J. C. Chem. Comm. 2009, 6729-6731. (g) Berben, L. A.; Peters, J.C. Chem. Comm. 2010, 398-400. (h) Baffert, C; Artero, V.; Fontecave, M. Inorg. Chem., 2007, 46, 1817-1824. (i) Bernhardt, P. V.; Jones, L. A. Inorg. Chem. 1999, 38, 5086-5090. (j) Pantani, O., Naskar, S., Guillot, R., Millet, P., Anxolabehere-Mallart, E. and Aukauloo, A. Angew. Chem. Intl. Ed. 2008, 47, 9948-9950. (k) Razavet, M.; Artero, V.; Fontecave, M. Inorg. Chem, 2005, 44, 4786-4795. (1) Fihri, A.; Artero, V.; Razavet, M; Baffert, C; Leibl, W.; Fontecave, M. Angew. Chem. Intl. Ed. 2008, 47, 564-567. (m) Dempsey, J. L.;

Brunschwig, B. S.; Winkler, J. R., Gray, H. . Acc. Chem. Res. 2009, 42, 1995-2004. (n) Dempsey, J. L.; Winkler, J. R.; Gray, H. B. J. Am. Chem. Soc. 2010, 132, 16774-16776. (o) Dempsey, J. L.; Winkler, J. R.; Gray, H. B. J. Am. Chem. Soc. 2010, 132, 1060-1065. (p) Du, P.; Schneider, 1; Luo, G.; Brennessel, W.W.; Eisenberg, R. Inorg. Chem. , 2009, 48, 4952- 4962. (q) Du, P.; Knowles, K.; Eisenberg, R. J Am. Chem. Soc, 2008, 130, 12576-12577. (r) Fihri, A.; Artero, V.; Pereira, A.; Fontecave, M. Dalton Trans. 2008, 41, 5567-5569. (s) Probst, B.; Kolano, C. Hamm, P.; Alberto, R. Inorg. Chem., 2009, 48, 1836-1843. (t)

Lakadamyali, F.; Erwin, R. Chem Commun, 2011, 47, 1695-1697. (u) Szajna-Fuller, E. and Bakac, A. Eur. J. Inorg. Chem. 2010, 2488-2494. (v) Pantani, O.; Anxolabehere-Mallart,E.; Aukauloo, A.; Millet, P. Electrochem. Commun. 2007, 9, 54-58.

⁷ ( ρΜομ-8)₂8₂ΟΙ₂ as electrocatalyst for H₂ production: Appel, A. M., DuBois, D. L.; Rakowski-DuBois M. J Am. Chem. Soc. 2005, 127, 12717-12726.

⁸ [(Py₅Me₂)Mo(CF₃S0₃)]⁺ is a precatalyst for H⁺ electroreduction: Karunadasa, H. I.; Chang, C. J.; Long, J. R. Nature, 2010, 464, 1329-1333.

⁹ (a) Efros, L. L.; Thorp, H. H.; Brudvig, G. W.; Crabtree, R. H. Inorg. Chem. 1992, 31, 1722-1724. (b) Bernhardt, P. V.; Lawrence, G. A.; Sangster, D. F. Inorg.Chem. 1988, 27, 4055-4059.

¹⁰ (a) [Ni(P^Ph ₂N^C6H4 ₂)₂]²⁺ Complexes as Electrocatalysts for H₂ Production: Kilgore, U. J.; Roberts, J. A. S.; Pool, D. H.; Appel, A. M.; Stewart, M. P.; Rakowski-DuBois, M.;

Dougherty, W. G.; Kassel, W. G.; Bullock, R. M.; DuBois, D. L. J Am. Chem. Soc, 201 1, 133 , 5861-5872. (b) Similar complexes have also been shown to perform C0₂

electroreduction as well as H₂ production: Rakowski-DuBois, M.; DuBois D. L. Acc. Chem. Res. 2009, 42, 1974-1982.(c) Helm,M.L.; Stewart, M.P.; Bullock, R.M.; Rakowski DuBois, M.; DuBois, D.L.; Science 201 1, 12. 863-866.

u Ni and Co cobaloximes: Jacques, P.-A.; Artero, V. ; Pecaut, J.; Fontecave, M. Proc. Natl. Acad. Sci. 2009,106, 20627-20632.

¹² Stubbert, B. D.; Peters, J.C.; Gray, H. B. J. Am. Chem. Soc. 2011 Doi: 10.1021/ja2078015. ^I3David Morales-Morales; Craig Jensen; The Chemistry ofPincer Compounds 2007, Elsevier.

¹⁴ Manuel, T. D.; Rohde J.-U. J. Am. Chem. Soc. 2009, 131, 15582-15583.

¹⁵ Pool, D. H.; DuBois, D. L. J. Organomet. Chem., 2009, 694, 2858-2865.

b. Wilson, A. D.; Newell, R. H.; McNevin, M. J.; Muckerman, J.T.; Rakowski DuBois, M.; DuBois, D. L. J. Am. Chem. Soc, 2006, 128, 358-366.

¹⁶ Since not all precatalyst molecules are simultaneously in contact with the electrode throughout the CV experiment, the voltammetric rate of H₂ formation is higher than the observed gas evolution observed in a headspace analysis.

¹⁷ Luca, O. R.; Wang, T.; Konezny, S. J.; Batista, V. S.; Crabtree, R. H. New J. Chem. 2011, 35, 998-999.

¹⁸ (a) Wang, T.; Brudvig, G.; Batista, V. S. J. Chem. Theory Comput. 2010, 6, 755-760; (b) Wang, T.; Brudvig, G. W.; Batista, V. S. J. Chem. Theory Comput. 2010, 6, 2395-2401.

¹⁹ Gaussian 09, Revision A.l , Frisch, M. J.; et al. Gaussian, Inc., Wallingford CT, 2009. See the ESI for the complete reference.

²⁰ (a) Chirik P. J.; Wieghardt, K. Science 2010, 327, 794-795. (b) Bouwkamp, M. W.;

Bowman, A. C; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc, 2006, 128, 13340-13341. (c) Budzelaar, P. H. M.; de Bruin, B.; Gal, A. W. Inorg. Chem., 2001, 40, 4649-4655. (d) Sazama, G. T.; Betley, T. A. Inorg. Chem. 2010, 49, 2 12-2524. (e) A case of a redox active ligand participation in olefin binding was highlighted: Crabtree, R. H. Science 2001, 106, 56- 57. (f) Wang, K; Stiefel, E. I. Science, 2001, 291, 106-109. The data analysis in this paper was revised by (g) Geiger, W. E. Inorg. Chem., 2002, 41, 136-139.

²¹ (a) Huynh, M. H. V.; Meyer, T. J. Chem. Rev. 2007, 107, 5004-5064. (b) Costetin, C; Robert, M.; Saveant, i.-U. Acc. Chem.2524 J.-M. Acc. Chem. Res. 2010, 43, 1019-1029. Example 2

Ligand Preparation, Solubility Enhancement, and Derivatives with

Surface Tethering Capability:

Synthesis and Ligand Preparation:

Representative procedure: 4mmol of the appropriate pyridine were dissolved in 20mL ethanol along with 2mmol 2,6-diacetylpyridine. The solution was immersed in an oil bath preheated to 65°C with magnetic stirring. The solids must be completely dissolved before the next step for optimum yields. In the next step, 50mg KOH were dissolved in 2mL methanol and added to the stirring solution. The solution was refluxed (air condenser) for the time stated when a large amount of solid had precipitated. [ 4 hours for N5*, 30 mins N7, 1 hour N5] No inert atmosphere is necessary.

i) The N5 ligand is isolated as a bright yellow solid in 94% yield. The initial report report of the N5 ligand (R. P. Thummel, Y. Jahng Inorg Chem, 1986, 26, 2527-2524) reports spectroscopic characterization that corresponds to our data. See Figure 1 A.

ii) The 5* ligand was isolated as an off-white solid in 84%yield: an unprecedented yield. iii) The N7 ligand was isolated as a yellow solid in 87%yield: an unprecedented yield.

See Figure IB.

Nickel halide complexes:

Representative procedure: 0.42mmol (131mg) Ni(dme)Br₂ and 0.42mmol of the pincer ligand were added to a 50mL round bottom flask along with a magnetic stir bar. After the headspace was evacuated and backfilled with Ar three times, 20mL anhydrous CH₂C1₂ were added. The suspension was stirred for 6 hours at room temperature. The resulting brown suspension was filtered and the solid was washed three times with lOmL diethyl ether. See Figure 1C. Diagram 4 shows the general form of the ligand set.

3. Ni₂(OH)2(L)₂ compounds:

Representative procedure for N₅: 0.3mmol ligand N₅ in 5mL MeCN (dissolved with sonication) were added to a 50mL round bottom flask along with a magnetic stir bar and 0.42mmol green Ni(OH₂)6(BF₄)₂. Te mixture becomes an orange-brown suspension. After 5mL of water are added, the solution becomes homogeneous. The solution was then divided into three parts and left to stand. After 10 days, the combined solid residue was washed with water. 65mg golden flakes of product were recovered (32%).

i) Proposed structure for Ni₂(OH)₂(Ns)₂: golden flakes. (BF₄)₂

H402

Elemental analysis: calcd. C: 50.66%; H: 2.83%; N 14.07%; F: 15.26%; Found (sample dried to constant weight): C: 50.34%, H:2.58%, N:13.90%, F:15.20%; ii) A similar procedure was applied to the NiN₅* system but heating of the mixture was needed. This procedure yielded a Ni₂F₂ fluoride dimer presumably as a result of BF₄ hydrolysis. Other routes are being sought to try to make the Ni₂(OH₂)N₅* product.

Representative Procedure for N7 : 0.16mmol ligand and 0.16mmol Ni(OH₂) (BF₄)₂) were added to 600mL water. After stirring the resulting suspension at 60°C for 3 hours, a pale yellow solution is formed that was concentrated to 50mL on a rotary evaporator. The tan solid that precipitated was isolated by filtration. (98mg, 85%>)

Proposed structure for Ni₂(OH)₂(N₇)₂: tan solid

Elemental analysis: calcd. C: 44.07%; H, 2.72% ; F, 14.67%; N 18.93%; Found: C: x%; H:y%; N:z%; F: 14.46%

Ni(N₇)₂ (BF₄) was isolated as red crystals after recrystallization from MeOH/CH₂Cl₂ and MeCN (vapor diffusion).

Description of the H cell:

The cathode chamber had a working volume of 50mL (0.04 raM in the respective catalyst for the catalytic runs) and 20 mL for the cathode chamber. The two chambers were separated by a coarse frit. Stock solutions were prepared of the respective electrolytes. In the catalytic run the cathode was charged with 50 mL stock solution and catalyst (4.5μη ο1) and the anode with 20 mL of the same solution. The cathode and anode chamber solutions were sparged with inert gas for 5 min prior to starting the experiment.

The working electrode was a BASi RVC electrode referenced vs. Ag/AgCl (KCl_sat). The counter electrode is a 2.5cm x 2.5cm Pt mesh.

Background runs were performed after the catalytic runs with the cathode containing 50 mL of the stock solution with no catalyst and the anode 20 mL of the stock solution.

Quantitative Mass Spectrometry calibration of voltage response against H₂ detection was performed using 2.05%, 15%, 25% and 50% H₂/He custom prepared mixed gases from Tech Air and AirProducts Inc. Calibration. The average of duplicate catalytic run analyses are shown.

NiN₅Br₂

Volume H2 Volume H2 # Turnovers

Faradaic Cathode pH

Conditions detected (Background in

Yield % after 30 min (mL) subtraction) 30 min pHl

KC1/HC1

20.75 5.62 80% 55 6.34 -0.5V vs.

NHE

_PH4

phosphate

0.05M 13.84 13.84 45% 136 10.4 -0.8V vs.

NHE

pH7

phosphate

0.05M 0.077 0.077 1.03% 7.6 9.71 -UV vs.

NHE

NiN₅*Br₂

Volume H2 # Turnovers

Volume H2 Faradaic Cathode pH

Conditions (Background in

detected(mL) Yield % after 30 min subtraction) 30 min

pHl

KC1/HC1

29.54 14.42 96% 143 7.3 -0.5V vs.

NHE

pH4

phosphate

5.16 5.16 32% 51 9.05

0.05M

-0.8V vs. NHE

pH7

phosphate

0.05M No product observed 7.26 -l . lV vs.

NHE

Example 3

Further Exemplification of Qrganometallic Ni Pincer Complexes:

We describe further hereinafter electrocatalytic proton reduction and H₂ production by a series of organometallic nickel complexes with tridentate pincer ligands. The kinetics of H₂ production from voltammetry is consistent with an overall third order rate law: the reaction is second order in acid and first order in catalyst. Hydrogen production with 86-95% Faradaic yields was confirmed by gas analysis. Two proposed intermediates in the proton reduction cycle were isolated in a representative system and show a catalytic response akin to the parent compound.

Sustainable production of hydrogen is currently sought in the context of alternative energy strategies^lv,v and low-cost proton reduction catalysts are of interest. ^V1,v" The hydrogenase enzymes have served as inspiration for a large number of model compounds^vm'^lx,x and H⁺ reduction catalysts. Catalysts that model hydrogenases have now achieved essentially reversible H₂/H⁺ equilibrium" using bimetallic [Fe-Fe] and [Fe-Ni] active sites.

Large-scale commercial use of Pt, the most active proton reduction catalyst, is prohibitively costly.^x" Among nonprecious metals, catalysis by complexes of Co,™¹ Mo,^xlv'^xv ^^χνι,^χνιι,^χνι^η,^χι^χ ^_{QQn re}p_0rt_ed. Various factors affect the efficacy of these catalysts, including the solubility, stability, solvent and choice of proton source. For example, DuBois et al. notes an improvement of up to 60% in turnover number on addition of water to a nonaqueous system,¹⁴ while Fontecave et al.¹⁵ report drastic variations in turnover frequencies in macroscopic measurements of hydrogen production in a series of Co and Ni diimine dioxime catalysts in the presence of different proton donors.

A large body of work involving hydrogenase mimics, specifically Ni tetraphosphine complexes with pendant basic sites, has come from DuBois et al.™ These complexes not only act as electrocatalysts for proton reduction, but are also shown to catalyze the reverse process, hydrogen oxidation. The amine basic sites in the catalyst structure are thought to aid in stabilizing a bound hydrogen molecule, facilitate H₂ cleavage and mediate proton shuttling to and from the Ni metal center. The best system was shown to exhibit a turnover frequency of 106,000 s^_1 in acetonitrile, in the presence of 1.2 M water.™ When a closely related catalyst to DuBois' system was tethered to a carbon nanotube, a turnover rate of 20,000 h^"1, or -333 s^" ¹ was obtained from a 1 hour electrolysis. After 10 hours, the TON was 35,000, or—58 s^"1.^xxn Light-driven hydrogen production was also seen in aqueous conditions with a similar system.^xxlii

In organometallic chemistry, pincer ligands are widely used to support catalysts for a variety of different reactions, owing to the high stability they impart to transition metal complexes^XXIV and the tunability of their steric and electronic properties. Here, we report a series of Ni complexes having pincer ligands that give good activity as electrocatalysts for proton reduction.

Results and Discussion

Pincer catalysis

The pincer compounds depicted in Figure 1X3, although known,^{xxv,XXVI,x vu} have not previously been investigated in the context of proton reduction electrocatalysis. Their electrochemistry has now been studied in a standard 0.1 M acetonitrile/supporting electrolyte solution in order to determine their reduction potentials and establish any correlation between their catalytic activity and the ligand donor power (Figure 2X3). Our results confirm the expectation that a more strongly donor ligand framework stabilizes the higher oxidation state and makes metal reduction more difficult (See Figure 2X3 and the additional description hereinafter for details).

Previously, Meyer et al. showed that reduction of a Ni¹¹ complex with strong donor carbene ligands required large negative potentials^xxvm and we have assigned the two reduction waves of 1 (Figure 2X) to the N^/Ni¹ and NiVNi⁰ couples, although ligand participation cannot be completely excluded.^xxi Complexes 2 and 3 show similar electrochemical behavior. They both show a single reduction wave which we assign to a NP/Ni¹ couple. Complex 2 is more easily reduced than complex 3 which is consistent with the more donating framework of the PCP (complex 3) ligand.™

Incremental addition of aliquots of a 1 M HCl solution to complexes 1-3 led to a progressive increase in the current observed by voltammetry in this nickel pincer series (see ST). Figure 2X3 shows comparative cyclic voltammo grams of the pincer complexes in the presence of 20 uL 1 M HCl. Irreversible waves with increasingly more negative onset potentials were observed in the order: 1 < 2 < 3. As expected, increasingly donor ligands appear to provide more negative onset potentials. Plots of current density vs. applied potential were constructuted for catalysts 1-3 (Table 1X3 and S4) and show a lower overpotential at 1 mA cm^"2 for 1 (-0.273 V), than for 2 (-0.424 V) or 3 (-0.397 V).

The rates of H₂ production were measured from voltammetry (Table 1) using the method of DuBois and co-workers¹⁴ (described in detail in the SI). Our data are consistent with an overall third order rate law: the reaction is second order in acid and first order in catalyst. Catalyst 3 shows the highest overpotential in the series (-424 mV) and has a turnover frequency of 209 s^"1 under our conditions. Our systems appear to be slower catalysts in acetonitrile than DuBois' Ni tetraphosphine systems^14,17,18 but they operate at similar overpotentials. However, direct comparison is complicated due to differences in the exact experimental conditions.

Faradaic

Catalyst k (M V¹) TOF ( Yield^c

1 1.5* 10⁴ ± 2.3* 10³ 155 0.775 -0.273 V 86%±4%

2 0.55* 10" ± 0.32* 10³ 54.6 0.275 -0.397 V 90%±3%

3 2.1 * 10" ± 1.7* 10³ 209 1.045 -0.424 V 95%±2%

Table 1X3. Relative rates of proton reduction in a 0.1 M NBu₄BF₄ acetonitrile solution and overpotentials of catalysts 1-3 at 1 mA cm^"2. ^aCalculated for 0.1 M I-f, 5 mM catalyst, using the method described by DuBois et. α/.^χχχι determined from plots of Current density vs. applied potential constructed from chronoamperograms (dwell time: 60 sec) at progressively more negative potentials. The working electrode was a 3 mm diameter glassy carbon disk. Measurements were performed with magnetic stirring, using 5.2 mL acetonitrile solutions containing 200 μΐ, 1 M aqueous HC1 at catalyst concentrations of 0.2 mM. ^cFrom bulk electrolyses at -0.6 V vs. NHE, (details in the Experimental section and SI).

By analogy to the DuBois system, we propose the mechanism shown in Scheme 1 to describe the pathway for proton reduction. The PCP framework utilized to support the catalytically active compound 3 in acetonitrile provides a tractable representative system for the isolation of possible intermediates. We believe that the parent halide 3 is able to undergo a ligand exchange with the solvent. For the purposes of probing this hypothesis we were able to indepenently prepare 3-MeCN⁺ . The hydride complex 3-H was also synthetically accessibl literature methods.

H Reduct on H

Scheme 1. Proposed catalytic cycle proton reduction using 3-MeCN⁺ formed after loss of CI^" in MeCN.

The redox behavior of complex 3-H shown in Figure 3 was followed by cyclic voltammetry using 0.1 M NB i₄BF₄ and 0.2 mM catalyst precursor solution in a Schlenk cell under rigorously anhydrous conditions. The change from the halide in 3 to the much softer H^" ligation in 3-H causes a dramatic shift of the N^/Ni¹ couple by almost 1 V in the positive direction to -0.2 V vs. NHE. This wave also coincides with the onset of proton reduction catalysis in the parent halide 3, as seen in Figure 3X3. This observation is consistent with 3-H being an intermediate in the overall reaction. Upon incremental addition of 10 ]xL HCl to 3-H (see SI), bubbles are immediately observed followed by a steadily increasing current response analogous to the chloride compound under the same conditions. We, therefore, assign the observed electrochemistry to the [3-H]⁺/[3-H]° step (Scheme 1) and its participation in the catalytic cycle.

The third order rate law derived for catalysts 1-3 (see SI for more details) is consistent with the turnover limiting step in our catalytic cycle involving protonation of a Ni- H intermediate. We cannot exclude the possibility that the limiting step is in fact the loss of a metal-bound H₂; however, protonation of 3-H with one equivalent of HBF₄ in acetonitrile results in the clean formation of the acetonitrile supported cation 3-MeCN⁺ (Figure 4X3, Eq 1), which can also be prepared through the abstraction of CI^" from 3. When 3-H was protonated in a J. Young NMR tube, H₂ was detected by 1H NMR spectroscopy. This is consistent with our proposed mechanism involving protonation of the hydride, to generate a short-lived dihydrogen complex [3-¾]⁺ in Scheme 1, followed by coordination of solvent to close the cycle. A molecular hydrogen complex is a probable intermediate in the conversion of 3-H to 3-MeCN⁺ but is presumably too unstable to observe spectroscopically in acetonitrile in the timescale of the experiment. In the absence of protons, 3-MeCN⁺ shows an irreversible Niⁿ Ni' wave at a potential intermediate to that of 3 and 3-H (Figure 4X3) which we believe corresponds to the [3-MeCN]⁺/[3MeCN]° in the catalytic cycle depicted in Scheme 1. The irreversibility and magnitude of the wave is caused by the inability of the acetonitrile ligand to stabilize the Ni¹ center being formed (See SI for CVs at different scan rates).

In acidic conditions, 3-MeCN⁺ shows an increasing reductive current response akin to 3-H and 3 (Figure 3). The onset of the catalytic wave is in the same region as the Niⁿ/Ni^[ wave exhibited by 3-H, supporting our hypothesis that the solvento complex is involved in the catalytic cycle."™¹

Conclusions

A series of Niⁿ pincer complexes are active catalysts for electrochemical proton reduction. Bulk electrolysis experiments followed by macroscopic determination of the quantity of H₂ produced demonstrated good Faradaic yields (86%-95%). Two of the possible intermediate species were isolated and shown to be catalytically active using the PCP ligand framework. Further tuning of the ligand will be explored to look for even more active catalysts that operate at lower overpotentials.

Experimental Section General Methods:

All reagents were received from commercial sources and used without further purification unless otherwise specified. Solvents were dried by passage through a column of activated alumina followed by storage under dinitrogen. NMR spectra were recorded at room temperature on Bruker AMX-400 or 500 MHz spectrometers unless otherwise specified. Chemical shifts are reported with respect to residual internal protio solvent for 1H and ^I3C{'H} NMR spectra and to an external standard for ³¹P{1H} spectra (85% H₃P0₄ at 0.0 ppm). Literature procedures were utilized to synthesize compounds 2, 3, 3-H. ^22-25 Elemental analyses were performed by Robertson Microlit Inc.

Electrochemical Experiments, Bulk Electrolyses and Gas Analysis

Cyclic voltammograms (CVs) in acetonitrile were collected on glassy carbon electrodes (3 mm diameter from Bioanalytical Systems) with a platinum wire counter electrode and a silver wire reference electrode (referenced to NHE with ferrocene as external standard Ei_/2=690 mV vs. NHE). Measurements were performed in 0.1 M NBU₄BF₄ solutions at 2 mM concentration of the respective complexes. CVs were recorded after additions of 4* 10 1 M HC1 via volumetric syringe. CV measurements of 3-H were performed in a custom Ar filled Schlenk voltammetry cell equipped with a septum injection port. All other CVs were recorded after rigorous exclusion of air via Argon purge. Data workup was performed on OriginPro v8.0988 and AfterMath Data Organizer Version 1.2.3383.

Plots of current density vs. applied potential were constructured from

chronoamperometry experiments (dwell time: 60 seconds) at potentials lower than 0 V (us NHE) at a glassy carbon electrode in a single-chamber, three-electrode configuration. The experiments were performed in 5.2 mL of 0.1 M NBU₄BF₄ acetonitrile solutions containing 0.2 mM catalyst and 200 of a 1 M aqueous HC1 solution. Vigorous magnetic stirring was used to avoid diffusion limitations from concentration gradients at the working electrode.

The order of reaction with respect to acid was determined by analysis of cyclic voltammograms of 2 mM catalyst solutions with different acid concentrations. To a 0.1 M NBu₄BF₄ acetonitrile solution containing 2 mM catalyst (from the CVs of which i_p was determined at different scan rates), 5 increments of an aqueous 1 M HC1 solution (also containing 0.1 M NBu₄BF₄) were added. The voltammograms thus collected at each acid concentration (for several scan rates) were used to obtain catalytic currents (denoted i_c). The ratio of i_c/i_p was plotted against acid concentration for the different scan rates. From the approximately linear behavior at each scan rate, we conclude that our system obeys a rate law which is second order in acid. The slopes of the scan rate-dependent data were then used to calculate the third order rate constants. Graphs and description of the methods are available in the SI.

The order in catalyst was determined from voltammograms collected at 100 mV/s with 3.5 mL of a 0.1 M NBu₄BF₄ acetonitrile solution with a concentration of 2.82 mM catalyst and 4 mM acid. Increments of 0.5 mL of a 0.1 M NBu₄BF₄ acetonitrile solution were added and the dilutions were adjusted to maintain a constant acid concentration. We assumed no significant variations in the acid concentration throughout the experiment. From each voltammogram, i_c was plotted against catalyst concentration. From the linearity of the resulting graph, we conclude that at reasonably low catalyst concentration first order behavior is observed. Graphs and description of the methods are available in the SI.

Controlled potential headspace H₂ detection experiments were performed in a custom built two cylinder 50 mL bulk electrolysis H cell anode/cathode chamber separated by a coarse frit. The working electrode was a reticulated vitreous carbon electrode from

Bioanalytical Systems MF-2077 referenced vs. Ag/AgCl (KCl_sat). Headspace H₂ detection was performed at the Yale Department of Geology on a calibrated mass spectrometer: dual inlet Thermo Finnegan MDT 253 and an air-tight bulk electrolysis H Cell equipped with a sampling port. 1 mL volumes of gas were compressed in the bellow and then opened to the mass spectrometer.

Synthesis and Characterization of New Compounds and Experimental Procedure Synthesis of [{(MesIm)₂Py}NiBr]⁺[0₂CCF₃r (1)

This complex was prepared from 2,6-bis(3-mesitylimidazol-2-ylidene)pyridine disilver dibromide using a modified procedure based on that reported by K. Inamoto et al.²² The bis- carbene disilver dibromide was prepared from the corresponding pyridinium bis-

22

mesitylimidazolium dibromide (502 mg, 1 equiv) in the presence of excess Ag₂0 (903mg, 4.7 equiv.) in 1 ,2-dichloroethane under reflux overnight. After cooling, the desired compound was subsequently recrystallized from acetone and pentane giving 300mg of product (44% yield). To a suspension of NiBr₂(DME) (37.0 mg, 0.12 mmol) in 14 mL of CH₂C1₂ under a stream of argon was added the bis-carbene disilver dibromide (100 mg, 0.12 mmol, 1 equiv.) dissolved in 5 mL of CH₂C1₂ at 0 °C. The resulting light brown reaction mixture was stirred for 16 h at room temperature in the dark, followed by addition of silver trifluoroacetate (31.2 mg, 0.14 mmol, 1.2 equiv.) After an additional 24 h of stirring the yellow/brown mixture was transferred to a pad of celite via canula and filtered to remove insoluble silver bromide. The filtrate was concentrated in vacuo and washed three times with 5 mL of Et₂0 to remove excess Ag0₂CCF₃, giving 1 as a light brown solid (56 mg, 72% Yield). Anal, found (calcd) for C₃₁H₂₉BrF₃N₅Ni0₂: C 53.52 (53.25), H 3.94 (4.18), N 10.09 (10.02) %. 1H NMR (400 MHz, CD₂C1₂): 8.46 (2H, br, Ar-CH_Im), 8.42 (1H, br t, Ar-CH_py), 8.10 (2H, br d, Ar-CH_py), 6.92 (2H ,br, Ar-CHi_m), 6.90 (4H, s, Ar-CH_Mes), 2.28 (6H, s, Ar-p-CH₃), 2.07 (12H, s, Ar-o- CH ppm. ¹³C-{1H} NMR (125.8 MHz, CD₂C1₂): 164.1, 151.7, 140.2, 134.8, 130.4, 130.4, 129.2, 127.3, 119.1, 110.4, 21.3, 18.1 ppm. Synthesis of [(PCP)Ni(NCCH₃)]⁺[BF₄]^" (3-MeCN⁺)

To (PCP)NiCl (3) (40 mg, 82 μηιοΐ) and AgBF₄ (16 mg, 82 μηιοΐ) was added 1 mL of acetonitrile at 25 °C. The reaction was stirred overnight and then filtered through celite. The solvent removed was in vacuo to give [(PCP)Ni(NCCH₃)]⁺[BF₄]^" (3-MeCN⁺) as a yellow- orange powder (47 mg, 99% Yield). Single crystals suitable for x-ray crystallography were grown from saturated acetonitrile solutions containing 3-MeCN⁺ (see S7). HR FT-ICR MS: Found (calcd for C₂₆H₄₆ iP₂): m/z = (M)⁺ 492.2467 (492.2459), (M-ACN)⁺ 451.2193 (451.2196). ^!H NMR (NCCD₃, 500.0 MHz):□ 6.94 - 6.98 (3H, m, AxH), 3.32 (4H, t, PG¾Ar, J = 4.2Hz), 1.37 (36H, t, PC(C¾)₃, J = 6.6Hz). ^I3C-{1H} NMR (NCCD₃, 125.8 MHz):□ 154.6 (t, J = 11.4Hz), 153.0 (t, J = 13.9Hz), 135.9 (s), 128.0 (s), 123.4 (t, J = 9.1Hz), 35.9 (t, J = 7.6Hz), 33.4 (t, J = 13.7Hz), 29.7 (s). ³¹P-{^!H} NMR ( CCD₃, 161.9 MHz):

80.1ppm.

Protonation of [(PCP)NiH] (3-H)

To [(PCP)NiH] (3-H) (3 mg, 6.2 μηιοΐ) dissolved in 0.5 mL d₃-acetonitrile in a J. Young NMR tube was added 1 eq of HBF₄ in 10 μΐ H₂0 at 25°C. Mixing of the two solutions occurred only after the NMR cap was replaced. 1H NMR spectroscopy displayed a

characteristic resonance for H₂ along with peaks corresponding to [(PCP)Ni(NCCH₃)]⁺[BF₄]^" (3-MeCN⁺).

Supporting Information Available

Additional cyclic voltammograms, further experimental details, details of the crystal structure determination of 3-MeCN⁺ and xyz coordinates and energies of optimized structures are available free of charge via the Internet.

References for Example 3:

¹ Artero, V.; Fontecave, M. Coord. Chem. Rev. 2005, 249, 1518-1535.

² Vincent, K. A.; Parkin, A.; Armstrong, F. A. Chem. Rev. 2007, 107, 4366-4413.

³ Crabtree, R. H. Energy Production and Storage: Inorganic Chemical Strategies for a Warming World, 2010, Wiley-VCH, p 3-19.

⁴ Rosi, N. L.; Eckert, J.; Eddaoudi, M.; Vodak, D. T.; Kim, J.; O'Keeffe, M.; Yaghi, O. M. Science, 2003, 300, 1 127-1130. ⁵ Barton, B. E.; Whaley, M. C; Rauchfuss, T. B.; Gray, D. L. J. Am. Chem. Soc. 2009, 131, 6942-6943.

⁶ Gloaguen, F.; Lawrence, J. D.; Rauchfuss, T. B.; Benard, M.; Rohmer M. Inorg. Chem. 2002, 41, 6573-6582.

⁷ (a) Gloaguen, F.; Rauchfuss, T. B. Chem. Soc. Rev. 2009, 38, 100-108. (b) Losse, S.; Vos, J. G.; Rau, S. Coord. Chem. Rev. 2010, 254, 2492-2504.

⁸ Rakowski-DuBois, M.DuBois, M. R; DuBois D. L. Acc. Chem. Res. 2009, 42, 1974-1982.

⁹ (a) Bockris, J. O'M.; Conway, B. E. Trans. Faraday Soc. 1949, 45, 989-999. (b) Ezaki, H.; Morinaga, M.; Watanabe, S. Electrochimica Acta, 1993, 38, 557-564.

¹⁰ Stoichiometric cobalt mediated H₂ production: (a) Connolly, P.; Espenson, J. H. Inorg. Chem. 1986, 25, 2684-2688. (b) Koelle; U.; Ohst, S. Inorg. Chem. 1986, 25, 2689-2694. (c) Chao,T.-H.; Espenson, J. H. J. Am. Chem. Soc, 1978, 100, 129-133. Cobalt electrocatalysis (d) Hu X.; Brunschwig B. S.; Peters J. C. J. Am. Chem. Soc. 2007, 129, 8988-8998. (e) Kellett, R. M.; Spiro, T. G. Inorg. Chem. 1985, 24, 2373-2377. (f) Szymczak, N. K.; Berben, L. A.; Peters, J. C. Chem. Commun. 2009, 6729-6731. (g) Berben, L. A.; Peters, J. C. Chem. Commun. 2010, 398-400. - supported electrocatalyst (h) Baffert, C; Artero, V.; Fontecave, M. Inorg. Chem., 2007, 46, 1817-1824. (i) Bernhardt, P. V.; Jones, L. A. Inorg. Chem. 1999, 38, 5086-5090. (j) Pantani, O., Naskar, S., Guillot, R., Millet, P., Anxolabehere-Mallart, E.; Aukauloo, A. Angew. Chem. Intl. Ed. 2008, 47, 9948-9950. (k) Razavet, M.; Artero, V.; Fontecave, M. Inorg. Chem, 2005, 44, 4786-4795. (1) Fihri, A.; Artero, V.; Razavet, M.; Baffert, C; Leibl, W.; Fontecave, M. Angew. Chem. Intl. Ed. 2008, 47, 564-567. (m)

Stubbert, B. D.; Peters, J. C; Gray H. B.; J. Am. Chem. Soc, 2011, 133, 18070-18073.

Cobalt photocatalysis (n) Dempsey, J. L.; Brunschwig, B. S.; Winkler, J. R., Gray, H. B. Acc. Chem. Res. 2009, 42, 1995-2004. (o) McNamara, W.R.; Han, Z.; Alperin, P.J.; Brennessel, W.W.; Holland, P. L.; Eisenberg, R. Journal of the American Chemical Society 2011, 133, 15368-15371. Studies of photogenerated intermediates: (o) Dempsey, J. L.; Winkler, J. R.; Gray, H. B. J. Am. Chem. Soc. 2010, 132, 16774-16776. (p) Dempsey, J. L.; Winkler, J. R.; Gray, H. B. J. Am. Chem. Soc. 2010, 132, 1060-1065. (q) Du, P.; Schneider, J.; Luo, G.;

Brennessel, W.W.; Eisenberg, R. Inorg. Chem., 2009, 48, 4952-4962. (r) Du, P.; Knowles, K.; Eisenberg, R. J. Am. Chem. Soc, 2008, 130, 12576-12577. (s) Fihri, A.; Artero, V.;

Pereira, A.; Fontecave, M. Dalton Trans. 2008, 41, 5567-5569. (t) Probst, B.; Kolano, C. Hamm, P.; Alberto, R. Inorg. Chem., 2009, 48, 1836-1843. (q) Lakadamyali, F.; Erwin, E. Chem. Commun. 2011, 47, 1695-1697. (u) Szajna-Fuller, E.; Bakac, A. Eur. J. Inorg. Chem. 2010, 2488-2494 (v) Pantani, O.; Anxolabehere-Mallart, E.; Aukauloo, A.; Millet, P.

Electrochem. Commun. 2007, 9, 54-58. - both Co and Ni glyoximes.

¹¹ {CpMo(D-S)}₂S₂CH₂ as electrocatalyst for H₂ production: Appel, A. M., DuBois, D. L.; Rakowski-DuBois M. J Am. Chem. Soc. 2005, 127, 12717-12726.

¹² [(Py₅Me₂)Mo(CF₃S0₃)]¹⁺: Karunadasa, H. I.; Chang, C. J.; Long, J. R. Nature, 2010, 464, 1329-1333.

¹³ (a) Macrocyclic Nickel compound: Efros, L. L.; Thorp, H. H.; Brudvig, G. W.; Crabtree, R. H. Inorg. Chem. 1992, 31, 1722-1724. (b) Electrochemistry of Nickel tetrazamacrocycles: Bernhardt, P. V.; Lawrence, G. A.; Sangster, D. F. Inorg.Chem. 1988, 27, 4055-4059.

¹⁴ (a) [Ni(P^ph ₂N^C6H4X2)₂]²⁺ Complexes as Electrocatalysts for H₂ Production: Kilgore, U. J.; Roberts, J. A. S.; Pool, D. H.; Appel, A. M.; Stewart, M. P.; Rakowski-DuBois, M.; Dougherty, W. G.; Kassel, W. G.; Bullock, R. M.; DuBois, D. L. J Am. Chem. Soc, 2011, 133 , 5861-5872. (b) Similar complexes have also been shown to perform C0₂ electroreduction as well as H₂ production: Rakowski-DuBois, M.; DuBois D. L. Acc. Chem. Res. 2009, 42, 1974-1982.

¹⁵ Ni and Co cobaloximes: Jacques, P.-A.; Artero, V.; Pecaut, J.; Fontecave, M. Proc. Natl. Acad. Set 2009, 106, 20627-20632.

¹⁶ Angamuthu, R.; Bouman, E. Phys. Chem. Chem. Phys. 2009, 11, 5578-5583.

1 7

For a series of papers discussing the role of proton relays in Ni electrocatalysis: (a) Yang, J. Y; Bullock, R. M.; Shaw, W. J.; Twamley, B.; Fraze, K.; DuBois, D. L.; Simone, R.; Rousseau, R.; DuPuis, M.; Rakowski-DuBois, M. J. Am. Chem. Soc. 2009, 131, 5935-5945. (b) Yang, J. Y.; Shentan, C; Kassel, W. S.; Bullock, R. M.; DuBois, D. L.; Simone, R.; Rousseau, R.; DuPuis, M.; Rakowski-DuBois, M. Chem. Commun. 2010, 46. (d) Rakowski- DuBois, M; DuBois, D. M.; Chem. Soc. Rev. 2009, 38, 62-72. (c) DuBois, D. L.; Morris, B. R. Eur. J. Inorg. Chem. 2011, 1017-1027. (e) Jain, A.; Lense, S.; Linehan, J. C; Raugei, S.; Herman, C; DuBois, D. L. Inorg. Chem. 2011, 50, 4073. The Co analogous systems has been reported (f) Wiedner, E. S.; Yang, J. Y,; Dougherty, W. J.; Kassel, W. Bullock, R. M.; Rakowski-DuBois, M.; DuBois, D. L. Organome tallies 2010, 29, 5390-5401.

1⁸Helm, M.L; Stewart, M.P.; Bullock, R. M.; Rakowski DuBois, M.; DuBois, D.L.; Science, 2011 ,333, 863-866.

¹⁹ Le Goff, A,; Artero, V.; Jousselme, B.; Tran, P. D.; Guilet, N.; Metaye, R; Fihri, A.; Palacin, S.; Fontecave, M. Science 2009, 326, 1384-1387.

²⁰ McLaughlin, M.P.; McCormick, T.M.; Eisenberg, R.; Holland, P. Chem. Commun., 2011, 47, 7989-7991. David Morales-Morales; Craig Jensen; The Chemistry of Pincer Compounds 2007, Elsevier.

²² Inamoto, K.; Kuroda, I.; Kwon, E.; Hiroya, K.; Doi, T. J. Organomet. Chem. 2009, 694, 389-396.

²³ Gomez-Benitez, B.; Baldovino-Pantaleon, O.; Herrera-Alvarez, C; Toscano, R. A.; Morales-Morales, D. Tett. Lett. 2006, 47, 5059-5062.

²⁴ Boro, B. J.; Duesler, E. N.; Goldberg, K. I., Kemp, R. A. Inorg. Chem. 2009, 48, 5081- 5087.

²⁵ (a) Hu, X.; Castro-Rodriguez, I.; Meyer, K. Chem. Commun. 2004, 19, 2164-2165. (b) An example of NiⁿH species supported by a meridional tridentate ligand has been reported: He, T.; Andino, J. G.; Gao, X.; Fullmer, B. C; Caulton, K. G. J. Am. Chem. Soc. 2010, 132, 910- 911.

²⁶ Two recent reports of redox active carbenes: (a) Tennyson, A. G.; Lynch, V. M.; Bielawski, C. W. J. Am. Chem. Soc, 2010, 132, 9420-9429. (b) Dzik, W. I.; Zhang, X. P.; de Bruin, B. Inorg. Chem., 2011, 50, 9896-9903.

²⁷ Zhu, K.; Achord, P.D.; Zhang, X.; Krogh-Jespersen, K.; Goldman, A. S. J. Am. Chem. Soc, 2004, 126, 13044-13053.

²⁸ (a) Pool, D. H.; DuBois, D. L. J. Organomet. Chem., 2009, 694, 2858-2865. (b) Wilson, A. D.; Newell, R. H.; McNevin, M. J.; Muckerman, J. T.; Rakowski DuBois, M.; DuBois, D. L. J. Am. Chem. Soc, 2006, 128, 358-366.

²⁹ Electrodeposition was not observed. Background runs performed before and after running bulk electrolyses did not show evidence of significant surface alterations.

Example 4

Catalysis using Ni pincer complexes in non-aqueous conditions

As explained above, If H₂ is to be a fuel in environmentally friendly alternative energy strategies, ^XXX1 '^XXX1V more sustainable sources of H₂ are required. Elemental Pt is currently the best catalyst for the reduction of protons to H₂ ^xv but its low abundance and high cost make it unsuitable for large scale commercial use.³ A range of different transition metal complexes which can act as either electrocatalysts or photocatalysts have now been found, including systems involving Co,^xxxvi Mo^{xxxvii,xxxviii} andNi.^xxxix'^XI'^XIi However, improved catalysts that can operate in aqueous conditions with a cheap first row transition metal are of interest. Hydrogenase enzymes can bring about reversible FTTH;? equilibration and many synthetic systems have been developed in an attempt to mimic their bimetallic [Fe-Fe] or [Fe-Ni] active sites.^xllI'^xlm,xhv Although heavy metals like Rh, Ir and Os have been effectively utilized for H₂ activation,^xlv first row metals are preferred due to high abundance and low cost. Recent work involving hydrogenase mimics, specifically Ni tetraphosphine complexes with pendant basic sites, has come from DuBois et a/.^xlvl These complexes not only act as electrocatalysts for proton reduction, but also catalyze the reverse process, H₂ oxidation. The amine basic sites in the ligand structure aid in stabilizing a bound ¾ molecule, facilitate H₂ cleavage and mediate proton shuttling to and from the Ni center. Scheme 1 depicts an example of such a catalyst, DB-l,^xlv" a Ni tetraphosphine with a pendant amine basic site in its periphery. The first step involves reduction to a Ni¹ species (DB-2), which is subsequently protonated at the amine basic site to form a Ni¹ cation (DB-3). Proton transfer to Ni then occurs to generate a Ni^m hydride (DB-4). DB-4 undergoes a second le^" reduction to form DB-5 which is subsequently protonated. The protonation occurs on the nearby pendant amine, and the proximity of the amine to the hydride in Niⁿ-H (DB-6) is proposed to aid in both the binding and release of H₂. This allows the system to operate in the oxidative as well as the reductive direction. When a closely related catalyst to DuBois' system was tethered to a carbon nanotube, a turnover rate of 20,000 per hour, or—333 per second was obtained in a 1 hour electrolysis run; a 10 hour electrolysis run showed a TON of 35,000, or -58 per second.^xlvm Recently the DuBois system has also been harnessed for light-driven H₂ production in conjunction with a photosensitizer and electron donor to give a maximum TON of 2700.^xlix

Scheme 1. Catalytic cycle for proton reduction by a Ni-tetraphosphine with a pendant basic site.¹⁷

Unfortunately, the DuBois molecular system requires acetonitrile or acetonitrile/water as the solvent and proton reduction catalysts that are soluble and active in water are relatively rare. ' Such systems might find applications in fuel cells or for H₂ storage. To date, the Mo catalyst,¹⁸ [(Py₅Me₂)Mo(CF₃S0₃)]⁺, with approximately 1600 turnovers per hour at pH 7 in a solution phase preparative experiment, is the most effective water soluble catalyst. A Co analogue has also been reported¹' but catalysts incorporating first row transition metals generally give lower turnover numbers.

Pincer ligands are attractive because they impart high stability to the resulting transition metal catalysts and facilitate tuning of ligand properties.¹" Here we report that a series of Ni complexes supported by pincer ligands give good activity as electrocatalysts for proton reduction in both acetonitrile and aqueous conditions where solubility permits. In addition we find activity at low overpotential for pyridyl-2,6-bisimine pincer ligands that can show ligand-centered redox behavior. Experimental and computational studies have helped us understand the mechanism of proton reduction and suggest that a key step for these complexes involves proton coupled electron transfer (PCET).

Results and Discussion

Catalysis using Ni pincer complexes in non-aqueous conditions.

The Ni pincer complexes selected, shown in Figure lX4,^hu'^hv'^lv'^lvIhave not previously been investigated for proton reduction catalysis. With the idea that ligand-centered redox- activity might be useful for promoting proton reduction, complexes 4 and 5 were of particular interest since these pincers have indeed been shown to undergo ligand-centered redox processes. ^lvu The relative current response of compounds 1-5 was studied in a standard 0.1 M NBu₄BF₄ acetonitrile solution in order to determine the reduction potentials of the complexes. We sought to establish a correlation between their catalytic activity and the ligand donor power (Figure 2X4). Our results show that a more strongly donor ligand framework stabilizes higher oxidation states and makes metal reduction more difficult (See Figure 2X4 and the SI for details).

Previously, Meyer et.al. showed that reduction of a Ni¹¹ complex with strong donor carbene ligands required large negative potentials^lvm and we have tentatively assigned the two reduction waves of 1 (Figure 2X4) to the N^/Ni¹ and NiVNi⁰ couples, although ligand participation cannot be excluded.¹¹" The related complexes 2 and 3 show similar

electrochemical behavior. They both show a single reduction wave which we assign to a N^/Ni¹ couple. Complex 2 is more easily reduced than complex 3 which is consistent with the more donating framework of the PCP (complex 3) ligand. ^lx Ligand-centered reduction of the NNN pincer ligands of compounds 4 and 5 has been reported, as well as EPR evidence for an organic radical as the product of one-electron reduction.²⁷ Assuming this is the case here, the comparable voltammetric reduction waves that occur just above 0 V for complexes 4 and 5 may be associated with ligand-centered processes. If so, the second reduction events at ~ -0.5 V vs. NHE can be attributed to a N^/Ni¹ couple.

Incremental addition of 1 M HC1 to complexes 1-5 led to a progressive increase in the voltammetric current due to the production of H₂ (see below for the detection of H₂ and the SI for more details). Figure 2X4 shows comparative cyclic voltammo grams of the pincer complexes in the presence and absence of 20 of 1 M HC1. In order to more accurately assess catalyst performance, a series of chronoamperometry experiments were performed and Tafel plots were constructuted to determine overpotentials for 1-5 (Table 1 and SI). A current density of 1 mA cm^" was chosen for this determination. Catalysts with redox active ligands: 4 and 5 show the lowest overpotentials: 0.223V for 4 and 0.193V for 5 versus 0.273V for 1, 0.424 V for 2 and 0.397 V for 3. We assign the discrepancy between the observed

voltammetric onset of catalysis and measured overpotentials to differences in diffusional parameters between the complexes.

The rates of H₂ production were measured from voltammetry (Table 1) using the method of DuBois and co-workers³¹ (described in detail in the SI). Our data is consistent with an overall third order rate law; the reaction is second order in [acid] and first order in

[catalyst]. Catalysts active at low overpotentials 4 (-0.233 V) and 5 (-0.193 V) have a turnover frequency of 105 s^"1 and 68 s^"1 respectively, while catalyst 3 which shows the highest overpotential in the series (-424 V), has a turnover frequency of 209 s^"1 under our conditions. Our systems appear to be slightly slower catalysts in acetonitrile than DuBois' Ni tetraphosphine systems¹⁰'^16,31 but complexes 4 and 5 operate at lower overpotentials.

However, direct comparison is complicated due to differences in the exact experimental conditions and the systems can be considered broadly comparable. TOF Potential at

1 1.55* 10⁴± 2.35* 10³ 155 0.775

2 0.546* 10⁴ ± 0.327* 10³ 54.6 0.275 -0.397V

3 2.09* 10⁴ ± 1.7Ή0³ 209 1.045 -0.424V

4 1.05* 10⁴ ± 2.08*10³ 105 0.525 -0.233V

5 0.680* 10⁴± 1.03* 10³ 68 0.34 -0.193V

Table 1. Relative rates of proton reduction in a 0.1 M NBu₄BF₄ acetonitrile solution and overpotentials of catalysts 1-5 at 1 mA cm^"2. Calculated for 0.1 M H⁺, 5 mM catalyst, using the method described by DuBois et. al.^{1x1 b}Determined from Tafel Plots constructed from chronoamperograms (dwell time: 60 sec) at progressively more negative potentials at a glassy carbon electrode with magnetic stirring, using 5.2 mL acetonitrile solutions containing 200 μΐ_^ 1 M aqueous HCl at catalyst concentrations of 0.2 mM.

Catalysis using Ni pincer complexes 4 and 5 in aqueous organic and aqueous conditions.

Not many electrocatalytic molecular systems that have been studied in detail in water, presumably because of low solubility of most relevant complexes, or possibly poor behavior in water. Given the present interest in proton reduction catalysts that operate in an aqueous environment,⁸ we tested the catalytic ability of 1-5 in water. Catalyst 4 is sufficiently soluble for screening under purely aqueous conditions, while 5, although insoluble in pure water, is soluble in MeCN/H₂0 (1 :3 v:v). Compounds 1-3 were not soluble under these conditions. Incremental addition of acid to aqueous or organic/aqueous solutions of 4 or 5 both yielded the expected progressive increase in current response (Figures 3X4 and 4X4). Although neither 4 nor 5 showed high reduction currents in neutral water (or aq. MeCN in the case of 5), we were able to investigate the proton reduction activity of 4 in aqueous conditions at low pH because of its solubility.

Unlike the CV of 5, which does not show any resolvable features in neutral aqueous conditions (Figure 4X4), at pH 7 in 0.1 M aqueous KCl, the CV of 4 shows a quasi-reversible reduction wave at -0.62 V (Figure 3). As aliquots of 1 M HCl are added, we observed marked incremental increases in current but the return oxidation wave is completely lost. Gas bubbles are visible at the basal plane graphite surface during the CV experiment with 40 uL of acid solution (Figure 3X4). The presence of an uneven biphasic gas/liquid interface at the electrode and a continuous change in surface area available in this diffusion-controlled measurement may cause the unusual crossing of the waves near 0.7 V in the reduction and return oxidation scans. As H₂ is produced, less of the reducible substrate becomes available at the working electrode surface, thus causing a decrease in current response on the return wave in the oxidation direction. Alternatively the curve crossing could be caused by the facile reduction of a Ni^ra-H intermediate at a comparable or positive potential compared to 4.

However, DFT calculations (vide infra) suggest that the formation of a Ni^m species is unfavorable and this explanation is unlikely. In the presence of acid, compound 5 displayed similar current response to 4, suggesting that with the exception of differences in solubility, both compounds are comparable for proton reduction (Figure 4X4).

Bulk electrolysis experiments were performed under the catalytic conditions to determine the macroscopic amount of H₂ produced by 4 in water (see supporting

information); in order to maximize ¾ production for detection, a high surface area reticulated vitreous carbon working electrode was selected for this experiment. Specifically, a 50 mL 0.1 M KC1/HC1 solution (pH 1) with 0.2 mM precatalyst was held at -1.1 V vs. NHE for one hour. At the end of the experiment the pH of the solution was 1.15, indicating that protons had been consumed. After subtraction of the charge corresponding to background H₂ production without added catalyst (80 C), 132 C were consumed, equivalent to the theoretical production of 0.68 millimoles of H₂. From quantitative mass spectrometry, 0.65 millimoles of H₂ were detected after background subtraction. Thus, we achieve H₂ production under these conditions with a Faradaic efficiency of >95%. This data corresponds to a minimum of 65 mole H₂ per mole of catalyst precursor per hour. Previously reported aqueous preparative proton reduction turnover frequencies with untethered molecular complexes of first row

99

metals show less than 17 mole H₂ per mole of catalyst per hour. However we are unable to directly compare our system with other reported examples due to the different methods used for determining TOF. Cell configurations vary widely across research groups, posing severe difficulties for comparison of results across systems. .

DFT Catalytic Cycle and Mechanistic Experiments

In order to investigate the mechanism of H₂ evolution using complexes 1-5, we performed ab initio calculations on 3 and 4, as representative examples. As in previous studies,^,x" we applied DFT/B3LYP to characterize the structural and spin/electronic properties of the reaction intermediates and perform free energy calculations.¹"¹¹¹ Gas phase free-energy changes were calculated at the B3LYP/cc-pVTZ level, using minimum energy structures obtained at the DFT B3LYP/LANL2DZ level of theory, and then corrected by solvation free-energy calculations with the LANL2DZ basis set, using the Polarizable Continuum Model (ε = 78.4 for water) as implemented in Gaussian 09.

Our calculations predict that in aqueous conditions, 4 will readily lose one Br^" to form a square planar complex with a single Br^" ligand (AG for ligand loss is -4.7 kcal mol^"1).

Subsequently, the Br^" ligand can easily exchange with water (AG for replacement of Br^" with water is -6.3 kcal mol^"1) (Scheme 2). This is consistent with experimental results indicating that dissolution of 4 in <¾-DMSO results in the formation of a diamagnetic species, most likely a 4-coordinate square planar dication with coordinated DMSO (see the SI for more information). Given that the catalysis is performed in water, the DFT results suggest that the catalytic cycle begins with a square planar Ni¹¹ water-ligated complex and that 4 is only a precatalyst.

Scheme 2. Energies for ligand substitution in 4 in aqueous conditions.

The first step in the catalytic cycle (Figure 5X4) could involve either protonation of the Ni¹¹ starting material to give a formally Ni^IV hydride or reduction of [4'(OH₂)]²⁺. Both experimental and theoretical data suggest that reduction can be thermo dynamically favored over protonation to the unfavorable Ni^IV oxidation state. The lowest energy reduction product is the square planar, water-ligated, complex [4'(OH₂)]⁺,'^xlv with high-spin density on the pincer ligand, as shown by the spin-distribution in Figure 6. However, our DFT calculations indicate that an isomer of [4'(OH₂)]⁺ with the unpaired electron on the Ni is close in energy to the isomer with the ligand centered radical and within the errors of DFT we can not differentiate between the two. Therefore, we can not be definite in our assignment of the reduction waves, shown in Figure 2X4, as being either ligand centered or metal centered.

Once [4'(OH₂)]⁺ has been formed, three main pathways are available: (a) further reduction to give a Ni¹ species with a ligand-centered radical, (b) protonation of [4'(OH₂)]⁺ to give either a Ni^IV hydride with a ligand-centered radical or a Ni¹¹¹ hydride with a neutral ligand, or (c) PCET to give a Ni¹¹ hydride with a neutral ligand. Our DFT and experimental results suggest that the second reduction of 4 to give a Ni¹ species and a ligand-centered radical is energetically disfavored and requires a significantly greater reduction potential than that needed during catalysis (vide supra); pathway (a) has therefore been dismissed.

Protonation of [4'(OH₂)]⁺ could result in the formation of two different Ni hydride isomers, cw-[4'(OH₂)₂H]²⁺ or /ra¾s-[4'(OH₂)₂H]²⁺ that differ in the coordination geometry around Ni. In cis- [4' (OH₂)₂H] , the hydride is coplanar with the NNN ligand with two axially-bound water ligands (which stabilize the higher oxidation state), while in trans- [4 '(OH₂)₂H]²⁺ the hydride is perpendicular to the NNN ligand. The spin density analysis indicates that both cis- [4'(OH₂)₂H]²⁺ and trara-[4'(OH₂)₂H]²⁺ are formally Ni¹¹¹ hydrides (Figure 6), suggesting that aNi^IV species with a ligand-centered radical is not accessible. However, protonation to form cw-[4'(OH₂)₂H]²⁺ or tram-[4'(OH₂)₂H]²⁺is significantly uphill energetically,

(AG_aq([4'(OH₂)]⁺/ cw-[4'(OH₂)₂H]²⁺) = 40 kcal mol^"1 (1.73 eV) and AG_aq([4'(OH₂)]⁺/trara- [4'(OH₂)₂H] ) = 26 kcal mol^" (1.13 eV)), and is not spontaneous. On this basis we propose that the PCET pathway (c) is the most likely route that converts [4'(OH₂)]⁺ into the square planar Ni¹¹ hydride [4'(H)]⁺. The free energy requirement for [4'(OH₂)]⁺→[4'(H)]⁺

conversion thus includes (1.13 + 0.059 pH) eV to protonate [4'(OH₂)]⁺ to give trans- [4'(OH₂)₂H] minus the excess free energy (4.60 eV) due to reduction of trans- [4'(OH₂)₂H] into [4'(H)] . Therefore, by coupling protonation and reduction in PCET,

[4*(OH₂)]⁺ is converted into [4'(H)]⁺ at a potential (1.13 + 0.059 pH - 3.60) V. This very large driving force force^lxv of Δ0_3ς([4'(ΟΗ₂)]⁺/[4·(Η)]⁺ = -80 kcal mol^"1 (-3.47 V vs NHE) at pH = 0 is consistent with our measured KIE (k_H/ko) of 4.2(1). The catalytic cycle is completed by protonation of [4*(H)]⁺ to give the Ni¹¹ dihydrogen complex [4'(H₂)]²⁺)

(AG_aq([4*(H)]⁺/ [4*(H₂)]²⁺) = 9 kcal mol^-1) and subsequent H₂ evolution, with the H₂ ligand substituted by water to regenerate [4'(OH₂)] . Notably, the highly reactive nature of the radical intermediates proposed to intervene in the cycle has been previously documented: a related species to the one-electron reduced catalyst precursor 4 has been shown to undergo oxidative demethylation upon exposure to air.³² Our experiments were performed under rigorous Ar purge and we believe that such a degradation pathway is only accessible in the presence of adventitious oxidizing species, such as aerial dioxygen.^lxY1

Our calculated catalytic cycle for proton reduction using 3 (Figure 7X4) involves similar intermediates as proposed for 4 (Table 2 compares the relative energies of the intermediates). However, there are several important differences in the relative energies and electronic structures of the intermediates. In an analogous fashion to 4, the solvento complex [3'(MeCN)]⁺ is calculated to be the starting point for catalysis using 3. Subsequently, one electron reduction of [3'(MeCN)]⁺ generates the Ni¹ species [3'(MeCN)] (Figure 8), which is consistent with our assignment of the first reduction in the CV of 3 to the N^/Ni¹ couple. Protonation of [3'(MeCN)] to form the Ni¹¹¹ hydride [3Ή]⁺ is downhill in energy by -7.2 kcal mol^"1, which suggests that PCET is not required in the PCP system. Presumably the fact that [3'(MeCN)] is neutral makes it significantly easier to protonate than cationic [4'(OH₂)]⁺. Surprisingly, the coordination of additional MeCN to [3'H]⁺ is disfavored and the lowest energy structure is a planar four coordinate species rather than the octahedral six coordinate complex calculated for 4 (see SI for more details). In order to complete the cycle the Ni¹¹¹ intermediate [3'H]⁺ is reduced by one electron to give the square planar Ni¹¹ hydride [3'(H)], which undergoes protonation to give the dihydrogen complex [3'(H₂)]⁺, followed by loss of H₂ to regenerate [3' (MeCN)] ⁺.

The third order rate law experimentally determined for catalysts 1-5 suggests that the turnover limiting step in our catalytic cycles involve protonation of the Ni-H intermediate ([4'(H)]⁺ to [4'(H₂)]²⁺ in Figure 5 or [3'(H)J to [3*(H₂)]⁺ in Figure 7X4). We cannot exclude the limiting step being loss of H₂ (for example [4'(H₂)]²⁺ ~» [4'(OH )]²⁺), however, at least in the case of 3, relative energies indicate that this should be a spontaneous process.

Relative Energy Relative Energy

Compound Compound

(kcal mol ) (kcal mol^" )

[4'(OH₂)]²⁺ 0 [3'(MeCN)]⁺ ¾

[4'(OH₂)]⁺ -95.7 [3'(MeCN)] -64.8

/r ra-[4'(OH₂)₂H]²⁺ -56.1

OT-[4'(OH₂)₂H1²⁺ -69.7 [3'(H)]⁺ -72.0

[4'(H)]⁺ -184.5 [3*(H)] -177.6

[4'(H₂)]²⁺ -175.6 [3'(H₂)]⁺ -192.7

Table 2. Comparison of relative energies of intermediates from DFT calculations of intermediates in proton reduction using 3 and 4.

The DFT calculations indicate that there are four key common intermediates present in the catalytic cycle for proton reduction using both 3 and 4: (i) the Ni¹¹ solvento species

[4'(OH₂)]²⁺ or [3'(MeCN)]⁺, (ii) the postulated Ni¹¹ solvento complex with a ligand-centered radical [4*(OH₂)]⁺ or the Ni¹ species [3'(MeCN)], (iii) the Ni¹¹ hydrides [4'(H)]⁺ or [3'(H)], and (iv) the Ni¹¹ dihydrogen complexes [4'(H₂)]²⁺ or [3'(H₂)]⁺. Attempts to isolate and spectroscopically characterize these intermediates using the NNN pincer ligand were complicated by the high reactivity of these NNN pincer complexes in water. However the PCP framework, provided a more tractable system to compare with the proposed

intermediates. The hydride complex [3'(H)] was prepared by literature methods. Complex

[3 '(H)] was expected to undergo reduction at a more positive potential than complex 3. The redox behavior of complex [3 '(H)] shown in Figure 9 was followed by cyclic voltammetry using 0.2 mM catalyst in a 0.1 M NBu₄BF₄ acetonitrile solution in a Schlenk cell under rigorously anhydrous conditions. Indeed, the change from the halide in 3 to the much softer H^" ligation in [3'(H)] causes a dramatic shift of the N^/Ni¹ couple by ~1 V in the positive direction to -0.2 V vs. NHE. This wave also coincides with the onset of proton reduction catalysis in the parent halide, 3 as seen in Figure 9X4. This observation is consistent with

[3'(H)] being an intermediate in the overall reaction. Upon incremental addition of 10 μΐ, HC1 to [3'(H)], bubbles are immediately observed as is also seen in the case of other metal hydrides, followed by a steadily increasing current response analogous to the chloride compound under the same conditions.

Protonation of [3'(H)J with one equivalent of HBF₄ in acetonitrile results in the clean formation of the acetonitrile cation [3'(MeCN)]⁺ (Eq 1), which can also be prepared through the abstraction of CI^" from 3. When ]3'(H)] was protonated in a J. Young NMR tube, H₂ was detected by ^!H NMR spectroscopy. This is consistent with our proposed mechanism involving protonation of the hydride [3'(H)], to generate the dihydrogen complex [3'(H₂)]⁺, followed by coordination of solvent to generate [3'(MeCN)]⁺. A dihydrogen complex is a probable intermediate in the conversion of [3'(H)] to [3'(MeCN)]⁺ but is presumably too unstable to observe spectroscopically in acetonitrile on the timescale of the experiment. In the absence of protons, [3'(MeCN)]⁺ shows an irreversible N^'/Ni¹ wave at a potential intermediate to that of 3 and 3-H (Figure 10X4). We assign the irreversibility and magnitude of the wave to the inability of the acetonitrile ligand to stabilize the Ni¹ center being formed.

In acidic conditions [3'(MeCN)]⁺ shows an increasing reductive current response akin to [3'(H)] and 3 (Figure 1 1). The onset of the catalytic wave is in the same region as the

Ni'W wave exhibited by [3'(H)], supporting our hypothesis that the solvento complex is involved in the catalytic cycle. Our results with the PCP complexes are consistent with our proposed mechanism for proton reduction. Conclusions

Pursuant to the present invention, we have studied a series of Ni pincer complexes and compared their relative electrochemical response to addition of increments of acid. All of the compounds studied were active catalyst precursors for electrochemical proton reduction and we believe this is the first report of pincer complexes being operationally homogeneous catalysts for this reaction. The pincer complexes 4 and 5, which incorporate a redox active NNN ligand showed the lowest overpotentials and were also able to reduce protons at high rates: 105 s^"1 and 68 s^"1, respectively, in acetonitrile and in aqueous conditions. Bulk electrolysis experiments followed by macroscopic determination of the quantity of H₂ produced, demonstrated that complex 4 catalyzed the formation of at least 65 mole ¾ per mole of catalyst per hour (-1.1 V vs. NHE in 50 mL 0.1 M KC1/HC1 solution pH 1 with 0.2 mmol catalyst), which makes it one of the most active catalysts reported to date for aqueous proton reduction. However, a reliable comparison of ¾ evolution measurements across different systems is difficult because of the wide array of experimental conditions used in proton reduction.¹¹ Factors that change between systems include experimental setups, electrolysis potentials and choice of solvent and proton source. Furthermore, some reports lack details about some of the collection parameters, such as the electrolysis time. This leads to an inability to compare the macroscopic rates of H production in different systems.

DFT studies suggest that the mechanism of proton reduction for 3 and 4 are similar and that there are several common indermediates. In the case of 3 several of the proposed intermediates were isolated and shown to be catalytically active. At this stage we believe that although a key PCET step is required for 4, it is not required for 3. Further work will look to tune the properties of the pincer ligand to create even more active catalysts for proton reduction.

Experimental Section

General Methods

All reagents were received from commercial sources and used without further purification unless otherwise specified. Solvents were dried by passage through a column of activated alumina followed by storage under dinitrogen. NMR spectra were recorded at room temperature on Bruker AMX-400 or 500 MHz spectrometers unless otherwise specified. Chemical shifts are reported with respect to residual internal protio solvent for 1H and NMR spectra and to an external standard for ³¹P NMR spectra (85% H₃P0₄ at 0.0 ppm). Literature procedures were utilized to synthesize compounds 2, 3, 3-H, 4 and 5.^35-38 Elemental analyses were performed by Robertson Microlit Inc.

Electrochemical Experiments, Bulk Electrolyses and Head Space Analysis

Cyclic voltammograms (CVs) in aqueous solution were collected using a 0.09 cm² basal plane graphite working electrode prepared by the method of Blakemore et. al.,³² a platinum wire counter electrode and an aqueous Ag/AgCl, KCl^lxv" (sat) reference. In acetonitrile, a glassy carbon electrode was used as the working electrode and a Ag wire reference (referenced vs. NHE with ferrocene as external standard). For the aqueous electrochemistry 3 mL 0.1 M KCl aqueous solutions were used, with incremental amounts of acid added (1 M HC1, 40, 90, 1 0 Κ). In the case of 5, 1 mL acetonitrile was needed for solubilization. Electrochemistry in acetonitrile was performed in 3 mL 0.1 M NBu₄BF₄ at 2 mM concentration of the respective complexes. Basal plane graphite electrodes were preferred over glassy carbon electrodes in aqueous conditions due to low background currents in the absence of catalysts. Background CVs are provided in the SI. CVs were recorded after addition of 4* 10 1 M HC1 via volumetric syringe. CV measurements of [3'(H)] were performed in a custom Ar filled Schlenk voltammetry cell equipped with a septum injection port. All other CVs were recorded after rigorous exclusion of air via Argon purge. Data workup was performed on OriginPro v8.0988 and AfterMath Data Organizer Version

1.2.3383.

Kinetic isotope studies studies were carried out with a Pine AFCBP1 Bipotentiostat and a Pine MSR variable-speed rotator. The reference was a Ag/AgCl electrode

(Bioanalytical Systems, Inc.) and the counter electrode was a platinum wire. The disc

(surface area: 0.07 cm²) material was basal plane graphite (with stabilizing resin around the graphite; from Pine). The disc was assembled in a Pine E6-series ChangeDisk Setup. The disk voltage was held at 0.5 V vs NHE for 30 min. Data was collected in 12 mL of a 0.1 M KCl solution containing 1 mM catalyst 4 and 20% HC1 or 20% DC1 (procured from

Cambridge Isotopes) at 500 rpm.

Tafel Plots were constructured from chronoamperometry experiments (dwell time: 60 seconds) at potentials lower than 0 V (vs NHE) at a glassy carbon electrode in a single- chamber, three-electrode configuration. The experiments were performed in 5.2mL of a 0.1 M NBu₄BF₄ acetonitrile solution containing 0.2 mM catalyst and 200 of a 1 M aqueous HC1 solution. Magnetic stirring was used to avoid diffusion limitations from concentration gradients at the working electrode. The order of reaction with respect to acid was determined by analysis of cyclic voltammograms of 2 mM catalyst solutions with different acid concentrations. To a 0.1 M NB 1₄BF₄ acetonitrile solution containing 2 mM catalyst (from the CVs of which i_p was determined at different scan rates), 5 μΐ. increments of an aqueous 1 M HC1 solution (also containing 0.1 M NBU4BF₄) were added. The voltammograms thus collected at each acid concentration (for several scan rates) were used to obtain catalytic currents (denoted i_c). The ratio of iji^' _p was plotted against acid concentration for the different scan rates. From the linear behavior at each scan rate, we conclude that our system obeys a rate law which is second order in acid. The slopes of the scan rate-dependent data were then used to calculate the third order rate constants. Graphs and description of the methods are available in the SI.

The order in catalyst was determined from voltammograms collected at 100 mV/s with 3.5 mL of an acetonitrile 0.1 M NBu₄BF₄ solution with a concentration of 2.82 mM catalyst and 4 mM acid. Increments of 0.5 mL of a 0.1 M NBu₄BF₄ acetonitrile solution were added and the dilutions were adjusted to maintain a constant acid concentration. We assumed no large variations in the acid concentration throughout the experiment. From each voltammogram, i_c was plotted against catalyst concentration. From the linearity of the resulting graph we conclude that at reasonably low catalyst concentration first order behavior is observed. Graphs and description of the methods are available in the SI.

Controlled potential headspace H₂ detection experiments were performed in a custom built two cylinder 50 mL bulk electrolysis H cell anode/cathode chamber separated by a coarse frit. The working electrode was a BASi RVC electrode referenced vs. Ag/AgCl (KCl_sat). Headspace H₂ detection was performed at the Yale Department of Geology on a calibrated mass spectrometer: dual inlet Thermo Finnegan MDT 253 and an air-tight bulk electrolysis H Cell equipped with a sampling port. 1 mL volumes of gas were compressed in the bellow to 10% then opened to the mass spectrometer.

Computational Methods

Density functional calculations were carried out using Gaussian 09.^lxvm Gas phase free energy changes were calculated at the B3LYP/cc-pVTZ level, using minimum energy structures obtained at the B3LYP/LANL2DZ level of theory. Changes in the free energy in aqueous and non-aqueous conditions AG(soln) from reactants R to products P were found using the thermodynamic cycle: AG(g)

R(g) ·► P(g)

AG(soln)

R(soln) P(soln) where:

AG(soln) = AG(g) + AG^p _soW - AG^R _solv,

and AG(g) = AH(g) - ΓΔ5¾) is the free energy state transition in the gas phase.

Enthalpy changes in the gas phase H(g) - AHscF + AHzpi- + AHr were obtained from changes in the DFT self-consistent field energy AHscF vibrational frequency calculations that yield changes in the zero point energy AHzpE, corrections for molecular entropy changes S(g), and corrections due to changes in thermal enthalpy AHj. The solvation free energies AGsoiv for species other than solvated H⁺ and Br^" were computed using the Polarizable Continuum Model as implemented in Gaussian 09 based on the gas-phase geometries with dielectric constants of ε - 78.4 and 35.7 for water and MeCN, respectively, for the continuum solvating medium and using the cc-pVTZ basis set for all atoms except Ni for which the LANL2DZ basis set was used. We used experimentally determined solvation free energies of AG^H+so (aq) = -264.61 kcal mol^"' and AG^H+ _solv(MeCN) = -260.20 kcal mol^"1 for the proton^!xix and AG^Br" _soiv(aq) = -70.75 kcal mol^"1 for bromide.^lxx The latter is an average of three experimental free energies of solvation discussed in Ref. 42.

Synthesis and Characterization of New Compounds and Experimental Procedure Synthesis of [{(MesIm)₂Py}NiBr]⁺[TFA]^" (1)

This complex was prepared from 2,6-bis(3-mesitylimidazol-2-ylidene)pyridine disilver dibromide³⁵ using a modified procedure based on that reported by K. Inamoto et. al.³⁵ To a suspension of NiBr₂(DME) (37.0 mg, 0.12 mmol) in 14 mL of CH₂C1₂ under a stream of argon was added the bis-carbene disilver dibromide (100 mg, 0.12 mmol, 1 equiv.) dissolved in 5 mL of CH₂C1₂ at 0 °C. The resulting light brown reaction mixture was stirred for 16 h at room temperature in the dark, followed by addition of silver trifluoroacetate (31.2 mg, 0.14 mmol, 1.2 equiv.) After an additional 24 h of stirring the yellow/brown mixture was transferred to a pad of celite via canula and filtered to remove insoluble silver bromide. The filtrate was concentrated in vacuo and washed three times with 5 mL of Et₂0 to remove excess Ag0₂CCF₃, giving 1 as a light brown solid. (56 mg, 72% Yield). Anal, found (calcd) for C₃₁H₂₉BrF₃N₅Ni0₂: C 53.52 (53.25), H 3.94 (4.18), N 10.09 (10.02) %. 1H NMR (400 MHz, CD₂C1₂): δ 8.46 (2H, br, Ar-CH_ta), 8.42 (1H, br t, Ar-CH_py), 8.10 (2H, br d, Ar-CH_py), 6.92 (2H ,br, Ar-CHi_m), 6.90 (4H, s, Ar-CH_Mes), 2.28 (6H, s, Ar-p-G¾), 2.07 (12H, s, Ar-o- CH pm. ^uC-{^lH} NMR (125.8 MHz, CD₂C1₂): 6 164.1, 151.7, 140.2, 134.8, 130.4, 130.4, 129.2, 127.3, 119.1, 110.4, 21.3, 18.1 ppm.

Synthesis of [(PCP)Ni(NCCH₃)]⁺ [BF₄]^" ([3'(MeCN)]⁺)

To [(PCP)NiCl] (3) (40 mg, 82 μηιοΐ) and AgBF₄ (16 mg, 82 μηιοΐ) was added 1 mL of acetonitrile at 25 °C. The reaction was stirred overnight and then filtered through celite. The solvent removed was in vacuo to give [(PCP)Ni(NCCH₃)]⁺[BF₄]^" ([3'(MeCN)]⁺ as a yellow- orange powder. (47 mg, 99% Yield). Single crystals suitable for x-ray crystallography were grown from saturated acetonitrile solutions containing [3'(MeCN)]⁺ (see SI for more details). HR FT-ICR MS: Found (calcd for C₂₆H₄₆NNiP₂): m/z = (M)⁺ 492.2467 (492.2459), (M- MeCN)⁺ 451.2193 (451.2196). 1H NMR (NCCD₃, 500.0 MHz): δ 6.94 - 6.98 (3H, m, AxH), 3.32 (4H, t, PC¾Ar, J = 4.2Hz), 1.37 (36H, t, PC(C¾)₃, J = 6.6Hz). ¹³C-{1H} NMR

(NCCD₃, 125.8 MHz): 6154.6 (t, J = 11.4Hz), 153.0 (t, J = 13.9Hz), 135.9 (s), 128.0 (s), 123.4 (t, J = 9.1Hz), 35.9 (t, J = 7.6Hz), 33.4 (t, J = 13.7Hz), 29.7 (s). ³¹P-{1H} NMR

(NCCD₃, 161.9 MHz): 80.1ppm.

Protonation of [(PCP)NiH] ([3'(H)])

To [(PCP)NiH] ([3'(H)]) (3 mg, 6.2 μπιοΐ) dissolved in 0.5 mL d₃-acetonitrile in a J. Young NMR tube was added 1 eq of HBF₄ in 10 iL H₂0 at 25°C. Mixing of the two solutions occurred only after the NMR cap was replaced. H NMR spectroscopy displayed a characteristic resonance for H₂ at δ 4.57 along with peaks corresponding to

[(PCP)Ni(NCCH₃)]⁺ [BF₄]^" ([3'(MeCN)]⁺)^Ixxi (see above for 1H NMR chemical shifts).

NMR of 4 in </₆-DMSO

Compound 4 (4 mg) was dissolved in 0.6 mL i¾-DMSO (Cambridge Isotope Laboratories ampule). 1H NMR (400 MHz, i ₆-DMSO) δ 8.50 (d, J - 7.8, 2H, pyridine _META H), 8.16 (t, J = 7.8, , pyridine _para._H), 7.13 (d, J = 7.5, 4K, Ar„eta H), 6.96 (t, J = 7.5, 2R Ar_{para H}), 2.22 (s, 6H, Me Acetyl , 2.01 (s, 12H, Me _Ar). Supporting Information Available

Additional cyclic voltammograms, further experimental details, details of the crystal structure determination of [3'(MeCN)]⁺ and xyz coordinates and energies of optimized structures are available free of charge via the Internet at http://pubs.acs.org.

References for Example 4

² Navarro, R. M.; Pena, M. A.; Fierro, J. L. G. Chem. Rev. 2007, 107, 3952-3991.

³ Artero, V.; Fontecave, M. Coord. Chem. Rev. 2005, 249, 1518-1535.

⁵ (a) Bockris, J. O'M.; Conway, B. E. Trans. Faraday Soc, 1949, 45, 989-999. (b) Ezaki, H.; Morinaga, M.; S. W atanabe Electrochimica Acta, 1993, 38, 557-564.

⁶ (a) Koelle, U.; Ohst, S. Inorg. Chem. 1986, 25, 2689. (b) Connolly, P.; Espenson, J. H. Inorg. Chem. 1986, 25, 2684-2688. (c) Chao, T-H.; Espenson, J.H. J. Am. Chem. Soc , 1978, 100, 129-133. (d) Kellett, R. M.; Spiro, T. G. Inorg. Chem. 1985, 24, 2373-2377. (e) Hu X.; Brunschwig B. S.; Peters J. C. J. Am. Chem. Soc 2007, 129, 8988-8998. (f) Szymczak, N. K.; Berben, L. A.; Peters, J. C. Chem. Comm. 2009, 6729-6731. (g) Berben, L. A.; Peters, J.C. Chem. Comm. 2010, 398-400. (h) Baffert, C; Artero, V.; Fontecave, M. Inorg. Chem., 2007, 46, 1817-1824. (i) Bernhardt, P. V.; Jones, L. A. Inorg. Chem. 1999, 38, 5086-5090. Q) Pantani, O., Naskar, S., Guillot, R., Millet, P., Anxolabehere-Mallart, E. and Aukauloo, A. Angew. Chem. Intl. Ed. 2008, 47, 9948-9950. (k) Razavet, M.; Artero, V.; Fontecave, M. Inorg. Chem, 2005, 44, 4786-4795. (1) Fihri, A.; Artero, V.; Razavet, M.; Baffert, C; Leibl, W.; Fontecave, M. Angew. Chem. Intl. Ed. 2008, 47, 564-567. (m) Dempsey, J. L.;

Brunschwig, B. S.; Winkler, J. R., Gray, H. B. Acc. Chem. Res. 2009, 42, 1995-2004. (n) Dempsey, J. L.; Winkler, J. R.; Gray, H. B. J. Am. Chem. Soc. 2010, 132, 16774-16776. (o) Dempsey, J. L.; Winkler, J. R.; Gray, H. B. J. Am. Chem. Soc. 2010, 132, 1060-1065. (p) Du, P.; Schneider, J.; Luo, G.; Brennessel, W.W.; Eisenberg, R. Inorg. Chem., 2009, 48, 4952- 4962. (q) Du, P.; Knowles, K.; Eisenberg, R. J. Am. Chem. Soc , 2008, 130, 12576-12577. (r) Fihri, A.; Artero, V.; Pereira, A.; Fontecave, M. Dalton Trans. 2008, 41, 5567-5569. (s) Probst, B.; Kolano, C. Hamm, P.; Alberto, R. Inorg. Chem., 2009, 48, 1836-1843. (t) Lakadamyali, F.; Erwin, R. Chem Commun, 2011, 47, 1695-1697. (u) Szajna-Fuller, E. and Bakac, A. Eur. J. Inorg. Chem. 2010, 2488-2494 (v) Pantani, O.; Anxolabehere-Mallart,E.; Aukauloo, A.; Millet, P. Electrochem. Commun. 2007, 9, 54-58. ⁷ (C lvk^-S)2S2CH2 as electrocatalyst for H₂ production: Appel, A. M., DuBois, D. L.;

Rakowski-DuBois M. J Am. Chem. Soc. 2005, 127, 12717-12726.

[(Py5Me₂)Mo(CF₃S0₃)]⁺ as electrocatalyst for H⁺ reduction: Karunadasa, H. I.; Chang, C. J.; Long, J. R. Nature, 2010, 464, 1329-1333.

^I0(a) [Ni(P^ph ₂N^C6H4x ₂)₂]²⁺ Complexes as Electrocatalysts for H₂ Production: Kilgore, U. J.; Roberts, J. A. S.; Pool, D. H.; Appel, A. M.; Stewart, M. P.; Rakowski-DuBois, M.;

Dougherty, W. G.; Kassel, W. G.; Bullock, R. M.; DuBois, D. L. J. Am. Chem. Soc, 2011, 133 , 5861-5872. (b) Similar complexes have also been shown to perform C0₂

electroreduction as well as H₂ production: Rakowski-DuBois, M.; DuBois D. L. Acc. Chem. Res. 2009, 42, 1974-1982.

¹¹ Ni and Co cobaloximes: Jacques, P.-A.; Artero, V. ; Pecaut, J.; Fontecave, M. Proc. Natl. Acad. Sci. 2009,106, 20627-20632.

12 Nickel-Iron Dithiolato Hydrides as electrocatalysts: Barton, B. E.; Whaley, M. C;

Rauchfuss, T. B.; Gray, D. L. J. Am. Chem. Soc. 2009, 131, 6942-6943.

¹³ Bimetallic Iron carbonyl dithiolates as H₂ production electrocatalyts: Gloaguen, F.;

Lawrence, J. D.; Rauchfuss, T. B.; Benard, M.; Rohmer M. Inorg. Chem. 2002, 41, 6573- 6582.

¹⁴ (a) Gloaguen, F.; Rauchfuss, T. B. Chem. Soc. Rev. 2009, 38, 100-108. (b) Losse, S. ; Vos, J. G.; Rau, S. Coord. Chem. Rev., 2010, 254, 21-22, 2492-2504.

¹⁵ (a) Sellmann, D.; Rackelmann, G.H.; Heinemann, F.W. Chem. Eur. J. 1997, 3, 2071. (b) Lough, A.J.; Park, S.; Ramachandran, R.; Morris, R. H. J. Am. Chem. Soc. 1994, 116, 8356. (c) Morris, R. H. Can. J. Chem. 1996, 74, 1907.

¹⁶ For a series of papers discussing the role of proton relays in Ni electrocatalysis: (a) Jain, A.; Lense, S.; Linehan, J. C; Raugei, S.; Herman, C; DuBois, D. L. Inorg. Chem. 2011, 50, 4073. (b) DuBois, D. L.; Morris, B. R. Eur. J. Inorg. Chem. 2011, 1017-1027. (c)Yang, J. Y.; Shentan, C; Kassel, W. S.; Bullock, R. M.; DuBois, D. L.; Simone, R.; Rousseau, R.;

DuPuis, M.; Rakowski-DuBois, M. Chem. Comm. 2010, 46. (d) Rakowski-DuBois, M.;

DuBois, D. M.; Chem. Soc. Rev. 2009, 38, 62-72. (e) Yang, J. Y.; Bullock, R. M.; Shaw, W. J.; Twamley, B.; Fraze, K.; DuBois, D. L.; Simone, R.; Rousseau, R.; DuPuis, M.; Rakowski- DuBois, M. /. Am. Chem. Soc. 2009, 131, 5935-5945. Comparison between a Co and the analogous Ni system was also made available: (f) Wiedner, E. S.; Yang, J. Y,; Dougherty, W. J.; Kassel, W. Bullock, R. M.; Rakowski-DuBois, M.; DuBois, D. L. Organometallics 2010, 29, 5390-54

¹⁷ The original work depicts electron transfer in the oxidation direction: Curtis, C. J.;

Miedaner, A.; Ciancanelli, R.; Ellis, W. W.; Noll, B. C; Rakowski-DuBois, M; DuBois, D. L. Inorg. Chem., 2003, 42, 216-227.

¹⁸ Le Goff, A,; Artero, V.; Jousselme, B.; Tran, P. D.; Guilet, N.; Metaye, R.; Fihri, A.;

Palacin, S.; Fontecave, M. Science 2009, 326, 1384-1387.

¹⁹ McLaughlin, M. P.; McCormick, T. M.; Eisenberg, R.; Holland, P. L. Chem. Commun. , 2011, 47, 7989-7991.

²⁰ (a) O'Hayre, R. P.; Cha, S.-W.; Coellela, W. G.; Prinz, F. B. Fuel Cell Fundamentals 2009, 2^nd Edition, Wiley and Sons, (b) Dinca, M., Long J.R. Angew. Chem. Int. Ed. 2008, 47, 6766 - 6779(c) For our concept of Virtual Hydrogen Storage: Crabtree, R. H. Energy Environ. Set, 2008, 1, 134. (d) Luca O. R.; Wang T.; Konezny S. J.; Batista V. S.; Crabtree R. H. New J. Chem., 2011, 35, 998-999 for proof of principle work. An organic analyte causes

incompatibilities with the proton exchange membrane (swelling) and also adds resistance thus decreasing the overall voltage output of our cell.

²¹ Sun, Y.; Bigi, J. P.; Piro, N. A.; Tang, M. L.; Long, J. R.; Chang, C. J. J. Am. Chem. Soc, 2011, 735, 9212-9215.

²² David Morales-Morales; Craig Jensen; The Chemistry of Pincer Compounds 2007,

Elsevier.

²³ Inamoto, K.; Kuroda, I.; Kwon, E.; Hiroya, K.; Doi, T. J. Organomet. Chem. 2009, 694, 389-396.

²⁴ Gomez-Benitez, B.; Baldovino-Pantaleon, O.; Herrera-Alvarez, C; Toscano, R. A.;

Morales-Morales, D. Tett. Lett. 2006, 47, 5059.

²⁵ Boro, B. J.; Duesler, E. N.; Goldberg, K. I., Kemp, R. A. Inorg. Chem. 2009, 48, 5081- 5087.

²⁶ Suzuki, H.; Matsumura, S.; Satoh, Y.; Sogoh, K.; Yasuda, H. React. Funct. Polym. 2004, 77-91.

²⁷ Manuel, T. D.; Rohde J.-U. J. Am. Chem. Soc. 2009, 131, 15582-15583.

²⁸ (a) Hu, X.; Castro-Rodriguez, I.; Meyer, K. Chem. Commun. 2004, 19, 2164-2165. (b) An example of NiⁿH species supported by a meridional tri dentate ligand has been reported: He, T.; Andino, J. G.; Gao, X.; Fullmer, B. C; Caulton, K. G. J. Am. Chem. Soc. 2010, 132, 910- 911. Two recent reports of redox active carbenes: (a) Tennyson, A. G.; Lynch, V. M.;

Bielawski, C. W. J. Am. Chem. Soc, 2010, 132, 9420-9429. (b) Dzik, W. I.; Zhang, X. P.; de Bruin, B. Inorg. Chem., Article ASAP, DOI: 10.1021/ic200043a.

Zhu, .; Achord, P. D.; Zhang, X.; Krogh-Jespersen, K.; Goldman, A. S. J. Am. Chem. Sbc., 2004, 126, 13044-13053.

³¹ (a) Pool, D. H.; DuBois, D. L. J. Organomet. Chem., 2009, 694, 2858-2865. (b) Wilson, A. D.; Newell, R. H.; McNevin, M. J.; Muckerman, J.T.; Rakowski DuBois, M.; DuBois, D. L. J. Am. Chem. Soc, 2006, 128, 358-366.

³² Blakemore, J. D.; Schley, N. D.; Balcells, D.; Hull, J. F.; Olack, G. W.; Incarvito, C. D.; Eisenstein, O.; Brudvig, G. W.; Crabtree, R. H. J. Am. Chem. Soc. 2010, 132, 16017-16029.

³³ Data were smoothed using the 5 point adjacent averaging in Origin 8.5.

³⁴ Peris, E.; Lee Jr., J. C; Rambo, J. R.; Eisenstein, O.; Crabtree, R. H. J. Am. Chem. Soc. 1995, 77, 3485;

³⁵ Lough, A. J.; Park, S.; Ramachandran, R.; Morris, R. H. J. Am. Chem. Soc. 1994, 116, 8356.

³⁶ (a) Chirik P. J.; Wieghardt, K. Science 2010, 327, 794-795. (b) Bouwkamp, M. W.;

Bowman, A. C; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc, 2006, 725, 13340-13341. (c) Budzelaar, P. H. M.; de Bruin, B.; Gal, A. W. Inorg. Chem., 2001, 40, 4649-4655. (d) Sazama, G. T.; Betley, T. A. Inorg. Chem. 2010, 49, 2512-2524. (e) A case of a redox active ligand participation in olefin binding was highlighted: Crabtree, R. H. Science 2001, 106, 56- 57. (f) Wang, K; Stiefel, E. I. Science, 2001, 297, 106-109. The data analysis in this paper was revised by (g) Geiger, W. E. Inorg. Chem., 2002, 41, 136-139.

³⁷ (a) Huynh, M. H. V.; Meyer, T. J. Chem. Rev. 2007, 107, 5004-5064. (b) Costetin, C; Robert, M.;

Saveant, J.-M. Acc. Chem. Res. 2010, 43, 1019-1029.

We did not see electrodeposition processes. Background runs performed after running bulk electrolysis and voltammetry do not show any new features on the electrode surface for any of the catalysts. Backgrounds are available in the SI.

³⁹ Ag/AgCl Cl_Sat = 0.197 vs. NHE; Bard, A.; Faulkner, L. R. Electrochemical Methods Fundamentals and Applications, 2004, 2^nd Edition, Wiley and Sons.

⁴⁰ Gaussian 09, Revision A.l, Frisch, M. J.; et al. Gaussian, Inc., Wallingford CT, 2009. See the SI for the complete reference.

⁴¹ (a) In water: Liptak, M. D.; Shields, G. C. J. Am. Chem. Soc. 2001, 123, 7314-7319. (b) In acetonitrile: Kelly, C. P.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B 2007, 777, 408-422. ⁴² Hummer, G.; Pratt, L. R.; Garca, A. E. J. Phys. Chem. 1996, 100, 1206-1215.

⁴³ Fulmer, G. R; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.; Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I. Organometallics, 2010, 29, 2176-2179.

Claims

What is claimed is:

1. A process comprising electrocatalytically reducing a liquid organic hydrogen carrier (LOHC) in a direct organic fuel cell/flow battery by contacting the LOHC with a

homogeneous redox catalyst under aqueous or non-aqueous acidic conditions and at a temperature of between about 100° C to about 300° C, said catalyst comprising a tndendate, redox-active pincer ligand complex containing a group 10 transition metal and one or two coordinating groups bound to said transition metal.

2. The process of claim 1, wherein the group 10 transition metal is selected from the group consisting of Ni, Pd, Pt, and Ds.

3. The process of claim 2, wherein there are two coordinating groups, each of which is a halogen.

The process of claim 3, wherein each coordinating group

The process of claim 1, wherein the pincer ligand complex is defined by the formula

X_\ and X₂ are the same or different and are a halogen, provided that either (but not both) of Xi or X₂ may be absent;

Ri and R₂ are independently selected from the group consisting of a C₅-Ci₄ aryl or a C₅-Ci heteroaryl, said C₅-C₁₄ aryl or a C₅-CH heteroaryl being optionally substituted with one or more substituents selected from the group consisting of Ci-C₆ alkyl, Ci-C₆ alkenyl, and Ci-Ce alkynyl;

R₃ and R4 are independently selected from the group consisting of optionally substituted Cj- C₆ alkyl, optionally substituted Ci-C^ alkenyl, and optionally substituted CrC₆ alkynyl;

or Ri and R₃ and R₂ and R4, together with the nitrogen to which they are bound, form an optionally substituted, single ring or fused two or three ring C₅-Ci₄ heteroaryl, said single ring heteroaryl optionally containing one or two additional ring heteroatoms selected from the group consisting of N and 0, and said two or three fused ring heteroaryl optionally containing between one, two, or three additional ring heteroatoms selected from the group consisting of N and O;

R₅ and ¾ are absent or are independently selected from the group consisting of substituted or unsubstituted Ci-C₆ alkyl, Ci-C₆ alkenyl, and Ci-C₆ alkyny, or R₅ and R , together with the carbon atoms to which they are bound, form an optionally substituted C₅-C₁₀ aryl or a C₅-Ci₀ heteroaryl; and

n is 0 or 1.

6. The process of claim 5, wherein: M is Ni; Xi and X₂ are Br; and R₅ and R^ are absent.

7. The process of claim 1, wherein the pincer ligand complex is defined by the formula

(II)

8. The process of claim 1 , wherein the pincer ligand complex is defined by the formula (III):

9. The process of claims 7 or 8, wherein the LOHC is contacted with the homogeneous redox catalyst in an aqueous HC1 solution.

10. The process of claim 1, wherein subsequent to the reduction of the LOHC, reduced LOHC is electrohydrogenated in the presence of the redox catalyst using protons generated by electrooxidation of water at a cathode of the direct organic fuel cell/flow battery.

1 1. A process comprising electrocatalytically reducing a liquid organic hydrogen carrier (LOHC) in a direct organic fuel cell/flow battery by contacting the LOHC with a

homogeneous redox catalyst under aqueous or non-aqueous acidic conditions and at a temperature of between about 100° C to about 300° C, said catalyst comprising a tridendate, redox-active pincer ligand complex having the formula (I)(A):

Ri and R₂ are independently selected from the group consisting of a C₅-Q₄ aryl or a C₅-Ci₄ heteroaryl, said C₅-Ci₄ aryl or a C₅-Ci₄ heteroaryl being optionally substituted with one or more substituents selected from the group consisting of Ci-C₆ alkyl, Q-Ce alkenyl, and Q-C₆ alkynyl;

R₃ and R₄ are independently selected from the group consisting of optionally substituted Q- C₆ alkyl, optionally substituted Ci-C₆ alkenyl, and optionally substituted Ci-C₆ alkynyl; or Ri and R₃ and R₂ and R4, together with the nitrogen to which they are bound, form an optionally substituted, single ring or fused two or three ring C₅-C₁₄ heteroaryl, said single ring heteroaryl optionally containing one or two additional ring heteroatoms selected from the group consisting of N and O and said two or three fused ring heteroaryl optionally containing between one, two, or three additional ring heteroatoms selected from the group consisting of N and O;

R₅ and R6 are absent or are independently selected from the group consisting of substituted or unsubstituted Ci-C₆ alkyl, Q-Q alkenyl, and Ci-C₆ alkyny, or R₅ and R^, together with the carbon atoms to which they are bound, form an optionally substituted C₅-C₁₀ aryl or a Cs-Cio heteroaryl; n is 0 or 1 ; and

Y is selected from the group consisting of a Lewis acid, a complex Lewis acid, and the ligand of a coordination compound with the metal acting as the Lewis Acid.

12. A direct organic fuel cell/flow battery fuel cell system in which a liquid organic hydrogen carrier (LOHC) is electrocatalytically reduced by contacting the LOHC with a homogeneous redox catalyst under aqueous or non-aqueous acidic conditions and at a temperature of between about 100° C to about 300° C, said catalyst comprising a tridendate, redox-active pincer ligand complex containing a group 10 transition metal and one ot two coordinating groups bound to said transition metal.

13. The system of claim 12, wherein subsequent to the reduction of the LOHC, the reduced LOHC is electrohydrogenated in the presence of the redox catalyst using protons generated by electrooxidation of water at a cathode of the direct organic fuel cell/flow battery.

14. The process of claims 1-8, wherein the LOHC is a partially or fully hydrogenated, nitrogen-containing aromatic heterocycle.

15. The process of claim 14, wherein the nitrogen-containing aromatic heterocycle is selected from the group consisting of decalin, 2-aminopyridine, 4-methylpyrimidine, dipyrimidinemethane, dimethyltetrazine, dipirimidine, diazacarbazole, alkylcarbazole, 4- aminopyridine, dipyrazinemethane, tripyrazinemethane, tripyrazineamine, dipyrazine, tetrazacarbazole, isoquinoline, di(2-pyridyl)amine, quinazoline, perhydro-N-ethylcarbazole, tetrahydroisoquinoline, perhydro-dibenzofuran, perhydro-indole, N-methyl perhydro-indole, 4, 4'-bipiperidine, 4-amino-methylpyrimidine, and aminomethylcyclohexane, or a combination of two or more of the foregoing.

16. The process of claim 12, wherein the LOHC is a partially or fully hydrogenated, nitrogen-containing aromatic heterocycle.

17. The process of claim 16, wherein the nitrogen-containing aromatic heterocycle is selected from the group consisting of decalin, 2-aminopyridine, 4-methylpyrimidine, dipyrimidinemethane, dimethyltetrazine, dipirimidine, diazacarbazole, alkylcarbazole, 4- aminopyridine, dipyrazinemethane, tripyrazinemethane, tripyrazineamine, dipyrazine, tetrazacarbazole, isoquinoline, di(2-pyridyl)amine, quinazoline, perhydro-N-ethylcarbazole, tetrahydroisoquinoline, perhydro-dibenzofuran, perhydro-indole, N-methyl perhydro-indole, 4, 4'-bipiperidine, 4-amino-methylpyrimidine, and aminomethylcyclohexane, or a

combination of two or more of the foregoing.

18. The process of claim 11 , wherein the LOHC is a partially or fully hydrogenated, nitrogen-containing aromatic heterocycle.

19. The process of claim 18, wherein the nitrogen-containing aromatic heterocycle is selected from the group consisting of decalin, 2-aminopyridine, 4-methylpyrimidine, dipyrimidinemethane, dimethyltetrazine, dipirimidine, diazacarbazole, alkylcarbazole, 4- aminopyridine, dipyrazinemethane, tripyrazinemethane, tripyrazineamine, dipyrazine, tetrazacarbazole, isoquinoline, di(2-pyridyl)amine, quinazoline, perhydro-N-ethylcarbazole, tetrahydroisoquinoline, perhydro-dibenzofuran, perhydro-indole, N-methyl perhydro-indole, 4, 4'-bipiperidine, 4-amino-methylpyrimidine, and aminomethylcyclohexane, or a

combination of two or more of the foregoing.

20. A supported homogeneous redox catalyst comprising a tridendate, redox-active pincer ligand complex containing a group 10 transition metal and one ot two coordinating groups bound to said transition metal.

21. The catalyst of claim 20, wherein the group 10 transition metal is selected from the group consisting of Ni, Pd, Pt, and Ds.

22. The catalyst of claim 20, wherein there are two coordinating groups, each of which is a halogen.

23. The catalyst of claim 22, wherein each coordinating group is Br.

24. The catalyst of claim 20, wherein the pincer ligand complex is defined by the formula (I):

Xi and X₂ are the same or different and are a halogen, provided that either (but not both) of Xi or X₂ may be absent;

Ri and R₂ are independently selected from the group consisting of a C₅-Ci₄ aryl or a C5-C₁₄ heteroaryl, said C5-C14 aryl or a C₅-C₁₄ heteroaryl being optionally substituted with one or more substituents selected from the group consisting of C[-C₆ alkyl, C[-C₆ alkenyl, and C C₆ alkynyl;

R₃ and R4 are independently selected from the group consisting of optionally substituted Cj- C₆ alkyl, optionally substituted Ci-C₆ alkenyl, and optionally substituted Ci-C₆ alkynyl; or Ri and R₃ and R₂ and R4, together with the nitrogen to which they are bound, form an optionally substituted, single ring or fused two or three ring C₅-Ci₄ heteroaryl, said single ring heteroaryl optionally containing one or two additional ring heteroatoms selected from the group consisting of N and 0 and said fused two or three ring heteroaryl optionally containing between one, two, or three additional ring heteroatoms selected from the group consisting of N and 0;

R₅ and Rs are absent or are independently selected from the group consisting of substituted or unsubstituted Ci-C₆ alkyl, Ci-C₆ alkenyl, and Ci-C₆ alkyny, or R₅ and ¾, together with the carbon atoms to which they are bound, form an optionally substituted C₅-Cio aryl or a C₅-C₁₀ heteroaryl;

and n is 0 or 1.

25. The catalyst of claim 24, wherein: M is Ni; Xj and X₂ are Br; and R₅ and R₆ are absent. The catalyst of claim 24, wherein the pincer ligand complex has the formula (II)

28. A supported homogeneous redox catalyst comprising a tridendate, redox-active pincer ligand complex pincer ligand complex having the formula (I)(A):

Ri and R₂ are independently selected from the group consisting of a C5- 4 aryl or a C₅-Ci₄ heteroaryl, said C₅-Ci₄ aryl or a C₅-Ci₄ heteroaryl being optionally substituted with one or more substituents selected from the group consisting of Ci-C alkyl, C[-C alkenyl, and Q-Q alkynyl;

R₃ and R4 are independently selected from the group consisting of optionally substituted Q- C₆ alkyl, optionally substituted C[-C₆ alkenyl, and optionally substituted Ci-C₆ alkynyl; or Ri and R₃ and R₂ and R4, together with the nitrogen to which they are bound, form an optionally substituted, single ring or fused two or three ring C5-Q4 heteroaryl, said single ring heteroaryl optionally containing one or two additional ring heteroatoms selected from the group consisting of N and O and said fused two or three ring heteroaryl optionally containing between one, two, or three additional ring heteroatoms selected from the group consisting of N and O;

R₅ and ¾ are absent or are independently selected from the group consisting of substituted or unsubstituted Q-C₆ alkyl, Ci-Ce alkenyl, and Cj-C₆ alkyny, or R₅ and R₆, together with the carbon atoms to which they are bound, form an optionally substituted C₅-Cio aryl or a C5-C₁0 heteroaryl; n is 0 or 1 ; and

29. The catalyst of claim 20, wherein the pincer ligand complex is selected from the group of compounds of the formula:

The catalyst of claim 20, wherein the pincer ligand complex is defined by the formula

Ri and R are independently selected from the group consisting of a C5-C14 aryl or a C₅-C₁₄ heteroaryl, said C5-Q4 aryl or a C5-C14 heteroaryl being optionally substituted with one or more substituents selected from the group consisting of Ci-C₆ alkyl, C[-C₆ alkenyl, and Cj-C₆ alkynyl; and R₅ and R are absent or are independently selected from the group consisting of substituted or unsubstituted C_\-Ce alkyl, Ci-C₆ alkenyl, and Ci-C₆ alkynyl.

31. A supported homogeneous redox catalyst comprising a tridendate, redox-active pincer ligand complex having the formula (V):

R₁₀, Rn, and R₁₂ are selected from the group consisting of mono or di-Ci-C6 alkyl, mono or di-Ci-C₆ alkenyl, and mono or di-Q-Ce alkynyl (for Ri₀ and Rn, most preferably (tBu)₂, and for R₁₂, most preferably Me); and

Y is a Lewis, a complex Lewis acid, or the ligand of a coordination compound with the metal acting as the Lewis Acid.

32. The catalyst of claim 30, wherein R₁₀ and Rn are (tBu)₂, and R₁₂ is Me.

33. A process comprising electrocatalytically reducing a liquid organic hydrogen carrier (LOHC) in a direct organic fuel cell/flow battery by contacting the LOHC with a

homogeneous redox catalyst under aqueous or non-aqueous acidic conditions and at a temperature of between about 100° C to about 300° C, said catalyst comprising a tridendate, redox-active pincer ligand complex having the formula (IV):

Ri R2 (iv)

X₁ and X₂ are the same or different and are a halogen, provided that either (but not both) of Xi or X₂ may be absent;

Ri and R₂ are independently selected from the group consisting of a C₅-Ci₄ aryl or a C₅-Ci₄ heteroaryl, said C₅-Ci₄ aryl or a C₅-Ci₄ heteroaryl being optionally substituted with one or more substituents selected from the group consisting of Cr alkyl, Cj-Q alkenyl, and C i-C alkynyl; and

R5 and P6 are absent or are independently selected from the group consisting of substituted or unsubstituted Ci-C₆ alkyl, Ci-C₆ alkenyl, and Ci-C₆ alkynyl.

34. A process comprising electrocatalytically reducing a liquid organic hydrogen carrier (LOHC) in a direct organic fuel cell/flow battery by contacting the LOHC with a

homogeneous redox catalyst under aqueous or non-aqueous acidic conditions and at a temperature of between about 100° C to about 300° C, said catalyst comprising a tridendate, redox-active pincer ligand complex having the formula (V):

wherein M is a group 10 transition metal selected from the group consisting of Ni, Pd, Pt, and

Ds;

Rio, Ri i, and Ri₂ are selected from the group consisting of mono or di-Ci-C₆ alkyl, mono or di-C₁-C₆ alkenyl, and mono or di-Ci-C₆ alkynyl; and

35. The catalyst of claim 33, wherein Rio and Rn are (tBu)₂, and R_i2 is Me.

36. The process of claim 1 , wherein the pincer ligand complex is selected from the group consisting of compounds 1-4, where x is 1 , 2, or 3 :

37. The catalyst of claim 20, wherein the pincer ligand complex is selected from the group consisting of compounds 1-4, where x is 1 , 2, or 3 :

38. A composition comprising a partially or fully hydrogenated, nitrogen-containing aromatic heterocycle as a liquid organic hydrogen carrier in combination with a catalyst according to any of claims 20-32.

39. The composition according to claim 38 wherein said heterocycle is selected from the group consisting of decalin, 2-aminopyridine, 4-methylpyrimidine, dipyrimidinemethane, dimethyltetrazine, dipirimidine, diazacarbazole, alkylcarbazole, 4-aminopyridine, dipyrazinemethane, tripyrazinemethane, tripyrazineamine, dipyrazine, tetrazacarbazole, isoquinoline, di(2-pyridyl)amine, quinazoline, perhydro-N-ethylcarbazole,

tetrahydroisoquinoline, perhydro-dibenzofuran, perhydro-indole, N-methyl perhydro-indole, 4, 4'-bipiperidine, 4-amino-methylpyrimidine, aminomethylcyclohexane, 1,2,3,4- tetrahydroquinaldine, 2,3-dimethyltetrahydroquinoxaline or a mixture thereof.

40. The composition according to claim 39 wherein said heterocycle is 1,2,3,4- tetrahydroquinaldine, 2,3-dimethyltetrahydroquinoxaline or a mixture thereof.

41. A composition comprising a homogeneous redox catalyst in combination with 1,2,3,4-tetrahydroquinaldine, 2,3-dimethyltetrahydroquinoxaline or a mixture thereof.

42. The process of claim 1, wherein said transition metal is nickel and subsequent to the reduction of the LOHC, the LOHC is dehydrogenated to release energy from the direct organic fuel cell/flow battery.

43. The process of claim 42 wherein said LOCH is dehydrogenated in the presence of nanoparticles of nickel oxide (Ni0₂) in solution or as a solid at an anode of the direct organic fuel cell/flow battery.

44. The process of claim 1, wherein said dehydrogenated LOHC is subsequently electrohydrogenated in the presence of the redox catalyst using protons generated by electrooxidation of water at a cathode of the direct organic fuel cell/flow battery.

45. The process of claim 42, wherein said dehydrogenated LOHC is subsequently electrohydrogenated in the presence of the redox catalyst using protons generated by electrooxidation of water at a cathode of the direct organic fuel cell/flow battery.

46. The process of claim 43, wherein said dehydrogenated LOHC is subsequently electrohydrogenated in the presence of the redox catalyst using protons generated by electrooxidation of water at a cathode of the direct organic fuel cell/flow battery.