Ruthenium H2

Article pubs.acs.org/IC

Solvent-Dependent Thermochemistry of an Iridium/Ruthenium H2 Evolution Catalyst Kelsey R. Brereton,† Catherine L. Pitman,† Thomas R. Cundari,‡ and Alexander J. M. Miller*,† †

Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3290, United States Department of Chemistry and CASCaM, University of North Texas, Denton, Texas 76203, United States

‡

S Supporting Information *

ABSTRACT: The hydricity of the heterobimetallic iridium/ ruthenium catalyst [Cp*Ir(H)(μ-bpm)Ru(bpy)2]3+ (1, where Cp* = η5-pentamethylcyclopentadienyl, bpm = 2,2′-bipyrimidine, and bpy = 2,2′-bipyridine) has been determined in both acetonitrile (63.1 kcal mol−1) and water (29.7 kcal mol−1). Hydride 1 features a large increase in the hydride donor ability when the solvent is changed from acetonitrile to water. The acidity of 1, in contrast, is essentially solvent-independent because 1 remains strongly acidic in both solvents. On the basis of an X-ray crystallographic study, spectroscopic analysis, and time-dependent density functional theory calculations, the disparate reactivity trends are ascribed to substantial delocalization of the electron density onto both the bpm and bpy ligands in the conjugate base of 1, [Cp*Ir(μ-bpm)Ru(bpy)2]2+ (3). The H2 evolution tendencies of 1 are considered in the context of thermodynamic parameters.

■

INTRODUCTION The bimetallic hydride [Cp*Ir(H)(μ-bpm)Ru(bpy)2]3+ (1, Scheme 1) has been implicated as an intermediate in the

water spurred us toward thermochemical and electronic structure studies. The transformations provided in Scheme 1 are all proposed to involve an iridium hydride intermediate. The thermochemistry of the metal−hydride bond may therefore provide valuable insight into the catalytic reactivity. The thermodynamic hydricity (ΔG°H−),13 the free energy required to heterolytically cleave the metal−hydride bond and release H−, can predict or explain the reactivity of metal hydride complexes.14−20 Several experimental methods to measure ΔG°H− have been established, including the common practice of establishing equilibrium with H2 or another hydride donor/acceptor pair of known hydricity.21,22 Alternatively, a “potential−pKa” thermochemical cycle can provide hydricity values without the need for a known reference species. Equations 1−4 show how the hydricity (eq 4) can be determined through measurement of the pKa of the hydride (eq 1) and the reduction potential of its conjugate base (eq 2), along with the free energy for the reduction of a proton to a hydride (eq 3).

Scheme 1. Reactions Catalyzed by 1 in Water

catalytic dehydrogenation of formic acid to CO2 and H2,1 the 4e− reduction of oxygen,2 and the photocatalytic production of H2 in conjunction with a [Ru(bpy)3]2+ photosensitizer3 in an aqueous solvent. In formic acid dehydrogenation, the bimetallic catalyst (as the sulfate salt) produces H2 with turnover frequencies greater than 400 h−1 via an unusual hydrogen tunneling mechanism.1 The photocatalytic production of H2 from 1 was not observed without additional [Ru(bpy)3]2+, despite containing a fragment of the famous chromophore4−8 as part of its ligand scaffold. Instead of functioning as a light absorber itself, as observed in other Cp*Ir-supported hydride complexes,9−12 a separate [Ru(bpy)3]2+ chromophore initiates catalysis by electron transfer to the bimetallic species, which leads to 1 after protonation. The high catalytic activity of 1 in © XXXX American Chemical Society

[M−H]+ ⇌ [M] + H+

[M] ⇌ [M]2 + + 2e−

H+ + 2e− ⇌ H−

1.364(pK a)

(1)

− [− 46.12(E°)]

(2)

−46.12(E°H+ /H−)

[M−H]+ ⇌ [M]2 + + H−

ΔG°H−

(3) (4)

Received: September 14, 2016

A

DOI: 10.1021/acs.inorgchem.6b02223 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Despite sustained interest in developing aqueous catalytic reactions involving metal hydrides, hydricity measurements in water have only very recently seen significant development. As a result, hydricity comparisons between different solvents are also rare. A breakthrough in this area came when Creutz and co-workers determined the hydricity of two ruthenium complexes in water,23,24 as well as in acetonitrile (MeCN).25 More recently, the Berben group determined the hydricity of an iron electrocatalyst in both water and MeCN,19 and the groups of Yang18 and Appel26 established the hydricity of nickel hydrides in both solvents. We recently determined the hydricity of [Cp*Ir(bpy)(H)]+ in both MeCN10 and water.20 Solvent effects on hydricity can be substantial. In general, hydride complexes are more hydridic (lower hydricity value) in water.13 The role of ion solvation in this trend has been recognized as important in this regard.25 Ligands present in water can also influence the hydricity; water, hydroxide, or anions in the aqueous medium can bind the metal center as ligands following hydride release.20 The bimetallic system under investigation here can be conveniently studied in both MeCN and water, owing to the favorable solubility profile imparted by the doubly charged ruthenium polypyridyl unit. We have fully characterized several bimetallic intermediates proposed in the catalytic cycles for aqueous catalysis and established the thermodynamic hydricity of 1 in water and MeCN.

Figure 1. Spectrophotometric titration of 4 (red trace) to form 2-OH2 (purple trace) in 0.1 M sodium phosphate. The inset shows an absorbance at 560 nm (black circles) and a least-squares fitting to the Henderson−Hasselbalch equation (red line) to give pKa = 7.0 ± 0.1.

titration confirms that the IrIII d6 center is ligated by water and that the aqua complex will be dominant below pH 7. Electrochemical data for the ruthenium/iridium complexes have not been reported previously. The reduction potential of 2-OH2 was determined using cyclic voltammetry (CV), as shown in Figure 2 in a pH 5 phosphate buffer. The reduction

■

RESULTS AND DISCUSSION Characterization of [Cp*Ir(OH2)(μ-bpm)Ru(bpy)2]4+ (2OH2) in Water. The catalyst in prior studies was the sulfate salt of 2-OH2.1 Initial efforts to isolate [Cp*Ir(OH2)(μ-bpm)Ru(bpy)2][SO4]2 were complicated by sulfate coordination to the iridium center. In order to avoid this speciation and to maintain the solubility in both MeCN and water, the corresponding trifluoromethanesulfonate (triflate, SO3CF3−) salt was targeted. Stirring [Cp*Ir(OH2)3][SO3CF3]2 and [Ru(bpy)2(bpm)][SO3CF3]2 in water adjusted to pH 2 with dilute triflic acid under N2 afforded the dark forest green aqua complex 2-OH2 in 94% yield. Adding Na2SO4 to the triflate salt of 2-OH2 in D2O resulted in equilibrium mixtures of 2-OH2 and [Cp*Ir(SO4)(μ-bpm)Ru(bpy)2][SO4] 2-SO4 (Figure S2). The Cp* resonance (1H NMR δ 1.71 in D2O) of 2-SO4 is consistent with that in Fukuzumi’s prior report.1 The acidity of the aqua ligand of 2-OH2 was measured to establish the protonation state of the complex under the experimental conditions. The pKa of 2-OH2 was determined by a spectrophotometric titration in which acid was added to the hydroxo complex [Cp*Ir(OH)(μ-bpm)Ru(bpy)2]3+ (4) to form the aqua complex 2-OH2. The change in the absorbance at three different wavelengths (400, 560, and 600 nm) as a function of the pH was fit to the Henderson−Hasselbalch equation to provide a pKa of 7.0 ± 0.1 (Figure 1, inset). Structurally related aqua complexes [Cp*Ir(bpy-X)(OH2)]2+ (bpy-X = 4,4′-X-2,2′-bipyridine, X = H, OMe, COO−), [(η6C6Me6)Ru(bpy-X)(OH2)]2+ (X = H, OMe), and [(η6cymene)Ru(bpy-COO)(OH2)]2+ have acidities similar to that of 2-OH2, with pKa values in the range of 7−10.20,27,28 The concentrations of iridium and ruthenium were determined by inductively coupled plasma mass spectrometry, providing a molar extinction coefficient of 2-OH2 at λmax (410 nm) of 1800 M−1 cm−1; the extinction coefficient of 4 was also determined at λmax (415 nm) to be 1200 M−1 cm−1. The

Figure 2. Cyclic voltammogram of 2-OH2 in 0.1 M sodium phosphate adjusted to pH 5. A glassy carbon working electrode, a Ag/AgCl reference electrode, and a platinum wire counter electrode were used. The scan rate was 100 mV s−1.

feature, E1/2 = −0.19 V vs NHE, was assigned as a quasireversible 2e− reduction on the basis of the narrow peak-topeak separation (ΔEp = 49 mV). A Randles−Sevcik plot showed a linear dependence of the peak current on the square root of the scan rate, indicating a diffusion-limited reduction (Figure S5). To confirm that the reduction is a 2e− process, controlled potential electrolysis of 2-OH2 was conducted in a pH 7 phosphate buffer at −0.54 V vs NHE. During electrolysis, 1.6e− of current was passed as the dark green solution turned a darkpurple-red color [for details, see the Supporting Information (SI), section I.c.]. 1H NMR spectra of the electrolyzed solution B

DOI: 10.1021/acs.inorgchem.6b02223 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry contained diagnostic resonances for [Cp*Ir(μ-bpm)Ru(bpy)2]2+ (3), and UV−vis spectra also showed characteristic structured bands between 600 and 900 nm consistent with 3. The pKa of 2-OH2 indicates that an aqua ligand will be present under the experimental conditions, such that the reduction potential corresponds to the 2e− reduction of aqua complex 2OH2 to the reduced species 3 (which is not ligated by water). The standard reduction potential of 2-OH2 is the same as that of [Cp*Ir(OH)(bpy-COO)]− 20 and occurs at potentials less negative than expected for a ruthenium polypyridine-based reduction. Additional features at −0.6 and −1 V vs NHE are tentatively assigned to ruthenium polypyridine-based reductions (Figure S6). Characterization of Reduced Species 3. We also set out to independently synthesize the reduced complex 3, which had only been characterized by 1H and 13C NMR spectroscopy and was implicated as an intermediate in the catalytic 4e− reduction of oxygen.2 A biphasic reaction mixture containing the dark green aqua complex 2-OH2 and formate in a pH 5 aqueous layer above a colorless CH2Cl2 organic layer was stirred for 4 h. Over the course of the reaction, the CH2Cl2 layer took on a dark purple red color (Scheme 2). Isolation of the organic layer

Figure 3. Structural representation of 3 with ellipsoids drawn at the 50% probability level. The asymmetric unit contains another molecule of 3 and four triflate ions that are not shown, along with diethyl ether and MeCN solvents. Hydrogen atoms are omitted for clarity. Selected distances (Å): Ir1−N1 1.998(6), Ir1−N2 1.989(6), Ru1−N3 2.089(6), Ru1−N4 2.081(6), Ru1−N5 2.061(6), Ru1−N6 2.052(6), Ru1−N7 2.050(6), Ru1−N8 2.056(6).

Scheme 2. Synthesis of 3 from 2-OH2

provided pure 3 in 75% yield. The 1H NMR spectrum of 3 shows the anticipated increase in symmetry resulting from a geometric rearrangement around iridium that places the bpm ring perpendicular to the Cp* ring. Along with the change in symmetry, a dramatic upfield shift of one of the bpm resonances to δ 6.0 is diagnostic of the formation of 3. Purple crystalline needles suitable for X-ray diffraction were grown by vapor diffusion of diethyl ether into an MeCN solution of 3. The molecular structure is pseudo-C2-symmetric, with a C2 axis bisecting the C−C bond of bpm consistent with other low-valent Cp*M(bpy) complexes (Figure 3).29−31 The absorption spectrum of 3 features a structured band in the long-wavelength region of the UV−vis spectrum (Figure 4A) attributed to partial diimine ligand reduction.32−34 These broad structured absorbance features are common among reduced polypyridylruthenium and -iridium complexes.10,20,32,35−37 To probe the electronic structure of this reduced catalytic intermediate, density functional theory (DFT) and time-dependent density functional theory (TD-DFT) calculations were performed on 3. The singlet ground-state (S0) structure of 3 was optimized using the PBE, M06, and B3LYP functionals (LANL2DZ ECP basis set for the iridium and ruthenium atoms and 6-31G* for all other atoms), modeling water solvation with a polarized continuum model (PCM). A high degree of delocalization is observed in the highest occupied molecular orbital (HOMO),

Figure 4. (A) Experimental UV−vis spectrum (black line) and calculated (PBE functional) excitations (blue lines) from 3 with the described orbital involvement. (B) Orbitals involved in the lowestenergy transition at 735 nm.

lowest unoccupied molecular orbital (LUMO), and LUMO+2 (Figure 4B). The extensive orbital mixing of the HOMO is consistent with the reduction being delocalized across both the metal and ligand, somewhere intermediate between the formal extremes of IrI(bpm) and IrII(bpm)−1.32 From the optimized ground-state geometry, the absorption properties and excited states in water were explored by TDDFT. The first 50 excitations of 3 were calculated using the same three hybrid functionals in water solvent (for details, see C

DOI: 10.1021/acs.inorgchem.6b02223 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

above. The pKa of 1 in water was determined through spectrophotometric titration. Initial experiments titrating conjugate base 3 with dilute triflic or phosphoric acid led to the disappearance of the features assigned to 3, but the addition of a base to the solution did not restore the original spectrum (Figure S12). We hypothesize that the addition of triflic acid to solutions of 3 generates a high enough local pH that 1 is initially formed but undergoes further protonation to irreversibly generate H2 and the aqua complex 2-OH2. This reactivity is consistent with reports that 1 is an intermediate in catalytic H2 evolution activity.1,3 To avoid overprotonation, titration was performed using NaOH. Dissolving 3 in acidic solutions (pH 2−3) formed mixtures of hydride and conjugate base in situ (confirmed by 1 H NMR spectroscopy). A representative titration adding hydroxide to a pH 2.6 starting solution is shown in Figure 6.

the SI, section III.a), following similar methods used to explore [Cp*M(diimine)] complexes (M = Co, Rh, Ir).38,39 Good agreement was observed between the electronic transitions predicted by all three functionals. However, the best agreement with the experimental spectrum was observed using the PBE functional (Figure 4A). The molecular orbitals involved in all excitations calculated are outlined in the SI, section III.a. The transition at 735 nm is assigned to an excitation from the HOMO to a mixed state featuring LUMO and LUMO+2 character. Monometallic Cp*Ir(bpy)-based complexes feature bpy-centered π−π* transitions.38 For 3, contributions from both the bpm and bpy ligands are observed. All functionals show mixed contributions from both HOMO to LUMO and HOMO to LUMO+2 (LUMO+1 is also ruthenium and bpy π* based). Moving to higher energy, the transitions available to 3 at λ < 500 nm with the highest oscillator strengths originate from either the HOMO or the Ru d orbitals and access the π* system of the diimine rings. The calculations indicate a high degree of orbital mixing between the iridium and ruthenium centers, supporting potential cooperative involvement between both metal centers upon photoexcitation. Hydricity Determination of 1 in Water. The conjugate acid of the reduced complex 3, iridium hydride 1, was targeted via several synthetic procedures. The treatment of 2-OH2 with formate in acidic aqueous solutions1 led to decomposition. Reduction with NaBH4 in acidic aqueous solutions or methanol (MeOH) yielded similar results. Access to the desired hydride was finally obtained by dissolution of isolated samples of the reduced complex 3 in acidic, buffered solutions (pH 2−3). Protonation yielded 1, as evidenced by a diagnostic hydride resonance (δ −11.4 in D2O). In MeCN, a similar strategy was employed: protonation of 3 with protonated N,N-dimethylformamide, [HDMF][SO3CF3] (pKa = 6.1),40 yielded hydride 1 (δ −11.8 in CD3CN) along with the conjugate base DMF. To obtain the aqueous hydricity of 1, the potential−pKa thermochemical cycle of Figure 5 was employed. The necessary reduction potential (E° = −0.19 V vs NHE) was discussed

Figure 6. Spectrophotometric titration of 1 with NaOH to form 3. The inset shows an absorbance at 720 nm as a function of the pH, fit to the Henderson−Hasselbalch equation (red line).

Fitting of the absorbance at three wavelengths (511, 720, and 800 nm) from titrations beginning at pH 2.5 and 3 gave an average pKa of 3.1 ± 0.2 (Figure 6, inset). This value is slightly lower than the pKa of 3.9 reported by Fukuzumi et al. for the sulfate salt.1 The hydride complex 1 is surprisingly acidic for a transitionmetal complex.41,42 Whereas the aqua complex 2-OH2 has a pKa value similar to the monometallic species [Cp*Ir(bpyCOO)(OH2)], hydride 1 is 9 orders of magnitude more acidic than the corresponding monometallic hydride [Cp*Ir(bpyCOO)(H)]− (Table 1). Complex 1 is among the most acidic metal hydrides reported in water, with a pKa comparable to those of carbonylmolybdenum, -rhenium, and -tungsten hydrides.13,43,44 The dramatic increase in the acidity of the hydride, but not the aqua, relative to monometallic species is attributed to the formation of a highly stabilized conjugate base. As described above, DFT calculations suggest that the electron density in the reduced complex 3 can be stabilized by delocalization through the bpm and bpy π systems. The thermodynamic hydricity of 1 in water could be calculated based on the reduction potential and acidity data described above. According to eqs 1−4, which employ a newly

Figure 5. Thermochemical cycle used to determine the hydricity of 1 (L = water or MeCN). D

DOI: 10.1021/acs.inorgchem.6b02223 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Table 1. Comparison of Thermochemical Parameters of [Cp*Ir(bpy-COO)]2− (3) and [Cp*Ir(bpy-COO)]2− (5) E° (V vs NHE) [MIII(L)] + 2e− ⇆ [MI] + L pKa of aqua [MIII(OH2)] ⇆ [MIII(OH)] + H+ pKa of hydride [MIII(H)] ⇆ [MI] + H+ ΔG°H− (kcal mol−1) [MIII(H)] ⇆ [MIII(OH2] + H− a

3

5a

−0.19 (L = H2)

−0.19 (L = OH−)

7.0

7.6

3.1

12.4

29.7

32.0

1 + H 2O ⇌ 2‐OH 2 + H−

(5)

H 2 ⇌ H + + H−

(6)

1 + H+ + H 2O ⇌ 2‐OH 2 + H 2

(7)

On the basis of the hydricity of 1 and the accepted estimate for the heterolytic cleavage of H2 in water,13,24 H2 evolution is favorable by 4.5 kcal mol−1 at standard states (pH 0). These thermochemical predictions assume that the iridium center is ligated only by the solvent water after H2 release. If salts or buffer components bind iridium after hydride transfer, their metal binding strengths can contribute to the overall free energy.20 An NMR titration assessed the influence of the sulfate salts used in catalysis. Taking into account the free energy of sulfate substitution, ΔG°OH2→SO4 = −0.4 kcal mol−1, provides an effective hydricity of 1 to generate the sulfate complex 2-SO4, ΔG°H−(SO4) = 29.3 kcal mol−1. As is optimal for catalysis, no intermediates are prohibitively high in energy or so stable as to prevent further reactivity. The thermochemical values given in Scheme 3 are at the chosen standard state of pH 0, but the catalytic performance of 1 is fastest over a relatively narrow range of pH 3.5−4.5.1,3 With the well-defined equilibrium reactions of Scheme 3, the concentrations of key species can be tracked as a function of the pH. Below pH 3, the rate-determining step is hydride transfer from formate to form 1. As the solution pH drops below the pKa of formic acid, the amount of formate available for metal binding and hydride transfer diminishes. For example, at pH 2, formic acid will undergo only 2% proton dissociation, leaving little formate to drive the hydride-transfer reaction toward products. The large excess of formic acid relative to the catalyst will help mitigate this scenario. At higher pH (pH >5), the large concentration of formate leads to the rate-determining step becoming the protonation of 1 to release H2.1 In the H2 evolution step, the concentrations of 1 and 2 will be equal at pH 3.3 (for derivation, see the SI, section I.f.). As the pH increases, however, the equilibrium concentrations will redistribute to increase the concentration of 2 relative to 1, hindering H2 release. Furthermore, hydride 1 (pKa = 3.1) will be deprotonated to form 3 as the pH increases, further hindering H2 release. At pH 3, the system appropriately balances the acidity and hydricity parameters to optimize H2 evolution. Photochemical H2 production in water exhibits a similar behavior, with the highest quantum yield obtained at pH 3.6.3 A balance is again required in the system between the range of hydride stability and the pKa of ascorbic acid (pKa = 4.0) that reductively quenches the excited state of [Ru(bpy)3]2+. Reductive quenching will be best at pH >5, where the ascorbate concentration is maximal, but the pKa of the hydride (pKa = 3.1) requires more acidic conditions. The reduced chromophore [Ru(bpy)3]+ is a strong reductant, easily capable of reducing 2-OH2 to 3 (thermodynamically favorable by 25 kcal mol−1).45 Near pH 3, the bimetallic species will be protonated to form the hydride 1, which releases H2 in the same manner as that described above. Hydricity Determination in MeCN. Hydricity is most commonly determined in MeCN solution, and only a few complexes have well-defined hydricity values in both water and MeCN.10,13,18−20,23,46 To better compare our thermochemical data with other metal hydrides and to more fully develop the connection between hydricity scales in H2O and MeCN, the

Reference 20.

standardized value for the reduction of H+ to H−,13,24 the aqueous hydricity of 1 was determined, ΔG°H−(H2O) = 29.7 kcal mol−1. This hydricity represents the release of a hydride ion, and in water the solvation process includes the addition of a water ligand after hydride transfer (denoted by the parenthetical H2O in the free-energy expression). Hydride 1 is a stronger hydride donor [smaller ΔG°H−(H2O) value] than the monometallic hydride [Cp*Ir(bpy)(H)]+ [ΔG°H−(H2O) = 31.5 kcal mol−1],20 despite being tricationic. Relating Thermochemistry to Catalysis. The proposed intermediates in H2 evolution catalysis are depicted in Scheme 3 along with the newly collected thermochemical data. Each Scheme 3. Relevant Thermochemical Values of Proposed Catalytic Intermediates (Free Energies are Shown in Blue and Given in kcal mol−1; Hydricities are Shown in Green and Given in kcal mol−1; Reduction Potentials in V vs NHE and Acidities are Shown in Red)

step of a formic acid dehydrogenation cycle can be described thermochemically. After proton dissociation described by the pKa of formic acid, hydride transfer occurs from formate to iridium. Formate (ΔG°H− = 24.1 kcal mol−1 in water)24 is a better hydride donor than 1 (ΔG°H−(H2O) = 29.7 kcal mol−1 in water), so hydride transfer from formate to 2-OH2 is thermodynamically favorable by 5.6 kcal mol−1. Hydride 1 then reacts with a proton to release H2 and regenerate 2-OH2. The driving force for H2 evolution from 1 can be estimated using a thermochemical cycle based on eqs 5−7.13 E

DOI: 10.1021/acs.inorgchem.6b02223 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

0.081 Å is seen for the current system. The other changes in the bond lengths are also consistent both with the rhodium system and with free bpy.47−49 CV was used to obtain the reduction potential in MeCN. The reduced species 3 exhibits a 2e− oxidation at −0.52 V vs Cp2Fe+/0 in MeCN containing 0.1 M [Bu4N][PF6], with a ΔEp of 76 mV (Figure S11). The pKa of the hydride was determined in a 1H NMR experiment. The addition of varying amounts of [HDMF][SO3CF3] (pKa = 6.1 in MeCN)40 to CD3CN solutions of 3 produces equilibrium mixtures of 3 and hydride 1. The relative concentrations in four separate experiments with varying amounts of acid provide an average pKa of 5.5 ± 0.5 (Figure S13). Hydride 1 is almost 20 orders of magnitude more acidic than [Cp*Ir(bpy)(H)]+ in MeCN (pKa = 23.3).10 Furthermore, the acidity of hydride 1 hardly changes upon moving from water to MeCN. A difference of only 2 pKa units is quite uncommon for transition-metal hydrides, which typically shift by 6−8 pKa units upon moving between water and MeCN.43,44 Even with organic compounds, pKa values typically increase by at least 5 pKa units in MeCN versus water.50−52 Various predictive models were examined to help explain the unexpectedly small solvent dependence of the pKa. The pKa in MeCN is related to the pKa in water by the solvent-transfer free energies of each species, as depicted in Figure 9 and written in eq 8.

hydricity of 1 was determined in MeCN. The same potential− pKa thermodynamic cycle was utilized, which requires the reduction potential of [Cp*Ir(NCCH3)(μ-bpm)Ru(bpy)2]4+ (2-MeCN) and the pKa of 1. In MeCN, a different constant for the reduction of H+ is also used.21 The aqua ligand of 2-OH2 is readily displaced in MeCN solutions to form 2-MeCN. Large green block-shaped crystals of 2-MeCN were grown by the slow evaporation of an MeCN/ tetrahydrofuran solution (Figure 7). Structural comparisons

Figure 7. Structural representation of 2-MeCN with ellipsoids drawn at the 50% probability level. The asymmetric unit contains four triflate ions that are not shown, along with MeCN and tetrahydrofuran solvents. Hydrogen atoms are omitted for clarity. Selected distances (Å): Ir1−N9 2.063(4), Ir1−N2 2.135(4), Ir1−N9 2.063(4), Ru1−N3 2.059(4), Ru1−N4 2.088(4), Ru1−N5 2.072(4), Ru1−N6 2.063(4), Ru1−N7 2.061(5), Ru1−N8 2.064(4).

pK a(MeCN) = pK a(H 2O) + +

ΔG°tr(H2O → MeCN)(3) 1.364

−

ΔG°tr(H2O → MeCN)(H+) 1.364 ΔG°tr(H2O → MeCN)(1) 1.364

(8)

between oxidized and reduced complexes containing bipyridyltype ligands have been used to gain insight into differences in the electronic structures.47 Figure 8 compares the bond lengths in bpm in 2-MeCN and 3.

Figure 9. Free energy of transfer of a species from the gas phase into different solvent environments.

If one assumes that the solvent-transfer free energies of 1 and 3 are the same, a simplified semiempirical prediction can be made according to eq 9, which suggests that the pKa in MeCN will be 7.5 pKa units higher than the value in water (for further details, see the SI, section II.a).43 For 1, eq 9 overestimates the pKa in MeCN substantially (predicted to be 10.6 but found to be 5.5).

Figure 8. Bond length comparison between 2-MeCN (blue) and 3 (red) with differences in length upon reduction (black). Bonds that contract upon reduction are shown in red, and bonds that lengthen are shown in green.

The bonds that contract are consistent with what is observed for free bpy47 as well as for the related system Cp*Rh(bpy).48,49 For free bpy, the bridging C−C bond between the two pyridyl rings was shown to be the best indicator of ligand reduction, with a contraction from 1.490 to 1.431 Å upon 1e− reduction of the ligand. An even more dramatic contraction of

pK a(MeCN) = pK a(H 2O) + 7.5

(9)

Izutsu developed a method for estimating ΔG°tr based on a semiempirical calculation of ΔG°sv (eq 10).53 In eq 10, NA is Avogadro’s number, z is the charge of the species being F

DOI: 10.1021/acs.inorgchem.6b02223 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry solvated, e is the charge of an electron, r is the ionic radius of the species being solvated, and δs = rs/λs, where rs is the ionic radius of the spherical solvent and λs is the Wertheim polarization factor of the solvent. ΔG°sv =

−NAz 2e 2 ⎛ 1⎞ ⎜1 − ⎟ 4πε0 2(r + δs) ⎝ εr ⎠

Figure 10 shows how experimental hydricity changes upon moving from MeCN and water for 1 and several other

(10)

ΔG°tr(H2O → MeCN) = ΔG°sv (MeCN) − ΔG°sv (H 2O) (11)

The ionic radii of 1 and 3 were estimated from the average of several distances in the crystal structures of 1 and 2-MeCN, allowing ΔG°sv to be calculated for each species. The energy of transfer between solvents (ΔG° tr(H 2O→MeCN)) was then determined as the difference between ΔG°sv for each solvent (eq 11).53 The pKa of 1 in MeCN could then be predicted based on its pKa in water using eq 8.54 This method estimates the pKa in MeCN to be 8.7, still a substantial overestimate. We finally turned to DFT to better predict the effects of solvation on 1 and 3. Geometry optimization was performed in a vacuum, with implicit water solvation, and with implicit MeCN solvation. The free energy of solvation (ΔG°sv) and the free energy of transfer from water to MeCN (ΔG°tr(H2O→MeCN)) were determined from the data (for further details, see the SI, section II.c.). Using computed values of ΔG°tr in eq 8 provided pKa = 5.1 in MeCN, in almost perfect agreement with the experimental result. For this system, a continuum solvation method was a better model for experiment than semiempirical predictions. Combining the reduction potential and acidity data and adjusting for the 2e− reduction of a proton (79.6 kcal mol−1)21 gives ΔGoH−(MeCN) = 63.1 kcal mol−1. This value is 1 kcal mol−1 greater than the reported hydricity for [Cp*Ir(bpy)(H)]+ of 62 kcal mol−1.10 The transition metal hydricity typically correlates strongly with the reduction potential of the conjugate hydride acceptor (MeCN complex), with more negative potentials leading to strong hydride donors.55 The reduction potential of 2-MeCN is over 500 mV less negative than that of [Cp*Ir(bpy)(NCCH3)]2+, the bimetallic hydride 1 would be expected to be a very weak hydride donor, in the range of 68 kcal mol−1. However, the pKa of 1 is much more acidic than typical iridium hydrides, and this has an opposing effect on the hydricity (eqs 1−4). Thus, although the relevant reduction potential and pKa values are not similar, the bimetallic and monometallic hydrides end up with very similar hydricity values. Table 2 highlights the thermochemical parameters measured in both solvents. It is noteworthy that hydride transfer from formate is thermodynamically much more favorable in MeCN (∼20 kcal mol−1) than in water (∼5 kcal mol−1), which could have implications for catalysis involving 1 in MeCN.

Figure 10. Comparison of complexes with reported hydricities in MeCN and water (tpy = 2,2′:6′,2″-terpyridine).10,18,20,23,46

complexes. All of the hydrides are more hydridic (better hydride donors and lower ΔG°H−) in water than in MeCN. This is partially due to the different constants for the H+/H− reduction in each solvent, which is reflected in the hydricity of H2 in Figure 10. However, if this was the only factor contributing to the differences in the hydricities, there would be a systematic difference of 41.8 kcal mol−1 for each species. The charged species released upon hydride donation are better stabilized in water (ε = 78.4)53 than in MeCN (ε = 35.9),53 and additional factors such as hydrogen-bonding interactions could also be influencing this difference.46 The largest values of ΔΔG°H− are in the range 33−35 kcal mol−1 for 1, [HFe4N(CO)12]−,19 and [HNi(dmpe)2]+.26 Two of these complexes are large multimetallic species capable of efficient delocalization, but more detailed studies would be required to elucidate the factors that influence the magnitude of ΔΔG°H− between the two solvents.

■

CONCLUSIONS Several heterobimetallic catalytic intermediates have been isolated and characterized in solution using electrochemical methods, UV−vis spectroscopy, and NMR spectroscopy. Complexes 2-MeCN and 3 have also been structurally characterized by X-ray crystallography. The electronic structure of the reduced species 3 was investigated by DFT and TDDFT, revealing a significant degree of metal−ligand delocalization. The stability afforded by delocalization is responsible for the surprisingly strong acidity of 1. Solvation energies approximated using continuum solvation models in DFT calculations accurately predicted the subtle change in pKa upon moving from water to MeCN. This work includes the first reported structural, electrochemical, and thermodynamic characterization of heterobimetallic Cp*Ir/Ru complexes. The doubly charged ruthenium center provides good solubility in both polar organic solvents and water, enabling determination of the hydricity using a potential−pKa thermodynamic cycle in both water and MeCN. While the acidity of 1 remains relatively constant between solvents, the hydricity values are quite different in water and MeCN. The catalytic activity of 1 toward photochemical H2

Table 2. Comparison of the Thermochemical Values in Water and Acetonitrile E° of 2 (V) pKa of 1 ΔG°H− of 1 (kcal mol−1) a

water

MeCN

−0.19a 3.1 29.7

−0.52b 5.5 63.1

vs NHE. bvs Cp2Fe+/0. G

DOI: 10.1021/acs.inorgchem.6b02223 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

2H), 7.78 (d, J = 5.7 Hz, 2H), 7.58 (t, J = 5.3 Hz, 2H), 7.41 (dt, J = 18.4 and 6.6 Hz, 4H). Synthesis of [Cp*Ir(OH2)(μ-bpm)Ru(bpy)2][SO3CF3]4 (2-OH2). On the basis of a previous preparation in neutral water,1 [Cp*Ir(OH2)3][SO3CF3]2 (150 mg, 0.221 mmol) was dissolved in pH 2 water adjusted with triflic acid and added to a Schlenk flask. [Ru(bpy)2(bpm)][SO3CF3]2 (192 mg, 0.221 mmol) was then added to the flask with stirring. The mixture was stirred in pH 2 water (35 mL) for 2 h at room temperature. A dark green crystalline solid was isolated in vacuo in 94% yield. Reactions run at pH >359 resulted in the formation of [(Cp*Ir)2(μ-OH)3][SO3CF3] and incomplete conversion to the product. 1H NMR (600 MHz, D2O): δ 9.49 (dd, J = 19.2 and 5.6 Hz, 2H), 8.60 (dt, J = 24.1 and 8.1 Hz, 6H), 8.27 (d, J = 5.6 Hz, 1H), 8.21 (t, J = 7.9 Hz, 1H), 8.19−8.08 (m, 3H), 8.03 (dt, J = 11.0 and 5.7 Hz, 2H), 7.75 (dd, J = 32.0 and 5.7 Hz, 2H), 7.62−7.53 (m, 2H), 7.51−7.39 (m, 3H) 1.79 (s, 15H). 13C{1H} NMR (151 MHz, D2O): δ 167.15, 163.48, 163.25, 157.64, 157.46, 157.42, 157.29, 153.70, 152.59, 152.19, 151.80, 140.09, 140.02, 139.85, 139.83, 128.51, 128.45, 128.42, 128.33, 128.26, 128.16, 125.42, 125.32, 125.27, 92.00, 8.78. HRMS (ESI+). Calcd for [Cp*Ir(OH)(μ-bpm)Ru(bpy)2]3+·2Cl− + MeOH: m/z 1019.15. Found: m/z 1018.95. Synthesis of [Cp*Ir(μ-bpm)Ru(bpy)2][SO3CF3]2 (3). The aqua complex 2-OH2 (50 mg, 0.033 mmol) was dissolved in 2 mL of water, and a 5 M pH 5 formate solution (52.8 μL, 0.264 mmol) was added with stirring. Dichloromethane (4 mL) was added to the reaction mixture. The biphasic reaction mixture was stirred at room temperature for 4 h. After color transfer from the water layer to the dichloromethane layer was observed, the bright-violet organic layer was separated from the water layer. The pure product was isolated in vacuo in 75% yield. 1H NMR (600 MHz, D2O): δ 8.79 (d, J = 6.7 Hz, 2H), 8.48 (t, J = 8.5 Hz, 4H), 8.09−7.94 (m, 6H), 7.86 (d, J = 5.6 Hz, 2H), 7.32 (q, J = 7.3 Hz, 4H), 7.07 (d, J = 4.6 Hz, 2H), 6.09 (dd, J = 6.8 and 4.5 Hz, 2H), 2.02 (s, 15H). 13C{1H} NMR (101 MHz, D2O): δ 158.07, 157.62, 155.05, 152.61, 152.47, 148.60, 143.00, 137.96, 127.54, 127.22, 124.21, 108.49, 88.27, 9.33. Synthesis of [Cp*Ir(H)(μ-bpm)Ru(bpy)2][SO3CF3]3 (1). The hydride was prepared in situ using 2.8 equiv of [H-DMF][SO3CF3] to protonate 3 in MeCN and generate [Cp*Ir(H)(μ-bpm)Ru(bpy)2][SO3CF3]3 (1). 1H NMR (400 MHz, CD3CN): δ 9.13 (t, J = 5.6 Hz, 2H), 8.57 (d, J = 8.4 Hz, 2H), 8.53 (t, J = 7.7 Hz, 2H), 8.20 (t, J = 8.1 Hz, 1H), 8.17−8.09 (m, 7H), 7.76 (d, J = 5.7 Hz, 1H), 7.68 (t, J = 5.6 Hz, 2H), 7.66−7.61 (m, 1H), 7.56 (t, J = 6.6 Hz, 1H), 7.47 (d, J = 6.6 Hz, 3H), 1.95 (s, 15H), −11.81 (s, 1H). Computational Methods. Geometry optimizations and frequency calculations were done using the hybrid functionals M06, B3LYP, and PBE as implemented in Gaussian 0962 with the LANL2DZ ECP basis set63,64 for the iridium and ruthenium atoms and 6-31G* for all other atoms. The PCM or SMD implicit solvation models (water solvent) were employed for all calculations.

evolution and formate dehydrogenation was considered in the context of the newly available thermochemical values. The large difference in the hydricity of 1 between water and MeCN has the potential for use as a catalytic tool to modulate hydride reactivity in different solvent environments.

■

EXPERIMENTAL SECTION

General Considerations. All synthetic procedures were carried out using a nitrogen-filled glovebox or a nitrogen Schlenk line unless otherwise noted. Water was thoroughly degassed with nitrogen prior to use. Organic solvents were dried and degassed with argon using a Pure Process Technology solvent system. D2O and CD3CN were purchased from Cambridge Isotope Laboratories Inc. and used as received for air-stable samples. Dissolved oxygen was removed by sparging with nitrogen (D2O) or through three freeze−pump−thaw cycles (CD3CN) before being stored in a glovebox. [Cp*Ir(Cl)2]2,56 [Ru(bpy)2(Cl)2],57 [Ru(bpy)2(OH2)2][SO3CF3]2,58 and [Cp*Ir(OH2)3][SO3CF3]259 were synthesized according to literature procedures. 1 H NMR spectra were recorded on either a 400 or 600 MHz spectrometer at 25 °C. Chemical shifts are reported with respect to residual organic solvents60 or a dioxane internal standard in D2O. UV− vis spectra were obtained with either a Cary 60 spectrophotometer or an Ocean Optics USB2000+ spectrometer with a DT-MINI-2GS deuterium/tungsten halogen light source controlled by OceanView software. The solution pH was recorded with either an OrionStar A111 pH meter with a Beckman-Coulter pH probe or a Hach PHW47-SS ISFET probe and a Hach H170 portable meter. Electrochemical studies were performed using a Pine WaveNow potentiostat or a WaveDriver bipotentiostat controlled with Af termath software. For CV, an undivided three-electrode cell with a 3-mmdiameter glassy carbon electrode that was polished between scans with 0.05-μm alumina powder was used. Controlled potential electrolysis studies were performed in a pressure-bridged divided H-cell. A reticulated vitreous carbon electrode impaled on a graphite rod was utilized as the working electrode. For both experiments, a platinum wire counter electrode and a Ag/AgCl (3 M NaCl) reference electrode were used. All potentials are reported versus NHE by adding +0.21 V to the potentials measured versus Ag/AgCl.61 Inductively coupled plasma mass spectrometry using an Agilent Technologies 7500x series spectrometer at the UNC Biomarker Mass Spectrometry Facility was employed to determine the iridium and ruthenium concentrations for UV−vis. A calibration curve of iridium and ruthenium from 10 to 500 ppb was generated and used to determine the concentrations of the UV−vis samples. Mass spectrometry was carried out with a LTQ FT (ICR 7T; ThermoFisher, Bremen, Germany) mass spectrometer. Samples (in water/methanol solutions) were introduced via a microelectrospray source at a flow rate of 3 μL min−1. Xcalibur (ThermoFisher, Breman, Germany) was used to analyze the data. Molecular formula assignments were determined with Molecular Formula Calculator (version 1.2.3). Single-crystal X-ray diffraction data were collected on a Bruker APEX-II CCD diffractometer at 100 K with Cu Kα radiation (λ = 1.54175 Å). The structures were solved with the olex2.solve structure solution program using charge flipping and refined with the XL refinement program using least-squares minimization. For additional details, see the SI, section I.a. Synthetic Procedures. Synthesis of [Ru(bpy)2(bpm)][SO3CF3]2. bpm (43 mg, 0.272 mmol) was dissolved in water (5 mL) and added to a solution of [Ru(bpy)2(H2O)2][SO3CF3]2 (200 mg, 0.268 mmol) in water (5 mL) with stirring in a 20 mL scintillation vial. The mixture was stirred at room temperature for 5 h, and a brown-orange powder was isolated in vacuo. Excess bpm was removed by dissolving the crude product mixture in methanol and isolating the desired solid upon the addition of diethyl ether in 92% yield. 1H NMR (400 MHz, D2O): δ 9.07 (d, J = 5.0 Hz, 2H), 8.55 (dd, J = 8.3 and 3.7 Hz, 4H), 8.22 (d, J = 6.1 Hz, 2H), 8.08 (q, J = 8.5 Hz, 4H), 7.93 (d, J = 5.6 Hz,

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.6b02223. Additional experimental details, NMR spectra, cyclic voltammograms, UV−vis spectra, and detailed computational details including input files, optimized coordinates for all complexes, calculated electronic spectra of 3, and orbital compositions of excited states (PDF) Crystallographic data for 3 and 2-MeCN (CIF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. H

DOI: 10.1021/acs.inorgchem.6b02223 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

■

(15) Rakowski DuBois, M.; DuBois, D. L. The Roles of the First and Second Coordination Spheres in the Design of Molecular Catalysts for H2 Production and Oxidation. Chem. Soc. Rev. 2009, 38, 62−72. (16) Rakowski Dubois, M.; Dubois, D. L. Development of Molecular Electrocatalysts for CO2 Reduction and H2 Production/Oxidation. Acc. Chem. Res. 2009, 42, 1974−1982. (17) Helm, M. L.; Stewart, M. P.; Bullock, R. M.; Rakowski Dubois, M.; DuBois, D. L. A Synthetic Nickel Electrocatalyst with a Turnover Frequency above 100000 s−1 for H2 Production. Science 2011, 333, 863−866. (18) Tsay, C.; Livesay, B. N.; Ruelas, S.; Yang, J. Y. Solvation Effects on Transition Metal Hydricity. J. Am. Chem. Soc. 2015, 137, 14114− 14121. (19) Taheri, A.; Thompson, E. J.; Fettinger, J. C.; Berben, L. A. An Iron Electrocatalyst for Selective Reduction of CO2 to Formate in Water: Including Thermochemical Insights. ACS Catal. 2015, 5, 7140−7151. (20) Pitman, C. L.; Brereton, K. R.; Miller, A. J. M. Aqueous Hydricity of Late Metal Catalysts as a Continuum Tuned by Ligands and the Medium. J. Am. Chem. Soc. 2016, 138, 2252−2260. (21) Berning, D. E.; Noll, B. C.; DuBois, D. L. Relative Hydride, Proton, and Hydrogen Atom Transfer Abilities of [HM(diphosphine)2]PF6 Complexes (M = Pt, Ni). J. Am. Chem. Soc. 1999, 121, 11432−11447. (22) Curtis, C. J.; Miedaner, A.; Ellis, W. W.; DuBois, D. L. Measurement of the Hydride Donor Abilities of [HM(diphosphine)2]+ Complexes (M = Ni, Pt) by Heterolytic Activation of Hydrogen. J. Am. Chem. Soc. 2002, 124, 1918−1925. (23) Creutz, C.; Chou, M. H. Hydricities of d6 Metal Hydride Complexes in Water. J. Am. Chem. Soc. 2009, 131, 2794−2795. (24) Connelly, S. J.; Wiedner, E. S.; Appel, A. M. Predicting the Reactivity of Hydride Donors in Water: Thermodynamic Constants for Hydrogen. Dalton Trans. 2015, 44, 5933−5938. (25) Matsubara, Y.; Fujita, E.; Doherty, M. D.; Muckerman, J. T.; Creutz, C. Thermodynamic and Kinetic Hydricity of Ruthenium(II) Hydride Complexes. J. Am. Chem. Soc. 2012, 134, 15743−15757. (26) Connelly Robinson, S. J.; Zall, C. M.; Miller, D. L.; Linehan, J. C.; Appel, A. M. Solvent Influence on the Thermodynamics for Hydride Transfer from Bis(diphosphine) Complexes of Nickel. Dalton Trans. 2016, 45, 10017−10023. (27) Ogo, S.; Kabe, R.; Hayashi, H.; Harada, R.; Fukuzumi, S. Mechanistic Investigation of CO2 Hydrogenation by Ru(II) and Ir(III) Aqua Complexes under Acidic Conditions: Two Catalytic Systems Differing in the Nature of the Rate Determining Step. Dalton Trans. 2006, 4657−4663. (28) Liu, Z.; Habtemariam, A.; Pizarro, A. M.; Fletcher, S. A.; Kisova, A.; Vrana, O.; Salassa, L.; Bruijnincx, P. C. A.; Clarkson, G. J.; Brabec, V.; et al. Organometallic Half-Sandwich Iridium Anticancer Complexes. J. Med. Chem. 2011, 54, 3011−3026. (29) Ziessel, R.; Noblat-Chardon, S.; Deronzier, A.; Matt, D.; Toupet, L.; Balgroune, F.; Grandjean, D. Comparative Structural Studies of [(η5-Me5C5)IrI(2,2′-bipyridine-4,4′-COOH)] and [(η5Me5C5)IrIII{2,2′-bipyridine-4,4′-COO(CH2)3-pyrrole}Cl]ClO4. Acta Crystallogr., Sect. B: Struct. Sci. 1993, 49, 515−519. (30) Mork, B. V.; McMillan, A.; Yuen, H.; Tilley, T. D. The [(η5C5Me5)(2,2′-bipyridine)Ru]− Anion and Its Use in the Synthesis of Alkyl, Gallyl, and Stannyl Complexes of Ruthenium. Organometallics 2004, 23, 2855−2859. (31) Blakemore, J. D.; Hernandez, E. S.; Sattler, W.; Hunter, B. M.; Henling, L. M.; Brunschwig, B. S.; Gray, H. B. Pentamethylcyclopentadienyl rhodium complexes. Polyhedron 2014, 84, 14−18. (32) Ladwig, M.; Kaim, W. Electronic Structure of Catalytic Intermediates for Production of H2: (C5Me5)Ir(bpy) and Its Conjugated Acid. J. Organomet. Chem. 1992, 439, 79−90. (33) Kaim, W.; Reinhardt, R.; Waldhör, E.; Fiedler, J. Electron Transfer and Chloride Ligand Dissociation in Complexes [(C5Me5)ClM(bpy)]+/[(C5Me5)M(bpy)]n (M = Co, Rh, Ir; N = 2+,+,0,-): A Combined Electrochemical and Spectroscopic Investigation. J. Organomet. Chem. 1996, 524, 195−202.

ACKNOWLEDGMENTS The authors gratefully acknowledge funding from the National Science Foundation Center for Enabling New Technologies through Catalysis (Grant CHE-1205189) and the Royster Society of Fellows (fellowship to C.L.P.). We thank Peter White for his assistance with X-ray crystallography, Brandie Ehrmann for her assistance with mass spectroscopy, and Peter Cable for his assistance with analytical facilities provided by the UNC Biomarker Mass Spectrometry Facility, supported, in part, by a grant from the National Institute of Environmental Health Sciences (Grant P30ES010126). A sample of [HDMF] [SO3CF3] was generously provided by Brian McCarthy and Jillian Dempsey.

■

REFERENCES

(1) Fukuzumi, S.; Kobayashi, T.; Suenobu, T. Unusually Large Tunneling Effect on Highly Efficient Generation of Hydrogen and Hydrogen Isotopes in pH-Selective Decomposition of Formic Acid Catalyzed by a Heterodinuclear Iridium-Ruthenium Complex in Water. J. Am. Chem. Soc. 2010, 132, 1496−1497. (2) Fukuzumi, S.; Kobayashi, T.; Suenobu, T. Formic Acid Acting as an Efficient Oxygen Scavenger in Four-Electron Reduction of Oxygen Catalyzed by a Heterodinuclear Iridium-Ruthenium Complex in Water. J. Am. Chem. Soc. 2010, 132, 11866−11867. (3) Fukuzumi, S.; Kobayashi, T.; Suenobu, T. Photocatalytic Production of Hydrogen by Disproportionation of One-ElectronReduced Rhodium and Iridium-Ruthenium Complexes in Water. Angew. Chem., Int. Ed. 2011, 50, 728−731. (4) Kalyanasundaram, K. Photophysics, Photochemistry and Solar Energy Conversion with tris(bipyridyl)ruthenium(II) and Its Analogues. Coord. Chem. Rev. 1982, 46, 159−244. (5) Kalyanasundaram, K.; Grätzel, M. Applications of Functionalized Transition Metal Complexes in Photonic and Optoelectronic Devices. Coord. Chem. Rev. 1998, 177, 347−414. (6) Meyer, T. J. Chemical Approaches to Artificial Photosynthesis. Acc. Chem. Res. 1989, 22, 163−170. (7) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; von Zelewsky, A. Ru(II) Polypyridine Complexes: Photophysics, Photochemistry, Eletrochemistry, and Chemiluminescence. Coord. Chem. Rev. 1988, 84, 85−277. (8) Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Visible Light Photoredox Catalysis with Transition Metal Complexes: Applications in Organic Synthesis. Chem. Rev. 2013, 113, 5322−5363. (9) Ziessel, R. Photocatalysis. Mechanistic Studies of Homogeneous Photochemical Water Gas Shift Reaction Catalyzed under Mild Conditions by Novel Cationic Iridium (III) Complexes. J. Am. Chem. Soc. 1993, 115, 118−127. (10) Barrett, S. M.; Pitman, C. L.; Walden, A. G.; Miller, A. J. M. Photoswitchable Hydride Transfer from Iridium to 1-Methylnicotinamide Rationalized by Thermochemical Cycles. J. Am. Chem. Soc. 2014, 136, 14718−14721. (11) Barrett, S. M.; Slattery, S. A.; Miller, A. J. M. Photochemical Formic Acid Dehydrogenation by Iridium Complexes: Understanding Mechanism and Overcoming Deactivation. ACS Catal. 2015, 5, 6320− 6327. (12) Chambers, M. B.; Kurtz, D. A.; Pitman, C. L.; Brenneman, M. K.; Miller, A. J. M. Efficient Photochemical Dihydrogen Generation Initiated by a Bimetallic Self-Quenching Mechanism. J. Am. Chem. Soc. 2016, 138, 13509−13512. (13) Wiedner, E. S.; Chambers, M. B.; Pitman, C. L.; Bullock, R. M.; Miller, A. J. M.; Appel, A. M. Thermodynamic Hydricity of Transition Metal Hydrides. Chem. Rev. 2016, 116, 8655−8692. (14) Fraze, K.; Wilson, A. D.; Appel, A. M.; Rakowski DuBois, M.; Dubois, D. L. Thermodynamic Properties of the Ni−H Bond in Complexes of the Type [HNi(P2RN2R‘)2](BF4) and Evaluation of Factors That Control Catalytic Activity for Hydrogen Oxidation/ Production. Organometallics 2007, 26, 3918−3924. I

DOI: 10.1021/acs.inorgchem.6b02223 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

(53) Izutsu, K. Electrochemistry in Nonaqueous Solutions; Wiley-VCH: Weinheim, Germany, 2002. (54) Muckerman, J. T.; Skone, J. H.; Ning, M.; Wasada-Tsutsui, Y. Toward the Accurate Calculation of pKa Values in Water and Acetonitrile. Biochim. Biophys. Acta, Bioenerg. 2013, 1827, 882−891. (55) Berning, D. E.; Miedaner, A.; Curtis, C. J.; Noll, B. C.; Rakowski DuBois, M. C.; DuBois, D. L. Free-Energy Relationships between the Proton and Hydride Donor Abilities of [HNi(diphosphine) 2]+ Complexes and the Half-Wave Potentials of Their Conjugate Bases. Organometallics 2001, 20, 1832−1839. (56) White, C.; Yates, A.; Maitlis, P. M.; Heinekey, D. M. (η5Pentamethylcyclopentadienyl)Rhodium and -Iridium Compounds. Inorg. Synth. 1992, 29, 228−234. (57) Sullivan, B. P.; Salmon, D. J.; Meyer, T. J. Mixed Phosphine 2, 2′-bipyridine Complexes of Ruthenium. Inorg. Chem. 1978, 17, 3334− 3341. (58) Taher, D.; Thibault, M. E.; Di Mondo, D.; Jennings, M.; Schlaf, M. Acid-, Water- and High-Temperature-Stable Ruthenium Complexes for the Total Catalytic Deoxygenation of Glycerol to Propane. Chem. - Eur. J. 2009, 15, 10132−10143. (59) Cayemittes, S.; Poth, T.; Fernandez, M. J.; Lye, P. G.; Becker, M.; Elias, H.; Merbach, E. Mechanistic Investigation on the Water Substitution in the η5-Organometallic Complexes Cp*Ir(H2O)32+ and Cp*Rh(H2O)32+. Inorg. Chem. 1999, 38, 4309−4316. (60) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.; Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I. NMR Chemical Shifts of Trace Impurities: Common Laboratory Solvents, Organics, and Gases in Deuterated Solvents Relevant to the Organometallic Chemist. Organometallics 2010, 29, 2176−2179. (61) Smith, T. J.; Stevenson, K. J. Reference Electrodes. In Handbook of Electrochemistry; Zoski, C. G., Ed.; Elsevier: Amsterdam, The Netherlands, 2006; p 75. (62) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; et al. Gaussian 09, revision D.01; Gaussian, Inc.: Wallingford, CT, 2009. (63) Schuchardt, K. L.; Didier, B. T.; Elsethagen, T.; Sun, L.; Gurumoorthi, V.; Chase, J.; Li, J.; Windus, T. L. Basis Set Exchange: A Community Database for Computational Science. J. Chem. Inf. Model. 2007, 47, 1045−1052. (64) Feller, D. J. The Role of Databases in Support of Computational Chemistry Calculations. J. Comput. Chem. 1996, 17, 1571−1586.

(34) Braterman, P. S.; Song, J. I.; Kohlmann, S.; Vogler, C.; Kaim, W. Spectroelectrochemistry of Aromatic Ligands and Their Derivatives: III. Binuclear Transition Metal Complexes of CuI, Mo0, and ReI with 2, 2′-Bipyrimidine. J. Organomet. Chem. 1991, 411, 207−213. (35) Krejčik, M.; Záliš, S.; Ladwig, M.; Matheis, W.; Kaim, W. A Study of the Reduced States of Four Isomeric Bidiazines (Diaza-2,2′bipyridines) by UV-Vis/NIR-Spectroelectrochemistry. J. Chem. Soc., Perkin Trans. 2 1992, 2007−2010. (36) Kaim, W.; Ernst, S.; Kasack, V. ESR of Homo-and Heteroleptic Mono-and Dinuclear Tris(alpha-Diimine)ruthenium Radical Complexes. J. Am. Chem. Soc. 1990, 112, 173−178. (37) Záliš, S.; Krejčik, M.; Drchal, V.; Vlcek, A. A. The Sequence of Ligand Reductions in Heteroleptic Ruthenium-Diimine Complexes: Calculation of Redox Potentials as a Diagnostic Tool. Inorg. Chem. 1995, 34, 6008−6014. (38) Sieger, M.; Kaim, W.; Stufkens, D. J.; Snoeck, T. L.; Stoll, H.; Zalis, S. Reduced and Excited States of the Intermediates (αdiimine)(C5R5)Rh in Hydride Transfer Catalysis Schemes: EPR and Resonance Raman Spectroscopy, and Comparative DFT Calculations of Co, Rh and Ir Analogues. Dalton Trans. 2004, 3815−3821. (39) Hopmann, K. H. How Accurate Is DFT for Iridium-Mediated Chemistry? Organometallics 2016, DOI: 10.1021/acs.organomet.6b00377. (40) Izutsu, K. Acid−Base Dissociation Constants in Dipolar Aprotic Solvents; IUPAC Chemical Data Series; Blackwell Science: Oxford, U.K., 1990. (41) Morris, R. H. Estimating the Acidity of Transition Metal Hydride and Dihydrogen Complexes by Adding Ligand Acidity Constants. J. Am. Chem. Soc. 2014, 136, 1948−1959. (42) Hu, Y.; Shaw, A. P.; Estes, D. P.; Norton, J. R. Transition-Metal Hydride Radical Cations. Chem. Rev. 2016, 116, 8427−8462. (43) Dedieu, A. Transition Metal Hydrides; VCH Publishers: New York, 1992. (44) Morris, R. H. Brønsted − Lowry Acid Strength of Metal Hydride and Dihydrogen Complexes. Chem. Rev. 2016, 116, 8588− 8654. (45) Sutin, N.; Creutz, C. Properties and Reactivities of the Luminescent Excited States of Polypyridine Complexes of Ruthenium(II) and Osmium(II). In Advances in Chemistry; Wrighton, M. S., Ed.; American Chemical Society: Washington, DC, 1978; Vol. 168. (46) Creutz, C.; Chou, M. H.; Hou, H.; Muckerman, J. T. Hydride Ion Transfer from Ruthenium(II) Complexes in Water: Kinetics and Mechanism. Inorg. Chem. 2010, 49, 9809−9822. (47) Scarborough, C. C.; Sproules, S.; Weyhermüller, T.; Debeer, S.; Wieghardt, K. E. Electronic and Molecular Structures of the Members of the Electron Transfer Series [Cr(tbpy)3]n (n = 3+, 2+, 1+, 0): An Xray Absorption Spectroscopic and Density Functional Theoretical Study. Inorg. Chem. 2011, 50, 12446−12462. (48) Nakai, H.; Jeong, K.; Matsumoto, T.; Ogo, S. Catalytic C−F Bond Hydrogenolysis of Fluoroaromatics by [(η5-C5Me5)RhI(2,2′Bipyridine)]. Organometallics 2014, 33, 4349−4352. (49) Ogo, S.; Hayashi, H.; Uehara, K.; Fukuzumi, S. Crystal Structures of Organometallic Aqua Complexes [Cp*RhIII(bpy)(OH2)]2+ and [Cp*RhIII(6,6′-Me2bpy)(OH2)]2+ Used as Key Catalysts in Regioselective Reduction of NAD+ Analogues. Appl. Organomet. Chem. 2005, 19, 639−643. (50) Kaupmees, K.; Trummal, A.; Leito, I. Basicities of Strong Bases in Water: A Computational Study. Croat. Chem. Acta 2014, 87, 385− 395. (51) Raamat, E.; Kaupmees, K.; Ovsjannikov, G.; Trummal, A.; Kütt, A.; Saame, J.; Koppel, I.; Kaljurand, I.; Lipping, L.; Rodima, T.; et al. Acidities of Strong Neutral Brønsted Acids in Different Media. J. Phys. Org. Chem. 2013, 26, 162−170. (52) Kütt, A.; Movchun, V.; Rodima, T.; Dansauer, T.; Rusanov, E. B.; Leito, I.; Kaljurand, I.; Koppel, J.; Pihl, V.; Koppel, I.; et al. Pentakis(trifluoromethyl)phenyl, a Sterically Crowded and ElectronWithdrawing Group: Synthesis and Acidity of Pentakis(trifluoromethyl)benzene, -Toluene, -Phenol, and -Aniline. J. Org. Chem. 2008, 73, 2607−2620. J

DOI: 10.1021/acs.inorgchem.6b02223 Inorg. Chem. XXXX, XXX, XXX−XXX