Role of Ligand Protonation in Dihydrogen Evolution from a

Article pubs.acs.org/IC

Role of Ligand Protonation in Dihydrogen Evolution from a Pentamethylcyclopentadienyl Rhodium Catalyst Samantha I. Johnson,†,‡ Harry B. Gray,† James D. Blakemore,*,§ and William A. Goddard, III*,‡ †

Center for Chemical Innovation in Solar Fuels, California Institute of Technology, Pasadena, California 91125, United States Materials Research Center, California Institute of Technology, Pasadena, California 91125, United States § Department of Chemistry, University of Kansas, Lawrence, Kansas 66045-7582, United States ‡

S Supporting Information *

ABSTRACT: Recent work has shown that Cp*Rh(bpy) [Cp* = pentamethylcyclopentadienyl, bpy = 2,2′- bipyridine] undergoes endo protonation at the [Cp*] ligand in the presence of weak acid (Et3NH+; pKa = 18.8 in MeCN). Upon exposure to stronger acid (e.g., DMFH+; pKa = 6.1), hydrogen is evolved with unity yield. Here, we study the mechanisms by which this catalyst evolves dihydrogen using density functional theory (M06) with polarizable continuum solvation. The calculations show that the complex can be protonated by weak acid first at the metal center with a barrier of 3.2 kcal/mol; this proton then migrates to the ring to form the detected intermediate, a rhodium(I) compound bearing endo η4-Cp*H. Stronger acid is required to evolve hydrogen, which calculations show happens via a concerted mechanism. The acid approaches and protonates the metal, while the second proton simultaneously migrates from the ring with a barrier of ∼12 kcal/mol. Under strongly acidic conditions, we find that hydrogen evolution can proceed through a traditional metal−hydride species; protonation of the initial hydride to form an H−H bond occurs before migration of the hydride (in the form of a proton) to the [Cp*] ring (i.e., H−H bond formation is faster than hydride−proton tautomerization). This work demonstrates the role of acid strength in accessing different mechanisms of hydrogen evolution. Calculations also predict that modification of the bpy ligand by a variety of functional groups does not affect the preference for [Cp*] protonation, although the driving force for protonation changes. However, we predict that exchange of bpy for a bidentate phosphine ligand will stabilize a rhodium(III) hydride, reversing the preference for bound [Cp*H] found in all computed bpy derivatives and offering an appealing alternative ligand platform for future experimental and computational mechanistic studies of H2 evolution.

■

INTRODUCTION Efficient generation of fuels from renewable resources requires development of catalysts that can couple protons and electrons with minimal energy input and high rates. In hydrogen generation, two protons and two electrons are coupled together by a catalyst:

one operates with weak acids, while another is active with strong acids. Neither pathway involves formation of a traditional metal hydride.4 A model system noted for its ability to efficiently catalyze hydrogen evolution is based on the Cp*Rh(bpy) platform [Cp* = pentamethylcyclopentadienyl, bpy = 2,2′-bipyridine]. Straightforward synthetic routes to such Group 9 complexes bearing [Cp*] and other ancillary bidentate chelates were developed by Maitlis and co-workers, enabling widespread use of these compounds.11,12 Subsequently, Grätzel and Kölle demonstrated that the rhodium system could catalyze hydrogen generation,13 while Steckhan and co-workers demonstrated net hydride transfer to generate NADH from NAD+.14,15 Since these important early reports, research groups have been able to use this catalyst for electrochemical hydrogen generation, both in homogeneous solution and immobilized on electrode surfaces.16−20

2H+ + 2e− → H 2 Both heterogeneous and homogeneous catalysts show activity for the hydrogen evolution reaction; however, establishing the atomistic details of fuel formation is often difficult.1 Depending on the identity of the metal catalysts and conditions used, a variety of oxidation states and/or protonation states could be capable of generating hydrogen.2,3 Moreover, important recent work has demonstrated that “noninnocent” organic ligands in molecular catalysts can accelerate the rate of hydrogen evolution and therefore be intimately involved in enabling coupling of substrate protons and electrons.4−10 One catalyst of note, a nickel phlorin system augmented with a pendant base, has been shown to have two pathways for hydrogen evolution: © XXXX American Chemical Society

Received: July 18, 2017

A

DOI: 10.1021/acs.inorgchem.7b01698 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

mechanistic work, as protonated intermediates in hydrogen evolution are rarely stable.3,41 In this report, we apply density functional theory (DFT) to locate intermediates and transition states along pathways leading to H2 evolution with 1. We find a low-energy pathway for the initial protonation of 1 and establish a deeper understanding of the more complex second protonation event. Results obtained here explain how [η5-Cp*] and [η4Cp*H] can be usefully interconverted and implicate close ligand−metal cooperation in enabling effective H2 evolution catalysis on this platform. Importantly, the calculations reveal the possible roles of the intermediate rhodium(I) complex bearing [η4-Cp*H] in hydrogen evolution with Cp*Rh(bpy).

Hydrogen evolution in this system has been proposed to occur by initial reduction of the rhodium(III) precatalyst such as [Cp*Rh(bpy)(NCCH3)]2+ to form a rhodium(I) species, Cp*Rh(bpy) (1).13 This rhodium(I) complex was detected in the early chemical and pulse radiolytic work21 and has been recently cleanly isolated and crystallized.22,23 Reduction was proposed to be followed by protonation at the metal to form a rhodium(III) hydride as shown in Scheme 1. A second proton transfer followed by solvent recoordination was envisioned to yield H2 and close the catalytic cycle. Scheme 1. Hydrogen Evolution with Cp*Rh(bpy): A Rhodium(III) Hydride and Rhodium(I) Cp*H Complex Were Implicated by Experimental Work

■

METHODS

All calculations were performed using density functional theory. Geometry, frequency, and solvation calculations used the B3LYP functional,42,43 including the Grimme D3 dispersion correction.44 For Rh atoms, we used the Los Alamos angular momentum projected effective core potential with the 2-ζ basis set (LACVP**).45−47 The light elements were described with the 6-31G** basis set. All single point energy calculations were completed with the M06 meta-GGA functional,48 again modified with the D3 Grimme dispersion correction. Because the parameters in M06 were optimized to fit a large training set of molecular complexes and transition state barriers, M06 leads to more reliable barrier heights. Rh was again described with the small core effective core potential (ECP) but now treated with the 3-ζ LACV3P**++ basis set, augmented with f and diffuse functions. The 6-311G**++ basis was used for organic atoms.49,50 Solvation in acetonitrile was applied using the Poisson−Boltzmann polarizable continuum model51,52 with a dielectric constant of 37.5 and probe radius of 2.19 Å. To calculate the free energy of acetonitrile, we calculated the free energy for 1 atm ideal gas and subtracted the experimental energy of vaporization, 1.27 kcal/mol.53 Free energies were computed in according to the following equation:

Interestingly, treatment of 1 in CH3CN with a weak acid,24 triethylammonium (pKa = 18.8 in CH3CN), was recently shown to result in protonation of the [Cp*] ring.25 X-ray crystallography and NMR studies confirm that the site of protonation on the [Cp*] ligand is oriented toward the metal center; i.e., the added proton has an endo disposition. Even upon addition of multiple equivalents of triethylammonium, the protonated complex does not produce H2. However, when the Cp*H complex 3 is treated with a stronger acid, protonated dimethylformamide (pKa = 6.1 in CH3CN), H2 is evolved with unity yield. Importantly, the formation of complex 3 is also observed in aqueous solutions, suggesting a general role for Cp*H in the chemistry of metal complexes of this type.26 The observation of a catalytically competent η4-Cp*H complex in this work contradicts the usual notion from organometallic chemistry that [η5-Cp*] is a spectator ancillary ligand in catalytic systems; instead, close metal−ligand cooperation appears to be involved in effective catalysis here.27−29 Importantly, Cp*H has been demonstrated in previous work to serve as a ligand in several classes of transition-metal complexes.30−35 However, involvement of [Cp*H] in hydrogen evolution catalysis was not previously postulated for this system.36 Considering the high activity of [Cp*] metal complexes in a number of important catalytic systems37,38 as well as recent results by Peters and co-workers demonstrating a possible role for [Cp*H] moieties as H+/e− donors in N2 reduction catalysis,39 an understanding of the role of [Cp*H] in hydrogen evolution with 1 could lead to better catalysts that exploit noninnocent reactivity modes of [Cp*].40 Moreover, the observation of an intermediate in hydrogen evolution with 1 provides a unique opportunity for in-depth

G = EM06 + Gsolv + EZPE + H vib + HTR − T (Svib + Selec + STR ) In this equation, EM06 represents the electronic single point energy; Gsolv is the solvation energy, and EZPE represents the zero-point energy correction. HTR = (12/2kBT) is the translational and rotational enthalpy; Hvib is the vibrational enthalpy, and Svib, Selec, and STR are the vibrational, electronic, and translational and rotational entropies, respectively. Gas phase translational and rotational entropies were scaled by 54%, as suggested by Wertz, to account for liberation in the solvent.54 Zero point energies, enthalpic, and entropic effects were based on frequency calculations at the stationary points, corrected to room temperature. To validate calculations involving organic acids, the pKa values of triethylammonium (Et3NH+) and protonated dimethylformamide (DMFH+) were calculated in acetonitrile and compared to experiment. For these calculations, the value of the proton in solution was calculated from its gas phase free energy (GH+ = H − TS = 2.5kBT − T × 26.04 = −6.3 kcal/mol) plus the empirical free energy of solvation in water at concentration of 1 M (ΔGH+,solv = −265.9 + kBTln(24.5)), wherein kB is the Boltzmann constant and T is room temperature, as found by Tissandier et al.55 To account for solvation in acetonitrile, a value for the free energy of intersolvent proton transfer of 14.1 kcal/ mol was used in accordance with measurements by Roberts and coworkers.56,57 This yields a value of −256.2 kcal/mol for the free energy of the proton solvated in acetonitrile. Using this value, the pKa’s of Et3NH+ and DMFH+ in acetonitrile were calculated as 18.5 and 4.0, respectively, which compare well to the experimentally measured values of 18.8 and 6.1. In calculations involving the Rh complexes, the molecules bearing acidic protons were explicitly included, rather than by implicit inclusion of the energy of the free proton. All calculations were completed in version 8.4 of Jaguar.58 B


Article

Inorganic Chemistry

Scheme 2. Protonation at the Metal Center to Form the Traditional Hydride 2 is the Most Kinetically Feasible Pathway

■

RESULTS AND DISCUSSION Generation of [η4-Cp*H] by Protonation. We explored several pathways leading to the observed product of the first protonation event, η4-Cp*H-bound rhodium(I) (3), as shown in Scheme 2. In this report, all free energies are referenced to isolated 1 and Et3NH+ (and DMFH+ where noted in the second protonation mechanism). We calculated three thermodynamically accessible protonated compounds, differing by the site of protonation by Et3NH+. Formation of a traditional hydride bound to a formal rhodium(III) metal center (complex 2) is exergonic by 1.1 kcal/mol. Formation of 3X or 3, a rhodium(I) complex bearing exo or endo η4-Cp*H, respectively, is exergonic by 7.0 or 7.3 kcal/mol, respectively. Though 3 is the lowest in energy, in agreement with experimental observations,25,26 the energies of the two isomers are close enough in energy that the exo product could also exist. The geometries of 3 and 3X display a significant shift in the [Cp*] ring position with respect to the metal center, consistent with the pyramidalization of one ring carbon and formation of [Cp*H] bound as a bis-olefin to a formally rhodium(I) center. This conversion of [Cp*] to [Cp*H] is reflected in a change in the C−C bond lengths of the ring carbons. In 1, 5 equivalent C−C bonds are present with calculated lengths ranging from 1.425 to 1.448 Å. Upon conversion to 3 or 3X, the bond lengths become inequivalent and localized. Consistent with localization of π electron density in two distinct olefins, the C− C bonds that serve as ligands to rhodium(I) are measured to be 1.415 Å each, while the in-ring C−C bonds to the pyrimidalized carbon are 1.523 Å. The bound olefin bond lengths compare well with the experimental X-ray crystallographic data, which finds those C−C bonds lengths are 1.442(2) Å in 3.25,26 Experimentally, formation of this bound, chelating π-acceptor ligand is reflected in the electronic absorption spectrum of 3,

which suggests a more modest reduction potential compared to 1. X-ray crystallographic data show concomitant lengthening of the inter-ring C−C bond of bpy from 1.422(4) Å in 1 to 1.475(3) Å in 3. This was previously suggested to arise from partial bpy reduction in 1, enforced by the anionic π-donor [Cp*], whereas 3 is stabilized by the neutral bis-olefin πacceptor ligand [Cp*H]. This remarkable change should be reflected in the distribution of electron density in the complex, which is indeed confirmed by our calculations. Loss of electron density in the LUMO of bipyridine is reflected in a calculated lengthening of the inter-ring bpy C−C bond from 1.439 Å in 1 to 1.480 Å in 3, in good agreement with experimental results. The significant filling of the π-symmetry LUMO of bipyridine59 in the calculated structure of 1 is thus consistent with symmetry-allowed coupling to rhodium, which has been discussed in prior work on compounds of this type60 and is favored by the coplanarity of bpy with respect to the rhodiumcentered orbitals.22 Upon conversion to 3, the calculated and experimental structures confirm planarity between bpy and the Rh-centered orbitals is lost and that the Rh center is stabilized by formation of strongly π-accepting [η4-Cp*H]; both of these phenomena contribute to attenuated filling of the bpy π* level in 3. In line with this model, recently published electrochemical data for an analogue of 3, (Cp*H)Rh(bpy)Br, confirm the expected relative reduction potentials of 1 and 3.61 Consistent with this model, the frontier orbitals (Figure 1) of 1 are delocalized over the pz orbitals of bpy, the dyz orbital of the metal, and the py orbitals of the carbons in the [Cp*] ring. Upon protonation at the metal center to form 2, the HOMO is calculated to have contributions from the s orbital of the hydride ligand, the dz2 of rhodium, and the π system of the [Cp*] ligand. In 3, however, the HOMO is more localized on rhodium, in its dz2 orbital. This implies that the formal C


Article

Inorganic Chemistry

features a slightly shorter H−CCp* distance of 1.56 Å which, along with the reduced barrier, supports the steric argument against direct protonation at the ring from the endo position. The final option is direct protonation of the rhodium center, which is represented by TS 1c. At 3.2 kcal/mol, this transition state is significantly more favorable. At 1.41 Å, this transition state features the longest N−H distance among TS 1a, TS 1b, and TS 1c, demonstrating a late reaction with closer similarity to products than reactants. The origin of the late reaction and long N−H distance is likely due to steric factors because deep penetration of NEt3H+ into the cavity defined by the metal and [η5-Cp*] is disfavored. Along this line, inspection of the coordinates for TS 1a and TS 1c reveals a difference in the disposition of the NEt3H+ with respect to the Rh center and bpy ligand of 1. Specifically, there is plausibly a rather long (and therefore weak) interaction between bpy and the ethyl protons of NEt3H+ in the case of TS 1c. As the bipyridine ring system is significantly reduced in 1, we cannot discount the possibility of such an interaction in TS 1c that leads to stabilization over TS 1a that lacks this closer approach to bpy. Notably, the thermodynamically preferred complex is 3, despite the low kinetic barrier for direct metal protonation.55 Complex 2 features a long RhIII−H bond of 1.56 Å, implying a weak bond likely to undergo further reactivity in this framework.63 However, 2 is calculated to support a true hydride ligand: it does not bridge to the [Cp*] ring (H−CCp* = 2.72 Å; ∠ (H−RhCCp*) = 89.9°). Due to the long, weak Rh−H bond, we expected a transition state between 2 and 3, which we call TS 2. This transition state, TS 2, was found and is shown in Scheme 3. Starting with 2, the hydride can move to a bridging position between the rhodium center and [Cp*] with a barrier of 15.2 kcal/mol (see Scheme 3). It is exothermic to tautomerize the hydride to form the [endo-η4-Cp*H]-bound complex 3, a finding supported by lowtemperature 1H NMR experiments from Miller and coworkers.26 Notably, the formation of a strong C−H bond may contribute heavily to the stability of 3, helping to drive the proton transfer to the ring to completion. Finally, the thermally accessible conversion of 2 to 3 contrasts with the plausible

Figure 1. Frontier orbitals of complexes 1, 2, and 3. Partial reduction of bpy in 1 gives rise to bpy ligand contributions to the HOMO of 1, unlike in 3 where no major bpy reduction is implicated by experiment or computational modeling. An isovalue of 0.005 is used.

oxidation state of rhodium is indeed + I, stabilized by πaccepting [η4-Cp*H]. The clear bpy orbital contribution to the HOMO in 1 but not in 3 supports the notion that bpy reduction gives rise to contraction of the inter-ring C−C bond, as predicted and measured in a variety of environments.62 To more fully understand the processes that could give rise to formation of the protonated species, 2, 3, and 3X, we calculated transition states to determine the energetically preferred pathway. The first, TS 1a, involves direct protonation of a carbon atom in [η5-Cp*] by Et3NH+ to form 3 bearing [endo-η4-Cp*H]. In this transition state, the aromaticity of the [Cp*] ring is broken by the incoming proton, resulting in the pyramidalization of the nascent quaternary ring carbon. The distance between the incoming proton and this quaternary ring carbon in the transition state is 1.57 Å. The 19.9 kcal/mol barrier for this route indicates a kinetically slow reaction; TS 1a would be on-path for direct C−H activation with NEt3 as base in the reverse reaction, consistent with the high calculated barrier. This high barrier may have been increased by steric hindrance, as incoming Et3NH+ clashes with the bpy plane. With this rationalization, direct exo protonation from the “top” of the [Cp*] ring to form the analogous state, TS 1b, could be expected to be more favorable. Indeed, TS 1b has a lower energy barrier of 15.8 kcal/mol; however, this barrier is only 4 kcal/mol lower than that of TS 1a. This transition state

Scheme 3. After Formation of the Hydride, the Proton Can Undergo Transfer to the [Cp*] Ring via a Bridging Transition State, TS 2, Resulting in Formation of 3

D


Article

Inorganic Chemistry reverse reaction to convert 3 into 2 or 1; this reverse reaction has a barrier of 22.5 kcal/mol, virtually insurmountable at room temperature, and consistent with experimental observation of rapid formation of 3 from 1. Consistent with calculation of the thermally accessible state TS 2, we observe extension (delocalization) of the frontier orbitals onto the [η5-Cp*] ligand in 2, suggesting the presence of suitably energetic electron density that can serve to lure the proton away from the metal center to form a new C−H bond in close proximity to the metal center. This is further supported experimentally by low-temperature NMR spectra that show initial formation of a hydride species followed by formation of [endo-η4-Cp*H] bound to rhodium. This pathway also supports the experimentally observed exclusive presence of the endo isomer despite the nearly equivalent energies of isomers 3 and 3X. Modified 2,2′-Bipyridine Ligands. To gauge substituent effects on the electronic structure of the Cp*Rh(bpy) complex, functional group were substituted in the 4 and 4′ positions of the 2,2′-bipyridine ligand. The calculated differences in energy between the rhodium(III) hydrides and their tautomeric rhodium(I) complexes bearing [Cp*H] are plotted as a function of the Hammett constant (σp) of the functional groups at the 4 and 4′ positions in Figure 2a. Dimethylamino, methoxy, and trifluoromethyl groups were installed and compared with the parent complex, resulting in Hammett constants spanning a range from −0.83 to 0.54.64 The result shows a linear free energy relationship between ΔG and σp, suggesting that within this small family of bipyridine complexes, the perturbation of electronic structure is limited. There is little change in the Mulliken populations on the nitrogen, rhodium, and relevant carbon atoms in complexes with different functional groups, further confirming retention of the major features of the electronic structure across the series. Of note here, a computational study correlating the hydricities of Cp*Ir(bpy) complexes and σp has recently appeared in the literature, in which good agreement was obtained between measured and calculated hydricities.65 These results imply that, to favor the formation of a rhodium hydride, 4 and 4′ substituents with Hammett constants of ∼2.5 would be needed: only a strongly electron-donating substituent could achieve this on a bipyridine platform. In this family of complexes, it appears that electronic effects from its substituents do not markedly perturb the bipyridine ligand. As an aside, this feature makes this catalyst an attractive candidate for surface attachment: a Cp*Rh(bpy) catalyst bound to a surface would be predicted to perform similarly to the homogeneous analogue. Indeed, surface attachment of this catalyst has proven successful in prior work.16,20 Intuitively, one might expect the more electron-donating ligands to have a less exergonic ΔG between complexes 2 and 3 because more electron density on the metal would favor the formation of the hydride. Nonetheless, this is not seen. Reasons for this can be seen in Figure 2b, in which the energy to form derivatives of 2 and 3 from 1 are plotted against the Hammett constant. As the bpy ligand becomes more electron-withdrawing, the metal complex becomes overall more difficult to protonate. This effect is felt more strongly at the metal center because it is more tightly coupled to the bpy ligand substituents than the [Cp*] fragment. Consequently, the hydride becomes more difficult to form relative to the protonated [Cp*]; this stronger coupling between the rhodium center and the bpy ligand gives rise to the noticeably differing slopes of the trend lines.

Figure 2. (a) The linear correlation between the Hammett constant of functional groups on bpy and the energy difference between derivatives of complexes 2 and 3. (b) As the functional groups on bipyridine become more electron withdrawing, protonation to form both complexes becomes increasingly unfavorable, albeit with different slopes (energy of [RhIII−H] in purple and [(Cp*H)RhI] in blue).

Larger variations are made possible, however, by exploring bidentate ligands beyond bipyridine. Using the bidentate chelating 1,2-bis(dimethylphosphino)ethylene (dmpe) ligand instead of bpy in a model system, the ΔG is shifted just slightly in favor of the RhIII hydride, as shown in Scheme 4. Previously, Scheme 4. Hydride is Slightly Favored in Phosphine-Based Ligand Sets

a [Cp*Rh] diphosphine dihydride was studied; this complex generates free Cp*H upon heating.36 A more closely related hydride has also been experimentally observed in trace amount on a similar chelating diphosphine-based ligand platform.66 Examination of the Mulliken populations shows significantly less positive character on the rhodium atom, consistent with the E


Article

Inorganic Chemistry

Figure 3. Varying the H−H distance and Rh−H distances yields several critical points of interest, geometries of which can be seen in the lower structures a−c.

hydrogen is evolved both from 1 and 3. Understanding the role that the protonated Cp* moiety plays in this reaction is key, as it offers unusual new mechanistic opportunities that might be exploited in other systems. In isolation, the formation of the hydride species 2 from 3 encounters a barrier of 22.5 kcal/mol, which is quite high, as previously shown. Though a concerted route by which the protons approach simultaneously is conceivable, we could not locate an explicit transition state. A transition state using DMF as a proton shuttle was considered as well, but similarly cannot be found. However, a stepped coordinate-scan approach involving a scan of two bond length variables was successful in locating a reasonable pathway and will be applied here. The first variable was the Rh−H1 distance, scanned in increments of 0.076 Å, where H1 is the ring proton of the [Cp*H] ligand in 3. The second variable was the distance between the H1 and the proton of the acid, H2, scanned in increments of 0.115 Å. This resulted in 251 individual geometries and energies, as shown in the contour plot in Figure 3. From these geometries and their energies, several critical points emerged. Point A is a local minimum that represents the protonated Cp* complex 2 with the acid nearby. Point B results from proton H1 migration to form a loosely bound hydride (Rh−H distance of 1.697 Å); the acid proton H2 orients toward this species, suggesting this shallow minimum precedes protonation of the hydride. Point C reflects the system following protonation of the nascent hydride−dihydrogen

sigma-donor properties of the phosphine groups. Thus, in the systems bearing bidentate ligands, hydride tautomerization to form bound [Cp*H] may not be observed for all ligand sets but rather is apparently a characteristic of the bipyridine-bearing systems. The involvement of the low-lying π* orbital of bpy is implicated here. In prior experimental work, the ability for this π* level to accept electron density from the rhodium(I) center in 1 has been demonstrated by both spectroscopic60 and structural methods.22,62 Specifically, the bpy ligand can stabilize the rhodium(I) metal center by delocalization of electron density. Such coupling between the rhodium(I) metal center and the π* level of bpy is enforced and enhanced by the strongly π-donating [η5-Cp*] ligand.67 Based on these results, the metal−hydride interaction can be viewed as significantly weakened in bpy complexes in comparison to their diphosphine analogues. This weak bond is so destabilized thermodynamically that it can be exchanged for formation of a stronger C−H bond despite the cost of breaking the aromaticity of the [η5Cp*] ring. We suggest that future studies of hydrogen evolution with [Cp*Rh] complexes could usefully focus on use of chelating diphosphine ligands. Such studies may uncover additional experimental evidence regarding the role of hydrides in mechanisms of hydrogen evolution with these model catalysts. Protonation of the Cp*H Complex and H−H Bond Formation. In the presence of a stronger acid such as DMFH+ (protonated dimethylformamide, pKa = 6.1 in MeCN), F


Article

Inorganic Chemistry Scheme 5. Routes Involving a Second Protonation by HDMFa

a

Energies are in kcal/mol, and bond lengths are in Å.

(H2) has been released into the solution. The barrier for this is also low and featureless, suggesting that the dihydrogen adduct will be lost as quickly as it formed. Moving to increasing Rh−H distances along the y-axis shows a low energy, featureless profile, further suggesting the rapid loss of H2. Notably, a large barrier in which H2 is formed by direct protonation of 2 was calculated toward the top middle of Figure 3, which will be discussed later (vide infra). The path connecting A and B features a barrier of ca. 12 kcal/mol, which is lower than the defined barrier in TS 2 and is accessible at room temperature. The barrier is reduced by the simultaneous movement of the proton H1 of the Cp*H ligand in 3 toward the Rh metal center and the protons H1 and H2 toward each other. These calculations result in a fairly flat potential energy surface with a broad barrier region, though still with a thermally accessible barrier. Starting with 3, several alternative mechanisms can be envisioned. Direct acid attack at the protonated [Cp*H] can be seen in the top middle of Figure 3 (and explicitly in Scheme S1 of the Supporting Information), in which [η4-Cp*H] acts like an organic hydride donor. However, the HOMO of 3 is largely localized on the formally rhodium(I) center. As a result, there are no reactive electrons on the [Cp*H] ring, and the electron density to generate an equivalent of H− would have to pass through the π system of the said [Cp*H] ring. In line with this elaborate picture, the calculated barrier for the reaction is 43.8 kcal/mol: too high for a room temperature H2 evolution reaction.26 Notably, the calculated high barrier is in line with second order rate constants that have been measured for hydride transfer from aryl rings to carbocation hydride acceptors; these are often several orders of magnitude smaller than their transition-metal hydride counterparts.68−70

Other potential routes for this process are shown in Scheme 5. The [Cp*H] complex 3 can undergo a barrierless second protonation, as in the pathway shown in blue in Scheme 5 connecting intermediates 3 and 5. Complex 5 is the trans isomer of the doubly protonated complex and the lowest energy isomer of all possible combinations. It features a loosely coordinated acetonitrile opposite the hydride ligand. Notably, an analogue (not shown) without the bound acetonitrile is higher in energy by 8 kcal/mol. Stabilization of this species upon binding of acetonitrile is not surprising considering the presence of strongly π-accepting bis-olefin [Cp*H] on a formally rhodium(III) complex. The cis isomers with and without acetonitrile are higher in energy by 3.3 and 15.1 kcal/ mol, respectively. To establish the barrierless nature of the protonation to form 5, we carried out a relaxed coordinate scan of the Rh−H bond distance from 1.766 Å to the resting distance of 1.516 Å with a step size of 0.015 Å. The results of the scan are in Figure 4, in which the energy increases monotonically as the Rh−H distance decreases. From 5, the potential energy surface branches into two potential routes. The higher barrier route through TS 5 in Scheme S1 (see Supporting Information) involves a bond metathesis-like pathway, wherein the [Cp*] ring rotates and the proton from the ring is passed back to the metal. The geometry of this transition state is unique because the proton is not passed directly from overhead to the metalbound proton but rather from a slightly rotated position onto the Rh (dihedral angle ∠ H−Rh−C−H of 45.0°). A representation of this geometry is shown in Figure 5a. We expected that this would go on to form 6, the dihydrogen complex present as a short-lived intermediate before the thermodynamically favored release of hydrogen, shown by 7. G


Article

Inorganic Chemistry

Our attention turned next to calculation of H2 evolution paths from the “traditional” Rh−H species 2, with two possible calculated routes discussed here. The first, TS 3 (Figure 5b), features a side-on attack by DMFH+ to form the dihydrogen adduct 6. The energy of TS 3 shows that barrier to reaction is 7.3 kcal/mol, which is only 8.4 kcal/mol higher than 2. In TS 3, the Rh−H bond length is 1.71 Å, lengthened significantly from the Rh−H bond length of 1.56 Å in 2 but not as long as the Rh−H bonds in the product 6. The H−H bond length in the transition state is 0.91 Å, which relaxes to 0.783 Å in 6. This H−H distance is indicative of a hydrogen complex rather than a tightly bound H2 or dihydride complex.72 Thus, in this transition state, the system moves from bearing a hydride ligand to being a loosely bound H2 adduct. In agreement with these calculations, experimental studies have never detected this fleeting species with bound H2. Release of H2 and coordination of acetonitrile is exothermic and completes the net reaction sequence. A second route was calculated by implementation of a relaxed coordinate scan of the H−H bond length in a configuration with minimized hydride 3 and a nearby DMFH+ (Figure 6). This pathway features protonation of 3

Figure 4. Relaxed coordinate scan moving the Rh−H distance 0.015 Å each step.

However, at a barrier of 27.1 kcal/mol, this process is also unfeasible at room temperature. The published experimental data for H2 evolution in MeCN solvent used two equivalents of DMFH+ to trigger H2 evolution from 3 (which is produced in situ from 1 and NEt3H+). One equivalent of DMFH+ protonates the equivalent of NEt3 that remains in solution following the formation of 3. The second equivalent of added DMFH+ provides the proton needed for the formation of H2. This sequence is supported by gas chromatography measurements, in which addition of only a single equivalent of acid results in little H2 evolution.25 This is due to the single equivalent of NEt3 already present in the starting solution: this single equivalent of base is rapidly protonated before H−H bond formation can take place. It is important to note that if the base is removed, the reaction can progress without the additional second equivalent of acid. However, in the original experiment in the first moment following DMFH+ addition, some NEt3 could be available in solution to serve as a base. This unusual situation resembles those found in some heterogeneous systems in which protonation at one location can stabilize the system for a second protonation, only to be abstracted later.71 This pathway is shown by the alternative transition state TS 4, wherein NEt3 abstracts a proton from the ring to form the singly protonated hydride 2. The barrier for this is similarly high at 20.5 kcal/mol. The analogous transition state with DMF acting as base has a higher barrier of 24.5 kcal/mol, which is reasonable considering that NEt3H+ (pKa = 18.8) is much less acidic than DMFH+ (pKa = 6.1). Taken together, the energetics of these transition states rule out the “direct” participation of [Cp*H] as a pendant proton donor to a Rh−H species and suggest that the only path to H2 generation from the [Cp*H] complex is via the concerted path featured in Figure 3.

Figure 6. Potential energy surface along a decreasing H−H bond distance with acid attack from the lower face of the metal complex. Picture shows geometry at peak point.

from the sterically accessible “face” of the compound near the bipyridine ligand. The coordinate scan is referenced to 3 and the nearby DMFH+ (i.e., this configuration is taken as ΔG = 0). The H−H bond distance was then varied by 0.048 Å, and the new geometry was minimized at each step. This process was continued to allow a map of energies (including enthalpies and entropies) across a H−H distance ranging from 1.638 to 0.538

Figure 5. Relevant transition states (a) TS 5 and (b) TS 3. (c) Attack on hydride from the lower face of the metal complex. H


Article

Inorganic Chemistry

coupling between Rh and bipyridine upon protonation is driven by changing geometric and electronic factors, giving rise to metal−ligand cooperation that enables an efficient catalytic cycle for H−H bond formation. Along this line, we find that the [(Cp*H)Rh] complex is competent for hydrogen evolution upon exposure to DMFH+ and does so through a structure with a proton bridging between the metal center and the [Cp*] ligand. Thus, the transition state for H−H bond formation in the weak acid case is ∼12 kcal/mol. However, when this catalyst operates in strong acid such as excess DMFH+, the calculations suggest that the [Cp*H] structure will not be involved in hydrogen evolution. Thus, various pathways may operate for this catalyst; these depend upon acid strength. Taken together, these studies shed light on the roles that the common ligand [Cp*] can play in catalysis and point to ligand-based strategies that can be used to favor or disfavor its protonation in a catalytic cycle.

Å. Decreasing the H−H distance to 0.729 Å or less results in a large destabilization of the system due to atomic repulsion. As the H−H interatomic distance is increased, a shallow well centered at an H−H distance of 1.399 Å was found, but the energy of this well (∼1 kcal/mol) is smaller than the likely DFT error. This well is followed by a barrier in the potential energy surface of 3.1 kcal/mol at an H−H distance of 1.016 Å. This is an earlier transition state than the side-on attack because both the H−H distance and the H−O distance in DMFH+ are much shorter than the corresponding distances in the reactants. Similarly, the Rh−H distance of 1.66 Å in this configuration implies geometry more similar to that of 3. A representation of the complex at the maximum along the potential energy surface is seen in Figure 5c. As the H−H distance decreases with formation of the H−H bond, the barrier is overcome, leading to a calculated minimum of −1.2 kcal/mol at the H−H distance of 0.777 Å. In this state, the Rh−H distances are 1.86 and 2.08 for the original hydride and transferred proton, respectively. This geometry suggests that 6, a rhodium(III) dihydrogen adduct, would not be a stable species in solution. Summarizing, both of our calculated protonation routes with low barriers involve participation of both the hydride and the metal, a scheme seen previously in biological,73,74 heterogeneous,71 and inorganic75,76 systems. Further, both low-barrier pathways calculated to operate via the hydride 3 are lower in energy than protonation via the bridging pathway, TS 2. We expect that the [Cp*] ring will not participate in hydrogen evolution driven by strong acids such as DMFH+ because fast kinetics will drive the system toward rapid hydrogen evolution. Thus, the [Cp*H] compound 3 is likely not relevant to the mechanism of H−H bond formation in the strong acid case but is likely involved in H−H bond formation for the case of weaker acids. Solis et al. also identified a case in which two separate catalytic cycles can operate for a catalyst, with these being distinct for two acids of different strengths.4 Similar mechanistic branching was observed by Rountree et al.77 and Das et al.78 In future experimental mechanistic work on the system described here, we note also in our conclusion that multiple paths likely exist, and the chosen conditions will contribute to the active mechanism(s) of H−H bond formation. The presence of multiple pathways and/or off-cycle intermediates may be implied in the slow turnover associated with [Cp*Rh] catalysts for formate dehydrogenation, relative to their [Cp*Co] and [Cp*Ir] analogues.79,80

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b01698. Free energy surface for additional transition states and tabulated components of calculated energies (PDF) Geometric coordinates of all transition states and intermediates (XYZ)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]; Phone: +1-785-864-3019. *E-mail: [email protected]; Phone: +1-626-395-3093. ORCID

James D. Blakemore: 0000-0003-4172-7460 William A. Goddard III: 0000-0003-0097-5716 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank Dr. Jay Winkler, Dr. Robert Nielsen, Dr. Sijia Dong, Dr. Davide Lionetti, Yufeng Huang, and Sydney Corona for numerous helpful discussions. This work was supported by the US National Science Foundation through the CCI Solar Fuels Program (CHE-1305124) and the Resnick Sustainability Institute at Caltech (fellowship to S.I.J.). J.D.B. was supported during preparation of this manuscript by an award from the State of Kansas through the University of Kansas New Faculty General Research Fund.

■

CONCLUSIONS We investigated the formation of unusual [(Cp*H)Rh] compounds that can arise from protonation of Cp*Rh(bpy) and have now shown that protonation by a weak acid occurs first at the metal. The delivered proton can then bridge to the [Cp*] ring to form the observed endo protonated thermodynamic product. We considered several modifications of the bpy ligand to show that the thermodynamic product is always the protonated [Cp*] ligand due to the electronic structural features unique to bipyridine ligands. However, this situation changes with the use of diphosphine ligands, which instead prefer the traditional metal hydride, a useful finding for future studies of these model catalysts. In light of these calculations, we suggest that the ability of bipyridine ligands to accept electronic density into their π-symmetry LUMO from the reducing [Cp*Rh] fragment contributes heavily to the unique reactivity of these complexes with protons. Modulation of this

■

REFERENCES

(1) McKone, J. R.; Marinescu, S. C.; Brunschwig, B. S.; Winkler, J. R.; Gray, H. B. Earth-abundant hydrogen evolution electrocatalysts. Chem. Sci. 2014, 5, 865−878. (2) Artero, V.; Fontecave, M. Some general principles for designing electrocatalysts with hydrogenase activity. Coord. Chem. Rev. 2005, 249, 1518−1535. (3) Dempsey, J. L.; Winkler, J. R.; Gray, H. B. Mechanism of H2 Evolution from a Photogenerated Hydridocobaloxime. J. Am. Chem. Soc. 2010, 132, 16774−16776. (4) Solis, B. H.; Maher, A. G.; Dogutan, D. K.; Nocera, D. G.; Hammes-Schiffer, S. Nickel phlorin intermediate formed by proton-

I


Article

Inorganic Chemistry coupled electron transfer in hydrogen evolution mechanism. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 485−492. (5) Solis, B. H.; Maher, A. G.; Honda, T.; Powers, D. C.; Nocera, D. G.; Hammes-Schiffer, S. Theoretical Analysis of Cobalt Hangman Porphyrins: Ligand Dearomatization and Mechanistic Implications for Hydrogen Evolution. ACS Catal. 2014, 4, 4516−4526. (6) Loewen, N. D.; Thompson, E. J.; Kagan, M.; Banales, C. L.; Myers, T. W.; Fettinger, J. C.; Berben, L. A. A pendant proton shuttle on [Fe4N(CO)12]- alters product selectivity in formate vs. H2 production via the hydride [H-Fe4N(CO)12]. Chem. Sci. 2016, 7, 2728−2735. (7) O’Hagan, M.; Ho, M.-H.; Yang, J. Y.; Appel, A. M.; DuBois, M. R.; Raugei, S.; Shaw, W. J.; DuBois, D. L.; Bullock, R. M. Proton Delivery and Removal in [Ni(PR2NR′2)2]2+ Hydrogen Production and Oxidation Catalysts. J. Am. Chem. Soc. 2012, 134, 19409−19424. (8) Ho, M.-H.; Rousseau, R.; Roberts, J. A. S.; Wiedner, E. S.; Dupuis, M.; DuBois, D. L.; Bullock, R. M.; Raugei, S. Ab Initio-Based Kinetic Modeling for the Design of Molecular Catalysts: The Case of H2 Production Electrocatalysts. ACS Catal. 2015, 5, 5436−5452. (9) Wilson, A. D.; Shoemaker, R. K.; Miedaner, A.; Muckerman, J. T.; DuBois, D. L.; DuBois, M. R. Nature of hydrogen interactions with Ni(II) complexes containing cyclic phosphine ligands with pendant nitrogen bases. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 6951−6956. (10) Lacy, D. C.; Roberts, G. M.; Peters, J. C. The Cobalt Hydride that Never Was: Revisiting Schrauzer’s “Hydridocobaloxime”. J. Am. Chem. Soc. 2015, 137, 4860−4864. (11) Maitlis, P. M. (Pentamethylcyclopentadienyl)rhodium and -iridium complexes: approaches to new types of homogeneous catalysts. Acc. Chem. Res. 1978, 11, 301−307. (12) Nutton, A.; Bailey, P. M.; Maitlis, P. M. Pentamethylcyclopentadienyl-rhodium and -iridium complexes. Part 29. Syntheses and X-ray structure determinations of [{Rh(C5Me5)}2(OH)3]OH[middle dot]11H2O and [{lr(C5Me5)}2(OH)3]O2CMe[middle dot]14H2O and related complexes. J. Chem. Soc., Dalton Trans. 1981, 1997−2002. (13) Kölle, U.; Grützel, M. Organometallic Rhodium(III) Complexes as Catalysts for the Photoreduction of Protons to Hydrogen on Colloidal TiO2. Angew. Chem., Int. Ed. Engl. 1987, 26, 567−570. (14) Ruppert, R.; Herrmann, S.; Steckhan, E. Efficient indirect electrochemical in-situ regeneration of nadh:electrochemically driven enzymatic reduction of pyruvate catalyzed by d-ldh. Tetrahedron Lett. 1987, 28, 6583−6586. (15) Ruppert, R.; Herrmann, S.; Steckhan, E. Very efficient reduction of NAD(P)+ with formate catalysed by cationic rhodium complexes. J. Chem. Soc., Chem. Commun. 1988, 1150−1151. (16) Blakemore, J. D.; Gupta, A.; Warren, J. J.; Brunschwig, B. S.; Gray, H. B. Noncovalent Immobilization of Electrocatalysts on Carbon Electrodes for Fuel Production. J. Am. Chem. Soc. 2013, 135, 18288− 18291. (17) Lattimer, J. R. C.; Blakemore, J. D.; Sattler, W.; Gul, S.; Chatterjee, R.; Yachandra, V. K.; Yano, J.; Brunschwig, B. S.; Lewis, N. S.; Gray, H. B. Assembly, characterization, and electrochemical properties of immobilized metal bipyridyl complexes on silicon(111) surfaces. Dalton Trans. 2014, 43, 15004−15012. (18) Caix, C.; Chardon-Noblat, S.; Deronzier, A.; Moutet, J.-C.; Tingry, S. Pentamethylcyclopentadienyl)(polypyridyl) rhodium and iridium complexes as electrocatalysts for the reduction of protons to dihydrogen and the hydrogenation of organics. J. Organomet. Chem. 1997, 540, 105−111. (19) Chardon-Noblat, S.; Cosnier, S.; Deronzier, A.; Vlachopoulos, N. Electrochemical properties of [(C5Me5)RhIII(L)Cl]+ complexes (L = 2,2′-bipyridine or 1,10-phenanthroline derivatives) in solution in related polypyrrolic films. Application to electrocatalytic hydrogen generation. J. Electroanal. Chem. 1993, 352, 213−228. (20) Cosnier, S.; Deronzier, A.; Vlachopoulos, N. Carbon/poly {pyrrole-[(C5Me5)Rh(bpy)Cl]+} modified electrodes; a molecularlybased material for hydrogen evolution (bpy = 2,2′-bipyridine). J. Chem. Soc., Chem. Commun. 1989, 1259−1261.

(21) Kölle, U.; Kang, B.-S.; Infelta, P.; Comte, P.; Grätzel, M. Elektrochemische und pulsradiolytische Reduktion von (Pentamethylcyclopentadienyl)(polypyridyl)rhodium-Komplexen. Chem. Ber. 1989, 122, 1869−1880. (22) 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. (23) 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. (24) 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. (25) Quintana, L. M. A.; Johnson, S. I.; Corona, S. L.; Villatoro, W.; Goddard, W. A.; Takase, M. K.; VanderVelde, D. G.; Winkler, J. R.; Gray, H. B.; Blakemore, J. D. Proton−hydride tautomerism in hydrogen evolution catalysis. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 6409−6414. (26) Pitman, C. L.; Finster, O. N. L.; Miller, A. J. M. Cyclopentadiene-mediated hydride transfer from rhodium complexes. Chem. Commun. 2016, 52, 9105−9108. (27) Ben-Ari, E.; Leitus, G.; Shimon, L. J. W.; Milstein, D. Metal− Ligand Cooperation in C−H and H2 Activation by an Electron-Rich PNP Ir(I) System: Facile Ligand Dearomatization−Aromatization as Key Steps. J. Am. Chem. Soc. 2006, 128, 15390−15391. (28) Khusnutdinova, J. R.; Milstein, D. Metal−Ligand Cooperation. Angew. Chem., Int. Ed. 2015, 54, 12236−12273. (29) Zhang, J.; Leitus, G.; Ben-David, Y.; Milstein, D. Facile Conversion of Alcohols into Esters and Dihydrogen Catalyzed by New Ruthenium Complexes. J. Am. Chem. Soc. 2005, 127, 10840−10841. (30) Zamorano, A.; Rendón, N.; Valpuesta, J. E. V.; Á lvarez, E.; Carmona, E. Synthesis and Reactivity toward H2 of (η5-C5Me5)Rh(III) Complexes with Bulky Aminopyridinate Ligands. Inorg. Chem. 2015, 54, 6573−6581. (31) Green, M. L. H.; Pratt, L.; Wilkinson, G. 760. A new type of transition metal-cyclopentadiene compound. J. Chem. Soc. 1959, 0, 3753−3767. (32) Macias, R.; Holub, J.; Kennedy, J. D.; Stibr, B.; Thornton-Pett, M.; Clegg, W. Effects of metal-centre orbital control on cluster character and electron distribution between borane and hydrocarbon ligands; significance of the structures of [μ]-9,10-(SMe)-8,8-(PPh3)2nido-8,7-IrSB 9H9] and [[μ]-9,10-(SMe)-8-([η]4-C5Me5H)- nido8,7-RhSB9H9]. J. Chem. Soc., Dalton Trans. 1997, 149−152. (33) Enders, M.; Fernández, P.; Kaschke, M.; Kohl, G.; Ludwig, G.; Pritzkow, H.; Rudolph, R. Coordination chemistry of neutral quinolyland aminophenylcyclopentadiene derivatives. J. Organomet. Chem. 2002, 641, 81−89. (34) Tsai, W.-M.; Rausch, M. D.; Rogers, R. D. Improved Synthesis of Pentabenzylcyclopentadiene and Study of the Reaction between Pentabenzylcyclopentadiene and Iron Pentacarbonyl. Organometallics 1996, 15, 2591−2594. (35) Liu, L.-K.; Luh, L.-S. Consecutive Butylations on the Cyclopentadienyl Ring of the [(η5-C5H5)Fe(CO)2PPh3]+ Cation. Organometallics 1996, 15, 5263−5265. (36) Jones, W. D.; Kuykendall, V. L.; Selmeczy, A. D. Ring migration reactions of (C5Me5)Rh(PMe3)H2. Evidence for η3 slippage and metal-to-ring hydride migration. Organometallics 1991, 10, 1577− 1586. (37) 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. Half-Sandwich Iridium Complexes for Homogeneous Water-Oxidation Catalysis. J. Am. Chem. Soc. 2010, 132, 16017−16029. (38) Brintzinger, H.; Bercaw, J. E. Bis(pentamethylcyclopentadienyl)titanium(II). Isolation and reactions with hydrogen, nitrogen, and carbon monoxide. J. Am. Chem. Soc. 1971, 93, 2045−2046. (39) Chalkley, M. J.; Del Castillo, T. J.; Matson, B. D.; Roddy, J. P.; Peters, J. C. Catalytic N2-to-NH3 Conversion by Fe at Lower Driving Force: A Proposed Role for Metallocene-Mediated PCET. ACS Cent. Sci. 2017, 3, 217. J


Article

Inorganic Chemistry

software program with strengths in life and materials sciences. Int. J. Quantum Chem. 2013, 113, 2110−2142. (59) Creutz, C. Bipyridine Radical Ions. Comments Inorg. Chem. 1982, 1, 293−311. (60) 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. (61) Peng, Y.; Ramos-Garcés, M. V.; Lionetti, D.; Blakemore, J. D. Structural and Electrochemical Consequences of [Cp*] Ligand Protonation. Inorg. Chem. 2017, 1, DOI: 10.1021/acs.inorgchem.7b01895. (62) Gore-Randall, E.; Irwin, M.; Denning, M. S.; Goicoechea, J. M. Synthesis and Characterization of Alkali-Metal Salts of 2,2′- and 2,4′Bipyridyl Radicals and Dianions. Inorg. Chem. 2009, 48, 8304−8316. (63) Bakac, A. Aqueous rhodium(III) hydrides and mononuclear rhodium(II) complexes. Dalton Trans. 2006, 1589−1596. (64) Hansch, C.; Leo, A.; Taft, R. W. A survey of Hammett substituent constants and resonance and field parameters. Chem. Rev. 1991, 91, 165−195. (65) Brereton, K. R.; Bellows, S. M.; Fallah, H.; Lopez, A. A.; Adams, R. M.; Miller, A. J. M.; Jones, W. D.; Cundari, T. R. Aqueous Hydricity from Calculations of Reduction Potential and Acidity in Water. J. Phys. Chem. B 2016, 120, 12911−12919. (66) Carmona, D.; Cativiela, C.; Garcia-Correas, R.; Lahoz, F. J.; Lamata, M. P.; Lopez, J. A.; De Viu, M. P. L.-R.; Oro, L. A.; Jose, E. S.; Viguri, F. Chiral rhodium complexes as catalysts in Diels-Alder reactions. Chem. Commun. 1996, 1247−1248. (67) Lionetti, D.; Day, V. W.; Blakemore, J. D. Synthesis and Electrochemical Properties of Half-Sandwich Rhodium and Iridium Methyl Complexes. Organometallics 2017, 36, 1897−1905. (68) Horn, M.; Schappele, L. H.; Lang-Wittkowski, G.; Mayr, H.; Ofial, A. R. Towards a Comprehensive Hydride Donor Ability Scale. Chem. - Eur. J. 2013, 19, 249−263. (69) Mayr, H.; Lang, G.; Ofial, A. R. Reactions of Carbocations with Unsaturated Hydrocarbons: Electrophilic Alkylation or Hydride Abstraction? J. Am. Chem. Soc. 2002, 124, 4076−4083. (70) Cheng, T.-Y.; Brunschwig, B. S.; Bullock, R. M. Hydride Transfer Reactions of Transition Metal Hydrides: Kinetic Hydricity of Metal Carbonyl Hydrides. J. Am. Chem. Soc. 1998, 120, 13121−13137. (71) Huang, Y.; Nielsen, R. J.; Goddard, W. A.; Soriaga, M. P. The Reaction Mechanism with Free Energy Barriers for Electrochemical Dihydrogen Evolution on MoS2. J. Am. Chem. Soc. 2015, 137, 6692− 6698. (72) Crabtree, R. H. Dihydrogen Complexation. Chem. Rev. 2016, 116, 8750−8769. (73) Eilers, G.; Schwartz, L.; Stein, M.; Zampella, G.; de Gioia, L.; Ott, S.; Lomoth, R. Ligand versus Metal Protonation of an Iron Hydrogenase Active Site Mimic. Chem. - Eur. J. 2007, 13, 7075−7084. (74) Carroll, M. E.; Barton, B. E.; Rauchfuss, T. B.; Carroll, P. J. Synthetic Models for the Active Site of the [FeFe]-Hydrogenase: Catalytic Proton Reduction and the Structure of the Doubly Protonated Intermediate. J. Am. Chem. Soc. 2012, 134, 18843−18852. (75) Besora, M.; Lledos, A.; Maseras, F. Protonation of transitionmetal hydrides: a not so simple process. Chem. Soc. Rev. 2009, 38, 957−966. (76) Johnson, S. I.; Nielsen, R. J.; Goddard, W. A. Selectivity for HCO2− over H2 in the Electrochemical Catalytic Reduction of CO2 by (POCOP)IrH2. ACS Catal. 2016, 6, 6362−6371. (77) Rountree, E. S.; Dempsey, J. L. Potential-Dependent Electrocatalytic Pathways: Controlling Reactivity with pKa for Mechanistic Investigation of a Nickel-Based Hydrogen Evolution Catalyst. J. Am. Chem. Soc. 2015, 137, 13371−13380. (78) Das, P.; Ho, M.-H.; O’Hagan, M.; Shaw, W. J.; Morris Bullock, R.; Raugei, S.; Helm, M. L. Controlling proton movement: electrocatalytic oxidation of hydrogen by a nickel(II) complex containing proton relays in the second and outer coordination spheres. Dalton Trans. 2014, 43, 2744−2754.

(40) Kefalidis, C. E.; Perrin, L.; Burns, C. J.; Berg, D. J.; Maron, L.; Andersen, R. A. Can. a pentamethylcyclopentadienyl ligand act as a proton-relay in f-element chemistry? Insights from a joint experimental/theoretical study. Dalton Trans. 2015, 44, 2575−2587. (41) Marinescu, S. C.; Winkler, J. R.; Gray, H. B. Molecular mechanisms of cobalt-catalyzed hydrogen evolution. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 15127−15131. (42) Becke, A. D. Density Functional Thermochemistry, 3. The Role of Exact Exchange. J. Chem. Phys. 1993, 98, 5648−5652. (43) Lee, C. T.; Yang, W. T.; Parr, R. G. Development of the ColleSalvetti Correlation-Energy Formula into a Functional of the Electron Density. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785−789. (44) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010, 132, 154104. (45) Melius, C. F.; Goddard, W. A. Initio Effective Potentials for Use in Molecular Quantum Mechanics. Phys. Rev. A: At., Mol., Opt. Phys. 1974, 10, 1528−1540. (46) Melius, C. F.; Olafson, B. D.; Goddard, W. A. Fe and Ni AB initio effective potentials for use in molecular calculations. Chem. Phys. Lett. 1974, 28, 457−462. (47) Hay, P. J.; Wadt, W. R. Ab Initio Effective Core Potentials for Molecular Calculations - Potentials for K to Au including the Outermost Core Orbitals. J. Chem. Phys. 1985, 82, 299−310. (48) Zhao, Y.; Truhlar, D. G. The M06 suite of density functionals for main group thermochemistry, thermochemical kinetics, noncovalent interactions, excited states, and transition elements: two new functionals and systematic testing of four M06-class functionals and 12 other functionals. Theor. Chem. Acc. 2008, 120, 215−241. (49) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. SelfConsistent Molecular Orbital Methods. 20. Basis Set for correlated Wave-Functions. J. Chem. Phys. 1980, 72, 650−654. (50) Clark, T.; Chandrasekhar, J.; Spitznagel, G. W.; Schleyer, P. V. Efficient Diffuse Function-Augmented Basis Sets for Anion Calculations. 3. The 3-21+G Basis Set for 1st Row elements. J. Comput. Chem. 1983, 4, 294−301. (51) Marten, B.; Kim, K.; Cortis, C.; Friesner, R. A.; Murphy, R. B.; Ringnalda, M. N.; Sitkoff, D.; Honig, B. New Model for Calculation of Solvation Free Energies: Correction of Self-Consistent Reaction Field Continuum Dielectric Theory for Short-Range Hydrogen-Bonding Effects. J. Phys. Chem. 1996, 100, 11775−11788. (52) Tannor, D. J.; Marten, B.; Murphy, R.; Friesner, R. A.; Sitkoff, D.; Nicholls, A.; Honig, B.; Ringnalda, M.; Goddard, W. A. Accurate First Principles Calculation of Molecular Charge Distributions and Solvation Energies from Ab Initio Quantum Mechanics and Continuum Dielectric Theory. J. Am. Chem. Soc. 1994, 116, 11875− 11882. (53) Bridgeman, O. C.; Aldrich, E. W. Vapor Pressure Tables for Water. J. Heat Transfer 1964, 86, 279−286. (54) Wertz, D. H. Relationship between the gas-phase entropies of molecules and their entropies of solvation in water and 1-octanol. J. Am. Chem. Soc. 1980, 102, 5316−5322. (55) Tissandier, M. D.; Cowen, K. A.; Feng, W. Y.; Gundlach, E.; Cohen, M. H.; Earhart, A. D.; Coe, J. V.; Tuttle, T. R. The proton’s absolute aqueous enthalpy and Gibbs free energy of solvation from cluster-ion solvation data. J. Phys. Chem. A 1998, 102, 7787−7794. (56) Pegis, M. L.; Roberts, J. A. S.; Wasylenko, D. J.; Mader, E. A.; Appel, A. M.; Mayer, J. M. Standard Reduction Potentials for Oxygen and Carbon Dioxide Couples in Acetonitrile and N,N-Dimethylformamide. Inorg. Chem. 2015, 54, 11883−11888. (57) Roberts, J. A. S.; Bullock, R. M. Direct Determination of Equilibrium Potentials for Hydrogen Oxidation/Production by Open Circuit Potential Measurements in Acetonitrile. Inorg. Chem. 2013, 52, 3823−3835. (58) Bochevarov, A. D.; Harder, E.; Hughes, T. F.; Greenwood, J. R.; Braden, D. A.; Philipp, D. M.; Rinaldo, D.; Halls, M. D.; Zhang, J.; Friesner, R. A. Jaguar: A high-performance quantum chemistry K


Article

Inorganic Chemistry (79) Himeda, Y.; Onozawa-Komatsuzaki, N.; Sugihara, H.; Arakawa, H.; Kasuga, K. Half-Sandwich Complexes with 4,7-Dihydroxy-1,10phenanthroline: Water-Soluble, Highly Efficient Catalysts for Hydrogenation of Bicarbonate Attributable to the Generation of an Oxyanion on the Catalyst Ligand. Organometallics 2004, 23, 1480− 1483. (80) Badiei, Y. M.; Wang, W.-H.; Hull, J. F.; Szalda, D. J.; Muckerman, J. T.; Himeda, Y.; Fujita, E. Cp*Co(III) Catalysts with Proton-Responsive Ligands for Carbon Dioxide Hydrogenation in Aqueous Media. Inorg. Chem. 2013, 52, 12576−12586.

L


Role of Ligand Protonation in Dihydrogen Evolution from a

Recommend Documents