A Dinuclear Iridium(V,V) Oxo-Bridged Complex ... - ACS Publications

Mar 23, 2018 - ABSTRACT: We report a general method for the preparation and crystallization of highly oxidized metal complexes that are difficult to p...
5 downloads 8 Views 3MB Size
Download PDF
Article Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX

pubs.acs.org/IC

A Dinuclear Iridium(V,V) Oxo-Bridged Complex Characterized Using a Bulk Electrolysis Technique for Crystallizing Highly Oxidizing Compounds Dimitar Y. Shopov,#,† Liam S. Sharninghausen,#,† Shashi Bhushan Sinha,#,† Brandon Q. Mercado,† David Balcells,*,‡ Gary W. Brudvig,*,† and Robert H. Crabtree*,† †

Department of Chemistry, Yale University, 225 Prospect Street, New Haven, Connecticut 06520, United States Hylleraas Center for Quantum Molecular Sciences, Department of Chemistry, University of Oslo, P.O. Box 1033, Blindern, 0315 Oslo, Norway

‡

S Supporting Information *

ABSTRACT: We report a general method for the preparation and crystallization of highly oxidized metal complexes that are difficult to prepare and handle by more conventional means. This method improves typical bulk electrolysis and crystallization conditions for these reactive species by substituting oxidation-prone organic electrolytes and precipitants with oxidation-resistant compounds. Specifically, we find that CsPF6 is an effective inert electrolyte in acetonitrile, and appears to have general applicability to electrochemical studies in this solvent. Likewise, CCl4 is not only an oxidation-resistant precipitant for crystallization from MeCN but it also enters the lattice. In this way, we synthesized and characterized an Ir(V,V) mono-μ-oxo dimer which only forms at a very high potential (1.9 V vs NHE). This compound, having the highest isolated oxidation state in this redox-active system, cannot be formed chemically. DFT calculations show that the oxidation is centered on the Ir−O−Ir core and facilitated by strong electrondonation from the pyalk (2-(2-pyridinyl)-2-propanolate) ligand. TD-DFT simulations of the UV−visible spectrum reveal that its royal blue color arises from electron excitations with mixed LMCT and Laporte-allowed d−d character. We have also crystallographically characterized a related monomeric Ir(V) complex, similarly prepared by oxidizing a previously reported Ir(IV) compound at 1.7 V, underscoring the general applicability of this method.

■

INTRODUCTION The rising importance of water1 and CH2 oxidation catalysis has driven increased interest in high-oxidation-state inorganic chemistry.3 As yet, limited work is available on methods, protocols, and ligand systems for the synthesis, handling, and employment of reactive, highly oxidized complexes, in contrast to the extensive work on highly reduced species. The development of ligands able to withstand oxidation, while at the same time being strong enough electron donors to stabilize high oxidation states, as previously articulated by Collins,4 is one of the main challenges in high-valent coordination chemistry.4,5 In seminal work in this area, Collins and coworkers developed a family of oxidation-resistant macrocyclic ligands and outlined rules for predicting ligand stability to oxidation.4 In our prior reports, we have shown that the pyridine tert-alkoxide chelate ligand pyalk (Scheme 1), 2-(2pyridinyl)-2-propanolate, permits access to stable Rh(IV)6 and Ir(V)3f complexes, the ligand itself being remarkably oxidatively stable. In pyalk, the alkoxide donor group provides very strong σ- and π-donation to the metal, while the pyridine ligand enhances coordinative stability by remaining bound even if the alkoxide becomes protonated. Additionally, the methyl groups block oxidation at what would otherwise be a highly vulnerable © XXXX American Chemical Society

bridging carbon, and contribute to amphiphilic solubility properties. Iridium complexes of pyalk have provided both highly active water oxidation catalysts (WOCs),7 as well as an array of stable high oxidation states.3f,5 In prior work, we found that an Ir WOC, the “blue solution”, formed from Cp*Ir(pyalk)Cl or Ir(pyalk)(CO)2 with loss of the organometallic ligands, can be driven for WO either electrochemically or chemically.7 Although the blue solution has been intensively studied, no crystal structure of the resting state or any other likely intermediate has ever been obtained. The available spectroscopic evidence suggests that the catalyst resting state is a mixture of Ln(pyalk)IrIV−O−IrIV(pyalk)Ln bridged dinuclear isomers, containing a 1:1 Ir to pyalk ratio as well as a variety of ligands derived from the reaction medium. This led us to synthesize model complexes of the type Cl(pyalk)2IrIV−O− IrIV(pyalk)2Cl (1IV,IV).8,9 These were isolated in well-defined isomeric forms and identified by single crystal X-ray crystallography. Prior work10 has detailed the III,III, IV,IV, and IV,V oxidation states of the system, all capable of reversible Received: March 23, 2018

A

DOI: 10.1021/acs.inorgchem.8b00757 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Scheme 1. Four Chemically and Electrochemically Accessible Redox States of 1a

a

Adapted with permission from ref 10.

redox interconversion (Scheme 1). The IV,V state is unexpectedly stable for such a highly oxidized complex, a fact we attribute to delocalization across both metal centers. The Ir(V) oxidation state is implicated in Ir-catalyzed water and CH oxidation catalysis, where reactive Ir(V) oxo species are often invoked as key intermediates.11 However, there are few examples of well characterized Ir(V) coordination complexes, no doubt due to their reactivity and instability. Recent computational studies11a have proposed Ir(V,V) oxobridged species as key catalytic intermediates in WO catalysis by the “blue solution”; given the stability of Ir(IV,V) in our oxo-dimer model system,10 we sought to access the V,V state. While we saw a redox couple at 1.9 V vs NHE,10 the extremely oxidizing nature of the product posed difficulties for detailed study. Preparing and handling such reactive species, having potentials at the limit of the range accessible with common one-electron chemical oxidants,12 is a general challenge in the field. Not only was our oxidized product inaccessible with the usual chemical oxidants but it was also sensitive to water and most organic solvents. Here, we report a novel protocol for electrochemical preparation and crystallization which preserves this highly oxidized species by minimizing the involvement of reducing species. Specifically, bulk electrolysis in acetonitrile with CsPF6 electrolyte, followed by crystallization with CCl4 as precipitant, successfully yields the oxidized Ir(V,V) complex, and allows crystallographic characterization [Warning: CCl4 is highly toxic and should be handled with caution]. CCl4 may aid crystallization because it becomes incorporated in the lattice. The entire procedure involves only a single substance with any degree of oxidative vulnerability, MeCN, but this solvent fortunately proved inert under our conditions. The same technique also facilitated the preparation of an Ir(V) pyalk monomer, again inaccessible with standard oxidants and techniques; this method may therefore prove general for similar systems.

Figure 1. Cyclic voltammogram of 1IV,IV (∼0.2 mM) in DCM solution (scan rate 100 mV/s) with 0.1 M NBu4PF6 electrolyte. Adapted with permission from ref 10.

blue solution, consistent with the spectroelectrochemical results. This oxidized state was highly reactive, reverting to the purple 1IV,V state within minutes at room temperature or hours at −80 °C. At first it seemed that this instability would preclude X-ray crystallography. On further study, however, we found that 1V,V is highly water-sensitive, and its degradation was primarily due to atmospheric moisture. In contrast, the compound’s reactivity with MeCN and CH2Cl2 appears to be minimal, resulting in a room temperature lifetime of several hours under inert atmosphere. Unfortunately, the water sensitivity may not be associated with water oxidation to O2, since no O2 was detected by gas chromatography. The corresponding oxidized product must be solvent derived but we were unable to obtain evidence for its identity. We achieved successful bulk preparation and isolation of 1V,V by electrolysis under carefully selected conditions in a moisturefree glovebox with a high-purity solvent. Acetonitrile is the preferred solvent owing to its wide electrochemical window and ability to dissolve inorganic salts. As an electrolyte, we found that CsPF6, an oxidation-resistant salt seldom utilized in electrochemistry, proved sufficiently soluble (between 0.05 and 0.1 M at STP) to support adequate current. This allowed us to circumvent the use of potentially oxidizable organic salts, such as NBu4PF6. Although our conditions successfully afford 1V,V, it is present with other solution species and its reactivity makes further manipulations challenging. We therefore sought to crystallize the solution directly from the bulk electrolysis medium. In order to minimize introduction of contaminants, we chose vapor diffusion, which necessitated a precipitating solvent that is miscible with acetonitrile and highly oxidation-resistant. Of the solvents tried, only CCl4 had the right properties, being nonpolar and oxidatively stable. To further retard decom-

■

RESULTS AND DISCUSSION Characterization of 1V,V. Electrochemical Synthesis. Cyclic voltammetry of 1 in organic solutions shows a reversible or quasi-reversible redox couple at a potential of 1.9 V vs NHE, ∼1 V higher than the Ir IV,IV/IV,V couple (Figure 1). Spectroelectrochemistry (see SI, Figure S1) showed that this redox process is fully reversible, with no degradation on redox cycling despite the high potential. Chemical oxidants failed to give the same transformation so we restricted subsequent work to bulk electrolysis. Electrochemical oxidation of 1IV,IV or 1IV,V solutions at ca. 1.9 V vs NHE under ambient conditions led to formation of a deep B

DOI: 10.1021/acs.inorgchem.8b00757 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry position, we used reduced temperature (−20 °C). However, this proved to excessively slow the rate of vapor diffusion, given the relatively modest volatility of our solvents. We therefore moved to a sealable glass vessel which could be evacuated, allowing us to reduce the volume of the electrolytic medium and carry out the vapor diffusion itself under static reduced pressure, increasing the rate of gas diffusion. One potential concern here is that the excess electrolyte would interfere with crystallization. However, CsPF6 quickly precipitates from the solution in pure form, while the complex only crystallizes subsequently. In fact, the complex crystallized remarkably well: a small number of comparatively large (up to 1 mm) crystals of 1V,V formed, while the 1IV,V state remained fully dissolved even at a high CCl4/MeCN ratio. Because of the large size of the crystals, isolation of pure material was possible via manual separation of the two materials. Once crystallized, the compound can be handled more easily, the crystals showing no observable degradation under paraffin oil under ambient atmospheric conditions. Crystallography. X-ray diffraction data from the crystals gave a well-defined structure (Figure 2) that supports the

Figure 3. Comparison of (a) crystallographic and (b) DFT bond distances and angles for 1III,III, 1IV,IV, 1IV,V, and 1V,V.

ligand, consistent with positive charge buildup on the metal center as well as depletion of localized negative charge at the alkoxide group, given the 2.89 Å O···F separation. Also, as in our prior report of an Ir(V) mononuclear complex,3f we see incorporation of CCl4 into the lattice which may well help crystal growth. The trends in bond lengths and angles across the four oxidation states of 1 are shown in Figure 3. Bonds between Ir and anionic alkoxo- and oxo- ligands shorten as the Ir oxidation state increases from (III,III) to (V,V), as expected for metalcentered oxidation. In contrast, the Ir-pyridine bond lengths remain approximately constant on oxidation. This is likely because the neutral pyridine ligand is less affected by change in oxidation state. The Ir−O−Ir bond angle is bent in the (III,III) state but approaches linearity on oxidation to the (V, V) state, which may result from increased Ir−O−Ir π-bonding in the higher oxidation state complexes. The slight bending of the Ir− O−Ir angle on oxidation from 1IV,IV to 1V,V can perhaps be attributed to crystal packing effects. Computational Modeling. The geometric and electronic structures of 1V,V were fully optimized by means of DFT calculations (Figure 4), which converged into a ground state with singlet multiplicity. The optimized geometry is in good agreement with the crystal structure, with a low RMS deviation of 0.022 Å for the metal−ligand bond distances. Despite the high oxidation state, the main structural features of the system are preserved, including the μ-oxo bridge, which at 172.2° (169.3° in the crystal) is almost linear, with symmetric Ir−O Figure 2. Top: Thermal ellipsoid diagram of the crystal structure of 1V,V at 30% probability level. Hydrogen atoms have been omitted for clarity. Bottom: Space filling model of 1V,V.

proposed Ir(V,V) assignment. The structure shows the same connectivity and conformation as for the 1IV,V and 1IV,IV states, while there are now two PF6− counterions per dimer molecule, as expected for a dicationic complex. Furthermore, Ir−O bond lengths are shorter than those in the Ir(IV,V) state, by an average of 0.037 Å (Figure 3 and SI Table S1), consistent with metal-centered oxidation. The packing view (Figure 2) also shows that the PF6− counterions are tightly nestled into the groove between the pyridine and alkoxide groups of each

Figure 4. DFT-optimized structure (left) and LUMO (right) of 1V,V. Bond distances are given in Å. Hydrogen atoms were removed for clarity. C

DOI: 10.1021/acs.inorgchem.8b00757 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry distances (1.86 Å). The κ1-Cl and κ2-pyalk coordination environments are also preserved, with d(Ir−Cl) = 2.39 Å and d(Ir−O/N) = 1.94/2.04 Å (cis-μ-O) and 1.89/2.05 Å (trans-μO). In line with the high oxidation state of 1V,V, the Ir−O bond distances are significantly shorter than in the reduced complexes (Figure 3). The oxidation takes place at the bimetallic core, as shown by the localization of the LUMO upon the Ir−O−Ir moiety, which has a charge of +2.24 e. This is compensated by strong electron-donation from the pyalk ligand,13 as shown by the dramatic change of its charge, i.e., from −1.00 (free ligand) to +0.20 e (Ir-bound ligand in 1V,V), consistent with the close pyalk-PF6− contacts seen in the crystal structure (vide supra; Figure 2). In contrast, the pyalk charge is 0.00, −0.19 and −0.34 e in 1IV,V, 1IV,IV, and 1III,III, respectively. This wide range of charges is consistent with NBO analysis showing that the strongest ligand−metal π interaction is pO(pyalk) → dIr. The geometry optimization of the open-shell states of 1V,V yielded higher energies for both the triplet and quintet states, which are 6.2 and 10.8 kcal mol−1 above the singlet, respectively. This is consistent with the antibonding π* character of the LUMO in the singlet ground state, which is an out of phase combination of the metal d and oxygen p orbitals across the Ir−O−Ir bridge. Further, the triplet and quintet geometries yielded larger RMS deviations relative to the crystal structure (0.026 and 0.042 Å, respectively). Spectroscopic Characterization. X-ray photoelectron spectroscopy (XPS) of crushed crystals of [1V,V](PF6)2 show uncharacteristically broad Ir 4f peaks, which we suspect arise from partial reduction to the Ir(IV,V) state by free electrons, as previously observed for the 1IV,V state.10 Deconvoluting the spectrum based on the known peak positions for 1IV,V (Figure 5) supports the presence of a new set of peaks with binding

energies that are 1.1 eV higher, indicating a metal-centered oxidation. This binding energy rise is similar to that observed between 1IV,IV and 1IV,V; neither oxidation involves any change in the ligand set, whereas the difference between 1III,III and 1IV,IV is much smaller no doubt due to protonation (Table 1). Table 1. Binding Energies for the Ir-4f Region of All the Four Oxidation States of 1 complex III,III

1 1IV,IV 1IV,V 1V,V

binding energy (eV) 62.7, 63.1, 64.3, 65.4,

65.7 66.1 67.3 68.4

Figure 6. UV−visible absorption spectra of the four oxidation states of 1 in dichloromethane.

The UV−visible spectrum of 1V,V (Figure 6) was modeled by means of TD-DFT calculations including solvation by dichloromethane. Figure 7 shows the main electronic

Figure 5. Top: The deconvoluted Ir-4f region of the XPS spectrum of 1V,V. Bottom: Comparison of the Ir-4f region for the XPS spectrum of 1III,III, 1IV,IV, 1IV,V, and 1V,V (deconvolution) showing the increasing binding energy for higher oxidation states. The lower binding energy peak for all the four states has been highlighted for clarity.

Figure 7. UV−visible experimental absorbance (blue band) and TDDFT excitations (1−5 red lines; more details in Figure S5) with their oscillator strengths ( f) for complex 1V,V. For the sake of clarity, only excitations with f > 0.05 were included. D

DOI: 10.1021/acs.inorgchem.8b00757 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

shown by CAS calculations,8,9b but single-state DFT calculations indicate that the major absorptions are of d−d type, whereas LMCT bands are blue-shifted into the near-UV. In 1III,III (Figure 6), all absorptions are in the UV and consist only of metal-to-ligand charge transfers (MLCT), as the metals’ t2g levels are fully occupied. The trend of transitioning from d−d to LMCT character on oxidation from 1IV,IV to 1V,V is consistent with metal-centered charge depletion. The former state displays abnormally electron-rich Ir centers (as evidenced by the displacement of LMCT’s into the UV, which are typically in the visible region for Ir(IV) complexes), likely due to the π-interactions with the oxo. As the dimer is sequentially oxidized, the red-shifting of LMCT features signals a decreasing ligand-to-metal energy gap, indicating that the oxidation is metal-centered. Pure d−d excitation is also noted in the 1IV,V state, and is likewise redshifted to over 900 nm, but comes with a significantly lower absorptivity.10 Characterization of a Mononuclear Ir(V) Complex. The bulk electrolysis method used to prepare and crystallize 1V,V should be broadly applicable to other highly oxidized complexes. Indeed, we were able to apply this method to another previously reported complex,14a trans-IrIV(pyalk)2Cl2 (2IV, Figure 9), which also resists further oxidation with chemical oxidants. Broad-window cyclic voltammetry measurements revealed a second reversible redox couple at 1.7 V vs NHE (Figure 9), so we performed bulk electrolysis on the

transitions, for which the oscillator strength (f) is larger than 0.05, together with the experimental absorbance. The presence of multiple intense transitions, i.e., 1−5, scattered over the 400−600 nm range, accounts for the breadth of the absorption band. On the short wavelength side of the spectrum, transitions 1−3 have a strong ligand-to-metal charge transfer (LMCT) character, involving electron excitations from different nπ(pyalk,Cl) orbitals below the HOMO level to the π*(Ir− O−Ir) LUMO (Figure S5). On the long-wavelength side, the electronic transitions (4 and 5) are clearly different. Both involve orbitals closer to the HOMO−LUMO levels and, though they also yield charge transfer to the bimetallic core, electrons are promoted from orbitals in which the π contribution of the ligand (mainly Cl) mixes with that of the Ir−O−Ir moiety. This accounts for the stronger intensities of these transitions, with f = 0.2641 (4, at λexc = 507 nm) and 0.1024 (5, at λexc = 553 nm). The symmetry and spatial localization of the orbitals involved suggest that transitions 4 and 5 have a hybrid nature mixing LMCT and Laporte-allowed, conjugated d−d character. Comparing the spectral and electronic features of 1V,V with the lower oxidation states reveals a clear trend. The TD-DFT spectrum of 1IV,V has fewer transitions of high intensity, in line with its narrower absorption band (Figure 6). The most intense transition in 1IV,V ( f = 0.2827) takes place at a shorter wavelength (λexc = 468 nm) relative to that in 1V,V (λexc = 507 nm). The difference, Δλexc = 39 nm, is in agreement with the blue-shift observed in the experimental UV−visible spectra (ΔλMax = 68 nm). In addition, the transition in 1IV,V possesses significantly greater conjugated d−d character (Figure 8). The complex shape of the 1IV,IV spectrum can be due to the multireference character of the ground state of this species

Figure 8. Natural transition orbitals (NTOs at an isovalue of 0.05 (ea0−3)1/2) associated with the strongest electronic excitations of complex 1 in the (III,III), (IV,IV), (IV,V), and (V,V) oxidation states, predicted at the TD-DFT ωB97xd/LANL2TZ(f),6-311+G** level. Hydrogen atoms are removed for clarity.

Figure 9. (a) Cyclic voltammogram of 2 (∼0.2 mM) in ACN solution (scan rate 100 mV/s) with CsPF6 electrolyte. (b) Schematic of bulk electrolysis preparation of 2V. (c) Crystal structure of 2V. E

DOI: 10.1021/acs.inorgchem.8b00757 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

calculations. The energy-minimum nature of the optimized geometries was confirmed by calculating the vibrational frequencies, which also yielded the thermodynamic parameters, including the zero-point, thermal, and entropy energies. Electronic structures and energies were refined by means of single-point PBE0/LANL2TZ(f),6311+G** calculations19 on the optimized geometries. In all calculations, dispersion forces were modeled with the Grimme’s GD3BJ approach20 and numerical accuracy was increased by using the ultrafine (99,590) pruned grid. The stability of the converged electron densities was verified by single-point calculations removing orbital constraints. Local charges and donor−acceptor interactions were quantified by means of natural bond orbital analysis, using the NBO6 program.21 The TDDFT calculations included SMD(dichloromethane) solvation and were solved for 40 states with the ωB97xd functional.22 The canonical molecular orbitals associated with the electron excitations were located by computing natural transition orbitals.23 Physical Methods. Electrochemistry. Bulk electrolysis experiments were performed using a standard three-electrode setup carried out on a Princeton Applied Research VersaSTAT-4 or a Pine AFCBP1 bipotentiostat. Experiments were performed using a platinum mesh working electrode, an Ag wire pseudoreference electrode, and a platinum wire counter electrode. The counter electrode was placed in a small fritted tube with the same electrolyte solution as the main chamber. Experiments were performed in a dry glovebox under a N2 atmosphere. X-ray Photoelectron Spectroscopy (XPS). The XPS spectrum was collected using a monochromatic 1486.7 eV Al Ka X-ray source on a PHI VersaProbe II X-ray Photoelectron Spectrometer with a 0.47 eV system resolution. The energy scale was calibrated using Cu 2p3/2 (932.67 eV) and Au 4f7/2 (84.00 eV) peaks on a clean copper plate and a clean gold foil. The sample consisted of crushed crystalline 1V,V placed on a silicon wafer. Because of the moisture sensitivity of 1V,V, the sample was loaded onto the silicon wafer in a dry glovebox and transferred into the XPS instrument without exposure to the atmosphere. Samples prepared by drop casting solutions of 1V,V gave signals corresponding to 1IV,V formed by reduction of 1V,V. The shifts for the data collected were calibrated using the Si 2p peak (99.3 eV) from the Si wafer. Gas Chromatography. This was performed using a Varian 450 GC instrument, with He as the carrier gas. Synthesis. General. Unless otherwise specified, solvents and reagents were purchased from commercial sources and used as received. Honeywell CHROMASOLV LC-MS grade dry acetonitrile was used for the bulk electrolysis experiments. Compounds 1IV,IV and 2IV were prepared as described previously.8,14a All manipulations on 1V,V and 2V were carried out in a dry glovebox under N2 atmosphere. Warning: CCl4 is highly toxic and should be handled with caution and proper PPE. [1V,V][PF6]2-(CCl4)2. Due to the moisture sensitivity of 1V,V, all manipulations were carried out in a dry glovebox under N 2 atmosphere. A sample of 5 mg of 1IV,IV was dissolved in a saturated solution of CsPF6 in dry acetonitrile. Bulk electrolysis was performed on a stirred solution in a 5-dram vial with an applied potential of approximately 1.8 V vs a Ag wire pseudoreference electrode. The setup consisted of a Pt gauze working electrode and a Pt wire counter electrode, with the counter electrode placed in a small fritted tube filled with a saturated CsPF6 solution in dry acetonitrile. Over the course of 10−15 min of electrolysis, the solution gradually changed color from blue-green to purple followed by a second change to royal blue (see Figure S1). Electrolysis was stopped, the solution was agitated using a pipet to dissolve the remaining 1IV,IV starting material, and electrolysis was resumed until the solution was royal blue in color and a low, steady current was reached. The electrodes were removed and 1V,V was concentrated in vacuo by placing the vial into a large glass container with a Schlenk outlet and evacuating the container. CCl4 was then added to the outer container and the container was placed under static vacuum and stored at −20 °C. Crystals of 1V,V were obtained as dark blue blocks upon vapor diffusion by CCl4. Note: Complex 1V,V is highly sensitive to reduction, and introduction of slight impurities by

complex in the same manner as for 1. The solution changed in color from magenta to deep violet, and the product was crystallized successfully as before. The X-ray structure (Figure 9) indicates a cationic complex with one PF6− counterion. The pattern of bond length changes compared to the Ir(IV) state is again characteristic of metal-centered oxidation, and is very similar to the pattern in Figure 3: Ir−O bonds have contracted from 1.943 to 1.888 Å, Ir−Cl bonds have slightly contracted from 2.347 to 2.324 Å, and Ir−N bonds show no significant change. Based on this information, we assign this new species as the Ir(V) state of 2, [2V]PF6. As in the case of 1V,V, this compound is a powerful one-electron oxidant, as well as being sensitive to water.

■

CONCLUSIONS We have prepared and characterized an Ir(V,V) mono-μ-oxo dimer as well as an Ir(V) monomer using a novel bulk electrolysis technique. Both compounds are more highly oxidized and reactive forms of complexes we have previously reported,10,14a and which could not be prepared by chemical oxidation. The former complex appears to be the first isolated example of an Ir(V,V) oxo-bridged dimer, a species proposed over 30 years ago by Sykes et al.,15 as well as proposed as an intermediate in Ir water oxidation catalysis.11a,c With this addition, we have now characterized the dimer in four oxidation states, (III,III), (IV,IV), (IV,V), and (V,V) that show no observable degradation on redox cycling, demonstrating the extensive redox chemistry possible for Ir oxo dimers. The metal-centered character of the oxidation is supported by DFT and TD-DFT calculations, which also showed that the intense colors of these complexes are due to mixed LMCT and Laporte-allowed d−d electron excitations within the visible range of the spectrum. Our electrochemical method allows for the in situ formation and crystallization of highly oxidized complexes, which are too reactive to be prepared or handled by traditional means. Our approach combines the traditional bulk electrolysis and crystallization techniques to stabilize the reactive oxidized species by eliminating commonly used organic substances prone to oxidation. Specifically, we utilize acetonitrile with CsPF6 as an inert inorganic electrolyte, CCl4 as an inert precipitating solvent, and low-temperature low-pressure vapor diffusion conditions for maintaining the oxidized state during crystallization. We believe that this method will prove applicable to a wide variety of similar species and may lead to the isolation and characterization of other novel reactive compounds. Due to their structural stability, the reversible, one-electron transfer nature of their redox chemistry, the very high reduction potentials (1.9 and 1.7 V vs NHE), and their solubility in organic solvents, both complexes have possible application as powerful, organic-phase, one-electron oxidants. These could be used in applications where direct electrolysis is not feasible, such as for studying transient species.

■

METHODS

Computational Details. Calculations were carried out with the Gaussian0916 software package. Optimizations were performed without any constraint at the DFT PBE/LANL2TZ(f),6-311+G** level.17 Geometries were optimized either in gas phase or solution (dichloromethane with the continuum SMD model).18 The former were compared to the X-ray crystal structures, whereas the latter were used to model the UV−visible spectra by means of TD-DFT F

DOI: 10.1021/acs.inorgchem.8b00757 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry transferring to a new vial or adding fresh solvent led to significant reduction. [2V]PF6. Complex 2V was prepared and crystallized using the same method employed for 1V,V. A sample of 5 mg of 2IV was dissolved in a saturated solution of CsPF6 in dry acetonitrile in a short 5-dram vial. Bulk electrolysis was performed using moisture free conditions analogous to those used for 1V,V. A color change from magenta to violet was observed upon oxidation. The resulting solution containing 2V was concentrated in vacuo and crystallized under reduced pressure by vapor diffusion with CCl4 at −20 °C.

■

thank Prof. James Mayer, Yale for providing us access to the glovebox in his laboratory.

■

(1) (a) Lewis, N. S.; Nocera, D. G. Powering the planet: Chemical challenges in solar energy utilization. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 15729. (b) Blankenship, R. E.; Tiede, D. M.; Barber, J.; Brudvig, G. W.; Fleming, G.; Ghirardi, M.; Gunner, M. R.; Junge, W.; Kramer, D. M.; Melis, A.; Moore, T. A.; Moser, C. C.; Nocera, D. G.; Nozik, A. J.; Ort, D. R.; Parson, W. W.; Prince, R. C.; Sayre, R. T. Comparing photosynthetic and photovoltaic efficiencies and recognizing the potential for improvement. Science 2011, 332, 805. (c) Blakemore, J. D.; Crabtree, R. H.; Brudvig, G. W. Molecular catalysts for water oxidation. Chem. Rev. 2015, 115, 12974. (d) Hunter, B. M.; Gray, H. B.; Muller, A. M. Earth-abundant heterogeneous water oxidation catalysts. Chem. Rev. 2016, 116, 14120. (e) Gust, D.; Moore, T. A.; Moore, A. L. Solar fuels via artificial photosynthesis. Acc. Chem. Res. 2009, 42, 1890. (f) Kärkäs, M. D.; Verho, O.; Johnston, E. V.; Åkermark, B. Artificial photosynthesis: Molecular systems for catalytic water oxidation. Chem. Rev. 2014, 114, 11863. (g) Meyer, T. J.; Sheridan, M. V.; Sherman, B. D. Mechanisms of molecular water oxidation in solution and on oxide surfaces. Chem. Soc. Rev. 2017, 46, 6148. (h) Parent, A. R.; Sakai, K. Progress in base-metal water oxidation catalysis. ChemSusChem 2014, 7, 2070. (i) Kärkäs, M. D.; Åkermark, B. Water oxidation using earth-abundant transition metal catalysts: opportunities and challenges. Dalton Trans. 2016, 45, 14421. (j) Singh, A.; Spiccia, L. Water oxidation catalysts based on abundant 1st row transition metals. Coord. Chem. Rev. 2013, 257, 2607. (k) Fisher, K. J.; Materna, K. L.; Mercado, B. Q.; Crabtree, R. H.; Brudvig, G. W. Electrocatalytic water oxidation by a copper(II) complex of an oxidation-resistant ligand. ACS Catal. 2017, 7, 3384. (2) (a) Labinger, J. A.; Bercaw, J. E. Understanding and exploiting CH bond activation. Nature 2002, 417, 507. (b) A recent issue of Chem. Rev. was devoted to C-H activation. See: Crabtree, R. H.; Lei, A. Introduction: C-H activation. Chem. Rev. 2017, 117, 8481 and articles cited therein. (3) (a) Wang, G. J.; Zhou, M. F.; Goettel, J. T.; Schrobilgen, G. J.; Su, J.; Li, J.; Schloder, T.; Riedel, S. Identification of an iridium-containing compound with a formal oxidation state of IX. Nature 2014, 514, 475. (b) Gulliver, D. J.; Levason, W. The chemistry of ruthenium, osmium, rhodium, iridium, palladium and platinum in the higher oxidation states. Coord. Chem. Rev. 1982, 46, 1. (c) Jørgensen, C. K. Oxidation Numbers and Oxidation States; Springer: Heidelberg, Germany, 1969. (d) Hickman, A. J.; Sanford, M. S. High-valent organometallic copper and palladium in catalysis. Nature 2012, 484, 177. (e) Topczewski, J. J.; Sanford, M. S. Carbon−hydrogen (C−H) bond activation at PdIV: a Frontier in C−H functionalization catalysis. Chem. Sci. 2015, 6, 70. (f) Sharninghausen, L. S.; Sinha, S. B.; Shopov, D. Y.; Mercado, B. Q.; Balcells, D.; Brudvig, G. W.; Crabtree, R. H. Synthesis and characterization of iridium(V) coordination complexes with an N,Odonor organic ligand. Angew. Chem., Int. Ed. 2017, 56, 13047. (4) (a) Anson, F. C.; Christie, J. A.; Collins, T. J.; Coots, R. J.; Furutani, T. T.; Gipson, S. L.; Keech, J. T.; Krafft, T. E.; Santarsiero, B. D.; Spies, G. H. The design of multianionic chelating ligands for the production of inorganic oxidizing agents. Osmium coordination chemistry that provides stable potent oxidizing agents and stable potent reducing agents. J. Am. Chem. Soc. 1984, 106, 4460. (b) Collins, T. J. Designing ligands for oxidizing complexes. Acc. Chem. Res. 1994, 27, 279. (c) Warner, G.; Mills, M.; Enslin, C.; Pattanayak, S.; Panda, C.; Sen Gupta, S.; Ryabov, A. D.; Collins, T. J.; Panda, T. K. Reactivity and operational stability of N-tailed TAMLs through kinetic studies of the catalyzed oxidation of orange II by H2O2: synthesis and X-ray structure of an N-phenyl TAML. Chem. - Eur. J. 2015, 21, 6226. (d) Collins, T. J.; Ryabov, A. D. Targeting of high-valent iron-TAML activators at hydrocarbons and beyond. Chem. Rev. 2017, 117, 9140. (5) Michaelos, T. K.; Shopov, D. Y.; Sinha, S. B.; Sharninghausen, L. S.; Fisher, K. J.; Lant, H. M. C.; Crabtree, R. H.; Brudvig, G. W. A pyridine alkoxide chelate ligand that promotes both unusually high

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.8b00757. Spectroelectrochemistry, experimental setup for bulk electrolysis and crystallization, table for comparison of crystallographic bond lengths and angles, crystallographic and computational details (PDF) Accession Codes

CCDC 1589397−1589398 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. *E-mail: [email protected]. ORCID

Liam S. Sharninghausen: 0000-0002-2249-1010 Shashi Bhushan Sinha: 0000-0002-3212-7570 David Balcells: 0000-0002-3389-0543 Gary W. Brudvig: 0000-0002-7040-1892 Robert H. Crabtree: 0000-0002-6639-8707 Author Contributions #

Dimitar Y. Shopov, Liam S. Sharninghausen, and Shashi Bhushan Sinha contrubuted equally to this work. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work (D.Y.S., L.S.S., S.B.S.) was supported the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences under Award Number DE-SC0001059 as part of the Argonne-Northwestern Solar Energy Research (ANSER) Energy Frontier Research Center (spectroscopy and characterization) and under Award Number DEFG02-07ER15909 (synthesis). D.B. was supported by the Research Council of Norway through a Centre of Excellence grant (Grant No. 179568/V30) and the Norwegian Supercomputing Program (NOTUR) through a grant for computing time (Grant No. NN4654K). D.B. also acknowledges the EU REA for a Marie Curie Fellowship (Grant CompuWOC/618303). Dr. Min Li and the Yale Materials Characterization Core are acknowledged for collection of data by X-ray Photoelectron Spectroscopy. We G

DOI: 10.1021/acs.inorgchem.8b00757 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry oxidation Ssates and water-oxidation catalysis. Acc. Chem. Res. 2017, 50, 952. (6) Sinha, S. B.; Shopov, D. Y.; Sharninghausen, L. S.; Vinyard, D. J.; Mercado, B. Q.; Brudvig, G. W.; Crabtree, R. H. A stable coordination complex of Rh(IV) in an N,O-donor environment. J. Am. Chem. Soc. 2015, 137, 15692. (7) (a) Thomsen, J. M.; Sheehan, S. W.; Hashmi, S. M.; Campos, J.; Hintermair, U.; Crabtree, R. H.; Brudvig, G. W. Electrochemical Activation of Cp* Iridium complexes for electrode-driven wateroxidation catalysis. J. Am. Chem. Soc. 2014, 136, 13826. (b) Hintermair, U.; Sheehan, S. W.; Parent, A. R.; Ess, D. H.; Richens, D. T.; Vaccaro, P. H.; Brudvig, G. W.; Crabtree, R. H. Precursor transformation during molecular oxidation catalysis with organometallic iridium complexes. J. Am. Chem. Soc. 2013, 135, 10837. (c) Sheehan, S. W.; Thomsen, J. M.; Hintermair, U.; Crabtree, R. H.; Brudvig, G. W.; Schmuttenmaer, C. A. A molecular catalyst for water oxidation that binds to metal oxide surfaces. Nat. Commun. 2015, 6, 6469. (d) 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. (e) Huang, D. L.; Beltran-Suito, R.; Thomsen, J. M.; Hashmi, S. M.; Materna, K. L.; Sheehan, S. W.; Mercado, B. Q.; Brudvig, G. W.; Crabtree, R. H. New Ir bis-carbonyl precursor for water oxidation catalysis. Inorg. Chem. 2016, 55, 2427. (f) Schley, N. D.; Blakemore, J. D.; Subbaiyan, N. K.; Incarvito, C. D.; D’Souza, F.; Crabtree, R. H.; Brudvig, G. W. Distinguishing homogeneous from heterogeneous catalysis in electrode-driven water oxidation with molecular iridium complexes. J. Am. Chem. Soc. 2011, 133, 10473. (8) Sharninghausen, L. S.; Sinha, S. B.; Shopov, D. Y.; Mercado, B. Q.; Choi, B.; Roy, X.; Balcells, D.; Brudvig, G. W.; Crabtree, R. H. High oxidation state iridium mono-μ-oxo dimers related to water oxidation catalysis. J. Am. Chem. Soc. 2016, 138, 15917. (9) For previous examples of Ir(IV,IV) oxo-dimers see: (a) Chen, T. R.; Wu, F. S.; Lee, H. P.; Chen, K. H. C. Diiridium bimetallic complexes function as a redox switch to directly split carbonate into carbon monoxide and oxygen. J. Am. Chem. Soc. 2016, 138, 3643. (b) Lehman, M. C.; Pahls, D. R.; Meredith, J. M.; Sommer, R. D.; Heinekey, D. M.; Cundari, T. R.; Ison, E. A. Oxyfunctionalization with Cp*IrIII(NHC)(Me)(Cl) with O2: Identification of a rare bimetallic IrIV μ-Oxo intermediate. J. Am. Chem. Soc. 2015, 137, 3574. (c) Fortner, K. C.; Laitar, D. S.; Muldoon, J.; Pu, L.; Braun- Sand, S. B.; Wiest, O.; Brown, S. N. Ultrafast and ultraslow oxygen atom transfer reactions between late metal centers. J. Am. Chem. Soc. 2007, 129, 588. (10) Sinha, S. B.; Shopov, D. Y.; Sharninghausen, L. S.; Stein, C. J.; Mercado, B. Q.; Balcells, D.; Pederson, T. B.; Reiher, M.; Brudvig, G. W.; Crabtree, R. H. Redox activity of oxo-bridged iridium dimers in an N,O-donor environment: characterization of remarkably stable Ir(IV,V) complexes. J. Am. Chem. Soc. 2017, 139, 9672. (11) (a) Yang, K. R.; Matula, A. J.; Kwon, G.; Hong, J.; Sheehan, S. W.; Thomsen, J. M.; Brudvig, G. W.; Crabtree, R. H.; Tiede, D. M.; Chen, L. X.; Batista, V. S. Solution structures of highly active molecular Ir water-oxidation catalysts from density functional theory combined with high-energy X-ray scattering and EXAFS spectroscopy. J. Am. Chem. Soc. 2016, 138, 5511. (b) Boller, T. M.; Murphy, J. M.; Hapke, M.; Ishiyama, T.; Miyaura, N.; Hartwig, J. F. Mechanism of the mild functionalization of arenes by diboron reagents catalyzed by iridium complexes. Intermediacy and chemistry of bipyridine-ligated iridium trisboryl complexes. J. Am. Chem. Soc. 2005, 127, 14263. (c) Thomsen, J. M.; Huang, D. L.; Crabtree, R. H.; Brudvig, G. W. Iridium-based complexes for water oxidation. Dalton Trans. 2015, 44, 12452. (d) Paneque, M.; Poveda, M. L.; Salazar, V.; Taboada, S.; Carmona, E.; et al. C−H bond activation of thiophenes by Ir complexes of the hydrotris(3,5-dimethylpyrazolyl)borate ligand, TpMe2. Organometallics 1999, 18, 139. (e) Steegstra, P.; Busch, M.; Panas, I.; Ahlberg, E. Revisiting the redox properties of hydrous iridium oxide films in the context of oxygen evolution. J. Phys. Chem. C 2013, 117, 20975. (f) Balcells, D. Insight into metal-catalyzed water oxidation from a DFT perspective. Adv. Organomet. Chem. 2016, 65, 115.

(12) Parent, A. R.; Crabtree, R. H.; Brudvig, G. W. Comparison of primary oxidants for water-oxidation catalysis. Chem. Soc. Rev. 2013, 42, 2247. (13) For the sake of comparison across the oxidation series, this refers to the pyalk ligand trans to the μ-oxo ligand. (14) (a) Shopov, D. Y.; Rudshteyn, B.; Campos, J.; Batista, V. S.; Crabtree, R. H.; Brudvig, G. W. A full set of iridium(IV) pyridinealkoxide stereoisomers: highly geometry-dependent redox properties. Chem. Sci. 2017, 8, 1642. (b) Shopov, D. Y.; Rudshteyn, B.; Campos, J.; Batista, V. S.; Crabtree, R. H.; Brudvig, G. W. Stable iridium(IV) complexes of an oxidation-resistant pyridine-alkoxide ligand: highly divergent redox properties depending on the isomeric form adopted. J. Am. Chem. Soc. 2015, 137, 7243. (15) (a) Castillo-Blum, S. E.; Richens, D. T.; Sykes, A. G. New aqua ions of iridium in oxidation states (III), (IV), and (V). J. Chem. Soc., Chem. Commun. 1986, 14, 1120. (b) Castillo-Blum, S. E.; Richens, D. T.; Sykes, A. G. Oxidation of hexaaquairidium(III) and related studies: preparation and properties of iridium(III), iridium(IV), and iridium(V) dimers as aqua ions. Inorg. Chem. 1989, 28, 954. (16) 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.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision D.01; Gaussian, Inc.: Wallingford, CT, 2013. (17) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77, 3865. (18) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. Universal solvation model based on solute electron density and on a continuum model of the solvent defined by the bulk dielectric constant and atomic surface tensions. J. Phys. Chem. B 2009, 113, 6378. (19) Adamo, C.; Barone, V. Toward reliable density functional methods without adjustable parameters: The PBE0 model. J. Chem. Phys. 1999, 110, 6158. (20) Grimme, S.; Ehrlich, S.; Goerigk, L. Effect of the damping function in dispersion corrected density functional theory. J. Comput. Chem. 2011, 32, 1456. (21) Glendening, E. D.; Landis, C. R.; Weinhold, F. J. NBO 6.0: Natural bond orbital analysis program. J. Comput. Chem. 2013, 34, 1429. (22) Chai, J.-D.; Head-Gordon, M. Long-range corrected hybrid density functionals with damped atom-atom dispersion corrections. Phys. Chem. Chem. Phys. 2008, 10, 6615. (23) Martin, R. L. Natural transition orbitals. J. Chem. Phys. 2003, 118, 4775.

H

DOI: 10.1021/acs.inorgchem.8b00757 Inorg. Chem. XXXX, XXX, XXX−XXX