New Ir Bis-Carbonyl Precursor for Water Oxidation Catalysis

22 Feb 2016 - Liam S. Sharninghausen , Shashi Bhushan Sinha , Dimitar Y. Shopov , Bonnie Choi , Brandon Q. Mercado , Xavier Roy , David Balcells , Gar...
1 downloads 0 Views 2MB Size
Article pubs.acs.org/IC

New Ir Bis-Carbonyl Precursor for Water Oxidation Catalysis Daria L. Huang,§ Rodrigo Beltrán-Suito,§ Julianne M. Thomsen,§ Sara M. Hashmi,† Kelly L. Materna,§ Stafford W. Sheehan,‡ Brandon Q. Mercado,§ Gary W. Brudvig,§ and Robert H. Crabtree*,§ †

Department of Chemical and Environmental Engineering, Yale University, 9 Hillhouse Avenue, New Haven, Connecticut 06520, United States ‡ Catalytic Innovations LLC, 70 Crandall Road, P.O. Box 356, Adamsville, Rhode Island 02801, United States § Department of Chemistry, Yale University, 225 Prospect Street, New Haven, Connecticut 06520, United States S Supporting Information *

ABSTRACT: This paper introduces IrI(CO)2(pyalc) (pyalc = (2-pyridyl)-2-propanoate) as an atom-efficient precursor for Irbased homogeneous oxidation catalysis. This compound was chosen to simplify analysis of the water oxidation catalyst species formed by the previously reported Cp*IrIII(pyalc)OH water oxidation precatalyst. Here, we present a comparative study on the chemical and catalytic properties of these two precursors. Previous studies show that oxidative activation of Cp*Ir-based precursors with NaIO4 results in formation of a blue IrIV species. This activation is concomitant with the loss of the placeholder Cp* ligand which oxidatively degrades to form acetic acid, iodate, and other obligatory byproducts. The activation process requires substantial amounts of primary oxidant, and the degradation products complicate analysis of the resulting IrIV species. The species formed from oxidation of the Ir(CO)2(pyalc) precursor, on the other hand, lacks these degradation products (the CO ligands are easily lost upon oxidation) which allows for more detailed examination of the resulting Ir(pyalc) active species both catalytically and spectroscopically, although complete structural analysis is still elusive. Once Ir(CO)2(pyalc) is activated, the system requires acetic acid or acetate to prevent the formation of nanoparticles. Investigation of the activated bis-carbonyl complex also suggests several Ir(pyalc) isomers may exist in solution. By 1H NMR, activated Ir(CO)2(pyalc) has fewer isomers than activated Cp*Ir complexes, allowing for advanced characterization. Future research in this direction is expected to contribute to a better structural understanding of the active species. A diol crystallization agent was needed for the structure determination of 3.



INTRODUCTION Since our initial report in 2009,1 the Cp*IrIII series (Cp* = pentamethyl-cyclopentadienyl, C5Me5−) has been thoroughly investigated by many groups for efficient water oxidation and CH activation catalysis.2−21 Better understanding of the mechanistic details of these highly active precatalysts is desirable as these oxidation reactions are very relevant to sustainable energy efforts22−31 and potentially new synthetic methodologies.32,33 Upon mixing the Cp*Ir(pyalc)OH (pyalc = (2-pyridyl)-2-propanoate) (1, Figure 1) with aqueous oxidant, the Cp* is oxidatively degraded12,21,34,35 during

activation of the precatalyst. This releases ca. 1.8 equiv of acetic acid (AcOH) and gives a blue solution that acts as a resting state for catalysis. Unfortunately, oxidized Ir species resulting from activation of Cp*Ir complexes have never been successfully crystallized for structural work, but a joint experimental−computational study led to the suggestion that the suggested active species from 1 is the IrIV-μ-oxo dimer, 2.7 Although 1 is a competent precatalyst for electrochemical and chemical water oxidation, unambiguous characterization of 2 has been problematic.7,36 Our working hypothesis is that 2 is not the only Ir species in solution, but rather that a mixture of isomers results from the oxidation of 1. Our group has separated, isolated, and crystallographically characterized a full series of [IrIV(pyalc)2Cl2] isomers that do not interconvert easily as part of a continuing study in our lab.37 These monomeric IrIV complexes related to 2 give kinetically stable isomers. In addition to the possibility of isomeric forms for 2, it is also likely that when 1 is activated, the degradation products from the Cp* (primarily acetic acid and iodate) can also bind to the Ir in various ways.19,38 If many of these structures are

Figure 1. Cp*IrIII(pyalc)OH precursor (1), suggested IrIV-oxo dimer (2), and the new [IrI(CO)2(pyalc)] precursor (3). © XXXX American Chemical Society

Received: December 7, 2015

A

DOI: 10.1021/acs.inorgchem.5b02809 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

clusters,44 failed. We were eventually able to crystallize [Ir(CO)2(pyalc)] by slow evaporation at −20 °C for a month. However, the crystals obtained in this way were of poor quality and did not give high quality X-ray diffraction data (see Supporting Information). To circumvent this problem and improve the crystal quality, we turned to cocrystallization reagents. Since the pyalc O atom is strongly basic, it should be able to form strong H bonds with proton donors, a factor that might promote cocrystallization. Unfortunately, while H bond acceptors like triphenylphosphine oxide and pyridine derivatives are widely applicable,45−49 the most common H bond donors, carboxylic acid, amines, and amides, all led to reaction with 3. Instead, we found that short-chain diols HO(CH2)nOH (n = 3−5) and benzenedimethanols, until now unrecognized as small molecule cocrystallization agents, gave the largest cocrystals. Solutions having a 2:1 ratio of catalyst:diol in THF were allowed to evaporate at 4 °C for 1 week to give large orange-red crystals. The crystal structures could now be easily obtained. Crystal packing data showed that the diol OH groups indeed form strong H bonds (dO···O = 2.477(3) Å) with the basic pyalc O atoms of [Ir(CO)2(pyalc)] leading to tight and continuous packing (Figure 2). The data also showed that the

indeed present, it is not surprising that all crystallization attempts on 2 have thus far been unsuccessful. Additionally, the presence of the obligatory degradation products may also affect catalysis: excess iodate has been shown39 to inhibit electrochemical water oxidation (WO) catalysis, and AcOH has been similarly suggested16 to influence CH oxidation catalysis. These factors that complicate analytical study of the Cp* precursor led us to look for Ir(pyalc) derivatives that could be oxidatively activated without needing a large excess of oxidant and that would not form potential ligands such as acetate. An IrI dicarbonyl, 3, seemed best adapted for this role, since upon oxidation, the metal would not be able to retain the CO ligands. We now report the synthesis of a novel precursor [IrI(CO)2(pyalc)] (3) that avoids the presence of the Cp* and thus all of its oxidative degradation products. Upon activation of 3, the CO ligands are indeed quickly lost, and a blue solution is formed, thus eliminating the obligatory production of AcOH in the Cp* case and reducing the amount of periodate required to oxidize the placeholder ligand. Oxidative activation of 3 now allows us to examine the effect of AcOH on the activated blue species. The swifter activation of 3 relative to 1 is consistent with our expectation that CO ligands would depart more easily. Consistent with prior results that show inhibition by iodate,39 the resulting blue solution is also more active as a water oxidation catalyst, and NMR data also indicate that fewer isomers are produced from 3 than from 1. Although all crystallization attempts of the blue active species have so far failed, we were able to isolate and crystallographically characterize [Ir(pyalc)2I2] from the blue solution of activated 3, the iodide deriving from further reaction of the iodate. This is the first report of any IrIV compound crystallized from a solution of activated water oxidation catalyst precursor, although it is a degradation product incapable of water oxidation catalysis. The availability of a catalytically active blue solution from 3 permits us to determine the effect of added AcOH on the catalytic activities of the solution. The aim of this study is to present a comparison between the chemical and catalytic properties of these two precursors independent of structural analysis of the active species, which is under active investigation separately.

Figure 2. ORTEP crystal structure of 3 cocrystallized with 1,4butanediol with 50% thermal ellipsoids. Numbering is based on the asymmetric unit.

packing of the cocrystal was different from that of the crystals obtained without cocrystallization reagents. This difference is reflected in the physical properties: pure 3 is pale yellow but becomes orange or red when cocrystallized with the diols. While previously reported H donor cocrystallization reagents were often specifically designed with a target in mind,50,51 diols are expected to be applicable to a wider class of organometallic compounds and could prove generally useful. Activation of 3 and Confirmation of a Homogeneous Species. Since 3 was designed to resemble the Cp*Ir(pyalc)OH analogue, we set out to test if it activated to give the same oxidation product, 2. Addition of 3 to an aqueous NaIO4 solution quickly produced a deep blue solution, with the color being similar to 2 (Figure 3). Because the blue solutions from 1 and 3 have not been unambiguously characterized and confirmed to be the same species, we refer to the blue solution of activated 3 as 2′. Overall, 2′ has a much higher absorbance than 2. We believe that this is because 2′ has more IrIV species in solution than 2; the activation of 1 to 2 degrades Cp*, which produces acetic acid as well as methanol and formate. The latter two compounds are a powerful reductant and may reduce some of the active IrIV species to IrIII,39 which would diminish the absorption at ∼600 nm. Additionally, UV−vis time-course measurements show that 3 activates much more quickly than 1 (Figure 4). Compound 3 is likely oxidized through a much simpler route than has been documented for the Cp* analogue,



RESULTS Synthesis and Characterization of [Ir(CO)2(pyalc)] Precursor, 3. The pyalc ligand was kept as the chelate ligand because, with Cp* precatalysts, the precursor with pyalc (1) was the most active and robust for water oxidation. The [Ir(CO)2(pyalc)] precursor (3) was prepared by an adaptation of a previously published procedure that is applicable to a range of other [IrI(CO)2Ln]+ compounds.40−42 In this synthesis, [IrI(cod)Ln]+ (cod = 1,5-cyclooctadiene) is first prepared and subsequently treated with CO to give the dicarbonyl precatalyst [IrI(CO)2Ln]+. In the first stage, we successfully obtained the known [Ir(cod)(pyalc)] precursor by a new method. [Ir(cod)Cl]2 was dissolved in dichloromethane and added to an aqueous solution of sodium tert-butoxide and the pyalc ligand. The resulting [Ir(cod)(pyalc)] was isolated and then converted to the desired [Ir(CO)2(pyalc)] material by treatment with CO gas. The product obtained was a yellow powder that is benchstable as a solid for several weeks. Several methods were utilized in attempts to obtain suitable crystals of 3 for X-ray measurements, but standard methods proved unavailing. Even attempts using gels,43 a technique successfully used by others in our lab to crystallize multi-iridium B

DOI: 10.1021/acs.inorgchem.5b02809 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 5. UV−vis spectrum of 2′ and equimolar equivalents of iodate and acetic acid. Inset shows a closer view of the red-shift as more acetate and iodate are added.

Figure 3. UV−vis spectrum of 1 mM 1 (black) with 100 mM NaIO4 in H2O and 1 mM of 3 (red) with 100 mM NaIO4 in H2O, from 350 to 800 nm.

To ensure that 2′ remained homogeneous, we employed dynamic light scattering (DLS) and transmission electron microscope (TEM) methods. The DLS measurements revealed an interesting trend: when 3 was injected into aqueous NaIO4, small, dilute nanoparticles (NPs) formed after ∼30 min (Figure 6). However, when small aliquots of acetic acid were added, even as low as 50 equiv relative to the precatalyst, NP formation was delayed by 4 h. With 150 equiv of AcOH, no nanoparticles are observed even after 18 h. This stabilization effect is likely not just due to pH effects: when trifluoroacetic acid was added (150 equiv), NPs still formed after 11 h (Figure S3). Due to the ∼2 nm detection limit of DLS under the given conditions, we performed high resolution TEM measurements (Figures S5−S6), which supported the trend found by DLS. With no acetic acid added to the reaction, TEM images show large, obvious NP formation. Using the same preparatory techniques for the TEM grid, when 3 is predissolved with AcOH the NPs are not present. The AcOH likely stabilizes the Ir active species in two ways: first, the COOH group acts as an in situ chelate ligand which binds to 2′ and prevents aggregation; and second, the CH3 group serves as a CH oxidation source similar to Cp*, thereby reducing the rate by which the Ir centers may interact with each other and form NPs. When 100 equivalents of Na2SO4 are added to the DLS samples to act as a coordinating anion but not as a CH oxidation source, NPs still form after ∼30 min. When 100 equiv of tetrahydrofuran is added to act as a CH oxidation substrate (it is converted to the corresponding lactone) but not as a chelate ligand, NPs also form. AcOH, the main organic byproduct from the Cp* activation and previously thought to have interfered with catalysis, actually may be the reason behind the long-term stability of 2 noted in prior work.16 Chemical and Electrochemical Water Oxidation with 3. Although 1 and 3 do not give exactly the same species on oxidative activation, both precatalysts are capable of water oxidation (Table 1). We first tested a variety of chemical oxidants with 3. No acetic acid was added to these samples as NP formation only occurs after 30 min and the WO rates are determined on the basis of the first 30 s after catalyst injection. Sodium periodate, ceric(IV) ammonium nitrate (CAN), and

Figure 4. UV−vis kinetics of (black) 0.5 mM 1 with 50 mM NaIO4 in H2O monitored at 610 nm, and (red) 0.5 mM 3, 50 mM NaIO4, and 25 mM acetic acid in H2O, monitored at 595 nm under the same conditions. Acetic acid was added to prevent nanoparticle formation (see below). Absorbances were normalized.

supported by the fact that the activation of 3 follows zero-order kinetics with respect to oxidant (Figure S2). Closer inspection of the UV−vis spectrum of activated 3 revealed that the λmax of 2′ was slightly blue-shifted relative to 2. This λmax shift indicates that the coordination sphere around 2′ is different from that in 2. The activation of 3 requires only a small amount of periodate (as little as 2.5 equiv per Ir, Figure S2). In contrast, 1 requires at least 30 equiv of periodate, owing to the oxidative degradation of Cp* to give acetic acid and other products. The excess degradation products, especially iodate and acetate, are expected to bind to the Ir-oxo dimer19,38 so as to shift the absorbance slightly. Indeed, when external amounts of both acetic acid and iodate are added to 2′, the λmax red-shifts eventually to match the 610 nm value previously seen for 2 (Figure 5). Our new working hypothesis is that 2′ is different from but related to 2 and likely still has IrIV-oxo species in solution that are capable of oxidation catalysis. C

DOI: 10.1021/acs.inorgchem.5b02809 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 6. Mean particle size of solutions of 0.25 mM 3 with 250 mM NaIO4 and other additives: (left) various amounts of acetic acid and (right) sodium sulfate and tetrahydrofuran (THF). All sample volumes were kept consistent at 4 mL. Insets for each graph show a more detailed view of the initial NP formation kinetics. Large particles observed upon initial injection in the AcOH graph are due to bubble formation from water oxidation catalysis. This behavior has been documented in previous studies by our group:52 the bubbles eventually dissipate from solution and the particles observed are from Ir species in solution. For more detail, refer to the experimental details later in the paper.

Table 1. Rates of Catalytic Oxygen Evolution Using [Ir(CO)2(pyalc)] and Cp*Ir(pyalc)OH with Various Chemical Oxidantsa 3, [Ir(CO)2(pyalc)]

1, Cp*Ir(pyalc)OH ‑1

oxidant

rate (μmol O2/L min)

TOF (s )

rate (μmol O2/L min)

TOF (s‑1)

NaIO4 NaIO4 + 100 equiv acetic acidb CAN oxone NaClO H2O2

345 ± 15.0 329 ± 6.0 102 ± 18.4 0.65 ± 0.20 NR NR

5.8 ± 0.3 5.5 ± 0.1 1.7 ± 0.3 0.01 ± 0.003 NR NR

105 ± 5.5 105 ± 3.4 13.9 ± 1.1 NRc NR NR

1.8 ± 0.2 1.8 ± 0.1 0.23 ± 0.02 NR NR NR

Conditions: 1 μM [Ir], 50 mM oxidant in H2O, 25 °C. bEquivalents of additive are relative to mmol catalyst. For more information, refer to experimental details. cNR = no reaction.

a

oxone all gave oxygen evolution as measured by a Clark-type electrode. Sodium hypochlorite and hydrogen peroxide did not produce O2. While 3 with CAN and oxone showed modest or poor O2 production, 3 injected into NaIO4 evolved oxygen nearly three times as fast as the Cp* analogue under the same conditions: 3 has a TOF of 5.8 s−1, 1 has 1.7 s−1. Compound 3 continued to be a robust WO precatalyst even in the presence of 100 equiv of acetic acid, which was added to ensure that no NPs are formed during longer WO experiments. The blue solution from 3 continued to evolve dioxygen after several injections of oxidant over an hour, indicating that it is a longlived active species (Figure S9). We also tested the viability of 3 as an electrochemical catalyst. By cyclic voltammetry, we see a catalytic response around 1.4 V versus NHE using a gold electrode at pH 3.5 (Figure 7). These features were promising for electrochemical activation and oxygen evolution using 3. With adaptation of previously used methods from our group,39 3 was activated by a high surface area platinum working electrode in a solution of Na2SO4 and acetate buffer. Similar to chemical activation, electrochemical activation of 3 is much milder and faster relative to 1; 3 appears to lose its CO ligands after only 8 h of electrolysis at 1.0 V versus NHE, compared to 1, which requires 36−48 h of electrolysis at 1.45 versus NHE to fully degrade the Cp*. The milder oxidation conditions allowed us to also use a

Figure 7. Cyclic voltammogram of 1 mM 3 (black) in 0.25 M Na2SO4, pH 2.75, with a gold working electrode, Ag/AgCl reference, and Pt wire counter. Scans were taken at 500 mV/s. CV of blank electrolyte was taken under same setup (red). The reductive wave at 0.9 V is associated with the reduction of gold oxide at the surface of the working electrode.

reticulated vitreous carbon (RVC) working electrode to activate 3. Solutions from Pt-activated and RVC-activated 3 were D

DOI: 10.1021/acs.inorgchem.5b02809 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

these spectra, as the aromatic peaks are too broad and the aliphatic region is overwhelmed by the signals from acetic acid and other residual Cp* oxidation fragments. Careful examination of a 1H NMR spectrum of 2′, the result of 3 being activated with NaIO4, shows peaks that are much sharper and have a narrower range than those from 2 (Figure 9). This may be because the swifter activation of 3 relative to 1 allows for fewer Ir isomers to be produced. More importantly, with no organic byproducts from precursor activation, the aliphatic region for 2′ reveals clear evidence of several isomeric forms being present. While the isomer peaks from the aromatic region are broad and somewhat indistinct, a wide range of sharp, small peaks from 0.5 to 3 ppm clearly indicates that several similar species exist in solution, with one dominant species at 1.48 ppm. These aliphatic peaks are likely associated with the methyl groups of the pyalc. Even more promising, 1H NMR spectra of electrochemically activated 3, 2′e, showed even fewer peaks in both the aromatic and aliphatic regions (Figure S14). This reduction in the number of isomers may arise because the electrochemical activation of 3 is much milder in terms of oxidizing potential relative to NaIO4. Periodate may allow more thermodynamically unfavorable isomers to be formed as 3 loses its placeholder ligands and forms a multinuclear species. Detailed interpretation is complicated because although binuclear IrIV complexes could be diamagnetic by antiferromagnetic coupling, monomers and oddnuclearity clusters would be paramagnetic and not be expected to show any NMR signals. Diffusion-ordered NMR spectroscopy (DOSY) measurements were performed in an attempt to see if the signals from the proposed dinuclear blue species could be distinguished from those from monomers and organics; while qualitative analyses about complexes’ sizes can be derived on the basis of diffusion constants, the overlapping signals and fast relaxation times of the polymetallic species made it difficult to apply this method with confidence (Figure S15). No signals were detected by EPR spectroscopy. Nevertheless, the presence of fewer isomers in 2′ may make it possible to crystallize one of the Ir activation products. All prior crystallization efforts carried out on 2 failed, likely due to the presence of multiple isomers. In fact, a novel mononuclear compound, Ir(pyalc)2I2 (Figure 10), has indeed been isolated from a solution of 2′, the first species to be unambiguously characterized from an activated Ir precursor solution. Unfortunately, the compound isolated is not the blue solution species but rather a decomposition product. The iodide ligands must come from the NaIO4 oxidant employed in the activation step, with the resulting iodate eventually decomposing to I−/I2/ I3− in aqueous conditions. Even so, the fact that a single species was obtained from a mixture of isomers indicates that 3 may be a better choice than 1 as precatalyst for future crystallization attempts.

evaporated to dryness and redissolved in D2O to examine their 1 H NMR. The spectra were very similar, suggesting that the type of electrode used for bulk electrolysis does not affect the activation process and the Ir species produced (Figure S11). Both activated solutions were also deep blue in color, λmax = 600 nm, and showed no nanoparticles by DLS (Figure S4). Oxygen assays carried out on 2′e (the electrochemically formed blue solution from 3) show ∼95% Faradaic efficiency at 1.4 V versus NHE at pH 3.5 (Figure 8). At this pH, this

Figure 8. Evolved oxygen by Clark electrode from a 1 mM solution of 2′e, formed after 8 h of bulk electrolysis at 1.0 V in 0.25 M Na2SO4 and 100 mM acetate buffer. A gold electrode is held at 0 V for 10 min and then stepped to 1.4 V for 10 more minutes. The blue line indicates the experimental trace, and the black dotted line is the theoretical yield based on current passed.

corresponds to an overpotential of about 350 mV. Interestingly, the acetate buffer component of the bulk electrolyte does not significantly affect the Faradaic efficiency of the catalyst, despite the fact that the acetate CH bonds could be prone to oxidation. Acetate buffer was added to the bulk electrolyte to stabilize 2′e, as for the addition of acetic acid to chemically activated 2′. Yet, we only begin to observe a decrease in Faradaic efficiency when greater than 150 mM of acetate buffer is used. This likely indicates that excessive amounts of acetate act as a CH substrate to be oxidized, rather than an in situ ligand. Just as nanoparticles were detected in the DLS solution studies, we observed by SEM heterogeneous Ir deposition on the working electrode in the absence of acetate buffer (Figure S8). The deposition decreased as more acetate buffer was present, with 100 mM being the ideal concentration that limits deposition but does not affect Faradaic efficiency (Figure S9). Indications Concerning Different Isomers Present. In a previous study,16 1H NMR spectra of solutions of Cp* catalyst, 1, activated with NaIO4 have shown broad peaks in the aromatic region attributed to the pyalc ligand of a paramagnetic Ir species, originally suggested to be assigned to the dimer 2. More detailed measurements of these solutions show that these broad features consist of clusters of overlapping peaks with similar shifts (Figure S12), suggesting the presence of multiple, noninterconverting isomers. This would align well with our group’s demonstration that all possible IrIV(pyalc)2Cl2 and IrIV(pyalc)3 isomers can coexist without interconversion.37 Unfortunately, no dominant isomer can be identified from



CONCLUSION We report the synthesis of a new bis-carbonyl Ir precursor, Ir(CO)2(pyalc), as a precatalyst for catalytic water oxidation. This complex can be activated to an active WO species upon addition to aqueous oxidant, but requires acetate and/or acetic acid to remain homogeneous over extended periods of time (hours to days). Due to the lack of excess iodate, the activated species from Ir(CO)2(pyalc) evolves dioxygen at nearly three times the rate of the previously reported solution made from oxidation of the Cp*Ir(pyalc)OH precursor. The bis-carbonyl complex is also a competent electrochemical WO precatalyst, E

DOI: 10.1021/acs.inorgchem.5b02809 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 9. 1H NMR spectrum of 1.8 mM 3 activated by 10 equiv of NaIO4 in D2O. The peak at 3.2 ppm is a dimethylsulfone standard. Solution was dried and redissolved in D2O. Synthesis of [IrI(cod)(pyalc)]. This known compound was prepared by an alternative method. [Ir(cod)Cl]2 (0.4 mmol, 268.7 mg) was added to a Schlenk flask with stir bar. The flask was put under inert conditions, and 15 mL of DCM was added. NaOtBu (0.8 mmol, 76.9 mg) and pyalc ligand (0.8 mmol, 109.8 mg) were dissolved in 15 mL of H2O and degassed for 10 min. The aqueous solution was transferred to the Schlenk flask via cannula, and the biphasic system was vigorously stirred at room temperature for 1 h under positive pressure of N2. The aqueous layer slowly changed from a cloudy yellow to a clear pale yellow, while the organic layer remained a clear bright orange. The organic layer was collected via a separatory funnel, and the aqueous layer was washed with 2 × 10 mL DCM. The organic fractions were combined and dried with MgSO4. The solvent was removed in vacuo and produced a bright orange powder. Yield: 212 mg (68.6%). 1H NMR (400 MHz, CD2Cl2): δ = 8.92 (d, J = 5.7 Hz, 1H), 7.93 (t, J = 7.8 Hz, 1H), 7.48 (d, J = 8.1 Hz, 1H), 7.33(1, J = 7.3 Hz, 1H), 1.51 (s, 3H). 13C NMR (126 MHz, CD2Cl2): δ = 179.9, 176.3, 175.3, 151.2, 139.9, 123.9, 122.75, 86.48, 34.36. Synthesis of [IrI(CO)2(pyalc)]. [Ir(cod) (pyalc] (0.1 mmol, 43.8 mg) was added to a Schlenk flask and put under N2 before being dissolved in 15 mL of dry DCM added via syringe. CO gas was then bubbled through the reaction for 30 min. The reaction turned from clear orange to clear yellow. Upon completion, the solution was then filtered through a 0.2 μm PTFE syringe filter, and the solvent was

with activation requiring much milder conditions and taking place much faster than that for Cp*Ir(pyalc)OH. The resulting catalyst achieves near-100% Faradaic efficiency with low overpotentials. We find that the activation of both Ir(CO)2(pyalc) and Cp*Ir(pyalc)OH precursors produces a mixture of different Ir(pyalc) isomers which may explain the difficulty in crystallizing any activated blue solution intermediate up to now. The isolation of the Ir(pyalc) 2 I 2 decomposition product from a solution of activated Ir(CO)2(pyalc) encourages us to try to crystallize other relevant species in future.



EXPERIMENTAL SECTION

General Procedures. All organic solvents were dried using a Grubbs-type purification system. High purity Milli-Q water was used for all aqueous experiments. All chemicals were purchased from major suppliers and used without further purification. Syntheses were performed using standard Schlenk techniques under a dry atmosphere of N 2 . The 2-(2′-pyridyl)-2-propanoate (pyalc) ligand, 53 Cp*IrIII(pyalc)OH,6 and Cp*IrIII(pyalc)Cl15 were prepared according to previous literature procedures. 1H NMR spectra were collected at room temperature on a 400 MHz or a 600 MHz Varian spectrometer and referenced to the residual solvent signal (δ in ppm, J in Hz). F

DOI: 10.1021/acs.inorgchem.5b02809 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

(ALV-GmbB). g(τ) functions were fit by a second-order cumulant analysis to obtain average particle size. All solutions were passed through 0.2 μm Teflon PTFE filters prior to measurement or injection to remove large particles. Oxygen evolution assays with chemical oxidants were performed as previously described.7 Assays were conducted with a temperature controlled YSI 5300A Clark-type electrode which was inserted into a bubble-free, water-cooled jacket at 25 °C. A typical experiment used 5 mL of freshly prepared aqueous NaIO4 (50 mM) which was allowed to equilibrate for several minutes before inserting the electrode into the cell. The electrode was then allowed to stabilize before injecting 5 μL of 1 mM catalyst solution to start catalysis. If additives were used, they were added from a 1 or 10 mM aqueous stock solution while the cell was allowed to equilibrate with the atmosphere and temperaturecontrolled jacket. The procedure was repeated three times for reproducibility. Rates were determined on the basis of the first 30 s of data collection. Electrochemical studies were performed as previously described.39 Experiments were carried out on a Princeton Applied Research VersaSTAT-4, using a standard three-electrode setup. Reference and working electrodes were purchased from Bioanalytical Systems, Inc., unless otherwise noted. A Ag/AgCl reference electrode (+0.197 V vs NHE) was used for all measurements. Working electrodes for cyclic voltammograms and oxygen assays were either gold or platinum (0.017 cm2); counter electrodes were platinum wire. Electrodes were cleaned prior to use by thorough polishing with 1 μm diamond paste and cycled in 0.5 M sulfuric acid over potential range −0.375 to 1.8 V versus Ag/AgCl. Platinum electrodes were used to calibrate the oxygen collection efficiency both before and after oxygen assays for accurate Faradaic efficiency calculations. Gold electrodes were used for O2 evolution measurements; gold electrodes were chosen because they have minimal background water oxidation relative to platinum. The working electrode for bulk electrolysis was either a platinum gauze basket or a reticulated vitreous carbon electrode; the counter electrode was platinum mesh. Concentrations of iridium precatalyst 3 were 1 mM dissolved in 0.25 M Na2SO4 and 100 mM acetate buffer (50 mM acetic acid and 50 mM sodium acetate) bulk electrolyte. pH was adjusted using H2SO4 and NaOH. Oxygen evolution measurements for electrochemical experiments were adapted from a literature procedure39 and by using the same Clark-type electrode as described above.

Figure 10. ORTEP crystal structure of Ir(pyalc)2I2, a deactivation product crystallized from a modified solution of 2′ with 50% thermal ellipsoids. After activation of 1 mM 3 with 10 equiv of NaIO4 in 5 mL of H2O and 5 mL of tBuOH, 2 mM pyalc ligand was added and allowed to sit at room temperature for three months. Pyalc was believed to be a good chelating ligand that would aid in crystallizing an active intermediate, but proved to be so strong it broke the multinuclear species. removed in vacuo. The pale yellow powder was further dried under vacuum for 1 h. Yield: 37.4 mg (97%). [Ir(cod)(pyalc] can also be used directly from the biphasic reaction described above. The reaction was cannulated from the Schlenk flask into a degassed separatory funnel. The organic phase was emptied into the two-neck flask under vacuum attached to the funnel through a ground glass joint. CO gas was then bubbled through the reaction for 30 min. Once the reaction turned from clear orange to clear yellow, the yellow solid was obtained in the same manner as described previously. Yield from a 1.0 mmol scale reaction: 333.6 mg (86.3%). 1 H NMR (400 MHz, CD2Cl2): δ = 8.72 (d, J = 5.5 Hz, 1H), 7.93 (t, J = 7.7 Hz, 1H), 7.48 (d, J = 8.1 Hz, 1H), 7.33(1, J = 6.5 Hz, 1H), 1.51 (s, 6H). 13C NMR (126 MHz, CD2Cl2): δ = 179.23, 176.3, 175.3, 151.2, 139.6, 123.9, 122.8, 86.5, 34.4. ESI(+)MS calcd for (C10H12IrNO3 + H)+: 386.04, 384.04. Found: m/z = 386.04, 384.03. Anal. Calcd for C10H10IrNO3: C, 31.25; H, 2.62; N, 3.64. Found: C, 31.53; H, 2.57; N, 3.53. Analyses. Precatalyst 3 is not especially soluble in water, so for aqueous experiments 3 was sonicated for at least 5 min in solution. Solutions of 3 were prepared fresh prior to a given experiment. UV−vis spectra were recorded at 1 nm resolution on a Varian Cary50 using 1.0 cm quartz cuvette. Kinetic measurements were taken at every 5 s for 8 min. Baseline measurements were taken in either neat solvent or blank electrolyte solutions. Transmission electron microscopy (TEM) and scanning transmission electron microscopy coupled with energy dispersive X-ray spectroscopy (STEM-EDX) maps were taken using an FEI Tecnai Osiris 200 kV transmission electron microscope. Samples were prepared by flowing aqueous suspensions of the iridium-containing compounds through copper TEM grids, supported by ultrathin carbon on a lacey carbon support or silicon monoxide (Ted Pella). Dynamic lights scattering experiments were conducted on the basis of literature adaptations.6 Data were collected at 90° with a CGS5000F goniometer setup (ALV5000, ALV-GmbH), using a 532 nm incident laser (Coherent Verdi). Data were collected in intervals of 30 s for the first 3 h of the experiment and then at 5 min intervals afterward in a dark room. For the light scattering experiments of the electrolyzed samples, data were collected in intervals of 30 s for 15 min. Data analysis and sample preparation were adapted from literature procedure.6 Light scattering correlation functions g(τ) with microsecond resolution were collected by onboard correlator cards



ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.5b02809. Characterization data including NMR spectra, UV−vis spectra, and crystallographic structures (PDF) Crystallographic details for 3 + 1,4-butanediol (CIF) Crystallographic details for 3 (CIF) Crystallographic details for Ir(pyalc)2I2 (CIF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare the following competing financial interest(s): US Patent Application Number 14/317,906 by J.M.T., S.W.S., G.W.B, and R.H.C. contains intellectual property described in this article.



ACKNOWLEDGMENTS This research was supported by the Center for Catalytic Hydrocarbon Functionalization, Number DE-SC0001298 (D.L.H. and R.H.C.), and Argonne-Northwestern Solar Energy G

DOI: 10.1021/acs.inorgchem.5b02809 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

(21) Zuccaccia, C.; Bellachioma, G.; Bortolini, O.; Bucci, A.; Savini, A.; Macchioni, A. Chem. - Eur. J. 2014, 20 (12), 3446−3456. (22) 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. Science 2011, 332 (6031), 805−809. (23) Graetzel, M. Acc. Chem. Res. 1981, 14 (12), 376−384. (24) Gust, D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 2009, 42 (12), 1890−1898. (25) Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U. S. A. 2006, 103 (43), 15729−15735. (26) Meyer, T. J. Acc. Chem. Res. 1989, 22 (5), 163−170. (27) Que, L.; Tolman, W. B. Nature 2008, 455 (7211), 333−340. (28) Walter, M. G.; Warren, E. L.; McKone, J. R.; Boettcher, S. W.; Mi, Q.; Santori, E. A.; Lewis, N. S. Chem. Rev. 2010, 110 (11), 6446− 6473. (29) Young, K. J.; Martini, L. A.; Milot, R. L.; Snoeberger, R. C., III; Batista, V. S.; Schmuttenmaer, C. A.; Crabtree, R. H.; Brudvig, G. W. Coord. Chem. Rev. 2012, 256 (21−22), 2503−2520. (30) Brewster, T. P.; Ou, W. C.; Tran, J. C.; Goldberg, K. I.; Hanson, S. K.; Cundari, T. R.; Heinekey, D. M. ACS Catal. 2014, 4 (9), 3034− 3038. (31) Barrett, S. M.; Pitman, C. L.; Walden, A. G.; Miller, A. J. M. J. Am. Chem. Soc. 2014, 136 (42), 14718−14721. (32) Brewster, T. P.; Miller, A. J. M.; Heinekey, D. M.; Goldberg, K. I. J. Am. Chem. Soc. 2013, 135 (43), 16022−16025. (33) Chen, M. S.; White, M. C. Science 2007, 318 (5851), 783−787. (34) Savini, A.; Belanzoni, P.; Bellachioma, G.; Zuccaccia, C.; Zuccaccia, D.; Macchioni, A. Green Chem. 2011, 13 (12), 3360−3374. (35) Zuccaccia, C.; Bellachioma, G.; Bolaño, S.; Rocchigiani, L.; Savini, A.; Macchioni, A. Eur. J. Inorg. Chem. 2012, 2012 (9), 1462− 1468. (36) Junge, H.; Marquet, N.; Kammer, A.; Denurra, S.; Bauer, M.; Wohlrab, S.; Gärtner, F.; Pohl, M.-M.; Spannenberg, A.; Gladiali, S.; Beller, M. Chem. - Eur. J. 2012, 18 (40), 12749−12758. (37) Shopov, D. Y.; Rudshteyn, B.; Campos, J.; Batista, V. S.; Crabtree, R. H.; Brudvig, G. W. J. Am. Chem. Soc. 2015, 137 (22), 7243−7250. (38) Liao, R.-Z. Z.; Siegbahn, P. E. M. ACS Catal. 2014, 4 (11), 3937−3949. (39) Thomsen, J. M.; Sheehan, S. W.; Hashmi, S. M.; Campos, J.; Hintermair, U.; Crabtree, R. H.; Brudvig, G. W. J. Am. Chem. Soc. 2014, 136 (39), 13826−13834. (40) Sharninghausen, L. S.; Campos, J.; Manas, M. G.; Crabtree, R. H. Nat. Commun. 2014, 5, 5084. (41) Ugo, R.; La Monica, G.; Cenini, S.; Bonati, F. J. Organomet. Chem. 1968, 11, 159−166. (42) Claver, C.; Ruiz, A.; Masdeu, A. M.; Viñas, J.; Saballs, T.; Lahoz, F. J.; Plou, F. J. J. Organomet. Chem. 1989, 373 (2), 269−278. (43) Choquesillo-Lazarte, D.; García-Ruiz, J. M. J. Appl. Crystallogr. 2011, 44 (1), 172−176. (44) Sharninghausen, L. S.; Mercado, B. Q.; Crabtree, R. H.; Balcells, D.; Campos, J. Dalt. Trans. 2015, 44, 18403−18410. (45) Etter, M. C.; Baures, P. W. J. Am. Chem. Soc. 1988, 110 (2), 639−640. (46) Etter, M. C.; Reutzel, S. M. J. Am. Chem. Soc. 1991, 113 (7), 2586−2598. (47) Li, N.; Shi, L.; Wang, X.; Guo, F.; Yan, C. Int. J. Anal. Chem. 2011, 2011, 130102. (48) Aakeroy, C. B.; Seddon, K. R. Chem. Soc. Rev. 1993, 22 (6), 397−407. (49) Crabtree, R. H.; Eisenstein, O.; Sini, G.; Peris, E. J. Organomet. Chem. 1998, 567 (1−2), 7−11. (50) Desiraju, G. J. Chem. Sci. 2010, 122 (5), 667−675. (51) Etter, M. C. Acc. Chem. Res. 1990, 23 (4), 120−126. (52) Graeupner, J.; Hintermair, U.; Huang, D. L.; Thomsen, J. M.; Takase, M.; Campos, J.; Hashmi, S. M.; Elimelech, M.; Brudvig, G. W.; Crabtree, R. H. Organometallics 2013, 32 (19), 5384−5390.

Research (ANSER), Number DE-SC0001059 (J.M.T., R.-B.S., K.L.M. and G.W.B.); both agencies are Energy Frontier Research Centers funded by the U.S. Department of Energy. D.L.H acknowledges support by the National Science Foundation Graduate Research Fellowship Program under Grant No. DGE-1122492 and the Yale Dox Summer Research Fellowship. R.-B.S. R. B.-S. acknowledges support from the Consejo Nacional de Ciencia, Tecnologiá e Innovación Tecnológica (CONCYTEC), and the Pontificia Universidad Católica del Perú (PUCP). DLS measurements were conducted in the Yale Facility for Light Scattering. We thank Liam Sharninghausen for assistance with crystallography and Dr. Eric Paulson for assistance with DOSY NMR spectroscopy.



REFERENCES

(1) Hull, J. F.; Balcells, D.; Blakemore, J. D.; Incarvito, C. D.; Eisenstein, O.; Brudvig, G. W.; Crabtree, R. H. J. Am. Chem. Soc. 2009, 131 (25), 8730−8731. (2) Blakemore, J. D.; Schley, N. D.; Balcells, D.; Hull, J. F.; Olack, G. W.; Incarvito, C. D.; Eisenstein, O.; Brudvig, G. W.; Crabtree, R. H. J. Am. Chem. Soc. 2010, 132 (45), 16017−16029. (3) DePasquale, J.; Nieto, I.; Reuther, L. E.; Herbst-Gervasoni, C. J.; Paul, J. J.; Mochalin, V.; Zeller, M.; Thomas, C. M.; Addison, A. W.; Papish, E. T. Inorg. Chem. 2013, 52 (16), 9175−9183. (4) Grotjahn, D. B.; Brown, D. B.; Martin, J. K.; Marelius, D. C.; Abadjian, M.-C.; Tran, H. N.; Kalyuzhny, G.; Vecchio, K. S.; Specht, Z. G.; Cortes-Llamas, S. A.; Miranda-Soto, V.; van Niekerk, C.; Moore, C. E.; Rheingold, A. L. J. Am. Chem. Soc. 2011, 133 (47), 19024−19027. (5) Hetterscheid, D. G. H.; Reek, J. N. H. Chem. Commun. 2011, 47 (9), 2712−2714. (6) Hintermair, U.; Hashmi, S. M.; Elimelech, M.; Crabtree, R. H. J. Am. Chem. Soc. 2012, 134 (23), 9785−9795. (7) Hintermair, U.; Sheehan, S. W.; Parent, A. R.; Ess, D. H.; Richens, D. T.; Vaccaro, P. H.; Brudvig, G. W.; Crabtree, R. H. J. Am. Chem. Soc. 2013, 135 (29), 10837−10851. (8) Hong, D.; Murakami, M.; Yamada, Y.; Fukuzumi, S. Energy Environ. Sci. 2012, 5 (2), 5708−5716. (9) Kawahara, R.; Fujita, K.; Yamaguchi, R. J. Am. Chem. Soc. 2012, 134 (8), 3643−3646. (10) Lalrempuia, R.; McDaniel, N. D.; Müller-Bunz, H.; Bernhard, S.; Albrecht, M. Angew. Chem., Int. Ed. 2010, 49 (50), 9765−9768. (11) Lewandowska-Andralojc, A.; Polyansky, D. E.; Wang, C.-H.; Wang, W.-H.; Himeda, Y.; Fujita, E. Phys. Chem. Chem. Phys. 2014, 16 (24), 11976−11987. (12) Park-Gehrke, L. S.; Freudenthal, J.; Kaminsky, W.; DiPasquale, A. G.; Mayer, J. M. Dalt. Trans. 2009, No. 11, 1972−1983. (13) Savini, A.; Bellachioma, G.; Ciancaleoni, G.; Zuccaccia, C.; Zuccaccia, D.; Macchioni, A. Chem. Commun. 2010, 46 (48), 9218− 9219. (14) Savini, A.; Bucci, A.; Bellachioma, G.; Giancola, S.; Palomba, F.; Rocchigiani, L.; Rossi, A.; Suriani, M.; Zuccaccia, C.; Macchioni, A. J. Organomet. Chem. 2014, 771, 24−32. (15) Schley, N. D.; Blakemore, J. D.; Subbaiyan, N. K.; Incarvito, C. D.; D’Souza, F.; Crabtree, R. H.; Brudvig, G. W. J. Am. Chem. Soc. 2011, 133 (27), 10473−10481. (16) Sheehan, S. W.; Thomsen, J. M.; Hintermair, U.; Crabtree, R. H.; Brudvig, G. W.; Schmuttenmaer, C. A. Nat. Commun. 2015, 6, 6469. (17) Turlington, C. R.; White, P. S.; Brookhart, M.; Templeton, J. L. J. Am. Chem. Soc. 2014, 136 (10), 3981−3994. (18) Turlington, C. R.; White, P. S.; Brookhart, M.; Templeton, J. L. J. Organomet. Chem. 2015, 792, 81−87. (19) Wang, C.; Wang, J.-L.; Lin, W. J. Am. Chem. Soc. 2012, 134 (48), 19895−19908. (20) Zhou, M.; Schley, N. D.; Crabtree, R. H. J. Am. Chem. Soc. 2010, 132 (36), 12550−12551. H

DOI: 10.1021/acs.inorgchem.5b02809 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (53) Wong, Y.-L.; Yang, Q.; Zhou, Z.-Y.; Lee, H. K.; Mak, T. C. W.; Ng, D. K. P. New J. Chem. 2001, 25 (2), 353−357.

I

DOI: 10.1021/acs.inorgchem.5b02809 Inorg. Chem. XXXX, XXX, XXX−XXX