Abstract
Seven acetylacetonate (acac) metal complexes ranging from early transition metals to post-transition metals were examined by cyclic voltammetry in acetonitrile (MeCN), dichloromethane (DCM), tetrahydrofuran (THF), dimethyl sulfoxide (DMSO), and dimethylformamide (DMF). The electronic potential of any observed redox events is reported along with an analysis of the reversibility of those events across a range of scan rates. Group 8 compounds Fe(acac)3 Ru(acac)3 showed at least quasi-reversible reductions across all solvents while Ru(acac)3 also featured a reversible oxidation. The early and post-transition compounds VO(acac)2, Ga(acac)3 and In(acac)3 exhibited irreversible reductions, while TiO(acac)2 showed no redox activity within the examined potential ranges. Mn(acac)3 featured an oxidation that showed solvent-dependent reversibility, and a reduction that was irreversible in all examined solvents. DFT calculations indicated minimal solvent effects on the HOMO-LUMO gap for the majority of compounds, but a significant effect was observed for Ru(acac)3. This study serves as a valuable initial step for further examination of acetylacetonate metal complexes for applications as electrochemical internal standards, nanoparticle precursors, and electrocatalysts.
Keywords: Electroanalytical Electrochemistry, Organic Electrochemistry, Theory and Modelling
Introduction
Recent years have seen an increasing demand for renewable energy resulting from the heavy reliance on unsustainable fossil fuels that produce greenhouse gases.1 Many alternative energy molecules currently under examination are formed through electrochemical transformations, such as H2 formed through proton reduction, O2 formed through water oxidation, and various carbon compounds formed through CO2 reduction.2,3 Electrochemical studies require a standard reference to measure accurately the potential at which these transformations occur.4 Reliable reference electrodes have been well established for aqueous systems, but internal standards for organic solvents remain limited. The most common reference for electrochemical studies in organic solvents is ferrocene (Fc), as it features a reversible Fe3+/2+ couple at 0.7 V vs SHE, is stable under ambient conditions, low-cost, and is inert towards most analytes.5 Ruiz Aranzaes et al. explored metallocenes as internal references for cyclic voltammetry and found substituted decamethylferrocenes, decamethylcobaltocene, and [FeICp*(η6-C6Me6)] showed less solvent dependence than ferrocene on redox potential.6
To further aid in the search for alternative energy, a great deal of modern research is focused on identifying affordable electrocatalysts capable of producing alternative energy molecules at modest overpotentials.1 Although many catalysts are molecular and designed for their proposed function, others are nanoparticle precursors that deposit a catalytically active material on the surface of the electrode.7,8 At present, it is difficult to predict which compounds will yield a catalytically active deposit, so a screening approach may be useful as an initial step to identify such nanoparticle precursors.
The electrochemical properties of many transition metal acetylacetonate compounds were examined in the late 1970s through the 1980s in organic solvents that would go on to be used frequently in electrochemical research: acetonitrile (MeCN), dichloromethane (DCM), tetrahydrofuran (THF), dimethyl sulfoxide (DMSO), and dimethylformamide (DMF). Gritzner et al. reported a thorough examination of manganese(III) acetylacetonate, Mn(acac)3, in a wide array of organic solvents and electrolytes.9 The group 8 compounds iron(III) acetylacetonate, Fe(acac)3, and ruthenium(III) acetylacetonate, Ru(acac)3, have been examined in DCM, MeCN, THF, and DMF.10,11,12 In these reports, the data were collected using a silver electrode as the sole reference. The contemporary procedure for referencing electrochemical data in an organic solvent is to use an internal standard, such as ferrocene, to avoid drift during experiments.4 In this report, we both expand upon the current literature of metal acetyacetonate electrochemistry by examining more compounds in a wider array of organic solvents, as well as update it by utilizing ferrocene, employing variable scan rate, and supplementing our findings with Density Functional Theory (DFT) analysis.
This study screens the electrochemical properties of commercially available metal acetylacetonate compounds (Figure 1). Coordination compounds examined in this study include: Ru(acac)3, Fe(acac)3, TiO(acac)2, VO(acac)2, Ga(acac)3, In(acac)3, and Mn(acac)3. The acetylacetonate compounds were selected due to their commercial availability, solubility in organic solvents, as well as to provide samples from early, late, and post-transition metals. These compounds were examined by cyclic voltammetry (CV) in the organic solvents: MeCN, DCM, THF, DMSO, and DMF. This diverse set of solvents features a wide range of polarity, coordination capabilities, and dielectric constants.13 Each compound was then examined by DFT to look at the HOMO-LUMO energy gap, orbital diagrams and the distribution of electron density across the molecule. The influence of the solvents on the electrochemical properties of the metal complexes is described below.
Figure 1.
Trisacetylacetonato complex (left) bisacetylacetonatooxide complex (right)
Experimental Section
General Considerations.
All chemical compounds were purchased commercially and used as is. Tetra-n-butylammonium hexafluorophosphate (98%) was purchased from Thermo Scientific. Acetonitrile (≥99.9%), Tetrahydrofuran (HPLC grade, ≥99.7%), dimethyl sulfoxide (≥99.9%), and dimethylformamide (HPLC grade, ≥99.9%) were purchased from VWR Chemicals. Dichloromethane (≥99.5%) was purchased from Macron Fine Chemicals. Ruthenium(III) acetylacetonate, vanadium(IV) oxide acetylacetonate, and indium(III) acetylacetonate were purchased from Beantown Chemical. Ferrocene and iron(III) acetylacetonate were purchased from Tokyo Chemical Industry. Titanium(IV) oxide acetylacetonate and gallium(III) acetylacetonate were purchased from Strem Chemicals. Manganese(III) acetylacetonate was purchased from Acros Organics.
Electrochemical Measurements.
Electrochemical analyses were conducted at room temperature in a 10 mL glass cell equipped with a glassy carbon working electrode (3.0 mm diameter), a silver pseudo reference electrode (53 mm length), and a platinum counter electrode (0.58 cm2 approximate surface area), using a WaveNow potentiostat, all of which was purchased from Pine Research. The silver pseudo reference electrode was placed in a glass tube fitted with a porous glass tip, which was then filled with the electrolyte solution to be used in the cell (see below). The glassy carbon electrode was polished with alumina slurry then rinsed with deionized water and dried prior to each experiment.
An electrolyte stock solution of 250 mM tetrabutylammonium hexafluorophosphate [NBu4][PF6] was prepared in each of the following solvents: MeCN, DMF, DCM, and DMSO. For THF, a 200 mM solution of [NBu4][PF6] was prepared. In each experiment, 5 mL of the electrolyte stock solution was added to the cell, which was sparged with N2 gas. Whenever the cell was unsealed, the solution was sparged with N2 again. An open circuit potential (OCP) was performed while the solution was stirred with a magnetic stir bar and repeated with the addition of any compound to the solution. A CV was collected on the electrolyte solution, starting and ending at the measured OCP, at 250 mV/s and 3 segments within the following potential parameters for each solvent: THF −1.8 V to 1.6 V, MeCN −2.6 V to 2.3 V, DMF −2.4 to 1.6 V, DCM −1.6V to 1.8 V, DMSO −2.6 V to 1.4 V.
Solid acetylacetonate compound (0.015 mmol) was added to the electrochemical cell. The solution was stirred with a magnetic stir bar and an OCP experiment was performed. Using the result of OCP as the initial potential, a CV experiment collected at 250 mV/s within the solvent ranges describe above. If a cathodic peak was observed without an anodic peak on the return scan, or vice versa, the signal was deemed irreversible. If the compound exhibited a cathodic and an anodic peak, further CV experiments were performed at variable scan rates (VSR, reported in mV/s): 50, 100, 500, and 1000. If both the cathodic and anodic peaks were detected at all scan rates, the feature was deemed reversible. If the return wave diminished at 50 mV/s, but became more pronounced as scan rate increased, the feature was deemed quasi-reversible.
A stock solution of 1.0 mM ferrocene in the corresponding electrolyte solution was prepared. To the electrochemical cell, 100 μL of the Fc stock solution was added, followed by OCP and CV experiments collected at 250 mV/s. The analyte signal was referenced to the Fc signal. The E1/2 for all reversible and quasi-reversible signals was calculated by averaging the anodic and cathodic peak values, then subtracting the E1/2 value of Fc.
Computational Methods.
Ruthenium(III) acetylacetonate [Ru(acac)3] and indium(III) acetylacetonate [In(acac)3] were optimized using DFT B3LYP-lan12dz guess = mix and scf=xqc in vacuum.14 Ru(acac)3 was also optimized in the following solvents: acetonitrile (MeCN), dichloromethane (DCM), tetrahydrofuran (THF), dimethylsulfoxide (DMSO), and n,n-dimethylformamide (DMF) using scrf=(solvent=solvent name), In(acac)3 was optimized using the same parameters in MeCN and DMSO. Gallium(III) acetylacetonate [Ga(acac)3], iron(III) acetylacetonate [Fe(acac)3], manganese(III) acetylacetonate [Mn(acac)3], and vanadyl acetylacetonate [VO(acac)2] were optimized using DFT B3LYP/6-31g geom=connectivity in vacuum. Fe(acac)3 and VO(acac)2 was also optimized in the following solvents with scrf=(solvent=solvent name) for MeCN, DCM, THF, DMSO, and DMF. Ga(acac)3 was optimized using the same parameters in MeCN, DMF, and DMSO. All DFT orbital diagrams were generated with an isovalue of 0.2.
A zero-point energy correction (ZPE) is an adjustment made to the energy value calculated in Gaussian to account for the inherent vibrational energy that a system possesses as the total energy is calculated along with the translation, rotational, vibration, and electronic motion contributions to the energy. The zero point corrected total energy (E0) is the sum of the total electronic energy (Etot) and the zero point vibrational energy ZPVE: E0 = Etot + ZPVE using non-imaginary frequencies. ZPE was calculated by taking the total thermal energy in Kcal/mol and subtracting the zero-point correction.
Results
The results of the electrochemical analyses are listed in Table 1 with cyclic voltammograms available in the supplementary information. All of the examined compounds, save for TiO(acac)2, featured electrochemical signals.
Table 1.
Electrochemical signals of acetylacetonate compounds.
| MeCN | DCM | THF | DMSO | DMF | |
|---|---|---|---|---|---|
|
Ru(acac)3
Oxidation |
RO E1/2 = 0.61 |
RO E1/2 = 0.56 |
RO E1/2 = 0.54 |
RO E1/2 = 0.60 |
RO E1/2 = 0.59 |
|
Ru(acac)3
Reduction |
RR E1/2 = −1.15 |
RR E1/2 = −1.24 |
RR E1/2 = −1.36 |
RR E1/2 = −1.17 |
RR E1/2 = −1.21 |
| Fe(acac)3 | RR E1/2 = −1.05 |
RR E1/2 = −1.14 |
RR E1/2 = −1.27 |
QR E1/2 = −1.06 |
RR E1/2 = −1.11 |
| TiO(acac)2 | No Signal | No Signal | No Signal | No Signal | No Signal |
| VO(acac)2 | IO Epa = 0.63 |
QO E1/2 = 0.61 |
IO Epa = 0.59 |
IO Epa = 0.44 |
IO Epa = 0.45 |
| Ga(acac)3 | IR Epc = −2.68 |
No Signal | No Signal | IR Epc = −2.71 |
IR Epc = −2.71 |
| In(acac)3 | IR Epc = −2.65 |
No signal | No Signal | IR Epc = −2.59 |
IR Epc = −2.67 |
|
Mn(acac)3
Oxidation |
RO E1/2 = 0.56 |
QO E1/2 = 0.54 |
RO E1/2 = 0.51 |
QO E1/2 = 0.55 |
QO E1/2 = 0.55 |
|
Mn(acac)3
Reduction |
IR Epc = −0.80 |
IR Epc = −0.78 |
IR Epc = −1.00 |
IR Epc = −0.74 |
IR Epc = −0.79 |
Note: All potentials reported as V vs Fc+/0. RO = reversible oxidation, RR = reversible reduction, IO = irreversible oxidation, IR = irreversible oxidation, QR = quasi-reversible reduction, QO = quasi-reversible oxidation.
Ru(acac)3 featured two reversible electrochemical events. A reversible Ru4+/3+ oxidation was observed at approximately 0.6 V vs Fc+/0 in all tested solvents (Figure S1). The solvent identity had minimal effect on the potential of the reduction with a range of 0.05 V across solvents. A reversible Ru3+/2+ reduction was observed within the range of −1.4 to −1.1 V vs Fc+/0 for all solvents (Figure S3). The solvent effect was more pronounced for the reduction, yielding a range of 0.21 V across solvents.
Fe(acac)3 featured a Fe3+/2+ reduction within the range of −1.3 to −1.0 V vs Fc+/0 in all solvents (Figure S5). The feature is quasi-reversible in DMSO, showing a weaker anodic peak at 50 mV/s scan rate (Figure 2), but was found to be reversible in the other solvents. Changing the solvent yielded a variation of 0.22 V for the observed reduction.
Figure 2.
Variable scan rate cyclic voltammograms of the reduction of iron(III) acetylacetonate [Fe(acac)3]. Red trace = 50 mV/s, yellow trace = 100 mV/s, light green trace = 250 mV/s, dark green trace = 500 mV/s, blue trace = 1000 mV/s.
TiO(acac)2 showed no electrochemical signals within the examined solvent windows. This result is not surprising considering inert nature of Ti(IV) towards redox reactions. This also indicates that the acetylacetonate ligands are inactive towards redox under these conditions.
VO(acac)2 featured an irreversible V5+/4+ oxidation with Epa values from 0.44 to 0.63 V vs Fc+/0 (Figure S6). The compound displayed a significant variation between solvents. In DCM, this oxidation becomes increasingly reversible at higher scan rates, indicating quasi-reversibility (Figure S7).
Ga(acac)3 featured an irreversible reduction in MeCN, DMSO, and DMF with Epc values from −2.68 V to −2.71 V Fc+/0 (Figure S8). No signal was observed for DCM and THF, likely due to the event existing outside the solvent windows. Changing the solvent had minimal effect on the potential of this reduction, with a variation in Epc values of 0.03 V.
In(acac)3 featured an irreversible reduction in MeCN, DMSO, and DMF very similar to that of Ga(acac)3 within the range of −2.59 to −2.67 V vs Fc+/0 (Figure S9). No signal was observed for DCM and THF, again likely due to the event existing outside the solvent windows. The observed variation between solvents was 0.08 V.
Mn(acac)3 featured two electrochemical events. A Mn4+/3+ oxidation was observed at 0.5-0.6 V vs Fc+/0 in all solvents (Figure S10). The oxidation appears reversible in MeCN and THF at all scan rates (Figure S11). The oxidation appears irreversible at 50 mV/s scan rate in DCM, DMSO, and DMF, but becomes more reversible at greater scan rates, indicating quasi-reversibility in these solvents. Solvent variation had a negligible effect on the potential of this oxidation with a range of 0.05 V. The compound also features an irreversible Mn2+/3+ reduction in all solvents with Epc values −0.74 to −1.00 V vs Fc+/0 (Figure S12).
DFT calculations.
The HOMO-LUMO energy band gap was calculated to examine the differences between solvents and under vacuum, as it allows the energy gap to be calculated exactly in real space.15 Figure 3 shows the HOMO-LUMO energy gap under vacuum for Ga(acac)3 (4.86 eV), Fe(acac)3 (4.54 eV), VO(acac)2 (3.91 eV), Mn(acac)3 (2.68 eV), Ru(acac)3 (1.21 eV), and In(acac)3 (0.81 eV).
Figure 3.
HOMO-LUMO summary diagram with iron(III) acetylacetonate, gallium(III) acetylacetonate, manganese(III) acetylacetonate, indium(III) acetylacetonate, ruthenium(III) acetylacetonate, vanadyl(IV) acetylacetonate in vacuum.
The HOMO-LUMO energy gap for Fe(acac)3 does not vary much when solvents were added into the calculation. For MeCN (Figure S45) Fe(acac)3 had a 4.56 eV HOMO-LUMO energy gap. The energy gap was found to be the same in DCM (Figure S46), DMF (Figure S47), and THF (Figure S49) for Fe(acac)3. In DMSO, Fe(acac)3 had a HOMO-LUMO energy gap of 4.57 eV (Figure S48). The difference in energy gap for Fe(acac)3 only differed by 0.02 to 0.03 eV depending on the solvents incorporated into the basis set.
VO(acac)2 experiences similar trends in the HOMO-LUMO energy gap when solvents are added into the calculations. The energy gap in DCM (Figure S46) and THF (Figure S49) had the same value as under vacuum. When MeCN (Figure S45), DMF (Figure S47), and DMSO (Figure S48) are employed the HOMO-LUMO energy gap was found to be 3.98 eV. The difference in energy gap for VO(acac)2 differs by 0.03 eV for all solvents incorporated into the basis set.
Ga(acac)3 had the same HOMO-LUMO energy gap regardless of whether the calculations were done in vacuum, MeCN (Figure S45) or DMSO (Figure S48). Mn(acac)3 was found to have a HOMO-LUMO energy gap of 2.73 eV in MeCN (Figure S43), DMF (Figure S47), DMSO (Figure S48); Mn(acac)3 had a 2.72 eV HOMO-LUMO energy gap in THF (Figure S49) and DCM (Figure S46). With the addition of solvents, the energy gap difference was found to be 0.05-0.06 eV. In(acac)3 has a HOMO-LUMO energy gap of 0.77 eV in MeCN (Figure S45) and in DMSO (Figure S48). The energy gap for In(acac)3 shows a difference of 0.04 eV with the addition of solvents.
Ru(acac)3 displayed a more varied HOMO-LUMO energy gap than its transition metal counterparts. In DCM (Figure S46) and THF (Figure S49) the energy gap was close to the one calculated under vacuum at 1.20 eV with a minimal difference of 0.01 eV. However, in DMSO (Figure S48), DMF (Figure S47), and MeCN (Figure S45) the HOMO-LUMO energy gap was calculated to be 2.48 eV. Ru(acac)3 has the largest energy gap difference of 1.28 eV.
Discussion
Of the compounds examined in this study, the group 8 compounds Fe(acac)3 and Ru(acac)3 exhibited the most stability across solvents. These results are consistent with a previous study examining the electrochemistry of Ru(acac)3 in DCM, MeCN, THF, and DMF (Table 2).10 The electrochemical properties of Fe(acac)3 has also been reported separately in DCM and MeCN.11,12 Our results serve to augment the previous findings with referenced data across more solvents and with variable scan rates to elucidate reversibility more rigorously. Although both Fe(acac)3 and Ru(acac)3 were found to be soluble, stable, and reversible across all tested solvents, their utility as internal standards is questionable. The acac ligand, though a chelate, was prone to decomposition in the presence of water during our studies. An ideal standard would be robust and resistant to a wide range of conditions, a quality that has helped solidify ferrocene as the definitive internal standard for electrochemical analyses. Furthermore, Ru(acac)3 features two redox events, effectively doubling the probability of the standard overlapping with an analyte.
Table 2.
Comparison of Experimental Data to Literature Redox Potentials.
| Solvent | Experimental (V vs Fc+/0) | Literature (V vs. Fc+/0) | |
|---|---|---|---|
| Ru(acac)3 Oxidation | MeCN | E1/2 = 0.61 | E1/2 = 0.51* 10 |
| DCM | E1/2 = 0.56 | E1/2 = 0.60* 10 | |
| THF | E1/2 = 0.54 | E1/2 = 0.66* 10 | |
| DMSO | E1/2 = 0.60 | - | |
| DMF | E1/2 = 0.59 | Epa = 0.52* 10 | |
| Ru(acac)3 Reduction | MeCN | E1/2 = −1.15 | E1/2 = −1.23* 10 |
| DCM | E1/2 = −1.24 | E1/2 = −1.24* 10 | |
| THF | E1/2 = −1.36 | E1/2 = −1.19* 10 | |
| DMSO | E1/2 = −1.17 | - | |
| DMF | E1/2 = −1.21 | E1/2 = −1.27* 10 | |
| Fe(acac)3 | MeCN | E1/2 = −1.05 | E1/2 = −1.14* 10 |
| E1/2 = −1.04† 12 | |||
| DCM | E1/2 = −1.14 | E1/2 = −1.19* 11 | |
| THF | E1/2 = −1.27 | - | |
| DMSO | E1/2 = −1.06 | - | |
| DMF | E1/2 = −1.11 | - | |
| VO(acac)2 | MeCN | Epa = 0.63 | - |
| DCM | E1/2 = 0.61 | - | |
| THF | Epa = 0.59 | - | |
| DMSO | Epa = 0.44 | E1/2 = 0.43† 17 | |
| DMF | Epa = 0.45 | - | |
| Mn(acac)3 Oxidation | MeCN | E1/2 = 0.56 | E1/2 = 0.05‡ 9 |
| DCM | E1/2 = 0.54 | E1/2 = 0.10‡ 9 | |
| THF | E1/2 = 0.51 | - | |
| DMSO | E1/2 = 0.55 | E1/2 = 0.00‡ 9 | |
| DMF | E1/2 = 0.55 | E1/2 = 0.01‡ 9 | |
| Mn(acac)3 Reduction | MeCN | Epc = −0.80 | E1/2 = −1.01‡§ 9 |
| DCM | Epc = −0.78 | E1/2 = −1.11‡§ 9 | |
| THF | Epc = −1.00 | E1/2 = −0.92‡§ 9 | |
| DMSO | Epc = −0.74 | E1/2 = −1.23‡§ 9 | |
| DMF | Epc= −0.79 | E1/2 = −1.01‡§ 9 |
Note: *Data was reported as V vs Ag/AgCl, which was converted to V vs Fc+/0 by subtracting the reported value of Fc+/0 = 0.53 V vs. Ag/AgCl.10 †Data was reported as V vs SCE, which was converted to V vs Fc+/0 by subtracting the reported value of Fc+/0 = 0.382 vs SCE.6 ‡Data was reported as V vs bisbiphenylchromium+/0, which was converted first to V vs Ag+/0 by subtracting the reported value of Ag+/0 = 1.12 V vs bisbiphenylchromium+/0,16 then converted to V vs Fc+/0 by subtracting the reported value of Fc+/0 = 0.53 V vs. Ag/AgCl.§Data was collected using polarography.
VO(acac)2 displayed an irreversible oxidation event at approximately 0.5 V vs Fc+/0 while the analogous TiO(acac)2 displayed no redox events in any of the solvents used in this study. A previous study examining VO(acac)2 in DMSO reported similar results.17 The titanium compound in this study serves as a useful control to examine the redox potential of the acetylacetonate ligands. With no signal observed, it can be concluded that both the metal center and the ligands are inert under the experimental conditions of this study. Being the most similar compound, there is evidence to support that the oxidation observed for VO(acac)2 is metal-based. DFT studies were performed to explore this hypothesis further, as well as the nature of the other compounds of this study (see below).
The group 12 compounds, Ga(acac)3 and In(acac)3, both featured an irreversible reduction at very negative potential around −2.7 V vs Fc+/0. The peak current of this reduction is appreciably greater than that of the other examined compounds, exceeding 200 μA in some cases (Figure S9), which could indicate a multi-electron process. The established method for determining a multi-electron process is to measure the peak separation between the cathodic and anodic peaks, but the irreversible nature of this reduction forestalls such analysis.18 However, a previous electrochemical study of a gallium(III) 2,2’-bipyridine complex in DMSO and MeCN found that the gallium(III) was reduced to metallic gallium through the adsorbed ligand, leading credence to our hypothesis.19
Similar to Ru(acac)3, Mn(acac)3 displayed a Mn4+/3+ oxidation and a Mn3+/2+ reduction event. That the reversibility of the oxidation is solvent dependent suggests that the oxidized species may be interacting with DCM, DMSO, and DMF to yield an irreversible signal. This interaction occurs on a slow enough timescale to give a more reversible signal in the CV at greater scan rates. This Mn(IV) species appears to be inert towards MeCN and THF, yielding the reversible traces at all examined scan rates. The reduction appears irreversible in all solvents used in this study, making it difficult to speculate on the cause of the irreversibility. However, based on this evidence, it is possible that the chemical change following the reduction does not involve a solvent molecule. The compounds may warrant further investigation to determine the nature of their decomposition products or their utility as electrocatalysts. In particular, with the relevance of manganese to photosystem II, it should be examined for any activity towards water oxidation.20
The redox properties of Mn(acac)3 in organic solvents have been thoroughly explored by Gritzner et al. but conflict somewhat with our findings (Table 2).9 Similar to this report, they discuss both an oxidation and a reduction of Mn(acac)3, but their reported E1/2 value of the oxidation differs from ours by about −0.5 V. Gritzner et al. describe the necessity to use a special technique involving an unconventional internal reference, bisbiphenylchromium tetraphenylborate, to conduct their experiments. According to their report, the Mn(acac)3 oxidation occurs at a potential nearly identical to that of ferrocene, perhaps explaining the need for the exotic standard. Our results were consistent across all solvents and were further supported by DFT calculations (vide infra). The reduction of Mn(acac)3 is more consistent between our reports, specifically that is appears to be quasi-reversible, widely varied across solvents, and occurs at approximately −1 V vs. Fc+/0. Gritzner et al. utilized a different technique to acquire the reduction data, polarography, which may account for the deviations between our results. Their study serves as a useful comparison to compare our techniques as we expand upon their findings using variable scan rates and DFT studies to describe the electronic structure of these transformations.
The other compounds were found to be consistent with literature reports, where available (Table 2). Some values match perfectly and, outside of the case for Mn(acac)3 described above, the most significant differences seem to occur with THF, giving a maximum difference of 0.17 V for the reduction of Ru(acac)3. The greater variability observed for THF could be the result of the lower solubility of electrolyte, leading to lower conductivity and greater margins of error.
The impact of solvent on the analyte redox potentials appears to be more significant for reductions than oxidations. The E1/2 for the Ru(acac)3 and Mn(acac)3 oxidations shifted by a modest degree between solvents of about 0.05 V. Although the oxidation of VO(acac)2 shifted more significantly, interpretation less reliable at most are as one event was quasi-reversible and others irreversible, making a true E1/2 value difficult to determine. In contrast, the reduction potentials of Ru(acac)3 and Fe(acac)3, shifted 0.2 V and the Epc value of the Mn(acac)3 reduction showed a range of 0.26 V between DMSO and THF. It should be noted that DMSO has the highest dielectric constant of the solvents tested (ε = 47) and THF is among the lowest (ε = 7.6).13 Although the reductions observed in DMSO and THF seem to correlate to the dielectric constants, this trend does not appear to be consistent for the other solvents.
DFT Calculations.
The electronic structure and stability of several first-row transition metal acetylacetonate complexes has recently been studied by Fernandes et al.21 Their work helped to quantify the stability of each metal based on the number of acac by examining the HOMO energy. Similarly, in this study we found that the HOMO of Ga(acac)3 and In(acac)3 in vacuum displays all of the electron density on the acetylacetonate ligands, distributing it from the oxygen atoms connected directly to the transition metals and the connecting carbon atoms. A similar electron density distribution is seen in the LUMO of In(acac)3. In both the HOMO and LUMO of Ga(acac)3 we see that the electron density is almost equally distributed over the three ligands. In In(acac)3 most of the density is only on two ligands and a minor amount is on the connecting oxygen atoms of the third ligand. The LUMO of Ga(acac)3 shows equivalent, but isolated, electron density presences on all three ligands and the transition metal. Transition metal complexes Fe(acac)3 and Mn(acac)3 show density largely on one of the three acetylacetonate ligands with some density also on the metal. Electron density is predominately distributed over two of the ligands with minor density on the transition metal in the case of Ru(acac)3 and VO(acac)2. In the case of VO(acac)2 the electron density is predominately on the oxygen atoms of the acetylacetonate ligands with minimal electron density on the middle carbon of the ligand; furthermore there is only minor density present on the oxygen atom associated with the V=O bond. The LUMO of Fe(acac)3 has density shared equivalently over two of the ligands with a smaller amount displayed on the third ligand. In the case of Mn(acac)3 the oxygen atoms of the acetylacetonate ligands bearing most of the density in the LUMO. The carbon atoms of the ligands do retain some of the electron density and it is spread evenly between each of the three ligands. Ru(acac)3 and VO(acac)2 have similar electron density dispersed across two ligands in their LUMO. However, in the case of Ru(acac)3 there is more electron density on the transition metal than in the HOMO. VO(acac)2 has isolated density centralized on the carbon chain associated with the two ligands along with density shared between the transition metal and all the connected oxygen atoms.
The HOMO-LUMO orbital diagrams did not differ significantly when solvents were incorporated into the caclulations for Fe(acac)3, VO(acac)2, Mn(acac)3, Ga(acac)3, or In(acac)3. Ru(acac)3 experienced the largest difference upon the addition of solvents to the basis set calculations. When calculated in MeCN, DMF, and DMSO, Ru(acac)3 HOMO diagram displays electron density on all three ligands instead of two and more electron density can be seen on the transition metal itself. On one of the ligands the density is isolated from the metal and distributed across the carbon atoms only. For the other two ligands, the density is distributed across the connected carbonyl groups of the acetylacetonate ligand. In DCM, the HOMO orbital diagram showed the density localized primarily on one of the ligands with only minor density focused on the oxygen atoms of the other two ligands and the metal. When calculated in MeCN, DMF, and DMSO, Ru(acac)3 LUMO diagram displays a similar distribution to that of the HOMO diagram; however, the LUMO diagram has isolated electron density on the ligands and the ruthenium atom. The LUMO diagram in DCM shows most of the electron density on the transition metal and one acetylacetonate ligand. This electron density is shared between the Ru-acac bonds. There is very minor electron density present on the oxygen atoms on the remaining two ligands and minimal electron density on the carbon atoms. The HOMO and LUMO of Ru(acac)3 remained similar to the LUMO in vacuum when THF was used in the basis set.
Conclusion
Metal acetylacetonate complexes lend themselves well to electrochemical screening, as they are soluble in a wide range of organic solvents and are commercially available. This study provides initial electrochemical and DFT characterization for a range of metal compounds in the most common organic solvents utilized in electrochemistry. The observed solvent-dependent changes in the electrochemical behavior of these metal complexes show the importance of solvent selection in managing redox reaction reversibility. Each solvent's particular properties, such as polarity and coordination power, play an important role in defining the electrochemical properties of these metal complexes. Solvent-based DFT calculations show that there are minimal changes in the HOMO-LUMO energy gap for most of the metals discussed in the paper; however, it is interesting to note that Ru(acac)3 energy gap and orbital diagrams did vary based on which solvents were used. This further shows how solvents can affect the electrochemical properties. The results of this study provide a starting point for future electrochemical research on these compounds.
Supplementary Material
Acknowledgments
This publication was made possible by a grant from the National Institute of General Medical Sciences (GM103440) from the National Institutes of Health (NIH). Computational studies was made possible using the computational modelling software and supercomputer Cruntch3, supported by the National Science Foundation grant CHE-1531468. This publication’s contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. This project was also supported by Seed Grant funding from the Office of the Provost, Nevada State University.
References
- 1.Zhao X; Levell ZH; Yu S; Liu Y Atomistic Understanding of Two-dimensional Electrocatalysts from First Principles. Chem. Rev 2022, 122, 10675–10709. [DOI] [PubMed] [Google Scholar]
- 2.McCrory CC; Jung S; Ferrer IM; Chatman SM; Peters JC; Jaramillo TF; Benchmarking Hydrogen Evolving Reaction and Oxygen Evolving Reaction Electrocatalysts for Solar Water Splitting Devices. J. Am. Chem. Soc 2015, 137(13), 4347–4357. [DOI] [PubMed] [Google Scholar]
- 3.Weekes DM; Salvatore DA; Reyes A; Huang A; Berlinguette CP Electrolytic CO2 reduction in a flow cell. Acc. Chem. Res 2018, 51(4), 910–918. [DOI] [PubMed] [Google Scholar]
- 4.Elgrishi N, Rountree KJ, McCarthy BD, Rountree ES, Eisenhart TT, & Dempsey JL A Practical Beginner’s Guide to Cyclic Voltammetry. J. Chem. Educ 2018, 95, 197–206. [Google Scholar]
- 5.Gagne RR; Koval CA; Lisensky GC Ferrocene as an Internal Standard for Electrochemical Measurements. Inorg. Chem 1980, 19, September, 2854–2855. [Google Scholar]
- 6.Ruiz Aranzaes J; Daniel M-C; Astruc D Metallocenes as References for the Determination of Redox Potentials by Cyclic Voltammetry - Permethylated Iron and Cobalt Sandwich Complexes, Inhibition by Polyamine Dendrimers, and the role of Hydroxy-Containing Ferrocenes. Can. J. Chem 2006, 84, 288–299. [Google Scholar]
- 7.Malapit CA; Prater MB; Cabrera-Pardo JR; Li M; Pham TD; McFadden TP; Blank S; Minteer SD Advances on the Merger of Electrochemistry and Transition Metal Catalysis for Organic Synthesis. Chem. Rev 2022, 122, 3180–3218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Trindell JA; Duan Z; Henkelman G; Crooks RM Well-Defined Nanoparticle Electrocatalysts for the Refinement of Theory. Chem. Rev 2020, 120, 814–850. [DOI] [PubMed] [Google Scholar]
- 9.Gritzner G; Murauer H; Gutmann V Solvent and Salt Effects on the Redox Behaviour of Trisacetylacetonato Manganese(III). J. Electroanal. Chem 1979, 101, 185–200. [Google Scholar]
- 10.Tocher JH; Fackler HP Jr. Electrochemical Investigations of Several Transition Metal Tris(acetylacetonate) Complexes. Inorganica Chim. Acta 1985, 102(2), 211–215. [Google Scholar]
- 11.Beaver BD; Hall LC; Lukehart CM; Preston LD Reactions of Coordinated Molecules. XXVII. Cyclic Voltammetry of Several Transition Metal Metalla-acetylacetonate Complexes. Inorganica Chim. Acta 1981, 47, 25–30. [Google Scholar]
- 12.Richert SA; Tsang PKS; Sawyer DT Ligand-Centered Redox Processes for MnL3, FeL3, and CoL3 Complexes (L = Acetylacetonate, 8-Quinolinate, Picolinate, 2,2’-Bipyridyl, 1,10-Phenanthroline) and for Their Tetrakis(2,6-dichlorophenyl)porphinato Complexes [M(Por)]. Inorg. Chem 1989, 28, 2471–2475. [Google Scholar]
- 13.Wilson JN The Dielectric Constants of Polar Liquids. Chem. Rev 1939, 25, March, 377–406. [Google Scholar]
- 14.Gaussian 16, Revision C.01, Frisch MJ; Trucks GW; Schlegel HB; Scuseria GE; Robb MA; Cheeseman JR; Scalmani G; Barone V; Petersson GA; Nakatsuji H; Li X; Caricato M; Marenich AV; Bloino J; Janesko BG; Gomperts R; Mennucci B; Hratchian HP; Ortiz JV; Izmaylov AF; Sonnenberg JL; Williams-Young D; Ding F; Lipparini F; Egidi F; Goings J; Peng B; Petrone A; Henderson T; Ranasinghe D; Zakrzewski VG; Gao J; Rega N; Zheng G; Liang W; Hada M; Ehara M; Toyota K; Fukuda R; Hasegawa J; Ishida M; Nakajima T; Honda Y; Kitao O; Nakai H; Vreven T; Throssell K; Montgomery JA Jr.; Peralta JE; Ogliaro F; Bearpark MJ; Heyd JJ; Brothers EN; Kudin KN; Staroverov VN; Keith TA; Kobayashi R; Normand J; Raghavachari K; Rendell AP; Burant JC; Iyengar SS; Tomasi J; Cossi M; Millam JM; Klene M; Adamo C; Cammi R; Ochterski JW; Martin RL; Morokuma K; Farkas O; Foresman JB; Fox DJ; Gaussian, Inc., Wallingford CT, 2016. [Google Scholar]
- 15.Bredas J Mind the Gap! Mater. Horiz 2014, 1, 17–19. [Google Scholar]
- 16.Orlik M; Gritzner G Standard Electrode Reaction Rates of bis(biphenyl)chromium(I) in N,N-dimethylformamide. J. Electroanal. Chem 1997, 421, 121–128. [Google Scholar]
- 17.Asri Nawi M; Riechel TL Electrochemical Studies of Vanadium(III) and Vanadium(IV) Acetylacetonate Complexes in Dimethyl Sulfoxide. Inorg. Chem 1981, 20, 1974–1978. [Google Scholar]
- 18.Sampson MD; Nguyn AD; Grice KA; Moore CE Rheingold AL; Kubiak CP; Manganese Catalysts with Bulky Bipyridine Ligands for the Electrocatalytic Reduction of Carbon Dioxide: Eliminating Dimerization and Altering Catalysis. J. Am. Chem. Soc 2014, 136, 5460–5471. [DOI] [PubMed] [Google Scholar]
- 19.Chung Y; Lee C-W Electrochemistry of Gallium. J. Electrochem. Sci. Technol 2013, 4(1), 1–18. [Google Scholar]
- 20.McEvoy JP; Brudvig GW Water-Splitting Chemistry of Photosystem II. Chem. Rev 2006, 106(11), 4455–4483. [DOI] [PubMed] [Google Scholar]
- 21.Fernandes GFS; Pontes MAP; Machado FBC; Ferrao LFA Electronic Structure and Stability of Transition Metal Acetylacetonates TM (AcAc)n (TM = Cr, Fe, Co, Ni, Cu; n= 1, 2, 3). Comput. Theor. Chem 2022, 1207, 113502–113511. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.