Manoeuvring organo-electrocatalytic selective CO₂ reduction to CO by terpyridine derivatives: DFT mechanistic exploration

Abstract

The field of sustainable energy is rapidly growing, with electro-catalytic CO₂ reduction becoming a key focus. This study examines the effectiveness of substituted terpyridine derivatives (L1–L6) as electrocatalysts for CO₂ reduction in a DMF and H₂O mixture. Using polypyridyl ligands for CO₂ electrochemical transformation into valuable products is relatively rare. Our investigation shows that terpyridine ligands are highly effective in converting CO₂ into CO and H₂via a 2H⁺/2e⁻ reduction process, with an overpotential of 650 to 850 mV vs. SCE. Notably, catalyst L6 demonstrated impressive faradaic efficiency (FE) for CO and H₂, with 74% and 0.04% at −1.8 V vs. SCE, respectively, after 3 hours of electrolysis. All these catalysts, L1–L6, exhibit over 90% selectivity for CO generation. The DFT studies reveal that the mechanism follows EECC pathways, where ‘E’ represents electrochemical steps and ‘C’ denotes chemical steps.

---

Introduction

The growing global demand for energy, primarily from non-renewable fossil fuels, contributes to air pollution and environmental toxins by continuing to emit carbon dioxide.^1–5 Since CO₂ is a thermodynamically inert molecule, its high overpotential makes reduction extremely difficult. On the other hand, electrochemical CO₂ reduction has emerged as a widely used and economically feasible technique for converting greenhouse gases into useful commodities and fuels and using intermittent renewable energy.^6–8 Reducing CO₂ to other chemicals, such as CO, (HCOO⁻/HCOOH), HCHO, CH₃OH, and CH₄, requires complex multi-electron processes.^9–11 Molecular homogeneous catalysts are more selective than heterogeneous ones. They are essential catalysts due to this molecular property. Ruthenium, rhenium, rhodium and iridium-based catalysts have shown great promise in electrocatalytic CO₂ reduction.^12–15 However, from an economic point of view, the practical use of noble metal complexes is not viable for electrocatalytic CO₂ reduction. Due to the growing demand for sustainable catalysis, we must focus on developing molecular catalysts that use readily available components to create a sustainable and economically feasible future.¹⁶ Metal-free electrocatalysts, composed solely of organic compounds devoid of metal centers, present distinct benefits in cost, abundance, and environmental footprint. These catalysts frequently demonstrate a wide range of functional groups that can enhance the electrochemical reduction of CO₂ with exceptional selectivity. A promising path for sustainable chemical transformations has been previously described by certain groups that used organic catalysts, such as pyridinium salts and dihydropyridines (DHPs), to electrochemically reduce carbon dioxide.¹⁷ The ability of benzoate derivatives to reduce CO₂ was examined by Gambino et al. with faradaic efficiencies of up to 99%; the study, which was carried out at a mercury (Hg) pool electrode in 0.1 M TBAP in DMF, found that monocarboxylic acid was the main result from benzoates along with dicarboxylic acid formation.¹⁸ Bocarsly and his team made a major breakthrough in the electrochemical reduction of CO₂ in 1994 when they showed that the pyridinium ion could catalyze the creation of methanol on a hydrogenated platinum electrode. With a respectable faradaic efficiency of 30%, the investigation demonstrated the active involvement of the 1e⁻ reduced species originating from the pyridinium ion. Significantly, the absence of methanol formation occurred when pyridinium was substituted with N-methylpyridinium ions, highlighting the unique catalytic behaviour linked to various pyridinium derivatives in the electrochemical reduction of CO₂.¹⁹ Aromatic nitriles and esters have the ability to preferentially convert CO₂ to oxalate, which makes them promising homogeneous catalysts for CO₂ reduction in aprotic environments, as later revealed by Gennaro et al. The catalytic activity is investigated by cyclic voltammetry, which shows nucleophilic addition to CO₂ that creates a bond that regenerates the catalyst.²⁰ With 0.5 M KCl at pH = 4.7, Portenkirchner et al. carried out a study showing how pyridine and pyridazine aid in the reduction of CO₂. With a faradaic efficiency of 14% for the pyridine system and 3.6% for the pyridazine system, methanol was the final product that was separated from this process.²¹ Choudhury and his colleagues recently proposed a novel family of dicationic chemical compounds (PAC1–PAC4) as possible NADP⁺ mimics. By catalyzing electrochemical CO₂ reduction (eCO₂RR) to formic acid, these chemicals can be electrochemically reduced to produce NADPH mimics. Up to 35 ± 3% faradaic efficiency was attained with the PAC3 system. This marks a significant advancement in the use of metal-free organic catalysts to move from sub-stoichiometric to catalytic eCO₂RR processes.²²

A viable substitute that addresses issues with metal scarcity and environmental effects is the use of metal-free organocatalysts. Sustainable electrochemical CO₂ reduction will be developed as a result of improving the performance of metal-free organocatalysts and comprehending their intricate mechanisms. The continuous research into metal-free and metal-complex catalysts highlights the multidisciplinary initiatives meant to reduce the negative effects of CO₂ emissions on the environment while enhancing the potential of green and sustainable chemistry. Expanding on a similar design concept, we provide a novel class of organic CO₂ reduction catalysts, [R-Tpy], where R is a phenyl derivative and Tpy is 4′-methyl-2,2′:6′,2′′-terpyridine. Since the terpyridine framework has a large π-conjugation, it is simple to add different functional groups to adjust its structural and electrical characteristics. Through multi-electron transfer and improved catalytic activity and selectivity, this versatility improves its capacity to stabilize radical intermediates during CO₂ reduction. The electrochemical study involves featuring the general formula [R-Tpy], incorporating six unique terpyridine derivatives: [Cl-Tpy] (L1), [OMe-Tpy] (L2), [Ph-Tpy] (L3), [QCl-Tpy] (L4), [Me-Tpy] (L5), and [CN-Tpy] (L6), where QCl and Ph represent 3-chloroisoquinoline and phenyl groups, respectively. These compounds display remarkable selectivity towards the reduction of CO₂ to CO, along with a negligible amount of H₂ generation (Scheme 1).

Scheme 1 Organo-electrocatalysis of CO₂ to CO by R-Tpy.

Results and discussion

Synthesis and characterization of catalysts

Syntheses. The catalysts [R-Tpy], [Cl-Tpy] (L1), [OMe-Tpy] (L2), [Ph-Tpy] (L3), [QCl-Tpy] (L4), [Me-Tpy] (L5), and [CN-Tpy] (L6), were synthesized using a reported procedure (Scheme 2).^23,24 Terpyridine is a tridentate redox-active chelating ligand, which was initially isolated by Morgan and Burstall.^23–26 We have synthesized six terpyridine ligands (L1–L6), with different substitutions at the 4′ position. The structure of the synthesized catalysts was confirmed through spectroscopic analyses, including ¹H NMR, ¹³C NMR, and mass spectrometry (Fig. S2–S19).

Scheme 2 Reaction scheme of Tpy derivatives (L1–L6).

Electrochemical studies

The electrochemical properties of the catalysts were investigated by cyclic voltammetry using a three-electrode system with a glassy carbon electrode used as the working electrode, a calomel electrode as the reference, and a Pt wire as the counter electrode, and 0.1 M TBAP (tetra butyl ammonium perchlorate) was used as a supporting electrolyte under an inert atmosphere in a dry DMF solution. The voltammograms of all the ligands (L1–L6) are shown in Fig. 1. In L1, there is one quasi-reversible peak at E_1/2 = −1.78 (E_p,a = −1.71, E_p,c = −1.85) V vs. SCE, and other than that, there are three cathodic peaks at −1.44 V, −1.65 V, and −2.2 V vs. SCE and one anodic peak at −2.1 V vs. SCE. All these peaks were ascribed to the Tpy/Tpy˙⁻ reduction process (Fig. 1).

Fig. 1 (A)–(F) CVs of 1 mM L1–L6 in dry DMF using 0.1 M TBAP as a supporting electrolyte under an N₂ atmosphere.

The other three ligands L2, L3, and L5 show similar behavior, one sharp reduction peak appeared at −2.0 V, with two oxidation peaks at E_p,a = −1.93 V and −1.72 V, E_p,a = −1.86 V and −1.65 V, and E_p,a = −1.87 V and −1.69 V concerning L2, L3, and L5, respectively. These three ligands showed similar nature in CV due to the presence of electron donating groups OMe- and Me- in the ligand scaffold (Fig. 1). On the other hand, L4 shows three reduction peaks and two oxidation peaks at E_p,c = −1.68 V, −1.76 V, and −2.1 V, and E_p,a = −1.98 V and −1.63 V, respectively. However, L6 exhibits four reduction and oxidation peaks because of the CN moiety present in L6 at E_p,c = −1.51 V, −1.65 V, −1.85 V, and −2.07 V, and E_p,a = −1.96 V, −1.7 V, −1.5 V, and −1.31 V, respectively (Fig. 1). Further, we have checked the number of electron transfers in this electrochemical process. The coulometry of catalysts L1, L4, and L6 at −2.0 V and L2, L3, and L5 at −2.1 V vs. SCE reveals that two electrons are involved during electrochemical conversion (as shown in Fig. S21–S26). Moreover, the non-terpyridine-based ligand 2,2′-bipyridine (bpy) exhibits three cathodic-based peaks at E_p,c = −1.86 V, −2.2 V, and −2.54 V, and three anodic-based peaks at E_p,a = −2.0 V, −0.76 V, and −0.3 V. Similarly, 1,10-phenanthroline (phen) shows two cathodic-based peaks at E_p,c = −1.68 V and −2.02 V, and two anodic-based peaks at E_p,a = −0.6 V and −0.25 V (Fig. S27). A minimal increase in catalytic current has been noted when 2.2 M H₂O has been added to the reaction mixture under an N₂ atmosphere. Upon increasing the water concentration beyond 2.2 M, the catalytic current exhibits no further enhancement and remains essentially constant. This observation suggests that the electrode surface has reached a saturation limit with respect to proton availability, such that additional water does not contribute to an increase in proton-coupled electron transfer kinetics (Fig. S28). Moreover, it has been noted that catalytic current saturates rapidly under an N₂ atmosphere upon addition of a small amount of phenol (PhOH) as the proton source indicating faster saturation of proton availability at the electrode surface (Fig. S29). It has been noted that peaks become shifted cathodically when potential has been measured with respect to ferrocene. The organo-catalyst L3 exhibits one sharp reduction peak appearing at −2.49 V and two oxidation peaks at E_p,a = −2.35 V and −2.1 V vs. Fc^+/0 (Fig. S30C).

Electrochemical CO₂ reduction

The evaluation of catalysts for electrochemical CO₂ reduction centered on using terpyridine derivatives L1–L6. The experimental conditions involved a solution of DMF saturated with CO₂ gas at a concentration of approximately 0.20 M. A critical aspect of this process was the hydration of CO₂, resulting in carbonic acid H₂CO₃. The generation of H₂CO₃ and its subsequent role as a proton donor were essential for facilitating the CO₂RR.^27,28 As shown in the cyclic voltammograms of catalysts L1–L6 recorded in CO₂-saturated DMF solution, the cathodic peak at −1.85 V shifts to −1.64 V vs. SCE for L1 (Fig. 2(A)), the peak at −2.0 V shifts to −1.76 V vs. SCE for L2 (Fig. 2(B)), the peak at −1.95 V shifts to −1.78 V vs. SCE for L3 (Fig. 2(C)), the peak at −1.7 V shifts to −1.61 V vs. SCE for L4 (Fig. 2(D)), the peak at −1.97 V shifts to −1.80 V vs. SCE for L5 (Fig. 2(E)), and the peak at −1.79 V shifts to −1.65 V vs. SCE for L6 (Fig. 2(F)). However, the peak shifts more cathodically (E_p,c = −2.35 V vs. Fc^+/0) for L3 when measured with respect to ferrocene as a reference under a CO₂ atmosphere. The addition of 2.2 M H₂O as a proton source shifts the cathodic potential to E_p,c = −2.30 V vs. Fc^+/0 when measured under a CO₂ atmosphere (Fig. S30D).

Fig. 2 (A)–(F) CVs of 1 mM catalysts L1–L6 in DMF using 0.1 M TBAP as a supporting electrolyte under N₂ and CO₂ atmospheres and CO₂ with 2.2 M H₂O.

The 100–300 mV positive shift for L1–L6 in the first cathodic process in the presence of CO₂ exhibited cathodic current enhancement, as illustrated in Fig. 2. The mid-wave potential for this electrochemical process was determined to be E_1/2 = −1.73 V, −1.71 V, −1.69 V, −1.78 V, −1.73 V, and −1.87 V vs. SCE for L1–L6, respectively (Fig. S31). The L3 catalyst required less potential, and L6 had a higher potential than the other catalysts during the CO₂RR. Cyclic voltammetry experiments were conducted under several conditions to elucidate further the electrocatalytic process, including N₂, CO₂, and CO₂ with 2.2 M H₂O in DMF. In each experiment, CO₂ saturation was achieved after a 15 minute purge, revealing a moderate increase in catalytic current in the presence of 2.2 M H₂O (Fig. 2).²⁹ Moreover, upon CO₂ purging a very minimal increase in catalytic current has been noted for both bpy and phen (Fig. S27B and D), implying that both catalysts are not effective for CO₂ reduction as compared to terpyridine ligands. The absence of reactivity with bpy and phen can be attributed to their distinct electronic and structural features compared to terpyridine. Terpyridine, with three pyridyl units, provides stronger stabilization of radical/cationic intermediates and a more favorable geometry for transition-state organization. In contrast, bpy (bidentate) and phen (rigid, sterically congested) lack the same electronic and conformational flexibility, preventing efficient stabilization of the reactive intermediates. Addition of a small amount of a strong proton source such as PhOH leads to faster saturation of catalytic current (Fig. S32) in contrast to H₂O (Fig. 2(A)–(F)).

Controlled potential electrolysis

Controlled potential electrolysis (CPE) was conducted at a glassy carbon plate used as a working electrode with a surface area of 1.0 cm², utilizing 0.05 mM catalysts L1–L6 in a CO₂-saturated DMF/H₂O mixture and employing a custom-made electrolyzer with 0.1 M TBAP as the supporting electrolyte. The CPE experiments, spanning three hours or until complete catalyst deactivation, revealed a noteworthy pattern in charge. The potential was systematically varied from −1.6 V to −1.9 V vs. SCE, revealing an increased charge build-up with higher applied potentials (Fig. S33). Upon conducting the rinse test after 3.0 h of electrolysis, followed by subsequent CPE without a catalyst at −1.9 V vs. SCE, no gaseous products were evolved during the bulk electrolysis of blank solution (electrolysis without using a catalyst) or in the rinse test (Fig. S34), which was further confirmed by gas chromatography. GC was employed to analyze the products formed during the electrolysis process. The analysis confirmed the presence of CO₂-reducing products such as CO and H₂. To validate the identification of CO and H₂ produced by CO₂ reduction, standard gases of CO and H₂ were injected into the GC system to determine their retention times (Fig. S35). Gas chromatography plots for samples labelled L1 to L6 at a specific potential of −1.6 V vs. SCE are provided in Fig. S36–S41. Moreover, the GC plots for detection of CO and H₂ using bpy and phen ligands are shown in Fig. S42. During the electrolysis process, gaseous samples were collected using a gas-tight syringe after 3.0 h of electrolysis. The collected samples were then injected into the gas inlet of the GC for analysis. CO and H₂ production was quantified at each potential during electrolysis using response factors. The quantities of CO and H₂ produced at a potential of −1.7 V vs. SCE are depicted in Fig. 3, providing a visual representation of the amount of each gas generated using organocatalysts L1–L6.

Fig. 3 Amount of CO (A) and H₂ (B) produced after 3.0 h of CPE at −1.7 V vs. SCE of L1–L6.

Further, the gaseous products were confirmed using an Omni Star™ Gas Analysis System GSD 320 (Pfeiffer) quadrupole mass spectrometer. Isotopic labelling experiments for catalysts L1–L6 were carried out using an online mass analyzer, employing both ¹²CO₂ and ¹³CO₂ isotopes.³⁰ During controlled potential electrolysis (CPE), the gas analyzer's sensor was connected to the headspace of the electrochemical cell. The detected products included hydrogen (H₂, m/z = 2), ¹²CO (m/z = 28), and ¹³CO (m/z = 29) in the reaction under ¹²CO₂ and ¹³CO₂ atmospheres for the L1–L6 catalysts (Fig. S43–S47). A typical online mass and gas measurement during the reduction of ¹³CO₂ by catalysts L1 (Fig. 4(A)–(C)) and L4 (Fig. 4(D)–(F)) is depicted in Fig. 4. The highest faradaic efficiencies for CO and H₂ were 59.0% and 0.02% for L6 at −1.6 V vs. SCE after 3 hours of electrolysis (Table 1).

Fig. 4 The CPE under ¹³CO₂ using L1 as a catalyst: (A) H₂ detection, (B) ¹³CO detection, (C) and ¹³CO₂ detection using an online gas and mass analyzer. The CPE under ¹³CO₂ using L4 as a catalyst: (D) H₂ detection, (E) ¹³CO detection, and (F) ¹³CO₂ detection using an online gas and mass analyzer.

Table 1 Catalytic comparison for CO₂ reduction among L1–L6 at −1.6 V vs. SCE

Catalyst	CO selectivity (%)	FE for CO (%)	H2 selectivity (%)	FE for H2 (%)
L1	99.6	44.0	0.500	0.20
L2	99.9	22.0	0.046	0.01
L3	99.8	27.0	0.140	0.04
L4	99.9	29.0	0.070	0.02
L5	99.5	46.0	0.050	0.02
L6	99.9	59.0	0.0340	0.02

---

The faradaic efficiency, TON, and TOF for CO and H₂ of all the catalysts (L1–L6) after 3 h of CPE at varied potentials of −1.6 V to −1.9 V vs. SCE are summarized in Tables S2–S13 and those of bpy and phen are summarized in Tables S14 and S15 (detailed calculations are given in the SI).

Catalytic stability

The stability of catalysts was essential to ensure the efficiency and durability of the CO₂RR process. A stable catalyst can maintain its activity over extended periods of operation. In this context, UV-vis spectroscopy was employed to analyze 0.5 mM L1–L6 before and after bulk electrolysis (as shown in Fig. S10A), indicating no significant changes in the spectra (Fig. S48). This suggests that the L1–L6 organocatalysts remained stable under the catalytic conditions of CO₂ reduction. After 3 hours of electrolysis, a rinse test was conducted for all the catalysts (Fig. S34), confirming both the homogeneous nature of the catalytic conditions and the stability of the L1 to L6 organocatalysts throughout the process. Further, we have taken CVs before and after bulk electrolysis. No spectral changes were observed during the process (Fig. S49). All the catalysts exhibited stability throughout the catalytic process, confirming their reliability in the CO₂RR. Furthermore, the homogeneity of the catalysts was validated using ESI mass spectrometry (Fig. S50–S55). The obtained spectra were consistent with the mass spectra recorded prior to electrolysis, confirming the stability of the catalysts.

Furthermore, cyclic voltammetry performed over 30 consecutive cycles under both N₂ and CO₂ atmospheres did not reveal the appearance of any new peaks for any of the ligands (Fig. S57, S60 and S62), further supporting their stability and homogeneity. Under an N₂ atmosphere, the catalytic current remains constant after the addition of a proton source over successive cycles. This clearly indicates that the H₂O concentration remains constant during the repetitive cycles. In contrast, when the solution is purged with CO₂, a gradual decrease in the catalytic current is observed upon repetitive cycles (Fig. S60 and S62), which we attribute to the saturation of CO₂ at the electrode surface.

Kinetic parameters and catalytic efficiency

We investigated the catalytic kinetics of catalysts L1–L6, examining the impact of variables such as catalyst concentration and water content under different conditions. The study demonstrated a linear relationship between the inverse square root of the scan rate (ν^−1/2) and the ratio of catalytic peak current under CO₂ (i_c) to peak current under N₂ (i_p), represented as (i_c/i_p) (Fig. S56–S62). At higher scan rates, a negative shift in the plateau potential was observed (Fig. S56, S58, S59 and S61). To further probe the catalytic mechanism, scan rate-dependent studies were performed (Fig. S56, S58, S59 and S61). In both N₂ and CO₂ atmospheres, no saturation of the catalytic current was observed, even at high scan rates, suggesting rapid and efficient electron transfer to CO₂. Under N₂, only a minimal increase in current was detected with an added proton source, in line with gas chromatography (vide infra), which confirmed negligible H₂ evolution. Furthermore, the ratio of catalytic to peak current (i_c/i_p) decreased with increasing scan rate from 200–2000 mV s⁻¹ (Fig. S63), indicating diffusion-limited catalysis, as expected for a system where substrate (CO₂) mass transport governs the overall catalytic current.

To study the kinetics of electrocatalytic CO₂ reduction, we analyzed the cyclic voltammograms of catalysts L1–L6 with varying catalyst concentrations. We found that the rate of CO₂ reduction increased linearly with higher catalyst concentrations (Fig. 5 and S64). The relationship between the catalytic current (i_cat) and its components is given by the equation i_cat = n_catFA[cat](Dk_CO2[CO₂])^1/2, where n_cat = 2 for CO₂ reduction to CO, F = 96 500 C mol⁻¹, A = 0.071 cm², D is the diffusion coefficient, k_CO2 is the rate constant, and [CO₂] is the CO₂ concentration in DMF. Moreover, there is a very negligible shift in the onset potential upon raising the catalyst concentration from 0.25 mM to 1.25 mM under a CO₂ atmosphere (Fig. 5). Additionally, the current increased with the addition of H₂O, which acts as a proton source in CO₂ reduction. The choice of proton source significantly affects the efficiency, selectivity, and overall performance of the CO₂ reduction reaction. Acidic environments help protonate intermediates like H₂CO₃, speeding up the reaction rates and optimizing CO₂ conversion into desired products.

Fig. 5 (A)–(F) CVs of catalysts L1–L6 in DMF using 0.1 M TBAP as a supporting electrolyte under N₂ (only 0.25 mM catalysts) and CO₂ atmospheres with 2.2 M H₂O [varying catalyst concentrations from 0.25 mM to 1.25 mM].

Foot of the wave analysis and catalytic Tafel plots

Investigating catalytic systems systematically through the utilization of turnover frequency and overpotential presents a significant challenge due to various factors, including intrinsic catalytic activities, dependence on transport parameters, and ohmic drops. Savéant et al. introduced a logical and non-destructive technique known as foot-of-the-wave analysis (FOWA), which is applied to routine cyclic voltammetry (CV) observations to estimate TOF quickly. This method focuses solely on the catalytic reaction rate in the electrode's reaction–diffusion layer, disregarding contributions from other phenomena like product inhibition, substrate consumption, and catalyst deactivation.³¹

The analytical approach from cyclic voltammetry (CV) measurements involves comparing data obtained in the absence and presence of CO₂, relying on establishing a linear relationship between the current ratio (i/i_p) and a potential-dependent term expressed as 1/{1 + exp[(nF/RT)(E − E_cat/2)]}. Here, i represents the catalytic current when CO₂ is present, while i_p denotes the peak current in the absence of CO₂. Furthermore, F denotes the Faraday constant, R is the gas constant, T is the absolute temperature, n is the number of electrons required for CO₂ to CO conversion, E is the applied potential, and E_cat/2 is the wave catalytic current. Plotting these terms against each other in the foot-of-the-wave region of the CV yields a linear relationship with a specific slope, which equals the expression 2.24(RT/nFν)^1/2(k_FOWA)^1/2, where ν represents the scan rate in V s⁻¹. From this slope, the catalytic rate constant, k_FOWA, or TOF_max, for L1–L6 was found to be 1105 s⁻¹, 724 s⁻¹, 1589 s⁻¹, 2211 s⁻¹, 1197 s⁻¹ and 2511 s⁻¹, respectively (Fig. S65).

The subsequent analysis revealed slope values and their respective catalytic rate constants for L1–L6, indicating that L6 and L4 exhibit higher catalytic rates than the others. This difference can be attributed to the electron density, steric effects, and redox properties of L6 and L4 likely contributing to their superior catalytic activity towards electrochemical CO₂ reduction compared to the other terpyridine derivatives. In-depth analysis involves the correlation of the turnover frequency, a kinetic parameter, with the thermodynamic overpotential (η) derived from the standard reduction potential for the CO₂ reduction process and the applied potential (E). This relationship is expressed as . By plotting TOF against overpotential, a catalytic Tafel plot is generated, providing valuable insights into the electrochemical CO₂ reduction process. In the investigation, the onset potential (E_onset) for L1–L6 was determined to be −1.46 V, −1.41 V, −1.43 V, −1.46 V, −1.42 V, and −1.45 V vs. SCE, respectively. Additionally, the overpotential (η) and the catalytic half-wave potential (E_cat/2) were evaluated for each complex. These parameters provide crucial information for understanding each catalyst's catalytic activity and performance in electrochemical CO₂ reduction (Table 2).

Table 2 List of onset potential, half-wave potential, TOF_max and overpotential of L1–L6

Catalyst	E onset (V vs. SCE)	E 1/2 (V vs. SCE)	Slope	TOFmax (s^−1)	Overpotential (η/mV)
L1	−1.46	−1.73	27	1105	730
L2	−1.41	−1.72	22	724	715
L3	−1.43	−1.68	32	1589	687
L4	−1.46	−1.78	38	2211	780
L5	−1.42	−1.73	28	1197	730
L6	−1.45	−1.87	40	2511	870

---

For L1–L6, the overpotentials were 730 mV, 715 mV, 687 mV, 780 mV, 730 mV, and 870 mV, respectively. Correspondingly, the catalytic half-wave potentials (E_1/2) were determined to be −1.73 V, −1.72 V, −1.68 V, −1.78 V, −1.73 V, and −1.87 V vs. SCE for L1–L6, respectively.

Utilizing catalytic Tafel plots as a benchmarking tool offers valuable insights into the intrinsic catalytic activities of various catalysts, aiding in identifying optimal operating conditions. The primary objective is to improve the catalytic rate (TOF) while minimizing overpotential, thus enhancing the efficiency of the electrochemical CO₂ reduction process. An ideal catalyst is anticipated to demonstrate a catalytic Tafel plot that is shifted diagonally to the upper left, indicating superior performance and efficiency (Fig. 6). Upon comparing the catalytic Tafel plots for each catalyst within a single diagram, it becomes evident that L3 is suitable for CO₂ reduction.

Fig. 6 The catalytic Tafel plots of L1–L6 for the comparison of catalytic activity.

The overpotential associated with L3 is measured to be 687 mV, suggesting that the multifaceted molecular structure of L3 endows it with a unique combination of properties that collectively enhance its catalytic activity and selectivity in electrochemical CO₂ reduction. This compound represents a promising avenue for developing efficient CO₂ conversion catalysts, contributing to advancing sustainable energy technologies.

Spectro-electrochemical studies

Investigating terpyridine derivatives for electrochemical CO₂ reduction involves a comprehensive UV-vis spectroelectrochemistry study. This technique allows for real-time monitoring of the electronic changes occurring in the catalysts during the reduction process. In this study, UV-SEC was employed to analyze the behaviour of Tpy derivatives (L1–L6) during CO₂ reduction. The experiments were conducted in a DMF and H₂O mixture solution, using 0.0156 mM (L1–L6) with 0.1 M TBAP as a supporting electrolyte under N₂ and CO₂ atmospheres at −1.9 V vs. SCE (Fig. 7 and S66). Significant changes in the UV-vis spectra of the Tpy derivatives were observed after applying the potential during CPE. Initially, the Tpy derivatives exhibited characteristic absorption bands corresponding to their electronic transitions. During the reduction process, these bands underwent notable shifts and intensity changes, indicating the progression of the catalytic reaction. We noticed during UV-SEC that all the catalysts show a new band at around 310–380 nm, and the molar extinction value appears at around 3000 M⁻¹ cm⁻¹ for both conditions. This happened due to the 2e⁻ reduced R-Tpy transferring electrons to the CO₂ molecule, and charge transfer occurred from the terpyridine nitrogen atom to CO₂ due to n → π* transition. The appearance of new absorption bands and the shifting of existing bands provided insights into the mechanistic pathways of CO₂ reduction by the terpyridine derivatives.

Fig. 7 The spectral changes during spectro-electrochemical studies for 0.0156 mM under a CO₂ atmosphere at −1.9 V vs. SCE using 0.1 M TBAP as a supporting electrolyte for (A) L1, (B) L2, (C) L3, (D) L4, (E) L5, (F) L6.

Infrared spectroelectrochemistry is a powerful tool for investigating the mechanistic pathways of electrochemical CO₂ reduction. Organocatalysts have shown promise in CO₂ reduction due to their ability to facilitate multi-electron transfer steps and stabilize reactive intermediates. This technique helps to track the formation and consumption of CO₂-derived species, such as carbonates, carboxylates, or carbon monoxide. In situ FT-IR spectra were recorded for all organocatalysts (L1–L6) using a high-sealing, optically transparent thin-layer electrochemical (OTTLE) cell equipped with CaF₂ windows. A Pt mesh was employed as the working electrode, a Pt wire as the auxiliary electrode, and an Ag wire as the pseudo-reference electrode. The measurements were carried out with a 0.5 mM solution of the complex in 4.9 mL DMF containing 0.1 mL H₂O and 0.1 M TBAP. Prior to data collection, the solution was purged with N₂ followed by CO₂ for 15 minutes. Controlled potential electrolysis was performed at −1.8 V vs. SCE. All the organocatalysts L1–L6 exhibit bands that are indicative of C=O stretching vibrations in protonated carboxylic acid species (N–CO₂H) progressively appearing at υ_str = 1700–1725 cm⁻¹ corresponding to the formation of N–CO₂H species (Fig. 8 and S67).³²

Fig. 8 ΔTransmittance vs. wavenumber plots, where the initial spectra before electrolysis have been subtracted from the observed spectra during electrolysis for 0.5 mM (A) L3 (red initial and blue final) and (B) L5 (blue initial and red final) under a CO₂ atmosphere at −1.8 V vs. SCE using 0.1 M TBAP as a supporting electrolyte.

Density functional theory (DFT) analysis

DFT³³ calculation based on Gaussian 16 (ref. 34) was performed to gain insight into the mechanism of the H₂ formation and CO₂ reduction reaction and the effect of different electron donating and withdrawing functional groups during the reaction. Geometry optimization and frequency calculations of all intermediates and transition states were performed using the MO6L functional.³⁵ For instance, the formation of intermediate species could be inferred from the emergence of new bands in the UV-vis spectra. This approach underscores the potential of terpyridine derivatives as effective catalysts for CO₂ reduction, contributing to the advancement of sustainable energy technologies.

Diffusion and polarization functions were utilized along with the Pople basis set^36,37 6-31+G(d,p) to optimize all the geometries. The solvent effect of the reaction was considered by the self-consistent reaction field approach using the SMD solvation model³⁸ in DMF solvent. The computational details are provided in the SI. In the electrochemical CO₂ to CO formation, the reaction proceeds via several proton and electron transfer steps; hence, it is crucial to consider experimental potential and pH during the calculation of free energy and reduction potential.³⁹ In this study, all free energy profile diagrams were constructed by implementing the experimental pH of 7.4 and applied potential of −1.60 V (vs. SCE). The H₂ formation and CO₂ reduction reactions were studied in the presence of six different reagents (L1 to L6). The possible reaction mechanism for the H₂ formation reaction is shown in Scheme 3. The optimized structures for the intermediates are provided in Fig. S68.

Scheme 3 Possible reaction mechanism of the H₂ formation reaction on different catalysts at pH = 7.4 and an applied potential of −1.60 V (SCE) (all values are in kcal mol⁻¹).

The H₂ formation reaction generally proceeds through two electron and two proton addition reactions. Here, we consider two consecutive reduction reactions followed by two proton addition paths. The first reduction reaction of Int-1 is exogenic in nature and generates Int-2. The potential corrected free energy changes for the reaction vary from 1.9 kcal mol⁻¹ to −8.4 kcal mol⁻¹. The free energy of reduction of the para methoxybenzene derivative is the least exogenic with 1.9 kcal mol⁻¹, and para cyano-benzene is the most exogenic. The result indicates that the electron-withdrawing substitution in the R group (like L6, L1, and L4) accelerates the reduction compared to the electron-donating substitution (L5 and L2). Int-2 is reduced to Int-3 in the subsequent reduction step, and the second reduction reaction is less exogenic than the first one. As expected, the second reduction reaction follows a similar substitution effect on reduction energy trends. In the protonation reaction, proton addition takes place at the N center to make a N–H bond in Int-4 with a 1.01 Å bond distance. All calculated pK_a values of each protonation step are provided in Table S16.

Interestingly, the dehydrogenation reaction is more favourable for the electron-donating group than for the electron-withdrawing group. All transition states are confirmed by IRC calculation (Fig. S69). The reaction may proceed in another path through a nine-membered TS-2 transition state. In all cases, the activation energy viaTS-2 is more than the reaction viaTS-1. The reaction viaTS-1 is the most favourable reaction path (Fig. S70).

The DFT mechanistic study indicates that the most favourable path for the two proton and two electron addition reaction proceeds viaInt-5 protonation at two adjacent N-centers. The substrates are crucial in protonation, reduction, and activation energy in dehydrogenation reactions. Protonation at para methoxybenzene is the most favourable, with a reaction energy of −41.9 kcal mol⁻¹. In the second protonation reaction, protonation occurs at adjacent N to make Int-5 or at the third nitrogen to make Int-6. In all the cases, the protonation at the adjacent N is more feasible than that at the third nitrogen atom (Table S17). After adding two protons and two electrons, we explore the dehydrogenation reaction. In the blue path, two N–H bonds dissociate via a six-membered TS-1 transition state to make an H₂ molecule, followed by regenerating the active reagent Int-1. The activation energy for the dehydrogenation reaction varies for different reagents from 35.9 kcal mol⁻¹ to 40.3 kcal mol⁻¹ (Fig. 9).

Fig. 9 Transition states and relative energy (a) transition state structures of the CO₂ reduction reaction and (b) competition of the activation barrier for CO₂ reduction and H₂ formation reactions.

The high activation energy barrier for the H₂ formation reaction indicates a significant competition between the HER and CO₂ reduction reaction. The reagents L1–L6 participate in the CO₂ reduction reaction in the experimental study. Here, a detailed mechanism is proposed in Scheme 4 based on the experimental and theoretical studies. Like previous hydrogen formation reactions, Int-3 is generated after taking two electrons.

Scheme 4 A possible CO₂ reduction reaction path in the presence of different non-metal catalysts at pH = 7.4 and an applied potential of −1.60 V (SCE) (all values are in kcal mol⁻¹).

When CO₂ is introduced, Int-3 undergoes a reaction with CO₂ to make Int-7. The CO₂ addition reaction is exogenic for L3, L5, L2, and L1, with respective reaction energies of −5.0, −5.7, −6.3, and −4.8 kcal mol⁻¹ (Fig. S71). The reaction is endothermic for L6 and L4 with respective reaction energies of 1.5 and 2.9 kcal mol⁻¹. The electron-donating groups are more efficient for the CO₂ capture reaction. Next, two consecutive proton additions occur to regenerate Int-1 with the formation of CO and H₂O. In the first protonation reaction, captured CO₂ in Int-7 takes one proton to generate Int-8, and the step is exothermic for all the catalysts. In the second protonation reaction, the OH group of Int-8 takes one proton and converts it into water and CO with a regenerating catalyst. The reaction proceeds through the TS-3 transition state with an activation barrier of 8.7, 7.1, 9.5, 8.3, 10.8, and 11.4 kcal mol⁻¹ for catalysts L1–L6, respectively. The free energy plot of the CO₂ reduction reaction is illustrated in Fig. 10. The overall reaction is exothermic with a −96.4 kcal mol⁻¹ reaction energy.

Fig. 10 Energy profile diagram of the CO₂ reduction reaction to CO on different non-metal catalysts at pH = 7.4 and an applied potential of −1.60 V (SCE) in the presence of the L1 catalyst.

After two reductions, the reduced intermediate can participate in the H₂ evolution reaction, and in the presence of CO₂, CO₂ capture and CO formation take place. These non-metal catalysts are efficient for CO₂ reduction reactions with a competitive H₂ evolution process. Fig. 9(b) shows that the energy barrier for the CO₂ to CO formation reaction is lower than that for the competitive H₂ formation reaction for all six catalysts. The high activation energy barrier for the dehydrogenation step indicates that the CO₂ reduction reaction is more feasible than the H₂ formation reaction. Our experimental studies (vide supra, Fig. 5) are also nicely fitted with the DFT study in the presence of CO₂ reduction reactions that are more feasible than H₂ formation.

Catalytic comparison with reported catalysts

The catalytic performances of L1–L6 for CO₂ reduction were evaluated under identical electrochemical conditions (0.1 M Bu₄NClO₄ in DMF at −1.6 V vs. SCE), providing insights into their activity and selectivity. Among the terpyridine-based catalysts, L6 exhibited the highest faradaic efficiency for CO production at 59%, followed closely by L5 at 46%. L1 achieved an FE of 44%, while L4 and L3 produced CO with efficiencies of 29% and 27%, respectively. L2 displayed the lowest CO FE at 22% (Table 3). These results indicate that structural modifications, particularly the incorporation of electron-withdrawing or conjugation-extending substituents, significantly impact catalytic performance. When compared to reported catalysts, the terpyridine-based systems operate at a moderate potential of −1.6 V vs. SCE with competitive efficiencies. Benzoate derivatives and aromatic nitriles/esters, while achieving selectivity for oxalate (C₂O₄²⁻) of 96–99%, require significantly more negative potentials (−1.98 to −2.2 V vs. Ag/AgI and SCE). Pyridinium and pyridine/pyridazine systems excel under milder aqueous conditions and produce MeOH with faradaic efficiencies of 14–30% at potentials as low as −0.55 to −0.75 V vs. SCE. The PAC3 system, an extensively π-conjugated heterohelicene, demonstrates a FE of 35% for HCOOH production at −2.26 V vs. Fc⁺/Fc.

Table 3 The comparison of the catalytic performances with reported organocatalysts

Catalyst	Electrochemical conditions	Reported potential (V)	Products and selectivity	Faradaic efficiency (%)	CPE time (h)	Ref.
Where PAC3 = bis-imidazolium embedded extensively π conjugated heterohelicene.
L1	0.1 M Bu4NClO4 in DMF	−1.6 V vs. SCE	CO (99.6%) and H2 (0.5%)	CO (44.0) and H2 (0.2)	3.0	This work
L2	0.1 M Bu4NClO4 in DMF	−1.6 V vs. SCE	CO (99.9%) and H2 (0.046%)	CO (22.0) and H2 (0.01)	3.0	This work
L3	0.1 M Bu4NClO4 in DMF	−1.6 V vs. SCE	CO (99.8%) and H2 (0.14%)	CO (27.0) and H2 (0.04)	3.0	This work
L4	0.1 M Bu4NClO4 in DMF	−1.6 V vs. SCE	CO (99.9%) and H2 (0.07%)	CO (29.0) and H2 (0.02)	3.0	This work
L5	0.1 M Bu4NClO4 in DMF	−1.6 V vs. SCE	CO (99.5%) and H2 (0.05%)	CO (46.0) and H2 (0.02)	3.0	This work
L6	0.1 M Bu4NClO4 in DMF	−1.6 V vs. SCE	CO (99.9%) and H2 (0.03%)	CO (59.0) and H2 (0.02)	3.0	This work
bpy	0.1 M Bu4NClO4 in DMF	−1.8 V vs. SCE	CO (98.87%) and H2 (1.12%)	CO (34.0%) and H2 (0.4%)	1.0	This work
phen	0.1 M Bu4NClO4 in DMF	−1.8 V vs. SCE	CO (99.8%) and H2 (0.19%)	CO (36.0%) and H2 (0.7%)	1.0	This work
Benzoate derivatives	0.1 M Bu4NBr in DMF	−1.98 V vs. Ag/AgI	Dicarboxylic acid	75.0	—	18
Pyridinium ion over hydrogenated Pd electrodes	0.5 M NaClO4 at pH 5.4	−0.55 V vs. SCE	MeOH	30	19.0	19
Aromatic nitriles and esters	0.1 M Bu4NClO4 in DMF	−2.2 V vs. SCE	C2O4^2−	96	—	20
Pyridine and pyridazine	0.5 M KCl at pH 4.7	−0.75 V vs. SCE	MeOH	14 and 4	—	21
PAC3	0.1 M TBAPF6 in 10% H2O/MeCN	−2.26 V vs. Fc^+/Fc	HCOOH	35	6.0	22

---

Overall, L5 and L6 stand out among the terpyridine-based catalysts due to their efficiency and moderate operating potential, offering a promising metal-free alternative for selective CO₂ reduction to CO. Compared to other systems, their performance underscores the potential of structural tuning in organic frameworks to achieve competitive catalytic activity and selectivity.

Conclusion

The present study explains the synthesis and spectroscopic characterization of L1–L6 terpyridine derivatives. These catalysts were harnessed for electrocatalytic CO₂ reduction in a DMF/water mixture, utilizing TBAP as a supporting electrolyte. Controlled potential electrolysis (CPE) was conducted for all the catalysts, culminating in the generation of CO as the predominant end-product, exhibiting over 90% selectivity. A comparative analysis of the surface phenomenon unveiled a superior rate constant for L6 compared to others. Nevertheless, scrutinizing the catalytic Tafel plots revealed a heightened catalytic efficiency for L3 over all the other catalysts. The oblique displacement of the catalytic Tafel plot towards the upper left quadrant accentuates enhanced performance, indicating L3 as a preferred candidate for CO₂ reduction. Furthermore, upon delving into the bulk phenomenon, L6 manifested augmented catalytic efficiency relative to the others, verified by accompanying density functional theory (DFT) calculations. In this context, the electrochemical and chemical (EECC) mechanistic paths are favoured.

Abbreviations

CO2RR	CO2 reduction reaction
CPE	Controlled potential electrolysis
DFT	Density functional theory
FE	Faradaic efficiency
FOWA	Foot of the wave analysis
GC	Gas chromatography
HER	Hydrogen evolution reaction
SEM/EDX	Scanning electron microscopy with energy dispersive X-ray spectroscopy
SPE	Spectro electrochemistry
TBAP	Tetra butyl ammonium perchlorate
bpy	2,2′-Bipyridine
phen	1,10-Phenanthroline

Author contributions

SkSA: conducted the experiments and wrote the initial draft. KM: conducted the theoretical studies and wrote the manuscript. DR: curated the data and wrote the initial draft. TNM: conducted the labelling experiments. PKJ: conducted the experiments. AB: conducted the experiments. BSM: provided the resources for the DFT calculations and modified the manuscript. PK: provided the resources for online gas and mass analyses and labelled ¹³CO₂ studies. SKP: conceived the idea, provided the resources, and wrote and modified the draft manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

Details of materials and instrumentation, characterisation of the compounds, electrochemistry experiments, bulk electrolysis, gas evolution and cartesian coordinates are available in the supplementary information (SI).

Supplementary information: the electronic spectra, HRMS, magnetic, electrochemical, GC, and geometry-optimized coordinates are provided in the SI. See DOI: https://doi.org/10.1039/d5cy01027f.

Acknowledgements

This work was conducted at the Artificial Photosynthesis Laboratory, Department of Chemistry and Chemical Biology, IIT(ISM) Dhanbad, Jharkhand, India. It was financially supported by CSIR grant 01(3062)/21/EMR-II awarded to SKP. The authors acknowledge the CRF, IIT(ISM) Dhanbad for the HRMS and SEM/EDX data. The theoretical studies were conducted at IIT Hyderabad. The online gas and mass analyses were conducted at IISER Tirupati.

References

1. M. Aresta, A. Dibenedetto and A. Angelini, Chem. Rev., 2014, 114(3), 1709–1742.
2. M. Robert, ACS Energy Lett., 2016, 1, 281–282.
3. S. Ma and P. J. Kenis, Curr. Opin. Chem. Eng., 2013, 2(2), 191–199.
4. S. Zhang, Q. Fan, R. Xia and T. J. Meyer, Acc. Chem. Res., 2020, 53(1), 255–264.
5. S. J. Davis, K. Caldeira and H. D. Matthews, Sci., 2010, 329(5997), 1330–1333.
6. K. Talukdar, S. Sinha Roy, E. Amatya, E. A. Sleeper, P. Le Magueres and J. W. Jurss, Inorg. Chem., 2020, 59(9), 6087–6099.
7. M. Cheng, Y. Yu, X. Zhou, Y. Luo and M. Wang, ACS Catal., 2018, 9(1), 768–774.
8. G. Centi and S. Perathoner, Catal. Today, 2009, 148(3–4), 191–205.
9. W. H. Wang, Y. Himeda, J. T. Muckerman, G. F. Manbeck and E. Fujita, Chem. Rev., 2015, 115(23), 12936–12973.
10. J. L. White, M. F. Baruch, J. E. Pander III, Y. Hu, I. C. Fortmeyer, J. E. Park, T. Zhang, K. Liao, J. Gu, Y. Yan and T. W. Shaw, Chem. Rev., 2015, 115(23), 12888–12935.
11. C. Costentin, S. Drouet, M. Robert and J.-M. Savéant, Science, 2012, 338(6103), 90–94.
12. (a) C. W. Machan, M. D. Sampson and C. P. Kubiak, J. Am. Chem. Soc., 2015, 137(26), 8564–8571; (b) H. Ishida, K. Fujiki, T. Ohba, K. Ohkubo, K. Tanaka, T. Terada and T. Tanaka, J. Chem. Soc., Dalton Trans., 1990,(7), 2155–2160.
13. (a) M. R. Madsen, J. B. Jakobsen, M. H. Rønne, H. Liang, H. C. D. Hammershøj, P. Nørby, S. U. Pedersen, T. Skrydstrup and K. Daasbjerg, Organomettalics, 2020, 39(9), 1480–1490; (b) M. L. Clark, P. L. Cheung, M. Lessio, E. A. Carter and C. P. Kubiak, ACS Catal., 2018, 8(3), 2021–2029.
14. (a) C. M. Bolinger, B. P. Sullivan, D. Conrad, J. A. Gilbert, N. Story and T. J. Meyer, J. Chem. Soc., Chem. Commun., 1985,(12), 796–797; (b) C. M. Bolinger, N. Story, B. P. Sullivan and T. J. Meyer, Inorg. Chem., 1988, 27(25), 4582–4587.
15. S. T. Ahn, E. A. Bielinski, E. M. Lane, Y. Chen, W. H. Bernskoetter, N. Hazari and G. T. R. Palmore, Chem. Commun., 2015, 51(27), 5947–5950.
16. R. Francke, B. Schille and M. Roemelt, Chem. Rev., 2018, 118(9), 4631–4701.
17. (a) E. B. Cole, P. S. Lakkaraju, D. M. Rampulla, A. J. Morris, E. Abelev and A. B. Bocarsly, J. Am. Chem. Soc., 2010, 132, 11539–11551; (b) P. K. Giesbrecht and D. E. Herbert, ACS Energy Lett., 2017, 2(3), 549–555.
18. S. Gambino, G. Filardo and G. Silvestri, J. Appl. Electrochem., 1982, 12(5), 549–555.
19. G. Seshadri, C. Lin and A. B. Bocarsly, J. Electroanal. Chem., 1994, 372(1–2), 145–150.
20. A. Gennaro, A. A. Isse, J. M. Savéant, M. G. Severin and E. Vianello, J. Am. Chem. Soc., 1996, 118(30), 7190–7196.
21. E. Portenkirchner, C. Enengl, S. Enengl, G. Hinterberger, S. Schlager, D. Apaydin, H. Neugebauer, G. Knör and N. S. Sariciftci, ChemElectroChem, 2014, 1(9), 1543–1548.
22. P. Karak, S. K. Mandal and J. Choudhury, J. Am. Chem. Soc., 2023, 145(13), 7230–7241.
23. E. C. Constable, J. Lewis, M. C. Liptrot and P. R. Raithby, Inorg. Chem., 1990, 178(1), 47–54.
24. J. Patel, K. Majee, E. Ahmad, A. Vatsa, B. Das and S. K. Padhi, ChemistrySelect, 2017, 2(1), 123–129.
25. G. T. Morgan and F. H. Burstall, J. Chem. Soc., 1932, 3, 20–30.
26. G. Morgan and F. H. Burstall, J. Chem. Soc., 1937, 347, 1649–1655.
27. C. Costentin, S. Drouet, M. Robert and J. M. Savéant, Science, 2012, 338(6103), 90–94.
28. S. Roy, B. Sharma, J. Pécaut, P. Simon, M. Fontecave, P. D. Tran, E. Derat and V. Artero, J. Am. Chem. Soc., 2017, 139(10), 3685–3696.
29. S. S. Akhter and S. K. Padhi, Eur. J. Inorg. Chem., 2022, 2022(20), e202200206.
30. (a) M. Zabilskiy, V. L. Sushkevich, D. Palagin, M. A. Newton, F. Krumeich and J. A. van Bokhoven, Nat. Commun., 2020, 11, 2409; (b) R. Kumar, R. Babu, S. Chakrabortty, V. Madhu and E. Balaraman, J. Org. Chem., 2024, 89, 14720–14739; (c) P. Bhardwaj, T. Devi and P. Kumar, Inorg. Chem. Front., 2023, 10, 7285–7295; (d) Y. H. Hong, Y. M. Lee, S. Fukuzumi and W. Nam, Chem, 2024, 10, 1755–1765; (e) Kulbir, C. S. A. Keerthi, S. Beegam, S. Das, P. Bhardwaj, M. Ansari, K. Singh and P. Kumar, Inorg. Chem., 2023, 62, 7385–7392.
31. C. Costentin, S. Drouet, M. Robert and J. M. Savéant, J. Am. Chem. Soc., 2012, 134(27), 11235–11242.
32. (a) S. C. Cheng, C. A. Blaine, M. G. Hill and K. R. Mann, Inorg. Chem., 1996, 35(26), 7704–7708; (b) V. Nikolaou, Z. M. Luo, M. Gil-Sepulcre, J. W. Wang, O. Rüdiger, I.-F. Ardoiz, M. Nicaso, J.-B. Buchholz and A. Llobet, Inorg. Chem., 2025, 64, 11796–11806.
33. W. Kohn and L. J. Sham, Phys. Rev., 1965, 140(4A), A1133–A1138.
34. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M. S. Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman and D. J. Fox, Gaussian 16, Revision C.01, Gaussian, Inc., Wallingford CT.
35. Y. Zhao and D. G. Truhlar, Theor. Chem. Acc., 2008, 120(1), 215–241.
36. R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem. Phys., 2008, 72(1), 650–654.
37. M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S. Gordon, D. J. DeFrees and J. A. Pople, J. Chem. Phys., 1982, 77(7), 3654–3665.
38. A. V. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. B, 2009, 113(18), 6378–6396.
39. Y. Sakaguchi, A. Call, K. Yamauchi and K. Sakai, Dalton Trans., 2021, 50(44), 15983–15995.

---

Footnote

† These authors contributed equally to this work.

---