Electrocatalytic Hydrogen Evolution Using Cyano-Substituted Triaryl Corrole Antimony(III) Complexes

Abstract

Developing molecular electrocatalysts with controllable and predictable properties remains a central challenge in hydrogen evolution reaction (HER) catalysis. Herein, four Sb(III) corrole complexes (1–4) bearing zero to three p-cyano-substituted meso-phenyl groups (-CN Ph) were synthesized to investigate the effect of electron-withdrawing substituents on their catalytic HER performance, in which complexes 2–4 are newly reported. All prepared complexes were well characterized via UV–vis, NMR, HRMS, and XPS. SEM–EDS and UV–vis analyses indicated their uniform dispersion and excellent stability under organic and neutral aqueous solvent electrolysis conditions. When using TsOH as the proton source in DMF, complex 4 exhibited the highest activity with a TOF of 42.19 s⁻¹ at an overpotential of 895 mV. In mixed aqueous–organic media, the Faradaic efficiency of complex 4 reached 85.5%. The HER activity increases with the increasing number of cyano groups, and this observation has been rationalized via DFT calculations, which indicates a ligand-centered reduction and supports a possible ECEC pathway for the HER. These results highlight that cyano functionalization can modulate the electronic properties of Sb(III) corroles, thereby enhancing HER performance. This is helpful for designing efficient Sb(III) corrole-based HER catalysts.

1. Introduction

Driven by global “dual carbon” targets, the transition to clean energy, including hydrogen, solar, and wind technologies, has become a central strategy to address fossil fuel depletion and environmental pollution [1]. Hydrogen, with high energy density, abundant availability, and zero-emissions, has therefore emerged as a major research focus [2,3]. While various hydrogen production strategies, such as photocatalysis [4,5] and hydrogen carrier decomposition [6,7,8,9,10,11], have been explored, water electrolysis is widely considered a green and sustainable approach for hydrogen production and provides a well-defined electrochemical platform that is often used for studying molecular hydrogen evolution reaction (HER) catalysts, as it allows molecular properties to be related to catalytic behavior under controlled conditions [12]. However, its implementation is limited by slow kinetics and high overpotentials [13]. Developing electrocatalysts delivering high activity and stability at low overpotentials is crucial to enhance HER efficiency [14]. Although platinum-based catalysts perform excellently, their high cost and limited availability motivate the development of non-precious-metal catalysts [15,16]. Molecular electrocatalysts provide well-defined active sites and tunable electronic structures, making them attractive platforms for mechanistic studies and rational catalyst design [17].

Corrole, a non-natural aromatic tetrapyrrole tri-anionic macrocycle, has higher electron density than its porphyrin counterparts. It may stabilize metal ions in higher oxidation states [18]. Metal corrole complexes have been widely investigated for electrocatalytic HER, especially transition-metal corroles such as cobalt [19,20], copper [21,22,23], manganese [24,25], and iron complexes [26,27,28]. Recently, corrole complexes of main-group elements, including tin [29] and gallium [30], as well as non-metallic elements such as silicon [31] and phosphorus [32], have also gained increasing interest in electrocatalysis.

Modification at the meso-positions is widely recognized as an effective strategy for regulating the catalytic activities of corrole complexes via electronic effects [33,34]. In general, the introduction of strong electron-withdrawing groups, such as -CF₃ [35], -C₆F₅ [36], and -NO₂ [37], enhances HER performance. For example, Liang et al. [38,39,40] showed that meso-substituted electron-withdrawing groups lead to lower overpotentials, larger i_c/i_p value (i_c is the catalytic peak current after acid addition, and i_p is the initial reduction current without acid), and reduced electrochemical impedance, while Liu et al. [41] found that in phosphorus corroles, increasing the number of meso-pentafluorophenyl groups decreases both i_c/i_p and the apparent rate constant (k_obs), an effect opposite to the generally observed HER trend. In addition, Lei et al. [42] reported that corrole bearing the most electron-withdrawing substituents exhibits the highest catalytic stability.

Antimony(III) corrole was first synthesized in 2000 [43]. Although antimony porphyrins had been used as molecular HER electrocatalysts as early as 2017 [44], it was not until 2024 that Sb(III) corroles were firstly reported for the HER, which showed that the HER activity of Sb(III) corrole is dependent on the peripheral substituents [45]. In Sb(III) corroles, changes in meso-substitution have been shown to influence redox properties and catalytic performance, highlighting their electronic tunability [46,47]. Our previous studies on cobalt [48] and tin corroles [49] demonstrated that p-cyano-substituted meso-phenyl groups (-CN Ph) can affect the electronic properties of the corrole framework significantly and enhance catalytic activity. Motivated by these findings, we herein wish to examine the effect of the number of -CN Ph on the electrocatalytic HER performance of Sb(III) corroles. The molecular structures of the prepared Sb(III) corrole complexes are depicted in Figure 1.

2. Results and Discussion

2.1. Structural Characterization

Based on literature methods [50,51], Sb(III) corroles bearing 0–3 -CN Ph were synthesized. Complexes 2–4 are newly prepared, whereas complex 1 has been previously reported. All complexes were characterized via nuclear magnetic resonance (NMR) and high-resolution mass spectrometry (HRMS). The spectroscopic data are consistent with the expected corrole frameworks and molecular formulas, with representative spectra shown in Figures S1–S24 and summarized in Table S1.

The UV–visible spectroscopy (UV–vis) absorption spectra of freebase and complexes 1–4 were recorded in dichloromethane (DCM) over the range of 300–800 nm (Figure 2). All freebase corroles display a strong Soret band within 415–425 nm, along with two to three weaker Q bands within 500–700 nm, characteristic of corrole-based π–π* transitions (S₀ → S₂ and S₀ → S₁, respectively) [52]. Coordination with Sb(III) induces a notable redshift of the Soret band and enhanced Q-band intensity, consistent with previous reports [45,50]. The intense Soret band shows a progressive red shift from complexes 1 to 4, consistent with the increasing electron-withdrawing effect of the -CN Ph groups, which modulates the π-electronic structure of the corrole macrocycle and lowers the π→π* transition energy. A distinct splitting of the Soret band is observed, which arises from the Sb(III) atom positioned out of the macrocycle plane, inducing structural distortion and symmetry reduction [51]. In contrast, Sb(V) corroles typically feature a nearly coplanar metal center and display an unsplit Soret band [53,54]. Interestingly, the degree of Soret band splitting gradually diminishes with increasing numbers of -CN Ph substituents. This trend, similar to that reported in earlier studies, likely arises from changes in the in-plane electronic dipole components of the complex [46,55]. These spectral features confirm the formation of the freebase and corresponding Sb(III) corrole complexes while revealing the substituent-dependent modulation of their electronic structures. Detailed absorption data are summarized in Table S2.

X-ray photoelectron spectroscopy (XPS) was used to investigate the elemental composition and antimony oxidation state in complexes 1–4. The Sb 3d spectra (Figure 3a) exhibit two characteristic peaks at ~530 eV (Sb 3d_5/2) and ~540 eV (Sb 3d_3/2), consistent with the documented binding energies for Sb(III). Notably, the Sb 3d_5/2 (~530 eV) partially overlaps with O 1s (~532 eV) [45,56]. The N 1s spectra (Figure 3b) show two observable components: one at ~399 eV, corresponding to the pyrrolic nitrogen of the corrole macrocycle, and another at ~400 eV, assigned to the cyano nitrogen, in agreement with prior reports [48,49]. Additionally, increasing the number of cyano substituents leads to higher Sb 3d binding energies, likely reflecting the inductive electron-withdrawing effect of -CN groups on the Sb center. The C 1s and XPS survey spectra of the complexes are provided in the Supporting Information (Figure S25). Overall, the XPS analysis confirms the successful coordination of antimony to the freebase corroles. The corresponding full width at half maximum (FWHM) values used for peak deconvolution are summarized in Table S3.

2.2. Cyclic Voltammetry (CV) Studies

CV experiments were performed on complexes 1–4 (0.5 mM) to investigate the effect of cyano substitution on their redox behavior. Measurements were conducted in DMF with TBAP (0.10 M) serving as the supporting electrolyte, and all potentials were calibrated against the Fc/Fc⁺ couple (Figure S26). The CV profiles recorded covering −1.95–0.30 V at 100 mV·s⁻¹ (Figure 4) display two similar redox couples located near −1.7 V and 0.1 V for all complexes. These processes correspond to single-electron corrole-centered reduction and oxidation, consistent with previous studies reporting that the Sb(III) oxidation state remains unchanged during electrochemical cycling [46,57]. The features at ~−1.7 V and ~0.1 V are therefore ascribed to the [Sb corrole]/[Sb corrole]⁻ and [Sb corrole]/[Sb corrole]⁺ couples, respectively. A weak reduction peak near −1.5 V appears from the second scan onward for all complexes, likely arising from intermediates generated during oxidation at ~0.1 V [46]. Extending the negative scan beyond −1.6 V for complexes 3 and 4 also reveals a weak oxidation peak around −0.8 V, plausibly originating from intermediates formed during the ~−1.7 V reduction process. A clear substituent effect is evident: the [Sb corrole]/[Sb corrole]⁻ couple shifts progressively toward more positive potentials with increasing numbers of cyano groups. This trend reflects the strong electron-withdrawing nature of cyano groups, which decreases the π-electron density of the corrole macrocycle and facilitates reduction at less negative potentials.

To further understand the electronic structure of complexes 1–4, density functional theory (DFT) calculations were performed. As shown in Figure 5a, the optimized geometries of complexes exhibit the characteristic triarylcorrole frameworks, indicating that 4-cyanophenyl introductions do not induce significant structural distortions. Frontier molecular orbital analysis (Figure 5b) shows that the HOMO, HOMO−1, LUMO, and LUMO+1 are predominantly localized on the corrole ligand, with negligible contribution from the Sb center, suggesting the ligand-centered nature of the redox processes [58]. The calculated frontier orbital energies exhibit systematic lowering with increasing cyano substitutions. Specifically, cyano groups lower the HOMO energy, making electron removal less favorable and shifting the oxidation potential anodically, while the lowering of the LUMO energy enhances electron acceptance, consistent with the positive shifts observed in reduction potentials [59]. These results show that cyano groups may significantly modulate the electronic structures of the corrole ligands, aligning with the observed redox behavior (Table S4).

To assess the electron-transfer behavior of complexes 1–4, CV profiles were recorded at scan rates of 100–400 mV·s⁻¹ (Figure S27). Analysis of the peak current dependence according to the Randles–Sevcik equation (Equation (1)) shows a linear increase of i_p with v^1/2, demonstrating that the redox processes are diffusion-controlled [60].

$${\mathit{i}}_{\mathit{p}}=\mathbf{0.4463}{\mathit{n}}_{\mathit{p}}\mathit{F}\mathit{A}{\mathit{c}}_{\mathit{c}\mathit{a}\mathit{t}}{\left({\displaystyle \frac{{\mathit{n}}_{\mathit{p}}\mathit{F}\mathit{v}\mathit{D}}{\mathit{R}\mathit{T}}}\right)}^{\mathbf{1}\mathbf{/}\mathbf{2}}.$$

(1)

In Equation (1), i_p denotes the peak current, n_p is the number of electrons transferred, F is the Faraday constant (96,485 C·mol⁻¹), A is the electrode surface area (0.07 cm²), c_cat is the concentration of complexes (0.5 mM), R is the gas constant (8.3145 J·mol⁻¹·K⁻¹), T is the temperature, D is the diffusion coefficient, and v is the scan rate.

2.3. Electrocatalytic Hydrogen Evolution of Antimony Corroles in Organic Media

The HER catalytic performance of Sb(III) corroles, which is strongly dependent on proton donor acidity, was investigated in an organic medium using acetic acid (AcOH, pK_a = 13.5 [61]), trifluoroacetic acid (TFA, pK_a = 6.0 [62]), and p-toluenesulfonic acid (TsOH, pK_a = 2.6 [63]).

With AcOH, a weak proton donor, only small shifts in peak potentials and slight increases in current were observed for complexes 1–4 (Figure S28), indicating that the HER proceeds inefficiently under weakly acidic conditions due to limited proton availability, which hampers the rapid protonation of the reduced species [Sb corrole]⁻ and slows the catalytic cycle.

With TFA, pronounced changes emerged (Figure 6). Increasing acid concentration caused the oxidation wave of the [Sb corrole]/[Sb corrole]⁻ couple to diminish, while the corresponding reduction wave shifted anodically with a clear increase in peak current. These features are characteristic of an EC mechanism (E: electron transfer, C: proton coupling), in which the one-electron-reduced species [Sb corrole]⁻ undergoes rapid protonation. The pronounced increase in the reduction peak near −1.9 V suggests the involvement of additional EC steps at more negative potentials.

To further assess the role of the Sb center in enhancing HER activity, a comparative study was performed using 36 equivalents of TFA as the proton source. CV profiles were recorded for an unmodified glassy carbon electrode, the freebase corrole (TCC, 0.5 mM), and the corresponding complex 4 (TCC-Sb, 0.5 mM) (Figure S29). Upon addition of TFA, the bare electrode displayed negligible current, and TCC showed only a weak reduction response, indicating minimal catalytic contribution from the ligand framework alone. In contrast, complex 4 exhibited a noticeably higher reduction current, reflecting its improved electrocatalytic performance. These results clearly demonstrate that incorporation of the Sb(III) center notably promotes the HER activity of Sb(III) corrole systems.

With TsOH, a stronger proton donor than TFA, complexes 1–4 exhibited larger positive shifts in their reduction peaks, accompanied by a further increase in catalytic current (Figure 7). These electrochemical features indicate that the EC sequence operates more efficiently. This enhanced activity arises from the much lower pK_a of TsOH, which provides a higher effective proton concentration in the reaction medium, thereby facilitating rapid and sustained protonation and promoting the catalytic cycle. Moreover, the CV profiles show that the Sb corroles exhibit similar electrochemical behavior in both TFA and TsOH, indicating that the HER proceeds through the same mechanism.

To better assess the effect of cyano substitution on the HER activity of complexes 1–4, the peak current ratio (i_c/i_p) and turnover frequency (TOF) were analyzed. The i_c/i_p ratio reflects the relative catalytic response under identical electrochemical conditions and is commonly used to assess electron–proton coupling behavior during catalysis. The results show that i_c/i_p rises with higher proton concentrations and exhibits excellent linear correlations. Under identical acid concentrations, the catalytic activity follows the order 4 > 3 > 2 > 1 (Figure 8), reflecting improved catalytic performance associated with an increasing number of -CN Ph substituents. The k_obs was extracted using Equation (2), and the TOF, numerically equivalent to k_obs, directly represents the intrinsic catalytic rate [64]. As shown in Figure S30, the TOF values follow the same substituent-dependent trend as i_c/i_p, with complex 4 achieving a maximum of 42.19 s⁻¹ under 36 equivalents of TsOH, highlighting the beneficial effect of cyano substitution. The observed substituent-dependent trend suggests that cyano substitution effectively modulates the electronic structure of the corrole, which in turn facilitates electron transfer during the catalytic process.

$${\mathit{k}}_{\mathit{o}\mathit{b}\mathit{s}}=\mathit{k}{\left[{\mathit{H}}^{+}\right]}^{\mathit{x}}=\mathbf{1.94}\mathit{v}{\left({\displaystyle \frac{{\mathit{i}}_{\mathit{c}}}{{\mathit{i}}_{\mathit{p}}}}\right)}^{\mathbf{2}}.$$

(2)

Overpotential, commonly used to evaluate HER efficiency, is defined according to Equations 3 and 4 [65]. In both TFA and TsOH media, the overpotentials of complexes 1–4 follow the trend 1 > 2 > 3 > 4. Under 36 equivalents of TFA, complex 4 displays the smallest overpotential (777 mV). This behavior can be rationalized by the incorporation of -CN Ph, whose strong inductive electron-withdrawing effect modulates the electronic structure of the corrole framework, resulting in a more favorable catalytic reduction potential. Together with the i_c/i_p and TOF results, these observations support the beneficial role of -CN Ph in Sb(III) corrole HER performance. All catalytic parameters are summarized in Table S5.

$$\mathsf{\eta }={\mathit{E}}_{\mathit{H}\mathit{A}}^{\mathbf{0}}-\mathit{E}$$

(3)

and

$${\mathit{E}}_{\mathit{H}\mathit{A}}^{\mathbf{0}}={\mathit{E}}_{{\mathit{H}}^{+}}^{\mathbf{0}}-\left({\displaystyle \frac{\mathbf{2.303}\mathit{R}\mathit{T}}{\mathit{F}}}\right){\mathit{p}\mathit{K}}_{\mathit{a}},$$

(4)

where ${\mathit{E}}_{\mathit{H}\mathit{A}}^{\mathbf{0}}$ is the standard H⁺/H₂ potential in DMF (−0.62 V), and pK_a denotes the acid dissociation constant in DMF. Based on these parameters, the calculated proton-coupled potentials are ${\mathit{E}}_{\mathit{T}\mathit{F}\mathit{A}}^{\mathbf{0}}$= −0.97 V and ${\mathit{E}}_{\mathit{T}\mathit{s}\mathit{O}\mathit{H}}^{\mathbf{0}}$= −0.77 V.

To place the present system within the context of molecular HER catalysis, a comparison with representative catalysts based on main-group elements and transition metals is summarized in Table S6. Under comparable conditions, TCC-Sb(4) exhibits HER activity within the range of these molecular systems, with moderate overpotentials and TOFs. The results further indicate that HER activity in Sb corrole systems can be systematically modulated through electronic effects introduced by macrocycle substitution, providing guidance for future catalyst design.

Controlled potential electrolysis (CPE) was performed to assess the efficiency and stability of complexes 1–4. The complexes were subjected to 1 h of electrolysis at −1.8 V in TFA and TsOH. After subtracting the background charge from a blank electrode (Figure S31), the total accumulated charge of the complexes followed the trend 4 > 3 > 2 > 1 (Figure 9), with complex 4 achieving the highest corrected charge of 410 mC in TsOH, reflecting its superior catalytic performance. This substituent-dependent trend is consistent with the activity sequence derived from TOF analyses, further supporting the positive effect of cyano substitution on the sustained HER performance of Sb(III) corrole complexes.

Stability assessment in organic media indicated that the UV–vis spectra and CV profiles of complexes recorded before and after 1 h of CPE showed no noticeable changes, confirming molecular stability (Figures S32 and S33). The recyclability of the complexes was further evaluated via CVs recorded over 30 consecutive cycles, which remained essentially unchanged (Figure S34). Electrode reusability was verified using CV profiles of bare glassy carbon electrodes recorded before and after CPE and after 30 consecutive CV cycles in the presence of the catalyst, indicating that the electrode surface remains stable under repeated operation (Figures S35 and S36).

To further rationalize the excellent durability of the corrole complexes, DFT calculations were carried out to examine a possible demetallation pathway (Tables S7–S10). The calculated free-energy change for metal dissociation is substantially high, indicating that demetallation of the Sb corrole is energetically unfavorable under the electrochemical conditions.

Electrochemical impedance spectroscopy (EIS) was employed to probe the interfacial charge-transfer behavior of complexes 1–4 (Figure 10). The Nyquist plots display a semicircle in the high-to-medium frequency region associated with the charge-transfer resistance (R_ct). In both TFA and TsOH, the R_ct values decrease from complex 1 to 4 and are consistently lower in TsOH at the same acid concentrations, indicating more favorable charge-transfer kinetics for complexes with a higher number of -CN Ph substituents. The low-frequency oblique line suggests that mass-transport limitations are not the dominant contribution under the present conditions, consistent with prior reports on molecular catalysts [30,66]. Equivalent-circuit fitting yields R_ct values of 673, 621, 578, and 525 Ω in TFA, and 631, 571, 513, and 476 Ω in TsOH for complexes 1–4, respectively. The reduced R_ct values are indicative of more efficient interfacial charge-transfer behavior, which is generally beneficial for sustaining higher catalytic currents during the HER, consistent with the observed activity trends. A summary of the equivalent-circuit fitting parameters is provided in Tables S11 and S12.

2.4. Proposed Mechanism for Hydrogen Evolution Catalyzed by Antimony Corroles

Compared with the spectra recorded without acid, the UV–vis profiles of complexes 1–4 showed no discernible changes upon addition of 36 equivalents of TsOH (Figure S32). This observation indicates that the acidic medium does not directly protonate the complexes to form [Sb(III)Cor]-H. Instead, the catalytic process likely requires an initial one-electron reduction to generate the active species, which then undergoes proton-coupled steps.

In addition, Tafel plots of complex 4 derived from CV measurements (Figure S37) exhibit slopes of 110.76 mV dec⁻¹ with TFA and 96.99 mV dec⁻¹ with TsOH, suggesting that the HER is likely controlled by the initial proton–electron transfer (Volmer step) [21,30,67]. A complementary kinetic isotope effect (KIE) experiment was performed using deuterated trifluoroacetic acid (TFA-d₁) as the proton source for complex 4. The KIE value was determined to be 4.49, further confirming that proton transfer participates in the rate-limiting step (Figure 11). These observations are consistent with a Volmer–Heyrovsky-type hydrogen evolution pathway.

In light of these findings, we propose a plausible catalytic pathway for complexes 1–4, where TFA or TsOH acts as the proton donor (Figure 12). The HER is expected to proceed via an ECEC pathway. Specifically, the Sb(III) corrole complex undergoes an initial one-electron reduction to generate the active intermediate [Sb(III)Cor]⁻ (I → II), which rapidly engages in a proton-coupling step to afford the hydride species [Sb(III)Cor] ^∙H (II → III), completing the first EC event. A subsequent reduction then produces [Sb(III)Cor]H⁻ (III → IV), which reacts with a second proton via a Heyrovsky-type process to produce H₂ and restore the original species (IV → I), thereby accomplishing the second EC step.

To obtain more complementary spectroscopic evidence, UV–vis spectra of complex 4 were recorded under reducing conditions, followed by the addition of a proton source (Figure S38). Upon reduction, complex 4 showed characteristic changes, including attenuation of the bands at 447 and 461 nm, the appearance of a new maximum at ~430 nm, and the emergence of a distinct absorption band at 700–720 nm, which is a characteristic feature of ligand-centered corrole radical anions [47,68]. Interestingly, addition of the proton source led to an obvious new blue-shifted spectrum, which may be temporally assigned as the formation of int2 (Figure 12). These observations provide more information on the reduced and protonated intermediates involved in the HER. Definitive identification of these species is a topic waiting for future studies.

To further understand the HER mechanism, DFT calculations were carried out. Spin density analysis of the one-electron-reduced intermediate int1 (Figure S39) reveals that the unpaired electron is predominantly localized at the 5- or 15-positions of the macrocycle. Combined with the frontier orbital distributions on the macrocycles (Figure 5b), a ligand-centered HER mechanism is supported rather than metal-centered involvement. Based on the proposed mechanism and previous reports [30,47], protonation-site screening was performed (Scheme S1) using the most active catalyst TCC-Sb(4) as a model complex (M1–M13, Figure S40), and M5 was found to have the lowest energy (Figure 13a, Table S13), indicating that protonation at the 15-position is the most favorable. Accordingly, int2 is proposed to be 15-protonated for all complexes.

The Gibbs free energy profile along the HER pathway is shown in Figure 13b. The first one-electron reduction step (formation of int1) exhibits the highest energy requirement and is therefore the rate-determining step. Notably, this energetic barrier is significantly lower than that calculated for demetallation, further supporting that catalytic turnover is kinetically favored over catalyst decomposition. The calculated energy changes for this step are in the order 1 > 2 > 3 > 4, consistent with the experimentally observed anodic shifts in onset potentials, and the energy-ordering for the subsequent steps is the same. The second reduction step (int2 → int3), which is associated with H-H bond formation, is thermodynamically uphill for all complexes, with free energy changes decreasing in the order 1 > 2 > 3 > 4, indicating a progressively lower energetic penalty for this formation. The subsequent H₂ release from int3 is exergonic and follows the same ordering, suggesting increasingly facile H₂ desorption from complexes 1 to 4. These thermodynamic trends may rationalize the experimentally observed TOF trend (4 > 3 > 2 > 1). Overall, the calculated thermodynamic trend (4 > 3 > 2 > 1) agrees well with the experimental HER activity order, with complex 4 being the most active among them (Tables S7–S10).

To evaluate the influence of cyano substitution on the aromaticity of the corrole macrocycle, NICS(1)_zz calculations were performed for complexes 1–4 (Scheme S2, Table S14). The negative NICS(1)_zz values indicate antiaromatic character, which is consistent with their nonplanar geometries as reflected by the molecular planarity parameter (MPP) and span of deviation from plane (SDP) [69]. The extent of this antiaromaticity was further examined using localized orbital locator for π-electrons (LOL-π, Figure 14) [70,71] and multicenter bond order (MCBO) analyses, which reveal that π-electron density is mainly localized on individual pyrrolic units and interrupted delocalization over the macrocycle, indicative of weak and localized (anti)aromatic character. Therefore, cyano substitution appears to have little effect on the intrinsic aromaticity of the corrole framework while primarily modulating the electronic density distribution.

Electrostatic potential (ESP) [72,73] (Figure 15) and atomic dipole-corrected Hirshfeld atomic charge (ADCH) analyses (Table S15) [74,75] are utilized to estimate the effects of cyano introductions on electronic density. As described, as more cyano is introduced, the macrocycles and Sb atoms show decreasing electronic densities. It is noted that the lessened electronic density is more favorable for the electron-accepting steps, especially the most energy-consuming step of int1 formation, which is in agreement with the experimental results.

To examine whether the substituent-dependent trend observed for cyano groups is general, additional DFT calculations were carried out for Sb corrole models bearing zero to three nitro (-NO₂) or trifluoromethyl (-CF₃) substituents at the meso-phenyl positions. The calculated HOMO/LUMO energies (Figures S41 and S42) and ESP maps (Figures S43 and S44) show progressive LUMO energy lowering and increased electron deficiency of the corrole framework with increasing numbers of electron-withdrawing groups, similar to the cyano series. The same trends have been reported for Sb corroles with varying meso-groups [46,47], consistent with our findings. These results indicate that the enhanced HER activity arises from a general electron-withdrawing effect rather than a cyano-specific phenomenon.

2.5. Electrocatalytic Hydrogen Evolution of Antimony Corroles in Neutral Aqueous Media

To evaluate the practical applicability of complexes 1–4 as electrocatalysts, we further examined their HER performance under neutral aqueous conditions. Experiments were performed in a 0.1 M phosphate solution under neutral conditions, using a DMF/H₂O mixture (1:2, v/v) to enhance solubility and enable effective catalysis. All potentials were measured versus a saturated KCl Ag/AgCl electrode.

The concentration-dependent HER activity of complexes 1–4 was investigated (Figure S45). The introduction of the catalysts markedly increased the electrochemical current and shifted the initial reduction potential positively. The relative activities of the complexes follow the order 4 > 3 > 2 > 1, in agreement with the trends observed in organic media (Figure 16).

CPE (2 min, 1138–1538 mV) was performed to assess the HER activity of complexes 1–4. All charge values were corrected using blank electrolysis data (Figure S46). The accumulated charge rises with overpotential and follows the order 4 > 3 > 2 > 1, with complex 4 achieving 53.78 mC at 1538 mV (Figure S47). TOF values calculated from Equations (5) and (6) [76] exhibited the same trend, although the idealized assumptions regarding catalyst distribution and electron utilization at the electrode interface may result in conservative estimates of activity (Figure S48).

$$\mathit{O}\mathit{v}\mathit{e}\mathit{r}\mathit{p}\mathit{o}\mathit{t}\mathit{e}\mathit{n}\mathit{t}\mathit{i}\mathit{a}\mathit{l}=\mathit{A}\mathit{p}\mathit{p}\mathit{l}\mathit{i}\mathit{e}\mathit{d} \mathit{p}\mathit{o}\mathit{t}\mathit{e}\mathit{n}\mathit{t}\mathit{i}\mathit{a}\mathit{l}+\mathbf{0.059}\mathit{p}\mathit{H}+\mathbf{0.199}$$

(5)

and

$$\mathit{T}\mathit{O}\mathit{F}={\displaystyle \frac{∆\mathit{C}}{\mathit{F}{\mathit{n}}_{\mathbf{1}}{\mathit{n}}_{\mathbf{2}}\mathit{t}}},$$

(6)

In these equations, applied potential refers to the electrolysis voltage, ΔC represents the blank-corrected charge, F is the Faraday constant (96,485 C·mol⁻¹), n₁ is the number of electrons required to produce 1 mol of H₂, n₂ is the catalyst amount, and t is the electrolysis time.

Gas chromatography (GC) was applied to qualitatively and quantitatively evaluate H₂ evolution for complexes 1–4 (Figure 17). Electrolysis was performed in a phosphate medium containing 5.0 μM catalyst at −1.8 V for 1 h. Visible bubble formation on the electrode and within the bulk solution confirmed the occurrence of the HER. Quantification using an external calibration curve (Figure S49) revealed H₂ volumes of 0.36, 0.68, 0.94, and 1.41 mL for complexes 1–4, respectively. The corresponding Faradaic efficiencies, calculated according to Equations (7) and (8) [77], were 70.84%, 76.36%, 82.3%, and 85.45%. These results demonstrate that catalytic performance improves progressively with increasing cyano substitution. All calculations were performed after subtracting the blank background. The associated catalytic parameters are summarized in Table S16.

$$\mathit{P}{\mathit{V}}_{{\mathit{H}}_{\mathbf{2}}}={\mathit{n}}_{{\mathit{H}}_{\mathbf{2}}}\mathit{R}\mathit{T}$$

(7)

and

$$\mathit{F}\mathit{a}\mathit{r}\mathit{a}\mathit{d}\mathit{a}\mathit{i}\mathit{c} \mathit{e}\mathit{f}\mathit{f}\mathit{i}\mathit{c}\mathit{i}\mathit{e}\mathit{n}\mathit{c}\mathit{y} \mathit{f}\mathit{o}\mathit{r} {\mathit{H}}_{\mathbf{2}}={\displaystyle \frac{\mathit{F}{\mathit{n}}_{{\mathit{H}}_{\mathbf{2}}}\mathit{z}}{\mathit{Q}}}\times \mathbf{100}\%,$$

(8)

where F is the Faraday constant (F = 96,485 C·mol⁻¹), ${\mathit{n}}_{{\mathit{H}}_{\mathbf{2}}}$ is the amount of H₂ produced during electrolysis, z is the number of electrons transferred per H₂ molecule (z = 2), and Q denotes the total charge accumulated during the 1 h electrolysis.

The stability of complexes 1–4 in neutral aqueous media was assessed via 8 h of CPE at −1.8 V. Molecular stability is supported by nearly constant current densities and unchanged UV–vis spectra (Figure S50). Electrode stability was confirmed using SEM and EDS analyses after long-term electrolysis, showing no detectable changes in morphology or composition (Figure S51). These results indicate that both the molecular catalysts and the electrode exhibit excellent durability under aqueous HER conditions, supporting that the HER activity predominantly arises from the molecular Sb corroles rather than from heterogeneous species formed on the electrode.

3. Materials and Methods

All reagents and solvents were purchased from commercial suppliers (Macklin, Shanghai, China; Aladdin, Shanghai, China; Energy Chemical, Shanghai, China) and were of analytical grade (≥98%) unless otherwise stated. Pyrrole (≥98.00%, Acros Organics, Geel, Belgium) was freshly distilled prior to use. UV–vis absorption spectra were acquired using a PerkinElmer LAMBDA 365+ spectrophotometer with DCM/DMF as the solvent at room temperature. ¹H NMR and ¹³C NMR spectra were recorded on a Bruker AVANCE III HD 500 MHz spectrometer using CDCl₃ as the solvent. High-resolution mass spectra (HRMS) were obtained using a Bruker Maxis Impact (ESI) instrument with methanol as the solvent. X-ray photoelectron spectroscopy (XPS) was performed on a Thermo Scientific K-Alpha⁺ spectrometer with monochromatic Al Kα radiation (150 W). Binding energies were corrected against the C 1s peak at 284.8 eV, and the spectra were background-corrected using the Shirley method.

Electrochemical measurements. Electrochemical experiments were conducted on a CHI 660 E electrochemical workstation. Prior to each measurement, the solutions were purged with high-purity N₂ to remove dissolved oxygen. All measurements were carried out at room temperature. Organic-phase measurements were performed in DMF with 0.10 M TBAP using a three-electrode configuration consisting of a 3 mm glassy carbon working electrode, an Ag/AgNO₃ reference, and a carbon rod counter. Mixed aqueous measurements were performed in a DMF/H₂O (1:2, v/v) electrolyte with 0.1 M KCl and 0.25 M KH₂PO₄, with the pH adjusted to neutral using KOH. A 1 cm × 1 cm glassy carbon plate, an Ag/AgCl reference, and a platinum wire counter were used in the three-electrode setup. CV, CPE, EIS, and GC were employed for electrochemical characterization. CV measurements were conducted over the potential window of −1.95 to 0.3 V at a scan rate of 100 mV s⁻¹, whereas in the presence of acid, the potential range was adjusted to −1.95 to 0 V. Hydrogen gas generated during electrolysis was measured qualitatively and quantitatively using an Agilent 7890 A gas chromatograph.

Computation details. Density functional theory (DFT) calculations were performed. Geometrical optimizations for all following species (Tables S17–S55) were conducted in Gaussian C. 01 [78], at the ωB97XD/SDD level for the Sb atom and def2-SVP for others [79,80] in the N,N-dimethylformamide (DMF) implicit solvent model produced by the integral equation formalism polarizable continuum model (IEFPCM) [81]. Frequency calculations were performed at the same level to ensure the optimized structures at the minimum of the potential surfaces and to obtain more accurate thermodynamic quantities. All electron realistic single point energy (SPE) calculations were then performed in DMF (SMD model) [82] with ORCA 6.1.1 [83,84], considering scalar realistic effects in the second-order Dougals–Kroll–Hess (DKH2) method at ωB97M-V/SARC-DKH-TZVPP [85,86] for the Sb atom and DKH-def2-TZVPP for others [80]. NICS(1)_zz [87,88,89] calculations were performed for complexes 1–4 at the ωB97M-V/x2c-TZVPPall [90,91] using ORCA 6.1.1, considering the scalar realistic effects within the X2C approach. We placed a ghost atom in the mean plane center of the macrocycle for complexes 1–4 and set an extremely tight S-type basic function (S 1 1 100,000) for it. All of the following wave function analyses, including ESP and ADCH calculations, were completed using Multiwfn_3.8 [92,93], and the results were visualized using VMD 1.9.3. [94]. For comparison, the same procedures were implemented on Sb(III) corrole complexes in which the cyano was replaced by trifluoromethyl or nitro. Gibbs free energy changes along the HER pathway were obtained using Shermo_2.6 [95], applying a frequency scaling factor 0.9490 (ωB97XD/6-31G(d)).

Based on previously reported procedures [50,51], we synthesized a series of Sb corrole complexes. Scheme 1 illustrates the synthesis of dipyrromethane intermediates, while Scheme 2 depicts the synthetic route to corrole ligands and subsequent metalation to afford complexes 1–4.

Synthesis of phenyl-substituted dipyrromethane (a)

Freshly distilled pyrrole (160 mL) and benzaldehyde (1.52 g) were mixed until a homogeneous solution was obtained. TFA (150 µL) was rapidly injected below the liquid surface as a catalyst, and the reaction proceeded for 1 h under light exclusion, then quenched with Et₃N (300 µL). Residual pyrrole was removed under reduced pressure at 140 °C to yield a black viscous residue. The crude product was purified on 300–400 mesh silica gel using HEX/DCM (3:1, v/v), yielding the desired compound a (2.2 g, 69.1%).

Synthesis of 4-Cyanophenyl-Dipyrrolidine (b)

The target product was prepared analogously to a, with 4-cyanobenzaldehyde (1.87 g) used in place of benzaldehyde (1.52 g). The crude material was purified on silica gel (300–400 mesh) using HEX/DCM (1:2, v/v) to yield the desired compound b (2.98 g, 84.53%).

Synthesis of 5,10,15-Triphenylcorrole (TPC)

Compound a (2.5 mmol, 0.56 g) and benzaldehyde (1 mmol, 0.11 g) were reacted with diluted HCl (3.6%, 100 mL) in methanol (100 mL) at room temperature under light exclusion for 2 h. The reaction mixture was extracted with DCM and saturated brine, dried over anhydrous Na₂SO₄, and concentrated. The residue was dissolved in DCM, treated with p-chloranil (0.74 g), and stirred in the dark for 14 h. Excess p-chloranil was separated via wet-column chromatography, and the crude product was further purified on 300–400 mesh silica gel using HEX/DCM (6:1, v/v) and recrystallized from DCM/HEX to yield the desired compound TPC (98 mg, 18.72%). ¹H NMR (500 MHz, Chloroform-d) δ 8.88 (t, J = 7.6 Hz, 4H), 8.61–8.47 (m, 4H), 8.36 (d, J = 7.2 Hz, 4H), 8.17 (d, J = 6.7 Hz, 2H), 7.78 (dt, J = 41.0, 7.5 Hz, 9H). ¹³C NMR (126 MHz, Chloroform-d_, with 0.5 equiv of TFA-d₁) δ 145.98, 143.45, 138.70, 137.58, 136.88, 135.93, 135.74, 135.44, 134.97, 130.44, 129.98, 129.82, 129.10, 128.77, 125.86, 125.54, 122.88, 119.39, 119.23. HR-MS: m/z calculated for C₃₇H₂₆N₄⁺: 527.2230 [M+H]⁺; found: 527.2241.

Synthesis of 5,15-Diphenyl-10-(4-cyanophenyl) corrole (BPCC)

BPCC was prepared analogously to TPC, with 4-cyanobenzaldehyde (1 mmol, 0.131 g) replacing benzaldehyde. The crude product was purified via HEX/EA (10:1, v/v) and recrystallized from DCM/HEX to yield the desired compound BPCC (191 mg, 34.69%). ¹H NMR (500 MHz, Chloroform-d, with 0.5 equiv of TFA-d₁) δ 8.90 (s, 4H), 8.56 (s, 2H), 8.44 (s, 2H), 8.31 (d, J = 30.9 Hz, 6H), 8.05 (s, 2H), 7.78 (d, J = 42.9 Hz, 6H). ¹³C NMR (126 MHz, Chloroform-d, with 0.5 equiv of TFA-d₁) δ 152.28, 142.15, 140.93, 137.45, 136.94, 136.24, 136.12, 135.82, 134.91, 134.49, 132.30, 131.56, 130.17, 130.10, 125.74, 125.25, 122.12, 119.78. HR-MS: m/z calculated for C₃₈H₂₅N₅⁺: 552.2183 [M+H]⁺; found: 552.2190.

Synthesis of 10-Phenyl-5,15-bis(4-cyanophenyl) corrole (PBCC)

PBCC was prepared analogously to TPC, with compound a replaced by b (0.618 g, M = 247.3). The crude product was purified using HEX/EA (6:1, v/v) and recrystallized from DCM/HEX to yield the desired compound PBCC (210 mg, 36.49%). ¹H NMR (500 MHz, Chloroform-d) δ 8.97 (s, 2H), 8.82 (s, 2H), 8.65–8.42 (m, 8H), 8.14 (d, J = 21.8 Hz, 5H), 7.78 (s, 4H). ¹³C NMR (151 MHz, Chloroform-d) δ 149.47, 144.28, 142.31, 139.19, 135.43, 134.82, 131.89, 128.66, 127.56, 126.37, 122.56, 120.23, 119.28, 116.95, 114.65, 113.40, 111.30, 100.52. HR-MS: m/z calculated for C₃₉H₂₄N₆⁺: 577.2135 [M+H]⁺; found: 577.2143.

Synthesis of 5,10,15-Tris(4-cyanophenyl) corrole (TCC)

TCC was prepared analogously to TPC, with compound a replaced by b (0.618 g, M = 247.3) and benzaldehyde replaced by 4-cyanobenzaldehyde (0.131 g, M = 131.13). The crude product was purified using HEX/EA (5:1, v/v) and recrystallized from DCM/HEX to yield the desired compound TCC (232 mg, 36.63%). ¹H NMR (500 MHz, Chloroform-d) δ 9.07 (s, 2H), 8.87 (s, 2H), 8.6–8.43 (m, 8H), 8.30 (s, 2H), 8.11 (d, J = 20.1 Hz, 6H). ¹³C NMR (151 MHz, Chloroform-d) δ 146.87, 144.12, 135.56, 135.45, 132.01, 131.30, 127.27, 126.45, 123.25, 119.22, 117.40, 111.77, 111.53, 109.68. HR-MS: m/z calculated for C₄₀H₂₃N₇⁺: 602.2088 [M+H]⁺; found: 602.2095.

Synthesis of TPC-Sb(1)

TPC (100 mg) was reacted with SbCl₃ (0.674 g) in pyridine (10 mL) at 100 °C under light exclusion for 2 h until complete consumption of the freebase corrole was confirmed via TLC. After solvent removal under reduced pressure, the residue was redissolved in DCM, dry-loaded onto silica, and purified on 300–400 mesh silica gel using HEX/DCM (10:1, v/v). The purified fraction was recrystallized from DCM/HEX to yield the desired compound TPC-Sb(1) (97 mg, 79%). ¹H NMR (500 MHz, Chloroform-d) δ 9.14 (dd, J = 40.7, 4.4 Hz, 4H), 8.75 (dd, J = 22.5, 4.3 Hz, 4H), 8.26 (s, 5H), 8.00 (s, 1H), 7.76 (dt, J = 23.9, 7.8 Hz, 9H). ¹³C NMR (126 MHz, Chloroform-d) δ 143.75, 141.85, 140.62, 138.98, 138.05, 134.89, 134.51, 134.11, 128.94, 127.74, 127.67, 127.47, 127.45, 125.63, 124.72, 123.64, 117.42, 115.98, 109.61. HR-MS: m/z calculated for C₃₇H₂₃N₄Sb: 644.0955; found: 644.0960.

Synthesis of BPCC-Sb(2)

BPCC (100 mg) was reacted with SbCl₃ (0.643 g) following the procedure for TPC-Sb. The crude product was purified via HEX/DCM (2:1, v/v) and recrystallized from DCM/HEX to yield the desired compound BPCC-Sb(2) (98 mg, 80.98%). ¹H NMR (500 MHz, Chloroform-d) δ 9.12 (d, J = 4.1 Hz, 2H), 9.06 (d, J = 4.7 Hz, 2H), 8.72 (d, J = 4.1 Hz, 2H), 8.57 (d, J = 4.6 Hz, 2H), 8.18 (s, 6H), 7.98 (d, J = 7.4 Hz, 2H), 7.71 (dt, J = 13.8, 7.1 Hz, 6H). ¹³C NMR (126 MHz, Chloroform-d) δ 149.33, 147.14, 145.59, 142.95, 140.29, 139.06, 138.79, 138.13, 135.18, 134.46, 131.27, 129.51, 127.86, 127.81, 126.23, 124.61, 124.52, 124.14, 123.73, 123.63, 119.44, 119.22, 119.15, 119.00, 118.18, 116.37, 114.22, 111.22, 107.01. HR-MS: m/z calculated for C₃₈H₂₂N₅Sb: 669.0908; found: 669.0902.

Synthesis of PBCC-Sb(3)

PBCC (100 mg) was reacted with SbCl₃ (0.615 g) following the procedure for TPC-Sb. The crude product was purified using HEX/DCM (1:1, v/v) and recrystallized from DCM/HEX to yield the desired compound PBCC-Sb(3) (98 mg, 81.19%). ¹H NMR (500 MHz, Chloroform-d) δ 9.19 (d, J = 4.1 Hz, 2H), 8.97 (d, J = 4.7 Hz, 2H), 8.75 (d, J = 4.7 Hz, 2H), 8.69 (d, J = 4.1 Hz, 2H), 8.31 (s, 5H), 8.03 (d, J = 8.2 Hz, 4H), 7.69 (s, 4H). ¹³C NMR (126 MHz, Chloroform-d) δ 152.27, 149.33, 147.80, 147.74, 147.21, 145.52, 143.56, 141.27, 139.45, 139.05, 138.83, 137.56, 135.99, 135.17, 131.56, 128.21, 127.83, 127.61, 125.91, 125.57, 124.93, 124.61, 124.51, 124.13, 123.62, 119.32, 119.22, 119.00, 117.02, 114.91, 114.22, 111.59, 111.24, 96.28. HR-MS: m/z calculated for C₃₉H₂₁N₆Sb: 694.0860; found: 694.0858.

Synthesis of TCC-Sb(4)

TCC (100 mg) was reacted with SbCl₃ (0.59 g) following the procedure for TPC-Sb. The crude product was purified using HEX/DCM (1:2, v/v) and recrystallized from DCM/HEX to yield the desired compound TCC-Sb(4) (101 mg, 84%). ¹H NMR (500 MHz, Chloroform-d) δ 9.21 (d, J = 4.1 Hz, 2H), 9.01 (d, J = 4.7 Hz, 2H), 8.72 (d, J = 4.1 Hz, 2H), 8.66 (d, J = 4.7 Hz, 2H), 8.31 (s, 5H), 8.08–7.96 (m, 7H). ¹³C NMR (126 MHz, Chloroform-d) δ 147.22, 146.54, 145.20, 142.81, 138.69, 137.71, 135.52, 135.04, 131.63, 131.42, 128.80, 126.16, 124.92, 124.61, 124.13, 119.22, 117.43, 115.65, 114.22, 111.87, 111.77, 108.57. HR-MS: m/z calculated for C₄₀H₂₀N₇Sb⁺: 720.0891 [M+H]⁺; found: 720.0865.

4. Conclusions

In summary, four Sb(III) corrole complexes bearing zero to three -CN Ph substituents were prepared, and their HER performance was evaluated in both organic and neutral aqueous media. The HER activity follows an order of 1 < 2 < 3 < 4, indicating that increasing number of cyano substitution may progressively enhance catalytic response by altering the electronic properties of the corrole macrocycle. This enhancement is manifested by higher catalytic currents and turnover frequencies, together with lower overpotentials and charge-transfer resistances. All complexes maintained excellent stability and electrochemical durability under operational conditions. DFT calculations rationalize the experimentally observed substituent-dependent HER activity trend and further support an ECEC catalytic pathway. This work shows that cyano functionalization is an effective electronic-tuning strategy for improving HER activity of Sb(III) corroles and is helpful for the design of efficient new Sb(III) corrole-based molecular electrocatalysts.