Next Article in Journal
Mechanical Properties of Biodegradable Fibers and Fibrous Mats: A Comprehensive Review
Previous Article in Journal
Study on the Electro-Optical Properties of Polymer-Dispersed Liquid Crystals Doped with Cellulose Nanocrystals
Previous Article in Special Issue
Novel Fe(II)-Based Supramolecular Film Prepared by Interfacial Self-Assembly of an Asymmetric Polypyridine Ligand and Its Electrochromic Performance
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Molecules
Volume 30
Issue 15
10.3390/molecules30153274
molecules-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Oxygen Reduction by Amide-Ligated Cobalt Complexes: Effect of Hydrogen Bond Acceptor

by
Zahra Aghaei
1,
Adedamola A. Opalade
1,
Victor W. Day
2 and
Timothy A. Jackson
1,*
1
Department of Chemistry and Center for Environmentally Beneficial Catalysis, University of Kansas, Lawrence, KS 66045, USA
2
X-Ray Crystallography Laboratory, University of Kansas, Lawrence, KS 66045, USA
*
Author to whom correspondence should be addressed.
Molecules 2025, 30(15), 3274; https://doi.org/10.3390/molecules30153274
Submission received: 30 June 2025 / Revised: 1 August 2025 / Accepted: 1 August 2025 / Published: 5 August 2025
(This article belongs to the Special Issue Metal Complexes: Synthesis, Characterization and Applications)
Browse Figures
Versions Notes

Abstract

The ability of earth-abundant metals to serve as catalysts for the oxygen reduction reaction is of increasing importance given the prominence of this reaction in several emerging technologies. It is now recognized that both the primary and secondary coordination environments of these catalysts can be modulated to optimize their performance. In this present work, we describe two CoII complexes [CoII(PaPy2Q)](OTf) (1) and [CoII(PaPy2N)](OTf) (2) that catalyze chemical and electrochemical dioxygen reduction. Both 1 and 2 contain CoII centers in a N5− coordination environment, but 2 has a naphthyridine group that places a nitrogen atom in the secondary coordination sphere. Solid-state X-ray crystallography and solution-state spectroscopic measurements reveal that, apart from this second-sphere nitrogen in 2, complexes 1 and 2 have essentially identical properties. Despite these similarities, 2 performs the chemical reduction of dioxygen ~10-fold more rapidly than 1. In addition, 2 has an enhanced performance in the electrochemical reduction of dioxygen compared to 1. Both complexes yield a significant amount of H2O2 in the chemical reduction of dioxygen (>25%). The enhanced catalytic performance of 2 is attributed to the presence of the second-sphere nitrogen atom, which might enable the efficient protonation of cobalt–oxygen intermediates formed during turnover.

1. Introduction

The catalytic reduction of dioxygen (O2) is important in energy storage technologies, fuel cells, and metal–air batteries [1]. The combination of the oxygen reduction reaction (ORR) and fuel oxidation generates an electromotive force capable of powering vehicles, electronic devices, and homes [2]. Molecular oxygen can undergo reduction via the following two distinct pathways:
O2 + 2H+ + 2e− → H2O2
O2 + 4H+ + 4e− → 2H2O
The conventional process for generating hydrogen peroxide (the so-called anthraquinone process) is energy-demanding. An alternative process for H2O2 production, which is potentially more sustainable, is the electrochemical, two-electron/two-proton reduction of O2 to H2O2 (Equation (1)) [3]. The four-electron/four-proton reduction of dioxygen to generate water (Equation (2)) is used in fuel cells [4]. The catalytic reduction of molecular oxygen is also important in aerobic respiration in biology, where cytochrome c oxidase converts dioxygen to water [5]. Other examples of dioxygen reduction in biology include amine oxidase and galactose oxidase enzymes, which reduce O2 to hydrogen peroxide [6], and multi-copper oxidases that reduce O2 to H2O [7].
The selectivity of the ORR towards water or hydrogen peroxide by metal catalysts is influenced by several factors, such as the electronic and steric features of the ligand and the primary and secondary coordination spheres of the metal center. Because the coordination spheres of homogeneous metal catalysts can be more easily tuned than their heterogeneous analogues, studies of molecular catalysts have played an important role in understanding the mechanism of dioxygen reduction [8].
Currently, platinum nanoparticles are the benchmark for ORR catalysts. However, the scarcity and expense of platinum have motivated the development of cost-effective alternatives based on first-row transition metals [9,10]. Cobalt complexes are being increasingly recognized as promising ORR catalysts. For instance, Duboc et al. demonstrated that a bimetallic Co-based complex supported by a bis-thiolate N2S2-donor ligand was capable of electrochemically and chemically catalyzing the reduction of dioxygen to water, using octamethylferrocene (Me8Fc) as a reducing reagent and 2,6-lutidinium tetrafluoroborate (LutHBF4) as the proton source [11]. Although numerous studies have explored the mechanisms and activities of earth-abundant metal catalysts in reducing O2, there is still an incomplete understanding of how particular changes in the metal primary and secondary coordination spheres can be used to tailor catalyst performance [12,13]. Extensive research has focused on the development of Fe-based catalysts that incorporate secondary coordination sphere effects to boost catalyst performance in the ORR [14,15,16,17,18,19]. In contrast, although numerous molecular cobalt catalysts have been developed to reduce O2 via either the 2e−/2H+ or 4e−/4H+ pathways, only a few studies have demonstrated how secondary coordination sphere interactions influence the performance of cobalt ORR catalysts [20,21]. Studies have shown that the steric hindrance of access to the cobalt center [22] and the presence of cationic groups [23] in the second coordination sphere can play important roles in modulating ORR catalyst performance.
Several recent studies have shown how the ORR activity and selectivity of cobalt catalysts can be tuned by appending Lewis bases in the secondary coordination sphere. For example, Machan reported catalytic dioxygen reduction to water in MeOH by a CoIII complex supported by a bipyridine-bis-phenolate (N2O2) ligand (Scheme 1a) [24]. No catalytic activity was observed in MeCN. A modified version of this catalyst with methoxy groups near the O2 binding site reduced dioxygen to H2O2 in MeOH and was catalytically active in MeCN (Scheme 1b) [24]. Paria and co-workers described a CoIII catalyst supported by a bis-pyridine-bis-oxime ligand (Scheme 1c) [8]. In phosphate buffer at pH = 4, the catalyst favored the 4e−/4H+ pathway to give H2O. The selectivity was altered at a higher pH, with ~75% water formation at pH = 7 and 65% H2O2 generation at pH = 11. It was proposed that the oxime moiety in the second coordination sphere of this catalyst acts as a proton shuttle [8]. More recently, Paria and co-workers expanded on this initial complex by generating a series of CoIII complexes using the bis-pyridine-dioxime framework but with the aryl moiety appended with different substituents (Scheme 1d–h). All complexes reduced O2 through the 4e–/4H+ pathway, but the complexes bearing ortho-NHMe2+ and ortho-OMe substituents (Scheme 1d,e) had turnover frequencies (TOF) increased by ~1000- and ~250-fold compared to the unsubstituted parent complex [25].
In this present study, we report two CoII complexes with amide-containing, anionic N5 ligands (Scheme 2) that act as catalysts in the reduction of molecular oxygen. The CoII complexes [CoII(PaPy2Q)]+ (1) and [CoII(PaPy2N)]+ (2) have identical first coordination spheres (PaPy2QH (N,N-Bis(2-pyridylmethyl)amine-N-ethyl-2-quinolinecarboxamide) and PaPy2NH (N-(2-(bis(pyridine-2-ylmethyl)amino)ethyl)-1,8-naphthyridine-2-carboxamide)). However, the PaPy2N ligand of complex 2 includes a naphthyridine moiety that introduces a nitrogen atom into the secondary coordination sphere.
The presence of this Lewis base gives complex 2 the ability to engage in proton delivery and/or hydrogen bonding interactions with cobalt–oxygen intermediates that might form during dioxygen reduction. A comparison of the ORR activities of these complexes, therefore, provides a probe of how a very specific modification of the secondary coordination sphere influences catalysis. Compared to 1, complex 2 demonstrates a nearly 10-fold increase in the rate of dioxygen reduction, has an approximately 23-fold higher efficiency based on TOFi values, and exhibits more activity in the electrochemical reduction of dioxygen.

2. Results

2.1. Formation and Characterization of Complexes [CoII(PaPy2Q)](OTf) (1) and Na[CoII(PaPy2N)](OTf)2 (2)

Complexes 1 and 2 were prepared by adding Co(OTf)2 to PaPy2QH and PaPy2NH in methanol under a nitrogen atmosphere. The solid-state structures of 1 and 2 were determined by X-ray crystallography (Figure 1). Each complex has a CoII center in a distorted trigonal bipyramidal environment (τ = 0.73 and 0.67 for 1 and 2, respectively). The PaPy2Q and PaPy2N ligands are bound in a pentadentate fashion, with the pyridyl moieties giving N(pyridine)–Co–N(pyridine) angles of 118° and 114°, respectively.
For 1, the crystal structure shows a non-coordinating triflate counter anion, with the closest Co···O distance of 5.67 Å. The unit cell of complex 2 shows Na+ ions from the NaOtBu used during its preparation. Two Na+ ions and four triflate counter anions form a bridge between separate units of 2, with the O(amide) groups of each unit of 2 serving as ligands to Na+ (Supplementary Materials, Figure S1).
An overlay plot of 1 and 2 (Supplementary Materials, Figure S1) reveals essentially identical structures. Specifically, all nonhydrogen atoms superimpose with a root-mean-square deviation of 0.146 Å. The displacement of Co from the N2, N4, N5 mean plane is identical (0.36 Å) in both molecules, and the N1, N2, and N3 groupings are essentially perpendicular to the N2, N4, N5 triangle, as one would expect for a trigonal bipyramid (see Figure S1 for coordination polyhedra). A comparison of the Co–N bond lengths between 1 and 2 shows only minor differences (Table 1). For example, the Co–N1 bond in 2 (2.0806(19) Å) is slightly shorter than that of 1 (2.107(3) Å), which could reflect a difference between the naphthyridine and quinoline ligands. Indeed, the crystal structures of the [MnII(PaPy2Q)](OTf) and [MnII(PaPy2N)](OTf) complexes show that the Mn–N(quinoline) bond in 1 is longer than the Mn–N(naphthyridine) bond in 2 (2.357(2) and 2.228(3) Å, respectively) [26]. The Co–N2 bond of 1, which involves the amide nitrogen, is slightly shorter than that of 2, which might reflect the involvement of the O(amide) group in 2 with the Na+ ions (Supplementary Materials, Figure S1). The other Co–N2, Co–N3, and Co–N4 distances are quite similar for 1 and 2 (Table 1). Given the very similar Co–N bond lengths between 1 and 2, the biggest difference is in the presence of N6 in the naphthyridine group in 2. The positioning of this atom is near a cleft between the pyridine ligands, which could be a potential site of O2 binding and reduction. Consequently, complex 2 has the potential for a possible proton relay in the second coordination sphere that is lacking in 1. Such second-sphere proton relays have previously been shown to be very effective in enhancing the efficacy of the oxygen reduction reaction [8,25,27,28].
The electronic absorption spectra of 1 and 2 dissolved in MeCN are quite similar (Figure 2). Each spectrum shows a broad band with a low intensity near ~650–700 nm, a sharper, more intense feature near ~520–550 nm, and a broader, higher-intensity band at ~450 nm. Specifically, complex 1 exhibits electronic absorption bands at 650, 540, and 460 nm (ε = 50, 180, and 150 M−1 cm−1, respectively). Complex 2 has electronic absorption bands at 650, 545, and 450 nm (ε = 20, 115, and 190 M−1 cm−1, respectively). In each case, these bands arise from ligand-field transitions of the CoII centers, which are quite sensitive to the metal coordination geometry [29]. Assuming an idealized C3v point group for 1 and 2, we can assign the three absorption bands as excitations from the 4A2 ground state to 4E, 4A2(P), and 4E(P) excited states, in ascending order. These assignments follow those of Lever, who reported that high-spin, trigonal bipyramidal CoII complexes have three ligand-field absorption bands in the visible spectral region, with 4A2 → 4E at 990–670 nm, 4A2 → 4A2(P) at 750–580 nm, and 4A2 → 4E(P) at 580–470 nm [29]. The similarity in the energies and intensities of the electronic absorption bands is consistent with 1 and 2 having the same CoII geometry in MeCN solution. We also performed experiments to determine if MeCN coordinates to the CoII centers in 1 and 2, which would generate six-coordinate species. Because the electronic absorption spectra of 1 and 2 in the non-coordinating solvent CH2Cl2 have band energies identical to those in MeCN (Supplementary Materials, Figure S2), we conclude that MeCN does not coordinate to the CoII centers and they remain five-coordinate.
Figure 1 and Figure 2 were obtained from ESI-MS and 1H NMR experiments. The ESI-MS data for 1 and 2 at 25 °C show prominent peaks at m/z of 455.12 and 456.11, respectively (SI Figure S2). These results are in excellent agreement with the calculated m/z of [CoII(PaPy2Q)]+ (455.12) and [CoII(PaPy2N)]+ (456.11). The 1H NMR spectra of 1 and 2 in MeCN-d3 are quite similar (Table 1 and Figure S3), with each showing nine peaks in the downfield region (150–10 ppm) and one peak in the upfield region (~−6 to −10 ppm). This large range of chemical shift values is attributed to hyperfine interactions between the protons of the PaPy2Q and PaPy2N supporting ligands and the paramagnetic CoII centers. Although the development of detailed assignments for these peaks is beyond the scope of this work, the similar chemical shift patterns and identical number of hyperfine-shifted peaks suggest similar structures for 1 and 2 in solution. On the basis of the X-ray crystal structures for 1 and 2 showing equivalent pyridyl groups, we would predict 13 proton resonances for 1 and 12 for 2. The appearance of fewer than the predicted number of resonances is not unexpected. Protons closest to the CoII center could be fast-relaxing and too broad to observe. Alternatively, some proton resonances might appear in the diamagnetic region and be obscured by solvent peaks.
The perpendicular-mode X-band EPR spectra of frozen solutions of 1 and 2 in MeCN at 7.5 K show positive signals at g ≈ 4.4 and broad, negative signals at higher fields (Table 1 and Supplementary Materials, Figure S4). These signals are similar to those observed for other high-spin CoII complexes [30] and indicate an S = 3/2 ground state. We further probed the spin states of 1 and 2 by performing Evans NMR experiments in MeCN-d3 at 25 °C (Table 1 and Figure S5). From these experiments, we observed magnetic moments of 2.6 μB for 1 and 2.9 μB for 2. (In calculating the magnetic moments for 1 and 2, we did not include diamagnetic corrections, which are usually small for these types of complexes.) These values are lower than expected for the spin-only magnetic moments of an S = 3/2 system (~3.9 μB) [31]. It is possible that the magnetic moments for 1 and 2 were lower than expected due to reactions with dioxygen that would give product(s) with lower spin states. However, larger ranges of magnetic moments of 2.3–4.3 μB have been reported for CoII complexes [32], which have been attributed to a range of factors [33].

2.2. Reduction of Dioxygen with 1 and 2

2.2.1. Chemical Reduction of Dioxygen

We explored the ability of 1 and 2 to catalytically reduce dioxygen in an air-saturated MeCN solution at 25 °C using a chemical reductant. We chose decamethylferrocene (Me10Fc) as the reductant, with trifluoroacetic acid (TFA) as acid. The reactions used a large excess of reductant (1.1 mM) and acid (15 mM) compared to the cobalt catalyst (0.02 mM). The catalytic ORR activity was determined by monitoring the oxidation of Me10Fc to Me10Fc+ at 780 nm (ε = 495 M−1 cm−1) by UV–Vis spectroscopy (Figure 3) [34].
The addition of 0.02 mM of 2 to a solution of Me10Fc (1.1 mM) and TFA (15 mM) under air led to the complete (100%) conversion of Me10Fc to Me10Fc+ in 5.2 min (Figure 3, bottom). Complex 1 also reacted with 1.1 mM Me10Fc and 15 mM TFA, but in this case, the complete conversion of Me10Fc to Me10Fc+ was not observed until 60 min (Figure 3, top). Control reactions in the absence of complexes 1 and 2 gave the maximum formation of Me10Fc+ at 90 min (Supplementary Materials, Figure S6). The difference in Me10Fc+ formation rates between 1 and 2 is attributed to the presence of the naphthyridine group in 2 that can engage in second-sphere hydrogen bonding interactions. To provide a detailed comparison of the performances of 1 and 2, we calculated TOFi (initial turn over frequency) values for each catalyst with respect to the concentration of Me10Fc+ (Supplementary Materials, Figure S7). On the basis of these values, complex 2 (TOFi = 1200 h−1) reacted ~23 times more rapidly than complex 1 (TOFi = 70 h−1).
To assess the robustness of 1 and 2 as catalysts, we performed cycles of catalytic experiments using 0.02 mM Co complex, 15 mM TFA, and 0.55 mM (0.36 mmoles) Me10Fc. After Me10Fc+ had reached maximal formation, 0.36 mmoles of Me10Fc was added to the same reaction mixture. Each addition resulted in the complete conversion of the added Me10Fc to Me10Fc+ (Supplementary Materials, Figure S8). An increase in the reaction time was observed for the second and third Me10Fc additions. Specifically, the reaction time for 1 increased by ~3 min for each addition of Me10Fc. For 2, the reaction time increased by ~5 min. These observations imply that some fraction of the catalyst might have decayed. However, the reaction times are still less than those observed in the absence of a catalyst (120 min when using 15 mM TFA and 0.55 mM Me10Fc), implying that a significant portion of the catalyst remained active over multiple cycles.
The Ti-TPyP (oxo [5,10,15,20-tetra(4-pyridyl)porphyrinato]titanium(IV)) reagent was utilized to quantify the amount of hydrogen peroxide produced from the reduction of dioxygen in the reactions of 1 and 2. From these experiments, we observed 18 ± 6% formation of H2O2 for 1 and 27 ± 7% H2O2 formation for 2. The percentages are based on the theoretical 100% yield of H2O2 based on the concentration of 1.1 mM Me10Fc (Equation (3)), as follows:
O2 + 2 Me10Fc + 2 CF3CH2COOH → 2 Me10Fc+ + H2O2 + 2 CF3CH2COO−
Because hydrogen peroxide is known to react with certain ferrocene derivatives, including Me10Fc [35,36,37], the values of H2O2 reported here should be viewed as lower limits of the actual amount of H2O2 produced in these reactions. In support, we repeated the catalytic reduction of O2 in the presence of 2 at −40 °C while keeping all other conditions the same as those in our experiments at 25 °C (Supplementary Materials, Figure S9). The lower temperature was chosen to minimize the decomposition of H2O2. We observed 42 ± 5% H2O2 under these conditions. This result suggests that catalysts 1 and 2 predominantly generate hydrogen peroxide, which decays in the presence of Me10Fc at higher temperatures. To further verify this possibility, we repeated the catalytic reactions with H2O2 (1.1 mM) instead of dioxygen. In the first experiment (Supplementary Materials, Figure S10), H2O2 (1.1 mM) was added to an MeCN solution of Me10Fc (1.1 mM), TFA (15 mM), and complex 2 (0.02 mM) under a nitrogen atmosphere at 25 °C. After approximately 30 s, we observed ~100% conversion of Me10Fc to Me10Fc+ and detected 52 ± 3% of hydrogen peroxide in the final solution. This near 50% conversion of H2O2 is consistent with two equiv. of Me10Fc consumed per one equiv. of H2O2 consumed (i.e., Me10Fc is a one-electron reductant, and the reduction of H2O2 to H2O requires two electrons). A second experiment performed in the absence of a catalyst (Supplementary Materials, Figure S11) revealed ~100% formation of Me10Fc+ in roughly 30 s, with ~56 ± 9% of hydrogen peroxide remaining. These data suggest that H2O2 is rapidly reduced to H2O by Me10Fc. On this basis, we conclude that H2O2 is the dominant product of dioxygen reduction by 1 and 2 under the described conditions.

2.2.2. Electrocatalytic ORR by Complexes 1 and 2

The abilities of 1 and 2 to electrocatalytically reduce dioxygen were investigated with cyclic voltammetry (CV). These experiments were performed in an MeCN solution of 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6) at room temperature. A background scan of the electrolyte solution shows no current with a magnitude greater than -3 mA in the potential range from 0 to −1.2 V vs. Fc/Fc+ (Figure 4, left inset). A scan with only electrolyte and TFA (5 mM) under a nitrogen atmosphere shows a slight increase in current magnitude to approximately −15 μA near 1.2 V (Figure 4, left, and Supplementary Materials, Figure S12). The magnitude of this current increased only slightly for solutions of electrolyte, TFA, and a catalytic amount (0.2 mM) of either 1 or 2 under nitrogen (Figure 4, left). Large changes are observed in the CV traces when the same experiments with 0.1 mM solutions of 1 and 2 were performed in an oxygen-saturated solution. In the presence of a catalytic amount of 2 (0.1 mM) (Figure 4, right, red trace), an irreversible peak is observed at −0.71 V vs. Fc+/Fc, exhibiting a large catalytic current (more than −100 μA). Similarly, a CV scan for a solution of electrolyte, TFA, and 1 (0.1 mM) in the presence of oxygen shows an irreversible peak at −0.77 V vs. Fc+/Fc with a current near −60 μA (Figure 4, right, blue trace). Notably, the catalytic current of 2 was ~1.6-fold larger than that of 1 (Figure 4, left). This difference in the magnitude of the produced catalytic current is potentially due to the difference in the second coordination spheres of these two complexes (Figure 1).
The scan rate dependence of the catalytic current was evaluated by CV experiments in an oxygen-saturated solution in the presence of 5 mM TFA at room temperature. Both cobalt complexes showed a linear relationship between the square root of the scan rate vs. peak current (Figure 5), indicating that the electrocatalytic process was diffusion-controlled [8,38,39]. This result suggests that 1 and 2 perform electron transfer to reduce O2 as homogeneous species and not as adsorbed catalysts on the electrode surface. If adsorption was occurring, the peak current would be expected to show a linear relationship with the scan rate [39].

3. Discussion

The controlled reduction of dioxygen is critical to both biological and industrial processes. In a key biological example, dioxygen serves as a proton and electron acceptor during cellular respiration in aerobic organisms. In this multi-step process, the reduction of O2 to H2O is coupled with proton transfer across a membrane, driving ATP synthesis [40]. Emerging energy technologies such as fuel cells and metal–air batteries utilize the reduction of O2 to H2O to convert chemical to electrical energy [2]. In both systems, catalysts are required to coordinate the delivery of multiple protons and electrons to O2. At present, the most effective synthetic catalysts are carbon-supported platinum nanoparticles [2]. The high cost and low abundance of platinum causes scalability issues with fuel cells, motivating the development of new catalysts with earth-abundant metals [13,41,42,43,44]. To develop these new catalysts, we must understand the specific structural and electronic factors that control (a) the efficiency of O2 reduction and (b) the selectivity for forming H2O. Structure–function performance metrics are most readily obtained for homogeneous catalysts. The knowledge gained from these studies can be applied to optimize heterogeneous catalysts, which are more practical in fuel cells. While certain catalyst design elements, such as potential sites of protonation (Lewis bases) near the O2 binding site, have been shown to affect both O2 reduction efficiency and selectivity, more systematic studies are needed to link particular molecular features with increases in performance.
In this present study, we generated a pair of cobalt complexes (1 and 2; see Figure 1), which allowed us to evaluate the role of a Lewis base in close proximity to the metal center in oxygen reduction selectivity and efficiency. Both X-ray crystallography and spectroscopic measurements revealed that 1 and 2 had very similar CoII coordination environments. However, the naphthyridine moiety in 2 incorporated a Lewis base in the second coordination sphere. Several previous studies have established that second-sphere Lewis bases can enhance the rate of dioxygen reduction and can also modulate the selectivity for H2O versus H2O2 production [8,25,28,29]. These second-sphere bases likely assist in the delivery of protons to Co–oxygen intermediates. Scheme 3 shows a general mechanism for oxygen reduction by a mononuclear Co catalyst [24]. Oxygen reduction is initiated by the binding of O2 to the CoII complex to give a CoIII-superoxide (CoIII-OO.) intermediate. This intermediate is protonated and reduced (either in a concerted or stepwise fashion), generating a CoIII-hydroperoxo (CoIII-OOH) intermediate. In the last step, the second protonation and reduction occur and are followed by the cleavage of the Co-O bond to hydrogen peroxide, regenerating the CoII species.
In our experiments probing the ability of 1 and 2 to reduce oxygen using the reductant Me10Fc, we observed that 2 achieved the maximum formation of Me10Fc+ 10-fold faster than 1. Given that the only difference between these complexes was the presence of the second-sphere Lewis base in 2, we attribute this difference to the role of this Lewis base in facilitating the reduction of dioxygen. We also studied the ability of both complexes to reduce dioxygen electrochemically. CV measurements showed that 2 had a ~1.6-fold larger catalytic current compared to 1. Although the Lewis base in 2 had a large influence on the rate of dioxygen reduction, the effect on product selectivity appeared more limited. An analysis of the reaction mixtures following the chemical reduction of dioxygen at 25 °C revealed 20 ± 6% and 30 ± 7% formation of H2O2 for 1 and 2, respectively. In both cases, this extent of H2O2 formation should be viewed as a lower limit, given the potential for Me10Fc to facilitate the reduction of H2O2 to H2O. The reduction of dioxygen to H2O2 is well-known for cobalt ORR catalysts [24,44,45,46,47,48,49,50], although shifts from all N-coordination to mixed N/O or N/S coordination appear to increase the selectivity for H2O [11,24].
What makes complex 2 a more efficient catalyst than 1? In a recent study of a different mononuclear Co ORR catalyst, Machan and co-workers demonstrated that the proton-transfer steps to the Co–oxygen intermediates were rate-determining [24]. Thus, any modifications to the catalyst architecture that would facilitate these steps could increase catalytic performance. We postulate that, in the presence of an acid such as TFA, complex 2 exists as an equilibrium of protonated forms, with N6 (Figure 1) as the site of protonation. The ability of 2 to harbor protons near the Co center could enhance the rate of proton-transfer steps, thereby increasing its catalytic efficiency for dioxygen reduction. Alternatively, or in addition, the ability of 1 to perform dioxygen reduction could be suppressed by the relatively close proximity of H24 (Supplementary Materials, Figure S1) to the Co center. This hydrogen is 3 Å from the Co center and, therefore, could slightly disfavor dioxygen coordination relative to 2.
While the comparison of 1 and 2 provides a convenient probe for the influence of second-sphere effects on the reduction of dioxygen, the overall catalytic performance of these complexes is not as high as that of other homogeneous complexes. For example, a recently reported thiolate-ligated dicobalt catalyst had a TOFi = 3000 h−1 for the reduction of O2 to water, using Me8Fc as a reductant and LutHBF4 as the proton source [11]. This TOFi is significantly faster than those of 1 and 2 (TOFi = 70 and 1200 h−1, respectively). Thus, catalysts that incorporate advantageous features of the primary and secondary coordination spheres of these systems might be best suited for achieving a high performance.

4. Materials and Methods

4.1. General Methods and Instrumentation

All chemicals were utilized as purchased from commercial suppliers unless stated otherwise. The PaPy2QH (N,N-Bis(2-pyridylmethyl)amine-N-ethyl-2-quinolinecarboxamide) and PaPy2NH (N-(2-(bis(pyridine-2-ylmethyl)amino)ethyl)-1,8-naphthyridine-2-carboxamide) ligands were synthesized by previously reported methods [26,51].
Electronic absorption measurements were acquired by an Agilent 8453 UV–Vis system (Agilent, Santa Clara, CA, USA) or a Varian Cary 50 Bio UV–Vis spectrophotometer (Varian, Cranford, NJ, USA). The reaction temperature was regulated using a Quantum Northwest TC 1 temperature controller and stirrer (Quantum Northwest, Liberty Lake, WA, USA) or a Unisoku cryostat and stirrer (Unisoku, Hirakata, Japan). 1H NMR measurements were collected with a Bruker DRX 400 MHz spectrometer (Bruker, Billerica, MA, USA). The hyperfine-shifted 1H NMR spectra were obtained over an expanded spectral window (from 150 to −100 ppm) using 1000 scans to enhance S/N. Spline-based baseline correction was obtained using multipoint spline adjustment in the MestReNova software (version 16.0.0-39276). ESI-MS (electrospray ionization mass spectrometry) measurements were carried out on an LCT Premier Micromass electrospray time-of-flight mass spectrometer (Micromass, Cary, NC, USA). The X-band EPR analysis was performed using a Bruker EMXplus spectrometer (Bruker, Billerica, MA, USA) equipped with an Oxford ESR900 continuous-flow liquid helium cryostat and controlled via an Oxford ITC503 temperature unit (Oxford Instruments, Concord, MA, USA).

4.2. Synthesis of 1 and 2

4.2.1. Synthesis of [CoII(PaPy2Q)](OTf) (1)

In a glovebox, 0.351 g (0.88 mmol) of PaPy2QH was dissolved in 20 mL of CH3OH. A solution of 0.085 g (0.88 mmol) of NaOtBu dissolved in 20 mL of CH3OH was added dropwise to the solution of PaPy2QH under vigorous stirring. The mixture was stirred continuously for 15 min. Then, 0.316 g (0.88 mmol) of CoII(OTf)2 in 20 mL of CH3OH was added dropwise to the mixture. A cloudy brown solution was obtained. The mixture was stirred continuously for 24 h. The reaction was stopped, and the resulting brown solution was filtered using a syringe filter. The solvent was removed completely from the filtrate, leaving behind a reddish-brown precipitate. The solid was dissolved in approximately 10 mL of MeCN and layered with ether for recrystallization. After the ether layer and MeCN layer diffused into each other to form a single layer with precipitated solids, the solvent was carefully decanted, leaving behind some reddish-brown solid. The precipitate was dried in vacuo, washed with ether, and repeatedly dried two times. Repeated recrystallization of this reddish-brown solid in MeCN produced 0.462 g of 1 (87% yield). Crystals suitable for X-ray crystallography were obtained by the slow vapor diffusion of ether into a concentrated acetonitrile solution of 1.

4.2.2. Synthesis of Na[CoII(PaPy2N)](OTf)2 (2)

In a glovebox, 0.429 g (1.1 mmol) of PaPy2NH was dissolved in 25 mL of CH3OH. A solution of 0.104 g (1.1 mmol) of NaOtBu dissolved in 20 mL of CH3OH was added dropwise to the solution of PaPy2NH under vigorous stirring. The mixture was stirred continuously for 15 min. Then, 0.385 g (1.1 mmol) of CoII(OTf)2 in 20 mL of CH3OH was added dropwise to the mixture. A pink-brown solution was obtained. The mixture was stirred continuously for 24 h. The reaction was stopped, and the resulting brown solution was filtered using a syringe filter. The solvent was removed completely from the filtrate, leaving behind a reddish-brown precipitate. The solid was dissolved in approximately 15 mL of MeCN and layered with ether for recrystallization. After the ether layer and MeCN diffused into each other to form a single layer with precipitated solids, the solvent was carefully decanted, leaving a reddish-brown solid. The solid was dried in vacuo, washed with ether, and repeatedly dried two times. Repeated recrystallization of this reddish-brown precipitate in MeCN produced 0.524 g of 2 (80% yield). Crystals suitable for X-ray crystallography were obtained by the slow vapor diffusion of ether into a concentrated acetonitrile solution of 2.

4.3. XRD Characterizations

Complete sets of unique reflections were collected with monochromated CuKα radiation for single-domain crystals of both compounds. Totals of 2619 ([CoII(PaPy2Q)][OTf]) and 3931 (Na[CoII(PaPy2N)][OTf]2) 1–wide ω- or ϕ-scan frames were collected with a Bruker APEX II CCD (Bruker, Billerica, MA, USA)area detector. Frame-counting times of 5–8 s were used for [CoII(PaPy2Q)][OTf] and 4–6 s for Na[CoII(PaPy2N)][OTf]2. X-rays for both structures were provided by a Bruker MicroStar (Bruker, Billerica, MA, USA)microfocus rotating anode operating at 45 kV and 60 mA and equipped with an APEX II CCD (Bruker, Billerica, MA, USA)detector and Helios multilayer X-ray optics. Preliminary lattice constants were obtained with the Bruker program SMART. Integrated reflection intensities for both crystals were produced using the Bruker program SAINT [52]. Both data sets were corrected empirically for variable absorption effects using SADABS [52]. SHELXT [53] was used to solve each structure using intrinsic phasing/dual-space techniques. All stages of weighted full-matrix least-squares refinement were conducted using Fo2 data with SHELXL v2017 [54].
The final structural model for both structures incorporated anisotropic thermal parameters for all non-hydrogen atoms and isotropic thermal parameters for all hydrogen atoms. In order to experimentally distinguish between the PaPy2Q and PaPy2N ligands, the hydrogen atom bonded to the C24 atom in the PaPy2Q ligand of 1 was located from a difference Fourier and included in the structural model as an independent isotropic atom whose parameters were allowed to vary in least-squares refinement cycles. The C24–H24 bond length in 1 was refined to a final value of 0.86(5) Å. No significant electron density was observed at a reasonable sp2-hybridized position around the corresponding nitrogen atom in the PaPy2N ligand of 2. The remaining ligand hydrogen atoms in both structures (except H24 in 1) were placed at idealized riding model sp2- or sp3-hybridized positions with C–H bond lengths of 0.95–0.99 Å. The isotropic thermal parameters of the idealized hydrogen atoms in both structures were fixed at values 1.2 times the equivalent isotropic thermal parameter of the carbon atom to which they are covalently bonded. The relevant crystallographic and structure refinement data from these models for both compounds are given in Tables S1 and S2. Due to physical limitations of the instrument with Cu radiation, it was difficult to obtain triclinic and monoclinic Cu data sets with 100% completeness without remounting the data crystal. Both data sets, therefore, had more than a hundred reflections each that were not measured.
An alternate refinement model for 1 was proposed by a reviewer who noted that “SHELXL uses spherical atomic factors (so called ‘IAM model’) for all atoms. This results in significant bias in hydrogen atom location; among other things: C–H bonds are shorter by 0.1–0.2 A. Therefore, refinement of H atom locations using X-ray data with this model results in a good fit but not in the correct position of hydrogen atoms. The simplest correction for achieving correct hydrogen positions is to assign a (neutron diffraction) value of 1.083 Å to all C–H bonds. Alternatively, non-spherical factors from DFT calculations for each specific structure can be calculated and used for refinement. The resulting bias is less than 0.02 A (which is within the usual standard deviation for X-H of 0.02–0.04 A).” The reviewer ran olex-refine with NoSpherA2 (ORCA5.0, r2SCAN, BASIS SET: def2-TZVP, MULTIPLICITY: 4). The final C24–H24 bond length from this refinement was 1.05(5) Å, a more realistic value than 0.86(5) Å.
Of course, the most important point is that all reasonable models should, and do, in this case, indicate the unequivocal presence of H24 as part of a C–H moiety in 1 and its absence in 2, where the C–H group is replaced by an isoelectronic and chemically useful N atom.

4.4. Catalytic Reduction of Dioxygen

4.4.1. Chemical Catalytic ORR by UV–Vis Spectroscopy

Measurements of UV–Vis spectroscopy were performed in a quartz cuvette with a 1 cm pathlength at 25 °C under air. In total, 300 μL of a 7 mM stock solution of Me10Fc was introduced to 1635 μL of acetonitrile (MeCN) in the cuvette, followed by the addition of 25 μL of TFA from a 1.2 M stock solution. After approximately five seconds, to allow for the mixing of TFA and Me10Fc, 40 μL of the catalyst solution (1 mM of stock solution in MeCN) was added to the mixture. In the control experiments, 40 μL of MeCN was added to the mixture instead of the catalyst solution.

4.4.2. Electrocatalytic ORR by CV

Electrochemical measurements were conducted using a three-electrode system equipped with a glassy carbon working electrode, a platinum wire counter electrode, and a non-aqueous reference electrode (containing supporting electrolyte solution). All solutions were prepared in 0.1 M TBAPF6 in MeCN solution. For each experiment, 50 μL of a 0.3 M TFA stock solution was added to the cell containing 2.8 mL of the supporting electrolyte solution, followed by the addition of 150 μL of a 5 mM catalyst stock solution prepared under an inert atmosphere. This mixture was saturated with oxygen by passing oxygen gas through the solution. In order to evaluate the activity of the catalyst in the absence of oxygen, experiments were carried out with identical conditions under a nitrogen atmosphere.

4.5. Quantification of Produced H2O2 from Chemical Catalytic ORR

To determine the amount of H2O2 generated in each catalytic dioxygen reduction reaction, spectroscopic titrations were performed with an acidic solution of [TiO(TPyPH4)]4+ (Ti-TPyP reagent) [55,56]. The reagent solution was prepared by dissolving [TiO(TPyP)] (≥90%, TCI) (3.44 mg) in 100 mL of 0.05 M aqueous HCl. This solution is light-sensitive and was stored at 5 °C in a refrigerator.
A small portion (15 μL) of each sample solution (collected 10 min after reaction completion) was added to a mixture of the Ti-TPyP reagent (250 μL), 4.8 M perchloric acid (250 μL), and water (235 μL) at 20 °C. After 5 min, the resulting solution was diluted to a final volume of 2.5 mL with water, and the absorbance at λ = 433 nm was measured in a 1 cm pathlength UV–Vis cuvette. In a similar manner, a blank solution was prepared by adding 15 μL of the solvent (acetonitrile) instead of the sample solution. All experiments were repeated three times to obtain reproducible data in the 2% range. A decrease in the sample absorbance is indicative of the presence of H2O2 in the solution. (ΔAsample(i) = Ablank − Asample(i)). To perform the calibration, the same procedure was conducted by adding 15 μL of aqueous solutions of H2O2 (in the range from 100 to 500 μL, prepared via the dilution of a standard solution of 9.66 mM) instead of sample solutions. By plotting the ΔA values (ΔAcal(i)) against the concentration of H2O2 in the final sample solution ([H2O2]final sample) (Supplementary Materials, Figure S13), Equation (4) is obtained as follows:
∆A = 0.1546 [H2O2]final sample + 0.042
The amount of H2O2 in the final solution ([H2O2]final sample) was determined from Asample(i) via Equation (4).

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/molecules30153274/s1, Figure S1: X-ray crystal structure of 2, Figure S2: Top: ESI-MS of 1 and 2. Bottom: UV–Vis spectra of 1 and 2 in CH2Cl2, Figure S3: 1H NMR of 1 and 2, Figure S4: Perpendicular-mode EPR data of 1 and 2, Figure S5: Evans NMR spectra of 1 and 2, Figure S6: UV–vis spectra of control experiment, Figure S7: linear initial rate fit of time profile of ORR in the presence of 1 and 2, Figure S8: Time profile of catalytic ORR after multiple addition of Me10Fc, Figure S9: UV–Vis spectra of ORR by 2 at −40 °C, Figure S10: UV–Vis spectra of H2O2 reduction in the presence of 2 at 25 °C under N2, Figure S11: UV–Vis spectra of H2O2 reduction in the absence of catalyst at 25 °C under N2, Figure S12: Background cyclic voltammograms of TFA under nitrogen and under oxygen, Figure S13: Plot of ΔAcal(i) vs. [H2O2]final sample, Table S1: Crystal data and structure refinement for [Co(C24H22N5O)][O3SCF3] (1), Table S2: Crystal data and structure refinement for [Na][Co(C23H21N6O)][O3SCF3]2 (2).

Author Contributions

Conceptualization and methodology Z.A., A.A.O. and T.A.J.; formal analysis, Z.A., A.A.O. and V.W.D.; investigation, Z.A. and A.A.O.; writing—original draft preparation, Z.A., A.A.O. and T.A.J.; writing—review and editing, Z.A. and T.A.J.; project administration, T.A.J.; funding acquisition, T.A.J. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the U.S. National Science Foundation (CHE-2154955 to T.A.J.). EPR spectra were acquired on an instrument obtained through U.S. NSF Grant CHE-0946883. Support for the NMR instrumentation was provided by NIH Shared Instrumentation Grant #S10OD016360. XRD data were obtained on an instrument obtained through U.S. NSF Grant CHE-0923449.

Data Availability Statement

Crystallographic data are available through the Cambridge Crystallographic Data Centre (CCDC) at https://www.ccdc.cam.ac.uk/ with structure codes 2465463 (complex 1), and 2465462 (complex 2).

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Zhang, Y.-L.; Goh, K.; Zhao, L.; Sui, X.-L.; Gong, X.-F.; Cai, J.-J.; Zhou, Q.-Y.; Zhang, H.-D.; Li, L.; Kong, F.-R.; et al. Advanced Non-Noble Materials in Bifunctional Catalysts for ORR and OER toward Aqueous Metal–Air Batteries. Nanoscale 2020, 12, 21534–21559. [Google Scholar] [CrossRef]
  2. Pegis, M.L.; Wise, C.F.; Martin, D.J.; Mayer, J.M. Oxygen Reduction by Homogeneous Molecular Catalysts and Electrocatalysts. Chem. Rev. 2018, 118, 2340–2391. [Google Scholar] [CrossRef]
  3. Siahrostami, S.; Verdaguer-Casadevall, A.; Karamad, M.; Deiana, D.; Malacrida, P.; Wickman, B.; Escudero-Escribano, M.; Paoli, E.A.; Frydendal, R.; Hansen, T.W.; et al. Enabling Direct H2O2 Production through Rational Electrocatalyst Design. Nat. Mater. 2013, 12, 1137–1143. [Google Scholar] [CrossRef] [PubMed]
  4. Debe, M.K. Electrocatalyst Approaches and Challenges for Automotive Fuel Cells. Nature 2012, 486, 43–51. [Google Scholar] [CrossRef] [PubMed]
  5. Bhunia, S.; Ghatak, A.; Dey, A. Second Sphere Effects on Oxygen Reduction and Peroxide Activation by Mononuclear Iron Porphyrins and Related Systems. Chem. Rev. 2022, 122, 12370–12426. [Google Scholar] [CrossRef] [PubMed]
  6. Whittaker, J.W. Free Radical Catalysis by Galactose Oxidase. Chem. Rev. 2003, 103, 2347–2364. [Google Scholar] [CrossRef] [PubMed]
  7. Solomon, E.I.; Heppner, D.E.; Johnston, E.M.; Ginsbach, J.W.; Cirera, J.; Qayyum, M.; Kieber-Emmons, M.T.; Kjaergaard, C.H.; Hadt, R.G.; Tian, L. Copper Active Sites in Biology. Chem. Rev. 2014, 114, 3659–3853. [Google Scholar] [CrossRef]
  8. Das, A.; Ali, A.; Gupta, G.; Santra, A.; Jain, P.; Ingole, P.P.; Paul, S.; Paria, S. Catalytic Four-Electron Reduction of Oxygen to Water by a Molecular Cobalt Complex Consisting of a Proton Exchanging Site at the Secondary Coordination Sphere. ACS Catal. 2023, 13, 5285–5297. [Google Scholar] [CrossRef]
  9. Luo, J.; Tian, X.; Zeng, J.; Li, Y.; Song, H.; Liao, S. Limitations and Improvement Strategies for Early-Transition-Metal Nitrides as Competitive Catalysts toward the Oxygen Reduction Reaction. ACS Catal. 2016, 6, 6165–6174. [Google Scholar] [CrossRef]
  10. Gewirth, A.A.; Varnell, J.A.; DiAscro, A.M. Nonprecious Metal Catalysts for Oxygen Reduction in Heterogeneous Aqueous Systems. Chem. Rev. 2018, 118, 2313–2339. [Google Scholar] [CrossRef]
  11. Sun, L.; Gutierrez, J.; Goff, A.L.; Gatineau, D.; Jackson, T.A.; Gennari, M.; Duboc, C. A Non-Macrocycle Thiolate-Based Cobalt Catalyst for Selective O2 Reduction into H2O. ChemCatChem 2024, 16, e202400270. [Google Scholar] [CrossRef]
  12. Cook, E.N.; Machan, C.W. Bioinspired Mononuclear Mn Complexes for O2 Activation and Biologically Relevant Reactions. Dalton Trans. 2021, 50, 16871–16886. [Google Scholar] [CrossRef] [PubMed]
  13. McCormick, M.J.; Machan, C.W. Developing Homogeneous First Row Early Transition Metal Catalysts for the Oxygen Reduction Reaction. Dalton Trans. 2024, 53, 16807–16814. [Google Scholar] [CrossRef]
  14. Nishiori, D.; Menzel, J.P.; Armada, N.; Reyes Cruz, E.A.; Nannenga, B.L.; Batista, V.S.; Moore, G.F. Breaking a Molecular Scaling Relationship Using an Iron–Iron Fused Porphyrin Electrocatalyst for Oxygen Reduction. J. Am. Chem. Soc. 2024, 146, 11622–11633. [Google Scholar] [CrossRef]
  15. Bhunia, S.; Ghatak, A.; Rana, A.; Dey, A. Amine Groups in the Second Sphere of Iron Porphyrins Allow for Higher and Selective 4e–/4H+ Oxygen Reduction Rates at Lower Overpotentials. J. Am. Chem. Soc. 2023, 145, 3812–3825. [Google Scholar] [CrossRef]
  16. Martin, D.J.; Mayer, J.M. Oriented Electrostatic Effects on O2 and CO2 Reduction by a Polycationic Iron Porphyrin. J. Am. Chem. Soc. 2021, 143, 11423–11434. [Google Scholar] [CrossRef]
  17. Bhunia, S.; Rana, A.; Ghosh Dey, S.; Ivancich, A.; Dey, A. A Designed Second-Sphere Hydrogen-Bond Interaction That Critically Influences the O–O Bond Activation for Heterolytic Cleavage in Ferric Iron–Porphyrin Complexes. Chem. Sci. 2020, 11, 2681–2695. [Google Scholar] [CrossRef] [PubMed]
  18. Ghatak, A.; Bhakta, S.; Bhunia, S.; Dey, A. Influence of the Distal Guanidine Group on the Rate and Selectivity of O2 Reduction by Iron Porphyrin. Chem. Sci. 2019, 10, 9692–9698. [Google Scholar] [CrossRef]
  19. Bhunia, S.; Rana, A.; Roy, P.; Martin, D.J.; Pegis, M.L.; Roy, B.; Dey, A. Rational Design of Mononuclear Iron Porphyrins for Facile and Selective 4e–/4H+ O2 Reduction: Activation of O–O Bond by 2nd Sphere Hydrogen Bonding. J. Am. Chem. Soc. 2018, 140, 9444–9457. [Google Scholar] [CrossRef]
  20. Dinda, S.; Siddiqui, A.I.; Behera, S.; Mondal, B. Demystifying the Role of Covalently Attached Electron–Proton Transfer Mediator in Oxygen Reduction Reaction Catalyst. ACS Catal. 2023, 13, 12643–12647. [Google Scholar] [CrossRef]
  21. Sinha, S.; Ghosh, M.; Warren, J.J. Changing the Selectivity of O2 Reduction Catalysis with One Ligand Heteroatom. ACS Catal. 2019, 9, 2685–2691. [Google Scholar] [CrossRef]
  22. Xu, Y.; Jin, X.; Zhang, J.; Qin, H.; Mei, B.; Liu, T.; Lei, S.; Li, S.; Bo, Y.; Li, X.; et al. Regulating Steric Effect of Cobalt Corroles for Promoted Oxygen Electrocatalysis. ACS Catal. 2024, 14, 14350–14355. [Google Scholar] [CrossRef]
  23. Gao, Y.; Lei, H.; Bao, Z.; Liu, X.; Qin, L.; Yin, Z.; Li, H.; Huang, S.; Zhang, W.; Cao, R. Electrocatalytic Oxygen Reduction with Cobalt Corroles Bearing Cationic Substituents. Phys. Chem. Chem. Phys. 2023, 25, 4604–4610. [Google Scholar] [CrossRef] [PubMed]
  24. Nichols, A.W.; Cook, E.N.; Gan, Y.J.; Miedaner, P.R.; Dressel, J.M.; Dickie, D.A.; Shafaat, H.S.; Machan, C.W. Pendent Relay Enhances H2O2 Selectivity during Dioxygen Reduction Mediated by Bipyridine-Based Co–N2O2 Complexes. J. Am. Chem. Soc. 2021, 143, 13065–13073. [Google Scholar] [CrossRef]
  25. Das, A.; Santra, A.; Kumari, A.; Ghosh, D.; Paria, S. Breaking the Scaling Relationship for Oxygen Reduction Reaction Using Molecular Cobalt Complexes. J. Am. Chem. Soc. 2025, 147, 6549–6560. [Google Scholar] [CrossRef]
  26. Opalade, A.A.; Hessefort, L.; Day, V.W.; Jackson, T.A. Controlling the Reactivity of a Metal-Hydroxo Adduct with a Hydrogen Bond. J. Am. Chem. Soc. 2021, 143, 15159–15175. [Google Scholar] [CrossRef]
  27. Chang, C.J.; Loh, Z.-H.; Shi, C.; Anson, F.C.; Nocera, D.G. Targeted Proton Delivery in the Catalyzed Reduction of Oxygen to Water by Bimetallic Pacman Porphyrins. J. Am. Chem. Soc. 2004, 126, 10013–10020. [Google Scholar] [CrossRef]
  28. McGuire, R., Jr.; Dogutan, D.K.; Teets, T.S.; Suntivich, J.; Shao-Horn, Y.; Nocera, D.G. Oxygen Reduction Reactivity of Cobalt(II) Hangman Porphyrins. Chem. Sci. 2010, 1, 411–414. [Google Scholar] [CrossRef]
  29. Gliemann, G. A. B. P. Lever: Inorganic Electronic Spectroscopy, Vol. 33 Aus: Studies in Physical and Theoretical Chemistry, Elsevier, Amsterdam, Oxford, New York, Tokio 1984. 863 Seiten, Preis: $113, 50. Berichte Bunsenges. Phys. Chem. 1985, 89, 99–100. [Google Scholar] [CrossRef]
  30. Wang, C.-C.; Chang, H.-C.; Lai, Y.-C.; Fang, H.; Li, C.-C.; Hsu, H.-K.; Li, Z.-Y.; Lin, T.-S.; Kuo, T.-S.; Neese, F.; et al. A Structurally Characterized Nonheme Cobalt–Hydroperoxo Complex Derived from Its Superoxo Intermediate via Hydrogen Atom Abstraction. J. Am. Chem. Soc. 2016, 138, 14186–14189. [Google Scholar] [CrossRef]
  31. Stoufer, R.C.; Smith, D.W.; Clevenger, E.A.; Norris, T.E. Complexes of Cobalt(II). I. On the Anomalous Magnetic Behavior of Some Six-Coordinate Cobalt(II) Complexes. Inorg. Chem. 1966, 5, 1167–1171. [Google Scholar] [CrossRef]
  32. Thuery, P.; Zarembowitch, J. Spin State of Cobalt(II) in Five- and Six-Coordinate Lewis Base Adducts of N,N’-Ethylenebis(3-Carboxysalicylaldiminato)Cobalt(II). New Spin-Crossover Complexes. Inorg. Chem. 1986, 25, 2001–2008. [Google Scholar] [CrossRef]
  33. Wang, L.; Gennari, M.; Cantú Reinhard, F.G.; Gutiérrez, J.; Morozan, A.; Philouze, C.; Demeshko, S.; Artero, V.; Meyer, F.; de Visser, S.P.; et al. A Non-Heme Diiron Complex for (Electro)Catalytic Reduction of Dioxygen: Tuning the Selectivity through Electron Delivery. J. Am. Chem. Soc. 2019, 141, 8244–8253. [Google Scholar] [CrossRef]
  34. Mangue, J.; Gondre, C.; Pécaut, J.; Duboc, C.; Ménage, S.; Torelli, S. Controlled O2 Reduction at a Mixed-Valent (II,I) Cu2S Core. Chem. Commun. 2020, 56, 9636–9639. [Google Scholar] [CrossRef]
  35. Lin, J.; Xin, Q.; Gao, X. Real-Time Detection of Hydrogen Peroxide Using Microelectrodes in an Ultrasonic Enhanced Heterogeneous Fenton Process Catalyzed by Ferrocene. Environ. Sci. Pollut. Res. Int. 2015, 22, 11170–11174. [Google Scholar] [CrossRef]
  36. Lubach, J.; Drenth, W. Enolization and Oxidation: II. Oxidation of Ferrocene by Molecular Oxygen and Hydrogen Peroxide in Acidic Media. Recl. Trav. Chim. Pays-Bas 1973, 92, 586–592. [Google Scholar] [CrossRef]
  37. Fomin, F.M.; Zaitseva, K.S. Mechanism of the Oxidation of Ferrocene and Its Derivatives with Hydrogen Peroxide: A New Effect. Russ. J. Phys. Chem. 2014, 88, 466–470. [Google Scholar] [CrossRef]
  38. Zhang, X.; Chen, G.-S.; Liu, H.-C.; Zhu, M.-J.; Xie, M.-Y.; Cen, M.-S.; Li, Q.-J.; Wang, T.-S.; Zhang, H.-X. Bidirectional O2 Reduction/H2O Oxidation Boosted by a Pentadentate Pyridylalkylamine Copper(II) Complex. Electrochimica Acta 2023, 446, 142099. [Google Scholar] [CrossRef]
  39. Passard, G.; Dogutan, D.K.; Qiu, M.; Costentin, C.; Nocera, D.G. Oxygen Reduction Reaction Promoted by Manganese Porphyrins. ACS Catal. 2018, 8, 8671–8679. [Google Scholar] [CrossRef]
  40. Yoshikawa, S.; Shimada, A. Reaction Mechanism of Cytochrome c Oxidase. Chem. Rev. 2015, 115, 1936–1989. [Google Scholar] [CrossRef] [PubMed]
  41. Bullock, R.M.; Chen, J.G.; Gagliardi, L.; Chirik, P.J.; Farha, O.K.; Hendon, C.H.; Jones, C.W.; Keith, J.A.; Klosin, J.; Minteer, S.D.; et al. Using Nature’s Blueprint to Expand Catalysis with Earth-Abundant Metals. Science 2020, 369, eabc3183. [Google Scholar] [CrossRef] [PubMed]
  42. Cook, E.N.; Machan, C.W. Homogeneous Catalysis of Dioxygen Reduction by Molecular Mn Complexes. Chem. Commun. 2022, 58, 11746–11761. [Google Scholar] [CrossRef]
  43. Ghosh, A.C.; Duboc, C.; Gennari, M. Synergy between Metals for Small Molecule Activation: Enzymes and Bio-Inspired Complexes. Coord. Chem. Rev. 2021, 428, 213606. [Google Scholar] [CrossRef]
  44. Passard, G.; Ullman, A.M.; Brodsky, C.N.; Nocera, D.G. Oxygen Reduction Catalysis at a Dicobalt Center: The Relationship of Faradaic Efficiency to Overpotential. J. Am. Chem. Soc. 2016, 138, 2925–2928. [Google Scholar] [CrossRef]
  45. Monte-Pérez, I.; Kundu, S.; Chandra, A.; Craigo, K.E.; Chernev, P.; Kuhlmann, U.; Dau, H.; Hildebrandt, P.; Greco, C.; Van Stappen, C.; et al. Temperature Dependence of the Catalytic Two- versus Four-Electron Reduction of Dioxygen by a Hexanuclear Cobalt Complex. J. Am. Chem. Soc. 2017, 139, 15033–15042. [Google Scholar] [CrossRef]
  46. Wang, Y.-H.; Pegis, M.L.; Mayer, J.M.; Stahl, S.S. Molecular Cobalt Catalysts for O2 Reduction: Low-Overpotential Production of H2O2 and Comparison with Iron-Based Catalysts. J. Am. Chem. Soc. 2017, 139, 16458–16461. [Google Scholar] [CrossRef]
  47. Wang, Y.-H.; Goldsmith, Z.K.; Schneider, P.E.; Anson, C.W.; Gerken, J.B.; Ghosh, S.; Hammes-Schiffer, S.; Stahl, S.S. Kinetic and Mechanistic Characterization of Low-Overpotential, H2O2-Selective Reduction of O2 Catalyzed by N2O2-Ligated Cobalt Complexes. J. Am. Chem. Soc. 2018, 140, 10890–10899. [Google Scholar] [CrossRef]
  48. Wang, Y.-H.; Mondal, B.; Stahl, S.S. Molecular Cobalt Catalysts for O2 Reduction to H2O2: Benchmarking Catalyst Performance via Rate–Overpotential Correlations. ACS Catal. 2020, 10, 12031–12039. [Google Scholar] [CrossRef]
  49. Lv, B.; Li, X.; Guo, K.; Ma, J.; Wang, Y.; Lei, H.; Wang, F.; Jin, X.; Zhang, Q.; Zhang, W.; et al. Controlling Oxygen Reduction Selectivity through Steric Effects: Electrocatalytic Two-Electron and Four-Electron Oxygen Reduction with Cobalt Porphyrin Atropisomers. Angew. Chem. Int. Ed. 2021, 60, 12742–12746. [Google Scholar] [CrossRef] [PubMed]
  50. Yang, J.; Li, P.; Li, X.; Xie, L.; Wang, N.; Lei, H.; Zhang, C.; Zhang, W.; Lee, Y.-M.; Zhang, W.; et al. Crucial Roles of a Pendant Imidazole Ligand of a Cobalt Porphyrin Complex in the Stoichiometric and Catalytic Reduction of Dioxygen. Angew. Chem. Int. Ed. 2022, 61, e202208143. [Google Scholar] [CrossRef]
  51. Eroy-Reveles, A.A.; Leung, Y.; Beavers, C.M.; Olmstead, M.M.; Mascharak, P.K. Near-Infrared Light Activated Release of Nitric Oxide from Designed Photoactive Manganese Nitrosyls:  Strategy, Design, and Potential as NO Donors. J. Am. Chem. Soc. 2008, 130, 4447–4458. [Google Scholar] [CrossRef]
  52. Krause, L.; Herbst-Irmer, R.; Sheldrick, G.M.; Stalke, D. Comparison of Silver and Molybdenum Microfocus X-Ray Sources for Single-Crystal Structure Determination. J. Appl. Crystallogr. 2015, 48, 3–10. [Google Scholar] [CrossRef]
  53. Sheldrick, G.M. SHELXT–Integrated Space-Group and Crystal-Structure Determination. Acta Crystallogr. Sect. A Found. Adv. 2015, 71, 3–8. [Google Scholar] [CrossRef] [PubMed]
  54. Sheldrick, G.M. Crystal Structure Refinement with SHELXL. Acta Crystallogr. Sect. C Struct. Chem. 2015, 71, 3–8. [Google Scholar] [CrossRef] [PubMed]
  55. Matsubara, C.; Kawamoto, N.; Takamura, K. Oxo[5,10,15,20-Tetra(4-Pyridyl)Porphyrinato]Titanium(IV): An Ultra-High Sensitivity Spectrophotometric Reagent for Hydrogen Peroxide. Analyst 1992, 117, 1781–1784. [Google Scholar] [CrossRef]
  56. Takamura, K.; Matsubara, C.; Matsumoto, T. Reaction Specificity of a Titanium(IV)-Porphyrin Complex to Hydrogen Peroxide in View of an Ab Initio Study. Anal. Sci. 2008, 24, 401–404. [Google Scholar] [CrossRef]
Molecules 30 03274 sch001
Scheme 1. Structures of Co ORR catalysts with second-sphere Lewis bases. Top (a,b): Machan complexes based on N2O2 ligand [24]. Bottom (c–h): Paria complexes supported by bis-pyridine-bis-oxime framework [8,25].
Scheme 1. Structures of Co ORR catalysts with second-sphere Lewis bases. Top (a,b): Machan complexes based on N2O2 ligand [24]. Bottom (c–h): Paria complexes supported by bis-pyridine-bis-oxime framework [8,25].
Molecules 30 03274 sch001
Molecules 30 03274 sch002
Scheme 2. Structures of PaPy2QH and PaPy2NH ligands (left) and complexes 1 and 2 (right).
Scheme 2. Structures of PaPy2QH and PaPy2NH ligands (left) and complexes 1 and 2 (right).
Molecules 30 03274 sch002
Molecules 30 03274 g001
Figure 1. X-ray crystal structures of 1 (left) and 2 (right) showing 50% probability thermal ellipsoids. The counter ions, solvent molecules, and hydrogen atoms were removed for clarity.
Figure 1. X-ray crystal structures of 1 (left) and 2 (right) showing 50% probability thermal ellipsoids. The counter ions, solvent molecules, and hydrogen atoms were removed for clarity.
Molecules 30 03274 g001
Molecules 30 03274 g002
Figure 2. Electronic absorption spectra of 2.0 mM solutions of 1 (left) and 2 (right) in MeCN at 25 °C.
Figure 2. Electronic absorption spectra of 2.0 mM solutions of 1 (left) and 2 (right) in MeCN at 25 °C.
Molecules 30 03274 g002
Molecules 30 03274 g003
Figure 3. UV–Vis spectrum of the formation of Me10Fc+ in the presence of a catalytic amount (0.02 mM) of 1 (left) and 2 (right) in MeCN under air. The initial spectra are in purple, intermediate spectra are gray, dashed traces, and final spectra are blue (left) and red (right). Conditions: 1.1 mM Me10Fc and 15 mM TFA at 25 °C. Insets: Time trace of the formation of Me10Fc+ at 780 nm.
Figure 3. UV–Vis spectrum of the formation of Me10Fc+ in the presence of a catalytic amount (0.02 mM) of 1 (left) and 2 (right) in MeCN under air. The initial spectra are in purple, intermediate spectra are gray, dashed traces, and final spectra are blue (left) and red (right). Conditions: 1.1 mM Me10Fc and 15 mM TFA at 25 °C. Insets: Time trace of the formation of Me10Fc+ at 780 nm.
Molecules 30 03274 g003
Molecules 30 03274 g004
Figure 4. (Left): CVs of 0.2 mM solutions of complexes 1 (red) and 2 (blue) in MeCN, under nitrogen, in the presence of TFA (5 mM). Inset: Zoomed in cyclic voltammograms under nitrogen. (Right): Cyclic voltammograms of 0.1 mM solutions of the complexes 1 (red trace) and 2 (blue trace) in MeCN saturated with oxygen, in the presence of TFA (5 mM). All CVs are conducted using a 3 mm glassy carbon working electrode and a Pt wire counter electrode at a scan rate of 0.1 V/s.
Figure 4. (Left): CVs of 0.2 mM solutions of complexes 1 (red) and 2 (blue) in MeCN, under nitrogen, in the presence of TFA (5 mM). Inset: Zoomed in cyclic voltammograms under nitrogen. (Right): Cyclic voltammograms of 0.1 mM solutions of the complexes 1 (red trace) and 2 (blue trace) in MeCN saturated with oxygen, in the presence of TFA (5 mM). All CVs are conducted using a 3 mm glassy carbon working electrode and a Pt wire counter electrode at a scan rate of 0.1 V/s.
Molecules 30 03274 g004
Molecules 30 03274 g005
Figure 5. Top left (a): CV of 0.1 mM complex 1, in a solution of 5 mM TFA in MeCN saturated with oxygen at different scan rates. Down left (c): CV of 0.1 mM complex 2, saturated with oxygen in a solution of 5 mM TFA in MeCN at different scan rates. Right (b,d): A plot of catalytic current vs. square root of scan rate of complex 1 (b) and 2 (d).
Figure 5. Top left (a): CV of 0.1 mM complex 1, in a solution of 5 mM TFA in MeCN saturated with oxygen at different scan rates. Down left (c): CV of 0.1 mM complex 2, saturated with oxygen in a solution of 5 mM TFA in MeCN at different scan rates. Right (b,d): A plot of catalytic current vs. square root of scan rate of complex 1 (b) and 2 (d).
Molecules 30 03274 g005
Molecules 30 03274 sch003
Scheme 3. General proposed mechanism for ORR by mononuclear Co complexes.
Scheme 3. General proposed mechanism for ORR by mononuclear Co complexes.
Molecules 30 03274 sch003
Table 1. Cobalt–ligand bond lengths (Å) from X-ray crystallography, UV–Vis absorption band maxima (nm), and intensities (M−1 cm−1), 1H NMR chemical shifts (ppm), EPR g values, and solution magnetic moments (μB), for complexes 1 and 2.
Table 1. Cobalt–ligand bond lengths (Å) from X-ray crystallography, UV–Vis absorption band maxima (nm), and intensities (M−1 cm−1), 1H NMR chemical shifts (ppm), EPR g values, and solution magnetic moments (μB), for complexes 1 and 2.
Property12
L = PaPy2QL = PaPy2N
Cobalt–ligand bond lengths
Co–N12.107 (3)2.0806 (19)
Co–N21.935 (3)1.959 (2)
Co–N32.182 (3)2.1914 (19)
Co–N42.036 (3)2.041 (2)
Co–N52.058 (3)2.0503 (19)
λmax (ε)650 (50)650 (20)
540 (180)545 (115)
460 (150)450 (190)
1H NMR149.1141.5
132.6121.2
79.995.7
77.490.4
65.673.0
55.565.9
53.953.8
44.345.8
−1.5−0.4
−7.4−5.9
g~4.4~4.4
~2.0~1.9
μeff2.62.9
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Aghaei, Z.; Opalade, A.A.; Day, V.W.; Jackson, T.A. Oxygen Reduction by Amide-Ligated Cobalt Complexes: Effect of Hydrogen Bond Acceptor. Molecules 2025, 30, 3274. https://doi.org/10.3390/molecules30153274

AMA Style

Aghaei Z, Opalade AA, Day VW, Jackson TA. Oxygen Reduction by Amide-Ligated Cobalt Complexes: Effect of Hydrogen Bond Acceptor. Molecules. 2025; 30(15):3274. https://doi.org/10.3390/molecules30153274

Chicago/Turabian Style

Aghaei, Zahra, Adedamola A. Opalade, Victor W. Day, and Timothy A. Jackson. 2025. "Oxygen Reduction by Amide-Ligated Cobalt Complexes: Effect of Hydrogen Bond Acceptor" Molecules 30, no. 15: 3274. https://doi.org/10.3390/molecules30153274

APA Style

Aghaei, Z., Opalade, A. A., Day, V. W., & Jackson, T. A. (2025). Oxygen Reduction by Amide-Ligated Cobalt Complexes: Effect of Hydrogen Bond Acceptor. Molecules, 30(15), 3274. https://doi.org/10.3390/molecules30153274

Article Metrics

Back to TopTop