Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2023 Feb 16;13(5):3109–3119. doi: 10.1021/acscatal.2c05951

Exploring the Parameters Controlling Product Selectivity in Electrochemical CO2 Reduction in Competition with Hydrogen Evolution Employing Manganese Bipyridine Complexes

Wanwan Hong †, Mahika Luthra ‡, Joakim B Jakobsen †, Monica R Madsen †, Abril C Castro ‡, Hans Christian D Hammershøj †, Steen U Pedersen §, David Balcells ‡, Troels Skrydstrup ††,*, Kim Daasbjerg ∥,*, Ainara Nova ‡,⊥,*
PMCID: PMC9990071  PMID: 36910875

Abstract

graphic file with name cs2c05951_0012.jpg

Selective reduction of CO2 is an efficient solution for producing nonfossil-based chemical feedstocks and simultaneously alleviating the increasing atmospheric concentration of this greenhouse gas. With this aim, molecular electrocatalysts are being extensively studied, although selectivity remains an issue. In this work, a combined experimental–computational study explores how the molecular structure of Mn-based complexes determines the dominant product in the reduction of CO2 to HCOOH, CO, and H2. In contrast to previous Mn(bpy-R)(CO)3Br catalysts containing alkyl amines in the vicinity of the Br ligand, here, we report that bpy-based macrocycles locking these amines at the side opposite to the Br ligand change the product selectivity from HCOOH to H2. Ab initio molecular dynamics simulations of the active species showed that free rotation of the Mn(CO)3 moiety allows for the approach of the protonated amine to the reactive center yielding a Mn-hydride intermediate, which is the key in the formation of H2 and HCOOH. Additional studies with DFT methods showed that the macrocyclic moiety hinders the insertion of CO2 to the metal hydride favoring the formation of H2 over HCOOH. Further, our results suggest that the minor CO product observed experimentally is formed when CO2 adds to Mn on the side opposite to the amine ligand before protonation. These results show how product selectivity can be modulated by ligand design in Mn-based catalysts, providing atomistic details that can be leveraged in the development of a fully selective system.

Keywords: carbon dioxide reduction, hydrogen evolution reaction, manganese bipyridine complexes, density functional theory calculations, mechanisms, electrocatalysis

Introduction

Catalysis is crucial for modern society and life in general. Catalytic processes occur in the mitochondria, where glucose is converted into energy, in refineries, where crude oil is converted into fuels, in the polymer industry, facilitating the production of plastics for computers, houses, and clothes, and in the pharmaceutical industry, where vital medicine is produced. In fact, 85% of all products manufactured have been produced with the assistance of a catalyst, and in 90% of all chemical processes, at least one catalyst is employed.1 Despite the huge number of existing catalysts, it is still of high priority to develop new and more efficient ones that with low cost, low energy consumption, and low environmental impact selectively convert building blocks into target compounds.2 Electrocatalysts show great promise in this respect, as they can be powered by renewable energy sources while operating at ambient temperature and pressure.3 For instance, electrocatalysts are widely used in the electrolysis of water, where water is decomposed into oxygen and hydrogen gas, of which the latter is a promising carbon-neutral fuel that does not emit harmful exhaust gases.4,5 Furthermore, research regarding electrochemical CO2 reduction is advancing with the prospect of decreasing CO2 pollution while converting the greenhouse gas into value-added fuels and chemicals.6−8

A recurring challenge within catalysis is to obtain a high selectivity toward a single product. Within the family of metal-based molecular catalysts, a lot of research has focused on secondary coordination sphere effects and how different functional groups influence the selectivity of the complexes.12−16 Recently, we showed how the selectivity of CO2 reduction by Mn bipyridine (bpy) catalysts (1a*–c*, Figure 1) changes from CO, when hydroxyl (1b*) or alkyl groups (1c*) are close to the metal center, to HCOOH, when amines (1a*) are placed in the vicinity of the metal.9 As presented in Scheme 1, the amines function as proton shuttles, transferring protons from the ligand to the metal center, thereby facilitating the formation of a Mn hydride, 4, which is the key intermediate in the HCOO–/HCOOH and H2 pathways. Additionally, the presence of the amine functionalities is also known to stabilize the carboxylate intermediate, 7, by forming an intramolecular hydrogen bond, which then favors CO formation (blue pathway, Scheme 1).17−19 Since amines accelerate all three pathways, the selectivity originates from other contributing factors, such as the nature of the metal,20 geometrical structure,21 hydricity,22 and pKa.23 In the case of Mn complexes, CO has been the dominant product for the nonamine-containing complexes (1b* and 1c*).

Figure 1.

Figure 1

Open in a new tab

Chemical structures of reference complexes (1a*, 1b*, and 1c*) reported in our previous work9 and complexes 1a, 1b, and 1c reported herein.

Scheme 1. Pathways for CO (blue), H2 (green), and HCOO– (red) Formation with Mn Bipyridine Complexes Bearing Amines in the Secondary Coordination Sphere Previously Reported and Proposed in This Work.

Scheme 1

Open in a new tab

The arc connects the amine-bearing pendants in the secondary coordination sphere, while only one of the sides is included for simplicity. LO and HO stand for low overpotential and high overpotential, respectively.9−11

Since the abovementioned pathways always compete with each other, especially when there is a bifurcation between the H2 and HCOO– pathways from a common Mn hydride intermediate, we explored how the structure affects the selectivity of the catalysts within the Mn bpy family. The structure has already been shown to affect the catalytic activity, as an unsubstituted Mn bpy complex forms less reactive dimers upon reduction,24 which is avoided by introducing bulky substituents on the ligands.25 However, the fundamental understanding of how the molecular structure and product distribution are linked is still missing. To investigate this correlation more carefully, three different Mn tricarbonyl complexes with macrocyclic bipyridine ligands were synthesized (1a–c, Figure 1). Using our previously reported complex 1a*, bearing tertiary amines in the secondary coordination sphere, as a reference catalyst, here, we aim to explore the effect of locking all the amine groups in a relatively enclosed structure by introducing a macrocyclic linker. Furthermore, we examine if the central heteroatom of the linker (N or O, for complexes 1a and 1b, respectively) affects the product distribution, while the size of the macrocyclic ring is evaluated with the one-atom shorter complex 1c.

With these complexes in hand, the different possible pathways proposed in Scheme 1 are scrutinized to get a clearer mechanistic understanding of the factors affecting the product distribution. Specifically, the order of competing protonation and CO2 insertion reactions is considered, and the energy barriers of these steps are evaluated by density functional theory (DFT) calculations. Furthermore, information about the flexibility of the molecular structures is gained through ab initio molecular dynamics (AIMD). This fundamental study builds on an electrochemical assessment, taking complex 1a as a starting point, in which the key intermediates formed during reduction are identified using infrared spectroelectrochemistry (IR-SEC). Finally, based on the mechanistic insight obtained from our combined experimental and computational studies, we propose a general guide for rational ligand design.

Results and Discussion

Synthesis and Characterization of Complexes

The three macrocyclic complexes (1a–c) were synthesized as exemplified for 1a in Scheme 2. The synthesis was envisioned to proceed through a double reductive amination from dialdehyde A using a small excess of the corresponding di- or triamines. To minimize the formation of larger ring sizes or polymers, the reaction was performed using a low concentration of dialdehyde A (55 mM) in tetrahydrofuran (THF), and a syringe pump was employed to slowly add the amines over five hours (Scheme 2a). After stirring for 48 h, full conversion of dialdehyde A was obtained in all three cases, and the macrocyclic ligands could be obtained in 13–48% yield after purification. Complexes 1a–c were obtained by metalation of the respective ligands using Mn(CO)5Br in THF at 55 °C yielding the complexes as yellow solids in good yields (Scheme 2b). The complexes were analyzed by 1H NMR, 13C NMR, ATR-IR, and high-resolution mass spectrometry (HR-MS). Furthermore, the 1H NMR and 13C NMR chemical shifts of complex 1a were calculated and found to be in good agreement with the experimental values (Table S1).

Scheme 2. Synthetic Protocol Employed Toward Complex 1a.

Scheme 2

Open in a new tab

Figure 2 shows the single-crystal X-ray diffraction structure of 1a, and its crystallographic data are listed in the Supporting Information (Section 2). In the solid state, the Mn center adopts a facial octahedral geometry, having the modified bpy ligand and two carbonyl ligands in the equatorial plane, while the bromide is placed in the axial position together with the third carbonyl group pointing in opposite directions. Interestingly, the Mn–Br bond is in the direction opposite to the aliphatic amine ligand. The bpy plane is slightly distorted (N–C–C–N dihedral angle, φ = 11.6°), while the adjacent phenyl rings are rotated out of the plane 60–80°. The linker between the two phenyl rings is situated below the plane of the bpy ring system. This is an interesting observation as it implies that no amines are in close vicinity to the Mn center, which we previously have shown is important for the product selectivity.9 The single-crystal X-ray diffraction structures of 1b and 1c are shown in Figure 2, and crystallographic data are listed in the Supporting Information (Section 2). The coordination sphere of Mn is facial octahedral for both complexes, and the general observations mentioned above for 1a also hold true for 1b and 1c. Only one isomer with the Br in the opposite direction to the ligand (exo) is observed for 1a–c in solution, which is also supported by DFT calculations (TPSSh-D3/def2SVP//TPSSh-D3/def2TZVP, def2TZVPD, see Section 3 in the Supporting Information for details). It was found that the exo isomers of 1a, 1b, and 1c were 8.4, 8.2, and 7.4 kcal mol–1, respectively, more stable than the corresponding endo isomers, characterized by Br and amine pointing in the same direction (Scheme S1).

Figure 2.

Figure 2

Open in a new tab

Molecular structures of 1a, 1b, and 1c obtained from single-crystal X-ray diffraction studies. Mn = maroon, Br = green, C = gray, N = blue, and O = red; thermal ellipsoids at 50% probability. Hydrogen atoms are omitted for clarity.

AIMD Simulations

Formation of the hydride intermediate in the H2 pathway (Scheme 1) requires the proximity of the chelating-ligand amines to the metal center, which seems to be excluded by the crystal structure of 1a (Figure 2). However, this reaction is preceded by the two-electron reduction of 1a and the concomitant elimination of bromide, yielding an anion species 2a–, and the protonation of the amine moiety of the latter to form 3a (Scheme 1 and Figure 3). Hence, we studied the fluxional behavior of the ligand in these two species by means of DFT AIMD (PBE-D3/DZVP) as implemented in the CP2K program26−28 followed by static DFT calculations.

Figure 3.

Figure 3

Open in a new tab

Time evolution of (top) the Mn···N distances (Å) for 2a– and (bottom) the Mn···N(H) and Mn···H(N) distances for the three protonated isomers of the corresponding 3a-NX (X = M, S1, or S2) complexes. The three amine N atoms of the chelating ligand are labeled as NM (middle one) and NS1 and NS2 (the two on the sides).

Figure 3 shows the time evolution (within a production trajectory of 25 ps) of the Mn···H distances for complex 2a–, involving the three amine N atoms of the ligand: the middle one (NM) and the two on the sides (NS1 and NS2). The side N atoms were initially considered different because the ligand is asymmetric in the crystal structure. However, their dynamic behavior should be equivalent at longer trajectories.

The AIMD simulation of 2a– reflects the dynamic structure of the chelating ligand, in which the variation of the Mn···N distances has wide amplitudes [1.86 (NM), 2.13 (NS1), and 1.54 Å (NS2)] and large average values [6.74 (NM), 4.61 (NS1), and 5.29 Å (NS2)]. These distances are indeed long and similar to those observed in the crystal structure of 1a [7.06 (NM), 4.73 (NS2), and 5.10 Å (NS2)]. Further, the NM atom, which corresponds to the most flexible amine, is located at the position furthest from the metal center. Interestingly, this behavior is reversed when 2a– is protonated to 3a (Figure 3); i.e., the shortest average Mn···N(H) distance is observed when the proton binds to the middle NM atom, yielding an average value of 3.39 Å (measured after the sudden change at ∼2 ps). For all three protonated isomers, 3a-NM, 3a-NS1, and 3a-NS2, the time evolution of the Mn···N(H) distances is strongly correlated to that of the corresponding Mn···H(N).

Therefore, the shortest average Mn···H(N) distance was also observed for NM, with a small value of 2.38 Å, suggesting the presence of a hydrogen bond between Mn and the H–N moiety. This is consistent with the dramatic shortening of the average Mn···NM distance by 3.35 Å upon protonating 2a–, and the narrowing of its amplitude to 0.78 Å. In contrast, the Mn···NS(H) and Mn···H(NS) distances in 3a-NS1/2 fluctuate, as indicated by their large amplitudes in the range of 2.31 and 3.09 Å. This indicates that even though the formation of a hydrogen bond is feasible (shortest Mn···H(NS1) distance = 2.30 Å), this interaction is weaker than that in 3a-NM (shortest Mn···H(NM) distance = 1.87 Å) (Table S2).

The fluxional behavior of 2a– and 3a-NX (X = M, S1, and S2) can be ascribed to the rotation of the Mn(CO)3 core (Figure S1). The AIMD simulation of 2a– shows that this rotation causes an oscillation of the metal axial vacancy in between two positions: one pointing to the triamine bridge (endo) and the other to the opposite direction (exo). In the endo form, the interaction between the N atoms and the vacancy is repulsive because both moieties have lone pairs (Figure S2). Conversely, in the 3a-NM complex, this interaction becomes attractive, which yields a Mn···H–N hydrogen bond. The exo to endo flip can be observed during the initial 5 ps of the AIMD of 3a-NM (Figure 3), in which both the Mn···NM(H) and Mn···H(NM) distances undergo a sudden shortening of ∼3 Å.

The calculations thus show that, despite the long distance observed in the crystal structure of 1a, the middle amine can approach the metal center in the crucial intermediate 3a, facilitating the formation of the hydride complex 4a. The less flexible side positions could also allow the formation of a hydride, but it seems less preferred.

Electrochemical and Spectroscopic Evaluation

Electrochemical experiments were conducted to determine how the macrocyclic bipyridine ligands influence the electrocatalytic properties of the manganese complexes. Cyclic voltammograms of complexes 1a–c are recorded in 0.1 M Bu4NBF4/MeCN solutions under first Ar- and then CO2-saturated conditions (Figure 4 and Figures S3 and S4). Analogously to our findings for complex 1a*–c*,9 all three complexes present one reduction wave at around −1.69 V vs Fc+/Fc under Ar atmosphere. According to the previous mechanistic interpretation,9,25 this reduction wave is the result of a consecutive electron transfer–chemical reaction–electron transfer (ECE) mechanism where the first one-electron reduction of Mn(bpy-R)(CO)3Br (complex 1 in Scheme 1) is followed by the dissociation of Br–, generating a neutral intermediate 2.11 The second electron is transferred from the electrode to 2 instantly since 2 is formed very close to the electrode surface due to the rapid dissociation of Br– and is easier to reduce than 1 itself. Thus, the overall two-electron reduction gives rise to the anionic state 2–,29 as shown in Scheme 1. The process is diffusion-controlled with unaltered electron stoichiometry according to the Randles–Ševčík equation (Figure S5).25

Figure 4.

Figure 4

Open in a new tab

Cyclic voltammograms recorded on 1.5 mM 1a at a GC electrode (diameter = 1 mm) using ν = 0.1 V s–1 in Ar- or CO2-saturated 0.1 M Bu4NBF4/MeCN.

The above interpretation was also confirmed by DFT calculations (see the Supporting Information for further details) taking 1a as an example (Scheme 3). The first reduction from 1a to 1a– has been calculated to take place at E1 = −1.81 V vs Fc+/Fc, after which Br– is easily expelled to form 2a with ΔG = −7.1 kcal mol–1. The analysis of the spin density in 1a– (Table S3) shows that the electron is localized at the bpy ligand as found in similar complexes.30−32 The second reduction generating the catalytically active intermediate 2a– has been computed to occur at E2 = −1.59 V vs Fc+/Fc. Taken together, these three processes yield a two-electron reduction taking place at E3 = −1.60 V vs Fc+/Fc. This overall potential E3 is positively shifted compared to E1 due to the follow-up dissociation of Br–, which is in good accordance with the experimental potential of −1.69 V vs Fc+/Fc.

Scheme 3. Computational Study on the Reduction Path of 1a Comprising Two Kinds of Amine Moieties in the Ligand.

Scheme 3

Open in a new tab

All potential values are calculated relative to Fc+/Fc (see the Supporting Information for details). Protonation steps assume that the proton source is CF3CH2OH/CO2.

Infrared spectroelectrochemistry analysis of 1a, in the absence of CO2 and a proton source, further uncovers the formation of anionic species 2a– since the CO stretches shift from 2021, 1936, and 1906 cm–1 (1a) to 1909 and 1807 cm–1 during the voltammetric sweeping (Figure S6), in alignment with previous reports.25,33,34 In addition, the stretch at 1982 cm–1 and the bump at 1884 cm–1 stem from manganese hydride 4a (vide infra in Schemes 4 and 5).9,35 Its appearance is a consequence of the amine-bearing ligand, which shuttles protons from residual water in the electrolyte to the metal center to generate 4a. Hence, we assign the faint peak at around −2.10 V vs Fc+/Fc in Figure 4 under Ar to the reduction of 4a.36,37 A similar mechanism was outlined by us for complex 1a*.9

Scheme 4. DFT-Calculated Energy Profile for the Reduction of CO2 to Three Different Products: H2 (Green), HCOO– (Red), and CO (Blue) Using the endo Profile and the Middle (NM) Amine Moiety as a Proton Shuttle,,,

Scheme 4

Open in a new tab

The endo profile refers to the pentacoordinate complex with the ligand in the same direction as the metal vacant site.

All energies are considered at a redox potential of −1.60 V vs Fc+/Fc.

Wiggly arrows show the calculated reduction potentials for 4a, 5a, and 8a. Energy profiles after reduction are displayed in Scheme 5 (at −1.87 V vs Fc+/Fc) and Scheme S6 (at −1.95 V vs Fc+/Fc).

Numbers in the energy profile indicate relative Gibbs free energy in kcal mol–1.

Some intermediates have been omitted for clarity. The complete energy profiles are presented in Scheme S9.

Scheme 5. DFT-Calculated Energy Profile Starting from 4a– to Form H2 (Green) and HCOO– (Red) Using the endo Profile and the Middle (NM) Amine Moiety as a Proton Shuttle,

Scheme 5

Open in a new tab

All energies are relative to a redox potential of −1.87 V vs Fc+/Fc.

Numbers in the energy profile indicate relative Gibbs free energy in kcal mol–1.

Now addressing the CO2-saturated electrolyte ([CO2] ≈ 0.28 M),38 significant changes occur to the cyclic voltammetric response. Notably, the oxidation waves of the anionic species disappear (see Figure 4 and Figures S3 and S4). A prepeak appears at −1.55 V vs Fc+/Fc for 1a and 1c, which is associated with the reduction of the solvent-coordinated cation originating from partial solvolysis of 1a and 1c.24 This behavior was also reported for analogous Mn complexes.35,37 In addition, DFT calculations are in agreement with this assignment. We found that the solvent-coordinated complex, 2a+-MeCN, is also generated through a two-electron oxidation process of 2a– at E4 = −1.22 V vs Fc+/Fc followed by the exergonic coordination of MeCN with ΔG = − 6.4 kcal mol–1, which is reduced at a potential of E5 = −1.36 V vs Fc+/Fc (Scheme 3). At the same time, a dramatic current enhancement is detected at −2.10 V vs Fc+/Fc. This is in accordance with a high-overpotential pathway, in which the hydride species 4 is reduced and subsequently enters the catalytic cycle.9 The trace crossing in Figure 4 and Figures S3 and S4 indicates an acceleration of the catalytic behavior on the reverse scan (see Section 5 in the Supporting Information for a detailed explanation).39 The specific catalytic effect of residual water (∼0.034 M) in combination with CO2 (Figure 4) is discussed thoroughly in Section 6 of the Supporting Information. Upon sequential addition of either 2,2,2-trifluoroethanol (TFE, pKa = 35.4 in MeCN,40Figure S7) or 2-propanol (iPrOH, pKa ≈ 42 in MeCN,37Figure S8), the catalytic current continues increasing, until it levels off or even drops once very high concentrations of TFE (2.0 M) or iPrOH (1.0 M) are employed. Thus, under these conditions, the CO2 reduction reaction is independent of the proton concentration but limited by the regeneration of the catalyst, as often observed for catalytic reactions.9,25

Upon introducing CO2 to the IR-SEC experiments of 1a, a mixture of 2a– (1909 and 1807 cm–1) and 4a (1982 and 1884 cm–1) is observed during reduction (Figure S9), where the significant increase in the amount of 4a formed can be attributed to the acidification of the solution induced by CO2, i.e., CO2 increases the proton-donating ability of residual water. Introducing a proton source (TFE or iPrOH) makes the signals assigned to 2a– vanish. In contrast, the signals from 4a are strong, highlighting how proton sources facilitate the hydride generation (Figures S10 and S11). As depicted in Scheme 1, 4 is a vital intermediate in the catalytic cycle, where it either is further protonated to release H2 or combines with CO2 to generate a formato complex 5.22,41,42 A first assessment of the possible CO2 reduction products can be gained from the IR-SEC spectra recorded in the region of 1750–1500 cm–1 with either TFE or iPrOH as a proton source. The bands at 1691 (Figure S12) and 1660 cm–1 (Figure S13) are assigned to the C=O stretches of trifluoroethyl- and isopropyl carbonates, respectively, while the signal at 1609 cm–1 in both cases is attributed to the formation of HCOO–, which is produced when CO2 inserts into the Mn–H bond.9,37,43 In addition, the signal at 1609 cm–1 shows a weaker intensity when TFE is present compared with that in the case of iPrOH, which could be related to the diminished generation of HCOO–.

To gain further insights, we turned to controlled potential electrolysis (CPE) to analyze the product distribution on a longer time scale than cyclic voltammetry and IR-SEC experiments. CPE was performed for 1 h at −2.25 V vs Fc+/Fc (∼150 mV more negative than the reduction of the hydride species) for 1a–c using either 2.0 M TFE or 1.0 M iPrOH as a proton source. All experiments were carried out in a two-chamber H-cell following the same procedure as reported in our previous work.9,20 While the reference complex, 1a*, produces HCOOH (or HCOO–) as the dominant product,9 complexes 1a–c show high selectivity for H2 (Figure 5 and Table S5). In general, only small amounts of HCOO– and CO are produced for all three complexes when TFE is employed as a proton donor; the amounts of HCOO– increase slightly when using the weaker proton donor, iPrOH. This observation can be explained by the competing pathways described in Scheme 1, where 4 either reacts with CO2 to form HCOO– or H+ to produce H2, of which the latter option is favored when stronger proton donors are used.14,35,37,44 This result also suggests that the highest energy barriers for the formation of these two products are very similar (vide infra).

Figure 5.

Figure 5

Open in a new tab

Product distributions obtained after 1 h of electrolysis (at −2.25 V vs Fc+/Fc) of 1.5 mM 1a*(9) or 1a–c in CO2-saturated 0.2 M Bu4NBF4/MeCN containing either 2.0 M TFE (fully colored) or 1.0 M iPrOH (shaded colored) as a proton source.

Specifically, complexes 1a and 1b with the same linker length but different central hetero atoms (N and O, respectively) show both a high Faradaic efficiency for H2 (FEH2) of ∼66% with 2.0 M TFE and ∼48% with 1.0 M iPrOH. The FEHCOO- also displays a decent value of ∼29% for 1a and 1b with 1.0 M iPrOH. These results indicate that, even in the absence of the middle N, the side N in 1b can act as proton shuttle favoring the formation of the intermediate 4b. This is also consistent with the AIMD simulations using 3a-NS1/2 (Figure 3). Shortening the macrocyclic ring by leaving the central heteroatom out (1c) leads to an enhancement of FEH2, which reaches a maximum of 73% with 2.0 M TFE. At the same time, FEHCOO– is suppressed regardless of the proton source. CO is also produced in negligible quantities (FECO = 2–6%) in all cases.

In a control experiment using no catalyst, we noted that only ∼2.9 μmol of H2 was produced in the presence of 2.0 M TFE during the 1 h CPE at -2.25 V vs Fc+/Fc. This is significantly less than the yield of H2 generated in the presence of 1a (Table S6), thus ruling out the possibility that background reactions contribute to the high FEH2. As mentioned earlier, 4 can be detected by IR-SEC both for complex 1a* with an open structure9 and 1a with a closed macrocyclic structure (Figures S10 and S11), making the HCOO–/H2-generating pathways viable. To account for the change in product selectivity with 1a* and 1a–c, the mechanism for the formation of H2, CO, and HCOO– was studied by DFT calculations.

DFT Mechanistic Studies

The mechanism of CO2 reduction starts with the anionic intermediate 2a–, whose lowest energy isomer was selected from AIMD simulations and optimized using static DFT calculations in solvent (Scheme S2). Three scenarios were considered for the formation of the three reaction products (H2, CO, and HCOO–): two starting with the protonation of the amines (NM and NS), which favors the formation of a Mn hydride (4a) in an endo conformation (see Scheme 4 and Schemes S3 and S4), and one starting with the direct CO2 addition or protonation of 2a– in an exo conformation (with CO2, or H, and the amines in opposite sides, see Scheme S5).

Scheme 4 outlines the reaction energy profiles for the formation of H2, CO, and HCOO– considering the protonation of the middle N as the first step, which is the one that provides the lowest energy barrier for the formation of H2. This profile has been poised at a calculated potential of −1.60 V vs Fc+/Fc, which is the potential required to reduce 1a to 2a–, as shown in Scheme 3. As shown previously by us,9 the amine is protonated by TFE with the assistance of CO2 yielding 3a-NM and CF3CH2OCOO–, through an energy barrier of 8.8 kcal mol–1. From here, proton transfer to the metal center to form 4a is preferred over direct CO2 addition to the Mn center to form 7a-NM by 1.2 kcal mol–1. Furthermore, CO2 addition appears to be endergonic and reversible, whereas the proton transfer is exergonic and irreversible. Once the reactive 4a is generated, there are two competing pathways: (i) CO2 insertion into the Mn–H bond leading to the Mn–OCHO formato complex (5a) for final HCOO– generation (HCOO– pathway in red)8,42 and (ii) protonation of the amine moiety facilitating H2 production (H2 pathway in green).22,45 The energy barrier of CO2 insertion (TS-4) to form 10a is 19.2 kcal mol–1; 10a, in which the formate is bonded to Mn by the hydrogen atom (Mn–HCO2) is 18.2 kcal mol–1 less stable than the one bonded via the oxygen atom (5a). Initially, one TS connecting 10a and 5a was found at 8.8 kcal mol–1, with Mn···H and Mn···O distances of 2.49 and 3.27 Å, respectively. However, upon considering the possibility of connecting 5a and 2a+, we found TS-7, which also connects 10a and 5a but with an energy of 2.6 kcal mol–1, and longer Mn···H and Mn···O distances (3.14 and 4.04 Å, respectively). An alternative mechanism for the direct formation of 5a from 3a-NM, as it has been suggested for enzymes,46−48 was also considered without success.

Following the H2 pathway, the energy difference between TS-5 (−2.5 kcal mol–1) and 4a (−11.5 kcal mol–1), which is 9.0 kcal mol–1, shows that the barrier for protonation of the amine to form 6a+-NM is lower than that of CO2 insertion. Liberation of H2, however, requires the proton transfer from the amine to the hydride, which has an energy barrier of 15.1 kcal mol–1 (TS-8). Overall, the formation of H2 and HCOO– has similar barriers, as shown by the energies of TS-8 (6.2 kcal mol–1) and TS-4 (7.7 kcal mol–1), respectively, with generation of H2 being slightly preferred. This result is consistent with the experimental findings that H2 formation is preferred over HCOO– (FEH2 = 66% vs FEHCOO– = 29%). On the other hand, the formation of CO (blue pathway) continues by proton transfer (TS-6) from the Mn–CO2 intermediate (7a-NM) to form a hydroxyl–carbonyl complex, 8a, which is a well-known intermediate in CO2 to CO reduction processes.10,49 Subsequent protonation of 8a (1.1 kcal mol–1) followed by proton transfer, forming a tetracarbonyl adduct and final liberation of CO and H2O (4.0 kcal mol–1), has a barrier of 10 kcal mol–1 but is endergonic by ∼3 kcal mol–1. The CO pathway has consistently higher energies when compared to the H2 and HCOO– pathways.

Another interesting observation from Scheme 4 is that the H2 pathway is reversible relative to 4a. Since the reaction takes place under electrochemical conditions, 4a can be reduced to 4a–, with a calculated redox potential of −1.87 V vs Fc+/Fc (Scheme 4), which is consistent with the appearance of the high current density increment in the cyclic voltammograms under CO2 (−2.10 V vs Fc+/Fc, Figure 4), where 4a– is generated as the bifurcation between the H2 and HCOO– competing pathways. Hence, the pathways for HCOO– and H2 formation were also calculated from this reduced intermediate, 4a– (Scheme 5). It is seen that the energy barriers relative to 4a– for H2 formation (TS-15 = 10.4 kcal mol–1) and CO2 insertion (TS-13 = 14.7 kcal mol–1) are lowered by 7.3 and 4.5 kcal mol–1, respectively, when compared to Scheme 4, where these barriers are 17.7 (TS-8) and 19.2 kcal mol–1 (TS-4) relative to 4a. Consistently, the barrier for H2 formation is lowered more than that for HCOO– generation, leading H2 to be the dominant product. Additionally, it was seen that product formation in Scheme 5 was exergonic and irreversible, pointing to a reaction that takes place via 4a–. This finding is in line with the high overpotential pathway reported for complex 1a*.9

The results shown in Scheme 4 were compared with those considering the protonation of the side amine (Schemes S3 and S4). The main differences observed are as follows: (i) Although the protonation of the side amine is thermodynamically favored over the middle amine (ΔG = −2.5 and 1.0 kcal mol–1, respectively), proton transfer to the Mn center is preferably assisted by the middle amine, with a lower energy barrier (ΔG‡ = 21.2 kcal mol–1 vs ΔG‡ = 13.6 kcal mol–1)—this finding is consistent with the AIMD simulations, which showed that the middle proton gets closer to the Mn center than the protons on the sides; (ii) the barrier for CO2 coordination (CO pathway, TS-18) is lower than that for the amine-to-Mn proton transfer (H2 pathway, TS-17) when the side amine is protonated, whereas the opposite holds true for the middle amine; (iii) after the formation of 4a, both the middle and side amines have similar barriers for the H2 and HCOO– pathways, with a slight preference for the former.

Finally, we compared the energy profiles shown in Scheme 4 with those obtained with the exo isomer formed by the rotation of the Mn(CO)3 core in the AIMD calculations (Scheme S5). Compared with the endo isomer, where the formation of 4a is assisted by the NM amine, the exo isomer must go through a direct protonation of the metal to form 4aex with an energy barrier of 17.2 kcal mol–1, which is 9.7 kcal mol–1 lower than the direct protonation of the metal in the endo isomer (Scheme S9). Conversely, for the CO pathway, a direct addition of CO2 to Mn takes place in the exo isomer with an energy barrier of 14.9 kcal mol–1 (TS-27, Scheme S5), which is favored over the direct addition of CO2 to Mn in the endo isomer by 3.2 kcal mol–1 (TS-33, Scheme S6). As a result, while the formation of 4a (ΔG‡ = 14.6 kcal mol–1) is favored over direct CO2 addition (ΔG‡ = 18.1 kcal mol–1) in the endo case, the opposite happens in the exo profile, with a preference for CO2 addition (ΔG‡ = 14.9 kcal mol–1) over hydride formation (ΔG‡ = 17.2 kcal mol–1). These results suggest that the cyclic ligand, once protonated, forms a cagelike structure that protects the Mn center from CO2 addition, hindering the formation of CO. Despite this, a minor concentration of CO is formed due to the addition of CO2 in the side opposite to the ligand before protonation (exo isomer). This mechanistic picture is in line with the results obtained experimentally in the CPE measurements.

The preference for H2 formation over HCOO– can be ascribed to the same effect. In 4a, the cagelike structure of the ligand hinders the addition of CO2. Indeed, the opposite selectivity is observed with the open system 1a*. To corroborate this hypothesis, we compared the energy barriers obtained for the protonation of hydride 4a and CO2 addition for 1a and 1a* (Scheme S7). These calculations showed that both processes have lower energy barriers in the open system 1a*. The protonation of 4a* has an energy barrier that is 3.9 kcal mol–1 lower than that of complex 4a, while the CO2 addition barrier is significantly lower (by 14.5 kcal mol–1), consistent with a change in selectivity from mainly HCOO– for 1a* to mainly H2 for 1a (Figure 5). This result was further supported by controlled potential electrolysis of 1a* and 1a in the presence of 2.0 M TFE under Ar atmosphere (i.e., without CO2), which yielded H2 as the only product. Similar results were obtained for 1b and 1c as electrocatalysts under these experimental conditions (Table S7). Hence, the less efficient production of HCOO– by 1a–c is, most likely, attributed to the low accessibility of CO2. The macrocyclic ligand around the metal center induces steric hindrance for the insertion of CO2 into the metal hydride. At the same time, protons can still be delivered to it by the amine-bearing ligands, explaining the preference for H2 generation over CO2 reduction. Along this line, it would be expected that the hindrance associated with CO2 insertion to hydride will be incremented for 1c when a more rigid ring is formed by removing the middle nitrogen atom. Indeed, this is reflected experimentally in the lowest FE for HCOO– (3%) obtained for 1c among 1a–c.

Conclusions

Rational design of molecular complexes is crucial for the development of highly selective and efficient catalysts, but the factors affecting these parameters are not uncovered completely yet. This work addressed how the geometric structure of Mn-based complexes affects the competing protonation and CO2 insertion steps in the electrocatalytic CO2 reduction reaction. Depending on the sequence of these steps, different catalytic intermediates are formed, leading to the formation of CO, when CO2 reacts directly with the metal center, or HCOO–/H2 if a metal hydride is formed during the first step.

To investigate how spatial geometry around the catalytic site affects the formation of these products, three different Mn(bpy-R)(CO)3Br complexes (1a–c, Figure 1) were designed and synthesized. Via a combined theoretical and experimental study, we found several parameters that influence the selectivity of the catalysts. First, ab initio molecular dynamics simulations revealed that the three carbonyl groups rotate freely around the metal center after a 2e– reduction of the complex, leaving the molecular structure more flexible than first anticipated. The outcome of the CO2 reduction reaction is thus dependent on the stability of the two major conformers, where the catalytic site points either in the same (endo) or in the opposite direction (exo) of the macrocyclic ligand. In the former case, amine protonation and subsequent proton transfer to the metal center are preferred, favoring the formation of a metal hydride, whereas CO2 attacks directly from the unhindered side in the latter case to generate a manganese hydroxyl–carbonyl complex, the key intermediate for CO production. Hence, the reactivity cannot be rationalized based on the experimentally derived crystal structure due to the fluxional behavior of the carbonyl groups, which to the best of our knowledge has not been reported previously for CO2-reducing Mn tricarbonyl complexes. A second factor that affects the mechanism and therefore also the product distribution is the position of the amine functionalities on the macrocyclic ligand. In this respect, ab initio molecular dynamics simulations showed that the distance between Mn and the protonated amines is shortest for the middle amine of complex 1a, lowering the energy barrier for proton transfer to the metal center compared to the case where the side-amine was protonated. Thus, hydride formation can be enhanced by tuning the distance between the metal center and the proton shuttles. Finally, the accessibility of CO2 as well as protons to the active site should always be kept in consideration, as it is crucial for switching reaction pathways from one to another.

These computational findings align well with the experimental results, where H2 was the dominant product for all three catalysts (1a, 1b, and 1c), while only small amounts of CO and HCOO– were produced due to disfavored CO2 addition reactions before and after hydride formation, respectively. Formation of HCOO– was further suppressed with complex 1c, where the linker was shortened by one atom, thus limiting CO2 insertion and favoring a second protonation after formation of the hydride. The observed product distribution is furthermore in contrast with our previous work, where complex 1a* with a similar but “open” amine ligand produced HCOO– as the dominant product, substantiating the importance of considering the geometric structure when designing new molecular catalysts.

Acknowledgments

This research is financially supported by the Novo Nordisk Foundation CO2 Research Center (grant no. NNF21SA0072700), the Danish National Research Foundation (grant no. DNRF118; grant no. DNRF-93), the European Union’s Horizon 2020 research and innovation program under the Marie Sklodowska-Curie grant agreement no. 859910, the European Union’s Horizon 2020 research and innovation program under grant agreement no. 862179, and NordForsk (no. 85378). This publication reflects the views only of the authors, and the Commission cannot be held responsible for any use, which may be made of information contained therein. W.H. (CSC no. 202006370038) thanks the China Scholarship Council for financial support. M.L., D.B., A.C.C., and A.N. acknowledge support from the Research Council of Norway through the Centre of Excellence (no. 262695) and D.B., A.C.C., and A.N. for its FRINATEK program (nos. 325003, 325231, and 314321, respectively). M.L. and A.C.C. thank the Norwegian Metacenter for Computational Science (NOTUR) for computational resources (no. nn4654k).

Glossary

Abbreviations

AIMD

ab initio molecular dynamics

CPE

controlled potential electrolysis

DFT

density functional theory

HER

hydrogen evolution reaction

IR-SEC

infrared spectroelectrochemistry

STAB–H

sodium triacetoxyborohydride

TFE

2,2,2-trifluoroethanol

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acscatal.2c05951.

  • Experimental details; X-ray crystallographic data; supporting figures (S1–S19), schemes (S1–S9), and tables (S1–S9); computational details; synthesis procedures and spectroscopic data (PDF)

  • X-ray crystallographic data for 1a–c in CIF format and Cartesian coordinates (XYZ)

Author Present Address

∇ Unisense A/S, Langdyssen 5, 8200 Aarhus N, Denmark (M.R.M.)

Author Present Address

# Department of Chemistry, Technical University of Denmark, 2800 Kgs. Lyngby, Denmark (J.B.J.)

Author Contributions

W.H., M.L., and J.B.J. contributed equally. The manuscript was written with contributions from all authors. All authors have given approval to the final version of the manuscript.

The authors declare no competing financial interest.

Notes

Computational results are available in the ioChem-BD repository50 and can be accessed via https://doi.org/10.19061/iochem-bd-6-192.

Supplementary Material

cs2c05951_si_001.pdf (5MB, pdf)
cs2c05951_si_002.xyz (532.5KB, xyz)

References

  1. Fletcher S.; Tsang E.. https://www.chem.ox.ac.uk/catalysis (accessed May 31, 2022).
  2. Heuson E.; Froidevaux R.; Itabaiana I.; Wojcieszak R.; Capron M.; Dumeignil F. Optimisation of Catalysts Coupling in Multi-Catalytic Hybrid Materials: Perspectives for the Next Revolution in Catalysis. Green Chem. 2021, 23, 1942–1954. 10.1039/D0GC04172F. [DOI] [Google Scholar]
  3. Jin S.; Hao Z.; Zhang K.; Yan Z.; Chen J. Advances and Challenges for the Electrochemical Reduction of CO2 to CO: From Fundamentals to Industrialization. Angew. Chem., Int. Ed. 2021, 60, 20627–20648. 10.1002/anie.202101818. [DOI] [PubMed] [Google Scholar]
  4. Khan M. A.; Zhao H.; Zou W.; Chen Z.; Cao W.; Fang J.; Xu J.; Zhang L.; Zhang J. Recent Progresses in Electrocatalysts for Water Electrolysis. Electrochem. Energy Rev. 2018, 1, 483–530. 10.1007/s41918-018-0014-z. [DOI] [Google Scholar]
  5. Anwar S.; Khan F.; Zhang Y.; Djire A. Recent Development in Electrocatalysts for Hydrogen Production through Water Electrolysis. Int. J. Hydrogen Energy 2021, 46, 32284–32317. 10.1016/j.ijhydene.2021.06.191. [DOI] [Google Scholar]
  6. Liang H. Q.; Beweries T.; Francke R.; Beller M. Molecular Catalysts for the Reductive Homocoupling of CO2 towards C2+ Compounds. Angew. Chem., Int. Ed. 2022, 61, e202200723 10.1002/anie.202200723. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dibenedetto A.; Nocito F. The Future of Carbon Dioxide Chemistry. ChemSusChem 2020, 13, 6219–6228. 10.1002/cssc.202002029. [DOI] [PubMed] [Google Scholar]
  8. Francke R.; Schille B.; Roemelt M. Homogeneously Catalyzed Electroreduction of Carbon Dioxide-Methods, Mechanisms, and Catalysts. Chem. Rev. 2018, 118, 4631–4701. 10.1021/acs.chemrev.7b00459. [DOI] [PubMed] [Google Scholar]
  9. Rønne M. H.; Cho D.; Madsen M. R.; Jakobsen J. B.; Eom S.; Escoudé É.; Hammershøj H. C. D.; Nielsen D. U.; Pedersen S. U.; Baik M.-H.; Skrydstrup T.; Daasbjerg K. Ligand-Controlled Product Selectivity in Electrochemical Carbon Dioxide Reduction Using Manganese Bipyridine Catalysts. J. Am. Chem. Soc. 2020, 142, 4265–4275. 10.1021/jacs.9b11806. [DOI] [PubMed] [Google Scholar]
  10. Riplinger C.; Sampson M. D.; Ritzmann A. M.; Kubiak C. P.; Carter E. A. Mechanistic Contrasts between Manganese and Rhenium Bipyridine Electrocatalysts for the Reduction of Carbon Dioxide. J. Am. Chem. Soc. 2014, 136, 16285–16298. 10.1021/ja508192y. [DOI] [PubMed] [Google Scholar]
  11. Ngo K. T.; McKinnon M.; Mahanti B.; Narayanan R.; Grills D. C.; Ertem M. Z.; Rochford J. Turning on the Protonation-First Pathway for Electrocatalytic CO2 Reduction by Manganese Bipyridyl Tricarbonyl Complexes. J. Am. Chem. Soc. 2017, 139, 2604–2618. 10.1021/jacs.6b08776. [DOI] [PubMed] [Google Scholar]
  12. Loewen N. D.; Berben L. A. Secondary Coordination Sphere Design to Modify Transport of Protons and CO2. Inorg. Chem. 2019, 58, 16849–16857. 10.1021/acs.inorgchem.9b03102. [DOI] [PubMed] [Google Scholar]
  13. Nichols A. W.; Machan C. W. Secondary-Sphere Effects in Molecular Electrocatalytic CO2 Reduction. Front. Chem. 2019, 7, 397. 10.3389/fchem.2019.00397. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Franco F.; Cometto C.; Ferrero Vallana F.; Sordello F.; Priola E.; Minero C.; Nervi C.; Gobetto R. A Local Proton Source in a [Mn(bpy-R)(CO)3Br]-type Redox Catalyst Enables CO2 Reduction Even in the Absence of Brønsted Acids. Chem. Commun. 2014, 50, 14670–14673. 10.1039/C4CC05563B. [DOI] [PubMed] [Google Scholar]
  15. Roy S. S.; Talukdar K.; Jurss J. W. Electro- and Photochemical Reduction of CO2 by Molecular Manganese Catalysts: Exploring the Positional Effect of Second-Sphere Hydrogen-Bond Donors. ChemSusChem 2021, 14, 662–670. 10.1002/cssc.202001940. [DOI] [PubMed] [Google Scholar]
  16. Roy S.; Sharma B.; Pécaut J.; Simon P.; Fontecave M.; Tran P. D.; Derat E.; Artero V. Molecular Cobalt Complexes with Pendant Amines for Selective Electrocatalytic Reduction of Carbon Dioxide to Formic Acid. J. Am. Chem. Soc. 2017, 139, 3685–3696. 10.1021/jacs.6b11474. [DOI] [PubMed] [Google Scholar]
  17. Fernández S.; Cañellas S.; Franco F.; Luis J. M.; Pericàs M. À.; Lloret-Fillol J. The Dual Effect of Coordinating −NH Groups and Light in the Electrochemical CO2 Reduction with Pyridylamino Co Complexes. ChemElectroChem 2021, 8, 4456–4465. 10.1002/celc.202100859. [DOI] [Google Scholar]
  18. Chapovetsky A.; Do T. H.; Haiges R.; Takase M. K.; Marinescu S. C. Proton-Assisted Reduction of CO2 by Cobalt Aminopyridine Macrocycles. J. Am. Chem. Soc. 2016, 138, 5765–5768. 10.1021/jacs.6b01980. [DOI] [PubMed] [Google Scholar]
  19. Jakobsen J. B.; Rønne M. H.; Daasbjerg K.; Skrydstrup T. Are Amines the Holy Grail for Facilitating CO2 Reduction?. Angew. Chem., Int. Ed. 2021, 60, 9174–9179. 10.1002/anie.202014255. [DOI] [PubMed] [Google Scholar]
  20. Madsen M. R.; Jakobsen J. B.; Rønne M. H.; Liang H.; Hammershøj H. C. D.; Nørby P.; Pedersen S. U.; Skrydstrup T.; Daasbjerg K. Evaluation of the Electrocatalytic Reduction of Carbon Dioxide using Rhenium and Ruthenium Bipyridine Catalysts Bearing Pendant Amines in the Secondary Coordination Sphere. Organometallics 2020, 39, 1480–1490. 10.1021/acs.organomet.9b00815. [DOI] [Google Scholar]
  21. Talukdar K.; Sinha Roy S.; Amatya E.; Sleeper E. A.; Le Magueres P.; Jurss J. W. Enhanced Electrochemical CO2 Reduction by a Series of Molecular Rhenium Catalysts Decorated with Second-sphere Hydrogen-bond Donors. Inorg. Chem. 2020, 59, 6087–6099. 10.1021/acs.inorgchem.0c00154. [DOI] [PubMed] [Google Scholar]
  22. Ceballos B. M.; Yang J. Y. Directing the Reactivity of Metal Hydrides for Selective CO2 reduction. Proc. Natl. Acad. Sci. 2018, 115, 12686–12691. 10.1073/pnas.1811396115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Waldie K. M.; Ostericher A. L.; Reineke M. H.; Sasayama A. F.; Kubiak C. P. Hydricity of Transition-Metal Hydrides: Thermodynamic Considerations for CO2 Reduction. ACS Catal. 2018, 8, 1313–1324. 10.1021/acscatal.7b03396. [DOI] [Google Scholar]
  24. Bourrez M.; Molton F.; Chardon-Noblat S.; Deronzier A. [Mn(bipyridyl)(CO)3Br]: an Abundant Metal Carbonyl Complex as Efficient Electrocatalyst for CO2 Reduction. Angew. Chem., Int. Ed. 2011, 50, 9903–9906. 10.1002/anie.201103616. [DOI] [PubMed] [Google Scholar]
  25. Sampson M. D.; Nguyen A. D.; Grice K. A.; Moore C. E.; Rheingold A. L.; Kubiak C. P. Manganese Catalysts with Bulky Bipyridine Ligands for the Electrocatalytic Reduction of Carbon Dioxide: Eliminating Dimerization and Altering Catalysis. J. Am. Chem. Soc. 2014, 136, 5460–5471. 10.1021/ja501252f. [DOI] [PubMed] [Google Scholar]
  26. Kühne T. D.; Iannuzzi M.; Ben M. D.; Rybkin V. V.; Seewald P.; Stein F.; Laino T.; Khaliullin R. Z.; Schütt O.; Schiffmann F. CP2K: An Electronic Structure and Molecular Dynamics Software Package - Quickstep: Efficient and Accurate Electronic Structure Calculations. J. Chem. Phys. 2020, 152, 194103 10.1063/5.0007045. [DOI] [PubMed] [Google Scholar]
  27. Hutter J.; Iannuzzi M.; Schiffmann F.; VandeVondele J. cp2k: Atomistic Simulations of Condensed Matter Systems. WIREs Comput. Mol. Sci. 2014, 4, 15–25. 10.1002/wcms.1159. [DOI] [Google Scholar]
  28. VandeVondele J.; Krack M.; Mohamed F.; Parrinello M.; Chassaing T.; Hutter J. Quickstep: Fast and Accurate Density Functional Calculations Using a Mixed Gaussian and Plane Waves Approach. Comput. Phys. Commun. 2005, 167, 103–128. 10.1016/j.cpc.2004.12.014. [DOI] [Google Scholar]
  29. Sampson M. D.; Kubiak C. P. Manganese Electrocatalysts with Bulky Bipyridine Ligands: Utilizing Lewis Acids To Promote Carbon Dioxide Reduction at Low Overpotentials. J. Am. Chem. Soc. 2016, 138, 1386–1393. 10.1021/jacs.5b12215. [DOI] [PubMed] [Google Scholar]
  30. Mondal B.; Ye S. Hidden Ligand Noninnocence: A Combined Spectroscopic and Computational Perspective. Coord. Chem. Rev. 2020, 405, 213115 10.1016/j.ccr.2019.213115. [DOI] [Google Scholar]
  31. Zhang Y.-Q.; Chen J.-Y.; Siegbahn P. E. M.; Liao R.-Z. Harnessing Noninnocent Porphyrin Ligand to Circumvent Fe-Hydride Formation in the Selective Fe-Catalyzed CO2 Reduction in Aqueous Solution. ACS Catal. 2020, 10, 6332–6345. 10.1021/acscatal.0c00559. [DOI] [Google Scholar]
  32. Wang X.-Z.; Meng S.-L.; Chen J.-Y.; Wang H.-X.; Wang Y.; Zhou S.; Li X.-B.; Liao R.-Z.; Tung C.-H.; Wu L.-Z. Mechanistic Insights Into Iron(II) Bis(pyridyl)amine-Bipyridine Skeleton for Selective CO2 Photoreduction. Angew. Chem., Int. Ed. 2021, 60, 26072–26079. 10.1002/anie.202107386. [DOI] [PubMed] [Google Scholar]
  33. Smieja J. M.; Sampson M. D.; Grice K. A.; Benson E. E.; Froehlich J. D.; Kubiak C. P. Manganese as a Substitute for Rhenium in CO2 Reduction Catalysts: the Importance of Acids. Inorg. Chem. 2013, 52, 2484–2491. 10.1021/ic302391u. [DOI] [PubMed] [Google Scholar]
  34. Machan C. W.; Sampson M. D.; Chabolla S. A.; Dang T.; Kubiak C. P. Developing a Mechanistic Understanding of Molecular Electrocatalysts for CO2 Reduction using Infrared Spectroelectrochemistry. Organometallics 2014, 33, 4550–4559. 10.1021/om500044a. [DOI] [Google Scholar]
  35. Franco F.; Cometto C.; Nencini L.; Barolo C.; Sordello F.; Minero C.; Fiedler J.; Robert M.; Gobetto R.; Nervi C. Local Proton Source in Electrocatalytic CO2 Reduction with [Mn(bpy-R)(CO)3Br] Complexes. Chem. – Eur. J. 2017, 23, 4782–4793. 10.1002/chem.201605546. [DOI] [PubMed] [Google Scholar]
  36. Rønne M. H.; Madsen M. R.; Skrydstrup T.; Pedersen S. U.; Daasbjerg K. Mechanistic Elucidation of Dimer Formation and Strategies for Its Suppression in Electrochemical Reduction of Fac–Mn(bpy)(CO)3Br. ChemElectroChem 2021, 8, 2108–2114. 10.1002/celc.202100279. [DOI] [Google Scholar]
  37. Madsen M. R.; Rønne M. H.; Heuschen M.; Golo D.; Ahlquist M. S. G.; Skrydstrup T.; Pedersen S. U.; Daasbjerg K. Promoting Selective Generation of Formic Acid from CO2 Using Mn(bpy)(CO)3Br as Electrocatalyst and Triethylamine/Isopropanol as Additives. J. Am. Chem. Soc. 2021, 143, 20491–20500. 10.1021/jacs.1c10805. [DOI] [PubMed] [Google Scholar]
  38. Jitaru M. Electrochemical Carbon Dioxide Reduction - Fundamental and Applied Topics. J. Univ. Chem. Technol. Metall. 2007, 42, 333–344. [Google Scholar]
  39. Wu Q.; Li M.; He S.; Xiong Y.; Zhang P.; Huang H.; Chen L.; Huang F.; Li F. The Hangman Effect Boosts Hydrogen Production by a Manganese Terpyridine Complex. Chem. Commun. 2022, 58, 5128–5131. 10.1039/D2CC00757F. [DOI] [PubMed] [Google Scholar]
  40. Clark M. L.; Cheung P. L.; Lessio M.; Carter E. A.; Kubiak C. P. Kinetic and Mechanistic Effects of Bipyridine (bpy) Substituent, Labile Ligand, and Brønsted Acid on Electrocatalytic CO2 Reduction by Re(bpy) Complexes. ACS Catal. 2018, 8, 2021–2029. 10.1021/acscatal.7b03971. [DOI] [Google Scholar]
  41. Saha P.; Amanullah S.; Dey A. Selectivity in Electrochemical CO2 Reduction. Acc. Chem. Res. 2022, 55, 134–144. 10.1021/acs.accounts.1c00678. [DOI] [PubMed] [Google Scholar]
  42. Kinzel N. W.; Werle C.; Leitner W. Transition Metal Complexes as Catalysts for the Electroconversion of CO2 : An Organometallic Perspective. Angew. Chem., Int. Ed. 2021, 60, 2–62. 10.1002/anie.202006988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Heimann J. E.; Bernskoetter W. H.; Hazari N. Understanding the Individual and Combined Effects of Solvent and Lewis Acid on CO2 Insertion into a Metal Hydride. J. Am. Chem. Soc. 2019, 141, 10520–10529. 10.1021/jacs.9b05192. [DOI] [PubMed] [Google Scholar]
  44. Bhattacharya M.; Sebghati S.; VanderLinden R. T.; Saouma C. T. Toward Combined Carbon Capture and Recycling: Addition of an Amine Alters Product Selectivity from CO to Formic Acid in Manganese Catalyzed Reduction of CO2. J. Am. Chem. Soc. 2020, 142, 17589–17597. 10.1021/jacs.0c07763. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Wiedner E. S.; Chambers M. B.; Pitman C. L.; Bullock R. M.; Miller A. J.; Appel A. M. Thermodynamic Hydricity of Transition Metal Hydrides. Chem. Rev. 2016, 116, 8655–8692. 10.1021/acs.chemrev.6b00168. [DOI] [PubMed] [Google Scholar]
  46. Yang J. Y.; Kerr T. A.; Wang X. S.; Barlow J. M. Reducing CO2 to HCO2– at Mild Potentials: Lessons from Formate Dehydrogenase. J. Am. Chem. Soc. 2020, 142, 19438–19445. 10.1021/jacs.0c07965. [DOI] [PubMed] [Google Scholar]
  47. Chen L.; Guo Z.; Wei X. G.; Gallenkamp C.; Bonin J.; Anxolabehere-Mallart E.; Lau K. C.; Lau T. C.; Robert M. Molecular Catalysis of the Electrochemical and Photochemical Reduction of CO2 with Earth-Abundant Metal Complexes. Selective Production of CO vs HCOOH by Switching of the Metal Center. J. Am. Chem. Soc. 2015, 137, 10918–10921. 10.1021/jacs.5b06535. [DOI] [PubMed] [Google Scholar]
  48. Zhang Y.-Q.; Wang Z.-H.; Li M.; Liao R.-Z. Understanding the Chemoselectivity Switch in CO2 Reduction Catalyzed by Co and Fe Complexes Bearing a Pentadentate N5 Ligand. J. Catal. 2022, 414, 277–293. 10.1016/j.jcat.2022.09.002. [DOI] [Google Scholar]
  49. Sinopoli A.; La Porte N. T.; Martinez J. F.; Wasielewski M. R.; Sohail M. Manganese Carbonyl Complexes for CO2 Reduction. Coord. Chem. Rev. 2018, 365, 60–74. 10.1016/j.ccr.2018.03.011. [DOI] [Google Scholar]
  50. Álvarez-Moreno M.; de Graaf C.; López N.; Maseras F.; Poblet J. M.; Bo C. Managing the Computational Chemistry Big Data Problem: The ioChem-BD Platform. J. Chem. Inf. Model. 2015, 55, 95–103. 10.1021/ci500593j. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

cs2c05951_si_001.pdf (5MB, pdf)
cs2c05951_si_002.xyz (532.5KB, xyz)

Articles from ACS Catalysis are provided here courtesy of American Chemical Society

RESOURCES