Kinetic isotope effect offers selectivity in CO₂ reduction

Abstract

A binuclear Ni complex with N,O donors catalyzes CO₂ reduction via its Ni(I) state. The product distribution when H₂O is used as a proton source shows similar yields for CO, HCOOH and H₂. However, when D₂O is used, the product distribution shows a ∼65% selectivity for HCOOH. In situ FTIR indicates that the reaction involves a Ni–COO* and a Ni–CO intermediate. Differences in H/D KIEs on different protonation pathways determine the selectivity of CO₂ reduction.

---

The reduction of CO₂ to chemicals having commercial value has taken centre stage in multi-electron multi-proton catalysis. Apart from the obvious appeal of the chemical conversion of a greenhouse gas to value added chemicals, the inherent challenges in the reaction have attracted the attention of several research groups across the world. Substantial progress has been made in the last few years and several molecular and material catalysts have been developed for this purpose and several of these show considerable reactivity.^1–14

Reduction of CO₂ selectively to any of the several possible products poses a challenge. The 2e⁻/2H⁺ reduction of CO₂ can produce either HCOOH or CO and competes with the reduction of protons to form H₂.^15–20 Several groups have contributed to the current understanding of the mechanism of the reactions involved.^8,21–26 The reduced metal centre binds either CO₂ or H⁺ to produce M–CO₂ or M–H species (Scheme 1). The M–H species can be further protonated to generate H₂ or it can transfer the H⁻ to CO₂ to generate formate. The M–CO₂ species on the other hand generates a M–COOH species upon protonation and then can undergo either a M–C bond cleavage to generate formate or a C–O bond cleavage to generate CO. These species have been observed chemically as well as by using spectro-electrochemistry and characterized.^6,20,27

Scheme 1 Selectivity for 2e⁻/2H⁺ reduction of CO₂via the M–COOH pathway.

Selectivity in 2e⁻/2H⁺ reduction has been demonstrated by several groups and it has been proposed that the electronic structure of the M–COOH species plays a vital role in the process.²⁶ When the M–C bond is very covalent, the protonation of the OH-group is favored resulting in C–O bond cleavage and release of CO, while in cases where the M–C bond is not very covalent, the M–C bond cleaves generating HCOOH. The covalency of the M–C bond can be tuned by the choice of the metal as well as the spin states. Using a ligand that has a protonation site has provided relief from the competing HER.^20,28,29

In addition to the design of the ligand and the choice of metal controlling the electronic structure of the catalyst, selectivity can also be accessed by kinetic control. For example, at lower concentrations of protons (as well as by using weaker proton sources), the HER is naturally suppressed. In a curious case, the product distribution of 2e⁻/2H⁺ CO₂ reduction by Pd/C was demonstrated to depend on the isotope of the proton source used. In situ X-ray characterization verifies the formation of Pd–H species. Substituting H₂O with D₂O reduces the rate of the competing hydrogen evolution reaction (HER) due to lower D⁺ concentration and enhances the carbon dioxide reduction reaction (CO₂RR), leading to a shift from 48% CO in H₂O to 76% CO in D₂O.³⁰ Clearly, the kinetic isotope effect (KIE) involved in the reduction of CO₂ was different from the KIE involved in the reduction of H⁺ leading to different selectivities in H₂O and D₂O. Similar changes in product distribution were observed during CO₂ reduction by cobalt phthalocyanine immobilized on a coordinating polymer support.³¹ A significant kinetic isotope effect (KIE) was reported for the decomposition of HCOOH to form CO₂ and H₂ by an Ir/Ru complex.³² Another study of isotope-controlled selectivity by quantum tunneling results in different chemical reactivity of cyclopropylmethylcarbene.³³ While isotope effects have been reported in the kinetics of electrochemical CO₂ reduction using molecular catalysts, changes in selectivity have not been reported and whether such changes in selectivity can be obtained in a molecular catalyst under homogeneous conditions are yet to be evaluated.

In this manuscript, we report the synthesis of a novel binuclear Ni catalyst, [(Py2ald)₂Ni₂](BPh₄)₂ (1(BPh₄)₂), and explores its application in CO₂ reduction by using water as the proton source. The ligand, Hpy2ald (3-{[Bis(2-pyridinylmethyl)amino]methyl}-2-hydroxybenzaldehyde), was synthesized following literature report.³⁴ The catalyst shows no selectivity for either CO or HCOOH and exhibits a competing HER in H₂O. However, in D₂O the catalyst exhibits 65% selectivity for HCOOH. In situ spectro-electrochemical FTIR provides insight into the mechanism of CO₂ reduction.

Cyclic voltammetry (CV) measurements conducted with 0.5 mM 1(BPh₄)₂ in N₂ saturated acetonitrile solution unveiled two closely positioned peaks at −1.82 V and −2.05 V vs. Fc⁺/Fc⁰, respectively (Fig. 1A and inset). To identify the associated reduction process, CV was conducted using the free ligand HPy2ald under identical conditions. It was observed that the peak at −2.05 V vs. Fc⁺/Fc⁰ is linked to ligand reduction, potentially involving the reduction of the aldehyde group. Supporting evidence for this transformation is found in corroborating Fourier-transform infrared spectroscopy coupled with spectro-electrochemistry (FTIR-SEC) data. For 1(BPh₄)₂, when the working electrode is held at −2.05 V vs. Fc⁺/Fc⁰, the peak corresponding to the –CHO group at 1633 cm⁻¹ diminishes (Fig. 2A, inset), signifying its reduction, while a new peak in the 2500–2700 cm⁻¹ range emerges, corresponding to the formation of the –CH₂OH group (Fig. 2A). More importantly, the complete reduction of the –CHO vibration indicates that both –CHO groups are being reduced. The additional peak at −1.82 V vs. Fc⁺/Fc⁰ is associated with the Ni(II/I) reduction. The reduction of Ni^II to Ni^I was confirmed by FTIR-SEC under one atmosphere CO (Fig. S6, ESI†). Ni^II does not bind CO, but when FTIR-SEC at −1.85 V vs. Fc⁺/Fc° was conducted, a clear Ni^I–CO vibration at 1980 cm⁻¹ was observed. Similar carbonyl species were previously reported in a binuclear cobalt complex bridged by pyrazole.¹¹

Fig. 1 Cyclic voltammetry of 0.5 mM 1(BPh₄)₂ (green) and free ligand (purple trace) in N₂ saturated acetonitrile solution, and its catalytic nature in CO₂ saturated acetonitrile in the absence (blue trace) and presence of H₂O (red trace). The inset provides a zoomed-in view of the potential range from −0.4 to −2.4 V in an N₂ atmosphere (A). Comparison of the CV trace of the Ni-complex in CO₂ and N₂ saturated acetonitrile in the presence of 5% (V/V) H₂O shows electrocatalytic CO₂ reduction along with the HER (B). 100 mM tetrabutylammonium perchlorate (TBAP) was used as the supporting electrolyte.

Fig. 2 FTIR-SEC spectra of 1(BPh₄)₂ (4 mM) in N₂ saturated acetonitrile while keeping the potential at −2.05 V vs. Fc⁺/Fc⁰ showing conversion of –CHO to –CH₂OH. The inset shows the zoomed-in view of the 1550–1800 cm⁻¹ region (A); in situ FTIR-SEC spectra (CPE at −2.2 V vs. Fc⁺/Fc⁰) of the intermediates of 1(BPh₄)₂ (4 mM) in acetonitrile using ¹²CO₂ (blue trace) and ¹³CO₂ (red trace) as substrates. The inset shows the zoomed-in difference spectra (¹³CO₂–¹²CO₂) of the 1850–2050 cm⁻¹ region, showing a shift of 44 cm⁻¹ (B). 100 mM tetrabutylammonium perchlorate (TBAP) was used as the supporting electrolyte.

In the presence of saturated CO₂ and 5% (V/V) H₂O, a strong reduction current is observed at −2.2 V vs. Fc⁺/Fc⁰ with a pre-catalytic feature at −1.65 V vs. Fc⁺/Fc⁰ which indicates CO₂ binding to the reduced Ni^I (Fig. 1A). The electrocatalytic current at −2.2 V vs. Fc⁺Fc⁰ is not observed in the absence of CO₂ and represents electrocatalytic CO₂ reduction (Fig. 1B). These observations lead to the conclusion that the species responsible for CO₂ binding is Ni(I) but the electrocatalysis proceeds after further reduction of Ni^III–CO₂⁻/Ni^III–COOH (represented as Ni^III–COO*). In the presence of 5% (V/V) H₂O in an N₂ atmosphere, a catalytic current is observed with an onset at −2.3 V vs. Fc⁺/Fc⁰, indicating the occurrence of the hydrogen evolution reaction (HER). The absorption spectra of 1(BPh₄)₂ before and after CPE show an 18% reduction of the bridging phenolate to Ni charge transfer, not present in a monomer (Fig. S11 and S12, ESI†) after 30 minutes.

To delve into the reaction mechanism and identify the intermediates involved, FTIR-SEC was carried out in a CO₂-saturated acetonitrile solution in an OTTLE cell. While keeping the working electrode at −2.2 V vs. Fc⁺/Fc⁰, the FTIR-SEC results unveiled the presence of two isotope-sensitive vibrations associated with ¹³CO₂ (Fig. 2B). A peak observed at 1633 cm⁻¹, which corresponds to a Ni^I–COO* intermediate, shifted to 1588 cm⁻¹ when a heavier isotope of CO₂ (¹³CO₂) was used. Another peak at 1980 cm⁻¹ was also observed, corresponding to the Ni^I–CO species previously observed, which is a likely precursor to the carbon monoxide gas detected during the gas analysis. This peak shifted to 1936 cm⁻¹ when ¹³CO₂ was used (Fig. 2B, inset). Based on these observations, a plausible mechanism is proposed (Scheme 2). Further evidence for the mechanism is marshalled from kinetic isotope effects discussed later.

Scheme 2 Reaction mechanism of different pathways of CO₂/CO (black arrow) and CO₂/HCOOH (blue arrow) along with the HER (green arrow). The relative KIE (KIE*) is shown for all the reaction pathways.

Controlled potential electrolysis (CPE) was conducted using a glassy carbon electrode with a larger surface area (2 cm²) (Fig. 3A). After electrolysis of 0.5 mM of 1(BPh₄)₂ in CO₂ saturated acetonitrile solution in the presence of 5% (V/V) H₂O, the gaseous products generated were collected from the headspace and were subjected to analysis using gas chromatography (GC) equipped with a thermal conductivity detector (TCD). The GC-TCD analysis revealed the presence of two gases: hydrogen and carbon monoxide (Fig. 3B and Fig. S8, ESI†). The faradaic yields were determined to be 32% for hydrogen and 23% for carbon monoxide. An additional 36% of the product was identified as formate, when the solution, after electrolysis, was extracted with water and subsequently analysed by ion chromatography (Fig. 3C and Fig. S9, ESI†). Reduction of the metal may lead to formation of Ni nanoparticles on the electrode. The rinse test of the working electrode after CPE, however, does not show any catalytic activity (Fig. S10, ESI†). The results from the CPE indicate that 1(BPh₄)₂ exhibits no particular selectivity for either of the three 2e⁻/2H⁺ reduced products, i.e. H₂, CO and HCOOH (Fig. 3D). The mechanistic pathway likely involved in 2e⁻ CO₂ reduction (Scheme 1) by 1(BPh₄)₂ would include key intermediates like Ni–COOH and Ni–CO. The Ni–COOH species would lead to CO after a C–O bond cleavage which upon protonation of the C-atom of the Ni^I–COOH species will release HCOOH. Formation of H₂ entails protonation of the reduced Ni centre (Scheme 1) forming a Ni–H species. The same Ni–H species can produce HCOOH by reducing CO₂ (Scheme 1). The different possible pathways involved may or may not involve protonation in the rate determining step (rds). A rds where protons are involved generally show the H/D KIE. In fact, CO₂ reduction as well as the competing hydrogen evolution has been reported to exhibit H/D KIEs^{30,31,35–37} and, in general, pathways involving metal hydrides exert substantial KIEs. The KIEs can both offer insight into the mechanism involved and result in changes in product distribution, i.e. selectivity.

Fig. 3 CPE performed at −2.2 V vs. Fc⁺/Fc⁰ varying the heavier isotope of the proton source for 1/2 hour using a glassy carbon plate as the working electrode, Pt wire as the counter electrode and Ag/AgCl as the reference electrode. Charge consumption (A) followed by product detection using gas chromatography-TCD (B) and ion chromatography (C) and their faradaic yields (D) are compared for both H₂O and D₂O (¹²CO₂/H₂O – red trace, ¹²CO₂/D₂O – blue trace).

In pursuit of a deeper understanding of the mechanism underlying carbon dioxide reduction, which yields both carbon monoxide and formic acid as reduced products, a kinetic isotope effect (KIE) investigation was undertaken. On using heavy water (D₂O) instead of H₂O, the overall catalytic current was decreased (Fig. 3A). The products detected using GC and IC (ion chromatography) were H₂, CO and HCOOH. However, the selectivity for HCOOH increases from 36% to a staggering 63% when D₂O is used instead of H₂O. The FY of CO and H₂ shrinks from 32% and 23% to 12% and 16%, respectively. The charges consumed by H₂, CO and HCOOH in H₂O were 3.26C, 2.35C and 3.67C, whereas in D₂O they were 1.33C, 1.00C and 5.23C, respectively (Table 1). Thus, in D₂O the production of H₂ and CO is suppressed relative to H₂O, whereas the production of HCOOH is enhanced. These results present a unique case where the selectivity between the three products of CDR, namely H₂, CO and HCOOH, for a molecular catalyst is shifted drastically by the use of isotopes.

Table 1 Relative KIEs calculated for the three processes

Charge consumed	H2O		D2O		Relative KIE (H/D)
	10.2		8.3
Products	Faradaic yield (%)	Charge (C)	Faradaic yield (%)	Charge (C)
H2	32	3.26	16	1.33	3.5
CO	23	2.35	12	1.00	3.3
HCOOH	36	3.67	63	5.23	1

---

The results indicate that H₂ evolution and CO production are inhibited in the presence of D₂O relative to H₂O, while HCOOH production is accelerated. Since these reactions are competitive, the absolute currents being consumed cannot be used to derive a H/D KIE for the individual reactions. However, a relative KIE (KIE* in Scheme 2) (Table 1) can be obtained relative to the HCOOH formation reaction. The reduction of CO₂ to CO exhibits a relative KIE of 3.3, while the H₂ evolution shows a relative KIE of 3.5 relative to any implicit KIE HCOOH may have. Formation of CO requires two protons and it is likely that the second protonation step (protonation of the M–COOH species) is the rds, as suggested by its accumulation during catalysis, resulting in a H/D KIE. The relative KIE observed here is similar to the KIE observed in CO₂ reduction to CO by Co–thiolate complexes where the protonation of the –OH end of a M–COOH species is determined to be the rds.¹⁷ Similarly, the formation of H₂ involves two protonation steps and the second protonation (i.e. protonation of M–H) is likely the rds that earns the H/D KIE. A similar KIE was recorded for H₂ evolution by Ni complexes where the protonation of the Ni–H species was the rds.³⁸

The formic acid produced can be formed either via Ni–H attack on CO₂ or via C-protonation of the Ni–COOH intermediate. The H/D KIE reported for the hydride pathway is substantial (5–10)³⁷ because of the shift of the hydride from the metal to the CO₂ in the TS. The fact that the HCOOH increase in D₂O contradicts this possibility in the case of 1(BPh₄)₂. Rather, the hydrolysis of Ni–COOH via C-protonation seems to be the most logical pathway.

In summary, differential H/D KIEs associated with the two competing 2e⁻/2H⁺ CO₂ reduction pathways and that of the HER allow obtaining ∼65% selectivity in HCOOH formation. This is the first example where selectivity in CO₂ reduction taking advantage of KIEs could be demonstrated in a molecular system.

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

A. D. acknowledges SERB research grant no. CRG/2021/00154 and DST/TMD(EWO)/IC5-2018/03(G). A. M. acknowledges CRG/2022/001931 for funding. S. P., S. A., and S. G. acknowledge the CSIR for senior research fellowship.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. S. Overa, B. H. Ko, Y. Zhao and F. Jiao, Acc. Chem. Res., 2022, 55, 638–648.
2. J. Qiao, Y. Liu, F. Hong and J. Zhang, Chem. Soc. Rev., 2014, 43, 631–675.
3. H.-R. “Molly” Jhong, S. Ma and P. J. A. Kenis, Curr. Opin. Chem. Eng., 2013, 2, 191–199.
4. Q. Chang, Y. Liu, J.-H. Lee, D. Ologunagba, S. Hwang, Z. Xie, S. Kattel, J. H. Lee and J. G. Chen, J. Am. Chem. Soc., 2022, 144, 16131–16138.
5. H. Kang, A. Staples-West, A. Washington, C. Turchiano, A. Cooksy, J. Huang and J. Gu, ChemCatChem, 2023, 15, e202300576.
6. A. J. Morris, G. J. Meyer and E. Fujita, Acc. Chem. Res., 2009, 42, 1983–1994.
7. R. Cao, ChemSusChem, 2022, 15, e202201788.
8. S. Amanullah, P. Saha, A. Nayek, M. E. Ahmed and A. Dey, Chem. Soc. Rev., 2021, 50, 3755–3823.
9. S. Fang, M. Rahaman, J. Bharti, E. Reisner, M. Robert, G. A. Ozin and Y. H. Hu, Nat. Rev. Methods Prim., 2023, 3, 61.
10. M. Bonchio, J. Bonin, O. Ishitani, T.-B. Lu, T. Morikawa, A. J. Morris, E. Reisner, D. Sarkar, F. M. Toma and M. Robert, Nat. Catal., 2023, 6, 657–665.
11. A. Bohn, J. J. Moreno, P. Thuéry, M. Robert and O. Rivada-Wheelaghan, Chem. – Eur. J., 2023, 29, e202202361.
12. S. Amanullah, P. Gotico, M. Sircoglou, W. Leibl, M. J. Llansola-Portoles, T. Tibiletti, A. Quaranta, Z. Halime and A. Aukauloo, Angew. Chem., Int. Ed., 2023, e202314439.
13. P. Gotico, Z. Halime, W. Leibl and A. Aukauloo, ChemPlusChem, 2023, 88, e202300222.
14. C. Zhang, P. Gotico, R. Guillot, D. Dragoe, W. Leibl, Z. Halime and A. Aukauloo, Angew. Chem., Int. Ed., 2023, 62, e202214665.
15. L. Chen, Z. Guo, X.-G. Wei, C. Gallenkamp, J. Bonin, E. Anxolabéhère-Mallart, K.-C. Lau, T.-C. Lau and M. Robert, J. Am. Chem. Soc., 2015, 137, 10918–10921.
16. J. D. Froehlich and C. P. Kubiak, Inorg. Chem., 2012, 51, 3932–3934.
17. S. Dey, M. E. Ahmed and A. Dey, Inorg. Chem., 2018, 57, 5939–5947.
18. M. D. Sampson and C. P. Kubiak, J. Am. Chem. Soc., 2016, 138, 1386–1393.
19. C. G. Margarit, N. G. Asimow, C. Costentin and D. G. Nocera, ACS Energy Lett., 2020, 5, 72–78.
20. S. Amanullah, P. Saha and A. Dey, J. Am. Chem. Soc., 2021, 143, 13579–13592.
21. J. A. Barrett, C. J. Miller and C. P. Kubiak, Trends Chem., 2021, 3, 176–187.
22. C. Costentin, M. Robert and J.-M. Savéant, Chem. Soc. Rev., 2013, 42, 2423–2436.
23. Y. Yamazaki, H. Takeda and O. Ishitani, J. Photochem. Photobiol., C, 2015, 25, 106–137.
24. E. Fujita, Coord. Chem. Rev., 1999, 185–186, 373–384.
25. E. Fujita, C. Creutz, N. Sutin and B. S. Brunschwig, Inorg. Chem., 1993, 32, 2657–2662.
26. P. Saha, S. Amanullah and A. Dey, Acc. Chem. Res., 2022, 55, 134–144.
27. B. Mondal, A. Rana, P. Sen and A. Dey, J. Am. Chem. Soc., 2015, 137, 11214–11217.
28. A. Chapovetsky, M. Welborn, J. M. Luna, R. Haiges, T. F. I. I. I. Miller and S. C. Marinescu, ACS Cent. Sci., 2018, 4, 397–404.
29. M. E. Ahmed, A. Rana, R. Saha, S. Dey and A. Dey, Inorg. Chem., 2020, 59, 5292–5302.
30. J. H. Lee, B. M. Tackett, Z. Xie, S. Hwang and J. G. Chen, Chem. Commun., 2020, 56, 106–108.
31. Y. Liu and C. C. L. McCrory, Nat. Commun., 2019, 10, 1683.
32. S. Fukuzumi, T. Kobayashi and T. Suenobu, J. Am. Chem. Soc., 2010, 132, 1496–1497.
33. A. Nandi, D. Gerbig, P. R. Schreiner, W. T. Borden and S. Kozuch, J. Am. Chem. Soc., 2017, 139, 9097–9099.
34. I. A. Koval, D. Pursche, A. F. Stassen, P. Gamez, B. Krebs and J. Reedijk, Eur. J. Inorg. Chem., 2003, 1669–1674.
35. N. Devi, C. K. Williams, A. Chaturvedi and J. “Jimmy” Jiang, ACS Appl. Energy Mater., 2021, 4, 3604–3611.
36. S. Dey, T. K. Todorova, M. Fontecave and V. Mougel, Angew. Chem., Int. Ed., 2020, 59, 15726–15733.
37. S. Roy, B. Sharma, J. Pécaut, P. Simon, M. Fontecave, P. D. Tran, E. Derat and V. Artero, J. Am. Chem. Soc., 2017, 139, 3685–3696.
38. D. Hong, Y. Tsukakoshi, H. Kotani, T. Ishizuka, K. Ohkubo, Y. Shiota, K. Yoshizawa, S. Fukuzumi and T. Kojima, Inorg. Chem., 2018, 57, 7180–7190.

---

Footnotes

† Electronic supplementary information (ESI) available: experimental details, control electrochemical experiments and structural data. CCDC 2208720 and 2310122. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d3cc06336d

‡ These authors contributed equally.

---