Zwitterionic alcoholic solutions for integrated CO₂ capture and hydrogenation

Abstract

An integrated CO₂ capture and conversion (ICCU) process using zwitterionic bases (ZBs) in alcohol solvents was developed. Inner salts composed of a 1,3-dimethylimidazolium cation covalently attached to a phenolate moiety react with CO₂ to form alkyl carbonates (AKCs) in alcohols. The ZBs yield up to 70% AKCs and retain chemisorbed CO₂, whereas tertiary amines with similar conjugate acid pK_a values capture only small amounts of CO₂ and quickly release it in the alcohols tested. The AKCs were hydrogenated to HCOO⁻ or methanol, depending on the reaction temperature and the Ru catalyst selected. Using isopropanol as the solvent, the ICCU process combining 1,3-dimethyl-2-(4-oxyphenyl)imidazolium (ZB-p) and cis-[Ru(dppm)₂Cl₂] quantitatively converted the respective AKC to formate at 50 °C and 10 bar H₂ after 35 minutes of reaction. This represents one of the most active ICCU systems reported to date, with a turnover frequency of 660 h⁻¹ under these mild conditions. Methanol was obtained with a 66% yield after 20 hours by hydrogenation of the AKC of ZB-p in ethylene glycol using the catalyst Ru-MACHO-BH at 140 °C and 70 bar H₂.

---

Green foundation

1. Using CO₂ as a feedstock for chemical production is a green chemistry challenge aimed at mitigating global warming. Most CCU strategies require pure CO₂, which involves energy-intensive purification. The integrated CO₂ capture and conversion (ICCU) approach employs zwitterionic phenolate bases (ZBs) dissolved in alcohols to form alkyl carbonates (AKCs), which are then hydrogenated by Ru catalysts into formate or methanol, thereby saving energy.

2. Compared to traditional amines, ZBs offer significant advantages. They retain CO₂ as AKCs in various alcohols and are non-volatile. The AKCs are efficiently hydrogenated by Ru catalysts into formate or methanol, providing more energy-efficient pathways than direct CO₂ hydrogenation methods.

3. Future improvements could involve optimizing reaction conditions to minimize CO₂ desorption, developing strategies for catalyst reuse, and elucidating the reaction mechanism. These efforts will reduce energy demands, improve process economics, and expand green CO₂ transformation methods.

Introduction

One of the most pressing challenges today is mitigating climate change caused by the rising levels of greenhouse gases in the atmosphere. CO₂ can participate in various types of chemical reactions to produce commercially valuable commercial compounds.^1,2 This strategy is known as carbon capture and utilization (CCU). There are essentially two types of processes: incorporation and reductive.³ In the incorporation process, the carbon oxidation state is not altered, and the most important examples include the production of urea used as fertilizers, the synthesis of salicylic acids used in various sectors of the pharmaceutical industry, the synthesis of inorganic carbonates that are incorporated into the cement industry or used as fillers, and the production of cyclic carbonates and polycarbonates. However, industrial CCU processes⁴ are still in their early stages, as they do not yet use the large amount of CO₂ that would be necessary to prevent a 2 °C increase in the average pre-industrial revolution temperature. Therefore, there is a need to develop reductive pathways that produce commercially valuable chemical compounds.

In the reductive process, carbon can be partially reduced to intermediate oxidation states or fully reduced to −4, as seen in methane production. Various catalytic systems that reduce CO₂ have been reported.^5–13 However, most CCU processes require CO₂ gas with a high degree of purity.¹⁴ There are few sources of carbon dioxide with sufficient purity to be used directly, and most of the time it needs to be separated from other gases, as in the case of exhaust gases from industries and thermal power plants, or even from the atmosphere. Current purification methods are based on the selective removal of CO₂ from gas mixtures using liquid or solid sorbents.¹⁵ Liquid-based, solid-supported, or membrane-based technologies are the most widely used. The sorption process is exothermic, but the desorption of CO₂ and gas compression can account for up to 90% of the energy consumption during purification.¹⁶

This energy demand can be avoided through integrated carbon capture and conversion (ICCU, Scheme 1). ICCU involves using the species (Sorbent·CO₂) formed by the reaction or dissolution of CO₂ in the sorbent as reagents to produce chemical compounds (Scheme 1a).^17,18 In an ideal system, the sorbent is regenerated at the end of the process while the products from the CO₂ reduction are purified.

Scheme 1 Strategies for the reductive ICCU to produce C₁ chemicals. (a) General scheme of reductive ICCU, (b) ICCU to formate involving the intermediate AKC. (c) ICCU to methanol involving the intermediate AKC.

Several sorbents·CO₂ adducts can be reduced to C₁ organic compounds. Examples of sorbents that react with CO₂ include:¹⁷ (i) epoxides, which form cyclic carbonates; (ii) 2-aminoalcohols, which form oxazolidinones; (iii) primary or secondary amines, which form carbamic acid derivatives or (bi)carbonates; (iv) tertiary amines, which form (bi)carbonates; (v) inorganic bases, which form (bi)carbonates. Cyclic carbonates¹⁹ and oxazolidinones²⁰ are reduced to diols or aminoethanols, respectively, with methanol as the final CO₂ hydrogenation product. Carbamic acid derivatives and (bi)carbonates can also be reduced to the respective formate salts, formamides, formate esters or further reduced to methanol.²¹

This manuscript focuses on the use of bases (B): that capture CO₂ exclusively in the form of alkoxy carbonates (AKC) in alcoholic solvents. The resulting adduct is hydrogenated to formate (Scheme 1b) or further hydrogenated to methanol (Scheme 1c), depending on the catalyst and reaction conditions. He and co-workers first reported the formation of AKC and its use as an intermediate for in situ CO₂ hydrogenation.²² A mixture of carbamates and AKC was obtained using polyethyleneimine as an absorber, which produced formate as the main product when RhCl₃ with CyPh₂P ligands served as the catalyst precursor. In 2014, the Heldebrandt and He groups reported the hydrogenation of AKC using amidinium or guanidinium bases dissolved in methanol.^23,24 In these studies, Rh and Ru catalysts with phosphine ligands were used to hydrogenate methyl carbonates, resulting in formate or methyl formate. Prakash's group further explored this strategy by testing various bases (amines or hydroxides) and alcohols (primarily ethylene glycol) for the formation of methanoate and methanol, employing both homogeneous^6,25,26 and heterogeneous catalysts.^27–29

In this work, we present an ICCU scheme for producing formate and methanol through the hydrogenation of CO₂ captured by zwitterionic bases (ZBs), ZB-p and ZB-m (Fig. 1). These compounds selectively form AKC by reacting with CO₂ in alcoholic solutions, which are then hydrogenated by Ru molecular catalysts using H₂ (Scheme 2). Unlike amines such as MDEA, TEA, and TMEDA, which have similar conjugate acid pK_a (pK_aH) values (Fig. 1), ZBs can retain captured CO₂ in alcohol solution. Moreover, AKC hydrogenation in isopropanol using cis-[Ru(dppm)₂Cl₂] catalysts [where dppm = bis(diphenylphosphino)methane] proceeded faster with ZBs than with the stronger base DBU.

Fig. 1 Bases tested for the ICCU process described in this work. pK_aH is the conjugated acid pK_a of the base in aqueous solutions.

Scheme 2 ICCU through catalytic hydrogenation using ZBs.

Results and discussion

Our group previously described aqueous solutions of ZB-p and ZB-m³⁰ as promising CO₂-capturing agents that could replace classical amines in the scrubbing process due to their negligible loss and virtually non-volatile nature. Solutions of these compounds in alcohols can capture CO₂ and form alkyl carbonates suitable for the ICCU process. The system was initially studied using ethylene glycol (EG), which had previously been used with tertiary amines or metal hydroxides for CO₂ capture followed by hydrogenation.^6,25 Solutions of ZB-p and ZB-m in EG (0.5 mmol dissolved per mL of solvent) reacted at room temperature under a constant pressure of 1.6 bar of CO₂, yielding 60–70% monoalkyl carbonate as confirmed by ¹H NMR spectra and gravimetry (Table S3†), and both techniques gave the same results within ±5% uncertainty.

Alkyl carbonates of ZB-p solutions in EG were fully converted after 1 hour of reaction in batch mode at 70 °C, 70 bar of H₂, with 0.8 mol% cis-[Ru(dppm)₂Cl₂] as the catalyst precursor (Table S4, entry 4†). Just the formate product was detected in the liquid phase, with yields going from 81% to 100% when extending the reaction time from 1 to 20 hours (Table S4, entries 2–4†). Only AKC decomposition was observed when attempting to hydrogenate without a catalyst (Table S4, entry 1†). Reactions conducted at 50 °C for 1 h reduced AKC conversion to 52%, and formate yield decreased to 37% (Table 1, entry 5). Extending the reaction time to 2 hours at 50 °C slightly improved alkyl carbonate conversion to 62% and increased the HCOO⁻ yield to 45% (Table S4, entry 6†).

Table 1 Effect of catalyst precursor and H₂ pressure on the ICCU process using ZB-p as capturing agent^a

Entry	Catalyst precursor^b	H2 pressure (bar)	AKC conversion (%)	Formate yield (%)
a Reaction conditions: 4.0 mL of a fresh ZB-p solution (0.5 mmol dissolved in 1 mL of EG) containing captured CO2, 0.010 mmol catalyst precursor, stirred at 600 rpm for 65 min, 50 °C. b The molecular structure of the catalyst and its abbreviation is shown in Fig. S1.†
1	—	0	17 ± 3	0
2	[RuCl2(PPh3)3]	70	38 ± 1	0
3	Ru-SNS	70	31 ± 4	13 ± 4
4	Ru-MACHO-BH	70	29 ± 2	25 ± 2
5	cis-[Ru(dppm)2Cl2]	70	52 ± 5	37 ± 5
6	cis-[Ru(dppm)2Cl2]	10	61 ± 4	49 ± 4
7	cis-[Ru(dppm)2Cl2]	5	45 ± 3	43 ± 3
8	cis-[Ru(dppm)2Cl2]	0	18 ± 3	0
9	Ru-MACHO	70	69 ± 8	58 ± 8
10	Ru-MACHO	5	44 ± 3	14 ± 3
11	Ru-MACHO	0	25 ± 1	0

---

Since substrate conversion remained incomplete after 1 h at 50 °C and 70 bar of H₂, these conditions were deemed appropriate for evaluating the effects of different catalyst precursors and H₂ pressure (Table 1, entries 2–5 and 9). At 50 °C, 17% of AKC decomposes (Table 1, entry 1), and no formate was detected, indicating CO₂ desorption from the solution into the reactor gas phase. The alkyl carbonate conversion and HCOO⁻ yields varied significantly depending on the Ru complex used. In all cases, alkyl carbonate conversion exceeded methanoate yield, and no other side products were observed in the liquid phase by ¹H and ¹³C{¹H} NMR. [RuCl₂(PPh₃)₃] enhanced CO₂ desorption, showing a 38% conversion (Table 1, entry 2). However, unlike the other complexes tested, [RuCl₂(PPh₃)₃] was unable to produce HCOO⁻ or methanol under the tested conditions. cis-[Ru(dppm)₂Cl₂] and Ru-MACHO were the most effective catalyst for converting AKC (Table 1, entries 5 and 9) and demonstrated higher formate yields compared to the other Ru complexes tested.

Therefore, the ICCU process using these two catalysts was further explored. The hydrogen transfer mechanism was ruled out, as reactions performed without H₂ produced no reduction products after 1 h at 50 °C (Table 1, entries 8 and 11). However, as observed for [RuCl₂(PPh₃)₃], Ru-MACHO appears to promote CO₂ desorption, as higher AKC conversion was observed if compared to the catalyst-free reaction (Table 1, entries 1 and 11). Variations in H₂ pressure had different impacts on the ICCU process with these complexes. Both alkyl carbonate conversion and formate yield decreased as pressure dropped to 5 bar when Ru-MACHO was used (Table 1, entries 9 and 10). Conversion decreased from 69% to 44% and yields from 58% to 14%. In contrast, when cis-[Ru(dppm)₂Cl₂] was used, decreasing H₂ pressure from 70 bar to 5 bar did not significantly alter the results within experimental uncertainty (Table 1, entries 5–7), with an average conversion of 43%. Interestingly, an increase in yield was observed at a 10 bar H₂ pressure (Table 1, entry 6), with 49% formate yield at 61% AKC conversion—values similar to those observed for Ru-MACHO at 70 bar.

The ICCU process with cis-[Ru(dppm)₂Cl₂] was further explored, as lower pressures can be used without significant loss in activity, and the complex is easy to prepare due to the commercially availability of the dppm ligand. Reaction time profiles were obtained at 5, 10 and 70 bar of H₂ (Fig. 2 and Fig. S34–36†). Both conversion and yield increase up to 50 minutes of reaction at 50 °C with 0.8 mol% of catalyst. After this period, carbonate consumption slows and reaches a plateau around 65%. Conversion remains higher than HCOO⁻ yield until 125 minutes, after which conversion and yield tend to equalize at 10 and 70 bar. However, the methanoate yield reached a maximum of 45% in the reaction performed at 5 bar and remained constant even when the reaction time was extended to 1205 minutes (Fig. S34†). This last observation is intriguing, as a pressure drop of approximately 2 bar was noted in all experiment, and the reactor being pressurized at the end of each reaction.

Fig. 2 Alkyl carbonate conversion and formate yield time profile for the ICCU process using cis-[Ru(dppm)₂Cl₂]and 10 bar H₂ pressure. Reaction conditions: 4.0 mL of a fresh ZB-p solution (0.5 mmol dissolved in 1 mL of EG) containing captured CO₂, 0.010 mmol cis-[Ru(dppm)₂Cl₂], 50 °C, 600 rpm stirring.

Various alcohols and water, classified as green solvents,³¹ were tested for the ICCU reaction (Table 2). CO₂ sorption could only be quantified by gravimetry using EG, due to its low vapor pressure. All other solvents volatilize during the removal of physisorbed CO₂, complicating gravimetric analysis. Carbonate species were quantified by using NMR techniques. AKC was identified by signals from hydrogen atoms on the carbon adjacent to the carbonate group, and bicarbonate in aqueous solution was quantified by ¹³C NMR.³⁰ In all cases, the relaxation time constant of longitudinal magnetization (T₁) was determined for all H in the molecules (or carbon in aqueous solutions), and NMR experiments were conducted with a delay time (d₁) set to five times the T₁ value of the nucleus with the longest relaxation time. At room temperature and ∼0.5 mol L⁻¹, the AKC signal and the solvent overlapped when NMR samples were prepared using the same procedure as for EG. As result, reactions with different solvents required specific sample preparation procedure, which are detailed in the ESI (Table S1†). Solutions of ZB-p yielded 70% of the respective alkyl carbonate in primary alcohols (Table 2, entries 1–3), or bicarbonate in water (Table 2, entry 4). The isopropanol solution absorbed less CO₂, with 63% of ZB-p being consumed (Table 2, entry 5). CO₂ capture in a ∼0.5 mol L⁻¹ solution of ZB-p in 4-methyl-2-pentanol (MIBC) led to the formation of a white precipitate, preventing the use of this alcohol in the current study.^32–34

Table 2 Effect of the solvent for the ICCU process using ZB-p as capturing agent^a

Entry	Solvent	CO2 capture^b (%)	Alkyl carbonate conversion (%)	Formate yield (%)	TON	TOF (h^−1)
a Reaction conditions: 4.0 mL of a fresh ZB-p solution (0.5 mmol dissolved in 1 mL of solvent) containing captured CO2, 0.010 mmol cis-[Ru(dppm)2Cl2], 35 min, 50 °C, 600 rpm stirring. b CO2 capture means the yield of alkyl carbonate at the capture step. c Determined by ^13C{^1H} NMR.
1	Ethylene glycol	70	25 ± 1	19 ± 1	27	46
2	Ethanol	71	47 ± 4	15 ± 1	21	36
3	n-Butanol	70	77 ± 5	77 ± 6	1.1 × 10^2	1.9 × 10^2
4	Water	70^c	2 ± 1	0	0	0
5	Isopropanol	63	100	91 ± 2	1.1 × 10^2	1.9 × 10^2

---

The solvent screening for the ICCU process was conducted at 50 °C and 10 bar of H₂ for 35 minutes, using cis-[Ru(dppm)₂Cl₂] as the catalyst. Under these conditions, a 25% alkyl carbonate conversion was observed with EG, allowing for a meaningful comparison of turnover frequencies (TOF) across different solvents. Water proved unsuitable for the ICCU process under these conditions, as only 2% of bicarbonate was converted and no HCOO⁻ was detected (Table 2, entry 4). Among all primary alcohols, butanol exhibited the highest alkyl carbonate conversion (77%) and produced equivalent amounts of methanoate. In reaction with isopropanol, AKC was fully converted, yielding 91% formate, (Table 2, entry 5), with a turnover number (TON) of 1.1 × 10² and a TOF of 1.9 × 10² h⁻¹. These TON and TOF values are comparable to those obtained with n-butanol, making either of these solvents suitable for improving the ICCU process. Isopropanol was selected because it dissolves higher amounts of ionic species due to its higher dielectric constant compared to n-butanol.

The CO₂ sorption capacity of ZB-p and ZB-m dissolved in isopropanol is much higher than conventional amines MDEA, TEA, and TMEDA, which have similar pK_aH. Indeed, AKC was only observed for ZB-p and ZB-m and DBU, (Table S3, entries 3, 6 and 10†). In situ NMR measurements using high-pressure cells show that only 3% of the base is transformed in the respective AKC with 1.6 bar of pure CO₂ and the base is fully regenerated when pressure is released (Fig. S29–S32†). This last observation is in agreement with what has been reported previously.³⁵ The solubility of the AKC derived from ZB-p and ZB-m in isopropanol is limited to ∼0.5 mol L⁻¹ and sometimes precipitation occurs when CO₂ is captured by 1.0 mol L⁻¹ solutions of ZB-p or 0.75 mol L⁻¹ of ZB-m. A gentle heating at 50 °C desorbs a little of CO₂, and stable solutions are attained. Moreover, the higher the pK_aH of the chosen base, the higher is AKC yield. A ∼1 mol L⁻¹ solution of bases form a maximum of 53% AKC at equilibrium for ZB-p, followed by 70% for ZB-m and 86% for DBU. These three bases are suitable for the ICCU process, while traditional tertiary amines used previously²⁵ will be unproductive as CO₂ is released after depressurization.

Fig. 3 shows the AKC conversion and HCOO⁻ yield for ICCU systems employing the bases ZB-p and ZB-m and DBU with varying amounts of the catalyst precursor cis-[Ru(dppm)₂Cl₂]. Reactions were performed at fixed AKC concentration of 0.50 mmol of AKC per mL of solvent to enable fair comparison among different bases. Reactions were conducted at 50 °C, 35 min, and 10 bar H₂. Catalyst to substrate percentages were 0.10, 0.25 and 0.50 mol% and were calculated in relation to AKC. Additionally, desorption reactions of all three systems are also shown (hatched bars), in which H₂ and catalyst were excluded. As expected, the base with the most acidic conjugated acid releases more CO₂, and 35% AKC decomposes for ZB-p (pK_aH = 8.5), followed by 11% for ZB-m (pK_aH = 8.9) and only 5% for DBU (pK_aH = 13.5). Hydrogenation at low catalyst concentration showed an interesting behaviour. For all three bases, AKC conversion is higher than HCOO⁻ yield with 0.10% of cis-[Ru(dppm)₂Cl₂], indicating that isopropyl carbonate decomposition to CO₂ is competing with hydrogenation pathway. The presence of Ru complex promotes CO₂ evolution, as the sum of desorption and formate yield does not account completely to the transformation of AKC. Higher amounts of catalyst increase hydrogenation rates, equalizing substrate conversion and formate yield. Conversion and yield are slightly higher for 0.50% runs, though good results are observed with 0.25%. The most active system measured was using ZB-p along with 0.25% of Ru, showing a TON of 3.8 × 10² and TOF of 6.6 × 10² h⁻¹, Table S5, entry 3.†

Fig. 3 AKC conversion and formate yield for the ICCU process using ZB-p (green), ZB-m (blue) and DBU 3 (gray) in isopropanol and varying cis-[Ru(dppm)₂Cl₂] % in relation to AKC. Reaction conditions: 50 °C, 35 min, 4 mL solution with [AKC] = 0.50 mol L⁻¹, 35 min, 10 bar H₂.

The TON towards formate is relatively small when compared with those reported in the literature (Table S6† compiles the results of ICCU by catalytic hydrogenation to formate and methanol using molecular catalysts). These values are limited because higher catalyst-to-substrate ratios are required to avoid AKC decomposition. However, the TOF is among the highest reported to date for ICCU to HCOO⁻, especially considering the reaction temperature and H₂ pressure.

ZB can be recovered after the reaction by two different approaches: (i) passing the reaction mixture through an anion exchange resin, and (ii) extract ZB and adducts using water from alcohols that are poorly miscible with H₂O (e.g., n-butanol), followed by ZB regeneration using an anion exchange resin. Both methods successfully recovered ZB-p, which retained its ability to capture CO₂ (Table S7†). On the other hand, the catalyst lost its catalytic activity during the regeneration step (Table S7†). Most articles involving ICCU hydrogenation processes only report the recovery of the base (Table S6†).

Finally, methanol was produced using Ru-MACHO-BH and ZB-p in EG. Methanol was obtained with a 66% yield after 20 h by hydrogenating AKC at 140 °C under 70 bar H₂, with 0.5 mol L⁻¹ of ZB-p and 0.80% of Ru-MACHO-BH, Table S4, entry 9.†

Conclusions

Zwitterionic bases dissolved in alcohols produce AKCs by reacting with CO₂ at room temperature. ZB conversion to AKC ranges from 50–70%, depending on the pK_aH of the chosen base and the alcohol used. The adduct is easily hydrogenated to formate in a wide range of alcohols. On the other hand, tertiary amines such as MDEA, TEA, and TMEDA, which have similar pK_aH values to ZBs, do not significantly capture CO₂ and, therefore, are not suitable for ICCU using simple alcohols. The optimal catalytic conditions were found by combining ZB-p with isopropanol and hydrogenating the AKC at 50 °C and 10 bar H₂ with 0.25% cis-[Ru(dppm)₂Cl₂], which produced formate with a TON of 3.8 × 10² and a TOF of 6.6 × 10² h⁻¹, and are among the most active systems reported to date, especially at such low temperatures and H₂ pressure. Methanol was obtained with a 66% yield after 20 h by hydrogenating AKC at 140 °C under 70 bar H₂, with 0.5 mol L⁻¹ of ZB-p and 0.80% of Ru-MACHO-BH.

ICCU through the hydrogenation of AKC seems to proceed via direct alkyl carbonate hydrogenation rather than CO₂ formed by desorption during heating. Experiments with low catalyst loadings indicate that decomposition of AKC competes with hydrogenation, and the desorption is detrimental to the ICCU process. Ideally, hydrogenation to formate should be performed at lower temperatures to disfavour CO₂ evolution. This is fundamental to the quantitative conversion of AKC to HCOO⁻. Methanol formation requires higher temperatures, and an ideal ICCU to methanol should be performed in two steps: a lower temperature step to obtain formate and a higher temperature step to produce methanol. Studies are currently being conducted to investigate reaction mechanisms, focusing on the formation of MeOH.

Author contributions

PM and JD: conceptualization; BAC: data curation; PM, JD and FPS: funding acquisition; PM, BAC, MLCZ, FPS and WV: investigation; PM and BAC: methodology; PM, JD and FPS: project administration; PM: supervision; PM and BAC: writing – original draft; all authors: writing – review & editing.

Data availability

The data supporting this article have been included as part of the ESI.†

Data for this article, including NMR fid files are available at Open Science Framework at https://osf.io/62zjt/?view_only=41215f9972c6452db7d9fba58d815fea.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Authors would like to thank Dr Graciane Marin and Dr Günter Ebeling, both from UFRGS, for providing ZB-p and ZB-m in the beginning of this work and for helpful discussions regarding ZB purification. Authors would like to also thank CNPq (grants INCT-Catálise 444061/2018-5 and 406260/2018-4), Finep (Grant 2439/22), FAPERGS (grant 22/2551-0000386-9), and Petrobras (Grant 5850.0109122.18.9) for financial support During the preparation of this work, the authors used ChatGPT 4.0 for text proofreading of the main text and the translation of the text in ESI,† and Adobe Illustrator for image improvements. After using these tools, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

References

1. M. Peplow, Nature, 2022, 603, 780–783.
2. J. Dupont and P. Lozano, Angew. Chem., Int. Ed., 2024, e202416459.
3. J. Klankermayer, S. Wesselbaum, K. Beydoun and W. Leitner, Angew. Chem., Int. Ed., 2016, 55, 7296–7343.
4. R. Chauvy and G. De Weireld, Energy Technol., 2020, 8, 2000627.
5. A. Weilhard, M. I. Qadir, V. Sans and J. Dupont, ACS Catal., 2018, 8, 1628–1634.
6. R. Sen, A. Goeppert, S. Kar and G. K. S. Prakash, J. Am. Chem. Soc., 2020, 142, 4544–4549.
7. A. Bordet, S. El Sayed, M. Sanger, K. J. Boniface, D. Kalsi, K. L. Luska, P. G. Jessop and W. Leitner, Nat. Chem., 2021, 13, 916–922.
8. A. Moazezbarabadi, D. Wei, H. Junge and M. Beller, ChemSusChem, 2022, 15, e202201502.
9. M. I. Qadir and J. Dupont, Angew. Chem., Int. Ed., 2023, 62, e202301497.
10. C. J. Koch, A. Goeppert and G. K. Surya Prakash, ChemSusChem, 2024, 17, e202301789.
11. T. Neves-Garcia, M. Hasan, Q. Zhu, J. Li, Z. Jiang, Y. Liang, H. Wang, L. M. Rossi, R. E. Warburton and L. R. Baker, J. Am. Chem. Soc., 2024, 146, 31633–31646.
12. S.-T. Bai, G. De Smet, Y. Liao, R. Sun, C. Zhou, M. Beller, B. U. W. Maes and B. F. Sels, Chem. Soc. Rev., 2021, 50, 4259–4298.
13. J. Dupont, B. C. Leal, P. Lozano, A. L. Monteiro, P. Migowski and J. D. Scholten, Chem. Rev., 2024, 124, 5227–5420.
14. A. Otto, T. Grube, S. Schiebahn and D. Stolten, Energy Environ. Sci., 2015, 8, 3283–3297.
15. B. Gurkan, X. Su, A. Klemm, Y. Kim, S. Mallikarjun Sharada, A. Rodriguez-Katakura and K. J. Kron, iScience, 2021, 24, 103422.
16. M. Li, E. Irtem, H.-P. Iglesias van Montfort, M. Abdinejad and T. Burdyny, Nat. Commun., 2022, 13, 5398.
17. L. Zhao, H.-Y. Hu, A.-G. Wu, A. O. Terent'ev, L.-N. He and H.-R. Li, J. CO2 Util., 2024, 82, 102753.
18. M. Zanatta, ACS Mater. Au, 2023, 3, 576–583.
19. J. Kothandaraman and D. J. Heldebrant, RSC Adv., 2020, 10, 42557–42563.
20. J. R. Khusnutdinova, J. A. Garg and D. Milstein, ACS Catal., 2015, 5, 2416–2422.
21. S. Kar, A. Goeppert and G. K. S. Prakash, Acc. Chem. Res., 2019, 52, 2892–2903.
22. Y. N. Li, L. N. He, A. H. Liu, X. D. Lang, Z. Z. Yang, B. Yu and C. R. Luan, Green Chem., 2013, 15, 2825–2829.
23. Y. N. Li, L. N. He, X. D. Lang, X. F. Liu and S. Zhang, RSC Adv., 2014, 4, 49995–50002.
24. M. Yadav, J. C. Linehan, A. J. Karkamkar, E. van der Eide and D. J. Heldebrant, Inorg. Chem., 2014, 53, 9849–9854.
25. R. Sen, C. J. Koch, A. Goeppert and G. K. S. Prakash, ChemSusChem, 2020, 13, 6318–6322.
26. J. Kothandaraman, A. Goeppert, M. Czaun, G. A. Olah and G. K. S. Prakash, Green Chem., 2016, 18, 5831–5838.
27. C. J. Koch, V. Galvan, A. Goeppert and G. K. S. Prakash, Green Chem., 2023, 25, 1803–1808.
28. R. Sen, C. J. Koch, V. Galvan, N. Entesari, A. Goeppert and G. K. S. Prakash, J. CO₂ Util., 2021, 54, 101762.
29. Z. Suhail, C. J. Koch, A. Goeppert and G. K. S. Prakash, Langmuir, 2024, 40, 5401–5408.
30. G. L. P. Aydos, G. Marin, G. Ebeling, F. P. dos Santos, B. C. Leal, R. D. Zink, B. A. Vargas, P. Migowski, R. Stieler, B. B. de Araújo, P. Gonçalves, H. K. Stassen, L. S. Pereira, A. P. Musse and J. Dupont, ChemSusChem, 2023, 16, e202300971.
31. D. Prat, A. Wells, J. Hayler, H. Sneddon, C. R. McElroy, S. Abou-Shehada and P. J. Dunn, Green Chem., 2016, 18, 288–296.
32. M. Scott, B. B. Molinos, C. Westhues, G. Franciò and W. Leitner, ChemSusChem, 2017, 10, 1085–1093.
33. K. R. Ehmann, J.-M. Ji, K. Dinsing, C. R. Maier, A. J. Vorholt and W. Leitner, ACS Sustainable Chem. Eng., 2024, 12, 15569–15577.
34. N. Guntermann, G. Franciò and W. Leitner, Green Chem., 2022, 24, 8069–8075.
35. J. E. Rainbolt, P. K. Koech, C. R. Yonker, F. Zheng, D. Main, M. L. Weaver, J. C. Linehan and D. J. Heldebrant, Energy Environ. Sci., 2011, 4, 480–484.

---

Footnote

† Electronic supplementary information (ESI) available: All experimental details (including materials, methods and equations used), NMR spectra and supporting tables with supporting data. See DOI: https://doi.org/10.1039/d4gc05917d

---