A dual-functional catalyst for selective dicarbamate synthesis via oxidative carbonylation: enhanced methoxylation for suppressing urea polymer formation

Abstract

High-yield dicarbamate synthesis via oxidative carbonylation has been challenging due to substantial urea polymer formation. A Pd/CeO₂ catalyst was successfully synthesized dicarbamates while suppressing urea polymer formation. The superior methoxylation reactivity of Pd/CeO₂, which originated from Pd–CeO₂ synergism, facilitated the conversion of the urea species, thereby preventing urea accumulation.

---

With the growing concern over the climate crisis and environmental issues, the demand for low-carbon and eco-friendly production is continuously increasing. Polyurethane (PUR), an important petrochemical with wide applications such as packaging, furniture, coatings, adhesives, and thermoplastics,^1,2 is also subject to these environmental demands. PUR is traditionally synthesized by the reaction of polyols and diisocyanates. The green synthesis of polyols is well-established, achieving a significant reduction in the carbon footprint by up to 3 kg per kg of CO₂ incorporated,³ whereas diisocyanate synthesis still relies on petroleum-derived chemicals with high carbon footprints and extremely toxic phosgene.^4,5 Therefore, it is essential to develop green diisocyanate synthesis processes that employ low-carbon materials and avoid the use of phosgene for eco-friendly PUR production.

Non-phosgene diisocyanates can be practically obtained through the thermal decomposition of dicarbamates, indicating that the efficient and sustainable synthesis of dicarbamates is important.^6,7 Dicarbamates can be synthesized by reacting nitro- or amino compounds with carbonyl sources, such as dimethyl carbonate (DMC), CO₂, and CO.^8–12 The synthesis route using DMC is advantageous due to its mild reaction conditions and high carbamate yields. However, this method requires additional processes for DMC synthesis, such as transesterification of cyclic carbonates with methanol and oxidative carbonylation of methanol.^7,13 The direct synthesis of dicarbamates from CO₂ is highly desirable and promising, given that CO₂ is abundant and nontoxic. However, due to the high thermodynamic stability and low reaction kinetics of CO₂, high operating temperatures and pressures are required, and more than equivalent quantities of environmentally and industrially unfavored substances such as organic halides and metal alkoxides would be necessary.^14–21 Numerous efforts have been made to overcome these drawbacks of the CO₂ route, including the use of dehydrating agents to shift the thermodynamic equilibrium¹⁸ and adoption of alcohol as a more sustainable coupling reagent.²⁰ However, these approaches still require complex reaction systems or excess reagents.

The oxidative carbonylation (OC) route, which uses CO as a carbonyl source, presents a more industrially viable method for dicarbamate synthesis. The advantages of the OC route include: (1) high atom efficiency by incorporating two C1 building blocks (CO and methanol); (2) relatively high yields, which are beneficial for practical applications; and (3) CO₂ utilization, as CO can be readily produced from CO₂ by reactions such as reforming, hydrogenation, and the reverse Boudouard reaction (CO₂ + C → 2CO).²² Recently, as part of the Carbon2Polymer project in Germany, dicarbamate synthesis via oxidative carbonylation was conducted.^4,6,23 In this process, industrially relevant aromatic dicarbamates, toluene-2,4-dicarbamate (TDC) and methylene diphenyl-4,4′-dicarbamate (MDC), were successfully synthesized with palladium-supported heterogeneous catalysts via OC. Moreover, the feasibility of processes for the diisocyanate production via the OC route was demonstrated through a basic process design.⁴

Oxidative carbonylation typically provides fairly high carbamate yields. 79.5% selectivity of methyl N-phenyl carbamate (MPC) was achieved on a Pd/C catalyst system with 80% conversion of aniline.²⁴ A polymer-supported palladium–manganese bimetallic catalyst exhibited a higher yield of ethyl N-phenyl carbamate (90%).²⁵ This superior performance of oxidative carbonylation has encouraged various studies in terms of catalysts, reaction conditions, and additives.^24–28 However, most of the research on oxidative carbonylation has been devoted to monocarbamate synthesis, despite dicarbamates, which have two functional groups, being essential in polymerization reactions for PUR.^24,29–32Only a few studies have dealt with dicarbamate synthesis via OC.^33–36 Leitner's group recently reported a novel catalytic system with heterogeneous Pd catalysts for dicarbamate synthesis. This system yielded 44% TDC and 24% MDC.^6,37 They elucidated the reaction network by thoroughly analyzing products and investigated the effect of the support on dicarbamate yields. Another key finding of this study was the identification of yield limitations not observed in the monocarbamate synthesis. They suggested that a major cause of the limited dicarbamate yields was the formation of a considerable amount of solid precipitates, mainly consisting of urea derivatives.

The formation of solid precipitates (urea polymer) during dicarbamate synthesis can be explained by the OC reaction pathway. The dicarbamate synthesis consists of diamine carbonylation, which forms urea intermediates (eqn (1)), and subsequent urea methoxylation to convert the urea intermediates to carbamates (eqn (2)).^38–40

This indicates that when the methoxylation rate is slower than the carbonylation rate, the urea intermediates continue to grow and produce high-molecular-weight and unreactive urea polymer (Scheme 1). Therefore, developing a catalytic system that balances the rates of carbonylation and methoxylation by accelerating the methoxylation is crucial in reducing the formation of urea polymer.

Scheme 1 Dicarbamate synthesis via oxidative carbonylation of diamines.

Herein, we identified catalytic and reaction characteristics necessary to balance the carbonylation and methoxylation rates in dicarbamate synthesis via OC. By conducting carbamate synthesis under various reaction conditions and analyzing the individual reaction rates of carbonylation and methoxylation, we found that enhancing the methoxylation rate is critical in minimizing the formation of urea polymer. It is well known that catalyst supports significantly influence the physicochemical properties of active sites and can directly participate in chemical reactions. Thus, we investigated the effects of the catalyst support on methoxylation reactivity using several supports (Al₂O₃, SiO₂, and CeO₂) that are typically used in catalytic reactions. Notably, the interaction between Pd and CeO₂ increased the methoxylation reactivity by activating urea and methanol, leading to the successful synthesis of aromatic dicarbamates using the Pd/CeO₂ catalyst with minimal solid precipitate formation.

We employed Pd catalysts supported on metal oxides for the oxidative carbonylation of 2,4-toluenediamine (TDA) (Fig. 1a). All the prepared catalysts were characterized using X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and transmission electron microscopy (TEM) (Fig. S1–3†). TDA is completely consumed and converted over all the catalysts, but TDC yields vary considerably depending on the catalysts. Pd/CeO₂ shows the highest TDC yield among the supported Pd catalysts, and the yield is even higher than that obtained using PdCl₂, which is known to be an effective catalyst for the oxidative carbonylation of aniline.^40–42 The higher performance of Pd/CeO₂ is notable since PdCl₂, a homogeneous catalyst, has many more accessible Pd sites than Pd/CeO₂ (Fig. S1 and S3†). This result indicates that the number of exposed Pd sites is not the only factor that results in higher TDC productivity. Moreover, the different TDC yields depending on the supports indicate that the catalyst support plays an important role in TDC synthesis.

Fig. 1 TDC synthesis over homo- and heterogeneous Pd catalysts using different (a) catalysts, (b) temperature and (c) pressure. Reaction conditions: Pd/CeO₂ catalyst, Pd (2.8 mol%), TDA (0.03 M), NaI (30 mol%), CO/O₂ = 4, pressure (10 bar), temperature 135 °C, time 5 h. Reaction conditions for (d): pressure (5 bar). Reaction conditions for Pd/CeO₂(×2): Pd (5.6 mol%), NaI (60 mol%), time 10 h. The yield was determined using HPLC.

Insoluble solid matter was observed after the reaction with PdCl₂. Conversely, organic solid matter was rarely observed for Pd/CeO₂, as confirmed by thermogravimetric analysis (TGA) of the spent catalysts (Fig. 2). The weight loss in TGA measurements is due to thermal oxidation of organic solids that cannot be removed by solvent washing or drying. Pd/CeO₂ showed the lowest weight loss, indicating the lowest organic solid formation, followed by Pd/Al₂O₃ and later Pd/SiO₂. The TDC yield of Pd/CeO₂ was maintained consistently even after five recycling runs (Fig. S4†). However, Pd/Al₂O₃ and Pd/SiO₂ showed an obvious decline in the TDC yields with each cycle. This was probably due to the residual organics on the catalyst surface, hindering the interaction between the active sites and reactants. As described above, this organic solid could possibly be urea polymer, which does not further react to produce TDC (Scheme 1).⁶ Since the urea polymer with high molecular weight forms when the amine carbonylation is faster than urea methoxylation, it is necessary to suppress urea polymer formation by controlling the reaction rates of carbonylation and/or methoxylation to obtain a high TDC yield.

Fig. 2 TGA curves of fresh and spent Pd catalysts.

The effects of the reaction conditions on the reaction rates of carbonylation and methoxylation were investigated (Fig. 1b–d). The TDC yields for different reaction temperatures are shown in Fig. 1b. The yield increases in the order of 110 °C < 160 °C < 135 °C. TGA results showed that this order is closely related to the amount of urea polymer formation (Fig. S5†). The amount of urea polymer, as determined by weight loss in TGA, was 22.6, 5.5, and 8.9% after the reactions at 110, 135, and 160 °C, respectively. The reaction temperature showing the highest TDC yield exhibited the lowest weight loss and vice versa. The carbonylation reaction for urea synthesis from aniline is known to take place at around 100 °C.^39,43,44 However, MPC synthesis (aniline carbonylation + 1,3-diphenylurea (DPU) methoxylation) requires a much higher temperature of 150 °C,^29,31 indicating a higher reaction temperature for methoxylation than carbonylation. We can infer that the cause of the largest amount of urea polymer formation and the lowest TDC yield at 110 °C is the slower methoxylation rate relative to the carbonylation rate at this reaction temperature. Considering that the amount of urea polymer increases again at 160 °C, TDC synthesis is also not favourable at an excessively high temperature, and it appears that amine carbonylation and urea methoxylation rates are well balanced at 135 °C in our catalytic system.

The TDC yield and reaction pressure show an inverse relationship (Fig. 1c). Namely, the TDC yield increases as the reaction pressure decreases. In general, high pressure accelerates the overall reaction rate of carbamate synthesis, which is attributed to the increased CO and O₂ solubility.^45,46 However, this is also related to the respective reaction rates of amine carbonylation and urea methoxylation. The reaction pressure of 5 bar might be high enough for amine carbonylation to proceed and low enough to hinder urea polymer formation. Conclusively, we proved the significant impact of the catalysts and reaction conditions on urea polymer formation and TDC yield. This finding suggests that adjusting catalytic properties and the reaction conditions is a viable strategy to achieve a balance between the carbonylation and methoxylation rates.

Finally, we compared the TDC yields over PdCl₂ and Pd/CeO₂ under the optimized conditions (Fig. 1d). The highest TDC yield of 53% was obtained over Pd/CeO₂. In contrast, PdCl₂ shows only a 27% TDC yield under the same reaction conditions. Since the conversion was 100% on both catalysts, the difference in the TDC yield is due to a difference in methoxylation ability. In addition, when the amounts of catalysts and reaction time were increased twofold, the TDC yield was increased by up to 80%. As shown above, the urea polymer formation is a decisive hurdle in TDC synthesis, and the different reaction rates between carbonylation and methoxylation are the most important factor in this regard.

We thus carried out MPC synthesis from aniline to reveal the rate difference between amine carbonylation and urea methoxylation over Pd/CeO₂ and PdCl₂ (Fig. 3). Aniline, which has a single amine functional group, was applied as a substrate. The MPC synthesis from aniline via oxidative carbonylation proceeds via two sequential reactions: carbonylation, which converts aniline to DPU, and methoxylation, which converts DPU to MPC, which are identical processes to TDC synthesis (Scheme 1). In MPC synthesis over Pd/CeO₂, aniline conversion and MPC yield rapidly increased from 20 to 120 min, showing 96% conversion and 91% yield. The similar slopes of the aniline conversion and MPC yield indicate that the carbonylation and methoxylation rates are similar on Pd/CeO₂. The yield of the intermediate DPU increased initially, but decreased after 40 min, resulting in a high yield of MPC. On the contrary, different reaction rates between carbonylation and methoxylation were confirmed on the PdCl₂ catalyst. The slope of the MPC yield is lower than that of the aniline conversion, indicating a slow methoxylation rate. As a result, the yield of the intermediate DPU steadily increased up to 60 min and further decreased much more slowly compared to that of Pd/CeO₂. These experiments provide evidence that the high MPC yield of Pd/CeO₂ is essentially attributable to the fast methoxylation rate. The fast methoxylation rate of Pd/CeO₂ favorably affected the TDC yield, as confirmed in Fig. 1.

Fig. 3 Time course of MPC synthesis over Pd/CeO₂ and PdCl₂. Reaction conditions: aniline (0.06 M), Pd (1.4 mol%), NaI (30 mol%), temperature 135 °C, pressure 5 bar. Conversion and yield were determined using GC-MS.

To elucidate the effects of the Pd catalyst supports on the methoxylation rates, DPU methoxylation was tested using Pd/CeO₂, Pd/Al₂O₃, and Pd/SiO₂ (Table 1). As described above, this reaction can reflect the methoxylation reaction in the TDC synthesis. The MPC yields follow the order Pd/CeO₂ > Pd/Al₂O₃ > Pd SiO₂, which correlates with the TDC yields (Fig. 1a). These results demonstrate that the methoxylation of the urea intermediate is the key step in the overall dicarbamate synthesis, and the high activity of Pd/CeO₂ in TDC synthesis is caused by the accelerated methoxylation. The roles of the individual catalyst components for the carbonylation and methoxylation were investigated. The carbonylation only proceeds in the presence of Pd and NaI. The absence of either one results in no conversion of amine (Table S1†), indicating that Pd–I acts as the active site for the carbonylation reaction. This finding is highly consistent with previous research.^24,47 Analysis of entries 4–6 in Table 1 reveals that using CeO₂ by itself or adding NaI does not lead to any improvement in the methoxylation reaction. However, a significant increase in carbamate yield is observed for the Pd/CeO₂ catalyst. This indicates that the synergistic interaction between Pd and CeO₂ has a positive impact on the methoxylation rate.

Table 1 DPU methoxylation using different catalysts and NaI

Entry	Catalyst	NaI (mol%)	MPC yield (%)
1	Pd/CeO2	—	95
2	Pd/Al2O3	—	81
3	Pd/SiO2	—	45
4	Pd/CeO2	30	94
5	CeO2	30	66
6	—	—	67

---

An XPS analysis was performed to identify the origin of the high methoxylation rate of the Pd/CeO₂ catalyst. Pd 3d XPS spectra consisting of Pd 3d_5/2 and Pd_3/2 are presented in Fig. 4a, and the Pd²⁺/Pd⁰ ratio was calculated through peak fitting deconvolution. Pd/CeO₂ exhibits a higher Pd²⁺ ratio (42%) than Pd/SiO₂ (32%) and Pd/Al₂O₃ (31%). Fig. 4b shows the Ce 3d spectra of CeO₂ and Pd/CeO₂. The oxidation state of CeO₂ was identified as Ce⁴⁺ (filled in yellow) with a small portion of Ce³⁺ (filled in purple). Conversely, Pd/CeO₂ shows obviously increased Ce³⁺ peaks compared to CeO₂. This high ratio of Pd²⁺ to Ce³⁺ in the Pd/CeO₂ catalyst is due to the substitution of Ce⁴⁺ by Pd²⁺, as reported in previous research.⁴⁸ The effect of Pd²⁺ and Ce³⁺ on the methoxylation rates can be inferred from the literature. According to Kostic's report, Pd²⁺ efficiently catalyzes the alcoholysis of urea, which is a similar process to the methoxylation of the urea intermediate.⁴⁹ The activation of the carbonyl moiety in urea by Pd²⁺ facilitates the nucleophilic attack of methanol, resulting in an increased methoxylation rate, which ultimately leads to a high yield of TDC. Somorjai et al. reported the effect of Ce³⁺ on methanol decomposition. They demonstrated that Ce³⁺ sites with a neighboring oxygen vacancy (Ce³⁺–Ov–Ce³⁺) enhance the adsorption capacity of methanol and stabilize adsorbed methoxy species well. The reactivity of Ce³⁺ with methanol is beneficial to the methoxylation reaction in TDC synthesis. Based on these reports, it can be concluded that the superior methoxylation activity of Pd/CeO₂ originates from its high ratios of Pd²⁺ and Ce³⁺. This electronic structure of Pd/CeO₂ remained stable even after the reaction (Fig. S2†), contributing to the consistent catalyst reactivity (Fig. S4†).

Fig. 4 (a) Pd 3d XPS spectra of reduced Pd/CeO₂, Pd/Al₂O₃ and Pd/SiO₂. (b) Ce 3d XPS spectra of reduced CeO₂ and Pd/CeO₂. (c) FT-IR spectra of the catalysts after sequential exposure to methanol/N₂ and N₂ flow at 50 °C for 1 h.

In situ diffuse reflectance infrared transform spectroscopy (DRIFTS) experiments were conducted with a methanol flow to elucidate the difference in the methoxylation rate for the different catalysts. Methanol/N₂ gas was introduced into the catalyst for 30 min, followed by N₂ purging for 1 h at 50 °C (the detailed experimental procedure is described in the ESI†). Strong gaseous methanol peaks, which make it difficult to observe chemisorbed species, were mainly observed under the methanol/N₂ flow (Fig. S6†). After N₂ purging to remove the gaseous methanol and physisorbed species, the chemisorbed species were observed. Fig. 4c shows Fourier transform infrared (FT-IR) spectra in the range of the C–O vibration of the methoxy species after N₂ purging. Three major peaks are observed at 1098, 1043, and 1022 cm⁻¹, corresponding to monodentate (1098 cm⁻¹) and multidentate (1043 and 1022 cm⁻¹) bridging methoxy species.^50–52

As can be seen in results, Pd/CeO₂ shows much higher intensity than the other catalysts. Relatively small peaks are observed for Pd/Al₂O₃. Pd/SiO₂ exhibits no characteristic adsorption peak. These results indicate that the abundance of the surface methoxy species follows the order Pd/CeO₂ > Pd/Al₂O₃ > Pd/SiO₂. The large amount of surface methoxy species on Pd/CeO₂ promotes urea methoxylation (refer to Table 1), thereby increasing the TDC yield (refer to Fig. 1a). We propose a reaction mechanism for dicarbamate synthesis via OC, based on the findings of this study and relevant literature. Initially, Pd metal is activated by NaI to form a Pd–I bond, then reacts with CO and amine, leading to the formation of a carbamoyl species. Subsequently, the carbamoyl species reacts with another amine to produce the urea intermediate. This urea intermediate then undergoes reaction with methanol to yield dicarbamate and diamine. The surface hydrogen generated by the deprotonation of amine is removed by oxygen, leading to the regeneration of the catalyst surface. The proposed reaction mechanism is illustrated in Scheme S1.†

We also applied our catalytic system to other diamine and alcohol substrates (Scheme 2). Pd/CeO₂ successfully synthesizes MDC, which is a precursor of MDI, occupying more than 40% of the PU market, via the oxidative carbonylation of 4,4′-methylenediphenyl diamine (MDA) (1). In contrast to that of aromatic amines, the synthesis of aliphatic amines did not proceed efficiently. A 1,6-hexanedicarbamate (2) yield of only 32% was observed, and isophorone dicarbamate (3) was not synthesized at all. Although the yield of aliphatic dicarbamates was low, the amine species was completely converted. This result indicates that the low aliphatic dicarbamate yield is most likely due to the high nucleophilicity of the aliphatic amine, which promotes urea polymer formation. The chain length of the alcohol greatly affected the dicarbamate yield (4–6). As the alkyl chain becomes longer, the corresponding dicarbamate yields are lower. It appears that short chain alcohols are better in terms of acting as a nucleophile for the substitution reaction. In addition, it is supposed that the longer the alcohol chain is, the lower the urea intermediate solubility is, which may inhibit alcoholysis, making it difficult to convert to carbamate.⁵³

Scheme 2 Oxidative carbonylation using various amines and alcohols.

In this study, we attempted to synthesize dicarbamates in high yields by alleviating the formation of urea polymer and to elucidate the relationship between the dicarbamate yield and urea polymer formation. Pd/CeO₂ showed a higher yield of TDC than the other catalysts, including homogeneous PdCl₂. A fairly high TDC yield of 80% was achieved over Pd/CeO₂ under mild reaction conditions (135 °C and 5 bar). The catalysts with high TDC yield showed low urea polymer formation and vice versa. Given that urea polymer can be formed when the methoxylation rate is slower than the carbonylation rate, the small amount of urea polymer over Pd/CeO₂ was caused by the fast methoxylation rate. Indeed, the superior methoxylation reactivity of Pd/CeO₂ was confirmed in the MPC synthesis and DPU methoxylation reaction. XPS and DRIFT experiments were conducted to investigate the origin of the superior methoxylation of Pd/CeO₂. XPS results indicated that Pd/CeO₂ had high ratios of Pd²⁺ and Ce³⁺, which are known to be beneficial for the activation of urea and methanol, respectively. In the DRIFT experiment under a methanol flow, a much higher amount of surface methoxy species was observed in Pd/CeO₂ compared to the other catalysts. The abundant methoxy species on the catalyst surface can promote the methoxylation rate, which in turn leads to a high yield of dicarbamates. We believe that these findings will encourage the development of efficient catalytic systems for eco-friendly non-phosgene isocyanate production.

Author contributions

S. H: data curation, formal analysis, investigation, methodology, writing – original draft. Y-W. Y: formal analysis, investigation, resources, writing – review & editing. K. J: investigation, resources. M. I: investigation, resources. J-A. L: investigation, resources. J. H. L: methodology, funding acquisition, supervision, validation, writing – review & editing. J. H. P: conceptualization, methodology, project administration, funding acquisition, supervision, validation, writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by Institutional Research Program of KRICT (KK2411-20) and Technology Innovation Program (20225A10100040) funded by the Ministry of Trade, Industry & Energy.

References

1. D. K. Chattopadhyay and K. V. S. N. Raju, Prog. Polym. Sci., 2007, 32, 352–418.
2. H. M. C. C. Somarathna, S. N. Raman, D. Mohotti, A. A. Mutalib and K. H. Badri, Constr. Build. Mater., 2018, 190, 995–1014.
3. N. von der Assen and A. Bardow, Green Chem., 2014, 16, 3272–3280.
4. T. Kaiser, A. Rathgeb, C. Gertig, A. Bardow, K. Leonhard and A. Jupke, Chem. Ing. Tech., 2018, 90, 1497–1503.
5. S. A. Cucinell and E. Arsenal, Arch. Environ. Health, 1974, 28, 272–275.
6. C. Hussong, J. Langanke and W. Leitner, Chem. Ing. Tech., 2020, 92, 1482–1488.
7. P. Wang, S. Liu and Y. Deng, Chin. J. Chem., 2017, 35, 821–835.
8. O. Kreye, H. Mutlu and M. A. R. Meier, Green Chem., 2013, 15, 1431–1455.
9. Y. Cao, Y. Chi, A. Muhammad, P. He, L. Wang and H. Li, Chin. J. Chem. Eng., 2019, 27, 549–555.
10. Y. Dai, Y. Wang, J. Yao, Q. Wang, L. Liu, W. Chu and G. Wang, Catal. Lett., 2008, 123, 307–316.
11. R. Juárez, A. Corma and H. García, Top. Catal., 2009, 52, 1688–1695.
12. D.-L. Sun, J.-Y. Luo, R.-Y. Wen, J.-R. Deng and Z.-S. Chao, J. Hazard. Mater., 2014, 266, 167–173.
13. H. Wang, B. Lu, X. Wang, J. Zhang and Q. Cai, Fuel Process. Technol., 2009, 90, 1198–1201.
14. Q. Zhang, H.-Y. Yuan, N. Fukaya, H. Yasuda and J.-C. Choi, ChemSusChem, 2017, 10, 1501–1508.
15. D. Riemer, P. Hirapara and S. Das, ChemSusChem, 2016, 9, 1916–1920.
16. Q. Zhang, H.-Y. Yuan, N. Fukaya and J.-C. Choi, ACS Sustainable Chem. Eng., 2018, 6, 6675–6681.
17. K. A. Alferov, Z. Fu, S. Ye, D. Han, S. Wang, M. Xiao and Y. Meng, ACS Sustainable Chem. Eng., 2019, 7, 10708–10715.
18. Y. Gu, A. Miura, M. Tamura, Y. Nakagawa and K. Tomishige, ACS Sustainable Chem. Eng., 2019, 7, 16795–16802.
19. Q. Zhang, H.-Y. Yuan, N. Fukaya, H. Yasuda and J.-C. Choi, Green Chem., 2017, 19, 5614–5624.
20. Q. Zhang, H.-Y. Yuan, X.-T. Lin, N. Fukaya, T. Fujitani, K. Sato and J.-C. Choi, Green Chem., 2020, 22, 4231–4239.
21. M. Abla, J.-C. Choi and T. Sakakura, Green Chem., 2004, 6, 524–525.
22. P. Lahijani, Z. A. Zainal, M. Mohammadi and A. R. Mohamed, Renewable Sustainable Energy Rev., 2015, 41, 615–632.
23. W. Leitner, G. Franciò, M. Scott, C. Westhues, J. Langanke, M. Lansing, C. Hussong and E. Erdkamp, Chem. Ing. Tech., 2018, 90, 1504–1512.
24. S. P. Gupte and R. V. Chaudhari, J. Catal., 1988, 114, 246–258.
25. B. Wan, S. Liao and D. Yu, Appl. Catal., A, 1999, 183, 81–84.
26. S. Fukuoka, M. Chono and M. Kohno, J. Chem. Soc., Chem. Commun., 1984, 399–400.
27. H. Alper and F. W. Hartstock, J. Chem. Soc., Chem. Commun., 1985, 1141–1142.
28. B. Chen and S. S. C. Chuang, J. Mol. Catal. A: Chem., 2003, 195, 37–45.
29. S. Kanagasabapathy, A. Thangaraj, S. P. Gupte and R. V. Chaudhari, Catal. Lett., 1994, 25, 361–364.
30. F. Ferretti, E. Barraco, C. Gatti, D. R. Ramadan and F. Ragaini, J. Catal., 2019, 369, 257–266.
31. F.-M. Mei, L.-J. Chen and G.-X. Li, Appl. Organomet. Chem., 2010, 24, 86–91.
32. P. Toochinda and S. S. C. Chuang, Ind. Eng. Chem. Res., 2004, 43, 1192–1199.
33. F. Shi and Y. Deng, J. Catal., 2002, 211, 548–551.
34. F. Shi and Y. Deng, Chem. Commun., 2001, 443–444.
35. F. Shi, Y. Deng, T. SiMa and H. Yang, J. Catal., 2001, 203, 525–528.
36. V. Valli and H. Alper, Organometallics, 1995, 14, 80–82.
37. C. Hussong, J. Langanke and W. Leitner, Green Chem., 2020, 22, 8260–8270.
38. F. Ragaini, Dalton Trans., 2009, 6251–6266.
39. B. Gabriele, G. Salerno, R. Mancuso and M. Costa, J. Org. Chem., 2004, 69, 4741–4750.
40. A. Krogul and G. Litwinienko, J. Mol. Catal. A: Chem., 2015, 407, 204–211.
41. H. Alper, G. Vasapollo, F. W. Hartstock, M. Mlekuz, D. J. H. Smith and G. E. Morris, Organometallics, 1987, 6, 2391–2393.
42. P. Giannoccaro, J. Organomet. Chem., 1987, 336, 271–278.
43. S. Liu, X. Dai, H. Wang and F. Shi, Green Chem., 2019, 21, 4040–4045.
44. S. T. Gadge, E. N. Kusumawati, K. Harada, T. Sasaki, D. Nishio-Hamane and B. M. Bhanage, J. Mol. Catal. A: Chem., 2015, 400, 170–178.
45. F. Wu, Q. Zhao, L. Tao, D. Danaci, P. Xiao, F. A. Hasan and P. A. Webley, J. Chem. Eng. Data, 2019, 64, 5609–5621.
46. K. Fischer and M. Wilken, J. Chem. Thermodyn., 2001, 33, 1285–1308.
47. A. A. Kelkar, D. S. Kolhe, S. Kanagasabapathy and R. V. Chaudhari, Ind. Eng. Chem. Res., 1992, 31, 172–176.
48. Y.-Q. Su, J.-X. Liu, I. A. W. Filot, L. Zhang and E. J. M. Hensen, ACS Catal., 2018, 8, 6552–6559.
49. N. V. Kaminskaia and N. M. Kostić, Inorg. Chem., 1998, 37, 4302–4312.
50. Z. Qi, L. Chen, S. Zhang, J. Su and G. A. Somorjai, J. Am. Chem. Soc., 2021, 143, 60–64.
51. E. Finocchio, M. Daturi, C. Binet, J. C. Lavalley and G. Blanchard, Catal. Today, 1999, 52, 53–63.
52. A. Badri, C. Binet and J.-C. Lavalley, J. Chem. Soc., Faraday Trans., 1997, 93, 1159–1168.
53. F.-M. Lee and L. E. Lahti, J. Chem. Eng. Data, 1972, 17, 304–306.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4gc01045k

---