Thank you for visiting Nature.com. The version of browser you are using has limited CSS support. For best results, we recommend that you use a newer version of your browser (or disable Compatibility Mode in Internet Explorer). In the meantime, to ensure ongoing support, we are showing the site without styling or JavaScript.
Formic acid is one of the most promising candidates for long-term storage of liquid hydrogen. Here we present a series of new ruthenium clamp complexes with the general formula [RuHCl(POP)(PPh3)] using commercially available or easily synthesized xanthos-type tridentate POP clamp ligands. We used these complexes to dehydrogenate formic acid to produce CO2 and H2 under mild, reflux-free conditions using the ionic liquid BMIM OAc (1-butyl-3-methylimidazolium acetate) as a solvent. From the point of view of the maximum turnover frequency, the most effective catalyst is the [RuHCl(xantphos)(PPh3)]Ru-1 complex known in the literature, which has a maximum turnover frequency of 4525 h-1 at 90 °C for 10 min. The post-conversion rate was 74%, and the conversion was completed within 3 hours (>98%). On the other hand, the catalyst with the best overall performance, the novel [RuHCl(iPr-dbfphos)(PPh3)]Ru-2 complex, promotes complete conversion within 1 h, resulting in an overall turnover rate of 1009 h-1. In addition, catalytic activity was also observed at temperatures up to 60 °C. In the gas phase, only CO2 and H2 were observed; CO was not detected. High-resolution mass spectrometry showed the presence of N-heterocyclic carbene complexes in the reaction mixture.
The growing market share of renewable energy and its variability has led to demand for industrial-scale energy storage technologies in the power, thermal, industrial and transport sectors1,2. Hydrogen is considered one of the most abundant energy carriers3, and liquid organic hydrogen carriers (LOHCs) have recently become the focus of research, offering the promise of storing hydrogen in an easily processed form without the problems associated with pressurization or cryogenic technologies4. ,5,6. Due to their physical properties, much of the existing transportation infrastructure for gasoline and other liquid fuels can be used to transport LOHC7,8. The physical properties of formic acid (FA) make it a promising candidate for hydrogen storage with a hydrogen weight content of 4.4%9,10. However, published catalytic systems for formic acid dehydrogenation typically require the use of volatile organic solvents, water or pure formic acid,11,12,13,14 which may necessitate the use of solvent vapor separation techniques such as condensation, which can lead to problems in consumer applications. applications, additional load. This problem can be overcome by using solvents with negligible vapor pressure, such as ionic liquids. Previously, our working group demonstrated that the ionic liquid butylmethylimidazolium acetate (BMIM OAc) is a suitable solvent in this reaction using the commercially available fixing complex Ru-PNP Ru-MACHO type 15. For example, we demonstrated FA dehydrogenation in a continuous flow system using BMIM OAc , achieving a TON of over 18,000,000 at 95°C. Although some systems have previously achieved high TON, many have relied on volatile organic solvents (such as THF or DMF) or additives used (such as bases). In contrast, our work actually uses non-volatile ionic liquids (ILs) and no additives.
Hazari and Bernskoetter reported the dehydrogenation of formic acid (FA) at 80 °C using a Fe-PNP catalyst in the presence of dioxane and LiBF4, achieving an impressive turnover number (TON) of approximately 1,000,00016. Laurenci used a Ru(II)- complex catalyst TPPPTS in a continuous FA dehydrogenation system. This method resulted in almost complete FA dehydrogenation with very little traces of CO detected at 80 °C17. To further advance this field, Pidko demonstrated the reversible dehydrogenation of FA using Ru-PNP clamp catalysts in DMF/DBU and DMF/NHex₃ mixtures, achieving TON values of 310,000 to 706,500 at 90 °C18. Hull, Himeda and Fujita studied a binuclear Ir complex catalyst in which KHCO3 and H2SO4 were sacrificed, alternating CO2 hydrogenation and FA dehydrogenation. Their systems achieved TONs of over 3,500,000 and 308,000 respectively for hydrogenation at 30°C, CO2/H2 (1:1), 1 bar pressure and for dehydrogenation between 60 and 90°C19. Sponholz, Junge and Beller developed a Mn-PNP complex for reversible CO2 hydrogenation and FA dehydrogenation at 90 °C20.
Here we used an IL approach, but instead of using Ru-PNPs, we explored the use of Ru-POP catalysts, which to our knowledge have not been previously demonstrated in this regard.
Due to their excellent metal-ligand coupling (MLC), amino-PNP clamp complexes based on Noyori-type concepts with interacting secondary amino functional groups 21 (such as Ru-MACHO-BH) are generally becoming increasingly popular in some small molecule operations. Popular examples include CO22, hydrogenation of alkenes and carbonyls, transfer hydrogenation23 and acceptorless dehydrogenation of alcohols24. It has been reported that N-methylation of PNP clamp ligands can completely stop the catalyst activity25, which can be explained by the fact that amines serve as proton sources, which is an important requirement during the catalytic cycle using MLC. However, the opposite trend in formic acid dehydrogenation was recently observed by Beller, where N-methylated Ru-PNP complexes actually showed better catalytic dehydrogenation of formic acid than their unmethylated counterparts26. Since the former complex cannot participate in MLC via the amino unit, this strongly suggests that MLC, and hence the amino unit, may play a less important role in some (de)hydrogenation transformations than previously thought.
Compared to POP clamps, the ruthenium complexes of POP clamps have not been sufficiently studied in this area. POP ligands have traditionally been used primarily for hydroformylation, where they act as chelating ligands rather than the characteristic bidentate bite angle of approximately 120° for clamping ligands, which have been used to optimize selectivity for linear and branched products27,28,29. Since then, Ru-POP complexes have rarely been used in hydrogenation catalysis, but examples of their activity in transfer hydrogenation have previously been reported30. Here we demonstrate that the Ru-POP complex is an efficient catalyst for the dehydrogenation of formic acid, confirming Beller’s discovery that the amino unit in the classical Ru-PNP amine catalyst is less important in this reaction.
Our study begins with the synthesis of two typical catalysts with the general formula [RuHCl(POP)(PPh3)] (Fig. 1a). To modify the steric and electronic structure, dibenzo[b,d]furan was selected from commercially available 4,6-bis(diisopropylphosphino) (Fig. 1b) 31. The catalysts studied in this work were synthesized using a general method developed by Whittlesey32, with using the [RuHCl(PPh3)3]•toluene33 adduct as a precursor. Mix the metal precursor and POP clamp ligand in THF under strictly anhydrous and anaerobic conditions. The reaction was accompanied by a significant color change from dark purple to yellow and gave a pure product after 4 hours of reflux or 72 hours of reflux at 40°C. After removing the THF in vacuo and washing twice with hexane or diethyl ether, the triphenylphosphine was removed to give the product as a yellow powder in high quantitative yield.
Synthesis of Ru-1 and Ru-2 complexes. a) Method of synthesis of complexes. b) Structure of the synthesized complex.
Ru-1 is already known from the literature32, and further characterization focuses on Ru-2. The 1H NMR spectrum of Ru-2 confirmed the cis configuration of the phosphine atom in the ligand of the hydride pair. The peak dt plot (Fig. 2a) shows 2JP-H coupling constants of 28.6 and 22.0 Hz, which are within the expected range of previous reports32. In the hydrogen decoupled 31P{1H} spectrum (Fig. 2b), a 2JP-P coupling constant of approximately 27.6 Hz was observed, confirming that both the clamp ligand phosphines and PPh3 are cis-cis. In addition, ATR-IR shows a characteristic ruthenium-hydrogen stretching band at 2054 cm-1. For further structural elucidation, the Ru-2 complex was crystallized by vapor diffusion at room temperature with quality sufficient for X-ray studies (Fig. 3, Supplementary Table 1). It crystallizes in the triclinic system of space group P-1 with one cocrystalline benzene unit per unit cell. It exhibits a wide P-Ru-P occlusal angle of 153.94°, which is significantly wider than the 130° occlusal angle of the bidentate DBFphos34. At 2.401 and 2.382 Å, the Ru-PPOP bond length is significantly longer than the Ru to PPh3 bond length of 2.232 Å, which may be a result of the wide backbone snack angle of DBFphos caused by its central 5-ring. The geometry of the metal center is essentially octahedral with an O-Ru-PPh3 angle of 179.5°. The H-Ru-Cl coordination is not entirely linear, with an angle of approximately 175° from the triphenylphosphine ligand. Atomic distances and bond lengths are listed in Table 1.
NMR spectrum of Ru-2. a) Hydride region of the 1H NMR spectrum showing the Ru-H dt signal. b) 31 P{ 1 H} NMR spectrum showing signals from triphenylphosphine (blue) and POP ligand (green).
Structure of Ru-2. Thermal ellipsoids are displayed with a 70% probability. For clarity, the cocrystalline benzene and hydrogen atoms on the carbon have been omitted.
To evaluate the ability of the complexes to dehydrogenate formic acid, reaction conditions were selected under which the corresponding PNP-clamp complexes (e.g., Ru-MACHO-BH) were highly active15. Dehydrogenation of 0.5 ml (13.25 mmol) formic acid using 0.1 mol% (1000 ppm, 13 µmol) ruthenium complex Ru-1 or Ru-2 using 1.0 ml (5.35 mmol) ionic liquid (IL) BMIM OAc (table-figure) 2; Figure 4);
To obtain the standard, the reaction was first carried out using the precursor adduct [RuHCl(PPh3)3]·toluene. The reaction is carried out at a temperature from 60 to 90 °C. According to simple visual observations, the complex could not be completely dissolved in IL even with prolonged stirring at a temperature of 90°C, but dissolution occurred after the introduction of formic acid. At 90°C, a conversion of 56% (TOF = 3424 h-1) was achieved within the first 10 minutes, and almost quantitative conversion (97%) was achieved after three hours (entry 1). Reducing the temperature to 80 °C reduces the conversion by more than half to 24% after 10 min (TOF = 1467 h-1, entry 2), further reducing it to 18% and 18% at 70 °C and 60 °C, respectively 6% (entries 3 and 4). In all cases, no induction period was detected, suggesting that the catalyst may be reactive species or that the conversion of reactive species is too rapid to be detected using this data set.
After precursor evaluation, Ru-POP clamp complexes Ru-1 and Ru-2 were used under the same conditions. At 90°C, high conversion was immediately observed. Ru-1 achieved 74% conversion within the first 10 minutes of the experiment (TOFmax = 4525 h-1, entry 5). Ru-2 showed slightly less but more consistent activity, promoting 60% conversion within 10 min (TOFmax = 3669 h-1) and full conversion within 60 min (>99%) (entry 9). It is noteworthy that Ru-2 is significantly superior to the precursor metal and Ru-1 at full conversion. Therefore, while the metal precursor and Ru-1 have similar TOFoverall values at reaction completion (330 h-1 and 333 h-1, respectively), Ru-2 has a TOFoverall of 1009 h-1.
Ru-1 and Ru-2 were then subjected to a temperature change in which the temperature was gradually reduced in 10 °C increments to a minimum of 60 °C (Fig. 3). If at 90°C the complex showed immediate activity, almost complete conversion occurred within an hour, then at lower temperatures the activity dropped sharply. The conversion of Py-1 was 14% and 23% after 10 min at 80°C and 70°C, respectively, and after 30 min it increased to 79% and 73% (entries 6 and 7). Both experiments showed a conversion rate of ≥90% within two hours. Similar behavior was observed for Ru-2 (entries 10 and 11). Interestingly, Ru-1 was slightly dominant at the end of the reaction at 70 °C with a total TOF of 315 h-1 compared to 292 h-1 for Ru-2 and 299 h-1 for the metal precursor.
A further decrease in temperature to 60 °C led to the fact that no conversion was observed during the first 30 minutes of the experiment. Ru-1 was significantly inactive at the lowest temperature at the beginning of the experiment and subsequently increased activity, indicating the requirement of an activation period during which the Ru-1 precatalyst is converted into catalytically active species. Although this is possible at all temperatures, 10 minutes at the beginning of the experiment was not sufficient to detect an activation period at higher temperatures. Similar behavior was found for Ru-2. At 70 and 60 °C, no conversion was observed during the first 10 min of the experiment. It is important to note that in both experiments, carbon monoxide formation was not observed within the detection limit of our instrument (<300 ppm), with H2 and CO2 being the only products observed.
Comparison of formic acid dehydrogenation results obtained previously in this working group, representative of the state of the art and using Ru-PNP clamp complexes, showed that the newly synthesized Ru-POP clamp has activity similar to its PNP counterpart 15. While clamp The PNP achieved RPMs of 500-1260 h-1 in batch experiments, the new POP clamp achieved a similar TOFovertal value of 326 h-1, and TOFmax values of Ru-1 and 1590 h-1 were observed. respectively, are 1988 h-1 and 1590 h-1. Ru-2 is 1 at 80 °C, Ru-1 is 4525 h-1 and Ru-1 is 3669 h-1 at 90 °C, respectively.
Temperature screening of formic acid dehydrogenation using Ru-1 and Ru-2 catalysts. Conditions: 13 µmol catalyst, 0.5 ml (13.25 mmol) formic acid, 1.0 ml (5.35 mmol) BMIM OAc.
NMR is used to understand reaction mechanisms. Since there is a very significant difference in 2JH-P between hydride and phosphine ligands, the focus of this study is on the hydride peak. For Ru-1, a typical dt pattern of the hydrogenation unit was found during the first 60 minutes of dehydrogenation. Although there is a significant downfield shift from −16.29 to −13.35 ppm, its coupling constants with phosphine are 27.2 and 18.4 Hz, respectively (Figure 5, Peak A). This is consistent with all three phosphines in which the hydrogen ligand is in the cis configuration and suggests that the ligand configuration is somewhat stable in the IL for approximately one hour under optimized reaction conditions. The strong downfield shift may be due to the elimination of chlorinated ligands and the formation of the corresponding acetyl-formic acid complexes, the in situ formation of the d3-MeCN complex in the NMR tube, or the formation of the corresponding N-heterocycles. explained. Carbene (NHC) complex. During the dehydrogenation reaction, the intensity of this signal continued to decrease, and after 180 minutes the signal was no longer observed. Instead, two new signals were discovered. The first shows a clear dd pattern occurring at -6.4 ppm (peak B). The doublet has a large coupling constant of about 130.4 Hz, indicating that one of the phosphine units has moved relative to the hydrogen. This may mean that the POP clamp is rearranged into a κ2-P,P configuration. The appearance of this complex late in catalysis may indicate that this species leads to deactivation pathways over time, forming a catalyst sink. On the other hand, the low chemical shift suggests that it may be a dihydrogenous species15. The second new peak is located at -17.5 ppm. Although its fold is unknown, we believe it is a triplet with a small coupling constant of 17.3 Hz, indicating that the hydrogen ligand binds only to the phosphine ligand of the POP clamp, also indicating the release of triphenylphosphine (peak C). It can be replaced by another ligand, such as an acetyl group or an NHC formed in situ from the ionic liquid. Dissociation of PPh3 is further indicated by a strong singlet at -5.9 ppm. in the 31P{1H} spectrum of Ru-1 after 180 minutes at 90 °C (see additional information).
Hydride region of the 1H NMR spectrum of Ru-1 during the dehydrogenation of formic acid. Reaction conditions: 0.5 ml formic acid, 1.0 ml BMIM OAc, 13.0 µmol catalyst, 90 °C. NMR was taken from MeCN-d 3 , 500 μl of deuterated solvent, approximately 10 μl of the reaction mixture.
To further confirm the presence of active species in the catalytic system, high resolution mass spectrometry (HRMS) analysis of Ru-1 was performed after injection of formic acid for 10 min at 90 °C. This suggests the presence of species devoid of the chlorine ligand precatalyst in the reaction mixture. as well as two NHC complexes, the putative structures of which are shown in Figure 6. The corresponding HRMS spectrum can be seen in Supplementary Figure 7.
Based on these data, we propose an intrasphere reaction mechanism similar to that proposed by Beller, in which N-methylated PNP clamps catalyze the same reaction. Additional experiments excluding ionic liquids did not show any activity, so its direct involvement seems necessary. We hypothesize that activation of Ru-1 and Ru-2 occurs through chloride dissociation followed by possible NHC addition and triphenylphosphine dissociation (Scheme 1a). This activation in all species has previously been observed using HRMS. IL-acetate is a stronger Bronsted base than formic acid and can strongly deprotonate the latter35. We hypothesize that during the catalytic cycle (Scheme 1b), active species A bearing NHC or PPh3 are coordinated through formate to form species B. Reconfiguration of this complex to C ultimately results in the release of CO2 and the trans-dihydrogen complex D The subsequent protonation of the acid into a dihydro complex with the previously formed acetic acid to form dihydro complex E is similar to the key step proposed by Beller using N-methylated PNP clamp homologues. In addition, an analogue of the complex E L = PPh3 was previously synthesized by a stoichiometric reaction using Ru-1 in a hydrogen atmosphere after extraction of chloride with sodium salt. Removal of hydrogen and coordination of formate provides A and completes the cycle.
A mechanism for the intrasphere reaction of formic acid dehydrogenation using the fixing complex Ru-POP Ru-1 is proposed.
A new complex [RuHCl(iPr-dbfphos)(PPh3)] has been synthesized. The complex was characterized by NMR, ATRIR, EA and X-ray diffraction analysis of single crystals. We also report the first successful application of Ru-POP pincer complexes in the dehydrogenation of formic acid to CO2 and H2. Although the metal precursor achieved similar activity (up to 3424 h-1), the complex reached a maximum turnover frequency of up to 4525 h-1 at 90 °C. Moreover, at 90 °C, the new complex [RuHCl(iPr-dbfphos)(PPh3)] achieved a total time of flight (1009 h-1) to complete formic acid dehydrogenation, which is significantly higher than that of the metal precursor (330 h-1). and the previously reported complex [RuHCl(xantphos)(PPh3)] (333 h-1). Under similar conditions, the catalytic efficiency is comparable to that of the Ru-PNP clamp complex. HRMS data indicate the presence of a carbene complex in the reaction mixture, although in small quantities. We are currently studying the catalytic effects of carbene complexes.
All data obtained or analyzed during this study are included in this published article [and supporting information files].
Azarpour A., Suhaimi S., Zahedi G. and Bahadori A. A review of the shortcomings of renewable energy sources as a promising source of future energy. Arab. J. Science. engineer. 38, 317–328 (2013).
Moriarty P. and Honery D. What is the global potential for renewable energy? update. support. Energy Rev 16, 244–252 (2012).
Rao, P. C. and Yoon, M. Potential liquid organic hydrogen carrier (Lohc) systems: a review of recent advances. Energy 13, 6040 (2020).
Niermann, M., Beckendorf, A., Kaltschmitt, M. and Bohnhoff, K. Liquid organic hydrogen carriers (LOHC) – evaluation based on chemical and economic properties. internationality. J. Hydrogen energy. 44, 6631–6654 (2019).
Teichmann, D., Arlt, W., Wasserscheid, P. and Freimann, R. Future energy sources based on liquid organic hydrogen carriers (LOHC). energy environment. science. 4, 2767–2773 (2011).
Niermann, M., Timmerberg, S., Drunert, S. and Kaltschmitt, M. Liquid organic hydrogen carriers and alternatives for the international transport of renewable hydrogen. update. support. Energy Ed. 135, 110171 (2021).
Rong Y. et al. International technical and economic analysis of hydrogen storage and transportation from a hydrogen production plant to a hydrogenation terminal station. J. Hydrogen energy. 1–12 https://doi.org/10.1016/j.ijhydene.2023.01.187 (2023).
Guo, J. et al. Formic acid as a potential on-board hydrogen storage method: development of homogeneous noble metal catalysts for dehydrogenation reactions. Seuss Chemistry Chemistry. 14, 2655–2681 (2021).
Muller, K., Brooks, K., and Autry, T. Hydrogen storage in formic acid: a comparison of process options. Energy fuel. 31, 12603–12611 (2017).
Wang, Z., Lu, S.M., Li, J., Wang, J. and Li, Q. Iridium complex with N,N’-diimine ligand has unprecedented high formic acid dehydrogenation activity in water. Chemical. – EURO. J. 21, 12592–12595 (2015).
Hong D. et al. Synergistic effect of heterobinuclear IrIII-MII complexes on the catalytic release of H2 during the dehydrogenation of formic acid in water. Inorganic matter. Chemical. 59, 11976–11985 (2020).
Fink K., Laurenci G.A. and a precious catalyst for the rhodium-catalyzed dehydrogenation of formic acid in water. EURO. J.Inorg. Chemical. 2381–2387 (2019).
Seraj, JJA, et al. An efficient catalyst for the dehydrogenation of pure formic acid. Nat. communicate. 7, 11308 (2016).
Piccirelli L. et al. Multifunctional catalysis of CO2 hydrogenation-dehydrogenation using the Ru-PNP/ionic liquid system. J. Am. Bitch. 145, 5655–5663 (2023).
Belinsky E.A. et al. Dehydrogenation of formic acid with Lewis acid using an iron catalyst on a Pinzer support. J. Am. Bitch. 136, 10234–10237 (2014).
Henriks V., Juranov I., Autissier N. and Laurenci G. Dehydrogenation of formic acid on homogeneous Ru-TPPTS catalysts: unwanted CO formation and its successful removal with PROX. catalyst. 7, 348 (2017).
Filonenko G.A. etc. Efficient and reversible hydrogenation of carbon dioxide into formate using the ruthenium catalyst PNP-Pinzer. Chemistry Cat chemistry. 6, 1526–1530 (2014).
Hull, J. et al. Reversible hydrogen storage using carbon dioxide and proton-switched iridium catalysts in aqueous media at moderate temperatures and pressures. Nat. Chemical. 4, 383–388 (2012).
Wei, D. et al. The Mn-Pincer complex is used for the reversible hydrogenation of carbon dioxide to formic acid in the presence of lysine. Nat. vitality. 7, 438–447 (2022).
Piccirilli L., Pinheiro D.L. and Nielsen M. Pincer Recent advances in transition metal catalysts for sustainable development. catalyst. 10, 773 (2020).
Wei, D., Junge, H. and Beller, M. Amino acid systems for carbon dioxide capture and catalytic use for formate production. Chemical. science. 12, 6020–6024 (2021).
Subramanian M. et al. General and selective homogeneous ruthenium transfer hydrogenation, deuteration and methylation of functional compounds with methanol. J. Cutler. 425, 386–405 (2023).
Ni Z., Padilla R., Pramanik R., Jorgensen M.S.B. and Nielsen M. Base-free and acceptor-free dehydrogenating coupling of ethanol to ethyl acetate using PNP complexes. Dalton’s span. 52, 8193–8197 (2023).
Fu, S., Shao, Z., Wang, Y., and Liu, Q. Manganese-catalyzed upgrading of ethanol to 1-butanol. J. Am. Bitch. 139, 11941–11948 (2017).