Mild Reduction of Carbon Dioxide to Methane with Tertiary Silanes Catalyzed by Platinum and Palladium Silyl Pincer Complexes

Pincing CO2: Group 10 (Pd, Pt) silyl pincer complexes in combination with B(C6F5)3 can facilitate the efficient catalytic conversion of CO2 to methane with a tertiary silane as the reductant under mild conditions. The reduction involves the formation of Pd and Pt formatoborate complexes. The large-scale combustion of fossil fuels (coal, petroleum, natural gas) has led to a significant and alarming rise in levels of atmospheric CO2 over the past several decades, and this trend is anticipated to continue.1 As a greenhouse gas, CO2 is a significant contributor to global warming, and thus the current high level of anthropogenic CO2 in the atmosphere is of great concern.1 Although efforts to capture and sequester CO2 are being pursued, such efforts have so far proven relatively costly and remain in their infancy.2 A complementary strategy for addressing both the high levels of atmospheric CO2 and the dwindling supply of fossil fuels available to meet our energy needs is the recycling of CO2 by conversion into a hydrocarbon fuel that is suitable for use in our current energy infrastructure. As such, there is significant interest in the development of efficient methodologies for the reduction of CO2 to methanol and/or methane, with the ultimate goal of achieving a carbon-neutral catalytic process.1c, 3, 4 In this context, the pursuit of homogeneous catalysts for the conversion of CO2 to methanol and/or methane is an area of growing interest,4–8 and to date, only a handful of such catalyst systems have been developed.9–15 With respect to methanol formation, Milstein and co-workers recently demonstrated the facile hydrogenation of CO2-derived organic carbonates, carbamates, and formates to methanol utilizing PNN pincer Ru complexes as catalysts.10 Building on this pioneering work, Sanford and co-workers reported a three component cascade catalysis route for the hydrogenation of CO2 to form methanol,9d and Leitner and co-workers demonstrated that a single Ru phosphine catalyst is also effective in this transformation.9e There has also been significant interest in the development of efficient reduction pathways that do not rely on hydrogen for the reduction of CO2 to the methoxide level. Utilizing a POCOP pincer NiII hydride catalyst, Guan and co-workers demonstrated the hydroboration of CO2 with catecholborane to give methoxyboryl species with a relatively high turnover frequency.12 Furthermore, although the Ir-catalyzed hydrosilylation of CO2 to form methoxysilane species was initially reported over twenty years ago, albeit with low efficiency,13a more recently Ying and co-workers reported a metal-free N-heterocyclic carbene catalyzed process for the reduction of CO2 with diphenylsilane to form the corresponding methoxysilane with significantly improved turnover numbers and turnover frequencies.13b, 13c Focusing on the complete reduction of CO2 to methane, prior to our commencing work in this area, only three examples of such homogeneous catalyst systems had been reported.14, 15 Matsuo and Kawaguchi disclosed the use of bis(phenoxo) Zr alkyl complexes, which in conjunction with the highly Lewis acidic borane B(C6F5)3, catalyzed the reaction of CO2 with hydrosilanes to form methane.14a Piers and co-workers subsequently disclosed a catalyst system based on the frustrated Lewis pair (FLP) 2,2,6,6-tetramethylpiperidine (TMP)/B(C6F5)3, which reacts with CO2 and Et3SiH to form a formatoborate species that is subsequently hydrosilylated in the presence of excess B(C6F5)3 to form methane.14b Lastly, Wehmschulte and co-workers recently reported that the highly reactive Lewis acid Et2Al+ catalyzes the conversion of CO2 to methane as well as other hydrocarbons upon reaction with hydrosilanes, albeit at elevated temperatures and long reaction times.14c In an effort to diversify the catalyst portfolio available for the reduction of CO2 to methane, we became interested in identifying a complementary late metal catalyst system for this transformation. We envisioned that a platinum group metal catalyst might offer increased stability and decreased sensitivity to protic impurities relative to a d0 metal alkyl or a cationic Group 13 alkyl species, while still achieving suitable activity in CO2 reduction chemistry. We were particularly encouraged by the utility of such late metal species in the reduction of CO2 to methanol, and considered that the identification of an appropriately configured late metal catalyst could provide an entry point towards complementary processes leading to formation of methane. Herein we report our results detailing the utility of soluble, well-defined Group 10 (Pd, Pt) silyl pincer complexes, in combination with B(C6F5)3, for the efficient catalytic conversion of CO2 to methane using tertiary silanes as the reductant. While this manuscript was in preparation, Brookhart and co-workers reported a related example of a cationic [(POCOP)IrIII] species that functions as a catalyst for the conversion of CO2 to methane using hydrosilanes.15 We have previously described the synthesis, structural features and bond activation reactivity of a variety of platinum group metal pincer complexes supported by tridentate phosphinosilyl ligands of the type [κ3-(2-R2PC6H4)2SiMe]− (R-PSiP, R=alkyl, aryl),16 including the synthesis of square planar Group 10 (Ni, Pd, Pt) complexes that underwent facile cleavage of SiH, SiCl and SiC (sp2 and sp3) bonds.16c, 16e Furthermore, [(Ph-PSiP)PdII] complexes have been shown to be useful in the catalytic hydrocarboxylation of allenes and 1,3-dienes with CO2 under mild conditions as well as in the synthesis of diborylalkenes via dehydrogenative borylation.17 Efforts to prepare a PtII hydride complex of the type [(Cy-PSiP)PtH] led to the formation of [(Cy-PSi(μ-H)P)Pt] (1), which was identified on the basis of NMR and IR spectroscopic data as a bis(phosphino) Pt derivative of (Cy-PSiP)H that features η2-SiH coordination involving the tethered silicon fragment (Scheme 1).16c The Pd analogue (2) of 1 was prepared by a similar route involving treatment of [(Cy-PSiP)PdCl] with LiEt3BH (Scheme 1). Synthesis of (Cy-PSiP)Pt and Pd Complexes (only one resonance structure is shown for formatoborate complexes 5–6′). In an effort to further explore the reactivity of these unusual η2-SiH complexes, we investigated their ability to undergo hydride abstraction. Interestingly, treatment of complexes 1 and 2, respectively, with the strong Lewis acid B(C6F5)3 afforded the hydride abstracted products 3 and 4 (Scheme 1), as evidenced by the appearance of a sharp 11B NMR resonance in each case (for 3: δ=−25.4 ppm, d, 1JBH=92 Hz, [D8]THF; for 4: δ=−23.3 ppm, d, 1JBH=77 Hz, [D6]benzene) that featured BH coupling. We considered that complexes 3 and 4 offered an intriguing entry point for studying CO2 fixation due to parallels between these complexes and the FLP-derived salt [TMPH][HB(C6F5)3], which reacts with CO2 to form a formatoborate species that has been implicated in the stoichiometric hydrogenation of CO2 to methanol as well as the catalytic formation of methane upon reaction with excess B(C6F5)3 and Et3SiH.14b Treatment of a benzene solution of either complex 3 or 4 with CO2 gas (ca. 1 atm) resulted in the immediate formation of the corresponding Pt and Pd formatoborate complexes (Scheme 1; M=Pt, 5; M=Pd, 6). The 1H NMR spectra of complexes 5 and 6 each contain a diagnostic formate resonance at 8.65 and 8.49 ppm, respectively, whereas the 13C NMR spectra of these complexes feature a resonance at 170.8 and 171.7 ppm, respectively, corresponding to the formate carbon. The formatoborate complexes 5 and 6 were readily distinguished from the corresponding Pt and Pd formate species (Scheme 1; M=Pt, 7; M=Pd, 8) on the basis of their 1H, 13C and 31P NMR spectroscopic features (e.g., the 1H NMR resonance corresponding to the formate proton for 7, δ=9.81 ppm, with Pt satellites, 3JHPt=40 Hz; for 8, δ=9.20 ppm). Furthermore, unlike complex 5, which was isolated in 88 % yield, the Pt formate complex 7 was only observed under a CO2 atmosphere and reformed 1 upon removal of CO2 (Scheme 1). Although the Pd formate complex 8 proved isolable, treatment of 8 with one equivalent of B(C6F5)3 resulted in quantitative (by 1H and 31P NMR spectroscopy) conversion to 6 (Scheme 1). Similar insertion of CO2 into a hydroborate BH bond has previously been observed for main group fragments involving FLPs.11b, 14b Few related examples of formatoborate complexes involving transition metal species have been reported, and all previously reported examples appear to result exclusively from the initial formation of a metal formate species that subsequently forms a formate–borane adduct.18 Furthermore, Bercaw and Labinger have reported that, in the case of [HNi(dmpe)2]+, although trialkyl boranes facilitate the formation of formate–borane adducts upon reaction with CO2, B(C6F5)3 exhibited only hydride transfer from Ni to B and no CO2 reduction.18 Thus, it appears that the facile reduction of CO2 by 3 and 4 to form formatoborate species is unprecedented. In an effort to determine if less Lewis acidic boranes would facilitate similar stoichiometric CO2 reduction chemistry, complexes 1 and 2 were each treated with one equivalent of BPh3. Although 1 did not appear to undergo a reaction with BPh3, complex 2 reacted readily at room temperature to quantitatively (by 31P NMR spectroscopy) afford 4′, the BPh3 analogue of 4, which was isolated in 95 % yield (Scheme 1). Treatment of 4′ with CO2 (1 atm) resulted in the formation of the desired formatoborate complex 6′, which features a characteristic formate 1H NMR resonance at δ=8.85 ppm ([D6]benzene), as well as a 13C NMR resonance at δ=173.3 ppm corresponding to the formate carbon (Scheme 1). Complex 6′ was also accessed by the treatment of the Pd formate species 8 with one equivalent of BPh3. Although a BPh3 analogue of 3 was not available, we were able to access a related platinum formatoborate complex (5′) by treating a mixture of 1 and BPh3 with CO2 (Scheme 1). Complex 5′ was isolated in 90 % yield and, like 6′, features both 1H and 13C NMR resonances consistent with a formate group. We postulate that 5′ forms via the in situ formation of 7, which can subsequently react with BPh3 to form the observed formatoborate complex. Treatment of a [D6]benzene solution of either 5 or 6 with four equivalents of Me2PhSiH resulted in the quantitative (by 31P NMR spectroscopy) formation of 3 or 4, respectively, with the concomitant evolution of methane and (Me2PhSi)2O (Scheme 1), which were observed in the 1H NMR spectra of the reaction mixtures at δ=0.14 and 0.32 ppm, respectively. Having thus observed the stoichiometric reduction of CO2 to methane, we pursued a catalytic variant of this reaction, using in situ generated 3 or 4 (from treatment of 1 or 2 with one equivalent of B(C6F5)3) as the catalyst (Table 1). Using Me2PhSiH as the reductant, catalytic reactions were carried in fluorobenzene at 65 °C, 1 atm of CO2, and a catalyst loading of 0.065 mol % relative to silane. The amount of (Me2PhSi)2O produced was determined on the basis of calibrated GC data and turnover numbers (TON) were determined based on mol of silane reacted per mol of catalyst. Using 3 as the catalyst afforded 1063 turnovers after 4 h, whereas the Pd analogue at the same loading resulted in 469 turnovers with heating at 85 °C (Table 1, entries 1 and 6). In the case of the more active Pt-based catalyst 3, the turnover frequency (TOF) after 0.5 h at 65 °C was determined to be 956 h−1. As expected, the TOF was observed to decrease over the course of the reaction (TOF=616 h−1 after 1 h and 266 h−1 after 4 h). Notably, the use of BPh3 in place of B(C6F5)3 under comparable conditions afforded no catalytic turnover. Entry Catalyst Catalyst [mol %] Silane Time [h] (R3Si)2O [mmol][b] TON[c] 1 3 0.065 Me2PhSiH 4 3.4 1063[d] 2[e] 3 0.016 Me2PhSiH 8 5.7 1781 3[f] 3 0.016 Me2PhSiH 16 6.9 2156 4 3 0.065 Et3SiH 4 0.07 22 5[g] 3 0.065 Me2PhSiH 4 0 0 6[h] 4 0.065 Me2PhSiH 4 1.5 469 7[i] NaH/B(C6F5)3/15-c-5 0.4 Me2PhSiH 4 1.2 57 8 LiEt3BH/B(C6F5)3 0.1 Me2PhSiH 4 0.1 16 9 1 0.0065 Me2PhSiH 4 0 0 10 2 0.0065 Me2PhSiH 4 0 0 In an effort to assess the lifetime of the catalyst (0.016 mol % 3 relative to the starting amount of silane), we carried out experiments in which a fresh charge of CO2 (1 atm) was reintroduced to the reaction vessel after an initial run of 4 h at 65 °C. Subsequent heating for an additional 4 h at the same temperature resulted in a TON of 1781 after a total of 8 h heating (entry 2). After two additional rounds of reintroducing CO2 (16 h total heating at 65 °C), a TON of 2156 was achieved, indicating that catalyst activity does diminish with time (entry 3). The bulkier silane Et3SiH led to significantly decreased activity (entry 4). Control experiments carried out in the absence of CO2 confirmed that the observed silyl ether formation cannot be attributed to side reactions involving, for example, adventitious water or O2 (entry 5). Further control experiments confirmed the key role of both the platinum group metal complex and B(C6F5)3 in achieving the efficient catalytic reduction of CO2 (entries 7–10). Notably, the catalytic productivity (TON) and rates (TOF) exhibited by 3 are comparable in magnitude to those observed by Brookhart and co-workers in their recently reported cationic [(POCOP)IrIII] system.15 A proposed catalytic cycle for the reduction of CO2 utilizing complexes such as 3 and 4 and tertiary silanes is shown in Scheme 2. Based on the observed stoichiometric reactivity, we postulate that the abstracted hydride species 3 and 4 initially react with CO2 to form the corresponding formatoborate complexes 5 and 6, respectively, which subsequently react with two equivalents of silane to reform 3 and 4 and generate one equivalent of CH2(OSiR3)2. Evidence for the intermediacy of the bis(silyl)acetal species was observed from the treatment of complexes 5 and 6 with one equivalent of either Me2PhSiH or Et3SiH. In both cases, this reaction resulted in the complete consumption of silane and afforded about 50 % conversion (by 31P NMR spectroscopy) to the abstracted hydride species 3 and 4. The corresponding bis(silyl)acetal was also observed in the reaction mixture by 1H NMR spectroscopy [δ=5.0 ppm, s, CH2(OSiR3)]. We propose that subsequent reduction of the bis(silyl)acetal to form methane and the corresponding bis(silyl)ether is achieved by the previously reported route involving B(C6F5)3 mediated hydrosilylation.14a, 14b, 19, 20 Support for this latter series of steps is drawn from the observation that the replacement of B(C6F5)3 with the less Lewis acidic borane BPh3 does not lead to the formation of methane and bis(silyl)ether. Rather, upon the treatment of 5′ or 6′ with four equivalents of Me2PhSiH, only the formation of CH2(OSiPhMe2)2 was observed. Proposed catalytic cycle for the reduction of CO2 to CH4. In summary, we have demonstrated that Group 10 (Pd, Pt) silyl pincer complexes in combination with B(C6F5)3 can facilitate the efficient catalytic conversion of CO2 to methane with a tertiary silane as the reductant under mild conditions. The reduction involves the formation of Pd and Pt formatoborate complexes, which have been isolated and characterized. This reactivity represents a rare example of transition metal catalyzed CO2 reduction to methane. Encouraged by these initial findings, we are pursuing the development of increasingly effective pincer-based platinum group metal catalysts for this transformation. General information: All reactions were performed under nitrogen in an MBraun glove-box or by using standard Schlenk techniques. Gas chromatography was performed on a Shimadzu GC-2014 equipped with a SGE BP-5 30 m, 0.25 mm I.D. column. For conversions and yields given on the basis of GC experiments, the data were corrected by calibration with dodecane as an internal standard, and product identity was confirmed by comparison with authentic samples. Chemicals were purchased from Aldrich, Gelest, and Strem and used as received. Dry, oxygen-free solvents were used unless otherwise indicated. The compounds [(Cy-PSi(μ-H)P)Pt] (1)16c and [(Cy-PSiP)PdCl]16e were prepared according to literature procedures. Typical procedure for the catalytic reduction of CO2 with hydrosilane: All catalytic runs were conducted under a nitrogen atmosphere in resealable glass reaction cells (50 mL) containing a magnetic stir-bar and fitted with a gas-tight PTFE stopcock. Dodecane (0.50 mL) was added to a solution of 1 (0.005 g, 0.0064 mmol) and B(C6F5)3 (0.003 g, 0.0064 mmol) in C6H5F (2 mL). The solution was transferred to a reaction cell and was degassed by two freeze-pump-thaw cycles. CO2 (1 atm) was introduced. Under a purge of CO2, neat Me2PhSiH (1.50 mL, 9.78 mmol) was added to the reaction mixture via syringe. The sealed reaction cell was subsequently heated for 4 h at 65 °C. Once the reaction was complete the cell was submerged in a room temperature water bath and quickly exposed to vacuum and refilled with N2 gas. In the glove-box, the reaction mixture was filtered through a silica plug and was subsequently analyzed by use of gas chromatographic methods. We are grateful to the NSERC of Canada (including a Discovery Grant for L.T.) and Dalhousie University for their generous support of this work. We also thank Dr. Michael Lumsden (NMR3, Dalhousie) for his assistance. Detailed facts of importance to specialist readers are published as ”Supporting Information”. Such documents are peer-reviewed, but not copy-edited or typeset. They are made available as submitted by the authors. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.