Hexacoordinated tin complexes catalyse imine hydrogenation with H₂

Abstract

Frustrated Lewis pair (FLP) hydrogenation catalysts predominantly use alkyl- and aryl-substituted Lewis acids (LA) that offer a limited number of combinations of substituents, limiting our ability to tune their properties and, ultimately, their reactivity. Nevertheless, main-group complexes have numerous ligands available for such purposes, which could enable us to broaden the range of FLP catalysis. Supporting this hypothesis, we demonstrate here that hexacoordinated tin complexes with Schiff base ligands catalyse imine hydrogenation via activation of H_2(g). As shown by hydrogen–deuterium scrambling, [Sn(^tBu₂Salen)(OTf)₂] activated H_2(g) at 25 °C and 10 bar of H₂. After tuning the ligands, we found that [Sn(Salen)Cl₂] was the most efficient imine hydrogenation catalyst despite having the lowest activity in H_2(g) activation. Moreover, various imines were hydrogenated in yields up to 98% thereby opening up opportunities for developing novel FLP hydrogenation catalysts based on hexacoordinated LA of main-group elements.

---

Frustrated Lewis pairs (FLPs)¹ combining bulky Lewis acids (LAs) with Lewis bases (LBs) catalyse imine hydrogenation via H₂ activation (Fig. 1A).^2–4 Tuning their electronic and steric properties improves the FLP hydrogenation activity, expands the substrate scope^5,6 and, in some cases, imparts water tolerance.^7–10 Notwithstanding these outcomes, tuning predominantly involves triaryl-substituted Lewis acids^8,11,12 of boron, aluminium, gallium and indium¹³ with a narrow margin of manoeuvre, preventing us from further enhancing their reactivity. As a case in point, some authors argue that we cannot electronically tune the BAr₃ LAs any further.⁷

Fig. 1 (A) Reaction scheme of FLP-catalysed imine hydrogenation, (B) previously reported Sn-based LAs used in FLP-catalysed hydrogenations, and (C) L₄SnX₂ catalysts developed for H₂ activation and imine hydrogenations in this study.

FLP reactivity has nevertheless been further modified by using group 15 and 14 LAs.^14–21 In particular, R₃SnX (R = alkyl or aryl, X = halogen, ⁻OTf, ⁻NTf₂, and ⁻ClO₄) Lewis acids (Fig. 1B) in their cationic form are isolobal to group 13-based LAs,¹⁵ so they can act as their direct substitutes. In fact, tin-based FLPs effectively hydrogenate many functional groups and small molecules,^{14–16,22–25} but lack the diversity of metal complexes.

In addition to simple tetravalent alkyl- and aryl-substituted LAs, tin(IV) also forms hexacoordinated complexes with various ligands, including bipyridine,²⁶ benzoylpyridine²⁷ and Schiff bases.²⁸ In particular, Schiff base ligands can stabilize various tin oxidation states, and their complexes are known LA catalysts.^28,29 Moreover, they have been used in asymmetric catalysis,^30,31 which is currently a prominent target of FLP chemistry.^32,33 Despite the fact that a few main group complexes act as hydrogenation catalysts,^34–36 we hypothesized that hexacoordinated tin complexes with Schiff base ligands may be applied as LAs in FLPs to catalyse imine hydrogenation.

In this study, we report that hexacoordinated complexes of tin with Schiff base ligands in the form L₄SnX₂, containing N and O donor atoms (Fig. 1C), activate H₂ gas and act as hydrogenation catalysts for imine reduction. The presence of labile or hemi-labile axial ligands X⁻, such as triflate (OTf⁻) and chloride, promotes the formation of a vacant site on the LA metal centre, which is necessary for efficient catalysis.

Given the large positive polarization of tin(IV) and their labile triflate ligand(s), [Sn(^tBu₂Salen)(OTf)₂] (1-(OTf)₂) and [Sn(^tBu₂Salen)Cl(OTf)] (1-Cl(OTf)) showed high Lewis acidity, assessed using the Guttmann–Beckett (GB) method (AN = 83.6 and AN = 71.8, respectively), where 1-(OTf)₂ presumably dissociates both triflate ligands and 1-Cl(OTf) dissociates the triflate but retains the chloride to form the Lewis acidic ions [Sn(^tBu₂Salen)]₂⁺ and [Sn(^tBu₂Salen)Cl]⁺, respectively. Their Lewis acidity is similar to that of B(C₆F₅)₃ (AN = 78.1). B(C₆F₅)₃ is frequently used in FLP hydrogenations and is known to activate H₂ with bases as weak as THF or dioxane.^9,10,37,38 Calculation of hydride affinities (HIA) (Table 1, entries 1 and 2), following a protocol described by Greb et al., also indicates that HIA is much higher than B(C₆F₅)₃ (HIA = 481 kJ mol⁻¹).³⁹ Accordingly, 1-(OTf)₂ and 1-Cl(OTf) may activate H₂ together with Lewis bases (LBs) comparable to FLPs based on B(C₆F₅)₃ if a favourably oriented encounter complex forms.

Table 1 Guttmann–Beckett Lewis acidity measurement and hydride affinities of the tested complexes and their ions

Entry	Lewis acid	AN^a	Ionic form^b	HIA (kJ mol^−1)^c
a Acceptor number. b Probable ionic fragments used for the calculation of hydride affinities assume dissociation of at least one ligand and all weakly coordinating anionic ligands. c Hydride ion affinities (HIA) were calculated at the DSD-PBEB86-D3BJ/def2-QZVP level.
1	1-OTf2	83.6	[Sn(^tBu2Salen)]^2+	1170
2	1-Cl(OTf)	71.8	[Sn(^tBu2Salen)Cl]^+	660
3	1-Cl2	0	[Sn(^tBu2Salen)Cl]^+	660
4	2-Cl2	0	[Sn(Salen)Cl]^+	688
5	3-Cl2	0	[Sn(Salophen)Cl]^+	701
6	4-Cl2	0	[Sn(^tBu2Salophen)Cl]^+	682

---

As expected, 1-(OTf)₂ and 1-Cl(OTf) activated H₂ gas in the presence of THF, as shown by hydrogen–deuterium scrambling to H₂ and D₂ at 10 bar and at 25 and 60 °C, respectively (Fig. 2). Conversely, the dichloride complex [Sn(^tBu₂Salen)Cl₂] (1-Cl₂) does not possess a free binding site, at least at 25 °C, and in effect has the measured AN = 0 and the hydride affinity of 660 kJ mol⁻¹ (Table 1, entry 3) remains masked by the Cl⁻ ligands that must dissociate to reveal the cationic LA site. In line with Cl⁻ coordination and the lack of binding site at low temperatures, 1-Cl₂ failed to activate H₂, even with additional base, 2,4,6-collidine, or DABCO, at temperatures between 25 and 60 °C. These results indicate that H₂ activation requires a free binding site via a labile ligand.

Fig. 2 ²D NMR spectra of HD scrambling by 1-(OTf)₂ in THF-d₈ at 25 °C and 10 bar of HD. The blue trace is T = 0 h, and the red trace is T = 19 h. The peak at 4.45 ppm corresponds to D₂ gas, suggesting HD scrambling. Comparable spectra are obtained with 1-Cl(OTf) (ESI†).

Under the reaction conditions used for H₂ activation, however, neither of the complexes 1-(OTf)₂ and 1-Cl(OTf)₂ displayed any hydrogenation activity, suggesting that hydride transfer could be hindering the desired catalytic reactivity in line with the high HIA of 1170 and 660 kJ mol⁻¹ respectively and associated poor hydride donor ability. Catalytic hydrogenation of the FLP model substrate, N-tert-butyl-1-phenylmethanimine, was observed only at 180 °C and 50 bar of H₂ in sulfolane with 1-OTf₂, 1-Cl(OTf) and even 1-Cl₂, in 10, 49 and 29% yield, respectively (Table 2, entries 1–3). These findings suggest that chloride becomes a sufficient leaving group for H₂ activation at high temperatures, albeit less so than triflate.

Table 2 Optimization table for N-tert-butyl-1-phenylmethanimine reduction by L₄SnX₂ complexes with Schiff base ligands

Entry	Catalyst	Solvent	Temperature (°C)	Yield (%)
Reaction conditions: N-tert-butyl-1-phenylmethanimine (1 mmol), reaction solvent (4 mL), catalyst (0.05 mmol), H2 (50 bar), 17 h. All reactions were performed in triplicate, quantifying the reaction product by ^1H NMR with CH2Br2 as the internal standard and confirming the structure by ESI-MS.
1	1-(OTf)2	Sulfolane	180	10
2	1-Cl(OTf)	Sulfolane	180	49
3	1-Cl2	Sulfolane	180	29
4	2-Cl2	Sulfolane	180	53
5	3-Cl2	Sulfolane	180	42
6	4-Cl2	Sulfolane	180	28
7	1-(OTf)2	Toluene	180	30
8	1-Cl(OTf)	Toluene	180	86
9	1-Cl2	Toluene	180	84
10	2-Cl2	Toluene	180	98
11	1-Cl2	Toluene	150	NR
12	2-Cl2	Toluene	150	27
13	2-Cl2	Collidine	180	85

---

Replacing the bulky -^tBu groups with -H on 1-Cl₂ to form [Sn(Salen)Cl₂] (2-Cl₂) improved the product yield from 29 to 53% (Table 2, entries 3 and 4), implicating steric hindrance in the low hydrogenation activity of the complexes. This hypothesis was tested with [Sn(Salophen)Cl₂] (3-Cl₂) and [Sn(^tBu₂Salophen)Cl₂] (4-Cl₂), which yielded the desired product in 42 and 28% yield, respectively (Table 2, entries 5 and 6), thus confirming that –^tBu groups on complexes 1-Cl₂, and 4-Cl₂ hinder substrate access to the metal centre. Moreover, the –^tBu-substituted complexes 1-Cl₂ and 4-Cl₂ had similar yields of the target product, at 29 and 28%, respectively (Table 2, entries 3 and 6) and the HIA of the best catalyst 2-Cl₂ without the –^tBu groups and the worst 4-Cl₂ with –^tBu groups are almost identical (Table 1, entries 4 and 6), further demonstrating that steric hindrance of –^tBu groups is the limiting factor of the activity of chloride complexes. Attempts to calculate an optimized FLP structure were only successful for 2-Cl₂ (Fig. 4B) as the –^tBu groups on 1-Cl₂ and 4-Cl₂ prevented FLP formation.

Substituting sulfolane for toluene improved the catalytic performance of 1-OTf₂, 1-Cl(OTf) and 1-Cl₂ reaching 30, 86 and 84% yields of the desired amine, respectively (Table 2, entries 7–9), thus approximately doubling and trebling those obtained in sulfolane (Table 2, entries 1, 2 and 3). Among other solvent effects, this can be attributed to the enhanced solubility of H₂ gas in toluene⁴⁰ that can promote tin hydride formation via the Le Chatelier principle. As a result, 1-Cl(OTf) and 1-Cl₂ showed similar activities. The failure to further improve the yield of 1-(OTf)₂ was attributed to the instability of this catalyst under these reaction conditions. 2-Cl₂ also demonstrated improved activity in toluene and the desired product was obtained in 98% yield (Table 2, entry 10). The use of 2,4,6-collidine as the solvent, which can also act as an FLP base and enhance H₂ activation, decreased the yield from 98 to 85% (Table 2, entries 10 and 13). Lowering the temperature to 150 °C, in toluene, also decreased the yield and confirmed 2-Cl₂ as the best catalyst (Table 2, entries 11 and 12). Based on these results, we established the optimal reaction conditions, which were 180 °C, toluene, 50 bar of H₂ and 17 hours over the catalyst 2-Cl₂ (5 mol%).

Under optimal reaction conditions, 2-Cl₂ was used as the catalyst to assess the substrate scope of the reaction (Fig. 3). The model substrate N-tert-butyl-1-phenylmethanimine was converted into the corresponding amine (1) in 98% yield. Introduction of functional group(s) onto the benzene ring such as 4-chloride-, 4-trifluoromethyl- or 4,5-dimethoxy- decreased the reaction yields to 61 (2), 42 (3) and 50% (5) respectively or halted the reaction in the case of 4-nitro- (4), and 2-hydroxo- (6 and 7). As shown by mass spectrometry, nitro- and hydroxy-substitution completely inhibited imine reduction possibly due to the slow but preferential –NO₂ group reduction to –NH₂ and to the relative Brønsted acidity of phenol(s), respectively. Phenols can protonate the hydride formed in the reaction and hence reverse H₂ activation. Simultaneous reduction of an alkene was also observed and N-(tert-butyl)-N-cinnamylamine (13) was obtained in only 10% yield, while the rest was further hydrogenated to remove the alkene moiety.

Fig. 3 Substrate scope of the imine hydrogenation reaction with H₂ over 2-Cl₂. Reaction conditions: imine (1 mmol), toluene (4 mL) and LA (5 mol%), H₂ (50 bars), 180 °C, 17 h. All yields were determined by ¹H NMR with CH₂Br₂ as the internal standard, and all structures were confirmed by ESI-MS. (a) 70 bar (b) Extended reaction time to 48 h.

Fig. 4 (A) X-ray structure of 2-Cl₂ and (B) calculated FLP structure at the DSD-PBEB86-D3BJ/def2-QZVP level.

Substitution of the –^tBu group for –Hex, Cy, –Ph, –Bn or other unfunctionalized aliphatic or aromatic hydrocarbon substituents resulted in the corresponding amine formation in 6 to 82% yield (8 to 17). Noteworthy is the tolerance of ortho-methyl substitution, which yielded the corresponding amine 11 in 82% yield, whereas ortho-diisopropyl inhibited the reactivity and the corresponding amine 17 was obtained in 6% yield.

Nevertheless, further reaction trends are difficult to establish and analysis of side reactions, reaction mass balance and catalyst stability indicate that the desired reaction yield is a balance between the rate of substrate hydrogenation, its decomposition and catalyst deactivation, which decomposes to an inactive mixture of metallic tin, tin oxides and ligand fragments under the reaction conditions.

In toluene, the reaction substrate also acts as the Lewis base to activate H₂ because the tin(IV) complexes lack this ability on their own. Furthermore, a ligand must also dissociate from the coordinatively saturated complexes (Fig. 4A) to generate a Lewis acidic site, and generate an active FLP catalyst (Fig. 4B) as shown in our H₂ activation studies (Fig. 2). Based on these observations and on previous literature,^4,41–44 we propose a catalytic cycle for hydrogenation over hexacoordinate tin(IV) complexes (Scheme 1).

Scheme 1 Catalytic cycle proposed for imine hydrogenation with H₂ catalysed by tin(IV)-Schiff base complexes.

In the initial phase (Scheme 1, I), a ligand X⁻ (Cl⁻ or OTf⁻) dissociates from the complex to reveal an LA site. Interaction with the imine then forms the active FLP catalyst (II) and splits H₂ yielding L₄SnHX and [imineH][X] (III). The protonation activates the imine towards hydride transfer from the metal centre to the iminium double bond (IV).⁴³ The produced amine reversibly binds the tin(IV) complex, which slows down the reaction as demonstrated by the addition of the product (1 mmol) to the reaction, which decreased the conversion of the starting imine from 85% to 30% in a 6-hour reaction. Eventual dissociation of the produced amine then regenerates the active LA with a free binding site (V).

In conclusion, hexacoordinated tin(IV) complexes with Schiff base and labile or hemi-labile axial ligands can be used as LA components of FLPs for H₂ activation and imine hydrogenation. H₂ activation takes place with bases as weak as THF at 25 °C and can be reversible, as shown by HD scrambling. Nevertheless, imine reduction only occurs at temperatures ≥ 150 °C and optimally at 180 °C, suggesting that hydride transfer hinders the reaction. The best catalyst is [Sn(Salen)Cl₂] (2-Cl₂) even though dichloride complexes have the lowest ability to activate H₂ among all complexes tested in this study. In effect, various imines are hydrogenated in yields ranging from 6 to 98% in toluene depending on substrate substitution and functionalization. The high variability of these complexes opens up opportunities for developing hydrogenation catalysts based on hexacoordinated compounds of main-group elements.

The authors thank the Czech Science Foundation (GAČR 21-27431M) and Charles University Research Centre program No. UNCE/24/SCI/010 for funding the study. We also thank Carlos V. Melo for editing the manuscript. The research used computational resources of the Centro de Computación de Alto Desempeño – CCAD of Universidad Nacional de Cordoba – UNC (https://ccad.unc.edu.ar/), part of Sistema Nacional de Computación de Alto Desempeño – SNCAD of the Ministry of Science, Technology and Innovation, Argentina.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. G. C. Welch, R. R. San Juan, J. D. Masuda and D. W. Stephan, Science, 2006, 314, 1124–1126.
2. D. W. Stephan and G. Erker, Angew. Chem., Int. Ed., 2010, 49, 46–76.
3. D. W. Stephan, Acc. Chem. Res., 2015, 48, 306–316.
4. D. J. Scott, M. J. Fuchter and A. E. Ashley, Chem. Soc. Rev., 2017, 46, 5689–5700.
5. D. W. Stephan, Science, 2016, 354, aaf7229–aaf7229.
6. J. Lam, K. M. Szkop, E. Mosaferi and D. W. Stephan, Chem. Soc. Rev., 2019, 48, 3592–3612.
7. É. Dorkó, B. Kótai, I. Pápai, T. Soós, M. Szabó and A. Domján, Angew. Chem., Int. Ed., 2017, 56, 9512–9516.
8. D. J. Scott, T. R. Simmons, E. J. Lawrence, G. G. Wildgoose, M. J. Fuchter and A. E. Ashley, ACS Catal., 2015, 5, 5540–5544.
9. T. Mahdi and D. W. Stephan, J. Am. Chem. Soc., 2014, 136, 15809–15812.
10. D. J. Scott, M. J. Fuchter and A. E. Ashley, J. Am. Chem. Soc., 2014, 136, 15813–15816.
11. V. B. Saptal, G. Juneja and B. M. Bhanage, New J. Chem., 2018, 42, 15847–15851.
12. V. Fasano and M. J. Ingleson, Chem. – Eur. J., 2017, 23, 2217–2224.
13. M. Xu, J. Possart, A. E. Waked, J. Roy, W. Uhl and D. W. Stephan, Philos. Trans. Royal Soc. A, 2017, 375, 20170014.
14. A. Paparakis, R. C. Turnell-Ritson, J. S. Sapsford, A. E. Ashley and M. Hulla, Catal. Sci. Technol., 2023, 13, 637–644.
15. D. J. Scott, N. A. Phillips, J. S. Sapsford, A. C. Deacy, M. J. Fuchter and A. E. Ashley, Angew. Chem., Int. Ed., 2016, 55, 14738–14742.
16. J. Sapsford, D. Scott, N. Allcock, M. Fuchter, A. Ashley and C. Tighe, Adv. Synth. Catal., 2018, 360, 1066–1071.
17. P. Sarkar, S. Das and S. K. Pati, Chem. – Asian J., 2022, 17, e202200148.
18. T. Thorwart, D. Hartmann and L. Greb, Chem. – Eur. J., 2022, 28, e202202273.
19. T. A. Kinder, R. Pior, S. Blomeyer, B. Neumann, H. G. Stammler and N. W. Mitzel, Chem. – Eur. J., 2019, 25, 5899–5903.
20. P. Holtkamp, F. Friedrich, E. Stratmann, A. Mix, B. Neumann, H. G. Stammler and N. W. Mitzel, Angew. Chem., Int. Ed., 2019, 58, 5114–5118.
21. J. M. Bayne and D. W. Stephan, Chem. Soc. Rev., 2016, 45, 765–774.
22. R. T. Cooper, J. S. Sapsford, R. C. Turnell-Ritson, D. H. Hyon, A. J. P. White and A. E. Ashley, Philos. Trans. Royal Soc. A, 2017, 375, 20170008.
23. G. R. Whittell, E. I. Balmond, A. P. M. Robertson, S. K. Patra, M. F. Haddow and I. Manners, Eur. J. Inorg. Chem., 2010, 3967–3975.
24. J. S. Sapsford, D. Csókás, R. C. Turnell-Ritson, L. A. Parkin, A. D. Crawford, I. Pápai and A. E. Ashley, ACS Catal., 2021, 11, 9143–9150.
25. A. Paparakis and M. Hulla, ChemCatChem, 2023, 15, e202300510.
26. Y. M. Ahmed and G. G. Mohamed, Inorg. Chem. Commun., 2022, 144, 109864.
27. A. Pérez-Rebolledo, G. M. de Lima, N. L. Speziali, O. E. Piro, E. E. Castellano, J. D. Ardisson and H. Beraldo, J. Organomet. Chem., 2006, 691, 3919–3930.
28. H. Jing, S. K. Edulji, J. M. Gibbs, C. L. Stern, H. Zhou and S. B. T. Nguyen, Inorg. Chem., 2004, 43, 4315–4327.
29. A. M. Abu-Dief and I. M. A. Mohamed, J. Basic Appl. Sci., 2015, 4, 119–133.
30. M. Palucki, P. J. Pospisil, W. Zhang and E. N. Jacobsen, J. Am. Chem. Soc., 1994, 116, 9333–9334.
31. S. De, A. Jain and P. Barman, ChemistrySelect, 2022, 7, e202104334.
32. W. Meng, X. Feng and H. Du, Acc. Chem. Res., 2018, 51, 191–201.
33. W. Meng, X. Feng and H. Du, Chin. J. Chem., 2020, 38, 625–634.
34. Y. Liang, J. Luo, Y. Diskin-Posner and D. Milstein, J. Am. Chem. Soc., 2023, 145, 9164–9175.
35. H. Elsen, C. Färber, G. Ballmann and S. Harder, Angew. Chem., Int. Ed., 2018, 57, 7156–7160.
36. A. Friedrich, J. Eyselein, H. Elsen, J. Langer, J. Pahl, M. Wiesinger and S. Harder, Chem. – Eur. J., 2021, 27, 7756–7763.
37. D. J. Scott, M. J. Fuchter and A. E. Ashley, Angew. Chem., Int. Ed., 2014, 53, 10218–10222.
38. L. J. Hounjet, C. Bannwarth, C. N. Garon, C. B. Caputo, S. Grimme and D. W. Stephan, Angew. Chem., Int. Ed., 2013, 52, 7492–7495.
39. E. Philipp and L. Greb, ChemPhysChem., 2021, 22, 935–943.
40. J. J. Simnick, H. M. Sebastian, H.-M. Lin and K.-C. Chao, J. Chem. Eng. Data, 1978, 23, 339.
41. P. Pérez, D. Yepes, P. Jaque, E. Chamorro, L. R. Domingo, R. S. Rojas and A. Toro-Labbé, Phys. Chem. Chem., 2015, 17, 10715–10725.
42. S. Grimme, H. Kruse, L. Goerigk, G. Erker, S. Grimme, H. Kruse, L. Goerigk and G. Erker, Angew. Chem., Int. Ed., 2010, 49, 1402–1405.
43. J. Paradies, Eur. J. Org. Chem., 2019, 283–294.
44. P. A. Chase, T. Jurca and D. W. Stephan, Chem. Commun., 2008, 1701–1703.

---

Footnote

† Electronic supplementary information (ESI) available: Catalyst synthesis and analysis; NMRs, MS, IRs, sXRD, and Lewis acidity measurements. CCDC 2330171 and 2333521. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d3cc05878f

---