Nicole A.
Torquato
a,
Joseph M.
Palasz
a,
Quentin C.
Bertrand
b,
Felix M.
Brunner
a,
Thomas
Chan
a,
Milan
Gembicky
a,
Anthony A.
Mrse
a and
Clifford P.
Kubiak
*a
aDepartment of Chemistry and Biochemistry, University of California, San Diego, 9500 Gilman Dr, La Jolla, CA 92093, USA. E-mail: ckubiak@ucsd.edu
bDepartment of Chemistry, Yale University, New HavenCT 06520USA
First published on 5th September 2022
Treatment of the trichlorotin-capped trinuclear nickel cluster, [Ni3(dppm)3(μ3-Cl)(μ3-SnCl3)], 1, with 4 eq. NaHB(Et)3 yields a μ3-SnH capped trinuclear nickel cluster, [Ni3(dppm)3(μ3-H)(μ3-SnH)], 2 [dppm = bis(diphenylphosphino)methane]. Single-crystal X-ray diffraction, nuclear magnetic resonance (NMR) spectroscopy, and computational studies together support that cluster 2 is a divalent tin hydride. Complex 2 displays a wide range of reactivity including oxidative addition of bromoethane across the Sn center. Addition of 1 eq. iodoethane to complex 2 releases H2 (g) and generates an ethyltin-capped nickel cluster with a μ3-iodide, [Ni3(dppm)3(μ3-I)(μ3-Sn(CH2CH3))], 4. Notably, insertion of alkynes into the Sn–H bond of 2 can be achieved via addition of 1 eq. 1-hexyne to generate the 1-hexen-2-yl-tin-capped nickel cluster, [Ni3(dppm)3(μ3H)(μ3-Sn(C6H11))], 5. Addition of H2 (g) to 5 regenerates the starting material, 2, and hexane. The formally 44-electron cluster 2 also displays significant redox chemistry with two reversible one-electron oxidations (E = −1.3 V, −0.8 V vs. Fc0/+) and one-electron reduction process (E = −2.7 V vs. Fc0/+) observed by cyclic voltammetry.
Vapor diffusion of diethyl ether into a THF solution of 2 at −35 °C resulted in brownish-red X-ray quality crystals. X-ray diffraction revealed that cluster 2 crystallized in the triclinic space group P and features a μ3-Sn atom capping a triangular face consisting of three nickel atoms supported by three bridging dppm ligands (Fig. 1). Although the Sn–H hydrogen atom in 2 could not be precisely located in the difference electron-density map, the Sn–H linkage was unequivocally confirmed by 1H and 119Sn NMR spectroscopy (vide infra). Density functional theory (DFT) calculations and natural bond orbital (NBO) analysis also support this structural assignment (see ESI† for details).
Compared to 1, 2 exhibits substantially shorter Sn–Ni bond distances and considerably longer Ni–Ni bond distances (Table 1). While many low-valent Sn-hydrides form dimers,3,4,92 does not, likely as a result of the trinuclear Ni3(dppm)3 framework providing considerable steric bulk, an effect seen in other monomeric Sn–H complexes.27,28 The structure of 2 shows slight variations in Ni–Sn bond distances. Additionally, the Ni1–Ni3 bond distance is significantly shortened in comparison to the other Ni–Ni bonds (Fig. 1 and Table 1). This implies that, in the solid-state, the μ3 binding of the Sn–H is not entirely symmetric, and the interactions of Sn–H may be more localized on one nickel center. Similar variations in Ni–Sn bond distances are seen in the starting material, 1.25,26 These structural features are also observed in the results of DFT calculations. DFT and natural bond order (NBO) analysis for complex 2 (B3LYP/LANL2DZ) reveal that the Sn–H interacts principally with one nickel atom (Fig. 2, see also ESI†).
Bond length [Å]/angle [°] | 1 | 2 | 3 | 4 | 5 |
---|---|---|---|---|---|
Ni1–Ni2 | 2.4593(18) | 2.7445(6) | 2.4104(9) | 2.6840(5) | 2.5011(10) |
Ni1–Ni3 | 2.4829(18) | 2.4449(5) | 2.4020(9) | 2.8886(5) | 2.5285(10) |
Ni2–Ni3 | 2.4825(16) | 2.7315(6) | 2.3858(9) | 2.8227(5) | 2.6655(11) |
Sn1–Ni1 | 2.6118(14) | 2.5187(5) | 2.7261(8) | 2.4423(4) | 2.4324(8) |
Sn1–Ni2 | 2.5960(16) | 2.5502(5) | 2.5716(7) | 2.4419(4) | 2.5282(8) |
Sn1–Ni3 | 2.6185(16) | 2.5348(5) | 2.7447(7) | 2.4607(4) | 2.4724(8) |
∠C–Sn–Ni1 | 143.57(8) | 162.02(16) | |||
∠C–Sn–Ni2 | 140.24(8) | 136.59(17) | |||
∠C–Sn–Ni3 | 132.11(8) | 125.91(16) |
Further details of the molecular and electronic structure were elucidated by 31P, 1H, and 119Sn NMR spectroscopy. The 31P NMR spectroscopy of 2 in benzene-d6 revealed symmetrical binding of the μ3-SnH and μ3-H capping groups on the NMR timescale as demonstrated by an isolated sharp singlet resonance at 23.4 ppm. The Ni–H resonance was located in the 1H NMR at −0.86 ppm and it appears as a heptet, featuring significant two-bond coupling (2JP−H) to the six phosphorus atoms coordinated to the nickel atoms. In addition, the Sn–H 1H NMR resonance was located at −0.42 ppm and it appears as a singlet. The assignment of this resonance as that belonging to the hydrogen-bound Sn was evidenced by the proton-coupled 119Sn NMR spectrum which, at room temperature, features a doublet at 2939.4 ppm (see ESI: S4†). The 1H decoupled 119Sn NMR spectrum contained a singlet resonance, indicating that there is one hydrogen atom bound to Sn, which is also further confirmed by integration of the peaks in the 1H NMR spectrum.
Satellite peaks arising from coupling of hydrogen to the 117Sn and 119Sn nuclei with 1J(Sn–H) values near 1900 Hz are normally observed for tetravalent Sn(IV) hydrides.3,29,30 In contrast, divalent Sn(II) hydride complexes typically exhibit much smaller coupling constants (1J(Sn–H)ca. 100 Hz) because in these Sn(II) hydrides most of the s-orbital electron density is localized to the lone pair on Sn.4,31–33 The 1J(117/119Sn–1H) coupling constant of 2 was found to be 360 Hz, consistent with a divalent Sn center. This assignment is corroborated by NBO analysis, which predicts that the Sn hybridization used to form the Sn–H linkage in 2 has about 20% s-character at Sn. Furthermore, DFT analysis indicates the presence of a lone pair on Sn in the HOMO−1 (Fig. 2). In addition, preliminary Mössbauer results on 2 depict a +2 oxidation state Sn center. Overall, NMR, computational studies, and preliminary Mössbauer results suggest that the Sn can be considered as sp2 hybridized with most of the s-orbital density localized on a lone pair on Sn.
To support experimentally the assignment of 2 as a Sn(II) hydride with a lone pair, we performed reactivity studies. Alkyl halides are known to undergo oxidative addition to Sn(II) complexes that feature a lone pair.34 Addition of 1 eq. BrCH2CH3 to 2 in THF at −20 °C, results in the oxidative addition of BrCH2CH3 at the Sn center and formation of [Ni3(dppm)3(μ3-H)(μ3-Sn((Br)(H)(CH2CH3)))], 3 in 75% yield (Scheme 2).
The formation of 3 is characterized by a significant upfield shift in the 31P NMR spectrum compared to 2, with a resonance located at 4.2 ppm in benzene-d6. As for the 1H NMR signals, the Ni–H resonance shifts upfield to −4.70 appearing as a heptet, and the Sn–H resonance shifts downfield to 8.78 ppm, presumably a result of the close proximity to the Sn–Br bond. There is significant broadening in the 31P and 1H NMR spectra due to the break in symmetry from the three different functional groups attached to Sn. In comparison to 2, the proton-coupled 119Sn NMR of 3 depicts an upshifted doublet with a resonance located at 247.8 ppm and a 1J(Sn–H) = 993.6 Hz, suggesting a larger s orbital contribution from the Sn into the H atom bond (Table 2, also see ESI: S9†). The 1J(Sn–H) value of 3 is comparable to that reported by the Rivard group for a Sn(II) dihydride transition metal coordinated complex, IPr–SnH2–W(CO)5 (IPr = [(HCNAr′)2C:]; Ar′ = 2,6-iPr2C6H3).16
31P, δ (ppm) | Ni–H, δ (ppm) | Sn–H, δ (ppm) | 119Sn, δ (ppm) | 1 J (Sn–H) (Hz) | |
---|---|---|---|---|---|
2 | 23.4 | −0.86 | −0.42 | 2939.4 | 346.7 |
3 | 4.2 | −4.80 | 8.77 | 248.0 | 993.6 |
4 | 34.1 | 2115.1 | |||
5 | 31.0 | 3.54–3.62 | 1837.8 |
Vapor diffusion of pentane into a THF solution of 3 at −20 °C resulted in red colored X-ray quality crystals (Fig. 3). The hydride attached to Sn was not precisely located. There appears to be significant asymmetry of the μ3-SnHBr(CH2CH3) binding to the trinuclear nickel core, suggested by the significant elongation of the Sn–Ni1 and Sn–Ni3 bonds versus the Sn–Ni2 bond (Table 1). Furthermore, the data indicates an overall shortening of the Ni–Ni bond distances more closely resembling that of cluster 1 than 2 (Table 1).
Surprisingly, in an analogous reaction to forming cluster 3, treatment of 2 with 1 eq. ICH2CH3, releases H2 and generates an CH3CH2Sn-capped, nickel cluster with a μ3 iodide [Ni3(dppm)3(μ3-I)(μ3-Sn(CH2CH3))], 4, in 94% yield (Scheme 2). The formation of 4, is characterized by a significant downfield shift from the starting material in the 31P NMR to 34.1 ppm in benzene-d6, which is flanked by satellites arising from 2-bond coupling to the 119Sn nuclei (2J(Sn,P) = 98.8 Hz), while the 119Sn NMR shifts upfield to 2115.14 ppm (Table 2, see ESI: S14†).
Layering of pentane over a benzene solution of 4 at 22 °C resulted in brownish-red X-ray quality crystals (Fig. 3). The Sn–Ni bond distances and the C–Sn–Ni bond angles reveal an almost linear binding orientation of the μ3-SnCH2CH3 to the center of the trinuclear nickel core. Furthermore, the shortened Sn–Ni bond distances: 2.4423(4), 2.4419(4), and 2.4607(4), are of similar length to other transition metal coordinated stannylyne complexes in the literature, which primarily exhibit a linear arrangement unless there are considerable steric demands due to the ligand environment.35–40 We will report separately results of Mössbauer spectroscopy and computational studies to support this assignment.
Hydrostannylation is a well-known transformation in organic chemistry.41 Hydrostannylation utilizing tin(II) species has been demonstrated. These reactions feature transition metal and radical catalyst-free cis-hydrostannylation of the alkynes under mild conditions.42–44 Therefore, it was of interest to evaluate the reactivity of complex 2 with alkynes. 1 eq. of 1-hexyne was added to 2 in a benzene solution, which resulted in insertion to yield 1-hexen-2-yl-Sn-capped nickel cluster [Ni3(dppm)3(μ3-H)(μ3-Sn(C6H11))], 5, in 98% yield (Scheme 3). A study from the Power group demonstrated Sn(II) hydrostannylation of primary alkynes resulted in Sn binding to the terminal position.42 However, complex 5 exhibits a distinct preference to bind to the secondary position.
Scheme 3 Synthesis of [Ni3(dppm)3(μ3-H)(μ3-Sn)(C6H11)], 5, and hydrogenation to [Ni3(dppm)3(μ3-H)(μ3-SnH)], 2. |
The formation of 5, is characterized by an upfield shift in the 31P NMR to 31.0 ppm from the starting material, 2. In comparison to 2, an upshifted Ni–H resonance in the 1H NMR was observed at −3.54 to −3.62 ppm, while the 119Sn NMR resonance shifted upfield to 1837.8 ppm. Vapor diffusion of pentane into a diethyl ether solution of 5 at −20 °C results in brown X-ray quality crystals (Fig. 3). The μ3-Sn(C6H11) features a significant bend from a 180-degree angle to the center of the three nickel clusters, suggesting the presence of a lone pair on the Sn, similar to cluster 2 (Table 1). This is supported by variations in the Sn–Ni bond distances in complex 5, in contrast to 4. Additionally, the Sn–Ni and Ni–Ni bond distances in 5 is similar to those of complex 2 (Table 1).
After isolating complex 5, hydrogenation of the 1-hexen-2-yl fragment and regeneration of 2 was attempted. Upon addition of H2 (g) to complex 5, complex 2 and hexane are formed in 21% yield by 1H NMR. In this process, an intermediate species is formed at −4.5 ppm via31P NMR and there is a shift in the Ni–H resonance to −6.2 ppm via1H NMR, before the regeneration of complex 2 (see ESI: S19–S21†). Thus far, this intermediate has proved elusive due to the chemical instability in solution. NMR stability studies of 5 in THF-d8 at 22 °C found that 92% of the initial integration of 1H NMR features were lost over 24 hours, leading to a variety of unidentified species (see ESI: S22 and S23†). Repeat studies in benzene-d6 yielded similar conclusions. This is an unexpected result, as previous studies demonstrating tin(II) hydrostannylation of alkynes only studied the formation of stable tin alkene species and have not shown the hydrogenation chemistry of these products.42–44 The literature indicates that, Sn has primarily been utilized as a stoichiometric reagent, however, this study demonstrates a platform that may lead to catalytic application of alkyne reduction by Sn–H utilizing H2 (g).41
Finally, the oxidative addition chemistry revealed by the novel low valent μ3-SnH capped trinuclear nickel cluster 2 raises the question of electronics pertaining to this species. Cyclic voltammetry (CV) of a solution of 2 in THF revealed two reversible, one-electron oxidations (at E = −0.8, −1.3 V vs. [Cp2Fe]0/+) and one partially-reversible one electron reduction process (E = −2.7 V vs. [Cp2Fe]0/+) (see ESI for CV: S27–S31†). Complex 2 is a formally 44-electron trinuclear cluster of the closo structural class and is therefore electronically deficient in accordance with Wade's rules.45,46 To elucidate the nature of these oxidation events, we computed molecular orbitals of 2 (Fig. 2). Analysis of the computed MOs suggests that the oxidation events involve depleting electron density from the d-orbital based Ni–Ni antibonding interactions, thereby providing relief from antibonding instability. The reduction process is best described as the introduction of electron density into the LUMO, which features electron density shared between the Ni3 core and Sn and d-orbital based Ni–Ni antibonding interactions. This is expected to increase Ni–Ni antibonding instability in the complex, which is consistent with the partial irreversibility observed for the redox process observed in the CV. Together, the electrochemical and DFT results of 2 suggest frontier d-orbitals interactions centered at the Ni3 core. Given that the coordination environment around the Ni3 cluster does not allow for activation by the Ni3 center, we propose these charge states may be mediated through the Sn center, potentially enabling the oxidative addition and unique hydrogenation chemistry displayed here.
Footnote |
† Electronic supplementary information (ESI) available. CCDC 2173131, 2169137, 2169160 and 2102266. For ESI and crystallographic data in CIF or other electronic format see https://doi.org/10.1039/d2sc04042e |
This journal is © The Royal Society of Chemistry 2022 |