Redox flexibility in a germanium hydride manifold: hydrogen shuttling via oxidative addition and reductive elimination

Abstract

We report the synthesis of a trimetallic mixed-valence Ge(I)/Ge(II)/Ge(III) trihydride, which presents a structural novel motif among systems of the type (XMH)_n (M = group 14 metal). In terms of reactivity (Ar^NiPr2)GeGe(Ar^NiPr2)(H)Ge(Ar^NiPr2)(H)₂ can act as a source of both the Ge(II) and Ge(IV) hydrides via Ge–H reductive elimination from the central metal centre involving two different regiochemistries.

---

Subvalent germanium hydrides (i.e. hydrides in formal metal oxidation states <4) have played a central role in establishing methodologies for bond activation and catalysis by main group compounds.¹ The digermyne (Ar^DippGe)₂ reported by Power and co-workers in 2005 (where Ar^Dipp = 2,6-Dipp₂C₆H₃ and Dipp = 2,6-ⁱPr₂C₆H₃) offered the first example of facile main group dihydrogen activation, generating a mixture of hydride products (Ar^DippGeH₃, (Ar^DippGeH₂)₂ and (Ar^DippGeH)₂) depending on the reaction stoichiometry.² Orbital comparisons with transition metal complexes offered a rationale for initial H₂ activation occurring at one of the germanium centres in (Ar^DippGe)₂,³ and structural validation of this type of unsymmetrical Ge(I)/Ge(III) mixed-valence dihydride was subsequently obtained for {Ar*(Me₃Si)N}GeGe(H)₂{N(SiMe₃)Ar*} (where Ar* = 4-Me-2,6-(Ph₂CH)₂C₆H₂).⁴ Two-coordinate Ge(II) hydride systems can be accessed by employing an even greater degree of steric bulk, as in the case of (for example) {Ar*((^tBuO)₃Si)N}GeH,⁵ and concurrent access to a Ge-H bond and a vacant coordination site is thought to be important in related systems which act as very active catalysts for carbonyl hydroboration.⁶

In the presence of less sterically demanding X ligands, aggregation of [XGeH] monomers is common (Scheme 1); dimerization via Ge=Ge bond formation is prevalent for germanium (type II systems),^2,4b,7 while H-bridged structures of the type XSn(μ-H)₂SnX are known for heavier tin congeners.^7a,8,9 Unsymmetrical Ge(I)/Ge(III) systems (type III) can also formally be regarded as a product assembled from two [XGeH] monomers via Ge–H oxidative addition at a Ge(II) centre.

Scheme 1 Conceptual link between various isomeric forms of (putative) germanium(II) hydrides.

In recent work we have been interested in the use in low-valent group 14 chemistry of hemi-labile pincer ligands, 2,6-(R₂NCH₂)₂C₆H₃ (R = Et, ⁱPr), and have shown how these can be exploited to control catenation, and to effect reversible E–H bond activation and the reversible uptake of CO₂.¹⁰ Here we show that these supporting ligands can be used to support a novel mixed-valence variant of [XGeH]_n, in the form of a trimetallic aggregate of type (X)Ge-Ge(X)(H)-Ge(X)(H)₂. This system formally contains Ge(I), Ge(II) and Ge(III) centres, and, remarkably, can act as a source of both Ge(II) and Ge(IV) hydrides by reductive Ge–H elimination from the central metal centre with two different regio-chemistries. As such, it offers a demonstration of unprecedented redox flexibility within a germanium hydride manifold.

The reaction of Ar^NiPr2GeCl (1; Ar^NiPr2 = 2,6-(ⁱPr₂NCH₂)₂-C₆H₃)^10a with K[HBEt₃] in toluene at room temperature over a period of 3 h gives rise to a single new species, 2 (Scheme 2), characterized by ¹H and ¹³C NMR spectra indicating a low degree of molecular symmetry.‡ The ¹³C spectrum, for example, features four ⁱPr CH and eight ⁱPr CH₃ signals, together with four resonances due to the methylene carbons of the –CH₂N tethers. By means of comparison, the corresponding ¹³C spectrum at room temperature for ‘simple’ mononuclear species (such as 1 itself)^10a typically features one ⁱPr CH, two ⁱPr CH₃ and one –CH₂N signals. In addition, the ¹H NMR spectrum of the product contains three signals (each integrating to 1H) assigned to germanium-bound hydrogen atoms, namely doublets at δ_H = 4.70 and 4.88 (with coupling constants of 7.1 and 9.7 Hz, respectively) and a doublet of doublets at δ_H = 4.99 (J = 7.1, 9.7 Hz). 2 can also be prepared from precursors containing Ge–O bonds via metathesis reactions with pinacolborane. Thus, reactions of the naphthoxide derivative Ar^NiPr2Ge(ONaph-1) (3) or the formate complex Ar^NiPr2Ge(OC(O)H) (4) with HBpin yield (1-NaphO)Bpin and HC(O)OBpin, respectively,¹¹ together with the same Ar^NiPr2Ge-containing species. The identity of the product was definitively established by a combination of elemental microanalysis, IR spectroscopy and X-ray crystallography (Fig. 1).

Scheme 2 Synthesis of trinuclear germanium hydride 2 from either Ge–Cl or Ge–O containing precursors via metathesis with B–H bonds.

Fig. 1 Molecular structure of 2 in the solid state as determined by X-ray crystallography. ⁱPr groups shown in wireframe format and most hydrogen atoms omitted for clarity; thermal ellipsoids plotted at the 40% probability level. Key bond lengths (Å): Ge(1)–Ge(2) 2.531(1), Ge(2)–Ge(3) 2.433(1), Ge(1)–N(1) 2.166(1), Ge(1)–C(1) 1.996(1), Ge(2)–C(21) 1.991(2), Ge(3)–C(41) 1.973(1), Ge(2)–H(2) 1.45(2), Ge(3)–H 1.43(2), 1.47(2).

The solid-state structure shows 2 to be a trinuclear species, (Ar^NiPr2)GeGe(Ar^NiPr2)(H)Ge(Ar^NiPr2)(H)₂, featuring a chain of three metal atoms, two of which engage in no short contacts (<4 Å) with the amine donors of the pendant ligand arms. The third germanium centre, Ge(1), is coordinated by a single amine donor, with the associated Ge–N distance (2.166(1) Å) being similar to those found for the terminal, N-donor ‘capped’ metal centres in di- or tetra-nuclear Ge(I) chains featuring the same (or related) pincer ligands.^10a The Ge(1)–Ge(2) separation (2.531(1) Å) is consistent with a relatively long single bond (cf. 2.5052(3) Å for the corresponding linkage in (Ar^NiPr2Ge)₄ and ca. 2.40 Å for the sum of the covalent radii.^10a,12 The lack of coordinated amine arms at either Ge(2) or Ge(3) suggests the presence of metal-bound hydrogen atoms, and notwithstanding the uncertainties in the location of hydrogens by X-ray crystallography, the presence of two H atoms at Ge(3) and one at Ge(2) is consistent with (i) peaks located in the difference Fourier map; (ii) the presence of three Ge–H stretching bands (at 1981, 2003 and 2042 cm⁻¹) in the solid-state IR spectrum of 2; and (iii) with the coupling patterns observed for the three germanium bound hydrogens in the ¹H NMR spectrum of 2. A similar pattern of resonances (two doublets and a doublet of doublets) has been reported previously for {(HCDippN)₂B}Ge(H)₂Ge(H){N(SiMe₃)₂}{B(NDippCH)₂},¹² with the larger doublet splitting (9.7 Hz for 2, cf. 12.0 Hz) being attributed to the geminal ²J_HH coupling, and the smaller one (7.1 Hz for 2, cf. 2.8 Hz) to the trans³J_HH coupling. The two compounds feature similar alignments of the Ge–H bonds across the Ge₂ unit in the solid state (2: H(2)–Ge(2)–Ge(3)–H torsion angles of 166.1, 77.2°, cf. 175.6 and 46.7°).¹³

Assignment of formal oxidation states to the metal centres in 2, implies the presence of a central Ge(II) unit (labelled Ge(2)), flanked by Ge(I) and Ge(III) centres bearing zero and two hydrogen atoms, respectively (i.e. Ge(1) and Ge(3)). Such as description is also consistent with the shorter Ge(2)–Ge(3) distance (2.433(1) Å, cf. 2.531(1) Å for Ge(1)–Ge(2)), reflecting the smaller covalent radius associated with Ge(III) over Ge(I).

The structure of 2 represents a novel motif added to the family of known isomeric hydride systems of the type (XGeH)_n (X = aryl, amido etc.). In addition to monomeric and Ge=Ge bonded digermene Ge(II) systems (e.g.I and II, Scheme 1),^2,4b,5,7,9 unsymmetrical Ge(I)/Ge(III) derivatives of the type XGeGe(H)₂X (III) have been postulated as key intermediates in the activation of H₂ by digermynes,³ and have been structurally characterized for X = N(SiMe₃)Ar*.⁴ Conceptually, type III systems could be regarded as dimeric species, (XGeH)₂, formed via formal oxidative addition of the Ge–H bond of one monomeric hydrido-germylene to the germanium centre of another (Scheme 1). By extension, a subsequent Ge–H activation step at the unsaturated metal centre of the resulting (germyl)germylene could then generate a species akin to 2, featuring a linear chain of three Ge centres. With this in mind – and given the reversibility demonstrated recently for E–H oxidative addition at related Sn centres^10b – we were interested to probe the scope for 2 to act as a source of monomeric germanium hydride species. Given the presence of both Ge(I) and Ge(III) centres in the terminal positions in 2, we hypothesized that this system might act as a source of germanium hydride moieties in different formal oxidation states via Ge–H reductive elimination from the central metal centre occurring via different regio-chemistries (Scheme 3).

Scheme 3 Potential Ge–H reductive elimination processes leading to the cleavage of 2.

In the case of related tin(II) systems, it has been shown that reaction with CO₂ (to give a tin formate complex) offers a viable route for trapping hydride species of the type Ar^NiPr2SnH.^10b With this in mind, we examined the reactivity of 2 towards CO₂, aiming to probe the viability of the trinuclear framework to act as a synthon for [Ar^NiPr2GeH]. In the event, this reaction proceeds rapidly and quantitatively at room temperature to generate the formate complex Ar^NiPr2Ge{OC(O)H} (4; Scheme 4). The same compound can also be generated via the metathesis reaction between Ar^NiPr2GeCl and K[O₂CH], and its molecular structure (and κ¹ coordination mode of the formate ligand) confirmed crystallographically (Fig. 2). Interestingly, while the formation of 4 in this way is consistent with the idea of 2 acting as the synthetic equivalent of three molecules of Ar^NiPr2GeH, it is noteworthy that this reaction is chemically reversible, such that 4 can be re-converted into 2 (and HC(O)OBpin) by the action of pinacolborane.

Scheme 4 Chemically reversible cleavage of 2 into Ge(II) products by the insertion of CO₂ into Ge–H bonds.

Fig. 2 Molecular structure of 4 in the solid state as determined by X-ray crystallography. ⁱPr groups shown in wireframe format and most hydrogen atoms omitted for clarity; thermal ellipsoids plotted at the 40% probability level. Key bond lengths (Å): Ge(1)–C(1) 1.993(2), Ge(1)–O(1) 1.952(1), O(1)–C(21) 1.273(2), C(21)–O(2) 1.223(3), Ge(1)–N 2.341(1), 2.832(1).

Ge–H reductive elimination from 2 in the opposite sense to generate a Ge(IV) hydride species can also be demonstrated (Schemes 3 and 5). Thus, the reaction of 2 with the hydride abstraction agent [Ph₃C][B(C₆F₅)₄] in 1,2-difluorobenzene leads to precipitation of the insoluble Ge(I) tetramer (Ar^NiPr2Ge)₄,^10a accompanied by the formation of the cation [Ar^NiPr2GeH₂]⁺, as the [B(C₆F₅)₄]⁻ salt (5; Scheme 5). 5 has been characterized by standard spectroscopic methods and its structure in the solid state confirmed by X-ray crystallography (Fig. 3). The presence of two germanium-bound hydrogen atoms within an overall trigonal bipyramidal metal coordination environment is implied by the presence of (i) a single resonance (integrating to 2H) in the ¹H NMR spectrum at δ_H = 5.92 ppm and (ii) bands due to Ge–H stretching modes at 2162 and 2179 cm⁻¹ in the IR spectrum of solid 5. The bond lengths associated with the germanium centre (d(Ge(1)–C(1)) = 1.895(2) Å, Ge(1)–N 2.263(2), 2.272(1) Å) are somewhat shorter than those found in (for example) 1, presumably reflecting the smaller size and greater Lewis acidity of Ge(IV) over Ge(II), and the effect of the overall cationic charge.

Scheme 5 Cleavage of 2 leading to the formation of a Ge(IV) dihydride cation via reaction with [Ph₃C][B(C₆F₅)₄].

Fig. 3 Molecular structure of 5 in the solid state as determined by X-ray crystallography. ⁱPr groups shown in wireframe format and solvent molecule/most hydrogen atoms omitted for clarity; thermal ellipsoids plotted at the 40% probability level. Key bond lengths (Å): Ge(1)–C(1) 1.895(2), Ge(1)–H 1.46(2), 1.49(2), Ge(1)–N 2.263(2), 2.272(1).

In conclusion, we report the synthesis of a trimetallic mixed-valence Ge(I)/Ge(II)/Ge(III) trihydride, the structure of which represents a novel motif among systems of the type (XMH)_n (M = group 14 metal). (Ar^NiPr2)GeGe(Ar^NiPr2)(H)Ge(Ar^NiPr2)(H)₂ (2) can be viewed conceptually as being formed from monomeric [(Ar^NiPr2)GeH] units by successive Ge–H oxidative addition processes occurring at Ge(II), via the intermediacy of the (germyl)germylene, (Ar^NiPr2)GeGe(Ar^NiPr2)(H)₂. Consistent with this idea, 2 can act as a source of the [(Ar^NiPr2)GeH] fragment in reactions with CO₂ (generating the formate Ar^NiPr2Ge{OC(O)H}, 4), and 4 can be re-converted back into 2via Ge–O/B–H metathesis with pinacolborane. On the other hand, Ge–H reductive elimination from 2 occurring with a different regiochemistry can be used as a source of the Ge(IV) hydride (Ar^NiPr2)GeH₃, which undergoes hydride abstraction with [Ph₃C][B(C₆F₅)₄] to generate the trigonal bipyramidal cation [Ar^NiPr2GeH₂]⁺. The chemically reversible and regiochemically flexible nature of this Ge–H activation chemistry therefore facilitates unprecedented redox flexibility within a germanium hydride manifold.

AC carried out the synthetic experiments and spectroscopic characterization of new compounds; AH and JH carried out the crystallographic measurements; SA managed the project and wrote the manuscript.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. For a recent review of main group hydride chemistry, see: M. M. D. Roy, A. A. Omaña, A. S. S. Wilson, M. S. Hill, S. Aldridge and E. Rivard, Chem. Rev., 2021, 121, 12784–12965.
2. G. H. Spikes, J. C. Fettinger and P. P. Power, J. Am. Chem. Soc., 2005, 127, 12232–12233.
3. P. P. Power, Nature, 2010, 463, 171–177.
4. (a) J. Li, C. Schenk, C. Goedecke, G. Frenking and C. Jones, J. Am. Chem. Soc., 2011, 133, 18622–18625 . For a related system formed in the presence of a Lewis base, see: ; (b) A. F. Richards, A. D. Phillips, M. M. Olmstead and P. P. Power, J. Am. Chem. Soc., 2003, 125, 3204–3205.
5. T. J. Hadlington, B. Schwarze, E. I. Izgorodina and C. Jones, Chem. Commun., 2015, 51, 6854–6857.
6. T. J. Hadlington, M. Hermann, G. Frenking and C. Jones, J. Am. Chem. Soc., 2014, 136, 3028–3031.
7. (a) T. J. Hadlington, M. Hermann, J. Li, G. Frenking and C. Jones, Angew. Chem., Int. Ed., 2013, 52, 10199–10203; (b) J. A. Kelly, M. Juckel, T. J. Hadlington, I. Fernández, G. Frenking and C. Jones, Chem. – Eur. J., 2019, 25, 2773–2785.
8. (a) E. Rivard, R. C. Fischer, R. Wolf, Y. Peng, W. A. Merrill, N. D. Schley, Z. Zhu, L. Pu, J. C. Fettinger, S. J. Teat, I. Nowik, R. H. Herber, N. Takagi, S. Nagase and P. P. Power, J. Am. Chem. Soc., 2007, 129, 16197–16208; (b) Y. Peng, B. D. Ellis, X. Wang and P. P. Power, J. Am. Chem. Soc., 2008, 130, 12268–12269.
9. For a more wide-ranging theoretical discussion of M(II) hydrides of Ge and Sn see: G. Trinquier, J. Am. Chem. Soc., 1990, 112, 2130–2137.
10. (a) A. Caise, L. P. Griffin, A. Heilmann, C. McManus, J. Campos and S. Aldridge, Angew. Chem., Int. Ed., 2021, 60, 15606–15612; (b) A. Caise, A. E. Crumpton, P. Vasko, J. Hicks, C. McManus, N. H. Rees and S. Aldridge, Angew. Chem., Int. Ed., 2022, 61, e202114926; (c) A. Caise, L. P. Griffin, A. Heilmann, C. McManus, J. Campos and S. Aldridge, Angew. Chem., Int. Ed., 2022, 61, e202117496 . For examples of the chemistry of related ligands see: ; (d) R. Jambor, B. Kašná, K. N. Kirschner, M. Schürmann and K. Jurkschat, Angew. Chem., Int. Ed., 2008, 47, 1650–1653; (e) M. Novák, M. Bouška, L. Dostál, A. Růžička, A. Hoffmann, S. Herres-Pawlis and R. Jambor, Chem. – Eur. J., 2015, 21, 7820–7829; (f) M. Novák, L. Dostál, Z. Růžičková, S. Mebs, J. Beckmann and R. Jambor, Organometallics, 2019, 38, 2403–2407.
11. (a) C. Bibal, S. Mazières, H. Gornitzka and C. Couret, Polyhedron, 2002, 21, 2827–2834; (b) E. A. Romero, J. L. Peltier, R. Jazzar and G. Bertrand, Chem. Commun., 2016, 52, 10563–10565.
12. B. Cordero, V. Gómez, A. E. Platero-Prats, M. Revés, J. Echeverría, E. Cremades, F. Barragán and S. Alvarez, Dalton Trans., 2008, 2832–2838.
13. M. Usher, A. V. Protchenko, A. Rit, J. Campos, E. Kolychev, R. Tirfoin and S. Aldridge, Chem. – Eur. J., 2016, 22, 11685–11698.

---

Footnotes

† Electronic supplementary information (ESI) available: Full synthetic/characterizing date’ representative spectra; CIFs. CCDC 2242325–2242328. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d3cc01411h

‡ Synthetic and characterizing data for new compounds can be found in the ESI.† Crystallographic data for compounds 2, 3, 4 and 5 can be obtained from the CCDC (citing reference numbers 2242325–2242328).

---