Atom-economic access to cationic magnesium complexes

Abstract

Cationic alkaline-earth complexes attract interest for their enhanced Lewis acidity and reactivity compared with their neutral counterparts. Synthetic protocols to these complexes generally utilize expensive specialized reagents in reactions generating multiple by-products. We have studied a simple ligand transfer approach to these complexes using (NacNac)MgR and ER₃ (NacNac = β-diketiminate anion; E = group 13 element; R = aryl/amido anion) which demonstrates high atom economy, opening up the ability to target these species in a more sustainable manner. The success of this methodology is dependent on the identity of the group 13 element with the heavier elements facilitating faster ligand exchange. Furthermore, while this reaction is successful with aromatic ligands such as phenyl and pyrrolyl, the secondary amide piperidide (pip) fails to transfer, which we attribute to the stronger 3-centre-4-electron dimerization interaction of Al₂(pip)₆.

---

Introduction

Interest in cationic alkaline-earth metal complexes has recently been growing on account of their high Lewis acidity, interesting reactivity and potential exploitation as sustainable-metal catalysts.¹ A common approach to their development is to incorporate a bulky bidentate β-diketiminate (BDI or NacNac) monoanion as the stabilizing ligand² due to their ease of synthesis, steric/electronic tunability, and the sizeable steric protection they can offer to the cationic metal centre. These ligands were already particularly prevalent in s-block chemistry in for example the stabilization of low-valent [(Mg(I), Ca(I)] or reactive hydrido-equipped alkaline-earth metal complexes,³ the stabilization of dihydropyridyl ligands and hydrogen-rich B/N ligands⁴ and the development of magnesium complexes for deprotonation or C–H/C–F bond activation and consequent stabilization of the resulting sensitive products.⁵ However, it is now established that these ligands are not as innocent as they were initially thought to be, with recent alkali–metal research in this area showing that NacNac is susceptible to attack at the backbone site⁶ on account of the HOMO being located at the C_γ position, which renders it nucleophilic.⁷

If such species are genuinely contenders to become sustainable reagents it is not enough for them to simply be accessible but rather, they must be sustainably accessible via facile methodologies using inexpensive reagents, in high yields and with minimal or ideally no waste products. For example, some textbook examples of these complexes, [(NacNac)Mg]⁺ [X]⁻ were prepared on sub-mM scales by reaction of (NacNac)MgnBu with the specialized reagents [CPh₃]⁺ [X]⁻, resulting in elimination of Ph₃CH and 1-butene by-products ([X]⁻ = weakly coordinating anions [B(C₆F₅)₄]⁻ and [Al(OC(CF₃)₃)₄]⁻Scheme 1a).⁸

Scheme 1 Synthetic methodologies employed to access NacNac supported cationic magnesium complexes: (a) previous work; (b) this work.

Recently we have been interested in exploiting ligand exchange reactions as a means of delivering high atom economy in the synthesis of magnesium aluminate complexes with application as magnesium battery electrolytes and thus postulated that a similar approach may be beneficial here (Scheme 1b).⁹ We report now our findings, discussing both the advantages and limitations of this methodology.

Results and discussion

We initiated this study by reacting (^DippNacNac)MgPh·THF (Dipp = 2,6-diisopropylphenyl, see Fig. S1–S3†)¹⁰ with the Lewis acid AlPh₃·OEt₂¹¹ in a 1 : 1 stoichiometric ratio in THF solution according to Scheme 1b. This was done on an NMR scale in a J. Youngs tube so that the reaction could be monitored using ²⁷Al NMR spectroscopy, exploiting the tetrahedral symmetry of the putative anion which should deliver a sharp and narrow resonance. This was realised in such a resonance at 133 ppm consistent with the emergence of the [AlPh₄]⁻ anion.¹² The broad nature of the AlPh₃ resonance meant that loss of starting material could not be adequately monitored and so this study was supplemented with ¹H NMR data to determine when the reaction was complete. Although necessarily busy on account of converting one multi-proton containing species into another similar species, the area around 5 ppm in the ¹H NMR spectrum was informative as only the singlet of the NacNac γ-H resonates in this region. After two days at room temperature the ratio of reactant : product was nearing 50 : 50 and so heat was applied for a further week. However, this only improved the ratio to 40 : 60 after 2 days, suggesting either that the reaction has concluded, or was in equilibrium. This reaction was repeated on a millimolar scale in the hope that recrystallization of the target SSIP [(^DippNacNac)Mg]⁺ [AlPh₄]⁻ (1, Scheme 2) would shift this equilibrium to the right. The product after attempted recrystallization from a hexane/THF mix was a yellow oil which contained only a few colourless crystals. Single crystal X-ray analysis revealed these to be the charge-separated complex [Mg·6THF]²⁺ 2[AlPh₄]⁻ (1a, Fig. 1) which we believe is a minor Schlenk equilibrium¹³ product of the targeted complex 1. However, its anionic moiety unequivocally confirms that ligand (phenyl) transfer from Mg to Al has occurred as intended, supporting the ²⁷Al NMR observations.

Scheme 2 Proposed Schlenk equilibrium of heteroleptic complex 1 to give small quantities of homoleptic Mg complexes 1a and 1b.

Fig. 1 Molecular structure of [Mg·6THF]²⁺ 2[AlPh₄]⁻ (1a) with hydrogen atoms, minor disordered component of a THF backbone and one of the two aluminate anions omitted for clarity and with thermal ellipsoids drawn at 50% probability. Selected bond lengths (Å) and angles (°): Al–C, 2.006(3)–2.015(3), av. 2.011; Mg–O, 2.082(2)–2.122(2), av. 2.098; C–Al–C, 106.7(1)–113.1(1), av. 109.4; cis O–Mg–O, 88.78(7)–90.81(7), av. 90.00.

A ¹H NMR spectrum of the yellow oil in C₆D₆ solution was highly complicated, indicative of the presence of multiple species. However, focusing on the 4.8–5.0 ppm region, where the NacNac backbone (γ) hydrogen atom resonates, this displayed four distinct resonances. The two smaller resonances at 4.88 and 4.86 ppm can be assigned to ^DippNacNac(H) and (^DippNacNac)MgPh starting material respectively. The resonance at 4.97 ppm is tentatively assigned to Mg(^DippNacNac)₂ (1b) by comparison with the spectrum of the pure compound reported by Harder to resonate at 5.01 ppm in the related NMR solvent C₆D₅CD₃.^2a The most intense resonance at 4.83 ppm is thus assigned to the desired cation in 1. Further evidence came from the ²⁷Al NMR spectrum which again displayed a sharp singlet at 133 ppm diagnostic of the highly symmetrical tetraphenylaluminate anion.¹² We are confident that this resonance does not represent the crystallographically verified magnesium bisaluminate species 1a as we have prepared this rationally (see ESI, Fig. S4–S6†) and it is too insoluble to display any resolvable resonances in C₆D₆, with highly polar DMSO required to dissolve it sufficiently to record usable spectra.

Next, we decided to study the effect of the group 13 element upon the ligand transfer reaction and so turned our attention to the lighter congener BPh₃. However, this corresponding reaction was found to be much slower than that with AlPh₃. Consequently, we monitored this borane reaction at 60 °C in d₈-THF by ¹H and ¹¹B NMR spectroscopy over a period of 8 days. The success of the reaction was evidenced again at the 5 ppm region of the ¹H NMR spectrum where loss of the resonance corresponding to the NacNac γ-CH group of the starting material (4.91 ppm) was accompanied by downfield growth of the product resonance, [(^DippNacNac)Mg]⁺ [BPh₄]⁻ (2) at 5.11 ppm and a small resonance of (^DippNacNac)₂Mg (5.00 ppm, Fig. 2 left, see Fig. S7 and S8† for full spectra) as seen previously in the alane reaction.

Fig. 2 Section of ¹H (left) and ¹¹B NMR spectra (right) of NMR scale reaction of ^DippNacNacMgPh and BPh₃ in d₈-THF showing the region diagnostic of the C_γ-H of the ^DippNacNac species; all values in ppm.

This result was mirrored by the ¹¹B NMR spectrum which showed gradual loss of the broad resonance of BPh₃ at 52.8 ppm and the growth of a sharp singlet of the [BPh₄]⁻ anion at −6.6 ppm (Fig. 2, right).¹⁴ While this reaction was a success, the long timescale required to prepare the boron complex 2versus aluminate 1 inspired us to pursue a gallium-centred Lewis acid. Triphenylgallium was therefore synthesised (see Fig. S9–S11†) and reacted with ^DippNacNacMgPh in d₈-THF. Fig. 3 shows the ¹H NMR spectra of both starting materials (GaPh₃·OEt₂ and ^DippNacNacMgPh), followed by the reaction of both after 15 minutes (annotated in Fig. S12†). From these it can be seen by the lack of Mg-Ph peaks at approximately 6.6 ppm, that all of the starting ^DippNacNacMgPh has been consumed. New phenyl resonances can also be seen at 6.82, 6.91 and 7.57 ppm indicating the presence of a new M-Ph unit corresponding to the tetraphenylgallate anion of [(^DippNacNac)Mg]⁺ [GaPh₄]⁻ (3). Small resonances are still visible from the GaPh₃·OEt₂ starting material suggesting that this reactant was in slight excess in this reaction. It is clear from these observations that the reactivity increases as group 13 is descended.

Fig. 3 ¹H NMR spectra of (a) GaPh₃·OEt₂, (b) ^DippNacNacMgPh and (c) the reaction between GaPh₃·OEt₂ and ^DippNacNacMgPh to yield 3 in d₈-THF after 15 minutes.

Returning to an aluminium-centred Lewis acid, we probed the effect of modifying the electronics of the transferred ligand. Specifically, p-tolyl was chosen since modification at the para-position is unlikely to sterically prohibit the coordination of four ligands around the aluminium centre as desired in the product. The starting materials (^DippNacNac)Mg(p-Tol)·THF and Al(p-Tol)₃·OEt₂ were prepared in an analogous manner to the original aryl substituents (see Fig. S13–S19†). NMR spectroscopic monitoring in d₈-THF showed that the reaction to yield [(^DippNacNac)Mg]⁺ [Al(p-Tol)₄]⁻ (4) was considerably slower than that of the original phenyl transfer reaction. Specifically, the growth of a sharp singlet in the ²⁷Al NMR spectrum at 133 ppm took much longer, and ¹H monitoring (specifically focusing on the emergence of the new resonance at 5.09 ppm in comparison to the original NacNac resonance at 4.90 ppm, see Fig. S20 and S21†) informed that only 10% conversion had occurred after 56 hours at room temperature. Heating at 60 °C for a week failed to make further change to this ratio, while variable temperature NMR studies gave no indication of an equilibrium between starting materials and products. With such a poor conversion, it seemed unwise to scale up and pursue a fractional crystallization of the product and so at this stage we turned to an alternative class of ligand, switching from aryl ligand transfer to (secondary) amido ligand transfer. Keeping with a six-membered cyclic ligand, we utilized the piperidide (NC₅H₁₀, pip) ligand for this task. [(^DippNacNac)Mg(pip)]₂ was prepared according to literature procedures,¹⁵ while [Al(pip)₃]₂ was prepared via salt metathesis of lithium piperidide with AlCl₃ in a 3 : 1 ratio in diethyl ether. Crucially, the presence of Lewis donor ether as the reaction medium did not break this up into monomeric AlR₃·OEt₂ as was seen with the triarylaluminium complexes studied earlier, but rather recrystallisation from Et₂O or THF yielded dimeric unsolvated [Al(pip)₃]₂.¹⁶ This was also witnessed by Smith et al. who prepared this dimer by direct deprotonation of piperidine with alane in THF, with the product remaining dimeric even in solution.¹⁷ The reaction of these two substrates in THF showed no evidence of ligand transfer with NMR analysis demonstrating only starting material. This may be due to the persistent dimeric nature of the neutral aluminium starting material which contains stronger 3-centre 4-electron dimerizing interactions which inhibits its ability to act as a ligand acceptor from the magnesium complex.

Consequently, given the relative success of phenyl transfer, we next studied the aromatic amide pyrrole, from the basis that its nitrogen lone pair would be unavailable for Lewis pair donation on account of it contributing to the aromaticity of the anion. Thus, a THF solution of Al(pyr)₃·OEt₂ (pyr = pyrrolyl, C₄H₄N⁻) was introduced to a THF solution of (^DippNacNac)Mg(pyr)·THF (see Fig. S22–S27†) and this mixture was stirred at room temperature for two hours. Removing solvent under reduced pressure and attempting to re-dissolve the residue in hexane, gave a cloudy solution. This was filtered and cooled to −20 °C to give colourless crystals of the desired product, [(^DippNacNac)Mg·2THF]⁺ [Al(pyr)₄]⁻ (5) in a 50% crystalline yield (Fig. 4), although NMR analysis suggests that the solution yield is much higher (vide infra). Gratifyingly, this product confirms that high atom economy has been achieved here as we have simply transferred a pyrrolyl anion from Mg to Al with loss of Al-coordinated Et₂O to form the desired charge-separated magnesium aluminate complex.

Fig. 4 Ion pair structure of [(^DippNacNac)Mg·2THF]⁺ [Al(pyr)₄]⁻ (5) with hydrogen atoms and minor disordered component of a THF backbone omitted for clarity and with thermal ellipsoids drawn at 50% probability. Selected bond lengths (Å) and angles (°): Al–N, 1.842(2)–1.861(2) av. 1.851; Mg1–N5, 2.024(2); Mg1–N6, 2.006(2); Mg1–O1, 1.991(2); Mg1–O2, 1.987(2); N–Al–N, 107.5(1)–111.7(1) av. 109.4; N5–Mg1–N6, 96.31(7); O1–Mg1–O2, 99.33(7); N5–Mg1–O1, 120.17(7); N5–Mg1–O2, 112.46(7); N6–Mg1–O1, 113.72(7); N6–Mg1–O2, 115.99(7).

To the best of our knowledge the anionic moiety, [Al(pyr)₄]⁻ has never been reported before although tetrahedral tetra(cyclic-amido)aluminate structures such as [Al(1,4-dihydropyridyl)₄]⁻¹⁸ and calix[4]pyrrole aluminates¹⁹ have appeared in the literature. The cationic moiety of 5 comprises a four-coordinate magnesium centre coordinated by the two nitrogen atoms of the NacNac ligand and two THF molecules. There is asymmetry in the coordination of the two THF molecules such that the Mg centre is heavily distorted from tetrahedral with one THF molecule lying almost in the plane of the NacNac ligand and the other lying perpendicular to it. This has been witnessed previously in neutral ether-solvated (^DippNacNac)MgR complexes (R = Ph,¹⁰ C₆F₅,^5b C₆F₄CF₃^5b). Despite being in markedly different environments, these THF molecules display very similar O–Mg bond lengths [1.987(2) and 1.991(2) Å respectively]. These values are intermediate between the O–Mg values in (^DippNacNac)MgOtBu·2THF [2.048(2) and 1.844(2) to neutral THF and anionic ⁻OtBu respectively]²⁰ demonstrating the increased Lewis acidity of the magnesium in its cationic form.

The veracity of this structure was supported in solution by ¹H, ¹³C and ²⁷Al NMR spectroscopy (Fig. S28–S30†). For example, the symmetric aluminate anion is represented by a sharp singlet in the ²⁷Al NMR spectrum at 98.1 ppm. The ¹H and ¹³C NMR spectra are as anticipated, with the CH backbone resonance of 5 at 4.71 ppm coming very close to that assigned to the similar complex 1 (4.83 ppm). However, despite multiple attempts we were never able to obtain a purity higher than 80%, with the remaining 20% of material being the starting material, (^DippNacNac)Mg(pyr)·THF. Efforts to increase purity included multiple fractional recrystallizations, rapidly precipitating the product out of solution, longer reaction times in different solvents and including an excess of Al(pyr)₃ to ensure complete reaction of the Mg starting material, but all proved unsuccessful. It is perhaps unsurprising that this starting material co-crystallizes with the desired product given the similarities in their structures, as a five-membered anion (pyrrolyl) is ultimately replaced with a neutral five-membered ligand (THF). Finally, we determined the Lewis acidity of the Mg cation in 5via the Gutmann–Beckett method²¹ by addition of a small amount of Et₃P=O to the NMR sample followed by recording of the ³¹P NMR chemical shift. This methodology provides an Acceptor Number for the Lewis acid on a scale of 0–100. Complex 5 gave a chemical shift of 70.8 ppm for an Acceptor Number of 65.9, which is comparable for the same Mg cation paired with a [B(C₆F₅)₄]⁻ anion (72.8 ppm; AN 70.3)^1a despite lack of fluorination of the aryl substituents.

Conclusion

In this study, we have demonstrated that β-diketiminate supported magnesium cationic complexes can be prepared in an atom economical way by ligand transfer of a monodentate anion from Mg to a group 13 Lewis acid. While this approach represents a valuable and potentially highly useful methodology since it utilises only cheap and easy to prepare starting materials, there are also some limitations to the approach that must be considered. Schlenk equilibria may be a factor as the desired final products are heteroleptic magnesium complexes of general formula [(NacNac)Mg]⁺ [ER₄]⁻. The identity of the central group 13 element of the Lewis acid influences the rate of the reaction as demonstrated using EPh₃ (E = B, Al, Ga), whereby the ligand transfer reaction proceeds faster as the group is descended. The aggregation state of the Lewis acid is also a factor in the outcome as no reaction was seen using tris-piperidyl aluminium, which dimerizes via 3-centre 4-electron bonding (even in the presence of Lewis basic ethers), whereas tris-pyrrolyl aluminium, whose nitrogen lone pair forms part of the aromatic manifold of the ring, proceeds smoothly and in good yield.

Whilst this ligand transfer methodology is not a general approach to cationic magnesium complexes, it does represent a viable potential route and its cost-effectiveness and atom-economy means that it is worthy of consideration on a case-by-case basis, bearing the above limitations in mind.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We gratefully acknowledge Prof. Robert Mulvey and Jennifer Lynch for interesting discussions on Mg-NacNac chemistry.

References

1. (a) J. Pahl, S. Brand, H. Elsen and S. Harder, Chem. Commun., 2018, 54, 8685–8688; (b) S. Brand, H. Elsen, J. Langer, W. A. Donaubauer, F. Hampel and S. Harder, Angew. Chem., Int. Ed., 2018, 57, 14169–14173; (c) L. Garcia, M. D. Anker, M. F. Mahon, L. Maron and M. S. Hill, Dalton Trans., 2018, 47, 12684–12693; (d) A. Friedrich, J. Pahl, H. Elsen and S. Harder, Dalton Trans., 2019, 48, 5560–5568; (e) J. Pahl, A. Friedrich, H. Elsen and S. Harder, Organometallics, 2018, 37, 2901–2909; (f) K. Thum, A. Friedrich, J. Pahl, H. Elsen, J. Langer and S. Harder, Chem. – Eur. J., 2021, 27, 2513–2522.
2. (a) S. Harder, Organometallics, 2002, 21, 3782–3787; (b) S. P. Sarish, S. Nembenna, S. Nagendran and H. W. Roesky, Acc. Chem. Res., 2011, 44, 157–170; (c) L. Bourget-Merle, M. F. Lappert and J. R. Severn, Chem. Rev., 2002, 102, 3031–3066.
3. (a) S. P. Green, C. Jones and A. Stasch, Science, 2007, 318, 1754–1757; (b) S. J. Bonyhady, C. Jones, S. Nembenna, A. Stasch, A. J. Edwards and G. J. McIntyre, Chem. – Eur. J., 2010, 16, 938–955; (c) T. X. Gentner, B. Rösch, G. Ballmann, J. Langer, H. Elsen and S. Harder, Angew. Chem., Int. Ed., 2019, 58, 607–611; (d) A. S. S. Wilson, M. S. Hill and M. F. Mahon, Organometallics, 2019, 38, 351–360; (e) H. Elsen, J. Langer, M. Wiesinger and S. Harder, Organometallics, 2020, 39, 4238–4246.
4. (a) M. S. Hill, D. J. MacDougall and M. F. Mahon, Dalton Trans., 2010, 39, 11129–11131; (b) D. J. Liptrot, M. S. Hill, M. F. Mahon and D. J. MacDougall, Chem. – Eur. J., 2010, 16, 8508–8515; (c) M. S. Hill, M. Hodgson, D. J. Liptrot and M. F. Mahon, Dalton Trans., 2011, 40, 7783–7790; (d) M. S. Hill, G. Kociok-Köhn, D. J. MacDougall, M. F. Mahon and C. Weetman, Dalton Trans., 2011, 40, 12500–12509; (e) J. Intemann, M. Lutz and S. Harder, Organometallics, 2014, 33, 5722–5729; (f) P. Bellham, M. D. Anker, M. S. Hill, G. Kociok-Köhn and M. F. Mahon, Dalton Trans., 2016, 45, 13969–13978; (g) C. Weetman, M. S. Hill and M. F. Mahon, Polyhedron, 2016, 103, 115–120; (h) C. Weetman, M. S. Hill and M. F. Mahon, Chem. – Eur. J., 2016, 22, 7158–7162; (i) J. Spielmann, D. F.-J. Piesik and S. Harder, Chem. – Eur. J., 2010, 16, 8307–8318; (j) C. Jones, S. J. Bonyhady, S. Nembenna and A. Stasch, Eur. J. Inorg. Chem., 2012, 2596–2601.
5. (a) S. E. Baillie, V. L. Blair, T. D. Bradley, W. Clegg, J. Cowan, R. W. Harrington, A. Hernan-Gomez, A. R. Kennedy, Z. Livingstone and E. Hevia, Chem. Sci., 2013, 4, 1895–1905; (b) C. Bakewell, A. J. P. White and M. R. Crimmin, J. Am. Chem. Soc., 2016, 138, 12763–12766; (c) L. Davin, R. McLellan, A. Hernan-Gomez, W. Clegg, A. R. Kennedy, M. Mertens, I. A. Stepek and E. Hevia, Chem. Commun., 2017, 53, 3653–3656; (d) L. Davin, R. McLellan, A. R. Kennedy and E. Hevia, Chem. Commun., 2017, 53, 11650–11653; (e) C. Bakewell, B. J. Ward, A. J. P. White and M. R. Crimmin, Chem. Sci., 2018, 9, 2348–2356; (f) M. Garçon, A. J. P. White and M. R. Crimmin, Chem. Commun., 2018, 54, 12326–12328; (g) L. Davin, W. Clegg, A. R. Kennedy, M. R. Probert, R. McLellan and E. Hevia, Chem. – Eur. J., 2018, 24, 14830–14835.
6. (a) R. M. Gauld, R. McLellan, A. R. Kennedy, J. Barker, J. Reid and R. E. Mulvey, Chem. – Eur. J., 2019, 25, 14728–14734; (b) C. Camp and J. Arnold, Dalton Trans., 2016, 45, 14462–14498; (c) F. A. LeBlanc, A. Berkefeld, W. E. Piers and M. Parvez, Organometallics, 2012, 31, 810–818; (d) J. R. Lynch, A. R. Kennedy, J. Barker, J. Reid and R. E. Mulvey, Helv. Chim. Acta, 2022, 105, e202200082.
7. (a) D. W. Randall, S. D. George, P. L. Holland, B. Hedman, K. O. Hodgson, W. B. Tolman and E. I. Solomon, J. Am. Chem. Soc., 2000, 122, 11632–11648; (b) W. Ren, S. Zhang, Z. Xu and X. Ma, Dalton Trans., 2019, 48, 3109–3115.
8. A. Friedrich, J. Eyselein, J. Langer and S. Harder, Organometallics, 2021, 40, 448–457.
9. S. A. Brown, S. A. Cussen, R. Kennard, S. Marchesini, J. J. Pryke, A. Rae, S. D. Robertson, R. N. Samajdar and A. J. Wain, Chem. Commun., 2022, 58, 12070–12073.
10. A. P. Dove, V. C. Gibson, P. Hormnirun, E. L. Marshall, J. A. Segal, A. J. P. White and D. J. Williams, Dalton Trans., 2003, 3088–3097.
11. S. Krieck, H. Görls and M. Westerhausen, Organometallics, 2008, 27, 5052–5057.
12. O. Mizrahi, N. Amir, E. Pollak, O. Chusid, V. Marks, H. Gottlieb, L. Larush, E. Zinigrad and D. Aurbach, J. Electrochem. Soc., 2008, 155, A103–A109.
13. (a) W. Schlenk and W. Schlenk Jr., Chem. Ber., 1929, 62, 920–924; (b) D. Seyferth, Organometallics, 2009, 28, 1598–1605; (c) R. M. Peltzer, O. Eisenstein, A. Nova and M. Cascella, J. Phys. Chem. B, 2017, 121, 4226–4237; (d) P. Rinke, H. Görls and R. Kretschmer, Inorg. Chem., 2021, 60, 5310–5321.
14. B. Wrackmeyer, Z. Nat. B., 2015, 70, 421–424.
15. L. J. Bole, L. Davin, A. R. Kennedy, R. McLellan and E. Hevia, Chem. Commun., 2019, 55, 4339–4342.
16. L. Sandig-Predzymirska, J. Ortmeyer, J. Wagler, E. Brendler, F. Habermann, M. Anders, M. Felderhoff and F. Mertens, Dalton Trans., 2020, 49, 17689–17698.
17. M. A. Andrianarison, M. C. Ellerby, I. B. Gorrell, P. B. Hitchcock, J. D. Smith and D. R. Stanley, J. Chem. Soc., Dalton Trans., 1996, 211–217.
18. (a) P. T. Lansbury and J. O. Peterson, J. Am. Chem. Soc., 1963, 85, 2236–2242; (b) K. Hensen, A. Lemke, T. Stumpf, M. Bolte, H. Fleischer, C. R. Pulham, R. O. Gould and S. Harris, Inorg. Chem., 1999, 38, 4700–4704.
19. (a) F. Ebner, H. Wadepohl and L. Greb, J. Am. Chem. Soc., 2019, 141, 18009–18012; (b) F. Ebner, L. M. Sigmund and L. Greb, Angew. Chem., Int. Ed., 2020, 59, 17118–17124; (c) F. Ebner, P. Mainik and L. Greb, Chem. – Eur. J., 2021, 27, 5120–5124.
20. M. H. Chisholm, J. C. Huffman and K. Phomphrai, J. Chem. Soc., Dalton Trans., 2001, 222–224.
21. (a) U. Mayer, V. Gutmann and W. Gerger, Monatsh. Chem., 1975, 106, 1235–1257; (b) M. A. Beckett, G. C. Strickland, J. R. Holland and K. S. Varma, Polymer, 1996, 37, 4629–4631; (c) P. Erdmann and L. Greb, Angew. Chem., Int. Ed., 2022, 61, e202114550.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 2073561–2073565 and 2288826–2288828. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d3dt02669h

---