Cationic dinuclear complexes [M₂(PCP)₂μ-Cl][GaCl₄] of the group 10 elements. metallophilic interactions and catalytic dehydrogenation of Me₂NHBH₃

Abstract

The facile synthesis of the cationic dinuclear group 10 complexes [M₂(PCP)₂μ-Cl]⁺ (M = Ni, Pd, Pt) by transmetallation from a simple Ga precursor is reported (PCP = 2,6-(Ph₂P)C₆H₃). Their use for the catalysed dehydrogenation of Me₂NHBH₃ shows that Ni has a higher reactivity than Pt, whereas Pd is inactive.

---

The stability of organometallic species relies essentially on a judicious choice of ligands and substituents, which provide kinetic and electronic stabilization to the metal atom. For this purpose, bidentate 2-diorganophosphinophenyl carbanions, [2-R₂PC₆H₄]⁻ (I, R = Ph, i-Pr) are frequently employed for a large number of main group and transition metal complexes (Fig. 1).¹ Recently, we introduced the related bis(2,6-diphenylphosphino)phenyl carbanion, [2,6-(Ph₂P)₂C₆H₃]⁻ (II),² as a tridentate “PCP” ligand, which is also isoelectronic with the previously known neutral bis(2,6-diphenylphosphino)pyridine, 2,6-(Ph₂P)₂C₅H₂N (III), ligand (Fig. 1).³ So far, the PCP ligand was applied for the preparation of photoactive coinage metal complexes, [Cu₄(PCP)₃]⁺,⁴ and linear and cyclic [Au₄(PCP)₂(tht)₂]²⁺, as well as a mixed-metal pincer complex, e.g. HgAu₂(PCP)Cl₃.² In an effort to broaden the scope of the PCP ligand towards main group chemistry, we have now reacted the lithium reagent 2,6-(Ph₂P)₂C₆H₃]₂Li⁵ with GaCl₃ to give rise to the formation of [2,6-(Ph₂P)₂C₆H₃]₂GaCl (1), which was isolated as colourless crystals in 95% yield (Scheme 1).

Fig. 1 Bi- and tridentate P-based ligands.

Scheme 1 Synthesis of 1.

Notably, variations of the stoichiometric ratio did not give the mono- or trisubstituted Ga-species in significant amounts. The molecular structure of 1 shows that only one of the four P atoms is engaged in the first coordination sphere of the Ga atom (Fig. 2). The spatial arrangement may thus be described as distorted tetrahedral. The primary Ga–P bond length of 1 (2.567(1) Å) is substantially shorter than the remaining secondary Ga–P contacts (3.071(5), 3.382(5), 3.653(7) Å), which, however, are still within the van der Waals limit (4.2 Å). In solution, the P atoms are equivalent on the NMR time scale at room temperature, as the ³¹P NMR spectrum (CDCl₃) of 1 shows only one signal at δ = −3.0 ppm. No qualitative change was observed at −30 °C. With the aim to prepare Ga-based Z-type complexes,⁶ we reacted 1 with two equivalents of Ni(dme)Cl₂, Pd(MeCN)₂Cl₂ and Pt(MeCN)₂Cl₂, respectively. Surprisingly, in all cases rapid transmetallation occurred and the cationic dinuclear group 10 complexes⁷ [M₂(PCP)₂μ-Cl][GaCl₄] (2M, M = Ni, Pd, Pt; PCP = 2,6-(Ph₂P)₂C₆H₃) formed, which were isolated as red, orange and red crystals in 92%, 72% and 64% yield, respectively (Scheme 2). The molecular structures of 2Ni, 2Pd and 2Pt reveal complete separation of the [M₂(PCP)₂μ-Cl]⁺ cations and the [GaCl₄]⁻ anions, which formed from the expelled Ga(iii) ions of 1 (Fig. 3). The spatial arrangement of the metal atoms is distorted square planar and defined by cis-P₂CCl donor sets. The two metal atoms in the dinuclear motifs are bridged by the Cl atoms. The M–M distances of 1Ni (2.713(1) Å), 1Pd (2.807(1) Å) and 1Pt (2.818(1) Å) are indicative of metallophilic interactions⁸ and increase with the atomic number of the group 10 metals.

Fig. 2 Molecular structure of 1 showing 50% probability ellipsoids and the atomic numbering scheme. Selected bond lengths: Ga1–C10 1.975(2), Ga1–C60 1.994(2), Ga1–Cl1 2.250(1), Ga1–P1 2.567(1) Å.

Scheme 2 Synthesis of 2M (M = Ni, Pd, Pt) and complexation reaction to give 1·PtCl₂.

Fig. 3 Molecular structures of 2Ni (top), 2Pd (middle) and 2Pt (bottom) showing 50% probability ellipsoids and the atomic numbering scheme. Selected bond lengths of 2Ni: Ni1–C10 1.925(2), Ni1–Cl1 2.247(1), Ni1–P1 2.149(1), Ni1–P4 2.232(1), Ni2–C60 1.910(2), Ni2–Cl1 2.219(1), Ni2–P2 2.230(1), Ni2–P3 2.141(1), Ni1–Ni2 2.713(1) Å. Selected bond lengths of 2Pd: Pd1–C10 2.042(8), Pd1–Cl1 2.396(2), Pd1–P1 2.252(2), Pd1–P4 2.360(2), Pd2–C60 2.052(5), Pd2–Cl1 2.415(2), Pd2–P2 2.354(2), Pd2–P3 2.236(2), Pd1–Pd2 2.807(1) Å. Selected bond lengths of 2Pt: Pt1–C10 2.040(2), Pt1–Cl1 2.448(1), Pt1–P1 2.222(1), Pt1–P4 2.334(1), Pt2–C60 2.032(2), Pt2–Cl1 2.418(1), Pt2–P2 2.330(1), Pt2–P3 2.227(1), Pt1–Pt2 2.818(1) Å.

The outer P atoms of the structural motifs are situated in trans-position of the Cl atoms, whereas the inner P atoms are located in trans-position of the C atoms. The outer M–P bond lengths of 2Ni (2.141(1), 2.149(1) Å), 2Pd (2.236(2), 2.252(2) Å) and 2Pt (2.222(1), 2.227(1) Å) are generally shorter than the inner M–P bond lengths of 2Ni (2.230(1), 2.232(1) Å), 2Pd (2.354(2), 2.360(2) Å) and 2Pt (2.330(1), 2.334(1) Å), which is consistent with the larger trans-effect of the Pt–C bond compared to the Pt–Cl bond.⁹ The fact that the respective Pt–P bond lengths are shorter than the Pd–P bonds is attributed to the relativistic contraction of the 6s orbital of Pt.⁸ The ³¹P NMR spectra (CDCl₃) of 2Ni, 2Pd, 2Pt show two distinctively different signals of equal intensity, which suggests that the molecular structures are retained in solution. The less shielded signals of 2Ni (δ = 9.2 ppm), 2Pd (δ = 8.2 ppm) and 2Pt (δ = 13.0 ppm) are assigned to the inner P atoms, whereas the more shielded signals of 2Ni (δ = −60.5 ppm), 2Pd (δ = −76.3 ppm) and 2Pt (δ = −76.1 ppm) belong to the outer P atoms. The assignment is based on the observed ¹J(³¹P–¹²³Pt) couplings of 1Pt of the inner (2073 Hz) and outer (4081 Hz) P atoms under the assumption that the magnitude inversely correlates with the bond lengths. The reaction of 1 with a different Pt-source, namely, Pt(Et₂S)₂Cl₂, in an equimolar ratio, indeed afforded the Z-type complex 1·PtCl₂, which was isolated as pale yellow crystals in 59% yield (Scheme 2). Attempts to prepare similar Ni and Pd complexes failed and gave only 2Ni and 2Pd as principal products.

The molecular structure of 1·PtCl₂ confirms the putative square planar arrangement of the Pt atoms, which adopts a cis-configuration of the two P atoms (Fig. 4). The Ga atom is situated perpendicular to Pt square plane. The Pt–Ga bond length (3.078(1) Å) suggests that the occupied dz² orbital at the Pt atom tentatively donates electron density to the vacant orbital at the Ga atom suggesting Z-type coordination. The Ga–P bond length of 1·PtCl₂ (2.543(1) Å) is shorter than in the parent 1 (2.567(1) Å). The ³¹P NMR spectrum (CDCl₃, 243 K) of 1·PtCl₂ shows four equally intense signals. The two more deshielded signals at δ = 30.8 and 24.7 showing ¹J(³¹P–¹²³Pt) couplings of 3430 Hz and 3016 Hz and are assigned to the P atoms involved in the coordination to the Pt atom. The two more shielded signals are doublets centred at δ = −1.7 and −10.1 ppm both showing the same J(³¹P–³¹P) couplings (102 Hz), which are indicative for through space interactions¹⁰ of the two remaining P atoms via their electron lone pairs. This is quite unexpected in light of the large P⋯P distance (3.731(9) Å).

Fig. 4 Molecular structure of 1·PtCl₂ showing 50% probability ellipsoids and the atomic numbering scheme. Selected bond lengths: Ga1–C10 2.009(2), Ga1–C60 1.994(2), Ga1–Cl1 2.190(1), Ga1–P1 2.543(1), Pt1–Cl2 2.359(1), Pt1–Cl3 2.337(1), Pt1–P2 2.258(1), Pt1–P3 2.251(1), Pt1–Ga1 3.078(1) Å.

The most interesting structural feature of the cations 2M (M = Ni, Pd, Pt) are the metallophilic interactions, which were analyzed in more detail by using the electron density based bond descriptors AIM,¹¹ NCI¹² and IGMH.¹³ AIM analysis give rise to bond critical points between the metal–metal bond path for 2Pd and 2Pt, but is absent in case of the lighter 2Ni. However, the electron density minimum found along the direct Ni–Ni axis (0.22 e Å⁻³) is in line with the values found for the electron densities at the bcp for 2Pd (0.26 e Å⁻³) and 2Pt (0.32 e Å⁻³, Table S2, ESI†). These values are in the range found for other nickelophilic (0.18–0.36 e Å⁻³),¹⁴ palladophilic (0.21 e Å⁻³)¹⁵ and bimetallophilic interactions of the 6th period (0.28 e Å⁻³).¹⁶ The delocalization index δ(M|M) is also increasing from 2Ni (0.17) over 2Pd (0.26) to 2Pt (0.37), indicating metallophilic interactions in all three metal species, with increasing strengths from Ni to Pt. This is further corroborated by inspection of the IGMH partitioning. Each metal atom is defined as an individual fragment and the interaction between the two fragments are depicted in Fig. 5, showing attractive blue colored areas between the metals which are increasing from Ni to Pt. The NCI approach is also capable to unravel the weak non-covalent metallophilic interactions, however in the present case it is more difficult to distinguish the metallophilic interactions from other inter- and intramolecular interactions (Fig. S11–S13, ESI†)¹⁷ and therefore the IGMH method might be preferred in cases of weak interactions, being in the vicinity of other interactions.

Fig. 5 IGM based on a Hirshfeld partition of the molecular density of 2M (B3PW91/6-311+G(2df,p)). Each metal atom defines one fragment. IGMH iso-surfaces at s(r) = 0.01 colour coded with sign(λ2)ρ in a.u. Blue surfaces refer to attractive forces and red to repulsive forces. Green indicates weak interactions.

The hydrogen-rich dimethylamine borane, Me₂NHBH₃, undergoes thermally induced dehydrogenation at 130 °C to give quantitatively the aminoborane (Me₂NBH₂)₂ comprising a four-membered ring structure.¹⁸ While no reaction occurs below 100 °C, Manners et al. have shown that the dehydrogenation can be effectively catalyzed by addition of 0.5–5 mol% of [M(1,5-cod)(μ-Cl)]₂ (M = Rh, Ir; 1,5-cod = 1,5-cyclooctadiene) or RhCl₃·3H₂O within 24 h at 45 °C or even lower temperatures at longer reaction times.^19,20 Bimetallic complexes offer interesting opportunities for catalytic reactions.²¹ The interplay of two metal atoms in close proximity to each other often leads to cooperative effects and higher activities compared to single site catalysts. In a first foray, we investigated the catalytic dehydrogenation of Me₂NHBH₃ at 70 °C.

Typically, NMR scale experiments with 4 mol% of 2Ni, 2Pd and 2Pt were carried out in 1,2-difluorobenzene and the progress was monitored by recording ¹¹B NMR spectra. At the end of the reaction the spectrum showed almost exclusively a triplet centered at δ = 5.3 ppm with ¹J(¹¹B–¹H) coupling of 112 Hz that was unambiguously assigned to the cyclic (Me₂NBH₂)₂. The reaction proceeded three times faster upon using 2Ni (6 h) than 2Pt (18 h), whereas no reaction occurred in the presence of 2Pd (Scheme 3). Approximate turnover frequencies were determined as TOF = 4.16 h⁻¹ (2Ni) and 1.38 h⁻¹ (2Pt). The ³¹P NMR spectra of the reaction mixtures indicated that 2Ni (δ = 26 ppm) and 2Pt (δ = 29 ppm) changed into new yet unassigned species. In effort to further invest the kinetics of the dehydrogenation, the hydrogen gas evolution was monitored under the optimized reaction conditions for 2Ni and 2Pt (Fig. 6). The increase of the net pressure correlates well to the corresponding ¹¹B NMR data.

Scheme 3 Dehydrogenation of Me₂NHBH₃ catalysed by 2Ni and 2Pt in 1,2-difluorobenzene at 70 °C.

Fig. 6 Net pressure generation of the 2Ni (blue) and 2Pt (orange) catalyzed dehydrogenation of Me₂NHBH₃ in 1,2-difluorobenzene at 70 °C.

Again, 2Ni shows three times faster reaction times than 2Pt. After an initial sharp rise in pressure the reaction progressed approximately linear within the experiment. Notably, the use of 1·PtCl₂ also drives the cyclization, but is very slow (96 h).

In summary, the reaction of [2,6-(Ph₂P)₂C₆H₃]₂GaCl (1) with Ni(dme)Cl₂, Pd(MeCN)₂Cl₂ and Pt(MeCN)₂Cl₂, respectively, surprisingly proceeded with transmetallation and gave rise to the formation of cationic dinuclear group 10 complexes [M₂(PCP)₂μ-Cl][GaCl₄] (2M, M = Ni, Pd, Pt; PCP = 2,6-(Ph₂P)₂C₆H₃)] featuring a rare series of attractive metallophilic interactions between group 10 elements.²² The counterion [GaCl₄]⁻ was formed from the expelled Ga atom of 1. Only the reaction of 1 with Pt(Et₂S)₂Cl₂, afforded an expected Z-type complex, namely 1·PtCl₂. The dehydrogenation of Me₂NHBH₃ to give (Me₂NBH₂)₂, is catalysed upon addition of 2Ni and 2Pt, whereas 2Pd is inactive. Efforts are underway to broaden the scope of applications in catalysis and bond activation.

We are grateful to the Deutsche Forschungsgemeinschaft (DFG) for financial support. This work is dedicated to the memory of Professor Ian Manners.

Data availability

Spectroscopic and crystallographic data including figures of NMR spectra as well as the results of DFT calculations are given in the ESI.†

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. M. A. Bennett, S. K. Bhargava, N. Mirzadeh and S. H. Priver, Coord. Chem. Rev., 2018, 370, 69–128.
2. M. Olaru, J. F. Kögel, E. Lork, S. Mebs, M. Vogt and J. Beckmann, Chem. – Eur. J., 2020, 26, 275–284.
3. (a) A. L. Balch, Prog. Inorg. Chem., 1993, 41, 239–329; (b) E. S. Smirnova, J. M. Muñoz Molina, A. Johnson, N. A. G. Bandeira, C. Bo and A. M. Echavarren, Angew. Chem., Int. Ed., 2016, 55, 7487–7491.
4. (a) M. Olaru, E. Rychagova, S. Ketkov, Y. Shynkarenko, S. Yakunin, M. V. Kovalenkoc, A. Yablonsiky, B. Andreev, J. Beckmann and M. Vogt, J. Amer. Chem. Soc., 2020, 142, 373–381; (b) G. Smolentsev, C. J. Milne, A. Guda, K. Haldrup, J. Szlachetko, N. Azzaroli, C. Cirelli, G. Knopp, R. Bohinc, S. Menzi, G. Pamfilidis, D. Gashi, M. Beck, A. Mozzanica, D. James, C. Bacellar, G. F. Mancini, A. Tereshchenko, V. Shapovalov, W. M. Kwiatek, J. Czapla-Masztafiak, A. Cannizzo, M. Gazzetto, M. Sander, M. Levantino, V. Kabanova, E. Rychagova, S. Ketkov, M. Olaru, J. Beckmann and M. Vogt, Nat. Commun., 2020, 11, 2131.
5. F. Meyer, T. Kuzmera, E. Lork, M. Vogt, J. Beckmann and Z. Allgem, Anorg. Chem, 2021, 647, 1890–1895.
6. S. Furan, M. Molkenthin, K. Winkels, E. Lork, S. Mebs, E. Hupf and J. Beckmann, Organometallics, 2021, 40, 3785–3796.
7. (a) P. Buchwalter, J. Rosé and P. Braunstein, Chem. Rev., 2015, 115, 28–126; (b) E. Lu and S. T. Liddle, Group 10 Metal-Metal Bonds (Chapter 10) in Molecular Metal-Metal Bonds, Compounds, Synthesis, Properties. ed. S. T. Liddle, Wiley-VCH, Weinheim, Germany, 2015, 325–395.
8. (a) P. Pyykkö, Chem. Rev., 1997, 97, 597–636; (b) H. Schmidbaur and A. Schier, Chem. Soc. Rev., 2008, 37, 1931–1951; (c) M. J. Katz, K. Sakai and D. B. Leznoff, Chem. Soc. Rev., 2008, 37, 1884–1895; (d) S. Sculfort and P. Braunstein, Chem. Soc. Rev., 2011, 40, 2741–2760; (e) H. Schmidbaur and A. Schier, Chem. Soc. Rev., 2012, 41, 370–412; (f) H. Schmidbaur and A. Schier, Angew. Chem., Int. Ed., 2015, 54, 746–784.
9. M. D. Risi, W. Henderson and G. C. Saunders, Polyhedron, 2024, 255, 116992.
10. J. C. Hierso, Chem. Rev., 2014, 114, 4838–4867.
11. R. W. F. Bader, Atoms molecules. a quantum theory, Cambridge University Press, Oxford UK, 1991.
12. E. R. Johnson, S. Keinan, P. Mori-Sánchez, J. Contreras-García, A. J. Cohen and W. Yang, J. Am. Chem. Soc., 2010, 132, 6498–6506.
13. (a) C. Lefebvre, G. Rubez, H. Khartabil, J.-C. Boisson, J. Contreras-García and E. Hénon, Phys. Chem. Chem. Phys., 2017, 19, 17928–17936; (b) T. Lu, J. Comput. Chem., 2022, 43, 539–555.
14. A. S. Novikov, Inorg. Chim. Acta, 2018, 483, 21–25.
15. S. Furan, E. Lork, S. Mebs, E. Hupf and J. Beckmann, Z. Anorg. Allg. Chem., 2020, 646, 856–865.
16. R. Gericke, M. A. Bennett, S. H. Privér and S. K. Bhargava, Inorg. Chem., 2023, 62, 8846–8862.
17. S. Furan, M. Vogt, K. Winkels, E. Lork, S. Mebs, E. Hupf and J. Beckmann, Organometallics, 2021, 40, 1284–1295.
18. A. B. Burg and C. L. Randolph, J. Am. Chem. Soc., 1949, 71, 3451–3455.
19. C. A. Jaska, K. Temple, A. J. Lough and I. Manners, Chem. Commun., 2001, 962–963.
20. C. A. Jaska, K. Temple, A. J. Lough and I. Manners, J. Am. Chem. Soc., 2003, 125, 9424–9434.
21. (a) J. Campos, Nat. Rev. Chem., 2020, 4, 696–702; (b) M. Navarro, J. J. Moreno, M. Peréz-Jiménez and J. Campos, Chem. Commun., 2022, 58, 11220–11235.
22. O. Zheng, S. Borsley, G. S. Nichol, F. Duarte and S. L. Cockroft, Angew. Chem., Int. Ed., 2019, 131, 12747–12753.

---

Footnote

† Electronic supplementary information (ESI) available: M = Ni, Pd, Pt. PCP = 2,6-bis(diphenylphosphino)phenyl. CCDC 2378087–2378091. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4cc04296d

---