Two-electron oxidation of the quadruply bonded paddlewheel compounds M2(dpa)4 (dpa = 2,2′-dipyridylamido; M2 = Cr2, Mo2, MoW, W2) either severs the metal–metal bond entirely (Cr2: two antiferromagnetically coupled Cr(III) centers) or leaves a diamagnetic, heavily stretched, multiconfigurational σ²π²δ² metal–metal linkage (Mo2, MoW, W2). Further oxidation with an oxygen-atom donor yields unusual linear M···M≡O oxo dications whose metal–metal separations exceed single-bond covalent radii and whose reactions with substrates such as trialkylphosphines have been reported to proceed by single-electron (SET/radical) pathways rather than classical two-electron oxygen-atom transfer (OAT). What remains open is a quantitative account of the structural, electronic, and environmental factors that set the boundary between bond retention and cleavage, the factors that select between OAT and SET manifolds in the bimetallic oxo species, and a converged electronic-structure description of residual metal–metal bonding in the borderline stretched regime.

Background

Dinuclear compounds of the group-6 metals can carry a metal–metal quadruple bond, conventionally described by the orbital filling σ²π⁴δ² (one σ, two π, and one δ bonding combination of the metal d orbitals, occupied by four bonding electron pairs); classic examples are the "paddlewheel" complexes M2X4 or M2(L–N,N')4 in which four bridging ligands wrap around the M2 unit. A standard supporting ligand is dpa⁻, the deprotonated 2,2′-dipyridylamide, which binds through an amido nitrogen and can additionally chelate or leave pendant a pyridyl donor, so that the ligand set can reorganize when the electron count of the metal core changes. Oxidizing a quadruply bonded M2(dpa)4 core removes electrons from the δ/δ* manifold; naive electron counting predicts a triply bonded σ²π⁴ dication with only slight bond elongation. Brogden et al. (Inorg. Chem. 2014, 53, 4777–4790) showed experimentally that reality is stranger: for M2 = Mo2, MoW, and W2, the two-electron-oxidized dications [M2(dpa)4]²⁺ retain diamagnetic, highly stretched metal–metal separations (2.7156–2.7981 Å, elongated by more than 0.5 Å relative to the neutral quadruple bonds and far beyond normal triple-bond lengths), while the Cr2 analogue undergoes complete Cr–Cr bond cleavage (3.197 Å), leaving two essentially independent Cr(III) centers that couple antiferromagnetically with exchange constant J = −75(2) cm⁻¹ from fitting the temperature-dependent magnetic susceptibility to a Heisenberg spin Hamiltonian (the standard exchange-coupling model used to extract the magnetic coupling constant J between two metal-center spins from susceptibility data). Multiconfigurational calculations — complete-active-space self-consistent field followed by second-order perturbation theory (CASSCF/CASPT2), here with six electrons in six M–M bonding/antibonding orbitals — describe the retained Mo/W bonds as strongly multiconfigurational singlets dominated by a σ²π²δ² configuration (weights 24–42%) with effective bond orders (EBOs, a CAS-based bond-order measure) of 1.01–1.59, i.e., "stretched" partial triple bonds. Treating the dications of Mo2, MoW, and W2 with m-chloroperoxybenzoic acid as an oxygen-atom donor effects a further two-electron oxidation plus oxygen-atom transfer (OAT, the transfer of a single O atom, a formal two-electron event) to give the linear M···M≡O oxo dications [M2O(dpa)4]²⁺, in which one metal bears a short terminal M≡O bond (≈1.66–1.70 Å, EBO ≈ 2.85) while the M···M separation (3.07–3.13 Å) exceeds the sum of single-bond covalent radii; calculations find only minimal residual M–M interaction. The reactivity of these oxo species is unusual: in earlier work on [W2O(dpa)4]²⁺ (JACS 2011), excess tri-tert-butylphosphine in acetonitrile regenerates the quadruply bonded W2(dpa)4 by an overall four-electron reduction described as including oxygen-atom transfer and ascribed to the cooperation of W–O and W–W multiple bonding; mechanistic studies of this reaction proposed that the net transformation proceeds through a single-electron-transfer (SET) manifold with phosphinyl radical intermediates — a solvent kinetic isotope effect (faster in CD3CN by ≈2.1) and cyclic voltammetry argued for rate-limiting outer-sphere electron transfer — in place of the classical clean two-electron OAT to phosphine oxide known for mononuclear metal-oxo complexes. The one-electron oxidized trication [W2O(dpa)4]³⁺ (Chem. Commun. 2015) was also reported to react with phosphines and sulfides by SET, and its geometry and EPR spectrum (unpaired electron localized on the oxo-bound W(V) center) proved strongly dependent on the chosen density functional, exposing the fragility of single-determinant density functional theory (DFT) for these weakly metal–metal-bonded, metal-oxo systems. These observations define a coherent set of unresolved questions at the intersection of metal–metal bonding and multielectron redox chemistry: where exactly the oxidation-driven transition from stretched-but-bonded to cleaved cores lies, what makes a proximate second metal turn an oxo into a one-electron oxidant instead of an oxygen-atom donor, and whether this "two inorganic functional groups in one molecule" design can be exploited catalytically.

Problem Statement

For the group-6 dipyridylamido dimers M2(dpa)4 (dpa = 2,2′-dipyridylamido; M2 = Cr2, Mo2, MoW, W2) and their oxidized derivatives, determine, quantitatively and mechanistically: (1) the structural, electronic, and environmental factors that decide whether two-electron oxidation of the quadruply bonded core yields a retained but heavily stretched, diamagnetic, multiconfigurational metal–metal bond (as for Mo2, MoW, W2: σ²π²δ² ground states, separations 2.72–2.80 Å) or complete bond cleavage to weakly coupled M(III) centers (as for Cr2: 3.197 Å, J = −75 cm⁻¹), including how this boundary depends on metal identity across the homologous Cr/Mo/W series; and (2) for the linear M···M≡O oxo dications [M2O(dpa)4]²⁺ and their one-electron oxidized trications, the factors that select between classical two-electron oxygen-atom transfer to a substrate and one-electron (SET/radical) oxidation of the substrate — as observed with trialkylphosphines — together with the elementary steps, intermediates, and kinetics of both manifolds. A convergent quantitative electronic-structure description of the residual metal–metal bonding in the borderline stretched regime is an integral part of the objective, since it underpins the quantitative interpretation of (1) and (2). Treat the objective as resolved when it enumerates and ranks the structural, electronic, and environmental factors that decide the retained-bond-versus-cleavage outcome and maps that ranking onto the Cr/Mo/W series; enumerates the elementary steps and rate-limiting features that select between the OAT and SET/radical manifolds for the oxo dications and trications and states which manifold is predicted to dominate under the cited conditions; and reports, for the residual metal–metal bonding of the [M2(dpa)4]²⁺ dications (M2 = Mo2, MoW, W2), a bounded target quantity — the effective bond order (EBO) computed within a stated CASSCF/CASPT2 active space and basis set — together with the active-space and basis-set convergence behaviour of that quantity across the three M–M pairs; any factor that cannot be ranked, or any electronic-structure quantity that cannot be converged within the stated accuracy, must be reported as such with the evidence of non-convergence rather than replaced by an unquantified description.

References

  1. Brogden, D. W.; Turov, Y.; Nippe, M.; Li Manni, G.; Hillard, E. A.; Clérac, R.; Gagliardi, L.; Berry, J. F. Oxidative Stretching of Metal–Metal Bonds to Their Limits. Inorg. Chem. 2014, 53, 4777–4790. DOI 10.1021/ic5007204. https://doi.org/10.1021/ic5007204
  2. Nippe, M.; Victor, E.; Berry, J. F. Oxidation Chemistry of Axially Protected Mo2 and W2 Quadruply Bonded Compounds. Inorg. Chem. 2009, 48, 11889–11895. DOI 10.1021/ic901965b. https://doi.org/10.1021/ic901965b
  3. Nippe, M.; Goodman, S. M.; Fry, C. G.; Berry, J. F. Chemically Reversible Four-Electron Oxidation and Reduction Utilizing Two Inorganic Functional Groups. J. Am. Chem. Soc. 2011, 133, 2856–2859. DOI 10.1021/ja1114248. https://doi.org/10.1021/ja1114248
  4. Brogden, D. W.; Berry, J. F. Not all density functionals are created equal: the case of the missing electron in the oxidized [W–W≡O]⁷⁺ core. Chem. Commun. 2015, 51, 9153–9156. DOI 10.1039/c5cc02917a. https://doi.org/10.1039/c5cc02917a
  5. Brogden, D. W.; Berry, J. F. Heterometallic Second-Row Transition Metal Chain Compounds in Two Charge States: Syntheses, Properties, and Electronic Structures of [Mo–Mo–Ru]⁶⁺/⁷⁺ Chains. Inorg. Chem. 2015, 54, 7660–7665. DOI 10.1021/acs.inorgchem.5b01370. https://doi.org/10.1021/acs.inorgchem.5b01370
  6. Dolinar, B. S.; Berry, J. F. Influence of Lewis acid charge and proximity in Mo Mo⋯M linear chain compounds with M = Na⁺, Ca²⁺, Sr²⁺, and Y³⁺. Polyhedron 2016, 103, 71–78. DOI 10.1016/j.poly.2015.09.028. https://doi.org/10.1016/j.poly.2015.09.028
  7. Corcos, A. R.; Pap, J. S.; Yang, T.; Berry, J. F. A Synthetic Oxygen Atom Transfer Photocycle from a Diruthenium Oxyanion Complex. J. Am. Chem. Soc. 2016, 138, 10032–10040. DOI 10.1021/jacs.6b05942. https://doi.org/10.1021/jacs.6b05942
  8. Berry, J. F.; Roy, M. 2-Aminopyridine and Related Ligands to Support Metal–Metal Bonded Compounds. In Comprehensive Coordination Chemistry III, 2021, pp 406–427. DOI 10.1016/B978-0-08-102688-5.00075-1. https://doi.org/10.1016/B978-0-08-102688-5.00075-1