Significant progress has been made in the past 10–15 years on the design, synthesis, and properties of multimetallic coordination complexes with heterometallic metal–metal bonds that are paramagnetic. Several general classes have been explored including heterobimetallic compounds, heterotrimetallic compounds of either linear or triangular geometry, discrete molecular compounds containing a linear array of more than three metal atoms, and coordination polymers with a heterometallic metal–metal bonded backbone. We focus in this Review on the synthetic methods employed to access these compounds, their structural features, magnetic properties, and electronic structure. Regarding the metal–metal bond distances, we make use of the formal shortness ratio (FSR) for comparison of bond distances between a broad range of metal atoms of different sizes. The magnetic properties of these compounds can be described using an extension of the Goodenough–Kanamori rules to cases where two magnetic ions interact via a third metal atom. In describing the electronic structure, we focus on the ability (or not) of electrons to be delocalized across heterometallic bonds, allowing for rationalizations and predictions of single-molecule conductance measurements in paramagnetic heterometallic molecular wires.

The new heterometallic chain compounds Mo Ni(dpa) Cl (1) and [Mo Ni(dpa) Cl ]OTf (2) (dpa=2,2'-dipyridylamine) have been prepared and studied by crystallography and magnetic susceptibility, among other methods. Oxidation of 1 to 2 removes an electron from the multiply bonded Mo unit, consistent with the formulation of 2 containing a (Mo ) ⋅⋅⋅(Ni) core. While 1 contains an S=1, pseudo-octahedral Ni ion, 2 has an S=3/2 ground state, in which the two Ni unpaired electrons, one in a localized δ-orbital and one in a heavily delocalized σ -orbital are joined by an unpaired electron in a Mo-Mo δ-orbital. The S=3/2 ground state is persistent to 300 K, evidencing strong ferromagnetic coupling of the Mo and Ni spins with J≥150 cm . This ferromagnetic interaction occurs via delocalization of a σ -electron across all three metal atoms, forcing ferromagnetic alignment of electrons in orthogonal Ni and Mo δ-symmetry orbitals. We anticipate that this new means of coupling spins can be used as a design principle for the preparation of new compounds with high spin ground states.

Clean axial ligand substitution reactions of heterometallic extended metal atom chains (HEMACs) supported by the dpa ligand (dpa = 2,2'-dipyridylamine) have been synthetically challenging due to side reactions that alter the trimetallic core. Following the hypothesis that a heterometallic core containing second-row transition metals would be more robust toward ligand substitution, we report the synthesis of three new heterotrimetallic compounds, MoNi(dpa)(OTf) (1), MoNi(dpa)(NCS) (2), and MoNi(dpa)(NCSe) (3) that are obtained cleanly and in good yield. Compound 1 may be synthesized either directly by reaction of Ni(OTf) with Mo(dpa) (4) or indirectly, by reaction of MoNi(dpa)Cl (5) with 2 equiv of TlOTf. Axial ligand substitution on 1 via solutions containing NaNCS or KNCSe afford 2 or 3, respectively. X-ray crystal structures of 1, 2, and 3 present short Mo-Ni distances of 2.458(8)Å /2.47(1) Å, 2.548(1), and 2.546(1), respectively. Density functional theory (DFT) calculations indicate a 3-center 3-electron σ bonding interaction between the Mo quadruply bonded core and the Ni in both 1 and 2. These complexes were analyzed by SQUID magnetometry, supporting the presence of a high spin Ni center with S = 1.

Understanding the fundamental properties governing metal− metal interactions is crucial to understanding the electronic structure and thereby applications of multimetallic systems in catalysis, material science, and magnetism. One such property that is relatively underexplored within multimetallic systems is metal−metal bond polarity, parameterized by the electronegativities (χ) of the metal atoms involved in the bond. In heterobimetallic systems, metal−metal bond polarity is a function of the donor−acceptor (Δχ) interactions of the two bonded metal atoms, with electropositive early transition metals acting as electron acceptors and electronegative late transition metals acting as electron donors. We show in this work, through the preparation and systematic study of a series of Mo 2 M(dpa) 4 (OTf) 2 (M = Cr, Mn, Fe, Co, and Ni; dpa = 2,2′dipyridylamide; OTf = trifluoromethanesulfonate) heterometallic extended metal atom chain (HEMAC) complexes that this expected trend in χ can be reversed. Physical characterization via single-crystal X-ray diffraction, magnetometry, and spectroscopic methods as well as electronic structure calculations supports the presence of a σ symmetry 3c/3e − bond that is delocalized across the entire metal-atom chain and forms the basis of the heterometallic Mo 2 −M interaction. The delocalized 3c/3e − interaction is discussed within the context of the analogous 3c/3e − π bonding in the vinoxy radical, CH 2 CHO. The vinoxy comparison establishes three predictions for the σ symmetry 3c/3e − bond in HEMACS: (1) an umpolung effect that causes the Mo−M interactions to become more covalent as Δχ increases, (2) distortion of the σ bonding and non-bonding orbitals to emphasize Mo−M bonding and de-emphasize Mo−Mo bonding, and (3) an increase in Mo spin population with increasing Mo−M covalency. In agreement with these predictions, we find that the Mo 2 •••M covalency increases with increasing Δχ of the Mo and M atoms (Δχ Mo−M increases as M = Cr < Mn < Fe < Co < Ni), an umpolung of the trend predicted in the absence of σ delocalization. We attribute the observed trend in covalency to the decreased energic differential (ΔE) between the heterometal d z 2 orbital and the σ bonding molecular orbital of the Mo 2 quadruple bond, which serves as an energetically stable, "ligand"-like electron-pair donor to the heterometal ion acceptor. As M is changed from Cr to Ni, the σ bonding and nonbonding orbitals do indeed distort as anticipated, and the spin population of the outer Mo group is increased by at least a factor of 2. These findings provide a predictive framework for multimetallic compounds and advance the current understanding of the electronic structures of molecular heteromultimetallic systems, which can be extrapolated to applications in the context of mixed-metal surface catalysis and multimetallic proteins.

A one‐electron‐oxidized Mo2Ni(dpa)4Cl2 complex that displays strong ferromagnetic coupling between the two terminal metal atoms is depicted on the frontispiece and is represented as people talking on the phone about the physical characterization methods used to determine an S=3/2 ground state. The cat in between is not actively communicating, but is thinking about density functional theory calculations also used to support the experimental findings. The findings suggest a new means of spin coupling that may be valuable for the design of high‐spin molecules. For more information please see the Communication by J. A. Chipman and J. F. Berry on page 1494 ff.