Binuclear Cobalt Paddlewheel-Type Complexes: Relating Metal–Metal Bond Lengths to Formal Bond Orders

Abstract

Paddlewheel-type complexes are prominent among experimentally known binuclear cobalt complexes and incorporate substituted formamidinate, guanidinate, and carboxylate ligands in digonal, trigonal, and tetragonal arrays around the bimetallic core. Such complexes are modeled here by density functional theory using unsubstituted ligands, extending the whole set to incorporate a variety of metal oxidation states and spin multiplicities. The DFT results for ground state cobalt–cobalt bond lengths and ground state spin multiplicity of the model complexes are often quite close to the experimental results for the corresponding substituted complexes. The three series of complexes often exhibit parallel trends with regard to effects of change in the metal oxidation state and spin multiplicity. The formamidinate and guanidinate series show marked resemblances. The lowered symmetry in many model trigonal complexes implies that such deviations in the experimentally known congeners arise from the inherent electronic structure. For ground state species, the DFT results provide Co–Co bond orders (BOs) from MO occupancy considerations. Further, using a revised electron bookkeeping method, Co–Co formal bond order (fBO) values from 0.0 to 2.0 are assigned to all of the 85 complexes studied. The computed Co–Co bond lengths fall into distinct ranges according to the formal bond order values (from 0.5 to 2)

Lantern‐Type Divanadium Complexes with Bridging Ligands: Short Metal−Metal Bonds with High Multiple Bond Orders

“…This shortening of metalÀ metal bond lengths is apparently a feature of many DFT results, as seen in some recent studies on various binuclear lantern-type complexes. [36][37][38][39]…”

“…DFT studies on other paddlewheel-type complexes (dicobalt, dimanganese, diiron and dichromium) reveal the same disparity between experimental and DFT-derived metalÀ metal bond distances for the more highly correlated triple, quadruple and quintuple metalÀ metal bonds owing to overestimation of electron correlation by DFT. [36][37][38][39] 3. Conclusions 1.…”

Binuclear Lantern Complexes of All First-Row 3d Metals Scandium to Zinc: Metal–Metal Bond Lengths and Bond Orders, with Measures of Intrinsic Metal–Metal Bond Strength

“…This density functional theory (DFT) study was applied to homobimetallic lantern-type complexes (digonal, trigonal, and tetragonal) of all the first-row 3d metals scandium to zinc. Lantern binuclear complexes of the metals vanadium to nickel had been investigated earlier by DFT using the LAN2LDZ basis set. − The current study employs two functionals with the improved cc-pVTZ basis set. This study also includes the “outer” 3d metals–scandium and titanium on one side, and copper and zinc on the other.…”

“…Experimentally characterized dicobalt lantern-type complexes include tetragonal, trigonal, and digonal complexes, besides numerous axially coordinated tetracarboxylates . Other noncarbonyl dicobalt complexes with diverse ligands are also known …”

Lantern‐type dinickel complexes: An exploration of possibilities for nickel–nickel bonding with bridging bidentate ligands

“…Ni Ni bonds of order 0.5 are associated with a shorter R NiNi range of 2.433 ± 0.076 Å, along with an increase in the W NiNi range (0.34 ± 0.05).Ni Ni single bonds (fBO = 1) in non-digonal complexes display the range 2.404 ± 0.099 Å, which is somewhat below the value of 2.48 Å for a Ni Ni single bond on the basis of the estimate of 1.24 Å for the nickel covalent radius 12. This may point to a tendency of the M06-L functional to underestimate metal metal bond lengths, as noted for divanadium, dichromium, dimanganese, diiron, and dicobalt lantern-type complexes [39][40][41][42][43]. It could also arise from the derivation of the nickel covalent radius of 1.24 Å from a database containing a large number of carbonyl-containing complexes, for which the metal metal bond lengthening effect of the carbonyl ligands operates.…”