| Abstract
| - A theoretical analysis of known and hypothetical doubly bridged Cu(I)−Cu(I) dimers and other d10−d10 analogues is presented. DFT/B3LYP calculations as well as a qualitative approach are used to describe different stereochemistries, distortional trends, and direct M−M interaction in frameworks of the type LnCu(μ-X)2CuLm (X = H, halide, CO, acetylide; L = two-electron-donor ligand; m = 1, 2; n = 1, 2). The greatest attention is devoted to the few complexes with acetylide bridges, which are classified in different categories.
- DFT/B3LYP calculations on known and hypothetical doubly bridged Cu(I)−Cu(I) dimersand other d10−d10 analogues have been carried out. The bridging ligands may be only σdonors (hydrides) or have added π-donor (halides) or π-acceptor (carbonyls, as yet unknown)capabilities. In particular, the few reported LnCu(μ-C⋮CR)2CuLm frameworks have beeninvestigated. The latter are symmetric (type c) or asymmetric (types a and b), dependingon the nature and number of terminal ligands (n = 1, 2; m = 1, 2). Beside the accurategeometric and energetic computations, the nature of the chemical bonding is explored interms of perturbation theory arguments (EHMO approach). Thanks to the σ donor power ofthe bridges, electron density is driven into the bonding combinations (σ and π) of emptymetal s and pπ orbitals. In the presence of π-donor ligands, population of the correspondingσ* and π* levels occurs and the M−M bond vanishes. In contrast, insufficient back-donationfrom copper d orbitals prevents the formation of bridged carbonyl dimers and trigonal-planarmonomers are favored. A case study is that of the heterobinuclear d10−d10 complex (CO)2Cu(μ-CO)2Co(CO)2, where the lone pairs of the CO bridges are preferentially directed towardcobalt for electronegativity reasons. A similar situation is highlighted for the model (PH3)2Cu(μ-C⋮CH)2Cu(PH3) (type b), where both bridges orient toward the unique fragment(PR3)Cu because of the different hybridization of L2M and LM σ orbitals. In the speciesLnCu(μ-C⋮CH)2CuLn (n = 2 or n = 1, type a or c), the potential energy surface for thesymmetric to asymmetric rearrangement of the central Cu2C2 ring is quite flat. However, asymmetric Cu2(μ-C⋮CR)2 framework is achieved with η2-bound alkynes (type c). This isattributable to the π* levels of the latter ligands, which stabilize the metal pπ orbitals involvedin bridge bonding. The asymmetric Cu2C2 arrangement is preferred again in models wherethe terminal alkynes are substituted for by single phosphine ligands.
|
