Search

Density functional theory (DFT) at the generalized gradient approximation (GGA) level has been applied for the analysis of the bond between group 10 metals and N-heterocyclic carbene (NHC) in complexes [MX3(NHC)] � (M = Ni, Pd, Pt, X = H, Cl, I). For comparative purposes, similar calculations have been performed for analogous pyridine complexes [MX3(py)] � (py = pyridine). Full geometry optimizations have been performed for all complexes. The role of the M–L p interaction was investigated by the aid of respectively, energy decomposition analysis, Hirshfeld atomic charge variation, molecular orbital considerations and bond order decomposition analysis. The p-bonding contribution increases in the order I < Cl < H, and Pt < Pd < Ni. Most significantly, the absolute p-acceptor ability of the NHC in these complexes is larger than that of pyridine. However, due to the dominant r donor interactions, the relative contribution, that is the p/r ratio, is predicted to be smaller.

The ground and excited-state properties of [Ru(bpy)(tpy)dmso] 2 + (bpy = 2,2'-bipyridine, tpy = 2,2':6',2"-terpyridine; dmso = dimethyl sulfoxide) have been studied by the means of density functional theory (DFT). In particular, the singlet ground state and the potential energy surface of the lowest triplet were investigated along the coordinate involved in the S → O linkage isomerization of dmso. The time-dependent-DFT approach (TDDFT) was used to interpret the absorption spectra of the system, while a ΔSCF procedure was applied to compute the emission spectra. The good agreement between computed and experimental spectra highlights the power of DFT approaches in the description of complex transition metal containing systems. In addition, this method allows the full description of the ground and excited potential energy surfaces of [Ru(bpy)(tpy)-dmso] 2 + which can only be roughly derived from experimental data, thus providing clues for further improvement in the engineering of phototriggering materials.

In many theoretical studies dealing with the selective complexation of trivalent actinides with respect to trivalent lanthanides, the method of calculation is assessed by comparing computed geometries with crystal structures that are often available. Yet, the selectivity is better rationalized through thermodynamic data, as enthalpy and entropy terms. In this article, we have theoretically modeled competing complexation reactions of [Ce(terpy)3]3+, [U(terpy)3]3+, [Ce(MeBTP)3]3+ and [U(MeBTP)3]3+ systems (terpy = 2,2′:6′2″-terpyridine; MeBTP = methyl-2,6-di(1,2,4-triazin-3-yl)pyridine) within the framework of the Density Functional Theory. Our calculations manage to qualitatively account for the experimental relative stabilities of terpy and MeBTP complexes, and in particular for the better coordinating strength of MeBTP for trivalent uranium. We also show by comparing the MeBTP ligand with its non-alkylated form (HBTP) that model systems often used in quantum chemistry must be carefully chosen when energetic comparisons are undertaken.

Experimental uncertainties concerning the coordination mode of trivalent plutonium in concentrated LiCl have led us to theoretically evaluate the f-f transitions of a series of rare earth aquo and chloro complexes. The calculation of Pr(III), U(III), Np(III), and Pu(III) systems' spectra was undertaken using the LFDFT (ligand field density functional theory) route that combines the backgrounds of ligand field (LF) theory with Kohn-Sham orbitals. LF parameters are fitted to previous DFT calculations, thus preventing the use of empirical data. The f-f transitions values are globally well predicted, but the lack of accurate experimental references can sometimes hinder reliable comparisons. Despite this, the nephelauxetic effect from aquo to chloro complexes is clearly observed through both spectral red shifts and the decrease in F2, the Slater-Condon parameter. Accordingly, this work provides the first theoretical characterization of covalency in trivalent f elements through their electronic spectra.

The structure and nature of the metal-metal bonding interaction in the cationic complexes [(eta6-C6Me6)2Ru2(mu2-H)3]+ (1), [(eta6-C6Me6)2Ru2(mu2-H)2(mu2-1,4-SC6H4Br)]+ (2), [(eta6-C6Me6)2Ru2(mu2-H)(mu2-1,4-SC6H4Br)2]+ (3), and [(eta6-C6Me6)2Ru2(mu2-1,4-SC6H4Br)3]+ (4) have been studied at the density functional theory (DFT) level using molecular orbital (MO) theory, bond order (BO) analysis, bond decomposition energy (BDE), electron localization function (ELF), and Laplacian of the density methods. The results show that there is no direct bond between the two ruthenium atoms in 1-4, the MO interaction within the diruthenium backbone being stabilized by the bridging ligands. For complex 1, the ELF clearly shows that the bond within the diruthenium backbone is through the three bridging hydride ligands, which act as a sort of glue by forming three-center two-electron bonds characterized by (Ru, H, Ru) basins with 1.8 e mostly located in the H atomic basin.

Recently, linear scaling construction of the periodic exact Hartree-Fock exchange matrix within the Γ-point approximation has been introduced [J. Chem. Phys. 122, 124105 (2005)]. In this article, a formalism for evaluation of analytical Hartree-Fock exchange energy gradients with respect to atomic positions and cell parameters at the Γ-point approximation is presented. While the evaluation of exchange gradients with respect to atomic positions is similar to those in the gas phase limit, the gradients with respect to cell parameters involve the accumulation of atomic gradients multiplied by appropriate factors and a modified electron repulsion integral (ERI). This latter integral arises from use of the minimum image convention in the definition of the Γ-point Hartree-Fock approximation. We demonstrate how this new ERI can be computed with the help of a modified vertical recurrence relation in the frame of the Obara-Saika and Head-Gordon-Pople algorithm. As an illustration, the analytical gradients have been used in conjunction with the QUICCA algorithm [K. Németh and M. Challacombe, J. Chem. Phys. 121, 2877 (2004)] to optimize periodic systems at the Hartree-Fock level of theory.

Energetics and the electronic structure of various types of single-walled carbon nanotubes have been investigated by using Density Functional Theory. Armchair [n,n], zigzag [n,0] and chiral [n,m] C40H20 nanotubes have been considered. Calculations show that the armchair isomer is the most stable among the three types and they further reveal the factors that stabilize this isomer. Nucleus-independent chemical shift calculations indicate the aromaticity of the individual hexagonal rings in the carbon nanotubes and explain the extent of electron delocalization in them.

The research activity within our laboratory of computational chemistry at the University of Fribourg is presented. In this review, a brief outline of a recently proposed Ligand Field Density Functional Theory (LFDFT) model for single nuclear and its extension to dimer transition metal complexes is given. Applications of the model to dinuclear complexes are illustrated for the interpretation of exchange coupling in the bis-?-hydroxo-bridged dimer of Cu(II) and to the description of the quadruple metal-metal bond in Re2Cl82?. The analysis of the chemical bonding is compared with results obtained using other approaches, i.e. the Extended Transition State model and the Electron Localization Function. It is shown that the DFT supported Ligand Field Theory provides consistent description of the ground and excited state properties of transition metal complexes.

The Ligand Field (LF) model in molecular science or the Crystal Field model in condensed matter science has been introduced more than eighty years ago. Since then, this theory plays a central role each time that molecules containing d- and/or f-elements with open shells are adressed. No doubt, this fact is related to the dominant localisation of the frontier orbitals within the metal-sphere. This common feature enables us to describe approximately the electronic structure of these molecules using orbitals that are centred on a single atom and to treat their interaction with the chemical environment essentially as a perturbation. Another reason for the success of this simple theory is the fact that the more accurate molecular orbital theory does generally over-estimate covalence of transition metal atoms and thus yields wave functions that are too delocalized. We give here a survey of the development of LF theory since its introduction simultaneously by Hans Bethe and John Hasbrouck van Vleck more than eighty years ago. Over the years, LF theory was a semi-empirical model with adjustable parameter until the end of last century when we introduced non-empirical LF theory that is based on DFT calculations. The results of this first-principle prediction are in very good agreement with the experimental observations. Sample calculations of tetrahedral and octahedral Cr-complexes and hexa-acquo Ni(II)-complexes are used to validate the new model and to analyse the calculated parameters