What is the "best" atomic charge model to describe through-space charge-transfer excitations?

CEISAM, UMR CNRS 6230, BP 92208, Université de Nantes, 2, Rue de la Houssinière, 44322 Nantes, Cedex 3, France.
Physical Chemistry Chemical Physics (Impact Factor: 4.49). 03/2012; 14(16):5383-8. DOI: 10.1039/c2cp40261k
Source: PubMed


We investigate the efficiency of several partial atomic charge models (Mulliken, Hirshfeld, Bader, Natural, Merz-Kollman and ChelpG) for investigating the through-space charge-transfer in push-pull organic compounds with Time-Dependent Density Functional Theory approaches. The results of these models are compared to benchmark values obtained by determining the difference of total densities between the ground and excited states. Both model push-pull oligomers and two classes of "real-life" organic dyes (indoline and diketopyrrolopyrrole) used as sensitisers in solar cell applications have been considered. Though the difference of dipole moments between the ground and excited states is reproduced by most approaches, no atomic charge model is fully satisfactory for reproducing the distance and amount of charge transferred that are provided by the density picture. Overall, the partitioning schemes fitting the electrostatic potential (e.g. Merz-Kollman) stand as the most consistent compromises in the framework of simulating through-space charge-transfer, whereas the other models tend to yield qualitatively inconsistent values.

Download full-text


Available from: Denis Jacquemin, Aug 26, 2014
  • Source
    • "Since the definition of partial atomic charges is not strict, various models of partial atomic charges can, however, differ significantly in the reliability of their predictions. Therefore, an evaluation of the performance of a partial atomic charge model for a given problem is necessary before using it for making reliable predictions232425262728. In this work, various models of partial atomic charge are tested to establish their efficiency for computations on diheteroaryl ketones and thioketones. "

    Preview · Article · Jan 2016
  • Source
    • "All optimizations were calculated without any symmetry constraints using 6-31G(d,p) basis set on Gaussian09 software package [40]. In order to investigate partial atomic charge, the Merz–Singh–Kollman [41] approach with B3LYP/6- 31G(d,p) was employed. The vertical excitation energy and electronic absorption spectra were simulated using TDDFT with B3LYP, BHandHlyp, PBE0, and CAM-B3LYP hybrid functionals. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The geometries and electronic structures of organic dye sensitizers, CCT1A, CCT2A, CCT3A, CCT1PA, and CCT2PA comprising double-donor groups, π-spacer, and acceptor group forming D–D–π–A system, were studied using DFT and TDDFT. The calculated results have shown that TDDFT calculation using a newly-designed functional which takes into long-range interaction, CAM-B3LYP, was reasonably capable of predicting the excitation energies and the absorption spectra of the molecules. The adsorption of these dyes on the TiO2 anatase (1 0 1) surface and the electron injection mechanism were also investigated using a dye-(TiO2)38 cluster model, employing PBE and TD-CAM-B3LYP calculations, respectively. The adsorption energy (Eads) of CCTnA (n = 1–3) was calculated to be −15.26, −18.93, and −20.12 kcal/mol respectively, indicating strong adsorption of dye to a TiO2 surface by carboxylate groups. These calculated results suggested that the CCT3A is a promising candidate for highly efficient DSSCs. It was shown that the electron injection mechanism occurs by direct charge-transfer transition in a dye-TiO2 interacting system, resulted in the stronger electronic coupling strengths of the anchoring group of the dyes and the TiO2 surface which corresponded to higher observed Jsc as expected in CCT3A dye. Through a combined theoretical and experimental investigation we have shown that the trend of charge-injection efficiency in dye-sensitized solar cells constituted from dyes is determined by the adsorption energy of dye-(TiO2)38 complexes.
    Full-text · Article · Mar 2013 · Organic Electronics
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Anthocyanidins are a class of π-conjugated systems responsible for red, blue, and purple colors of plants. They exhibit the capacity of aggregation in the presence of other natural compounds including flavonols. Such complexations induce color modulation in plants, which is known as copigmentation. It is largely driven by π-interactions existing between pigments and copigments. In this work, the energies of copigmentation–complexation and self-association are systematically evaluated for an anthocyanidin/flavonol couple prototype (3-O-methylcyanidin/quercetin). To describe noncovalent interactions, DFT-D appears mandatory to reach a large accuracy. Due to the chemical complexity of this phenomenon, we also aim at assessing the relevance of both B3P86-D2 and ωB97X-D functionals. The benchmarking has shown that B3P86-D2 possesses enough accuracy when dealing with π–π interactions with respect to both spin component scaled Møller–Plesset second-order perturbation theory post Hartree–Fock method and experimental data. UV–vis absorption properties are then evaluated with time-dependent DFT for the different complexes. The use of range-separated hybrid functionals, such as ωB97X-D, helped to correctly disentangle and interpret the origin of the UV–vis experimental shifts attributed to the subtle copigmentation phenomenon.
    Full-text · Article · May 2012 · Journal of Chemical Theory and Computation
Show more