Our objectives are: to model and understand observed properties of molecules and materials in terms of their atomic structures, electronic states and energy levels; to use this understanding to design new materials with specific properties; to guide and stimulate new experiments and synthesis techniques. We are particularly interested in: surface and interface systems, e.g. organic monolayers on metal surfaces or solid-liquid interfaces between water or aqueous solutions and metal oxide surfaces; understanding the relation between structure, electronic properties and (photo)reactivity; photovoltaics (conversion of sunlight to electrical power), photocatalysis (light-assisted catalytic reactions) and, generally, electron-transfer processes at the interface between a solid material and an adsorbed molecular species.

The primary tool of our studies is the First-Principles Molecular Dynamics (FPMD) approach, which combines an ab-initio quantum-chemical method with reliable predictive capabilities, namely Density Functional Theory (DFT), with Molecular Dynamics. This allows us not only to investigate static structures, but also to access finite temperature dynamical properties. Efficient and reliable techniques like the "string method" (combined with FPMD) are used to determine reaction pathways and transition states. Time Dependent DFT (TDDFT) is used to study the response of photo-excited systems, e.g. optical absorption spectra. The use of hybrid functional is explored to deal with situations where self-interaction effects are important (e.g. defects in oxide materials).

Recent studies include the determination of reaction pathways for the attachment of organic molecules to silicon surfaces; characterizing the structure and reactivity of titania nanoparticles towards various molecular probes (photocatalysis); investigating the electron injection mechanism from a photo-excited dye sensitizer to a TiO2 nanoparticle (useful for understanding the functioning of photoelectrochemical cells); determining the electronic structure of metal-molecule-metal junctions, useful for understanding electron transport through organic monolayers and molecular devices.

Towards a molecular scale understanding of Photocatalysis on Metal Oxides.

Heterogeneous photocatalysis on semiconducting metal oxides is a promising, yet often inefficient, solution for many energy-related processes. It allows using sunlight for the destruction of highly toxic molecules and remediation of pollutants; for the selective, synthetically useful redox transformations in specific organic compounds; the production of hydrogen, and the conversion of solar energy to electric power. UV-irradiation of TiO2 results in the excitation of an electron-hole pair. The excited species can trigger oxidation/reduction reactions at the surfaces, often through formation of radicals from adsorbed water, hydroxyls, or oxygen. Electron-hole recombination competes with photocatalytic reactions and can be prevented by fast trapping of the excited species at special surface sites or adsorbates. A matching of the material’s band gap to the solar spectrum and a specific tailoring of surface trapping sites is key to making the overall process more efficient and selective. In general, photocatalytic systems are very complex and do not allow a definite conclusion of the relative role of the different radical species, trapping sites, the influence of defects and impurities, the role of the crystallographic phase and/or orientation, and parameters such as particle size and shape. An increased molecular-level understanding of photocatalytic processes could help tailor systems for specific applications and increase their overall efficiency.

In collaboration with the experimental group of Prof. Ulrike Diebold at Tulane University, we are conducting a combined experimental and theoretical study on well- characterized model systems, where the influence of these various factors can be singled out. We mainly focus on TiO2, the most promising photocatalyst. We consider single crystals with different orientations, modifications and dopants, as well as nanoscopic TiO2 clusters on various substrates. We use a first principles approach based on Density Functional Theory, whose predictive capabilities for the systems of interest have already been verified. Determination of the surface structures and molecular adsorption geometries requires extensive exploration of potential energy surfaces, for which we employ ab-initio molecular dynamics (MD) techniques. MD simulations are carried out to test the stability of the calculated structures and study finite temperature effects. Electronic properties, including charge and spin density distributions, electrostatic potential maps, projected densities of states, STM images, etc. are studied. Reaction pathways and barriers are determined using an extension of the ab- initio MD that has been implemented by our group. For problems involving electronic excitation, Time Dependent DFT (TDDFT) methods is used.

Project funded by DOE.

Organic monolayers on and between metal surfaces

Charge transport across nanometer scale metal-molecule-metal junctions is one of the central themes in the rapidly developing field of molecular electronics. Experiments and theoretical studies have identified two main conduction mechanisms in these systems. In one case, the molecule bridging the metal electrodes has states in resonance with the electrode states near the Fermi energy Ef, leading to electron transport similar to that in conventional metals (at small voltages). In the second mechanism, Ef is located between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO): although no resonant state is present, the exponential tail of the HOMO or LUMO at Ef, broadened by the interaction between the molecule and the electrodes, opens a non-resonant tunneling transport channel.
As a first step toward a better understanding of charge transport in molecular systems, we have recently investigated the electronic properties of various metal-molecule-metal junctions under zero bias, including the prototypical Au(111)/CnS2/Au(111) system of n-alkanedithiols, CnS2 (n=4,8,12), between two flat (111) gold electrodes, with chemical S-Au bonds at both ends, and molecule-tip contacts formed by conjugated and mixed saturated-conjugated molecules adsorbed on a Au(111) substrate. For Au(111)/CnS2/Au(111), our results show that the LDOS at Ef is mainly determined by the number of methylene units, with a much weaker dependence on the separation between the electrodes, thus providing support to the idea that for these systems the tunneling is mainly through bonds.
Future work will include the use of a newly developed theoretical approach to study the current flow in metal-molecule-metal junctions at finite voltages. We are also interested in studying the effect of water and possible contaminants on the electronic and transport properties of the molecular junction.

Project supported by NSF through Grant DMR02-13706 to the MRSEC- Princeton Center for Complex Materials.

Organic Functionalization of Semiconductor Surfaces

The modification of semiconductor surfaces by the attachment of unsaturated organic molecules has been a topic of increasing interest over the last ten years. The idea that combining organic chemistry with semiconductor technology could bring new advances and/or applications has been an important motivation for the activity in this field. A variety of methods for preparing films of organic molecules on different semiconductors, particularly on silicon, have been developed. Among these, a particularly promising approach is a surface chain reaction of terminally unsaturated molecules with hydrogen-terminated Si surfaces. For the prototype case of alkene molecules reacting with H-Si(100), for instance, this reaction mechanism has been shown to lead to the formation of linear nanostructure of molecules attached to the surface through C-Si bonds.
Using First Principles String Molecular Dynamics -- an efficient method for finding reaction pathways which was recently developed in our group -- we have studied the surface chain reaction mechanism for different alkene, alkyne and aldehyde molecules interacting with hydrogenated (111) and (100) surfaces in which one Si dangling bond is initially present. We compared unconjugated and conjugated molecules, in order to understand how the terminal conjugation influences the reaction viability. We found interesting differences in the reaction pathways of all these molecules, and explained a variety of experimental observations.
These studies will be extended to different semiconductor surfaces, for which experimental results have become available. New functionalization mechanisms, e.g. optically activated functionalization of fully H-saturated Si surfaces, are being considered. Of particular interest is the study of how the attachment to the surface affects the molecular functionality, e.g. in the case of some biomolecules.