Density Function Study Of The Interaction Between Silver Clusters And Atoms

Metal clusters and nanoparticles often have excellent catalytic activities and their catalytic properties are very sensitive to the number of atoms in the clusters. Due to the rapid development of the computational techniques and various high-efficiency methods, the first-principle calculation studies play more and more important role in materials simulation and explanation of experimental phenomenon. The quantum size effects of the clusters are depended on the relations between the electronic structures and cluster sizes, thus the theoretical studies of the relations can help to understand and predict the size effects of clusters and nanoparticles.The ground state of Ag atom has a closed Ad shell and a single s valence electron, hence it could be seen as 'alkali-like' metals. Small silver clusters usually have very different physical and chemical properties from silver bulk material. There are experiments have proved that the active sites in Ag catalysis are only composed of only few Ag atoms. In recent years, atomic and molecular chemisorption on small silver clusters in the gas phase has been an active field of experimental and theoretical research. Both experimental and theoretical studies on the binding of oxygen, carbon-monoxide and hydrogen on Ag clusters have also appeared. Our groups have done lots of researches on the partial oxidation of methanol to formaldehyde on silver surface, and studied the modification on metal surface by using situ scanning tunneling microscopy (STM), low-energy electron diffraction (LEED) and X-ray photoelectron Spectroscopy (XPS). From the experimental results, the introduction of halogen atoms will enchance the reactivity and the selectivity of the oxidation of methanol. Three kinds of Ag(110) surface modification (chemisorption of halogen atoms, pre-adsorption of oxygen and electrochemical modification) and their effects to adsorption of methanol on the Ag(110) surface have also been investigated by theoretical calculations. So based on the previous work, in this thesis, the bindings of atoms on neutral and charged silver clusters are extensively studied by means of Density Functional Theory (DFT) , with cluster model approximation.To date, the Density Functional Theory (DFT) is a practical method to study the metallic systems, which has been rapidly developed recently. The reliability DFT methods is governed by the quality of the approximate exchange-correlation (XC) energy functional E_xc[ρ]. In the last decade, a number of new functionals have been suggested, often by forcing the required DFT characteristics to match experimental or high-level ab initio results. Considering the large number of DFT functionals available today, it is quite important that the applicability and accuracy of functionals must be documented, so they can be applied appropriately to electronic structure problems. Compared with the assessment of DFT functionals for organic systems, the assessment of DFT methods on the systems containing transition and coinage metal atoms seems to manifest a more complex picture. Many of these studies involve no more than two atoms of the 3d and Ad transition-metals, comparative studies of new functionals for larger transition metal clusters are still lacking due to the high computational demands. For such systems, reliable predictions of the structures and energies are of paramount importance.In the present work, as silver clusters are chosen as the model systems, we first tested 23 DFT functionals, including both GGA and hybrid, are adopted in our work on the neutral and ionic Ag_n (n≤4) clusters. The reliability of these functionals is evaluated by comparing predicted DFT equilibrium geometries, vibrational frequencies, vertical and adiabatic ionization potentials (VIP and AIP), and vertical detachment energies (VDE) with available experimental data or high-level ab initio results.We find that DFT methods incorporating the uniform electron gas limit in the correlation functional part, namely those with Perdew's correlation functionals (PW91, PBE, P86, TPSS), Becke's B95, and the Van Voorhis-Scuseria functional VSXC, generally perform better than the other group of functionals, e.g. those incorporating the LYP correlation functional and variations of the B97 functional. Strikingly, these two groups of functionals can produce qualitatively different results for the Ag₃ and Ag₄ clusters. The energetic properties and vibrational frequencies of Ag_n are also evaluated by the different functionals. The present study shows that the choice of DFT methods for heavy metals may be critical. It is found that the PW91PW91 functional has some advantages for predicting the range of properties. So we chose the PW91PW91 functional to study the following three parts:1, the structures and the properties of bare silver clusters: a number of DFT investigations of the geometries, physical properties and electronic structures of small neutral and charged silver clusters are already available in the literatures. In this work, for each cluster size and charge state, several initial structures of Ag_n,^0,±1 (n = 2～7) have been optimized. Our calculated low-energy structures of both the neutral and the charged clusters are in general consistent to the published literatures. The comparison of our calculations (the ionization potential (IP) and electron affinity (EA) for silver clusters) with experiments shows that PW91PW91 functional give satisfactory results.2, the binding of fluorine, chlorine, bromine, or iodine atom binding to small neutral, anionic, or cationic silver clusters Ag_n,^0,±1 (n = 2～7) has been studied by using the PW91PW91 density functional method. It was found that the binding of halogen atoms on the lowest-energy bare clusters does not always produce the lowest-energy complexes. In addition, the binding of halogen can greatly change the geometries of the silver clusters in some cases. The most stable structures of complexes, Ag_nX^0,±1 (X=F, C(?), Br, I) often have similar structures and binding patterns. Among various possible adsorption sites, bridge site is energetically preferred for the neutral Ag_n while top site is energetically more preferred for the anionic Ag_n with n≤6. For cationic clusters, adsorptions on bridge and face sites have similar binding energies, which are much larger than those on top sites. Natural bond orbital analyses show that irrespective of charge state, electrons always transfer from silver atoms to adsorbate and silver acts like alkali metals in the interaction with halogen atom. It was also found that halogen atoms bind more strongly with odd-electron bare clusters than with even-electron bare clusters. These patterns reveal that even-electron clusters are more stable than odd-electron clusters. The binding energy and electron transfer ability are in the order: F>Cl>Br>I, which is consistent to the electronegativity order.3, small neutral, anionic, and cationic silver cluster hydrides Ag_nH and anionic HAg_nH (n = 1 - 7) have been studied using the PW91PW91 density functional method. It was found that the most stable structure of the Ag_nH complex (neutral or charged) does not always come from that of the lowest-energy bare silver cluster plus an attached H atom. Among various possible adsorption sites, bridge site is energetically preferred for the cationic and most cases of neutral Ag_n. For anionic Ag_n^-, the top site is preferred for smaller Ag_n within n≤4 while bridge site is preferred for bigger clusters. After binding the second H atom, the obtained lowest energy structures HAg_nH^- are those with two H atoms shared with only one Ag atom in anionic clusters. Natural bond orbital analysis shows that irrespective of charge state, electrons always transfer from silver atoms to adsorbate. Significant odd-even alternation patterns that hydrogen atoms bind more strongly with odd-electron bare clusters than that with even-electron bare clusters can be observed.