A bond-pair model Hamiltonian developed previously for systems consisting of interacting atoms is applied to describe atom-surface interactions. By proposing a mixed basis set involving localized adatom orbitals ͕ ␣ ͖ and extended surface states ͕ k ͖, and by application of a mean-field approximation, the Hamiltonian is reduced to the form of the single-particle Anderson model. The resulting model Hamiltonian is free from adjustable parameters. These parameters include both the effects of electronic interactions between the atom and the solid and those arising from the lack of orthogonality between the adsorbate and substrate orbitals. The nonlocal exchange contributions are treated consistently within the Hartree-Fock method, while valencelike and corelike band states are also taken into account. This model is applied to consider the interaction of hydrogen with metals ͑Al, Li, and Na͒. The results for chemisorption are in good agreement with those obtained by other theoretical approaches based on either the density functional theory or embedding cluster methods, as well as with existing experimental data. In addition, the calculation of the shifts and widths of the adsorbate levels in an ample range of separation distances are also in good agreement with those obtained by using atomic physics techniques. ͓S0163-1829͑98͒10031-0͔
The combined effect of finite potential barriers and dielectric mismatch between dot and matrix on excitonic properties of semiconductor quantum dots has been studied. To avoid the unphysical divergence in the self-polarization energy which arises for the simplest and profusely adopted steplike model of the dielectric interface, we proposed a realistic (finite size) smooth profile for the dielectric interface. We have found that the excitonic binding energy can be either higher than the corresponding one to complete confinement by infinite barriers or essentially zero for a wide range of dot sizes depending on the thickness of the dielectric interface. 73.20 Dx,77.55.+f
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.