Ames Laboratory Condensed Matter Physics Group

Atomistic Simulation Highlights(2000-2001)

Rocking and Rotation of a Si-Si, Ge-Ge, and Si-Ge Addimer on the Si(001)

Z.Y. Lu, F. Liu, C. Z. Wang, X. R. Qin, B.S. Swartzenruber, M. G. Lagally, and K.M. Ho, Phys. Rev. Lett. 85, 5603(2000).

We have performed extensive first-principles calculations to study the rocking and rotation of an addimer (Si-Si, Ge-Ge, or Si-Ge) on the top of a dimer row on the Si(001) surface. The energy barriers of the rocking motion for the addimers are computed to be less than 0.2 eV and STM experiment should not be able to resolve such fast rocking dynamics. Our calculations further suggest that the rocking dynamic appearance of an addimer on the Si(001) observed by STM is actually due to the 180 degree rotation of a mixed Ge-Si addimer (path 1 in the figure). The energy barrier of such a rotation is around 0.74 eV, in good agreement with experimental measurement.

Diffusion and Intermixing of Si-Ge Addimers on the Si(001) Surface

Z.Y. Lu, C. Z. Wang, and K. M. Ho, Surf. Sci. 506, L282(2002).

Recent experiment discovered an interesting reversible intermixing process involving the exchange of the Ge atom in an adsorbed Si-Ge dimer on the Si(001) surface with a substrate Si atom. We have performed first-principles total energy calculations to study the atomistic mechanisms of diffusion and intermixing in this system. Our calculation suggests that intermixing is triggered by the diffusion of the addimer on the surface. The energy barriers for the diffusion and intermixing events obtained from our calculations are in good agreement with experiment.

Correlated Piecewise Diffusion of a Ge Addimer on the Si(001) Surface

Z. Y. Lu, C. Z. Wang, and K. M. Ho, Phys. Rev. B62, 8104(2000)

We have performed extensive first-principles calculations to study the binding and diffusion of a Ge addimer on the Si(001) surface. The diffusion of a Ge addimer along the lowest-energy-barrier pathway (path 1 in the figure) is found to be piecewise but with strong correlation. This correlated piecewise diffusion pathway is also favored by a Si addimer on the Si(001) surface, with an energy barrier in excellent agreement with experimental measurement.

Structure and Electronic Properties of the Stepped Si(111)-(7x7) Surface

M. Hupalo, B. J. Min, C. Z. Wang, K. M. Ho and M.C. Tringides, Phys. Rev. Lett. 84, 2877(2000)

We have studied the atomistic structure and electronic properties of the stepped Si(111)-(7x7) surface in collaboration with the experiment group at Ames Lab (M. Tringides). Our study shows that the observed dependence of the STM images of the stepped Si(111)-(7x7) surface as the function of tunneling voltage is strongly correlated with the local electronic density of states of the adatoms on the stepped surface which is different from that of the flat surface. Our studies also reveal several new surface bands induced by the step. Atomic relaxation responsible for the rearrangement of electronic structure at the step has been investigated. The calculation results provide useful information for a better understanding the images and spectra from STM measurement.

Reconstruction of the Mo(100) Surface: a Tight-Binding Study

H. Haas, C. Z. Wang, K. M. Ho, M. Fahnle, C. Elsasser, Surf. Sci. 457, L397(2000).

The Mo (100) surface has been investigated by our recently developed environment-dependent tight-binding model. The predicted reconstruction of the Mo(100) surface is similar to the one observed in diffraction experiments. The electronic structure and Fermi-surface contours of the surface states obtained by our calculations seem to favor the charge-density-wave mechanism as the driving force for the reconstruction.

Atomistic Simulation of Laser-Induced Graphitization on a Diamond (111) Surface

C. Z. Wang, K. M. Ho, M. D. Shirk, P. A. Molian, Phys. Rev. Lett. 85, 4092(2000).

We have performed tight-binding molecular dynamics studies of laser-induced graphitization on the diamond (111) surface. Our simulation results suggest that the quality of the laser-treated diamond surfaces is dependent on the length of laser pulse being used. Under nanosecond or longer laser pulses, the diamond (111) surface is found to graphitize via formation of graphite-diamond interfaces, leading to a dirty surface after the laser treatment. By contrast, with femtosecond laser pulses, graphitization of the surface is found to occur layer by layer, resulting in a clean surface after the process. This atomistic picture provides an explanation of recent experimental observations. The physical origin of the different behavior of the electronically-excited diamond surface and thermal-excited diamond surface is also investigated.