Ames Laboratory Condensed Matter Physics Group

Atomistic simulation highlights (1998)

Transferable Tight-Binding Parameters

Our main effort this year has been toward the development of more accurate tight-binding parameters for Carbon, Silicon, and Molybdenum systems. We have used the Cray vector machines at NERSC in this effort to find parameters suitable for a wide range of structures so that we can investigate a broad range of phenomena from surface dynamics to amorphous structures.
In the next year, we will use the new Silicon parameters to study many processes involved in the homogenous and heterogenous growth of films on Silicon wafers. Understanding the dynamics of Si adatoms and addimers on Si surfaces is essential to understanding the low temperature epitaxial Si crystal growth process, where the system is far from equilibrium and the surface morphology is determined mainly by the rates of competing atomic transitions. Understanding this Si growth is an essential starting point for controlled growth of Si-Ge heterostructures and other more complex systems. Over the next year, we will concentrate our efforts on developing a complete understanding of the homo-epitaxial growth process, from studying the island formation process to understanding the role of steps in the growth process, including diffusion of adatoms and addimers on stepped and flat surfaces.

Car-Parinello Pseudopotential Calculations

We have been using the Cray vector machines, and more recently the Cray T3E, to perform ab initio investigations of clusters and surfaces. The Car-Parinello code has also been vital to the development of accurate tight-binding parameters for Silicon and other systems.
We have used this code to find the lowest energy configurations for a series of small Silicon clusters, and to try to understand the electronic structure of these clusters. A genetic search algorithm is used to find these low energy configurations using a less accurate classical or tight-binding model, then the best candidates are further relaxed using the Car-Parinello code to get a more accurate measure of the ground state energy. Using this approach, we have found many structures for Silicon clusters of upto 20 atoms that had not previously been know. We can also use this code to investigate their stability as a function of temperature, and we can look at their electronic structure to help understand their stability in the presence of other elements and clusters. The picture to the right is an isosurface of the charge density for a Silicon 13 cluster, showing the electronically active site in deep purple and blue on the top. Future work will be geared toward Ge and Sn clusters, where the parallel Car-Parinello code on the T3E should allow us to work with clusters of upto 50 atoms. This should allow us to elucidate some of the basic formation trends for clusters in general, and to compare the trends for the different elements. Adapting the code to handle charged clusters will further our ability to compare our work to experimental results.
Running the Car-Parinello code on the T3E will also allow us to tackle many surface structures and surface dynamics that were previously unreachable, or could only be done by using shorter cutoffs that hurt the accuracy of the results. We are currently doing a complete Si(7x7) electronic structure calculation to make clear all the surface states and their relation with STM images observed in experiments. We plan also to do ab initio MD simualtions for interesting and intricate surface phase transitions from c(4x2) or p(2x2) to (1x1) on Si(100) surface and from (7x7) to(1x1) on Si(111) surfaces.

References:

K. M. Ho, A. A. Shvartsburg, B. C. Pan, Z. Y. Lu, C. Z. Wang, J. Wacker, J. L.Fye, and M. F. Jarrold, “Structures of medium sized silicon clusters,” Nature 392, 582(1998).

G. D. Lee, C. Z. Wang, Z. Y. Lu, and K. M. Ho, “Addimer diffusion between trough and dimer row on Si(100),” Phys. Rev. Lett. 81, 5872(1998)

H. Haas, C. Z. Wang, M. Fahnle, C. Elsasser, and K. M. Ho, “Environment-dependent tight-binding model for molybdenum,” Phys. Rev. B57, 1461(1998).

C. Z. Wang, B. C. Pan, M. S. Tang, H. Haas, M. Sigalas, G. D. Lee, and K. M. Ho, “Environment-dependent tight-binding model,” MRS Symposium Proc. 491, 211(1998).