Silicon AD-Dimer Binding on Si(100)
- PDF / 793,810 Bytes
- 5 Pages / 414.72 x 648 pts Page_size
- 106 Downloads / 228 Views
PETER J. BEDROSSIAN*, ARTHUR P. SMITH**, AND HANNES JONSSON** *Lawrence Livermore National Laboratory, L-350, Livermore CA 94551 "**Department of Chemistry, BG-10, University of Washington, Seattle WA 98195
ABSTRACT Isolated Si ad-dimers nucleate predominantly on Si(100) substrate dimer rows at room temperature. Using tunneling microscopy (STM), we find that, while ad-dimers oscillate between two mutually orthogonal orientations, the geometry with the ad-dimer bond perpendicular to the substrate's dimer bonds is the more stable. This observation confirms a prediction of ab initio calculations with a gradient correction for the energies of ad-dimers in various configurations, which differs from that of the local density approximation (LDA). INTRODUCTION The development of reliable simulation of crystal growth depends critically on an accurate description of the interaction among individual adatoms and between adatoms and the substrate, particularly at the stage of growth in which individual adatoms nucleate precursors for island growth. Models of layerwise, homoepitaxial growth on Si(100)-2xl generally assume that deposited Si adatoms initially nucleate individual dimers, which in turn become precursors for the growth of dimer rows. 1-7 The identification of the preferred binding topology for a solitary ad-dimer from among various inequivalent topologies is therefore central to realistic modeling of Si growth on Si(100). Different ad-dimer configurations have been predicted on the basis of first-principles calculations using LDA and gradient corrections. Using tunneling microscopy (STM), we identify the preferred binding configuration directly and confirm the prediction of ab initio calculations with a gradient correction, in what we believe to be the first experimental test of a case in which the gradient correction offers a qualitatively different result from LDA. METHODS The experiments employed an ultra-high vacuum (UHV) chamber with base pressure below 10-10 torr. The Si(100) surface was initially prepared by heating to 1250C, after which it yielded a strong 2x 1 periodicity in Low-Energy Electron Diffraction (LEED) and tunneling microscopy (STM). Si was deposited on a room-temperature Si(100) surface from a thoroughly-degassed, heated wafer at =8x1O- 4 monolayers per second (ML/s), during which the chamber pressure remained below 2x10- 10 torr. After deposition, the sample was transferred in situ to a commercial STM (Omicron Associates) for imaging. All images were acquired with constant tip-sample current, with greater tip-sample separation represented by a brighter shade of grey. The images presented below were acquired by biasing the tunnel junction of the STM with the tip negative with respect to the sample, so that electrons tunnel from the tip into empty sample surface states. This biasing scheme enables the identification of individual atoms on Si(100), as opposed to dimer bonds which would be imaged under opposite junction polarity. 35 Mat. Res. Soc. Symp. Proc. Vol. 389 01995 Materials Research Socie
Data Loading...
