Modeling of Atomistic Processes in Semiconductors: from Defect Signatures to a Hierarchy of Annealing Mechanisms

  • PDF / 1,175,674 Bytes
  • 11 Pages / 612 x 792 pts (letter) Page_size
  • 68 Downloads / 152 Views

DOWNLOAD

REPORT


0978-GG08-01

Modeling of Atomistic Processes in Semiconductors: from Defect Signatures to a Hierarchy of Annealing Mechanisms Michel Bockstedte1,2 Fisica de Materiales, Universidad de Pais Vasco, Apto 1072, San Sebastian, Spain 2 Theoretische Festkörperphysik, Universität Erlangen-Nürnberg, Staudtstr. 7B2, Erlangen, 91058, Germany 1

ABSTRACT The modeling of atomistic processes in semiconductors based on the density functional theory is outlined. The role of intrinsic defects in the self and dopant diffusion, as well as in the dopant activation is investigated for the case of silicon carbide. The identification of the microscopic origin of experimental defect centers by calculated defect signatures establishes a link between theoretical modeling and experiments.

INTRODUCTION Point defects in semiconductors are pivotal for the materials application in electronic devices. Beneficial substitutional impurities enable as dopants the p-type or n-type conduction. Well defined grown-in or implanted dopant profiles are the corner stones of device engineering. Un-intentional impurity-related deep defect centers and intrinsic defects counteract these efforts. They compensate dopants and as defects mediating the dopant diffusion degrade dopant profiles. Albeit unwanted and at the same time unavoidable, they play a vital role in the dopant activation and the annealing of process-induced defects. Unraveling of the fundamental mechanisms of diffusion and defect kinetics is a focus of both experimental and theoretical research. The modeling of atomistic processes in semiconductors aims to derive a microscopic picture of the activation and diffusion of dopants based on ab initio methods combined with a thermodynamic analysis and tools to investigate rate processes. Commencing with the most fundamental intrinsic defects and typical dopants, the modeling investigates the nature of the defects and their complexes. The identification of experimental defect centers by means of calculated defect signatures, such as hyperfine tensors and local vibrational modes, establishes an important link between the predictions of the modeling and their verification in experiments. The analysis of the diffusion mechanisms of intrinsic defects forms the corner stone of the understanding of the vacancy and interstitial mediated dopant diffusion and the annealing kinetics and the formation of thermally stable vacancy or interstitial aggregates.

The methods utilized in the modeling will be briefly outlined and applied to the compound semiconductor silicon carbide. In SiC owing to the compound nature of the material two different kind of vacancies and interstitials exists, that are proven to possess distinct physical properties with profound consequences for the defect kinetics and dopant diffusion. The identification of the silicon and carbon vacancy, the di-vacancy and the carbon vacancy-antisite complexes that evolves from the metastability of the silicon vacancy provide the basis for the discussion of the annealing mechanisms of vacancies and inters