Self-Diffusion in Intrinsic and Extrinsic Silicon Using Isotopically Pure 30 Silicon Layer
- PDF / 113,278 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 38 Downloads / 226 Views
SELF-DIFFUSION IN INTRINSIC AND 30 ISOTOPICALLY PURE SILICON LAYER
EXTRINSIC
SILICON
USING
Yukio Nakabayashi, Hirman I. Osman, Toru Segawa, Kazunari Toyonaga, Satoru Matsumoto, Keio Univ., Dept. of Electronics and Electrical Engineering, Yokohama JAPAN Junichi Murota, Tohoku Univ. Res. Inst. of Electrical Communication, Sendai, JAPAN Kazumi Wada, Massachusetts Institute of Technology, Dept. of MS&E, Cambridge Takao Abe, Sin-Etsu Handootai, Isobe R&D Center, Gunma, JAPAN ABSTRACT Silicon self–diffusion coefficients were measured in intrinsic and extrinsic silicon from 870 to 1070°C using isotopically pure 30Si layer. 30Si diffusion profiles are determined by secondary ion mass spectrometry. The temperature dependence of intrinsic diffusion coefficient in bulk Si is obtained. Comparing it in heavily As-doped or B-doped Si, it is found that Si self-diffusion is entirely mediated by interstitialcy mechanism at lower temperatures below 870°C. INTRODUCTION As device dimension shrinks with increasing degree of integrations, accurate prediction and precise control of dopant profiles such as junction depths become very important, and many parameters are necessary for simulation models. Since dopant atoms diffuse by interaction with point defects such as vacancies and self-interstitials, and most Si-LSI fabrication processes such as oxidation and ion-implantation perturb the equilibrium point-defect concentrations, understanding the behavior of point defects and derivation of accurate Si self-diffusivity are essential for device technology. Self-diffusion in Si can be viewed as a limiting case of dopant diffusion, in which diffusion atoms carry no excess charge and introduce no distortion in the lattice. Thus the study of self-diffusion is very important for the understanding of diffusion mechanism of dopant atoms. To date, Si self-diffusion has been investigated by many researchers [1-4] on account of its scientific and technological importance. However, the results on Si self-diffusion reported by them are less consistent compared with Ge self-diffusion, due to the difficulty of Si self-diffusion experiment caused by the very short half-life of the usually used radioactive tracer 31Si, high 30Si background concentration (3.1%) and 30Si implantation-induced radiation damage. In such situation, there is a common recognition that combining the lower and higher temperature data, a kink exists at about 1000°C in the Arrhenius plot of Si self-diffusion and it indicates a change in the mechanism. Seeger and Chik [6] suggested that both vacancy and interstitialcy mechanisms contribute to the nonlinearity of the Arrhenius plot. More specifically they proposed that in Si at lower temperatures self-diffusion mainly occurs via vacancies, whereas at higher temperatures it J3.3.1
is dominated by the interstitialcy mechanism. Further Gösele et al. [7] calculated these contributions from diffusion data of Au or Ni in Si. According to the calculation, the interstitialcy component is larger than vacancy component at higher temperatures
Data Loading...
