• PDF / 262,037 Bytes
• 6 Pages / 420.48 x 639 pts Page_size
• 99 Downloads / 274 Views

DOWNLOAD

REPORT

---

ON THE TEMPERATURE DEPENDENCE OF RESISTIVITY OF POLYCRYSTALLINE SILICON FILMS MARK S. RODDER' AND DIMITRI A. ANTONIADIS" 'Texas Instruments Inc., Dallas, TX 75265 "Massachusetts Institute of Technology, Cambridge, MA 02139 ABSTRACT It is shown that the grain boundary (GB) in polycrystalline-silicon (poly-Si) films need not be modeled as a temperature-dependent potential barrier or as an amorphous region to explain the temperature (T) dependence of resistivity (p) in p-type poly-Si films at low T. Specifically, we consider that QB defect states allow for the tunneling component of current to occur by a two-step process. Incorporation of the two-step process in a numerical calculation of p vs. T results in excellent agreement with available data from 100 K to 300 K. INTRODUCTION A new conduction mechanism in poly-Si films is proposed which considers that the grain boundary (GB) defect states allow for a tunneling component of current to occur by a two-step process. The influence of the two-step process on conduction is dominant at low temperature (T); as such, this paper will compare numerical results with available low-T data for resistivity (p) versus T from Lu et al [1]. PREVIOUS MODELS The models of interest are those previously proposed to explain the data of Lu which shows a downward bending of the In p vs. 1/kT at low T [1]. Two different conduction models have been proposed to explain this downward bending of the In p vs. 1/kT. The first model considered that transport is due to thermionic emission and thermionic field emission (TE/TFE) with the GB proposed to be a potential barrier with height qx (in eV) and width 6 (in nm) being fitting parameters whose values decrease as T decreases due to a proposed reduction in phonon scattering [1]. This is unsatisfying since the GB width is quite small in comparison to typical grain sizes and thus, phonon scattering at the GB is not likely to be a dominant limitation to transport [2]. The second model to explain the downward bending of the In p vs. 1/kT proposed that transport is due to drift/diffusion (D/D) of carriers to an amorphous (a) GB and then conduction through the a-GB region via conventional a-conduction mechanisms (3]. This model is also unsatisfying for the following reasons. First, the bottleneck to transport to the GB is, in general, not solely D/D but also TE/TFE [4]. Second, the consideration of the GB as an a-region is not supported by electron microscopy. Third, even if one was to consider the GB as an a-region with width on the order of 1 nm, it is not clear that conventional models for conduction in an infinite amedium would always apply. For example, the variable-range hopping process occurring in an infinite a-medium may not be applicable to an (assumed) a-GB since the most probable hopping distance in conventional theory may be larger than the GB width. Mat. Res. Soc. Symp. Proc. Vol. 106. 11988 Materials Research Society

78

PRESENT MODEL The present model is formulated in one-dimension (valid when the depletion width W satisfies W