Preparation of High Conductive Cubic Boron Nitride Thin Films by in-situ zinc Doping
- PDF / 83,826 Bytes
- 3 Pages / 612 x 792 pts (letter) Page_size
- 12 Downloads / 229 Views
0955-I11-01
Preparation of High Conductive Cubic Boron Nitride Thin Films by in-situ zinc Doping Kenji Nose and Toyonobu Yoshida The University of Tokyo, Tokyo, 113-8656, Japan
ABSTRACT Silicon and zinc were doped to cubic boron nitride (cBN) thin films by sputtering bulk dopant sources during the film deposition. Concentrations of these two elements showed different in-plane distributions, presumably caused by the difference of the sticking probability on the growing surface. Silicon impurity did not behave as an active dopant in cBN. However, zinc doping enhanced the electric conductivity from 10-8 to 10-2 (Ω−1cm-1) as the zinc concentration increased from undoped to approximately 2 %. INTRODUCTION Much attention has been paid to deep-ultraviolet light emitting diodes and high power electronic devices. Cubic boron nitride (cBN) is a potential candidate for these applications because it has unique properties such as the ultra wide bandgap and the high chemical stability. In the deposition of cBN thin films, varieties of plasma [1-3]- and ion beam-assisted [4,5] deposition methods have been established until now. Based on these thin film processes, electric and optoelectric properties of cBN thin films are studied in detail recently. Especially, a lot of efforts have been made to the development of doping methods to actualize semiconducting properties, such that magnesium [6] and beryllium [7] doping to cBN have already been reported. Unfortunately, no clear change of the electric conduction has been achieved yet. EXPERIMENT In this study, a sputter doping [8,9] was attempted to dope silicon and zinc impurities to cBN. Thin rod-shaped solids of pure silicon and zinc were sputtered during the film deposition by applying –200 V DC bias in a phase-regulated RF bias sputtering system. The arrangement of the rod, the substrate and the target is illustrated in Fig. 1. Numerical calculations of the dopant distribution were performed by a Monte Carlo method based on the random scattering of dopant elements in the atmosphere of argon. Impurity concentrations were measured in-depth and inplane by secondary ion mass spectroscopy (SIMS, Cameca, IMS-4f) and X-ray photoelectron spectroscopy (XPS, Shimazu, Kratos). Electric conductivities and hall voltages were measured by the van der Pauw method using Resitest 8300 (Toyo) from room temperature to 570 K. The carrier type was identified by the Seebeck measurement.
RESULTS & DISCUSSIONS Silicon and zinc concentrations were controlled from zero to approximately 4 % by adjusting the position of the rod tip from x=110 to 0 mm in Fig. 1. In-plane XPS measurements clarified that silicon concentration differed 5 times between positions of x= -10 and x=10 mm when the rod tip was fixed at x=0 mm. On the other hand, the difference of zinc concentration was only 20 % under the same condition. The Monte Carlo simulation could well explain the different in-plane distributions by using different sticking probabilities. That is, the zinc distribution was well simulated by the sticking probabi
Data Loading...
