The Effect of Annealing Temperature on the Morphology of Stacking Faults in Czochralski Silicon
- PDF / 2,471,756 Bytes
- 5 Pages / 417.6 x 639 pts Page_size
- 2 Downloads / 204 Views
Inc.
Narayan, and Tan,eds. Defects in Semiconductors
381
THE EFFECT OF ANNEALING TEMPERATURE ON THE MORPHOLOGY OF
STACKING FAULTS IN CZOCHRALSKI SILICON R. F. PINIZZOTTO and H. F. SCHAAKE Central Research Laboratories, Texas Instruments,
Inc.,
Dallas,
TX 75265
ABSTRACT Nucleation and growth of stacking faults formed in CZ silicon during oxygen precipitation have been studied using x-ray topography, TEM and FTIR. Samples were annealed in argon for various 0 times at 550%C and 750 C followed by a 16 hour anneal in dry oxy0 gen at 1000'C. In samples annealed at 550 C, the stacking faults were several layers thick with colonies of precipitates at their 0 centers. The faults in samples annealed at 750 C contained only one particle and were single in nature. It is proposed that the faults are formed by thin oxygen precipitate platelets and that the different morphologies are due to different oxygen precipitation rates. The platelets are probably a modified cristobalite, as determined by micro-diffraction results.
INTRODUCTION Stacking faults in Czochralski silicon have been studied for many years and are the subject of numerous scientific papers. !One way these stacking faults are formed is during the precipitation of oxygen in silicon due to various heat treatment cycles. Some authors report that the stacking faults are generated and nucleated at one oxygen precipitate particle and then grow by the capture of the interstitials formed to accomodate the volume change caused by the oxidation progress [1]. Others have said that colonies or groups of particles are found on the faults [2]. No one has proposed a mechanism that can reconcile the two types of morphologies. In this paper, we report the results of our studies of both types of faults. A model explaining their possible relationship is proposed. We also report on micro-difraction results that imply the material on the faults is a modified cristobalite phase of Si0 2 . EXPERIMENTAL DETAILS The material used in these experiments was (100) p-type boron doped, 5-10 Q -cm silicon grown by the Czochralski process in a commercially built crystal puller. The slices were all cut from the top of a single boule. A resistivity stabilization anneal was not performed. Other than the crystal growth process itself, the wafers were not subjected to any heat treatments not specifically mentioned in this report. Fourier transform infrared spectroscopy was used to measure the oxygen and carbon concentrations. The radial distribution of these impurities was measured by stepwise scanning a 3-mm aperture across the slices. The oxygen concentration was found to vary by a factor of 2 from the edge to the center of the slice with the maximum concentration being 1.0 x 101 cm -3, using the calibration of Kaiser and Keck r3]. The carbon concentration was below the detectability of our FTIR apparatus and is estimated 3 to be < 2 x 1016 cm . The samples were annealed in slice form in a two step process. The first step was designed to form nuclei which could survive latter higher temperatu
Data Loading...
