Temperature Dependence of the Growth and Saturation of Light-Induced Defects in a-Si:H
- PDF / 395,844 Bytes
- 6 Pages / 420.48 x 639 pts Page_size
- 92 Downloads / 147 Views
TEMPERATURE DEPENDENCE OF THE GROWTH AND SATURATION OF LIGHT-INDUCED DEFECTS IN a-Si:H
M. Isomura(a), N. Hata(b) and S. Wagner Department of Electrical Engineering, Princeton University Princeton, NJ, 08544
ABSTRACT We report new experimental data on the light-soaking of a-Si:H with a Kr-ion laser at an optical generation rate G of at least 4x10 2 1 to 3x10 2 2 cm- 3s-1. We studied the temperature and intensity dependence of the saturation of the defect density and found that the saturation value of light-induced defects (Nsat) is insensitive to temperature and light intensity below about 90*C. Above 90'C Nsat drops with increasing temperature. This behavior can be explained within the defect pool model by a limited number of defect sites coupled with an annealing process. INTRODUCTION Many researchers are working on the Staebler-Wronski effect [1] because lightinduced defects in a-Si:H pose a serious problem to device applications, and their understanding is of fundamental importance to a-Si:H. However, we have neither a clear understanding of the defect mechanism nor an unequivocal method for its evaluation. Therefore, it is important to acquire precise quantitative information about the effect. The saturated light-induced defect density Nsat [2] is a useful parameter for the evaluation of the stability of a-Si:H. Saturation can be reached within a few hours by soaking with light from a Kr-ion laser (X=647nm), which is absorbed nearly uniformly and at 3 W/cm 2 produces a carrier generation rate of 3x10 2 2 cm- 3 s"1 . Because of the phenomenon of saturation, the measured defect density Nsat is independent of spatial or temporal variations of the light intensity. Therefore, Nsat is a more reliable and useful criterion for stability than defect densities measured before saturation, because the rate of buildup of the defect density is affected very much by the conditions of light-soaking. We also found Nsat clearly correlated with the defect growth rate measured at typical sunlight intensity, so that we can estimate the defect growth rate dNs/dt once we have determined Nsat [3]. Therefore, the light-soaking history of a material may be evaluated in a few hours, and feedback on stability can be provided quickly to the deposition laboratory. In electronic grade a-Si:H samples obtained from many different laboratories and deposited by different techniques, Nsat rises with the Tauc gap (Eopt) or E 04 and with the total hydrogen concentration (CH), but is not correlated with the initial defect density or the Urbach energy [4]. The Nsat values measured so far lie between 4x10 16 and 2xl0 17cm- 3 with apparent lower limits that depend on Eopt and CH [3]. Therefore, Eopt and CH probably are important for stability. The observation of thermal annealing during light-soaking at high temperature, however, accentuates the question of the mechanism for defect saturation in our method. It could be the exhaustion of sites which are convertible to defects, or the attainment of a steady state between annealing and generation of def
Data Loading...
