Evidence For Trap-Conversion Induced Instability In Amorphous Silicon
- PDF / 159,150 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 113 Downloads / 190 Views
A12.5.1
Evidence For Trap-Conversion Induced Instability In Amorphous Silicon Vikram L. Dalal, Puneet Sharma and Abdul Aziz Dept. of Electrical and Computer Engr. and Microelectronics Research center Iowa State University, Ames, Iowa 50011, USA ABSTRACT It has been shown recently that there are two distinct types of recovery during annealing of amorphous Silicon after degradation due to light soaking. It has been postulated that the two different kinetics of annealing point to the existence of two different types of states, with perhaps one state being charged dangling bonds and the other state being neutral dangling bonds. To see if two kinds of states exist, in this paper, we study the kinetics of degradation within the first 100 seconds, and also study the entire absorption curve at all degradation times. An analytical model is derived for early time degradation based on the conversion of a D- state into a neutral dangling band by absorption of a light generated ( the trap-to-dangling bond conversion model of Adler) and the experimental data of degradation versus light intensity fit the predictions of the model very well. The model also predicts that the Adler-type negatively charged defect states, which have a negative correlation energy, upon conversion will transform into Do states at a higher energy, and therefore, there should be a decrease in absorption corresponding to states closer to the valence band, and an increase in absorption corresponding to states near the mid-gap. For the films where such D- states are deliberately introduced by using a small oxygen ( a donor atom) leak , we see strong evidence for such a behavior in absorption, with a decrease in the 1.3-1.4 eV photon energy range, and an increase in the 1.1 eV photon energy range. The increase in Do corresponds well with the decrease in photo-conductivity, even at the earliest times. INTRODUCTION It has been known for some time that two types of states are created in a-Si:H upon degradation, the “fast” states and “slow” states, the designation being based on whether they anneal rapidly or slowly upon thermal annealing [1]. Recently, Heck et al[2], Stradins [3] and Wronski et al[4] have shown that indeed, the behavior of annealing of photo-conductivity and of fill factors of solar cells indicates that there are two distinct types of states in the material. In particular, one does not find a one-to-one inverse correlation between midgap defect density measured by absorption at ~1.2 eV below the conduction band and photo-conductivity, as the standard single defect model would suggest. It has been postulated that perhaps the fast states are charged defects. Adler proposed many years ago that there were charged defects present in the material because of structural inhomogeneities in the material, and that these charged defects were negatively correlated in energy, i.e. the D- state was below the Do or D+ state [5]. He further proposed that the D- and D+ states would convert into neutral Do dangling bonds, at a different energetic location, upon lig
Data Loading...
