Temperature Dependence of the Photoinduced Degradation and Annealing in a-Si:H
- PDF / 147,111 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 92 Downloads / 133 Views
Temperature Dependence of the Photoinduced Degradation and Annealing in a-Si:H Niko Schultz and P. Craig Taylor University of Utah, Salt Lake City, Utah 84112 ABSTRACT We investigated the temperature dependence of the production and annealing kinetics of the light induced defect states in a:Si:H by electron spin resonance (ESR). At low temperatures (T ~ 25 K) the silicon dangling bond production is about half as efficient as it is at 300 K. Defects, which are created below about 100 K, almost entirely anneal at room temperature. A sample of a-Si:H, which is subjected to several photo-excitation and annealing cycles, shows a very slow increase of both the degraded and annealed defect densities. The difference in the spin densities between the annealed and degraded states decreases with an increasing number of degradation/annealing cycles. INTRODUCTION Photo-induced degradation is a well studied and characteristic feature of hydrogenated amorphous silicon (a-Si:H) [1-3]. In fact, there are several photo-induced effects which have many manifestations, including the decrease of solar cell efficiencies and the degradation of transistor properties. From an observed proportionality between the low energy absorption, which is ascribed to deep defects, and the electron spin resonance (ESR) at g = 2.0055 [4] it has been concluded that the photoinduced spins originate from silicon dangling bonds, i.e., undercoordinated Si atoms. Therefore, the ESR resonance at g = 2 is a measure of the dangling bond density, and hence of the degradation in a-Si:H. It is also well known that photoinduced degradation is primarily a reversible process. Annealing at temperatures above 200°C for at least 30 min is a common recipe to restore the original defect density of an unexposed sample. The cycle of photodegradation and annealing can be repeated many times with only reversible changes [5]. EXPERIMENTAL DETAILS The a-Si:H sample consisted of a stack of four a-Si:H films on quartz substrates with a spin density of about 5 × 1015 cm-3 including the surface spin density. The sample was mounted inside an ESR cavity equipped with openings to irradiate the sample inside the cavity. All experiments were performed on a standard ESR spectrometer (Bruker, Model 200D-SRC) at Xband. We investigated the ESR spin density by employing a second harmonic technique, i.e., by detecting the ESR signal at twice the magnetic field modulation frequency, which was set to 50 kHz. Except where otherwise stated, all ESR measurements were performed at about 50 K. For all degradation measurements a Xe arc lamp was employed. The light passed through a 580 nm long pass filter and had an intensity of approximately 100 mW/cm2 at the sample. In every degradation measurement the sample was irradiated for 10 h. Annealing of the sample at temperatures, T, below 300 K took place inside the cavity by reduction of the He cooling gas
A3.4.1
flow. For annealing at T > 300 K the sample was removed from the cavity and heated for 30 min in a nitrogen atmosphere at about T = 200°C (470
Data Loading...
