The Effect of Light Soaking on the Photoconductivity Response in a-Si:H
- PDF / 299,428 Bytes
- 6 Pages / 420.48 x 639 pts Page_size
- 100 Downloads / 147 Views
THE EFFECT OF LIGHT SOAKING ON THE PHOTOCONDUCTIVITY RESPONSE IN A-SI:H
QIU CHANGHUA, WU WENHAO, ZHAO SHIFU AND HAN DAXING Institute of Physics, Academia Sinica, P.O.Box 603,
Beijing,
PRC
ABSTRACT The temperature dependences of response time tr and steady state photoconductivity (PC) were used to deduce the DOS at energies above the dark Fermi level. The tr and PC of annealled state A and light soaked state B were measured from 115K to 300K. Light soaking causes degradation of PC. Compared with state A, the PC response of state B is faster at low temperatures, but is slower at high temperatures. The difference between state A and state B was interpreted by a decrease of shallow traps and an increase of deep traps.
Introduction It was known as SWEI that light soaking induces degradation of PC and lowering of Fermi level. SWE was explained by an increase of dangling bond 2 states near midgap. But there are evidences for changes of two kinds of 3 centers or donor-like and acceptor-like states . So it is an unsolved problem whether only dangling bond states are induced or not. The response behaviour of PC is determined by generation, trapping and recombination, so investigation of PC response and steady state PC could bring out information about traps and recombination centers. The steady state PC and its temperature dependence were used to characterize the electron density ns, the electron lifetime T and the position of quasi-Fermi
level Eq.
These quantities
were obtained
Tp = e ns p
(la)
ns = G T
(lb)
G = p F(1-R)(1-exp(-ctd))/d
(2)
Eq = kTln(ns/Nc) = kTln(Op /(eANc)).
(3)
from relations
2 Set the conduction band edge Ec=O. Reasonable values of 1=lcm /(Vs), r1=1 6 20 3 and Nc= .2xlO /cm were used. When the steady state was approached, the distribution of electrons 5 6 is characterized by Eq, with a distribution function '
f(E) = exp(-(E-Eq)/kT),
E>Eq.
The density of electrons distributed in the traps is nt = f' f(E)g(E)dE, Eq
(4) given by (5a)
3
where g(E) 1/(cm eV) is the continuous density of gap states at energy E. Usually, g(E) varies more slowly than f(E), therefore, the trapped electrons are in the vicinity of Eq, nt
- g(Eq)kT/e .
Mat. Res. Soc. Symp. Proc. Va. 70. 1986 Materials Research Society
(5b)
232
The approach to steady state was used to characterize electron trapping. In our experiments, a mechanical shutter was used to turn on the illumination. The approach to steady state takes much longer time than the electron lifetime, because of large number of trap states. The response time tr is defined by the time when6 the photocurrent rises to 90% of its steady state value. tr is given by tr = (l+nt/ns)T = (nt/ns)T ,
(6a)
insert of eq.(ib) tr
(6b)
= nt/G .
According to eq.(6b) and eq.(5b), the measurement of response time tr gives the density of trapped electrons nt and the DOS at the energy of Eq. The quasi-Fermi level Eq could be moved towards the conduction band edge Ec significantly by lowering temperature, so the combined measurements of the temperature dependences of
Data Loading...
