A Barrier to Trap Filling in CuIn 1-x Ga x Se 2
- PDF / 447,101 Bytes
- 5 Pages / 612 x 792 pts (letter) Page_size
- 50 Downloads / 190 Views
B5.24.1
A Barrier to Trap Filling in CuIn1-xGaxSe2 David L. Young, Kannan Ramanathan, Miguel Contreras, Jehad Abushama, Richard S. Crandall National Renewable Energy Laboratory, 1617 Cole Blvd., Golden, CO 80401 ABSTRACT Voltage pulses of variable length were applied to CuIn1-xGaxSe2/CdS (0 < x < 1) junction solar cells. The resulting transient capacitance emission signal was recorded for several minutes. The amplitude of the capacitance emission signal increased linearly with the log of pulse time. These data do not follow the standard model for trap capture and emission of carriers. Instead they follow a simple electrostatic model based on electrostatic charging of traps. INTRODUCTION The CuIn1-xGaxSe2 material system is one of the leading candidates for low-cost absorbers for thin-film solar cells achieving efficiencies over 19% [1]. Device performance correlates with the density [2,3] and energy [4] of defect levels detected by capacitance techniques with a maximum efficiency occurring at x ~ 0.3. Theoretical studies have assigned defect transition energy levels in the bandgap to specific crystalline defects[5]. However, metastabilities in the material system complicate the assignments of experimental and theoretical defect levels [6-10]. The nature of electronic traps is commonly studied using deep-level transient spectroscopy (DLTS) methods. DLTS approaches can determine defect energy levels relative to band edges (majority or minority trap), as well their trapping cross-section (s) [11,12]. The charge emission-rate and s are important quantities for determining whether a defect functions as a recombination center or a shallow trap. This knowledge is especially important for photovoltaic (PV) materials because recombination limits the available photocurrent of the solar cell. The most reliable method to determine s is to measure the density of trapped charge as a function of trap-filling pulse time (tp) in a junction device structure. The density of filled traps (Nf) should increase linearly with time at short tp and finally saturate when charge is being emitted as fast as it is being captured. This process should obey the following equation [11]: tp Ê - ˆ tc ˜ Á N f ( t ) = N o 1- e , Á ˜ Ë ¯
(1)
where No is the saturated defect density and tc is the characteristic time containing s. This paper outlines a failed attempt to determine s in CuIn1-xGaxSe2 using the above † equation. Instead we find that Ns(t) obeys the expression: Ê t ˆ N s ( t ) = Ao lnÁ1+ p ˜ , Ë to ¯
† Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 14 Aug 2017 at 10:49:51, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1557/PROC-763-B5.24
(2)
B5.24.2
where Ao is a constant and to is a characteristic time. Similar behavior was observed in aSi [13], GaAs [14], and Ge0.3Si0.7/Si [15]. A model [13] explaining these results is based on the realization that the traps are clustered, rather than uniformly distributed. Trap-filling charges thes
Data Loading...
