Investigating Minority-Carrier Traps in p-type Cu(InGa)Se 2 Using Deep Level Transient Spectroscopy
- PDF / 163,703 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 44 Downloads / 191 Views
F5.35.1
Investigating Minority-Carrier Traps in p-type Cu(InGa)Se2 Using Deep Level Transient Spectroscopy Steven W. Johnston, Jehad A. M. AbuShama, and Rommel Noufi National Renewable Energy Laboratory, Golden, CO 80401, U.S.A. ABSTRACT Measurements of p-type Cu(InGa)Se2 (CIGS) using deep-level transient spectroscopy (DLTS) show peaks associated with minority-carrier traps, even though data were collected using reverse bias conditions not favorable to injecting minority-carrier electrons. These DLTS peaks occur in the temperature range of 50 to 150 K for the rate windows used and correspond to electron traps having activation energies usually in the range of 0.1 to 0.2 eV for alloys of CIS, CGS, and CIGS. The peak values also depend on the number of traps filled. For short filling times of 10 µs to 100 µs, a small peak appears. As the DLTS filling pulse width increases, the peak increases in response to more traps being filled, but it also broadens and shifts to lower temperature suggesting that a possible series of trap levels, perhaps forming a defect band, are present. The peaks usually saturate in a timeframe of seconds. These filling times are sufficient for electrons to fill traps near the interface from the n-type side of the device due to a thermionic emission current. Admittance spectroscopy data for the same samples are shown for comparison. INTRODUCTION CuInGaSe2 (CIGS) single-junction solar cells have achieved conversion efficiencies exceeding 19% [1]. Alloys from CuInSe2 (CIS) to CuGaSe2 (CGS) can span a band gap range from 1.04 eV to 1.7 eV. So, besides CIGS becoming a worthy candidate for efficient singlejunction thin-film solar cells, various compositions may also be useful for potentially evenhigher-efficiency multi-junction solar cells [2]. Measuring and characterizing defects is one approach toward understanding and improving limited performance. Here, we show deep level transient spectroscopy (DLTS) [3] and admittance spectroscopy data for CIGS, CIS, and CGS samples. The samples measured here represent the growth processes that have produced highefficiency solar cells. The particular samples that were measured have near-record efficiencies of 18.5%, 14.5%, and 10.2% for CIGS, CIS, and CGS, respectively. EXPERIMENT DLTS data were collected using a Fourier-transform based system from Accent Optical Technologies [4]. Samples were reverse biased, and pulses to zero bias were used to fill traps. Positive peaks correspond to capturing minority carriers. The data were plotted on Arrhenius plots and analyzed using the standard conventions [3,5]. Admittance spectroscopy data were collected at zero bias using an Agilent 4294A impedance analyzer with a frequency range from 40 Hz to over 1 MHz. The amplitude of the ac signal was 50 mV. Data were analyzed using the methodology developed by Walter et al. [6].
F5.35.2
While admittance spectroscopy could be measured on the 0.43-cm2 samples, the DLTS capacitance measurement is limited to about 1000 pF, and the sample areas were reduced to about 0.02 cm2 by m
Data Loading...
