Optical characterization of copper indium gallium diselenide thin films

Hebert, Damon

Permalink

https://hdl.handle.net/2142/42271 Copy

Description

Title

Optical characterization of copper indium gallium diselenide thin films

Author(s)

Hebert, Damon

Issue Date

2013-02-03T19:29:53Z

Director of Research (if dissertation) or Advisor (if thesis)

Rockett, Angus A.

Doctoral Committee Chair(s)

Rockett, Angus A.

Committee Member(s)

• Abelson, John R.
• Bishop, Stephen G.
• Lyding, Joseph W.

Department of Study

Materials Science & Engineerng

Discipline

Materials Science & Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Date of Ingest

2013-02-03T19:29:53Z

Keyword(s)

• solar cell
• copper indium gallium diselenide (CIGS)
• Cu(In,Ga)Se2
• photoluminescence
• cathodoluminescence
• thin film
• electron backscatter diffraction (EBSD)
• band tail
• grain boundary
• emission
• sodium
• anneal
• Cadmium sulfied (CdS)

Abstract

Cu(In,Ga)Se2 (CIGS) and its alloys are the leading choice for thin film photovoltaic absorber layers due to their high performance in devices, low degradation, high optical absorption coefficient and high tolerance to off-stoichiometry and intrinsic defects. Film conductivity and recombination losses are controlled by intrinsic point defect concentrations, especially in the near-surface space-charge region of the heterojunction. Despite the amount of research already performed on CIGS alloys, their optoelectronic properties, defect chemistry and recombination mechanisms are still poorly understood. The focus of this dissertation is to optically characterize a selection of CIGS absorber layers fabricated by various techniques in order to better understand the radiative emission and defect physics. This work aims to identify the defects responsible for recombination and their relation to grain boundaries and band edge fluctuations, which limit device performance. This study used photoluminescence (PL) spectroscopy, photoluminescence excitation (PLE) spectroscopy, and cathodoluminescence (CL) to study radiative emissions from a variety of Cu-poor CIGS thin films. Three general types of CIGS films were analyzed. Polycrystalline layers deposited on Mo-coated soda lime glass, polycrystalline layers deposited on metal foil, and epitaxial films grown on (100) and (111) GaAs were analyzed in this work. This work concludes that the donor-acceptor pair recombination model used in most interpretations of CIGS emission should be replaced with a model that accounts for high compensation and band edge fluctuations, which is shown to be undoubtedly the case in Cu-poor CIGS. Within this model, the most commonly observed emissions were explained as free-to-bound types, specifically iii band-to-impurity (BI) and tail-to-impurity (TI) types. Band tail width was measured by PLE. A correlation was established between band tail width and device efficiency. CIGS absorber layers that produced devices of higher performance showed narrower band tails. CL and PL showed an additional deep emission in Na-free films, not present in Nacontaining films grown in parallel. It is concluded that most grain boundaries in CIGS act as collection areas for point defects and point defect clusters but also are more or less inactive with respect to recombination due to their built-in electrostatic hole barrier. Spectral and spatial emission characteristics were studied on plan-view CIGS surfaces that were covered with a ~50 nm thick CdS film by chemical bath deposition (CBD). It is concluded that spectral changes that others have observed in the emission of CdS-treated films is a result of the CBD process itself and not the resulting film or the formation of the heterojunction. The effect of low temperature (~180°C) air annealing on the emission characteristics of CdS/CIGS thin films was studied by cryogenic infrared and visible PL. Spectral shape was not significantly affected by annealing for either film, but PL intensity did show some dependence on anneal time for both films, which led to an estimate of an optimal time window of 3-10 hours for low temperature annealing.

Graduation Semester

2012-12

Permalink

http://hdl.handle.net/2142/42271

Copyright and License Information

Copyright 2012 Damon Hebert

Owning Collections