Return to search

Spatially resolved opto-electric measurements of photovoltaic materials and devices

The objective of this study is to characterize and analyse defects in solar cell devices. Materials used to fabricate solar cells are not defects free and therefore, there is a need to investigate defects in cells. To investigate this, a topographical technique was developed and employed which uses a non-destructive methodology to analyse solar cells. A system was built which uses a technique based on a laser beam induced current (LBIC). LBIC technique involves focusing light on to a surface of a solar cell device in order to create a photo-generated current that can be measured in the external circuit for analyses. The advantage of this technique is that it allows parameter extraction. Parameters that can be extracted include short-circuit current, carrier lifetime and also the external and internal quantum efficiency of a solar cell. In this thesis, LBIC measurements in the form of picture maps are used to indicate the distribution of the localized beam induced current within solar cells. Areas with low minority carrier lifetime in solar cells are made visible by LBIC mapping. Surface reflection intensity measurements of cells can also be mapped using the LBIC system developed in this study. The system is also capable of mapping photo-generated current of a cell below and above room temperature. This thesis also presents an assessment procedure capable of assessing the device and performance parameters with reference to I-V measurements. The dark and illuminated I-V characteristics of solar cells were investigated. The illuminated I-V characteristics of solar cells were obtained using a defocused laser beam. Dark I-V measurements were performed by applying voltage across the cell in the dark and measuring a current through it. The device parameters which describe the behaviour of I-V characteristic were extracted from the I-V data using Particle Swarm Optimization (PSO) method based on a one-and two-diode solar cell models. Solar cells of different technologies were analysed, namely, single-crystalline (c-Si) and multicrystalline (mc-Si) silicon, Edge-defined Film-fed Growth Si (EFG-Si) and Cu(In,Ga)(Se,S)2 (CIGSS) thin film based cells. The LBIC results illustrated the effect of surface reflection features and material defects in the solar cell investigated. IQE at a wavelength of 660 nm were measured on these cells and the results in general emphasised the importance of correcting optical losses, i.e. reflection loss, when characterizing different types of defects. The agreement between the IQE measurements and I-V characteristics of a cell showed that the differences in crystal grains influence the performance of a mc-Si cell. The temperature-dependence of I-V characteristics of a CIGSS solar cell was investigated. The results showed that, for this material, the photo response is reduced at elevated temperatures. In addition to LBIC using a laser beam, solar spectral radiation was employed to obtained device performance parameters. The results emphasised the effect of grain boundaries as a recombination centres for photo-generated hole-pairs. Lastly, mesa diode characterizations of solar cells were investigated. Mesa diodes are achieved by etching down a solar cell so that the plateau regions are formed. Mesa diodes expose the p-n junction, and therefore mesa diode analysis provides a better way of determining and revealing the fundamental current conduction mechanism at the junction. Mesa diodes avoid possible edge effects. This study showed that mesa diodes can be used to characterize spatial non-uniformities in solar cells. The results obtained in this study indicate that LBIC is a useful tool for defect characterization in solar cells. Also LBIC complements other characterization techniques such as I-V characterization.

Identiferoai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:nmmu/vital:10520
Date January 2010
CreatorsThantsha, Nicolas Matome
PublisherNelson Mandela Metropolitan University, Faculty of Science
Source SetsSouth African National ETD Portal
LanguageEnglish
Detected LanguageEnglish
TypeThesis, Doctoral, PhD
Formatix, 99 leaves ; 30 cm, pdf
RightsNelson Mandela Metropolitan University

Page generated in 0.0026 seconds