Search

Photocurrent response mapping is a powerful imaging technique for assessing defects and losses in photovoltaic devices. However, it has not enjoyed widespread application because high‐resolution measurements of large samples can last several hours, while weak signals from micrometer‐sized laser spots require lock‐in amplification. An alternative approach presented recently is the use of digital micromirror devices combined with compressed sensing theory. There are significant benefits when using such methods, such as signal amplification, undersampling options, and simplified measurement systems. Nevertheless, high computational requirements have limited the experimental application of this method to low‐resolution outputs. Herein, the mathematical background and the experimental approach toward megapixel resolution, ultrafast compressed sensing current mapping are presented, overcoming previous computational and experimental barriers. A high‐power digital light processing projection system is developed for the experimental application. Solutions to computational issues, sampling optimization, and measurement strategies are presented and the flexibility of the system regarding the sizes of photovoltaic devices that can be measured is demonstrated. [ABSTRACT FROM AUTHOR]

Spatial characterisation methods for photodetectors and other optoelectronic devices are necessary for determining local performance, as well as detecting local defects and the non-uniformities of devices. Light beam induced current measurements provide local performance information about devices at their actual operating conditions. Compressed sensing current mapping offers additional specific advantages, such as high speed without the use of complicated experimental layouts or lock-in amplifiers. In this work, the signal amplification advantages of compressed sensing current mapping are presented. It is demonstrated that the sparsity of the patterns used for compressive sampling can be controlled to achieve significant signal amplification of at least two orders of magnitude, while maintaining or increasing the accuracy of measurements. Accurate measurements can be acquired even when a point-by-point scan yields high noise levels, which distort the accuracy of measurements. Pixel-by-pixel comparisons of photocurrent maps are realised using different sensing matrices and reconstruction algorithms for different samples. The results additionally demonstrate that such an optical system would be ideal for investigating compressed sensing procedures for other optical measurement applications, where experimental noise is included. [ABSTRACT FROM AUTHOR]