Development of a time-resolved photoluminescence imaging system using a compressed sensing approach

Baltušis, Aidas (2025) Development of a time-resolved photoluminescence imaging system using a compressed sensing approach. PhD thesis, University of Glasgow.

Abstract

In this work, a novel measurement technique for time-resolved photoluminescence (TRPL) imaging of semiconductors is developed using a compressed sensing approach. TRPL provides crucial insights into the charge carrier lifetimes of semiconductor materials, which directly reflect material quality and influence the efficiency of electron-hole recombination processes—key factors in the performance of devices such as solar cells and LEDs. However, conventional TRPL techniques are limited in their ability to efficiently capture high spatial resolution data, often requiring slow, point-by-point measurements that are time-intensive and unsuitable for large or non-uniform samples. The compressed sensing method developed in this work overcomes these limitations by enabling simultaneous measurement of multiple spatial points, reducing the number of measurements required while maintaining high-resolution imaging and accuracy.

The first part of the thesis develops a comprehensive simulation framework for compressed sensing TRPL measurements, beginning with simple 1D models and expanding to complex simulations in both the spectral and temporal domains. The simulation follows a statistical approach, integrating key factors such as photon pile-up and noise to ensure realistic modelling of experimental conditions. A reconstruction algorithm is also developed to handle the high-dimensional data generated from compressed sensing TRPL, forming the foundation for the design and implementation of the experimental system. The simulation results show that in some cases, as few as 2% of measurements can be sufficient compared to conventional methods, demonstrating the effectiveness of compressed sensing.

The second part of this thesis focuses on the design and construction of a TRPL measurement system, specifically developed for imaging semiconductor samples with emission in the near-infrared (NIR) spectral range using a raster scanning approach. The experimental system was setup using 640 nm pulsed laser for excitation and two photomultiplier tube detectors, allowing detection between 700-1600 nm. Point measurements were conducted on cadmium telluride (CdTe) and copper indium gallium selenide (CIGS) samples, while full raster-scan measurements were performed on CdTe. These experiments demonstrated the system’s capability but highlighted challenges such as laser light leakage and divergence of the laser beam.

In the final part of the thesis, the experimental system is modified, to demonstrate a proof-of-concept compressed sensing TCSPC imaging system for acquiring TRPL maps of semiconductor materials and devices. The TRPL imaging results obtained using the compressed sensing approach are compared with those acquired through a conventional point-by-point method over the same excitation area. The feasibility of this methodology is clearly demonstrated, highlighting reductions of 50% or more in measurement acquisition time. Additionally, the benefits and challenges of the experimental prototype system are presented and thoroughly discussed, laying the groundwork for further improvements in system design and application.

Overall, these results provide a pathway toward an improved approach to TRPL imaging, with the first example of a compressed sensing TRPL system. While this thesis primarily focuses on photovoltaic applications, the findings are applicable to other wavelengths and material systems. Future work can build on this foundation to overcome the remaining limitations and further enhance the technique's capabilities.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Additional Information: Primary funding from Physical Sciences Research Council (EPSRC). Financial support in the form of iCASE studentship from National Physical Laboratory (NPL) and IQE plc.
• Subjects: T Technology > T Technology (General)
• Colleges/Schools: College of Science and Engineering > School of Engineering
• Funder's Name: Engineering and Physical Sciences Research Council (EPSRC), National Physical Laboratory (NPL), IQE plc
• Supervisor's Name: Sweeney, Professor Stephen
• Date of Award: 2025
• Depositing User: Theses Team
• Unique ID: glathesis:2025-85308
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 12 Jun 2025 09:31
• Last Modified: 07 Jul 2025 14:26
• Thesis DOI: 10.5525/gla.thesis.85308
• URI: https://theses.gla.ac.uk/id/eprint/85308

Related URLs

• Conference proceeding
• Article DOI
• Conference item