Physics-informed machine learning for TCO-layer thickness prediction and process analysis from multi-spectral images
Source abstract
Abstract
We present a machine learning model for a robust and fast evaluation of thickness maps of Transparent Conducting Oxide (TCO) layers based on multichannel spectral images only. The model is applicable for the quality inspection of heterojunction solar cells with textured surfaces and an amorphous silicon layer stack beneath the TCO layer. Within our physics-informed approach, synthetic data are created online for model training by simulating reflection maps for given TCO thickness variations. The developed method determines a full-scale TCO-thickness map in 1 s from inline measurable RGB image data. The spatially resolved analysis allows inline quality inspection of the thickness distributions. Additionally, the thickness profile at the edges is inspected with high spatial resolution in Silicon heterojunction solar cell precursors, where TCO edge exclusion at the rear side is required to avoid shunting. We demonstrate our approach by quantifying the completeness and masking area for narrow masks, which is a process optimization step for increasing cell efficiency. We derive sorting criteria for an early-stage process control regarding shunts and quantify the influence of the positioning accuracy of the mask on the short-circuit current. • Physics-informed machine learning model trained purely with synthetic data. • Fast and full-scale thickness determination of TCO layers from inline measurements. • Quality inspection for heterojunction solar cells with textured surfaces. • Early-stage process control regarding edge exclusion properties.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Industrial Vision Systems and Defect Detection · Electron and X-Ray Spectroscopy Techniques · Machine Learning in Materials Science
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (16)
Show source-provided references
1. Infrared light management in high-efficiency silicon heterojunction and rear-passivated solar cells10.1063/1.4772975 ↗
2. Influence of cell edges on the performance of silicon heterojunction solar cells10.1016/j.solmat.2022.111605 ↗
3. Improved layer properties combined with light soaking enabling for 23% efficient silicon heterojuction solar cells
4. Characterization of thin-film structures of silicon heterojunction solar cells with inline reflectance spectroscopy
5. Inline characterization of ultrathin amorphous silicon stacks in silicon heterojunction solar cell precursors with differential reflectance spectroscopy10.1109/JPHOTOV.2023.3301132 ↗
6. S. Kumar, C. Diestel, S. Al-Hajjawi, J. Schmitz, M. Hemsendorf, J. Haunschild, S.J. Rupitsch, and S. Rein, “Inline mapping of amorphous silicon layer thickness of heterojunction precursors using multispectral imaging”, in SiliconPV 2024, 14th International Conference on Crystalline Silicon Photovoltaics.10.52825/siliconpv.v2i.1324 ↗
7. Rapid and accurate thin film thickness extraction via UV-vis and machine learning
8. Machine learning powered ellipsometry10.1038/s41377-021-00482-0 ↗
9. Thickness and refractive index measurements of a thin-film using an artificial neural network algorithm10.1088/1681-7575/acb70d ↗
10. Non-destructive thickness characterisation of 3D multilayer semiconductor devices using optical spectral measurements and machine learning10.37188/lam.2021.001 ↗
11. A self-consistent hybrid model connects empirical and optical models for fast, non-destructive inline characterization of thin, porous silicon layers10.1051/epjpv/2022035 ↗
12. Byrnes · 2020
13. Reflection of normally incident light from silicon solar cells with pyramidal texture10.1002/pip.1050 ↗
14. Yun · 2007 · 298
15. Ronneberger · 2015
16. “Lessons learned from 25 Years production Technology research & development” 6 pages/8th world conference on photovoltaic