New machine learning process monitoring methodologies based on spectroscopic techniques for inline inspection of thin film PV industrial manufacturing processes (MALMO)

Abstract

Industrial production of thin-film photovoltaic (TFPV) devices presents a high degree of complexity in which small deviations from standard manufacturing conditions result in defective products and a waste of materials, energy and time. This can be minimized through early detection of deviations using in-line process monitoring (PM) tools. X-ray fluorescence (XRF), Raman and photoluminescence spectroscopies are well-suited for PM of TFPV device manufacturing providing compositional, structural and optoelectronic information in a non-destructive fast way. However, their use for PM requires developing methodologies that allow generating useful information that correlates with the fabrication process. The MALMO project aims at developing and demonstrating in an industrial environment advanced statistical and machine learning PM methodologies for their implementation in a XRF-Raman-photoluminescence characterization platform in an existing industrial Cu(In,Ga)Se2 solar foil roll-to-roll manufacturing pilot line. The implementation of these methodologies will allow optimizing the fabrication process and will be extrapolable to other industrial applications.

Acknowledgements

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under Marie Skłodowska-Curie grant agreement No. 801342 (Tecniospring INDUSTRY) and the Government of Catalonia’s Agency for Business Competitiveness (ACCIÓ).