Controlled thin film growth through machine-learning based closed-loop feedback with online X-ray scattering analysis

In recent years, X-ray scattering has received a significant boost due to the increased use of machine learning strategies in the analysis of data acquired at synchrotron sources. This data has been collected based on proposal MI-1462. We leveraged ML based real-time data analysis integrated into modern beamline environments. Integrating with BLISS all online fit results are saved together with raw data in this dataset. As a proof-of-concept, we grew organic molecular thin films of Alq3 (C27H18AlN3O3) and PTCDI-C8 (N,N′-dioctyl-3,4,9,10-perylene tetracarboxylic diimide). The growth is monitored with X-ray reflectivity and we established a closed loop between real-time, ML-based online data analysis and the sample environment to tailor the deposition process of organic thin films on molecular monolayer level.