Project: High-efficiency heterojunction nano-photocatalysts for visible-light-driven degradation of CECs using machine learning (WP1)
Host institution: KU Leuven
Supervisor(s): Raf Dewil (PhD promoter)
Objectives: To synergistically integrate machine learning with nanophotocatalysis, augment contaminant mineralization, and utilize solar energy.
DC1 will employ advanced machine learning algorithms to optimize the design of Z-scheme nano-photocatalysts, with focus on optimizing light absorption, charge dynamics, and surface reactions crucial for the degradation of CECs under visible-light irradiation. The understanding of photoactivity-structure relationships is vital for predicting the performance of nano-photocatalysts in breaking down various CECs efficiently. The task involves utilizing density functional theory calculations to feed the AI models, which will guide the experimental design towards achieving maximum degradation efficiency. The AI-driven methodology will analyse parameters such as photocatalyst dose, pH, illumination time and CEC concentration, to establish optimal conditions for photocatalytic performance. By creating a robust multivariate statistical model, DC1 will map the relationship between photocatalyst characteristics and performance metrics. This innovative approach aims to enhance the photocatalytic degradation efficiency and to model the reaction kinetics for various CECs.
Expected results: Novel nano-photocatalyst to efficiently degrade CECs under visible-light. Knowledge on the degradation mechanism of organic pollutants. AI-based predictive model on pollutant degradation.
Planned secondments:
- WSU (Sup.: L. Sheppard): M18-20; 3 months: Design of high rate solar photocatalysis reactors;
- WATCH (Sup.: D. Du Pasquier): M28-31; 3 months: training in embryotoxicity assessment of nanoparticles using Xenopus Eleuthero embryo Thyroid Assays (XETA)
Enrolment in Doctoral degree: KUL Arenberg Doctoral School of Science, Engineering and Technology (BE)
