AI-Driven Biohybrid Photocatalysis for CO₂ Conversion & Solar Fuels

Aim

Workshop Objectives

Workshop Structure

📅 Day 1: Biohybrid Nanocatalysis & Artificial Photosynthesis

• Introduction to biohybrid nanocatalysis and semiconductor–biological systems
• Natural, artificial, and semi-artificial photosynthesis
• Semiconductor photocatalysis, band-gap engineering, and solar-energy harvesting
• Charge generation, separation, recombination, and interfacial electron transfer
• Semiconductor–microbe and semiconductor–enzyme interfaces
• Extracellular electron transfer and biological utilization of photogenerated electrons
• Key photocatalytic materials including TiO₂, ZnO, g-C₃N₄, BiVO₄, Fe₂O₃, and CdS
• Material properties governing photocatalytic activity, stability, and biohybrid performance
• Materials databases and computational screening of semiconductor photocatalysts

🛠️ Hands-on:

• Explore semiconductor materials using Materials Project datasets
• Analyse band gap, composition, stability, density, and electronic properties
• Compare and filter candidate photocatalysts based on activity, stability, and toxicity-related criteria
• Visualise structure–property relationships using Python
• Prepare a shortlisted semiconductor dataset for biohybrid-system design

🧰 Tools Covered:

Google Colab, Python, Materials Project, mp-api, pymatgen, pandas, NumPy, Matplotlib, and semiconductor datasets

📅 Day 2: AI-Guided CO₂ Conversion & Biohybrid Performance Prediction

• Solar-driven CO₂ conversion, carbon fixation, and microbial electrosynthesis
• Photobiocatalytic CO₂ reduction and semiconductor–microbe electron-transfer mechanisms
• Microbial and enzymatic pathways for carbon conversion and reducing-power generation
• Production of solar fuels and value-added chemicals including hydrogen, formate, acetate, alcohols, and organic acids
• Relationship between material properties, biological conditions, and conversion performance
• Development of AI-ready biohybrid nanocatalysis datasets
• Feature engineering for material, biological, and operating parameters
• Random Forest and XGBoost models for CO₂-conversion and product-yield prediction
• Model evaluation using R², MAE, and RMSE
• Explainable AI and SHAP-based interpretation of structure–property–performance relationships

🛠️ Hands-on:

• Integrate semiconductor, biological, and operating-condition data into an AI-ready dataset
• Prepare features including band gap, catalyst type, pH, temperature, light intensity, and reaction conditions
• Train and compare Random Forest and XGBoost prediction models
• Evaluate model performance and interpret influential parameters using SHAP
• Rank semiconductor–biological combinations based on predicted CO₂-conversion potential

🧰 Tools Covered:

Google Colab, Python, pandas, NumPy, scikit-learn, XGBoost, SHAP, Matplotlib, and biohybrid nanocatalysis datasets

📅 Day 3: Safe-by-Design Biohybrids & Sustainable Optimization

• Reactive oxygen species, oxidative stress, and microbial biocompatibility
• Heavy-metal leaching, nanomaterial toxicity, and semiconductor–microbe interactions
• Material degradation, long-term stability, catalyst recyclability, and regeneration
• Balancing photocatalytic efficiency with biological compatibility and safety
• Safe-by-design principles for biohybrid nanocatalysts
• Sustainability, life-cycle considerations, and scale-up of artificial-photosynthesis systems
• Multi-objective optimization of efficiency, stability, toxicity, cost, and sustainability
• Pareto optimization and performance trade-offs for candidate selection
• AI-assisted ranking of biohybrid systems for sustainable biomanufacturing
• Future directions in autonomous and AI-guided biohybrid-material discovery

🛠️ Hands-on:

• Integrate material, AI-predicted performance, safety, and sustainability data from Days 1 and 2
• Compare high-efficiency and safe-and-stable photocatalyst candidates
• Perform multi-criteria and Pareto-based candidate ranking
• Generate comparative plots and optimization visualisations
• Identify semiconductor–biological combinations with the best balance of performance, safety, and sustainability
• Develop a preliminary safe-by-design biohybrid nanocatalyst shortlist for CO₂ conversion and solar biomanufacturing

🧰 Tools Covered:

Google Colab, Python, pandas, scikit-learn, XGBoost, SHAP, Matplotlib, multi-objective optimization workflows, and candidate-ranking templates

Who Should Enrol?

Important Dates

Workshop Outcomes