Unveiling Competitive Adsorption in TiO₂ Photocatalysis through Machine-Learning-Accelerated Molecular Dynamics, DFT,...

Area of Science:

• Environmental Science
• Materials Science
• Physical Chemistry

Background:

• Solar photocatalysis is crucial for water treatment but hindered by cosolute interference.
• Nontarget organic molecules adsorb onto photocatalysts, reducing pollutant degradation efficiency.

Purpose of the Study:

• To computationally predict the inhibitory effects of organic molecules on titanium dioxide (TiO2) photocatalysis.
• To understand the molecular interactions governing competitive adsorption during pollutant degradation.

Main Methods:

• Employed Density Functional Theory (DFT) and machine learning interatomic potentials (MLIPs).
• Investigated Langmuir-Hinshelwood adsorption dynamics for various cosolutes and para-chlorobenzoic acid (pCBA).
• Utilized MLIPs for extended simulations of adsorption behavior.

Main Results:

• Trimesic acid and gallic acid showed the highest inhibition of pCBA degradation, followed by succinic acid.
• Molecular structure, including spatial arrangement and electronic interactions, significantly influenced adsorption and inhibition.
• MLIPs enabled deeper insights into dynamic adsorption processes over longer timescales.

Conclusions:

• Molecular-level interactions are critical for optimizing photocatalytic water treatment.
• Computational modeling, especially MLIPs, provides a powerful tool for predicting and understanding cosolute effects.
• Findings pave the way for designing more robust photocatalysts for environmental applications.