0
Article ? AI-assigned paper type based on the abstract. Classification may not be perfect — flag errors using the feedback button. Tier 2 ? Original research — experimental, observational, or case-control study. Direct primary evidence. Human Health Effects Marine & Wildlife Remediation Sign in to save

Machine learning algorithm for modeling oxytetracycline adsorption kinetics on microplastics in marine environments

Journal of Hazardous Materials Advances 2025 1 citation ? Citation count from OpenAlex, updated daily. May differ slightly from the publisher's own count.
Jiayi Xin, Hongyan Xing

Summary

Microplastics in the ocean don't just float — they actively adsorb and concentrate other pollutants like antibiotics, potentially acting as vectors that deliver these chemicals into marine organisms. This study built a mathematical model of how the antibiotic oxytetracycline adsorbs onto microplastics in seawater, and used quantum machine learning to dramatically speed up the computational modeling of these complex, multi-component interactions. Faster and more accurate models of microplastic-pollutant binding behavior could improve our ability to assess the true toxicological risk that plastic-chemical combinations pose to marine life.

Marine plastic waste undergoes physicochemical degradation into microplastics, which accumulate and release toxic pollutants via adsorption–desorption kinetics, enabling tissue invasion and physiological disruption in marine organisms. This study established a mathematical model to characterize the adsorption kinetics of typical pollutants on microplastics in marine environments using convection diffusion reaction (CDR) equation. The adsorption of microplastics in marine environments exhibits complex nonlinear dynamics, characterized by coupled transport reaction processes. Traditional numerical methods result in high computational costs during the processing. By combining quantum state encoding with machine learning, a microplastic adsorption kinetics algorithm based on quantum machine learning (QML) is proposed, which achieves acceleration in handling the nonlinear dynamics of multi-component adsorption systems. Quantum computing, through its properties of quantum superposition and entanglement, can process high-dimensional data in parallel. It has great potential in tasks such as feature extraction, classification, and regression, which can improve computational efficiency and model performance. The simulation results show that the proposed algorithm has good performance in the adsorption kinetics and isotherm fitting of typical adsorbents by microplastics in marine environments, providing theoretical and data support for studying the influence of surface property changes of microplastics on the behavior of coexisting pollutants in marine environments. • A convection diffusion equation for marine microplastic adsorption kinetics is proposed. • The multi-component characteristics of microplastic solutes is considered. • Quantum states can encode multiple component information simultaneously. • The parallel learning capability of machine learning can reduce complexity.

Share this paper