Steepest-Entropy-Ascent Framework for Predicting Arsenic Adsorption on Graphene Oxide Surfaces -- A Case Study

TL;DR

This paper introduces a thermodynamic framework based on steepest-entropy-ascent quantum thermodynamics to predict arsenic adsorption on graphene oxide, capturing both equilibrium and transient behaviors without empirical rate laws.

Contribution

The study develops a novel, fully predictive model using SEAQT framework for arsenic adsorption, integrating quantum thermodynamics and neural networks for accurate, non-empirical predictions.

Findings

Equilibrium capacities match experimental data within 5%.

Model accurately predicts adsorption kinetics.

Framework provides thermodynamically consistent predictions.

Abstract

Water contamination by arsenic(V) constitutes a major public-health concern, underscoring the need for models that capture both equilibrium and transient adsorption behaviour. A framework that can do so is the steepest-entropy-ascent quantum thermodynamic (SEAQT) framework, which is used here to describe the uptake of As(V) on graphene oxide (GO) across pollutant concentrations of 25-350 mg/L. A non-equilibrium equation of motion derived from the steepest-entropy-ascent principle for a five-component system (water, arsenic, two GO functional groups, and protons is solved with an energy eigenstructure generated by a Replica-Exchange Wang-Landau algorithm and then extrapolated to relevant contaminant concentrations via an artificial neural network. Without recourse to empirical rate laws, the model predicts the time-dependent adsorption capacity, the stable-equilibrium arsenic concentration, and the pH dependence of removal efficiency. Equilibrium capacities are reproduced within 5 % of experimental isotherms, and the characteristic adsorption time aligns with the reported kinetics. These results indicate that SEAQT framework provides a thermodynamically consistent, fully predictive tool for designing and optimising adsorbent-based water-treatment technologies.