Explainable Machine Learning for Methylene Blue Removal Under Irradiation: Assessing Nominal Cu-Loading in AC@NiO

Abstract

The time-dependent removal performance under irradiation of activated-carbon-supported nickel oxide (AC@NiO) samples containing nominal Cu-loadings of 0%, 3%, 5%, 7.5%, and 10% was examined using experimental concentration data and an explainable machine-learning (ML) framework. Reaction time and nominal Cu-loading were used as predictors. Preliminary screening with Ct/C0 as the common target showed that Random Forest (RF) outperformed K-Nearest Neighbors, Multi-Layer Perceptron, and Support Vector Regression. The target-specific regularized RF models were fitted for Ct/C0 and ln(C0/Ct); normalized apparent removal efficiency was derived deterministically as η/100=1−Ct/C0. Across 10 controlled random seeds, the mean leave-one-out cross-validation R2 values were 0.9691±0.0013 for Ct/C0 and the derived η/100, and 0.9152±0.0090 for ln(C0/Ct). A fixed chronological evaluation, trained at t≤100 min and evaluated at 100<t≤120 min, yielded mean R2 values of 0.9668±0.0073 and 0.8890±0.0163, respectively. Because the composition-specific RF predictions were constant across this boundary interval, these scores are interpreted as later-time boundary diagnostics rather than evidence of temporal extrapolation. Point-estimate SHAP, impurity-based, and permutation-based analyses assigned greater predictive importance to reaction time, whereas moving-block bootstrap results showed that this ranking was sensitive to temporal resampling. Among the five tested compositions, the nominal 5% Cu-containing sample exhibited the most favorable apparent removal profile. This finding is specific to the investigated conditions and does not establish a universal optimum or a causal physicochemical mechanism.