Post ferric-substitution detection method optimization for Ni(II)-organic complexes measurement: Simulation,...

Nickel (Ni(II)) complexes, especially those formed with strong ligands such as ethylenediaminetetraacetic acid (EDTA), are difficult to quantify due to their low environmental concentrations and weak ultraviolet (UV) absorbance. These characteristics limit the effectiveness of conventional spectrophotometric methods. Among indirect detection strategies, Fe(III) substitution methods has emerged as a viable approach. However, the associated parameters have not been systematically optimized, resulting in limited sensitivity and practical application. In this study, we systematically refine the Fe(III) substitution approach by simulation-guided experimental design, machine-learning based variables importance analysis, and predictive modeling analysis. Thermodynamic simulations and density functional theory (DFT) calculations guided experimental design. Under optimized conditions, the method achieved a detection limit as low as 1 × 10⁻³ mM for Ni-EDTA. Application in surface water, groundwater, and electroplating wastewater showed strong linearity (R² > 0.96) and good matrix tolerance. In addition, machine learning models were utilized to interpret variable importance and predict recovery performance. Notably, Random Forest Regression (RFR) model demonstrated superior predictive performance (R² = 0.951) and revealed that both pH and water bath duration are critical factors. This research successfully develops and optimizes a reliable Fe(III) substitution method for environmental monitoring of Ni complexes. The combined approach represents a significant advancement in water quality analysis and provides a promising strategy for addressing the challenges posed by Ni(II) complexes in complex aqueous environments.