Experimental and Machine Learning Investigation of Poly‐ε‐caprolactone‐MXene Composites for Methylene Blue Capture

Abstract The upcycling of polymeric waste into high‐performance adsorbents presents a sustainable strategy for wastewater remediation. In this study, a nanofibrous adsorbent was fabricated via electrospinning using recovered poly(ε‐caprolactone) (PCL) waste sourced from 3D printing residues. To enhance adsorption performance, Ti 3 C 2 T x MXene and a microwave‐assisted ionic liquid‐functionalized composite (MW‐Ti 3 C 2 T x ‐IL) were incorporated, respectively, resulting in the development of 3PM and 3PMIL. MW‐assisted functionalization significantly improved methylene blue (MB) adsorption, with 3PMIL demonstrating an adsorption capacity of 360 ± 2.65 mg/g, which substantially surpassed that of pristine PCL (105.81 ± 0.013 mg/g) and 3PM (291.93 ± 1.57 mg/g). Adsorption kinetics followed a pseudo‐second‐order model, while the Langmuir isotherm indicated monolayer chemisorption, attributed to electrostatic interactions and enhanced surface functionalities. A machine learning (ML) model, trained using bootstrapped datasets, accurately predicts adsorption performance (R 2 = 0.9), with predictions closely aligning with experimental values. Correlation analysis reveals that m/v ratio, dosage, pH, and dye concentration are key descriptors influencing adsorption behavior. Structural and surface analyses confirm successful MXene incorporation and functionalization, contributing to enhanced adsorption. The nanofibers retain biodegradability in simulated environments, highlighting their environmental compatibility. This work presents a scalable, data‐driven strategy for designing biodegradable, high‐performance adsorbents from plastic waste, advancing sustainable solutions for wastewater treatment.