OICC Press International Journal of Industrial Chemistry 2228-5547 17 1 2026 03 31 Enzymatic Urea Biosensor on Screen-Printed Carbon Electrodes Based on a Hierarchical Polypyrrole/Reduced Graphene Oxide Scaffold with Machine Learning- Assisted Analysis 10.57647/ijic.2026.1701.03 EN Roozbeh Jamaliyan Department of Chemical Engineering, Qu.c., Islamic Azad University, Quchan, Iran Afshin Farahbakhsh Department of Chemical Engineering, Qu.c., Islamic Azad University, Quchan, Iran Javad Mohebbi Najm Abad Department of Chemical Engineering, Qu.c., Islamic Azad University, Quchan, Iran Susan Khosroyar Department of Chemical Engineering, Qu.c., Islamic Azad University, Quchan, Iran Journal Article 2026 03 31 This study presents a cost-effective enzymatic urea biosensor developed on a screen-printed carbon electrode (SPCE) using a sequential layer-by-layer modification strategy. A conductive polypyrrole (PPy) scaffold was electrochemically deposited to mitigate the restacking of reduced graphene oxide (rGO) sheets, creating a high-surface-area matrix for the covalent immobilization of urease (Urs). Evaluated in phosphate-buffered saline (PBS), the biosensor exhibited a linear response from PBS blank (0 mM) to 30 mM urea, with a calculated limit of quantification (LOQ) of 12.17 µM, a limit of detection (LOD) of 3.65 µM, and an area-normalized sensitivity of 3.68 µA·mM⁻¹·cm⁻². The device demonstrated high precision with relative standard deviations (RSDs) below 3%, robust storage stability (93% retention after 4 weeks), and good selectivity against physiological interferents. Beyond experimental testing, an explainable machine learning (ML) workflow was used as a supportive post hoc analysis to analyze the influence of fabrication variables and to benchmark urea concentration prediction against conventional electrochemical calibration. Overall, this work presents an engineering-optimized PPy/rGO/Urs modification strategy compatible with disposable SPCE fabrication for urea sensing in PBS and provides a transparent, ML-assisted data-analysis framework for sensor optimization and interpretation under controlled bench-validation conditions.