Halloysite nanotube/black phosphorene nanohybrid modified screen-printed carbon electrode as an ultra-portable electrochemical sensing platform for smartphone-capable detection of maleic hydrazide with machine learning assistance

Abstract

With the assistance of machine learning (ML), black phosphorene (BP) stabilized by silver nanoparticles (AgNPs) is used to modify halloysite nanotube (HNT) to obtain highly conductive nanomaterials, HNT/BP-AgNPs, which are morphologically characterized and elementally analyzed. Artificial neural network (ANN) and least squares support vector machine (LS-SVM) are adopted for the intelligent and rapid analysis of maleic hydrazide (MH). An ultra-portable electrochemical sensor bases on HNT/BP-AgNPs modifying screen-printed carbon electrode (SPCE), smartphone and mini-palm potentiostat for detection of MH in the linear range 0.7–55 μM with limit of detection (LOD) of 0.3 μM. For comparison, a traditional electrochemical sensor is fabricated by glass carbon electrode (GCE), desktop computer and large electrochemical potentiostat, and the linear range is 0.3–600 μM with low LOD of 0.1 μM. The ultra-portable electrochemical sensor combined with ML for the detection of MH in sweat potato and carrot gain satisfactory recoveries.