Intelligently optimized electrospun polyacrylonitrile/poly(vinylidene fluoride) nanofiber: Using artificial neural networks

Abstract

In this work, an intelligent approach is applied for the first time in the modelling and optimization of electrospun Polyacrylonitrile/Poly(Vinylidene Fluoride) (PAN/PVdF) nanofiber properties. A genetic algorithm-based computer code was developed to optimize the architecture of an artificial neural network, by which an accurate model was developed for the prediction of nanofiber diameter, the standard deviation of nanofiber diameter and porosity of electrospun membrane. Electrospinning of polyacrylonitrile/poly(vinylidene fluoride) (PAN/PVdF) was applied to obtain a quantitative relationship between selected electrospinning parameters (namely applied voltage, solution concentration, and PVdF composition) and nanofiber diameter, the standard deviation of nanofiber diameter and porosity of electrospun membrane. The morphology and nanofiber diameter were investigated by field emission scanning electron microscopy (FESM). The range of produced nanofiber diameters was from 116 to 379 nm. It seemed that the nanofiber diameter and standard deviation of nanofiber diameter decrease with PVdF composition and increase with solution concentration. The applied voltage had no important effect on the nanofiber diameters. The porosity of the electrospun membrane decreases with solution concentration and increases with PVdF composition.

Keywords

• polymer composites
• artificial intelligence
• electrospinning
• genetic algorithm-based computer code
• nanofiber diameter