Fermentation process optimisation based on ANN and RSM for xylitol production from areca nut husk followed by xylitol crystal characterisation

Abstract

Areca nut husk is an agricultural bio-waste with massive potential for producing value-added chemicals by conversion of xylose present in it through fermentation. This study aimed to optimise the production of xylitol from areca nut husk hemicellulosic acid hydrolysate (as the substrate) by using Candida tropicalis (MTCC 6192). Response Surface Methodology (RSM) and Artificial Neural Network (ANN) were compared for fermentation process optimisation in modelling and prediction. Agitation, pH, temperature, xylose concentration and operation time were considered variable parameters. The maximum concentration of xylitol obtained was around 9.96 g/L under designed conditions. Crystallisation analysis was conducted to separate pure xylitol crystal from fermented broth, followed by characterisation of xylitol crystal such as X-ray diffraction (XRD), Field emission scanning electron microscopy (FESEM), Thermogravimetric Analysis (TGA), Differential scanning calorimetry (DSC) and Raman analysis. DSC analysis revealed that the melting process of commercial xylitol crystals after crystallisation (CXCAC) was observed at 93.71 °C with enthalpies of 210.7 J/g. However, almost similar results were observed for xylitol crystals from fermentation broth (XCFB), with a melting point at 91.73 °C and enthalpies of 220 J/g.