Abstract
In this work, response surface methodology and adaptive neuro-fuzzy inference system approaches were used to predict and model effect of extraction conditions of pectin from medlar fruit (Mespilus germanica L.). The pectin extracted at optimized conditions (89 degrees C, 4.83 h and 4.2 pH) could be classified as high methoxyl pectin. Sugar composition analysis showed that pectin was mainly composed of D-galacturonic acid, c-arabinose, c-rhamnose, D-galactose and D-glucose. Fourier Transform Infrared Spectroscopy, RAMAN and nuclear magnetic resonance spectra confirmed molecular structure, revealing presence of D-galacturonic acid backbone. X-ray diffraction patterns revealed an amorphous structure. Differential scanning calorimetry showed endothermic (123 degrees C) and exothermic peaks (192 degrees C). Thermogravimetric analysis revealed three decomposition regions, 50-225 degrees C, 225-400 degrees C and 400-600 degrees C. Steady and dynamic shear analyses revealed that pectin had a pseudo-plastic behavior with storage (G') and loss (G '') modulus increasing with increment in frequency, indicating viscoelastic structure more predominantly elastic than viscous.