Performance of amidoxime-modified poly(acrylonitrile- Co-acrylic acid) for removal of boron in aqueous solution

Boron is an essential micronutrient for plants, humans and animals, which is also an important component in various industries. Along with the wide spread of boron application, more boron waste pollutes the water sources, and leads to a series of environment and health problems. Adsorption is the most efficient technique among many boron removal technologies; which can treat solutions containing a very low concentration of boron. This project aimed to produce amidoxime-modified poly(acrylonitrile-co-acrylic acid) (AO-modified poly(AN-co-AA)) with optimised method and to investigate the performance of AO-modified poly(AN-co-AA) on the adsorption of boron ions in batch operation. Batch adsorption was conducted at the physiochemical parameters of pH, adsorbent dosage and initial boron concentration. The isotherms and kinetics of adsorption data were studied at various initial boron concentration. Meanwhile, the Artificial Neural Network (ANN) was simulated from experimental data and applied to optimize, develop and create prediction models for boron adsorption by AO-modified poly(AN-co-AA). As a result, the optimised synthesis method at 55 ºC gave yield of 77% and the conversion of nitrile group to amidoxime at pH = 8 was 78%. The optimal operating condition for boron batch adsorption were determined as initial pH ≈ 7. Besides, the process reached its equilibrium at adsorbent dosage of 4.2 g/L and initial concentration of 41 mg/L. The best fit model for adsorption isotherm was Sips model with heterogeneity factor (n) = 0.7611. In kinetic study, the adsorption data was well fitted in Pseudo-second order equation with equilibrium rate constant (k2) = 0.0807 ± 0.0112 g. mg−¹.min−¹. In modelling section, both feed-forward and recurrent artificial neural network (ANN) have been simulated to predict the adsorption potential of synthesised polymer. Among several models, radial basis function (RBF) with orthogonal least square (OLS) algorithm displays good prediction on boron adsorption behaviour with mean square error (MSE) and coefficient of determination (R²) at 0.000209 and 0.9985, respectively. The adsorption equilibrium is reached within 57 min with the maximum adsorption capacity at 15.23 ± 1.05 mg/g. The results indicate that the AO-modified poly(AN-co-AA) is a potential and effective adsorbent for boron removal from aqueous solution.

Text FK 2019 102 - ir.pdf Download (1MB)

Actions (login required)

View Item