Forecasting The Adsorption Capacity Of Organic Dye By Using Zirconium-Based Metal-Organic Framework (MOF): Comparison Studies Between Response Surface And Neural Network Models
Loading...
Date
2021-07-01
Authors
Poopathi, Veshmen
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
The factors affecting the adsorption capacity of Zirconium Metal Organic
Framework were analyzed which includes the pH, contact time, amount of adsorbent
and initial dye concentration. The experiment was run based on central composite
design (CCD) in response surface methodology (RSM). The experimental results were
used to investigate the effect of input factors on the adsorption capacity of Zirconium
MOF and to develop a model to predict system performance. According to the
response surface plot, higher adsorption capacity of Zirconium MOF can be achieved
with less adsorbent and a higher dye concentration. RSM was used to create a
mathematical model, and the model's performance was evaluated using analysis of
variance (ANOVA). Another neural network model was created using MATLAB’s
neural network toolbox and Mathematica's net operation and predictor function. The
adsorption capacity of Zirconium MOF was predicted using a mathematical and neural
network model. Due to a shortage of experimental data for neural network training,
the mathematical model generated in RSM had a higher accuracy in predicting the
output response, with an R2 of 0.97 and an RMSE of 2.87. RSM performed numerical
optimization for the adsorption capacity of Zirconium MOF to determine the best
operating conditions. The maximum adsorption capacity of Zirconium MOF (46.75
mg/g) was found to be at pH 7, contact time of 70 min, adsorbent amount of 10 mg,
and initial dye concentration of 44.99 mg.