Process modeling of sonocatalytic degradation of caffeine using CeO2 via black box method / Siti Nurfarahin Nordin and Nur Fadzeelah Abu Kassim
In order to meet human demands, the pharmaceutical industries are increasing over the years. Caffeine (C8H10N4O2), representative as one of the pharmaceuticals and personal care products (PPCPs) was considered to be contaminating to humans and other aquatic life which has exerted water pollutions cr...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2020
|
| Subjects: | |
| Online Access: | https://ir.uitm.edu.my/id/eprint/81583/1/81583.pdf https://ir.uitm.edu.my/id/eprint/81583/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In order to meet human demands, the pharmaceutical industries are increasing over the years. Caffeine (C8H10N4O2), representative as one of the pharmaceuticals and personal care products (PPCPs) was considered to be contaminating to humans and other aquatic life which has exerted water pollutions crisis. In this study, the mathematical modeling of sonocatalytic degradation of caffeine using CeO2 was developed via artificial neural networks. The artificial neural network (ANN) was employed for developing the suitable modeling of the CeO2 catalyst in determining the efficiency of sonocatalytic degradation of caffeine using CeO2 (%). The parametric conditions of this study involved initial pH of caffeine, initial concentration of caffeine (g/L), and dosage of CeO2 (g/L). Thus, a three-layered feed-forward back propagation neural network with 12 neurons in the hidden layer was built to give the optimal results on the efficiency of sonocatalytic degradation of caffeine using CeO2. ANN predicted high accuracy in which R2, MSE, and MAE values were 0.996, 0.3109, and 0.07885 respectively. It was also revealed that the ANN model was provided excellent predictive performance by giving the highest value of R2. |
|---|