Skoči na glavni sadržaj

Izvorni znanstveni članak

https://doi.org/10.15255/KUI.2023.004

Removal of Chlortetracycline Chlorhydrate by Photo-Fenton Process: Experimental Study and ANN Modelling

Nabila Boucherit ; Biomaterials and Transport Phenomena Laboratory (LBMPT), Yahia Fares University, Médéa, Alžir
Salah Hanini ; Biomaterials and Transport Phenomena Laboratory (LBMPT), Yahia Fares University, Médéa, Alžir
Abdellah Ibrir orcid id orcid.org/0000-0003-0332-1398 ; Materials and Environment Laboratory (LME), Faculty of Technology, Yahia Fares University, Médéa, Alžir
Maamar Laidi ; Biomaterials and Transport Phenomena Laboratory (LBMPT), Yahia Fares University, Médéa, Alžir *
Mohamed Roubehie-Fissa orcid id orcid.org/0000-0002-9154-6409 ; Biomaterials and Transport Phenomena Laboratory (LBMPT), Yahia Fares University, Médéa, Alžir

* Dopisni autor.


Puni tekst: engleski pdf 1.021 Kb

str. 627-637

preuzimanja: 130

citiraj


Sažetak

The present work aimed to study the feasibility of photo-Fenton oxidation for the degradation of chlortetracycline chlorhydrate (CTC) in aqueous solutions, as well as the modelling of system behaviour by artificial neural networks. The removal performance of CTC oxidation by the Photo-Fenton process was studied under solar radiation. Different parameters were studied, such as pH (3 to 5), and initial concentrations of CTC (0.1 to 10 mg l–1), hydrogen peroxide (1.701 to 190.478 mg l–1), and ferrous ions (2.8 to 103.6 mg l–1). Results showed that a high removal efficiency of 92 % was achieved at pH 3 under optimised conditions, such as 10 mg l–1 of CTC, 127.552 mg l–1 of H2O2, and 36.4 mg l–1 of Fe2+. The transformation of CTC molecules was proved by UV-visible and HPLC analyses, which showed that almost no CTC molecules were remaining in the treated solution. A multi-layer perceptron artificial neural network has been developed to predict the experimental removal efficiency of CTC based on four dimensionless inputs: molecular weight, and initial concentrations of CTC, hydrogen peroxide and ferrous ions. The best network has been found with a high determination coefficient of 0.9960, and with a very acceptable root mean square error 0.0108. In addition, the global sensitivity analysis confirms that the most influential parameter for the CTC removal by photo-Fenton oxidation is the initial concentration of ferrous cations with a relative importance of 33 %.

Ključne riječi

artificial neural networks; multi-layer perceptron; chlortetracycline chlorhydrate; modelling; photo-oxidation

Hrčak ID:

309806

URI

https://hrcak.srce.hr/309806

Datum izdavanja:

11.11.2023.

Podaci na drugim jezicima: hrvatski

Posjeta: 592 *