Application of Artificial Neural Network and Shrinking Core Model for Copper (Ii) and Lead (Ii) Leaching from Contaminated Soil Using Ethylenediaminetetraacetic Acid

ABSTRACT

This work examined the kinetics of Copper (II) and Lead (II) leaching in EDTA acid. The impacts of various factors such as pH, acid concentration, soil-to-solution ratio, and stirring speed were investigated to improve the leaching conditions and assess the leaching kinetics. According to the findings, copper (II) and lead (II) leaching was enhanced by an increase in leaching temperature, stirring speed and acid concentration, as well as a drop in solid-liquid ratio and pH. The results indicated that after 120 minutes of leaching in EDTA acid with a concentration of 1.5 M and a solid/liquid ratio of 5 g/200 mL, pH of 3, and stirring speed of 375 rpm, 90.98% of copper (II) and 95.43% of lead (II) were removed from the sample. Diffusion across the product layer was the rate-controlling phase during the leaching; according to the experimental results, a diffusion-controlled model best matched. The activation energy was determined to be 15.58 kJ/mol and 15.70 kJ/mol for copper (II) and lead (II), indicating the diffusion-controlled process. The ANN Multi-layer, Feed-forward, and Back-Propagation Learning algorithms are trained to optimize the leaching process parameters. The ANN algorithm was developed with two neurons as output layers corresponding to copper (II) and lead (II) leaching recovery, 15 hidden layers, and 5 input variables describing the leaching parameters. The optimal trained neural network illustrates the validation, testing, and training steps with R² values of 0.992, 0.994 and 0.999, respectively. The leaching mechanism was validated by (XRF) X-Ray Fluorescence, (FTIR) infrared spectroscopy and (SEM) scanning electron microscope analysis of the original sample.

KEYWORDS:

• Artificial neural network
• shrinking core
• leaching, copper (II)
• lead (II)
• EDTA

Abbreviation

ANN Artificial neural network

EDTA ethylenediaminetetraacetic acid

RMSE Root means square errors

ARE Average relative errors

MSE Mean square errors

BP Back-propagation

LM Levenberg-Marquardt model

ML Multilayer perceptron

SCM Shrinking core model

EGTA Ethylene glycol bistetraacetic acid

HEDTA Nyhydroxyethylenediamine triacetic acid

(XRF) X-Ray Fluorescence

(FTIR) infrared spectroscopy

(SEM) scanning electron microscope

Data statement

The data supporting this study’s findings are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Author Statement

Musamba Banza: conceptualization, methodology, formal analysis, investigation, data curation, writing, writing- original draft.

Hilary Rutto: validation, formal analysis, investigation, writing-reviewing, and editing.

Tumisang Seodigeng: validation, formal analysis, investigation, and editing.

Disclosure statement

The authors declare that they have no known competition for financial interests or personal relationships that could have influenced the work reported in this paper.

Additional information

Funding

The authors have no funding to report.