| Abstrakt: |
Nowadays, the recycling of metals from electrical and electronic waste is of great relevance due to its direct and indirect impact on environmental, social, and economic fields. Therefore, this study, conducted at the laboratory level, focuses on the recovery of copper from printed circuit boards through dynamic acid leaching in an H2SO4-O2 system, with the stirring rate controlled as the main parameter. Initially, the metallic pins were characterized by SEM-EDS, revealing that they consist of 7.56 wt% of copper, the predominant element serving as the base material. A thin gold film (79 wt%) is deposited on the copper to enhance its electrical conduction properties. In the subsequent leaching step, a random sample of 10 g was taken in a 500 mL volume, with an acid concentration of 0.03 M. The system was heated to 298.15 K under an oxygen partial pressure of 101.3 kPa. The stirring rate was varied from 450 to 1000 rpm, resulting in a maximum copper concentration of 645.294 ppm in the solution. The experimental constants were calculated for low (0–60 min) and high (60–240 min) chemical attack times, yielding ranges of 0.026 to 0.923 and 0.019 to 2.577 min− 1, respectively. On the other hand, one of the main outcomes of this research lies in the implementation of an artificial neural network to intelligently model the experimental process. It exhibited a mean squared error, correlation coefficient, and determination coefficient of 0.99690. Artificial neural networks emerge as an exceptional tool in predicting hydrometallurgical processes. This innovative application not only optimizes copper recovery but also ensures a cost-effective and environmentally friendly management of electronic waste. In the same way, it is possible to generate models of problems through learning. For all the aforementioned reasons, in the present work, an artificial neural network is developed to predict the dissolution of Cu in an electronic waste leaching process, considering the stirring rate as a key factor. Highlights: • Electronic waste as secondary source of precious and non-precious metals. • Copper is obtained from recycled computer printed circuit boards through an environmentally friendly chemical process. • Artificial neural networks enable design and fault finding in complex systems by validating, aggregating and analyzing data. • Application of an artificial neural network for processing, identification and modeling of dynamic acid leaching systems. • A novel combined hydrometallurgical and artificial neural network process for the recovery of copper from electronic waste. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink