We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Projecting the sorption capacity of heavy metal ions onto microplastics in global aquatic environments using artificial neural networks
Summary
Machine learning models accurately predicted how much heavy metals like cadmium, lead, and zinc would adsorb onto microplastics in rivers, lakes, and oceans, based on factors like metal concentration and water salinity. Aged microplastics showed higher metal sorption capacity than virgin plastics, and predicted values matched real-world field measurements.
Microplastics pollution and their interaction with heavy metal ions have gained global concern. It is essential to develop models to predict the sorption capacity of heavy metal ions onto microplastics in global aquatic environments, and to connect the laboratory study results with the field measurement results. In this paper, the artificial neural networks (ANN) models were established based on literature data. for The results showed that the ANN model could predict the sorption capacity of heavy metal ions (including Cd, Pb, Cr, Cu, and Zn) onto microplastics in the global environments with high correlation coefficient (R) values (0.926∼0.994). The predicted sorption capacity was influenced by the initial concentration of heavy metal ions and the salinity in surrounding water. The predicted sorption capacity in rivers and lakes was higher than that in the ocean. Aged microplastics had higher affinity to heavy metal ions than virgin microplastics. The predicted sorption capacity of Cd, Pb, and Zn ions onto large microplastics (5 mm) was less than 0.12 μg/g. The predicted amount was in agreement with the field measurement results, suggesting that the laboratory studies can provide useful information for projecting the sorption capacity of heavy metal ions onto microplastics in global aquatic environments.