Machine Learning Application, Isotherm Modelling and Adsorption Kinetics for the Removal of Chromium (VI) using Carissa Carandas

Abstract

The primary goal of this study is to investigate and predict the capacity of low cost bio-sorbent Carissa carandas in removing chromium from industrial waste water using analytical techniques and machine learning techniques such as Radial Basis Function Neural Networks. The adsorption technique has been widely used to separate metal ions in bulk for many years, and it is well known to researchers. The adsorption capacity of Carissa carandas for the removal of Cr(VI) ions from local industrial wastewater was investigated in this paper as an application to real-time wastewater treatment using various analytical instruments such as Atomic Adsorption Spectroscopy (AAS), Scanning Electron Microscope (SEM), Fourier Transform Infrared Spectroscopy (FTIR) and X-ray diffraction (XRD). In this paper, we proposed machine learning based radial basis function neural network, which is used to model the obtained experimental results to estimate the percentage removal of Cr(VI) for different unknown metal ion concentrations.The adsorption of Cr(VI) from local industrial waste water by Carissa carandas prepared by chemical method, it was treated with the standard test procedures, analyzed using Langmuir, Freundlich isotherms and Pseudo-first and second order kinetic models. Adsorbent dose, initial adsorbate concentration, pH, and contact time all had an impact on the percentage of adsorption. According to the research's findings, Carissa carandas was able to adsorb hexavalent chromium to a maximum of 96% at 5 ppm, 120 minutes, 1 g/L of adsorbent dose, pH = 5, and 95.81% using ANN for the same parameters. The Carissa carandas are widely distributed and freely accessible in many locations, thus the results of this research will be economically and effectively beneficial for the health of the ocean and land environments.