Removal of Antidiabetic Pharmaceutical (Metformin) Using Graphene Oxide Microcrystalline Cellulose (GOMCC): Insights to Process Optimization, Equilibrium, Kinetics, And Machine Learning

Abstract

Cellulose is a basic material for the manufacture of filters and adsorbent materials, as well as for the removal of contaminants, especially in food, pharma and other industries. The goal of this research is to see how a novel arrangement of biomass resources (hybrid biomass) can be used to improve the characteristics and behavior of microcrystalline cellulose (MCC) produced by the traditional method of acid hydrolysis. Chemical oxidation cum exfoliation was used to synthesize graphene oxide (GO) nanoparticles, which were then composited with microcrystalline cellulose (GOMCC) and used for metformin (MFM) adsorption from an aqueous medium. The GOMCC characterized using particle size distribution, X-ray analysis (XRD), Raman spectra analysis, Fourier transform infrared spectroscopy (FTIR), fieldemission scanning electron microscopy (FESEM) analysis, and transmission electron microscopy (TEM). The GOMCC that had been prepared was mesoporous and had a reasonable surface area (196 m 2 /g), with a pore volume of 0.134 cm 3 /g and pore width of 17.85 nm. Batch-mode adsorption was performed at various temperatures (288, 303, and 318 K), GOMCC amounts (50, 100, and 150 mg), MFM concentrations (30, 50, and 70 mg/L), and pH values (4.5, 6.5, and 8.5) and were investigated with the Box-Behnken statistical design. The machine learning, models�Support Vector Machine (SVM), Gaussian Process Regression (GPR), Regression Trees (TREE), and Ensemble of Regression Trees (Ensemble)�are used to predict the adsorptive removal of MFM onto GOMCC from the BBD design input. Different isotherm equations and various kinetic models were used to assess the sorption data and ideal values were determined using the sum of normalized errors methodology. MFM maximal sorption capacity was investigated to be 132.10 mg/g. The kinetics revealed that the pseudo-first-order model fit the data exactly. MFM sorption was found to be spontaneous (ΔG°) and exothermic (ΔH°) in a thermodynamic analysis. The chemisorption of MFM onto GOMCC was followed by subsequent pore diffusion, which was exothermic and spontaneous. The main mechanisms responsible for the removal of MFM were found to be π−π interactions, alone interactions, sulfur interactions, hydrophobic and hydrogen bonding.