Linear and Nonlinear Regression Analysis for the Adsorption of Remazol Dye by Romanian Brewery Waste By-Product, Saccharomyces cerevisiae

Earth's water balance and economy are becoming increasingly fragile due to overpopulation, global warming, severe environmental pollution and both surface and groundwater pollution. Therefore, it is essential to find solutions to the problems of water scarcity and water pollution. In this research, an experiment was designed to optimize the technique for the adsorption of Remazol Red F3B (RR) dye by lyophilized brewery yeast waste from the fermentation process. Moreover, we proved that brewery yeast is a great adsorbent. Batch adsorption experiments were carried out for optimization of different initial parameters, such as initial dye concentration (5-1000 mg/L), amount of yeast (0.5-2.5 g), pH (3-11) and temperature (20 to 40 degrees C). Furthermore, the structure and elemental composition of the adsorbent were analyzed with SEM, EDS and FTIR before and after biosorption. The best fits for the mathematical isotherm models in the case of the linear form were the Langmuir I and Freundlich models (R-2 = 0.923 and R-2 = 0.921) and, for the nonlinear form, the Khan model (R-2 = 0.9996) was the best fit. The pseudo-second-order kinetic model showed the best fit for both linear (plotting t/q(t) vs. t) and nonlinear forms, are the calculated q(e) values were similar to the experimental data.

Automated summary

The water balance and economy of Earth are becoming increasingly fragile due to overpopulation, global warming, severe environmental pollution, and water pollution. Finding solutions to water scarcity and water pollution is essential. In this research, an experiment was designed to optimize the adsorption technique of Remazol Red F3B dye using lyophilized brewery yeast waste. It was proved that brewery yeast is a great adsorbent. Different initial parameters were optimized, and the adsorbent's structure and elemental composition were analyzed. The mathematical isotherm models, such as Langmuir I, Freundlich, and Khan models, showed good fits. The pseudo-second-order kinetic model also provided a good fit.