Environmental Advances (Apr 2024)
Optimal Adsorption of a Binary Dye Mixture of Basic Yellow 2 and Rhodamine B using Mixture-Process Variable Design, Ridge Analysis and Multi-Objective Optimization
Abstract
The adsorption of a mixture of Basic Yellow 2 and Rhodamine B dyes was studied using the MIL-101(Cr)-SO3H metal organic framework adsorbent. The adsorption experiments were implemented using mixture-process variable design. The statistically significant factors affecting total adsorption capacity (qTotal) of MILS for the two dyes and their combined percentage removal (PRTotal) were identified. It was concluded that in addition to main factors, the binary interactions between pH, temperature and dosage also played a major role in influencing the adsorption performance. Different mechanisms such as π-π interactions, steric hindrance, electrostatic interactions, and hydrogen bonding were proposed to be responsible for the adsorption. The total adsorption capacities at the global optimal conditions for 1:3, 1:1 and 3:1 Basic Yellow 2 – Rhodamine B dye mixtures were found to be 209.8, 218.5 and 216.2 mg/g respectively. The classical analytical ridge analysis approach led to optimal solutions, some of which were outside the experimental design space. This limitation was overcome in the numerical ridge analysis using composition constraints. The improved ridge analysis approach identified the locus of local optimum total adsorbent loadings and percentage removals along the steepest ascent path enroute to the global optimum solution. These were identified within and on the boundaries of the feasible experimental design space. Using the ridge analysis path, conditions less severe in pH, temperature, and adsorbent dosage than those corresponding to the global optimum conditions could be identified without significant loss in objective. This approach has high potential for application in wastewater treatment processes to identify better as well as suitable operating conditions than the current settings when there are changes to feed compositions of the pollutants, and/or process conditions (e.g., pH, and temperature). Conditions of pH, temperature and adsorbent dosage that maximized qTotal could be different from those that maximized PRTotal and hence a compromise solution had to be found. Multi-objective optimization using an in-house developed elliptical method could reliably generate the Pareto fronts and identify suitable compromise solutions for different proportions of dyes in the mixture.