Predictive and correlative modeling of breakthrough curves in fixed-bed adsorption of organic contaminants

The predictive and correlative performance of a mechanistic breakthrough model for fixed-bed adsorption of organic contaminants was systematically evaluated. The model assumes constant-pattern behavior, Langmuir equilibrium, and intraparticle mass transfer represented by the linear driving force (LDF) approximation. Published breakthrough data for paracetamol, Reactive Blue 5G, and phenol were analyzed using two parameterization strategies. In the predictive mode, equilibrium and kinetic parameters were specified a priori from independent batch experiments or engineering correlations. In the correlative mode, these parameters were estimated directly from breakthrough data. The predictive approach captured the overall shape of the breakthrough curves but exhibited limited quantitative accuracy. Adjustment of the adsorption capacity was required for phenol, and the LDF rate coefficient required modification in all cases. In contrast, the correlative approach consistently produced excellent agreement with experimental data across all systems. Mathematical examination further indicates that the model structure inherently accommodates moderately tailing breakthrough behavior.