Comparative analysis of fixed-bed sorption models using phosphate breakthrough curves in slag filter media

Abstract

Fixed-bed kinetic sorption (Bohart-Adams, Thomas, Yoon-Nelson, Clark, Wolborska, and modified dose-response) models are commonly used to simulate breakthrough curves (BTCs) from fixed-bed systems. However, more caution should be taken in using these models. Some researchers misused the equation, which is a totally different type from the original model, as a simplified model. Others used the same equation expressed in different forms as an independent model. The aim of this study was to clarify the fixed-bed sorption models via comparative analysis using the phosphate BTCs in slag filter media. For the analysis, the breakthrough data for phosphate (initial phosphate concentration = 1.0 and 2.0 mg/L) sorption in fixed-bed columns (inner diameter = 2.5 cm and column length = 10, 20, and 30 cm) were obtained from the experiments. The original Bohart-Adams model was simplified in the literature to the convergent- and divergent-type models in order to be used for the BTC analysis. However, the divergent-type model, which is equivalent to the Wolborska model, should not be the type of Bohart-Adams model used, because it behaves totally different from the original model. Also, the Thomas and Yoon-Nelson models should not be used simultaneously with the Bohart-Adams model, because they are equivalent to the simplified convergent-type Bohart-Adams model, and the parameters of both of the models (k(T), q(0), k(YN), and tau) can easily be calculated from the Bohart-Adams model parameters (k(BA) and N-0). The Bohart-Adams, Clark, and modified dose-response models could describe the BTCs relatively well with a high determination coefficient and a low chi-square coefficient. From this study, the Bohart-Adams, Clark, and modified dose-response models are recommended for the BTC analysis, because these models can provide useful design parameters (k(BA), N-0, Z(0), t(b), and q(0)) for the fixed-bed systems.