Adsorption of para chloro meta xylenol (PCMX) in composite adsorbent beds: Parameter estimation using nonlinear least square technique

Phenols and phenolics are considered to be potentially hazardous water pollutants. Para chloro meta xylenol (PCMX) is the major component of several disinfectants, produced commercially by internationally well-known pharmaceutical companies. Adsorption of PCMX in various combinations of adsorbents, both natural and artificial, has been studied. Specially, water hyacinth (Eichhornia crassipes [Mart] Solms-Laubach) was used as an adsorbent in combination with powdered activated carbon (PAC) and granular activated carbon (GAC). Prediction of PCMX adsorption capacities is important in design considerations. Various equilibrium adsorption isotherms have been studied using experimental data. Equilibrium was achieved after approximately 24 h. Five isotherm models, namely, Tempkin isotherm, Freundlich isotherm. Langmuir isotherm, Redlich-Peterson isotherm and Toth isotherm were studied and parameterized. Respective parameters for each of the models were estimated by the nonlinear Levenburg-Marquardt parameter estimation method, based on the ((2) minimization. Model performance was ascertained using uncertainty analysis on both the parameters, k(1) and k(2), for each of the five models. The models were simultaneously ranked. The Tempkin isotherm was ranked I in case of all the 3 samples with extremely low uncertainties on both parameters k(1) and k(2). Depending on the nature of the sample and its range of concentration levels, isotherm model by Langmuir scored 2 in the ranking scale on 2 occasions out of 3 with low uncertainties on both k(1) and k(2). (C) 2009 Elsevier B.V. All rights reserved.