Objectives:In this study, we used liquid chromatograph-organic carbon detector-organic nitrogen detector (LC-OCD-OND) to evaluate adsorption and breakthrough characteristics of NOM fractions (biopolymers (BP), humic substances (HS), building blocks (BB) and low molecular weight organic substances (LMW-O)) according to the various characteristics of the different materials of granular activated carbons (GACs).Methods:Breakthrough characteristics, adsorption capacity and partition coefficients were evaluated by NOM fractions (BP, HS, BB, and LMW-O) using a lab-scale GAC adsorption column filled with coal-, coconut- and wood-based GAC. The GAC column test was operated with 10 minutes empty bed contact time (EBCT). The pore characteristics of each GAC were evaluated using an automated gas sorption analyzer (Autosorb iQ3, Quantachrome, USA) and the concentrations of NOM fractions in the influent and effluent were analyzed using chromatography LC-OCD-OND (Model 8, DOC-Labor, Germany).Results and Discussion:NOM adsorption capacity was evaluated for different materials of laboratory scale GAC adsorption column test. To study the adsorption behavior of individual NOM fractions according to the operation time, NOM was fractionated into BP, HS, BB and LMW-O by LC-OCD-OND, and the individual NOM fractions were quantified. Higher MW like BP was not adsorbed to GAC, in contrast, HS, BB, and LMW-O were well removed during the initial operation period, the concentrations in the effluent gradually increased as increase the operation period until reaching to the pseudo steady-state. Poor removal of BP in GAC adsorption may be a result of blocking the pores with large MW BP and hinder the access to the pores. However, in the case of HS, BB, and LMW-O, as the molecular size decreased, these organic matters easily access to the pores inside of GAC. It was confirmed through the partition coefficient that the adsorption capacity of these NOM fractions increased in proportion to the MW. In addition, in order to achieve a high NOM removal efficiency in the GAC adsorption process, not only the specific surface area, pore volume, and pore width of the GAC must be large, but also the pH<sub>zpc</sub> must be higher than the neutral pH level.Conclusions:In order to achieve a high NOM removal efficiency in the GAC adsorption process, not only the specific surface area, pore volume, and pore width of the GAC must be large, but also the pH<sub>zpc</sub> must be higher than the neutral pH level. In addition, in the NOM fractions, BP were not adsorbed to GAC, while the adsorption capacity of the remaining NOM fractions increased as the MW of the NOM fractions decreased. LMW-O was the most adsorbed, followed by BB, HS and BP. BP and HS play an important role in the membrane fouling that are introduced a lot into domestic and foreign water treatment plants. This study showed that the BP was not removed by the adsorption mechanism of the GAC process. In addition, HS was adsorbed and removed at the beginning of the operation, but the adsorption capacity of HS decreased rapidly as the operation period increased compared to other NOM fractions. Therefore, the GAC adsorption process is not expected to be an effective pre-treatment technology for reducing membrane foulants. Previous studies showed that the yields of DBPs (µmol・DBP/µmol・C) in the high MW humic and low MW non-humic fractions are similar. Therefore, it is suggested that the GAC adsorption process is more effective for DBP precursor control in water containing a larger percentage of LMW NOM.
Read full abstract