Recently, the production of taste and odor (T&O) compounds is a common problem in water industry. Geosmin is one of the T&O components in drinking water. However, geosmin is hardly eliminated through the conventional water treatment systems. Among various advanced processes capable of removing geosmin, adsorption process using granular activated carbon (GAC) is the most commonly used process. As time passes, however GAC process changes into biological activated carbon (BAC) process. There is little information on the BAC process in the literature. In this study, we isolated and identified microorganisms existing within various BAC processes. The microbial concentrations of BAC processes examined were 3.5×105 colony forming units (CFU/g), 2.2×106 CFU/g and 7.0×105 CFU/g in the Seongnam plant, Goyang plant and Goryeong pilot plant, respectively. The dominant bacterial species were found to be Bradyrhizobium japonicum, Novosphingobium rosa and Afipia broomeae in each plants. Removal efficiencies of 3 μg/L geosmin by the dominant species were 36.1%, 36.5% and 34.3% in mineral salts medium(MSM) where geosmin was a sole carbon source.

The purpose of this study was to investigate the removal of ammonium nitrogen by biological nitrification in raw water containing LAS using BAC. At batch teats, LAS removal by ozone followed the first order reaction, and the rate constants(k) by ozone dose 1, 3mg/min.L were $0.040min^{-1}$, $0.062min^{-1}$ respectively. Therefore, the more ozone was dosed, the higher LAS was removed The reaction between ozone and ammonium nitrogen also followed the first order, and rate constants(k) at pH7,8 and 9 were $8.9{\times}10^{-4}min-1$, $3.8{\times}10^{-3}min^{-1}$, and $2.9{\times}10^{-2}min^{-1}$ respectively at ozone dose of 3mg/min.L . Therefore, ammonium nitrogen was little removed by ozone under neutral pH of 7. The continuous flow apparatus had four sets composed of a ozone contacter and a GAC column. Through continuous filtration test for 50days, the following conclusions were derived; (1) LAS was removed 23%, 30% respectively by ozone dose 1, 3mg/L, and was not detected in all column effluents during the period of experiment. Therefore, it appeared that adsorption capacities of each column still remained. (2) Ammonium nitrogen concentration after ozone contact varied little in raw Water because pH of raw water was from 6 to 7, and was transfered to nitrite and nitrate within GAC columns as the result of staged nitrification. After 30days, nitrite was not detected in all column effluents due to biological equilbrium between nitro semonas and nitrobacter Average removals of ammonium nitrogen in each column after the lapse of 30days were the following; ${\cdot}$ column A (ozone dose 3mg/L, EBCT 9.5min): about 100% ${\cdot}$ column B (ozone dose 1mg/L, EBCT 9.5min): 91% ${\cdot}$ column C (ozone dose 3mg/L, EBCT 14.2min): about 100% ${\cdot}$ column D (ozone dose 0mg/L, EBCT 9.5min): 53% Though column A and C reached nitrification of about 100%, column C (longer EBCT than column A) was more stable than column A. (3) After backwash, nitrification reached steady state within 5 to 8 hours. Therefore, nitrification was not greatly affected by backwash. (4) According to the nitrification capacity in depth of column A, C, where 100% nitrification occured. LAS was removed within 20cm, while ammonium nitrogen required more depth to be removed by nitrification.

In Korea, many drinking water treatment plants (DWTPs) have introduced and are going to introduce biological activated carbon (BAC) process to treated dissolved organic matter (DOM) in water which are difficult to control by conventional water treatment processes. Even though more decade have passed since introduced BAC in Korea, most of BAC operating method was followed to the modified sand filter operating manuals. In case of BAC backwashing, many DWTPs set the periods of backwashing about 3∼5 days. In this study, we have collected data to set the proper BAC backwashing periods from both pilot-plant and real DWTPs. We had measured heterotrophic plate count (HPC), turbidity, water temperature, dissolved organic carbon (DOC) and headloss from just after backwashing to the next backwashing time for two years. Considering water quality factors, the BAC run time from backwashing to the next backwashing could extend more 30 days without water quality deterioration if the head loss do not reach the limited level which depends on each BAC facilities’ condition. It means the BAC treated water could be saved in the proportion of extended the backwashing period to the existing backwashing period.

The concentration of organic compounds was analyzed at each step of BAC process though BDOCtotal/rapid/slow. Further, bacteria communities and biomass concentrations measured FISH and ATP methods were analyzed. The bed volume (BV) of steady state is different from that of based on assessment of organic compounds removal. Bed volumes in DOC, BDOCrapid and BDOCtotal/slow removal at steady state were around 27,500 (185.8 day), 15,000 (101.4 day) and 32,000 (216.2 day), respectively. A biomass didn't change after the bed volume reached 22,500 (152.0 day) according to analyzing ATP concentration of bacteria. The concentration of ATP was 2.14 μg/g in BV 22,500 (152.0 day). The total bacterial number was 4.01±0.4×107 cells/g at the bed volume 1,150 (7.8 day) (the initial operation) and the number of bacteria was 9.27±0.2×109 at the bed volume 58,560 395.7 day) that increased more than 200 times. Bacterial uptrend was reduced and bacterial communities were stabilized since BV 18,720 (126.5 day). When BV were 1,150 (7.8 day), 8,916 (60.2 day), 18,720 (126.5 day), 31,005 (209.5 day), 49,632 (335.3 day), 58,560 (395.7 day), a proportion of total bacteria for the Eubacteria were 60.1%, 66.0%, 78.4%, 82.0%, 81.3% respectively. γ-Proteobacteria group was the most population throughout the entire range. The correlation coefficient (r2) between Eubacteria biomass and ATP concentration was 0.9448.

The bacterial community structure in biological activated carbon (BAC) process in drinking water treatment plant was investigated by Fluorescent in situ Hybridization (FISH) with rRNA-targeted oligonucleotide probe. Samples were collected at different three points in BAC process every month for one year. They were hybridized with a probe specific for the alpha, beta, gamma subclass of the class Proteobacteria, Cytophaga- Flavobacteria group and Gram-positive high G+C content (HGC) group. Total numbers of bacteria in BAC process counted by 4',6-diamidino-2-phenylindole (DAPI) staining were 5.4×1010 (top), 4.0×1010 (middle) and 2.8×1010 cells/ml (bottom). The number of the culturable bacteria was from 1.0×107 to 3.6×107 cells/ml and the culturability was about 0.05%. The faction of bacteria detectable by FISH with the probe EUB338 was about 83% of DAPI counts. Gamma and alpha subclass of the class Proteobacteria were predominant in BAC process and their ratios were over 20% respectively. In top and middle, alpha, beta and gamma subclass of the class Proteobacteria competed with each other and their percentages was changed according to the season. In bottom, gamma subclass of the class Proteobacteria was predominant all through the year. It could be successfully observed the seasonal distribution of bacterial community in biological activated carbon process using FISH.

The objective of this study was to clarify the characteristics of the removed micropollutant since the breakthrough of adsorption ability was occurred in biological activated carbon(BAC) process. The removal efficiency of DOC (Dissolved Organic Carbon) was 36 % in the breakthrough of BAC occurred by NOM (Natural Organic Matter). The most of removal DOC was found out the adsorbable and biodegradable DOC (A&BDOC). But it was not clear to remove by any mechanism because A&BDOC have simultaneously the adsorption of activated carbon and biodegradation by microorganism in BAC. The removal of bromophenol was examined with BAC and rapid sand filter, for investigation of DOC removal mechanism in the breakthrough of BAC. In this experiment, BAC filter has been operated for 20 months for the treatment of reservoir water. The BAC filter was already exhausted by NOM. Bromophenol, adsorbable and refractory matter, was completely removed by BAC filter. Therefore, it might be removed by the adsorption in BAC. Adsorption isotherms of bromophenol were compared to two BACs which was preloaded with 500 daltons and 3,000 daltons of NOM. BAC preloaded with 3,000 daltons of NOM was not decreased to the adsorbability of bromophenol but BAC preloaded with 500 daltons of NOM was greatly decreased to it. These result indicated that NOM of low molecular weight can be removed by adsorption after a long period of operation and the breakthrough by NOM in BAC. Therefore, micropollutants might be removed through adsorption by saturated BAC.

Improvement of water quality and Investigation of bacterial characteristics have been conducted in a pilot plant using biological activated carbon (BAC) in water treatment process at the downstream of the Nakdong River. Most of water control parameters were highly improved after passing through BAC. Approximately 54% of dissolved organic carbon was removed in coal-based BAC process. Bacterial biomass and bacterial production appeared 9.8×108 CFU/g and 7.l㎎-C/㎥·hr in coal-based BAC, respectively. Predominant bacteria species grown in BAC were identified as Pseudomonas, Flavobacterium, Alcaligenes, Acinetobacter and Aeromonas species. Particularly Pseudomonas vesicularis was dominant in both coal-based and coconut-based BACs, while Pseudomonas cepacia was dominant in wood-based BAC.

The advanced oxidation process (AOP) using ozone combined with hydrogen peroxide and ultraviolet treatment were evaluated for biodegradable dissolved organic carbon (BDOC) formation and dissolved organic carbon (DOC) removal. Oxidation treatment were conducted alone or combination with ozone, hydrogen peroxide and ultraviolet processes. Ozone dosage of ozone process was varied from 0.5㎎/ℓ·min to 5㎎/ℓ·min. Ozone/hydrogen peroxide process was done using 20㎎/ℓ of hydrogen peroxide concentration. Ozone/ultraviolet process was irradiated with 12mW/㎠ of density and 254nm. Ozone dosage was varied from 0.5㎎/ℓ·min to 5㎎/ℓ·min at the ozone/hydrogen peroxide and ozone/ultraviolet processes too. Contact time of all the process was 20 minutes. Oxidation treatment were performed on microfiltration effluent samples. BDOC formation was reached to an optimum at ozone dosage of 1.5㎎/ℓ·min in the ozone/hydrogen peroxide process and 1㎎/ℓ·min in ozone/ultraviolet process, after which BDOC formation was decreased at higher ozone dosages. But BDOC formation was increased with ozone dosages increasing in ozone process. The efficiency of DOC removal was higher AOPs than ozone process. Ozone/ultraviolet proces was the highest for DOC removal efficiency in each process. THMFP removal efficiency by ozone/ultraviolet process was higher than that by each of ozone process and ozone/hydrogen peroxide process.

In order to research the adsorption removal characteristics of trace organic by-products in disinfection of drinking water by biological activated carbon(BAC), water samples disinfected with Cl_2, O_3 and ClO_2 after treatment by fluidized-bed system with water added with humic acid(l0㎎/L) were investigated the formation and the removal of trihalomethanes (THMs), and the trace organic by-products by gas chromatography(GC) & gas chromatography/mass selective detector(GC/MSD). Control was used by activated carbon(AC) and water added with hurnic acid(HA). The results were summarized as follow ; The THMs removal effect of BAC by chlorination was in lower 90 % than that of control(HA), the sorts of oxidants formed by Cl_2, O3 and ClO_2 were that O_3 was very fewer than Cl_2 or ClO_2 and that ClO_2 was fewer than Cl_2 The trace organic by-products were esters and phthalates etc. Based on results above, it is concluded that BAC was appeared the more desirable adsorption-degradation removal characteristics than that of AC.