Korean J. Chem. Eng.

, 20(6), 1054-1065 (2003)

REVIEW

Biofilter in Water and Wastewater Treatment

Durgananda Singh Chaudhary, Saravanamuthu Vigneswaran, Huu-Hao Ngo, Wang Geun Shim* and Hee Moon*
Faculty of Engineering, University of Technology, Sydney (UTS), PO Box 123, Broadway, NSW 2007, Australia *Faculty of Applied Chemistry, Chonnam National University, Kwangju 500-757, Korea
Abstract Biofilter is one of the most important separation processes that can be employed to remove organic pollutants from air, water, and wastewater. Even though, it has been used over a century, it is still difficult to explain theoretically all the biological processes occurring in a biofilter. In this paper, the fundamental of biological processes involved in the biofilter is critically reviewed together with the mathematical modeling approach. The important operating and design parameters are discussed in detail with the typical values used for different applications. The most important parameter which governs this process is the biomass attached to the medium. The relative merits of different methods adopted in the measurement of the biomass are discussed. The laboratory-and full-scale applications of the biofilter in water and wastewater treatment are also presented. Their performances in terms of specific pollutant removal are highlighted. Key words: Biofilter, Biomass, Organics, Water, Wastewater

INTRODUCTION Filtration is one of the most important treatment processes used in water and wastewater treatment. In water treatment, it is used to purify the surface water for potable use whereas in wastewater treatment, the main purpose of filtration is to produce effluent of high quality so that it can be reused for various purposes. Any type of filter with attached biomass on the filter-media can be defined as a biofilter. It can be the trickling filter in the wastewater treatment plant, or horizontal rock filter in a polluted stream, or granular activated carbon (GAC) or sand filter in water treatment plant. Biofilter has been successfully used for air, water, and wastewater treatment. It was first introduced in England in 1893 as a trickling filter in wastewater treatment [Metcalf and Eddy, 1991], and since then, it has been successfully used for the treatment of domestic and industrial wastewater. Originally, biofilter was developed using rock or slag as filter media, however at present, several types and shapes of plastic media are also used. There are a number of small package treatment plants with different brand names currently available in the market in which different shaped plastic materials are packed as filter media and are mainly used for treating small amount of wastewater (e.g. from household or hotel). Irrespective of its different names usually given based on operational mode, the basic principle in a biofilter is the same: biodegradations of pollutants by the micro-organisms attached onto the filter media. Use of a biofilter in drinking water treatment (especially with granular activated carbon as filter media) was felt necessary only after the discovery of the re-growth of micro-organisms in water distribution pipe lines few decades ago. It has been observed that the inner surface of water distribution pipelines carrying potable water is coated with layers of biomass in few years of service period [Van der Kooij et al., 1982; LeChevallier and Lowry, 1990; Bou To whom correspondence should be addressed. E-mail: [email protected]

wer and Crowe, 1988]. The biodegradable organic matter (BOM), NH4+, Fe2+, Mn2+, NO2 , dissolved H2 and several other reduced species of sulfur are the most pertinent components that can cause bacterial regrowth on the water distribution pipelines [Rittmann and Huck, 1989]. Due to the regrowth of the microbial mass in the pipelines, the drinking water is considered biologically not stable. Even though there is no direct evidence of its instant health and hazardous side effects, use of such drinking water in long run cannot be assured to be safe. Besides the by-products of chlorine disinfection, disinfections by-products (DBPs) are often carcinogenic and harmful. The biological treatment especially by granular activated carbon (GAC) biofilter has been found effective in removing organic substances that can cause the microbial growth in the pipe lines, and is normally recommended to be included in the water treatment processes after ozonation [Bouwer and Crowe, 1988; Hozalski et al., 1995; Ahmad and Amirtharajah, 1998; Carlson and Amy, 1998]. Bacterial masses attached onto the filter media as biofilm oxidize most of the organics and use it as an energy supply and carbon source. Removal of the organic matters not only impairs microbial regrowth but also reduces taste and odor, the amount of organic precursor (available to form disinfection by-products, corrosion potential) and other micropollutants of health and aesthetic concern. Because of its wide range of application, many studies have been done on biofiltration system in last few decades (Table 1). However, theoretically it is still difficult to explain the behavior of a biofilter. The growth of different types of microorganisms in different working conditions makes it impossible to generalize the microbial activities in a biofilter. The biofilters operated at different filtration rates and influent characteristics can have diverse efficiency for different target pollutants. Besides, due to some of the operational drawbacks of the biofilter such as performance fluctuation, maintenance of biomass, and disinfection adequacy of the biofilter effluent, research on biofiltration process has become imperative. This paper mainly focuses on the theoretical and modeling aspects, and the performance of the biofilter in removing organics,
1054

Biofilter in Water and Wastewater Treatment

1055

Table 1. Summary of the past studies on biofiltration system with water and wastewater Researcher, source Ahmad et al. [1998], water Boon et al. [1997], wastewater Carlson and Amy [1998], water Hozalski and Bouwer [1998], synthetic water, NOM Yang et al. [2001], aquaculture water Niquette et al. [1998], water Filter medium Anthracite+ sand Granite, blastfurnace slag Anthracite Experimental parameter Organic AOC-P17, AOC-NOX, NPOC, turbidity BOD, ammonia, SS Biomass HPC None Major observation Backwashing technique and hydraulic transient can affect the performance of a biofilter. Performance a biofilter depends on organic load ing rate, temperature, and filter design configuration. DOC removal is controlled by biomass. The filter acclimatized at higher HLR had a substantially higher cumulative biomass. Biomass accumulation is not impaired by backwash with water Characteristics of filter media are more critical than the flow scheme to the biofilter in affecting the performance of the biofilter. Shut down of biofilter promotes anaerobic conditions reducing the quality of the effluent. The biofilter should be backwashed when anaerobic condition occurs. Removal efficiency of a biofilter depends on EBCT, not on filtration rate GAC contained 3-8 times more biomass than anthracite or sand

DOC, BDOC

Phospholipid analysis HPC bacterial count None

Glass beads+ sand Plastic media3 different shapes GAC

TOC

BOD5, SS, NH3-N, 3 NO3-N, NO2-N, PO4 DOC, DO, NH3, NO2

Bacterial count

Servais et al. [1994], water Wang et al. [1995a, b]

GAC Anthracite+ sand, GAC+ sand, sand

DOC, BDOC, NBDOC TOC, BDOC, aldehydes, AOC-NOX, THM and TOX formation potential

14

C-Glucose respiration Phospholipid analysis

AOC=Assimilable organic carbon, BOD5 =Biochemical oxygen demand, SS=suspended solid, DBP=Disinfection by-product, DOC=Dissolved organic carbon, DO=Dissolved oxygen, BDOC=Biodegradable dissolved organic carbon, NBDOC=Non-biodegradable dissolved organic carbon, THMFP=Trihalomethane formation potential, TOXFP=Total organic halide formation potential, HPC=Heterotrophic plate count, NPOC=Non-purgeable organic carbon, HLR=hydraulic loading rate, NOX=Nitrogen oxides.

nutrients, and some specific pollutants from water and wastewater. THEORETICAL BACKGROUND 1. Fundamentals of Biological Process In a biofiltration system, the pollutants are removed due to biological degradation rather than physical straining as is the case in normal filter. With the progression of filtration process, microorganisms (aerobic, anaerobic, and facultative bacteria; fungi; algae; and protozoa) are gradually developed on the surface of the filter media and form a biological film or slime layer known as biofilm. The development of biofilm may take few days or months depending on the influent organic concentration. The crucial point for the successful operation of a biofilter is to control and maintain a healthy biomass on the surface of the filter. Since the performance of the biofilter largely depends on the microbial activities, a constant source of substrates (organic substance and nutrients) is required for its consistent and effective operation. There are three main biological processes that can occur in a biofilter, (i) attachment of microorganism, (ii) growth of microorganism and (iii) decay and detachment of microorganisms. As the success of a biofilter depends on the growth and maintenance of microorganisms (biomass) on the surface of filter media, it is necessary to

understand the mechanisms of attachment, growth and detachment on the surface of the filter media. 1-1. Attachment of Microorganisms The mechanisms by which microorganisms can attach and colonize on the surface of the filter media of a biofilter are (i) transportation, (ii) initial adhesion, (iii) firm attachment, and (iv) colonization [Van Loosdrecht et al., 1990]. The transportation of microorganisms to the surface of the filter media is further controlled by four main processes, (a) diffusion (Brownian motion), (b) convection, (c) sedimentation due to gravity, and (d) active mobility of the microorganisms. As soon as the microorganisms reach the surface, initial adhesion occurs which can be reversible or irreversible depending upon the total interaction energy, which is the sum of Van der waals force and electrostatic force. The DLVO (Derjaguin-Landau-Verwey-Overbeek) theory is often used to describe the adhesion of the microorganisms on the surface of the filter media. The processes of firm attachment and colonization of microorganisms depend on influent characteristics (such as organic type and concentration) and surface properties of the filter media. The steric effects, hydrophobicity of the microorganisms, contact angle, and electrophoretic mobility values are taken into consideration to estimate the attachment of microorganisms on the surface of filter media. 1-2. Substrate Utilization and Biomass Growth
Korean J. Chem. Eng.(Vol. 20, No. 6)

1056

D. S. Chaudhary et al.

A biofilm is an accumulation of microorganism onto a surface. Since the microorganisms are attached to the surface, the supply of organics or substrate (food) to the microorganisms in a biofilm is mainly controlled by the bulk and surface transport phenomena. The substrate must be transported from the bulk liquid to the biofilms outer surface where it has to diffuse into the biofilm for its metabolism. The factors that influence the rate of substrate utilization within a biofilm are (i) substrate mass transport to the biofilm, (ii) diffusion of the substrate into the biofilm, and (iii) utilization kinetics within the biofilm. The other key factors that affect the performance of a biofilm process are the growth yield of the substrate and the physical factors affecting the biofilm detachment. The substrate transport from the bulk liquid to the outer surface of the biofilm can be described by Ficks first law [Eq. (1)]:
AD (S Ss ) J = ------------------------Ld (1)

Table 2. Detachment rate expressions (adopted from Hozalski, 1996) Detachment rate [ML 2T 1] kd Xf Lf kd Xf (Lf)2 kd Xf kd Xf Lf 0.58 // Lf (k/d +kd g) kd gave Xf (Lf)2 Reference Chang and Rittmann [1987], Rittmann [1989] Wanner and Gujer [1986] Bakke et al. [1990] Rittmann [1982] Speitel and DiGiano [1987] Peyton and Characklis [1992]

The substrate transport to the microorganism inside the biofilm by molecular diffusion can be described by Ficks second law [Eq. (2)] and the substrate utilization by the Monod expression [Eq. (3)].
Sb rd = D f -------- Z2
2

(2) (3)

kXf Sb ru = --------------Ks + Sb

The factors that can affect the growth of biomass in non-steady state conditions are given by Eq. (4) [Rittmann and Brunner, 1984].
YkX f Sb -A L rg = ----------------Ks + S b f f (4)

1-3. Detachment of Biomass The success of a biofilter mainly depends on the efficient maintenance of biomass attached to the filter media. Biomass detachment is one of the most important mechanisms that can affect the maintenance of biomass in a biofilter. Erosion, abrasion, sloughing, grazing or predation, and filter backwashing are the mostly observed and literally discussed detachment mechanisms. Erosion of biomass occurs due to the fluid shear whereas abrasion of biomass is the process of scraping the biocell off the surface by collision of external particle. Similarly, large patches of biomass are detached by sloughing, and a part of biomass especially on the outer surface of the biofilm may be lost due to the grazing of protozoa. Evaluation of the biomass loss due to filter backwashing is very important in operational point of view. Backwash bed expansion, mode of backwash such as air scour, filter effluent or chlorinated water backwash may affect biomass during backwashing. However, previous studies have shown that the effective biomass which is mainly responsible for the organic removal is not lost during normal filter backwash [Chaudhary et al., 2001; Ahmad and Amirtharajah, 1998]. Most of studies have been concentrated on biomass loss due to shear stress only. A summary of the reported biomass loss due to shear stress in GAC biofilter is presented in Table 2. 2. Mathematical Modeling There are very few models reported in the literature that can predict the performance of a biofilter. Most of these models are based on the assumption of steady state condition [Rittmann, 1990; RittNovember, 2003

mann and Manem, 1992; DiGiano and Speitel, 1993]. Rittmann and McCarty [1980] first introduced a steady-state biofilm model in which, the mass transport and the microbial kinetics were expressed by Ficks second law and Monod equation respectively. It was assumed that minimum bulk substrate concentration (Smin) is required to maintain the steady-state biofilm in the filter. The model describes the fundamental biological processes but does not take into account the biofilm growth with time. Chang and Rittmann [1987] developed a model for the kinetics of biofilm on activated carbon (BFAC) incorporating film mass transfer, biodegradation, and adsorption of a substrate, as well as biofilm growth. All the fundamental biological processes have been included in this model. However, the non-steady state condition due to backwashing, change in the filter bed porosity and hence the filter depth have not been considered in this model. The concept of dimensionless empty bed contact time (EBCT), which allows comparison of results among different surrogate parameters such as AOC and BDOC (Biodegradable dissolved organic carbon) was developed by Zhang and Huck [1996a] utilizing the steady-state biofilm model of Rittmann and MaCarty [1980]. Huck et al. [1994] developed a first order biofilm model. The model is more practical than accurate in predicting the performance of a biofilter. It assumes that the organic removal in a biofilter is directly proportional to the influent concentration. Billen et al. [1992] developed the CHABROL model to predict BDOC removal. The model incorporates the major microbial processes and substrates of different biodegradabilities. The model showed that BDOC removal is directly proportional to influent BDOC and EBCT. The three kinds of interaction, (i) interaction with dissolved organic matter, (ii) interaction with the solid support, and (iii) the mortality and grazing of bacteria have been incorporated in the model to describe the dynamics of the bacterial community colonizing on the support (filter media). The model consists of six state variables, namely: concentrations of rapidly and slowly hydrolysable biodegradable macromolecules of organic substances, concentrations of directly usable monomeric substances, bacterial biomass actively attached to the solid support, the bacterial biomass reversibly adsorbed to the support, and free bacterial biomass in interstitial water. The model is capable of relating the macroscopic functioning of biofilters to the kinetics of the basic microbiological processes. It can predict the fixed bacterial biomass and the biodegradable organic matter in the effluent from the characteristics of influent water for a given values of contact time and temperature. The model was also calibrated and validated with pilot and full size filters run in the Neuilly-

Biofilter in Water and Wastewater Treatment

1057

sur-Marne and Choisy-le-Roi plants, France. Most of the above-mentioned models describe the biological processes in a biofilter. There are no complete models that can predict the efficiency of the biofilter at different operating conditions. Boon et al. [1997] conducted pilot-scale biofilter experiments with sewage and developed empirical equations to predict the biochemical oxygen demand (BOD5) and ammonia removal by the biofilters. It was observed that organic and hydraulic loading rates of the biofilter can limit the organic removal efficiency of the biofilter. Hozalski and Bouwer [2001a] developed a numerical model called BIOFILT, to simulate the non-steady state behavior of biologically active filters used for drinking water treatment. The model is capable of simulating substrate (biodegradable organic matter) and biomass (both attached and suspended) profiles in a biofilter as a function of time. The model also has capability to simulate the effects of a sudden loss in attached biomass due to filter backwash on substrate removal efficiency [Hozalski and Bouwer, 2001b]. The model is very practical and it incorporates most of the fundamental processes of the biofiltration. Some of the limitations reported on this model are: (i) It is a single substrate model, (ii) It assumes that there is no mixing of the filter media during backwashing, (iii) It does not incorporate the adsorption of substrate that occurs when GAC is the filter media of the biofilter, and (iv) The model requires data on parameters to perform the simulation. Despite these limitations, the BIOFILT is the first model that incorporates the backwashing effect in the simulation, and should be considered as the most practical model to date. The simulated results of different fractions of a BOM mixture in a full-scale biofilter plant are shown in Fig. 1. Although, all the models described above are successful in modeling the fundamental biological processes of a biofilter, the other important parameters that need to be addressed in the biofilter model are change in: (i) filter bed porosity, (ii) surface area, and (iii) bed depth. The model should be able to predict the long-term performance of the biofilter at different operating conditions such as organic and hydraulic loadings changes. Alonso et al. [2001] developed a dynamic mathematical model for the biodegradation process of volatile organic compounds (VOCs) using diatomaceous earth biological support media (Celite 6 mm R-635 Bio-Catalyst Carrier). The effect of nitrate concentration, reactor backwashing, and change

Fig. 2. Effect of biofilm thickness on simulation results [Chaudhary, 2003].

in bed porosity and specific surface area of the filter media have been incorporated in the model. Chaudhary [2003] modeled the long-term performance of a GAC biofilter with low strength synthetic wastewater incorporating both initial adsorption and biodegradation processes. The model was based on the fundamental mechanisms of transport of substrate in the bulk liquid, biofilm growth, transport, and biodegradation within the biofilm, and adsorption on activated carbon. The effect of biofilm thickness on simulation n results is shown in Fig. 2. 2-1. Description of State Variables The main components that need to be incorporated in the model are: (i) substrate in the liquid bulk liquid, (ii) biomass suspended in the bulk liquid, (iii) substance diffusion and biodegradation in biofilm, (iv) biofilm growth and decay, and (v) change in bed porosity, specific surface area and bed depth. 2-1-1. Substrate in the Bulk Liquid The unsteady-state material balance on the substrate in the bulk liquid is represented by the advection-diffusion equation with adsorption and reaction terms [Eq. (5)]
C 2 C C ----------- = D ax -------- 2 t z bio ads z (5)

with initial and boundary conditions

C =C0 dC - = ( C z = 0 C z = 0 ) at z = 0 D ax = -----dz
+

dC ------ = 0 at z = L dz

(5a)

The last two terms of the Eq. (5) represent the substrate removal rates by biodegradation and adsorption respectively, and are given by:
C bio = k max -------------- Xs Ks + C 3N ads = (1 ) ---------------3 4 Rp . (5b)

(5c)

Fig. 1. Simulated removal of different fractions of a BOM mixture in a full-scale biofilter [Hozalski and Bouwer, 2001a].

2-1-2. Biomass Suspended in the Bulk Liquid The suspended biomass in the bulk liquid can be represented by
Korean J. Chem. Eng.(Vol. 20, No. 6)

1058

D. S. Chaudhary et al.

Eq. (6).
X s k max C 1 ' -------- X + ---------- Kdc ------- a'f X f = Y ------------- s ' t Ks + C (6)

with initial and boundary conditions,

Xs=Xs0 and z=0 (6a)

2-1-3. Biofilm Diffusion and Biodegradation Biofilm diffusion and biodegradation of the substrate is given by Eq. (7).
Sb S b kmax Sb X f --------------------- = D f --------t Ks + Sb x2
2

(7)

It is assumed that the substrate diffuses through biofilm where it is biodegraded by the microorganisms. 2-1-4. Biofilm Growth and Decay Since the biofilm growth rate is directly related to the biological activity, the cell growth rate (which is the sum of the cell production rate due to degradation and its decay rate) can be written as in Eq. (8).
dLf L Y kmax S ------- = --------------------- btot dr 0 K +S dt s
f

(8)

2-1-5. Change in Bed Porosity, Specific Surface Area and Bed Depth The growth of biofilm in the bed alters the bed porosity, the specific surface area, and the filter medium depth as the bioadsorption proceeds. These changes are given by the Eqs. (9), (10), and (11) respectively.
Lf 3 n L f 2 Lf -+ ----- --------- + ' = 1 ( 1 0) 1 R p 4 R p 2 R p 3 3( 1 0 ) Lf Lf +2 - 1 + ----- ( 2 n) ----a'f = -----------------Rp 2 Rp Rp
3 2 L 1 0 + L f n L f L f + -- ----- 2 ----- 3 ---- = ---------- 1 ----L0 1 ' R p 4 Rp Rp

(9)

(10)

(11)

GAC (an adsorptive media) can be a better choice than anthracite or sand (non-adsorptive media) for the removal of organic substances from tertiary wastewater or surface water [LeChevallier et al., 1992; Wang et al., 1995a, b]. A GAC filter might have less specific surface area (surface area per unit volume of filter) available for microbial attachment than a sand filter because the effective size of sand is usually smaller than GAC. Further the size of GAC micropores (1-100 nm) seem to be too small for microorganisms (typically greater than 200 nm in diameter) penetration inside these micropores (AWWA research and technical committee report, 1981). However, the macroporous structure and irregular surface of GAC offer more appropriate sites for biomass attachment. GAC can adsorb and retain slowly biodegradable components that can be biodegraded by the attached microbial mass leading to continuous bioregeneration of the GAC. It also provides protection from shear loss of biomass. Wang et al. [1995a] found the mesoporous GAC surface texture more suitable for biomass attachment than macroporous and microporous GAC. The biofilter media should provide: (i) a suitable surface for quick biomass growth, (ii) larger surface area for biomass growth, and (iii) good surface texture to hold biomass against shear and sloughing. The effect of types of media on the performance of the biofilter is shown in Fig. 3. 2. Empty Bed Contact Time (EBCT) The contact time, usually expressed as empty bed contact time (EBCT), is a key design and operating parameter of a biofilter. Zhang and Huck [1996b] have introduced the concept of dimensionless contact time incorporating EBCT, specific surface area of the medium, substrate diffusivity and rate of biodegradation. Usually the percentage removal of organic substances increases with increase in contact time up to an optimum value. Both the filter depth and hydraulic loading can be changed to increase the EBCT. Previous studies have shown that the contact time (and not the hydraulic loading) is the key variable responsible for organic removal. For a given EBCT, organic removal is independent of hydraulic loading in the range typically used in rapid filtration [Servais et al., 1994; Carlson

DESIGN CONSIDERATION The parameters that can affect the performance of a biofilter are the characteristics of filter media, hydraulic and organic loading rate, and filter backwash techniques. Other factors that can influence the performance of a biofilter are the temperature and the presence of oxidants, i.e. O3, H2O2, Cl2, and NH4Cl etc. in the influent [Urfer et al., 1997; Goel et al., 1995]. These factors should be carefully studied before designing a biofiltration system. 1. Filter Media The economical and efficient operation of a biofilter highly depends on the characteristics of its filter media. While selecting the filter media, one should also consider the source and concentration of targeted pollutants. For the treatment of primary wastewater, the right choice of the filter media can be the blast furnace slag or granite or synthetic media depending upon the volume of wastewater, whereas for the treatment of tertiary wastewater, air stream containing VOCs or for removing offensive organic substances from the drinking water supply line, GAC or anthracite or filter coal or sand could be the better choice. Previous studies have shown that
November, 2003

Fig. 3. Effect of types of filter media on the performance of biofilter [Billen et al., 1992].

Biofilter in Water and Wastewater Treatment

1059

Fig. 4. Effect of EBCT on the performance of biofilter (Chabrol model) [Laurent et al., 1999].

are usually hydrophobic in nature, are attached to filter media (GAC) with a greater force than non-biological clay particle (measured as turbidity). The difference in the detachment of these particles during backwashing should be taken into account while selecting or optimizing the backwashing of a biofilter. Previous researches have shown no major loss of biomass during backwash of the biofilter [Ahmad et al., 1998; Lu and Huck, 1993]. Servais et al. [1991] backwashed the GAC biofilter with air scour and water routinely every 50-100 hours of continuous run, but no significant difference in vertical biomass profiles before and after backwash was observed. 4. Temperature The effect of temperature on the bacterial activity on the biofilter and hence the performance of the biofilter is shown in Fig. 5. The activities of bacterial community adapted at 10 oC and 20 oC were found to increase with increase in temperature in range of 1030 oC. BIOFILTER STATE VARIABLE PARAMETERS AND THEIR MEASUREMENT 1. Substrate Biofilters are used for many purposes. It can be used for the treatment of primary wastewater, tertiary wastewater or for the treatment of potable water. Measurement of biofilter state variable parameters depends on the purpose of the use of the biofilter. If the purpose of the biofilter is to treat the primary wastewater, then the parameters that should be measured are BOD, COD, SS etc. However, when the biofilter is used for the tertiary wastewater treatment, then the organic level such as TOC could be an appropriate parameter to be measured. Similarly, the main purpose of the use of a biofilter in potable water treatment is to reduce the chlorine demand or disinfection by-product formation potential and the bacterial re-growth potential, and its measurements are expressed in terms of BDOC and AOC. Since the measurement of AOC or BDOC is of specific nature, precise measurement methods are required. Some of the commonly used methods for the measurement of AOC and BDOC as reported by Huck [1990] are: (a) Van der Kooij method, (b) Kemmy method, (c) US-EPA method, (d) Werner method, (e) Jago-Stanfield method, and (f) Billen-Servais method. In Van der Kooij method, the AOC concentration is expressed as g acetate C eq/L, whereas in Kemmy and US-EPA methods, colony-forming units (cfc/ml) is measured and then converted into AOC g/L and coliform growth response (CGR) respectively. In Werner and Jagostanfield methods, the bacterial cell concentration is measured in terms of turbidity or adenosine triphosphate (ATP) concentration. Billen-Servais method measures the biodegradable dissolved organic carbon (BDOC). 2. Biomass Growth The performance of a biofilter depends on the biomass attached to the filter media. The biomass growth and its maintenance over the surface of the filter media, on the other hand, depend mostly on the surface characteristics of the filter media itself. As mentioned earlier, different media can have different biomass growth rate and biomass retention capacity. GAC, sand, anthracite, blast-furnace slag and floating polypropylene pellets are some of the common biofilter media used in the water and wastewater treatment. Other factors that can affect the biomass accumulation are the filtration
Korean J. Chem. Eng.(Vol. 20, No. 6)

and Amy, 1995]. Huck et al. [1994] showed that the organic removal efficiency of a biofilter could be approximated by a first-order model. Servais et al. [1992] reported a linear increase in BDOC removal with the increase in EBCT between 10-30 min of the biofilter. The past studies [Price, 1994; Hozalski et al., 1995], which showed very little or no effect of EBCT on organic removal efficiency of a biofilter, might be due to partial acclimatization of the biofilter. The effect of EBCT on the performance of biofitler is shown in Fig. 4. 3. Filter Backwash It is important to select an appropriate filter backwashing technique for successful operation of a biofilter. The biomass attached to the filter media has to be carefully maintained during backwashing [Ahmad et al., 1998; Bouwer and Crowe, 1998; Bablon et al., 1988; Graese et al., 1987; Miltner et al., 1995]. Ahmad and Amirtharajah [1998] found that biological particles (measured as heterotrophic plate counts and cellular adenosine triphosphate), which

Fig. 5. Percentage of bacterial activity with respect to maximal activity with respect to temperature [bacterial community adapted at 10 oC ( ) and 20 oC ( )] [Billen et al., 1992].

1060

D. S. Chaudhary et al.

rate, filter backwashing techniques, and the organic content of the influent wastewater. Most of the studies with natural surface water showed that 3-months period is required for a GAC filter to retain maximum amount of biomass [Servais et al., 1994; Ahmad and Amirtharajah, 1998]. Several methods are adopted in practice to measure the biomass attached to the filter media depending on the availability of the analytical facilities. Usually for the biofilter used in water treatment facilities, the amount of biomass is relatively small (in microgram) and hence precise methodology for the biomass measurement is required. Ahmad et al. [1998] however, used heterotrophic plate count (HPC) to measure the biomass growth in the biofilter. The heterotrophic bacteria were enumerated by using the spread plate method according to the STANDARD METHODS [1989] section 9215C. The growth medium used was R2A agar, and incubation conditions were 20 oC for seven days. Wang et al. [1995a] used phospholipids analysis to estimate the biomass in the biofilter. About 0.5 g of GAC filter media with attached biomass were taken from the filter, and washed with dechlorinated tap water to remove the suspended solids so that the measured mass would be only the attached biomass. Basically, the method is to extract the organically bound phosphorous and then it is digested to inorganic phosphate which can be quantified by colorimetric measurement. The amount of biomass is reported as nmol lipid-P/g dry filter media (1 nmol lipid-P is equivalent to about 108 bacteria of the size of E.coli). Servais et al. [1994] suggested that it would not be possible to enumerate the bacteria attached onto activated carbon due to size and surface irregularity of the GAC, and developed a new approach to estimate the bacterial biomass. In this method, bacterial activity is measured under standard conditions, and then it is related to size of active bacterial population through the glucose the respiration rate, and finally correlated to g C of biomass by a conversion factor of 1.1 g C of bacterial biomass per nanomole of glucose respired per hour [Servais et al., 1991]. Chaudhary et al. [2001] used total dry weight method to measure the biomass in a GAC biofilter acclimatized with synthetic wastewater (Fig. 6). This method is simple and more practical to measure the biomass of relatively large quantity. Maximum biomass mea-

sured was 0.09 g per g of GAC after 42 days of operation. A minimum biomass concentration of 0.036 g per g of GAC was observed after 63 days of continuous operation. Another set of experiments conducted with filtration rate of 2.5 m/h showed a biomass concentration of 0.1 g per g of GAC in 30 days of continuous filter run. The amount of biomass accumulation thus found to be dependent on hydraulic loading rate (HLR) and the organic concentration. Carlson and Amy [1998] also found the biomass concentration profile as a function of HLR. The higher the loading rate, the greater was the initial biomass and deeper the penetration into the filter bed. The biomass concentration profile thus appears to be the most critical parameter in the design of biofiltration system. Two important changes that can be observed due to the biomass coating on the outer surface of the GAC pellet are: (i) a decrease in the fixed bed porosity, and (ii) an expansion of the GAC bed of the biofilter. The maximum bed expansion of 1.14 cm (equivalent to 22.8% expansion) was observed by Chaudhary et al. [2001] after 42 days operation. A study of GAC biofilter at Neuilly-sur-Marne water treatment plant, France [Servais et al., 1994] showed that for a given empty bed contact time (EBCT), biological removal of organic matter in GAC filters is independent of filtration rate in the range of 6-18 m/ h. They also found some decrease in biomass after 100 days of operation. However, the average biomass in the filter operating at different filtration rates but at identical contact times remained almost constant. Ong et al. [1999] working with high strength wastewater (BOD5= 389 mg/L) in an ultra-compact biofilm reactor observed 52.5% and 32.8% decrease in biomass after 38 days and 94 days of filter run respectively. The decrease in biomass in the biofilter may be due to die-off of microorganisms and its subsequent removal during backwashing. Despite the decrease in the biomass, the removal efficiency of the biofilter is not impaired, and it continues to produce consistent quality of effluent [Chaudhary et al., 2001; Ahmad et al., 1998]. APPLICATION OF BIOFILTER Biofilter can be employed either as a primary treatment unit or secondary unit in the wastewater treatment system. When the amount of wastewater is relatively small and hence a complete treatment can be accomplished in one tank (package treatment plant) which has been partitioned for pretreatment, biofiltration, and sedimentation processes (Fig. 7). Various types and shapes of plastic materials are used as the biofilter media. This type of package treatment plant is widely used to treat on-site household and industrial waste-

Fig. 6. Biomass accumulations in the GAC biofilter (Filtration rate =1 m/h, average influent TOC=3.5 mg/L) [Chaudhary et al., 2001].
November, 2003

Fig. 7. Schematics of the package biofiltration system for the treatment of household wastewater.

Biofilter in Water and Wastewater Treatment

1061

Fig. 8. Process diagram of the trickling biofiltration system for domestic wastewater treatment.

Fig. 10. Schematics of biofiltration system for surface water treatment.

water. Biofilter has successfully been used as a trickling filter for the domestic wastewater treatment. It can be used with and without other biological treatment processes depending on the characteristics of the influent, and the effluent quality requirement (Fig. 8). The rock, slag or plastic materials are used as the trickling biofilter media. The application options of trickling biofilter vary with the treatment objectives, the media type, and the nature of the other treatment units in the process train. It can be used for roughing, carbon oxidation, combined carbon oxidation and nitrification with different arrangements of two or more biofilters units. The advantages of using bio-trickling filter over the conventional activated sludge process are (i) less operational cost, (ii) less area requirement, (iii) well stabilized sludge (no sludge bulking or floating problem). In advanced wastewater treatment, biofilter can be used along with conventional physico-chemical processes such as coagulationflocculation, filtration and sedimentation (Fig. 9). The conventional filter and the biofilter units can be combined together depending on the suspended solid concentration. Since the main purpose of the biofilter is to remove the dissolved organics, the suspended particles are removed in conventional filter before subjecting the wastewater to the biofiltration system. The biofilter has similarly been assessed by many researchers as an essential part of surface water treatment for potable to reduce the microbial growth in the distribution pipe lines, corrosion potential and the disinfection by-products [Bouwer and Crowe, 1988; Carlson and Amy, 1998]. Normally, GAC biofilter is recommended to use in the surface water treatment, as in GAC biofilter the organics are removed by both adsorption and biodegradation mechanisms (Fig. 10).

PERFORMANCE OF BIOFILTER In this section, the performance of biofilter in removing organics measured in different terms such as TOC and BOD5, and some specific pollutants are discussed. 1. Surface Water Treatment A study of GAC biofiltration system conducted at Neuilly-surMarne treatment plant, France by Servais et al. [1994] using three pilot filters with varying bed depth and filtration velocity, but similar empty bed contact time (EBCT) showed that organic removal efficiency of the GAC filter for a given EBCT is independent of filtration velocity in the range of 6-18 m/h. The organic removal efficiency of the filters is shown in Table 3. This study indicates that the removal of the biodegradable organic carbon increases with the biomass growth on the GAC surface, however the nonbiodegradable organic removal efficiency of the filter decreases. An investigation of the GAC system at the Palo Alto Reclamation Facility, USA revealed that biofilter could have pseudo-steady state removal of 50% for the first year, 24% for the second year and 14% for the third year [Summers and Roberts, 1984]. Adsorption of organics and biological degradation of the organics adsorbed onto the activated carbon are two major mechanisms for the consistent removal of organics in the GAC biofiltration system. The performance of biofilters at different plants in removing organics (TOC, DOC and AOC), ammonia, and nitrate are summarized in Tables 4, 5, and 6 respectively.

Table 3. Removal of biodegradable and nonbiodegradable dissolved organic carbon (BDOC and NBDOC) from the GAC pilot filters (adopted from Servais et al., 1994) Filter I II III I II III Influent (mgC/L) BDOC 0.4 0.4 0.4 0.41 0.41 0.41 NBDOC 1.1 1.1 1.1 After 7 months 1.32 1.32 1.32 Removal % BDOC 50 45 40 56 51 49 NBDOC 49 47 43 5 7 5

Fig. 9. Schematics of biofiltration system for advanced wastewater treatment.

Filter I: GAC bed depth=1 m, filtration velocity=6 m/h, EBCT= 10 min. Filter II: GAC bed depth=2 m, filtration velocity=12 m/h, EBCT=10 min. Filter III: GAC bed depth=3 m, filtration velocity= 18 m/h, EBCT=10 min
Korean J. Chem. Eng.(Vol. 20, No. 6)

1062

D. S. Chaudhary et al.

Table 4. TOC removal efficiency of biofilters (adopted from Bouwer and Crowe, 1988) Mode of operation

Aerated biofilter (top layer GAC), full-scale
Fluidized bed, pilot-scale
Rapid sand filtration, full-scale
Biologically active GAC filter, full-scale
Biologically active GAC filter, full-scale (advanced
wastewater treatment for ground water recharge) *Preozonation employed.

Location Annet sur Marne, France Medmenham, UK The Netherlands Mulheim, Germany Orange county, California, USA

Influent concentration 3.2 mg TOC/L 2.8 mg BOD5/L 23-500 g AOC/L 1.8-2.6 mg DOC/L 25 mg DOC/L

Reduction % 38 * 29 3-84 75* 20

Table 5. Ammonia removal efficiency of biofilters (adopted from Bouwer and Crowe, 1988) Mode of operation

Aerated biofilter, full-scale
Fluidized bed, pilot-scale
Rapid sand filtration, full-scale
Biologically active GAC filter, full-scale
Biologically active GAC filter, full-scale

Location Annet sur Marne, France Medmenham, UK Mulheim, Germany Mulheim, Germany Rouen la Chapelle, France

Influent concentration mg NH4-N/L 4 2.0-2.5 1.0 0.33 1.36

Reduction % 97 100 100 94 78

Table 6. Nitrate removal efficiency of biofilters (adopted from Bouwer and Crowe, 1988) Mode of operation

Biofilter, full-scale
Biofilter, full-scale
Fluidized bed, full-scale
Fluidized bed, pilot-scale (26-45 mg methanol/L)

Location Eragny, France Eragny, France Stevenage, UK Bucklesham, UK

Influent concentration, mg NO3-N/L 37.9 13.5 15.0 14.0

Reduction % 84 50 63 100

2. Low Strength Wastewater An experimental study conducted by Chaudhary et al. [2001] at Environmental R&D Laboratory at University of Technology showed that GAC biofilter can be operated for a long time without regeneration of GAC. In this study, synthetic wastewater was prepared using three organic and seven inorganic substances [Organics: glucose, peptone, yeast extract; Inorganics: MnSO4, CaCl2, NaHCO3, NaCl, MgSO47H2O, KH2PO4, (NH4)2SO4], and the GAC bed was acclimatized with relatively lower filtration rate (1 m/h). The organic removal efficiency of the biofilter remained constant at 50-55% even after 77 days of continuous run (Fig. 11). The daily backwash adopted to avoid the physical clogging of

Fig. 11. TOC removal efficiency of the GAC biofilter (Filtration rate=1 m/h, average initial TOC concentration=3.5 mg/ L) [Chaudhary et al., 2001].
November, 2003

the biofilter did not seem to affect the organic removal efficiency of the filter. From the laboratory-scale filter study, Hozalski and Bouwer [1998] also found that biomass accumulation is not impaired by backwash with water. In their experiments, the organic removal efficiency of the biofilter was found to be unchanged after the backwash. Some of the biomass may naturally be lost during backwashing of the filter but the loss of biomass can create more sites for adsorption of organics and thus impairment is balanced. This can happen when the adsorption capacity of GAC is not fully exhausted. The effects of both influent organic concentration and filtration rate on the organic removal efficiency of the biofilter were experimentally investigated [Chaudhary et al., 2001]. It was observed that with increased filtration rates, the effluent quality became inferior to that with lower filtration rate (at which the filter was acclimatized) but the organic removal pattern remained unchanged with time. It might be due to the fact that when the hydraulic loading rate of the biofilter was increased, the EBCT is decreased and the increased organic mass loading exceeded the ability of the biomass to assimilate the available biodegradable organic substances resulting in substandard effluent quality. It should be noted that the filter column was acclimatized with relatively low concentration of organics (TOC of 3.5 mg/L) and low filtration rate of 1 m/h for the gradual growth of biomass in the filter media. The performance of the biofilter improved slightly when the influent TOC concentration was increased to 6.8 mg/L. The obvious reason for this improvement could be the increased biological activity of the microorganism. The first order steady-state model devel-

Biofilter in Water and Wastewater Treatment

1063

oped by Huck et al. [1994] also showed that the organic removal efficiency of the biofilter is approximately directly proportional to the influent organic concentration. However, when the influent concentration was increased from 6.8 mg/L to 11.2 mg/L, the removal efficiency of the filter was higher initially and then decreased with time. The experimental results thus indicates that the biomass profile is the most critical parameter in the design of a biofiltration system, and that the biofilter should be operated as close to steady-state conditions as possible to achieve optimum organic removal efficiency. The sudden increase in the flow rate and influent concentration can change the efficiency of the biofilter temporarily, but if the steady-state biomass condition is allowed to develop, the organic efficiency of the biofilter would be equivalent to that of the organic or hydraulic loading rate at which the filter is first acclimatized. LeChevallier et al. [1992] and Prevost et al. [1992] have also observed decreases in the organic carbon removal with the decrease in EBCT. LeChevallier et al. [1992] found an increase in TOC removal from 29 to 51.2 percent when EBCT was increased from 5 to 20 min i.e. when the filtration velocity was decreased by four times. However, Carlson and Amy [1998] have reported from their pilot scale experimental studies that organic removal in a biofilter is limited either by biodegradable organic matter (BOM) formation or biomass concentration, not by filter operating parameters. They also found that optimum organic removal efficiency of the biofilter was at the loading rate to which the filter was acclimatized, and if the steady-state biomass conditions were allowed to develop, even at higher hydraulic loading rate, the removal efficiency of biofilter would increase to that found at the lower hydraulic loading rate (at which the filter was first acclimatized). 3. High Strength Wastewater A full-scale study was conducted by Boon et al. [1997] employing six biofilter columns of different diameters (6-26 m) with blastfurnace slag and granite as filter media. The performance of the biofilters is summarized in Table 7. The BOD5 and ammonia-N removal efficiency of the filters varied from 85%-97% and 55%-98% respectively. CONCLUSION 1. Biofilter can effectively be used in an economical manner to produce high quality of effluent due to its consistent TOC removal efficiency, long operational life and simplicity in operation. 2. The biological activity led to a consistent effluent organic con-

centration over a long period of time. The daily backwash usually adopted to ease the filter bed seems to have no effect on the biomass growth rate, and hence the effluent quality. Its performance however can be affected by the filtration rate and the influent organic concentration, suggesting that the biofilter should be operated in the same conditions at which it is acclimatized for its optimum and consistent organic removal efficiency. 3. A correct choice of filtration rate and GAC medium depth with appropriate backwash can lead to a long-term operation with consistent and superior effluent quality. 4. The mathematical model should incorporate the biofilter parameters estimated for different operating conditions (such as acclimatization filtration rate and initial organic concentration) to verify the adaptability of the model in practice. NOMENCLATURE A Af a'f btot C C0 D Dax Df J Ks k kmax Kdc kd k/d // kd L Ld Lf Lo n N r Rp : surface area normal to the filter media [m2] : biofilm surface area [m2] : specific surface area of the pellet with biomass [m2] : total shear and decay loss [s 1] : liquid phase organic concentration [mg/L] : initial liquid phase organic concentration [mg/L] : molecular diffusion coefficient for the substrate in the bulk liquid phase [m2/s] : axial dispersion coefficient [m2/s] : molecular diffusion coefficient within biofilm [m2/s] : substrate flux into the biofilm [mg/m2/s] : monod half-velocity coefficient [mg/L] : maximum specific rate of substrate utilization [mg of substrate/mg of biomass/s] : maximum rate of substrate utilization [mg/mg/s] : decay coefficient [s 1] : detachment rate coefficient [expression dependent units] : detachment rate coefficient [expression dependent units] : detachment rate coefficient [expression dependent units] : bed depth with biofilm [m] : diffusion layer thickness [m] : biofilm thickness [m] : initial bed depth [m] : number of pellet in contact : substrate uptake rate of the biofilm [mg/m2/s] : radial distance measured from the center of the pellet [m] : radius of the pellet [m]

Table 7. Performance of biofilter for the treatment of high strength wastewater [Boon et al., 1997] Biofilter Flow diameter/size (L/s) (m) 16 22 26 10216.8 6 6 3.24 7.3 27.2 92.4 0.92 0.44 Settled sewage Alkalinity (mg/L) 142 240 300 171 165 331 pH 6.6-8.2 7.2-7.6 7.4-7.8 7.0-7.7 6.9-7.3 7.0-8.0 Ammonia.N (mg/L) 19.1 15.7 22.5 17.4 23.5 44.0 SS Alkalinity BOD5 (mg/L) (mg/L) (mg/L) 108 72 136 130 146 265 84 77 118 102 126 146 9 100 120 73 79 166 pH 5.8-6.2 7.6-7.7 7.5-7.7 6.9-7.4 6.7-7.2 7.0-8.0 Final Effluent Ammonia-N BOD5 SS (mg/L) (mg/L) (mg/L) <2.2 <2.0 <2.3 <2.5 14.6 19 11 7.2 6.3 14 24 32 36 17.5 20 30 45 53

Korean J. Chem. Eng.(Vol. 20, No. 6)

1064

D. S. Chaudhary et al.

rd rg ru S Sb Ss t x Xf Xs Xs0 Y Z z

: rate for substrate accumulation due to diffusion [mg/m3/s] : rate of biomass growth within the biofilm [mg/s] : rate of substrate utilization in the biofilm [mg/m3/s] : substrate concentration in the bulk solution [mg/L] : substrate concentration in the biofilm [mg/L] : substrate concentration at the outer surface of the biofilm [mg/L] : time [s] : distance perpendicular to the pellet [m] : biomass density of biofilm [mg/L] : suspended biomass concentration [mg/L] : initial suspended biomass concentration [mg/L] : yield coefficient [mg/mg] : distance normal to the biofilm surface [m] : distance along the biofilter length [m]

Greek Letters : filtration efficiency : bed porosity ' : bed porosity with biofilm o : initial bed porosity : empty bed contact time [s] : biofilm shear loss coefficient [s 1] g : specific biomass growth rate [s 1] gave : average specific growth rate [s 1] : fluid velocity [m/s] : fluid shear stress [Nm 2] REFERENCES
Ahmad, R., Amirtharajah, A., Al-Shawwa, A. and Huck, P. M., Effects of Backwashing on Biological Filters, J. AWWA, 90(12), 62 (1998). Ahmad, R. and Amirtharajah, A., Detachment of Particles during Biofilter Backwashing, J. AWWA, 90(12), 74 (1998). Alonso, C., Suidan, M. T., Kim, B. R. and Kim, B. J., Dynamic Mathematical Model for the Biodegradation of VOCs in a Biofilter: Biomass Accumulation Study, Env. Sci. Tech., 32(20), 3118 (1998). AWWA Research and Technical Practice Committee on Organic Contaminants, An Assessment of Microbial Activity of GAC, J. AWWA, 73(8), 447 (1981). Bablon, G. P., Ventresque, C. and Ben Aim, R., Developing a SandGAC Filter to Achieve High Rate Biological Filtration, J. AWWA, 80(12), 47 (1988). Bakke, R., Characklis, W. G., Turakhia, M. H. and Yeh, A.-I., Modelling a Monopopulation Biofilm System: Pseudomonas Aeruginosa, In Biofilms. Characklis, W. G. and Marshall, K. C., eds., New York, Wiley (1990). Billen, G., Servais, P., Bouillot, P. and Ventresque, C., Functioning of Biological Filters Used in Drinking Water Treatment Plant-the Chabrol Model, J. W SRT-Aqua, 41(4), 231 (1992). Boon, A. G., Hemfrey, J., Boon, K. and Brown, M., Recent Developments in the Biological Filtration of Sewage to Produce High-Quality Nitrified Effluents, J. Chart. Insti. Water Envi. Manage., 11(12), 393 (1997). Bouwer, E. J. and Crowe, P. B., Biological Processes in Drinking Water Treatment, J. AWWA, 80(9), 82 (1988). Carlson, K. H. and Amy, G. L., BOM Removal during Biofiltraiton,
November, 2003

J. AWWA, 90(12), 42 (1998). Carlson, K. H. and Amy, G. L., Relative Importance of EBCT and HLR on the Removal of BOM during Biofiltration, Proceedings AWWA WQTC, New Orleans (1995). Chang, H. T. and Rittmann, B. E., Mathematical Modelling of Biofilm on Activated Carbon, Env. Sci. Tech., 21(3), 273 (1987). Chaudhary, D. S., Vigneswaran, S., Ngo, H. H. and Jegatheesan, V., Applicability of Granular Activated Carbon (GAC) Biofilter in Wastewater Treatment and Reuse, Proceedings of Asian Waterqual 2001, IWA Asia-Pacific Regional Conference, Fukuoka, Japan, 2, 685 (2001). Chaudhary, D. S., Adsorption-Filtration Hybrid Systems in Wastewater Treatment and Reuse, PhD Dissertation, University of Technology, Sydney, Australia (2003). DiGiano, F. A. and Speitel Jr., G. E., Biofilm Computer Models: Time for Practical Applications, J. AWWA, 82(5), 26 (1993). Goel, S., Hozalski, R. M. and Bouwer, E. J., Biodegradation of NOM: Effect of NOM Source and Ozone Dose, J. AWWA, 87(1), 90 (1995). Graese, S. L., Snoeyink, V. L. and Lee, R. G., Granular Activated Carbon Filter-Adsorber Systems, J. AWWA, 79(12), 64 (1987). Hozalski, R. M. and Bouwer, E., Deposition and Retention of Bacteria in Backwashed Filters, J. AWWA, 90(1), 71 (1998). Hozalski, R. M. and Bouwer, E., Non-steady State Simulation of BOM Removal in Drinking Water Biofilters: Model Development, Water Res., 35(1), 198 (2001a). Hozalski, R. M. and Bouwer, E., Non-steady State Simulation of BOM Removal in Drinking Water Biofilters: Application and Full-scale Validation, Water Res., 35(1), 211 (2001b). Hozalski, R. M., Goel, S. and Bouwer, E., TOC Removal in Biological Filters, J. AWWA, 87(12), 40 (1995). Hozalski, R. M., Removal of Biodegradable Organic Matter in Drinking Water Biofilters: Experimental Studies and Model Development, PhD dissertation, Johns Hopkins University, Baltimore, Maryland, USA (1996). Huck, P. M., Measurement of Biodegradable Organic Matter and Bacterial Growth Potential in Drinking Water, J. AWWA, 82(7), 78 (1990). Huck, P. M., Zhang, S. and Price, M. L., BOM Removal During Biological Treatment: A First Order Model, J. AWWA, 86(6), 61 (1994). Laurent, P., Prevost, M., Cigana, J., Niquette, P. and Servais, P., Biodegradable Organic Matter Removal in Biological Filters: Evaluation of the Chabrol Model, Water Res., 33(6), 1387 (1999). LeChevallier, M. W. and Lowry, C. D., Disinfecting Biofilms in a Model Distribution System, J. AWWA, 82(7), 87 (1990). LeChevallier, M. W., Becker, W. C., Schorr, P. and Lee, R. G., Evaluating the Performance of Biologically Active Rapid Filters, J. AWWA, 84(4), 136 (1992). Lu, P. and Huck, P. M., Evaluation of the Methods for Measuring Biomass and Biofilm Thickness in Biological Drinking Water Treatment, Proceedings AWWA WQTC, Miami (1993). Metcalf and Eddy Inc., Wastewater Engineering: Treatment, Disposal and Reuse, 3rd edition, Revised by Tchobanoglous, G. and Burton, F., McGraw-Hill, Inc., Singapore (1991). Miltner, R. J., Summers, R. S. and Wang, J. Z., Biofiltration Performance: Part 2, Effect of Backwashing, J. AWWA, 87(12), 64 (1995). Niquette, P., Prevost, M., Servais, P., Beaudet, J. F., Coallier, J. and Lafrance, P., Shutdown of BAC Filters: Effects on Water Quality, J. AWWA, 90(12), 53 (1998).

Biofilter in Water and Wastewater Treatment

1065

Ong, S. L., Ng, W. J., Lee, L. Y., Lui, Y. and Hu, J. Y., Effect of Biofilm Morphology on the Performance of an Ultra-compact Biofilm Reactor for Sewage Treatment, Proceedings Asian Waterqual99, 7th IAWQ Asia-Pacific Regional Conference, Taipei, Taiwan, 1, 53 (1999). Peyton, B. M. and Characklis, W. G., Kinetics of Biofilm Detachment, Water Sci. Tech., 26(9-11), 1995 (1992). Prevost, M., Coallier, J., Mailly, J., Desjardins, R. and Duchesne, D., Comparison of Biodegradable Organic Carbon Techniques for Process Control, J. Water SRT-Aqua, 41(3), 141 (1992). Price, M. L., Ozone and Biological Treatment of DBP Control and Biological Stability (90649), AWWARF, Denver. Rittmann, B. E., Analyzing Biofilm Processes Used in Biological Filtration, J. AWWA, 82(12), 62 (1990). Rittmann, B. E. and Brunner, C. W., The Non-steady State Biofilm Process for Advanced Organics Removal, J. Water Poll. Con. Fed., 56(7), 874 (1984). Rittmann, B. E. and Huck, P. M., Biological Treatment of Public Water Supplies, CRC Critical Reviews in Environmental Control, 19, 119 (1989). Rittmann, B. E. and Manem, J. A., Development and Experimental Evaluation of a Steady-state, Multispecies Biofilm Model, Biotech. Bioeng., 39, 914 (1992). Rittmann, B. E. and McCarty, P. L., Model of Steady-state Biofilm Kinetics, Biotech. Bioeng., 22, 2343 (1980). Rittmann, B. E., Detachment from Biofilms, In Structure and Function of Biofilms, Edited by W. G. Characklis and P. A. Wilderer, New York, Wiley (1989). Rittmann, B. E., The Effect of Shear Stress on Biofilm Loss Rate, Biotech. Bioeng., 24, 501 (1982). Servais, P., Billen, G. and Bouillot, P., Biological Colonization of Granular Activated Carbon Filters in Drinking-Water Treatment, J. Environ. Eng., 120(4), 888 (1994). Servais, P., Billen, G., Bouillot, P. and Benezet, M., Pilot Study of Bio-

logical GAC Filtration in Drinking Water Treatment, J. Water SRTAqua, 41(3), 163 (1992). Servais, P., Billen, G., Ventresque, C. and Bablon, G. P., Microbial Activity in GAC Filters at the Choisy-le-Roi Treatment Plant, J. AWWA, 83(2), 62 (1991). Speitel, G. E. and DiGiano, F. A., Biofilm Shearing under Dynamic Conditions, J. Environ. Eng., 113(3), 464 (1987). Standard Methods for the Examination of Water and Wastewater, APHA-AWWA-WEF, Washington, D. C. (1989). Summers, R. S. and Roberts, P. V., Simulation of DOC Removal in Activated Carbon Beds, J. Environ. Eng., 110(2), 73 (1984). Urfer, D., Huck, P. M., Booth, S. D. J. and Coffey, B. M., Biological Filtration for BOM and Particle Removal: A Critical Review, J. AWWA, 89(12), 83 (1997). Van der Kooij, D., Visser, A. and Hijnen, W. A. M., Determining the Concentration of Easily Assimilable Organic Carbon in Drinking Water, J. AWWA, 74(10), 540 (1982). Van Loosdrecht, M. C., Influence of Interfaces on Microbial Activity, Microbial Reviews, 54(1), 75 (1990). Wang, J. W., Summers, R. C. and Miltner, R. J., Biofiltration Performance: Part 1, Relationship to Biomass, J. AWWA, 87(12), 55 (1995a). Wang, J. Z., Summers, R. S. and Miltner, R. J., Biofiltration Peroformance: Part 2, Effect of Backwashing, J. AWWA, 87(12), 64 (1995b). Warnner, O. and Gujer, W. A., Multispecies Biofilm Model, Biotech. Bioeng., 27, 314 (1986). Yang, L., Chou, L. and Shieh, W. K., Biofilter Treatment of Aquaculture Water for Reuse Applications, Water Res., 35(13), 3097 (2001). Zhang, S. and Huck, P. M., Removal of AOC in Biological Water Treatment Processes: A Kinetic Modeling Approach, Water Res., 30(5), 1195 (1996a). Zhang, S. and Huck, P. M., Parameter Estimation for Biofilm Processes in Biological Water Treatment, Water Res., 30(2), 456 (1996b).

Korean J. Chem. Eng.(Vol. 20, No. 6)