Hzhzhzhd1 4hhh

Multiple parallel flow streams, four or more units per flow stream, single-stage units n

LJ n

Figure 5.14 Various schemes of staging RBC units.

The biomass of the RBC.

If an RBC is supplied with secondary influent, the unit will be divided into four sections. The first section will not be able to accomplish nitrification, because of a high content of organic matter and, therefore, no nitrifying population will be able to develop.

Both nitrification and organic oxidation will be carried out in the second section. The waste water content of ammonium is high and, therefore, the nitrification is relying upon on the oxygen content in the waste water and on the size of the nitrifying biomass, developed in relation to the size of the heterotrophic biomass.

In the third section most of the organic load in the waste water is oxidized, so that this section will function as the nitrifying section. As in trickling filter processes, nitrification will only proceed after the carbon concentration has been substantially reduced. Only the oxygen content in the waste water will limit the nitrifying rate.

In the fourth section the ammonium content is so low that it is not the oxygen content that limits the nitrification rate, but the content of ammonia itself. It is, therefore, important to have full control over the organic content in the different parts of an RBC plant, and to design at least four or five modules in series, if nitrification is required, because as illustrated in Section (3.13) ammonia itself can Inhibit the nitrification rate.

5.7.2 Equations for Modelling the RBC Reactor

Matsuo and Yamamoto (1985); Watanabe (1985) and Gujer and Boiler (1990) have all modelled the process of RBC units.

Gujer and Boiler (1990) proposed a model containing two levels: a microscopic level and a macroscopic level. The microscopic level considered the transport and reaction processes within the biofilm. The macroscopic level described the system as a whole. Mixing conditions within the individual compartments, influent and effluent transport processes, gas exchange processes, exchange of substrate, nutrients and biomass within the biofilm, and reactions catalyzed by biomass in suspension were considered as Important factors in the performance of the RBC.

Three submodels, a kinetic model, a biofilm-model, and a reactor compartment model were proposed to take account of the above factors.

The kinetic sub-model.

The equations proposed by Gujer and Boiler (1990) for the kinetic sub-model were the same as outlined in Section 5.4 for the biofilm kinetics, depending on either zero order or half order kinetics.

The biofilm sub-model.

The biofilm sub-model takes the following variables into account: the dissolved components, the particulate components, the removed biomass, the surface floccula-tion and the thickness of the biofilm. The different equations used in this sub-model are shown in Table 5.9 and the relevant constants are given in Table 5.10.

The reactor sub-model.

The reactor sub-model takes Into consideration the rotation of the RBC and the design of the reactor compartment.

Variation in the concentration of dissolved oxygen in the depth of the biofilm, due to rotation of the RBC, depends upon the processes of diffusion and reaction. The depth of penetration of dissolved components due to molecular diffusion is given by:


L| = Depth of penetration of compartment i during time t in metres.

D|" = Effective diffusion coefficient within the biofilm, assumed to be 80% of the value in pure water.

Table 5.9 The different equations used in the biofilm sub-model for the RBC.

Process Equation

Transport of dissolved components where J = 0 for z = L,

Dissolved component d si dJs

dt dz S1

Symbols z = depth of biofilm; z = 0 at surface and z = LB at support material.

J = flux of component i due to molecular diffusion within the biofilm.

Sj (z) = concentration of dissolved component i at biofilm depth z.

Dj = effective diffusion coefficient within the biofilm, assumed to be 80% of the value in pure water.

rs j = transformation rate of the dissolved component i per unit volume of biofilm.

Process Equation

Particulate components dxj dj, . dt dz where

Surface flocculation



Xj = concentration of particulate species i within the biofilm.

X,ot = sum of all particulate species concentration.

Jzi = flux of particulate species i within thé biofilm.

rxj = rate of production of particulate species i within the biofilm.

JFL0C i = flux of particulate material i flocculated form from the bulk liquid to the surface of the biofilm.

kFLOC = flocculation mass transfer coefficient.

Process Equation

JSHEAR,I - KSHEAR " LB ' XB,i kshear = 10 d"1 for primary effluent

= 0,05 "1 for secondary and tertiary effluent.

Biofilm thickness


^shear i = *'ux particulate material i, sheared from thé surface of the biofilm to the reactor bulk liquid.

kshear i = Shear rate constant.

XB j = concentration of particulate material i, at the surface of the biofilm (z=0)

Table 5.10 Different constants proposed for use in the Biofilm sub-model.

Symbol Unit

Heterotrophic organisms:

Maximum growth rate Saturation coefficient for COD for NH4+ -N for 02 for HCO3" for NO3

Denitrification coefficient DEN 0,70 g/m3

Decay rate bH 0,35 d"1

Yield coefficient (COD/COD) YH 0,57 g/g

Fraction particulate decay product f, 0,08

Nitrogen content of biomass (N/COD) 0,06 g/g

Decay product 0,05 g/g


Maximum growth rate Hmax,N 0,35 d 1 Saturation coefficient for NH4-N KNH4 0,70 g/m3

for HCO3" 0,20Mol/m3

Decay rate bN 0,05 d"1

Yield coefficient (C0D/N02" -N) 0,18 g/g


Maximum growth rate H nb 0,60 d"1

Ks knh4+

ho ha hno

Saturation coefficient

for NH4+-N


0,05 g/nri

for N02"-N


0,50 g/m:

for 02


0,10 g/m:

Decay rate


0,09 d"1

Yield coefficient (C0D/N03"-N)


0,06 g/g

Diffusion coefficients within a biofilm correlated for temperature (10 °C) and reduction to 80% of values in pure water.

Dissolved oxygen 106 • 10"6 m2/d

Degradable COD 31-10"6m2/d

Nitrate 85-10"6m2/d

Rate constants for biofilm surface reactions

Flocculation rate KFL0C 0,10 d"1

Shear rate constants KSHEAR

Primary effluent 0,10 d"1

Secondary and tertiary effluent 0,05 d"1

Gujer and Boiler (1990) demonstrate that It Is only at a very slow rotation speeds and a low concentration of residual pollutants, that the effect of the rotation of the support material must be considered.

RBC's today are usually compartmentalized; each drum of rotating surface areas is calculated as an individual reactor compartment.

For the Nth reactor compartment the substrate balance is written as:

= <£>+*) * (CJilM-Ci>w) +rliK*VK-JltK*AH (5-24)


VN = Volume of reactor N (bulk water phase) in m3.

C| N = Bulk concentration of dissolved or suspended component i in reactor N in kg per m3.

Q + R = Influent and recycle flow rate in m3 per hour.

rj N = Rate of production of component i within the bulk liquid in kg i/m3 * day.

J| N = Flux of component i into the biofilm of reactor N, in kg / m2 *


An = Surface area of support material in reactor A in m2.

The entire reactor system can then be modelled as a series of reactors with the option of recirculation of effluent from the last compartment to the first one. It is also possible to reverse the flow, either from the first to the last or, alternatively, from the last to the first reactor.

5.7.3 The Application of the RBC

Rotating Biological Contactors are popular in small-scale waste water treatment plants (< 100 P.E.), because of their easy maintenance, low sludge production, and low power requirements. One of the key costs in the production of an RBS is the support material. The RBC is therefore, ideal for the treatment of small volumes, which cannot easily be connected to a central treatment system for economic or geographic reasons.

The RBC can be used as a combined oxidation and nitrification system for secondary effluent, or as a tertiary nitrifying system depending on the composition of the influent waste water. Examples of anaerobic denitrification with an RBC are also presented in the literature (Hosomi et al. 1991).

The different functions of the RBC in the nitrogen removal processes are listed below:

A) Combined oxidation and nitrification with an RBC unit.

Treatment of industrial waste water or small scale waste water treatment plants, with an RBC treatment plant, will usually provide a unit with combined oxidation of carbonic material and nitrification of ammonium to nitrate. Because the waste water is a mixture with high carbon and ammonium contents which act as substrate for both oxidizing and nitrifying bacteria, the bacteria will compete for the space on the RBC disks.

B) Nitrification with an RBC unit.

Using the RBC as a tertiary nitrifying treatment plant has been shown to be highly efficient. The RBC units can, therefore, be expected to be used for final refinement, in an effort to reach present effluent standards for nitrogen content, because they can be integrated into existing flow schemes.

C) Denitrification with an RBC unit.

Denitrification in RBC systems may be observed in the following two situations: 1) Addition of oxygen may be reduced, either by nearly complete submersion of the rotating contactors, or by maintaining an atmosphere poor in oxygen. This situation will allow denitrification even at low levels of the organic loading, if nitrate is fed to the reactor, for example via recirculation of nitrate from the effluent waste water.

2) Denitrif¡cation may occur in the depth of a biofilm, where oxygen has fallen to insignificant levels. This requires high concentration of organics and nitrate to secure denitrifying conditions in the depth of a biofilm. Since the recirculation of nitrate results in dilution of the concentration of carbon compounds, this situation may only occur with industrial waste water where concentration of carbon compounds is high.

D) Simultaneous nitrification and denitrification (SND) with an RBC unit.

Masuda etal. (1991) reported the loss of nitrogen in an RBC plant treating the leachate from a sanitary landfill located at Miyazaki, Japan. The loss of nitrogen was appreciable during the summer. The RBC was covered by a hood, and during the summer the oxygen pressure in the hood was 1.8 to 1.9 atm., which was a little less than in an unhooded RBC. Matsuda and his co-workers measured the production of nitrogen gas from the biofilm, using a covered RBC, to be able to observe denitrification. In order to explain the loss of nitrogen, the authors made the following hypothesis: Nitrifiers and denitrifiers co-exist in a biofilm; the denitrifiers become active, if the transfer rate of oxygen to the biofilm decreases sufficiently to result in the formation of a micro-anaerobic environment.

Halling-Sorensen and Hjuler (1992:1993) have observed the same occurrence in a submerged filter, using clinoptilolite as the media.(See section 5.8.1).

Masuda and his co-workers conducted a series of experiments to discover the factors which influence simultaneous nitrification and denitrification (SND). They showed that the highest capacity of the SND in the RBC unit, in midsummer, was a total conversion of 130 g/m3 NH4+ - N to 80 g/m3 gaseous nitrogen N2, and the remaining 50 g/m3 was converted to N03" - N. The efficiency of SND was accordingly 61,5 %.

5.7.4 Recent Development in the RBC Technology

Much effort has been made to enhance both capacity and effluent quality of the RBC treatment systems because in the near future, full nitrification will have to be adopted in many treatment plants in order to reach present effluent standards. Using the RBC as a tertiary treatment step, it can in most cases be integrated into existing flow schemes.

Boiler etal. (1990) proposed a two-stage nitrifying RBC, including precipitation, primary clarification and a solid separation step; this was after the use of the first BOD-removing RBC's using a cloth filter, and finally a nitrifying RBC with the possibility of reversing the flow so as to obtain a higher utilization of the surface of the biomass carrier throughout the RBC. (see Fig. 5.15). This design also provides a better possibility for fluctuations of ammonia in the waste water because of the higher nitrification potential. This type of RBC plant, with filtration before nitrification and two-sided loading of waste water, requires about 40% less surface area and volume than a conventional RBC with one-sided flow, where a nitrification capacity of 1,8-2,9 g N/ m2 * d, is established.

To further enhance the capacity of the RBC, Wanner et al. (1990) proposed a packed-cage RBC. This is an RBC which is a combination of suspended and fixed-film biomass. The discs or groups of discs from a conventional RBC were replaced with a cage, packed with a random medium. The cage is equipped with tubular aeration and mixing elements.

The combination of suspended and fixed film biomass should enhance the capacity for nitrification and lower the the cost, because aeration of the activated sludge is separated from the rotation of the RBC and, therefore, an external source of air is avoided, and a large biomass is developed. This combination should make this design suitable for plants handling the treatment of 500 to 800 P.E.

To improve the effluent quality of the RBC process, Tanaka et al. (1991) investigated the behaviour of the fine particles throughout the processes; they found that an increase in the hydraulic retention time in the RBC reduced the amount of fine particles and increased the amount of coarse suspended solids, which are easily removed by clarification.

5.7.5 Nitrogen Loading Capacity and Removal Efficiency of the RBC-process

If nitrification is desired, loading rates should be reduced to 0,03 to 0,08 m3 / m2 • d. This is about one third of the capacity of an RBC when applied only to the removal of organics.

From primary clarifier

for BOD5 Removal

Cloth filter

Backwash water

Figure 5.15 The two-stage nitrifying RBC, including precipitation, primary clarification and a solids separation step after the first BOD removing RBC's with a cloth filter and lastly an RBC for nitrification. (From Gujer and Boiler 1990).

The following mass balance equation can be used to calculate the removal of ammonium per m2 filter per day as an average for the whole filter. Figure 5.18 show the relationship between nitrification capacity and temperature in an RBC unit.


Q = loading rate for waste water in m3 /day.

A'= the total disc area in m2.

10 15 20 25

Temperature °C

Figur 5.16 The relationship between nitrification capacity and temperature in an RBC unit (partly after EPA (1984) and La Cour Jansen and Henze (1990)).

Figure 5.16 show the relationship between nitrification capacity and temperature in an RBC unit, and Table 5.11 some examples of nitrifying removal rate for the RBC using different types of waste water. Table 5.11 show the removal rate with different applications of the RBC.

Table 5.11 The removal rate for the RBC using different types of waste water.

Wastewater Nitrification type rate g N /m2 • d maximum minimum

Domestic 1,69 2,56

Percolate 2,42 2,66

Fertilizer industry 2,36 2,67

Leather industry 2,35 2,62

Sewage water 1,53 1,97

Source La Cour Jansen and Henze (1990)

Table 5.12 The removal rate with different applications of the RBC.


Treatment step


Temp. "C

Process rate nitrification gN/nf d




800 P.E.



Boiler etal. (1990)



secondary (combined RBC and activated sludge)

less than 100 P.E.



Wanner et al.(1990)

Combined nitrification/oxidation

compact RBC




Ahn and Chang (1991)

Combined nitrification/oxidation

RBC with simultaneous nitrification and denitrification




Matsuda etal. (1991)

5.7.6 Advantages and Disadvantages of the Nitrifying RBC

The following points summarize the major advantages and disadvantages connected with the use of the RBC in the process of nitrogen removal in treatment plants.


1. Only a small land area is required.

2. Ability to obtain a high content of biomass per m3 or m2 of disc because of the highly developed disc units and, therefore, the lower contact time with the waste water.

3. Simple operation of the equipment.

4. Ability to handle shock loads and, therefore, suitable for treatment of highly concentrated industrial waste water.

5. Ability to achieve a high degree of waste water purification, including nitrification.

6. Good performance even with a low oxygen level in bulk waste water because most oxygen is absorbed during rotation in the air phase.

7. Good performance of tertiary nitrification and, therefore, a solution to introduction of full biological removal of nitrogen in existing plants.

8. Using an RBC unit, pumping large amounts of waste water is avoided, because the water is passed slowly through the basin, where the contactor is rotating.

1. Enclosures are necessary to protect against low temperatures, rain and wind.

2. High capital cost.

3. Upsets can and do occur, because of too great wash-out of biofilm.

5. Most RBC's are mainly designed for BOD removal, although some nitrification may occur in some plants.

4. Lack of documented operating experience.

5.8 Submerged Filters

Submerged biological filters (also known as biological aerated filters or contact aerators) are filters where the fixed material upon which the biofilm develops is continously submerged in the waste water they treat being treated.

The use of submerged filters has received renewed interest because of the development of plastic media and other sorts of filter media upon which large quantities of bacteria can grow. Also it has been shown that submerged biological filters may be very efficient at nitrification. Dillon and Thomas (1990).

Submerged filters (Fig. 5.1) can be designed both in up-flow or down-flow modes. In both cases, there is often a combination of both fixed-film and suspended growth between the filter media. Air is supplied to provide oxygen to the microorganisms, to promote mixing, and to scrub excess biofilm from the filter media to prevent irregular sloughing and plugging problems.

Because of the large biomass concentration, the contact time is often, low compared to other treatment systems to achieve the same efficiency.

Only few submerged filters are installed for nitrification. But indications are that they can be cost effective from both a capital standpoint and an operation and maintenance standpoint, that they reduce land area requirements, and that used, as teritiary treatment, have an efficiency of up to 90 percent of nitrogen removal with very low retention times.

Examples of the use on full size plants, of a biological fixed-film reactor for combined oxidation and nitrification treatment step for treatment of municipal waste water, are the systems Biocarbone and Biofor developed by respectively OTV and Degremont. (Dillon and Thomas 1990; Gilles 1990; Mange and Gros 1990; Paffoni et al. 1990, and Rogella and Bourbigot 1990).

The Biocarbone process use grain-sized biodagene (expanded schist) as bedvolume and the Biofor use spherical biolite as bedvolume.

The Biocarbone is a counter-current, granular media, aerobic filter with a water down-flow and an air up-flow. Its name is related to the earlier use of activeted carbon as matrix. Biofor is an abbreviation form from Biological Oxygenated Reactor. Biofor is defined as an aerobic treatment using fixed biomass on a 1-5 mm granualr medium with an upflowing co-current of injected air and water (Paffoni et al. 1990).

These two processes primaryly differ in the fluid direction, namely a cocurrent in the Biofor process and a counter-current in the Biocarbone process.

The case study presented involves the use of simultaneous nitrification and denitrification (SND) with an upflow fixed bed, applying clinoptilolite as matrix. The combination of nitrification and denitrification in one single reactor has been described in the literature (Matsuda et al. 1987; 1991) The development of the process described has been conducted at the Section of Environmental Chemistry in the Royal Danish School of Pharmacy in Copenhagen, Denmark. During the summer 1992 the first pilot plant project was built as tertiary treatment step of slaughterhouse waste water.

5.8.1 Case Study;

Simultaneous Nitrification and Denitrification (SND) as Tertiary Treatment Step, Using a Submerged Biofilter of Clinoptilolite


The potential for using a simultaneous nitrification and denitrification (SND) upflow fixed bed reactor (UFBR) as a tertiary treatment step for removing nitrogen from waste water is examined in this case study. Clinoptilolite (with a grain size of 2.0-4.0 mm) was used as supporting medium for the bacterial growth. As indicated in Chapter 9 on ion-exchange, clinoptilolite is a natural zeolite which selectively sorbs NH4+. Furthermore the media has a porous surface, and has a high specific surface area, ideal for bacterial growth. The removal of the adsorbed ammonium from the zeolite by nitrifying bacteria allows regeneration of the zeolite surface and thus enables the same zeolite to be used repeatedly.

Thus the purpose of this case study is to explain the mechanisms and show the results of a single-stage simultaneous nitrification and denitrification (SND) reactor that biologically transforms ammonium-N to nitrogen gas, with ethanol as electron donor for denitrification.

Laboratory reactors were constructed (Fig. 5.17) of plexiglass tubes and used in three different runs using clinoptilolite as media. The loads conducted during the 3 different experimental runs, each of the duration of six months, are presented in Table 5.14.

Table 5.13 Denitrification rates depending of nature of medium surface in packed-bed coloumns

Nature of

Media trade name

Specific surface area

Denitrification rates

medium surface


g/m3 d (at stated temp0 C)

High porosity corrugated sheet modules or dumped media

Kock Flexirings

2,13 3,34

45 (13°C), 53 (15°C), 54 (17°C), 136 (20°C), 115 (21°C), 61 (23°C),

336 (27°C)

Envirotech Surface


40 (10-23°C)

Intalox Saddelse


216-417 (20°C)

Rashig Rings


192-304 (25°C) 100-120 (20°C)

Filter A and B

J1,651 Polprasert and 13,11/ Park (1986)

47,8 495 (25-35 °C)

Low-porosity media

200-400 (20 °C)

After: EPA (1975) and Metcalf and Eddy (1979)

After: EPA (1975) and Metcalf and Eddy (1979)

Outline of the 3 experimental runs:

Run 1: Waste water containing Ammonium-N, and COD (Chemical oxygen demand) in the influent waste water.

Run 2: Waste water containing Ammonium-N, Nitrate-N and COD in the influent. The Nitrate-N was introduced to the waste water to see if denitrification could proceed. The clinoptilolite media do not bind nitrate-N. Run 3: Waste water containing Ammonium-N without applying COD to the waste water. This should prevent denitrification from proceeding, and the influent ammonium-N should be recovered as nitrate-N.

A pilot-plant to treat, tertiary stage, industrial waste water using clinoptilolite as media, were built, at the Island of Fyn in Denmark.

Results and discussion

As indicated in Chapter 9 clinoptilolite is a natural zeolite, which selectively sorbs NH4+. The ionbinding capacity is 1.3 meq/g media (Jorgensen etal. 1976). The efficiency of fresh support matrix, is therefore high until the ion-exchange capacity is used up. The removal of ammonium from the waste water will thus decline until the introduced nitrifying biomass becomes sufficient to convert all of the influent ammonium.

The step of biomass development is critical, because if the developed ratio of nitrosomonas and nitrobacter is out of balance, breakthrough of nitrite-N (N02") will appear in the waste water and inhibit the development of nitrosemonas. If the biomass concentrations of the two bacteria species are adjusted, the nitrifying efficiency is raised. Figure 5.18 show the removal efficiency during the 10 first days of biomass development, on previously unused clinoptilolite. Because of its ionbinding capacity, clinoptilolite will bind nearly all ammonium in the first few days. After the capacity is used up, a breakthrough of ammonium will appear until the concentration of nitrifying biomass will be able to convert some of the ammonium to nitrate.

When nitrate-N is developed during nitrification and suitable conditions (anoxic and carbon source) exist, denitrifying bacteria will be developed and nitrate can be converted to nitrogen gas.

Figure 5.19 show the first six days running of a 30 mg/l ammonium influent on a clinoptilolite reactor. The first three days of treatment, a breakthrough of nitrate was observed in effluent samples, because of lack of denitrifying bacteria. For first day, 5.2 mg/l of nitrate-N was found. This amount declined the following days, due to the rapid development of denitrifying bacteria.


Experimental RUN 1.

Simultanous nitrification and denitrification (SND) of waste water containing ammonium and a organic source, measured as COD, in the influent waste water, was conducted during RUN 1.

Synthetic waste water

Gas Collection Bottle

20 cm

Figure 5.17 Laboratory reactor used to conduct experimental Runs 1 to 3.

Synthetic waste water

Gas Collection Bottle

20 cm

Figure 5.17 Laboratory reactor used to conduct experimental Runs 1 to 3.

Table 5.14 Loads conducted during Run 1 to 3 applying clinoptilolite as media.

Run 1

Run 2

Run 3

NH/ -N mg/1

30=> 1000


30=> 1000

N03-N mg/1

up to 180

up to 70

COD int. mg/L





7,7 - 7,8

7,7 - 7,8

7,2 - 7,5

Flow L/hours

0,B => 5,3

0,9 => 3,0

1,2 => 5,0

Oxygen conc. mg/1

2 - 3


2 - 3

Reactor media





Grain size mm

2 - 5


2 - 5

Void volume liter




Bed volume liter




Bed/void volume





Average pore










Reactor volume 1




Reactor high




Reactor dia




meter m.



samplingports. mm




Number of samp





SND occurred




0 0

Post a comment