High Performance Nitrogen Removal System using Macro-porous Cellulose Carrier

pp. 159-166 in Traditional Technology for Environmental Conservation and Sustainable Development in the Asian-Pacific Region

Proceedings of the UNESCO - University of Tsukuba International Seminar on Traditional Technology for Environmental Conservation and Sustainable Development in the Asian-Pacific Region, held in Tsukuba Science City, Japan, 11-14 December, 1995.

Editors: Kozo Ishizuka, D. Sc. , Shigeru Hisajima, D. Sc. , Darryl R.J. Macer, Ph.D.


Copyright 1996 Masters Program in Environmental Sciences, University of Tsukuba. Commercial rights are reserved, but this book may be reproduced for limited educational purposes. Published by the Master's Program in Environmental Science and Master's Program in Biosystem Studies, University of Tsukuba, 1996.

Masatoshi Matsumura,
Institute of Applied Biochemistry, University of Tsukuba, JAPAN

Tetsuya Yamamoto, Hiroshi Tsubota, Osafumi Ito, Elvira G. Suto, University of Tsukuba, Japan;
Kazuhiro Shinabe, Kubota Co., Ltd., Japan;
Ryuji Uno, Kimiaki Yasuda, Biomaterial Co., Ltd., Japan


Abstract

Macro-porous cellulose carrier (AQUACEL) was applied for immobilization of nitrifying and denitrifying bacteria to develop a practical nitrogen removal system. The results of nitrification experiments revealed that the particle and pore sizes of the carrier have a significant influence on the nitrification process. When the optimized carrier (particle size of 1 mm and pore size of 500 m) was employed, the complete ammonium oxidation was attained at a high ammonium loading rate of 12 kg-N/m3-carrier/d. For the porous carriers with pore sizes larger than 360 m, oxygen transfer inside the carrier was enhanced by convection. The newly developed reactors using hydrodynamic jet flow and centrifugal force were effective in distributing the floating carriers homogeneously in the reactor. Complete denitrification was confirmed even at a high nitrate loading rate of 12 kg-N/m3-carrier/d. This AQUACEL system is currently being applied effectively to denitrification of actual wastewater discharged from an electroplating factory.

Key words : nitrification, denitrification, porous carrier, immobilization, hydrodynamic jet flow, effectiveness factor

Introduction

Modern society utilizes lakes, marshes and reservoirs as principal and stable resources for its domestic, industrial and agricultural water supply. Rapid expansion of the economic activities and urbanization imposes tremendous pressure on these enclosed water bodies. In spite of introduction of wastewater regulation for large scale factories and domestic wastewater treatment plants the water quality in these water resources becomes worse and worse by eutrophication. The eutrophication in these enclosed water bodies is mainly caused by increasing nutrient wasteloads from many scattered small scale point sources of pollution like domestic household and small factories with a displacement less than 20 m3/day. These small-scale point sources of pollution are not restricted by any wastewater regulation, and the wastewater containing excessive nutrient salts like nitrogen and phosphorus is being discharged without effective treatment.

Elimination of eutrophication requires the development of a compact but highly efficient system of nitrification and denitrification. In the biological nitrogen removal system, the nitrification process generally becomes the limiting step because of the extremely low growth rate of obligate autotrophic nitrifying bacteria. Previous works conducted to improve nitrogen removal rates include the use of immobilized cells accomplished by entrapment. However, in these cases, most of the cells existed only in the narrow region near the surface of the gel beads due to the limited oxygen transfer rate inside them. Consequently, this resulted in the poor volume efficiency of the gel volume. When the entrapment method was applied to immobilization of denitrifying bacteria, floatation of gel beads caused by the evolution of nitrogen gas resulted. The entrapment method also requires specialized equipment and sufficient time to prepare an adequate number of gel beads for wastewater treatment.

The purpose of this research work is to establish a practical system of nitrification and denitrification which is suitable for small scale point sources of pollution. In this study the effectiveness of a positively-charged macro-porous carrier in the immobilization of nitrifying and denitrifying bacteria by adsorption was investigated. This carrier, "AQUACEL", is made of foamed cellulose with continuous macro-pores. This structure is intended to allow convection of the liquid inside the carrier resulting in oxygen transfer rates increase. The increase in oxygen transfer rate leads to high cell concentration inside the carrier. More stable adhesion can be attained as a result of electrostatic forces. To counter the problem of carrier floatation in the denitrification process, a new reactor equipped with a hydrodynamic jet flow drive was developed. This new reactor distributes the floating carriers in the reactor homogeneously without any mechanical damage to the carriers. The performance of this novel denitrification system was evaluated by using an actual wastewater discharged from an electroplating factory.

Materials and methods

Physical properties of porous carriers

The surface of the AQUACEL carrier was treated with polyethyleneimine (PEI) to provide an ion exchange capacity (IEC) of 0.8-1.3 meq/g stabilized by crosslinking. The effective surface area is high (3-7 m2/g by BET adsorption isotherm method), a void fraction of 97%, and an apparent density of about 1.05 kg-wet/dm3. The optimum structure of AQUACEL carrier in cubic form was determined by changing the particle and pore sizes to about 1-5 mm and 100-1260 m, respectively. Cubic PVF (polyvinylformal) carriers with homogenous pore sizes ranging from 60 to 350 m and a particle size of 3 mm were also employed to study the possibility of enhancement of oxygen transfer into the porous carriers by convection.

Reactors for nitrification and denitrification

A laboratory scale airlift reactor which consisted of two compartments: one for aeration and the other for carrier settling, was used for continuous nitrification (Fig. 1). The working volume of the reactor was 2.3 dm3, with air/oxygen enriched air supplied through a sintered glass ball at a volumetric flow rate of 1.8 /min. The carrier volume was 4.4-7.0% of the working volume of reactor.

Two different reactors were used for denitrification. Fig. 2 shows the reactor with a hydrodynamic jet flow drive. The floating carriers were suctioned into a pipe set at the center of the reactor by downward liquid flow which was induced by driving jet flow through an ejector. The suctioned carriers were transferred to the bottom part of the reactor, and then discharged tangentially against the vessel wall. This tangential flow caused rotation of the liquid in the reactor, which enhanced homogenous distribution of carriers. The working volume of this reactor was 12.5 dm3, with a 20% volume fraction of the carriers. This laboratory scale reactor was mainly used for basic research works to determine the optimum operating conditions for denitrification. The other reactor with a working volume of 500 dm3 employed for denitrification of actual wastewater discharged from an electroplating company was a mixing tank equipped with an elbow at the end of the downflow tube (Fig. 3). The floating carriers in this reactor flew into the downflow tube together with the downward liquid flow induced by centrifugal force of the elbow pipe.


T: Feed tank,
W: Water bath,
P: Peristaltic pump,
R: Reactor,
C: Carrier,
F: Flow meter,
E: DO electrode,
D: DO controller,
A: Air pump,
N: Gas cylinder
Fig. 1. Experimental set-up for continuous nitrification.


Working volume 12. 5 dm3
Filling ratio 18.4%
Carrier volume 2.3 dm3

1. Ejector
2. Liquid circulation pump
3. Wire screen
4. DO sensor
5. pH sensor
6. Overflow tube


Fig. 2. Schematic diagram of ejector type bioreactor.


Working volume 500 dm3
Filling ratio 20.0%
Carrier volume 100 dm3

1. Influent
2. Wire screen
3. Mixer
4. Effluent
5. Stabilizer
6. Downflow tube

Fig. 3. Schematic diagram of mixing tank reactor with downflow tube.

Experimental procedures for nitrification and denitrification

The start-up phase of the AQUACEL process was relatively simple. The nitrifying and denitrifying seed bacteria were prepared from a sludge generated from a nitrogen treatment plant. Dried AQUACEL was soaked in sludge suspension for 3 h inside the reactors to immobilize the bacteria. During this time the carriers swelled, and the bacteria were homogeneously trapped in the carriers. The sludge suspension was replaced with fresh synthesized medium containing ammonium or nitrate ion, and then batch cultivated for a few days under aerobic and anaerobic condition for nitrification and denitrification, respectively. The continuous nitrification and denitrification experiments were initiated by pumping synthesized feed medium containing ammonium and nitrate ion into the respective reactors. The feed rate was gradually increased until the maximum nitrogen removal rate was attained.

The basic composition of the synthesized feed medium for nitrification and denitrification are shown in Table 1. Sodium bicarbonate at a concentration of 937.2 mg/l is equivalent to an alkalinity of 534 CaCO3 mg/l and the ratio of alkalinity to ammonium nitrogen (alkalinity/N ratio) was fixed at 14.0 throughout this research work. When the concentration of ammonium ion in feed medium was increased, the concentration of NaHCO3 was also increased according to this ratio to avoid alkali deficiency. Denitrification requires several trace elements and a carbon source. In this work, four different carbon sources, namely, acetate, ethanol, methanol and isopropanol were used as electron donor for the denitrification. The concentrations of these carbon sources were fixed at 1.5 times the theoretical requirement calculated from the stoichiometric relationship between reduction of nitrate and the hydrolysis of these carbon sources. To avoid contamination of the feed medium inside the storage tank, the carbon source and the salt solution were separately supplied. These solutions were fed into the reactor at the same flow rate.


Table 1. Basic composition of basal media (synthetic wastewater).

Fig. 4. Experimental set-up for measuring effectiveness factor of immobilized cells.

Determination of effectiveness factor

The experimental set-up for the measurement of the effectiveness factor for the immobilized cells is shown in Fig.4. The air-saturated medium for nitrification was continuously fed from the medium storage tank (T) into the vessel (V) through the peristaltic pump (P). The carriers immobilizing cells and the medium in the vessel were stirred by a magnetic stirrer, and the DO concentration in the vessel was measured with a galvanic DO electrode (E). The oxygen uptake rate of the immobilized can be expressed as:

where VL= volume of the medium in the vessel, CL=DO concentration in the vessel, CL*=DO concentration in the inlet medium, F=flow rate of the medium, R=distance from the center of carrier to the surface, n=number of the carriers, De=effective oxygen diffusion coefficient in the carrier, CA=local DO concentration inside the carrier, and r=distance from the center of the carrier.

(NOTE EQUATIONS ARE NOT ON-LINE HERE)
At a steady state, VL(dCL/dt)=0, therefore

The effectiveness factor of the immobilized cells represents the ratio of oxygen uptake rate by the immobilized cells to that obtainable when there is no intraparticle resistance. This is represented as:

where CAS=DO concentration at the surface of the carriers, Vm=maximum oxygen uptake rate, and Km=Michaelis constant. From Eqs. 2 and 3, the effectiveness factor of the immobilized cells can be rewritten as:

where CL=CAS since the partition coefficient of oxygen was assumed to be 1.0. The effectiveness factor of the immobilized cells was thus calculated from

Analyical methods

Ammonia was measured by colorimetric methods using an Ammonia Test Kit (Wako Pure Chemicals, Japan) based on indophenol reaction. Nitrite and nitrate were measured with an ion chromatograph analyzer (Yokokawa, Model IC 100, Japan). Analysis of cations was done using a Jarrel ash inductively-coupled argon plasma analyzer (Model 975, USA). The number of nitrifiers and denitrifiers was measured by MPN (most probable number) method.

Results and Discussion

Start-up and stability of AQUACEL system

A long term nitrification experiment with the AQUACEL carrier (particle size 3 mm, pore size 500 m, IEC 1.13 meq/g) which lasted for more than eighteen months was successfully conducted without any degradation of the carriers. The hydraulic retention time (HRT: 2.5-10 h), alkalinity (CaCO3 130-1340 g/dm3), dissolved oxygen (DO: 1.3 - 7.5 g/m3), and NH4-N influent concentration (40-60 g/m3) were varied. Figure 5 shows the time course of NH4-N concentration in the early phase after the start-up of continuous nitrification. Twenty days after the start-up, the inlet ammonium ions of 50 g/m3 was completely oxidized under a HRT of 5 h, and the ammonium oxidation rate attained the maximum value of 14 kg-N/m3-carrier within 60 days. After confirming the maximum ammonium oxidation rate, the alkalinity and DO were varied to investigate their effects on ammonium oxidation rate.

Comparing the results obtained with cells entrapped in polyethylene glycol cubes (PEG), the effect of DO on ammonium oxidation rate is shown in Fig. 6. In both cases, ammonium oxidation rate was obviously influenced by DO. It increased with increasing DO until it reaches a point where it levels off in spite of DO supply. The maximum ammonium oxidation rate of AQUACEL immobilized cells was more than twice that of cells entrapped in polyethylene cubes (PEG).


Fig. 5. Time course of NH4-N concentration and pH in the early phase of long-term nitrification experiment using AQUACEL.

Temp. : 30 C
Ammonium loading rate : 13.2 kg-
N/m3-carrier/d
Alkalinity : 400 CaCO3 mg/L
Carrier
Size : 3x3x3 mm, Shape : Cubic
Mean pore size : 500 m
Ion Exchange Capacity : 1.13 meq/g

Fig. 6. Effect of dissolved oxygen concentration on ammonium oxidation rate.

Effect of particle size on maximum ammonium loading rate

The maximum ammonium loading rates which allow the complete ammonium oxidation were measured during continuous nitrification using AQUACEL carriers with a pore size of 100 m and particle sizes of 1, 3, and 5 mm. The continuous operation was started with a NH4-N loading rate of 1 kg-N/m3-carrier/d and was gradually increased up to 25 kg-N/m3-carrier/d by increasing NH4-N influent concentrations up to 170 g/m3 under a constant HRT of 3 h. Loading rates lower than 5 kg-N/m3-carrier/d resulted in no differences in ammonium oxidation rate (Fig. 7). Loading rates higher than 8 kg-N/m3-carrier/d resulted in residual ammonium ion and accumulation of nitrite ion using carriers with particle sizes of 3 and 5 mm. Complete ammonium removal was attained in the 1 mm carrier even at a high loading rate of 12 kg-N/m3-carrier/d. At completion of the experiment, the number of cells was determined using the most probable number (MPN) method. The differences in the concentration of nitrifiers among the carriers used are shown in Table 2. The low cell concentration of nitrite oxidizers in the 3 and 5 mm carriers must be caused by the depletion of oxygen inside the carriers and the growth inhibition caused by high residual ammonia and accumulated nitrite.

Temperature 25 C
Relative alkalinity 14.4 kg-CaCO3 /kg-NH4-N
Carrier pore size 100 m
Fig. 7. Effect of carrier size on ammonium oxidation rate.

Effect of pore size on ammonium oxidation rate

The effect of pore size on ammonium loading rate was measured using AQUACEL carriers with pore sizes ranging from 100 to 1260 m and PVF carriers with pore sizes ranging from 60 to 350 m. No differences in ammonium oxidation rates were observed among the AQUACEL carriers (Fig. 8-A). PVF carriers with a pore size of 60 m showed lower oxidation rates compared to carriers with pore size larger than 100 m even at low ammonium loading rates (about 4 kg-N/m3-carrier/d) (Fig. 8-B). These differences in the ammonium oxidation rates between AQUACEL and PVF carriers probably result from their characteristic pore size distribution. The PVF carrier has a narrow pore size distribution while AQUACEL has a broad one.

Table 2. Cell number of nitrifiers in the carriers.

Fig. 8. Effect of mean pore size on ammonium oxidation rate. A: AQUACEL B: PVF

Using the PVF carriers operated under an ammonium loading rate of 3 kg-N/m3-carrier/d, the effectiveness factors were measured at different rotating speed of magnetic stirrer under a constant DO of 3 mg/l (Fig. 9). More obvious influence of mixing intensity was observed in the effectiveness factor of the carrier with pore size of 360 m. This means that the oxygen transfer by convection may be expected for the carriers with pore size larger than 360 m.

Fig.9. Effect of agitation speed on effectiveness factor.

Application of AQUACEL to denitrification

Through continuous denitrification experiments using the downflow column reactor, suitable electron donors for denitrification with AQUACEL carrier was first selected among methanol, ethanol, isopropanol and acetate. When acetate and ethanol were used as electron donors, the start-up of the process was rapid; however, several problems were encountered. These problems included excessive cell formation and generation of a viscous foam, which made stable and long-term operation difficult. Isopropanol resulted in the accumulation of nitrite in the culture broth. No problems were encountered using methanol so it was selected as the most suitable electron donor for the AQUACEL system. The essential trace metal elements and their minimum concentrations were determined and long term continuous denitrification was conducted by using the reactor shown in Fig 2. In this reactor the distribution of the carriers was highly improved, and complete denitrification was confirmed even at high nitrate loading rate of 12 kg-N/m3--carrier/d (Fig. 10). In this experiment, the feed concentration of nitrate was changed in the region from 20 to 70 mol/m3, and HRT was gradually decreased from 25 to 3.0 h.

Fig.10. Effect of nitrate loading rate on

nitrate reduction rate.

Using the reactor shown in Fig.3, the AQUACEL system was applied for denitrification of actual wastewater discharged from an electroplating factory. This factory discharges industrial wastewater containing nitrate (22-600 g-N/m3) and nitrite (20-400 g-N/m3) with a displacement of 50 m3/d. In spite of the vigorous fluctuation in the influent concentrations of NO3-N and NO2-N, the system was able to keep the effluent NO3-N concentration at lower than 5 g/m3 and completely removed NO2-N (Fig. 11).

Fig.11. Time course of NO3-N and NO2-N concentration during the

continuous denitrification using electroplating wastewater.

Conclusion

The AQUACEL system demonstrates significant improvement in nitrification and denitrification rates, and was confirmed to be more practical compared to the system using entrapped immobilized cells. The experimental results on nitrification showed that oxygen transfer rate continued to limit nitrification performance in AQUACEL carriers. Establishment of an excellent AQUACEL system requires more detailed studies on oxygen transfer into the carriers, and development of a reactor which can induce the forced convection inside the carrier.
Back to contents list
To the next chapter
Back to books published by Eubios Ethics Institute
Back to Eubios Ethics Institute Home Page