PII: S0043-1354(00)00442-5

Wat. Res. Vol. 35, No. 7, pp. 17141722, 2001 # 2001 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0043-1354/01/$ - see front matter

SIMULTANEOUS CARBON AND NITROGEN REMOVAL FROM HIGH STRENGTH DOMESTIC WASTEWATER IN AN AEROBIC RBC BIOFILM
A. B. GUPTA and S. K. GUPTA*
Centre for Environmental Science and Engineering, Indian Institute of Technology, Powai, Mumbai 400 076, India (First received 1 July 2000; accepted in revised form 5 September 2000) Abstract}High strength domestic wastewater discharges after no/partial treatment through sewage treatment plants or septic tank seepage eld systems have resulted in a large build-up of groundwater nitrates in Rajasthan, India. The groundwater table is very deep and nitrate concentrations of 500750 mg/l (113169 as NO N) are commonly found. A novel biolm in a 3-stage lab-scale rotating biological 3 contactor (RBC) was developed by the incorporation of a sulphur oxidising bacterium Thiosphaera pantotropha which exhibited high simultaneous removal of carbon and nitrogen in fully aerobic conditions. T. pantotropha has been shown to be capable of simultaneous heterotrophic nitrication and aerobic denitrication thereby helping the steps of carbon oxidation, nitrication and denitrication to be carried out concurrently. The rst stage having T. pantotropha dominated biolm showed high carbon and NH+-N removal rates of 8.725.9 g COD/m2 d and 0.811.85 g N/m2 d for the corresponding loadings of 4 10.032.0 g COD/m2 d and 1.03.35 g N/m2 d. The ratio of carbon removed to nitrogen removed was close to 12.0. The nitrication rate increased from 0.81 to 1.8 g N/m2 d with the increasing nitrogen loading rates despite a high simultaneous organic loading rate. However, it fell to 1.53 g N/m2 d at a high load of 3.35 g N/m2 d and 32 g COD/m2 d showing a possible inhibition of the process. A simultaneous 4463% removal of nitrogen was also achieved without any signicant NO N or NO N build-up. The second and third 2 3 stages, almost devoid of any organic carbon, acted only as autotrophic nitrication units, converting the + NH4 -N from stage 1 to nitrite and nitrate. Such a system would not need a separate carbon oxidation step to increase nitrication rates and no external carbon source for denitrication. The alkalinity compensation during denitrication for that destroyed in nitrication may also result in a high economy. # 2001 Elsevier Science Ltd. All rights reserved Key words } aerobic denitrication, autotrophic nitrication, carbon removal, biolm, heterotrophic nitrication, high strength domestic sewage, nitrogen removal, Thiosphaera pantotropha

INTRODUCTION

High strength domestic wastewater discharges in certain arid areas of the world may cause an alarming increase in groundwater nitrate levels. Acute water scarcity in such areas results in the generation of a highly concentrated sewage, which may be nally discharged on soil after partial treatment centrally or through individual units. Not many of the treatment plants are designed to remove nitrogen from sewage and the euent is normally utilised for surface irrigation. The ammonical and organic forms of nitrogen present in the wastewater get oxidised in upper soil zones which allow sucient oxygen transfer from atmosphere and join the groundwater as nitrates. The disposal of domestic wastewaters

*Author to whom all correspondence should be addressed. Tel.: +91-22-5767853; fax: +91-22-572-3480; e-mail: skgupta@cc.iitb.ernet.in

in areas not served by sewer systems is almost exclusively by use of septic tank and seepage elds. Euents from septic tanks generally contain 5070 mg N/l with about 75% of the nitrogen present as ammonium and 25% as organic nitrogen. The euent from septic tank is usually discharged to aerobic seepage elds, where ammonium and organic nitrogen are transformed to nitrate which may be transported to groundwater (Walker et al., 1973; Starr and Sawhney, 1980). Abeliovich (1987) indicated high COD ($1000 mg/L) and high NH -N ($75 mg/l) concentrations in raw 4 sewage and in the stored domestic wastewater reservoirs in Israel used for summer irrigation. In Jaipur (Rajasthan, India) the raw sewage received at the activated sludge plant has a BOD of 600800 mg/l and NH -N concentration of 80 110 mg/l during 4 summers when the water shortage in acute (personal communication, PHED Rajasthan). The treated wastewater having a high nitrogen concentration is

1714

Removal of C&N from High Strength Domestic Waste

1715

discharged into a lake resulting in steep rise in groundwater nitrates in the vicinity. In a study, over 40% well-water samples were found to contain more than 11.3 mg/l NO -N, a limit set by the WHO for 3 drinking water (Mathur and Kumar, 1990). The situation in the whole arid and semi-arid region of Rajasthan is alarming. The Central Groundwater Board published a nitrate contour map of Rajasthan which indicates many belts encompassed by 750, 600 and 500 mg/l nitrate concentration contours (Gupta, 1988). Because nitrate contamination of groundwater is generally pandemic, the costs associated with remediating groundwater are high and the risk to human health especially of infant methaemoglobinaemia is involved, the matter has to be looked into seriously (Shuval and Gruner, 1977; Bouchard et al. 1992). Biological nitricationdenitrication is one of the most economical processes for nitrogen removal from municipal wastewaters. RBCs have been tried extensively for single-stage carbon removal and nitrication as well as for separate stages in series for BOD removal and nitrication from municipal wastewaters (Weng and Molof, 1974; Stover and Kincannon, 1975; Murphy et al., 1977). Eects of various operating parameters like turbulence (Kugaprasatham et al., 1992), disc rotation speed (Friedman et al., 1979), hydraulic conditions (Kugaprasatham et al., 1991), organic particulate matter (Figueroa and Silverstein, 1992), and recirculation (Klees and Silverstein, 1992) on nitrifying biolms have been studied in details. Some literature is also available on simultaneous nitrication and denitrication in micro-aerobic lms (Masuda et al., 1991; Watanabe et al., 1992) but only one report on a simultaneous organics and nitrogen removal in a fully aerobic RBC biolm is available (Gupta et al., 1994). This paper describes the use of an aerobic RBC system for simultaneous carbon removal, nitrication and denitrication using Thiosphaera pantotropha, a heterotrophic nitrier and aerobic denitrier in mixed bacterial biolm. Isolated during studies on a desulphurizing, denitrifying euent treatment system, T. pantotropha was found capable of myxotrophic and heterotrophic growth on a wide range of substrates (Robertson and Kuenen, 1983). Ample literature has been published, after the isolation of this bacterium by Robertson and Kuenen (1983), on its peculiar enzyme system to explain the properties of heterotrophic nitrication (Robertson et al., 1988a,b, 1989) and aerobic denitrication (Robertson and Kuenen, 1984a,b, 1988a; Bell et al., 1990; Bell and Ferguson, 1991; Berks et al., 1993; Moir et al., 1993). Two important points to note were: (i) The specic nitrifying activity of heterotrophic nitriers is reported to be 103104 times lower than that of autotrophs. However, the ammo-

nia-oxidising rates of T. pantotropha are only 10103 times lower than the autotrophs (Geraats et al., 1990). While growing as heterotroph the growth rate of it tends to be much higher than those for the autotrophs (the mmax for Nitrosomonas europaea is about 0.030.05/h, and that of T. pantotropha can be as high as 0.4/h under same growth conditions) giving it a competitive advantage (Robertson et al., 1988b). (ii) The aerobic denitrication rates were much higher than heterotrophic nitrication rates in chemostat studies with axenic cultures of T. pantotropha at all dilution rates indicating extra capacity of this bacterium to take nitrate or nitrite coming from other routes apart from its nitrication path (Robertson et al., 1988b). The above considerations may be utilised in a mixed culture having autotrophic nitriers and T. pantotropha along with other heterotrophs to evolve an optimum carbon and nitrogen removal system. Reports of the use of this bacterium in suspended and immobilised forms in axenic cultures for carbon removal, nitrication, and denitrication are available (Robertson et al., 1988b; Geraats et al., 1990; Hoojimans et al., 1990). van Neil et al. (1993) studied the competition between a heterotrophic nitrier T. pantotropha and an autotrophic nitrier Nitrosomonas europaea for ammonia in chemostat cultures. They varied the C : N ratio between 1.9 and 10.4 and found that a higher C : N ratio favoured the growth of T. pantotropha, but it was able to outcompete N. europaea for ammonia only at a value of 10.4. At DO below 10 mM, the autotroph became oxygen limited and the heterotroph dominated the culture. When the dilution rate increased from 0.04 to 0.067/h, N. europaea could not maintain itself successfully in the chemostat. The only work on mixed cultures has been reported for a highstrength fertilizer industry waste water (Gupta et al., 1994).

MATERIALS AND METHODS

The experimental set-up A three-stage RBC with acrylic circular plates was employed. The details of RBC are given in Table 1 while Fig. 1 shows a sketch of the set-up. Composition of synthetic media Two media concentrations were used for dierent experimental runs. The rst medium containing 1000 mg/l COD and 110 mg/l NH -N, had two variations in carbon 4 source. The rst (sewage 1a) had a combination of sodium acetate and glucose while the second (sewage 1b) had only sodium acetate as the carbon source. The second concentration containing 500 mg/l COD and 55 mg/l NH -N 4 (sewage 2), had only sodium acetate as the carbon source. The composition of dierent media is shown in Table 2, while Table 3 gives the average characteristics of these media.

1716 Development of biolm

A. B. Gupta, S. K. Gupta Theoretical determination of T. pantotropha and Nitrosomonas biomasses Presence of T. pantotropha in biolm was conrmed by streaking the lm on agar plates of medium used by Robertson and Kuenen (1983) for growing T. pantotropha. Few randomly picked colonies were subjected to biochemical tests specic to T. pantotropha (Robertson and Kuenen, 1983). Scanning electron micrographs also showed the presence of chains of cocci similar to those observed with T. pantotropha axenic cultures. However, it was dicult to quantify the T. pantotropha biomass in the lm. A theoretical procedure was adopted to assess the biomasses of T. pantotropha and Nitrosomonas in the rst stage of RBC by using the following steps:

T. pantotropha pure culture was obtained from Dr. L.A. Robertson (Department of Microbiology, University of Delft, Netherlands). It was grown in liquid medium having composition of sewage 1a. The trough of RBC was inoculated with this culture and lled up completely with the same synthetic sewage. The shaft of RBC was rotated at 5 rpm. Initially about 20% wastewater was replaced daily with the fresh wastewater which was increased to 50% replacement after 5 days. A thin microbial lm was formed on the plates in 1 week. After 15 days, continuous feeding was started at an HRT of 2 days. After another week it was shifted to an HRT of one day. The reactor was operated at three dierent HRTs before switching over to sewage 1b and sewage 2 successively. For each medium composition observations were taken at three dierent HRTs.

(i)

Sampling and analysis Euent samples were collected from the sampling ports of the three stages, provided at the bottom of the trough. These were analysed for various parameters. All parameters were measured on alternate days except for pH and DO which were monitored daily. Ammonianitrogen was measured by Nesslerisation, nitrate nitrogen by brucine sulphate, nitritenitrogen by diazotisation method, and COD by closed reux method as given in Standard Methods (APHA, 1989). DO was measured by YSI DO probe, pH by pH meter and alkalinity by titrating against a standard HCl solution. Suspended solids in the euent were also measured once in 4 days.

Since DO in the rst stage was always greater than 1.5 mg/l, all observed denitrication was attributed to the aerobic denitrication activity of T. pantotropha. (ii) The ammonical nitrogen loss due to assimilation, stripping etc. was neglected. This assumption is justied as Gonenc and Harremoes (1985) while working on simultaneous carbon removal and nitrication indicated that an average of 8% of total nitrogen did not appear in the euent. In an experiment where T. pantotropha and Nitrosomonas were cocultured in a chemostat, the maximum ammonium removal by assimilation at an inuent C : N ratio of 3.7 was 15% (Robertson and Kuenen, 1992). (iii) All unit activities of T. pantotropha and Nitrosomonas were taken corresponding to a

Table 1. Summary of the dimensions of RBC experimental unit Parameter No. of stages No. of disks/stage Disk diameter Disk thickness Disk spacing Shaft diameter Eective surface area/stage Total eective surface area of the unit Eective trough capacity/stage Total trough capacity/stage Submergence Rotational speed Units Number Number cm cm cm cm m2 m2 l l % Rpm Value Table 3 14 25.0 0.2 1.5 5.0 1.2 3.6 8 24 32 5 2. Composition of dierent synthetic concentrations studied wastewater

Compound

Concentration in (mg/l) Sewage 1a Sewage 1b 450 } 2100 300 150 100 Sewage 2 225 } 1050 150 75 50

NH4Cl Glucose CH3COONa3 H2O Na2HPO4 KH2PO4 MgSO4

450 625 708 300 150 100

Fig. 1. Schematic diagram of the experimental setup.

Removal of C&N from High Strength Domestic Waste

1717

C : N ratio of 3.7 for an experiment involving co-culture of T. pantotropha and Nitrosomonas (van Niel et al., 1993), and aerobic denitrication rates were corrected for the observed DO values.

Table 3. Wastewater characteristicsa Compound Sewage 1a COD NH -N 4 C/N Total PO3 P 4 PH Alkalinity (as CaCO3)
a

Value Sewage 1b 1000 110 3.32 102 7.4 600 Sewage 2 500 55 3.32 51 7.2 400

1000 110 3.32 102 7.4 300

All values are in mg/l except pH.

(iv) The heterotrophic nitrication rate of T. pantotropha was 2.6 n mol N/1010 cells/min which was equivalent to 0.0765 mg N/d/mg biomass, as 1 mg of T. pantotropha biomass has been shown to be equal to 1.46 1010 cells. (v) The aerobic denitrication rate of T. pantotropha was 3.4 n mol N/1010 cells/min which was equivalent to 0.1 mg N/d/mg biomass at a DO level of 3.8 mg/l. This value was increased by 10% as the observed DO levels in the rst stage were in a range of 1.52.5 mg/l and aerobic denitrication activity increases as DO decreases (Robertson et al., 1988b). (vi) The autotrophic nitrication rate of Nitrosomonas was 610 n mol N/1010 cells/min which was equivalent to 12.17 mg N/d/mg biomass, as 1 mg of Nitrosomonas biomass has been shown to be equal to 0.Tab99 1010 cells.

Table 4. Average monitored pseudo steady-state values of various inuent and euent parameters for sewage 1a and 1b Sample Sewage 1a I E1 E2 E3 I E1 E2 E3 I E1 E2 E3 Sewage 1b I E1 E2 E3 I E1 E2 E3 I E1 E2 E3 HRT (h) COD (mg/l) NH -N (mg/l) 4 NO -N (mg/l) 2 NO -N (mg/l) 3 pH Alkalinity (mg/l) SS (mg/l) Temp. (8C)

24

18

15

1000 510 510 510 1000 510 510 510 1000 510 510 510 1000 30 510 510 1000 120 100 10 1000 190 120 120

105 38 34 30 105 36 28 25 105 40 38 36 111 30 3 0.6 105 35 11 2 109 61 27 17

0.40 0.05 0.03 0.60 0.14 0.00 0.9 0.6 0.10

3.0 7.0 11.0 0.00 7.0 9.5 3.0 6.0 8.5

7.0 6.8 6.7 7.5 6.9 6.6 6.4 7.4 6.9 6.8 6.7 7.4 7.9 7.5 7.2 7.25 8.32 8.20 8.28 7.35 8.19 8.10 7.28

7.4 140 80 70 300 110 80 70 320 140 50 50 640 470 280 230 833 523 428 380 856 666 428 273

28

22 27

24 26

24 27

24

0.21 1.90 1.90 0.80 5.90 4.60 ND 4.80 4.00

0.3 15.0 17.0 1.0 9.0 20.0 ND 17.0 24.0

20 27

18

20 24

15

28

Note: I, Inuent, E1, E2, E3, euent from stages 1, 2 and 3 respectively; ND, not detectable.

Table 5. Average monitored pseudo steady-state values of various inuent and euent parameters for sewage 2 Sample HRT (h) 24 COD (mg/l) 519 82 82 82 519 82 82 82 519 82 82 82 NH -N 4 (mg/l) 54 14 10 5 56 22 21 16 54 30 25 22 NO -N 2 (mg/l) NO -N 3 (mg/l) DO pH (mg/l) 7.16 7.78 7.76 7.68 7.33 8.15 8.12 8.08 7.22 8.05 7.98 7.98 Alkalinity (mg/l) 407 200 197 152 381 271 262 245 366 328 319 290 SS (Mg/l) Temperature (8C) 23

I E1 E2 E3 I E1 E2 E3 I E1 E2 E3

0.29 0.56 0.18 6.66 7.03 8.66 0.06 2.53 5.99

17 18 22 4 4 7 0.3 1 2

2.5 4.0 5.9 1.8 3.8 4.9 1.5 3.5 5.0

18 19

18

24 22

15

10

1718

A. B. Gupta, S. K. Gupta
NAD1, Net overall alkalinity destroyed per unit of NH -N removed from the system. 4 NAD2, Net overall alkalinity destroyed per unit of NH -N removed after correction for alkalinity generated during denitrication assuming that the rate of alkalinity generation during denitrication is half the rate its 4 destruction during nitrication.
b a

Overall% Overall% Total NAD1a euent total NH -N 4 removed nitrogen nitrogen (mg/l) removal

(vii) The total nitrication and denitrication in mg/ d in stage 1 were derived from observation on sewages 1b and 2 at dierent HRTs. (viii) T. pantotropha biomass was calculated by dividing total denitrication by unit denitrication rate of T. pantotropha. (ix) The heterotrophic nitrication contribution in mg/d was obtained by multiplying the T. pantotropha biomass obtained in the previous step by unit heterotrophic nitrication rate of T. pantotropha. (x) Autotrophic nitrication in mg/d was obtained by subtracting heterotrophic nitrication contribution (step (ix)) from total nitrication (step (iii)). (xi) Nitrosomonas biomass was obtained by dividing autotrophic nitrication in mg/d by unit nitrication rate of Nitrosomonas.

Hydraulic loading rate (l/m2 d)

Overal

6.66 8.88 10.66 20.00 26.66 32.00 12.03 12.66 16.90 6.35 7.12 9.75 3.72 4.42 6.37 19.39 27.05 45.43 82.5 74.16 58.22 99.45 97.69 84.07 0.73 0.91 0.97 0.05 0.23 0.32 0.54 0.64 1.07 1.6 1.8 1.5 0.74 0.93 1.16 2.2 2.8 3.3 20.0 26.6 32.0 19.4 23.4 25.9

Stage 1

g COD removed/g NH -N 4 removed in First stage

NAD2b

RESULTS AND DISCUSSION

Table 4 summarises the average pseudo-steadystate values of various parameters for sewages 1a and 1b at dierent HRTs. At each HRT the reactor was run for about 3 weeks. It was observed that the pseudo-steady-state as characterised by consistency in euent parameters was attained within the rst week of the run. It was seen that more than 80% carbon was exhausted in the rst stage though some organics were observed to leak over to second and third stages at higher loading rates. The growth of T. pantotropha was thus limited by the carbon source and carbon played an important role in the rst stage only. It was observed that with sewage 1a the rst stage accounted for a total nitrogen removal of 6065% which could be attributed primarily to simultaneous heterotrophic nitrication and aerobic denitrication capacity of T. pantotropha along with autotrophic nitrication in this stage. Stages 2 and 3 were almost devoid of any carbon source and hence of T. pantotropha, since this organism can survive autotrophically only on reduced sulphur compounds. These stages acted as autotrophic nitrication units which oxidised a part of the remaining ammonical nitrogen to nitrate nitrogen. Nitrication in these stages was probably hampered by low alkalinity in wastewater as no external source for alkalinity was used. It was decided to switch over to sewage 1b with sodium acetate as the only carbon source. This sewage showed higher alkalinity in the inuent by about 300 mg/l which reected in the improved performance of stages 2 and 3. The euent ammonical nitrogen levels reduced signicantly. The biolm in the rst stage was thick, o-white in colour, whereas in the second and third stages it was very thin and brownish.

Table 6. Derived parameters from Tables 4 and 5

Stage 1 Stage 2 Stage 3 Overall

NH -N Loading (g N/m2 d) 4

NH -N Loading (g N/m2 d) 4

Overall

Stage 1

First stage organics (g COD/m2 d)

Removal

Loading

Sewage 1b 24 18 15 Sewage 2 24 18 15

HRT (h)

10.0 13.3 16.0

8.7 11.6 13.9

1.1 1.5 1.7

0.36 0.50 0.58

0.8 0.9 0.8

0.07 0.02 0.13

0.10 0.14 0.10

0.33 0.36 0.35

90.50 71.43 60.01

49.23 43.81 45.17

27.6 31.68 29.81

5.17 3.36 2.33

7.10 4.85 3.74

10.74 12.76 17.21

20.00 26.66 32.00

6.66 8.88 10.6

Removal of C&N from High Strength Domestic Waste

1719

The nitrication rate increased slightly from 1.61 to 1.85 g N/m2 d when the HRT was reduced from 24 to 18 h but decreased at an HRT of 15 h to 1.53 g N/m2 d. This may be due to two reasons: (a) A decrease in the HRT which resulted in a higher organic load the competition shifts in favour of T. pantotropha as against autotrophic nitriers thus reducing the autotrophic nitrication contribution. (b) Initially an increase in organic load accompanied by a decrease in DO increases the heterotrophic nitrication rate of T. pantotropha (Robertson et al., 1988b) which compensates for the reduction in autotrophic nitrication but at very high loading rates the DO probably was too low and the heterotrophic nitrication rate decreased. It was decided to monitor DO in the reactor for further studies. The above observations indicate a tremendous advantage of this system over the conventional set-ups which was possibly due to the introduction of T. pantotropha in the biolm. A high nitrication rate (1.51.8 g/m2 d) was obtained in the rst stage despite a high simultaneous COD loading rate of 2032 g/m2 d whereas Boller et al. (1994) from experiments on a RBC biolm under dierent organic loads, reported that nitrication started only when organic loads were below 15 g COD/m2 d and was fully developed at about 8 g COD/m2 d. Bovendeur et al. (1990) also indicated the nitrication rate to decrease by 0.015 g NH -N=m2 d per g/m2 d of COD 4 removed where the initial nitrication rate at zero COD was 0.575 g/m2 d.

ammonium to nitrite or nitrate which appeared in the nal euent. The NO -N and NO -N concen2 3 tration ranged from 0.04.61 to 8.5824.08 mg/l respectively. The pH was observed to decrease in successive stages due to destruction of alkalinity during nitrication touching a low value of 6.4 with sewage 1a. But after the switch over to sewages 1b and 2, it always remained over 7.0. Euent SS were in a range

Fig. 2. Relation between overall NH -N loading and 4 removal rates.

Another interesting point to note was that a nitrogen removal of about 4463% was observed in the rst stage with almost no nitrite and nitrate accumulation. In a conventional aerobic system designed for carbon removal and nitrication by Gonenc and Harremoes (1985) the nitrogen balance between the inuent and euent indicated that on an average only 8% of inuent nitrogen did not appear in the euent. In the present study the second and third stages were able to nitrify the remaining

Fig. 3. Relation between NH -N loading and removal rates 4 for stage 1.

Table 7. Nitrogen balance across the reactor for sewage 1b & 2 HRT (h) Inuent Sewage 1b 24 18 15 Sewage 2 24 18 15 110.80 104.70 108.75 54.37 56.4 54.4 Total nitrogen (mg/1) Stage 1 30.8 37.5 60.8 31.0 32.8 30.2 Stage 2 20.2 26.5 49.6 28.9 32.9 28.7 Stage 3 19.4 27.0 45.4 27.6 31.7 29.8 % nitrogen lost Stage 1 72.2 64.2 44.1 43.0 41.7 44.4 Overall 82.5 74.1 58.2 46.9 43.8 45.2

1720

A. B. Gupta, S. K. Gupta
Percent AN 60.0 52.2 33.9 31.0 22.7 30.5 390 523 645 1335 1496 1280 Note: Tp, T. pantotropha; Nm, Nitrosomonas, AN, Autotrophic Nitrication, HN, Heterotrophic nitrication. 10.0 13.3 16.0 20.0 26.6 32.0 1.09 1.50 1.74 2.21 2.80 3.25 976 1096 974 1935 2222 1840 562 753 928 1920 2151 1840 0.11 0.11 0.11 0.11 0.11 0.11 5.1 6.8 8.4 17.4 19.5 16.7 586 572 330 599 727 561 0.05 0.05 0.03 0.05 0.06 0.05 106 145 312 356 325 363 40.0 47.8 66.1 69.0 67.3 69.5

(i)

The calculated dry biomass of T. pantrotropha increased from 5.1 to 19.5 g as the organic

Organic loading rate (g COD/m2 d)

Table 8 lists the various values computed for the determination of the biomasses of T. pantotropha and Nitrosomonas in the rst stage biolm. Some of the salient points to be noted were

Nitrogen loading rate (gN/m2 d)

Theoretical determination of T. pantotropha and Nitrosomonas biomasses

Total nitrication (mg/d)

2028 mg/l and the wastewater temperature varied between 24 and 288C. Though there was a substantial nitrogen removal, the total euent nitrogen was quite high (19.345.4 mg/l) and it was felt that a once-through system for this high strength waste would not be sucient. The system was switched over to sewage 2 after working with sewages 1a and 1b at three dierent HRTs each. Table 5 shows the average monitored pseudosteady-state values for sewage 2. At every HRT the constancy in euent parameters was observed in 56 days. The euent nitrogen was still very high (27.631.6 mg/l). The nitrication rates probably suered due to lowering of ambient temperature which resulted in a decrease in the wastewater temperature. Table 6 lists various derived parameters from Tables 4 and 5 for sewages 1b and 2. The organic removal rates in the rst stage ranged from 8.7 to 25.9 g COD/m2 d while the ammonium removal rates varied between 0.81 and 1.85 g NH -N=m2 d. The ratio of COD removed to 4 NH -N removed was close to 12. The overall nitric4 ation rates ranged from 0.320.97 g NH -N=m2 d for 4 the corresponding loading rates of 0.361.16 of NH -N=m2 d. The euent nitrogen was still above 4 25 mg/l and an overall nitrogen removal of 4382% was observed. The alkalinity consumption per unit of NH -N nitried ranged from 2.74 to 9.75. 4 Fig. 2 depicts overall NH -N removal rate vs. 4 loading rate for sewages 1b and 2 which yields a straight line relation. A similar graph is plotted for the rst stage separately which indicates a change in the order of reaction at higher loading rates and a possible substrate inhibition (Fig. 3). Table 7 shows the nitrogen balance across the RBC which indicates that 4172% nitrogen was lost in the rst-stage euent which could primarily be attributed to aerobic denitrication by T. pantotropha. There was a little loss (54%) in the subsequent stages for sewage 2. The nitrogen loss in stages 2 and 3 put together for sewage 1b ranged from 10 to 14% which may be due to the leakage of carbon from rst stage at high loading rates thereby helping T. pantotropha to survive and denitrify. An overall nitrogen removal of 4382% due to aerobic denitrication along with a signicant nitrication in this study establish the superiority of the present system over conventional ones.

Table 8. Theoretical estimation of T. pantotropha and Nitrosomonas biomasses

Total denitrication (mg/d)

Denitrication rate Tp mg N/d/mg biomass

Tp biomass (g)

HN (mg/d)

AN (mg/d)

N m biomass g

(Tp biomass)/N m biomass

Percent HN

Removal of C&N from High Strength Domestic Waste

1721

loading rate increased from 10 to 26.6 g COD/ m2 d while that of Nitrosomonas uctuated in a close range of 0.0270.060 g. (ii) As the organic loading rate increased from 10 to 32 g COD/m2 d, the ratio of T. pantrotropha biomass to Nitrosomonas biomass increased from 106 to 363. (iii) The percent heterotrophic nitrication contribution increased from 40 to 69.5 as the organic loading rate increased from 10 to 32 g COD/ m2 d. This was due to an increase in both T. pantotropha biomass and the ratio of T. pantotropha biomass to Nitrosomonas biomass as the organic loading increased.

CONCLUSIONS

It was possible to achieve over 43% removal of inuent nitrogen from two concentrations of high strength synthetic wastewater along with over 90% COD removal and 60% nitrication. Very high organic removal rates (8.725.9 g COD/m2 d) and high nitrication rates (1.531.85 g NH -N=m2 d) 4 in the rst stage indicated that an increase in rst stage area can result in a highly improved performance of the system for a simultaneous carbon removal, nitrication and denitrication under aerobic conditions by introducing T. pantotropha in a xed biolm.

REFERENCES

Abeliovich A. (1987) Nitrifying bacteria in wastewater reservoirs. Appl. Environ. Microbiol. 53, 754760. APHA (1989) Standard Methods for the Examination of Water and Wastewater, 17th ed. American Public Health Association, New York. Bell L. C. and Ferguson S. J. (1991) Nitric and nitrous oxide reductases are active under aerobic conditions in cells of. Thiosphaera pantotropha. Eur. J. Biochem. 273, 423427. Bell L. C., Richardson D. J. and Ferguson S. J. (1990) Periplasmic and membrane bound nitrate reductases in Thiosphaera pantotropha. FEBS Lett. 265, 8587. Berks B. C., Baratta D., Richardson D. J. and Ferguson S. J. (1993) Purication and characterisation of a nitrous oxide reductase from Thiosphaera pantotropha. Eur. J. Biochem. 272, 467476. Boller M., Gujer W. and Tschui M. (1994) Parameters aecting nitrifying biolm reactors. Water Sci. Technol. 29, 111. Bouchard D. C., Williams M. K. and Surampalli R. Y. (1992) Nitrate contamination of groundwater sources and potential health eect. J. Am. Water Works Assoc. 84, 8590. Bovendeur J., Zwaga A. B., Lobee B. G. J. and Blom J. H. (1990) Fixed biolm reactors in aquacultural water recycle systems: Eect of organic matter elimination on nitrication kinetics. Water Res. 24, 207213. Figueroa L. A. and Silverstein J. (1992) The eect of particulate organic matter on biolm nitrication. Water Environ. Res. 64, 728733. Friedman A. A., Robbins L. E. and Woods R. C. (1979) Eect of disc rotational speed on biological contactor eciency. J. Water Pollut. Control Fed. 51, 26782689.

Geraats S. G. M., Hooijmans C. M., van Niel E. W. J., Robertson L. A., Heijnen J. J., Luyben K. C. A. M. and Kuenen J. G. (1990) The use of a metabolically structured model in the study of growth, nitrication and denitrication by Thiosphaera pantotropha. Biotechnol. Bioengng. 36, 921930. Gonenc I. E. and Harremoes P. (1985) Nitrication in rotating disc systems-I: criteria for transition from oxygen to ammonia rate limitation. Water Res. 19, 11191127. Gupta A. B. (1988) Fluoride and nitrate problems in Rajasthan, P.G. Diploma, Dissertation, University of Rajasthan. Jaipur, India. Gupta S. K., Raja S. and Gupta A. B. (1994) Simultaneous nitrication and denitrication in RBC. Environ. Technol. 15, 143150. Hooijmans C. M., Geraats S. G. M., van Niel E. W. J., Robertson L. A., Heijnen J. J. and Luyben K. C. A. M. (1990) Determination of growth and coupled nitrication/ denitrication by immobilized Thiosphaera pantotropha using measurement and modeling of oxygen proles. Biotechnol. Bioengng. 36, 931939. Klees R. and Silverstein J. (1992) Improved biological nitrication using recirculation in rotating biological contactors. Water Sci. Technol. 26, 545553. Kugaprasatham S., Nagaoka H. and Ohgaki S. (1991) Eect of short-term and long-term changes in hydraulic conditions on nitrifying biolm. Water Sci. Technol. 23, 14871494. Kugaprasatham S., Nagaoka H. and Ohgaki S. (1992) Eect of turbulence on nitrifying biolms at non-limiting substrate conditions. Water Res. 26, 16291638. Masuda S., Watanabe Y. and Ishiguro M. (1991) Biolm properties and simultaneous nitrication and denitrication in aerobic rotating biological contactors. Water Sci. Technol. 23, 13551363. Mathur Y. P. and Kumar P. (1990) Concerns and trends in environmental levels of nitrates. Ind. J. Environ. Health. 32, 97108. Moir J. W. B., Baratta D., Richardson D. J. and Ferguson S. J. (1993) The purication of a cd1}type nitrite reductase from, and the absence of a copper}type nitrite reductase from, the aerobic denitrier Thiosphaera pantotropha; the role of pseudoazurin as an electron donor. Eur. J. Biochem. 212, 377385. Murphy K. L., Sutton P. M., Wilson R. W. and Jank B. E. (1977) Nitrogen control: design considerations for supported growth systems. J. Water Pollut. Control Fed. 49, 549557. Robertson L. A. and Kuenen J. G. (1983) Thiosphaera pantotropha gen. Nov. sp. Nov., a facultatively anaerobic, facultatively autotrophic sulphur bacterium. J. Gen. Microbiol. 129, 28472855. Robertson L. A. and Kuenen J. G. (1984a) Aerobic denitrication: a controversy revived. Arch. Microbiol. 139, 351354. Robertson L. A. and Kuenen J. G. (1984b) Aerobic denitrication}old wine in new bottles?. Antonie van Leeuwenhoek J. Microbiol. Serol. 50, 525544. Robertson L. A. and Kuenen J. G. (1988a) Heterotrophic nitrication in Thiosphaera pantotropha: oxygen uptake and enzyme studies. J. Gen. Microbiol. 134, 857863. Robertson L. A. and Kuenen J. G. (1992) Nitrogen removal from water and waste. In Microbial control of pollution. eds J. C. Fry, G. M. Gadd, R. A. Herbert, C. W. Jones and I. A. Watson-Craik, pp. 227267. Society for General Microbiology Ltd. Robertson L. A., van Niel E. W. J., Torremans R. A. M. and Kuenen J. G. (1988b) Simultaneous nitrication and denitrication in aerobic chemostat cultures of Thiosphaera pantotropha. Appl. Environ. Microbiol. 54, 28122818. Robertson L. A., Cornelisse R. and Kuenen J. G. (1989) The eect of thiosulphate and other inhibitors of

1722

A. B. Gupta, S. K. Gupta autotrophic and heterotrophic nitriers for ammonia in chemostat cultures. 2, 521525. Walker W. G., Bouma J., Keeney D. R. and Olcott P. G. (1973) Nitrogen transformation during subsurface disposal of septic tank euent in sands II Groundwater quality. J. Environ. Qual. 2, 521525. Watanabe Y., Masuda S. and Ishiguro M. (1992) Simultaneous nitrication and denitrication in micro-aerobic biolms. Water Sci. Technol. 46, 511522. Weng C. N. and Molof A. H. (1974) Nitrication in the biological xed lm rotating disc system. J. Water Pollut. Control Fed. 46, 16741685.

autotrophic nitrication on heterotrophic nitriers. Antonie van Leeuwenhoek J. Microbiol. Serol. 56, 301309. Shuval H. I. and Gruner N. (1977) Infant methaemoglobinaemia and other health eects of nitrates in drinking water. Prog. Water Technol. 8, 183193. Starr J. L. and Sawhney B. L. (1980) Movement of nitrogen and Carbon from a septic tank system draineld. Water Air Soil Pollut. 13, 113123. Stover E. L. and Kincannon D. F. (1975) One-step nitrication and carbon removal. Water Sewage Works, 123, 6669. van Neil E. W. J., Arts P. A. M., Wesselink B. J., Robertson L. A. and Kuenen J. G. (1993) Competition between