Indian Journal of Chemical Technology Vol. 8, Janu:Jry 200 I, pp. 62-67

Studies in nitrification of synthetic fertilizer wastewater in an upflow biofilter

S Chandravathanam & D V S Murthy Environmental Engineering Laboratory, Department of Chemical Engineering,

Indian Institute of Technology, Madras, Chennai 600 036, India

Received 31 March 1999; accepted 19 May 2000

The upn ow :Jcratcd biofiltcr was used to invest igate the nitrification performance of fertili zer-industry wastewater. A syntheti c composi ti on of ammoniu m nitroge n concent ration in the range of 500-2500 mg/L was considered as a representative composi ti on of ac tual fertilizer wastewater and was studied at a HRT of 24 h (corresponding appli ed am mon ium load range was 0.5-2.5 kg NH/-N!m3.d). For the innuent concentration increase from 500 to 2500 mg/L/Nl-1/ ­N, the ammonium remova l efficiency variation was observed to be in the range of 81-58 %. The max imum ammoniu m removal efficiency of 8 1 % was observed for the applied ammonium load of 0.5 kg NH/-Nim3/d The relati ve lower effi ciencies were attributed to the probable substrate (ammoni um) and/or product (nitrite) inhi bition. The experiments were conducted in non-backwash conditions of the biofiltcr. The results were found to be satisfactory.

In the interest of environmental protection, wastewater treatment faciliti es today must meet increasi ngly st ringent quality standards, with nearly constan t effi ciency. At the same time, treatment facilities are being install ed closer to the urban areas and sometimes even form part of the city scape. This new trend, often justified on economic grounds, affects the choice of the technology to be used 1• To comply with new effluent discharge standards of 10-20 mg TN/L, as stated by European Directive2

,

nitrogen removal has become increas ingly important. The nitrification req uirements usually lead to a reactor volume four times larger than that for carbon removal on l/. Because of hi gh specific surface area and intensely aerated environment, granular filters provide favourable conditions for autotrophic bacteria attachment and so nitrification becomes independent of sludge age requirement4

• The slow growth rate of nitri fiers is counterbalanced by their attachment on media3 The nitrifying filters were supposed to nitrify and po li sh the res idual organic pollution contained in the secondary purified water5

. Nitrification inhibition \Vas usually not a problem in the treatment of domestic wastewater. In wastewaters with a high unoxidized nitrogen content, such as fertili zer was tewaters, complete nitrification might not occur because of free ammonia or free nitrous acid inhibition. Nitrite oxidizer micro-organisms are more sensitive than the ammonia oxidizers to free ammonia inhi bition, this being the main responsible factor causing the nitrite accumulation6

. In the present study, the suitability of the up flow biofilter system for the

nitrification treatment of fertilizer industrial wastewater was carried-out.

Experimental Procedure

Materials

The schematic diagram of the upflow aerated biofilter fabricated for the present study is shown in Fig. 1. The square shaped biofilter was made up of 4 mrn perspex sheet, of dimensions, length, width and height of 50, 50 and 40 ern respectively. The bottom II ern height of the biofilter was occupied with the water inlet and the air distribution parts. The biofilter was packed randomly with 4000 numbers of polyurethane foam cubes, of 2 ern size (foam porosity: 0.4), to a height of 14 em, between two perspex perforated plates of 3 mm thickness. At the top portion of the biofilter immediately after the packing at 25 em height, a sampling port (I 0) was provided for the treated wastewater. A provision was made at the bottom portion (4) of the biofilter for the entry of process wastewater to be treated. An aquarium aerator of capacity 176 Llh supplied the required air.. The uniform air di stribution was ensured with 6 numbers of stone air distribution-aids. Wastewater was pumped into the reactor by means of a peristaltic pump. The effective volume of the biofilter was 45.4 L with the bed porosity of 0.57. The carbon requirement for cell synthesis and alkalinity requirement for nitrification were met with sodi um bicarbonate.

CHANDRA VATHANAM & MURTHY: NITRIFICATION OF SYNTHETIC FERTILIZER WASTEWATER 63

I. PERSPEX REACTOR 2. PUF PA(Kii'KJ 3. DRAIN ~- WASTEWATER INL ET

5. AIR INLET

6. AIR OISTRIBUTOR 7. AOUAM AERATER

8. PERISTALTIC PUMP 9. FEED TANK

I 0. WASTEWA TER OUTLET

I I . PERFORATED PLATES

Fig. !- Schematic diagram of upnow biofi lter

Start-up operation of the biofilter The biofilter was inoculated by feeding the sewage

water of aeration tank of Nesapakkam Sewage Treatment Plant, Chennai, India, for two weeks, at the rate of 5 Llh, in the batch mode. Later on, the population of nitrifiers was increased by pumping enrichment culture medium of a specified composition as given in Table I, fo r about three weeks at the rate of 1.89 Llh in batch mode. When the enough biomass concentration was observed visibly , and was in conformation with the literature of 30 days of acclimation8

, the acclimation was stopped after fi ve weeks and continuous ex periments were started with the fert ili zer wastewater to be studied. The pH of the influent wastewater was adjusted to 8.0 with sod ium bicarbonate.

Wastewater characteristics A sy nthetic composition of NH/ -N as (NH4) 2S04

in the range of 500-2500 mg/L (based on the composition of fertilizer wastewater\ was studied as the synthetic fertili zer was tewater with no presence of urea, and dissolved solids. The COD was maintained below detectab le limits.

Method The required composition (500-2500 mg/L

NH/ -N) of wastewater was prepared in the feed tank and then it was pumped into the biofi lter at the required flow rate ( 1.89 Llh corresponding to the

Table !-Enrichment culture medium composition

Constituents g/L

(N H4)2S04 2.0 K2HP04 1.0 MgS04 0.5 FeS04 0.4 NaCI 0.4 CaC03 1.0 MgC03 1.0

hydrau lic retention time of 24 h). The wastewater entered the biofilter at its bottom portion, in order to be an up flow type. The samples of treated effl uents were collected from the sampling port of the biofil ter for the estimation of ammonium, nitrite and nitrate nitrogens, di ssolved oxygen and pH as the case may be. All the experiments were conducted in non­backwash conditions of the biofil ter.

Analytical methods All analyses and preservation techniques were in

accordance with those outlined in Standard Methods 10

• The NH/ -N analyses were done with Nesslerization method. The N03- - N was determined by ultra-violet spectrophotometric method. The N02- -

N was estimated by Diazotization method . The DO concentration was measured by Winkler's method.

Results and Discussion Fig. 2 shows the effluent NH/ -N concentration

profi le with time for the five different inlet ammonium nitrogen concentrations viz. , 500, 1000, 1500, 2000 and 2500 mg/L. In each case, three different phases of increasing profile followed by decreasing profile and a subsequent steady stale profile of effl uent ammonium nitrogen concentration can be identified. When the run was started, the available ammon ium concentration was almost nil in the reactor. As time progresses, more and more fresh feed entered the biofilter and more residuals find their way into the effluent. The increasing trend is in accordance with the investigations of high strength synthetic ammonium chloride wastewaters studied by Cecen and Orak6 and Cecen and Gonenc 11

. The decreas ing trend can be explained as fo llows. The supporting medi a used in the present study was a spongy foam material. The residual concentration of ammonium present in the reactor was slowly getting adsorbed on the surface of the foam particles , therefore the residual concentration appearing in the effl uent decreased. In some in vesti gations6

·11

, the supporti ng media used were pal l nn gs and

64 INDIAN J CHEM TECI-INOL., JANUARY 2001

E

u c 0 u

2:

1600.00 ~------------,

1200.00

800.00

400.00

000 100.00 200.00 Time (h)

-'. IClXJ pprn

• lj(tl p'Pfl'

* ZQ)Opp<n

+ UWp P'"

300.00 400.00

Fig. 2-Variation of efn uent ammonium nitrogen concent ration \v ith time for varying inn uen t ammonium nit rogen concent rations (Syste m: Synthetic Ferti li zer Wastewater)

~

"' .s u c

B 2:

M 0 2: ,_. c ~

~

~ ...

800.00 ~---------------,

600.00 -

40000

200.00

lnflumt NH• •. N a:r'IC.

... ICJ:Ci pprn

• 1$Qlpptt~

* ~,,...,

+ bOO pP"'

0 00 -~:---,------,----.-----J 0.00 100.00 200.00

Time (h) 300.00 400.00

Fi g. 3-Variat ion of efn uenl nitrite nitrogen concen trati on with time for varying innuent ammonium nitrogen concentrati ons (System: Synthetic Fertili zer Wastewater)

polyethylene cylinders which were having poli shed surfaces, Jack adsorption property compared to foam and so a si milar decreasing trend was not observed in their in vestigations. Liu and Capdeville 12 reported a similar decreasing trend of effluent ammonium

2~.00 ,----------------------------,

200.00

:::r >:. ~ u 160.00 c 0

2:

,... 0 z: ~

c ~ lnft ud'11 NH~ ·· Nro-;c. ~

c;: ~ • JOOF P'"

.4. lCOOppm

• lii)Jppm

* ~OClOppm

0.00 100.00 200.00 300.00 400.00 Time (h)

Fig. 4-Yariati on of efnuent nitrate nitrogen concentration with time for varying innuenl ammonium nitrogen concentrati ons (System: Synthetic Ferti li zer Wastewater)

ni trogen concentration in water ni tri fication process (using positively charged polystyrene films as the supporting medium) . As the inl et concentrat ion of ammonium nitrogen increases (500-2500 mg/L), the effluent ammonium nitrogen concentrat ion increased (95-1 043 mg/L) and the time to attain the steady-state also increased (70-320 h). Si milar studies6 on ammonium chloride wastewater system had reveal ed that the steady-state effluent ammonium nitrogen concentration of 230 mg/L was attained afterl20 h for the inlet ammonium, nitrogen concentration of 487 mg/L compared to the result of 95 mg/L steady-state effl uent ammonium nitrogen concentration, which was attained in 70 h for the feed ammonium concentration of 500 mg/L in the present study. The higher performance of the present system can be attributed to the high porosity of the supporting material.

Fig. 3 shows the effluent nitrogen concentration profile with time for five different in fl uent ammonium nitrogen concentrations. The increasi ng profile was due to the reason that more and more feed enters the reactor. The decreasing profile of nitrite nitrogen concentration can be attributed to the subsequent nitrate formation and presumed adsorption on foam cubes. The steady-state concentrations of nitrite nitrogen were in the range of 80-395 mg/L. The high concentration of effluent nitrite nitrogen can be attributed to the inhibition of nitrite ox idizers caused by the high applied ammoni tm nitrogen

CHANDRAVATHANAM & MURTHY: NITRIFICATION OF SYNTHETIC FERTILIZER WASTEWATER 65

3.00 -,-----------------,

., "' E z + 'I." 2.00 2:

'"' "' ;

0

E

" 0

E 1.00 E ..

"" ~ E

w

000 --;-1 - --- --.------r-----1

0.00 1.00 2.00 3.00 Appl ied ammonium load (kg NH4 +. Nfm3.d)

Fig. 5- Variation of eliminated ammonium load with applied ammonium load (Sys tem: Synthetic Fertil izer Wastewater)

. .. >'

2000 0<> ln nucnt NJf 4 ... N c.Q, , .:; _ ( m liV'l)

--

Fi g.6-Yari ati on of percentage ammonium removal with innuent ammonium nitrogen concentrati on (System Fertilizer Wastewater)

concentrations. In the study of ammonium sulphate system, Fdz, Polanco et al. 13 observed that, for an increase of feed ammonium concentration from 50 to 150 mg NH/ -N/L, the ammonium oxidizers activity increased (Kmax from 100 to 110 mg NH/ -N/gVAS.h) and the nitrite oxidizers activity decreased (Kmax from 20 to I 0 mg NO-rN/g V AS.h) with the consequent accumulation of nitrite (as shown by the increase of the ratio of Kmax of the ammonium oxidizers from 6 to 10). However in the present study the oxidizers activity was not determined.

Fig. 4 shows the effl uent nitrate nitrogen concentration profile with time. The three regions of decreasing, increasing and steady-state profiles can be seen from the figure. Due to the initi al increase of ammonium and nitri te nitrogen concentrations, the decrease in the concentration of effl uent nitrate nitrogen was as ex pected. In the case of 500 mg/L inlet concentration of ammoni um nitrogen, the

Fig.7-SEM pictures of nitrifying biofilm showing the presence of spherical and rod shaped bacteria

effluent nitrate nitrogen concentration decreased initially as in other cases and attained the steady-state value faster. The increase of the inlet concentration of ammonium nitrogen (500-2500 mg/L) , caused to increase the steady-state concentration of effluent nitrate nitrogen from 65 to 226 mg/L. In the studies6

for applied ammonium nitrogen concentration of 487 mg/L, the steady-state concentration of nitrate nitrogen was 30-50 mg/L at 3.2 mg/L DO, which is very much less than the present value of 65 mg/L which was attained at a DO of 6.4 mg/L and an inl et of 500 mg/L. ammonium nitrogen concentration

The linear variation of ammonium removal load (0.4- 1.5 kg NH/ -N/m3.d) with the applied ammoniu m load (0 .5-2.5 kG NH/-N/m3.d) is shown in Fig. 5.

66 INDIAN J CHEM TECHNOL., JANUARY 2001

The maximum ammonium removal load of 1.5 kg NH/-N!In3.d was obtained for the applied load of 2.5 kG NH/-N/m3.d. This is in accordance with the studies made by Rogalla et a/.2 and Rogall and Bourbigot4 on the full scale sewage treatment plants for the applied ammonium load ranges of 0.1-2.1 and 0.2-2 kG NH/-N/m3.d respectively and Peladan et a/. 8 on the synthetic ammonium ch loride wastewater system for the applied ammonium load range of 0.5-4 kG NH/-N/m3.d.

The percentage ammonium removal is defined as,

% Ammonium removal= Intluent ammonium Effluent ammonium nitrogen concentration -nitrogen concentration

Influent ammonium nitrogen concentration

. . . ( I )

The percentage ammonium removal as defined in Eq. ( I) was plotted against influent NH/-N concentration, and this can be seen from Fig. 6. As the inlet ammonium concentration increases (for the increase or appli ed ammonium load from 0.5-2 .5 kG NH/ ­N/m3.d), the percentage ammonium removal decreased from 81 to 58 %, which is comparable with the studies made by Ch ui et a/.4 on the synthetic ammonium chloride wastewater system, where, for the variation of applied ammonium load from 0.25 to 1.0 kG NH/-N!m3.d, a decrease in ammonium removal efficiency from 86 to 41 % was observed. A similar trend was also observed in the investigation of Kraft and Seyfried 15 on municipal wastewater treatment, where for the applied ammon ium loads of 0.9, 1.0 and 1.32 kG NH/ -N!Jn3.d, a decrease 111

ammonium removal efficiencies as 88,69 and 64 % respectively were reported.

The Scanning Electron Microscope (SEM) study of the sludge sample (Fig. 7) has shown the presence of Bacill us and Cocci type bacteria in the biofilter (structural similarity with the nitrifying bacteria).

Conclusions

(i) An increase of ammonium nitrogen concentration in the inlet of the biofilter (500-2500mg/L, in terms of ammonium load 0.5-2 .5 kG NH/ -

N!Jn3.d) decreased the ammoni um removal efficiency (8 1-58 % ).

(i i) The maximum ammonium removal efficiency of 81 % was obtained for the applied ammonium load of 0.5 kG NH/ -N/m3.d.

(iii) By maintaining a higher level of DO, at nearly same HRT, the possibility of obtaining higher ammonium removal efficiency was observed.

(iv) The relatively lower efficiencies of the present system, can be attributed to the substrate (ammonium) and/or product (n itrite) inhibition and there is a need to dilute higher nitrogen concentration in fertilizer wastewaters to obtain reasonably higher removal efficiencies in the biofi lter.

References

Pujol R, Hamon M, Kandel X & Lemme! H, Water Sci Techno/., 29( 1994) 33.

2 Rogall a F, Lamonche A, Specht W & Kl eiber B, Water Sci Techno/, 29( 1994) 207.

3 PafToni C, Gousailles M, Rogalla F & Gilles P, Water Sci Techno/, 22( 1990) 18 1.

4 Rogalla F & Bourbigot M, Water Sci Techno/, 22( 1990) 273.

5 Vedry B, Paffon i C, Gousailles M & Bernard C, Water Sci Techno/, 29( 1994) 39.

6 Ceeen F & Orak E, J Chem Tech Biotechno/, 65( 1996) 229.

7 Pelezer M J, Can C C S & Klieg N G, Microbiology, 5th edition (New York, Tala McGraw- Hill), 1993.

8 Peladan J G, Lemme! H & Pujol R, Water Sci Tech, 34( 1996) 347.

9 Mahajan S P, Pollution Control in Process Industries , 2nd edition (Tala McGraw-Hill, Delhi) 1990.

10 Standard methods for the Examination of Water and Wastewater, 19th edi ti on, A WWA, WPC F, Washington, D.C, 1995.

II Cecen F & Gonenc I E, Water Em•iron , 67( 1995) 132.

CHANDRAVATHANAM & MURTHY : NITRIFICATION OF SYNTHETIC FERTILIZER WASTEWATER 67

12 Liu Y & Capdcville B, Water Res, 30( 1996) 1645.

13 Fdz-Polanco F, Villaverde S & Garcia P A, Water Sci Techno/, 34( 1996), 37 1.

14 Chu i P C,Terashima Y, Tay J H & Ozaki H. Water Sci Tecllllol, 34( 1996) 187.

15 Kraft A & Seyfried C F, Water Sci Techno/, 23( 1991 ) 1783.