Biological nutrient removal by a sequencing batch reactor (SBR) using an internal organic carbon source in digested piggery wastewater

(1)

Biological nutrient removal by a sequencing batch reactor

(SBR) using an internal organic carbon source in digested

piggery wastewater

D. Obaja, S. Mac

e, J. Mata-Alvarez

*

Departament d’Enginyeria Quımica i Metallurgia, Universitat de Barcelona, Martı i Franques, no. 1, pta. 6, 08028 Barcelona, Spain Received 5 September 2003; accepted 5 March 2004

Available online 16 April 2004

Abstract

Experiments in a lab-scale SBR were conducted to demonstrate the feasibility of using an internal carbon source (non-digested pig manure) for biological nitrogen and phosphorus removal in digested piggery wastewater. The internal C-source used for denitriﬁcation had similar eﬀects to acetate. 99.8% of nitrogen and 97.8% of phosphate were removed in the SBR, from an initial content in the feed of 900 mg/l ammonia and 90 mg/l phosphate.

Keywords: Piggery wastewater treatment; Internal carbon source; Sequencing batch reactor; Nutrient removal; Nitriﬁcation; Denitriﬁcation; Biological phosphorus removal

1. Introduction

Biological denitrification is a reliable method for nitrogen removal from wastewater (Abufayed and Sch-roeder, 1986). Heterotrophic bacteria use the available carbon source. Since nitrified liquor is usually deficient in organic carbon and the low carbon source level limits the biological denitrification process, sufficient organic carbon sources must be provided for proper denitrifi-cation. In addition, for proper biological phosphorus removal, an easily biodegradable carbon source is nee-ded at the P release stage. Effluents from pig manure anaerobic digesters still contain a large amount of COD and a large nutrient load (ammonia and phosphate). Though a small fraction of the COD consists of volatile fatty acids and could be used for denitrification, easily biodegradable COD is clearly needed, given the amounts of nitrogen and phosphorus. Synthetic chemi-cals, such as methanol or acetic acid, are expensive, but are quite effective. However, the use of non-digested pig manure as a source of easily biodegradable carbon is an alternative and cheaper method. The use of wastewater

as an electron donor for denitrification in sequencing batch reactors (SBRs) was suggested by Pallis and Irvine (1985) and implemented for piggery wastewaters by Bortone et al. (1992). This approach has aroused greater interest recently, especially the use of domestic waste-water with the organic fraction of municipal solid waste (OFMSW) as carbon source. For instance, Beccari et al. (1998) studied the potential of readily biodegradable COD obtained from acidogenic fermentation of the OFMSW as electron donor for denitrification. Experi-ments at an SBR pilot plant used domestic wastewater from the wastewater treatment plant at Fusina (Venice). The results showed a remarkable improvement both in denitrification rate and in flexibility of the response to influent load peaks. The main advantage of this ap-proach is economic, as no expensive external carbon source is required. Other studies have followed this idea of using residual effluents as substitutes for chemical carbonaceous sources (e.g. Monteith et al., 1980; Skrinde and Bhagat, 1982; Bernet et al., 1996; Lee et al., 1997; Ra et al., 2000; Graja and Wilderer, 2001; Cervantes et al., 2001).

This study aimed to establish an approach to removing nutrients from digested pig manure in an SBR with internal easily degradable organic matter and to explore the feasibility of this more cost-eﬃcient removal

*

Corresponding author. Tel.: 1305; fax: +34-93-402-1291.

E-mail address:[email protected](J. Mata-Alvarez).

(2)

system. SBRs provide a simple and even economical way of treating piggery wastewaters. SBR systems offer substantial benefits over alternative conventional flow systems, because of their flexibility and capacity to meet various treatment objectives (Norcross, 1992; Ketchum, 1997).

In the experiments, non-digested pig manure was added at the beginning of the denitrification stage, to-gether with acetic acid in different ratios. Reactor per-formance was then studied under three operating conditions (P1, P2 and P3), with different proportions of external (acetic acid) and internal (volatile fatty acids of pig manure) carbon.

The appropriate C/N ratios were established with care, so as to achieve complete denitriﬁcation without any carbon source excess. Thus, both waste of time and oxygen consumption were avoided.

2. Methods

2.1. Substrate and readily biodegradable C-source Raw and digested pig manure was obtained from a piggery at Caldes de Montbui, 30 km from Barcelona. The latter had been treated in an industrial anaerobic digester on the farm. The eﬄuent from the digester was centrifuged (Beckman model J2-21) at 4000 rpm for 15 min at 5 °C, to remove most of the suspended solids. The supernatant was used as substrate for the experi-ments. The non-digested pig manure was used after centrifugation as an easily biodegradable carbon source. Before being fed into the reactor, the supernatant was diluted with tap water to the desired concentration (see Fig. 1 for feed preparation). Table 1 shows the charac-teristics of the supernatant and of the non-digested pig manure.

2.2. Experimental device

The SBR, a cylindrical tank with a volume of 3 l, was made of Pyrex glass and ﬁtted with mixing and air sparging systems (Fig. 2). It was complemented by two peristaltic pumps (Cole Parmer Instruments, Model

number 7553-85, Chicago), one for feeding, the other for drawing oﬀ eﬄuent and excess sludge. The SBR opera-tion cycles were controlled by programmable timers.

A data acquisition system (model CRISON pHrocon 18, Barcelona) was used for the continuous recording of the mixed liquor temperature and of pH. Optimum pH was set at 8.1. Higher pH was corrected by mechanical addition of hydrochloric acid HCl (0.5 N).

Piggery wastewater reservoir NH₄+_-N concentration = 1600 mg/l (Table 1) Feed reservoir NH -N concentration = ₄+ 900 mg/l To be diluted with SBR effluent (here, with tap water)

Fig. 1. Feed preparation: after feed is prepared, it is introduced in the reactor. As there are 2 l remaining from the previous cycle, the initial

concentration of NHþ 4-N is 300 mg/l. Speed Regulator pH meter Timers T1 T2 T3 T4 Influent Effluent Air HCl Stirrer DIFFUSER PT100 pH REACTOR SBR

Fig. 2. Diagram of the SBR. Table 1

Characterisation of the digested and non-digested piggery wastewater supernatant after centrifugation

Parameters Digested piggery wastewater Non-digested piggery wastewater No. of analyses pH 8.42 8.45 – TS (g/l) 11.21 13.42 5 VS (g/l) 5.35 6.23 5 TSS (g/l) 2.58 3.10 5 VSS (g/l) 1.96 2.2 5 VFA (mg/l) 1050 5275 3 BOD5(mg/l) 1730 3250 4 COD (mg/l) 3085 7450 8 Alkal7:74(mg/l) 1183 1450 5 Alkal3:53(mg/l) 5226 6530 5 Total N (mg/l) 1650 785 8 NHþ 4-N (mg/l) 1600 720 6 NO 3-N (mg/l) 0 0 6 NO 2-N (mg/l) 0 0 6 PO3 4 -P (mg/l) 147 120 6

(3)

2.3. Analytical methods

Chemical oxygen demand (COD), biochemical oxy-gen demand (BOD5), alkalinity, total nitrogen (TN) and

solids (total suspended solids (TSS), volatile suspended solids (VSS), total solids (TS) and volatile solids (VS)) were all analysed following standard methods (APHA, 1992).

Nitrogen compounds, nitrates, nitrites and phos-phates were analysed with an ionic chromatograph (KONIK model KNK-500-A Series, column Waters IC. Pak) under the following conditions: automatic injector

(Kontron model HPLC autosampler 465), 80ll volume

sample, conductivity detector (Wescan) for concentra-tions between 2 and 50 mg/l, and UV detector (Kontron HPLC 332) for low concentrations (0.1–2 mg/l).

Ammonium was determined by an ammonia-speciﬁc electrode (Crison, model pH 2002, Barcelona).

Volatile fatty acids (VFA) were analysed by gas chromatography (HP 5890 Series II, ﬂame ionisation detector) under the following conditions: column from 120 to 170°C, injector at 280°C, detector at 300°C and volume sample 1ll. Helium was employed as carrier gas. Samples were centrifuged at 10,000 rpm for 10 min and ﬁltrated through 0.45lm to remove suspended sol-ids prior to being fed to the chromatographic columns.

3. Results and discussion

3.1. Optimisation of the cycle length

In a preliminary study using acetic acid as carbon source (Obaja et al., 2003), a set of optimal conditions

for removing nutrients from digested pig manure was established. This consisted of three cycles of 8 h per day. The maximum initial NHþ₄-N concentration that could be treated in the reactor was 500 mg/l, with a hydraulic retention time (HRT) set at one day. Thus, the total N removed was 1500 mg NHþ₄-N/l day.

However, later optimisation experiments (Obaja, 2002) improved removal eﬃciency. In these experiments, several cycle lengths, 4, 8, 12 and 24 h, were tested. In all of them, a constant quantity of eﬄuent was fed per day (1/ 3 of reactor volume). As the number of cycles per day was increased and the quantity of substrate fed per day re-mained constant, the total amount of substrate fed was increased and the HRT was decreased.

The results obtained in these optimisation experi-ments are shown in Table 2 and Fig. 3. As can be seen, the best yields in terms of nitriﬁcation rate were ob-tained in 4-h cycles. With this strategy, the maximum initial NHþ₄-N concentration that could be treated in the reactor was found to be 300 mg/l per cycle with an HRT set at 0.5 days. Thus, the total N removed in 6 cycles/day was 1800 mg NHþ4-N/l day. Therefore, this was the

ini-tial strategy chosen to remove the ammonium and or-ganic matter from the inﬂuent with an internal carbon source for denitriﬁcation.

3.2. Optimisation of the dose of easily degradable carbon Reactor performance was studied under three oper-ating conditions (experiments P1, P2 and P3), of which external (acetic acid) and internal (volatile fatty acids of pig manure) carbon made up, respectively, 75% and 25%, 50% and 50%, and 0% and 100% (% expressed in terms of mg of acetic acid). All these experiments were Table 2

Inﬂuence of the cycle length Cycle’s length (h) Initial NHþ 4-N concentration in reactor (mg/l) Average nitriﬁcation rate (mg NHþ 4-N/l h) NHþ 4-N removal (%) NO x-N removal (%) PO3 4 -P removal (%) Maximum N removal (mg N-NHþ 4-N/day l) HRT (days) 4 300 150 99.6 100 98.4 1800 0.5 8 500 119.4 99.7 100 97.3 1500 1 12 700 87 99.8 100 98.6 1400 1.5 24 1300 68 99.6 100 98 1300 3 0 100 200 300 400 0 1 2 3 4 Time (h) Species concentration (mg/l) NH₄-N (mg/L) NO₃-N (mg/L) NO2-N (mg/L) PO4-P (mg/L)

(4)

conducted at 25 °C (estimated as a minimum for the temperature of the digester eﬄuent after centrifugation in a full-scale plant).

Each cycle began with an anaerobic stage of 1 h (the reactor was stirred but not aerated) for phosphorus re-moval. Between the second and third hour of the cycle there was an aerobic stage, with stirring and aeration, for nitrification. The fourth hour was anoxic for deni-trification, adding initially the easily biodegradable carbon source, with stirring but no aeration. The cycle should have finished at this stage. However, as the main effect of adding the internal carbon source was to

in-crease NHþ

4-N concentration in the reactor, the cycle

length had to be extended. In fact, three additional hours were needed (intermittently aerobic and anoxic every 30 min) to remove this nutrient, resulting in an HRT of 0.87 days. The operating conditions of this new cycle length are shown in Table 3.

After 7 h, the sludge was drawn, with stirring and aeration throughout the stage to ensure a homogeneous purge. Thus, SRT (or sludge age) was kept constant and equal to 11 days in all the experiments (Bortone et al., 1994). The solids then settled and liquid was drawn (approximately 1/3 of the reactor supernatant in each cycle including the volume of the purge) in order to maintain the HRT level set. Then a new cycle started. 3.2.1. Period 1

In this period, the internal carbon source was estab-lished as 25% from the VFA supplied by the wastewater and 75% as acetic acid.

During the aerobic stage, NHþ

4-N concentration

de-creased from 300 to 1 mg/l (i.e. 99.7% NHþ₄-N removal). The average nitriﬁcation rate was 149.5 mg/l h, a ﬁgure higher than other authors’ results, which oscillate mainly between 3 and 150 mg/l h (Bortone et al., 1992,

1994; Andreottola et al., 1997; Lee et al., 1997; Su et al., 1997; Tilche et al., 1999, 2001; Edgerton et al., 2000; Ra et al., 2000).

According to Bernet et al. (1996), at the denitriﬁca-tion stage, the C/N ratio needed for a complete removal

of NO

3-N and NO2-N is 1.7. Therefore, at the third

hour of the cycle (anoxic stage), 1130 mg/l of VFA were added (25% coming from non-digested pig manure and 75% acetic acid). Fig. 4A shows the proﬁles of NHþ4-N,

NO

3-N and NO2-N during this cycle.

During this phase, which lasted 1 h, nitrate concen-tration decreased from 257 to 0 mg NO

3-N/l and all the

VFA introduced were consumed. When the 1130 mg/l of VFA were introduced into the reactor, 74 mg/l of NHþ

4

-N from the internal carbon source were introduced at the same time. To remove these, another nitrification stage was added, resulting in a nitrate concentration of 64 mg/l. As this stage had to be shorter, only 1 h was programmed for nitrification, as against the 2 h of the previous stage. For complete denitrification, 153 mg/l of VFA were necessary. Following the same strategy for supplying easily degradable carbon, 35 mg/l of NHþ4-N

were introduced at the same time, which again was re-moved through a new nitriﬁcation stage that produced

31 mg/l of NO

3-N. By this method, following the cycles

consisting of successive phases of nitriﬁcation–denitriﬁ-cation, all the nutrients were completely removed. Easily degradable C-source was added six times during the entire cycle. Fig. 4A and B show the complete set of stages (7 h overall) required for a complete removal of nitrogen, using 25% of VFA coming from non-digested pig manure.

The ﬁrst stage of phosphate removal, during the ﬁrst hour of the cycle, saw an increase in phosphate con-centration from 29.2 to 31.9 mg PO3

4 -P/l. During the

two hours of the aerobic stage, this concentration de-creased from 31.9 to 2.9 mg/l: 91% phosphate removal (see Fig. 4C). The value of the P/C ratio was 0.045. 3.2.2. Periods 2 and 3

All these results for P1 are summarised in Table 4 and Fig. 4A–C. The results for P2 and P3 are also reported in Table 4 and are shown in Figs. 5A–C and 6A–C, respectively.

As Figs. 4–6 and Table 4 show, results were very similar and good in the three periods: nutrient removals ranged from 97.8 to 99.9. In all periods, profiles of ammonia, nitrite and nitrate in the reactor were com-parable. Introduction of easily biodegradable carbon source at the beginning of the anoxic phase led to a high rate of denitrification, with all nitrate removed within the first hour of this phase.

The almost identical performance of 100% internal carbon feeding source and 25% internal/75% external carbon feeding source suggests that the internal carbon source (non-digested pig manure) is a viable choice for Table 3

Operational parameters of the lab-scale SBR used in this study

Parameters (in the reactor) Experiment

P1 P2 P3

Volume (l) 3 3 3

Inﬂuent ﬂow rate (l/day) 3.43 3.43 3.43

Cycle length (h) 7 7 7

Number of cycles per day 3.43 3.43 3.43

Inﬂuent volume per cycle (l) 1 1 1

mg of NHþ

4-N removed per

day and litre of reactor

1028 1028 1028 HRT (day) 0.87 0.87 0.87 SRT (day) 11 11 11 pH <8.1 <8.1 <8.1 Temperature (°C) 30 30 30 % of internal C-sourcea ₂₅ ₇₅ ₁₀₀ Number of easily biodegradable C-source additions during the cycle

6 6 6

(5)

enhancing SBR performance. The extra nitrogen added by the internal carbon source oﬀers no impediment to either nitrogen or phosphorus removal. Comparison of

these results with those obtained in the operation with-out internal C-source (Table 4) shows that results are very similar in terms of removal yields, but are better

0 50 100 150 200 250 300 0 1 2 3 4 5 6 7 Time (h) S pec ie s c onc ent ra ti on (m g/ L) NH4-N NO3-N NO2-N 0 10 20 30 40 0 1 2 3 4 5 6 7 Time (h) S pec ie s c onc ent ra ti on (m g/ L) S pec ie s c onc ent ra ti on (m g/ L) PO4-P 0 500 1000 1500 0 1 2 3 4 5 6 7 Time (h) VFA (A) (B) (C)

Fig. 4. Proﬁles of the concentrations of the diﬀerent species of N and P in a single cycle of the SBR operation during period 1.

Table 4

Results of the three periods studied and of a previous experiment with acetic acid as carbon source

Period 1 Period 2 Period 3 Previous operation without

internal C-source Nitrification Influent NHþ 4-N concentration (mg/l) 300 301 300 300 Effluent NHþ 4-N concentration (mg/l) 1 3 3 4.3

Average nitriﬁcation rate (mg/l h) 149.5 149 149.5 74.6

% Removal 99.9 99.8 99.8 99.7 Denitriﬁcation Initial NO 3-N concentration (mg/l) 262 260 257 264 Final NO 3-N concentration (mg/l) 0 0 0 0 Consumed VFA (mg/l) 447 442 437 450

% of VFA coming from the non-digested pig manure 25 75 100 0

Phosphorus removal

PO3

4 -P concentration (mg/l) at the beginning of the

anaerobic stage

29.2 28.3 28 29.6

PO3

4 -P concentration (mg/l) at the end of the anaerobic

stage

31.9 30.2 30.1 41.2

PO3

4 -P concentration (mg/l) at the end of the aerobic stage 2.9 3.2 3.2 3.2

% Removal 98.1 97.8 97.8 97.8

(6)

with an internal carbon source, as the average nitriﬁ-cation rate obtained is around double.

Good nutrient removal results are not surprising, since acetate constitutes 50% of products from the fer-mentation of pig manure (Lee et al., 1997) and is rapidly consumed and degraded by heterotrophic bacteria.

However, it has to be stressed that, in all the periods using internal carbon source, three additional hours were needed to remove the added nutrients. In conse-quence, the length of the cycle and thus the HRT need to be extended considerably.

3.3. Batch experiments

In each period studied, the biomass in the activated sludge was characterised by the determination of ammonia utilization rate (AUR), nitrogen utilization rate (NUR) and oxygen utilization rate (OUR). The AUR test can also be used to identify possible inhibitory eﬀects from wastewater.

AUR, NUR and OUR increase when the percentage of internal carbon source increases. In fact, when the % of non-digested pig manure is higher, more nutrients are added, which gives higher rates of AUR, NUR and OUR because of the greater activity of the viable bio-mass (Table 5). Higher AUR, NUR and OUR show

that functional groups in activated sludges are sensitive to the speciﬁc characteristics of their substrate.

Table 5 also compares the average nutrient removal rate, the AUR, the NUR and the OUR of the experi-ment with an internal carbon source, and the values obtained in a previous experiment (Obaja et al., 2003), in which acetic acid was the sole carbon source. Using

the same concentration of NHþ

4-N, AUR, NUR and

OUR were higher in experiments with non-digested pig manure used as internal C-source than in experiments with acetic acid. Thus, the type and dose of organic carbon source aﬀect the type of bacteria that develop and the organism growth rate, making the viable bio-mass more active (Tam et al., 1994).

3.4. Economic aspects

Comparison of experiments with internal and exter-nal carbon sources for denitriﬁcation makes it clear that the advantage of an internal carbon source is the saving from not having to use an expensive easily biodegrad-able carbon source. However, there is another factor to take into account, and this has a negative eﬀect on the economics of this approach.

As Table 3 shows, in experiments using the internal

C-source, 1028 mg of NHþ4-N per day and litre of

0 50 100 150 200 250 300 0 1 2 3 4 5 6 7 Time (h) S pec ie s c onc ent ra ti on (m g/ L) S pec ie s c onc ent ra ti on (m g/ L) S pec ie s c onc ent ra ti on (m g/ L) NH4-N NO3-N NO2-N 0 10 20 30 40 0 1 2 3 4 5 6 7 Time (h) PO4-P 0 200 400 600 800 1000 1200 0 1 2 3 4 5 6 7 Time (h) VFA (A) (B) (C)

(7)

reactor can be removed in 3.43 cycles (7 h long), feeding 1/3 of the reactor volume in each cycle. However, a much better yield in terms of N removal per volume of reactor is obtained with acetic acid used as carbon source. With this latter chemical, 1800 mg of NHþ

4-N

per day and litre of reactor can be removed, using six cycles per day (4 h long) (see Table 2). This means that, if the internal C-source is used, a larger amount (75% more) is needed to remove the same amount of NHþ

4-N.

In addition, if fresh pig manure is to be used for the

substrate needed for denitriﬁcation, equipment to pre-pare this substrate (ﬁlter, pump, etc.) should also be provided. In consequence, higher investment is needed, if no raw acetic acid (or methanol) is used for the anoxic step. The two aspects need to be balanced, though the high prices of chemicals should make the period needed to recover the additional investment short. The resource conservation implicit in the use of an internal C-source is also a factor––not an economic one, but a question of sustainability––that should be taken into account. Table 5

Average of nutrient removal rates, values of AUR, NUR and OUR, and percentage of nutrient removals obtained during the three periods

Parameters Units P1 P2 P3 Acetic acid as carbon

source (Obaja et al., 2003)

Ammonium removal rate mg NHþ

4-N g VSS

1_h1 _57.7 _62.5 _68.2 _24.3

Nitrate removal rate mg NO

3-N g VSS

1_h1 _98.8 _107.0 _116.8 _105.6

Phosphate removal rate mg PO3

4 -P g VSS 1_h1 _5.2 _5.6 _6.1 _3.9 AUR mg NHþ 4-N g VSS 1_h1 _31.6 _32.7 _34.2 _27.5 NUR mg NO 3-N g VSS 1_h1 _38.7 _40.1 _42.5 _31.1 OUR mg O2g VSS1min1 3.5 3.6 3.7 1.7 Ammonium removal % 99.9 99.8 99.8 99.7 Nitrate removal % 99.9 99.8 99.9 99.9 Phosphate removal % 98.1 97.8 97.8 97.8 0 50 100 150 200 250 300 0 1 2 3 4 5 6 7 Time (h) Species concentration (mg/L) NH₄-N NO3-N NO2-N 0 5 10 15 20 25 30 35 0 1 2 3 4 5 6 7 Time (h) Species concentration (mg/L) PO4-P 0 200 400 600 800 1000 1200 0 1 2 3 4 5 6 7 Time (h) Species concentration (mg/L) VFA (A) (B) (C)

(8)

4. Conclusions

The sequencing batch reactor is an eﬃcient tool for biological carbon and nutrient removal, capable of achieving eﬄuents with very low nitrogen and phos-phorus concentrations from highly concentrated waste-waters. With digested piggery wastewater, nitrogen and phosphorus removal yields are around 100% and 98%, respectively.

The time cycle and the duration of each single phase within the time cycle must be designed properly, in order to optimise the removal of nutrients. The choice of hydraulic residence times and cellular retention times will depend on this optimisation.

Denitrification can be seen not only as a way of removing nitrogen pollution, but also as an efficient method of removing organic carbon. Complete denitri-fication was obtained when the C/N ratio was equal to or higher than 1.7.

The experiments carried out demonstrated the feasi-bility of using non-digested pig manure as an easily biodegradable carbon source for denitriﬁcation and dephosphatation. The process is as simple as when acetic acid or methanol is used. Indeed, the same excellent nutrient removal is obtained, and the additional carbon

and NHþ

4-N introduced can also be removed.

The denitrification rate is affected by the type and dose of organic carbon source used. Higher mean re-moval rates are obtained with an internal carbon source. This can be explained by the fact that different types of organic carbon sources influence the type of bacteria that develop, organism growth rate, nitrate reduction and the degree of accumulation of intermediate by-products (Tam et al., 1992, 1994).

The main advantage of using an internal carbon source is the saving in chemicals. This has a very positive eﬀect on the plant’s operating costs. However, an internal carbon source needs higher investment due to the substantial increase in volume needed and the nature of the unit for preparing the substrate for denitriﬁcation.

References

Abufayed, A.A., Schroeder, E.D., 1986. Performance of SBR/denitri-ﬁcation with primary sludge carbon source. Journal Water Pollu-tion Control FederaPollu-tion 58, 387–397.

Andreottola, G., Bortone, G., Tilche, A., 1997. Experimental valida-tion of a simulavalida-tion and design model for nitrogen removal in sequencing batch reactors. Water Science and Technology 35 (1), 113–120.

APHA, 1992. Standard Methods for the Examination of Water and Wastewater, 16th ed. American Public Health Association, Wash-ington, DC.

Beccari, M., Cecchi, F., Majone, M., Musacco, A., 1998. Improved biological nutrient removal resulted from the integrated treatment

of wastewater and municipal solid waste. Annali di Chimica 88 (11–12), 773–781.

Bernet, N., Habouzit, F., Moletta, R., 1996. Use of an industrial eﬄuent as a carbon source for denitriﬁcation of a high-strength wastewater. Applied Microbiology and Biotechnology 46, 92–97. Bortone, G., Gemilli, S., Ramabaldi, A., Tilche, A., 1992.

Nitriﬁca-tion, denitriﬁcation and biological phosphate removal in sequenc-ing batch reactors treatsequenc-ing piggery wastewater. Water Science and Technology 26 (5–6), 977–985.

Bortone, G., Malaspina, F., Stante, L., Tilche, A., 1994. Biological nitrogen and phosphorus removal in an anaerobic/anoxic sequenc-ing batch reactor with separate bioﬁlm nitriﬁcation. Water Science and Technology 30 (6), 303–313.

Cervantes, F.J., De la Rosa, D.A., Gomez, J., 2001. Nitrogen removal

from wastewaters at low C/N ratios with ammonium and acetate as electron donors. Bioresource Technology 79, 165–170.

Edgerton, B.D., McNevin, D., Wong, C.H., Menoud, P., Barford, J.P., Mitchell, C.A., 2000. Strategies for dealing with piggery eﬄuent in Australia. Water Science and Technology 41 (1), 123– 126.

Graja, S., Wilderer, P.A., 2001. Characterization and treatment of the liquid eﬄuents from the anaerobic digestion of biogenic solid waste. Water Science and Technology 43 (3), 265–274.

Ketchum Jr., L.H., 1997. Design and physical features of sequencing batch reactors. Water Science and Technology 35 (1), 11–18. Lee, S.-I., Park, J.-H., Ko, K.-B., Koopman, B., 1997. Eﬀect of

fermented swine wastes on biological nutrient removal in sequenc-ing batch reactors. Water Research 31 (7), 1807–1812.

Monteith, H.D., Bridle, T.R., Sutton, P.M., 1980. Industrial waste carbon sources for biological denitriﬁcation. Progress in Water Technology 12, 127–144.

Norcross, K.L., 1992. Sequencing batch reactors––an overview. Water Science and Technology 26, 2523–2526.

Obaja, D., 2002. Pig manure nutrient removal in a SBR. PhD thesis, Faculty of Chemistry, University of Barcelona.

Obaja, D., Mace, S., Costa, J., Sans, C., Mata-Alvarez, J., 2003. Nitriﬁcation, denitrifcation and biological phosphorus removal in a sequencing batch reactor treating piggery wastewater. Bioresource Technology 87 (1), 103–111.

Pallis, J.C., Irvine, R.L., 1985. Nitrogen removal in a low-loaded single tank sequencing batch reactor. Journal Water Pollution Control Federation 57, 82–86.

Ra, C.S., Lo, K.V., Shin, J.S., Oh, J.S., Hong, B.J., 2000. Biological nutrient removal with an internal organic carbon source in piggery wastewater treatment. Water Research 34 (3), 965–973.

Skrinde, J.R., Bhagat, S.K., 1982. Industrial wastes as carbon sources in biological denitriﬁcation. Journal Water Pollution Control Federation 54, 370–377.

Su, J.J., Kung, Ch.-M., Lin, J., Lian, W.-Ch., Wu, J.-F., 1997. Utilization of sequencing batch reactor for in situ piggery waste-water treatment. Journal of Environmental Science and Health (Part A) 32 (2), 391–405.

Tam, N.F.Y., Wong, Y.S., Leung, G., 1992. Signiﬁcance of external carbon sources on simultaneous removal of nutrients from waste-water. Water Science and Technology 26, 1047–1055.

Tam, N.F.Y., Leung, G.L.W., Wong, Y.S., 1994. Eﬀects of external carbon loading on nitrogen removal in sequencing batch reactors. Water Science and Technology 30, 73–81.

Tilche, A., Bacilieri, E., Bortone, G., Malaspina, F., Piccinini, S., Stante, L., 1999. Water Science and Technology 40 (1), 199– 206.

Tilche, A., Bortone, G., Malaspina, F., Piccini, S., Stante, L., 2001. Biological nutrient removal in a full-scale SBR treating piggery wastewater: results and modelling. Water Science and Technology 43 (3), 363–371.