Microsoft Word - 82Lambri.doc CCHHEEMMIICCAALL EENNGGIINNEEEERRIINNGG TTRRAANNSSAACCTTIIOONNSS VOL. 38, 2014 A publication of The Italian Association of Chemical Engineering www.aidic.it/cet Guest Editors: Enrico Bardone, Marco Bravi, Taj Keshavarz Copyright © 2014, AIDIC Servizi S.r.l., ISBN 978-88-95608-29-7; ISSN 2283-9216 Aerobic Granular Sludge for Leachate Treatment Gaetano Di Bellaa, Michele Torregrossa*b a Facoltà di Ingegneria e Architettura dell’Università Kore di Enna, Cittadella Universitaria 94100 Enna, Italy. b Dipartimento di Ingegneria Civile, Ambientale, Aerospaziale, dei Materiali, Università degli Studi di Palermo, michele.torregrossa @unipa.it The treatment of municipal landfill leachate by means of aerobic granular sequencing batch reactors (GSBRs) was investigated. The paper reports the results from an experimental campaign lasted 100 days, which has been divided into three periods: cultivation of granular sludge (70 days), operation with semi-fresh (15 day) and diluted landfill leachate (15 day). Two different GSBR configurations were used: a Sequencing Batch Bubble Column reactor and a Sequencing Batch Airlift Reactor. All reactors were operated at Volume Loading Rates (VLRs) between 4.8 and 7.2 gCOD/(m 3 ·d). The Chemical Oxygen Demand (COD) removal efficiency varied between 80% and 90% under operation with synthetic wastewater feeding. On the other hand, the COD removal performance decreased to 40-50 % with semi-fresh leachate and to 50-60% with diluted leachate. Regarding nitrogen removal, after granules formation, the performance were satisfactory only when the reactors were fed with synthetic wastewater. Contrarily, the obtained results underline that a specific pre-treatment of ammonium must be applied in order to optimize nitrogen removal. However, the observed results indicate that the landfill leachate can be potentially treated in GSBR bioreactors. 1. Introduction Aerobic granulation is a new environmental biotechnology for treating a wide variety of wastewater (among others Adav et al., 2008). Indeed, this technology has shown a great potential in wastewater treatment due to several advantages represented by excellent biomass settle-ability, high biomass retention, good ability to treat high organic loading rate (Liu and Tay, 2006a). The aerobic granular sludge can be easily obtained in sequencing batch reactors with different configuration. In particular, the mostly used Granular Sequencing Batch Reactor (GSBR) configurations are: Sequencing Batch Bubble Column reactor (SBBC) and Sequencing Batch Airlift Reactor (SBAR) (Beun et al., 2002). Generally, the operational conditions imposed by a sequential procedure allow to operate a selective process between fast and slow settling biomass. Consequently, due to bacteria self-immobilization, the selected biological aggregates will gradually change in real stable granules (Liu and Tay, 2002). This initial selection phase can be called “cultivation phase” and it is considered as a physical screening step (or physical settling-washing out action) that operates a “selection pressure” on the biomass in the reactor (Liu and Tay, 2006b). Thus, only the aggregate that becomes big and dense enough to quickly settle would be retained in the reactor. In general, the cultivation phase is carried out gradually decreasing the settling time-length, in order to improve the selection pressure efficiency (Torregrossa et al., 2007). Granules formed in such a way are compact and characterized by an outer spherical shape. Further, the structure of stable granule is characterized by different layers: in the inner part of the granules, it can be observed the presence of heterotrophic population and, due to oxygen diffusion limitation inside the granules, it is possible to establish a denitrification process; in the intermediate layer, autotrophic biomass is dominant; in the outer layer, where oxygen and organic substances are highly available, heterotrophic growth occurs (Jin et al., 2008). The correct development of granulation, and consequently the actual granule stratification, depends on some important factors and conditions (Kim et al., 2008): feast and famine alternation, related to the duration of external substrates feeding; duration of the starvation phase; superficial air velocity; hydrophobic condition of the mixed liquor; organic loading rate. For these reasons, the granules can be formed after an unspecified time. Although aerobic granulation in GSBR has been extensively investigated, most of the previous studies concerning the cultivation of aerobic granules have been carried out with synthetic wastewater (Wey et al., 2012). Only few studies with real wastewater are available to better characterize the aerobic granulation process with domestic sewage (Kim et al., 2008; Ni DOI: 10.3303/CET1438083 Please cite this article as: Di Bella G., Torregrossa M., 2014, Aerobic granular sludge for leachate treatment, Chemical Engineering Transactions, 38, 493-498 DOI: 10.3303/CET1438083 493 et al., 2009). Reports of successful aerobic granulation systems were also obtained with other real readily biodegradable wastewater such as dairy effluent, malting wastewater, brewery wastewater (Schwarzenbeck et al., 2004 and 2005). Contrarily, the treatment of toxic and refractory wastewater is still lacking (Wey et al., 2012). In this context, the paper reports a case study that describes the organic carbon and nitrogen removal during cultivation period with synthetic wastewater in aerobic granular sludge SBR, and a subsequent treatment of landfill leachate at different organic and ammonium loading rates. 2. Materials and Methods 2.1 Experimental set-up Two different reactors, named R1 and R2, were used. In particular, the entire experimental period lasted 100 days, which has been divided into three sub-periods: • Period I (70 days) - cultivation of granular sludge with synthetic wastewater; • Period II (15 day) - operation with "semi-fresh" leachate; • Period III (15 day) - operation with leachate "strongly diluted" with synthetic wastewater. During Period 1 SBBC and SBAR configuration were used. Contrarily, during Periods II and III, both reactors were operated under SBBC configuration. All the reactors, were constituted by a working volume of 3.5 L. Other features and the scheme of experimental installation are reported in Figure 1. Real Wastewater RISER DO pH PLC SV DO pH Sludge withdrawal EFFLUENT SV = Solenoid valve B = Blower pH = pH probe DO = Dissolved Oxigen probe SV D = air diffuser Sr P PB SBAR SBBC D D Real Wastewater Sludge withdrawal Sr= wastewater screenning Wastewater collected from full scale plant Reactor features Units SBBC SBAR External cylinder Riser Height mm 860 860 600 Level water table mm 700 740 740 Internal diameter mm 80 80 54 External diameter mm 90 90 60 Bottom reactor – riser distance mm 30 Volumetric exchange ratio % 50 50 Working volume L 3.5 3.5 influent Figure 1 Pilot plant layout: geometric characteristics of reactors The average characteristics of synthetic wastewater (Period I) and leachate (Period II and III) are shown in Table 1. In particular, leachate was collected from Palermo Municipal landfill. Preliminarily, aged and fresh leachates were mixed and stored, in order to maintain similar characteristics during the whole experimentation. Subsequently, during Period II, it was used leachate with a pollutants concentration greater than ones in the Period III, in order to study the potential occurrence of granule breaking effect due to the different nature of the substrates in the influent (after initial cultivation). More specifically, in order to obtain a "semi-fresh" mixture a weak dilution (with tap water) was applied (in order to maintain an average concentration of 9600 mg/L). Contrarily, in Period III, in order to evaluate the possible improvement of system performance, and the eventual new granulation, a "strong" dilution of leachate was applied (until reaching a average COD concentration of 4800 mg/L). The synthetic wastewater was only used during cultivation and in R2 during period II. This last condition was applied in order to compare the performance of two reactors fed with different wastewaters, maintaining the same Volumetric Loading Rate (VLR). The plant was started-up with an initial settlement time (tS) of 7 minutes that was decreased to 3 minutes after only 2 weeks (Torregrossa et al., 2007). The cyclic operation of SBR has been automated using a programmable logic controller (PLC). In order to change the VLR during the experimentation, without changing the COD concentration in the wastewater, the cycle time was changed (generally, longer times were applied when feeding leachate, according to Wey et al., (2012)). More specifically, in order to balance the VLR, the reactors have been operated as follows: 8 daily cycles of 3 h each, during cultivation (Period I) and when synthetic wastewater was fed in R2; 1 daily cycles of 24 hours and 3 cycles of 8 hours when "semi-fresh" or "diluted" leachate were fed, respectively. In table 2, all operational conditions during the whole experimentation are summarized. 494 Table. 1. Composition of synthetic and landfill leachate fed to the GSBRs Parameter Period I Period II Period III Syntetic wastewater [mg/L] R1(SBAR) R2(SBBC) R1(SBBC) R2(SBBC) R1(SBBC) R2(SBBC) COD 598 ± 50 598 ± 50 1185 ± 101 CODsol 162 ± 13 162 ± 13 340 ± 24 N-NH4 25 ± 2.4 25 ± 2.4 51 ± 8.1 N-NO3 0.3 ± 0.1 0.3 ± 0.1 0.3 ± 0.1 N-Ntot 30 ± 2.4 30 ± 2.4 55 ± 9.4 Leachate Weak diliution COD 9738 ± 450 CODsol 2800 ± 125 N-NH4 1960 ± 82 N-NO3 0.5 ± 0.4 N-Ntot 3700 ± 285 Leachate Strong diliution COD 4560 ± 165 4560 ± 165 CODsol 1540 ± 175 1540 ± 175 N-NH4 945 ± 54 945 ± 54 N-NO3 0.3 ± 0.4 0.3 ± 0.4 N-Ntot 1845 ± 175 1845 ± 175 Table. 2. Operational condition during the period I, II and III. Experimen tal period Period duration Range Type Cycle Duration of single SBR phase VLR tF tAIR tS tDS days days h min min min min min kgCOD·m -3 ·d -1 I Synthetic wastewater 70 1-7 SBAR 180 5 163 7 5 2.4 R1 8-14 SBAR 180 5 165 5 5 2.4 15-21 SBAR 180 5 167 3 5 2.4 II Leachate Weak dil. 15 71-85 SBBC 1440 5 1427 3 5 4.8 III Leachate Strong dil. 15 86-100 SBBC 720 5 707 3 5 4.8 I Cultivation 70 1-7 SBBC 180 5 163 7 5 2.4 R2 8-14 SBBC 180 5 165 5 5 2.4 15-21 SBBC 180 5 167 3 5 2.4 II Synthetic wastewater 15 71-85 SBBC 180 5 167 3 5 4.8 III Leachate Strong dil. 15 86-100 SBBC 480 5 467 3 5 7.2 tF – Fill time; tAIR – Aeration time; tS – Settling time; tF – Fill time; tDS – effluent discharge time 2.2 Analytical procedure In both experimental periods, the chemical-physical parameters analysed were: total COD, soluble COD (CODsol), total suspended solids (TSS), volatile suspended solids (VSS), ammonium nitrogen (NH4-N) and nitrate nitrogen (NO3-N), with regards to influent, mixed liquor and effluent. In particular, the biomass concentration (in term of TSS), in the reactor and in the effluent, has been determined by filtrating the homogeneous sample using a 0.45 μm filter and drying the filter for at least 24 h at 105°C. All the other chemical-physical analyses have been carried out according to the Standard Methods [APHA, 1998]. Sludge characteristics have been analyzed by means of microscopic and stereoscopic observations. In particular, microbiological analysis have been carried every two-three days. More specifically, a sample of granular sludge (50 mL) has been analysed through a stereoscope that acquired the microscopic images, using an opportune enlargement (relating to granule sizes). The images has been analysed by a specialized IA (IM50 by Leica). It is important to specify that the granule dimensions have not been evaluated through a statistical measure, but rather as average value of the maximum dimensions observed in the mixed liquor sample. 3. Results 3.1 Cultivation Period The cultivation phase was aimed to the granular sludge production: for this reason, as discussed above, the plant was fed with synthetic wastewater. By analyzing the granulation phenomenon as well as on the basis of the large number of measurements, some interesting aspects can be drawn. In particular, as underlined by the average data shown in Figure 2, the reactor R1 in SBAR configuration, showed a slightly better granulation phenomenon compared to that one occurred in the reactor R2 in SBBC configuration. Briefly: 1. the removal of the organic substance (in terms of COD) was satisfactory in both reactors and improved with the growth of biomass, reaching values higher than 85%, at the end of Period I; 2. nitrification (N) occurred efficiently only when the granules reached an average size greater than 0.5- 0.6 mm (after the 22nd day in R1 and after the 29th day in R2); 3. denitrification (DN) have been particularly affected by the growth and stratification of the granules and only at the end of Period I the best results have been achieved: in any case, the denitrification was 495 always greater in the granules of reactor R1 (SBAR), because they were larger and (obviously) better stratified than granules in R2. Cultivation Period from to ηCOD ηN ηDN size TSS EPStot day day % % % mm g/L mg/gVSS 0 7 77.7592 53.0259 15.5584 0 9.55933 113.251 8 14 80.602 64.5907 25.9756 0.2 13.547 142.674 15 21 84.6455 68.1376 17.3247 0.39 13.687 213.892 22 28 87.9462 72.4552 42.1735 0.5 19.1565 337.409 29 49 89.8198 76.436 63.0129 0.63667 23.0181 271.599 50 70 91.0502 77.2789 81.4889 0.93 17.1748 228.116 0 7 76.087 46.2048 8.1319 0 8.18219 71.9758 8 14 79.097 57.7362 21.1685 0.1 19.4384 219.625 15 21 83.1104 64.4001 25.0222 0.32 18.1848 293.365 22 28 86.9565 68.9198 33.924 0.44 14.4807 252.658 29 49 87.3416 73.7428 50.7918 0.55333 18.2208 311.917 50 70 88.709 76.036 70.7455 0.85667 22.55 261.141 R1 R2 1 mm 1 mm 1 mm 1 mm R1 (8) R1 (28) R2 (8) R1 (29) Figure 2 Granule sizes during cultivation phase Therefore, the observed results confirmed the usefulness of the airlift in the column during the cultivation phase (Di Bella and Torregrossa, 2013), since a more regular shear stress can be obtained. Moreover, it appeared that the mechanical stress, due to the presence of the airlift, initially caused a greater production of EPST. The latter acts as "glue" for other biological aggregates, facilitating the initial formation and growth of granules. Unfortunately, the analysis of EPST was not performed in the subsequent periods, due to the influence of leachate compounds in the colorimetric analysis. 3.2 Experience with leacheate As previously discussed, starting from Period II, both reactors were configured as SBBC (by removing the airlift in R1). Furthermore, in order to properly compare the reactors, the granular sludge of R1 (the "best" one) was redistributed in both reactors with a starting concentration of about 11 g/L in terms of TSS. More specifically, only a portion of granular sludge of R1 was collected (after a preliminary selection of the granules with best settle-ability). The only differences were: the type of wastewater (leachate or synthetic) and the operating conditions (VLR), as already shown in the table 2. 3.2.1. Granulation The available studies on granulation in GSBR systems fed with landfill leachate are very limited, especially under aerobic conditions. The study of Wei et al. (2012), already cited in this paper, was based on the experimental observation of the performance achieved thanks to a start-up period carried out directly with leachate: although useful, this choice has limited the size of the granules to values lower than 0.5 mm. On the contrary, in the present experience, the evolution of the granulation resulting by organic and ammonium shock loads has been studied: starting from an ideal environment for the granules growth (synthetic wastewater), it was "abruptly" fed leachate (more or less diluted). The aim was to study what happened to the granules. Figure 2 shows the trend of the average size of the granules and the concentration of the biomass (in terms of SS). As shown in Fig. 2a (Period II) the average granule sizes tended, as expected, to decrease when the leachate was fed to the system. On the contrary, in the R2 (Fig 2b) the granule sizes were maintained (apart from the immediate shock due to the jump of VLR, from 2.4 to 4.8 kg COD/(m 3 d)) and even increased at the end of Period II. From day 85 (Period III), the granulation showed a different trend: in the reactor R1 there was a formation of new granules and the unstructured ones have begun to "grow"; while in R2 the granules have begun to break down. This was probably due to the fact that the leachate was fed in R1 when the biomass was partially acclimated, but with concentrations of leachate lower than ones in the Period I (maintaining the value of VLR, via the reduction of the cycle from 24 to 12 hours). On the contrary, in reactor R2, in Period III it was fed real leachate to the biomass previously acclimatized with synthetic wastewater (low concentrations but at a higher load, from 4.8 to 7.2 kg COD/(m3 d), with 3 cycles of 8 hours daily). However, despite the high stress imposed, the granule breaking was limited and less evident, probably because the granulation and stratification were more efficient. Obviously, the SS concentration trends in the reactors were influenced by the rupture of the granules and cyclical conditions of "washout". This, of course, had obvious repercussions on the performance of the process, as will be shown below. 496 3.2.2. COD and ammonium removal The performances, in terms of COD removal, are shown in Fig 3. 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 70 75 80 85 90 95 100 S S c on ce nt ra ti on [ g/ L] G ra nu le S iz e [m m ] Time [day] TSS Granule Size 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 70 75 80 85 90 95 100 S S c on ce nt ra ti on [ g/ L] G ra nu le S iz e [m m ] Time [day] TSS Granule Size 1 1 1 mm 1 mm 1 mm 1 mm a) b) Figure 3 Granule sizes and TSS concentration in R1 (a) and R2 (b) -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 70 75 80 85 90 95 100 P er fo rm an ce [% ] C O D [m g/ L] Time [day] [COD] influent [COD] effluent COD removal LEACHATE (PERIOD II) VLR = 2.4 kgCOD/m2/day LEACHATE (PERIOD III) VLR = 4.8 kgCOD/m2/day -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 70 75 80 85 90 95 100 P er fo rm an ce [% ] C O D [m g/ L] Time [day] [COD]in [COD]out COD removal SYNTHETIC (PERIOD II) VLR = 4.8 kgCOD/m2/day LEACHATE (PERIOD III) VLR = 7.2 kgCOD/m2/day Figure 4 Organic removal performance It is important to underline that the experimentation was carried-out under dynamic conditions during Period II and III. However, the results showed that a stable organic was achieved in both reactors, but the COD removals are meanly low: 45-50% in R1 and 60-65 % in R2. In fact, the organic removal was strongly dependent on the changing of influent, from synthetic to leachate. Only in R2 during Period II, with synthetic wastewater, the COD removal are very good, always higher than 80% despite the high COD concentration (about 1200 g/L). Regarding the leachate feeding in Period III, according to observation reported above, the data show that the granulation phenomenon start again in R1 when leachate was fed. Summarizing, from the 70th to 85th day of R1 the removal efficiency decreased from 70% to 20% due to strong deflocculation of the granules (and washout of biomass). In the same period, R2 was fed with synthetic wastewater with a COD concentration of 1185 ± 105 mg/L and COD removal fluctuated in the range 78-88%. In Period III, the COD removal in R1 increased reaching a value equal to 45%, thanks to the establishment of a new phenomenon of granulation. Contrarily, the COD removal in R2 decreased but only to 55%, despite the high organic loading rate and the changing of the influent (from synthetic to leachate). On the other hand, Figure 3 shows the results in terms of Nitrification (N) and Denitrification (DN) performance in both reactors. In general, biological nitrogen removal normally involves two separated step: aerobic nitrification of ammonium to nitrate; anoxic denitrification of nitrate to nitrogen gas. In the granular sludge process the simultaneous nitrification and denitrification (SND) is generally developed (Di Bella and Torregrossa, 2013). In this context, the denitrification-nitrification activity depends on granule size because, without dissolved oxygen control, the oxygen penetration change with granules diameter. In particular, the SND process is satisfactory when synthetic wastewater was fed: both denitrification and nitrification performance higher than 70-80% (data are not reported). On the other hand, when the leachate was used as influent, the nitrogen conversion occurred at "high ammonium concentration": in this condition, according to Wey et al., (2012), nitrifying bacteria in aerobic granules was enriched, while the heterotrophs lost their competitive dominance, which anticipated the simultaneous ammonium oxidation and organic biodegradation. Besides this aspect, due to ammonium oxidation production, the activities of both ammonium oxidizing bacteria (AOB) were inhibited, resulting in the accumulation of ammonium (lower nitrification efficiency) in both typical cycles. Nitrate concentration was also increasing due to decrease in denitrifying activity. For this reason, the denitrification performance resulted not satisfactory (<20% in R1 and <40% in R2), despite the granular sizes are not very small. this result was similar to what observed for nitrification. Probably, for leachate treatment in GSBR, a specific pre-treatment for ammonium removal must be applied, in order to reduce the ammonium concentration in the effluent. 497 -100 -80 -60 -40 -20 0 20 40 60 80 100 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 70 75 80 85 90 95 100 Pe rf or m an ce [% ] NH 4-N [m g/ L] Time [day] [N-NH4] in [N-NH4]out Nitrification -100 -80 -60 -40 -20 0 20 40 60 80 100 0 20 40 60 80 100 120 140 160 180 200 70 75 80 85 90 95 100 Pe rf or m an ce [% ] NO 3-N [m g/ L] Time [day] [N-NO3] in [N-NO3]out DENitrification LEACHATE (PERIOD II) LEACHATE (PERIOD III) LEACHATE (PERIOD II) LEACHATE (PERIOD III) -100 -80 -60 -40 -20 0 20 40 60 80 100 0 200 400 600 800 1000 1200 1400 1600 1800 2000 70 75 80 85 90 95 100 Pe rf or m an ce [% ] NH 4-N [m g/ L] Time [day] [N-NH4]in [N-NH4]out Nitrification -100 -80 -60 -40 -20 0 20 40 60 80 100 0 20 40 60 80 100 120 140 160 180 200 70 75 80 85 90 95 100 Pe rf or m an ce [% ] NO 3-N [m g/ L] Time [day] [N-NO3] in [N-NO3]out DENitrification SYNTHETIC (PERIOD II) LEACHATE (PERIOD III) SYNTHETIC (PERIOD II) LEACHATE (PERIOD III) R1 R1 R2 R2 Figure 5 Nitrogen removal performance 4. Conclusion The aim of this work has been to observe the phenomenon of the aerobic granulation in two GSBR fed with landfill leachate The result confirm the possibility to obtain aerobic granule size feeding landfill leacheate. However, this study has confirmed the results reported in the literature highlighting the decisive role of ammonium concentration: in particular, the organic and nitrogen removal tend to decrease when ammonium concentration increase. For this reason, a specific ammonium pre-treatment must be applied before aerobic granular sludge application. Furthermore, the initial cultivation of granular should be carried out with small increment of lecheate, in order to optimize the bacteria acclimatation and specification. Reference Adav S.S., Lee D.J., Show K.Y., Tay J.H., 2008. Aerobic granular sludge: recent advances. Biotechnology Advances vol. 26, pp. 411–423. APHA-AWWA-WEF 1998., Standard Methods for the Examination of Water and Wastewater, 20th ed. American Public Health. Beun J.J., van Loosdrecht M. C. M. and Heijnen J.J., 2002. Aerobic granulation in a sequencing batch airlift reactor. Water Research vol. 36, pp. 702-712. Di Bella G., Torregrossa M., 2013 Simultaneous nitrogen and organic carbon removal in aerobic granular sludge reactors operated with high dissolved oxygen concentration. Bioresource Tecnhology vol. 142 (2), pp. 706-713. Jin R.C., Zhenga P, Mahmooda Q., Zhanga L., 2008. Performance of a nitrifying airlift reactor using granular sludge. Separation and Purification Technology vol. 63, pp. 670–675. Kim I.S., Kim S.M., Jang A., 2008. Characterization of aerobic granules by microbial density at different COD loading rates. Bioresource Technology vol. 99, pp. 18–25. Liu Y.Q., Tay J.H., 2006a. Variable aeration in sequencing batch reactor with aerobic granular sludge. Journal of Biotechnology vol. 124, pp. 338–346. Liu Y.Q., Tay J.H., 2006b. Cultivation of aerobic granules in a bubble column and an airlift reactor with divided draft tubes at low aeration rate. Biochemical Engineering Journal vol. 34, pp. 1-7 Liu Y.Q., Tay J.H., 2002. The essential role of hydrodynamic shear force in the formation of biofilm and granular sludge. Water Research vol. 36, pp. 1653–1665. Ni B.J, Xie W.M, Liu S.G., Yu H.Q., Wang Y.Z., Wang G., Dai X.L., 2009. Granulation of activated sludge in a pilot-scale sequencing batch reactor for the treatment of low-strength municipal wastewater. Water Research vol. 43, pp. 751-761. Schwarzenbeck N., Borges, J.M., Wilderer, P.A., 2005. Treatment of dairy effluents in an aerobic granular sludge sequencing batch reactor. Appl. Microbiol. Biotechnol. 66, 711–718. Schwarzenbeck N., Erley, R., Mc Swain, B.S., Wilderer, P.A., Irvine R.L., 2004. Treatment of malting wastewater in a granular sludge sequencing batch reactor (SBR). Acta Hydroch. Hydrob. 32, 16–24. Torregrossa M., Di Bella G., Viviani G., Gnoffo A., 2007. Performances of a granular sequencing batch reactor (GSBR). Water Science and Technology vol. 55 (8-9), pp. 125-133. Wei Y., Ji M., Li R., Qin F., 2012. Organic and nitrogen removal from landfill leachate in aerobic granular sludge sequencing batch reactors Waste Management 32, 448–455 498