Bioprocess and Biosystems Engineering

, Volume 41, Issue 8, pp 1115–1120 | Cite as

Effects of intermittent aeration periods on a structured-bed reactor continuously fed on the post-treatment of sewage anaerobic effluent

  • Bruno Garcia Silva
  • Márcia Helena Rissato Zamariolli Damianovic
  • Eugenio Foresti
Research Paper


This study assessed the simultaneous nitrification and denitrification processes and remaining organic matter removal from anaerobic reactor effluent treating wastewater in a single reactor. A structured-bed reactor, with polyurethane foam as support media, was subjected to intermittent aeration and effluent recirculation. Aerated/non-aerated periods varied in the range of 2/1–1/3 h. The chemical oxygen demand (COD) in the effluent remained between 26 and 42 mg L−1 throughout all the aeration conditions. Aeration periods of 1/2 h removed 80 and 26% of Total Kjeldahl Nitrogen and Total Nitrogen, respectively. A low solid production was observed during the 300 days of operation, resulting in a solid retention time of 139 days. The results indicate that the non-aerated periods generated alkalinity that favored nitrification, maintaining low COD concentrations in the effluent. The structured bed reactor presented a low solid production and effluent loss below 20 mgSSV L−1, similar to concentrations obtained in secondary decanters.


Sewage Anaerobic effluent Fixed bed reactor Intermittent aeration SND 


Water scarcity, more restrictive discharge standards, population growth, and urbanization require the development of wastewater treatment technologies that are easy to operate, inexpensive, and compact. In developing countries, anaerobic reactors are used to treat wastewater, because they do not use energy for aeration and have low-area requirements. However, anaerobic treatment is unable to completely remove organic matter and nitrogen remains as ammonia, often requiring post-treatment to meet discharge standards. Therefore, post-treatment technologies are sought that reach the tertiary level, which are expensive and complex to operate.

The available literature offers a wide range of systems for the post-treatment of effluents from Up-flow Anaerobic Sludge Blanket (UASB) reactors. Among these are: biological filters, submerged aerobic biofilters, wetlands, sequencing batch reactors, primary chemical treatment, zeolite column treatment, and dissolved air flotation systems [1]. For nitrogen removal, there is a demand for systems that integrate nitrification and denitrification in a single unit so as to avoid adding an external carbon source while at the same time economizing on alkalinizers.

The simultaneous nitrification and denitrification (SND) process occurs when the nitrification and denitrification reactions take place in a single operational unit. There are some advantages of systems that operate with SND compared to the conventional nitrogen removal systems such as: fewer treatment units; maintenance of unchanged operating conditions, not needing monitoring and control equipment between the aerated zone and the anoxic zone; reduction of operating costs; and lower demand for oxygen and alkalinity [2, 3].

The structured-bed reactor subjected to recirculation and intermittent aeration (SBRRIA) enables the SND process. This process occurs due to the establishment of anoxic and aerobic microbial communities in the fixed biofilm [4]. Thus, the system needs to receive an adequate amount of biodegradable organic matter for the denitrifying activity, as well as oxygen and alkalinity for the nitrifying activity. This paper reports on the aeration period as an operational parameter to balance oxygen and nitrate, as electron acceptors, to remove nitrogen and save alkalinity in the post-treatment of UASB reactor effluent.

Materials and methods

Experimental setup and operational strategy

The system comprised two serial reactors: a UASB followed by a SBRRIA (Fig. 1), fed by domestic sewage from a predominantly residential area. The UASB reactor had a cylindrical shape, 70 cm in height and a total volume of 11.2 L, and the SBRRIA reactor also had a cylindrical shape, but had a 14 cm inner diameter and was 90 cm high. The structured bed (support material) consisted of: 13 polyurethane foam cylinders measuring 3 cm in diameter and 70 cm high fixed by PVC rods. After inserting the support material, the working volume of the reactor was 9.6 L and 87.5% porosity. The SBRRIA was subjected to a recirculation ratio equal to 3. The flow rate to the UASB was 1.10 L.h−1, resulting in a Hydraulic Retention Time (HRT) of 9 h, whereas the flow rate to the SBRRIA was 0.75 L.h−1 and the HRT was approximately 12 h. A flow regulating reservoir was needed to ensure that anaerobic effluent would always be available. Oxygen was supplied by an air compressor unit which was connected to a porous stone so as to allow diffusion of the air into the liquid medium.

Fig. 1

Schematic representation of experimental setup. (1) UASB reactor, (2) UASB feed pump, (3) gas effluent, (4) reservoir, (5) excess discharge, (6) structured-bed reactor subjected to recirculation and intermittent aeration (SBRRIA) feed pump, (7) SBRRIA, (8) Effluent, (9) recirculation pump, and (10) air pump

Following the methodology described by Zaiat et al. (1994) [5], the support material was inoculated with a mixture (50% v/v) of sludge derived from activated sludge with a nitrifying activity and also anaerobic sludge from the UASB.

After inoculation, the reactor underwent a period a 24-day operation under continuous aeration, at which time complete nitrification was established. After nitrification was established, the non-aeration period was gradually extended to 1/3 h to favor denitrification, and thus TN removal. Table 1 shows the duration of each operational stage.

Table 1

Aeration time cycle conditions

Aeration period (aerated/non-aerated)

Duration (days)

Continuous aeration


2/1 h


1/1 h


0.75/1.25 h


1/2 h


1/3 h


Physico-chemical analyses

The system monitoring consisted of carrying out the following analyses: pH, alkalinity, Total Kjeldahl Nitrogen (TKN), NH4+ (ammoniacal nitrogen), NO2 (nitrite), NO3 (nitrate), TSS (total suspended solids), VSS (volatile suspended solids), dissolved oxygen (DO), and chemical oxygen demand (COD). The NO3, NO2−, and NH4+ analyses were carried out using a Dionex ion chromatography (ICS 5000®), equipped with a conductivity detector and two different columns (IonPac® AG23 Anion-Exchange Column and IonPac® CG12A Cation-Exchange Column), operating at 30 °C temperature. The flow adopted was 1.0 mL min− 1 and the mobile phase for determining the anions consisted of a calcium carbonate and calcium bicarbonate solution (4.5 and 0.8 mM, respectively). A solution of concentrated sulphuric acid (40 mM) was used as the mobile phase to determine the NH4+. Total Nitrogen (TN) was calculated as the sum of TKN, nitrite and nitrate. The alkalinity was determined according to Dilallo and Albertson [6], modified by Ripley et al. [7]. All the other analyses were carried out according to Standard Methods for the Examination of Water and Wastewater [8].

Results and discussion

Nitrogen removal and COD

The reactor was maintained under continuous aeration for 24 days. Under this condition, 9.5 mg L−1 of TKN was observed in the effluent, which was related to the consumption of 98% of the alkalinity, establishing complete nitrification. After the nitrification was established, the aeration period was gradually reduced to favor denitrification and, consequently, nitrogen removal. Table 2 shows the concentration of nitrogen compounds, as well as the efficiency of ammonia oxidation and TN removal, as the non-aeration time was increased. Nitrite and nitrate were not detected in the UASB effluent. Nitrite analyses were kept below the detection limit in all samples throughout the experimental period.

Table 2

TN concentrations in the influent and effluent nitrogen compounds in the SBRRIA

Aeration period

UASB effluent TN (mg L− 1)

SBBRIA effluent TN (mg L− 1)

SBBRIA effluent N-NO3 (mg L− 1)

SBBRIA effluent TKN (mg L− 1)

Ammonia oxidation efficiency (%)

TN removal efficiency (%)

Continuous aeration

36 ± 6 (10)

30 ± 7 (10)

17 ± 8 (10)

14 ± 5 (10)

65 ± 10

23 ± 15

2/1 h

38 ± 5 (10)

29 ± 3 (10)

22 ± 3 (10)

6 ± 3 (10)

80 ± 10

26 ± 9

1/1 h

39 ± 6 (8)

30 ± 4 (7)

20 ± 6 (8)

10 ± 7 (7)

75 ± 16

24 ± 10

0.75/1.25 h

38 ± 4 (9)

27 ± 2 (9)

17 ± 6 (10)

10 ± 4 (9)

68 ± 15

28 ± 6

1/2 h

34 ± 6 (11)

24 ± 4 (11)

18 ± 3 (11)

6 ± 2 (11)

82 ± 7

29 ± 10

1/3 h

22 ± 9 (8)

16 ± 7 (68)

3 ± 5 (8)

13 ± 6 (8)

42 ± 19

16 ± 17

Values represent the mean ± standard deviation (number of samples)

Total nitrogen (TN) = TKN + N-NO2 + NO3

The UASB effluent remained in the range of 85 to 150 mg DQO L−1, even when there were large variations of influent concentration. The small variations of UASB effluent ensured the stability of the SBRRIA, which presented the remaining COD throughout all the operational stages in the range from 26 to 42 mg L−1 (Table 3). The high organic matter removal capacity in SBRRIA was demonstrated in the previous studies [9, 10, 11]. These studies report that the remaining COD in the range of 24–69 mg L−1 was obtained for post-treatment of poultry slaughterhouse effluent and synthetic effluent with different carbon sources.

Table 3

Mean values of COD concentration in each operation phase

Aeration cycle

Raw sewage

UASB effluent

SBRRIA effluent

Total COD removal efficiency (%)

Continuous aeration

285 ± 77 (10)

138 ± 31 (9)

37 ± 11 (9)

97 ± 3

2/1 h

445 ± 272 (9)

128 ± 24 (9)

42 ± 16 (9)

87 ± 7

1/1 h

534 ± 120 (8)

147 ± 36 (7)

47 ± 19 (7)

91 ± 3

0.75/1.25 h

625 ± 231 (10)

108 ± 29 (10)

23 ± 9 (10)

96 ± 2

1/2 h

201 ± 52 (11)

120 ± 24 (10)

36 ± 20 (10)

82 ± 10

1/3 h

162 ± 48 (6)

86 ± 28 (6)

42 ± 13 (6)

73 ± 5

Values represent the mean ± standard deviation (number of samples)

The low availability of the remaining organic matter as an electron donor for denitrification was observed in all the experimental phases in the UASB reactor effluent (C/N ratio of 3.04). Due to this condition, no significant differences were found in the TN removal efficiency, reaching the maximum of 29 ± 10% in the 1/2 h phase. For the aeration periods of 1/3 h, the ammonia oxidation efficiency dropped to 42 ± 19%, therefore, showing the limit of oxygen availability for nitrification.

The results showed that the increase of up to 2 h of the non-aerated period, associated with the greater availability of electron donors, did not increase the TN removal even while maintaining high ammonia oxidation efficiencies. For periods of non-aeration of 3 h, damage to nitrification occurred without increasing denitrification of the nitrified fraction. There is a clear dependence between the aeration and non-aeration periods for the availability of electrons and maintenance of the microorganism activity in the same unit.

He et al. [12] demonstrated that TN removal increased from 30.33 to 89.90% when using the adequate amount of organic matter, which increased the C/N ratio from 6.12 to 21.29. This occurred, because increasing the organic matter can provide enough electrons for denitrification. According to the authors, a large amount of organic matter is used by the AHB during the diffusion process into the biofilm, while the rest might not meet the requirements for denitrification. The same phenomenon was observed by Chen et al. [13], reaching 43.9% of TN removal in an aerated biofilter treating synthetic wastewater at a C/N ratio of 3 (29.93 mg COD L−1 and 10.02 mg N-NH3 L−1). In that research, the increase of the C/N ratio to 8, increased the TN removal to 50.54%, maintaining the ammonia oxidation at 73.18%.

Although the 2/1, 0.75/1.25, and 1/2 h conditions presented similar ammonia oxidation efficiencies, the 1/2 h phase stands out as it maintained the efficiency at an average of 82 ± 7%. This value is 17% higher than the average obtained in the continuous aeration phase, possibly due to the increase in alkalinity availability, as the denitrification process generates alkalinity.

Most systems for the TN removal from sanitary sewage use sequential units, subject to effluent recirculation [14], or with batch feeding [3, 15, 16]. In general, the denitrification step requires adding an external carbon source. Few papers presented in the literature refer to systems in which only endogenous sources are used. For example, in the biofilter system composed by anoxic and aerobic units with recirculation [17], only 21% of the influent nitrogen was converted into gaseous nitrogen and 61% of the nitrogen remained as nitrate. The C/N ratio influent was 3.61 (112 mg total COD L−1 and 31 mg NTK L−1), therefore, limiting the denitrification process.


The effluent from the UASB reactor presented little variation concerning pH and alkalinity, whereas the alkalinity effluent to the SBRRIA is directly related to the TKN oxidation efficiency of the respective phase, because for every 1 mg of N-NH4+ oxidized to nitrite or nitrate, 7.14 mg CaCO3 L−1 of alkalinity is consumed. On the other hand, during the denitrification process, 3.57 mg CaCO3 L−1 is generated for every 1 mg of N–NO3 consumed. However, the observed denitrification efficiencies were low, and then, alkalinity generation reached a maximum of 65 mg L−1 in phase 1/2 h. Consumption and a generation of alkalinity were calculated from the mean of the influent and effluent TKN, N–NH4+, N–NO2−, and N–NO3 concentrations in each phase. Partial nitrification was not taken into account, because no nitrite was detected. The ammonia oxidation over nitrite (ANAMMOX) was also not considered due to the low concentrations of nitrite and ammonia. It should be noted that part of the ammonia is also used for anabolism. However, the requirement for ammonia for synthesis is about 1% [18] and the heterotrophic growth was considered low by the low organic load applied. Thus, the assimilation of ammonia by biomass is considered negligible. Table 4 shows the mass balance of the consumption and alkalinity generation for each phase.

Table 4

Mass balance related to the alkalinity consumed and generated in each phase

Aeration period

SBBRIA influent alkalinity (mg L− 1)

Nitrogen nitrifieda (mg L− 1)

Nitrogen ammonifiedb (mg L− 1)

Nitrogen denitrifiedc (mg L− 1)

Alkalinity consumptiond (mg L− 1)

Alkalinity generatione (mg L− 1)

Theoretical effluent alkalinity (mg L− 1)

Measured effluent alkalinity (mg L− 1)

Continuous aeration

176 ± 32







21 ± 39

2/1 h

172 ± 25







8 ± 9

1/1 h

200 ± 20







21 ± 47

0.75/1.25 h

221 ± 32







43 ± 38

1/2 h

180 ± 20







21 ± 8

1/3 h

157 ± 47







99 ± 51

aN–NH4+ nitrified = (TKN influent–TKN effluent)

bN-org ammonified = [(TKN influent − N–NH4+ influent)−(TKN effluent − N–NH4+ effluent)]

cN-NO3-denitrified = [N–NO3 influent + (TKN influent − TKN effluent) − N–NO3 effluent]

dEach 1 mg of N–NH4+-nitrified consumes 7.14 mgCaCO3 L−1 of alkalinity

eEach 1 mg de N–NO3-denitrified generates 3.57 mgCaCO3 L−1 of alkalinity

fEach 1 mg de N-org ammonified generates 3.57 mgCaCO3 L−1 of alkalinity

It can be observed that the alkalinity consumed, calculated by the reaction stoichiometry in the 2/1, 1/1, and 1/2 h phases, is higher than the influent alkalinity to the SBRRIA, indicating that the alkalinity generated during the denitrification process helped increase TKN oxidation efficiency when compared to the continuous aeration phase.

It is worth mentioning that, conventionally, effluent treatment systems that operate with continuous aeration require higher energy consumption to achieve high efficiency of COD removal. However, when the aim is nitrification, intermittent aeration may be an interesting option as non-aeration periods may help nitrification by generating alkalinity through denitrification.

In the SBRRIA system with intermittent aeration, organic matter can be removed by aerobic respiration during the aerated phase, or by heterotrophic denitrification in the anoxic phase. It is recommended that the effluent organic matter from the anaerobic reactor be used mainly for denitrification, which occurs in the absence of free oxygen. Thus, by adopting the intermittent aeration strategy, the oxygen limitation allowed some of the organic matter to be used for denitrification, therefore, maintaining the ammonia oxidation efficiency at a high level.

Solids production

Table 5 presents the average of the solids analyses at the collection points, as well as the solids retention efficiency. The concentrations of solids obtained from the SBRRIA effluent are comparable with values found in the effluent from secondary decanters in activated sludge systems. The volatile suspended solids within the reactor were estimated at 24.48 g. Based on this estimation, the solid retention time resulted in 138.8 days.

Table 5

Average concentration of solids in the sewage, UASB, and SBRRIA effluent


Volatile suspended solids (VSS)

Fixed suspended solids (FSS)

Total suspended solids (TSS)

Raw sewage

167 ± 120 (11)

36 ± 29 (11)

203 ± 141 (11)

UASB effluent

27 ± 11 (11)

16 ± 31 (11)

43 ± 37 (11)

SBRRIA effluent

14 ± 9 (11)

5 ± 8 (11)

20 ± 10 (11)

Solids retention efficiency

79 ± 22% (11)

24 ± 10% (10)

79 ± 22% (11)

Values represent the mean ± standard deviation (number of samples)

Considering that the solids which adhered to the SBRRIA biofilm remained constant, the observed yield coeficiente (Yobs) at 0.17 gVSS. gCOD−1 was also estimated. This value is lower than the one obtained for the submerged aerobic filter treating UASB reactor effluent (0.4 gVSS gCOD−1) [19].

The low cell growth observed was related to the high solids retention time (138 days), the low F/M ratio (0.09 gCOD gVSS−1 d−1). According to Habermacher et al., Habermacher et al. [20] increasing the retention time of solids from 25 to 80 days provided a 36% decrease in Yobs in activated sludge systems. Values even lower than those found in this study (0.12 gVSS gCOD−1) were reported in on a membrane reactor with complete retention of solids and showed a decreasing tendency in relation with the concentration of sludge in the reactor [21].


The SBRRIA subjected to an HRT of 12 h removed residual organic matter from the UASB effluent treating wastewater in both aeration conditions (continuous and intermittent). Under intermittent aeration, there was efficient nitrification up to 82 ± 7%, allowing the TN removal of 29 ± 10% TN and additional alkalinity generation. The best TN removal of 29 ± 10% occurred at 1/2 h condition due to the low electron donor availability in the UASB effluent (C/N ratio was 3.03). The structured-bed reactor with intermittent aeration presented low solid production (solids retention time is equal to 138.8 days and observed cellular yield of 0.17 gVSS gCOD−1). The value of the relation between aerated and non-aerated periods determined the efficiency of the involved processes and the concentrations of different nitrogen compounds in the effluent. The intermittent aeration strategy in SBRRIA has shown potential for reduction of alkalinity demand, energy consumption for aeration, and the amount of excess sludge.



This study was supported by FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo, Brasil) and CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico, Brasil).


  1. 1.
    Khan AA, Gaur RZ, Tyagi VK et al (2011) Sustainable options of post treatment of UASB effluent treating sewage: A review. Resour Conserv Recycl 55:1232–1251. CrossRefGoogle Scholar
  2. 2.
    Canto CSA, Rodrigues JAD, Ratusznei SM et al (2008) Feasibility of nitrification/denitrification in a sequencing batch biofilm reactor with liquid circulation applied to post-treatment. Bioresour Technol 99:644–654. CrossRefGoogle Scholar
  3. 3.
    Münch EV, Lant P, Keller J (1996) Simultaneous nitrification and denitrification in bench-scale sequencing batch reactors. Water Res 30:277–284. CrossRefGoogle Scholar
  4. 4.
    Wijffels RH, Tramper J (1995) Nitrification by immobilized cells. Enzyme Microb Technol 17:482–492. CrossRefGoogle Scholar
  5. 5.
    Zaiat M, Cabral AKA, Foresti E (1994) Horizontal-flow anaerobic immobilized sludge reactor for wastewater treatment: conception and preliminary perfor- mance evaluation. Braz J Chem Eng 11:33–42Google Scholar
  6. 6.
    Dilallo R, Albertson OE (1961) Volatile acids by direct titration. J Water Pollut Control Fed 33:356–365Google Scholar
  7. 7.
    Ripley LE, Boyle WC, Converse JC (1986) Improved alkalimetric monitoring for anaerobic digestion of high-strength wastes. Water 58:406–411Google Scholar
  8. 8.
    APHA (2005) Standard methods for the examination of water and wastewater, Centennial Edition. American Public Health Association. ISBN-13: 9780875530475Google Scholar
  9. 9.
    Santos CED, Moura RB, Damianovic MHRZ., Foresti E (2016) In fluence of COD/N ratio and carbon source on nitrogen removal in a structured-bed reactor subjected to recirculation and intermittent aeration (SBRRIA). J Environ Manage 166:519–524. CrossRefPubMedGoogle Scholar
  10. 10.
    Barana AC, Lopes DD, Martins TH et al (2013) Nitrogen and organic matter removal in an intermittently aerated fixed-bed reactor for post-treatment of anaerobic effluent from a slaughterhouse wastewater treatment plant. J Environ Chem Eng 1:453–459. CrossRefGoogle Scholar
  11. 11.
    Moura RB, Damianovic MHRZ., Foresti E (2012) Nitrogen and carbon removal from synthetic wastewater in a vertical structured-bed reactor under intermittent aeration. J Environ Manage 98:163–167. CrossRefPubMedGoogle Scholar
  12. 12.
    He S, Xue G, Wang B (2009) Factors affecting simultaneous nitrification and de-nitrification (SND) and its kinetics model in membrane bioreactor. J Hazard Mater 168:704–710. CrossRefPubMedGoogle Scholar
  13. 13.
    Chen D, Yu C, Bao T et al (2016) Simultaneous removal of nitrogen and phosphorus using autoclaved aerated concrete particles in biological aerated filters. Desalin Water Treat 57:19402–19410. CrossRefGoogle Scholar
  14. 14.
    Paredes D, Kuschk P, Mbwette TSA et al (2007) New aspects of microbial nitrogen transformations in the context of wastewater treatment—a review. Eng Life Sci 7:13–25. CrossRefGoogle Scholar
  15. 15.
    Moraes BS, Orrú JGT, Foresti E (2013) Nitrogen and sulfide removal from effluent of UASB reactor in a sequencing fed-batch biofilm reactor under intermittent aeration. J Biotechnol 164:378–385. CrossRefPubMedGoogle Scholar
  16. 16.
    Sun L, Wang Z, Wei X et al (2015) Enhanced biological nitrogen and phosphorus removal using sequencing batch membrane-aerated biofilm reactor. Chem Eng Sci 135:559–565. CrossRefGoogle Scholar
  17. 17.
    Rocher V, Paffoni C, Gonçalves A et al (2012) Municipal wastewater treatment by biofiltration: comparisons of various treatment layouts. Part 1: assessment of carbon and nitrogen removal. Water Sci Technol 65:1705–1712. CrossRefPubMedGoogle Scholar
  18. 18.
    Henze M, Van Loosdrecht MCM, Ekama GA (2008) Biological wastewater treatment: principles, modeling and design. Iwa Pub. ISBN-13: 9781843391883Google Scholar
  19. 19.
    Pujol R, Canler JP, Iwema A (1992) Biological aerated filters - an attractive and alternative biological process. Water Sci Technol 26:693–702CrossRefGoogle Scholar
  20. 20.
    Habermacher J, Benetti AD, Derlon N, Morgenroth E (2015) The effect of different aeration conditions in activated sludge - Side-stream system on sludge production, sludge degradation rates, active biomass and extracellular polymeric substances. Water Res 85:46–56. CrossRefPubMedGoogle Scholar
  21. 21.
    Pollice A, Laera G, Blonda M (2004) Biomass growth and activity in a membrane bioreactor with complete sludge retention. Water Res 38:1799–1808. CrossRefPubMedGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  • Bruno Garcia Silva
    • 1
  • Márcia Helena Rissato Zamariolli Damianovic
    • 1
  • Eugenio Foresti
    • 1
  1. 1.Biological Processes Laboratory, São Carlos School of Engineering (EESC)University of São Paulo (USP)São CarlosBrazil

Personalised recommendations