The paper presents results of study of anoxic ammonia oxidation at low temperature conducted at JSC Mosvodokanal, Moscow, Russia. The study was carried out in two reactors, 5 l each, operated at the temperature of 5–10 °С. Both reactors were inoculated with the soil, collected from waste water sludge landfill, that presumably, contained low-temperature adapted nitrifying bacteria. Reactor No. 1 contained nitrifying bacteria only. Reactor No. 2 was further inoculated by anammox bacteria. Filtrate from digested sludge belt thickeners was added into the reactors to achieve a final N-NH4 concentration of 70–95 mg/l. The reactors were operated as sequencing batch reactors. After 90 days of incubation maximal nitrification rate in reactor No. 1 was 1.4 mg N-NH4/g VSS*h, and in reactor No. 2–1.0 mg N-NH4/g VSS*h. Estimated doubling time of nitrifying bacteria was 45 days. Total mineral nitrogen removal in the 1st reactor was 20% (via process of heterotrophic denitrification), and in the 2nd – 60% (via both the processes of heterotrophic and autotrophic nitrate reduction). Through the process of autotrophic denitrification (anammox), two times as much nitrogen was removed, compared to the heterotrophic denitrification process. Anammox process rate was 0.4 mg N-NH4/g VSS*h.

A process of autotrophic anoxic ammonia oxidation (АААО) is used for the treatment of waste water with high concentrations of ammonia and low concentrations of organic matter. Nitrite is used as ammonia oxidant. The autotrophic nitrogen removal technology is a combination of two processes: partial nitrification, which involves oxidizing half of ammonia to nitrite, and a process of anoxic ammonia oxidation by nitrite, called ‘anammox’. The advantages of this technology are as follows: high efficiency (removal of over 90% of nitrogen), no organic carbon demand, low sludge yield and lower nitrogen removal cost (Jetten et al. 1999). Nowadays, anammox bacteria – based processes of waste water treatment are introduced in more than 100 WWTPs all over the world. The most known specific technologies are SHARON-ANAMMOX, DEMON, Canon, ANITA-Mox, DeAmmon, OLAND and some others (Van Hulle et al. 2010).

The AAAO studies have been carrying out at JSC Mosvodokanal since 2006. Two technologies were developed. AAAO process for the treatment of filtrate of digested sludge belt thickeners in a two-reactor pilot plant had optimal temperature range of 15–35 оС (Danilovich et al. 2010). One-reactor technology had optimal temperature range of 28–35 °С (Kozlov et al. 2014). Heating of large amounts of waste water is a power-consuming and financially undesirable process. Therefore, for the development of AAAO process it is important to run it at low ambient temperatures of 10–20 оС or even below. The Paques company is one of the leaders in the development of AAAO technology. They are working on the development of a low temperature АААО process (http://www.dutchwatersector.com/news-events/news/5796-cold-anammox-starts-nitrogen-removal-from-low-temperature-waste-water-at-wwtp-dokhaven.html). There are some other works describing anammox process at low temperatures of 12–18 °С (Yang et al. 2011; Winkler et al. 2012; Hu et al. 2013).

Implementation of АААО process is very important for the countries with cold climate, including Russia, particularly for nitrogen removal from high-ammonia loaded effluents, like from sludge treatment facilities, sludge beds, where temperature ranges between 6 and 20 °С. It is important to note that complete AAAO process is realized by two distinct groups of bacteria both equally important for this process, nitrifying and anammox bacteria. Nitrifying bacteria are very sensitive to low temperatures (Henze et al. 2000). Nitrification rate decreases 14 times as temperature changes from 25 to 6 °C (Danwille 1995).

The objective of this study was to explore a possibility to run ammonia oxidation process at 5–10 °С.

A set-up for activated sludge cultivation and waste water treatment consisted of two plastic reactors, 5 l each, placed into thermostat KS-200 with the temperature maintained at the level of 5–10 °С. Both reactors were inoculated with soil samples, taken from the sludge treatment plant (sludge landfill, Moscow region), that presumably, contained low temperature adapted nitrifying bacteria, as the samples were taken in deep autumn. Reactor No. 1 contained nitrifying bacteria only. Reactor No. 2 was further inoculated with activated sludge from the anoxic ammonia oxidation pilot plant, described in (Hramenkov et al. 2013), which contained a new anammox bacteria candidatus Аnammoxomicrobium moscowii with the temperature optimum of 15–35 °С. They were able to exhibit anammox activity at the temperatures below 10 °С (Figure 1).

Figure 1

Relative activity of anammox bacteria in the activated sludge, used for bio-reactor No. 2 inoculation.

Figure 1

Relative activity of anammox bacteria in the activated sludge, used for bio-reactor No. 2 inoculation.

Close modal

The concentration of О2 in reactor No. 1 was maintained at 1–4 mg/l, in reactor No. 2 – at 0.5–2 mg/l. Filtrate from digested sludge belt thickeners was added into the reactors to achieve a final ammonia nitrogen concentration of 70–95 mg/l. The medium pH was maintained at 7.8–8.3, which was an optimal value for this type of anammox bacteria, by adding СаСО3. The reactors were operated as SBR (sequencing batch reactors), involving the following stages: sedimentation of sludge was 1 h, treated filtrate discharge was 10 min, fresh filtrate loading was 1 min, aeration was 4–7 days. The characteristics of untreated and treated filtrates, along with the reactor performance parameters are shown in Table 1. For pollutants concentrations determination common methods were used (APHA AWWA WEA 2012). Specific activities of anammox and nitrifying bacteria were determined in the end of experiment by the method of ‘sacrificed flasks' as described in (Hramenkov et al. 2013). Discrepancy between nitrification rates measured directly and calculated from ammonia removal in the reactors were 15% or less.

Table 1

Reactors No. 1 and No. 2 performance parameters

 Reactor No. 1Reactor No. 2
TSS, g/l 2.3–2.6 3.3–3.6 
pH 7.8–8.3 7.8–8.3 
Dissolved oxygen mg/1–4 0.5–2 
Temperature, °C 5–10 5–10 
N- NH4 supplied water, mg/l 70–95 70–95 
N- NH4 treated water, mg/l 2–4 10–15 
N-mineral total, treated water, mg/l 60–70 30–40 
Nitrification rate, mg N-NH4/g VSS*h (calculated/directly determined) 1.4/1.28 1.05/1.1 
Anammox process rate, mg N-NH4/g VSS*h 0.4 
 Reactor No. 1Reactor No. 2
TSS, g/l 2.3–2.6 3.3–3.6 
pH 7.8–8.3 7.8–8.3 
Dissolved oxygen mg/1–4 0.5–2 
Temperature, °C 5–10 5–10 
N- NH4 supplied water, mg/l 70–95 70–95 
N- NH4 treated water, mg/l 2–4 10–15 
N-mineral total, treated water, mg/l 60–70 30–40 
Nitrification rate, mg N-NH4/g VSS*h (calculated/directly determined) 1.4/1.28 1.05/1.1 
Anammox process rate, mg N-NH4/g VSS*h 0.4 

N-mineral total was expressed as the sum of N-NH4 + N-NO2 + N-NO3.

The AAAO process implementation depends on the activity of two groups of bacteria: aerobic ammonia oxidizing bacteria, which oxidize ammonia to nitrite, and the anammox bacteria, which oxidize ammonia by nitrite. During this study it was important to demonstrate the activity of both groups at low temperatures.

Dynamics of ammonium concentrations in incoming and purified water, and the content of total mineral nitrogen in purified water are shown in Figure 2.

Figure 2

The content of ammonium (as N-NH4) in the incoming water (1), and in the treated water: in the reactors No. 1 (2) No. 2 (3) and total mineral nitrogen Ntot min in treated water in reactor No. 1 (4) and No. 2 (5) in the course of experiment. Ntot min were expressed as the sum of N-NH4 + N-NO2 + N-NO3.

Figure 2

The content of ammonium (as N-NH4) in the incoming water (1), and in the treated water: in the reactors No. 1 (2) No. 2 (3) and total mineral nitrogen Ntot min in treated water in reactor No. 1 (4) and No. 2 (5) in the course of experiment. Ntot min were expressed as the sum of N-NH4 + N-NO2 + N-NO3.

Close modal

Active nitrification took place in both reactors, however, at different rates. In reactor No. 1, inoculated only by soil nitrification rate was higher (Figure 3), due to the higher oxygen content in the reactor (1–4 mg/l versus 0.5–2 mg/l in reactor No. 2). During the first 20 days of the experiment nitrification rate in both reactors were similar and grew slightly – from 0.24 to 0.33 mg of N-NH4/gVSS*h. During this period, the bacteria were undergoing an adaptation to a new environment. In the following 70 days of incubation in the reactor No. 1 nitrification rate rose to 1.4 mg N-NH4/g VSS*h. The reactor No. 2 nitrification rate was 40% lower. Nitrification rate, as defined in the direct experiment, was 1.28 mg and 1.1 mg N-NH4/g VSS*h for reactors No. 1 and No. 2, respectively, which corresponds to the real rates of the process in the reactors.

Figure 3

The specific rate of oxidation (removal) of ammonium (as N-NH4) in reactor No. 1 (1) and No. 2 (2) at a temperature of 5–10 °С.

Figure 3

The specific rate of oxidation (removal) of ammonium (as N-NH4) in reactor No. 1 (1) and No. 2 (2) at a temperature of 5–10 °С.

Close modal

As the velocity of any microbiological process is proportional to a number of microorganisms which catalyses this process, then the specific rate of ammonia oxidation is proportional to the content of ammonium oxidizing bacteria in the activated sludge. From the curves shown in Figure 3, we can calculate the doubling time of nitrifying bacteria – using the Excell software and approximating the exponential growth curve. Doubling time of nitrifying bacteria in reactor No. 1 was 45 days. For reactor No. 2 calculation was not carried out, because it would be incorrect due to the parallel existence of competitive process (anammox).

The resulting specific ammonia oxidation rate is substantially lower than in the aerobic bioreactors operated at temperatures 18–25 °C (2–4 mg N-NH4/g VSS*h) due to decrease of all biological processes at lower temperatures.

As a result of increase the activity of psychrophilic nitrifying bacteria, the ammonium content in treated water of reactor No. 1 at day 90th was 2–4 mg/l. In water treated in reactor No. 2 this parameter was 10–15 mg/l, despite the presence of two groups of bacteria that oxidize ammonium. This testifies to the leading role of the nitrification process in the removal of ammonia at low temperatures. Data show that ammonium removal efficiency reaches 95%.

In addition to the removal of ammonia in the reactors decrease of the total amount of mineral nitrogen was also observed. Total mineral nitrogen was calculated as the amount of ammonia, nitrite and nitrate nitrogen (Figures 2 and 4). Noticeable removal of mineral nitrogen in the reactor No. 1 began at the day 20, and in the reactor No. 2 - from the very beginning of cultivation. After 90 days of the experiment reactor No. 1 removed approximately 20% of the incoming mineral nitrogen, and the reactor No. 2–60%, i.e. 3 times more.

Figure 4

Fraction of total mineral nitrogen Ntot min removed in reactor No. 1 (1) and in reactor No. 2 (2) at a temperature of 5–10 °С, expressed as a fraction of the amount of supplied mineral nitrogen.

Figure 4

Fraction of total mineral nitrogen Ntot min removed in reactor No. 1 (1) and in reactor No. 2 (2) at a temperature of 5–10 °С, expressed as a fraction of the amount of supplied mineral nitrogen.

Close modal

It is obvious that heterotrophic denitrification took place in the first reactor in which nitrate and nitrite formed from ammonia oxidized organic substances present in the filtrate. The multiplication of these bacteria took some time, about 3 weeks, and that led to the absence of nitrogen removal in the first 3 weeks. The possibility of nitrification–denitrification process in filtrate of digested sludge at higher temperatures by specific heterotrophic microflora was shown earlier (Khar'kina et al. 2010).

The reactor No. 2 was inoculated by activated sludge with anammox bacteria capable of ammonium nitrite oxidation at low temperatures, so total mineral nitrogen removal rate was 3 times higher. By the end of this experiment total mineral nitrogen removal rate reached a value of 60%. Moreover, nitrogen removal began in the first days, because anammox bacteria were adapted to the filtrate of digested sludge.

The maximum rate of anammox process was 0.4 mg N-NH4/g VSS*h, which is significantly lower (30–100 times) than the specific rate of anammox process carried out by the same anammox bacteria in previous work (Hramenkov et al. 2013). This low rate is explained by two factors. 1. Activated sludge with anammox bacteria was stored for 2 years without food and some part of the bacteria died out. 2. Decrease of all chemical and biochemical reactions with decrease of temperature. Temperature coefficient of Vant Hoff is 2–4 for change the temperature to 10 °C. Accordingly, when the temperature drops from 30 to 10 °C, the rate of any (bio)chemical reaction can be lower 2–16 times. In (Hu et al. 2013), the rate of cold anammox reached 40 nmol N/min mg protein or 34 mg N/g protein h. Assuming the protein content in activated sludge of 50% (Henze et al. 2000) the resulting anammox process rate would be 17 mg N/g VSS h. That is 40 times higher than in our work. This significant difference is possibly due to the different anammox bacteria content and activity in sludge. In cited work (Hu et al. 2013) fresh activated sludge was used containing up to 80% of anammox bacteria and it was incubated at 12 °C for 268 days, while in our case anammox bacteria content does not exceed 5%, the temperature was lower with a shorter incubation period.

The data obtained prove the possibility of ammonia oxidation process at temperatures below 10 °C. This opens up broad prospects for application of anoxic ammonium oxidation process for treatment of ammonia-rich waste water, i.e. reject waters of sludge treatment plants and industrial waste water in the areas with low average annual temperature.

  1. A possibility of anammox process implementation at temperatures below 10 °С has been shown.

  2. Ammonia aerobic oxidation rate reaches the value of 1.4 mg N/g VSS*h.

  3. Ammonium removal efficacy reaches 98%. The nitrifying bacteria doubling time at 5–10 °С was 45 days.

  4. Anammox rate reaches the value of 0.4 mg N/g VSS*h.

  5. Total mineral nitrogen removal efficacy reaches 60%.

  6. The obtained results forms a basis for nitrogen removal technologies developing from nitrogen-rich waste waters in the areas with low average annual temperatures.

The work was supported by the Ministry of Education and Sciences of the Russian Federation in the scope of Federal Targeted Programme ‘Research and Development in Priority Fields of S&T Complex of Russia in 2014–2020’, according to Subsidies Agreement no. 14.607.21.018 of June 5, 2014, project RFMFFI60714X0018.

Danilovich
D. A.
Kozlov
M. N.
Moyzhes
O. V.
Nikolaev
Yu A.
Kazakova
E. A.
2010
Anaerobic ammonia oxidation for nitrogen removal from concentrated waste water
.
Water Supply and Sanitary Technique
4
,
49
54
(in Russian)
.
Henze
M.
Harremoes
P.
La Cour Jansen
J.
Arvin
E.
2000
Wastewater Treatment. Biological and Chemical Processes
, 3rd edn.
IWA
,
London
.
Hramenkov
S. V.
Kozlov
M. N.
Kevbrina
M. V.
Dorofeev
A. G.
Kazakova
E. A.
Grachev
V. A.
Kuznetsov
B. B.
Polyakov
D. Y.
Nikolaev
Y. A.
2013
A novel bacterium carrying out anaerobic ammonium oxidation in a reactor for biological treatment of the filtrate of wastewater fermented sludge
.
Microbiology
82
(
5
),
628
636
.
Hu
Z.
Lotti
T.
de Kreuk
M.
Kleerebezem
R.
van Loosdrecht
M.
Kruit
J.
Jetten
M. S.
Kartal
B.
2013
Nitrogen removal by a nitritation-anammox bioreactor at low temperature
.
Applied Environmental Microbiology
79
(
8
),
2807
2812
.
doi: 10.1128/AEM.03987-12
.
Jetten
M. S. M.
Strous
M.
Van de Pas-Schoonen
K. T.
Schalk
J.
Van Dongen
U. G. J. M.
Van de Graaf
A. A.
Logemann
S.
Muyzer
G.
Van Loosdrecht
M. C. M.
Kuenen
J. G.
1999
The anaerobic oxidation of ammonium
.
FEMS Microbiology Reviews
22
,
421
437
.
Khar'kina
O. V.
Nikolaev
Yu A.
Dorofeev
A. G.
Kazakova
E. A.
2010
Nitrogen removal from return flows from digested sludge treatment facilities by the mean of nitri-denitrification without additional carbon source
.
Water Supply and Sanitary Technique
10
,
60
64
(in Russian)
.
Kozlov
M. N.
Kevbrina
M. V.
Nikolaev
Yu A.
Dorofeev
A. G.
Grachev
V. A.
Kazakova
E. A.
Aseeva
V. G.
Zharkov
A. V.
2014
One-reactor technology of nitrogen removal from waste waters
.
Water Supply and Sanitary Technique
5
,
54
63
(in Russian)
.
Standard Methods for the Examination of Water and Wastewater
2012
22nd edn.
American Public Health Association/American Water Works Association/Water Environment Federation
,
Washington DC, USA.
Van Hulle
S. W. H.
Vandeweyer
H. J. P.
Meesschaert
B. D.
Vanrolleghem
P. A.
Dejans
P.
Dumoulin
A.
2010
Engineering aspects and practical application of autotrophic nitrogen removal from nitrogen rich streams
.
Chemical Engineering Journal
162
,
1
20
.
Western Fertilizer Handbook, California Fertilizer Association
.
1995
Interstate Publishers Inc.
,
Danville. Illinois
.
Yang
J.
Zhang
L.
Hira
D.
Fukuzaki
Y.
Furukawa
K.
2011
High-rate nitrogen removal by the anammox process at ambient temperature
.
Bioresource Technology
102
(
2
),
672
676
.
doi: 10.1016/j.biortech.2010.08.039
.