Aleksandra Wur F

F

ACTORS INCREASING EFFICIENCY OF

DEAMMONIFICATION PROCESS FOR

NITROGEN REMOVAL FROM MAINSTREAM

WASTEWATER

Aleksandra Wur

© Aleksandra Wur 2014 Degree Project Masters Level

In association with the research group Water, Sewage and Waste Technology Division of Land and Water Resources Engineering

Royal Institute of Technology (KTH) SE-100 44 STOCKHOLM, Sweden

S

S

Allt eftersom kraven på renat avloppsvatten höjs så utvecklas nya reningsprocesser. Traditionell kväverening görs med nitrifikation och denitrifikation och genom att undersöka dessa metoder närmare både angående optimala parametrar och gränsvärden kan nya mer effektiva processer tas fram. En av dessa processer är Anammox, som kom fram för ungefär tjugo år sedan. Det behövs dock mer kunskap om processen för att den ska klara av kraven som ställs på storskalig vattenrening i dag. I detta examensarbete gjordes två satsvisa tester i labskala, detta för att undersöka vilka faktorer som leder till en ökad effektivisering av kvävereduktionen, i huvudströmmen av deammonifikations processen, med låg ammoniak koncentration och temperatur. I första testet undersöktes det vilka inhibitorer som hade störst påverkan på aktiviteten av de nitrat oxiderande bakterierna (NOB) i suspenderade slammet (detta då dessa bakterier inte gynnar deammonifikations processen). Bakterie aktiviteten analyserades genom upptagshasigheten av syre (OUR). Glödresten (VSS) för de olika inhibitorerna var också analyserat så att aktiviteten kunde relateras till gram (g) biomassa. De testade inhibitorerna var: fri salpetersyrlighet (FNA), fri ammoniak (FA), natrium klorat (NaClO3), friskt UASB (upflow anaerobic sludge blanket)

utflöde och myrsyra.

I andra testet undersöktes vilka faktorer som ledde till en minskning av NOB-bakterierna. De valda faktorerna var pH, löst syre (DO, dissolved oxygen), och total mängd ammoniak-kväve (TAN, total ammonia nitrogen). Dessa faktorer kunde lätt kontrolleras och användes för att göra en testserie med programmet MODDE. Det gjordes 34 tester med mikroorganismer som var placerade på Kaldnes ringar.

S

E

Requirements concerning the quality of treated wastewater are still grow-ing. Therefore, it is necessary to develop new purification processes. The most common method for nitrogen removal from wastewater is a tradi-tional nitrification - denitrification. Knowledge about application of this processes, the optimal parameters and limits allows creation of new, more efficient methods. One of them is the Anammox process, which was discovered about twenty years ago. Currently, it requires more de-tailed knowledge to be able to use it to mainstream wastewater and ob-tain satisfactory results in accordance with legal requirements.

In this work, two types of lab-scale batch tests were carried out in order to find factors for increasing nitrogen removal efficiency of deammonifi-cation mainstream wastewater treating process in conditions of low am-monia concentration and low temperature.

In the first batch test suspended sludge was used in tests and it was checked which inhibitor has the greatest impact on the nitrite oxidizing bacteria (NOB) activity (presence of NOB bacteria is not desired in de-ammonification process). The activity of various groups of bacteria was tested by the Oxygen Uptake Rate (OUR). In addition, the volatile sus-pended solids (VSS) was determined to relate the activity result to gram (g) of biomass. Inhibitors that have been tested are: free nitrous acid (FNA), free ammonia (FA), sodium chlorate (NaClO3), fresh UASB

ef-fluent (upflow anaerobic sludge blanket) and formic acid.

A

I would like to sincerely thank to my supervisor Professor Elżbieta Płaza for giving me the opportunity to participate in interesting project and to realize this master thesis. I appreciate your support and help during my stay in Stockholm.

I would like to thank to my adviser PhD Józef Trela for guidance and support.

I would like to express my gratitude to PhD Karol Trojanowicz and PhD Andriy Malovanyy for showing me how research work looks like and for supporting me in my work. Thank you for yours valuable and competent comments and for being friendly for me.

I am also grateful to Swedish Environmental Research Institute (IVL) workers and Master students: Susanna Berg and Vera Apostolopoulou. Thank you for great atmosphere at Hammarby Sjostadsverket, work with you was a pleasure.

Finally, I would like to thank to my parents, siblings, brother-in-law and friends for help, encouragement and support.

Last but not least, my boyfriend Łukasz – thank you for your love, pa-tience and just being with me.

A

Anammox Anaerobic ammonium oxidation

ATU Allylthiourea C4H8N2S

AOB Ammonium Oxidizing Bacteria

COD Chemical Oxygen Demand

Deammonification Partial nitritation/Anammox

DO Dissolved Oxygen

FA Free Ammonia

FNA Free Nitrous Acid

H Heterotrophs

HRT Hydraulic Retention Time

IFAS Integrated fixed-biofilm activated sludge

MBBR Moving Bed Biofilm Reactor

N2 Nitrogen gas

NOB Nitrite Oxidizing Bacteria

N-NH4 Nitrogen in Ammonium form

N-NO2 Nitrogen in Nitrtite form

N-NO3 Nitrogen in Nitrate form

OUR Oxygen Uptake Rate

R1 Pilot-plant scale Reactor 1

R2 Pilot-plant scale Reactor 2

SBR Sequence Batch Reactor

TAN Total Ammonia Nitrogen

TSS Total Suspended Solids

UASB Upflow anaerobic sludge blanket

T

C

Summary in Swedish iii

Summary in English v

Acknowledgments vii

Abbreviations and symbols ix

Table of Content xi

Abstract 1

1. Introduction 1

2. Traditional nitrification and denitrification 1

3. Anammox process 2

3.1. Different types of used reactors and biomass growth systems 2

3.2. Partial nitritation/Anammox process 3

3.3. Parameters influencing partial nitritation/Anammox process 6

4. Aim of study 7

5. Materials and methods 7

5.1. Hammarby Sjöstadsverket 7

5.2. Experimental setup at Hammarby Sjöstadsverket 7

5.3. Strategy 1 – inhibition of NOB 8

5.3.1. Batch test 8

5.3.2. Inhibitors used during experiments 9

5.3.3. Oxygen Uptake Rate tests (OUR) 9

5.3.4. Volatile Suspended Solids (VSS) 10

5.4. Strategy 2 – suppression of NOB 10

5.4.1. Experimental planning with MODDE 11

5.4.2. Batch tests 12

5.4.3. Chemical analyses 13

6. Results and discussion 13

6.1. Irreversible inhibition of NOB 13

6.1.1. Free nitrous acid 14

6.1.2. Free ammonia 15

6.1.3. NaClO3 15

6.1.4. Fresh UASB effluent 16

6.1.5. Formic acid 17

6.1.6. Summary of all inhibition tests 17

6.2. Preferential conditions for NOB suppression 19

6.2.1. Set of experiments built with MODDE program 19

6.2.2. Model validation 20

6.2.3. Scaled and Centered coefficients 21

6.2.4. Effect of ammonia nitrogen, DO and pH on NOB suppression 21

6.2.5. Summary of batch tests and modelling results 23

7. Conclusions 25

References 26

Appendix I – Oxygen Uptake Rate test data I

Appendix II – Data and results for VSS XX

Appendix III – Data and results for batch tests (preferential conditions for NOB

A

In recent years, the use of Anammox process for wastewater treatment has been thor-oughly investigated. Currently, a major challenge is to use this process for the main-stream.

The aim of this study is to find factors increasing efficiency of the deammonification process for nitrogen removal from mainstream wastewater in conditions of low am-monia concentration and low temperatures.

Two types of lab-scale batch tests were done and obtained results were analysed sepa-rately. In the first lab-scale batch test suspended sludge was used and series of OUR tests were carried out. Inhibitors used during experiments were: FNA, FA, NaClO3,

fresh UASB effluent and formic acid. The best results, after all tests obtained for using the free nitrous acid as an inhibitor. Results shows that NOB bacterial activity was in-hibited, while AOB activity was still high. The second type of lab-scale batch test was used to check interactions between factors which have impact for the NOB suppres-sion. Selected factors were: pH, DO and TAN and these factors were used to plan a series of experiments with MODDE application. After series of 34 experiments, re-sults showed that this method is not effective for low concentrations of TAN and another, more efficient strategy is needed. New strategy should reduce the NOB activ-ity or increase the activactiv-ity of Anammox.

It is difficult to find a good strategy to carry out this process because many factors are affecting it. Using the results, it is necessary to conduct further research, which will give indications to use the deammonification process for mainstream wastewater and will let to achieve good results.

Key words: partial nitritation/Anammox process; lab-scale reactors; batch tests; low temperatures; low ammonia concentration.

1. I

Removal of nitrogen compounds from wastewater is carried out mainly by biological methods, which use the processes of nitrification and deni-trification. However, microorganisms involved in these processes are sensitive to the inhibitory effect of various factors. Knowledge about the impact of various parameters on the bacteria activity is important to be able to improve processes and increase their productivity. The necessity to eliminate nitrogen compounds from wastewater is related to the re-quirements concerning the quality of treated wastewater.

Results of research on mainstream wastewater treatment with deammon-ification process show, that the competition between AOB and NOB for oxygen and competition between NOB bacteria that compete with anammox bacteria for NO2- are the main factor limiting reaching high

efficiency of nitrogen removal. This competition is very dependent on temperature. This task is even more challenging in one step systems where simultaneous nitritation/Anammox process is ongoing in the same reactor. Furthermore, AOB and Anammox bacteria activity signifi-cantly decreases with temperature drop under 25°C.

2. T

Nitrification is a two-step process, carried out under aerobic conditions. In the first stage, called nitritation, ammonium (NH4) is oxidized to

ni-trite (NO2) by ammonia oxidizing bacteria (AOB). Nitratation is a

sec-ond stage of the process and is csec-onducted by nitrite oxidizing bacteria (NOB). During this part nitrite is oxidized to nitrate (NO3).

2NH4+ + 3O2 → 2NO4- +4H+ + 2H2O (nitritation) (1)

2NO2- + O2 → 2NO3- (nitratation) (2)

Denitrification is carried out under anaerobic conditions by anaerobic microorganisms of the Pseudosomonas genus. These microorganisms utilize oxygen from the nitrate or nitrite respiration, and the contained in the wastewater organic compounds as food substances.

These processes require significant energy input for aeration during nitri-fication and supply of an external carbon source for denitrifying bacteria. Moreover, the legal standards defining the maximum allowable concen-tration of nitrogen in the effluent discharged into the receivers are be-coming more stringent. Consequently, new solutions are needed (Makuch, 2009).

3. A

The Anammox process is biological, autotrophic process carried out un-der anaerobic conditions. During the process there is a total conversion of ammonium ions to nitrogen gas without providing the organic carbon from external source (van de Graaf, 1996). Nitrite ions are electron ac-ceptors in the biological oxidation to nitrogen gas (Fig. 1).

The process is mostly applied in the wastewater treatment with high COD concentrations and high concentrations of ammonium nitrogen. Combination of the process of anaerobic wastewater treatment, wherein the organic matter is converted to methane is an effective and economi-cal solution for this type of wastewater (European Commision, 2004). The advantage is that the process does not require an external organic carbon source and CO2 emission into the atmosphere is lower than 85%,

compared to the nitrification and denitrification. Another advantage of this process is the lower energy requirement compared to traditional methods of nutrients removal by more than 35%. Using this process as a method of nitrogen compounds removal leads to significant reduction in operating costs. Anammox has a chance to completely replace the deni-trification and save half of costs of oxygenating wastewater during nitri-fication (Makuch, 2009). Process was discovered about twenty years ago and so far its use for wastewater streams with high temperatures and high concentrations of ammonia has been thoroughly investigated (Isaka et al, 2006; Vazquez-Padin et al, 2009; Dapena-Mora et al, 2004; Third et al, 2005).

In recent years, many researchers started work on improving this pro-cess, using it for different types of sewage and on its application at lower temperatures (Vazquez-Padin et al, 2011; Hendrickx et al. 2012, Szatkowska et al, 2007; Ma et al, 2013).

3.1. Different types of used reactors and biomass growth systems

Hendrickx applied 4.5 L gas-lift reactor for autotrophic nitrogen removal process. Reactor was inoculated with granular sludge and operated at 20°C during 253 days. Experiment results showed that growth of anammox bacteria is possible in this conditions and a nitrogen removal rate of 0.26 g N/(Lxd) was achieved (Hendrickx et al, 2012).

Vazquez-Padin used also the granular sludge for biomass growth and applied it to the SBR reactor. In the SBR reactor, CANON process was carried out at two different temperatures, influent and DO concentra-tions. In the reactor operated at 20°C influent was 225 mg N/L and DO concentration of 3.1 mgO2/L. Experiment was also conducted at 15°C

with 175 mgN/L in the influent and DO concentration of 2.2 mgO2/L.

Achieved nitrogen removal rate was 0.5 g N/(Lxd) and 0.2 g N/(Lxd), respectively (Vazquez-Padin et al, 2011).

Szatkowska conducted studies on partial nitrification and Anammox processes in one-stage system and maintained it at 20°C in the moving-bed biofilm reactor. Obtained nitrogen removal efficency was 46% (Szatkowska et al, 2007).

A similar removal efficiency of nitrogen obtained de Clippeleir et al. (2011). OLAND process for nitrogen removal was used and conducted at 25°C and at low HRT times. A short review of the literature concern-ing the different types of Anammox reactors operated at a temperature of 27°C or lower was done (Table 1).

3.2. Partial nitritation/Anammox process

Another name for partial nitritation/Anammox process is Deammonifi-cation. This is a two-step process, consisting of the partial nitritation and Anammox process. In the first step ammonium is oxidized to nitrite. Then, the Anammox bacteria under anaerobic conditions use ammoni-um and nitrite ions as substrates and produce nitrogen gas (Yang, 2012). NH4+ +0.75O2 +HCO3-→0.5NH4+ +0.5NO2- +CO2 +1.5H2O (3)

NH4+ +1.32NO2- +H+ →1.02N2+0.26NO3- +2H2O (4)

NH4+ +0.85O2 →0.44N2+0.11NO3- +1.43H2O+0.14H+ (5)

Combination of nitritation and Anammox process for nitrogen removal from wastewater allows for a significant costs reduction, compared to a conventionally nitrification and denitrification. Estimated costs of

Table 1 Overview of different types of anammox reactors operated at a temperature 27°C or lower (modified after Hendrickx et al, 2012). Biomass growth system Reactor type Type of wastewater Operation time d Volume l ( m3) T o C pH HRT h DO mg O2/l Influent mg N/L Loading rate g N/(L d) (g N/ m2d) Removal rate g N/(L d) [kg N/m3d] (g N/m2d1) Nitrogen removal efficien-cy % Reference granular

sludge air lift synthetic 253 4.5 20 7.5-8.2 5.3 0.0 69 0.31 0.26 84 Hendrickx et al, 2012 granular sludge SBR sludge digester supernatant 1120 1.5 20 7.5-8.2 6 3.1 225 0.9 0.5 56 Vazquez-Padin et al, 2011 15 7.5-8.2 6 2.2 175 0.7 0.2 29 activated

sludge UASB sewage 120 8.5 27 n.a. 4.6 n.a. 46 0.47 0.4 88 Ma et al, 2011

activated

sludge UASB synthetic 200 8 16 n.a. 0.28 3.7-6.5 n.a. n.a. [2.28] n.a. Ma et al, 2013

activated

sludge UFBBR sewage ~100 1

Table 1 Overview of different types of anammox reactors operated at a temperature 27°C or lower (modified after Hendrickx et al, 2012) - continued from previous page.

wastewater treatment can be reduced by about 60% due to the lower costs of aeration.

Additionally, there is no need to add an external carbon source due to the autotrophic nature of the process, and thus increase the low biomass. Moreover, application of this process allows to reduce CO2 emission

significantly (Miksch et al, 2010).

In the literature, combination of partial nitritation and Anammox can be described in different ways. The name depends if the process is carried out in one or two reactors (Table 2).

3.3. Parameters influencing partial nitritation/Anammox process

There are many factors affecting the course of the process and its effi-ciency. The most important of these include: temperature, pH, oxygen concentration, HRT, organic carbon, biomass concentration, suspended solids and other.

Highest efficiency of the Anammox process can be reached at tempera-tures between 20 - 43°C and the optimum temperature for the process is 37°C. At temperatures above 45°C biological activity of Anammox bacteria is inhibited. Lowering the temperature to 37°C does not restore the activity of the bacteria (Dosta et al, 2008). For better understanding of the effect of temperature on the process more and more experiments are carried out at temperatures lower than the optimum. Ma et al. (2013) used the 8.0 L UASB reactor and led the process by 200 days at 30°C and 16°C. Obtained results showed high nitrogen removal. At 30°C and HRT = 0.12 h nitrogen removal rate was 5.72 kg N/m3 _{d, while the}

temperature 16°C and HRT = 0.28 h nitrogen removal rate was 2.28 kg N/m3 _{d. Furthermore, emission of N2O from the reactor was}

very low.

The pH value suitable for the Anammox bacteria is in the range between 6.7 - 8.3 (with optimum value 8.0), (van Hulle et al, 2010).

Concentration of dissolved oxygen has a huge influence on the activity of AOB and NOB bacteria.

In partial nitritation/Anammox conducted in one reactor system AOB bacteria use DO for nitrite production. NO2-N are needed for

Anam-mox reaction and at low concentrations of DO nitrite deficiency can oc-cur. In contrast, high concentrations of DO to influence bacterial growth NOB activity. Furthermore, high DO concentrations cause bigger activi-ty of NOB bacteria which presence is not desired in this process.

Table 2 Different naming of partial nitritation /Anammox process (Cema, 2009).

Process name Number of

reac-tors Alternative proces name Two-reactor

nitritation/anammox 2

SHARON1 – anammox Two stage OLAND2 Two stage deammonifiation

one-reactor nitritation/anammox 1 Aerobic deammonification OLAND2 CANON3 deammonification SNAP4 DEMON5 DIB6

1 - Sustainable High rate Ammonium Removal Over Nitrite, 2 – Oxygen-Limited Autotrophic Nitrification Denitrification, 3 –Completely Autotrophic Nitrogen removal Over Nitrite, 4 –Single-stage Nitrogen removal using the Anammox and Partial nitritation, 5 – names only refers to the process in a SBR under pH-control,

4. A

This Master Thesis is a part of the “Deammonification project” and it is based on one-reactor partial nitritation/Anammox process.

The aim of this study was to find factors increasing efficiency of deam-monification process for nitrogen removal from mainstream wastewater in conditions of low ammonia concentration and low temperature (below 50 mgN-NH4/L and below 25°C, respectively). This aim can be

reached due two different applied types of batch tests: • Irreversible inhibition of NOB

• Providing preferential conditions for NOB suppression

To achieve the objectives during the experimental work it was necessary to:

• Review literature about deammonification process treating mainstream wastewater

• Review previous Master’s Thesis which were also a part of “Deammon-ification project”

• Find most promising inhibitor of all used for NOB bacteria in sus-pended sludge

• Find preferential conditions for NOB suppression (based on correla-tions between controlled variables such as TAN, pH and DO concentra-tion)

It was also necessary to obtain practical and theoretical knowledge about: • Chemical analysis

• Batch tests

• Calibrating and operating offline measurements devices • Preparing different chemical solutions

• Using MODDE application.

5. M

5.1. Hammarby Sjöstadsverket

Hammarby Sjöstadsverk is a research center in water purification tech-nology. It has been funded by the Stockholm Water AB and the Swedish government and has been officially opened in 2003. The research center is located in Stockholm, on top of Henriksdals wastewater treatment plant and is operated by Royal Institute of Technology (KTH) and IVL Svenska Miljöinstitutet (www.sjostad.ivl.se). In this research facility are conducted various studies concerning new technologies used in the puri-fication of wastewater. This place gives a lot of opportunities for stu-dents, doctoral, research workers, who have the possibility to extend their knowledge by working with qualified employees.

5.2. Experimental setup at Hammarby Sjöstadsverket

The experimental setup consists of two MBBR reactors (R1 and R2), in which the process is conducted in a variety of conditions, among others, at various temperatures (Fig. 2).

Reactors are equipped with on-line meters, such as conductivity, pH, re-dox, temperature and dissolved oxygen concentration. Except for dis-playing the current value of individual parameters on the display, these values are also transmitted and stored on hard drive. In addition both of the reactors are equipped with stirrers and aeration system.

5.3. Strategy 1 – inhibition of NOB

One of the possibilities of increasing ratio between AOB and NOB ac-tivity, which could be easily tested by OUR tests, is to use system with sludge recirculation and irreversibly inhibit NOB bacteria in suspended sludge (Fig. 3).

5.3.1. Batch test

Sludge from R1 was used in experiments. It was taken from settler and let sediment for about 60 minutes (Fig. 4). Then the liquid phase was taken away carefully and only the concentrated part was used for the tests. VSS of the sludge was determined to relate the activity result to g of biomass. Activity of AOB and NOB was determined with OUR tests. Liquid medium for the tests was based on effluent from R1. Big batch of effluent was collected (enough for all tests planned for the day). Ammo-nium was analysed and supplemented with NH4HCO3 to reach NH4-N

concentration of 50 mg/L and supplemented with NaNO2 to

concentra-tion of 15 mg NO2-N/L (“OUR” solution). pH of the liquid was in the

range of 7.7-7.8.

Test with every inhibitor consisted of 3 parts: a) Test of uninhibited biomass (just after collection)

b) Test of biomass after the period of contact with an inhibitor (inhibitor present during the test)

c) Test of biomass after period of inhibition, thoroughly washed in water There were also made several tests with storage of biomass for the same period of time as the inhibition period and OUR test was done after that (reference sample).

5.3.2. Inhibitors used during experiments

• FNA: artificial solution containing NH4Cl (500 mg NH4-N/L) and

NaNO2 (500 mg NO2-N/L). Big excess was prepared, divided into two

parts and adjusted pH with H2SO4 to the 6.0 and 6.5 value. Two 2.7 L

bottles with 300 mL of sludge and 2.4 L prepared inhibitor was left for overnight mixing. Sludge was in contact with inhibitor for about 17 hours.

• FA: big excess of supernatant was taken and pH was checked (pH=8.51). Bottle with 2.7 L volume with 500 mL of sludge and 2.2 L of supernatant was left for overnight mixing. Sludge was in contact with supernatant for about 18 hours.

• NaClO3: 300mL of sludge was used. 1.18mL of NaClO3 was added into

biomass and left for 10 minutes. After this time biomass was washed and tested.

• fresh UASB effluent: UASB effluent was taken. Bottle with 2.7 L vol-ume with 500 mL of sludge and 2.2 L of UASB effluent was left for overnight mixing. Sludge was in contact with UASB effluent for about 18 hours.

• formic acid: tests were performed for the three different concentrations of formic acid. Bottles with 2.7 L volume with 500 mL of sludge and 2.0 L of “OUR solution” was left for overnight mixing. Prepared sam-ples had the following concentrations: 200 mg/L, 100 mg/L and 50mg/L. The pH was also measured and obtained values: 5.98, 6.69, 7.06 for these concentrations. Sludge was in contact with formic acid for about 18 hours.

5.3.3. Oxygen Uptake Rate tests (OUR)

In order to test the activity of different types of bacteria in the suspend-ed sludge, series of OUR tests were carrisuspend-ed out. During measurement ac-tivity of the heterotrophic bacteria (responsible for denitrification), AOB (take part in the nitritation) and NOB (take part in the nitratation) can be checked. OUR test is based on measurements of the NH4+-N oxidation

rate. Inhibition of bacterial groups occurred by adding inhibitors such as NaClO3 and ATU (Surmacz – Górska et al, 1996).

For the test, three-necked bottle with a capacity of 1.27 L was used. Liquid medium was prepared from R1 outflow, transferred into the bot-tle and filled about 70-80% of the capacity. Then, the botbot-tle was placed on a magnetic stirrer set in a water bath. Water bath was operated at 25°C. After about 40 minutes the temperature in the bottle stabilized, oxygen concentration was measured. If the DO concentration was too low, sample was aerated to concentration about 9 mg/L. Afterwards a volume of suspended sludge was added. Liquid medium was also trans-ferred to the bottle so that no headspace remains. DO electrode (Hach Lange DO meter) inserted in the middle neck. Other two necks were sealed with a rubber stoppers with needle. 5 mL of NaClO3 (inhibition of

NOB) and 5mL of ATU (inhibition of AOB) were added through needle after 7-10 minutes and 7-11 minutes (Fig. 5). The whole test lasted about 15-30 minutes (duration time depended on used inhibitor). The oxygen concentration was measured every 10 seconds and the values were sent to the computer and saved as a text file by a special application (Fig. 6).

5.3.4. Volatile Suspended Solids (VSS)

VSS was determined to relate the activity result of different kinds of bac-teria to g of biomass. Total Suspended Solids (TSS) was also measured. TSS measurement result tells us about the amounts of the components retained on the filter during filtration, which means that they are larger than the pores in the filter used in the filtration. VSS measurement tells us about the organic matter, which was on the filter after ignition of the sample.

The dried filter was weighed with the aluminum plate. The same filter used for filtration of a volume of suspended sludge. After filtration, the filter was transferred back into the aluminum plate and placed for about one hour into the oven at 105°C.

After one hour, the aluminium plate with filter was weighed again. The next step was to place the weighted sample in the oven, this time at 550°C and leave it there for about 45 minutes. After this time the sample was weighed again. TSS and VSS calculations were based on two equa-tions.

TSS = (B – A)/(D * 1000) [g/L] (6)

VSS = (B – C)/(D * 1000) [g/L] (7)

A – weight of the filter and aluminium plate before filtration,

B – weight of the filter and aluminium plate after 60 minutes in the oven at 105°C [mg],

C - weight of the filter and aluminium plate after 45 minutes in the oven at 550°C [mg],

D – volume of the sludge used for filtration [L].

5.4. Strategy 2 – suppression of NOB

bacteria. One of this factors is FA concentration above level of 0.1 mg/L that inhibits growth of NOB at the same time do not suppress both Anammox and AOB bacteria. Concentration of FA depends on two factors: total ammonia concentration and pH of wastewater. There-fore by controlling pH in the reactor we can control FA concentration and at the same time induce suppression of NOB. Another parameter that can be used to outcompete NOB is concentration of DO. For the further research the following factors have been chosen:

- total ammonia concentration (TAN), - pH,

- DO concentration during aerated phase of intermittent aeration cycle.

5.4.1.Experimental planning with MODDE

MODDE is a dedicated software for using different experimental design methods (for example Central Composite Face method) and for results elaborating (Response Surface Modelling method). It enables to find in-teractions between different parameters of the system and to build em-pirical models of the system.

The first step before starting the series of batch tests was to plan their program. MODDE application, version 7.0 was used for this purpose. Work started with designing the set of batch tests experiments utilizing Project Design function of MODDE. DO concentration, TAN concen-tration and pH were defined as independent factors with minimum and maximum values and units of factor (Table 3).

The next step was to define response factor. In this work, it was defined as N-NO3r/N-NH4r, with units specified as %.

As the model type, response surface modelling (RSM) was selected. In the MODDE program there are different experiments to choose from in order to build the model. The more experiments we make, the more ac-curate model we get. CCF (Central Composite Face) model was used (with 3 central points).

Table 3 Values used to create a matrix in MODDE application.

-1 0 1

TAN [mgN/l] 5 25 45

pH [-] 7 7.8 8.6

DO concentration [mgO2/l] 1 1.5 2

In the next step the application creates a worksheet that contains a list of all the experiments. This list is made up of various combinations of fac-tors defined including their values. Received a list of 34 experiments that had to be done.

After conducting all the experiments, the worksheet has been completed with the obtained results. Afterwards model was fitted. In this work the Multiple Linear Regression (MLR) was used As a result of fitting, Plot Summary was displayed. It is a bar graph that shows how well the model has been fitted to data. The diagram consists of four bars. The first bar is the R2 _{and it shows how consistent are the results of the experiments}

with the results given by the model. Another bar - Q2 _{- shows the}

frac-tion of variafrac-tion of the response that can be predicted by the model. The values of R2 _{and Q}2 _{should be close to 1, then we are sure that the model}

is well fitted and has good predictions. The third bar is the Model Validi-ty (measures the validiValidi-ty of the model). The last bar named Reproducibil-ity measures the variation of the response under the same conditions as compared to the total variation of responses. Overall, all four bars should have a value close to 1 (the higher all bars, the better the model is).

Sometimes there are problems with the fitting of the model. This means that the results of the experiments are outliers (errors occurred during the experiment and had an impact on the obtained results). Outliers can be identified by the Normal Probability Plot of residuals. Response value identified in this way can be excluded from the model (model must be fitted again). After these steps, the model is ready to use. When we get a well-fitted model, we are able to calculate the response for each point with a defined range of factors that were used to build the model. Coefficient Plot was also used. This graph shows the coefficients for the selected response level of confidence. This results are displayed as an er-ror bars.

It is also possible to make plots which show the relationship between all of used factors and their influence on the response. You can select 2D or 3D graphs that clearly show a lot of information and make it easy to be understood (Malovanyy, 2009).

5.4.2. Batch tests

All experimental studies were conducted in 2 batch type reactors with volume of 1 L each. In the course of the tests reactors were intermittent-ly aerated with aquaristic pump (30 minutes of aerated phase and

135 minutes of non-aerated phase). About 400 mL of Kaldnes rings were collected from R2 and separated to the two batch reactors (Fig. 7). The supernatant was taken from R2 inlet and diluted to the previously set NH4-N concentration level and the pH adjusted/decreased to the set

level. Collected about 10 mL of the sample and filtrated it through a 0.45 µm syringe filters. First 2-3 mL of filtrate discarded. Analysed NH4-N , NO2-N, NO3-N, alkalinity at the beginning of the test. Batch

reactors were placed on magnetic stirrers in water bath and temperature was maintained at 16 - 17°C with use of ice cartridges inside water bath. Bottles filled to the cap with prepared supernatant. Afterwards remaining oxygen was removed by purging with N2 gas. DO meters were used to

control concentration of dissolved oxygen (Fig. 8). When the concentra-tion of DO decreased to about 0.05 mg/L and temperature stabilized six hour test was started. At the end of the last non-aerated phase, 10 mL of the sample was collected from each reactor and filtered using a syringe filters. Analysis NH4-N , NO2-N, NO3-N, alkalinity was done. The pH

concentration in both reactor was also measured. The entire measure-ment procedure was repeated for all controlled variables in accordance to the list of experiments from MODDE.

5.4.3. Chemical analyses

Chemical analysis were performed with standard spectrophotometric methods. Dr. Lange cuvettes were used to this purpose: NH4-N:

LCK303 (B#14023), NO2-N: LCK342 (B#13347), NO3-N: LCK340

(B#14023), Alkalinity: LCK362 (B#14008).

The filtered samples were used to prepare the cuvettes, following the manufacturer's instructions (Fig. 9). After the time required for the dis-coloration of samples, spectrophotometer used to read the obtained val-ues (Fig. 9).

6. R

In this chapter results will be presented and analysed. This section is also divided into two parts due to the applied two types of tests for increase the ratio between AOB and NOB activity.

6.1. Irreversible inhibition of NOB

During this study several OUR tests using different inhibitors were car-ried out. For the tests, were used: FNA, FA, NaClO3, fresh UASB

efflu-ent and formic acid. Graphs of individual tests and the table with the de-tailed results included at the end (Appendix I and II).

6.1.1.Free nitrous acid

Supernatant has ammonium concentration in the range of 800-1200 mg N/L and alkalinity molar ratio of around 1.The alkalinity content is enough for oxidizing roughly half of ammonium to nitrite. Consumption of alkalinity during ammonium oxidation leads to decrease of pH to around 6-6.5. In this study artificial solution containing NH4Cl

(500 mg NH4-N/L) and NaNO2 (500 mg NO2-N/L) was used with pH

set to 6 and 6.5. Due to mixing of this solution with sludge nitrite centration decreased and pH increased. During the first test nitrite con-centration was 444 mg NO2-N/L and pH=6.88, which corresponds to

FNA concentration of 153.4 μg N/L (temperature of mixed biomass: 19.2°C). During the second test nitrite concentration was 444 mg NO2-N/L and pH=6.64, which corresponds to FNA

concentra-tionof 265,2 μg HNO2-N/L (temperature of mixed biomass: 19.4°C).

After overnight mixing, in the bottle filled with artificial solution with pH=6.5, nitrite concentration did not change. pH decreased to 6.61 which corresponds to FNA concentration of 281,8 μg HNO2-N/L

(temperature: 19.7°C). In the bottle filled with artificial solution with pH=6.0, pH decreased to 6.52 which corresponds to FNA concentration of 344 μg HNO2-N/L (temperature: 20.0°C).

Results of tests during which the biomass was in contact with the inhibi-tor showed that activity of all types of bacteria was inhibited (Fig. 10). In both batch reactors NOB were a leading group, which is not desirable. After washing the biomass AOB bacteria regenerate its activity, which means that the inhibition is reversible. In contrast, the activity of NOB has been stopped. Comparing the biomass with different pH a slight dif-ference in activity of all types of bacteria can be seen. Test showed, that NOB can be inhibited with FNA concentrations between 281.8 μg HNO2-N/L and 344 μg HNO2-N/L with no effect on AOB.

At pH=6.0 FNA content is higher as compared with pH = 6.5 which is also reflected in minimally higher NOB inhibition.

Fig. 8 DO meters used during the tests.

6.1.2. Free ammonia

Supernatant has ammonium concentration 750 mg N/L and pH=8.51, which corresponds to FA concentration of 83.4 mg NH3-N/L

(temperature: 19.7°C).

OUR test with the use of biomass with inhibitor showed that AOB activity was significantly inhibited (Fig. 11). All night contact of biomass with the supernatant resulted in a growth of NOB activity and hetero-trophs. Heterotrophs can’t be so high at this FA concentration, the rea-son of this values are probably mistakes with VSS analysis. After washing the biomass, noted that AOB regenerated and it is the most active group in this test.

6.1.3.NaClO3

Sodium chlorate inhibits oxidation of NO2 to NO3 without negative

im-pact on the oxidation of NH4 to NO2. NaClO3 as an inhibitor has also

negative impact for AOB bacteria, when is in 15 minutes contact.

Ten minutes biomass contact with the inhibitor caused a decrease of the activity of all groups of bacteria (Fig. 12).

Fig. 10 Oxygen Uptake Rate tests results for FNA inhibition.

Despite the short contact with inhibitor, AOB activity occurred noticea-ble drop, which may be related with the presence of inhibitor during the washing of the biomass.

Biomass was washed with tap water twice (sedimentation took about an hour, then the liquid phase was withdrawn by syringe and the tube). After washing NOB recovered its activity and AOB’s activity percentage drop was significant compared to a representative sample. The obtained test results are not satisfactory. For better understanding the properties of inhibiting NaClO3 it is necessary to perform further experiments with

higher concentrations of the inhibitor.

6.1.4.Fresh UASB effluent

Concentration of sulfides (S2-_{) is 8.05 mg/L in fresh UASB effluent.}

Concentration in inflow to R1 is 2.7 mg/L, which is because of oxida-tion during storage period.

The test results showed no inhibition of AOB bacteria (Fig. 13). In the test with biomass with inhibitor, AOB activity increased slightly, but there was also an increase in the activity of the two other groups present in the biomass.

Fig. 12 Oxygen Uptake Rate tests results for NaClO3 inhibition.

After the last test with washed biomass can be seen a clear increase in the activity of AOB and NOB bacteria compared to the test with an un-inhibited biomass. Using a UASB effluent as an inhibitor does not give good results.

6.1.5.Formic acid

Tests on the activity of three groups of bacteria using formic acid as an inhibitor present in the biomass during the test showed that it affects all bacteria. There are visible differences in the various concentrations used. The largest decrease in activity was observed when 200 mg/L of formic acid was used (Fig. 14).

Eilersen et al. (1994) showed that the concentration higher than 100 mg/L completely inhibited the activity of NOB without having ef-fect on AOB. However, in this tests the concentration also afef-fects the AOB. After washing biomass, results showed that the biomass is regen-erated. There was a noticeable increase in activity of all bacteria. The predominant group in all assays was AOB bacteria, but at concentrations of 200 mg/L and 100 mg/L the activity was not significantly greater than NOB

bacteria. The increase in activity of bacteria may be due to errors in the VSS. The best results were obtained with washed biomass, which was inhibited with 50 mg/L of formic acid.

6.1.6.Summary of all inhibition tests

After all tests the best results are visible for tests with using the free ni-trous acid as an inhibitor.

It is clearly evident that NOB bacterial activity was inhibited, while AOB activity was high.

Ratio between AOB:NOB shows clearly that the best results were ob-tained for the washed biomass with pH = 6.0 (Table 4).

Use of ammonium or nitrite as inhibitors is very useful, because it is not necessary to use additional chemicals. It has been checked only in night-long contact with inhibitor.

Table 4 Results of all OUR tests. Name AOB [gO2/ gVSS*d] NOB [gO2/ gVSS*d] H [gO2 /gVSS*d] AOB:NOB F re e n it ro u s a c id Uninhibited biomass 0.290 0.191 0.194 1.52 Biomass without washing pH=6.5 0.012 0.037 0.016 0.32 Washed biomass pH=6.5 0.332 0.065 0.044 5.11 Biomass without washing pH=6.0 0.006 0.019 0.016 0.32 Washed biomass pH=6.0 0.277 0.034 0.038 8.15 F re e a m m o n ia Uninhibited biomass 0.368 0.131 0.341 2.81 Supernatant without washing 0.152 0.292 0.568 0.52 Washed supernatant 0.255 0.210 0.124 1.21 N a C lO 3 Uninhibited biomass 0.368 0.131 0.341 2.81 Washed biomass + NaClO3 0.210 0.105 0.085 2.00 F re s h U A S B e ff lu e n t Uninhibited biomass 0.368 0.131 0.341 2.81 UASB effluent without washing 0.625 0.762 1.106 0.82

Washed UASB effluent 0.636 0.449 0.383 1.42

Most of OUR tests were performed with three repetitions (some 1-2 only). If performed more repetitions, the results would be better, more accurate. In some cases, the activity of specific groups of bacteria in-creased comparing to the reference sample. This is probably due to an incorrect determination of VSS. The study is a preliminary test in order to check which of used substances most efficiently inhibits growth of NOB bacteria. For the tests with different substances as inhibitors, sys-tem with sludge recirculation was used and inhibition of NOB occurred in suspended sludge. Probably, this method of inhibition will be difficult to apply to Kaldnes rings, due to the fact that the growth of microorgan-isms occur in the biofilm. It is difficult to inhibit only the NOB bacteria growth without negative affect on the AOB and Anammox activity. It is required to conduct further tests that will be aimed at a closer examina-tion of how the contact time of the inhibitor affects the level of inhibi-tion, as well as check how long this inhibition will be maintained for. Selected inhibitor should also be biodegradable.

6.2. Preferential conditions for NOB suppression

During the research, a series of 34 batch tests was carried out in the fol-lowing planned experiments list from MODDE program. The variable values in the assays were: pH, concentration of DO and concentration of TAN. Detailed results are presented in tables (Appendix III).

6.2.1.Set of experiments built with MODDE program

According to the instructions in section 4.5.1. Project Design was run in MODDE application. Proposed experiments were made and obtained values of response were added into worksheet (Table 5).

Results of the batch tests are reliable. However, comparing them with pi-lot scale R2 current results, they had systematically higher values of the response parameter. In R2 at present at TAN=10 mgN/L, pH=8 and

Table 5 List of parameters of experiments with obtained values.

DO concentration during the aeration phase = 1 mgO2/L - the value of

the coefficient N-NO3r/N-NH4r is about 60%. The same values were

added to the MODDE application, the model predicted that the rate N-NO3r/N-NH4r is about 83%.Way of the aeration of batch reactors

had an impact for this results. Aeration during aeration phase was con-trolled manually. It was hard to keep it in the set value and the value had changed in the range of ±0.2 mgO2/L. Intermittent aeration and FA

concentration were used as a method of suppression of NOB bacteria. The non-aerated phase lasted 135 minutes, the dissolved oxygen concen-tration decreased with time. It was assumed that after 45 minutes of the duration of this phase, DO concentration will decrease to about 0.05 mgO2/L. It means, that the anoxic conditions should be created

(without the presence of DO), in which the NOB activity was inhibited. During the tests it was recorded that the concentration of DO in the non-aerated phase decreased to a maximum of 0.3 mgO2/L (at pilot

scale R2, DO concentration during non-aerated phase decrease to about 0.01 mgO2/L). Therefore, NOB had a substrate for the growth and their

residual activity still was present.

6.2.2. Model validation

Response surface methodology is useful to find interactions between selected, controlled variables and response variable. Standard response surface methodology equation with three general variables is the follow-ing: Y=54.0233–23.8877x1–2.70453x2+4.02343x3+11.3379x12 +4.21672x22_+6.82228x32_{–2.87484x1x2+2.93735x1x3} +0.0523376x2x3 (8) Y = (N-NO3r/N-NH4r) x1 = TAN, x2 = pH, x3 = DO

After fitting the model with use of MLR method system indicated exper-iments: N11, N16 and N22 as outliers. Further explanation of outliers is in chapter 4.5.1. These experiments were excluded from the data pro-cessed with MODDE and the model was fit again. This gave the Sum-mary Plot for the model (Fig. 15).

Values for bars in Summary Plot as follows: R2_{=0.92, Q}2_{=0.83, Model}

Validity=0.67, Reproducibility>0.89. The values of R2 _{and Q}2 _{are close}

to 1, so the model is well fitted to the data and is able to explain and predict (N-NO3r/N-NH4r)within the experiments with high precision.

Overall, all four bars have a high values (the higher all bars, the better the model is).

6.2.3. Scaled and Centered coefficients

After fitting the model the Coefficient Plot was screened in order to verify which factors influence the response N-NO3r/N-NH4r .

According to the Coefficient Plot (Fig. 16) the greatest impact on the re-sponse have: TAN concentration and DO concentration. The pH also affects N-NO3r/N-NH4r value, but to a lesser extent. Interaction

be-tween factors also influence the response. The relationship bebe-tween TAN and pH affects the concentration of FA, which is an inhibitor of NOB bacteria. Relationship between TAN and DO was also noted.

6.2.4. Effect of ammonia nitrogen, DO and pH on NOB suppression

The relationships between factors are also clearly visible in the graphs. The chart below shows the N-NO3r/N-NH4r ratio dependence on two

parameters: DO and TAN. Displayed charts for three different values of the pH concentration (Fig. 17).

Noticeable is higher inhibition of NOB for low DO concentration, high TAN values and high pH (with increasing concentration of ammonia

NOB inhibition is higher). The lowest value of the response

N-NO3r/N-NH4r obtained for: TAN=45 mgN/L, pH=8.6,

DO=1.0 mgO2/l and it is 39.9%.

Due to the fact that TAN concentration is one of the factors which had the greatest influence on the response, graph for dependence of N-NO3r/N-NH4r from TAN was drawn (Fig. 18). pH was fixed in the

highest value and DO concentration in the lowest.

It is clear that with increasing concentration of TAN with constant pH set at 8.6 and DO=1.0 mgO2/L, NOB activity is the lowest.

MODDE application allows also to display 3D charts. Example graph for dependence between TAN and DO at constant pH=7.8 for response (Fig. 19).

Another factor which had an impact on response is DO concentration. The chart below shows the N-NO3r/N-NH4r ratio dependence on two

parameters: pH and TAN. Displayed charts for three different values of the DO concentration.

In addition, NOB bacteria compete with AOB for dissolved oxygen, which is required for AOB bacteria in the production of nitrite. DO concentration has a significant effect on the activity of both bacterial groups. However growth rate of AOB is higher at low DO concentra-tions compared to NOB.

For a better presentation that DO concentration influences the response factor, graph for dependence of N-NO3r/N-NH4r from DO was drawn

(Fig. 21). pH and DO were fixed in the highest value.

It is noticeable that increasing of DO concentration during the aeration phase (at constant pH=8.6 and TAN=45 mgN/L), has no effect on the NOB suppression efficiency.

Relationship between pH and DO also affects N-NO3r/N-NH4r value,

but to a lesser extent. The chart below shows the response dependence on two parameters: pH and DO. Displayed charts for three different values of the TAN concentration (Fig. 22).

Fig. 17. Relationship between ratio of (N-NO3r/N-NH4r) and DO and

TAN: A) pH=7, B) pH=7.8, C) pH=8.6.

Fig. 18 (N-NO3r/N-NH4r) as a function of TAN with: pH fixed at 8.6 and

The method used for NOB suppression is not effective for lower con-centrations of ammonia nitrogen. With increasing concentration of am-monia nitrogen a noticeable reduction of NOB activity is visible. For the highest tested concentrations of TAN=45 mgN/L, the lowest DO=1.0 mgO2/L and the highest pH=8.6, NOB suppression is most

ef-fective.

6.2.5. Summary of batch tests and modelling results

The scope of this part of the research was to check the possibility of finding simple factors, which can be easily monitored and controlled during reactor operation and which have an impact on the NOB activity. The chosen factors were FA (dependent on TAN and pH) and DO.

Fig. 19 Relation between (N-NO3r/N-NH4r) – response factor

TAN and DO at pH=7.8.

MODDE program was used to design a series of tests which could check impact of DO, TAN and pH and their possible interactions on suppression of NOB. After all the experiments it was found that this method is not effective for low concentrations of TAN. Low concentra-tions of ammonia nitrogen must be maintained due to work-related re-quirements for wastewater treatment plants (TAN<10 mgN/L). There-fore, another strategy is needed. Strategy should reduce the NOB activity or increase the activity of Anammox . This can be achieved for example by bioaugumentation, which means supplementation of reactor with more active biomass or with biomass with different structure. Another possibility is to change the system to a system with suspended sludge re-circulation (IFAS). Finding most promising solutions is a challenge, but it is crucial for the application of Anammox to the mainstream. Further research is needed, probably with the use of more and different parame-ters impacting the process which can be easily controlled.

Fig. 22 Relationship between ratio of (N-NO3r/N-NH4r) and pH and DO:

A) TAN=5 mgN/L, B) TAN=25 mgN/L, C) TAN=45 mgN/L.

Fig. 21 (N-NO3r/N-NH4r) as a function of DO with: pH fixed at 8.6 and

7. C

Research work for this Master Thesis was conducted on the suspended sludge from R1 and on the Kaldnes rings from R2. Studies have been carried out during more than three months and two types of batch tests were done. All the calculations and analysis are based on own research. This allowed to formulate the following conclusions for the two types of performed tests:

Irreversible inhibition of NOB:

• Different inhibitors were tested, the highest NOB inhibition without strong effect for AOB activity were obtained for free nitrous acid used as an inhibitor

• It is difficult to inhibit only the NOB bacteria growth without negative affect on the AOB and Anammox activity

• OUR test is helpful tool for monitoring the evolution of NOB, AOB and Heterotrophs activity

• Further research work is required with checking how contact time in-fluences the level of inhibition and how long the inhibition effect is sus-tained for.

Providing preferential conditions for NOB suppression:

• A proper choice of intermittent aeration can be a limiting factor for NOB activity, increasing DO concentration during to the aeration phase has no advantageous effect on the NOB suppression efficiency

• Proposed method is not effective for lower concentrations of ammonia nitrogen (TAN= 5mgN/L)

• Knowledge about effect of the chosen parameters and their interactions on the process, gained with MODDE application may be useful to con-trol and operate the MBBR reactor

R

Cema G. 2009. Comparative study on different Anammox systems. TRITA-LWR PhD Thesis 1053, 84p.

Dapena-Mora A., Campos J.L., Mosquera-Corral A., Jetten M.S.M., Mendez R. 2004. Stability of the ANAMMOX process in a gas-lift re-actor and a SBR, Journal of Biotechnology 110:159-170.

De Clippeleir H., Yan X., Verstraete W., Vlaeminck S.E. 2011. OLAND is feasible to treat sewage-like nitrogen concentrations at low hydraulic residence times, Environmental Technology 90:1537–1545. De Clippeleir H., Vlaeminck S.E., De Wilde D., Daeninck K., Mosquera

M., Boeckx P., Verstraete W., Boon N. 2013. One-stage partial ni-tritation/anammox at 15 °C on pretreated sewage: feasibility demon-stration at lab-scale, Environmental Biotechnology, 97:10199–10210. Dosta J., Fernandez I., Vazquez-Padin J.R., Mosquera-Corral A.,

Campos J.L., Mata-Alvarez J., Mendez R. 2008. Short- and long-term effects of temperature on the Anammox process, Journal of Hazardous

Materials 154:688-693.

Eilersen A.M., Henze M., Kloft L. 1994. Effect of volatile fatty acid and trimethylamine on nitrification in activated sludge, Water Research 28(6):1329-1336.

European Commision, 2004. Water Cycle and Soil-related aspects: WATER TECHNOLOGIES: RESULTS AND OPORTUNITIES,

Global Change and Ecosystems European Communities, Belgium, 2004.

Gao D-W, Lu J-C, Liang H. 2014. Simultaneous energy recovery and au-totrophic nitrogen removal from sewage at moderately low tempera-tures, Environmental Biotechnology, 98:2637–2645.

Hendrickx T.L.G, Wang Y., Kampman C., Zeeman G., Temmink H., Buisman C.J.N. 2012. Autotrophic nitrogen removal from low strength waste water at low temperature, Water Research 46:2187-2193. Isaka K., Sumino T., Tsuneda S. 2006. High nitrogen removal

perfor-mance at moderately low temperaturę utilizing anaerobic ammonium oxidation reactions, Journal od Bioscience and Bioengineering 103(5):486-490.

Ma B., Zhang S., Zhang L., Peng Y., Wang J., Wang S., Wang S., Peng Y. 2011. The feasibility of using a two-stage autotrophic nitrogen removal process to treat sewage, Bioresource Technology 102:8331–8334. Ma B., Peng Y., Zhang S., Wang J., Gan Y. Chang J., Wang S., Wang S.,

Zhu G. 2013. Performance of anammox UASB reactor treating low strength wastewater under moderate and low temperatures, Bioresource

Technology 129:606–611.

Makuch A. 2009. Biologiczne usuwanie związków azotu ze ścieków za-wierających sulfonamidy, Rozprawa doktorska 210p.

Malovanyy A. 2009. Monitoring and application of Anammox process in one stage Deammonification system, TRITA-LWR Degree Project 09:20, 74p.

Miksch K., Sikora J. 2010. Biotechnologia ścieków, Wydawnictwo Naukowe

PWN, 237p.

Persson F., Sultana R., Suarez M., Hermansson M., Plaza E., Wilén B. 2014. Structure and composition of biofilm communities in a moving bed biofilm reactor for nitritation–anammox at low temperatures,

Surmacz-Gorska J, Gernaey K, Demuynck C, Vanrolleghem P, Ver-straete W. 1996. Nitrification monitoring in activated sludge by oxy-gen uptake rate (OUR) measurements, Pergamon 30:1228-1236. Szatkowska B., Cema G., Plaza E., Trela J., Hultman B. 2007. A

one-stage system with partial nitritation and Anammox processes in the moving-bed biofilm reactor, Water Science and Technology 55:19–26. Third K.A., Paxman J., Schmid M., Strous M., Jetten M.S.M.,

Cord-Ruwisch R.C. 2005. Enrichment of Anammox from Activated Sludge and Its Applicaion in the CANON Process, Microbial Ecology 49:236-244.

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.

Vazquez-Padin J.R., Fernandez I., Morales N., Campos J.L., Mosquera-Corral A., Mendez R. 2011. Autotrophic nitrogen removal at low temperature, Water Science and Technology 63(6):1282-1288.

Vazquez-Padin J.R., Figueroa I., Mosquera-Corral A., Campos J.L., Mendez R. 2009. Post-treatment of effluents from anaerobic digesters by the Anammox process, Water Science and Technology 60:1135-1143. Wójcik W. 2011. Evaluation of microbiological activity during the

de-ammonification process for nitrogen removal, TRITA-LWR Degree

Project 11:26, 95p.

A

I

–

O

U

R

Fig. 23 OUR results, Uninhibited biomass, 2014-03-10, Rep. I.

Table 6 Summarized OUR results, Uninhibited biomass, 2014-03-10.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-10 Uninhib-ited biomass -0.00187 -0.00143 -0.00056 -0.00087 -0.00044 1.98 2 -0.00206 -0.00141 -0.00055 -0.00086 -0.00065 1.32 3 -0.00188 -0.00133 -0.00056 -0.00077 -0.00055 1.40

Table 7 Summarized OUR results, Biomass without washing, pH=6.5, 2014-03-11.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-11 Biomass without washing pH=6.5 -0.00033 -0.00019 -0.0001 -0.00009 -0.00014 0.64 2 -0.00026 -0.00007 -0.00005 -0.00002 -0.00019 0.11

Fig. 26 OUR results, Biomass without washing, pH=6.5, 2014-03-11, Rep. I.

Table 8 Summarized OUR results, Biomass without washing, pH=6.0, 2014-03-11.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-11 Washed biomass pH=6.5 -0.00012 -0.00006 -0.00003 -0.00003 -0.00006 0.50 2 -0.0003 -0.00017 -0.00015 -0.00002 -0.00013 0.15

Fig. 28 OUR results, Biomass without washing, pH=6.0, 2014-03-11, Rep. I.

Table 9 Summarized OUR results, Washed biomass, pH=6.5, 2014-03-11.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-11 Washed biomass pH=6.5 -0.00166 -0.00138 -0.00015 -0.00123 -0.00028 4.39 2 -0.00147 -0.00129 -0.00016 -0.00113 -0.00018 6.28

Fig. 30 OUR results, Washed biomass, pH=6.5, 2014-03-11, Rep. I.

Table 10 Summarized OUR results, Washed biomass, pH=6.0, 2014-03-11.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-11 Washed biomass pH=6.0 -0.00107 -0.00101 -0.00012 -0.00089 -6E-05 14,83 2 -0.00116 -0.001 -0.00012 -0.00088 -0.00016 5.50

Fig. 32 OUR results, Washed biomass, pH=6.0, 2014-03-11, Rep. I.

Fig. 34 OUR results, Uninhibited biomass, 2014-03-13, Rep. I.

Fig. 35 OUR results, Uninhibited biomass, 2014-03-13, Rep. II.

Table 11 Summarized OUR results, Uninhibited biomass, 2014-03-13.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-13 Uninhib-ited biomass -0.00331 -0.00283 -0.0015 -0.00133 -0.00048 2.77 2 -0.00372 -0.00315 -0.00147 -0.00168 -0.00057 2.95 3 -0.00379 -0.00315 -0.00142 -0.00173 -0.00064 2.70

Fig. 37 OUR results, Washed biomass + NaClO3, 2014-03-13, Rep. I.

Table 12 Summarized OUR results, Washed biomass+ NaClO3,

2014-03-13.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-13 Biomass + NaClO3 (washed ) -0.00129 -0.00094 -0.00028 -0.00066 -0.00035 1.89 2 -0.00112 -0.00086 -0.00026 -0.0006 -0.00026 2.31 3 -0.00112 -0.0008 -0.00021 -0.00059 -0.00032 1.84

Fig. 39 OUR results, Washed biomass + NaClO3, 2014-03-13, Rep. III.

Table 13 Summarized OUR results, Supernatant without washing, 2014-03-14.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-14 Super-natant without washing -0.00154 -0.00123 -0.00097 -0.00026 -0.00031 0.84 2 -0.00168 -0.00117 -0.00093 -0.00024 -0.00051 0.47 3 -0.00156 -0.001 -0.00078 -0.00022 -0.00056 0.39

Fig. 41 OUR results, Supernatant without washing, 2014-03-14, Rep. II.

Fig. 43 OUR results, UASB effluent without washing, 2014-03-14, Rep. I.

Fig. 44 OUR results, UASB effluent without washing, 2014-03-14, Rep. II.

Table 14 Summarized OUR results, UASB effluent without washing, 2014-03-14.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-14 UASB effluent without washing -0.00433 -0.0028 -0.00205 -0.00075 -0.00153 0.49 2 -0.00327 -0.002 -0.00129 -0.00071 -0.00127 0.56 3 -0.00248 -0.00229 -0.0015 -0.00079 -0.00019 4.16

Fig. 46 OUR results, Washed supernatant, 2014-03-14, Rep. I.

Table 15 Summarized OUR results, Washed supernatant, 2014-03-14.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-14 Washed superna-tant -0.00086 -0.00058 -0.00018 -0.0004 -0.00028 1.43 2 -0.00099 -0.00061 -0.00021 -0.0004 -0.00038 1.05

Fig. 48 OUR results, Washed UASB effluent, 2014-03-14, Rep. I.

Table 16 Summarized OUR results, Washed UASB effluent, 2014-03-14.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-14 Washed UASB effluent -0.00108 -0.00078 -0.00023 -0.00055 -0.0003 1.83 2 -0.00109 -0.00077 -0.00022 -0.00055 -0.00032 1.72

Fig. 50 OUR results, Uninhibited biomass, 2014-03-17, Rep. I.

Table 17 Summarized OUR results, Uninhibited biomass, 2014-03-17.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-17 Uninhib-ited biomass -0.00336 -0.00227 -0.00141 -0.00086 -0.00109 0.79 2 -0.00316 -0.00262 -0.00102 -0.0016 -0.00054 2.96 3 -0.00298 -0.00256 -0.00118 -0.00138 -0.00042 3.29

Fig. 52 OUR results, Uninhibited biomass, 2014-03-17, Rep. III.

Table 18 Summarized OUR results, Biomass + formic acid (200 mg/l), 2014-03-18.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB

1 2014-03-18 Biomass + formic acid (200 mg/L) -0.00043 -0.00038 -0.00032 -0.00006 -0.00005 1.20

Fig. 54 OUR results, Washed biomass + formic acid (200 mg/l), 2014-03-18, Rep. I.

Table 19 Summarized OUR results, Washed biomass + formic acid (200 mg/l), 2014-03-18.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-18 Washed biomass + formic acid (200 mg/L) -0.00122 -0.00077 -0.00025 -0.00052 -0.00045 1.16 2 -0.00115 -0.00076 -0.00027 -0.00049 -0.00039 1.26

Table 20 Summarized OUR results, Biomass + formic acid (100 mg/l), 2014-03-18.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB

1 2014-03-18 Biomass + formic acid (100 mg/L) -0.00096 -0.00054 -0.00031 -0.00023 -0.00042 0.55

Table 21 Summarized OUR results, Washed biomass + formic acid (100 mg/l), 2014-03-18.

Date Name NOB+

AOB+H AOB+H H AOB NOB AOB:NOB 1 2014-03-18 Washed biomass + formic acid (100 mg/L) -0.00174 -0.00117 -0.00033 -0.00084 -0.00057 1.47 2 -0.00175 -0.00125 -0.00039 -0.00086 -0.0005 1.72

Fig. 57 OUR results, Washed biomass + formic acid (100 mg/l), 2014-03-18, Rep. I.