gcema PhD thesis corrected - DiVA portal290602/FULLTEXT01.pdf · TRITA-LWR PhD Thesis 1053 ISSN 1650-8602 ISRN KTH/LWR/PHD 1053-SE ISBN 978-91-7415-501-3 ... Thanks go also to master

TRITA-LWR PhD Thesis 1053 ISSN 1650-8602 ISRN KTH/LWR/PHD 1053-SE ISBN 978-91-7415-501-3

COMPARATIVE STUDY ON DIFFERENT

ANAMMOX SYSTEMS

Grzegorz Cema

October 2009

Grzegorz Cema TRITA LWR PhD Thesis 1053

ii

Comparative study on different Anammox systems

iii

iii

Dedicated to my parents Małgorzata and Paweł

Harvard Law:

Under the most rigorously controlled conditions of pressure, temperature, humidity, and other variables, the organism will do as it damn well pleases


iv


v

v

ACKNOWLEDGEMENTS

This thesis would not be possible without my colleagues and the support of my friends and fam-ily. I would like to thank to my supervisor Prof. Elżbieta Płaza for giving me the opportunity to join a special research team and to participate in the deammonification project. I appreciate your support, constructive advices and suggestions. Thank you for many ‘brain storm’ discussions. I would like to express my gratitude to my second supervisor Prof. Joanna Surmacz-Górska, for guiding me through the process of becoming a researcher. I would like to thank for the right amount of freedom, supporting me in my work and for finding her office door always open for any questions I had. Thank you, for many fruitful and stimulating discussions. I acknowledge Dr Józef Trela for the leadership of deammonification project and devoting lots of his time on discussion about the Anammox tests. I would like to sincerely thank to Prof. Bengt Hultmant for his guidance and support. I respect your huge knowledge and never ending ideas. Beata Szatkowska and Luiza Gut were two other Ph.D. students involved in the deammonifica-tion project and sharing with me experimental and analytical work. Beata, there is not enough space in this thesis to write down all the support and help I got from you in my time in Stock-holm. However, most of all thank you for your friendship. Luiza, I am indebted to you for intro-ducing me and explaining the research conducted in Stockholm. Thank you also for your friend-ship that developed during our Ph.D. studies. I also respect your professionalism. I would like to thank Ania Raszka for FISH analyses and the most important for her great friendship. I hope we can do a real good project together again and I hope for some mountains trips. Special thanks to Maja Długołęcka for many our helpful and stimulating discussions. Maja, you are also an excel-lent discussion partner over the everyday ‘cup of coffee’. Thanks go also to master students Giampaolo Mele, Aleksandra Pietrala, Anna Chomiak and Arkadiusz Stachurski, for their help in experiments. Giampaolo, you wrote that there was no need to mention that the “something of me” was in your thesis. I could say the same, but there is “something of you all” in this thesis and without your help, it would be impossible to finish it. You are also more friends than colleagues. Special thanks have to be done to Jan Bosander, Expert Process Engineer at the Himmerfjäden WWTP for all his help, continuous availability. His competence and great professionalism have been one of the best tools in the practical things during my research. He and the staff of the WWTP created an unforgettable working environment. Dr Ewa Zabłocka-Godlewska, I appreciate your support in microbiological analyses. Monica Löwén, Marek Tarłowski and Katarzyna Radziszewska, thanks for their aid in laboratory works. I am very grateful to Aira Saarelainen and Elżbieta Tarłowska for pleasant help with administra-tive matters. Jerzy Buczak thank you for the help with computer problems and being so friendly. Dr Lesław Płonka, thank you for your help with computer issues, but mainly for being a friend more than a colleague. Many people at the Environmental Biotechnology Department (EBD) deserve my gratitude. Hereby I would like to thank especially to: Ewa, Jarek, Dorota, Ola and Sebastian. Thank you for your friendship. Of course, I would like to thank you all the members of the EBD for friendliness. Most of all, my family deserves the biggest appreciation. To my Mother, for all her support, en-couragement and love. My brother Qba and his wife Ula for being always supportive. Qba, you are the best brother. To all my family for helping me every time I need it. Last but not least, my


vi

love Oleńka – thank you for your patience, support and understanding and just for being with me! This thesis was realized as a joint PhD study at Royal Institute of Technology (KTH), Stockholm, and Silesian University of Technology, Gliwice. In Poland: study was funded by Ministry of Science and Higher Education (Project reference number 1 T09D 030 30 - 0359/H03/2006/30). In Sweden: Financial support was obtained from VA-FORSK, SYVAB, J. Gust Richert Founda-tion and Lars Eric Lundbergs Foundation. The pilot plant was build by PURAC AB and has been operated with co-operation of the Royal Institute of Technology (KTH) and SYVAB. I wish to acknowledge all people, whom I might have not mentioned here and who have - either directly or indirectly – affected my professional life. Thank you

Stockholm, October 2009.


vii

vii

TABLE OF CONTENT

ACKNOWLEDGEMENTS ........................................................................................................................... V

ACRONYMS AND ABBREVIATIONS ....................................................................................................... IX

LIST OF PAPERS ......................................................................................................................................... XI

ABSTRACT ..................................................................................................................................................... 1

I – INTRODUCTION .................................................................................................................................... 2

II – BACKGROUND ...................................................................................................................................... 2

II-1. The Anammox process .................................................................................................................................. 2 Nitrogen cycle and the discovery of the Anammox process .............................................................................................................. 2 Brief process overview ..................................................................................................................................................................... 4 Application of the Anammox bacteria .......................................................................................................................................... 5 Summary....................................................................................................................................................................................... 9

II-2. Landfill leachate - characteristics and treatment methods for nitrogen elimination ........................................ 9 Landfill leachate generation ......................................................................................................................................................... 10 Landfill leachate composition ....................................................................................................................................................... 10 Landfill leachate treatment review – nitrogen removal .................................................................................................................. 11 Summary..................................................................................................................................................................................... 15

II-3. Reject water - characteristics and treatment method for nitrogen removal .................................................. 15 Chemical and physical methods .................................................................................................................................................... 16 Biological treatment ..................................................................................................................................................................... 17 Summary..................................................................................................................................................................................... 18

III – AIM OF THE THESIS ......................................................................................................................... 19

IV – MATERIALS AND METHODS ........................................................................................................... 19

IV-1. Membrane assisted bioreactor (MBR) ......................................................................................................... 19 Reactor operation ......................................................................................................................................................................... 19 Analytical procedure .................................................................................................................................................................... 20 Membrane cartridge ..................................................................................................................................................................... 21 Batch tests ................................................................................................................................................................................... 21

IV-2. Moving Bed Biofilm Reactor (MBBR) – Two step process ........................................................................ 21 Reactor operation ......................................................................................................................................................................... 21 Biofilm carrier material ............................................................................................................................................................... 22 Analytical procedure .................................................................................................................................................................... 23 Batch tests ................................................................................................................................................................................... 23

IV-3. Moving Bed Biofilm Reactor (MBBR) – one step process .......................................................................... 24 Reactor operation ......................................................................................................................................................................... 24 Determination of the dry weight of biomass developed on Kaldnes cerrier ....................................................................................... 25 Batch tests ................................................................................................................................................................................... 25 Oxygen uptake rates tests ............................................................................................................................................................ 25

IV-4. Rotating Biological Contactor (RBC) .......................................................................................................... 26 Reactor operation ......................................................................................................................................................................... 26 Analytical procedure .................................................................................................................................................................... 28 FISH – Fluorescent in situ Hybridization .................................................................................................................................. 28 Denitrifying bacteria analysis ....................................................................................................................................................... 29


viii

Batch tests ................................................................................................................................................................................... 29

V – RESULTS AND DISCUSSION ............................................................................................................. 29

V-1. Membrane Assisted Bioreactor (MBR) ......................................................................................................... 29 Process performance evaluation ..................................................................................................................................................... 29 Nitrogen conversion ..................................................................................................................................................................... 33

V-2. Moving Bed Biofilm Reactor – two step process ......................................................................................... 34 Process performance evaluation ..................................................................................................................................................... 34 Assessment of bacterial activity in biofilm and activated sludge ..................................................................................................... 37 Estimation of kinetic parameters ................................................................................................................................................. 39

V-3. Moving Bed Biofilm Reactor –from two-step towards one-step process ..................................................... 40 V-4. Moving Bed Biofilm Reactor – one-step process ......................................................................................... 41

Process performance evaluation ..................................................................................................................................................... 41 Influence of conditions in the pilot-plant on nitrogen removal dynamics ......................................................................................... 44 Dissolved oxygen influence on the nitrogen removal rate ................................................................................................................ 44 Evaluation of kinetic parameters ................................................................................................................................................. 45

V-5. Rotating Biological Reactor – two step process............................................................................................ 46 Process performance evaluation ..................................................................................................................................................... 46 Kinetic evaluation of process ......................................................................................................................................................... 49 Looking for bacteria populations ................................................................................................................................................. 49

V-6. Rotating Biological Reactor – one-step process ........................................................................................... 51 Process performance evaluation ..................................................................................................................................................... 51 Nitrogen conversion ..................................................................................................................................................................... 52 Kinetic evaluation of process ......................................................................................................................................................... 52 Looking for bacteria populations ................................................................................................................................................. 53

VI – SYSTEMS COMPARISON ................................................................................................................... 54

VII – CONCLUSIONS ................................................................................................................................. 58

Membrane assisted BioReactor (MBR) ....................................................................................................................................... 58 Moving Bed Biofilm Reactor (MBBR) – two-step process ............................................................................................................ 59 Moving Bed Biofilm Reactor (MBBR) – one-step process ............................................................................................................ 59 Rotating Biological Contactor (RBC) – two-step process .............................................................................................................. 60 Rotating Biological Contactor (RBC) – one-step process .............................................................................................................. 60 General ....................................................................................................................................................................................... 61

VIII – FUTURE RESEARCH ....................................................................................................................... 61

REFERENCES ............................................................................................................................................. 63


ix

ix

ACRONYMS AND ABBREVIATIONS

A – disc surface area, [m2] AAOB – anaerobic ammonium oxidizing bacteria Anammox – Anaerobic Ammonium Oxidation AOB – aerobic Ammonium Oxidizing Bacteria AOX – adsorbable organic halogen ATP – Adenosine5’-triphosphate ATU – allylthiourea B – dry weight of biomass developed on carriers, [mg d.w.] BABE – Bio Augmentation Batch Enhanced BOD – biochemical oxygen demand, [g O2 m-3] CANON – Completely Autotrophic Nitrogen Removal Over Nitrite COD – chemical oxygen demand, [g O2 m-3] d – mass of 50 kaldnes carriers after drying, [mg] d.w. – dry weight DEAMOX – DEnitrifying AMmonium OXidation DEMON – pH-controlled deammonification system, names only refers to the process in a

SBR denammox – DENitrification-anAMMOX process DIB – Deammonification in Internal-aerated Biofilm system DO – Dissolved oxygen, [g O2 m-3] Dwa – diffusivity coefficient of electron acceptor in water Dwd – diffusivity coefficient of electron donor in water e – mass of 50 kaldnes carriers after washing, [mg] ET – actual evaporative losses from the bare-soil/evapotranspiration losses from a

vegetated surface FISH – Fluorescent in situ Hybridization GFP – granular floating polystyrene H – heterotrophs HDPE – high density polyethylene HRT – hydraulic retention time K – proportionality coefficient KB – the saturation value constant, [g m-2d-1] KI – Haldane inhibition coefficient, [g m-3] KIA – Aiba inhibition coefficient, [g m-3] KM – Michaelis constant [g m-3] L – leachate production M – mol MAP – magnesium ammonium phosphate MBBR – Moving Bed Biofilm Reactor MBR – Membrane assisted BioReactor MLSS – Mixed Liquors Suspended Solids, [g l-1]


x

MSW – municipal solid wastes mwa – molecular weight of electron acceptor mwd – molecular weight of electron donor NO – nitrite oxide NOB – aerobic Nitrite Oxidizing Bacteria OLAND – Oxygen-Limited Autotrophic Nitrification Denitrification OUR – Oxygen Uptake Rate P – precipitation PBS – phosphate-buffered saline PFC – polyurethane foam cubes Q – inflow rate, [m3 d-1] r – substrate utilization rate, [g m-3d-1] R – surface run-off rA – substrate utilization rate, [g m-2d-1] RBC – Rotating Biological Contactor rpm – revolutions per minute S – substrate concentration, [g m-3] Sba – bulk liquid electron acceptor substrate concentration Sbd – bulk liquid electron donor substrate concentration SBR – sequencing batch reactor Se – effluent substrate concentration, [g m-3] Sharon – Single reactor system for High Ammonium Removal Over Nitrite Si – inflow substrate concentration, [g m-3] SNAP – Single-stage Nitrogen removal using the Anammox and Partial nitritation SS – Suspended Solids, [g l-1] TOC – total organic carbon, [g O2 m-3] UASB – Upflow Anaerobic Sludge Bed reactor VFA – volatile fatty acids Vmax – the maximum utilization rate constant, [g m-2d-1]; [g m-3d-1] VSS – Volatile Suspended Solids, [g l-1] WWTP – WasteWater Treatment Plant XOcs – xenobiotic organic compounds ΔUs – change in soil moisture storage ΔUw – change in moisture content of the refuse components υa – molar stoichiometric reaction coefficient for electron acceptor (moles) υd – molar stoichiometric reaction coefficient for electron donor (moles


xi

xi

LIST OF PAPERS

This thesis is based on the following papers, which are appended at the end of the thesis and referred to by their Roman numbers:

I. Cema G., Plaza E., Surmacz-Górska J., Trela J., Miksch K. (2005). Study on evaluation of kinetic parameters for Anammox process. In: Proceedings of the IWA Specialized Conference Nutrient Management in Wastewater Treatment Processes and Recycle Streams, Krakow Poland, 19-21 September 2005, 379-388.

II. Cema G., Szatkowska B., Plaza E., Trela J., Surmacz-Górska J. (2006). Nitrogen removal rates at a technical-scale pilot plant with the one-stage partial nitritation/Anammox process. Water Science and Technology, 54(8), 209-217.

III. Szatkowska B, Cema G., Plaza E., Trela J., Hultman B. (2007). One-stage system with partial nitritation and Anammox processes in moving-bed biofilm reactor. Water Science and Technology, 55(8-9), 19-26.

IV. Cema G, Płaza E., Trela J., Surmacz-Górska J., (2008). Dissolved oxygen as a factor in-fluencing nitrogen removal rates in a one-stage system with partial nitritation and Anam-mox process. In: Proceedings of the IWA Biofilm Technologies Conference, 8 – 10 January 2008, Singapore. Submitted for publication in Water Science and Technology.

V. Cema G., Pietrala A., Płaza E., Trela J., Surmacz-Górska J. (2009). Activity assessment and kinetic parameter estimation in single stage partial nitritation/Anammox. Submitted for publication in Journal of Hazardous Materials.

VI. Cema G., Wiszniowski J., Żabczyński S., Zabłocka-Godlewska E., Raszka A., Surmacz-Górska J. (2007). Biological nitrogen removal from landfill leachate by deammonification assisted by heterotrophic denitrification in a rotating biological contactor (RBC). Water Science and Technology, 55(8-9), 35-42.

VII. Cema G., Raszka A., Stachurski A., Kunda K., Surmacz-Górska J., Płaza E. (2009). A one-stage system with partial nitritation and Anammox processes in Rotating Biological Contactor (RBC) for treating landfill leachate. In: Proceedings of the IWA Conference Processes in Biofilms: Fundamentals to Applications, Davis, USA, 13-16 September 2009. Submitted for publication in Journal of Hazardous Materials.

Other publications related to this research not appended in the thesis: International journals/books

Cema G., Wiszniowski J., Żabczyński S., Zabłocka-Godlewska E., Raszka A., Surmacz-Górska J., Płaza E., (2008). Simultaneous nitrification, anammox and denitrification in aerobic rotating biological contactor (RBC) treating landfill leachate. In: Management of pollutants emission from landfills and sludge. Pawłowska & Pawłowski (eds). Taylor & Francis Group, London, 211-218. Conference publications

Żabczyński S., Raszka A., Cema G., Surmacz-Górska J. (2009). Nitrifiers populations and kinetic parameters analysis of membrane – assisted bioreactors. In: Proceedings of the IWA 2nd Specia-lized Conference Nutrient Management in Wastewater Treatment Processes. Kraków, Poland, 6-9 September 2009.


xii

Cema G., Płaza E., Surmacz-Górska J. and Trela J. (2005). Activated sludge and biofilm in the Anammox reactor – cooperation or competition? In: Integration and optimization of urban sanitation systems, Joint Polish-Swedish Seminars, Cracow, 2005, TRITA-LWR.REPORT 3018, 129 – 138. Cema G., Surmacz-Górska J. and Miksch K. (2004). Implementation of anammox process in the membrane assisted bioreactor. In: Integration and optimization of urban sanitation systems, Joint Polish-Swedish Seminars, Stockholm, 2005, TRITA-LWR.REPORT 3017, 81-92. Surmacz- Górska J., Cema G., and Miksch K. (2004). Deamonification process in membrane assisted bioreactors. In: Integration and optimization of urban sanitation systems, Joint Polish-Swedish Semi-nars, Wisła, 2003, TRITA-LWR.REPORT 3007, 81-91. Reports & compendia

Trela J., Płaza E., Hultman B., Cema G., Bosander J., Levlin E. (2008). Evaluation of one-stage deammonification. VA-Forskrapport nr 2008-18, (in Swedish) Trela J., Hultman B., Płaza E., Szatkowska B., Cema G., Gut L., Bossander J. (2006). Develop-ment of a basis for design, operation and process monitoring of deammonification at municipal wastewater treatment plants. VA-Forskrapport nr 2006-15, (in Swedish).


1

ABSTRACT

The legal requirements for wastewater discharge into environment, especially to zones exposed to eutrophication, lately became stricter. Nowadays wastewater treatment plants have to manage with the new rules and assure better biogenic elements’ removal, in comparison with the past. There are some well-known methods of diminishing concentrations of these compounds, but they are ineffective in case of nitrogen-rich streams, as landfill leachate or reject waters from dewatering of digested sludge. This wastewater disturbs conventional processes of nitrification-denitrification and raise necessity of building bigger tanks. The partial nitritation followed by Anaerobic Ammonium Oxidation (Anammox) process appear to be an excellent alternative for traditional nitrification/denitrification. The process was investigated in three different reactors – Membrane Bioreactor (MBR), Moving Bed Biofilm Reactor (MBBR) and Rotating Biological Contactor (RBC). The process was evaluated in two options: as a two-stage process performed in two separate reactors and as a one-stage process. The two-step process, in spite of very low ni-trogen removal rates, assured very high nitrogen removal efficiency, exceeding even 90% in case of the MBBR. However, obtained results revealed that the one-step system is a better option than the two-step system, no matter, what kind of nitrogen-rich stream is taken into consideration. Moreover, the one-step process was much less complicated in operation. Performed research confirmed a hypothesis, that the oxygen concentration in the bulk liquid and the nitrite produc-tion rate are the limiting factors for the Anammox reaction in a single reactor. In order to make a quick and simple determination of bacteria activity, the Oxygen Uptake Rate (OUR) tests were shown as an excellent tool for evaluation of the current bacteria activity reliably, and without a need of using expensive reagents. It was also shown, that partial nitritation/Anammox process, could be successfully applied at temperatures much lower than the optimum value. Performed Fluorescent in situ Hybridization (FISH) analyses, proved that the Anammox bacteria were mainly responsible for the nitrogen removal process.

Key words: Anammox; biofilm system; landfill leachate; nitrogen removal; reject water; removal rates


2

I – INTRODUCTION

Nowadays, water is one of the most precious components on earth. It is part of all living cells and is a key resource in society. Over 70% of our planet is covered by surface water. However, around 97% is comprised of salty water in the oceans. The rest is freshwa-ter. Most of this, about 69%, is locked up in the glaciers and icecaps. From the remaining freshwater, most occurs as a ground water and only about 0.3% is found as surface water in lakes and rivers. With growing peo-ple population and the same increasing hu-man consumption of water combined with increasing pollution of water sources, careful use, management, treatment and water reuse becomes therefore absolutely essential. One of the problems associated with fresh-water pollution is nutrients discharge into surface water causing acceleration of the eutrophication process. Although, natural eutrophication may take a thousand of years, due to human activity, this process is rapidly intensified by increasing aquatic plant nutri-ent inputs to water bodies. Thus, the term “eutrophication” has become synonymous with “excessive fertilization” or the input of sufficient amounts of aquatic plant nutrients, which causes the growth of excessive amounts of algae and/or aquatic macro-phytes in a water body (Petts, 2005). During last few decades, a huge numbers of land waters areas all over the world have been affected by eutrophication. It is also a serious problem concerning the Baltic Sea, which due to its special geographical and clima-tological characteristics is highly sensitive to the environmental impacts of human activi-ties. Hence, wastewater, in its untreated form, cannot be discharged directly into environ-ment and there is a need for its appropriate treatment. Nitrogen is essential for living organisms as a part of proteins; however, it is also one of the nutrients causing eutrophication problems. Nowadays, the biological methods are com-monly used for treatment of municipal wastewater and some industrial sewage. In most wastewater treatment plants (WWTP), nitrogen is removed by biological nitrifica-

tion/denitrification. These processes proceed well with typical municipal wastewater. Nev-ertheless, there are also nitrogen-rich waste-water streams like landfills leachate or reject waters from dewatering of digested sludge for which, traditional nitrifica-tion/denitrification can be generally ineffec-tive due to free ammonia inhibition and unfavourable biodegradable carbon content for denitrification. Because of high require-ments for oxygen and necessity of addition external carbon source, treating such nitro-gen-rich streams with traditional nitrifica-tion/denitrification would become expensive and not sustainable. Ammonia can be also removed by physical/ chemical processes. However, they have several disadvantages like odour production, air pollution or high cost of chemicals. Partial nitritation followed by Anaerobic Ammonium Oxidation (Anammox) may be an alternative for such streams. In the partial nitritation, only half of ammonium is con-verted to nitrite and then ammonium and nitrite are transformed into nitrogen gas in the Anammox reaction. The ammonium is oxidized under anoxic conditions with nitrite as electron acceptor. Hence, the combination of partial nitritation with the Anammox process results in reduction of energy con-sumption for aeration and additionally no external electron donor has to be added. For these reasons, it is a very interesting way of wastewater management with comparison to traditional nitrification/denitrification.

II – BACKGROUND

II-1. The Anammox process

As the Anammox process is object of study in this thesis, the introduction to the process and its characteristics organisms are de-scribed briefly in this chapter.

Nitrogen cycle and the discovery of the Anammox process In nature inorganic nitrogen atoms can exist in different oxidation states from -3 (NH4

+) to +5(NO3

-). Most of the nitrogen com-pounds representing these oxidation states can be converted to each other through mi-


3

crobial activity (Kartal, 2008). The turnover of nitrogen in biosphere is known as the nitrogen cycle (Fig. 1). Nitrogen oxidation state is changed by different microorganisms, that carry out catabolic reactions (nitritation, nitratation, denitrification, dissimilatory ni-trate reduction and Anaerobic Ammonium Oxidation (Anammox), anabolic reactions (ammonium uptake, assimilatory nitrate re-duction and nitrogen fixation), and ammoni-fication (Dapena Mora, 2007). In the beginning of the 20th century, most of reactions depicted in the N-cycle were al-ready known for a long time, and the N-cycle was assumed to be complete. In this com-plete N-cycle, there was no reaction account-ing for the possibility of the anaerobic oxida-tion of ammonium (Kartal, 2008). In 1977, Engelberd Broda used thermodynamic calcu-lation – standard free energy values of chemical reaction – to make a prediction on the existence of chemolitoautotrophic bacte-ria capable of oxidizing ammonium using nitrite as electron acceptor. These bacteria were known as “Lithotrophs missing from

nature”. Mulder et al. (1995) experimentally confirmed Broda’s prediction two decades later. However, they hypothesized that am-monium conversion was nitrate-depended. Van de Graaf et al. (1995) proved the biologi-cal character of the process. Van de Graaf and co-workers (1996) showed that presence of nitrite as electron acceptor is essential for Anammox activity and not nitrates as it was initially supposed. In 1999, Strous et al. (1999), basing on analysis of the 16S rRNA gene sequence, identified “missing lith-otrophs” as a new, autotrophic member of the order Planctomycetales. They were found in both wastewater treatment plants and natural systems (Table 1) (Zhang et al., 2007). Almost 30 – 50% of gaseous nitrogen production is attributed to the Anammox bacteria in nitro-gen cycle (Dalsgaard et al., 2005; Arrigo, 2005; Op den Camp et al., 2006). Their dis-tinct phenotypic characteristics involve red colour, budding production, crateriform structure on the cell surface, intracellular compartment “anammoxosome”, and intra-cytoplasmic membrane containing ladderane

Fig. 1. Scheme of the nitrogen cycle.

Norg.

NH3

NH2OH NO

NO2

NO3

N2

N2O

N2H4

Nitrificationn

Denitrification

Fixation

Anammox

Assimilation

Ammonification

Dissimilation

Genus Species Source Brocadia Candidatus Brocadia anammoxidans Wastewater Candidatus Brocadia fulgida Wastewater Kuenenia Candidatus Kuenenia stuttgartiensis Wastewater Scalindua Candidatus Scalindua brodae Wastewater Candidatus Scalindua wagneri Wastewater Candidatus Scalindua sorokinii Seawater Others Candidatus Jettenia asiatica Not reported Candidatus Anammoxoglobus propionicus Synthetic water

Table 1. Anammox bacteria discovered up-to-date (after Zhang et al., 2007).


4

lipid. As a special organelle in the cell, anammoxosome was considered to have three functions: (1) providing a place for catabolism; (2) generating energy for ATP synthesis through proton motive force across the anammoxsome membrane; (3) protecting the bacteria from the proton diffusion and intermediate toxicity due to their imperme-able membranes (Zhang et al., 2007).

Brief process overview In the Anammox process, ammonium is converted with nitrite as electron acceptor in a ratio 1:1.32, respectively, to dinitrogen gas (Strous et al., 1998) (eq. 1).

+−−+ +++ 0.13H0,066HCO1,32NONH 324

15,00.5232 NO0,066CH0,26NO1,02N ++→ −

O2.03H2+ (1) The main product of the anaerobic ammonia oxidation is dinitrogen gas, nevertheless around 10% of nitrogen in the influent is converted to nitrate nitrogen. General nitro-gen balance shows ammonium to nitrite to nitrate ratio of 1:1.32:0.26. The Anammox bacteria have very strong affinity for their substrates, ammonium and nitrite. The affin-ity constant values for ammonium and nitrite are below 5µM (Kartal et al., 2007). Substantial uncertainty exists on the interme-diates in the catabolism (van der Star, 2008). Based on the 15N-labelling experiments hy-drazine (N2H4) was identified as an interme-diate of the process. The occurrence of free hydrazine, a rocket fuel, in microbial nitrogen metabolism is rare, if not unique (Kartal e al., 2007). Van de Graaf et al. (1997) based on 15N-labbeling experiments proposed a possi-ble metabolic pathway for anaerobic ammo-

nium oxidation. Ammonium is oxidized by hydroxylamine (NH2OH) to form hydrazine. Reducing equivalents derived from N2H4 then reduce nitrite to form hydroxylamine and N2 (Fig. 2A). Nitrate formation could generate reducing equivalents for biomass growth. Strous et al. (2006) based on genomic analysis of Kuenenia Stuttgartiensis indicated that nitrite oxide (NO) could also be an in-termediate. According to this new pathway, nitrite was first reduced to nitric oxide; am-monium was then combined with NO to form hydrazine, which was later oxidized to dinitrogen gas (Kartal, 2008)(Fig. 2B). Differ-ent metabolic pathway was proposed by Kartal (2008). According to his suggestion Anammox catabolism starts with one elec-tron reduction of NO2

- to NO, this is poten-tially followed by a three electron reduction of NO to NH2OH. This step could be fol-lowed by the condensation of hydroxylamine and ammonia to form hydrazine and the to dinitrogen gas (Fig. 2C). Generally, nitrite is not directly converted to hydrazine but via hydroxylamine and/or nitrite oxide (van der Star, 2008). Figure 2 shows a schematic representation of three possible metabolic pathways. Recent studies showed that Anammox bacte-ria were capable of nitrate reduction with organic acids as electron donors and the same out-compete heterotrophic denitrifiers for these compounds. The end product of nitrate reduction by Anammox bacteria is dinitrogen gas. It was also showed that Anammox bacteria are also able to reduce nitrate to ammonium using organic acids as electron donor (Fig. 3). In this way, Anam-mox bacteria are capable of producing their

NO2ˉ

N2

N2H4

NH4+ NO

NO2ˉ

N2H4

NH4+ NH2OH

NO

NO2ˉ

N2

N2H4

NH4+ NH2OH

N2

A CB

Fig. 2. Different hypo-theses on the Anam-mox catabolic path-way. Additional potential intermediate are: A) hydroxylamine, B) nitric oxide, C) or hydroxylamine and nitric oxide (van der Star, 2008).


5

own ammonium (and nitrite) to perform their “standard” catabolism (Kartal et al., 2007; Kartal, 2008, van der Star, 2008). Egli et al. (2001) demonstrated that the Anammox bacteria were active within the range of temperatures from 6 to 43°C with the optimum at 37°C. For the optimal tem-perature, the pH range is between 6.5 and 8.5 (Gut, 2006). Inhibition studies showed, that Anammox bacteria are reversibly inhibited by very low levels (< 1µM) of oxygen concentra-tions and irreversibly inhibited by high nitrite concentrations (>10 µM). Egli et al. (2001) showed that Kuenenia Stuttgartiensis has a higher, but still low, tolerance to nitrite. When the nitrite concentration was more than 5 mM for a longer period (12h), the Anammox activity was completely lost. How-ever, the activity could be restored by addi-tion of trace amounts (±50 µM) of the Anammox intermediate, hydrazine (Li et al., 2004; Op den Camp et al., 2007; Kartal et al.,

2007). It was also demonstrated that com-mon denitrification substrates methanol and ethanol severely inhibited Anammox bacteria at a concentration below 1 mM. High salinity up to 30 g l-1 salt concentration cause reversi-ble inhibition (Kartal et al., 2007).

Application of the Anammox bacteria Operation and investment cost of wastewater treatment plant can be decreased by using innovative technologies based on new bio-logical conversion methods. Due to negative environmental aspects of nitrogen discharge to recipients and increasingly stringent efflu-ent standards, the effective nitrogen removal is necessary. Biological removal of high ni-trogen concentrations from wastewater is very expensive when there is a lack of biode-gradable organic carbon. Increasing require-ments concerning nitrogen concentration in treated wastewater and increasing cost of the treatment exert a necessity of development a new method for biological nitrogen removal. Recently, Anammox process was developed and proposed as a new technology for treat-ing streams containing high concentration of ammonia nitrogen and low concentration of organic carbon. However, the Anammox process requires nitrite as electron acceptor for anaerobic oxidation of ammonium, and for its application in wastewater treatment, different setups are used to provide nitrite: 1-reactor or 2-reactors systems. The common purpose in the application of all the systems is providing Anammox bacteria with nitrite (Kartal et al., 2007). Generally, part of am-

NH4+

NO

NO2

N2 N2O

NO3

N2

denitrification

dissimilatory nitrate reduction

Anammox

Fig. 3. Two possible routes of nitrate reduc-tion by Anammox bacteria (Kartal, 2008).

NH4+ + oxygen → NO2

-

NH4+ + NO2

- → N2

nitrifiers

Anammox One aerated reactor

Aerated reactor

Non aerated reactor

NH4+ + oxygen → NO2

-

NH4+ + NO2

- → N2

nitrifiers

Anammox

NO3- +sulfide/COD→ NO2

-

denitrifiers

NH4++ NO2

- →N2

Anammox

One non‐aerated reactor

A B C

Fig. 4. Simple scheme illustrating different Anammox configurations and different sources of nitrite: A) Nitritation and Anammox in Two-reactors in series, B) Nitritation and Anammox in one single reactor, C) Partial denitrification of nitrates to nitrites with the Anammox process in one non-aerated reactor.


6

monium is converted to nitrite and then the remaining ammonium and the formed nitrite is converted to dinitrogen gas by Anammox bacteria. Additionally, recently a new process was developed which combines the anaerobic ammonium oxidation with denitrifying condi-tions using sulphide as an electron donor for the production of nitrite from nitrate within anaerobic biofilm (Kalyuzhnyi et al., 2006). In Figure 4 there is shown the Anammox proc-ess in different configurations and different sources of nitrite. The processes of nitritation and Anammox were observed and studied in different con-figurations, types of reactors and under vari-ous conditions. Parallel research, performed by a few research groups in different coun-tries, has led to several names for processes where Anammox organisms play a major role (Table 2). This situation leads to an unclear terminology in the literature. Van der Star et al. (2007) proposed to clarify this situation by using the following descriptive terms:

• The anammox process for the anoxic con-version of ammonium and nitrite to dini-trogen gas.

• One-reactor nitritation-anammox process as the occurrence of the nitrite production and the anammox process in one reactor.

• Two-reactor nitritation-anammox process for the partial oxidation of ammonium to ni-trite in an aerated reactor, followed by an anoxic reactor, where only anammox process takes place.

• One-reactor denitrification-anammox process for the anoxic processes of denitrification from nitrate to nitrite, combined with the anammox process.

• The anammox reactor for the reactor in which only the anammox process takes place.

• Anammox organism: the dedicated organ-isms capable of performing the anammox process.

Two reactor nitritation-Anammox Partial nitritation/Anammox process is based on two processes. First assumed, that ammo-

Process name Number of reactors

Source of nitrite Alternative process names

Two-reactor Nitritation-anammox

2 NH4+ Nitritation SHARONa – anammox

Two stage OLANDb Two stage deammonifiation

One-reactor Nitritation-anammox

1 NH4+ Nitritation Aerobic deammonification

OLANDb CANONc Aerobic/anoxic deammonification deammonification SNAPd DEMONe DIBf

Two-reactor denitrification - anammox

1 NO3¯ Denitrification

DEAMOXg denammoxh

anammoxi a – acronym of Sustainable High rate Ammonium Removal Over Nitrite, b – acronym of Oxygen-Limited Autotrophic Nitrification Denitrification, c – acronym of Completely Autotrophic Nitrogen removal Over Nitrite, d – acronym of Single-stage Nitrogen removal using the Anammox and Partial nitritation, e – names only refers to the process in a SBR under pH-control, f - acronym of Deammonification in Internal-aerated Biofilm system, g – DEnitrifying AMmonium OXidation h – DENitrification-anAMMOX process i – System where anammox was found originally. Whole process was originally designated as “anammox”

Table 2. Process options and names for nitrogen removal systems involving the Anammox proc-ess (after van der Star et al., 2007)


7

nium is partly oxidized to nitrite in the partial nitritation stage and then nitrite react with remaining ammonium in the Anammox stage. The nitritation of ammonium to nitrite is conducted by aerobic ammonium oxidizing bacteria (AOB), a total nitrification should be avoided and the effluent should contain around 50% of the ammonium and 50% of nitrite. Different strategies can be used to selective retention of AOB bacteria in the system and to prevent further nitrite oxida-tion to nitrate by aerobic nitrite oxidizers, including the control of temperature, hydrau-lic retention time, alkalinity, the pH-value, dissolved oxygen concentration in the reactor as well as the amount of free ammonia (Paredes et al., 2007; Zhang et al., 2007). The combination of two processes - partial nitrita-tion and Anammox - in two reactors in series is illustrated in Figure 5. Van Dongen et al. (2001a) showed that the Sharon (Single reactor system for High Am-monium Removal Over Nitrite) process could be successfully combined with the Anammox, creating two-stage process for treating reject water originating from dewa-tering of digested sludge. When the Sharon reactor is used to provide the feed for the Anammox process, only 50% of the ammo-nium needs to be converted to nitrites. Since

sludge liquor generally contains enough alka-linity (in the form of bicarbonate) to com-pensate for the acid production if only 50% of the ammonium is oxidized. In this man-ner, exact ratio for full nitrogen removal in the Anammox process can be obtained (van Dongen et al., 2001a). One reactor Nitritation-Anammox The ability of bacterial cultures to create biofilm brings a possibility to enhance bio-logical wastewater treatment efficiency. Moreover, the ability of Anammox and Ni-trosomonas species to grow within the same biofilm layer enabled to design a one-stage system for nitrogen removal. Simultaneous performance of nitritation and Anammox processes can lead to a complete autotrophic nitrogen removal in one single reactor. In a one-stage process ammonium oxidizers in the outer layer of the biofilm can co-exist with the Anammox organisms present in the inner layer. In this way, oxygen that inhibits the Anammox process is consumed in the outer layer of the biofilm and Anammox bacteria are protected from oxygen. The combination of these two processes - partial nitritation and Anammox- in one reactor is illustrated in Fig. 6. Simultaneous nitritation and Anammox were observed and studied in various types of reactors under different conditions. Aeration devices and reactor configuration determine the transfer of air to the bulk phase. A trans-fer from the bulk phase over a boundary layer to the biofilm limits oxygen transfer to the bacteria. Also the limitation determined by hydrodynamics conditions is very impor-tant (van Hulle et al., 2003). In the moving bed bioreactor, the oxygen concentration has a great influence on the nitrification rate when the oxygen is rate-limiting (Hem et al., 1994). Also, it was proved that nitrite produc-tion rate is the rate-limiting step for the Anammox process in a single-stage system (Szatkowska et al., 2007). Intermittent aera-tion can also be used to secure a suitable ratio of oxygen and oxygen free conditions in the biofilm.

influent

Partial nitritation (Partial SHARON)

Anammox

effluent

Fig. 5. Scheme of the two-stage partial nitritation/Anammox process.


8

A combination of partial nitrita-tion/Anammox process can also be establish in one single reactor under oxygen limited conditions what is principle of the so-called CANON process. The CANON process was investigated as an alternative to use conven-tional activated sludge for treatment of wastewater limited by organic carbon sub-strate (Third, 2003). Appropriate ammonium and DO (dissolved oxygen) concentration enable the consumption of oxygen by AOB (aerobic Ammonium Oxidizing Bacteria) to an extent in which DO concentration is not over the threshold toxic to the Anammox bacteria. The oxygen-limited conditions be-low 0.5% air saturation provide an adequate environment on a stable interaction between Nitrosomonas-as aerobic microorganisms and Planctomycete-like anaerobic bacteria (Sliekers et al., 2002; Ahn, 2006). The growth of NOB (aerobic Nitrite Oxidizing Bacteria) (and subsequent nitrate production) is pre-vented due to their lower affinity for oxygen compared to AOB and for nitrite compared to Anammox bacteria. Subsequently, the produced nitrite, an inhibitor to AOB, is used as an electron acceptor by the Anammox bacteria (Zhang et al., 2007; Vázquez-Padín et al., 2008). The obtaining of the micro-aerobic

conditions for the CANON process can be achieved in different kind of systems like SBR and gas-lift (Vázquez-Padín et al., 2008). One-reactor denitrification-Anammox Recently, a new process was developed. It is a combination of the anaerobic ammonium oxidation with denitrifying conditions using sulphide as an electron donor for production of nitrite from nitrate within anaerobic biofilm (Kalyuzhnyi et al., 2006). The princi-pal flow diagram of this concept for treat-ment of high strength, strong nitrogenous and sulphate bearing wastewater is shown in Fig. 7. In the first stage of this process, anaerobic mineralization of organic nitrogen takes place. Next, the effluent from this reactor (rich in ammonia and sulphide) is partly fed to the nitrifying reactor to generate mainly nitrate and the rest directly to the DEAMMOX reactor. In the final DEAMOX stage, both flows are mixed together for consecutive realisation of nitrite production mainly from nitrate using sulphide as an electron donor and for the Anammox proc-ess (Kalyuzhnyi et al., 2006; Szatkowska, 2007).

Partial nitritation

wastewater

aerobiczone

anaerobic zone

carrier

Fig. 6. Scheme of one-step partial nitrita-tion/Anammox process within the biofilm.

Anaerobic

Reactor (AR)

Nitrifying Reactor

(NR) DEAMOX reactor

influent effluent NO3

- + (NO2-)

NH4+, (NS-)

Fig. 7. Flow diagram of the DEAMOX concept (Kalyuzhnyi et al., 2006).


9

Summary Application of the Anammox process in wastewater treatment can lead to significant reduction of operational costs. Compared to conventional nitrification-denitrification dependent nitrogen removal systems, the Anammox allows over 50% of the oxygen to be saved (only half of the ammonia has to be oxidized to nitrite instead of full oxidation to nitrates). Furthermore, because The Anam-mox is an autotrophic process, the problem regarding the supply of an electron donor (to support conventional denitrification) is cir-cumvented and no organic carbon source is needed. Additionally, Anammox bacteria oxidize ammonium under anoxic conditions with nitrite as the electron acceptor, and converse energy for CO2 fixation. This is in great concern because taxes on CO2 may even incur further significant cost in future if WWTPs are not excluded from this charge. Hence, the cost and CO2 emission are re-duced by 60% to 90%, respectively (Fux, 2003; Op den Camp et al., 2007; Kartal et al., 2007). The Anammox process is particularly suited for high nitrogen loaded industrial wastewa-ters that lack a carbon source. Different ammonium-rich streams (piggery manure, urine, digested fish canning effluents, tannery wastewater, landfill leachate, sludge liquors) have been studied with regard to the Anam-mox process application for its treatment. As the application of the Anammox process for landfill leachate and sludge liquors originating from dewatering of digested sludge, treat-ment is an objective of this study, the charac-teristics and different treatment methods for nitrogen elimination from these streams are introduced briefly here.

II-2. Landfill leachate - characteristics and treatment methods for nitrogen eli-mination

In most countries, sanitary landfilling is the most common way to eliminate municipal solid wastes (MSW) (Renou et al., 2008). Up to 95% of total MSW collected worldwide is disposed of in landfills (Kurniawan et al., 2006). In Europe, the number of permitted or legal landfills appears to have declined due

to implementation of the EU Landfill Direc-tive (1999/31/EC) (Kohler N. & Perry, 2005). Nevertheless, municipal waste landfill-ing is still a very important issue in the waste management system in Europe and the rest of the world. Waste disposal to landfills, in general, is an easy and low-cost waste management option but it raises environmental concerns. During the process of waste degradation, landfills produce waste products in three phases. These are solid (i.e., degraded waste); liquid (i.e., leachate, which is water polluted with wastes); and gas (usually referred to as landfill gas) (Butt et al., 2007). The major potential environmental impact related to landfill leachate generation is pollution of groundwa-ter and surface water (El-Fadel et al., 1997; Kjeldsen et al., 2002). In Poland due to stricter regulations (Act on Wastes of 27 April 2001 with following changes), which transposed EU legislation requirements on waste management into the Polish national legislation, the amount of deposited wastes has to be decreased. Im-plementing this legislation is connected with the necessity of taking up a number of im-portant actions, including limiting the amounts of biodegradable waste sent to the dumping sites. However, because of eco-nomical issues, landfills are the most attrac-tive disposal route for municipal solid waste in Poland (Wiszniowski 2006a). About 90% of municipal solid waste is currently disposed of in landfill sites (“Environment 2007” by Central Statistical Office). Many existing landfills are of an ageing design with no properly designed foundations, so the leachate can easily penetrate into the sur-rounding groundwater (Suchecka et al., 2006). The problem with landfill leachate produc-tion and management is one of the most important issues associated with the sanitary landfills. Environmental regulations require controlling the leachate level, which means that excess leachate must be removed and disposed of. Because of variable leachate composition from different landfills, leachate treatment methods have not been unified so far (Kulikowska and Klimiuk, 2007).


10

Landfill leachate generation Leachate is produced when water and/or other liquids seep through the wastes depos-ited in a landfill. Its production is the result of precipitation, surface runoff, infiltration, storage capacity, etc. (Heyer and Stegmann, 2002). Biochemical conditions, seasonal water regime of the landfill and changes in the solid waste composition affect both the quality and the quantity of these wastewaters (Gut, 2006). The water balance on the landfill site can be summarized as follows (Blakey, 1992): L = P – R – ΔUs − ET – ΔUw (2) Where: L – leachate production, P – precipitation, R – surface run-off, ΔUs – change in soil mois-ture storage, ET – actual evaporative losses from the bare-soil/evapotranspiration losses from a vegetated surface, ΔUw – change in moisture content of the refuse components. In addition, the climate has a great influence on leachate generation because the input of precipitation and loses through evaporation. Moreover, leachate production depends also on the nature of the wastes themselves (Re-nou et al., 2008).

Landfill leachate composition The landfill leachate is very high and complex polluted wastewater. The mixtures of high organic and inorganic contaminants may be found there as a result of biological, chemical and physical processes at landfills, which are combined with waste composition and land-fill water regime (Heyer and Stegmann, 2002; Poznyak et al., 2008). The composition of landfill leachate depends on many various factors like age of landfill, climate, nature of deposited wastes and also varies in composi-tion from site to site. Landfill leachate con-tains four main groups of compounds (Chris-tensen et al., 2001; Kjeldsen et al., 2002): • Dissolved organic matter – expressed as

Chemical Oxygen Demand (COD) or Total Organic Carbon (TOC), including CH4, volatile fatty acids and more refrac-tory compounds,

• Inorganic macrocomponents – Ca2+, Mg2+, Na+, K+, NH4

+, Fe2+, Mn2-, Cl-, SO4

2-, and HCO3-,

• Heavy metals – Cd, Cr, Cu Pb, Ni and Zn,

• Xenobiotic organic compounds (XOCs) – originating from households or indus-trial chemicals and present in relatively low concentrations in the leachate (usu-ally less than 1.0 g m-3 of individual com-pound). These compounds include, among others, a variety of aromatic hy-drocarbons, phenols, chlorinated aliphatic and adsorbable organic halogens (AOX).

Leachate composition may also be character-ized by different toxicity, determined using toxicological tests (Vibrio fischeri, Daphnia similes, Artemia salina etc.), which proved indirect information on the content of pol-lutants that may be harmful to a particular class of organisms (Kjeldsen et al., 2002; Renou et al., 2008). Toxicity is a consequence of contaminants mixture, their synergistic or antagonistic effects, and different physical-chemical properties, and toxicity tests may thus give more information about potential environmental impact than do chemical analyses alone (Marttinen et al., 2002). The toxicity tests have confirmed the potential dangers of landfill leachate and the necessity of treating it (Kjeldsen et al., 2002; Renou et al., 2008). There are many factors affecting the quality of the leachate, however among many others, the age of the landfill in particular influences the composition of the leachate (Renou et al., 2008). Data presented by Kulikowska and Klimiuk (2007) indicate that the landfill age has a significant effect especially on organic compounds and variation of these parame-ters with time may have important implica-tions in leachate management. Three types of leachate can be classified by landfill age: young, intermediate and stabilized (Amok-rane et al., 1997; Poznyak et al., 2008). Gener-ally, young landfills contain large amounts of readily biodegradable organic matters and as a result of rapid anaerobic fermentation of this matter leachate normally contains high concentration of volatile fatty acids (VFA).


11

With time, when landfill enters the methano-genic phase, the biodegradable fraction of organic pollutants decreases and the VFA are converted to biogas. Consequently, organic matter and BOD (Biochemical Oxygen De-mand) to COD ratio decreases significantly and the organic compounds are dominated by refractory compounds (Welander et al., 1998; Neczaj et al., 2007). In contrast, the concentration of ammonia does not decrease, and often constitutes a major long-term pollutant in leachate (Kjeldsten et al., 2002). Authors also suggested that neither heavy metals nor xenobiotic organic compounds, but ammonia would be the most concern in the long run as theory and model simulations show. The main sources of nitrogen are proteins, which accounts for approximately 0.5% of dry weight of municipal solid wastes. The hydrolysis of the polypeptide chains is disadvantaged in energetic terms and this is apparently the reason for the slow kinetics of protein hydrolysis, which in turn causes the slow release of ammonia. Nitrogen can trig-ger off eutrophization in receiving water-courses and therefore its removal from landfill leachate, e.g., by biological treatment, is required (Jokela et al., 2002).

Landfill leachate treatment review – nitrogen removal As mentioned above, landfills leachate char-acteristics depend on several factors, as type of wastes collected, seasonal variation of the precipitation, the age of landfill and others. These factors show the complexity of this wastewater and therefore indicate that there is no universal solution for its treatment. According to Renou et al. (2008), conven-tional landfill leachate treatment can be per-formed in three ways: • leachate transfer – recycling and com-

bined treatment with domestic sewage, • chemical and physical methods, • biological treatment – aerobic and an-

aerobic processes. Leachate transfer Recycling of the leachate back through the top has been largely used in the past decade, because it was one of the least expensive

options available. There are many advantages of operating a landfill as a bioreactor. The leachate recycling not only improves their quality, but also shortens the time required for waste stabilization. Among others, addi-tional advantages are: in situ leachate treat-ment and improvement of the landfill gas production rate, which may be favourable for energy recovery. This will tend to produce stabilized leachate containing relatively low concentration of biodegradable organic car-bon but high concentrations of ammonia and persistent organic compounds (Knox, 1985; Jianguo et al., 2007; Renou et al., 2008). How-ever, Price et al. (2003) showed that it is pos-sible to remove ammonia from leachate by ex-situ nitrification of ammonia followed by usage of the landfill as an anaerobic bioreac-tor for denitrification. Few years ago, the treatment of landfill leachate together with municipal wastewater was a common solution. However, this op-tion is not advised due to presence of organic inhibitory compounds and accumulation of hazardous compounds from the leachate, which consequently leads to reduce treatment efficiency and increase the effluent concen-tration (Waleander et al., 1998; Renou et al., 2008). Moreover, Aktas and Çeçen (2001) observed nitrification inhibition and nitrite accumulation to about 85 – 100% of the total NOx-N, when leachate was mixed with do-mestic wastewater. Chemical and physical methods Because of toxic nature of stabilized leachate, these effluents are difficult to deal with and biological processes are very inefficient. Therefore, alternative technologies based on physical-chemical stages are required (Rivas et al., 2004). These processes include reduction of suspended solids, colloidal particles, float-ing material, colour and toxic compounds by flotation, coagulation/flocculation, adsorp-tion, chemical oxidation and air stripping. Physical-chemical treatments for the landfill leachate are used in addition at the treatment line (pre-treatment or last purification) or to treat a specific pollutant (e.g. stripping - for ammonia removal) (Renou et al., 2008). Specifically, ammonia has been identified, as one of the major toxicants to microorganisms


12

in the treatment system, suggesting that pre-treatment prior to the biological treatment system is required to reduce the concentra-tion of NH4-N (Kim et al., 2007). Generally, it is a well known fact, that volatile fatty acid content decrease with landfill age and neither biological nitrification nor denitrification is not appropriate due to low COD to NH4-N ratio (the lack of sufficient electron donors in leachate and the high energy requirements for aeration) (He et al., 2007). The most common physical-chemical method for ammonia removal from leachate is air stripping which allows removing up to 93% of ammonia (Li et al., 1999; Marttinen et al., 2002; Renou et al., 2008). If this method is to be efficient, the medium needs to have high pH value and the contaminated gas phase must be treated with either H2SO4 or HCl. A major concern about ammonia air stripping is releasing NH3 into the atmos-phere, which causes severe air pollution if ammonia cannot be properly absorbed by neither H2SO4 nor HCl. Other drawbacks are the calcium carbonate scaling of the stripping tower, when lime is used for pH adjustment, and the problem of foaming which imposes to use a large stripping tower (Li et al., 1999). Additionally, since the leachate from an aged landfill contains high alkalinity just like a strong pH buffering system, the pH variation before and after stripping, will consume a large amount of alkali and acid (Li et al., 1999). Moreover, Marttinen et al. (2002) reported that in some cases stripping and ozonation increased toxicity in spite of COD and ammonia removal. This may be a result of oxidation of specific organic compounds to more toxic ones. This is on great impor-tance in case following biological treatment or discharges into environment. As an alternative to eliminate high level of NH4-N in leachate, the precipitation of NH4-N by forming magnesium ammonium phos-phate (MAP, struvite, MgNH4PO4·6H2O) can be applied. Kim and co-workers (2007) demonstrated that struvite precipitation is an excellent pre-treatment process. Formation of magnesium ammonium phosphate, a crystal with a solubility as low as 0.0023 g per 100 ml H2O, has been considered to be an

effective method for removal of ammonium, because of its high reaction rate and low residual ammonium concentration (Li et al., 1999). On the other hand, in spite of very high ammonia removal exceeding even 98%, struvite precipitation may be expensive due to high cost of chemicals, especially magne-sium chloride (Ozturk et al., 2003; Calli et al., 2005; He et al., 2007). However He et al. (2007) demonstrated that about 44% of chemical cost might be saved by using the MAP decomposition residues as the sole magnesium and phosphate sources. Addi-tionally, Li et al. (1999) pointed out that high salinity formed in the treated leachate during precipitation by using MgCl2·6H2O and NaHPO4·12H2O, which may affect microbial activity in the following biological processes. Other solution for ammonium removal from landfill leachate is ion exchange as an alterna-tive treatment option. The ion exchange is more competitive to other methods because of little influence of the low temperature. Clinoptilolite, one of natural zeolites, was found very effective in removing ammonia from water and wastewater (Wang et al., 2006). Zeolite is known to possess a higher selective ion-exchange capability for ammo-nium ion than Ca2+ and Mg2+, even when the concentration of the latter is higher than the former (Junga et al., 2004). Nevertheless, the presence of competitive ions such as K+, Na+, Ca2+ and Mg2+ in landfill leachate can reduce the ammonium adsorption capacity and increase equilibrium-making time. How-ever, experimental results indicate that am-monia can be removed by 84% from leachate using clinoptilolite as an ion exchanger (Kiet-lińska and Renman, 2005). Biological treatment In spite of stable treatment effects, and pref-erable adaptability to the changes of wastewa-ter quality and quantity, physical/chemical methods have several shortcomings: odour, air pollution, high chemical costs, high-energy consumption and excess sludge pro-duction (Bae et al., 1997; Liang and Liu, 2007). The main reason to select a biological process for nitrogen removal is the lower price compared to the physicochemical methods (Dapena Mora, 2007). The biologi-


13

cal nitrification/denitrification is probably the most efficient and the cheapest process to eliminate nitrogen from leachate. How-ever, specific toxic substances and/or pres-ence of bio-refractory organics can inhibit biological treatment (Wiszniowski et al., 2006b). Moreover, there are other problems associated with biological nitrogen removal. Young landfills contain higher concentration of biodegradable organic matter and ammo-nia, what is conductive factor for biological denitrification. On the other hand, the ma-ture landfill leachate contains relatively lower concentration of degradable organic material but higher concentrations of ammonia and the same the external carbon source addition is needed. The activity of nitrifying bacteria is a function of temperature, pH, ammonia concentration and nitrifying biomass concen-tration. The growth rate of nitrifying micro-organisms is also slow and might be inhibited by metals and hazardous materials. The low amount of bioavailable phosphorus in landfill leachate may limit the nitrification process or cause bulking sludge problems. Additionally, landfill sites exist in a wide range of envi-ronments, including areas with cool climate and achieving nitrification at low temperature requires large aeration basin volume or high biomass concentration in the aeration unit (Hoilijoki et al., 2000; Isaka et al., 2007). Ilies and Mavinič (2001) investigated the nitrification and denitrification processes at operating temperatures down to 10ºC in the activated sludge system. The nitrification process appeared to be unaffected (to any great extent) by a decrease in ambient tem-peratures to 14ºC. However, when tempera-ture dropped to 10ºC, the nitrification inten-sity drop between 10 and 30% was observed. Authors primarily attributed nitrification inhibition to low operating temperature, although they identified also other possible factors like too short aerobic hydraulic reten-tion time (HRT) and increasing level of ni-trous acid associated with low pH-value. But, Hoilijoki et al. (2000) proved that nitrification is feasible even at temperature as low as 10, 7 and even 5ºC. However, at 5ºC complete nitrification was obtained in the activated sludge system with addition of carrier mate-

rial, whereas activated sludge without carriers was able to remove only 61% of ammonium. Authors suggested that nitrifying microorgan-isms were attached to the carrier material and somehow stabilised the process at low tem-perature. Also Welander et al. (1997) con-ducted nitrification of landfill leachate at temperature of 10ºC, using plastic carrier material for biofilm growth. Authors proved that using suspended-carrier biofilm technol-ogy brings no risk for loss of biomass due to separability problems. Moreover, low tem-peratures have only weak negative effect on nitrification rate. They suggested that this phenomenon could be explained by oxygen diffusion into the biofilm, so the decreased specific reaction rate at lower temperatures is masked by deeper penetration of oxygen in the biofilm, resulting in larger mass of active nitrifiers. Also Jokela et al. (2002) confirmed that nitrification of landfill leachate at low temperatures at range of 10 and even 5ºC is possible using biofilm systems. The most important issue concerning N removal is to ensure appropriate C to N ratio. The optimum COD/NO3-N ratio for denitrification depends on the nature of the carbon source and on operating conditions of the system. Generally, the biological process is especially efficient in treatment young landfill leachate rich in volatile fatty acids (Wiszniowski et al., 2006b). For leachate characterized by BOD5/COD ratio above 0.5, Surmacz-Górska (2000) proposed three systems for biological treatment: membrane bioreactor, rotating biological contactor and system with pre-denitrification and microor-ganisms immobilized on suspended carrier. The best results were obtained in the system consisted of activated sludge with pre-denitrification and microorganisms immobi-lized on suspended carrier. However, the post-denitrification with the external source of organic carbon is recommended to re-move remaining nitrites and nitrates. Im and co-workers (2001) proposed anaerobic-aerobic system including simultaneous methanogenesis and denitrification to treat organic and nitrogen compounds in imma-ture leachate. The BOD5/COD and C/N ratio were 0.44 and 14, respectively and com-


14

plete N removal was achieved with raw land-fill leachate as a carbon source. Much more complicated is situation with mature leachate containing relatively low concentration of biodegradable organic material but high con-centration of ammonia. For that reason a supplementary source of organic carbon is needed (Wiszniowski et al., 2006b; Zhang et al., 2007). Kaczorek and Ledakowicz (2006) achieved the 99% removal of inorganic ni-trogen compounds using sodium acetate as external carbon source in two-sludge system with secondary denitrification. Ilies and Mav-inič (2001) used methanol as supplementary carbon source for denitrification in 4-stage Bardenpho system (without anaerobic stage). Additionally, the authors reported inhibition of denitrification when operating temperature dropped from 20 to 17ºC and finally to 10ºC. Welander and co-workers (1998) accom-plished denitrification using initially acetic acid and later methanol as external carbon source. The study was performed under realistic conditions (variations in temperature and leachate composition, etc.), and the tem-perature of the leachate varied between 10 and 26ºC. One of the main advantages of this system compared to activated sludge is weak temperature dependence. In addition, Louki-dou and Zoubolis (2001) performed denitrifi-cation using methanol and found out that the attached-growth biomass treatment method may be an interesting option comparing to the conventional activated sludge process. Due to shortage of organic carbon source for denitrification of mature landfill leachate and cost of external carbon source addition it was important to obtain technology that facili-tates enhanced nitrogen removal under the condition of low carbon source. Denitrifica-tion via nitrite instead of nitrate saved the oxygen requirement for nitrite oxidation and 40% of the carbon demand. A biodegradable COD to NO2-N ratio greater than 2.5 may ensure complete denitrification when short-cut nitrification takes place (Bae et al., 1997; Canziani et al., 2006; Zhang et al., 2007). The oxidation of ammonium to nitrite has usually been carried out using the Sharon technol-ogy. However, this technology could not be suitable for treatment of influents with vari-

able nitrogen loads, such as landfill leachate, because of possible stability problems when receiving ammonium loading shocks (Ganigué et al., 2007). Canziani and co-workers (2006) tested nitrogen removal from mature leachate by partial nitritation in mem-brane bioreactor and subsequent denitrifica-tion in a moving bed biofilm reactor. It was shown that a stable partial nitritation of am-monium to nitrite might be accomplished with temperature higher than 30ºC and free ammonia concentration higher than 2.5 g m-3. However, a low dissolved oxygen concentra-tion, kept under 0.5 g m-3, remains the key control parameter for partial nitritation. Un-der these conditions; it was possible to oxi-dize more than 85% of influent nitrogen to nitrite. Peng and co-workers (2008) reported other possibility for controlling partial nitrita-tion of ammonium to nitrite. They showed that the main factor achieving and maintain-ing nitritation is a proper range of free am-monia concentration obtained by dilution recycled final effluent. It inhibits nitrite oxi-dizing bacteria but not ammonium oxidizing bacteria. Isaka et al. (2007) observed a partial nitrification performed by nitrifying bacteria entrapped in a gel carrier. In the study, nitri-fication reactor was operated at high dis-solved oxygen concentration levels; low temperature, and infinite sludge retention time, therefore partial nitrification was possi-ble because of free ammonia inhibition. One more method for controlling the oxidation of ammonium to nitrite is presented by Ganigué and co-workers (2007). In their study, the ammonium conversion to nitrite was achieved and controlled by alkalinity avail-ability. Nevertheless, in this study occurred only partial ammonium oxidation to nitrite occurred, because the researchers wanted the Anammox process to take place. Due to the limitations of biological processes associated with low C/N ratio, high-energy consumption and unstable running, there was a need to search for new treatment methods. It appeared that the Anammox process could be a good alternative for traditional nitrifica-tion denitrification of mature landfill leachate by reducing oxygen and carbon source re-quirements and assuring high nitrogen re-


15

moval efficiency. However, in spite that the Anammox process was known in the mid nineties, there are still very few papers con-cerning the process for landfill leachate treatment. In the end of that decade, Hele-mer and Kunst (1998) observed a loss of inorganic nitrogen of up to 90% in the nitri-fication step of rotating biological contactor under low DO conditions. Moreover, Siegrist and co-workers (1998) also observed exten-sive loss of nitrogen (up 70% of ammonium oxidized) in a nitrifying rotating biological contactor treating ammonium rich landfill leachate. Authors suggested two hypotheses for this autotrophic nitrogen removal: auto-trophic denitrification by Nitrosomonas or by the Anammox process. Performed detailed analyzes of composition and spatial structure of the microbial community in the biofilm on the RBC (Egli et al., 2001; Egli et al., 2003) proved that the Anammox bacteria were exclusively present in the deeper part of the biofilm. Zhang and Zhou (2006) used a bench scale UASB reactor for removal of nitrogen from leachate by means of the Anammox process. The results showed that the average removal efficiencies of ammo-nium, nitrite and total nitrogen were 87.5%, 74.9% and 79.6% respectively, corresponding to the average ratio of removed nitrite-to-ammonium equal to 1.14 during the steady phase of the Anammox activity. Liang and Liu (2008) used bench scale up-flow fixed bed biofilm reactors for two-step partial nitritation and the Anammox process to remove nitrogen. About 60% of ammonium and 64% of nitrite nitrogen were simultane-ously removed in the Anammox reactor.

Summary It is important to note that the selection of the most suitable treatment methods for landfill leachate depends on their characteris-

tics, technical applicability and constraints, effluent discharge alternatives, cost-effectiveness, regulatory requirements and environmental impact. A combination of physicochemical and biological treatments is required to achieve effective removal of NH4–N and COD with a substantial amount of biodegradable organic matter. In most cases, physicochemical treatments are suit-able for pre-treatment of stabilized leachate to complement the biological degradation process (Kurniawan et al., 2006). In case of the Anammox process, there is a need to make more detailed research concerning on dependence the process on temperature decrease.

II-3. Reject water - characteristics and treatment method for nitrogen removal

Not only landfill leachate is with reason con-sidered as problematic. As it turns out, recy-cled streams within the wastewater treatment plant may also contribute to magnitude ni-trogen load in the inlet. Liquors arising from dewatering of sludge by belt presses, centri-fuges or alternative dewatering measures are referred to as sidestreams (Thornton et al., 2007). Usually during anaerobic sludge diges-tion, organic carbon is partially converted to methane gas, while about 50% of the nitro-gen, bound in the sludge, is released as am-monium (Siegrist, 1996; van Dongen et al., 2001a). This sludge liquor contains relatively high concentration of ammonium nitrogen (typically 200 – 700 g m-3) and a relatively low content of biodegradable organic matter. Additionally in the reject water the ratio of HCO3

-:NH4+ is normally equal to 1.1:1 (van

Dongen et al., 2001a; van Dongen et al., 2001b; Thorton et al., 2007). In Table 3 the example of average reject water composition is presented.


16

Despite of the fact that the volumetric su-pernatant flow is only around 2% of the total influent wastewater, contains up to 25% of the total nitrogen load into WWTP. Addi-tionally it is usually returned to the head of the sewage treatment works (Siegrist, 1996; Janus and van der Roest, 1997; Thornton et al., 2007; Dosta et al., 2007). This is especially dubious in case when the latter has limited aeration/nitrification/denitrification capacity. Conventional biological extension requires additional volume of aeration tanks and con-sequently a substantial investment (Janus and van der Roest, 1997; Volcke et al., 2006). In some cases enlarge or modifying the opera-tion of the mainstream processes already on site and this may be the most cost-effective options. However, some works may not have this option available and on sites, where sidestreams comprise a significant proportion of the total ammonium loading on the main-stream processes, a possible option is to treat directly the sidestream (Thornton et al., 2007). Generally, the chemical, physical and biologi-cal processes are feasible to recycle or elimi-nate ammonium from reject waters.

Chemical and physical methods One of the methods is precipitation of am-monium in the form of magnesium ammo-nium phosphate MgNH4PO4 (MAP) by addi-tion of phosphoric acid and magnesium oxide. The process is according to following reaction when the thermodynamic solubility product is exceeded (Siegrist, 1996; Çelen and Türker, 2001):

O6HPOMgNHO6HPONHMg

244

2344

2

⋅→→+++ −++

(3)

Çelen and Türker (2001) developed a two-step purification process to recover MAP

crystals from impurities of the anaerobic digester. First, precipitates were dissolved in acid and the pollutions were removed by centrifugation. The clarified supernatant was re-precipitated by adjusting its pH with caus-tic. It was shown that in the two steps proc-ess white MAP crystals could be obtained with over 85% recovery. However, to per-form precipitation the pH-value must be increased up to 9 usually with NaOH solu-tion, what generates additional cost. Gener-ally, struvite precipitation may be expensive because of high chemical cost, especially magnesium chloride (Ozturk et al., 2003). Besides this, sludge disposal costs also in-crease since use of MAP could increase the mass of dry solids for disposal by 50%. Addi-tionally, the lacks of reliable ammonia moni-tor for such dirty liquors make the process difficult to control (Jeavons et al., 1998). As an alternative to eliminate high level of NH4-N from reject water, there is used am-monia air or steam stripping. Ammonia in reject water is mainly present as ammonium ion. By raising the pH-value the ammonium is converted to ammonia, which is readily soluble in water. When contacted with a gaseous phase, the ammonia will be trans-ferred from water phase to the gaseous phase. The main difference between air and steam stripping is the treatment of the am-monia-rich gaseous phase (Janus and van der Roest, 1997). In order to achieve an increase of the reject water pH-value above 10, NaOH or Ca(OH)2 is added. Sludge flocks and precipitated as CaCO3 due to pH in-crease and have to be removed in a pre-sedimentation step (Siegrist, 1996). Siegrist (1996) demonstrated NH3 removal efficiency up to 97% at supernatant temperature be-tween 10 – 22°C. The greatest advantage of

Component Unit Values SS g m-3 675 COD g m-3 1500 – 2000 NH4

+-N g m-3 800 – 900 P-total g m-3 19.3 HCO3

- g m-3 3000 HCO3

-/N ratio mol HCO3- (mol NH4

+-N)-1 0.98 Temperature °C 35 pH - 8.2

Table 3. Example of average reject water composition (Dosta et al., 2007).


17

the ammonia stripping is relatively simple operation not affected by wastewater fluctua-tion if pH and temperature remain stable (Szatkowska, 2007). A major concern about ammonia air stripping is releasing NH3 as it was mentioned before. Other available method of treatment is ion exchange by materials selective to the ammo-nium ion. Most of the research undertaken has studied the effectiveness of naturally occurring materials, mainly zeolites. Today the most suitable zeolite for this process is clinoptilolite, a selective aluminosilicate of volcanic provenance, which shows an am-monium exchange capacity in the range 0.94 - 21.52 g NH4

+ N kg-1 (Thornton et al., 2007). Mackinnon et al. (2003) demonstrated that ion exchange using a new material, known by the name MesoLite, shows strong potential for the removal of ammonia from sludge liquors and an opportunity to concurrently reduce phosphate levels. MesoLite shows an ammonium exchange capacity in the range 45 - 55 g NH4

+ N kg-1. Using MesoLite materi-als, over 90% reduction of ammonia was achieved in the side stream. Thornton and co-workers (2007) confirmed that MesoLite is highly selective for the ammonium ion. Authors showed that over 95% of it was removed from belt press liquors with an initial ammonium nitrogen concentration exceeding 600 g m-3. The ion exchange is more competitive comparing to others meth-ods because of little dependency at the low temperatures.

Biological treatment Biological treatment is generally considered as the best available treatment of sludge liquors. However, high strength ammonia liquors are proved to be difficult to treat biologically, due to the toxicity of ammonia and its oxidation products at particular pH values. The alkalinity in digested sludge liquor is generally not sufficient to allow full nitrifi-cation and an additional alkalinity source needs to be added to maintain the pH (Jeavons et al., 1998). Siegrist (1996) investigated separate intermit-tent nitrification/denitrification of digester supernatant. Because degradable organic

content of the supernatant was low, therefore a carbon source (methanol) had to be added for appropriate denitrification performance. The process temperature was close to the inlet temperature of supernatant which was about 20 – 30 °C. Moreover, author stated that addition of excess sludge from super-natant treatment would enhance performance of nitrification in wastewater treatment in winter. Hwang et al. (2000) analysed two types of upflow biofilm reactor packed with granular floating polystyrene (GFP) and polyurethane foam cubes (PFC) to investigate the denitrification performance in reject waters. The results showed that very high denitrification rate was achieved in both types of biofilm reactor because the high concentration of volatile fatty acids contained in reject water provided the effective donors for denitrification. Operational costs of the biological nitrogen removal process stem from oxygen and or-ganic matter requirements for nitrification and denitrification, respectively. Due to the fact, that nitrite is intermediary compound in both steps: nitrification and denitrification, it would be convenient to procure a partial nitrification up to nitrite and then denitrifica-tion starting from nitrite. This approach will produce savings in the oxygen needs during nitrification, a reduction in the denitrification organic matter requirements, plus a decrease in surplus sludge production (Ciudad et al., 2004). Fux et al. (2004a) investigated nitrogen removal by nitritation followed by denitrifica-tion with carbon addition in the SBR reactor. They obtained efficient and robust nitrogen elimination at a total hydraulic retention time of 1 day via the nitrite pathway. The removal efficiency was amounted to 85 – 90% and ethanol was used as electron donor for deni-trification at a ratio of 2.2 gCOD g-1 N re-moved. Another way to eliminate nitrogen from the sludge liquor is the Sharon process. Due to a short retention time of biomass (approximately 1 day) and high temperature (35°C), the nitrite oxidisers are washed out and only nitrites are formed in the Sharon reactor. The process is performed without sludge retention. For oxidation of ammo-nium to nitrite, 25% less oxygen is necessary


18

than for the oxidation of ammonium to ni-trate. For denitrification of nitrite, 40% less methanol is needed than for denitrification of nitrate. Denitrification is used to control the pH-value (Hellinga et al., 1998; van Dongen et al., 2001a). The first full-scale application of the Sharon process has been constructed at the Rotterdam Dokhaven WWTP. It was possible to achieve an N-removal efficiency of 90%, and the process was stable despite load variations and some process distur-bances. The process requires high tempera-ture; however, heat production by biological conversion appeared to be significant, due to the high inlet concentration, and contributes to the optimal operating temperature of 30 - 40°C (Mulder et al., 2001). The BABE (Bio Augmentation Batch En-hanced) process is another technique for treatment of sludge liquor. The principle of this process is to implement nitrification reactor in the sludge return line, the so-called BABE reactor. It can be fed with an internal N-rich flow. Also, limited amount of acti-vated sludge from main process is recycled over the BABE reactor. Hence, the nitrifica-tion capacity of an activated sludge process can be augmented by the addition of nitrifiers cultivated in the BABE reactor. Practical results of the full-scale application of the BABE technology showed that the augmen-tation effect of applying the BABE process improved the nitrification rate of the sludge with almost 60% (Salem et al., 2003; Salem et al., 2004). Since the Anammox bacteria were discov-ered, a new alternative appeared for treat-ment of sludge liquor (Gut, 2006). In order to relieve the main wastewater treatment plant, reject water treatment with a combined Sharon-Anammox process seems to be a promising option. The Sharon process can be operated without denitrification, what is especially interesting in view of coupling the former with the Anammox process (van Dongen et al., 2001a). The simulation results indicate that significant improvements of the effluent quality of the main wastewater treatment plant can be realized. The best results were obtained by means of cascade feedback control of the Sharon effluent ni-

trite-to-ammonium ratio through setting an O2-set-point that is tracked by adjusting the air flow rate. Additionally, an economic analysis of operating cost index shows that implementation of this operation mode war-rants the associated investment costs (Volcke et al., 2005; Volcke et al., 2006). Van der Star et al. (2007) described the first full-scale Anammox reactor. The operation was com-pared with parameters previously reported in studies on laboratory scale. The maximum attained conversion of 9.5 kg N m-3d-1 was limited by the available influent load and was not a maximum volumetric conversion of the Anammox reactor. Rosenwinkel et al. (2005) investigated the simultaneous process in the Moving Bed Biofilm Reactor (MBBR) treating reject wa-ter. It was possible to remove up to 80% of the ammonia nitrogen load from sludge dewatering on Hattingen WWTP. The nitro-gen removal from reject water led to higher stability of the activated sludge tanks of the WWTP and reduced nitrogen load in the influent (Rosenwinkel and Cornelius, 2005). Recently, a full-scale application of partial nitritation/Anammox process was success-fully achieved in SBR reactor at the wastewa-ter treatment plant in Strass, Austria (In-nerebner et al., 2007). This pH-controlled system reached the designed elimination capacity of 300 kg N day-1 at the end of 2004. Both aerobic and anaerobic ammonia-oxidizing bacteria were enriched in the bio-mass and due to the high NH4-N/COD ratio of 2.5 to 3 in the influent flow, autotrophic processes were dominant. The process was operated in a SBR system providing intermit-tent aeration governed by three control mechanisms: time-, pH-, and dissolved oxy-gen (DO) control.

Summary Generally, separate sludge dewatering liquor treatment mostly requires new tanks or a comprehensive modification of available reactors. The overall investment costs are therefore significantly reduced if the separate treatment operates at high volumetric reac-tion rates. Furthermore, the dewatering liq-uor normally needs to be stored prior sepa-


19

rate treatment, as the sludge dewatering facili-ties are often operated intermittently (Fux et al., 2004a). With a wide variety of technolo-gies for reject water treatment, the question rises, which process is the best? For each treatment processes there is case-specific limitation, which sets the main goals for the side-stream treatment. Besides pure process engineering aspects, also other aspects such as: start-up time, risk of failure, flexibility and others have to be taken into account (van Loosdrecht and Salem, 2005).

III – AIM OF THE THESIS

Biological nitrogen removal used to purify wastewater with high ammonium content can become a major cost factor of wastewater treatment, in particular when the wastewater contains little amounts of biologically de-gradable carbon compounds. Moreover, the operational and investment cost can be saved with the use of new biological conversion methods and technology. Partial nitrita-tion/Anammox is an excellent example of such new process, where oxygen and aeration energy are largely reduced and no external carbon source is required. The partial nitrita-tion/Anammox process has the potential to be more cost-efficient than nitrifica-tion/denitrification. The process can be applied for treatment of ammonium-rich wastewaters such as landfill leachate or sludge liquor originating from dewatering of di-gested sludge. The aim of these studies was to investigate nitrogen removal from ammonium-rich streams with application of the Anammox process. The main goal of the investigations was to study the process performance and to esti-mate nitrogen removal rates in the partial nitritation/Anammox process applied in different systems. The process was estab-lished in three different reactors (Membrane assisted BioReactor, Moving Bed Biofilm Reactor and Rotating Biological Contactor). The main objectives of this study were: • To compare different systems applied

for the Anammox process,

• To establish potential process bottle-necks and influence of operation pa-rameters on process performance,

• Estimation of kinetic constants of nitro-gen removal process by usage of batch tests.

The more specific objectives of the research included: • studying the Anammox process in Mov-

ing Bed Biofilm Reactor treating reject water originating from dewatering of di-gested sludge, and to obtain long-lasting and stable Anammox process,

• studying the Anammox process in the Rotating Biological Contactor treating real landfill leachate, and to obtain long-lasting and stable Anammox process,

• finding correlations between the factors that influence the functioning of the Anammox bacteria and to determine suitable conditions in order to obtain maximum removal rates,

• defining the most important parameters influencing bacterial activity to oxidize ammonium with nitrite to nitrogen gas,

• investigating the optimal oxygen condi-tion for purpose of maximum nitrogen removal rate by batch tests application,

• determination of the respiration rate for the partial nitritation process by applica-tion of OUR (oxygen uptake rate) tests with addition of ATU (inhibition of ni-tritation step) and NaClO3 (inhibition of nitratation step),

• developing the Anammox process in the Membrane assisted BioReactor.

IV – MATERIALS AND METHODS

IV-1. Membrane assisted bioreactor (MBR)

Reactor operation The Anammox process for nitrogen removal from synthetic wastewater was investigated in the laboratory scale Membrane assisted Bio-Reactor (MBR). The liquid volume in the reactor was 0.036 m3 and the reactor was provided with synthetic wastewater by


20

peristaltic pump and the same way permeate was sucked out. The effluent pump was connected with flat sheet membrane cartridge (Fig. 8). Synthetic medium was made of tap water and additionally adjusted by NaHCO3, Na2HPO4, NaNO2 and NH4Cl in order to reach nitrite-to-ammonium ratio of 1.32:1 (Strous et al. 1998). To check the reactor’s management with virtual medium, at the end of the ex-periment the landfill leachate from the mu-nicipal landfill in Gliwice (Poland) was added to the synthetic medium up to 10% of its volume. The reactor was equipped with a heater to keep temperature stable above 30ºC, and with vertical stirrer to ensure proper mixing. During the research period, the Hydraulic Retention Time (HRT) and the synthetic

medium composition were gradually changed. The scheme of the MBR is shown in Figure 8, while the changes in operational conditions of the MBR are listed in Table 4.

Analytical procedure During the research period, the samples were taken from influent, effluent and the mixed liquor three times a week. The efficiency of the biological treatment was followed in terms of the general parameters such as: COD (dichromate method), ammonia (Kjeltec System 1026 Tecator), nitrite and nitrate (colorimetric method). Moreover, the process was monitored by measurements of the following parameters: flow rate, pH, temperature, dissolved oxygen and the bio-mass concentration in the bulk liquid. The free ammonia and free nitrous acid concen-trations were calculated according to An-thonisen et al. (1976). Additionally, respiratory activity of the first and the second stage of the nitrification were measured as described by Surmacz-Górska et al. (1996). Mensuration of dissolved oxygen uptake by the bacterial culture and during the subsequent addition of the selective inhibi-tors was conducted (Fig. 9). A completely closed reactor vessel (volume = 110 ml) was used for the measurements. The reactor was equipped with oxygen electrode connected with the recorder. Mixed liquors samples were collected from the MBR reac-tor, aerated by shaking the sample, and trans-ferred to the vessel, which was carefully closed, with no remaining air bubbles. During the measurement, samples were mixed using a magnetic stirrer.

Fig. 8. Scheme of the investigated MBR system.

Parameter Unit value Reactor volume Membrane cartridge surface

m3

m2 0.036 0.1

Flow rate m3 d-1 0.009 – 0.023 Hydraulic retention time (HRT) d 4 – 1.52 COD0 g O2 m-3 4.9 – 60 NH4

+ - N0 g m-3 17.8 – 74.5 NO2

- - N0 g m-3 21.8 – 98.1 Biomass concentration g MLSS l-1 3.5 – 12.4 Temperature ºC 33.8 ± 0.6 pH - 7.9 ± 0.1

Table 4. Operation conditions of the MBR.


21

Sodium chlorate (2ml of 12% NaClO3 solu-tion) was used as an inhibitor of the nitrite oxidizers whereas allylthiourea (2ml of 0.028% ATU solution) as an inhibitor of ammonium oxidizing bacteria. The respira-tory activity of heterotrophs was also calcu-lated as the remaining oxygen uptake after additions of inhibitors.

Membrane cartridge KUBOTA membrane cartridge type 203 was used. Membrane sheets are ultrasonic-welded on both surfaces of the membrane panel. They are made of chlorinated polyethylene with nominal 0.4 µm pore size. Effective surface area was equal to 0.1 m2. Between the panel and the membrane sheets, a spacer is laid to distribute filtered liquid into series of channels that lead to a nozzle on top of the cartridge. Moreover, the spacer prevents membrane sheets from sticking onto mem-brane panel (Shino et al., 2004).

Batch tests Two batch tests were performed during the research period in order to confirm presence of the Anammox process in the reactor and

to determine nitrogen removal rates. Tests were performed in 2 L reactor. Activated sludge from membrane assisted bioreactor was stirred for 24 hours without aeration and feeding in order to remove substrate and starving the sludge. The water bath was used to keep the same temperature like in the reactor. After 24 hours, sludge was thickened to 0.2 L and then reactor was filled with synthetic feeding media up to 2 L. Collected samples were immediately analysed for inor-ganic nitrogen compounds. Additionally, temperature, dissolved oxygen and pH were measured during the batch tests.

IV-2. Moving Bed Biofilm Reactor (MBBR) – Two step process

Reactor operation The Moving Bed Biofilm Reactor (MBBR) was constructed by the PURAC Company and was located at the Himmerfjärden Waste Water Treatment Plant (WWTP), southwest of Stockholm by the Himmerfjärden bay. The pilot plant was running since 2002. Re-sults from the start-up period are presented in Trela et al., (2004) and Szatkowska (2004). The technical-scale pilot plant was designed for studies of the two step partial nitrita-tion/Anammox process in two reactors of 2.1 m3 each, in series. A partial nitritation process was in the first reactor with Hydrau-lic Retention Time (HRT) of 2 days followed by the Anammox process in the second reactor with HRT of 3 days. Reactors were filled, to 50% of their volume, with Kaldnes biofilm carriers. The influent reject water, from dewatering of the digested sludge, was continuously pumped to the first reactor and

Nitrite

oxidation

+ Ammonia oxidation

+ Organic carbon

oxidation

Inhibited by NaClO3

Ammonia oxidation

+ Organic carbon

oxidation

Inhibited by ATU

Organic carbon

oxidation

Fig. 9. Schematic representation of the action of NaClO3 and ATU on the respirato-ry activity (Surmacz-Górska et al., 1996).

Effluent

Settling tank (buffer)

External recirculation

R2R1 Deoxidising column

Conductivity

Air

Conductivity

DO Settling tank

Dilution with tap water

Conductivity

Settling tank pH, T pH, T

Internal recirculation

Fig. 10. Pilot plant configuration; R1 – partial nitritation reactor, R2 – Anammox reactor ( inter-nal and external recirculation introduced in the transition period from 2-step towards 1-step), (based on Gut, 2006).


22

next the Anammox reactor was fed with effluent from the first step (Fig. 10). The effluent from the first reactor was diluted with tap water. The nitrogen influent load to the Anammox reactor was increased during February – July 2004 period by stepwise decrease of the flow rate of the tap water used. In the start-up period of the Anammox reactor operation, the pH was adjusted by dosage of a Na2CO3 and NaHCO3 solution to keep pH-value around 8. Both reactors were divided into three zones and each zone had a mechanical stirrer. To keep appropriate temperature conditions, the heaters were situated in the first zone of each reactor. Additionally, in the partial nitritation reactor blowers were used, to ensure aeration for ammonium oxidation. The technical details of pilot plant are presented in Table 5.

Biofilm carrier material The Kaldnes carriers were made of a high-

density polyethylene HDPE. Their density was equal to 0.96 kg l-1. The standard types, used in this research, had dimension of 9.1 mm in diameter and the length of 7.2 mm. Each Kaldnes carrier is shaped as a small cylinder with a cross on the inside of the cylinder and “fins” on the outside (Fig. 12). The carriers were suspended in the reactors and were in continuous movement. Since the biomass is growing primary on the protected surface on the inside of the carriers (Rusten et al., 2006), only the effective biofilm surface area of the carriers is given in Table 5. The total surface area consists of both inner and outer surfaces, while the effective surface area is that where biofilm seems to be at-tached. Bur the outer surface of the carrier is continuously cleaned from the biofilm by the frequent collisions between carriers' ele-ments. Application of these carriers allowed achieving total specific surface area of the biofilm equal to 690 m2 in 1 m3 of carriers

Fig. 11. Photo of the technical-scale pilot plant. Partial nitrita-tion reactor on the right side and the Anammox reactor on the left side.

Parameter Unit value No. of reactors - 2 Volume of each reactor m3 2.1 No. of zones in each reactor - 3 Reactor 1 filling by Kaldnes carriers % 46 Reactor 2 filling by Kaldnes carriers % 51 Effective surface area in R1 m2 483.5 Effective surface area in R2 m2 530.2 Volume of settling tank (buffer) m3 0.8 Volume of settling tanks m3 0.125

Table 5. Technical parameters of technic-al scale pilot plant –two-stage process.


23

and an effective specific surface area of the biofilm equal to 500 m2 in 1 m3 of carriers (Welander et al., 1998; Ødegaard et al., 2000; Ødegaard et al., 2006).

Analytical procedure During the study, samples of both influent and effluent (each reactor) were collected and analysed for inorganic nitrogen forms, alka-linity, total nitrogen and Chemical Oxygen Demand (COD). Chemical analyses were conducted using the Dr. Lange test tubes. Samples from batch tests were analysed using AQUATEC-TECATOR 5400 ANALYZER (flow-injection system based on VIS spectro-photometry). Before analyses, all samples were filtrated through 25-μm prefilter fol-lowed by a 0.45-μm filter. Moreover, the process was monitored on a daily basis by manual measurements of physical parameters (flow rate, pH, temperature, conductivity and dissolved oxygen concentration in the bulk liquid) in the influent, effluent and within the reactor. Moreover, a biofilm thickness was measured. For measurement, several Kaldnes carriers were randomly collected from the pilot plant. To get a proper picture of the biofilm, the carriers were cut into thin slices. Then the carriers were scanned with a resolu-tion of 2400 pixels per cal. and the biofilm thickness was measured using Gimp2 soft-ware. Twenty different measurements of biofilm thickness were made for each carrier. The physical parameters were measured using following equipments:

• pH-meter – WTW pH 330, • DO probe – YSI 550A (YSI incorpo-

rated), • thermometer – HI 9063 microcomputer

K-thermo couple thermometer (HANNA instruments, Labassco),

• conductivity meter – TDS Meter (HACH).

Additionally, the pilot plant was equipped with on-line devices. The on-line equipment consisted of Cerlic BB2 device measuring the oxygen concentration. The pilot plant was also provided with two conductivity meters (Dr Lange Analon Cond 10), located in the settling tanks.

Batch tests In order to estimate nitrogen removal rates in different periods of pilot plant operation, several batch tests for studies of nitrogen uptake rates were performed. One series of these tests were divided into four groups (Table 6) in order to recognize significance of the activated sludge and Kaldnes biofilm. In each test, in one bottle, condensed sludge from the Anammox reactor was used while medium used in the second bottle was differ-ent in each group. In the first group, Kaldnes carriers and sludge from R2 were used whereas in the second group Kaldnes carriers and filtrated supernatant. In the third group, Kaldnes was rinsed out in order to remove sludge covering the carriers during pulling out from the reactor. The first time the Kaldnes were rinsed out in tap water and the second time by filtered supernatant (to elimi-nate negative impact of tap water on biofilm) having the same temperature like in the pilot. Additionally, to confirm previous results, in the fourth group three tests were made: one with activated sludge, second with medium from R2 and the third with rinsed Kaldnes carriers and filtered supernatant. Additionally for estimation of the Michaelis-

Fig. 12. Photos of the Kaldnes carrier ma-terial and biofilm on it.

Group 1 Group 2 Group 3 Group 4 Test 1 and Test 2 Test 3 and Test 4 Test 5 and Test 6 Test 7

S K+S S K (NR) S K (R) S K+S K (R)

Table 6. Different batch tests combinations (K – Kaldnes; S – concentrated sludge; NR – not rinsed; R – rinsed).


24

Menten kinetic parameters for the Anammox reaction, series of four batch tests were per-formed. The values of Vmax and KM were estimated separately for ammonium and nitrite using Lineweaver-Burk, Eadie-Hofstee and Hanes-Woolf. The procedure of test performance is presented in Paper I.

IV-3. Moving Bed Biofilm Reactor (MBBR) – one step process

Reactor operation The pilot plant experiment has been run for more than 3 years as a two-stage process where partial nitritation and Anammox took place in two separate reactors (see chapter IV-2). The reactors might be operated in series or parallel. In March 2005 the pilot plant was modified and an experiment with recirculation of a nitrate-rich Anammox effluent to the nitritation reactor was started (Fig. 10). It was intended to denitrify nitrates in the first zone of reactor 1; zones 2 and 3 of the nitritation reactor were connected and in zone 1 the aeration was ceased to obtain oxygen-free conditions for denitrification. After a few weeks, nitrogen loss was ob-served in reactor 1. It was due to the fact that Anammox bacteria found excellent condi-tions to develop its bacteria culture in zone 1. Hence, Anammox was seeded throughout the reactor. Then recirculation was stopped and aeration was switched on again in zone 1 to develop a two-step biofilm layer, an outer layer with oxygen-rich conditions for nitrifi-ers and an inner layer for the Anammox growth. At first the reactor was divided into two zones (zones 2 and 3 were connected), later on all partitions were opened and reac-tor was set as completely mixed reactor. The reactor was filled up to 50% of its volume with Kaldnes biofilm carriers, as biofilm

carriers providing an effective surface area of 250 m2 m-3 of the reactor volume. The Kald-nes carriers in the reactor were in a continu-ous movement due to work of vertical mixers and air supply from the bottom of the reac-tor. The MBBR was supplied with reject water from sludge dewatering after anaerobic digestion in the preliminary phase, the hy-draulic retention time (HTR) was equal to 1 day. To estimate influence of a nitrogen load increase in the following period HRT was shortened to 16 hours. Then, the HRT was again increased to one day. The technical details of pilot plant are presented in Table 7. Previous work in this area (Szatkowska and Płaza, 2006), demonstrated that the Anam-mox process could be operated at a tempera-ture range below 30-35oC, therefore, the process proceeded at a natural temperature of incoming supernatant. Only in winter time an additional heater was supplied to keep temperature above 20ºC. The scheme of the pilot plant is presented in Figure 13. Analytical procedure and biofilm carriers were described above in chapter IV-2.

Settlingtank (buffer)

Conductivity

Air

DO Settling tank

Conductivity

pH, T

Fig. 13. Flow diagram of the pilot-plant with one-stage partial nitritation and Anammox processes.

Parameter Unit value No. of reactors - 1 Volume of reactor m3 2.1 No. of zones in reactor - 2, later 1 Reactor filling by Kaldnes carriers % 46 effective surface area m2 483.5 Volume of settling tank (buffer) m3 0.125 Volume of settling tank m3 0.125

Table 7. Technical parameters of technical scale pilot plant – one-stage process.


25

Determination of the dry weight of biomass developed on Kaldnes carrier To evaluate dry weigh of the biomass at-tached to the carriers, 50 carriers were taken from the reactor and dried in 105°C up to stable weight. Each series consisted of three independent repeats in order to avoid poten-tial measurement errors. Then the biofilm was carefully removed under water stream. In order to completely get rid of remaining biomass, the carriers were placed in vessel with 2M sodium hydroxide solution. A week later, carriers were rinsed again by tap water stream, dried in 105°C and weighted again. The dry weight of biomass developed on carrier can be expressed as follows:

50e-dB = (4)

Where: d – mass of 50 Kaldnes carriers after drying [mg]; e – mass of 50 Kaldnes carriers after washing the biomass from carriers and drying [mg]; B – the dry weight of biomass devel-oped on carrier [mg d.w.]

Batch tests In order to estimate nitrogen removal rates in different periods of the pilot plant operation, five series of batch tests for studies of nitro-gen uptake rates were performed. The spe-cific objective of each series is presented in Table 8. All batch tests were performed in a one-litre vessel that was filled with 50% Kaldnes carri-ers from the pilot plant reactor. An effective specific biofilm surface of 0.25 m2 per 1 dm3 of a reactor bottle was provided. The bottles

were placed in a water bath to keep the tem-perature constant. Additionally, the magnetic stirrers were used to assure appropriate mix-ing of medium during the tests. The vessels could be supplied with air and/or additions of nitrite. Samples were taken every half hour for measurement of inorganic nitrogen com-ponents. Parallel to the sampling, measure-ments of pH-value, DO, conductivity and temperature were performed. The specific conditions of each series of test were differ-ent and specific test performance according to tests objectives can be found in Paper II, III, IV and V.

Oxygen uptake rates tests Six series of OUR test were performed. Se-ries 1-4 were conducted on fresh medium, taken directly form the pilot-plant. Series 5th and 6th were conducted on medium after one and two days starving, respectively. At every test day, the suspended solids and volatile suspended solids were analyzed. Moreover, immobilized biomass concentration was analyzed. The tests were performed in a tightly closed vial equipped with a DO probe and a stirrer, and lasted about 10 minutes each. The con-centrations of sodium chlorate and allylthio-urea were 17 mM and 43 µM, respectively. The concentration of the former compound was chosen according to Gut et al. (2005). The research proved that the dose of sodium chlorate at level of 20 mM proposed by Sur-macz-Górska et al. (1996) had an inhibitory effect on ammonia oxidizing bacteria in this particular system, and a lower one was de-termined.

Series Tests amount Objective Paper

no

1 19 • Estimation of the influence of conditions in the pilot-plant on nitrogen removal rates. II

2 8 • Test with addition of allylthiourea to examine more closely the nature of nitrogen removal mechanism. II

3 12 • Estimation of the nitrogen removal rate at oxygen rich, oxygen free and oxygen rich with artificial nitrite conditions. III

4 20 • Study of the influence of different dissolved oxygen concentrations on nitrogen removal rates. IV

5 32 • Evaluation of bacteria population activity and estimation of nitrogen removal kinetic parameters. V

Table 8. Objectives of the following batch tests series.


26

The aim of the tests performed on fresh medium was to evaluate changes in bacteria activity during the pilot-plant performance. Whereas the tests performed on starved medium were made in order to evaluate kinetic parameters of the aerobic ammonium oxidizing bacteria (AOB). The detailed test performance can be found in Paper V. For calculation of inhibitory effect observed due to substrate excess, the Haldane model was used (eq. 5).

I

2

M

maxA

KSSK

SVr

++

⋅= (5)

Where: rA – the substrate utilization rate [g m-2d-1]; Vmax – the maximum substrate uptake rate [g m-2d-1]; S – the substrate concentration [g m-3]; KM – half-saturation (Michaelis) coeffi-cient [g m-3]; KI – the Haldane inhibition coefficient [g m-3]. The Aiba model (eq. 6) was also checked, however, the correlation coefficient in each case was worse then in the Haldane model.

Moreover, in few cases it was impossible to fit Aiba model at all. The Aiba model:

⎟⎟⎠

⎞⎜⎜⎝

⎛−

+⋅

=IAM

maxA K

SexpSKSV

r (6)

Where: rA – the substrate utilization rate [g m-2d-1]; Vmax – the maximum substrate uptake rate [g m-2d-1]; S – the substrate concentration [g m-3]; KM – half-saturation (Michaelis) coeffi-cient [g m-3]; KIA – the Aiba inhibition coeffi-cient [g m-3].

IV-4. Rotating Biological Contactor (RBC)

Reactor operation A lab-scale Rotating Biological Contactor (RBC) with partially immersed discs was used. The RBC consisted of three equally sized stages. For each stage there were four discs fixed to the centre horizontal shaft. The effective disc area in each stage was 0.87 m2 and the disc submergence was 41%. A devel-oped surface area for biofilm was provided with doormat (PE) fixed to discs. Oxygen

Effluent

Influent

Drive motor

External carbon source

discs Effluent

Influent

Drive motor

External carbon source

discsFig. 14. Flow scheme of the RBC.

Fig. 15. Photos of the lab-scale rotating biological contactor.


27

transfers from the air to the RBC unit in three ways: oxygen absorption at the liquid film over the biofilm’s surface when the biofilm is in the air; direct oxygen transfer happening at the air–water interface caused by the turbulence created by the rotator movement; and direct oxygen absorption by the microorganisms during the air exposure (Rodgers and Zhan, 2003). The RBC unit was covered by polystyrene foam to prevent the growth of algae by light exclusion. The scheme of the RBC is shown in Figure 14 and the design characteristics of the RBC are listed in Table 9. The contactor was supplied with municipal landfill leachate and artificial NaNO2 and NH4Cl solutions to reach high nitrogen con-centration and appropriate excess of nitrite over ammonium oscillating around the NO2:NH4 ratio 1.3:1 for Anammox process. Nitrogen loading rates applied to the first stage of the RBC were gradually increased from 3 to 6 g N m-2d-1. Three research peri-ods can be differentiated basing on the in-creasing nitrogen concentration in the influ-ent (period I: day 1 to 36; period II: day 36 to 91; period III: day 91 to 176). Additionally, an appropriate dose of sodium bicarbonate NaHCO3 was added to neutralize the alkalin-ity decrease during the nitrification process. Glucose, as an external carbon source, was introduced in the third part of the contactor

in order to denitrify nitrites and nitrates after the nitrification and Anammox process. In the later period, the glucose dosage was stopped and the characteristics of the influent was changed. For development of the simul-taneous nitritation/Anammox process, the contactor was supplied only with the ammo-nium nitrogen and no longer supplied with additional nitrite. In the first period, the contactor was provided with leachate from old landfill site in Gliwice. The leachate was collected from the pond and later on from the concrete chamber being a retention tank. Due to the fact, that ammonium concentra-tion in the leachate was low, not exceeded 100 g N-NH4

+ m-3, the NH4Cl solution was added to reach high nitrogen concentration around 700 ± 50 g N-NH4

+ m-3. In the later period of research, the leachate form Gli-wice’s landfill was replaced by leachate from young landfill in Zabrze. The change of the leachate feeding the contactor was gradual. The leachates from both landfill sites were mixed gradually in the following ratios: 2:1, 1:1, 1:2, 0:1 for leachate from Gliwice and Zabrze, respectively. As a matter of fact that the landfill leachate from Zabrze was charac-terized by high concentration of ammonia nitrogen exceeding 1000 g N-NH4

+ m-3, the addition of artificial NH4Cl solutions was stopped. The characteristics of pure landfill leachates from landfill sites in Gliwice and Zabrze are presented in Table 10.

Parameter Unit value No. of stages - 3 No. of discs per stage - 4 Total number of discs - 12 Disc diameter m 0.225 Total surface area available for growth m2 2.61 Disc submergence % 41 Liquid volume m3 0.014

Table 9. Characteris-tics of the rotating biological contactor.

Parameter Unit Gliwice Zabrze pond chamber NH4

+-N g m-3 < 10 100 1100 NO2

--N g m-3 0,5 0,1 0,1 NO3

--N g m-3 76 17 0,7 COD g O2 m-3 250 370 1500 BOD g O2 m

-3 100 120 950

Table 10. The charac-teristics of the leachate from landfill sites In Gliwice and Zabrze.


28

Analytical procedure During the research, samples were collected from the influent, effluent and each stage of the contactor twice a week. The efficiency of the biological treatment was followed in terms of the general parameters such as: COD (dichromate method), ammonia and organic nitrogen (Kjeltec System 1026 Teca-tor), nitrite, and nitrate. Moreover, the proc-ess was monitored by measuring other pa-rameters: flow rate, pH, temperature and dissolved oxygen (DO). No specific heating was applied and the temperature was kept constant at 17 ± 2.4 ºC.

FISH – Fluorescent in situ Hybridization Detached biofilm samples were fixed with paraformaldehyde solution (4% paraformal-dehyde in phosphate-buffered saline (PBS), pH 7.2) at 4 °C for 3 hours and washed sub-sequently in PBS. Fixed samples were stored in PBS: ethanol (1:1) solution at -20 °C. In situ hybridization was performed as described previously by Daims et al. (2005). 16S rRNA targeted fluorescence labelled oligonucleotide

probes; the sequences and targeted sites are listed in Table 11. The probes EUB338, EUB338 II and EUB338 III were mixed together (EUB338 mix) in proportion 1:1:1 in order to detect all bacteria. The probes, chosen from the probeBase database (Loy et al., 2003), were 5’ labeled with the dye FLUOS (5(6)-carboxyfluorescein-N-hydroxysuccinimide ester), Cy3 or Cy5. Both the probes and unlabeled competitor oligonucleotides were obtained from Biomers, Ulm, Germany. Prior to microscope observations samples were embedded in Citifluor (Citifluor Ltd, UK) to reduce the fluorochrome fading (bleaching). A scanning confocal microscope (Zeiss LSM 510) equipped with an Ar-ion laser (488nm) and two HeNeLasers (543nm and 633nm) were used to examine the mi-crobial community. Image processing was performed using the standard software pack-age delivered with the instrument (Zeiss LSM version 3.95).

probe target organisms probe sequence target site FA %

Ref.

NEU

most halophilic and halotolerant Nitrosomonas spp.

CCC CTC TGC TGC ACT CTA 653 – 670 40 1

CTE-competitor NEU

TTC CAT CCC CCT CTG CCG 653 – 670 1

Claster 6a192 Nitrosomonas oligotropha lineage

CTT TCG ATC CCC TAC TTT CC 192 – 211 35 2

Competitor claster 6a192 CTT TCG ATC CCC TGC TTC

C 192 – 211 2

Ntspa662 genus Nitrospira GGA ATT CCG CGC TCC TCT 662-679 35 3

Competitor Ntspa662 GGA ATT CCG CTC TCC TCT 662-679 3

NIT3 Nitrobacter spp. CCT GTG CTC CAT GCT CCG 1035 –

1052 40 4

Competitor NIT3 CCT GTG CTC CAG GCT CCG 1035 –

1052 4

EUB338 most Bacteria GCT GCC TCC CGT AGG AGT 338 - 355 35 5 EUB338 II Planctomycetales GCA GCC ACC CGT AGG TGT 338 - 355 35 6EUB338 III Verrucomicrobiales GCT GCC ACC CGT AGG TGT 338 - 355 35 6Pla46 Planctomycetales GAC TTG CAT GCC TAA TCC 46 - 63 30 7

Amx820

anaerobic ammonium-oxidizing bacteria, Candidatus `Brocadia anammoxidans' and Candidatus `Kuenenia stuttgartiensis'

AAA ACC CCT CTA CTT AGT GCC C 820 - 841 40 8

Ref: 1 – Wagner et al., 1995, 2 – Adamczyk et al., 2003, 3 – Daims et al., 2001, 4 – Wagner et al., 1996, 5 – Amann et al., 1990, 6 – Daims et al., 1999, 7 – Neef et al., 1998, 8 – Schmid et al., 2001.

Table 11. List and description of probes used for the analysis.


29

Denitrifying bacteria analysis Material used for microbiological analysis was collected from three stages of the contactor. Two 10 g samples of biofilm were prepared from each stage. One of them was used for determination of dry mass. The second sam-ple was inserted into Erlenmeyer flask con-taining 90 ml of 0.85% NaCl and then was vigorously shaken on the rotary shaker for 15 min. Afterwards the sample was left for 2 min. to allow sedimentation of large particles and was then used for preparing various dilutions. Suspensions prepared in this way were used for isolation of denitrifying bacte-ria. For titre determination of denitrifying bacteria liquid Giltay medium was used. Inoculated samples were incubated at 26 ºC for 48 hours. After incubation from positive samples 0.1 ml of inoculum was taken and transferred onto nutrient agar plates for isolation of pure denitrifying bacteria cul-tures. Inoculated plates were incubated at 26 ºC for 48 hours. After incubation, single cultures of bacteria were isolated and purified on Petri dishes with nutrient agar. Pure strains were used for inoculation of the liquid Giltay medium to recheck their ability for denitrification and their morphology, Gram reaction and oxidase reaction were deter-mined. For identification of oxidase-positive bacteria strains API 20 NE (standardized system for non-enteric bacteria strains) was used and for oxidase-negative bacteria strains API 20 E (standardized system for enteric bacteria strains).

Batch tests During operation of the lab-scale rotating biological contactor, it became possible to perform some batch tests for studies of ni-trogen uptake rates. The most essential thing associated with this RBC reactor was that test had to be done directly in the reactor. It was impossible to take part of the rotating disc for test without destroying of the whole reactor. That was the main disadvantage of these tests as they could negatively influenced the continuous process occurring in the reactor. One day before the test, the feed of the leachate to the contactor was stopped in order to starve the bacteria and to remove remaining COD to exclude heterotrophic denitrification during the test. Additionally, the first stage of the contactor was tightly separated from the rest of the reactor. After 24 hours, NaNO2 and NH4Cl solutions mix-ture were added to the first stage of the con-tactor to obtain initial ammonium and nitrite concentrations. The test lasted two hours and samples were collected every 15 minutes and analyzed for inorganic nitrogen forms. Addi-tionally, temperature, dissolved oxygen and pH were measured during the batch test.

V – RESULTS AND DISCUSSION

V-1. Membrane Assisted Bioreactor (MBR)

Process performance evaluation Implementation of the Anammox process into the Membrane Assisted BioReactor (MBR) was the aim of performed

Parameter Unit Average Min. Max. St. dev. SamplesNo.

NH4-N in. g m-3 49.5 17.8 74.5 14.5 53 NO2-N in. g m-3 53.9 21.8 98.1 20.3 53 NO2-N/NH4-N - 1.08 0.59 1.44 0.18 53 N removal efficiency % 38.9 0 74.9 16.4 53 COD0 gO2 m-3 43.9 4.9 60 17.2 12 Biomass concentration kg MLSS m-3 7.3 3.5 12..4 2.8 38 Flow rate m3 d-1 0.016 0.009 0.023 0.005 47 HRT d 2.58 1.52 4.1 0.9 37 Temperature ºC 33.9 32 35.8 0.7 52 pH-value in. - 8.0 7.6 8.1 0.1 52 pH-value reactor - 7.9 7.5 8.1 0.1 52

Table 12. Characteristics of the influence medium and operational parameters of the MBR reac-tor.


30

experiment. Therefore, it was necessary to provide condition favourable for growth of appropriate bacteria. The bacteria cultivation took place at the temperature above 30°C, at very low dissolved oxygen concentration (below 0.3 g O2 m-3) and at average pH-value in the influent corrected and maintained around 8. The pH in the reactor was usually the same as in the influent, what agrees with theories (Schalk et al., 1998, Siegrist et al., 1998). Exceptional occurrence of the process’ brake-downs caused the pH drop, as the first stage of nitrification prevailed over the Anammox. The temperature was kept above 30ºC, with average value of 33.8 ± 0.6ºC. In the influent

medium a very low content of biodegradable organic carbon was maintained in order to prevent over-growth of the heterotrophs. Moreover, the nitrite-to-ammonium ratio in the influent to the reactor was around 1:1 which is close to the stoichiometric value of 1.32:1 (Strous et al., 1998). During the ex-periment, the nitrogen load to the reactor was gradually increased from 0.01 to 0.09 kg N m-3d-1. The operational parameters and characteristics of the influent medium are presented in Table 12. At the beginning of the experiment, nitrite nitrogen was the main product of ammonia oxidation (Fig. 16).

0

20

40

60

80

100

120

140

160

180

200

1 23 54 78 96 105 114 126 138 148 161 170 180Time [days]

Nitr

ogen

con

cent

ratio

n[g

N m

-3]

inorg N in inorg N out N-NO3 outN-NO2 out N-NH4 out

HRT

Fig. 16. Nitrogen conversion during start-up of the Anammox process in the membrane assisted bioreactor (Cema et al., 2004).

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

1 23 54 78 96 105 114 126 138 148 161 170 180 189

Time [days]

NO2-N

/NH

4-N

ratio

0

20

40

60

80

100

120

140

N re

mov

al [g

m-3

] and

N

rem

oval

effi

cien

cy[%

]

removal % of removal NO2-N/NH4-N ratio

Fig. 17. Nitrogen removal and nitrogen removal efficiency in the membrane assisted bioreactor (Cema et al., 2004).


31

During first 30 days of the experiment, nitro-gen removal efficiency decreased from 27 to 10 % (Fig. 17). It seems most probable, that this reduction of nitrogen removal efficiency was mainly caused by high nitrite nitrogen concentration in the reactor equal to 64.8 g NO2-N m-3 on average. This concentration was even higher than 60 g NO2-N m-3 re-ported by Fux et al. (2002) as the toxic value for the Anammox bacteria activity, even causing its loss. Therefore, the total nitrogen concentration in the influent had to be re-duced from 100 g N m-3 to 40 g N m-3. As a result of this change, the nitrite concentration in the reactor dropped significantly to level below 6 g NO2-N m-3. At the same time nitrate nitrogen was the main product of ammonia oxidation. In order to suppress the nitrate production, the inorganic nitrogen load to the reactor was raised. For this pur-pose, total nitrogen concentration in the influent to the reactor was increased gradually up to the value around 100 g N m-3. Fur-

thermore, hydraulic retention time (HRT) was gradually diminished from 4 to 1.5 days. The changes, made the nitrate production drop and slight increase of nitrite and ammo-nia concentration was noticed. Total nitrogen removal efficiency again increased within 14 days from 4.1% to 63.8%. Moreover, very intensive gas production in the reactor was observed. Due to the fact, that nitrite nitro-gen was being completely removed in the reactor, the nitrite-to-ammonium ratio in the influent was changed from 1:1 to 1:32 ac-cording to stoichiometric value. After this change, nitrogen removal efficiency increased to the maximum value equal to 74.9%. Since then, process break-down and drastically drop of nitrogen removal efficiency to 25.4% were observed (Figure 17). Also, gas produc-tion in the reactor stopped. Moreover, nitrite nitrogen concentration rose to value of 74 g NO2

¯-N m-3, which is a toxic value for the Anammox activity (Fux et al., 2002, Schmidt et al., 2003). It seems probable, that such an

y = 0.5202x - 0.0054R² = 0.9656

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0 0.02 0.04 0.06 0.08

N re

mov

al ra

te [k

g N

m-3

d-1]

N loading rate [kg N m-3d-1]

y = 0.8654x - 0.0245R² = 0.1599

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.07 0.08 0.09 0.10

N re

mov

al ra

te [k

g N

m-3

d-1]

N loading rate [kg N m-3d-1]

0

2

4

6

8

10

12

14

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0 20 40 60 80 100 120 140 160 180 200

MLS

S [g

l-1]

[kg

N m

-3d-

1 ]

Time [days]

NItrogen removal rate Nitrogen loading rate MLSS

A B

C

Fig. 18. A) Correlation between nitrogen loading and nitrogen removal rate, B) Corre-lation between nitro-gen loading and nitro-gen removal rate (for stable level of loading after 121st day of expe-riment), C) variations of nitrogen loading rate, nitrogen removal rate and biomass con-centration in the MBR (Cema et al., 2004).


32

unexpected breakdown of the Anammox process was caused by very intensive growth of algae (Chlorophyta) on the wall of the reac-tor. Due to high nitrite concentration, influ-ent nitrogen load had to be reduced to nearly 50% (Figure 16). Additionally, reactor was covered with aluminium foil to protect from the light and algae growth. These changes caused increase of nitrogen removal effi-ciency to 51%. However, the process was very unstable. On the other hand, it seems most probable that the salt precipitation which interfered with microbial activity could be other reason of the nitrogen removal efficiency decrease (Trigo et al., 2006). This could be an explanation of such sudden breakdown of the process and its further instability. The other possibility of these problems was the fact, that the mineral elements like K, Fe or Mg was not added to the influent medium even as it is obvious that they were present in the tap water used for the influent medium preparation. Nevertheless, it was decided to introduce these mineral elements to the reac-tor by addition of real landfill leachate to the synthetic wastewater up to 5-10% of its vol-ume. This caused slight increase of the nitro-gen removal, which reached 44%, on average, at the end of research period. Figure 18a shows that the nitrogen removal rate increased parallel with growth of the nitrogen-loading rate. On the one side, this may be due to adaptation of the bacteria; on the other hand, when the nitrogen-loading rate was stable between 86th and 110th day of the experiment, also nitrogen removal rate was stable despite biomass concentration

increase (Fig. 18c). After increasing of the nitrogen-loading rate, the intensive rise of nitrogen removal rate was observed and also nitrogen removal efficiency has been im-proved. It is highly probable, that the Anammox bacteria work better with higher capacity. The correlation between nitrogen loading rate on the stable level 0.075 – 0.09 kg N m-3d-1 and nitrogen removal rate (Fig. 18b) is very low. It is rather unlikely that these nitrogen removal rates are correlated with nitrogen loading rate. Therefore, the good linear correlation was found only within the low range of the nitrogen loads. For high range of the nitrogen-loading rate, the mechanism of nitrogen removal is more complex, what indicated that there were additional inhibitors. It could be possible, that such a high nitrogen-loading rate was too high for bacteria activity and it was the first signal of the process breakdown after 142nd days of the experiment. The other possibility is that the Anammox bacteria had lost in a competition for ammonia with aero-bic ammonium oxidizing bacteria. Interesting information about microorgan-isms’ activity gave Oxygen Uptake Rate (OUR) measurements (Fig. 19). The Anammox process is strictly anaerobic process; however, no one has grown pure cultures of these bacteria in the laboratory. Other microorganisms are essential to re-move one or more toxic products – nitrite, oxygen, organic matter or free radicals – or they might be required to provide essential nutrient (Mohan et al., 2004). Membrane assisted bioreactor used for start-up of the Anammox process was used earlier for nitri-

0

2

4

6

8

10

12

0 25 50 75 100 125 150 175 200

OU

R [g

O2

m-3

h-1]

Time [days]

OUR Ammonium oxidizing bacteriaOUR Nitrite oxidizing bacteria

Fig. 19. Oxygen up-take rate of ammo-nium oxidizing and nitrite oxidizing bacte-ria in the MBR (Cema et al., 2004).


33

fication of high ammonia nitrogen concentra-tion, therefore it is possible that the nitrifiers were still present in the reactor. Measure-ments of OUR activity of ammonium oxidiz-ers and nitrite oxidizers gave much lower results than in earlier research (Surmacz-Górska et al. 2004), but these bacteria were still present in the MBR. There were some relationship between OUR of the ammonium and nitrite oxidizers and nitrite and nitrate nitrogen concentration in the reactor. It was especially clearly seen during the process breakdown. At the same time nitrite concen-tration significantly increased and the OUR of the ammonium oxidizers increased con-siderably and along with increase of OUR of nitrite oxidizers. Also increase of nitrate concentration was observed (Fig. 19). More information about process stoichiome-try, especially in relation to results of OUR test, are given by the ratio of nitrate produc-tion to ammonium consumption (NO3

-/NH4+), as well as the ratio between

nitrite conversion and ammonium conversion (NO2

-/NH4+) (Fig. 20) (the ratios were calcu-

lated as the molar ratio of nitrogen). The average conversion ratio of nitrite and ammonium nitrogen during the whole re-search was 1.05 ± 0.69. However, from start of the nitrogen removal to process break-down in 148th day, the average ratio was equal to 1.31 ± 0.26, which is similar to value stemming from the Anammox stoichiometry

(Strous et al., 1998). On the other hand, the conversion ratio of nitrate production to ammonium conversion was 0.76 ± 0.31 what was much higher than the stoichiometric one. This phenomenon showed that high nitrite oxidizers activity occurred in the reactor, what is in agreement with results obtained in OUR tests. Interesting information could be seen from the correlation between nitrogen removal efficiency and nitrate production to ammonium conversion ratio in period from start of the nitrogen removal to process breakdown in 148th day. Along with nitrogen removal efficiency increase, the nitrate pro-duction to ammonium conversion ratio was decreasing (Fig. 20B) whereas at the same time, nitrite to ammonium conversion ratio was stable amounted to 1.31. These facts indicate that with decreasing activity of nitri-fiers, the Anammox activity and probably amount of the Anammox bacteria was rising. After process breakdown, the nitrate produc-tion to ammonium conversion ratio rapidly increased up to 1.04 in average, consequently leading to loss of the Anammox activity.

Nitrogen conversion The tests were carried out in order to verify whether the Anammox process did take in the reactor. During first hours of the test, the nitrogen removal was not observed. At the same time, the parallel removal of ammo-nium and nitrite nitrogen with increase of

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

0 25 50 75 100 125 150 175 200Time [days]

NO

2- /NH

4+ and

NO

3- /NH

4+

NO2/NH4 NO3/NH4 y = -0.0163x + 1.5011

R2 = 0.6142

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

20 30 40 50 60 70 80

N removal efficiency [%]

NO3- /N

H4+

A B

Fig. 20. A) Conversion ratio of nitrite and ammonium and between nitrate production and am-monium conversion (negative values – nitrite build up took place), B) Relationship between nitrogen removal efficiency and conversion ratio of nitrate production and ammonium conver-sion (data from 78th to 148th day – from start of the nitrogen removal to process breakdown).


34

nitrates concentration was observed. Despite the fact, that there was no nitrogen removal on the beginning of the tests, after 24 hours, decrease of inorganic nitrogen concentration was noticed on the level of 34.5 and 24%, respectively. Nitrates were the main product of nitrogen oxidation. The nitrite-to-ammonium removal rate ratio during the whole test was equal to 2.59, and was much higher than the stoichiometric one for the Anammox process. It confirms the result of OUR tests that the aerobic nitrite oxidizing bacteria are still present and active in the reactor; but on the other hand, the Anam-mox process occurs in the MBR. It seems probable that lower than in the MBR nitro-gen removal efficiency and none nitrogen removal during first few hours of the tests were caused by too high nitrite concentration in the reactor, at the beginning of the tests. Nitrogen removal started after partial nitrite consumption by nitrite oxidizers (Fig. 21A and B).

V-2. Moving Bed Biofilm Reactor – two step process

Process performance evaluation The partial nitritation/Anammox system was designed as a two-step process consisting of an initial partial nitritation reactor (R1) fol-lowed by an Anammox reactor (R2), where the nitrogen removal took place. The process was operated at a temperature above 30°C to keep favorable conditions for growth of the Anammox bacteria for which 37°C is the optimal temperature (Egli et al., 2001). The pH-value in the effluent was within the range from 7.6 to 9.2 and was higher than in the influent (equal to 7.3±0.4). This phenome-non is in accordance with theory of the Anammox process, in which a certain pH-value increase is expected due to con-sumption of hydrogen ions during cell syn-thesis. Consequently, there was no need for pH-value adjustment by dosage of a Na2CO3 and NaHCO3 solutions as it was in the start-

0

20

40

60

80

100

120

140

0 4 8 12 16 20 24

N c

once

trat

ion

[g N

m-3

]

Time [hours]

NO3-N NO2-NNH4-N Total inorg. Ninorg. N t=0

0

20

40

60

80

100

120

140

0 4 8 12 16 20 24

N c

once

trat

ion[

g N

m-3

]

Time [hours]

NO3-N NO2-NNH4-N inorg. Ninorg. N - t=0

A B

Fig. 21. Example of nitrogen conversion in the batch tests, A) 125th day of the expe-riment, B) 132nd day of the experiment (Cema et al., 2004).

Table 13. Characteristics of the influent medium and operational parameters of the two-stage MBBR reactor.


NH4-N in g m-3 124.4 32.3 201.0 31.0 41 NO2-N in g m-3 157.3 71.5 220.5 29.5 41 NO2-N/NH4-N - 1.3 0.7 2.2 0.3 41 N removal efficiency % 86.0 49.4 97.4 8.6 53 COD0 gO2 m-3 185.4 100.0 374.0 76.3 20 Suspended solids g SS m-3 670 335 1165 312 13 Flow rate m3 d-1 0.52 0.36 0.85 0.04 184 HRT d 3.1 1.8 5.1 0.3 184 Temperature reactor ºC 31.4 21.4 42.7 2.7 199 pH-value in - 7.3 5.7 8.5 0.4 199 pH-value out - 8.3 7.6 9.2 0.3 199


35

up period of the Anammox reactor operation (Gut, 2006). The DO concentration was kept at a low level of 0.14 ± 0.07 g O2 m-3 in spite that the reactor was not airtight. Moreover, the nitrite-to-ammonium ratio in the influent to the reactor was kept around the stoichi-ometric value equal to 1.32:1. Table 13 shows the parameters of the Anammox MBBR and characteristics of the influent medium. The Anammox reactor worked as a moving-bed reactor that combined activated sludge and biofilm cultures. The Anammox reactor was operated steadily obtaining high removal of total inorganic nitrogen with an average value of 254 g N m-3, which corresponded to average efficiency of 87%. The reactor was loaded with nitrogen with the average value of 0.28 g N m-2d-1 corresponded to an aver-age nitrogen removal rate amounting to 0.25 g N m-2d-1. The maximum obtained nitrogen removal rate was equal to 0.39 g N m-2d-1 (Fig. 22). During the process, the ammonium and nitrite nitrogen were removed almost com-pletely and the nitrate nitrogen was the main nitrogen form in the effluent from the reac-tor. The nitrate nitrogen formation was measured to 6.9% (on average) of the re-moved inorganic nitrogen with comparison to theoretically expected value of 11%. Dif-ferences between experimental data and theory can be explained by the fact that the heterotrophic bacteria were also present in

the reactor. However, the Anammox process was the main cause of nitrogen removal due to average low drop in the COD concentra-tion equal to 63.7 g O2 m-3. The stoichiomet-ric use of COD in heterotrophic denitrifica-tion (neglecting cell synthesis) is 1.72 g COD per gram of reduced nitrite nitrogen, what indicated that less than 15% of nitrogen could have been removed due to heterotro-phic denitrification. Also other factors, like presence of nitrite and ammonia in an opti-mal ratio and efficient biomass retention (biofilm system), indicate dominant role of the Anammox process. Moreover presence of the Anammox bacteria (Brocadia anammoxi-dans and Kuenenia stuttgartiensis) were proved by means of using the FISH technique (FISH analysis for the pilot plant reactor were made thank to courtesy of the research group from the Delft University of Technology, Kluyvert Laboratory for Biotechnology, the Nether-lands). Later tests, performed by a research group of the Gotheborg University, Depart-ment of Chemistry, Sweden, confirmed that Brocadia anammoxidans was present in the Anammox reactor (Gut, 2006). Despite of fluctuations in nitrogen load and variable nitrite-to-ammonium ratio in the influent (Fig. 22), there were no problems with keep-ing stable values of nitrogen, below 50 g m-3, in the effluent from the reactor. However, too high nitrite-to-ammonium ratio has an adverse influence on the Anammox process.

0,00,20,40,60,81,01,21,41,61,82,02,22,4

0

20

40

60

80

100

120

0 25 50 75 100 125 150 175 200 225 250 275 300

NO2-N

/NH 4

-N r

atio

and

Nitr

ogen

lo

ad a

nd r

emov

al ra

te [

g m

-2d-1

]

N re

duct

ion

effic

ienc

y [%

]

Time [days]

Ninorg. reduction NO2-N/NH4-N ratio N load N removal rate

Fig. 22. Variations of nitrite-to-ammonium ratio, nitrogen removal efficiency and nitrogen load and removal rate.


36

The nitrite-to-ammonium ratio exceeding 2.0 caused significant decrease of nitrogen re-moval efficiency. The ratio of nitrate production to ammonium consumption (NO3

-/NH4+), as well as the

ratio between nitrite conversion and ammo-nium conversion (NO2

-/NH4+) gave some

additional information about the process stoichiometry (Fig. 23). The average conversion ratio of nitrite and ammonium (NO2

-/NH4+) during the whole

research period was 1.35 ± 0.27 and was a little bit higher than these values found by Strous and co-workers (1998). Additionally, the conversion ratio between nitrate produc-tion and ammonium conversion was 0.15 ± 0.07 what was much lower than the stoichiometric value of 0.26. These data suggested that the Anammox process was

dominating in the system; however, the lower production of nitrates and higher conversion of nitrite than stemming from stoichiometry was mainly caused by some heterotrophic denitrification process. During the research period, concentration of suspended solids in the Anammox reactor was very changeable (Fig. 24). The average values were 1.57 kg m-3 and 1.14 kg m-3 for SS and VSS respectively. It was obvious that such high concentration of suspended bio-mass must have had some impact on process performance. Owing to stable results, it seems probable that the activated sludge had rather positive influence on the process.

0.0

0.5

1.0

1.5

2.0

2.5

0 25 50 75 100 125 150 175 200 225 250 275 300Time [days]

Conv

ersi

on ra

tios:

NO

2- :NH

4+ and

NO

3- /NH

4+

NO3/NH4 NO2/NH4NO2

-/NH4+ = 1.32 according

to stoichiometry

NO3-/NH4

+ = 0.26 according to stoichiometry

Fig. 23. Conversion ratio of nitrite and ammonium and between nitrate production and ammo-nium conversion.

0

1

2

3

4

5

6

z1 z3 z1 z3 z1 z3 z1 z3 z1 z3 z1 z3 z1 z3 z1 z3 z1 z3 z1 z3 z1 z3 z1 z3 z1 z3 z1 z3

1 31 50 64 78 92 106 120 132 150 199 251 274 295Slud

ge c

once

ntra

tion

[kg

m-3

]

Time [days]

SS VSS Fig. 24. Variations of the suspended solids (SS) and volatile sus-pended solids (VSS) in the Anammox reactor (Cema et al., 2005).


37

Assessment of bacterial activity in biofilm and activated sludge To confirm the hypothesis of the positive effect of activated sludge on the process, as well as to check if the Anammox process occurs mainly in biofilm, several batch tests were performed. In Table 14 concentrations of SS, VSS and nitrogen removal rates are presented for each batch test. In Figure 25 example profiles of nitrogen conversion during batch tests are presented. Generally, the highest nitrogen removal rates

were obtained in the test with condensated activated sludge. However, in these tests the nitrification activity and oxidation of ammo-nia nitrogen prevailed over nitrite removal. Additionally, high pH drop was noticed, what confirms the presence and activity of aerobic ammonium oxidizing bacteria. Nevertheless, removal of total inorganic nitrogen indicated, that the Anammox bacteria could also live in the suspended biomass. In the test with not rinsed out Kaldnes carriers and filtrated su-pernatant, the VSS concentration was equal to 0.23 kg m-3 as the sludge stuck to the

Nitrogen removal rate Sludge concentration gN m-3d-1 gN/gVSS·d gSS/L gVSS/L

Group 1

Test 1 Sludge 82.8 0.069 3.30 1.20 K + S 140.6 n/a 0.73 0.37

Test 2 Sludge 283.2 0.102 3.80 2.77 K + S 197.7 n/a 1.20 0.88

Group 2

Test 3 Sludge 62.6 0.020 4.50 3.21 Kaldnes

(NR) 77.9 n/a 0.39 0.30

Test 4 Sludge 136.1 0.051 3.77 2.67 Kaldnes

(NR) 76.0 n/a 0.33 0.23

Group 3

Test 5 Sludge 58.0 0.030 2.60 1.96 Kaldnes (R) 0 n/a 0.05 0.04

Test 6 Sludge 139.7 0.052 3.90 2.70

Kaldnes (R) 0 n/a 0.08 0.06 Group 4

Test 7 Sludge 91.2 0.031 3.86 2.90 K + S 37.4 n/a 0.31 0.26

Kaldnes (R) 0 n/a 0.02 0.02

Table 14. Nitrogen removal rates and sludge concentrations in batch tests (K=S – Kaldnes carrier and sludge; NR – not rinsed; n/a – not ana-lysed).

20

25

30

35

40

45

50

55

60

0 30 60 90 120150180210240270

N c

once

ntra

tion

[g m

-3]

Time [min]

N-NH4 N-NO2 N-NO3

20

25

30

35

40

45

50

55

60

0 30 60 90 120150180210240270

N c

once

ntra

tion

[g m

-3]

Time [min]

N-NH4 N-NO2 N-NO3

5

15

25

35

45

55

65

0 60 120 180 240 300 360 420

N c

once

ntra

tion

[g m

-3]

Time [min]

N-NH4 N-NO2 N-NO3

A C B

Fig. 25. Example of nitrogen conversion during batch test A) for condensed sludge, B) Test for Kaldnes carriers and filtrated supernatant, C) Test for Kaldnes carriers (rinsed-out) and filtrated supernatant.


38

Kaldnes carriers while they were taken out of the liquor. In these tests, the average nitrogen removal rates were lower than in the tests with condensed sludge. Also, the ratio of nitrite-to-ammonium removal rate was below 1 what indicated that there was also some nitrifiers’ contribution in the biofilm popula-tion. Based on calculations of nitrogen re-moval by activated sludge, and assuming that sludge, in bottle with Kaldnes carriers, works in the same way like activated sludge alone, it was proved that mainly biofilm is responsible for nitrogen removal. The results for the tests with rinsed Kaldnes carriers and filtrated supernatant (Fig. 25C) were surprising, be-cause nitrogen removal was not detected. Instead, minor oxidation of ammonium to nitrite was noticed. High oxygen concentra-tion during the test could be the main reason of this situation, which could inhibit the Anammox process. Before the test, the bottle was flushed with nitrogen gas to obtain oxy-gen free condition. However, during the experiment oxygen concentration exceeded 0.5 g O2 m-3. It seems most probable that nitrifiers, which are present mostly in acti-vated sludge, play a role of oxygen removers.

The amount of nitrifiers in the biofilm on Kaldnes carriers is insufficient to perform deoxidation. Interesting is also, that in the tests with the activated sludge, dissolved oxygen concentration was always a little bit lower than in the tests with the mixture of Kaldnes carriers and sludge. These results, might confirm the hypothesis that nitrifiers are present mainly in the activated sludge. Both in the mixture of Kaldnes carriers and sludge and in the tests with the activated sludge alone, it could be observed that along with the increase of VSS concentration, also nitrogen removal rate was increasing (Fig. 26A and 26B). However, much smaller con-centration of VSS in the tests performed with combined biofilm and activated sludge is related to much higher nitrogen removal than that observed in the test with activated sludge only. It can indicate two hypotheses: firstly, the Anammox bacteria are present in higher percentage in biocenosis of biofilm than in the biocenosis of activated sludge. Secondly, it is possible that nitrifiers, present in the sludge, create favourable conditions for Anammox bacteria on the biofilm by con-

y = 291.18x + 2.6776R2 = 0.8959

0

50

100

150

200

250

300

0 0.2 0.4 0.6 0.8 1

VSS [g l-1]

N re

m. [

g N

m-3

d-1]

y = 115.65x - 176.69R2 = 0.9027

0

50

100

150

200

250

300

1.9 2.1 2.3 2.5 2.7 2.9VSS [g l-1]

N re

m. [

g N

m-3

d-1]

y = 115.65x - 176.69R2 = 0.9027

0

50

100

150

200

250

300

1.9 2.1 2.3 2.5 2.7 2.9VSS [g l-1]

N re

m. [

g N

m-3

d-1]

0

30

60

90

120

150

180

2.382.772.671.96 2.7 2.9

VSS [g VSS l-1]

N re

mm

oval

rate

[g N

m-3

d-1

]

0

0.015

0.03

0.045

0.06

0.075

0.09

Nitr

ogen

to V

SS ra

tio

[gN

/gVS

S]

N rem. gN/gVSS

A

D

B

C

Fig. 26. A) Correlation between VSS concen-tration and average N removal rate in the tests with mixture of Kaldnes carriers and sludge, B) Correlation between VSS concen-tration and average N removal rate in the tests with condensed sludge, C) Relation-ship between N con-centration (in t = 0) to VSS ratio and N re-moval rate (test with condensed sludge), D) Relationship between VSS concentration, N removal rate and N to VSS ratio in the tests with condensed sludge.


39

sumption of oxygen that can diffuse into the liquid, which do not happen in activated sludge. On Figure 26C and 26D, relationships be-tween nitrogen to VSS ratio and nitrogen removal rates are presented for the tests performed only with the concentrated sludge alone. The decrease of nitrogen removal rate along with the increase of nitrogen to VSS ratio was observed in different tests. Because in none of the tests with the activated sludge there were no problems with sustaining proper oxygen concentration, it seems most probable that it is mainly due to increase nitrogen load in the tests. Moreover, with the increase in nitrogen load, the nitrogen re-moval rates were decreasing. It could be a little surprising that at such low nitrogen load during the test the efficiency in nitrogen removal deteriorated. Explanation of this could be that in the batch tests very high concentration of nitrogen forms are reached at the beginning of the test, which is signifi-cantly higher than in the system with con-tinuous flow. Such high concentration of the nitrogen could suppress the bacteria respon-sible for nitrogen removal. Cooperation or competition? Performed batch tests proved that in the Anammox reactor nitrifiers were present. It seems probable that it is mainly due to the spreading of nitrifying bacteria from the partial nitritation reactor to the second reac-tor. Obtained results show that nitrifiers are mostly in the activated sludge and their amount on biofilm is insignificant. The re-sults of these batch tests were also confirmed by OUR tests performed by Gut (2006). By OUR test it was also proved that AOB cul-ture was more dynamic than NOB culture. Some substrate competition between nitrifi-ers and the Anammox bacteria could be possible. On the other hand, the tests showed that nitrifiers are responsible for oxygen consumption. Oxygen diffusion into mixed liquor can inhibit the nitrogen re-moval. It appears that it is rather some coop-eration of different type of bacteria than competition, what was proven by the per-formed batch tests. The fact that under the period of tests’ execution nitrogen removal

process efficiency in the pilot plant operated at Himmerfjärden WWTP was on average 84% also confirms this hypothesis. The moving-bed system is adequate to gain cooperation of many bacterial cultures in removing nitrogen. The results from this study on the Anammox reactor demonstrated that there is the nitrifying activity present in the Anammox reactor and it is concentrated chiefly in the activated sludge. The coopera-tion of activated sludge and biofilm on Kald-nes carriers is responsible for total effect of nitrogen removal but Anammox activity focuses on biofilm on Kaldnes carriers.

Estimation of kinetic parameters One of the main aims of this thesis was to estimate the kinetic parameters of nitrogen removal. Due to this reason several batch test were performed to evaluate bacteria activity in the reactor and to determine kinetic con-stants of ammonium and nitrite removal for the Anammox process were calculated (Pa-per I). Three different methods were used for determination of the kinetic parameters, and the highest correlation was reached for the Hanes-Woolf method. According to this method, value of Km and Vmax for ammonium removal was 5.74 gNH4-N m-3 and 77.52 g NH4-N m-3d-1 (0.31 g NH4-N m-2d-1) and for nitrite removal 6.53 gNO2- N m-3 and 90.09 gNO2- N m-3d-1 (0.36 gNO2- N m-2d-1), re-spectively. Since, in the reactor were present Anammox bacteria as well as nitrifiers and heterotophs, the obtained results of the ki-netic parameters refer to the complex sys-tems more than to the particular microorgan-ism. In such system, the interaction between different microorganisms may have a big influence on the received results. Received Michaelis constant both for nitrite and am-monium nitrogen are quite high. Additionally, van Dongen and co-workers (2001a) discov-ered that a biofilm thickness of even 0.2 mm is yet active. Up to this depth, the conver-sions can be calculated without taking into account diffusion limitation. These facts indicate that the mass transfer effect in the biofilm and the laminar layer above the biofilm is negligible compared to the Micha-elis constant.


40

V-3. Moving Bed Biofilm Reactor –from two-step towards one-step process

Originally, the technical-scale pilot-plant was designed for studies of two-stage partial nitritation/Anammox process in two reactors in series. Stable partial nitritation is an essen-tial precondition for Anaerobic Ammonium Oxidation in a two-step process (Fux, 2003). The HRT fluctuated from 1.6 to 3.3 days, with an average value equal to 2 ± 0.3 days. The reactor was supplied directly with reject water from dewatering of digested sludge that contained high concentration of ammonium nitrogen varying from 260 to 917 g m-3 with an average value amounted to 609.4 ± 167 g m-3 (Fig. 27). The average ammonium loading rate was 1.08 g NH4-N m-2d-1 and the specific nitrite production rate during stable partial nitrita-tion period was 0.57 g NO2-N m-2d-1. During

the whole period, it was possible to achieve nitrite-to-ammonium ratio in the effluent from the reactor equal to 1.3 ± 0.2, which is close to stoichiometry of the Anammox reaction. It was feasible to reach such favour-able ratio by combination of several factors that favour ammonium oxidizing bacteria over nitrite oxidizers. The sludge liquor, used in presented study, had an influent alkalinity-to-ammonium ratio around 1.3 - 1.4 resulting in 58 ± 7% ammonium oxidation ensuring proper feeding media for the Anammox reactor. Drop in pH-value, presence of free ammonia that exceeded 20 g NH3 m-3 (in the first zone of the reactor) and free nitrous acid up to 5 g HNO2 m-2 (in zone 2 and 3 of R1), temperature exceeding 30°C allowed to sup-press nitrite-oxidizing bacteria. As a result, the nitrate nitrogen concentration in the effluent from the partial nitritation reactor not exceeded 20 g m-3 on average. Fux and

0

100

200

300

400

500

600

700

800

900

1000

0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450

Time [days]

Nitro

gen

conc

entra

tion

[g m

-3] in NH4-N out NH4-N out NO2-N out NO3-N N inorg out

external recirculation

Fig. 27. Nitrogen conversion in the partial nitritation reactor.

0

100

200

300

400

500

600

700

800

275 300 325 350 375 400 425 450

Tota

l nitr

ogen

co

ncen

tratio

n [g

m-3

]

Time [days]

influent effluent Fig. 28. Total inorgan-ic nitrogen in the influent and effluent from partial nitritation reactor.


41

co-workers (2004b) reported that after eleven months of operation of partial nitritation MBBR reactor, significant nitrate production occurred. Authors explained this phenome-non as a result of the adaptation of existing nitrite oxidizers or the accumulation of a new species. The average DO concentration was 3 g O2 m-3 and decrease of this value to 1.5 g O2 m-3 caused only temporal decrease of nitrate production and it increased again within 40 days. Jianlong and Ning (2004) demonstrated that partial nitritation was steadily obtained at DO concentration of 1.5 g O2 m-3. In our research, it was possible to obtain stable partial nitritation with only minor production of nitrates with the average DO concentration equal to 1 g O2 m-3. Gen-erally, at low DO, Nitrosomonas is growing faster than Nitrobacter, so nitrite will be enriched (Rosenwinkel and Cornelius, 2005). In 339th day of the experiment, the external recirculation from the effluent of the Anam-mox reactor to the first zone of the partial nitritation reactor was applied. Simultane-ously the aeration of the first zone of the R1 was swiched-off. The aim of introduced changes was to remove the nitrate nitrogen generated in the Anammox process by het-erotrophic denitrification by remaining or-ganic acids from anaerobic digestion. How-ever, introduced recirculation initiated a significant reduction of nitrogen (Fig. 28) that could not be explained by heterotrophic denitrification. The average nitrogen removal efficiency was equal to 37% during that pe-riod. It seems that created condition turned out to be favourable to Anammox microorganisms. High nitrogen removal in the first reactor was observed just after one-month period of recirculation. Such fast observed activity of the Anammox bacteria could be explained by two phenomena. At first, the Anammox microorganisms were already present in the nitrifying biofilm. It was proved by the FISH analysis that in partial nitritation reactor except Nitrosomonas sp. also some Anammox bacteria were present in the biofilm. More-over, the external recirculation allowed seed-ing the partial nitritation reactor with the Anammox bacteria derived from the Anam-

mox reactor. Rosenwinkel and Cornelius (2005) described the same phenomenon that after decrease of DO concentration in the partial nitritation reactor the nitrogen loss occurred which was not caused by heterotro-phic denitrification (based on the COD bal-ance). The authors suggested that it is neces-sary to first build up a biofilm structure in the carrier material, which can be in the next step seeded with the Anammox microorganisms. Due to the high nitrogen removal in the first reactor, it was decided to change process from two-step into one-step process.

V-4. Moving Bed Biofilm Reactor – one-step process

Process performance evaluation Both the partial nitritation and the Anammox processes took place in a single reactor. Table 15 presents statistical evaluation of the pa-rameters measured during operation of the pilot plant. During the operational period, a pH-value in the effluent from the reactor was equal to the influent value. While the partial nitritation reaction takes place, a large decrease in pH value is normally observed, whereas a rise of pH value is characteristic for the Anammox reaction as cell synthesis occurs. Simultane-ous performance of both processes resulted in ions compensation and therefore the pH drop in the one-stage system was minimal. The pH level was rather constant, with the average values of 7.84 ± 0.11 in the influent to the reactor and 7.84 ± 0.24 in the outlet of the system. The sporadic drop of the pH value in the outlet was mainly as a result of stops in the inflow to the pilot plant. Temperature is a very important factor influ-encing the biological processes. It is espe-cially key factor for the Anammox process, which has optimum temperature on the level of 37 oC. When the reactor was operated as a partial nitritation reactor in two-step process, the average temperature exceeded 30oC. However, due to the fact that it was proved that the Anammox process can be operated at a temperature below the range of 30-35oC (Szatkowska and Płaza, 2006; Szatkowska, 2007), the process did not required additional


42

heating as the digester supernatant tempera-ture was at 25 ± 2.4ºC. Due to this fact, the system worked under natural temperature of incoming supernatant with the average tem-perature in the reactor equal to 24.2 ± 2.6 ºC. Additional heater was supplied only during the winter period to keep temperature stable. The dissolved oxygen (DO) concentration (Fig. 29), as the most vital one for proper course of the simultaneous partial nitrita-tion/Anammox process, was monitored over the period described. Since ammonium nitrogen concentration in the bulk liquid is much higher than the oxy-gen or nitrite nitrogen concentration, ammo-nium diffusion will not limit the process. Nitrite is produced in outer layer by nitrifiers

but is mainly consumed by the Anammox culture in inner layer what means that oxygen is therefore the main limiting factor control-ling the overall rate of the partial nitrita-tion/Anammox process in biofilm reactors. Particularly, it should be taken under consid-eration that dissolved oxygen is also respon-sible for inhibition of the Anammox process. Under high values of oxygen concentration process efficiency can be significantly re-duced. The average value of dissolved oxygen concentration during operational period until 656th day of research was 2.3 ± 0.8 g O2 m-3. A big fluctuation, which can be observed in Fig. 29, was mainly due to centrifuges break-downs, which resulted in no influent coming to the system, as well as electricity break-downs that caused diffusers stops. After 656th


NH4-N in g m-3 577.0 351.0 945.0 87.4 102 N removal efficiency % 58.7 32.8 90.4 12.6 102 COD0 gO2 m-3 209.9 177.0 271.0 27.1 14 Suspended solids g SS m-3 261 87 950 267 10 Flow rate m3 d-1 2.0 1.5 2.9 0.3 102 HRT d 1.1 0.7 2.7 0.3 197 Temperature reactor ºC 24.7 17.0 31.9 2.44 397 pH-value in - 7.84 7.38 8.30 0.11 377 pH-value out - 7.84 6.87 8.60 0.24 389 DO reactor gO2 m-3 2.03 0.07 7.13 0.95 391

Table 15. Characteristics of the influent medium and operation parameters of the one-stage MBBR reactor.

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800

Time [days]

DO c

once

ntra

tion

[g

m-3

]

average DO concentration 2.3±0.8 Interm ittent aeration

Fig. 29. Variations of dissolved oxygen concentration in the reactor. From 656th day of research, intermittent aeration was applied (dashed line stands for average values during aeration mode).


43

day of research, the intermittent aeration system was tested with cyclic turn on/off the aeration system. The aeration mode was 35 minutes with 25 minutes mode without aera-tion. After 723rd day of research, the cycles were changed to 30 minutes of aeration and 30 minutes without. The average oxygen concentration during aeration time was equal to 3.1 ± 0.6 g O2 m-3. During that period, values of DO were more stable in spite of more often breakdown noticed in the pilot plant. The process was also monitored by conduc-tivity measurements since Szatkowska (2004, 2007) proved, that it an excellent indicator for process monitoring (Figure 30). The removal of two main ions coexisting in supernatant as ammonium and hydrogen

carbonate resulted also in conductivity deple-tion. The drop in conductivity value in the reactor is because of ammonia oxidation to nitrite and in the same time alkalinity con-sumption in the nitritation process as well as nitrogen loss during the Anammox process. It can be also observed that curves of con-ductivity measurements at the influent and the effluent run parallel. The system was supplied directly with super-natant that contained high concentrations of ammonium varying from 351-945 g m-3 (Fig. 31) with an average value amounted to 577 ± 87.4 g m-3 (Table 15). During the process, part of ammonium was oxidised to nitrite that reacted with remaining ammonium to dinitrogen gas and nitrates. The total inor-ganic nitrogen elimination for the whole

0

1

2

3

4

5

6

7

8

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800

Time [days]

Cond

uctiv

ity [m

S cm

-1]

inout

Fig. 30. Variations of the conductivity in the inflow and effluent from the reactor.

0

100

200

300

400

500

600

700

800

900

1000

1 36 71 94 113 127 154 197 241 283 316 350 388 435 479 522 570 612 675 724 765

Nitro

gen

conc

entra

tion

[g N

m-3

]

Time [days]

NH4-N in N inorg. out NO2-N out NH4-N out NO3-N out

Fig. 31. The nitrogen variations in the partial nitritation/Anammox reactor.


44

analysed period was 58.7 ± 12.6 % on aver-age. At the same time the average removal of ammonium nitrogen amounted to 66.1 ± 13.9%, what indicates that, in the outlet from the pilot plant, there was still high concentra-tion of ammonia. The mean value of ammo-nium nitrogen in the effluent from the reac-tor was equal to 198.6 ± 98.6 g NH4-N m-3. During the operational period, about 38.5 g m-3 of nitrates was produced, what was around 10.1% of the removed ammonium nitrogen. This value is very close to a stoichiometric one, which is equal to 11%. Figure 31 shows the nitrogen variations in the partial nitritation/Anammox reactor. The detailed description of the pilot plant operations is described in Paper II, III, IV and V.

Influence of conditions in the pilot-plant on nitrogen removal dynamics One of the crucial things concerning design of reactor for nitrogen removal is their stabil-ity in case of hydraulic and substrate concen-tration shocks. It is especially relevant in case of the Anammox process, which is known as very sensitive one. In order to investigate of the sudden change in the hydraulic regime in the pilot plant, several batch tests were per-formed. It was shown, that the shock change in the flow rate and the same in the substrate loading rate to the reactor could temporarily decrease the nitrogen removal efficiency and the same – the nitrogen removal rates (Paper

II). However, after short time, the adaptation to the new condition was observed. The average nitrogen removal rate was higher than compared to the period with lower flow rate. Batch tests confirmed, that in spite of initial negative effect of increased flow rate, the system could easily adapt to the new conditions. The investigations performed by Jin et al. (2008), showed the similar results concerning tolerance of the reactor to the flow rate shock. Additionally, authors stated that the tolerance of Anammox reactor to substrate concentration shock could be weaker than to flow rate shock. In practice, it is suggested that mixing characteristics is a key factor in selection of reactor configura-tions for Anammox and an equalizing tank or recycling line is necessary to cope with se-rious influent substrate concentration shock (Strous et al., 1997; Jin et al., 2008). In order to more detailed examine of nitro-gen elimination mechanism, the tests with addition of allylthiourea (selective inhibitor of Nitrosomonas bacteria) were performed. These tests proved that mainly the Anammox bacteria are responsible for nitrogen removal (Paper II).

Dissolved oxygen influence on the nitrogen removal rate In simultaneous partial nitritation/Anammox process, nitrite availability for the Anammox bacteria has a great impact on overall nitro-gen removal rates. Generally, it is dependent

00.5

11.5

22.5

33.5

44.5

55.5

119 day 126 day 128 day 162 day 119 day 126 day 128 day 162 day 119 day 126 day 128 day 162 day

no DO DO DO+NO2

N re

mov

al r

ate

[g N

m-2

d-1]

Nrem NH4-N NO2-N

Fig. 32. Juxtaposition of the tests with anoxic conditions (noDO), aerobic conditions (DO) and aerobic conditions with addition of a NaNO2 solution (DO+NO2-N).


45

on the nitritation rate. In order to determine the significance of the presence of the nitrite on the overall nitrogen removal rate, several batch tests were performed. Estimation of the nitrogen removal rate at oxygen rich and oxygen free conditions was made. Moreover, additional series of batch tests in aerobic condition with addition of a NaNO2 solution were performed. Investigation of the influ-ence of additional nitrite supply to the batch volume was the main aim of these tests (Pa-per III). Figure 32 shows the results from all batch tests. Obtained results proved, that the nitrite concentration seems to be the rate-limiting factor for the simultaneous nitrita-tion/Anammox process. Moreover, the tests showed that the best results were obtained for those performed under anoxic conditions. It could be explained by competition for substrate by aerobic and anaerobic ammo-nium oxidizers or by partial penetration of the oxygen into the inner, Anammox layer of the biofilm. These results revealed that more detailed investigation on dissolved oxygen influence is necessary. Due to this reason the impact of the oxygen concentration on the nitrogen removal rates was examined (Paper IV). It turned out, that the highest nitrogen removal rates were obtained for the dissolved oxygen concentration around 3 g O2 m-3. However, it can be stated, that the dissolved oxygen concentration above 2 g O2 m-3 had not significant effect on the nitrogen removal rates. On the other hand, at a DO concentra-tion of 4 g O2 m-3 increases of nitrite and nitrate nitrogen concentration in the batch reactor were noticed. It was also observed that increase of biofilm thickness during the operational period had no influence on nitro-gen removal rates in the pilot plant.

Evaluation of kinetic parameters Previous tests confirmed that in the single stage partial nitritation/Anammox process, the activity of the aerobic ammonium oxidiz-ing bacteria and the nitrite production rated had a great impact on the overall reactor performance (Paper III and IV). For that reason, it was necessary to evaluate a simple method, allowing to monitor the activity of aerobic ammonium oxidizing bacteria. The oxygen uptake rate (OUR) measurements turned out to be an excellent tool for estima-tion of the activity of different microorgan-ism (Paper V). Application of OUR tests allows measuring the bacteria activity quickly and with good repeatability, and without a need of using expensive chemicals. Other tests were designed to determine kinetic parameters of reactions performed by aerobic organisms in bacterial community. The expe-riments showed that due to very high inhibi-tion coefficient the aerobic ammonium oxi-dizing bacteria seems to be very resistant to high substrate concentration. Generally it can be stated, that inhibition effect of ammonia nitrogen can be neglected. Additionally, for evaluation of the different bacteria activity and to determine kinetic parameters of nitrogen removal several series of batch test were performed (Paper V). Tests were performed in different periods of pilot-plant operation in order to evaluate the influence of long time process operation on kinetic parameters. The results of calculated parameters are presented in table 16. Obtained results of Vmax, are significantly higher than the results obtained for two-step process (Paper I). These results showed that one-step simultaneous partial nitrita-tion/Anammox process is a better option that two-step system.

series Vmax KM KI

g m-2d-1 g (g d.w.)-1d-1 gN-NH4 m-3 gN-NH4 m-3 5th 1.64 0.13 3.01 1417 6th 2.03 0.10 3.38 1519

Table 16. Kinetic pa-rameters estimated for batch tests.


46

V-5. Rotating Biological Reactor – two step process

Process performance evaluation The RBC was run for six months at 17 ± 2.4ºC which was much lower than the opti-mum temperature reported for the Anam-mox process (Paper VI). However, Siegrist et al. (1998) observed a significant nitrogen loss (ranging from 27 to 68%) in RBC during the treatment of stabilized leachate at tempera-tures ranging from 15 to 20ºC, but they were not sure whether it resulted from the Anam-mox process or from autotrophic denitrifica-tion. Table 17 presents statistical evaluation of the parameters measured during operation of the RBC.

During the operational period, a slight de-crease in pH-value between influent and effluent was noticed. The pH-value in the influent was equal to 8.1 ± 2.4 whereas in the effluent it was 7.9 ± 0.7. During the partial nitritation reaction, a large decrease in pH value is normally observed, while a rise of pH value is characteristic for the Anammox reaction due to cell synthesis. The decrease of pH value indicates that there is intensive first step of nitrification in the first stage of the reactor. Nitrogen loading rates applied to the first stage of the RBC were gradually increased from 3 to 6 g N m-2d-1, and based on the increasing nitrogen concentration in the influent three research period can be differ-entiated (Fig. 33). The influent ammonium


NH4-N in g m-3 421.0 176.0 658.8 148.8 44 NO2-N in g m-3 524.1 215.4 830.0 181.0 44 NO2-N/NH4-N - 1.3 0.7 2.3 0.3 44 N removal efficiency % 71.7 4.4 98.4 18.6 41 COD0 gO2 m

-3 669.2 200.0 1700.0 217.5 42 Flow rate m3 d-1 0.004 0.002 0.006 0.001 44 HRT d 3.3 2.2 6.9 0.8 44 Temperature reactor ºC 17.0 12.0 23.6 2.7 43 pH-value in - 8.1 7.0 8.6 2.4 41 pH-value out - 7.9 6.5 9.0 0.7 43 DO I stage gO2 m

-3 2.6 1.6 4.4 0.7 31

Table 17. Characteristics of the influent medium and operational parameters of the two-stage RBC.

0

200

400

600

800

1000

1200

1400

1600

1 8 18 25 32 38 50 57 64 71 81 88 106 113 121 127 136 143 148 155 165 171

Nitro

gen

conc

entra

tion

[g N

m-3

]

Time [days]

influent Stage I effluent

Period IIIPeriod IIPeriod I

Fig. 33. Inorganic nitrogen removal in RBC.


47

and nitrite nitrogen concentration were gradually increased and remained within the range of 176 – 658.8 g N m-3 and 215.4 - 839.0 g N m-3 respectively (Fig. 33). The inorganic nitrogen removal efficiency in the first stage of the contactor was 58 ± 17% on average during the whole operating time and it was comparable during subsequent periods I, II and III (58, 55 and 57% respectively). This indicates that the nitrogen removal efficiency was independent of the nitrogen load in the applied range. Most probably, the Anammox process was the main cause of nitrogen removal (Paper VI).

The process of nitrogen removal predomi-nated in the first stage of the contactor, pro-viding 88 and 95% ammonium and nitrite nitrogen removal, respectively, during the whole period of operation. In Figure 34 an example of nitrogen profiles in the RBC unit in period I (day 18), period II (day 53) and period III (day 116) are shown. These pro-files confirm that the process of inorganic nitrogen removal predominated in the first stage of the contactor. Moreover, it was associated with high nitrate production as a result of aerobic nitrification. Further re-moval of remaining nitrates took place in the third stage of the contactor where glucose as external carbon source was added. Addition-ally, the nitrite-to-ammonium ratio was also changeable in the influent and varied from 0.7 to 2.3 (Paper VI). However, it seems that this variation had no influence on the inor-ganic nitrogen removal efficiency. Additional information about process stoichiometry are given by the ratio between nitrate production and ammonium consump-tion (NO3

-/NH4+), as well as the ratio be-

tween nitrite conversion and ammonium conversion (NO2

-/NH4+) (Fig. 35). The aver-

age conversion ratio of nitrite-to-ammonium was equal to 1.46, varied from 0.7 to 4.3, and was higher than value described by Strous and co-workers (1998). However, this ratio was comparable to 1.43 obtained in the rotat-ing biological contactor by Wyffels and co-workers (2003). Substantial fluctuations of

0

200

400

600

800

1000

1200

1400

In

St. I

St. I

I

St. I

II

Out In

St. I

St. I

I

St. I

II

Out In

St. I

St. I

I

St. I

II

Out

18 53 116

Nitr

ogen

Com

poun

ds a

nd C

OD

[g N

m-3]

NO3-N NO2-N NH4-N COD

Fig. 34. Example of nitrogen profile in the RBC (days: 18, 53 and 116).

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

0 20 40 60 80 100 120 140 160 180Time [days]

Conv

ersi

on ra

tios:

NO

2- :NH

4+ and

NO

3- /NH

4+

NO2/NH4 NO3/NH4

NO2-/NH4


NO3-/NH4


Fig. 35. Conversion ratio of nitrite and ammonium and between nitrate production and ammo-nium conversion.


48

the ratio can be explained by the competition between nitrifiers and Anammox bacteria. The occurrence of nitrification is also con-firmed by the drop in pH from 8.1 in the influent to 7.1 in stage I. The conversion ratio between nitrate production and ammo-nium conversion was 0.84 ± 0.44 what was much higher than the stoichiometric value of 0.26 (Strous et al., 1998). This phenomenon shows that high nitrite oxidizers activity occurred in the first stage of the contactor. Based on the Anammox process stoichiome-try, no more than 30% of produced nitrates were due to anaerobic ammonium oxidation activity. The ammonium, nitrite and inorganic nitro-gen removal efficiencies in the first stage of the RBC unit are shown in Figure 36. During the whole research period, the average am-monium and nitrite nitrogen removal effi-ciency were on the level of 87.8 ± 15.8% and 94.8 ± 4.8%. The maximum ammonium and nitrite re-moval rates were 3.0 and 3.9 g N m-2d-1, respectively. The maximum inorganic nitro-gen removal rates were 5.8 g N m-2d-1 (0.93 kg N m-3d-1) with an average value of 2.8 g N m-2d-1 (Paper VI). The aeration of treated medium resulted in the DO concentration amounting to 2.6 ± 0.7 g O2 m-3 in the first stage of the contactor. In the second stage the DO concentration rose to 4.8 ± 0.9 g O2 m-3 probably as a result of

earlier consumption of substrates for nitrifi-cation in the first stage and there was no reaction in the second stage of the contactor. The aerobic conditions, in stage I, were con-ducive to nitrification proceeding in the sys-tem, and resulted in nitrate production amounting to 50% of nitrogen inflow. Ac-cording to the stoichiometry of the Anam-mox reaction (Strous et al. 1998), nitrate production should be equal to 11% nitrogen inflow, and the difference was caused by presence of nitrite oxidizers in the biofilm. Nitrobacter and Nitrospira were the species responsible for oxidation of nitrite to nitrate and they competed for substrate with the Anammox bacteria. Between the 130th and the 150th day of the experiments, a temporary breakdown of the process performance was observed. It is difficult to explain, because no operating problem was noticed. It is conceivable that the Anammox bacteria are inhibited (reversi-bly) if exposed to aerobic conditions (Schmidt et al., 2003). For two weeks before this breakdown, the ammonium concentra-tion in stage I dropped much below 20 g NH4-N m-3; this could lead to the process inhibition due to the deeper penetration of oxygen into the biofilm. Siegrist et al. (1998) also observed this phenomenon. It could be also caused by the increase of nitrite nitrogen up to 90 g m-3 on the 123rd day of the ex-periment, however, on the contrary high peaks of nitrite concentration on the 67th and

0

10

20

30

40

50

60

70

80

90

100

1 15 25 36 50 60 71 85 106 116 127 141 148 163 171

Nitro

gen

rem

oval

[%]

Time [days]% N inrg removal % NH4-N removal % NO2-N removal

Fig. 36. Ammonium, nitrite and inorganic nitrogen removal efficiency in the first stage of the RBC.


49

71st days (around 90 g m-3) did not cause such problems. The Anammox process (for the type Brocadia Anammoxidans) is irreversibly inhibited by nitrite at concentrations exceed-ing 70 g N m-3 for longer period (Schmidt et al., 2003; Dapena Mora 2007). Egli and co-workers (2001) showed that Kuenenia stutt-gartiensis has a tolerance to nitrite even up to 180 g N m-3. However, there is no informa-tion about short-term effects of high nitrite nitrogen concentration in the Anammox reactor. This could mean that the process is insensitive to short-term high nitrite concen-trations in the reactor (even up to 100 g NO2-N m-3 for several hours). It resulted in only a temporary decrease of nitrogen re-moval rates. Therefore, the breakdown of the process efficiency could be due to the over-lap of these two phenomena: deeper oxygen penetration into the biofilm due to low am-monium concentration in the bulk liquid, as well as high nitrite concentration. The other possibility is that the Candidatus Kuenenia stuttgatiensis was the dominating Anammox bacteria in the reactor and consequently there were more resistant to high nitrite concentra-tion. The final nitrogen removal was performed in the stage III of the RBC unit by using a sup-plementary biodegradable organic (glucose) as an external carbon source for denitrifica-tion. The average COD/N ratio was equal to 6.2 ± 0.7. The removal of nitrate entering the third stage of the contactor was 34 % on average, which corresponded to a nitrogen removal rate of 0.68 g NO3-N m-2d-1. The average efficiency of inorganic nitrogen re-moval achieved in the whole RBC was 77%.

Kinetic evaluation of process It was shown that the Stover-Kincannon model can be used to describe the ammo-nium and nitrite removal rates in the RBC (Paper VI). However, it appeared that the Stover-Kincannon model was not appropri-ate for the process of the inorganic nitrogen removal (correlation coefficient R = 0.62). This suggested the existence of different enzyme systems acting in the nitrogen trans-formations. Probably the lower correlation was due to participation of other microorgan-isms in nitrogen conversion. For instance Nitrosomonas is able to deammonify and, perhaps to a small extent, heterotrophic denitrifiers able to use small amounts of biodegradable carbon in the influent. The estimated values of the maximum substrate utilization rates (16.72 and 44.05 g m-2d-1 for ammonium and nitrite nitrogen, respectively) showed the possibility of a still higher nitro-gen load to the RBC.

Looking for bacteria populations The presence of the Anammox bacteria belonging to Candidatus Brocadia anammoxidans and/or Candidatus Kuenenia stuttgatiensis in the first two stages of the contactor was also proved (by FISH analyses), confirming the main role of the Anammox process in the nitrogen removal in the first stage of the RBC (Paper VI). Additionally, microbial analysis confirmed he presence of nitrifiers in the biofilm (Table 18). These data suggested that the two simultaneous processes of am-monium oxidation and Anammox could occur in one single RBC unit. Referring to studies of Egli et al. (2003), it was proved that within the RBC biofilm depth, ammonium oxidizing bacteria are on

Nitrosomonas

oligotropha lineage

Most halophilic and halotolerant

Nitrosomonas spp.

Nitrobacter spp.

genus Nitrospira

Anaerobic ammonium-oxidizing bacteria,

Candidatus ‘Brocadia anammoxidans’ and

Candidatus ‘Kuenenia stuttgartiensis’

Stage I - +++ ++ +++ +++ Stage II + ++ ++ ++ + Stage III + + - + -

Scale: (-) – absence; (+) – few; (++) – middle; (+++) – high availability

Table 18. Identified nitrifiers and Anammox bacteria.


50

the outer layer, and the Anammox bacteria were only detected in the lower part of the biofilm, defined by penetration depth of the oxygen into the biofilm. Using commercial identification kits API 20NE and API 20E, in the biofilm from all three stages of the reactor, the denitrifying bacteria strains have been identified (Table 19). Microbial analysis in the stage I of RBC re-vealed the presence of denitrifying bacteria such as: Acinetobacter calcoaceticus, Pseudomonas fluorescence and Alcaligenes xylodoxidans, fol-lowed by: Pseudomonas fluorescence, Pseudomonas alcaligenes, Proteus vulgaris in stage II (Table 19). However, the biodegradable organic carbon to nitrogen ratio was unfavourable for het-erotrophic denitrification. Therefore, hetero-trophic denitrification could not be responsi-ble for nitrogen removal in the first stage of the RBC. The presence of Nitrosomonas spp. in the biofilm suggested that part of the ni-trogen loss could be also attributed to denitri-fying Nitrosomonas cells located in the lower part of the biofilm. They could use ammo-nium as electron donors and nitrite as elec-tron acceptor in case of oxygen lack (Bock et al., 1995; Helmer and Kunst, 1998). The results confirmed presence of Pseudomo-nas considered as a species responsible for denitrification. It is in agreement with the recent results of Ju et al. (2005) who found that Pseudomonas earuginosa is able to denitrify under aerobic condition when DO concen-tration exceeded 1 g O2 m-3. Likewise, in an early observation, Bang et al. (1995) reported the presence of aerobic denitrification in the RBC reactor even when the DO concentra-tion exceeded 3 g O2 m-3 in the treated me-dium. Aeromonas salmonicida, Aeromonas hydro-

phila were the dominating denitrifying strains in the third stage of the contactor. They are Gram-negative, ubiquitous organisms. They are widespread in the mixed liquor of acti-vated sludge plant when they are involved in the degradation of organic matter. They are known as incompletely denitrifying hetero-trophic bacteria showing strong reduction of nitrates to nitrites (Drysdale et al., 1999). Moreover, some authors have observed that these bacteria preferred oxic conditions for growth but were still able to produce nitrites from nitrates reduction while simultaneously utilising oxygen as a final electron acceptor. The domination of denitrifiers responsible for partial denitrification of nitrates only to nitrites and the presence of nitrite oxidizers in the stage III may be an explanation of the low efficiency of nitrate removal. The average oxygen concentration in the bulk liquid was 3.0 ± 1.5 g O2 m-3; this created favourable conditions for nitrifiers that oxidized nitrites produced by Aeromonas spp.

Bacteria identified Stage I Stage II Stage IIIAcinetobacter calcoaceticus + - - Pseudomonas fluorescence + + - Alcaligenes xylodoxidans + - - Pseudomonas alcaligenes - ++ - Proteus vulgaris - + -Aeromonas salmonicida - - ++ Aeromonas hydrophila - - ++ Pseudomonas earuginosa - - + Shigella spp - - + Acinetobacter lwoffi - - +

++ - Dominating strain

Table 19. Identified denitrifying bacteria strains.

Fig. 37. Photos of the biofilm in the denitri-fication part of the RBC.


51

Glucose appears to be the least efficient carbon source in comparison with the methanol and acetic acid used in previous research. Efficiencies of 83 to 97% in nitro-gen removal respectively were achieved for the both organic carbon sources (unpub-lished data). Additionally, it seems that with glucose as an external carbon source, intensive biofilm growth accompanied nitrate reduction and aerobic degradation. In the third stage of the RBC very intensive growth was observed, and discs were completely covered with a white, fluffy biofilm (Fig. 37) Moreover, different denitrifying bacteria strains were noticed when methanol and acetic acid were applied (unpublished data). For biofilm fed with methanol the following bacteria strains have been identified: Aeromo-nas hydrofilla, Aeromonas salmonicida (dominat-ing strain), Aeromonas mausoucida, Pseudomonas aeruginosa (dominating strain), Pseudomonas fluorescence and for biofilm fed with acetic acid: Pseudomonas shigellides, Pseudomonas alcaligenes and Aeromonas sorbia. During the operational period it was proved that, the process is insensitive to short-term high nitrite concentrations in the reactor (even up to 100 g NO2-N m-3 for several hours) (Paper VI).

V-6. Rotating Biological Reactor – one-step process

Process performance evaluation The Rotating Biological Contactor (RBC) treating real landfill leachate from two Polish landfill sites was running for 8 months with average operational temperature equal to 20.6 ± 1.1°C. Three research periods can be dif-ferentiated based on the increasing nitrogen concentration in the influent (period I: day 1 to 45; period II: day 45 to 159; period III: day 159 to 230). (Fig. 38). During the first one, the start up of the Anammox process took place. The second one was the period of the stable work of the reactor. The last one was the period when the process inhibition oc-curred (Paper VII). The reactor was supplied with landfill leachate that contained high concentration of ammonium nitrogen varying from 891 to 1562 g m-3 with an average value amounted to 1173 ± 234 g m-3 (Fig. 38). In the third period of reactor operation, strong process inhibition was noticed (Paper VII). The nitrite nitrogen concentration in the reactor exceeded 100 g m-3 and due to this, the nitrogen concentration in the influ-ent, and the same nitrogen load, had to be decreased from above 1000 g m-3 to 300 – 400 g m-3. After process breakdown, changes in biofilm characteristics in the first stage of the contac-tor were noticed. During stable process op-

0

200

400

600

800

1000

1200

1400

1600

1800

1 16 31 46 61 76 91 106 121 136 151 166 181 196 211 226Time [days]

N c

once

ntra

tion

[g N

m-3

]

influent Stage I effluent

PerioPeriod II Period IIIPeriod I

Fig. 38. Inorganic nitrogen removal in RBC.


52

eration, intensive biofilm growth was ob-served. Process inhibition caused significant biofilm detachment from the disc, and at the same time biofilm changed colouring from light to very dark brown (Fig. 39).

Nitrogen conversion Additional information about process per-formance in the first stage of the RBC gave batch tests carried out in second and third period of the experiment. Fig. 40A and B shows the nitrogen conversion during the batch tests. In both test, the ammonium nitrogen re-moval during the whole test was higher than nitrite nitrogen while in the Anammox proc-ess should be the opposite situation accord-ing to the stoichiometric reaction. However, two subsequent processes – aerobic oxida-tion to nitrite and the Anammox process, used ammonium nitrogen. It was also inter-esting that the nitrogen removal was higher during first 40 minutes of the test. This phe-nomenon could be explained by very high nitrite nitrogen concentration during the batch tests, which exceeded 100 g m-3 what caused the Anammox process inhibition and the same nitrite removal rate after 40 minutes

decreased. Interesting is also comparison of the test from second and third period of experiment. In the test performed in the third period, a slight increase of nitrite nitrogen concentration after 60 minutes was observed, whereas in second period nitrites concentra-tion decreased for whole test. It could be concluded, that mainly the Anammox proc-ess was inhibited during the breakdown of the process performance, and it seemed that partial nitritation was not affected.

Kinetic evaluation of process The analysis of data indicated that the Stover-Kincannon model well described the process of ammonium and inorganic nitrogen re-moval (Paper VII). Due to the fact, that obtained value of KB (the saturation con-stant), the equation become the equation of the first order reaction (González-Martínez and Duque-Luciano, 1991). The Stover-Kincannon equation can be rewritten as:

ASQKr i

A⋅

⋅= (7)

Where: K – is a proportionality coefficient.

Fig. 39. Photos of the biofilm in the first stage of the contactor A) 69th day, Period II –stable process perfor-mance, B) 222th day, Period III – after process inhibition.

0

20

40

60

80

100

120

140

160

180

200

0 20 40 60 80 100 120 140

Nitr

ogen

con

cent

ratio

n [g

m-3]

Time [min]

NH4-N NO2-N NO3-N

0

20

40

60

80

100

120

140

160

180

200

0 20 40 60 80 100 120 140

Nitr

ogen

con

cent

ratio

n [g

m-3]

Time [min]

N-NH4 N-NO2 N-NO3

A B

Fig. 40. Example of nitrogen conversion in the batch tests A) Period II, B) Period III.


53

B

max

KVK = (8)

The relation between load and the removal rate results is linear for the tested values of load to the reactor, according to the model proposed by Eckenfelder (González-Martínez and Duque-Luciano, 1991).

Looking for bacteria populations Microbial analysis (FISH) confirmed the coexistence of nitrifiers and the Anammox bacteria belonging to Candidatus Brocadia anammoxidans and/or Candidatus Kuenenia stuttgatiensis in the rotating biological contac-tor (Paper VII). These data confirm the main role of the simultaneous partial nitrita-tion/Anammox process in the nitrogen re-moval. In table 20, 21 and 22 result of bacte-ria population analysis in the first, second and third stages of the contactor are presented, respectively. During the whole research period the Nitro-somonas sp. were the main aerobic ammonium oxidizers all stages of the contactor. The

Nitrosomonas oligotropha were present only during the first period of the experiment and in the first stage of the contactor also at the beginning of the second period, and it did not develop to a larger sized population under the operating conditions of the reactor. Egli and co-workers (2003) analysed changes in bacteria population during the start-up of the simultaneous partial nitrita-tion/Anammox process in the RBC. They identified Nitrosomonas eurpaea/eutropha, Nitro-somonas urea/oligotropha and Nitrosomonas com-munis in the reactor, however, N. euro-pea/eutropha become the dominating stain in the reactor. This might indicating, that also in presented research this bacteria strain was dominating aerobic ammonium oxidizer in the system. Nitrobacter sp. and Nitrospira sp. were identified as bacteria responsible for aerobic nitrite oxidation. However, theses bacteria were identified only during the first period of the experiment when nitrates were the main product of ammonium oxidation. In the later period when nitrate production was much

Day AOB NOB AAOB; B.

anammoxidans and/or K.

stuttgartiensis Nitrosomonas sp. Nitrosomonas

oligotropha Nitrobacter sp. Nitrospira sp.

16 +++ ++ ++ + - 44 ++++ ++ +++ + - 75 ++++ + - - - 110 ++++ + - - + 135 +++ - - - +++ 169 +++ - - - +++ 200 +++ - - - +++ 236 ++++ - - - +++

Scale: (-) – absence; (+) – few; (++) – middle; (+++) – high availability; (++++) - dominant

Table 20. Identified nitrifiers and the Anammox bacteria in the first stage of the contactor (AOB – ammonium oxidizing bacteria; NOB – nitrite oxidizing bacteria; AAOB – anaerobic ammo-nium oxidizing bacteria).





16 +++ + ++ + - 44 +++ ++ +++ + - 75 +++ + - - - 110 +++ + - - + 135 +++ - - - +++ 169 +++ - - - +++ 200 +++ - - - +++ 236 ++++ - - - +++


Table 21. Identified nitrifiers and the Anammox bacteria in the second stage of the contactor.


54

lower, probably some other bacteria, which were not identified, were responsible for second step of nitrification. Generally, nitrite oxidizing bacteria were present in much lower quantity than the aerobic ammonium oxidizers. The presence of Candidatus Brocadia anam-moxidans and/or Candidatus Kuenenia stuttgatien-sis, especially in the first two stages of the RBC, was also proved. However, they were detected in the middle of the second period in 110 day of experiment, whereas, high nitrogen removal efficiency exceeding 70% were form 45th day. It can be as a result of the methodology of samples collection for FISH analysis. For bacteria analysis, detached biofilm samples from bottom of the reactor were collected. Taking under consideration, that the Anammox bacteria are present in the

inner part of the biofilm (Egli et al., 2003), it may explain such late detection of these bacteria. The Anammox bacteria predomi-nated mainly in the first and second stage of the contactor, however they were also de-tected in the last stage. Interesting is, that amount of the Anammox bacteria became stable after process breakdown. It was possi-ble that there was rather inhibition of bacteria activity than cell’s death.

VI – SYSTEMS COMPARISON

Table 23 compiles results from performed studies with operation of one- and two-step partial nitritation/Anammox in different reactors – MBR, MBBR and RBC.

During the performed research, it was possi-ble to achieve high nitrogen removal effi-ciency in all systems. In all systems the





16 +++ + +++ + - 44 +++ + +++ + - 75 +++ - - - - 110 +++ - - - + 135 +++ - - - + 169 +++ - - - + 200 +++ - - - + 236 +++ - - - +


Table 22. Identified nitrifiers and the Anammox bacteria in the third stage of the contactor.

parameter unit MBR** MBBR***two-step

MBBRone-step

RBC two-step

RBC****one-step

N rem. % 47.5±12.7 86.0±8.6 58.7±12.6 58 ±17* 74.6±9.5* N-load kg m-3d-1 0.06±0.02 0.07±0.01 0.6±0.1 0.9±0.24* 1.12±0.23* N-load g m-2d-1 - 0.28±0.06 2.42±0.06 4.74±1.1* 5.99±1.24* N rem. rate kg m-3d-1 0.03±0.018 0.06±0.01 0.34±0.08 0.53±0.20* 0.83±0.32* N rem. rate g m-2d-1 - 0.25±0.05 1.39±0.31 2.80±1.20* 2.55±1.73* NH4-N in g m-3 49.5±14.5 124.4±31.0 577.0±87.4 421.0±148.8 755.5±380.9 NO2-N in g m-3 53.8±20.3 157.3±29.5 - 524.1±181.0 - NO2-Nconv/NH4-Nconv - 1.05±0.69 1.35±0.27 - 1.46±0.57 - NO3-Nprod/NH4-Nconv - 0.8±0.46 0.15±0.07 0.1±0.05 0.84±0.44* 0.32±0.29* pH-value in - 8.0±0.1 7.3±0.4 7.8±0.1 8.1±2.4 8.1±0.5 pH-value out - 8.0±0.1 8.3±0.3 7.8±0.2 7.9±0.7 8.3±0.7 DO g m-3 <0.2 0.14±0.07 2.03±0.95 2.6±0.7* 1.8±0.7* Temp. reactor °C 33.9±0.7 31.4±2.7 24.7 17.0±2.7 20.6±1.5

* - for the first stage of the RBC where the Anammox process took place; ** - For MBR results for period from 78th to 148th day – from start of the nitrogen removal to process breakdown; *** - Results only for the Anammox reactor; **** - Results for stable process operation – period II.

Table 23. Results and operational parameters for all described systems.


55

Anammox process took place what was confirmed by FISH analysis (with except for MBR – no FISH assay was performed). Reac-tors were supplied with different feeding media. The MBR was supplied with synthetic wastewater, MBBR with reject water from dewatering of digested sludge and the RBC was supplied with municipal landfill leachate from two landfill sites. The highest average nitrogen removal efficiency equal to 86% during the whole research was achieved for two-step partial nitritation/Anammox proc-ess in the Moving Bed Biofilm Reactor, but the N-load was very low. The lowest value was achieved for the MBR and was equal to 47.5 ± 12.7%. However it should be stated that in the MBBR reactor, the Anammox process was already working, whereas in the MBR, the start-up of the process took place. In Figure 41 the comparison of ammonium, nitrite and inorganic nitrogen removal effi-ciency in different system are presented. For the MBR, the results for period from 78th to 148th day are presented – from start of the nitrogen removal to process breakdown. Additionally, also for RBC one-step system, the results from stable process operation are presented. Generally, the highest removal efficiency was achieved for the MBBR two-step process, and the lowest for MBR reac-

tor. Very high removal efficiency of ammo-nium and nitrite nitrogen achieved for the MBBR two-step and RBC two-step and the values were on the similar level. However, in the RBC reactor the inorganic nitrogen re-moval efficiency was much lower than in the MBBR and was equal to 58 ± 17% on aver-age during the whole operating time. The difference between these two systems was mainly in the nitrate production to am-monium conversion ratio. In the MBBR two-step, this ratio was equal to 0.15± 0.07 and was lower than the stoichiometric one, whereas in the two-step RBC the ratio was equal to 0.84 ± 0.44 indicating high nitrite oxidizers activity in the first stage of the contactor. It should be stated that in the two-step RBC the DO concentration was much higher, on the level of 2.6 g m-3, while in the two-step MBBR it was 0.14 g m-3 on average. The DO concentration in the two-step RBC was comparable rather with values obtained in single stage systems. Comparing single stage systems, where partial nitritation and Anammox took place in one single reactor, the nitrogen removal efficiency was higher in the RBC than MBBR, and was equal to 75.7% on average (in the second period) compared to 58.7% achieved in the MBBR.

0

20

40

60

80

100

120

MBR MBBR two-step MBBR one-step RBC two-step RBC one-step

rem

oval

effi

cien

cy [

%]

N inorg. NH4-N NO2-N

Fig. 41. Ammonium, nitrite and inorganic nitrogen removal efficiency – comparison of different systems. (For MBR results for period from start of the nitrogen removal to process breakdown; for RBC one-stage results for stable process operation – period II; for RBC two-stage as well one-stage results are for the first stage of the contactor where Anammox process took place).


56

It can also be noticed (from standard devia-tion values), that during whole operational period in the RBC much higher fluctuation in the nitrogen removal efficiency was recorded. However also there were higher fluctuations in influent ammonium concentration than in the one-step MBBR. It also seems that char-acteristics of the landfill leachate influenced the process performance in a higher extend than the reject water. Obtained values of nitrogen removal rates in MBBR system, showed that one-step partial nitritation/Anammox process is a better option for reject water treatment that two-step process. The average inorganic nitrogen removal rate in two-step MBBR system was 0.25 ± 0.05 g N m-2d-1 (0.06 ± 0.01 kg N m-

3d-1) and was five time lower than in one-step MBBR system where achieved 1.39 ± 0.31 g N m-2d-1 (0.34 ± 0.08 kg N m-3d-1). Figure 42A and B show the nitrogen removal rates in all investigated systems. Vázquez-Padín and co-workers (2009) ob-served similar difference in nitrogen removal rates between two-step and one step process. They used sequencing batch reactors (SBR) for treating supernatant of an anaerobic digester. The combined system allowed the removal of nitrogen loading rates around

0.08 kg N m-3d-1, whereas in single unit a mean nitrogen removal rate of 0.8 kg N m-3d-1 was observed. Conducted studies showed that nitrogen removal rates in the two-step RBC system (equal to 2.8 ± g N m-2d-1 and 0.53 ± 0.20 kg N m-3d-1, in the first stage of the contactor) were lower than for one-step process (4.57 ± 1.10 g N m-2d-1 and 0.83 ± 0.17 kg N m-3d-1, in the first stage of the contactor for stable process operation - period II). Additionally in the two-step RBC higher nitrate production to ammonium conversion ratio was noticed, and it was 0.84 ± 0.44 and 0.32 ± 0.29 (0.17 for stable process operation) for two-step and one-step respectively. In the two-step RBC nitrite nitrogen from influent was oxi-dized by nitrite oxidizers to nitrates, whereas in one-step process nitrite oxidizers loosed the competition for nitrite with the Anam-mox bacteria. During the whole research, the highest removal rates were achieved for the one-step RBC and were 2.5 time higher that for one-step MBBR system. AOB and Anammox bacteria are slow grow-ing microorganisms which optimal tempera-ture of operation is around 30°C. For this reason, Anammox and CANON process were mostly applied to effluents with tem-

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

MBR MBBR two-step

MBBR one-step

RBC two-step

RBC one-step

N re

mov

al ra

te [k

g N

m-3

d-1]

0

0,5

1

1,5

2

2,5

3

3,5

4

4,5

5

MBR MBBR two-step

MBBR one-step

RBC two-step

RBC one-step

N re

mov

al ra

te [g

N m

-2d-1

]

A B

Fig. 42. A) Nitrogen removal rates in kg N m-3d-1 – comparison of different systems; B) Nitrogen removal rates in g N m-2d-1 - comparison of different systems. (For MBR results for period from start of the nitrogen removal to process breakdown; for RBC one-stage results for stable process operation – period II; for RBC two-stage as well one-stage results are for the first stage of the contactor where Anammox process took place).


57

peratures higher than 30°C (Vázquez-Padín et al., 2009). Therefore, the MBR and two-step MBBR systems were operated at tempera-tures exceeding 30°C. However, Szatkowska and Płaza (2006) and Dosta et al. (2008) demonstrated that the impact of temperature on Anammox process is significant in case of sudden decrease of temperature. Moreover, authors proved that bacteria exposed for low temperatures for longer period are able to adapt to new conditions and work more efficiently. Due to this reason, the one-step MBBR system worked under natural tem-perature of incoming reject water with the average temperature in the reactor equal to 24.3 ± 2.6°C. An additional heat was sup-plied only during winter period to keep tem-perature stable. Moreover, the average opera-tional temperature for the one-step as well two-step RBC system was even lower and was equal to 20.6 and 17°C for two-step and one-step system, respectively. Also Siegrist et al. (1998) and Egli et al. (2001) observed a significant nitrogen removal in the RBC at temperature below 20°C. Recently a few authors demonstrated that the Anammox process can be operated at temperature be-low 20°C. Isaka and co-workers (2008) dem-onstrated that removal activity decreased gradually with decreasing operating tempera-ture, however, it still occurred at 6°C with nitrogen conversion rate equal to 0.36 kg N m-3d-1 with comparison to 2.8 and 6.2 at 22 and 32°C, respectively. Authors stated that

the temperature dependence is strongly af-fected by the type of Anammox culture used. In table 24 there is a comparison of condi-tions (temperature and DO concentration) and results in different systems for one-step partial nitritation/Anammox process. Due to the fact, that the Anammox bacteria are reversibly inhibited by very low concen-trations (<1 µM) of oxygen (Kartal et al., 2007), the Anammox process in the MBR reactor and two-step MBBR was operated without any aeration with DO concentration in the bulk liquid below 0.2 g O2 m-3. Addi-tionally, nitrifiers present in both reactors managed with the increased oxygen concen-tration. Different situation took place in the two-step RBC reactor, in which frequent rotation of the media by the shaft supplies oxygen to the microorganisms both in the biofilm and in the bulk liquid. As a result, the average DO concentration in the first zone of the contactor was equal to 2.6 ± 0.7 g O2 m-3. Such a high DO concentration resulted in significant nitrate nitrogen pro-duction. For one-step process, the partial nitritation is a bottleneck of overall reaction. Generally, nitrification rate in fixed-film systems is often limited by the bulk liquid DO concentration (Metcalf and Eddy, 2004). Additionally, DO concentration is one of the most often discussed parameter when eco-nomic factors are considered in treatment plant operation (Szatkowska, 2007). On the other hand, oxygen affects the Anammox

Reactor type Application T DO N rem N rem Ref

°C g m-3 % kg m-3d-1

(g m-2d-1)

MBBR Sludge liquor 28.0-29.0 0.8 – 2.0 71 - 75 (1.2 – 2.2) 1. MBBR Sludge liquor 20 - 35 0 - 4 80 (2 – 2.5) 2. MBBR Sludge liquor 24.3 2.03 58.7 0.34 (1.39) 3. RBC Landfill leachate 20.6 1.8 75 0.83 (4.57) 4. RBC Synthetic wastewater 29.1 0.57 89 (7.4) 5. RBC Landfill leachate 15 - 20 1 - 3 30 - 70 (0.4 – 1.2) 6. RBC High salinity s.w. n.d. <1 84 0.48 7. SBR Synthetic wastewater 30.0 0.1 n.d. 0.08 8. SNAP biofilm reactor Synthetic wastewater 35.0 2 - 3 60 - 80 n.d. 9. Gas-lift reactor Synthetic wastewater n.d. 0.5 42 1.5 10. SBR Sludge liquor 18 - 24 0.5 85 0.45 11.

1. Seyfried et al., 2002; 2. Rosenwinkel et al., 2005b; 3. This study; 4. This study; 5. Pynaert et al., 2003 ; 6. Siegrist et al., 1998; 7.Windey et al., 2005; 8. Third. et al., 2005 ; 9. Furukawa et al., 2006; 10. Sliekers et al., 2003; 11. Vázquez-Padín et al., 2008; n.d. - no data; s.w. – synthetic wastewater.

Table 24. Comparison of conditions and nitrogen removal in a one-stage nitrogen removal process.


58

process since it is a process inhibitor. The average DO oxygen concentration measured in the bulk liquid was equal to 2.03 g O2 m-3

for one-step MBBR reactor and 1.8 g O2 m-3

for one-step RBC system. However, as it was mentioned before, the higher nitrogen re-moval rates were obtained for the RBC. Other researchers who worked with the MBBR technology report DO concentration between 0.8 and 2 g m-3 (Hippen, 2000; Sey-fried et al., 2002). Under such condition, it was possible to reach 75 and 70% of nitrogen removal, respectively. At the DO value equal to 5.9 g m-3, the nitrogen removal amounted only to 10%. According to Johansson et al. (1998), the nitrogen efficiency removal of 60-70% can be obtained under the DO concen-tration below 1 g m-3. The full-scale MBBR in Germany where one-stage partial nitrita-tion/Anammox process was applied gave 80% of nitrogen removal efficiency at DO in the range of 0-4 g m-3 (Rosenwinkel and Cornelius, 2005). Oxygen condition and nitrogen removal rates presented in this the-sis are comparable to results reported by Seyfried et al. (2002). However, the German experiments were run under higher tempera-tures. Pynaert et al (2003) demonstrated ni-trogen removal in the RBC at the level of 89% at the very high nitrogen surface load ranging from 4 to 8 g N m-2d-1. In compari-son to RBC presented in this study, they achieved DO concentration equal to 0.57 g O2 m-3. However, their reactor was operated at temperature 29.1°C, what was significantly higher value than 20.6°C presented in this study. Conducted experiments on different systems with both two- and one-stage partial nitrita-tion/Anammox process demonstrated that one-stage system is a better option for nitro-gen removal from ammonium rich streams irrespective of kind of wastewater. It was possible to obtain higher nitrogen removal rates and moreover, one-step system was less complicated in operation as there was no need for nitrite-to-ammonium control in the influent. Additionally, there was no risk with overloading the system with NO2-N, how-ever, the production of NO2-N in the reactor should be controlled as nitrite nitrogen is

process inhibitor in higher concentration. The one-step process seems to be more cost effective since there need only one reactor and due to higher nitrogen removal capacity, the smaller reactor volume can be used. Moreover, process can be operated under natural temperature of incoming sludge liq-uor and in case of RBC reactor even at tem-peratures below 20°C, therefore the expenses could be reduced.

VII – CONCLUSIONS

The combined partial nitritation/Anammox process was investigated in three different reactors. Additionally process was investi-gated in two different system configurations as one- and two-step process. The partial nitritation/Anammox process could be suc-cessfully applied to treat nitrogen rich streams as landfill leachate and reject water originating from dewatering of digested sludge. Performed research allowed to show differences among used systems and to point out the best system solution. This research has shown that during the start-up period, the Anammox process is very sensitive and unstable. The conclusion are summarised briefly below:

Membrane assisted BioReactor (MBR) 1. The experiments carried out proved that

retention of biomass in Membrane as-sisted BioReactor and long sludge age give possibility for implementation of the Anammox process.

2. At the temperature above 30°C, dissolved oxygen concentration below 0.3 g O2 m-3 and at very low contents of biodegradable organic compounds within 5 months of process operation over 75% of nitrogen removal was reached and very intensive gas production was observed.

3. The research confirmed that very long time, over fourth months, is needed for implementation of the Anammox process.

4. Performed OUR tests bear out presence of ammonium and nitrite oxidizing bacte-ria in the Anammox reactor.


59

Moving Bed Biofilm Reactor (MBBR) – two-step process 1. Performed research proved, that in spite

of observed changes in activities of differ-ent groups of bacteria in Anammox reac-tor, it is possible to obtain stable results of nitrogen removal and high process effi-ciency of nitrogen removal equal to 86.0 ± 8.6%.

2. There is a cooperation of both bacterial cultures (Aerobic ammonium oxidizers and Anammox bacteria) in activated sludge and biofilm in the nitrogen conver-sions. Due to the fact, that nitrifiers play role of oxygen removers, it was possible to establish stable Anammox process in the reactor.

3. In the batch test with both Kaldnes and sludge, the nitrogen removal capacity is much higher in the tests with concen-trated sludge only. The contribution of biofilm on Kaldnes carriers made up the efficient Anammox performance.

4. It is interesting to notice that nitrifiers are outcompeted by Anammox bacteria in the biofilm so that their population is very small, but still active.

5. The kinetic constants of ammonium and nitrite removal for the Anammox process in pilot plant at the Himmerfjäden WWTP were calculated. The best correlation coef-ficients were found for Hanes-Woolf method. The average value of Km and Vmax for ammonium removal was 6.18 gNH4-N m-3 and 78.21 gNH4-N m-3d-1 and for nitrite removal 6.76 gNO2-N m-3 and 90.80 gNO2-N m-3d-1, respectively.

6. Obtained results of kinetic parameters were not stable during the research period, probably because of most probable very dynamic changes in activities and biomass amounts of different groups of bacteria, which are responsible for Anammox and nitrification process in the reactor.

7. High free ammonia and nitrite concentra-tion during the batch tests (not observed in the pilot) affected nitrifiers and Anam-mox populations and proved that such type of batch tests are not excellent simu-lation of condition in the Anammox pilot

plant although the basic conditions like DO concentration, pH and temperature were similar.

Moving Bed Biofilm Reactor (MBBR) – one-step process 1. The biofilm with bacterial culture was able

to perform two processes simultaneously. Adequate hydraulic retention time com-bined with proper oxygen conditions and nitrogen loads are essential for high nitro-gen removal rates.

2. The maximum removal rate, obtained for pilot plant amounted to 2.6 g N m-2d–1, with the average value equal to 1.39 ± 0.31 g N m-2d-1, during the whole research period.

3. The coexistence of aerobic and anaerobic ammonium oxidizers within the biofilm was confirmed by FISH analysis.

4. In conducted batch tests it was shown that the best results are achieved for oxy-gen concentration around 3 g O2 m-3 with the average nitrogen removal rates on the level of 1.8 ± 0.31 g N m-2 d-1 The highest nitrogen removal rate in batch tests were at DO concentration equal to 3.15 g O2 m-3 and amounted to 2.09 g N m-2d-1. At DO concentration equal to 4 g O2 m-3 increase of nitrite con-centration in the batch reactor was no-ticed, and in the same time decrease of in-organic nitrogen removal rates.

5. It was proved, that the Oxygen Uptake Rate tests, as well as the batch tests are great tools that can be applied to the esti-mation of kinetic parameters of reaction performed by nitrifiers and heterotrophs involved in the one-stage partial nitrita-tion/Anammox process.

6. The kinetic constants of ammonium re-moval for the partial nitrita-tion/Anammox process in pilot plant at the Himmerfjäden WWTP were calcu-lated. The best correlation coefficients were found for Lineweaver-Burk and Hanes-Woolf methods. The average value of KM and Vmax for ammonium removal was 17.92 g NH4-N m-3 and 6.45 gNH4-N m-2d-1. Six months later, measured Vmax was on the similar level equal to


60

6.92 g NH4-N m-2d-1, whereas KM was sig-nificantly higher and equal to 37.84 g NH4-N m-3.

7. It was shown, that both aerobic ammo-nium oxidizers and the ANAMMOX bac-teria reveal strong affinity to the substrate, and relatively high resistance for high con-centrations of it. The nitrite oxidizing bac-teria were much more sensitive, and thus they will always be partly inhibited, which is good from the point of view of the best partial nitritation/Anammox process per-formance.

8. Additionally, the activity of ammonia oxidizing nitrifiers was very close to the maximal velocity, which indicates, that the operational conditions serve them good.

9. The highest nitrogen removal rate in the batch test simulating the simultaneous partial nitritation/Anammox process amounted to 2.96 g N m-2d-1, for simula-tion of the Anammox process only it was equal to 5.18 g N m-2d-1.

10. The batch tests showed that the nitrite production seems to be the rate-limiting step for the Anammox reaction in a single reactor.

Rotating Biological Contactor (RBC) – two-step process 1. Removal of total nitrogen has been

achieved with a combination of different processes: nitrification, Anammox and heterotrophic denitrification processes in one RBC. Moreover, it is also possible that some part of nitrogen was removed due to autotrophic denitrification con-ducted by Nitrosomonas spp.

2. FISH analysis confirmed coexistence of nitrifiers and the Anammox bacteria in one reactor. Additionally used API tests proved presence of aerobic denitrifying bacteria in the reactor.

3. The maximum removal rates for ammo-nium and nitrite nitrogen amounting to 3.0 g N m-2d-1 (0.56 kg N m-3d-1) and 3.9 g N m-2d-1 (0.76 kg N m-3d-1) respec-tively, moreover the maximum inorganic nitrogen removal rates (5.0 g N m-2d-1 (0.93 kg N m-3d-1)) has been achieved

4. The Stover-Kincannon model can be used to describe the ammonium and nitrite re-moval rates in the RBC. The estimated values of the maximum substrate utiliza-tion rates (16.72 and 44.05 g N m-2d-1 for ammonium and nitrite nitrogen respec-tively) showed the possibility of still higher nitrogen load to the RBC.

5. The rotating biological contactor with the Anammox process can be successfully op-erated even at temperatures below 20 °C.

6. The process was insensitive to short-term high nitrite concentrations in the reactor of up to 100 g NO2-N m-3 for several hours.

Rotating Biological Contactor (RBC) – one-step process 1. Removal of total inorganic nitrogen has

been achieved with combination of nitrifi-cation, Anammox and denitrification processes.

2. FISH analysis confirmed coexistence of nitrifiers and the Anammox bacteria in one reactor.

3. The maximum inorganic nitrogen removal rates amounting to 6.1 g N m-2d-1 (1.14 kg N m-3d-1) in the first stage of the contactor. The average inorganic nitrogen removal rate during the stable process op-eration was equal to 4.45 g N m-2d-1 (0.83 kg N m-3d-1). However, during the whole research period the average inor-ganic nitrogen removal rate during the stable process operation was equal to 2.55 g N m-2d-1 (0.46 kg N m-3d-1).

4. The RBC reactor could be highly loaded with ammonium nitrogen and is able to treat medium of very high its concentra-tion even exceeding 1500 g N m-3.

5. Generally, the process of nitrogen re-moval predominated in the first stage of the contactor, providing 92 and 61% of ammonium and inorganic nitrogen re-moval, respectively, during the whole pe-riod of operation.

6. The Stover-Kincannon model can be used to describe the inorganic and ammonium nitrogen removal in the RBC. The values of Vmax and KB were estimated as 264.2


61

and 346.4 g N m-2d-1, respectively for in-organic nitrogen and 64.94 and 64.80 g N m-2d-1, respectively, for ammo-nium nitrogen in the first stage of the con-tactor.

7. The obtained value of KB for nitrogen removal was much higher than tested loading rates indicating inorganic nitrogen removal according to the first order reac-tion. The relation between load and the removal rate results is linear for the tested values of load to the reactor.

8. It was proved that RBC with the simulta-neous partial nitritation/Anammox proc-ess can be successfully operated even at temperatures around 20 °C.

9. The high nitrite concentration can inten-sify the inhibition effect of the toxic sub-stances present in the landfill leachate.

General 1. The partial nitritation/Anammox process

can be successfully applied for separate treatment of reject water from dewatering of digested sludge as well as for nitrogen removal from landfill leachate.

2. Obtained values of nitrogen removal rates showed that one-step simultaneous partial nitritation/Anammox process is a better option than two-step system irrespective of kind of wastewater.

3. The obtained nitrogen removal rates were higher in the one-step system which was additionally less complicated in operation and there was no need for nitrite-to-ammonium control in the influent.

4. In spite of the low nitrogen removal rates, the two-step MBBR system assured the highest nitrogen removal efficiency and stable long-term process operation.

5. Partial nitritation/Anammox process can be successfully applied at temperatures much lower than the optimum value.

6. The oxygen concentration in the bulk liquid and the nitrite production rates seem to be the rate-limiting step for the Anammox reaction in a single reactor.

7. High free ammonia and nitrite concentra-tion during the batch tests (not observed in the pilot) affected nitrifiers and Anam-

mox populations and proved that such type of batch tests are not excellent simu-lation of condition in the Anammox pilot plant although the basic conditions like DO concentration, pH and temperature were similar.

VIII – FUTURE RESEARCH

It has been shown that the systems combin-ing partial nitritation/Anammox process can be successfully applied for separate treatment of different types of nitrogen rich wastewa-ter. Due to increasing pollution of water sources and stricter regulations, interest in new processes like the Anammox is rising. There is still a lot of problems, which need to be solved. The following aspects may be investigated in future. • The slow growth rate of the Anammox

organisms is one of the most important of the process bottlenecks. The cultiva-tion of slow-growing microorganisms re-quires efficient retention time of biomass. The process start-up can be impeded by insufficient biomass build-up and the same, the full-scale implementation re-quires a longer start-up period compared to other nitrogen removal technologies. Therefore, one of the most important things in future research is to speed-up this stage.

• The Anammox process is sensitive to high oxygen and nitrite concentrations. In case of lost of the bacteria activity a lot of time is needed for process recovery. That is why, there is a need to draw up a moni-toring system for process performance as well as for control of quality of seeding media to prevent process inhibition by some toxics substances.

• There is strong need to identify a toxic substances present in landfill leachate and in others wastewater to prevent future process inhibition.

• In biofilm systems, biokinetic parameters such as substrate affinities, maximum growth rate or maintenance need can not be well assessed due to diffusion limita-tion (van der Star, 2008). Therefore, the enrichment of the Anammox bacteria in


62

suspended culture, give the great oppor-tunity for the research on Anammox physiology. The Membrane Assisted bio-reactor seems to be promising and excel-lent tool for study of the Anammox bac-teria.

• Due to an influence of mass transfer in biofilm, the biofilm kinetics should be ex-amined.

• Process optimization by using of the mathematical modelling would be an in-teresting task, which allows also checking the influence of different condition with-out risk of process inhibition.


63

REFERENCES

Adamczyk J., Hesselsoe M., Iversen N., Horn M., Lehner A., Nielsen P. H., Schloter M., Roslev P., Wagner M. (2003). The isotope array, a new tool that employs substrate-mediated la-belling of rRNA for determination of microbial community structure and function. Ap-plied and Environmental Microbiology, 69, 6875-6887.

Ahn J.-H. (2006). Sustainable nitrogen elimination biotechnologies: A review. Process Biochemis-try, 41 (8), 1709-1721.

Aktas O., Çeçen F. (2001). Nitrification inhibition in landfill leachate treatment and impact of activated carbon addition. Biotechnology Letters 23, 1607 – 1611.

Amann R.I., Binder B.J., Olson R.J., Chisholm S.W., Devereux R., Stahl D.A. (1990). Combina-tion of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Applied and Environmental Microbiology, 56, 1919-1925.

Amokrane A., Comel C., Veron J. (1997). Landfill leachates pretreatment by coagulation-flocculation. Water Research, 31 (11), 2775 – 2782.

Anthonisen A.C., Loehr R.C., Prakasam T.B.S., Srinath E.G., (1976). Inhibition of nitrification by ammonia an nitrous acid. Journal WPCF (Water Pollution Control Federation), 48 (5), 835 – 852.

Arrigo K.R. (2005). Marine microorganisms and global nutrients cycles. Nature, 437, 349-355. Bae J-H., Kim S-K., Chang H-S. (1997). Treatment of landfill leachates: ammonia removal via

nitrification and denitrification and further COD reduction via nitrification and denitrifi-cation and further COD reduction via Fenton’s treatment followed by activated sludge. Water Science and Technology 36 (12), 341 – 348.

Bang D.-Y., Watanabe Y., Noike T. (1995). An experimental study on aerobic denitrification with polyvinyl alcohol as a carbon source in biofilms. Water Science and Technology, 32 (8), 235 – 242.

Blakey N.C. (1992). “Model Prediction of Landfill Leachate Production” in: Landfilling of waste: leachate. Edited by Christensen T.H., Cossu R., Stegmann R., Elsevier Applied Science, London and New York, Elsevier Science Publishers LTD.

Bock E., Schmidt I., Strüven R., Zart D. (1995). Nitrogen loss caused by denitrifying Nitrosomo-nas cells using ammonium or hydrogen as electron donors and nitrite as electron accep-tor. Archives of Microbiology, 163, 16-20.

Broda E. (1977). Two kinds of lithotrophs missing in nature. Zeitschrift für allgemeine Mikrobiologie, 17 (6), 491 – 493.

Butt T.E., Lockley E., Oduyemi K.O.K. (2007). Risk assessment of landfill disposal sites – State of the art. Waste Management, 28, 952 – 964.

Calli B., Mertoglu B., Inanc B. (2005). Landfill leachate management in Istanbul: applications and alternatives. Chemosphere 59, 819 – 829.

CanzianiR., Emondi V., Garavaglia M., Malpei F., Pasinetti E., Buttiglieri G. (2006). Effect of oxygen concentration on biological nitrification and microbial kinetics in a cross-flow membrane bioreactor (MBR) and moving-bed biofilm reactor (MNNR) treating old land-fill leachate. Journal of Membrane Science 286, 202 – 212.

Çelen I., Türker M. (2001). Recovery of ammonia as struvite from anaerobic digester effluents. Environmental Technology, 22, 1263-1272.

Cema G., Płaza E., Surmacz-Górska J. and Trela J. (2005). Activated sludge and biofilm in the Anammox reactor – cooperation or competition? In: Integration and optimization of urban sanitation systems, Joint Polish-Swedish Seminars, Cracow, 2005, TRITA-LWR.REPORT 3018, 129 – 138.


64

Cema G., Surmacz-Górska J. and Miksch K. (2004). Implementation of anammox process in the membrane assisted bioreactor. In: Integration and optimization of urban sanitation systems, Joint Polish-Swedish Seminars, Stockholm, 2004, TRITA-LWR.REPORT 3017, 81-92.

Christensen T.H., Kjeldsen P., Bjerg P.L., Jensen D.L., Christensen J.B., Baun A., Albrechtsen H-J., Heron G. (2001). Biogeochemistry of landfill leachate plumes. Applied Geochemistry, 16 (7-8), 659-718.

Ciudad G., Robilar O., Muñoz P., Ruiz G., Chamy R., Vergara C. and Jeison D. (2005). Partial nitrification of high ammonia concentration wastewater as a part of a shortcut biological nitrogen removal process. Process Biochemistry, 40, 1715-1719.

Daims H, Stoecker K, Wagner M. (2005). Fluorescence in situ hybridization for the detection of prokaryotes. /In/ Advanced Methods in Molecular Microbial Ecology, (Osborn AM, Smith CJ, ed.). Bios-Garland, Abingdon, UK., 213-239.

Daims H., Nielsen J.L., Nielsen P.H., Schleifer K.H., Wagner, M. (2001). In situ characterization of Nitrospira-like nitrite-oxidizing bacteria active in wastewater treatment plants. Applied and Environmental Microbiology, 67, 5273-5284.

Daims H., Brühl A., Amann R., Schleifer K.-H., Wagner M. (1999). The domain-specific probe EUB338 is insufficient for the detection of all Bacteria: Development and evaluation of a more comprehensive probe set. Systematic and Applied Microbiology, 22, 434-444.

Dalsgaard T., Thamdrup B., Canfield D.E. (2005). Anaerobic ammonium oxidation (anammox) in the marine environment. Research in Microbiology, 156 (4), 457-464.

Dapena Mora A., (2007). Wastewater treatment by anammox process: A short-circuit in the natural nitrogen cycle. PhD thesis, Departamento de Ingeniería Química, Universidare de Santiago de Compostela.

Dosta J., Fernández I., Vázquez-Padín J.R., Mosquera-Corral A., Campos J.L., Mata-Álvarez J., Méndez R. (2008). Short- and long-term effects of temperature on the Anammox proc-ess. Journal of Hazardous Materials, 154 (1-3), 688-693.

Dosta J., Galí A., Benabdallah El-Hadj T., Macé S., Mata-Álvarez J. (2007). Operation and model description of a sequencing batch reactor treating reject water for biological nitrogen re-moval via nitrite. Bioresource Technology, 98, 2065 – 2075.

Drysdale G.D., Bux F., Kasan H.C. (1999). Denitrification by heterotrophic bacteria during acti-vated sludge treatment. Water SA, 25 (3), 357–362.

Egli K., Bosshard F., Werlen C., Lais P., Siegrist H., Zehnder A.J.B., van der Meer J.R. (2003). Microbial Composition and Structure of a Rotating Biological Contactor Biofilm Treating Ammoniu-Rich Wastewater without Organic Carbon. Microbial Ecology 45, 419 – 432.

Egli K., Fanger U., Alvarez P.J.J., Siegrist H., van der Meer J.R., Zehnder A.J.B. (2001). Enrich-ment and characterisation of an anammox bacterium from rotating biological contactor treating ammonium-rich leachate. Archives of Microbiology 175, 198 – 207.

El-Fadel M., Findikakis A.N., Leckie O.J. (1997). Environmental Impacts of Solids Wastes Land-filling. Journal of Environmental Management, 50, 1 – 25.

Fux C., Velten S., Carozzi V., Solley D., Keller J. (2004a). Efficient and stable nitritation and denitritation of ammonium-rich sludge dewatering liquor using an SBR with continuous loading. Water Research, 40, 2765 – 2775.

Fux C, Huang D., Moti A., Siegrist H. (2004b). Difficulties in maintaining long term partial nitri-tation of ammonium-rich sludge digester liquids in a moving-bed biofilm reactor (MBBR). Water Science and Technology, 49 (11-12), 53-60.

Fux C. (2003). Biological nitrogen elimination of ammonium-rich sludge digester liquids. PhD Thesis, Swiss Federal Institute of Technology Zurich. DISS. ETH NO. 15018.


65

Fux C., Boehler M., Huber P., Brunner I., Siegrist H. (2002). Biological treatment of ammonium-rich wastewater by partial nitritation and subsequent anaerobic ammonium oxidation (anammox) in a pilot plant. Journal of Biotechnology, 99, 295 – 306.

Furukawa K., Lieu P.K., Tokitoh H., Fujii T. (2006). Development of single-stage nitrogen re-moval using anammox and partial nitritation (SNAP) and its treatment performances. Water Science and Technology 53 (6), 83-90.

Ganigué R., López H., Balaguer m.D., Colprim J. (2007). Partial ammonium oxidation to nitrite of high ammonium content urban landfill leachates. Water Research 41, 3317 – 3326.

González-Martínez S., Duque-Luciano J. (1991). Aerobic submerged biofilm reactors for waste-water treatment. Water Research, 26 (6), 825-833.

Gut L. (2006). Assessment of a partial nitritation/Anammox system for nitrogen removal. Trita-LWR. LIC:2034, licentiate thesis.

Gut L., Płaza E., Długołęcka M., Hultman B. (2005). Partial nitritation process assessment. Vat-ten, 61, 175-182.

He S., Zhang Y., Yang M., Du W., Harada H. (2007). Repeated use of MAP decomposition resi-dues for the removal of high ammonium concentration from landfill leachate. Chemos-phere, 66, 2233 – 2238.

Hellinga C., Schellen A.A.J.C., Mulder J.W., van Loosdrecht M.C.M., Heijnen J.J. (1998). The SHARON process: an innovative method for nitrogen removal from ammonium-rich wastewater. Water Science and Technology, 37 (9), 135-142.

Helmer Ch., Kunst S. (1998). Simultaneous nitrification/denitrification in an aerobic biofilm system. Water Science and Technology, 37 (4 – 5), 183 – 187.

Hem L.J., Rusten B., Ødegaard H. (1994). Nitrification in a moving bed biofilm reactor. Water Research, 28 (6), 1425 – 1433.

Heyer K.–U., Stegmann R. (2002). Leachate management: leachate generation, collection, treat-ment and costs. Available from: http://ifas-hamburg.de/pdf/leachate.pdf

Hippen A., Helmer C., Kunst S., Rosenwinkel K- H., Seyfried, C. F. (2000). Sixth years of practi-cal experiences with aerobic/anoxic deammonification in biofilm systems. IWA Confer-ence, Paris, France, 3-7, July 2000.

Hoilijoki T.H., Kettunen R.H., Rintala J.A. (2000). Nitrification of anaerobically pretreated mu-nicipal landfill leachate at low temperature. Water Research, 34 (5), 1435 – 1446.

Hwang Y., Yoneyama Y., Noguchi H. (2000). Denitrification characteristics of reject water in upflow biofiltration. Process Biochemistry, 35, 1241 – 1245.

Ilies P., Manivic D.S. (2001). The effect of decreased ambient temperature on the biological nitri-fication and denitrification of a high ammonia landfill leachate. Water Research 35 (8), 2065 – 2072.

Im J-H., Woo H-J., Choi M-W., Han K-B., Kim C-W. (2001). Simultaneous organic and nitrogen removal from municipal landfill leachate using an anaerobic-aerobic system. Water Re-search 35 (10), 2403 – 2410.

Innerebner G., Heribert Insama H., Franke-Whittlea I.H., Wett B. (2007). Identification of anammox bacteria in a full-scale deammonification plant making use of anaerobic ammo-nia oxidation. Systematic and Applied Microbiology, 30 (5), 408-412.

Isaka K., Date Y., Kimura Y., Sumina T., Tsuneda S. (2008). Nitrogen removal performance using anaerobic ammonium oxidation at low temperatures. FEMS Microbiology Letters, 282, 32-38.


66

Isaka K., Yoshie S., Sumino T., Inamori Y., Tsuneda S. (2007). Nitrification of landfill leachate using immobilized nitrifying bacteria at low temperatures. Biochemical Engineering Journal, 37, 49 – 55.

Janus H.M., van der Roest H.F. (1997). Don’t reject the idea of treating reject water. Water Science and Technology, 35 (10), 27 – 34.

Jeavons J., Stokes L., Upton J., Bingley M. (1998). Successful sidestream nitrification of digested sludge liquors. Water Science and Technology, 38 (3), 111 – 118.

Jianguo J., Guodong Y., Zhou D., Yunfeng H., Zhonglin H., Xiangming F., Shengyong Z., Cha-oping Z. (2007). Pilot-scale experiment on anaerobic bioreactor landfills in China. Waste Management, 27, 893 – 901.

Jianlong W., Ning Y. (2004). Partial nitrification under limited dissolved oxygen conditions. Proc-ess Biochemistry, 39 (10), 1223-1229.

Jin R-c., Hu B-l., Zheng P., Qaisar M., Hu A-h., Islam E. (2008). Quantitative comparison of stability of ANAMMOX process in different reactor configurations. Bioresource Technology, 99 (6), 1603-1609.

Johansson P., Nyberg A., Beier M., Hippen A., Seyfried C. F. and Rosenwinkel K-H. (1998). Cost efficient sludge liquor treatment. In: Joint Polish–Swedish Reports, Report No 3. Royal Insti-tute of Technology, Stockholm, TRITA-AMI Report 3048, 65-72.

Jokela J.P.Y., Kettunenb R.H., Sormunena K.M., Rintalaa J.A. (2002). Biological nitrogen re-moval from municipal landfill leachate: low-cost nitrification in biofilters and laboratory scale in-situ denitrification. Water Research, 36, 4079 – 4087.

Ju L.-K., Chen F., Xia Q. (2005). Microaerobic denitrifica-tion of Pseudomonas aeruginosa by online NAD(P)H fluo-rescence. Journal of Industrial Microbiology and Biotechnology, 32, 622 -628.

Junga J-Y., Chunga Y-C., Shinb H-S., Sona D-H. (2004). Enhanced ammonia nitrogen removal using consistent logical regeneration and ammonium exchange ofzeolite in modified SBR process. Water Research, 38, 347 – 354.

Kaczorek K., Ledakowicz S. (2006). Kinetics of nitrogen removal from sanitary landfill leachate. Bioprocess and Biosystems Engineering, 29, 291 – 304.

Kalyuzhnyi S., Gladchenko M., Mulder A., Versprille B. (2006). DEAMOX-New biological ni-trogen removal process based on anaerobic ammonia oxidation coupled to sulphide-driven conversion of nitrate into nitrite. Water Research, 40, 3637-3645.

Kartal B. (2008). Ecophysiology of the anammox bacteria. PhD thesis. Department of Microbiology, Radbound University Nijmegen. Nijmege, ISBN 978 90 90227 14 6.

Kartal B., van der Star W.R.L., Schmid M.C., van de Pas-Schoonen K., Picioreau C., Abma W.R., Op den Camp H., Jetten M.S.M., van Loosdrecht M., Strous M. (2007). Anammox Proc-ess: State Of The Art. In: Proceeding of the CLONIC Final Workshop, 19 – 20th April 2007 Barcelona, Spain, 8-21.

Kietlińska A., Renman G. (2005). An evaluation of reactive filter media for treating landfill leachate. Chemosphere, 61, 933 – 940.

Kim D., Ryu H-D., Kim M-S., Kim J., Lee S-J. (2007). Enhancing struvite precipitation on poten-tial for ammonia nitro gen removal in municipal landfill leachate. Journal of Hazardous Ma-terials, 146, 81 – 85.

Kjeldsen P., Barlaz M.A., Rooker A.P., Baun A., Ledin A., Christensen T.H. (2002). Present and Long-Term Composition of MSW Landfill Leachate: A Review. Critical Reviews in Envi-ronmental Science and Technology, 32 (4), 297 – 336.

Knox K. (1985). Leachate treatment with nitrification of ammonia. Water Research, 19 (7), 895 – 904.


67

Kohler N., Perry E. (2005). Report on Implementation of the Landfill Directive in the 15 Mem-ber States of the European Union 04523371. REF ENV.A.2/ETU/2004/0016.

Kulikowska D., Klimiuk E. (2007). The effect of landfill age on municipal leachate composition. Bioresource Technology, Article in press.

Kurniawan T.A., Lo W.H., Chan G.Y. (2006). Physico-chemical treatments for removal of recal-citrant contaminants from landfill leachate. Journal of Hazardous Materials. B129, 80 – 100.

Li X.Z., Zhao Q.L. Hao X.D. (1999). Ammonium removal from landfill leachate by chemical precipitation. Waste Management, 19, 409 – 415.

Li X., Zen G., Rosenwinkel K.H., Kunst S., Weichgrebe D., Cornelius A., Yang Q. (2004). Start up of deammonification process in one single SBR system. Water Science and Technology, 50 (6), 1-8.

Liang Z., Liu J. (2008). Landfill leachate treatment with a novel process: Anaerobic ammonium oxidation (Anammox) combined with soil infiltration system. Journal of Hazardous Materi-als, 151, 202 – 212.

Liang Z., Liu J-x. (2007). Control factors of partial nitritation for landfill leachate treatment. Jour-nal of Environmental Science, 19, 223 – 229.

Loy A., Horn M., Wagner M. (2003). probeBase - an online resource for rRNA-targeted oligonu-cleotide probes. Nucleic Acids Research, 31, 514-516.

Loukidou M. X., Zouboulis A. I. (2001). Comparison of two biological treatment processes using attached-growth biomass for sanitary landfill leachate treatment. Environmental Pollution, 111 (2), 273 – 282.

Mackinnon I.D.R., Barr K., Miller E., Hunter S., Pinel T. (2003). Nutrient removal from waste-waters using high performance materials. Water Science and Technology, 47 (11), 101 – 107.

Marttinen S.K., Kettunen R.H., Sormunen K.M., Soimasuo R.M., Rintala J.A. (2002). Screening of physical-chemical methods for removal of organic material, nitrogen and toxicity from low strength landfill leachates. Chemosphere, 46, 851 – 858.

Metcalf & Eddy (2004). Wastewater Engineering. Treatment- Disposal-Reuse. McGraw-Hill Int. Edi-tions, Civil engineering Series. Fourth edition.

Mohan S., Dennison V., Schmid M., Richardson D., Cole J., (2004). We all needfriends – but why can’t Anammox bacteria live alone? In: Proceedings of the “EU 5th framework IcoN Symposium: Anammox: new sustainable N-removal from waste water”, Ghent, 19-24 September.

Mulder J. W., van Loosdrecht M.C.M., Hellinga C., van Kempen R. (2001) Full-scale application of the SHARON process for treatment of rejection water of digested sludge dewatering. Water Science and Technology, 43 (11), 127 – 134.

Mulder A., van de Graaf A.A., Robertson L.A., Kuenen J.G. (1995). Anaerobic ammonium oxi-dation discovered in denitrifying fluidized bed reactor. FEMS Microbiology Ecology, 16, 177-184.

Neczaj E., Kacprzak M., Lach J., Okoniewska E., (2007). Effect of sanitation on combined treatment of landfill leachate and domestic sewage in SBR reactor. Desalination, 204, 227 – 233.

Neef A., Amann R., Schlesner H., Schleifer K.-H. (1998). Monitoring a widespread bacterial group: in situ detection of planctomycetes with 16S rRNA-targeted probes. Microbiology, 144, 3257-3266.

Op den Camp H.J.M., Jetten M.S.M. Strous M. (2007). “Anammox” in: Biology of the nitrogen cycle. Edited by Bothe H., Ferguson S.J. Newton W.E., Elsevier B. V., Amsterdam, Nether-lands.


68

Op den Camp H.J.M., Kartal B., Guven D., van Niftrik L.A.M.P., Haaijer S.C.M., van der Star W.R.L., van de Pas-Schoonen K.T., Cabezas A., Ying, Z., Schmid M.C., Kuypers M.M.M., van de Vossenberg J., Harhangi H.R., Picioreanu C., van Loosdrecht M.C.M., Kuenen J.G., Strous M. and Jetten M.S.M., (2006) Global impact and application of the anaerobic ammonium-oxidizing (anammox) bacteria. Biochemical Society Transactions, 34, part 1, 174-178.

Ozturk I., Altinbas M., Koyuncu I., Arikan O., Gomec-Yangin C. (2003). Advanced physico-chemical treatment experiences on young municipal landfill leachates. Waste Management, 23, 441 – 446.

Paredes D., Kuschk P., Stange F., Müller R. A., Köser H. (2007). Model experiments on improv-ing nitrogen removal in laboratory scale subsurface constructed wetlands by enhancing the anaerobic ammonia oxidation. Water Science and Technology, 56 (3), 145 – 150.

Peng Y., Zhang S., Zeng W., Zheng S., Mino T., Satoh H. (2008). Organic removal by denitrita-tion and methanogenesis and nitrogen removal by nitritation from landfill leachate. Water Research, 42, 883 – 892.

Petts G. (2005). “Environmental flows” in: Water Encyclopedia, Volumes 1 – 5 edited by Lehr J., Keeley J., Lehr J., Kingery T.B., John Willey & Sons. New Jersey, United States of Amer-ica.

Poznyak T., Bautista G.L., Chairez I., Córdova R.I., Rios L.E. (2008). Decomposition of toxics pollutants in landfill leachate by ozone after coagulation treatment. Journal of Hazardous Materials, 151, 202 – 212.

Price G.A., G., Barlaz M.A., Hater G.R. (2003). Nitrogen management in bioreactor landfills. Waste Management, 23, 675 – 688.

Pynaert K., Smets B.F., Wyffels S., Beheydt D., Siciliano S.d., Vestraete W. (2003). Characteriza-tion of an Autotrophic Nitrogen-Removing Biofilm from a Higly Lab-Scale Rotating Bio-logical Contactor. Applied and Environmental Microbiology, 69, 3626 – 3635.

Renou S., Givaudan J. G., Poulain S., Dirassouyan F., Moulin P. (2008). Landfill leachate treat-ment: Review and opportunity. Journal of Hazardous Materials, 150, 468 – 493.

Rivas F.J., Beltán F., Carvalho F., Acedo B., Gimeno O. (2004). Stabilized leachates: sequential coagulation-flocculation + chemical oxidation process. Journal of Hazardous Materials B116, 95 – 102.

Rodgers M., Zhan X.-M. (2003). Moving-medium biofilm reactors. Reviews in Environmental Science and Bio/Technology, 2, 213-224.

Rosenwinkel K-H., Cornelius A., Ruhrverband D.T. (2005). Full-scale application of the deam-monification process for treatment of sludge water. In: Proceeding of the IWA Specialized Conference: Nutrient Management in Wastewater Treatment, Processes and Recycle Streams, 19 – 21 September 2005, Kraków, Poland, 483 – 491.

Rosenwinkel K-H., Cornelius A. (2005). Deammonification in the Moving-Bed Process for Treatment of Wastewater with High Ammonia Content. Chemical Engineering and Technol-ogy, 28 (1), 49-52.

Rusten B., Eikebrokk B., Ulgenes Y., Lygren E. (2006). Design and operations of the Kaldnes moving bed biofilm reactors. Aquacultural Engineering, 34, 322 – 331.

Salem S., Berends D.H.J.G., van der Roest H.F., van der Kuij R.J., van Loosdrecht M.C.M. (2004) Full-scale application of the BABE technology. Water Science and Technology, 50 (7), 87-96.

Salem S., Berends D.H.J.G., Heinen J.J., Van Loosdrecht M.C.M. (2003). Bio-augmentation by nitrification with return sludge. Water Research, 37, 1794 – 1804.


69

Seyfried C. F., Rosenwinkel K. H., Hippen A. (2002). Deammonification: a cost-effective treat-ment process for nitrogen-rich wastewaters. WEFTEC 2002 Proceedings 75th Annual Con-ference and Exposition. McCormick Place, Chicago, Illinois, USA, September 28-October 2, 2002.

Schalk J., Oustad H., Gijs Kuenen J., Jetten M.S.M. (1998). The anaerobic oxidation of hydrazine: a novel reaction in microbial nitrogen metabolism. FEMS Microbiology Letters, 158, 61 – 67.

Schmidt I., Sliekers O., Schmid M., Bock E., Fuerst J., Kuenen J.G., Jetten M.S.M., Strous M. (2003). New concepts of microbial treatment processes for the nitrogen removal in wastewater. FEMS Microbiology Reviews, 772, 1 – 12.

Schmid M., Schmitz-Esser S., Jetten M.S.M., Wagner M. (2001). 16S-23S rDNA intergenic spacer and 23S rDNA of anaerobic ammonium-oxidizing bacteria: implications for phylogeny and in situ detection. Environmental Microbiology, 3, 450 – 459.

Shino H., Nishimori K., Kawakami S., Uesaka T., Izumi K. (2004). Introduction to Submerged Flat Sheet Membrane Applications and its Future to be Headed. In. Ultrapure Water Asia 2004, March 2004.

Siegrist H., Reithaar S., Koch G., Lais P. (1998). Nitrogen loss in nitrifying rotating contactor treating ammonium-rich wastewater without organic carbon. Water Science and Technology, 38 (8-9), 241 – 248.

Siegrist H. (1996). Nitrogen removal from digester supernatant – comparison of chemical and biological methods. Water Science and Technology, 34 (1-2), 399 – 406.

Sliekers A.O., Third K.A., Abma, W., Kuenen J.G., Jetten, M.S.M. (2003). CANON and Anam-mox in a gas-lift reactor. FEMS Microbiology Letters, 218 (2), 339-344.

Sliekers A.O., Derwort N., Campos Gomez J.L., Strous M., Kuenen J.G., Jetten M.S.M. (2002). Completely autotrophic nitrogen removal over nitrite in one single reactor. Water Research, 36, 2475-2482.

Strous M., Pelletier E., Mangenot S., Rattei T., Lehner A., Taylor M.W., Horn M., Daims H., Bartol-Mavel D., Wincker P., Barbe V., Fonknechten N., Vallenet D., Segurens B., Schenowitz-Truong C., Medigue C., Collingro A., Snel B., Dutilh B.E., Op den Camp H.J.M., Van der Drift C., Cirpus I., Van de Pas-Schoonen K.T., Harhangi H.R., Van Nif-trik L., Schmid M., Keltjens J., Van de Vossenberg J., Kartal B., Meier H., Frishman D., Huynen M.A., Mewes H-W., Weissenbach J., Jetten M.S.M., Wagner M., Le Paslier D. (2006). Deciphering the evolution and metabolism of an Anammox bacterium from a community genome. Nature, 440, 790-794.

Strous M., Fuerst J.A., Kramer E.H.M., Logemann S., Muyzer G., van de Pas-Schoonen K.T., Webb R., Kuenen J.G., Jetten, M.S.M. (1999). Missing lithotroph identified as new planc-tomycete. Nature, 400, 446-448.

Strous M., Heijnen J., Kuenen J. G., Jetten M.S.M. (1998). The sequencing batch reactor as a powerful tool for the study of slowly growing anaerobic ammonium-oxidizing microor-ganisms. Applied Microbiology and Biotechnology, 50, 589-596.

Strous M., van Gerven E., Zheng P., Kuenen J.G., Jetten M.S.M. (1997). Ammonium removal from concentrated waste streams with the anaerobic ammonium oxidation (ANAMMOX) process in different reactor configurations. Water Research, 31, 1955–1962.

Suchecka T., Lisowski W., Czykwin R., Piatkiewicz W. (2006). Landfill leachate: water recovery in Poland. Filtration and Separation, 43 (5), 34-38.

Surmacz- Górska J., Cema G., and Miksch K. (2004). Deamonification process in membrane assisted bioreactors. In: Integration and optimization of urban sanitation systems, Joint Polish-Swedish Seminars, Wisła, 2003, TRITA-LWR.REPORT 3007, 81-91.


70

Surmacz-Górska J. (2000). Usuwanie zanieczyszczeń organicznych oraz azotu z odcieków powstających w wysypiskach odpadów komunalnych (Removal of organics and nitrogen from landfill leachate). Inżynie-ria Środowiska, 44, Zeszyty Naukowe Politechniki Śląskiej (in Polish).

Surmacz – Górska J., Gernaey K., Demuynck C., Vanrolleghem P., Verstreate W. (1996). Nitrifi-cation monitoring in activated sludge by oxygen uptake rate (OUR) measurements. Water Research, 30 (5), 1228 – 1236.

Szatkowska B, Cema G., Plaza E., Trela J., Hultman B. (2007). One-stage system with partial nitritation and Anammox processes in moving-bed biofilm reactor. Water Science and Tech-nology, 55 (8-9), 19-26.

Szatkowska B., Plaza E. (2006). Temperature as a factor influencing the Anammox process per-formance. Water and Environmental Management Series, Young Researchers 2006, 51-58.

Szatkowska B. (2007). Performance and Control of Biofilm Systems with Partial Nitritation and Anammox for Supernatant Treatment. TRITA-LWR PhD Thesis 1035.

Szatkowska B. (2004). Treatment of ammonium-rich wastewater by partial nitritation/Anammox in a biofilm system. Licentiate thesis, Royal Institute of Technology, Stockholm, Sweden. TRITA-LWR.LIC 2023.

Third K.A., Paxman J., Schmid M., Strous M., Jetten M.S.M., Cord-Ruwisch R. (2005). Treatment of nitrogen-rich wastewater using partial nitrification and Anammox in the CANON process. Water Science and Technology, 52 (4), 47-54.

Third K. (2003). Oxygen management for optimisation of nitrogen removal in a sequencing batch reactor. PhD Thesis, School of Biological Sciences and Biotechnology, Murdoch University, Western Australia.

Thornton A., Pearce P., Parsons S.A. (2007). Ammonium removal from digested sludge liquors using ion exchange. Water Research, 41, 433 – 439.

Trela J., Płaza E., Szatkowska B., Hultman B., Bosander J., Dahlberg A-G. (2004). Deammonifi-kation som en ny process för behandling av avloppsströmmar med hög kvävehalt. Vatten, 60 (2) 119-127 (in Swedish).

Trigo C., Campos J.L., Garrido J.M., Mendez R. (2006). Start-up of the Anammox process in a membrane bioreactor. Journal of Biotechnology, 126, 475 – 487.

Wagner M., Rath G., Koops H.P., Flood J., Amann R. (1996). In situ analysis of nitrifying bacte-ria in sewage treatment plants. Water Science and Technology, 34, 237-244.

Wagner M., Rath G., Amann R., Koops H.-P., Schleifer K.-H. (1995). In situ identification of ammonia-oxidizing bacteria. Systematic and Applied Microbiology, 18, 251-264.

Wang Y., Liu S., Xu Z., Han T., Chuan S., Zhu T. (2006). Ammonia removal from leachate solu-tion using natural Chinese clinoptilolite. Journal of Hazardous materials, B136, 735 – 740.

Welander U., Henrysson, T., Welander, T. (1998). Biological nitrogen removal from municipal landfill leachate in a pilot scale suspended carrier biofilm process. Water Research, 32 (5), 1564 – 1570.

Welander U., Henrysson, T., Welander, T. (1997). Nitrification of landfill leachate using sus-pended-carrier biofilm technology. Water Research, 31 (9), 2351 – 2355.

Windey K., Do Bo I., Vestraete W. (2005). Oxygen-limited autotrophic nitrification-denitrification (OLAND) in a rotating biological contactor treating high-salinity wastewa-ter. Water Research, 39, 4512 – 4520.

Wiszniowski J. (2006a). Combination of the biological and photocatalytic methods in the process of municipal landfill leachate treatment. PhD thesis, Environmental Biotechnology Department, The Sile-sian University of Technology, Gliwice, Poland.


71

Wiszniowski J., Robert D., Surmacz-Górska J., Miksch K., Weber J.V. (2006b). Landfill leachate treatment methods: A review. Environmental Chemistry Letters, 4, 51-61.

Wyffels S., Pynaert K., Boeckx P., Vestraete W., Van Cleemput O. (2003). Identification and quantification of nitrogen removal contactor by 15N tracer techniques. Water Research, 37, 1252 – 1259.

van de Graaf A.A., de Bruijn P., Robertson L.A., Jetten M.S.M., Kuenen J.G. (1997). Metabolic pathway of anaerobic ammonium oxidizing microorganism in a fluidized bed reactor. Microbiology, UK., 143, 2415-2421.

van de Graaf A.A. , de Bruijn P., Robertson L.A., Jetten M.S.M., Kuenen J.G. (1996). Autotro-phic growth of anaerobic ammonium-oxidizing micro-organisms in a fluidized bed reac-tor. Microbiology, UK, 142, 2187-2196.

van de Graaf A.A., Mulder A., de Bruijn P., Jetten M.S.M., Robertson L.A., Kuenen J.G. (1995). Anaerobic oxidation of ammonium is a biologically mediated process. Applied and Envi-ronmental Microbiology, 61, 1246-1251.

van der Star W. (2008). Growth and Metabolism of Anammox Bacteria. PhD Thesis, ISBN: 978-90-9022771-9.

van der Star W., Abma W.R., Blommers D., Mulder J.W., Tokutomi T., Strous M., picioreau C., Van Loosdrecht M.C.M. (2007). Startup of reactor for anoxic ammonium oxidation: Ex-periences from the first full-scale anammox reactor in Rotterdam. Water Research, 41 (18), 4149-4163.

van Dongen L. G.J.M., Jetten M.S.M., van Loosdrecht, M.C.M. (2001a) The combined Sharon/Anammox Process, A sustainable method for N-removal from sludge water. STOWA Re-port, IWA Publishing, London, UK.

van Dongen L. G.J.M., Jetten M.S.M., van Loosdrecht M.C.M. (2001b). The Sharon-Anammox process for treatment of ammonium rich wastewater. Water Science and Technology, 44 (1), 153-160.

van Loosdrecht M.C.M., Salem S. (2005). Biological treatment of sludge digester liquor. In: Pro-ceeding of the IWA Specialized Conference: Nutrient Management in Wastewater Treatment, Processes and Recycle Streams, 19 – 21 September 2005, Kraków, Poland, 13 – 22.

Vázquez-Padín J., Fernádez I., Figueroa M., Mosquera-Corrala A., Campos J-L., Méndez R. (2009). Applications of Anammoxnext term based processes to treat anaerobic digester supernatant at room previous temperature. Bioresource Technology, 100 (12), 2988-2994.

Vázquez-Padín J.R., Pozo M.J., Jarpa M., Figueroa M., Franco A., Mosquera-Corral A., Campos J.L, Méndez R. (2008). Treatment of anaerobic sludge digester effluents by the CANON process in an air pulsing SBR. Journal of Hazardous Materials – in press Available online 24 November 2008.

Volcke E.I.P., Gernaey K.V., Vrecko D., Jeppsson U., van Loosdrecht M.C.M., Vanrolleghem P.A. (2006). Plant-wide (BSM2) evaluation of reject water treatment with a SHARON-Anammox process. Water Science and Technology, 54 (8), 93-100.

Volcke E.I.P., Van Hulle S.W.H., Donckels B.M.R., van Loosdrecht M.C.M., Vanrolleghem P.A. (2005). Coupling the SHARON process with Anammox: Model-based scenario analysis with focus on operating costs. Water Science and Technology, 52 (4), 107 -115.

Zhang H-g., Zhou S-q. (2007). Treating leachate mixture with anaerobic ammonium oxidation technology. Journal of Central South University of Technology, 13 (6), 663 – 667.

Zhang S-j., Peng Y-z., Wang S-y., Zheng S-w., Guo J. (2007). Organic matter and concentrated nitrogen removal by shortcut nitrification and denitrification from mature municipal land-fill leachate. Journal of Environmental Sciences, 19, 647 – 651.


72

Ødegaard H. (2006). Innovation in wastewater treatment: the moving bed biofilm process. Water Science and Technology, 53 (9), 17-33.

Ødegaard H., Gisvold B., Strickland J. (2000). The influence of carrier size and shape in the mov-ing bed biofilm process. Water Science and Technology, 41 (4-5), 383-391.

“Environment 2007” collective study of the Central Statistical Office (GUS), Poland. Available form: www.stat.gov.pl

Documents

gcema PhD thesis corrected - DiVA portal290602/FULLTEXT01.pdf · TRITA-LWR PhD Thesis 1053 ISSN 1650-8602 ISRN KTH/LWR/PHD 1053-SE ISBN 978-91-7415-501-3 ... Thanks go also to master