Anaerobic ammonium oxidation (anammox)
Anaerobic ammonium oxidizing bacteria (AnAOB) are known to oxidize ammonium to nitrogen gas with nitrite as electron acceptor under anoxic conditions, forming nitrate in the process (Figure 1.5, pathway 4). This transformation is done in three consecutive, coupled reactions with two intermediates, nitric oxide (NO) and hydrazine (N2H4). First, nitrite is reduced to NO with nitrite reductase (nir) which is then, as we understand, condensed with ammonium to form hydrazine, via a αβγ multi-enzyme complex hydrazine synthase (HZS) (Dietl et al., 2015). Finally, hydrazine is oxidized to dinitrogen gas by hydrazine dehydrogenase (hdh) (Kartal et al., 2012).
All reported AnAOB belong to the order of the Brocadiales (phylum Planctomycetes) and can be divided into five “Candidatus” genera. Four of these genera (Ca. Kuenenia, Ca. Brocadia, Ca. Anammoxoglobus and Ca. Jettenia) have been enriched from activated sludge (Strous, et al., 1999). The fifth (Ca. Scalindua) has mostly been detected in natural habitats such as marine sediments and oxygen minimum zones (Jetten, et al., 2009). Of the above-mentioned genera, Ca. Brocadia and Ca. Kuenenia are most abundant in engineered anammox systems, however a consensus is yet to be found on how key drivers such as high/low affinity for substrate, sensitivity to low temperature and/or to inhibitors etc. impact the niche differentiation between different AnAOB genera (see Chapter 3).
At first it was considered that AnAOB were obligate autotrophs and thus unable to convert organic carbon substrate. However, recent studies have reported that some AnAOB species including Ca. Jettenia asiatica, Ca. anammoxoglobus propionicus, Ca. Brocadia fulgida and Ca. Kuenenia stuttgartiensis have the capacity to oxidize smaller organic electron donors such as acetate and propionate with nitrate as an electron acceptor while forming ammonium with nitrite as intermediate (Güven et al., 2005; Kartal et al., 2007; Kartal et al., 2008; Shu et al., 2016). This organotrophic capacity has advantages for wastewater treatment as it would allow to further reduce nitrate in the effluent. Also, since the oxidized organic carbon is completely oxidized to CO2 and not incorporated into biomass, the biomass yield is reduced and less excess sludge is produced (Boran et al., 2007; Winkler et al., 2012). It is easy to imagine how this can potentially affect the competition between AnAOB and HB when organic carbon is present.
• Autotrophic denitrification which is the reduction of nitrate and/or nitrite to nitrogen gas using Hydrogen (produced in situ from fermentation) or Sulphur-compounds as electron donors. This process could potentially replenish the nitrite pool and increase the nitrogen removal by (Speth et al., 2016). This process can be an interesting approach for wastewaters with low bCOD/N since the (expensive) addition of external COD could be avoided by introducing more elegant solutions e.g. the use of packed bed reactors containing limestone (Vandekerckhove et al., 2018).
• Complete ammonium oxidation (Comammox) is the combined ammonium and nitrite oxidation performed by a single organism during growth via ammonium oxidation to nitrate. This process, performed by certain species of the Nitrospira genus, has been reported to co-exist with the anammox process under oxygen limiting conditions (Daims et al., 2015; van Kessel et al., 2015). However, given their oligotrophic lifestyle and low specific activities (Kits et al., 2017), it is unlikely that these organisms would be significantly abundant in PN/A systems.
Table 1.1 Overall stoichiometry of the 4 key microbial processes (nitritation, nitratation, denitrification and anammox) as well as combinations conventionally applied during Biological nitrogen removal (nitrification/denitrification and partial nitritation/anammox) (after Desloover, 2013; Vlaeminck, 2009).
Until now, several PN/A strategies have been proposed to steer microbial competition, but some are not yet reproduced and lack general consensus. These strategies aimed at (1) promoting growth and activity of AerAOB, AnAOB, and tollerating the activity of nitrite and nitrate reducing heterotrophs (HBNOX-) while suppressing the aerobic activity of NOB, we label this as “ON/OFF” control; and (2) washing-out NOB and heterotrophs from the reactors, while selectively retaining (and seeding) AerAOB and AnAOB, labelled as “IN/OUT” control (Figure 1.7).
Studies based on the ON/OFF control strategy implemented specific oxygen and/or substrate supply patterns or controlled exposure to certain inhibitors to steer the metabolic state of the different microbial groups in the process. Some examples of strategies are:
(a) maintaining a residual ammonium concentration (i.e. 2-4 mg N L-1) for efficient NOB suppression in PN/A (Poot et al., 2016). It allows sufficient oxygen limitation in biofilms which helps obtain nitritational granular reactors and protect AnAOB from oxygen inhibition (Lotti et al., 2015). In floccular systems, the specific growth rate of AerAOB is promoted to ensure that the dissolved oxygen (DO) is the rate limiting parameter during aeration (Isanta et al., 2015; Third et al., 2001).
(b) aeration control using either Continuous low DO-setpoints (< 0.2 mg O2 L-1) to minimize AnAOB oxygen inhibition, and increase their competitiveness for nitrite in the biofilm (Pérez et al., 2014; Wett et al., 2013) or Intermittent aeration or so-called “transient anoxia” to balance
the periodic supply of oxygen to exploit the nitratational lag (minimum 15-30 min. anoxic) (Laureni et al., 2016; Morales et al., 2016), complete nitrite consumption in the anoxic phase and limit AnAOB inhibition by oxygen (Gilbert et al., 2014a); Seuntjens D. et al., submitted). Typically, higher DO-setpoints (> 1.5 mg O2 L-1) are used to maximize activity of AerAOB over NOB (Han et al., 2016; Isanta et al., 2015; Kornaros et al., 2010; Third et al., 2001).
(c) exposing flocs to inhibitors such as free ammonia (FA) and free nitrous acid (FNA) to suppressed NOB with 80-90% nitritation in a floccular reactor (Agrawal et al., 2017b). As FA and FNA cannot reach inhibitory concentrations in the mainstream, a return-sludge treatment, that exposed thickened flocs from the clarifier, has been proposed.
IN/OUT control refers to selective control of the sludge retention time (SRT) of different sludge fractions. This is especially relevant under mainstream conditions where lower temperatures (10-15°C) lower the growth rates and activities of the desired organisms. Long biofilm SRT are required to retain AnAOB due to their slow growth rate, especially under low-temperature mainstream conditions (SRT =70d at 15°C, >100d at 10 °C) (Pérez et al., 2014; Yang et al., 2017). Therefore, biofilm-based reactors have been used, mainly as granule (Wett et al., 2013), or carrier material (Wang et al., 2017) configurations. In contrast, a short enough flocculent SRT to selectively washout NOB, yet retain AerAOB (Kornaros et al., 2010; Third et al., 2001); Seuntjens et al., unpublished). To bring these conflicting worlds together, one-stage hybrid systems (= granule/biofilm + floc) (Lemaire et al., 2014; Pérez et al., 2014) have also been validated to achieve simultaneous, short-floc and long-biofilm SRT, allowing NOB washout from suspension and AnAOB retention in the biofilm. Another strategy might be the separation of nitritation and anammox in a two-stage approach. Special attention should also be paid to the proper control of (in)organic suspended solids (SS) entering the reactor since these SS or the flocculent used to improve their settleability in the previous stage, are reported to attach to AnAOB biomass inside the reactor, resulting in a loss of activity (Yamamoto et al., 2008).
IN/OUT + ON/OFF control = reactor solution
The possible combination of various strategies belonging to the “ON/OFF” and/or “IN/OUT” approaches have been advocated for NOB out-selection. For instance, in a pilot study (Third et al., 2001) based on suspended biomass, operated at 25oC, a combination of short aerobic SRT, intermittent aeration at high DO concentration and residual ammonium was successful for NOB wash-out. In a granular biomass reactor (Malovanyy et al., 2015) operated at 15oC, shorter SRT of the flocculent fraction with continuous aeration at low DO set-point also demonstrated NOB wash-out. Intermittent aeration at a low DO set-point, and strict SRT to just retain AerAOB and wash-out NOB also worked in a hybrid reactor (suspended and carrier-based biomass, (Lawson et al., 2017).
Figure 1.7 – Strategies for design and operation of a one- or two-stage partial nitritation/anammox (PN/A) reactor. NOB: Nitrite oxidizing bacteria. AerAOB: aerobic ammonium oxidizing bacteria. AnAOB: Anoxic ammonium oxidizing bacteria. HB: Heterotrophic bacteria. HBNOx˗: Heterotrophic bacteria reducing nitrite or nitrate. SRT: Sludge retention time. bCOD/N: biodegradable chemical oxygen demand over nitrogen (Agrawal et al., 2018).
Uncontrollable wastewater parameters requiring process adaptation
Residual organic carbon
Municipal wastewater has a high carbon to nitrogen ratio (around 6-10 gCODtot/gNtot), because of this, a first stage aimed at maximal recovery of organic carbon (known as “C-Stage”) has been proposed in the past few years. This should also ease the implementation of mainstream PN/A (known as “N-Stage”). Nevertheless, the presence of heterotrophic bacteria is inevitable in mainstream PN/A processes (Speth et al., 2016), due to the availability of residual COD in the effluent of the C-stage or soluble metabolic products (SMP) released from AerAOB and AnAOB. This can allow fast growing HBNOX- to compete with the slow growing AnAOB for nitrite and space. On the upside HBNOX- can potentially (a) increase nitrogen removal efficiency by denitrifying the nitrate produced by the anammox reaction (Jenni et al., 2014) and (b) suppress NOB by reducing nitrite availability (Third et al., 2001). However, there is no consensus about the impact of organic carbon on the balance between anammox and denitrification and how it is impacted by low temperatures. Answering these questions is a key step towards successful MRM for mainstream PN/A.
Table of contents :
CHAPTER I: INTRODUCTION
1.2. FROM WASTEWATER TREATMENT PLANT (WWTP) TO WATER RESOURCE RECOVERY FACILITY (WRR)
1.3. KEY MICROBIAL PROCESSES FOR BIOLOGICAL NITROGEN REMOVAL (BNR)
1.4. MAINSTREAM PN/A: FROM ECOPHYSIOLOGY TO PROCESS TECHNOLOGY THROUGH MICROBIAL RESOURCE MANAGEMENT (MRM)
1.6. RESEARCH QUESTIONS AND THESIS OUTLINE
CHAPTER II INSTANT COLD TOLERANCE IMPACTED BY ANAMMOX GENUS RATHER THAN BY AGGREGATE SIZE
2.2 MATERIALS AND METHODS
2.2.1. Types of biomass
2.2.2. Determination of specific ammonium removal rate (SARR)
2.2.3. Size fractionation and disaggregation treatment
2.2.4. Determination of activation energy and temperature coefficient
2.2.5. Analytical methods
2.2.6. Microbial community analysis: 16S rRNA gene amplicon sequencing
2.3.1 Biomass composition and aggregate size
2.3.2. Effect of temperature on AnAOB activity
Role of aggregate size
2.7. CONFLICTS OF INTEREST
CHAPTER III ENRICHMENT AND ADAPTATION YIELD HIGH ANAMMOX CONVERSION RATES UNDER LOW TEMPERATURES
3.2. MATERIAL AND METHODS
3.2.1 Set-up and operation of the reactors
3.2.2 Biomass inoculum mix
3.2.3 Anammox activity and chemical analyses
3.2.4 Particle size distribution of the biomass aggregates
3.2.5 Microbial community analyses
3.3.1. Reactor performance
126.96.36.199. Start-up of the reactors
188.8.131.52. Activity evolution at constant temperature (30°C)
184.108.40.206. Activity evolution at decreasing temperature (30°C to 10°C)
3.2 BIOMASS AGGREGATE SIZE AND MICROBIAL COMMUNITY ANALYSIS
3.2.1 Biomass particle size distribution
3.2.2 Evolutions in the microbial community
3.4.1 Enrichment and adaptation favoring high specific activities
3.4.2 Potential AnAOB genus niche differentiation
3.4.3 Towards implementation of partial nitritation/anammox
CHAPTER IV IMPACT OF SLOWLY BIODEGRADABLE ORGANIC CARBON ON THE COMPETITION BETWEEN ANAMMOX BACTERIA AND DENITRIFIERS AT DIFFERENT TEMPERATURES.
IMPACT OF SLOWLY BIODEGRADABLE ORGANIC CARBON ON THE COMPETITION BETWEEN ANAMMOX BACTERIA AND DENITRIFIERS AT DIFFERENT TEMPERATURES.
4.2 MATERIAL AND METHODS
4.2.1 Set-up and operation of the reactors
4.2.3 Anammox activity and chemical analyses
4.2.4 Microbial community analyses
4.3.1. Reactor performance
220.127.116.11. Activity evolution at high temperature (30°C)
18.104.22.168. COD removal
4.3.2 Biomass aggregate size and microbial community analysis
22.214.171.124 Transition from granular system to hybrid system with flocs
126.96.36.199 Evolutions in the microbial community
4.4.1 Competition between AnAOB and HB for nitrite during
4.4.2 Differential SRT favors retention of AnAOB over HB at low temperature
CHAPTER V CONCLUSIONS
5.2. MAJOR FINDINGS
5.3. IMPLICATIONS FOR MICROBIAL RESOURCE MANAGEMENT
5.4. DESIGN CHOICES FOR MAINSTREAM PN/A SYSTEMS