Chapter 1 gives a short overview of the history of aerobic granular sludge technology and finishes with an outline of the thesis. Chapter 2 deals with segregation of biomass as a function of height of the sludge bed. Phosphate accumulating organisms were found to dominate at the bottom of the sludge bed, whereas Glycogen accumulating organisms dominated at the top of the sludge bed. By selective removal of glycogen accumulating organisms dominated sludge from the top of the sludge bed, more than 95% P removal efficiencies were achieved at 30°C. Based on current knowledge this is the first process in which a stable biological P-removal could be maintained at 30°C. In chapter 3 the selective sludge removal was studied within this enhance biological phosphorous removal (EBPR) system in more detail. At 30°C the removal of bottom sludge from the PAO-rich part of the sludge bed in minor proportions did not negatively affect P-removal and allowed to obtain biomass with a lower ash content. The research further shows that at 20ºC, selective removal of PAOs was not crucial for a stable bioP removal and biomass can be removed equally throughout the sludge bed. Our results indicate that high ash content and density of bottom granules positively correlated with the presence of PAO-dominated granules.
In chapter 4 we studied the competition for nitrite between nitrite oxidizing bacteria (NOB) and anaerobic ammonium oxidizing bacteria (Anammox) at low temperatures. White granules were dominated by NOB bacteria and were mainly located at the top of the settled sludge bed whereas red granules were dominated by Anammox bacteria and were located at the bottom. Granules from the top of the sludge bed were smaller and therefore had a larger aerobic volume fraction. These smaller granules also furthermore had a lower density then larger granules and consequently a slower settling rate. Selective sludge removal from the top of the settled sludge bed selectively removed NOB resulting in an increased overall biomass specific N-conversion. This forms an option for obtaining a stable Anammox process at lower temperatures in municipal wastewater treatment systems. Chapter 5 investigates the effect of granular density on the settling velocity of individual granules. The granule was divided in different layers each occupying a certain volume fraction consisting out of either bacteria, extra polymeric substances or precipitates. The density of each fraction was estimated experimentally. Each volume fractions was coupled with the corresponding densities to calculate a total density of a granule. This was used to calculate settling velocities. Results revealed that Phosphate accumulating organisms (PAO) had a higher density than glycogen accumulating organisms leading to significantly higher settling velocities for PAO dominated granules explaining earlier observations of the segregation of the granular sludge bed inside reactors. The model showed that a small increase in the volume fraction of precipitates (1-5%) strongly increased the granular density and thereby the settling velocity. For nitritation-anammox granular sludge the settling model shows that density differences are not very important and segregation of the biomass in the bed is mainly caused by variations in granule radius.
In chapter 6 we showed that the temperature and ionic strength dependent density and viscosity changes of water have great impact on settling velocity of granular sludge. The corresponding slow settling of small granules at decreased water viscosities and increased water densities as caused by a lower temperature can be an important reason for the reported troublesome start-up of granular sludge reactors at low temperatures. Settling velocities also decreased with increasing salt concentrations. Changes in salt concentration will cause a strong time dependent effect of settling of granules due to the slow diffusion of salts into the granules. Conductivity and temperature measurements can therefore be used as an additional operational factor to stabilize and improve biomass retention in granular sludge technology.
Chapter 1 gives a short introduction in microbial ecology. Chapter 2 of the second book reports the differences in the microbial community composition of flocculent sludge and granular sludge. Their total bacterial community composition were very dissimilar whereas the community assessment showed that both systems had on average a similar species richness entropy, and evenness, suggesting that although the bacterial groups where very dissimilar a same stability in microbial community and function was obtained. The AOB population showed more unevenness than it was the case for the total bacterial populations. A correlation between the ammonium oxidizing bacterial population and changes in ammonium removal efficiency as well as temperature was found for both systems, whereas the bacterial population correlated with total nitrogen removal efficiencies.
Chapter 3 aims to unravel the reasons for the disproportion in the ratio of AOB and NOB in aerobic granular sludge. In this study, we analysed the nitrifying microbial community (ammonium-oxidizing bacteria - AOB and nitrite-oxidizing bacteria - NOB) within three different aerobic granular sludge treatment systems as well as within one flocculent sludge system. Fluorescent in situ hybridization (FISH) and quantitative-PCR (qPCR) showed that Nitrobacter was the dominate NOB in acetate fed aerobic granules. In the conventional system, both Nitrospira and Nitrobacter were present in similar amounts. This suggested that the growth of Nitrobacter within aerobic granular sludge was partly uncoupled from the lithotrophic nitrite supply from AOB. This was supported by activity measurements which showed a 3 fold higher nitrite oxidizing capacity than ammonium oxidizing capacity. Based on these findings, two hypotheses were considered: either Nitrobacter grew mixotrophically by acetate-dependent dissimilatory nitrate reduction (ping-pong effect) or a nitrite oxidation/nitrate reduction loop (nitrite loop) occured in which denitrifiers reduced nitrate to nitrite supplying additional nitrite for the NOB apart from the AOB. The disproportion of the amount of AOB and NOB in granular sludge should be investigated further to confirm the hypothesis made in this work.
In chapter 4 the specific solid retention time for different bacteria within flocculent and granular sludge was determined. Samples were collected from reactor and effluent sludge and the number of a specific bacterial group was evaluated in respect to the total bacterial community by the means of quantitative polymerase chain reaction (qPCR). Operational data were combined with molecular techniques and the SRT of each individual microorganism could be calculated. It was further observed that protozoa were grazing on the bacterial community within the system indicating that they have the potential to shorten the specific SRT of bacteria. Archea were not found in the flocculent system but were present in small amounts within the granular system. Chapter 5 gives outlook about possible applications molecular techniques in wastewater treatment.