Search

Two pilot-scale trickling filters were constructed and tested for manganese removal from potable water, using different fractions of silicic gravel as support media (mono- and multilayer filter). Manganese oxidation in drinking water was found to be cause by both biological oxidation and heterogeneous catalytic paths. Mixed culture populations were used to inoculate the trickling filters and the feed manganese concentrations and volumetric flow rates (VFRs) were between 0.6–2.0 mg/l and 500–2000 ml/min, respectively. The monolayer filter was flooded for high VFRs, and it was very effective for all conditions tested (100% removal efficiency, up to 2850 mg Mn/day). The multilayer filter was less effective for high manganese concentrations but it could remove up to 3250 mg Mn/day. A new mathematical model was developed assuming heterogeneous autocatalytic and biological as the main oxidation manganese paths. First order kinetics was used to describe the heterogeneous catalytic oxidation, while Monod-type kinetics was used to describe the net biological manganese oxidation. The simplicity of the pilot-scale design, the lack of need for an external mechanical aeration source and the ability to predict operation of the system offers a very attractive solution for manganese removal from potable water.

The kinetics of phenol biodegradation in draw-fill suspended and attached growth reactors by olive pulp bacteria were investigated. The Haldane equation was adopted in order to describe phenol biodegradation in the draw-fill suspended-growth flask reactors. Kinetic constants of the Haldane equation obtained from batch culture experiments with phenol concentration ranging from 13 to 280 mg l−1, were μmax=0.054 h−1, KS=19.23 mg l−1, and Ki=36.95 mg l−1. However, in the draw-fill experiments at the packed-bed reactor, phenol inhibitory effects were minimized increasing the inhibitory constant value Ki′ at 1571 mg l−1. The model predicted phenol biodegradation in the packed-bed reactor adequately, for all initial phenol concentrations tested, apart from the highest phenol concentration of 1990 mg l−1, where clogging problems and insufficient aeration in the filter prevailed.

The efficiency of packed bed reactors on the biological degradation of phenol was studied in this work. In particular, the effect of the specific surface area of the support material and the operating mode on biodegradation efficiency were determined. Phenol removal was studied using two pilot-scale packed bed reactors loaded with different support materials, i.e., calcite gravel and plastic tubes. The draw-fill operating mode was tested for the two reactors, while in the case of the plastic tube packed-bed reactor the draw-fill operating mode with recirculation was also investigated. Indigenous bacteria from olive pulp were enriched and used as inoculum for the reactors. The enrichment of the bacterial community capable for phenol degradation was based on natural selection. The maximum removal rate achieved at the two reactors filled with gravel support material and plastic tubes under draw-fill operation was found to be 25.3 g phenol/d and 14.03 g phenol/d respectively, for a feed phenol concentration of 500 mg/l. Increase of feed phenol concentration to about 2750 mg/l demonstrated inhibitory effects on bacterial growth, while the corresponding removal rate was reduced. The operating mode of draw-fill with recirculation for the packed-bed reactor loaded with plastic tubes, resulted in a maximum removal rate of 29.9 g phenol/d for feed phenol concentration of about 500 mg/L, which even exceeded the maximum recorded removal rate of the gravel media reactor, under draw-fill operation.

The purpose of the present study was to estimate Cr(VI) removal through biological mechanisms in biofilm reactors operated in SBR operating mode with recirculation using different filter media types. The choice of the support material is of great importance since chromate reduction is followed by the formation of sediments, which causes obstruction of the flow along the filter depth. In order to overcome this awkwardness, we tested the performance of two different support materials, i.e. plastic media and calcitic gravel, in pilot-scale trickling filters. The feed concentrations of Cr(VI) examined were about 5, 30 and 100 mg/l, while the concentration of the organic carbon was constant at 400 mg/l, in order to avoid carbon limitations in the bulk liquid. Plastic media showed better performance at the different Cr(VI) concentrations examined, compared to the gravel media. The removal rates for the plastic media achieved were 4.23±0.18, 3.62±0.1 and 3.3±0.08 g Cr(VI)/d, for feed concentration of Cr(VI) about 5, 30 and 100 mg/l respectively, while for the gravel media the corresponding removal rates were 4.11±0.09, 3.52±0.06 and 2.5±0.07 g Cr(VI)/d.

Pilot scale trickling filters were constructed and tested in order to study biological removal of ammonia, iron and manganese from potable water. The effect of the size of the support material on nitrification performance was studied extensively. The mean size of the gravel and hence, the specific surface area was found to be critical for optimal nitrification operation. A steady-state model developed in previous work was used to predict filter's performance. The model was very accurate only for the gravel size for which maximum nitrification rates were observed. The effect of the operational conditions on the physico-chemical and combined physico-chemical and biological iron oxidation was also studied. It was found that the contribution of biological oxidation is significant, increasing filter's efficiency by about 6% and reducing the required filter depth by about 40%. Manganese biological removal was studied using gravel with small mean diameter, thus providing high specific surface area. Feed concentrations up to 4.0 mg/l were treated sufficiently. Finally, experiments were performed to investigate the simultaneous removal of ammonia, iron and manganese. Experimental results showed that the combined, as well as the simultaneous removal of the aforementioned pollutants, can be achieved by single-step filtration.

The capability of olive fruit bacteria to remove chemical oxygen demand (COD) and phenolic compounds from olive mill wastewater (OMW) using flasks and packed bed reactors was tested. Batch aerobic experiments were performed in flask reactors, with OMW at different dilutions (20%, 50%, and 100%). The maximum phenolic and dissolved COD removal reached up to 82–90% for the dilutions of 20%, 50%, and 100%, in 11, 23, and 30 days, respectively. Experiments were also conducted in a bench and a pilot-scale packed bed reactor, operated in the laboratory and at an olive mill, respectively. At 281C the maximum phenolic and dissolved COD removal, recorded at the bench-scale reactor, reached about 60% and 70%, respectively, for a time period of only 27 h. At the pilot-scale reactor, where the maximum daily temperature varied between 6 and 201C, the maximum phenolic and d-COD removal reached about 55% and 70% in 5 and 3 days, respectively. These results indicate an economically feasible method for OMW biodegradation. r 2007 Elsevier Ltd. All rights reserved.

The survival of free-living nitrogen-fixing microbial populations in the natural ecosystem is crucial for the system maintenance and productivity due to the unique role of these organisms in the global biogeochemical cycle of nitrogen. The dynamics of a nitrogen-fixing microbial population grown at various conditions in a chemostat, together with a competitive (for the common resources) population, was studied through bifurcation analysis of a mathematical model of the system. When the carbon source is found in abundance in the feed, then the competition for this nutrient is low. High amounts of ammonium nitrogen (a substance that inhibits growth of the nitrogen-fixing population) are assimilated for the growth needs of both populations. Under these conditions the nitrogen-fixing population can survive (alone or together with its competitor) in a wide range of parameter values and operating conditions and in some cases nitrogenase synthesis occurs. When the inflow medium contains low carbon substrate concentrations, high competition occurs for this nutrient. In these conditions the nitrogen-fixing population can survive only if it has the competitive advantage over its competitor. However, if the inflow medium contains high ammonium concentrations, the nitrogen-fixing population is inhibited and loses its competitive advantage. Under these conditions, only nitrogen-fixing populations, which are able to establish amensalistic interactions can survive in the system.

A pilot-scale trickling filter was constructed and tested for iron removal from potable water. Iron removal was found to be caused by both biological and physico-chemical iron oxidation. The extent of each oxidation type was assessed. The system was inoculated with mixed culture and its performance was tested under continuous operation at 14 °C. Feed iron concentrations and volumetric flow rates (VFR), ranged between 1–4 mg/l and 1000–3000 ml/min (225–680 m 3 /m 2 day), respectively. First order kinetics was used to describe the physico-chemical iron oxidation while Monod-type kinetics was used to describe the net biological iron oxidation. An increase of VFR from 1000 to 3000 ml/min reduced the filter's physico-chemical removal efficiency from 93 to 80%, while it remained constant for all the iron feed concentrations (1–4 mg/l) that were tested at each VFR. Bio-oxidation improved filter efficiency by about 5–6%. In all cases there was very good agreement between model predictions and experimental data. A series of experiments were also performed in order to investigate the simultaneous ammonia and iron oxidation. It was found that ammonia influence on iron removal becomes substantial only for high iron and ammonia concentrations. The simplicity of the pilot-scale design, the lack of the need for an external mechanical aeration source and the ease for the design and prediction of the system operation offers a very attractive solution for iron removal from potable water.

The progress of hydrogenotrophic denitrification in batch assays with a mixed culture was examined. The aim was to study the performance of the culture for hydrogenotrophic denitrification of drinking water at various NO3− concentrations. The kinetics of the process were studied in batch experiments. The rates of the process were found to be inhibited at high nitrate concentrations. The nitrite concentration remained at very low values throughout the process but it still affected biomass growth. A kinetic model was developed and its kinetic parameters were determined from the batch experiments. The growth kinetics could be very well described by using expressions for double nutrient limitation (nitrate, nitrite) with inhibition from nitrate. The proposed model is capable of describing accurately enough cellular growth, nitrate and nitrite utilization and utilization rates under anoxic conditions in the presence of various concentrations of nitrate in the range of 7 to about 200 mg N/l.

Phenol removal was studied using batch cultures of free suspended cells and attached growth processes. Indigenous bacteria from olive pulp were enriched and used as inoculum for the filter and a suspended-growth flask reactor. A pilot-scale packed bed reactor was constructed and operated for biological phenol removal from industrial wastewater. The packed bed reactor was found to be more resistant to high phenol concentrations and led to significantly higher removal rates than the suspended-growth reactor. Two different operating modes were used to investigate the optimal performance of the filter, i.e., continuous and draw-fill. The latter was found to achieve removal rates up to 12.65 g phenol/(l d), while the continuous operating mode achieved removal rates only up to 0.082 g phenol/(l d).The low operating cost combined with the high phenol removal rates indicates that the above technology may offer a feasible solution to a serious environmental problem.

The strategy of alternative prey (switching), regulated by a catabolic repression control-like (CRCL) mode, in which the desirable prey acts as repressor on the predator's attack on the alternative prey, was studied in a two-prey–one-predator chemostat system. In this system, alternative prey has to survive predation and competition for nutrients from the desirable prey. Therefore, when the alternative prey has no competitive advantage for the common substrate over the desirable prey, its survival depends on the protection offered by the desirable prey via CRCL. In this case, CRCL allows the coexistence of both desirable and alternative prey and predator populations. However, when the alternative prey has the competitive advantage over the desirable prey, CRCL negatively affects both the state of survival of the desirable prey and the coexistence state.

Due to the nitrogenous oxygen demand and the carcinogenic chloramines that are formed, ammonia needs to be removed from potable water before it can be consumed. Ammonia may be removed chemically or biologically. Biological ammonia removal (nitrification) is carried out by two different nitrifying bacteria, Nitrosomonas and Nitrobacter. Due to the low specific growth rates of these bacteria, high nitrification rates can be secured only in attached growth systems such as trickling filters. This allows retaining the slow-growing nitrifiers even at high hydraulic loadings. The choice of filter medium is of great importance because it provides the surface for bacterial growth and determines the air and water flow characteristics of the filter. Recirculation may enhance nitrification in trickling filters for high hydraulic and ammonia loads, whereas iron and manganese inhibit nitrification. Modeling of nitrification in trickling filters allows proper design of the filters and prediction of their performance. Such mathematical models have been used to develop operating diagrams that may be used to ensure safe and complete nitrification of potable water for a wide range of flow rates and ammonia feed concentrations. Keywords: nitrification; potable water; trickling filters; modeling; support media; recirculation; operating diagram

We studied the steady-state behavior of a mathematical model of a nitrifying trickling filter. In particular, we studied the effect of the operating conditions of the filter on the complete and safe nitrification. We presented the results with the help of the operating diagram of the system and we determined the range of operating conditions resulting in optimal operation. We also computed biofilm thickness along the filter depth and concentration profiles inside the biofilm, and compared them with experimental results found in the literature. The comparison shows very good qualitative agreement between model predictions and experimental data.

Biodegradation in porous media is studied with carefully controlled and well-characterized experiments in model porous media constructed of etched glass. Porous media of this type allow visual observation of the phenomena that take place at pore scale. An aqueous solution of five organic pollutants (toluene, phenol, o-cresol, naphthalene and 1,2,3-trimethylbenzene) was used as a model NAPL (representing creosote). The bacteria used were Pseudomonas fluorescens, which are indigenous (even predominant) in many contaminated soils. The maximum aqueous concentrations of the specific organic substances, below which biodegradation becomes possible, were determined as a function of temperature from toxicity experiments. Visualization experiments were made under various flow velocities and organic loadings to study the morphology and thickness of the biofilm as a function of the pore size and the distance from the entrance, and the efficiency of biodegradation. The efficiency of biodegradation decreased as the aqueous concentration of NAPL at the inlet increased and/or as the flow velocity increased. The thickness of biofilm decreased as the distance from the inlet increased and/or the pore diameter decreased. A quasi-steady-state theoretical model of biodegradation was used to calculate the values of the mesoscopic biochemical rates and to predict the profile of NAPL concentration in the porous medium and the thickness of biofilm in pores. The agreement between experimental data and model predictions is quite satisfactory.

We analyze a mathematical model of a simple food web consisting of one predator and two prey populations in a chemostat. Monod's model is employed for the dependence of the specific growth rates of the two prey populations on the concentration of the rate-limiting substrate and a generalization of Monod's model for the dependence of the specific growth rate of the predator on the concentrations of the prey populations. We use numerical bifurcation techniques to determine the effect of the operating conditions of the chemostat on the dynamics of the system and construct its operating diagram. Chaotic behavior resulting from successive period doublings is observed. Multistability phenomena of coexistence of steady and periodic states at the same operating conditions are also found.

Although hemochromatosis and pathological situations due to chronic iron overload have been extensively described, there is little information about the influence of iron on other trace elements in the cell. The aim of this study was to investigate changes in the concentration of zinc, manganese, and copper in the liver, spleen, and brain of rats after iron overload. Iron overload in Wistar rats was achieved by iron-supplemented diet or by intraperitoneal or intravenous injection of polymaltose iron. Iron, zinc, manganese, and copper were determined by atomic absorption spectrophotometry. Iron overload in rats, regardless of the route of its application, resulted in an increase not only of iron but also of zinc and manganese in the liver and the spleen, whereas the content of these metals in the brain did not change. The copper content of the liver, spleen, and brain remained the same after iron overload. The increase of zinc and manganese in the liver and spleen following iron overload was probably a result not only of increased intestinal absorption but also of increased uptake from the cell. This is also supported by the fact that no increase in the zinc and manganese concentrations occurred in the brain since, despite iron overload, the iron content remained constant.

Manganese removal using a biological trickling filter was investigated. Manganese removal was found to be caused by both biological and chemical manganese oxidation. The extent of each oxidation type was assessed. The performance of the trickling filter was tested under both continuous and sequencing batch reactor operation. The effectiveness and throughput for each operational mode were determined as a function of retention time and the advantages of each operational mode were investigated. It was found that the continuous operational mode leads to higher percentage of manganese removal but lower throughput rates when compared with a sequencing batch reactor operation with the same feed concentration and retention time. A series of experiments was also performed in order to investigate the interactions between ammonia, iron and manganese removal when simultaneously present in a biological filter. For low ammonia concentrations there is no serious inhibition of manganese removal. For higher ammonia concentrations inhibition of manganese removal becomes substantial. The presence of iron affects both ammonia and manganese removal negatively, while ammonia and manganese do not significantly affect iron removal.

The present work was conducted to study integrated chromium removal from aqueous solutions in horizontal subsurface (HSF) constructed wetlands. Two pilot-scale HSF constructed wetlands (CWs) units were built and operated. One unit was planted with common reeds (Phragmites australis) and one was kept unplanted. Influent concentrations of Cr(VI) ranged from 0.5 to 10mg/L. The effect of temperature and hydraulic residence time (8-0.5 days) on Cr(VI) removal were studied. Temperature was proved to affect Cr(VI) removal in both units. In the planted unit maximum Cr(VI) removal efficiencies of 100% were recorded at HRT's of 1 day with Cr(VI) concentrations of 5, 2.5 and 1mg/L, while a significantly lower removal rate was recorded in the unplanted unit. Harvested reed biomass from the CWs was co-composted with olive mill wastes. The final product had excellent physicochemical characteristics (C/N: 14.1-14.7, germination index (GI): 145-157%, Cr: 8-10mg/kg dry mass), fulfills EU requirements and can be used as a fertilizer in organic farming.

A dynamic model which describes nitritification and nitratification in trickling filters has been developed. The model predicts the concentration profiles of ammonia, nitrite and nitrate along the filter depth and along the biofilm depth, as a function of the operating parameters, in batch as well as in continuous operation. The model can also predict the biofilm thickness as a function of filter depth and time. The batch version of the model predicts that the best moment to start the continuous operation is the moment when Nitrobacter concentration is maximum and nitrite concentration is minimum, while the continuous version of the model predicts that nitrite accumulation would be observed only when a significant ammonia concentration is observed at the filter outlet. A pilot-scale trickling filter was used in order to verify the predicted results. In all cases there was very good agreement between predicted and experimental results.

A dynamic model for nitrifying trickling filters is developed, based on material balances in the biofilm and the bulk liquid. The model predicts the profile of ammonia as a function of the operating parameters (volumetric flow rate and feed ammonia concentration) and the biofilm thickness as a function of filter depth and time.

The bioreduction of hexavalent chromium from aqueous solution was carried out using suspended growth and packed-bed reactors under a draw-fill operating mode, and horizontal subsurface constructed wetlands. Reactors were inoculated with industrial sludge from the Hellenic Aerospace Industry using sugar as substrate. In the suspended growth reactors, the maximum Cr(VI) reduction rate (about 2 mg/L h) was achieved for an initial concentration of 12.85 mg/L, while in the attached growth reactors, a similar reduction rate was achieved even with high initial concentrations (109 mg/L), thus confirming the advantage of these systems. Two horizontal subsurface constructed wetlands (CWs) pilot-scale units were also built and operated. The units contained fine gravel. One unit was planted with common reeds and one was kept unplanted. The mean influent concentrations of Cr(VI) were 5.61 and 5.47 mg/L for the planted and unplanted units, respectively. The performance of the planted CW units was very effective as mean Cr(VI) removal efficiency was 85% and efficiency maximum reached 100%. On the contrary, the unplanted CW achieved very low Cr(VI) removal with a mean value of 26%. Both attached growth reactors and CWs proved efficient and viable means for Cr(VI) reduction.

A practical graphical method for the design of nitrifying trickling filters for potable water treatment is presented. The method allows the selection of the key design parameters (support media and filter dimensions) based on appropriate design curves. In the sequel, the possible advantages of water recirculation are examined theoretically and experimentally. Both theoretical and experimental results shown that an increase in ammonia bioconversion is expected for high or intermediate values of hydraulic loading and high values of inlet ammonia concentration, whereas at low hydraulic rates, recirculation has an adverse effect.

Ozonation of textile industrial wastewaters was examined in CSTR and batch studies. Experiments were done for various types of textile wastewaters and for different reaction and retention times. Much better results were obtained using ozonation after the coagulation-precipitation stage. Also a mathematical model was developed, able to describe wastewater decolonization and COD reduction. In all cases the comparison between the model predictions and the experimental data was satisfactory.

Rice hull hydrolysate was used as feedstock for microbial lipids production using the oleaginous fungus Mortierella isabellina. Kinetic experiments were conducted in C/N ratios 35, 44 and 57 and the oil accumulation into fungal biomass was 36%, 51.2% and 64.3%, respectively. A detailed mathematical model was used in order to describe the lipid accumulation process. This model was able to predict reducing sugar and nitrogen consumption, fat-free biomass synthesis and lipid accumulation. Neutral lipids constitute the predominant lipid fraction, while the major fatty acids were oleic, palmitic and linoleic acid. Fatty acids of long aliphatic chain were not detected, thus the microbial oil produced is a promising feedstock for biodiesel production.

Attached growth biological systems provide certain advantages over suspended growth ones. The high biomass concentration and biofilm structure provide resistance to high hydraulic and organic loadings, and toxicity effects. Attached growth processes can be applied to treat water and wastewater. This article presents applications of attached growth systems for ammonia, nitrate, iron, and manganese removal from potable water, as well as for chromium and olive mill wastewater (OMW) treatment. Biological water treatment can increase polluted water reservoir exploitation. Aerobic and autotrophic conditions are required for ammonia, iron, and manganese oxidizers, while anoxic and hydrogenotrophic conditions favor water denitrification. Attached growth systems achieve very high reduction rates for Cr(VI) and they could be effective even for high organic loadings, as in the case of OMW. The design of appropriate reactor configurations can enhance the rates of natural biological processes. Knowledge of microbial populations and the relationships between them would help provide the appropriate environmental conditions for maximum microbial growth and pollutant degradation. More than any other biological system, attached growth systems require an interdisciplinary approach. Attached growth systems achieve high biodegradation rates in compact configurations for a variety of pollutants. Their dominance in wastewater- and water-treatment systems in the next decades should be considered definite.

In this study we report the isolation of four denitrifying bacteria from a batch reactor, where the progress of hydrogenotrophic denitrification was examined. Only three of the strains had the ability to use hydrogen as electron donor. In the present work, kinetic batch experiments were carried out in order to study the dynamic characteristics of pure and defined mixed cultures of hydrogen-oxidizing denitrifying bacteria, under anoxic conditions, in a defined synthetic medium, in the presence of nitrates. Kinetic models were developed and the kinetic parameters were determined from the batch experiments for each bacterium separately. The behavior of mixed cultures and the interactions between the bacteria were described using kinetic models based on the kinetic models developed for each bacterium separately and their predictions were compared with the results from mixed culture experiments. The mathematical models that were developed and validated in the present work are capable of describing the behavior of the bacteria in pure and mixed cultures, and in particular, the kinetics of nitrate and nitrite reduction and cell growth.

A pilot-scale trickling filter was constructed and tested for biological chromium(VI) removal from industrial wastewater. Indigenous bacteria from industrial sludge were enriched and used as inoculum for the filter. Sodium acetate was used as carbon source and it was found to inhibit chromate reduction at high concentrations. Three different operating modes were used to investigate the optimal performance and efficiency of the filter, i.e. batch, continuous and SBR with recirculation. The latter one was found to achieve removal rates up to 530 g Cr(VI)/m2 d, while aeration was taking place naturally without the use of any external mechanical means. The low operating cost combined with the high hexavalent chromium reduction rates indicates that this technology may offer a feasible solution to a very serious environmental problem.

We study a model of three microbial populations competing for three complementary nutrients in a single chemostat. By using methods of numerical bifurcation theory we analyze the model equations and determine the effect of the model parameters on the dynamics of the system. The main question to be answered is whether there exist conditions under which the three populations can coexist in a stable state in the chemostat. The analysis shows that coexistence can be obtained as a stable steady state but also as a stable periodic state for a wide range of operating conditions of the chemostat.

We examine the question of coexistence of three microbial populations competing for two complementary rate-limiting substrates in configurations of interconnected chemostats. It is known that coexistence of two populations competing for two rate-limiting substrates is possible in a single chemostat, but coexistence is not possible when three populations are involved. We examine whether coexistence of three populations becomes possible by considering configurations of two or three interconnected chemostats, thus allowing for effects of spatial heterogeneity. Computational analysis of the model equations indicates that in the case of two chemostats coexistence is possible only for specific discrete parameter values where the system is structurally unstable and therefore the coexistence state is not practically attainable, whereas in the case of three chemostats coexistence is possible for a whole range of parameter values where the system is structurally stable and therefore the coexistence state can be realized in practice.

It is known that, when two microbial populations competing for a single rate-limiting nutrient are grown in a spatially uniform environment, such as a single chemostat, with competition being the only interaction between them, they cannot coexist, but eventually one of the two populations prevails and the other becomes extinct. Spatial heterogeneity has been suggested as a means of obtaining coexistence of the two populations. A configuration of two interconnected chemostats is a simple model of a spatially heterogeneous environment. It has been shown that, when Monod's model is used for the specific growth rates of the two populations, steady-state coexistence can be obtained in such systems for wide ranges of operating conditions. In the present work, we study a model of microbial competition in configurations of interconnected chemostats and we show that, if a substrate inhibition model is used for the specific growth rates of the two populations, coexistence in a periodic state is also possible. The analysis of the model is done by numerical bifurcation theory methods.