Removal of micropollutants and ecotoxicity during combined biological activated carbon and ozone (BO3) treatment

Ozonation is a viable option to improve the removal of micropollutants (MPs) in wastewater treatment plants (WWTPs). Nevertheless, the application of ozonation is hindered by its high energy requirements and by the uncertainties regarding the formation of toxic transformation products in the process. Energy requirements of ozonation can be reduced with a pre-ozone treatment, such as a biological activated carbon (BAC) filter, that removes part of the effluent organic matter before ozonation. This study investigated a combination of BAC filtration followed by ozonation (the BO3 process) to remove MPs at low ozone doses and low energy input, and focused on the formation of toxic organic and inorganic products during ozonation. Effluent from a WWTP was collected, spiked with MPs (approximately 1 µg/L) and treated with the BO3 process. Different flowrates (0.25-4 L/h) and specific ozone doses (0.2-0.6 g O3/g TOC) were tested and MPs, ecotoxicity and bromate were analyzed. For ecotoxicity assessment, three in vivo (daphnia, algae and bacteria) and six in vitro CALUX assays (Era, GR, PAH, P53, PR, andNrf2 CALUX) were used. Results show that the combination of BAC filtration and ozonation has higher MP removal and higher ecotoxicity removal than only BAC filtration and only ozonation. The in vivo assays show a low ecotoxicity in the initial WWTP effluent samples and no clear trend with increasing ozone doses, while most of the in vitro assays show a decrease in ecotoxicity with increasing ozone dose. This suggests that for the tested bioassays, feed water and ozone doses, the overall ecotoxicity of the formed transformation products during ozonation was lower than the overall ecotoxicity of the parent compounds. In the experiments with bromide spiking, relevant formation of bromate was observed above specific ozone doses of approximately 0.4 O3/g TOC and more bromate was formed for the samples with BAC pre-treatment. This indirectly indicates the effectivity of the pre-treatment in removing organic matter and making ozone more available to react with other compounds (such as MPs, but also bromide), but also underlines the importance of controlling the ozone dose to be below the threshold to avoid formation of bromate. It was concluded that treatment of the tested WWTP effluent in the BO3 process at a specific ozone dose of 0.2 g O3/g TOC, results in high MP removal at limited energy input while no increase in ecotoxicity, nor formation of bromate was observed under this condition. This indicates that the hybrid BO3 process can be implemented to remove MPs and improve the ecological quality of this WWTP effluent with a lower energy demand than conventional MP removal processes such as standalone ozonation.

Microbial dynamics and bioreactor performance are interlinked with organic matter removal from wastewater treatment plant effluent

Optimizing bioreactor performance for organic matter removal can achieve sustainable and energy-efficient micropollutant removal in subsequent tertiary treatment. Bioreactor performance heavily depends on its resident microbial community; hence, a deeper understanding of community dynamics is essential. The microbial communities of three different bioreactors (biological activated carbon, moving bed biofilm reactor, sand filter), used for organic matter removal from wastewater treatment effluent, were characterized by 16S rRNA gene amplicon sequence analysis. An interdependency between bioreactor performance and microbial community profile was observed. Overall, Proteobacteria was the most predominant phylum, and Comamonadaceae was the most predominant family in all bioreactors. The relative abundance of the genus Roseococcus was positively correlated with organic matter removal. A generalized Lotka-Volterra (gLV) model was established to understand the interactions in the microbial community. By identifying microbial dynamics and their role in bioreactors, a strategy can be developed to improve bioreactor performance.

The effect of organic matter fractions on micropollutant ozonation in wastewater effluents

Organic matter (OM) is the most important factor influencing the effectivity and efficiency of micropollutant (MP) ozonation in wastewater effluents. The importance of the quantity of OM is known, because of this, total organic carbon (TOC) is generally used to determine the required ozone dose for any water sample. Still, the effect of OM type on MP ozonation is not well understood. In this study, effluents from five wastewater treatment plants were collected and the organic matter in these effluents was fractionated using membranes (F1-4) and resin (HI, HOA, HON and HOB). Fractions were diluted to the same TOC concentration, spiked with MPs and ozonated at three ozone doses. Our results show that all five effluents had comparable OM compositions and similar MP removal, confirming the suitability of OM quantity (TOC) to compare the ozone requirements for wastewater effluents. From the 19 analysed MPs, three groups were identified that showed similar removal behaviour. The strongest differences between the groups were observed around MP ozone reactivities of 10², 10⁴ and 10⁶ M^-1 s^-1. This indicates the presence of three OM groups in the samples that interfere with the removal of different MPs. MP removal in the resin fraction HON were higher for MPs with high and medium ozone reactivity, indicating a low interference of OM in this fraction with MP ozonation. OM in the resin fractions HOA and HI showed higher interference with MP ozonation. Therefore, removing the HOA and HI fractions prior to ozonation would result in a lower required ozone dose and a more efficient removal of the MPs. MP removal correlated with the OM characteristics A300, SR and fluorescence component comp 2. These characteristics can be used as inline tools to predict the required ozone dose in water treatment plants.

Performance of full scale constructed wetlands in removing antibiotics and antibiotic resistance genes

Additional treatment of wastewater, such as constructed wetlands (CWs), is a possible solution to reduce the discharge of antibiotics and antibiotic resistance genes (ARGs) from households and industry to the environment. This study aims to investigate the occurrence and removal of antibiotics and ARGs by two full scale CWs operated at different hydraulic retention times (HRT), namely 1 day and 3 days. Both CWs were receiving the same wastewater treatment plant (WWTP) effluent. Temporally and spatially distributed sampling of water and sediment was conducted for one year and samples were analyzed for antibiotics and ARGs by using LC-MS/MS and qPCR. Results showed that both CWs removed antibiotics significantly with a comparable overall removal of 28%-100%, depending on the type of antibiotics. However, some of the antibiotics showed higher concentration after the CW treatment. Five antibiotics (tiamulin, tylosin, oxytetracycline, sulfamethoxazole and trimethoprim) were the most abundant (>1500 ng/l on average) in winter. Meanwhile, ermB was the most abundant (average of 5.0 log) in winter compared to summer (average of 3.5 log). Other ARGs did not show a significant increase or decrease between winter and summer. ARGs were removed from the wastewater by 0.8 to 1.5 log. The HRT did not influence the removal of either the antibiotics or the ARGs. A strong correlation was found between sul genes and intI1. The results also revealed a positive and a negative relationship from sampling point 1 to sampling point 5: a positive relation between abundance of antibiotics, ARGs, and of NO₃-N, NH₄-N, TP, COD and a negative relation between antibiotics, ARGs and temperature. This relationship showed the effect between antibiotics and ARGs concentrations with physicochemical parameters and nutrients. The ability of CWs to reduce the input of micropollutants into the environment makes CWs a potential post treatment to WWTP.

Improving removal of antibiotics in constructed wetland treatment systems based on key design and operational parameters: A review

While removal of antibiotics in constructed wetland treatment systems (CWTS) has been described previously, few studies examined the synergistic effect of multiple design and operational parameters for improving antibiotic removal. This review describes the removal of 35 widely used antibiotics in CWTS covering the most common design parameters (flow configuration, substrate, plants) and operational parameters (hydraulic retention time/hydraulic loading rates, feeding mode, aeration, influent quality), and discusses how to tailor those parameters for improving antibiotic removal based on complex removal mechanisms. To achieve an overall efficient removal of antibiotics in CWTS, our principal component analysis indicated that optimization of flow configuration, selection of plant species, and compensation for low microbial activity at low temperature is the priority strategy. For instance, a hybrid-CWTS that integrates the advantages of horizontal and vertical subsurface flow CWTS may provide a sufficient removal performance at reasonable cost and footprint. To target removal of specific antibiotics, future research should focus on elucidating key mechanisms for their removal to guide optimization of the design and operational parameters. More efficient experimental designs (e.g., the Box-Behnken design) are recommended to determine the settings of the key parameters. These improvements would promote development of this environmentally friendly and cost-efficient technology for antibiotic removal.

Effects of salinity on the treatment of synthetic petroleum-industry wastewater in pilot vertical flow constructed wetlands under simulated hot arid climatic conditions

Petroleum-industry wastewater (PI-WW) is a potential source of water that can be reused in areas suffering from water stress. This water contains various fractions that need to be removed before reuse, such as light hydrocarbons, heavy metals and conditioning chemicals. Constructed wetlands (CWs) can remove these fractions, but the range of PI-WW salinities that can be treated in CWs and the influence of an increasing salinity on the CW removal efficiency for abovementioned fractions is unknown. Therefore, the impact of an increasing salinity on the removal of conditioning chemicals benzotriazole, aromatic hydrocarbon benzoic acid, and heavy metal zinc in lab-scale unplanted and Phragmites australis and Typha latifolia planted vertical-flow CWs was tested in the present study. P. australis was less sensitive than T. latifolia to increasing salinities and survived with a NaCl concentration of 12 g/L. The decay of T. latifolia was accompanied by a decrease in the removal efficiency for benzotriazole and benzoic acid, indicating that living vegetation enhanced the removal of these chemicals. Increased salinities resulted in the leaching of zinc from the planted CWs, probably as a result of active plant defence mechanisms against salt shocks that solubilized zinc. Plant growth also resulted in substantial evapotranspiration, leading to an increased salinity of the CW treated effluent. A too high salinity limits the reuse of the CW treated water. Therefore, CW treatment should be followed by desalination technologies to obtain salinities suitable for reuse. In this technology train, CWs enhance the efficiency of physicochemical desalination technologies by removing organics that induce membrane fouling. Hence, P. australis planted CWs are a suitable option for the treatment of water with a salinity below 12 g/L before further treatment or direct reuse in water scarce areas worldwide, where CWs may also boost the local biodiversity. Graphical abstract.

Fate of antibiotics and antibiotic resistance genes during conventional and additional treatment technologies in wastewater treatment plants

Information on the removal of antibiotics and ARGs in full-scale WWTPs (with or without additional treatment technology) is limited. However, it is important to understand the efficiency of full-scale treatment technologies in removing antibiotics and ARGs under a variety of conditions relevant for practice to reduce their environmental spreading. Therefore, this study was performed to evaluate the removal of antibiotics and ARGs in a conventional wastewater treatment plant (WWTP A) and two full-scale combined with additional treatment technologies. WWTP B, a conventional activated sludge treatment followed by an activated carbon filtration step (1-STEP® filter) as a final treatment step. WWTP C, a treatment plant using aerobic granular sludge (NEREDA®) as an alternative to activated sludge treatment. Water and sludge were collected and analysed for 52 antibiotics from four target antibiotic groups (macrolides, sulfonamides, quinolones, tetracyclines) and four target ARGs (ermB, sul 1, sul 2 and tetW) and integrase gene class 1 (intI1). Despite the high removal percentages (79-88%) of the total load of antibiotics in all WWTPs, some antibiotics were detected in the various effluents. Additional treatment technology (WWTP C) showed antibiotics removal up to 99% (tetracyclines). For ARGs, WWTP C reduced 2.3 log followed by WWTP A with 2.0 log, and WWTP B with 1.3 log. This shows that full-scale WWTP with an additional treatment technology are promising solutions for reducing emissions of antibiotics and ARGs from wastewater treatment plants. However, total removal of the antibiotics and ARGS cannot be achieved for all types of antibiotics and ARGs. In addition, the ARGs were more abundant in the sludge compared to the wastewater effluent suggesting that sludge is an important reservoir representing a source for later ARG emissions upon reuse, i.e. as fertilizer in agriculture or as resource for bioplastics or bioflocculants. These aspects require further research.

Pilot-scale hybrid constructed wetlands for the treatment of cooling tower water prior to its desalination and reuse

Cooling towers are responsible for a large part of the industrial fresh water withdrawal, and the reuse of cooling tower water (CTW) effluents can strongly lower industrial fresh water footprints. CTW requires desalination prior to being reused, but various CTW components, such as total organic carbon (TOC), conditioning chemicals and total suspended solids (TSS) hamper physico-chemical desalination technologies and need to be removed from the CTW. A cost-efficient and robust pre-treatment is thus required, which can be provided by constructed wetlands (CWs). The present study is the first study that determined the CTW pre-treatment efficiency of hybrid-CWs and the impact of winter season and biocides in the CTW on the pre-treatment efficiency. The most efficient CW flow type and dominant removal mechanisms for CW components hampering physico-chemical desalination were determined. Subsurface flow CWs removed PO₄^3-, TSS and TOC as a result of adsorption and filtration. Vertical subsurface flow CWs (VSSF-CW) excelled in the removal of benzotriazole as a result of aerobic biodegradation. Horizontal subsurface flow CWs (HSSF-CW) allowed the denitrification of NO₃^- due to their anaerobic conditions. Open water CWs (OW-CWs) did not contribute to the removal of components that hamper physico-chemical desalination technologies, but do provide water storage options and habitat. The biological removal processes in the different CW flow types were negatively impacted by the winter season, but were not impacted by concentrations of the biocides glutaraldehyde and DBNPA that are relevant in practice. For optimal pre-treatment, a hybrid-CW, consisting of an initial VSSF-CW followed by an OW-CW and HSSF-CW is recommended. Future research should focus on integrating the hybrid-CW with a desalination technology, e.g. reverse osmosis, electrodialysis or capacitive ionization, to produce water that meets the requirements for use as cooling water and allow the reuse of CTW in the cooling tower itself.

Anaerobic biodegradation of pharmaceutical compounds coupled to dissimilatory manganese (IV) or iron (III) reduction

Pharmaceuticals in water have adverse effects on aquatic environment. Anaerobic pharmaceutical biodegradation coupled to dissimilatory manganese(Mn) (IV)- or iron(Fe) (III)-oxides reduction is potentially efficient but unexplored. In this study, batch experiments were performed using different Mn(IV) and Fe(III) species with a microbial inoculum pre-cultivated with 15 mM chemically-synthesized Mn(IV) and 10 mg L^-1 metoprolol. Results show 26% caffeine and 52% naproxen are degraded with Mn(IV) as terminal electron acceptor and insignificant biodegradation for other pharmaceuticals tested. Reduction of Mn(IV) from drinking water treatment is coupled to anaerobic biodegradation of metoprolol and propranolol, resulting in removal efficiencies of 96% and 31%, respectively. The results indicate that adsorption contributes to the pharmaceutical removal during the first 10 days of incubation, while biodegradation is the main removal mechanism in the whole period. Fe(III) can also be used as electron acceptor in anaerobic pharmaceutical biodegradation. Over half of the added metoprolol is degraded with both chemically-synthesized Fe(III) and Fe(III)-citrate as terminal electron acceptors. However, this process did not occur when using Fe(III) from drinking water treatment or Fe(III)-based sorbents. This study indicates that anaerobic pharmaceutical biodegradation coupled to dissimilatory Mn(IV) or Fe(III) reduction is possible, and promising for application to cleaning wastewater treatment plant effluents.

Non-target screening reveals the mechanisms responsible for the antagonistic inhibiting effect of the biocides DBNPA and glutaraldehyde on benzoic acid biodegradation

The desalination and reuse of discharged cooling tower water (CTW) as feed water for the cooling tower could lower the industrial fresh water withdrawal. A potential pre-treatment method before CTW desalination is the use of constructed wetlands (CWs). Biodegradation is an important removal mechanism in CWs. In the present study, the impact of the biocides 2,2-dibromo-2-cyanoacetamide (DBNPA) and glutaraldehyde on the biodegradation process by CW microorganisms was quantified in batch experiments in which benzoic acid was incubated with realistic CTW biocide concentrations. DBNPA had a stronger negative impact on the biodegradation than glutaraldehyde. The combination of DBNPA and glutaraldehyde had a lower impact on the biodegradation than DBNPA alone. UHPLC-qTOF-MS/MS non-target screening combined with data-analysis script 'patRoon' revealed two mechanisms behind this low impact. Firstly, the presence of glutaraldehyde resulted in increased DBNPA transformation to the less toxic transformation product 2-bromo-2-cyanoacetamide (MBNPA) and newly discovered 2,2-dibromopropanediamide. Secondly, the interaction between glutaraldehyde and DBNPA resulted in the formation of new products that were less toxic than DBNPA. The environmental fate and toxicity of these products are still unknown. Nevertheless, their formation can have important implications for the simultaneous use of the biocides DBNPA and glutaraldehyde for a wide array of applications.

Benzotriazole removal mechanisms in pilot-scale constructed wetlands treating cooling tower water

The reuse of discharged cooling tower water (CTW) in the cooling tower itself could reduce fresh water intake and help mitigating fresh water scarcity problems. However, this requires desalination prior to its reuse, and hindering fractions, such as conditioning chemicals, should be removed before desalination to obtain a higher desalination efficiency. Constructed wetlands (CWs) can provide such a pre-treatment. In this study, the mechanisms underlying the removal of conditioning chemical benzotriazole (BTA) in CWs was studied using an innovative approach of differently designed pilot-scale CWs combined with batch removal experiments with substrate from these CWs. By performing these combined experiments, it was possible to determine the optimal CW design for BTA removal and the most relevant BTA removal processes in CWs. Adsorption yielded the highest contribution, and the difference in removal between different CW types was linked to their capability to aerobically biodegrade BTA. This knowledge on the main removal mechanisms for BTA allows for a CW design tailored for BTA removal. In addition, the outcomes of this research show that performing batch experiments with CW substrate allows one to determine the relevant removal mechanisms for a given compound which results in a better understanding of CW removal processes.

Retention soil filters for the treatment of sewage treatment plant effluent and combined sewer overflow

Retention soil filters (RSFs) are vertical flow constructed wetlands. They are mainly used for the treatment of combined sewer overflow or stormwater, and not operated during dry weather conditions. However, RSFs have been successfully tested as continuous post treatment for sewage treatment plant effluents. In this paper we present a new approach, namely dual usage of the retention soil filter. During dry weather the RSF is used for the polishing treatment of sewage treatment plant effluent and during overflow events, the retention soil filter treats the combined sewer overflow. This study was conducted at two pilot RSFs that were fed with sewage treatment effluent for four years. Removal of TOC, DOC, nutrients and 21 organic micropollutants was determined during six months at different sequences of regular effluent and overflow treatment conditions. TOC, DOC and nutrients, appearing in high concentration in combined sewer overflow, were effectively removed, and metformin and caffeine micropollutants showed >99% removal. Residues from this combined sewer treatment that were sorbed on filter material or stored in pore water were washed out directly after treatment when STP effluent infiltration was initiated. This effect declined within 20 h after combined sewer overflow treatment. Dry periods of 18 h between combined sewer and sewage treatment plant effluent feeding counteracted the wash out effects. The highest removal efficiency was found in the beginning of the feeding time of 28 h, indicating that shorter feeding cycles enhance the overall efficiency of the RSF. Finally, the results show that a single RSF system can successfully reduce emissions of TOC, DOC, nutrients and micropollutants to surface waters from two different emission pathways, i.e. from regular treated effluents and storm related sewer overflows. In conclusion, the dual usage of RSF is a promising approach and ready for upscaling and implementation.

Diffusion of hydrophilic organic micropollutants in granular activated carbon with different pore sizes

Hydrophilic organic micropollutants are commonly detected in source water used for drinking water production. Effective technologies to remove these micropollutants from water include adsorption onto granular activated carbon in fixed-bed filters. The rate-determining step in adsorption using activated carbon is usually the adsorbate diffusion inside the porous adsorbent. The presence of mesopores can facilitate diffusion, resulting in higher adsorption rates. We used two different types of granular activated carbon, with and without mesopores, to study the adsorption rate of hydrophilic micropollutants. Furthermore, equilibrium studies were performed to determine the affinity of the selected micropollutants for the activated carbons. A pore diffusion model was applied to the kinetic data to obtain pore diffusion coefficients. We observed that the adsorption rate is influenced by the molecular size of the micropollutant as well as the granular activated carbon pore size.

In situ removal of four organic micropollutants in a small river determined by monitoring and modelling

Organic micropollutants (OMPs) are widely detected in surface waters. So far, the removal processes of these compounds in situ in river systems are not yet totally revealed. In this study, a combined monitoring and modelling approach was applied to determine the behaviour of 1-H benzotriazole, carbamazepine, diclofenac and galaxolide in a small river system. Sewage treatment plant effluents and the receiving waters of the river Swist were monitored in 9 dry weather sampling campaigns (precipitation < 1 mm on the sampling day itself and <5 mm total precipitation two days before the sampling) during different seasons over a period of 3 years. With the results gained through monitoring, mass balances have been calculated to assess fate in the river. With the DWA Water Quality Model, OMP concentrations in the river were successfully simulated with OMP characteristics gained through literature studies. No removal was determined for 1-H benzotriazole and carbamazepine, whereas diclofenac showed removal that coincided with light intensity. Moreover, modelling based on light sensitivity of diclofenac also suggested relevant degradation at natural light conditions. These two approaches suggest removal by photodegradation. The highest removal in the river was detected for galaxolide, presumably due to volatilisation, sorption and biodegradation. Furthermore, short-term concentration variability in the river was determined, showing that daily concentration patterns are influenced by dynamics of sewage treatment plant effluent volumes and removal processes in the river.

Biological regeneration of manganese (IV) and iron (III) for anaerobic metal oxide-mediated removal of pharmaceuticals from water

Applying manganese(IV)- or iron(III)-(hydr)oxides to remove pharmaceuticals from water could be attractive, due to the capacity of these metal oxides to remove pharmaceuticals and be regenerated. As pharmaceutical removal under anaerobic conditions is foreseen, Mn(IV) or Fe(III) regeneration under anaerobic conditions, or with minimum oxygen dosage, is preferred. In this study, batch experiments are performed to investigate (1) Mn(IV) and Fe(III) regeneration from Mn(II) and Fe(II); (2) the pharmaceutical removal during biological Mn(IV) and Fe(III) regeneration; and (3) anaerobic abiotic pharmaceutical removal with different Mn(IV) or Fe(III) species. Results show that biological re-oxidation of reduced Mn(II) to Mn(IV) occurs under oxygen-limiting conditions. Biological re-oxidation of Fe(II) to Fe(III) is obtained with nitrate under anaerobic conditions. Both bio-regenerated Mn(IV)-oxides and Fe(III)-hydroxides are amorphous. The pharmaceutical removal is insignificant by Mn(II)- or Fe(II)-oxidizing bacteria during regeneration. Finally, pharmaceutical removal is investigated with various Mn(IV) and Fe(III) sources. Anaerobic abiotic removal using Mn(IV) produced from drinking water treatment plants results in 23% metoprolol and 44% propranolol removal, similar to chemically synthesized Mn(IV). In contrast, Fe(III) from drinking water treatment plants outperformed chemically or biologically synthesized Fe(III); Fe (III) from drinking water treatment can remove 31-43% of propranolol via anaerobic abiotic process. In addition, one of the Fe(III)-based sorbents tested, FerroSorp^®RW, can also remove propranolol (20-25%). Biological regeneration of Mn(IV) and Fe(III) from the reduced species Mn(II) and Fe(II) could be more effective in terms of cost and treatment efficiency.

Retention soil filter as post-treatment step to remove micropollutants from sewage treatment plant effluent

Retention soil filters (RSFs) are a specific form of vertical flow constructed wetlands for the treatment of rain water and/or wastewater. We have tested 3 pilot RSFs to investigate removal of dissolved organic carbon (DOC) and 14 different organic micropollutants (OMPs) from the effluent of a large scale sewage treatment plant (STP). Two of them were operated as conventional RSF with material (sand with CaCO₃ and organic matter) from two different full-scale RSFs. The third pilot RSF contained filter material (sand with CaCO₃) with additional biochar in the upper layer (0-10 cm) and granulated activated carbon (GAC) in the lower layer (60-90 cm). The filters were planted with Phragmites australis. The RSFs were operated and monitored for 3 years, and water samples were taken regularly at inflow, outflows and in 3 depths within the filters. In total 523 samples were taken. In the conventional RSF, best median removal was detected for galaxolide, diclofenac 4-hydroxy, metoprolol and clarithromycin (75-79%). No removal was seen for sulfamethoxazole and carbamazepine. The DOC and OMP removal in the conventional RSFs was best in the upper layer with highest organic matter content, increased in time over the three years of operation and also with extended contact time. In the effluent of the RSF with GAC, 10 out of the 14 OMPs could not be detected; 4 OMPs were detected, but only metformin with removal < 80%, thus showing a more efficient removal than the conventional RSF. A decrease in DOC removal was detected in the GAC layer (>88% to 60%) over the 2.5 years of operation. Biochar was most effective in OMP removal in the first operational year. It can be concluded that the increasing removal efficiency of the conventional RSF material - also present in the RSF with biochar and GAC - might mitigate the reduced efficiency of the sorbent additives biochar and GAC. This enables to extend the operational lifetime of the filters with acceptable removal rates. Finally, our study demonstrates that an RSF with GAC shows an enhanced removal of OMPs, which is a suitable post-treatment step for STPs.

Fate and distribution of pharmaceutically active compounds in mesocosm constructed wetlands

Removal of pharmaceutically active compounds (PhACs) in constructed wetlands (CWs) is a complex interplay of different processes. We studied fate and distribution of seven PhACs (caffeine, CAF; naproxen, NAP; metoprolol, MET; propranolol, PRO; ibuprofen, IBP; carbamazepine, CBZ; diclofenac, DFC) in mesocosm CWs and effects of irradiation via pre-photocatalysis, substrate composition (mainly sediment) through addition of litter (dead plant biomass), and plants. CWs showed high removal of CAF, NAP, MET, PRO, and IBP (79-99%). All seven PhACs were detected in substrate and plant tissues as well as IBP intermediates. Estimated PhAC mass balance showed that sorption dominated PRO removal in CWs while other PhACs were mainly removed by biodegradation and/or phytodegradation. Pre-photocatalysis significantly increased removal of PhACs except for CAF and IBP, and decreased accumulation of PhACs in substrate and plant tissues of the following wetland compartment. Litter addition in CW significantly enhanced removal of PRO and CBZ via biodegradation and/or phytodegradation. Plants played an essential and positive role in removing PhACs, resulting from direct phytoremediation and indirectly enhancing sorption and biodegradation. Our study provides knowledge to understand removal mechanisms of PhACs in CWs and to potentially enhance PhAC removal by developing pre-photocatalysis, adding dead plant biomass, and optimizing vegetation.

Evaluation of attenuation of pharmaceuticals, toxic potency, and antibiotic resistance genes in constructed wetlands treating wastewater effluents

The performance of constructed wetlands (CWs) in the removal of pharmaceutically active compounds (PhACs) is generally evaluated on the basis of chemical analysis. In this work, we used a combination of chemical, toxicological, and molecular analyses to assess the attenuation of PhACs, toxic potency and antibiotic resistance genes (ARGs) in a field study of three CWs serving as tertiary treatment of wastewater treatment plants. First, 17 PhACs were analysed chemically, of which 14 were detected and seven at concentrations >0.1μg/l. Even though some of the individual PhACs were moderately or highly removed in the CWs investigated, median removal of overall PhACs was approximately 50% in the vertical subsurface flow CW (VSF-CW) with a lower hydraulic loading rate while the removal in the other two free water surface flow CWs (SF-CWs) was negligible. Second, toxic potency of wastewater extracts was assessed in a range of bioassays. Estrogenicity was overall attenuated in CWs, while the neurotoxic potency of wastewater extracts did not decrease after passage through the two CWs investigated. Third, the VSF-CW and one of the SF-CW showed a positive removal of an integrase gene and three ARGs tested. The increased concentrations of ARGs in the other SF-CW, as well as the increase of total bacteria in all CWs, may relate to regrowth of resistance-carrying bacteria. Finally, multivariate analysis shows that most PhACs are positively correlated to the observed toxic potency. Additionally, low removal of organics and nutrients seems to parallel with low removal of PhACs. ARGs positively correlated with organics, nutrients and some PhACs, and the integrase gene but not to the respective antibiotics. The insufficient removal of PhACs, toxic potency, and ARGs indicates the need of an optimal design of CWs as tertiary treatment facilities.

Effects of dissolved organic matter and nitrification on biodegradation of pharmaceuticals in aerobic enrichment cultures

Natural dissolved organic matter (DOM) and nitrification can play an important role in biodegradation of pharmaceutically active compounds (PhACs) in aerobic zones of constructed wetlands (CWs). This study used an enrichment culture originating from CW sediment to study the effect of DOM and nitrification on aerobic biodegradation of seven PhACs. The enriched culture degraded caffeine (CAF), metoprolol (MET), naproxen (NAP), and ibuprofen (IBP) with a consistent biodegradability order of CAF>MET>NAP>IBP. Biodegradation of propranolol, carbamazepine, and diclofenac was insignificant (<15%). CAF biodegradation was inhibited by the easily biodegradable DOM. Conversely, DOM enhanced biodegradation of MET, NAP, and IBP, potentially by contributing more biomass capable of degrading PhACs. Nitrification enhanced biodegradation of NAP and IBP and mineralization of the PhAC mixture as well as less biodegradable DOM, which may result from co-metabolism of ammonia oxidizing bacteria or enhanced heterotrophic microbial activity under nitrification. MET biodegradation was inhibited in the presence of nitrification. DOM and nitrification effects on PhAC biodegradation in CWs gained from this study can be used in strategies to improve CW operation, namely: designing hydraulic retention times based on the biodegradability order of specific PhACs; applying DOM amendment; and introducing consistent ammonium streams to increase removal of PhACs of interest.

Anoxic conditions are beneficial for abiotic diclofenac removal from water with manganese oxide (MnO2)

This is the first study examining pharmaceutical removal under anoxic conditions with MnO₂. This study compares the abiotic removal of seven pharmaceuticals with reactive MnO₂ particles in the presence of oxygen (oxic conditions) and in the absence of oxygen (anoxic conditions). Due to the novelty of pharmaceutical removal under anoxic conditions, the influence of phosphate buffer, pH, and MnO₂ morphologies is also examined. Results show that over 90% of diclofenac is removed under anoxic conditions. Additionally, we found that (1) anoxic conditions are beneficial for diclofenac removal with MnO₂, (2) phosphate buffer affects the pharmaceutical removal efficiencies, (3) higher pharmaceutical removal is obtained at acidic pH compared to that at neutral or alkaline conditions, and (4) amorphous MnO₂ removes pharmaceuticals better than crystalline MnO₂. The pharmaceutical molecular structure and properties, MnO₂ properties especially reactive sites of the MnO₂ surface, are important for degradation kinetics. This study provides a fundamental basis towards understanding pharmaceutical degradation with MnO₂ under anoxic conditions, and development of a cost-effective, sustainable technology for removal of pharmaceuticals from water.

Pharmaceutical biodegradation under three anaerobic redox conditions evaluated by chemical and toxicological analyses

Biodegradation of pharmaceutically active compounds (PhACs) in the subsurface layer of constructed wetlands (CWs) under various anaerobic redox conditions is rarely studied. In this study, CW sediment microbial populations were enriched for PhAC biodegrading organisms. Biodegradation effectivity of a mixture of six PhACs (caffeine, CAF; naproxen, NAP; metoprolol, MET; propranolol, PRO; ibuprofen, IBP; carbamazepine, CBZ) and single compounds (CAF, NAP) was investigated under nitrate reducing, sulfate reducing, and methanogenic conditions using chemical and toxicological analyses. Biodegradation efficiencies varied strongly among the six PhACs and three redox conditions chosen. CAF and NAP were completely biodegraded under sulfate reducing and methanogenic conditions whereas biodegradation efficiencies of the other PhACs were much less (MET, PRO <20%; IBP, CBZ, negligible). CAF and NAP showed significantly lower biodegradation under nitrate reducing conditions than under the other two redox conditions. No difference was found in biodegradation efficiencies of CAF and NAP when present as single compound, or as a mixture with other PhACs. Different intermediates were observed, indicating different biodegradation pathways under different redox conditions and when the PhACs were present as single compound or in a mixture. From toxicological perspective, toxicity of PhACs and/or their intermediates to Vibrio fischeri was attenuated during the biodegradation process. Chemical and toxicological data showed positive correlations in principle component analysis, by which potentially toxic PhACs and intermediates are indicated for further ecotoxicological hazard assessment.

Anaerobic degradation of a mixture of MtBE, EtBE, TBA, and benzene under different redox conditions

The increasing use of biobased fuels and fuel additives can potentially change the typical fuel-related contamination in soil and groundwater. Anaerobic biotransformation of the biofuel additive ethyl tert-butyl ether (EtBE), as well as of methyl tert-butyl ether (MtBE), benzene, and tert-butyl alcohol (TBA, a possible oxygenate metabolite), was studied at an industrially contaminated site and in the laboratory. Analysis of groundwater samples indicated that in the field MtBE was degraded, yielding TBA as major product. In batch microcosms, MtBE was degraded under different conditions: unamended control, with medium without added electron acceptors, or with ferrihydrite or sulfate (with or without medium) as electron acceptor, respectively. Degradation of EtBE was not observed under any of these conditions tested. TBA was partially depleted in parallel with MtBE. Results of microcosm experiments with MtBE substrate analogues, i.e., syringate, vanillate, or ferulate, were in line with the hypothesis that the observed TBA degradation is a cometabolic process. Microcosms with ferulate, syringate, isopropanol, or diethyl ether showed EtBE depletion up to 86.5% of the initial concentration after 83 days. Benzene was degraded in the unamended controls, with medium without added electron acceptors and with ferrihydrite, sulfate, or chlorate as electron acceptor, respectively. In the presence of nitrate, benzene was only degraded after addition of an anaerobic benzene-degrading community. Nitrate and chlorate hindered MtBE, EtBE, and TBA degradation.

Marine snow increases the adverse effects of oil on benthic invertebrates

After the Deepwater Horizon oil spill, a MOSSFA (Marine Oil Snow Sedimentation and Flocculent Accumulation) event took place, transporting an estimated 14% of total released oil to the sediment, and smothering parts of the benthic ecosystem. This microcosm study describes the effects of oiled artificial marine snow on benthic macroinvertebrates. Corophium volutator survival was reduced by 80% in oil-contaminated snow. Hydrobia ulvae survival was reduced by 40% in oil-contaminated snow, possibly due to consumption of oiled snow. Macoma balthica was sensitive to marine snow, addition of oil slightly decreased survival. This study reveals trait-dependent sensitivity to oil with or without marine snow. The main drivers for organismal response to marine snow and oil are motility, sensitivity to hypoxia and oil toxicity, and feeding habits. Adverse effects of MOSSFA events on benthos will have consequence for the benthic-pelagic habitat and food chain, and should receive more attention in oil spill management.

Oil biodegradation: Interactions of artificial marine snow, clay particles, oil and Corexit

During the Deepwater Horizon (DwH) oil spill, interactions between oil, clay particles and marine snow lead to the formation of aggregates. Interactions between these components play an important, but yet not well understood, role in biodegradation of oil in the ocean water. The aim of this study is to explore the effect of these interactions on biodegradation of oil in the water. Laboratory experiments were performed, analyzing respiration and n-alkane and BTEX biodegradation in multiple conditions containing Corexit, alginate particles as marine snow, and kaolin clay. Two oil degrading bacterial pure cultures were added, Pseudomonas putida F1 and Rhodococcus qingshengii TUHH-12. Results show that the presence of alginate particles enhances oil biodegradation. The presence of Corexit alone or in combination with alginate particles and/or kaolin clay, hampers oil biodegradation. Kaolin clay and Corexit have a synergistic effect in increasing BTEX concentrations in the water and cause delay in oil biodegradation.

Metabolism of Ibuprofen by Phragmites australis: Uptake and Phytodegradation

This study explores ibuprofen (IBP) uptake and transformation in the wetland plant species Phragmites australis and the underlying mechanisms. We grew P. australis in perlite under greenhouse conditions and treated plants with 60 μg/L of IBP. Roots and rhizomes (RR), stems and leaves (SL), and liquid samples were collected during 21 days of exposure. Results show that P. australis can take up, translocate, and degrade IBP. IBP was completely removed from the liquid medium after 21 days with a half-life of 2.1 days. IBP accumulated in RR and was partly translocated to SL. Meanwhile, four intermediates were detected in the plant tissues: hydroxy-IBP, 1,2-dihydroxy-IBP, carboxy-IBP and glucopyranosyloxy-hydroxy-IBP. Cytochrome P450 monooxygenase was involved in the production of the two hydroxy intermediates. We hypothesize that transformation of IBP was first catalyzed by P450, and then by glycosyltransferase, followed by further storage or metabolism in vacuoles or cell walls. No significant phytotoxicity was observed based on relative growth of plants and stress enzyme activities. In conclusion, we demonstrated for the first time that P. australis degrades IBP from water and is therefore a suitable species for application in constructed wetlands to clean wastewater effluents containing IBP and possibly also other micropollutants.

Chemical dispersants: Oil biodegradation friend or foe?

Chemical dispersants were used in response to the Deepwater Horizon oil spill in the Gulf of Mexico, both at the sea surface and the wellhead. Their effect on oil biodegradation is unclear, as studies showed both inhibition and enhancement. This study addresses the effect of Corexit on oil biodegradation by alkane and/or aromatic degrading bacterial culture in artificial seawater at different dispersant to oil ratios (DORs). Our results show that dispersant addition did not enhance oil biodegradation. At DOR 1:20, biodegradation was inhibited, especially when only the alkane degrading culture was present. With a combination of cultures, this inhibition was overcome after 10days. This indicates that initial inhibition of oil biodegradation can be overcome when different bacteria are present in the environment. We conclude that the observed inhibition is related to the enhanced dissolution of aromatic compounds into the water, inhibiting the alkane degrading bacteria.

Effect of granular activated carbon concentration on the content of organic matter and salt, influencing E. coli activity and survival in fluidized bed disinfection reactor

Granular activated carbon (GAC) is used in water treatment systems, typically to remove pollutants such as natural organic matter, volatile organic compounds, chlorine, taste, and odor. GAC is also used as a key component of a new technology that combines a fluidized bed reactor with radio frequency electric fields for disinfection. So far, the effects of GAC on bacteria in these fluidized bed reactors are unclear. This paper describes a systematic study of the physico-chemical changes in five microbial media compositions caused by different concentrations (23-350 g/L) of GAC, and the effects of these physico-chemical changes on the metabolic activity and survival of a model microorganism (Escherichia coli YMc10) in a fluidized bed reactor. The chemical adsorption taking place in suspensions with specific GAC changed nutritional, osmotic, and pH conditions in the investigated microbial media (LB, diluted LB, PBS, diluted PBS, and tap water), leading to a decay of the metabolic activity and survival of E. coli. Especially media that are poor in organic and mineral compounds (e.g., PBS) with suspended GAC showed a concentration decay of 3.5 Log CFU/mL E. coli after 6 h. Organic compounds depletion and severe pH variation were enhanced in the presence of higher GAC concentrations.

Stimulation of hexachlorocyclohexane (HCH) biodegradation in a full scale in situ bioscreen

The feasibility of a bioscreen for the in situ biodegradation of HCH and its intermediates is demonstrated at a contaminated site in The Netherlands, via the discontinuous addition of methanol as electron donor. An infiltration system was installed and operated at the site over a length of 150 m and a depth of 8 m, to create an anaerobic infiltration zone in which HCH is converted. The construction of the infiltration system was combined with the redevelopment of the site. During passage through the bioscreen, the concentration of HCH in the groundwater decreased from 600 μg/L to the detection limit of the individual HCH isomers (0.01 μg/L) after one year of operation. The concentration of the intermediate biodegradation products benzene and chlorobenzene increased and achieved steady state values of respectively 800 and 2700 μg/L. Benzene and chlorobenzene were treated aerobically on site in an existing wastewater treatment plant. By changing the infiltration regime, it is conclusively shown that HCH removal is the result of the biological degradation and stimulated by the addition of methanol as electron donor. To our knowledge, this is the first successful field demonstration of the stimulated transformation of HCH to intermediates in a full scale anaerobic in situ bioscreen, combined with an aerobic on site treatment to harmless end products.

Genome analysis and physiological comparison of Alicycliphilus denitrificans strains BC and K601(T.)

The genomes of the Betaproteobacteria Alicycliphilus denitrificans strains BC and K601^T have been sequenced to get insight into the physiology of the two strains. Strain BC degrades benzene with chlorate as electron acceptor. The cyclohexanol-degrading denitrifying strain K601^T is not able to use chlorate as electron acceptor, while strain BC cannot degrade cyclohexanol. The 16S rRNA sequences of strains BC and K601^T are identical and the fatty acid methyl ester patterns of the strains are similar. Basic Local Alignment Search Tool (BLAST) analysis of predicted open reading frames of both strains showed most hits with Acidovorax sp. JS42, a bacterium that degrades nitro-aromatics. The genomes include strain-specific plasmids (pAlide201 in strain K601^T and pAlide01 and pAlide02 in strain BC). Key genes of chlorate reduction in strain BC were located on a 120 kb megaplasmid (pAlide01), which was absent in strain K601^T. Genes involved in cyclohexanol degradation were only found in strain K601^T. Benzene and toluene are degraded via oxygenase-mediated pathways in both strains. Genes involved in the meta-cleavage pathway of catechol are present in the genomes of both strains. Strain BC also contains all genes of the ortho-cleavage pathway. The large number of mono- and dioxygenase genes in the genomes suggests that the two strains have a broader substrate range than known thus far.

Anaerobic benzene degradation under denitrifying conditions: Peptococcaceae as dominant benzene degraders and evidence for a syntrophic process

An anaerobic microbial community was enriched in a chemostat that was operated for more than 8 years with benzene and nitrate as electron acceptor. The coexistence of multiple species in the chemostat and the presence of a biofilm, led to the hypothesis that benzene-degrading species coexist in a syntrophic interaction, and that benzene can be degraded in syntrophy by consortia with various electron acceptors in the same culture. The benzene-degrading microorganisms were identified by DNA-stable isotope probing with [U-(13) C]-labelled benzene, and the effect of different electron donors and acceptors on benzene degradation was investigated. The degradation rate constant of benzene with nitrate (0.7 day(-1) ) was higher than reported previously. In the absence of nitrate, the microbial community was able to use sulfate, chlorate or ferric iron as electron acceptor. Bacteria belonging to the Peptococcaceae were identified as dominant benzene consumers, but also those related to Rhodocyclaceae and Burkholderiaceae were found to be associated with the anaerobic benzene degradation process. The benzene degradation activity in the chemostat was associated with microbial growth in biofilms. This, together with the inhibiting effect of hydrogen and the ability to degrade benzene with different electron acceptors, suggests that benzene was degraded via a syntrophic process.

Degradation of 4-n-nonylphenol under nitrate reducing conditions

Nonylphenol (NP) is an endocrine disruptor present as a pollutant in river sediment. Biodegradation of NP can reduce its toxicological risk. As sediments are mainly anaerobic, degradation of linear (4-n-NP) and branched nonylphenol (tNP) was studied under methanogenic, sulphate reducing and denitrifying conditions in NP polluted river sediment. Anaerobic bioconversion was observed only for linear NP under denitrifying conditions. The microbial population involved herein was further studied by enrichment and molecular characterization. The largest change in diversity was observed between the enrichments of the third and fourth generation, and further enrichment did not affect the diversity. This implies that different microorganisms are involved in the degradation of 4-n-NP in the sediment. The major degrading bacteria were most closely related to denitrifying hexadecane degraders and linear alkyl benzene sulphonate (LAS) degraders. The molecular structures of alkanes and LAS are similar to the linear chain of 4-n-NP, this might indicate that the biodegradation of linear NP under denitrifying conditions starts at the nonyl chain. Initiation of anaerobic NP degradation was further tested using phenol as a structure analogue. Phenol was chosen instead of an aliphatic analogue, because phenol is the common structure present in all NP isomers while the structure of the aliphatic chain differs per isomer. Phenol was degraded in all cases, but did not affect the linear NP degradation under denitrifying conditions and did not initiate the degradation of tNP and linear NP under the other tested conditions.

Physiological and phylogenetic characterization of a stable benzene-degrading, chlorate-reducing microbial community

A stable anoxic enrichment culture was obtained that degraded benzene with chlorate as an electron acceptor. The benzene degradation rate was 1.65 mM benzene per day, which is similar to reported aerobic benzene degradation rates but 20-1650 times higher than reported for anaerobic benzene degradation. Denaturing gradient gel electrophoresis of part of the 16S rRNA gene, cloning and sequencing showed that the culture had a stable composition after the seventh transfer. Five bacterial clones were further analyzed. Two clones corresponded to bacteria closely related to Alicycliphilus denitrificans K601. The three other clones corresponded to bacteria closely related to Zoogloea resiniphila PIV-3A2w, Mesorhizobium sp. WG and Stenotrophomonas acidaminiphila. DGGE analysis of cultures grown with different electron donors and acceptors indicated that the bacterium related to Alicycliphilus denitrificans K601 is able to degrade benzene coupled to chlorate reduction. The role of the other bacteria could not be conclusively determined. The bacterium related to Mesorhizobium sp. WG can be enriched with benzene and oxygen, but not with acetate and chlorate, while the bacterium related to Stenotrophomonas acidaminophila grows with acetate and chlorate, but not with benzene and oxygen. As oxygen is produced during chlorate reduction, an aerobic pathway of benzene degradation is most likely.

Benzene Degradation Coupled with Chlorate Reduction in a Soil Column Study

Perchlorate and chlorate are electron acceptors that during reduction result in the formation of molecular oxygen. The produced oxygen can be used for activation of anaerobic persistent pollutants, like benzene. In this study chlorate was tested as potential electron acceptor to stimulate benzene degradation in anoxic polluted soil column. A chlorate amended benzene polluted soil column was operated over a period of 500 days. Benzene was immediately degraded in the column after start up, and benzene removal recovered completely after omission of chlorate or a too high influent chlorate concentration (22 mM). Mass balance calculations showed that per mole of benzene five mole of chlorate were reduced. At the end of the experiment higher loading rates were applied to measure the maximal benzene degradation rate in this system; a breakthrough of benzene was not observed. The average benzene degradation rate over this period was 31 micromol l(-1) h(-1) with a maximal of 78 micromol l(-1) h(-1). The high degradation rate and the necessity of chlorate indicate that oxygen produced during chlorate reduction indeed is used for the activation of benzene. This is the first column study where benzene biodegradation at a high rate coupled with anaerobic chlorate reduction is observed.

Hydrogen threshold concentrations in pure cultures of halorespiring bacteria and at a site polluted with chlorinated ethenes

Halorespiring microorganisms are not only able to oxidize organic electron donors such as formate, acetate, pyruvate and lactate, but also H(2). Because these microorganisms have a high affinity for H(2), this may be the most important electron donor for halorespiration in the environment. We have studied the role of H(2)-threshold concentrations in pure halorespiring cultures and compared them with mixed cultures and field data. We have found H(2)-threshold values between 0.05 and 0.08 nM for Sulfurospirillum halorespirans, S. multivorans and Dehalobacter restrictus under PCE-reducing and nitrate-reducing conditions. The reduction of PCE and TCE can proceed at H(2) concentrations of below 1 nM at a polluted site. However, for the reduction of lower chlorinated ethenes a higher H(2) concentration is required. This indicates that the measured H(2) concentration in situ can be an indicator of the extent of anaerobic reductive dechlorination.