The aim of this study was to assess the environmental impact of microbial fuel cells (MFCs) implemented in constructed wetlands (CWs). To this aim a life cycle assessment (LCA) was carried out comparing three scenarios: 1) a conventional CW system (without MFC implementation); 2) a CW system coupled with a gravel-based anode MFC, and 3) a CW system coupled with a graphite-based anode MFC. All systems served a population equivalent of 1500 p.e. They were designed to meet the same effluent quality. Since MFCs implemented in CWs improve treatment efficiency, the CWs coupled with MFCs had lower specific area requirement compared to the conventional CW system. The functional unit was 1 m3 of wastewater. The LCA was performed with the software SimaPro® 8, using the CML-IA baseline method. The three scenarios considered showed similar environmental performance in all the categories considered, with the exception of Abiotic Depletion Potential. In this impact category, the potential environmental impact of the CW system coupled with a gravel-based anode MFC was around 2 times higher than that generated by the conventional CW system and the CW system coupled with a graphite-based anode MFC. It was attributed to the large amount of less environmentally friendly materials (e.g. metals, graphite) for MFCs implementation, especially in the case of gravel-based anode MFCs. Therefore, the CW system coupled with graphite-based anode MFC appeared as the most environmentally friendly solution which can replace conventional CWs reducing system footprint by up to 20%. An economic assessment showed that this system was around 1.5 times more expensive than the conventional CW system.
Microbial Fuel Cells implemented in Constructed Wetlands (CW-MFCs) show limited performance. Geobacter Lovleyi has been demonstrated to be one of the predominant bacterial species in active CW-MFCs. The aim of this study was to characterize the growth of G.Lovleyi so as to identify if it could be a source for the observed CW-MFCs low performances. To this aim, G. Lovelyi was grown under three different electron donors (acetate, lactate and formate) and two electron acceptors (fumarate and Fe(III) citrate). G. Lovleyi growing and electron transfer characteristics was also studied by inoculating it in double chambered MECs (anodes poised at 31, 450 and 771 mV). Results showed that its growth was supported by acetate, with doubling times of 4.4±0.1 and 8±0.1 hours for fumarate and Fe(III) citrate as electron acceptors, respectively. G. Lovleyi was also demonstrated to be highly intolerant to oxygen, requiring cysteine as a reducing agent. In contrast, formate and lactate did not support cell growth even in the presence of cysteine. Maximum currents achieved were that of 0.08 mA and 0.26 mA for the MECs operated at 450 mV and 771 mV, respectively. However, no current was observed at 31mV. Confocal laser scanning microscopy (CLSM) analysis showed poor electrode coverage, indicating that G. Lovleyi did not attach to the electrode effectively. According to these results, low performances of CW-MFCs could by at least partially explained by the inability of G. lovleyi to oxidize the wide range of metabolites present in CW, to tolerate even trace oxygen concentrations or to efficiently attach to electrodes surface.
Clogging in HSSF CW may result in a reduction of system's life-span or treatment efficiency. Current available techniques to assess the degree of clogging in HSSF CW are time consuming and cannot be applied on a continuous basis. Main objective of this work was to assess the potential applicability of microbial fuel cells for continuous clogging assessment in HSSF CW. To this aim, two replicates of a membrane-less microbial fuel cell (MFC) were built up and operated under laboratory conditions for five weeks. The MFC anode was gravel-based to simulate the filter media of HSSF CW. MFC were weekly loaded with sludge that had been accumulating for several years in a pilot HSSF CW treating domestic wastewater. Sludge loading ranged from ca. 20 kg TS·m- 3 CW·year- 1 at the beginning of the study period up to ca. 250 kg TS·m- 3 CW·year- 1 at the end of the study period. Sludge loading applied resulted in sludge accumulated within the MFC equivalent to a clogging degree ranging from 0.2 years (ca. 0.5 kg TS·m–3CW) to ca. 5 years (ca. 10 kg TS·m–3CW). Results showed that the electric charge was negatively correlated to the amount of sludge accumulated (degree of clogging). Electron transference (expressed as electric charge) almost ceased when accumulated sludge within the MFC was equivalent to ca. 5 years of clogging (ca. 10 kg TS·m–3CW). This result suggests that, although longer study periods under more realistic conditions shall be further performed, HSSF CW operated as a MFC has great potential for clogging assessment.
The cathode of microbial fuel cells (MFCs) implemented in constructed wetlands (CWs) is generally set in close contact with water surface to provide a rich oxygen environment. However, water level variations caused by plants evapotranspiration in CWs might decrease MFC performance by limiting oxygen transfer to the cathode. Main objective of this work was to quantify the effect of water level variation on MFC performance implemented in HSSF CW. For the purpose of this work two MFCs were implemented within a HSSF CW pilot plant fed with primary treated domestic wastewater. Cell voltage (E-cell) and the relative distance between the cathode and the water level were recorded for one year. Results showed that E-cell was greatly influenced by the relative distance between the cathode and the water level, giving an optimal cathode position of about 1 to 2 cm above water level. Both water level variation and E-cell were daily and seasonal dependent, showing a pronounced day/night variation during warm periods and showing almost no daily variation during cold periods. Energy production under pronounced daily water level variation was 40% lower (80 +/- 56 mWh/m(2) . day) than under low water level variation (131 +/- 61 mWh/m(2) . day). Main conclusion of the present work is that of the performance of MFC implemented in HSSF CW is highly dependent on plants evapotranspiration. Therefore, MFC that are to be implemented in CWs shall be designed to be able to cope with pronounced water level variations.
This work aimed at determining the amount of energy that can be harvested by implementing microbial fuel cells (MFC) in horizontal subsurface constructed wetlands (HSSF CWs) during the treatment of real domestic wastewater. To this aim, MFC were implemented in a pilot plant based on two HSSF CW, one fed with primary settled wastewater (Settler line) and the other fed with the effluent of a hydrolytic upflow sludge blanket reactor (HUSB line). The eubacterial and archaeal community was profiled on wetland gravel, MFC electrodes and primary treated wastewater by means of 16S rRNA gene-based 454-pyrosequencing and qPCR of 16S rRNA and mcrA genes. Maximum current (219 mA/m(2)) and power (36 mW/m(2)) densities were obtained for the HUSB line. Power production pattern correlated well with water level fluctuations within the wetlands, whereas the type of primary treatment implemented had a significant impact on the diversity and relative abundance of eubacteria communities colonizing MFC. It is worth noticing the high predominance (13-16% of relative abundance) of one OM belonging to Geobacter on active MFC of the HUSB line that was absent for the settler line MFC. Hence, MFC show promise for power production in constructed wetlands receiving the effluent of a HUSB reactor. (C) 2015 Elsevier Ltd. All rights reserved.
Methane is emitted in horizontal subsurface flow constructed wetlands (HSSF CWs) during wastewater treatment. The objective of this work was to determine the influence of primary treatment and organic loading rate on methane emissions from constructed wetlands. To this aim, methane emissions from a HSSF CW pilot plant were measured using the closed chamber method. The effect of primary treatment was addressed by comparing emissions from wetlands receiving the effluent of an anaerobic (HUSB reactor) or a conventional settler as primary treatments. Alternatively, the effect of organic loading was addressed by comparing emissions from wetlands operated under high organic loading (52 g COD m (2) day (1)) and low organic loading (17 g COD m (2) day (1)). Results showed that methane emission rates were affected by the type of primary treatment and, to a lesser extent, by the organic loading applied. Accordingly, lower redox conditions and slightly higher organic loading of a wetland receiving the effluent of a HUSB reactor resulted in methane emissions twelve times higher than those of the wetland fed with primary settled wastewater. Moreover, systems subjected to three times higher organic loading than that recommended lead to higher methane emission rates, although high data variability resulted in no statistically significant differences.
Sediment microbial fuel cell (sMFC) represents a variation of the typical configuration of a MFC in which energy can be harvested via naturally occurring electropotential differences. Moreover, constructed wetlands show marked redox gradients along the depth which could be exploited for energy production via sMFC. In spite of the potential application of sMFC to constructed wetlands, there is almost no published work on the topic. The main objective of the present work was to define the best operational and design conditions of sub-surface flow constructed wetlands (SSF CWs) under which energy production with microbial fuel cells (MFCs) would be maximized. To this aim, a pilot plant based on SSF CW treating domestic sewage was operated during six months. Redox gradients along the depth of SSF CWs were determined as function of hydraulic regime (continuous vs discontinuous) and the presence of macrophytes in two sampling campaigns (after three and six months of plant operation). Redox potential (EH) within the wetlands was analysed at 5, 15 and 25 cm. Results obtained indicated that the maximum redox gradient was between the surface and the bottom of the bed for continuous planted wetlands (407.7 ± 73.8 mV) and, to a lesser extent, between the surface and the middle part of the wetland (356.5 ± 76.7 mV). Finally, the maximum redox gradients obtained for planted wetlands operated under continuous flow regime would lead to a power production of about 16 mW/m2.
Non-homogeneous mixing of methane (NHM) within closed chambers was studied under laboratory conditions. The experimental set-up consisted of a PVC vented chamber of 5.3 litres of effective volume fitted with a power-adjustable 12 V fan. NHM within the chamber was studied according to fan position (top vs lateral), fan airflow strength (23 vs 80 cubic feet per minute) and the mixing time before sample withdrawal (5, 10, 15 and 20 minutes). The potential bias of methane flux densities caused by NHM was addressed by monitoring the difference between linearly expected and estimated flux densities of ca. 400, ca. 800 and ca. 1,600 mg CH4.m-2 d-1. Methane within the chamber was under non-homogeneous conditions. Accordingly, methane concentrations at the bottom of the chamber were between 20 to 70% higher than those recorded at the middle or top sections of the chamber, regardless of fan position, fan air-flow strength or time before sample withdrawal. NHM led to notable biases on flux density estimation. Accordingly, flux density estimated from top and middle sampling sections were systematically lower (ca. 50%) than those expected. Flux densities estimated from bottom samples were between 10% higher and 25% lower than expected, regardless of the flux density considered.
'Subsurface-flow constructed wetlands have become a very popular cost effective and green technology for treatment of waste water throughout Europe and the rest of the world. Original predictions over the longevity of constructed wetlands were approximately 50 -100 years (Conley et al 1991). However, it has become disappointingly apparent that these systems are clogging and have on average a lifetime of less than 10 years (Griffin et al 2008). Currently when a wetland becomes clogged, the whole site has to be refurbished and the reeds regrown, which takes several years and has significant economic consequences for the operators.
Our project ARBI aims to develop and trial an Autonomous Reed Bed Installation containing a magnetic resonance probe, can be deployed in several locations within a wetland to give measurements at different depths. By measuring the relaxation times of MR, sufficient information can be obtained to determine the clog state of the gravel bed of a wetland. This would enable the operators to isolate those areas of the bed where the problem resides and make a partial intervention, without the need to remove and re-plant the whole reed bed. When the system is developed, we will have potential for application in other water treatment systems based on subsurface flows like: slow rate sand filters, and river bank filtration.
The project will have major benefits for those organisations who would like to install reedbeds but have resisted doing because of concerns over performance and maintenance costs. These will increase the potential size of the market for reedbed installers and benefit the SMEs in the consortium who currently are operating in a constrained market which has not achieved its true potential.'