- Email: [email protected]
Evaluation of Sedum as driver for plant microbial fuel cells in a semi-arid green roof ecosystem
Ecological Engineering 108 (2017) 203–210
Contents lists available at ScienceDirect
Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng
Evaluation of Sedum as driver for plant microbial fuel cells in a semi-arid green roof ecosystem Natalia F. Tapiaa,b, Claudia Rojasc, Carlos A. Bonillaa,b, Ignacio T. Vargasa,b, a b c
MARK
⁎
Centro de Desarrollo Urbano Sustentable (CEDEUS), Chile Department of Hydraulic and Environmental Engineering, Pontificia Universidad Católica de Chile, 7820436 Santiago, Chile Institute of Agricultural Sciences, University of O’Higgins, Rancagua, Chile
A R T I C L E I N F O
A B S T R A C T
Keywords: Green roofs Plant microbial fuel cells Pyrosequencing Succulent plant
A plant microbial fuel cell (PMFC) is a novel and sustainable bioelectrochemical technology that takes advantage of the organic compounds released through the plant roots (exudates) to produce power via electrochemically active bacteria. This technology has been implemented in flooded ecosystem; however, their use in arid or semiarid environments has been less explored. Thus, the use of plant adapted to conditions of less water availability could expand the application of PMFCs. In this study, seven Sedum species (succulent plants) were evaluated for their ability to generate energy under non-saturated conditions. Electrochemical results confirmed power generation by the Sedum species tested, with reactors using Sedum hybridum achieving the highest power density (92 μW m−2), followed by reactors tested with S. rupestre (15.5 μW m−2). The lowest performance was observed in systems using S. spurium (< 1 μW m−2). Total organic carbon from root exudates did not have a direct effect on PMFCs performance. Pyrosequencing analysis revealed higher abundance of bacteria in the family Micrococcaceae in anodic biofilms of S. hybridum (13.47 ± 2.69%) reactors than in biofilms of S. rupestre (2.28 ± 0.52%), and S. spurium (2.29 ± 1.94%) PMFCs. Although the current magnitudes observed in our study were lower than those reported for flooded PMFCs, a positive relationship between water content of plant growth media and power outputs was observed. After irrigation, current and water gradually decrease over time with a similar rate, showing a correlation coefficient between both variables of 0.95 ± 0.01. Moreover, it encourages further testing of PMCFs as potential indicators of soil water content in semiarid green roofs where a better water-use efficiency is needed.
1. Introduction Microbial fuel cells (MFCs) are bioelectrochemical systems that use electrochemically active bacteria (EAB) to catalyze reactions at one or both electrodes (the anode and cathode). Initially, MFCs were conceived to transform chemical energy stored in wastewater into electricity. However, the progress achieved for this technology in recent years has expanded its potential applications (Cao et al., 2009; Logan, 2010). One example of it is the sediment- or soil-based microbial fuel cell (SMFC), which uses organic matter stored in sediments as a source of energy for EAB. This technology has been considered an innovative opportunity to provide power to electronic devices in remote locations, where it is difficult and expensive to replace batteries (Donovan et al., 2008). Nevertheless, the availability of organic matter represents a limitation for this technology during long-term operation. A sustainable alternative to supply organic matter to SMFC devices is the use of plants
that can provide the fuel for these systems throughout photosynthesis. This type of technology has been named plant microbial fuel cells (PMFCs). In a PMFC, the plant generates organic compounds released through it roots (exudates), which are later used by EAB for power production (Helder et al., 2013a). The implementation of PMFCs has been mostly tested in water-saturated environments with plant adapted to this condition (Table 1). Among them, the main species tested in PMFCs since this technology emerged in 2008 have been: Glyceria maxima (Strik et al., 2008; Timmers et al., 2012a; Timmers et al., 2013); Spartina anglica (Helder et al., 2010; Helder et al., 2013b) and Oryza sativa (Chen et al., 2012; De Schamphelaire et al., 2010; Kaku et al., 2008). All of these species have in common their capacity to grow under flooded conditions as a result of their adapted tissue called aerenchyma, which enables them to carry oxygen to their roots under such environments (Ap Rees and Wilson, 1984; Justin and Armstrong, 1991; Maricle and Lee, 2002).
⁎ Corresponding author at: Department of Hydraulic and Environmental Engineering, Pontificia Universidad Católica de Chile, Av. Vicuña Mackenna 4860, 7820436 Macul, Santiago, Chile. E-mail address: [email protected] (I.T. Vargas).
http://dx.doi.org/10.1016/j.ecoleng.2017.08.017 Received 8 November 2016; Received in revised form 14 July 2017; Accepted 15 August 2017 0925-8574/ © 2017 Elsevier B.V. All rights reserved.
Ecological Engineering 108 (2017) 203–210
Saturation
Saturation Saturation Saturation Not saturation
120
112 65 90 360
10 – – – – 4.52 105 – – – – 2.71
22 163 18 0.09
– 100 – 6000
O2, Ferricyanide O2 O2 O2
Helder et al., 2010 Chiranjeevi et al., 2012 Lu et al., 2015a,b This study
The water content in PMFCs has been reported as a key factor for power generation as this condition allows to maintain the anode electrode under anoxic conditions and facilitate ion transport to the cathode electrode (Chiranjeevi et al., 2012). Indeed, significant drop in voltage (from 634 mV to 0 mV) and current density (200 mA/m2 to 125 mA/ m2) have been previously shown in PMFCs after irrigation ceased (Chiranjeevi et al., 2012; Li et al., 2016). Thus, it is not surprising that PMFCs are currently mainly used in environments under saturated conditions like wetlands, with plant species adapted to such environments (Table 1). However, PMFCs can also be implemented in engineered environments such green roofs, representing a benefit over other sources of energy generated from biomass (Strik et al., 2008). In spite of the potential to extend this technology to other environments, performance of PMFCs in green roofs operated under conditions different than saturation is virtually unknown. In conditions different than humid ecosystem of the northern hemisphere, where green roofs originally started, their benefits including reducing runoff, improving air quality and increasing biodiversity are still highly valued (Lu et al., 2015a). However, the implementation of green roofs in arid and semiarid zones requires alternative plant species to support these systems. Under such environmental conditions, succulent plants, capable of retaining water are among the vegetation most frequently used. Among them, some species in the Sedum genus have been previously used due to their low water requirements and tolerance to droughts. They are able to resist these conditions due to their adaptation to a special photosynthetic pathway, known as Crassulacean Acid Metabolism (CAM), which allow them to keep their stomata closed during the day (avoiding water evaporation) and open them at night to fix carbon dioxide (Farrell et al., 2012). Soil conductivity and water content could limit the PMFCs performance in arid and semi-arid environments (Domínguez-Garay et al., 2013; Wang et al., 2012). An increment in soil water content could result in a reduction of the internal resistance of the system, allowing an increment in the current generated by the PMFC. In this type of climates, the water is a critical resource that should be efficiently used, particularly in green infrastructure used as a driver to transform existing cities towards a next generation of net-zero (water and energy) urban settlements (Lehmann, 2014). The use of commercial moisture sensors allows a more efficient and sustainable use of water. However, this type of sensors is still relatively expensive for domestic and commercial applications, particularly in developing countries. Hence, the study of the influence of variable water contents and the quality of soil or substrate used to support plant growth are needed and expand the application of PMFCs, not only in search of a new source of energy, but also for the development of a low-cost water content biosensor. In this study, we assessed the performance of PMFCs supported by Sedum plant species under unsaturated conditions and thus the potential to implement this technology in green roof environments of semiarid regions. We hypothesized that these plants would provide organic matter through their roots to be used by EABs in electrodes of constructed PMFCs. For this purpose, seven Sedum species were tested in terms of their capability to generate electricity and organic exudates. Three representative PMFCs were also evaluated for the microbial community developed on anode and cathode electrodes.
Rhizodeposits Rhizodeposits and soil Rhizodeposits Rhizodeposits and soil A.anomala P. setaceum C. indica S. hybridum
Natronocella Beijerinckiaceae Rhizobiales Bacteria Bacteria Geobacteraceae Anaerolineaceae Micrococcaceae Rhizodeposits and soil O. sativa
Saturation Saturation Saturation Rhizodeposits Rhizodeposits Rhizodeposits S. anglica S. anglica O. sativa
78 154 134
– 52 –
6
156
O2
Timmers et al., 2010 Helder et al., 2010 De Schampeliare, 2008, 2010 Kaku et al., 2008 O2 O2, Ferricyanide Ferricyanide 1800 – – 22 21 21 – – – 141 – 44
79 222 33
525 900 67 80 4 – 153 – Saturation Saturation 67 – Rhizodeposits Rhizodeposits G. maxima G. maxima
Bacteria Geobacteraceae Ruminococcaceae Comamonadaceae Bacteria Bacteria Desulfobulbus Geobacteraceae Archaea
32 164
Max Avg Avg
Max
Power density (mW m−2) Current density (mA/ m2) Water condition Operation time (days) Microbial community Electron donor Plant
Table 1 Performance PMFCs reported in literature and results of Sedum species with the best performance in this study are shown.
Internal resistance (ohm)
O2 O2
Electron acceptor
Ref.
Strik et al., 2008 Timmers et al., 2012a,b
N.F. Tapia et al.
2. Materials and methods 2.1. Screening of sedum species in plant microbial fuel cells A preliminary screening test was first carried out to evaluate the potential of using Sedum species in PMFCs. Eight reactors were run in this experiment, seven contained one individual seedling S. album, S. hybridum, S. kamtschaticum, S. reflexum, S. rupestre, S. sexangulare or S. spurium plant species and one was left unplanted to represent a control pot (Fig. 1). These species additionally were used in a pilot green roof constructed at the San Joaquín Campus, Pontificia Universidad Católica 204
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Fig. 1. A: The PMFC Reactors with each one of the seven Sedum species used in the experiment: (1) S. kamtschaticum, (2) S. rupestre, (3) S. album, (4) S. hybridum, (5) S. spurium, (6) S. sexangulare, (7) S. reflexum, and (8) control (no plant) B. Diagram of the PMFC reactor configuration during operation.
conductivity sensor GS3 (Decagon®, Pullman, USA).
de Chile (PUC), Santiago, Chile (Reyes et al., 2016). The PMFC reactors consisted of 450 mL plastic pots, with holes at the bottom to allow free drainage. Each pot was filled with 400 mL of plant growth substrate collected in May 2013 from a green roof under construction (before plants establishment) at the same Campus (S1). The substrate consisted of a mineral matrix of sand, silt, and clay amended with organic matter (Table S.1). Electrodes made of a circular graphite fiber felt of 92 mm in diameter and 6 mm thick (MudWatt, California, USA) were initially installed; cathode electrodes were buried at a depth of 7 cm and anode electrodes were placed on the surface of the substrate. The buried electrodes were inoculated with mud, from a pond used as a water supply for irrigation located at the same University’s campus, once placed at the bottom of the pots. The reactors were operated during an initial startup period of 120 days for system adjustment (from December 2013 to April 2014). During this time, plant irrigation was performed weekly by adding approximately 150 mL of tap water in the upper part of the reactor until the water drained for the bottom pot. From day 60–110 the irrigation system was changed to a continuous irrigation mode in order to use irrigation conditions similar to green roof (rainfall events and sprinkler irrigation) and evaluate the effect in plant performance according to irrigation strategy. After 120 days, the upper graphite fiber felt electrodes (anodes) were replaced with activated carbon granules (3 mm in diameter; 35 cm3 volume) and a graphite rod (as connection), for a better adhesion of electrode to the substrate during the periods without feeding water. Titanium wires (23 cm in length; 0.5 mm in diameter) were used to connect the two electrodes to an external resistor of 1 kΩ (Fig. 1). The reactors were maintained under controlled temperature conditions and continuously monitored over a period of 360 days (from December 2013 to December 2014). The average room temperature during this period was 20 ± 2 °C (except between days 260 and 350 when reactor was exposed to 32 ± 2 °C). Light periods extended for 14 h per day, with an average radiation of 23 ± 2 W m−2, and they were measured every 10 min by a solar radiation sensor (Decagon®, Pullman, USA). During the evaluation period (day 120 and beyond) irrigation was performed weekly by adding 100 mL of tap water or until saturation was reached. Cell voltage was measured every 10 min with a data acquisition system (2700; Keithley, USA). Power and current densities were normalized by the area of the cathode electrode (0.0067 m2), where the cathodic reaction occurred. The polarization curve and the power density curves were computed by varying the external resistance from 10 Ω to 100 kΩ every 30 min or until the system reached a stable voltage. Substrate water content, temperature and electrical conductivity were measured by a temperature, humidity and ground
2.2. Evaluation of PMFCs performance using sedum hybridum After the preliminary test was conducted, new plants for the species used in the reactors showing the best performance were obtained for further characterization. New PMFCs reactors consisted of three pots with Sedum hybridum and one pot with no plant as control. The plants and growing media were obtained from a second green roof established at the San Joaquín Campus, which was already under operation using the same Sedum species evaluated in the screening test (Reyes et al., 2016). Electrodes consisted of cathodes made of graphite fiber felt and anodes of activated carbon granules (as described above). Anodes were inoculated with mud obtained from the same pond the inoculum for the screening test was sampled. The PMFCs were maintained under the same conditions that screening test, with an average room temperature of 20 ± 2 °C over period of 215 days (from December 2014 to July 2015) and light period of 14 h per day. Irrigation was performed weekly by adding 150 mL of tap water or until water drained by bottom pot. Cell voltage was measured same that previous experiment, every 10 min, and current densities also were normalized by the area of the cathode. 2.3. Plant root exudate analysis A separated analysis was conducted on August 2015 to evaluate root exudates. For root exudate measurements, new plant seedlings (n = 3) of similar size and of each of the seven species tested in the screening test, were obtained from the same plant nursery. Plants were weighed after substrate particles were gently removed by manually shaking their roots. To further remove substrate particles from roots, they were rinsed with tap water and dipped into 30 mg L−1 of chloramphenicol for 2 h to minimize microbial growth (Subbarao et al., 1997). Roots were then washed again with tap water and sterile deionized water to remove residues of the antibiotic used. To collect exudates, plant roots were submerged in sterile centrifuge tubes containing 50 mL of deionized water and then aseptically stored at 20 °C for 48 h. Each tube was covered with aluminum foil to avoid light exposure. After this time, plants were removed from the tubes and exudate solutions were collected. Each solution was acidified to pH ∼ 2.4 using 6 M HCl and maintained at 4 °C until total organic carbon (TOC) was measured using a Shimadzu TOC-L CPH equipment (Shimadzu, Japan). The TOC contents of each solution were normalized by the measured weight of fresh plant biomass. 205
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Alignment was performed using PyNAST (Caporaso et al., 2010a) against the Greengenes core set (http://greengenes.lbl.gov/), and then filtered to remove gap positions.
2.4. DNA extraction and pyrosequencing The electrode samples for microbial community analysis were taken from the PMFCs of the screening test, with the highest, the second highest, and the lowest power density generation after 360 days of system operation. DNA extraction was performed using the Power Soil® DNA isolation kit (MoBio Laboratories, Inc., Carlsbad, CA) following the manufacturer’s instructions. Each of the selected PMFCs was dismantled under sterile conditions to obtain the electrode samples. The upper electrodes (anodes) were placed on sterile petri dishes from where pieces of graphite rod and granules were randomly selected to obtain duplicate samples of approximately 0.27 g for each of the three PMFCs examined. The cathode samples were aseptically sectioned to obtain eight small pieces (∼0.1 g) of graphite fiber felt per reactor. These subsamples were slightly moistened using sterile deionized water, previous to DNA extraction to avoid excessive absorption of extraction solutions. DNA samples extracted from each of the small graphite felt fiber subsamples were randomly selected and poled together to obtain two composite DNA samples per cathode electrode. The DNA quality and concentration were determined using a NanoDrop 2000c Spectrophotometer (Thermo Fisher Scientific Inc.). All samples were stored at −80 °C prior to the pyrosequencing analysis. The pyrosequencing analysis was performed using a 454 Genome Sequencer FLX Titanium System (Roche Diagnostics, Indianapolis, IN, USA) at the Research and Testing Laboratory (Lubbock, Texas. USA). The amplicon library was prepared for each sample using a bacteriaspecific primer set, 28F (5′-GAGTTTGATCNTGGCTCAG −3′) and 519R (5′- GTNTTACNGCGGCKGCTG −3′) flanking the regions V1-V3 of the 16S rRNA gene (Frank et al., 2013). The PCR reactions were performed in 25 μL reaction volumes with the Qiagen HotStar Taq master mix (Qiagen Inc., Valencia, California), 1 μL (5 μM) of each forward and reverse primer and 1 μL of the DNA extract. The reaction conditions consisted of an initial denaturation step at 95 °C for 5 min followed by 35 cycles, each consisting of denaturation at 94 °C for 30 s, primer annealing at 54 °C for 40 s, and an extension at 72 °C for 1 min with a final extension at 72 °C for 10 min and a 4 °C hold. After the PCR amplification, the products were visualized with eGels (Like Technologies, Grand Island, New York). The amplicons were then pooled equimolar, and each pool was cleaned and selected for sizes using Agencourt AMPure XP (BeckmanCoulter, Indianapolis, Indiana) as directed by the Roche 454 protocols (454 Life Sciences, Branford, Connecticut). The selected pools were then quantified and diluted for use in the emPCR reactions. These reactions were performed and subsequently enriched following the established manufacturer’s protocols (454 Life Sciences).
2.6. Statistical analysis Alpha diversity analysis included the total number of observed OTUs, Good’s sample coverage, Shannon and inverse Simpson diversity indices obtained from normalized data after ten iterations performed per index. Beta diversity, based on phylogenetic information, was assessed by a Principal Coordinate analysis (PCoA) on unweighted UniFrac distances. Differences among the Sedum species were evaluated comparing the alpha diversity and TOC analysis. These were evaluated with a one-way Analysis of Variance (ANOVA) and Tukey’s test using Microsoft Excel data analysis. P-values less than 0.05 were considered significant. 3. Results and discussion 3.1. PMFCs performance in the screening test Over the period of 150 days of system operation the seven PMFCs reactors tested (screening test) showed power generation, as opposed to the control reactor that ceased to produce recordable values (S2). This could have been due to the consumption by microorganisms of initial organic matter available in the substrate (2.85%). Hence, for the subsequent operation, the power obtained from the PMFC reactors was assumed to be mainly due to the chemical energy provided by the plant as exudates. Our results showed no dependency of current production on light periods, thus the circadian oscillation reported in previous studies (Helder et al., 2013b; Kaku et al., 2008) using S. anglica y O. sativa was not observed for the Sedum PMFCs tested in this study under controlled laboratory conditions. Power density curves performed on day 345 of operation (Fig. 2) indicated that S. hybridum displayed the highest performance at the final stage of the experiment, reaching a maximum of 92 μW m−2. S. rupestre, S. sexangulare and S. album followed with maximum power densities of 15.5 μW m−2; 8.4 μW m−2; and 2.4 μW m−2, respectively. The lowest performances were achieved by S. kamtschaticum, S. spurium
2.5. Processing and analysis of pyrosequence data The sequences obtained in this study have been deposited into the NCBI Sequence Read Archive (SRA) with the accession No. SRP062553. They were processed and analyzed using the Quantitative Insights Into Microbial Ecology (QIIME) program (Caporaso et al., 2010b) through the MacQIIME version 1.8.0. Sequences were denoised by removing the reads that did not match to barcodes or primers, had homopolymers longer than 6 bp. and quality scores lower than 25 (Quince et al., 2011). Only those filtered sequences between 200 bp and 1000 bp in length were included for further analysis. The sequences were then assigned to operational taxonomic units (OTUs) at 97% similarity using the de novo OTU picking protocol with usearch to additionally identify and remove singletons and chimeras (Edgar, 2010; Edgar et al., 2011). A representative sequence for each OTU was selected for downstream analysis. Taxonomic assignment was done using the uclust method and the Greengenes taxonomy database (DeSantis et al., 2006; McDonald et al., 2012) released on August 2013 (13.8 v) at a 90% confidence threshold. The unassigned sequences, referring to those sequences that could not be classified at the domain level, and sequences classified as chloroplast and mitochondria, were removed from the dataset.
Fig. 2. Power curves conducted to seven Sedum species tested in this study after 345 days of operation.
206
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Fig. 3. P-MFCs performance over time. Evaluation with distinct irrigation regimes: drip irrigation (A) (days 99–110) and weekly irrigation (B) (days 235–246).
and S. reflexum, all of which had values lower than 1 μW m−2. This difference between species is due to the performance at the final stage of PMFC reactors, where S. hybridum shows a higher current density in relation to S. rupestre (Figure S.2). As it was expected, the low water content in the soil affects the mobility for cation transport, increasing the internal resistance of the system (Chiranjeevi et al., 2012). Internal resistances were estimated from polarization curves, reaching values of 15 kΩ for S. rupestre and S. sexangulare, and 5.34 kΩ for S. hybridum. These values are higher compared to previous PMFC studies (Table 1). The application of drip and weekly irrigation resulted in differences in current generation (Fig. 3). While under drip irrigation relatively constant currents were observed, under weekly irrigation there was a sudden increase of current, observed after each irrigation event. After irrigation, soil moisture measured in S. hybridum and S. rupestre PMFC, reached up 40% v/v, decreasing at similar rate that the produced current, until values around 5% v/v corresponding to negligible current values. The correlation analysis between both variables, after the maximum current was reached, showed r values of 0.96 and 0.94 for S. rupestre and S. hybridum, respectively (Figure S.3). This suggests a close relationship between power generation and soil moisture content, with higher power obtained at higher soil moisture. This result is in agreement with data presented by Chiranjeevi et al. showing a significantly drop in the open circuit voltage when water supplementation was stopped, following by an increase when irrigation was restored (Chiranjeevi et al., 2012).
3.3. Analysis of root exudates The total organic carbon analysis of the root exudates obtained from the seven Sedum species tested in this study showed relatively low concentrations, ranging from 1.87 to 16.59 mg L−1 per gram of fresh plant. Similar values were observed for almost all species tested; only S. sexangulare had a significantly higher concentration (9.99 ± 5.75 mg L−1 per gram of fresh plant) according to Tukey’s test (p < 0.05) (Table 2). The similarity observed in carbon content among root exudates suggests that this parameter would not have a direct positive effect on PMFC performance, as the reactor resulting in the greatest current production (S. hybridum) did not necessarily achieved the highest TOC content (4.29 ± 1.29 mg L−1 per gram of fresh plant). It is likely that more than the total amount of carbon released by plant roots, the composition of these exudates could be more important for power generation as it has been previously reported that EAB have different affinity for different organic compounds (Chae et al., 2009). We acknowledge though that exudation process can be influenced by many factors, such as collection time, damage inflicted on the root, collection medium and temperature, among others (Pinton et al., 2007), which might have influenced our results. In addition, due to the long time of collection of exudates, some easily degradable organic compounds could have been decomposed, causing the low values detected in our study. 3.4. Analysis of microbial community According to Pinton et al. (2007) the composition of exudates depends of each plant species. Thus, Sedum species tested in this study could exudate different organic compounds, generating differences in the community composition and the subsequent PMFC performance. Since, no current was observed on the control reactor, electrodes of three Sedum species (representatives of the highest, second highest and the lowest power density) were analyzed, showing slight differences in microbial composition and diversity.
3.2. PMFCs performance using S. hybridum To assess reproducibility of the observed results obtained by the screening test, triplicated P-MFC using S. hybridum were run in the laboratory. Reactors operated under weekly irrigation mode show maximum current densities (after of 150 days of the start-up) of 2.36, 1.58 and 1.54 mA/m2, comparable with the average current density of 2.71 mA/m2 obtained with the screening test and follow the same water-content dependence observed in the first set of experiments. Interestingly, after about 150 days of operation the control reactor also stopped current production supporting the hypothesis that power obtained is provided by the plant as exudates (Fig. 4). Even though the power densities obtained in this study are considerable lower than those reported for PMFCs operated under water saturation (Timmers et al., 2012b), the observed relationship between the water content in the substrate and current generated by the Sedum PMFC reactors, could be used as water content indicator expanding the potential applications of these systems. Moreover, it encourages further testing of PMCFs as potential sensor for soil water content in actual semi-arid green roofs. This opportunity could impact the sustainability of urban settlements by including low-cost technology for monitoring the soil water content, leading to a better management of water, energy and land.
3.4.1. Bacterial community composition A total of 12 phyla were detected in anode samples (upper electrodes) and 13 in cathode samples (lower electrodes) (Table S.2), with seven dominant groups identified in anodes (Fig. 5). Proteobacteria was the most abundant phylum in anode (44.8% to 67.1%) and cathode (29.2% to 59.9%) electrodes, followed by Actinobacteria (13.5% to 38.98% in anodes; and 7.26% to 15.38% in cathodes) and Chloroflexi (3.07% to 16.03% in anodes; and 3.47% to 15.58% in cathodes). These three phyla have been previously reported as the dominant bacterial groups in electrodes of PMFCs, along with Acidobacteria and Bacteroidetes, which were also detected in our study (Lu et al., 2015b). Proteobacteria in the family Comamonadaceae and the genus Paracoccus were among the most abundant members in anode electrodes (Table 3). Both Comamonadaceae and Paracoccus have a negative impact on power generation, in the presence of oxygen or nitrate, since 207
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Fig. 4. Triplicated P-MFC experiment using Sedum hybridum. The experiment shows the observed behavior for PMFCs of Fig. 3 (and S2) (A). After a starup period of 150 days, while no current was recorded for the control reactor (no plant), the three reactors with S. hybridum show increases in current after irrigation (B).
Table 2 Total Organic Carbon in the root exudate of 21 plants from the seven Sedum species.
Table 3 Relative abundances of the main bacterial taxonomic groups at genus or higher taxa levels in the samples of the anode (A1, A2) and cathode (C1,C2) electrodes of the three Sedum species: S. rupestre, S. hybridum and S. spurium.
Total Carbon Organic (mg L−1 per gr of plant) Specie
Plant 1
Plant 2
Plant 3
Mean ± SD
S. S. S. S. S. S. S.
5.74 1.96 6.03 1.87 6.31 4.79 3.12
3.93 4.42 16.59 2.19 5.86 7.54 2.13
3.23 2.84 7.36 3.55 3.96 5.56 2.57
4.29 3.07 9.99 2.54 5.38 5.96 2.61
hybridum rupestre sexangulare album spurium kamtschaticum reflexum
± ± ± ± ± ± ±
1.29 1.25 5.75 0.89 1.25 1.42 0.49
Relative Abundance (%) a a b a a a a
Phylum: Proteobacteria Family Comamonadaceae Genus Paracoccus Family Rhodobacteraceae Genus Pseudomonas Genus Rubellimicrobium Genus Rhodobacter Phylum: Actinobacteria Family Micrococcaceae Order Acidimicrobiales Order Actinomycetales Phylum Chloroflexi Genus Ardenscatena
Different letters indicate statistically significant differences (p < 0.05) according to Tukey’s test.
Phylum: Proteobacteria Family Ectothiorhodospiraceae Phylum: Acidobacteria Order DS-18 Phylum: Actinobacteria Order 0319-7L14 Phylum Nitrospirae Order Nitrospirales
S. hybridum
S. rupestre
S. spurium
A1
A2
A1
A2
A1
A2
10.42 9.31 3.34 1.65 1.69 5.35
4.70 6.11 3.46 7.52 1.23 2.65
12.48 12.16 5.32 1.89 1.52 1.90
19.99 9.33 4.89 3.30 1.77 0.61
5.68 10.23 6.15 6.28 5.18 1.98
3.49 10.42 6.38 3.46 4.95 1.22
15.37 3.34 6.06
11.57 5.34 4.65
2.65 5.76 1.16
1.91 6.32 1.02
0.92 2.16 0.39
3.67 2.29 2.20
2.32 C1
11.27 C2
7.27 C1
10.76 C2
14.45 C1
10.27 C2
0.03
19.87
16.31
0.02 8.40
5.37
8.58
4.35
6.98
9.93
2.58
4.74
3.01
5.74
0.65
1.21
4.04
5.29
0.95
5.16
1.25
0.47
they decrease the availability of compounds that could be used by EABs (Kawaichi et al., 2013; Timmers et al., 2012a; Zhang et al., 2014). In cathode electrodes, members in the family Ectothiorhodospiraceae were especially abundant in S. spurium reactors. However, for the other two cathodes (S. hybridum and S. rupestre), this family was not observed. It is worth noticing that one of the most commonly reported EABs, Geobacter spp., (De Schamphelaire et al., 2010; Timmers et al., 2012a) was not detected in our study (Table 3). Among the Actinobacteria phylum, members in the Micrococcaceae family were especially abundant in the anode of the S. hybridum. Among this group, the species Kocuria rhizophila has been reported as a bacterium capable of generating electricity from glucose oxidation (Luo et al., 2015). Thus, the possible presence of members of the Micrococcaceae family as K. rhizophila or unreported exoelectrogenic species could explain the better performance of the reactor with S. hyridum. In the phylum Chloroflexi, the most abundant group was the Ardenscatena genus, which has a negative impact in power generation in presence of oxygen. The implementation of Sedum plants in PMFCs did not seem to affect the typical microbial composition of soil systems. In these environments, the Proteobacteria, Chloroflexi and Actinobacteria, among other
Fig. 5. Relative abundances of bacterial phyla in: (A) anode samples (upper electrodes), and (B) cathode samples (lower electrodes) of S. hybridum, S. rupestre, and S. spurium species. The bacterial group others consisted of sequences classified at the bacterial domain level only (≤0.1%) and other phyla comprising less than 0.5% of classified sequences.
208
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Table 4 The alpha diversity in the upper (A1, A2) and lower (C1, C2) electrodes corresponding to S. rupestre (P2), S. hybridum (P4) and S. spurium (P5). The observed OTUs, the sample coverage and the Shannon and inverse Simpson diversity indices were calculated from the normalized datasets (2245 sequences per each sample). The bacterial OTUs were generated at 97% sequence similarity. Diversity indices
P2A1 P2A2 P4A1 P4A2 P5A1 P5A2 P2C1 P2C2 P4C1 P4C2 P5C1 P5C2
Number of total sequences
Number of observed OTUsa
Sampling coverage
Shannon
7427 4127 2245 4536 5191 6913 3253 3102 4998 5025 4645 3647
344.7 ± 8.78 296 ± 5.44 275 ± 0 282.9 ± 5.30 323.6 ± 5.32 312.5 ± 6.17 644.4 ± 6.60 668.8 ± 7.08 699.6 ± 9.42 740.2 ± 13.76 486.8 ± 6.56 606.8 ± 5.94
0.93 0.95 0.96 0.95 0.95 0.95 0.88 0.88 0.86 0.84 0.90 0.88
6.66 6.05 6.35 6.39 6.55 6.81 8.27 8.44 8.49 8.53 6.95 7.45
b c a a d ad ab ac bc abc bd cd
± ± ± ± ± ± ± ± ± ± ± ±
Inv-Simpson 0.04 0.05 9.36e-16 0.02 0.02 0.03 0.02 0.02 0.04 0.04 0.04 0.05
c d a a ad bc ab ac b b bd abc
36.51 ± 1.73 18.06 ± 1.1 36.34 ± 7.45e-15 41.30 ± 1.02 34.31 ± 0.89 52.48 ± 2.65 97.53 ± 5.04 141.73 ± 4.85 126.90 ± 8.86 127.62 ± 6.52 21.55 ± 1.15 28.34 ± 1.47
a b a a a bc ab ac c c b d
Mean ± Standard Deviation values with different letters in the same column indicate statistically significant differences (p < 0.05) according to Tukey’s test. a Total number of OTUs identified per sample.
phyla also identified in this study are typically reported as abundant and ubiquitous members of the microbial community (Fierer et al., 2012; Janssen, 2006; Lauber et al., 2009). Moreover, the presence of Actinomycetales, Rhizobiales, Pseudomonas and Rhodobacter representatives, which are groups known to play a role in organic matter decomposition or nutrient cycling in soils (Maier and Pepper, 2009), could indicate the lower disturbance of PMFCs on soil functioning. This phenomenon takes special relevance in the event of PMFCs implementation in green roofs that are already under operation, and where active soil microbial communities have been already established.
3.4.2. Diversity analysis The total number of curated sequences obtained per samples fluctuated from 2245 and 7427, with coverage ranging from 0.88 and 0.95 (Table 4). The highest bacterial diversity, as estimated by number of OTUs, the Shannon and Inv-Simpson indices (Table 4), was achieved in the cathode electrodes, which is consistent with the highest number of phyla detected in these electrodes. The greatest abundance and diversity of bacteria observed in cathode electrodes could have been likely due to their initial inoculation with mud pond. On average, the reactor evidencing the greatest electrochemical performance (S. hybridum) was the one showing the highest diversity indices among cathode electrodes. On the other hand, this reactor did not necessarily have the highest bacterial diversity among anode electrodes (Table 4). Similarly, the lowest diversity indices in cathode electrodes were achieved in the reactor with the worst performance (S. spurium). The Principal Coordinate Analysis (PCoA) of unweighted UniFrac distances showed differences between anodic and cathodic microbial communities (Fig. 6). The first axis explains 45.9% of the difference on the microbial community and clearly separates out communities from the two electrodes. It is expected that different microbial communities will be found as they perform different functions catalyzing distinct chemical reactions. The second axis explains 8.3% of the variation in the data and positions communities of the S. spurium reactor apart from those of the S. hybridum and S. rupestre reactors. This might suggest the development of a community in the whole reactor that could have resulted in the poor performance of it. The separation among cathode communities could be likely due to the presence of different Sedum species, as they were all initially inoculated with the same mud. The use of different Sedum species can cause variations in the composition of the exudates (Pinton et al., 2007), and thus, it would be expected that the communities developed in the electrodes of each Sedum type differ in some degree.
Fig. 6. Beta diversity of the microbial community between the samples based on the PCoA implemented in QIIME. Squares represent the anode electrode samples (A1, A2), and the circle represents the cathode electrode samples (C1, C2) of the three Sedum species: S. rupestre (P2, green), S. hybridum (P4, red) and S. spurium (P5, blue). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
4. Conclusion This paper demonstrates that Sedum species are feasible to use in the PMFCs. Although current magnitudes achieved with the Sedum PMFCs were lower than those reported for flooded systems, power generation indicates that this technology could be implemented in semi-arid ecosystems. S. hybridum showed the best performance among tested species with power density six times higher than S. rupestre. Pyrosequencing showed differences among anodes of the three reactors analyzed. The bacterial family Micrococcaceae was the most abundant in the anode of S. hybridum unlike the other anodes analyzed. This result opens the possibility of future research to particular groups of Sedum-associated bacteria with potential electrochemical activity. Additionally, community analysis suggests that a potential implementation of Sedum PMFCs in natural or engineered environments could not affect the microbial composition and nutrients fluxes in the soil. Finally, the observed relationship between the current generated by the PMFCs and the water content in the soil could help to expand this technology for monitoring irrigation in semiarid environments, as a result of an efficient and 209
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Buisman, C.J.N., Potting, J., 2013a. Electricity production with living plants on a green roof: environmental performance of the plant-microbial fuel cell. Biofuels. Bioprod. Biorefin. 7, 52–64. Helder, M., Strik, D.P., Timmers, R.A., Raes, S.M., Hamelers, H.V., Buisman, C.J., 2013b. Resilience of roof-top plant-microbial fuel cells during dutch winter. Biomass Bioenergy 51, 1–7. Janssen, P.H., 2006. Identifying the dominant soil bacterial taxa in libraries of 16S rRNA and 16S rRNA genes. Appl. Environ. Microbiol. 72, 1719–1728. Justin, S.H.F.W., Armstrong, W., 1991. Evidence for the involvement of ethene in aerenchyma formation in adventitious roots of rice (Oryza sativa L.). New Phytol. 118, 49–62. Kaku, N., Yonezawa, N., Kodama, Y., Watanabe, K., 2008. Plant/microbe cooperation for electricity generation in a rice paddy field. Appl. Microbiol. Biotechnol. 79, 43–49. Kawaichi, S., Ito, N., Kamikawa, R., Sugawara, T., Yoshida, T., Sako, Y., 2013. Ardenticatena maritima gen. nov., sp. nov., a ferric iron- and nitrate-reducing bacterium of the phylum ‘Chloroflexi’ isolated from an iron-rich coastal hydrothermal field, and description of Ardenticatenia classis nov. Int. J. Syst. Evol. Microbiol. 63, 2992–3002. Lauber, C.L., Hamady, M., Knight, R., Fierer, N., 2009. Pyrosequencing-based assessment of soil pH as a predictor of soil bacterial community structure at the continental scale. Appl. Environ. Microbiol. 75, 5111–5120. Lehmann, S., 2014. Low carbon districts: mitigating the urban heat island with green roof infrastructure. City Cult. Soc. 5, 1–8. Li, X., Wang, X., Zhao, Q., Wan, L., Li, Y., Zhou, Q., 2016. Carbon fiber enhanced bioelectricity generation in soil microbial fuel cells. Biosens. Bioelectron. 85, 135–141. Logan, B., 2010. Scaling up microbial fuel cells and other bioelectrochemical systems. Appl. Microbiol. Biotechnol. 85, 1665–1671. Lu, J., Yuan, J.-g., Yang, J.-z., Chen, A.-k., Yang, Z.-y., 2015a. Effect of substrate depth on initial growth and drought tolerance of Sedum lineare in extensive green roof system. Ecol. Eng. 74, 408–414. Lu, L., Xing, D., Ren, Z.J., 2015b. Microbial community structure accompanied with electricity production in a constructed wetland plant microbial fuel cell. Bioresour. Technol. 195, 115–121. Luo, J., Li, M., Zhou, M., Hu, Y., 2015.. Characterization of a novel strain phylogenetically related to Kocuria rhizophila and its chemical modification to improve performance of microbial fuel cells. Biosens. Bioelectron. 69, 113–120. Maier, R.M., Pepper, I.L., 2009. In: Gerba, R.M.M.L.P.P. (Ed.), Chapter 4 − Earth Environments. Environmental Microbiology, second edition. Academic Press, San Diego. Maricle, B.R., Lee, R.W., 2002. Aerenchyma development and oxygen transport in the estuarine cordgrasses Spartina alterniflora and S. anglica. Aquat. Bot. 74, 109–120. McDonald, D., Price, M.N., Goodrich, J., Nawrocki, E.P., DeSantis, T.Z., Probst, A., Andersen, G.L., Knight, R., Hugenholtz, P., 2012. An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea. ISME J. 6, 610–618. Pinton, R., Varanini, Z., Nannipieri, P., 2007. The Rhizosphere: Biochemistry and Organic Substances at the Soil-Plant Interface, second edition. CRC Press. Quince, C., Lanzen, A., Davenport, R., Turnbaugh, P., 2011. Removing noise from pyrosequenced amplicons. BMC Bioinf. 12, 38. Reyes, R., Bustamante, W., Gironás, J., Pastén, P.A., Rojas, V., Suárez, F., Vera, S., Victorero, F., Bonilla, C.A., 2016. Effect of substrate depth and roof layers on green roof temperature and water requirements in a semi-arid climate. Ecol. Eng. 97, 624–632. Strik, D.P.B.T.B., Hamelers, H.V.M., Snel, J.F.H., Buisman, C.J.N., 2008. Green electricity production with living plants and bacteria in a fuel cell. Int. J. Energy Res. 32, 870–876. Subbarao, G.V., Ae, N., Otani, T., 1997. Genotypic variation in iron-, and aluminumphosphate solubilizing activity of pigeonpea root exudates under P deficient conditions. Soil Sci. Plant Nutr. 43, 295–305. Timmers, R.A., Rothballer, M., Strik, D.P., Engel, M., Schulz, S., Schloter, M., Hartmann, A., Hamelers, B., Buisman, C.J.N., 2012a. Microbial community structure elucidates performance of Glyceria maxima plant microbial fuel cell. Appl. Microbiol. Biotechnol. 94, 537–548. Timmers, R.A., Strik, D.P.B.T.B., Hamelers, H.V.M., Buisman, C.J.N., 2012b. Characterization of the internal resistance of a plant microbial fuel cell. Electrochim. Acta 72, 165–171. Timmers, R.A., Strik, D.P.B.T.B., Hamelers, H.V.M., Buisman, C.J.N., 2013. Electricity generation by a novel design tubular plant microbial fuel cell. Biomass Bioenergy 51, 60–67. Wang, X., Cai, Z., Zhou, Q., Zhang, Z., Chen, C., 2012. Bioelectrochemical stimulation of petroleum hydrocarbon degradation in saline soil using U-tube microbial fuel cells. Biotechnol. Bioeng. 109, 426–433. Zhang, G., Zhang, H., Ma, Y., Yuan, G., Yang, F., Zhang, R., 2014. Membrane filtration biocathode microbial fuel cell for nitrogen removal and electricity generation. Enzyme Microb. Technol. 60, 56–63.
sustainable water management. Acknowledgments This work is part of the microbial fuel cells project in CEDEUS (Centro de Desarrollo Urbano Sustentable), in which we want to acknowledge Dr. Sergio Vera’s team for their help and for the access to the Laboratorio de Infraestructura Vegetal (LIVE). This research was funded by the CEDEUS center CONICYT/FONDAP/15110020. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ecoleng.2017.08.017. References Ap Rees, T., Wilson, P.M., 1984. Effects of Reduced Supply of Oxygen on the Metabolism of Roots of Glyceria maxima and Pisum sativum. Z. Pflanzenphysiol. 114, 493–503. Cao, X., Huang, X., Liang, P., Xiao, K., Zhou, Y., Zhang, X., Logan, B.E., 2009. A new method for water desalination using microbial desalination cells. Environ. Sci. Technol. 43, 7148–7152. Caporaso, J.G., Bittinger, K., Bushman, F.D., DeSantis, T.Z., Andersen, G.L., Knight, R., 2010a. PyNAST: a flexible tool for aligning sequences to a template alignment. Bioinformatics 26, 266–267. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., Fierer, N., Pena, A.G., Goodrich, J.K., Gordon, J.I., Huttley, G.A., Kelley, S.T., Knights, D., Koenig, J.E., Ley, R.E., Lozupone, C.A., McDonald, D., Muegge, B.D., Pirrung, M., Reeder, J., Sevinsky, J.R., Turnbaugh, P.J., Walters, W.A., Widmann, J., Yatsunenko, T., Zaneveld, J., Knight, R., 2010b. QIIME allows analysis of highthroughput community sequencing data. Nat. Meth. 7, 335–336. Chae, K.-J., Choi, M.-J., Lee, J.-W., Kim, K.-Y., Kim, I.S., 2009. Effect of different substrates on the performance, bacterial diversity, and bacterial viability in microbial fuel cells. Bioresour. Technol. 100, 3518–3525. Chen, Z., Huang, Y.-c., Liang, J.-h., Zhao, F., Zhu, Y.-g., 2012. A novel sediment microbial fuel cell with a biocathode in the rice rhizosphere. Bioresour. Technol. 108, 55–59. Chiranjeevi, P., Mohanakrishna, G., Mohan, S.V., 2012. Rhizosphere mediated electrogenesis with the function of anode placement for harnessing bioenergy through CO2 sequestration. Bioresour. Technol. 124, 364–370. De Schamphelaire, L., Cabezas, A., Marzorati, M., Friedrich, M.W., Boon, N., Verstraete, W., 2010. Microbial community analysis of anodes from sediment microbial fuel cells powered by rhizodeposits of living rice plants. Appl. Environ. Microbiol. 76, 2002–2008. DeSantis, T.Z., Hugenholtz, P., Keller, K., Brodie, E.L., Larsen, N., Piceno, Y.M., Phan, R., Andersen, G.L., 2006. NAST: a multiple sequence alignment server for comparative analysis of 16S rRNA genes. Nucleic Acids Res. 34, W394–W399. Domínguez-Garay, A., Berná, A., Ortiz-Bernád, I., Esteve-Núñez, A., 2013. Silica colloid formation enhances performance of sediment microbial fuel cells in a low conductivity soil. Environ. Sci. Technol. 47, 2117–2122. Donovan, C., Dewan, A., Heo, D., Beyenal, H., 2008. Batteryless, wireless sensor powered by a sediment microbial fuel cell. Environ. Sci. Technol. 42, 8591–8596. Edgar, R.C., Haas, B.J., Clemente, J.C., Quince, C., Knight, R., 2011. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27, 2194–2200. Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461. Farrell, C., Mitchell, R.E., Szota, C., Rayner, J.P., Williams, N.S.G., 2012. Green roofs for hot and dry climates: interacting effects of plant water use: succulence and substrate. Ecol. Eng. 49, 270–276. Fierer, N., Leff, J.W., Adams, B.J., Nielsen, U.N., Bates, S.T., Lauber, C.L., Owens, S., Gilbert, J.A., Wall, D.H., Caporaso, J.G., 2012. Cross-biome metagenomic analyses of soil microbial communities and their functional attributes. Proc. Natl. Acad. Sci. U. S. A. 109, 21390–21395. Frank, K.L., Rogers, D.R., Olins, H.C., Vidoudez, C., Girguis, P.R., 2013. Characterizing the distribution and rates of microbial sulfate reduction at Middle Valley hydrothermal vents. ISME J. 7, 1391–1401. Helder, M., Strik, D.P., Hamelers, H.V., Kuhn, A.J., Blok, C., Buisman, C.J., 2010. Concurrent bio-electricity and biomass production in three Plant-Microbial Fuel Cells using Spartina anglica, Arundinella anomala and Arundo donax. Bioresour. Technol. 101, 3541–3547. Helder, M., Chen, W.-S., van der Harst, E.J.M., Strik, D.P.B.T.B., Hamelers, H.V.M.,
210
Contents lists available at ScienceDirect
Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng
Evaluation of Sedum as driver for plant microbial fuel cells in a semi-arid green roof ecosystem Natalia F. Tapiaa,b, Claudia Rojasc, Carlos A. Bonillaa,b, Ignacio T. Vargasa,b, a b c
MARK
⁎
Centro de Desarrollo Urbano Sustentable (CEDEUS), Chile Department of Hydraulic and Environmental Engineering, Pontificia Universidad Católica de Chile, 7820436 Santiago, Chile Institute of Agricultural Sciences, University of O’Higgins, Rancagua, Chile
A R T I C L E I N F O
A B S T R A C T
Keywords: Green roofs Plant microbial fuel cells Pyrosequencing Succulent plant
A plant microbial fuel cell (PMFC) is a novel and sustainable bioelectrochemical technology that takes advantage of the organic compounds released through the plant roots (exudates) to produce power via electrochemically active bacteria. This technology has been implemented in flooded ecosystem; however, their use in arid or semiarid environments has been less explored. Thus, the use of plant adapted to conditions of less water availability could expand the application of PMFCs. In this study, seven Sedum species (succulent plants) were evaluated for their ability to generate energy under non-saturated conditions. Electrochemical results confirmed power generation by the Sedum species tested, with reactors using Sedum hybridum achieving the highest power density (92 μW m−2), followed by reactors tested with S. rupestre (15.5 μW m−2). The lowest performance was observed in systems using S. spurium (< 1 μW m−2). Total organic carbon from root exudates did not have a direct effect on PMFCs performance. Pyrosequencing analysis revealed higher abundance of bacteria in the family Micrococcaceae in anodic biofilms of S. hybridum (13.47 ± 2.69%) reactors than in biofilms of S. rupestre (2.28 ± 0.52%), and S. spurium (2.29 ± 1.94%) PMFCs. Although the current magnitudes observed in our study were lower than those reported for flooded PMFCs, a positive relationship between water content of plant growth media and power outputs was observed. After irrigation, current and water gradually decrease over time with a similar rate, showing a correlation coefficient between both variables of 0.95 ± 0.01. Moreover, it encourages further testing of PMCFs as potential indicators of soil water content in semiarid green roofs where a better water-use efficiency is needed.
1. Introduction Microbial fuel cells (MFCs) are bioelectrochemical systems that use electrochemically active bacteria (EAB) to catalyze reactions at one or both electrodes (the anode and cathode). Initially, MFCs were conceived to transform chemical energy stored in wastewater into electricity. However, the progress achieved for this technology in recent years has expanded its potential applications (Cao et al., 2009; Logan, 2010). One example of it is the sediment- or soil-based microbial fuel cell (SMFC), which uses organic matter stored in sediments as a source of energy for EAB. This technology has been considered an innovative opportunity to provide power to electronic devices in remote locations, where it is difficult and expensive to replace batteries (Donovan et al., 2008). Nevertheless, the availability of organic matter represents a limitation for this technology during long-term operation. A sustainable alternative to supply organic matter to SMFC devices is the use of plants
that can provide the fuel for these systems throughout photosynthesis. This type of technology has been named plant microbial fuel cells (PMFCs). In a PMFC, the plant generates organic compounds released through it roots (exudates), which are later used by EAB for power production (Helder et al., 2013a). The implementation of PMFCs has been mostly tested in water-saturated environments with plant adapted to this condition (Table 1). Among them, the main species tested in PMFCs since this technology emerged in 2008 have been: Glyceria maxima (Strik et al., 2008; Timmers et al., 2012a; Timmers et al., 2013); Spartina anglica (Helder et al., 2010; Helder et al., 2013b) and Oryza sativa (Chen et al., 2012; De Schamphelaire et al., 2010; Kaku et al., 2008). All of these species have in common their capacity to grow under flooded conditions as a result of their adapted tissue called aerenchyma, which enables them to carry oxygen to their roots under such environments (Ap Rees and Wilson, 1984; Justin and Armstrong, 1991; Maricle and Lee, 2002).
⁎ Corresponding author at: Department of Hydraulic and Environmental Engineering, Pontificia Universidad Católica de Chile, Av. Vicuña Mackenna 4860, 7820436 Macul, Santiago, Chile. E-mail address: [email protected] (I.T. Vargas).
http://dx.doi.org/10.1016/j.ecoleng.2017.08.017 Received 8 November 2016; Received in revised form 14 July 2017; Accepted 15 August 2017 0925-8574/ © 2017 Elsevier B.V. All rights reserved.
Ecological Engineering 108 (2017) 203–210
Saturation
Saturation Saturation Saturation Not saturation
120
112 65 90 360
10 – – – – 4.52 105 – – – – 2.71
22 163 18 0.09
– 100 – 6000
O2, Ferricyanide O2 O2 O2
Helder et al., 2010 Chiranjeevi et al., 2012 Lu et al., 2015a,b This study
The water content in PMFCs has been reported as a key factor for power generation as this condition allows to maintain the anode electrode under anoxic conditions and facilitate ion transport to the cathode electrode (Chiranjeevi et al., 2012). Indeed, significant drop in voltage (from 634 mV to 0 mV) and current density (200 mA/m2 to 125 mA/ m2) have been previously shown in PMFCs after irrigation ceased (Chiranjeevi et al., 2012; Li et al., 2016). Thus, it is not surprising that PMFCs are currently mainly used in environments under saturated conditions like wetlands, with plant species adapted to such environments (Table 1). However, PMFCs can also be implemented in engineered environments such green roofs, representing a benefit over other sources of energy generated from biomass (Strik et al., 2008). In spite of the potential to extend this technology to other environments, performance of PMFCs in green roofs operated under conditions different than saturation is virtually unknown. In conditions different than humid ecosystem of the northern hemisphere, where green roofs originally started, their benefits including reducing runoff, improving air quality and increasing biodiversity are still highly valued (Lu et al., 2015a). However, the implementation of green roofs in arid and semiarid zones requires alternative plant species to support these systems. Under such environmental conditions, succulent plants, capable of retaining water are among the vegetation most frequently used. Among them, some species in the Sedum genus have been previously used due to their low water requirements and tolerance to droughts. They are able to resist these conditions due to their adaptation to a special photosynthetic pathway, known as Crassulacean Acid Metabolism (CAM), which allow them to keep their stomata closed during the day (avoiding water evaporation) and open them at night to fix carbon dioxide (Farrell et al., 2012). Soil conductivity and water content could limit the PMFCs performance in arid and semi-arid environments (Domínguez-Garay et al., 2013; Wang et al., 2012). An increment in soil water content could result in a reduction of the internal resistance of the system, allowing an increment in the current generated by the PMFC. In this type of climates, the water is a critical resource that should be efficiently used, particularly in green infrastructure used as a driver to transform existing cities towards a next generation of net-zero (water and energy) urban settlements (Lehmann, 2014). The use of commercial moisture sensors allows a more efficient and sustainable use of water. However, this type of sensors is still relatively expensive for domestic and commercial applications, particularly in developing countries. Hence, the study of the influence of variable water contents and the quality of soil or substrate used to support plant growth are needed and expand the application of PMFCs, not only in search of a new source of energy, but also for the development of a low-cost water content biosensor. In this study, we assessed the performance of PMFCs supported by Sedum plant species under unsaturated conditions and thus the potential to implement this technology in green roof environments of semiarid regions. We hypothesized that these plants would provide organic matter through their roots to be used by EABs in electrodes of constructed PMFCs. For this purpose, seven Sedum species were tested in terms of their capability to generate electricity and organic exudates. Three representative PMFCs were also evaluated for the microbial community developed on anode and cathode electrodes.
Rhizodeposits Rhizodeposits and soil Rhizodeposits Rhizodeposits and soil A.anomala P. setaceum C. indica S. hybridum
Natronocella Beijerinckiaceae Rhizobiales Bacteria Bacteria Geobacteraceae Anaerolineaceae Micrococcaceae Rhizodeposits and soil O. sativa
Saturation Saturation Saturation Rhizodeposits Rhizodeposits Rhizodeposits S. anglica S. anglica O. sativa
78 154 134
– 52 –
6
156
O2
Timmers et al., 2010 Helder et al., 2010 De Schampeliare, 2008, 2010 Kaku et al., 2008 O2 O2, Ferricyanide Ferricyanide 1800 – – 22 21 21 – – – 141 – 44
79 222 33
525 900 67 80 4 – 153 – Saturation Saturation 67 – Rhizodeposits Rhizodeposits G. maxima G. maxima
Bacteria Geobacteraceae Ruminococcaceae Comamonadaceae Bacteria Bacteria Desulfobulbus Geobacteraceae Archaea
32 164
Max Avg Avg
Max
Power density (mW m−2) Current density (mA/ m2) Water condition Operation time (days) Microbial community Electron donor Plant
Table 1 Performance PMFCs reported in literature and results of Sedum species with the best performance in this study are shown.
Internal resistance (ohm)
O2 O2
Electron acceptor
Ref.
Strik et al., 2008 Timmers et al., 2012a,b
N.F. Tapia et al.
2. Materials and methods 2.1. Screening of sedum species in plant microbial fuel cells A preliminary screening test was first carried out to evaluate the potential of using Sedum species in PMFCs. Eight reactors were run in this experiment, seven contained one individual seedling S. album, S. hybridum, S. kamtschaticum, S. reflexum, S. rupestre, S. sexangulare or S. spurium plant species and one was left unplanted to represent a control pot (Fig. 1). These species additionally were used in a pilot green roof constructed at the San Joaquín Campus, Pontificia Universidad Católica 204
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Fig. 1. A: The PMFC Reactors with each one of the seven Sedum species used in the experiment: (1) S. kamtschaticum, (2) S. rupestre, (3) S. album, (4) S. hybridum, (5) S. spurium, (6) S. sexangulare, (7) S. reflexum, and (8) control (no plant) B. Diagram of the PMFC reactor configuration during operation.
conductivity sensor GS3 (Decagon®, Pullman, USA).
de Chile (PUC), Santiago, Chile (Reyes et al., 2016). The PMFC reactors consisted of 450 mL plastic pots, with holes at the bottom to allow free drainage. Each pot was filled with 400 mL of plant growth substrate collected in May 2013 from a green roof under construction (before plants establishment) at the same Campus (S1). The substrate consisted of a mineral matrix of sand, silt, and clay amended with organic matter (Table S.1). Electrodes made of a circular graphite fiber felt of 92 mm in diameter and 6 mm thick (MudWatt, California, USA) were initially installed; cathode electrodes were buried at a depth of 7 cm and anode electrodes were placed on the surface of the substrate. The buried electrodes were inoculated with mud, from a pond used as a water supply for irrigation located at the same University’s campus, once placed at the bottom of the pots. The reactors were operated during an initial startup period of 120 days for system adjustment (from December 2013 to April 2014). During this time, plant irrigation was performed weekly by adding approximately 150 mL of tap water in the upper part of the reactor until the water drained for the bottom pot. From day 60–110 the irrigation system was changed to a continuous irrigation mode in order to use irrigation conditions similar to green roof (rainfall events and sprinkler irrigation) and evaluate the effect in plant performance according to irrigation strategy. After 120 days, the upper graphite fiber felt electrodes (anodes) were replaced with activated carbon granules (3 mm in diameter; 35 cm3 volume) and a graphite rod (as connection), for a better adhesion of electrode to the substrate during the periods without feeding water. Titanium wires (23 cm in length; 0.5 mm in diameter) were used to connect the two electrodes to an external resistor of 1 kΩ (Fig. 1). The reactors were maintained under controlled temperature conditions and continuously monitored over a period of 360 days (from December 2013 to December 2014). The average room temperature during this period was 20 ± 2 °C (except between days 260 and 350 when reactor was exposed to 32 ± 2 °C). Light periods extended for 14 h per day, with an average radiation of 23 ± 2 W m−2, and they were measured every 10 min by a solar radiation sensor (Decagon®, Pullman, USA). During the evaluation period (day 120 and beyond) irrigation was performed weekly by adding 100 mL of tap water or until saturation was reached. Cell voltage was measured every 10 min with a data acquisition system (2700; Keithley, USA). Power and current densities were normalized by the area of the cathode electrode (0.0067 m2), where the cathodic reaction occurred. The polarization curve and the power density curves were computed by varying the external resistance from 10 Ω to 100 kΩ every 30 min or until the system reached a stable voltage. Substrate water content, temperature and electrical conductivity were measured by a temperature, humidity and ground
2.2. Evaluation of PMFCs performance using sedum hybridum After the preliminary test was conducted, new plants for the species used in the reactors showing the best performance were obtained for further characterization. New PMFCs reactors consisted of three pots with Sedum hybridum and one pot with no plant as control. The plants and growing media were obtained from a second green roof established at the San Joaquín Campus, which was already under operation using the same Sedum species evaluated in the screening test (Reyes et al., 2016). Electrodes consisted of cathodes made of graphite fiber felt and anodes of activated carbon granules (as described above). Anodes were inoculated with mud obtained from the same pond the inoculum for the screening test was sampled. The PMFCs were maintained under the same conditions that screening test, with an average room temperature of 20 ± 2 °C over period of 215 days (from December 2014 to July 2015) and light period of 14 h per day. Irrigation was performed weekly by adding 150 mL of tap water or until water drained by bottom pot. Cell voltage was measured same that previous experiment, every 10 min, and current densities also were normalized by the area of the cathode. 2.3. Plant root exudate analysis A separated analysis was conducted on August 2015 to evaluate root exudates. For root exudate measurements, new plant seedlings (n = 3) of similar size and of each of the seven species tested in the screening test, were obtained from the same plant nursery. Plants were weighed after substrate particles were gently removed by manually shaking their roots. To further remove substrate particles from roots, they were rinsed with tap water and dipped into 30 mg L−1 of chloramphenicol for 2 h to minimize microbial growth (Subbarao et al., 1997). Roots were then washed again with tap water and sterile deionized water to remove residues of the antibiotic used. To collect exudates, plant roots were submerged in sterile centrifuge tubes containing 50 mL of deionized water and then aseptically stored at 20 °C for 48 h. Each tube was covered with aluminum foil to avoid light exposure. After this time, plants were removed from the tubes and exudate solutions were collected. Each solution was acidified to pH ∼ 2.4 using 6 M HCl and maintained at 4 °C until total organic carbon (TOC) was measured using a Shimadzu TOC-L CPH equipment (Shimadzu, Japan). The TOC contents of each solution were normalized by the measured weight of fresh plant biomass. 205
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Alignment was performed using PyNAST (Caporaso et al., 2010a) against the Greengenes core set (http://greengenes.lbl.gov/), and then filtered to remove gap positions.
2.4. DNA extraction and pyrosequencing The electrode samples for microbial community analysis were taken from the PMFCs of the screening test, with the highest, the second highest, and the lowest power density generation after 360 days of system operation. DNA extraction was performed using the Power Soil® DNA isolation kit (MoBio Laboratories, Inc., Carlsbad, CA) following the manufacturer’s instructions. Each of the selected PMFCs was dismantled under sterile conditions to obtain the electrode samples. The upper electrodes (anodes) were placed on sterile petri dishes from where pieces of graphite rod and granules were randomly selected to obtain duplicate samples of approximately 0.27 g for each of the three PMFCs examined. The cathode samples were aseptically sectioned to obtain eight small pieces (∼0.1 g) of graphite fiber felt per reactor. These subsamples were slightly moistened using sterile deionized water, previous to DNA extraction to avoid excessive absorption of extraction solutions. DNA samples extracted from each of the small graphite felt fiber subsamples were randomly selected and poled together to obtain two composite DNA samples per cathode electrode. The DNA quality and concentration were determined using a NanoDrop 2000c Spectrophotometer (Thermo Fisher Scientific Inc.). All samples were stored at −80 °C prior to the pyrosequencing analysis. The pyrosequencing analysis was performed using a 454 Genome Sequencer FLX Titanium System (Roche Diagnostics, Indianapolis, IN, USA) at the Research and Testing Laboratory (Lubbock, Texas. USA). The amplicon library was prepared for each sample using a bacteriaspecific primer set, 28F (5′-GAGTTTGATCNTGGCTCAG −3′) and 519R (5′- GTNTTACNGCGGCKGCTG −3′) flanking the regions V1-V3 of the 16S rRNA gene (Frank et al., 2013). The PCR reactions were performed in 25 μL reaction volumes with the Qiagen HotStar Taq master mix (Qiagen Inc., Valencia, California), 1 μL (5 μM) of each forward and reverse primer and 1 μL of the DNA extract. The reaction conditions consisted of an initial denaturation step at 95 °C for 5 min followed by 35 cycles, each consisting of denaturation at 94 °C for 30 s, primer annealing at 54 °C for 40 s, and an extension at 72 °C for 1 min with a final extension at 72 °C for 10 min and a 4 °C hold. After the PCR amplification, the products were visualized with eGels (Like Technologies, Grand Island, New York). The amplicons were then pooled equimolar, and each pool was cleaned and selected for sizes using Agencourt AMPure XP (BeckmanCoulter, Indianapolis, Indiana) as directed by the Roche 454 protocols (454 Life Sciences, Branford, Connecticut). The selected pools were then quantified and diluted for use in the emPCR reactions. These reactions were performed and subsequently enriched following the established manufacturer’s protocols (454 Life Sciences).
2.6. Statistical analysis Alpha diversity analysis included the total number of observed OTUs, Good’s sample coverage, Shannon and inverse Simpson diversity indices obtained from normalized data after ten iterations performed per index. Beta diversity, based on phylogenetic information, was assessed by a Principal Coordinate analysis (PCoA) on unweighted UniFrac distances. Differences among the Sedum species were evaluated comparing the alpha diversity and TOC analysis. These were evaluated with a one-way Analysis of Variance (ANOVA) and Tukey’s test using Microsoft Excel data analysis. P-values less than 0.05 were considered significant. 3. Results and discussion 3.1. PMFCs performance in the screening test Over the period of 150 days of system operation the seven PMFCs reactors tested (screening test) showed power generation, as opposed to the control reactor that ceased to produce recordable values (S2). This could have been due to the consumption by microorganisms of initial organic matter available in the substrate (2.85%). Hence, for the subsequent operation, the power obtained from the PMFC reactors was assumed to be mainly due to the chemical energy provided by the plant as exudates. Our results showed no dependency of current production on light periods, thus the circadian oscillation reported in previous studies (Helder et al., 2013b; Kaku et al., 2008) using S. anglica y O. sativa was not observed for the Sedum PMFCs tested in this study under controlled laboratory conditions. Power density curves performed on day 345 of operation (Fig. 2) indicated that S. hybridum displayed the highest performance at the final stage of the experiment, reaching a maximum of 92 μW m−2. S. rupestre, S. sexangulare and S. album followed with maximum power densities of 15.5 μW m−2; 8.4 μW m−2; and 2.4 μW m−2, respectively. The lowest performances were achieved by S. kamtschaticum, S. spurium
2.5. Processing and analysis of pyrosequence data The sequences obtained in this study have been deposited into the NCBI Sequence Read Archive (SRA) with the accession No. SRP062553. They were processed and analyzed using the Quantitative Insights Into Microbial Ecology (QIIME) program (Caporaso et al., 2010b) through the MacQIIME version 1.8.0. Sequences were denoised by removing the reads that did not match to barcodes or primers, had homopolymers longer than 6 bp. and quality scores lower than 25 (Quince et al., 2011). Only those filtered sequences between 200 bp and 1000 bp in length were included for further analysis. The sequences were then assigned to operational taxonomic units (OTUs) at 97% similarity using the de novo OTU picking protocol with usearch to additionally identify and remove singletons and chimeras (Edgar, 2010; Edgar et al., 2011). A representative sequence for each OTU was selected for downstream analysis. Taxonomic assignment was done using the uclust method and the Greengenes taxonomy database (DeSantis et al., 2006; McDonald et al., 2012) released on August 2013 (13.8 v) at a 90% confidence threshold. The unassigned sequences, referring to those sequences that could not be classified at the domain level, and sequences classified as chloroplast and mitochondria, were removed from the dataset.
Fig. 2. Power curves conducted to seven Sedum species tested in this study after 345 days of operation.
206
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Fig. 3. P-MFCs performance over time. Evaluation with distinct irrigation regimes: drip irrigation (A) (days 99–110) and weekly irrigation (B) (days 235–246).
and S. reflexum, all of which had values lower than 1 μW m−2. This difference between species is due to the performance at the final stage of PMFC reactors, where S. hybridum shows a higher current density in relation to S. rupestre (Figure S.2). As it was expected, the low water content in the soil affects the mobility for cation transport, increasing the internal resistance of the system (Chiranjeevi et al., 2012). Internal resistances were estimated from polarization curves, reaching values of 15 kΩ for S. rupestre and S. sexangulare, and 5.34 kΩ for S. hybridum. These values are higher compared to previous PMFC studies (Table 1). The application of drip and weekly irrigation resulted in differences in current generation (Fig. 3). While under drip irrigation relatively constant currents were observed, under weekly irrigation there was a sudden increase of current, observed after each irrigation event. After irrigation, soil moisture measured in S. hybridum and S. rupestre PMFC, reached up 40% v/v, decreasing at similar rate that the produced current, until values around 5% v/v corresponding to negligible current values. The correlation analysis between both variables, after the maximum current was reached, showed r values of 0.96 and 0.94 for S. rupestre and S. hybridum, respectively (Figure S.3). This suggests a close relationship between power generation and soil moisture content, with higher power obtained at higher soil moisture. This result is in agreement with data presented by Chiranjeevi et al. showing a significantly drop in the open circuit voltage when water supplementation was stopped, following by an increase when irrigation was restored (Chiranjeevi et al., 2012).
3.3. Analysis of root exudates The total organic carbon analysis of the root exudates obtained from the seven Sedum species tested in this study showed relatively low concentrations, ranging from 1.87 to 16.59 mg L−1 per gram of fresh plant. Similar values were observed for almost all species tested; only S. sexangulare had a significantly higher concentration (9.99 ± 5.75 mg L−1 per gram of fresh plant) according to Tukey’s test (p < 0.05) (Table 2). The similarity observed in carbon content among root exudates suggests that this parameter would not have a direct positive effect on PMFC performance, as the reactor resulting in the greatest current production (S. hybridum) did not necessarily achieved the highest TOC content (4.29 ± 1.29 mg L−1 per gram of fresh plant). It is likely that more than the total amount of carbon released by plant roots, the composition of these exudates could be more important for power generation as it has been previously reported that EAB have different affinity for different organic compounds (Chae et al., 2009). We acknowledge though that exudation process can be influenced by many factors, such as collection time, damage inflicted on the root, collection medium and temperature, among others (Pinton et al., 2007), which might have influenced our results. In addition, due to the long time of collection of exudates, some easily degradable organic compounds could have been decomposed, causing the low values detected in our study. 3.4. Analysis of microbial community According to Pinton et al. (2007) the composition of exudates depends of each plant species. Thus, Sedum species tested in this study could exudate different organic compounds, generating differences in the community composition and the subsequent PMFC performance. Since, no current was observed on the control reactor, electrodes of three Sedum species (representatives of the highest, second highest and the lowest power density) were analyzed, showing slight differences in microbial composition and diversity.
3.2. PMFCs performance using S. hybridum To assess reproducibility of the observed results obtained by the screening test, triplicated P-MFC using S. hybridum were run in the laboratory. Reactors operated under weekly irrigation mode show maximum current densities (after of 150 days of the start-up) of 2.36, 1.58 and 1.54 mA/m2, comparable with the average current density of 2.71 mA/m2 obtained with the screening test and follow the same water-content dependence observed in the first set of experiments. Interestingly, after about 150 days of operation the control reactor also stopped current production supporting the hypothesis that power obtained is provided by the plant as exudates (Fig. 4). Even though the power densities obtained in this study are considerable lower than those reported for PMFCs operated under water saturation (Timmers et al., 2012b), the observed relationship between the water content in the substrate and current generated by the Sedum PMFC reactors, could be used as water content indicator expanding the potential applications of these systems. Moreover, it encourages further testing of PMCFs as potential sensor for soil water content in actual semi-arid green roofs. This opportunity could impact the sustainability of urban settlements by including low-cost technology for monitoring the soil water content, leading to a better management of water, energy and land.
3.4.1. Bacterial community composition A total of 12 phyla were detected in anode samples (upper electrodes) and 13 in cathode samples (lower electrodes) (Table S.2), with seven dominant groups identified in anodes (Fig. 5). Proteobacteria was the most abundant phylum in anode (44.8% to 67.1%) and cathode (29.2% to 59.9%) electrodes, followed by Actinobacteria (13.5% to 38.98% in anodes; and 7.26% to 15.38% in cathodes) and Chloroflexi (3.07% to 16.03% in anodes; and 3.47% to 15.58% in cathodes). These three phyla have been previously reported as the dominant bacterial groups in electrodes of PMFCs, along with Acidobacteria and Bacteroidetes, which were also detected in our study (Lu et al., 2015b). Proteobacteria in the family Comamonadaceae and the genus Paracoccus were among the most abundant members in anode electrodes (Table 3). Both Comamonadaceae and Paracoccus have a negative impact on power generation, in the presence of oxygen or nitrate, since 207
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Fig. 4. Triplicated P-MFC experiment using Sedum hybridum. The experiment shows the observed behavior for PMFCs of Fig. 3 (and S2) (A). After a starup period of 150 days, while no current was recorded for the control reactor (no plant), the three reactors with S. hybridum show increases in current after irrigation (B).
Table 2 Total Organic Carbon in the root exudate of 21 plants from the seven Sedum species.
Table 3 Relative abundances of the main bacterial taxonomic groups at genus or higher taxa levels in the samples of the anode (A1, A2) and cathode (C1,C2) electrodes of the three Sedum species: S. rupestre, S. hybridum and S. spurium.
Total Carbon Organic (mg L−1 per gr of plant) Specie
Plant 1
Plant 2
Plant 3
Mean ± SD
S. S. S. S. S. S. S.
5.74 1.96 6.03 1.87 6.31 4.79 3.12
3.93 4.42 16.59 2.19 5.86 7.54 2.13
3.23 2.84 7.36 3.55 3.96 5.56 2.57
4.29 3.07 9.99 2.54 5.38 5.96 2.61
hybridum rupestre sexangulare album spurium kamtschaticum reflexum
± ± ± ± ± ± ±
1.29 1.25 5.75 0.89 1.25 1.42 0.49
Relative Abundance (%) a a b a a a a
Phylum: Proteobacteria Family Comamonadaceae Genus Paracoccus Family Rhodobacteraceae Genus Pseudomonas Genus Rubellimicrobium Genus Rhodobacter Phylum: Actinobacteria Family Micrococcaceae Order Acidimicrobiales Order Actinomycetales Phylum Chloroflexi Genus Ardenscatena
Different letters indicate statistically significant differences (p < 0.05) according to Tukey’s test.
Phylum: Proteobacteria Family Ectothiorhodospiraceae Phylum: Acidobacteria Order DS-18 Phylum: Actinobacteria Order 0319-7L14 Phylum Nitrospirae Order Nitrospirales
S. hybridum
S. rupestre
S. spurium
A1
A2
A1
A2
A1
A2
10.42 9.31 3.34 1.65 1.69 5.35
4.70 6.11 3.46 7.52 1.23 2.65
12.48 12.16 5.32 1.89 1.52 1.90
19.99 9.33 4.89 3.30 1.77 0.61
5.68 10.23 6.15 6.28 5.18 1.98
3.49 10.42 6.38 3.46 4.95 1.22
15.37 3.34 6.06
11.57 5.34 4.65
2.65 5.76 1.16
1.91 6.32 1.02
0.92 2.16 0.39
3.67 2.29 2.20
2.32 C1
11.27 C2
7.27 C1
10.76 C2
14.45 C1
10.27 C2
0.03
19.87
16.31
0.02 8.40
5.37
8.58
4.35
6.98
9.93
2.58
4.74
3.01
5.74
0.65
1.21
4.04
5.29
0.95
5.16
1.25
0.47
they decrease the availability of compounds that could be used by EABs (Kawaichi et al., 2013; Timmers et al., 2012a; Zhang et al., 2014). In cathode electrodes, members in the family Ectothiorhodospiraceae were especially abundant in S. spurium reactors. However, for the other two cathodes (S. hybridum and S. rupestre), this family was not observed. It is worth noticing that one of the most commonly reported EABs, Geobacter spp., (De Schamphelaire et al., 2010; Timmers et al., 2012a) was not detected in our study (Table 3). Among the Actinobacteria phylum, members in the Micrococcaceae family were especially abundant in the anode of the S. hybridum. Among this group, the species Kocuria rhizophila has been reported as a bacterium capable of generating electricity from glucose oxidation (Luo et al., 2015). Thus, the possible presence of members of the Micrococcaceae family as K. rhizophila or unreported exoelectrogenic species could explain the better performance of the reactor with S. hyridum. In the phylum Chloroflexi, the most abundant group was the Ardenscatena genus, which has a negative impact in power generation in presence of oxygen. The implementation of Sedum plants in PMFCs did not seem to affect the typical microbial composition of soil systems. In these environments, the Proteobacteria, Chloroflexi and Actinobacteria, among other
Fig. 5. Relative abundances of bacterial phyla in: (A) anode samples (upper electrodes), and (B) cathode samples (lower electrodes) of S. hybridum, S. rupestre, and S. spurium species. The bacterial group others consisted of sequences classified at the bacterial domain level only (≤0.1%) and other phyla comprising less than 0.5% of classified sequences.
208
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Table 4 The alpha diversity in the upper (A1, A2) and lower (C1, C2) electrodes corresponding to S. rupestre (P2), S. hybridum (P4) and S. spurium (P5). The observed OTUs, the sample coverage and the Shannon and inverse Simpson diversity indices were calculated from the normalized datasets (2245 sequences per each sample). The bacterial OTUs were generated at 97% sequence similarity. Diversity indices
P2A1 P2A2 P4A1 P4A2 P5A1 P5A2 P2C1 P2C2 P4C1 P4C2 P5C1 P5C2
Number of total sequences
Number of observed OTUsa
Sampling coverage
Shannon
7427 4127 2245 4536 5191 6913 3253 3102 4998 5025 4645 3647
344.7 ± 8.78 296 ± 5.44 275 ± 0 282.9 ± 5.30 323.6 ± 5.32 312.5 ± 6.17 644.4 ± 6.60 668.8 ± 7.08 699.6 ± 9.42 740.2 ± 13.76 486.8 ± 6.56 606.8 ± 5.94
0.93 0.95 0.96 0.95 0.95 0.95 0.88 0.88 0.86 0.84 0.90 0.88
6.66 6.05 6.35 6.39 6.55 6.81 8.27 8.44 8.49 8.53 6.95 7.45
b c a a d ad ab ac bc abc bd cd
± ± ± ± ± ± ± ± ± ± ± ±
Inv-Simpson 0.04 0.05 9.36e-16 0.02 0.02 0.03 0.02 0.02 0.04 0.04 0.04 0.05
c d a a ad bc ab ac b b bd abc
36.51 ± 1.73 18.06 ± 1.1 36.34 ± 7.45e-15 41.30 ± 1.02 34.31 ± 0.89 52.48 ± 2.65 97.53 ± 5.04 141.73 ± 4.85 126.90 ± 8.86 127.62 ± 6.52 21.55 ± 1.15 28.34 ± 1.47
a b a a a bc ab ac c c b d
Mean ± Standard Deviation values with different letters in the same column indicate statistically significant differences (p < 0.05) according to Tukey’s test. a Total number of OTUs identified per sample.
phyla also identified in this study are typically reported as abundant and ubiquitous members of the microbial community (Fierer et al., 2012; Janssen, 2006; Lauber et al., 2009). Moreover, the presence of Actinomycetales, Rhizobiales, Pseudomonas and Rhodobacter representatives, which are groups known to play a role in organic matter decomposition or nutrient cycling in soils (Maier and Pepper, 2009), could indicate the lower disturbance of PMFCs on soil functioning. This phenomenon takes special relevance in the event of PMFCs implementation in green roofs that are already under operation, and where active soil microbial communities have been already established.
3.4.2. Diversity analysis The total number of curated sequences obtained per samples fluctuated from 2245 and 7427, with coverage ranging from 0.88 and 0.95 (Table 4). The highest bacterial diversity, as estimated by number of OTUs, the Shannon and Inv-Simpson indices (Table 4), was achieved in the cathode electrodes, which is consistent with the highest number of phyla detected in these electrodes. The greatest abundance and diversity of bacteria observed in cathode electrodes could have been likely due to their initial inoculation with mud pond. On average, the reactor evidencing the greatest electrochemical performance (S. hybridum) was the one showing the highest diversity indices among cathode electrodes. On the other hand, this reactor did not necessarily have the highest bacterial diversity among anode electrodes (Table 4). Similarly, the lowest diversity indices in cathode electrodes were achieved in the reactor with the worst performance (S. spurium). The Principal Coordinate Analysis (PCoA) of unweighted UniFrac distances showed differences between anodic and cathodic microbial communities (Fig. 6). The first axis explains 45.9% of the difference on the microbial community and clearly separates out communities from the two electrodes. It is expected that different microbial communities will be found as they perform different functions catalyzing distinct chemical reactions. The second axis explains 8.3% of the variation in the data and positions communities of the S. spurium reactor apart from those of the S. hybridum and S. rupestre reactors. This might suggest the development of a community in the whole reactor that could have resulted in the poor performance of it. The separation among cathode communities could be likely due to the presence of different Sedum species, as they were all initially inoculated with the same mud. The use of different Sedum species can cause variations in the composition of the exudates (Pinton et al., 2007), and thus, it would be expected that the communities developed in the electrodes of each Sedum type differ in some degree.
Fig. 6. Beta diversity of the microbial community between the samples based on the PCoA implemented in QIIME. Squares represent the anode electrode samples (A1, A2), and the circle represents the cathode electrode samples (C1, C2) of the three Sedum species: S. rupestre (P2, green), S. hybridum (P4, red) and S. spurium (P5, blue). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
4. Conclusion This paper demonstrates that Sedum species are feasible to use in the PMFCs. Although current magnitudes achieved with the Sedum PMFCs were lower than those reported for flooded systems, power generation indicates that this technology could be implemented in semi-arid ecosystems. S. hybridum showed the best performance among tested species with power density six times higher than S. rupestre. Pyrosequencing showed differences among anodes of the three reactors analyzed. The bacterial family Micrococcaceae was the most abundant in the anode of S. hybridum unlike the other anodes analyzed. This result opens the possibility of future research to particular groups of Sedum-associated bacteria with potential electrochemical activity. Additionally, community analysis suggests that a potential implementation of Sedum PMFCs in natural or engineered environments could not affect the microbial composition and nutrients fluxes in the soil. Finally, the observed relationship between the current generated by the PMFCs and the water content in the soil could help to expand this technology for monitoring irrigation in semiarid environments, as a result of an efficient and 209
Ecological Engineering 108 (2017) 203–210
N.F. Tapia et al.
Buisman, C.J.N., Potting, J., 2013a. Electricity production with living plants on a green roof: environmental performance of the plant-microbial fuel cell. Biofuels. Bioprod. Biorefin. 7, 52–64. Helder, M., Strik, D.P., Timmers, R.A., Raes, S.M., Hamelers, H.V., Buisman, C.J., 2013b. Resilience of roof-top plant-microbial fuel cells during dutch winter. Biomass Bioenergy 51, 1–7. Janssen, P.H., 2006. Identifying the dominant soil bacterial taxa in libraries of 16S rRNA and 16S rRNA genes. Appl. Environ. Microbiol. 72, 1719–1728. Justin, S.H.F.W., Armstrong, W., 1991. Evidence for the involvement of ethene in aerenchyma formation in adventitious roots of rice (Oryza sativa L.). New Phytol. 118, 49–62. Kaku, N., Yonezawa, N., Kodama, Y., Watanabe, K., 2008. Plant/microbe cooperation for electricity generation in a rice paddy field. Appl. Microbiol. Biotechnol. 79, 43–49. Kawaichi, S., Ito, N., Kamikawa, R., Sugawara, T., Yoshida, T., Sako, Y., 2013. Ardenticatena maritima gen. nov., sp. nov., a ferric iron- and nitrate-reducing bacterium of the phylum ‘Chloroflexi’ isolated from an iron-rich coastal hydrothermal field, and description of Ardenticatenia classis nov. Int. J. Syst. Evol. Microbiol. 63, 2992–3002. Lauber, C.L., Hamady, M., Knight, R., Fierer, N., 2009. Pyrosequencing-based assessment of soil pH as a predictor of soil bacterial community structure at the continental scale. Appl. Environ. Microbiol. 75, 5111–5120. Lehmann, S., 2014. Low carbon districts: mitigating the urban heat island with green roof infrastructure. City Cult. Soc. 5, 1–8. Li, X., Wang, X., Zhao, Q., Wan, L., Li, Y., Zhou, Q., 2016. Carbon fiber enhanced bioelectricity generation in soil microbial fuel cells. Biosens. Bioelectron. 85, 135–141. Logan, B., 2010. Scaling up microbial fuel cells and other bioelectrochemical systems. Appl. Microbiol. Biotechnol. 85, 1665–1671. Lu, J., Yuan, J.-g., Yang, J.-z., Chen, A.-k., Yang, Z.-y., 2015a. Effect of substrate depth on initial growth and drought tolerance of Sedum lineare in extensive green roof system. Ecol. Eng. 74, 408–414. Lu, L., Xing, D., Ren, Z.J., 2015b. Microbial community structure accompanied with electricity production in a constructed wetland plant microbial fuel cell. Bioresour. Technol. 195, 115–121. Luo, J., Li, M., Zhou, M., Hu, Y., 2015.. Characterization of a novel strain phylogenetically related to Kocuria rhizophila and its chemical modification to improve performance of microbial fuel cells. Biosens. Bioelectron. 69, 113–120. Maier, R.M., Pepper, I.L., 2009. In: Gerba, R.M.M.L.P.P. (Ed.), Chapter 4 − Earth Environments. Environmental Microbiology, second edition. Academic Press, San Diego. Maricle, B.R., Lee, R.W., 2002. Aerenchyma development and oxygen transport in the estuarine cordgrasses Spartina alterniflora and S. anglica. Aquat. Bot. 74, 109–120. McDonald, D., Price, M.N., Goodrich, J., Nawrocki, E.P., DeSantis, T.Z., Probst, A., Andersen, G.L., Knight, R., Hugenholtz, P., 2012. An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea. ISME J. 6, 610–618. Pinton, R., Varanini, Z., Nannipieri, P., 2007. The Rhizosphere: Biochemistry and Organic Substances at the Soil-Plant Interface, second edition. CRC Press. Quince, C., Lanzen, A., Davenport, R., Turnbaugh, P., 2011. Removing noise from pyrosequenced amplicons. BMC Bioinf. 12, 38. Reyes, R., Bustamante, W., Gironás, J., Pastén, P.A., Rojas, V., Suárez, F., Vera, S., Victorero, F., Bonilla, C.A., 2016. Effect of substrate depth and roof layers on green roof temperature and water requirements in a semi-arid climate. Ecol. Eng. 97, 624–632. Strik, D.P.B.T.B., Hamelers, H.V.M., Snel, J.F.H., Buisman, C.J.N., 2008. Green electricity production with living plants and bacteria in a fuel cell. Int. J. Energy Res. 32, 870–876. Subbarao, G.V., Ae, N., Otani, T., 1997. Genotypic variation in iron-, and aluminumphosphate solubilizing activity of pigeonpea root exudates under P deficient conditions. Soil Sci. Plant Nutr. 43, 295–305. Timmers, R.A., Rothballer, M., Strik, D.P., Engel, M., Schulz, S., Schloter, M., Hartmann, A., Hamelers, B., Buisman, C.J.N., 2012a. Microbial community structure elucidates performance of Glyceria maxima plant microbial fuel cell. Appl. Microbiol. Biotechnol. 94, 537–548. Timmers, R.A., Strik, D.P.B.T.B., Hamelers, H.V.M., Buisman, C.J.N., 2012b. Characterization of the internal resistance of a plant microbial fuel cell. Electrochim. Acta 72, 165–171. Timmers, R.A., Strik, D.P.B.T.B., Hamelers, H.V.M., Buisman, C.J.N., 2013. Electricity generation by a novel design tubular plant microbial fuel cell. Biomass Bioenergy 51, 60–67. Wang, X., Cai, Z., Zhou, Q., Zhang, Z., Chen, C., 2012. Bioelectrochemical stimulation of petroleum hydrocarbon degradation in saline soil using U-tube microbial fuel cells. Biotechnol. Bioeng. 109, 426–433. Zhang, G., Zhang, H., Ma, Y., Yuan, G., Yang, F., Zhang, R., 2014. Membrane filtration biocathode microbial fuel cell for nitrogen removal and electricity generation. Enzyme Microb. Technol. 60, 56–63.
sustainable water management. Acknowledgments This work is part of the microbial fuel cells project in CEDEUS (Centro de Desarrollo Urbano Sustentable), in which we want to acknowledge Dr. Sergio Vera’s team for their help and for the access to the Laboratorio de Infraestructura Vegetal (LIVE). This research was funded by the CEDEUS center CONICYT/FONDAP/15110020. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ecoleng.2017.08.017. References Ap Rees, T., Wilson, P.M., 1984. Effects of Reduced Supply of Oxygen on the Metabolism of Roots of Glyceria maxima and Pisum sativum. Z. Pflanzenphysiol. 114, 493–503. Cao, X., Huang, X., Liang, P., Xiao, K., Zhou, Y., Zhang, X., Logan, B.E., 2009. A new method for water desalination using microbial desalination cells. Environ. Sci. Technol. 43, 7148–7152. Caporaso, J.G., Bittinger, K., Bushman, F.D., DeSantis, T.Z., Andersen, G.L., Knight, R., 2010a. PyNAST: a flexible tool for aligning sequences to a template alignment. Bioinformatics 26, 266–267. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., Fierer, N., Pena, A.G., Goodrich, J.K., Gordon, J.I., Huttley, G.A., Kelley, S.T., Knights, D., Koenig, J.E., Ley, R.E., Lozupone, C.A., McDonald, D., Muegge, B.D., Pirrung, M., Reeder, J., Sevinsky, J.R., Turnbaugh, P.J., Walters, W.A., Widmann, J., Yatsunenko, T., Zaneveld, J., Knight, R., 2010b. QIIME allows analysis of highthroughput community sequencing data. Nat. Meth. 7, 335–336. Chae, K.-J., Choi, M.-J., Lee, J.-W., Kim, K.-Y., Kim, I.S., 2009. Effect of different substrates on the performance, bacterial diversity, and bacterial viability in microbial fuel cells. Bioresour. Technol. 100, 3518–3525. Chen, Z., Huang, Y.-c., Liang, J.-h., Zhao, F., Zhu, Y.-g., 2012. A novel sediment microbial fuel cell with a biocathode in the rice rhizosphere. Bioresour. Technol. 108, 55–59. Chiranjeevi, P., Mohanakrishna, G., Mohan, S.V., 2012. Rhizosphere mediated electrogenesis with the function of anode placement for harnessing bioenergy through CO2 sequestration. Bioresour. Technol. 124, 364–370. De Schamphelaire, L., Cabezas, A., Marzorati, M., Friedrich, M.W., Boon, N., Verstraete, W., 2010. Microbial community analysis of anodes from sediment microbial fuel cells powered by rhizodeposits of living rice plants. Appl. Environ. Microbiol. 76, 2002–2008. DeSantis, T.Z., Hugenholtz, P., Keller, K., Brodie, E.L., Larsen, N., Piceno, Y.M., Phan, R., Andersen, G.L., 2006. NAST: a multiple sequence alignment server for comparative analysis of 16S rRNA genes. Nucleic Acids Res. 34, W394–W399. Domínguez-Garay, A., Berná, A., Ortiz-Bernád, I., Esteve-Núñez, A., 2013. Silica colloid formation enhances performance of sediment microbial fuel cells in a low conductivity soil. Environ. Sci. Technol. 47, 2117–2122. Donovan, C., Dewan, A., Heo, D., Beyenal, H., 2008. Batteryless, wireless sensor powered by a sediment microbial fuel cell. Environ. Sci. Technol. 42, 8591–8596. Edgar, R.C., Haas, B.J., Clemente, J.C., Quince, C., Knight, R., 2011. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27, 2194–2200. Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461. Farrell, C., Mitchell, R.E., Szota, C., Rayner, J.P., Williams, N.S.G., 2012. Green roofs for hot and dry climates: interacting effects of plant water use: succulence and substrate. Ecol. Eng. 49, 270–276. Fierer, N., Leff, J.W., Adams, B.J., Nielsen, U.N., Bates, S.T., Lauber, C.L., Owens, S., Gilbert, J.A., Wall, D.H., Caporaso, J.G., 2012. Cross-biome metagenomic analyses of soil microbial communities and their functional attributes. Proc. Natl. Acad. Sci. U. S. A. 109, 21390–21395. Frank, K.L., Rogers, D.R., Olins, H.C., Vidoudez, C., Girguis, P.R., 2013. Characterizing the distribution and rates of microbial sulfate reduction at Middle Valley hydrothermal vents. ISME J. 7, 1391–1401. Helder, M., Strik, D.P., Hamelers, H.V., Kuhn, A.J., Blok, C., Buisman, C.J., 2010. Concurrent bio-electricity and biomass production in three Plant-Microbial Fuel Cells using Spartina anglica, Arundinella anomala and Arundo donax. Bioresour. Technol. 101, 3541–3547. Helder, M., Chen, W.-S., van der Harst, E.J.M., Strik, D.P.B.T.B., Hamelers, H.V.M.,
210