- Email: [email protected]
Comparison in performance of sediment microbial fuel cells according to depth of embedded anode
Bioresource Technology 127 (2013) 138–142
Contents lists available at SciVerse ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Comparison in performance of sediment microbial fuel cells according to depth of embedded anode Junyeong An a, Bongkyu Kim a, Jonghyeon Nam a, How Yong Ng b, In Seop Chang a,⇑ a b
School of Environmental Science and Engineering, Gwangju Institute of Science and Technology (GIST), 261 Cheomdan-gwagiro, Buk-gu, Gwangju 500-712, Republic of Korea Division of Environmental Science and Engineering, Faculty of Engineering, National University of Singapore, Block EA #03-12, 9 Engineering Drive 1, Singapore 117576, Singapore
h i g h l i g h t s " We investigated the relation of anode-embedding depth with the power output of SMFC. " As the anode depth was increased, the internal resistances of SMCs were increased. " Nevertheless, the SMFC performances were increased as the anode depth was increased. " The anode potential could be a primary parameter for determining the anode depth.
a r t i c l e
i n f o
Article history: Received 1 August 2012 Received in revised form 24 September 2012 Accepted 26 September 2012 Available online 12 October 2012 Keywords: Sediment microbial fuel cell Microbial fuel cell Benthic MFC Redox potential Anode depth
a b s t r a c t Five rigid graphite plates were embedded in evenly divided sections of sediment, ranging from 2 cm (A1) to 10 cm (A5) below the top sediment layer. The maximum power and current of the MFCs increased in depth order; however, despite the increase in the internal resistance, the power and current density of the A5 MFC were 2.2 and 3.5 times higher, respectively, than those of the A1 MFC. In addition, the anode open circuit potentials (OCPs) of the sediment microbial fuel cells (SMFCs) became more negative with sediment depth. Based on these results, it could be then concluded that as the anode-embedding depth increases, that the anode environment is thermodynamically and kinetically favorable to anodophiles or electrophiles. Therefore, the anode-embedding depth should be considered an important parameter that determines the performance of SMFCs, and we posit that the anode potential could be one indicator for selecting the anode-embedding depth. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction A sediment microbial fuel cell (SMFC) is a device used to harvest electricity by locating the anode in sediment phase and placing the cathode in an oxygen-rich water phase (Lowy and Tender, 2008; Rezaei et al., 2007). In the SMFC anode, the family Geobacteraceae is a known anodophile that directly transfers electrons to the anode electrode (Holmes et al., 2004) and sulfur- or iron-reducing bacteria are classified as electrophiles that indirectly donate electrons to the anode via sulfur and iron redox cycles (Holmes et al., 2004). However, these anodic microbial reactions do not occur equally through all layers of the sediment. In general, the water/ sediment boundary is under oxygen rich conditions; thus, oxygen diffuses into the sediment and organic matter in the sediment is aerobically consumed based on the aerobic respiration of heterotrophic microorganisms within a depth 0.5 cm below the sediment ⇑ Corresponding author. Tel.: +82 2 715 3278; fax: +82 2 715 2434. E-mail address: [email protected] (I.S. Chang). 0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2012.09.095
surface (Tsertova et al., 2011). Then, as oxygen levels are depleted relative to the sediment depth, a series of rather stable horizontal gradients can be established within the sediment, in which elec 2 tron acceptors such as NO 3 , NO2 , metal oxides, SO4 , and CO2 (for methanogenesis) are consumed (Nealson and Myers, 1992; Nealson and Stahl, 1997; Londry and Suflita, 1999; Kim et al., 1997). To date, in order to make SMFC installation in the sediment of a natural environment as easy as possible, most sediment MFCs are installed using only anode and cathode electrodes without using chambers or membranes (Lowy and Tender, 2008; Rezaei et al., 2007; Tender et al., 2008). Accordingly, their performance can be affected by a number of factors inherently existing in the natural environment. In particular, embedding the anode at a depth at which O2, NO 3 , or NO2 exists would cause a partial loss of organics by microorganisms that prefer O2, NO 3 , or NO2 as thermodynamically more favorable electron acceptors in the sediment (Nealson and Myers, 1992; Londry and Suflita, 1999; DiChristina, 1992), resulting in a decline of the anode performance. Hence, if the anode
J. An et al. / Bioresource Technology 127 (2013) 138–142
electrode is embedded at a depth at which anodophiles or electrophiles such as metal or sulfate-reducing bacteria are activated, the anode performance might be enhanced. However, it is not easy to identify the sediment depths at which these electron acceptors do not affect the anode performance of SMFCs, because their concentrations vary from sediment to sediment (Nealson and Myers, 1992; Londry and Suflita, 1999; Kim et al., 1997). Even for the acceptors in sediment from the same site, concentrations can vary at different depths and are also affected by any change in the environment around the sediment (DiChristina, 1992; Froelich et al., 1979; Häggblom and Bossert, 2003). This means that the corresponding microbial physiology also changes, with variances found at different sediment depths, sediment sites, and sediment environments. Furthermore, it has been reported that the organic content in most sediments is usually low (between 0.4 and 2.2%) (Chiou and Kile, 2000), resulting in a mass transfer limitation of SMFCs. For these reasons, most SMFC researchers have focused on simple ways to increase the anode performance without considering the sediment nature or overall microbial physiology. For example, in attempts to overcome the limited organic content of sediment, Shantaram et al. (2005) employed an abiotic sacrificial anode, leading to an increase in the anode performance through the corrosion of a manganese alloy. And in other studies, Lowy and Tender (2008) treated anode electrodes with anthraquinone1,6-disulfonic acid in an attempt to enhance the redox reactions of sulfate and sulfide, and Rezaei et al. (2007) enhanced MFC performance by adding chitin into the anode electrodes as a supplementary organic matter. However, that a common nature exists for different types of sediment in a natural environment should be more fully considered; there is a decrease in the redox potential of sediment corresponding to the depth. In sediment, the existence of electron 2 acceptors such as O2, NO 3 , NO2 , metal oxides, SO4 , and CO2 (for CH4 reduction) can lead to different microbial pathways and energy yields (Wang and Francis, 2005). This difference in energy yield then leads to a spatial separation of the redox zones – the higher the energy yield from the microbial pathway, the higher the redox potential maintained. Consequently, the location of where oxygen reduction occurs characterizes a relatively higher redox potential than where denitrification occurs (Kim et al., 1997), and the denitrification pathway establishes a relatively higher redox potential than either the metal or sulfate reduction (Nealson and Stahl, 1997). As a result, the redox potential of sediment might decrease with sediment depth, being a general characteristic of the redox potential profile in sediment. In this system, the redox potential is the sum of the electrode potential and pH effect, such that (Garrels and Christ, 1990):
aA þ bB $ cC þ dD þ n½e þ h½Hþ ;
ð1Þ
Eh ¼ E0 0:05916=n logð½Aa ½Bb =½Cc ½Dd Þ þ ð0:05916 h=nÞpH;
ð2Þ
where A and B are the reactants in reaction (1), C and D are the products of the reaction, n is the number of electrons involved, Eh is the redox potential, E0 is the standard potential, and a, b, c, d, and h denote the reaction coefficients. When based on Eq. (2), it can be seen that decrease in the redox potential implies that there is also an increase in the production of protons and electrons or a decrease in the electrode potential. Hence, it is expected that as the anode-embedding depth becomes deeper, more electrons are produced or that a more negative anode potential would be produced. As such, in this study, we: 1) compare the performances of SMFCs having anode electrodes horizontally embedded at different depths in a homogeneous sediment; 2) investigate relationships among the anode-embedding depth, anode performance, and anode
139
potential; and 3) recommend that the anode potential of SMFCs could be a primary parameter for determining the anode depth that optimizes power generation. 2. Methods 2.1. Electrode preparation Thin rigid graphite plates (1 mm thickness, 10 cm length, 10 cm width) were used as the anode electrode in order to prevent physical deflection and folding when being embedded in sediment. The middle edge of each plate was inserted into the gap between a folded titanium plate (1 mm thickness, 3 cm length, 3 cm width), and connected to an electrical wire. The titanium plate and wire were then sealed with poly epoxy to make them waterproof. For the cathode electrode, graphite felt (2.54 cm thickness, 10 cm length, 10 cm width) was used to ensure a sufficient cathode area for the oxygen reduction reaction (ORR). In the same manner as the anode electrode preparation, the felts were connected to titanium plates and wires. 2.2. Installation of MFCs and operating conditions Sediment (0–15 cm depth) and local reservoir water were collected from the Jangseong Reservoir (near Gwangju, Korea). The collected sediment was physically mixed in a laboratory for 2 h to ensure that the organic and inorganic contents of the sediment were homogeneous. Then, to construct duplicate SMFCs, 10 rigid graphite plates, denoted as A1–A5, were installed at even intervals in an aquarium, and 10 acrylic frames were used to fix the anode electrodes at different depths. As sediment was added to the bottom of the aquarium, the anode electrodes were horizontally inserted into the frame closest to the bottom. Note that each section contained one anode; the anode electrodes were installed as follows: A1 (2 cm), A2 (4 cm), A3 (6 cm), A4 (8 cm), and A5 (10 cm) below the sediment surface. After the installation of all anode electrodes, the sediment was evenly packed. The all cathodes were subsequently installed at a height 2 cm above the sediment surface (see Supplementary Information for further details). Thereafter, air-saturated fresh reservoir water was circulated in the aquarium, and maintained at a 5.18 ± 0.29 mg/L DO concentration in order to remove cathode limitations caused by lack of oxygen (An et al., 2010). Note also that the organic content in the sediment was 7.2 ± 0.9%, as measured by standard methods and that the MFCs were operated under ambient temperature (20–22 °C). 2.3. Analyses The open-circuit voltage (OCV) and closed-circuit voltage (CCV) of the MFCs were measured using a data acquisition system (Multimeter 2700, Keithly Instruments Inc., Cleveland, OH) that was connected to a personal computer. During these tests, the time interval for data acquisition was varied from 10 s to 30 min, depending on the experiment being performed. Current–voltage (I–V) curves of the tested MFCs were measured by employing a variable resistance that ranged from 40 kO to 200 O. The current (I) was then calculated based on Ohm’s law, I = V/R, where V is the voltage difference between the anode and cathode, and R is the external circuit resistance used during experiments. Sequential (step) decreases in the external resistance ranged from 40 kO to 600 O; each resistance was applied for 15 min, and data were acquired every 10 s. To observe differences in the internal resistance according to the depth of the embedded anode, the solution resistance was estimated from electrochemical impedance spectroscopy (EIS) measurements obtained using Autolab with an FRA–
140
J. An et al. / Bioresource Technology 127 (2013) 138–142
ADC impedance module (PGSTAT302, Eco Chemie, Utrecht) for a high frequency range of 100 kHz to 0.01 Hz. The onset voltages (two-electrode configuration) were OCVs with a ± mV amplitude (An et al., 2011). Here, the DO concentration and pH were measured using a 4-star DO pH meter (Thermo Fisher Scientific Inc., MA).
3. Results and discussion 3.1. Change in anode performance according to anode depth After all MFCs reached a stable 0.84 ± 0.01 V OCV, the MFCs were converted to closed-circuit mode using a 1 kO external resistance. After connection of the external loads, the MFC CCVs sharply increased from 0.13 to 0.26 V for 1 d, but thereafter, the developed CCVs stabilized at different levels; the deeper the anode was installed, the higher the voltage produced (Fig. 1). In the figure, the A5 MFC is seen to produce a 400% higher CCV than the A1 MFC. To further investigate the reason for this voltage difference, the MFCs were converted into open-circuit mode and the performance curves measured. After 1 d of conversion, the OCVs stabilized at different levels (i.e., onset voltages of Fig. 2a). When classifying the OCVs relative to their anode and cathode potentials, it is seen that the OCV differences were caused by differences in the anode open circuit potentials (OCPs; anode onset potentials of Fig. 2b): anode A1 was 0.20 V, anode A2 was 0.27 V, anode A3 was 0.32 V, anode A4 was 0.38 V, and anode A5 was 0.44 V. In contrast, all cathode OCPs were quite similar due to the circulation of air-saturated water. These results indicate that the anode potential is more negative as the anode-embedding depth becomes deeper, similar to the general nature of change in the redox potential of most sediment (Nealson and Stahl, 1997; Froelich et al., 1979; Wang and Francis, 2005). When the I–V curves were measured, the maximum power and current of the MFCs increased with respect to depth order (the deeper, the higher); the power and current density of the A5 MFC (at a 10 cm depth) were 2.2 times and 3.5 times higher, respectively, than those of the A1 MFC (at a 2 cm depth). Hence, it can be concluded that as the anode-embedding depth becomes deeper, more electrons and a more negative anode potential are produced, suggesting that the anode environment is both thermodynamically and kinetically favorable to anodophiles and electrophiles. These results thus imply that the anode-embedding depth is an important parameter that determines the anode performance of SMFCs and that the anode potential could be one indicator for selecting the anode-embedding depth.
Fig. 1. Voltage development of sediment MFCs under closed-circuit mode with 1 kO external resistance for 12 d.
Fig. 2. (A) Performances of sediment MFCs measured after 12 d of closed-circuit operation; (B) electrode potential plots of anode (hollow) and cathode (gray) from I–V curves, plotted every 15 min.
3.2. Comparison in development of anode potentials by sequential decreases in the external resistance The A5 MFC displayed the maximum performance due to the performance of the anode A5, implying that the sediment environment at a 10 cm depth is the most thermodynamically and kinetically favorable to anodophiles and electrophiles. To obtain evidence supporting this statement, real-time changes in the anode overpotential according to the current change were investigated by plotting the I–V curves of the Fig. 2 as V–T (time) curves, using data measured every 10 s. Note that the I–V curves were obtained using a step-down of the external resistance, ranging from 40 kO to 600 O (An et al., 2011). Under high resistances (40 and 30 kO), after an anode potential immediately increased on connection of each resistance, all anode potentials developed almost equally (Fig. 3). However, when resistances from 20 to 3 kO were applied, the anodes potentials of A1, A2, A3, and A4 gradually increased. All cathode potentials were almost equally developed under the applied resistances, implying that the electron production rate of the anodes did not catch up to the electron consumption rate in the counter cathodes. This behavior of anodes A1, A2, A3, and A4 became even clearer after applying 2 kO to 600 O resistances. In contrast, the potential of anode A5 evenly developed under 20 and 3 kO. Thereafter, when 2, 1 kO, and 600 O were connected, the anode potential of A5 temporally increased, but with time it declined toward a negative value.
141
J. An et al. / Bioresource Technology 127 (2013) 138–142
Table 2 Internal resistances of MFCs obtained from EIS: internal resistances between the water–sediment interface and anodes, and between the cathodes and the water– sediment interface.
Fig. 3. Potential change by sequential (step) decrease of external resistance from 40 kO to 600 O; anode overpotentials, potential difference before and after connection of each resistance for 10 s, are presented in Table 1.
This result demonstrates that as time passes the electron transfer rate to the A5 anode gradually increases and catches up with the demand for electrons at the cathode, meaning that substrate oxidation increases in the anode (Bard and Faulkner, 2001), thereby confirming that the activity of anodophiles or electrophiles in anode A5 increased over time. As such, we posit here that this result could be indirect evidence that the anodophiles or electrophiles could prefer anode A5 as the final electron acceptor, embedded at the depth of 10 cm below the water/sediment boundary interface. Table 1 displays the anode overpotentials that were obtained from differences of the anode potentials before and after connection of each resistance for an initial 10 s. In the table, as the depth of the embedded anode becomes deeper, the anode overpotentials increase. This increase is possibly due to an increase of the internal resistance caused by the increased distance between the anode and cathode. Accordingly, we also investigated how much the internal resistance increased as the anode depth became deeper. 3.3. Compression in internal resistances according to anodeembedding depths To compare changes in the internal resistance of MFCs at different anode depths, EIS was used to measure the solution resistance. As presented in Table 2, the internal resistances between cathodes and the water/sediment interface of MFCs range from 20 to 34 O, probably due to the uneven water/sediment boundary interface. The internal resistance between the water/sediment interface and anodes increases as the anode-embedding depth becomes deeper. The increase in the internal resistance results in an increase of the total internal resistances occurring in the MFCs. Hence, it is clear that as the anode electrode becomes more deeply installed, Table 1 Anode overpotentials, the anode potential difference before and after connection of each resistance for initial 10 s, obtained during sequential (step) decrease of external resistance from 40 kO to 600 O. External resistance
Anode overpotentials (mV) Anode A1
Anode A2
Anode A3
Anode A4
Anode A5
40 kO 30 kO 20 kO 3 kO 2 kO 1 kO 600 O
3.9 0.8 9.9 30.7 18.2 43.6 39.2
4.6 1.2 10.9 34.3 20.2 49.4 43.8
4.4 1.3 11.7 36.6 21.7 53.1 48.0
5.4 1.7 12.1 40.8 23.7 59.9 52.5
6.4 1.9 15.7 48.3 27.9 69.7 60.8
MFCs (anode embedding depth)
Internal resistance (X)
MFC MFC MFC MFC MFC
110 ± 9.6 148 ± 10.5 165 ± 15.1 182 ± 11.6 199 ± 13.2
w/A1 w/A2 w/A3 w/A4 w/A5
(2 cm) (4 cm) (6 cm) (8 cm) (10 cm)
Water–sediment Cathode and water– interface and anode sediment interface 34.0 ± 2.8 25.5 ± 2.6 25.8 ± 4.23 20.2 ± 3.26 25.3 ± 2.04
Total internal resistance (X)
144 ± 12.4 173 ± 13.1 191 ± 19.5 203 ± 14.9 224 ± 15.2
MFCs overcome a relatively higher internal resistance in order to produce more power; our results suggest that embedding an anode within a 10 cm depth can offset the increase in anode overpotential caused by an increase in the anode-embedded depth. Recently, Song et al. (2010) embedded a graphite plate (127 cm2 surface area) at a depth 4 cm under sediment and obtained a maximum power density of 3.15 mW/m. The anode OCP they maintained was approximately 140 mV versus the saturated calomel electrode, and 93 mV versus the saturated Ag/AgCl electrode; the internal resistance between the anode and cathode was 214 X. In this study, the MFC with an A5 anode (graphite plate, 200 cm2 surface area, embedded at 10 cm depth) produced a 14.5 mW/m maximum power density. The A5 anode maintained an OCP of 422 mV against the saturated Ag/AgCl reference electrode, even after 24 d of initiation (Fig. 2b); the internal resistance between the A5 anode and cathode was 224 ± 15.2 X. When based on our results, it is expected that previous researchers (Song et al., 2010) could obtain more power from their SMFCs if they embedded the anode at a depth of between 4 and 10 cm, where the anode potential was maintained at approximately 420 mV versus the saturated Ag/AgCl electrode. Hence, we suggest that information pertaining to the anode potentials could be informative for selecting the depth for embedding the anode in sediment. In this study, we limited the anode-embedding depth to a range from 2 to 10 cm, thus it is not clear whether more power can be generated from a sediment depth of more than 10 cm. However, it is thought that if the depth is increased to more than 10 cm, the internal resistance will be further increased and may lower the SMFC performance, even though the anode performance is increased. Therefore, it is important to identify the trade-off point between the internal resistance (or anode overpotential) and the MFC power output.
4. Conclusion Based on the results, we concluded that as the anode-embedding depth increases, more electrons and a more negative anode potential are produced, i.e., that the anode environment is thermodynamically and kinetically favorable to anodophiles and electrophiles. Therefore, the anode-embedding depth should be considered an important parameter that determines the performance of SMFCs, and we posit that the anode potential could be one indicator for selecting the anode-embedding depth.
Acknowledgements This work was supported by a grant from Doyak (R0A-2008000-20088-0) funded by the National Research Foundation (NRF) of the Korean government (MEST), and by the Gwangju Institute of Science and Technology (GIST) institutional research program.
142
J. An et al. / Bioresource Technology 127 (2013) 138–142
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.biortech.2012.09. 095. References An, J., Moon, H., Chang, I.S., 2010. Multiphase electrode microbial fuel cell system that simultaneously converts organics coexisting in water and sediment phases to electricity. Environ. Sci. Technol. 44 (18), 7145–7150. An, J., Jeon, H., Lee, J., Chang, I.S., 2011. Bifunctional silver nanoparticle cathode in microbial fuel cells for microbial growth inhibition with comparable oxygen reduction reaction activity. Environ. Sci. Technol. 45 (12), 5441–5446. Bard, A.J., Faulkner, A.R., 2001. Electrochemical Methods: Fundamentals and Application, second ed. Wiley & Sons, New York (Chapter 3). Chiou, C.T., Kile, D.E., 2000. Contaminant sorption by soil and bed sediment; U.S. Department of the Interior, U.S. Department of the Interior, U.S. Geological Survey: Reston, VA. DiChristina, T.J., 1992. Effects of nitrogen and nitrite in dissimilatory ion reduction by Shewanella putrefaciens 200. J. Bacteriol. 174 (6), 1891–1896. Froelich, P.N., Klinkhammer, G.P., Bender, M.L., Luedtke, N.A., Heath, G.R., Cullen, D., Dauphin, P., Hammond, D., Hartman, B., Maynard, V., 1979. Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis. Geochim. Cosmochim. Acta 43 (7), 1075–1090. Garrels, R.M., Christ, C.L., 1990. Solutions, Minerals and Equilibria. Jones and Bartlett, London. Häggblom, M.M., Bossert, I.D. (Eds.), 2003. Dehalogenation: Microbial Processes and Environmental Applications. Kluwer Academic Publishers, Boston, MA, pp. 42– 48. Holmes, D.E., Nicoll, J.S., Bond, D.R., Lovley, D.R., 2004. Potential role of a novel psychrotolerant member of the family Geobacteraceae, Geopsychrobacter electrodiphilus gen. nov., sp nov., in electricity production by a marine sediment fuel cell. Appl. Environ. Microbiol. 70 (10), 6023–6030.
Kim, D.H., Matsuda, O., Yamamoto, T., 1997. Nitrification, denitrification and nitrate reduction rates in the sediment of Hirosima Bay. Japan. J. Oceanogr. 55 (4), 317– 324. Londry, K.L., Suflita, J.M., 1999. Use of nitrate to control sulfide generation by sulfate-reducing bacterial associated with oily waste. J. Ind. Microbiol. Biotechnol. 22 (6), 582–589. Lowy, D.A., Tender, L.M., 2008. Harvesting energy from the marine sediment-water interface III. Kinetic activity of quinone- and antimony-based anode materials. J. Power Sources 185 (1), 70–75. Nealson, K.H., Stahl, D.A., 1997. Microorganism and biogeochemical cycles: what can we learn from layered microbial communityes? In: Nealson, K.H., Banfield, J.F. (Eds.), Geomicrobiology: Interactions between Microbes and Minerals, 35. Mineralogical Society of America, Washington, D.C, pp. 5–34. Nealson, K.H., Myers, C.R., 1992. Microbial reduction of manganese and Iron: new approaches to carbon cycling. Appl. Environ. Microbiol. 58 (2), 439–443. Rezaei, F., Richard, T.L., Brennan, R.A., Logan, B.E., 2007. Substrate-enhanced microbial fuel cells for improved remote power generation from sedimentbased systems. Environ. Sci. Technol. 41 (11), 4053–4058. Shantaram, A., Beyenal, H., Raajan, R., Veluchamy, A., Lewandowski, Z., 2005. Wireless sensors powered by microbial fuel cells. Environ. Sci. Technol. 39 (13), 5037–5042. Song, T.S., Yan, Z.S., Zhao, Z.W., Jiang, H.L., 2010. Removal of organic matter in freshwater sediment by microbial fuel cells at various external resistances. J. Chem. Technol. Biotechno. 85 (11), 1489–1493. Tsertova, N., Kisand, A., Tammert, H., Kisand, V., 2011. Low seasonal variability in community composition of sediment bacteria in large and shallow lake. Environ. Microbiol. Rep. 3 (2), 270–277. Tender, L.M., Gray, S.A., Groveman, E., Lowy, D.A., Kauffman, P., Melhado, J., Tyce, R.C., Flynn, D., Petrecca, R., Dobarro, J., 2008. The first demonstration of a microbial fuel cell as a viable power supply: powering a meteorological buoy. J. Power Sources 179 (2), 571–575. Wang, Y., Francis, A.J., 2005. Evaluation of microbial activity for long-term assessment of deep geologic nuclear waste repositories. J. Nucl. Radiochem. Sci. 6 (1), 43–50.
Contents lists available at SciVerse ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Comparison in performance of sediment microbial fuel cells according to depth of embedded anode Junyeong An a, Bongkyu Kim a, Jonghyeon Nam a, How Yong Ng b, In Seop Chang a,⇑ a b
School of Environmental Science and Engineering, Gwangju Institute of Science and Technology (GIST), 261 Cheomdan-gwagiro, Buk-gu, Gwangju 500-712, Republic of Korea Division of Environmental Science and Engineering, Faculty of Engineering, National University of Singapore, Block EA #03-12, 9 Engineering Drive 1, Singapore 117576, Singapore
h i g h l i g h t s " We investigated the relation of anode-embedding depth with the power output of SMFC. " As the anode depth was increased, the internal resistances of SMCs were increased. " Nevertheless, the SMFC performances were increased as the anode depth was increased. " The anode potential could be a primary parameter for determining the anode depth.
a r t i c l e
i n f o
Article history: Received 1 August 2012 Received in revised form 24 September 2012 Accepted 26 September 2012 Available online 12 October 2012 Keywords: Sediment microbial fuel cell Microbial fuel cell Benthic MFC Redox potential Anode depth
a b s t r a c t Five rigid graphite plates were embedded in evenly divided sections of sediment, ranging from 2 cm (A1) to 10 cm (A5) below the top sediment layer. The maximum power and current of the MFCs increased in depth order; however, despite the increase in the internal resistance, the power and current density of the A5 MFC were 2.2 and 3.5 times higher, respectively, than those of the A1 MFC. In addition, the anode open circuit potentials (OCPs) of the sediment microbial fuel cells (SMFCs) became more negative with sediment depth. Based on these results, it could be then concluded that as the anode-embedding depth increases, that the anode environment is thermodynamically and kinetically favorable to anodophiles or electrophiles. Therefore, the anode-embedding depth should be considered an important parameter that determines the performance of SMFCs, and we posit that the anode potential could be one indicator for selecting the anode-embedding depth. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction A sediment microbial fuel cell (SMFC) is a device used to harvest electricity by locating the anode in sediment phase and placing the cathode in an oxygen-rich water phase (Lowy and Tender, 2008; Rezaei et al., 2007). In the SMFC anode, the family Geobacteraceae is a known anodophile that directly transfers electrons to the anode electrode (Holmes et al., 2004) and sulfur- or iron-reducing bacteria are classified as electrophiles that indirectly donate electrons to the anode via sulfur and iron redox cycles (Holmes et al., 2004). However, these anodic microbial reactions do not occur equally through all layers of the sediment. In general, the water/ sediment boundary is under oxygen rich conditions; thus, oxygen diffuses into the sediment and organic matter in the sediment is aerobically consumed based on the aerobic respiration of heterotrophic microorganisms within a depth 0.5 cm below the sediment ⇑ Corresponding author. Tel.: +82 2 715 3278; fax: +82 2 715 2434. E-mail address: [email protected] (I.S. Chang). 0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2012.09.095
surface (Tsertova et al., 2011). Then, as oxygen levels are depleted relative to the sediment depth, a series of rather stable horizontal gradients can be established within the sediment, in which elec 2 tron acceptors such as NO 3 , NO2 , metal oxides, SO4 , and CO2 (for methanogenesis) are consumed (Nealson and Myers, 1992; Nealson and Stahl, 1997; Londry and Suflita, 1999; Kim et al., 1997). To date, in order to make SMFC installation in the sediment of a natural environment as easy as possible, most sediment MFCs are installed using only anode and cathode electrodes without using chambers or membranes (Lowy and Tender, 2008; Rezaei et al., 2007; Tender et al., 2008). Accordingly, their performance can be affected by a number of factors inherently existing in the natural environment. In particular, embedding the anode at a depth at which O2, NO 3 , or NO2 exists would cause a partial loss of organics by microorganisms that prefer O2, NO 3 , or NO2 as thermodynamically more favorable electron acceptors in the sediment (Nealson and Myers, 1992; Londry and Suflita, 1999; DiChristina, 1992), resulting in a decline of the anode performance. Hence, if the anode
J. An et al. / Bioresource Technology 127 (2013) 138–142
electrode is embedded at a depth at which anodophiles or electrophiles such as metal or sulfate-reducing bacteria are activated, the anode performance might be enhanced. However, it is not easy to identify the sediment depths at which these electron acceptors do not affect the anode performance of SMFCs, because their concentrations vary from sediment to sediment (Nealson and Myers, 1992; Londry and Suflita, 1999; Kim et al., 1997). Even for the acceptors in sediment from the same site, concentrations can vary at different depths and are also affected by any change in the environment around the sediment (DiChristina, 1992; Froelich et al., 1979; Häggblom and Bossert, 2003). This means that the corresponding microbial physiology also changes, with variances found at different sediment depths, sediment sites, and sediment environments. Furthermore, it has been reported that the organic content in most sediments is usually low (between 0.4 and 2.2%) (Chiou and Kile, 2000), resulting in a mass transfer limitation of SMFCs. For these reasons, most SMFC researchers have focused on simple ways to increase the anode performance without considering the sediment nature or overall microbial physiology. For example, in attempts to overcome the limited organic content of sediment, Shantaram et al. (2005) employed an abiotic sacrificial anode, leading to an increase in the anode performance through the corrosion of a manganese alloy. And in other studies, Lowy and Tender (2008) treated anode electrodes with anthraquinone1,6-disulfonic acid in an attempt to enhance the redox reactions of sulfate and sulfide, and Rezaei et al. (2007) enhanced MFC performance by adding chitin into the anode electrodes as a supplementary organic matter. However, that a common nature exists for different types of sediment in a natural environment should be more fully considered; there is a decrease in the redox potential of sediment corresponding to the depth. In sediment, the existence of electron 2 acceptors such as O2, NO 3 , NO2 , metal oxides, SO4 , and CO2 (for CH4 reduction) can lead to different microbial pathways and energy yields (Wang and Francis, 2005). This difference in energy yield then leads to a spatial separation of the redox zones – the higher the energy yield from the microbial pathway, the higher the redox potential maintained. Consequently, the location of where oxygen reduction occurs characterizes a relatively higher redox potential than where denitrification occurs (Kim et al., 1997), and the denitrification pathway establishes a relatively higher redox potential than either the metal or sulfate reduction (Nealson and Stahl, 1997). As a result, the redox potential of sediment might decrease with sediment depth, being a general characteristic of the redox potential profile in sediment. In this system, the redox potential is the sum of the electrode potential and pH effect, such that (Garrels and Christ, 1990):
aA þ bB $ cC þ dD þ n½e þ h½Hþ ;
ð1Þ
Eh ¼ E0 0:05916=n logð½Aa ½Bb =½Cc ½Dd Þ þ ð0:05916 h=nÞpH;
ð2Þ
where A and B are the reactants in reaction (1), C and D are the products of the reaction, n is the number of electrons involved, Eh is the redox potential, E0 is the standard potential, and a, b, c, d, and h denote the reaction coefficients. When based on Eq. (2), it can be seen that decrease in the redox potential implies that there is also an increase in the production of protons and electrons or a decrease in the electrode potential. Hence, it is expected that as the anode-embedding depth becomes deeper, more electrons are produced or that a more negative anode potential would be produced. As such, in this study, we: 1) compare the performances of SMFCs having anode electrodes horizontally embedded at different depths in a homogeneous sediment; 2) investigate relationships among the anode-embedding depth, anode performance, and anode
139
potential; and 3) recommend that the anode potential of SMFCs could be a primary parameter for determining the anode depth that optimizes power generation. 2. Methods 2.1. Electrode preparation Thin rigid graphite plates (1 mm thickness, 10 cm length, 10 cm width) were used as the anode electrode in order to prevent physical deflection and folding when being embedded in sediment. The middle edge of each plate was inserted into the gap between a folded titanium plate (1 mm thickness, 3 cm length, 3 cm width), and connected to an electrical wire. The titanium plate and wire were then sealed with poly epoxy to make them waterproof. For the cathode electrode, graphite felt (2.54 cm thickness, 10 cm length, 10 cm width) was used to ensure a sufficient cathode area for the oxygen reduction reaction (ORR). In the same manner as the anode electrode preparation, the felts were connected to titanium plates and wires. 2.2. Installation of MFCs and operating conditions Sediment (0–15 cm depth) and local reservoir water were collected from the Jangseong Reservoir (near Gwangju, Korea). The collected sediment was physically mixed in a laboratory for 2 h to ensure that the organic and inorganic contents of the sediment were homogeneous. Then, to construct duplicate SMFCs, 10 rigid graphite plates, denoted as A1–A5, were installed at even intervals in an aquarium, and 10 acrylic frames were used to fix the anode electrodes at different depths. As sediment was added to the bottom of the aquarium, the anode electrodes were horizontally inserted into the frame closest to the bottom. Note that each section contained one anode; the anode electrodes were installed as follows: A1 (2 cm), A2 (4 cm), A3 (6 cm), A4 (8 cm), and A5 (10 cm) below the sediment surface. After the installation of all anode electrodes, the sediment was evenly packed. The all cathodes were subsequently installed at a height 2 cm above the sediment surface (see Supplementary Information for further details). Thereafter, air-saturated fresh reservoir water was circulated in the aquarium, and maintained at a 5.18 ± 0.29 mg/L DO concentration in order to remove cathode limitations caused by lack of oxygen (An et al., 2010). Note also that the organic content in the sediment was 7.2 ± 0.9%, as measured by standard methods and that the MFCs were operated under ambient temperature (20–22 °C). 2.3. Analyses The open-circuit voltage (OCV) and closed-circuit voltage (CCV) of the MFCs were measured using a data acquisition system (Multimeter 2700, Keithly Instruments Inc., Cleveland, OH) that was connected to a personal computer. During these tests, the time interval for data acquisition was varied from 10 s to 30 min, depending on the experiment being performed. Current–voltage (I–V) curves of the tested MFCs were measured by employing a variable resistance that ranged from 40 kO to 200 O. The current (I) was then calculated based on Ohm’s law, I = V/R, where V is the voltage difference between the anode and cathode, and R is the external circuit resistance used during experiments. Sequential (step) decreases in the external resistance ranged from 40 kO to 600 O; each resistance was applied for 15 min, and data were acquired every 10 s. To observe differences in the internal resistance according to the depth of the embedded anode, the solution resistance was estimated from electrochemical impedance spectroscopy (EIS) measurements obtained using Autolab with an FRA–
140
J. An et al. / Bioresource Technology 127 (2013) 138–142
ADC impedance module (PGSTAT302, Eco Chemie, Utrecht) for a high frequency range of 100 kHz to 0.01 Hz. The onset voltages (two-electrode configuration) were OCVs with a ± mV amplitude (An et al., 2011). Here, the DO concentration and pH were measured using a 4-star DO pH meter (Thermo Fisher Scientific Inc., MA).
3. Results and discussion 3.1. Change in anode performance according to anode depth After all MFCs reached a stable 0.84 ± 0.01 V OCV, the MFCs were converted to closed-circuit mode using a 1 kO external resistance. After connection of the external loads, the MFC CCVs sharply increased from 0.13 to 0.26 V for 1 d, but thereafter, the developed CCVs stabilized at different levels; the deeper the anode was installed, the higher the voltage produced (Fig. 1). In the figure, the A5 MFC is seen to produce a 400% higher CCV than the A1 MFC. To further investigate the reason for this voltage difference, the MFCs were converted into open-circuit mode and the performance curves measured. After 1 d of conversion, the OCVs stabilized at different levels (i.e., onset voltages of Fig. 2a). When classifying the OCVs relative to their anode and cathode potentials, it is seen that the OCV differences were caused by differences in the anode open circuit potentials (OCPs; anode onset potentials of Fig. 2b): anode A1 was 0.20 V, anode A2 was 0.27 V, anode A3 was 0.32 V, anode A4 was 0.38 V, and anode A5 was 0.44 V. In contrast, all cathode OCPs were quite similar due to the circulation of air-saturated water. These results indicate that the anode potential is more negative as the anode-embedding depth becomes deeper, similar to the general nature of change in the redox potential of most sediment (Nealson and Stahl, 1997; Froelich et al., 1979; Wang and Francis, 2005). When the I–V curves were measured, the maximum power and current of the MFCs increased with respect to depth order (the deeper, the higher); the power and current density of the A5 MFC (at a 10 cm depth) were 2.2 times and 3.5 times higher, respectively, than those of the A1 MFC (at a 2 cm depth). Hence, it can be concluded that as the anode-embedding depth becomes deeper, more electrons and a more negative anode potential are produced, suggesting that the anode environment is both thermodynamically and kinetically favorable to anodophiles and electrophiles. These results thus imply that the anode-embedding depth is an important parameter that determines the anode performance of SMFCs and that the anode potential could be one indicator for selecting the anode-embedding depth.
Fig. 1. Voltage development of sediment MFCs under closed-circuit mode with 1 kO external resistance for 12 d.
Fig. 2. (A) Performances of sediment MFCs measured after 12 d of closed-circuit operation; (B) electrode potential plots of anode (hollow) and cathode (gray) from I–V curves, plotted every 15 min.
3.2. Comparison in development of anode potentials by sequential decreases in the external resistance The A5 MFC displayed the maximum performance due to the performance of the anode A5, implying that the sediment environment at a 10 cm depth is the most thermodynamically and kinetically favorable to anodophiles and electrophiles. To obtain evidence supporting this statement, real-time changes in the anode overpotential according to the current change were investigated by plotting the I–V curves of the Fig. 2 as V–T (time) curves, using data measured every 10 s. Note that the I–V curves were obtained using a step-down of the external resistance, ranging from 40 kO to 600 O (An et al., 2011). Under high resistances (40 and 30 kO), after an anode potential immediately increased on connection of each resistance, all anode potentials developed almost equally (Fig. 3). However, when resistances from 20 to 3 kO were applied, the anodes potentials of A1, A2, A3, and A4 gradually increased. All cathode potentials were almost equally developed under the applied resistances, implying that the electron production rate of the anodes did not catch up to the electron consumption rate in the counter cathodes. This behavior of anodes A1, A2, A3, and A4 became even clearer after applying 2 kO to 600 O resistances. In contrast, the potential of anode A5 evenly developed under 20 and 3 kO. Thereafter, when 2, 1 kO, and 600 O were connected, the anode potential of A5 temporally increased, but with time it declined toward a negative value.
141
J. An et al. / Bioresource Technology 127 (2013) 138–142
Table 2 Internal resistances of MFCs obtained from EIS: internal resistances between the water–sediment interface and anodes, and between the cathodes and the water– sediment interface.
Fig. 3. Potential change by sequential (step) decrease of external resistance from 40 kO to 600 O; anode overpotentials, potential difference before and after connection of each resistance for 10 s, are presented in Table 1.
This result demonstrates that as time passes the electron transfer rate to the A5 anode gradually increases and catches up with the demand for electrons at the cathode, meaning that substrate oxidation increases in the anode (Bard and Faulkner, 2001), thereby confirming that the activity of anodophiles or electrophiles in anode A5 increased over time. As such, we posit here that this result could be indirect evidence that the anodophiles or electrophiles could prefer anode A5 as the final electron acceptor, embedded at the depth of 10 cm below the water/sediment boundary interface. Table 1 displays the anode overpotentials that were obtained from differences of the anode potentials before and after connection of each resistance for an initial 10 s. In the table, as the depth of the embedded anode becomes deeper, the anode overpotentials increase. This increase is possibly due to an increase of the internal resistance caused by the increased distance between the anode and cathode. Accordingly, we also investigated how much the internal resistance increased as the anode depth became deeper. 3.3. Compression in internal resistances according to anodeembedding depths To compare changes in the internal resistance of MFCs at different anode depths, EIS was used to measure the solution resistance. As presented in Table 2, the internal resistances between cathodes and the water/sediment interface of MFCs range from 20 to 34 O, probably due to the uneven water/sediment boundary interface. The internal resistance between the water/sediment interface and anodes increases as the anode-embedding depth becomes deeper. The increase in the internal resistance results in an increase of the total internal resistances occurring in the MFCs. Hence, it is clear that as the anode electrode becomes more deeply installed, Table 1 Anode overpotentials, the anode potential difference before and after connection of each resistance for initial 10 s, obtained during sequential (step) decrease of external resistance from 40 kO to 600 O. External resistance
Anode overpotentials (mV) Anode A1
Anode A2
Anode A3
Anode A4
Anode A5
40 kO 30 kO 20 kO 3 kO 2 kO 1 kO 600 O
3.9 0.8 9.9 30.7 18.2 43.6 39.2
4.6 1.2 10.9 34.3 20.2 49.4 43.8
4.4 1.3 11.7 36.6 21.7 53.1 48.0
5.4 1.7 12.1 40.8 23.7 59.9 52.5
6.4 1.9 15.7 48.3 27.9 69.7 60.8
MFCs (anode embedding depth)
Internal resistance (X)
MFC MFC MFC MFC MFC
110 ± 9.6 148 ± 10.5 165 ± 15.1 182 ± 11.6 199 ± 13.2
w/A1 w/A2 w/A3 w/A4 w/A5
(2 cm) (4 cm) (6 cm) (8 cm) (10 cm)
Water–sediment Cathode and water– interface and anode sediment interface 34.0 ± 2.8 25.5 ± 2.6 25.8 ± 4.23 20.2 ± 3.26 25.3 ± 2.04
Total internal resistance (X)
144 ± 12.4 173 ± 13.1 191 ± 19.5 203 ± 14.9 224 ± 15.2
MFCs overcome a relatively higher internal resistance in order to produce more power; our results suggest that embedding an anode within a 10 cm depth can offset the increase in anode overpotential caused by an increase in the anode-embedded depth. Recently, Song et al. (2010) embedded a graphite plate (127 cm2 surface area) at a depth 4 cm under sediment and obtained a maximum power density of 3.15 mW/m. The anode OCP they maintained was approximately 140 mV versus the saturated calomel electrode, and 93 mV versus the saturated Ag/AgCl electrode; the internal resistance between the anode and cathode was 214 X. In this study, the MFC with an A5 anode (graphite plate, 200 cm2 surface area, embedded at 10 cm depth) produced a 14.5 mW/m maximum power density. The A5 anode maintained an OCP of 422 mV against the saturated Ag/AgCl reference electrode, even after 24 d of initiation (Fig. 2b); the internal resistance between the A5 anode and cathode was 224 ± 15.2 X. When based on our results, it is expected that previous researchers (Song et al., 2010) could obtain more power from their SMFCs if they embedded the anode at a depth of between 4 and 10 cm, where the anode potential was maintained at approximately 420 mV versus the saturated Ag/AgCl electrode. Hence, we suggest that information pertaining to the anode potentials could be informative for selecting the depth for embedding the anode in sediment. In this study, we limited the anode-embedding depth to a range from 2 to 10 cm, thus it is not clear whether more power can be generated from a sediment depth of more than 10 cm. However, it is thought that if the depth is increased to more than 10 cm, the internal resistance will be further increased and may lower the SMFC performance, even though the anode performance is increased. Therefore, it is important to identify the trade-off point between the internal resistance (or anode overpotential) and the MFC power output.
4. Conclusion Based on the results, we concluded that as the anode-embedding depth increases, more electrons and a more negative anode potential are produced, i.e., that the anode environment is thermodynamically and kinetically favorable to anodophiles and electrophiles. Therefore, the anode-embedding depth should be considered an important parameter that determines the performance of SMFCs, and we posit that the anode potential could be one indicator for selecting the anode-embedding depth.
Acknowledgements This work was supported by a grant from Doyak (R0A-2008000-20088-0) funded by the National Research Foundation (NRF) of the Korean government (MEST), and by the Gwangju Institute of Science and Technology (GIST) institutional research program.
142
J. An et al. / Bioresource Technology 127 (2013) 138–142
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.biortech.2012.09. 095. References An, J., Moon, H., Chang, I.S., 2010. Multiphase electrode microbial fuel cell system that simultaneously converts organics coexisting in water and sediment phases to electricity. Environ. Sci. Technol. 44 (18), 7145–7150. An, J., Jeon, H., Lee, J., Chang, I.S., 2011. Bifunctional silver nanoparticle cathode in microbial fuel cells for microbial growth inhibition with comparable oxygen reduction reaction activity. Environ. Sci. Technol. 45 (12), 5441–5446. Bard, A.J., Faulkner, A.R., 2001. Electrochemical Methods: Fundamentals and Application, second ed. Wiley & Sons, New York (Chapter 3). Chiou, C.T., Kile, D.E., 2000. Contaminant sorption by soil and bed sediment; U.S. Department of the Interior, U.S. Department of the Interior, U.S. Geological Survey: Reston, VA. DiChristina, T.J., 1992. Effects of nitrogen and nitrite in dissimilatory ion reduction by Shewanella putrefaciens 200. J. Bacteriol. 174 (6), 1891–1896. Froelich, P.N., Klinkhammer, G.P., Bender, M.L., Luedtke, N.A., Heath, G.R., Cullen, D., Dauphin, P., Hammond, D., Hartman, B., Maynard, V., 1979. Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis. Geochim. Cosmochim. Acta 43 (7), 1075–1090. Garrels, R.M., Christ, C.L., 1990. Solutions, Minerals and Equilibria. Jones and Bartlett, London. Häggblom, M.M., Bossert, I.D. (Eds.), 2003. Dehalogenation: Microbial Processes and Environmental Applications. Kluwer Academic Publishers, Boston, MA, pp. 42– 48. Holmes, D.E., Nicoll, J.S., Bond, D.R., Lovley, D.R., 2004. Potential role of a novel psychrotolerant member of the family Geobacteraceae, Geopsychrobacter electrodiphilus gen. nov., sp nov., in electricity production by a marine sediment fuel cell. Appl. Environ. Microbiol. 70 (10), 6023–6030.
Kim, D.H., Matsuda, O., Yamamoto, T., 1997. Nitrification, denitrification and nitrate reduction rates in the sediment of Hirosima Bay. Japan. J. Oceanogr. 55 (4), 317– 324. Londry, K.L., Suflita, J.M., 1999. Use of nitrate to control sulfide generation by sulfate-reducing bacterial associated with oily waste. J. Ind. Microbiol. Biotechnol. 22 (6), 582–589. Lowy, D.A., Tender, L.M., 2008. Harvesting energy from the marine sediment-water interface III. Kinetic activity of quinone- and antimony-based anode materials. J. Power Sources 185 (1), 70–75. Nealson, K.H., Stahl, D.A., 1997. Microorganism and biogeochemical cycles: what can we learn from layered microbial communityes? In: Nealson, K.H., Banfield, J.F. (Eds.), Geomicrobiology: Interactions between Microbes and Minerals, 35. Mineralogical Society of America, Washington, D.C, pp. 5–34. Nealson, K.H., Myers, C.R., 1992. Microbial reduction of manganese and Iron: new approaches to carbon cycling. Appl. Environ. Microbiol. 58 (2), 439–443. Rezaei, F., Richard, T.L., Brennan, R.A., Logan, B.E., 2007. Substrate-enhanced microbial fuel cells for improved remote power generation from sedimentbased systems. Environ. Sci. Technol. 41 (11), 4053–4058. Shantaram, A., Beyenal, H., Raajan, R., Veluchamy, A., Lewandowski, Z., 2005. Wireless sensors powered by microbial fuel cells. Environ. Sci. Technol. 39 (13), 5037–5042. Song, T.S., Yan, Z.S., Zhao, Z.W., Jiang, H.L., 2010. Removal of organic matter in freshwater sediment by microbial fuel cells at various external resistances. J. Chem. Technol. Biotechno. 85 (11), 1489–1493. Tsertova, N., Kisand, A., Tammert, H., Kisand, V., 2011. Low seasonal variability in community composition of sediment bacteria in large and shallow lake. Environ. Microbiol. Rep. 3 (2), 270–277. Tender, L.M., Gray, S.A., Groveman, E., Lowy, D.A., Kauffman, P., Melhado, J., Tyce, R.C., Flynn, D., Petrecca, R., Dobarro, J., 2008. The first demonstration of a microbial fuel cell as a viable power supply: powering a meteorological buoy. J. Power Sources 179 (2), 571–575. Wang, Y., Francis, A.J., 2005. Evaluation of microbial activity for long-term assessment of deep geologic nuclear waste repositories. J. Nucl. Radiochem. Sci. 6 (1), 43–50.