- Email: [email protected]
Increased power density from a spiral wound microbial fuel cell
Biosensors and Bioelectronics 41 (2013) 894–897
Contents lists available at SciVerse ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Short communication
Increased power density from a spiral wound microbial fuel cell Boyang Jia 1, Dawei Hu 1, Beizhen Xie, Kun Dong, Hong Liu n Laboratory of Environmental Biology and Life Support Technology, School of Biological Science and Medical Engineering, Beihang University, Beijing 100191, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 July 2012 Received in revised form 18 September 2012 Accepted 25 September 2012 Available online 5 October 2012
Using Microbial fuel cell (MFC) to convert organic and inorganic matter into electricity is of great interest for powering portable devices, which is now still limited by the output of MFC. In this study, a spiral wound MFC (SWMFC) with relatively large volume normalized surface area of separator (4.2 cm2/ml) was fabricated to enhance power generation. Compared with double-membrane MFC (DMMFC) and conventional double chamber MFC (DCMFC), the power density of SWMFC increased by 42% and 99% resulted from its lower internal resistance. Besides larger separator area, the better performance of SWMFC benefited from its structure sandwiching the cathodes between two separators. This point was proved again by a comparison of another DCMFC and a triple chamber MFC (TCMFC) as well as a simulation using finite element method. Moreover, the feature of SWMFC was more convenient and compact to scale up. Therefore, SWMFC provides a promising configuration for high power output as a portable power source. & 2012 Elsevier B.V. All rights reserved.
Keywords: Microbial fuel cell Spiral wound Mathematical model
1. Introduction A microbial fuel cell (MFC) is a novel energy technology which could generate electricity from various organic matters (Logan et al., 2006; Osman et al., 2010, 2011). Powering small portable appliances, for example BOD sensors, is one of the potential applications of MFC. While the power output of MFC needs to be enhanced in limited space (Chang et al., 2005; Shantaram et al., 2005; Willner, 2002). The configuration of MFC can be diversified (Logan et al., 2006), such as cubic structures (Liu and Logan, 2005a), tubular structures (He et al., 2006) and so on. The power output might be further enhanced by improving the configuration of MFC. Increasing the area of separator and electrodes in a constant volume was an effective way to improve the power output (Oh and Logan, 2006). For instance, separator and electrode were often placed at one end in most of the cubic reactors; however the power density was doubled by adding another separator electrode assembly at the other end of the cubic reactors (Fan et al., 2007, Zhang et al., 2009). But in most of the current configurations, the dimensions of reactors will be enlarged greatly with the increase of separator area, hampering the proportional growth of volumetric power density. In this study, a spiral wound structure, which had been applied successfully in lithium-ion cell (also called Swiss-roll construction) (Pletcher and Walsh, 1993), was utilized in microbial fuel cell. This
n
Corresponding author. Tel./fax: þ 86 10 82339837. E-mail address: [email protected] (H. Liu). 1 Both authors contributed equally to this work.
0956-5663/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bios.2012.09.051
special MFC configuration was able to enhance the power output by decreasing its internal resistance. In addition, the feature of SWMFC was more convenient and compact to scale up and provided a promising configuration as a portable power source.
2. Materials and methods 2.1. MFC configuration Carbon cloth anodes were (Q-EFB-30-075D, Nice Special Carbon Corporation, China) treated by ammonia gas (Cheng and Logan, 2007), and cathodes were made of carbon paper (HCP135, Hesen Corporation, China) with 0.5 mg/cm2 platinum (Cheng et al., 2005). Cation exchange membrane (CEM) (CMI-7000, Membrane International, USA) was served as a separator. The SWMFC was constructed as in Fig. 1. Anode of 19 cm (L) 2.6 cm (H) was sandwiched between the two separators with the dimensions of 22 cm (L) 4 cm (H) and 20 cm (L) 4 cm (H), respectively. The bottom and the top of the two separators were plugged up with silicone strips in size of 0.5 cm (W) 0.5 cm (H). Two tubes were sited at both ends of the anode chamber as the inlet and outlet, respectively. Four pieces of carbon paper with 1.2 cm (W) 3.4 cm (H) were inserted into the cathode chamber which was formed by winding the two separators. To seal the bottom the SWMFC was immersed in 0.5 cm deep adhesive (Silicone rubber, Nanda 704, China); and then the top was sealed by pouring the adhesive into the adhesive channel (Fig. 1a) enclosed by two separators and silicone strip. This adhesive would
B. Jia et al. / Biosensors and Bioelectronics 41 (2013) 894–897
895
was the projected surface area of the cathode electrode. Power was calculated as P (W) ¼V2/Rex and power density was respectively normalized by the projected surface area of the cathode (mW/m2) and by the liquid volume (W/m3). The polarization curve was obtained according to the reference (Logan et al., (2006)). Impedance measurements (Zennium, Zahner, Germany) were conducted at the open circuit voltage over a frequency range of 106–10 1 Hz with a sinusoidal perturbation of 5 mV amplitude. The analysis of resistances herein was according to the reference (He et al., 2006). Conductivity of PBS was measured by conductivity meters (HI 8633, HANNA, Italy). Digital simulation was carried out by means of MatLab/PDE toolbox using finite element method (Chandrupatla and Belegundu 2002).
3. Results and discussion 3.1. Comparison of electricity generation and internal resistance
Fig. 1. Schematic of the spiral wound MFC (SWMFC): sectional view (a), stereogram (b), top view with dimension (c) and graph (d).
coat on silicon strips and seal the gap between the separator and silicon strips. A double-membrane MFC (DMMFC) was designed for the comparison of SWMFC and conventional MFC. It could be regarded as the unfolded type of SWMFC. Anode was placed into the channel composed by two identical separators, and then silicone strip and adhesive were used to seal around the channel. Afterwards, the whole part was immersed into a sink, and four pieces of cathode were inserted into the gap between the separator and the sink (Fig. S1 a and b). A double chamber MFC (DCMFC) was constructed according to the configuration described in the reference (Rabaey et al., 2005) (Fig. S1 c and d). These three MFCs (SWMFC, DMMFC and DCMFC) were regarded as group 1 and all physical parameters were shown in Table S1. To further interpret the advantages of SWMFC, group 2, which was comprised of a DCMFC2 and a triple chamber MFC (TCMFC), were constructed (Fig. S1 e and f). In TCMFC, anode was cut into two equal parts and placed into different anode chambers, respectively. One half of the CEM surfaces (two pieces) were coated with adhesive (Silicone rubber, Nanda 704, China) to keep the same separator area with DCMFC2. 2.2. Operation condition All the anodes were inoculated with 10 mL effluent from a well operated MFC, which was initially inoculated by anaerobic sludge from a domestic wastewater treatment plant (Beijing, China). Then acetate (1 g/L) in 50 mM phosphate buffer solution (PBS) (Logan et al., 2007) containing mineral (12.5 mL/L) and vitamin (5 mL/L) solutions (Balch et al., 1979) were fed into the anode chambers. 50 mM PBS was used as the catholyte. Aeration was used to supply dissolved oxygen in the cathode chambers. All the reactors were operated as batch mode under ambient temperature (25 1C72 1C) conditions with 1000 O resistance. Electrolyte was refilled every 40 h in group 1 and 96 h in group 2. 2.3. Measurement and calculation The output voltage (V, mV) of all the MFCs was monitored by data acquisition system (USB-1608FS, Measurement Computing Corporation, USA). Current density was calculated as j (A/m2) ¼ V/(RexAcat), where Rex (O) was the external resistance, and Acat (m2)
During the steady-state in group 1 the maximum output voltages which lasted for at least 24 h of SWMFC, DMMFC and DCMFC were similar (0.47 V, 0.44 V and 0.43 V, respectively) (Fig. 2a). However, the maximum power density of SWMFC (77.3 W/m3 and 1414 mW/m2) was much higher than that of DMMFC (54.4 W/m3 and 995 mW/m2) and DCMFC (38.8 W/m3 and 709 mW/m2) (Fig. 2b). The maximum power density of DMMFC was 40.2% higher than that of DCMFC. The insufficient area of separator has been reported as one of the limiting factor for MFC performance (Oh and Logan, 2006). In our case, the fact that the separator area of DMMFC (139 cm2) was 7 times larger than that of DCMFC (19.6 cm2) might be the main reason for the distinction of maximum power density. In addition, the DMMFC was assembled according to the following configuration: cathode—separator–anode—separator—cathode (Fig. S1b). Similar MFC configuration has been reported for the ability of lowering the internal resistance and increasing power density (Fan et al., 2007). Therefore, the large area of separator and the special configuration contributed mainly for the higher power density. Moreover, in the DCMFC of this case, the anode was folded, which would make the actual effective area of electrode not as large as the other two. It might also result in the lower power density. Although the separator areas of SWMFC and DMMFC were almost equal and the anode chambers were all enclosed by two separators, the maximum power density of the former was 42.9% higher than that of the latter. Thus, the improvement of power density might result from its spiral wound structure, in which the cathodes were sandwiched between two separators (Fig. 1c). Group 2 was established to exam the effect of this structure on power density accordingly. The physical parameters of DCMFC2 and TCMFC were almost identical except that the cathode was also enclosed by two separators in TCMFC (Table S1 and Fig S1 e and f). It was shown that the maximum power density of TCMFC (27.8 W/m3 and 101.7 mW/m2) was 39% higher than that of DCMFC2 (20 W/m3 and 73 mW/m2). Consequently, it is reasonable to believe that the configuration as SWMFC and TCMFC could enhance the power density by sandwiching the cathode between separators. Internal resistances of the MFCs were measured to investigate the cause of higher power density in SWMFC. The Nyquist plot showed that diffusion and charge-transfer resistances played a similar role in all MFCs. However, there was great diversity among ohmic resistance, including the resistance caused by electrolyte, separator and electrode (Fig. S2 and Table S1). For group 1, the ohmic resistance of DMMFC was 44% lower than that of DCMFC, which might be owing to the larger separators area in DMMFC (He et al. 2006; Oh and Logan, 2006). While for SWMFC and DMMFC, with similar separator and electrode areas,
896
B. Jia et al. / Biosensors and Bioelectronics 41 (2013) 894–897
cathode between two separators, as in SWMFC and TCMFC, could increase the power density by decreasing the ohmic resistance. 3.2. Modeling of the different MFCs To further confirm the influence of the structure on internal resistance, their respective internal resistance was calculated using finite element method. In our case, two assumptions were applied to simplify the approach: the released and accepted amounts of protons in anode and cathode were equal per unit area and per unit time, respectively; separators were regarded as a solution which conductivities were different with PBS. The kinetic equation of the current density (j) in MFC could be expressed as follows: @j=@t ¼ rj þ Q
ð1Þ
where Q is a current source, r is a hamilton operator. j is related to the electric field (E) and electric potential (V) through: j ¼ sE ¼ s rV
ð2Þ
where s is conductivity of electrolyte solution in MFC. Substituting Eq. (2) into Eq. (1) yield: 2
@j=@t ¼ s r V þ Q
ð3Þ
The potential distribution in MFC with the steady operation could be derived in terms of the elliptic Poisson’s equation: 2
s r V þ Q ¼ 0:
ð4Þ
Referring to Ohm’s law and direct derivation, the spatial distribution of internal resistance (R) in MFC could be specified as follows: Rðx,yÞ ¼ rV=j:
ð5Þ
Hence, the ohmic resistance of each MFC (ri) could be calculated as follows: Z Z ri ¼ rdxdy, ð6Þ Oi
where Oi indicated the specific geometrical structure of each MFC. Digital simulation was carried out to investigate the distribution of j and V in different MFCs, and numerically integrated R to obtain internal resistance (some details were presented in Fig. S3). The triangular mesh with 105 grids was created and refined via adaptive refinement to solve the Eqs. (2) and (4), and the simulation results of group 1 were shown in Fig. 3. The j and V were calculated and expressed as currentlines of the vector field of j and the equipotential lines of V. As is well known, the electric flux line is originated from anode and terminated on cathode, its density is directly proportional to field intensity or potential gradient. Specific structure of SWMFC resulted in the electric flux lines distributed unevenly in inner space of SWMFC, i.e. the
Fig. 2. Voltage generation with 1000 O (a), power density (b) and I–V curve(c) measured by changing the circuit external resistances.
the ohmic resistance of the former was 29% lower than that of the latter. This decrease in SWMFC might mainly result from electrolyte resistance, but it was uncertain whether the decrease was benefit from its special structure. It was because the electrode spacing could affect the electrolyte resistance as well (Liu et al., 2005b). This distance was difficult to control in SWMFC for the anode was a curve surface. However, it could be regulated precisely in group 2. The result in Table S1 showed that the ohmic resistance of TCMFC was 19 O less than that of DCMFC2. It proved that sandwiching the
Fig. 3. Simulation results of group 1 by finite element method with MatLab/PDE toolbox.
B. Jia et al. / Biosensors and Bioelectronics 41 (2013) 894–897
electric flux lines concentrated in the regions of spiral wound anode near the cathode, hence the field intensity or potential gradient in these regions were larger than those in the regions relatively far from the cathode and vise versa. Based on simulation results as well as Eqs. (5) and (6), the ohmic resistance of each MFC could be obtained from double spline interpolation and integration. The results showed similar trends to experimental values (ri in Table S1) and indicated the structure sandwiching cathode between separators could decrease the ohmic resistance.
897
special electrode distribution (cathodes were sandwiched between the separators) which decreased the ohmic resistance. Moreover, the SWMFC was more compact and suitable to scale up than the other two configurations.
Acknowledgments This work was supported by grants from the National Nature Science Foundation of China (30970716).
3.3. Characteristics of SWMFC Appendix A. Supporting information Dimensional change is a significant issue as scaling up MFCs, especially portalbe MFCs. But it could be ignored in SWMFC compared with the other two configurations. For instance, if the liquid volumes of anode are enlarged from 30 mL to 60 mL at fixed heights and widths, the lengths will be extended by 6 cm in DCMFC and 15 cm in DMMFC. However, the diameter will only be increased by 0.5 cm in SWMFC. Moreover, the length/height and length/width ratios of DMMFC reached to 5.5 and 44, respectively. These ratios might lead to the waste of space and the instability of the structure. On contrary, the diameter/height ratio was only about 1 in the SWMFC. Therefore, the SWMFC was more compact and suitable to scale up. However, several challenges should be overcome to improve SWMFC. The cathode made from flexible material could be adopted because carbon paper could not be bended and attached to the separator which may influence the distance between electrodes. A method prepared cathode by electrochemically depositing Pt nanoparticles on a CNT-textile may be a way to resolve this problem (Xie et al., 2011). Moreover, the dimensions of handmade SWMFC could not be controlled precisely. The distance between two separators might be varied in the rolling process because of the deformations of separators, which could raise concerns about leaking. Some kinds of plane filler (for example, rubber pad) may be helpful during manufacturing process to control the deformations. In addition, the material of separator should be optimized in next work to lower the cost and improve the performance of MFC. J-cloth applied in a MFC with double cloth electrode assemblies MFC provided a promising choice in SWMFC (Fan et al., 2012).
4. Conclusions In this paper, the SWMFC was investigated as a potential configuration for improving the power output and the scaling up capacity. Compared with DMMFC and DCMFC, the volumetric power density of SWMFC was increased by 42% and 99%. It was mainly caused by that the relatively larger separator area and
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2012.09.051.
References Balch, W., Fox, G., Magrum, L., Woese, C., Wolfe, R., 1979. Microbiological Reviews 43, 260–296. Chandrupatla, T.R., Belegundu, A.D., 2002. Introduction to Finite Elements in Engineering, third ed. Prentice Hall, New Jersey. Chang, I.S., Moon, H., Jang, J.K., Kim, B.H., 2005. Biosensors and Bioelectronics 20, 1856–1859. Cheng, S., Liu, H., Logan, B., 2005. Environmental Science and Technology 40, 364–369. Cheng, S., Logan, B., 2007. Electrochemistry Communications 9, 492–496. Fan, Y., Hu, H., Liu, H., 2007. Journal of Power Sources 171, 348–354. Fan, Y., Han, S.-K., Liu, H., 2012. Energy and Environmental Science 5 (8), 8273. He, Z., Wagner, N., Minteer, S.D., Angenent, L.T., 2006. Environmental Science and Technology 40, 5212–5217. Liu, H., Logan, B., 2005a. Environmental Science and Technology 38, 4040–4046. Liu, H., Cheng, S., Logan, B., 2005b. Environmental Science and Technology 39, 658–662. Logan, B., Cheng, S., Watson, V., Estadt, G., 2007. Environmental Science and Technology 41, 3341–3346. Logan, B., Hamelers, B., Rozendal, R., Schroder, U., Keller, J., Freguia, S., Aelterman, P., Verstraete, W., Rabaey, K., 2006. Environmental Science and Technology 40, 5181–5192. Oh, S.E., Logan, B., 2006. Applied Microbiology and Biotechnology 70, 162–169. Osman, M.H., Shah, A.A., Walsh, F.C., 2010. Biosensors and Bioelectronics 26, 953–963. Osman, M.H., Shah, A.A., Walsh, F.C., 2011. Biosensors and Bioelectronics 26, 3087–3102. Pletcher, D., Walsh, F.C., 1993. Industrial Electrochemistry, senond ed. Blackie Academic & Professional. ¨ Rabaey, K., Boon, N., Hofte, M., Verstraete, W., 2005. Environmental Science and Technology 39, 3401–3408. Shantaram, A., Beyenal, H., Veluchamy, R.R.A., Lewandowski, Z., 2005. Environmental Science and Technology 39, 5037–5042. Willner, I., 2002. Science 298, 2407–2408. Xie, X., Pasta, M., Hu, L., Yang, Y., McDonough, J., Cha, J., Criddle, C.S., Cui, Y., 2011. Energy and Environmental Science 4, 4. Zhang, X., Cheng, S., Wang, X., Huang, X., Logan, B., 2009. Environmental Science and Technology 43, 8456–8461.
Contents lists available at SciVerse ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Short communication
Increased power density from a spiral wound microbial fuel cell Boyang Jia 1, Dawei Hu 1, Beizhen Xie, Kun Dong, Hong Liu n Laboratory of Environmental Biology and Life Support Technology, School of Biological Science and Medical Engineering, Beihang University, Beijing 100191, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 July 2012 Received in revised form 18 September 2012 Accepted 25 September 2012 Available online 5 October 2012
Using Microbial fuel cell (MFC) to convert organic and inorganic matter into electricity is of great interest for powering portable devices, which is now still limited by the output of MFC. In this study, a spiral wound MFC (SWMFC) with relatively large volume normalized surface area of separator (4.2 cm2/ml) was fabricated to enhance power generation. Compared with double-membrane MFC (DMMFC) and conventional double chamber MFC (DCMFC), the power density of SWMFC increased by 42% and 99% resulted from its lower internal resistance. Besides larger separator area, the better performance of SWMFC benefited from its structure sandwiching the cathodes between two separators. This point was proved again by a comparison of another DCMFC and a triple chamber MFC (TCMFC) as well as a simulation using finite element method. Moreover, the feature of SWMFC was more convenient and compact to scale up. Therefore, SWMFC provides a promising configuration for high power output as a portable power source. & 2012 Elsevier B.V. All rights reserved.
Keywords: Microbial fuel cell Spiral wound Mathematical model
1. Introduction A microbial fuel cell (MFC) is a novel energy technology which could generate electricity from various organic matters (Logan et al., 2006; Osman et al., 2010, 2011). Powering small portable appliances, for example BOD sensors, is one of the potential applications of MFC. While the power output of MFC needs to be enhanced in limited space (Chang et al., 2005; Shantaram et al., 2005; Willner, 2002). The configuration of MFC can be diversified (Logan et al., 2006), such as cubic structures (Liu and Logan, 2005a), tubular structures (He et al., 2006) and so on. The power output might be further enhanced by improving the configuration of MFC. Increasing the area of separator and electrodes in a constant volume was an effective way to improve the power output (Oh and Logan, 2006). For instance, separator and electrode were often placed at one end in most of the cubic reactors; however the power density was doubled by adding another separator electrode assembly at the other end of the cubic reactors (Fan et al., 2007, Zhang et al., 2009). But in most of the current configurations, the dimensions of reactors will be enlarged greatly with the increase of separator area, hampering the proportional growth of volumetric power density. In this study, a spiral wound structure, which had been applied successfully in lithium-ion cell (also called Swiss-roll construction) (Pletcher and Walsh, 1993), was utilized in microbial fuel cell. This
n
Corresponding author. Tel./fax: þ 86 10 82339837. E-mail address: [email protected] (H. Liu). 1 Both authors contributed equally to this work.
0956-5663/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bios.2012.09.051
special MFC configuration was able to enhance the power output by decreasing its internal resistance. In addition, the feature of SWMFC was more convenient and compact to scale up and provided a promising configuration as a portable power source.
2. Materials and methods 2.1. MFC configuration Carbon cloth anodes were (Q-EFB-30-075D, Nice Special Carbon Corporation, China) treated by ammonia gas (Cheng and Logan, 2007), and cathodes were made of carbon paper (HCP135, Hesen Corporation, China) with 0.5 mg/cm2 platinum (Cheng et al., 2005). Cation exchange membrane (CEM) (CMI-7000, Membrane International, USA) was served as a separator. The SWMFC was constructed as in Fig. 1. Anode of 19 cm (L) 2.6 cm (H) was sandwiched between the two separators with the dimensions of 22 cm (L) 4 cm (H) and 20 cm (L) 4 cm (H), respectively. The bottom and the top of the two separators were plugged up with silicone strips in size of 0.5 cm (W) 0.5 cm (H). Two tubes were sited at both ends of the anode chamber as the inlet and outlet, respectively. Four pieces of carbon paper with 1.2 cm (W) 3.4 cm (H) were inserted into the cathode chamber which was formed by winding the two separators. To seal the bottom the SWMFC was immersed in 0.5 cm deep adhesive (Silicone rubber, Nanda 704, China); and then the top was sealed by pouring the adhesive into the adhesive channel (Fig. 1a) enclosed by two separators and silicone strip. This adhesive would
B. Jia et al. / Biosensors and Bioelectronics 41 (2013) 894–897
895
was the projected surface area of the cathode electrode. Power was calculated as P (W) ¼V2/Rex and power density was respectively normalized by the projected surface area of the cathode (mW/m2) and by the liquid volume (W/m3). The polarization curve was obtained according to the reference (Logan et al., (2006)). Impedance measurements (Zennium, Zahner, Germany) were conducted at the open circuit voltage over a frequency range of 106–10 1 Hz with a sinusoidal perturbation of 5 mV amplitude. The analysis of resistances herein was according to the reference (He et al., 2006). Conductivity of PBS was measured by conductivity meters (HI 8633, HANNA, Italy). Digital simulation was carried out by means of MatLab/PDE toolbox using finite element method (Chandrupatla and Belegundu 2002).
3. Results and discussion 3.1. Comparison of electricity generation and internal resistance
Fig. 1. Schematic of the spiral wound MFC (SWMFC): sectional view (a), stereogram (b), top view with dimension (c) and graph (d).
coat on silicon strips and seal the gap between the separator and silicon strips. A double-membrane MFC (DMMFC) was designed for the comparison of SWMFC and conventional MFC. It could be regarded as the unfolded type of SWMFC. Anode was placed into the channel composed by two identical separators, and then silicone strip and adhesive were used to seal around the channel. Afterwards, the whole part was immersed into a sink, and four pieces of cathode were inserted into the gap between the separator and the sink (Fig. S1 a and b). A double chamber MFC (DCMFC) was constructed according to the configuration described in the reference (Rabaey et al., 2005) (Fig. S1 c and d). These three MFCs (SWMFC, DMMFC and DCMFC) were regarded as group 1 and all physical parameters were shown in Table S1. To further interpret the advantages of SWMFC, group 2, which was comprised of a DCMFC2 and a triple chamber MFC (TCMFC), were constructed (Fig. S1 e and f). In TCMFC, anode was cut into two equal parts and placed into different anode chambers, respectively. One half of the CEM surfaces (two pieces) were coated with adhesive (Silicone rubber, Nanda 704, China) to keep the same separator area with DCMFC2. 2.2. Operation condition All the anodes were inoculated with 10 mL effluent from a well operated MFC, which was initially inoculated by anaerobic sludge from a domestic wastewater treatment plant (Beijing, China). Then acetate (1 g/L) in 50 mM phosphate buffer solution (PBS) (Logan et al., 2007) containing mineral (12.5 mL/L) and vitamin (5 mL/L) solutions (Balch et al., 1979) were fed into the anode chambers. 50 mM PBS was used as the catholyte. Aeration was used to supply dissolved oxygen in the cathode chambers. All the reactors were operated as batch mode under ambient temperature (25 1C72 1C) conditions with 1000 O resistance. Electrolyte was refilled every 40 h in group 1 and 96 h in group 2. 2.3. Measurement and calculation The output voltage (V, mV) of all the MFCs was monitored by data acquisition system (USB-1608FS, Measurement Computing Corporation, USA). Current density was calculated as j (A/m2) ¼ V/(RexAcat), where Rex (O) was the external resistance, and Acat (m2)
During the steady-state in group 1 the maximum output voltages which lasted for at least 24 h of SWMFC, DMMFC and DCMFC were similar (0.47 V, 0.44 V and 0.43 V, respectively) (Fig. 2a). However, the maximum power density of SWMFC (77.3 W/m3 and 1414 mW/m2) was much higher than that of DMMFC (54.4 W/m3 and 995 mW/m2) and DCMFC (38.8 W/m3 and 709 mW/m2) (Fig. 2b). The maximum power density of DMMFC was 40.2% higher than that of DCMFC. The insufficient area of separator has been reported as one of the limiting factor for MFC performance (Oh and Logan, 2006). In our case, the fact that the separator area of DMMFC (139 cm2) was 7 times larger than that of DCMFC (19.6 cm2) might be the main reason for the distinction of maximum power density. In addition, the DMMFC was assembled according to the following configuration: cathode—separator–anode—separator—cathode (Fig. S1b). Similar MFC configuration has been reported for the ability of lowering the internal resistance and increasing power density (Fan et al., 2007). Therefore, the large area of separator and the special configuration contributed mainly for the higher power density. Moreover, in the DCMFC of this case, the anode was folded, which would make the actual effective area of electrode not as large as the other two. It might also result in the lower power density. Although the separator areas of SWMFC and DMMFC were almost equal and the anode chambers were all enclosed by two separators, the maximum power density of the former was 42.9% higher than that of the latter. Thus, the improvement of power density might result from its spiral wound structure, in which the cathodes were sandwiched between two separators (Fig. 1c). Group 2 was established to exam the effect of this structure on power density accordingly. The physical parameters of DCMFC2 and TCMFC were almost identical except that the cathode was also enclosed by two separators in TCMFC (Table S1 and Fig S1 e and f). It was shown that the maximum power density of TCMFC (27.8 W/m3 and 101.7 mW/m2) was 39% higher than that of DCMFC2 (20 W/m3 and 73 mW/m2). Consequently, it is reasonable to believe that the configuration as SWMFC and TCMFC could enhance the power density by sandwiching the cathode between separators. Internal resistances of the MFCs were measured to investigate the cause of higher power density in SWMFC. The Nyquist plot showed that diffusion and charge-transfer resistances played a similar role in all MFCs. However, there was great diversity among ohmic resistance, including the resistance caused by electrolyte, separator and electrode (Fig. S2 and Table S1). For group 1, the ohmic resistance of DMMFC was 44% lower than that of DCMFC, which might be owing to the larger separators area in DMMFC (He et al. 2006; Oh and Logan, 2006). While for SWMFC and DMMFC, with similar separator and electrode areas,
896
B. Jia et al. / Biosensors and Bioelectronics 41 (2013) 894–897
cathode between two separators, as in SWMFC and TCMFC, could increase the power density by decreasing the ohmic resistance. 3.2. Modeling of the different MFCs To further confirm the influence of the structure on internal resistance, their respective internal resistance was calculated using finite element method. In our case, two assumptions were applied to simplify the approach: the released and accepted amounts of protons in anode and cathode were equal per unit area and per unit time, respectively; separators were regarded as a solution which conductivities were different with PBS. The kinetic equation of the current density (j) in MFC could be expressed as follows: @j=@t ¼ rj þ Q
ð1Þ
where Q is a current source, r is a hamilton operator. j is related to the electric field (E) and electric potential (V) through: j ¼ sE ¼ s rV
ð2Þ
where s is conductivity of electrolyte solution in MFC. Substituting Eq. (2) into Eq. (1) yield: 2
@j=@t ¼ s r V þ Q
ð3Þ
The potential distribution in MFC with the steady operation could be derived in terms of the elliptic Poisson’s equation: 2
s r V þ Q ¼ 0:
ð4Þ
Referring to Ohm’s law and direct derivation, the spatial distribution of internal resistance (R) in MFC could be specified as follows: Rðx,yÞ ¼ rV=j:
ð5Þ
Hence, the ohmic resistance of each MFC (ri) could be calculated as follows: Z Z ri ¼ rdxdy, ð6Þ Oi
where Oi indicated the specific geometrical structure of each MFC. Digital simulation was carried out to investigate the distribution of j and V in different MFCs, and numerically integrated R to obtain internal resistance (some details were presented in Fig. S3). The triangular mesh with 105 grids was created and refined via adaptive refinement to solve the Eqs. (2) and (4), and the simulation results of group 1 were shown in Fig. 3. The j and V were calculated and expressed as currentlines of the vector field of j and the equipotential lines of V. As is well known, the electric flux line is originated from anode and terminated on cathode, its density is directly proportional to field intensity or potential gradient. Specific structure of SWMFC resulted in the electric flux lines distributed unevenly in inner space of SWMFC, i.e. the
Fig. 2. Voltage generation with 1000 O (a), power density (b) and I–V curve(c) measured by changing the circuit external resistances.
the ohmic resistance of the former was 29% lower than that of the latter. This decrease in SWMFC might mainly result from electrolyte resistance, but it was uncertain whether the decrease was benefit from its special structure. It was because the electrode spacing could affect the electrolyte resistance as well (Liu et al., 2005b). This distance was difficult to control in SWMFC for the anode was a curve surface. However, it could be regulated precisely in group 2. The result in Table S1 showed that the ohmic resistance of TCMFC was 19 O less than that of DCMFC2. It proved that sandwiching the
Fig. 3. Simulation results of group 1 by finite element method with MatLab/PDE toolbox.
B. Jia et al. / Biosensors and Bioelectronics 41 (2013) 894–897
electric flux lines concentrated in the regions of spiral wound anode near the cathode, hence the field intensity or potential gradient in these regions were larger than those in the regions relatively far from the cathode and vise versa. Based on simulation results as well as Eqs. (5) and (6), the ohmic resistance of each MFC could be obtained from double spline interpolation and integration. The results showed similar trends to experimental values (ri in Table S1) and indicated the structure sandwiching cathode between separators could decrease the ohmic resistance.
897
special electrode distribution (cathodes were sandwiched between the separators) which decreased the ohmic resistance. Moreover, the SWMFC was more compact and suitable to scale up than the other two configurations.
Acknowledgments This work was supported by grants from the National Nature Science Foundation of China (30970716).
3.3. Characteristics of SWMFC Appendix A. Supporting information Dimensional change is a significant issue as scaling up MFCs, especially portalbe MFCs. But it could be ignored in SWMFC compared with the other two configurations. For instance, if the liquid volumes of anode are enlarged from 30 mL to 60 mL at fixed heights and widths, the lengths will be extended by 6 cm in DCMFC and 15 cm in DMMFC. However, the diameter will only be increased by 0.5 cm in SWMFC. Moreover, the length/height and length/width ratios of DMMFC reached to 5.5 and 44, respectively. These ratios might lead to the waste of space and the instability of the structure. On contrary, the diameter/height ratio was only about 1 in the SWMFC. Therefore, the SWMFC was more compact and suitable to scale up. However, several challenges should be overcome to improve SWMFC. The cathode made from flexible material could be adopted because carbon paper could not be bended and attached to the separator which may influence the distance between electrodes. A method prepared cathode by electrochemically depositing Pt nanoparticles on a CNT-textile may be a way to resolve this problem (Xie et al., 2011). Moreover, the dimensions of handmade SWMFC could not be controlled precisely. The distance between two separators might be varied in the rolling process because of the deformations of separators, which could raise concerns about leaking. Some kinds of plane filler (for example, rubber pad) may be helpful during manufacturing process to control the deformations. In addition, the material of separator should be optimized in next work to lower the cost and improve the performance of MFC. J-cloth applied in a MFC with double cloth electrode assemblies MFC provided a promising choice in SWMFC (Fan et al., 2012).
4. Conclusions In this paper, the SWMFC was investigated as a potential configuration for improving the power output and the scaling up capacity. Compared with DMMFC and DCMFC, the volumetric power density of SWMFC was increased by 42% and 99%. It was mainly caused by that the relatively larger separator area and
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2012.09.051.
References Balch, W., Fox, G., Magrum, L., Woese, C., Wolfe, R., 1979. Microbiological Reviews 43, 260–296. Chandrupatla, T.R., Belegundu, A.D., 2002. Introduction to Finite Elements in Engineering, third ed. Prentice Hall, New Jersey. Chang, I.S., Moon, H., Jang, J.K., Kim, B.H., 2005. Biosensors and Bioelectronics 20, 1856–1859. Cheng, S., Liu, H., Logan, B., 2005. Environmental Science and Technology 40, 364–369. Cheng, S., Logan, B., 2007. Electrochemistry Communications 9, 492–496. Fan, Y., Hu, H., Liu, H., 2007. Journal of Power Sources 171, 348–354. Fan, Y., Han, S.-K., Liu, H., 2012. Energy and Environmental Science 5 (8), 8273. He, Z., Wagner, N., Minteer, S.D., Angenent, L.T., 2006. Environmental Science and Technology 40, 5212–5217. Liu, H., Logan, B., 2005a. Environmental Science and Technology 38, 4040–4046. Liu, H., Cheng, S., Logan, B., 2005b. Environmental Science and Technology 39, 658–662. Logan, B., Cheng, S., Watson, V., Estadt, G., 2007. Environmental Science and Technology 41, 3341–3346. Logan, B., Hamelers, B., Rozendal, R., Schroder, U., Keller, J., Freguia, S., Aelterman, P., Verstraete, W., Rabaey, K., 2006. Environmental Science and Technology 40, 5181–5192. Oh, S.E., Logan, B., 2006. Applied Microbiology and Biotechnology 70, 162–169. Osman, M.H., Shah, A.A., Walsh, F.C., 2010. Biosensors and Bioelectronics 26, 953–963. Osman, M.H., Shah, A.A., Walsh, F.C., 2011. Biosensors and Bioelectronics 26, 3087–3102. Pletcher, D., Walsh, F.C., 1993. Industrial Electrochemistry, senond ed. Blackie Academic & Professional. ¨ Rabaey, K., Boon, N., Hofte, M., Verstraete, W., 2005. Environmental Science and Technology 39, 3401–3408. Shantaram, A., Beyenal, H., Veluchamy, R.R.A., Lewandowski, Z., 2005. Environmental Science and Technology 39, 5037–5042. Willner, I., 2002. Science 298, 2407–2408. Xie, X., Pasta, M., Hu, L., Yang, Y., McDonough, J., Cha, J., Criddle, C.S., Cui, Y., 2011. Energy and Environmental Science 4, 4. Zhang, X., Cheng, S., Wang, X., Huang, X., Logan, B., 2009. Environmental Science and Technology 43, 8456–8461.