A double-chamber denitrifying microbial fuel cell (MFC), using boric acid-borate buffer solution as an alternative to phosphate buffer solution, was set up to investigate the influence of buffer solution concentration, temperature and external resistance on electricity generation and pollutant removal efficiency. The result revealed that the denitrifying MFC with boric acid-borate buffer solution was successfully started up in 51 days, with a stable cell voltage of 205.1 ± 1.96 mV at an external resistance of 50 Ω. Higher concentration of buffer solution favored nitrogen removal and electricity generation. The maximum power density of 8.27 W/m3 net cathodic chamber was obtained at a buffer solution concentration of 100 mmol/L. An increase in temperature benefitted electricity generation and nitrogen removal. A suitable temperature for this denitrifying MFC was suggested to be 25 °C. Decreasing the external resistance favored nitrogen removal and organic matter consumption by exoelectrogens.

INTRODUCTION

The microbial fuel cell (MFC) is considered to be a promising sustainable technology to effectively solve problems of energy shortage and water pollution since it can simultaneously facilitate electricity recovery and pollutant removal from wastewater (Du et al. 2007). As a promising technology for wastewater treatment, the MFC could be used to remove not only organic matter but also nitrogen from wastewater (Zhu et al. 2013). Clauwaert et al. (2007) first performed a complete cathodic denitrification without any power input in a denitrifying MFC (with biocathode), which demonstrated the feasibility of MFC being applied for simultaneous nitrate removal in the cathode and organic matter removal in the anode from wastewater.

Phosphate buffer solution (PBS) has been widely used in MFC studies to maintain a suitable pH for exoelectrogens and to increase conductivity of the solution (Clauwaert et al. 2007; Du et al. 2007). However, addition of high concentrations of PBS in MFC is expensive, especially for the application in wastewater treatment. Moreover, the phosphate would contribute to eutrophication of water bodies if it was discharged without being removed from the effluent. Therefore, it is imperative to find alternative buffer solutions for MFC. Several pH buffer solutions such as bicarbonate, 2-[N-morpholino] ethane sulfonate, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid and piperazine-N,N′-bis [2-ethanesulfonate] had been confirmed as effective alternatives to PBS in a single-chamber MFC with an air-cathode (Fan et al. 2007; Nam et al. 2010). Boric acid-borate solution is another potential alternative to PBS because it is chemically stable and does not interfere with biochemical reactions. Qiang et al. (2011) reported that MFC with an appropriate concentration of boric acid-borate buffer solution could greatly enhance the electron recovery rate compared to that without buffer solution.

So far, alternative buffer solutions to PBS were mostly studied in single-chamber MFC and have not been previously reported in denitrifying MFC. In this experiment, a double-chamber denitrifying MFC was started up with a boric acid-borate buffer solution. Afterward the impacts of buffer solution concentration, temperature and external resistance on the electricity generation characteristics and pollutant removal performance of denitrifying MFC were investigated.

MATERIALS AND METHODS

MFC construction

A denitrifying MFC was constructed with two rectangular chambers that served as anode and cathode chambers. A proton exchange membrane (Nafion 117, DuPont, Wilmington, Delaware, USA) was placed between the anode and the cathode chamber. Each chamber was filled with granular graphite (diameter: 2–6 mm, porosity: ∼50%) as electrode and inserted by a graphite rod. The eventual volumes of net cathodic and anodic chamber (NCC and NAC) were both 161 cm3. The cathode and anode rods were connected with a manual variable resistor to close the circuit. The pretreatment process of proton exchange membrane and granular graphite was stated previously (Li & Zhang 2012). The anodic and cathodic influent were, respectively, pumped into corresponding chambers at the same flow rate of 0.35 mL/min, which resulted in the same HRT of 7.7 h. Both anodic and cathodic liquids were recirculated at a rate of 20 mL/min for mixing. The experiment was carried out at 30 ± 0.5 °C unless specified.

Inoculation and synthetic wastewater

The anodic and cathodic chambers were, respectively, inoculated with anaerobic and anoxic sludge from Longwangzui wastewater treatment plant, Wuhan, China. The granular graphite was immersed in corresponding sludge for 24 h to absorb microbes before being loaded in the denitrifying MFC. The basic substrate of synthetic wastewaters fed to the anode and cathode consisted of MgSO4 · 7H2O (0.1 g/L), CaCl2 (0.015 g/L) and 1 mL/L trace elements as described by Rabaey et al. (2005). In addition, CH3COONa (chemical oxygen demand (COD) = 300 mg/L), NaNO3 (-N = 35 mg/L) and NaHCO3 (1 g/L) were used as an anodic electron donor, cathodic electron acceptor and inorganic carbon source, respectively. Unless specified, the anodic and cathodic influent pH was, respectively, maintained at 7.30 ± 0.05 and 7.65 ± 0.05 using 100 mmol/L boric acid-borate buffer solution.

Experimental protocol

During start-up of the denitrifying MFC, the external resistance was fixed at 100 Ω in the initial stage and adjusted to 50 Ω on Day 26. After stable electricity production and pollutant removal was achieved, the denitrifying MFC was run at varied buffer solution concentrations (0, 50 and 100 mmol/L). At each buffer solution concentration, the polarization curve and internal resistance were obtained under steady state, as evidenced by steady cell voltage being obtained (Watson & Logan 2011). The maximum current density obtained in the polarization curve test was estimated as the limiting current. Then the external resistance was set at a level where the maximum power density and limiting current being, respectively, achieved for at least 24 h, during which stable pollutant removal performance was tested. To optimize suitable temperature, the denitrifying MFC was in turn run at different temperatures (15, 20, 25, 30 and 35 °C), and the electrogenesis parameters and pollutant removal performance of denitrifying MFC were studied as above. Afterwards, the denitrifying MFC was run at a fixed temperature of 25 ± 0.5 °C and varied external resistances (200, 150, 100, 50, 25, 10 and 5 Ω). At each external resistance, the denitrifying MFC was run for at least 12 h during which stable electrogenesis parameters and pollutant removal performance were tested.

Analysis

COD, nitrate, nitrite and pH were determined according to Standard Methods (APHA 1992). All analyses were carried out in triplicate except in the start-up stage. With corresponding external resistance (R), the cell voltage (U) was recorded every 4 min using a UT71D intelligent digital multimeter (Uni-Trend Technology Co., Ltd, Dongguan, China). The anode potential was monitored with an Ag/AgCl reference electrode (assumed to be +0.197 V versus Standard Hydrogen Electrode, SHE). All potentials were reported versus SHE. The current (I) and power (P) were calculated as follows: I = U/R, P = I2R, and normalized by the NCC volume. Polarization curves were obtained by measuring the stable voltage and anode potential with the external resistances being varied from 9,000to 5 Ω. Anodic Coulombic efficiency was calculated as a ratio of the total Coulombs (calculated based on the current) to the theoretical amount of Coulombs (calculated based on COD oxidation) (Li & Zhang 2012).

RESULTS AND DISCUSSION

Start-up of the denitrifying MFC

As shown in Figure 1(a), the cell voltage increased slowly before Day 14, but rapidly increased from less than 30 mV to 260 mV followed by a stable level. Along with the increase of cell voltage, the anode effluent COD concentration decreased significantly from Day 14 and reached a stable level of less than 20 mg/L after Day 31 (Figure 1(b)). This implied that the exoelectrogens growth rate was accelerated after 14 days adaptation. After the external resistance was adjusted from 100 to 50 Ω on Day 26 for further promoting the growth of exoelectrogens (Liu et al. 2005), COD removal increased little while the anodic Coulombic efficiency gradually increased from 14.1 to 22.2%. During the whole start-up stage, the cathode effluent nitrite was mostly less than 0.12 mgN/L but that shortly accumulated up to 0.55 mgN/L after the external resistance was adjusted from 100 to 50 Ω (Figure 1(c)). The possible reasons for short nitrite accumulation after the external resistance was increased is that the increase of current would provide more electrons for denitrification while nitrite reduction needs more time to adapt to this change than nitrate reduction to nitrite.

Figure 1

Variation of: (a) cell voltage, (b) COD, and (c) nitrogen during the start-up stage.

Figure 1

Variation of: (a) cell voltage, (b) COD, and (c) nitrogen during the start-up stage.

After 51 days operation, a stable cell voltage of 205.1 ± 1.96 mV was obtained at an external resistance of 50 Ω. The corresponding removal rates of COD and -N were 0.854 ± 0.020 kg/(m3NAC.d) and 6.17 ± 1.906 g/(m3NCC.d), respectively (data not shown). A previous experiment in a similar denitrifying MFC with PBS showed that a stable cell voltage of 420 mV with corresponding -N removal rate of 38 g/(m3NCC.d) were obtained at an external resistance of 50 Ω after 42 days start-up (Li & Zhang 2012). By comparison, it took a longer start-up time and obtained lower cell voltage and nitrogen removal rate in this study. One possible reason is the inhibiting effect of boric acid-borate buffer solution at high concentrations on the microbe (Qiang et al. 2011). The cathodic effluent pH (equal to pH of catholyte) reached 8.20 ± 0.05 in this study, which greatly deviated from the suitable pH 6.5–7.5 for denitrifier. Clauwaert et al. (2009) reported that the nitrogen removal increased from 0.22 to 0.50 kg -N/(m3NCC.d) when the pH of catholyte was decreased from 8.3 to 7.2. Therefore, another possible reason for inferior MFC performance in this study was the higher pH of catholyte. The open circuit potential of anode was around −281.2 mV, which was close to the standard redox potential of the couple CO2/CH3COOH (E′ = −280 mV). It implied that the anodic exoelectrogens were well developed and the limitation of the cathode should be responsible for low electricity generation.

Performance of denitrifying MFC at different buffer solution concentrations

As shown in Figure 2, a maximum power density of 5.67, 7.47 and 8.24 W/m3 NCC was attained at buffer solution concentrations of 0, 50 and 100 mmol/L, respectively. The corresponding external resistances, which could be estimated as corresponding internal resistances, were 67, 56 and 48 Ω, respectively. The anode and cathode solution conductivity were 0.506 and 0.899 ms/cm, respectively, under buffer free conditions, while they increased to 1.761 and 1.960 ms/cm under 100 mmol/L buffer solution condition. It means that a high solution conductivity resulting from high buffer solution concentration could reduce the internal resistance and favor electricity generation. Puig et al. (2012) found the same result that the maximum power density increased from 1.3 to 4.4 W/m3 NCC as the anode and cathode influent conductivity were, respectively, increased from 1.021 and 1.039 ms/cm to 3.98 and 4.078 ms/cm.

Figure 2

Polarization curves at different buffer solution concentrations.

Figure 2

Polarization curves at different buffer solution concentrations.

A high concentration of buffer solution could also improve buffer capacity to maintain a suitable pH for exoelectrogens. The effluent pH of anode was maintained at 7.0–7.5 at different buffer solution concentrations, but that of the cathode increased with the decrease of buffer solution concentration and reached 9.06 ± 0.04 under buffer-free conditions, which was greatly deviated from the suitable pH for denitrifier. As a result, the COD removal rate fluctuated little, but the maximum -N removal rate increased from 6.26 ± 0.402 to 7.97 ± 0.284 g/(m3NCC.d) as the buffer solution concentration was increased from 0 to 100 mmol/L (Table 1). Therefore, a high concentration of buffer solution favored electricity generation and nitrogen removal but has little effect on the COD removal. This was consistent with report of Puig et al. (2012). However, they found higher -N removal rate (25.4 g/(m3NCC.d)) at anodic and cathodic solution conductivities of 1.799 and 2.343 ms/cm mainly because of the different buffer solution (PBS) being applied. High concentration of boric acid-borate buffer solution would also incur water pollution. Adsorption and membrane separation technology would be investigating removal or recovery of boric acid from effluent in future research.

Table 1

Pollutant removal of denitrifying microbial fuel cell at different buffer solution concentration and external resistance

Boric acid-borate buffer solution (mmol/L) External resistance (Ω) COD removal rate [kg/(m3NAC.d)] NO3-N removal rate [g/(m3NCC.d)] 
67a 1.011 ± 0.009 5.22 ± 0.164 
5b 1.024 ± 0.004 6.26 ± 0.402 
50 56a 1.027 ± 0.008 5.41 ± 0.805 
50 5b 1.028 ± 0.017 6.33 ± 1.215 
100 48a 1.032 ± 0.006 6.54 ± 0.402 
100 5b 1.035 ± 0.001 7.97 ± 0.284 
Boric acid-borate buffer solution (mmol/L) External resistance (Ω) COD removal rate [kg/(m3NAC.d)] NO3-N removal rate [g/(m3NCC.d)] 
67a 1.011 ± 0.009 5.22 ± 0.164 
5b 1.024 ± 0.004 6.26 ± 0.402 
50 56a 1.027 ± 0.008 5.41 ± 0.805 
50 5b 1.028 ± 0.017 6.33 ± 1.215 
100 48a 1.032 ± 0.006 6.54 ± 0.402 
100 5b 1.035 ± 0.001 7.97 ± 0.284 

aExternal resistance where the maximum power density was achieved.

bExternal resistance where the limiting current was achieved.

Performance of denitrifying MFC at different temperatures

Polarization curves showed that the maximum power density of 5.61, 6.86, 7.89, 8.23 and 8.76 W/m3 NCC was attained at temperatures of 15, 20, 25, 30 and 35 °C, respectively (data not shown). The corresponding external resistances were, respectively, 80, 60, 60, 50 and 50 Ω, which could be estimated as corresponding to the internal resistance. Gonzalez del Campo et al. (2013) reported the same result that the maximum power density increased with the temperature being stepwise increased from 20 to 40 °C. It was thought that the high temperature could not only improve electrochemical activity of exoelectrogens and lower the reaction energy peak but also reduce the internal resistance and enhance the output of power density (Ahn & Logan 2010; Gonzalez del Campo et al. 2013). However, once the temperature was increased to a certain extent and exceeded the suitable range for microorganisms, the increase of power density would be impeded. As a result, the maximum power density increment in the lower temperature range (15–25 °C) was greater than for the higher temperature range (25–35 °C). When the cell voltage at different temperatures with an external resistance of 50 Ω was listed (data not shown), it showed a similar trend: the cell voltage increment (from 191.6 to 246.7 mV) in the lower temperature range (15–25 °C) was greater than that (from 246.7 to 265.6 mV) in the higher temperature range (25–35 °C).

As shown in Table 2, the temperature had little effect on COD removal rate. Nevertheless, lower temperature led to weaker power generation (as mentioned above). It implied that anodic exoelectrogens were more sensitive than non-exoelectrogens to the change of temperature and non-exoelectrogens consumed more COD at the lower temperature. It was found that the nitrogen removal rate at an external resistance where the limiting current was achieved was higher than that at an external resistance where the maximum power density was achieved for each temperature. The reason is this lower external resistance led to more electrons being transferred to the cathode for denitrification. The nitrate removal rate at lower temperature (15 and 20 °C) was lower than that at higher temperature (25, 30 and 35 °C), which was in agreement with the suitable temperature range (25–35 °C) for the denitrifier. Moreover, higher temperature can also decrease dissolved oxygen and provide an anoxic environment to facilitate nitrogen removal (Cheng et al. 2011). The maximum -N removal rate of 7.54 ± 0.204 g/(m3NCC.d) was achieved at an external resistance of 4 Ω and temperature of 25 °C, with a corresponding COD removal rate of 0.906 ± 0.014 kg/(m3NAC.d). As mentioned above, the maximum power density increment slowed after the temperature was higher than 25 °C. Hence, the suitable temperature for this denitrifying MFC was suggested to be 25 °C.

Table 2

Pollutant removal of denitrifying microbial fuel cell at different temperatures and external resistances

Temperature (°C) External resistance (Ω) COD removal rate [kg/(m3NAC.d)] NO3-N removal rate [g/(m3NCC.d)] 
15 80a 0.887 ± 0.012 2.53 ± 0.201 
15 5b 0.889 ± 0.008 4.22 ± 0.201 
20 60a 0.886 ± 0.004 4.05 ± 0.308 
20 5b 0.887 ± 0.003 4.84 ± 0.604 
25 50a 0.859 ± 0.006 6.05 ± 0.503 
25 4b 0.906 ± 0.014 7.54 ± 0.204 
30 50a 0.857 ± 0.002 5.83 ± 0.594 
30 5b 0.886 ± 0.003 6.54 ± 0.604 
35 50a 0.861 ± 0.019 4.62 ± 0.101 
35 5b 0.926 ± 0.005 5.97 ± 0.402 
Temperature (°C) External resistance (Ω) COD removal rate [kg/(m3NAC.d)] NO3-N removal rate [g/(m3NCC.d)] 
15 80a 0.887 ± 0.012 2.53 ± 0.201 
15 5b 0.889 ± 0.008 4.22 ± 0.201 
20 60a 0.886 ± 0.004 4.05 ± 0.308 
20 5b 0.887 ± 0.003 4.84 ± 0.604 
25 50a 0.859 ± 0.006 6.05 ± 0.503 
25 4b 0.906 ± 0.014 7.54 ± 0.204 
30 50a 0.857 ± 0.002 5.83 ± 0.594 
30 5b 0.886 ± 0.003 6.54 ± 0.604 
35 50a 0.861 ± 0.019 4.62 ± 0.101 
35 5b 0.926 ± 0.005 5.97 ± 0.402 

aExternal resistance where the maximum power density was achieved.

bExternal resistance where the limiting current was achieved.

Performance of denitrifying MFC at different external resistances

As shown in Table 3, the -N removal rate increased from 3.06 ± 0.302 to 7.54 ± 0.201 g/(m3NCC.d) when the external resistance was decreased from 200 to 5 Ω. The reason for this is that lower external resistance led to higher current and favored denitrification (Li et al. 2013). Although the COD removal rate remained at 0.856 ± 0.002 to 0.873 ± 0.004 kg/(m3NAC.d), the anodic Coulombic efficiency increased gradually from 9.6 to 42.8% as the external resistance was decreased from 200 to 5 Ω. This indicated that the external resistance had little effect on COD removal but lower external resistance led to a higher proportion of COD being consumed by exoelectrogens. The highest power density of 6.5 W/m3 NC was obtained at an external resistance of 50 Ω. Clauwaert et al. (2007) reported a higher power density of 8 W/m3 NCC in a similar denitrifying MFC with 50 mmol/L PBS. This means that the denitrifying MFC operated with 50 mmol/L PBS would gain an advantage, for power generation, over that with 100 mmol/L boric acid-borate buffer solution. Cha et al. (2009) reported a lower power density of 3.7 W/m3 NCC and a higher -N removal rate of 98 g/(m3NCC.d) in a similar denitrifying MFC without additional buffer solution. The possible reason for higher nitrate removal rate in research of Cha et al. is that the cathode influent formerly contained organic matter and nitrate, which would have enriched heterotrophic denitrifier in the cathode for more nitrate removal. Zhu et al. (2013) found that removal of the membrane from denitrifying MFC would result in a lower power density (0.0712 W/m3 NCC) and nitrate removal rate (0.186 g/(m3NCC.d)) even though PBS was used.

Table 3

Performance of the denitrifying microbial fuel cell at different external resistances

External resistance (Ω) COD removal rate [kg/(m3NAC.d)] NO3-N removal rate [g/(m3NCC.d)] Effluent NO2-N (mg/L) Power density (W/m3 NCC) Anodic Coulombic efficiency (%) 
200 0.867 ± 0.008 3.06 ± 0.302 0.02 ± 0.000 4.4 ± 0.01 9.6 ± 0.08 
150 0.859 ± 0.003 3.56 ± 0.402 0.02 ± 0.018 5.3 ± 0.03 12.3 ± 0.08 
100 0.856 ± 0.002 5.55 ± 0.201 0.03 ± 0.008 6.4 ± 0.07 16.5 ± 0.05 
50 0.862 ± 0.003 6.05 ± 0.503 0.02 ± 0.002 6.5 ± 0.25 23.5 ± 0.36 
25 0.856 ± 0.015 6.19 ± 0.302 0.02 ± 0.002 5.8 ± 0.18 31.6 ± 0.04 
10 0.846 ± 0.023 6.69 ± 0.201 0.03 ± 0.000 3.9 ± 0.25 41.3 ± 0.21 
0.873 ± 0.004 7.54 ± 0.201 0.03 ± 0.002 2.2 ± 0.31 42.8 ± 2.82 
External resistance (Ω) COD removal rate [kg/(m3NAC.d)] NO3-N removal rate [g/(m3NCC.d)] Effluent NO2-N (mg/L) Power density (W/m3 NCC) Anodic Coulombic efficiency (%) 
200 0.867 ± 0.008 3.06 ± 0.302 0.02 ± 0.000 4.4 ± 0.01 9.6 ± 0.08 
150 0.859 ± 0.003 3.56 ± 0.402 0.02 ± 0.018 5.3 ± 0.03 12.3 ± 0.08 
100 0.856 ± 0.002 5.55 ± 0.201 0.03 ± 0.008 6.4 ± 0.07 16.5 ± 0.05 
50 0.862 ± 0.003 6.05 ± 0.503 0.02 ± 0.002 6.5 ± 0.25 23.5 ± 0.36 
25 0.856 ± 0.015 6.19 ± 0.302 0.02 ± 0.002 5.8 ± 0.18 31.6 ± 0.04 
10 0.846 ± 0.023 6.69 ± 0.201 0.03 ± 0.000 3.9 ± 0.25 41.3 ± 0.21 
0.873 ± 0.004 7.54 ± 0.201 0.03 ± 0.002 2.2 ± 0.31 42.8 ± 2.82 

Nitrite accumulation (2 mgN/L) was reported in a previous study of denitrifying MFC (Lefebvre et al. 2008). Literature suggests that adding some organic matter in the cathode of denitrifying MFC reduces the nitrite accumulation (Qiang et al. 2011). However, no nitrite accumulation was found and the effluent nitrite was lower than 0.03 mgN/L at each external resistance in this study. One possible reason for this happening could be that the electrons provided for cathodic denitrification in this study were sufficient. Another possible reason is that the inner layer of biofilm attached on the graphite granule provided a suitable anoxic condition for denitrifiers, which alleviated the inhibiting impact of dissolved oxygen on denitrification (Li & Zhang 2012).

CONCLUSIONS

Denitrifying MFC with boric acid-borate buffer solution was started up successfully in 51 days with a stable cell voltage of 205.1 ± 1.96 mV at an external resistance of 50 Ω. The concentration of buffer solution and temperature had little effect on COD removal, but a higher concentration of buffer solution and temperature benefited electricity generation and nitrogen removal. The suitable temperature for this denitrifying MFC was suggested to be 25 °C. Decreasing the external resistance had little effect on COD removal, but favored nitrogen removal and organic matter consumption by exoelectrogens.

ACKNOWLEDGEMENTS

The Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry (No. 2011-1568), supported this work. The authors wish to acknowledge Prof. Mihret Dananto for his comments and help with revising the manuscript.

REFERENCES

REFERENCES
APHA
1992
Standard Methods for the Examination of Water and Wastewater
,
18th edn
.
American Public Health Association
,
Washington, DC
.
Cheng
S.
Xing
D.
Logan
B. E.
2011
Electricity generation of single-chamber microbial fuel cells at low temperatures
.
Biosensors and Bioelectronics
26
(
5
),
1913
1917
.
Clauwaert
P.
Rabaey
K.
Aelterman
P.
De Schamphelaire
L.
Ham
T. H.
Boeckx
P.
Boon
N.
Verstraete
W.
2007
Biological denitrification in microbial fuel cells
.
Environmental Science & Technology
41
(
9
),
3354
3360
.
Clauwaert
P.
Desloover
J.
Shea
C.
Nerenberg
R.
Boon
N.
Verstraete
W.
2009
Enhanced nitrogen removal in bio-electrochemical systems by pH control
.
Biotechnology Letters
31
(
10
),
1537
1543
.
Gonzalez del Campo
A.
Lobato
J.
Caňizares
P.
Rodrig
M. A.
Fernandez Morales
F. J.
2013
Short-term effects of temperature and COD in a microbial fuel cell
.
Applied Energy
101
,
213
217
.
Lefebvre
O.
Al-Mamun
A.
Ng
H. Y.
2008
A microbial fuel cell equipped with a biocathode for organic removal and denitrification
.
Water Science and Technology
58
(
4
),
881
885
.
Li
J.-T.
Zhang
S.-H.
2012
Basic research on denitrifying microbial fuel cell
.
China Environmental Science
32
(
4
),
617
622
.
Nam
J.-Y.
Kim
H.-W.
Lim
K.-H.
Shin
H.-S.
Logan
B. E.
2010
Variation of power generation at different buffer types and conductivities in single chamber microbial fuel cells
.
Biosensors and Bioelectronics
25
(
5
),
1155
1159
.
Puig
S.
Coma
M.
Desloover
J.
Boon
N.
Colprim
J.
Balaguer
M. D.
2012
Autotrophic denitrification in microbial fuel cells treating low ionic strength waters
.
Environmental Science & Technology
46
(
4
),
2309
2315
.
Qiang
L.
Yuan
L.-J.
Ding
Q.
2011
Influence of buffer solutions on the performance of microbial fuel cell electricity generation
.
Huanjing Kexue
32
(
5
),
1524
1528
.
Rabaey
K.
Ossieur
W.
Verhaege
M.
Verstraete
W.
2005
Continuous microbial fuel cells convert carbohydrates to electricity
.
Water Science and Technology
52
(
1–2
),
515
523
.
Watson
V. J.
Logan
B. E.
2011
Analysis of polarization methods for elimination of power overshoot in microbial fuel cells
.
Electrochemistry Communications
13
(
1
),
54
56
.
Zhu
G.
Onodera
T.
Tandukar
M.
Pavlostathis
S. G.
2013
Simultaneous carbon removal, denitrification and power generation in a membrane-less microbial fuel cell
.
Bioresource Technology
146
(
1–6
),
1
6
.