Appl Biochem Biotechnol (2014) 173:472–485 DOI 10.1007/s12010-014-0854-x

Municipal Solid Waste Landfill Leachate Treatment and Electricity Production Using Microbial Fuel Cells Lisa Damiano & Jenna R. Jambeck & David B. Ringelberg

Received: 3 October 2013 / Accepted: 10 March 2014 / Published online: 27 March 2014 # Springer Science+Business Media New York 2014

Abstract Microbial fuel cells were designed and operated to treat landfill leachate while simultaneously producing electricity. Two designs were tested in batch cycles using landfill leachate as a substrate without inoculation (908 to 3,200 mg/L chemical oxygen demand (COD)): Circle (934 mL) and large-scale microbial fuel cells (MFC) (18.3 L). A total of seven cycles were completed for the Circle MFC and two cycles for the larger-scale MFC. Maximum power densities of 24 to 31 mW/m2 (653 to 824 mW/m3) were achieved using the Circle MFC, and a maximum voltage of 635 mV was produced using the larger-scale MFC. In the Circle MFC, COD, biological oxygen demand (BOD), total organic carbon (TOC), and ammonia were removed at an average of 16%, 62%, 23%, and 20%, respectively. The larger-scale MFC achieved an average of 74% BOD removal, 27% TOC removal, and 25% ammonia reduction while operating over 52 days. Analysis of the microbial characteristics of the leachate indicates that there might be both supportive and inhibiting bacteria in landfill leachate for operation of an MFC. Issues related to scale-up and heterogeneity of a mixed substrate remain. Keywords Landfill leachate . Leachate treatment . Municipal solid waste . MFC

Introduction Historically, landfills have been the primary form of waste management throughout the world, and in many locations, this is still the case. For example, in the United States of America (USA), of the 250 million tons of waste produced in 2010, 136 million tons (54.2%) went to

L. Damiano University of New Hampshire, Durham, NH 03824, USA J. R. Jambeck (*) University of Georgia, Athens, GA 30602, USA e-mail: [email protected] D. B. Ringelberg US Army ERDC-CRREL, Hanover, NH 03755, USA Present Address: L. Damiano Sandborne Head and Associates, Concord, NH 03301, USA

Appl Biochem Biotechnol (2014) 173:472–485


landfills. While this number has been on the decline (from 89% in 1980), historical landfills still exist, and landfills will likely remain a disposal option into the future [33]. Landfill leachate is liquid that emanates from the landfill system either produced by the waste within the system or occurring from a past infiltration, i.e., groundwater or rainfall. Leachate must be collected and managed per most regulations throughout the world. In the USA, Federal regulations require that municipal solid waste (MSW) landfills are lined and leachate be collected and managed to protect human health and the environment [34]. Outside sources such as groundwater infiltration, precipitation, and/or surface drainage into the landfill can contribute significantly to the leachate volume. There are four primary categories of compounds in MSW landfill leachate: dissolved organic matter, inorganic macrocomponents, trace metals, and xenobiotic compounds [17]. Leachate management may consist of recycling back into the landfill, evaporation, treatment followed by disposal, or direct discharge to a municipal wastewater collection system. Some wastewater treatment facilities have the capability and capacity to accept landfill leachate with no treatment; however, some require pre-treatment, and the large organic load of leachate can be difficult and expensive to manage. If organic loading and/or ammonia is to be reduced, the traditional leachate treatment options reviewed here require energy, time, and cost, often with no additional benefit. Microbial fuel cells (MFCs) can be used to treat landfill leachate without additional energy input, while also producing a small amount of power. Bacteria that are capable of this exocellular electron transfer to an electrode can do this by artificial mediators, self-produced mediators, or through direct contact [25, 9]. The electrons flow through wiring and a resistor to produce electrical current and therefore direct electricity [26]. Various other studies have examined both unaltered and amended landfill leachate as substrate in MFCs, evaluating various levels of treatment as well [38, 18, 10, 32, 7]. In a study related to this work, a voltage of 542 mV was produced from a landfill leachate MFC with a power density of 3 mW/m2 (94 mW/m3) with a coloumbic efficiency (CE) of 4 to 17%. Chemical oxygen demand (COD) removals were in the range of 2 to 43%; biological oxygen demand (BOD) removal was 50%; TOC removal was 26%, and removals of ammonia were from 18 to 70% [6]. While these results were promising, this work includes a different design (horizontal instead of vertical), which allows for better contact of the leachate with the cathode. Additionally, an analysis of the microbial characteristics of the leachate was completed, indicating that there might be both supportive and inhibiting bacteria in landfill leachate for operation of an MFC.

Materials and Methods Leachate Landfill leachate was collected from Cell III, Phases 1 and 2 of the Turnkey Recycling and Environmental Enterprises (TREE) facility in Rochester, NH. Phases 1 and 2 of Cell III were opened in 1995 and 1996, respectively, with a third phase opened in 1997. At TLR III (Phases 1 and 2), leachate was collected directly from the pumping station. The leachate was transported in either 2 or 19 L HDPE plastic containers and placed in the MFCs within an hour of arrival at the laboratory, with the exception of cycles 2b and 4b of the Circle MFC, which were stored at 8–9 ° C for 1 week prior to use. All leachates were characterized before use in the MFC to generate baseline data to compare with the output leachate characteristics. Circle MFC Design The circular MFC was oriented horizontally, improving the contact between the leachate and the cathode over previous designs. The anode chamber was made from a 1000 mL Nalgene plastic cylindrical container, with a working volume of 934 mL. The


Appl Biochem Biotechnol (2014) 173:472–485

anode was constructed using a 11×11×0.9525 cm dense fine-grain graphite plate and nine 0.48 cm diameter by 7.5 cm long graphite rods. The plate was cut into an x-shape, and the rods were attached with silver epoxy (EE129-4, Epo-tek) as illustrated in Fig. 1c and d. The anode had a total surface area of 258 (0.258 m2). The cathode was composed of wet-proofed woven carbon cloth coated with 1 mg/cm2 platinum and was constructed based on previous research [26, 4]. The open-air cathode had a diameter of 8 cm for a total of 50 cm2. The cathode sat 1 cm above the anode when installed. The center of the pre-fabricated lid of the plastic container was removed, and the cathode was sealed in place using aquarium-grade 100% silicon. Silver epoxy (EE129-4, Epo-tek) was used to connect insulated copper wire to both the anode and carbon cloth. Once the lid was tightened into place, this system was placed on its side to create constant contact between the cathode and leachate, as pictured Fig. 1a. Larger-Circle MFC A larger MFC was created to determine electrical output and treatment capabilities of a scaled-up MFC (Fig. 1b). The anode chamber was made from a 5-gallon highdensity polyethylene bucket with a diameter of 28 cm and 34 cm height. There was a total volume of 1.89 L, with a working volume of 1.83 L. The anode was constructed using a medium extruded graphite plate 30.5×30.5×0.635 cm and nine 1.27 cm diameter by 30.5 cm long, fine, extruded graphite rods. The plate was cut into an x-shape, and the rods were attached with silver epoxy (EE129-4, Epo-tek). The anode had a total surface area of 1,942 cm2 (1.942 m2). The cathode was constructed in the same manner as described for the Circle MFC. The cathode was allowed to float on the surface of the leachate to allow constant contact in this upright system and had a diameter of 30 cm (707 cm2). However, overlapping on the sides of the container was allowed to minimize air infiltration. The cathode sat 1 cm above the anode when installed. Silver epoxy (EE129-4, Epo-tek) was used to connect insulated copper wire to both the anode and carbon cloth. Electrical Components An electrical breadboard was used for the wiring of both systems. A capacitor to compensate for electrical noise within the system was used, along with a 1Ω resistor to compensate for the resistance of the data acquisition unit and a 470 Ω resistor to provide a load for the system [6]. The internal resistance of this system was 400–500Ω. For optimum operation of an MFC, the external resistance should be equal to that of the internal resistance [1], which, in this case, correlated well.



Fig. 1 MFC designs (a) Circle MFC, vol. of 934 mL (b) larger-scale MFC, vol. of 1.83 L (c, d) Illustrations of graphite anode configuration used in both the Circle and large-scale designs, not to scale

Appl Biochem Biotechnol (2014) 173:472–485


MFC Operation Leachate was used as both the substrate and innoculum in this research. No additional anaerobic bacteria or nutrients were added to the system. Leachate was added and removed between cycles with caution to limit disturbance of any biofilm formation on the anode. No cleaning was done between consecutive cycles of MFC operation, so that continual growth of an exoelectrogen community could be achieved. A cycle of operation for these MFCs began with the addition of recently sampled leachate into system. The cycle typically ended when voltage produced dropped below 50 mV. Three sets of data were obtained: electrical production, leachate treatment, and microbial characterization. The MFC was operated in batch mode (although because of evaporation and utilization, leachate was added to the MFC in a cycle when necessary), and data were collected for each cycle of operation; referred to as ‘cycle’ in the following sections and consecutively numbered, starting with 1. Leachate taken from the landfill closed cell, Cell III, Phase 1, is designated ‘b’ while leachate from the landfill Cell III, Phase 2, is designated ‘c’. Electrical Measurements and Calculations A data acquisition unit (National Instruments USB 6210) was used with a desktop computer and software to measure and record data from the microbial fuel cells using a LabView program. Current was calculated according to the Equation V=I×Rext, where V=voltage (millivolts); I=current (milliamperes); and Rext =external resistance [25]. Current was normalized by the cathode area, since the anode materials had an abundance of surface area (geometric and internal) to calculate current density. Power density (for area and volume) is calculated by P (watts per square meter or cubic meter)=V2/(A or Vol×Rext), where V=voltage (volts); A=area of cathode or anode; Vol=volume of reactor; and Rext =external resistance [25]. In this case, power density is reported with normalization to the cathode area and the reactor volume. Columbic efficiency is calculated by equation CE =(8×I×t)/(F×VolA ×ΔCOD) where I=average current over time, t (amperes); t=time of cycle (seconds); F=Faraday’s constant (96,500C/mol); VolA =volume of anode compartment (liters), and ΔCOD=change in COD concentration over time, t (grams per liter) [25]. Polarization curves illustrate how well the MFC can maintain voltage as a function of current production. To obtain the polarization curve and power density curve, the external resistance was varied from 40,000Ω to 10Ω. Voltage was recorded for each resistance once readings had stabilized. Leachate Characterization For all measurements, data were recorded for the leachate prior to input into the MFC system and after treatment. Further description of residence times can be found in the results section of this paper. Temperature (Celsius, ±0.15° accuracy), pH (±0.2 units accuracy), oxidation reduction potential (ORP) (millivolts, ±20 mV accuracy), dissolved oxygen (milligrams per liter and percent, ±2% accuracy), conductivity (millisiemens per centimeter, ±0.5% accuracy), and specific conductivity (microsiemens per centimeter) were all measured using a YSI 556 MPS probe. Other analyses included: COD (accuracy, 778– 822 mg/L COD, 95% confidence limits of distribution, run in triplicate) with Hach method 8000 [11]; BOD (detection limit 5 mg/L) with Standard Method 5210 B [2]; TOC (detection limit 20 mg/L) with US EPA SW-846 9060 [35]; ammonia (detection limit, 250 mg/L) with Standard Method 4500 [2]; alkalinity (detection limit, 5 mg/L) with Standard Method 2320 B [2]; nitrite (2 mg/L detection limit), nitrate (0.5 mg/L detection limit), sulfate (25 mg/L detection limit), and chloride (25 mg/L detection limit) with US EPA Method 300.0 A [31]; total phosphorus and phosphate (detection limit 0.5 mg/L) with US EPA Method 365.3 [36]; sulfide (±2% accuracy, run in triplicate) with US EPA Method 376.2 using Hach Method 813 [12]. Replicates and blanks were included in discrete sample runs. Inductively coupled plasmaatomic emission spectroscopy was used to detect cations and inorganic (trace) metals in the


Appl Biochem Biotechnol (2014) 173:472–485

leachate. Samples were analyzed for the presence and concentration of aluminum, antimony, arsenic, barium, calcium, cobalt, chromium, iron, magnesium, manganese, nickel, selenium, silver, vanadium, and zinc. NIST standards, calibration blanks, and calibration verifications were used for each analysis to ensure quality of the data. The calibration verifications and NIST standards were included at least every 20 samples to ensure the calibration remained consistent over the entire analysis, and that various labs, conducting the same trace metal analysis, were detecting similar concentrations of the same solution. Solution matrix spikes were performed to make sure elemental interferences were not affecting the detection capabilities of the analysis. Microbial Analysis The microbial communities of four different leachate/biofilm samples were fingerprinted using terminal restriction fragment length polymorphisms. Details of methods used are provided in Damiano [5]. A sample of leachate from TLR II, Phase 1, which was unable to produce electricity (non-producing), was tested to determine if the microbial community could be inhibiting electrical results. Leachate from TLR III, Phase 2, was also analyzed prior (pre-MFC) to entering the MFC system as well as after (post-MFC) running a complete cycle in the Circle MFC. A sample of biofilm scraped from the anode of the Circle MFC was also tested when the MFC had been consistently running for approximately 2 months with landfill leachate.

Results and Discussion Voltage Production Seven continuous cycles of operation were completed over 3.5 months with the circle cell (Fig. 2a and b). The larger circle cell was operated in two cycles for a total of 65 days. The start/peak/end voltage, time-to-peak voltage, and cycle times are provided in Damiano [5]. Some of the voltage versus time plots mimic the phases that are typical in bacterial growth. The growth process begins with a lag phase as bacteria become accustomed to the environmental conditions, and little growth is observed. This phase is followed by exponential growth of the microbial population and then a stationary phase where little growth is seen, but living cells are maintained. Lastly, a negative growth phase occurs if no new nutrients or carbon source are supplied to the bacteria. The absence of a lag phase in some cycles could be the result of an existing microbial community within the system (no cleaning of the MFCs was conducted between consecutive cycles of operation). Once the bacteria begin to die from the exhaustion of the carbon source and/or nutrients in the leachate, electricity generation begins to decrease as well. Circle MFC cycles 3b and 4b had significantly shorter run times than other cycles. This is likely from the decreased COD levels of the influent leachate in these cycles (908 and 1,075 mg/L, respectively). Some of the variations in voltage readings are because of evaporation from the system and the subsequent additions of leachate that caused a return to optimum operating conditions (creating more of a “fed-batch” reactor in some cases). The maximum peak voltage of all the cycles was 534 mV, with leachate (b), cycle 4b. Cycle 7c was significantly longer than any previous cycle; the BOD content of the influent leachate was twofold the original amount, increasing from 180 to 200 mg/L to 430 mg/L. Voltage generation was sustained for nearly a month before the MFC was taken offline (still producing 110 mV) because of timing constraints of providing a microbial sample for analysis. While the large-scale MFC increased the volume of substrate to 19 L, significant internal resistance is added to the system, with both protons and electrons having longer paths to travel to complete the circuit and reaction, so linear increases in voltage are not observed. Voltage was produced by the scaled-up MFC but quickly decreased, most likely due to the use of new

Appl Biochem Biotechnol (2014) 173:472–485


materials for MFC construction and necessary acclimation of the bacteria (Fig. 2c). A second cycle was completed where voltage was maintained >52 days and had a peak of 635 mV (Fig. 2c). Power Density and Coulombic Efficiency Power densities for previous research with Square MFCs were 3 mW/m2 and 94 mW/m3 [6]. Maximum power densities for the Circle MFC design of this research improved to 24 to 31 mW/m2 or 669 to 844 mW/m3. Because of the large internal resistance, the power density of the large scale MFC was insignificant; however, it still performed well for leachate treatment, as will be discussed in subsequent sections. With varying architectures and operation differences, a direct comparison of results to other research utilizing landfill leachate or other wastewaters is not well defined, but a comparison of power densities, among other parameters, from other studies is presented in Table 1. One study obtained a power density of 6,817.4 mW/m3, using landfill leachate in a small 40 mL volume single-chamber MFC, using dilute leachate and anaerobic sludge inoculum; the greater power density of this research can be attributed to the smaller-scale MFC and the amended leachate substrate [38]. Another study contained a pyrrhotite-coated graphite-cathode, which significantly lowered the internal resistance of the cell resulting in a power density of 4,200 mW/m3 [18]. The Circle MFC in this study outperformed the max power densities of a tubular MFC (0.9 L volume) with a membrane and a continuous feed of leachate (maximum power density of 1.38 mW/m2 [10]) and a single-chamber column cell (power density of 344 mW/m3 [32]). In addition, two polarization curves (illustrating how well the MFC can maintain a voltage as a function of current production) were produced for the Circle MFC (Fig. 3).


B 600 Cycle 1b Cycle 2b Cycle 3b Cycle 4b

Voltage (mV)

500 400 300 200

400 300 200 100

100 0

Cycle 5c Cycle 6c Cycle 7c

500 Voltage (mV)


0 2 3 4 5 6 7 8 9 1011121314151617


0 2 4 6 8 10 12 14 16 18 20 22 24

Time (days)

Time (days)



Voltage (mV)

600 500 400 300 200 Cycle 1c Cycle 2c

100 0

0 4 8 12 16 20 24 28 32 36 40 44 48

Time (days)

Fig. 2 Voltage versus time plots of Circle MFC for (a) Cycles 1b–4b, (b) Cycles 5c–7c and (c) larger-scale MFC, Cycles 1–2c

Landfill leachate

Landfill leachate

Landfill leachate

Circular with activated carbon anode

Circular with biochar anode









388 40






Vol. (mL)


43 74.7%±5.5% 28.6%±8.9% 27%±16%

49.2±2.3 (699±33 mW/m3) 40.4±12 (575±168 mW/m3) 31 (824 mW/m3)




In units of milliwatts per cubic meter

Removal of BOD






57–66c 78

NR 3.4






CE (%)

50–70 70–98






COD removal (%)





9 6,817.4b






Power density (mW/m2, except noted)

NR no resulta Two chamber cell with pyrrhotite-coated graphite-cathode with Fenton’s reagents oxidizing biorefractory organic compounds

Landfill leachate

Landfill leachate


Landfill leachate



Landfill leachate


Swine wastewater


Domestic wastewater

Swine wastewater


Domestic wastewater Landfill leachate (dilute)

Paper wastewater


Tubular Cylinder

Brewery wastewater




Shape/architecture (all single chamber cells, except noted)

Table 1 Comparison of results of this study with values from the literature

This research







[19] [38]







478 Appl Biochem Biotechnol (2014) 173:472–485

35 30 25 20 15 10 5 0

Power Density (mW/m2)

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

30 25 20 15 10 5 0

Power Density (mW/m2)


0.0000 0.0001 0.0001 0.0002 0.0003 0.0006 0.0011 0.0021 0.0026 0.0047 0.0077 0.0096 0.0097 0.0118 0.0109 0.0101

Cell Voltage (V)

Appl Biochem Biotechnol (2014) 173:472–485

Current Density (mA/cm2) Power Density














0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0.0000

Cell Voltage (V)

Polarization Curve

Current Density (mA/cm2) Polarization Curve

Power Density

Fig. 3 Power density and polarization curves for the Circle MFC

Coulombic efficiency, CE, often used to describe the efficiency of MFC systems, was calculated for all cycles of each MFC. Values ranged from 7.9 to 41% for the Circle design and were 5.2% for the larger-scale design. When complex substrates are used in MFCs, CE is calculated based upon COD removals as a representation of the amount of organic degradation achieved by the system. The work of You et al. recorded a CE of 3.4%, with previous work by Ganesh and Jambeck [7], resulting in a CE of 1.27%±0.61% (Table 1). MFC Leachate Characterization Water Quality Parameter Treatment Increases in pH were in the range of 6% to 16%, with one small decrease occurring in the Circle MFC, cycle 2b (−3%). pH decreased for the larger-scale MFC (Table 2). While it is common for the pH to change at the anode of an MFC during operation, it is generally a decrease to a more acidic level due to the incomplete transfer of protons to the cathode and a resulting buildup of protons at the anode which reduces pH. However, in other research using landfill leachate, a slight increase in system pH was observed during MFC operation, attributed to a possible removal of acidic components present in the leachate, such as volatile fatty acids, during MFC operation [10, 7]. Leachate is a highly buffered system as well, and sulfides in the leachate can accept protons and limit pH decrease [14]. For all designs and cycles of the MFCs, there was a consistent decrease in alkalinity during the total cycle time. This would suggest that buffering of the system was occurring while the MFC was operating. If protons are accumulating at the cathode, a decrease in alkalinity can occur as the system is attempting to remain in equilibrium. Furthermore,


Appl Biochem Biotechnol (2014) 173:472–485

Table 2 Summary of influent, effluent, and percent difference values for leachate characterization of Circle and larger-scale MFC systems Circle MFC

Temperature (°C)

Larger MFC




% Difference




% Difference








−12% 11%









Conductivity (mS/cm)









DO (mg/L)









ORP (mV)





7 4







2,130±907 238±131

1,780±811 90±28

−16% −62%

2 2






2,386 305

2,444 78

2% −74%




































30% −25%


















Total phosphorous Sulfide

4 3

28±28 0.24±0.04

5.6±3.0 0.1±0.07

−80% −58%

2 2

11.5 0.22

6.2 0.2

−47% −18%

ammonia and phosphate were removed by the MFC, so if these were contributing to alkalinity, this would cause a decrease. Sulfide can also accept protons in leachate, and this decreased from the MFC, which could contribute to alkalinity reduction [14]. There has been substantial research that illustrates conductivity is a key factor in the efficiency of an MFC system. Conductivity, through the increase in ionic strength, has been artificially increased in many wastewaters and artificial substrates that are used in MFCs [13, 20, 21, 24, 30]. One of the major benefits of using landfill leachate as a substrate in MFCs is that conductivity is relatively high. Influent conductivity was in the range of 11.2 to 17.3 mS/ cm. Effluent readings were in the range of 10.4 to 19.1 mS/cm, creating a decrease in levels ranging from 1 to 17% (with one exception in cycle 2b where conductivity was increased). Landfill leachate is generally an anaerobic substrate with low DO levels, which were variable throughout sampling and testing of the MFCs in this research. Increases in DO concentrations within the MFC systems during operation would be due to the system being open to the air. An aerobic zone could have formed near the cathode and resulted in increases of DO. For the cycles where a decrease in DO occurred, anaerobic conditions within the system were more efficient, and a larger anaerobic zone was allowed to control the system and reduce DO. ORP can be a useful measure of the state of the system being tested, as it measures the tendency of a solution to gain or lose electrons. Although landfills are generally anaerobic systems, one exception to this anaerobic condition was the influent leachate from TLR II, Phase 2, during Cycles 3 and 4b for the Circle MFC. These influent values were +52.5 and + 73 mV. This suggests that the leachate entering the MFC was actually aerobically active. Interestingly, these cycles also had significantly shorter total cycle time for both of the MFC designs. However, a connection cannot be definitively made between ORP and cycle time because of the dynamic values of other constituents. For all other incoming leachate samples, a

Appl Biochem Biotechnol (2014) 173:472–485


range of negative ORP values were obtained; however, ORP values of effluent leachate did vary and had an overall increase in each MFC (Table 2). To produce electricity, MFCs must have anaerobic conditions for the appropriate bacteria to grow within the system; however, these results, along with the DO results, do suggest that the systems had aerobic zones. COD, BOD, and TOC The Circle COD removal range was 15 to 49% (mean=16% reduction, Table 2); however, in Cycles 5 through 7c, COD increased during the cycle of the MFC. For the same leachate c, the larger-scale MFC had an increase in COD of 6.6% and a decrease of 3.1%. There are several inorganics that can exhibit COD in leachate, such as sulfides that can interfere in COD being a measure of just the organics of a system [14]. BOD removals for these same cycles were 53 to 72% for the Circle MFC, and 47 to 86% for the larger-scale. While the COD removals are lower than those recorded by other landfill leachate in MFC research (70 to 98%), BOD is in the range of 57 to 66% removal that has been previously reported [6, 1]. COD removals for other MFC systems utilizing different wastewaters are shown in Table 1; however, direct comparison is confounded because of highly variable operating conditions and architectures. Small volume systems (

Municipal solid waste landfill leachate treatment and electricity production using microbial fuel cells.

Microbial fuel cells were designed and operated to treat landfill leachate while simultaneously producing electricity. Two designs were tested in batc...
959KB Sizes 0 Downloads 3 Views