Electricity generation from untreated fresh digestate with a cost-effective array of floating microbial fuel cells

Chemical Engineering Science 198 (2019) 108–116

Contents lists available at ScienceDirect

Chemical Engineering Science journal homepage: www.elsevier.com/locate/ces

Electricity generation from untreated fresh digestate with a costeffective array of floating microbial fuel cells Sara Monasterio Martinez a,b, Mirella Di Lorenzo b,⇑ a b

Exergy Ltd., Office IV7 1st Floor, Coventry Innovation Village CUTP, Cheetah Road, Coventry CV1 2TL, UK Biosensors, Bioelectronics and Biodevices Centre (C3Bio) and Department of Chemical Engineering, University of Bath, Bath BA2 7AY, UK

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


An innovative floating microbial fuel

cell system, characterised by a simple and cost-effective design, is reported.
The anode is treated to enhance the wettability and surface area.
Bioelectricity generation from fresh digestate is demonstrated.
A stack of up to 10 MFCs, sharing the same electrolyte and electrically connected in parallel was assembled.
A linear power output increase was observed with the sequential additions of MFC units in the stack.

a r t i c l e

i n f o

Article history: Received 30 September 2018 Received in revised form 18 December 2018 Accepted 26 December 2018 Available online 19 January 2019 Keywords: Agri-food waste Fresh digestate Microbial fuel cells Stack Bioelectricity

a b s t r a c t Over one billion tons of food waste is generated every year worldwide. This waste represents a substantial part of municipal solid waste and is usually incinerated. The effective integration of the anaerobic digester process with the microbial fuel cell technology is an ecofriendly and promising solution for the treatment of food waste, which leads to clean energy generation and by-products of industrial interest. In this context, we here report the development of a floating air-cathode microbial fuel cell device and demonstrate electricity generation from fresh digestate, directly collected from an anaerobic digester effluent with no pre-treatment, used as the electrolyte, fuel and source of electroactive bacteria. The floating fuel cells are characterised by a simple yet innovative design. No metal catalyst is used at the cathode, and the use of a membrane is not required thanks to a natural vertical stratification of microorganisms in the digestate that prevents oxygen diffusion to the anode. Both the wettability and the surface area of the anode are enhanced with a two-step pre-treatment, which enhances the electrochemical performance of the electrode, leading to an oxidation peak twice greater than the non-treated electrode. The individual microbial fuel cell unit generated a power peak of 0.043 ± 0.001 mW, which increased linearly by connecting several units electrically in parallel in a stack and reached the value of 0.43 mW (corresponding to 51 ± 2 mW m3) with ten units. Considering the simplicity and affordability of the design proposed, which facilitates upscaling, this work paves the way for a promising and environmentally friendly alternative to food waste incineration. Crown Copyright Ó 2019 Published by Elsevier Ltd. All rights reserved.

1. Introduction

⇑ Corresponding author. E-mail address: [email protected] (M. Di Lorenzo). https://doi.org/10.1016/j.ces.2018.12.039 0009-2509/Crown Copyright Ó 2019 Published by Elsevier Ltd. All rights reserved.

Worldwide, around 30% of food is wasted throughout the agrifood supply chain (FAO, 2018). In Europe alone, more than 88 million tons of food are wasted annually, at a cost estimated to be over

S.M. Martinez, M. Di Lorenzo / Chemical Engineering Science 198 (2019) 108–116

€140 billion (Stenmark et al., 2016). These numbers are expected to increase in the following years, considering the forecasted onethird world population growth by 2050 (World Resources Institute, 2013). Every day, food waste (FW) treatment systems pose consistent environmental, public and economic issues, due to high energy/water demands, high emissions leading to climate change and increasing costs (Thyberg and Tonjes, 2016, Stenmark et al., 2016). The high moisture content of FW generally results in a lower heating value when compared with other materials. As such, waste to energy incineration (WTE) is considered an energetically unfavourable treatment for FW and it also produces harmful dioxins (Press and Academy, 2006). FW, however, can become an energy source if managed in landfills with methane collection systems, through composting or in anaerobic digesters (Karthikeyan et al., 2017). Anaerobic digestion (AD) is widely used for FW management, amongst other organic waste recycling approaches (i.e. gasification or pyrolysis) (De Baere, 2000; Gao et al., 2017). This technology reduces the generation of greenhouse gas emissions and the energy consumption related to WTE processes, while promoting bio-fertiliser recovery. In an AD-based treatment plant, the collected and accepted food is screened and fed into the pasteurisation process. Then, the pasteurised waste goes to the AD process where microorganisms break down the FW in an oxygen-free atmosphere. One of the obtained products is biogas, mainly consisting of a methane-rich biogas, which can be harvested and used as fuel (Holm-Nielsen et al., 2009). The high amounts of nutrient-rich slurry endproduct of AD, known as fresh digestate, is widely employed by farmers as an organic bio-fertiliser (Costa et al., 2015). Fresh digestate, however, has great potential as bioenergy source, which should be further explored. Recently, microbial fuel cells (MFCs) have gained great attention as a promising renewable and environmentally friendly biomass-based technology for the treatment of organic matter in waste (Nastro et al., 2015). With respect to other treatment technologies, MFCs not only offer several environmental (low carbon footprint), economic (low operational costs) and operational (self-generation of microorganisms, mild operating conditions) advantages, but they also present the unique ability to directly generate useful energy (Rabaey and Verstraete, 2005). MFCs can be fed with a wide range of organic compounds thanks to the flexibility of bacteria metabolism and the ability to use several organic compounds for chemical energy recovery (Bose et al., 2018; Pham et al., 2006). Activated sludge has shown to be a good source of microorganisms for MFCs treating liquid wastes, proving effective for biological and chemical oxygen demand reduction (Rodrigo et al., 2007; Singh et al., 2018). (Semi-)solid/viscous wastes has also been explored as fuel source for MFCs. In particular, Mohan and Chandrasekhar (2011) demonstrated bioelectricity generation from composite food waste fermentation with a solid phase MFC. Examples in the literature on the MFC technology for agri-food waste treatment are summarised in Table 1. The integration of AD with MFC has been previously suggested as an effective way to further treat wastewaters and enhance the fuel efficiency of the system with greater energy generation (Kim et al., 2015). A recent review also suggests the use of microbial electrolysis cells to help balance pH and volatile fatty acids concentration in anaerobic digesters (Yu et al., 2018). Still, the use of fresh digestate (with no pre-treatment) as feedstock in MFCs has not been fully explored yet. In this context, we here report an innovative stack of floating air-cathode microbial fuel cells, and demonstrate energy harvesting from untreated fresh digestate, composed for a 50–60% of kerbside collection from city councils, and for the rest of supermarket FW. The digestate is used as fuel, electrolyte and bacteria source for the MFCs. Therefore, no external bacterial inoculum or carbon

109

source is added to the system. The MFC unit has a simple yet effective design, which consists of two graphite felt electrodes (the anode and the cathode), with an immerged anode and a cathode floating onto the digestate surface. No membrane nor an expensive oxidation reduction reaction metal catalyst is used at the cathode. Consequently, the overall system is cost-effective and easy to scale-up. We first investigate the performance of a single MFC unit. Afterwards, to demonstrate power output scale-up in a practical application set-up, an array of up to ten floating MFCs, electrically connected in parallel, is built-up and tested. 2. Materials and methods 2.1. Reagents and digestate All reagents used were of analytical grade and used without further purification. They were all purchased from Alfa Aesar and Sigma-Aldrich, unless otherwise specified. Digestate from an anaerobic digester (Wessex Water Treatment Plant in Avonmouth, Bristol, UK), which processes food waste, was used as the feedstock, electrolyte and bacteria source in the MFCs, without undergoing any pre-treatment. Table 2 summarises the physico-chemical properties of the digestate collected. The Chemical Oxygen Demand, COD, of the digestate was measured with a COD high range EPA approved dichromate method (Rice et al., 2017) (Hannah Instruments, COD reagents high range). The pH and conductivity of the digestate were measured by using a Thermo Scientific Orion Star A325 probe. The percentage in water content of digestate, WCdigestate, was determined from the weight difference before and after a drying process, according to Eq. (1):

WC digestate ð%Þ ¼

  W1  W2  100% W2

ð1Þ

where W1 is the weight of the digestate sample (g) and W2 is the weight of the digestate sample (g) respectively before and after being oven-dried overnight at 105 °C (Reeb and Milota, 1999). The dry solids content was calculated as follows (Rice et al., 2017):

Dry Solidsð%Þ ¼

W3  100% W2

ð2Þ

where W3 is the weight (g) of the digestate sample after gasified at 550 °C, obtained by incubating overnight the sample in a thermostatically controlled muffle oven (Thermo Scientific Thermolyne Muffle Furnace). 2.2. MFC design A catalyst-free membrane-less air-cathode MFC design was used (Fig. 1). The system consists of two electrodes held in place with nylon screws (RS components) at a fixed distance of 1.5 cm from each other. Both anode and cathode consisted of a 5  5 cm2 rectangular piece of graphite felt, GF (0.7 cm thickness, Online Furnace Services Ltd). To increase the exposed area of the electrodes, nine holes of 4 mm diameter each, at a distance of 1 cm from each other, were pierced onto the GF, which led to a projected area of 22.73 cm2. During the operation, the anode was immersed into the digestate, while the cathode was exposed to air (Fig. 1). To prevent the cathode from soaking into the digestate and allow the system to float, a rectangular piece of styrene foam (8 cm length  8 cm width  0.4 cm height), was used as shown in Fig. 1A and B. Titanium wire (0.25 mm dia., Alfa Aesar) was used for electrical contacts. Insulation tubing (RS, 2:1, 1.6 mm dia  1.2 m) was used to protect the contacts from short-circuiting.

110

Table 1 Summary of MFCs operated with food-based waste.

c

Source inoculum

Electrode materialsa

Catholyte

Membrane

Catalyst (cathode side)

Pmax (mW m2)

imax (mA m2)

Ref.

Kitchen garbage/bamboo

No

No

58–60

N/A

Moqsud et al. (2014)

Atmospheric O2

NafionÒ

No

1–10

25–95

Swine wastewater

No

Atmospheric O2

No

Pt

261

1400

Venkata Mohan et al. (2010b) Min et al. (2005)

Canteen-based composite

Anaerobic sludge

Atmospheric O2

NafionÒ

No

2.25–5

40–90

Goud et al. (2011)

Rice mill

Anaerobic sludge

Anode: carbon fibre Cathode: carbon fibre Anode: graphite plate Cathode: graphite plate Anode: carbon paper Cathode: carbon paper Anode: graphite plate Cathode: graphite plate Anode: stainless steel mesh Cathode: graphite plate

Atmospheric O2

Dairy wastewater

Commercial micro-organisms (not further specified) Anaerobic mixed consortia

Aerated tap water

No

(a) 40 (b) 7

(a) 120 (b) 35

Behera et al. (2010)

Kerbside Food

Anaerobic sludge

Atmospheric O2

(a) Earthen pot (b) NafionÒ ZirfonÒ

Pt

2

6

Li et al. (2016)

Vegetable

Anaerobic sludge

Atmospheric O2

NafionÒ

No

15–27

120–125

Corn stover biomass

Domestic wastewater

Atmospheric O2

nsb

Pt

300–400

1000

Venkata Mohan et al. (2010a) Wang et al. (2009)

Meat packing wastewater (MPW)

MPW

Atmospheric O2

No

Pt

140

700

Heilmann and Logan (2006)

Gelatin

Activated sludge

Atmospheric O2

Terracotta

Activated carbon with PTFEc

235

532

Ieropoulos et al. (2017)

Unless specified in Table 1, the MFCs work in a single chamber configuration. ns = not specified. Polytetrafluoroethylene.

Anode: carbon cloth Cathode: carbon cloth Anode: graphite plate Cathode: graphite plate Anode: plain carbon paper Cathode: carbon cloth Anode: non-wet proofed Toray carbon paper Cathode: wet proofed Toray carbon paper Anode: carbon fibre veil with 30 g/mb carbon loading Cathode: carbon fibre veil

S.M. Martinez, M. Di Lorenzo / Chemical Engineering Science 198 (2019) 108–116

a b

Waste

S.M. Martinez, M. Di Lorenzo / Chemical Engineering Science 198 (2019) 108–116 Table 2 Summary of the physico-chemical characteristics of the digestate used. Parameter

Value

COD (mg L1) pH Conductivity (mS) Water content (%) Dry solids (%) Volatile acids (mg L1) Volatile solids (mg L1)

9932.42 ± 91.46 7.72 ± 0.51 6.96 ± 0.53 55.39 ± 7.83 3.65 ± 0.05 6615 ± 357 68.59 ± 0.41

2.3. Anode pre-treatment Prior to being used as the anode, the graphite felt was pretreated to enhance the hydrophilicity of the carbon nanofibres, with a procedure adapted from Feng et al. (2010). For this purpose, the electrode was soaked for 48 h at room temperature either in ethanol or acetone, as specified. Afterwards, the electrode was exposed to an aqueous solution of ammonium peroxydisulfate (0.87 M) and sulfuric acid (1.88 M) for 15 min and finally, thermally treated in a muffle furnace at 450 °C for 30 min in air atmosphere (temperature ramp: 10 °C min1). After each pre-treatment step, the anodes were thoroughly washed five times for 5 min each with distilled water. The so-treated electrodes were stored in MilliQ water at room temperature until used. Cyclic voltammetry tests were performed using a PGSTAT 302 (Metrohm-Autolab, The Netherlands) to assess the electroactvity of the electrode at each stage of treatment. The tests were performed in a 104 M ferricyanide solution in 0.1 M KNO3, at a scan rate of 5 mV s1, in a con-

111

ventional three-electrode set up, with GF as the working electrode, Pt wire as the counter electrode (1.5 mm dia.) and saturated calomel electrode, SCE (CHI150, CH Instruments, Inc.), as the reference electrode. The electroactive surface area, ESA, of the treated and nontreated GF was determined by CV tests run at different scan rates (5, 10, 20, 30, 50, 100, 200, 300, 400 and 600 mV s1), in the three-electrode set-up described above. The ESA was estimated from the height of the oxidation current peak, Iox p , by using the Randles-Sevcik equation for reversible processes (Eq. (3)):

  5  n1=3  A  D1=2  C  t1=2 Iox p ¼ 2:6910

ð3Þ

where n is the number of electrons transferred in the redox reaction (n = 1); D is the diffusion coefficient of ferricyanide (7.60  106 cm2 s1) (Stevens et al., 2001), C is the bulk concentration of the redox probe (1  107 mol cm3), and m is the scan rate (V s1). The hydrophobic nature of the GF was studied by the sessile drop technique (contact angle system, Dataphysics), by placing a drop of water (5 lL) onto the GF surface. The contact angle, h, was then measured with the SCA 20 software. 2.4. MFC operation Firstly, the performance of a single MFC unit was investigated. With this purpose, the fuel cell was fitted into a 500 mL beaker filled with digestate (Vdigestate = 300 mL). To polarise the cell, the anode and the cathode were connected through an external load, Rext, of 510 O. The cell voltage over time was monitored with a data

Fig. 1. (A) Sketch to-scale and (B) photograph of the MFC used in this study. (C) Sketches to-scale of the experimental set up used in this study, corresponding to pond 1 (P1) hosting two MFCs, and pond 2 (P2) hosting 10 MFCs.

112

S.M. Martinez, M. Di Lorenzo / Chemical Engineering Science 198 (2019) 108–116

Fig. 2. Cyclic voltammograms of the non-pre-treated (GF0) and pre-treated (GFA, GFE and GFE-T) anodes in 0.1 mM ferricyanide solution with 0.1 M KNO3, at a scan rate of 5 mV s1. GF0 refers to a GF electrode that has not been pre-treated (control). GFA and GFE refer to a GF electrode that has been soaked for 48 h in acetone and ethanol, respectively. GFE-T refers to a GF electrode that underwent the solvent and acid-heat treatment process. Current density refers to the electrode projected area: 1 cm2.

acquisition system (ADC-24 Pico data logger, Pico technology, UK). Ohm’s law (I = V/Rext) was used to calculate the current, I, (where V indicates the cell voltage). This test was performed in replicates. Polarisation tests were performed by in situ linear sweep voltammetry (LSV) measurements, scanning the cell potential from the open circuit voltage (OCV) to zero at a scan rate of 5 mV s1. Before the LSV test, the cells were left under open circuit for 2 h. This time was required for the fuel cells to reach a pseudo-steady state and a stable OCV. The power and current densities were normalised to the projected anode surface area. The internal resistance was estimated from the slope of the ohmic region in the polarisation curve, as previously described (Logan et al., 2006). The morphology of the anodic biofilm was visualised by Scanning Electron Microscopy, SEM, (Jeol JSM-6480LV). The biofilm was fixed as described in the Supplementary Information. All samples were coated with gold prior to imaging. To scale up the bioelectricity generation process, several MFC units were fitted into a PVC box to create a stack. As shown in Fig. 1C, two configurations were used. Pond 1 (P1, 25 cm length  10 cm width  10 cm height) consisted of a box hosting two MFCs, while Pond 2 (P2, 46 cm length  30 cm width  10 cm height) hosted 10 MFCs. In both configurations the fuel cells shared the digestate, which was filled out up to a volume of 0.9 L in P1 and of 8.2 L in P2. Each pond was topped up with fresh digestate on a weekly basis to counteract volume losses by evaporation. During the enrichment phase, the MFCs in the pond were individually connected to the external resistor (Rext = 510 X). Once a steady state cell voltage value was observed, after approximately a week, the MFCs were electrically stacked in parallel to scale up the power output, as shown in Fig. S1 in the Supplementary Information. All the experiments were performed in duplicates and conducted at room temperature, which was maintained at approximately 18 °C with an air conditioning system. 3. Results and discussion 3.1. Anode pre-treatment Prior to being used as the anode in the MFC, the GF was pretreated with a two-step process to enhance the wettability and

the surface area of the carbon fibres. GF has indeed a hydrophobic nature, with a contact angle of 150.77 ± 6.64° (Fig. S2A). As such, if untreated, most of its surface area would not be reached by the electrolyte. The wettability is also important for both the electrochemical performance of the electrode and the biofilm attachment (Artyushkova et al., 2015; Sarjit et al., 2015). For example, it has been recently demonstrated that by switching from a hydrophobic to a hydrophilic electrode the electron transfer activity of the electroactive bacterium Shewanella Ioihica PV-4 increased more than five times (Ding et al., 2015). This increase is a result of enhanced microdomain molecular interactions between positively charged polarities and hydrophilic surfaces at the microbe/electrode interface that facilitate biofilm formation and its conductivity (Ding et al., 2015). The wettability of the GF anode was enhanced through the use of a solvent, such as acetone or ethanol, followed by exposure to ammonium peroxydisulfate to remove any organic impurities (first step of the pre-treatment). Fig. 2 compares the voltammetric behaviour of the GF electrode treated with ethanol (GFE) or acetone (GFA) with respect to the non-treated electrode (GF0). As shown, GF0 generates the lowest current density (oxidation current density peak = 8.76 A m2), while when the electrode was exposed to ethanol the peak increased up to 12.7 A m2. The latter value was 14.41% higher than what observed for GFA, and 44.98% higher than the case of no treatment (GF0). The CV curve obtained with the GFE electrode is also characterised by much better defined redox peaks, which are almost identical and have a very small potential difference. On the basis of these results, ethanol was selected as the solvent for the first step of pre-treatment. It has been recently demonstrated that ethanol ca successfully enhance the wettability of graphite felt and remove any trapped air within the fibres (Smith et al., 2015). The ethanol-treated GF, had a contact angle as low as 21.57 ± 3.92° (Fig. S2B). The second step of the anode pre-treatment consisted of a thermal treatment to increase the electrode surface area (leading to GFE-T). This step further increased the oxidation current density generated, which was 60% larger than the value obtained with GFE, with no variation in the voltage difference. Several thermal treatments have been reported for the enhancement of the anode performance in MFCs. These include the addition of mediators and metals (Lowy et al., 2006; Park and Zeikus, 2003), or the modification of the carbon surface (Cheng and Logan, 2007; Feng et al., 2010). Cheng and Logan demonstrated a 20% enhancement in the power output when a carbon cloth electrode was treated at 700 °C with ammonia gas treatment (Cheng and Logan, 2007). Feng et al. reported a 34% increase in the power output with a combined acid-thermal treatment to a carbon electrode (Feng et al., 2010). SEM images of the GFE and GFE-T anodes (Fig. 3) suggest that the second step of the pre-treatment increases the surface area of the graphite fibres by generating cracks in their structure. This increase favours the mass transfer/electron transfer mechanisms (Chouler et al., 2016; Di Lorenzo et al., 2010), which is demonstrated by the greater redox peaks obtained with GFE-T shown in Fig. 2. The surface area enhancement also benefits the bacteria attachment in the MFC and, therefore, biofilm development, thus leading to better electrochemical performance (Guo et al., 2013). The electroactive surface area of the treated and non-treated graphite felt electrode was estimated as described in the Experimental Section. Fig. S3 in the Supplementary Information shows a linear relationship between scan rate and the height of the oxidation current peaks obtained with each electrode. It resulted that after the second step of pre-treatment the ESA was 1.2 times larger than the electrode not treated, and 1.1 times larger than the electrode that underwent only the first step of treatment.

113

S.M. Martinez, M. Di Lorenzo / Chemical Engineering Science 198 (2019) 108–116

B

A

1 m

1 m Fig. 3. SEM images of the GFE (A) and (B) GFE-T electrodes. (SEM magnification: 10,000).

Fig. 4. Enrichment of the anodic biofilm with digestate, under an external resistance of 510 X. Experimental data were interpolated with the modified Gompertz model (Eq. (4)). Error bars refer to two replicates.

3.2. Testing the performance of the MFC unit A floating air-cathode MFC device was developed. A very simple and cost-effective design was implemented, with no metal catalyst at the cathode and no external membrane. Both the cathode catalyst and the membrane account for the largest cost in the MFC technology. Consequently, there is an increasing interest in exploring alternative materials (Chouler et al., 2017, 2016) and/or adopting strategies to reduce the capital cost of MFCs and facilitate large scale applications. Particularly attractive are membrane-less devices. These devices have also the advantage of reducing the internal ohmic resistance of the system, which is related to the resistance to the transfer of protons across the membrane (Liu and Logan, 2004). The MFC was tested in a non-treated AD effluent (fresh digestate), which acted as bacteria source, fuel and electrolyte. Initially, the performance of individual MFC units was investigated. With this purpose, the MFCs were fitted into a beaker filled with 300 mL of digestate, enriched and operated individually. Since the cathode is directly exposed to digestate, both the anode and the cathode can be colonised by bacteria. It has been previously suggested that in membrane-less MFCs, the cathodic biofilm acts as a living membrane that prevents oxygen diffusion to the anode (Chouler et al., 2017). On the other hand, since no aeration nor agitation is implemented in the system, it is expected that a vertical stratification with different redox zones naturally occurs in the digestate. Only the surface layer of the digestate is aerobic, with microorganisms consuming oxygen faster than it can penetrate into the digestate. Below the aerobic layer, successive layers,

where nitrate, iron and manganese, sulfate, and carbon dioxide respectively, are used as electron acceptors or oxidants. This energetic hierarchy of electron accepting processes has been defined as the thermodynamic ladder (Bethke et al., 2011). This natural stratification removes the need for a membrane in the MFC, provided that the electrode spacing is large enough. On the other hand, if the electrode spacing is too wide, the internal resistance of the system increases and consequently the power output is reduced (Ahn et al., 2014). In this work, a distance of 1.5 cm between the two electrodes was used, as preliminary tests showed that this was the best compromise for power generation (data not shown). Nevertheless, a more in-depth study on the optimal electrode spacing, along with an analysis of vertical microbial distribution in the digestate is required for the system optimisation. Fig. 4 shows the voltage versus time generated over a period of 31 days. As shown, after three days of acclimatisation, an exponential growth of anodic biofilm was observed, which corresponded to a 54.2% reduction in the COD value, starting from an initial value of 15,000 mg L1 (Table S1 in the Supplementary Information). After eight days of enrichment, the system was topped up with fresh digestate on a weekly basis to counteract volume losses by evaporation. This might be one of the reasons why the COD value stabilised to 9858 ± 121 mg L1 after approximately four days. After 11 days of operation, a pseudo-steady state was observed, with a constant current density of 104.76 ± 1.19 mA m2, which corresponded to a power density output value of 25.02 ± 0.60 mW m2. It was therefore assumed that after this period of time the biofilm enrichment onto the electrodes surface was completed (Babauta et al., 2012; Zhao et al., 2009). Polarisation tests, performed after four weeks of operation, showed a maximum power output of 76.88 ± 16.61 mW m2 (Fig. S4). The modified Gompertz model (Eq. (4)) was used to derive the maximum specific growth rate, mmax, by considering that the limiting factor for electricity production during transient operations is the bacteria growth (Mateo et al., 2018):

   lmax  e E ¼ Emax  exp  exp  ðk  t Þ þ 1 Emax

ð4Þ

where Emax is the maximum voltage achievable under close circuit conditions, k is the lag time and e is the Napier’s constant (2.71828). It resulted that the mmax was 0.00575 d1, considering the experimental values of Emax equal to 0.234 ± 0.008 V at 510 X and k of 3 days (Fig. 4). So far, only a few studies report the use of non-treated AD effluents as feedstock and electrolyte in MFCs. Premier et al. (2013) have reported the integration of a tubular air-cathode MFC with a methanogenesis treatment step to remove volatile fatty acids (VFAs) and the associated odours in digestate, as well as generating electricity. Fradler et al. (2014) have reported the generation of

114

S.M. Martinez, M. Di Lorenzo / Chemical Engineering Science 198 (2019) 108–116

A

B

1 m

1 m

100 m

100 m

Fig. 5. SEM images of the GF anodic surface before being in contact with fresh digestate (A), and after a month of operation (B). (SEM magnifications: 200; and for the insets 10,000).

A

B

0.5

0.5

0.4

0.4

0.3

0.3

Ecell / V

P / mW

P2

0.2 0.1 0

P1

0.1

P1 0

P2

0.2

1

2

3

I / mA

4

5

6

0

0

1

2

3

4

5

6

I / mA

Fig. 6. Power (A) and polarisation (B) curves of the two MFC stacks studied, P1 and P2, obtained by electrically connecting in parallel two (P1) and 10 (P2) MFC units in digestate after 25 days of operation.

0.1

P / mW

0.08 0.06 0.04

0.02 0

2

4

6

8

10

Number of MFC units Fig. 7. Effect of stacking several MFC units in parallel on the overall power generated. The first dot in the graph refers to the stacking of two cells to up to 10 cells stacked.

5.5 W m3 from anaerobic digester sludge with a tubular twochambered four-module MFC reactor in continuous flow mode. Weld and Singh (2011) reported the integration of the MFC technology with an anaerobic digester, showing a maximum power density output of 0.13 W m2. A direct comparison of performance between these systems and the MFC that we report here is, however, difficult, due to differences in design and working conditions. All these studies refer in fact to the use of Pt catalysts at the cathode and the use of a membrane, while our system is characterised by a much simpler and cost-effective design. Moreover, our MFC is operated under static (batch) conditions.

The anodic biofilm after 30 days of operation was visualised by SEM (Fig. 5). As shown, the GF fibres were covered with a dense network of filaments that interconnect the bacteria cells (dimensions: 1–2 lm of length and 0.5 of width), which are populated by spherical- and rod-shaped bacteria. This coverage demonstrates that the treated GF fibres are a good support for the anodic biofilm, due to a better substrate accessibility within the fibres and the enhanced roughness of the fibres after the treatment. Nonetheless, the cell density decreases gradually with inward-depth. This trend is typically observed in 3D electrodes, typically characterised by denser biofilm onto the electrode surface (Chae et al., 2009). A further investigation by advanced DNA sequencing method, would allow the identification of the species developed in the biofilm and the species distribution within the electrode thickness 3.3. Scaling up the power output Taking into account the thermodynamic limitation of a single microbial fuel cell (1.1 V at OCV) (Zielke, 2006), the most promising strategy to scale up the power output generated with this technology is to arrange multiple MFC units into stacks (Ieropoulos et al., 2013). In this study, we have investigated two staking configurations, identified as P1 and P2, in which the MFC units were electrically connected in parallel and share the same electrolyte (digestate). Firstly, a system with only two MFC units was considered (P1). Subsequently, up to 10 units were connected together (P2). The performance of the resulting stacks was investigated in terms of overall power generation. Fig. 6 shows the overall power and cell potential generated by the two stacks after 25 days of operation.

S.M. Martinez, M. Di Lorenzo / Chemical Engineering Science 198 (2019) 108–116

The current output was constant over time as a consequence of a stable value in the COD of the digestate. As observed in Fig. S5, after a 52% initial decay in the COD during the first five days of operation in batch, the COD of the digestate was maintained constant at 9945.81 ± 0.28 mg L1 by topping up on a daily basis the system with fresh digestate. As observed from Fig. 6A, P2 reached a maximum power of 0.43 mW at a current of 2.5 mA, while P1 produced 0.089 mW as maximum power at a current of 0.58 mA. The overall maximum power output generated by P2 was, therefore, approximately five times larger than the one generated by P1 and, nearly 10 times larger than what generated by a single unit (0.045 mW). Successful power scale-up by arranging multiple MFCs in stack has been already demonstrated (Ieropoulos et al., 2008; Ledezma et al., 2013). Fig. 7 shows that the power output (calculated from the recorded voltage at the steady state) increase is linear with the number of MFC units connected in parallel. This is the result of a functional increase in the electrode surface area per volume ratio, resulting from stacking together the MFC units, with consequently larger surface available for the electrochemical reactions and a higher volumetric power density. From the polarisation tests, it results that activation and ohmic losses dominate in both P1 and P2 (Fig. 6B). Activation losses are associated with slow reaction rates taking place at the electrode surface, while ohmic losses are related to the internal resistance, Rint, which might result from ions-flow resistance trough the digestate and from any electrical resistance in the system (i.e. electrodes and connections). When the MFC units are operated in a parallel configuration, the MFC stack tends towards the lowest Rint (Chouler et al., 2016). In P1, the internal resistance, RP1 int, is reduced to 307.6 X, compared to 624.5 ± 50.8 X obtained with a single cell. For the case of P2, RP2 int is further reduced to 65.5 X. Since the focus of the study was on energy generation only, the ability of the MFC arrays to further treat the digestate was not investigated. Moreover, except for the first five days, during which the arrays were operated in batch, a fed-batch operation was considered. To counteract volume losses by evaporation, the pond hosting the array was in fact topped-up on a daily basis with untreated fresh digestate, as reported in the Methods section. The COD of the digestate in the pond is the only parameter that has been regularly measured during operation, and, as shown in Fig. S5, it was kept constant during the fed-batch operation. 4. Conclusions We report an innovative floating microbial fuel cell system, characterised by a simple and cost-effective design and we demonstrate bioelectricity generation from fresh digestate without the addition of an external carbon source or bacterial inoculum. Prior to being used in the MFC, the anode electrode underwent a twostage treatment to enhance the wettability of the carbon fibres (stage 1) and the specific surface area (stage 2). It resulted that stage 1 decreased the contact angle of 85.7% and enhanced the electroactivity of 4%. Further electroactivity increase of 9% was obtained after stage 2. The performance of the MFC unit was first tested individually and then in a stack of up to 10 MFCs, sharing the same electrolyte and electrically connected in parallel. A linear power output increase was observed with the sequential additions of MFC units in the stack, reaching a peak value of 0.43 mW (corresponding to a power density of 51 ± 2 mW m3) at a current of 2.5 mA. Over a billion tons of food is wasted every year worldwide. This waste is typically incinerated or dumped in open area, posing sev-

115

ere issues to public health and to the environment. An appropriate management of food waste could include the integrated use of anaerobic digestion and microbial fuel cells for waste treatment while generating products of interest, such as fertilisers and bioenergy. The simplicity of the design implemented in this study overcomes the cost limitations of many MFC devices previously reported that use either expensive membrane and/or Pt or other expensive catalysts at the cathode. As such, not only the manufacturing process and the system scale-up is drastically simplified, but also the proposed stack of MFCs can pave the way for a sustainable and environmentally-friendly food waste treatment which is affordable also to the poorest areas in the world. Declaration of interests The authors declared that there is no conflict of interest. Acknowledgements This work has been developed within the AgroCycle project, funded by the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 690142. The authors would like also to thank: Ursula Potter and Diana Lednitzky for their help with the SEM imaging, Microscopy and Analysis Suite, University of Bath (UK); Wesley Wong from Wessex Water (UK) for kindly providing us the digestate used in this work. Appendix A. Supplementary material Supplementary data to this article can be found online at https://doi.org/10.1016/j.ces.2018.12.039. References Ahn, Y., Hatzell, M.C., Zhang, F., Logan, B.E., 2014. Different electrode configurations to optimize performance of multi-electrode microbial fuel cells for generating power or treating domestic wastewater. J. Power Sources 249, 440–445. https:// doi.org/10.1016/j.jpowsour.2013.10.081. Artyushkova, K., Cornejo, J.A., Ista, L.K., Babanova, S., Santoro, C., Atanassov, P., Schuler, A.J., 2015. Relationship between surface chemistry, biofilm structure, and electron transfer in Shewanella anodes. Biointerphases 10, 019013. https:// doi.org/10.1116/1.4913783. Babauta, J., Renslow, R., Lewandowski, Z., Beyenal, H., 2012. Electrochemically active biofilms: facts and fiction. A review. Biofouling. https://doi.org/10.1080/ 08927014.2012.710324. Behera, M., Jana, P.S., More, T.T., Ghangrekar, M.M., 2010. Rice mill wastewater treatment in microbial fuel cells fabricated using proton exchange membrane and earthen pot at different pH. Bioelectrochemistry 79, 228–233. https://doi. org/10.1016/j.bioelechem.2010.06.002. Bethke, C.M., Sanford, R.A., Kirk, M.F., Jin, Q., Flynn, T.M., 2011. The thermodynamic ladder in geomicrobiology. Am. J. Sci. 311, 183–210. https://doi.org/10.2475/ 03.2011.01. Bose, D., Kandpal, V., Dhawan, H., Vijay, P., Gopinath, M., 2018. Energy recovery with microbial fuel cells: bioremediation and bioelectricity, p. 7–33. https://doi.org/ 10.1007/978-981-10-7413-4_2. Chae, K.J., Choi, M.J., Lee, J.W., Kim, K.Y., Kim, I.S., 2009. Effect of different substrates on the performance, bacterial diversity, and bacterial viability in microbial fuel cells. Bioresour. Technol. 100, 3518–3525. https://doi.org/10.1016/j. biortech.2009.02.065. Cheng, S., Logan, B.E., 2007. Ammonia treatment of carbon cloth anodes to enhance power generation of microbial fuel cells. Electrochem. commun. 9, 492–496. https://doi.org/10.1016/j.elecom.2006.10.023. Chouler, J., Bentley, I., Vaz, F., O’Fee, A., Cameron, P.J., Di Lorenzo, M., 2017. Exploring the use of cost-effective membrane materials for Microbial Fuel Cell based sensors. Electrochim. Acta 231, 319–326. https://doi.org/10.1016/ j.electacta.2017.01.195. Chouler, J., Padgett, G.A., Cameron, P.J., Preuss, K., Titirici, M.M., Ieropoulos, I., Di Lorenzo, M., 2016. Towards effective small scale microbial fuel cells for energy generation from urine. Electrochim. Acta 192, 89–98. https://doi.org/10.1016/ j.electacta.2016.01.112. Costa, A., Ely, C., Pennington, M., Rock, S., Staniec, C., Turgeon, J., 2015. Anaerobic Digestion and its Applications. United States Environ. Prot. Agency, p. 15.

116

S.M. Martinez, M. Di Lorenzo / Chemical Engineering Science 198 (2019) 108–116

De Baere, L., 2000. Worldwide state-of-the-art of anaerobic digestion of solid waste. Water Sci. Technol. 41, 283–290. Di Lorenzo, M., Scott, K., Curtis, T.P., Head, I.M., 2010. Effect of increasing anode surface area on the performance of a single chamber microbial fuel cell. Chem. Eng. J. 156, 40–48. https://doi.org/10.1016/j.cej.2009.09.031. Ding, C.M., Lv, M.L., Zhu, Y., Jiang, L., Liu, H., 2015. Wettability-regulated extracellular electron transfer from the living organism of shewanella loihica PV-4. Angew. Chem. – Int. Ed. 54, 1446–1451. https://doi.org/10.1002/ anie.201409163. FAO, 2018. Food loss and waste reduction. Food loss food waste 26–27. Feng, Y., Yang, Q., Wang, X., Logan, B.E., 2010. Treatment of carbon fiber brush anodes for improving power generation in air-cathode microbial fuel cells. J. Power Sources 195, 1841–1844. https://doi.org/10.1016/j. jpowsour.2009.10.030. Fradler, K.R., Kim, J.R., Shipley, G., Massanet-Nicolau, J., Dinsdale, R.M., Guwy, A.J., Premier, G.C., 2014. Operation of a bioelectrochemical system as a polishing stage for the effluent from a two-stage biohydrogen and biomethane production process. Biochem. Eng. J. 85, 125–131. https://doi.org/10.1016/j. bej.2014.02.008. Gao, A., Tian, Z., Wang, Z., Wennersten, R., Sun, Q., 2017. Comparison between the technologies for food waste treatment. Energy Proc. 105, 3915–3921. https:// doi.org/10.1016/j.egypro.2017.03.811. Goud, R.K., Babu, P.S., Mohan, S.V., 2011. Canteen based composite food waste as potential anodic fuel for bioelectricity generation in single chambered microbial fuel cell (MFC): bio-electrochemical evaluation under increasing substrate loading condition. Int. J. Hydrogen Energy 36, 6210–6218. https://doi.org/ 10.1016/j.ijhydene.2011.02.056. Guo, K., Freguia, S., Dennis, P.G., Chen, X., Donose, B.C., Keller, J., Gooding, J.J., Rabaey, K., 2013. Effects of surface charge and hydrophobicity on anodic biofilm formation, community composition, and current generation in bioelectrochemical systems. Environ. Sci. Technol. 47, 7563–7570. https://doi. org/10.1021/es400901u. Heilmann, J., Logan, B.E., 2006. Production of electricity from proteins using a microbial fuel cell. Water Environ. Res. 78, 531–537. https://doi.org/10.2175/ 106143005X73046. Holm-Nielsen, J.B., Al Seadi, T., Oleskowicz-Popiel, P., 2009. The future of anaerobic digestion and biogas utilization. Bioresour. Technol. 100, 5478–5484. https:// doi.org/10.1016/j.biortech.2008.12.046. Ieropoulos, I., Greenman, J., Melhuish, C., 2008. Microbial fuel cells based on carbon veil electrodes: Stack configuration and scalability. Int. J. Energy Res. 32, 1228– 1240. https://doi.org/10.1002/er.1419. Ieropoulos, I., Theodosiou, P., Taylor, B., Greenman, J., Melhuish, C., 2017. Gelatin as a promising printable feedstock for microbial fuel cells (MFC). Int. J. Hydrogen Energy 42, 1783–1790. https://doi.org/10.1016/j.ijhydene.2016.11.083. Ieropoulos, I.A., Greenman, J., Melhuish, C., 2013. Miniature microbial fuel cells and stacks for urine utilisation. Int. J. Hydrogen Energy 38, 492–496. https://doi.org/ 10.1016/j.ijhydene.2012.09.062. Karthikeyan, O.P., Mehariya, S., Chung Wong, J.W., 2017. Bio-refining of food waste for fuel and value products. Energy Proc. 136, 14–21. https://doi.org/10.1016/ j.egypro.2017.10.253. Kim, T., An, J., Jang, J.K., Chang, I.S., 2015. Coupling of anaerobic digester and microbial fuel cell for COD removal and ammonia recovery. Bioresour. Technol. 195, 217–222. https://doi.org/10.1016/j.biortech.2015.06.009. Ledezma, P., Greenman, J., Ieropoulos, I., 2013. MFC-cascade stacks maximise COD reduction and avoid voltage reversal under adverse conditions. Bioresour. Technol. 134, 158–165. https://doi.org/10.1016/j.biortech.2013.01.119. Li, H., Tian, Y., Zuo, W., Zhang, J., Pan, X., Li, L., Su, X., 2016. Electricity generation from food wastes and characteristics of organic matters in microbial fuel cell. Bioresour. Technol. 205, 104–110. https://doi.org/10.1016/j. biortech.2016.01.042. Liu, H., Logan, B.E., 2004. Electricity generation using an air-cathode single chamber microbial fuel cell in the presence and absence of a proton exchange membrane. Environ. Sci. Technol. 38, 4040–4046. https://doi.org/10.1021/es0499344. Logan, B.E., Hamelers, B., Rozendal, R., Schröder, U., Keller, J., Freguia, S., Aelterman, P., Verstraete, W., Rabaey, K., 2006. Microbial fuel cells: methodology and technology. Environ. Sci. Technol. 40, 5181–5192. https://doi.org/10.1021/ es0605016. Lowy, D.A., Tender, L.M., Zeikus, J.G., Park, D.H., Lovley, D.R., 2006. Harvesting energy from the marine sediment-water interface II. Kinetic activity of anode materials. Biosens. Bioelectron. 21, 2058–2063. https://doi.org/10.1016/j. bios.2006.01.033. Mateo, S., Mascia, M., Fernandez-Morales, F.J., Rodrigo, M.A., Di Lorenzo, M., 2018. Assessing the impact of design factors on the performance of two miniature microbial fuel cells. Electrochim. Acta 297, 297–306. https://doi.org/10.1016/ j.electacta.2018.11.193. Min, B., Kim, J.R., Oh, S.E., Regan, J.M., Logan, B.E., 2005. Electricity generation from swine wastewater using microbial fuel cells. Water Res. 39, 4961–4968. https:// doi.org/10.1016/j.watres.2005.09.039. Mohan, S.V., Chandrasekhar, K., 2011. Solid phase microbial fuel cell (SMFC) for harnessing bioelectricity from composite food waste fermentation: influence of

electrode assembly and buffering capacity. Bioresour. Technol. 102, 7077–7085. https://doi.org/10.1016/j.biortech.2011.04.039. Moqsud, M.A., Omine, K., Yasufuku, N., Bushra, Q.S., Hyodo, M., Nakata, Y., 2014. Bioelectricity from kitchen and bamboo waste in a microbial fuel cell. Waste Manag. Res. 32, 124–130. https://doi.org/10.1177/0734242X13517160. Nastro, R.A., Falcucci, G., Hodgson, D.M., Minutillo, M., Trifuoggi, M., Guida, M., Avignone-Rossa, C., Dumontet, S., Jannelli, E., Ulgiati, S., 2015. Utilization of Agro-industrial and Urban Waste as Feedstock in Microbial Fuel Cells (MFCs). Glob. Clean. Prod Conf. Park, D.H., Zeikus, J.G., 2003. Improved fuel cell and electrode designs for producing electricity from microbial degradation. Biotechnol. Bioeng. 81, 348–355. https://doi.org/10.1002/bit.10501. Pham, T.H., Rabaey, K., Aelterman, P., Clauwaert, P., De Schamphelaire, L., Boon, N., Verstraete, W., 2006. Microbial fuel cells in relation to conventional anaerobic digestion technology. Eng. Life Sci. 6, 285–292. https://doi.org/10.1002/ elsc.200620121. Premier, G.C., Kim, J.R., Massanet-Nicolau, J., Kyazze, G., Esteves, S.R.R., Penumathsa, B.K.V., Rodríguez, J., Maddy, J., Dinsdale, R.M., Guwy, A.J., 2013. Integration of biohydrogen, biomethane and bioelectrochemical systems. Renew. Energy 49, 188–192. https://doi.org/10.1016/j.renene.2012.01.035. Press, N.A., Academy, N., 2006. Waste Incineration and Pubic Health. . Rabaey, K., Verstraete, W., 2005. Microbial fuel cells: novel biotechnology for energy generation. Trends Biotechnol. 23, 291–298. https://doi.org/10.1016/j. tibtech.2005.04.008. Reeb, J., Milota, M., 1999. Moisture content by the oven-dry method for industrial testing weight of water weight of wood MC –. Wdka, 66–74. https://doi.org/ 10.1016/j.apsusc.2013.11.072. Rice, E.W., Baird, R.B., Eaton, A.D., 2017. Standard methods for the evaluation of water and wastewater 23rd Ed. EPUB. Rodrigo, M.A., Cañizares, P., Lobato, J., Paz, R., Sáez, C., Linares, J.J., 2007. Production of electricity from the treatment of urban waste water using a microbial fuel cell. J. Power Sources 169, 198–204. https://doi.org/10.1016/j. jpowsour.2007.01.054. Sarjit, A., Mei Tan, S.A., Dykes, G., 2015. Surface modification of materials to encourage beneficial biofilm formation. AIMS Bioeng. 2, 404–422. https://doi. org/10.3934/bioeng.2015.4.404. Singh, H.M., Pathak, A.K., Chopra, K., Tyagi, V.V., Anand, S., Kothari, R., 2018. Microbial fuel cells: a sustainable solution for bioelectricity generation and wastewater treatment. Biofuels 7269, 1–21. https://doi.org/10.1080/ 17597269.2017.1413860. Smith, R.E.G., Davies, T.J., Baynes, N.D.B., Nichols, R.J., 2015. The electrochemical characterisation of graphite felts. J. Electroanal. Chem. 747, 29–38. https://doi. org/10.1016/j.jelechem.2015.03.029. Stenmark, Å., Jensen, C., Quested, T., Moates, G., 2016. Estimates of European food waste levels, IVL-report C 186. . Stevens, N.P.C., Rooney, M.B., Bond, A.M., Feldberg, S.W., 2001. A comparison of simulated and experimental voltammograms obtained for the [Fe(CN) 6] 3-/4couple in the absence of added supporting electrolyte at a rotating disk electrode. J. Phys. Chem. A 105, 9085–9093. https://doi.org/10.1021/jp0103878. Thyberg, K.L., Tonjes, D.J., 2016. Drivers of food waste and their implications for sustainable policy development. Resour. Conserv. Recycl. 106, 110–123. https:// doi.org/10.1016/j.resconrec.2015.11.016. Venkata Mohan, S., Mohanakrishna, G., Sarma, P.N., 2010a. Composite vegetable waste as renewable resource for bioelectricity generation through noncatalyzed open-air cathode microbial fuel cell. Bioresour. Technol. 101, 970– 976. https://doi.org/10.1016/j.biortech.2009.09.005. Venkata Mohan, S., Mohanakrishna, G., Velvizhi, G., Babu, V.L., Sarma, P.N., 2010b. Bio-catalyzed electrochemical treatment of real field dairy wastewater with simultaneous power generation. Biochem. Eng. J. 51, 32–39. https://doi.org/ 10.1016/j.bej.2010.04.012. Wang, X., Feng, Y., Wang, H., Qu, Y., Yu, Y., Ren, N., Li, N., Wang, E., Lee, H., Logan, B. E., 2009. Bioaugmentation for electricity generation from corn stover biomass using microbial fuel cells. Environ. Sci. Technol. 43, 6088–6093. https://doi.org/ 10.1021/es900391b. Weld, R.J., Singh, R., 2011. Functional stability of a hybrid anaerobic digester/ microbial fuel cell system treating municipal wastewater. Bioresour. Technol. 102, 842–847. https://doi.org/10.1016/j.biortech.2010.09.002. World Resources Institute, 2013. Creating a sustainable food future: a menu of solutions to sustainably feed more than 9 billion people by 2050. World Resour. Rep. 2013–14, 130. https://doi.org/10.1016/S0264-8377(03)00047-4. Yu, Z., Leng, X., Zhao, S., Ji, J., Zhou, T., Khan, A., Kakde, A., Liu, P., Li, X., 2018. A review on the applications of microbial electrolysis cells in anaerobic digestion. Bioresour. Technol. 255, 340–348. https://doi.org/10.1016/j. biortech.2018.02.003. Zhao, F., Slade, R.C.T., Varcoe, J.R., 2009. Techniques for the study and development of microbial fuel cells: an electrochemical perspective. Chem. Soc. Rev. 38, 1926–1939. https://doi.org/10.1039/b819866g. Zielke, E.A., 2006. Thermodynamic analysis of a single chamber microbial fuel cell. Analysis, 2.