1
Water formation at the cathode and sodium recovery using Microbial fuel cells (MFCs)
2
Iwona Gajdaa, John Greenmana,b, Chris Melhuisha, Carlo Santoroc,d, Baikun Lic,d, Pierangela
3
Cristianie, Ioannis Ieropoulosa,b*
4
5
6
7
8
9
10
aBristol
Robotics Laboratory, Block T , UWE, Bristol, Coldharbour Lane, Bristol BS16 1QY
bSchool
cDepartment
of Civil and Environmental Engineering, University of Connecticut, 261 Glenbrook rd, Storrs, CT 06269, USA
dCenter
eRSE
of Life Sciences, UWE, Bristol, Coldharbour Lane, Bristol BS16 1QY
for Clean Energy Engineering, University of Connecticut,44 Weaver rd, Storrs, CT 06269, USA
– Ricerca sul Sistema Energetico S.p.A., Environment and Sustainable Development Dept., Via Rubattino 54, 20134
Milan, Italy
*ioannis.ieropoulos@brl.ac.uk
11
12
Abstract
13
Microbial Fuel Cells (MFCs) utilise biodegradable carbon compounds in organic waste to
14
generate electric current. The aim of this work was to enhance MFC performance by using
15
low cost and catalyst (platinum)-free cathode materials. The results showed that the range of
16
Pt-free cathodes including activated carbon, plain carbon fibre veil with and without
17
microporous layer (MPL) in two-chamber MFC generated power with simultaneous catholyte
18
generation in the cathode chamber. This is the first time to report a clear catholyte formation
19
on the cathode half cell, which was directly related to MFC power performance. The
20
importance of this phenomenon may be attributed to the oxygen reduction reaction, water
21
diffusion and electroosmotic drag. The synthesised catholyte in situ on the open-to-air
22
cathode appeared to be sodium salts (9% concentration), which was recovered from the
23
anolyte feedstock containing sludge and sodium acetate. An overlooked benefit of catholyte
24
formation and accumulation contributes greatly to the overall wastewater treatment, water
25
recovery, bioremediation of salts and carbon capture.
26
Keywords: Microbial Fuel Cell (MFC), carbon veil cathodes, Microporous Layer (MPL),
27
electroosmotic drag, oxygen reduction reaction, wet scrubbing
28
1. Introduction
29
The world of the 21st century is facing shortages in fresh water as well as increased electricity
30
demand, which is inspiring the development of new, alternative technologies of wastewater
31
treatment and power generation. One of the most promising solutions, addressing both
32
challenges, is the Microbial Fuel Cell (MFC) technology, whose bio-catalytic activity of
33
microbes digesting organic matter allows the direct generation of electricity [1]. Biomass in
34
wastewater contains a significant amount of energy, and its utilisation as a power source may
35
have an important role in helping to secure a sustainable future. The most attractive aspect of
36
the process is that the generation of electricity from wastewater is resulting in cleaner waste.
37
The Microbial Fuel Cell (MFC) comprises an anode and a cathode chambers, separated by a
38
proton selective membrane. The electrons are generated by the metabolic pathways of the
39
microbes digesting biomass into CO2, electrons and protons. For each electron that is
40
produced, an equivalent proton must be transported to the cathode through the electrolyte,
41
which sustains the current flow. The oxygen reduction reaction (ORR) at the
42
membrane/cathode interface consumes oxygen and generates water or hydrogen peroxide.
43
According to Kinoshita [2], under alkaline conditions, the H2O2 is then transformed from both
44
electrochemical and chemical processes [2].
45
Peroxide pathway in alkaline solutions
46
O2 +H2O +2e- → HO2- +OH-
47
Followed by reduction of peroxide:
48
HO2- + H2O +2e- → 3 OH-
49
or decomposition :
50
2 HO2- - → 3 OH- + O2
51
Air cathode MFCs are the most promising configuration for practical applications including
52
wastewater treatment and power generation due to their improved electrical power output and
53
operational simplicity [3]. However, the ORR of a working MFC in comparison to chemical
54
fuel cells is limited due to the MFC neutral pH and ambient operating temperature. Overall,
55
the cathode is being usually the limiting factor for power generation, which hinders the
56
overall system performance. [4, 5]. The utilization of platinum as a cathode catalyst would
57
certainly enhance the overall performances but it would elevate the total MFC cost.
58
Unfortunately, the platinum would have also been poisoned by a non-sterile environment in a
59
relatively short amount of time [6]; therefore there is a critical need for low-cost catalysts for
60
future long-term MFC applications. Materials with high surface area and different surface
61
treatments have been used to increase the rate of ORR and improve the overall MFC
62
performance. For example, granular graphite, micro-porous layer and activated carbon, have
63
been reported to support cathodic oxygen reduction due to the very large specific surface area
64
[7-9] and high porosity that would also improve the water management.
65
In Proton Exchange Membrane Fuel Cells (PEMFCs) the water management is critical for
66
achieving high performance. The polymer exchange membrane requires to be sufficiently
67
hydrated to maintain high ionic conductivity. During fuel cell operation, water molecules
68
migrate through the membrane due to electro-osmotic drag and diffusion and water is also
69
being generated at the cathode/membrane interface due to the ORR. If the water generated is
70
not removed, cathode flooding may hinder the oxygen transport on the catalyst active sites
71
and severely penalize the output [10]. In PEMFCs, the liquid water transport mechanism is
72
one of the most dominant parameter influencing the performance of the cathode and current
73
density-based loss distribution [11]. Water is also essential for the operation of MFC,
74
providing an electrolyte bridge between anolyte, membrane and the cathode, however, there
75
are very few MFC works studying water production (via oxygen reduction), water transport
76
from the anode and its sufficiency for supporting cathode reaction [12]; thus, in the open to
77
air cathode design an external supply of water is necessary [13]. In addition, it has been found
78
that in MFCs cations, other than protons, are more likely to complete the circuit via
79
transportation from the anode to the cathode [14] allowing their recovery from wastewater.
80
With the ionic movement that will drag water molecules through electroosmosis, water also
81
can be recovered. Similar concept has been recently utilized for bioelectrochemical systems
82
with the purpose of desalinization of water [15]. Moreover it has also been recognised that
83
the flooding of the cathode is a negative factor in MFCs [16] attributing it to hindered oxygen
84
mass transfer by catalyst sites filled with water and pursuing a direction of development for
85
anti-flooding binders [17]. However, from the wastewater treatment point of view, the
86
formation and accumulation of catholyte might serve as a huge advantage to the MFC
87
technology if catholyte flooding is correctly managed. This can be achieved through the
88
cathode electrode design and its configuration within the reactor. The generated or
89
transported water from the anolyte to the cathode chamber in the MFC system plays an
90
important role in wastewater treatment as an additional waste cleanup method, synthesis and
91
bioremediation (recovery) of elements.
92
This work aims to: (i) explore the water transport phenomenon in the dual-chamber MFC
93
system in relation to power production in order to understand the mechanism of catholyte
94
flow on carbon porous electrodes as an opportunity for biosynthesis, remediation and
95
recovery of useful components from the wastewater. (ii) Present for the first time the
96
catholyte formation in situ on the range of carbon based cathode electrodes; and (iii) analyse
97
the produced catholyte in terms of its salt composition and other elements from the anodic
98
wastewater, and its potential for carbon capture in wet scrubbing open to air cathode MFCs.
99
100
2. Materials and Methods
101
2.1.MFC design and operation
102
Twelve tested MFC reactors comprised 25mL anode chambers and 25mL cathode chambers
103
as previously described [18] separated by a cation exchange membrane (CMI Membranes
104
International, USA). Already established and well matured anodes were used from previous
105
MFC experiments, which have been running for at least 6 months under pseudo steady-state
106
conditions. Anode electrodes were made from carbon fibre veil with a carbon loading of
107
20g/m2 (PRF Composite Materials, Poole, Dorset, UK) and had a total macro surface area of
108
270cm², folded into 3D rectangular cuboids (geometric surface area of 36cm2) in order to fit
109
into the chamber and be fully suspended in the anolyte fluid. The cathode electrodes were
110
used as shown in Table 1 and placed in transparent MFC chambers with 2 vents (0.3mm
111
diameter) at the top. The anolyte was recirculated using a 16-channel peristaltic pump (205U,
112
Watson Marlow, UK) with a flow rate of 47mL/h. The tubing was made of silicone, its length
113
was 40cm from the pump to the MFC, 40cm from the MFC to the 1L anolyte reservoir and
114
90cm from the reservoir back to the pump, periodically supplemented with fresh sludge
115
mixed with 20mM sodium acetate as a carbon-energy (C/E) source, in order to maintain C/E-
116
replete conditions and avoid having the anode as the limiting half-cell.
117
2.2.Cathode electrodes preparation
118
The same folded carbon veil electrode used for the anode was used as the control for the
119
cathode (CV) and was the only type of 3D cathode electrode used. Microporous layer on
120
carbon cloth and on carbon veil was prepared as previously described [19]. Activated carbon
121
(activated charcoal, BET area of 802 m2g-1, Calgon, Pittsburgh, PA) was prepared with a
122
loading of 60±2 mgAC cm-2 and PTFE (20%wt) were mixed using a blender and pressed on a
123
30% wt PTFE treated carbon cloth (Fuel Cell Earth) that was used as the current collector in
124
MFCs. The AC cathodes were prepared under pressure force (1400 psi) for 2 minutes and
125
then heated at 200°C for 1 hour [20].
126
The cathode chamber had the additional hollow space at the bottom and included a syringe to
127
allow the collection of the catholyte produced (see Fig.2). All 12 MFCs were divided in 4
128
experimental groups of identical triplicates (Table 1). No catalysts or buffers were used.
129
Table 1. Types of electrodes in the open to air cathode half cells.
Electrode Name
Electrode Type
Cathode type, size
CV
Carbon veil (Control) folded
3Dimensional (folded), 270cm2
MPL
Carbon cloth with Microporous layer
2D 10cm2
CV MPL
Carbon veil with Microporous layer
2D 10cm
AC
Activated carbon on carbon cloth
2D 10cm2
130
131
2.3. Measurement and Calculation
132
Voltage (V) was measured in millivolts (mV) and monitored using an ADC-24 Channel Data
133
Logger (Pico Technology Ltd., Cambridgeshire, U.K.), recorded data were processed using
134
GraphPad Prism® version 5.01 software package (GraphPad, California, U.S.A.). Current in
135
amperes (A) was determined using Ohm’s law, I=V/R, where (R) is the external resistor load
136
in ohms (Ω). Power in watts (W) was calculated using Joule’s law P=IV. Polarisation
137
experiments were performed by connecting a variable resistor, with a range between
138
30,000Ω-10Ω. Resistance was changed every 3 minutes, with data recorded every 30
139
seconds. Anodes were inoculated with anaerobic activated sludge provided by the Wessex
140
Water Scientific Laboratory (Saltford, UK). Sludge was mixed with 20mM acetate and used
141
as feedstock. Anodes were connected to the 16-channel peristaltic pump (205U, Watson
142
Marlow, Falmouth, UK) with the use of silicon tubing and 1L of anolyte bottle to maintain
143
anolyte recirculation at flow rate of 47 mL/h.
144
2.4. Analysis of the accumulated catholyte”water”
145
The pH was measured with Hanna 8424 pH meter (Hanna, UK) and the conductivity with
146
470 Jenway conductivity meter (Camlab, UK). Dry weight of precipitated salts was
147
determined by drying 0.5mL of catholyte over 48h and weighing the dry mass. Samples were
148
prepared for SEM microscopy by sputter coating in gold using an Emscope SC500 sputter
149
coating unit. Images were observed and captured using a Philips XL30 scanning electron
150
microscope (SEM). Energy dispersive x-ray (EDX) analysis was performed (Philips XL30
151
SEM) and it was used to determine elements present in precipitated cathodic salts. Detection
152
limits are typically 0.1-100%wt. The ICP-OES (Varian Inc. Vista-Pro ICP-OES using Axial
153
Plasma) was used for metal analysis of the catholyte samples. GC-MS was performed to
154
determine organic compounds in the collected catholyte by liquid extraction 1:1 with ethyl
155
acetate. XRD (X-ray Diffraction) analysis on precipitated salts from the catholyte was
156
determined using powder measurements, performed on a Bruker D8 Advance Diffractometer
157
with the results being analysed using EVA software package (Bruker, UK). For the single
158
crystal measurement, a unit cell check was performed on a Bruker Microstar Cu-anode 4-
159
circle diffractometer with the results being analysed with APEX II (Bruker, UK)
160
3. Results and Discussion
161
3.1. Polarisation experiments of cathode materials
162
To assess the maximum power point (MPP), a polarisation experiment was performed; Fig.1
163
presents the polarisation data where the maximum power recorded for the different electrodes
164
was: AC 498μW; MPL 419 μW; CVMPL 313 μW and CV121 μW. From these, it can be
165
concluded that the 2D pressed cathode type materials such as AC, MPL, MPL CV are
166
performing 4, 3.4, and 2.5-fold better, respectively compared to the control 3D cathode (CV).
167
The improved performance might have been due to the higher specific surface area of the
168
activated carbon in AC and MPL and the better contact with the PEM, which might have
169
resulted in an improved ORR rate.
MPL Power
MPL Voltage
CV Power
CV Voltage
600
400
AC Power
AC Voltage
500
300
400
300
200
Voltage[mV]
Power[W]
CV MPL Power
CV MPL Voltage
200
100
100
0
0
1000
2000
4000
5000
0
6000
Current (µA)
170
171
172
3000
Figure 1. Polarisation data of all tested MFCs (mean value of tiplicate MFCs)
173
3.2. Catholyte accumulation
174
Catholyte formation on the surface of the cathode electrode was first observed as droplets,
175
visible to the naked eye (Fig. 2, left) and further on as accumulation of the catholyte from all
176
the tested MFCs. The newly formed catholyte was consistently clear in colour (Fig.2, right)
177
and when evaporated left a residue suggesting high salt content, which is soluble in water
178
(Fig.2, centre). At the same time, anolyte loss was observed. To explore the catholyte
179
accumulation phenomenon, all MFCs were connected to 2.4kΩ, 1.2kΩ, 600Ω and 300 Ω
180
resistors. The 300Ω resistor was chosen as the closest to the internal resistance for MPP,
181
which was derived from the polarisation experiment. The catholyte was collected after 72h of
182
steady state MFC operation under external load conditions. The dry mass of collected
183
precipitant was estimated on the basis of evaporation of collected catholyte samples of equal
184
volumes. The maximum salt concentration calculated for each cathode (see Figure 3) was
185
shown to be: CV 40.1 g/L, AC 57.2 g/L, MPL 93.8g/L, CV MPL 86.2 g/L and shows
186
correlation with power performance.
187
Figure 2. MFC with droplets of water forming on the cathode and collected liquid in the
syringe(left), cathode chamber dissassembled with salts percipitation(centre),catholyte liquid
collected (right).
Salt concentration[g/L]
60
100
40
0
1200
1200
2400
CV MPL
AC
Power[W]
Salt concentration[g/L]
100
60
100
1200
Resistance value[]
2400
Power[W]
Salt concentration[g/L]
300
100
80
200
60
100
40
0
300 600
1200
Salt concentration[g/L]
80
200
300 600
300 600
Resistance value[]
Salt concentration[g/L]
Power[W]
40
2400
40
191
60
100
Resistance value[]
300
0
80
200
0
300 600
100
Salt concentration[g/L]
80
200
Power[W]
Salt concentration[g/L]
300
100
Salt concentration[g/L]
Power[W]
CV
Power[W]
Power[W]
MPL
300
Power[W]
188
189
190
2400
Resistance value[]
192
193
Figure 3. Salt concentration in produced catholyte calculated by dry mass of salts evaporated
from the catholyte formed in all 4 experimental MFC groups under various load conditions.
194
3.3. Relationship between catholyte accumulation and power generation of MFCs
195
Voltage and produced catholyte “water” were monitored during a 72h period at each resistor
196
value. In terms of cathode material, the highest power was recorded for AC 262μW, MPL
197
159μW, MPL CV 152μW at 300Ω and CV 88μW at 600Ω (Figure 4). Catholyte
198
accumulation was recorded after 72h under each resistance value. The data show that water
199
was produced from all cathode materials, at different levels, and was dictated by the level of
200
power as well as by the cathode material. The maximum current was produced from AC
201
934μA, MPL 728μA, CV MPL 712μA and CV 540μA. In terms of power generation, the best
202
performing cathode was the activated carbon (AC) with a maximum power density of
203
262mW/m2 (normalised to the total macro cathode surface area). The relationship between
204
power generation and the catholyte accumulation is shown in Figure 4, as a function of
205
external resistance.
CV
Power[W]
Catholyte produced[mL]
300
2.5
Power[W]
1.0
100
0.5
0
300 600
1200
2400
2.0
200
1.5
1.0
100
0.5
0.0
0
300 600
Resistance value[ ]
AC
Power[W]
Catholyte produced[mL]
2.5
Power[W]
1.5
1.0
100
0.5
1200
Resistance value[ ]
0.0
2400
0.0
Power[W]
2.5
2.0
200
1.5
1.0
100
0.5
0
300 600
1200
2400
Water produced[mL]
200
300 600
2400
Catholyte produced[mL]
300
Water produced[mL]
2.0
0
1200
Resistance value[ ]
Power[W]
CV MPL
300
2.5
Water produced[mL]
1.5
Water produced[mL]
2.0
200
Power[W]
Catholyte produced[mL]
300
Power[W]
MPL
0.0
Resistance value[ ]
206
207
Figure 4. Power average over 72h period in relation to amount of catholyte collected after
208
72h steady state response under various external resistances: 2.4kΩ, 1.2kΩ, 600Ω and 300Ω.
209
These findings seem to be similar to the previously reported water loss through the
210
membrane, which was also dependent on the value of the external load resistance, suggesting
211
that the major water transport phenomena is related to electro-osmotic drag of water through
212
the membrane [12, 21]. However, this has only been reported as a loss of anolyte volume and
213
furthermore this is the first time that newly synthesised catholyte of such volumes, is actually
214
collected from MFCs. This has also been reported for chemical PEM fuel cells [22], which
215
traditionally employ water management techniques to avoid flooding and improve
216
performance. This data shows that water formation actually improves power generation
217
which has an opposite effect to the conventional notion of cathode flooding and showing
218
cathode half cell architecture as flowing system.
219
220
3.4. Analysis of the accumulated catholyte
221
3.4.1. Conductivity and pH
CV
pH
100
Conductivity[mS/cm 2]
pH
40
11
20
10
300 600
1200
Conductivity[mS/cm 2]
60
11
40
10
300 600
AC
pH
2
Conductivity[mS/cm ]
100
pH
60
40
11
20
300 600
1200
Resistance value[ ]
2400
0
pH
Conductivity[mS/cm 2]
100
80
12
60
40
11
20
10
300 600
1200
2400
Conductivity[mS/cm 2]
12
13
Conductivity[mS/cm 2]
80
10
2400
pH
CV MPL
1200
Resistance value[ ]
Resistance value[ ]
13
100
80
12
0
2400
pH
Conductivity[mS/cm 2]
60
Conductivity[mS/cm 2]
80
12
13
pH
MPL
13
0
Resistance value[ ]
222
223
Figure 5. Conductivity and pH values of catholyte collected after 72h under various external
224
load conditions: 2.4kΩ, 1.2kΩ, 600Ω and 300Ω.
225
The conductivity and pH behaved in a similar manner, with the highest conductivity and pH
226
recorded for the highest power levels and water volumes. The pH of the collected samples in
227
the range of 10.6-12.7 suggests high caustic content; Figure 5 shows the measured pH and
228
conductivity values of the collected catholyte from MFCs under the different resistance loads.
229
The data suggest that improved power generation and subsequent higher ion exchange rate
230
between the anode and the cathode drives the OH- accumulation on the cathode surface,
231
therefore the more power produced by the MFC, the more caustic catholyte is collected. The
232
higher reaction rates at MPT (300Ω) would suggest that CO2 buffering (reacting with OH-)
233
would be suppressed compared with the lower power generation (i.e. slower reaction rates)
234
recorded under 2.4kΩ or 1.2kΩ. As previously reported, in the aqueous cathode half cells,
235
high pH of the catholyte is generating a large membrane pH gradient between the anolyte and
236
the catholyte, causing a significant potential loss [23], which could be compensated by the
237
increase in cathodic conductivity [24]. In the present study the excess of catholyte is
238
constantly removed from the cathode surface by gravity (Fig 2) preventing cathode flooding,
239
salt precipitation and high pH gradient build-up, therefore not affecting the MFC
240
performance.
241
3.4.2 EDX and SEM analysis
100
80
Cathode
surface
%
60
40
Catholyte
precipitation
20
0
242
Na
Si
S
K
Ca
Al
243
244
245
Figure 6. EDX profile representing % of detected elements in precipitated salts from the
electrode surface and the evaporated liquid sample(left) and SEM image of the catholyte
crystals(right).
246
Given that the anolyte contained 20mM sodium acetate, it would be expected to find sodium
247
in the precipitated salts. The precipitated salts were crystalline, transparent or white and
248
soluble in water (Fig 2 and Fig 6). The EDX analysis of the salts that precipitated on the
249
surface of the electrode and from the evaporated catholyte solution, show 96.7% and 93.6%
250
sodium content, respectively. This suggests that cations such as sodium (from sludge and
251
sodium acetate feedstock) are being recovered from the anode to the cathode.
252
3.4.3 GC MS and ICP OES analysis
253
The GC-MS analysis detected no organic content in the collected liquid. With the exception
254
of Na, which was off-scale for all the samples, and K which was off-scale for AC, the ICP
255
OES profile is shown in Figure 7. The analysis showed varying amounts of potassium,
256
magnesium and calcium primarily, depending on cathode material, and very small amounts of
257
aluminium and zinc. Although the data were not directly proportional with the level of power
258
produced by each material, the highest peaks were recorded from the most powerful AC
259
cathode.
260
261
Figure 7. Detected trace elements in catholyte samples (mg/L) by ICP OES. Sodium was off
262
the measurable scale in all tested samples (*), as well as Potassium in sample AC (*).
263
3.4.4 XRD analysis
264
X-ray diffraction pattern analysis of all crystallised catholyte samples, appear to be made up
265
from very similar materials. Analysis using single crystal X-ray diffraction suggests that the
266
primary component of the material has a monoclinic cell of a = 21.0, b = 3.5, c =10.7 β =
267
106.41. A database search of materials with a known unit cell of this value with a Na content,
268
suggests a mineral phase known as Trona or Na3H(CO3)2.H2O.
269
Recent developments in the field of BES research have shifted from power production to
270
chemical synthesis, where the energy is stored as chemicals in the cathode and valuable
271
compounds are produced. Here, it is suggested that it may be possible to integrate both
272
energy generation and chemical synthesis as it appears that both processes are inherently
273
linked. The recovery of caustic soda from wastewater and acetate solution used as anolyte has
274
been previously reported [25], by using an energy-consuming MEC and a catholyte
275
containing NaCl. Here, as no catholyte solution was used, the electricity produced by the
276
MFCs is directly linked to the synthesis of caustic catholyte in situ. Similar carbonate salt
277
deposits were previously reported on the cathode electrode-membrane assembly, however
278
this accumulation was hindering the overall MFC performance [26]. Similarly, in cases of
279
single chamber membrane-less MFCs, the biocathode, free of platinum, seemed to be limited
280
by carbonate deposits other than pH higher than 10 [27, 28]. Here, it is suspected that the salt
281
accumulation is prevented due to the effective water transport and production, which removes
282
salts away from the membrane by hydrodynamic flow and is washing away potentially
283
accumulating salts off from the cathode surface. Therefore, the catholyte flow in the present
284
MFC may be also beneficial in keeping the cathode pores free of precipitants as well as it
285
could enhance the water management of the membrane.
286
287
3.5. Significance of catholyte accumulation to environmental cleanup
288
In principle, the catholyte production is mainly due to the water transport and thus the volume
289
of catholyte produced should equal the volume of anolyte lost (minus evaporative losses)
290
considering negligible the contribution of water produced through ORR due to low current
291
generation. Figure 5 shows that up to 2mL of catholyte can be produced over 3 days of
292
operation, from a 25mL sludge filled anode, which theoretically suggests that it would take
293
(for example) 16 days to produce 10.6mL of clear catholyte. If by the end of this period the
294
anolyte is completely depleted (which we expect to be the case), this would suggest a 40%
295
recovery rate with a 60% evaporation rate. This is only a prediction based on visual
296
inspection of anolyte evaporative losses, and further work is required to validate this
297
phenomenon. Apart from sodium which was added as sodium acetate in the feedstock, the
298
other detected elements through the ICP OES analysis included potassium, magnesium and
299
calcium (Fig. 7). It is therefore suspected that these cations must have come from the
300
wastewater through the PEM, and in effect this demonstrates a type of recovery. This would
301
be valuable in extracting useful elements from highly polluted/toxic waste streams.
302
303
3.6. Sodium recovery through bioproduction
304
It has already been hypothesised that a 25mL anolyte could produce 10.6mL of catholyte.
305
This would be the equivalent of 426L of catholyte recovered from 1m3 of anolyte. Taking
306
into account the concentration (5-9%) and composition (90% sodium) of salts in the obtained
307
catholyte, it may be suggested that in theory, 45-81g of sodium per L of catholyte can be
308
recovered, which would be the equivalent of 19.1-34.5kg of salts from 1m3 of anolyte, at
309
larger scale. The analysis of the salt deposits suggests the mineral trona
310
Na3(CO3)(HCO3)•2H2O or Na2CO3-NaHCO3:2H2O which is one of the natural forms of
311
sodium carbonate and it is a double salt of sodium carbonate and sodium bicarbonate. Its
312
potential could be quite high, due to the abundance and low price of sodium-precursors that
313
are normally used in the production of battery components [29].
314
315
3.7. Control of catholyte alkalinity
316
The typical pH of the collected catholyte throughout the experiment was 10.6-12.7. During
317
the experiment the cathode chamber was closed with only 2 open to air vents left at the top of
318
the reactor. As it was observed by obstructing the vents and preventing the ambient air to
319
flow through the chamber, the pH of the catholyte collected subsequently increased to >13.
320
This indicates that by limiting the air flow and consequently the CO2 flux and its buffering
321
properties, it is possible to affect the alkalinity of the catholyte. Moreover, elevating the
322
already high pH of the catholyte simply by limiting the air flow is suggesting the possible use
323
of the synthesised caustic solution as disinfectant and may prevent biofouling of the
324
membrane by limiting the growth of biofilm on the cathode side [30].
325
326
3.8. Carbon capture
327
Removal of gaseous components through contact with a liquid is known as “wet scrubbing”.
328
Carbon dioxide absorption from atmospheric air using alkaline solution has been explored in
329
the 1940s [31] and used at large/industrial scales [32]. In the air extraction process, alkaline
330
sodium solvent reacts chemically with the entrapped CO2. Therefore, it is suspected that in
331
the present MFC cathode, an air flow (and consequently its CO2 content) played an important
332
role in controlling the alkalinity of the caustic catholyte formed in the porous open to air
333
cathode, and the further formation of sodium salts such as: trona, carbonate and bicarbonate
334
of sodium. Furthermore, it has already been proposed that an addition of CO2 is an important
335
pre-condition for the formation of trona deposits [33]. Sodium carbonate is commonly
336
employed in the (i) glass and ceramic, (ii) petroleum, (iii) aluminium, (iv) paper, (v) soap,
337
(vi) detergent and (vii) caustic soda industries. With the proposed MFC set up, caustic
338
solution (and its carbonate content) may be directly synthesised on the MFC cathode to
339
capture atmospheric CO2 and allow sodium recovery from the anolyte. This would allow the
340
MFC to be a truly carbon negative technology.
341
342
Conclusions
343
It has been shown for the first time that Microbial Fuel Cells have the ability to synthesise
344
and extract useful elements on different catalyst-free carbon-based cathodes with the
345
important advantage of electricity generation during this process. The level of power
346
performance depends on the cathode material used however the in situ synthesis of catholyte
347
appears to be independent of the electrode material used and it shows a significant correlation
348
with its electrical performance. This is the first time that carbon-capture via chemical
349
synthesis of liquid catholyte containing valuable minerals, actually improved (and not
350
hindered) the power generation of MFCs, which strongly demonstrates the benevolent
351
potential of the MFC technology.
352
Acknowledgement
353
The work was funded by the Engineering and Physical Sciences Research Council EPSRC
354
CAF EP-I004653/1 and EP/L002132/1. Authors would like to thank Dr David Patton, Mr
355
Paul Bowdler from the University of the West of England and Dr Christopher H. Woodall
356
from the University of Bristol for the valuable expertise and analysis of the catholyte
357
samples.
358
References
359
360
[1] R.M. Allen, H.P. Bennetto, Microbial fuel-cells:Electricity production from
carbohydrates, Appl. Biochem. Biotechnol. 39/40 (1993) 27-40.
361
[2] K. Kinoshita, Carbon: electrochemical and physicochemical properties, (1988).
362
363
364
[3] Park, D.H. and Zeikus, J.G. (2003) Improved fuel cell and electrode designs for producing
electricity from microbial degradation. Biotechnology and Bioengineering [online]. 81 (3),
pp.348-355.
365
366
[4] R. Holze, W. Vielstich, The Kinetics of Oxygen Reduction at Porous Teflon‐Bonded Fuel
Cell Electrodes, J. Electrochem. Soc. 131 (1984) 2298-2303.
367
368
[5] H. Rismani-Yazdi, S.M. Carver, A.D. Christy, O.H. Tuovinen, Cathodic limitations in
microbial fuel cells: An overview, J. Power Sources. 180 (2008) 683-694.
369
370
371
[6] C. Santoro, A. Stadlhofer, V. Hacker, G. Squadrito, U. Schröder, B. Li. Activated Carbon
Nanofibers (ACNF) as cathode for Single Chamber Microbial fuel cells (SCMFCs). Journal
of Power Sources, 243 (2013) 499-507
372
373
[7] S. Freguia, K. Rabaey, Z. Yuan, J. Keller, Non-catalyzed cathodic oxygen reduction at
graphite granules in microbial fuel cells, Electrochim. Acta. 53 (2007) 598-603.
374
375
376
[8] C. Santoro, A. Agrios, U. Pasaogullari, B. Li, Effects of gas diffusion layer (GDL) and
micro porous layer (MPL) on cathode performance in microbial fuel cells (MFCs), Int J
Hydrogen Energy. 36 (2011) 13096-13104.
377
378
379
[9] H. Dong, H. Yu, X. Wang, Q. Zhou, J. Feng, J. A novel structure of scalable air-cathode
without Nafion and Pt by rolling activated carbon and PTFE as catalyst layer in microbial
fuel cells. Water Res. 46 (2012) 5777-5787.
380
381
[10] L. Yu, W. Chen, M. Qin, G. Ren, Experimental research on water management in proton
exchange membrane fuel cells, J. Power Sources. 189 (2009) 882-887.
382
383
[11] D. Natarajan, T. Van Nguyen, Three-dimensional effects of liquid water flooding in the
cathode of a PEM fuel cell, J. Power Sources. 115 (2003) 66-80.
384
385
386
[12] J.R. Kim, G.C. Premier, F.R. Hawkes, R.M. Dinsdale, A.J. Guwy, Development of a
tubular microbial fuel cell (MFC) employing a membrane electrode assembly cathode, J.
Power Sources. 187 (2009) 393-399.
387
388
389
[13] F. Zhang, K.S. Jacobson, P. Torres, Z. He, Effects of anolyte recirculation rates and
catholytes on electricity generation in a litre-scale upflow microbial fuel cell, Energy &
Environmental Science. 3 (2010) 1347-1352.
390
391
[14] R.A. Rozendal, H.V. Hamelers, C.J. Buisman, Effects of membrane cation transport on
pH and microbial fuel cell performance, Environ. Sci. Technol. 40 (2006) 5206-5211.
392
393
[15] C. Forrestal, P. Xu, Z. Ren, Sustainable Desalination using a microbial capacitive
desalination cell, Energy and Environmental Science 5, 2012, 7161-7167.
394
395
[16] Wilkinson, S. (2000) “Gastrobots”—Benefits and Challenges of Microbial Fuel Cells in
FoodPowered Robot Applications. Autonomous Robots 9 (2), pp.99-111.
396
397
398
[17] F. Zhang, G. Chen, M.A. Hickner, B.E. Logan, Novel anti-flooding poly
(dimethylsiloxane)(PDMS) catalyst binder for microbial fuel cell cathodes, J. Power Sources.
(2012).
399
400
[18] I. Gajda, J. Greenman, C. Melhuish, I. Ieropoulos, Photosynthetic cathodes for Microbial
Fuel Cells, Int J Hydrogen Energy. (2013).
401
402
403
[19] C. Santoro, Y. Lei, B. Li, P. Cristiani, Power generation from wastewater using single
chamber microbial fuel cells (MFCs) with platinum-free cathodes and pre-colonized anodes,
Biochem. Eng. J. 62 (2012) 8-16.
404
405
406
407
[20] C. Santoro, K. Artyushkova, S. Babanova, I. Ieropoulos, M. Grattieri, P. Cristiani, S.
Trasatti, B. Li, P. Atanassov. Parameters Characterization and Optimization of Activated
Carbon (AC) Cathodes for Microbial Fuel Cell Application. Bioresource Technology, Under
review
408
409
[21] L. Zhuang, S. Zhou, Y. Wang, C. Liu, S. Geng, Membrane-less cloth cathode assembly
(CCA) for scalable microbial fuel cells, Biosensors and Bioelectronics. 24 (2009) 3652-3656.
410
411
412
[22] Y.H. Park, J.A. Caton, An experimental investigation of electro-osmotic drag
coefficients in a polymer electrolyte membrane fuel cell, Int J Hydrogen Energy. 33 (2008)
7513-7520.
413
414
415
[23] R.A. Rozendal, H.V. Hamelers, K. Rabaey, J. Keller, C.J. Buisman, Towards practical
implementation of bioelectrochemical wastewater treatment, Trends Biotechnol. 26 (2008)
450-459.
416
417
418
[24] R. Rozendal, T. Sleutels, H. Hamelers, C. Buisman, Effect of the type of ion exchange
membrane on performance, ion transport, and pH in biocatalyzed electrolysis of wastewater,
Water Science and Technology. 57 (2008) 1757-1762.
419
420
421
[25] K. Rabaey, S. Butzer, S. Brown, J. Keller, R.A. Rozendal, High current generation
coupled to caustic production using a lamellar bioelectrochemical system, Environ. Sci.
Technol. 44 (2010) 4315-4321.
422
423
424
[26]T.H Pham, J.K. Jang, H.S. Moon, I.S. Chang, B.H. Kim, Improved performance of
microbial fuel cell using membrane-electrode assembly, Journal of microbiology and
biotechnology. 15 (2005) 438-441.
425
426
[27] P. Cristiani, M. Carvalho, E. Guerrini, M. Daghio, C. Santoro, B. Li, Cathodic and
anodic biofilms in single chamber microbial fuel cells, Bioelectrochemistry. 92 (2013) 6-13.
427
428
429
[28] C. Santoro, M. Cremins, U. Pasaogullari, M. Guilizzoni, A. Casalegno, A. Mackay, B.
Li, Evaluation of Water Transport and Oxygen Presence in Single Chamber Microbial Fuel
Cells with Carbon-Based Cathodes, J. Electrochem. Soc. 160 (2013) G3128-G3134.
430
431
[29] M.D. Slater, D. Kim, E. Lee, C.S. Johnson, Sodium‐Ion Batteries, Advanced Functional
Materials. (2012).
432
433
[30] J. Xu, G. Sheng, H. Luo, W. Li, L. Wang, H. Yu, Fouling of proton exchange membrane
(PEM) deteriorates the performance of microbial fuel cell, Water Res. 46 (2012) 1817-1824.
434
435
[31] N.A. Spector, B.F. Dodge, Removal of carbon dioxide from atmospheric air, Trans. Am.
Inst. Chem. Eng. 42 (1946) 827-848.
436
437
[32] K.S. Lackner, P. Grimes, H. Ziock, Capturing carbon dioxide from air, (2001) In: 24th
Annual Technical Conference on Coal Utilization, Clearwater, FL.
438
[33] S. Earman, F.M. Phillips, B.J. McPherson, The role of “excess” CO2 in the formation of
439
trona deposits, Appl. Geochem. 20 (2005) 2217-2232.