International Journal of
Environmental Research
and Public Health
Article
Concurrent Phosphorus Recovery and Energy
Generation in Mediator-Less Dual Chamber
Microbial Fuel Cells: Mechanisms and
Influencing Factors
Abdullah Almatouq 1,2, * and Akintunde O. Babatunde 1
1
2
*
Hydro-Environment Research Centre, Energy and Environment Theme,
Cardiff University School of Engineering, Queen’s Buildings, The Parade, Cardiff CF24 3AA, UK;
[email protected]
Kuwait Institute of Scientific Research, P.O. Box 24885, Safat 13109, Kuwait
Correspondence: [email protected]; Tel.: +44-29-2087-0076; Fax: +44-29-2087-4939
Academic Editors: Rao Bhamidiammarri and Kiran Tota-Maharaj
Received: 29 December 2015; Accepted: 23 March 2016; Published: 28 March 2016
Abstract: This study investigated the mechanism and key factors influencing concurrent phosphorus
(P) recovery and energy generation in microbial fuel cells (MFC) during wastewater treatment.
Using a mediator-less dual chamber microbial fuel cell operated for 120 days; P was shown to
precipitate as struvite when ammonium and magnesium chloride solutions were added to the
cathode chamber. Monitoring data for chemical oxygen demand (COD), pH, oxidation reduction
potential (ORP) and aeration flow rate showed that a maximum 38% P recovery was achieved; and
this corresponds to 1.5 g/L, pH > 8, ´550 ˘ 10 mV and 50 mL/min respectively, for COD, pHcathode ,
ORP and cathode aeration flow rate. More importantly, COD and aeration flow rate were shown to
be the key influencing factors for the P recovery and energy generation. Results further show that the
maximum P recovery corresponds to 72 mW/m2 power density. However, the energy generated at
maximum P recovery was not the optimum; this shows that whilst P recovery and energy generation
can be concurrently achieved in a microbial fuel cell, neither can be at the optimal value.
Keywords: bio-electrochemical system; phosphorus; phosphorus recovery; microbial fuel cell; struvite
1. Introduction
Microbial fuel cells (MFCs) are systems that convert chemical energy in an organic substrate in
wastewater into electrical energy. Phosphorus (P) is one of the key pollutants that can be removed
during wastewater treatment in MFC with the concurrent generation of energy. P is essential for crop
growth, and approximately 60% of P used around the world comes from non-renewable sources [1].
Therefore, there is increasing interest in finding new sustainable sources of P. Wastewater is a rich
source of nutrients that can be used as a sustainable source of P. Magnesium ammonium phosphate
hexahydrate (MAP or struvite) is an efficient slow release fertilizer [2]. The mechanism of struvite
precipitation is highly dependent on solution pH (pH > 8). The precipitation occurs in equimolecular
concentration of magnesium (Mg), ammonium (NH4 ) and P; and these combine with water to form
struvite [3]. P can be recovered as struvite from different waste streams including reject wastewater
and digester effluent; and it can be achieved using several methods, including: chemical addition,
carbon dioxide stripping or electrolysis [4–6]. However, the chemical addition is a costly process and
the chemicals used to raise the pH can account for up to 97% of struvite cost [6,7].
Recent research findings on MFCs have demonstrated their potential to recover P as struvite [7–9].
P can be recovered exclusively through precipitation in MFC because P compounds are not involved in
Int. J. Environ. Res. Public Health 2016, 13, 375; doi:10.3390/ijerph13040375
www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2016, 13, 375
2 of 12
RecentRes.
research
findings
on375
MFCs have demonstrated their potential to recover P as struvite
Int. J. Environ.
Public Health
2016, 13,
2 of [7–
12
9]. P can be recovered exclusively through precipitation in MFC because P compounds are not
involved in electron transfer via reduction—oxidation (REDOX) reactions [10]. P recovery as struvite
electron
transfer
via reduction—oxidation
[10]. P recovery
as struvite using
MFCs
using MFCs
involves
cathode reactions, (REDOX)
whereby reactions
water is consumed
and hydroxide
is generated
involves
cathode
reactions,
whereby
water
is
consumed
and
hydroxide
is
generated
due
to
electrons
due to electrons transferred from the anode to the cathode (Equations (1) and (2)):
transferred from the anode to the cathode (Equations (1) and (2)):
(1)
Anode chamber: CH3 COO- + 4 H2 O → 2 HC + 9 H+ + 8 e9H` ` 8e´
Anodechamber : CH3 COO´ ` 4H2 O Ñ 2HCO´
3 `
Cathode chamber: 2H O O
4e →
4 OH
(1)
(2)
´
Cathodechamber
: 2H2 Oin
`O
4e´around
Ñ 4OHthe
The generated hydroxide
leads to increase
the
cathode, and P starts(2)
to
2 `pH
precipitate. P recovery from different wastewater sources has been widely studied using single
The generated hydroxide leads to increase in the pH around the cathode, and P starts to precipitate.
chamber MFCs [5,8,9]. Ichihashi and Hirooka [9] reported 27% P recovery as struvite from swine
P recovery from different wastewater sources has been widely studied using single chamber MFCs [5,8,9].
wastewater. It was noted that precipitates formed around the cathode, which indicates that the pH
Ichihashi and Hirooka [9] reported 27% P recovery as struvite from swine wastewater. It was noted that
around the cathode is higher than at any other place in the MFC. In addition, synthetic wastewater
precipitates formed around the cathode, which indicates that the pH around the cathode is higher than
was used in a single chamber MFC to identify the effect of Mg and NH4 concentration on P recovery
at any other place in the MFC. In addition, synthetic wastewater was used in a single chamber MFC to
as struvite [8]. Dosing ammonium and magnesium for struvite precipitation is essential to form
identify the effect of Mg and NH4 concentration on P recovery as struvite [8]. Dosing ammonium and
struvite; as the concentrations of NH4 and Mg increases, more P is precipitated. In another study, a
magnesium for struvite precipitation is essential to form struvite; as the concentrations of NH4 and
3-stage single chamber MFC was used to recover 78% P as struvite from human urine [4]. Using
Mg increases, more P is precipitated. In another study, a 3-stage single chamber MFC was used to
urine as a substrate increased the recovered P because urea hydrolysis increased the pH and created
recover 78% P as struvite from human urine [4]. Using urine as a substrate increased the recovered P
an alkaline environment which accelerated the precipitation. From these studies, it has been shown
because urea hydrolysis increased the pH and created an alkaline environment which accelerated the
that MFC has the potential to recover P as struvite from different wastewater sources. However,
precipitation. From these studies, it has been shown that MFC has the potential to recover P as struvite
most of these studies were conducted in single chamber MFCs, where pH buffering may occur due
from different wastewater sources. However, most of these studies were conducted in single chamber
to the accumulation of protons and hydroxide ions in the same electrolyte; this could limit the P
MFCs, where pH buffering may occur due to the accumulation of protons and hydroxide ions in the
recovery. To overcome this limitation, dual chamber MFCs are increasingly being used. They have
same electrolyte; this could limit the P recovery. To overcome this limitation, dual chamber MFCs are
better separation between the anode and cathode chambers; this leads to pH splitting which creates
increasingly being used. They have better separation between the anode and cathode chambers; this
an alkaline environment around the cathode to improve the P recovery process [10]. Such a 2-stage
leads to pH splitting which creates an alkaline environment around the cathode to improve the P
dual chamber MFC was operated in this study to recover P in the form of struvite. The specific
recovery process [10]. Such a 2-stage dual chamber MFC was operated in this study to recover P in
objectives of the study were: (i) to understand the mechanisms of P recovery as struvite in
the form of struvite. The specific objectives of the study were: (i) to understand the mechanisms of P
dual-chamber MFC; and (ii) to identify the influencing factors for the P recovery.
recovery as struvite in dual-chamber MFC; and (ii) to identify the influencing factors for the P recovery.
2.2. Materials
Materials and
and Methods
Methods
2.1.MFC
MFCConfiguration
Configuration
2.1.
Twosets
setsofofdual-chamber
dual-chamberH-type
H-typebottles
bottles(Adams
(Adams &
& Chittenden
Chittenden Scientific
Scientific Glass,
Glass, Berkeley,
Berkeley, CA,
CA,
Two
USA),
were
used
to
construct
the
MFCs
(see
Figure
1).
USA), were used to construct the MFCs (see Figure 1).
Figure
Figure1.1.Experimental
Experimentalset-up
set-upof
ofthe
thedual-chamber
dual-chamberMFC.
MFC.
Int. J. Environ. Res. Public Health 2016, 13, 375
3 of 12
Anode and cathode electrodes were made of carbon cloth 2.5 ˆ 5 cm (projected area of 25 cm2 )
with a volume of 300 mL for each chamber. The cathode contained a Pt catalyst (0.5 mg/cm2 10% Pt
on Carbon Cloth Electrode) to improve cathode performance, whilst the anode is plain carbon cloth.
Both electrodes were connected with a titanium wire (0.5 mm, purity > 99.98%, Alfa Aesar, Heysham,
UK). A Nafion membrane (Nafion 117#, Sigma Aldrich, London, UK) was placed in the middle of the
anode and the cathode. The membrane was pre-treated by boiling in H2 O2 (30%) and deionized water,
followed by 0.5 M H2 SO4 and deionized water, each for 1 h, and thereafter it was stored in deionized
water prior to being used. The MFCs were maintained at 20 ˝ C in a temperature controlled room.
2.2. MFC Experimental Design and Start-Up
The anode chambers were inoculated with a 1:1 mixture of activated sludge obtained from
a wastewater treatment plant in Cardiff and anolyte medium (mainly containing acetate). The external
resistance was 1000 Ω at the initial stage of the operation, and it was reduced to 430 Ω where the
system produced the maximum power in the polarization test. During the start-up-period, the cathode
chamber was filled with 50 mM phosphorus buffer, and it was continuously aerated using an aquarium
pump. Once the MFC achieved stable electricity production for at least three cycles, the phosphorus
buffer was replaced with anolyte effluent from another batch.
The MFCs were operated in 2-stage feed-batch mode at room temperature. In the 1st stage,
the synthetic wastewater was fed to the anode chamber for COD removal. At the end of each
cycle, the effluent from the anode chamber was filtered and directed into the cathode chamber
(2nd stage) for P recovery (see Figure 1). Anode influent pH was adjusted to pH = 7 and an aquarium
pump was used to supply air continuously to the cathode chamber in order to provide oxygen as
electron acceptor. Media replacement was carried out in a glove box to maintain the anaerobic
environment for the anode biofilms. The synthetic wastewater contained 1 g/L sodium acetate;
KH2 PO4 , 0.65 g/L; K2 HPO4 , 0.65 g/L; KCl, 0.74 g/L; NaCl, 0.58 g/L; NH4 Cl, 0.375 g/L; MgSO4 ¨ 7H2 O,
0.1 g/L; CaCl2 ¨ 2H2 O, 0.1 g/L; 0.1 mL/L of a trace element mixture containing (per litre): iron(II) sulfate
heptahydrate (FeSO4 ¨ 7H2 O), 1 g; zinc chloride (ZnCl2 ), 0.07 g; manganese(II) chloride tetrahydrate
(MnCl2 ¨ 4H2 O), 0.1 g; boric acid (H3 BO3) , 0.006 g; calcium chloride hexahydrate (CaCl2 ¨ 6H2 O), 0.13 g;
copper (II) chloride dihydrate (CuCl2 ¨ 2H2 O), 0.002 g; nickel (II) chloride (NiCl2 ¨ 6H2 O), 0.024 g;
sodium molybdate (Na2 MoO4 ¨ 2H2 O), 0.036 g; cobalt(II) chloride (CoCl2 ¨ 6H2 O), 0.238 g and vitamins.
The synthetic wastewater simulated reject wastewater during conventional wastewater treatment
which usually has high P concentration. After the start-up period, four operational parameters (COD
concentration, anolyte and catholyte volume and cathode dissolved oxygen concentration) were
monitored in order to determine their impacts on energy generation and P recovery.
2.3. Analytical Methods
The voltage across the external resistance was recorded every 15 min using ADC-20 data logger
system (Pico Technology, Saint Neots, UK), which was connected to a computer. Based on the recorded
voltage, current (I) and power (P = IV) were calculated according to Ohm’s law. The current density
and power density were calculated by dividing the current and the power by the anode area (25 cm2 )
Coulombic efficiency (CE) was calculated by integrating the measured current relative to the theoretical
current previously described by Equation (3) [11]. The polarization curves were determined by varying
the external resistance from 10,000 to 10 Ω with an interval of 15 min to obtain a stable voltage reading:
şt
M 0 I dt
CE “
Fbv An ∆COD
(3)
where M = 32, the molecular weight of oxygen, F is Faraday’s constant, b = 4 is the number of electrons
exchanged per mole of oxygen, vAn is the volume of liquid in the anode compartment, and ∆COD
Int. J. Environ. Res. Public Health 2016, 13, 375
4 of 12
is the difference between anode COD influent and anode COD effluent over time tb (cycle duration).
COD removal efficiency was calculated using Equation (4):
COD removal e f f iciency “
anode COD in f luent ´ anode COD e f f luent
ˆ 100
anode COD in f luent
(4)
The chemical oxygen demand (COD), total nitrogen (TN), ammonium (NH4 -N), nitrate (NO3 ),
nitrite (NO2 ) and total phosphorus (TP) were measured using a DR3900 spectrophotometer (HACH,
Salford, UK) according to its operating procedures. Water quality analyses were conducted to assess
the wastewater treatment efficiency of the MFC. pH and oxidation-reduction potential (ORP) were
recorded and monitored in real time using a multi-channel parameter data recorder (EA Instruments,
London, UK). Probes for measuring pH and ORP were connected to the data logger which was
connected to a personal computer.
2.4. Phosphorus Precipitation in MFC
The theoretical P concentration in the synthetic wastewater was approximately 8 mM at pH 7. For P
precipitation, ammonium chloride and magnesium chloride solutions with the theoretical concentration
of 8 mM were pumped into the cathode chamber using a peristaltic pump. The ammonium and
magnesium solution was pumped with a flow rate of 6 mL/day. The flowrate was chosen to achieve
a 1:1:1 molar ratio of struvite precipitation NH4 :Mg:P. Before and after each cycle, the cathode chamber
was washed with deionised water 3 times, and then cleansed and dried properly to remove any
attached precipitates on the chamber walls. After each precipitation cycle, the used cathode was
removed for maintenance and replaced with new ones. Cathode maintenance is essential to remove P
precipitates from cathode surface, as the precipitates reduce cathode performance and dissolution
treatment increases cathode performance to their initial level. The used cathodes were firstly soaked in
deionised water, and thereafter in 10 mM 2-(N-morpholino) ethanesulfonic acid (MES) buffer adjusted
to pH 5.5 with NaOH, and then finally rinsed with deionised water. At the end of each cycle, the
catholyte was filtered using a 0.2 um filter membrane (Fisher Scientific, Pittsburgh, UK). The precipitate
was collected and analysed by X-ray diffraction (XRD). The recovered P was calculated using Equation (5):
Pin ´ Pout “ P filter
precipitates
` P attached
to cathode and cell
(5)
where Pin = P concentration in the cathode influent, Pout = P concentration in the cathode effluent,
and P filter precipitates = the recovered P. Prior to the start of the experiment, checks were conducted to
confirm that P cannot be precipitated without an MFC. A small beaker filled with the same synthetic
wastewater (pH = 7 and 8 mM of P) and the same concentration of ammonium chloride and magnesium
chloride was pumped into the beaker at the same rate (6 mL/day) to achieve a molar ratio of (1:1:1).
No precipitates were observed anywhere in the beaker during the operation time. Thereafter, solution
pH was gradually increased using 1 M NaOH. Precipitation started to occur when the pH reached 7.58.
2.5. Statistical Analyses
Statistical analyses were performed using Pearson coefficient analysis to determine any correlation
between the different operational conditions and the monitored parameters (anode and cathode pH,
anode and cathode ORP). In particular, identifying the relationship between different operational
conditions and cathode pH will lead to improved understanding of maximising P recovery in MFC.
3. Results and Discussion
3.1. Phosphorus Recovery and Electricity Generation in Dual Chamber MFC
After the start-up period, stable electricity production was achieved in the system. At the
beginning of each batch (batch duration = 48 h), electricity production peaked with a current density
Int. J. Environ. Res. Public Health 2016, 13, 375
Int. J. Environ. Res. Public Health 2016, 13, 375
5 of 12
5 of 12
of (324 mA/m2) and decreased with time until the end of each batch. The maximum power density
was achieved at COD = 0.7 g/L, and the system achieved a maximum power density of 198 mW/m2
of (324 mA/m2 ) and decreased with2 time until the end of each batch. The maximum power density
with a current density of 0.49 mA/m .
was achieved at COD = 0.7 g/L, and the system achieved a maximum power density of 198 mW/m2
Due to the degradation of organic matter at the anode, electrons are released and transferred
with a current density of 0.49 mA/m2 .
through the external resistance to the cathode. Current is generated during the transfer of electrons
Due to the degradation of organic matter at the anode, electrons are released and transferred
from anode to cathode. At the cathode, water is consumed and hydroxide is generated; this increases
through the external resistance to the cathode. Current is generated during the transfer of electrons
the pH around the cathode electrode (pH > 8). Precipitates around the cathode were observed after a
from anode to cathode. At the cathode, water is consumed and hydroxide is generated; this increases
short period of adding Mg and NH4 solution to the cathode chamber. Cathode effluent was filtered
the pH around the cathode electrode (pH > 8). Precipitates around the cathode were observed after
at the end of each batch and the precipitations were analysed using XRD (see Figure 2). P was
a short period of adding Mg and NH4 solution to the cathode chamber. Cathode effluent was filtered at
recovered in the dual chamber MFC as magnesium ammonium phosphate hexahydrate (struvite) at
the end of each batch and the precipitations were analysed using XRD (see Figure 2). P was recovered
the cathode chamber, by dosing 8 mM of NH4Cl and MgCl2 (0.76 g/L and 0.42 g/L, respectively) to
in the dual chamber MFC as magnesium ammonium phosphate hexahydrate (struvite) at the cathode
achieve a molar ratio (Mg: NH4: P: 1:1:1).
chamber, by dosing 8 mM of NH4 Cl and MgCl2 (0.76 g/L and 0.42 g/L, respectively) to achieve
a molar ratio (Mg: NH4 : P: 1:1:1).
Figure 2. XRD
XRDpatterns
patternsofofprecipitates
precipitatesin in
cathode
chamber,
precipitated
sample;
thethe
cathode
chamber,
(a) (a)
precipitated
sample;
andand
(b)
(b)
magnesium
ammonium
phosphate
standard.
magnesium
ammonium
phosphate
standard.
It was
greater
thethe
precipitates
at the
the less
generated
by the
It
was observed
observedthat
thatthe
the
greater
precipitates
at cathode,
the cathode,
theenergy
less energy
generated
MFC.
This
suggests
that
precipitates
at
the
cathode
obstruct
the
mass
transfer
of
ions
and
oxygen
[8].
by the MFC. This suggests that precipitates at the cathode obstruct the mass transfer of ions and
Current[8].
density
wasdensity
negatively
by Paffected
precipitates
at the cathode
where
low current
oxygen
Current
wasaffected
negatively
by P precipitates
at chamber,
the cathode
chamber,
where
2) was observed
density
(62
mA/m
with
high
P
precipitates
(Figure
3).
During
phase
1
(COD
= 0.71
2
low current density (62 mA/m ) was observed with high P precipitates (Figure 3). During
phase
2
g/L), the
maximum
density
achieved
wasachieved
370 mA/m
pH was
7.4. Small
(COD
= 0.7
g/L), thecurrent
maximum
current
density
wasand
370average
mA/m2cathode
and average
cathode
pH
amount
of
struvite
was
precipitated
in
the
cathode
chamber
(2%–8%
of
P).
In
phase
2
(COD
=
1
g/L),
was 7.4. Small amount of struvite was precipitated in the cathode chamber (2%–8% of P). In phase 2
the current
density
decreased
significantly
assignificantly
compared to
1, where
average
cathode
pH
(COD
= 1 g/L),
the current
density
decreased
as phase
compared
to phase
1, where
average
increasedpH
to increased
7.8 and struvite
precipitate
from
8% to 26%.
cathode
to 7.8 and
struviteincreased
precipitate
increased
from 8% to 26%.
In
phase
3
(COD
=
1.5
g/L),
the
decrease
in
current
density
was
significantly
more
as as
a result
of
In phase 3 (COD = 1.5 g/L), the decrease in current
density
was
significantly
more
a result
the
increased
amount
of
precipitates
on
the
cathode
electrode
(Figure
4).
Average
cathode
pH
of the increased amount of precipitates on the cathode electrode (Figure 4). Average cathode pH
increased and
and peaked
peaked at
at 8.3.
8.3. The
phase 3
3 with
with aa maximum
maximum
increased
The maximum
maximum P
P recovery
recovery was
was achieved
achieved in
in phase
recovery
of
38%.
These
results
implies
that
the
greater
the
precipitates
at
the
cathode,
the
less
recovery of 38%. These results implies that the greater the precipitates at the cathode, the less energy
energy
generated by
by the
the MFC.
MFC.
generated
Int. J. Environ. Res. Public Health 2016, 13, 375
Int.
J. Environ.
Res.
Public
Health
2016,
375
Int.
J. Environ.
Res.
Public
Health
2016,
13,13,
375
6 of 12
6 of
6 of
1212
P recovery=
2%-8%
P recovery=
2%-8%
P recovery=
8%-26%
P recovery=
8%-26%
Phase
Phase
1 1
Phase
Phase
2 2
P recovery=
20%-38%
P recovery=
20%-38%
400
400
Phase
Phase
3 3
Current density (mA/m22)
Current density (mA/m )
350
350
300
300
250
250
200
200
150
150
100
100
5050
00
00
2020
4040
6060
8080
100
100
Time
(days)
Time
(days)
Figure 3.
3. Impact
Impact of
of P
P recovery
recovery on
on current
current density.
Figure
density.
Figure
3. Impact
of P recovery
on current
density.
Figure
4.
Evidence
phosphorus
precipitation
on
the
cathode.
Figure
4. 4.
Evidence
ofof
phosphorus
precipitation
onon
the
cathode.
Figure
Evidence
of
phosphorus
precipitation
the
cathode.
3.2. COD
Removal
and
Coulombic Efficiency
3.2.
Removal
and
Coulombic
3.2.COD
COD
Removal
and
CoulombicEfficiency
Efficiency
The system
was operated
at three
different COD
concentrations (0.7,
1.0 and
1.5 g/L)
simulate
The
toto
simulate
Thesystem
systemwas
wasoperated
operatedatatthree
threedifferent
differentCOD
CODconcentrations
concentrations(0.7,
(0.7,1.0
1.0and
and1.5
1.5g/L)
g/L)
to
simulate
COD
concentrations
of
reject
wastewater.
The
anode
COD
removal
efficiency
ranged
from
70%toto
COD
concentrations
of
reject
wastewater.
The
anode
COD
removal
efficiency
ranged
from
70%
COD concentrations of reject wastewater. The anode COD removal efficiency ranged from 70% to 90%.
90%.AtAtlow
low CODconcentration,
concentration,the
theaverage
average removalefficiency
efficiencywas
was90%
90% andatathigh
highCOD
COD
90%.
At low
COD COD
concentration, the average
removalremoval
efficiency was 90% and
at highand
COD concentration,
concentration,the
theaverage
averageremoval
removalefficiency
efficiencywas
was70%.
70%.Increasing
Increasinginfluent
influentCOD
CODconcentration
concentration
concentration,
the average removal efficiency was 70%. Increasing influent COD concentration decreased COD
decreased
COD
removal
efficiency.
The
average
coulombic
efficiency
was
10%
and
decreased with
decreased
COD
removal
The
average
coulombic
efficiency
10%
and
it it
decreased
removal efficiency.
The efficiency.
average coulombic
efficiency
was 10%
and itwas
decreased
with
increasingwith
COD
increasingCOD
CODconcentration.
concentration. Similarfindings
findings werereported
reportedbybySleutels
Sleutels etal.al.[12],
[12], whereit itwas
was
increasing
concentration.
Similar findingsSimilar
were reported bywere
Sleutels et al. [12],
where it et
was noted where
that increasing
notedthat
thatincreasing
increasingsubstrate
substrateconcentration
concentrationleads
leadstotodecreases
decreasesininthe
thecoulombic
coulombicefficiency.
efficiency.
noted
substrate concentration leads to decreases in the coulombic efficiency. Furthermore, increasing cathode
Furthermore,increasing
increasingcathode
cathode pHdue
due tohydroxide
hydroxide generationleads
leadstotoa adecrease
decreaseininelectricity
electricity
Furthermore,
pH due to hydroxide
generationpH
leads toto
a decrease in generation
electricity generation,
and this then
decreases
generation,
and
this
then
decreases the
coulombic
efficiency
[13]. This
result
demonstrates
that
COD
generation,
and
this
then
decreases
coulombic
efficiency
result
demonstrates
that
COD
the coulombic
efficiency
[13]. Thisthe
result
demonstrates
that [13].
CODThis
concentration
is an important
factor
concentration
is
an
important
factor
for
obtaining
high
coulombic
efficiency
and
energy
generation
concentration
an important
for obtaining
highgeneration
coulombicin
efficiency
andcoulombic
energy generation
for obtainingishigh
coulombicfactor
efficiency
and energy
MFCs. The
efficiency
in
MFCs.
The
coulombic
efficiency
can
be
improved
by
using
a better
configuration
dual
chamber
incan
MFCs.
The
coulombic
efficiency
can
be
improved
by
using
a
better
configuration
ofof
dual
chamber
be improved by using a better configuration of dual chamber MFC. The type used
in the
current
MFC.
The
type
used
in
the
current
study
is
the
H-type
which
is
known
to
be
a low-efficiency
system
MFC.
The
type
used
in
the
current
study
is
the
H-type
which
is
known
to
be
a
low-efficiency
system
study is the H-type which is known to be a low-efficiency system due to the limited surface area for
due
to
the
limited
surface
area
for
ion
exchange
membranes,
and
the
long
distance
between
due
the limited
surface and
areathe
forlong
ion distance
exchange
membranes,
and [14].
the In
long
distance
ion to
exchange
membranes,
between
electrodes
addition,
thebetween
diffusion
electrodes[14].
[14].InInaddition,
addition,the
thediffusion
diffusionofofoxygen
oxygenfrom
fromthe
thecathode
cathodeand
andfrom
fromthe
thepH
pHand
andORP
ORP
electrodes
probesport
portdecrease
decreasethe
thecoulombic
coulombicefficiency
efficiencyshowing
showingthat
thatthe
theoxygen
oxygencould
couldwork
workasasananelectron
electron
probes
acceptor
the
anode
chamber.
acceptor
inin
the
anode
chamber.
Int. J. Environ. Res. Public Health 2016, 13, 375
7 of 12
of oxygen from the cathode and from the pH and ORP probes port decrease the coulombic efficiency
showing that the oxygen could work as an electron acceptor in the anode chamber.
Int. J. Environ. Res. Public Health 2016, 13, 375
3.3. Effect
of Different Operational Parameters
7 of 12
3.3. Effect
of Different
The
performance
ofOperational
MFCs canParameters
be influenced by different operational parameters such as: COD
concentration,
chamber
volume
and
catholyte
aeration
flowoperational
rate. It isparameters
thereforesuch
veryas:important
The performance of MFCs can be
influenced
by different
COD
to understand
thechamber
impact volume
of these
on MFC
Thevery
influence
oftoCOD
concentration,
andparameters
catholyte aeration
flowperformance.
rate. It is therefore
important
understand
the impact
of these
parameters
MFC performance.
The
influence
of COD
concentration
which
acts as fuel
for the
bacteria,on
electrolyte
volume and
cathode
air flow
rate on
concentration
which
acts
as
fuel
for
the
bacteria,
electrolyte
volume
and
cathode
air
flow
rate
on
MFCs electricity generation have been studied [13,15–17]. In general, the studies show that increasing
MFCs
electricity
generation
have
been
studied
[13,15–17].
In
general,
the
studies
show
that
COD concentration and cathode aeration leads to increased energy generation; however, the influence
increasing
COD on
concentration
aeration
leads to increased energy generation; however,
of these
parameters
P recoveryand
hascathode
not been
investigated.
the influence of these parameters on P recovery has not been investigated.
3.3.1. P recovery at Different Substrate Concentrations
3.3.1. P recovery at Different Substrate Concentrations
160
35
140
30
120
25
100
20
80
15
60
P recovery %
Current density (mA/m2)
Different COD concentrations were used to identify the impacts of COD concentration on P
Different COD concentrations were used to identify the impacts of COD concentration on P
recovery.
The COD
concentration
at at
the
anode
from0.7
0.7toto1.51.5
g/L
(organic
loading
recovery.
The COD
concentration
the
anodechamber
chamber varied
varied from
g/L
(organic
loading
rate =rate
0.35= to
0.75
g COD/L/day).
It Itwas
the anolyte
anolyteCOD
CODleads
leadstotoincrease
increase in
0.35
to 0.75
g COD/L/day).
wasobserved
observedthat
thatincreasing
increasing the
the recovered
P at thePcathode
(Figure
5). As5).the
concentration
increased
from
0.70.7g/L
in the recovered
at the cathode
(Figure
AsCOD
the COD
concentration
increased
from
g/Lto
to 1.5
1.5 g/L,
the recovered
P increased
from 7%
to 7%
38%.
Similarly,
as influent
COD
concentration
at the
g/L, the recovered
P increased
from
to 38%.
Similarly,
as influent
COD
concentrationincreased
increased at
theanode
anode,
anode oxidation
(Figure
6). This that
implies
that matter
organicdegradation
matter
anode,
oxidation
reactions reactions
increasedincreased
(Figure 6).
This implies
organic
degradation
due
to substratefor
availability
for the microorganism.
As a result
the organic
increased
due toincreased
substrate
availability
the microorganism.
As a result
of theof organic
matter
matter
degradation,
electrons
are
liberated
and
transferred
from
the
anode
to
the
cathode
[11].
degradation, electrons are liberated and transferred from the anode to the cathode [11].
Current
density
(mA/m2)
Recovered P
10
40
5
20
0
0
700
1000
1500
Influent COD (mg/L)
Figure
5. Impact
ofofCOD
densityand
andP Precovery.
recovery.
Figure
5. Impact
CODon
on current
current density
Increasing COD concentration leads to increase in the transfer of electrons from the anode to the
Increasing COD concentration leads to increase in the transfer of electrons from the anode to the
cathode chamber. This implies that more electrons are available at the cathode for oxygen reduction
cathode
chamber. This implies that more electrons are available at the cathode for oxygen reduction
reactions; this, in turn facilitates struvite formation. Increased oxygen reduction reactions at the
reactions;
this,
in to
turn
facilitates
formation.
Increased
reduction
at the
cathode
leads
increases
in thestruvite
generation
of hydroxide
ions; andoxygen
this increased
the pHreactions
to >8 at the
cathode
leads
to
increases
in
the
generation
of
hydroxide
ions;
and
this
increased
the
pH
to
>8
at the
cathode (Figures 6 and 7).
cathode (Figures
6 and
7).cathode pH increased as COD concentration increased (Figure 7). The average
It was noted
that
pH increased
frompH
7.4 increased
at COD = 0.7
g/L toconcentration
8.3 at COD = 1.5
g/L. Energy
generation
Itcathode
was noted
that cathode
as COD
increased
(Figure
7). The was
average
negatively
affected from
by P 7.4
precipitation
theg/L
cathode
COD
concentration
increased, was
cathode
pH increased
at COD =in0.7
to 8.3chamber.
at CODAs
= 1.5
g/L.
Energy generation
more precipitates
cathode were
observed.
Thechamber.
current density
decreased
with increasing
negatively
affected byonP the
precipitation
in the
cathode
As COD
concentration
increased,
COD concentration due to the amount of precipitates on the cathode surface (Figure 6).
more precipitates on the cathode were observed. The current density decreased with increasing COD
concentration due to the amount of precipitates on the cathode surface (Figure 6).
Int.Int.
Environ.
Res.
Public
Health
2016,
13,
375
J. Environ.
Res.
Public
Health2016,
2016,13,
13,375
375
Int.
J.J.Environ.
Res.
Public
Health
of
8 of88 12
of 12
12
Figure6.6.Impact
Impact of
of COD
COD on current
pH.
Figure
current density,
density,anode
anodeORP,
ORP,anode
anodeand
andcathode
cathode
pH.
Figure 6. Impact of COD on current density, anode ORP, anode and cathode pH.
8.5
Average cathode pH
Average cathode pH
8.5
8
8
7.5
7.5
7
7
0.7
0.7
1
Influent COD (g/L)
1
1.5
1.5
Influent
COD (g/L)
Figure 7. Impact of COD
concentrations
on cathode pH.
Figure
7. Impact
Impact
of COD
COD
concentrations
onthe
cathode
pH. and this obstructed the
The precipitates onFigure
the cathode
covered
theconcentrations
surface area of
cathode;
7.
of
on
cathode
pH.
mass transfer of ions and oxygen. This finding is consistent with the finding of Hirooka and
Ichihashi
[8], which on
showed
that thecovered
electricity
generated
byarea
MFCs
with
precipitate
was
lower
thanthe
The
the
surface
area
of the
cathode;
andand
this
obstructed
The precipitates
precipitates
onthe
thecathode
cathode
covered
the
surface
of the
cathode;
this
obstructed
of MFCs
without
precipitate.
mass
transfer
of
ions
and
oxygen.
This
finding
is is
consistent
with
thethat
mass
transfer
of ions
and
oxygen.
This
finding
consistent
withthe
thefinding
findingofof Hirooka
Hirooka and
and
Ichihashi
Ichihashi [8],
[8], which
which showed
showed that
that the
the electricity
electricity generated
generated by
by MFCs
MFCs with
with precipitate
precipitate was
was lower
lower than
than
3.3.2.
P
Recovery
at
Different
Anode
and
Cathode
Volumes
that
that of
of MFCs
MFCs without
without precipitate.
precipitate.
Electrolyte volume is another important parameter for P recovery. Three different volumes
3.3.2.
P
Recovery
at Different
Different
Anode
and
Cathode Volumes
Volumes
3.3.2.
P
at
and
Cathode
(180, Recovery
225,
and 270
mL) wereAnode
used to
investigate
the impact of anode and cathode volumes on P
recovery
and
energy
generation.
The
volumes
were
chosen
based
the total
volume
of the
chamber
Electrolyte volume isisanother
forfor
P recovery.
Three
different
volumes
(180,
Electrolyte
anotherimportant
importantparameter
parameter
P on
recovery.
Three
different
volumes
and
the
electrode
surface
area.
225, and
investigate
the impact
of anode
cathode
volumesvolumes
on P recovery
(180,
225,270
andmL)
270were
mL) used
weretoused
to investigate
the impact
ofand
anode
and cathode
on P
Increasing the electrolyte volume leads to increases in the P precipitated. The greater the
and energy
The volumes
were chosen
onbased
the total
volume
of the chamber
and the
recovery
andgeneration.
energy generation.
The volumes
werebased
chosen
on the
total volume
of the chamber
volume of synthetic wastewater available at the anode and the cathode, the more P that can be
electrode
surface area.
and
the electrode
surface area.
recovered from the solution. By increasing anode chamber volume, more organic matter is offered
Increasing the
leads
to increases
in the
precipitated.
The greater
the volume
Increasing
the electrolyte
electrolytevolume
volume
leads
to increases
inPthe
P precipitated.
The greater
the
for the organism at the anode for oxidation, and consequently; more reduction reactions occurred at
of synthetic
wastewater
available at
the anode
and
the
cathode,
the cathode,
more P that
can
be recovered
from
volume
of
synthetic
wastewater
available
at
the
anode
and
the
the
more
P
that
can
be
the cathode. Moreover, the impact of changing chamber volume on cathode pH was not significant.
the solution.
By the
increasing
anode
chamber volume,
more organic
matter
is offered
the organism
at
recovered
from
solution.
By increasing
anode chamber
volume,
more
organicfor
matter
is offered
the
anode
for
oxidation,
and
consequently;
more
reduction
reactions
occurred
at
the
cathode.
Moreover,
for the organism at the anode for oxidation, and consequently; more reduction reactions occurred at
the cathode.
impact ofMoreover,
changing the
chamber
volume
on cathode
pH volume
was not on
significant.
the
impact
of changing
chamber
cathode pH was not significant.
Int. J. Environ. Res. Public Health 2016, 13, 375
Int. J. Environ. Res. Public Health 2016, 13, 375
9 of 12
9 of 12
The average power density output (COD = 0.7 g/L) of the MFC under three different volumes
The average power density output (COD = 0.7 g/L) of the MFC under three different volumes
(180, 225, 270 mL) were 25, 20 and 18 mW/m22, respectively. Increasing anode chamber volume from
(180, 225, 270 mL) were 25, 20 and 18 mW/m , respectively. Increasing anode chamber volume
from
180 mL
to 270 mL leads to decreases in the average power density from 25 mW/m2 to 18 mW/m2 .
180 mL to 270 mL leads to decreases in the average power density from 25 mW/m2 to 18 mW/m2. The
The decrease
inpower
powerdensity
density
occurred
the amount
of P precipitates
270 was
mL greater
which was
decrease in
occurred
due due
to thetoamount
of P precipitates
in 270 mLin
which
greater
than
180
mL.
Anode
and
cathode
volumes
have
an
impact
on
energy
generation,
as
increasing
than 180 mL. Anode and cathode volumes have an impact on energy generation, as increasing
the
the volume
thepower
power
density
(Figure
volumeleads
leads to
to decreases
decreases ininthe
density
(Figure
8). 8).
Figure
8. Maximum
power
densityand
and PP recovery
recovery %
electrolyte
volumes.
Figure
8. Maximum
power
density
%atatdifferent
different
electrolyte
volumes.
Achieving high power density and high P recovery concurrently in the system would require
Achieving
highand
power
density and
high P recovery
system would
require
further research
development
as P precipitation
on theconcurrently
cathode leadsin
to the
deteriorating
electricity
further
researchinand
as P precipitation
onpower.
the cathode leads to deteriorating electricity
production
thedevelopment
MFC; and this reduces
the generated
production in the MFC; and this reduces the generated power.
3.3.3. P Recovery at Different Aeration Flow Rates at the Cathode
3.3.3. P Recovery at Different Aeration Flow Rates at the Cathode
Cathode aeration is an important parameter for system optimization. The microbial fuel cells
were operated
under
controlled
dissolved
oxygen
differentThe
aeration
flow rates
werewere
Cathode
aeration
is an
important
parameter
for concentrations;
system optimization.
microbial
fuel cells
usedunder
to examine
the impact
of aeration
onconcentrations;
both energy generation
and
P recovery.
operated
controlled
dissolved
oxygen
different
aeration
flow Identifying
rates were the
used to
impacts
of aeration
flow rates
MFC
performance
is important
for scalingIdentifying
up the system
examine
the impact
of aeration
ononboth
energy
generation
and P recovery.
the and
impacts
assessing the operational cost. Previous studies have shown that cathode aeration has a great impact
of aeration
flow rates on MFC performance is important for scaling up the system and assessing the
on energy generation, where the current increased as the DO concentration increased. These results
operational cost. Previous studies have shown that cathode aeration has a great impact on energy
show the importance of supplying the optimal amount of oxygen to obtain optimal performance of
generation,
where the current increased as the DO concentration increased. These results show the
the MFC [18]. The average current density of the MFC under three different aeration flow rates (no
importance
of
supplying
the optimal
amount
oxygen
to obtain
optimal performance
of theinMFC
2, respectively.
aeration, 50
and 150 mL/min)
were
92, 127of
and
155 mA/m
Applying aeration
the [18].
The average
current
density
of
the
MFC
under
three
different
aeration
flow
rates
(no
aeration,
50 and
cathode chamber with a flow rate of (150 mL/min) increased the average current density by more
2
2
150 mL/min)
and
155 mA/m
respectively.
aeration
theincreased,
cathode chamber
than 40% were
(from92,
92 127
to 155
mA/m
) (Figure, 9).
In addition,Applying
the generated
powerinwas
and
with similar
findings
by Mashkour
anddensity
Rahimnejad
[18]. than 40% (from 92 to
with athese
flowwere
rateinofagreement
(150 mL/min)
increased
the average
current
by more
2 ) (Figure
With
regard 9).
to PInrecovery,
it was
that increasing
aeration
flow rate
the cathode
to
155 mA/m
addition,
the found
generated
power was
increased,
andatthese
were inleads
agreement
increase
in
cathode
pH,
where
oxygen
is
being
used
as
an
electron
acceptor
and
hydroxide
ions
are
with similar findings by Mashkour and Rahimnejad [18].
produced [10]. With (50 mL/min) flow rate, cathode pH reached 8.5, which was optimal for P
With
regard to P recovery, it was found that increasing aeration flow rate at the cathode leads
recovery as struvite; whereas with no aeration, cathode pH reached 7.5 (Figure 10). The difference in
to increase in cathode pH, where oxygen is being used as an electron acceptor and hydroxide ions
P recovery between the flow rates of 50 mL/min and 150 mL/min was not significant; however, there
are produced
[10]. With (50 mL/min) flow rate, cathode pH reached 8.5, which was optimal for
was an increase of 25% in power density when the aeration flow rate increased from 50 to 150
P recovery
as
struvite;
whereas
aeration,
cathode
pH cannot
reachedsupply
7.5 (Figure
10).
The difference
mL/min. These results
showwith
thatno
passive
or low
aeration
enough
oxygen
for the in
P recovery
between
the from
flow rates
of 50chamber.
mL/minThese
and 150
mL/min
was not
however,
transferred
protons
the anode
findings
also show
thatsignificant;
cathode aeration
is anthere
was an
increase of
25%for
in energy
powergeneration
density when
aeration flow rate increased from 50 to 150 mL/min.
influencing
factor
and Pthe
recovery.
These results show that passive or low aeration cannot supply enough oxygen for the transferred
protons from the anode chamber. These findings also show that cathode aeration is an influencing
factor for energy generation and P recovery.
Int.Int.
J. Environ.
Res.
J. Environ.
Res.Public
PublicHealth
Health2016,
2016,13,
13,375
375
of 12
10 10
of 12
Int. J. Environ. Res. Public Health 2016, 13, 375
10 of 12
Figure
9.
Impact
aerationflow
flowrate
rate on
on
energy
and
PPrecovery.
Figure
9.9.Impact
of
aeration
flow
rate
onenergy
energygeneration
generation
and
P
recovery.
Figure
Impact
ofof
aeration
generation
and
recovery.
Figure 10. Impact of aeration flow rate on cathode pH.
Figure10.
10. Impact
Impact of aeration
Figure
aerationflow
flowrate
rateon
oncathode
cathodepH.
pH.
3.4. Statistical Analyses
3.4.
StatisticalAnalyses
Analyses
3.4.
Statistical
The correlations between the monitored parameters and MFC performance was studied to
identify
the influencing
factorsthe
on monitored
MFC performance
for P recovery
and
energy generation.
Table 1 to
The correlations
between
parameters
and MFC
performance
was studied
The correlations between the monitored parameters and MFC performance was studied to identify
showsthe
that
COD significantly
anode pH, for
cathode
pH and
The Table
more 1
identify
influencing
factors onaffected
MFC performance
P recovery
andcurrent
energydensity.
generation.
the influencing
factors
on chamber,
MFC performance
for pH
P recovery
and energy
generation.
Table 1 has
shows
substrate
the anode
the higher
the
at the
cathode.
cathode
aeration
an that
shows
that in
COD
significantly
affected
anode
pH,
cathode
pHMoreover,
and current
density.
The more
COD impact
significantly
affected
anode
pH,
cathode
pH
and
current
density.
The
more
substrate
in
the
anode
onthe
cathode
where supplying
amount
of oxygen
to the
cathode
leads to
an
substrate in
anodepH
chamber,
the higherthe
theoptimal
pH at the
cathode.
Moreover,
cathode
aeration
has
an
chamber,
the
higher
the
pH
at
the
cathode.
Moreover,
cathode
aeration
has
an
impact
on
cathode
pH
increase
in the pH.
pH the
leads
to P precipitation
as struvite
and
deterioration
impact
on cathode
pH High
wherecathode
supplying
optimal
amount of oxygen
to the
cathode
leads toinan
electricity
generation
due to
the precipitates
on
cathode
Theincrease
correlation
analysis
where
supplying
the
amount
of oxygen
toPthe
the
cathodeelectrode.
leads
to an
in the
pH. High
increase
in the
pH.optimal
High
cathode
pH
leads to
precipitation
as struvite
and deterioration
in
suggests
that COD
and cathode aeration
are the
key
influencing factors
for P recovery
and energy
cathode
pH
leads
to
P
precipitation
as
struvite
and
deterioration
in
electricity
generation
due
to the
electricity generation due to the precipitates on the cathode electrode. The correlation analysis
generation.
Designing
the
system based
on these findings
cansuggests
help to achieve
the optimal
condition
precipitates
on the
cathode
electrode.
The correlation
that
and cathode
aeration
suggests that
COD
and cathode
aeration
are the keyanalysis
influencing
factors
forCOD
P recovery
and energy
for
energy
generation
and
P
recovery.
aregeneration.
the key influencing
factors
for
P
recovery
and
energy
generation.
Designing
the
system
based
Designing the system based on these findings can help to achieve the optimal condition on
these
findings
can help and
to achieve
the optimal condition for energy generation and P recovery.
for energy generation
P recovery.
Table 1. Correlation analysis for the MFC performance.
Variable
Aeration
Flow analysis
Anodefor
Cathode
Anode
Cathode
Table
1.1.Correlation
for the
the
MFCperformance.
performance.
Table
Correlation
analysis
MFC
Rate
Aeration- Flow
Aeration flow rate
Aeration
Variable
Variable
Anode
pH
Anode pH
0.457
*
Flow Rate Rate
Cathode
pH
0.619
Aeration
flow rate
- **
Aeration Anode
flow
rate
Anode
ORP
0.111
pH
0.457
* AnodeCathode
pH
0.457 *
ORP
0.070
Cathode pH
0.619 **
Cathode pH
0.619 **
0.783 **
I
density
0.317
Anode
ORP
0.111´0.362 *
Anode
ORPCOD
0.111
Cathode
ORP
0.070
I Density
pH
pH
ORP
ORP
Anode
Cathode
Anode
Cathode
I Density
Cathode
pH pH
Cathode
ORP - I Density
- Anode ORP
pH
ORP
ORP
0.783
--- - **
- -−0.362
−0.330
- -- - -*
- 0.094
0.101
0.015
0.783 **
-**
- −0.095
−0.533
−0.236
−0.099
−0.362
*
−0.330 ´0.330
0.929
0.803
−0.275
- -−0.468-** 0.101**
0.015**
0.094
Cathode ORP
0.070
0.101
0.015
0.094
** Correlation
at the
0.01 level (2-tailed).
***Correlation
significant
0.05 level (2-tailed).
I density is significant
0.317
−0.236is´0.095
−0.095 at the−0.099
I density
0.317
´0.533 ** −0.533
´0.236
´0.099
- 0.929 **
0.929
** ** 0.803 **´0.275
−0.275
-−0.468
** **
CODCOD
0.803
´0.468
** Correlation
is significant
atatthe
Correlationisissignificant
significant
0.05
level
(2-tailed).
** Correlation
is significant
the0.01
0.01level
level(2-tailed).
(2-tailed); ** Correlation
at at
thethe
0.05
level
(2-tailed).
Int. J. Environ. Res. Public Health 2016, 13, 375
11 of 12
4. Conclusions
A mediator-less dual chamber microbial fuel cell was used to investigate concurrent P recovery
and energy generation. Due to the high pH around the cathode, the MFC forms precipitates containing
phosphorus in the catholyte and on the cathode surface. The main component of the precipitate
was determined by X-ray diffraction analysis to be magnesium ammonium phosphate hexahydrate.
P recovery at the cathode ranged from 1% to 38%; and correlation analysis indicated that COD and
aeration flow rate are the key factors influencing P recovery and energy generation. Further studies are
recommended to minimize the energy loss due to the precipitates on the cathode surface. This could
be focused on a movable cathode that can be replaced once the voltage starts to decrease, or finding
a specific material that prevents any particle attachment on the cathode surface.
Acknowledgments: The Authors would like to acknowledge the technical staff at the Cardiff University School
of Engineering for supporting them. The first author would like to thank Kuwait Institute for Scientific Research
for their financial support.
Author Contributions: Abdullah Almatouq designed and operated the systems to generate the performance
and operation data. Data analysis and initial manuscript drafting was undertaken by Abdullah Almatouq.
Akintunde O. Babatunde provided the supervision and academic guidance for the work and the manuscript
writing, and also supervised the data analysis.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Cooper, J.; Lombardi, R.; Boardman, D.; Carliell-Marquet, C. The future distribution and production of
global phosphate rock reserves. Resour. Conserv. Recycl. 2011, 57, 78–86. [CrossRef]
Güney, K.; Weidelener, A.; Krampe, J. Phosphorus recovery from digested sewage sludge as map by the help
of metal ion separation. Water Res. 2008, 42, 4692–4698. [CrossRef] [PubMed]
Doyle, J.D.; Parsons, S.A. Struvite formation, control and recovery. Water Res. 2002, 36, 3925–3940. [CrossRef]
You, J.; Greenman, J.; Melhuish, C.; Ieropoulos, I. Electricity generation and struvite recovery from human
urine using microbial fuel cells. J. Chem. Technol. Biotechnol. 2015, 91, 647–654. [CrossRef]
Cusick, R.D.; Ullery, M.L.; Dempsey, B.A.; Logan, B.E. Electrochemical struvite precipitation from digestate
with a fluidized bed cathode microbial electrolysis cell. Water Res. 2014, 54, 297–306. [CrossRef] [PubMed]
Jaffer, Y.; Clark, T.; Pearce, P.; Parsons, S. Potential phosphorus recovery by struvite formation. Water Res.
2002, 36, 1834–1842. [CrossRef]
Cusick, R.D.; Logan, B.E. Phosphate recovery as struvite within a single chamber microbial electrolysis cell.
Bioresour. Technol. 2012, 107, 110–115. [CrossRef] [PubMed]
Hirooka, K.; Ichihashi, O. Phosphorus recovery from artificial wastewater by microbial fuel cell and its effect
on power generation. Bioresour. Technol. 2013, 137, 368–375. [CrossRef] [PubMed]
Ichihashi, O.; Hirooka, K. Removal and recovery of phosphorus as struvite from swine wastewater using
microbial fuel cell. Bioresour. Technol. 2012, 114, 303–307. [CrossRef] [PubMed]
Kelly, P.T.; He, Z. Nutrients removal and recovery in bioelectrochemical systems: A review. Bioresour. Technol.
2014, 153, 351–360. [CrossRef] [PubMed]
Logan, B.E.; Hamelers, B.; Rozendal, R.; Schröder, U.; Keller, J.; Freguia, S.; Aelterman, P.; Verstraete, W.;
Rabaey, K. Microbial fuel cells: Methodology and technology. Environ. Sci. Technol. 2006, 40, 5181–5192.
[CrossRef] [PubMed]
Sleutels, T.H.; Darus, L.; Hamelers, H.V.; Buisman, C.J. Effect of operational parameters on coulombic
efficiency in bioelectrochemical systems. Bioresour. Technol. 2011, 102, 11172–11176. [CrossRef] [PubMed]
Cheng, S.; Logan, B.E. Increasing power generation for scaling up single-chamber air cathode microbial
fuel cells. Bioresour. Technol. 2011, 102, 4468–4473. [CrossRef] [PubMed]
Zhang, F.; He, Z. Simultaneous nitrification and denitrification with electricity generation in dual-cathode
microbial fuel cells. J. Chem. Technol. Biotechnol. 2012, 87, 153–159. [CrossRef]
Aelterman, P.; Versichele, M.; Marzorati, M.; Boon, N.; Verstraete, W. Loading rate and external
resistance control the electricity generation of microbial fuel cells with different three-dimensional anodes.
Bioresour. Technol. 2008, 99, 8895–8902. [CrossRef] [PubMed]
Int. J. Environ. Res. Public Health 2016, 13, 375
16.
17.
18.
12 of 12
Jadhav, G.; Ghangrekar, M. Performance of microbial fuel cell subjected to variation in pH, temperature,
external load and substrate concentration. Bioresour. Technol. 2009, 100, 717–723. [CrossRef] [PubMed]
Nam, J.Y.; Kim, H.W.; Lim, K.H.; Shin, H.S. Effects of organic loading rates on the continuous electricity
generation from fermented wastewater using a single-chamber microbial fuel cell. Bioresour. Technol. 2010,
101, S33–S37. [CrossRef] [PubMed]
Mashkour, M.; Rahimnejad, M. Effect of various carbon-based cathode electrodes on the performance of
microbial fuel cell. Biofuel Res. J. 2015, 2, 296–300. [CrossRef]
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons by Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).