Urine transduction to usable energy: A modular MFC approach for smartphone and remote system charging

Abstract

This study reports for the first time the full charging of a state-of-the-art mobile smartphone, using Microbial Fuel Cells fed with urine. This was possible by employing a new design of MFC that allowed scaling-up without power density losses. Although it was demonstrated in the past that a basic mobile phone could be charged by MFCs, the present study goes beyond this to show how, simply using urine, an MFC system successfully charges a modern-day smartphone. Several energy-harvesting systems have been tested and results have demonstrated that the charging circuitry of commercially available phones may consume up to 38% of energy on top of the battery capacity. The study concludes by developing a mobile phone charger based on urine, which results in 3h of phone operation (outgoing call) for every 6h of charge time, with as little as 600mL (per charge) of real neat urine.

Full text

Urine transduction to usable energy: A modular MFC approach for
smartphone and remote system charging
Xavier Alexis Walter
a,
⇑
, Andrew Stinchcombe
a
, John Greenman
b
, Ioannis Ieropoulos
a,
⇑
a
Bristol BioEnergy Centre (B-BiC), Bristol Robotics Laboratory, T-Block, Frenchay Campus, University of the West of England (UWE), Bristol BS16 1QY, United Kingdom
b
Microbiology Research Laboratory, Department of Biological, Biomedical and Analytical Sciences, Faculty of Applied Sciences, Frenchay Campus, University of the West of
England, Bristol BS16 1QY, United Kingdom
highlights

1st application of a MFC design that
can be scaled-up without power-
density losses.

1st full charge of a basic phone within
42 and 68 h employing neat urine as
MFC-fuel.

1st full charge of smartphone within
68 and 82 h employing neat urine as
MFC-fuel.

1st charging system allowing 1 h
45 min phone call per 3 h of charge.
graphical abstract
article info
Article history:
Received 25 April 2016
Received in revised form 24 May 2016
Accepted 8 June 2016
Available online xxxx
Keywords:
Phone-charging system
Membrane-less MFCs
Energy management
Sustainable energy
abstract
This study reports for the first time the full charging of a state-of-the-art mobile smartphone, using
Microbial Fuel Cells fed with urine. This was possible by employing a new design of MFC that allowed
scaling-up without power density losses. Although it was demonstrated in the past that a basic mobile
phone could be charged by MFCs, the present study goes beyond this to show how, simply using urine,
an MFC system successfully charges a modern-day smartphone. Several energy-harvesting systems have
been tested and results have demonstrated that the charging circuitry of commercially available phones
may consume up to 38% of energy on top of the battery capacity. The study concludes by developing a
mobile phone charger based on urine, which results in 3 h of phone operation (outgoing call) for every
6 h of charge time, with as little as 600 mL (per charge) of real neat urine.
Ó2016 The Authors. Published by Elsevier Ltd. This is an open accessarticle under the CC BY license (http://
creativecommons.org/licenses/by/4.0/).
1. Introduction
Microbial Fuel Cells (MFCs) are a bio-electrochemical technol-
ogy converting organic waste into electricity [1]. An MFC com-
prises two electrodes, a cathode and an anode, and relies on the
capacity of certain microorganisms to use the anode as their end-
terminal electron acceptor in their anaerobic respiration. The tech-
nology’s intrinsic characteristics are, low power – compared to
established technologies, wide range of organic matter (otherwise
considered as waste) can be used as fuel, low fiscal and mainte-
nance costs, stacking of MFC collectives required for exploitable
power levels and durability with robustness. Over the last three
decades, research in the field has focused on (i) improving the
power density of individual units [2,3], (ii) reducing the cost of
each unit [4–6], (iii) assembling units into stacks to reach
http://dx.doi.org/10.1016/j.apenergy.2016.06.006
0306-2619/Ó2016 The Authors. Published by Elsevier Ltd.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
⇑
Corresponding authors.
E-mail addresses: xavier.walter@brl.ac.uk (X.A. Walter), andrew.stinchcom-
be@brl.ac.uk (A. Stinchcombe), john.greenman@uwe.ac.uk (J. Greenman), ioannis.
ieropoulos@brl.ac.uk (I. Ieropoulos).
Applied Energy xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Applied Energy
journal homepage: www.elsevier.com/locate/apenergy
Please cite this article in press as: Walter XA et al. Urine transduction to usable energy: A modular MFC approach for smartphone and remote system
charging. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.06.006

exploitable power [7,8], (iv) widening the range of potential fuels
[9–11], and (v) demonstrating the implementation of this biotech-
nology into practical applications [12–18]. The results we present
here focus on the latter aspect.
Previous research demonstrated the possibility of charging a
basic mobile phone with urine as the fuel for MFCs (battery par-
tially charged up to 3.7 V in 24 h) [17]. This previous study opened
up the possibility of employing urine in remote locations for
telecommunication purposes as well as demonstrated that the
simplicity of the MFC design – i.e. ceramic membranes – was con-
ducive to mass production. In the present study, the focus is also
on the implementation of the MFC technology as a power supply
for telecommunication purposes, however with significant
advances, bringing the application closer to potential deployment.
The aims of the study were therefore: (i) to fully charge the battery
of both a basic and a smart phone, up to 4.2 V (full charge), (ii) to
have a system independent of any external electrically powered
peripherals (i.e. pump), (iii) to fuel the MFC system with the
amount of urine of a single individual and (iv) to have minimal
footprint and maintenance requirements.
To reach useful levels of power, pluralities of small MFCs need
to be assembled into stacks or cascades. By assembling fluidically
isolated small MFCs and electrically connecting them into paral-
lel/series configurations, both the voltage and current can be
increased [8]. The size of individual MFCs plays an important role
in the performance of the collective, and although units can be
enlarged, this is often at the detriment of power density [8,19]. This
is primarily due to diffusion limitations and sub-optimal volume-
to-surface-area (SAV) ratios, which can be improved by decreasing
the size of the MFC reactor and thus maximally exploiting the
(inter)active surfaces, where the microbial biofilms transfer elec-
trons [8,19,20]. Provided that these parameters are respected in
the re-design of a MFC architecture, then it should be possible to
achieve equal power densities with both unit enlargement (to a
certain extent) and miniaturisation [21]. This could be achieved
by employing a medium size chassis, housing anode and cathode
electrodes without a membrane, exposed to the same electrolyte
[21]; in this setup, the anode is fully submerged to the fuel,
whereas the cathode is only partially exposed to the bulk urine.
Such MFC ‘‘module” can be considered as a collective of 20 MFCs
connected in parallel, and in the present study, 6 such modules
were used and connected in series. Such scaling up refers to the
individual MFC elements within each box, which have been put
together in a collective manner as a modular stack, indicating
how they can be multiplied to produce higher power and treat
higher feedstock volumes. The results demonstrate for the first
time that this design can be employed to power practical applica-
tions by charging various types of cell phones. This system was
recently shown to be robust and stable over 6 months, in a series
electrical configuration – without any cell reversal [21].
An anaerobic ammonium abstraction, measured in the anaero-
bic part of the cathode, gave rise to the hypothesis that in such a
setup, part of the cathode operation is ultimately in the nitrogen
redox cycle [21]. It seems that the anaerobic part of the cathode
was functioning as a denitrifying biocathode [22,23]. A similar
mechanism has recently been reported by Li et al. (2016) who
identified an electroactive ammonium oxidation process [24] and
the use of an anaerobic nitrite-reducing cathode that was added
to a classical MFC set-up. To achieve our aims, 6 modules were
assembled, connected in series and employed to charge a basic
mobile phone, a smartphone and in addition be implemented as
a bespoke mobile phone charger, which provided 1 h 45 min of
call-time for every 3 h of charging, only consuming 600 mL of
urine as fuel every 6 h. It has to be noted that depending on the
type of phone, electronic circuitry was built to manage the charge
cycles.
2. Materials and methods
The whole system comprised 3 parts: the feeding mechanism
representing the ‘urination’ of a single individual (Fig. 1a), a cas-
cade of MFCs as the energy transducer and power source (Fig. 1b),
and electronic circuitry managing the power produced by the MFCs
to charge the cell phones (Fig. 1c).
2.1. Feeding mechanism
For any system and especially for a live biofilm system, the fuel
supply rate is a critical factor [21]. As the ultimate aim of this study
is the deployment of MFCs in real environments, the feeding mech-
anism needed to operate without any energy requirement. Hence,
the pump was only present to (i) simulate an individual urinating
and (ii) to provide a regular feeding pattern. Fuel was pumped into
a tube that tilted, when the volume of urine reached 200 mL, into a
bigger container, which included a syphon. This combined feeding
mechanism was set to distribute 600 mL bursts (2.5 L min
1
)
every 6 h (2.4 L d
1
). In field deployment conditions, such gravity
feeding mechanism implies that a height of 80 cm below the uri-
nal‘s floor is required.
Fresh urine was anonymously collected daily from healthy indi-
viduals and pooled together (pH 6.4–7.4). The reservoir tank was
filled daily with 2.5 L of urine during weekdays, and with 9 L dur-
ing weekends. Hence the urine pumped in the feeding mechanism
aged consistently between 1 and 3 days (pH 8.4–9.3).
2.2. Microbial fuel cell cascade
The MFC cascade consisted of 6 modules whereby the outflow
of upstream module was the inflow to the module downstream
(Fig. 1). The height of the whole setup was 77 cm, including the
Fig. 1. Illustration of the different sub-systems of the setup employed to charge
mobile phones. (a) Feeding mechanism, (b) cascade of MFCs, and (c) various energy
harvesting circuitry allowing optimum mobile phone charging.
2X.A. Walter et al. /Applied Energy xxx (2016) xxx–xxx
Please cite this article in press as: Walter XA et al. Urine transduction to usable energy: A modular MFC approach for smartphone and remote system
charging. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.06.006

10 mm spacers between each box. Containers of 5 L in volume
(H WL: 12 17 27 cm; ‘‘EuroStacking containers”; Plastor
Limited, Berkshire, UK) were used that had 20 MFCs with the cath-
odes separated from the anodes by a 5 mm spacer. The cathodes
and anodes were kept in place by vertical terracotta plates. The
modules were initially inoculated with the anolyte effluent from
established MFCs that had been running under long-term continu-
ous flow conditions with urine. The displacement volume of each
module was of 1.75 L, which is the parameter used for normalising
power densities; the total displacement volume of the whole stack
was 10.5 L. Thus the hydraulic retention time of the whole system
was 105 h, based on a feeding rate of 2.4 L a day. The MFCs were
operating in ambient temperature conditions (22.0 ± 0.5 °C).
As described in [21], cathodes were made from a carbon based
micro-porous layer [5,25], and the anodes from carbon veil fibres
(20 g m
2
; PRF Composite Materials Poole, Dorset, UK). The total
surface area of the anode and cathode, for each module, was
30,000 cm
2
and 3612 cm
2
respectively [21].
2.3. Energy management and phone charging
During this study, various phones, battery capacities and charg-
ing modes were tested (Table 1). In the case of the basic phone, for
comparison purposes, the same model was used as in our previous
study [17] (Samsung GT-E2121B). This model of cell phone has a
lithium ion battery, 3.7 V and 1000 mA h (BMP). A smart phone
(SP) (Samsung Galaxy S I9000), equipped with a lithium ion battery
(3.7 V, 1600 mA h), was also charged. Both the BMP and the SP
were modified to allow a direct charge of the battery (mBMP and
mSP, respectively).
The phone modification was carried out by creating direct
ground and positive connections to the handset battery, thus
bypassing the proprietary phone charging circuitry, although some
parasitic effects may have still occurred since the circuitry was not
physically removed.
The electronics used for the direct charging experiments
included a Texas Instruments bq25504EVM-674 – Ultra Low Power
Boost Converter and an LM385Z-2.5 voltage reference. The
bq25504EVM-674 was configured to the recommended settings
for lithium ion charging and to the use of an external reference
source (LM385Z-2.5) to control the input voltage of the MFCs at
2.5 V (Fig. 2a). A 6800
l
F capacitor was connected to the ‘‘VSTOR
input” of the TI board to help with signal smoothing and harvester
regulation. The efficiency of the harvesting system was measured
using a power supply and resistive loads. The average efficiency
of the harvesting system was calculated at 85% (2.5 V/40 mA Input,
3.2–4.2 V Output).
The electronics used for charging the smartphone (SP) included
two Maxwell BCAP3000 P270 super capacitors and a custom made
Charge Control Board (CCB: Fig. 2b). The bq25504EVM-674 was
modified to give a higher output voltage of 5 V to increase the
energy stored in the super capacitors and therefore more energy
transfer per charge to the phone battery. The method of adjust-
ment for the upper output voltage limit (VBAT_OV) of the
BQ25504EVM was to replace resistors R4 (Rov1) and R3 (Rov2)
using the following equation:
VBAT OV ¼3
2VBIAS 1þR
OV2
R
OV1

VBIAS is a regulated output from the device of 1.25 V (±0.04 V). The
sum of the resistors is recommended to be no higher than 10 M
X
.
Based on these calculations R4 was set to 6.2 M
X
and
R3 = 3.6 M
X
. Both of these resistors were of 1% tolerance. This
would result in an upper output limit of 5.104 V, which was suffi-
cient for our purposes.
The CCB consisted of a MAX917 comparator configured with
additional hysteresis (Output goes high at 5 V and low at 4.5 V).
The output of the MAX917 was connected to a PSMN0R9-30YLD
N-channel MOSFET switch chosen for its very low ON-resistance
(RDSon) of less than 1 m
X
@ > 4.5 V Vgs. When the MAX917
Table 1
Different phones employed with corresponding battery capacity and charging mode.
Denomination Phone Battery
(mA h)
Charge mode
Modified Basic Mobile Phone
(mBMP)
Samsung
E2121
1000 Direct to
battery
Basic Mobile Phone (BMP) 1000 Normal
Plug-in mobile SIM Phone
(PMSP)
150 Direct to
battery
Modified Smart Phone (mSP) Samsung
Galaxy S
1650 Direct to
battery
Smart Phone (SP) 1650 Normal
Fig. 2. Schematic of the energy management circuitry. (a) Circuitry employed for directly charging of the phone (BMP, mBMP, PMSP, mSP) and consisting of a TI energy
harvesting board. (b) Added circuitry comprising two Maxwell super capacitors and one hysteresis board to charge the SP.
X.A. Walter et al. /Applied Energy xxx (2016) xxx–xxx 3
Please cite this article in press as: Walter XA et al. Urine transduction to usable energy: A modular MFC approach for smartphone and remote system
charging. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.06.006

output goes high, the load is switched ON. When it goes low, the
load is switched OFF. The two 3000 F Maxwell super capacitors
were connected together in series and bottom balanced before
any initial charging.
As already mentioned, the objectives were to fully charge a
mobile phone (up to 4.2 V) and find the most appropriate energy
management system for such a full charge, as well as develop a
system capable of powering telecommunication apparatus in a
remote area. To achieve the latter, a ‘‘plug-in mobile SIM phone”
(PMSP, Table 1) was employed: a Samsung GT-E2121B modified
with a 150 mA h lithium battery (Overlander, UK) that was directly
charged from the energy harvesting circuitry (i.e. bypassing the
phone built-in energy management circuitry).
2.4. Data capture and polarisation experiments
Voltage output was monitored against time using an Agilent LXI
34972A data acquisition/switch unit (Farnell, UK). Measurements
were recorded every minute. Recorded raw data were processed
and analysed using Sigma Plot v11 software (Systat Software Inc,
London, UK). The polarisation scans were performed, using an
automated resistorstat [26], and the resistive load values ranged
from 38,000
X
to 4
X
. Each resistor was connected for a period
of 10 min.
3. Results and discussion
3.1. Urine as a power source
The 6 MFCs modules, which were assembled into the single cas-
cade stack, were identical to the large ones described in [21].At
first, 3 modules were connected in series, with 3 additional mod-
ules progressively added. Over the period of 6 months, during
which these modules were electrically connected in series, no cell
reversal was observed (Fig. 3), although the output of module A
decreased to 125 mV during the period between days 118 and
132. Despite this and without any manual attempt to recover,
the output from this module returned to the normal level of ca.
330 mV, at day 132. As the stack output was always kept at 2.5 V
by the reference point of the microchip and bq25504-EVM, the
voltage of the other modules (B–F) compensated for the voltage
reduction of module A.
The polarisation experiment showed that the cascade of 6 mod-
ules was able to supply a maximum of 139 mW at 1.7 V under a
load of 21
X
(Fig. 4). However, to limit the harvesting loading,
the system output voltage was fixed at 2.5 V by the reference chip
and bq25504-EVM. Referring to the polarisation experiment, this
set voltage corresponds to a power of 114 mW under a 53
X
load.
This power rating was further confirmed by the power output
levels measured during the various charging cycles (from 95 to
120 mW, at 2.5 V).
3.2. Direct battery charging
Under the same conditions as previously reported [17]
(i.e. Samsung GT-E2121B; 1000 mA h battery, direct battery charg-
ing), the full charge of the battery was achieved in 42 h (Fig. 5)
whilst treating 4.2 L of urine. This result illustrates a clear effi-
ciency improvement compared to the 24 h needed to reach 3.7 V
Fig. 3. Voltage monitoring of the different modules (A–F) when connected in series
(black curves). At the beginning only 3 modules were connected in series. From day
38 to day 57 the stack was under a 57
X
load (average of 2.47 V, 43.3 mA, 107 mW).
The arrow indicates when the energy management circuitry (Fig. 2a) was connected
to the stack and maintained its output voltage at 2.5 V. All data points from open-
circuit conditions (i.e. when the charge cycle would finish) were removed.
Fig. 4. Polarisation and power curves from the cascade of 6 modules. Each resistor
was connected for a period of 10 min.
Fig. 5. Battery charge of the mBMP phone turned OFF. Voltage, power and current
of the Stack during charge (mBMP). The battery was directly charged by the energy
harvesting board that was set at 4.2 V (i.e. stop charging battery).
4X.A. Walter et al. /Applied Energy xxx (2016) xxx–xxx
Please cite this article in press as: Walter XA et al. Urine transduction to usable energy: A modular MFC approach for smartphone and remote system
charging. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.06.006

in the previous study: here this charge level was reached within
104 min.
The performance and cost (£360) of the present system, results
in an equivalent cost analysis of £8.64/milliwatt/litre of urine,
which is lower compared to previous reports (£13.38) [17]. Hence
the present study pushes boundaries further by demonstrating
that the set system is an economic, reliable and relatively compact
power source for telecommunication apparatus.
The same apparatus was employed to charge the battery of a
smart phone (Samsung Galaxy S I9000; battery of 1650 mA h). As
with the conventional mobile phone, the output of the energy har-
vesting board was directly connected to the battery. Results indi-
cate that the MFC cascade could fully charge the smart phone in
68 h when the phone was OFF (Fig. 6a). As expected, when the
phone was turned ON, it took longer to fully charge the battery
(82 h; Fig. 6b). Nonetheless, results demonstrate that the stack of
MFC was sufficient to provide energy for smart phones to function
as well as to charge their battery.
Since the stack was continuously providing the same amount of
energy (105 mW; Fig. 6a and b), the charge time differences
reflect the amount of energy consumed by the phone when turned
ON. Calculations indicate that 25.7 kJ – over 68 h – was required to
charge the battery, when the phone was OFF. Conversely, when the
phone was ON, 30.9 kJ was needed to fully charge the battery –
over 82 h. Hence, it can be assumed that the stand-by consumption
was 5.2 kJ. Interestingly, after being in open circuit for 17 min the
power output of the stack increased from 105 mW to 120 mW. This
14.3% power increase resulted in a 13.7% faster charge. This illus-
trates and supports the previous report that intermittent loading
increases the efficiency of the membrane-less ceramic-based MFCs
(capacitive-properties) [27].
3.3. Conventional battery charging
To examine further the implementation of MFCs as a power
source for telecommunication purposes, we investigated charging
the mobile phone through their built-in energy management cir-
cuitry (i.e. ‘‘normally”).
The conventional mobile phone (BMP; 1000 mA h battery) was
turned OFF and fully charged in 68 h (Fig. 7). Compared to a direct
connection of the battery, using the BMP built-in charging circuitry
increased the charging time by 26 h (Figs. 7 and 5). This charge
cycle of 68 h at 110 mW corresponds to a total energy of 26.9 kJ.
That is 11.8 kJ more than the direct charge from the battery
(Fig. 5). This result implies that the BMP built-in circuitry required
11.8 kJ to charge 15.1 kJ in the battery.
Fig. 6. Monitoring of the battery charge of the mSP phone (modified Samsung Galaxy S I9000). Data shown are the voltage of the phone’s battery and the stack’s power and
current during charging (mSP; stack being maintained at 2.5 V output). (a) Direct charge of the 1650 mA h battery of the turned OFF mSP. (b) Direct charge of the battery with
the mSP turned ON. The difference of charging time is due to the power consumption of the running phone, thus illustrating that the cascade produced more energy than
what was required for the phone to function.
Fig. 7. Battery charge of the BMP phone turned OFF (conventional phone). The
figure shows the battery voltage and the stack power and current. The phone was
charged through its micro-USB port, i.e. by the phone built-in proprietary energy
management circuitry.
X.A. Walter et al. /Applied Energy xxx (2016) xxx–xxx 5
Please cite this article in press as: Walter XA et al. Urine transduction to usable energy: A modular MFC approach for smartphone and remote system
charging. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.06.006

Following the same approach, the cascade was used to charge
the smart phone that was switched OFF (SP; 1650 mA h battery).
However, although the built-in charging circuitry recognised the
power source as sufficient to charge the phone (charge symbol vis-
ible), measurements of the battery voltage indicated that no charge
was occurring. Hence, a bespoke circuit was built to charge the
phone through its built-in proprietary circuitry. The system that
was implemented comprised 2 super-capacitors and an external
hysteresis board (EHB) inserted between the harvesting board
and the SP. The hysteresis mechanism of the harvesting board
was set to stop harvesting energy when the output voltage (one
of the super-capacitors) was above 5.0 V. The EHB was set to dis-
charge the energy from super-capacitors to the phone between
the voltage levels of 5.0 V and 4.5 V.
The full charge of the SP took 110 h (Fig. 8a), during which the
MFC cascade produced an average power of 98 mW at 2.5 V. At the
beginning of the experiment, the hysteresis board was unstable
and was triggered before the super-capacitors reached 5.0 V
(Fig. 8a). Despite this and even if irregular, the current flowing
towards the phone (425 mA; Fig. 8b) was sufficient to charge the
SP. As illustrated (Fig. 8b), towards the end of the charge, the cur-
rent flow gradually decreased as the battery was getting full.
This intermittent-charge required a total energy of 38.8 kJ
(98 mW over 110 h). In comparison, charging directly the battery
required 25.7 kJ (Fig. 6a). This means that the stand-by energy
was 13.2 kJ. Without questioning the efficiency of the phone’s
built-in charging circuitry, results indicate that either more MFCs
are needed to reach the required power, or the phones to be
charged have to be adapted to this specific charging system.
As illustrated by the time needed to fully charge either the con-
ventional phone (BMP) or the smart-phone (SP), the size of the sys-
tem has to be increased to practically charge modern phones that
have higher power capacity than the one employed here. Such a
larger charging system reaching USB-1 output implies (i) double
the number of modules to reach 5 V, and (ii) 10 times larger com-
partmentalized modules to produce 500 mA, thus removing the
need for any energy management circuitry. In the present study,
it is the lack of fuel (urine) that prevented the scale-up of the
MFC charging system. Nonetheless, the presented stack (i) was able
to fully charge a smart-phone for the first time, and (ii) was suffi-
cient to power an adapted off-grid plug-in SIM phone.
3.4. Adapted plug-in mobile SIM phone
The aim for the implementation of such a charging mechanism
is to provide power in remote areas, which also means appropri-
ately customizing the telecommunication device. Hence, to adapt
the phone, the charging/discharging duty cycle needs shortening
– i.e. to charge reasonably quickly to a practical level that would
allow for daily autonomy. Because the MFC cascade was tailored
to a single individual use, its size increase was not desired. More-
over, the energy harvesting board – needed for the cascade home-
ostasis and optimum operation – was redundant considering the
phones’ proprietary built-in charging circuitry. Hence, as a proof-
of-concept, the mBMP (i.e. bypassing built-in energy management
circuitry) was employed in combination with a smaller battery
(150 mA h). This phone was designated as a ‘‘plug-in mobile SIM
phone” (PMSP).
Results showed that the battery was fully charged in approxi-
mately 6 h (Fig. 9a), which interestingly corresponds to the refu-
elling time (600 mL every 6 h). However, because it would not be
practical to empty the battery and wait another 6 h before using
the phone again (i.e. emergency call), the device was left
plugged-into its urine-fuelled MFC power supply. In addition, it
Fig. 8. Charging of the Samsung Galaxy S I9000 through its USB port (SP). (a) Power produced by the stack and voltage illustrating the charge/discharge cycles of the
supercapacitor. The pictures indicate the charge state of the phone (yellow dots indicate when pictures were taken). (b) Current delivered by the supercapacitors during the
discharge cycles. The time between each discharge cycle (approx. 10 h of using the phone) is not represented. (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)
6X.A. Walter et al. /Applied Energy xxx (2016) xxx–xxx
Please cite this article in press as: Walter XA et al. Urine transduction to usable energy: A modular MFC approach for smartphone and remote system
charging. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.06.006

was considered that on regular use, the battery should not be
allowed to drop below 3.8 V (charge half-time). With these consid-
erations in mind, increasingly longer phone calls were made,
ensuring to stop communications as soon as the battery voltage
reached 3.8 V, which corresponded to half the charging time of
the PMSP’s battery.
Results show that the longest call lasted 105 min before the bat-
tery voltage reached the 3.8 V (Fig. 9b). Plotting the call time with
the corresponding charge time indicated that the charge/use duty
cycle was 2 to 1: Y= 2.295 X± 5.5 min (R
2
= 0.962). Ystands for
the charging time and Xfor the call length. In this experiment, the
cascade of 6 modules delivers sufficient energy to run the phone.
However, this energy was insufficient to power the communication
itself, hence the need for the smaller battery, which in this case was
acting as an energy storage/buffer for communication. Practically,
these results show that a single individual urinating 600 mL every
6 h would have 1 h 45 min of phone conversation every 3 h, and
another 1 h 45 min of communication in a case of emergency.
4. Conclusions
Results demonstrated that the MFC technology is sufficiently
mature for implementation in out-of-the-lab applications. The sys-
tem developed here – consisting of the feeding mechanisms, the
MFC cascade and the energy harvesting circuitry – demonstrated
for the first time the full charge of mobile phones, including smart
phones, with neat urine as fuel. Moreover, it was shown that this
system could be tailored to specific needs, in the present case to
the need of a single individual. At the moment this system has a
charge/discharge duty cycle of 2 to 1. Nevertheless, the results pre-
sented here are the first demonstration, in real conditions, of a
complete MFC-based system able to power telecommunication
devices in remote areas.
Acknowledgements
This work was funded by the EPSRC under the New Directions
grant no. EP/L002132/1.
References
[1] Potter MC. Electrical effects accompanying the decomposition of organic
compounds. Proc R Soc B 1911;84:260–76.
[2] Schroder U, Niessen J, Scholz F. A generation of microbial fuel cells with
current outputs boosted by more than one order of magnitude. Angew Chem
Int Ed 2003;42:2880–3.
[3] Qiao Y, Li CM, Bao S-J, Bao Q-L. Carbon nanotube/polyaniline composite as
anode material for microbial fuel cells. J Power Sources 2007;170:79–84.
[4] Kim HJ, Park HS, Hyun MS, Chang IS, Kim M, Kim BH. A mediator-less microbial
fuel cell using a metal reducing bacterium, Shewanella putrefaciense. Enzyme
Microb Technol 2002;30:145–52.
[5] Gajda I, Stinchcombe A, Greenman J, Melhuish C, Ieropoulos I. Ceramic MFCs
with internal cathode producing sufficient power for practical applications. Int
J Hydrogen Energy 2015;40:14627–31.
[6] Chaudhuri SK, Lovley DR. Electricity generation by direct oxidation of glucose
in mediatorless microbial fuel cells. Nat Biotechnol 2003;21:1229–32.
[7] Aelterman P, Rabaey K, Pham HT, Boon N, Verstraete W. Continuous electricity
generation at high voltages and currents using stacked microbial fuel cells.
Environ Sci Technol 2006;40:3388–94.
[8] Ieropoulos I, Greenman J, Melhuish C. Microbial fuel cells based on carbon veil
electrodes: stack configuration and scalability. Int J Energy Res
2008;32:1228–40.
[9] Liu H, Ramnarayanan R, Logan BE. Production of electricity during wastewater
treatment using a single chamber microbial fuel cell. Environ Sci Technol
2004;38:2281–5.
[10] Cercado-Quezada B, Delia M-L, Bergel A. Treatment of dairy wastes with a
microbial anode formed from garden compost. J Appl Electrochem
2010;40:225–32.
[11] Velasquez-Orta SB, Curtis TP, Logan BE. Energy from algae using microbial fuel
cells. Biotechnol Bioeng 2009;103:1068–76.
[12] Wilkinson S. ‘‘Gastrobots” – benefits and challenges of microbial fuel cells in
food powered robot applications. Auton Robot 2000;9:99–111.
[13] Ieropoulos I, Melhuish C, Greenman J. Artificial metabolism: towards true
energetic autonomy in artificial life. Lect Notes Artif Int 2003;2801:792–9.
[14] Chang IS, Moon H, Jang JK, Kim BH. Improvement of a microbial fuel cell
performance as a BOD sensor using respiratory inhibitors. Biosens Bioelectron
2005;20:1856–9.
[15] Tender LM, Gray SA, Groveman E, Lowy DA, Kauffman P, Melhado J, et al. The
first demonstration of a microbial fuel cell as a viable power supply: powering
a meteorological buoy. J Power Sources 2008;179:571–5.
[16] Su L, Jia W, Hou C, Lei Y. Microbial biosensors: a review. Biosens Bioelectron
2011;26:1788–99.
[17] Ieropoulos IA, Ledezma P, Stinchcombe A, Papaharalabos G, Melhuish C,
Greenman J. Waste to real energy: the first MFC powered mobile phone. Phys
Chem Chem Phys 2013;15:15312–6.
[18] Winfield J, Chambers LD, Rossiter J, Stinchcombe A, Walter XA, Greenman J,
et al. Fade to green: a biodegradable stack of microbial fuel cells.
ChemSusChem 2015;8:2705–12.
[19] Ren H, Lee H-S, Chae J. Miniaturizing microbial fuel cells for potential portable
power sources: promises and challenges. Microfluid Nanofluid
2012;13:353–81.
[20] Walter XA, Forbes S, Greenman J, Ieropoulos IA. From single MFC to cascade
configuration: the relationship between size, hydraulic retention time and
power density. Sustain Energy Technol Assess 2016;14:74–9.
[21] Walter XA, Gajda I, Forbes S, Winfield J, Greenman J, Ieropoulos I. Scaling-up of
a novel, simplified MFC stack based on a self-stratifying urine column.
Biotechnol Biofuel 2016;9:93.
[22] Xiao Y, Zheng Y, Wu S, Yang Z-H, Zhao F. Bacterial community structure of
autotrophic denitrification biocathode by 454 pyrosequencing of the 16S rRNA
Gene. Microb Ecol 2015;69:492–9.
[23] Li W, Zhang S, Chen G, Hua Y. Simultaneous electricity generation and
pollutant removal in microbial fuel cell with denitrifying biocathode over
nitrite. Appl Energy 2014;126:136–41.
[24] Li Y, Williams I, Xu Z, Li B, Li B. Energy-positive nitrogen removal using the
integrated short-cut nitrification and autotrophic denitrification microbial fuel
cells (MFCs). Appl Energy 2016;163:352–60.
[25] Santoro C, Agrios A, Pasaogullari U, Li B. Effects of gas diffusion layer (GDL) and
micro porous layer (MPL) on cathode performance in microbial fuel cells
(MFCs). Int J Hydrogen Energy 2011;36:13096–104.
[26] Degrenne N, Buret F, Allard B, Bevilacqua P. Electrical energy generation from a
large number of microbial fuel cells operating at maximum power point
electrical load. J Power Sources 2012;205:188–93.
[27] Walter XA, Greenman J, Ieropoulos IA. Intermittent load implementation in
microbial fuel cells improves power performance. Bioresour Technol
2014;172:365–72.
Fig. 9. Charge and discharge curves of the Plug-in mobile SIM Phone (PMSP; turned ON). (a) Voltage of the 150 mA h battery illustrating the consistency in the charge cycles.
(b) Temporal voltage measurements showing the relationship between the discharge time (outgoing phone calls; coloured in grey) and the charging time.
X.A. Walter et al. /Applied Energy xxx (2016) xxx–xxx 7
Please cite this article in press as: Walter XA et al. Urine transduction to usable energy: A modular MFC approach for smartphone and remote system
charging. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.06.006