Int. J. Electrochem. Sci., 14 (2019) 4557 – 4570, doi: 10.20964/2019.05.

01

International Journal of
ELECTROCHEMICAL
SCIENCE
www.electrochemsci.org

Potassium Permanganate as an Oxidant for a Microfluidic

Direct Formate Fuel Cell
Chunmei Liu1,*, Haihua Liu2, Lei Liu3
1
Institute of Vehicle and Transportation Engineering, Henan University of Science and Technology,
Luoyang 471003, Henan Province, China
2
School of Energy and Architecture, Xi'an Aeronautics University, Xi'an 710000, Shaanxi Province,
China
3
China Nonferrous Metals Processing Technology Co., Ltd., Luoyang 471003, Henan Province, China
*
E-mail: [email protected]

Received: 18 January 2019 / Accepted: 22 February 2019 / Published: 10 April 2019

A simple Y-type microfluidic fuel cell (MFC) was constructed with graphite plates as the electrodes.
Unlike the formic acid/potassium permanganate MFCs previously reported, an alkaline sodium
formate solution (a fuel) and an acid potassium permanganate solution (KMnO4, an oxidant) were
adopted by our MFC. The flowing pattern in the main channel was a typical laminar flow. The effects
of three parameters such as the concentrations of KMnO4, the concentrations of H2SO4 (a supporting
catholyte), and the flow rates of the reactants on the power output of the MFC were studied. The
experimental results showed that the cell performance first improved but then decreased with
increasing each parameter. The optimum MFC exhibited a maximum power density of 4.51 mW cm -2
and a limiting current density of 6.67 mA cm-2 with 50 mM KMnO4, 0.5 M H2SO4 and a reactant flow
rate of 30 mL h-1. Although the power densities and limiting current densities of our MFC are not the
highest in the MFCs using KMnO4 as oxidants, its open circuit voltage remains at the top. Our work
demonstrates that it is feasible to construct MFCs using alkaline formate and acidic permanganate
solutions, which diversifies the types of MFCs containing KMnO4 as the oxidant.

Keywords: Microfluidic fuel cells; Potassium permanganate; Supporting catholyte concentration;

Reactant flow rate; Open circuit voltage

1. INTRODUCTION

Due to the increasing power demand for modern portable electronics such as laptops, cell
phones and medical instruments, micro-sized fuel cells are attracting worldwide research efforts as
promising miniaturized power sources [1]. Microfluidic fuel cells (MFCs), also known as laminar
flow-based fuel cells or membraneless fuel cells, have served as a major member in the family of
Int. J. Electrochem. Sci., Vol. 14, 2019 4558

micro-sized fuel cells for more than a decade [2]. MFCs exploit the co-laminar flow feature of the two
fluids in microscale channels to naturally separate the anolyte and catholyte without the need for high
cost physical barriers such as proton exchange membranes (PEMs). This membraneless configuration
can not only avoid a series of problems related to the membranes such as difficulty in water
management, fuel crossover, and membrane fouling, but also gain the higher power density than the
micro PEM fuel cell [3]. Thus, the development of laminar flow-based MFCs is highly desirable for
making high performance and economically portable power sources.
The fuel and oxidant solutions flow laminarly in parallel with each other in a microchannel of a
MFC. The supporting electrolytes in the two streams can increase electrical conductivity of the streams
to promote ionic transport. The mixing of the two fluids takes place only at the liquid-liquid interface
in the middle of the microchannel. In addition, the fuels, oxidants and supporting electrolytes could be
optimized independently in the MFCs. The electrolytes at the anode and cathode can be both acidic or
alkaline media, or different acidic/alkaline media. It is reported that MFCs with the alkaline anode
solutions and the acidic cathode solutions led to a very high open circuit voltage [4].
At present, a variety of fuels have been used in MFCs such as formic acid [4-5], methanol [6],
hydrogen gas [7], hydrogen peroxide [8], vanadium-based redox species [9-10] and formate [11-12].
Among these fuels, formate is proven to be an efficient and effective fuel due to its advantages
including its carbon-neutral characteristic [13], high theoretical potential [14], no poisoning effect in
alkaline media [15], and easy to store and transport [11].
The performance of MFCs is mainly limited by the rate of reactant transport to the active sites
of the electrodes [16], especially for the cathodic half cell reactions [17]. The dissolved oxygen as an
oxidant has the major limitations of limited solubility (2 ~ 4 mM) in water and low diffusivity (~ 2 ×
10-5 cm2 s-1). If using air as the oxidant for the cathode reaction (such the cathode coined as "air-
breathing cathode"), both the oxidant concentration and the diffusivity (~ 0.2 cm2 s-1) can be
significantly augmented. The first MFC with an air-breathing cathode was presented by Ferrigno [2].
To prevent fuel crossover and facilitate ion transport for air-breathing cathodes, blank electrolytes as
supporting electrolytes are needed. However, besides the aforementioned merits of the air-breathing
cathodes, some drawbacks exist. One of them is the oxygen reduction at the cathode that demands
expensive noble metals (e.g. Pt) as catalysts.
Instead of using air as the oxidant, the performance of MFCs can also be improved with other
liquid oxidants such as H2O2 [18], NaClO4 [19], vanadium redox species [20-21] and KMnO4 [22-25].
In the presence of Pt catalyst, H2O2 and NaClO4 can self-decompose to release oxygen gas and
chlorine gas. These bubbles could interrupt the liquid-liquid interface or the laminar flow between the
two streams, leading to severe mixing of the two streams and significantly reducing the cell
performance. Vanadium redox species generate no gaseous products when undergoing redox reactions
and their reactions do not need catalysts, but vanadium element is toxic and must be strictly prevented
from leaking. Considering the trend of miniaturization and integration for future MFCs, an ideal
oxidant needs to equip with a high electrochemical activity, large solubility, no gases production
during a redox reaction and environmental friendliness.
Potassium permanganate (KMnO4) is one possible candidate oxidant. It has many advantages
such as a high oxidation ability, large redox potential and the proceeding of its reduction reaction with
Int. J. Electrochem. Sci., Vol. 14, 2019 4559

no catalyst. Previous researches using KMnO4 as the oxidant for MFCs mainly investigated the
different cell constructions such as the radial flowing of the two streams [22], the three inlets [23], and
a bridge-shaped microchannel cross-section [24], or with electrospun CNx nanofibers as the cathode
[25].
Unlike other researches adopting KMnO4 as the oxidant in acidic electrolytes of the MFCs [22-
25, 26], the MFC in this study involves the alkaline sodium formate solution and the acidic KMnO4
solution as the fuel and oxidant, respectively. The usage of the mixed supporting electrolyte in this
experiment is to promote the electrochemical reactions on the electrodes and obtain a high open circuit
voltage. In this study, the effects of the oxidant concentrations, the supporting catholyte concentrations
and flowing rates on the MFC performance are investigated.

2. EXPERIMENTAL CONSTRUCTION AND METHODS

2.1 MFC construction

(a) (b)

Figure 1. Schematic of the microfluidic fuel cell using graphite plates as electrodes ((a) the assembled
MFC, (b) exploded view of the MFC. Note: The wedge and flake are made of PMMA. Here,
the two plates are shown in yellow just to make them clear.).

In this experiment, a Y-shaped MFC was constructed, as schematically shown in Figure 1. The
actual picture of the MFC is attached to the manuscript as Figure S1 of the Supporting Materials. The
cell was comprised of two PMMA (poly methyl methacrylate) plates and two 1 mm thick graphite
plates as the electrodes. The microchannel of the MFC was lined by the sides of the two graphite plates
placed in parallel, with the dimensions of 40 mm × 2 mm × 1 mm. The two small PMMA plates were
put at the two ends of the microchannel to separate the electrodes and seal this channel. The effective
surface area of the electrode was 0.4 cm2 (40 mm × 1 mm) [27, 28]. Two 2 mm inlets were at the two
sides of the top plate and proceed to connect the microchannel, while one 3 mm outlet was mounted at
the bottom of the bottom plate. The whole device was eventually assembled by six 4 mm screw bolts.
Pd particles serving as anode catalysts were loaded onto the one side of the anode electrode.
The Pd plating solution is 1.0 wt. % PdCl2 in 1 M HCl solution. The electrodeposition of the Pd
catalyst involved a first of applying a potential of 0.0 V vs. Ag/AgCl reference electrode (0.198 V vs.
standard hydrogen electrode, SHE) for 5 minutes, followed by immersing in the 5% Nafion solution
Int. J. Electrochem. Sci., Vol. 14, 2019 4560

for 5 minutes. This process was repeated for five times, until the Pd catalyst loading was 5 mg cm-2
[28, 29]. The electrodeposition was carried out by a Zennium electrochemical workstation (Zahner,
Germany) via a three electrode mode, with the graphite plate as the working electrode, an Ag/AgCl
electrode in saturated KCl solution as the reference electrode, and a Pt sheet (10 mm × 50 mm × 0.2
mm) as the counter electrode.

2.2 The visual experiment for the laminar flowing

The visual experiment for the in-parallel laminar flow of the two streams in the microchannel
was recorded by Nikon D7000 (Japan) camera equipped with an AFS micro Nikon 105 mm optical
lens.

2.3 Performance evaluation

The fuel and oxidant solution were delivered into the microchannel of the cell by a dual pass
injection syringe pump (LSP02-1B, Baoding Longer Precision Pump Co., Ltd., China). The fuel
solution was a mixture of 2 M sodium formate (HCOONa) and 2 M NaOH solution blended in a 1:1
volume ratio. The oxidant stream was composed of different concentrations of the KMnO4 solution as
the oxidant and H2SO4 solution as the supporting electrolyte. The cell performance was gained by a
chronoamperometry method using an electrochemical workstation (Zahner, Germany). The cell
voltages were varied from the open circuit to the short circuit with a decreasing potential step of 200
mV. Each cell voltage was kept for 3 minutes. The currents were recorded at the end of each potential
step. The electrode potentials were the potential difference between the each electrode and the
Ag/AgCl reference electrode located near the outlet. The schematic figure of experimental
measurement system is shown as Figure S2 of the Supporting Material. The current density and power
density were normalized to the effective electrode surface area (0.4 cm2).
The electrochemical reaction at the anode is:
HCOO- + 3OH- → CO32- + 2H2O + 2e-, E0= -1.17 V vs. SHE (1)
There are different routes for the cathode reaction of KMnO4 reduction, the main reduction
route in acidic media is:
MnO4- + 8H+ + 5e- → Mn2++ 4H2O, E0= 1.507 V vs. SHE (2)
-
The generated Mn(II) ions can be oxidized by MnO4 in acidic solutions, producing insoluble
MnO2 [30]:
2MnO4- + 3Mn2+ + 2H2O →5MnO2+ 4H+, E0= 0.46 V vs. SHE (3)
The theoretical open circuit voltage (OCV) of this MFC is 2.677 V (assuming the fuel and the
oxidant are completely oxidized and reduced, respectively).
Int. J. Electrochem. Sci., Vol. 14, 2019 4561

3. RESULTS AND DISCUSSION

3.1 Visual experiment of the parallel flowing

MFCs exploit the feature of two streams in a laminar flowing in a microchannel. Therefore, the
laminar flow of the two streams is a prerequisite for the normal operation of MFCs. The Reynold
number (Re) is a characteristic value to judge the flowing state, which is calculated according to the
formula below [31]:
Re= vDe/μ
where v is the average velocity of flow on the transversal surface of the microchannel (m s-1),
De is the equivalent diameter (m), De= 4A/Χ [A transversal surface area of the main channel (m2) and X
wetted perimeter (m)], μ is the dynamic viscosity coefficient of the solution (m2 s-1), here μ is
approximately treated as the viscosity coefficient of water at 25 ℃ (μ = 0.871×10-6 m2 s-1 [31]). The
calculated Re values under the different flow rates are listed in Table 1. Since all Re values are below 6,
the flowing state in the microchannel is typically laminar.

Table 1. Re values under the different flow rates

Flux (mL h-1) 10 20 30 40

Re 1.27 2.54 3.80 5.07

To further verify the laminar flow state of the two streams in the microchannel, the visual
experiment was carried out. To clearly visualize the flowing state, deionized water and deionized water
dyed with red ink were adopted as the two streams, respectively. In the reactor comprised of the two
graphite plates without the Pd catalysts, the two fluids were simultaneously sent into the reactor by a
syringe pump. The photographs of the visual experiment are shown in Figure 2. The two water streams
in the main channel under various flow rates ranging from 10 mL h-1 to 40 mL h-1 do not mix with
each other, which is consistent with the calculated Re values.

Figure 2. Digital photos showing the two fluids in the microchannel under different flow rates: ((a) 10
mL h-1, (b) 20 mL h-1, (c) 30 mL h-1, (d) 40 mL h-1).

3.2 Effects of KMnO4 concentrations on the cell performance

KMnO4 as an oxidant in our MFC plays an important role in determining the cell performance.
The effects of KMnO4 concentrations ranged from 12.5 mM to 100 mM on the cell performance and
electrode potentials are shown in Figure 3 and Figure 4, respectively (the flow rate is kept constant at
Int. J. Electrochem. Sci., Vol. 14, 2019 4562

20 mL h-1 and the H2SO4 solution concentration is 0.5 M). Table 2 enlists the OCVs, the peak power
densities and the limiting current densities of the MFC under the different KMnO4 concentrations.

Figure 3. Polarization curves (left Y-axis) and power density curves (right Y-axis) at various KMnO4
concentrations with 0.5 M H2SO4 as a supporting electrolyte under a flow rate of 20 mL h-1.

Figure 4. Electrode potential curves at different KMnO4 concentrations containing 0.5 M H2SO4 as a
supporting electrolyte under a flow rate of 20 mL h-1.

Based on the results of Figure 3 and Table 2, the OCVs of the MFCs are all more than 1.78 V
with different KMnO4 concentrations. These OCVs are less than the theoretical cell voltage (2.677 V),
which can be attributed to the two reasons. The first one is the mixing between the fuel and oxidant
solutions at the intersection of the two fluids happened. The second one is existence of side reaction of
(3) that results in MnO2. Nevertheless, these OCVs in the mixed media are higher than those (about
1.45 V) of the MFCs with all acidic media reported previously [31].
Int. J. Electrochem. Sci., Vol. 14, 2019 4563

Table 2. OCVs, peak power densities and limiting current densities under the different KMnO4
concentrations

KMnO4 concentrations OCV Peak power Densities Limiting current densities

(mM) (V) (mW cm-2) (mA cm-2)
12.5 1.78 2.37 3.22
25 1.81 3.22 4.71
50 1.82 4.06 5.85
100 1.83 3.80 5.22

Figure 3 illustrates that the cell performance first increases and then decreases with increasing
the oxidant concentrations. When the KMnO4 concentration increases from 12.5 mM to 50 mM, the
peak power density rises from 2.37 mW cm-2 to 4.06 mW cm-2 by 71.31%. This result can be ascribed
to the higher oxidant concentration that not only supplies more oxidants, but also improves the
transport of the oxidant to the cathode electrode. The latter promotes the electrochemical reaction rate
on the cathode and improves the overall cell performance. While the oxidant concentration is further
increased to 100 mM, the cell performance startes to decrease. The maximum power density and the
maximal limiting current density is 3.80 mW cm-2 and 5.22 mA cm-2 at 100 mM KMnO4, respectively.
Although the higher oxidant concentration improves the oxidant ability and decreases the cathode
resistance as discussed before, the more insoluble MnO2 are inevitably produced from the direct
reaction of KMnO4 and HCOONa in the inner diffusion zone and from the secondary reaction of
KMnO4 reduction. The yielded MnO2 could clog the microchannel and prevent the oxidant from
contacting with the cathode. This negative impact exacerbates under highly concentrated KMnO 4
solutions and long term operation of the MFC.
From the electrode potential plots shown in Figure 4, the cathode performance shows an
obvious mass transfer limitation at 12.5 mM KMnO4, where the cathode potential decreases sharply.
When the KMnO4 concentration increases to 50 mM, the higher oxidant concentration improves the
mass transfer of the oxidant and facilitates the mass transfer at the cathode. At 50 mM KMnO4, the cell
performance reaches the best. When further increasing the KMnO4 concentration to 100 mM, the
cathode potential reduces slightly and the anode potential slightly increases, compared to the potentials
at 50 mM KMnO4. This phenomenon is due to the production of insoluble MnO2 that clogs the
microchannel and hinders mass transfer of the reactants.

3.3 Effects of H2SO4 concentrations on the cell performance

H2SO4 as the supporting electrolyte in the catholyte provides the H+ ions that not only improve
the electrical conductivity of the catholyte, but also promote the oxidation activity of KMnO4.
The effects of H2SO4 concentrations (0.25, 0.5, 1.0 M) on the cell performance and electrode
potentials are illustrated in Figure 5 and Figure 6, respectively (the flow rate is kept constant at 20 mL
h-1 and the concentration of KMnO4 is 50 mM). And the OCVs, the peak power densities and the
limiting current densities of the MFC under different H2SO4 concentrations are summarized in Table 3.
Int. J. Electrochem. Sci., Vol. 14, 2019 4564

Figure 5. Polarization curves (left Y-axis) and power density curves (right Y-axis) at different H2SO4
concentrations of 50 mM KMnO4 and a flow rate of 20 mL h-1.

Figure 6. Electrode potential curves under different H2SO4 concentrations at 50 mM KMnO4 and a
flow rate of 20 mL h-1.

Table 3. OCVs, peak power densities and limiting current densities under the different H2SO4
concentrations

H2SO4 concentrations (M) OCV (V) Peak power Densities Limiting current
(mW cm-2) densities (mA cm-2)
0.25 1.67 1.41 3.58
0.5 1.82 4.06 5.85
1.0 1.80 3.13 4.25

Concluded from Figure 5 and Table 3, the peak power density and the limiting current density
of the cell with 0.5 M H2SO4 is 4.06 mW cm-2 and 5.85 mA cm-2, respectively, increasing by 1.87
Int. J. Electrochem. Sci., Vol. 14, 2019 4565

times and 0.96 times than those of the MFCs with 0.25 M H2SO4. The enhancement is due to the
increased supporting electrolyte concentration that accelerates the mass transfer of cathode reactions
and reduces the ohmic resistance at the cathode. In addition, the reduction of KMnO4 at the cathode is
aided by H+ and enhanced with the augment of H+ concentrations. If the supporting electrolyte
concentration is low, the H+ concentration near the cathode surface is small, leading to slow transfer
rate of H+ to the cathode electrode and compromising cell performance. At 1.0 M H2SO4, the cell
performance decreases appreciably at the intermediate and high current densities, except that the cell
performance is a little higher than that of cell at 0.5 M H2SO4 at the low current density (< 2.08 mA
cm-2). This phenomenon could be explained by the poisoning effect of SO42- on the catalyst activity of
Pd catalysts [32]. Degradation in the catalytic activity could increase the anode potential to the positive
values, deteriorating the anode performance thus decreasing the cell performance.

3.4 Effects of flow rates on the cell performance

The flow rate is one of the key factors dictating the cell performance by influencing the flowing
state and the mass transfer of the reactants and products. The effects of flow rates from 10 mL h -1 to 40
mL h-1 on the cell performance and electrode potentials are investigated, with results presented in
Figure 7 and Figure 8, respectively (the concentrations of KMnO4 and H2SO4 solutions are kept
constant at 50 mM and 0.5 M, respectively). The OCVs, the peak power densities and the limiting
current densities of the MFC under the different flow rates are summarized in Table 4.
From Figure 7 and Table 4, the cell performance increases with the flow rate from 10 mL h -1 to
30 mL h-1. It is mainly due to the improvement in the flow rate which is feasible to not only transport
the reactants to the electrodes, but also take the products out of the main channel. These improve the
mass transfer and promote the electrochemical reactions on the electrodes. The peak power density and
limiting current density of the MFC at a flow rate of 30 mL h-1 is 4.51 mW cm-2 and 6.67 mA cm-2,
respectively, 2.03 times and 1.47 times higher than those of MFCs at 10 mL h -1. When the flow rate is
further increased to 40 mL h-1, the cell performance decreases. This decrement is because that the
further increased flow rate can cause a hydraulic instability [27-29] in the microchannel. This fluid
instability destroys the liquid-liquid interface of the laminar flow and mixes the two streams.
Therefore, little electricity can be gained since the oxidants and fuels react directly in the fluids rather
than through the external electric circuit.
It is noted that the OCV of the cell at 10 mL h-1 is about 1.65 V, lower than 1.82 V and 1.83 V
at 20 mL h-1 and 30 mL h-1, respectively. The reason is that the slow flow rate likely induces transverse
diffusion of the two streams, as observed in previous works [33]. The OCV at 40 mL h-1 is 1.76 V, less
than 1.82 V at 20 mL h-1, due to the serious mixing of the fuel and oxidant streams at the high flow
rate.
Int. J. Electrochem. Sci., Vol. 14, 2019 4566

Figure 7. Polarization curves (left Y-axis) and power density curves (right Y-axis) at different flow
rates with 50 mM KMnO4 containing 0.5 M H2SO4 as a supporting electrolyte.

Figure 8. Electrode potential curves at different flow rates with 50 mM KMnO4 containing 0.5 M
H2SO4 as a supporting electrolyte.

Table 4. OCVs, peak power densities and limiting current densities under the different flow rates

flow rates OCV Peak power Densities Limiting current densities

(mL h-1) (V) (mW cm-2) (mA cm-2)
10 1.65 1.49 2.70
20 1.82 4.06 5.85
30 1.83 4.51 6.67
40 1.76 3.69 5.22

As seen from the electrode potentials of Figure 8, at the flow rate of 10 mL h -1, the cathode is
under serious mass transfer limitation, because the reactants could not be supplied in time. At 30 mL h-
1
, the cell performance is the best owing to the improvement in the mass transfer with the increase in
Int. J. Electrochem. Sci., Vol. 14, 2019 4567

the flow rate. While at 40 mL h-1, the larger flow rate disturbs the laminar liquid-liquid separation
interface and mixes the oxidant and the fuel, leading to considerable decrease in the cathode potentials
and increase in the anode potentials.
To compare our cell performance with other reported MFCs with KMnO4 as the oxidant, the
peak power densities, limiting current densities of these MFCs are presented in Table 5. From Table 5,
we can see that except the MFC by López-Montesinos [19], the peak power density of our MFC is the
highest, suggesting the MFC with an alkaline sodium formate as the fuel and an acidic KMnO4 as the
oxidant is feasible for achieving a high performance. In addition, our MFC has an outstanding
advantage of a simple construction and no need of cathode catalyst. In the further study to enhance our
MFC performance, we can adopt other flowing patterns such as the sequential flow to improve the
mass transfer in the cell.

Table 5. Peak power densities and limiting current densities of the MFCs with KMnO4 as the oxidant

MFC Configuration or Electrode Cathode Peak power Limiting current

features Catalyst Densities densities
(mW cm-2) (mA cm-2)
Our Y-type No 4.51 6.67
Choban [26] Y-type Pt black 2.4 8
Sun [23] Three inlets Pt 0.7 2.6
Salloum [22] Sequential flow Pt black 2.8 5
López-Montesinos [24] Bridged shaped No 26 35
Jindal [25] Electrospun CNx nanofibers Pt and Au 3.43 9.79
as cathode

4. CONCLUSIONS

In this study, a Y-shaped MFC with two graphite plates as the electrodes was assembled and
studied. This device had advantages such as easy to be constructed and feasible to a visual experiment
for the laminar flowing in the microchannel. The microfluidic fuel cell adopted an alkaline sodium
formate as the fuel and an acidic KMnO4 as the oxidant. The effects of the oxidant concentrations, the
supporting catholyte concentrations and the flow rates on the MFC performance were systematically
investigated. Increasing the three parameters is observed to first increase but then decrease the cell
performance. The optimal condition to achieve the best cell performance is determined to be 50 mM
KMnO4, 0.5 M H2SO4 and a flow rate of 30 mL h-1. Under these conditions, the MFC displays the
maximal power density of 4.51 mW cm-2 and the largest limiting current density of 6.67 mA cm-2.
Further increasing the KMnO4 concentration to 100 mM lowers the cell performance due to the
formation of a large amount of insoluble MnO2 product from reduction of KMnO4 that clogs the
microchannel, and hinders the mass transfer of the reactants to the electrode and the products out from
the microchannel. At 1.0 M H2SO4, the cell performance declines, because more SO42- ions diffuse to
the anode that poison the Pd catalyst. When the flowing rate is more than 30 mL h -1, hydraulic
instability happens and destroys the laminar liquid-liquid interface, seriously mixing the two fluids and
decreasing the cell performance.
Int. J. Electrochem. Sci., Vol. 14, 2019 4568

In view of the insoluble MnO2 product from the reduction reaction of KMnO4, this MFC
construction might be improved by adding another inlet to introduce the third electrolyte to separate
the KMnO4 solution and HCOONa solution.

CONFLICTS OF INTEREST
There are no conflicts to declare.

ACKNOWLEDGEMENTS
This work is supported by the National Natural Science Foundation of China (No.51506046).

SUPPORTING MATERIAL:

S1 Photograph of a completed microfluidic fuel cell

Figure S1. Photograph of a completed microfluidic fuel cell

Figure S1 shows the actual configuration of the Y-type MFC. The two silicone tubes and one
silicone tube were glued by silica gel at the two inlets and the one outlet, respectively. The two end
plates were glued on the main body of our MFC.

S2 Experiment measurement system

Figure S2 shows the schematic configuration of the Y-type MFC measurement system adopting
the external reference electrode.
In this study, the polarization curves were gained by a chronoamperometry method using the
electrochemical workstation (Zahner, Germany) with a two-electrode configuration. Individual anodic
and cathodic potentials were obtained versus an external Ag/AgCl electrode in saturated KCl solution
as the reference electrode by in situ measuring the voltage difference between the electrode and the
reference electrode. The reference electrode was put near at the outlet of the MFC. This configuration
Int. J. Electrochem. Sci., Vol. 14, 2019 4569

was also used in other works [27, 34]. It is worth noting that the porous ceramics at the front end of the
reference electrode should be inserted into the electrolyte from the outlet of the MFC.

Figure S2. Schematic configuration of the Y-type MFC measurement system using the external reference electrode

References

1. C. K. Dyer, J. Power Sources, 106 (2002) 31-34.

2. R. Ferrigno, A. D. Stroock, T. D. Clark, M. Mayer and G. M. Whitesides, J. Am. Chem. Soc., 124
(2002) 12930-12931.
3. J. L. Cohen, D. A. Westly, A. Pechenik and H. D. Abruňa , J. Power Sources, 139 (2005) 96-105.
4. N. D. Mota, D. A. Finkelstein, J. D. Kirtland, C. A. Rodriguez, A. D. Stroock and H. D. Abruňa, J.
Am. Chem. Soc., 134 (2012) 6076-6079.
5. R. S. Jayashree, L.Gancs, E. R. Choban, A. Primak, D. Natarajan, L. J. Markoski and P. J. A.
Kenis, J. Am. Chem. Soc, 127 (2005) 16758-16759.
6. A. S. Hollinger, R. J. Maloney, R. S. Jayashree, D. Natarajan, L. J. Markoski and P. J. A. Kenis, J.
Power Sources, 195 (2010) 3523-3528.
7. J. An, Y. B. Kim, J. Park, T. M. Gür and F. B. Prinz, Nano Lett., 13 (2013) 4551-4555.
8. S. A. M. Shaegh, N. T. Nguyen, S. M. M. Ehteshami and S. H. Chan, Energy Environ. Sci., 5
(2012) 8225-8228.
9. E. Kjeang, B. T. Proctor, A. G. Brolo, D. A. Harrington, N. Djilali and D. Sinton, Electrochim.
Acta, 52 (2007) 4942-4946.
10. L. Li, G. Nikiforidis, M. K. H. Leung and W. A. Daoud, Appl. Energ., 177 (2016) 729-739.
11. E. Kjeang, R. Michel, D. A. Harrington, D. Sinton and N. Djilali, Electrochim. Acta, 54 (2008)
698-705.
12. Y. S. Li, H. Wu, Y. L. He, Y. Liu and L. Jin, J. Power Sources, 287 (2015) 75-80.
13. X. Cheng, R. Chen, X. Zhu, Q. Liao, X. He, S. Li and L. Li, Int. J. Hydrogen Energy, 41 (2016)
2457-2465.
14. L. An, T. S. Zhao, S. Y. Shen, Q. X. Wu and R. Chen, Int. J. Hydrogen Energy, 35 (2010) 4329-
4335.
15. X.W. Yu and A. Manthiram, Appl. Catal. B Environ., 165 (2015) 63-67.
Int. J. Electrochem. Sci., Vol. 14, 2019 4570

16. M. H. Chang, F. Chen and N. S. Fang, J. Power Sources, 159 (2006) 810-816.
17. E. R. Choban, P. Waszczuk, and P. J. A. Kenis, Solid-State Lett., 8 (2005) A348-A352.
18. F. L. Chen, M. H. Chang and C. W. Hsu, Electrochim. Acta, 52 (2007) 7270-7277.
19. E. Kjeang, R. Michel, D. A. Harrington, D. Sinton and N. Djilali, Electrochim. Acta, 54 (2008)
698-705.
20. M.-A. Goulet, M. Skyllas-Kazacos and E. Kjeang, Carbon, 101 (2016) 390-398.
21. M.-A. Goulet, O.A. Ibrahim, W. H. J. Kim and E. Kjeang, J. Power Sources, 339 (2017) 80-85.
22. K. S. Salloum, J. R. Hayes, C. A. Friesen and J. D. Posner, J. Power Sources, 180 (2008) 243-
252.
23. M. H. Sun, G. V. Casquillas, S. S. Guo, J. Shi, H. Ji, Q. Ouyang and Y. Chen, Microelectron. Eng.,
84 (2007) 1182-1185.
24. P. O. López-Montesinos, N. Yossakda, A. Schmidt, F. R. Brushett, W. E. Pelton and P. J. A. Kenis,
J. Power Sources, 196 (2011) 4638-4645.
25. A. Jindal, S. Basu, N. Chauhan, T. Ukai, D. S. Kumar and K. T. Samudhyatha, J. Power Sources,
342 (2017) 165-174.
26. E. R. Choban, L. J. Markoski, A. Wieckowski and P. J. A. Kenis, J. Power Sources, 128 (2004) 54-
60.
27. D. D. Ye, Y. Yang, J. Li , X. Zhu, Q. Liao and B. Zhang, Biomicrofluidics, 9 (2015) 064110-1-
064110-8.
28. D. D. Ye, Y. Yang, J. Li, X. Zhu, Q. Liao, B. W. Deng and R. Chen, Int. J. Hydrogen Energy, 38
(2013) 15710-15715.
29. B. Zhang, D. D. Ye, J. Li, X. Zhu and Q. Liao, J. Power Sources, 214 (2012) 277-284.
30. F. A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, 4th ed., John Wiley & Sons, 1980.
31. A. Li, S. H. Chan and N.-T. Nguyen, J. Micromech. Microeng., 17 (2007) 1107-1115.
32. R. S. Jayashree, M. Mitchell, D. Natarajan, L. J. Markoski and P. J. A. Kenis, Langmuir, 23 (2007)
6871-6874.
33. M.-A. Goulet and E. Kjeang, Electrochim. Acta, 140 (2014) 217-224.
34. X. Zhu, B. Zhang, D. D. Ye, J. Li and Q. Liao, J. Power Sources, 247 (2014) 346-353.

© 2019 The Authors. Published by ESG (www.electrochemsci.org). This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/4.0/).