i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e8

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/he

Carbon black / PTFE composite hydrophobic gas

diffusion layers for a water-absorbing porous
electrolyte electrolysis cell

Yuki Terayama a,*, Takamasa Haji b, Shoichi Furukawa b,

Munemitsu Nomura b, Masamichi Nishihara a,c,
Stephen Matthew Lyth a,d,e, Yoshitsugu Sone f,g, Hiroshige Matsumoto a
a
International Institute for Carbon-Neutral Energy Research (WPI-I2CNER), Kyushu University, 744 Motooka,
Nishi-ku, Fukuoka, 819-0395, Japan
b
Department of Hydrogen Energy Systems, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka, 819-0395, Japan
c
Next-Generation Fuel Cell Research Center (NEXT-FC), Kyushu University, 744 Motooka, Nishi-ku, Fukuoka,
819-0395, Japan
d
Platform of Inter/Transdisciplinary Energy Research (Q-PIT), Kyushu University, 744 Motooka, Nishi-ku, Fukuoka,
819-0395, Japan
e
Energy2050, Department of Mechanical Engineering, University of Sheffield, S10 2TN, UK
f
Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency, 3-1-1 Yoshinodai, Chuo-ku,
Sagamihara-shi, Kanagawa, 252-5210 Japan
g
SOKENDAI, 3-1-1 Yoshinodai, Chuo-ku, Sagamihara-shi, Kanagawa, 252-5210 Japan

article info abstract

Article history: The characteristics of a water-absorbing porous electrolyte electrolysis cell, in which
Received 13 October 2017 pressurized water is injected directly into the electrolyte layer, are investigated. High water
Received in revised form support force is required for the gas diffusion layer (GDL) in this novel cell design, and
14 November 2017 therefore here we report a new type of hydrophobic GDL comprising an acetylene black
Accepted 6 December 2017 (AB) and poly(tetrafluoroethylene) (PTFE) composite film. The method of preparation of the
Available online xxx AB/PTFE slurry, film formation methods, and the AB/PTFE weight ratio were investigated
and optimized. The ball-milling and transfer method were suitable for preparing uniform
Keywords: AB/PTFE slurry and successfully covering AB/PTFE film without any cracks on micro-porous
Water electrolysis layer coated carbon paper, respectively. An investigation about different PTFE weight ratios
Gas diffusion layer against AB from 0.1 to 6 showed a serious trade-off character between electrical resistance
Polymer composite film R, gas permeability V0 , and water support force Plim. The 1/2.5 of AB/PTFE weight ratio was
Proton conduction most optimal, which showed to have most equivalent R (2.5 U cm2), V0 (136 mL atm1 cm2
min1), and Plim (0.25 MPa). We also confirmed that fabricated GDL with optimal condition
was worked as the blocking layer against water injected through electrolyte layer and
pressurized by nitrogen gas, and as gas-permeation layer for generated hydrogen gas in
water electrolysis test.
© 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

* Corresponding author.
E-mail address: terayama@i2cner.kyushu-u.ac.jp (Y. Terayama).
https://doi.org/10.1016/j.ijhydene.2017.12.045
0360-3199/© 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Terayama Y, et al., Carbon black / PTFE composite hydrophobic gas diffusion layers for a water-
absorbing porous electrolyte electrolysis cell, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/
j.ijhydene.2017.12.045
2 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e8

electrolysis cell (Fig. 1 (b) [11]). This water-absorbing electrol-

Introduction
Hydrogen production is one of the key topics in terms of
bringing about a sustainable hydrogen-based society, and
ending our dependence on fossil fuels [1e3]. Over the past
thirty years, water electrolysis has become attractive as a
highly practical, relatively low cost, and convenient method to
produce hydrogen gas [4]. Ideally, renewable electrical power
generated from solar cells or wind turbines should be utilized.
Eventually, the hydrogen gas produced by water electrolysis is
expected to be reconverted to back electrical power by the use
of fuel cells [5]. Therefore, the development of water elec-
trolysis cells with higher performance is of crucial importance
for the future.
Polymer electrolyte water electrolysis (PEWE), using Nafion
membranes, is one of the most promising methods for
hydrogen production [6,7]. Fig. 1 (a) shows a schematic image
of conventional PEWE cell, which consist of dense polymer
electrolyte (e.g. Nafion), electrocatalyst layer (e.g. Platinum or
Iridium oxide inks), gas diffusion layer (e.g. carbon paper).
Water is supplied from outside of anode electrode to the
electrolyte layer, and then generated protons move to cathode
electrode during water electrolysis. However, PEWE cells still
have some practical problems, such as the poor stability of
Nafion at temperatures higher than 80  C [8]; dehydration of
the electrolyte layer [9]; and decomposition of carbon supports
at the oxygen-evolution electrode [10]. To resolve these issues,
recently developed a novel water-absorbing porous electrolyte
ysis cell consists of a hydrophilic inorganic porous electrolyte,
an electrocatalyst layer, and a hydrophobic gas diffusion layer
(GDL). We previously reported a water-absorbing porous
electrolyte electrolysis cell using sulfonated titanium oxide
particles as the electrolyte layer. We obtained a cell voltage of
3.0 V at 20 mA cm2, and a hydrogen gas evolution rate similar
to the theoretical rate calculated by Faraday's law [11].
Compared with conventional PEWE cells, this new type of
cell has several advantageous points. These include: (1) direct
injection of water into the porous electrolyte layer to avoid
dehydration of the layer during operation of water electrolysis
because of continuous supplying water through porous elec-
trolyte; (2) use of a water tank for water electrolysis with
applied pressure to produce pressurized hydrogen or oxygen
gas effectively because compressed hydrogen or oxygen gas
was effectively produced by controlling differential pressure
between water tank and collected gas tank; and (3) a possi-
bility to reduce concentration overvoltage in the electrode
layer because electrocatalyst layer was supplied a water from
porous electrolyte layer and separation of generated gas to
outside of the cell in hydrophobic electrocatalyst layer. In
general, the researches of high pressure PEM electrolyzer to
produce purified and compressed hydrogen gas were reported
[12], however these reported cells were difficult for compli-
cated cell designs and high cost. On the other hand, there is
large possibility to design simple structure in our suggested
water absorbing porous electrolyte electrolysis cell, which is
expected to be substitution from PEM electrolyzer cell in the

Fig. 1 e Schematic images of different water electrolysis cells: (a) a conventional PEWE cell consist of dense polymer
electrolyte layer (e.g. Nafion), electrocatalyst layer, and carbon paper type gas diffusion layer. Water was supplied from
outside of anode electrocatalyst layer. (b) Water-absorbing porous electrolyte cell consist of hydrophilic porous electrolyte
layer, controlled hydrophobic electrocatalyst layer, and fully hydrophobic gas diffusion layer. Water was injected for porous
electrolyte layer directly.

Please cite this article in press as: Terayama Y, et al., Carbon black / PTFE composite hydrophobic gas diffusion layers for a water-
absorbing porous electrolyte electrolysis cell, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/
j.ijhydene.2017.12.045
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e8 3

future. In order to fabricate a cell taking advantage of these

points, the gas diffusion layer (GDL) should have higher water
support force against the applied water pressure than con-
ventional cells, high hydrogen/oxygen gas permeability, and
low electrical resistance. Therefore, new types of hydrophobic
GDLs must be designed.
In previous studies outside the field of electrolysis, Shimizu
et al. reported a hydrophobic GDL prepared by pressing a
composite powder of acetylene black (AB) and poly(tetra-
fluoroethylene) (PTFE) on nickel mesh, followed by hot
pressing at 370  C. The cathodic and anodic electrode perfor-
mance/stability was evaluated in a metal-air battery [13].
Kitagawa et al. reported contact angle and water-vapor
permeability of PTFE/AB hydrophobic layers coated on com-
mercial carbon papers with microporous layers (MPLs) [14]. In
a recent patent, porous PTFE films were prepared by a rolling
method with controllable drawing ratios and directions,
which showed reasonable water support force and gas
permeability for gas filters [15]. Although these reported GDLs
showed appropriate performance for batteries, fuel cells, and
gas filters, the GDLs have not been applied in water-absorbing
Fig. 2 e Structure of a water-absorbing porous electrolyte
porous electrolyte electrolysis cell. Requirement of specifica-
electrolysis single cell. In this cell, porous electrolyte layer
tion for GDL of water absorbing porous electrolyte electrolysis
was sandwiched between smaller cathode and larger
cell was blocking against injected water by back pressure at
anode electrode (electrocatalyst layer and electrolyte layer).
1.0 MPa, higher gas permeability of generated hydrogen and
The water was directly injected and pressurized by back-
oxygen gas than 18 mL atm1 cm2 min1, and lower electrical
pressure. The generated gas was separated to hydrogen
resistance than 0.1 U cm2. We recognized that it was difficult
and oxygen gases in the electrocatalyst layers.
to achieve preparing a GDL with these satisfied properties
such as water support force, gas permeability, and electrical
resistance in our previous study [11]. The difficult point of
achieve it are optimization of these three parameter because density to produce a large amount of hydrogen and oxygen
of trade-off characteristic between conductive carbon and gases.
hydrophobic materials. Since typical materials such as carbon In order to realize the all required functions for a water-
black and PTFE have a different affinity for the solvent as absorbing porous electrolyte electrolysis cell, we developed a
dispersion media and each other, it resulted that there are bespoke GDL using AB and PTFE composite films. In order to
non-uniform state and many large cracks in GDL leading to achieve higher water support force, gas permeability, and
poor performances. In order to resolve the problem about sufficient electron conductivity, we focus on the uniformity of
non-uniformity and cracking of GDL, we designed a novel gas the AB and PTFE mixture using several mixing methods; on
diffusion layer using AB/PTFE mixing by ball-milling and film the substrates for making GDLs; on film formation methods;
forming by transfer method on MPL coated carbon paper as on the weight ratios of AB/PTFE; and on the hot-pressing
support substrate. The point of this study is that carbon paper conditions. Here, we report results of GDL preparation and
supported the film toughness and electrical conductivity of discuss the relationship between performance and prepara-
GDL, the control of trade-off character of GDL is expected to be tion methods. Finally, hydrogen production is carried out
simple and easy process, which is different compared with using the prepared GDLs in a water-absorbing porous elec-
other previous studies. trolyte electrolysis cell.
Our electrolysis cell concept is outlined in Fig. 2. Water is
directly injected into the porous electrolyte layer using a back
pressure of 1.0 MPa. This water permeates into the electro- Experimental
catalyst layers where the hydrogen evolution reaction (HER)
and oxygen evolution reaction (OER) process take place. The Materials
hydrophobic GDL provides electrical contact, allows the
generated gases to escape the cell, but also blocks water, Acetylene Black (AB, Denka Company Ltd.), PTFE dispersion
maintaining the high cell pressure and preventing leakage. (Polyflon PTFE-D, 60 wt% aqueous solution, Daikin Industries
This is why a higher water support force is required. In Ltd.), Triton-X (98%, Kishida Chemical Co., Ltd.), and ethanol
contrast, in conventional electrolysis cells (Fig. 1 (a)), water is (99.8%, Kishida Chemical Co., Ltd.) were used as received.
injected from outside of anode side electrode to supply the Distilled water (15.0 MUcm) was purified with Pure Water
water for electrolyte layer and cathode electrode. In this Equipment Elix Advantage3 (Merck Millipore Co.). Commercial
concept of conventional electrolysis cell, the high water carbon paper (TGP-H-090, Toray Industries, Inc.) and micro-
support force is not required but the dehydration of elec- porous layer (MPL)-coated carbon paper (GDL29BC, Chemix
trolyte layer take place in case of operation with high current Co. Ltd.) were purchased from Chemix Co. Ltd.

Please cite this article in press as: Terayama Y, et al., Carbon black / PTFE composite hydrophobic gas diffusion layers for a water-
absorbing porous electrolyte electrolysis cell, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/
j.ijhydene.2017.12.045
4 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e8
Preparation and transfer of AB/PTFE slurries
2.0 g of AB powder was added to 1.0 g of Triton-X, 95 mL of
distilled water, and 5 mL of ethanol. This dispersion was then
stirred for 15 min at room temperature, and ultra-sonicated
for 15 min (Sonic Technology Co. Ltd., GSD150AT). 50 g of
the resulting mixture and 50 g of f2.0 mm YTZ balls (Nikkato
Co. Ltd.) were added into 100 mL a plastic bottle with a cap and
milled using a rolling machine for 1 day at room temperature,
to crush the aggregated AB powders. Different amounts of
PTFE dispersion in aqueous solution were added into the
milled AB mixture in the bottle, and then again milled for
60 min. The milled mixture solution of AB and PTFE was
filtered using a cellulose acetate filter membrane with 5.0 mm
average pore size (Sartrious Japan Co. Ltd, Minisart 17594 K)
and a plastic syringe to remove the aggregated parts of the
mixture completely.
The obtained AB/PTFE slurry was uniformly sprayed onto
the un-glazed surface of aluminum foil (Fig. 3), using an air
brush (Co. Tamiya Inc., SPRAY-WORK HG AIRBRUSH III). In the
case of e.g. aluminum foil, this was placed on a hotplate at
200  C, and then in a vacuum oven at 200  C for 15 min to
evaporate the volatile solvents and remove residual Triton-X.
The AB/PTFE film was then transferred from the aluminum
foil to the MPL-coated carbon paper by hot-pressing at
240 kg cm2 and at 360  C for 3 min. After cooling at room
temperature for 3 min, the AB/PTFE-coated carbon paper was
immersed into 6 wt% HCl aqueous solution to completely
remove aluminum foil from the surface. Finally, the resulting
GDL was washed by distilled water several times and dried at
room temperature.
Evaluation of GDLs
The morphology was observed by scanning electron micro-
scope (SEM, Topcon sm-350). Electrical resistance (R, U cm2)
was evaluated perpendicular to the plane using a galvanostat
(Hokuto Denko Co., Potentiosta/Galvanostat HA-151B) to apply
a constant DC current (Fig. 4 (a)). The gas permeability was
measured using a home-made test cell (Fig. 4 (b)) giving the
pressure loss from 0.01 MPa to 0.25 MPa using Ar gas with a
precisely controllable regulator. The Ar gas permeability was
measured using a soap-film flow mater (HORIBA STEC Ltd.,
Film Flow Meter VP-2). The normalized gas permeability rate
V’ (mL$atm1$cm2$min1) was calculated from the obtained
values V (mL$min1) as follows:
V
V' ¼
A,P
where A is the contact area between the GDL and Ar gas (cm2),
and P is the applied pressure loss (atm).
Fig. 3 e Preparation of AB/PTFE GDL composite films on MPL-coated carbon paper.
Please cite this article in press as: Terayama Y, et al., Carbon black / PTFE composite hydrophobic gas diffusion layers for a water-
absorbing porous electrolyte electrolysis cell, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/
j.ijhydene.2017.12.045
The water permeability was also measured using a home-
made test cell (Fig. 4 (c)). Purified water was injected into the
cell and pressure was applied using nitrogen gas, controlled by
a regulator from 0 to 1.0 MPa. The applied pressure just before
water leakage was defined as the limiting pressure Plim (MPa).
Water electrolysis test
Membrane electrode assemblies (MEAs) for water electrolysis
were prepared by the followed procedure. GDL disks (AB/PTFE
films on MPL-coated carbon paper) were prepared with di-
ameters of f12 and f23 mm. Pt and IrO2 (Chemix Co. Ltd)
electrocatalyst layers were sprayed onto the GDLs on a hot-
plate, at 70  C. The mass loadings of Pt and IrO2 were 0.5 and
1.5 mg cm2, respectively. Nafion membrane (NR-211, DuPont)
was sandwiched between the two electrocatalyst-coated
GDLs, and the assembly was then hot-pressed at 100 kg cm2
and 140  C, for 1 min.
Current-voltage (i-V) characteristic and hydrogen gas evo-
lution rate for water electrolysis were then evaluated. Elec-
trical voltage was measured by galvanostat (Hokuto Denko
Co., Potentiosta/Galvanostat HA-151B) in constant DC current
mode. The amount of generated hydrogen gas was evaluated
by gas chromatography (Agilent, 3000 A Micro GC, column:
Molsieve, 14 m  320 mm  12 mm, 20 mL min1 Ar gas flow),
and were determined by calculation using a calibration curve
of 1 % H2/Ar gas. The theoretical evolution rates were calcu-
lated using Faraday's law.
Results and discussion
Evaluation of AB/PTFE mixing and deposition method
AB/PTFE slurries were first prepared by several different
mixing methods. These were: magnetic stirring, ultra-
sonication, and ball-milling. In case of magnetic stirring,
many aggregates were formed in the mixture, and therefore
this method was determined to be unsuitable. After ultra-
sonication, the extent of aggregation was decreased, but the
mixture was not sufficiently homogeneous for spraying using
the air brush. This indicates that simple stirring and ultra-
sonication are not sufficient to break the strong interaction
between AB particles in aqueous solution. Therefore, ball
milling using zirconia beads was used, resulting in a uniform
mixture without significant AB aggregate content. After well-
dispersed AB slurries were obtained, PTFE dispersion was
(1) added and a second ball-milling step was performed. After
filtration of this mixture, a homogeneous black slurry without
aggregates was obtained. Therefore, ball-milling was adopted
as the main method to prepare our AB/PTFE slurries.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e8 5

Fig. 4 e Setup of (a) electrical resistance, (b) gas permeability, and (c) water support force measurements.

AB/PTFE GDLs were prepared on bare carbon paper, or
MPL-coated carbon paper, either by direct spraying, or by the
film transfer method (Table 1, Method 1e4). In Method 1e2,
the AB/PTFE slurry was sprayed directly onto the carbon
paper, or MPL-coated carbon paper using hot-plate at 200  C,
and the water support force was evaluated. These direct-
sprayed GDLs showed no water support force. SEM images
revealed that these samples had large cracks on the surface
(Fig. 5 (c) and (d). These cracks are probably formed due to
the large pores that exist between the fibers of the carbon
paper, which disrupt the uniform formation of the sprayed
slurry and allow water to penetrate. In the alternative
method, AB/PTFE slurries were sprayed onto aluminum foils
and then transferred onto the carbon paper by hot pressing
(Method 3e4). When the AB/PTFE film was transferred onto
bare carbon paper (Method 3), the water support force was
still negligible due to the small cracks on the surface (Fig. 5
(e)). However, when the AB/PTFE film was transferred from
aluminum foil to the MPL-coated carbon paper (Method 4),
the water support force was as high as 0.3 MPa. SEM imaging
revealed that cracks were not observed in these final trans-
ferred films (Fig. 5 (f)). This is most likely because the MPL
acts as an intermediate layer between the carbon paper and
the AB/PTFE composite film, resulting in a less damaging
transfer process. From these results, it is concluded that a
suitable method to prepare AB/PTFE composite films is to use
ball-milling during slurry preparation, spraying onto
aluminum foil to prepare uniform films, and transfer onto
MPL-coated carbon paper.
Effect of hot-pressing temperature on AB/PTFE performance
Hot-pressing temperature strongly influences the morphology
and performance of composite GDLs. AB/PTFE films trans-
ferred to the GDLs without hot-pressing was weak, and broke
easily during evaluation. Therefore, melting of the PTFE was
deemed necessary to improve the water support force. Fig. 5 (f,
g, and h) shows SEM images of AB/PTFE composite films
transferred onto MPL-coated carbon papers, then hot-pressed
at 280, 320, and 360  C. Before hot-pressing (Fig. 5, (b)), large
>50 mm cracks were observed, which are expected to
contribute to lower water support force. The transfer method
was therefore used to suppress the formation of these cracks.
After hot-pressing at 320 and 280  C (Fig. 5, (g) and (h)), cracks
were still observed, although the size was smaller than after
the direct spraying method (Fig. 5 (c and d)) or transfer method
on bare carbon paper (Fig. 5 (e). These films also displayed very
low water support force of less than 0.01 MPa. On the other
hand, no cracks were observed after hot-pressing at 360  C
(Fig. 5 (f)), and water leakage was not observed at 0.3 MPa
applied water pressure. The improved morphology is a result
of heating the layer to a temperature higher than the melting
point of PTFE (327  C) [16]. At lower hot-pressing tempera-
tures, the PTFE retains its particulate morphology, with sub-
micron diameter, and the cracks in the MPL-coated carbon
paper cannot be completely covered by the AB/PTFE com-
posite film. By hot-pressing at 360  C, the melted PTFE com-
posite can flow into, and seal the cracks under high pressure,
resulting in a more uniform film. At temperatures higher than
360  C, PTFE decomposes to generate harmful gases [17].
These results suggest that 360  C is a suitable temperature for
hot-pressing of the AB/PTFE composite films on MPL-coated
carbon paper substrates.
Effect of AB/PTFE weight ratio on performance
Fig. 6 shows electrical resistance, R, versus AB/PTFE weight
ratio of the prepared GDLs. The resistance of the GDLs with
PTFE low weight ratio from 0.1 to 2.0 was approximately
constant at 1.3e2.0 U cm2. In this case, the high ratio of
electronically conducting AB carbon particles are physically
connected throughout the film, resulting in low electrical
resistance. On the other hand, at higher PTFE ratios over 2.5,
the resistance dramatically increases, as the connectivity be-
tween AB particles is disrupted by insulating PTFE. The R
values of prepared GDLs are relatively higher than 0.1 U cm2

Table 1 e Preparation of GDLs with AB/PTFE using different substrates and film formation methods.
Method Substrate Slurry AB/PTFE(wt/wt) Film formation Appearance Water support force
1 Bare Carbon Paper 1/6 Spraying Many Large Cracks 0.0 MPa
2 MPL-Coated Carbon Paper 1/6 Spraying Many LargeCracks 0.0 MPa
3 Bare Carbon Paper 1/6 Transfer Many Small Cracks 0.0 MPa
4 MPL-Coated Carbon Paper 1/6 Transfer No Cracks 0.3 MPa

Please cite this article in press as: Terayama Y, et al., Carbon black / PTFE composite hydrophobic gas diffusion layers for a water-
absorbing porous electrolyte electrolysis cell, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/
j.ijhydene.2017.12.045
6 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e8

Fig. 5 e SEM images of: (a) bare carbon paper before transfer method, (b) MPL-coated carbon paper before transfer method,
After preparation by spray method on (c) bare carbon paper and (d) MPL-coated carbon paper, and after preparation by
transfer method on (e) bare carbon paper, and MPL-coated carbon paper with different hot-press temperature at (f) 280  C, (g)
320  C and (h) 360  C.

of targeted value for water absorbing porous electrolyte elec-
trolysis cell. We supposed that uniformity of AB/PTFE was not
so enough and have to be improved using other mixing
method or additive agent. The ball-milling with stronger
power and additive fluorinated polymer with low melting
point might be required for more uniform state, and will tried
to be investigated in the future.
Fig. 7 shows the effect of AB/PTFE ratio on gas permeability,
V’. The films with low PTFE ratio show approximately con-
stant gas permeability in the range 500e560 mL atm1 cm2
min1. On the other hand, the gas permeability dramatically
decreases at intermediate PTFE ratios (from 1.0 to 3.0). Above a
PTFE ratio of 3.0, the gas permeability is less than
10 mL atm1 cm2 min1. The AB/PTFE films with high PTFE
content contain dense polymer layers, which prevent gas
permeation through the GDL. Therefore, the AB/PTFE content
should be carefully controlled when designing GDLs which
require high gas permeability.
Fig. 8 shows the maximum water pressure, Plim, of the AB/
PTFE films. In case of films with low PTFE ratios (from 0.1 to
2.0), the maximum water pressure has approximately con-
stant values in the range 0.02e0.03 MPa. As the PTFE ratio
increases (from 2.5 to 6.0), the Plim value also increases up to a
much larger value of 0.30 MPa. This result indicates that the
PTFE concentration strongly affects the water support force of
the GDLs. Therefore, GDL films with higher PTFE ratio are
required for suitable water barrier properties at high water
pressure.

Please cite this article in press as: Terayama Y, et al., Carbon black / PTFE composite hydrophobic gas diffusion layers for a water-
absorbing porous electrolyte electrolysis cell, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/
j.ijhydene.2017.12.045
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e8 7

Fig. 8 e Water permeability, Plim, versus PTFE weight ratio

Fig. 6 e Electrical resistance, R, versus PTFE weight ratio in
in AB/PTFE films transferred onto MPL-coated carbon
AB/PTFE films transferred onto MPL-coated carbon paper.
paper.

The results described here clearly indicate that there is a
serious trade-off between electrical resistance, gas perme-
ability, and water support force for optimal GDL performance.
On the basis of these results, we determined an optimized
GDL composition of AB/PTFE ¼ 1/2.5. This GDL has a most
equivalent performance such as low electrical resistance of
2.5 U cm2, a suitable gas permeability of 136 mL atm1 cm2
min1, and a water support force of 0.25 MPa.
Water-absorbing porous electrolyte electrolysis cell using the
optimized AB/PTFE GDL
Fig. 9 shows the i-V curve and hydrogen evolution rate ob-
tained from the water-absorbing porous electrolyte electrol-
ysis cell at 25  C, with the optimized AB/PTFE hydrophobic
MEA. In this experiment, the water in the cell was pressurized
by N2 gas (Fig. 2) and was successfully distributed to the sur-
face of the prepared GDL through the porous electrolyte
membrane. In the low current density region (~30 mA cm2),
the voltage obtained from the water electrolysis test increased
with increasing current density from 1.65 to 1.76 V. At higher
current density (>30 mA cm2), the increase of the voltage was
smaller, and the voltage finally reached 1.97 V at 175 mA cm2.
This result indicates that the obtained performance from our
water electrolysis system is reasonable compared with our
previous PEWE system [11]. However, a voltage of 1.97 V at
175 mA/cm2 is significantly higher than that of previously
reported PEWE system (1.7 V at 200 mA cm2) [18]. This is due
to the much higher resistance of AB/PTFE GDLs prepared in
this work (2.5 U cm2) which is more than 100 times higher
than that of the un-coated carbon paper used previously [19].
The hydrogen gas evolution rate in Fig. 9 is about 70% of the
theoretical value calculated from Faraday's Law. Therefore,
~30 % of the generated hydrogen gas did not diffuse

Fig. 9 e i-V characteristics and H2 gas evolution rate for the

water electrolysis using the optimized AB/PTFE MEA at 25

C, DP ¼ 0.10 MPa. The AB/PTFE weight ratio was 1/2.5.
Closed squares (-), open triangles (△), and dotted lines
Fig. 7 e Gas permeability, V′, versus PTFE weight ratio in are the i-V characteristic, experimental H2 evolution rate,
AB/PTFE films transferred onto MPL-coated carbon paper. and theoretical H2 evolution rates, respectively.

Please cite this article in press as: Terayama Y, et al., Carbon black / PTFE composite hydrophobic gas diffusion layers for a water-
absorbing porous electrolyte electrolysis cell, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/
j.ijhydene.2017.12.045
8 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e8
conventionally out of the cell to be evaluated by gas chro-
matography. This indicates that some of the generated
hydrogen diffused into the water chamber through the elec-
trolyte, or that the hydrogen reacts with the generated oxygen
gas near the anode. These problems suggest that the GDL layer
and/or electrocatalyst layer is not sufficiently hydrophobic.
Therefore, GDLs and electrocatalyst layers with optimal hy-
drophobicity for this water-absorbing porous electrolyte
electrolysis cell will be developed in the near future.
Conclusions
We reported the preparation of a new hydrophobic GDL for a
water-absorbing porous electrolyte electrolysis cell. Several
key properties of the GDL such as electrical resistance, gas
permeability, and water support force were evaluated, and
optimized conditions for GDL fabrication and composition
were determined. To fabricate GDLs with high performance
requires homogeneous mixing of electron-conducting AB and
insulating PTFE, which was achieved by preparing the slurry
in aqueous solution with Triton-X surfactant, and by using
ball-milling. Transfer method using MPL coated carbon paper
was most suitable way to ensure highly uniform and crack-
free AB/PTFE films, which were formed on aluminum foil by
spray deposition, then transferred onto MPL-coated carbon
paper by hot-pressing at 360  C. From evaluation of GDL with
different AB/PTFE weight ratio from 1/0.1 to 1/6 using gas
permeability, electrical resistance, and water permeability
tests, the optimal composition of AB/PTFE was determined to
be 1/2.5, which showed most equivalent performances such
as low electrical resistance of 2.5 U cm2, a suitable gas
permeability of 136 mL atm1 cm2 min1, and a water sup-
port force of 0.25 MPa. We also confirmed that MEAs fabricated
using these AB/PTFE GDLs worked a blocking layer against
injected water by back-pressure and gas permeation layer for
generated hydrogen gas effectively in a water-absorbing
porous electrolyte electrolysis system. However, at present,
relatively high voltage (1.97 V at 175 mA cm1) and lower
hydrogen evolution rate than 70% were obtained compared
with theoretical values calculated from Faraday law, due to
insufficient hydrophobicity. Therefore, in future work we will
focus on this point.
Acknowledgements
This work was supported by CREST (Creation of Innovative
Core Technology for Manufacture and Use of Energy Carriers
from Renewable Energy, No. JPMJCR1442), JST, Japan.
references
[1] Nahar G, Mote D, Dupont V. Hydrogen production from
reforming of biogas: review of technological advances and an
Indian perspective. Renew Sustain Energ Rev
2017;76:1032e52. https://doi.org/10.1016/j.rser.2017.02.031.
[2] Dutta S. A review on production, storage of hydrogen and its
utilization as an energy resource. J Ind Eng Chem
2014;20:1148e56. https://doi.org/10.1016/j.jiec.2013.07.037.
Please cite this article in press as: Terayama Y, et al., Carbon black / PTFE composite hydrophobic gas diffusion layers for a water-
absorbing porous electrolyte electrolysis cell, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/
j.ijhydene.2017.12.045
[3] Valente A, Iribarren D, Dufour J. Life cycle assessment of
hydrogen energy systems: a review of methodological
choices. Int J Life Cycle Assess 2017;22:346e63. https://
doi.org/10.1007/s11367-016-1156-z.
[4] Yilmaz F, Balta MT, Selbas‚ R. A review of solar based
hydrogen production methods. Renew Sustain Energ Rev
2016;56:171e8. https://doi.org/10.1016/j.rser.2015.11.060.
[5] Schmittinger W, Vahidi A. A review of the main parameters
influencing long-term performance and durability of PEM
fuel cells. J Power Sources 2008;180:1e14. https://doi.org/
10.1016/j.jpowsour.2008.01.070.
[6] Babic U, Suermann M, Büchi FN, Gubler L, Schmidt TJ. Critical
reviewdidentifying critical gaps for polymer electrolyte
water electrolysis development. J Electrochem Soc
2017;164:F387e99. https://doi.org/10.1149/2.1441704jes.
[7] Fouda-Onana F, Chandesris M, Me deau V. Investigation on
the degradation of MEAs for PEM water electrolysers part I:
effects of testing conditions on MEA performances and
membrane properties. Int J Hydrog Energ 2016;41:16627e36.
https://doi.org/10.1016/j.ijhydene.2016.07.125.
[8] Washida S, Hirano M, Hisatome N, Shimizu K. Hydrogen
production by solid polymer membrane water electrolysis
under high-temperature and high-pressure. J Jpn Inst Energ
2002;81(5):322e7. https://doi.org/10.3775/jie.81.322.
[9] Gon~ i-Urtiaga A, Presvytes D, Scott K. Solid acids as
electrolyte materials for proton exchange membrane (PEM)
electrolysis: review. Int J Hydrog Energ 2012;37:3358e72.
https://doi.org/10.1016/j.ijhydene.2011.09.152.
[10] Gonza  lez-Huerta RG, Ramos-Sa  nchez G, Balbuena PB.
Oxygen evolution in Co-doped RuO2 and IrO2: experimental
and theoretical insights to diminish electrolysis
overpotential. J Power Sources 2014;268:69e76. https://
doi.org/10.1016/j.jpowsour.2014.06.
[11] Kim S-J, Sakai T, Oda H, Hamaguchi J-I, Okuyama Y,
Matsuka M, et al. Water electrolysis using water-absorbing
porous electrolyte consisting of a sulfonated nanotitania
proton conductor. Electrochemistry 2012;80(4):246e8. https://
doi.org/10.5796/electrochemistry.80.246.
[12] Harada H et.al. US Patent No. US 7,048,839 B2.
[13] Shimizu Y, Matsuda H, Miura N, Noboru Yamazoe N. Bi-
functional oxygen electrode using large surface area. J
Electrochem Soc 1990;137:3430e3. https://doi.org/10.1149/
1.2086234.
[14] Kitahara T, Nakajima H, Inamoto M, Morishita M. Novel
hydrophilic and hydrophobic double microporous layer
coated gas diffusion layer to enhance performance of
polymer electrolyte fuel cells under both low and high
humidity. J Power Sources 2013;234:129e38. https://doi.org/
10.1016/j.jpowsour.2013.01.150.
[15] Ishii K, Ayusawa S etal. Japan patent Kokai 2013-253214 a (19,
December, 2013).
[16] Kindl P, Sormann H. Investigations on the phase-transition
of PTFE at the melting-point by positron-annihilation.
Physica Status Solidi A Appl Res 1981;66(2):627e33. https://
doi.org/10.1002/pssa.2210660227.
[17] Kagiya T, Yokoyama N, Ueno T. Pyrolysis of r-ray irradiated
poly(tetrafluorethylene) powder. Nihon Kagaku Kaishi
1974;1974(9):1777e81. https://doi.org/10.1246/
nikkashi.1974.1777.
[18] Millet P, Alleau T, Durand R. Characterization of membrane-
electrode assemblies for solid polymer electrolyte water
electrolysis. J Appl Electrochem 1993;23:322e31. https://
doi.org/10.1007/BF00296687.
[19] Li H, Fujigaya T, Nakajima H, Inaba A, Ito K. Optimum
structural properties for an anode current collector used in a
polymer electrolyte membrane water electrolyzer operated
at the boiling point of water. J Power Sources 2016;332:16e23.
https://doi.org/10.1016/j.jpowsour.2016.09.086.