Improved deuterium separation factor for the iron electrode prepared in a magnetic field

Article abstract of DOI:10.1016/j.electacta.2004.04.018

The electrolytic deuterium separation factors were measured in KOH aqueous solution at 298K for the two iron films electrodeposited in no magnetic field (film (a)) and a magnetic field of 0.5T (film (b)). The separation factors were higher for film (b) (9.8-12.3) than those for film (a) (8.7-10.2). From the exchange current density and the real surface area, it was found that film (b) has higher electrocatalytic activity for the hydrogen evolution reaction. The SEM images showed cracks on the whole surface for film (b), which suggests that the strain energy is stored. The improved electrocatalytic activity for the HER is explained by the preferred adsorption of proton due to the strain energy in the film. The increase of the deuterium separation factor is also explained by the electrocatalysis activated by the strain energy.

Full text of DOI:10.1016/j.electacta.2004.04.018

Electrochimica Acta 49 (2004) 4181–4187
Improved deuterium separation factor for the iron
electrode prepared in a magnetic field
H. Matsushima, T. Nohira1, Y. Ito∗,2
Department of Fundamental Energy Science, Graduate School of Energy Science, Kyoto University, Sakyo-ku, Kyoto 606-8501, Japan
Received 20 December 2003; received in revised form 18 February 2004; accepted 11 April 2004
Available online 28 May 2004
Abstract
The electrolytic deuterium separation factors were measured in KOH aqueous solution at 298 K for the two iron films electrodeposited in
no magnetic field (film (a)) and a magnetic field of 0.5 T (film (b)). The separation factors were higher for film (b) (9.8–12.3) than those for
film (a) (8.7–10.2). From the exchange current density and the real surface area, it was found that film (b) has higher electrocatalytic activity
for the hydrogen evolution reaction. The SEM images showed cracks on the whole surface for film (b), which suggests that the strain energy
is stored. The improved electrocatalytic activity for the HER is explained by the preferred adsorption of proton due to the strain energy in the
film. The increase of the deuterium separation factor is also explained by the electrocatalysis activated by the strain energy.
© 2004 Elsevier Ltd. All rights reserved.
Keywords: Deuterium separation; Water electrolysis; Real surface area; Magnetic field; Strain energy
1. Introduction
ported that the electrodeposited iron had the highest value
around 10.
Heavy water (D2O) is indispensable as a moderator
in fission reactors and together with T (tritium) for the
fusion reaction and for the heavy water nuclear reactor.
There are several methods to produce heavy water commer-
cially, e.g. hydrogen sulphide–water exchange process [1],
ammonia–hydrogen exchange process [1] and water elec-
trolysis process [2]. Among them, water electrolysis has the
highest separation factor, therefore, it plays a crucial role in
the final separation stage. It is, therefore, very important to
develop a cathode material having a high separation factor.
In the practical electrolysis plant, mild steel has been em-
ployed as a cathode material for deuterium separation. Its
separation factor has been reported to be within the range of
4–7 in 28 wt.% KOH solution [3]. Pure iron is known to be
one of the superior materials having high separation factors
[4–7]. Brun et al. [6] investigated the electrolytic separation
factor at various cathode materials in 10 wt.% KOH and re-
Bockris et al. [8–10] theoretically investigated the deu-
terium separation factor using quantum mechanics. The
general explanation taken in the literatures is that the deu-
terium separation factor depends largely upon two factors.
One is tunnel effect of the hydrogen isotope. The other is
a difference in the zero point energy of the O–H and O–D
adsorbed on the metal, which provides an extra effect to the
tunnel effect. However, such quantum mechanical approach
seems to be possible only for well-clarified electrodes like
mercury and single crystals of noble and transition met-
als. It is very difficult to conduct the quantum mechanical
approach on the most of the practical systems, e.g. poly-
crystalline, various surface microstructure, oxide film and
metal alloy. On the other hand, it is well known that the
deuterium separation factor is generally high for the elec-
trodes having a high activity for the hydrogen evolution
reaction (HER) [11,12]. Therefore, the separation factor is
expected to be improved even for the same kind of met-
als by increasing the electrocatalysis for the HER. In fact,
Rowland [13] measured the separation factors for the iron
electrodes with different surface conditions in 2.0 M KOH
solution and reported that the separation factor for the bare
iron was higher than that for the iron covered with an oxide
∗
Corresponding author. Tel.: +81-7575-35827; fax: +81-7575-35906.
E-mail addresses: nohira@energy.kyoto-u.ac.jp (T. Nohira),
y-ito@energy.kyoto-u.ac.jp (Y. Ito).
1
Co-corresponding author.
ISE member.
2
0013-4686/$ – see front matter © 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2004.04.018

4182
H. Matsushima et al. / Electrochimica Acta 49 (2004) 4181–4187
film. The cause of this difference was explained by the
higher electrocatalytic activity of the bare iron for the HER.
On the other hand, several factors have been reported for
increasing the HER electrocatalysis for the same kind of
metals. For example, Doyle et al. [14] studied the effect of
microstructure of nickel electrode on the HER electrocataly-
sis in 0.1 M NaOH solution and reported that characteristics
of grain boundary and nanocrystalline defects considerably
influenced the electrocatalytic activity. Ohmori [15] inves-
tigated the HER in 0.1 M NaOH by using nickel electrodes
annealed in various temperatures and reported that the elec-
trocatalytic activity depended on the strain energy. This
was explained by the occurrence of alkaline metal deposi-
tion which adsorbed on the electrode active site instead of
hydrogen atom.
On the basis of the above reported results and consider-
ations, the authors considered that a new electrode mate-
rial having a high separation factor would be obtained by
controlling the microstructure and the strain for the elec-
trodeposited iron film. In our previous study, iron films
were electrodeposited in uniform magnetic fields (0.5–5 T)
and no magnetic field [16], and remarkable difference was
found concerning the texture. That is, biaxial texture evo-
lution occurred in magnetic fields (>0.5 T) while uniaxial
texture evolution proceeded in no magnetic field. It is easy
to estimate that characteristics of the grain boundaries
are different from each other for the two iron films. In
addition, larger strain should be left for the film electrode-
posited in a magnetic field, because iron is a ferromagnetic
metal.
manent magnet (NEOMAX, Sumitomo Special Metals) was
superimposed parallel to the electrode planes.
The texture of the electrodeposited iron film was measured
by XRD using Cu K␣ line (X’Pert, Philips, 50 kV, 40 mA).
The morphology was observed by SEM (S-2600H, Hitachi).
2.2. Measurement of polarization curves for the HER and
the real surface areas
Fig. 1 shows the electrolytic cell for water electrolysis.
The measurement was conducted in 1.0 mol l−1 KOH aque-
ous solution at 298 K. The catholyte was separated from the
anolyte by a membrane filter. The total volume of the elec-
trolyte was 150 cm3. The cathode was the electrodeposited
iron film (10 mm×5 mm×0.02 mm). The anode was a sheet
of platinum (20 mm × 20 mm × 0.1 mm, Pt 99.9%, Nilaco).
A reference electrode was the reversible hydrogen electrode
(RHE).
The real surface area of the iron electrode was deter-
mined by ac impedance and initial potential–relaxation
methods. The impedance data were acquired by frequency
response analyzer (SI 1260, Solartron). The frequency range
was from 10 kHz to 0.1 Hz. The impedance spectra were
fitted by ZPLOT software (Scribner Associates Inc). The
potential–relaxation curves were recorded by digital oscil-
loscope (NR-2000, Keyence) following the interruption of
various cathodic current densities. The measurement time
range was 10−6 s to 10−1 s. The non-linear potential curves
were fitted by KaleidaGraph software (Synergy Software).
In the present paper, the authors actually measured the
deuterium separation factor for the iron films electrode-
posited in a magnetic field and no magnetic field. The iron
films were characterized from the viewpoints of the surface
morphology, the texture, the kinetic parameters for the HER
and the real surface area. Based on these results, the cause
of the difference in the separation factor was discussed con-
sidering the grain boundary energy and the strain energy.
2.3. Measurement of deuterium separation factors
The same electrolytic cell shown in Fig. 1 was used for
the deuterium separation. The electrolyte was 1.0 mol l−1
KOH with 10 at.% deuterium adjusted by heavy water (D2O
99.9%, Cambridge Isotope Laboratories). The deuterium
separation was conducted galvanostatically at apparent
current densities from 50 to 400 mA cm−2. The evolved
2. Experimental
2.1. Preparation of iron electrodes
The electrodeposition apparatus and method have been
described in our previous report [16]. Thus, only several
important points are mentioned here. The cathode was a
sheet of copper (10 mm × 10 mm × 0.2 mm, Cu 99.99%,
Nilaco). The anode was a sheet of iron (10 mm × 10 mm ×
0.2 mm, Fe 99.99%, Nilaco). An electrolyte composition
was 0.90 mol l−1 FeSO4·7H2O, 0.15 mol l−1 FeCl2·4H2O
and 0.43 mol l−1 NH4Cl. The pH was adjusted to 1.5 with
H2SO4. The solution temperature was maintained at 298 K.
The iron electrodeposition was conducted galvanostatically
at 10 mA cm−2 until the amount of electrical charge reached
150 C cm−2. A uniform magnetic field (0.5 T) by the per-
Fig. 1. Schematic electrolytic cell for water electrolysis. A: Sampling
bottle; B: cathode (electrodeposited Fe, 1 cm2); C: Luggin probe; D: anode
(Pt, 8 cm2); E: membrane; F: liquid junction; G: reference; H: Pt black.

H. Matsushima et al. / Electrochimica Acta 49 (2004) 4181–4187
4183
hydrogen gas was stored in a sampling bottle. The gath-
ered gas was flowed through the silica gel to remove water
impurity from the hydrogen gas. A quadrupole mass spec-
trometer (AQA-100 MPX, ANELVA) was used to measure
the ratio of deuterium to protium.
iron grains was irregular and angular. In film (b), the surface
was roundish and smooth. However, there were no large
differences in grain size and surface roughness between the
two iron films (Fig. 3).
The metallographic structure was also investigated by the
X-ray pole figure measurement. There was a remarkable dif-
ference in the film texture. Fig. 4 shows the pole figures of
iron (1 1 0) plane. The magnetic field was applied parallel to
the substrate surface and its direction was indicated by an
arrow in the figure. In film (a), there was an annular sym-
metric pattern at the angle of 30◦ to the direction normal to
the substrate plane. The angle of 30◦ indicated that ꢀ2 1 1ꢁ
plane was preferentially oriented vertical to the film growth
direction (electric field direction). The circle pattern indi-
cated that (1 1 0) planes faced arbitrary directions randomly.
That is, uniaxial texture evolution along (2 1 1) axis occurred
in no magnetic field. On the other hand, in film (b), there
was an obvious crystal orientation in same direction of the
3. Results and discussion
3.1. Structure of iron electrodes
In the present paper, two types of the iron film electrodes
were prepared to study the magnetic field effect on elec-
trode characteristics: film (a), iron film electrodeposited in
no magnetic field (0 T) and film (b), iron film electrode-
posited in a magnetic field (0.5 T).
The surface morphology was observed by SEM, as shown
in Fig. 2. In film (a), the surface was uniformly covered
with iron grains and no crack was observed. In film (b), the
cracks were found on the whole surface while no distinct
change was observed in the grain size. Moreover, there is
a possibility of the existence of small cracks that cannot
be recognized by the present SEM. The occurrence of the
cracks suggests the existence of the strain energy induced
by a high magnetic field.
In order to investigate surface morphology in detail, AFM
observation was conducted. In film (a), the surface of the
Fig. 2. SEM images of the iron films electrodeposited in ferrous sulfate
solution at 10 mA cm−2. Film (a) iron electrodeposited in 0 T, film (b)
iron electrodeposited in 0.5 T.
Fig. 3. AFM images of the iron films electrodeposited in ferrous sulfate
solution at 10 mA cm−2
.

4184
H. Matsushima et al. / Electrochimica Acta 49 (2004) 4181–4187
100
10
1.0
0.1
0.01
0.0
0.1
0.2
-h
0.3
0.4
0.5
/ V
Fig. 5. Steady state polarization curves for the HER at films (a) and (b)
in 1.0 mol l−1 KOH solution at 298 K. (᭿) Film (a) and (᭹) film (b).
3.3. Real surface area measurements
The i0 value is determined by both the intrinsic electro-
catalytic activity and the real surface area of the electrode.
In order to distinguish these factors, the real surface area
was estimated from the interfacial double-layer capacitance
by following two methods.
One is the ac impedance method. It was carried out for the
HER in 1.0 mol l−1 KOH solution at four overpotentials: (1)
−0.40 V (2) −0.35 V (3) −0.30 V and (4) −0.25 V. Fig. 6
shows complex-plane plots for the HER at films (a) and (b).
It can be seen that the center of the semicircles is depressed
in both films. The depression is caused by two capacitances
interaction, the interfacial double-layer capacitance, Cdl, and
Fig. 4. Pole figures of iron ꢀ1 1 0ꢁ plane at films (a) and (b) electrodeposited
at 10 mA cm−2. An arrow is a magnetic field direction.
magnetic field at the angle of 30◦. It was found that (1 1 0)
plane was oriented to the magnetic field direction in addition
to (2 1 1) plane orientation. That is, biaxial texture evolution
along (2 1 1) and (1 1 0) axes occurred in a magnetic field.
the hydrogen adsorption pseudocapacitance, C . [19–21].
␾
The impedance data of the HER can be represented by
the equivalent circuit shown in Fig. 7 [22], where Rs is
the solution resistance, R1 the charge transfer resistance,
R2 the adsorption resistance. In the present study, the con-
stant phase element is used for both capacitances in or-
der to compensate for non-homogeneity of the electrode
surface.
In Fig. 6, the solid curves represent the best fitting curves
while the fitted values of the equivalent circuit elements are
listed in Table 2. It was found that the fitting curves are
agreed well with experimental plots. According to Table 2,
the Cdl values were almost constant regardless of the over-
potential. The values of films (a) and (b) were 230 5 and
450 15 ␮F cm−2, respectively.
3.2. Polarization curves for the HER
The HER at films (a) and (b) was examined in 1.0 mol l−1
KOH solution at 298 K. Prior to the polarization measure-
ment, the films were polarized cathodically at −0.20 V ver-
sus RHE for 1 h in order to remove pre-existing oxide film on
the electrode surface. Fig. 5 shows steady state polarization
curves. The measurement was conducted potentiostatically
from −0.05 to −0.50 V, changing the electrode potential by
decrements of 0.05 V. Steady state current was obtained af-
ter 60 s electrolysis at each potential. From these plots, it
is clear that both electrodes have one well-defined Tafel re-
gion. The kinetic parameters obtained from the Tafel plots
are listed in Table 1. The Tafel slopes, b, were almost same
for the both films. They were in good agreement with the
reported value, 110–160 mV, for iron electrode in alkaline
solution [17,18]. On the other hand, the apparent exchange
current density, i0, was different from each other. The i0
value of film (b) was about five times higher than that of
film (a).
The other is potential–relaxation method. The open circuit
potential, just after interrupting the cathodic current density,
Table 1
Values of the kinetic parameters for the HER at films (a) and (b)
Type of film
b (mV)
i0 (mA cm−2
)
(a)
(b)
143
138
1.92 × 10−2
7.98 × 10−2

H. Matsushima et al. / Electrochimica Acta 49 (2004) 4181–4187
4185
is expressed by the following equation [23]
6
5
4
ꢀ
ꢁ
ꢀ
ꢁ
dη
dt
αηF
−Cdl
= i0 exp
(1)
RT
where α is the transfer coefficient and F the Faraday con-
stant. Integration of Eq. (1) gives
3
2
η(t) = a − b log(t + τ)
(2)
4
1
0
3
2
where a = −b log(2.3i0/bCdl), b = 2.3RT/αF and the in-
tegration constant τ = bCdl/2.3it=0. Therefore, η(t) should
be a linear function of log t for t ꢂ τ. Also, from the inte-
gration constant τ
1
0
1
2
3
4
5
6
Z’ (real) / Ωcm2
film (a)
6
5
4
3
2.3it=0τ
Cdl
=
(3)
b
Consequently, the constants b and τ are evaluated by fitting
Eq. (2) to the experimental data so that Cdl can be calculated
from Eq. (3).
The potential was measured at four different cathodic
current densities: (1) 6.5 mA cm−2 (2) 10 mA cm−2 (3)
15 mA cm−2 and (4) 20 mA cm−2. The current was main-
tained for 90 s until the electrode potential reached the
steady state value. Then, the open circuit potential transient
was recorded after interrupting the current.
2
1
0
4
3
1
2
0
1
2
3
4
5
6
Z’ (real) / Ωcm2
film (b)
Fig. 8 shows the potential–relaxation curves for the HER
at films (a) and (b). The circle points and the solid curves
Fig. 6. Experimental (symbols) and fitted (solid curves) complex plane
plots for the HER at four overpotentials (1) −0.40 (2) −0.35 (3) −0.30
and (4) −0.25 V. (Solution: 1.0 mol l−1 KOH, temperature: 298 K).
Cdl
C
RS
R1
R2
Fig. 7. Equivalent circuit used for analysis of the ac impedance data in
Fig. 6.
Table 2
Values of the equivalent circuit elements for the data of the ac impedance
measurement
Type
of film
−η (V) Rs (ꢀ) R1 (ꢀ) C1 (␮F) R2 (ꢀ) C2 (mF)
(a)
(a)
(a)
(a)
0.25
0.30
0.35
0.40
0.605
0.606
0.596
0.623
0.505
0.509
0.453
0.441
225
239
224
227
5.74
2.17
1.12
0.41
5.91
2.48
1.97
1.08
(b)
(b)
(b)
(b)
0.25
0.30
0.35
0.40
0.612
0.622
0.628
0.638
0.173
0.152
0.136
0.123
465
461
476
445
6.66
2.25
1.56
0.73
59.6
50.8
45.2
32.5
Fig. 8. Experimental (symbols) and fitted (solid curves) poten-
tial–relaxation transient curves for the HER at four cathodic current den-
sities: (1) 6.5 (2) 10 (3) 15 and (4) 20 mA cm−2. (Solution: 1.0 mol l−1
KOH, temperature: 298 K).

4186
H. Matsushima et al. / Electrochimica Acta 49 (2004) 4181–4187
stand for experimental data and fitted curves, respectively.
Since the experimental data are well fitted by Eq. (2), it
is possible to calculate the double-layer capacitance on the
iron film. The experimental conditions and the calculated
capacitances are listed in Table 3. The values of films (a)
and (b) were 200 10 and 350 10 ␮F cm−2, respectively.
The capacitances were constant regardless of the applied
current density for the both films. These values agreed ap-
proximately with those obtained by the ac impedance mea-
surement (Table 2).
13
12
11
10
9
Since the Cdl of film (b) is about twice larger than that
of film (a), the real surface area of film (b) can be con-
sidered to be about twice larger. This might be explained
by cracks that increase the surface area contacting the elec-
trolyte. In this case, small cracks that could not recognized
in Fig. 2 would play a major role. On the other hand,
the i0 of film (b) is about five times higher than that of
film (a). This cannot be explained sufficiently only by the
difference of the real surface area. Therefore, it is sug-
gested that film (b) has the higher electrocatalytic activity for
the HER.
8
0
100
200
300
400
500
Current Density / mA cm-2
Fig. 9. Dependencies of the deuterium separation factor on the current
density in 1.0 mol l−1 KOH solution including 10 at.% deuterium at 298 K.
Error bars are standard deviations. (᭿) Film (a) and (᭹) film (b).
For film (a), the value of α increased from 8.7 to 10.2
as the current density increased. The values are higher than
that of the mild steel used in the commercial plant [3]. This
difference might be explained by the impurities in the steel.
It was reported that carbon and other metals impurities might
act as the catalyst for the unfavorable exchange reaction at
the cathode surface as follows [26],
3.4. Deuterium separation
The electrolytic deuterium separation was conducted gal-
vanostatically. The deuterium separation factor, α, is defined
by the following isotopic abundance ratio [24]
HDO + H2 → HD + H2O
(5)
([D]/[H])l
α =
(4)
([D]/[H])g
In film (b), the effective current density decreased by about
one-half from the surface area correction. The values of α
were 9.8–12.3, which is higher than those of film (a). The
highest value 12.3 was obtained at 210 mA cm−2. This is
the highest separation factor among all kinds of materials at
298 K.
On the basis of above results, it was found that the iron
electrode prepared in a magnetic field had high electrocat-
alytic activity for the HER and is superior in deuterium sepa-
ration. There are two conceivable factors that could increase
the electrocatalytic activity.
Firstly, the effect of grain boundary energy is consid-
ered. The grain boundaries have been reported to be ac-
tive sites for the HER [14,27]. This is explained by that
the grain boundaries, whose defects are known as coordi-
natively unsaturated sites, generally possess higher energy
than normal coordination. Therefore, the iron electrode hav-
ing larger grain boundary energy might be more favorable
for the HER. On the other hand, it is reported that the
grain boundary energy is larger for larger misorientation an-
gle [28]. Since iron crystal is highly oriented at film (b)
where the biaxial texture evolution occurred, the misorien-
tation angle of film (b) is expected to be smaller than that
of film (a). In this case, grain boundary energy of film (b)
would be smaller than that of film (a), which means that
film (b) should be inactive for the HER. Since the experi-
mental results contradict the above inference, it is suggested
where [D] and [H] are atomic fractions of deuterium and
protium, the subscript (l) refers to the liquid phase and (g)
to the gas phase.
In our experiment α is measured after continuous water
electrolysis for 8 h since the electrolytic separation requires
the duration time to reach a steady state [25]. Fig. 9 shows
dependencies of the separation factor on the current density
at 298 K. In the figure, circle and square plots stand for
films (a) and (b), respectively. The correction for the increase
of surface area was made for the current density of film
(b). Error bars are standard deviations of three independent
experimental data at each current density.
Table 3
The fitted values of i0 and η in Eq. (2) and the calculated values of Cdl
for the data of the initial potential–relaxation measurement
Type of film
i0 (mA cm−2
)
−η (V)
Cdl (␮F)
(a)
(a)
(a)
(a)
6.5
10
15
0.312
0.362
0.401
0.448
195
197
209
192
20
(b)
(b)
(b)
(b)
6.5
10
15
0.313
0.344
0.377
0.410
362
346
357
362
20

H. Matsushima et al. / Electrochimica Acta 49 (2004) 4181–4187
4187
that the grain boundary energy is not essentially relevant
Acknowledgements
to the increase of the electrocatalytic activity in the present
case.
Authors wish to thank Prof. Y. Fukunaka for helpful dis-
cussions throughout the study. Authors wish to express our
sincere thanks to Prof. T. Yasuda for the use of a permanent
magnet and Prof. S. Kikuchi for the use of an X-ray diffrac-
tometer for texture measurements. This study was partly
supported by the 21st COE program (Establishment of COE
on Sustainable-Energy System) from the Ministry of Edu-
cation, Culture, Sports, Science and Technology in Japan.
Then, the effect of strain energy is discussed. It was re-
ported that the strain energy of the electrode accelerated the
HER in 0.1 M NaOH solution [15]. The mechanism was ex-
plained by that the adsorption of proton is preferred to that
of alkaline metal ions by the strain energy stored in the elec-
trode surface. As described in Section 3.1, the occurrence of
cracks (Fig. 2(b)) suggested that the strain energy is stored
in film (b). Therefore, the increase of the separation factor
could be explained by the electrocatalysis activated by the
strain energy.
References
[1] J.L. Aprea, Int. J. Hydrogen Energy 27 (2002) 741.
[2] D.Lj. Stojic, S.S. Miljanic, T.D. Grozdic, M.M. Jaksic, Bull. Elec-
trochem. 12 (1996) 436.
4. Conclusion
[3] M. Benedict, T.H. Pigford, H.W. Levi (Eds.), Nuclear Chemical
Engineering, McGraw-Hill, New York, 1981, p. 743.
[4] M. Hammerli, W.J. Olmstead, B.L. Muju, F.R. Smith, J. Electroanal.
Chem. 43 (1973) 45.
The two iron films electrodeposited in no magnetic field
(film (a)) and a magnetic field of 0.5 T (film (b)) were ex-
amined with the aim of developing an electrode material
having a high deuterium separation factor.
[5] P.R. Rowland, Nature 218 (1968) 945.
[6] J. Brun, Th. Varberg, W. Gundersen, R. Solli, D.K.N.V.S. Forhandl,
29 (1956) 5.
[7] B. Topley, H. Eyring, J. Chem. Phys. 2 (1934) 217.
[8] J.O’M. Bockris, S. Srinivasan, D.B. Matthews, Discuss. Faraday Soc.
39 (1965) 239.
[9] J.O’M. Bockris, D.B. Matthews, J. Chem. Phys. 44 (1966) 298.
[10] J.O’M. Bockris, E. Gileadi, R. Haynes, J. Electroanal. Chem. 19
(1968) 446.
[11] H. Kita, J. Electrochem. Soc. 113 (1966) 1095.
[12] M. Salmon, B.E. Conway, J. Phys. Chem. 68 (1964) 2009.
[13] P.R. Rowland, J. Electroanal. Chem. 32 (1971) 109.
[14] D.M. Doyle, G. Palumbo, K.T. Aust, A.M. Elsherik, U. Erb, Acta
Metallurg. Mater. 43 (1995) 3027.
[15] T. Ohmori, J. Electroanal. Chem. 172 (1984) 123.
[16] H. Matsushima, T. Nohira, I. Mogi, Y. Ito, Surf. Coat. Technol. 179
(2004) 245.
[17] L.M. Vracar, D.M. Drazic, J. Electroanal. Chem. 265 (1989) 171.
[18] H. Alemu, K. Juttner, Electrochim. Acta 33 (1988) 1101.
[19] D.A. Harrington, B.E. Conway, Electrochim. Acta 32 (1987) 1703.
[20] L. Bai, D.A. Harrington, B.E. Conway, Electrochim. Acta 32 (1987)
1713.
[21] R.D. Armstrong, M. Henderson, J. Electroanal. Chem. 39 (1972) 81.
[22] L. Bai, L. Gao, B.E. Conway, J. Chem. Soc. Faraday Trans. 89
(1993) 235.
[23] R. Simpraga, G.T. Filho, S.Y. Qian, B.E. Conway, J. Electroanal.
Chem. 424 (1997) 141.
(1) The SEM and AFM images of the films showed no large
differences in the grain size and the surface roughness
except for that film (b) contained cracks on the whole
surface, suggesting that the strain energy is stored in the
iron film.
(2) From the X-ray pole figure measurement, the bi-
axial texture evolution was confirmed for film (b)
while the uniaxial texture evolution was found for
film (a). The grain boundary energy is considered
to be small for film (b) since iron crystal is highly
oriented.
(3) The i0 values for the HER in 1.0 mol l−1 KOH solution
at 298 K were 1.92 × 10−2 and 7.98 × 10−2 mA cm−2
for film (a) and film (b), respectively. The real surface
area was found to be about twice larger for film (b)
than film (a) by the Cdl measurements. These results
conclude that the film (b) has the higher electrocatalytic
activity for the HER even after the correction for the
real surface area.
(4) The electrolytic deuterium separation factors mea-
sured in 1.0 mol l−1 KOH at 298 K were higher for
film (b) (9.8–12.3) than those for film (a) (8.7–
10.2).
(5) The improved electrocatalytic activity for the HER is
explained not by the grain boundary energy, but by the
strain energy. The increase of the deuterium separation
factor is also explained by the electrocatalysis activated
by the strain energy.
[24] L.I. Krishtalik, V.M. Tsionsky, J. Electroanal. Chem. 31 (1971) 363.
[25] D.Lj. Stojic, S.S. Miljanic, T.D. Grozdic, N.M. Bibic, M.M. Jaksic,
Int. J. Hydrogen Energy 16 (1991) 469.
[26] J. Brun, W. Gundersen, T. Varberg, D.K.N.V.S. Forhandl, 30 (1957)
29.
[27] J.Y. Huot, M.L. Trudeau, R. Schulz, J. Electrochem. Soc. 138 (1991)
1316.
[28] R.L. Colombo, J. Howard, J. Mater. Sci. 4 (1969) 753.