Hydrogen evolution reaction on gold single-crystal electrodes in acid solutions

Article abstract of DOI:10.1021/jp9831987

The hydrogen evolution reaction on the low index planes of single-crystal Au electrodes has been studied in HClO₄ solutions with the aim of ascertaining the extent of the sensitivity of the reaction rate to the surface crystallographic structure. The use of the hanging meniscus rotating disk technique allowed to carry out all measurements under rotation, thus avoiding blocking effects produced by hydrogen bubbles and accumulation of dissolved hydrogen in the interfacial region during potential excursions into relatively high overpotentials. In contrast with previously published results showing only slight variations in activity with Au crystal face, a pronounced structural sensitivity is reported for the first time. The catalytic activity presents the following sequence Au(111) > Au(100) > Au(100). ? 1998 American Chemical Society.

Full text of DOI:10.1021/jp9831987

J. Phys. Chem. B 1998, 102, 10931-10935
10931
Hydrogen Evolution Reaction on Gold Single-Crystal Electrodes in Acid Solutions
Joelma Perez and Ernesto R. Gonzalez*,†
Instituto de Qu ´ı mica de S a˜ o Carlos, USP, CP 780, 13560-970, S a˜ o Carlos (SP), Brazil
Hebe M. Villullas*,‡
INFIQC, Depto. F ´ı sico Qu ´ı mica, Fac. Ciencias Qu ´ı micas, UNC, Agencia Postal 4,
CC 61, 5000 C o´ rdoba, Argentina
ReceiVed: July 28, 1998; In Final Form: October 17, 1998
The hydrogen evolution reaction on the low index planes of single-crystal Au electrodes has been studied in
HClO solutions with the aim of ascertaining the extent of the sensitivity of the reaction rate to the surface
4
crystallographic structure. The use of the hanging meniscus rotating disk technique allowed to carry out all
measurements under rotation, thus avoiding blocking effects produced by hydrogen bubbles and accumulation
of dissolved hydrogen in the interfacial region during potential excursions into relatively high overpotentials.
In contrast with previously published results showing only slight variations in activity with Au crystal face,
a pronounced structural sensitivity is reported for the first time. The catalytic activity presents the following
sequence Au(111) > Au(100) > Au(110).
1
. Introduction
effects on the (111) and (210) faces.26 A small dependence of
kinetics on the crystal plane was reported, but unfortunately,
The hydrogen evolution reaction (HER) has been widely
the results of those investigations were found to depend on the
previous potential history, were poorly reproducible, showing
variations in currents up to 25% for some of the faces, and
present contradictions in the order in which catalytic activity is
investigated on many different electrode materials and in a
variety of electrolytic solutions and experimental conditions.
1
However, despite the large number of published papers about
the HER, some fundamental aspects remain unresolved. For
instance, the question of whether this reaction is sensitive to
the crystallographic orientation of the electrode surface is still
unclear. Although the importance of structure-sensitive data to
24,25
observed.
For the HER on Ag single-crystal electrodes, only
two papers were published. A structure-insensitive kinetics was
2
7
found by Lorenz et al., while a recent report by Trasatti and
co-workers shows significant differences in the behavior of the
elucidate the characteristics of the HER was recognized in the
2
8
low index planes in perchloric acid.
2
1960s in the pioneer work of Piontelli and other electrochem-
It is also worth mentioning that all the work that has been
ists, early studies of the HER on single-crystal electrodes were
carried out on poorly defined surface structures. The develop-
ment of surface preparation techniques that give well-defined
carried out for the HER on Au and Ag single crystals2
4-28
was
done using stationary electrodes, for which accumulation of
hydrogen bubbles is difficult to avoid even when the applied
overpotential is kept relatively low. Thus, the discrepancies
among reported data and the considerable current variations
found in some works could be due, at least partially, to the
blocking effect of hydrogen bubbles and/or to the influence of
the diffusion of dissolved hydrogen away from the interface.29
On the other hand, the hanging meniscus rotating disk (HMRD)
3
single-crystal surfaces did not begin until the mid-1970s.
Unfortunately, more recent studies of the HER on well-ordered
single-crystal surfaces are still relatively scarce and insufficient
to establish any significant correlation between HER kinetics
and surface structure.
While hydrogen adsorption on Pt single crystals has been
extensively studied and found to be very sensitive to crystal
30
plane,4
-13
technique makes very easy the use of cylindrical single crystals
only a few papers have been devoted to the HER on
31,32
1
4-21
as rotating electrodes.
The use of the HMRD is very simple,
Pt single crystals.
of the HER is insensitive to the surface structure,
recently significant structural sensitivity for the HER on low-
Most of them reported that the kinetics
1
4-19
requiring only a careful control of the meniscus height to avoid
lateral wetting. This introduces undesired contributions to the
measured current from the side of the single crystal, which
present orientations different from the one being investigated
and uncertainties in the evaluation of the effective electrode
and only
index Pt single crystals in acid as well as in alkaline solutions
has been reported,2
0,21
with the (110) face being the most active.
Even fewer papers have been dedicated to study the HER on
single-crystal electrodes with weak catalytic activity, like Au
and Ag. To our knowledge, after the two papers published some
3
3,34
area.
The small amount of published papers and the discrepancies
in the reported results clearly show that a definitive correlation
between the kinetics of the HER in acid media and surface
structure has not yet been established unambiguously and that
further research on this subject is necessary. In this work, the
HER on Au single crystals is studied on rotating electrodes by
using the HMRD technique, which improves the experimental
conditions to study this reaction in a wider range of overpo-
22,23
3
5 years ago by Peraldo Bicelli et al.,
only three studies of
2
4-26
the HER on gold single crystals have been reported.
Two
of them were devoted to the study of the influence of surface
structure on the HER rates2
4,25
and the third one to temperature
†
‡
E-mail: ernesto@iqsc.sc.usp.br.
E-mail: mercedes@fisquim.fcq.unc.edu.ar.
1
0.1021/jp9831987 CCC: $15.00 © 1998 American Chemical Society
Published on Web 12/05/1998

10932 J. Phys. Chem. B, Vol. 102, No. 52, 1998
Perez et al.
tentials. This is necessary in order to gain a better insight into
the effects of surface structure on the kinetics and to clarify
some of the discrepancies found in previously published works.
2. Experimental Section
All experimental work was performed at IQSC-USP. High-
purity (99.999%) gold single crystals with the orientations (111),
(
(
110), and (100) were purchased from Metal Crystals and Oxides
Cambridge, England) with 7 mm length and 5 mm diameter.
Mechanical polishing of the crystals was done in specially
constructed supports using fine emery paper (Microcut 1200)
and several alumina powders down to 0.05 µm from Buehler
(
Illinois, USA). The orientation of the surfaces was kept to better
31
than 0.5° using procedures described elsewhere.
The single crystals were mounted as HMRD electrodes as
previously described.30 To avoid lateral wetting of the electrodes,
the meniscus height was carefully controlled by means of a dial
gauge set at 3.1 mm for the three electrodes.
Electrochemical measurements were carried out in a con-
ventional cell made out of Pyrex glass with a lid of Teflon.
The auxiliary electrode was gold foil. In all experiments, a
reversible hydrogen electrode (RHE) in the working solution
was used as reference. Immediately before each experiment,
the electrodes were cleaned by “flame annealing” for about
5-10 min and cooled in an Ar atmosphere. Before transferring
the electrode to the cell, the surface was protected with a drop
of the working solution. All experiments were done in 0.1 M
HClO4 (Merck, Suprapur) solutions that were deareated with
pure nitrogen. The HMRD electrodes were controlled with a
Pine Instruments setup and a EG&G Princeton Applied Research
Figure 1. (a) Cyclic voltammetry for Au (111) in N
2
-saturated 0.1 M
2
73 A potentiostat.
-1
4
HClO , sweep rate is 50 mV s . (b) Polarization curves for hydrogen
-
1
evolution at different rotation rates, sweep rate is 10 mV s
.
3
. Results and Discussion
the case of extended cycling involving relatively high values
This study of the HER on Au single crystals was carried out
-2
-
1
of current density (ca. 10 mA cm ). In contrast, it was reported
by means of slow potential sweeps (typically 10 mV s ). The
currents were found to be independent of sweep rate between
2
5
by Hamelin and Weaver that changes in the CV profiles of
oxide formation and reduction were observed for the Au(100)
face after extended potential cycling into the hydrogen region.
These variations were taken as evidence of surface defects.
However, it has been demonstrated that variations such as that
-1
5
and 20 mV s indicating quasi-steady-state conditions in that
range. Different prepolarization regimes were employed. Suc-
cessive cycles to different positive potential limits, including
formation and reduction of a monolayer of oxide or just within
the double-layer region, as well as sequential sweeps to different
negative potential limits were recorded. In all cases, results for
the HER were found to be independent of the previous potential
2
5
observed in that work can be associated to the removal of
3
1
trace impurities from the surface.
For all the faces, sequential sweeps into the hydrogen
evolution region were also recorded at different rotation rates
ranging between 50 and 1600 rpm. Several successive cycles
were recorded and found to be identical for all the values of
the rotation rate used and independent of potential regime. The
polarization curves obtained at different rotation speeds are
shown in Figures 1b-3b. Since hydrogen currents do not
increase with rotation rate, it seems clear that for the experi-
mental conditions employed (i) the HER is kinetically controlled
(ii) the removal of hydrogen bubbles from the surface is efficient
even at 50 rpm or, more likely, bubble formation is impeded
25
history. In contrast, it was previously reported that currents
were significantly sensitive to the potential cycling conditions
in HClO4, while in solutions containing LiClO4 and NaClO4,
even larger variations (up to 25% for the smallest currents) were
observed on some faces. Moreover, currents were found to be
higher when measurements in the hydrogen region took place
immediately after a cycle of oxide formation-reduction was
completed than after long delay periods between the oxide
formation-reduction cycle and the measurement of hydrogen
evolution currents.
36
by the lack of nucleation centers, (iii) the solution is relatively
free from active impurities, which otherwise would be carried
to the electrode surface by convection.
On the other hand, it has already been established that
voltammetric curves for the formation and reduction of a
monolayer of oxide, on the low-index planes of Au single-crystal
electrodes in HClO4 solutions, exhibit characteristic features that
can be used as an indication of the general physical state of the
However, for the (111) face, slightly lower currents are
observed at higher rotation rates, probably caused by the
unavoidable presence, even in the cleanest solution, of trace
35
electrode surface and also serve as criterion of solution clean-
3
1
31
liness. Such curves were obtained in the working electrolyte
before and after the measurements into the hydrogen evolution
region and are shown in Figures 1a-3a. Essentially identical
voltammetric profiles were obtained before and after several
successive cycles into the hydrogen evolution region, even in
impurities. From these results, a rotation rate of 50 rpm was
selected for the kinetic measurements. It is worthwhile to note
that the small decrease in the HER current for the (111) face
produced by rotation is the only effect observed in the present
work that can be associated with the influence of trace impurities

Hydrogen Evolution Reaction on Gold Electrodes
J. Phys. Chem. B, Vol. 102, No. 52, 1998 10933
Figure 2. (a) Cyclic voltammetry for Au (100) in N
2
-saturated 0.1 M
Figure 3. (a) Cyclic voltammetry for Au (110) in N
2
-saturated 0.1 M
-
1
HClO
evolution at different rotation rates, sweep rate is 10 mV s
4
, sweep rate is 50 mV s . (b) Polarization curves for hydrogen
-1
HClO
evolution at different rotation rates, sweep rate is 10 mV s
4
, sweep rate is 50 mV s . (b) Polarization curves for hydrogen
-
1
.
-1
.
on the kinetics of the HER. In contrast, it seems likely that the
dependence of hydrogen currents on the prepolarization regime
and the changes in the voltammetric profiles of oxide formation
and reduction observed after measurements in the hydrogen
evolution region reported by Hamelin and Weaver25 could have
been caused by impurities. Also, it should be mentioned that
the variations reported were quite significant for some of the
faces even though all data were obtained using stationary
electrodes. From the comparison of the results presented here,
which were independent of the potential history and showed
that voltammetric curves remained unaltered after the electrode
was polarized into the HER region, with those previously
2
5
published, it seems clear that the surfaces and solutions used
in the present work were cleaner.
Figure 4 shows the polarization curves for the HER obtained
at 50 rpm for the three Au low-index faces. Taking into account
the results discussed, it is conclued that the influence of the
surface crystallographic orientation on the HER rate is absolutely
clear. Additionally, it should be stressed that such a pronounced
variation of activity with the crystal face has never been reported
in previous work. As it can be clearly seen in Figure 4, Au
single crystals present the following sequence of activity toward
the HER: (111) > (100) > (110). As discussed above,
previously published data for the HER on Au single crystals
seem to have been obtained using polarization regimes that could
have favored contamination.2 It remains difficult to explain
however, why the small variation of activity with crystal face
observed previously was in just the opposite order of that
showed in Figure 4.
Figure 4. Polarization curves for hydrogen evolution on the low-index
gold single-crystal electrodes in N -saturated 0.1 M HClO , sweep rate
2
4
-
1
is 10 mV s , rotation rate is 50 rpm.
Even more interesting is the fact that the reported data show
that catalytic activity toward the HER on Ag single crystals
exhibits a sequence identical to that shown for Au in Figure 4,
with the (111) face being the most active. This activity sequence
for Ag was found to correlate well with the corresponding values
of the potential of zero charge, and such a correlation seems to
4,25
37
28
apply also for Au. As pointed out by Trasatti et al., the HER
rate increases with atomic density of the surface; that is, catalytic
activity follows the increase of the electron work function.
In addition, despite the fact that most of the published studies
of the HER on single-crystal surfaces on several materials have
not revealed any pronounced influence of structure on the
It is quite interesting to note though that a recent study of
28
the HER on Ag single crystals in acidic solutions has shown
that catalytic activity is sensitive to crystallographic orientation.
1
,14-19,22-25,27
38
reaction rate,
Horiuti and Kita showed, from a

10934 J. Phys. Chem. B, Vol. 102, No. 52, 1998
Perez et al.
necessary in order to get a complete understanding of the
fundamental aspects of this reaction. It will clearly be of interest
to perform measurements of the HER on single-crystal faces
of Au and other metals using more powerful techniques such
4
8
as impedance spectroscopy.
4
. Conclusions
A large dependence of the kinetics of the hydrogen evolution
reaction on gold electrodes in HClO4 solutions with the
crystallographic orientation of the surface was found for the
first time. The reaction seems to proceed through the same
mechanism on the three faces studied. The catalytic activity
increases with atomic density of the surface and presents the
following sequence: Au(111) > Au(100) > Au(110).
Figure 5. Tafel plots for hydrogen evolution on the low-index gold
Acknowledgment. The authors thank, for financial support,
the Funda c¸ a˜ o de Amparo a Pesquisa do Estado de S a˜ o Paulo
(FAPESP, Proc 97/02471-6 and 95/0692-0). They also thank
single-crystal electrodes in N
data in Figure 4.
2 4
-saturated 0.1 M HClO obtained from
the Conselho Nacional de Desenvolvimento Cient ´ı fico e Tec-
nol o´ gico (CNPq) and the Financiadora de Estudos e Projetos
(FINEP).
TABLE 1: Exchange Current Densities for HER in the
Low-Potential Region
jo/A cm-2
this work,
.1 M HClO
from ref 24,a
1 M HClO
from ref 25,b
0.1 M HClO
4
References and Notes
0
4
4
(1) Appleby, A. J.; Chemla, M.; Kita, H.; Bronoel, G. Encyclopedia
-
-
-
-
7
7
7
7
-7
-7
-7
-7
Au (110)
Au (100)
Au (111)
Polycryst
0.3 × 10
1.0 × 10
0.7 × 10
0.8 × 10
1.4 × 10
of Electrochemistry of the Elements; Bard, A., Ed.; 1982; Vol. IX.
-7
0.5 × 10
2.5 × 10
1.4 × 10
1.0 × 10
(
2) Piontelli, R.; Scientific Papers 1935-1971; Accad. Nazionale dei
-7
0.7 × 10
Lincei, Ist. Lombardo Accad. Scienze e Lettere and Accad. Scienze di
Torino: Milano, 1974; Vol. III.
-7
1.2 × 10
(
3) Schoeffel, J. A.; Hubbard, A. T. J. Electroanal. Chem. 1977, 49,
330.
4) Ishikawa, R. M.; Hubbard, A. T. J. Electroanal. Chem. 1976, 69,
a
Calculated using empirical parameters to fit the experimental curve.
Estimated from Figure 3 of ref 25.
2
3
1
b
(
17.
(5) Clavilier, J.; Faure, R.; Guinet, G.; Durand, R. J. Electroanal. Chem.
theoretical estimation of the recombination rates of adsorbed
hydrogen atoms on the low-index planes of Ni, that a large effect
of crystallographic structure on the reaction rate should be
expected in the order (111) . (100) . (110).
980, 107, 205.
(
6) Clavilier, J. J. Electroanal. Chem. 1980, 107, 211.
(7) Wagner, F. T.; Ross, P. N. J. Electroanal. Chem. 1983, 150, 141.
(
8) Aberdam, D.; Durand, R.; Faure, R.; El-Omar, F. Surf. Sci 1986,
71, 303.
9) Love, B.; Seto, K.; Lipkowski, J. J. Electroanal. Chem. 1986, 199,
1
Figure 5 shows the Tafel plots obtained from the polarization
curves of Figure 4. The Tafel slope varies gradually from of
ca. -0.06 at low overpotentials to -0.12 V/dec at high
overpotentials, in about the same way for the three faces studied.
Thus, it is apparent that the HER proceeds through the same
reaction mechanism on the three crystal faces and that the
differences observed in the values of the currents for the HER
at any given potential are caused by the influence of surface
structure on the rate of elementary steps. Identical behavior was
observed for a polycrystalline electrode. Even though the Tafel
slope varies continuously with electrode potential, exchange
current densities were estimated for the potential region were
slope is ca. -0.060 V/dec and are shown in Table 1 compared
with literature values.
(
219.
(10) Al Jaaf-Golze, K.; Kolb, D. M.; Scherson, D. J. Electroanal. Chem.
986, 200, 353.
11) Clavilier, J.; Armand, D.; Sun, S. G.; Petit, M. J. Electroanal. Chem.
1
1
3
(
986, 205, 267.
(12) Tripovic, A. V.; Adzic, R. R. J. Electroanal. Chem. 1986, 205,
35.
(
13) Markovic, N.; Hanson, M.; McDougall, G.; Yeager, E. J. Elec-
troanal. Chem. 1986, 214, 555.
(14) Physhnogreva, I. I.; Skundin, A. M.; Vasilev, Yu; Bagostki, V. S.
Electrokhimiya 1970, 6, 142.
(
15) Schulander, S.; Rosen, M.; Flinn, D. J. Electrochem. Soc. 1970,
17, 1251.
16) Seto, K.; Ianello, A.; Love, B.; Lipkowski, J. J. Electroanal. Chem.
1987, 226, 351.
17) Protopopoff, E.; Marcus, P. J. Chim. Phys. Phys.-Chim. Biol. 1991,
8, 1423.
1
(
(
8
Despite the large amount of data for the HER on polycrys-
talline Au electrodes that can be found in the literature, published
results are rather inconsistent. Although one value of Tafel slope
has been usually reported, values are found to vary from -0.027
(
18) Kita, H.; Ye, S.; Gao, Y. J. Electroanal. Chem. 1992, 334, 351.
(19) G o´ mez, R.; Fern a´ ndez Vega, A.; Feliu, J. M.; Aldaz, A. J. Phys.
Chem. 1993, 97, 4769.
20) Markovic, N. M.; Sarraf, S. T.; Gasteiger, H. A.; Ross, P. N., Jr.
J. Chem. Soc., Faraday Trans. 1996, 92, 3719.
21) Markovic, N. M.; Grgur, B. N.; Ross, P. N., Jr. J. Phys. Chem. B
(
3
9-45
to -0.15 V/dec.
Kuhn and Byrne observed two Tafel
(
slopes, -2.3RT/2F at low current densities with a sharp
1997, 101, 5405.
4
6
(22) Peraldo Bicelli, L.; Graziano, M. R. Istituto Lombardo (Rend. Sc.)
A 1962, 96, 98.
transition to -2 × 2.3RT/F at high current densities. More
recently, Conway and Bai have also reported a Tafel slope of
(23) Peraldo Bicelli, L.; Romagnani, C. Istituto Lombardo (Rend. Sc.)
-
0.06 V/dec for the HER on polycrystalline gold at relatively
A 1963, 97, 155.
4
7
(24) Brug, G. J.; Sluyters-Rehbach, M.; Sluters, J. H.; Hamelin, A. J.
Electroanal. Chem. 1984, 81, 245.
low overpotentials. In view of this variety of reported
Tafel slopes, it is surprising that although no great influence of
surface structure was found in previous studies of the HER on
Au single-crystal electrodes, smoothly curved Tafel plots with
slopes changing from ca. -0.06 to about -0.12 V/dec were
(25) Hamelin, A.; Weaver, M. J. Electroanal. Chem. 1987, 223, 171.
(26) Hamelin, A.; Stoicoviciu, L.; Chang, S.; Weaver, M. J. Electroanal.
Chem. 1991, 307, 183.
(
27) Abd. El-Hamin, A. M.; Juttner, K.; Lorenz, W. J. J. Electroanal.
Chem. 1980, 106, 193.
28) Doubova, L. M.; Daolio, S.; Trasatti, S. 192nd Electrochem. Soc.
Meeting, 48th ISE Meeting, Paris, 1997, Abstract 1056.
2
2-25
reported.
Although some mechanistic interpretation of
these values has been proposed,
(
24,47
further studies seem to be

Hydrogen Evolution Reaction on Gold Electrodes
J. Phys. Chem. B, Vol. 102, No. 52, 1998 10935
(
(
63.
29) Conway, B. E.; Bai, L. J. Electroanal. Chem. 1986, 198, 149.
30) Cahan, B. D.; Villullas, H. M. J. Electroanal. Chem. 1991, 307,
Elektrokhimiya 1977, 13, 717). In the HMRD configuration, an insulating
shroud is not used and, thus, such sites for nucleation are not present.
(37) Lecoeur, J.; Andro, J.; Parsons, R. Surf. Sci. 1982, 114, 320.
2
1
1
3
3
(
38) Horiuti, J.; Kita, H. J. Res. Inst. Catal. Hokkaido UniVersity 1964,
(31) Cahan, B. D.; Villullas, H. M.; Yeager, E. B. J. Electroanal. Chem.
1
2, 122.
991, 306, 213.
(
(
39) Bockris, J. O’M.; Parsons, R. Trans. Faraday Soc. 1948, 44, 860.
40) Pentland, N.; Bockris, J. O’M.; Sheldon, E. J. Electrochem. Soc.
(
32) Perez, J.; Villullas, H. M.; Gonz a´ lez, E. R. J. Electroanal. Chem.
997, 435, 179.
33) Villullas, H. M.; L o´ pez Teijelo, M. J. Electroanal. Chem. 1995,
84, 25.
1
957, 104, 182.
(
(
(
(
(
41) Schuldiner, S.; Hoare, J. P. J. Phys. Chem. 1957, 61, 705.
42) Ives, D. J. G. Can. J. Chem. 1959, 37, 213.
43) Schmid, G. M. Electrochim. Acta 1967, 12, 449.
44) Sasaki, T.; Matsuda, A. Chem. Lett. J. Inst. Catal. Hokkaido
(34) Villullas, H. M.; L o´ pez Teijelo, M. J. Electroanal. Chem. 1995,
85, 39.
(
35) Hamelin, A. Modern Aspects of Electrochemistry; Conway, B.,
Bockris, J. O’M., Eds.; Plenum: New York, 1986; Chapter 1, Vol. 16.
36) It was demonstrated by photographic measurements that in very
UniVersity 1973, 21, 157.
(45) Sasaki, T.; Matsuda, A. Chem. Lett. J. Inst. Catal. Hokkaido
UniVersity 1981, 29, 113.
(
clean electrolytes no hydrogen bubbles form on the surface of a conventional
rotating disk electrode even up to 10 A/cm . It was also shown that bubbles
nucleate only at the metal-PTFE edge (Ludwig, F.; Sen, R.; Yeager, E. B.
(46) Kuhn, A. T.; Byrne, M. Electrochim. Acta 1971, 16, 391.
(47) Conway, B. E.; Bai, L. Electrochim. Acta 1986, 31, 1013.
(48) Harrington, D. A.; Conway, B. E. Electrochim. Acta 1987, 32, 1703.
2