вход по аккаунту


Tuning Reaction Rates by Lateral Strain in a Palladium Monolayer.

код для вставкиСкачать
Pseudomorphic monolayers can be
formed on single-crystal substrates by the
deposition of palladium. Depending on the lattice
parameters of the substrate, the palladium monolayer is compressed
or dilated, effects which alter its properties, for example, the adsorption
of hydrogen or the electrooxidation of formic acid. The changes in
these properties can be studied electrochemically. For more information see the Communication by L. A. Kibler et al. on the following
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200462127
Angew. Chem. Int. Ed. 2005, 44, 2080 – 2084
Tuning Reaction Rates by Lateral Strain in a
Palladium Monolayer**
Ludwig A. Kibler,* Ahmed M. El-Aziz, Rdiger Hoyer,
and Dieter M. Kolb
One main goal of current electrocatalysis studies is to
establish the relationship between the surface structure of
an electrode material and its catalytic activity for a given
reaction.[1] Because of constraints set by most experimental
techniques, catalytic phenomena on surfaces are commonly
studied with model reactions on well-ordered metal surfaces.
Such studies contribute to a general understanding of
structure–reactivity relationships. In addition, model surfaces
are also used in theory to elucidate basic principles and trends
in catalysis, for example, by density functional theory (DFT)
Commonly employed model surfaces comprise singlecrystal electrode surfaces, which are mainly characterized by
their chemical composition, their crystallographic orientation, and the amount and the type of defects. These
parameters are crucial for the kinetics of electrode reactions
that involve adsorption of intermediates.[3, 4]
More recently, the role of surface strain in catalysis,
particularly as a means of tuning the catalytic activity, has
attracted interest.[5–7] Such strained surfaces can be realized
by the deposition of a metal onto a single-crystal substrate
that has different lattice constants. In the case of palladium
(1 1) commensurates are formed, that is, pseudomorphic
monolayers, this topic is reported herein.
Palladium overlayers on early-transition-metal surfaces
can have altered physical and chemical properties.[8–10] However, there was not always consensus concerning the origin of
such changes.[11] A rehybridization of the d-band without
considerable charge transfer from one metal to the other
seems to be the key to understanding this phenomenon.[11]
Recently, it has been shown by DFT calculations that the
electronic properties of a surface can be modified considerably by changing nearest-neighbor separations, such as in
pseudomorphic overlayers.[12–18] This effect is related to a shift
in the d-band center, the energetic position of which is crucial
for determining the physical and chemical properties of the
surface.[13] Accordingly, a change in the geometry of the
surface lattice will influence the bond strength of an
adsorbate. However, most of the calculated structures
cannot easily be fabricated or are not stable under real
conditions. As an extreme case, Koper et al. calculated the
effect of changing the atomic separations in an unsupported(!) platinum monolayer on its adsorption properties.[19]
In this way, the effect of pure lateral strain has been
computed, since substrate effects are excluded, and valuable
physical trends are obtained.
The effect of local strain, which is induced by subsurface
argon bubbles, on the adsorption strength of oxygen atoms
has been examined by scanning tunneling microscopy
(STM).[20] Other methods, such as photoelectron[10] and
vibrational spectroscopy,[21] have been used to reveal the
altered properties of pseudomorphic overlayers. However,
these methods give only indirect information about adsorption strengths.
Although of practical importance, the consequences of
lateral strain within metal overlayers on electrocatalytic
reactions have to date not been studied systematically. Only
in a few investigations, was the influence of pseudomorphic
growth on the catalytic activity considered.[5, 7, 22] Recent
electrochemical studies with epitaxially grown palladium
overlayers on Au(111) surfaces involved adsorption and
absorption of hydrogen,[23] formic acid oxidation,[5, 22] oxygen
reduction,[24] as well as formaldehyde[25] and CO oxidation.[4]
In all these cases, systematic changes in the activity with
palladium overlayer thickness were reported.
Herein, we present an electrochemistry study with
pseudomorphic palladium monolayers on seven different
single-crystal electrode surfaces of hexagonal orientation. We
give clear experimental evidence for a change in electrochemical properties of the (1 1) palladium monolayer with
the lattice parameter, in full accord with theory.[13] The
processes studied are hydrogen adsorption and the electrooxidation of formic acid.
The electrochemical behavior of a freshly prepared,[4, 26]
massive Pd(111) electrode in 0.1m H2SO4 is most conveniently
characterized by a cyclic voltammogram (Figure 1). Positive
of 0.6 V (all potentials are measured and quoted against the
[*] Dr. L. A. Kibler, Dr. R. Hoyer, Prof. Dr. D. M. Kolb
Abteilung Elektrochemie
Universitt Ulm
89069 Ulm (Germany)
Fax: (+ 49) 731-50-25409
Dr. A. M. El-Aziz
National Research Center, Physical Chemistry Department
El-Tahrir Street 12622 Dokki, Cairo (Egypt)
[**] Financial support by the Fonds der Chemischen Industrie is greatly
Angew. Chem. Int. Ed. 2005, 44, 2080 –2084
Figure 1. Cyclic voltammogram for a clean and well-ordered Pd(111)
electrode in 0.1 m H2SO4. Scan rate v = 10 mVs 1. The curve shows the
first scan, starting at 0.1 V versus SCE in negative direction.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
saturated calomel electrode SCE) surface oxidation starts,
with an anodic current peak at 0.8 V. This potential region is
usually avoided in our experiments, since irreversible surface
defects are generated during reduction of the oxide at 0.4 V.[27]
Negative of 0 V, the electrochemical behavior of palladium is
dominated by absorption of hydrogen into the bulk metal.
However, with well-ordered Pd(111) surfaces a sharp peak
around 0.07 V emerges for concomitant hydrogen adsorption and sulfate desorption.[28, 29] An ordered sulfate adlayer is
formed on the Pd(111) surface at + 0.09 V[29] giving rise to a
small voltammetric peak (Figure 1).
While the surface preparation for massive palladium
single-crystal electrodes needs careful control of annealing
and cooling conditions, the preparation of pseudomorphic
palladium monolayers can easily be done by electrochemical
deposition. This process has been thoroughly studied for
palladium on Au(111)[23, 30, 31] and Pt(111),[5, 27, 32, 33] and recently
also for Rh(111),[34] Ru(0001),[35, 36] Ir(111), Re(0001) and
PtRu(111).[36] Palladium is deposited onto these single-crystal
surfaces from PdCl2 or PdSO4 containing acidic solutions. A
palladium monolayer can be formed either by underpotential
deposition (upd) as in the case of Au, Pt, or Rh surfaces, or by
controlled dissolution of multilayers (Ru, Ir, Re). A transition
to a rather three-dimensional growth is observed after the
formation of one monolayer in all cases except for Au(111),
where two complete monolayers can be obtained.[31]
Figure 2 shows a series of STM images for the growth of
the first pseudomorphic palladium monolayer on Au(111).[31]
The topography of the substrate is nicely reproduced by the
monoatomic high-palladium layer (Figure 2 d). Note that
there are no signs of alloy formation for all the systems
under study. The pseudomorphy of palladium monolayers on
Au(111)[37] and Pt(111)[38] was verified by in situ surface X-ray
diffraction measurements. For the other systems, the formation of a pseudomorphic monolayer is inferred from in situ
STM measurements, which reveal two-dimensional growth,
and from cyclic voltammetry.
While the electrochemical behavior of the pseudomorphic
palladium layers is markedly different from that of a massive
Pd(111) electrode, bulk properties are approached for thicker
overlayers in all cases. Thin Pd overlayers show a high
overpotential for the absorption of hydrogen.[23] As a
consequence, characteristic voltammetric peaks arising from
hydrogen adsorption can be readily observed, which are
usually masked by the large currents for hydrogen absorption
into bulk palladium.
Figure 3 shows the positive-potential scans of cyclic
voltammograms for pseudomorphic palladium monolayers
on Au(111), Pt(111), PtRu(111), Rh(111), Ir(111), Ru(0001),
Figure 3. Positive voltammetric sweeps for Pd(111) and pseudomorphic palladium monolayers (PdML) on seven different single-crystal
substrates in 0.1 m H2SO4, revealing a spectrum for hydrogen desorption. Scan rate v = 10 mVs 1.
Figure 2. Series of STM images (500 nm 500 nm) for palladium deposition onto Au(111) from 0.1 m H2SO4 + 0.1 mm [PdCl4]2 .[31] A monoatomic high step and several monoatomic high islands are seen on
the bare gold surface (a) at 0.55 V. Starting from these defects palladium islands grow on the surface (b, c). The full pseudomorphic
palladium monolayer reproduces exactly the topography of the
substrate (d).
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
and Re(0001) together with the corresponding curve for bulk
Pd(111). Sulfuric acid was used instead of perchloric acid as
electrolyte owing to the possibility of chloride formation by
electrocatalytic perchlorate reduction. The so-called CO
displacement method has shown that (bi)sulfate is adsorbed
on the palladium surface at potentials positive of the
characteristic peaks in Figure 3, and hydrogen at potentials
negative of them.[4, 39] Though influenced by anion adsorption,
the peak potentials serve as a measure for the hydrogenadsorption strength (the free enthalpy of adsorption can be
derived directly from them) in such a solution. We assume
that hydrogen adsorbs much stronger than sulfate, since the
hydrogen peak potentials for sulfuric acid are more positive
than the potentials of zero charge.[40]
Nørskov et al. have shown recently that d-band centers for
pseudomorphic overlayers, shift in energy with nearestneighbor separations, an effect which has a direct influence
on binding energies.[12, 13] Their model is reproduced in
simplified form in Figure 4 a. In Figure 4 b, the electrode
potentials for hydrogen desorption on the various Pd(111)
surfaces are plotted versus the d-band shift, taken from
Angew. Chem. Int. Ed. 2005, 44, 2080 –2084
The altered adsorption behavior of hydrogen on the
different palladium surfaces, suggests that reactions involving
adsorbed intermediates will also have different kinetic
parameters for the various palladium surfaces. We chose
formic acid oxidation as a test reaction.
Although platinum as the most prominent electrocatalyst
for many oxidation reactions, it is easily poisoned during the
oxidation of formic acid by intermediate carbon monoxide,
this undesired side reaction is strongly suppressed on
palladium electrodes. Figure 5 shows current–potential
Figure 4. a) The main origin of a shift in the d-band center ed is a
change in the interatomic distances within an overlayer. If the d-band
of a metal is more than half-filled, an expanded pseudomorphic monolayer will lead to an up-shift of ed owing to band narrowing and energy
conservation.[13] b) A plot of the hydrogen-desorption potentials versus
the shift of the d-band center, ded, shows a linear correlation as theoretically predicted.[13] For palladium on the Re(0001) substrate, no
hydrogen peak was measured (see text). The hydrogen peak at 0.1 V
for palladium on PtRu(111) suggests a shift of the d-band center ded
of about 0.1 eV, which has not yet been calculated.
ref. [13]. The linear correlation between both quantities is
clearly seen.
Cyclic voltammograms performed with palladium monolayers on Re(0001) revealed that there is no hydrogen
adsorption on such a surface in sulfuric acid solutions. An
explanation for this observation is also given in Figure 4 b:
From the calculation of the corresponding d-band shift
( 0.72 eV[41]), it is concluded that hydrogen is only weakly
bound to this surface and the hydrogen adsorption peak is
shifted into the hydrogen-evolution region and hence, escapes
detection. This example shows that the electronic modification of a palladium monolayer has essentially two contributions. 1) If the lattice parameter of the overlayer is different
from that of the surface of the bulk metal, the extent of metal–
metal bonding within the surface is changed and accompanied
by a shift in the d-band center (geometric effect). 2) If the
electronic interaction between the surface layer and the
substrate is large, the location of the d-band center is also
influenced by the so-called ligand effect. Since the d-band
center shifts systematically with the lattice (the Pd–Pd
separation) the geometric effect is dominant for all the
systems shown in Figure 4 except for palladium on Re(0001),
where the ligand effect seems to be more important.
Angew. Chem. Int. Ed. 2005, 44, 2080 –2084
Figure 5. Current–potential curves (positive sweeps) showing the oxidation of formic acid on Pd(111) and on pseudomorphic palladium
monolayers in 0.1 m H2SO4 + 0.2 m HCOOH. Scan rate v = 20 mVs 1.
The curve for palladium on Re(0001) is not shown, since the activity of
that surface is practically zero as in the case of palladium on
curves for the anodic oxidation of formic acid on seven
different palladium monolayers and on a massive Pd(111)
electrode in acidic media. Comparison with the results in
Figure 3, shows that the onset of formic acid oxidation follows
the same trend as the hydrogen-desorption peak potential.
This result means that, starting with palladium on Au(111),
the catalytic activity of a palladium monolayer can be
substantially increased by shifting center of the d-band
down by means of lateral compression (see Figure 4 b).
However, if the position of the d-band center becomes too
low, as in the case of Pd on Rh, Ir, Ru, and Re, the oxidation
currents decrease owing to a weaker binding of adsorbates
which reduces their coverage. This situation leads to a
maximum in catalytic activity when the ability of the
palladium surface to bind adsorbates is not too weak and
not too strong. The difference between the two classes of
palladium surfaces is eye-catching: While the more open
palladium monolayers show an overall high catalytic activity,
compression reduces the oxidation currents significantly.
For the first time, a systematic electrochemical study is
presented, which is in full agreement with the so-called dband model from Nørskov et al.[12, 13] The adsorption behavior
and the catalytic activity for formic acid electrooxidation of a
pseudomorphic palladium monolayer on various single-crystal electrodes—Au(111), Pt(111), PtRu(111), Rh(111),
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Ir(111), Ru(0001), Re(0001), and Pd(111)—vary dramatically
with lateral compression or dilatation. Characteristic voltammetric peaks have been taken as a measure for the hydrogenbinding energy on the pseudomorphic palladium monolayers.
There is a linear relation between the hydrogen-desorption
peak potential and the calculated shift in the d-band center.
The consequences for the electrooxidation of formic acid on
these surfaces are highlighted. Although at present we cannot
make any quantitative correlation with calculated d-band
shifts because of lack of knowledge about the actual reaction
paths, our systematic studies clearly show the great importance of nearest-neighbor separations in heterogeneous
catalysis. They also emphasize the ease of tuning reaction
rates by the electrochemical modification of catalysts.
Received: September 27, 2004
Revised: October 28, 2004
Published online: February 11, 2005
[31] L. A. Kibler, M. Kleinert, R. Randler, D. M. Kolb, Surf. Sci.
1999, 443, 19.
[32] G. A. Attard, A. Bannister, J. Electroanal. Chem. 1991, 300, 467.
[33] J. Inukai, M. Ito, J. Electroanal. Chem. 1993, 358, 307.
[34] R. Hoyer, L. A. Kibler, D. M. Kolb, Surf. Sci. 2004, 562, 275.
[35] S. R. Brankovic, J. McBreen, R. R. Adzic, Surf. Sci. 2001, 479,
[36] L. A. Kibler, A. M. El-Aziz, unpublished results. The PtRu(111)
electrode had a 1:1 composition and has been prepared to yield a
presumably pseudomorphic platinum overlayer on the bulk
[37] M. Takahasi, Y. Hayashi, J. Mizuki, K. Tamura, T. Kondo, H.
Naohara, K. Uosaki, Surf. Sci. 2001, 461, 213.
[38] M. J. Ball, C. A. Lucas, N. M. Markovic, V. Stamenkovic, P. N.
Ross, Surf. Sci. 2002, 518, 201.
[39] B. lvarez, V. Climent, A. Rodes, J. M. Feliu, Phys. Chem.
Chem. Phys. 2001, 3, 3269.
[40] A. M. El-Aziz, L. A. Kibler, D. M. Kolb, Electrochem. Commun.
2002, 4, 535.
[41] V. Pallassana, M. Neurock, L. B. Hansen, B. Hammer, J. K.
Nørskov, Phys. Rev. B 1999, 60, 6146.
Keywords: adsorption · electrochemistry · heterogeneous
catalysis · monolayers · oxidation · surface chemistry
[1] N. M. Markovic, P. N. Ross, Surf. Sci. Rep. 2002, 45, 117.
[2] J. P. Perdew, J. A. Chevary, S. H. Vosko, K. A. Jackson, M. R.
Pederson, D. J. Singh and C. Fiolhais, Phys. Rev. B 1992, 46, 6671.
[3] R. R. Adzic, A. V. Tripkovic, W. E. OGrady, Nature 1982, 296,
[4] A. M. El-Aziz, L. A. Kibler, J. Electroanal. Chem. 2002, 534, 107.
[5] M. Baldauf, D. M. Kolb, J. Phys. Chem. 1996, 100, 11 375.
[6] C. Xu, D. W. Goodman, J. Phys. Chem. 1996, 100, 245.
[7] L. A. Kibler, D. M. Kolb, Z. Phys. Chem. 2003, 217, 1265.
[8] J. A. Rodriguez, D. W. Goodman, J. Phys. Chem. 1991, 95, 4196.
[9] M. W. Ruckman, M. Strongin, Acc. Chem. Res. 1994, 27, 250.
[10] J. A. Rodriguez, Surf. Sci. Rep. 1996, 24, 223.
[11] J. A. Rodriguez, D. W. Goodman, Acc. Chem. Res. 1995, 28, 477.
[12] B. Hammer, J. K. Nørskov, Surf. Sci. 1995, 343, 211.
[13] A. Ruban, B. Hammer, P. Stoltze, H. L. Skriver, J. K. Nørskov, J.
Mol. Catal. A 1997, 115, 421.
[14] M. Mavrikakis, B. Hammer, J. K. Nørskov, Phys. Rev. Lett. 1998,
81, 2819.
[15] P. Liu, J. K. Nørskov, Fuel Cells 2001, 1, 192.
[16] A. Roudgar, A. Groß, Phys. Rev. B 2003, 67, 033 409.
[17] A. Schlapka, M. Lischka, A. Groß, U. Ksberger, P. Jakob, Phys.
Rev. Lett. 2003, 91, 016 101.
[18] A. Roudgar, A. Groß, J. Electroanal. Chem. 2003, 548, 121.
[19] T. E. Shubina, M. T. M. Koper, Electrochim. Acta 2002, 47, 3621.
[20] M. Gsell, P. Jakob, D. Menzel, Science 1998, 280, 717.
[21] E. Kampshoff, E. Hahn, K. Kern, Phys. Rev. Lett. 1994, 73, 704.
[22] L. A. Kibler, A. M. El-Aziz, D. M. Kolb, J. Mol. Catal. A 2003,
199, 57.
[23] M. Baldauf, D. M. Kolb, Electrochim. Acta 1993, 38, 2145.
[24] H. Naohara, S. Ye, K. Uosaki, Electrochim. Acta 2000, 45, 3305.
[25] H. Naohara, S. Ye, K. Uosaki, J. Electroanal. Chem. 2001, 500,
[26] A. Cuesta, L. A. Kibler, D. M. Kolb, J. Electroanal. Chem. 1999,
466, 165.
[27] R. Hoyer, L. A. Kibler, D. M. Kolb, Electrochim. Acta 2003, 49,
[28] N. Hoshi, K. Kagaya, Y. Hori, J. Electroanal. Chem. 2000, 485,
[29] L.-J. Wan, T. Suzuki, K. Sashikata, J. Okada, J. Inukai, K. Itaya, J.
Electroanal. Chem. 2000, 484, 189.
[30] H. Naohara, S. Ye, K. Uosaki, J. Phys. Chem. B 1998, 102, 4366.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 2080 –2084
Без категории
Размер файла
380 Кб
strait, reaction, palladium, tuning, lateral, monolayer, rate
Пожаловаться на содержимое документа