68
3 Potential- or Adsorbate-Induced Changes in Surface Stress …
measured for surface reconstruction or adsorption of electrolyte species on singlecrystal Au electrodes with the corresponding electrocapillary (surface tension vs.
potential) curves to make clear the difference between surface stress and surface
tension.
The changes in surface stress depend not only on potential but also on surface
charge density. The dependence of surface stress g on surface charge density q,
i.e., ζ g,q =
∂g
∂q
, is named “surface stress–surface charge density coefficient,” which
is characteristic for solid electrode surfaces. The changes in surface elastic strain
of a solid electrode induce the changes in electrode potential. The dependence of
potential E on surface elastic strain ε, i.e., ζ E,ε =
∂ E
∂ε
, is named “potential–surface
elastic strain coefficient,” which is also characteristics for solid electrode surfaces. It
has been theoretically derived that both coefficients are equivalent to each other (see
Eq. (1.124) in Sect. 1.8 of Chap. 1). The above two characteristic parameters can be
measured separately by several methods such as DSA and DECMA (see Sects. 2.3
and 2.4 of Chap. 2). We compare the values of ζ g,q and ζ E,ε measured separately for
a (111)-textured Au (111) film electrode in perchloric acid solution to confirm that
ζ g,q is equivalent to ζ E,ε . The sign-reversal of ζ g,q or ζ E,ε has been observed for nanoporous or (111)-textured Pt electrodes in acid solutions, depending on the potential
regions of hydrogen adsorption, electric double layer (or oxygen adsorption), and
oxide formation. We discuss the origin of the sign-reversal on the basis of the changes
in electronic structure of the electrode surfaces.
3.2 Surface Reconstruction
When a clean solid surface is created by cleavage in ultra-vacuum, the surface atoms
have an excess energy (i.e., surface energy) as compared to that of interior atoms
in solid since the number of the nearest-neighbor atoms decreases on the surface
to bring the increase in electronic charge that does not participate in bonding. As
a result, surface relaxation in which the bond length normal to the surface varies
slightly for atoms in the surface region takes place on the clean surface to minimize
the surface energy. Particularly, the clean surfaces of the face-centered cubic (fcc)
transition metals such as Au, Pt, and Pd are often subjected to surface reconstruction
in which surface atoms undergo a lateral displacement to form a two-dimensional
superlattice different from the interior lattice structure [1–3].
For example, the (100) surface of fcc transition metal reconstructs into a hexagonal close-packed (hcp) form due to a heat treatment (see Fig. 3.1a) [1–3]. The
reconstructed (100) surface is slightly buckled because of the significant structural
misfit between the hcp top layer and the underlying (100) plane, and it is named
“hex” structure [1, 2]. The atomic density of the reconstructed hex structure is higher
by 20–25% than the unreconstructed (100) surface. On the other hand, the (110)
surface reconstructs into a (1 × 2) missing-row structure, where every second row in
[001] direction is missing (see Fig. 3.1b) [1, 2, 4, 5]. The surface reconstruction of
the (100) and (110) surfaces accompanies the increase in atomic density of surface
3 Potential- or Adsorbate-Induced Changes in Surface Stress …
measured for surface reconstruction or adsorption of electrolyte species on singlecrystal Au electrodes with the corresponding electrocapillary (surface tension vs.
potential) curves to make clear the difference between surface stress and surface
tension.
The changes in surface stress depend not only on potential but also on surface
charge density. The dependence of surface stress g on surface charge density q,
i.e., ζ g,q =
∂g
∂q
, is named “surface stress–surface charge density coefficient,” which
is characteristic for solid electrode surfaces. The changes in surface elastic strain
of a solid electrode induce the changes in electrode potential. The dependence of
potential E on surface elastic strain ε, i.e., ζ E,ε =
∂ E
∂ε
, is named “potential–surface
elastic strain coefficient,” which is also characteristics for solid electrode surfaces. It
has been theoretically derived that both coefficients are equivalent to each other (see
Eq. (1.124) in Sect. 1.8 of Chap. 1). The above two characteristic parameters can be
measured separately by several methods such as DSA and DECMA (see Sects. 2.3
and 2.4 of Chap. 2). We compare the values of ζ g,q and ζ E,ε measured separately for
a (111)-textured Au (111) film electrode in perchloric acid solution to confirm that
ζ g,q is equivalent to ζ E,ε . The sign-reversal of ζ g,q or ζ E,ε has been observed for nanoporous or (111)-textured Pt electrodes in acid solutions, depending on the potential
regions of hydrogen adsorption, electric double layer (or oxygen adsorption), and
oxide formation. We discuss the origin of the sign-reversal on the basis of the changes
in electronic structure of the electrode surfaces.
3.2 Surface Reconstruction
When a clean solid surface is created by cleavage in ultra-vacuum, the surface atoms
have an excess energy (i.e., surface energy) as compared to that of interior atoms
in solid since the number of the nearest-neighbor atoms decreases on the surface
to bring the increase in electronic charge that does not participate in bonding. As
a result, surface relaxation in which the bond length normal to the surface varies
slightly for atoms in the surface region takes place on the clean surface to minimize
the surface energy. Particularly, the clean surfaces of the face-centered cubic (fcc)
transition metals such as Au, Pt, and Pd are often subjected to surface reconstruction
in which surface atoms undergo a lateral displacement to form a two-dimensional
superlattice different from the interior lattice structure [1–3].
For example, the (100) surface of fcc transition metal reconstructs into a hexagonal close-packed (hcp) form due to a heat treatment (see Fig. 3.1a) [1–3]. The
reconstructed (100) surface is slightly buckled because of the significant structural
misfit between the hcp top layer and the underlying (100) plane, and it is named
“hex” structure [1, 2]. The atomic density of the reconstructed hex structure is higher
by 20–25% than the unreconstructed (100) surface. On the other hand, the (110)
surface reconstructs into a (1 × 2) missing-row structure, where every second row in
[001] direction is missing (see Fig. 3.1b) [1, 2, 4, 5]. The surface reconstruction of
the (100) and (110) surfaces accompanies the increase in atomic density of surface