13.2 O-Cu(001) Surface Reaction
257
In 1993, Lederer et al. [44], Arvanitis et al. [48], Yokoyama et al. [49], asserted
that there are two phases during the O-Cu(001) reconstruction. One corresponds to
an unreconstructed state and the other a reconstructed phase. In the former case,
oxygen atom locates 0.8 Å above the first Cu layer to form an off-centered pyramid,
or a c(2 × 2)-O structure with oxygen shifting 0.1 Å away from the symmetry site
center. Upon reconstruction, the adsorbate locates 0.2 Å above the first Cu layer with
a MR (
√
2 × 2
√
2)R45°-O structure. In the two sequential phases, oxygen adsorbate
retains at the apical site of an off-centered pyramid and then at the apical site of a
distorted tetrahedron. The nature of the O-Cu bond was suggested as “predominantly
ionic for the first phase and more covalent for the reconstructed second phase.”
It became certain in 1996 when Fujita et al. [43] confirmed the bi-phase structures
with STM observations. In the first phase, oxygen sits in the near-center of the nextnearest-neighboring hollow site to form a pyramid at an oxygen exposure of 25 L
or lower. The precursor phase is composed of nanometric c(2 × 2)-O domains with
protruding boundaries. Upon increasing oxygen exposures, the c(2 × 2)-O evolves
into the (2
√
2 ×
√
2)R45°-O structure by every fourth row of Cu atoms missing. In
1998, Tanaka et al. [50] demonstrated that the two phases are reversible under the
bombardment of energetic Ar
+ beams.
In 1997, Sun [19], Sun and Bai [51] clarified that the MR formation results from the
process of oxide tetrahedron bond formation that breaks all bonds of the MR atoms
being evaporated without needing other perturbation. The two sequential phases
result from the transition from O
−1 to O
−2 under the specific Cu(001) geometric
condition [52].
Figure 13.1a shows the zigzag and the U-shape protruding domain boundaries
of the first O
−1 derived phase. Comparatively, the O
−2 derived “dumb-bell” shaped
protrusions in Fig. 13.1b bridge over the missing rows. Jensen et al. [45] interpreted
these protrusions as: “Pairing of Cu-O-Cu chains forms by displacing the Cu and/or
O atoms next to the missing row by about 0.35 Å towards the missing row; the OCu-O chain is linked by the delocalized antibonding states”. However, Sun affirmed
that the nonbonding electron lone pair “:” connects the O
−2 :Cu
p :O
−2 and the pairing
Cu
p
↔ Cu
p dipoles (p) crosses the missing row. The “↔” represents the repulsive
force between the dipoles [19].
From the STM image of Fig. 13.1b, the separation of the paired rows was estimated
at 2.9 ± 0.3 Å. The length of the bright spot was estimated at 5.1 Å; the height of
the bright spot was 0.45 Å; whereas, for the pure Cu(001) surface the protrusion
was about 0.3 Å in height [38]. These STM observations founded the H 2 O-like
tetrahedral models to specify the bond nature and its consequence on the individual
atomic states.
Having overcome the experimental difficulties, Hitchen, Thurgate, and Jennings
[53] collected high-resolution VLEED spectra from O-Cu(001) surfaces in energy
range from 6.0 to 16.0 eV, far lower than the energy of plasmon excitation. The
VLEED O-Cu(001) spectra displayed a dramatic intensity change that occurs at
oxygen exposure of 200–300 L. This dramatic change was related to the transition
from a c(2 × 2) to a (
√
2 × 2
√
2)R45° phase. Unfortunately, with either or the
combination of the two superstructures acceptable fit of the VLEED data from the
257
In 1993, Lederer et al. [44], Arvanitis et al. [48], Yokoyama et al. [49], asserted
that there are two phases during the O-Cu(001) reconstruction. One corresponds to
an unreconstructed state and the other a reconstructed phase. In the former case,
oxygen atom locates 0.8 Å above the first Cu layer to form an off-centered pyramid,
or a c(2 × 2)-O structure with oxygen shifting 0.1 Å away from the symmetry site
center. Upon reconstruction, the adsorbate locates 0.2 Å above the first Cu layer with
a MR (
√
2 × 2
√
2)R45°-O structure. In the two sequential phases, oxygen adsorbate
retains at the apical site of an off-centered pyramid and then at the apical site of a
distorted tetrahedron. The nature of the O-Cu bond was suggested as “predominantly
ionic for the first phase and more covalent for the reconstructed second phase.”
It became certain in 1996 when Fujita et al. [43] confirmed the bi-phase structures
with STM observations. In the first phase, oxygen sits in the near-center of the nextnearest-neighboring hollow site to form a pyramid at an oxygen exposure of 25 L
or lower. The precursor phase is composed of nanometric c(2 × 2)-O domains with
protruding boundaries. Upon increasing oxygen exposures, the c(2 × 2)-O evolves
into the (2
√
2 ×
√
2)R45°-O structure by every fourth row of Cu atoms missing. In
1998, Tanaka et al. [50] demonstrated that the two phases are reversible under the
bombardment of energetic Ar
+ beams.
In 1997, Sun [19], Sun and Bai [51] clarified that the MR formation results from the
process of oxide tetrahedron bond formation that breaks all bonds of the MR atoms
being evaporated without needing other perturbation. The two sequential phases
result from the transition from O
−1 to O
−2 under the specific Cu(001) geometric
condition [52].
Figure 13.1a shows the zigzag and the U-shape protruding domain boundaries
of the first O
−1 derived phase. Comparatively, the O
−2 derived “dumb-bell” shaped
protrusions in Fig. 13.1b bridge over the missing rows. Jensen et al. [45] interpreted
these protrusions as: “Pairing of Cu-O-Cu chains forms by displacing the Cu and/or
O atoms next to the missing row by about 0.35 Å towards the missing row; the OCu-O chain is linked by the delocalized antibonding states”. However, Sun affirmed
that the nonbonding electron lone pair “:” connects the O
−2 :Cu
p :O
−2 and the pairing
Cu
p
↔ Cu
p dipoles (p) crosses the missing row. The “↔” represents the repulsive
force between the dipoles [19].
From the STM image of Fig. 13.1b, the separation of the paired rows was estimated
at 2.9 ± 0.3 Å. The length of the bright spot was estimated at 5.1 Å; the height of
the bright spot was 0.45 Å; whereas, for the pure Cu(001) surface the protrusion
was about 0.3 Å in height [38]. These STM observations founded the H 2 O-like
tetrahedral models to specify the bond nature and its consequence on the individual
atomic states.
Having overcome the experimental difficulties, Hitchen, Thurgate, and Jennings
[53] collected high-resolution VLEED spectra from O-Cu(001) surfaces in energy
range from 6.0 to 16.0 eV, far lower than the energy of plasmon excitation. The
VLEED O-Cu(001) spectra displayed a dramatic intensity change that occurs at
oxygen exposure of 200–300 L. This dramatic change was related to the transition
from a c(2 × 2) to a (
√
2 × 2
√
2)R45° phase. Unfortunately, with either or the
combination of the two superstructures acceptable fit of the VLEED data from the