Substrate unreconstructed surface

In essence the adstoms simulate the missing half of the substrate. These impurity-stabilized unreconstructed surfaces [e.g., Pt(lOO) and Au(lOO)] have the structure known for the other stable clean unreconstructed metal surfaces. 

The top layer of the surface reconstructs to LEED/6/ form a compact hexagonal surface, with 6/5 the density of unreconstructed surface. The layer spacing expands by 14.6 5.2% from the bulk value of 1.92 A. Some top-layer atoms are buckled outward by up to an additional 0.2 0.02 A so the hexagonal layer can fit the square layer below. This is 1/2 to 2/3 of the buckling required to have top layer atoms in hard-sphere contact with all substrate atoms. 

An alternative to these simple, more or less unspecific, models is the use of ab initio electronic structure calculations, in a similar spirit as in the corresponding water-metal models [40, 49]. Most studies so far have been performed with the cluster model approach [114]. In this approach, a finite number of atoms is chosen to model a local site on the crystal surface. Usually, the geometry of the substrate is kept fixed. The size of the substrate cluster varies between 4 and 20 metal atoms. The geometrical arrangement of the surface atoms is usually taken from the unreconstructed surface. 

It is useful to define the tenns coverage and monolayer for adsorbed layers, since different conventions are used in the literature. The surface coverage measures the two-dimensional density of adsorbates. The most connnon definition of coverage sets it to be equal to one monolayer (1 ML) when each two-dimensional surface unit cell of the unreconstructed substrate is occupied by one adsorbate (the adsorbate may be an atom or a molecule). Thus, an overlayer with a coverage of 1 ML has as many atoms (or molecules) as does the outennost single atomic layer of the substrate. 

For a compound semiconductor to be useful as a substrate in studies of electrodeposition, it is desirable that clean, unreconstructed, stoichiometric surfaces be formed in solution prior to electrodeposition. For CdTe, the logical starting point is the standard wet chemical etch used in industry, a 1-5% Brj methanol solution. A CdTe(lll) crystal prepared in this way was transferred directly into the UHV-EC instrument (Fig. 39) and examined [391]. Figure 66B is an Auger spectrum of the CdTe surface after a 3-minute etch in a 1% Br2 methanol solution. Transitions for Cd and Te are clearly visible at 380 and 480 eV, respectively, as well as a small feature due to Br at 100 eV. No FEED pattern was visible, however. As described previously, a layer of solution is generally withdrawn with the crystal as it is dragged (emersed) from solution (the emersion layer). After all the solvent has evaporated, the surface is left with a coating composed of the... 

The surface coverage of an adsorbate is another important parameter in ordering. We shall use the common definition of coverage where one monolayer corresponds to one adsorbate atom or molecule for each unit cell of the clean, unreconstructed substrate surface. Thus, if an adsorbed undissociated carbon monoxide molecule bonds to alternating top-layer metal atoms exposed at the Ni(100) surface, we have a coverage of a half monolayer. 

Figure 2.7 In situ STM images illustrating the influence of the electrode potential on the surface structure of a Au(lOO) substrate 12.10]. System Au(100)/10 M H2SO4 at T = 298 K (a) and (c) = - 200 mV vs. SCE (b) and (d) E = 300 mV. Images (a) and (b) show atomic resolution of reconstructed (rows) and unreconstructed surfeces (c) and (d) represent larger scans emphasizing reconstruction rows and monoatomically thick islands, respectively. Reprintetl by permission of Kluwer Academic Publishers.

Several of these adsorbates, while forming surface alloys at low coverage on Cu 100 exhibit a tendancy to undergo de-alloying transitions. Dealloying refers to a reversal of the surface alloy formation process and consists of either a gradual or abrupt reduction of the concentration of adsorbate located in substitutional sites in the first substrate layer and formation of an overlayer structure above an unreconstructed substrate. Two systems have been studied in some detail Cu 100 /PbandCu 100 /Bi. 

For metal particles on ionic substrates one can differentiate between interactions with non-polar (stoichiometric) and polar surfaces. The former surfaces are represented by essentially unrelaxed (001) facets of oxides with rock-salt structure, such as MgO(OOl) which is widely utilized as metal support and was addressed in the previous subsection. Oxide supports exhibiting polar surfaces are also common. Clean polar surfaces are unstable and thus difficult to prepare unreconstructed, dehydroxylated, and free of defects [88]. Corundum, a-Al203, is a prototype of such metal oxides. Its most stable Al-terminated (0001) surface... 

The coverage of adsorbates on a given substrate is usually reported in monolayers (ML). Most often, 1 ML is defined as the number of atoms in the outermost atomic layer of the unreconstructed, i.e. bulk-terminated, substrate. Sometimes, however, 1 ML is defined as the maximum number of adsorbate atoms that can stick to a particular surface, which is termed the saturation coverage. The saturation coverage can be much smaller... 

Figure Bl.21.1 shows a number of other clean unreconstructed low-Miller-index surfaces. Most surfaces studied in surface science have low Miller indices, like (111), (110) and (100). These planes correspond to relatively close-packed surfaces that are atomically rather smooth. With fee materials, the (111) surface is the densest and smoothest, followed by the (100) surface the (110) surface is somewhat more open , in the sense that an additional atom with the same or smaller diameter can bond directly to an atom in the second substrate layer. For the hexagonal close-packed (hep) materials, the (0001) surface is very similar to the fee (111) surface the difference only occurs deeper into the surface, namely in the fashion of stacking of the hexagonal close-packed monolayers onto each other (ABABAB.. . versus ABCABC.. ., in the convenient layerstacking notation). The hep (1010) surface resembles the fee (110) surface to some extent, in that it also...

The maximum PI in the current vs. potential curve correlates with the substrate surface transition. The difference in zero charge potentials between the reconstructed (px. 3) and the unreconstructed (1x1) gold surfaces gives rise to an additional current contribution due to double layer charging. Figure ID shows the ideal terminated (1x1) surface with atomic resolution. Around 0.80 V a disorder/order transition takes place within the layer of adsorbed (hydrogen-)sulfate ions [27-30]. The forma-tion/dissolution of the so-called /3 x. 7) overlayer correlates with the... 

The geometric stmcture is used conventionally for identification of a particular metal/semiconductor surface phase. The surface structures are usually labeled in accordance with their periodicity with respect to the underlying semiconductor crystal plane. Two methods for the description of the two-dimensional lattices are used conventionally. The first one was proposed by Park and Madden [68P] and it consists in the determination of the matrix which establishes a hnk between the basic translation vectors of the surface under consideration and those of the ideal (unreconstructed) substrate sirrface. That is, if a and b are the basic translation vectors of the substrate lattice, while and As are the basic translation vectors of the surface phase, than they can be linked by the equations... 

The second method for the description of the two-dimensional lattices was proposed by Wood [64W]. This method is the most-used one now though it is less versatile than the above matrix notation. In the Wood description, the ratio of the siuface periods and those of the unreconstructed substrate are given as well as an angle of rotation which makes the unit mesh of the surface to be aligned with the basic translation vectors of the substrate. That is, if a certain adsorbate A induces on the B(hkl) surface the reconstruction with the basic translation vectors of... 

---