Surface reconstruction

from Wikipedia, the free encyclopedia
Surface reconstruction (schematic): The position and the symmetry of the atoms on the surface (red) are changed compared to the regular atomic lattice (blue) inside the solid.

In the case of an interface or surface of a crystal, surface reconstruction occurs when the atoms close to the surface are shifted from their positions in the spatial crystal lattice . In an ideal crystal the atoms are arranged in a regular lattice; if you mentally hop from one atom to another of the same type, the crystal lattice looks completely identical. On the surface of the real crystal, all the atoms are missing from the lattice sites on the outside. Since the atoms in the interior of the crystal are arranged regularly, this also applies to every flat surface; the basic structure is given by that of the volume crystal. If the atoms near the surface, i.e. in the top 1 or 2 layers, are shifted from their lattice positions in the interior of the crystal, we speak of surface reconstruction. If the atoms are only shifted perpendicular to the surface, one speaks of surface relaxation .

Surface reconstruction properties

Similar to the structure of three-dimensional crystals, surfaces form two-dimensional crystal structures . The structure of the non-reconstructed surface arises from the volume crystal structure and the crystal plane of the surface. In this surface the arrangement of the atoms corresponds to that in the corresponding crystal face of the bulk crystal, for example it has a square structure in a (001) plane of a crystal in the cubic crystal system . As with the volume crystal, Bravais grids provide a system for the various possible structures. There are a total of five two-dimensional Bravais grids .

A reconstruction has a superstructure when the size of the unit cell (EZ) is multiplied in one or both directions. If two differently reconstructed EZs of the unreconstructed surface alternate in one surface direction and the EZs are equal in the second direction, the size of the reconstructed EZ doubles in one direction, while it remains unchanged in the second direction a (2 × 1) superstructure of the surface. Non-integer superstructures can also result, for example a (√2, √2) superstructure in which the diagonals of the unreconstructed form the reconstructed unit cell.

The reason for the formation of surface reconstructions is the reduction in free energy . In many cases, a reconstructed surface also has the lowest total energy ( surface energy ) and is also the most favorable surface structure at absolute zero . On many materials, different surface structures can be produced under the same conditions, which can be energetically different. The different structures can be reached through different preparation methods.

Adsorbate atoms or molecules can change the surface reconstruction. Most of the time, the surface atoms rearrange themselves in such a way that an energetically favorable bond geometry is created for the adsorbate. Different adsorbate atoms lead to different surface structures, so the adsorption of hydrogen or oxygen on the (001) surface of tungsten , a surface with a square face-centered structure, leads to an increase or decrease in the transition temperature between the unreconstructed and the (√2, √2 ) -Tungsten surface. There does not have to be a transition temperature between different reconstructions of the same surfaces, so every reconstruction can also be metastable . At the same temperature, for example, depending on the treatment of the surface, different reconstructions of the same surface can be present.

Many superstructures on pure surfaces are abolished by adsorbates. For example, the unsaturated bonds on silicon surfaces can be saturated by hydrogen. The silicon atoms on the surface no longer combine to form dimers, which means that reconstructions with dimer formations are transferred back to surfaces without the resulting superstructures.

Determination of the surface structure

Various methods are available for determining the surface structure. Direct methods are the measurement of the surface with scanning microscopic instruments such as the scanning tunnel microscope or the atomic force microscope in atomic resolution. These methods determine the electron density or the altitude of the crystal atoms directly. With Low-Energy Electron Diffraction (LEED) one obtains diffraction images of the surface, from which one can infer the structure of the surface. The diffraction reflections of a (2x1) superstructure are half as far apart in one direction as one would expect from the unreconstructed surface, while in the other direction they are the expected distances. With LEED you can quickly determine that there is a superstructure and in which directions.

Typical reconstructions of some surfaces

Semiconductor surfaces

The surface reconstructions of semiconductor surfaces can usually be explained by the fact that the number of "cut" bonds (unsaturated bonds, English dangling bonds ) are minimized.

Si (100): When (mentally) cutting through a silicon crystal ( diamond lattice ) along the (100) plane, two bonds per silicon atom are broken. Pairs of adjacent silicon atoms bond to one another with one of the cut bonds and form so-called dimers. There remains only one unsaturated bond per silicon atom. The lattice is distorted in such a way that one atom is higher in the dimer, one is lower ( buckled dimer ); directly adjacent dimers are oriented in opposite directions. At room temperature, however, the dimers quickly change their orientation as a result of thermal excitations and appear symmetrical in the scanning tunneling microscope .

The stable surface of the Si (111) surface has a complex (7 × 7) reconstruction. The structure, the dimer adatom stacking fault (DAS) model, was proposed in 1985 by K. Takayanagi and co-workers, whereby measurements by Gerd Binnig and co-workers with the scanning tunneling microscope formed an important basis.

Metal surfaces

STM measurement of the reconstruction of the (100) face of a gold single crystal

With the surfaces of pure metals , reconstructions are less common than with semiconductors; the cubic face-centered precious metals of the 6th period , Ir , Pt and Au, are an exception . In these three metals, the (100) surfaces reconstruct and instead of the square lattice of the face-centered cubic structure, a hexagonally densely packed atomic layer forms on the surface. The (110) surfaces of these metals show a “ missing row ” reconstruction; every second close-packed row of atoms is missing. With gold also reconstructs the (111) surface; the top atomic layer is contracted (herringbone reconstruction ). These reconstructions are caused by the particularly low surface energy of hexagonal dense surfaces of these metals and a high tensile stress in the surface.

Non-conductor

In the case of non-conductors, in particular ion crystals and most oxides, a common reason for the formation of surface reconstructions is that the surfaces of macroscopic objects have to be largely electrically neutral (“charge compensated”). For example, an unreconstructed (111) surface of the NaCl lattice would be a polar surface, i.e. the top atomic layer would consist only of positive or only negative ions, which would lead to an energetically unfavorable, extremely high electric field . This can be avoided by a reconstruction in which some of the ions in the uppermost atomic layer are missing.

literature

  • K. Oura, VG Lifshits, AA Saranin, AV Zotov, M. Katayama: Surface Science: An Introduction . Springer-Verlag, Berlin 2003, ISBN 3-540-00545-5 .
  • Charles Kittel: Introduction to Solid State Physics . 14th edition. Oldenbourg Wissenschaftsverlag, Munich 2006, ISBN 3-486-57723-9 , p. 532 ( limited preview in the Google book search - original title: Introduction to Solid State Physics . Translated by Siegfried Hunklinger).
  • Andrew Zangwill: Physics at Surfaces . Cambridge University Press, Cambridge 1996, ISBN 0-521-34752-1 , pp. 207, 258, 259 .

Individual evidence

  1. a b Zangwill: Physics at Surfaces , 1996, p. 259
  2. ^ Zangwill: Physics at Surfaces , 1996, p. 96
  3. DJ Chadi: Atomic and Electronic Structures of Reconstructed Si (100) Surfaces . In: Physical Review Letters . tape 43 , no. 1 , 1979, p. 43-47 , doi : 10.1103 / PhysRevLett . 43.43 .
  4. K. Takayanagi, Y. Tanishiro, M. Takahashi, S. Takahashi: Structural analysis of Si (111) -7 × 7 by UHV transmission electron diffraction and microscopy . In: Journal of Vacuum Science & Technology A: Vacuum Surface Films . tape 3 , no. 3 , 1985, pp. 1502-1506 , doi : 10.1116 / 1.573160 .
  5. G. Binnig, H. Rohrer, Ch. Gerber, E. Weibel: 7 × 7 Reconstruction on Si (111) Resolved in Real Space . In: Physical Review Letters . tape 50 , no. 2 , 1983, p. 120-126 , doi : 10.1103 / PhysRevLett.50.120 .
  6. Jacek Goniakowski, Fabio Finocchi, Claudine Noguera: Polarity of oxide surfaces and nanostructures . In: Reports on Progress in Physics . tape 71 , no. 1 , 2008, p. 016501 , doi : 10.1088 / 0034-4885 / 71/1/016501 .