Atomic-resolution scanning tunnelling microscopic studies were made of porous surfaces after vacuum heat-treatment at 850C and above. The (100) surfaces exhibited some 2 x 1 reconstruction, as in the case of monocrystalline (100) surfaces, but (111) samples did not exhibit any clear reconstruction. This was thought to be because the column tops were too small for large structures, such as 7 x 7, to form. The atomic planes on the column tops were parallel to the substrate, for both (100) and (111) samples; in agreement with low-energy electron diffraction studies.

Atomically Resolved Structure of Vacuum-Heated Porous Silicon W.Li, I.Andrienko, D.Haneman: Journal of Physics - Condensed Matter, 2000, 12[15], L269-74