Atoms on surfaces often reorder into other structures that are different from the internal structure of a solid block of material (aka the bulk). They do this to minimize energy. For details consult the Wikipedia page and further resources.
Misleading experimental data
Current day UHV systems for nano-science need to be heated to above 200°C (often for hours to days) to drive out rest gas molecules (especially water) that are strongly adsorbed onto the walls.
Out of this reason many material surfaces that actually have sufficient activation energy barriers to prevent surface reconstruction at room temperature have only been observed in a reconstructed state. Having many images showing reconstructed (annealed) surfaces and view that sow unreconstructed ones can make it seem that almost all surfaces reconstruct which is not necessarily true.
Preventing / avoiding surface reconstruction
Since advanced mechanosynthesis will be done at room temperature (or more likely below) surfaces with activation energies far above room temperature will not reconstruct.
Just as it is the case with the prevention of surface diffusion one wants to avoid metallic bonds with weak directionality (low energy barriers against angular displacement) instead one wants strong covalent bonds (like in diamond / graphite) or ionic bonds.
Both single crystalline graphite (HOPG) and salt crystals have been imaged in unreconstructed states despite preceding severe heating.
In the case of the mechanosynthesis of diamond, lonsdaleite, moissanite and related structures one can prevent surface reconstruction by depassivating (hydrogen abstraction) only the spot where the next carbon moiety is planned to be deposited. Having everything hydrogen passivated in the final product helps keeping surface reconstruction from happening if the product gets heated.
In case of other gemstone like compounds that contain transition metal elements (like e.g. rutile TiO2 and hematite Fe2O3) there's a strong ionic component beside the covalent bonding component. (TODO: how much is known about surface reconstruction of these materials when not heated)
- Seamless covalent welding
- Surface diffusion
Section: Thermodynamics prevents one from having every atom at the place we want it - wrong for practical scales.
Common misconceptions about atomically precise manufacturing
- Surface passivation