Many metal diboride compounds like $\ce{MgB2}$, $\ce{TiB2}$, and $\ce{ReB2}$ have laminated crystal structures with alternating sheets of metal atoms and boron atoms. If we polish the surface to ensure that only one type of atoms are facing outwards, will the two surfaces display different physical (such as friction coefficient, water contact angle) or chemical properties? enter image description here

  • $\begingroup$ It would be rather exfoliation, like with graphene, not polishing. $\endgroup$
    – Mithoron
    Jul 23, 2023 at 21:20
  • $\begingroup$ Not sure if the specific effect you want to see would be easy, but in general, different crystal faces will show different atoms on the surface and are known to have different chemical properties. $\endgroup$
    – matt_black
    Jul 23, 2023 at 23:12
  • 1
    $\begingroup$ Simple examples are the passivation of metals by oxide layers or antireflection layers on optical surfaces or an oil slick on a still water. or cell membrane structure. The finesse is getting the layers right. $\endgroup$
    – jimchmst
    Jul 23, 2023 at 23:52

3 Answers 3


A good question, but not sure it's possible, at least by polishing the given structure: if the boride layer is tightly bonded, I'd imagine it would be hard to remove such a layer, leaving the isolated metal sitting in the interstices.

However, there are ways to create a surface with a specific monolayer atop, and monatomic layers do have strong effects on the chemical and electronic properties of materials. For example,see Novel monatomic layer clusters for advanced catalysis materials and Electronic Structures of Polymorphic Layers of Borophane. If you have access, see also Monatomic Two-Dimensional Layers, Monoatomic Layer Electronics Constructed by Graphene and Boron Nitride Nanoribbons,


Undoubtedly yes. Giamello et al.[1] give an example involving a very subtle effect of this nature on a magnesium oxide (100) surface (meaning a square array of magnesium and oxygen atoms lying along the faces of cubic unut cells).

Molecular oxygen is adsorbed onto the magnesium ions and reduced to $\ce{O2^-}$ (through electron donation from the strongly basic oxide ions). If the surface layer is perfect, with no defects, then the superoxide ion is only loosely held to one magnesium ion and relatively mobile on the surface. But if there is an oxygen anion vacancy, the $\ce{O2^-}$ ion will be joined to all five surrounding magnesium ions and anchored tightly into the vacancy. The two different modes appear as different adsorption rnergies for atmospheric oxygen.


  1. E. Giamello, D. Murphy, E. Garrone, A. Zecchina (1993). "Formation of superoxide ions upon oxygen adsorption on magnesium-doped magnesium oxide: An EPR investigation". Spectrochimica Acta Part A: Molecular Spectroscopy, Volume 49, Issue 9, Pages 1323-1330, ISSN 0584-8539, https://doi.org/10.1016/0584-8539(93)80040-H.
  • $\begingroup$ If you are looking for a less subtle effect---imagine cleaving NaCl along the (110) plane two ways: 1) Leaving leaving the surface as Cl; 2) Leaving the surface as Na. Of course, such surfaces would be stable in a vacuum, $\endgroup$ Jul 24, 2023 at 7:53


The example most surface scientists should be familiar with is that of alpha silicon carbide or $\ce{\alpha-SiC}$.

Alpha silicon carbide (α-SiC) is the most commonly encountered polymorph, and is formed at temperatures greater than 1700 °C and has a hexagonal crystal structure (similar to Wurtzite).

It has a hexagonal crystal structure, and while the $[0001]$ face is terminated with silicon atoms, the $[000 \bar{1}]$ face is terminated with carbon.

Further, if you anneal in vacuum you can generate a layer or bilayer of graphene from the carbon-terminated side and use that as a substrate for growing additional hexagonal 2D materials like TMDs.

If you have a c-axis polished $\ce{\alpha-SiC}$ crystal, the challenge is then to tell which side is the "silicon side" and which is the "carbon". There are several "tricks" and just for an example here's a patent Method for distinguishing Si surface from C surface of SiC (silicon carbide) wafer CN103630708A

Image from Role of structure of C-terminated 4H-SiC($000 \bar{1}$) surface in growth of graphene layers: Transmission electron microscopy and density functional theory studies (also DOI:10.1103/PhysRevB.85.045426)

from "Role of structure of C-terminated 4H-SiC(000-1) surface in growth of graphene layers: Transmission electron microscopy and density functional theory studies" http://dx.doi.org/10.1103/PhysRevB.85.045426

  • $\begingroup$ is it impossible for a crystal to form with two of the same side? how does one side know what the other is doing? $\endgroup$
    – user253751
    Jul 26, 2023 at 9:33
  • $\begingroup$ @user253751 Yes that drives me crazy too! I am not a chemist but here's my thinking. If we call the vertical direction in the images the c-axis, the tetragonal carbon (four bonds) has one bond parallel to the c-axis and the other three almost sideways to adjacent silicon atoms. The carbon can tolerate one "free" bond (that will connect to hydrogen or water or whatever sticks to it as soon as you take it out of vacuum) but it will not tolerate three "free" bonds. $\endgroup$
    – uhoh
    Jul 26, 2023 at 9:39
  • 1
    $\begingroup$ @user253751 Carbon atoms left in that state will move around, cluster, form graphene or do something else, but the silicon can't hang on to a carbon with one bond. So the ""building blocks are the almost flat layer of carbon and silicon triply bonded to each other, each with a single dangling bond in opposite directions. That unit seems inseparable. $\endgroup$
    – uhoh
    Jul 26, 2023 at 9:41
  • 1
    $\begingroup$ that makes sense; it wouldn't apply to all crystals, just this specific one $\endgroup$
    – user253751
    Jul 26, 2023 at 9:44
  • $\begingroup$ @user253751 yes exactly; there will probably be a whole family of materials that do this, but they will probably be hexagonal and have this kind of strong in-plane, weaker out-of-plane, buckled kind of structure. $\endgroup$
    – uhoh
    Jul 26, 2023 at 9:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.