gamma-Al2O3 have a same structure with MgAl2O4. The difference is that the gamma-Al2O3 has one-ninth of the Al sites are vacant, and it remains controversially where the Al cancancies locate.

gamma-Al2O3 have a same structure with MgAl2O4. The difference is that the gamma-Al2O3 has one-ninth of the Al sites are vacant, and it remains controversially where the Al cancancies locate.

Hello,

as far as I know gamma-Al2O3 is a defect spinel structure. Therefore the unit cell is highly probably cubic (but at the moment I dont have access to the crystal structure of gamma-alumina). In your drawing the unit cell seems to be at least not cubic. My advice is to look for crystal structure data of gamma-alumina and then set up the 110 surface in an appropriate drawing program.

Regards,

Wolfgang

Hello,

I tried to draw a 110 plane inside the unit cell of gamma-alumina. Maybe it helps.

Regards,

Wolfgang

Well done, Wolfgang! You hit the nail on the head!

I took gamma-Al2O3 from the ICSD data base and selected the phase with space group Fd-3m. You can slice this structure according to the (110) plane. You obtain an alternating sequence of thin slices looking like the two fig. below. In these fig., large spheres are O and small sphere are Al

alternatively, your can select a larger slice combining the two thin slices above and get the following

Please consider that {110} is related to the definition of the unit cell vectors and therefore of the crystal system. I am not absolutely sure which of the 1-1-0 combinations you are talking about since your cell in the CIF file is monoclinic, i.e. you have different combinations: {110}, {011} (both with a multiplicity of 4), and {101}, {-101} (both with a multiplicity of 2). All of these planes should deliver a different "surface".

Of course, you have to ask yourself whether the monoclinic CIF file matches your requirements. It might be that you are really interested in {110} indexed as cubic phase. Then the descriptions given by the upper comments are well done, although I often prefer like Wolfgang a perpendicular presentation in order to see the stacking. However, if you need to see the distribution within the plane Gervais presentation is very useful. But again: take into account that the indexing is related to a certain basis vector description. If you change this, also the plane looks different.

A last comment: Please always use the international convention of lattice plane indexing (Miller indexing): either (hkl) for a single plane or {hkl} for the entity of all symmetry equivalent planes. A simple use of thee numbers is inacceptable, even if you are adding "plane". If you use (hkl) or {hkl} you don't need to add "plane" since this the description already tells everybody. 

Hello, everyone, I am sorry for my late response.  Firstly, thanks for all of your enthusiastic replies. Recently I am searching literature for the crystal structure of  gamma-alumina. From published papers I find two crystals, one is cubic Fm-3m and the other has P21/m space group.  I present them as below

With the second one by Digne et al., I built the (110) by setting U= [0 0 1] and V=[1 -1 0] as below. These pictures are top and side views of the supercell I obtained. I am not sure if this model is reasonable. 

 I don't think it's right, many papers said the (110) surface area is "8.40 × 8.07 × 19.17 A^2". 

I don't know how to build the (110) surface too, if you know how to do it, can you share your experience with us? thx

"The unit cell for this structure is rotated 45 degrees relative to the standard FCC unit cell (eg. Zhou and Snyder, 1991). As a result, the (010) surface of this unit cell corresponds to the (110) surface of the FCC unit cell". You may check the following link  for response to a similar inquiry on VAP forum:

http://listserv.fysik.dtu.dk/pipermail/ase-users/2014-March/002144.html

HI Jian, do you have a citation or know the authors of the gamma-Al2O3 cif file you attached to the question.