I'm looking to do some image simulations on the effect of an amorphous layer on top of my samples. For this I need an atomic structure input file.
Is there are relatively simple way to produce such amorphous structures from the crystalline structure?
You could try to "add" some thermal vibrations, however, I don't think this would helpp that much. If you insert a "crystal structure" you immediately implement a translation symmetry which means...a crystalline phase and no amorphous structure. You could increase the volume of the translation symmetry, i.e. generate a multiple unit cell volume and the shuffle each position and reduce the symmetry to P 1 and increase the Debye-Waller factor. Then you still have a translation symmetry....
I mean, as long as you not need some real physical simulations but only an image of a non-translational symmetric structure it might be OK.
The local order in amorphous is similar to that found in the crystalline phase. It is of the order of 1 nm. When neighboring sets of atoms having this local order interact, they start forming small crystallites. Therefore, it is possible to construct a model for the amorphous phase by starting from the crystallography of the corresponding crystalline phase. The crystallite size in will the model should be of the order of 1 nm. It is necessary to introduce microstrain for modeling the displacement of the atoms from their ideal crystallographic positions. I tested this method by modeling amorphous silica by using the crystallography of crystobalite and it works well.
The above idea is not bad, my recent experience by modeling atomic clusters shows, clearly, the importance of this local order: atoms do not arrange at random, they follow a well defined local order depending of the atom types in the cluster.
- Badges
- Science topic
- Similar topics
- Computer Science
- Images