For the purpose of theoretically studying my samples (Lithium ferrite) which are already prepared. I would like perform DFT. The problem I face here is that my samples have formulas like for example Li0.45−x/2Ni0.1Mn0.1CoxFe2.35−x/2O4 with x ranging from 0 to 0.1 in steps of 0.02 which are in fraction. How to go about drawing their structure in GaussView prior to running DFT on Gaussian software?
Hello Rajkumar Shanajaoba Singh
You could try to find the crystal structure and identify the unit of interest. You can draw or import the structures. You can calculate on a single unit (cluster) where you truncate the molecule, but you must first be careful about what you want to study and whether performing runs on such a cluster is meaningful. You can also use a unit cell with periodic boundary conditions to perform runs of your interests.
Also, search the literature because many solid-state compounds, as you mentioned, have fractional valencies but well-defined unit structures.
I hope this helps.
For varrying the fraction, you need to take supercell of you material.
Lithium ferrite appears to have a formula of LiFe5O8, which gives a Li to O ratio of 1:8, compared to your sample in which the ratio is 1:9. Do you have a good crystal structure? If so, you could build a large supercell of that structure, then randomly assign Ni, Mn, and Co to sites with fractional occupancy. You may have to try several variations to get the lowest energy structure. If you don't have a crystal structure, you might try starting with a large supercell of lithium ferrite [something like (LiFe5O8)10] and replacing some of the Fe sites with Ni, Mn, and Co to stay as close to your desired stoichiometry as possible.
- Badges
- Science topic
- Similar topics
- Engineering
- Petroleum Engineering
- Oil