@Camilla In general, the lattice vectors are expressed in the CELL_PARAMETERS. Each row of the CELL_PARAMETERS matrix expresses one of the lattice vectors and each column of the CELL_PARAMETERS matrix expresses the Cartesian component of that lattice vector. For ibrav=1, the matrix of the CELL_PARAMETERS is a diagonal one.
However, I can see that you got a non-diagonal CELL_PARAMETERS matrix after the vc-relax calculation, which means you probably used cell_dofree='shape' or 'all'. So, the simple-cubic structure relaxed to a rhombohedral structure (I think). But if you calculate the angle, you will see it's 89.99492373 degrees. This is essentially a cubic structure
@Camilla could you rephrase your question here again? I think I am not getting exactly what you are asking?
@Earman No I didn't use cell_dofree at all as a variable. Normally I'm used to get always a diagonal CELL_PARAMETERS matrix. But this time the CELL_PARAMETERS matrix wasn't diagonal and I don't know why. And that's why I was confused how to get the lattice parameter from it in this case
@Earman I saw you comment about the creation of supercell. But actually the problem i'm facing is how to create a supercell from the input file without changing the cell parameters
@Camilla In rhombohedral crystal if the angles are 90 degrees, then it becomes a simple cubic lattice. That's why I don't think you have to worry about non-diagonal values since you have an almost perfect cubic crystal (89.99492373 degrees instead of 90 degrees)
@Camilla Let's examine a very simple case. Assume, you have a supercell with 1 unit cell that is of size 1*1*1. Also, you have 1 atom at (0,0,0) and lattice parameter a. Then to create a supercell of size 1*1*2, you need to change the lattice parameter to 2a, and then add 2 atoms. One at (0,0,0) and another at (a/2, a/2, a/2).
In a similar manner, to create a supercell you have to change the lattice parameters and number of atoms with their respective positions
If you could edit your question with a bit more detail about your input file and your expected supercell dimension, I might try to give an answer to your question
@Earman Yeah but the lattice parameter of the unit cell stills the same no matters how big is your unit cell
and the input file will always contain the lattice parameter of the unit cell even if you create a supercell. just the number of atoms increase, and more atoms coordinates will be generated in the input file.
@Camilla I am not sure about the working of CIF2CELL but in general, this shouldn't be the case. If you are trying to create a supercell twice the size of the unit cell, then the lattice parameter should be doubled. I usually create the input file manually
@Camilla would it be possible for you to share some of the relevant parts of the input file such as ibrav, celldm(i) / a,b,c,cosab etc.? so that I can understand the structure of your material
@Camilla okay. Then first of all do a 1*1*1 vc-relax calculation. Find the optimized lattice parameter. Let's see you find it X ang. Then for a 2*2*2 supercell, the lattice parameter will be a=2*X Ang (or celldm(1)=2*X*1.88973 because celldm is in bohr while a is in and). Also, you have to increase the number of atoms and their coordinate accordingly.
Then maybe you can do another vc-relax calculation just to check if your optimized lattice parameter remains the same i.e. 2*X. Then you can introduce the defects, and do another vc-relax calculation. Whatever you find (let's say Y), use this Y for later calculations
Just use twice the cell parameters. Or alternatively, use cif2cell to generate the supercell input file (atomic coordinates in alat). Then just edit the celldm(1) in your input file using a text editor.
I did something and I think it wokred I created a CIF file using my optimized lattice parameter using VESTA then I used CIF2CELL to create 2*2*2 supercell. As you said the lattice parameter increased
Excellent! I wasn't sure if it would work, but since it worked, do you want me to add it as an answer to your question? People with similar problem in future will easily find the solution from your question then
However my optimized lattice parameter was 6.160639979 Ang but after using it in VESTA to create a CIF file it became 6.16064 Ang . I wanted to save all the numbers after the comma, that was the problem that i'm facing exactly
@Camilla If VESTA is crunching the number, then what you can do maybe is to open the input file with a text editor after you are done with VESTA. And then add those numbers to the input file. The structure of the input file is pretty straightforward. You can find and replace "6.16064" with "6.160639979"
I have to go. Maybe we can talk later if there's still a problem.
@Earman I found that we can generate a supercell using Phonopy code from the input file. However I have some doubts That I need to clarify about its use.
I found in the internet a method to generate a supercell from the input file using Phonopy code. And I have a question if someone can clarify my doubts please. The method is as following:
We make an input file where the structure is defined by 'ibrav', 'nat', 'ntyp', 'ATOMIC_SPECIES', 'ATOMIC_POS...
I found in the internet a method to generate a supercell from the input file using Phonopy code. And I have a question if someone can clarify my doubts please. The method is as following:
We make an input file where the structure is defined by 'ibrav', 'nat', 'ntyp', 'ATOMIC_SPECIES', 'ATOMIC_POS...
I am trying to use Quantum Espresso in ASE, but I keep getting the following error. When I run pw.x directly from the terminal, it works, so Quantum-Espresso appears to be installed correctly and it's path is seen. I have no idea how to deal with this.
File "/home/lee/.local/lib/python3.10/site-...
Quantum ESPRESSO
Quantum ESPRESSO