Hey, I am no expert either but I will try to help with what I have learned so far.
1) Yeah, you can initiate the calculation with your original cell parameter and then go relaxing your cell and atoms. And then that final cell parameter can be used for single point calculation (scf) to find band gap which you can compare with your experimental data. 2) Cell parameter are also important and needed to relax. You can use cell_dofree="ibrav" in CELL card (check input file of pw.x). In this way your cell parameters will be relaxed keeping the lattice shape constant. 3) Choice of pseudopotential majorly depends on your XC functional. From what I have learned and observed, PBEsol is a good choice if one is focusing on lattice parameters and structure of cell. PBE is good for properties like bandgap, spin, etc. revPBE is good if you are dealing with Van Der Waal interactions, etc. You need to choose what you want to focus on and can refers to papers to check what suits you best. 4) So, in single point calculation output, at the end you can see that there will be two terms, total magnetization and absolute magnetisation. If there is zero value for total and non zero value for absolute it means there is magnetisation per atom but cancelling out each other to make it zero i.e., Antiferromagnetic. If both the values are non zero and same then all magnetic moments are aligned i.e. Ferromagnetic. 5) You can adjust your total magnetization using tot_magnetization. If you keep it 0, the calculation will be performed under this constraint and you might get an AFM. Similarly for FM, assign a value you find suitable. Then you can compare the energies from two different outputs. Also, since you are dealing with Transition metals, make sure you use DFT+U. You can either calculate the U parameter or can use that are available in already published works. Please let me know if you have any more doubts. Regards Yuvam Bhateja B.Tech. IIET Shibpur India On Mon, 28 Sep 2020, 2:40 am Ismail BENAICHA, <[email protected]> wrote: > dear Users, hope you are doing well. > my name is Ismail a Ph.D. Moroccan student. > Please I have some questions if you don't mind. > I am trying to study the magnetic behaviour of thin layers of ZnO doped > with transition metals (Co, Ni, Fe) using DFT calculation, I am a beginner > on DFT calculation using quantum espresso, my questions are: > 1-Sir/madam, if I want to combine an experimental result (E_gap from > UV-visble analysis and cell parameter from XRD analysis) with the DFT > calculation and compare them. Should I change the cell parameters of my > super cell in the inpult file with the ones from XRD analysis then run my > calculation (RELAX, SCF, BAND, DOS), or should I run my calculation > normally then compare the results that I get? > > 2- Sir/madam, If i want relax my super cell without losing my Ibrav > structure should I choose "vc-relax" or "relax" in ''calculation=" > > 3-Sir/madam, how do you choose the pseudopotential that you will be > working with and how do I choose the pseudopotential if I want to perform > a DFT calculation using GGA ? > > 4-Sir/madam, how I can decide that I have FM, AFM at the end of the > calculation, if I perform a spin-polarized calculation? > > 5- Sir/madam, where can I find energy difference DeltaE=E(AFM)-E(FM) in > the output of SCF file? > > > thank you in advance > -- > *Benaicha Ismail * > Ph.D student. Calculation and Modeling of Physical Properties of Solid > Materials at (LPMC-UIT. FS. Kénitra) > Email-1- :[email protected] > *Email-2-* :[email protected] > > > <https://www.facebook.com/profile.php?id=100011444409108> > <https://www.linkedin.com/in/ismail-benaicha-367782105/> . > _______________________________________________ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list [email protected] > https://lists.quantum-espresso.org/mailman/listinfo/users
_______________________________________________ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list [email protected] https://lists.quantum-espresso.org/mailman/listinfo/users
