Dear Quantum Espresso users, I'm trying to calculate phonons on cubic SrTiO3 by using ph.x (v.4.3.2). I relaxed the whole structure (vc-relax) and got reasonable lattice and internal parameters (experimental 3.905 A, calculated 3.92(PBE)). Afterwards I performed a scf calculations with this structure and then used the phonon code. The problem is that I'm obtaining mostly negative frequencies. I know that cubic STO is unstable, but negative values should only appear around GAMMA, M and R symmetry points. I've also done this calculation by using the frozen phonon method and in this case results are in good agreement with the literature. But the frequencies I got from the ph.x calculation seem to be completely wrong. I've tried changing parameters, using a dense k-point grid and a high cutoff, but results are still wrong. I don't have any clue about what is going wrong, could anybody help? I'm copying the input files below. Thank you in advance for your help. Best,
- Alejandro The relaxed structure: &control calculation='scf', restart_mode='from_scratch', pseudo_dir = '/global/homes/r/rebola/pseudo/', outdir='/scratch/scratchdirs/rebola/', prefix = 'sto', / &system ibrav=0, celldm(1)=7.46252850, nat=5, ntyp=3, ecutwfc=40.0, ecutrho = 400, / &electrons conv_thr = 1.0d-8 mixing_beta = 0.7 / ATOMIC_SPECIES Sr 87.62 Sr.pbe-nsp-van.UPF Ti 47.867 Ti.pbe-sp-van_ak.UPF O 15.9994 O.pbe-van_gipaw.UPF CELL_PARAMETERS (alat= 7.46252850) 0.993935603 0.000000000 0.000000000 0.000000000 0.993935603 0.000000000 0.000000000 0.000000000 0.993935603 ATOMIC_POSITIONS (alat) Sr 0.000000000 0.000000000 0.000000000 Ti 0.496967802 0.496967802 0.496967802 O 0.000000000 0.496967802 0.496967802 O 0.496967802 0.000000000 0.496967802 O 0.496967802 0.496967802 0.000000000 K_POINTS (automatic) 4 4 4 0 0 0 The phonon input file: phonons of STO &inputph tr2_ph=1.0d-14, ldisp=.true., nq1=2, nq2=2, nq3=2 amass(1)=87.62, amass(2)=47.867, amass(3)=15.9994, epsil=.true. prefix='sto', outdir='/scratch/scratchdirs/rebola/', fildyn='sto.dyn', / ~