Good day everyone!

How to make a calculation in quantum espresso using hybrid functional? What should be considered?

My errors were the following:

Error in routine electrons (1): dexx is negative! Check that exxdiv_treatment is appropriate for the system, or ecutfock may be too low.

BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES = RANK 1 PID 32448 RUNNING AT n01p087 = KILLED BY SIGNAL: 9 (Killed)

My test run:

&CONTROL

calculation = 'scf'

prefix = 'si_hybrid'

outdir = './tmp'

pseudo_dir = '.' ! Update this path

verbosity = 'high'

tprnfor = .true.

tstress = .true.

etot_conv_thr = 1.0e-6

forc_conv_thr = 1.0e-4

/

&SYSTEM

ibrav = 2 ! Face-centered cubic (FCC) lattice

celldm(1) = 10.20 ! Lattice parameter in Bohr (for Si)

nat = 2 ! 2 atoms in primitive cell

ntyp = 1 ! 1 type of atom (Si)

ecutwfc = 44 ! Wavefunction cutoff (Ry)

ecutrho = 400 ! Charge density cutoff (Ry)

ecutfock = 400

occupations = 'smearing' ! Smearing for better convergence

smearing = 'gaussian' ! Gaussian smearing

degauss = 0.01 ! Smearing width (Ry)

input_dft = 'PBE0' ! Hybrid functional (PBE0)

exx_fraction = 0.25 ! Fraction of exact exchange (default for PBE0)

exxdiv_treatment = 'gygi-baldereschi' ! Treatment of EXX divergence

/

&ELECTRONS

! Use diagonalization for hybrid functionals (required for exact exchange)

diagonalization = 'cg'

electron_maxstep = 200

mixing_mode = 'plain'

mixing_beta = 0.4

conv_thr = 1.0e-10 ! Tighter convergence threshold

/

ATOMIC_SPECIES

Si 28.0855 Si.pbe-n-kjpaw_psl.1.0.0.UPF ! Use PBE pseudopotential (compatible with hybrid)

ATOMIC_POSITIONS {alat}

Si 0.00 0.00 0.00

Si 0.25 0.25 0.25

K_POINTS {automatic}

4 4 4 0 0 0 ! Dense k-grid for hybrid calculations