is your molecule metallic ? what does this mean ? what is the density of state at the Fermi energy ? stefano
Fanshun-Meng wrote: > Dear ESPRESSO Users: > > these days, when i used pw.x and ph.x to find the phonon frequencies of a > molecular at gamma point, met a serious problem: > if I use "occupied = smearing" in SCF calculation, a part of the ph.x > output file as follows: > \# Pert. # 1: Fermi energy shift (Ry) = 0.1232E+13 0.1115E+11 > > if I use "occupied = tetrahedra" in SCF calculation, a part of the ph.x > output file as follows: > # Pert. # 1: Fermi energy shift (Ry) = 0.4332E+44 0.1115E+47 > > no matter which occupied methods were used, why the Fermi energy shift is so > large(E11 or E45)? > or I make a mistake? > thank you! > > my SCF input file: > > &CONTROL > calculation = 'scf' , > outdir = './result/' , > wfcdir = './result/' , > pseudo_dir = './espresso-4.1.2/pseudo/' , > prefix = 'ag2h2s' , > nstep = 300 , > / > &SYSTEM > ibrav = 0, > celldm(1) = 28.343722662, > nat = 5, > ntyp = 3, > ecutwfc = 40 , > ecutrho = 320 , > occupations = 'tetrahedra' , > degauss = 0.03 , > smearing = 'fermi-dirac' , > / > &ELECTRONS > / > CELL_PARAMETERS cubic > 1.000000000 0.000000000 0.000000000 > 0.000000000 1.000000000 0.000000000 > 0.000000000 0.000000000 1.000000000 > ATOMIC_SPECIES > Ag 107.86800 Ag.pbe-d-rrkjus.UPF > S 32.07000 S.pbe-van_bm.UPF > H 1.00800 H.pbe-van_bm.UPF > ATOMIC_POSITIONS angstrom > Ag 6.783586392 7.506654156 7.504688044 > Ag 9.369904699 7.525335176 7.518919416 > S 4.245485150 7.495289160 7.482481411 > H 4.063583478 7.391786920 8.822219040 > H 4.080059280 8.840841587 7.518732089 > K_POINTS automatic > 1 1 1 0 0 0 > my phonons PH.X input file: > > phono of h2s at Gamma > &INPUTPH > outdir = ./result/' , > prefix = 'ag2h2s' , > fildyn = 'ag2h2s.dynG' , > ldisp = .false., > trans = .true., > tr2_ph = 1.0E-14 , > / > 0.0 0.0 0.0 > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum >
