Dear all Quantum Espresso users and developers,<br><br>I am not able to reproduce the zone cente phonons in ZrO2 calculated<br>in PRL 78, 21, 1997.<br><br>My results, without LO-TO splitting(can be calculated, once accoustical mode=0 cm-1 ) are:<br>
<br>T1u: 14.27 cm-1<span style="color: rgb(255, 102, 0);"> <---------not zero</span><br>T1u: 219.6 cm-1<br>T2g: 550.35cm-1 <br><br> I am not able to make the accoustical mode=0 cm-1.<br> Although the born charges and dielectric constants are <br>
ok with relative error~1%.<br><br><br><span style="color: rgb(255, 0, 0);">Input files:-----------------------------------------------</span><br><br><span style="color: rgb(51, 51, 255);">scf file:</span><br>&control<br>
calculation='scf', ! relax,self consistent calc.<br> verbosity='high',<br> restart_mode='from_scratch', ! 'from_scratch',<br> prefix='ZrO2',<br> wf_collect=.true., ! collect wave func in ./ZrO2.save<br>
tstress=.true.,<br> tprnfor=.true.,<br> nstep=100, ! default=50<br> dt=20, !10, ! default=20<br> etot_conv_thr=1.0D-5, ! default=1.0D-4<br>
forc_conv_thr=1.0D-5, !1.0D-4, default=1.0D-3<br> pseudo_dir = './',<br> outdir='./',<br> /<br><br> &system<br> ibrav = 2,<br> celldm(1)=9.67 <br> nat=3, ntyp=2, <br>
! nbnd= ! can calculate itself<br> ecutwfc=30, !50, !in Ry, no default, initially=60<br> ecutrho=350, !350 !300, ! 8*50=400<br> ! nr1=14,<br>
! nr2=14,<br> ! nr3=14,<br> ! nr1s=14,<br> ! nr2s=14,<br> ! nr3s=14,<br> ! nosym=.true.,<br> nspin=1, ! 1,2,4, "noncolin=.true."<br> ! input_dft= ! read from PP files.<br>
/<br><br> &electrons<br> mixing_beta = 0.2, ! 0.7<br> conv_thr = 1.0d-12, ! default=1.0d-06, ! 1.0d-12,1.0d-8<br> diagonalization=cg ! 'cg',!'cg', 'david', ! <br>
<br> /<br><br>ATOMIC_SPECIES <br> Zr 0.0000 Zr.pbe-nsp-van.UPF<br> O 0.0000 O.pbe-van_ak.UPF<br><br>ATOMIC_POSITIONS (crystal)<br>Zr 0.0 0.0 0.0<br>O 0.25 0.25 0.25<br>O 0.75 0.75 0.75 <br><br>K_POINTS (automatic)<br>
3 3 3 0 0 0 !8 8 8 0 0 0 <br><br><span style="color: rgb(51, 51, 255);">ph file</span>:----------------------------<br> &inputph<br> tr2_ph=1.0d-14, !1.0d-14, ! threshold for self consistency=1.0d-12(default)<br>
alpha_mix(1)=0.2, ! default=0.7<br> prefix='ZrO2',<br> iverbosity=1,<br> epsil=.true., <br> trans=.true., ! default=true ,<br> asr=.true.,<br> zue=.true., ! eff charges<br>
amass(1)=91.22400,<br> amass(2)=15.99940,<br> outdir='./',<br> fildyn='ZrO2.dynG', ! file where the dyn matrix is written<br><br> /<br>0.0 0.0 0.0<br><br>Effect of convergences (plane waves, k point sampling, augmentation functions)<br>
is checked. Forces are zero and press~1 kbar.<br><br>Bulk mod~233 GPa, expt~267Gpa => error~13 %.<br><br>Also,in above paper LDA pp was used, but i used USPP. <br><br>So in nutshell:---<br> i think USPP has to be modified or norm conserving pp should be used.<br>
<br>I need your comments over this.<br><br>Thanks!!<br><br>With Regards,<br>Sonu Kumar<br><br>Phd Student<br>Physics Department<br>Indian Institute of Technology <br>Delhi-110016, India<br>web:-<a href="http://www.iitd.ac.in/">http://www.iitd.ac.in/</a><br>