Dear Lorenzo, > > I think this is unnormal. > Why do you think it is unnormal? In high-symmetry materials and q-points it > is perfectly normal to have onyl a few non-zero terms.
I think the value of matrix element should not be so large. I also check elph.0.00000.XXX.XX files, which also have very large value or illogical output. For example, for G point, part results of elph.0.0000 0.00000 0.0000 file is as follows. 0.000000 0.000000 0.000000 10 18 -0.126332E-05 -0.620414E-06 -0.154679E-06 0.107608E-05 0.147451E-05 0.169464E-05 0.408894E-05 0.645471E-05 0.698396E-05 0.107719E-04 0.144215E-04 0.183331E-04 0.194042E-04 0.296271E-04 0.410801E-04 0.831798E-04 0.131738E-03 0.169451E-03 Gaussian Broadening: 0.005 Ry, ngauss= 0 DOS = 6.629818 states/spin/Ry/Unit Cell at Ef= 10.374777 eV lambda( 1)=******** gamma=******** GHz lambda( 2)=******** gamma=******** GHz lambda( 3)=******** gamma=******** GHz lambda( 4)= 0.4250 gamma= 31.34 GHz lambda( 5)= 0.0505 gamma= 5.11 GHz lambda( 6)= 0.0273 gamma= 3.16 GHz lambda( 7)=601.8197 gamma=******** GHz lambda( 8)=350.6821 gamma=******** GHz ................. You can see that many star happen. Also I noticed that values of some dynamical matrix element were connected together as presented as below. Dynamical Matrix in cartesian axes q = ( -0.310971568 -0.136778208 0.258285840 ) ------ 0.00174672 10.51302809 (see here) 0.00037302-10.51496752 -0.00002975 10.51477879 (see here) 0.00037302-10.51496752 -0.00108098 10.51459634 -0.00012917-10.51442422 -0.00002975 10.51477879 (see here) -0.00012917-10.51442422 -0.00025486 10.51457255 1 3 -0.00246965 10.51720967 0.00068821-10.51300824 0.00012537 10.51051515 0.00049193-10.51388748 0.00159857 10.51472824 -0.00196467-10.51474139 0.00064832 10.51324607 0.00147879-10.51444247 0.00078773 10.51321004 1 4 10.51411329 0.00042246 -10.51421211 -0.00161006 10.51191958 -0.00028359 -10.51392554 0.00013895 10.51339735 0.00115788 -10.51875627 0.00071854 10.51353792 0.00004149 -10.51165045 -0.00113263 10.51466247 -0.00032748 1 5 -0.00105101 10.51151308 -0.00097114-10.51337506 -0.00087478 10.51823862 -0.00093486-10.51426146 0.00079562 10.51524911 0.00088042-10.51220062 -0.00008398 10.51493606 -0.00145129-10.51582557 -0.00120748 10.51479141 .............................. > > > > For example, at G point, It is > > > > 5.000000000000000E-003 0.762532099281104 6.62981783474504 > What are these numbers? This is contents of a2Fq2r.51-61 files > >Do Your phonon dispersion look ok in the first place? The phonon spectrum obtained by using calculated *.dyn1, *.dyn2,.... files is reasonable. > Did you actually compute the el-ph coupling and found it to be uncorrect? Yes. I obtained elph.*** files. Many outputted gamma values in elph.**.**.** files are uncorrect. Thanks a lot. Best wishes, Yanling Institute of solid state physics, cas, China. > > Dear all, > > I calculate el-ph interaction using espresso-5.0.1 package. My input file is > Ok, which is confirmed by Xcrysden code. I got a2fq2r.* file at each q point > selected. I noticed that large dynamical matrix elements were obtained, while > all elements are zero at some point. I think this is unnormal. > > For example, at G point, It is > > 5.000000000000000E-003 0.762532099281104 6.62981783474504 > 1 > > Dynamical Matrix in cartesian axes > > q = ( 0.000000000 0.000000000 0.000000000 ) > > 1 1 > 137.60553244 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 137.79644417 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 137.58884974 0.00000000 > 1 2 > 137.54385366 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 137.31507421 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 137.52874532 0.00000000 > 1 3 > 137.50539357 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 137.54339987 0.00000000 0.02295830 0.00000000 > 0.00000000 0.00000000 0.01311991 0.00000000 137.55061180 0.00000000 > 1 4 > 137.58193625 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 137.55989359 0.00000000 0.02824413 0.00000000 > 0.00000000 0.00000000 0.01505453 0.00000000 137.55171888 0.00000000 > 1 5 > 137.50539357 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 137.54339987 0.00000000 -0.02295830 0.00000000 > 0.00000000 0.00000000 -0.01311991 0.00000000 137.55061180 0.00000000 > ................ > ............... > > At second point calculated, it is > 5.000000000000000E-003 0.762532099268857 6.62981783246344 > 2 > > Dynamical Matrix in cartesian axes > > q = ( 0.000000000 0.000000000 0.258285840 ) > > 1 1 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 1 2 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 1 3 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 1 4 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 1 5 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 > ......................... > ......................... > > DO you meet this case? Could you tell me the reason or how to correct it? > > Part parameters in fit.in file are > &system > ibrav= 0, celldm(1)=9.37471363, nat= 6, ntyp= 2,nbnd = 36, > ecutwfc =50.0, ecutrho = 500, occupations='smearing',smearing='mp', > degauss=0.04, la2F=.true. > / > &electrons > conv_thr = 1.0d-11 > mixing_beta = 0.4 > / > ................ > part parameters in ph.in file: > > .Electron-phonon coefficients for ss > &inputph > tr2_ph=1.0d-14, > prefix='lll', > alpha_mix(1)=0.4, > fildvscf='**dv', > outdir='./tmp/', > fildyn='111.dyn', > electron_phonon='interpolated', > trans=.true., > ldisp=.true. > nq1=4, nq2=4, nq3=4 > / > > > Many thanks > > Yanling > > Institute of solid state physics, P.R. China. > > > ------ > ISSP -------------- next part -------------- An HTML attachment was scrubbed... URL: http://pwscf.org/pipermail/pw_forum/attachments/20121122/f3f4d19a/attachment.html