As a follow up on my previous post, I let the calculation run to completion on a less dense k-grid and saw the following output:<br><br><blockquote style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;" class="gmail_quote">
ik 2 ibnd 17 linter: root not converged 0.748E+35<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.421E+23<br> Non-scf u_k: avg # of iterations =137.0<br> Non-scf Du_k: avg # of iterations =200.0<br>
<br> Dielectric constant from finite-differences<br><br> ( NaN NaN NaN )<br> ( NaN NaN NaN )<br> ( NaN NaN NaN )<br>
<br> Computing Second order response <br> kpoint 1 ibnd 17 pcgreen: root not converged NaN <br> kpoint 2 ibnd 17 pcgreen: root not converged NaN <br> kpoint 2 ibnd 17 pcgreen: root not converged NaN <br>
kpoint 2 ibnd 17 pcgreen: root not converged NaN <br> kpoint 2 ibnd 17 pcgreen: root not converged NaN <br> kpoint 2 ibnd 17 pcgreen: root not converged NaN <br> kpoint 2 ibnd 17 pcgreen: root not converged NaN <br>
<br><br> iter # 1 <a href="http://av.it">av.it</a>.: 147.2<br> thresh= 0.100E-01 alpha_mix = 0.100 |ddv_scf|^2 = NaN <br> kpoint 1 ibnd 17 pcgreen: root not converged NaN <br> kpoint 1 ibnd 17 pcgreen: root not converged NaN <br>
kpoint 1 ibnd 17 pcgreen: root not converged NaN <br> kpoint 1 ibnd 17 pcgreen: root not converged NaN <br> kpoint 1 ibnd 17 pcgreen: root not converged NaN <br> kpoint 1 ibnd 17 pcgreen: root not converged NaN <br>
kpoint 2 ibnd 17 pcgreen: root not converged NaN <br> kpoint 2 ibnd 17 pcgreen: root not converged NaN <br> kpoint 2 ibnd 17 pcgreen: root not converged NaN <br> kpoint 2 ibnd 17 pcgreen: root not converged NaN <br>
kpoint 2 ibnd 17 pcgreen: root not converged NaN <br> kpoint 2 ibnd 17 pcgreen: root not converged NaN <br><br><br> iter # 2 <a href="http://av.it">av.it</a>.: 200.0<br> thresh= 0.100E-01 alpha_mix = 0.100 |ddv_scf|^2 = NaN <br>
kpoint 1 ibnd 17 pcgreen: root not converged NaN <br> kpoint 1 ibnd 17 pcgreen: root not converged NaN <br> kpoint 1 ibnd 17 pcgreen: root not converged NaN <br><br><br></blockquote>and so on. The code crashes at the end with the error:<br>
<br><blockquote style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;" class="gmail_quote">%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br> from broyden : error # 1<br>
factorization<br> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br><br></blockquote><div><br>but obviously the problem occurs well before this crash. Again, I am trying to calculate the Raman intensity for a material with a small indirect overlap. Should this not be possible? I know that DFPT is applicable (without modification) only for systems with a clear distinction between occupied and valence states (in order to clearly specify what the projectors over the occupied and unoccupied manifolds I guess?). For the indirect overlap case with fixed occupancies I would imagine that this would be possible. <br>
<br>Anyway, just wanted to report on the results of my first post. If anyone knows what I can do to calculate the intensity or simply if it should or shouldn't work (and/or "why") I would greatly appreciate it.<br>
<br>Best,<br>Brad Malone<br>UC Berkeley<br><br> </div><div class="gmail_quote">On Sun, Nov 22, 2009 at 11:36 AM, Brad Malone <span dir="ltr"><<a href="mailto:brad.malone@gmail.com">brad.malone@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">Hi, I'm interested in calculating the Raman intensity for a material with a small indirect band overlap within LDA (~0.3 eV). As has been mentioned many times on this forum, Raman intensities cannot be calculated for a pure metal. But should I expect to be able to calculate the intensity for a material with a small, indirect band overlap? When I attempted to do so (using lraman=.true. in the ph.x calculation) I received the following messages:<br>
<br><blockquote><blockquote style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;" class="gmail_quote"> Computing Pc [DH,Drho] |psi> <br><br> Derivative coefficient: 0.001000 Threshold: 1.00E-12<br>
ik 1 ibnd 17 linter: root not converged 0.578E+08<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.763E+05<br> ik 1 ibnd 17 linter: root not converged 0.193E+11<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.119E+05<br>
ik 1 ibnd 17 linter: root not converged 0.595E+09<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.182E+05<br> ik 1 ibnd 17 linter: root not converged 0.142E+31<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.224E+24<br>
ik 1 ibnd 17 linter: root not converged 0.143E+30<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.170E+25<br> ik 1 ibnd 17 linter: root not converged 0.765E+28<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.151E+25<br>
ik 1 ibnd 17 linter: root not converged 0.643E+09<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.460E+05<br> ik 1 ibnd 17 linter: root not converged 0.619E+10<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.547E+07<br>
ik 1 ibnd 17 linter: root not converged 0.685E+09<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.810E+04<br> ik 1 ibnd 17 linter: root not converged 0.290E+34<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.848E+19<br>
ik 1 ibnd 17 linter: root not converged 0.132E+33<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.333E+18<br> ik 1 ibnd 17 linter: root not converged 0.155E+34<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.133E+20<br>
ik 1 ibnd 17 linter: root not converged 0.146E+06<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.442E+05<br> ik 1 ibnd 17 linter: root not converged 0.135E+07<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.169E+06<br>
ik 1 ibnd 17 linter: root not converged 0.567E+08<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.493E+06<br> ik 1 ibnd 17 linter: root not converged 0.292E+27<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.518E+23<br>
ik 1 ibnd 17 linter: root not converged 0.882E+28<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.215E+24<br> ik 1 ibnd 17 linter: root not converged 0.104E+29<br> kpoint 1 ibnd 17 pcgreen: root not converged 0.328E+23<br>
ik 2 ibnd 17 linter: root not converged 0.585E+21<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.363E+23<br> ik 2 ibnd 17 linter: root not converged 0.505E+21<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.119E+22<br>
ik 2 ibnd 17 linter: root not converged 0.117E+21<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.267E+22<br> ik 2 ibnd 17 linter: root not converged 0.159E+33<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.283E+27<br>
ik 2 ibnd 17 linter: root not converged 0.367E+32<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.660E+27<br> ik 2 ibnd 17 linter: root not converged 0.876E+32<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.670E+28<br>
ik 2 ibnd 17 linter: root not converged 0.249E+32<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.125E+28<br> ik 2 ibnd 17 linter: root not converged 0.348E+31<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.206E+26<br>
ik 2 ibnd 17 linter: root not converged 0.141E+30<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.561E+25<br> ik 2 ibnd 17 linter: root not converged 0.520E+36<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.990E+24<br>
ik 2 ibnd 17 linter: root not converged 0.140E+37<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.721E+24<br> ik 2 ibnd 17 linter: root not converged 0.836E+35<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.690E+25<br>
ik 2 ibnd 17 linter: root not converged 0.246E+30<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.540E+24<br> ik 2 ibnd 17 linter: root not converged 0.517E+29<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.415E+25<br>
ik 2 ibnd 17 linter: root not converged 0.133E+30<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.202E+26<br> ik 2 ibnd 17 linter: root not converged 0.570E+27<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.103E+21<br>
ik 2 ibnd 17 linter: root not converged 0.270E+29<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.451E+22<br> ik 2 ibnd 17 linter: root not converged 0.381E+29<br> kpoint 2 ibnd 17 pcgreen: root not converged 0.346E+22<br>
ik 3 ibnd 17 linter: root not converged 0.118E+34<br></blockquote></blockquote><br>and so on for each ik in the calculation. The code never crashed and I stopped it prematurely due to concern about these messages (seeing values on the order of 1E+34 is never calming). The code was also taking a very long time to calculate each ik. From what I've read in the Lazzeri and Mauri paper (PRL 90, 036401, 2003) some of the methods used to calculate the Raman tensor can take a lot of linear response calculations and so maybe I shouldn't expect it to go extremely fast. However, this root convergence worries me. The 17th band, which is the only band that this problem occurs for that I've seen, is the first conduction band of the system and is the band that indirectly overlaps with the top of the valence band. Any help or comments would be greatly appreciated!<br>
<br>Thanks,<br><font color="#888888">Brad Malone<br>UC Berkeley<br></font><br>P.S. I used 'fixed' occupancies in the SCF run to avoid the 'elec. field' error in the ph.x calculation. The fixed occupancy is what I would expect for this material in experiment, as the GW bandstructure for this material predicts it to be semiconducting.<br>
<br>
</blockquote></div><br>