wien2triqs running

[blueway39]

Dear all.

Using the wien2triqs, I want to reproduce some results of SrVO3 as in paper - PRB 80, 085101(2009)

Python script for CTQMC solver is described in web site(http://ipht.cea.fr/triqs/doc/user_manual/wien2k/advanced.html)- 'A more advanced example', and I used that script for my dmft calculation. (named 'SrVO3.py')

Also, I referred 'SrVO3.indmftpr' file appeared at 'Introduction to dmftproj'.

All are same as the manual & web site, but I got an error message during 'run -qdmft' and calculation was stop.

(bottom part)

          time elapsed total : 281 seconds
        Solver Hybridization Expansion has ended.
        Total charge of impurity problem : 1.662963
        Mixing Sigma and G with factor 1.0
        DC for shell 0 and block up = 4.651852
        DC for shell 0 and block down = 4.651852
        DC energy for shell 0 = 2.20496597067
        forrtl: severe (24): end-of-file during read, unit 32, file                                                   /data1/joon/data/Wien2k/SrVO3/type01/DMFT/SrVO3_01/SrVO3_01.qdmft
        Image              PC                Routine            Line        Source
        lapw2              00000000005874ED  Unknown               Unknown  Unknown
        lapw2              0000000000585FF5  Unknown               Unknown  Unknown
        lapw2              00000000004D1CB9  Unknown               Unknown  Unknown
        lapw2              000000000048B3BA  Unknown               Unknown  Unknown
        lapw2              000000000048ABB0  Unknown               Unknown  Unknown
        lapw2              00000000004AFCFA  Unknown               Unknown  Unknown
        lapw2              00000000004118DC  qdmft_mp_readdata          56  qdmft_tmp_.F
        lapw2              00000000004640EC  MAIN__                    215  lapw2_tmp_.F
        lapw2              000000000040542C  Unknown               Unknown  Unknown
        libc.so.6          000000316E61D994  Unknown               Unknown  Unknown
        lapw2              0000000000405339  Unknown               Unknown  Unknown

I found out that SrVO3.qdmft is empty.

SrVO3.py is uploaded on web( https://github.com/blueway39/all/issues/1 )

I think there is problem in python script.

Please, give me an any advice.

[leopo]

lapw2 crushes because it cannot read your *.qdmft file, and that happens because actually you do not write it. You need to add few more lines at the end of the script, see http://ipht.cea.fr/triqs/doc/user_manual/wien2k/selfcons.html for details on how to do fully self-consistent calculations

[blueway39]

Thank you for your advice.

It works good with adding line in script.

[blueway39]

How long does it take the 'run -qdmft' calculation?

It is running over 350 iterations ( wien2k's cycle ) using above script in SrVO3.

And, just using ' run -qdmft -i N ' is good choice ?? (Is it reliable results?)

[aichhorn]

Welcome to the world of numerics! You are using a stochastic Solver for the solution of the AIM, so this will almost never stop automatically, stochastic fluctuations on the total energy will be larger than the 0.1 mRyd criterion of Wien2k. So you have to decide yourself, if your calculations are converged. You have to check, if the self-energy does not change from one loop to the next, also the other properties like chemical potential, impurity charge, and so on. We cannot give you a criterion that is universally valid. You have to decide yourself. Just to give you a hint, 350 iterations is too much. Normally, less then 100 are enough. But again, depends on the system, the parameters, and what you want to do.

[blueway39]

Oh, I should handle the iteration.

By the way, the material that I want to test the LDA+DMFT has two transition metals, and I read a line on web like below.

" In case of non-equivalent atoms, the correlation energy has to be calculated for all of them separately (FOR EXPERTS ONLY). "

Is there any document or tutorial to learn this case?