search

TRIQS / triqs_0.x

DEPRECATED -- This is the repository of the older versions of TRIQS
Other
11 stars 9 forks source link

error in running dmftproj #141

Closed wasimrajamondal closed 11 years ago

wasimrajamondal commented 11 years ago

I want to calculate spectral function of BaRuO3. I want to treat Ru 4d as correlated and taking into account O2p orbitals. I have created case.almblm file and case.dmftsym file after running wien2k with non-spin-polarised and non-magnetic calculation. I have constructed case.indmftpr for BaRuO3 which is given below.

3 ! Nsort 1 1 3 ! Mult(Nsort) 2 ! lmax complex ! choice of angular harmonics 2 1 1 ! l included for each sort 1 0 0 0 -0.07 0.14 !Energy window relative to E_f 0.6931133736

when I am running dmftproj I am getting this error:

At line 173 of file /home/raja/TRIQS/TRIQS-master/pytriqs/Wien2k/dmftproj/dmftproj.f (unit = 7, file = 'BaRuO3.indmftpr') Fortran runtime error: Bad value during integer read

leopo commented 11 years ago

Your input file seems to be completely wrong, please read carefully the description of dmftproj input file at pp. 24-26 of the dmftrproj tutorial

http://ipht.cea.fr/triqs/doc/user_manual/_downloads/TutorialDmftproj.pdf

wasimrajamondal commented 11 years ago

Thanks for your help. I was confused with ce-gamma phase hubbad1 example. There was only one ce atoms. sorry for the mistake. Now I have the following question: In the input file for some basis is used as spherical harmonics and for some atoms basis are used as cubic.

How can I understand which basis should I use for a particular atom?