Page 1 of 1
Lda+u band structure calculation
Posted: Fri Oct 15, 2010 2:43 pm
by albayrouni
I want to calculate the electronic band structure of iron using the LDA+U method in a paw framework. Can you tell me please what are the input variables that are specific to the paw framework and that must be added in the two datasets : scf and non scf cycles.
Thank you in advance.
Re: Lda+u band structure calculation
Posted: Fri Oct 15, 2010 7:07 pm
by blackburn
Hi,
There is a tutorial about LDA+U that explains the variable you need.
http://www.abinit.org/documentation/hel ... _ldau.htmlYou don't need to specify anything for the PAW, ABINIT detects it automatically with the pseudopotential. For the LDA+U, you need to specify (in scf and/or non-scf):
usepawu 1
lpawu 3 -1 -1 #this is an example in which you add U on the f orbitals on the first type of atom
upawu 4 0 0 eV #value of U
jpawu 0.4 0 0 eV #value of J
Simon
Re: Lda+u band structure calculation
Posted: Fri Oct 15, 2010 10:18 pm
by albayrouni
wfeelb.in- input file of band structure calculation
- (1.23 KiB) Downloaded 270 times
Thank you for the answer but I have a problem. I first calculated the cell parameter of Fe using the LDA+U method and I found a=5.435 B. Then, Iused this value of a with the initial density matrix that gave me this value to generate the band structure. Unfortunately, the eigen values were wrong in comparison with other calculations. Can you help me please.
Thank you in advance.
Albayrouni