[Wien] How to specify intraband contributions in case.injoint, case.inkram

[Jian-Xin Zhu]

Dear Prof. Blaha,

I figured out where I made a mistake to run "x kram".
In the file of case.inkram, I need to put in the corresponding number  
of Gammas for Drude terms (when I want to add intraband  
contributions), which was
instructed clearly in UG. So for two columns, after I fed in 0.1 0.1  
on the 5th line in case.inkram, the program
ran properly.

However, I still have two questions ---

1/
Why is case.intra always empty after I run x joint with SWITCH=6?
I noticed after running x joint with switch=6, the case.joint stores  
the real and imaginary parts of epsilon.
I suspect these are intraband contributions.
Surely when I run "x joint" again with switch=4, this set of results  
will be erased.
So I guess case.intra might be a placeholder for them (that is, I can  
copy case.joint to case.intra
and then start running "x joint" with switch=4). But this set of  
results, as I now store in case.intra, is not needed
when I run "x kram".

2/
 From the archive, you once made the following remark in relation to a  
question raised by Dr. Shima Kadkhodazadeh--
 >>PS: Intraband contributions are for metals ! Thus you should not even
 >>attempt them for GaAs.

I guess the reason is that GaAs is a band insulator.
But for metallic compounds (e.g.,  intermetallic heavy fermion metals,  
like CeCu_{6-x}Au_{x}),
I can still run "x joint" with switch=6 to get the intraband  
contributions and the plasma frequency.
Am I right?

Thanks for the advice,

Jianxin

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20090911/0a6f8b34/attachment.htm>