[DFTB-Plus-User] R: Multi-terminal device

[Alessandro Pecchia]

Hi

 

I fixed the problem with relaxations. The patch will be soon ready for
download from the BCCMS website.

 

I noticed that you do not include any contact surface on you calculations.
This may produce artifacts, especially as you increase bias.

Also you do not include any lateral periodicity. 

In any case, if you do include Au surfaces in the molecular region, keep in
mind that the sk file you are using is not any good for Au relaxations.

Therefore you have to keep fixed all Au atoms.

 

I am simulating a multi-terminal device. After finishing the calculations of
the contacts, I got  this error message in the calculation of the active
region:

WARNING <zextract_dns>: bad indeces specification

WARNING <zextract_dns>: bad indeces specification

WARNING <zextract>: bad indeces specification

Error in check_nzval: wrong row or column specification

 

This seems to be something wrong with the structure. Usually this error
comes when your structure is not partitioned properly.

Please send it to me. Thanks.

 

 

Also can I read the current at each terminal using this command only?

TerminalCurrents= {"source" "drain" "gate"}

 

No. At the moment you should specify explicitly two lines:

TerminalCurrents= EmitterCollector{"source" "drain" }

TerminalCurrents= EmitterCollector{"source" "gate"}

 

NOTE that I decided to change the tag in EmitterCollector since I think it
is more transparent emitter comes naturally first.

 

 

 

Thanks for your kind attention

Greetings

Ahmed

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20130111/0fbb6278/attachment.htm>