QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: Abdullah Fatih on November 24, 2011, 19:06

Title: There is a error in ATK 11.8
Post by: Abdullah Fatih on November 24, 2011, 19:06

Hi,
I am using ATK 11.8 current version.
When I am simulating graphene ribbon or graphene sheet with MGAA I see an error message end of the simulation.

I attached error message. You can see.

What is the reason of this problem?

Thanks,

Title: Re: There is a error in ATK 11.8
Post by: kstokbro on November 24, 2011, 20:35

can you post the input file, or send it to [email protected] such that we can reproduce the error

Title: Re: There is a error in ATK 11.8
Post by: Abdullah Fatih on November 24, 2011, 22:29

Ok. I attached sample input file.

Title: Re: There is a error in ATK 11.8
Post by: Anders Blom on November 24, 2011, 22:55

I wouldn't expect meta-GGA to be useful at all for this system. It makes most sense to use for "traditional" 3D insulators and semiconductors. That said, the error is something we can look into, but I don't think you should wait for a solution, but rather stick to usual LDA/GGA for graphene.

Title: Re: There is a error in ATK 11.8
Post by: Nordland on November 24, 2011, 23:10

First of all I would recommend reading this post I have posted on a separate issue: http://quantumwise.com/forum/index.php?topic=1341.msg6582#msg6582

Regarding your system, the metaGGA can be used in two ways - either with an automatic determination of the parameter c ( which you are using ) or by setting it manually. However the formula for calculation of C is ill-defined for non crystals, so therefore for molecules and periodic structures with a lot of vacuum, the parameter c will slowly become more and more ill-defined, and the Hamiltonian will have larger and larger fluctuation , and could potentially crash the calculation.

So if you want to use metaGGA for such as system you should first determine the a sample system in which you will determine the optimial c parameter by manual tuning, then when creating your calculation on your true system, use this parameter you found for your sample study. This works quite well.

However I think it is important to understand what is the need for metaGGA. metaGGA can be used for opening the band gap of a semi-conductor/insulator, where LDA would underestimate the band gap. Hence it has no real application for metals as it has no sane effect on materials which are truly metals.

Title: Re: There is a error in ATK 11.8
Post by: Abdullah Fatih on November 25, 2011, 21:55

Dear Anders Blom, I understood you but I am looking forward for a solution :)

Dear Nordland, Thanks a lot for useful informations about this problem. I will try your opinions..