QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: hkuieagle on May 10, 2010, 08:40

Title: Very werid problem!
Post by: hkuieagle on May 10, 2010, 08:40

Hi all. I encountered the problem when I was doing some calculations on two probe system of graphene.Then to be simple,I take an experiment on Lih2Li system.Below is the description of the problem:
There are two scripts in the attachment which are copyed from the ATK manual , the ONLY difference between them is in the "unit_cell"(I reverse the order of A and B vertor). Since  it's symmetrical between A and B,I don't think it'll make any difference. But the results differs by a mile!
For lih2li-setup-1.py : the current is 1.57e-05
For lih2li-setup-2.py : the current is 1.69e-12
Well ,How to explain ?

bwt: ATK 2008.02.0 Linux-i686 build 13600 2008-03-03

Title: Re: Very werid problem!
Post by: zdhlover on May 10, 2010, 09:05

Do you think you can build a strange bulk structure:[[  0. ,   8.7,   0. ],
                                                                          [  8.7,   0. ,   0. ],
                                                                          [  0. ,   0. ,  11.6]]*Angstrom as the second script ?

you'd better check whether your two-probe system is correct or not at first.

Title: Re: Very werid problem!
Post by: Anders Blom on May 10, 2010, 10:08

The systems are not 100% equivalent: one is right-handed and one is left-handed. You had better always make your unit cell right-handed to avoid problems.

Title: Re: Very werid problem!
Post by: hkuieagle on May 11, 2010, 03:27

Well, I don't think we can tell the right-hand unit-cell from the left-hand in reality (even I don't know how to create the difference between them in lab).

Title: Re: Very werid problem!
Post by: Anders Blom on May 11, 2010, 10:57

True. But numerically there's a difference.

Title: Re: Very werid problem!
Post by: youngjfly on May 11, 2010, 14:18

true :D