QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: qiuming on March 17, 2011, 09:27

Title: A problem with geo-optimization
Post by: qiuming on March 17, 2011, 09:27

Hi,
I have two problems.
One:While I optimize the distances between the core molecules with the electrodes, the mulliken popolations always come to nearly zero in some distances.
Such as, I want to know the total energies while the distances between them are 2.41, 2.42, 2.43, and 2.44A. Calculations show that the total energies are -5555.00000, -5555.10000, 1700000.000000, and -5555.20000eV. So, which results shoude be chosen. The mulliken populations for all elements in 2.43A are nearly zero.
I wonder know why.
The other: If the calculations with the total energy are not convergent after 100 or 1000 steps, what should I do?

Thanks a lot.

Title: Re: A problem with geo-optimization
Post by: Nordland on March 17, 2011, 09:35

What version are you using of ATK?

Title: Re: A problem with geo-optimization
Post by: qiuming on March 17, 2011, 13:19

0802 and 0810

Title: Re: A problem with geo-optimization
Post by: Nordland on March 17, 2011, 14:01

I would strongly suggest you to upgrade to newest version of ATK. Since many of these problems never arises in the newer version.