QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: pervanem on September 21, 2018, 14:20

Title: 2016.4 vs 2018.06 sp1
Post by: pervanem on September 21, 2018, 14:20

Hello,
I performed the NEGF calculation (IV-Characteristics)  and transmisionspectrum....
But results are different from each other...default settings are almost same 2018.06 with 2016.4..
Why? which is more accurate?
Can you help me?
 related pictures was attached.

Title: Re: 2016.4 vs 2018.06 sp1
Post by: pervanem on September 21, 2018, 14:23

continued

Title: Re: 2016.4 vs 2018.06 sp1
Post by: pervanem on September 21, 2018, 14:23

continued2

Title: Re: 2016.4 vs 2018.06 sp1
Post by: Petr Khomyakov on September 21, 2018, 15:53

There are at least two defaults that have been changed: pseudopotential/basis set (FHI/DZP in 2016.4) and electron temperature (300K in 2016.4) .

In 2018, 'Electron temperature' parameter is called 'Broadening' (which is 1000 K in 2018) to reflect the fact that this is not a real electron temperature, but rather a numerical parameter to make SCF convergence smoother with respect to the number of k-points.

In 2018, the default pseudopotential/basis set is pseudoDojo/Medium.