[QE-users] ultrasoft vs NC pseudopotentials for ionic conductivity

[Lorenzo Paulatto]

> I tried to calculate activation energy of ionic conductivity using 
> ultrasoft pseudopotentials in QE, and the same using NC pseudopotentials 
> in siesta. The greatest agreement with the data of the article [F. Ning, 
> S. Li, B. Xu, and C. Ouyang, Solid State Ionics 263, 46 (2014).] was 
> obtained in the second case. Does this mean that pseudopotentials are 
> better suited for such cases? Is it possible that these pseudopotentials 
> better withstand situations when the distances between the atoms are 
> less than their natural level? Is there any example for the 
> qunatum-espresso for this area?

Dear Maxim,
you're not giving enough details to get a reasonable answer. Why do you 
think the results are different? How much difference? Why do you think 
it is the pseudopotential formalism that matters? Why do you think that 
the norm conserving one is better than the other? How did you exactly do 
all the calculation, the reference you cite used DFT+U, did you do as 
well? Which value of U...



-- 
Lorenzo Paulatto - Paris