Heey Shadman Saquib

I am not familiar with DMol3 or CASTEP, but if you're looking for another method to compare, why not try Quantum ESPRESSO? QE is considered a standard tool in computational catalysis and is widely used for accurately calculating OER/HER reaction pathways. Plus, it’s open source. I have used it myself for OER on a Co-decorated Pt surface, and it worked just fine.

Best, Gabriel Vinicius

Hello Gabriel Vinicius

Thanks for your suggestion. I will carry out the simulations in QE as well.

Could you please share the sequence of calculations in QE for OER? Is it something like this?

'vc-relax' with pw.x > 'scf' with pw.x > phonon with ph.x ?

I am still quite a beginner in DFT calculations.

Glad to help, Shadman Saquib

When I did it, I went through:

'vc-relax' with pw.x > 'scf' with pw.x > 'nscf' with pw.x > 'phonon' with ph.x.

I just did NSCF before the 'phonon' calc. But now, looking back, I understand that NSCF refines the results from the SCF calculation at denser k-points, providing better resolution for DOS or band structure. For OER calculations, it doesn't really matter. I might be mistaken, but this is how I see it now.

So, the steps you mentioned look fine to me.

Best,

Gabriel Vinicius

There are a lot of differences between a local basis set code and a plane-wave basis set code, and each approach has advantages and disadvantages. It's very difficult to say why one is faster than another without any information about the quality of the two calculations. For example:

* Were there the same number of electrons in each calculation?

* Did you use the same k-point and q-point set for each calculation?

* Are they converged to approximately the same degree with respect to basis set size?

* How much vacuum do you have in the out-of-plane direction? Vacuum is "free" in a local basis set, but not in a plane-wave basis set