My Community

Hi,

Recently, I came across several papers that study the adsorption of certain molecules into 2D materials. They argue that by adding charge you can tune the selectivity of adsorption to certain species. I am curios whether this approach makes sense? They change the total number of electrons in the cell, calculate the total energy and take that as a reference and then they calculate the total energy of the adsorbed molecules+charge which is then used to calculate the binding energy at that charge, see for example:

http://pubs.acs.org/doi/abs/10.1021/jp510387h
http://pubs.acs.org/doi/abs/10.1021/ja400243r

The above results were obtained not using VASP, and I am wondering whether such calculations could be performed by VASP. The VASP manual clearly states "Total energies from charged slab calculations are hence useless, and can not be used to determine relative energies." How then are people using these total energies.

Comments on this are highly appreciated.

Thanks.

I had a look at the first paper. The idea is not bad. But how to arrange the switching between neutral and charge state in the laboratory?
The calculations seem to be correct (they have a typo in Eq. 1). There is, however, indication that something goes wrong.
The comparison of two neutral systems (Fig. 5a) gives the adsorption energy of N2 Eads=-9.51 kcal/mol.
The comparison of two negatively charged systems (Fig. 5b) gives the desorption energy Edes=-28.82 kcal/mol.
This means the introduction of an electron is connected with large change of energy, i.e. change of bonding.
Where is a change of bonding there is some barrier. This things are absolutely not investigated.
Without the analysis how the energetics changes with the introduction
of the electron are conclusions of the work useless.

Dear Admin,

Many thanks for your detailed reply. From your answer it seems valid then to use charged slabs to calculate adsorption energy and other phenomena related to the introduction of charge, what is the meaning then of the warning in the VASP's manual:

"Total energies from charged slab calculations are hence useless, and can not be used to determine relative energies."

Regards.

Hi hat343,

VASP uses a background charge to compensate the extra charge in the unit cell. So the potential created by this charge at one point in your unit cell is a function of the size of the unit cell.

However, I'd suggest you stick with neutral unit cells like admin pointed out before, because
- charged unit cells are unphysical (the name here is coulomb explosion)!
- You'll never able to compare it with experiments, because charged unit cells simply do not exist!
- You'll struggle with the old problem of our trade: 'garbage in, garbage out' (sorry for the harsh words!).
- If you work hard, use same unit cell sizes for all your comparisons, you may succeed with comparing between equally charged systems (e.g. removing one and only one proton from different positions).

The question you may try to answer is more like: how could I manipulate the local structure to achive the effect I'm after?!

Cheers,

alex