Hello teachers,
I have done adsorption of molecules on the substrate and calculated the work function, but whenever I do it for a defective system my work function decreases too much and the graph looks very weird. instead of a straight-line, it gives a huge bump.
1. I have added dipole corrections,
2. my vacuum thickness was already 30A, but still, I increased it to 48A but the result is the same.
I will be very grateful for your answers.
Thanks
Hello Zul,
Your potential plot looks as expected. The hump you are seeing at ~27A (x-axis) is in fact the dipole correction itself.
The plot between 0 and 5A gives you the vacuum level for the bottom surface (no molecule) and the plot between ~17 and ~27A gives you the vacuum level of the top surface (with the adsorbed molecule). The molecule you added to the top surface induces a strong dipole and therefore reduces the work function by almost 2 eV. Did you ionically relax the top surface atoms in your calculation? (Btw, 20A of vacuum should be more than enough)
Lastly, the work function is the difference between the vacuum energy level and the Fermi level.
Hope this helps!
Best,
Peter
Hi Peter Schindler , thanks for your explanatory answer.
1. yes I have relaxed the molecule before.
2. yes, the decrease in work function is almost 2eV. if I consider the vacuum energy in ~17-27A. will it be a reasonable value?
Thanks for your time.
Zul Qarnain
1) Great! You mean, you have relaxed the molecule by itself, right? You should also relax the surface slab with the molecule on top. For that, it would be enough to relax the topmost 2 or 3 atomic layers of the surface (plus the molecule) and keep the rest of the slab fixed.
2) Yes, the vacuum energy between ~17-27A is the appropriate value to pick. I am not sure what "0 eV" on your y-axis plot refers to, but you will probably need to subtract the Fermi energy to get the work function value.
-Peter
Peter Schindler thank you very much.
1) yes I have relaxed the molecule and substrate before adsorption, then did the adsorption process. but I am using MXene, M2X type which is already a three-layer surface.
2) For the second point I get that thanks.
Zul Qarnain Ok, great. You should definitely relax the combined structure (slab + molecule, in the same cell) before calculating the work function because the distance and orientation of the molecule will strongly affect the dipole. Further, the unit cell area (normal to the z-direction) will affect the effective dipole as well, which corresponds to the packing density of the molecule on the surface.
Good luck with your research!
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling