Dear researchers,
I am simulating the interaction of molecules with a nanoribbon in Quantum ESPRESSO. The issue I am encountering is the unwanted periodic repetition of the adsorbed molecules along the nanoribbon. Even after increasing the supercell size to reduce molecule–molecule interaction, the molecules still repeat along the nanoribbon, and this seems to affect the results.
Please note that I have applied a vacuum region of 20 Å in the two directions perpendicular to the nanoribbon, so the system is effectively non-periodic in those directions.
My question is: If the supercell is sufficiently large so that the molecules no longer interact with each other, does the periodic repetition still affect the nanoribbon itself? Specifically, how significant is this effect on the nanoribbon’s electronic properties, such as current or density of states?
Is it correct to assume that each molecule interacts with the nanoribbon independently? If so, does periodicity still introduce artificial effects, even when molecule–molecule interactions are negligible?
Finally, what strategies can I use to minimize or eliminate this issue within the limits of plane-wave DFT codes like Quantum ESPRESSO?
Thank you in advance for your insights and suggestions.
QE has a option that add fake, local electric potential gradient to nullify the total dipole moment of the cell. Check this If this is the case.
This is not something you can eliminate if not by changing code. QE is built around plane waves which by definition have an intrinsic periodicity. The only option is to keep increasing the supercell size of the nano ribbon until there is no more interaction. Exactly as you did in the other directions, I would estimate that your molecules need to be about 15-20 A away to reduce their mutual interaction.
On the other hand, your question is quite generic. How do you know that the replica of the molecule are affecting your results? If you do not specify the physical effect you see and why you believe it to be spurious is difficult to help you.
Roberto D'Agosta Thank you very much for your detailed explanation — it's really helpful. Yes, that is exactly my question: I would like to know whether the periodic images of the molecule (due to the intrinsic periodicity in QE) can affect the calculated current, density of states (DOS), or adsorption energy on a graphene nanoribbon.
My supervisor emphasizes that these periodic images likely lead to differences in the results — meaning what we calculate might not reflect the true adsorption energy of a single molecule on the nanoribbon.
Do you think this is correct? And is there any practical way to calculate the adsorption energy of just a single molecule (with minimal or no interactions from its periodic images) using Quantum ESPRESSO?
Also, I’d like to clarify: are the results such as the adsorption energy and DOS calculated for the entire nanoribbon system, or just for the supercell defined in the calculation?
I'd really appreciate your advice.
Jae Hwan Jeong Thank you very much for your time and response — I really appreciate it. Although it’s not exactly related to the specific question I had, I still found your input interesting and helpful.
- Badges
- Science topic
- Similar topics
- Computer Science
- Program