My Community

Hi all,

I performed STM simulation according to the following two steps: step1: scf ; step2: calculate STM within the energy window. The job is well finished. The CHGCAR calculated from SCF is correct. But unfortunately, when I set a energy window by set EINT, the CHGCAR and PARCHG is absolutely wrong. Why it happened ?

I have sent the Input and the figures for CHGCAR from scf and PARCHG from step2 as attachments

, please kindly check them and give me some suggestions. Thanks!

(The VASP I used is v6.3.0)

Best & Regards!

Jianguo

Hi Jianguo,

Thanks for reaching out. I took a look at your calculation but there is nothing obviously wrong with it. I'm trying to run your example but this is quite a heavy calculation. Do you encounter the same problem also in a smaller test system?

manuel_engel1 wrote: ↑Wed Feb 12, 2025 1:37 pm

Hi Jianguo,

Thanks for reaching out. I took a look at your calculation but there is nothing obviously wrong with it. I'm trying to run your example but this is quite a heavy calculation. Do you encounter the same problem also in a smaller test system?

Thanks for your kindly reply. I also tried the case of graphene, the same problem is detected (PARCHG seems wrong, but the STM image created from p4vasp using it seems correct). The input files are upload as attachment.

Ah! I have missed an INCAR tag that is responsible for the behavior you are seeing. ISTART=0 causes the code to start from scratch, and not from the WAVECAR file. You have to change it to ISTART=1 (or just leave it out, as that is the default). I have verified with the graphene calculation that this is the problem and removing the tag fixes it.

manuel_engel1 wrote: ↑Thu Feb 13, 2025 1:05 pm

Ah! I have missed an INCAR tag that is responsible for the behavior you are seeing. ISTART=0 causes the code to start from scratch, and not from the WAVECAR file. You have to change it to ISTART=1 (or just leave it out, as that is the default). I have verified with the graphene calculation that this is the problem and removing the tag fixes it.

Wow, I also found that yesterday!

Thanks again!

I'm happy to help. It was a bit easy to miss, but it's good to know that you figured it out.