Sorry i found the total energ for paw potential is about -212, so there's no problem here. I just don't know how to get the enough accuracy as the literature. Or is there any way that can make the converge faster?<br>
If the case is a copper chain in a carbon nanotube, should I still set a single cu atom here with large unit cell? Or should I set up a copper chain and do the structure optimization? I am not quite sure because in one unit cell there's only one cu in cnt(6,6) . <br>
Thank you for somebody who could clarify this.<br> Best regards.<br><br>Chengyu Yang<br> <font color="#888888">MMAE, University of Central Florida, USA</font><br><br>