[Pw_forum] problem in optimizing nanotubes in QE

Lorenzo Paulatto lorenzo.paulatto at impmc.upmc.fr
Wed Sep 16 09:25:20 CEST 2015

On Wednesday, September 16, 2015 08:46:32 AM Naiwrit Karmodak wrote:
> This is a 3,3 nanotube. I have also tried with 6,6 nanotube and the problem
> is same. Both david and cg option for diagonalization is not giving any way
> to optimized geometry. Please suggest a solution for this.

Dear Naiwrit,
changing the type of diagonalization (cg or david) can affect the convergence 
of a single scf step, but the structure optimization will not change.

If your structure breaks down it is because it is not stable, at least with 
the functional you are using. Visualizing your input with xcrysden I can see 
that all your carbon atoms are stuck together, which means that your input is 
wrong. My guess is that you wanted to specify the atoms positions as fractions 
of the crystall lattice vectors, in this case you have to use the "crystal" 

C             0.0863    0.1169    0.0443
C             0.9137    0.8831    0.132

Also note that specifying a k-point shift along the x and y directions is not 
correct, because you want wavefunctions which are periodic over the unit cell 
in those directions. I.e. use
K_POINTS  automatic
1 1 10  0 0 1

Finally, please disregards the advice from Bahadir, especially the one about 
smearing, as reducing smearing will not necessarily make convergence easier.

HTH, kind regards

Dr. Lorenzo Paulatto
IdR @ IMPMC -- CNRS & Université Paris 6
+33 (0)1 44 275 084 / skype: paulatz
23-24/4é16 Boîte courrier 115
4 place Jussieu 75252 Paris Cédex 05

More information about the users mailing list