CP2K Open Source Molecular Dynamics

User Tools

Site Tools

Trace:
exercises:2016_uzh_cmest:calculating_pdos

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
exercises:2016_uzh_cmest:calculating_pdos [2016/10/30 16:33] – tmuellerexercises:2016_uzh_cmest:calculating_pdos [2020/08/21 10:15] (current) – external edit 127.0.0.1
Line 6: Line 6:
   * hexagonal Boron Nitride   * hexagonal Boron Nitride
  
-Both feature the same 2D crystal structure with the only difference being the lattice constant and that one consists of only carbons and the other of boron and nitride.+Both feature the same 2D crystal structure with the only difference being the lattice constant and that one consists of only carbons and the other of boron and nitrogen.
  
 Similar to the previous exercise we write the coordinates in term of the unit cell: Similar to the previous exercise we write the coordinates in term of the unit cell:
Line 57: Line 57:
         # print all projected DOS available:         # print all projected DOS available:
         NLUMO -1         NLUMO -1
 +        # split the density by quantum number:
         COMPONENTS         COMPONENTS
       &END       &END
Line 69: Line 70:
       SYMMETRY HEXAGONAL       SYMMETRY HEXAGONAL
       PERIODIC XYZ       PERIODIC XYZ
-      # and replicate it +      # and replicate this cell (see text): 
-      MULTIPLE_UNIT_CELL 2 2 2+      MULTIPLE_UNIT_CELL 2 2 1
     &END CELL     &END CELL
 +    &TOPOLOGY
 +      # also replicate the topology (see text):
 +      MULTIPLE_UNIT_CELL 2 2 1
 +    &END TOPOLOGY
     &COORD     &COORD
       SCALED       SCALED
Line 79: Line 84:
     &KIND C     &KIND C
       ELEMENT C       ELEMENT C
-      BASIS_SET TZVP-MOLOPT-GTH+      BASIS_SET DZVP-MOLOPT-GTH
       POTENTIAL GTH-PBE       POTENTIAL GTH-PBE
     &END KIND     &END KIND
Line 87: Line 92:
 </code> </code>
  
-The replication of the unit cell before doing the calculation is necessary since we are currently sampling only at he $\Gamma$ point and we will otherwise do not have enough sampling points for the density of states. Another option (which we will look into in the next exercise) is to sample over k-points instead.+The replication of the unit cell is necessary since the program samples only at he $\Gamma$ point unless instructed otherwise and we will otherwise do get a meaningful sampling of the density of states (e.g. the grid over the Brillouin Zone will be too coarse). Another option (which we will look into in the next exercise) is to sample over k-points instead. 
 + 
 +What you will get in addition to the output file is a file named ''graphene_pdos-k1-1.pdos'' (to be precise, you will get one such file per atom kind but here we only have one, carbon) with a content similar to: 
 + 
 +<code> 
 + Projected DOS for atomic kind C at iteration step i = 0, E(Fermi) =    -0.061771 a.u. 
 +#     MO Eigenvalue [a.u.]      Occupation                 s                py                pz                px               d-2               d-1       
 +          d0               d+1               d+2 
 +       1         -0.798996        2.000000        0.83986852        0.00000000        0.00000000        0.00000000        0.00000000        0.00000000        0.16013148        0.00000000        0.00000000 
 +       2         -0.764498        2.000000        0.42353037        0.12725580        0.00000000        0.36601449        0.01397386        0.00000000        0.06436706        0.00000000        0.00485843 
 +[...] 
 +</code> 
 + 
 +The columns correspond to the orbitals present in the basis set (hence //projected// DOS). Normally you would now do a convolution plot using a gaussian to get a smooth DOS, but instead you can also simply plot it in a scatter or vertical line plot. 
 + 
 +Please also note the unit of the energy, it is in $E_h$. When looking at DOS plots you may want to convert it to Electronvolt instead.  
 + 
 +While some of the new options to help with convergence are of numerical nature, [[howto:static_calculation#adding_smearing|the smearing is not]]. 
 + 
 +  * Repeat the above calculation for the different multiple cells 3x3x1, 4x4x1, 5x5x1, 6x6x1 and 7x7x1 
 +  * Even though you are not required to do any smoothing, create the DOS plots for each simulation. 
 +  * Do you see why it is necessary to do the unit cell replication? Hints: does graphene have a band gap? Compare the plots for 3x3x1 and 5x5x1. 
 +  * .. which state ($s$, $p_x$, ..) is mainly responsible for that? 
 +  * Repeat the calculations for h-BN instead (use the DZVP-MOLOPT-SR-GTH basis set for boron instead of the DZVP-MOLOPT-GTH and change the lattice constant to $2.504 Å$). 
 +  * Create the plots again (remember: now there are two pdos files, one for each kind). What is the clear difference to graphene? 
 +  * Do you now understand why we need smearing?
exercises/2016_uzh_cmest/calculating_pdos.1477845234.txt.gz · Last modified: (external edit)