[Pw_forum] Cohesive energy

[Lorenzo Paulatto]

On Thu, January 24, 2008 12:30, Riccardo Sabatini wrote:
> The result i had is around -7.47 Ry.
> and there is listed a value of -7.52 Ry for the Si atom free.

What you have to do is break the symmetry of the isolated atom.

Silicon atomic configuration is [Ne] 3s2 3p2, nobody knows what it would
if it were alone in the universe, but its ground state configuration fills
the 3s orbital completely and then, following Hund's rule, put 1 electron
each in two p orbitals. Example:
3s   2
3p-x 1
3p-y 1

When you do the pw calculation for the isolated atom symmetry is actively
enforced so the occupations will look more like this:
3s    2
3p-x 2/3
3p-y 2/3
3p-z 2/3

In order to break the symmetry you have first to relax its enforcement by
specifying nosym=.true. and then you have to break it actively in the
initial position by using a non cubic cell, some random starting wfcs or
some other similar trick.

You can also specify band occupations by hand, e.g. (if you use nbnd=4):
occupations = 'from_input'
 [...]
OCCUPATIONS
1.   1. 1. 0.
this will suffice in breaking the symmetry.

> In the first case i
> suppose one has to divide the total energy by the numbers of atoms in
> the cell (2 in the first case and 8 in the second)

true

> but i can't figure out what is energy/cell

it's what you get printed on the screen after "Total energy:", in periodic
boundary condition you can't compute "Total" energy

>     Is it correct to define a energy for each atom in a cell in this way
> ? (If i remember correctly the difference in energy for the sc
> simulation and for the fcc arise from the difference in volume, one is 4
> times than the other not from the numbers of atoms).

The sc cell *contains* 4 fcc cells, has 4 times its volume and 4 times its
atoms. Its energy per cell is 4 times larger, so the ratio energy/volume
must be the same. Also keep in mind that volume for an fcc cell is
(alat**3)/4.

Please have a look at crystal structure in you favourite solid state book!

> caused by the number of plane waves used ?

Without seeing you input files I can't know, there may be many other
reasons, most likely numbers of k-points (you need 4 times as many
k-points in fcc than in sc).

bye, LP

P.S. please provide your affiliation.

-- 
Lorenzo Paulatto
SISSA  &  DEMOCRITOS (Trieste)
+39 040 3787 511
http://people.sissa.it/~paulatto/


----------------------------------------------------------------
  SISSA Webmail https://webmail.sissa.it/
  Powered by SquirrelMail http://www.squirrelmail.org/