[Pw_forum] CRYSTAL OPTIMISATION

[Davide Tiana]

Dear Marta,
I am not really expert in solid state and plane-wave calc (I've  
started making   plane wave calc few months ago), anyway here some  
comments that I hope could help you.

you are optimising the cell parameters. Did you previously optimise  
the atomic position?
Anyway, checking your output it seems that the 1st system, could  
probably converge in few steps (so run it again starting from the last  
point)

"
chpc-aw7:Downloads davidetiana$ grep "enthalpy old"  
CristalXX.vcrelax.pbe_6-60-500.out
      enthalpy old            =   -2073.6107364912 Ry
      enthalpy old            =   -2075.4476327812 Ry
      enthalpy old            =   -2075.8476883991 Ry
      enthalpy old            =   -2075.9872095194 Ry
      enthalpy old            =   -2076.0338351603 Ry
      enthalpy old            =   -2076.0481135064 Ry
      enthalpy old            =   -2076.0609253277 Ry
      enthalpy old            =   -2076.0737452991 Ry
      enthalpy old            =   -2076.0836740723 Ry
      enthalpy old            =   -2076.0866052168 Ry
      enthalpy old            =   -2076.0888066767 Ry
      enthalpy old            =   -2076.0912973209 Ry
chpc-aw7:Downloads davidetiana$ grep volume CristalXX.vcrelax.pbe_6-60-500.out
      unit-cell volume          =   15197.3430 (a.u.)^3
      new unit-cell volume =  15854.52474 a.u.^3 (  2349.39814 Ang^3 )
      new unit-cell volume =  16243.99458 a.u.^3 (  2407.11162 Ang^3 )
      new unit-cell volume =  16142.99843 a.u.^3 (  2392.14553 Ang^3 )
      new unit-cell volume =  15959.34517 a.u.^3 (  2364.93093 Ang^3 )
      new unit-cell volume =  15912.47810 a.u.^3 (  2357.98594 Ang^3 )
      new unit-cell volume =  15875.48556 a.u.^3 (  2352.50422 Ang^3 )
      new unit-cell volume =  15780.27042 a.u.^3 (  2338.39479 Ang^3 )
      new unit-cell volume =  15598.61538 a.u.^3 (  2311.47629 Ang^3 )
      new unit-cell volume =  15524.64761 a.u.^3 (  2300.51540 Ang^3 )
      new unit-cell volume =  15493.58732 a.u.^3 (  2295.91274 Ang^3 )
      new unit-cell volume =  15439.91245 a.u.^3 (  2287.95894 Ang^3 )
      new unit-cell volume =  15387.53554 a.u.^3 (  2280.19748 Ang^3 )
      new unit-cell volume =  15346.52555 a.u.^3 (  2274.12043 Ang^3 )
chpc-aw7:Downloads davidetiana$
"

I am a bit less confident with the second one since if the volume as  
in the previous case decrease rapidly, the enthalpy is not (however I  
usually relax the atomic position working with total energy so it can  
be different in this case). However, looking with Xcrysden it seems  
nothing weird happen:
"
chpc-aw7:Downloads davidetiana$ grep "enthalpy old"  
CristalXX.vcrelax.second.pbe_6-60-500.out
      enthalpy old            =   -2076.0940909869 Ry
      enthalpy old            =   -2076.0947666683 Ry
      enthalpy old            =   -2076.0958973172 Ry
      enthalpy old            =   -2076.0966745403 Ry
      enthalpy old            =   -2076.0974062802 Ry
      enthalpy old            =   -2076.0982401038 Ry
      enthalpy old            =   -2076.0993096777 Ry
      enthalpy old            =   -2076.1007114162 Ry
      enthalpy old            =   -2076.1024392593 Ry
      enthalpy old            =   -2076.1042708808 Ry
      enthalpy old            =   -2076.1055118520 Ry
      enthalpy old            =   -2076.1068005511 Ry
      enthalpy old            =   -2076.1081007153 Ry
chpc-aw7:Downloads davidetiana$ grep volume  
CristalXX.vcrelax.second.pbe_6-60-500.out
      unit-cell volume          =   15346.5256 (a.u.)^3
      new unit-cell volume =  15347.63578 a.u.^3 (  2274.28495 Ang^3 )
      new unit-cell volume =  15344.50987 a.u.^3 (  2273.82174 Ang^3 )
      new unit-cell volume =  15338.58618 a.u.^3 (  2272.94394 Ang^3 )
      new unit-cell volume =  15326.37995 a.u.^3 (  2271.13516 Ang^3 )
      new unit-cell volume =  15307.15200 a.u.^3 (  2268.28587 Ang^3 )
      new unit-cell volume =  15275.28713 a.u.^3 (  2263.56399 Ang^3 )
      new unit-cell volume =  15226.01665 a.u.^3 (  2256.26285 Ang^3 )
      new unit-cell volume =  15151.73791 a.u.^3 (  2245.25588 Ang^3 )
      new unit-cell volume =  15040.70206 a.u.^3 (  2228.80207 Ang^3 )
      new unit-cell volume =  14958.62861 a.u.^3 (  2216.64003 Ang^3 )
      new unit-cell volume =  14914.57199 a.u.^3 (  2210.11152 Ang^3 )
      new unit-cell volume =  14876.72700 a.u.^3 (  2204.50347 Ang^3 )
      new unit-cell volume =  14843.89863 a.u.^3 (  2199.63881 Ang^3 )
      new unit-cell volume =  14767.53018 a.u.^3 (  2188.32217 Ang^3 )
chpc-aw7:Downloads davidetiana$
"
The last suggestion is to reduce, at least at first instance, your cut-off.

Concerning this:
>
> 3- do you know if there is a limit on the number of atoms in teh
> conventional cell?
>

This is not strictly related with your problem. You can run it.
I am not a developer but, checking PW routines, I don't think so:

"
read_file.f90:  IF (nat > 0) CALL checkallsym( nat, tau, ityp, nr1, nr2, nr3 )
setup.f90:     IF (nat==0) CALL errore('setup','free electrons: nbnd  
required in input',1)
setup.f90:  IF (nat==0) THEN
"

good luck,
Davide