My Community

Posted: **Fri Nov 23, 2012 11:59 am**

I'm having problems getting reasonable values for the energy of Y1(g) , an isolated yttrium atom in vacuum.

I need this to calculate the energy of adsorption of Y on a surface. I'm using the supplied PBE POTCAR file for Y:

Code: Select all

 PAW_PBE Y_sv 06Sep2000
 11.0000000000000000 
parameters from PSCTR are:
VRHFIN =Y: 4s4p5s4d
LEXCH = PE
EATOM=1046.2256 eV,  76.8954 Ry

TITEL = PAW_PBE Y_sv 06Sep2000

I get large energies for any INCAR parameter sets I've tried. My POSCAR files is one Y atom in a box of 10 x 10 x 10 Angstrom.

Here are the things I've tried in INCAR and the energies I get:
non-spin polarized = 63.5eV
spin polarized = 73.4eV
spin polarized, GGA+U with J=0, U=3, I get energy = 91eV

I should be getting a value of |energy| <~ 4eV, what is going on? Is there an issue with the POTCAR file? Here is an example INCAR file I use:

Code: Select all

SYSTEM = Y in vacuum, for energy

#Electronic?Static?Optimization---------------
PREC = High 
NELM = 120
NELMIN = 6
EDIFF = 1E-6
IALGO = 48

#Ionic?Relaxtion------------------------------
NSW = 0 #?0?ionic?relaxation?steps
IBRION = -1  
ISMEAR = -5

#SPIN?&?GGA+U?parameters--------------------
ISPIN = 2
LDAU = .TRUE.
LDAUTYPE = 2
LDAUL = 2
LDAUU = 3
LDAUJ = 0

#Output?files?---------------
LWAVE = .FALSE.
LCHARG = .FALSE.

#Parallelisation?for?LINUX?cluster
NPAR = 1      #number?of?nodes.

<span class='smallblacktext'>[ Edited ]</span>

Posted: **Tue Nov 27, 2012 4:19 pm**

POTCAR is OK. Use another diagonalization (IALGO=38). Note that when using new 52 potentials numerical problems like this disappear.

Posted: **Mon Dec 24, 2012 6:23 pm**

Note to all: IGALGO=38 fixed the problem. Thanks!

My Community

PBE POTCAR for Y (yttrium)

PBE POTCAR for Y (yttrium)

PBE POTCAR for Y (yttrium)

PBE POTCAR for Y (yttrium)