Page 1 of 1

A problem with molecule NO

Posted: Mon May 02, 2016 11:14 am
by tonylxh
Hi, everbody

I calculated the isolated NO molecule, the total energy Etot=-12.026eV. But the experimental data of bond energy N=O is 607KJ/mol, is about 6.294eV. I do not know why, so I ask for your help. My setup is as follow:

INCAR
general:
SYSTEM = NO
ISTART = 0 ; ICHARG=2
ISMEAR = 0 ; SIGMA = 0.2
dynamic:
IBRION = 2
NSW = 100
POTIM = 0.1
EDIFFG=-0.01
ENCUT=400
PREC = High
################################
POSCAR
1.00000000000000
30.8200000000000003 0.0000000000000000 0.0000000000000000
0.0000000000000000 30.8200000000000003 0.0000000000000000
0.0000000000000000 0.0000000000000000 30.3150000000000013
O N
1 1
Selective dynamics
Direct
0.7378683101883881 0.9468784104989977 0.9995057216026794 T T T
0.7352416702863921 0.9874461343450991 0.0000494644823208 T T T
##########################
KPOINT
Gamma-point only
1
rec
0 0 0 1
##########################
POTCAR use normal PBE

##########################

Thank you for your help and look forward to your answer
Best wishes

Re: A problem with molecule NO

Posted: Mon May 02, 2016 3:06 pm
by alex
Hi!

The energy of formation is NOT equal to the total energy.

I'd strongly(!) suggest you read some basic textbooks about thermodynamics and quantum chemistry/physics.

Cheers,

alex

Re: A problem with molecule NO

Posted: Mon May 02, 2016 3:42 pm
by tonylxh
I also considered spin polarization. with ISPIN=2 Etot_NO=-12.2979eV,and E_O=-1.5925eV E_N=-3.1112eV, Ebond=7.5942eV, which is better a little bit, but still 1.3eV larger than the experimental value.

Re: A problem with molecule NO

Posted: Tue May 03, 2016 10:58 am
by tonylxh
alex wrote:Hi!

The energy of formation is NOT equal to the total energy.

I'd strongly(!) suggest you read some basic textbooks about thermodynamics and quantum chemistry/physics.

Cheers,

alex
hi alex

thanks for your answer

sure, you are right, "The energy of formation is NOT equal to the total energy." But if you calculate an isolated atom like O N in a normal way(without spin polarization), the energy is rather small comparing with the total energy of molecule, that's why I only mentioned the total energy (I am sorry that I forget to give the results with spin polarization, and I have added this part already). I think "formation energy" is not what I am talking about,but "binding energy". For formation energy, one must think about the reference.
Thank you again for your answer

Best wishes

Re: A problem with molecule NO

Posted: Tue May 03, 2016 3:26 pm
by alex
Hi tonylxh,

I see your point. Anyway, scientific discussion needs precision in both language and experiment. ;-)
Thanks for clarification. Try something like that for INCAR (see below).

One more thing - you have done right in your POSCAR already - is to destroy symmetry in one direction.
You also need to check you OUTCAR for the proper occupation.

Cheers,

alex

PS:

SYSTEM = O-atom

Electronic minimisation
GGA = PE
PREC = Accu
EDIFFG = 1.e-8
EDIFFG = -1.e-6
ENCUT = 400.
NELMIN = 4
NELM = 200 ! Maximum number of electronic SC steps
LMAXMIX = 6
ISPIN = 2
NUPDOWN = 2

DOS related values
SIGMA = 0.001
ISMEAR = 0 ! Gaussian smearing for big cells

Do not write huge files
LCHARG = .FALSE.
LWAVE = .FALSE.

Performance options
NPAR = 1 ! adjust to number of processors
ALGO = N

Re: A problem with molecule NO

Posted: Wed May 04, 2016 2:27 pm
by tonylxh
alex wrote:Hi tonylxh,

I see your point. Anyway, scientific discussion needs precision in both language and experiment. ;-)
Thanks for clarification. Try something like that for INCAR (see below).

One more thing - you have done right in your POSCAR already - is to destroy symmetry in one direction.
You also need to check you OUTCAR for the proper occupation.

Cheers,

alex

PS:

SYSTEM = O-atom

Electronic minimisation
GGA = PE
PREC = Accu
EDIFFG = 1.e-8
EDIFFG = -1.e-6
ENCUT = 400.
NELMIN = 4
NELM = 200 ! Maximum number of electronic SC steps
LMAXMIX = 6
ISPIN = 2
NUPDOWN = 2

DOS related values
SIGMA = 0.001
ISMEAR = 0 ! Gaussian smearing for big cells

Do not write huge files
LCHARG = .FALSE.
LWAVE = .FALSE.

Performance options
NPAR = 1 ! adjust to number of processors
ALGO = N
Hi Alex

Thank you very much for your answer. I tried to calculate the energy of O atom with INCAR you suggusted, but the energy does not change much(less than 0.05eV). I am wondering if there is any mistake I made when I calculated the total energy of NO. Do you find any mistake of my input files?
Thanks again for your help

Best wishes
tony

Re: A problem with molecule NO

Posted: Mon May 09, 2016 2:53 pm
by alex
Hi tony,

you got the wrong occupation in your calculation. If you do it right, you'll end up at around -1.9 eV for the total energy of an O atom (400eV/O standard PAW/PBE).

cheers,

alex

Re: A problem with molecule NO

Posted: Wed May 11, 2016 11:11 am
by tonylxh
alex wrote:Hi tony,

you got the wrong occupation in your calculation. If you do it right, you'll end up at around -1.9 eV for the total energy of an O atom (400eV/O standard PAW/PBE).

cheers,

alex
Hi alex

Thank you very much for your help

Best wishes
tony