Positive cohesive energy for a well test geometry

Message

magnon963 · #1 Post by **magnon963** » Mon May 23, 2011 4:11 pm

Dear VASP users,

I'm checking BN system having 32 atoms with VASP 5.2.11 and intel compiler. The POSCAR file can't be wrong as it's simple hexagonal supercell. But it's giving me positive cohesive energy. And I know that it can not be correct. For Si with 32 atom cell it's wroking nicely but when I try it for BN surface it fails.

If anyone can guess what could be the reason. And how to sort this problem out.

Thanks a lot for you answer.

best regards,
PS

INCAR file:

Start parameter for this run:
ISTART = 0 job : 0-new 1-cont 2-samecut
ICHARG = 2 charge : 1-file 2-atom 10-const
INIWAV = 1 electr : 0-lowe 1-rand

Electronic Relaxation 1
ENCUT = 550 eV
NELM = 220
NELMIN = 0
NELMDL = -12 # of ELM STEP m
EDIFF = 1E-05 stopping cretiria for ELM
GGA = PE
LDIAG = .F.
LORBIT = 11
MAGMOM = 32*0.5

INIMIX = 0
AMIX = 0.1
BMIX = 0.00001
AMIX_MAG= 0.4
BMIX_MAG= 0.000001
MAXMIX = -8

Ionic Relaxation
NSW = 180 number of steps for IOM #
EDIFFG = -1.E-02 stopping-criterion for IOM
ISIF = 2
ISYM = 2
ISPIN = 2
LCORR = T
IWAVPR = 1
IBRION = 1
POTIM = 0.4
ALGO = N

NPAR = 16
NSIM = 1
LPLANE = .TRUE.

DOS related values:
SIGMA = 0.02
ISMEAR = 2 broad. in eV, -4-tet -1-fermi 0 gaus

NBAND = 96
LREAL = Auto
-------------------------------------------------------------------------------------
POSCAR
C sheet with 32 atoms--Configur
1.00000000000000
9.9481 0.0000 0.0000
0.0000 8.7043 0.0000
0.0000 0.0000 15.0000
32 ! C
Cart
0.01534 8.69561 0.00000
4.98940 8.69558 0.00000
1.26010 6.51383 0.00000
6.23412 6.51377 0.00000
2.50237 8.69560 0.00000
7.47642 8.69555 0.00000
3.74708 6.51379 0.00000
8.72116 6.51375 0.00000
0.01717 5.76819 0.00000
4.99127 5.76809 0.00000
2.50424 5.76809 0.00000
7.47829 5.76804 0.00000
1.25927 7.94548 0.00000
6.23330 7.94542 0.00000
3.74628 7.94543 0.00000
8.72031 7.94539 0.00000
0.01925 4.37072 0.00000
4.99330 4.37063 0.00000
0.01797 1.52218 0.00000
4.99203 1.52214 0.00000
1.26167 2.22465 0.00000
6.23575 2.22459 0.00000
2.50497 1.52217 0.00000
7.47905 1.52211 0.00000
2.50628 4.37064 0.00000
7.48036 4.37058 0.00000
1.26271 3.67254 0.00000
6.23679 3.67248 0.00000
3.74976 3.67245 0.00000
8.72383 3.67242 0.00000
3.74871 2.22456 0.00000
8.72278 2.22451 0.00000
-------------------------------------------------------
Automatic mesh
0 ! number of k-points = 0 ->automatic generation scheme
Gamma ! generate a Gamma centered grid
5 5 1 ! subdivisions N_1, N_2 and N_3 along recipr. l. vectors
0. 0. 0. ! optional shift of the mesh (s_1, s_2, s_3)
----------------------------------------------------------

I am using GGA-PBE POTCAR

-----------------------------------------------------------
OSZICAR

N E dE d eps ncg rms rms(c)
DAV: 1 0.185107797915E+04 0.18511E+04 -0.34430E+04 4992 0.883E+02
DAV: 2 0.135649697695E+04 -0.49458E+03 -0.43669E+03 6224 0.213E+02
DAV: 3 0.125940179195E+04 -0.97095E+02 -0.63077E+02 7072 0.794E+01
DAV: 4 0.125237853903E+04 -0.70233E+01 -0.36265E+01 6576 0.222E+01
DAV: 5 0.125202771687E+04 -0.35082E+00 -0.21108E+00 8192 0.593E+00
DAV: 6 0.125202228372E+04 -0.54332E-02 -0.43197E-02 6816 0.732E-01
DAV: 7 0.125202197656E+04 -0.30715E-03 -0.30128E-03 8368 0.187E-01
DAV: 8 0.125202196671E+04 -0.98488E-05 -0.97093E-05 5840 0.246E-02
DAV: 9 0.125202196585E+04 -0.86018E-06 -0.85753E-06 5328 0.735E-03
DAV: 10 0.125202196576E+04 -0.91651E-07 -0.91716E-07 4816 0.350E-03
DAV: 11 0.125202196574E+04 -0.25576E-07 -0.25263E-07 4496 0.221E-03
DAV: 12 0.125202196572E+04 -0.13767E-07 -0.13794E-07 4128 0.164E-03
DAV: 13 0.125202196571E+04 -0.85947E-08 -0.87233E-08 3824 0.122E-03
DAV: 14 0.125202196571E+04 -0.58340E-08 -0.58831E-08 3744 0.966E-04
DAV: 15 0.125202196570E+04 -0.41064E-08 -0.41198E-08 3744 0.754E-04
DAV: 16 0.125202196570E+04 -0.28699E-08 -0.29110E-08 3696 0.610E-04
DAV: 17 0.125202196570E+04 -0.20514E-08 -0.20944E-08 3696 0.485E-04
DAV: 18 0.125202196570E+04 -0.17353E-08 -0.14974E-08 3664 0.399E-04
DAV: 19 0.125202196570E+04 -0.11019E-08 -0.10888E-08 3648 0.320E-04
DAV: 20 0.125202196570E+04 -0.87675E-09 -0.78946E-09 3648 0.266E-04
DAV: 21 0.125202196570E+04 -0.27603E-09 -0.57878E-09 3648 0.214E-04
DAV: 22 0.125202196569E+04 -0.80036E-09 -0.42507E-09 3600 0.179E-04 0.930E+01
DAV: 23 0.347249842825E+04 0.22205E+04 -0.72837E+01 5712 0.281E+01 0.825E+01
DAV: 24 0.746490575155E+04 0.39924E+04 -0.65087E+04 9760 0.353E+02 0.201E+02
DAV: 25 0.725622605393E+04 -0.20868E+03 -0.88574E+03 7104 0.836E+02 0.213E+02
DAV: 26 0.670722726666E+04 -0.54900E+03 -0.23758E+03 7520 0.384E+02 0.224E+02
DAV: 27 0.619218718657E+04 -0.51504E+03 -0.77523E+02 8784 0.262E+02 0.228E+02

alex · #2 Post by **alex** » Mon May 23, 2011 7:32 pm

I'd guess you go better with only 50% of INCAR. Quite bulky file.

Do you know every switch you are setting and possible interactions with other ones??

Your maximum mixing is close to zero. Quite weird.

If you need all that, you probably have a very complex electronic structure.
But this is _not_ the case for either C32 or (BN)16. Start all over.

Cheers,

Alex

#3 Post by **admin** » Mon May 23, 2011 7:48 pm

have you checked the interatomic distances in your sheet (written in OUTCAR)? do they correspond to the C--C bondlength in Graphene?

magnon963 · #4 Post by **magnon963** » Tue May 24, 2011 9:18 am

Thank you for your reply.

But the problem is: I did the same calculation at some other place with vasp.5.2 and vasp.4.6. I got correct answers. Now when I'm at new place, and working with vasp.5.2.11, it's not working and giving me + energies. I've attached POSCAR file, and one can test it, all the inter atomic distance are perfect. Convergence was also very slow so I reduced mixing parameters.

If this problem is related to wrong installation then it would have given wrong answer for smaller systems also.

I'm very confused. Please suggest me some way to sort out this problem.

best regards,
PS