La2CuO4 DOS for LDA & LDA+U

Message

b_lack · #1 Post by **b_lack** » Mon Sep 24, 2012 8:17 am

I have a question related to LDA and LDA+U on La2CuO4,
paper says that just using LDA would give a metal conditions where by adding Hubbard correction it would became Insulator.

Wei et al., PRB 49 (1994), 12159

When I tried using VASP the result is already insulator (LDA), even without adding Hubbard corrections, now I kind a bit confuse, which one should I believe?

POSCAR
======
La Cu O
1.00
3.7793000000 0.0000000000 0.0000000000
0.0000000000 3.7793000000 0.0000000000
0.0000000000 0.0000000000 13.2000000000
4 2 8
Direct
0.0000000000 0.0000000000 0.3609000150
0.0000000000 0.0000000000 0.6390999560
0.5000000000 0.5000000000 0.8609000440
0.5000000000 0.5000000000 0.1390999850
0.0000000000 0.0000000000 0.0000000000
0.5000000000 0.5000000000 0.5000000000
0.0000000000 0.5000000000 0.0000000000
0.5000000000 0.0000000000 0.0000000000
0.5000000000 0.0000000000 0.5000000000
0.0000000000 0.5000000000 0.5000000000
0.0000000000 0.0000000000 0.1840000000
0.0000000000 0.0000000000 0.8159999850
0.5000000000 0.5000000000 0.6840000150
0.5000000000 0.5000000000 0.3159999850

INCAR
====
PREC = High
IBRION = 2
NSW = 0
ISIF = 2
ISMEAR = 0
SIGMA = 0.05
LREAL = Auto

KPOINTS
======
LCO
0
M
12 12 24
0 0 0

jlbettis · #2 Post by **jlbettis** » Mon Sep 24, 2012 8:56 pm

Is La2CuO4 experimentally determined to be an insulator? What version of VASP are you using? What version is Wei et al. using?

b_lack · #3 Post by **b_lack** » Tue Sep 25, 2012 9:59 am

Yes La2CuO4 are Insulator experimentally, but theoretically metal, and that the reason Hubbard corrections needed for that (LDA+U) so it would show Insulator case

My version is 4.6, but Wei et al didn't used VASP, but I'm just curious and a bit confuse also, ^_^