The ABINIT Users Mailing List ( CLOSED )

["D. R. Hamann" <drh@bell-labs.com>]

Dear ysesta(?),

I'm not sure about your convergence problems, but are you using 
pseudopotentials with model core charges to deal with the 
Louie-Froyen-Cohen non-linear core correction? Failing to use this 
systematically overemphasizes the tendency towards magnetism in 3d 
materials.

Also, 800eV (29 Har) is a rather low cutoff for 3d.

Don Hamann

ysesta00@stud.fme.vutbr.cz wrote:

> Dear Abinit users, 
> I am doing some computations about Ni-Ti alloys at B2 structure and I have little problem with it. My computations have too many iterations (25 - 50 it is depending on alat) and predict magnetic moment (not observed experimentally). I think, that this number of iterations is too large. I have tried to use all three type of pseudopotencials (Troullier-Martins LDA, FHI LDA and GGA), but all of them predict mag. moment. Is there someone who has experience with such computations? 
> Could it be better if I generated my own file with pseudopotential using "The OPIUM pseudopotentials generator"? (I haven’t done this before). Actually, when I computed pure Ti, using FHI LDA and GGA type pseudopotentials, it predicted mag. moment, only Troulier-Martins LDA predicted correct nonmagnetic properties. 
> Thanks for answer.
>
> This is my input file for Abinit (non-magnetic case):
> #---structure_definition--#
> acell 3*3.02 Angstrom
> andeg 90 90 90
> xred           
>     0.0 0.0 0.0
>     0.5 0.5 0.5
> ntypat 2
> znucl 22 28
> natom 2
> typat 1 2
> #nsym 1
> #symafm -1
> #chkprim 0
>
> #---energy_k-point-definition_mixing-energy_...--#
> tsmear 0.01
> iscf 2
> diemix 0.15 #when diemix is higher than 0.2 Abinit dont converge for this case
> ecut 800 eV
> kptopt 1
> ngkpt 15 15 15
> nstep 150
> toldfe 1.0d-5 eV
> #prtdos 2
> fband 1.5
>
> #---magnetic_computation--#
> occopt 4
> #nsppol 2
> #nspden 2
> #nspinor 1
> #nband 1 1
> #spinat 0.0 0.0 0.5 0.0 0.0 0.5
>
> #---optimalization_structure--#
> #optcell 1
> #ionmov 3
> #ecutsm 0.5
> #ntime 35
> #dilatmx 1.1
>
> #---unsorted_definition--#
> diemac 1.0d+6
>  


--
D. R. Hamann                            Phone: 908-582-4454
Director, Theoretical Materials         Fax:   908-582-4868
 Physics Research (retired)             email: drh@physics.bell-labs.com
Bell Laboratories 
Lucent Technologies
700 Mountain Ave, Room 1D-371
Murray Hill, NJ 07974-0636 USA