The ABINIT Users Mailing List ( CLOSED )

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

Dear Rongqin,

The restrictions you quote deal specifically with the strain perturbation,
rfstrs>0, new in 4.2 and 4.3.  Other, older perturbations (phonon and
electric field) may have some restrictions, but not all of the list you
quote.  For strain, spin polarization should be available in 4.4, and
all self-consistency algorithms will be available for metals.  (All are
presently available for insulators.)

Best regards,
Don Hamann

Wu Rongqin wrote:

Dear all,

 

When I read abini（respfn）, it tells something like this:

Limitations of the present implementation :

• Insulators and metals are possible, but some caution is in order for metals : Only simple mixing (iscf=2) or Anderson mixing (iscf=3) may be used for self-consistency. diemix<=0.5 is recommended.
• LDA only. GGA xc is not yet coded.
• Non-spin-polarized systems. Spin polarization has been carried along in the code modifications, but not tested.
• No spin-orbit coupling. This has not been coded, and appears to be quite challenging.
• Legendre polynomial treatment of non-local psps (useylm=0) only.
• Symmetry is presently used to skip redundant k points in the BZ sum, but not to skip redundant strain perturbations.

The 3^rd point tells that non-spin-polarized systems can be done by abinit, but there is a publication using abinit to do response calculation:

Physical review B, volume 65, 214302, by M.Veithen and Ph.Ghosez.

In which the ferroelectric phase of LiNbO3 is calculated.

So I want to know on earth whether abinit, the lasted version, can work on spin-polarized system or not.

 

Rongqin Wu

-- 
D. R. Hamann				Phone: 908-582-4454
Director, Theoretical Materials    	Fax:   908-582-4702
  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