The ABINIT Users Mailing List ( CLOSED )

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

Dear Zhufeng,

It looks to me as if you didn't set "intstrflag 1" in your anaddb data.  
This will prevent the relaxed-ion elastic tensor from being calculated 
properly, or perhaps at all.  The internal strain (intstr) will only 
effect C44 for this lattice, and must reduce it.  So try this first.  
You shouldn't have to redo your RF run. 

The variables "polflag," "relaxat,"  "relaxstr," "istrfix," "nstrfix," 
and "targetpol" are not relevant to an elastic constant calculation and 
should not be set.

A few other comments:  Since you only want elastic output (and Si isn't 
piezoelectric anyway), you can eliminate the dataset 2 ddk calculation, 
and eliminate "rfelfd" and "getddk" from the RF strain/atomic 
displacement calculation.  Rather than fit an EOS to get the relaxed 
lattice constant, you can use "ionmov 2" and "optcell 1" in a GS 
calculation (see tutorial 3).

Your k sample and cutoff look fine to me, but HGH psp's are VERY hard.  
You could try another psp.  If you use an FHI psp with a nonlinear core 
correction, be sure to set "optnlxccc 2" in both GS and RF 
calculations.  If you still don't like your results compared to 
experiment, try GGA instead of LDA (available for strain in 4.5.2, see 
Test_v4/t58-59).

Regards,
Don Hamann


zfhou@mail.edu.cn wrote:

> Dear Dr. Hamann,
>
> Thanks very much!
>
> According to your suggestion and the exmaples in Test_v4, I have calculated 
> the elastic constant of Si again by abinis and anaddb.
>
> The C11, and C12 that I got is  160.77 GPa and 63.12 GPa,respectively, which 
> are very close the experiment value 166.0 GPa and 64 GPa [1], respectively. 
> But The C44 is 108.4GPa and is larger than the experiment value 79.6GPa [1].
>
> The warning comments from the outputs tell me that I should increase the 
> cutoff and k mesh.  But I think that the ecutoff and k-points sampling I used 
> is sufficient for this calculation.  So I think the reason is from other 
> parameters in my input file.
> But until now I haven't found out the real reason.
>
> The folloing is the warning in my output file.
>
> "phonon modes warning- :
> accoustic sum rule violation met:the eigenvalues of accoustic mode
> are too large at Gamma point
> increase cutoff energy or k-points sampling."
>
>
> I used the HGH pseudopotential, CA-LDA (ixc =2 ), the lattice constant of 
> 10.1738 a.u.(in experiment, a0=10.261 a.u. for Si) which is optimized and 
> fitted by EOS, Ecut of 12 Ha, k mesh of 8 x 8 x 8 corresponding to 60 k 
> points in IBZ.
>
> By EOS, I got the bulk modulus of silicon (diamond strucutre) is 95.497 GPa, which is very close to the experimental value of 98 GPa (B = (C11+2C12)/3.0) for cubic crystals). 
>
> [1]. A. Darys and J. Kundrotas "handbook on physical properties Ge, Si, GaAs  and 
> InP" (Vilnius: Science and Encyclopedia Publisher)
>
> Regards
> zhufeng
>
> ---------the input for abinis--------
> ndtset 3 
> # Set 1 : initial self-consistency
> kptopt1   1
> tolvrs1 1.0d-18
> iscf1 5
>
> #set 2 the ddk 
> rfelfd2 2  # only derivative of the gs wf
> rfdir2 1 1 1
> nqpt2 1
> qpt2 0.0 0.0 0.0 #only the gamma point
> getwfk2 -1
> kptopt2 2
> iscf2 -3
> tolwfr2 1.0d-20
>
> # Set 3 : response-function strain calculation
> rfphon3 1
> rfatpol3 1 2  
> rfelfd3 3
> rfdir3 1 1 1
> nqpt3 1
> qpt3 0 0 0
> rfstrs3  3
> getwfk3 -2
> getddk3 -1
> kptopt3 2
> tolvrs3 1.0d-16
> iscf3 5
>
> #comon input data
> diemac   12.0
>   ecut   12.0
>   ixc 2
>   iscf   5
>  nband   12
>  ngkpt   8  8  8
> nloalg   4
>  nstep   24
> prtvol   10
> nshiftk 4                       
> shiftk 0.5 0.5 0.5
>       0.5 0.0 0.0
>       0.0 0.5 0.0
>       0.0 0.0 0.5
>
> #Definition of the SCF procedure
> acell 3*10.1738
> rprim  0.0  0.5  0.5    
>       0.5  0.0  0.5    
>       0.5  0.5  0.0   
> ntypat   1 
> typat   2*1
> natom   2
> xred           
>   0.0  0.0  0.0       
>   0.25 0.25 0.25     
> znucl 14 
> --------------------END----------------------
> ------------Input for anaddb----------
> # General informations
> #**********************
>   rfmeth  1
>   enunit  0
>   eivec   1
>   asr     1
>   chneut  2
>   slectz  0
>
> # Flags
> #*******
>   dieflag  2
>   ifcflag  0
>   thmflag  0
> elaflag  3 
>
> polflag 1
> relaxat 1
> relaxstr 1
> nstrfix 2
> istrfix 1 2
> targetpol 0.0 0.0 0.45
>
> # Frequence information
> #***********************
>   nfreq   100
>   frmin   0 
>   frmax   2E-3
>
> # Interatomic force constant information
> #****************************************
>   dipdip  1
>   nspher  0
>   ifcout  0
>   ifcana  0
>   natifc  2
>   atifc   1 2
>
> # Thermal information  (needed if thmflg is 1)
> #**********************************************
>   nchan       500
>   nwchan      10
>   dostol      0.1 
>   thmtol      0.05
>   ntemper     300
>   tempermin   10.0
>   temperinc   1.0
>
> # Wavevector grid number 1
> #**************************
>   brav  1
>   ngqpt 4 4 4
>   nqshft 1
>   q1shft 0 0 0
>
> # Wavevector grid number 2
> #**************************
>   ngrids 12 
>   ng2qpt 24 24 24
>
> # Wavevector list number 1 
> #**************************
>   nph1l 1
>   qph1l 0 0 0 1
>
> # Wavevector list number 2 (Gamma point only, with limiting direction
> #           in cartesian coordinates. )
> #*********************************************************************
>   nph2l 3
>   qph2l 1.0  0.0  0.0  0
>         0.0  1.0  0.0  0
>         0.0  0.0  1.0  0 
> --------END-----------------------------------
>
>  

--
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