The ABINIT Users Mailing List ( CLOSED )

[Vincent Chevrier <vincent.chevrier@dahn.phys.dal.ca>]

Dear Bernard,
thank you for your reply. I've rerun with toldfe=1e-9. But etotal does not
always agree with Etotal(DC). See below, in all cases tsmear=0.01 and the
runs only differ in the parameters stated.

toldfe=1e-5, tphysel=0
    >>>>>>>>> Etotal= -2.24785447837035E+02
    >>>> Etotal (DC)= -2.24785888234037E+02
    etotal            -2.2478510911E+02

toldfe=1e-5, tphysel=0.002131
    >>>>>>>>> Etotal= -2.24785745210972E+02
    >>>> Etotal (DC)= -2.24784864246079E+02
    etotal            -2.2478486425E+02

toldfe=1e-9, tphysel=0
    >>>>>>>>> Etotal= -2.24785888288957E+02
    >>>> Etotal (DC)= -2.24785888278458E+02
    etotal            -2.2478510947E+02

toldfe=1e-9, tphysel=0.002131
    >>>>>>>>> Etotal= -2.24784859172166E+02
    >>>> Etotal (DC)= -2.24784864956616E+02
    etotal            -2.2478486496E+02

when tphysel=0.002131 then etotal = Etotal(DC), but when tphysel=0 then
etotal != Etotal(DC). Why is this, or more specifically what is:

1 - the cause of the difference between Etotal(DC) and etotal
2 - the effect of tphysel on the difference between Etotal(DC) and etotal

Thanks
Vincent

> Dear Vincent,
>
> With iscf=17 (the default value of iscf),  ETOT
> is equal to the so-called "double counting" expression of the energy
written "Etotal (DC)" in the .out file (you could check it in your .out
file).
>
> Moreover, Etotal and Etotal (DC) (both written in the .out file) are
expected to agree if
> your calculation is well converged.
> In your case, your convergence criterion
> is weak (toldfe 1.d-05) so Etotal and Etotal (DC)
> are not equal. You could
> increase you convergence criterion to check this.
>
> Best regards
> Bernard
>
>
> Vincent Chevrier a écrit :
>> Hi,
>> I am wondering what the difference is between Etotal and etotal for PAW
calcs. I can't remember ever seeing a difference stated on the ABINIT
website. I'm especially interested in knowing how tphysel affects this
difference.
>> I completed two runs with the same structure the only difference being
tphysel=0 and tphysel=0.00213.
>> >From the .out file:
>> For tphysel=0:
>>     Kinetic energy  =  1.72081044902126E+01
>>     Hartree energy  =  1.17823500013938E+01
>>     XC energy       = -3.59810665254476E+01
>>     Ewald energy    = -1.07436906857906E+02
>>     PspCore energy  =  8.93535605840166E+00
>>     Loc. psp. energy= -4.47984809673791E+01
>>     Spherical terms = -7.44950150296981E+01
>>     >>>>> Internal E= -2.24785658830422E+02
>>     -kT*entropy     =  2.10993387856660E-04
>>     >>>>>>>>> Etotal= -2.24785447837035E+02
>> and
>>     etotal  -2.2478510911E+02
>> For tphysel=0.002131
>>     Kinetic energy  =  1.72075689848333E+01
>>     Hartree energy  =  1.17845516746048E+01
>>     XC energy       = -3.59818457081452E+01
>>     Ewald energy    = -1.07436906857906E+02
>>     PspCore energy  =  8.93535605840166E+00
>>     Loc. psp. energy= -4.48011025224259E+01
>>     Spherical terms = -7.44958705044103E+01
>>     >>>>> Internal E= -2.24788248875047E+02
>>     -kT*entropy     =  2.50366407550064E-03
>>     >>>>>>>>> Etotal= -2.24785745210972E+02
>> and
>>     etotal  -2.2478486425E+02
>> below is the input file. Can anybody help me understand:
>> 1 - the cause of the difference between Etotal and etotal
>> 2 - the effect of tphysel on the difference between Etotal and etotal
Thanks
>> Vincent
>> Input file:
>>     ndtset   2
>>    tphysel1  0.002131
>>    tphysel2  0.0
>>     occopt   4
>>     tsmear   0.01
>>  pawecutdg   15
>>       ecut   15
>>     kptopt   1
>>    nshiftk   1
>>     shiftk   0.5 0.5 0.5
>>   kptrlatt   6  0  0
>>              0  3  0
>>              0  0 10
>>      nstep   50
>>     toldfe   1.0d-5
>>     ntypat   2
>>      znucl   3 14
>>      natom   34
>>      typat   1  1  1  1  2  1  1  1
>>              2  1  1  1  2  1  1  1
>>              2  1  1  1  1  2  1  1
>>              1  2  1  1  1  2  1  1
>>              1  2
>>      acell   1.5065427965E+01  2.8615352526E+01  8.4083802112E+00 Bohr
rprim   1.0 0.0 0.0
>>              0.0 1.0 0.0
>>              0.0 0.0 1.0
>>       xred   7.4996953092E-13  5.0000000000E-01  0.0000000000E+00
>>              9.6480532055E-02  1.9420479791E-01  0.0000000000E+00
2.7080350417E-01  3.4610078252E-01  0.0000000000E+00
1.5292044305E-01  2.5787557735E-02  0.0000000000E+00
4.1748605305E-01  1.6031572532E-01  0.0000000000E+00
2.5955975937E-01  9.5181187368E-02  5.0000000000E-01
4.0978302750E-01  2.5635768786E-01  5.0000000000E-01
9.6925526499E-02  3.9393640703E-01  5.0000000000E-01
4.2837868374E-01  4.3177429281E-01  5.0000000000E-01
9.0351946795E-01  8.0579520209E-01  0.0000000000E+00
7.2919649583E-01  6.5389921748E-01  0.0000000000E+00
8.4707955695E-01  9.7421244227E-01  0.0000000000E+00
5.8251394695E-01  8.3968427468E-01  0.0000000000E+00
7.4044024063E-01  9.0481881263E-01  5.0000000000E-01
5.9021697250E-01  7.4364231214E-01  5.0000000000E-01
9.0307447350E-01  6.0606359297E-01  5.0000000000E-01
5.7162131627E-01  5.6822570719E-01  5.0000000000E-01
5.0000000000E-01  0.0000000000E+00  0.0000000000E+00
4.0351946795E-01  6.9420479791E-01  0.0000000000E+00
2.2919649583E-01  8.4610078252E-01  0.0000000000E+00
3.4707955695E-01  5.2578755773E-01  0.0000000000E+00
8.2513946947E-02  6.6031572532E-01  0.0000000000E+00
2.4044024063E-01  5.9518118737E-01  5.0000000000E-01
9.0216972503E-02  7.5635768786E-01  5.0000000000E-01
4.0307447350E-01  8.9393640703E-01  5.0000000000E-01
7.1621316266E-02  9.3177429281E-01  5.0000000000E-01
5.9648053206E-01  3.0579520209E-01  0.0000000000E+00
7.7080350417E-01  1.5389921748E-01  0.0000000000E+00
6.5292044305E-01  4.7421244227E-01  0.0000000000E+00
9.1748605305E-01  3.3968427468E-01  0.0000000000E+00
7.5955975937E-01  4.0481881263E-01  5.0000000000E-01
9.0978302750E-01  2.4364231214E-01  5.0000000000E-01
5.9692552650E-01  1.0606359297E-01  5.0000000000E-01
9.2837868374E-01  6.8225707188E-02  5.0000000000E-01
>
>