Dear Stefano, The fcc unit cell isn’t it 4 atoms per cell? Actually I have tried to set the braves lattice to sc with same input. And it shows me a more better result. But for Pd, it can be easily find that it’s fcc structure. Maybe I misunderstand something. Please correct me. Thank you again!
xiao > On 20 Apr 2015, at 23:20, stefano de gironcoli <[email protected]> wrote: > > Dear Xiao, > there is a factor of 4 in the volume per unit-cell between sc and fcc. > just geometry. > I don't really know what you have calculated but the energy one should > feed to ev.x is the energy per unit cell as you have used in the pw.x > calculation. So if in the ev.x input you say that your cell is fcc then you > should use the energy computed with the fcc unit cell (1 Pd atom per cell). > If your energy was computed in the sc unit cell (that has 4 Pd atoms in the > unit cell) then you should specify sc in the ev.x input. just be consistent. > > stefano > > On 20/04/2015 22:31, Xiao wrote: >> Dear Stefano, >> Yes! I used a fcc unit cell. It should have a factor 4. Thank you!!! >> But can you explain why there is a 4 times relation between fcc and sc? >> I don’t really understand why it happens. >> >> best >> xiao >>> On 20 Apr 2015, at 22:09, stefano de gironcoli <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> Dear Xiao, >>> are you using the fcc unit cell (1 atom/cell) or the SC one (4 >>> atoms/cell) ? >>> that would make a factor of 4 in your bulk modulus. >>> stefano >>> >>> On 20/04/2015 21:34, 杨潇 wrote: >>>> Dear All, >>>> >>>> Sorry for my mistake of previous email! It should be Ry I used. >>>> >>>> As I am using the ev.x to fitting EOS on Pd. I got the wrong bulk modulus. >>>> Here is my procedure. Please help me check what’s wrong with my fitting? >>>> input data in Å and Ry. >>>> >>>> 3.7351 -1.455513098 >>>> 3.8129 -1.5041846 >>>> 3.8907 -1.527184909 >>>> 3.9685 -1.530334715 >>>> 4.0463 -1.51819883 >>>> 4.1241 -1.4945274 >>>> >>>> Lattice parameter or Volume are in (au, Ang) > Ang >>>> Assuming Angstrom >>>> Enter type of bravais lattice (fcc, bcc, sc, noncubic) > fcc >>>> Enter type of equation of state : >>>> 1=birch1, 2=birch2, 3=keane, 4=murnaghan > 2 >>>> then I get the output file like below: >>>> >>>> # equation of state: birch 3rd order. chisq = 0.3034D-10 >>>> # a0 = 7.4514 a.u., k0 = 6726 kbar, dk0 = 5.51 d2k0 = -0.001 emin = >>>> -1.53118 >>>> # a0 = 3.94309 Ang, k0 = 672.7 GPa, V0 = 103.430 (a.u.)^3, V0 = 15.327 >>>> A^3 >>>> >>>> ######################################################################### >>>> # Lat.Par E_calc E_fit E_diff Pressure Enthalpy >>>> # Ang Ry Ry Ry GPa Ry >>>> ######################################################################### >>>> 3.73510 -1.45551 -1.45551 0.00000 170.75 -0.43510 >>>> 3.81290 -1.50418 -1.50418 -0.00000 89.34 -0.93621 >>>> 3.89070 -1.52718 -1.52719 0.00001 30.14 -1.32358 >>>> 3.96850 -1.53033 -1.53033 -0.00001 -12.29 -1.61843 >>>> 4.04630 -1.51820 -1.51820 0.00001 -42.10 -1.83804 >>>> 4.12410 -1.49453 -1.49453 -0.00000 -62.42 -1.99662 >>>> ~ >>>> >>>> The k0 here is 672.7GPa, it should be 180GPa more or less. And I fitted >>>> same data with another software, it gives me 167GPa. >>>> >>>> So Please Help Me What’s Wrong With My Fitting? >>>> >>>> Thank you!!! >>>> >>>> best >>>> xiao >>>> >>>> >>>> _______________________________________________ >>>> Pw_forum mailing list >>>> [email protected] <mailto:[email protected]> >>>> http://pwscf.org/mailman/listinfo/pw_forum >>>> <http://pwscf.org/mailman/listinfo/pw_forum> >>> _______________________________________________ >>> Pw_forum mailing list >>> [email protected] <mailto:[email protected]> >>> http://pwscf.org/mailman/listinfo/pw_forum >>> <http://pwscf.org/mailman/listinfo/pw_forum> >> >> >> _______________________________________________ >> Pw_forum mailing list >> [email protected] <mailto:[email protected]> >> http://pwscf.org/mailman/listinfo/pw_forum >> <http://pwscf.org/mailman/listinfo/pw_forum> > _______________________________________________ > Pw_forum mailing list > [email protected] > http://pwscf.org/mailman/listinfo/pw_forum
_______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
