sorry for late reply. so what is your suggestion for getting the accurate value for cohesive energy for W. I also tried another BCC material Ta and its cohesive energy is also more than 22% higher than the reported value (kittel). But when i calculated the cohesive energy for FCC Ni and Al the difference from experimental result is around 6 % (on the higher side)
Thanks in advance Vicky singh Bangalore On Tue, Dec 7, 2010 at 2:28 PM, Gabriele Sclauzero <sclauzer at sissa.it>wrote: > Dear Mohnish > > Il giorno 06/dic/2010, alle ore 07.51, mohnish pandey ha scritto: > > Dear Vicky! > Check the convergence of energy of isolated atoms also. > Normally convergence threshold for energy for isolated species is higher > than the crystal system. > > > I'm sorry, but I hope you'll let me disagree on this point. What makes you > say this? > > Check for different values of ecut and use the converged value. > > > One usually uses the same ecutwfc value for both bulk and atomic > calculations, in order to take advantage of some error cancellation. Indeed, > if you plot the difference between the bulk and atomic total energies as a > function of ecutwfc, you should see that it converges much faster than any > of the two energies taken alone. > > And next thing is you are doing calculations for bulk system so all the > degrees of freedom should be relaxed for geometry optimization so instead of > flag "relax" use "vc-relax".. The former one will converge the forces but > not the stress and latter one converges the stress also... > > > I agree that the "vc-relax" option is useful, but maybe not so much in this > case. I think that bulk W is bcc, hence it can be studied with one atom per > cell and there are no internal parameters to relax. You can simply scan the > total energy as a function of the lattice parameter to get the equilibrium > value, as Vicky already did. > > There are many other issues that come in, especially in the atomic > calculation (see recent discussions on the forum). Even when you get a > converged DFT value, in some cases this might be considerably far from the > experimental value. > > Best regards, > > GS > > On Sun, Dec 5, 2010 at 11:53 AM, vicky singh <kirtinandan07 at > gmail.com>wrote: > >> Hi >> >> I am trying to calculate the cohesive energy for tungsten. the code is >> attached. I am able to get the lattice parameter of 3.19 ang while >> experimental is 3.16 angs. Then i calculated the cohesive energy of w by >> calculating the energy of isolated w atom by changing lattice parameter up >> to 25. I got the cohesive energy of 11.1 ev/atom while reported value is 8.9 >> ev/atom. This is way too high. can anybody help me. I have varied the K >> point as well as The E_cut but no advantage. >> >> vicky singh >> research student >> Bangalore >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> > > > -- > Regards, > MOHNISH, > ----------------------------------------------------------------- > Mohnish Pandey > Y6927262,5th Year dual degree student, > Department of Chemical Engineering, > IIT KANPUR, UP, INDIA > ----------------------------------------------------------------- > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > > ? Gabriele Sclauzero, EPFL SB ITP CSEA > * PH H2 462, Station 3, CH-1015 Lausanne* > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20101217/554922d3/attachment.htm
