[Pw_forum] Cu100 WRONG PHONON FREQUENCIES

[Andrea Dal Corso]

On Mon, 2005-04-04 at 21:45 -0500, Alcantara Ortigoza, Marisol wrote:
> Dear all,
> 
> I have done phonon calculations on Cu(100) with both PBE and PZ
> pseudopotentials to compare them with other calculations (using LDA
> pseudopotential, 11 layers) and experimental results. These previous
> calculations are in good agreement with experiments and were done with a
> different code. 
> 
> So, I did 11 layers c(1x1) using PZ and,  9 and 13 layers using PBE
> 
> The relaxation of the systems was in good agreement with both experiment
> and the other calculation in both cases. The problems came with the
> phonon calculations. 
> 
> ==============================================================
> 
> X point
> 
> 9 layers PBE
> 
> Highest phonon frequency : OK
> Surface phonons (S1,S2, S6) are from 1-4 meV above the expected values,
> S4 is OK.
> 
> 13 layers PBE
> 
> Highest phonon frequency : 1 meV below expected value
> Surface phonons (S1,S2,S4) are 2-3 meV above ( 2 meV  below) the
> expected values, S6 is OK.
> 
> 11 layers PZ
> 
> Highest phonon frequency : 2 meV above expected value
> Surface phonons are all 2 meV above the expected values
> 
> ===================================================================
> 
> M point
> 
> 9 layers PBE
> 
> Highest phonon frequency : 2 meV above expected value
> Surface phonons (S1,S2,L1) are from 1 meV above (below) the expected
> values
> 
> 13 layers PBE
> 
> Highest phonon frequency : 2 meV above expected value
> Surface phonons are OK
> 
> 11 layers PZ
> 
> Highest phonon frequency : 0.5 meV above expected value
> Surface phonon S1 is 2 meV above the expected value.
> =====================================================================
> 
> The relaxation of these systems were performed until the forces were
> below ~0.0008 Ryd/au
> 
> K-point mesh 15x15x1
> Ecutwfc = 30 for PBE (40 for PZ)
> Ecutrho = 360 for PBE (480 for PZ)
> Width of intersurface vacuum space =  14 A
> 
> The phonon calculations were perforemed with
> 
>   tr2_ph=1.0d-12,

Please check the convergence with respect to this parameter.
If you are not applying the Acoustic sum rule low frequencies 
at X and M can be quite sensitive to tr2_ph.



>   alpha_mix(1) = 0.3
> 
> 
> So, I would like to know if you have done similar calculations and, if
> that is the case, the results you have found. Otherwise, I would like to
> know if from your experience with other systems you would expect one may
> get wrong phonon frequencies in this case. Or, maybe, you might detect
> something wrong within my calculation which could be problem.
> 
> Thank you very much.
> 
> Best regards.
> Marisol Alcantara Ortigoza.
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
-- 
Andrea Dal Corso                    Tel. 0039-040-3787428
SISSA, Via Beirut 2/4               Fax. 0039-040-3787528
34014 Trieste (Italy)               e-mail: dalcorso at sissa.it