<html>Thank you very much.<br />I will try this and return the results (at least 1 week of calculation).<br />For the 'bad' ASR, there is the problem that the defect modes are so close to the acoustic modes that there is a hybridization between them.<br />I suppose that it should drastically affect the ASR.<br /><br />Regards,<br />Antoine Jay<br /><br />Le Samedi, Octobre 30, 2021 11:46 CEST, Lorenzo Monacelli <lorenzo.monacelli@roma1.infn.it> a écrit:<br /> <blockquote type="cite" cite="3313649d-67ec-ee28-108b-9d7a1ed8eefb@roma1.infn.it">Dear Antonine Jay,<br /><br />By looking at your frequencies, they seem not well converged. Consider<br />that in a perfectly converged calculation the ASR should be respected<br />and its application should affect only transnational modes. However, all<br />the phonon frequencies change considerably there. Consider increasing<br />the tr2_ph value to 10^-20 or higher in the ph.x input script and<br />eventually also the wfc/density cutoff. Converging soft modes when a lot<br />of atoms and few symmetries are present in the system could be really<br />cumbersome with DFPT.<br /><br />Bests,<br /><br />Lorenzo<br /><br /><br />On 29/10/21 19:58, Tamas Karpati wrote:<br />> Dear Antoine Jay,<br />><br />> I think that in case of multiple imaginary eigenvalues the lower<br />> energy of a conformer<br />> may be less indicative. On the other hand, frozen phonons might be<br />> both needed and to be avoided,<br />> unless you consider a sufficient number of surrounding atoms. Have you<br />> checked if inclusion<br />> of more atoms helps the imaginaries fade away? Did you try to toss O where the<br />> imaginary eigenvectors suggest? This way imaginaries can be sorted out<br />> one after the other<br />> (lucky ones get rid of them at once :)<br />><br />> Hope this helps,<br />> t<br />><br />><br />><br />><br />> On Fri, Oct 29, 2021 at 10:10 AM Antoine Jay <ajay@laas.fr> wrote:<br />>> Dear users,<br />>> I would like to perform a phonon calculation at Gamma on a 216-atoms crystalline silicon supercell in which has been added 1 single oxygen (PBE functional).<br />>> This interstitial O is located 'nearly' in the middle of two Si, and unfortunately, many local minima are possible for its position: C2, C1, C1h, D3d, each of them being separated by a very small energy barrier and a small distance as already shown in Countinho-2000:<br />>> https://sci-hub.mksa.top/10.1103/physrevb.62.10824<br />>><br />>> When I calculate phonons of the C1h (that has the smallest energy) within DFPT, I obtain imaginary eigenvalues in the dynamical matrix.<br />>> Of course, when I look at the corresponding eigenvectors, I see that they are exactly located on the O.<br />>> Then I manually do a frozen phonon calculation using a displacement equal to the eigenvector and I see a positive curvature of the parabola meaning that the system is well at a minimum.<br />>> Increasing the size of the displacement in my Frozen phonon permits me to see the double minima switching between C2 and C1h.<br />>> I suppose that the size of the perturbation in DFPT is too high...? Where can I change it is the code?<br />>><br />>> Best regards,<br />>><br />>> Antoine Jay<br />>><br />>><br />>> Inputs in attachment<br />>> Results:<br />>> Forces are over converged: Total force = 0.00049 Ry/au<br />>> Frequencies after DFPT are these<br />>> freq ( 1- 1) = -118.6 [cm-1] --> ?<br />>> freq ( 2- 2) = -114.5 [cm-1] --> ?<br />>> freq ( 3- 3) = -99.2 [cm-1] --> ?<br />>> freq ( 4- 4) = -38.0 [cm-1] --> ?<br />>> freq ( 5- 5) = -37.7 [cm-1] --> ?<br />>> freq ( 6- 6) = -37.7 [cm-1] --> ?<br />>> freq ( 7- 7) = 87.7 [cm-1] --> A' I+R<br />>> freq ( 8- 8) = 91.5 [cm-1] --> A' I+R<br />>> freq ( 9- 9) = 91.8 [cm-1] --> A'' I+R<br />>><br />>> and after simple ASR:<br />>> # mode [cm-1] [THz] IR<br />>> 1 -107.94 -3.2359 0.0000<br />>> 2 -38.83 -1.1640 0.0000<br />>> 3 -10.69 -0.3206 0.0000<br />>> 4 -2.67 -0.0801 0.0000<br />>> 5 8.82 0.2643 0.0000<br />>> 6 29.09 0.8720 0.0000<br />>> 7 97.81 2.9324 0.0000<br />>> 8 98.94 2.9661 0.0000<br />>> 9 98.97 2.9670 0.0000 _______________________________________________<br />>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)<br />>> users mailing list users@lists.quantum-espresso.org<br />>> https://lists.quantum-espresso.org/mailman/listinfo/users<br />> _______________________________________________<br />> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)<br />> users mailing list users@lists.quantum-espresso.org<br />> https://lists.quantum-espresso.org/mailman/listinfo/users<br />_______________________________________________<br />Quantum ESPRESSO is supported by MaX (www.max-centre.eu)<br />users mailing list users@lists.quantum-espresso.org<br />https://lists.quantum-espresso.org/mailman/listinfo/users<br /> </blockquote><br /><br /><br /> </html>