The ABINIT Users Mailing List ( CLOSED )

[Yong Liu <occupierliu@yahoo.com.cn>]

Dear Duan:

       From your input file of RF calculation, the ecut is not good enough to generate a convergent result even I cann't figuer out the type of your pseudopental. You must test the convergence of the ecut carefully before the GS or RF calculation. Have you done the fully structral optimization? Please do it before your RF calculation. The un-relaxed structral parametres will generate the unstable or wrong results, for instance, the relaxed elastic constaits, c11 and c33, in your anaddb output file are negitive which are so not physical reasonable. And I guess the acoustic phonon frequencies at the gamma point are negative, this means your structure is unstable and not suitable for the RF calculation, especially for the properties calculation corresponding to the acoustic behavior such as elastic and piezolectric constants.

duan_semi <abinit_duan@126.com> 写道：

Dear abinitusers,

           I want to calculate the Proper piezoelectric constants (clamped ion and relaxed ion) of tetragonal phase of PbTiO3. I set my input files just following the example of AlAs in tutorial (the input file for RF calculation ,that for "anaddb" and the output file after using "anaddb" are the attached files). the calculation of RF is successful and the "DDB" file is gotten. After the analysis of "anaddb"  is finished, there is a warning message in the output file of "anaddb":

    "piezo9 : WARNING -
   Unstable eigenvalue detected in force constant matrix at Gamma point
   The system under calculation is physically unstable."

I think the problem maybe affect the precision and validity of   piezoelectric constants. Is what I think is right? what I should do to solve it?

Best regards,

WD Duan      

 

 

 

Sincerely.
Yong Liu

---

雅虎免费邮箱-3.5G容量，20M附件