[QE-users] Absorption spectrum using TDDFPT (turbo_lanczos.x)

[Giuseppe Mattioli]

Dear Vivek
You simply cannot run TDDFPT calculations in systems requiring  
occupations='smearing' for the convergence of scf (by the way, you can  
have "metallic" systems that don't contain "metals", for example  
shallow donors/acceptors in heavily doped covalent semiconductors...).  
Keep also in mind that GGA-TDDF(P)T is not particularly suitable for  
the calculation of absorption spectra in periodic systems and, AFAIK,  
TDDFPT with EXX functionals and k-points is quite expensive. You  
should either use a GW-BSE approach (maybe using YAMBO on top of the  
QE GGA ground state), or rely on a less sophisticated  
independent-particle approach using epsilon.x.
HTH
Giuseppe

Quoting Vivek Christhunathan <vivekppn at gmail.com>:

> Dear Oscar,
>
> Thanks for the reply.
>>> turbo_lanczos doesn't work with metals, so you have to define
> coccupations='fixed'.
>
> I even tried running the SCF calculation using coccupations='fixed', but I
> can not disable the mentioned WARNINGs.
>
>>> But in this case your problem is that scf calculation is not converged.
>
> If my understanding is correct, you're telling me to keep running the SCF
> calculation until it converges, despite the WARNINGs. I tried that, but it
> isn't converged; the calculation is simply taking CPU hours.
>
> Thanks.
>
> Kind Regards,
>
> Vivek C
>
>
> __________________________________
>
> Vivek Christhunathan
>
> PhD researcher
>
> Mechanical Engineering
>
> College of Engineering and Informatics
>
> Alice Perry Engineering Building
>
> National University of Ireland Galway, Ireland
> E-mail: v.christhunathan1 at nuigalway.ie
> Phone: +353 899811181, +91 9600752742
>
>
>
> On Wed, Mar 30, 2022 at 5:41 PM Vivek Christhunathan <vivekppn at gmail.com>
> wrote:
>
>> Hello Everyone,
>>
>> I am calculating absorption spectrum calculation for the system "CaWO4 and
>> Cu-doped CaWO4". I could do the turbo_lanczos.x for the CaWO4 system and
>> even plotted the absorption spectrum using turbo_spectrum.x, but I could
>> not do the same for the Cu-doped CaWO4 system because turbo_lanczos.x stops
>> me by showing the following message "Error in routine lr_readin  
>> (1): turboTDDFT
>> is not extended to metals". If that is the case, how did it work for the
>> CaWO4 system? Because CaWO4 has metal in it.
>>
>> Then I try to run the SCF calculation without specifying occupations= '',
>> smearing= '', and degauss= _, the following have appeared in the output file
>>
>> gamma-point specific algorithms are used
>>
>> Message from routine setup:
>> *the system is metallic, specify occupations*....
>> .
>> .
>> .
>>
>> *WARNING: integrated charge=  1544.80000000, expected=  1545.00000000*
>>
>> If I include "occupations=" in my input file, the warning disappears, but
>> not able to do turbo_lanczos.x. I am stuck in this loop and I am still
>> struggling with this.
>>
>> Input and output files are attached here for your kind perusal.
>>
>> Any help would be appreciated.
>>
>> Thanks in advance.
>>
>> Input:
>>
>>
>> Kind Regards,
>>
>> Vivek C
>>
>>
>> __________________________________
>>
>> Vivek Christhunathan
>>
>> PhD researcher
>>
>> Mechanical Engineering
>>
>> College of Engineering and Informatics
>>
>> Alice Perry Engineering Building
>>
>> National University of Ireland Galway, Ireland
>> E-mail: v.christhunathan1 at nuigalway.ie
>> Phone: +353 899811181, +91 9600752742
>>
>>



GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: <giuseppe.mattioli at ism.cnr.it>