Thank you so much dear Caterina for answering

1. My first question was surmounted with your help.

2. My mean of the second problem is:
EPSILON_XXX.OUT of "BSE-singlet" shows some excitonic features below the band gap. But with a shifting equal to the difference between the first peak in "RPA" and the first one in "BSE", my "BSE" and "RPA" spectra were graphically very similar and coincided. Not only for MgTe, but also for more organic and inorganic cases that I tested.
I understand that for some cases it is reasonable (for example in Anthracene, as pointed in C. Hummer et al work), but not for all structures.
For a better conclusion, I have brought my input files:
input file of MgTe for "BSE-singlet":

<input>

<title>MgTe_BSE_Singlet</title>

<structure speciespath="/opt/exciting/species" autormt="false">

<crystal scale="12.3153454079">
<basevect>0.0 0.5 0.5</basevect>
<basevect>0.5 0.0 0.5</basevect>
<basevect>0.5 0.5 0.0</basevect>
</crystal>

<species speciesfile="Mg.xml" rmt="2.2">
<atom coord="0.00 0.00 0.00"/>
</species>
<species speciesfile="Te.xml" rmt="2.0">
<atom coord="0.50 0.50 0.50"/>
</species>

</structure>

<groundstate

rgkmax="8"
gmaxvr="14"
xctype="GGA_PBE"
nempty="30"
ngridk="10 10 10"
maxscl="300">
</groundstate>

<xs xstype="BSE"
rgkmax="7"
ngridq="4 4 4"
ngridk="4 4 4"
vkloff="0.05 0.097 0.14"
nempty="100"
lmaxapwwf="3"
gqmax="2.7"
broad="0.003674933"
tevout="true" nosym="true">

<energywindow intv="0.0 2.0"
points="2400" />

<screening screentype="full"
nempty="120" />

<BSE bsetype="singlet"
nstlbsemat="1 13 1 16"
nstlbse="1 13 1 16"
aresbse="false"/>

<qpointset>
<qpoint>0.0 0.0 0.0</qpoint>
</qpointset>
</xs>

</input>

and input file of MgTe for "RPA":

<input>

<title>MgTe_Indipendent_Particle_Approximation</title>

<structure speciespath="/opt/exciting/species" autormt="false">

<crystal scale="12.3153454079">
<basevect>0.0 0.5 0.5</basevect>
<basevect>0.5 0.0 0.5</basevect>
<basevect>0.5 0.5 0.0</basevect>
</crystal>

<species speciesfile="Mg.xml" rmt="2.2">
<atom coord="0.00 0.00 0.00"/>
</species>
<species speciesfile="Te.xml" rmt="2.0">
<atom coord="0.50 0.50 0.50"/>
</species>

</structure>

<groundstate

rgkmax="8"
gmaxvr="14"
xctype="GGA_PBE"
nempty="30"
ngridk="10 10 10"
maxscl="300">
</groundstate>

<xs xstype="BSE"
rgkmax="7"
ngridq="4 4 4"
ngridk="4 4 4"
vkloff="0.05 0.097 0.14"
nempty="100"
lmaxapwwf="3"
gqmax="0.0"
broad="0.003674933"
tevout="true" nosym="true">

<energywindow intv="0.0 2.0"
points="2400" />

<screening screentype="full"
nempty="120" />

<BSE bsetype="IP"
nstlbsemat="1 13 1 16"
nstlbse="1 13 1 16"
aresbse="false"/>

<qpointset>
<qpoint>0.0 0.0 0.0</qpoint>
</qpointset>
</xs>

</input>

Dear Hajar,

thank you for posting the input files.

I notice that the first one corresponds to a full BSE calculation (BSE-singlet) while the second to the independent particle approximation, which is different from RPA. In this case you are just considering the diagonal (single particle) transitions, without including the the exchange term, as in the RPA. Also mind that in the two input files you are using different values of rgkmax.

This given, I still believe that there is no problem related to the code. If the two spectra look "graphically very similar" this might most likely be a material dependent feature. I suggest to study carefully your systems and check for comparable results in the literature.

If you have further problems related to exciting, just let us know, and we will do our best to help you out.

Best regards,
Caterina Cocchi
(exciting team)

Thank you so so much
I will check all my cases carefully, again, for certainty.

All the best for you

Hajar

Dear Caterina
I have an another question, please:

Is it possible in exciting-code to calculate "BSE" results using the eigenvectors and eigenvalues obtained by many body G0W0 ?

Thanks

Dear Hajar,

of course it is possible to solve the BSE equation starting from the G₀W₀ electronic structure. To do so, you have to include the gw element in your input file, before xs. To calculate the G₀W₀ electronic structure of your system, follow the related tutorial.

Best regards,
Caterina Cocchi
(exciting team)

Thank you so much dear Caterina

Best wishes

Hajar