[QE-users] How i can get correct band gap by GGA+U method ?

[Giuseppe Mattioli]

Dear Mohammed
I've found 3.9 eV for Mn (with PAW PP) a few years ago for a very  
similar layered Mn oxide, and applied the same value to Mn(IV) in  
monoclinic Ca2Mn3O8 (J. Am. Chem. Soc. 2015, 137, 10254, see SI). This  
said, if you want to have a better Eg you must apply the U correction  
also to O 2p (you may want to read Phys. Rev. X 2015, 5, 011006, where  
MnO2 is discussed in the framework of a "generalized U" method).  
However, keep in mind two things:
1) When dealing with (almost) full bands such as that arising from O  
2p in Mn oxides, you should not use linear response to calculate U,  
and you should use some semiempirical fitting instead (if it pleases  
you...).
2) If your fitting parameter is the band gap, then be careful to check  
that you have a good position of *both* band edges, maybe by comparing  
results obtained with the same PPs and same cell parameters with those  
obtained by using, eg. the HSE functional. It may happen (it happened  
to me in many cases where inter-site contributions as U+V could not be  
avoided, like GaAs or Si) that the U correction lowers the VB edge  
*and* the CB edge, maybe leading to a correct gap in a wrong overall  
potential landscape.
HTH
Giuseppe

Quoting Lorenzo Paulatto <paulatz at gmail.com>:

>> I have calculated the cell parameters and band-gap energy by GGA  
>> and GGA+U for CaMnO3 materiels (orthorhombic structure ). I tried  
>> two U values, one from the literature (U1=4.18 eV) and I have  
>> calculated the other one by the first principal method with the  
>> quantum espresso package (U2=5.7722 eV ).  The band-gap energies  
>> whose I have calculated by GGA, GGA+U1 and GGA+U2 are 0.71eV, 0.80  
>> eV and 0.88 eV respectively. However, these values are very small  
>> in comparison to the experimental result Eg=3.07 eV.
>
> Hello, I would proceed in two directions:
>
> 1. examine the band structure with projwfc, in order to determine  
> which is the character of the states close to the gap (es. Mn d, or  
> Oxygen O...). The applying a +U to those states can open the gap.
>
> 2. add an arbitrary but not absurd U to all species and see if I can  
> manage to open the gap in this way or if it is absolutely impossible
>
>
> cheers
>
>
>>
>> U2  was calculated by hp.x code in QE (version 6.5) package with  
>> 2*1*2 and 2*2*2 kpoints. In addition,  2*1*2 and 2*2*2 gave the  
>> same U2 value .
>>
>> you can find my input files below
>>
>> --
>> *Best regards,*
>> -------------------------------------
>> *Mohammed Lamhani,* Ph.D.
>> /LS3M Laboratory, Polydisciplinary Faculty of Khouribga, Sultan  
>> Moulay Slimane University of Beni Mellal, Morocco./
>>
>>
>> _______________________________________________
>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
>> users mailing list users at lists.quantum-espresso.org
>> https://lists.quantum-espresso.org/mailman/listinfo/users



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>