Re: [Wien] regarding k-mesh and tolf

2018-01-28 Thread Xavier Rocquefelte 
Here is a link towards a paper which treats such systems using GGA and 
HSE06. Indeed, with GGA the band gap is nearly zero for some of them.


https://arxiv.org/pdf/1706.08674.pdf

The point is that it seems to me that it will not be too much 
problematic although your band gap is zero if the bands are simply 
touching at the Fermi level.


You must check your band structure first.

In the paper they show HSEO6 band structures which clearly confirm that 
GGA+U or onsite-hybrid will not be useful.


One option for you could be to do band structure using GGA and 
Full-hybrid (HSE06 or PBE0) to compare and discuss the differences. Then 
if the difference is only band gap reduction, you could use GGA for the 
phonons calculations. In addition, it will allow you to estimate the 
residual forces using both GGA and hybrid, which will be one more 
confirmation (or not) that GGA is sufficient for your system.


Cheers

Xavier


Le 27/01/2018 à 16:01, chin Sabsu a écrit :

Sir, it is A2BX6,  x is halogen.

Sent from Yahoo Mail on Android 



On Sat, Jan 27, 2018 at 7:23 PM, Xavier Rocquefelte
 wrote:
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at

http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html



___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html


___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] regarding k-mesh and tolf

2018-01-28 Thread chin Sabsu 
 Thanks Sir for your valuable time for me.I will try to use PBE0. I am using 
PBE0  first time and do not know how much time wien2k will take for it. I will 
update here if I get an update.
Anyway, I have calculated phonon bands using PBESol for my two systems (one 
showing zero bandgaps and another showing some band gap) with 1x1x1 supercell. 
The list is not allowing to send docs me so the same I am sending you in your 
private email (anyone else interested can ask me for the bands). The phonon 
band (case_1.pdf , the one which giving zero band gap) is showing that 
structure is thermodynamically not stable while another case (case_2.pdf; the 
one which is showing some band gap) is thermodynamically stable. Anyway, more 
comments can be made after having accurate/PBE0 calculations. I am also sending 
you FORCE_SETS file.
Again thanks a lot for your kind help.
Regards
Chin S.

 
 
On Sunday 28 January 2018, 2:52:34 PM IST, Xavier Rocquefelte 
 wrote:  
 
  
Here is a link towards a paper which treats such systems using GGA and HSE06. 
Indeed, with GGA the band gap is nearly zero for some of them. 
 
 
https://arxiv.org/pdf/1706.08674.pdf
 
The point is that it seems to me that it will not be too much problematic 
although your band gap is zero if the bands are simply touching at the Fermi 
level. 
 
 
You must check your band structure first. 
 
 
In the paper they show HSEO6 band structures which clearly confirm that GGA+U 
or onsite-hybrid will not be useful. 
 
 
One option for you could be to do band structure using GGA and Full-hybrid 
(HSE06 or PBE0) to compare and discuss the differences. Then if the difference 
is only band gap reduction, you could use GGA for the phonons calculations. In 
addition, it will allow you to estimate the residual forces using both GGA and 
hybrid, which will be one more confirmation (or not) that GGA is sufficient for 
your system. 
 
 
Cheers
 
Xavier
 
 
 Le 27/01/2018 à 16:01, chin Sabsu a écrit :
  
Sir, it is A2BX6,  x is halogen.
 
 Sent from Yahoo Mail on Android 
 
  On Sat, Jan 27, 2018 at 7:23 PM, Xavier Rocquefelte 
 wrote:   
___
 Wien mailing list
 Wien@zeus.theochem.tuwien.ac.at
 http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
 SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
  
 
  
 ___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
 
 
 ___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
  ___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] regarding k-mesh and tolf

2018-01-28 Thread chin Sabsu 
Dear Xavier Sir,
Sorry, I understood my mistake in my phonon bands. I must go for higher 
supercell distorted structures and the present bands are not the accurate one. 
However, the rest queries are still looking for the answer.
PBE0 is good for optimization?From your experience how much time it will take 
for optimization? Due to limited facilities, your answer will help me a lot.
One alternative for me is (as adviced by you in a thread in quite past), I will 
optimize the structure from Wien2k and the phonon calculations will be carried 
out on QE.
Thank you very much for the excellent and productive reply.
Chin S.

 

On Sunday 28 January 2018, 3:31:32 PM IST, chin Sabsu  
wrote:  
 
  Thanks Sir for your valuable time for me.I will try to use PBE0. I am using 
PBE0  first time and do not know how much time wien2k will take for it. I will 
update here if I get an update.
Anyway, I have calculated phonon bands using PBESol for my two systems (one 
showing zero bandgaps and another showing some band gap) with 1x1x1 supercell. 
The list is not allowing to send docs me so the same I am sending you in your 
private email (anyone else interested can ask me for the bands). The phonon 
band ( case_1.pdf , the one which giving zero band gap) is showing that 
structure is thermodynamically not stable while another case (case_2.pdf; the 
one which is showing some band gap) is thermodynamically stable. Anyway, more 
comments can be made after having accurate/PBE0 calculations. I am also sending 
you FORCE_SETS file.
Again thanks a lot for your kind help.
Regards
Chin S.

 
 
On Sunday 28 January 2018, 2:52:34 PM IST, Xavier Rocquefelte 
 wrote:  
 
  
Here is a link towards a paper which treats such systems using GGA and HSE06. 
Indeed, with GGA the band gap is nearly zero for some of them. 
 
 
https://arxiv.org/pdf/1706.08674.pdf
 
The point is that it seems to me that it will not be too much problematic 
although your band gap is zero if the bands are simply touching at the Fermi 
level. 
 
 
You must check your band structure first. 
 
 
In the paper they show HSEO6 band structures which clearly confirm that GGA+U 
or onsite-hybrid will not be useful. 
 
 
One option for you could be to do band structure using GGA and Full-hybrid 
(HSE06 or PBE0) to compare and discuss the differences. Then if the difference 
is only band gap reduction, you could use GGA for the phonons calculations. In 
addition, it will allow you to estimate the residual forces using both GGA and 
hybrid, which will be one more confirmation (or not) that GGA is sufficient for 
your system. 
 
 
Cheers
 
Xavier
 
 
 Le 27/01/2018 à 16:01, chin Sabsu a écrit :
  
Sir, it is A2BX6,  x is halogen.
 
 Sent from Yahoo Mail on Android 
 
  On Sat, Jan 27, 2018 at 7:23 PM, Xavier Rocquefelte 
 wrote:   
___
 Wien mailing list
 Wien@zeus.theochem.tuwien.ac.at
 http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
 SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
  
 
  
 ___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
 
 
 ___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

[Wien] 'SELECT' - no energy limits found for atom 2 L= 1

2018-01-28 Thread chin Sabsu 
Dear Peter Sir and any Expert in the form,
Before writing this error I have searched the forum and FAQ and really do not 
find a way to overcome the issue. On FAQ you have written a nice explanation 
but I failed to understand it. The most efficient method I found in the past 
reduction of rmt but now it is also not solving my error. I also used -in1new 2 
option with "run_lapw ... -min" command. Itried to change all 0.30 in case.in1 
by 0.80 but it was not helpful,


The error is coming for Cubic perovskite when I optimize the structure with 
"run_lapw -min" option (while with min_lapw I am not getting any error).
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
**  Error in Parallel LAPW1
**  LAPW1 STOPPED at Mon Jan 29 01:28:36 IST 2018
**  check ERROR FILES!
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0     
 'SELECT' - no energy limits found for atom   2  L= 1  
 'SELECT' - E-bottom -200.0   E-top -200.0  


As the error comes from lapw1 so below is scf1 data;
cat *.scf1 gives
:LMAX-WF:   10   Non-Spherical LMAX:   4

  ATOMIC SPHERE DEPENDENT PARAMETERS FOR ATOM  Cs    
:e__0001: OVERALL ENERGY PARAMETER IS    0.3000
  OVERALL BASIS SET ON ATOM IS LAPW
:E2_0001: E( 2)=    0.3000
 APW+lo
:E2_0001: E( 2)=   -2.6539   E(BOTTOM)=   -3.027   E(TOP)=   -2.281  1  2   340
 LOCAL ORBITAL
:E0_0001: E( 0)=    0.7000
 APW+lo
:E0_0001: E( 0)=   -2.3380   E(BOTTOM)=   -3.764   E(TOP)=   -0.912  4  5   248
 LOCAL ORBITAL
:E1_0001: E( 1)=    0.7000
 APW+lo
:E1_0001: E( 1)=   -1.3099   E(BOTTOM)=   -3.057   E(TOP)=    0.437  3  4   276
 LOCAL ORBITAL

  ATOMIC SPHERE DEPENDENT PARAMETERS FOR ATOM  Sn    
:e__0002: OVERALL ENERGY PARAMETER IS    0.3000
  OVERALL BASIS SET ON ATOM IS LAPW
:E1_0002: E( 1)=    0.3000
 APW+lo


Below id mu case.in1

WFFIL  EF= 0.5   (WFFIL, WFPRI, ENFIL, SUPWF)
  7.00   10    4 (R-MT*K-MAX; MAX L IN WF, V-NMT
  0.30    6  0  (GLOBAL E-PARAMETER WITH n OTHER CHOICES, global APW/LAPW)
 2    0.30 0. CONT 1
 2   -5.07 0.0001 STOP 1
 0    0.30 0. CONT 1
 0   -1.66 0.0010 CONT 1
 1    0.30 0. CONT 1
 1   -0.66 0.0010 CONT 1
  0.30    5  0  (GLOBAL E-PARAMETER WITH n OTHER CHOICES, global APW/LAPW)
 1    0.30 0. CONT 1
 1   -6.09 0.0001 STOP 1
 2    0.30 0. CONT 1
 2   -1.55 0.0010 CONT 1
 0    0.30 0. CONT 1
  0.30    5  0  (GLOBAL E-PARAMETER WITH n OTHER CHOICES, global APW/LAPW)
 2    0.30 0. CONT 1
 2   -3.33 0.0001 STOP 1
 0    0.30 0. CONT 1
 0   -1.00 0.0010 CO

Re: [Wien] Charged cells using hybrid functionals

2018-01-28 Thread Marcelo Barbosa 
Dear Prof. Blaha,

Thank you very much for the fix, now it seems to be working fine.

Best regards,
Marcelo

> On 24 Jan 2018, at 17:21, Peter Blaha  wrote:
> 
> Yes, you are right, mBJ works with charged cells.
> 
> Here is the proposed fix for charged cells and hybrid-DFT:
> 
> In  lapw0.F   search for:
> 
> if(supn.ne.0.d0) write(6,*) ' CHARGED CELL with ',supn
> 
> and replace this line by:
> 
>  if(supn.ne.0.d0) then
> if(GRR_SWITCH .eqv. .true.) then
> supn=0.d0
> write(6,*) 'CHARGED CELL switched off for a -grr calculation'
> else
> write(6,*) ' CHARGED CELL with ',supn
> endif
>  endif
> 
> Regards
> 
> On 01/24/2018 09:41 AM, Marcelo Barbosa wrote:
>> That’s great news, thank you very much!
>> However, you also mentioned mBJ but isn’t it possible to use mBJ for charged 
>> cells at the moment?
>> I previously used it to improve the density of states and band structure 
>> after doing structural relaxation with PBE for charged cells and the 
>> simulation runs without errors and the results seem to be good.
>> Should I be more careful about them (or, even worse, not trust them)?
>> Best regards,
>> Marcelo
>>  
>>> On 24 Jan 2018, at 16:35, Peter Blaha  wrote:
>>> 
>>> We are working on this and probably I can send a fix very soon.
>>> 
>>> The decomposition of E-tot into kinetic, coulomb and xc Energy (switch KXC) 
>>> will still not be possible, but we can enable charged cells for hybrids or 
>>> mBJ calculations.
>>> 
>>> Regards
>>> 
>>> On 01/24/2018 04:27 AM, Marcelo Barbosa wrote:
 Dear Gavin,
 Thank you very much for your answer.
 Since hybrid functional calculations for charged cells is still not 
 possible, do you think that a plausible approach to compare the EFG of 
 different charged states using hybrid functionals could be done by 
 calculating the variation in each EFG orbital component from PBE to hybrid 
 and then consider that a similar variation in each component could be 
 expected for the charged state, therefore enabling an estimation of the 
 EFG for the charged state?
 Best regards,
 Marcelo
> On 12 Jan 2018, at 10:23, Gavin Abo  > wrote:
> 
> In SRC_lapw0/inputpars.F, it looks like SWITCH3 corresponds with putting 
> either KXC or screened exchange-correlation energy and no 
> exchange-correlation potential (e.g., EX_SPBE VX_SPBE VX_NONE VC_NONE) in 
> case.in0_grr.
> 
> The message "Charged cell and SWITCH3 not possible", and "Charged cells 
> AND E-tot decomposition not possible" message that you likely also have 
> in case.output0_grr, seems to be the same as the post at the link you 
> provided.  So yes, it seems that hybrid functional calculations for 
> charged cells is still not possible.
> 
> On 1/11/2018 12:08 AM, Marcelo Barbosa wrote:
>> Dear Sirs,
>> 
>> I’m trying to calculate a charged cell using hybrid functionals but the 
>> following error shows up:
>> 
>> "Charged cell and SWITCH3 not possible”
>> 
>> Looking at the mailing list archive, I only found this link from 2013 
>> about this subject 
>> (https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg09546.html)
>>  but I’m not sure if the problem is related since I don’t know what 
>> SWITCH3 corresponds to.
>> Is it still not possible to use hybrid functionals for charged cells?
>> 
>> Thank you for your help.
>> 
>> Best regards,
>> Marcelo
> ___
> Wien mailing list
> Wien@zeus.theochem.tuwien.ac.at 
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
 ___
 Wien mailing list
 Wien@zeus.theochem.tuwien.ac.at
 http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
 SEARCH the MAILING-LIST at:  
 http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>>> 
>>> -- 
>>> 
>>>  P.Blaha
>>> --
>>> Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
>>> Phone: +43-1-58801-165300 FAX: +43-1-58801-165982
>>> Email: bl...@theochem.tuwien.ac.atWIEN2k: http://www.wien2k.at
>>> WWW:   http://www.imc.tuwien.ac.at/TC_Blaha
>>> --
>>> ___
>>> Wien mailing list
>>> Wien@zeus.theochem.tuwien.ac.at
>>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>> SEARCH the MAILING-LIST at:  
>>> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>>

Re: [Wien] Wien2k workshop 2018 ?

2018-01-28 Thread Peter Blaha 

Dear wien2k users,

Since I already got several requests, let me announce the date and 
location of our wien2k workshop in 2018.


   25th WIEN2k-workshop
   Boston College, Boston, US, 12-16.June 2018

 (More info later in the mailing list and at www.wien2k.at)

It is  a great opportunity for beginners to get familiar with a
powerful but complex package like WIEN2k.

Best regards
Peter Blaha

Am 26.01.2018 um 18:31 schrieb Ilias Miroslav, doc. RNDr., PhD.:

Dear Professor Blaha,

please would you organize the Wien2k worshop also in this year 2018 ? If 
yes, will it be held in September (2018)  ?


Yours,

Miro Ilias



--
--
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300 FAX: +43-1-58801-165982
Email: bl...@theochem.tuwien.ac.atWIEN2k: http://www.wien2k.at
WWW: 
http://www.imc.tuwien.ac.at/tc_blaha- 


___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html