Re: [Wien] Doubts regarding the volume optimization of a triclinic cell

2021-10-24 Thread Anupriya Nyayban 
Dear Prof. Gavin and Prof. Marks,
Thank you for clearing it to me. I have used the wrong formula (
http://webmineral.com/help/CellDimensions.shtml#.YXTy5SbhUlI) for the
volume of a triclinic cell and this made me confused.

-- 
With regards
Anupriya Nyayban
Ph.D. Scholar
Department of Physics
NIT Silchar
___
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] Doubts regarding the volume optimization of a triclinic cell

2021-10-23 Thread Laurence Marks 
Also, the relevance of the results goes as exp(-time*1,000,000) so after 11
years the Professor has probably moved to another country.

I am 100% certain that you are worrying about details which do not matter:
a) DFT lattice parameters are rarely better than 0.5%.
b) DFT energies are rarely better than 4kJ/atom mole in accuracy.
c) Electronic states are rarely better than 0.1 eV, as DFT is ground state.

Accept that you calculations are only good approximations.

"All things appear and disappear because of the concurrence of causes and
conditions. Nothing ever exists entirely alone: everything is in
relationship to everything else."

-- Gautama Buddha
In Dwight Goddard, Buddha, Truth and Brotherhood (1934)

---
Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu
"Research is to see what everybody else has seen, and to think what
nobody else has thought" Albert Szent-Györgyi

On Sat, Oct 23, 2021, 6:09 AM Gavin Abo  wrote:

> I follow that your lattice parameters a=18.508 Bohr, b=39.835 Bohr,
> c=40.199 Bohr, α=119.701 degrees, β=103.309 degrees, and γ=90.0 degrees
> using the equation at [1] give you
>
> Volume_triclinic = 18.508 Bohr*39.835 Bohr*40.199 Bohr*sqrt(1-cos(119.701
> degrees *(π radians/180 degrees))^2-cos(103.309*(π radians/180
> degrees))^2-cos(90.0*(π radians/180 degrees))^2+2*cos(119.701 degrees *(π
> radians/180 degrees))*cos(103.309 degrees *(π radians/180
> degrees))*cos(90.0 degrees *(π radians/180 degrees))) ≈ 24823 Bohr^3.
>
> I also remember reading that the :VOL in case.scf is for the primitive
> unit cell [2-4].
>
> Regarding the 20805.429 Bohr^3, I don't seem to be seeing enough
> information regarding that.  Thus, I'm unable to follow where you are
> getting that from.
>
> I have seen the post at [5] describing how a triclinic optimization could
> be done.
>
> However, in addition to that post I have read other posts about how it
> requires too many calculations for triclinic such that it not likely
> reasonably possible to do [6-8].
>
> For example, say your advising professor says he needs a triclinic
> optimization one week from now.  However, say it takes 1 day to run and
> complete a single calculation on your initial triclinic structure.  Now, if
> you fixed b, c, α, β, γ and varied lattice constant a with four different
> values you would expect those calculations to take about 4 days.  However,
> to fully optimize the structure you would also need to vary b, c, α, β, and
> γ as well such that you would need to do 4096 calculations in total as
> explained in [5].
>
> Thus, you could estimate that it would take 4096 days or about 11 years
> (=4096/365.25).  So it not possible to meet the professors deadline, but
> the calculation is possible but would take 11 years.  However, another
> problem with that could be that you run out of hard drive space and cannot
> afford the cost to buy more hard drives to store the data from all those
> 4096 calculations.
>
> Furthermore, if you are like me, where I sometimes mess up my calculation
> setup during initialization (init_lapw).  I would probably check the result
> after 11 years to find out I setup the calculation wrong and have to start
> over.
>
> So for me it would probably end up similar to the scenerio in The
> Hitchhiker's Guide to the Galaxy where my GIGO [9] mistake during
> initialization would end with a result like 42 [10].
> [1] https://www.globalsino.com/EM/page3033.html
> 
> [2]
> https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg20184.html
> 
> [3]
> https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg14487.html
> 
> [4]
> https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg19755.html
> 
> [5]
> https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg10777.html
> 
> [6]
> https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg01914.html
>

Re: [Wien] Doubts regarding the volume optimization of a triclinic cell

2021-10-23 Thread Gavin Abo 
I follow that your lattice parameters a=18.508 Bohr, b=39.835 Bohr, 
c=40.199 Bohr, α=119.701 degrees, β=103.309 degrees, and γ=90.0 degrees 
using the equation at [1] give you


Volume_triclinic = 18.508 Bohr*39.835 Bohr*40.199 
Bohr*sqrt(1-cos(119.701 degrees *(π radians/180 
degrees))^2-cos(103.309*(π radians/180 degrees))^2-cos(90.0*(π 
radians/180 degrees))^2+2*cos(119.701 degrees *(π radians/180 
degrees))*cos(103.309 degrees *(π radians/180 degrees))*cos(90.0 degrees 
*(π radians/180 degrees))) ≈ 24823 Bohr^3.


I also remember reading that the :VOL in case.scf is for the primitive 
unit cell [2-4].


Regarding the 20805.429 Bohr^3, I don't seem to be seeing enough 
information regarding that.  Thus, I'm unable to follow where you are 
getting that from.


I have seen the post at [5] describing how a triclinic optimization 
could be done.


However, in addition to that post I have read other posts about how it 
requires too many calculations for triclinic such that it not likely 
reasonably possible to do [6-8].


For example, say your advising professor says he needs a triclinic 
optimization one week from now.  However, say it takes 1 day to run and 
complete a single calculation on your initial triclinic structure.  Now, 
if you fixed b, c, α, β, γ and varied lattice constant a with four 
different values you would expect those calculations to take about 4 
days.  However, to fully optimize the structure you would also need to 
vary b, c, α, β, and γ as well such that you would need to do 4096 
calculations in total as explained in [5].


Thus, you could estimate that it would take 4096 days or about 11 years 
(=4096/365.25).  So it not possible to meet the professors deadline, but 
the calculation is possible but would take 11 years.  However, another 
problem with that could be that you run out of hard drive space and 
cannot afford the cost to buy more hard drives to store the data from 
all those 4096 calculations.


Furthermore, if you are like me, where I sometimes mess up my 
calculation setup during initialization (init_lapw).  I would probably 
check the result after 11 years to find out I setup the calculation 
wrong and have to start over.


So for me it would probably end up similar to the scenerio in The 
Hitchhiker's Guide to the Galaxy where my GIGO [9] mistake during 
initialization would end with a result like 42 [10].


[1] https://www.globalsino.com/EM/page3033.html
[2] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg20184.html
[3] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg14487.html
[4] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg19755.html
[5] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg10777.html
[6] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg01914.html
[7] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg10505.html
[8] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg21450.html

[9] https://en.wikipedia.org/wiki/Garbage_in,_garbage_out
[10] 
https://en.wikipedia.org/wiki/Phrases_from_The_Hitchhiker%27s_Guide_to_the_Galaxy#The_Answer_to_the_Ultimate_Question_of_Life,_the_Universe,_and_Everything_is_42


On 10/23/2021 1:18 AM, Anupriya Nyayban wrote:

Dear Prof. Blaha and Dr. Bhamu,

Thank you Prof. Blaha for the valuable suggestion and I have followed 
all the steps as you have mentioned for the optimization. But still I 
have doubt when I am going to calculate the optimized lattice 
parameters from the optimized volume. The initial volume calculated 
using the lattice parameters before the volume optimization (values I 
have mentioned earlier) is 20805.429 Bohr^3. The scf file (before the 
volume optimization) shows the lattice parameters (LATT) and volume 
(infact this volume is the volume before inducing the mixing  into the 
2*2*2 supercell) to be "a=18.508, b=39.835, c=40.199 Bohr; α=119.701◦, 
β=103.309◦, γ=90.0◦" and "24823.466 Bohr^3", respectively. I am 
confused about the initial volume which one I should consider and why 
is this dissimiarity?



--
With regards
Anupriya Nyayban
Ph.D. Scholar
Department of Physics
NIT Silchar___
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] Doubts regarding the volume optimization of a triclinic cell

2021-10-23 Thread Anupriya Nyayban 
Dear Prof. Blaha and Dr. Bhamu,

Thank you Prof. Blaha for the valuable suggestion and I have followed all
the steps as you have mentioned for the optimization. But still I have
doubt when I am going to calculate the optimized lattice parameters from
the optimized volume. The initial volume calculated using the lattice
parameters before the volume optimization (values I have mentioned earlier)
is 20805.429 Bohr^3. The scf file (before the volume optimization) shows
the lattice parameters (LATT) and volume (infact this volume is the volume
before inducing the mixing  into the 2*2*2 supercell) to be "a=18.508,
b=39.835, c=40.199 Bohr; α=119.701◦, β=103.309◦, γ=90.0◦" and "24823.466
Bohr^3", respectively. I am confused about the initial volume which one I
should consider and why is this dissimiarity?


-- 
With regards
Anupriya Nyayban
Ph.D. Scholar
Department of Physics
NIT Silchar
___
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] Doubts regarding the volume optimization of a triclinic cell

2021-10-22 Thread Dr. K. C. Bhamu 
Dear Anupriya,
I would first do a full optimization (vc-relax) in other code (VASP/QE) for
such a large system and then using this fully optimized structure, I will
do an ion relaxation (run_lapw  -min) in Wien2k.

If you want to do everything using Wien2k, please follow Prof. Peter's
response.

Happy computing

Regards
Bhamu

On Sat, Oct 23, 2021, 3:27 AM Peter Blaha 
wrote:

> A full optimization of a bigger triclinic structure is hardly possible
> with WIEN2k.
>
> It depends a lot on what the purpose of your calculations is, but:
>
> a) Most importantly, optimize the position of the atoms using the
> forces:   run_lapw  -min ...
>
> b)  eventually I'd next do a volume optimization (x optimize, option 1)
> Do not forget the   -min switch in run_lapw, otherwise the results are
> nonsense. Volume changes give the largest change in E-tot.
> Note, that other separate options of x optimize are rather useless for
> this purpose. One cannot optimize alpha,beta,gamma for fixed a,b,c. This
> is nonsense.
>
> c) eventually you may use   optimize_abc_lapw -t 3 ...
> This optimizes a,b,c independently but simultaneously. Again don't
> forget the -min switch.
>
> For such a large cell, such calculations can be very time consuming when
> you do not use well adapted parameters (RKmax, k-points) and a good
> parallelization strategy on a sufficient number of cores.
>
> Am 22.10.2021 um 04:54 schrieb Anupriya Nyayban:
> > Dear users and experts,
> >
> > I have a triclinic structure (P1 space group) with lattice parameters
> > a=18.508, b=39.835, c=40.199 Bohr; α = 119.701 ◦ , β = 103.309 ◦ , γ =
> > 90.0 ◦.  I have the following doubts
> > 1) The volume obtained with these lattice parameters as well angles
> > (with the formula for triclinic cell) is not the same as the volume
> > (24823.466 Bohr^3) found in the scf file (VOL).
> > 2) The mailing list suggests volume optimization by a) varying a, b, c
> > while α, β, γ are constant; b) varying α, β, γ with fixed a, b, c. I
> > have done volume optimization by varying volume with constant a:b:c. But
> > I am not able to find the suitable switch for the case b) in the
> > optimize.job script.
> > Your valuable suggestion is highly appreciated!!
> >
> > Thank you!!
> > --
> > With regards
> > Anupriya Nyayban
> > Ph.D. Scholar
> > Department of Physics
> > NIT Silchar
> >
> > ___
> > 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
> >
>
> --
> --
> 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
> -
> ___
> 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] Doubts regarding the volume optimization of a triclinic cell

2021-10-22 Thread Peter Blaha 
A full optimization of a bigger triclinic structure is hardly possible 
with WIEN2k.


It depends a lot on what the purpose of your calculations is, but:

a) Most importantly, optimize the position of the atoms using the 
forces:   run_lapw  -min ...


b)  eventually I'd next do a volume optimization (x optimize, option 1)
Do not forget the   -min switch in run_lapw, otherwise the results are 
nonsense. Volume changes give the largest change in E-tot.
Note, that other separate options of x optimize are rather useless for 
this purpose. One cannot optimize alpha,beta,gamma for fixed a,b,c. This 
is nonsense.


c) eventually you may use   optimize_abc_lapw -t 3 ...
This optimizes a,b,c independently but simultaneously. Again don't 
forget the -min switch.


For such a large cell, such calculations can be very time consuming when 
you do not use well adapted parameters (RKmax, k-points) and a good 
parallelization strategy on a sufficient number of cores.


Am 22.10.2021 um 04:54 schrieb Anupriya Nyayban:

Dear users and experts,

I have a triclinic structure (P1 space group) with lattice parameters 
a=18.508, b=39.835, c=40.199 Bohr; α = 119.701 ◦ , β = 103.309 ◦ , γ = 
90.0 ◦.  I have the following doubts
1) The volume obtained with these lattice parameters as well angles 
(with the formula for triclinic cell) is not the same as the volume 
(24823.466 Bohr^3) found in the scf file (VOL).
2) The mailing list suggests volume optimization by a) varying a, b, c 
while α, β, γ are constant; b) varying α, β, γ with fixed a, b, c. I 
have done volume optimization by varying volume with constant a:b:c. But 
I am not able to find the suitable switch for the case b) in the 
optimize.job script.

Your valuable suggestion is highly appreciated!!

Thank you!!
--
With regards
Anupriya Nyayban
Ph.D. Scholar
Department of Physics
NIT Silchar

___
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



--
--
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
-
___
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] Doubts regarding the volume optimization of a triclinic cell

2021-10-22 Thread Anupriya Nyayban 
Dear users and experts,

I have a triclinic structure (P1 space group) with lattice parameters
a=18.508, b=39.835, c=40.199 Bohr; α = 119.701 ◦ , β = 103.309 ◦ , γ = 90.0
◦.  I have the following doubts
1) The volume obtained with these lattice parameters as well angles (with
the formula for triclinic cell) is not the same as the volume (24823.466
Bohr^3) found in the scf file (VOL).
2) The mailing list suggests volume optimization by a) varying a, b, c
while α, β, γ are constant; b) varying α, β, γ with fixed a, b, c. I have
done volume optimization by varying volume with constant a:b:c. But I am
not able to find the suitable switch for the case b) in the optimize.job
script.
Your valuable suggestion is highly appreciated!!

Thank you!!
-- 
With regards
Anupriya Nyayban
Ph.D. Scholar
Department of Physics
NIT Silchar
___
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