[Wien] a note on lattice parameter optimization
Fecher, Gerhard
fecher at uni-mainz.de
Thu Feb 8 15:57:39 CET 2018
As someone who is doing both, experiment and calculations, I do not agree for the following reasons
a) For example, in the Pearson database you will find many compounds where (in case of "free" parameters) the atom positions of the prototype are given
but not the "real" ones of the actual compound. You have always to check the original literature whether or not the "complete" structure was determined.
One must indeed check if the conclusions from published data make sense, I know enough that don't.
(I prefer own data instead of checking whether the XRD data of someone else make sense)
Note if you start with completely wrong parameters for a full optimization, then you might run into a wrong minimum that is not the ground state.
b) The most measurements are performed at "room temperature" (something or everything between 288 and 311 K) and the low
temperature lattice parameters or even the "real" structure might considerably differ from that.
c) To compare the size of the band gaps, one needs to know how they were measured. I guess a band gap determined
from high temperature thermoelectric data (as is usual in that community) has nothing to do with that found in calculations.
(Indeed, the crystal structure for such high temperature data are nearly never given)
If there is any method that gives the "correct" gap for Si, for example, this does not mean that the band gap is correct for every compound one may think about,
and this concerns both, experiment and calculation.
Ciao
Gerhard
DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy:
"I think the problem, to be quite honest with you,
is that you have never actually known what the question is."
====================================
Dr. Gerhard H. Fecher
Institut of Inorganic and Analytical Chemistry
Johannes Gutenberg - University
55099 Mainz
and
Max Planck Institute for Chemical Physics of Solids
01187 Dresden
________________________________________
Von: Wien [wien-bounces at zeus.theochem.tuwien.ac.at] im Auftrag von Peter Blaha [pblaha at theochem.tuwien.ac.at]
Gesendet: Donnerstag, 8. Februar 2018 15:30
An: wien at zeus.theochem.tuwien.ac.at
Betreff: Re: [Wien] a note on lattice parameter optimization
There are 3 options:
a) Use the experimental structure parameters and positions.
However, in MANY cases, where only powder samples exist, or light
elements (maybe even H) are next to heavy ones, the experimental
positions are VERY insecure (+ disorder, vacancies, ...).
b) in such cases, use the experimental lattice parameters (usually they
are much more accurate than any theory), but optimize the positions
theoretically.
c) Try to be "ab initio": optimize lattice parameters AND positions
simultaneously (and not only lattice parameters). This is the "ideal
theory approach", but be aware that due to our DFT errors, the results
may not coincide with experiment.
> The closest fit to reality is the experimental parameters/positions.
>
> _____
> Professor Laurence Marks
> "Research is to see what everybody else has seen, and to think what
> nobody else has thought", Albert Szent-Gyorgi
> www.numis.northwestern.edu <http://www.numis.northwestern.edu>
>
> On Feb 8, 2018 7:43 AM, "Dr. K. C. Bhamu" <kcbhamu85 at gmail.com
> <mailto:kcbhamu85 at gmail.com>> wrote:
>
>
> Dear Prof. Marks,
>
> Thank you very much for the conceptual reply.
>
> I am unable to answer the second point now but please see for 1st
> and 3rd point:
>
> I agree that some time hybrid calculations may fail for some
> systems. May you also want to point out that with element from high
> Z a -SO correction is needed. But in some reputed papers (Nano.
> letters) it is mention that -SO does not affect much for my system.
>
>
> Regarding, PBESol: in two of PRB papers of Prof. Peter, it is
> mentioned that PBESol and WC-GGA are better for optimization
> "PhysRevB.79.155107.pdf", PHYSICAL REVIEW B 79, 085104 2009 . The
> same trend I am getting where LDA is underestimating the lattice
> parameter optimization and PBE is overestimating for a list of
> materials that I have testes.
>
> WC-GGA and PBESol are giving the optimized lattice parameters in the
> vicinity of Exp lattice parameters (~0.01-~0.02 Ang reduced). So I
> used PBESol with mBJ.
>
> Yes, I also do agree that according to UG, LDA or PBE are good for
> mBJ but as PBESol gave me well-optimized results so I used PBESol
> with mBJ.
>
>
> So I am not able to find out the solution to my issue; Why the
> bandgap is drastically reducing if I use "double relaxed structure":
>
> 1. relaxed>> 2. optimization of lattice parameter without
> relaxing>>> 3. again relaxing the structure (using relaxed structure
> from step 1 and optimized lattice parameter from step 2.).
>
> I am sure the procedure used above seems wrong but I am not able to
> justify the "double relaxation of the structure.
>
>
> regards,
>
> Bhamu
>
>
>
>
>
>
>
> On Thu, Feb 8, 2018 at 6:48 PM, Laurence Marks
> <L-marks at northwestern.edu <mailto:L-marks at northwestern.edu>> wrote:
>
> Inlined, please think:
>
> On Feb 8, 2018 7:08 AM, "Dr. K. C. Bhamu" <kcbhamu85 at gmail.com
> <mailto:kcbhamu85 at gmail.com>> wrote:
>
> Dear Prof. Marks,
>
> I believe I am having to understand about the mentioned points:
>
>
> 1) Why is your number for the band gap "wrong", whereas what
> someone else calculated with HSE is "right"?
>
> Ans: Maybe you want to ask me why the band gap from my
> calculations should be fine instead of HSE. If so, then I
> shall mention that hybrid calculations are already a well
> know approach to predict the band gap of inorganic systems.
>
>
> Does that mean HSE is right?
>
>
> 2) What is the purpose of optimizing internal and cell
> parameters?
>
> Ans: To get the minimum forces on the system and by doing
> total optimization the system would be in the most stable
> ground state and can give accurate properties if the most
> suitable approach is applied!!
>
> Why?
>
>
> 3) Are the cell and internal parameters of PBEsol the same
> as what they should be with mBJ? What does "should be" mean
> here?
>
> Ans: Yes, structure file is same for PBESol and mBJ.
> According to me you wanted to ask me whether I have applied
> any changes in mBJ calculation, like relaxation (which is
> not possible and I did not apply).
> I just did PBESol scf >> save_lapw case_scf>>> one scf
> cycle>> save_lapw mbjx_grr >>> spacified the mBJ approach
> and finally mBJ scf.
>
> Why should a PBEsol refined structure be optimal for mBJ?
>
>
> regards
> Bhamu
>
>
> On Feb 8, 2018 3:41 AM, "Dr. K. C. Bhamu"
> <kcbhamu85 at gmail.com <mailto:kcbhamu85 at gmail.com>> wrote:
>
> No, I said the calculated gal is ~0.3eV less then the
> reported gap: HSE.
>
>
> I tested this on more than 10 systems and in each case
> obsessed this band gap discrepancy.
>
> Regards
> Bhamu
>
> On Feb 8, 2018 3:05 PM, <tran at theochem.tuwien.ac.at
> <mailto:tran at theochem.tuwien.ac.at>> wrote:
>
> A band gap of 0.3 eV is very small and a change of
> lattice constant may
> be enough to close the band gap.
>
> On Thursday 2018-02-08 10:30, Dr. K. C. Bhamu wrote:
>
> Date: Thu, 8 Feb 2018 10:30:29
> From: Dr. K. C. Bhamu <kcbhamu85 at gmail.com
> <mailto:kcbhamu85 at gmail.com>>
> Reply-To: A Mailing list for WIEN2k users
> <wien at zeus.theochem.tuwien.ac.at
> <mailto:wien at zeus.theochem.tuwien.ac.at>>
> To: A Mailing list for WIEN2k users
> <Wien at zeus.theochem.tuwien.ac.at
> <mailto:Wien at zeus.theochem.tuwien.ac.at>>
> Subject: Re: [Wien] a note on lattice parameter
> optimization
>
> Yes, I used PBESol followed by mBJ.
>
> On Feb 8, 2018 2:28 PM,
> <tran at theochem.tuwien.ac.at
> <mailto:tran at theochem.tuwien.ac.at>> wrote:
> Hi,
>
> Did you use mBJ to calculate the band gap
> in both procedures?
>
> FT
>
> On Thursday 2018-02-08 09:35, Dr. K. C.
> Bhamu wrote:
>
> Date: Thu, 8 Feb 2018 09:35:08
> From: Dr. K. C. Bhamu
> <kcbhamu85 at gmail.com <mailto:kcbhamu85 at gmail.com>>
> Reply-To: A Mailing list for WIEN2k
> users <wien at zeus.theochem.tuwien.ac.at
> <mailto:wien at zeus.theochem.tuwien.ac.at>>
> To: A Mailing list for WIEN2k users
> <Wien at zeus.theochem.tuwien.ac.at
> <mailto:Wien at zeus.theochem.tuwien.ac.at>>
> Subject: [Wien] a note on lattice
> parameter optimization
>
> Hello,
>
> Using experimental lattice
> parameters(a Ang), I optimized the structure
> with two options.
>
> 1. Directly optimized the structure
> without min_lapw ... and the optimized lattice
> parameters are in the vicinity of Exp a (~0.01
> Ang accuracy) and the calculated band gap (from
> mBJ2) is in agreement with reported value ~0.3
> eV less.
>
> 2. First relaxed the structure and
> then optimized this relaxed structure (without
> min_lapw ....). In this case the optimized
> lattice parameters are reduced by ~0.05 Ang and
> the band gap vanished or largely reduced.
>
>
> Why is the second way of doing band
> structure not working?
>
> I tested it on many systems. In some
> case system became metal and in some cases band
> gap largely reduced when I used the fully
> relaxed structure for optimization.
>
>
> Thanks,
>
>
> Bhamu
>
>
>
>
>
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> <mailto:Wien at zeus.theochem.tuwien.ac.at>
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__zeus.theochem.tuwien.ac.at_mailman_listinfo_wien&d=DwMFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0&m=jl0lHi-CtjMpp__lsTB6ZTdAJ0Xy47bf44CPXSBt7QE&s=xezPeLzVIzkyNQZdVyrkfiQm7d4wtaXeocIv-y-KdUY&e=>
> SEARCH the MAILING-LIST at:
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.mail-2Darchive.com_wien-40zeus.theochem.tuwien.ac.at_index.html&d=DwMFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0&m=jl0lHi-CtjMpp__lsTB6ZTdAJ0Xy47bf44CPXSBt7QE&s=1EhCR6okBj_tGDzqdgSAxNSZftNyujAJxvs8z3T7sq4&e=>
>
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> <mailto:Wien at zeus.theochem.tuwien.ac.at>
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__zeus.theochem.tuwien.ac.at_mailman_listinfo_wien&d=DwMFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0&m=jl0lHi-CtjMpp__lsTB6ZTdAJ0Xy47bf44CPXSBt7QE&s=xezPeLzVIzkyNQZdVyrkfiQm7d4wtaXeocIv-y-KdUY&e=>
> SEARCH the MAILING-LIST at:
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.mail-2Darchive.com_wien-40zeus.theochem.tuwien.ac.at_index.html&d=DwMFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0&m=jl0lHi-CtjMpp__lsTB6ZTdAJ0Xy47bf44CPXSBt7QE&s=1EhCR6okBj_tGDzqdgSAxNSZftNyujAJxvs8z3T7sq4&e=>
>
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> <mailto:Wien at zeus.theochem.tuwien.ac.at>
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__zeus.theochem.tuwien.ac.at_mailman_listinfo_wien&d=DwMFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0&m=aQoRvD9xu6e3OubdKBDmPnflUbmo41M7nwDfegZLBrg&s=1Y5DBgtCcLyu85SWwEL5PbwOT8-nRgYp67SVCwRNdgk&e=>
> SEARCH the MAILING-LIST at:
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.mail-2Darchive.com_wien-40zeus.theochem.tuwien.ac.at_index.html&d=DwMFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0&m=aQoRvD9xu6e3OubdKBDmPnflUbmo41M7nwDfegZLBrg&s=CQF8_t9GbaY-SgmuduDNbdIs1KsKQPhsf3uKTaKSE_Y&e=>
>
>
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> <mailto:Wien at zeus.theochem.tuwien.ac.at>
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__zeus.theochem.tuwien.ac.at_mailman_listinfo_wien&d=DwMFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0&m=z2imeuRA6sJvWdqAkT6cOQmIr4m1WaEPLeB9V35U0-s&s=57H6rQOa1RsoKvmFTtb-NwiERfFsGhwLmysPYkbq9w4&e=>
> SEARCH the MAILING-LIST at:
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.mail-2Darchive.com_wien-40zeus.theochem.tuwien.ac.at_index.html&d=DwMFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0&m=z2imeuRA6sJvWdqAkT6cOQmIr4m1WaEPLeB9V35U0-s&s=w_-ZCOm-zUhiAwUC0G7770aAu5wVXRuY4YyZZys13q4&e=>
>
>
>
>
>
> _______________________________________________
> Wien mailing list
> Wien at 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: blaha at theochem.tuwien.ac.at WIEN2k: http://www.wien2k.at
WWW: http://www.imc.tuwien.ac.at/TC_Blaha
--------------------------------------------------------------------------
_______________________________________________
Wien mailing list
Wien at 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
More information about the Wien
mailing list