[DFTB-Plus-User] Optimizing polyyne chain in DFTB+ for low BLA range: Need guidance

Wed Jun 14 18:41:08 CEST 2023

Hello Khadidja,

In the case of polyenes (i.e. the alternating double-single kind), LDA/GGA and DFTB substantially underestimate BLA, but the hybrids or +U approach increase this into the experimental ranges.

For polyyenes (tripple-single), I haven't tested them. But a test for a pure C chain in a periodic geometry with DFTB2 mio, and starting from a symmetry broken geometry, gives bond ratios of 0.48:0.52. This is at low temperatures and has a narrow band gap. Starting from the symmetric BA, I get a slightly higher energy structure (10 meV / C atom) with a BLA of 0.

Which geometry optimizer, electronic temperature and parameter sets are you using, and numerically how does the BLA compare to the reference values?

Regards

Ben

________________________________
From: DFTB-Plus-User <dftb-plus-user-bounces at mailman.zfn.uni-bremen.de> on behalf of Khadidja Chaib Edderraa <k.chaibedderraa at univ-chlef.dz>
Sent: 13 June 2023 17:46
To: User list for DFTB+ related questions <dftb-plus-user at mailman.zfn.uni-bremen.de>
Subject: Re: [DFTB-Plus-User] Optimizing polyyne chain in DFTB+ for low BLA range: Need guidance

Hello,
 Regarding the use of corrections (dftb+U and hybrid functional ) to dftb+ in order to obtain results closer to the experimental or previous dft calculations, I used the two corrections that you provided me as a solution, but the obtained results always show an increase in BLA, which is already considered greater than what it should be compared to the results obtained using dft. I want to know if dftb+ gives large values to the BLA ?, or are there other controls that should be added to control giving smaller values to the BLA (0.01 Å- 0.04 Å) ?
Can I use external slater koster files that are compatible with the polyyne chains I'm studying? I hope you can help me with this because I could not proceed with this research especially when I tried to compare my results with previous dft results.
Thank you for your time.

Regard,

On Mon, Jun 5, 2023, 14:27 Benjamin Hourahine <benjamin.hourahine at strath.ac.uk<mailto:benjamin.hourahine at strath.ac.uk>> wrote:
Hello Khadidja,

There are some examples in the test set, as well as a description in the manual.

The current release (22.2) supports hybrid functionals for molecular systems, and requires the ob2 parameter set. You will need to add to the hamiltonian block:

Hamiltonian = DFTB {
  SCC = Yes
  RangeSeparated = LC {
    Screening = MatrixBased {}
  }
.
.
.
}

The alternative of the +U methods do not require a specific parameter set and is available for periodic and molecular systems. The value of U is somewhat empirical, there is an argument that it should be a fraction of the atomic Hubbard U values depending on which functional is being used (https://doi.org/10.1021/JP070173B)

Hamiltonian = DFTB {
  SCC = Yes
OrbitalPotential = {
   Functional = {FLL}
   C = {
     Shells = {1 2} # sp block on the C
     UJ = 0.18235
   }
   #Functional = {pSIC}
   #C = {
   #  Shells = {1 2} # sp block on the C
   #  UJ = 0.1
   #}
  }
.
.
.
}

For polyacetylene, the above values gave reasonable BLA.

Regards

Ben

Dr. B. Hourahine
Senior Lecturer and Director of Student Support
Department of Physics, SUPA
University of Strathclyde, John Anderson Building,
107 Rottenrow, Glasgow G4 0NG, UK.
+44 141 548 2325, benjamin.hourahine at strath.ac.uk<mailto:benjamin.hourahine at strath.ac.uk>

[University of Strathclyde]

THE QUEEN’S ANNIVERSARY PRIZES
2019 & 2021
For Higher and Further Education

UNIVERSITY OF THE YEAR
2012 & 2019
Times Higher Education

SCOTTISH UNIVERSITY OF THE YEAR
2020
The Times & The Sunday Times

The University of Strathclyde is a charitable body, registered in Scotland, number SC015263.
Please consider the environment before printing this e-mail.

________________________________
From: DFTB-Plus-User <dftb-plus-user-bounces at mailman.zfn.uni-bremen.de<mailto:dftb-plus-user-bounces at mailman.zfn.uni-bremen.de>> on behalf of Hafid KHALFOUN <h.khalfoun at univ-chlef.dz<mailto:h.khalfoun at univ-chlef.dz>>
Sent: 02 June 2023 23:27
To: User list for DFTB+ related questions <dftb-plus-user at mailman.zfn.uni-bremen.de<mailto:dftb-plus-user at mailman.zfn.uni-bremen.de>>
Subject: Re: [DFTB-Plus-User] Optimizing polyyne chain in DFTB+ for low BLA range: Need guidance

Dear Benjamin, thank you for your response .
How do we include these corrections ?  On the skf files ? on the dftb_in.hsd file by adding a new block ?
How do we build this hybrid functional ?
Can you be more explicit , please ? Can I find some issues on this topic ?
thanks

Le ven. 2 juin 2023 à 16:01, Benjamin Hourahine <benjamin.hourahine at strath.ac.uk<mailto:benjamin.hourahine at strath.ac.uk>> a écrit :
Hello Khadidja,

BLA is fairly sensitive to functional in DFT, with mean-field functionals (LDA/GGA) often getting this incorrect. DFTB2 or 3 has the same issues, so as with DFT, either hybrid functionals or +U corrections may be needed.

The BLA for polyacetylene is definitely closer to the experimental values with these corrections. Hybrids should be used with ob2, while +U can be applied on top of the mio set.

Regards

Ben

Dr. B. Hourahin
Senior Lecturer and Director of Student Support
Department of Physics, SUPA
University of Strathclyde, John Anderson Building,
107 Rottenrow, Glasgow G4 0NG, UK.
+44 141 548 2325, benjamin.hourahine at strath.ac.uk<mailto:benjamin.hourahine at strath.ac.uk>

[University of Strathclyde]

THE QUEEN’S ANNIVERSARY PRIZES
2019 & 2021
For Higher and Further Education

UNIVERSITY OF THE YEAR
2012 & 2019
Times Higher Education

SCOTTISH UNIVERSITY OF THE YEAR
2020
The Times & The Sunday Times

The University of Strathclyde is a charitable body, registered in Scotland, number SC015263.
Please consider the environment before printing this e-mail.

________________________________
From: DFTB-Plus-User <dftb-plus-user-bounces at mailman.zfn.uni-bremen.de<mailto:dftb-plus-user-bounces at mailman.zfn.uni-bremen.de>> on behalf of Khadidja Chaib Edderraa <k.chaibedderraa at univ-chlef.dz<mailto:k.chaibedderraa at univ-chlef.dz>>
Sent: 02 June 2023 13:29
To: dftb-plus-user at mailman.zfn.uni-bremen.de<mailto:dftb-plus-user at mailman.zfn.uni-bremen.de> <dftb-plus-user at mailman.zfn.uni-bremen.de<mailto:dftb-plus-user at mailman.zfn.uni-bremen.de>>
Subject: Re: [DFTB-Plus-User] Optimizing polyyne chain in DFTB+ for low BLA range: Need guidance

CAUTION: This email originated outside the University. Check before clicking links or attachments.
Hello,
I am writing to you to provide further details regarding the structure I used before optimization for the chain. The structural parameters I employed were d1=1.51 Å and d2=1.22 Å. However, after performing the relaxation, I obtained a BLA value of approximately 0.11 Å, which is significantly larger than the range predicted by DFT (0.01-0.05 Å).

On Fri, Jun 2, 2023, 12:10 Khadidja Chaib Edderraa <k.chaibedderraa at univ-chlef.dz<mailto:k.chaibedderraa at univ-chlef.dz>> wrote:
Dear ,

I hope this email finds you well. I am currently working on optimizing a polyyne chain using DFTB+ and aiming to achieve a low range of bond length alternation (BLA) in my system. After exploring the available resources, I believe your expertise in DFTB+ could provide valuable insights and guidance for my research.

To provide some context, my research focuses on  ensuring that the molecular structure represents the most stable configuration and optimizing the polyyne chain is crucial for achieving BLA values  that align closely with experimental and other simulation data.

The specific details of the polyyne chain I'm studying are as follows:
- Number of carbon atoms: 2 carbon
- Parameterisation data (Slater-Koster files)  used to perform calculations : 3ob, mio, ob2, matsi
- Geometry optimization:  using  conjugate gradient algorithm but have not obtained the desired results.

My main questions are as follows:
1. What are the recommended optimization settings or parameters for a polyyne chain in DFTB+ to achieve a low BLA range?
2. Are there any specific convergence criteria that I should consider during the optimization process?
3. Are there any DFTB+ features or functionalities that could be particularly useful for optimizing polyyne chains and reducing the BLA range?

Thank you very much, and I look forward to your response.

Sincerely,
Khadidja
k.chaibedderraa at univ-chlef.dz<mailto:k.chaibedderraa at univ-chlef.dz>
_______________________________________________
DFTB-Plus-User mailing list
DFTB-Plus-User at mailman.zfn.uni-bremen.de<mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de>
https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user
_______________________________________________
DFTB-Plus-User mailing list
DFTB-Plus-User at mailman.zfn.uni-bremen.de<mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de>
https://mailman.zfn.uni-bremen.de/cgi-bin/mailman/listinfo/dftb-plus-user
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20230614/8e6e2886/attachment.htm>