[CP2K-user] [CP2K:19568] Re: Significant discontinuity in EOS curver for FeO

Kejiang Li lyam_likej at 126.com
Sat Nov 25 01:37:22 UTC 2023

Previous message (by thread): [CP2K-user] [CP2K:19566] Re: Significant discontinuity in EOS curver for FeO
Next message (by thread): [CP2K-user] [CP2K:19532] Problem with GW core electron calculations
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Dear Matthias,

Thanks a lot for your suggestion.

I have done the tests with different R_CUTOFF radius for the VDW_POTENTIAL
from 10.05 to 25.05 A. You can see that the increase of R_CUTOFF did not
improve the EOS curve for FeO. And it is indeed the VDW_POTENTIAL that
causes the discontinuity or asymmetry of the EOS curve of FeO, while all
the other parameters I tested did not have an obvious influence on the EOS
curve of FeO.

Maybe we can conclude that the empirical VDW_POTENTIAL is not applicable to
FeO, even though I have not yet understand the fundamental mechanism behind
this issue as VDW_POTENTIAL is a long-range interaction and should not have
an obvious influence on the short-range interaction. Anyway, it tells us
that we should be careful while using VDW_POTENTIAL for iron-containing
solid phase or any other system.

Many thanks for your continuous suggestions.

Best regards,
Kejiang
[image: IMG_0115.png]

On Saturday, November 25, 2023 at 3:14:58 AM UTC+8 Krack Matthias wrote:

Dear Kejiang

the default R_CUTOFF radius for the VDW_POTENTIAL is only 20 bohr (10.5 A)
which can lead to a non-negligible noise in that energy contribution for a
growing cell size, because the growth of the pair (neighbor) lists is not
smooth (continuous). You can try a much larger R_CUTOFF radius, e.g. 25 A,
for a smoother behavior.

I don’t think that a vdW correction is crucially needed for bulk FeO, but
for FeO surfaces with H2/H2O adsorbed it makes sense most likely.

Best

Matthias

*From: *cp... at googlegroups.com <cp... at googlegroups.com> on behalf of
Kejiang Li <lyam_... at 126.com>

*Date: *Thursday, 23 November 2023 at 23:16
*To: *cp2k <cp... at googlegroups.com>

*Subject: *Re: [CP2K:19561] Re: Significant discontinuity in EOS curver for
FeO

Dear Matthias,

I really appreciate your patience in responding to my question point by
point.

I researched the literature and saw no published EOS results for FeO using
cp2k.

I finally found that it was the VDW_POTENTIAL that led to the significant
discontinuity in my EOS curve. After eliminating the VDW_POTENTIAL
potential in all my test codes, CP2K provides a continuous EOS curve that
is quite normal. Please see the attached figure.

I never expected that VDW_POTENTIAL might have such an apparent lousy
effect on the FeO structure. What do you think might be the possible
fundamental mechanism for this issue? As I want to simulate the reaction
between FeO surface and H2/H2O in the future, I think it might be necessary
to include VDW_POTENTIAL.

Could you please provide your opinion regarding whether
including VDW_POTENTIAL in my future calculations is necessary?

Thanks a lot.

Best,
Kejiang

On Thursday, November 23, 2023 at 11:41:34 PM UTC+8 Krack Matthias wrote:

Dear Kejiang

· 1. You didn't set the VDW_POTENTIAL. Don't we need to
consider VDW_POTENTIAL for FeO?

Not sure if that is needed for bulk FeO. Did you check the work of other
authors? Did they apply such a vdW correction for bulk FeO?

· 2. You didn't set the initial value for MAGNETIZATION. Is this also
not necessary for FeO?

The &BS section does the same, i.e. sets the initial on-site electronic
configuration for the iron atoms, and it allows for a more detailed control.

· 3. You didn't set the UKS and MULTIPLICITY. Is the default setting
for UKS and MULTIPLICITY enough for FeO?

Yes, it is for AFM FeO as I already explained.

Matthias

On Thursday, November 23, 2023 at 5:39:55 PM UTC+8 Krack Matthias wrote:

Dear Kejiang

I have added a howto page <https://www.cp2k.org/howto:dft_u> to the CP2K
wiki for DFT+U calculations with an EOS calculation for cubic FeO as an
example.

HTH

Matthias

*From: *cp... at googlegroups.com <cp... at googlegroups.com> on behalf of
Kejiang Li <lyam_... at 126.com>
*Date: *Wednesday, 22 November 2023 at 07:19
*To: *cp2k <cp... at googlegroups.com>
*Subject: *[CP2K:19547] Re: Significant discontinuity in EOS curver for FeO

Dear all,

Here, I also attached the test results with different sizes of supercells.
There is still a sharp decrease or discontinuity on the left side of the
lowest point.

Could anyone provide any guidance?

Thanks a lot.

Best regards,

Kejiang

On Monday, November 20, 2023 at 9:00:11 AM UTC+8 Kejiang Li wrote:

Dear CP2K community,

I am calculating the EOS for FeO with a 2*2*2 supercell. But there is
always a significant discontinuity (sharp decrease) in the range of 4.3-4.4
angstrom lattice constant in the EOS curves, as shown in the figures below.

I have done the following tests:
- the convergence of the energy with different cutoffs;
- using different functionals, including PBE, HSE, and SCAN;
-including HSE with different HF fractions;
-using different magnetization and +U settings.
However, all the tests produce similar results with an apparent
discontinuity (4.3-4.3 A) .

This discontinuity is not observed in the Quantum Espresso results
conducted by myself and in the results of literature that generally used
VASP. I believe that the CP2K program might cause this problem.

Can you give me more suggestions on how to do further tests?

Here, I also attached one sample code I used during my tests.

Thanks a lot.

Best regards,
Kejiang
The University of Science and Technology Beijing

*Error! Filename not specified.*

*Error! Filename not specified.Error! Filename not specified.*

.
To view this discussion on the web visit
https://groups.google.com/d/msgid/cp2k/92048ac5-9f6d-43c3-9147-d21660d5ea6an%40googlegroups.com
<https://groups.google.com/d/msgid/cp2k/92048ac5-9f6d-43c3-9147-d21660d5ea6an%40googlegroups.com?utm_medium=email&utm_source=footer>
.

To view this discussion on the web visit
https://groups.google.com/d/msgid/cp2k/cbfe4a0e-b678-4ad7-809e-bd11dec37e4en%40googlegroups.com
<https://groups.google.com/d/msgid/cp2k/cbfe4a0e-b678-4ad7-809e-bd11dec37e4en%40googlegroups.com?utm_medium=email&utm_source=footer>
.

To view this discussion on the web visit
https://groups.google.com/d/msgid/cp2k/9696539d-2ebb-484c-8a70-cabeedfa3e31n%40googlegroups.com
<https://groups.google.com/d/msgid/cp2k/9696539d-2ebb-484c-8a70-cabeedfa3e31n%40googlegroups.com?utm_medium=email&utm_source=footer>
.

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/268b7b6f-aea6-4031-8dcd-9704669c7517n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20231124/1d053703/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: IMG_0115.png
Type: image/png
Size: 127913 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20231124/1d053703/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vdw2.png
Type: image/png
Size: 174567 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20231124/1d053703/attachment-0003.png>

Previous message (by thread): [CP2K-user] [CP2K:19566] Re: Significant discontinuity in EOS curver for FeO
Next message (by thread): [CP2K-user] [CP2K:19532] Problem with GW core electron calculations
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

More information about the CP2K-user mailing list