LiMnO2 isolated cluster

[Victor Volkov]

Dear users

I am exploring various Li/Mn/O2 isolated (non-periodic) clusters.

Here is a usual-regular fail for the high spin LiMnO2 case

according the material project LiMnO2_Pmmn_mp-754881.

This is a case of high spin NEL setting.

I report this that you would not need to try.

I am preparing testing antiferromagnetic NEL cases

and magnetization representations as well.

I believe, understanding the ocean of CP2K collapses

to spot the islands of stable solutions

for Li/Mn/O2 (and analogous composites) is important

for humanity to progress.

Do not be too strict on my drive: I live for year on my savings without any support

in the country of no humility.

I hope to stimulate students and researchers to post such questions

to be berserk and help CP2K excellence.

Victor  

 

[Krack Matthias]

Dear Victor

 

Converging such a small isolated cluster with a lot of unsaturated oxygen atoms (i.e. dangling bonds) at the surface will be very difficult. And even if you can converge it, the structure will most likely undergo strong deformations during a structure relaxation.

 

Best

 

Matthias

 

--
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+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/32737a60-546d-447d-b19e-d6f0898053e9n%40googlegroups.com.

[Victor Volkov]

Dear Matthias,

thank you for the kind reply

an for the explanations.

I was thinking about the dangling bonds and about the nature of Manganese ions.

Here is my current adaptation for LiMn3O4 material project mp-34650:

where

1) there are no dangling bonds

2) all Mn2+ and Mn3+ are patterned to  sum-up a singlet.

3) DFT+U is introduced.

In respect to smallness and deformations.

Well, actually, in bio-inorganic systems, often (if not almost always), experiments concern

properties of finite systems.

A polypeptide associating with a surface of a certain facet "never happens",

though the PBC formalism is mesmerizingly inviting, indeed.

This is why, I believe addressing structure and electronics of such small clusters,

as I try to manipulate with, are more likely to attract attention in the near future

as "realistic" imaginative representations to model electromagnetism upon

organic binding.

Also, I think, computing optical properties when stepping away from the systems under PBC

might be more feasible and realistic.

 

Thank you.

With best regards,

Victor

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/ZRAP278MB082744AB9C952BB0FE80047BF42DA%40ZRAP278MB0827.CHEP278.PROD.OUTLOOK.COM.

[Krack Matthias]

Dear Victor

 

I doubt that you will get rid of all dangling bonds, because the actual oxidation and/or spin states of the atoms come from your “wish list” in order meet your (chemical) expectations based on the imposed stoichiometry. Anyway, you should adjust a few things in your input just to have any hope of success,  i.e. convergence:

• Use “analytic” or “MT” as PSOLVER
• Use “LSD”
• Try a larger smearing temperature, e.g. 1000-2000 K, that value is not related to the ambient temperature
• Use a smaller mixing ALPHA, e.g. 0.1 instead of 0.4.

 

HTH

 

Matthias

 

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CAK%2B-SSiQB_F%2BVKQS_oupp91ir3bLLQYDO3G079rsp%3DUSPhPmwg%40mail.gmail.com.

[Victor Volkov]

Dear Matthias,

thank you for the advises.

I appreciate.

> I doubt that you will get rid of all dangling bonds,

Indeed, this a nice challenge together with:

a) searching material projects to try shaping relevant structures;

b) managing right stoichiometry;

c) managing proper bonding network according to valencies.

The Li4Mn12O16_adoptation.xyz

is one of current trials.

  

With best regards,

Victor

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/ZRAP278MB0827C9A4CD229100185476E0F42DA%40ZRAP278MB0827.CHEP278.PROD.OUTLOOK.COM.

[Victor Volkov]

Dear Matthias:

good evening.

The example I try to deal with makes it apparent

that the CP2K fundamental trump-card (at present) is the OT protocol for SCF.

May be, if any possible improvement - this would concern "only" the SCF.

(This reminds me the scf=xqc magic in "the alternative reality".)

So, giving a job for SCF developer might be the key for the future.

:-)

Please, forgive me: I wish the CP2K team would forgive me.

You are great !

However, I attach comparative sets of files for the LiMn3O4 case.

The performance is strikingly different.

Or, a proper match or "two miss-conceptions" happened to compensate each other

to yield a "suitability".

If you find a moment, a curiosity and a pit of energy to comment or criticize, I would appreciate, much.

If one of those is damaged - please, forgive and disregard.

Nice weekend.

With best,

Victor

PS:

After two days, I am still "hurt" by your perspective on the dangling bonds optimization.

:-)

Well, I think, any metal-organic system consists nothing but a stet of dangling bonds, usually.

My data base accounts about 1000 systems: normal modes, mulliken, esp, 

natural transition orbitals - all done proper.

All of them are just dangling bonds about a metal,

save the "core" metallocenes, roussin's salts and some  "core" cubanes.

Of course, trying to step from a solid state to a metal-organic is a puzzle.

Such chemistry blows my mind, in fact.

Victor

 

 

 

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/ZRAP278MB0827C9A4CD229100185476E0F42DA%40ZRAP278MB0827.CHEP278.PROD.OUTLOOK.COM.