Spring constant value

Dear Prof. Gale,

Can the spring constant values which we specify in the code be different compared to that in GULP library and previously reported literature ?

In my case, when I specify the spring constant value of Cu in Cu doped yttrium oxide to be 40, my code works… wherein when I specify it as 999999 which is specified in previously reported literature (not for the same doping), it does not work.

Could you please clarify the same?

Spring constants are parameters & so there is no special value - it can be whatever is fitted for a given material. A value of 999999 means a rigid ion but in old codes they expected all ions to have shells or none. In GULP just get rid of the shell when you see a value like this and make the potential act on the core.

Thanks Prof.Gale for this explanation.

I have one more query.

So, it is not necessary that all the atoms specified must have core and shells but it can be whatever it gets fitted to.

Am I right in this statement ?

Kindly clarify

Not sure I fully understand the statement. Atoms don’t have to shells, but if you don’t include shells you obviously can’t fit them. Physical intuition should tell you which ions are likely to be polarisable or not.

Thank you professor for this note.

While specifying the core and shell positions, is it necessary to specify the core and shell positions of all the atoms in a compound, or is it fine if some atoms have only cores. Like in a five element compound , four atoms have core and shells and one has only core.

I understand how core - shell model works, but asking for a clarification in specifying the co-ordinates.

Kindly clarify me this.

Thank you professor for this note.

While specifying the core and shell positions, is it necessary to specify the core and shell positions of all the atoms in a compound, or is it fine if some atoms have only cores. Like in a five element compound , four atoms have core and shells and one has only core.

I understand how core - shell model works, but asking for a clarification in specifying the co-ordinates.

Kindly clarify me this.

I think we’re going round in circles! The answer to this was back in the previous replies.

Thank you professor Gale for the response. Yes, I get it now. The example codes of GULP also helped me a lot.