I'd like to use the third cumulant option in Artemis, but considering the Debye model. Artemis has a Debye function for the second cumulant (Debye-Waller factor) but not to the third. How could this be done in Artemis? Kleber
Hi Kleber:
I think of the Debye model as a model for the stiffness of the bond
between two atoms that is like a spring. The vibration depends on the
length of the spring (distance between the atoms) and the stiffness of
the spring as related through the Debye temperature. Above the Debye
temperature the spring is weak and below it is strong. This all works
fairly well to describe a symmetrical change in the bond length
between the two atoms. It is symmetrical because the bond shortens
and lengths by the same amount about a relaxed spring length.
A third cummulant adds another freedom in that the amount that the
spring shortens is less than the amount that the spring lengthens.
This may be the case because the atoms centers repel strongly so the
spring may not be able to shorten as much as it is able to lengthen.
Very weak bonds like those in solution often have third cummulants.
The Debye model does not contain information about a third cummulant.
The spring like nature of the bond is not simply related to the
skewing of the vibrational amplitude toward larger values.
HTH
Shelly
On Tue, May 11, 2010 at 7:42 AM, Kleber Daum Machado
I'd like to use the third cumulant option in Artemis, but considering the Debye model. Artemis has a Debye function for the second cumulant (Debye-Waller factor) but not to the third. How could this be done in Artemis?
Kleber _______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Hi Kleber,
I'd like to use the third cumulant option in Artemis, but considering the Debye model. Artemis has a Debye function for the second cumulant (Debye-Waller factor) but not to the third. How could this be done in Artemis?
The debye and eins functions in Artemis/Ifeffit calculates sigma2 only. One could use the debye (or eins) function for sigma2 and either allow the third cumulant to vary freely or constrain it to be based on the value of sigma2. There are a few approaches to this in the literature, but I'm not aware of a general solution. Would that approach be sufficient, or do you have a particular model relating the third cumulant to sigma2 in mind? --Matt
participants (3)
-
Kleber Daum Machado
-
Matt Newville
-
Shelly Kelly