Hi Matt, Concerning the third cumulant, my problem is to use a particular model for it. Some articles (Frenkel and Rehr, PRB 48 (1993) 585; Araujo et al, PRB 74 (2006) 184102) show how to use the Einstein model on Artemis, since these articles show the expressions that should be defined for sigma^2(T) and C3(T) for this model in Artemis (eqs. 11 and 18 of Frenkel and eqs. 2 and 3 of Araujo). In fact, I have already tested them on my analysis, using the eins function for sigma^2 and defining the C3(T) expression given in those articles in Artemis. I'd like to test Debye model too, and my problem is to use Debye model for sigma^2(T) and C3(T). For sigma^2(T) Artemis has the debye function already predefined. I'd like to use/write an expression for C3(T) considering Debye model, but I didn't find anything about such expression on the literature. Kleber
On Wednesday 12 May 2010 01:59:02 pm Kleber Daum Machado wrote:
Hi Matt,
Concerning the third cumulant, my problem is to use a particular model for it. Some articles (Frenkel and Rehr, PRB 48 (1993) 585; Araujo et al, PRB 74 (2006) 184102) show how to use the Einstein model on Artemis, since these articles show the expressions that should be defined for sigma^2(T) and C3(T) for this model in Artemis (eqs. 11 and 18 of Frenkel and eqs. 2 and 3 of Araujo). In fact, I have already tested them on my analysis, using the eins function for sigma^2 and defining the C3(T) expression given in those articles in Artemis. I'd like to test Debye model too, and my problem is to use Debye model for sigma^2(T) and C3(T). For sigma^2(T) Artemis has the debye function already predefined. I'd like to use/write an expression for C3(T) considering Debye model, but I didn't find anything about such expression on the literature.
The paper by Frenkel and Rehr has formulas in it relating the cumulants to other parameters, like sigma^2 and an expansion coefficient. You can write any math expression you want as the value for a path parameter. Indeed, the eins function is not very complicated and could easily be implemented as a path parameter math expression. But eins is a common one, so Matt included it in Ifeffit. Ifeffit (and hence Artemis) don't have any special functions for implementing what you want, but the beauty of Ifeffit is that you get to write down any math expression that embodies the physics of what you are trying to measure. So, if you have a cumulant model in mind -- just do it! B -- Bruce Ravel ------------------------------------ bravel@bnl.gov National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 My homepage: http://xafs.org/BruceRavel EXAFS software: http://cars9.uchicago.edu/~ravel/software/exafs/
Kleber, Correlated Debye Model (CDM) was proposed by Sevillano, Meuth and Rehr in 1979 in a Phys Rev B artcile. It also shows how to find sigma2(T) from the CDM- and also from Einsten model, by the way. In the former case, you take phonon density of states in Debye approximation (where it is proportional to the square of the mode frequency) and projoect it on the direction of the bond between the atoms, so the modified projected density of states is used to calculate vibrational amplitudes. Similarly, it can be used to calculate not vibrational amplitudes but third cumulant but it has not been done with CDM to my knowledge. It was done only for the Einstein moedl in the subsequent papers though, as you and Bruce wrote. Anatoly ________________________________ From: ifeffit-bounces@millenia.cars.aps.anl.gov on behalf of Kleber Daum Machado Sent: Wed 5/12/2010 1:59 PM To: ifeffit@millenia.cars.aps.anl.gov Subject: [Ifeffit] Debye model on Artemis Hi Matt, Concerning the third cumulant, my problem is to use a particular model for it. Some articles (Frenkel and Rehr, PRB 48 (1993) 585; Araujo et al, PRB 74 (2006) 184102) show how to use the Einstein model on Artemis, since these articles show the expressions that should be defined for sigma^2(T) and C3(T) for this model in Artemis (eqs. 11 and 18 of Frenkel and eqs. 2 and 3 of Araujo). In fact, I have already tested them on my analysis, using the eins function for sigma^2 and defining the C3(T) expression given in those articles in Artemis. I'd like to test Debye model too, and my problem is to use Debye model for sigma^2(T) and C3(T). For sigma^2(T) Artemis has the debye function already predefined. I'd like to use/write an expression for C3(T) considering Debye model, but I didn't find anything about such expression on the literature. Kleber _______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
participants (3)
-
Bruce Ravel -
Frenkel, Anatoly -
Kleber Daum Machado