Weight Percentages for Multiple Phase PDF

danialf64

Hi,

I am performing pair distribution function refinement on Co/TiO2/Mn catalyst system. CT data was obtained so the capillary was removed.

My question is regarding how to calculate the weight percentages for different phases in the PDF. According to this tutorial, https://community.dur.ac.uk/john.evans/topas_workshop/tutorial_pdf_sno2.htm, the weight percentages are proportional to the molecular weight for a 2-phase PDF. Is the same true for pdf with more than two-phases? Also, is it the molecular weight of the unit cell or empirical formula?

I used the formulas from the tutorial, but just need to make sure that it is correct to use them.

Any help will be much appreciated.

rowlesmr

Hi Danial

I don't know how it would work with PDF, but as a guide, in "normal" XRD Rietveld analysis, the relative weight fractions of phases are given by

w_a = (sMV)_a / (Sum((sMV)_i, i = 1..N)

where s is the phase's scale factor, M is the unit cell mass*, and V is the unit cell volume.

See http://scripts.iucr.org/cgi-bin/paper?S0021889887086199 (the canonical reference) and https://doi.org/10.1107/S0021889887009415 (the other one - check out the submission dates!)

* Sometimes written as ZM, where Z is the number of formula units, and M is the mass of a formula unit.

Matthew

johnsoevans

For PDF the final point of the tutorial has what you want. Scale factor is proporotional to mole fraction. If you look at the INP file at https://community.dur.ac.uk/john.evans/topas_workshop/answers_topas/SnO2_2MoO3_dofr.inp, the scale is converted to a mass by multiplying by the RMM. massTotal is determined by summing the masses of all phases and weight_percent by massPhase/massTotal*100. You can just use the empirical formula as everything will scale. You should be able to generalise to >2 phases.

danialf64

Hi Matthew and John,

Thank you so much for the responses. It makes sense now!

Kindest regards,
Danial