It's called the Reference Intensity Ratio method - paper in the link. Read about RIR method.

Rietveld refinement will NOT help you find additional phases.

http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=8602057

Dear Sir, Rietveld refinement is useful to find the additional phases.

It's called the Reference Intensity Ratio method - paper in the link. Read about RIR method.

Rietveld refinement will NOT help you find additional phases.

http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=8602057

Hi Ian, I wouldn't go that far :-).

Rietveld refinement (better: "full pattern refinement" since from my point of view Rietveld should be used exclusively for a structure refinement and not for a "simple" scaling of phase intensities) can be sometime very helpful in order to find minor phases using the difference plot since there residual peaks become better visible than using a peak search or simply humans eye.  But in general it is true: you need to know all phases, except you are only interested in the relative fraction od two or several phases in a mixture (e.g. polluted by an amorphous phase which you don't need to know, except you destroy the crystallinity of a phase during milling). Also weak-scattering phase can only generate low-intense reflections but their amount within a mixture can be clearly higher that their intensity suggest.

The reference intensity ratio can be only a rough instrument since for the same phase e.g. quartz, different values are given in databases. Which one you finally use will decide about your phase fraction. Therefore, full pattern refinement as last step is currently the most reliable tool, assuming that you did not change the (detectable) composition of your sample during sample preparation.  

How to determine chemical composition of a rock by X-Ray power Diffraction pattern