May be the following reference is useful:

Klug and Alexander, X-ray diffraction procedures, New York, 1974.

As long as the crystals are small enough there is no big difference to a powder. The only recommendation is that you shouldn't use Cu radiation because of the fluorescence. It doesn't mean that it doesn't work, but you loose a lot of intensity. If you have a secondary monochromator you even can filter out this part of radiation. But the easier way is to use another X-ray tube.

The application of programs like Fullprof - i.e. full pattern analysis software is recommended. You need to know all phases in order to determine the phase fraction. But this is equivalent to powder diffraction.

Be careful with sample preparation. it might be that by grinding (what is commonly possible) you transform phases by the energy you are introducing by these processes. Of course you can have texture, but also this is a big challange in powder diffraction so that the respective consequences are known and discussed in XRD books. A famous standard publication is the recommended Klug, Alexanderm but there are many books which are useful as well, especially since there new tools are discussed which were still unknown 60 years ago. 

http://cdn.intechopen.com/pdfs-wm/33188.pdf

is a useful article. Even then if you have questions please ask me at : [email protected]

U can use the Rietveld package Quanto available from the Institute of Crystallography, Bari for phase quantification. I have used the software on dual phase steels and the results were quite comparable to those from VSM and EBSD.

https://books.google.iq/books?id=lvYbBQAAQBAJ&pg=PA144&lpg=PA144&dq=measurement+the+quantity+of+phases+in+metals&source=bl&ots=Nbc1cZnRZG&sig=QnSRO8tt_Gfcim0R7m0NUouUBy0&hl=en&sa=X&ei=XmCHVLH8O4rxUr-HgoAF&ved=0CCIQ6AEwAg#v=onepage&q=measurement%20the%20quantity%20of%20phases%20in%20metals&f=false

read this book

best regard

For better answer, please specify your material, exactly. For example, the direct comparison method can be used for quantitative phase measurement in stainless steels. Also, check it: http://www.sciencedirect.com/science/article/pii/S092150931401466X

All answers given above are correct

rietveld analysis using fullprof suite is the best

Rietveld analysis is actually for crystal structure refinement using powder XRD data. What you mean is full pattern analysis (therefore the name Fullprof). Moreover, Fullprof is only one option. There are several commercial and free software packages which allhave their special advantages. I would recommend a real fundamental pattern approach as used in BGMN. There no mathematical functions are assumed for the description of peak profiles. They will be derived from the experimental setup of your equipment and prevents meaningless minima refined. 

http://www.innovationservices.philips.com/sites/default/files/materials-analysis-xrd.pdf

Dear Sudhakar,

The best method for metals with XRD is to prepare a very clean polished sample for doing a complete spectra of powder XRD in reflexion mode.

Then you should use the Rietveld method to quantify the ammount of each phase. There are several especialized textbooks about the Rietveld method, which will explain you how to do it. If you have a general service or a laboratory in your University devoted to XRD, probably they will know how to do it. I send you here attached two papers in which I applied the Rietveld method to quantify the ammount of phases in an Al-Li alloy. Your material is different but the method should be the same.

You should pay attention to the texture of your material. Texture add an important difficulty to quantify the ammount of each phase. Hopefuly, in your case the austempered ductile iron probably donot exhibit a strong texture.

I hope this could be useful for you.

Dear Jose, only a short comment which has been pointed out by many other colleagues as well (but somewhere else): don't use spectra for diffractograms. A spectra is a linear distribution of energies. Looks similar but is something else. You also wouldn't call a quad a car only because it also has 4 wheels. Personally (and this is indeed very personally) I am not happy with mixing the Rietveld method with full pattern analysis, although mathematically the refinement algorithms are comparable. Rietveld developed this code for crystal structure refinement using neutron powder diffraction data. Much later (!) people realized that this refinement can be also applied for the refinement of the scaling factor for each phase which made it possible to quantify mictures of phases. The challange there is the peak overlap of different phases which becomes really difficult if phases generate anomalous peak broadening. Nevertheless, it is much better than the formerly used techniques like specific mixtures of phases and interpolations, or RIR procedures. 

Regarding your papers I only wnat point out that your material can be analysed by the mainly used Cu radiation. For Fe containing phases you will have a lot of fluorescence, i.e. a huge "constant" background and a comparatively bad signal/background ratio. Therefore, mineralogists usually uses Co radiation to prevent this effect, but then you perhaps get some problems with other elements like Cr which is quite common in steel, although the amount is often not that high.

May be the following reference is useful:

- A. Guinier, Théorie et technique de la radiocristallographie, Ed. Dunod, (Paris,    France, 1956).

- you can use also this article: