Retained Austenite is commonly investigated by short wave length in order to get as many as possible reflections, The reason is the typically existing texture which needs to be considered for e reasonable quantification. Cr you should use for residual stress only. There intense reflections appear at high Bragg angles which increases the precision of the observable shift dramatically.

Regarding your sample size: 5mm should not make any problem, even for W or a similar short wavelength radiation if you are using a collimator not bigger than 1mm. With 2mm you perhaps hit more than the sample. For Cr it is no problem, but you cannot quantify the intensity in a meaningful way by only 3 peaks. In so far, any discussion about the background is obsolete. Look for another radiation!    

thank you for responding to the question ... it is very detailed and knowledgeable...i would also like to have your views on another question regarding intensity of reflected radiation.

If the intensity of reflected radiation is enough to reveal the peaks of austenite and ferrite and to quantify a phase fraction. Does it matters if reflected intensities have lower cps.

for eg: when i use a 2mm collimator i get a higher intensity of reflections and when i use a 0.5 mm collimator, i get lower intensity of the reflections. However, the relative intensities of a particular phase behave in the same fashion in both cases and thus the quantified phase fractions remains the same. Therefore, can one say that the intensities using 2mm collimator are more accurate than using a collimator of 0.5mm.

I have tried to make it easy for you to understand... i hope you understand the question and would like to hear your view on it ...

Thank you.

Lets say, it is statistically more precise. It depends on the better signal to noise ratio. However, more important is that you cover a bigger range with a higher number of diffracting grains. In other words, if you have a 2 and a 1 mm collimator the covered area is of the 1mm is only 1/4 of the 2mm, so that the intensity is 4 times lower. Assuming a constant noise your signal to noise ration is 4 times better for the 2mm collimator (approximately :-)). You could compensate this with a 4 times longer acquisition time using the 1mm pinhole but the results won't be the same since the number of grains in the illuminated area is smaller. BUT: in order to measure with high precision you should not use primary pin holes in this dimension 2mm. I would recommend to use 1x4mm slit. But it really depends on the wave length (Bragg angles). All parameters are connected to each other and so a simple recommendation is not easy.

Hi Arbab Rehan,

According to your question, it happened to me to calculate the percentage of retained austenite using the Guinier's method (RX); you can found some details in the following article (published in open access);

M.L. FARES, M. ATHMANI, Y. ABIDI: Surface characteristics analysis of gas carburized new hot working tool steel. e-Journal of Surface Sciences and Nanotechnology, (2011) 09, 430-437.

@ M.L. Fares: Residual austenite determination is a "simple" determination of a phase using powder diffraction. Assuming that the crystal size is suitable (what is only roughly given in case of a bulk material) there is currently only one reliable technique: full pattern refinement. The complexity of this technique already indicates where are the weak points and which requirements have to be fulfilled. Older techniques can work and are sometimes underestimated but in this case I believe (!) that full pattern refinement is the only preferred technique. However, one need to keep the limits in mind: enough reflections, low noise, phases are known, simple or preferably no texture.

I am not sure how big are the errors of your analysis in the paper. Since in 1956 Guinier couldn't use Rietvelds idea I guess he is using simple intensity rations, perhaps a averaging of peak doublets, but this can give you in maximum a glue or some kind of relative changes between the different materials. What do you think?

How did you determined the austenite concentration? I did not find a more detailed description in the text.  

@G. Nolze: Rightly, I have no doubt that the used method in this paper contains some errors in the results. The latter were obtained by the method of Guinier according to an established program for this purpose. You also could consult another article (see the attached file below) where I presented an example of data:

@ M.L. Fares: Could you shortly describe what did you do with the XRD diffractograms (background)? How did you extracted the intensities, and how did you calculated the intensity ratio between ferrite (martensite) and residual austenite in order to determine the phase fraction? Many thanks!!!

@G. Nolze: I am very sorry because I can not give you further clarification regarding your request, except to say that the the values determination of the various intensities was performed using a software (associated with XRD diffractograms ) and these values were directly injected into another calculation program compiled on the basis of the Guinier's method. For your information purposes, all the manipulations were be done at the Ecole des Mines de Saint-Etienne (France) so that we could not have direct access to the "root" of such program. However, iIf you are thoroughly interested (and it seems that is the case), please contact directly "SMS center" of this School.

Well, I have to admit that the term "Guinier's method" does not tell me anything, which is perhaps my fault. It might be that I know it under another term (or I don't use any special term for it since it is some basic information in XRD). Therefore I was asking for. Nevertheless, if you recommend here to use it for the determiantion of residual austenite you certainly roughly know how it works, right? And since you cited it in your paper you should have read it anyway.

@G. Nolze:

Let me tell you that your last criticism is not appropriate...Nevertheless between scientists I can explain myself:

Guinier's method is a method similar to others (e.g. “Laue method”) to which further explanations can be found, by way of example, in the following documents.

 1. P. Keller: Quantitative, röntgenographische Phasenanalyse verschiedener Rosttypen, Materials and Corrosion (1967), 18(10), 865–878 (Wiley).

 2. Martin Priese:Praparation und strukturelle Charakterisierung der perowskitverwandten Uberstrukturphasen ScF3 und (La0.8Sr0.2)0.92 MnO3, DISSERTATION zur Erlangung des Doktorgrades des Fachbereichs Chemie der Universität Hamburg, 2002

 So, as I already mentioned, the results (e.g. diffracted beams intensities, diffraction angles ... etc.) derived from this method (see Table in the article) have been directly treated using software specially designed to reach the determination of the percentages of phases in presence. To be clear, in my study I was only interested by the determination of the latter without explicitly look how they have been found since the lab technicians experts were charged for it. It's true, I should have, as you specify, look to have more information on this, but at this point it depended on my time-consuming.

I would like to thank you all for your views... it has really helped me a lot ..