Your absorption will depend upon the chemical species present, whether or not you're close to an absorption edge for one of those species, and the density of the specimen. The only variable that's changing with pressure is the density of the gas/specimen in your pressure cell (and of course your density will likely change across the phase transformation).

Given the compressibility of your solid should be small (although I don't know what pressure range you're working over or what your solid is), perhaps you can get away with approximating your density as constant for the purpose of the absorption correction? A convenient calculator for absorption in capillary transmission can be found here

http://11bm.xray.aps.anl.gov/absorb/absorb.php

I believe the correct absorption correction type in GSAS is the type 0 function. Check the manual for the correct units.

Good luck!

Thanks for your reply Peter Metz

               I am familiar with the absorption calculator that you have linked, this can be used for the XRD pattern at ambient conditions and the type 0 function in GSAS . But my concern is how can i extend this idea to  high pressure data, were the sample thickness and density is varying, i am not able estimate thickness. Also the phase fraction will be varying across the phase transformation region.

if you do not know the shape and size of the sample that was illuminated you will not be able to a proper absorption correction. If you get negative thermal parameters that is a clear sign that your absorption correction is wrong as these are strongly coupled. If your beam size is smaller than the sample size this also complicated the correction.

As was already said, shape and size affect the absorption, however the most important are the elements of your sample. So, if you check the absorption, by guessing some parameters in the above mentioned webpage, maybe is even negligible ( mR < 5) so you don´t really need to take care about it. However is the absorption is high, even when you include in your favorite refinement program (I don´t know how it is done in GSAS), it will be only an approximation. Even in that case, I suggest you to use the same fixed value for the absorption parameter, for all the measured powder pattern.

The question then is what was the x-ray energy, or more importantly the absorption length of the sample at that energy. This will determine whether it is negligible or significant.

Is there a need for such corrections in XRR measurements from 0.4 to 1.4 degrees range, for example? 

The sampling area changes by orders of magnitude in this geometry.

Thank you all for your comments.

@ Rodrigo Castillo I have tried what ever u suggested,  and the thing is that i am getting a value mR

The determination of the mass absorption coefficients for each chemical element and for a given wavelength used, can be located in the International Tables for Crystallography (I use an old version, Volume III, pages 162-165).

The expression used is μi = d Σi Pi (μi / ρ), where (μi / ρ) = mass absorption coefficient, and d = density (g/cm3).

It's not the mass absorption coefficients that are the problem. It's the fact that the sample shape, size, and volume illuminated by the beam is not known, which makes it very difficult to get a valid correction that takes into account the 2theta dependency and allows correct structure factors to be obtained for high angle peaks. For determining thermal parameters this is key.

Maybe u should try the preferred orientation correction before u refine to the thermal and atomic displacement.

Can u attach ur data?

             Calculating the mass absorption coefficients can be done in this , but as Lawrence Margulies told, the problem is with the geometry of the sample. Now I am trying to estimate the shape of the sample and illuminated area, and searching for a possibility to model it with pressure variation.