If you ask to quantify I suppose you have crystalline + amorphous phases. Now, if the amorphous can be modelled with the same local structure as the crystalline part (mostly with polymers or organics) you can use the method as explained by Maykel. If you need to refine consider to bound the two structures before changing the crystallite size and r.m.s. microstrain (you select the phase object in the tree and use the "equal to" to bound to the previous phase you have duplicated. Then you remove the bound for the crystallite size and microstrain to change their values. Technically this correspond to the Le Bail approximation.

If the amorphous is a silica glass you can use the Le Bail approximation for the silica glass. In the Maud examples there is one, mixed silica glass and Al2O3. It is called the sio250.par the analysis file to load.

Otherwise the best method to quantify the amorphous if you have a powder sample (and only powder) is to use an internal standard. You add a know amount of a standard phase (I use Si or Al2O3, what is better at not overlapping much), from 10 to 30% should be sufficient. Then you measure the diffraction pattern and use Maud (but any other Rietveld software can be used as well) to quantify the crystalline phases ignoring the amorphous (either with some very large gaussian peaks in the background to fit the bumps or the interpolated background).

The you use the standard formula to get the amount of amorphous from what you analysed. If Wa is the amorphous fraction you are looking for, Ws is the amount of standard you added and Wm the amount of standard you got from the Rietveld analysis (they do not correspond as you where ignoring the amorphous) then you calculate:

Wa = (1- Ws/Wm) / (1 - Ws)

is very simple, accurate and require only one measurement. If you have a bulk not a powder you can use the Le Bail approximation but you need the amorphous local structure as explained before.

Duplicate the phase, select the phase, click Edit selected object (toolbar shortcut), click Macrostructure, click Option in Size- Strain model ( isotropic), change the crystallite size value so that the amorphous peak broadening close to the amorphous obs. hump, set the R.m.s to zero, key in the micro absorption*, click Compute Spectra to see the result.

*Brindley, G. W., Philos. Mag., 36, pp. 347-369, 1945 or Taylor, J. C. and Matulis, C. E., J. Appl. Cryst., 24, pp. 14-17, 1991,

thank you so much dear Maykel for clear and informative response 

for amorphous material, there should be no peak in xrd, but in case of mixed phase you will get a hump in your xrd. using maud you can remove this as a background using maud or fullprof or powderX.

If you ask to quantify I suppose you have crystalline + amorphous phases. Now, if the amorphous can be modelled with the same local structure as the crystalline part (mostly with polymers or organics) you can use the method as explained by Maykel. If you need to refine consider to bound the two structures before changing the crystallite size and r.m.s. microstrain (you select the phase object in the tree and use the "equal to" to bound to the previous phase you have duplicated. Then you remove the bound for the crystallite size and microstrain to change their values. Technically this correspond to the Le Bail approximation.

If the amorphous is a silica glass you can use the Le Bail approximation for the silica glass. In the Maud examples there is one, mixed silica glass and Al2O3. It is called the sio250.par the analysis file to load.

Otherwise the best method to quantify the amorphous if you have a powder sample (and only powder) is to use an internal standard. You add a know amount of a standard phase (I use Si or Al2O3, what is better at not overlapping much), from 10 to 30% should be sufficient. Then you measure the diffraction pattern and use Maud (but any other Rietveld software can be used as well) to quantify the crystalline phases ignoring the amorphous (either with some very large gaussian peaks in the background to fit the bumps or the interpolated background).

The you use the standard formula to get the amount of amorphous from what you analysed. If Wa is the amorphous fraction you are looking for, Ws is the amount of standard you added and Wm the amount of standard you got from the Rietveld analysis (they do not correspond as you where ignoring the amorphous) then you calculate:

Wa = (1- Ws/Wm) / (1 - Ws)

is very simple, accurate and require only one measurement. If you have a bulk not a powder you can use the Le Bail approximation but you need the amorphous local structure as explained before.

We used Maud to determinate the amorphous phases in

http://www.sciencedirect.com/science/article/pii/S0022309303003120?np=y

http://www.sciencedirect.com/science/article/pii/S0167577X0600348X

I hope it can help.

S. Palomares

I have a mixure of Fe, Ni, Ta and B nano-powders milled for 120h. And I am analyzing the data using MAUD. To see the amorphous content do I need to duplicate all phases as explained by Maykel Manawan?