Have you tried X-ray absorption around the Fe L edges? XAS is quite powerful in determining different valencies and possibly the local environment (however, I would vote for resonant inelastic X-ray scattering for dd excitations that can give you more information). Quantitatively, it would be tricky for these techniques - they might give you the ratio between Fe with different valencies.

Spectroscopy is never really easy... But with iron, XAS is quite unique. Still, unlike with XPS, you will need access to synchrotron radiation to get your experimental data.

You may want to consult our 2004 paper:

"Probing composition and interfacial interaction in oxide passivated core-shell iron nanoparticles by combining x-ray absorption and magnetic circular dichroism"

K Fauth, E Goering, G Schutz, LT Kuhn, J. Appl. Phys. 96 (2004):399-403, for an example of such an analysis. The paper quotes relevant earlier material and it may be interesting to watch out for some of the subsequent work which cited ours. Some more material is covered in Mod. Phys. Lett. B 21 (2007):1179-1187.

Concerning XPS, I would guess that it should be much easier to find adequate literature. But when it comes down to analyzing experimental data, then expert expertise is essential.

Both, XPS and XAS (though to a somewhat lesser extent) are surface sensitive methods. If the dimensions of the grains in your powder exceed the probing depth of these methods, then both may totally miss the composition of your material if it is inhomogeneous (different near grain surfaces thanfurther inside).

OK Yousef, It is always useful to use thermal analysis to discard the presence of oxydroxide phases like goethite and to quantify it. XRD is quite good for detecting and quantify magnetite and hematite. Mossbauer spectroscopy is excellent to determine small differences between Fe phases and to detect deviations from ideallity in short rank behaviour.

Have come back to propose Mössbauer spectroscopy and see it' been done...

A combination of Mössbauer Spec, temperature dependent magnetization measurements would help you a lot. Look at my paper titled : Size dependent structural and magnetic properties of FeO-Fe3O4 nanoparticles, in Nanoscale 2013.

Fe-57 Mössbauer spectroscopy is effectively an excellent technique for this purpose. The problem is that in many of these iron oxide phases you'll have Fe3+ in an octahedral environment of six oxygens, which can hardly be distinguished by XAS.

In Mössbauer spectroscopy, they can be distinguished by the different local magnetic fields at the iron site. Have a look at geological studies or work on natural clays.

It might be necessary to work at low temperatures, if the particles of your powder are so fine that they show superparamagnetic behaviour at room temperature.

Rietveld Refinement, since you have more than 3%wt of the second phase for a two-phase system using Cu-Ka. There are a lot of programs available for this purpose currently, but, of course due to the non triviality of the method all the parameters must to be manipulated carefully. I use Jana2006, a free program for powder and single crystal structure solution and refinemnent. By Jana2006 you can get rather than a quantitative result, also you can get parameters related to the structure quality (modulation, tensors, defects and so on).

Thanks all for sharing your great hands-on experience and the valued literature.

If you are interested in synchrotron spectroscopy, please visit the following site -

http://www-als.lbl.gov/index.php/beamlines/beamlines-directory.html

and contact the relevant beamline scientist (beamline contact) for more information (for example, 8.0.1)

XRFS analysis of iron oxide is one of the most accurate methods used in commercial labs.

With XRFS, samples are pulverized and fine discs are prepared by fusion with mixture of lithium metaborate and lithium tetraborate along with none wetting agent. Fusion is at 1000 deg. TGA must be performed to account for loss on ignition. These prepared glass discs can then be analyed by xrfs. However, xrf operators must be well experienced to do instrument calibration and method validation.

Thanks, Saki. I think XRF can give you the total iron regardless of oxidation state.

It does give total iron but in addition gives iron as FeO and Fe2O3. Wet chemical method can be used for FeO but comparatively, xrf is preferred because it us fast and convenient for sample preparation.