Yes, check this work for 5 nm nanoparticles... Temperature dependent Mössbauer.

Roca, A. G., Marco, J. F., Morales, M. del P. & Serna, C. J. Effect of Nature and Particle Size on Properties of Uniform Magnetite and Maghemite Nanoparticles. J. Phys. Chem. C 111, 18577-18584 (2007).

Juan

It would help if in addition to changing the temperature, you could apply a magnetic field.

Thank you, very much for the information. The paper you suggest is well written. Still I have a doubt regarding the conclusion from the spectra. Generally when we oxydized Fe3O4, core/shell of Fe3O4/gamma-Fe2O3 forms initially. In that case the signature of gamma Fe2O3 will be showing up in IR- spectrum. Can we conclude it to be gamma- Fe2O3 alone seeing the spectra. Regarding the Mossbauer spectra of Fe3O4 below Verwey temperature I feel very difficult to deconvolute into 5 spectrum, one from the tetrahedral site and 4 from the octahedral sites untill an unless magnetic field is applied.

Certainly the spectrum of the M5 and G5 mentioned in the suggested paper there is some diference at 16 K. This may be due to coarsening along with oxidation particles since even in the RT Spectrum the doublet is resolve in the case of RT.

If Mossbauer would not do if you may see if you would find difference in the (magneto-)optical spectra as for Fe3O4 the dmoinant spectra are intervalenec charge transitions9 see W.F.J. Fontijn et al . Phys Rev, B. 56 (1997) 5432.

Note that you should take great care in preparing the particles as they could well be some mixture

Hi, Electron energy loss spectroscopy (EELS) in TEM can give you good information particularly for nanoparticles less than 10 nm.

This paper could be a good reference,

http://prb.aps.org/abstract/PRB/v69/i12/e125415

y-Fe2O3 is considered a cation deficient Fe3O4. As a result, the saturation magnetization of Fe3O4 is slightly higher than y-Fe2O3. The sizes of your particles make magnetic or electrical characterization problematic. Fe3O4 has a lattice structure change around 125K, where there is an electrical resistance change (10^2) and easy magnetization axis change. Either can be witnessed in an R v. T, or M v. T figure. But, size is known to suppress this transition at 125K (Verwey Transition). So the absence of the tell-tale Verwey Transition doesn't necessarily rule out the presence of Fe3O4. Another option might be thermogravimetry (TG). I personally have not used TG though.

Hi,

from my point of view, as already pointed out by Jason Morales, you can not avoid the coexistence of Fe2O3 and Fe3O4 when it comes to nanoparticles; it merely results from the presence of vacancies, defects on the structure, dangling bonds, and probably the effect of the curvature as well, the density and even the composition of each particle is unique for each size. The exact assignment of observed FeOx to Fe2O3/Fe3O4 is unfeasible, its structural and magnetic properties differ from any bulk iron oxide phase as it is usually the case for passive films (see for instance M. F. Toney et al. Phys. Rev. Lett. 79, 4282, 1997 and also S. Couet et al. ibid 101, 056101, 2008). This may be one of the reasons for the usually observed decrease of magnetization at saturation on decreasing the particle size.

Thus, only when the size is beyond a critical diameter you may find a Verwey transition in the M(T) figure... on the other hand, Mossbauer could not tell you whether you have a core/shell Fe3O4/Fe2O3 structure, a mixture of particles, etc. The same applies for other techniques such as XAS.

In our case we have used HRTEM statistics to identify the Fe2O3-to-Fe3O4 ratio as shown in http://prb.aps.org/abstract/PRB/v74/i5/e054430

Just as mentioned before: there is a continuum of composotions in the spinel phase spanning all the way from maghemite to magnetite.

Just as with electron loss spectroscopy in a TEM, x-ray absorption has proved useful in detecting Fe species in nanocrystalline iron oxides.

The following 1999 work

http://onlinelibrary.wiley.com/doi/10.1002/%28SICI%291521-3951%28199909%29215:1%3C797::AID-PSSB797%3E3.0.CO;2-D/abstract

may be taken as a very good introduction, it has been quoted by quite a few more recent papers, some of which extend the analysis to magnetic characterization.

If you do XANES spectroscopy at the Fe-K edge, you could safely distinguish maghemite from magnetite by the edge position. The edge position for Fe(III) is at higher energies with respect to Fe(II). Passing from magnetite to maghemite, the shift in energy of the absorption edge is ca. 2.5 eV, and therefore is quite easy to determine. In addition, if you have Fe(II) in maghemite, comparison with standard samples will allow to determine the presence and amount. It works for particles down to 2-3 nm in size.