Skip to main content

Iron sesquioxide under pressure: formation of a new iron oxide polymorph


High-pressure synchrotron-based XRD, EXAFS and SMS experiments combined with ab-initio theoretical calculations unveiled the behaviour of compressed epsilon-Fe2O3. Above 27 GPa, a possible spin crossover (HS to IS) related to the formation of a new iron sesquioxide polymorph (epsilon’-phase) was observed.

  • Share

Iron sesquioxide (Fe2O3) is one of the most studied materials and is continuously showing its importance in very different fields from geophysics to biomedicine and in various technological applications [1]. This material is found in five different crystalline phases at ambient conditions and each of them reveal different and exciting properties. The epsilon phase, in particular, has exhibited paramount magnetic properties (giant coercivity) and has recently been found in basaltic rocks as a nanomineral [2]. The latter discovery may imply that the presence of this material in the Earth’s interior is underestimated; however, this material should be stable at extreme conditions of pressure and temperature. This compound is also considered unique because it is the only ordered iron sesquioxide polymorph that contains crystallographic iron in tetrahedral coordination without the presence of oxygen vacancies.

ε-Fe2O3 crystallises in an orthorhombic structure, which is formed by four polyhedral units: a regular octahedron, two distorted octahedra and a regular tetrahedron (see Figure 1a). X-ray absorption fine structure (EXAFS) under pressure performed at beamline BM23 confirmed this arrangement by measuring the different average interatomic distances up to 27 GPa, where a sudden change occurred. The deformation of the different polyhedral units was also estimated by the analysis of the Debye-Waller factor associated to each average Fe-O bond length. Synchrotron-based angle-dispersive X-ray diffraction measurements of ε-Fe2O3 under pressure were performed at beamline ID27, characterising its pressure-volume equation of state and revealing a large structural stability (up to 27 GPa). This result is in good agreement with the EXAFS analysis and ab-initio theoretical calculations. This phase remains stable up to 27 GPa, which is the limiting pressure to access the upper Earth’s mantle. This phase was also found to be dynamically stable at 25 GPa and 1800 K by ab-initio theoretical simulations.

Structure layout of epsilon iron oxide and experimental (black circles) and theoretical (blue squares) pressure-volume equation of state of epsilon iron oxide

Figure 1. a) Structure layout of ε-Fe2O3. b) Experimental (black circles) and theoretical (blue squares) pressure-volume equation of state of ε-Fe2O3.

Above 27 GPa, a pressure-induced volume collapse occurs (see Figure 1b). This transition has been observed in several iron sesquioxide polymorphs and is mainly associated to a high-spin-to-low-spin crossover. The volume collapse is nicely described by ab-initio theoretical calculations but the theoretically-calculated magnetic moment in each polyhedral unit is incompatible with the low-spin state at pressures above 27 GPa. Thus, changes in structure were analysed by XRD and EXAFS, and changes in the spin state of the different polyhedral units were analysed experimentally through synchrotron-based Mössbauer spectroscopy (SMS) at beamline ID18.

Regarding structural changes, the regular octahedron and one of the irregular octahedral units were found to remain roughly in the same configuration while the other irregular octahedron and the regular tetrahedron became distorted octahedral units, closer to a 5+1 coordination (see Figure 2). The electronic distribution of the 3d levels of iron was studied by SMS under pressure (see Figure 2). The spectrum at ambient conditions reproduced nicely the results obtained in the literature [3]. The analysis of the evolution on the spin state and the coordination of the four inequivalent irons of this compound reveals a change above 27.5 GPa compatible with that observed by XRD and EXAFS. On the other hand, the analysis of the Mössbauer parameters obtained from the spectrum at 36.5 GPa, revealed the lack of iron in tetrahedral coordination and the formation of a new strongly distorted octahedron. These Mössbauer parameters indicate that this new polyhedral unit is compatible with iron in an intermediate state (IS), which has never been observed in iron sesquioxide compounds but has been reported in pentacoordinated iron in complex compounds [4], which coincides with the coordination observed in the structural study (5+1 coordination).  

Comparison between structural and electronic evolution of epsilon iron oxide under pressure

Figure 2. Comparison between structural and electronic evolution of ε-Fe2O3 under pressure.

In this work, the spin crossover transition responsible for the volume collapse observed above 27 GPa and the structural mechanism driving it have been identified. The transition allowed the high pressure phase to be assigned to a new polymorph of the iron sesquioxide compound named the ε’-phase. This experimental and theoretical joint study characterised the structural, electronic and magnetic behaviour of the ε-Fe2O3 under compression for the first time.


Principal publication and authors
Stability and nature of the volume collapse of ε-Fe2O3 under extreme conditions, J.A. Sans (a), V. Monteseguro (b,c), G. Garbarino (b), M. Gich (d), V. Cerantola (b), V. Cuartero (b,e), M. Monte (b), T. Irifune (f,g), A. Muñoz (h) and C. Popescu (i), Nature Communications 9, 4554 (2018); doi: 10.1038/s41467-018-06966-9.
(a) Instituto de Diseño para la Fabricación y Producción Automatizada, MALTA Consolider Team, Universitat Politècnica de València, Valencia (Spain)
(b) ESRF
(c) ICMUV. MALTA Consolider Team, Universitat de València, Burjassot (Spain)
(d) Institut de Ciència de Materials de Barcelona (ICMAB-CSIC), Bellaterra (Spain)
(e) Centro Universitario de la Defensa de Zaragoza, Zaragoza (Spain)
(f) Ehime University, Matsuyama (Japan)
(g) Earth-Life Science Institute, Tokyo Institute of Technology, Tokyo (Japan)
(h) Departamento de Física, Instituto de Materiales y Nanotecnología, MALTA Consolider Team, Universidad de La Laguna, San Cristóbal de La Laguna (Spain)
(i) ALBA-CELLS, Barcelona (Spain)


[1] V. Urbanova et al., Chemistry of Materials 26, 6653 (2014).
[2] S. Lee & H. Xu, Minerals 8, 97 (2018).
[3] M. Popovici et al., Chemistry of Materials 16, 5542 (2004)
[4] A.K. Patra et al., Inorganic Chemistry 45, 7877 (2006).