Disentangling the ferroelectric phases of epitaxial hafnia
Johanna van Gent ∗, Ewout van der Veer, Yulei Li, Daniel A. Chaney, Beatriz Noheda · 2026
Read Disentangling the ferroelectric phases of epitaxial hafnia by Johanna van Gent ∗, Ewout van der Veer, Yulei Li, Daniel A. Chaney, Beatriz Noheda (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-15081.txt.
Introduction
van Gent Gonzalez and colleagues settle a dispute over the crystal phases of ferroelectric hafnia films. Standard probes struggle to separate the polar orthorhombic phase from a reported rhombohedral phase, yet device behavior hinges on which symmetry holds. Using 3D synchrotron diffraction maps of films in both phases, plus temperature and electrical checks, they confirm two truly distinct phases. The result clarifies the traits each phase brings to ultrathin devices.
Catalog record
| CREATOR | Johanna van Gent ∗, Ewout van der Veer, Yulei Li, Daniel A. Chaney, Beatriz Noheda |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32125 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.15081
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Disentangling the ferroelectric phases of epitaxial hafnia Johanna van Gent ∗, Ewout van der Veer, Yulei Li, Daniel A. Chaney, Beatriz Noheda Abstract Since its discovery, ferroelectric hafnia has been extensively studied due to its CMOS-compatibility and ability to remain polarized at sub-10 nm thicknesses. The ferroelectric behaviour is generally attributed to a polar orthorhombic ”OIII” phase. However, a second polar phase with rhombohedral symmetry (R-phase) has also been reported in epitaxial films. The nature of the R-phase remains disputed due to the subtle differences with the OIII-phase when probed by standard thin film characterisation techniques. Given the functional properties of ferroelectrics are crucially determined by the crystal symmetry, resolving this matter is imperative. In this work, we settle the controversy through extensive 3D reciprocal space surveys made possible via synchrotron-based grazing incidence diffraction from epitaxial films of both phases. These experiments, together with direct comparison of their temperature dependence and electrical responses, conclusively establish them as two distinct phases and provide insight into their key characteristics. Ferroelectric materials contain significant promise for non-volatile random access memory, operating faster and more energy-efficiently than existing alternatives [ 29 ] . However, this potential has been limited for figurehead ferroelectrics like PbZr 1-X Ti x O 3 and BaTiO 3 due to depolarization below 10 nm thicknesses and poor CMOS compatibility [ 22 , 28 , 29 ] . In 2011 a metastable phase of HfO 2 , a material already well-integrated into CMOS technologies as a high-k dielectric, was reported to exhibit nanoscale ferroelectricity [ 3 ] , overcoming both key limiting issues. Ferroelectric behaviour is attributed to a non-centrosymmetric orthorhombic “OIII-phase”, most commonly described by space group Pca2_{1} , Fig. 1 a. This peculiar structure, previously observed in partially-stabilized zirconia (Mg 0.1 Zr 0.9 O 2 ) [ 17 ] , is characterised by a unit cell in which half the oxygen ions contribute to the spontaneous polarization parallel to the b -axis, while the other half form a non-polar spacer layer [ 3 ] . Despite many varying hypotheses [ 18 , 12 , 19 ] , the origin of the uniquely scalable behaviour of ferroelectric hafnia remains unclear. In this context, epitaxial films have been presented as a better platform to investigate the polar nature of the ferroelectric phase when compared to their polycrystalline counterparts [ 33 , 37 , 20 ] due to reduced structural complexity and a minimization of extrinsic effects. However, reports of a second set of metastable ferroelectric phases with rhombohedral symmetry (R-phase), assigned to space groups R3m (Fig. 1 b) and R3 , for films grown epitaxially on SrTiO 3 (001) [ 37 ] and GaN(0001) [ 2 ] , respectively, revealed an additional nuance. In these reports, the R-phases were primarily identified through an elongation of the out-of-plane body diagonal (d 111 ) with respect to all (equal) in-plane body diagonals due to large compressive epitaxial strain (Fig. 1 d), with the 1/3 multiplicity of the {111} reflections implying a rhombohedral space group. The unlikelihood of a strain-induced tetragonal-rhombohedral or orthorhombic-rhombohedral phase transition suggests that the R-phase is most likely a distortion of the cubic fluorite structure [ 37 ] , therefore following a different transformation path from the OIII phase reported to evolve from the tetragonal fluorite phase [ 3 , 23 , 27 ] . Figure 1 : Schematic of a) the R3m structure and domains in both rhombohedral and hexagonal settings b) the [111]-oriented Pca2_{1} OIII-phase and domains, differences between axes length exaggerated for clarity. c) Laboratory source \theta/2\theta ‘specular’ scan of the HZO ||LSMO||STO(001) system compared to theoretical predictions for different hafnia polymorphs. d) Pole figure around specular (111) reflection ( \lambda =1.541 Å) and corresponding pole scans compared to the specular (OOP) reflection ( \lambda =1.378 Å) for original R-phase system, adapted from Wei et al. [ 37 ] . The discovery of the R-phase in hafnia-based films was and remains controversial for a number of reasons. Firstly, the formation energies for the R-phases are significantly higher than those of the OIII-phase [ 37 , 13 ] . Secondly, the ferroelectric behaviour of R-phase films resembles that of the OIII-phase in both P r values and scalability, contradicting significantly lower theoretical P r predictions and leading some to suggest a ”rhombohedrally-distorted orthorhombic” phase [ 39 ] . This controversy has remained active due to the difficulty distinguishing the two phases via conventional thin film diffraction approaches, namely; ‘specular’ θ/2θ measurements, texture (pole figure) and in-plane measurements, and classical reciprocal space mapping (RSMs). As shown in Fig. 1 c, specular studies are insufficient due to the small variation of the lattice parameters relative to a large peak width. The combination of texture and in-plane measurements responsible for the original discovery (Fig 1 d), as well as RSMs [ 26 ] , can all provide greater insight. However, such measurements have stringent resolution and flux requirements, are rarely performed, and often limited to exploring a small number of reflections. As a result, the current state of epitaxial hafnia literature has become exceedingly confusing. Epitaxial Hf 0.5 Zr 0.5 O 2 grown by pulsed laser deposition (PLD) on SrTiO 3 (001) with La 1-x Sr x MnO 3 (LSMO) buffer layers under similar conditions are reported as rhombohedral [ 37 , 26 ] , orthorhombic [ 20 , 32 ] and “rhombohedrally-distorted orthorhombic” [ 39 , 16 ] . Since the existence and orientation of the polarization in ferroelectrics is determined by the crystal symmetry, clarifying the structure of these systems is key to achieve full control of the functional properties. To remove the remaining ambiguity surrounding the epitaxial R- and OIII-phases requires the expansion of our ”field of view” in reciprocal space beyond that provided by the aforementioned diffraction techniques. Here we employ synchrotron based, grazing incidence diffraction with a large area detector to collect extensive and uninterrupted 3D reciprocal space volumes, an approach whose power has been demonstrated recently in other epitaxial systems [ 4 , 40 , 1 ] . This technique allows for the reconstruction of arbitrary reciprocal space planes and volumes, providing, for the first time, a complete and undistorted view of reciprocal space for the epitaxial hafnia based systems. In order to clearly compare the two proposed phases, PLD-grown epitaxial hafnia-based films on both SrTiO 3 (STO) substrates, reported to be polar rhombohedral [ 37 , 26 ] , and on Y 0.095 Zr 0.905 O 2 (YSZ) substrates, only reported to exhibit the polar orthorhombic phase [ 33 ] were explored. The ferroelectric properties of both systems have been characterised to confirm structure-functionality links, and their respective transformation pathways have been explored via in situ high temperature measurements. Taken together these studies constitute a comprehensive insight into the similarities and key differences of both ferroelectric phases of epitaxial hafnia. Comparison of ’as grown’ diffraction Figure 2 : a) In-plane ( HKI2 ) and b) out-of-plane ( H0\bar{H}L ) reciprocal space reconstructions for a 10 nm YHO (0001)_{\mathrm{H}} film grown on STO(001)/LSMO. Left/right panels show data overlaid by R-/OIII-phase simulations, respectively. Sphere colour represents origin phase and domain, with diameter depicting relative predicted intensity. For a) the simulation has been partially removed to display