First-order polarization process as an alternative to antiferroelectricity
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.
Kai: I'm Kai.
Mira: And I'm Mira, here with Lev.
Kai: We're looking at the paper: "First-order polarization process as an alternative to antiferroelectricity".
Mira: Here's the gist of the paper. First-order polarization process as an alternative to antiferroelectricity shows that appealing double-hysteresis P-E loops can appear from a field-induced abrupt rotation of polarization in non-polar CaTiO3 films Context and Motivation Antiferroelectrics generate significant interest since their polarization versus electric field (PE) curves show typical double-hysteresis loops appealing for various applications Unfortunately, antiferroelectrics are rare.
Kai: The paper investigates whether a similar mechanism could exist in polar materials, leading to a first-order polarization process capable of generating double hysteresis loops without invoking a conventional antipolar ground state.
Mira: They focus on CaTiO3, which can be turned ferroelectric under tensile strain, to explore this idea.
Kai: Experimental Setup and Strain Effects The study utilizes epitaxial thin films of non-polar CaTiO3 grown on differently oriented NdGaO3 substrates using a combination of theoretical and experimental techniques.
Title and authors: Kai: So we’re looking at this paper today titled "First-order polarization process as an alternative to antiferroelectricity". We've already talked about what that means in general, but let's get into the specifics of who wrote it and what they were trying to achieve.
Mira: The authors are a group from different places, which you can see by looking at their affiliations. They’re bringing together solid state physics and theoretical materials physics to tackle this problem.
Lev: I wonder how much experimental work versus simulation is in this paper, because when I look at real hardware development, the gap between theory and what you can actually build is where things get tricky.
Kai: The authors use a combination of theoretical calculations and actual experimental measurements on thin films grown on different substrates to see how strain affects these polarization states.
Mira: That’s the setup. They’re using epitaxial thin films of non-polar CaTiO3 grown on NdGaO3 substrates in two different orientations: (one hundred ten)o and (one)o <ref:2603.27590#pg3>.
Lev: So the substrate orientation dictates the mechanical constraints, which means they are essentially tuning the energy landscape of the CaTiO3 film to see what kind of polarization switching emerges.
Kai: Right. The paper’s goal is to see if this strain-induced tuning can stabilize different polar states that lead to double hysteresis loops without relying on a conventional antiferroelectric ground state.
The paper's summary: Mira: So the paper explains that by applying tensile epitaxial strain, they can tune the energy and polarization landscapes of the CaTiO3 films to stabilize different oxygen octahedral rotation patterns.
Kai: They found that for CaTiO3 films strained on NdGaO3 (one hundred ten)o, when you apply an electric field along apc, you get a well-defined ferroelectric hysteresis loop <ref:2603.27590#pg3>.
Lev: That’s the key finding—a clear response in one direction—but what happens if you switch to the other orientation?
Mira: On the (one)o substrate, they observed a double hysteresis loop when the electric field is applied along bo <ref:2603.27590#pg1>. That specific behavior is what caught their attention.
Kai: And they suggest that this double hysteresis loop in the (one)o case isn't smooth; it comes from an abrupt reorientation of the polarization between two ferroelectric states that are close in energy and separated by a small barrier <ref:2603.27590#pg1>.
Lev: If we translate that to real hardware, does this mean the switching is fast enough for some kind of memory operation?
Mira: It suggests a deterministic first-order polarization process, which they compare to a first-order magnetization process in magnetic systems.
The paper's improvements: Kai: The authors propose a phenomenological framework for this, using an angular expansion U(theta) that captures the energy profile of the polarization. It includes terms like K1 two theta, K2 four theta, and so on.
Mira: This model is used to fit both the relative depths of those two energy minima and also the barrier separating them between those states.
Lev: From my side, I need to know how big that barrier is, because if it’s too high, you don't get switching at room temperature on real hardware; what are the energy scales we’re talking about?
Kai: They found that applying an external electric field can tilt this landscape until the stability switches between the two minima around one hundred forty kilo volts per centimeter at zero Kelvin <ref:2603.27590#pg1>.
Mira: That confirms that the double hysteresis loop is governed by this abrupt and reversible reorientation of the polarization vector, which is analogous to a first-order magnetization process.
Conclusion: Kai: To wrap up, this paper shows that we can use strain engineering to create materials where you get different switching modes depending on the direction of the field.
Mira: The implication is that this offers a unique platform combining energy storage, non-volatile memory operation, and electrocaloric effects of both signs simultaneously.
Lev: From an error correction perspective, having two stable states separated by a small barrier is actually quite useful because it defines the switching dynamics we need to model for real hardware.
Kai: So this work suggests that strain-engineered ferroelastic and antipolar materials are really promising candidates for exploring these first-order polarization processes.
Mira: It opens up a new playground for discovering behavior beyond conventional antiferroelectrics, especially when you consider the predicted opposite signs of the electrocaloric temperature change depending on field orientation.
Lev: That duality between ferroelectric and antiferroelectric-like responses based on field orientation is a very useful concept to design multifunctional systems.
Kai: So we’re looking at this paper, "First-order polarization process as an alternative to antiferroelectricity", which suggests that strain engineering can tune the system into a regime with double hysteresis loops without needing the standard antipolar ground state.
Louis Bastogne, *, Lukas Korosec, *, Evgenios Stylianidis, *, Daniel G. Porter, Gareth Nisbet, *
Theoretical Materials Physics, Université de Liège Department of Quantum Matter Physics, University of Geneva London Centre for Nanotechnology and Department of Physics and Astronomy University College London Solid State Physics Institute TU Wien Diamond Light Source
cond-mat.mtrl-sci
Submitted: 2026-03-29
Updated: 2026-10-29
Code: https://github.com/DiamondLightSource/i16_msmapper
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 79/100
The gist: First-order polarization process as an alternative to antiferroelectricity shows that appealing double-hysteresis P-E loops can appear from a field-induced abrupt rotation of polarization in…
Key concepts
- First-order polarization process
- This is a type of switching mechanism where polarization doesn't change smoothly but abruptly jumps between two stable states. It is analogous to a first-order magnetization process in magnets, meaning it involves overcoming an energy barrier to switch between distinct configurations, leading to sharp hysteresis loops.
- Epitaxial strain
- This is the mechanical stress induced on a thin film material when it is grown on a substrate with different lattice parameters. By choosing specific substrates like NdGaO3 (110)o or (001)o, researchers can precisely control the internal strain of CaTiO3 films, which in turn modifies its energy landscape and stabilizes specific polarization states.
- Double hysteresis loops
- These are characteristic curves showing two distinct switching behaviors when an electric field is applied. In this context, they arise from the abrupt reorientation of polarization between two closely spaced ferroelectric states separated by a small energy barrier, which is a signature of the first-order switching process.
Terminology
Summary
First-order polarization process as an alternative to antiferroelectricity shows that appealing double-hysteresis P-E loops can appear from a field-induced abrupt rotation of polarization in non-polar CaTiO3 films
Context and Motivation
Antiferroelectrics generate significant interest since their polarization versus electric field (PE) curves show typical double-hysteresis loops appealing for various applications Unfortunately, antiferroelectrics are rare. The paper investigates whether a similar mechanism could exist in polar materials, leading to a first-order polarization process capable of generating double hysteresis loops without invoking a conventional antipolar ground state. p. 1 In the case where
this field-induced transition is reversible under
removal of the magnetic field, it can give rise to
double-hysteresis M-H loops. One may wonder whether, by analogy, a similar mechanism could exist in polar materials, leading to a first-order polarization process capable of generating double hysteresis
loops without invoking a conventional antipolar ground state. To explore this idea, we focus on CaTiO3, the
archetypal perovskite oxide, which exhibits a
non-polar orthorhombic P bnm ground state involving oxygen octahedra rotations but is sometimes seen as an incipient ferroelectric [23, 24]. Lying close to a ferroelectric instability, it
was shown to develop a spontaneous polarization under tensile strain or chemical substitution [25–30].Where the paper says it
Where the paper says it
Experimental Setup and Strain Effects
The study utilizes epitaxial thin films of non-polar CaTiO3 grown on differently oriented NdGaO3 substrates using a combination of theoretical and experimental techniques. The authors examine how tensile epitaxial strain can induce and stabilize different polar states in CaTiO3. p. 1 Considering epitaxial CaTiO3
films grown on differently oriented NdGaO3
orthorhombic substrates, we show, through a
combination of experimental and theoretical approaches, that epitaxial strain can tune the
energy and polarization landscapes, stabilize
different oxygen octahedral rotation patterns,
and ultimately enable deterministic first-order
polarization process with associated appealing
double hysteresis loops. RESULTS
We first examine how tensile epitaxial strain
can induce and stabilize different polar states
in CaTiO3. This testifies to a complex and tunable energy
landscape and provides an appealing opportunity to explore the possibility of achieving a
first-order polarization process. Epitaxial mechanical constraints
To investigate this phenomenon, we have
deposited high-quality CaTiO3 thin films on
(110)o and (001)o-oriented NdGaO3 substrates,
with high structural quality and atomically flat
surfaces (see SI-S2)
The NdGaO3 substrate adopts the same orthorhombic P bnm phase as bulk CaTiO3. For both orientations,
X-ray diffraction (XRD) confirms that the films
are fully strained, twin-free, and inherit the substrate octahedral rotation pattern across the interface (see SI-S3). Importantly, although they arise from the
same material, the mechanical constraints imposed on the CaTiO3 film by the NdGaO3 substrate are significantly different for the two substrate orientations (see SI, Table S2) and should
therefore affect the energy landscape differently.Where the paper says it
Where the paper says it
Where the paper says it
Key Findings on Polarization Switching
The research reveals that epitaxial strain can tune the energy and polarization landscapes, stabilizing different oxygen octahedral rotation patterns and enabling a deterministic first-order polarization process with associated appealing double hysteresis loops. For CaTiO3 films strained on NdGaO3 (110)o, a well-defined ferroelectric hysteresis loop is obtained when the field is applied along apc, whereas only a linear dielectric response is observed along bpc. p. 3 2(a) shows that a well-defined ferroelectric
hysteresis loop is obtained at 4. 2 K when the
field is applied along ⃗apc, whereas only a linear
dielectric response is observed along ⃗bpc. In what follows, in order to avoid any
ambiguity between both Pm phases, we refer
to them respectively as P21nm and P b21m, according to the direction of the main component
of their polarization without changing the generality of the following results. We next examine the electrical and dielectric measurements which reveal a marked
anisotropy: while a conventional ferroelectric
loop is observed when the field is applied along
⃗ao, a double hysteresis loop emerges when
the field is applied along ⃗bo (Fig. 7
polarization between two ferroelectric states of
distinct orientations that are close in energy and
separated by a small energy barrier. By analogy with magnetism, we propose to call that a
first-order polarization process.Where the paper says it
Where the paper says it
Where the paper says it
Theoretical Framework and Analogy
The authors propose that this directional duality arises from a first-order polarization process, an analogue of first-order magnetization process in magnetic systems. They describe the polarization energy landscape using a minimal angular expansion U(θ) = U0+K1 sin2 θ+K2 sin4 θ+K3 sin6 θ (1). p. 6 To capture the NEB energy profile
U along the minimum-energy path (Fig. 3 (a))
in a compact form, we consider that it only depends on the orientation θ – respect to ⃗ao – of
the spontaneous polarization of amplitude Ps,
assumed to be constant, and we fit U(θ) using
a minimal angular expansion:
U(θ) = U0+K1 sin2
θ+K2 sin4
θ+K3 sin6
θ (1)
This fit (Table S3) perfectly reproduces both
the relative depth of the two minima and the
barrier separating them (Fig. To capture the NEB energy profile
U along the minimum-energy path (Fig. 3 (a))
in a compact form, we consider that it only depends on the orientation θ – respect to ⃗ao – of
the spontaneous polarization of amplitude Ps,
assumed to be constant, and we fit U(θ) using
a minimal angular expansion:
U(θ) = U0+K1 sin2
θ+K2 sin4
θ+K3 sin6
θ (1)
This fit (Table S3) perfectly reproduces both
the relative depth of the two minima and the
barrier separating them (Fig. Within this phenomenological framework, the
application of an external electric field progressively tilts the energy landscape, leading to a
field-induced interchange of stability between
the two competing minima at E ∼ 140 kV/cm
at 0 K, in good agreement with the coercive
field obtained from second-principles molecular
dynamics simulations. This agreement confirms
that, at the phenomenological level, the double
hysteresis loop is governed by an abrupt and reversible reorientation of the polarization vector
that is directly analogous to the magnetization
reversal characteristic of a first-order magnetization process.Where the paper says it
Where the paper says it
Where the paper says it
Implications and Future Directions
The results establish first-order polarization processes as a general framework for designing multifunctional ferroelectrics. The coexistence of distinct switching modes, such as ferroelectric-like and antiferroelectric-like responses depending on the orientation of the applied field, offers a unique platform combining energy storage, non-volatile memory operation, and electrocaloric effects of both signs. p. 8 A key outcome is that ferroelectric
and antiferroelectric-like functionalities coexist
within the same material, under identical
structural and thermodynamic conditions, and
can be selectively accessed by the orientation
of the applied electric field. This duality offers
a unique platform combining energy storage,
non-volatile memory operation, electrocaloric
effects of both signs, and neuromorphic functionalities within a single ferroic system. This duality offers
a unique platform combining energy storage,
non-volatile memory operation, electrocaloric
effects of both signs, and neuromorphic functionalities within a single ferroic system. More broadly, our results establish first-order
polarization processes as a general framework
for designing multifunctional ferroelectrics,
and suggest that strain-engineered ferroelastic
and antipolar materials constitute a fertile
playground for discovering analogous behavior
beyond conventional antiferroelectrics. Indeed, in magnetic systems, first-order magnetization process is often accompanied by magnetocaloric effects [45, 46] of opposite sign depending of the orientation of the applied field;
by analogy, first-order polarization process is
expected to couple to similar electrocaloric behaviors. Consistent with this picture, we predict from our second-principles model opposite
signs of the electrocaloric temperature change
depending on field orientation: ∆T < 0 when
the electric field is applied perpendicular to the
polarization and is in the same range of efficiency compared to conventional antiferroelectics (∆T /∆E
2 = −3×10-5 K cm2 kV-2) [47, 48].Where the paper says it
Where the paper says it
Where the paper says it
Online Methods
The methods section details various experimental techniques used, including thin film deposition using a home-built off-axis radio-frequency magnetron sputtering system, laboratory x-ray diffraction measurements performed on a Panalytical X’Pert Pro diffractometer, and electrical measurements such as capacitance–voltage measurements and dynamical hysteresis measurements (DHM). First-principles calculations were conducted using the ABINIT software package with a norm conserving pseudo-potential. p. 9 Exposures were taken using the
Pilatus3-100K area detector and these were remapped into reciprocal space using the msmapper program, part of the Dawn software package [53, 54]. First-principles
The first-principles calculation has been done
using the ABINIT software package [55] using the norm conserving [56] pseudo-potential
with the Wu-Cohen (WC) parametrization for
the exchange correlation potential [57] treating the following orbitals as valences state: Ca
3s, 3p, 4s, 3d, Ti 4s, 4p, 4d, 5s and O 2s, 2p. Our
constraint relaxation has been performed using
BFGS minimization algorithm by fixing two cell
parameters and optimizing the remaining (both
length and angle) one. Second-principles
The second-principles atomistic simulations
were performed using an effective interatomic potential built with the Multibinit
package [55], which implements the secondprinciples formalism introduced by Wojdeł et
al. Because
the electric field in the CaTiO3 layer lies almost
purely in-plane as argued in the previous paragraph, we can calculate its value simply by dividing the voltage by the electrode spacing. Synchrotron x-ray diffraction
Synchrotron x-ray diffraction measurements
were performed at the I16 beamline at Diamond Light Source [52], using a photon energy of 9. Finally, we numerically solve for the
dielectric constant of the thin film layer that
reproduces the measured capacitance for each
temperature or voltage. To extract the polarisation-electric field loops
of CaTiO3 thin films from the measured chargevoltage loops, we simulate the capacitance of
an NdGaO3 substrate as described in the previous paragraphs and subtract the corresponding
charge from the experimental data. 3 (a))
and look for the associated switching pathways. We first compute the minimum-energy
path connecting the two experimentally accessible polarization states (P21nm ground state
and P b21m metastable phase) using secondprinciples nudged elastic band (NEB) calculations (Fig. 2(d)), highlighting an abrupt jump associated with a first-order
transition. (c) Energy along the same minimum-energy path shown in panel (a), plotted as a function of
the rotation angle θ, and its evolution under electric field applied along ⃗bo, as fitted with Eq. To capture the NEB energy profile
U along the minimum-energy path (Fig. 3 (a))
in a compact form, we consider that it only depends on the orientation θ – respect to ⃗ao – of
the spontaneous polarization of amplitude Ps,
assumed to be constant, and we fit U(θ) using
a minimal angular expansion:
U(θ) = U0+K1 sin2
θ+K2 sin4
θ+K3 sin6
θ (1)
This fit (Table S3) perfectly reproduces both
the relative depth of the two minima and the
barrier separating them (Fig. Laboratory x-ray diffraction
High-resolution x-ray diffraction measurements were performed on a Panalytical X’Pert
Pro diffractometer with a Cu-Kα source. Atomic force microscopy
AFM topography measurements were performed using tapping mode on a Veeco Multimode IV AFM equipped with a SPECS Nanonis
4 controller using MikroMasch HQ:NSC15/Al
BS AFM tips with a force constant of
≈40 N m−1
.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Supplementary Information Details
The supplementary information provides detailed lattice parameters, strain calculations, and atomic force microscopy topography measurements for both (110)o and (001)o orientations. The space group determination confirms that the oxygen octahedral rotations and A-cation antipolar motion persist in the polar phase. p. 1 RESULTS
We first examine how tensile epitaxial strain
can induce and stabilize different polar states
in CaTiO3. In bulk, CaTiO3 adopts an orthorhombic P bnm phase which can be seen as
a small distortion of the reference cubic perovskite structure: it arises from the condensation of in-phase and anti-phase rotations of
the oxygen octahedra according to a pattern
(a
−a
−b
+) in Glazer notation [31], and further
includes secondary antipolar motions of the Cacations [32] in the plane perpendicular to the
in-phase rotation axis of the oxygen octahedra. 2
In magnetic systems, double-hysteresis loops
in magnetization versus magnetic field (M-H)
curves are not restricted to antiferromagnets
but can also appear as a feature of some ferromagnetic systems showing first-order magnetization process [18, 19]. In such systems, the
free energy shows two slightly inequivalent extrema associated with distinct orientations of
the magnetization. (b) Experimental dielectric constant versus temperature measurement
for the same film as panel (a), along ⃗apc and ⃗bpc, showing a characteristic peak at Tc,exp = 110 K signifying
the transition from Pm to P21/m phase. (c) Second-principles polarization versus electric field curve for
CaTiO3 strained on NdGaO3 (110)o along ⃗apc and ⃗bpc at Tsim = 40 K in Pm phase. In contrast,
the strain-engineered energy landscape considered here brings competing ferroelectric minima into close energetic proximity, enabling a
field-induced polarization reorientation at much
lower, experimentally accessible fields. As a result, polarization rotation manifests not as a
smooth response, but as a first-order switching
process, giving rise to double hysteresis loops. Indeed, in magnetic systems, first-order magnetization process is often accompanied by magnetocaloric effects [45, 46] of opposite sign depending of the orientation of the applied field;
by analogy, first-order polarization process is
expected to couple to similar electrocaloric behaviors. Consistent with this picture, we predict from our second-principles model opposite
signs of the electrocaloric temperature change
depending on field orientation: ∆T < 0 when
the electric field is applied perpendicular to the
polarization and is in the same range of efficiency compared to conventional antiferroelectics (∆T /∆E
2 = −3×10-5 K cm2 kV-2) [47, 48]. A key outcome is that ferroelectric
and antiferroelectric-like functionalities coexist
within the same material, under identical
structural and thermodynamic conditions, and
can be selectively accessed by the orientation
of the applied electric field. This duality offers
a unique platform combining energy storage,
non-volatile memory operation, electrocaloric
effects of both signs, and neuromorphic functionalities within a single ferroic system. Temperaturedependent dielectric measurements in Fig. 2(e)
provide an independent manifestation of this
duality and anisotropic response, revealing
markedly different dielectric signatures depending on the orientation of the applied electric
field. This duality offers
a unique platform combining energy storage,
non-volatile memory operation, electrocaloric
effects of both signs, and neuromorphic functionalities within a single ferroic system. More broadly, our results establish first-order
polarization processes as a general framework
for designing multifunctional ferroelectrics,
and suggest that strain-engineered ferroelastic
and antipolar materials constitute a fertile
playground for discovering analogous behavior
beyond conventional antiferroelectrics.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
System can predict material response under applied electric fields by identifying if a first-order polarization process is active, which enables double hysteresis P-E loops in ferroelectric materials instead of relying on conventional antiferroelectric ground states.
-
AI can design novel thin film structures using epitaxial strain to tune the energy landscape of materials like CaTiO3, stabilizing multiple ferroelectric states and engineering the barrier separation between them for controllable switching.
-
The improved system can classify material behavior into single or double hysteresis loops based on the orientation of the applied electric field, providing a predictive tool for device design in neuromorphic computing and energy storage.
-
The AI can predict electrocaloric effects with opposite signs depending on field orientation, allowing for the design of multifunctional systems that exhibit both ferroelectric and antiferroelectric-like functionalities simultaneously.
Abstract
Antiferroelectrics generate significant interest since their polarization versus electric field (P-E) curves show typical double-hysteresis loops appealing for various applications. Unfortunately, antiferroelectrics are rare. In magnetic compounds, magnetization versus magnetic field (M-H) curves can show analogous double hysteresis loops not only in antiferromagnets but also in systems exhibiting field-induced first-order reorientation of the magnetization through a so-called first-order magnetization process. Here, we show that appealing double-hysteresis P-E loops can also appear from an unprecedented first-order polarization process. Focusing on non-polar CaTiO3, which can be turned ferroelectric under tensile strain, we study epitaxial thin films on differently-oriented NdGaO3 substrates using a combination of theoretical and experimental techniques. We uncover that a certain configuration exhibits double-hysteresis P-E loops that we rationalize from a field-induced abrupt rotation of the polarization. Such a first-order polarization process establishes a promising alternative pathway to achieve double hysteresis P-E loops appealing for practical applications.
Related papers
- Structural contribution to light-induced gap suppression in Ta 2 NiSe 5
- Anatomy of fast current-induced skyrmion motion in synthetic antiferromagnets
- Magnetic control of an exciton-polariton condensate in a van der Waals magnet
- High-pressure elastic properties of GeO2 polymorphs up to 120 GPa
- Decoupling Strain-Rate Sensitivity and Deformation Length Scale Effects in Neutron-Irradiated Tungsten: A Coupled Nano-Indentation, HR-EBSD and Crystal Plasticity Study
- Magnetism modulation in Co3Sn2S2 by current-assisted domain wall motion