Atomistic origin and strain control of the finite-temperature dielectric response in BaTiO3

Inside the tiny, invisible world of crystals, there exists a special kind of material that acts like a sponge for electricity. These are ferroelectrics, substances that can hold an electric charge and release it on command, making them the heart of everything from the capacitors in your phone to the sensors in medical imaging devices. For decades, scientists have known that these materials change their ability to store electricity as they heat up or cool down, but the exact reason why has remained a mystery hidden deep within their atomic structure. The question has always been: what is actually happening to the individual atoms inside the crystal when the temperature changes? Is the material getting softer, or are the tiny internal magnets simply turning in different directions?

A team of researchers at the University of Tokyo has finally looked inside this atomic world to find the answer, focusing on a classic material called barium titanate. Using powerful computer simulations that act like a high-speed movie of atoms in motion, they discovered that the material's ability to store electricity does not depend on how far the atoms move or how strongly they are tilted. Instead, the key lies in how easily the atoms can be coaxed to turn their direction when an electric field is applied. The study reveals that as the material gets hotter, the atoms do not shrink or change their average angle; rather, they simply become much harder to rotate. This finding offers a new, clear picture of how these materials work, moving the explanation from abstract waves of vibration to the concrete, physical turning of atomic groups.

Barium titanate is a well-known crystal that has been studied for over eighty years. It is famous for undergoing a series of shape changes as it is heated, shifting from one geometric arrangement to another. In its tetragonal phase, which exists at temperatures just below the point where it loses its special electric properties, the material is highly sensitive to heat. As the temperature rises, its ability to store electrical charge drops dramatically. Scientists have long tried to explain this drop using theories about "soft modes," which describe the collective vibration of atoms. However, these theories describe the material as a whole and do not show what a single group of atoms is doing when an electric field pushes on it. The researchers wanted to see the real-space behavior of the atoms to understand the true origin of this temperature dependence.

To solve this, the team used a sophisticated computer model based on machine learning. This model, trained on detailed quantum mechanical calculations, allowed them to simulate the movement of thousands of atoms over time while applying electric fields. They watched how the titanium and oxygen atoms, which form the core of the material's electric behavior, responded to heat and to the pull of an electric field. They defined a specific way to measure the local structure: the distance between a titanium atom and the center of the oxygen cage surrounding it, and the angle this distance makes with the main direction of the crystal's electric field.

The researchers first looked at what happens as the material is heated. They found that the size of the atomic shift and the average angle of the atoms remained surprisingly steady, even as the material's ability to store electricity plummeted. The atoms did not move closer together or tilt significantly more or less on average. The change was not in their position or their average tilt, but in their willingness to move when pushed. When an electric field was applied, the atoms at lower temperatures would shift their orientation to align with the field, creating a strong electrical response. As the temperature rose, this shifting became much more difficult. The atoms stayed in roughly the same place and kept the same average tilt, but they stopped turning as freely. The researchers quantified this by measuring how much the distribution of atomic angles changed when the electric field was reversed. This "orientational response" dropped sharply as the temperature increased, matching the drop in the material's ability to store electricity perfectly.

The team then tested if this rule held true when they changed the material in a different way. Instead of heating it, they squeezed and stretched the crystal from the sides, a process known as applying strain. They found that stretching the crystal made it much better at storing electricity, while squeezing it made it worse. In this case, the atoms did change their average tilt angle; they leaned more or less depending on the stretch. However, the orientational response still tracked the electrical performance perfectly. Whether the change was caused by heat or by mechanical strain, the material's ability to store electricity was directly linked to how easily the local atomic groups could be rotated by an electric field.

This discovery provides a unified explanation for two different ways of controlling the material. It shows that the macroscopic property of dielectric response is not about the atoms getting bigger or smaller, nor is it solely about their average tilt. It is about the freedom of the atoms to rotate. The study suggests that the "softness" of the material is actually an angular softness. The atoms are like tiny compass needles that are always present, but their ability to swing toward a new direction is what determines the material's electrical strength. When the material is hot, the thermal energy makes it hard for the electric field to turn these needles. When the material is strained, the physical shape of the crystal changes how easily those needles can turn.

The researchers validated their findings by comparing their simulation results with known experimental data, ensuring their model was accurate. They also checked that their results were not just an artifact of the way they constrained the computer simulation, confirming that the relationship between the atomic turning and the electrical response is a fundamental property of the material. By connecting the large-scale electrical behavior to the specific, local motion of atoms, this work bridges the gap between abstract theories of vibration and the tangible reality of atomic movement. It suggests that to design better materials for future electronics, scientists should focus not just on how much atoms move, but on how freely they can turn.