Physicists have discovered that ordinary ice exhibits flexoelectric properties, generating electricity when subjected to bending deformations. The scientists attributed this behavior to a ferroelectric phase transition in a near-surface layer just 15-20 nanometers thick. The physicists' work also pointed to the possible contribution of flexoelectricity to charge separation in thunderclouds. The results of the study were published in the journal Nature Physics.
Although physicists have already discovered 19 crystalline modifications of water ice, the properties of even the most common ordinary ice (also known as ice-Ih) are far from fully understood. For example, the question of its electromechanical properties remains open: researchers have repeatedly noted the connection between atmospheric electricity and ice crystals in clouds. However, the Bernal-Fowler rules prohibit ice-Ih from generating free charges under compression or tension, since the hydrogen atoms in such a crystal do not exhibit a long-range order structure. In other words, despite the hexagonal lattice built by oxygen atoms, this type of ice is a cluster of randomly oriented dipoles, which contradicts the essence of piezoelectricity. Therefore, the origin of electric charges must lie elsewhere.
Physicists from Spain, China, and the United States, led by Xin Wen of Xi'an Jiaotong University, have suggested that ice-Ih can generate charges through the flexoelectric effect—a phenomenon in which there is a relationship between the polarization of a material and its deformation gradient.
To test their hypothesis, the scientists fabricated capacitors from two gold-coated aluminum plates, freezing a layer of ultrapure water approximately two millimeters thick between them. The physicists first conducted piezoelectric measurements to confirm the non-piezoelectric nature of the ice samples. They then used a dynamic mechanical analyzer to create a three-point bending strain in the capacitors with a maximum stress of 0.006 gigapascals, fixing the edges of the capacitor and applying a force in the middle. This approach induced a flexoelectric response in the material, which the authors measured in a temperature range from 143 to 273 Kelvin.
At temperatures above 248 Kelvin, flexoelectricity significantly increased, which the physicists attributed to the transition of ice into quasi-liquid layers, characteristic of cases where the material is in a state just prior to melting. Such layers contain a large number of mobile charge-transferring ions. In the range of 203–248 Kelvin, the flexoelectric coefficient exhibited constant properties with a weighted average value of 1.14 ± 0.13 nanocoulombs per meter. This result was similar to that of dielectric ceramics, and the flexocoupling coefficient (flexoelectric coefficient divided by permittivity), equal to 1.29 ± 0.15 volts, fell within the range for intrinsic flexoelectricity in solids.
However, at temperatures below 203 Kelvin, the flexoelectric coefficient increased again, reaching a peak of 7.6 nanocoulombs per meter at 164.6 ± 1.7 Kelvin. The authors noted that such a temperature dependence had previously been observed only in ceramic materials with pronounced ferroelectric properties. Consequently, the physicists hypothesized that this flexoelectric maximum was caused by a ferroelectric phase transition confined to the near-surface region of the material. Further confirmation of this hypothesis was provided by the measured butterfly-shaped hysteresis loop, as well as the calculated Helmholtz free energy, which showed a shift in the Curie temperature to 164.6 Kelvin at a skin depth of 14.6 nanometers (the experimental estimate was 20.3 nanometers).
Furthermore, the authors emphasized that their numerical estimates explained the separation of electric charges in thunderclouds: due to the collision of grains with ice crystals, both materials deform and exchange flexoelectric charges—the grains become negatively charged, while the ice becomes positively charged. However, this model greatly simplifies all the intermediate processes and, according to the physicists, requires further clarification.
We wrote earlier about how metastable water turned into ice-VII and only then into ice-VI.
