Flexoelectricity is a rather obscure effect whereby a dielectric can become electrically polarized when bent and, conversely, can bend when polarized -- sort of like the piezoelectric effect, but different.
By training I'm a physicist and I like to stay in touch with the field. This is partly because advances in physics continue to fascinate me, partly because it's a change from the daily grind, and partly because, sometimes, I actually find something of relevance to EDA or design. Flexoelectricity, which is not the latest PG&E rate plan, is a bit of all three.
Flexoelectricity is a rather obscure effect whereby a dielectric can become electrically polarized when bent and, conversely, can bend when polarized. Wait -- isn't that the piezoelectric effect? Well sort of, but the flexoelectric effect is different. I'll explain and, in order to make this explanation simpler, I will abbreviate the flexoelectric effect to FEE and the piezoelectric effect to PEE (sorry, but it's the obvious acronym).
First, PEE is restricted to crystals with a limited set of symmetries where FEE is not (the effect has also been observed in liquid crystals, polymer films and bio-membranes). Second, coupling in FEE is between polarization and strain gradient, rather than homogenous strain (the case for PEE). And third, PEE is usually much larger than FEE on macroscopic scales, which -- until recently -- relegated FEE to being little more than a curiosity. What makes FEE interesting now is that it grows significantly at the strain gradients seen at nano-scales, so much so that it can become larger than PEE.
The theory behind FEE is a little more complex than for PEE. In PEE, you start with balanced electric dipole moments in a lattice under no stress. When you apply a flat stress, dipoles in the lattice move out of balance (they are squeezed inward in one direction and perhaps outward in the other directions); hence you get an electric polarization. In FEE you apply a bending stress (hence a strain gradient). One way to visualize the impact is to think of a body-centric (or face-centric) crystal structure (although FEE is not restricted to these cases). As the outer ions bend downward, the central ion is squeezed upward, resulting again in a dipole imbalance and hence polarization. (For visualization on other crystal structures, you will have to find a better source than me.)
All of this is good to know because this effect can be exploited in MEMS devices (e.g., accelerometers) and in nano-generators -- particularly energy-harvesting devices. And, equally important, it opens up a wider range of materials than can be used for PEE. Energy-harvesting is a hot topic in the brave new era of the Internet of Things (IoT), so expect to see growing discussion around this area.
FEE is also of great interest in understanding electro-mechanical effects in soft bio-materials, for example in understanding how hairs in the outer ear convert sound to electrical impulses and -- conversely -- how polarization can cause membranes to bend. FEE is also potentially interesting in the design of medical transducers (assuming it can deliver on higher efficiency than PEE).
For the present time, you probably shouldn't be planning your next MEMS or energy-harvesting design based on FEE. All work on this topic still seems to be in the lab (e.g., NCSU and the Department of Condensed Matter Physics in Geneva have publications in this area), and observed behavior has yet to match expectations. That said, some PEE nano-generators are now suspected to be at least in part dependent on FEE. This sort of feels like special pleading by the FEE apologists, but maybe time will prove them right.