Flexoelectricity is still a fairly obscure but –to a growing cohort of researchers- interesting property of some materials whereby they generate electricity upon bending. “Some materials” is a bit of an understatement: ALL materials except for metals can in theory be flexoelectric, and flexoelectricity has already been measured in ceramics, polymers, biomaterials such as bones, and semiconductors. The latter includes halide perovskites. Halide perovskites are a family of metal-organic semiconductors that is “hot” in solar energy research, having reached photovoltaic efficiency levels that took silicon many decades to reach. They also have their fair share of problems such as fast degradation, but that has not eclipsed their appeal. Moving on then.
A couple of years ago, I received an unusual request from Longlong Shu, someone whom I did not know at the time. A group of scientists led by him in Nanchang had measured the flexoelectricity of halide perovskite thin films, and the values seemed “giant”. Being a bit of an authority in flexoelectricity (which is not as impressive as it may sound: this is not yet a crowded field!) they wanted to know my opinion. My opinion was actually rather dismissive: I thought that thin films are no good for a first report of flexoelectricity in any material, because clamping to the substrate complicates disentangling the effect of strain gradient (flexoelectricity) from that of strain (piezoelectricity). Additionally, the flexoelectric analysis needed to incorporate the semiconductor nature of halides.
To his credit, instead of getting annoyed with me (if he did, he never told me!), Longlong took the criticisms on board and, some months later, wrote back with a new set of measurements, this time performed on large, high quality single crystals. It is also to his credit that, for the new experiments, he pulled together a large team of collaborators from different universities, and they all chipped in on the back of someone else’s hunch. Massimiliano Stengel was also brought on board to help with the theory. Science has these beautiful collaborative moments sometimes.
The results confirmed that the flexoelectricity of halides is large, so the result was now ready for reporting. However, being well known for their photovoltaic properties, it seemed silly not to try to combine them with flexoelectricity... The rest is history, and it is reported in our Nature Materials paper.
We found that shining light upon a halide perovskite while bending it enhances its flexoelectricity by several orders of magnitude. As a matter of fact, the effective flexoelectric coefficient of halide perovsites under illumination is the highest ever reported for any material. We also show that “photoflexoelectricity” can be explained by the effect of mechanical deformation on the electrode-semiconductor Schottky barriers, and it should therefore be a general property of ALL semiconductors. To illustrate this, we have measured it also in an oxide, SrTiO3, under UV light.
The bottom line is that photovoltaic materials –including halide perovskites- are also electromechanical and photoelectromechanical, so they can provide energy in more ways than one. And working with them is energizing.
Figure: schematic of a photoflexoelectric device consisting of a halide crystal (pictures of actual crystals below) with transparent electrodes, subject to the simultaneous input of mechanical bending and light. The output is a large photo-enhanced flexoelectricity, i.e. photoflexoelectricity.