Perovskites may work because their messy crystals build their own electric fields

Lead-halide perovskites have been a maddening contradiction for photovoltaic researchers. The materials are cheap to fabricate and deliver efficiencies that rival silicon, yet their crystal structure is a mess—full of inhomogeneities that should, by all conventional semiconductor logic, kill performance. The elegance of silicon lies in its near-perfect crystalline order. Perovskites seemed to succeed in spite of their chaos.

A team at the Institute of Science and Technology Austria (ISTA) has now shown that the chaos is the point. Focusing on methylammonium lead bromide (MAPbBr₃) in its cubic phase to eliminate low-symmetry effects, the researchers used nonlinear optical excitation to generate charge carriers. They observed a reproducible zero-bias photocurrent—an unambiguous signal that internal electric fields, not an external voltage, were doing the work of separating electrons and holes.

This is where the story gets physically counterintuitive. Previous efforts pinned perovskite performance on ferroelectricity, a property that creates a permanent electric polarization. But cubic-phase crystals lack the symmetry to support classic ferroelectricity, making that explanation a non-starter. The ISTA group, publishing in Nature Communications, identified a different phenomenon: flexoelectric polarization at strain-induced domain walls.

The source material also shows that flexoelectricity couples strain gradients to electric polarization. In a material riddled with mesoscopic structural inhomogeneities, those gradients are everywhere. The domain walls between differently oriented crystal regions act like microscopic power plants, generating the fields that shepherd charge carriers across long distances. The material’s apparent disorder is actually a dense network of built-in charge-separation junctions.

“There were many conjectures about the origin of charge separation in perovskites,” the ISTA researchers note, referencing the field’s long fixation on tetragonal-phase explanations. “This naive explanation completely failed to account for the fact that certain cubic perovskites like MAPbBr₃ also exhibit comparable performance.” The new work provides a general mechanism that works across phases, making it far more useful as a design principle.

The practical implication is a shift in how engineers think about defects. Instead of trying to eliminate every structural inhomogeneity—a costly and often futile exercise in solution-processed materials—the goal becomes controlling the density, orientation, and strain state of domain walls to tune the internal field landscape. It turns a fabrication liability into a performance lever.