Piezoresponse force microscopy (PFM) applies voltage through a nanoscale tip and measures the resulting deformation. When crystallized Sb₂S₃ shows clear hysteretic loops under PFM, the natural conclusion is ferroelectricity — a switchable electric polarization that deforms the crystal when you push it.
The signal is real. The hysteresis is real. The spatial contrast between crystalline and amorphous regions is vivid and reproducible. But Jungk et al. (arXiv:2603.13221) show that the response originates primarily from non-piezoelectric contributions: electrochemical strain, charge injection, and electrostatic interactions between the tip and the heterogeneous surface.
The crystalline phase isn't deforming because polarization is switching. It's deforming because it conducts differently from the glass around it. The measurement picks up the electrical difference between phases and renders it as a mechanical signal that looks exactly like ferroelectricity.
This matters because Sb₂S₃ is a phase change material — the same class that powers rewritable optical storage and emerging non-volatile memory. If it were genuinely ferroelectric, that would add a switchable functionality worth billions in device applications. The phantom response doesn't just mislead a paper; it misleads an industry.
The deeper problem: PFM is designed to detect piezoelectric coupling. When you point a piezoelectric detector at a material with complex electrical heterogeneity, it faithfully reports something. The instrument doesn't distinguish between a genuine coupling and an artifact that produces the same output. The measurement technique selects for the answer it was built to find.
Phase change materials are electrically heterogeneous by definition — that's their function. Pointing PFM at them is like testing for earthquakes with a seismograph placed on a vibrating floor. You'll always get a signal. The question is whether the signal means what you think it means.