Reconfigurable ferroelectric chiral nanostructures enable fast-switchable optical spatial differentiation.

Analog spatial differentiation is an emerging computational paradigm. By virtue of high speed and low-power consumption, optical method plays an important role in data compression, microscopy and computer vision. However, most developed optical differentiators are static and lack the reconfigurability of differentiation functions. Herein, we propose a reconfigurable space-variant ferroelectric chiral nanostructure to dynamically control the optical differentiation. Via switching the polarity of external electric field, 1st-order/2nd-order spatial differentiation or bright-field imaging can be actively selected with an ultra-short response time down to 62 μs. Edges of biological cells, as well as intensity objects, can be well identified, while their direct imaging is also achievable synchronously. Such fast-switchable differentiator shows excellent reliability and reversibility for over 1.8 million cycles and over 200 days. This work advances the ingenious building of ferroelectric nanostructures, and offers an important glimpse into their potential in neuromorphic photonics, biomedical microscopy and artificial intelligence.