Universal scaling of resolution with photon number in superresolution fluorescence microscopy

Abstract

Superresolution fluorescence microscopy techniques beat the diffraction limit, enabling ultra-high resolution imaging in biological physics and nanoscience. In all cases that have been studied experimentally, the resolution scales inversely with the square root of some parameter that measures the number of photons used. However, this ubiquitous limit arises from very distinct mechanisms in different approaches, raising the question of whether it is a fundamental limit that cannot be exceeded, or merely a coincidence of the techniques studied thus far. We demonstrate that, under very general assumptions that encompass essentially all fluorescence microscopy situations, the known resolution limit is indeed universal. Our model considers experiments that build up an image via any arbitrary sequence of steps compatible with our assumptions of (1) light that exhibits shot noise and (2) molecules that can be modeled with rate equations. A detailed examination of our assumptions shows that exceeding this resolution limit will require the use of quantum optical effects, pointing to an avenue for future innovation.