It does not "break" the diffraction limit. The title is inaccurate. The actual paper doesn't say they've broken the diffraction limit. They only achieve a subwa

It does not "break" the diffraction limit. The title is inaccurate. The actual paper doesn't say they've broken the diffraction limit. They only achieve a subwavelength focal point. I'll make an attempt to mansplain this... The actual diffraction limit depends on wavelength (lamda) and the numerical aperture (NA) of the lens. Diffraction limit is ~ lamda/(2*NA). In traditional objective lenses used in confocal scanning microscopes, the diffraction limit has been ~0.5 microns. If you increase NA, it also means decreasing working distance-- to the point where the focal point meets the surface of the lens. They've managed to increase NA with their zone plate, but since their "lens" thickness is "200 nm", they still have a useable working distance.