The required refractive index for glycerol immersion objectives is slightly higher than for silicone oil immersion objectives, which have recently been used to enable refractive index matching for live cell confocal spinning disc microscopy 19. No stepwise procedure is required, it is very inert, and many mounting media for fluorescence microscopy contain glycerol anyway.

Their advantage is that embedding biological samples in glycerol-containing media is easier feasible than in TDE. With NAвЂ‰=вЂ‰1.35, their numerical aperture is higher than those of water-immersion objectives, but not as high as with oil immersion. Glycerol-immersion objectives may be another alternative. But this method has the disadvantage that for an exact refractive index matching, the aqueous medium has to be replaced stepwise with TDE to avoid artefacts due to osmosis. For oil-immersion objectives with their high NA of 1.4 or more, a refractive index match may be achieved by embedding the sample in 2,2вЂ™-thiodiethanol (TDE) instead of aqueous buffers 17, 18. But due to the lower refractive index of water, water-immersion objectives have a lower NA (typically 1.2), which increases the size of the PSF and reduces the number of collected photons, effects that are disadvantageous for the localisation precision. For example, when water-immersion objectives were employed for 3D STORM with astigmatic detection 16, they yielded better results than oil-immersion objectives. However, a more straightforward approach is to prevent the refractive index mismatch and the associated degrading effects by matching the refractive indices of immersion medium, cover glass and sample. This was successfully implemented for astigmatic detection 14 and interferometric detection 15.

One approach to circumvent the degrading effects of refractive index matching is to include a wavefront correction mechanism into the system. Furthermore, wavefront deformations will distort the interference pattern when interferometric detection is used. For single molecule localisation, this is unfavourable as both effects degrade the localisation precision. This leads to a refractive index mismatch between cover glass and watery sample, which causes both aberrations and loss of signal 13. However, objective lenses with high NA which are required for good optical resolution mostly use oil as immersion medium due to the resulting high NA, whereas biological samples usually have to be imaged in aqueous buffers. The localisation precision surpasses the resolution limit, but nevertheless depends on the size of the microscopeвЂ™s point spread function (PSF) and therefore its numerical aperture (NA). The localisation precisions achievable with these methods were investigated theoretically 11 and a direct experimental comparison of biplane and astigmatic detection was performed for a microscope with a single water-immersion objective 12, but not for two-objective schemes such as interferometric detection. They all involve some means to make the single molecule images more z-dependent, for example by inserting a cylindrical lens into the detection beam path which adds astigmatic distortion to the images 6, imaging two distinct axial planes simultaneously 7 or by capturing the fluorescent light with two objectives and bringing it to interference 8, 9, 10, which causes z-dependent intensity variations. Originally, this approach was limited to 2D imaging, but subsequently various techniques have been developed to also obtain the axial ( z) coordinate and reconstruct 3D images. Spatial separation of the fluorescent emitters allows to obtain the dye molecule coordinates, and therefore to resolve the stained structures, with a precision at least ten-fold better than the optical resolution.

Single molecule localisation microscopy techniques as stochastic optical reconstruction microscopy (STORM) 1, direct stochastic optical reconstruction microscopy (dSTORM) 2, photoactivated localisation microscopy (PALM) 3, fluorescence photoactivation localisation microscopy (FPALM) 4 and ground state depletion followed by individual molecule return (GSDIM) 5 elegantly circumvent the diffraction limit of optical microscopy by imaging single fluorescent molecules which stochastically switch between a fluorescent and a non-fluorescent state.

Immersive detection of npc