Development of a standing-wave fluorescence microscope with high nodal plane flatness

This article reports about the development and application of a standing-wave fluorescence microscope (SWFM) with high nodal plane flatness. As opposed to the uniform excitation field in conventional fluorescence microscopes as SWFM uses a standing-wave pattern of laser light. This pattern consists of alternating planar nodes and antinodes. By shifting it along the axis of the microscope a set of different fluorescent structures can be distinguished. Their axial separation may just be a fraction of a wavelength so that an SWFM allows distinction of structures which would appear axially unresolved in a conventional or confocal fluorescence microscope. An SWFM is most powerful when the axial extension of the specimen is comparable to the wavelength of light. Otherwise several planes are illuminated simultaneously and their separation is hardly feasible. The objective of this work was to develop a new SWFM instrument which allows standing-wave fluorescence microscopy with controlled high nodal plane flatness. Earlier SWFMs did not allow such a controlled flatness, which impeded image interpretation and processing. Another design goal was to build a compact, easy-to-use instrument to foster a more widespread use of this new technique. The instrument developed uses a green-emitting helium-neon laser as the light source, a piezoelectric movable beamsplitter to generate two mutually coherent laser beams of variable relative phase and two single-mode fibres to transmit these beams to the microscope. Each beam is passed on to the specimen by a planoconvex lens and an objective lens. The only reflective surface whose residual curvature could cause wavefront deformations is a dichroic beamsplitter. Nodal plane flatness is controlled via interference fringes by a procedure which is similar to the interferometric test of optical surfaces. The performance of the instrument was tested using dried and fluorescently labelled cardiac muscle cells of rats. The SWFM enabled the distinction of layers of stress fibres whose axial separation was just a fraction of a wavelength. Layers at such a small distance would lie completely within the depth-of-field of a conventional or confocal fluorescence microscope and could therefore not be distinguished by these two methods. To obtain further information from the SWFM images it would be advantageous to use the images as input-data to image processing algorithms such as conceived by Krishnamurthi et al. (Proc. SPIE, 2655, 1996, 18-25). To minimize specimen-caused nodal plane distortion, the specimen should be embedded in a medium of closely matched refractive index. The proper match of the refractive indices could be checked via the method presented here for the measurement of nodal plane flatness. For this purpose the fluorescent layer of latex beads would simply be replaced by the specimen. A combination of the developed SWFM with a specimen embedded in a medium of matched refractive index and further image processing would exploit the full potential of standing-wave fluorescence microscopy.