In contrast to single-mode optical fibers that are massively used in telecommunications, multimode fibers deliver light along a series of concurrent optical paths known as optical modes. Employing a number of such modes in order to parallelize the delivery of information is, however, very difficult. Although the first attempts to do so date back to the early 1980s, this so-called mode-division multiplexing is becoming an increasingly important research effort in order to cope with ever growing demands on Internet traffic. The difficulty lies in unpredictable coupling of power between such modes as they propagate along hundreds of kilometers long waveguides exposed to unstable conditions. This typically limits communication to only a very few modes.
However, in parallel with applications in telecommunications, we are witnessing another steeply growing research effort with a great promise for imaging. By employing these newly developed methods, we can unscramble highly disordered images that have been sent down multimode fibers, thus making for the narrowest possible image-guiding systems. Recently, numerous studies have demonstrated direct imaging of microscopic objects or even living specimens via channels of only few tens of micrometers wide. This was possible due to fast progression of holographic methods providing control of light propagation within turbid media. While alternative schemes have been identified, a typical approach uses a spatial light modulator, first for a characterization of the way in which multimode fibers distort the optical modes and then using this information to pre-shape the wavefront of a laser beam, before it is coupled into the fiber. This way, a series of diffraction-limited foci behind the fiber are generated one at a time, which efficiently raster-scan a given object while collecting either retro-reflected light or a fluorescence emission signal.
Due to high modal and material dispersion, the related approaches based on fluorescence have been performed with highly coherent CW laser sources in a single-photon regime. The new contribution by the group from EPFL shows that the spatial shaping of the wavefront can also be achieved with lasers operating in the pulsed regime. By time-gating of the reference signal, the characterization of the fiber has been extended into the temporal domain, which allows the researchers to concentrate the laser power in both space and time. Moreover, this study shows how one can shape the temporal profile of the pulses; for example splitting the power of a single pulse between two consecutive pulses.
Although there are still numerous challenges, these advances significantly broaden the spectrum of applications this young and promising field can offer. Currently, the main limitation here remains the lack of flexible operation: any bending or looping of the fiber results in changes to the way optical modes are delivered, thus rendering imaging heavily impaired. Nevertheless, imaging in stationary geometries is already very advantageous given that with this technique one can acquire imaging through up to two orders of magnitude smaller area when compared to GRIN lenses, fiber bundles, etc. The method is very likely to find applications in numerous areas across life sciences where minimally invasive in-vivo imaging is highly desirable.
You must log in to add comments.