OSA's Digital Library

Spotlight on Optics

Spotlight on Optics


  • December 2012

Optics InfoBase > Spotlight on Optics > Raman probes based on optically-poled double-clad fiber and coupler

Raman probes based on optically-poled double-clad fiber and coupler

Published in Optics Express, Vol. 20 Issue 27, pp.28563-28572 (2012)
by Anna Chiara Brunetti, Walter Margulis, and Karsten Rottwitt

Source article Abstract | Full Text: XHTML | Full Text: PDF

Spotlight summary: “You really ought to get more fiber” is an increasingly common refrain told to doctors by many biomedical imaging research groups. Whereas fibers are still often considered primarily as pipes for light, many of the elements needed for a microscope, such as lenses, confocal apertures, beam splitters, dichroic mirrors, and light sources, can all be made from a fiber platform. Numerous groups have reported on single and multi-photon fluorescence microscopes, CARS microscopes, SHG and THG microscopes, and optical coherence tomography (OCT), all based largely on fiber elements. The usual arguments in favor of fiber devices apply here: they enable remote delivery, alignment free operation, are robust against mechanical shocks, can be used in a wide range of temperature, humidity, and chemical environments, etc. In applications where one is trying to detect weak signals, fibers also present a drawback in that nearly the entire optical path is in glass, so autofluorescence and Raman scattering can develop into a large background which needs to be mitigated. Double clad fibers are well-suited for this purpose, as they spatially isolate the excitation and collection channels.

Building upon this work, researchers at the Technical University of Denmark and Acreo AB have proposed an all-fiber Raman probe. They employ a double-clad fiber design with a coupler, which enables them to tackle the problem of how, with a single fiber probe, to simultaneously obtain a high intensity excitation beam and a high collection efficiency of the Raman signal, and be able to spatially separate the signal beam path for spectral analysis. Like multiphoton fluorescence or THG, Raman scattering is a nonlinear process with low efficiency in most materials, so one would want to deliver a high intensity, small area mode to the sample to maximize the Raman signal. However, whereas the excitation beam is coherent and directional, the spontaneous Raman signal is emitted in all directions, so to maximize the total signal one needs to collect over a large area. Double-clad fibers are provide this functionality, where the excitation pump is delivered through the single mode core and the Raman signal is collected through the large, multimode inner cladding. As compared to collecting through the single mode core, in this geometry much of the background generated in the fiber by the pump is removed from the signal collection arm. The backwards propagating Raman signal is then tapped off to a second fiber via a very simple coupler, made by stripping off the outer cladding of two fiber sections, twisting them around each other, and index matching the surrounding environment with oil. The coupler acts only on the cladding modes, while leaving the core unaffected. One can then send light from the output arm directly to a spectrometer without the need for splitters or dichroic filters, and without any further background signal being generated by backwards propagating pump light, since only cladding light is coupled back.

The one added wrinkle for making an all-fiber Raman probe is that Raman scattering efficiency scales with w4 and most spectrometers are based on silicon CCDs. Ideally one would want to use visible light for the excitation, whereas most fiber lasers operate at 1 ?m or higher. To solve this problem, the authors propose poling their fiber, so that they can generate green light via SHG from a pump at 1064 nm, which is a wavelength available in commercial fiber lasers. In the experimental work shown here, they stop one step shy of an all-fiber device and use a YAG laser for their pump, and they use optical poling, so their conversion efficiency is less than 0.2%. Nonetheless, they generate enough green light to observe a Raman signal from a fluid reservoir of dimethyl sulfoxide, and successfully measure not just the strongest Raman lines from their sample, but most of the weaker ones as well. The results could of course be improved with the use of a fiber laser, and with thermal poling, which has a greater SHG efficiency. This work shows the potential for impact in chemical and biological sensing for both clinical and industrial settings, and highlights an important area for which there is a need for novel fiber-based sources at short wavelengths.

--Paul Steinvurzel

Technical Division: Light–Matter Interactions
ToC Category: Spectroscopy
OCIS Codes: (060.2340) Fiber optics and optical communications : Fiber optics components
(060.2370) Fiber optics and optical communications : Fiber optics sensors
(190.2620) Nonlinear optics : Harmonic generation and mixing
(300.6450) Spectroscopy : Spectroscopy, Raman

Posted on December 20, 2012

Add Comment
You must log in to add comments.

Browse Journals / Lookup Meetings

Browse by Journal and Year


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Previous Spotlights