A proposed optoelectronic apparatus would perform the combined functions of a confocal microscope and a Raman spectrometer. It would be used to acquire Raman-spectral-image and/or monochromatic-image data from mineral and/or biological specimens with high three-dimensional spatial resolution. The Raman-spectral-image data could be used to identify materials present at various locations within the specimens.

The Confocal Single-Mode-Fiber-Optic Raman Microspectrometer would incorporate several improvements over prior Raman probes, including scanning Raman spectrometers that contain multimode optical fibers.

By virtue of the confocal-microscope aspect of its design and unlike prior Raman probes, the proposed apparatus would offer sufficient spatial resolution for imaging of microscopic objects and sufficient depth discrimination to enable sectioning; that is, the apparatus could be used to construct the equivalent of three-dimensional images from confocal-microscope scans in three dimensions. The spectrometer portion of the apparatus would be compact, relative to prior Raman spectrometers of equivalent spectral resolution. The design of the apparatus would also implement a unique solution to the problem of discriminating between Raman-scattered light and laser light used to excite Raman scattering - a difficult problem in that the Raman spectral shift can be small.

The figure depicts one version of the proposed apparatus. Light from a laser or a laser diode would be launched into a single-mode optical fiber configured as an input port (port 1) of a nominal 50/50 fiber-optic directional coupler. One of the output ports (port 3) of the coupler would not be used. Another single-mode optical fiber configured as an output port (port 2) would couple the laser light into a compact scanning head that would contain an objective lens assembly. The light diverging from the output end of this fiber would be focused by the objective lens onto a small spot on the surface of a specimen (or, optionally in the case of a semitransparent specimen, into a small subsurface volume). Light reflected from the specimen (including Raman-scattered light) would be focused by the objective lens assembly into the fiber, where it would travel back toward the 50/50 coupler. The portion of the reflected light coming out of port 4 of the 50/50 coupler would be split by a 10/90 fiber-optic splitter; the weaker output would be sent to a photodetector and the stronger to a Raman spectrometer.

This apparatus would differ substantially from prior Raman probes in which scanning heads are coupled by use of multimode optical fibers. Because of its single-mode nature and small diameter (a few micrometers), the core of the optical fiber ending in the scanning head could be considered a pinhole, which, in combination with the objective lens, would afford the resolution needed for confocal microscopy with the depth discrimination needed for three-dimensional imaging and spectroscopy of semitransparent objects. In contrast, coupling by use of multimode optical fibers results in much coarser resolution - both laterally and in depth.

Another important difference between this apparatus and prior Raman probes would lie in the manner of discriminating between Raman-scattered and laser light. In other Raman probes, laser light is often rejected from spectrometer-input paths by use of combinations of notch and edge filters. In the proposed system, a Bragg grating incorporated into the core of the optical fiber going to the spectrometer would serve as a high-resonance-quality (high-Q), in-line rejection filter that would block light in a narrow band centered at the laser wavelength while passing the remainder of the spectrum. (A Bragg grating could function in this way only within a single-mode optical fiber; it could not do so in a multimode fiber.) The degree of rejection of laser light could be more than 80 dB.

It would be necessary to compensate for the temperature sensitivities of the narrow-band rejection filter and the laser because if the laser wavelength were to drift from the rejection wavelength, then too much laser light would get through to the spectrometer. The use of fiber Bragg gratings would offer a convenient solution to this temperature-compensation problem: Another Bragg grating, of relatively low reflectivity, located in the optical fiber between the laser and port 1 of the 50/50 directional coupler, would be used to lock the laser wavelength; the laser wavelength would be controlled by the reflection band of this grating, which could be made to match the rejection band of the Bragg grating in the fiber going to the spectrometer. The portions of the optical fibers containing these gratings could be mounted in contact with a common heat sink and thereby maintained at the same temperature.

Yet another notable aspect of the proposed apparatus would be the aforementioned relative compactness of the spectrometer. This compactness would be achieved by a novel design featuring only two reflective surfaces, one of which would be a convex diffraction grating shaped and blazed by electron-beam lithography in poly(methyl methacrylate).

In the version of the apparatus depicted in the figure, the fiber end could be translated along the fast direction of the scan relative to the specimen by use of a microelectromechanical (MEM) scanning mechanism. In an alternative version, the fiber end would be held stationary and a MEM scanning mechanism within the head would translate a small scanning corner reflector that would function in conjunction with a stationary folding mirror and an objective lens assembly. The other directions of the scan can be provided by ordinary piezoelectrics or other means.

This work was done by Pantazis Mouroulis, Mehdi Vaez-Iravani, and Frank Hartley of Caltech for NASA's Jet Propulsion Laboratory. NPO-20932


This Brief includes a Technical Support Package (TSP).
Confocal Single-Mode-Fiber-Optic Raman Microspectometer

(reference NPO20932) is currently available for download from the TSP library.

Don't have an account? Sign up here.