Raman Microscope construction.

Raman Microscope Part List

•  1.  Helium-Neon laser 10 mW from THORLABS, polarized model HNL100L-EC.

• 1b. Laser power supply from THORLABS.

• 1c. Geoptik GK2 micro motion guide mount, normally used to guide a telescope. Here, it allows changing the position of the laser by tilting it around two axes, one horizontal and one vertical.

• 1d. Mechanical device allowing the adjustment of the vertical position of the laser ( coming from an old photographic enlarger). The direction of the laser tube is not changed during the motion.

• 1e. Horizontal adjustment of the laser position with an ADM dovetail bar.

•  2.  Laser-line band pass filter 632.8 nm. Edmund Optics NT47-494 integrated to the laser head.
•  3.  Spatial filter assembly Edmund Optics NT62-769. The pinhole normally delivered with the device has been replaced by a second 9 mm  objective Edmund Optics NT45-047 especially designed for He-Ne laser.
•  4.  Microscope analyzer used to attenuate the laser light in order to facilitate the positioning the sample manually into the laser spot looking through the microscope eyepieces.
•  5.  Microscope objective. Best results have been obtained with a 50X metallographic, NA 0.85 from Olympus.
•  6.  Rock sample.
•  7.  Modified flip Mirror MEADE M647 used as a beam splitter holder equipped with a Laser-Line Dichroic Plate Beam splitter Edmund Optics NT64-287 632.8 nm.
•  8.  Laser line edge filter Edmund Optics NT47-506 mounted in an Olympus U-OPA orthoscopic adapter.
•  9.  Neon light bulb for the wavelength calibration of the spectrograph.
•  10. Flip mirror MEADE M644. The light from the laser or from the neon bulb is sent to spectrograph.
•  11. Spectrograph entrance slit.
•  12-13. Modified Jobin-Yvon H20 monochromator. Concave holographic grating 1200 lines per mm.
•  14a. Bellow between spectrograph exit slit and camera objective lens.
•  14b. Olympus bellow for the adjustment of spectrum image on CCD sensor.
•  15. Camera lens, Olympus 50mm macro f2.
•  16. CCD Camera Atik 16HR with power supply 16b.
•  17. Olympus photo Eyepiece 2.5X used to refocus Raman light into the spectrograph entrance slit.
•  18. Diodes illuminator for examination of the sample.
•  19. X-axis metric stage 100mm travel from Edmund Optics NT56-338. This device is used to position exactly the focused laser spot into spectrograph entrance slit.
•  20. Lens used to see the laser spot on the spectrograph entrance slit plate.
•  21. Flip mirror cover.

Comments:

      One  important think when setting up a Raman microscope is to focus the laser beam at the center of the microscope field with a high power objective like the Olympus 50X NA 0.85. In this design, the position of the laser  can be moved in 4 directions (1c, 1d, 1e) : vertically and horizontally without changing the direction of the laser tube; the tube can also be tilted in a vertical and horizontal plane. Much resources can be found for such a design in the equipment sold for amateur astronomy. Good parts can been found at a price generally much lower than the professional microscopy components. For this mechanical assembly of the laser positioning, I used dovetail bars, a guiding telescope assembly, the focuser of an old photographic enlarger for the vertical motion and the stand of my old drilling machine repainted in black for a better look.

      The first design of this Raman microscope has been made without any spatial filter assembly. In that case, the laser beam reaches directly the microscope objective lens and is parallel to the optic axis. In that way there is minimum focusing action of the objective. Consider a beam very narrow and moving on the optic axis of a lens, it continues its way on the same line after passing through the lens, there is in fact nearly no focus if the width of the beam is sufficiently small. The image of this hypothetical beam is always a point whatever the distance to the lens. In practice this never happens because the beam has a finite width but when a narrow parallel beam is focused, the cone of light thus generated reach the focal plane with a small angle. If the light beam covers fully the pupil of the objective, the focused beam reaches the sample with a higher cone angle and the Z resolution is better. The small diagram below illustrates the Z resolution problem:

           This figure represents the light beam coming out the microscope objective, being focused onto the sample surface and diverging into a transparent crystal below the surface for two different light cone angles.  In case A on the figure, the beam is strongly focused on the sample plane. At small distance below the sample plane, the light is defocused and will be partly blocked by the entrance slit of the spectrograph. In case B, where the beam cone angle is small, the light remains in focus on a larger Z distance. The light coming from greater sample depth can enter the spectrograph so Z resolution is lower. For that reason, it is always advised for best results with Raman microscopy to use a high numerical aperture objective which increases the cone angle of the light beam.

             The beam splitter assembly (7)  and laser rejection filter (8) are described in more details on next page. As seen on the photograph above, this assembly is now part of the microscope column. To do that it is absolutely necessary to use an infinity corrected microscope's optics so accessories can be introduced between objective and microscope head without disturbing the image. The height of the beam splitter head is around 15 cm which is outside the specifications given by Olympus. Even in those conditions, I did not notice any significant image deformation.

             The second flip mirror (10) is equipped with a small neon bulb (9) for reference and to the left by a photo eyepiece to refocus the light onto spectrograph entrance slit. At the end of the beam splitter tube, a removable cylinder (an astronomical accessory from Orion) gives access to the entrance slit. With the help of an external lens (20) and the X-axis Stage (19), the focused spot can be precisely centered on the entrance slit by moving slightly the front of the spectrograph.

             The spectrograph (12) used in the present device is certainly not optimized but as I said before, I tried to use available components as far as possible to demonstrate the possibilities of Raman microscopy before optimizing all devices.  The Jobin-Yvon H20 used in this experimental arrangement is not a spectrograph but a scanning monochromator with entrance and exit slit. In normal use, the image of the entrance slit is formed in the plane of the exit slit. I have removed the exit slit to record a band spectrum and not a single wavelength  in the monochromator mode. It was not possible for mechanical reasons to place directly the CCD detector (16) in the image plane of the spectrograph. On the other hand, the direct imaging on a CCD detector with such a spectrometer optics is not suitable because the image provided by the concave grating is too large so only a small portion of the spectrum can be recorded at a time ( about 35 nm) and the image intensity is too faint for Raman spectroscopy. To overcome this problem, a demagnification with an additional lens  was necessary. I have chosen to photograph the spectrometer image ( virtual object ) as usually done in photo macrography that is with a bellow (14b) and macro lens (15) (Zuiko makro f2 for OM system). The biggest drawback of this optics is the vignetting produced by both bellows (14a and b). I'll describe this limitation later when presenting the first results.

            The CCD (16) detector is an astronomical camera cooled by Peletier around 30°C below ambient temperature ( Atik 16HR equipped with a SONY ICX285 CCD sensor 1392x1040 pixels ). I used this device first to make some astronomical pictures of faint nebulae. As it allows  to record images of faint objects barely visible with a telescope in the light polluted sky of Belgium, I decided to try it for Raman spectrography. I have reported below an example image I have made with this camera with a narrow band H alpha filter of such a faint object. (Rosette nebula). With such a camera, it is possible to make binning of the pixels that is to add the charge of some adjacent pixels directly on the CCD chip to improve the signal to noise ratio. I'm using a 8x1 binning to add 8 pixels in a column. This is necessary again due to the low intensities of Raman spectra.

             After setting up  the components, the optics must be aligned starting from the laser output. Both objectives of the spatial filter (3) must be aligned vertically with the 2 screws provided so that optic axes are perfectly collinear. The dichroic beam splitter (7) should make an angle of 45° with the laser beam in order to minimize transmitted laser intensity and thus maximize reflection. The position of the laser head is then adjusted (1c, 1d, 1e) so the beam is focused at the center of microscope stage (6). The first step is to bring the laser beam at the center of the field with the microscope objective being removed. Then in succession,  microscopes objectives of increasing power can be used with an adjustment of the laser beam at each step. Finally, the beam width is adjusted with the spatial filter micrometric screw to minimize laser spot size.

             With a properly aligned laser beam, the spectrometer can be tested with a high Raman scatterer. For that purpose, I'm using a section of the mineral crocoite (a lead chromate) which gives a very high Raman effect, more than sulfur. With this mineral, a Raman spectrum is collected in a few tens of a second so the spectrometer alignment can be set in real time.

Rosette nebula.