SEM-CL imaging

There are two ways you can do CL imaging with the SEM system.

The first is using the VP detector, which is light-sensitive. It is OK for brighly luminescent samples and broad areas. Remember that it's off to one side, so you will see shadowing effects. Also, areas farther from the detector will be darker, though higher magnification, multiple images, and stitching software can help with that problem, as will lowering the stage as far as possible.

The other method is to use PMT1. It uses the RPM to gather light from the luminescing sample. It can obtain panchromatic images (all colors, no color filter), or with the color wheel filters (controlled by the LabSpec software).

Getting ready

Turn on and optimize the SEM using the SE detector, turn on the CL hardware and software, and center the RPM.

RPM mirror schematic

The PMT of interest is in the little white box on the right. Rotate the thumbwheel so that it clicks with the ON label visible. That moves a mirror into the light path, directing light to the PMT.

Note: Remember to move the thumbwheel to OFF when you are finished. You can't collect spectra or center the mirror with it on.

Horiba controller box

The SEM brightness and contrast controls don't work with this system, so you use this controller box instead. You DO use the SEM scan and image acquisition controls, however, NOT the LabSpec software. LabSpec is here only used to center the mirror, and also to change the color wheel filter positions.

Press the STAND BY button. That turns the box on.

Press the HV button. That turns on the PMT1 high voltage.

Press the Load button on the screen to turn on the control interface. This screen shows bar graphs for contrast (left) and brightness (right). The two buttons at the top of the screen top give you coarse or fine control for the contrast and brightness control knobs.

The SEM should already be running and focused, and should be within a brightly and evenly luminescing material. Switch the detector to AUX1, which is PMT1. Adjust the control box brightness and contrast to see an image.

The image you see will probably look like nothing at first. Remember: 1) you are looking at a homogeneously luminescing surface, so there is no detail, and 2) the electron beam is rastering over a rectangle, not stationary at the RPM focal point, giving very inhomogeneous image brightness.

Also remember that CL light comes from a volume 1-2 microns across, not the much smaller electron beam spot. You can't get the resolution you might expect from an SE image. You can get better resolution with lower beam voltages, though you may need higher current, too.

And another thing. The luminescence electronic transitions can be pretty slow, especially compared to secondary or backscattered electrons, so images tend to streak out to the right. Slower scan speeds may be needed.

Slowly raise the stage using the Z-axis controls, maybe 0.2-0.3 mm from the RPM focus position. Your CL image should blurr, and the bright region should move to the right.

On LabSpec go to: Menu Maintenance / RPM adjustments / RPM_X / Offset. Change the X value to move the bright region of the image to the left.

Play around with the stage Z-axis position, and the LabSpec X-axis position, until the image field is more or less evenly illuminated, like this. Because of the focal characteristics of the parabolic mirror, perfection is probably not possible.

Using the SE detector, re-focus the SEM using the keyboard focus controls.

Line scan function to adjust constrast, brightness

Move the stage to a sample you want to image, and switch back to the PM1 detector.

Because slow scans are necessary, setting the best contrast and brightness for the PM1 detector is not easy using the whole image frame. Instead, on the SEM scannng tab, select the Line scan check box. That gives you a horizontal line across the field, which you can move up and down to cross dark and bright areas. Adjust contrast and brightness (Horiba control box) so the red signal intensity line intersects the graph top and bottom (maximum contrast). The slower the scan speed, the better this works. Remember that the electron beam is rastering across a single line, not the whole field, so even a slow speed is fast.

Once you have adjusted contrast and brightness, un-check the Line scan box and acquire your image in the normal SEM way. If it looks good, now you can make a color image.

In LabSpec, go to: Menu Acquisition / Instrument setup / Filter wheel, and select the blue filter.

Adjust contrast and brightness using the SEM line scan feature, and acquire a blue image.

Change to the green filter.

Adjust contrast and brightness using the SEM line scan feature, and acquire a green image.

Change to the red filter.

Adjust contrast and brightness using the SEM line scan feature, and acquire a red image.

Use third-party software to do any additional image processing, and combine the images to make a color RGB image.

This is blue-luminescing microcline containing a red-luminescing albite exsolution lamella. The lamella luminescence is patchy, possibly related to alteration.