Techniques for marking analysis points on thin sections...pros and cons

[D.]

Hi All,

I'm curious how different people physically mark points for analyses on their thin sections. I was taught with a metallic glass-cutting scribe pen, but that was a long time ago. I've grown accustomed to EDS mapping my entire section and then marking on the maps.

I know some people use permanent marker pens to draw on the sections. Is there a danger that putting the beam on it wil vaporize the ink and recondense it inside the machine?

What other techniques are used these days?

Happy Sunday/Monday   
Deon.

[Malcolm Roberts]

I encourage my users to carry out a flatbed scan of the uncoated thin section. Good quality and high res about 600 dpi. Then they print this out and mark up those areas of interest on the print out. I then reduce the scan size to ca 1-2 megs and digitize with picturesnap. This way there is no need to scratch anything or apply ink to the surface of the thinnie and navigation is a cinch.
Cheers,
Malc.

[Nick Bulloss]

I've used a diamond scribe to make a fiducial mark on the corner of the thin section (off the sample itself). Then I used the optical microscope stage to measure coordinates of the areas of interest which can then be translated and located in the probe. It is a little fiddly and now I prefer the scanning method described by Malc.
Cheers,
Nick

[neko]

We have Cameca Peak Sight 4, and loading offline images into there reliably is a pain (input the center and then the image dimensions and size in microns), so in addition to the thin-section scanning via flatbed scanning, I made some checkerboard images that are 2,500x4000 pixels, and I just told the cameca software to load that in 4 different times, and put a different center coordinate in for each one (25,000/2 x 40000/2, changing the signs for the specific quadrant). I saved those in the impDat file and now I have generic navigation aids for any given thin-section, and we can use an image editor to overlay the scan on the checkerboard image, line it up based on where things are in the probe, and you have an accurate navigation aid. If you have close-ups from microphotography you can layer those in as well.

I divided the image about 10x in the X direction for the checkerboard pattern. Plus, after you've lined things up between the scan and checkerboard, you can save that and import it back into Peak Sight with the exact same settings you loaded the plain checkerboard into, and it will be accurate enough until you remove the section. And can just adjust when you load a slide back in if the lack of accuracy drives you crazy.

Cheers,
Nick
[I'm pretty sure PfE just does all this automatically with far less work]

[John Donovan]

We have Cameca Peak Sight 4, and loading offline images into there reliably is a pain (input the center and then the image dimensions and size in microns), so in addition to the thin-section scanning via flatbed scanning, I made some checkerboard images that are 2,500x4000 pixels, and I just told the cameca software to load that in 4 different times, and put a different center coordinate in for each one (25,000/2 x 40000/2, changing the signs for the specific quadrant). I saved those in the impDat file and now I have generic navigation aids for any given thin-section, and we can use an image editor to overlay the scan on the checkerboard image, line it up based on where things are in the probe, and you have an accurate navigation aid. If you have close-ups from microphotography you can layer those in as well.

I divided the image about 10x in the X direction for the checkerboard pattern. Plus, after you've lined things up between the scan and checkerboard, you can save that and import it back into Peak Sight with the exact same settings you loaded the plain checkerboard into, and it will be accurate enough until you remove the section. And can just adjust when you load a slide back in if the lack of accuracy drives you crazy.

Cheers,
Nick
[I'm pretty sure PfE just does all this automatically with far less work]

Hi Nick,
But of course mon amie!   

There are several ways of getting this done in PfE. One method is the Stage mosaic window described here:

http://probesoftware.com/smf/index.php?topic=324.msg1659#msg1659

But probably the most popular method related to your endeavour, is our PictureSnap feature which is discussed in several topics on this forum, but here's a good place to start:

http://probesoftware.com/smf/index.php?topic=14.0

One cool aspect of PictureSnap is that it automatically handles sample rotation (e.g., round mounts) by utilizing three fiducial marks. Of course this also means that it also automatically handles changes in z position through interpolation.
John