There are many technical difficulties with using a SIMS (Secondary Ion Mass Spectrometer) instrument, in no small part because we use it to analyze Very Small Things. Take the calcite below, which is actually a decent size of a few tens of micrometers across (on the thin side for a human hair).
On the left is a cross-polarized view from a petrographic microscope. The right is a backscattered electron image from an electron microprobe.
I am investigating manganese-chromium isotopic ratios in meteorite calcite grains (and carbonates in general), and need to be able to analyze it with the SIMS. A good analysis requires avoiding cracks, inclusions, and other phases during analysis. Cracks, for example, tend to be “collection” areas for Mn or Cr, either through contamination during sample preparation or if there was low degrees of aqueous alteration mobilizing those elements and redepositing them along transport pathways. The beam size for my analyses will be about 5 micrometers, so there are a few places on the calcite grain I could analyze. But due to technical and geometric difficulties within the SIMS sample chamber, our main navigation tool is an optical camera where that same calcite grain looks like this (at end of arrow):
So we use ion imaging, where we raster the ion beam over the sample. But that raises another issue – the sample is carbon coated, so you have to ablate it away before you can see anything, which also can cause problems with conductivity if you isolate the area of interest from the carbon coating.
So instead, before we run SIMS analyses, we typically “burn” a hole through the carbon coat with a focused electron beam using a scanning electron microscope. The process involves zooming in to maximum on the spot to analyze and waiting for a few minutes for the electron beam to damage the carbon coat. Then, later during SIMS imaging, the ion beam hits the sample in that spot only, while the rest of the thin section is masked by the carbon coat. However, some phases are easily damaged by beams, devolatilizing or otherwise breaking down. Calcite is one of these more sensitive minerals. Focused beams devolatilizes it easily (loses CO2) and possibly disturbs the Mn/Cr ratio. A different method is needed.
Recently, Hope Ishii and John Bradley established the Advanced Electron Microscopy Center (website here) on campus, and they have a top of the line focused ion beam (FIB) instrument (FEI Helios NanoLab 660 model). It has the capability to mill out sections of samples to prepare ultra-thin slices (FIB sections) for high-resolution analyses. We have found it is also great for poking holes through the carbon coat at precise locations. It uses positively charged gallium ions to mill out a specified shape and to a precise depth. With this control we can selectively mill through the carbon coating, which is about 25 nanometers thick, without removing much sample. In the image below, the orange arrow is pointing to a 1 micrometer diameter hole drilled into the sample. We call them FIB marks.
Once we have all the grains we want marked, we can load them up and use ion imaging to find the exact location of the FIB mark. This involves roughly centering the image over the sample using the optical camera, then rastering a ~30 micrometer square at low beam current to ablate some material to view ablated ions. The following image is an example of ion imaging from a different grain. The brighter areas near the edge of the image are the borders of the calcite grain of interest, and the bright spot in the upper center is the FIB mark.
Once the beam is centered over the desired analysis spot, then it is time to analyze! And afterward, we verify where the beam hit the sample to make sure we didn’t miss. Sometimes it is obvious if we missed because it throws off the expected isotope ratios or element abundances. But as you can see below, all the preparation paid off and we hit where intended.
It is a lot of work for one analysis, but sample preparation and documentation is the key to a good analytical session.
poikiloblastic 