• Marisa Acosta

Perthite Modal Analysis

So, mostly I'm writing these days, which does not make for the best of blog posts. BUT , I did have a small foray into doing a modal abundance analysis on plagioclase lamellae in K-feldspar from a granulite (the purpose being to apply 2-feldspar microthermometry).

Perthite is a term used to describe the intergrowth of two feldspars (a background K-feldspar with blebs of albitic feldspar). This texture is the result of subsolidus exsolution - at high temperatures (when crystal lattices are generally more accommodating to ions of slightly different size) sodium and potassium readily substitute for one another, forming a solid solution between albite and K-feldspar end members. However, when the temperature of the system falls below the solvus two feldspars instead of one are the stable phase.

Here is an scanning electron microscope (SEM) - backscattered electron (BSE) image. The dark grey regions are the plagioclase lamellae and the lighter grey background area is the feldspar host.

Initially this project sent me down a rabbit hole of solid-solid nucleation mechanisms and kinetics, but I won't go into that here.

To do a simple modal abundance analysis of the two phases, I used a very simple workflow in ImageJ (aka Fiji).

After opening ImageJ, open the BSE image of interest and set the initial thresholds manually. The key here is to make sure the phase of interest and only that phase is colored in.

After having done so, recognize that the area that is colored in (black in the images shown) will be what is reported on in the next steps.

Set the measurements as shown below.

The "Area Fraction" is the the important one to check. Next, choose "Analyze" and save the output with an appropriate file name.

Re-threshold appropriately and analyze the second phase (or third or however many it is you are doing).

Rinse & Repeat.

Note that this method is suited for BSE images and some plane-polarized light petrographic images. It will not work with phases that are strongly pleochroic (without some clever manual manipulation on your part) or on cross-polarized light images with some crystals at extinction.


Recent Posts

See All