Could it be trithiane solids?

We had an interesting spent scavenger sample roll through our lab a few weeks back. This sample came from a long-time customer who has used our services for many years. Based on conversations and innuendo, it seems they believed they had trithiane solids in their system. They were getting a white precipitate—a goo, really—which acted as one might suspect trithiane solids would. However, based on all of the sample we’ve seen over time (and the reaction physics), we don’t believe trithiane can form in any appreciable quantity in a scavenging situation. Belief is good only until you find the sample that proves you wrong.

Here’s a picture of the sample. The customer’s sample—the vial with the white cap—contains three layers: a top organic layer with a light hydrocarbon; a bottom, aqueous, viscous layer that appears to be spent scavenger; and a middle layer with a “goo”. The top and bottom layers both contain low-percent MEA-dithiazine with no appreciable MEA-triazine.

(left) Trithiane mixed with aqueous scavenger and light hydrocarbons (right) Sample from scavenging facility

Although the customer did not explicitly say so, we are pretty sure they believe the middle layer to be trithiane. Many years (a decade!) ago, we tried to take a Raman spectrum of trithiane, but the fluorescence from the white powder was too strong to get a clean spectrum. This latest sample got us thinking about re-doing that work. So, we purchased some 1,3,5-trithiane. We added it to a vial with some light hydrocarbon (NPLs) and with some low-concentration MEA-triazine in water, and then mixed the vessel aggressively. Once the materials settled, the sample split into three layers: a top organic layer, a middle “goo”, and a bottom aqueous scavenger layer. Looking at the black-capped vial, it sure looks a lot like the customer’s sample.

Quantitative Raman Spectroscopy for trithiane solids identification

Neither the suspected trithiane solids “goo” nor the trithiane gave a good Raman spectrum. There is too much fluorescence, completely obscuring the Raman signal. But our bag of tricks goes deeper, so we switched from our normal %-level Quantitative Raman Spectroscopy methods to our trace-level, enhanced methods.

In the figure at the top, the green trace is trithiane. It has strong peaks in regions where we would expect it to have peaks: C-S bonds. The black trace is analytical-grade MEA-dithiazine with all the characteristic peaks we have come to know and love. The blue trace is the “goo”. It also has strong C-S bond peaks.

The “goo” has a couple peaks that align with dithiazine, but is not identical. This observation could mean it is a mixture of dithiazine and another C-S containing molecule. Or it could be some form of polymerized dithiazine. More importantly, the “goo” has no peaks that align with trithiane. Not only are the strongest peaks shifted from the “goo” spectrum, but trithiane also has a characteristic peak near 413 rel. wavenumbers that does not appear in either of the other spectra. In the literature, this peak has been assigned as a ring deformation, a symmetry that does not to exist in dithiazine.

This data is an example of what makes Quantitative Raman Spectroscopy so powerful. Not only do we have the ability to quantify constituents of triazine-based scavenger solutions, but also the ability to identify new or unknown compounds in a sample.

Share the Post:

Related Posts