Carbosulfan analysis via HPLC - Question regarding stability

[scottlandergrossman]

Hi there,

I have a question about the analysis of carbosulfan via HPLC with UV detection. Basically, I was wondering if anyone had any experience with carbosulfan breaking down in aqueous mobile phases, especially ones with low pH values (e.g. 2).

I have a suspicion this might be happening in my analysis based on some data that was collected, but I'm really not sure.

From what I've read, it seems like carbosulfan can turn into carbofuran through a number of different mechanisms when it's in the environment or in some animal, and one of the mystery peaks elutes around where carbofuran would elute, but since it also seems that there are folk interested in studying the presence of carbosulfan and its metabolites (carbofuran, for instance), it would be less than ideal if that transformation occurred in the instrument when the sample is injected into a starting condition with high aqueous content (which is the case in some of the papers I was able to find - and is my case).

Anyway, I was just wondering if anyone might have some insight into this issue.

Thanks!
Scott

[bisnettrj2]

http://www.capl.sci.eg/ActiveIngredient ... ulfan.html

Hydrolysed in aqueous media; DT50 0.2 h (pH 5), 11.4 h (pH 7), 173.3 h (pH 9)

Seems like degradation would be a problem, but I don't think you'd have a distinct peak at the RT of carbofuran, as I would expect the degradation to happen the entire time the analyte was on-column. What's your injection solvent? Can you run neutral (or preferably with a basic mobile phase?) and see if the peak for carbosulfan increases for a standard of the same concentration, with a decrease in the peak observed at the carbofuran RT?

[scottlandergrossman]

Okay...first of all, your quote is hilarious.

Second, thank you for the reply. Can I ask where you got your degradation time data? It's very helpful , and if it's from a resource I would have access to it would probably be good to...well...access it.

We could try using an aqueous phase with a neutral or basic pH to see what happens. That's a great suggestion, especially in light of the DT50 value at pH of 5.

The vast majority of my chromatography experience is with gas chromatography, and much of that was spent analyzing samples of explosives. HMX is a classic example of on-column degradation, and it gave me the same expectation that you seem to hold. But I wasn't sure how much I could extend that experience to this situation.

But with three distinct, very well shaped peaks that have a spread of 20 minutes between the most distant pair, it certainly did seem that if carbosulfan was degrading in the mobile phase it would almost have to be doing so immediately, probably before or at the very head of the column, and then stopping almost immediately. Otherwise I would have expected a cruddy peak shape for the carbosulfan with foothills extending most likely towards earlier retention times.

Another piece of data I can offer now, though, is that if the same sample (which is in acetonitrile) is analyzed on a C18 with an isocratic run of 90% AcN and 10% of the 20 mmol phosphoric acid in water produces a c-gram with just one peak at about 9 minutes (150 mm, 4.6 mm, 5 um; 1 mL/min), where the same sample run on a C8 under the same conditions produces a single peak at about 2 minutes.

The latter I just attributed to the lack of retention, but the former makes me wonder...

I wonder how much the degradation could be a pseudo-red-herring. Suppose it's not breaking down in the LC, but what if water somehow was in the raw material? I could be getting degradation in the bottle and just separating the peaks in the analysis.

Thanks so much for your help (and your humor - the laugh helped just as much as the data!).

Best,
Scott

[scottlandergrossman]

...one more thing.

If I'm seeing what would appear to be a raw material with 75% purity, then I'd be getting only 25% degradation. From the degradation time data, if it only takes 12 minutes for 50% of the stuff to degrade, then at pH 2 perhaps it's feasible that I really could be getting significant breakdown in just the first few minutes of the analysis when the mobile phase is mostly acidic water. The first peak doesn't appear until about 14 minutes into the analysis when the gradient is at about 50:50.

If the degradation slows considerably with the changing proportion of acidic water, then I guess the degradation really could basically "freeze" and it wouldn't be too terribly different than injecting a solution of three compounds.

Does that make sense?

[bisnettrj2]

Link to degradation data is at the top of my post.

I can't speak to the degradation rate as a proportion of aqueous:organic over the course of a gradient. I would attribute sharp Gaussian peaks to something present at injection or very near the injection, especially if they match authentic standards retention time. Are you using clear or amber injection vials? Is the standard stored in a freezer? Is it prepared from neat material or from a prepared standard from a standards supplier?

I used to analyze explosive by HPLC via EPA method 8330B. Tetryl is a very poor performer in water, heat, and if you look at it cross-eyed. Seems like carbosulfan is a similar type of analyte.

If you try a basic mobile phase (I'd go with 0.02% ammonium hydroxide in water: 0.02% ammonium hydroxide in methanol using your same gradient, basically because it's simple to make), make sure you use a column that can handle basic pH. I would be wary of using a pH 2 mobile phase on most reversed phase HPLC columns as well. I like the Phenomenex Gemini C18 for weird pH mobile phases - very robust. A Waters XBridge C18 or a XTerra MS C18 or an Agilent Extend-C18 can also handle high pH mobile phases.

Glad you got a laugh from that quote.