Avobenzone Stability?

[brett]

I've been having a bit of trouble with the stability of my Avobenzone in my HPLC method. I prepare a standard with 5 other organic sunscreens and when I inject the mix, I get %RSD's of <0.5% with the five others and a %RSD of >3% for my avobenzone. I've tried changing the diluent to all ACN, 35%water:65% organic, and a few things in between. I've also tried stabilizing it with an equal amount of octocrylene, but that didn't really seem to work. It also seems to degrade fairly consistently. Any ideas on how to fix this?

[Consumer Products Guy]

We had an SPF product that contained octocrylene, avobenzone, octisalate, and homosalate. I developed and cGMP-validated the procedure here, straightforward validation. We used isocratic elution, separated all five peaks (cis and trans for homosalate and calculated those individually) in under 7 minutes. We used a mobile phase different than yours, but that's proprietary.

We had good precision and accuracy on all five components, used a single standard containing all. Peak resolution was greater than 2.2 for all peaks, and this was regular HPLC on a 5.0cm length column too.

Concerning avobenzone specifically: we did find that the avobenzone was not as stable as the others when the standard mix was stored in the refrigerator; so the standard mix was made at least every two weeks.

We also know that the avobenzone degraded in the product prototype formulations due to the preservative, so the preservative system had to be changed.

[brett]

Our application and method sounds pretty similar. But without divulging anything you shouldn't, would you think that this issue has something to do with my diluent or mobile phase? Or is there something else I'm not seeing?

This problem only seems to be happening in my standard. The avobenzone peak area precision in my product samples appear to be fine, which I figure is due to the octocrylene or other avobenzone stabilizers in the formulation.

[Consumer Products Guy]

Our application and method sounds pretty similar. But without divulging anything you shouldn't, would you think that this issue has something to do with my diluent or mobile phase? Or is there something else I'm not seeing?

This problem only seems to be happening in my standard. The avobenzone peak area precision in my product samples appear to be fine, which I figure is due to the octocrylene or other avobenzone stabilizers in the formulation.

I can share this below. I can't say off hand if any of this would be the reason our RSDs and system suitability was always OK, and you're having issues. We made up the standards and samples in DMF. Instead of just water as the aqueous part of the mobile phase we used 0.5% acetic acid. Our sample preparation was 0.15 grams into 25 ml DMF, and calibration standard was made up at corresponding SPF levels as expected in the product, so chromatograms would practically be perfect overlays. We used about 313nm UV detection for all, UV lamp, 3 to 5 µl injection size.

One other thought: have you tried making up your calibration mix in a matrix of placebo product or similar non-SPF product?

Another "other thought": SPFs are big UV absorbers, and are in products at high levels. Make sure that your standards and samples are not too high, need to stay where UV detector is happy. And with SPF assay, rare to have any matrix peaks.

[brett]

I have tried adding the standards to a placebo but it apparently had no effect. My analyte signals are coming in at around 200 uV.

And from your suggestion, I tried substituting our aqueous mobile phase (0.1% H3PO4) in place of the water in the diluent but that was also ineffective.

I can't seem to understand why some days, everything looks perfect and other days the RSD is trash...that's really what bothers me.

A couple of things that are different than the methods I've used in the past is I am using a phenyl column with a guard column in front of it. You think my problems could be related to the column? It has been in use for around a year.

The downward trend of one vial injected 9 times is below (I think that's interesting and possibly significant):

[Consumer Products Guy]

Well, you've definitely demonstrated that there seems to be an avobenzone degradation, and pretty significant degradation. If your peaks look good, I doubt that it would be the chromatography or that the avobenzone is precipitating out of the standard mix.

I assume that you're using the same solvent for samples as for the standards. Did you detail exactly what solvents and how you are making standards? Are you using low-actinic autosampler vials or clear glass? We use low-actinic.

I think I would focus on the standard and try:
...making up avobenzone standard individually and seeing if that remains stable
...or at least try a different vendor's solvent
...or try a completely different solvent for making up standard