Avobenzone issues in Sunscreen method

[tom jupille]

If this were my problem, the first thing I'd address is the peak shape issue (you hit the nail on the head there!). To my eye, it looks like the avobenzone is binding strongly to something on the column and only bleeding off slowly (that would also explain why you don't see any peak at low levels).

So, first question: is this a brand-new column, or has it been used for something else (possible damage to the column surface?). If it's a used column, my first impulse would be to put in a new column. If it's a new column, try switching to a different C18 column -- either a "conventional" column (i.e., not solid-core) from PE or a solid-core column from a different source.

Until you get the tailing under control, anything else you do will be spinning wheels.

[brett]

The column I've been using was brand new. I actually used it for one run of a usp salicylic acid method but after that has been all sunscreens, and from the start has been missing the avobenzone peak.

Tom, that's pretty close to the conclusion I came to as well...I just wanted to see if I had missed something or if it was user error on my part. I still don't understand how avobenzone can interact with the column so differently than the other analytes though.

But, I think the general suggestion is to just buy another column? Any of you guys have any recommendations on another brand + specs?

[Consumer Products Guy]

But, I think the general suggestion is to just buy another column? Any of you guys have any recommendations on another brand + specs?

You asked - so I'd ask. I'd ask sales representatives of the chromatography companies you deal with for trial columns, let them know that if it looks good you'll be buying columns of other lot numbers for specificity/cGMP validations, and maybe transfer to production. I've done just this several times, and did end up validating methods using such columns.

[brett]

That is excellent advice. Thanks for everyone's help on this!

[juddc]

OK, I see something here that's not been addressed and I think it could be important. First, avobenzone undergoes a keto-enol tautomerism, which will affect your chromatography. Second, the rate at which that occurs is pH dependent. I found in my own work with avobenzone containing sunscreens that acetic acid does not yield a low enough pH to give good peak shape with regard to avobenzone. The diketo tautomer absorbs at ~267-nm if I recall correctly, and is clearly distinguished from the keto-enol, which absorbs at about 360 nm. What I see in your chromatogram of avobenzone alone are both tautomers: the diketo is at 1.5 min, you don't quite get to baseline, then your enol is your second larger peak at 7.5-8 min. You may be having a secondary column interaction with the enol; I've never used the column you're working with and haven't researched it, so I can't say for sure.

My suggestions:

1. Change your MP pH modifier from acetic acid to either phosphoric acid or TFA.
2. I found excellent and robust separation using a Waters Symmetry C8, 3.5uM column (3.0mm x 100mm) using a binary gradient between 0.1% aq H3PO4 and acetonitrile. This combination will yield a separation wherein the avobenzone elutes before the octinoxate and others, which is important because the diketo tautomer elutes before the keto-enol and (of course) never quite goes to baseline because the two are in equilibrium.

If I may be of further assistance, please feel free to post or email. I'll dig around and see if I can find the conditions I used to use.

Cheers!

CJ

[juddc]

Digging through the archives yielded this chromatogram. The small peak at 5.34 min is the diketo tautomer of avobenzone. The analytical wavelength was 330-nm and the instrumentation in question was a Waters Alliance 2695 with either a 2996 or 2998 PDA.

[Consumer Products Guy]

JuddC - nice chromatogram. Interesting, your octocrylene elutes last, and on my chromatogram posted earlier in this thread, our octocrylene elutes first. Just shows: there's more than one way to take a cat's skin off...

[juddc]

Thanks very much! It was a fun separation to develop - I recall the order of retention changing significantly during development based upon the solvent being used to elute. Methanol would give a completely different order than acetonitrile and THF would give yet a third order of elution. Your separation looks sharp as well - nice to have it that fast and isocratic. I did have a quicker method for simpler mixtures but it wouldn't pass specificity during forced degradation for the 5 component mix.

Note to OP - watch your sample prep as well. I found that acetonitrile yielded better & more consistent recovery than methanol and this might seem counterintuitive, but if you're working with a heavy lotion or cream with a high viscosity / polymer content, it might be wise to add just a bit of DI water to your flask once the sample is weighed into it in order to break the viscosity of the sample. If weighing perhaps ~100-200-mg of sample into a 50-mL flask, I'd add 500-ul to 1-mL of water, vortex that briefly, then add acetonitrile to complete the extraction.

[InChrom]

Hello, Has anyone tested impurities in sunscreen? I can't seem to find info on thermal degradation or stability study for sunscreen products. There are some info available for photodegradation but very limited data on the analytical method used.

[gtma]

I wonder if it should be consider like a preservative where impurities determination is not required. Any thoughts?

[Consumer Products Guy]

Hello, Has anyone tested impurities in sunscreen? I can't seem to find info on thermal degradation or stability study for sunscreen products. There are some info available for photodegradation but very limited data on the analytical method used.

Our products passed 6 months FDA stability at 40 C. Some sunscreen actives are esters so PH of the product cannot be high or those will saponify.