Chromatography Forum

In my experience, an ELS is an excellent detector to have in one's arsenal. They have their limitations in that they're not linear, you won't quantify anything that's volatile (obviously), your mobile phase additives need to be volatile, and injection to injection reproducibility is't as tight as a UV detector (though I've found ~1% RSD to be entirely attainable). I find mine to be quite rugged, however, and it's very useful for determining all sorts of analytes. Your calibration curves need to have a minimum of ~6 points and always run a system suitability before doing ANYTHING quantitative. You will find that your mobile phase composition will have a tremendous effect upon sensitivity as well as selectivity. Make sure that your water is up to snuff, keep your solvents clean, and have fun!

Thanky JUDDC for the advice.
But could you please briefly explain "why" injection precision is apparantly tenuos-- fails (or is not as reproducible) as we've observed with our UV Detector?
Is it the simple matter of droplet size from the eluatn or are there other confounders that inhere? And you noted the importance of preparing six cal curve points (usually we just make three and a blank)--is this a way to mitigate the issue of ''shakey" reproducibility?
We are awaiting price quotes on ELSD's now.

With regard to reproducibility, I'd guess that variation in droplet size is one issue and that peak shape is another. Because it's a nonlinear detector, I would think that you'd see greater area variations for peaks that are tailed or fronted than for those that are not. Generally, I've not found this to be a significant impediment to accurate quantification, though it's something to be aware of. The better you have your method optimized for peak shape, resolution, and retention time stability, the less variation you'll see in repeated injections. I know that sounds really obvious, but with an ELS method you need to be particularly aware. You'll need more that 3 points to calibrate, simply because the ELS response is quadratic and you need more points to deetermine a quadratic curve. I'll see if I can dig up some more info. In the meantime if you'd like to drop a line, feel free to send a note to christopher dot judd at basf dot com.

This article may be of interest:

http://chromatographyonline.findanalyti ... 9&pageID=5

Thank you for the reference reading; we are perusing it now. Thanks!

Been tasked with quantifying dimethicone in a body creme pilot batch. The upper managment says that FDA requires the testing for level of dimethicone and if it degrades faster than the other ingredients in the body creme.

With second look, I still think your management is interpreting the cGMP regulations wrong. Would the dimethicone be broken out as the active with the Drug Facts labeling? You state "if it degrades faster than the other ingredients in the body creme"; well, guess what: the active cannot degrade, you must do stability-indicating OTC tests to show it's stable, has nothing to do with whether other ingredients are stable.

As to HPLC-ELSD, I use it sometimes. But you may not get linearity to meet cGMP standards with it.

OK, the price quotes have come back from 3 vendors and the ELSD ranges from $21,000 to $24,500 USD. Used ones are ranging $18,500 to $21,000 USD. We are a small company with and the purchasing manager says this purchase "will not happen". So, I presume they will be content doing send-outs for PDMS quants. Nevertheless, this exploration was a learning experience. Thank you to all who offered input on this thread.

Well, it looks like it would be cheaper just to get an IR bench and a fixed pathlength cell!! I use it almost once a week, dimethicone in a lotion is a "label claim" product and I test it all the time. The extraction is a bit of a pain, but the results are always good. But I was thinking of dissolving the lotion in IPA or tolulene and using an RI detector...any thoughts?

Perspect to this issue, we are investigating tricking out this HPLC with an ELSD unit in lieu of the UV detector. Apparantly, PDMS do not contain a chromophore; however, quantifying the amount of scattered light generated by the sample residue per a fixed droplet size of mobile phase from which the analyte is evaporated may hold promise. But I realize none of this will be as easy as the manufacturer's say it will be . Any tips or user notes?

You will be much better of with a CAD Detector. CAD (Charged Aerosol Detection) will give you more information and sensitivity. ELSD will be okay but Dimethicones/PDMS is a profile of peaks and the CAD will give you a better profile to look at. When you run an actual sample you will soon realize that additives in the samples will interfere with parts of the Dimethicone profile and you will be better off with the sensitivity of the CAD for trying to integrate the clean portion of the profile. By profile I mean finger print-however you want to term it.

Get the book, The Analytical Chemistry of Silicones, by Dow Corning editor A Lee Smith, ISBN 0471516244. It is about $300, but I got one on an inter-library loan from the University of Oklahoma.

I have tried as well, and the problem that we run into is that our product has five polydimethylsiloxane compounds that overlap when we try to analyze them. For dimethicone alone, we use IR

Some success has been had by others using ICP-OES and microwave digesting with TMAH (tetramethylammoniumhydroxide) first. The article is found in: Fresenius' Journal of Analytical Chemistry, Volume 370, Numbers 2-3 / June, 2001, Pages 246-250.

Alltech has an HPLC method, but it appears more qualitative than quantitative, although it might be a start to base your analysis.


Brett

Been tasked with quantifying dimethicone in a body creme pilot batch. The upper managment says that FDA requires the testing for level of dimethicone and if it degrades faster than the other ingredients in the body creme.

With second look, I still think your management is interpreting the cGMP regulations wrong. Would the dimethicone be broken out as the active with the Drug Facts labeling? You state "if it degrades faster than the other ingredients in the body creme"; well, guess what: the active cannot degrade, you must do stability-indicating OTC tests to show it's stable, has nothing to do with whether other ingredients are stable.

As to HPLC-ELSD, I use it sometimes. But you may not get linearity to meet cGMP standards with it.

Your QA department sounds way more anal than mine. And that's tough to believe...

Try doing it via ICP. Its a lot of work but i was able to quantify it accurately.