We recently purchased an Alltech ELSD to determine Polydimethylsiloxane (PDMS) in one of our products. The method we are using was developed by a local university and uses a C8, 5 micron particle size, 100 angstrom pore size column with a chloroform/acetonitrile gradient. We are currently in the process of transferring it in house and are having some problems, these being variations in peak area with time. The peak area can be constant for some time and then at some point decreases. It appears to be related to the column age. Column cleanup results in an increase in area but not to original levels and requires several hours. We have tried a different column brand and seen the reverse. Because we have had very little experience with ELSD, we are not sure whether our problems are related to the ELSD or the column or both. It appears to be column? We have also never analysed PDMS in this way before. Are we using the right type of column. Literature i have read suggests that a larger pore size is required as molecular weight increases. Does anyone have any ideas or helpful hints relating to both ELSD and appropriate column requirements?

I would have probably done this analysis with a size exclusion column and refractive index detection. This is a vastly more robust approach(if it worked for your application) The response of ELSD is not a linear function of concentration so if the peak shape changes the response can change. However, I do not know if this factor is the source of your problem.

Could you tell us more about the sample as well as what the data are used for. Do you just need total PDMS, or are you trying to characterize the MW distribution?

Ask your supplier of PDMS for help on a method.

To determine if the column is the problem, remove it from the system, and see if the response is stable in flow injection mode. It could also be a volatility issue. Siloxanes, especially the low MW oligomers, have significant vapor pressure, and evaporation at high temperatures (>50C) could be causing you to lose sample to evaporation. Reduce the evaporation temperature 5-10 degrees. If volatility is an issue, your response should increase as temperature decreases.

Bill,

The sample we are testing is an antacid liquid with PDMS (simethicone) as one of the actives. The PDMS is extracted using chloroform, the chloroform dried with sodium sulphate, filtered then injected. We measure the total PDMS. The data will be used from stability of the product to release for sale.

I agree with you in that size exclusion would have probably been better. Unfortunately the method was not developed by us and i was not involved initially. Do you think that changing the column possibly to a larger pore size might help?

Con,
I can't offer any advice on doing this separation the way you are doing it as I have no experience with polymers on your column type. If I was going to try another column it would be a size exclusion column. they are so robust and it would be a quick isocratic separation. Call Polymer labs and ask for recommendation. Tell them you want a fast separation.

I probably would have just analyzed the extract for silica, if you have that capability or can send it out to do it.

Idd