Chromatography Forum

We had a method in place for detection of a trace analyte in ethanol solution. The matrix has changed in the case of some of the samples to include large amounts of glycerin. Glycerin elutes sufficiently later than the analyte of interest under the conditions. There is no problem separating these two when the glycerin is present at lower levels in the solution. When investigating why our analyte peak disappeared when testing the new matrix we discovered that it is now coeluting with the glycerol at the later retention time and in a manner that matches the broad tailing character of the glycerol solvent peak.

HP1 column - 25meter
IT = 35C, 1 minute
4C/min ramp

Injector temp is set at 320C - pulsed splitless injection 50psi - 0.5 minutes - These conditions are utilized to increase the amount of sample injected due to the trace levels of the analyte.

My thoughts are to vary the initial oven temperature and vary the ramp

Any input is appreciated.

Without looking at a few chromatograms, it is difficult to comment on what to expect. However, can you change to a wax type column? Glycerin is horrible on a methylsillicone column, but behaves fairlly well on wax. If your other analytes will show well on wax it should make life easier for you.

I'd have to agree with Don, even if you can get the separation figured out on a non-polar column like the HP1, you may struggle with peak shape and other problems with something as polar as glycerin. I'd try a more polar column.

Dusty,

If I am not mistaken, there is a post about a short, polar, pre-column used to hold up glycerin enough to effect the separation you are interested. Try searching glycerin or glycerol....

Best regards.

Thank you for the replies - a guard column or switching to a wax column could be an answer however this is one intermittent assay of many and column switching would be very undesirable. My understanding is that with the diffusion pump model I use I would need to vent and repump with either of these switches.

As for the peak shape and chromatographic properties of glycerol on the HP1, these issues would be moot if my analyte was not getting dragged into the glycerol soup by a couple of hundred RI units - originally I believed the detector would be off by the time glycerol elutes. This is the first time I've run into this effect involving a later eluting solvent. I have no trace of the analyte at the standard RI.

Hi Dusty

I have to agree with the earlier posts - you are going to have to change the column. The large amounts of glycerin are producing a short stretch of polar stationary phase that is retaining your analyte. I very much doubt that there is a way of fixing this by changing operating conditions.

So the question becomes how to change columns quickly without venting the MS. You could try one of the Eze Vent, No Vent variants that are available from most of the instrument companies and some of the independents.

Peter

OK - I'm convinced - thanks to everyone for saving me days or even weeks of monkeying around with injector and oven variables. Barring any further posts to the contrary it looks like a new column is needed to break up the love affair my analyte has with the large levels of glycerine present in the sample -

I'm going to try DBWAX as the literature indicates very large RI differences between the analyte and glycerin and ethanol which is another potential factor in the mix - If it works at the sensitivity I need I'll definitely be looking at the no-vent device from agilent, looks like it hooks into the electronics and creates an easy situation.

Dusty,

A thick film guard column at the front of your other column is perfectly do-able without venting if you are a bit efficient. The column acts as a big restrictor which allows lots of people to change septa and liners every day without venting their mass spectrometers. If you have everything there, ferrules, nuts, zero-volume union, column, you certainly could pop a length of pre-column on your current column (and pop it off when it was no longer needed.) You still need to get that other column but you do not necessarily have to change them every time.

Best regards.