Chromatography Forum

Utilize HP1 (I recognize not necessarily the best for these but has worked adequately in the past)
SPME headspace extraction with PDMS

Noticed in a recent project that these compounds were gone.

Have tried:

Removing a loop at the injector port
Removing the column end at the MSD in case any active sites had formed there
Switching to DVB/CAR/PDMS fiber for better extraction of these compounds
Changing liner (use 0.75 mm Supelco)
I had been cleaning and reusing liners but this last one was an old but never used liner I had in a drawer

I’ve been testing with a 100 ppb aqueous solution of these compounds and running headspace. I can’t even see them with SIM. I have ethyl butyrate in the mix and that always comes through fine. Instrument is a 7890A/5975C

Any suggestions would be appreciated

Try to prepare new fresh solution of compounds (1ppm etc.)

In the past, did you get good signal for your target molecules with 100% PDMS? Have you tried adding some salt to drive the phenols into the headspace? In my experience, it's not the first fiber I would choose for most things - especially at 100 ppb or less. In fact, it's the fiber I choose when I need less sensitivity. The "fiber-of-choice" for me for the vast majority of things is the carboxen/pdms/divinylbenzene-2cm phase. I am very surprised that didn't help you.

A colleague of mine had trouble with determining vanillin in a substantially aqueous matrix in the past. The phenol and aldehyde groups on the ring make it want to stay in the water much more than expected - especially when you consider there's a methoxy group on the ring as well. The need for the analysis went away before she could fully flush out a method for it. Maybe I'll give it another shot just to see what I can find out. I get very good results for other alkyl phenols (4-vinylphenol, 4-vinylguaiacol, 4-ethylphenol, 4-ethylguaiacol, cresols, etc.) in a substantially water matrix using cx/pdms/dvb-2cm.

I'd imagine a fiber with DVB would be better for vanillin as well as a db-5 column. Vanillins come off a moderately high temperature in the mid-high 200 deg C's. Could be the SPME fiber is going bad or there could be inlet or column activity (aldehydes can be prone to it).

Using the DVB/CAR/PDMS (grey) fiber and switching to immersion rather than headspace got me through this issue for now. Luckily the sample was amenable for immersion this time. I'll probably have to revisit this as that might not always be the case in the future. Thanks for all of the suggestions and help.

I tried heated headspace on a standard that was 136/114 ppb VAN/E-VAN in water. Couldn't get them using PDMS or CX/PDMS/DVB-2cm (60 °C incubation temperature). Tried immersion with a CX/PDMS - couldn't get them with that either. If I heat a few mg of each (neat) in an empty vial, I get them so I know I'd see them with my GCMS conditions. This pretty much confirms the problems my colleague had years ago when she was fooling with this. These are tough!

I tried adding 10% and 20% NaCl to the 136/114 standard and I still couldn't get them up into the headspace for a CX/PDMS/DVB-2cm to trap them (60 °C).

Are you using SIM or Scan. Ultimately though for such a high boiling point analyte a solvent extraction method would be better. Perhaps something like Quechers. I am looking into Agilent's new enhanced matrix removal for getting rid of lipid for my scoville/capsaicin method. The trouble is finding a good internal standard. Codeine is actually good one but given that it is a narcotic I am having trouble with it.

Using immersion and the grey fiber from Supelco I was able to use scan and did not have to resort to SIM