Ethylene glycol analysis

<p><span>Hi! I am curious to see if anyone out there has attempted to run ethylene glycol on the UPLC-MS as opposed to GC-MS. It's something my lab is looking into developing because we don't have a GC-MS. Does anyone have any ideas for a type of column to use or any other insight?</span></p><p></p><p><span>Thanks!</span></p><p><span>Kate</span></p>


  • Hello

    I think that you need to be careful as glycol can be a source of contamination for the entire LC MS system after you have introduced once, so be sure to run in low concentrations. I will ask our chemists about their experience. But there may be others on the community that can help too.


    Edited by: Elizabeth Hodgdon on Dec 5, 2008

  • Jim


    We have done this analysis with an ion-exclusion column, water as the mobile phase and refractive index detection (all HPLC mode). You can run this on your UPLC system as an HPLC analysis. I know this doesn't help you if you wanted to run at high linear velocities with a sub-2 micron column. However, it is an option.



  • Thanks, Liz, that's helpful to know. We are hoping to test serum samples for ethylene glycol (caused by consumption of anti-freeze). We've only done a little research on the possible method development. We haven't come across too many papers about it, though, so we are just trying to see if it's even feasible.

  • Kate:

    Here is first response from internal chemist...

    "along with being tough to retain (but not impossible) ethylene glycol has too low of a molecular wt ( 62.1) to offer any significant sensitivity in LC/MS, its going to get lost in the chemical noise. GCMS has much lower chemical noise (H2 and He carrier gasses don't contribute much). Polymeric forms (PEG) are easily seen in ES+ and easy to separate using simple gradients and C18 columns (probably the most common contaminant we see). And, you probably do not need GC/MS anyway, GC FID likely would work well enough".

    I shall gather more.


    Edited by: Elizabeth Hodgdon on Dec 6, 2008

  • Thanks, Jim! That's really helpful! :)

  • I have tried but could not get it to ionize. Whatever the reason. GCFID is the most common and easiest way to perform and is how we analyse it in our lab. Also way cheaper by GC. I wanted to try it on the LCMS to free up our GCFID since we also use it for NEV's. But no luck for me.


  • There are a few things to keep in mind if you do give ethylene glycol another chance by GC/MS. First, a moderate to high polarity column is needed to get sufficient retention. If all you have is a low polarity column like a 1 or a 5 then be careful of losing the peak in the solvent delay. Next, since the majority of the signal in the EI spectrum of ethylene glycol is between 29 and 33Da it is important to make sure the scan range includes this part of the mass scale. Its not unusual for EI scan methods to start at a low mass of 40 to 50Da. If you did that in this case you'd miss the detection of 80 to 90% of the available signal which would reduce the apparent response by at least 5 to 10X over what could be achieved by having the acquisition mass range start at 25Da. By starting at a low mass like 25Da you will start to see the air peaks (28/32) in the spectrum but they should be able to be subtracted out. I have attached the US FDA GC/MS method for detection of ethylene glycol in toothpaste for reference. This should help give you an idea of what a reasonable limit of detection should be. When looking at the example chromatograms in the document note that they are using a 20:1 split on a 1uL injection if you are trying to compare the response you've achieved.


    Doug S.