552.3 Method Development

I am currently developing a method for using EPA Method 552.3 on our 8890 w/uECD detector using He carrier gas and N2 make-up gas. My predecessor settled on using the columns from Agilent's App Note (DB-35ms UI and DB-XLB). The main thing I'm trying to determine is which type of liner would be best.

The EPA method was developed (in 2003) using 2-mm straight quartz liners. The App Note (which used SPE for pre-concentration that I won't be doing) used 4-mm Helix double taper liners. I've tried to reference others' methods online and found one that uses a Shimadzu GC with H2 carrier gas and used a Restek 3.5-mm single taper w/wool topaz liner.

There are several liners already in the lab (5190-2293 UI single taper w/wool), but I don't know if these were ordered by my predecessor or just offered as samples by Agilent. From what I've read, using liners with wool would just be offering unnecessary filtering and be adding activation sites, making it better to have no wool. So I assume I shouldn't use these liners.

The cost difference between the various Agilent 2-mm straight quartz liners and the 4-mm helix double taper liner are fairly large (4-5x higher for the double taper). Once my method has been developed and verified, I would likely only be running samples once a month, the run be mostly calibration standards and QC samples with up to 10 actual samples (all of which would be finished drinking water).

Is there any reason that I should go with the extra expense of the double taper liners over the straight liners that the 20 year old EPA method used? The App Note (that I've only skimmed so far) doesn't seem to offer any explanation of why they chose these liners except to stress the importance of inertness.

Additionally, the three different splitless 2-mm straight liners I am looking at actually decrease in price with increased deactivation/inertness. Does that make sense? Because of that, my current thought is to go with the 5190-6168 UI splitless 2-mm straight liners, but the webstore description says "for use with HS Transferline". I would still be okay to use them, right?

Thanks in advance for any assistance.

Parents
  • Hello all!

    I am currently also developing EPA 552.3, and have a question about the method.  This was the only forum that popped up when i searched this method. 

    The question i have is about doing a dilution, in the method it states that I can dilute the extract with the MTBE with the internal standard. When I run the diluted sample how would I account for the increased concentration of the internal standard? Or would the increased concentration not even make a difference if for example I do a 1:3 dilution with a final volume of 0.6mL?

    Any help with this will be appreciated!

  • An internal standard is always added prior to sample preparation, so that it undergoes the same extraction/preparation steps than the analytes. This normally means that after sample preparation you also lost a little ISTD. If that would not happen you could use an external standard for calibration.

    BTW: an internal standard therefore should be structurally and chemically similar to the analytes (textbook talk)

  • This question is method specific to EPA 552.3.  In the method there is a solution prepared that is MTBE made with the internal standard.  MTBE is the main solvent in the method and is used for the liquid-liquid extraction of the HAAs.  So, after the extraction is done and the sample is ran on the GC, if the sample needs a dilution the method states to use the MTBE with the internal standard solution to do a dilution with the extract.  My question is if I do that wouldn't the concentration of the Internal Standard increase and if so how would I be able to correct for that within the data analysis.

  • Yes, it would and it is totally illogical to do so. But I checked the method and could not find this. Which chapter/paragraph to you refer to?

  • The part in the method that refers to the dilution is in EPA Method 552.3 section 11.4.4

  • Thanks, I see. It is analytically wrong and your error bars will increase, better have another calibration point. But anyway, it's official and you can just follow the method. Seemingly they assume that nothing happens to the IS during sample preparation, but then an external standard calibration would also work.

    If you get an explanation from them, please post it here. TIA

Reply
  • Thanks, I see. It is analytically wrong and your error bars will increase, better have another calibration point. But anyway, it's official and you can just follow the method. Seemingly they assume that nothing happens to the IS during sample preparation, but then an external standard calibration would also work.

    If you get an explanation from them, please post it here. TIA

Children
No Data
Was this helpful?