Chromatography Forum

For my thesis, I am trying to analyze water samples for haloacetic acid concentrations.

I've got the methylation/extraction portion down - I am having trouble with developing a method. I've read up on the most recent EPA 552.3 method, as well as incorporated UMass' method (it's easier to follow and more concise). I am currently using an RTX-5M column, but both procedures call for a 1701 due to it's increased polarity. My concern is mainly because my surrogate standard and internal standards are not showing up in my 'zero' concentration calib. standard. I still have to run my other extracts that actually contain the HAA - it's also a pain to try to get Shimadzu Support to come down to school to help due to limited availability. A collegue of mine is analyzing trihalomethanes, and his peaks showed up no problem (he is also using a purge and trap).

Should I bother springing for the 1701 column? Or is this a problem with technique? Any suggestions? Please?

I have run the 552.2 method with a DB-5MS column with no major problems. There was a splitter after the inlet and a DB-XLB column in parallel so I don't remember if all there were any resolution problems. When running that method I advocate checking the extract pH before analysis. I learned that one the hard way.
If you are missing the trichloropropane peak you have a bigger issue than not enough column polarity. What is a "zero concentration" calibration standard used for?

Ahh I didn't even think of that - my other concern was that when preparing my calibration standards, my glassware was not 'clean' enough and thus there is some sort of interference/degredation happening. I am going to attempt to prepare them over and I'll be sure to keep an eye on the pH.

My "zero calibration standard" is a zero concentration of HAAs - it's just reagent water - I suppose you could call it a blank? I still take it through the extraction/derivatization process since that is what the method calls for. Thus, I am still adding my internal standard and surrogate standard to my calibration standard that technically shouldn't have any concentration of HAAs in it (I hope that makes sense).

Would my IS and/or SS not show up if the sample did not contain any of the analytes?

byla,

The -5 or 1701 is probably not your problem, especially since you are using MS (coelutions not such an issue) unless the IS/SS is so poorly retained that it is coming out before the MS turns on.

More likely extraction/derivatization or concentration issues. I would start by dervatizing just the standard and see what I get. I would also ask you for more details about your concentrations (especially since you are using MS.)

Best regards,

AICMM

AICMM -

I agree, I think it may have to do with my extraction/derivatization process. The concentrations I'm using for my calibration standards are 0, 25, 50, 100, 150, 200, 350, and 500 ppb. The concentration of my IS is 1.5 ppm and my SS is 1 ppm.

I will be re-making my standard solutions and calibration standards today - as I think more about my process, I don't think I was careful enough in injecting my analytes in reagent water. I noticed there was some turbulence (for lack of a better word) on the sides of the vial as I was making the standards - possible volatile escape?

Also, I know this is a rookie question, but I just want to be sure: when separating my aqueous and organic phases, the top 'layer' is my MTBE extract, right? I'm referring to pg 28-29, Section 11.2, of the EPA method here: http://www.epa.gov/ogwdw/methods/pdfs/m ... t552_3.pdf

I appreciate your feedback!!! This is all a great help!

byla,

Yes the top is MTBE.

When you speak of 25, 50 ppb and 1.5 and 1 ppm, do you mean in the water or do you mean at the instrument. Let me put it a different way. At the GC/MS are you shooting 25ng/mL? If you are, then you are only putting about 25 pg on column (assuming splitless and assuming 1 uL) and your MS is not sensitive enough to see this in full scan mode (maybe if you are running NCI...) The HAA methods are built around ECD detector which is very, very sensitive to the right compounds.

Even with 25 ppb at the water, you only get a 1:10 concentration factor so you are only putting on about 250 pg (best case) which is probably still a struggle for most MS.

Start out by making a mix at 25 ppm at the instrument and running that and see what you get.

BTW, your colleague can get away with it because of the purge and trap doing a concentration step.

Best regards,

AICMM

AICMM -

I did not even think of that - I am injecting 1 uL, splitless, but I am not exactly sure the concentration - I would assume it is what you had said (25 ng/mL). I am not sure of the exact specifics of our instrument, but I know that it is definitely not equipped with an ECD. A coworker of mine at work had recommended I make a range of concentrations in the ppm range to see if it would pick it up. However, my advisor is a little reluctant to do this, because he wanted to keep the concetrations as close to EPA MCLs as possible (which for one of the HAA is ~20 ppb) .. but I think I can convince him to take it another route and keep him happy.

I did make new standards last night at 25, 150, 350, and 500 ppb. I think I will make additional standards at 25, 50, and 100 ppm today as you suggested.

Should I also increase the concentration of my IS and SS? My coworker also suggested that these should be around 30-40 ppm each.

- Angela

This topic is a bit old, but I thought I might throw my two cents in. EPA 552.2 differs greatly from EPA 552.3 in respect to how the internal standard 1,2,3-TCP is added - in 552.2, it is added to the final extract; in 552.3, the IS is in the extraction solvent from the beginning. You may want to try adding your IS to a vial of 1-mL MTBE (without going through the extraction procedure) to give a 1 microgram per mL concentration (per 552.3), and 0.25 micrograms per mL (per 552.2). 1,2,3-TCP should elute near the middle of the run, right around BCAA and TCAA.

As for your advisor wanting to stay near the EPA MCLs - that's fine when you know where compounds elute and have optimized your chromatography and detection paramaters. However, when you are just starting out, shooting a 10 ppm standard on the MS should give you good confidence in identifying the correct RT, quant and qual ions, and will give you a rough approximation of how low you can go on the MS via the signal to noise ratio.

For reference, I run EPA 552.2, and my calibration curve ranges from 10-500 micrograms per liter (ppb) using an ECD; each point is extracted with each extraction batch, to correct for inefficiencies in methylation and extraction. I have a Varian 3400 and a Varian 3600, and am running an Agilent DB-XLB with either a Restek Stx-CLPesticides or Stx-CLPesticides2.