Chromatography Forum

Hello-

This chromatogram is a zoom in overlay is of a waste sample (blue) at a 400X dilution. The green is a 10 ppm Aroclor 1254 standard.



Would you dilute this further? Or would you say that this is conclusive that there is no Aroclor 1254 present?

I'm just trying to get some feedback from analysts with similar or more experience than me.

Thanks,
John

NOTE: The sample was acid-washed and treated with a Silica gel cartridge.

I just ascertained that the two large peaks at about 10.5 min in the chromatogram are alpha-Chlordane and gamma-Chlordane.

I thought that silica gel cartridges would remove chlorinated pesticides. Is the concentration of chlordane so high that it oversaturated the column?

Upon further investigation, heptachlor and 4,4'-DDT are also present in this silica treated sample extract.

Is there a way I can treat extracts that are high in chlordane so I don't end up with a large elevated detection limit?

Thanks,
John

No surrogates to gauge retention time shifts between the two runs?

For me, if the surrogates lined up, I would lean towards a non-detect, but I would also run it on a different column to confirm it.

I would also ask the project manager or whoever is running the show, what is the action level on this sample. That would drive my decision to dilute or not.

I only have <3 yrs experience with PCBs though.

Wow, a whole year has passed since my original post.

Still having issues with chlordane interfering with my PCB analysis. Sulfuric acid, digestion, silica gel and florisil don't seem to help. KMnO4 makes my extracts even more of a mess - my extracts become darker.

I contacted Restek tech support about this and still have yet to hear of any suggestions that could work. They suggested that a sulfuric wash should get rid of any pesticides, chlordane included. It doesn't.

Any suggestions?

John

PS> Let me know if I should re-upload a pic, don't know why the original pic above was deleted.

are these samples you're having problems with from the same site?

could you upload the photo again if you get a chance?

If the peaks interfere with the areas that you're evaluating for pcbs then you can't really do much other than dilute. Is it possible that you have sulfur contamination instead of the chlordanes?

Sorry for the delay -

Here are some pics from routine sample extracts (extracted with hexane, washed with H2SO4).

Sample (blue) overlaid with a 10 ppm Aroclor 1016/1260 standard (green):


The same sample at a 4X dilution after silica gel cartridge treatment (blue):


The treated sample (blue) overlaid with a 1 ppm pesticide standard (green):


The treated sample (green) overlaid with a Florisil treated portion of the same sample (blue):


This a routine type of sample for me - no one I've spoken to seems to have ever run into these types of samples.

As seen in the chromatogram with the sample and the pesticide standard, it's heptachlor, delta-BHC, gamma-Chlordane, and alpha-Chlordane which are unaffected by any clean-up method. Large amounts of these pesticides can force me to report elevated detection limits, which can cause problems.

Is there any way that these pesticides can be removed? I've not read anything stating so. And any technical support or applications chemist has not been able to offer any help past what I'm doing already.

Basically, no one has been able to offer suggestions other than "Use more Florisil."

Any ideas or suggestions?

Note: Potassium permanganate treatment actually makes my extracts darker and more difficult to see the separation.

Thanks,
John

Hi John,

I have tried hydroxylated polystyrene divenyl benzene SPE material (Isolute ENV+) for PCB/PAH/OCP clean up. This material has got affinity towards compounds with aromatic structures (like PCB, PAH etc). Using hexane as elution solvent, you would probably able to remove most of the unwanted compounds from your extract. See the pdf for some general idea about the separation mechanism in these types of polymeric materials.

http://www.chromatographyonline.com/lcg ... rticle.pdf

Good luck!

Praveen

Is there any way you can choose different peaks to quantitate your pcb? I know it's not ideal but if you're absolutely convinced that that aroclor is in there and there is other stuff giving you results of a high bias you can take those peaks out. I think you can generally quantitate with a minimum of 3 peaks. I really have no other thoughts on how to get rid of those extra peaks and I don't think it looks like Sulfur but you could always do a copper clean up to make sure?

These types of samples are actually fairly common - automobile waste is notorius - I have not seen clean up techniques that can routinely clear this stuff out, and the ECD is just not selective enough to only see the peaks of interest. If you use a confirmation column, you will have some luck ruling out suspect peaks. It helps you rule out an arochlor pattern if you can demonstrate it is not the same peak in a confirmation run. You may still be stuck with elevating the reporting limit, depending on the lab policy.

At some labs I have worked at, I would put together a selected ion method on a mass spectrometer - you can get comparable detection. This is only occasionally useful - the MS at unit mass resolution is often not selective enough to miss the coeluting peaks - better with some matrices, worse with others.

Tim

@pkutty: Thanks for the info, I contacted Biotage and had a pretty lengthy discussion with one of the staff chemists about this matter. The Isolute ENV cartridges are on the way!

@8jw8: I normally calibrate with the 4 most prominent peaks - sometimes the amount of pesticides in my sample completely saturate the detector. I've diluted up to 8000X for one particularly nasty sample.

@Yama001: Thanks for noting that these types of samples are common - I've not received any comments like that from any other applications chemist/technical support person that I've spoken to.

After emailing the above chromatograms to one company's tech support group, the reply was something like, "Sorry we can't help you further, as we don't really deal with real world samples too often in our company..."

Thanks again, your replies are greatly appreciated!

John