Chromatography Forum

Hello,
My wastewater/groundwater lab is trying to start implementing EPA 8270 for semivolatiles. We currently don't do any other organics analysis on GCMS and I have never written a method from EPA before.

I would *love* to us EPA 3511 (microextraction) for sample prep. but I'm having trouble reconciling it with 8270. For example: EPA 3510 concentrates from 1 L to 1 mL extract, but 3511 only concentrates from about 35 mL to a 2 mL extract. This will lower the extracted concentration of the surrogates, spikes, control standards, etc. pretty drastically.

So... I have two questions:
1. What kind of quantification ranges have you found if you've used these methods together? I'm so inexperienced in organics that I can't even decide what range to cover with my calibration curve.
2. Will the lower concentration factor from a microextraction cause me to run into the bottom of my detection limit, and if so, should I up the concentration of surrogate, spikes, etc. to compensate.

I'm just trying to get an idea of what other labs are doing for the practical application of this method--it seems way better than 3510 in so many ways!

It all depends on what regulatory detection and quantification limits do you need to achieve. Our lab does 3510 when a customer absolutely needs the 1000mL to 10mL (100-fold) concentration, but when he can do with a 30mL to 3mL (10-fold) concentration of the sample, we are very happy to use the easier 3511 microextraction.

I'm not 100% sure right now, I think I can check later, but I think our sVOC calibration range is 0.05 to 10ppm.

Another thing you should be vary of is the range of compounds 3511 and 3510 was tested on (by EPA), I think that 3510 was tested on more sVOCs than 3511, and you will need to validate your compounds of interest accordingly, if they are extracted with 3511 as well as they are extracted with 3510.

I'm not sure if I'm permitted to share specific data, but for our uses, 3511 was more than adequate, and to be honest, even better than 3510 with some compounds, although I suspect that this has to do specifically with operator proficiency, as 3510 is way more labor intensive (and prone to sample preparation errors) than 3511.

Thank you, that is very helpful!

I'll have to look for the thread where this was discussed but you can also use a combination of guard column and larger syringe injections to effectively get back that concentration factor.

Here is a very useful post from that other discussion...
viewtopic.php?f=2&t=50914&p=245374&hilit=3511#p245024

You will want to investiagate the link. "A 25ul injection can be done in a normal split/splitless injection port using CSR-LVI."

That was the one I was working with, and hope to get back to developing soon.

If switching from 3510 to 3511 you have to increase the instrument sensitivity to maintain the same reporting limits and detection limits. We are looking into large volume injections and possibly SIM methods to make the switch. If you are doing the full list semivolatiles method 8270 you will need to do both an acidic and a basic pH extraction. We found it is better to do a small volume 3510 extraction with both pH than just the 3511. We also received confirmation from the EPA that if we take samples in 100ml bottles, and do the extraction in the bottle, it is the same thing as doing the extraction in a separatory funnel, you just have to use a pipette to draw the DCM off the bottom of the bottle. You still have to do the extraction in triplicate for both pH ranges, but you still end up with using only about 25ml of extraction solvent which can easily be concentrated to 1ml final volume.

We have been using 3511 for over 25 years. With small modifications, it can be used for a wide range of volatile and semivolatile compounds. With a newer GC/MS system in SIM-SCAN mode, you can easily get sub-ppb RLs without LVI.
See this article:
“Simultaneous Determination of Volatile and Semivolatile Organic Compounds in Soil,” D.M. Mauro, Remediation, Vol. 10, Number 4, Autumn 2000.