I've been hearing a lot about SW-846 8015B/8260-like TPH-GRO by GCMS. Any opinions? What criteria would be used for QA/QC in regards to an environmental analysis? I still think an FID provides a more accurate picture of gasoline range organics in water/soil.

I have been out of the business for a while but this sounds like a waste of money and possibly a worse evil. FID is a detection method that will be sensitive to mass of analyte. MS will at best give results similiar to an FID in total ion current mode and at worst inaccurate results due to differences in ionization efficiencies. Looks to me like someone is not using the expensive toys and need to create a profit center. I now remember why I left environmental for the pharma business.

Having just done some air sample work using an MSD (don't have an FID in the lab) to analyze for 'paint thinner'-class residues in a house, I'd like to comment.

I found that the MSD structural information was useful, since this was a new problem, and allowed us to pick out compounds that were definitely not in the product under scrutiny. I felt I was pretty successful identifying the contribution of the paint thinner to the sample by looking at characteristic ions in standards as opposed to the samples. I amazed myself by later finding references indicating I had indeed ID'd materials common in homes. The quantities of these 'common' materials would not have changed the results significantly, though. The structure info is probably what's behind this, in any case.

But this all took lots of time. Unless this type of data is useful, and necessary (as it was for us)--such as samples that don't fit known patterns--why bother? It certainly wouldn't seem to be necessary for every sample. We certainly don't do MS analysis unless there is an advantage in doing so.

But if there are (lots of) MS systems out there doing nothing but burning electricity, as Anon 6:01 suggests, and the technical problems can be overcome, then why not? A well-done MS method might--and I'm guessing here--push detection limits down, if that's important, by (perhaps) eliminating interferences, while maintaining the basic outline of the method. Thousands of people must be familiar with this analysis, so it might be an easy step. It might be a way to leverage the money for an MS, which is in general more useful than an FID in most environmental work.

Questions back to you, Anon 5:59--wouldn't the 'Performance based' EPA method initiative dictate the QA/QC requirements? And why do you think the FID could provide a more 'accurate picture'? You don't mention that you have tried it, so are you saying you think the existing FID method provides an 'adequate' picture? I don't do EPA methods, and I know just enough to get into trouble. From a general knowledge of the fuel contamination problem, and the typical use of this analysis to address cleanup, I'd be inclined to agree that using MS instead of FID for this analysis is killing flies with a sledgehammer. A very expensive hammer!

You can do an accurate analysis by MS, even though the response factors for compounds will differ more than they will with an FID. The main issue is economics, there is very little money to be made doing this type of analysis, and the price per analysis is so low that a person would rapidly go broke doing it by MS instead of by FID. The states have almost universally settled on FID for this work, so no one is going to do a non-approved analysis unless there are factors such as the analysis done by the last anon.

About EPA Performance based methods, I have heard direct from the director of one of the EPA offices setting methods that use of a MS instead of a conventional detector is far beyond the scope of allowable changes. EPA has separate MS and conventional detector methods and will for the forseeable future.

In response to the last 2 messages... I agree it's a very expensive way to measure TPH-GRO. The only reason it's being considered is because we're running BTEX by GC/MS and, in California, our clients are being told the TPH-GRO can be added at very little (or no) incremental cost.

I agree with the "killing a fly with a sledgehammer" theory but believe the main issue to be sampling costs in regards to the analytical lab business. Its getting kind of cut throat and if by eliminating an extra container and an extra run for BTEX/GRO can save a buck,then labs will promote a method that may not be giving the best picture of a contaminated site to their client. They will, however, not be charging for the GRO run. (FYI - MTBE and Oygenates should be done by GCMS; no arguement here.)

The FID will still provide the best linearity for GRO in regards to contaminated gasoline sites. The data will also fit into the picture better when the site is being characterized for GRO/DRO and the other heavier range petroleum products.

I'd also like to comment on this TPH data usability.

It is very important to remember with all these GRO methods that state method will dictate the result. To include MTBE or not to include MTBE. To stop at C10 or C12. To subtract aromatics and oxygenates from aliphatic ranges?

Where does the madness end?