MtBE and its degradation products in water samples

[hplc.chem]

Dear all,

I am trying to find the right method for the analysis of MtBE and its degradation products in water samples. After searching the web, I noticed that most MtBE analyses are done by three techniques using GC:

1- Purge and Trap
2- Headspace
3- Direct Aqueous Injection

My problem is that I don’t have enough fund to make major modifications of our GC-MS. We have a Varian Saturn GC-MS (Ion Trap) with a manual injection system (1177 Injector).
The analysis goal is to identify and quantify MtBE and its degradation products.

Is there any SIMPLE method we can use for this purpose without modifying the GC-MS?
Do we need to get accessories for the GC-MS to use the above three methods?

Many thanks in advance.

[rb6banjo]

Only a minor modification to the injection port (to a narrow-bore liner) is all that's needed for SPME determination of low-concentrations of MTBE. See the Supelco SPME Applications guide:

http://www.sigmaaldrich.com/etc/mediali ... p/8652.pdf

[Don_Hilton]

Purge and trap takes a purge and trap sampler.

Headspace depends on the technique. It can be as simple as drawing a sample of headspace vapor from a capped headspace vial and injecting into the GC and can be done manally. Other variations can involve special samplers and some cost.

Injection of liquid water can be done - just like any other liquid sample, but injection of water has some issues related to degredation of some types of columns and transfer of inorganic residues to the GC inlet. There are numerous posts on this site discussing injection of water.

rb6banjo's suggestion is a good one. If you know which degredation products you are looking for, you can pick specific fibers to extract those components from water and it is failry straightforward.

[hplc.chem]

Thanks rb6banjo and Don_Hilton for the informative replies.

Are the FOUR methods similar in their quantification reproducibility, accuracy and ease of use?

[Don_Hilton]

I hate to sound like a lawyer - but the answer is: it all depends.

If you are looking to run only a few samples, you can work something out with liquid injections or SPME even before a purge and trap or a complex headspace sampler can be purchased, delivered, and installed.

Liquid injections typcially work well - if analytes are in the right range and the samples will not damage the column - with the atuosampler that is frequently present with a GC already in the lab. And, if there is no autosampler, manual injections work well - as long as you can stand sitting in front of the instrument to make them.

SPME works with some autosamplers or as a manual injection. If your autosampler is based on the CTS rail, there are some parts that can be added to handle SPME. For other autosamplers, you have to check the vendor literature. The CTS rail systems can also perform headspace sampling for gas vollumes of a mL or more from a headspace vial.

If you are looking at low concentrations, you may need to set up and use a purge and trap sampler - depending on the specific degredation products.

There are many options. And all work well for parirticular applications. The question is in the specifics of what your analysis is trying to do and what is the best way to get there.

[James_Ball]

I hate to sound like a lawyer - but the answer is: it all depends.

If you are looking to run only a few samples, you can work something out with liquid injections or SPME even before a purge and trap or a complex headspace sampler can be purchased, delivered, and installed.

Liquid injections typcially work well - if analytes are in the right range and the samples will not damage the column - with the atuosampler that is frequently present with a GC already in the lab. And, if there is no autosampler, manual injections work well - as long as you can stand sitting in front of the instrument to make them.

SPME works with some autosamplers or as a manual injection. If your autosampler is based on the CTS rail, there are some parts that can be added to handle SPME. For other autosamplers, you have to check the vendor literature. The CTS rail systems can also perform headspace sampling for gas vollumes of a mL or more from a headspace vial.

If you are looking at low concentrations, you may need to set up and use a purge and trap sampler - depending on the specific degredation products.

There are many options. And all work well for parirticular applications. The question is in the specifics of what your analysis is trying to do and what is the best way to get there.

Don is corrrect there. If you are looking for low ppb levels you will need the purge and trap or the SPME methods. Headspace can get to those low concentrations but mostly if you are using an head space autosampler that can take large volumes of headspace, the 1ml sample size will probably not get you to those low limits. Direct aqueous injections should be held to 1 or maybe 2 microliters so you will be more in the ppm sensitivity range. Also check to see how the water will affect your MS. I have never used and ion trap but I seem to recall that they do not like water at all.

[hplc.chem]

Many thanks for all of you.

The concntrations should be high enough as the analysis is for degradation products after catalytic procedure.

The column I have is CP SIL-8 CB and I have no Autosampler.

Does this information alter some suggestions?

[Don_Hilton]

It depends on what concentrations you are working at and if you know the degredation products you are looking for. And, it helps if you know what concentrations you are looking for. If you want to use the catalist to degreade MTBE at part per million levels, you are not sure what the products might be because there are other things in the water - and you need to account for 99.9% of the MTBE originally in the water - you have a nearly impossible problem. If you are looking at 25% MTBE in water and have expected products, like CH2C(CH3)2, HOCH3, and HOC(CH3)3 and you need to account for 90% of the MTBE added - an easier problem, but a 5% phenlymethylsilicone column is not what you want for the job.

And if you have a more agressive catylist, you are looking at things like CO, H2CO, CO2, H2CO2 and some other small molecules. I would look for a chromatographic column appropriate to your expected reaction products.

If your expected products are the alcohols and isobutene at percent levels in the solution, I would go for headspace injection with a GC syringe. (That would eliminate a lot of the water.) And a manual injection of 10 or 25 microliters from the headspace above several mL in a 2 cc headspace vial may be all you need.

If your catylist is something in solution - like a strong acid or a metal salt, you do not want to inject that into a GC. Anything that promotes a chemical reaction is likely to promote degredation of the GC column - and perhaps even damage the instrument.