Residual methyl alcohol by GCMS

[Pegry]

Hi Everybody,
I have to measure residual methanol (50-100 ppb level) in a polymer: PVP (polyvinylpyrrolidone). Unfortunately I don't have access to head space autosampler.
I think to start with solubilization of PVP in DMF (dimethyl-formamide) or dichloromethane and inject 1uL of solution (splitless mode). I have a WAX column (60m 0.25 I.D., 0.25u) or RTX35 (35% diphenyl 65%polysiloxane) and I want to start at 40°C and slowly heat up to 200°C. I think the best should be working in SIM mode monitoring the 32 (or 31) m/z but my fear is about the noise I get with this low ion.
I have access also to LCMS (ESI or APCI) but I think that GCMS is better in this case.

Have you some suggestions?


Thank you in advance for the help
Have a good day,
Davide

[CE Instruments]

Why not try manual headspace ? Just get a suitable bottle, equilibrate for 30 minutes in a lab oven and sample (using gloves) with a gas tight syringe. Injection will then be split injection with a low split ratio 10:1 and run full scan. Run at high levels until you are sure what peaks come out when and you will also see potential interferences should you need to go to SIM. I would not recommend looking for residual solvents in a polymer by injecting a solution as the polymer itself may not elute from your column ? If you can get a good method going with manual headspace and have lots of samples a new/secondhand autosampler will be cheap compared to the need to change liners/columns when you need splitless to have any chance of running with direct injection

[Pegry]

Thanks for the answer...

I think I couldn't do it by manual headspace bacause I need a good precision (CV%<6) and I have to do a quantitative work.
I think I will try SPME on the polymer dissolved in water..
I have no experience on SPME does anyone tryed it with residual methanol by GCMS??

Thank in advance for the help

Davide

[james little]

I have a friend that does a lot of SPME. He has much better luck as a semiquantitative/qualitative technique.

Might do better by the manual method on precision and accuracy from listening to my friends conversations.

[Okkie]

Thanks for the answer...

I think I couldn't do it by manual headspace bacause I need a good precision (CV%<6) and I have to do a quantitative work.
I think I will try SPME on the polymer dissolved in water..
I have no experience on SPME does anyone tryed it with residual methanol by GCMS??

Thank in advance for the help

Davide

SPME can do the trick as long as you use the same fibre for all analyses.
Or use an internal standard.
I mostly use the "normal" pdms fibre because this works on equilibrium.
If you need higher absorption, use a SPME fibre with carbon, but these fibres are less reproducable.

Why don't you do SPME in headspace?
This keeps the fibre clean. The fibre does concentrate your analyte so I do think you can analyse this.
IF the response is not high enough you could put your sample in a 3-way bottle and lead some nitrogen flow through your sample. Put the fibre in the exit flow stream. This will "empty" your sample and the methanol will be trapped on the fibre. This is almost dynamic headspace trapping.

Okkie

[JI2002]

Another option is Purge & Trap GC/MS, it can give you good precision and accuracy. If you have the instrument, it's worth a try.

[Noser222]

There may be some better methods out there, but this may work for you if you can extract the methanol into water. This method was developed for methyl ester determination in pectins.

Analytical Biochemistry 39, 418-428 (1971)

In short, they oxidize methanol to formaldehyde and then derivitize it with ammonia and 2-4-pentanedione to make 3,5-diacetyl-1,4-dihydro-2,6-dimethylpyridine (M+1 = 194). The original method used absorbance 412 to determine the amount of methanol. Using LC-MS eliminates interferences from other analytes.