I'm trying to detect low ppb levels of trichloroethylene (TCE) using GC-MS. I know there's a lot of literature out there, but I haven't found super detailed methods for the extraction process. I'm working with a GC-MS and do not have access to a purge-and-trap system or a solid phase extraction (SPE) system. I'm doing what I've read in the literature, but I can barely detect 10 ppm TCE when extracted into hexanes, and I should be able to detect 10 ppb. Based on other tests of other compounds, it doesn't seem to be a major issue with my GC-MS, but rather with my extraction method.
Here's what I'm doing so far for my TCE standards:
I'm using a Hamilton gas-tight (GT) syringe to add neat TCE to a serum vial filled with deionized water and capped with a PTFE/butyl septum and crimped shut with a 20 mm aluminum crimp seal. I let this 1000 mg/L stock solution stir and dissolve for an hour, then prepare other gas-tight vials for diluted stock solutions (50, 10, 1 mg/L). I let those equilibrate for an hour as well. All solutions are prepared on the day of the experiments. All vials are filled with DI water and a stir bar, then sealed, then the TCE (or TCE stock) is injected through the septum. So, the septum is punctured once.
I then prepare GC vials with 750 µL cold (4°C) n-hexane and cap the vials. I extract 750 µL of TCE solution using a GT syringe and add it to the hexanes, the cap. Then I vortex for 1 minute and let the solution settle for 5 minutes. Then I extract the hexane layer into a clean GC vial and cap it. I run this on the GC-MS with a 200°C inlet (split 1:1), column temperature of 35°C, and a run time of 5 minutes, running the MS in Scan and SIM mode. TCE elutes at 3.65 min. But I can't get more than noise on the total ion chromatogram (TIC) for 10 ppm and below, but that should yield a massive peak. As I get towards lower ppb range, that's when the signal-to-noise ratio should start getting worse. I'm stuck right now and would really appreciate any advice!
Hi Jordin,
We need some more information. What column are you using? I have used a GC/MS method similar to USEPA 551.1 for this work about 40 years ago when an MS was very fragile and expensive :-) Full MS scan (45-450 amu) detection limits were better than 10 picograms on column, with Selective Ion Monitoring (SIM) better than 1 pg. Your 10ppm solution (assuming efficient extraction) would put 5 nanograms on column.
A couple of points: the usual solvents for this work are pentane (BP 36C), MTBE (BP 55C), diethyl ether (BP 35C). Your hexane solvent has a BP of 69C which is close to TCE (87C). I'll make a guess that you are using a 25m x 0.25mm ID x 0.25µm BP5 column, which with a 1:1 split (or near "splitless") injection which will have a problem separating your TCE from the end of the solvent peak - If this is so, the instrument will turn its sensitivity down to protect itself from the large amount of hexane that may be still eluting with the TCE . A simple test would be to inject your 10ppm solution split 20:1 and look for the masses 130 and 132. The hexane peak will be much narrower, so it should be separated from the TCE - You are injecting "only" 500 picograms of TCE which the instrument will easily see (as I said you should be able to see
Hi Tim,
Thank you so much for your answer! The current column is an Agilent HP-5MS, 30mx0.250mm, 0.25 micron. It's not BP5, as other instrument users like this HP-5MS, but I could swap columns if I need to. I have been doing a 1:1 split, and didn't know that this could have a problem with separating out hexanes. I'll try your suggested test with a 20:1 split and SIM for 130 and 132, running for 20 min at 35C. I'll also try running with n-pentane, as I have some in my lab.
This is really helpful and I'm excited to try using pentane with your suggestions tomorrow. I'll also explore EPA 551.1--thanks for attaching it! I really appreciate your advice, and will report back how it goes.
Cheers!
Jordin
The column should be fine - The separation of a 5MS column has been designed to be almost identical to a BP5 - The "5" is historical and indicates that the stationary phase has the same polarity as a poly(95% dimethylsiloxane; 5% diphenylsiloxane) copolymer. The 5% phenyl groups in DB5 are attached to silicon atoms directly (Ph-Si-Ph), the methyl groups are similar (Me-Si-Me) and the silicone backbone has the structure -O-Si-O-Si-O-Si- The more stable 5MS stationary phase has some phenyl groups in a silphenylene backbone -O-Si-Ph-Si-O-Si-O-Si-
Similarly the more polar 17MS and 35MS phases are designed to be almost identical to BP17 and BP35, which contain 17% and 35% phenyl respectively.
Cheers, Tim
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts