pesticides

[pueblito]

Dear colleagues.
I´m trying to implement the analysis of organochlorine pesticides ( OCP) in a GC-MS sistem. but I have a problem.
I check the sistem with a mix of DFTPP, pentachlorophenol, benzidine and DDT. the sistem´s espectral response to DFTPP is in good agreement with EPA requirements. this is the same for pentachlorophenol and benzidine ( tailing factor)and DDT ( breakdown).
also I perform the 8270 SPCC with a minimun response factor of 0.15.
when I perform the calibration curve ( from 0.02 to 0.2 mg/l. And 0.05mg/l for internal standard ) for OCP I get a % RSD of Respose factor in the worst case of 20% ( always for methoxichlor).
but, when I analise OCP extracted (3510C) from desionized and river spiked water ( 0.1mg/l) I get erratic resposes for:
Dieldrin, Endrin, Endosulfan sulfate, DDD,DDT and methoxiclor ( the worst) they give a result of sometimes twice the spiked value( 0.2 rather than 0.1mg/l) and always a positive error !!! and this happens sometimes for a direct standard.
well, why does it happen? and why for some OCP ?( this do not occur for the BHC´s in the same standard or THM for a VOCOL column)
I use a four years old DB-5MS 30 x 0.25 x 0.25 column.
thank you very much for your answers.

[MarkRob]

This may not be the answer you're looking for but.

we've been doing OC Pest analyses for years and I tried to use epa 8270 for OC Pesticides and ran into trouble as well.

I find that I get incredible sensitivity and reproducability (RPD's less than 0.05 using new columns) on a dual ecd system using a column Y splitter and dual column coformation. This set-up is much cheaper than a GC-MSD system.

EPA 8270 (my least favorite method) has so many problems associated with it, I only use it when I absolutely have to. I could never achieve the sensitivity I need. I never tried to use SIM mode but I've realized that I dont need to use 8270 so I decided not to give myself the extra headaches. You may want to offer OC Pest by dual ECD (EPA 8081) with EPA 8270 confirmation.

What I've learned about my attempt to use 8270 for OC Pest is that although an MSD detector may offer abosolute identification, it is not always the best way to analyze specific types of compounds.

I welcome any comments.

[sassman]

How do you extract the samples? If you use SPE, do you dry the cartridge for a few minutes before eluting the analytes? Are the extracted samples in the exact same matrix as your standards? I am thinking that this may have something with a difference in your matrix (especially if the extracted sample might contain water). Of course check for calculation errors.

[pueblito]

Thats true, 8270 is very difficult.

[Bigbear]

I have found that the HP-5 works better than the DB-5 for 8270 compounds I can not explain it but it does. My columns usually die after a year. A 4 year old column may be part of your problem. Check the resolution for the following pairs: Benz[a]anthracene : Chrysene ( valley between should be>25% of peak height) , and Phenantherne : Anthracene ( should be baseline resolved). If not try some injection port maintenance and clip the column head, or change columns.

[bhuva31]

Regarding organochlorine pesticides, I prefer HP-5MS column using liner without glass wool to avoid degradation of pesticides.
Whatever may be the instrument senstivity, the sample preparation plays very crucial role. I prefer matrix spiking even in worst case matrix matching calibration to achieve results. This acceptable by CODEX ALIMENTARIUS COMMISSION also.
Cleanliness of liner, Trimming the injector end and column conditioning fresh calibration standards are very important along with tuned mass spectrometer to attain better results.

Happy Analytical days