Chromatography Forum

Hi, we run Micro liquid-liquid extraction GC method for acidic herbicide compounds in drinking water. We have ongoing problem with Dinoseb recovery. It’s random and we're trying to figure out what could be a problem. Is anybody else running this methods and has similar problems?

Dinoseb often doesn't behave well in that method. The nitro groups make it prone to loss. Unfortunately herbicide methods include a wide enough variety of compounds that the conditions are a compromise.

Thank you Steve.
I'm new to this method. I haven’t run it myself yet. I've been asked to "solve" this problem. It would be OK if recoveries were consistently low but they are all over the place. I was wondering if there is some "bench knowledge" I could use to start solve this mystery.

Sorry, I don't have the reference handy, but I have seen a paper where dinoseb losses are explained by conversion to a quinone form. Dinoseb is sensitive to conditions in the hydrolysis step which can result in low or no recovery. Also the methylation time may be too short under some conditions and that can result in low recoveries of the acids. Overall, these herbicide methods, I'll include 515.x and 8151, are a compromise. If you look at acceptance ranges for PT samples you will see for some compounds almost any reported amount is OK.

Thank you Steve. Looks like methylation time could be significant factor. I know there is a big difference in time of processing PT sample and regular batch samples (usually 20 samples).
Do you think you could track the paper you've mentioned?
Sorry to take your time.
Thank you very much for this conversation.

I recall doing 515.1 but not clearly. I do 515.4 which uses diazomethane
to derivitize ( pretty quick and effective).
Dinoseb is the most volatile herbicide we analyize for. I have found that if I am doing a PT that states that the dinoseb will be between 5 and 200, I extract a dilution that puts the maximum in my calibration. I have found that diluting for dinoseb post extraction does not work for me.

If you have not invested much time in developing 515.1 I would suggest
developing 515.4.

Ouch! I just read 515.1 and have done it sort of ( I did the Standard Method version)!
No wonder you are having problems with all the messing around the method entails.
Run don't walk to method 515.4. It is a quite simple "shake " and shoot. You don't use MeCl or ether.
It is as so:
measure 40 ml of sample in a septa vial, add 1 ml 4N NaOH. Leave for an hour with random shaking. Wash with 5 ml 10% MtBE/Hexane, discard organic. Add about 1-1.5 ml H2SO4 ( pH <1) and 16 g Na2SO4 extract with 4 ml MtBe containing internal standard. Transfer organic to clean vial and dry by adding some acidified Na2SO4 leave for at least an hour ( or over nite). Transfer into a clean vial and add 250-300ul diazomethane solution, leave for 30 min. Add silica gel ( 40-60 mesh) to
eliminate un used diazomethane leave for 30 min. Transfer to injection vial and shoot.

Thank you Bigbear for your frank advice. I've already suggested it to the boss. We'll see what happens.

Keep me posted. PM me if you need to.

Good luck.

I failed to mention that some good points to take to your boss are
Lower solvent usage ( difference could pay for development time)
515.4 takes less time

Just an update Bigbear... sorry for the silence. We are in the middle of EPA inspection so I need to wait but I got my boss to agree to run 515.4
Again thanks for the advice

Lucky you, we have our NELAC audit in Sept!!!!!
I'm surre you will like 515.4 lots better than .1..