I am treating a synthetic water sample spiked with Carbamazepine (CBZ) with non-thermal plasma. I am extracting with DCM and using GC/MS in SIM mode for increased sensetivity. CBZ has the following m/z in decreasing intensities 193, 236, 165. I have opted to use 236 for quantitative analysis instead of the base peak of 193. There are compounds with very simmilar chemical structures to CBZ like DiCBZ ( m/z of 193, 254, 210), Dihydrocarbamazepine (m/z of 194, 238, 180) and OXCBZ which have the a base peak of 193 or very close to 193 in the case of Dihydrocarbamazepine.
The reaction pathway initiated by plasma and CBZ is not known (to me at least) and i do not know if the plasma will alter the CBZ into one of these compounds as there is a confirmed reaction of OXCBZ metabolizes to DiCBZ, which can then form CBZ, through dehydration, when exposed to high temperatures. I do not know if the reaction is reversible or not.
Please advise on how scientific and logical this method.
Hello emmanuel,
In the past, I have studied CBZ removal with NTP removal in a DBD reactor. Just as in your case I also used a liquid-liquid extraction procedure with dichloromethane. GC-MS analyses were carried out on a standard HP-5MS column. In particular I found thar CBZ is very susceptible to degradation during injection on the GC column. A very low injection temperature (40 degrees) is typically recommended in order to avoid the decomposition of the compound during injection. The major decomposition product was iminostilbene. I didn’t detect other by-products but I should also note here that I was not particularly looking for by-products.
Detection of CBZ with GC-MS is very challenging due to its low volatility and themolability. In order to avoid thermal degradation as much as possible, you can try a derivation procedure. However this is quite time-consuming, and not always very succesfull. For this reason, I would like to advise you to abandon the idea of analysing CBZ on a GC-MS. Instead, liquid chromatography (LC-MS) should be preferred.
Besides this, it is not straightforward to make an accurate prediction about the by-products formed due to decomposition during plasma treatment. Likely, the type of species, and their concentration in the plasma depends on the type of reactor that is used, the feed gas (do you use oxygen, air, argon or something else to ignite the plasma?) and operational parameters (gas flow rate, water flow rate, power...)
Good luck,
Niels
