Chromatography Forum

Looking for advice. I am starting to layout and setup an LC-GC-MS system. Ideally we would use normal phase LC to interface to the GC (cmpds are not water soluble). We would like to prefocus onto the LC column as much as possible and likely take cuts thru out the LC separtion to the GC. The challenge arises because the samples are received in acetone. I havent done a lot of normal phase separations, but I am assuming that the cmpds will come out with the solvent front, or nearly, in a classical normal phase, hexane MP silica support, separation. I am looking for help or options on obtaining the "best" normal phase separation. My thoughts to try are: a size exclusion type separation to switch the sample solvent to a hexane mixture (assuming all cmpds are miscible), subambient LC? to lower strenght, small injection into a large column to "mix" the acetone away.....Any advice would be greatly appreciated. The other challenge is that we will have about ten cmpds to target, spanning polarity and volatility.

Thanks
Jon

Another option might be the at-column dilution technique, but you need to play around with the instrument set-up. Effectively, you inject the sample in acetone, dilute it post-injection and before the column with a stream of hexane, usually in a ratio of 10 or 20 to 1. I have never played with this trick in normal-phase chromatography, but we have used it a lot in RP, especially in prep RP.

It depends on how much acetone you intend to inject. If it is only 5-10 µL, the only bad effect will be somewhat broad peaks early in the chromatogram. Lately, I have been injecting samples in acetone or ethyl acetate on an RP column with >50% organic mobile phase. The peak distortion is noticeable, but not bad enough to ruin the chromatogram.

Years ago I used acetone as sample solvent to align with a supplier's GC extraction; I found if I kept the injection volume low enough that the acetone didn't cause broadening. So you may want to try smaller injection volumes to see if that helps the peak shapes, not a bad idea for any assay.