I derivatise aqueous samples containting sugars for GC-MS analysis by acetylating the hydroxyle groups of the sugars (forming alditol acetates). I've tested this and it works just fine. Recently, the GC-MS instrument has experienced some problems and we suspect it could be caused by samples which contain phosphates. Most of my samples don't containt phosphates but I add phosphates to some of them.

The question is: is it possible for inorganic orthophosphate and organic phosphates (e.g. acetyl phosphate) to be present in my final samples (after they have been derivatised)?

As I mentioned, I derivatise my sugars by reacting them with sodium borohydride (to reduce the keto/aldehyde groups into hydroxyles) and then I acetylate all hydroxyles using acetic anhydride. The final step is a liquid-liquid extraction where I add DCM to the aqueous mixtures; I recover the DCM phase after vigorous mixing and that's what I inject into the GCMS. So, even if I had organic/inorganic phosphates in my initial samples, I reckon they would tend to accumulate in the aqueous phase of the liquid-liquid extraction (which I discard).

It is possible at all that the phosphates are somehow 'choosing' to move into the DCM phase (where my analytes end up) insted of the aqueous one? In my opinion: no, as phosphates are very water-soluble. But I'd appreciate some feedback.

Thanks!!