Suppose that one is following a reaction in DMF solvent by TLC. Unless one removes it, the DMF would be expected to increase the Rf values of reactants and products, at least when the running solvent is less polar than DMF. Is there a convenient way to remove the DMF in such an instance? Thinking out loud, one might try a micro-extraction. One could heat the TLC plate to drive off the DMF so long as the product was not heat-sensitive.
My reason for asking is that we were reacting a UV-active thiol with a diethyl phosphonate bearing a tosylate group, produce a UV-active thiother. We ran the reaction for longer than the protocol indicated, and the TLC suggested that the starting material disappeared and that the product had a higher Rf value. I think that this was a misleading result.* We are working the reaction up, and two lines of evidence suggest that the reaction proceeded less than 50% to completion: mass of the crude product and TLC of the post-extraction material shows a significant amount of starting material. If we run the reaction again, we can use the DTNB test (Ellman's reagent) to look for loss of thiol (this can be done quantitatively). We can also use P-31 NMR to monitor the reaction (I wish we had done so for the reaction just completed). However, other possible reactions will not have these alternatives.
*I think that the DMF pushed the starting material and product together and we did not differentiate between them.