In Acta Chemica Scandinavia B 39 (1985) 79-91, lead author Rolf Carlson has made some suggestions for choosing a solvent for a reaction. He feels an important consideration will be to be sure that a sufficiently wide range of properties is explored. At the same time, he recognizes that solvents that have already been used successfully in the literature for similar or analogous reactions or those selected in the literature for reactions that seem likely to follow similar mechanisms will be attractive. Our experience also suggests that solvents that won’t dissolve the reactants usually fail. In most cases the liquid medium needs to be a single fluid.
His proposal produces a two-dimensional map using Eigenvector projections of the solvent descriptor space. I do not adequately know what that means and I don’t suppose it matters whether you do either. What matters is that if you choose one solvent from the central regions of each quadrant, such as chlorobenzene, diisopropyl ether, nitromethane and N-methyl-2-pyrollidone, there will be less probability that you will miss out on being directed towards an especially advantageous but less-obvious solvent choice. If it is obvious that a particular solvent picked from one or more of the quadrants will be unsuitable, choosing another from the same region can still maintain a diversity of solvent properties.
No comments:
Post a Comment