Purification of Chemical Products by Treatment with Mixtures of Solid Adsorbants: Identifying Useful Absorbants by a Combinatorial Method

Charcoaling as a means of removing small (<2%) amounts of contaminating substances from an organic preparation has long been known. Use of a plug of a solid adsorbant (such as silica or alumina) through which a solution of the slightly impure substance is passed for the purpose of purification is also long known.

There are other similar techniques. The use of these methods has been empirical, trial and error, one-at-a-time testing. In particular, the selection of a solvent to be used in combination with these solid adsorbants is empirical and this is a very real problem because it is often the combination of the correct choice of solvent and the correct choice of solid adsorbant, which produces the required purification.

Kilomentor now suggests the use of statistical methods and /or combinatorial chemistry to solve the problem. Thus although the problem has been long in existence, such a method has not been described and is not part of the common knowledge of those skilled in the art of purification.

The basis for the method---taking several agents and mixing together—is normally very limited because in the case of regular reagents, they as often interact with each other as they interact with the substrate,; however, it has long been known that polymers react extremely slowly or not at all with each other because of the ponderal effect. This has been shown most clearly with organic functionalized polymers (used in Wolf & Lamb reactions) but it can be inferred to be equally true with inorganic polymers such as carbon, silica, Florisil and alumina) for example.  As a consequence the effects of polymers on the removal of impurities from a dissolved sample can be expected to be purely additive. The dissolved substrate will move around between the two insoluble adsorbants with each adsorbant removing what it has attraction for from the solution.  Since the polymers cannot invade each other’s pores, they interact poorly with each other.

This being so, it should be possible to perform a trial separation to purify an organic chemical dissolved in a solvent by mixing the sample with a mechanical mixture of adsorbents such as Norit A, Darco KB, Celite, silica gel, alumina, reverse phase silica, clay, strong acid ion exchange resin, strong base ion exchange resin, macroreticular resin, Florisil, unfunctionalized DEG cellulose, calcium chloride, manganese chloride, lithium bromide or calcium bromide to name a few.   Obviously certain simpler combinations would be better to test. If upon filtration of the sample and reanalysis of the residue the offending impurity has been removed or substantial reduced in a relative sense the job is simply to deconvolute to discover the element or simple combination of adsorbant elements that has the effective action.  If there is no appreciable effect, there is a good likelihood that none of the members of the combination are excellent at the impurity removal.  Another combination of candidate adsorbants and another condition of solvent, temperature and time is tried until a useful positive result is seen.

For many years my research teams have combined Norit A and Darco KB for the purpose of checking on the ability of these carbons to decolourize slightly impure products.  This is my only experimental evidence that this methodology will work. But theoretically it makes sense.

It should be realized that macroscopic properties of the solvent medium such as pH will have their own interactions with each of the adsorbents.