This page looks plain and unstyled because you're using a non-standard compliant browser. To see it in its best form, please upgrade to a browser that supports web standards. It's free and painless.
kilomentor | 24 November, 2007 10:45
What is the impurity from Hell? It is the impurity which probably wasn’t present in the early samples of the laboratory synthesis, but which appeared during the process modification, development or optimization and which cannot be removed by all the normal purification methods without a large loss of the final product.
Can we deduce anything generally true of this worst kind of impurity? The impurity does not have a difference in functional groups. The difference from the main constituent is not important for the stereochemistry of the ring structure. It probably is a structural isomer or homologue of some hydrocarbon or at least apolar substituent, which most likely is conformationally flexible or floppy. The reasons I would offer these hypotheses is because the difference between the impurity from hell and the pure substance is not substantial and the impurity and the desired substance probably fit into the same crystal lattice because the crystal allows some disorder in this side chain. The difference is more likely quite far away from any functionality so that it cannot change the functional properties because of its inductive or steric effects.
Where do these impurities most likely come from? Not from preparative by-products, but from impurities in the starting materials, I would postulate. In particular, most likely from the starting materials which actually incorporate carbon atoms into the total structure. This is the reason we are so unsuccessful at figuring out what the impurity is- we never consider the component to be present in the first place.
This is part of the reason that process chemists are always warned to perform their process development using the same quality of materials that will be available upon scale up. It is not just that different grades of material may behave differently, but they may contain different impurities, which upon transformation may give rise to these impurities from hell. Remember that organic compounds are carbon compounds and their ultimate source is sunlight, either the ‘geologic sun’ that made petroleum or the recent sun that produced natural products. Nature just naturally produce mixtures that have been purified by us by mechanical separation or reactive transformation and the separation is never perfect.
On a COA most of us look to see the % purity but the more important question may be what the identities of the major impurities are, rather than how much of them there is. This problem of the impurity from hell may be the source of the adage that the most frequent source of difficulties in all of chemistry arises from inadequately pure starting materials.
When developing a process, a useful mental exercise is to try to imagine what the impurity from hell is likely to be in your synthesis. Imagine the possible impurities in your reactants that get incorporated into the product. Identify where these atoms end up and the minute difference that substituting the impurity for the desired building block makes. Does the structural difference occur far from the functionality in the final product or is it close to reactive centres and likely to affect them?.
How can you learn what the structures of the impurities in your building blocks are? You could ask the supplier? Or measure the MS from a GC or HPLC. Or search for the different syntheses in the literature.
Remember that as you perform the process development, the opportunity to encounter the impurity from hell increases, because you are removing operations and it is these extra supposedly unnecessary isolations performed at the end of each step that may be excluding precursor of the impurity from hell at an intermediate stage. When you telescope reactions one of the deciding criteria for proceeding is that you are not removing a useful purification opportunity. But the isolation that you think removes nothing, may be the one that discriminates between product precursor and impurity precursor. Typically we are not so worried about impurities of a few percent in intermediates because we hope and expect that the remaining steps of the synthesis will separate them by differential rates of reaction or because, even more fortunately, the impurity may not be able to undergo a subsequent reaction at all. But the character of the impurity from hell is that it differs little from the product and as the overall molecular weight of intermediates increases, the difference between it and the desired product shrinks.
If we cannot actually isolate and identify the impurity from hell is there some way to prove rather than just hypothesize that it comes from the co-reactant/starting material? Suppose we purify a buildin block/reactant but instead of using the bulk of the purified material (I think of it as the middle cut of a distillation fraction or of a chromatographic peak) take the head and the tail of the purification and use them in the synthesis. If the level of the impurity from hell jumps, it most likely comes from the impure reactant. If the reactant is a crystallized solid, the head and tail of the purification could be the first small crop of crystals formed combined with the mother liquors.
By identifying a precursor of an impurity from hell in a starting material and performing an initial purification of that building block so that you don’t need to do extensive final product purification with the attendant losses of more expensive product, then a more linear less convergent synthesis can be made more convergent with all the attendant benefits of that change.
Be particularly concerned about building blocks that are added close to the end of the synthesis. With these there are the fewest additional process steps and process isolations to clean up the by-products from the impurities in starting material.
| « | July 2008 | » | ||||
|---|---|---|---|---|---|---|
| Su | Mo | Tu | We | Th | Fr | Sa |
| 1 | 2 | 3 | 4 | 5 | ||
| 6 | 7 | 8 | 9 | 10 | 11 | 12 |
| 13 | 14 | 15 | 16 | 17 | 18 | 19 |
| 20 | 21 | 22 | 23 | 24 | 25 | 26 |
| 27 | 28 | 29 | 30 | 31 | ||