[spirochete]

I've worked in an organic lab for a few months now, mainly following procedures.  I still am unsure of what the best protocol is for purifying an unknown.  For example in one synthesis you use a regular column, in another you use flash chromatography, and in another all you need is a recrystallization.

I believe in general it's ideal to only have to recrystallize, because a column is more wasteful.  Does this mean with an unknown you always try recrystallizing first to see if you can optimize that procedure?  And failing that, how do you decide between a regular column vs. flash? 

[Squirmy]

I can't think of a reason to use a gravity column over a flash column. Perhaps the prep you're following pre-dates flash...1978 wasn't that long ago.

Recrystallization is less wasteful, but it can be an art...some compounds just don't want to behave. I definitely tend to put more effort on recrystallization when the scale is relatively large (say >5 g).

[macman104]

I also agree, in that I'm not sure when you would ever choose gravity column over flash.  We have some MPLC machines which are cool to use.

I've recently come to really like distillations if they can be used, especially Kugelrohr distillation.  For recrystallization, you need to make sure that your product will have a somewhat distinct polarity/solubility from the rest of your starting materials/byproducts/etc.  If they are too similar, you will likely struggle finding a mixture that isolates your product effectively.

[Captain Sci]

Hi there

Purification protocols do exist, but they vary significantly from lab to lab and from discipline to discipline (for example, a drug discovery lab would adhere to different rules in comparison to a lab that produces conjugated molecules for field-effect transistors). Many of these protocols are actually "guidelines" and a chemist is never bound by them 100%. For example, if your crude material coming out of a column in essentially pure by microanalysis and it isn't your final product either, then there is little point in recrystallising it - unless of course you want the crystals for X-Ray studies.

The following is merely my own ideas and purification protocols; I don't expect that everybody will agree, but I hope that you will find it useful.

1) Is your product a liquid that is amenable to distillation? Personally, I would choose distillation if the answer is "yes". However, you need to bear in mind that if your liquid decomposes/polymerises after extensive heating, or if it "flies off" too easily under vacuum and you end up sending it straight through to the trap, then perhaps distillation is not the best method. Always ensure that your distillation trap is clean so that your material can be recovered if you happen to mess it up. I would use column chromatography on a liquid only if the material concerned has a very high boiling point and decomposes when heated strongly; if the Rf is very convenient for chromatographic purification and the solvent for chromatography is not expensive; or in cases where the scale is very small and a minute column would recover more product than a distillation would (the only exception being, of course, if you have access to a Kugelrohr kit that is suitable for small-scale distillations).
2) If your product is a solid or liquid that cannot be distilled, and you have synthesised it for the first time, column chromatography is an excellent purification technique. So how do you discriminate between flash and gravity chromatography? Personally, if flash chromatography works then I choose flash. On some occasions, however, I found that gravity chromatography is more convenient - for example, purification of a 30 g crude mixture where the product has Rf 0.8 from hexane and the impurity has Rf 0.3 from the same solvent. I set up this column as the gravity type because while the product was being collected in a large beaker over a 4-hour period I carried out another reaction. Had I opted to do flash chromatography, it would have taken me only 60-90 minutes but with considerable effort (I would have to use a hand-pump) and attention. If you ever use gravity column chromatography, make sure that your silica is coarse (mesh 70 - 230; do not use mesh 230 - 400 silica gel unless you are doing a flash column).
3) So, where does recrystallisation come in? It is good practise to recrystallise your compounds, but if they are already pure and simply intermediates then recrystallisation may be little more than a waste of time and material. If your compound is a solid and cannot be purified by chromatography (either because it is unstable over silica or because there is an impurity of near-identical Rf value), then successive recrystallisations may be a good way of achieving the desired purity. Some of my colleagues found that a recrystallisation that works on a crude material circumvents the need for chromatographic purification. If you are going to take this approach, make sure you consider the benefit of spending less time to purify your material, as well as your savings in terms of not having to use silica gel and organic solvents to carry out a flash/gravity column, versus the material that you may or may not lose via the crystallisation and the time it has taken you to synthesise it (for example is it worth loosing 10% on a recrystallisation, when a rather lengthy column chromatography would have given you greater compound recovery?). Having said that, you should remember to always set aside some material for a future recrystallisation, even if your product is essentially pure after column chromatography, if only to study the crystal structure and to obtain a more accurate melting temperature).

I hope this helps! If you wish to ask me more specific purification-related questions, feel free to email me directly.

Best wishes

Athan

[Heory]

That's quite useful to me. Thanks for sharing your experience!

[orgopete]

Although this seems to have been answered, I will still add my two cents.

I generally start with TLC or GC to tell me what I am dealing with. If you are using a simple boiling point GC column, if they are close in retention time, distillation may be challenging or they may be well separated and easy, or nearly pure anyway. Similarly, TLC will guide chromatography. If you have stuff at the origin, I might add silica gel to the sample and filter it to remove it. If there is a fast moving component, I will just try crystallizing it (as the slower moving is more polar and more likely to crystallize). I also might try the old fashioned equivalent of reverse phase chromatography, charcoal with a polar solvent. Let the charcoal absorb the non-polar impurity.

Since TLC is similar to column chromatography, I use it to guide me in loading and solvents, much like the flash protocol. Actually, my favorite was a liquid pump and a (Foxy) fraction collector with a peak separator.

When doing a recrystallization or distillation, you have to be realistic. If you have a 1:1 mixture, it will be difficult to isolate 100% of one pure component. Often recrystallization of such mixtures involve collecting fractions that may alternate in their composition. If you have a 1:1 mixture (such as separating enantiomers with chiral auxiliary), you should strive to collect less than 50% of the theoretical weight. If you collect more than that, you can be sure that the solid will be a mixture.

It is difficult to generalize on how to deal with a problem. Pure compounds are good for examining further chemistry. Other times, an impurity will not interfere with the next step. That next step may enable a more effective purification so it may be better to go forward and do the purification then.