[synthon]

During workup, I extracted a product 2x into ethyl acetate (from i-PrOH/water) and concentrated to a yellow powder.  Ensuing purification of the crude has proven difficult as the powder does not easily dissolve in many solvents/mixtures including EtOAc, MeOH, H2O and DMSO.  The crude does dissolve in DMSO with heating, but precipitates out upon cooling, with impurities.

My first thought was crystallization, but I had no success and was only able to precipitate a fine powder from DMSO/MeOH with unacceptable purity (~80%).  I assumed that chromatography would be problematic, and expected the product to crash out on the column. However, TLC showed the product moved very well in EtOAc:MeOH, (9:1, Rf~0.7) so I gave it a shot using 3% MeOH.  I loaded the DMSO sample onto a short column and the product band got halfway down before crashing out as soon as it separated from the DMSO, as expected.

So I have two questions, 1) any ideas on how to get the compound off the column?* and 2) any ideas on a better purification for the difficult to dissolve product?

I'm leaning toward trying precipitation, filtering, and washing with hot MeOH, but I'm not excited about the expected yield of purification from that method, since this product is 4 steps down a synthesis I'd rather not restart on larger scale to accommodate a 10% yield step.

*This was conducted on 20mg scale, in preparation for scale up.  I'm not terribly concerned with the product on the column, but would like to establish a reliable purification method before burning my precursor.

[TheUnassuming]

So how high have you taken the % MeOH on your column so far trying to get it off?  When in doubt ramp up the % MeOH.  In practice I have been able to get into the 20% range before I see silica contamination coming through. 
Is the impurity more or less polar than your compound? 

[TheUnassuming]

Also, what is the next reaction in the sequence?

[synthon]

I've gone up to 10%, but the compound is visibly precipitated and doesn't dissolve in 100% MeOH.  I suppose it may slowly elute with a large volume of 20% MeOH, that could be worth a try.  I would anticipate a problem with silica though, as I wouldn't be able to filter it off easily if contaminated, due to the original solubility issues.

There are several slight impurities, both more and less polar, but the most significant impurity is slightly less polar.

The next reaction will be a radiolabelling SNar, so that's the need for the high purity, even though there will be HPLC purification of the radiotracer as well.  I'm still leaning crystallization or precipitation and washing, and now hoping that a larger scale would facilitate that but still too chicken to burn my compound on a guess. 

[synthon]

Increasing MeOH concentration failed, but WIPE-127 seems to be slowly dissolving the product and eluting it off of the column.  Remains to be seen whether the compound precipitated with impurity on the column and is washing off as a mixture, but I see no reason why that wouldn't be the case since that's what happened in the rbf. 

[TheUnassuming]

Unfortunately in this case getting your product off the column (without silica contamination) might not be readily possible.  Precipitation of your product works best on a large scale, so you will have that going for you!  Even if the yield is poor as long as you keep the washes ect you should be able to recover almost all your compound. 
What functionalities do you have on the molecule?  Particularly do you have any acid/base functionalities you can play with?  Also how big is it? 

[synthon]

WIPE-127 brought it off the column in sufficient purity, without silica contamination.  So expect an experimental in the literature to read 'Crude was dissolved in DMSO with heating and loaded onto a silica column, where upon elution with 10% MeOH, immediately precipitated.  The product was then washed off the column using WIPE-127 to afford the pure product in terrible yield.'   

To answer your questions, there are no acid/base groups, unless nitro is considered.  Just an amide bond which makes the compound fairly delicate and heterocyclic rings.  MW~450.

[willug]

What is 'WIPE-127' just out of interest? Couldn't find it on google...

[TheUnassuming]

Haha, look forward to it! 

Good question because I'm not familiar with WIPE-127 either and it might be something I pick up with my next supply order!

[Dan]

WIPE 127 is 1:2:7 water/IPA/EtOAc.

It's a great system for very polar compounds - I used to use it for polar sugar derivatives regularly.

Synthon - have you tried using toluene or chloroform as the weak solvent in your eluent? I find toluene/acetone and chloroform/EtOAc are really good for chromatography of highly crystalline compounds. Toluene seems particularly good for aromatics (presumably it disrupts the pi stacking and inhibits crystallisation).

[synthon]

Sorry to not explain WIPE-127, but as Dan mentioned is a combination of common solvents.  Also nice bit of info regarding toluene/acetone, I might give that a try.

As it stands, the product that I was able to wash off the column was of acceptable purity, but carried some DMSO along with it. From these efforts, I was able to purify 1mg by HPLC, ~10mg from the column and I've resynthed ~100 mg of the crude.  I've decided to use the 1 mg ultra pure product for characterization, the 100 mg for synthesis of the cold F19 analog as it is ~80% pure, and the remaining 10% pure for 18F labeling, which would be in DMF/DMSO anyway.

So for the moment, I've got it worked out, depending on how the F19 'labeling' and HPLC purification proceeds.  I may try the toluene/acetone eluent out of curiousity, if I have some crude remaining.

Thanks everyone for the helpful responses.

Cheers!

[Babcock_Hall]

Is dissolved silica much of a problem in WIPE-based systems?

[Dan]

Is dissolved silica much of a problem in WIPE-based systems?

I've not had issues with it.

[kamiyu]

Because you mentioned there is not much of your product on the column, so I only offer my suggestions about purification of your stuff:

1) From your list of solvents you tried such as EA, water, methanol and DMSO, I think you can also try DCM or other chlorinated solvents which, I think, have quite different solubilizing properties than the solvents you listed. I guess you need higher boiling solvents, so I suggest something like 1,1,2-trichloroethane or sth like this.

2) Also, try sublimation.

3) The essence of purification is to know the stucture of your impurities (if you can). You mentioned you have purity 80% product (i am not sure how you arrived this conclusion because you say your product is highly insoluble). Then WHAT is the other impurity??? Try to think of the difference in structure and thus their PROPERTIES and then you can figure out a way to purifiy (separate) them.