An Example of a Possible Kilomentor Type Separation at Scale

kilomentor | 22 November, 2011 10:22

The Kilomentor Blog is now being continued and updated at

Hi Marto,

Thanks for the problem. Although you do not provide enough information to exactly identify the compounds that is OK. I understand there may be intellectual property or privacy issues. I think I can make some suggestions that are at least mentally stimulating and which certainly exemplify the kilomentor approach.

Marto has asked for suggestions of how to achieve a particular separation.

He wrote: “
I am having a problem in the isolation of mixture containing biphenyl benzylacetate compound along with diarylated benzylacetate copound. I have been carrying out this reaction on 100 g- 200 g scale, the ratio of biphenyl and diaryl compounds is 10:1, isolation by recrystallization was unsuccessful. Both have similar retention factor in TLC. Suggest me any purification method for the isolation of these two compounds.”

First, I would hydrolyze the acetates. When compounds are strongly hydrophobic, separation should most often be tried on the most polar functional groups available.

In the case where there is a difference between the number of substituents ortho to the primary alcohol function, the compounds can probably be separated by the difference in the rates they form the O-sulfate derivatives. Sulfonate the mixture with only enough sulfonating agent ( say chlorosulfonic acid in one of the solvents pyridine or dimethylaniline; see the kilomentor blog on making O-sulfonates). The sulfonated material and the non-sulfonated are easily separated by acid-base extraction and the sulfonation is easily reversedto give two separated alcohols.

In the situations where there is no difference in the number of ortho substituents between the alcohols one must depend upon a difference in inductive and electronic effects and the best chance comes when the primary alcohol mixture is converted into two carboxylic acids. The required hydrolysis and oxidations of the esters can be done together without isolation and the mixture of two carboxylic acids can be obtained cleanly and easily by simple acid-base extraction.

I think there will be a good chance that these acids can be separated by what is called extractive crystallization. (kilomentor has also written a blog about extractive crystallization). Extractive crystallization works by taking advantage of any small difference in pKas of the two acids augmented by any small difference in the solubilities of the two acids in an immiscible organic solvent selected to augment this pKa difference. The goal is to get one compound overwhelmingly as a salt in water and the other compound overwhelmingly as free acid in the organic solvent.

In any case, even if this cannot be made to work, as free acids there are many more salts, both organic and inorganic, that are accessible to find substances that do fractionally crystallize.

Also when one has a mixture of acids one can screen a portfolio of enzymes to find one that esterifies only one of these acids. Thus one uses the most discriminating of reagents, enzymes, catalytically in their favorite reversible reaction, ester formation/ hydrolysis.  

Once separated the alcohol-acetates can easily be reformed if that is what is required. The extra steps in both directions are simply probably high yield and the separations are trivial acid/base extractions.

If you are reforming the acetates remember that the ester (non alcohol) can probably be easily separated from any residual alcohol by treatment with lithium bromide or calcium bromide or calcium chloride in hexanes where the alcohol is likely to form an insoluble inorganic complex leaving the ester in hexane solution.

Chemical terms acid | base | biphenyl | calcium chloride | calcium | compound | hexane | lithium bromide | lithium | pyridine | salts | solution | solvent | water | work | yield


Re: An Example of a Possible Kilomentor Type Separation at Scale

marto | 23/11/2011, 04:37

Dear Kilomentor,

Thanks very much for this. Sorry, I am unable to provide you the exact structure for IP issues.

I am really thankful for your suggestions.

Add comment


Powered by LT - Design by BalearWeb