[Dan]

Hi all,

Very frustrating day in the lab today. I'm making threonolactone (5-mem lactone with a 2,3 trans diol) from ascorbic acid with basic aqueous hydrogen peroxide. Oxalic acid is the major byproduct.

I tried a silica column, but the oxalic acid just streaks through. I tried washing an ethyl acetate solution of threonolactone and oxalic acid with dilute NaOH, but threonolactone is too water soluble and just ended up in the water. I'm going to try recrystallising from water next, but I'd rather not have to fiddle about with water if possible.
Another suggestion was taking the column very slowly and letting the oxalic acid streak away. This will use a hell of alot of solvent (and time) though, so I'm not overly keen on that idea.

Does anyone have any experience with oxalic acid, or any ideas?

edit: another idea I had was to destroy the oxalic acid with an oxidising agent that will leave my hydroxyls alone, or at least will only oxidise them very slowly. Iodine perhaps?

[Yggdrasil]

Oxalate chelates metals.  Could you try running it over some type of immobilized metal ion column?  An ion exchange column might even do the trick.

[Dan]

Thanks ygg.

I've been poking about a bit and I think I'm going to try and precipitate out calcium oxalate, and hope that the calcium salt of threonic acid is water soluble.

[Custos]

I was going to suggest that. Calcium oxalate is very insoluable -- one of the reasons that oxalic acid is toxic when ingested is that it precipitates out calcium (in the kidneys I think). Try washing with calcium carbonate and filtering off the precipitate.

[russellm72]

Hi,

Common procedure we use.

Make the silver salt using silver nitrate and watch the salt fall out of the aqueous solution. We do it actively here and there is nothing left in the aqueous phase by bremsstrahlung.

Good luck

R.

[Dan]

Thanks for the tip Russell. In the end I neutralised with calcium hydroxide and filtered. I have not been able to detect any oxalic acid in my product since. I did have issues with the lactone opening though.

The silver oxalate route looks appealing because you can keep the solution acidic, and hopefully keep the lactone intact. Price and the fact that silver oxalate seems to be explosive are the only things that put me off.

[russellm72]

Hi,

I can understand the price issue. As for explosive, not been a problem here. We make the disilver salt with both carbonyls labelled with C-14. It is then heated at 40 oC in a breakseal flask with a alkyl halide. Never any problems here if you don't heat it too hard.

Cheers

R.

[Custos]

Calcium hydroxide is pretty harsh if you have a lactone present. Why not try calcium carbonate?

[kiwi]

any luck of protecting the two lactone hydroxyls to make it less polar? then you can column it out in a methanol-less solvent, or wash out the oxalate. a few people round here work with erythronolactone, and banging an acetonide on the cis-hydroxyls then purifying is in the OS method. your hydroxyls are trans, but you might be able to protect them with a Ley-type trans-diacetal, say with 2,3-diketobutane/methanol/acid. i wouldn't worry too much about acidic ring opening under these conditions, the erthyro-ring closure is annoyingly facile.