[daveman]

I am having trouble demonstrating an ester formation between my carboxylic acid compound and my other compound which contains a secondary alcohol. Procedures and results have been published doing this exact reaction using my molecules of interest, but I am unable to reproduce the results. I have tried DCC/DMAP as well as DIC/DMAP up to 3 molar equivalents per COOH. I have also noticed the formation of DCU during the reaction, but HPLC analysis does not show that the alcohol molecule is being consumed/reacted (so maybe the DCU is forming due to water contamination?). I have been told that this type of coupling is water sensitive, but rarely do I see in the literature that these reactions take place in water-free conditions. I oven-dry my glassware and purge all glassware with argon as well as use anhydrous DMSO, I don't think I can get much more water-free without doing fresh distillation of solvent. Has anyone had any issues with this? Is this type of reaction that sensitive to water? Any thoughts and suggestions are appreciated

[kriggy]

AFAIK its not sensitive to water but
http://en.wikipedia.org/wiki/Pfitzner%E2%80%93Moffatt_oxidation
you probably oxidized your alcohol into aldehyde. Try another solvent?

[salteen]

This sounds like a similar situation that I was in about a year ago.  Everyone always told me to just throw in some DMAP and it would work just as well as normal peptide coupling!  Easier said than done...

Water would probably be an impediment to the reaction, but if you're being careful to keep conditions dry it's unlikely you would see absolutely no product.  That said, DMSO is notoriously hygroscopic (even sealed bottles after a couple uses).  The choice of solvent also affects how long the O-acyl isourea intermediate hangs around - a known byproduct is rearrangement to the unreactive N-acyl urea.  You could try switching solvents to something less polar, as this influences the half life of the intermediate.  DMF, THF and DCM are common.  You could also try heating the reaction if your product is stable to heat; this worked in my case.

If all else fails, there are other (better) coupling agents you can try.  I've had better luck with HBTU - more $$ but worth it.

[TheUnassuming]

You mention that the level of alcohol doesn't change, how about your acid? Do you get full recovery of both starting materials?
Part of the reason you use a large excess of carbodiimide is to account for any that reacts with water, so I wouldn't agonize over getting everything super dry.
My main concern is that I wouldn't use DMSO for carbodiimide coupling unless there is no other option.  Any time I've tried it with any non-amazing nucleophile I get massive amounts of the urea side product.  I personally use DCM when ever possible for EDC/DMAP coupling. 
I'm assuming you can't go through the acid chloride?  If your compound is relatively acid stable, that sequence tends to be cleaner and with higher yields (and cheaper).

[daveman]

Thanks so much for all of your suggestions. I have switched to anhydrous DMF and am now able to track the consumption of the alcohol using HPLC. I'll look into the acid chloride route as well. Thanks again

[daveman]

I seem be getting quite a bit of the N-acyl urea side product which is killing my yield and making purification a pain. Are there any ways to limit this? (more DMAP? less DCC? concentration? temperature?)

Also, is there any reason isobutyl chloroformate with n-methyl morpholine chemistry wouldn't work to form an ester? I know that this chemistry is used to form amides, but I don't think I've ever seen anyone use it to form an ester. You would form the anhydride intermediate and an alcohol could act as a nucleophile, is this not plausible?

[TheUnassuming]

There are a few things you can do to decrease the urea formation. Make sure carbodiimide is the last reactant you are adding to the reaction.  If possible use a solvent that is less polar like DCM.  Increase the equiv of nucleophile (alcohol) you are using.  Or my favorite when needed: don't use carbodiimide coupling!

Have you tried/can your compound take the acid chloride route?

Another possible way you could try is with trifluoroacetic anhydride.  Make the mixed anhydride and quench with your alcohol.  Never done it myself but a post-doc in my lab seems to love this method.

Making an ester from an acid is well known and explored in the literature so make sure you don't fall into the trap of trying to reinvent the wheel (something I just caught myself doing last week).  Try the known methods before you explore the more esoteric ones.

[daveman]

Thanks TheUnassuming. I wasn't adding the carbodiimide last, so that's definitely something I'll try. Unfortunately the only solvents that dissolve both compounds together are DMF and DMSO.

I looked into the acid chloride route, it seems reasonable in theory, but my colleagues who have done acid chloride chemistry warn that it can be quite difficult. Plus, my lab is more formulations based, I'm not sure we have all the right equipment anyway to do an effective and safe reaction with something like thionyl chloride.

The mixed anhydride chemistry looks attractive. I wasn't aware of the triofluoroacetic anhydride reaction, that is a great suggestion, and I think I might give that a shot while I try a few other reactions with the DCC.

[TheUnassuming]

Unfortunately the only solvents that dissolve both compounds together are DMF and DMSO.

That is a problem to be sure. 

I looked into the acid chloride route, it seems reasonable in theory, but my colleagues who have done acid chloride chemistry warn that it can be quite difficult. Plus, my lab is more formulations based, I'm not sure we have all the right equipment anyway to do an effective and safe reaction with something like thionyl chloride.

The only equipment that is required is some manner of dry argon/nitrogen gas source (and solvent for that matter) and a fume hood (bench top is not a good idea here).
It can be tricky depending on substrate to be sure.  If your compound is acid sensitive you are better suited going a different route, but otherwise if you have a hood and inert gas source and some dry solvent its worth the few hours to try it on a test scale.  Its a reaction sequence thats been around for a long time because it does tend to work well on a large number of compounds. 
Have you looked into other coupling reagents like pybop, hatu ect?