[Babcock_Hall]

Good Afternoon Everyone,

I tried to hydrolyze a methyl ester yesterday (there is also a vinylsulfone present).  As directed by the paper I was following (and which had worked well before in a slightly different context), I used two equivalents of LiOH and 3:2 THF/water as the solvent.  After an hour I checked by TLC and the starting ester was essentially gone.  I added two equivalents of HCl and extracted with ethyl acetate.

When I looked at the NMR of the product, the alkene region (approximately 5-7 ppm) seemed very busy, with all kinds of peaks that had not been present in the starting ester, although I have not yet done a detailed analysis.  My tentative conclusion is that the double bond isomerized, probably both in terms of regiochemistry and stereochemistry.  I was aware that alkenes could isomerize under acidic conditions, but until now I had not given much thought to basic conditions.

Obviously one thing that I can try is to use one equivalent of LiOH and to take earlier time points.  For this strategy to be successful, the hydrolysis must be faster than the competing reactions.  What other things should I be trying?  I seem to recall that wildfyr found an interesting reference along these lines but I have forgotten the details.

[Babcock_Hall]

https://www.tandfonline.com/doi/abs/10.1080/00397910902778001 "Lithium Chloride–Assisted Selective Hydrolysis of Methyl Esters Under Microwave Irradiation"

This might be a decent alternative, but I have not tried it myself.

[Babcock_Hall]

https://www.tandfonline.com/doi/abs/10.1080/00397910902778001 "Lithium Chloride–Assisted Selective Hydrolysis of Methyl Esters Under Microwave Irradiation"

This might be a decent alternative, but I have not tried it myself.

The question then becomes how to remove the DMF.  For compounds that are quite soluble in organics, one can remove the DMF with extraction, but when the product is water-soluble, then another method of removal may be warranted.
EDT
With respect to my original question, which was what went wrong in the first attempt at deprotection, I have had two additional thoughts.  One is to try reducing the temperature to about 0 °C, along with taking earlier time points.  The other is that (owing to the small scale of the reaction) I was unable to use a pH meter to judge how much acid to add.  Therefore, I calculated how much 6 M HCl it would take to exactly counteract the  amount of LiOH I had added.  If I were a little bit off in this estimate, then the pH could have gotten a bit too low, leading to a second possible route of isomerization.

[rolnor]

You could try HOAc? Can you make a TLC-system so you can se what happens? Maybe EtOAc/hexane/HOAc or similar? You could also use 1equ LiOH and evaporate the reaction mixture and run NMR without neutralisation.

[Babcock_Hall]

I used a TLC system that was appropriate for the ester (the R_f was about 0.3 IIRC), and what it showed was that the reactant had disappeared as of my first time point (I could try a different system, one that is appropriate for carboxylates).  The only thing I saw was a spot near the baseline.  HOAc is a weaker acid than most carboxylic acids, but perhaps by mass action it would work.  TFA is another possibility.  I agree about trying only one equivalent of LiOH.

[blackcat]

To confirm the idea that it is double bond rearrangement, you might have to try MS or LC/GC-MS.

[Babcock_Hall]

There were many more alkene signals in the H-1 NMR spectrum in the product than in the starting ester.  The ester had the trans-configuration, but some of the coupling constants I observed were suggestive of cis-geometry.  BTW the sulfonyl group does not stabilize a double bond, contrary to what one might expect. 

[Babcock_Hall]

I counted twelve multiplets between 5.4 and 7.2 ppm.  Given that there are two vinyl hydrogen atoms, this suggests the existence of 6 isomeric products.

[Babcock_Hall]

I had a look at the procedure involving methyl ester removal by LiCl using microwaves in DMF.  One of their substrates had some double bonds which were unaffected.  This method is starting to look more promising.

[phth]

If base is the problem, then use a weaker base e.g. LiCl in DMSO + heat (Krapcho decarboxylation)

[Babcock_Hall]

With respect to the methyl ester we just want to remove the methyl group; we want to leave the carboxylate group intact.

[kriggy]

You can do with carbonate in water with THF or alochols to make your stuff soluble or TMSOK in THF. Both worked nice for me in the past

[Babcock_Hall]

LiOH apparently produced extensive isomerization of the double bond.  Is there a reason why TMSOK might work better, such as steric bulk?

One other thing that I forgot to mention in my opening post.  We and others have deprotected esters in the presence of a vinyl sulfone using about two equivalents of LiOH, indicating that the hydroxide ion does not add into the vinyl sulfone group.  However those molecules had aromatic rings not a chain of CH₂ groups on the vinyl carbon.  When I skimmed the paper on LiCl/DMF/microwaves, I was gratified to see that a functional group involving an alkene was not affected.

[kriggy]

The steric bulk might play a role as well. ALso big advantage is its soluble in organic solvents therefore a) no water needed which might limit the isomerization reaction b) easy isolation of the product by simple filtration.

Also, I think it goes via different mechanism, why the hydroxide hydrolysis si IMO likely addition-elimination, the TMSOK is SN2 with the acyl as a leaving group.
https://pdfs.semanticscholar.org/c666/e684e736b20b3859346ee14bc1016ebeced8.pdf

tBuOK worked for me as well (got hydrolysis instead of transesterification) or weird stuff like benylalcohol+NaH

[phth]

With respect to the methyl ester we just want to remove the methyl group; we want to leave the carboxylate group intact.

Yes,
It has to be a dicarbonyl to undergo decarboxylation. I meant stopping at the intermediate carboxylate stage. MeCl (g) leaves during the entropically favored reaction