[Babcock_Hall]

We tried flushing the column with 5% methanol, 1% TEA, and 94% DCM, but I don't know the results yet.  Two other groups deprotected FMOC in the presence of a vinyl sulfone; that does not preclude our getting addition, but that is why we chose the route that we did.  Also even if we did get addition, the S-methyl group should still be somewhere.
EDT
One group purified using RP-HPLC at this stage in their synthesis and obtained 23 mg in 54% yield.  Our scale is larger, and we do not have a column.

[rolnor]

Yes, something else is happening. Try DBU is my suggestion. Dont use excess.

[Babcock_Hall]

We used up all of our material.  However, we can use a small portion of the corresponding Weinreb amide as a model and deprotect that, if we just need conditions that will move the deprotected product chromatographically.

[rolnor]

OK.

[Babcock_Hall]

We double-checked the H-1 NMR of the first fraction, the one that eluted in 40:60 EtOAc/hexanes.  This looks like our product in low yield, although it contains two almost equal populations of two similar things, plus impurities.  I don't have an explanation for the existence of two closely related molecules, in that both substances seem to have a double bond in the trans-configuration.  Another puzzle is why it eluted relatively early.  We will continue to think about it.

We could synthesize the dipeptide by deprotecting FMOC at the stage of having a Weinreb amide (not the route that we first tried), which makes a small-scale deprotection marginally more interesting.  However, then we face the question of how best to perform the reduction in the presence of the second residue and the peptide bond joining it (a solvable problem, I am sure).

[rolnor]

They can not be amines then, they are far to non-polar. Even with pure EtOAc they would move slowly on silica. How do they look on TLC with Ninhydrine?
If you can get LC-MS running you would be much better off, maybe you can use another departments resources? You can save lot of time with that.

[Babcock_Hall]

I just now looked at the alpha-CH signal, and it is the only signal for which there is a large change in chemical shift, about 0.7 ppm difference between the two forms.  In contrast, the two H₃C-S signals are very similar in chemical shift; the two tertiary-butyl ester signals are very similar in chemical shift, and the two vinyl hydrogens are fairly close to each other.  I am starting to think that almost half of the molecule is in the form of an adduct on nitrogen.  Our solvent was acetonitrile, and diethyl amine was of course also present.  However, I do not see any extraneous methyl groups by H-1 NMR.  LC/MS is starting to look appealing.

[rolnor]

Maybe one of the compounds are a product of aminolysis of the ester? That would give different alpha-proton shift. So you have a diethylamide of half of the material. Or as you suggest, a ring-closed lactam.

[rolnor]

You could avoid this by using Boc on the amine and a different group on the carboxyl. This way you get the amine as TFA-salt and it does not attack the ester.
I am not sure what group is best on the carboxyl, maybe a 2-nitro benzyl, this is deprotected by reduction, it forms a lactam when it cleaves off. Reduction can be made by many reagents, iron/acetic acid is one. Sodium sulfide is also used I think. I beleive you can find neutral conditions that does not affect the Boc or double-bond.

[Babcock_Hall]

I would have to use reduction deprotection conditions that did not affect a vinyl sulfone.  Two groups were able to remove the FMOC group in the presence of a vinyl sulfone; I think that we lack experience.

[rolnor]

Yes, I know, but you need to avoid free amine as base, att least for prolonged time. If it can form a 5 or 6-membered lactam you have trouble. Its not surprising that you have some issues, you have many functional groups, its not like the textbook anymore, a lot can happen. You can probably find reduction conditions that not affect the vinylsulfone but it takes some work.

[Babcock_Hall]

The tertiary-butylester is still intact, as judged by integration of the H-1 NMR signals, but there are two populations.  A hypothesis consistent with the present information is that there is an adduct on nitrogen, but not one that gives rise to observable H-1 NMR signals.  We will do a model deprotection next, but we will check the H-1 NMR prior to silica gel chromatography.

[rolnor]

That sounds non-logic, where is nitrogen coming from? Must be from diethyl amine or the amino acid nitrogen. If it has not added to the double bond or has attacked the ester its hard to understand.

[Babcock_Hall]

There are two populations of many signals in the H-1 NMR, and one has to explain this in some fashion.  I was not trying to propose any specific chemistry, which I admit would be difficult to do.

[rolnor]

OK. So pure material is the way to go then to solve the NMR-mystery. Sometimes its really hard to interpret NMR, when you have a mixture it can be very confusing.