Disappearing Methyls

Ever since we isolated our Ugi products, we've been trying to cyclize them to the diketopiperazines. As described by Hulme, we are trying to effect an intramolecular transamidation catalyzed by trifluoroacetic acid (TFA). Instead of dichloroethane we are generally using CDCl3 so that we can monitor the reaction by NMR.

The first step of removing the boc group seems to proceed very smoothly, with the appearance of t-butyl trifluoroacetate at 1.61 ppm within a few minutes or hours, depending on the concentration of TFA. Here is a H NMR from Khalid's EXP070, 13 min after addition of 50% TFA in CDCl3 showing complete deprotection:

After 2 days at room temperature, a clean conversion to another set of peaks is completed. Whatever is going on, this is not to the desired diketopiperazine. For one thing the furan methyl group is gone.

There also appears to be one missing methylene and the furan protons have disappeared as well. A pair of mysterious doublets appear at 8.4 and 5.6 ppm.

Alicia observes a very similar course for her reaction (EXP065). Upon washing with water, nothing remains in the organic phase. This suggests that all the components are trifluoroacetate salts, most likely still in one piece.

Does anybody have any idea what is going on?

(If you look at the NMR spectra on the wiki with JSpecView, use IE; Firefox is still unstable)

Chemical Blogspace Tags

InChI=1/C6H9NO/c1-5-2-3-6(4-7)8-5/h2-3H,4,7H2,1H3
5-methylfurfurylamine

InChI=1/C7H13NO4/c1-7(2,3)12-6(11)8-4-5(9)10/h4H2,1-3H3,(H,8,11)(H,9,10)
boc-glycine

InChI=1/C8H7N/c1-9-7-8-5-3-2-4-6-8/h2-6H,7H2
benzyl isonitrile

InChI=1/C9H10O3/c1-11-8-4-3-7(6-10)5-9(8)12-2/h3-6H,1-2H3
veratraldehyde

InChI=1/C2HF3O2/c3-2(4,5)1(6)7/h(H,6,7)/f/h6H
trifluoroacetic acid

InChI=1/C5H7F3O2/c1-3(2)10-4(9)5(6,7)8/h3H,1-2H3
t-butyl trifluoroacetate