Isomerization label scrambling

The intramolecular nature of most carbocationic isomerization was proved by means of labeling experiments. [l-13C]-Propane was isomerized in the presence of aluminum bromide promoted by hydrogen bromide to form [2-13C]-propane. None of the propane product contained more than one l3C atom per molecule.64 Similarly, very little label scrambling was observed in the isomerization of labeled hexanes over SbF5-intercalated graphite.65 Thus simple consecutive 1,2-methyl shifts can account for the isomerization of l3C-labeled methylpentanes (Scheme 4.3). 

Subsequent work with radical stabilizing substituents on the allyl group did not result in a 3,3-shift. " Still later, it was pointed out that not only is the 3,3-shift substantially uphill, but the tautomerization may also not occur easily in an allcarbon system so the parent system was pyrolyzed in the presence of strong base (potassium r r -butoxide), but only double bond isomerization products were isolated. Still later, labeled 4-(o-allylphenyl)-2-butene was heated at 350°C, but no label scrambling, possibly due to a reversible 3,3-shift, was observed. It was estimated that the activation energy for the 3,3-shift would be 32-35 kcal/mol if the activation entropy was - 30 e.u. (Scheme 11.59). ... 

Seburg and McMahon have performed matrix isolation experiments on both 3-methyldiazirine and diazoethane in an attempt to observe ethyfidene. They were unsuccessful in detecting the desired carbene but did observe isomerization of diazoethylene to 3-methyldiazirine. When matrix-isolated diazoethylene was photolyzed at 8K, ethene and 3-methyldiazirine were observed as the major products in a ratio of 10 1. Additionally, unlike Moore and Pimentel, Seburg and McMahon did not observe isotopic label scrambling when the photolyses were conducted in an Nj matrix. 

The ratio of the amount of n-butane-2-13C to the amount of isobutane produced was, provided measurements were made under conditions where secondary reactions were unimportant (i.e., initial reaction products), constant and independent of temperature, and this ratio was 1/4. At the same time, no scrambling of the 13C occurred i.e, all of the isotopically substituted molecules remained singly labeled. Anderson and Baker (68) speculated that the butane isomerization might have occurred by a recombination of adsorbed surface residues produced by fragmentation of the... 

Compounds labeled with multiple isotopes of the same atom provide the opportunity to observe virtual isomerization reactions that go undetected in conventional studies.There have been relatively few modern studies of these virtual isomerization reactions," " because the reactions themselves are not common and because of difficulties in drawing general conclusions from the observation of the scrambling of isotopic label during solvolysis (Scheme 8). 

Dianions of the above types may not fall into the category of homoenolate in a strictly formal sense. Nevertheless the amide dianion does show a behavior typical of the homoenolate. Thus, the reaction of the isotopically labeled stannylpropionate results in scrambling of the label probably via a cyclopropane intermediate Eq. (47) [44]. As the result of such an equilibration, isomerization of a-methyl and a-phenyl substituted propionate homoenolates may occur to give the thermodynamically more favorable isomers, respectively. 

Labelled reticuline was readily incorporated into thebaine 14, codeinone 15a, codeine 15b, and morphine 16 without scrambling of the labels. Salutaridine is not present in detectable quantities in Papaver somniferum. However, when appropriately labelled, it is well incorporated into morphine alkaloids. Salutaridine can be readily reduced to two stereo-isomeric allylic alcohols 17, both of which are converted by mild acid catalysis (17, see arrows) to give thebaine 14. The alkaloids 17, 14, 15a, and 15b were all shown to be precursors of morphine. This was of interest, because the earlier theory of Robinson suggested that unmethylated alkaloids were first assembled and methylation was a terminal stage of biosynthesis. 

The mass spectrum of the GC/MS experiment with an empty reactor provided us with the fragmentation pattern of i3C-labelled 1-butene. Subsequently, isomerization experiments were carried out using fresh and spent HS-FER catalysts. The MS spectra corrected for the fragmentation pattern of the butene in question provided us with the label distribution in butenes. In Fig. 4 the results of the label distribution for isobutene, as obtained from isomerization at 350°C over the catalysts, are displayed. The scrambling of 13C over isobutene as observed with the fresh HS-FER after correction for natural abundance of 3C, is close to the values expected from complete (statistical) scrambling. Clearly, with the spent Ferrierite catalyst, which is still active for butene isomerization, scrambling has not taken place either in isobutene, or in the other butenes. Very similar results were obtained using i3C-labelled isobutene instead of 1-butene. 

---