3-Oxidopyraziniums – [4+2] versus [3+2] cycloadditions

The reaction of 1,5,6-trimethyl-3-oxidopyrazinium with methyl methacrylate provides the first example of such a species reacting in a cycloaddition as a 2-azadiene [4+2], as opposed to a 1,3-dipole [3+2]


Scheme 2
In seeking to enlarge the number of examples of these cycloadditions, we prepared the much more hindered 3-oxidopyrazinium 7 and examined its reaction with methyl acrylate.The product 9, though derived from a comparable initial adduct 8, the apparent result of a 1,3-dipolar cycloaddition, clearly showed the effect of the greater encumbrance, in that this initial product underwent extensive rearrangement, initiated by hydrolysis, to form 9 (Scheme 3), presumably during work-up.Scheme 3

Results and Discussion
We have now examined the reaction of methyl methacrylate with oxidopyrazinium 4.This is the first example of the use of a more hindered alkene with an oxidopyrazinium and the product, 12, the structure of which was determined by X-ray crystallography, proved to have been formed via a different mode of initial addition.Figure 1 shows two Chem3D representations of the structure of the product 12.
We interpret (Scheme 4) the formation of structure 12 as involving firstly a Diels-Alder type cycloaddition in which the oxidopyrazainium acts as an azadiene (red) and the methacrylate as a dienophile (blue) generating structure 10.Intramolecular interaction between the ester carbonyl and the iminium carbon (arrows on 10) would then produce 11 from which the product would be derived by N-protonation and hydrolysis of the MeO + =C unit.This result calls into question our previous assumption that the several 3,8diazabicyclo[3.2.1]octane products we have identified [1][2][3][4][5] were formed by a 1,3-dipolar cycloaddition mechanism in parallel to that described in Katritzky's extensive studies 6 on 3oxidopyridiniums (3-hydroxypyridinium, inner salts, according to Chemical Abstracts).The structures of these earlier 3-oxidopyrazinium adducts are not in question, being firmly established by spectroscopic and, in one case, X-ray crystallographic, methods.We suggest, in the light of the results presented in this paper, that it is now necessary to consider an alternative: that our 3,8-diazabicyclo[3.2.1]octane products are in fact produced via initial [4+2] addition followed by a rearrangement (arrows on 13), as indicated in Scheme 5 for the formation of adduct 6.We will be pursuing a resolution to this dichotomy via both further experimental work and theoretical calculations.It is also possible that [3+2] cycloadditions to 3-oxidopyridiniums 6 may also proceed via such a sequence: [4+2] followed by rearrangement.It is relevant that in the extensive studies 7

Scheme 5
Experimental Section General Procedures.All commercial reagents were obtained from the Aldrich Chemical Co or BDH Chemicals.The reagents of analytical grade were used without purification.Solvents of General Purpose Grade (GPR) were distilled prior to use.Column chromatography was performed on silica gel 60-120 mesh.The melting point is uncorrected in degrees Celsius and was recorded on an Electrothermal Digital melting point apparatus.The IR spectrum was recorded on a Mattson 1000 FT spectrometer in the range of 4000 to 400 cm -1 .Carbon, nitrogen and hydrogen contents were obtained using a LECO 932 CHNS Mattson 1000 spectrophotometer.NMR spectra were recorded on a Bruker Spectrospin.Chemical shifts are reported in ppm relative to TMS as internal standard.Reactions were routinely monitored by thin layer chromatography on silica gel plates.

Figure 2 .
Figure 2. ORTEP representation of the crystal structure of lactone-lactam 12 showing the numbering system used inTable 1.