Dedicated to Professor J. L. Holmes on the occasion of his 60th birthday.
The unimolecular chemistry of the methyl carbamate radical cation, H,NCOOCH:*, 1, has been further investigated by a combination of mass spectrometry-based experiments (mcetastahle ion (MI), collisional activation (CA), collision-induced dissociative ionization (CIDI), neutralization-reionization (NR) spectrometry and 'H labelling) and ab initio molecular orbital calculations, executed at the MP3/6-31G*//4-31G level of theory and corrected for zero-point vibrational energies. Apart from the previously located maxima, i.e. H,NCOOCH: * , 1, the distonic ion H,NC(OH)OCH,+*, 2, hydrogen-bridged ions IH,N-C=O*.~H**-O=CH, 1 +*, 5, and I H,N-CH=O...
---H-*-O=C-HI "*, 7, there exist at least two other equilibrium structures, viz. the iminol ion H-N=C(OH)-OCH:', la, and the hydrogen-bridged species (H,C=O---H--*N(H)COH] +', 6a, which is closely related to ion 5. Although the iminol ion la lies only 30 kJ mol-' above 1, our calculations indicate that the barriers for its formation either directly from ionized methyl carbamate 1 via a 1,Shydrogen shift or indirectly via 1,lt-hydrogen shifts from the distonic ion 2 are too high to allow the iminol ion to be involved in the unimolecular chemistry of ionized methyl carbamate. This explains the earlier observation that there are no H-D exchange reactions prior to decomposition of ionized labelled methyl carbamate, in contrast to the related ion methyl acetate. However, attempts to generate the iminol ion by loss of CH,CN from CH,CH=N-NHCOOCH, produced the more stable distonic ion 2 instead, but it proved very difficult to assign its structure unequivocally because 2 can rapidly interconvert with 1 and so virtually identical dissociation characteristics ensue. Only by integration of results obtained from many experiments and from ab initio calculations could structure 2 be assigned. The distonic ion 2 can undergo two transformations: after stretching of the C -4 C H , bond the incipient formaldehyde can migrate within the electrostatic field of ionized hydroxyaminocarbene to the OH end to generate 5, but it can also migrate to the NH end to generate 6a. This explains the previous puzzling observation that H,NCOOCD:' forms both CD,OD' and CD,OH in CA and NR experiments. The calculations and experiments indicate that, although the ion is exceedingly difficult to characterize, the distonic ion 2 is the key intermediate for all the observed dissociations of methyl carbamate.