Characterization by theory of H-transfers and onium reactions of CH3CH2CH2N+H=CH2.

H-transfers by 4-, 5-, and 6-membered ring transition states to the pi-bonded methylene of CH3CH2CH2NH+=CH2 (1) are characterized by theory and compared with the corresponding transfers in cation radicals. Four-membered ring H-transfers converting 1 to CH3CH2CH=N+HCH3 (2) and CH3N+H=CH2 to CH2=NH+CH3 are high-energy processes involving rotation of the source and destination RHC= groups (R = H or C2H5) to near bisection by skeletal planes; migrating hydrogens move near these planes. The H-transfer 1 --> CH3C+HCH2NHCH3 (3) has a higher energy transition-state than 1 --> 2, in marked contrast to the corresponding relative energies of 4- and 5-membered ring H-transfers in cation-radicals. Six-membered ring H-transfer-dissociation (1 --> CH2=CH2 + CH2=N+HCH3) is a closed shell analog of the McLafferty rearrangement. It has a lower energy transition-state than either 1 --> 2 or 1 --> 3, but is still a much higher energy process than 6-membered ring H-transfers in aliphatic cation radicals. In contrast to the stepwise McLafferty rearrangement in cation radicals, H-transfer and CC bond breaking are highly synchronous in 1 --> CH3N+H=CH2 + CH2=CH2. H-transfers in propene elimination from 1 are ion-neutral complex-mediated: 1--> [CH3CH2CH2+ ---NH=CH2] --> [CH3C+HCH3 NH=CH2] --> CH3CH = CH2 + CH2=NH2+. Intrinsic reaction coordinate tracing demonstrated that a slight preference for H-transfer from the methyl containing the carbon from which CH2=NH is cleaved is due to CH2=NH passing nearer this methyl than the other on its way to abstracting H, i.e., some memory of the initial orientation of the partners accompanies this reaction.