International Journal of Mass Spectrometry and Jon Physics, 29 (1979) 205-222 205 0 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Eisevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands THE GAS PHASE ALLYL ANION J.H.J. DAWSON, A.J. NOEST and N&KM. NIBBERING Departmenr. of Organic Chemistry, University of Amsterdam, Nieu we Ach tergrach t 129, Amsterdam (The Netherlands) (First received 20 March 1978; in revised form 26 June 1978) ABSTRACT Ca(H/D); ions are generated in an ion cyclotron resonance mass spectrometer from the reaction of O- with labelled propenes and their reactions with nitrous oxide and car- bony1 fluoride are studied with a view to showing that little, if any, internal hydrogen scrambling takes place within the long lived ally1 anion. It is also reported that the reaction of 0’ with propene gives rise to the ions CsHr_ CHaCO’CH2 and CH$H=CHO- albeit as only a small proportion of the tota! reaction cross section. A reaction of the CaHr ion ia reported, which, under more favourable experimental conditions, might enable the structure of the ion to be determined. INTRODUCTION zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA It has been demonstrated beyond reasonable doubt that long lived C,H;- ions have the structure CH2=C- [l] and studies of the ion-molecule reac- tions of C+Ha- and CSH; [Z] have shown that both of these ions exist in two stable non-isomerising forms: ‘CH=C=CH- and ‘CHZ--C!=C-, and CH3-C-C’ and -CM-ECH respectively. In this paper the C3H; ions generated by proton abstraction from propene will be considered. It is expected that C,H; ions will have the planar ally1 struc@re CH,=CH-CH,‘, but there are some conceivable alternatives e.g. CH3-C=CH2, CH,-CH=CH- and ~H,-CH,+H-. The list is not exhaus- tive, for example the cluster ion HC!X!-.CH, has not been includet!, but it is rather doubtful if this or any of the other possibilities could be of reti them- ical significance. In fact, we are not aware of any reported observation of a stable CH2=CH- ion, and so doubt must be expressed concerning the two vinylic structures listed above. Theoretical calculations [3] have predicted the relationships between the ally], vinyl and cyclopropyl’ structures (see Fig. 1). It should be noted that no isomerization barriers have been calculated for the interconversion between the vinylic and the other CJH; structures. Under normal circumstances it would appear thzit both the cyclic and the allylic structure of the CJH; ion should exist, with little probability.of inter- conversion. There is also the point that the exothermicity of the ring-