POLYATOMIC MOLECULES 85 



(c) The ion formed in this initial break will probably have energy 

 enough to break down further. Loss of fragment C2H4, C3H6, etc., is 

 highly probable, the weakest C — C bond breaking. Also very highly 

 probable is the loss of hydrogens in pairs, a process for which AH is 

 only about 13^ volts, although there is certainly an activation energy 

 associated with the process of forming the H2 molecule. 



(d) In each breakdown of the ion, there is a chance of the charge 

 going with either fragment, the relative probabilities of obtaining the 

 two sets of products being given by the usual Boltzmann factor involving 

 the difference in ionization energy for the two fragments. If this 

 difference in ionization potential is over a few tenths of a volt, only the 

 more easily ionized fragment will be found. For example, the metastable 

 transition C4Hio"^ — ^ C2H4"'" + C2H6 is observed, but the transition 

 C4H10"'" — > C2H4 + C2H6''" is not observed, either metastable or other- 

 wise. 



(e) A few small peaks that must be due to doubly charged ions are 

 found; a much larger fraction of the doubly charged ions formed surely 

 immediately dissociates to give two singly charged ions. Although the 

 number of ions so produced will be but a small fraction of the total ioniza- 

 tion, it probably accounts for much of the relatively constant amount of 

 CH3 produced from all large hydrocarbons. 



It is obvious that the complexity of the situation makes these rules 

 have only the barest qualitative significance and that they are only of 

 limited applicability. The mass spectrum of cyclohexane (23) can 

 reasonably be interpreted in terms of these rules, but the mass spectrum 

 of benzene cannot without special consideration of the effect of the large 

 number of electrons. 



To sum up, the essential idea in the foregoing discussion is that the 

 process of ionization of the molecule is accompanied by the simultaneous 

 transfer of excess energy to the other electrons of the molecule, and that 

 instead of an immediate dissociation occurring this energy undergoes a 

 process of rearrangement, being transferred from the electronic to the 

 vibrational states of the molecular ion. As has already been mentioned, 

 we have at present no way of knowing the distribution function for this 

 excess energy. Moreover, we cannot be certain of how much of this 

 energy is transferred to the vibrational states of the molecule. It is 

 likely that at least some of the radiationless transitions are very rapid, 

 but there may, as already mentioned, be some transitions which are 

 much slower than the total lifetime of a molecule in the mass spectrom- 

 eter. The fact remains, however, that there will be some sort of dis- 

 tribution function for the vibrational energy and that in view of the 

 very large number of electronic states this function will approach a 



