76 CHEMICAL REACTIONS IN THE GAS PHASE 



energy curve diagram, gives only an upper limit for the amount of 

 energy necessary for a given process. That is, referring again to Fig. 1, 



^(A+) >:-D(AB) + -^(A) = -^(AB) + -D(AB+) 



In recent years, workers in mass spectra have succeeded remarkably 

 well in building instruments in which second-order reactions involving 

 ions and molecules are almost completely absent. This was not the case 

 in older instruments, and indeed some instruments were constructed 

 especially for the study of secondary reactions by differential pumping 

 arrangements which permitted independent control of the gas pressure 

 in the analyzer regions (14). The essential result of such work is that 

 the probability of charge exchange reactions, 



A+ + B -^ A + B+ 



varies inversely with the size of /(a) — 1(B), and only on the absolute 

 value of this difference if A+ has much kinetic energy. It may be very 

 large for 7(A) — /(b) ~ 0. For the special case of charge exchange where 

 A and B are the same, as 



N2+ + N2 -> N2 + N2+ 



the cross section for the reaction can be orders of magnitude greater 

 than expected from the kinetic theory diameters. 



Interesting applications of mass spectrometric data in combination 

 with thermal and kinetic data are contained in several papers by Eyring, 

 Hirschf elder, and Taylor (15). In the ori/io-para-hydrogen conversion 

 by alpha particles, 700-1000 molecules are converted per ion pair 

 formed (16). The ionization by alpha particles is actually largely due 

 to the secondary electrons of energies comparable to electron energies in 

 the mass spectrometer. From the ratio of ion pairs produced to alpha- 

 particle energy, and the mass spectrometric data on hydrogen, it is 

 deduced that the primary effect of the secondary electrons is to produce 

 approximately equal amounts of (2H) and £[2"^, and much smaller 

 amount of H+. Absolute reaction rate calculations indicate that the 

 reaction 



H2+ + H2 = H3 + + H 



is very rapid, and that the reaction 



H2 + H2+ = H2 + H + H+ 



is of no importance. The formation of "clusters," the formation of H~ 

 and Il2~, and the ortho-para conversion directly by alpha particles and 

 by H2"^, H3+, and H are shown to be negligible. Recombination of H3+ 



