PHYSICAL PRINCIPLES OF CHEMICAL REACTIONS 207 



translational energy of heavy particles (cf. Sect. 3-2c for explanation). 

 Quenching of the fluorescence by type III is also impossible, because the 

 lowest excitation energy of the rare gases far exceeds 4.86 ev. With rare 

 gas atoms, a small probability does exist for transfer of the d^Pi atoms 

 to the 6To state; this is evidenced by a prolonged afterglow. (The 



reaction 



O* + Ne -^ O + Ne 



has, however, been detected; the cross section is only 10^^^ cm^.) 



III. An example is 



Hg* + Tl ^ Hg + Tl* 



sensitized fluorescence of Tl* being observed. The only Tl Hnes which 

 occur are those having an excitation energy lower (or only slightly 

 greater) than 4.86 ev. Many other cases have been investigated in detail. 

 The cross sections for these processes are usually quite great. In the 

 case of the sensitized fluorescence by Hg* of Na vapor, the validity of 

 the first general principle mentioned (resonance) is convincingly demon- 

 strated by the much greater fluorescence from that excited state of Na 

 which has excitation energy (4.88 ev) closest to the excitation energy of 

 Hg*. 



IV. An example is 



Hg* + N2 -^ Hg (6^Po) + n; 



the N2 being left in an excited vibrational state. This process is proved 

 by observation of absorption lines originating in the 6^Po state, or by a 

 resulting delay in the fluorescence of Hg; Hg atoms in the 6^Po state have 

 such a small radiative transition probability that they live long enough 

 to be returned to the 6*Pi in a subsequent collision with an N2 molecule. 

 It is also demonstrated by the fact that, although N2 quenches the Hg 

 resonance radiation, when it is added to a mixture of Hg* and Tl it 

 increases the sensitized fluorescence. (The metastable 6^Po atoms are 

 virtually nonfluorescent, but readily excite Tl by process III; the sen- 

 sitized fluorescence is increased because its competition with Hg fluores- 

 cence is favored. The greater efficiency of metastable atoms for collisions 

 of the second kind is, as must be anticipated, a common phenomenon.) 



If O2 is used in place of N2, the fluorescence is quenched and cannot be 

 recovered ay sensitized fluorescence if Tl is added. The process is 

 probably 



Hg* + O2 -^ Hg (ground state) + O? 



V. No example of this type has, apparently, been observed for Hg*, 

 doubtless because 4.86 ev is smaller than the ionization potentials of 

 almost all atoms and molecules. The process is, however, a familiar 

 one in cases for which energy conservation is satisfied (Massey and 



% 



