PAPERS ON CHEMISTRY AND PHYSICS 139 



either ''primary" or " secondary". For "primary" 

 light reactions the number of molecules reacting per 

 quantum absorbed is either one or some small number. 

 In this series of reactions the free energy change is 

 either positive or very slightly negative (See Table I) 

 for those reactions for which the free energy change is 

 known. It seems probable that this could be stated as a 

 general law. 



TABLE L 



Primary Light Reactions (Bodenstein) 



2 HI = H, + I, ; AF% 8 = —630 



3 O a = 2 3 ; = +64800 

 2 NH 3 = N 2 + 3 H 2 ; = + 7820 



2 H„0 = 2 H 2 + 2 ; = + 10901 4 



SX = S/x ; — 1 ( ?) 



In the case of "secondary" light reactions, one 



quantum causes a large number of molecules to react. 



These reactions are almost always those which involve 



a large decrease in free energy, and the light seems 



more to play the role of a catalyst. 



TABLE II. 



Secondary Light Reactions (Bodenstein) 



Ho + a = 2 HC1 ; AF% S = —45384 



2 O s = 3 O, ; = — 64800 



2 H„0 2 = 2 H,0 + Oo ; = —56660 



CO -f Cl 2 == COCl 2 ; = — 16-260 



It would seem safe to predict, then, that only those re- 

 actions which involve a slight negative or a positive free 

 energy change will follow a photochemical equivalence 

 law. For the other reactions light seems to be capable of 

 starting a chain process which continues until it comes 

 to an accidental end. As Nernst 9 has pointed out an 

 "acceptor" which neither multiplies nor diminishes the 

 products of the primary reaction, but transforms them 

 directly into the equivalent quantity of finally measured 

 product should give rise to a reaction which obeys the 

 photochemical equivalence law. This point has been 

 studied by Pusch 10 , who studied the action of bromine on 



•Nernst. Zeit. Elektrochem.. 2k, 335 (1918). 

 "Pusch, ibid., 2',, 336 (1918). 



