64 PHOTOCHEMICAL PRINCIPLES 



by 6000-7000 A light. Furthermore, decay times of the order of many 

 seconds are not in the range to be expected for radical recombinations 

 at relatively high temperatures. Finally, it is difficult to reconcile such 

 a mechanism with the existence of three separate emissions of the 

 same wavelength. 



The excitation and decay of a long-lived triplet state, as in mecha- 

 nism (2), is incompatible with the observed definite temperature de- 

 pendence of the chloroplast luminescence, i.e., it is very unlikely that 

 lowering the temperature to — 100°C would increase the triplet life- 

 time to the order of hours. Furthermore, such a mechanism cannot 

 result in three separate emission acts having different time constants 

 but of the same wavelength. 



If enzymatic processes were involved here, as in mechanism (3), 

 cooling to — 140°C should decrease the rates of these processes to 

 essentially zero. This is not in accord with the fact that the rise time 

 and the concentration of unpaired spins are about the same at this 

 temperature as at 25 °C. Similarly, the presence of the 0.15-sec emis- 

 sion down to as low a temperature as — 100°C rules out the participa- 

 tion of enzymatic processes in either the forward or reverse transforma- 

 tions in this case. If, then, only the 2- and 15-sec emissions represent 

 chemical processes, one would expect that cooling, by preventing the 

 reaction leading to radical formation from taking place, would result 

 in a greater amount of energy appearing in the form of the 0.15-sec 

 decay. In fact, the emission at — 80°C is less than it is at room temper- 

 ature. Such a viewpoint is supported by the aging experiments men- 

 tioned earlier. Thus, if one assumes that the aging process involves the 

 inactivation of enzymes, the creation of centers (or radicals) for the 

 2- and 15-sec emission processes by enzymatic means should be re- 

 duced. This reduction of competitive processes should then lead to an 

 increase in the 0.15-sec emission intensity together with a concomitant 

 decrease in the 2- and 15-sec emission intensities. In fact, for aging 

 periods up to 8 hr, all three emission intensities are increased by the 

 same amount. 



We are thus left with mechanism (4) as the most likely explanation 

 for the phenomena we are reporting here. We shall next see how such 

 a scheme fits the data. Figure 8 shows a schematic representation of 

 the electronic energy bands in chloroplasts. Light is absorbed to pro- 



