ON ABSORPTION SPECTRA OF ORGANIC COMPOUNDS. 227 



solution is equal to that of the absorption band shown by that compound when 

 in the reactive phase. The same frequency therefore is characteristic of a given 

 substance in two solvents, in one of which it is exhibited as emission and in the 

 other as absorption. It is evident, therefore, that the constitution of each 

 compound is the same in the two cases. 



Very important conclusions may be drawn from these observations, namely, 

 that a given compound can exist in at any rate two phases which differ in 

 their reactivity and which are characterised by different absorption bands. 

 Also the absorption bands shown by the reactive phases are nearer to the red 

 end of the spectrum. It is therefore an obvious deduction that a definite 

 absorption band is associated with a definite type of reactivity. 



The next question to consider is whether an explanation of these facts can 

 be found. In the theories of absorption spectra given above no; reference is 

 made to the ultimate destination of the light which is being absorbed. It is 

 perfectly obvious that, unless the absorbing compound undergoes a photochemical 

 change, the total amount of energy absorbed must again be radiated. It is 

 equally evident that just as the light energy is absorbed at frequencies which 

 are characteristic of the absorbing substance, so also must this energy be 

 radiated at frequencies characteristic of the substance. Careful experiments 

 have proved that, provided the absorbing substance or its solution is free from 

 dust, there is no evidence of radiation at the frequencies which lie within tho, 

 absorption band. Clearly, therefore, the phenomenon of absorption is not one of 

 optical resonance, that is to say, the light energy absorbed by a substance is 

 radiated at frequencies which are not the same as those at which it has been 

 absorbed. Except in those cases where fluorescence or phosphorescence is 

 observed, the whole of the absorbed energy is radiated at frequencies which lie 

 in the infra-red region of the spectrum, and we have therefore — • 



Energy absorbed (visible or ultra-violet) = energy radiated (infra-red). 



This necessarily establishes a relationship between the various frequencies 

 exhibited by a substance in the infra-red, visible, and ultra-violet regions, and, 

 indeed, invites investigation of this relationship. 



It will be remembered that Planck formulated the theory that absorption 

 and radiation of energy are not continuous processes, but are discontinuous in 

 the sense that the energy is absorbed or emitted in a series of fixed amounts. 

 To these fixed amounts he gave the name of energy quanta, and he showed that 

 the size of the quantum is given by the product of the frequency into a universal 

 constant, the most recent value of which is 6-56 x lO"^'. According to this 

 theory, therefore, if a substance is absorbing light with a frequency of, say, 

 9 X 10", the process is not continuous, but each molecule absorbs a .series of 

 quanta, each of which is 9 x 10'^ X 6-56 x lO'^^, or 5-904 X lO'^ ergs. 

 Without discussion of the fundamental basis of this quantus theory it may be 

 applied to the problem of the absorption and radiation of energy by a molecule 

 when, as already explained, the total quantity of energy absorbed is radiated 

 at another and smaller frequency. Let a molecule absorb one quantum of light 

 energy at its absorbing frequency. This energy is then radiated at another 

 and smaller frequency, but it must be radiated as a whole number of quanta at 

 that frequency. It follows, therefore, that when a molecule is absorbing at 

 one frequency and radiating at another and smaller frequency, one quantum of 

 energy at the larger frequency must be equal to a whole number of quanta at 

 the smaller frequency. Finally, since the quantum is the product of the fre- 

 auency into the universal constant, the conclusion is reached that the absorbing 

 frequency must be an exact multiple of the radiating frequency. In other words, 

 the frequencies of each absorption band shown by a substance in the visible 

 and ultra-violet must, on the basis of Planck's theory, be an exact multipJe of a 

 frequency characteristic of that substance in the infra-red. It was not difficult 

 to test the validity of this deduction since the existence of characteristic 

 frequencies in the infra-red possessed by a substance can be proved by the 

 method of absorption spectra observations in that region, and indeed a o-reat 

 nu-^bor of substances had already been in vestisated in this manner. 



It may be stated at once tJiat the relation has been found to be true in *he 



Q3 



