ON ABSORPTION SPECTRA OF ORGANIC COMPOUNDS. 313 



In tliis type of reaction there is another large source of error in the calcula- 

 tion. It is assum€d that the energy is only absorbed by the reacting substance 

 at its molecular frequency in the sliort wave infra-red, and it is the amount of 

 energy radiated at this frequency only by the walls of the containing vessel 

 that is used as the basis of calculation. This assumption of monochromatic 

 absorption of energy by the phosphine is incorrect according to the phase theory, 

 since energy can also be absorbed by the gas at its atomic and intramolecular 

 frequencies, the smaller quanta tiius absorbed being summed up witliin the 

 molecule to form molecular quanta. The total number of molecular quanta 

 gained by the gas per second is, therefore, much larger tlian that calculated 

 from the radiation by the walls of energy of frequency equal to the molecular 

 frequency of the phosphine. When the increase in reaction velocity due to 

 this is taken into account, together with the resulting increase in the proportion 

 reabsorbed of the energy radiated during the second and third stages of the 

 reaction, the very great divergence of the observed results from those calculated 

 can easily be understood. 



An interesting result follows from the increase in the proportion reabsorbed 

 of the energy radiated during a reaction caused by an increase in the reaction 

 velocity. In an exothermic reaction, if this radiated energy is completely 

 reabsorbed the surrounding molecules will be completely activated and will 

 react. Under these conditions the absorption of a single energy increment, E, 

 by one molecule will be sufficient to cause an infinite number of molecules to 

 react, and the reaction will proceed as an explosion wave through the whole 

 mass of the substance. The criterion of an explosive reaction is, therefore, 

 that the proportion reabsorbed of the energy radiated during a reaction reach 

 a critical limit, so that the surrounding molecules are activated. The existence 

 of a limiting intensity of illumination below which explosive combination of 

 hydrogen and chlorine does not take place is well known. Again, the effect of 

 wire gauze in stopping an explosion wave in a gas is due to the absorption by 

 the metal of energy quanta from the gas molecules, which thereby lose their 

 critical increments of energy and therefore can no longer react. 



It may be claimed that the phase theory receives strong support from the 

 quantitative experiments described above. This support is further strengthened 

 by an extension of the principle of the reabsorption of ihe energy radiated 

 during a reaction. Let a substance A undergo a photochemical reaction when 

 it absorbs light of frequency Vi ; if a beam of light pass through a vessel 

 containing A it will be deprived of all rays of frequency Vi and will therefore 

 produce no effect when it enters a second vessel containing A. If a substance P, 

 which absorbs light of frequency V2, and is not thereby photochemically changed, 

 is mixed with A in the second vessel, this substance P will absorb rays of 

 frequency V2 and will radiate this energy at its molecular and atomic fre- 

 quencies. If the two substances A and P possess the same molecular or atomic 

 frequencies, this radiated energy can be absorbed by A, with the result that 

 some molecules of A will be activated and will undergo the same reaction as 

 they do when exposed to rays of their own phase frequency, Vi. 



The first example of this type of reaction, to which the name of 'photo- 

 catalysis ' has been given, was described by Daniels and Johnson (./. Annr. 

 C'hem. Soc, 43, 72 (1921)). These authors found that N2O5 is not affected 

 by blue light, but that, if some NO2 which absorbs this light is mixed with 

 the N2O5, the latter is decomposed when the mixture is exposed to blue light. 

 The energy absorbed by the NOo is radiated at its molecular and atomic fre- 

 quencies, and, since the two molecules have the same atomic frequencies, some 

 of this radiated energy is reabsorbed by the NoOs, which is thereby photo- 

 chemically decomposed. 



A second example of photocatalysis was found in the photochemical con- 

 version of carbonic acid into formaldehyde. This reaction normally takes place 

 when carbonic acid is exposed to light of very short wave-length, 200 /^i/i, but 

 it has been found possible by the use of a photocatalyst to realise this reaction 

 with the aid of visible light only. The ideal photocatalyst will obviously be 

 one which possesses exactly the same molecular frequency as the substance to 

 be photocatalysed, and, as previously explained, this is secured by using a 

 photocatalyst which forms an addition complex with the catalyst. In the case 

 of carbonic acid an excellent photocatalyst is found in malachite green, which 



