OXIDATION AND REDUCTION 581 



electrical phenomena are associated with the activation of oxygen, as shown by 

 the following facts. 



Oxidation and reduction do not always mean the addition or removal of 

 oxygen or hydrogen. The change from ferric to ferrous salts is a reduction, but 

 consists in the conversion of trivalent iron to bivalent iron. 



Haber (1898) has shown that the reducing action of hydrogen, developed on an electrode, 

 depends on the electrical potential there. In this way, a reduction process can be carried to a 

 particular point and no further. Thus, nitre-benzene can be reduced to azoxy -benzene and no 

 further, if the cathode potential is low. Dony-Henault (1900) also showed that alcohol can 

 be quantitatively oxidised as far as aldehyde, with a proper anode potential. 



The phenomena connected with the autoxidation of phosphorus and its effect 

 in condensing a steam jet were referred to above (page 31). We may note that 

 the effect consists in the production of " gas ions " and is not shown by the 

 products of the reaction, but only by some process taking place in the actual 

 course of the reaction itself. 



Ostwald (1890, p. 76) suggested that reduction means a diminution of charge, 

 that is, a loss of (negative) electrons. 



Oxygen, as we know, may be bi- or quadrivalent. In the peroxides it is 

 probably the latter, and, when split off in a particular way, it has unusually 

 powerful oxidising properties, a fact which is of great importance in physiological 

 oxidations. Its activity appears to depend on its readiness to give up its extra 

 charges. 



AUTOXIDATION 



The theory of the process which takes place in the spontaneous oxidation of 

 phosphorus, benzaldehyde, or other such substances, was suggested by Bach 

 (1897), and by Engler and Wild (1897), independently, and was adopted by 

 Ostwald (1900, 1). It has been observed as an experimental fact that, in such 

 reactions, there are formed simultaneously two oxides in equivalent proportion, 

 a lower oxide and a peroxide. Now, in the production of the former, energy is 

 given out, whereas the latter has a higher oxidation potential than the oxygen 

 gas, and requires energy to form it. This energy is derived from that afforded 

 by the production of the simple oxide. 



We saw, in speaking of "coupled reactions," that these reactions, in which 

 energy afforded by one reaction is used to enable another to take place, must be 

 capable of being expressed as parts of one and the same complete reaction ; we see, 

 then, why there is always a quantitative, equivalent relation between the simple 

 oxide and the higher oxide. Schonbein (see the monograph by Engler and 

 Weissberg, 1904, p. 9) was the first to point out that half the oxygen is used to 

 oxidise the substance itself, while the other half is "activated." The peroxide 

 obtained in the oxidation of phosphorus is ozone and its formation should be 

 described thus. In the process there is first formed a peroxide of phosphorus, 

 which then splits up into ozone, on the one hand, and a lower oxide of phosphorus 

 on the other hand. Thus : 



mP + nO 2 = P m O 2 (1) 



p m 2 -*.<>.* + r0 3 (2) 



(intermed. oxide) = (l wer oxide) + (higher oxide) 



Whether this intermediate oxide is to be detected or not depends on the rate of its 

 decomposition. 



The way in which the above reaction is described is that of Ostwald. It certainly is in 

 agreement with experimental facts and explains why the two products of an autoxidation are 

 always in equivalent proportion, since they were at one time combined in one substance. 

 Ostwald also points out that it is a general rule that the most unstable product of a reaction 

 makes its appearance first and is then decomposed. 



It is interesting to call to mind also what Larmor has pointed out (1908). If 

 we consider the great distance between the molecules of a gas as compared with 

 their own dimensions, it is easy to see that an impact between molecules takes 

 place only at a comparatively rare frequency. Suppose that the molecules are of 



