28 BELL SYSTEM TECHNICAL JOURNAL 



The corrosion reaction represented by the first reaction automatically 

 stops when the metal surface becomes covered with hydrogen atoms 

 and can only proceed when and as hydrogen is removed by one of the 

 processes above given. Some metals evolve molecular hydrogen if the 

 potential of the corrosion cell is slightly over that required to plate out 

 hydrogen atoms. In the greater number of cases, however, molecular 

 hydrogen is not liberated rapidly, i.e., as bubbles, unless the potential 

 is substantially higher than that required to plate out the atomic form. 

 The additional potential required to evolve molecular hydrogen 

 against the normal atmospheric pressure is termed "overvoltage." 

 In the case of metals of high hydrogen overvoltage, oxygen or oxidizing 

 substances are required to facilitate the removal of the hydrogen film 

 if corrosion is to proceed at an appreciable rate. Consequently cor- 

 rosion may be controlled by the rate at which oxygen reaches the metal 

 surface and "depolarizes" it. In many cases both processes are 

 operative. For example, when iron is corroded in dilute potassium 

 chloride solutions it is interesting to note that under a pressure of 

 one atmosphere of oxygen 1/13 of the total corrosion is accompanied 

 by the discharge of hydrogen gas and 12/13 by the oxidation of 

 hydrogen to form water. When the oxygen pressure on the system is 

 raised to 25 atmospheres by conducting the experiment in a closed 

 bomb the total rate of corrosion is increased 45-fold owing to the in- 

 creased rate of hydrogen oxidation, the rate of hydrogen evolution 

 being practically unaffected. In practical cases, as would be expected, 

 either the character of the corroding medium or the purity of the metal 

 may affect this ratio of hydrogen control to oxygen control. In tests 

 in which iron specimens were totally immersed in sea water exposed 

 to oxygen, about 35 per cent of the total corrosion was accompanied 

 by hydrogen evolution, whereas for similar specimens in a half- 

 normal solution of pure sodium chloride (which corresponds roughly 

 to the salt content of sea water) only 5.6 per cent of the corrosion was 

 of the hydrogen evolution type.'^^ The presence of one part per 

 million of impurity in zinc considerably accelerates the rate of corro- 

 sion 2^ mainly by stimulation of hydrogen evolution. 



The intensity with which a metal tends to send metal ions into 

 solution increases with the basic character of the metal. The presence 

 of ions of the metal in the solution may be considered to constitute a 

 force opposing this ionization tendency and the value of the resulting 

 equilibrium is known as the potential of the metal in that solution. 

 The value is constant or static only so long as there is no flow of 

 current between the metal and solution. The molal potentials and 

 normal potential of metals are their static potential in solutions of 



