Adsorption of organic corrosion inhibitors has been studied by adopting two methods, such as hydrogenation reaction with metallic catalysts, and potentiodynamic polarization determination with inhibitor-filmed electrodes.
Reduced Cu catalyst promoted hydrogenation reaction of caprylonitrile rapidly in the early stage but slowly later, while Cu-Cr-Ba oxide catalyst promoted with the constant rate. From these phenomena, produced octylamines may be adsorbed on metallic surface but not on metallic oxide surface.
Nickel catalyst stored in water could reduce acetone to isopropyl alcohol at 20°C, but hardly could one stored in alcohol or in ether at this temperature. The latter could promote hydrogenation of caprylonitrile more promptly than the former at 80°C. We would like to postulate that adsorbed water, whose oxygen atom was donating its lone pair of electrons to metallic surface of the catalyst, was pulled off by acetone with hydrogen-bridging, as shown in the following equations, revealing active points of Ni to acetone and the reaction proceeded with the production of isopropyl alcohol.
Meanwhile, caprylonitrile could not pull off the water and active points of Ni should be open to nitrile only by heating. Alcohol-filmed Ni catalyst on which the water was displaced with alcohol during the storage, as shown in the followings, acted as soon as the alcohol was desorbed from its surface at about 50°C.
To reveal active points of alcohol-filmed catalyst, some heating should be necessary for nitrile to be reduced. As adsorbed water was harder to be removed by heating than the other two, water-dipped Ni should be inferior to the others to proceed the hydrogenation of caprylonitrile to octylamines.
Potentiodynamic methods were adopted to know the behavior of amine to protect Cu and Fe, and it was found that lower molecular-weight amine, such as octylamine, demanded free molecules waiting for a chance of adsorption in the corrosive media, while higher molecular-weight amine, such as octadecylamine, could protect metals only with adsorbed molecules. These tendencies were more distinguishable for Cu than for Fe. Sometimes lower molecular-weight amine slightly corroded Cu by itself, while higher molecular-weight amine protected both Cu and Fe.

View full abstract

Arrhenius plots were made using data on uniaxially stressed specimens of Types 304, 303 Se, 201, and 431 stainless steels immersed in 35% MgCl₂ solution at various temperatures, pH of which had been adjusted to 3 at 22°C by adding HCl. The reciprocal of time-to-failure was taken as the rate of stress-corrosion cracking. Although this is not exactly correct, the results were in good agreement with the Arrhenius' equation. Furthermore, it was found that both the activation energy and logarithm of the rate coefficient, i. e. the pre-exponential term, increased in direct proportion to the stress-level applied and, in addition, the lines illustrating the above relations breaked at a stress-level corresponding to 0.03% proof yield strength of Types 304 and 303 Se steels. As for the other steels such a break was not evidently observed.
Arrhenius' equation describes only the temperature dependence of a reaction rate and does not include the entropy changes, so that the activation energy corresponds to the activation enthalpy. Thus, the stress-dependent term, associated with the rate coefficient but not with the enthalpy, should belong either to the entropy term, if it depends on stress, or to another rate coefficient of the probability expression in terms of the free energy, if it does not depend on stress.
The analysis of the results leads to the following conclusions: (1) if stress influences entropy, applied stress effectively decreases the activation free energy, and on the other hand the activation enthalpy increases with increasing stress; (2) if the entropy does not depend on stress, applied stress extraordinarily increases the frequency factor that depends on stress. This would mean that the density of active sites is augmented to very high order. Both the activation enthalpy and free energy increase with increasing stress.
In both cases, therefore, rate of stress-corrosion cracking increases with applied stress. It can not be made clear, however, on the basis of this kind of analysis, which view is true, and, furthermore, there would be a case in which stress could influence both the entropy and the frequency terms simultaneously. At present, one can only say that the stress-dependent term, associated with thermal excitement of atoms but not with potential energy, is a sort of environment-sensitive parameter that may be able to control the susceptibility to stress-corrosion cracking, although the detailed mechanism involved can not be clarified.

View full abstract

The rates of cathodic reduction of oxygen and hydrogen peroxide were compared for rusted lowalloy steel and mild steel by using the potential sweep method in neutral solution of 1M sodium sulfate.
The apparent reduction current of hydrogen peroxide decreased as the amount of rust-oxide increased by longer outdoor exposure, especially over thirty days.
It can be concluded that the rust layer formed on low-alloy steel shows greater contribution to the retardation of further atmospheric corrosion than does the rust on mild steel.

View full abstract