Abstract
Enameling steels can absorb hydrogen during enamel firing. The hydrogen dissolved in the lattice is released during the component cooling. It diffuses and accumulates in the steel-enamel interface, generating the detachment of small portions of the enamel. This phenomenon is known as fish-scale and represents the most dangerous damage in enameled steel product manufacturing. The microstructure of steel is characterized by the presence of precipitates, grain boundaries, voids, inclusions, etc., which can trap hydrogen, preventing it from diffusing. Therefore, the susceptibility of steel to fish-scale depends on the content, number, and type of trap sites. Electrochemical permeation tests represent the test traditionally used to study the phenomenon of diffusion and trapping in steel membranes. The test consists of the hydrogen generation on one face of the membrane and in the measurement of the hydrogen flow exiting from the opposite face. The latter is influenced by the trapping mechanism, therefore, through its determination, it is possible to determine the diffusion and trapping properties of hydrogen in steel. The hydrogen effective diffusion coefficient, i.e., the diffusion coefficient that the material would have if the diffusion were governed by Fick's laws, is traditionally used to obtain indications about the susceptibility of steel to fish-scale. It is influenced by the trapping phenomenon and represents a simple criterion for the quality assessment of steel. However, its determination does not allow an adequate description of the physics of the phenomenon leading to incorrect evaluations of the steel susceptibility to fish-scale. Steels with the same hydrogen effective diffusion coefficient can have completely different tendencies and fish scales. This work fits into this context and aims to provide a criterion for assessing the susceptibility of fish-scale steel by interpreting electrochemical permeation tests through a more accurate trapping model. Permeation curves carried out on different enameling steels were carried out and interpreted through a diffusion and trapping model based on the local equilibrium between trapped hydrogen and hydrogen upon normal lattice sites. The constitutive parameters of the model were identified and correlated with the manifestation of the fish-scale phenomenon. The validity domain of the local equilibrium hypothesis was discussed, a criterion for assessing fish-scale susceptibility was provided.