Investigation of parameters governing the corrosion protection efficacy of fusion bonded epoxy coatings

The primary cause of corrosion in transportation structures is due to chlorides which are applied to bridge decks as deicing salts. The direct cost of corrosion damage to the country's infrastructure is approximately $8.3 billion per year. One of the most common corrosion abatement methods in the United States is the barrier protection implemented through the application of fusion bonded epoxy coatings.

The purpose of this study was to investigate various coating and exposure parameters to determine their effects on the corrosion of reinforcing steel. The parameters investigated were: chloride content at the bar depth, coated bar corroded area, corrosion product color under the coating, epoxy coating adhesion, coating color, coating damage (holidays and holes), coating thickness, TGA, DSC and EDS analysis and SEM coating cracking investigation. This was accomplished by testing new coated bar specimens as well as specimens extracted from 27 bridge decks located in Virginia.

This study demonstrated the following: The extracted ECR coating samples presented extensive cracking compared to the new ECR samples in which the coating cracking was limited to only one sample. The DSC results showed that both the extracted samples as well as new samples are not fully cured during the manufacturing process. The coating degree of curing data also showed that the bars are insufficiently and unevenly heated prior to the application of the powder coating. Additionally, the samples investigated presented significant permanent adhesion loss with little or no epoxy coating residue present on the bar surface, while the EDS analysis showed that once adhesion is lost, corrosion will proceed unimpeded under the coating even in the absence of chlorides.

The parameters that presented a direct correlation with the observed corrosion activity were the number of holidays and the number of damaged areas per unit length of bar. This indicates that the passivation of the bare steel exposed to the concrete pore solution at the breaches in the epoxy coating is not the same as a bare bar under similar exposure conditions allowing it instead to corrode at lower concrete chloride concentration levels than bare bars.

The results also show a distinct loss of quality control in the handling and possibly storage of new coated bars. The new ECR samples had significantly higher damage density than the samples extracted from concrete even though the coating is damaged during the placement of the concrete, while there was no change in the number of holidays and cure condition.

Finally, the data presented further evidence that while limited, the non-destructive corrosion assessment methods available for bare steel reinforced structures may also be used on ECR reinforced structures. In particular, the corrosion rate measurements correlated reasonably well with the chloride concentrations at bar level. This indicates that while the chlorides may not influence the corrosion activity under the coating, they do influence the corrosion activity at breaches in the coating.