The previous study revealed that the performances of two liquid-applied, 100% solids epoxy pipeline coatings (coating A and coating B) in immersion conditions are greatly influenced by the types of abrasives used to prepare the surfaces. A further investigation was conducted to better understand the results: (i) coating A performed superior on steel grit and shot abrasive prepared surfaces compared to coating B; (ii) coating A was more susceptible to blistering on surfaces prepared with coal slags; (iii) both coatings performed poorly on surfaces prepared with glass grits; (iv) surfaces prepared with mineral slag and mineral blends abrasives generally showed better performance than those prepared with other non-metallic abrasives.
A literature review was conducted to identify the mechanism of epoxy coatings to steel substrate adhesion, the effect of water interaction with epoxy, and detection techniques for the absorbed water. The levels of embedment on the blasted surfaces were re-assessed using an improved approach and the potential for blistering due to soluble salt contaminants in the abrasives was examined. Focuses were given in experimental examination to monitor change in water uptakes, corresponding change in glass transition (Tg) temperatures, and the status of coating/substrate interfacial conditions (retained adhesion and presence of corrosion under microscope) when the representative coated panels of the two coatings were subjected in hot water immersion at 75°C. Subsequently, the results of these experiments led to possible explanations for the observed phenomena.
Dry abrasive blast cleaning is a commonly used surface preparation method to achieve the necessary surface cleanliness and roughness for the application of high-performance protective coatings. The effectiveness of abrasive blast cleaning depends on various factors, and some studies have indicated that the blasting abrasive used to clean the surface is crucial for the performance of the coating1,2. The previous work1 evaluated the performances of two liquid-applied epoxy pipeline coatings (coating A and coating B) applied to 15 different surfaces. The abrasive products used included metallic, glass, garnet, coal slags, mineral slag, and mineral blend abrasives. To support that it was blast material itself, not the resultant anchor profiles causing significant variation in coating performance, four of the abrasive products were intentionally used two different sizes and both new and recycled steel grits were included for blasting. The study by A. Recker and D. Corbin2 tested three high performance, high solids epoxy linings on surfaces produced by 12 blast media, including garnet, coal slag, steel slag, and glass / silica. It is interesting to note that the two different studies1,2 revealed the following common findings:
• Initial dry adhesion did not correlate with coating performance, as all abrasives selected for the studies could produce the required NACE No. 2/SSPC-SP 10 surfaces and the anchor profiles, and the prepared samples for performance testing all achieved very high initial adhesion (above 3,000 psi).
• Despite both studies made a great effort to characterize the surfaces produced by different blast media (anchor profile, roughness, peak counts/density, level of embedment, surface contamination, or SEM/EDX analysis, etc.), the cause of the variation in coating performance could not be interpreted due to lack of correlation between them.
• Mineral slag abrasive (identified as magnesium silicate by the other paper) consistently produced the best performing surface.
• The surfaces produced by other abrasives were found to be more selective with the test coatings, as indicated by significant variation in performance were observed, e.g., one coating performed well, but another coating on the same surface showed blistering and significant adhesion loss in the same tests.
