Cathodic disbonding of a thick polyurethane coating from steel in sodium chloride solution J.L. Luo a, *, C.J. Lin b , Q. Yang a , S.W. Guan c a Department of Chemical and Materials Engineering, University of Alberta Edmonton, AB T6G 2G6, Canada b Department of Chemistry, Xiamen University, Xiamen, China c Madison Chemical Industries Inc., 490 McGeachie Drive, Milton, ON L9Y 3Y5, Canada Received 14 August 1996; accepted 9 June 1997 Abstract The cathodic disbonding of a thick, pigmented polyurethane coating from steel in 3.5 wt.% NaCl solution was studied by using an electrochemical AC impedance technique. Double-cylinder electrolyte cells were designed to separate the measurements of cathodic disbonding process from the influence of the impedance of an artificial defect. It was found that for a thick, pigmented polyurethane coating, the more important transport pathway of the reactive species is along the coating/steel interface rather than through the coating. There existed a delay time for the cathodic disbonding process, and cathodic polarization was not a predominant factor in determining the cathodic disbonding behavior in the early stages. The thick polyurethane coating, which was applied on a well sand-blasted steel surface, had excellent resistance to cathodic disbonding.  1997 Elsevier Science S.A. Keywords: Cathodic disbonding; Thick polyurethane coating; Double-cylinder electrolyte cells; AC impedance 1. Introduction Cathodic disbonding is one of the main degradation pro- cesses for organic coatings on steel structures under catho- dic protection. The cathodic protection prevents corrosion but causes the cathodic surface to become strongly alkaline and generate hydrogen at exposed metal surfaces. The com- bined effect is to cause a loss of adhesion of the organic coating around damaged areas and to allow ingress of elec- trolyte. It is therefore essential that coatings be tested for their tendency to disbond under these conditions. Cathodic disbonding tests were developed in the 1960s, leading to the publication of ASTM G8 in 1969, in which an artificial defect was introduced to simulate the damage areas of a coating. Many other versions of the test based on the same principle have come into use to suit other conditions or types of coatings. Nevertheless, the mechanism of catho- dic disbonding of an organic coating has not been very clear and many inconsistent testing results were obtained in the industry. Over the past decade a few studies have been conducted to reveal the mechanism of cathodic disbonding by using DC and AC electrochemical techniques [1–6]. Several valu- able models have been proposed [7–12]. However, there are still two remaining problems that need to be considered. (1) Previous electrochemical measurements were per- formed directly on an artificial defect on a coated sample. When an artificial defect is made on an organic coating either by drilling or scribing, the electrical resistance of the coating exhibits almost the same as a short circuit. The measurements by AC or DC techniques characterize mainly the defect itself rather than the disbonding process. Therefore, the results thus obtained are usually different from those for a system naturally disbonded under cathodic protection. (2) Protective coatings being used in cathodically pro- tected structures are normally quite thick (300 mm) and also pigmented. The behavior of the cathodic disbonding of thick coatings would be pronouncedly different from that of thin and unpigmented coatings [13], which have been exten- sively studied in laboratories. In this work, the cathodic disbonding behavior of a thick, Progress in Organic Coatings 31 (1997) 289–295 0300-9440/97/$17.00  1997 Elsevier Science S.A. All rights reserved PII S0300-9440(97)00086-6 * Corresponding author.