Graduation Year


Document Type




Degree Granting Department

Civil Engineering

Major Professor

Alberto Sagues, Ph.D.


Cathodic behavior, Modeling, EIS, Galvanic corrosion, Durability


Aluminized steel Type 2 (AST2), often used for culvert pipes, is subject to corrosion which is the most important durability limitation factor. It was desired to determine if the outer aluminized layer will retain passivity and if protective galvanic action will develop. Thus, corrosion of unblemished and blemished AST2 surfaces was investigated in simulated natural waters. Experiments with unblemished specimens showed passive corrosion rates (~0.06 µm/yr) in scale-forming, 0.01 M Cl- solutions but sustained corrosion in other less protective media (with rates 3~10 µm/yr). Corrosion was manifested macroscopically by discoloration and few macro pits, but it likely proceeded also microscopically at the Fe-rich inclusion space scale. For blemished specimens, the aluminized coating galvanically protected to some extent the steel in all solutions. However, in 0.01 M Cl- solutions, protection was delayed until after some steel corrosion had occurred.

In some solutions, complete consumption of the outer aluminized coating around exposed steel was noted. Elsewhere, coating appearance was similar to that of the unblemished condition. Nominal durability projections made for 16-gage AST2 ranged from >100 yr for unblemished AST2 to ~10 yr for the blemished condition. The present findings were used as a first step in proposing refinements of presently used durability guidelines of AST2 culvert pipe. Cyclic cathodic polarization tests to examine O2 and H2 reduction at the Fe-rich inclusions showed significant hysteresis, more pronounced with decreasing scan rate. The effect was tentatively associated to the amount of Fe+2 being deposited during the downward scan, a hypothesis supported by results from a physical model. A static polarization model was formulated for the blemished configuration. Results matched experimental trends and permitted evaluating the effect of solution conductivity s beyond the experimental range.

Exposed steel corrosion rates at the steel were increasingly large for decreasing s. For the lowest s, corrosion rates at the exposed steel center were distinctly larger than at the edge, consistent with experiments. An impedance behavior model was also formulated. Results showed frequency dependent current distribution and predicted relatively small artifacts that were and not evident experimentally, but should be considered when exploring other system conditions.