PSI - Issue 64

Pat Rajeev et al. / Procedia Structural Integrity 64 (2024) 523–530 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

526

4

• Dealloying (selective removal of an element from an alloy by corrosion) • Pitting (localised attack where the rate of corrosion is greater in some areas than others) • Intergranular corrosion (localised attack at grain boundaries of a metal resulting in loss of strength and ductility) • Cracking As noted earlier, below ground region of a cable stay is more susceptible for corrosion comparing with the above ground region. Corrosivity of the soil governs the rate of corrosion below the ground level. The corrosivity of a soil relies on a number of factors such as moisture, dissolved salts, porosity, electrical conductivity and pH (Revie, 2008). Fig. 3(a) shows the reduction in the effective cross-sectional area of a cable stay at the ground level due to the progress of corrosion. Fig. 3(b) illustrates the presence of corrosion at the bottommost region of a cable stay. The increased moisture retention and the presence of salts at the bed log anchors could be potential factors contributing to the enhanced severity of corrosion at the anchor locations of cable stays.

(b)

(a)

Fig. 3. Corrosion of cable stays (a) reduction in cross-section at the ground level (b) presence of corrosion at the bottommost region of the steel rod

Different remedial measures are developed over the years to mitigate the effects of corrosion in metals. Use of protective coatings, alteration of soils and cathodic protection are the most widely used conventional techniques (Presuel-Moreno, 2008). Cable stays utilize protective Zinc coatings as a type of corrosion-resistant metallic coatings. Soil alteration refers to the process of reducing soil acidity by incorporating suitable materials, such as limestone. In cathodic protection, the surface corrosion of metal is prevented by designating it as the cathode within an electrochemical cell. After the initiation of corrosion, the rate of corrosion relies on number of factors and researchers have investigated the relationships between the service life of buried metals and corrosion severity. Following section briefly explains these corrosion models. 2.1. Corrosion pit depth models Since the early 1900s, researchers have been investigating the progression of corrosion in steels buried in soil, taking into account chemical, electrochemical, and physical aspects. However, developing analytical tools for predicting corrosion in soil has been a challenging task due to the inherent uncertainty and statistical variability in the corrosion amounts. The earliest and simplest corrosion models represented average corrosion rate as a linear function of time (Melchers, 2018). The uniform rate of corrosion is calculated in these models by dividing the corrosion loss or pit depth by the exposure time. Nonetheless, a major limitation of these initial models was the inability to incorporate effects of pertaining environmental and climatic conditions for the progress of corrosion. Hence, relying on the average corrosion rate as a linear function of time can be deceptive, and the accuracy of predictions becomes questionable. Experimental studies indicate that only short-term corrosion can be reasonably assumed to exhibit a linear variation with time, whereas long-term corrosion tends to follow a curvilinear trend (Southwell et al., 1976). Power law functions were later developed to address the shortcomings of linear uniform corrosion estimations. Coefficients of the power law functions were found by fitting to experimental data. The most extensively used and comprehensive database for metal corrosion in soils has been established by the U.S. National Bureau of Standards (NBS), which conducted corrosion experiments since 1910. In this extensive series of experiments, 128 sites across