Dana Medlin, a professor at the South Dakota School of Mining and Technology, discussed corrosion over much longer time periods. He has been working with the US Navy and the National Park Service to study corrosion of the U.S.S. Arizona, a battleship that was sunk during the Japanese attack on Pearl Harbor on December 7, 1941. The sunken vessel has been maintained as a memorial since 1962. Since the 1980s, underwater archaeologists have measured the buildup of marine material (called concretion) on the surface of the wreck.
Using these measurements, along with measurements of corrosion potential, acidity, and oxygen content, the researchers developed a simple model to predict the corrosion rate of the steel plates that make up the ship. This model has been validated with data from other shipwrecks, including the Civil War ironclad U.S.S. Monitor, and is now being correlated with data from the H.M.S. Titanic.
The corrosion rate of the World War II battleship is of more than just academic interest: there are still an estimated 500,000 gallons of oil aboard the sunken ship. A structural failure due to corrosion would release this oil into the harbor. Finite-element analysis performed at the National Institute of Standards and Technology indicates that "general collapse" of the steel plates will occur sometime around the year 2150. However, this model did not take the rivets that hold the steel plates together into account. If these rivets fail, the structure may "unzip" while the plates are still intact. A more detailed model, which takes the rivets into account, is currently being developed.
Computer modeling of corrosion processes was the theme of many presentations. One of the most ambitious projects was described by Erik Sapper of Boeing Research and Technology. Along with Australia's national science agency CSIRO, Boeing is trying to develop what it calls an "atoms to atmospheres to airplanes" multiscale corrosion model. This model will include everything from molecular simulations of corrosion inhibitor activity and microstructural simulations of metal alloy surfaces, all the way to global models that predict atmospheric conditions at different places around the world. This would make it possible to predict how much corrosion will occur on an airplane built from given materials and coatings as it flies along a given flight path.
While Sapper acknowledges that "this is not a small project, and it's not a quick project," he noted that "a lot of these models already exist; we're just threading the needle and connecting all these components." Currently, the main task is "handshake engineering": developing an interface that will allow all of the models to interact with one another. One day, this may lead to a practical design tool that engineers can use to explore corrosion in the virtual world before testing their designs in the real world. It seemed like a very appropriate topic for what was, after all, a virtual conference.