Back in 1943, NACE International was established in Houston, Texas, USA by 11 engineers focused on cathodic protection (CP) to address metal pipeline degradation. Now 36,000+ members strong, NACE has evolved into a worldwide organization that is involved in every industry and area of corrosion prevention and control. This year, NACE has been celebrating its 75th anniversary, a milestone made possible by the knowledge, expertise, and continued support of its members from around the globe. After honoring NACE’s history over the last year with classic MP articles from the past, guest editorials from NACE Past Presidents and Fellows, content about NACE’s evolution across all activities, and more, our latest issue is dedicated to the future of corrosion control as seen by experts in the field. In this roundtable special, panelists share their predictions on where the corrosion industry is going in the next 25 years and beyond.
For Part 3 of this series, the featured panelists are John R. Scully, FNACE, the Charles Henderson Chaired Professor of Materials Science and Engineering and the co-director of the Center for Electrochemical Science and Engineering in the School of Engineering and Applied Science at the University of Virginia, USA; Neil G. Thompson, Ph.D., FNACE, senior vice president and head of Pipeline Services at DNV GL USA, Inc., Dublin, Ohio, USA; and Jack Tinnea, the technical editor of Materials Performance.
John R. Scully
When thinking about the future of corrosion control technologies deployed by industry over the next 25 years and beyond, we should think about both new innovative developments on the horizon and also the state of maturity of existing corrosion control technologies that continue to evolve and become more sophisticated.
The stages of maturity mark the progress of existing corrosion control strategies. In my opinion, every strategy has a conceptual or notional phase, a proof-of-concept stage, and a qualitative implementation stage where the evolution of knowledge leads to the identification of certain metrics and target parameters that define the threshold for adequate control of a certain form of corrosion. At this point there is often a data-rich, computational implementation stage that becomes very quantified. The first stage is a point in development where the idea of the technology becomes widely accepted but it is still somewhat conceptual. For instance, in the evolution of CP as a corrosion control method, the concept became widely accepted that an actively corroding metal could be polarized to potentials equal to or more negative than its reversible electrode potential for metal dissolution and thereby achieve thermodynamic immunity. Or alternatively, cathodic polarization could be achieved to a potential between the freely corroding potential and the reversible electrode (Nernst) potential for metal dissolution and that condition could substantially lower the corrosion rate by slowing corrosion kinetics.
In the first stage there was the “concept.” The next stage was adaption of several highly useful quantitative criteria or specific “metrics” that concisely define successful control. Examples include the 100 mV instant off potential, and the protection potential criterion of –850 mV vs. copper/copper sulfate (Cu/CuSO4) electrode as well as other semi-quantitative “rules of thumb.” These metrics have become standardized and widely accepted. However, this stage was just one level of sophistication on the rung of progress. That is because such a criterion could not be achieved spatially across an entire cathodically protected structure at all locations—verification of the CP threshold level was only achieved at first at a few spots on the structure. In stage three, enough was known about the theory of CP and the metrics for success that this knowledge could be combined with growing computational capabilities to map the potential distribution across the entire structure. Computer generated finite element potential and current distributions are now fairly routine and can examine the spatial dependency of CP. Distributed remote sensors (i.e., reference electrodes) can monitor the potential at many locations. The potential distribution on a pipeline can now be explored in detail to assess CP levels.
A similar “set of stages” can be used to describe other corrosion control strategies such as coatings, safe limits for high-strength materials given the threat of hydrogen embrittlement in harsh environments, and choice of a corrosion-resistant alloy. Concepts lead to simple metrics. Simple metrics are now being replaced with more sophisticated models and tools. Stages of maturity define the current progress in most if not all corrosion control strategies.
What is next? There are many promising new corrosion control strategies yet to emerge. One benefits from the coming age of cyber physical systems. Here corrosion control will likely follow along the lines of the “smart cities internet of things” concept where many varieties of distributed sensors will in real time interrogate the corrosion “state of health” of a structure or system and algorithms or digital tools will make decisions either automatically or with owner inputs. In the future, many of the sensors needed will be powered by the nearby environment, harvesting energy from their surroundings without connection to the grid. The cyber physical world is here to stay and will likely expand into corrosion. These strategies will create enormous amounts of data. It is said that 90% of the world’s data has been generated in the last two years. This data collected is rich in information but too large to manage. Materials informatics, data sciences, and machine learning are but a few strategies that rely on such increasingly large amounts of data such as those generated from all those sensors. Data sciences approaches will be necessary to understand how to handle and interpret all that information but could establish relationships and trends impossible to see otherwise that could aid corrosion control. Such relationships might not be detected using conventional approaches. Data sciences approaches may also reveal relationships between environmental or material factors and corrosion that might not be discovered by conventional means.
All these possibilities and more point toward a bright and exciting future in the corrosion control industry.
Neil G. Thompson
In the 2017 Frank Newman Speller Award lecture, Narasi Sridhar described knowledge-based predictive analytics. I believe that knowledge-based analytics will make the largest difference in how we approach corrosion management. I am not referring to simple data trending, or data-centric correlative analysis (as Dr. Sridhar describes it), but combining correlative analysis with predictive modeling. These models can be empirical/semi-empirical models based on available data, expert-knowledge base models, or they can be mechanistic models based on scientific principles. Today, we are most successful using models to predict within a given variable space represented by the data. Predicting outside this variable space is very difficult.
As knowledge-based analytics continues to grow along with the analytical capabilities of data processing, the ability to combine data, empirical models, expert models, mechanistic models, and machine learning principles will allow for improved corrosion prediction in both variable space and time. Understanding the uncertainty associated with corrosion predictions will allow the engineer/operator to better understand how to utilize the information and when additional data may be required to decrease the uncertainty, if necessary. Moving toward the ability to provide near real time predictions will be critical to meeting expectations of operators and, in many cases, the public. For example, there is now almost zero tolerance for failures of any kind on energy pipelines. Environmentalists and the general public use these failures as justification to oppose new pipelines or shut down existing pipelines. I think our business as usual approach to corrosion control as well as integrity assessments will require significant changes and updating. The predictive capability of knowledge-based analytics, modeling, and machine learning tied to integrity assessment and near real time risk management will form the basis for these changes and the ability to predict pipeline critical conditions before a failure happens.
Jack Tinnea
Almost 50 years ago, I was working in my first job after graduation, doing research on cement chemistry for the civil engineering department of my alma mater. At that time, the University of Illinois had a Materials Engineering Group that included folks from ceramic, chemical, civil, and metallurgical engineering with occasional visits by someone from the physics or chemistry departments. Today, Illinois has a Materials Science and Engineering Department that rose from the merging of previous metallurgy and ceramic engineering departments.
Materials science can offer the practicing corrosion engineer many choices in controlling corrosion. I see the materials science field much like electrochemistry was in the time of Sir Humphry Davy when the father of CP made his greatest discovery: Michael Faraday. Yes, in the 1820s when Davy and Faraday were developing CP for the British Navy, our understanding of electrochemistry had advanced from Coulomb, Galvani, Priestly, and Volta, but in terms of where we are today our use of electricity was just getting started.
Nanotechnology
Early in my career, ICCP anodes were primarily cast iron or graphite. These were soon joined by platinum-clad anodes that evolved from finding a second marketplace for niobium-copper cored wires that were being manufactured for use in magnetic resonance image (MRI) scanners for hospital use. These MRI wires were soon found to be excellent platforms for ICCP if the wires were coated with a thin layer of platinum. In what seemed like just a few more years, mixed metal oxide (MMO) arrived. MMO anodes typically involved using ruthenium oxide (RuO2) or iridium oxide (IrO2) either singly or in various combinations to coat a titanium substrate. Titanium oxide (TiO2) has a similar rutile-type structure as RuO2 and IrO2 and similar ionic radii of Ti+4 (0.075 nm), Ru+4 (0.076 nm), and Ir+4 (0.077 nm) that allow the development of a tertiary solid solution that arguably could be considered an early form of nanotechnology.1
Today, science and engineering periodicals are filled with discussions of nanotechnology.2-3 Our understanding of nano-scale includes not just the manufacture of new products but our understanding of corrosion itself on the nanoscale.4 On the proactive side, as we better learn how to assemble materials atom-by-atom, we will produce amazingly corrosion-resistant
materials with a wide range of structural and thermal properties. Likewise, on the reactive side, it allows us to better understand why corrosion occurred in one location and not the other when on the macro-scale the two locations appear so very similar.
Corrosion engineers are familiar with anodes and cathodes, but recent advances in development in battery anodes and cathodes may lead to batteries that will easily power CP systems in remote locations with unimaginable service lives. Perhaps our grandchildren may develop a means to redirect or capture destabilizing ions present in the environment and in the process assist with in situ repair of protective oxide layers, like those present on stainless steels.
Cross-pollenating
Cross-pollenating is another area that likely will produce advances. For example, corrosion can be a problem with guitars, particularly electric guitars. Although this is not a topic frequently addressed by NACE, perspiration from the guitar player can cause corrosion of screws securing the pick guards and even can cause corrosion of the pickups (pups). Microphones are transducers that convert mechanical sound waves travelling through the air into variations in an electrical signal that get sent to the amplifier and then the speakers. The pups on an electric guitar are not microphones but are magnet-based transducers known as a variable reluctance sensor. Pups detect changes in the proximity of ferrous material, namely the steel strings of the guitar, so stringing your electric guitar with nylon strings will not work well—pups are not microphones.
Pups have a permanent magnetic core that is wrapped with thousands of turns of fine enameled copper wire. The fine enamel can age and start to flake. Add to that the close proximity of alnico or ferrite magnets, and it should be obvious to a corrosion engineer that perspiration could cause corrosion issues or result in partial short-circuiting of the pickup with a loss of performance. To protect the pups, some manufacturers employ just a wrap or two of electrical tape; others add an additional coat of enamel, lacquer, or epoxy to the outer face of the enameled copper wires. Still others use a mix of about 20% beeswax and 80% paraffin, melt the wax and then submerge the pup into the melted wax for 15 to 30 minutes or until the pup stops bubbling from the wax displacing air voids between with layers of the copper wire and the magnetic core. Over time, potting the pups with wax provides longer protection than the others, and problems can be easily corrected by melting the wax, whereas pups that are potted in epoxy and develop problems are typically just thrown away because removing the epoxy is likely to damage the windings.
In another area of potting and/or coating electrical components, epoxies, silicones, polyurethanes, polysulfides, and cyanoacrylates are used to protect printed circuit boards as are plasma and vapor applied coatings. Vapor phase corrosion inhibitors are also used.
Remote monitoring of major infrastructure elements often requires sensors to be embedded in concrete for as much as 80 years or more. To date, the published literature includes more than a few discussions of such embedded sensors failing in much less than that time interval.5-7 Achieving highly reliable embeddable sensors to fit the demands placed by 100-year infrastructure service lives might be facilitated through cross-pollenating. Perhaps by treating first with a vapor phase inhibitor, followed by potting with a material that provides good wetting characteristics to give excellent penetration, and that followed by a hydrophobic and/or alkaline-resistant coating would do the trick.
Conclusion
I believe that we are just at the start of an era of tremendous advances in our understanding of materials science. The next quarter century will bring not only amazing advances in materials themselves but in miniaturization that will greatly extend our ability to monitor and understand. Not all the real “wow” moments in technical advancement are in the past. There are infinite possibilities for similar breakthroughs for generations to come.
References
1 F. Moradi, Dehghanian, “Addition of IrO2 to RuO2 + TiO2 Coated Anodes and its Effect on Electrochemical Performance of Anodes in Acid Media,” Progress in Natural Science Materials International, Elsevier 24 (2014), p. 136.
2 Up Front, “X-Ray Microscopy Platform Probes Material Composition,” MP 57, 5 (2018): p. 11.
3 B. Valdez, M. Schorr, R. Salinas, “Food Industry: Equipment, Materials, and Corrosion,” MP 57, 5 (2018): p. 42.
4 A.M. El-Sherik, H.A. Ahwad, “Sour Corrosion: A Review of Current Gaps and Challenges,” MP 57, 5 (2018): p. 49.
5 C. Andrade, I. Martinez, “Embedded Sensors for the Monitoring of Corrosion Parameters in Concrete Structures,” Non-Destructive Testing in Civil Engineering Conference Proceedings (Nates, France: 2009).
6 K.R. Larsen, “Evaluating Sensors to Monitor Steel Corrosion in Concrete Structures,” MP 55 (2016).
7 S.K. Lee, et al., “Corrosion Monitoring of World’s Largest Tidal Power Plant,” MP 57, 11 (2018).