Reliability & Maintenance

Plant Maintenance: Efforts Pin Down Corrosion

Diverse efforts seek insights for understanding and combating the pervasive problem.

By Seán Ottewell, Editor at Large

Corrosion eats up 3–4% of global gross domestic product each year, according to a 2016 study by NACE International. That translates to an annual cost of about $2.5 trillion, says the Houston-based organization that focuses on corrosion prevention and control. However, corrosion remains an elusive as well as an expensive problem to pin down.

“Chemically we understand what corrosion is — but unfortunately it doesn’t occur uniformly at all. If it did, it would be easy enough to predict the rate of corrosion. What we need is well-controlled corrosion films to protect metals. It’s understanding why corrosion accelerates suddenly and takes place in a particular location that is crucial here,” explains Philip Withers, a professor of materials at the Royce Institute, University of Manchester, U.K.

“What is less well understood are specific features of corrosion. For example, corrosion at the atomistic level is a non-deterministic, stochastic process. So, the corrosion rate on a piece of equipment that is being used in exactly the same way from day to day will vary from day to day. Engineers don’t understand this way of thinking,” adds Stuart Lyon, AkzoNobel professor of corrosion control at the university.

In an effort to tackle this problem, last July BP, London, teamed up with the University of Manchester in a £5-million (≈$7-million) collaborative project. The funding is coming jointly from BP and the U.K.’s Engineering and Physical Sciences Research Council (EPSRC).

The project will bring together top researchers from the company, Imperial College London and the University of Cambridge — many of whom already work together on corrosion research through the six-year-old BP International Centre for Advanced Materials (BP-ICAM) at Manchester University — along with additional experts from the University of Leeds and University of Edinburgh.

This project stems from an earlier BP-ICAM effort which studied the fundamental processes that initiate corrosion.

“Manchester, Cambridge and Imperial have been working together for more than five years with BP looking at a range of advanced materials problems. But to solve these, we needed to bring in new skills. So, we recruited expertise from Leeds on tribocorrosion [material degradation due to the combined effect of wear and corrosion] and expertise from Edinburgh on how high pressure can affect the behavior of interfaces,” says Withers, who serves as principal investigator on the new project.

By combining this expertise with different skills in modelling and imaging as well as performing experiments under real-life conditions, the team hopes to answer three fundamental questions: What happens at the start of corrosion? How does it then propagate? And what occurs in tribocorrosion? Some of the basic understanding gained should enable improving current materials; the team will focus particularly on developing better coatings and inhibitors as well as wear-side lubricants and additives that can be used with them to extend equipment life (Figure 1).

By applying synchrotron radiation, among other techniques, the researchers hope to understand the very early stages of oxidation. Such radiation penetrates the surface of corrosion films and helps to show the importance of material stresses and densities on how protective layers break up in localized areas.

“Imaging is very important and we are now able to cheat the fundamental limits of the accuracy to get amazing resolutions. Fifteen-to-twenty years ago, for example, 20–30 microns was high resolution with X-ray imaging. Today, we are [at] the 50-nm scale. The great thing about using X-rays is that you can look through materials, so you get to see pits and other features and understand them at the sub-micron scale,” Withers explains.

Other imaging techniques used include atomic force microscopy, scanning electron microscopy and transmission electron microscopy (Figure 2).

Although the collaboration was announced in July, research really began in November. The team already has made progress: “What we have done is the basic modelling of the early stages of corrosion, looking at how structures change because, for example, the film gets thicker and this, in turn, affects diffusion and diffusion pathways. We have seen how corrosion films build up and this is very similar to the films that prevent wear. Further, when corrosion and wear occur together, the degradation accelerates and, so, we are looking at the interaction of the two. The interesting thing here is that one plus one can equal 1,000. This is because we can study the structure of corrosion and structure of wear individually — but acting together, their effect can be multiplied 1,000-fold.”

The team also is starting to build some of the rigs needed to study in-situ corrosion, including special experimental cells that replicate corrosion conditions in the field.

Saline environments are getting a special focus because BP has many pipes and plant equipment that are either in or near salty water. “That’ll lead us on to other studies, for example in the case of subsea pipelines, we can study the effects of different oxygen levels and different chlorine environments. On the tribocorrosion side, we are making up model lubricants and studying them, too,” adds Withers.

In fact, the project has been set up in a way to ensure that corrosion problems BP encounters in the field are fed directly to researchers via a team of company mentors who already work together in BP-ICAM. They have experience in many different specialities including upstream engineering, refining and lubricants.

“They help us to develop strategies about issues that are important to BP and also manage the flow of useful information between us and BP’s businesses. These mentors are really important because they push us to look not just at simple corrosion situations but also more-complex industrial-type situations. We also have a project manager from BP with lots of goals for us to achieve, for example plotting the gradual move from in-situ investigations of simple species to more demanding situations,” Withers notes.

This reflects the nature of the EPSRC funding, too, which came via the first round of a new initiative called prosperity partnerships. These are aimed specifically at bringing together industrial and academic expertise to solve industry-critical problems.

Coatings Concerns

AkzoNobel, Amsterdam, has been collaborating with the University of Manchester for more than 30 years. Not part of the BP-ICAM initiative, the company focuses more on the interaction between corrosion and coatings, and broader materials sciences issues.

In 2009, the company decided that corrosion protection was such an important area both for itself and its customers that it set up a specialist community of practice (COP). “My job is to manage the knowledge in this area — to highlight what we know and what we don’t know,” says Simon Gibbon, AkzoNobel COP leader in the field of corrosion protection at the university.

One of the outcomes of this decision was the 2012 launch of a more-focused collaboration to look at how corrosion interacts with existing coatings and then to use this knowledge both to improve their function and to develop new, improved coatings.

In the intervening years, the team has nailed a couple of things — particularly some hypotheses that were based on gut feeling, according to Lyon. “We’ve proved some and disproved others.”

For example, he says it’s quite easy to imagine that coating adhesion is really important and controls corrosion processes. Yet, it’s easy to find additives that increase adhesion but,in practice, actually reduce performance as well as some coating systems that show poor adhesion but provide superb performance, he cautions.

“So, this hypothesis is incorrect — sure the paint has to stick sufficiently to limit mechanical damage but beyond this there is no further benefit (and may be detriment) in corrosion protection by further increasing adhesion.”

Another idea is that damage gradually builds up in coatings during service until flaws join to create an easy pathway from the environment to the substrate. Because the polymers used in most paint systems are crosslinked networks, it’s been assumed that poorly crosslinked areas are most susceptible to water uptake and damage.

“However, for some coatings we have shown the opposite — more highly crosslinked parts of the polymer absorb more water. This counter-intuitive result was only obtained because, using our advanced analytical tools, we can probe the molecular composition of polymers at the nanoscale. So, this hypothesis may be correct but for the wrong reason. It’s important because you cannot accurately and reliably design a paint system based on incorrect hypotheses.”

One of the key chemical industry challenges Gibbon is tackling is corrosion under insulation (CUI). He notes: “This is particularly a problem caused by retrofits and if new builds aren’t done to standard. But how do you detect it?”

The corrosion might occur at a location that’s inaccessible or covered with a hard-to-safely-remove insulation layer. He knows of chemical companies that are fabricating entire buildings around very sensitive plant items to prevent exposure to water, drips from other pipes, joints, etc., that could lead to such corrosion. Another issue is humidity, particularly in cryogenic or other systems where condensation occurs, for example during plant downtime.

“So, it’s a complex challenge. We are working with experts in sensing technologies at Manchester to identify clever ways to incorporate intelligence into coatings so that local damage can be narrowed down to a limited area on the plant.”

“One way to prevent CUI, especially for new builds, is to ensure that the coatings are applied properly and the equipment they are used on is installed properly. It’s also important that the underlying metal meets the design spec. We’ve had incidents of coating failures which occurred because the metal manufacturer changed its process slightly and this, in turn, created surface issues,” adds Gibbon (Figure 3).

Lyon believes this message is getting through to the chemical industry, at least its more-enlightened companies: “The value added is definitely being appreciated and we are working together to create extra value for both Azko and the asset owners.”

However, most operating companies have dispensed within-house corrosion engineers and metallurgists, he points out, leading to a loss of knowledge that can prompt problems. For instance, he cites a company whose bronze shell-and-tube condenser failed after 22 years. A contractor suggested lining the tubes to plug the leaks. “However, what seemed like a good idea massively speeded up corrosion due to the increased flow rate such that the failure occurred again after just 18 months.”

He mentions a corn syrup producer in the U.S. as another cautionary example. To save on the costs of potable water being used in manufacturing, it switched to its own well water. “It sounds like a very logical decision by the plant manager but the potable water contained 100ppm of chloride ions while the well water contained 500ppm. The plant suffered a $3–4-million failure because of the resulting localized (pitting)corrosion problems,” he explains.

False economies also afflict painting. It’s common for a plant to opt for the cheapest quote when paying perhaps 20% more for professional applicators could double a coating’s life, stresses Lyon. This is one of the reasons that AkzoNobel has pioneered an industrial painting qualification with the Institute of Corrosion, Northampton, U.K.

The company also has launched a service for customers called Interplan in which AkzoNobel will either manage the corrosion protection of the assets involved or just provide advice as to when recoating is required (Figure 4).

“There are few corrosion issues on chemical plants that are not user-related,” cautions Gibbon.

Internal Resources

Unlike most chemical makers, BASF, Ludwigshafen, Germany, boasts an in-house materials engineering unit to support production processes. It covers all aspects of materials engineering connected with chemistry, including corrosion issues, and uses a range of non-destructive testing technologies.

However, the company goes outside when necessary. “In the case of exceptional and very specific problems, we cooperate with external partners such as universities and research facilities,” notes a spokeswoman.

One such partner is the Materials Technology Institute, St. Louis, Mo., where BASF is working with other chemical companies including Air Products, Sabic, DuPont, Shell, Air Liquide and Chevron on a range of research initiatives. One focuses on using software for thermodynamic modeling of corrosion and training engineers so they can predict the performance of alloys in corrosive environments and to improve the design of corrosion experiments. Another focuses on developing non-invasive monitoring techniques that can identify corrosion issues that lead to the deterioration of refractory linings.

BASF also works with the German Society for Corrosion Protection, Frankfurt, which acts as an interdisciplinary federation to bring together corrosion experts from industry and academia — with the aim of developing better tools to understand and deal with corrosion and its consequences.

“The benefit of these collaborations is that members affected with the same damage mechanisms work jointly together on mitigation approaches,” says the spokeswoman.


Ottewell2Seán Ottewell is Chemical Processing's Editor at Large. You can email him at sottewell@putman.net.