When it comes to corrosion under insulation (CUI), I tend to call it a "minefield," and for more than one reason. It’s not only a hidden phenomenon, but it's also one that can really come back to bite companies who don’t invest in a good CUI mitigation strategy, because sooner or later, those companies are going to get their names in the papers due to unwanted accidents, production loss and environmental damages. (Gain an introduction to CUI in Understanding Insulation Chemistry Proven to Inhibit CUI.)
But when the proverbial horse has bolted, a lot of asset owners want to close the stable doors by implementing rigorous measures. The problem is that they often only focus on a short-term solution, such as dismantling the entire thermal insulation; in some cases welding where wall thickness loss is critical; recoating equipment or piping and then reinsulating—in most cases with the same insulation system as before.
With CUI, it's best to think about long-term solutions and minimizing future failures. Of course, this takes an effort at all phases of the life cycle and starts with senior management allocating enough budget.
In this article, however, I'll only be focusing on what I think is the first step in the life cycle: engineering.
Engineering a Corrosion Under Insulation System
From an engineering point of view, we need to look at the whole system and make choices about:
- Metallurgy for equipment/piping
- Surface protection, such as coating or metalization
- Insulation material
- Cladding or jacketing
In this article, I will discuss items 3 and 4 in depth, and share recent insights and practices.
What Is Corrosion Under Insulation (CUI)?
In recent years, I’ve seen a variety of people dealing with CUI who have little or no background in the basic physical principles of corrosion. So, I'll start with a short summary:
CUI is a collective noun for various types of corrosion mechanisms, but it’s always caused by the presence of (rain) water containing chlorides and/or sulfates. For carbon steel, CUI only occurs in combination with a failed or damaged coating system, and can manifest as pitting corrosion or uniform wall thickness loss. For austenitic stainless steel, the most common form of corrosion is external chloride stress corrosion cracking.
Water, which eventually can accumulate onto the substrate and could form an electrolyte, can originate from various sources:
- Rainfall or heavy mist
- Drift from cooling towers
- Deluge systems
- Process leakage or spillage
- Condensation within the insulation systems
CUI can be expected on piping and equipment operating between 25°F (-4°C) and 347°F (175°C), but also in systems operating outside this range. (For example, cyclic temperatures or dead legs can increase susceptibility.) As said before, CUI is a hidden failure. It can occur locally or can affect a larger area.
Although corrosion rates for carbon steel are, in general, lower than chloride external stress corrosion cracking rates, for carbon steel—especially near salty, seaside environments—corrosion rates of up to 20 mils (0.5 mm) per year have been reported. (Discover more in What Causes Stress Corrosion Cracking In Pipelines.)
The Role of the Insulation System
It’s a common consensus among experienced people dealing with CUI that dry insulation systems simply don't exist in the long run. They therefore tend to be regarded as a bad influence. In addition, people have believed for decades that cladding/jacketing is 100% weatherproof and, in combination with elevated service temperatures, that water or moisture could never get trapped. This idea resulted in many cases in which piping, other that the basic shop primer, wasn’t even extra-coated.
Cladding/jacketing is primarily designed as weatherproofing and not as a vapor barrier. Depending on service temperature and ambient conditions, condensation within the insulation system may not be avoidable, and therefore needs to be addressed in the engineering phase. In other situations, water enters into the insulation system through failed or broken cladding/jacketing. (For more about this topic, read The Detrimental Effects of Wet Insulation in the CUI Range .) This can be caused by:
- Foot traffic
- Inadequate design
- Incorrect installation
- An insufficient maintenance strategy
Overlooking all possible CUI causes in relation to a consequence of the failure of piping systems or equipment, there's a justification for challenging the need for insulation.
The oil crisis during the 1970s brought new insights on energy savings and resulted in other design criteria for thermal insulation for the (petro) chemical industry. In some cases, this resulted in excessive insulating, which was not always economically feasible. However, recent geopolitical CO2 reduction goals persuaded many asset owners to re-evaluate these old goals and translate them into new company policies.
Below is a flow diagram that provides a few logical steps to understand whether insulation is necessary or could be replaced.
Reasons for Insulating
The first question that always has to be answered is the reason for insulating, which can be one or a combination of the following:
- Heat conservation or energy saving
- Process control
- Freeze protection/winterization
- Personnel protection
- Noise reduction
- Fire protection
These reasons determine the choice of insulation materials and the type of cladding/jacketing used. There are various insulation standards and guidelines around the world, but I would like to recommend the CINI Industrial Insulation Manual.
When going through the above flow diagram, we see that only insulation for personnel protection could perhaps be considered for removal and replacement with something like protective guards. However, because environmental and energy-saving goals have become more important in recent years, the removal of insulation even for this reason should be critically assessed.
Designing a Fit-for-Purpose Insulation System
There isn’t a "one-size-fits-all" insulation system. Therefore, insulation design should be more than just drawing up a specification. Pipeline engineers and equipment designers should make detailed insulation designs. Based on a consequence of failure assessment, piping and/or equipment with a higher ranking should lead to the creation of insulation systems with the lowest susceptibility. In order to do so, the following criteria need to be considered when designing insulation systems:
1. Choice of Insulation Materials
Insulation materials can be roughly subdivided into permeable (open-celled) and impermeable (closed-cell) materials. For systems below ambient conditions, where surface condensation or icing is possible, closed-cell materials like polyurethane (PUR/PIR) foam or cellular glass are often chosen, whereas for hot systems, mineral wool like stone or glass wool or expanded perlite are common products.
The choice also depends on local or historical grounds. For instance, Europe uses a lot of mineral wool for hot insulation, whereas in the United States, calcium silicate, perlite and cellular glass are more common. In specifications, it’s common practice not to refer to product names. Therefore, many insulation specifications refer to general technical requirements.
Important characteristics in relation to CUI include:
- Water absorption (ASTM C610 or ASTM C612)
- Leachable chlorides content (ASTM C871 or ASTM C795)
- Hydrophobic behavior
- Compressive strength (when foot traffic can be expected)
- Dimensional stability
2. Insulation Cladding or Jacketing
The first step is to determine the purpose for cladding or jacketing. There are several design criteria such as:
- The need for a vapor barrier (below ambient service temperatures)
- The need for weather protection and UV resistance
- Mechanical resistance
- Accessibility for maintenance or inspections
Cladding/jacketing can be subdivided into metal and non-metal, each of which has specific characteristics and a scope of applications. Although the above criteria determine the choice, this is also influenced by local available craftsmanship and practices. Another important part of cladding/jacketing is the use of caulking or sealants. The choice of whether all joints (longitudinal, circumferential as well as protrusions) shall be finished (or flashed) depends on how sheeting details are designed.
3. Local Geographic Conditions and Plant Layout
Seaside environments are different from inland environments, and arctic conditions differ from tropical ones. Downwind drift from cooling towers or frequent fire deluge drills are other major factors to consider. In Europe, the result of these considerations is that many sites are classified in the highest corrosion class. It's also important to create enough distance between piping and equipment to allow proper insulation installation and enable maintenance and inspection in the future.
4. Equipment, Piping and Tank Design Details
For pressure piping or equipment, standards and codes like ASME, API, BS and Lloyds are available. But details with regard to insulation design are often limited to things like clips or insulation support rings. In addition, some of these details can create potential ingress points. So, bridging the gap between mechanical design and insulation design is a giant leap forward in mitigating CUI. The following suggestions can significantly improve design when it comes to CUI:
- Collars on protruding tubes
- Vacuum rings that don’t retain water
- Lifting lugs that can easily be removed after installation
- Pipe support on high density (for further design details, I refer to the EFC CUI guideline and the CINI Industrial Manual)
5. Installation Procedures
Safety, health and environmental conditions are different depending on the country or region. So, it's important to verify the installation and application guidelines provided by the manufacturer with applicable legislation and rules. Health and safety requirements regarding things like fibers, dust and solvents can influence the choice.
6. Inspection and Maintenance Practices
Clients' own inspection procedures can require inspection plugs or access points for visual inspection or nondestructive evaluation (NDE). In addition, accessibility for maintenance purposes should be considered; for instance, making it possible to easily change gaskets can determine insulation design for valves. It is recommended that clients' best practices be used and that lessons learned from the use-phase regarding insulation be applied. QA/QC during erection is often left to the insulation contractors' organization. Recently I’ve seen many asset owners investing in independent autonomous QA/QC departments that are also drawing up inspection and test programs (ITP), in which critical "hold" and "witness" points are checked. These steps are vital when commissioning and setting up an inspection and maintenance strategy. (To learn more about inspections and evaluations, see our two part series CUI Detection Techniques for Process Pipelines.)
7. Life Cycle Costs (LCC) and Total Cost of Ownership (TCO)
These are two terms that, to some of us, will sound like senior management gibberish, but in my experience, companies with a good working CUI mitigation strategy have a maintenance manager who’s able to convince the senior management about the link between "overall equipment efficiency" and a CUI inspection and maintenance policy. And since CEOs talk in terms of money, these figures become important. One thing all studies have shown: thermal insulation has a return on investment (ROI) of less than two years. It’s been reported under every type of insulation material and cladding, and even in newer installations, that those installations that hold the least amount of water and dry most quickly result in the least amount of corrosion damage to equipment (NACE SP0198). As a logical consequence, impermeable, closed-cell insulation materials and vapor-tight barriers seem to be the best option. Even then, operational conditions (thermal expansion/contraction), foot traffic and failing caulking at protrusions can still damage these systems and cause water ingress. This brings me to my opinion that the following options should be considered:
- Option 1: Non-Contact System
Wet insulation that comes in contact with the substrate is the cause of all the above problems, so why not create a cavity between the insulation and substrate? This is called non-contact insulation. Although the idea is evident and being used by Statoil and Shell, some details still need to be addressed. For instance, the spacers used to create this cavity must not create a crevice to allow crevice corrosion. Depending on the service temperature, especially for vertical columns within this cavity, a vertical free hot air flow can create extra convection and consequently extra heat loss. However, this can easily be minimized by making compartments and can be compensated with a higher thickness.
- Option 2: An Aerated Insulation System
Moisture will always condense where the water vapor pressure rate is at its lowest and at the coldest spot. In thermal insulation, this is the cladding/jacketing. By creating an air cavity between the insulation material and cladding/jacketing, moisture not has only the possibility to freely condense, but also to find its way to the lowest point where it can escape through a drain hole.
Neither of these options is new, and they're implemented by companies like Shell, Statoil and Dow. Also, it’s documented in the CINI Industrial Manual as well as standards like NORSOK, DIN and AGI-Q. Despite this, it's not widely known within the insulation branch. Although some big asset owners are behind these systems, it’s important that independent testing give more scientific and reliable data that can lead to better standardization.
Life Cycle versus Investment
It's obvious that these options come with a surcharge in comparison to traditional ways of insulation. And without fundamental research and testing, it’s plausible that although water still can ingress, it has a way to get out. In other words, these systems are less susceptible to CUI. As a result, they contribute to expected life cycles. Will these systems increase inspection intervals? This depends on more factors. But with Option 1, endoscopic visual inspections are a possibility and large-scale insulation dismantling is no longer necessary. Broken or damaged cladding/jacketing still needs to be repaired, but in this case you can feel more confident that water that has gone in will sooner or later get out.
New Developments in CUI
The technical insulation market is not famous for its innovation. Nevertheless, some simple, bright ideas have been developed over the past years. For instance:
- Moisture detection systems
- Insulation with built-in wicking
The International Standard organization (ISO) has, under TC-67, formed a working group WG11, which is going to develop ISO/NP 19277 (Methods for Control of Corrosion Under Thermal Insulation and Fireproofing Materials). Part of this is to develop a standard that can be used to test and evaluate insulation systems and the effect of CUI. It's a necessary step in order to get worldwide acceptance and independent procedures, and open new ways for product development on a system level.
CUI Isn't New
CUI is nothing new and a lot of solutions in this article have been known for years. In my opinion, there are two major steps necessary to control CUI in the years to come: education and mindset. In the last decades, a lot of know-how has disappeared from the industry due to cutbacks, especially with asset owners. And, since it’s impossible to get a bachelor's degree in industrial technical insulation, this knowledge gap has been filled in by insulation manufacturers and contractors. But since their expertise is thermal insulation, and CUI is a corrosion issue, some bridges have to be built and knowledge shared. This starts with a changing mindset. CUI mitigation is a systems approach: metallurgic design, surface protection, insulation material and cladding/jacketing. By combining these disciplines within engineering, an important step forward can be made.
And, as member of ISO TC67/WG11, CINI and EFC, I see it as a personal challenge to build these bridges and keep on sharing knowledge.