What is Corrosion?

Corrosion has been defined in several different ways. Here are three examples, which contain some obvious similarities.

• A complex chemical or electrochemical process by which metal is destroyed through reaction with its environment

• The deterioration of a material, usually a metal, that results from a reaction with its environment

• The chemical or electrochemical reaction between a material, usually a metal, and its environment that produces a deterioration of the material and its properties

The similarities between these definitions suggest that corrosion involves:

a) deterioration of a material

b) a chemical or electrochemical process

c) a reaction or interaction with the immediate environment

Effects of Corrosion

Anyone who works in a power plant or industrial facility has experienced the operational effects of corrosion. Some of the effects of corrosion include the following.

• Loss of production, due to downtime for the repair of corrosion in piping, fittings, vessels, rotating equipment, etc.

• Leakage of process fluid to atmosphere or within a process vessel (such as an exchanger), leading to contamination of the environment or process fluids

• In extreme cases, rupture of pressurized piping or vessels, with possible consequences for the safety of the site personnel or the public

• Blockage of flow passages, whether piping or exchanger tubes, due to deposits of corrosion products

• Loss of heat transfer efficiency, due to corrosion products collecting on heat transfer surfaces

• High cost of installing and maintaining equipment (plus chemicals) necessary to monitor and control corrosion

• High cost of manpower to manage a corrosion program

• High cost of corrosion related repairs

• Increased equipment costs, due to over-design for corrosion allowance and corrosion prevention

Electrochemistry of Corrosion

Corrosion occurs in many forms, most of which will be described in Objective 2. However, the underlying cause of corrosion involves chemistry and, more specifically, electrochemistry.

Electrochemistry is a process that involves a flow of electrons (‘electro’) plus chemical reactions (‘chemistry’) between substances. In terms of corrosion, electrochemistry may be described as “chemical reactions occurring at the interface of a metal and an electrolyte, resulting in or caused by an electric current, which transfers electrons between the metal and the electrolyte.”

In simplest terms, an electrochemical reaction is one that involves the transfer of electrons. When electrons are lost from a metal (or are transferred from one area of a metal’s surface to another area of the same surface) corrosion occurs and the metal is weakened. Most metals readily react electrochemically with oxygen, water and other aqueous (water-containing) substances.

Oxidation, Reduction, and Redox

Oxidation, in chemistry, refers to any reaction in which atoms of a substance lose one or more electrons. Corrosion of a metal at a specific location occurs when the metal at that location loses electrons; that is, the metal ‘oxidizes’. During oxidation, electrically charged ions are created. These ions are available for chemical reaction with other ions. For example, in the oxidation of iron, a neutral (ie. no electrical charge) iron atom may oxidize to produce an iron atom that is missing two electrons. The change in the iron can be shown as:

Reduction is a chemical reaction in which the atoms of a substance gain one or more electrons. In corrosion, this usually involves the environment adjacent to the metal surface reacting with the metal and gaining electrons, at the expense of the metal itself, which loses electrons (ie. it corrodes). It may also involve one location on the surface of a metal gaining electrons (a process called ‘plating’) at the expense of another location, which loses electrons. Again using iron as an example, reduction of iron may be shown as an ion that is missing two electrons (Fe2+) receiving two electrons (2e–) and becoming a neutral atom (Fe).

Redox is a term derived from reduction and oxidation. It implies a reaction in which reduction and oxidation both occur. One substance in the reaction will be reduced, while the other will be oxidized. This is normally the case in a corrosion reaction, where electrons are transferred between the oxidized metal and the reduced environment. In other words, during corrosion there will be both an oxidation and a reduction reaction.

Oxygen and Hydrogen Reduction

It is relatively easy to understand the loss of ions (ie. oxidation) from a metal during corrosion. However, the reduction reactions, which occur in the substance that is reacting with the metal, may be more difficult to understand. In most industrial cases, that substance is a fluid and the nature of the fluid determines the nature of the reduction reaction.

If the fluid is acidic (ie. low pH, high hydrogen content), hydrogen ions in the fluid will adsorb onto the metal surface. Electrons in the metal are then free to react with these hydrogen ions. The result is a loss of metal ions (2e–) and the formation of hydrogen gas, as shown in the following reduction reaction.

If the fluid is neutral, there will be insufficient hydrogen ions to allow the above reaction. Instead, the electrons in the metal may react with oxygen in the presence of water. Air, dissolved in the fluid, makes oxygen available and this oxygen reacts with water and metal electrons (4e–) in a reduction reaction, to produce hydroxide.

Cathodic and Anodic Reactions

In electricity, the locations at which electrons are discharged or received are called the electrodes. Electrodes that discharge electrons are called anodes, while electrodes that receive electrons are called cathodes.

In electrochemistry, since oxidation involves the loss of electrons, the electrode at which oxidation occurs is the anode and the reaction at this location is called the anodic reaction. Therefore, the location on a metal surface where corrosion occurs (ie. metal is being oxidized) is the anode. Likewise, since reduction involves gaining electrons, the reduction of hydrogen and oxygen in the involved fluid is the cathodic reaction. The location (ie. the immediate interface between the fluid and the metal) at which this reduction occurs is called the cathode.

During corrosion of a single-element metal, all corrosion reactions involve one or more cathodic reactions, plus one anodic reaction. During corrosion of an alloyed metal, there may be more than one anodic reaction, since each alloying element may enter into a separate anodic reaction, thus corroding separately.

The Corrosion Cell

A corrosion cell is the combination of all components necessary for corrosion. All components must be present for corrosion to occur. For ions to flow, there must be a cathode, an anode, an electrical connection to allow electron flow, and an electrolyte on the surface of the metal. An electrolyte is a solution (eg. water, acid), which connects the anode and the cathode and readily conducts ions. Figure 1 illustrates the four components, plus the oxidation (anode) and reduction (cathode) reactions as described above.

Figure 1(a) shows the general relationship between the four components. Figure 1(b) shows corrosion reactions that may occur in an acidic medium. At the anode, Fe is oxidized, releasing Fe2+ into the electrolyte and releasing a flow of electrons through the electrical connection of the metal. Reduction of H+ and the release of H2 is occurring at the cathode.

Environment and Corrosion

“Environment”, in relation to corrosion, may be broadly defined as the direct and intimate surroundings of a material, which expose the material to substances or conditions that have the potential to interact chemically with the material. The nature of the environment to which a material is exposed determines the types and extent of corrosion that could occur. It also helps determine the best methods that should be applied to prevent or protect against corrosion.

There are two general environments that must be considered in any industrial facility.

• the external, natural environment, over which the facility has very little control, and

• the internal environment, which is largely a function of the fluids that are involved in the industrial process itself.

The internal environment, inside the piping, vessels, and equipment, is unique to each industry. It must be dealt with in industry-specific ways that consider all possible corrosion mechanisms associated with the unique processes, fluids, and materials. The external, natural environment is a common consideration for all industries and is usually subdivided into three components – air, water, and soil.

Air

The quality of the air around a facility is largely determined by the adjacent geography and land use. Common atmospheric influences on external corrosion at a facility include:

• Industrial or urban surrounding - Nearby industry emissions and urban pollution may cause acid rain (from sulphur dioxide and nitric oxide emissions). The facility itself may contribute directly, by virtue of the emissions it produces from the internal processes

• Rural surroundings - While this setting seems ideal, there may be corrosive contaminants in the air due to fertilizers and ammonia.

• Coastal location - Salt water (containing chlorides), is more corrosive than fresh water. The mist from a body of salt water can be very corrosive to the external coverings on vessels, piping, and buildings.

• High humidity and wet climate - Moisture in the air is corrosive, particularly when it condenses onto metal surfaces. Rain captures corrosive products in the form of acid rain.

• High temperature - Corrosion rates generally increase as temperature increases, unless the air is very dry.

• Excessive sunlight and other radiation - Radiation encourages corrosive elements in the environment to become active.

More detail on Atmospheric Corrosion is given later, in Objective 3.

Water

Water is brought into a facility for various purposes. The nature of the water is usually determined by the location of the facility in relation to common sources, including lakes, rivers, oceans, and underground wells. Contaminants in the water must be considered to avoid direct corrosion or to avoid deposits, which may encourage other corrosion mechanisms.

The most common corrosion-enhancing contaminants are dissolved gases (eg. oxygen, carbon dioxide, methane), dissolved solids, and biological micro-organisms. Dissolved solids interfere with corrosion inhibiting chemicals and increase the water’s electrical conductivity, which strengthens corrosion cells. Dissolved gases encourage pitting and acid formation. Micro-organisms contribute to under-deposit corrosion and enhance oxygen generation.

Soil

The nature of the soil within a facility and the interaction of the process with the soil can affect the corrosion of foundations and underground piping or vessels. Corrosion of buried structures is impacted by variations in soil properties and characteristics. Corrosion rates can vary widely, depending on the nature of the soil. While it is very difficult and inexact to quantify soil corrosion, it is generally accepted that soil becomes more corrosive when the following soil conditions increase:

• moisture content - water is the electrolyte required for electrochemical corrosion reactions

• oxygen content - oxygen participates in the cathodic corrosion reaction

• electrical conductivity - greater current flow through the soil means greater transfer of metal ions (corrosion)

• acidity - low pH soils are more corrosive to common materials, such as steel and cast iron.

• dissolved salts - chlorides increase soil conductivity and contribute to corrosion reactions.

• sulfate-reducing anaerobic bacteria - microbiologically influenced corrosion can occur very rapidly and most metals are susceptible.

Introduction

Corrosion of metals can be initiated and promoted by several different mechanisms, each of which is unique. The physical appearance of the corrosion created by each mechanism is distinct. The most common mechanisms are described here. Note that these mechanisms are often referred to as ‘forms of corrosion’ or ‘types of corrosion’.

Uniform vs. Localized Corrosion

The terms, ‘uniform’ and ‘localized’, may be considered more as recognizable characteristics of a particular corrosion, than as corrosion mechanisms. However, their differences and potential impacts are important to understand.

Uniform corrosion (also called general or generalized corrosion) is a corrosive action that occurs over an entire surface or over a relatively large section of the surface. The corrosion rate may be constant over the entire area. The pattern is usually a general, uniform thinning of the metal. The electrochemical process involves many microscopic anode and cathode sites on the surface. These sites continually alternate between anode and cathode behaviour, creating a uniform attack. With uniform corrosion, metal failure may take longer to occur, but if the surface becomes rough other forms of corrosion may be encouraged. Uniform corrosion can often be predicted and its progression is relatively easy to monitor. However, there is potential for the loss of a large amount of metal and the weakening of a large surface area.

Localized corrosion involves aggressive corrosive activity at very localized sites on a metal surface. Metal loss at these sites occurs at a faster rate than in the surrounding area, which may or may not corrode. The localization of attack may be due to inherent properties of the metal or the design, or to local breakdown of the corrosion protection. The electrochemistry of localized corrosion suggests that the anodic site is much smaller than the cathodic site, which results in a very high concentration of ions leaving the anode. Due to its localization, this corrosion is difficult to predict, monitor, and control. The danger of localized corrosion is that only a relatively small amount of metal loss can cause component failure.

Galvanic Corrosion

Galvanic corrosion (also called ‘dissimilar metal corrosion’) occurs when two different metals are immersed in a common electrolyte and there is a connection between the two metals, which allows an electric current to flow. The connection may be the two metals actually contacting each other or some other solid connection between them.

Three conditions must exist in order to produce galvanic corrosion.

1. There must be difference in electrochemical potential between two different metals or between locations within a single metal. Different metals have different electron structures; therefore, they have different electrical potentials and a potential difference exists between them. This potential difference (in volts) is the driving force that produces electron flow between the metals. If considering a single, unique piece of metal, dissimilarities within the metal itself are often sufficient to cause a potential difference.

2. The metals must be immersed in the same electrolyte, which provides a path for soluble metal ions to flow between them.

3. There must be a complete conducting path to allow the flow of electrons. The metals must be mechanically connected in some way to complete an electric circuit and allow current to flow.

Figure 2 shows the galvanic corrosion arrangement.

As Figure 2 demonstrates, corrosion occurs at the anode, which is the metal most susceptible to corrosion (ie. the less noble metal). Electrons are transferred from the anode, through the connector toward the cathode, which is the least susceptible to corrosion (ie. the more noble metal) (see “Galvanic Series” below, for explanation of “noble”). At the same time, the positive ions of the anodic metal flow into the electrolyte.

The rate at which galvanic corrosion occurs depends upon

• The difference in electrical potential between the two materials - higher potential difference means higher corrosion rate

• The type and concentration of the electrolyte

• The distance between the dissimilar materials - less distance means higher corrosion

• The ratio of cathode area to anode area - a higher cathode to anode ratio generally results in more severe corrosion

The Electromotive Series

Electrochemical potential (ie. voltage) is a characteristic of a metal that is determined by the metal’s atomic structure and availability of free electrons. It is different for each metal. If two different metals are connected in a galvanic cell, the one with the higher potential will act as the cathode and will drive current flow toward the other metal, which becomes the anode. Tables are available, which rank metals in terms of their electrode potentials, in volts. This ranking is referred to as the Electromotive Series. In general, the higher the electrical potential of a metal (ie. the higher in the series), the less likely it is to corrode. For example, copper (Cu), with an electromotive potential of +0.52 V, is less likely to corrode than aluminum (Al), which has a potential of -1.66 V. If these metals were together in a galvanic cell, the copper would act as the cathode and the aluminum as the corroding anode. The electron flow in this cell is from the Al to the Cu.

The Galvanic Series

Almost synonymous with the electromotive series and serving the same purpose in corrosion is the Galvanic Series. This is a ranking of metals, determined experimentally, which indicates the relative position of each metal in terms of nobility, where the more noble metals are less likely to corrode. In a galvanic cell, the less noble metal will be susceptible to galvanic corrosion. The more noble metal will be the cathode. Although the galvanic table indicates which metal will be anodic and which will be cathodic, it does not indicate the rate at which the galvanic action will occur. However, in general, the further apart two metals are in the galvanic series, the greater will be the potential current or electron flow and, therefore, the rate of corrosion.

Since the nature of an electrolyte has an effect on galvanic action, it follows that a galvanic series must be stated for the specific electrolyte in use. The relative position of a metal, and therefore its preference for anodic or cathodic action, could change, depending on the electrolyte.

The following list of metals represents a partial galvanic series (showing only the more common industrial metals) in which the electrolyte is salt water. The metals are listed from least noble to most noble. For example, magnesium is less noble than titanium and, in a corrosive atmosphere containing those two metals, would act as the corroding anode.

For alloyed metals, the factor that largely determines galvanic position is the actual composition of the metal alloy. Stainless steels, for example, exist in many different compositions and each has its own position in the series. The position is usually determined by the percentages of more noble and less noble metals in the alloy.

Concentration Cell Corrosion

Concentration cell corrosion is a localized and often very aggressive corrosion that occurs when the electrolyte in a corrosion cell becomes more concentrated at one location on a metal surface than in the adjacent environment. The more concentrated cell usually becomes anodic, while the surrounding environment is cathodic. Concentration difference can occur at any small irregularity in the surface where the electrolyte is stagnant, such as in a crevice or under a deposit.

Crevice Corrosion

The most common concentration cell corrosion is called crevice corrosion. Any crevice in the metal interface, such as gasket joints, fastener heads, loose coatings, pipe threads, and lap joints provides a location for a corrosion cell to concentrate. The micro-environment within the crevice may develop corrosion-enhancing characteristics that differ from the surroundings. These characteristics may include loss of corrosion inhibitor, oxygen depletion, increased acidity, and accumulation of more aggressive corrosion products (such as chlorides).

Since oxygen is an active participant in the cathodic reaction in many environments, the loss of oxygen in a crevice means that the normal cathodic oxygen-reduction reaction of the electrolyte cannot occur. This causes the crevice environment to become anodic. Corrosion then accelerates as the metal ions created within the crevice transport to the nearest cathodic surface. The anodic area inside the crevice is small compared to the cathodic area outside the crevice, which promotes rapid corrosion inside the crevice.

Under-Deposit Corrosion

When a deposit is allowed to accumulate on the surface of metal, the interface between the metal and the electrolyte at that location is interrupted. Like crevice corrosion, the surface can develop a concentration cell below the deposit. In boilers, for example, a concentration cell may be created when “wick boiling” occurs beneath a porous deposit on a heating surface, such as a boiler tube. Water in the boiler tube travels through tiny pores in the deposit and reaches the tube surface where it boils. The created steam migrates back out through channels in the deposit. This process concentrates corrosive products in a concentration cell, within and beneath the deposit.

Pitting

Pitting is a form of localized corrosion that is characterized by small cavities in the metal surface. Each cavity tends to propagate into the depth of the metal, rather than spreading out over any width, so the depth is generally greater than the width. Pits may be open and easily detected by visual inspection or they may be ‘hidden’ beneath a semi-permeable layer of corrosion products. While a surface may develop many pits, each individual pit is a separate, localized corrosion site.

The exact cause of pitting on a clean surface is not well understood. However, the accepted mechanism is very similar to crevice corrosion, in that a tiny irregularity in the metal surface becomes anodic and interacts electrochemically with the surrounding electrolyte. In some cases, the pit becomes covered by a film of corrosion products, which deprives it even more of oxygen, promoting further corrosion. In any case, the pit will continue to grow.

Oxygen Pitting is a form of pitting that occurs when free oxygen becomes attached to the metal surface. The surface in contact with the oxygen becomes cathodic while the adjacent surface becomes anodic. Corrosion products accumulate and eventually enclose the bubble of oxygen. This oxygen then becomes exhausted and the area under the corrosion cap reverts to an anode. Thereafter, corrosion continues as a concentration cell.

Deposition Corrosion is another mechanism that produces pitting. A fluid passing through a pipe may absorb metal ions from the pipe material. When the fluid then flows to a location where the metal is different and more anodic, the absorbed ions may be deposited onto the surface at this new location. This is called ‘plating’ and creates a situation with dissimilar metals. Microscopic, galvanic action occurs, causing pitting of the anodic material. One example of this is copper ions being absorbed by water and then being deposited on aluminum surfaces. Since the copper is more noble than the aluminum, the aluminum corrodes.

Conditions that promote pitting include

• small, local damage to the protective oxide film on the metal surface

• small, local damage to an applied protective coating

• small irregularities, during manufacture, in the metal itself.

• free oxygen at the metal surface

• microscopic areas of galvanic action

Figure 3 demonstrates how a pit may be created at a microscopic break in a protective oxide layer.

In fossil fuel-fired steam generating systems, deposition is usually in the form of metal oxides, which are produced through corrosion of equipment in the steam condensate and boiler feedwater system.

Selective Leaching

Selective leaching is a corrosion mechanism that occurs in some metal alloys. An alternate term for selective leaching is Dealloying. The process involves one element in the alloy being selectively removed by electrochemical interaction with the environment. This has the effect of making the alloy weaker, brittle, and porous. In one common example, called dezincification, zinc is selectively dissolved out of brass, which is an alloy of zinc and copper. Another example, which has proven to be a problem in underground water lines, is the removal of iron from cast iron. This leaves a much weaker metal, which often ruptures under internal fluid pressure.

Hydrogen Induced Corrosion

There are three corrosion mechanisms that are directly related to the presence of hydrogen at the surface of a metal. These are hydrogen embrittlement, hydrogen blistering, and hydrogen stress cracking.

1. Hydrogen Embrittlement

If the concentration of hydrogen, as by-product of corrosion, increases at the surface of steel, some of the hydrogen atoms (H) may diffuse into the steel. The hydrogen collects between the metal atoms and distorts their natural structure. This reduces the ability of the steel to deform elastically, causing a loss of ductility and an increase in the brittleness of the steel, which can lead to fracture of the metal.

2. Hydrogen Blistering

This process also involves hydrogen atoms diffusing into the interior of steel. In this case, atomic hydrogen collects at voids inside the steel. Here, the atoms bond together to form molecular hydrogen (H2). The molecules become too large to migrate out of the steel and become trapped. More hydrogen molecules accumulate to the point where internal pressure increases. This pressure can reach a point where metal is displaced outwards, causing fissures or blisters on the surface.

3. Hydrogen Stress Cracking

Again, atomic hydrogen diffuses into the steel and interacts with the atomic structure of the steel, at or near the grain boundaries of the metal. Pressure accumulates to the point where the metal structure separates, forming a crack. If corrosion is also active on the surface, the corrosion cell will migrate into the crack and widen it, in which case the process is called ‘hydrogen stress corrosion cracking’.

Mechanically Assisted Corrosion

Mechanically-assisted corrosion refers to a group of corrosion mechanisms, each of which is enhanced by some form of mechanical action. The most common mechanical conditions to contribute to corrosion are stress and fatigue.

1. Stress Corrosion Cracking

Stress corrosion cracking (SCC) of a material is due to the combination of:

• tensile stress (imposed by the operating conditions)

• susceptibility of the material to corrosion (ie. composition)

• corrosive substances to which the metal is exposed (ie. environmental conditions)

Stress corrosion cracking is a delayed failure process in which cracks start slowly and propagate slowly. Eventually the stresses in the remaining metal exceed the breaking strength and the metal suddenly fails.

The tensile stress required for stress corrosion cracking is usually much less than the yield stress of the metal. The stress may be produced by an external load, internal pressure, or by residual stresses from the manufacturing processes, such as welding, heat treatment, machining, or grinding.

Three common forms of SCC are chloride stress corrosion cracking, sulphide stress corrosion cracking, and caustic embrittlement.

Chloride Stress Corrosion Cracking

Stainless steels containing less than 30% nickel are susceptible to chloride stress corrosion cracking. The chloride attacks the metal along grain boundaries, causing corrosion in the form of cracks. Three components are necessary for chloride stress cracking to occur. An increase in any of these conditions will increase the probability and severity of corrosion.

1. chloride ions must be present,

2. oxygen must be present, and

3. the metal must be under tensile stress

The rate at which chloride stress corrosion cracking occurs will increase if the temperature increases or if the pH of the environment is less than 7.

Sulphide Stress Corrosion Cracking

Sulphide stress corrosion cracking occurs in steel and other high strength alloys when they are exposed to moist hydrogen sulphide environments. The following three conditions must be present.

1. hydrogen sulphide must be present,

2. water must be present, and

3. the metal must be high strength alloy steel under applied or residual tensile stress.

In fact, the ultimate mechanism created by sulphide stress cracking is hydrogen induced cracking. This is because sulphide ions in the aqueous hydrogen sulphide atmosphere inhibit the recombination of hydrogen atoms on the metal. The smaller hydrogen atoms then diffuse into the metal’s crystal structure, causing additional hydrogen induced cracking.

Caustic Embrittlement

Caustic embrittlement is the common term applied to “caustic stress corrosion cracking”. The mechanism is similar to chloride stress corrosion, except oxygen does not have to be present for caustic embrittlement to occur. Mild steels and stainless steels under tensile stress will crack if they are exposed to a caustic (ie. high pH) environment. Caustic concentrations must be very high to induce caustic stress cracking

2. Corrosion Fatigue

Fatigue occurs when a material is subjected to repeated loading and unloading stresses. Corrosion fatigue can result when a metal is subjected to these cycling stresses while in a corrosive environment. Boiler tubes subject to cyclic or fluctuating loads can be susceptible to this type of corrosion. If the load stresses are higher than a certain threshold, usually below the yield strength, microscopic cracks will begin to form at the surface. The excessive stresses may be due to load changes during cold start, forced cooling, or during shutdown or restart of a boiler where thermal stratification of water occurs along the tube length.

Corrosion is accelerated when the cycling stress causes the corrosion protection layer on the metal surface to be broken. Corrosion can then progress uninhibited to the point where it causes either generally weakening of the surface or localized pits. This weakening reduces the stress loads at which fatigue cracks may begin. When a primary corrosion fatigue crack does begin, the corrosion will propogate into the crack and will widen it. The net result is that fatigue failures will occur sooner in a corrosive environment than in a non-corrosive environment.

Flow Induced Corrosion

Corrosion may be established, and corrosion rates are enhanced, when the electrolyte is in motion. Flow-induced corrosion may be thought of as a mechanical process; however, when corrosive surroundings exist, it is also correct to consider this as a corrosion process. In fact, when the electrolyte flows, a combination of mechanical erosion and electrochemical corrosion occurs. Two common forms of this corrosion are erosion-corrosion and flow accelerated corrosion.

1. Erosion - Corrosion

Erosion-corrosion involves the repetitive creation and destruction of the protective surface film on a metal. When flow is within a designed limit, the surface film remains intact. However, if the flow exceeds a critical velocity surface friction and turbulence cause shear stresses which ultimately remove the film. Corrosion rates then increase. When flow returns to normal, there is opportunity for the protective film to re-establish. Surface corrosion is usually characterized by shallow pits or striations in the direction of flow. Flyash in biomass boilers induces relatively high erosion, due to high concentrations of alkali, sulphur, phosphorous, and chlorine.

2. Flow Accelerated Corrosion

Flow accelerated corrosion is a phenomena that has been responsible for several catastrophic incidents involving the sudden failure of high-pressure, high-temperature feedwater piping. It is characterized by a gradual, generalized thinning of a pipe wall within a specific area. This may lead to sudden failure when the pipe is no longer able to withstand the internal pressure.

Flow accelerated corrosion develops at locations in a pipe where there are direction changes or other disturbances in the flow. At these locations (such as feedwater and economizer elbows, reducers, tees, and steam attemperating lines) the pipe is continuously contacted by the feedwater. Also, at these locations the oxygen concentration in the feedwater is typically very low. The feedwater has been through the deaerator and had as much oxygen removed as possible.

When the dissolved oxygen content in the feedwater is very low, the protective magnetite layer (Fe2O3) on the internal pipe surface is not well established, because protective magnetite formation relies on a reducing reaction with dissolved oxygen in the water. At areas of high flow and turbulence, the water tends to dissolve the iron from the magnetite faster than it can be replenished, leaving an iron-deficient magnetite layer. In an attempt to maintain the magnetite layer, iron is continuously taken from the metal, causing continuous, general thinning in that area.

Microbiological Corrosion

Microbiologically Influenced Corrosion (MIC) can directly cause corrosion and can influence or accelerate many of the other corrosion mechanisms. Microbes (ie. bacteria) are a serious concern in industrial waters because they can grow and multiply at phenomenal rates. Bacteria are extremely hardy and, while they grow best in water, they can thrive in many other industrial solutions. They grow fastest when the pH is between 5.0 and 9.0, and in a wide temperature range, from -18°C to 82°C.

In order for bacteria to grow and influence corrosion, the following four environmental conditions must exist, along with the bacteria.

1. a metal surface to host the bacteria,

2. nutrients to feed the bacteria,

3. water, and

4. oxygen (although some type require minimal oxygen)

Bacteria can affect corrosion in several ways. Bacteria can

• directly affect anodic and cathodic reactions,

• react with protective surface layers,

• create corrosive conditions (by changing pH, dissolved oxygen, or compound concentrations),

• produce deposits, and

• secrete acidic fluids

Biological organisms can increase or decrease the oxygen content of the environment, thus affecting corrosion rates. For example, while steel tends to corrode uniformly over its surface when exposed to an aqueous environment, the corrosion rates are proportional to the dissolved oxygen delivered to the metal surface by the solution.

When a biofilm (ie. a deposit with a protective membrane) of bacteria forms on a metal surface, a unique micro-environment is created within the biofilm itself, which is much different than the surrounding environment. Within this micro-environment, changes may occur (in dissolved oxygen, pH, etc.) which lead to electrochemical reactions and an increase in corrosion rates in the vicinity of the biofilm.

Biological organisms can influence uniform corrosion, but are more likely to produce localized corrosion. A heavy deposit of organisms may temporarily reduce corrosion by forming a barrier, which prevents oxygen from contacting the metal surface. However, if the deposit of organisms becomes broken, oxygen is now free to reach the metal surface and localized corrosion may occur as local bacteria use this oxygen.

Bacteria, once established, can also be very difficult to eliminate. A big challenge in controlling MIC is the huge number of bacterial species, with a wide range of metabolic processes. Some species may oxidize specific metals (such as manganese, stainless steel, aluminum, copper alloys, nickel alloys), while others may produce organic acids, which then directly attack metal.

Types of Bacteria

There are three types of bacteria of interest in corrosion.

1. Aerobic Bacteria: these require oxygen to live and grow

2. Anaerobic Bacteria: these do not require oxygen to grow and some may even die in the presence of oxygen. However, some do use oxygen by breaking down available substances that contain oxygen, such as sulphate, nitrate, and carbonate.

3. Facultative Bacteria: these can grow with or without oxygen

1. Aerobic Bacteria

Iron bacteria are aerobic and can grow well in only trace amounts of oxygen. They form a sheath of ferric hydroxide around themselves as they grow, using soluble iron ions found in the water. The sheath can provide a site for the formation of an oxygen concentration cell. Iron bacteria do not take part directly in the corrosion process, but under the sheath that they create is a perfect site for anaerobic bacteria to act. This type of bacteria is often referred to as ‘slime’.

2. Anaerobic Bacteria

Sulphate-reducing bacteria (SRB) are very corrosive in an environment with very little or no oxygen (anaerobic). These bacteria reduce sulphates in the water, producing sulphides and subsequent production of hydrogen sulphide, which can cause the following four different corrosion problems.

• Pitting, directly under a colony of bacteria

• The generation of hydrogen sulphide (H2S) can make the water more acidic, raising the level of general corrosion

• Sulphide cracking (a form of stress corrosion cracking) and sulphide blistering

• “Sour corrosion” that forms insoluble iron sulphide, which itself forms more sites for further pitting

3. Facultative Bacteria

The term “facultative” is defined as the “ability to adapt to different conditions.” Facultative bacteria can live in aerobic or anaerobic conditions.

Many bacteria produce a dense layer of slime on a surface. Beneath the slime exists a perfect environment for facultative bacteria to thrive. Regardless of whether oxygen is present, bacterial corrosion will occur.

Atmospheric Corrosion

Atmospheric corrosion is the degradation of a metal surface when exposed to air and its pollutants. Atmospheric corrosion is a leading cause of metal failure and aesthetic damage and is an important factor in the service life of equipment and the durability of structural materials. There are three categories of atmospheric corrosion: dry, damp, and wet.

1. Dry Atmospheric Corrosion

In the absence of moisture, dry atmospheric corrosion is usually a very slow process. It is characterized by a stable film, which forms (adsorbs) on the surface of a metal in the presence of oxygen. For steels, this film is composed of iron oxide (FeO), created when the oxygen in the air chemically combines with iron from the metal. The simplified, overall chemical reaction is:

The film, often called a “passivating” film, is usually desirable because it creates a protective barrier, which prevents or slows further corrosion. A desirable film is free of defects and is self-repairing if damaged. Porous, less stable films are also possible, particularly when the atmosphere contains a sulphur compound, such as hydrogen sulphide. The presence of sulphur increases the possibility of defects and destroys the film’s ability to protect the metal.

2. Damp Atmospheric Corrosion

When the relative humidity of the atmosphere is about 70% or higher a very thin, invisible film of moisture forms on metal surfaces. This film acts as an electrolyte for the transfer of electrical current, thus leading to galvanic corrosion of the metal (due to local electrode potential differences, caused by imperfections, scratches, internal stresses, etc.). The corrosion rate increases as temperature or relative humidity increase.

Also, if salts are present in the moisture, the corrosion rate is increased even further since salts increase the current carrying ability of the electrolyte. Because salts are hygroscopic (i.e., they readily absorb water) any salt that is already on the metal surface will absorb moisture from the air and create the corrosive film even if the relative humidity is lower. One example is the way road salt promotes corrosion on an automobile.

3. Wet Atmospheric Corrosion

Wet atmospheric corrosion occurs when visible pockets or layers of water exist on a metal surface or if water becomes trapped in tiny imperfections in the surface. Again, the water acts as an electrolyte and iron is lost from the metal surface. The corrosion products may dissolve in the water and cause it to be even more conductive, thus increasing the corrosion rate. Even if the free water dries up, the corrosion products may remain wet and sit on the surface, causing corrosion to continue.

Industrial atmospheres are usually more corrosive than residential atmospheres, particularly if a sulphur-containing fuel is burned at the industrial site. One product of combustion is sulphur dioxide (SO2). This SO2 absorbs onto metal surfaces, forming sulphur trioxide (SO3) in the presence of oxygen. This SO3 then combines with water to produce sulphuric acid (H2SO4), which can dissolve protective films on the metal surface.

Stray Current Corrosion

Stray current corrosion is caused by an external source of current. Stray currents follow unintended electrical paths. Poor electrical conductors, damaged insulation, and dirty or loose electrical connections allow the electrical current to stray from its intended path. This current may travel through water, soil or any suitable electrolyte to find the path of least resistance, such as an underground pipe or structure.

Direct current (DC) is the most destructive source of stray current. Some sources are DC welders, large rectifiers, and improperly installed cathodic protection. Stray alternating current (AC) is not as destructive as stray DC. The greatest source of stray AC is from buried power lines. Figure 4 illustrates stray current corrosion.

A current path in the earth is provided by a low resistance object, in this case a metal pipeline. Current, straying from an impressed-current cathodic protection system will pass from the anode into the pipeline. The current flows along the pipeline for some distance before returning to the protected structure. The site where the stray current leaves the pipeline is anodic and corrosion occurs.

Galvanic corrosion and stray current corrosion are very similar, since both involve a protected cathodic site and a corroded anodic site. Lab analysis is often required to determine which one has occurred. However, a concentration of pits in an area where they are not normally found is an indication of stray current corrosion. Stray currents can be located by measuring voltage drops and current flows along the buried structure.

Intergranular Corrosion

Metals and metal alloys are made of microscopic, crystal-like structures, called grains, which bond together in random, irregular orientation, with boundaries between them. The boundaries, which form an interface between the grains, generally consist of atoms that have been displaced from their original crystals, plus other impurities. Intergranular corrosion occurs when the grain boundaries, which are less resistant to corrosion than the grains themselves, come under attack. Corrosion, in the form of metal loss, usually occurs along the boundary, thus weakening the metal structure. In some cases, the entire boundary around grains is lost, causing the grains to loosen and dislodge.

Intergranular corrosion is predominant in metal alloys (such as stainless steels or aluminum alloys), which are specific mixtures of different metals and are generally expected to be corrosion resistant. One or more of the alloying components (such as chromium in stainless steel), along with impurities, may migrate toward the boundaries. This is usually due to excessive temperatures (500°C to 800°C) that occur during improper heat treatment or welding.

The heat causes the grain boundaries and the adjacent grain areas (called the heat affected zone) to become sensitive to corrosion. The heat affected zone may also become weaker, due to depletion of one or more alloying elements. If this is caused by welding, it is termed ‘weld decay’. The most common prevention for intergranular corrosion is properly controlled heat treatment of the metal (after manufacture or welding), which helps to stabilize the grains and their boundaries.

Boiler Corrosion

Boilers are susceptible to corrosion in many areas and by many different mechanisms. This is due to water being the process fluid and to the operating conditions, which involve high temperature, pressure, and materials under stress. An industrial boiler has several different components, each with slightly different operating conditions, and with different corrosion concerns.

Magnetite Layer

Under normal conditions, a very thin, hard, protective layer of iron oxide, Fe3O4, (called magnetite) forms on the metal surfaces on the water side of a boiler. It forms when iron oxidizes and reacts with the water, as follows.

The magnetite (Fe3O4) passivates the metal surface, thus inhibiting further oxidation. In fact, there are usually two magnetite layers. An outer, porous layer is easily penetrated by water and aggressive ions, while an inner, less porous layer grows by diffusion of chemical ions through the outer layer. The magnetite layer grows to approximately 0.01 to 0.025 mm thick, at which point any further oxidation ceases. If periodic weakening or damaging of the layer occurs, it can be repaired by proper internal boiler water treatment. The appropriate pH level necessary to maintain the magnetite layer is in the range of 8.5 to 12.7. Heat transfer is not impaired by the layer, since it has high thermal conductivity.

There are three mechanisms that can deplete the magnetite layer: oxidation, caustic corrosion and hydrogen damage.

1. Oxidation

The largest source of boiler system corrosion is dissolved gases, which include oxygen, carbon dioxide and ammonia. Oxygen is the most aggressive. The degree of oxygen attack depends on the concentration of dissolved oxygen, the pH and the temperature of the water. Dissolved oxygen attacks the stable magnetite film according to the following chemical reaction.

The Fe2O3 (hematite) often appears as a hydrate (Fe2O3·H2O) that can act as both a base and a weak acid to promote pitting.

2. Caustic Corrosion (Caustic Gouging)

Caustic corrosion refers to the direct reaction of sodium with the metal in a boiler. It is most often seen in furnace tubes in regions of high heat fluctuation. Sodium hydroxide (NaOH) is added to boiler water in non-corrosive concentrations; however, conditions may exist in a boiler to induce caustic corrosion. These include steam blanketing and localized boiling.

• Steam blanketing occurs when a steam layer forms between the boiler water and the tube walls. When this happens, insufficient water contacts the tube surface to allow optimum heat transfer, which lowers boiler efficiency. The water that does reach the tube surface is rapidly boiled away, leaving behind a concentrated and corrosive caustic solution.

• Localized boiling occurs in boilers that use phosphate-treated water. In areas of high heat transfer, such as furnace tubes, porous deposits of phosphate may develop on the tube surface. Water then flows into the deposit and boils beneath it, leaving a concentrated caustic solution.

Caustic soda has high solubility so it remains in solution, but continues to concentrate until the pH becomes extremely high and localized attack begins. The magnetite layer is destroyed and the concentrated sodium hydroxide is then able to react with exposed boiler metal to produce, among other things, atomic hydrogen.

This caustic corrosion continues unless the porous deposits are removed or the caustic concentration is reduced to normal.

Caustic corrosion usually appears as irregular patterns and gouges. White salts may also appear in the metal sample. If caustic corrosion continues for an extended period, black magnetic iron oxide may be found in low flow areas, such as the mud drum. This is essentially due to the stripping away of the magnetite film.

3. Hydrogen Damage

If the boiler water becomes contaminated with acid (a low pH), the acid may dissolve the magnetite layer and attack the boiler metal. The corrosion reaction will produce hydrogen, which may penetrate the steel, and react with the carbon in the steel to form methane (CH4). The methane is too large to escape so gets trapped and exerts pressure between the metal grains, thus weakening the metal. This is referred to as hydrogen damage and the resulting failures are called brittle failures because the metal actually becomes brittle.

Economizers

During normal operating conditions, a well designed economizer delivers feedwater to the boiler drum at about 20°C below saturation temperature. During cold startup or hot restart of the boiler, the conditions inside the economizer are far from normal. The economizer inlet header, outlet header, and discharge line to the steam drum can be damaged if the economizer produces steam before feedwater flow is initiated. When feedwater enters, these sections of the economizer may experience thermal quenching, causing severe shock and possible stress cracks or fatigue at connections and support brackets.

• Pitting in the tubes or headers may occur due to free oxygen attack. The oxygen originates in poorly deaerated feedwater or from inadequate drainage and storage during boiler shutdown.

• Flow accelerated corrosion, due to a combination of excessive deaeration and high interface velocity at economizer elbows, may cause general thinning of the tubes as iron is dissolved into the feedwater. Conditions known to accelerate economizer tube thinning include:

• low pH,

• excessive oxygen scavenger chemicals,

• chemicals (such as chelants) that increase iron solubility, and

• thermal decomposition of organic material.

• Heat transfer surfaces are subject to corrosion product buildup and deposition of metal oxides.

• Corrosion can also occur on the gas side of the economizer when contaminants in the flue gas produce a low-pH environment. Condensing economizers are designed to resist the corrosive action of the liquids that condense out of the flue gas.

Superheaters

Superheaters remove the moisture content of the steam produced by the boiler. They are designed to operate within a wide range of corrosive environments, mainly due to the elevated temperatures at which they operate.

Superheaters are susceptible to corrosion in several ways.

• Superheater tube metal may experience oxidation when temperature becomes excessively high. Above 500°C the steel reacts directly with the steam, forming iron oxide and releasing hydrogen. This may lead to fissures and hydrogen blistering. Firing rates must be limited during startup so combustion gas at the superheater does not exceed 500°C until the boiler reaches operating pressure, all superheater tubes are cleared of condensate that may have accumulated during the shutdown, and steam flow through the superheater is confirmed. Overheating may also occur during low operating loads, when distribution of steam across the superheater inlet may not be even. A superheater tube will fail when its steam flow is blocked or reduced to the extent that the tube metal reaches its plastic flow temperature.

• At the superheater temperatures (around 538°C) in utility boilers, the formation of magnetite scale is greater than in the water components of the boiler. However, the scale does not adhere as easily to the surface. It tends to break down, particularly when the superheater cools after a shutdown. On restart, the magnetite may carry out with the steam and enter the steam turbine, where it can cause “hard particle erosion” of the turbine nozzles.

• Carryover from the steam drum into the superheater may cause deposits and tube corrosion. The most common carryover into superheaters is entrained water droplets. Boiler water usually contains boiler chemicals and other contaminants, such as oil, metallic copper, iron oxides, and silica. These may deposit inside superheater tubes, forming an insulating layer between the steam and the tube metal. The tube temperature may increase, promoting high temperature corrosion, both internally and externally. Direct contact spray attemperators (using boiler feedwater) are another contaminant source for the steam entering a superheater.

• Pitting of superheater tubes, particularly in pendant loops, is caused by exposure of water to oxygen during downtime. The water may be due to improper drainage of drainable superheaters, low spots in sagging horizontal tube banks, or a non-drainable superheater design. A superheater tube bank that remains flooded when the boiler is shut down develops corrosion cells. Some of the soluble deposits carried into the superheater are dissolved into the water and, when combined with the oxygen, form corrosion cells, usually wherever there is an interface between air and water. If unchecked this can result in pitting, which leads to pinhole leaks and potential rupture of the tube.

• Stress corrosion cracking is a common problem with superheaters made of austenitic stainless steel, which is ironically chosen for its corrosion resistance and strength. However, after long periods of exposure to high temperatures, cyclic stresses, and creep fatigue, the superheater headers may develop internal stress corrosion cracks, particularly in the ligaments between tube holes.

• External corrosion of superheater tubes is also a problem, particularly with boilers that burn solid fuels, such as coal and biomass.

Fuel Side Corrosion

Fossil fuels, except natural gas, contain impurities that promote corrosion on the flue gas side of boiler components. The impurities are usually compounds of sulphur, vanadium, and sodium. Solid fuels, including coal and biomass, produce additional gases, such as chlorine, along with solid combustion wastes (flyash), which present some unique corrosion opportunities.

Within a large, typical industrial boiler there are three temperature zones in which fuel side corrosion can occur.

• In the furnace. Here the generator tubes are subjected to a gas temperature of 1000°C – 1700°C and an internal tube temperature of 250°C – 400°C. Waterwall tube corrosion occurs in coal-fired units through the formation of pyrosulphates of sodium and potassium. In refuse-fired boilers, mixtures of chlorides of zinc, lead, iron, and sodium are the likely causes of corrosion and, due to the nature of the fuel, there may be other factors involved.

• In the superheater/reheater area. Here the gas temperature is approximately 650°C – 1000°C and the steam temperature is 500°C – 550°C. Superheater and reheater corrosion depends on components found in the fuel. The corrosive components are different for coal and heavy oil-fired boilers. Vanadium pentoxide mixtures with sodium oxide or sodium sulphate are the main corrosive compounds found in oil ash. For coal-fired boilers, sodium and potassium trisulphates react with iron to create high-temperature corrosion. In refuse-fired boilers, the deposits in the superheater banks are found to contain a high lead content.

• In the economizer/air heater area. As the flue gas passes through the economizer, it is cooled to approximately 150°C and the deposits have a higher percentage of phosphorous pentoxide. Then, as the flue gas passes through the air heater, it cools below the dew point of sulphuric acid, creating a very aggressive corrosion environment. In refuse-fired boilers the deposits from the economizer are predominantly calcium-rich sulphate material. It is believed that chlorine reacts with the scale, causing spalling and corrosion of the surfaces.

Dewpoint Corrosion

Most fossil fuels produce flue gases containing sulphur dioxide, sulphur trioxide, and water vapour. When these gases cool, the water vapour condenses and combines with the sulphur products to form sulphurous and sulphuric acids. The exact dewpoint depends on the concentration of the sulphur gases, but it is around 150°C. Any boiler component surfaces cooler than this (eg. air heater tubes, plates, casings, etc.) are susceptible to dewpoint corrosion. Flue gas leaks to atmosphere also reduce the temperature and cause corrosion at leak locations, such as openings to the furnace, support penetrations through the boiler casing, leaks around the superheater, reheater and economizer drain lines, piping penetrations, and the air preheater.

Fuel Ash Corrosion

Flyash in boiler flue gas increases corrosion. The flyash accumulates on surfaces throughout the flue gas path and acts as a sponge, collecting moisture, sulphuric acid, and other corrosive elements that originate in the fuel, thereby creating a corrosive environment.

Corrosion due to fuel ash occurs in oil, coal or refuse-fired boilers. Whether in the furnace, superheater, or reheater areas, the corrosion mechanisms are similar. The fuel ash deposit is a mixture of several compounds with low melting points. The mixture deposits on the tube surfaces and dissolves the protective iron oxide layer, exposing bare metal to the corrosive elements of the deposit.

In the case of furnace wall corrosion, the ash deposits contain mixtures of sodium and potassium pyrosulphates, with melting points between 300°C and 400°C. In coal-fired boilers, corrosion of superheaters and reheaters at 500°C and higher is attributed to trisulphates of sodium and potassium.

Heavy oil contains sodium, vanadium, sulphur (either elemental or as hydrogen sulphide), and other components, which produce an ash that melts at 480°C to 700°C. The molten deposits are highly corrosive. The deposits also convert SO2 to SO3, which then condenses and reacts with water vapour in the cooler sections at the back of the boiler, producing sulphuric acid (H2SO4). This can cause severe corrosion in the economizer, air preheater, and flue gas outlet. Water washing the fireside of an oil-fired boiler may also create acid corrosion, due to acids that form as the wash water evaporates on the tubes. To prevent fireside pitting, water washes should include a rinse with an alkaline solution.

In biomass-fired boilers, including municipal refuse boilers, chlorine is often found in the fuel (originating from polyvinyl chloride). The flue gas may contain chloride compounds (of iron, sodium, phosphate, zinc, and lead) with melting points as low as 175°C to 300°C. Several opportunities for corrosion are created by the chlorine and by these alkaline chlorides. Hydrochloric acid (HCl) may condense in low temperature areas of the boiler outlet. Free chlorine, available in the ash deposit, will react with and remove iron from the parent metal, forming FeCl2. The molten ash can react with and remove the magnetite layer, then attack the metal. Corrosion rates can be very high, with up to 12 mm of metal loss per year and failure in less than 2000 operating hours.

In addition to the above problems, fuel ash corrosion may cause fatigue cracks in the following manner. Some of the ash deposit melts, creating a thin liquid film between the tube surface and the solid ash deposit. The ash layer thickens until the film cannot support the increased mass; the ash layer falls off, exposing the tube surface to furnace heat. Local heat flux increases, causing a sudden increase in tube temperature and an area of high local stress. As the ash layer re-forms, it insulates the tube and reduces the metal temperature and stress. This cycling of temperature and stress repeats continuously, causing corrosion fatigue cracks.

Appearance of Fuel Ash Corrosion

For superheater and reheater tubes in an oil-fired boiler, the corrosion pattern depends on the mass and aerodynamics of the flue gas as it flows through the tube banks. In coal-fired boilers, an overheated area is indicated by a thickening of the oxide scale, which exhibits a series of cracks or grooves in the tube surface, often described as looking like “alligator hide.”

The external surfaces of waterwall tubes may show a series of circumferential grooves or cracks, while a cross-section of the tube may show shallow cracks or grooves through the tube wall. At the tip of a crack, the tube wall may be thin enough to produce a steam leak. In the burner area, an erosion pattern may be evident on the tubes, due to unburned coal particles, flame impingement, or both. In the furnace of a refuse burner, the waterwall corrosion usually appears as smooth, uniform wastage with generalized thinning of the tubes.

The difference between the ‘alligator-hide’ on superheater tubes and the cracking or grooving on waterwall tubes is due to the different heat flux and temperature spikes. In their hottest areas, the heat flux for waterwall tubes is three to four times that of superheater tubes. In biomass- and refuse-fired boilers, the heat flux is insufficient to cause high temperature peaks, so corrosion proceeds more uniformly.

Steam and Condensate Systems

Steam Condenser Corrosion

A steam condenser is a heat exchanger that is an intimate part of a steam-condensate cycle in a power plant. As such, any corrosion that occurs on the shell side can result in undesirable products being carried back to the boiler by the feedwater. The types of corrosion depend on whether the condenser tubes are made of copper alloys, stainless steel, or high performance alloys.

For condensers with copper alloy tubes, tube corrosion may occur as follows:

• Copper alloys (eg. brass) are readily corroded by ammonia. When hydrazine is used to assist oxygen scavenging, it breaks down into ammonia compounds, which enter the condenser with the steam. The ammonia combines with condensate and the resulting solution attacks the tubes at the center of the condenser, causing ammonia grooving. The copper that is removed from the tubes is carried with the condensate into the feedwater system and back to the boiler, where it may deposit on boiler tubes.

• In the presence of ammonia, the copper alloy tubes, which have a high residual stress, are susceptible to ammonia stress corrosion cracking, which leads to tube leaks.

• Corrosion on the water side is potentially due to erosion, under-deposit corrosion, and galvanic corrosion. Galvanic corrosion is encouraged by the cooling water, which acts as an electrolyte, and the dissimilar metals of the waterboxes and the tubes. To combat this, common practice is to install sacrificial anodes inside the water boxes.

For condensers with stainless steel tubes, corrosion may occur as follows.

• Stainless steel is not affected by boiler chemicals, including ammonia. However, stainless steels that contain nickel are subject to chloride stress corrosion cracking. This requires stress, chlorides, and a temperature of at least 300°C.

• Corrosion on the water side is due to crevice corrosion, pitting, or microbiological corrosion.

Feedwater Heater Corrosion

Feedwater heaters are designed to improve boiler efficiency by extracting heat from streams such as boiler water blowdown, turbine extraction steam, or turbine exhaust steam. Feedwater heaters are classed as either low-pressure (upstream of the feed pump) or high-pressure (downstream of the feed pump). They are also classed as open, in which the water contacts the heating fluid, or closed, in which the water doesn’t contact the heating fluid.

Regardless of design, the corrosion problems are similar, with the most common sources of corrosion being oxygen and improper pH. Due to temperature increase across the heater, incoming metal oxides are deposited in the heater and then released during changes in steam load and chemical balances. Stress corrosion cracking of welded components can also be a problem. Erosion is common in the shell side, due to high-velocity steam impingement on tubes and baffles.

Deaerator Corrosion

Deaerators present opportunities for several types of corrosion. This is due to their design, the flow patterns through them, and the mixing of water, steam, and gases. The common corrosion problems in deaerators include the following.

• Flow accelerated corrosion. In areas where water flow is high, the protective oxide layer may be removed from the surface, exposing it to continual loss of thickness.

• Oxygen pitting. Since free oxygen is released inside the deaerator, oxygen pitting may occur around the vent and at any location where the oxygen may become trapped. The pitting may occur in the shell or any of the internal components, such as tray supports.

• Erosion. Inlet baffles at the steam inlet and other locations where high velocity water impinges on the metal surface are susceptible to erosion corrosion.

• Corrosion fatigue cracking and stress corrosion cracking. This cracking problem has occurred in a large percentage of deaerators and their associated water storage compartments, usually after they have been in service for three or more years. All designs of deaerators, regardless of manufacturing methods or steels of construction are susceptible to cracking, which is believed caused by corrosion fatigue, often along with stress corrosion cracking.

Cracks mainly occur in the water storage compartments, at or below the water level. They begin at welds, corrosion pits, and heat-affected areas, most often at head-to-shell welds, but also on the internal surface of the shell where attachments are welded. Corrosion fatigue is caused by cyclic stresses due to startups and shutdowns, vibration, flow and level fluctuations. Stress corrosion cracking may occur in areas of high local stress, if caustic is carried in with the condensate.

Deaerator cracking can be minimized by maintaining steady operation (thus minimizing stresses), ensuring good control of oxygen scavenging chemicals, and maintaining all internal components in good working order. In addition, regular magnetic particle testing of internal welds is mandatory, along with immediate repair of any detected cracks.

Heat Exchanger Corrosion

Heat exchangers exist in many designs and for many different purposes, which involve heat exchange between many different fluids. The corrosion tendencies of a particular heat exchanger are mostly dependent on the fluids involved, the flow rates, the temperatures, and the metals used for exchanger construction. Regardless of purpose, there are several corrosion concerns that are common to most shell-and-tube exchangers.

• Crevice corrosion. If there is any gap (crevice) between the tubesheet and tube, deposits may collect and initiate corrosion in the crevice.

• Galvanic corrosion. To promote heat transfer, the tubes of a heat exchanger are often made of a different metal than the tubesheet or the shell. This presents opportunity for galvanic corrosion to occur. An effort is usually made to use metals that are as close as possible to each other in the galvanic series.

• Stress corrosion cracking. The nature of heat exchangers requires them to go through drastic temperature changes, particularly during startup and shutdown. In addition, the tubes and the shell are at different temperatures. Expansion and contraction at different rates may cause stresses to occur and after many such cycles the tubes may develop stress corrosion cracks. These are usually circumferential and may cause the tubes to leak.

• Pitting. This is highly dependent on the fluids in the exchanger. Entrained gases that may be released as the fluid temperature changes may attack the shell or the tubes.

• Erosion corrosion. High flow velocity, particularly through the tubes, has the potential to cause erosion of any protective film and of the bare metal. This may occur when the exchanger flow exceeds design or when some tubes become plugged, forcing excess flow through the remaining tubes. Turbulence as the flow enters the tubes will also cause erosion at the tube ends. Another area highly prone to erosion is at tube bends, such as in U-tube exchangers.

• Under-deposit and or microbiological corrosion. Operating an exchanger with the tube or shell flow below acceptable turndown rate may cause deposits to collect. If the fluids contain microbes, these will also deposit and create a corrosive environment.

Corrosion Monitoring

Corrosion monitoring involves a combination of continuous and intermittent activities for the purpose of obtaining comprehensive information on corrosion conditions over time. It represents a commitment to pro-actively quantify and control corrosion rates and to avoid conditions that could lead to corrosion-related failures. Real-time corrosion monitoring requires simultaneous linking of process parameter changes with measured corrosion information, from sensitive instrumentation. Full value is obtained from corrosion monitoring only if the information gathered is applied to an effective corrosion control program.

There are several methods used to monitor either the actual corrosion of the metal or the corrosive nature of the environment. Since each method has its limitations, it is common practice to use more than one method. The main methods used include visual inspection, corrosion coupons, corrosion probes, microbiological testing, chemical analysis, and non-destructive examination.

1. Visual Inspection

When a plant or a piece of equipment is shut down and opened, the opportunity is taken to inspect for internal corrosion. Quantitative measurements are taken carefully recorded. In terms of corrosion, the condition of an internal surface may be defined in one of the following ways.

No Corrosion - the metal surface appears to be entirely unaffected.

Definite Surface Corrosion - corrosive attack is deep enough to catch a knife blade. It may be further described as dulled, matte, or roughened.

Shallow Metal Attack - describes the removal of a perceptible, although barely measurable, amount of metal from the surface.

Pitted or Grooved - the metal is visibly removed to a measurable depth in the form of pits or grooves. The size, depth, shape, and distribution of pits and grooves are recognizable. The direction of grooving can also be seen.

Blistering or Scaling - thin layers of metal appear to have detached or sloughing off from the metal surface to form blisters or scale. This may indicate hydrogen diffusion.

Cracking - cracks may be visible to the naked eye, such that size, quantity and direction of cracks can be easily determined. Visible cracks are a serious, weakening condition and other, smaller cracks are likely present requiring other inspection methods over a wide area.

Visual inspection is also the primary method for external surfaces of piping, vessels, supports, etc, where atmospheric corrosion is a concern. It is also applied inside heaters, furnaces and flue gas paths, where products of combustion could create corrosive conditions. In all cases, the visual inspection may lead to further, more technical testing.

2. Corrosion Coupons

Corrosion coupons are carefully machined pieces of metal that are exposed directly to the process stream (ie. the corrosive environment). Coupons can periodically be removed and either visually inspected or sent to a lab to test for mass loss, corrosion rate, corrosion type and severity (eg. pitting). The results of these tests allow evaluation of one or more of the following.

• system corrosiveness

• material performance

• inhibitor performance

A coupon, made of the same metal as the pipe or vessel, is weighed to an accuracy of four decimal places before being inserted into the system and again after removal, to determine the mass of metal lost to corrosion. Coupons are often installed in pairs for redundant, comparative measurement, and are made from a variety of metal alloys, in several shapes and configurations.

• Flat plates, either solid or with holes drilled through,

• Round, metal rods, either singly or in groups of six or eight,

• Banded coupons, in which a band is wrapped around a conventional coupon, creating a crevice. The crevice creates a concentration cell so attack occurs under the band. This method is useful when oxygen corrosion is suspected.

• Spools or nipples are threaded or flanged into the system in such a way that they can be removed for evaluation.

Coupon Holders and Insulators

Coupons are securely mounted in such a way that they are electrically isolated from contact with all other metals, thus preventing galvanic corrosion (except when testing for galvanic corrosion). Mounting materials (brackets, bolts, and insulating materials) should be fully resistant to the environment to avoid loss of data or loss of electrical isolation. Holders must have isolation valves for positive isolation removal of the coupons. Some holders are retractable and can be withdrawn while the system is under full operating pressure.

Figure 5 is an example of an externally mounted corrosion test coupon rack, which provides a convenient means of monitoring the progress of corrosion in systems such as condensate lines and cooling water systems. The rack may be made from PVC or metal, depending on the system being monitored.

Figure 6 shows a retractable coupon (also called a slip-in rack), which allows coupons to be inserted and removed from the system without interrupting the process. It requires a gate valve and a nozzle, which serves as a retraction chamber. A coupon holding rod is contained in the retraction chamber, which is flanged to the isolation gate. The outside end contains a packing gland, through which the coupon holder passes.

A coupon is mounted on the rod and then drawn back into the retraction chamber. The chamber is bolted to the gate valve; the valve is then opened and the coupon is slipped through the gate valve and into the process stream.

Figure 7 shows an inserted rack coupon holder. It requires flanged connections that allow installation directly into the piping, with the coupons positioned across the flow path. In this design, the process must be shut down to install or remove the rack.

Coupon Handling

Test coupons are kept in a special treated envelope before and after exposure. Coupons must not be touched with bare fingers when being mounted in the coupon holder, since oil from a finger print interferes with corrosion. After removing a coupon from the rack, the coupon must be air dried, but not cleaned, before being placed in a treated envelope for delivery to the lab.

Chemical Inhibitor Performance

Corrosion coupons may be used to monitor the performance of a corrosion inhibitor. The uninhibited corrosion rate of a system is determined and used as a base corrosion rate. Once the base corrosion rate has been established, corrosion inhibitor feed is started and corrosion coupons are installed in the system. The performance of the inhibitor is usually expressed in terms of percent protection, where:

3. Corrosion Probes and Meters

Corrosion probes and meters are devices/instruments that can be inserted into the path of a process flow or into a vessel for the purpose of measuring the corrosiveness of the fluid. The probes are connected to external instruments, which provide a reading relative to the corrosion. Four common devices are electrical resistance probes, galvanic probe meters, linear polarization meters, and hydrogen probes.

Electrical Resistance Probes

The principle of electrical resistance can be used to indicate metal loss by corrosion. As the cross-sectional area of a conductor decreases, the electrical resistance increases. The electrical resistance of any conductor is given by

Using this principle, a current-carrying element (loop of wire) may be exposed to a corrosive environment. As it corrodes, its cross-sectional area decreases, causing its electrical resistance to increase. Electrical resistance corrosion rate meters operate on this principle. These meters use a probe with two similar elements - a corrosion element, which is exposed to the corrosive environment, and a reference element, which is inside the probe and protected from the corrosive environment. Since the resistance of the corrosion element is affected by temperature changes, the reference element provides compensation for any temperature variations. This arrangement is only useful in lower temperature environments. In medium to high temperatures, a variable temperature electrical resistance (VTER) probe, with associated electronics, is used. This type uses a single, corrosion element, with no need for a reference element.

Since the resistance probe does not require an electrolyte to carry current, it can be used effectively in gas systems to measure corrosion from stack gases. Accumulation of corrosion products, such as iron sulphide, on the probe can cause resistance to decrease, which falsely suggests a decrease in corrosion rate. The probe does not reliably measure pitting, so it must be removed regularly and visually checked for pitting.

Figure 8 shows a resistance probe. Resistance readings are plotted against time and the slope at any point on the plot may be converted to a corrosion rate.

Galvanic Probe Meters

Another type of meter is the galvanic probe meter, which uses the principle of galvanic action. It involves two dissimilar metals, immersed in the process fluid (which acts as an electrolyte) and connected electrically to a meter. A current flows due to the potential differences of the metals and this current is proportional to the corrosiveness of the system. A typical assembly contains brass and steel electrodes, connected to an ammeter, which measures current flow. It does not give a direct measurement of corrosion but detects changes in the corrosiveness of the system.

Linear Polarization Meters

Linear polarization meters indicate the corrosion rate of a coupon or electrode at the moment of the measurement. They are used to continuously monitor any fluctuations within a system (such as changes to corrosion inhibitor dosages) and their effects on the corrosion. Typical applications of these meters include:

• High pressure water systems,

• Cooling water systems,

• Condensate systems,

• Potable water treatment and distribution systems,

• Pulp and paper manufacturing, and

• Amine systems.

In this meter, a small current is caused to flow between a test electrode and a working electrode. An electrolyte, such as water, must be present to conduct current between the probes. The current flow alters the electrical potential of the “test” electrode. The corrosion rate (or corrosion current) is proportional to the test current divided by the change in potential when the potential change is small (20 mV or less). Sufficient current is applied to the test electrode to change its potential by a specific amount (for example, 10 mV); the current flow required to make the change is directly measured.

Linear polarization meters can have probes with either two or three electrodes. The reading from the two-electrode probe must be corrected for the resistivity of the process fluid. The potential change of the test electrode is measured with reference to the second, working electrode. The three-electrode meter requires no correction for solution resistivity. The third electrode has no current applied to it and is allowed to corrode freely. It serves as a reference electrode.

Figures 9 and 10 are simplified sketches of two- and three-electrode linear polarization meters.

Hydrogen Probes

Hydrogen probes measure the flow of the atomic hydrogen that is generated from corrosion and which passes through steel, causing hydrogen-induced corrosion (HIC). The areas most susceptible to HIC are carbon steel pressure vessels and piping that contain process fluids with H2S, cyanide, or arsenic. Hydrogen is also produced as a by-product of corrosion wherever water is in direct contact with the steel.

There are three basic types of hydrogen probes. The simplest consists of a thin-walled, carbon steel tube (hydrogen probe) with a solid rod inside the tube, forming a small annular space. This probe is inserted into the flow path. Any hydrogen atoms that escape from the carbon steel collect in the annular space and combine to form hydrogen molecules, which are too large to pass back into the process. As hydrogen gas accumulates, the pressure in the annular space increases and registers on an external pressure gauge.

The second type uses patch probes that are attached and sealed against the outside of the process piping wall. The principle is identical to the internal probes, except the path probes collect hydrogen atoms that penetrate to the outside of the pipe wall.

The third type is the palladium foil type electrochemical cell, which produces an electrical output proportional to the hydrogen evolution rate. Figure 11 illustrates this design.

4. Microbiological Testing

Generally, microbiological activity is determined from measurements used to detect chemical fouling and corrosion. These include analysis of corrosion coupons, conductivity, electrical resistivity, or evidence of pressure drops and heat transfer losses. Fouling probes or heat transfer can indicate the amount of surface deposits, but cannot identify the composition of the deposit. Similarly, the usual methods used to assess corrosion are not specific to biological activity.

The presence of a biofilm is determined from counts of biological organisms cultured from water samples. While microbiological culturing techniques can provide data on the densities of organisms found in process fluids, they require at least 24 hours for incubation, and the data obtained does not necessarily correlate with surface fouling. With no definitive means to determine the onset of biofilm formation, most facilities apply biocide treatments on a preset schedule or as needed, corresponding to visual evidence of growth. This uncertainty may lead to improper dosage of biocide to the process stream.

5. Chemical Analysis

Though not as effective as direct corrosion monitoring, a complete chemical analysis of makeup and cooling water may help detect corrosion products in the water. Since a primary indication of corrosion is the accumulation of iron in the process fluid, it is common practice to test for iron concentration. For chemical testing, both filtered and total iron testing is essential. Running a filtered iron test gives a snapshot of corrosion tendencies because any iron that does not pass through a filter paper can be safely assumed to be the result of corrosion.

Filtered iron testing is done by filtering a sample across a 0.2 micron filter pad and then testing the filtrate for iron. If the filtered sample is greater than 0.25 ppm iron, then corrosion control is inadequate. The filter paper can be saved as a qualitative measure of system cleanliness.

Total iron testing is done by boiling a sample with hydrochloric acid and hydrogen peroxide to dryness, then reconstituting it with distilled water. This procedure identifies all iron present in the water whether soluble or insoluble.

The insoluble, suspended iron removed during filtration of a sample is ferric iron which has usually flaked off a corrosion site and been transported in the water stream. The total iron measurement, minus the filtered iron reading, provides a measure of the soluble iron in ferrous form, which usually occurs naturally in the water. Insoluble iron is an indication that corrosion has occurred, while soluble iron is an indication of the corrosive potential of the water and of the likelihood that iron will deposit on downstream heat transfer surfaces.

6. Non-Destructive Examination

As the name implies, non-destructive examination involves methods by which the surface and internal integrity of a material may be tested without doing any damage to the material itself. The methods used are able to detect both surface and subsurface corrosion and, in some cases, can be used while the equipment is still in service. The most common methods of non-destructive examination include ultrasonic, radiographic, magnetic particle, and dye penetrant inspection.

Ultrasonic Inspection

In most cases, ultrasonic inspection can be used while the inspected equipment is still in service. During ultrasonic inspection, high frequency sound waves are transmitted into the material being inspected. These waves travel through the material until they encounter the opposite surface or an internal discontinuity. Material interfaces or discontinuities reflect the sound back to its origin, where it can be detected. The time interval between transmission of the ultrasonic wave and the arrival of the reflected wave back at the point of transmission is directly proportional to metal thickness or the depth of the discontinuity.

Ultrasonic waves are generated by transducers, which are constructed from piezoelectric materials, such as quartz. These materials are capable of expanding and contracting when subjected to a changing electrical field and will also produce an electrical potential when the material is placed under mechanical stress. Therefore, an ultrasonic transducer can be placed on a metal surface and used to both create and detect ultrasonic energy.

Guided Wave Testing (GWT)

Guided Wave Testing (GWT) is also known as Guided Wave Ultrasonic Testing (GWUT) or Long Range Ultrasonic Testing (LRUT). GWT is widely used to inspect many engineering structures, including metallic pipelines, boilers, and heat exchanger tubes. It uses very low ultrasonic frequencies (10-100 kHz) compared to conventional UT. This long range ultrasonic method uses mechanical stress waves, which propagate along an elongated structure while guided by its boundaries. The waves travel a long distance with minimal energy loss. A pipe or tube can be quickly screened and any discontinuities along the length are quickly identified. Where a possible defect is identified, follow-up inspection with conventional UT or other NDT methods is usually required to obtain detailed information on the nature and extent of the defect.

Radiographic Inspection

Radiographic inspection is a nondestructive tool used mostly for weld inspection, but can also be used to detect corrosion. It uses X-rays or gamma rays, which are emitted from a source and passed through the object being inspected, onto a piece of photographic film. When developed, the film is called a radiograph.

The amount of radiation that passes through a metal in a given length of time is inversely proportional to its thickness. This means that more radiation passes through a thinned, corroded area than through a thicker, undamaged area. The photographic film becomes darker as it is exposed to increased amounts of radiation, causing pits or corroded areas to show up as dark spots on the radiograph.

Radiographs are frequently taken in the horizontal plane to give a profile of pipe walls. They can also be taken in the vertical plane to establish the pattern of attack on the bottom or top of piping or vessels. Radiographic inspections can be carried out while equipment is in service.

Magnetic Particle Inspection

Magnetic particle inspection is primarily used to locate cracks that extend to the surface of a metal. It can only be used to inspect magnetic materials, such as steel. The metal to be inspected is magnetized using an electromagnet or a large permanent magnet. Cracks in the metal surface will break the magnetic lines of force and produce local leakage fields. When finely divided magnetic particles, such as iron powder, are sprinkled on the magnetized surface, they are attracted to the cracks by the localized fields, making it possible to determine their size and exact location.

Dye Penetrant Inspection

This method detects flaws or cracks that extend to the surface of a metal. A thin liquid with low surface tension is applied to the clean metal surface. By capillary action, the liquid draws into any cracks or surface discontinuities. Excess liquid is wiped from the surface and then a developer is applied. The developer acts like a blotter, drawing the liquid back out of the defect and making the defect easily visible.

There are two types of penetrant. One contains a red dye, which provides a good contrast with the white developer. The other contains a dissolved substance that makes the penetrant fluorescent, so the defects become easily visible under an ultraviolet light.

Design Considerations

The first opportunity to control and prevent corrosion within a facility occurs during the design stage. To ensure the integrity and safety of a facility, the designer must give foremost consideration to all aspects of corrosion and the specific factors that can affect corrosion throughout the process. In addition to the economic consideration discussed in Objective 1, the designers must consider and design for at least the following.

• External environments

• Internal environments

• Mechanical conditions

• Operating conditions

• Corrosion protection

• Corrosion monitoring

External Environments

Internal Environments

What are the internal fluids (liquid, vapour, or gas) within the process piping and vessels and how do they interact and flow? The corrosive nature/chemistry of each fluid in relation to the materials it will contact; entrained solids, erosion; flow patterns (turbulence, velocity, impingement); temperatures of fluids (affecting oxidation, condensation, etc.).

Mechanical Conditions

Operating Conditions

What are the expected operating conditions that might affect corrosion? Steady operation or “up and down”; extended down times; fluctuating pressures, temperatures and flows; inspection frequency; defined corrosion program and staff.

Corrosion Protection

What permanent systems and equipment must be installed to directly protect the facility against corrosion? Cathodic protection for underground piping (ground beds, impressed current, etc); internal coatings and linings for piping and vessels; thermal spray; external paints, coverings or wrappings (eg. underground lines), inhibitor injection system; chemical injection systems (for boilers), convenient and sufficient drainage points during shutdowns.

Corrosion Monitoring

What corrosion monitoring components must be designed into the facility? Strategically placed electrochemical corrosion probes and data collection equipment; installation points for corrosion coupons; convenient access openings for visual inspection (manways into vessels, handholes, cover plates, etc).

Operational Methods

In an operating facility, control and prevention of corrosion involves a combination of the operational strategies and the designed-in components. Operational strategies include a good, recognized corrosion management plan plus diligent application of corrosion inhibiting chemicals and operational procedures that minimize corrosion opportunities. Designed-in components, beyond material selection, include cathodic protection, deaeration, and protective coatings.

Corrosion Management

Every operating facility, in one way or another, is involved in corrosion management. Those with a clear corrosion management program are generally more successful at minimizing corrosion issues and thereby maximizing plant reliability, performance, and safety. The key components of a good corrosion management program include the following:

• Clearly understood policies and procedures

• Corrosion prediction and risk assessment

• Continuous measuring, monitoring and tracking

• Regular inspections and corrosion analysis

• Defined roles and responsibilities for the corrosion program

• Consultation with external corrosion “experts”

• Operational procedures that minimize corrosion

• Effective training of operating personnel

• Timely maintenance

Chemical Inhibitors

A corrosion inhibitor is a chemical substance that is added to a potentially corrosive environment where it reacts to decrease (or prevent) corrosion. The chemicals are injected at specified locations in a system. The operators must test regularly and adjust chemical feed to ensure that a minimum chemical residual concentration is always present. Some inhibitors are designed to inhibit the anodic reaction, while others inhibit the cathodic reaction.

Anodic inhibitors are also called passivating inhibitors, since they work by reacting with the metal surface in an oxidizing reaction to produce a protective oxide coating. This coating makes the anode passive (nonreactive with its environment). The most common anodic inhibitors are nitrite, molybdate, and orthophosphate.

Cathodic inhibitors (also called precipitating inhibitors) work by chemically precipitating a thin layer, which adheres directly to the metal surface, thus physically isolating the fluid (cathode) from the metal. Common cathodic inhibitors include zinc carbonate and calcium carbonate.

Corrosion inhibitors function by one of the following mechanisms:

• adsorbing as a thin protective film on the metal surface

• inducing the formation of a thick, unstable corrosion product

• forming a passive film on the metal surface

• changing the characteristics of the environment

• producing a protective precipitate, or

• removing a corrosion-enhancing constituent (eg. oxygen scavenging)

Inhibitors in Steam/Condensate Systems

Steam condensate systems require special attention to corrosion inhibition. This is because boilers generally release trace amounts of oxygen and carbon dioxide with the steam. Oxygen will react with metal to produce pits, while carbon dioxide will combine with water (when the steam condenses) to produce carbonic acid, lowering the pH of the condensate, which exposes the metal to acid attack. For corrosion inhibition, these systems use neutralizing amines, filming amines, or a combination of both.

Neutralizing amines are volatile, alkaline chemicals (such as ammonia, cyclohexylamine, morpholine) which react with carbonic acid to neutralize it, thus increasing the pH. They do not, however, protect the metal from oxygen attack. Control involves feeding the amine at a continuous rate that maintains a specified pH level (usually 8.5 – 9.5).

Filming amines physically isolate the metal from the water, by creating a very thin (one molecule thickness) film on the metal. This protects against both oxygen and carbon dioxide. Control involves maintaining a specified concentration of amine in the system, then monitoring the iron concentration or corrosion coupons.

Cathodic Protection

Cathodic protection is a method of reducing or eliminating the corrosion of metallic structures located in corrosive environments, by causing them to act as non-corroding cathodes. This is done by ensuring that the electrical potential of the protected surface remains less than that of other metals in the same corrosive environment.

There are two ways that cathodic protection can be achieved. One method involves the placing of sacrificial anodes in the same corrosive environment as the metal to be protected. These metals are more anodic (less noble) than the protected material, and corrode preferentially. The other cathodic protection method involves impressing an electric current on the corrosion cell so that electrons always flow toward the protected material, ensuring it is always cathodic. The primary application of cathodic protection is the protection of underground equipment, such as piping and storage tanks.

1. Sacrificial Anodes

When two metals of differing electrochemical potential are electrically connected to each other, and placed in a common electrolyte (such as soil), electrons will flow from the more active metal to the less active metal. If the more active metal happens to be a pipeline, then it becomes the anode and loses metal. Even without other metals in the near vicinity, an underground structure may corrode by developing interactive anodic and cathodic areas on its own surface. This corrosion can be shifted away from the structure and directed to sacrificial anodes, which are buried a few metres from the structure and connected directly to it by some conductive material. The anodes are made of a material that is significantly higher in the electromotive series than the protected structure.

Refer to Figure 12. Galvanic action causes electrons to flow from the magnesium or zinc anode, through the insulated connecting cable, to the pipeline. The magnesium or zinc ions that result from the loss of electrons are soluble, and enter the electrolyte (which in this case is the soil). The sacrificial anode therefore loses metal ions (corrodes) thus ensuring the pipeline does not. Carbonaceous backfill around the anode promotes a more uniform loss of metal.

Magnesium and zinc are the most common anode materials, since they are high in the electromotive series. The efficiency of zinc anodes is improved to nearly 90% by alloying with aluminum. Magnesium alloys are used primarily in soils and areas that have high resistance to current flow. Zinc alloys are used in highly conductive electrolytes at ambient temperatures and in applications where sparking must be avoided, such as in storage tanks for flammable liquids.

Sacrificial anodes may also be installed within vessels to protect their internal surfaces from corrosion. Such anodes may be found in tanks, condenser waterboxes, and heat exchanger headers.

Advantages of sacrificial anode systems include:

• No external power requirements

• No voltage regulation requirements

• Easy to install and replace anodes

• Minimum maintenance requirements

• Inexpensive if installed at time of pipeline construction

Disadvantages of sacrificial anode systems include:

• Differential potential depends on the electrolyte

• Small area of coverage per anode

• Poorly coated pipelines may require many anodes

• Protection reduced in highly-resistive soils

• Expensive if installed after pipeline construction

2. Impressed Current System

The nature of the environment may not produce a strong electrolyte, so that a galvanic current may not occur naturally. In this case, the current must be impressed (created) by an external power source. An impressed current system (Figure 13) uses consumable, semi-consumable, or non-consumable anodes, which are connected to a DC power source. The DC power source may be batteries or (commonly) rectified AC. The protected structure is always connected to the negative terminal of the DC power supply. Electron flow is from negative to positive. The DC power supply maintains a flow of electrons to the protected structure, keeping it cathodic regardless of soil conductivity. The DC power supply also maintains a flow of electrons from the anodes, thus maintaining their anodic behaviour regardless of soil conditions.

Advantages of an impressed current system

• Large areas can be protected by a single installation

• Can be designed for a wide range of voltage and current

• High current flow is available from a single groundbed

• Effectively protects bare and poorly coated pipelines

Disadvantages of an impressed current system

• Requires a power source which is subject to interruptions

• May cause cathodic interference problems

• Requires a higher level of maintenance

• Overprotection can damage protective coatings

Groundbeds

A groundbed is a system of many anodes spread in a pattern over a wide area. There are four general types of groundbed.

1. A conventional groundbed is a group of anodes installed about 100 m from the protected structure and spaced 5 m to 10 m apart. It distributes protective current over a wide area of the protected structure. It is also called a remote groundbed.

2. A distributed anode groundbed is a group of anodes installed along the protected structure, less than 10 m from the structure and spaced 10 m to 200 m apart. It is used to avoid interference from nearby structures and to better protect sections of bare or poorly coated metal.

3. A deep anode groundbed is a group of anodes installed vertically in a drilled hole at depths of 15 m or more. It is used when space isn’t available for a conventional groundbed or when the upper soil has high resistivity and the deeper soil has low resistivity.

4. A shallow vertical groundbed is a group of anodes installed vertically in a drilled hole at a depth of less than 15 m. It is used when horizontal space is very limited.

Monitoring and Maintenance

Cathodic protection systems require regular monitoring. In sacrificial systems, voltage readings indicate if the system voltage is declining, which indicates anode wastage. At a specified minimum voltage, the anodes must be replaced. The voltmeter must be very sensitive, since normal voltage levels are very low. For example, a buried magnesium anode protecting a carbon steel structure has a nominal voltage of only 0.85 V.

Impressed current systems must be monitored for current flow and adjusted when necessary to ensure the current is optimized for changing conditions. Anode wastage, variations in soil resistance (due to groundwater changes), and changes in the protected structure can be compensated for by adjustment of the DC current from the rectifier

Protective Coatings

Ideally, an engineering material (metal in particular) should have the necessary mechanical properties for its application plus be fully resistant to corrosion in its operating environment. Often this ideal is not possible and the mechanical properties must take precedence. Furthermore, the traditional methods of corrosion control may not be practical or possible. This is when a protective coating (or a protective lining) becomes necessary. The primary purpose of a protective coating is to physically isolate a metal surface from the environment, so that corrosion chemistry and corrosion cells cannot occur at the surface. The most common applications of coatings are external surfaces and vessel internals.

Metallic Coatings

Metallic coatings change the surface properties of the parent metal to those of the metal coating. The parent metal becomes a composite material, which exhibits properties not achievable by either material on its own. The coating provides a durable, corrosion resistant layer, while the parent metal provides the strength.

The most popular coatings are cadmium, chromium, nickel, aluminum, and zinc. Methods for applying a metallic coating include: electroplating, electroless plating, thermal spraying, hot dipping, chemical vapour deposition, ion vapour deposition, and galvanizing.

Galvanizing is the most widely used method for metallic coating. It involves coating carbon steel with a thin layer of zinc. Galvanized piping is mainly used for low temperature liquids. At temperatures below 77°C, the metallic zinc is the anode, but above 77°C the steel becomes the anode.

Nickel coatings are applied using a chemical process that depends upon the catalytic reduction of nickel ions in an aqueous solution, which contains a chemical reducing agent. The nickel deposits without the use of electrical energy (hence called ‘electroless plating’). Due to its exceptional corrosion resistance and hardness, this process is widely used to plate items like valves and pump parts to enhance the life of components that are exposed to severe service conditions, particularly in oil field and marine industries.

Metal cladding improves the corrosion resistance of a material by metallurgically bonding to its susceptible core material a surface layer of a metal or an alloy with good corrosion resistance. The cladding material is selected not only to have good corrosion resistance, but also to be anodic to the core alloy by about 80 to 100 mV. Therefore, if the cladding becomes damaged, or if the core alloy is exposed at drilled fastener holes, the cladding provides cathodic protection by sacrificially corroding.

Cladding is up to 80% cheaper than a solid pipe of the same material. Clad materials are widely used in the chemical industry, offshore oil production, oil refining, and the power generation industry.

Non-Metallic Coatings

Chemical treatment of a metal surface can produce a coating of metallic oxide that has better corrosion resistance than the metal itself. Examples of chemical treatments that produce a non-metallic coating include anodizing, nitriding, and phosphating. Chemical treatment of the metal surface is also used as a preparatory step prior to painting or coating the metal.

Many paints and liquid coatings have been developed to protect equipment from corrosion. The petroleum industry produces most of the basic ingredients for synthetic coatings. Paints and coatings have been developed for three main purposes; impermeability, corrosion inhibition, and cathodic protection.

Coatings designed for impermeability are designed to completely separate the metal from the corrosive environment and to have excellent adhesion to the metal. The coating used depends on the environment and the service requirements.

Coatings designed for corrosion inhibition function by providing a protective barrier that will react with the environment; for example, red lead based paints used on ships hulls and also bridge steelwork.

Coatings designed for cathodic protection usually contain a base paint (with zinc pigment added), which acts as an anode. Example – anti-rust paints.

Typical non-metallic coatings used in industry include:

• Epoxy resin coatings, applied as a liquid or powder

• Bituminous coatings, which may be applied hot or cold, including coal tar enamels and asphalt mastics

• Thermoplastic coatings, such as polyethylene, vinyl and plastisols

• Specialty coatings, which include urethanes, fluorocarbons, phenolics and polyesters

• Elastomeric coatings, including natural rubber, butyl rubber, neoprene, hypalon, and nitrile rubber

• Inorganic coatings and linings, which also include glass

Failure Analysis

Whenever a failure occurs, a structured effort must be applied to analyze the failure and determine the initiating cause. The failure analysis should be a step-by-step approach that attempts to identify the single event or condition that began a chain of events, leading to the failure. Unfortunately, failure analysis is a reactive procedure, in that failure has already occurred. However, the importance of finding the root cause is that, once discovered, it becomes the primary focus of attention in preventing recurrence of the failure. Eliminating the root cause removes or restricts the opportunity for all other events in the failure chain to occur.

Failure analysis techniques can be applied to other purposes, including, but not limited to, safety (such as accident or incident analysis) and quality control (such as product analysis), and production (such as procedure analysis). There are also many different analysis methods that may be applied and many different analysis tools that may be used. Regardless of the method, the underlying purpose is continuous improvement.

There are four main phases in a complete failure analysis; the component failure analysis, the root cause investigation, the root cause analysis, and the learning.

• Component failure analysis - This phase looks at the specific equipment that failed to identify the general nature or cause of the failure (eg. corrosion, fatigue, mechanical damage) and the operational events that may have influenced it.

• Root cause investigation - This is a totally objective and more in-depth discovery of the facts, leading to unbiased statements of what occurred leading up to the event, during the event, and surrounding the event. It includes human factors, without making any judgements.

• Root cause analysis - In this phase, ‘Why?” questions are asked, evidence and conditions (present and past) are analyzed, and a final statement of the root cause is formulated. This may include human causes and management system causes.

• Learning - Having discovered the root cause, this phase now determines what has been learned from the failure and what must be done to prevent a recurrence.

Corrosion Failure Analysis

After any corrosion-related failure it is extremely important to initiate a failure analysis process that will determine the root cause. The following guidelines for a corrosion analysis are paraphrased from the ASTM “Standard Guidelines for Corrosion-Related Failure Analysis”

Organize the analysis

• Conduct the analysis as early as possible after the failure, using every effort to protect any physical evidence

• Have a written plan for the investigation, indicating ALL required tasks

• Assign tasks and, if necessary, contract third-party expertise

Record conditions at the failure site

• If possible, examine the failure site before cleaning or altering the site in any way. Observe physical arrangements, odors, colors, textures, and condition of adjacent structures

• Take photos and videos. Photos should have labels that indicate position, size, etc. Take photos before and after sampling

• Make sketches and drawings of pertinent details at the site

• Interview personnel who may have witnessed the failure. Time, sounds, sights, etc. may be helpful

Determine operating conditions at time of failure

• Refer to operating logs and data logged information regarding the equipment operation (flow, temperature, pressure, etc)

• Trend operating conditions leading up to the failure

• Note any abnormal operating parameters before the failure

• Compare the operating parameters and conditions of similar equipment that has not failed

• Examine corrosion monitoring instruments, corrosion coupons, etc

Review construction and operation history

• Obtain original design drawings and specifications

• Obtain information regarding material specifications, methods of manufacture, welding procedures, heat treatments, surface treatments, etc.

• Review any modifications done since original fabrication (when, why and how they were done)

• Review operating history, especially any anomalies from normal parameters; any changes in process fluid specs or raw material sources (eg. water)

• Review maintenance, cleaning, and repair history

• Review the corrosion prevention procedures and operation records of all equipment (cathodic protection, coupons, inhibitor injection, etc.)

Obtain samples

• Include sample locations and collection methods in the investigation plan

• Avoid contamination of corrosion product samples, including by moisture, cutting fluids, etc.

• Ensure sample containers won’t react with the sample

• Make metal samples as large as possible for adequate testing

• Use non-metallic sampling tools, if possible, to avoid unnecessary damage to surfaces

• Take samples of process fluids from the failure area

• Follow specific procedures if microbiological samples are required.

• Avoid damage to any fracture surfaces while sampling

Evaluate samples

• Analyze deposits, corrosion products using appropriate analytical techniques and equipment

• Analyze metallurgical condition and structure of metallic samples, including welds. Includes metallographic examination of cross-sections

• Document the location and orientation of each specimen. Using photos, drawings, or description. Label each specimen.

Examine failure locations

• At the site of the failure, examine for pits, cracks, crevices, and general attack

• Take measurements to document surface chemistry, pit depths, crack dimensions, metal loss, etc.

• Use the appropriate examination techniques, such as light microscopes, scanning electron microscopes, x-ray spectrometers, fractographics (for fractures), cross-section examination (for intergranular failures)

Assess the corrosion-related failure

When all the data has been collected, it must be fully reviewed and assessed. Conclusions must be reached, based on solid engineering principles and taking into account all reported information, operational history, and examination data.

• Assess all observations from the failure locations, operational information, failed material examinations, and expert opinions

• Assess all out-of-spec materials, process stream, operating conditions

• Assess unusual or unexpected corrosion products

• Assess the type, extent, and rate of corrosion. Compare and assess against expected and historical corrosion rates

• From all assessments, apply root cause analysis to determine the root cause. Also identify additional contributing factors

File a Corrosion Failure Report

A comprehensive failure report must be presented, detailing the entire investigation, assessments, and conclusions, including any recommendations to prevent recurrence. The report should include the following topics.

• Precise description of the corrosion related failure

• Operating conditions at the time of failure

• Historical information related to the failure

• Concise descriptions of all deposit and metal samples taken (include photographs and sketches)

• Evaluations conducted (descriptions of all tests done on samples and at the failure site)

• Results of evaluations

• Conclusions (root cause of corrosion and failure; contributing factors)

• Recommendations (suggested corrective actions, if any)

• References; personnel involved in investigation

• Disposition of samples and records

1. Define the following as they apply to corrosion:

a) Oxidation

b) Reduction

c) Redox

2. Using appropriate sketch or sketches, explain a corrosion cell.

3. With respect to galvanic corrosion, state:

a) three conditions that must exist in order for it to occur

b) four conditions that affect the rate at which it occurs

4. Explain how deposits contribute to:

a) concentration cell corrosion

b) pitting

5. Explain three different types of Stress Corrosion Cracking.

6. Explain three categories of atmospheric corrosion.

7. Explain six ways in which superheater tubes may be exposed to corrosion.

8. What is fuel ash corrosion and how is it recognized?

9. Explain how steam condensers susceptible to corrosion if they have:

a) copper tubes, and

b) stainless steel tubes

10. What forms of corrosion are most likely in heat exchangers?

11. Briefly explain six methods used to monitor corrosion.

12. What operational methods can be used to control corrosion? Briefly explain each method.

13. Describe the four main phases of a corrosion failure analysis.

Self-Assessment

Question 1

?Fe ? 2e^- ? Fe²⁺? is an example of:

Reduction

Oxidation

Plating

Passivation

Question 2

Copper has an electrical potential of +0.52 Volts. Aluminum has an electrical potential of -1.66 Volts. If connected in a galvanic cell,

the potential difference would equal 1.14 Volts

the aluminum would provide electrons to the copper

the copper would be anodic to the aluminum

the aluminum would be more noble than the copper.

Question 3

Impressed current cathodic protection systems:

have AC power connected to the anodes and cathodes

protect structures by making them anodic

have the negative terminal of the power supply attached to the protected structure

prevent stray current corrosion

Question 4

In a typical corrosion cell, with water as the electrolyte, _______________ is produced at the __________________ as a result of the galvanic corrosion process.

hydrogen, cathode

hydrogen, anode

hydroxide, anode

reactive metal ions, cathode

Question 5

The process whereby corrosive products concentrate adjacent to boiler metal, beneath a porous heating surface deposit, is called:

oxygen pitting

crevice corrosion

selective leaching

wick boiling