Hydrogen-Related Stress Corrosion Cracking Stress corrosion cracking (SCC) may develop due to either highly localized anodic

dissolution processes or the specific effects of hydrogen. The latter case usually involves a localization of absorbed hydrogen atoms at specific trap sites, such as dislocations, grain boundaries, and microvoids within the metal. These trapping effects cause a deviation from the simple diffusion behavior, not only by immobilization, but also because interstitial sites and defect sites are differently affected when hosting hydrogen atoms.

(i) The Trapping of Hydrogen in Metals A description of the trapping process may be based on the chemical
potentials of the hydrogen absorbed in the metal.88 In the general case, the chemical potential of interstitial or diffusible hydrogen may be described by the equation where (H,diff) = chemical potential of diffusible absorbed hydrogen, (H,diff) = standard chemical potential of diffusible absorbed hydrogen, {H,diff} = activity of diffusible absorbed hydrogen. The chemical potential of trapped hydrogen may similarly be expressed by Under equilibrium conditions, the chemical potential of diffusible hydrogen equals the chemical potential of trapped hydrogen, i.e.,
Automated Methods of Corrosion Measurement 307

Equation (36) states that the activity of hydrogen at trap sites increases with decreasing chemical potential of the trap; hence a trap involves a site where the chemical potential is lower than the chemical potential associated with an interstitial site. Consequently, the concentration of hydrogen in a trap site is higher than the concentration found in the surroundings of the trap. Further mathematical treatments and classifications of traps have been presented in several papers.8793 The effect of traps on the rising transient hydrogen permeation behavior (Fig. 26) is an increase in the observed breakthrough time (due to the time needed to saturate the traps) and a steepening of the rising curve (due to a larger hydrogen concentration gradient at the exit surface when the traps are saturated and until steady state is attained).88 (ii) The Threshold Hydrogen Concentration for Stress Corrosion Cracking The electrochemical hydrogen permeation technique has been used in efforts to establish threshold hydrogen concentrations in steel below which no cracking occurs. The threshold concentration depends largely on the type of failure under investigation, the chemical and physical properties of the steel, and the magnitude of applied and residual stress. Hydrogen-induced cracking (HIC) may develop in the absence of external stresses and is caused by the recombination of absorbed hydrogen atoms at, for example, microvoids present in the steel. Very high pressures may build up as a result of the formation of hydrogen gas; if the pressure sets up great enough stress, crack initiation will result. The threshold hydrogen concentration in this case may be assessed simply by establishing a suitably large concentration gradient in Fig. 25) over the membrane. The subsurface hydrogen concentration just beneath the entry surface has to exceed the threshold concentration. This can be achieved

by adding a recombination poison to the catholyte. Hydrogen sulfide is known to be such a poison, which inhibits the Tafel recombination reaction, thereby increasing and according to Eq. (28), also Under steady-state conditions, the concentration of interstitially absorbed hydrogen can be established at any depth of the membrane (dotted line in Fig. 25). By subsequent metallographic examination of the mem308
Gregers Bech-Nielsen et al.

brane, the maximum depth where cracks can be found is noted and compared with the hydrogen concentration profile found in the permeation experiment. A more elegant variant was demonstrated by Coudreuse and Charles.94 By repeated permeation experiments, these authors managed to distinguish among diffusible, reversibly, and irreversibly trapped hydrogen and were thereby able to describe the threshold concentration as the sum of these three contributions. In cases where the propagation of hydrogen-related cracks depends on externally applied stresses, the SCC test may be performed using fracture mechanical test specimens. In this case, the direct measurement of the hydrogen concentration in the specimen is difficult, since the electrochemical permeation technique involves the creation of a hydrogen concentration gradient in the membrane. For this reason, the test specimen itself cannot act as the membrane due to the need for a uniform concentration of diffusible hydrogen during the test. Alternatively, as demonstrated by Robinson and Kilgallon,95 the hydrogen concentration in the test specimen may be assessed by placing a hydrogen permeation cell in the same environment as that used for the fracture mechanical test. By charging the membrane of the hydrogen permeation cell in the same manner as the test specimen, a parallel permeation experiment can be performed, which can be used to assess the hydrogen concentration present in the actual test specimen. This procedure yields a description of the critical stress intensity factor for stress corrosion cracking as a function of the concentration of hydrogen absorbed by the test specimen. Robinson and Kilgallon offer the following empirical relation found by constant displacement tests in various environments using double-cantilever beam (DCB) specimens prepared from a quenched and tempered high-strength, low-alloy offshore steel95 For comparison, the following relationship has been found for an X70 pipeline steel (slow strain rate test using compact tension specimens with the precrack located in the heat-affected zone of a weld)96
Automated Methods of Corrosion Measurement 309

The oil and gas sector in particular has an interest in such relationships, since offshore steel structures are cathodically protected and may be exposed in biologically active environments that involve activity of sulfate-reducing bacteria. These anaerobic bacteria produce sulfide, which in turn may enhance hydrogen uptake by the steel. There are similar concerns when pipelines are buried in anaerobic soil. The relations (37) and (38) can be compared with stress intensity factors actually present at flaws in the steel under operating conditions; the actual hydrogen concentration should be assessed as well, for example, by using hydrogen permeation units designed for this purpose.

Stress Corrosion Cracking

Stress corrosion cracking (SCC) occurs at points of stress. Usually, the metal or alloy is virtually free of corrosion over most of its surface, yet fine cracks penetrate through the surface at points of stress. The conditions necessary for SCC are: 1. A suitable environment (chemicals capable of causing SCC in carbon steel and low-alloy carbon steels) 2. A tensile stress, either residual or operational 3. Appropriate temperature and pH values One advantage of carbon steel is that SCC can be prevented by relieving stress after fabrication. Chemical species that induce SCC in carbon and low-alloy carbon steels, even at low concentrations include: hydroxides, gaseous hydrogen, gaseous chlorine, hydrogen chloride, hydrogen bromide, aqueous nitrate solutions, hydrogen sulfide gas, MnS and MnSe inclusions in the alloy, As, Sb, and Bi ions in aqueous solution, carbon monoxidecarbon dioxidewater gas mixtures. Many of these chemical systems will crack steel at room temperatures. Another chemical that causes SCC in steels is anhydrous ammonia. Alloys affected include carbon steel in storage tanks and ASTM A517 quenched and tempered steel in motor vehicle cargo tanks. Various grades of A517 steel contain small amounts of Cr, Ni, Mo, B, V, Ti, Zr, and Cu. This cracking can be alleviated by adding 0.2% water to the ammonia, eliminating air contamination from ammonia systems, and stress-relieving tanks or fabricating with hot formed or stress-relieved heads.

3.3 Sulfide Stress Cracking

Many corrosion processes produce hydrogen ions that combine with electrons from the base metal to form hydrogen atoms. Two such formed hydrogen atoms may combine to form a hydrogen molecule. The majority of such molecules will form hydrogen bubbles and float away harmlessly. However, a percentage of the hydrogen atoms will diffuse into the base metal and embrittle the crystalline structure. SSC will occur when a critical concentration of hydrogen is reached while a tensile stress exceeding a threshold level is present. Although H2S does not actively participate in the SSC reaction, sulfides act to promote the entry of the hydrogen atoms into the base metal.
78 Fundamentals of Metallic Corrosion: Atmospheric and Media Corrosion of Metals

The susceptibility of carbon steels to SSC is directly related to their strength or hardness levels. As carbon of low-alloy carbon steel is heattreated to progressively higher levels of hardness, the time to failure decreases rapidly for a given stress level. Temperature is also a factor. The most severe SSC takes place in the temperature range of 201208F (K7 to 498C). Above 1208F (498C), the diffusion rate of hydrogen is so fast that the hydrogen passes through the material so quickly that the critical concentration is not reached. Below 208F (K78C), the diffusion rate is so slow that the critical concentration is never reached. By carefully monitoring the processing of carbon and low-alloy carbon steels, and keeping the hardness level below 22 HRC, these steels will have acceptable resistance to SCC.

Hydrogen Damage
The body-centered crystal structures of carbon and low-alloy steels are susceptible to four types of hydrogen damage, two of which are lowtemperature

processes and two are high-temperature processes: Low-temperature Hydrogen blistering Hydrogen embrittlement High-temperature Decarburization Hydrogen attack The diffusion of hydrogen through steels to harm mechanical properties involves atomic or nascent hydrogen because molecular hydrogen cannot
Corrosion of Carbon and Low-Alloy Steels 79

diffuse through metals. Common sources of atomic hydrogen include corrosion (including the acid pickling of steel), misapplied cathodic protection, high temperature, moist atmospheres, electroplating, and welding. 3.5.1 Hydrogen Blistering During some acid services, such as acid pickling of steels, hydrogen atoms may penetrate the crystal lattice and collect in fissures or cavities in the steel. These atoms then combine into hydrogen gas molecules, eventually reaching pressures of several hundred thousand atmospheres and forming blisters on the steels surface. In petroleum process streams, this problem is promoted by so-called hydrogen evolution poisons such as sulfides, arsenic compounds, cyanides, and phosphorus-containing ions. In closed systems like pickling operations, chemical inhibitors are added to the acid to reduce the hydrogen penetration. 3.5.2 Hydrogen Embrittlement Another harmful effect of hydrogen penetration of steel is embrittlement, which is a more complicated metallurgical effect, possibly involving the interaction of hydrogen atoms with the tip of an advancing crack. For low-alloy steels, the alloys are most susceptible in their highest strength levels. Alloys containing nickel or molybdenum are less susceptible. If hydrogen is initially present in a steel, for example from electroplating, the hydrogen can be baked out. In fact, this embrittlement decreases with increasing service temperature, especially above 1508F (658C). Generally, hydrogen embrittlement is not usually a problem in steels with yield strengths below about 1000 MPa (150 ksi), but if hydrofluoric acid or hydrogen sulfide is present, the yield strength must be below 550 MPa (80 ksi) for good resistance. Welding conditions should be dry and lowhydrogen filler metal should be used to minimize hydrogen embrittlement. 3.5.3 Decarburization The hardness and strength of a steel depends on its carbon content. A loss of carbon (decarburization) lowers the tensile strength of steels and can be caused by moist hydrogen at high pressures and temperatures. Figure 3.6 shows the Nelson diagram that depicts the limit of service conditions for carbon and alloy steels in hydrogen services. 3.5.4 Hydrogen Attack High-temperature hydrogen attack refers to a reaction between hydrogen and a component of the alloy. For example, in steels, hydrogen reacts with iron carbide at high temperatures to form methane gas according to the
80 Fundamentals of Metallic Corrosion: Atmospheric and Media Corrosion of Metals

following reaction: CFe C 4HFe Z CH4: Because methane cannot diffuse out of steel, it accumulates and causes fissuring and blistering, thereby decreasing alloy strength and ductility. Alloy steels containing chromium and molybdenum are beneficial in such services because the carbides formed by the alloying elements are more stable than iron carbide and therefore resist hydrogen attack. It is noteworthy that water vapor and carbon dioxide at high temperatures can also decarburize steel.

3.6 Corrosion Fatigue

As the name implies, corrosion fatigue is affected by both the severity of corrosive conditions and mechanical, cyclical stress factors. Stress raisers such as notches, holes, weld defects, or corrosion pits can initiate fatigue cracks and a corrosive environment can reduce crack initiation time. For many materials, the stress range required to cause fatigue failure diminishes progressively with increasing time and with the number of cycles of applied stress. The first mention of the corrosion resistance of various alloys that had been formulated in which chromiumiron alloys were prepared appeared in 1820 in a published report by Stodart and M. Faraday. However, the maximum chromium content was below that required for passivity. Consequently, they narrowly missed discovering stainless steels. In 1821 in France, Berthier found that iron alloyed with large amounts of chromium was more resistant to acids than unalloyed iron. However, the alloys were high in carbon, brittle, and had no value as structural materials. During subsequent years, a variety of chromiumiron alloys were developed by several investigators who took advantage of the high strength and high hardness imparted by chromium. The inherent corrosion resistance of the alloys was not observed, primarily because the accompanying high carbon content impaired the corrosion properties. In 1904, Guillot of France produced low-carbonchromium alloys overlapping the passive composition range. Although he studied the metallurgical structure and mechanical properties of the chromiumiron alloys and the chromiumironnickel alloys, he did not recognize the outstanding property of passivity. The property of passivity, starting at a minimum of 12% chromium, was first described by Monnartz of Germany in 1908. He published a detailed account of the chemical properties of the chromiumiron alloys in 1911. Meanwhile, H. Brearly in England was attempting to develop iron-based chromium alloys to prevent erosion and fouling in rifle barrels. During his experiments, he noted their resistance to etching for metallographic examination. He observed that the 12% chromiumiron alloys did not etch with the usual nitric acid and other etching reagents, and that they did not rust over long periods of exposure to the atmosphere. He called these ferritic alloys stainless steel and recognized their possible use as cutlery materials. Simultaneously, Benno Strauss and Edward Maurer in Germany were investigating ironchromiumnickel compositions. They observed that the austenitic alloys containing 8% nickel were resistant to acid fumes, but the alloys were impractical because they cracked during any metalworking operation. However, Strauss restored ductility when he developed an
109

annealing heat treatment, followed by a water quench that dissolved the chromium carbides. Based on his experiments, Monnarty postulated that the passivity in stainless steels was caused by an invisible oxide film. This theory was not universally accepted. It was not until 1930 that his theory was proven electrochemically by H.H. Uhlig at the Massachusetts Institute of Technology. Stainless steels and similar chromium-rich alloys are characterized by their passivity. The general concept of passivity involves a base metal exhibiting the corrosion behavior of a more noble metal or alloy. For example, a piece of bare steel immersed in a copper sulfate solution develops a flash plating of metallic copper by a process known as cementation. If the bare steel is first immersed in a strong nitric acid solution, an invisible protective oxide layer is formed that prevents cementation and the steel is

said to have been passivated. Passivation of ferrous alloys containing more than 10.5% chromium is by the chromium addition. There are many stainless steel compositions, all of which have their own set of physical, mechanical, and corrosion resistance properties. In this chapter, we discuss the various types of corrosion to which stainless steels may be susceptible, keeping in mind that all compositions are not affected to the same degree, if at all. Specific corrosion problems and resistance of specific compositions will be discussed in succeeding chapters. Stainless steels are alloys of iron to which a minimum of 11% chromium has been added to provide a passive film to resist rusting when the material is exposed to weather. This film is self-forming and self-healing in environments where stainless steel is resistant. As more chromium is added to the alloy, improved corrosion resistance results. Consequently, there are stainless steels with chromium contents of 15, 17, and 20%, and even higher. Chromium provides resistance to oxidizing environments such as nitric acid and also provides resistance to pitting and crevice attack. Other alloying ingredients are added to further improve the corrosion resistance and mechanical strength. Molybdenum is extremely effective in improving pitting and crevice corrosion resistance. By the addition of copper, improved resistance to general corrosion in sulfuric acid is obtained. This will also strengthen some precipitationhardening grades. In sufficient amounts, however, copper will reduce the pitting resistance of some alloys. The addition of nickel will provide improved resistance in reducing environments and to stress corrosion cracking (SCC). Nitrogen can also be added to improve corrosion resistance to pitting and crevice attack and to improve strength. Niobium and titanium are added to stabilize carbon. They form carbides and reduce the amount of carbon available to form chromium carbide that can be deleterious to corrosion resistance. It is because of all of these alloying possibilities that so many types of stainless steel exist. It should also be kept in mind that the more alloying
110 Fundamentals of Metallic Corrosion: Atmospheric and Media Corrosion of Metals

elements used in the formulation, the greater will be the cost. Consequently, it is prudent to select the specific stainless steel composition that will meet the needs of the application. For example, it is not necessary to provide additional pitting resistance if the environment of the application does not promote pitting.

Corrosion Fatigue
Fatigue is a phenomenon that leads to cracking of a metal under repeated or fluctuating stresses at values below the tensile strength. Such fractures are progressive in nature, growing with the time exposure to the stress fluctuation. In corrosion fatigue, the fracture occurs sooner because of the combined effect of cyclic loading and corrosion at lower stress levels. Corrosion fatigue is difficult to predict because it varies with both alloy and environment; there is no specific environment that affects a particular alloy or alloy system. Occasional failures are experienced in applications in which austenitic stainless steel has replaced, for example, a carbon steel nozzle or pipe section. When a lighter gage of stainless steel replaces a heavier-walled steel section, either fatigue or corrosion fatigue may ensue because of vibrations that did not damage the carbon steel.