Stress Corrosion Cracking Of Pipeline Steels Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Stress Corrosion Cracking (SCC) has been considered as a vital threat to the safe in operation of gas or oil transmission pipelines. Because it may cause significant failures in pipeline, and these failures are usually low-probability, high-consequence events. To date, two forms of SCC penetration from the external surface of the pipeline have been recognized: one is the intergranular SCC, the other is a kind of transgranular SCC. The former one is also called ' high pH SCC ' and has been studied extensively. There are some well-accepted mechanisms developed for the high pH SCC, such as anodic dissolution and film rupture at crack tip. However, there are relatively limited understanding of the transgranular SCC, also known as ' near-neutral SCC '. In this work, the corrosion behaviour of pipeline steels in near-neutral solution has been studied using cyclic voltammetry based testing technique. The results showed strong evidence that the role of hydrogen was significant in near-neutral pH SCC. Moreover, Slow Strain Rate Testing (SSRT) technique were also applied to describe the microsttuctural effect in SCC.

1. Introduction

1.1 Background

Stress Corrosion Cracking (SCC) is a form of environmental assisted cracking. The environmental assisted cracking is known as the cracking in material that can be affected by the environment and the stress. Concretely, SCC can be defined as the interaction of a tensile stress and an aqueous environment acting on a susceptible metallic surface to initiate and propagate cracks [1]. The SCC is a delayed failure, it means the crack initials and propagates at a slow strain rate until the stress reaches the fracture strength. The sequence of events involved in the SCC process can be separated into three stages (shown in figure 1) [2]:

Crack initiation and 1 stage propagation

Stage 2 or steady-state crack propagation

Stage 3 crack propagation or final failure

Many researches have been conducted to develop mechanisms trying to explain the synergistic stress-environment interactions at the crack tip. In general, two basic mechanisms have been identified: anodic and cathodic mechanisms [2].

Fig.1 Schematic diagram of typical crack-propagation rate as a function of crack-tip stress-intensity behaviour illustrating the regions of stage 1, 2, and 3 crack propagation as well as identifying the plateau velocity and the threshold stress intensity [2].

Since the first observation of SCC on the external surface of a buried natural gas transmission pipeline in 1965, SCC of pipeline has occurred in several countries around the world and has been regarded as a major failure in some gas or oil transmission pipelines [3]. For example, a grave accident happened in Canada on 15 April 1996, which was caused by a SCC-induced rupture of the Trans Canada Pipeline locating at 10km south west of Winnipeg, a nearby house was totally damaged by the explosion and huge fireball [4]. Generally, buried pipelines are located within ever changing environmental conditions which may lead to a corrosive environment. Pipe coatings and cathodic protection are commonly used to fight against corrosion. However, SCC is still existing.

1.2 Types of stress corrosion cracking in pipeline steels

Due to the environment-dependence of SCC, the environmental conditions that are present during the crack propagation determines the type of SCC that can occur. So far, two types of SCC have been characterized, they are intergranular and transgranular SCC. Approximately, the pH value of the electrolyte will decide whether the cracking is intergranular or transgranular. Usually, high pH conditions cause intergranular SCC, also called classical SCC, while low pH conditions result in transgranular SCC with mixed modes around the crack tip. To date, a great deal of studies have been conducted focusing on the classical high pH intergranular SCC, and some possible mechanisms have been put forward. However, there is relatively limited researches on low pH SCC.

Intergranular stress corrosion cracking (IGSCC) of pipeline steels usually occurs in highly concentrated bicarbonate-carbonate () solution with high pH between 9-11. Fang et al [5] developed a well-accepted mechanism for the formation of high pH condition in pipeline steels, it can be described as: the buried pipelines make use of cathodic protection to prevent corrosion but, unfortunately, the cathodic current can be large enough to break the water into hydroxyl ions, thus increase the pH value. Then the high pH solution can absorb from the environment to form a complex bicarbonate-carbonate solution. Based on this high pH solution, more efforts have been spent focusing on the mechanisms of IGSCC. After the investigation of corrosion behaviour of X65 pipeline steel, Wang et al [4] provided a possible explanation, if the concentration of carbonate is high enough in the high pH solution to passivate the steel pipe surface, IGSCC can occur by anodic dissolution mechanism along the grain boundaries. Parkins [6] also developed a mechanism related to the rupture of oxide film at crack tip, as plastic strain ahead of the crack tip can stop the formation of protective oxide film around the crack tip, thus the propagation of crack can continue. This IGSCC can only occur in a very restricted range of electrochemical potentials (-550 mV vs a saturated calomel electrode [SCE] to -650 m as reported by Gonzalez-Rodriguez et al [7] ). Therefor, IGSCC may be reduced or avoided by controlling the electrochemical potentials and operating in lower temperature.

In contrast, transgranular stress corrosion cracking (TGSCC) can occur in pipelines under normal operating conditions, where the pH value of the electrolyte is much lower, usually about 5.5-7.5, so it is also called near-neutral pH SCC. This near-neutral electrolyte is generally characterized with anaerobic, dilute solutions. The formation of this near-neutral environment is related to pipe coating that is broken or disbonded from the external surface of the pipe, thus groundwater comes into contact with the surface, and resulting in the loss of cathodic protection. Roughly, the TGSCC has a a transgranular, quasi-cleavage crack morphology with very little branching [5]. Compared to classical IGSCC, there are relatively limited understanding on TGSCC.

In addition, Liu et al [8] recently published a paper about the investigation of corrosion behaviour of pipeline steel in acidic solution. In his report, an acidic soil with an average pH of 3.5-6.0, named ''Red Soil",was mentioned in several provinces in southeast China where some natural gas transmission pipelines were operating in those areas. The results of their experiment showed the process and mechanisms of SCC in acidic solution was mix-controlled by the anodic metal dissolution and ingress of hydrogen. In detail, anodic metal dissolution mechanism dominated the SCC behaviour at a relatively less negative electrochemical potential, and hydrogen began to take effect when the applied potential decreased more negatively. With the further negative shift of applied potential, the SCC of the steel follows completely a hydrogen-based mechanism, with a river-bed shaped brittle feature of the fracture surface.

1.3 The behaviour of near-neutral pH stress corrosion cracking

The underlying mechanism of near-neutral pH SCC, also called TGSCC, is still ambiguous. Recently, positive progress has been achieved. Cheng et al [9] showed strong evidence that hydrogen did involve in process of TGSCC. In their study, the near-neutral electrolyte, which has been described above in section 1.2, were always associated with high hydrogen permeation currents and sub-surface hydrogen concentrations. This observation indicates hydrogen evolution reaction does occur in TGSCC. Therefore, the role of hydrogen should be part of the TGSCC mechanism. Since the occurrence of hydrogen, the mechanisms for TGSCC will be much different from that of IGSCC, where hydrogen evolution reaction is impossible due to the high pH condition. Besides, Beavers' experimental result [10] mentioned the phenomenon that the crack growth rates in the heat-affected zone next to the weld line were some 30% higher than other regions. This indicates the microstructure or the residual stress in steel may influence the SCC process, the further understanding of this phenomenon may help improving the SCC resistance in the pipeline industry.

The purpose of present paper is to give a better understanding of the near-neutral pH SCC. The occurrence of hydrogen evolution reaction in TGSCC will be demonstrated and analysed using cyclic voltammetry measurement. Then slow strain rate testing (SSRT) technique will be applied to show the possible microstructural effect. Finally, A thermodynamic model will be introduced to calculate the crack growth rate, which can be contributed to predict the lifetime of the pipeline.

2. Experimental procedures

2.1 Investigation of the role of hydrogen: Cyclic voltammetry measurement

The working electrode used in this electrochemical test was made of a sheet of API (American Petroleum Institute) X-70 pipeline steel with chemical composition (wt%): C 0.06, Mn 1.44, Si 0.31, S 0.004, P 0.01, Ni 0.034, Cr 0.16, Mo 0.25, V 0.005, Cu 0.015, Ti 0.01, B 0.002 and Al 0.029. The metallurgical observation indicated the primary microstucture of X-70 steel was bainitic ferrite. Each specimen was ground with 600 grit emery paper on all faces. The edges were coated using a masking paint to prevent crevice corrosion between the epoxy mount and the electrode. Then all specimens were embedded in epoxy resin

manufactured by LECO leaving a working area of 0.155 . The working surface was subsequently polished with 3μm and 1μm diamond pastes, cleaned by distilled water and methanol. The test solution used was 0.01 M sodium bicarbonate solution that was typically used to simulate the dilute bicarbonate electrolyte trapped between the coating and the pipeline. The solution was made from analytic grade reagents (Fisher Scientific) and ultra-pure water (18 MΩ). All the tests were performed at ambient temperature (22). To reproduce the near-neutral pH environmental condition, the solution was purged with high purity (class 2.2) 5% balanced with gas for 2 h prior to and throughout the tests [11, 12].

The cyclic voltammogram (CV) measurements were performed on a three-electrode system through PINE bipotentiostat. X-70 steel was used as working electrode, saturated calomel electrode (SCE) was used as reference electrode and a coiled platinum wire was used as the counter electrode. All potentials data from the CV measurement were converted to values relative to the standard hydrogen electrode (SHE). Before the CV measurements, the specimen was cathodically polarized at -1.5 V for 3 min to remove the air-formed oxide film. The potential was scanned in the positive or negative direction between preset lower and upper switching potentials.

2.2 Investigation of microstructural effect: Polarization resistance measurement and Slow strain rate testing technique

The same API X-70 pipeline steel was also the testing material here. In order to obtain various microstructures and a range of hardness, various heat treatment were applied and listed below [13]:

The samples were cut from the X-70 steel skelp with average grain size about 5.8 μm. The microstructures were illustrated using metallographic microscope and listed in figure 2. The polarization resistance measurement was performed in a NS4 solution with chemical composition of (g/L) 0.11. 0.483, 0.137, 0.131 to simulate the near-neutral groundwater in Canada. During the test, the solution was bubbled with to ensure an anaerobic

Fig. 2 Typical microstructure of the

X-70 pipeline steel, (a) annealed, (b)

normalized, (c) quenched, (d) quenched

and tempered, (e) as-rolled [13]

condition and a pH value of 6.7. Before the test, the solution should be deaerated for at least 4 hours to remove the oxygen. The reason of the removal of oxygen would be discussed later. Polarization resistance measurement was then conducted using scanning potential range from -10mVto +10mV relative to open circuit potential condition, and the scanning rate was 0.125 mV/S. In addition, the test results were confirmed using electrochemical impedances spectroscopy (EIS).

Slow strain rate testiing (SSRT) technique was carried out to define the SCC resistance. The samples had a gauge length of 25 mm and a diameter of 3mm. The applied strain rate was . The SCC resistance was defined as the ratio of reduction in area measured in corrosive medium and air () [14].

3. Results and discussion

3.1 The role of hydrogen

The results of the CV measurement is plotted in figure 3 with different upper switching potentials at a potential sweep rate of 50mV/s. It is shown in the curves recorded, that the positive current keeps increasing with increased potential after a ' potential shoulder ', while a negative current peak is identified in the cathodic polarization curve. Approximately, the value of this negative potential current peak increases with increased upper limit potential.

Fig. 3 Cyclic voltammograms measured on X-70 steel with the different upper potential limits at a potential sweep rate of 50 mV/s in 0.01 M sodium bicarbonate solution. [11]

The result of CV measurement indicates that the mechanism of TGSCC should be much different to high pH IGSCC. In detaill, the anodic polarization behaviours of them are not the same. The polarization behaviour of steel in IGSCC was reported by Gu et al [15], which involved two-step oxidation: the pre-oxidation step about the formation of deposit layer and followed by the formation of . A stable oxide Fe film would form by further polarization. The propagation of crack could occur due to the rupture of this oxide film ahead of crack tip. In contrast, the polarization behaviour of TGSCC described by the CV measurement indicates that there is no formation of the stable oxide film on the surface of the steel in the near-neutral anaerobic environment. As shown in figure 3, the observation of the ' potential shoulder ' is related to the pre-oxidation step of the steel to form deposit layer, the electrochemical reaction is :

Then, the current density continues to increase with positively increased potential, i.e, the steel is still in active dissolution status, no passivation occurrs due to no stable oxide film forms. Moreover, the cathodic peaks marked in figure 3 can be attributed to the reduction of oxidation product formed before, and it could be summarized that the cathodic peak is more negative when the upper potential limit is more positive, as more oxidation product forms. So far, the analysis of CV measurement demonstrates that no oxide film formed on the surface of pipeline steel in TGSCC. Therefore, the film rupture mechanism becomes unsuitable for near-neutral pH SCC, and some mechanical factors, e.g. ingress of hydrogen might take effect in the process of TGSCC. Further CV measurement is conducted by plotting the cathodic Tafel slopes as a function of upper potential limits, as shown in figure 4. Typically, discharging of in a low pH solution has a value about -100 mV/decade, which is much higher than that marked in the figure (-450 - -500mV/decade). Possible explanation to this occurrence is that the hydrogen evolution reaction is dominated by reduction of water molecules in near-neutral pH condition, instead of reduction, because more cathodic over-potential is required in water reduction thus leads to a more negative Tafel slope [11].

Fig. 4 The cathodic Tafel slope as a function of upper potential limits [11]

To sum up, hydrogen acts as a critical role in near-neutral pH SCC, the reduction of water molecules plays a decisive role in the hydrogen evolution reaction. And the pre-oxidation step generates an unstable deposit layer and followed by either an electrochemical hydrogen recombination reaction or a hydrogen absorption reaction [12]. However, it should be emphasized that all these experiments were acting at anaerobic conditions, as the introduction of oxygen would generate a oxide film which is stable and can acte as inhibitor of the hydrogen evolution reaction.

3.2 Microstructural effect

The result of SSRT is shown in figure 5, it is obvious that the difference in microstructure could lead to different SCC resistance. Three groups with typical microstructure are highlighted in the figure. Generally speaking, the SCC resistance of specimens with same microstructure decreases with increasing yield strength.

Fig. 5 Relationship between SCC resistance and yield strength of X-70 steel [13]

The SCC data shows strong microstructure- dependent behaviour, it means that even the specimens have the same yield strength, the SCC resistance varies with different microstructure. Further discussion is developed based on the comparison between X70Q and the blue group. The SCC resistance of water-quenched specimen (X70 Q) is lower than as-rolled condition (blue group), although they all have a bainitic structure and water-quenching increases the yield strength. In addition, temper after water-quenching gives rise to great increase in SCC resistance with small drop in yield strength (See the data of X70 QT). Possible explanation to this phenomenon is focusing on the high micro-stresses induced by quenching due to excess carbon trapped interstitially [13]. In general, small cracks were reported to be more likely to initiate and propagate along the pearlite/ferrite boundaries [13], so the presence of pearlite-ferrite structure will imply poor SCC resistance. To sum up, pipeline industry prefers pipeline steel which has a fine-grained bainite+ferrite microstructure to that has a pearlite+ferrite microstructure due to better combination of SCC resistance and mechanical properties.

The result of polarization resistance measurement is also analysed in this paper, as shown in figure 6. The linear relationship of polarization resistance and SCC resistance indicates that, besides the ingress of hydrogen, anodic dissolution also contributes to the near-neutral pH SCC. Therefore, the SCC behaviour of pipeline steels in near-neutral solution can be identified as the result of combined action of stress, hydrogen, and anodic dissolution at the crack tip in near-neutral pH SCC, i.e. the TGSCC.

Fig. 6 Relationship between SCC resistance and polarization resistance. [13]

3.3 Thermodynamic model of TGSCC

In order to estimate the crack growth rate and to approximately predict the lifetime of the pipeline steels in near-neutral pH environment, a thermodynamic model of near-neutral pH SCC has been developed by Cheng [16]. The interactions of stress, hydrogen, and anodic dissolution at crack tip were fully illustrated in Cheng's model. The modelling starts from the electrochemical reactions, that occurs in TGSCC:

Anodic reaction: (1) , and

Cathodic reaction: (2)

stands for the hydrogen atoms adsorbed on the surface, and every two can combine into a : (3). Then these hydrogen can penetrate into the surface and transfer into absorbed hydrogen atoms : (4). Therefore, the whole electrode reaction can be described as:


In equation (5), x stands for the number of hydrogen atoms permeating into the steel. Now, the effect of stress will be taken into account. Yokobori et al [17] has published a model for the hydrogen movement at the crack tip as shown in figure 7, which indicated that the hydrogen atoms generated through cathodic reaction would diffuse toward the crack tip and the accumulate in the plastic zone, due to the stress concentration at the crack tip. Hence, the equation (5) should be sightly modified in the presence of stress , as followed:

, where y is also the number of hydrogen atoms but should be larger than x.

Fig. 7 Model for hydrogen diffusion and dissolution reaction at the crack-tip for a stressed steel specimen in deoxygenated, nearneutral pH solution [17]

Further modeling is to describe the free-energy change between equation (5) and (6), i.e. between unstressed and stressed steel. and , and , respectively represent the free-energy changes and electrochemical potentials of equation (5)and (6).



where G stands for the formation free-energy, n the number of electrons exchanged, F represents the Faraday constant. The free-energy change is used to calculate the anodic dissolution current density in reaction (5): (9)

The same as reaction (6): (10). Assuming the exchange current density remains constant, and the difference in free-energy change mainly results from the presence of stress, thus : (11), is the charge transfer coefficient. In general, where stands for the internal energy changes, S is the entropy, is strain energy density, and is the interaction energy between the lattice strain induced by hydrogen atoms and external stress field. As H=0 and the presence of stress, can be simplified as : (12) where W stands for molar weight, is principal stress, is density, E is Young's modulus. So the equation (11) can be re-written as :


where is the stress factor. Similar derivation is applied to all the factors and combine these factors together giving :


where is the effect of concentration difference of hydrogen atoms between unstressed and stressed samples. is the effect of hydrogen on the dissolution rate without stress. is the effect of stress on dissolution rate without hydrogen. is the effect of combined action of stress and hydrogen on the anodic dissolution. Finally, according to the crack growth rate in anodic dissolution mechanism : , the crack growth rate under the influence of stress and hydrogen can be described as In summary, this thermodynamic model is rough and more studies are required to give better understanding of this near-neutral pH SCC. So far, it is still difficult to precisely predicting the lifetime of these pipelines, so periodic detection is necessary to prevent SCC-induced accident.

4. Conclusion

The stress corrosion cracking of pipeline steels is introduced in this work, and more effort is spent to investigating the near-neutral pH SCC. Although the underlying mechanism is still incomplete, it is demonstrated that hydrogen plays a critical role in near-neutral SCC and accompanying with the anodic dissolution mechanism. However, the prediction of the lifetime of the pipelines is still difficult.