Investigation of Hydrogen Embrittlement Susceptibility and Fracture Toughness Drop After in Situ Hydrogen Cathodic Charging for an X65 Pipeline Steel

micromachines

Article Investigation of Hydrogen Embrittlement Susceptibility and Fracture Toughness Drop after in situ Hydrogen Cathodic Charging for an X65 Pipeline Steel

Helen P. Kyriakopoulou 1,*, Panagiotis Karmiris-Obrata ´nski 2,3 , Athanasios S. Tazedakis 4, Nikoalos M. Daniolos 4, Efthymios C. Dourdounis 4, Dimitrios E. Manolakos 3 1, and Dimitrios Pantelis † 1 Shipbuilding Technology Laboratory, School of Naval Architecture and Marine Engineering, National Technical University of Athens, 15780 Athens, Greece 2 School of Mechanical Engineering and Robotics, AGH University of Science and Technology, Mickiewicza 30, 30-059 Cracow, Poland; [email protected] 3 School of Mechanical Engineering, National Technical University of Athens, Heroon Polytechniou 9, 15780 Athens, Greece; [email protected] 4 Corinth Pipeworks S.A., Industrial Area of Voiotia, Thisvi, 32010 Domvraina, Greece; [email protected] (A.S.T.); [email protected] (N.M.D.); [email protected] (E.C.D.) * Correspondence: [email protected] This author passed away. † Received: 28 March 2020; Accepted: 18 April 2020; Published: 20 April 2020

Abstract: The present research focuses on the investigation of an in situ hydrogen charging effect during Crack Tip Opening Displacement testing (CTOD) on the fracture toughness properties of X65 pipeline steel. This grade of steel belongs to the broader category of High Strength Low Alloy Steels (HSLA), and its microstructure consists of equiaxed ferritic and bainitic grains with a low volume fraction of degenerated pearlite islands. The studied X65 steel specimens were extracted from pipes with 19.15 mm wall thickness. The fracture toughness parameters were determined after imposing the fatigue pre-cracked specimens on air, on a specific electrolytic cell under a slow strain rate bending loading (according to ASTM G147-98, BS7448, and ISO12135 standards). Concerning the results of this study, in the first phase the hydrogen cations’ penetration depth, the diffusion coefficient of molecular and atomic hydrogen, and the surficial density of blisters were determined. Next, the characteristic parameters related to fracture toughness (such as J, KQ, CTODel, CTODpl) were calculated by the aid of the Force-Crack Mouth Open Displacement curves and the relevant analytical equations.

Keywords: X65 pipeline steel; hydrogen embrittlement; in situ hydrogen cathodic charging; crack tip opening displacement test (CTOD); crack mouth open displacement (CMOD); fracture analysis

1. Introduction Over the last 15 years, the main targeting of the petroleum and hydrocarbon industry has been to encounter the phenomenon of hydrogen embrittlement and hydrogen induced toughness drop of the oil carrying pipeline steels during their operational cycle [1–5] The presence of molecular H2S in the internal environment of pipeline steels, attributed to the oil decreased quality, results in the diﬀusion and entrapment of atomic hydrogen into metal structure, which leads to the reinforcement of Hydrogen Induced Cracking (HIC) phenomenon, blistering eﬀect and development of brittle hydride

Micromachines 2020, 11, 430; doi:10.3390/mi11040430 www.mdpi.com/journal/micromachines Micromachines 2020, 11, 430 2 of 20

phases (FeH3)[5–9].The surface micro-cracking networks have a characteristic stepwise morphology, whereas blisters depending on their growth conditions may occur locally with a dome morphology or in an extended area with elongated and continuous morphology (parallel to the thermoplastic deformation axis) [9–12]. The blisters grow on lattice defects, where the hydrogen cations accumulate continuously. These microstructural sites are separated into irreversible or reversible traps, according to the energy needed thermodynamically for the hydrogen release process. Irreversible traps include vacancies, high angle grain boundaries, dislocation forests and non-metallic inclusions, while reversible traps include interstitial lattice sites, internal or external stacking faults, low angle grain boundaries and micro-twin boundaries [12–14]. Finally, the development of FeH3 hydrides was detected to occur mainly at energy upgraded locations such as the interfaces between non-metallic inclusions and matrix as well as the triple junctions. In particular, three micro mechanisms were proposed in the scientific field of physical metallurgy to justify the reduction of mechanical properties on pipeline steels due to the effect of atomic or molecular hydrogen, reported as Internal Pressure theory (IP), Hydrogen Enhanced Local Plasticity theory (HELP) and Hydrogen Induced Decohesion theory (HEDE). The first relates to the development of elevated internal pressures due to hydrogen adsorption on the micro-cavities of the microstructure, the second to the local increase of the microplasticity resulting in the excess of yield point and the third to the reduction of the interconnectivity between grain boundaries and boundaries of different phases and microstructural components [15–21]. The hydrogen embrittlement phenomenon leads to the deterioration of the chemico-mechanical integrity of the main metallurgical systems in the contemporary technological field. Its destructive effect was observed in alloy systems with Face Centered Cubic (FCC), Body Centered Cubic (BCC), and Hexagonal Close Packed (HCP) structures and is particularly pronounced in high strength low alloyed steels. The underlying mechanisms which contribute to this phenomenon are still not fully understood, despite the extensive research efforts in the oil and hydrogen gas pipeline transportation industry. For the pipeline steels the proposed embrittling micromechanisms mainly concern the formation of hydride phases, the reduction of the interfacial cohesive strength within the crystallographic structure due to the mitigation effect of hydrogen atoms (HEDE) and the hydrogen enhanced localized plasticity phenomenon (HELP). Hydride formation is favored by the presence of residual stress fields and is very common in metallurgical systems that exhibit a thermodynamic and kinematic tendency for hydride formation. During the evolution of Hydrogen Enhanced Decohesion mechanism (HEDE), the hydrogen cations accumulate within the high and low energy trapping centers of the crystal structure, reducing the cohesive energy strength within the grains of the matrix. This results in a severe reduction of the mechanical energy required to decode the crystallographic structure along certain crystallographic planes. As the more prone regions are considered the grain boundaries with high angle misorientation distribution factor and the interphase boundaries with elevated dislocation densities. The ductile surface of many metallurgical systems that had failed due to the penetration effect of hydrogen cations led to the support of a newly developed mechanism, named Hydrogen Enhanced Localized Plasticity (HELP). According to this mechanism, the propagation rate of dislocation networks and the dislocation core energy increases in areas with elevated concentrations of hydrogen cations, such as the micro-cracked and notched regions. Consequently, a significant increase in local plasticity is observed, which results in a high surficial density of shear slip bands and Ludders’ bands causing the overall failure of the structure. In recent published work M. B. Djukic, et al. investigated the hydrogen embrittlement susceptibility of a low carbon structural steel (grade 20-St.20 (GOST 1050-88)). This grade of steel is used for the construction of boiler tubes of fossil fuel power plants and is subjected to high temperature hydrogen attack and degradation of the mechanical properties due to hydrogen embrittlement process. Based on the experimental results the contribution of hydrogen enhanced decohesion (HEDE) and hydrogen enhanced localized plasticity (HELP) mechanisms is detected, depending on the local concentration of hydrogen cations, measured by means of Thermal Desorption Micromachines 2020, 11, 430 3 of 20

Spectroscopy. The coexistence of the main embrittlement mechanisms (HELP+HEDE) is confirmed by a mixed quasi-cleavage mode of fracture with the presence of brittle transgranular fracture surfaces into the ferrite islands (activity of HEDE mechanism) and microvoid coalescence fracture morphologies into the pearlitic microconstituents (activity of HELP mechanism) [22]. Rajwinder Singh et al. investigated the hydrogen influenced crack initiation and propagation under tensile and low cycle fatigue loadings in RPV steel. It is detected that under low cycle fatigue loading of the uncharged specimen, multiple microcrack networks initiated at the fractured surfaces from M3C carbides distributed along the bainitic ferrite lath boundaries. The microcracks initiated also from the non-coherent interphases between angular Al2O3-SiO2 inclusions and matrix material. In contrast for the specimens subjected to low cycle fatigue under hydrogen cathodic charging process the edges of cracks formed on the surface of the specimens during the first stage of LCF loading and propagated throughout the cross section with transgranular and intergranular fracture mode due to the evolution of HEDE mechanism. For the uncharged specimens a slip driven /transgranular fatigue crack propagation mode was detected counter to the hydrogen charged specimens, where the propagation of fracture was developed by the contribution of an intergranularly and transgranularly mode. Due to the hydrogen-impeded localized plasticity no striations are observed on the fractured surfaces of hydrogen charged specimens. In conclusion, different hydrogen embrittlement mechanisms were activated throughout the thickness of the specimens depending on the local hydrogen concentration level and the loading conditions [23]. M.B. Djukic, et al. investigated the hydrogen damage phenomenon in steels and proposed a hydrogen embrittlement model. The coexistence of HELP and HEDE embrittlement mechanisms was detected by the mixed mode of fracture by quasi-cleavage regions (QC), transgranular fracture features in ferrite islands (TG) and the evolution of microvoid coalescence phenomenon (MVC). For the lower hydrogen concentrations, the dominant fracture mechanism is considered to be HELP, due to the increased volume fraction of ductile MVC fracture features in the “quasi-cleavage” fracture surfaces of ferritic islands. For critical values of hydrogen concentration into the metal structure an expanded manifestation of HEDE mechanism is determined by sudden transition between ductile and brittle fracture modes. The restriction of HELP mechanism leads to the ductility drop and enhances the activity of HEDE mechanism by the severe reduction of cohesive strength between the interfacial grain boundaries. Finally, fracture toughness and microhardness values can be successfully correlated with the hydrogen cations’ concentration factor in microstructural sites. The fracture toughness of steel decreases with an increase of hydrogen concentration factor, regardless the type of the existing embrittlement mechanism [17–21,24]. Specifically for X52, X60, X65, X70, X80, and X100 pipeline steels it was found that during their operation in sour environment a significant decrease in ductility, maximum tensile strength, and fracture toughness parameters (KQ,KIC,Jel,Jpl, CTODel, CTODpl) takes place in relation to the adsorbed concentration of atomic or molecular hydrogen. Experimentally, the above phenomenon is studied either by ex situ or in situ hydrogen cathodic charging process during the Crack Tip Opening Displacement (CTOD) test, with selected parameters simulating the actual operating conditions. Particularly important has been proving the electrolyte composition (hydrogen content), the applied current density and the duration of the cathodic charging process. Specifically for the in situ hydrogen cathodic charging method, a higher decrease in toughness is observed primarily with increasing the applied current density, secondary with increasing the hydrogen content in the electrolytic environment and third with increasing the hydrogen cathodic charging duration. The volume fraction, the morphology and the distribution of non-metallic inclusions, carbides, and Martensite-Austenite islands (MAs), the dislocation density as well as the existence of a ferritic-pearlitic or a ferritic-bainitic or a mixed ferritic-pearlitic-bainitic microstructure in pipeline steels are the key factors that influence the Hydrogen Induced Cracking resistance performance. Chatzidouros et al. studied the toughness drop after the in situ hydrogen cathodic charging process during CTOD testing for API5L X52, X65 and X70 steels, with a ferritic-pearlitic, ferritic-bainitic and ferritic-mixed pearlitic-bainitic microstructure respectively [25,26]. The X65 pipeline steel was proved to be more prone Micromachines 2020, 11, 430 4 of 20

Micromachines 2020, 11, x 4 of 21 to the reduction of Jo and CTOD toughness parameters due to the adsorbed hydrogen concentration in the lattice structure than the other two grades of pipeline steels. For the maximum value of the applied was detected, which was calculated equal to 77.7%. For the less prone microstructures of X52 and X70 current density field (10 mA/cm2) a severe decrease of J toughness parameter was detected, which pipeline steels to the hydrogen embrittlement phenomenon,o it was observed that the ferritic islands was calculated equal to 77.7%. For the less prone microstructures of X52 and X70 pipeline steels to constituted the hydrogen diffusion channels and the ferrite-pearlite or ferrite-bainite interfaces the hydrogen embrittlement phenomenon, it was observed that the ferritic islands constituted the conducted as the main hydrogen trapping centers. hydrogen diffusion channels and the ferrite-pearlite or ferrite-bainite interfaces conducted as the main Through this study, interesting results were provided concerning the hydrogen induced hydrogen trapping centers. toughness decrease, the micromechanisms which contribute to failure, the morphology and surficial Through this study, interesting results were provided concerning the hydrogen induced toughness density of blisters and microcracks, the microhardness increment at pipeline steel surfaces, as far the decrease, the micromechanisms which contribute to failure, the morphology and surficial density of diffusion depth and the diffusion coefficient of hydrogen cations into the structure correlated with blisters and microcracks, the microhardness increment at pipeline steel surfaces, as far the diffusion the applied current density during the in situ hydrogen cathodic charging process. The novelty of depth and the diffusion coefficient of hydrogen cations into the structure correlated with the applied this paper is attributed to the use of an environmental friendly electrolyte consisted of 30 g/L NaCl current density during the in situ hydrogen cathodic charging process. The novelty of this paper and 3 g/L NH4SCN (which satisfies green chemistry’s requirements) for elevated values of the applied is attributed to the use of an environmental friendly electrolyte consisted of 30 g/L NaCl and 3 g/L current density field (10 and 20 mA/cm2) during the in situ hydrogen cathodic charging process and NH SCN (which satisfies green chemistry’s requirements) for elevated values of the applied current slow4 strain rate bending. The applied current densities exceeded the above of other researchers for density field (10 and 20 mA/cm2) during the in situ hydrogen cathodic charging process and slow X65 pipeline steel, which haven’ t been higher than 10 mA/cm2.These experiments took place in order strain rate bending. The applied current densities exceeded the above of other researchers for X65 to investigate the existence or not of a thermodynamic or electrokinetic potential threshold which pipeline steel, which haven’ t been higher than 10 mA/cm2. These experiments took place in order to leads to the hydrogen absorbed diminishing effect independently of the increment of the applied investigate the existence or not of a thermodynamic or electrokinetic potential threshold which leads to current density. According to the obtained experimental results, by increasing the applied current the hydrogen absorbed diminishing effect independently of the increment of the applied current density. density field during in situ hydrogen cathodic polarization process the fracture toughness was According to the obtained experimental results, by increasing the applied current density field during decreasing continuously. in situ hydrogen cathodic polarization process the fracture toughness was decreasing continuously. 2. Experimental Section 2. Experimental Section The dimensions of the crack tip opening displacement test (CTOD) specimen are shown in The dimensions of the crack tip opening displacement test (CTOD) specimen are shown in Figure 1. The Crack Tip Opening Displacement test took place in two separate stages. In the first Figure1. The Crack Tip Opening Displacement test took place in two separate stages. In the first stage, the controlled crack propagation rate during fatigue loading in air environment was realized stage, the controlled crack propagation rate during fatigue loading in air environment was realized (fatigue pre-crack) for the X65 pre-notched specimens. At this stage, the crack length was monitored (fatigue pre-crack) for the X65 pre-notched specimens. At this stage, the crack length was monitored using an Alternative Pulse Current Device (ACPD). As the fatigue crack propagated, the continuous using an Alternative Pulse Current Device (ACPD). As the fatigue crack propagated, the continuous variation of the electric voltage field and the resistance drop were measured. By means of a calibrated variation of the electric voltage field and the resistance drop were measured. By means of a calibrated transfer function, the gradient of the electric voltage field and the resistance drop were transformed transfer function, the gradient of the electric voltage field and the resistance drop were transformed into fatigue crack propagation length (Figure 2). into fatigue crack propagation length (Figure2). The pre-crack fatigue process of the specimens was carried out with a frequency of 10 Hz, while The pre-crack fatigue process of the specimens was carried out with a frequency of 10 Hz, while the parameters of the maximum-minimum load and the cycles are mentioned in Table 1. The different the parameters of the maximum-minimum load and the cycles are mentioned in Table1. The di fferent stages during the fatigue precrack process were developed in order to avoid the undesirable effects stages during the fatigue precrack process were developed in order to avoid the undesirable effects of of plasticity and crack blunting. plasticity and crack blunting.

Figure 1. Dimensions of the crack tip opening displacement test (CTOD) specimen. Figure 1. Dimensions of the crack tip opening displacement test (CTOD) specimen.

Micromachines 2020, 11, x 5 of 21

Table 1. Loading characteristics during fatigue precrack stage.

Loading Stage Cycles Fmax (kN) Fmin (kN) Target Set Point Amplitude First stage 9000 −14 −1.4 −7.7 6.3 Second stage 12,000 −10 −1.0 −5.5 4.5 Third stage 50,000 −6 −0.6 −2.7 3.3

In any case, when the crack length reached the critical value of 4mm, the loading mode Micromachines 2020, 11, 430 5 of 20 transformed to slow strain rate bending until fracture (1.6 × 10−11 s−1), after the specimen was placed in the specific electrolytic cell.

Micromachines 2020, 11, x 5 of 21

Table 1. Loading characteristics during fatigue precrack stage.

Loading Stage Cycles Fmax (kN) Fmin (kN) Target Set Point Amplitude First stage 9000 −14 −1.4 −7.7 6.3 Second stage 12,000 −10 −1.0 −5.5 4.5 FigureFigure 2. 2.CTOD CTOD testing testing by by the the applicabilityapplicability of a fatigue pre-crack pre-crack stage stage to to the the notched notched specimens specimens on on Third stage 50,000 −6 −0.6 −2.7 3.3 airair environment. environment. In any case, when the crack length reached the critical value of 4mm, the loading mode Table 1. Loading characteristics during fatigue precrack stage. transformedAt this final to stage slow ofstrain slow rate strain bending rate bending until fracture loading, (1.6 a× bimetallic10−11 s−1), after plate, the with specimen especially was placedadapted crack open displacement gauges (COD gauges), was applied on the surface of the notched area, so as in theLoading specific Stage electrolytic Cycles cell. Fmax (kN) Fmin (kN) Target Set Point Amplitude to record the expansion of the upper lip of the notch. As fracture was completed, by the use of the First stage 9000 14 1.4 7.7 6.3 Force-Crack Mouth Open Displacement− (F-CMOD)− curves and the− analytical fracture mechanic relationshipsSecond were stage calculated 12,000 the toughness10 parameters1.0 KQ, Jm, CTODel,5.5 CTODpl. 4.5 − − − MoreThird precisely, stage in the 50,000 electrolytic cell6 during the0.6 electrochemical 2.7charging process, 3.3 as cathode was imposed the pre-notched CTOD specimen− of X−65 steel and as anode− was selected a platinum coated titanium electrode. The electrolytic solution used in the cathodic polarization process was In any case, when the crack length reached the critical value of 4 mm, the loading mode transformed consisted of 0.2 M NaCl and 3 g/L NH4SCN (which11 reacts1 as a poison recombination inhibitor of to slow strainFigure rate 2. CTOD bending testing until by the fracture applicability (1.6 of10 a− fatigues− ),pre-crack after the stage specimen to the notched was placedspecimens in on the speciﬁc hydrogen cations on the metal surface). The× electrochemical system was stirred by a mixture of CO2 electrolyticair cell. environment. and N2 gases in order to stabilize the pH value between 5 and 5.2 and promote its electrochemical At this ﬁnal stage of slow strain rate bending loading, a bimetallic plate, with especially adapted activity.At During this final the stage in situ of slow hydrogen strain rate cathodic bending charging loading, aprocess bimetallic the plate, appl iedwith current especially density adapted was crack open displacement gauges2 (COD gauges), was applied on the surface of the notched area, so as stabilizedcrack open at 10 displacement and 20 mA/cm gauges, for (COD a cathodic gauges), polarization was applied duration on the surface of 26 hof (according the notched to area, Petroleum so as to record the expansion of the upper lip of the notch. As fracture was completed, by the use of Industryto record Standard the expansion ISO 3183) of the (Figure upper 3).lip ofThe the surface notch. Asof fracturethe cathodically was completed, charged by specimensthe use of the was theinsulated Force-CrackForce-Crack with Mouthpolyurethane,Mouth OpenOpen Displacement except of the (F-CMOD)notche (F-CMOD)d area, curves curves in orderand and the to the promoteanalytical analytical thefracture hydrogen fracture mechanic mechanic cation relationshipselectrodiffusionrelationships were wereprocess calculated calculated specifically the the toughness toughness in this area. parameters parameters KQ, ,JmJm, CTOD, CTODel, CTODel, CTODpl. pl. MoreMore precisely, precisely, in the in electrolyticthe electrolytic cell cell during during the the electrochemical electrochemical charging charging process, process, asas cathodecathode was imposedwas theimposed pre-notched the pre-notched CTOD specimenCTOD specimen of X65 of steel X65 andsteel as and anode as anode was was selected selected a platinum a platinum coated titaniumcoated electrode. titanium The electrode. electrolytic The electrolytic solution used soluti inon the used cathodic in the polarizationcathodic polarization process wasprocess consisted was of consisted of 0.2 M NaCl and 3 g/L NH4SCN (which reacts as a poison recombination inhibitor of 0.2 M NaCl and 3 g/L NH4SCN (which reacts as a poison recombination inhibitor of hydrogen cations on hydrogen cations on the metal surface). The electrochemical system was stirred by a mixture of CO2 the metal surface). The electrochemical system was stirred by a mixture of CO2 and N2 gases in order and N2 gases in order to stabilize the pH value between 5 and 5.2 and promote its electrochemical to stabilize the pH value between 5 and 5.2 and promote its electrochemical activity. During the in situ activity. During the in situ hydrogen cathodic charging process the applied current density was hydrogen cathodic charging process the applied current density was stabilized at 10 and 20 mA/cm2, for a stabilized at 10 and 20 mA/cm2, for a cathodic polarization duration of 26 h (according to Petroleum cathodicIndustry polarization Standard duration ISO 3183) of 26(Figure h (according 3). The surface to Petroleum of the cathodically Industry Standard charged ISO specimens 3183) (Figure was 3). The surfaceinsulated of thewith cathodically polyurethane, charged except specimens of the notche wasd insulated area, in order with to polyurethane, promote the excepthydrogen of the cation notched area, inelectrodiffusion order to promote process the specifically hydrogen in cation this area. electrodiffusion process specifically in this area.

(a) (b)

Figure 3. (a) Schematic illustration of the in situ hydrogen cathodic charging device, (b) Second stage of CTOD testing by the applicability of slow strain rate bending loading under in situ hydrogen cathodic charging conditions (polarization process).

(a) (b)

FigureFigure 3. (a )3. Schematic (a) Schematic illustration illustration of theof the in in situ situ hydrogen hydrogen cathodic cathodic chargingcharging device, device, (b (b) )Second Second stage stage of CTODof testing CTOD bytesting the applicabilityby the applicability of slow of strain slow ratestrain bending rate bending loading loading under under in situ in hydrogensitu hydrogen cathodic chargingcathodic conditions charging (polarization conditions (polarization process). process).

Micromachines 2020, 11, 430 6 of 20 Micromachines 2020, 11, x 6 of 21

3. Results and Discussion The microstructuremicrostructure of of X65 X65 pipeline pipeline steel steel consisted consisted of ferritic of ferritic and bainitic and bainitic grains withgrains high with volume high volumevolume fractionvolume offraction degenerated of degenerated pearlite islands. pearlite It island was alsos. It detected was also the detected existence the of existence limited proportion of limited ofproportion non-metallic of inclusionsnon-metallic (manganese inclusions sulfide, (mangane aluminumse sulfide, oxide) andaluminum embrittling oxide) Martensite-Austenite and embrittling phasesMartensite-Austenite (MAs). The average phases grain (MAs). size The of the average ferritic-bainitic grain size microstructure of the ferritic-bainitic was evaluated microstructure by means was of Electronevaluated Backscatter by means Diofff Electronraction technique Backscatte (EBSD)r Diffraction equal to technique 18 µm (Figure (EBSD)4). Theequal chemical to 18 μm composition (Figure 4). andΤhe mechanicalchemical composition properties and of X65 mechanical steel are properties shown in of Tables X65 2steel and are3. Theshown studied in Tables specimens 2 and 3. had The dimensionsstudied specimens 120 19had dimensions19 mm and 120 had × been19 × extracted19 mm and from had pipelines been extracted with external from pipelines diameter with 29 × × inexternal (736.6 diameter mm). These 29 in pipelines (736.6 mm). emanated These from pipelines the last emanated stage of from thermo-mechanical the last stage productionof thermo- processmechanical without production being subjectedprocess without to sour being service subjec conditions.ted to sour The service X65 pipeline conditions. steel The was X65 developed pipeline bysteel a continuouswas developed casting by a process continuous with casting the imposition process with of de-oxidation the imposition mechanism of de-oxidation of Al-killed mechanism during industrialof Al-killed refining during process. industrial refining process.

Figure 4. (a) Ferritic-Banitic microstructure of X65 pipeline steel by means of Optical Microscopy, (Figureb) Degenerated 4. (a) Ferritic-Banitic perlite islands microstructure and MAs micro of X65/constituency pipeline steel within by means the crystalographyc of Optical Microscopy, and (c) grain (b) orientaionDegenerated mapping perlite andislands averange and MAs grain micro/cons size by EBSDtituency analysis. within the crystalographyc and (c) grain orientaion mapping and averange grain size by EBSD analysis. Table 2. Chemical composition of X65 pipeline steel. Table 2. Chemical composition of X65 pipeline steel. C Si Mn P S V Nb Ti Fe C Si0.16% 0.45%Mn 1.65%P 0.02% 0.01%S 0.08%V 0.05% 0.06%Nb Bal. Ti Fe 0.16% 0.45% 1.65% 0.02% 0.01% 0.08% 0.05% 0.06% Bal. Table 3. Mechanical properties of X65 pipeline steel. Table 3. Mechanical properties of X65 pipeline steel. Yield Strength Tensile Strength Yield to Tensile Elongation Hardness Yield StrengthMin (KSI) Tensile StrengthMin (KSI) Min YieldRatio to (max)Tensile Ratio (%)Elongation (HV)Hardness Min (KSI) (KSI) (max) (%) (HV) 65 77 0.93 18 220 65 77 0.93 18 220

The effect of the microstructure on the hydrogen cations’ migration phenomenon and microcrack propagation velocity is mainly attribute to the existence of the lower bainite islands and the interfaces between the ferritic-bainitic and ferritic-degenereated perlitic grains.

Micromachines 2020, 11, 430 7 of 20

MicromachinesThe eff 2020ect, of 11 the, x microstructure on the hydrogen cations’ migration phenomenon and microcrack7 of 21 propagation velocity is mainly attribute to the existence of the lower bainite islands and the interfaces betweenThe islands the ferritic-bainitic of lower bainite and are ferritic-degenereated characterized by augmented perlitic grains. dislocation densities leading to the entrapmentThe islands effect of of lower atomic bainite hydrogen are characterized in the crystallographic by augmented texture. dislocation Additionally, densities the leading interfacial to the regionsentrapment between effect the of atomicferritic hydrogenand bainitic in the areas crystallographic are characterized texture. by Additionally,high misorientation the interfacial angle distributionregions between factor the and ferritic provoque and bainitic the supersaturation areas are characterized of crystal by structure high misorientation in molecular angle hydrogen distribution and thefactor development and provoque of extended the supersaturation microplastic of crystalfields. Consequently, structure in molecular these microstructural hydrogen and the regions development consist theof extendedmain crack microplastic initiation points fields. due Consequently, to the localized these excess microstructural of the critical regions stress consistintensity the factor main value. crack Ininitiation contrast, points the degenerated due to the perlite localized islands, excess which of the are critical characterized stress intensity by lower factor dislocation value. densities In contrast, as wellthe degeneratedas the interfacial perlite regions islands, between which arethe characterized ferritic-perlitic by regions, lower dislocation which are densitiescharacterized as well by as low the misorientationinterfacial regions angle between distribution the ferritic-perlitic factors, promote regions, the whichelectrodiffu are characterizedsion process byand low the misorientation propagation ofangle the developed distribution microcrack factors, promote networ theks. The electrodi microcrackffusion propagation process and rate the propagationincreases with of the increment developed ofmicrocrack the frequency networks. that Thethe defect microcrack coincides propagation the interfaces rate increases between with the the ferritic-perlitic increment of theand frequency ferritic- bainiticthat the areas. defect coincides the interfaces between the ferritic-perlitic and ferritic-bainitic areas. 2 AfterAfter applying applying the the in in situ situ hydrogen hydrogen cathodic cathodic charging charging process process at at 10 10 and and 20 20 mA/cm mA/cm2 currentcurrent densities,densities, a severesevere increase increase in surfacein surface hardness hardness was detectedwas detected (Figure (Figure5). This 5). e ff ectThis is attributedeffect is attributed according accordingto international to international literature literature to the interaction to the interaction between between dislocation dislocation networks networks and hydrogen and hydrogen Cottrell Cottrellatmospheres, atmospheres, to the creation to the of creation high volume of high fraction volume of hydrogenated fraction of vacancieshydrogenated (V-H clusters),vacancies and (V-H the clusters),expanded and development the expanded of interstitialdevelopment solid of solutioninterstitial [27 solid–35]. solution In particular, [27–35]. the In above particular, phenomenon the above is phenomenonattributed to theis attributed fact that hydrogenated to the fact that interstisial hydrog vacanciesenated interstisial of the crystalline vacancies structure of the impedecrystalline the structuremovement impede of dislocations, the movement resulting of dislocations, in their interlocking resulting and in immobilizationtheir interlocking eff ect.and The immobilization immobilized effect.dislocation The immobilized networks cannot dislocation act as networks active di cannotffusion ac channelst as active and diffusion the localized channels supersaturation and the localized in supersaturationhydrogen leads toin thehydrogen accumulation leads to of the residual accumulati stresses,on microplasticof residual stresses, zones and microplastic slip bands. zones As a result and slipof the bands. above, As the a result yield stressof the and above, the surfacethe yield micro stress hardness and the increase. surface Inmicro any case,hardness the highest increase. rate In of any the case,surface the microhardness highest rate of increment the surface was microhardnes detected durings increment the devolution was detected from the during uncharged the conditiondevolution to 2 fromthe cathodically the uncharged charged condition under to 10 the mA /cathodicallycm2 (in comparison charged to under the transition 10 mA/cm from (in the comparison charged condition to the 2 transitionunder a current from the density charged field condition of 10 to 20under mA/ cma current2). density field of 10 to 20 mA/cm ).

290

270

250

230

210

Microhardness (HV) 190

170 0 5 10 15 20 Current Density (mA/cm2)

Figure 5. Effect of the applied current density (10 and 20 mA/cm2) during the in situ hydrogen cathodic Figurecharging 5. processEffect of and the bending applied loading current on density the surface (10 and microhardness 20 mA/cm2 increment.) during the in situ hydrogen cathodic charging process and bending loading on the surface microhardness increment. Additionally, the diffusion depth of hydrogen cations into the microstructure of X65 steel, was validatedAdditionally, after in situ the hydrogen diffusion cathodicdepth of charging hydrogen process cations by into conducting the microstructure microhardness of X65 measurements steel, was validatedat the cross after section in ofsitu the specimens.hydrogen Withcathodic the increment charging of process the cathodic by chargingconducting current microhardness density from 2 measurements10 to 20 mA/cm at, the dicrossffusion section depth of ofthe hydrogen specimen cationss. With into the the increment crystal structureof the cathodic augmented charging from current50 to 180 densityµm (Figure from6 ).10 to 20 mA/cm2, the diffusion depth of hydrogen cations into the crystal structure augmented from 50 to 180 μm (Figure 6).

Micromachines 2020, 11, 430 8 of 20 Micromachines 2020, 11, x 8 of 21

Figure 6. Determination of hydrogen cations’ diffusion depth into the polycrystalline matrix by Figure 6. Determination of hydrogen cations’ diffusion depth into the polycrystalline matrix by conducting microhardness mesurements at the cross sections of the specimens, after in situ hydrogen conducting microhardness mesurements at the cross sections of the specimens, after in situ hydrogen cathodic charging polarization process during slow strain rate bending loading at (a) 10 mA/cm2 and cathodic charging polarization process during slow strain rate bending loading at (a) 10 mA/cm2 and (b) 20 mA/cm2 current density. (b) 20 mA/cm2 current density. By the use of second Fick’s law was calculated the diffusion coefficient of hydrogen cations into By the use of second Fick’s law was calculated the diffusion coefficient of hydrogen cations into the matrix structure. the matrix structure. D = x2/4t D = x2/4t where x: the hydrogen cations’ diffusion depth (cm) and t: the cathodic charging time duration (s). whereFor x: a the cathodic hydrogen charging cations’ current diffusion density depth field (cm) of 10and mA t: the/cm cathodic2 and in situcharging hydrogen time duration cathodic (s) charging For a cathodic charging current density field of 10 mA / cm2 and in situ hydrogen cathodic duration 26 h, the penetration depth of hydrogen cations’ in the matrix region was determined to be charging duration 26 h, the penetration depth of hydrogen cations’ in the matrix region was 50 µm. Accordingly, the diffusion coefficient was calculated as follows: determined to be 50 μm. Accordingly, the diffusion coefficient was calculated as follows: 2 2 11 2 1 D D= x= x/24/4t t= =0.0050 0.00502 / (4 × 2626 × 3600)3600) = =6.677356.67735 × 10−1011 cm− 2scm−1 s− × × × For a 20 mA/cm2 cathodic charging current density field and an in situ hydrogen cathodic For a 20 mA/cm2 cathodic charging current density field and an in situ hydrogen cathodic charging charging duration 26 h, the penetration depth of hydrogen cations was determined to be 180 μm. duration 26 h, the penetration depth of hydrogen cations was determined to be 180 µm. Accordingly, Accordingly, the diffusion coefficient was calculated as follows: the diffusion coefficient was calculated as follows: D = x2/4t = 0.001802 / (4 × 26 × 3600) = 8.65385 × 10−11 cm2s−1 2 2 11 2 1 D = x /4t = 0.00180 / (4 26 3600) = 8.65385 10− cm s− As it can be seen from Figure 7 the diffusion× × coefficient of hydrogen× atoms into crystalline structureAs it canduring beseen the in from situ Figure hydrogen7 the cathodic di ffusion charging coefficient process of hydrogen and bending atoms loading into crystalline until fracture structure −11 2 −1 −10 2 −1 duringwas icreased the in situlinearly hydrogen from 6.67735 cathodic × charging10 cm s process to 8.65385 and bending× 10 cm loadings by untilincreasing fracture the was current icreased 2 density field from 10 to 20 11mA/cm2 1. This observation10 can be2 correlated1 with the expansion of the linearly from 6.67735 10− cm s− to 8.65385 10− cm s− by increasing the current density field electrodiffusion fronts× due to the activation of hydrogen× cations’ migration effect by easy paths- from 10 to 20 mA/cm2. This observation can be correlated with the expansion of the electrodiffusion channels (ferrite-pearlite interfaces) and the excess of the electrostatic-thermodynamic barrier of fronts due to the activation of hydrogen cations’ migration effect by easy paths-channels (ferrite-pearlite reversible hydrogen traps (pearlite-bainite interfaces and low angle grain boundaries). interfaces) and the excess of the electrostatic-thermodynamic barrier of reversible hydrogen traps (pearlite-bainite interfaces and low angle grain boundaries). As it can be seen from Scanning Electron Microscopy images (Figure8), after the in situ hydrogen cathodic charging process both at 10 and 20 mA/cm2, in the surface structure of the X65 pipeline steel was detected micro-cracking phenomenon around the interfaces of blisters and non-metallic inclusions. These micro-cracks propagated by a crooked line morphology (stepwise micro-cracking) as the non-metallic inclusions and the blister formations act like electrokinetic barriers to the development of electrodiffusion process. More precisely these microstructural components act like high-energy irreversible traps and lead to the supersaturation in hydrogen of the matrix material, resulting in the excess of critical stress intensity factor value and the development micro-cracking phenomenon.

Micromachines 2020, 11, x 9 of 21

Micromachines 2020, 11, x 430 99 of of 21 20

Figure 7. Effect of the applied current density (10 mA/cm2 and 20 mA/cm2) on the diffusion coefficient of hydrogen cations into the ferritic-bainitic microstructure of X65 pipeline steel.

As it can be seen from Scanning Electron Microscopy images (Figure 8), after the in situ hydrogen cathodic charging process both at 10 and 20 mA/cm2, in the surface structure of the X65 pipeline steel was detected micro-cracking phenomenon around the interfaces of blisters and non- metallic inclusions. These micro-cracks propagated by a crooked line morphology (stepwise micro- cracking) as the non-metallic inclusions and the blister formations act like electrokinetic barriers to the development of electrodiffusion process. More precisely these microstructural components act like high-energy irreversible traps and lead to the supersaturation in hydrogen of the matrix material, resulting in the excess of critical stress intensity factor value and the development micro-cracking Figure 7. Effect of the applied current density (10 mA/cm2 and 20 mA/cm2) on the diffusion coefficient phenomenon. Figureof hydrogen 7. Effect cations of the into applied the ferritic-bainitic current density microstructure (10 mA/cm2 and of 20 X65 mA/cm pipeline2) on steel. the diffusion coefficient of hydrogen cations into the ferritic-bainitic microstructure of X65 pipeline steel.

Figure 8. (a) Stepwise micro-cracking and micro-cracking branching, (b) crack initiation points at hydrideFigure /8.non-metalic (a) Stepwise inclusion micro-cracking interfaces, an (cd,d micro-cracking) identification ofbranching, a non metallic (b) crack inclusion initiation at a high points bridge at formationhydride/ non-metalic from Electron inclusion Backscatter interfaces, Diffraction (c,d) technique identification (EBSD). of a non metallic inclusion at a high bridge formation from Electron Backscatter Diffraction technique (EBSD). As it can be seen from Figure9 the surficial density of microcracks was observed to increase 2 2 exponentially (from 2 cracks/cm to 9 microcracks/cm ) with the increment of the required polarization current density during the in situ hydrogen cathodic charging process. The rate of this increase appeared to be higher when switching from 10 to 20 mA/cm2 current density compared to switching from the uncharged condition to the charged condition under 10 mA/cm2 current density. In addition, the maximum width and maximum length of the microcracks were observed to increase continuously with the increment of the applied current density during cathodic polarization process. For a 10 mA/cm2 current density field, the maximum crack width was determined at 2.2 µm µ 2 and theFigure maximum 8. (a) Stepwise crack lengthmicro-cracking at 2.4 m. and Correspondingly micro-cracking branching, for a current (b) crack density initiation fieldof points 20 mA at /cm , the maximumhydride/ non-metalic crack width inclusion was determined interfaces, at(c, 2.75d) identificationµm and the of maximum a non metallic crack inclusion length at at 3.1 a highµm. It is, therefore,bridge conceivable formation from that Electron the growth Backscatter rate of Diffraction the microcracks technique is higher(EBSD). during the transition from the

Micromachines 2020, 11, x 10 of 21

As it can be seen from Figure 9 the surficial density of microcracks was observed to increase exponentially (from 2 cracks/cm2 to 9 microcracks/cm2) with the increment of the required polarization current density during the in situ hydrogen cathodic charging process. The rate of this increase appeared to be higher when switching from 10 to 20 mA/cm2 current density compared to switching from the uncharged condition to the charged condition under 10 mA/cm2 current density.

9 ) 2

Micromachines 2020, 11, x 10 of 21 3 Micromachines 2020, 11, 430 10 of 20 As it can be seen from Figure 9 the surficial density of microcracks was observed to increase exponentially (from(1/cm n Crack Density 2 cracks/cm2 to 9 microcracks/cm2) with the increment of the required polarization current density during the in situ hydrogen cathodic charging2 process. The rate of this uncharged state to the0 cathodically polarized state under 10 mA/cm compared to the corresponding 2 2 increaseone during appeared the transition to be0 higher from thewhen charging switching 5 process from at10 10 a currentto 20 mA/cm density 15 current ﬁeld of density 10 20 mA/ cmcomparedto that to of 2 2 switching20 mA/cm from(Figure the uncharged10). condition toCurrent the charged Density condition (mA/cm2) under 10 mA/cm current density.

9 )

Figure 9. Correlation2 between surficial density of microcracks and applied current density during in situ hydrogen cathodic charging and slow strain rate bending loading. 6 In addition, the maximum width and maximum length of the microcracks were observed to increase continuously with the increment of the applied current density during cathodic polarization 2 process. For a 10 mA/cm3 current density field, the maximum crack width was determined at 2.2 μm and the maximum crack length at 2.4 μm. Correspondingly for a current density field of 20 mA/cm2,

the maximum crack(1/cm n Crack Density width was determined at 2.75 μm and the maximum crack length at 3.1 μm. It is, therefore, conceivable that the growth rate of the microcracks is higher during the transition from the uncharged state0 to the cathodically polarized state under 10 mA/cm2 compared to the corresponding one during0 the transition 5 from the charging 10 process 15 at a current 20density field of 10 2 mA/cm2 to that of 20 mA/cm2 (Figure 10).Current Density (mA/cm ) Figure 9. Correlation between surﬁcial density of microcracks and applied current density during in

situ hydrogen cathodic charging and slow strain rate bending loading. Figure 9. Correlation between surficial density of microcracks and applied current density during in situ hydrogen cathodic charging and slow strain rate bending loading.

In addition, the maximum width and maximum length of the microcracks were observed to increase continuously with the increment of the applied current density during cathodic polarization process. For a 10 mA/cm2 current density field, the maximum crack width was determined at 2.2 μm and the maximum crack length at 2.4 μm. Correspondingly for a current density field of 20 mA/cm2, the maximum crack width was determined at 2.75 μm and the maximum crack length at 3.1 μm. It is, therefore, conceivable that the growth rate of the microcracks is higher during the transition from the uncharged state to the cathodically polarized state under 10 mA/cm2 compared to the corresponding one during the transition from the charging process at a current density field of 10 mA/cm2 to that of 20 mA/cm2 (Figure 10). Figure 10. (a) maximum crack length and (b) maximum crack width of surficial microcracks in conjuctionFigure 10. with(a) maximum the applied crack current length density and during (b) maximum in situ hydrogen crack width charging of surficial process andmicrocracks slow strain in rateconjuction bending with loading. the applied current density during in situ hydrogen charging process and slow strain rate bending loading. Through X-ray diffractometry, the existence of hydrides with FeH3 stoichiometry after in situ cathodic charging process both at 10 and 20 mA/cm2 was detected. In addition, the shift of the main di ffraction peaks, compared to those of the uncharged specimens, at lower diffraction angles is related to the development of compressive residual stresses within the crystallographic structure during the electrochemical process (Figure 11). In the surface area of the cathodically charged specimens of X65 pipeline steel, for a current density field of 10 mA/cm2, an increased volume fraction of blisters with dome-type morphology was detected. After the polarization process of the specimens at a current density field of 20 mA/cm2, the development of elongated blisters was observed, with extended microcrack branching networks at their surface. The microstructural locations correlated with the nucleation process of blisters were the energy upgraded areas, such as the triple junction locations, or the non-metallic inclusions and hydride interfaces (Figure 12)[36–38]. Figure 10. (a) maximum crack length and (b) maximum crack width of surficial microcracks in As it can be seen from Figure 13 the surficial density of blisters increases proportionally and conjuction with the applied current density during in situ hydrogen charging process and slow strain linearlyrate withbending the loading. cathodic polarization current density during the electrochemichal charging process. The rate of this increment is observed to be higher during the transition from 10 to 20 mA/cm2 current

Micromachines 2020, 11, x 11 of 21

MicromachinesThrough2020 X-ray, 11, 430 diffractometry, the existence of hydrides with FeH3 stoichiometry after in11 situ of 20 cathodic charging process both at 10 and 20 mA/cm2 was detected. In addition, the shift of the main diffraction peaks, compared to those of the uncharged specimens, at lower diffraction angles is relateddensity to compared the development to the transition of compressive from the residu unchargedal stresses condition within to the polarizedcrystallographic condition structure under 2 during10 mA /thecm electrochemicalcurrent density. process (Figure 11).

Figure 11. Identiﬁcation of phases by X-Ray Diﬀraction patterns (a) for the uncharged condition; 2 Figure(b,c) after 11. Identification cathodic polarization of phases at by 10 X-Ray and 20 Diffraction mA/cm current patterns density. (a) for the uncharged condition; (b,c) after cathodic polarization at 10 and 20 mA/cm2 current density.

Micromachines 2020, 11, x 12 of 21

In the surface area of the cathodically charged specimens of X65 pipeline steel, for a current density field of 10 mA/cm2, an increased volume fraction of blisters with dome-type morphology was detected. After the polarization process of the specimens at a current density field of 20 mA/cm2, the development of elongated blisters was observed, with extended microcrack branching networks at their surface. The microstructural locations correlated with the nucleation process of blisters were the energy upgraded areas, such as the triple junction locations, or the non-metallic inclusions and hydride interfaces (Figure 12) [36–38]. Micromachines 2020, 11, x 12 of 21

In the surface area of the cathodically charged specimens of X65 pipeline steel, for a current density field of 10 mA/cm2, an increased volume fraction of blisters with dome-type morphology was detected. After the polarization process of the specimens at a current density field of 20 mA/cm2, the development of elongated blisters was observed, with extended microcrack branching networks at their surface. The microstructural locations correlated with the nucleation process of blisters were the energy upgraded areas, such as the triple junction locations, or the non-metallic inclusions and Micromachines 2020, 11, 430 12 of 20 hydride interfaces (Figure 12) [36–38].

Figure 12. (a) Blister formations with a dome-type morphology after the in situ hydrogen cathodic charging process at 10 mA/cm2 current density under slow strain rate bending loading; (b) Elongated blisters with surficial micro-cracks and blistering on blistering nucleation effect after the in situ hydrogen cathodic charging process at 20 mA/cm2 current density under slow strain rate bending loading.

As it can be seen from Figure 13 the surficial density of blisters increases proportionally and linearly with the cathodic polarization current density during the electrochemichal charging process. The rateFigure of this 12. (incrementa) Blister formations is observed with to abe dome-type higher during morphology the transition after the from in situ 10 hydrogento 20 mA/cm cathodic2 current Figure 12. (a) Blister formations with a dome-type morphology after the in situ hydrogen cathodic densitycharging compared process to atthe 10 transition mA/cm2 current from the density uncharged under slow condition strain rate to the bending polarized loading; condition (b) Elongated under 10 charging process at 10 mA/cm2 current density under slow strain rate bending loading; (b) Elongated mA/cmblisters2 current with surﬁcialdensity. micro-cracks and blistering on blistering nucleation eﬀect after the in situ hydrogen blisters with surficial micro-cracks and blistering on blistering nucleation effect after the in situ cathodic charging process at 20 mA/cm2 current density under slow strain rate bending loading. hydrogen cathodic charging process at 20 mA/cm2 current density under slow strain rate bending loading. 14

As it can be seen12 from Figure 13 the surficial density of blisters increases proportionally and linearly with the cathodic10 polarization current density during the electrochemichal charging process. The rate of this increment is observed to be higher during the transition from 10 to 20 mA/cm2 current density compared to the8 transition from the uncharged condition to the polarized condition under 10 mA/cm2 current density.6 # Blisters 4

142

120 10 0 5 10 15 20 Current Density (mA/cm2) 8

Figure 13. Surﬁcial density6 of blister formations related to the applied current density during the in

# Blisters situ hydrogen cathodic charging process and slow strain rate bending loading. 4 Figure 13. Surficial density of blister formations related to the applied current density during the in Concerning the average growth size of blister formations, it seems to decrease from 180 µm situ hydrogen cathodic2 charging process and slow strain rate bending loading. to 110 µm by increasing the required cathodic polarization current density from 10 to 20 mA/cm2 0 (Figure 14). The above observation is attributed to the fact that by the increment of current density 0 5 10 15 20 ﬁeld during cathodic polarization process, more nucleation centers for blister formation are activated, Current Density (mA/cm2) thereby limiting their growth rate and mean size. For the X65 steel specimens subjected to CTOD mechanical testing in ambient air, the mechanical

values associated with the fracture toughness parameter were determined as follows: KQ = 42.070 MPa m1/2,J = 280.76 KN/m2, CTOD = 0.0061 mm, CTOD = 1.92 mm. · Figure 13. Surficial density of blisterel formations related to thepl applied current density during the in Forsitu thehydrogen specimens cathodic subjected charging toprocess in situ and hydrogen slow strain cathodic rate bending charging loading. process during slow strain rate bending, a signiﬁcant decrease in fracture toughness was observed, as it can be seen from the

characteristic mechanical values for applied current densities of 10 mA/cm2 and 20 mA/cm2. More precisely after the in situ hydrogen cathodic charging eﬀect at 10 mA/cm2 the main parameters correlated with toughness properties were calculated as follows: K = 35.04 MPa m1/2, J = 202.01 KN/m2, CTOD Q · el = 0.0054 mm, CTODpl =1.44 mm. Respectively after the in situ hydrogen cathodic charging procedure at 20 mA/cm2, the above parameters were identiﬁed as follows: K = 30.04 MPa m1/2, J = 175.8 KN/m2, Q · CTODel = 0.0037 mm, CTODpl = 0.98 mm (Figure 15). Micromachines 2020, 11, x 13 of 21

Concerning the average growth size of blister formations, it seems to decrease from 180 μm to 110 μm by increasing the required cathodic polarization current density from 10 to 20 mA/cm2 (Figure 14). The above observation is attributed to the fact that by the increment of current density field during cathodic polarization process, more nucleation centers for blister formation are activated, thereby limiting their growth rate and mean size.

180 m)

μ 160

140 Micromachines 2020, 11, x 13 of 21

Concerning the average120 growth size of blister formations, it seems to decrease from 180 μm to

110 μm by increasing the( ssize Mean Blisters required cathodic polarization current density from 10 to 20 mA/cm2 (Figure 14). The above observation100 is attributed to the fact that by the increment of current density field during cathodic polarization10 process, more nucleation 15 centers for blister formation 20 are activated, Micromachines 2020, 11, 430 13 of 20 thereby limiting their growth rate and meanCurrent size. Density (mA/cm2)

180 Figure 14. Mean size of blisters developed under the in situ hydrogen cathodic charging process (10 and 20 mA/cm2) and slow strain rate bending loading. m)

μ 160 For the X65 steel specimens subjected to CTOD mechanical testing in ambient air, the mechanical values associated with the fracture toughness parameter were determined as follows: KQ = 42.070 140 MPa·m1/2, J = 280.76 KN/m2, CTODel = 0.0061 mm, CTODpl = 1.92 mm. For the specimens subjected to in situ hydrogen cathodic charging process during slow strain rate bending, a significant120 decrease in fracture toughness was observed, as it can be seen from the

characteristic mechanical( ssize Mean Blisters values for applied current densities of 10 mA/cm2 and 20 mA/cm2. More 2 precisely after the in100 situ hydrogen cathodic charging effect at 10 mA/cm the main parameters correlated with toughness10 properties were calculated 15 as follows: KQ = 35.04 20MPa·m1/2, J = 202.01 2 KN/m , CTODel = 0.0054 mm, CTODpl =1.44Current mm. Density Respectively (mA/cm 2)after the in situ hydrogen cathodic charging procedure at 20 mA/cm2, the above parameters were identified as follows: KQ = 30.04 Figure 14. Mean size of blisters developed under the in situ hydrogen cathodic charging process MPa·m1/2, J = 175.8 KN/m2, CTODel = 0.0037 mm, CTODpl = 0.98 mm (Figure 15). (10 and 20 mA/cm2) and slow strain rate bending loading. Figure 14. Mean size of blisters developed under the in situ hydrogen cathodic charging process (10 and 20 mA/cm2) and slow strain rate bending loading.

For the X65 steel specimens subjected to CTOD mechanical testing in ambient air, the mechanical values associated with the fracture toughness parameter were determined as follows: KQ = 42.070 MPa·m1/2, J = 280.76 KN/m2, CTODel = 0.0061 mm, CTODpl = 1.92 mm. For the specimens subjected to in situ hydrogen cathodic charging process during slow strain rate bending, a significant decrease in fracture toughness was observed, as it can be seen from the characteristic mechanical values for applied current densities of 10 mA/cm2 and 20 mA/cm2. More precisely after the in situ hydrogen cathodic charging effect at 10 mA/cm2 the main parameters correlated with toughness properties were calculated as follows: KQ = 35.04 MPa·m1/2, J = 202.01 KN/m2, CTODel = 0.0054 mm, CTODpl =1.44 mm. Respectively after the in situ hydrogen cathodic charging procedure at 20 mA/cm2, the above parameters were identified as follows: KQ = 30.04 MPa·m1/2, J = 175.8 KN/m2, CTODel = 0.0037 mm, CTODpl = 0.98 mm (Figure 15).

Figure 15. Force-Crack Mouth Open Displacement (CMOD) curves of CTOD tested X65 pipeline steel (a) in the uncharged condition; (b) after in situ hydrogen cathodic charging at 10 mA/cm2 current density; (c) after in situ hydrogen cathodic charging at 20 mA/cm2 current density.

Consequently, during the transition from the uncharged state of X65 pipeline steel to the 2 cathodically polarized, for a current density field of 10 mA/cm , the parameters KQ, J, CTODel and CTODpl decreased by 16.71%, 28%, 11.48%, 25% respectively. During the transition from the charged condition for a current density field of 10 mA / cm2 to that with a current density field of 20 mA/cm2, the parameters KQ, J, CTODel and CTODpl decreased by 28.6%, 37.4%, 39.34%, 48.95% respectively. The rate of the drop of the main toughness parameters was considered to be higher during the transition from 10 to 20 mA/cm2 current density, as compared to the transition from the uncharged condition to the cathodically polarized condition under 10 mA/cm2 (Figure 16). The significant drop of the fracture toughness parameters after the in situ hydrogen cathodic charging process of X65 pipeline steel under a current density field of 10 mA/cm2, is attributed to the fact that the cathodic polarization procedure causes the supersaturation of surface layers in hydrogen cations, resulting in the development of extensive deformation fields. The consequent work hardening effect of the surface layers is correlated with an elevated microcrack surficial density, which is mainly attributed to the evolution of Hydrogen Enhanced Localized Plasticity (HELP), Hydrogen Induced Cracking (HIC) and Elastic Shielding Reactions mechanisms [39–41]. The HELP mechanism is based on the development of highly localized plastic deformation fields, around the hydrogen accumulation points. Moreover, the occurrence of localized plastic deformation fields favors the excess of critical Micromachines 2020, 11, 430 14 of 20

resolved shear stress (τCRSS), at an applied stress much lower than the actual yield strength of the examined material. This results in the material surrounding the localized plasticity zone (that has not yielded yet) to raise a significant barrier to slip action. As a consequence microcracks are developed in the meeting region of the plasticized and the non-deformed material. The HIC mechanism is Micromachinesbased on 2020 the, evolution 11, x of high internal pressure inside the matrixes as a result of the extremely high14 of 21 accumulation of hydrogen cations (which exceeds the critical localized chemical potential) [42]. This Figure 15. Force-Crack Mouth Open Displacement (CMOD) curves of CTOD tested X65 pipeline steel fact provokes the formation of over-pressurized hydrogen cavities, which lead the surrounding crystal 2 lattice(a) in to the deform uncharged plastically condition; favoring (b) after the formationin situ hydrogen of cracks cath evenodic without charging the at presence10 mA/cm of externallycurrent 2 applieddensity; load. (c) after When in situ the hydrogen HIC mechanism cathodic occurscharging near at 20 the mA/cm surface current of the density. specimen, it results in the development of blisters as the increased pressure impels the surrounding material to the surface. Consequently, during the transition from the uncharged state of X65 pipeline steel to the The latter embrittlement mechanism relies on the limitation of elastic interactions between various stress 2 cathodicallyaccumulation polarized, fields and for structure a current imperfections, density field such of 10 as mA/cm dislocation, the networks parameters and CottrellKQ, J, CTOD hydrogenel and CTODatmospherespl decreased (Elastic by 16.71%, shielding 28%, of stress 11.48%, centers 25%/ respectively.H-Shielding) [During43]. This the limitation transition is a from consequence the charged of 2 2 conditionthe ability for of a hydrogencurrent density atmospheres field of to 10 reconfigure mA / cm themselvesto that with around a current stress density concentration field of points20 mA/cm and , themoving parameters dislocations, KQ, J, CTOD resultingel and in CTOD the developmentpl decreased of by severe 28.6%, reaction 37.4%, forces 39.34%, to the 48.95% overall respectively. stress field. ForThe a critical rate of hydrogen the drop concentrationof the main toughness barrier within para themeters polycrystalline was considered matrix, to the be fieldhigher of repulsionduring the transitionforces tends from to10 be to attractive,20 mA/cm as2 current a result density, of the reaction as compared developed to the due transition to the extended from the hydrogenuncharged conditionaccumulation. to the cathodically The above phenomenon polarized condition causes dislocation under 10 coalescence mA/cm2 (Figure and favors 16). crack initiation [44].

Figure 16. Reduction rate of the main toughness parameters (a) KQ,(b) J,(c) CTODel, and (d) CTODpl Figurein function 16. Reduction with the rate applied of the current main density toughness during parameters the in situ ( hydrogena) KQ, (b) cathodic J, (c) CTOD chargingel, and process (d) CTOD and pl in functionslow strain with rate the bending applied loading. current density during the in situ hydrogen cathodic charging process and slow strain rate bending loading. Under the in situ hydrogen cathodic charging testing with a polarization ﬁeld of 20 mA/cm2, thereThe are significant two predominant drop of the embrittlement fracture toughness mechanisms parameters which provoke after the this in signiﬁcant situ hydrogen deterioration cathodic of the mechanical properties. The former is related to the nucleation and development of hydrides charging process of X65 pipeline steel under a current density field of 10 mA/cm2, is attributed to the (FeH ) within the crystalline structure and relies on the fact that hydrides are generally considered to fact that3 the cathodic polarization procedure causes the supersaturation of surface layers in hydrogen be brittle phases. Therefore the formation of hydrides leads to a considerable reduction in the critical cations, resulting in the development of extensive deformation fields. The consequent work stress intensity factor (KIC) for crack propagation and consequently favors the extended and rapid hardening effect of the surface layers is correlated with an elevated microcrack surficial density, crack development, causing the drop of the fracture toughness of X65 pipeline steel. which is mainly attributed to the evolution of Hydrogen Enhanced Localized Plasticity (HELP), Hydrogen Induced Cracking (HIC) and Elastic Shielding Reactions mechanisms [39–41]. The HELP mechanism is based on the development of highly localized plastic deformation fields, around the hydrogen accumulation points. Moreover, the occurrence of localized plastic deformation fields favors the excess of critical resolved shear stress (τCRSS), at an applied stress much lower than the actual yield strength of the examined material. This results in the material surrounding the localized plasticity zone (that has not yielded yet) to raise a significant barrier to slip action. As a consequence microcracks are developed in the meeting region of the plasticized and the non-deformed material. The HIC mechanism is based on the evolution of high internal pressure inside the matrixes as a result of the extremely high accumulation of hydrogen cations (which exceeds the critical localized chemical potential) [42]. This fact provokes the formation of over-pressurized hydrogen cavities, which lead

Micromachines 2020, 11, x 15 of 21

the surrounding crystal lattice to deform plastically favoring the formation of cracks even without the presence of externally applied load. When the HIC mechanism occurs near the surface of the specimen, it results in the development of blisters as the increased pressure impels the surrounding material to the surface. The latter embrittlement mechanism relies on the limitation of elastic interactions between various stress accumulation fields and structure imperfections, such as dislocation networks and Cottrell hydrogen atmospheres (Elastic shielding of stress centers / H- Shielding) [43]. This limitation is a consequence of the ability of hydrogen atmospheres to reconfigure themselves around stress concentration points and moving dislocations, resulting in the development of severe reaction forces to the overall stress field. For a critical hydrogen concentration barrier within the polycrystalline matrix, the field of repulsion forces tends to be attractive, as a result of the reaction developed due to the extended hydrogen accumulation. The above phenomenon causes dislocation coalescence and favors crack initiation [44]. Under the in situ hydrogen cathodic charging testing with a polarization field of 20 mA/cm2, there are two predominant embrittlement mechanisms which provoke this significant deterioration of the mechanical properties. The former is related to the nucleation and development of hydrides (FeH3) within the crystalline structure and relies on the fact that hydrides are generally considered to be brittle phases. Therefore the formation of hydrides leads to a considerable reduction in the critical Micromachines 2020, 11, 430 15 of 20 stress intensity factor (KIC) for crack propagation and consequently favors the extended and rapid crack development, causing the drop of the fracture toughness of X65 pipeline steel. The second one can be attributed to the fact thatthat as thethe currentcurrent densitydensity fieldfield increasesincreases duringduring cathodic polarizationpolarization process, process, the the di ffdiffusionusion coe fficoefcientficient of hydrogen of hydrogen cations cations into the into matrix, the thematrix, surficial the densitysurficial ofdensity microcrack of microcrack networks networks and blister and blister formations formations increases increases simultaneously, simultaneously, reinforcing reinforcing the decohesionthe decohesion effect effect between between ferritic ferritic and and bainitic bainit grains.ic grains. The The responsible responsible embrittlement embrittlement mechanism mechanism for thefor abovethe above observations observations is the is Hydrogen the Hydrogen Enhanced Enhanc Decohesioned Decohesion (HEDE) (HEDE) effect. This effect. phenomenon This phenomenon is based onis based the fact on that the atomic fact that hydrogen atomic penetrateshydrogen intopenetr theates crystalline into the structure crystallin ande structure occupies and intermediate occupies interstitialintermediate locations interstitial within locations the crystallographic within the lattice.crystallographic In these locations, lattice. In atomic these hydrogen locations, provokes atomic ahydrogen significant provokes reduction a significant in atomic cohesivereduction energy in atomic and cohesive strength. energy As a result,and strength. the formation As a result, of micro the defectsformation isfavored, of micro whose defects enlargement is favored, whose and volume enlargement fraction and increment volume catalyze fractionthe increment development catalyze of microcracksthe development [44–48 of]. microcracks [44–48]. The fracturedfractured surfacessurfaces of of the the specimens specimens that that had ha beend been submitted submitted under under CTOD CTOD testing testing both both at the at airthe andair and hydrogenated hydrogenated environment environment consisted consisted of the of fatigue-precrackedthe fatigue-precracked area ar andea and the the bended bended area. area. SpecificallySpecifically forfor the the uncharged uncharged and and CTOD CTOD tested test specimens,ed specimens, the fatigue the pre-crackedfatigue pre-cracked area appeared area toappeared consist ofto benchmarkconsist of benchmark formations, formations, with extended with widthextended and width intense and topographic intense topographic relief, expanded relief, aboutexpanded 45◦ fromabout the 45° loading from the axis loading (due to axis the (due excess to ofthe the excess critical of the resolved critical shear resolved stress shearτcrss). stress τcrss). The bended area consistedconsisted ofof equiaxialequiaxial craterscraters withwith homogeneoushomogeneous distributiondistribution andand sphericalspherical symmetry (Figure 1717).).

Figure 17. Fractured surfaces after CTOD testing of X65 steel for the uncharged condition (a)) fatiguefatigue pre-cracked area;area; ((bb)) bendedbended area.area.

For the specimens subjected to slow strain rate bending loading, under the effect of hydrogen cathodic charging process at a current density field of 10 mA/cm2 and for 20 mA/cm2, the extensive appearance of embrittled regions, such as quasi-cleavage facets, river pattern morphologies, fast fracture surfaces (FFS), teardrop ridges and stair-like features was detected. The stepwise microcracking phenomenon was also observed simultaneously by the appearance of micro-voids with fish eye morphology due to the evacuation of craters with trapped hydrogen (Figure 18). The fast fracture surfaces, the tear-drop ridges, and the quasi cleavage facets on the fractured surfaces of the CTOD tested specimens are indicative of the manifestation of the Hydrogen Enhanced Decohesion mechanism (HEDE). Correspondingly, the microvoid coalescence effect and the stair-like/saw-teeth embrittled surfaces confirm the development of Hydrogen Enhanced Localized Plasticity mechanism (HELP) and Elastic shielding reactions phenomenon. Finally, river patterns, micro-voids with fish eye morphology and stepwise micro-cracking are characteristic features of the Hydrogen Embrittlement effect (HE) and are mainly attributed to the mechanisms of Hydrogen Induced Cracking (HIC), Internal Pressure theory (IP) and the development of hydride phases and blister formations. Micromachines 2020, 11, 430 16 of 20 Micromachines 2020, 11, x 18 of 21

Figure 18. Fractured surfaces after CTOD testing under in situ hydrogen cathodic charging condition atFigure 20 mA 18./cm Fractured2,(a) river surfaces pattern after morphologies, CTOD testing (b) under stair-like in situ embrittled hydrogen surfaces, cathodic (c charging) decoded condition bands and 2 fractureat 20 mA/cm surfaces,, (a) ( driver) micro-void pattern morphologies, coalescence e (ﬀbect,) stair-like (e) quasi-cleavage embrittled surfaces, facets and (c) teardropdecoded bands surfaces and and (ffracture) voids withsurfaces, ﬁsh eye(d) morphologymicro-void coalescence and stepwise effect, micro-cracking. (e) quasi-cleavage facets and teardrop surfaces and (f) voids with fish eye morphology and stepwise micro-cracking. 4. Conclusions Author Contributions: Conceptualization, D.P.; Data curation, P.K.-O.; Formal analysis, H.P.K.; Funding acquisition,The novelty D.E.M.; of theMethodology, results of the H.P.K.; present Project research administration, work is attributed D.E.M.; Software, to the use P.K.-O.; of an environmentally Visualization, friendlyA.S.T., N.M.D. electrolyte and (greenE.C.D; Writing chemistry), – original and thedraft, imposition H.P.K., P.K.-O. of elevated and D.E.M.; cathodic Writing current – review densities & editing, during theH.P.K. in situ and hydrogen P.K.-O. cathodic charging process compared to other research works. In addition to the particularFunding: electrochemical This research received conditions no external imposed funding during the cathodic polarization process of this type

Micromachines 2020, 11, 430 17 of 20 of steel, it is the first time that the critical thermodynamic parameters such as hydrogen diffusion coefficient, the penetration depth of hydrogen cations, and surficial density of blisters and microcracks were determined. Finally, for a cathodic current density field of 20 mA/cm2, it is the first time that CTODel and CTODpl parameters were defined after the in situ hydrogen cathodic charging testing of X65 pipeline steel. The microstructure of X65 pipeline steel consisted of ferritic and bainitic grains with a high volume volume fraction of degenerated pearlite islands. The average grain size of the ferritic-bainitic microstructure was evaluated by Electron Backscatter Diffraction technique (EBSD) equal to 18 µm. By the use of X-ray Diffraction and Electron Backscattered Diffraction Technique it was detected the existence into the metal structure of FeH3 hydrides, after the in situ hydrogen cathodic charging process, both for 10 and 20 mA/cm2 applied current densities. The islands of lower bainite are characterized by augmented dislocation densities leading to the entrapment effect of atomic hydrogen in the crystallographic texture. The degenerated perlite islands, which are characterized by lower dislocation densities as well as the interfacial regions between the ferritic-perlitic regions, which are characterized by low misorientation angle distribution factors, promote the electrodiffusion process and the propagation of the developed microcrack networks

After applying the in situ hydrogen cathodic charging process at 10 and 20 mA/cm2 current • densities, a severe increase in metal surface hardness was detected. This effect is attributed to the interaction between dislocation networks and hydrogen Cottrell atmospheres, to the creation of high volume fraction of hydrogenated vacancies, and the development of interstitial solid solution. With the increment of the cathodic charging current density from 10 to 20 mA/cm2, the diffusion • depth of hydrogen cations into the X65 crystal structure augmented from 50 to 180 µm. The diffusion coefficient of hydrogen atoms into the crystalline structure during the in situ • hydrogen cathodic charging process and slow strain rate bending loading until fracture was icreased linearly from 6.67735 10 11 cm2s 1 to 8.65385 10 10 cm2s 1 by increasing the current × − − × − − density field from 10 to 20 mA/cm2. After the in situ hydrogen cathodic charging process both at 10 and 20 mA/cm2 in the surface • structure of the X65 pipeline steel was detected the microcracking phenomenon around the interfaces of blisters and non-metallic inclusions. The surficial density of microcracks was observed to increase exponentially (from 2 cracks/ cm2 to • 9 microcracks/cm2) with the increment of the required polarization current density during the in situ hydrogen charging process and slow strain rate bending loading. For a 10 mA/cm2 current density field, the maximum crack width was determined at 2.2 µm and the maximum crack length at 2.4 µm. Correspondingly for a current density field of 20 mA/cm2, the maximum crack width was determined at 2.75 µm and the maximum crack length at 3.1 µm. In the surface area of the cathodically charged specimens of X65 pipeline steel for a current • density field of 10 mA/cm2, an increased volume fraction of blisters with dome-type morphology was detected. After the polarization process of the specimens at a current density field of 20 mA/cm2, the development of elongated blisters was observed, with extended microcrack branching networks at their surfaces. Concerning the average growth size of blister formations, it seems to decrease from 180 µm to 110 µm by increasing the required cathodic polarization current density from 10 to 20 mA/cm2. For the X65 steel specimens subjected to CTOD mechanical testing in ambient air, the mechanical • 1/2 values associated with the fracture toughness were determined as follows: KQ = 42.070 MPa m , 2 · J = 280.76 KN/m , CTODel = 0.0061 mm, CTODpl = 1.92 mm. After the in situ hydrogen cathodic charging process at 10 mA/cm2 the main parameters correlated with toughness properties were calculated as follows: K = 35.04 MPa m1/2, J = 202.01 KN/m2, CTOD = 0.0054 mm, CTOD = Q · el pl 1.44 mm. Respectively after the in situ hydrogen cathodic charging procedure at 20 mA/cm2 the Micromachines 2020, 11, 430 18 of 20

above parameters were identified as follows: K = 30.04 MPa m1/2, J = 175.8 KN/m2, CTOD = Q · el 0.0037 mm, CTODpl = 0.98 mm. Consequently, during the transition from the uncharged state of X65 pipeline steel to the • 2 cathodically polarized under a current density field of 10 mA/cm , the parameters KQ, J, CTODel and CTODpl decreased by 16.71%, 28%, 11.48%, and 25%, respectively. During the transition from the charged condition under a current density field of 10 mA/cm2 to that with a current density 2 field of 20 mA/cm , the parameters KQ, J, CTODel and CTODpl decreased by 28.6%, 37.4%, 39.34%, and 48.95%, respectively. The significant drop of the fracture toughness parameters after the in situ hydrogen cathodic • charging process of X65 pipeline steel under a current density field of 10 mA/cm2, is attributed is related to the contribution of the Hydrogen Enhanced Localized Plasticity (HELP), Hydrogen Induced Cracking (HIC) and Elastic Shielding Reactions mechanisms After the in situ hydrogen cathodic charging testing under a polarization field of 20 mA/cm2 the • significant deterioration of the mechanical properties is attributed to the development of a high volume fraction of FeH3 hydrides and the evolution of the Hydrogen Enhanced Decohesion embrittlement mechanism (HEDE). The fractured surfaces of the specimens that had been submitted under CTOD testing both at the • air and hydrogenated environment consisted of the fatigue-precracked area and the bended area. For the specimens subjected to slow strain rate bending loading under the effect of hydrogen cathodic charging process at a current density of 10 and 20 mA/cm2 the extensive appearance of embrittled regions, such as quasi-cleavage facets, river pattern morphologies, fast fracture surfaces (FFS), teardrop ridges and stair-like features was detected.

Author Contributions: Conceptualization, D.P.; Data curation, P.K.-O.; Formal analysis, H.P.K.; Funding acquisition, D.E.M.; Methodology, H.P.K.; Project administration, D.E.M.; Software, P.K.-O.; Visualization, A.S.T., N.M.D. and E.C.D.; Writing – original draft, H.P.K., P.K.-O. and D.E.M.; Writing – review & editing, H.P.K. and P.K.-O. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Shih, D.S.; Robertson, I.M.; Birnbaum, H.K. Hydrogen embrittlement of a titanium: In situ TEM studies. Acta Metall. 1988, 36, 111–124. [CrossRef] 2. Westlake, D.G. A generalized model for hydrogen embrittlement. Amer. Soc. Met. 1969, 62, 1000–1006. 3. Beachem, C.D. A new model for hydrogen assisted cracking. Metall. Trans. 1972, 3, 437–451. 4. Birnbaum, H.; Sofronis, P. Hydrogen-enhanced localized plasticity—A mechanism for hydrogen-related fracture. Mater. Sci. Eng. A 1994, 176, 191–202. [CrossRef] 5. Sofronis, P.; Birnbaum, H.K. Mechanisms of the hydrogen dislocation-impurity interaction—Increasing shear modulus. J. Mech. Phys. Solids 1995, 43, 49–90. [CrossRef] 6. Bastien, P. Inﬂuence de l’ ecrouissage sur le frottement interieur de fer et de l’ acier, charge ou non en hydrogene. C. R. Hebd. Seances Acad. Sci. 1951, 232, 1845–1862. 7. McMahon, C. Hydrogen-induced intergranular fracture of steels. Eng. Fract. Mech. 2001, 68, 773–788. [CrossRef] 8. Fang, B.; Han, E.-H.; Wang, J.; Zhu, Z.; Ke, W. Hydrogen in stress corrosion cracking of X-70 pipeline steels in near-neutral pH solutions. J. Mater. Sci. 2006, 41, 1797–1803. [CrossRef] 9. Liang, P.; Li, X.; Du, C.; Chen, X. Stress corrosion cracking of X80 pipeline steel in simulated alkaline soil solution. Mater. Des. 2009, 30, 1712–1717. [CrossRef] 10. Cheng, Y.F. Analysis of electrochemical hydrogen permeation through X-65 pipeline steel and its implications on pipeline stress corrosion cracking. Int. J. Hydrogen Energy 2007, 32, 1269–1276. [CrossRef] Micromachines 2020, 11, 430 19 of 20

11. Torresislas, A.; Salinasbravo, V.; Albarran, J.; Gonzalezrodriguez, J. Effect of hydrogen on the mechanical properties of X-70 pipeline steel in diluted solutions at different heat treatments. Int. J. Hydrogen Energy 2005, 30, 1317–1322. [CrossRef] 12. Hardie, D.; Charles, E.; Lopez, A. Hydrogen embrittlement of high strength pipeline steels. Corros. Sci. 2006, 48, 4378–4385. [CrossRef] 13. Trasatti, S.P.; Sivieri, E.; Mazza, F. Susceptibility of a X80 steel to hydrogen embrittlement. Mater. Corros. 2005, 56, 111–117. [CrossRef] 14. Wang, R. Effects of hydrogen on the fracture toughness of a X70 pipeline steel. Corros. Sci. 2009, 51, 2803–2810. [CrossRef] 15. Zhang, T.; Chu, W.; Gao, K.; Qiao, L. Study of correlation between hydrogen-induced stress and hydrogen embrittlement. Mater. Sci. Eng. A 2003, 347, 291–299. [CrossRef] 16. Chen, W.; Kania, R.; Worthingham, R.; Van Boven, G. Transgranular crack growth in the pipeline steels exposed to near-neutral pH soil aqueous solutions: The role of hydrogen. Acta Mater. 2009, 57, 6200–6214. [CrossRef] 17. Chen, W.; Sutherby, R.L. Crack Growth Behavior of Pipeline Steel in Near-Neutral pH Soil Environments. Met. Mater. Trans. A 2007, 38, 1260–1268. [CrossRef] 18. Arafin, M.; Szpunar, J. Effect of bainitic microstructure on the susceptibility of pipeline steels to hydrogen induced cracking. Mater. Sci. Eng. A 2011, 528, 4927–4940. [CrossRef] 19. API 1104: Welding of pipelines and related facilities, 19th ed.; American Petroleum Institute: Washington, DC, USA, 1999. 20. American Petroleum Institute. API 579-1: Fitness-for-Service; American Petroleum Institute: Washington, DC, USA, 2007. 21. Mørk, K.; Collberg, L.; Bjømsen, T. Limit State Design in DNV96 Rules for Submarine Pipeline Systems: Background and Project Experience; Society of Petroleum Engineers (SPE): Houston, TX, USA, 1998. 22. British Standards Institution. Guide to Methods for Assessing the Acceptability of Flaws in Metallic Structures; British Standards Institution: London, UK, 2005. 23. Zapffe, C.; Sims, C. Hydrogen embrittlement, internal stress and defects in steel. Trans. AIME 1941, 145, 225–232. 24. Hayden, L.E.; Stalheim, D. ASME B31.12 Hydrogen Piping and Pipeline Code Design Rules and Their Interaction with Pipeline Materials Concerns, Issues and Research. In Proceedings of the ASME 2009 Pressure Vessels and Piping Conference. Volume 1: Codes and Standards, Prague, Czech Republic, 26–30 July 2009; pp. 355–361. 25. Chatzidouros, E.V.; Papazoglou, V.I.; Pantelis, D.I. Hydrogen effect on a low carbon ferritic – bainitic pipeline steel. Int. J. Hydrogen Energy 2014, 39, 18498–18505. [CrossRef] 26. Chatzidouros, E.; Traidia, A.; Devarapalli, R.; Pantelis, D.; Steriotis, T.; Jouiad, M. Effect of hydrogen on fracture toughness properties of a pipeline steel under simulated sour service conditions. Int. J. Hydrogen Energy 2018, 43, 5747–5759. [CrossRef] 27. Jemblie, L.; Olden, V.; Mainçon, P.; Akselsen, O.M. Cohesive zone modelling of hydrogen induced cracking on the interface of clad steel pipes. Int. J. Hydrogen Energy 2017, 42, 28622–28634. [CrossRef] 28. Olden, V.; Alvaro, A.; Akselsen, O.M. Hydrogen diffusion and hydrogen influenced critical stress intensity in an API X70 pipeline steel welded joint e experiments and FE simulations. Int. J. Hydrogen Energy 2012, 37, 11474–11486. [CrossRef] 29. Traidia, A.; Alfano, M.; Lubineau, G.; Duval, S.; Sherik, A. An effective finite element model for the prediction of hydrogen induced cracking in steel pipelines. Int. J. Hydrogen Energy 2012, 37, 16214–16230. [CrossRef]

30. Chong, T.-V.S.; Kumar, S.B.; Lai, M.O.; Loh, W.L. Eﬀects of wet H2S containing environment on mechanical properties of NACE grade FeMn steel pipeline girth welds. Eng. Fract. Mech. 2014, 131, 485–503. [CrossRef] 31. Hejazi, D.; Haq, A.; Yazdipour, N.; Dunne, D.; Calka, A.; Barbaro, F.; Pereloma, E.V. Eﬀect of manganese content and microstructure on the susceptibility of X70 pipeline steel to hydrogen cracking. Mater. Sci. Eng. A 2012, 551, 40–49. [CrossRef] 32. ASTM E1820. Standard Test Method for Measurement of Fracture Toughness; American Society for Testing and Materials: West Conshohocken, PA, USA, 2009. 33. Beltrán, M.; González, J.; Rivas, D.; Hernández, F.; Dorantes, H. On the role of microstructural properties on mechanical behavior of API-X46 steel. Procedia Struct. Integr. 2017, 3, 57–67. [CrossRef] Micromachines 2020, 11, 430 20 of 20

34. Huang, F.; Liu, J.; Deng, Z.; Cheng, J.; Lu, Z.; Li, X. Effect of microstructure and inclusions on hydrogen induced cracking susceptibility and hydrogen trapping efficiency of X120 pipeline steel. Mater. Sci. Eng. A 2010, 527, 6997–7001. [CrossRef] 35. Xue, H.B.; Cheng, Y.F. Hydrogen Permeation and Electrochemical Corrosion Behavior of the X80 Pipeline Steel Weld. J. Mater. Eng. Perform. 2012, 22, 170–175. [CrossRef] 36. Djukic, M.; Zeravcic, V.S.; Bakic, G.; Sedmak, A.; Rajicic, B. Hydrogen damage of steels: A case study and hydrogen embrittlement model. Eng. Fail. Anal. 2015, 58, 485–498. [CrossRef] 37. Singh, R.; Singh, V.; Arora, A.; Mahajan, D.K. In-situ investigations of hydrogen influenced crack initiation and propagation under tensile and low cycle fatigue loadings in RPV steel. J. Nucl. Mater. 2020, 529, 151912. [CrossRef] 38. Djukic, M.; Bakic, G.M.; Zeravcic, V.S.; Sedmak, A.; Rajicic, B. The synergistic action and interplay of hydrogen embrittlement mechanisms in steels and iron: Localized plasticity and decohesion. Eng. Fract. Mech. 2019, 216, 106528. [CrossRef] 39. ASME. ASME B31. Gas Transmission and Distribution Piping Systems; American Society of Mechanical Engineers: New York, NY, USA, 1999. 40. Tetelman, A.; Robertson, W. Direct observation and analysis of crack propagation in iron-3% silicon single crystals. Acta Met. 1963, 11, 415–426. [CrossRef] 41. Elboujdaini, M. Hydrogen-Induced Cracking and Sulfide Stress Cracking; Wiley: Hoboken, NJ, USA, 2011; pp. 183–194. 42. Birnbaum, H.K.; Moody, N.; Thompson, A.W. Hydrogen Effects on Materials Behavior. Min. Met. Mater. Soc. 1990, 12, 639–658. 43. Barrera, O.; Bombac, D.; Chen, Y.; Daff, T.D.; Galindo-Nava, E.; Gong, P.; Haley, D.; Horton, R.; Katzarov, I.; Kermode, J.R.; et al. Understanding and mitigating hydrogen embrittlement of steels: A review of experimental, modelling and design progress from atomistic to continuum. J. Mater. Sci. 2018, 53, 6251–6290. [CrossRef] 44. Stroh, A.N. A theory of the fracture of metals. Adv. Phys. 1957, 6, 418–465. [CrossRef] 45. Oriani, R.A.; Josephic, P.H. Equilibrium aspects of hydrogen-induced cracking of steels. Acta Metall. 1974, 22, 1065–1074. [CrossRef] 46. Oriani, R.; Josephic, P. Equilibrium and kinetic studies of the hydrogen-assisted cracking of steel. Acta Met. 1977, 25, 979–988. [CrossRef] 47. Troiano, A.R.; Hehemann, R.F. Stress Corrosion Cracking of Ferritic and Austenitic Stainless Steels. Hydrogen Embrittlement and Stress Corros. Cracking 1995, 231–248. 48. Campbell, J. The Mechanisms of Metallurgical Failure, On the Origin of Fracture; Butterworth-Heinemann: Birmingham, UK, 2020.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).