Hydrogen Embrittlement Resistance of Ferritic–Pearlitic Pipeline Steel with Non-Electrochemically Deposited Copper- or Nickel–Phosphorus-Based Coating

Abstract

This work deals with the effects of a non-electrochemically deposited copper- or nickel–phosphorus-based coating on the resulting resistance of traditional X42 grade pipeline steel against hydrogen embrittlement (HE). The susceptibility to HE was determined by the evaluation of the hydrogen embrittlement index (HEI) from the results of conventional room-temperature tensile tests using cylindrical tensile specimens. Altogether, three individual material systems were studied, namely uncoated steel (X42) and two coated steels, specifically with either a copper-based coating (X42_Cu) or a nickel–phosphorus-based coating (X42_Ni-P). The HEI values were calculated as relative changes in individual mechanical properties corresponding to the non-hydrogenated and electrochemically hydrogen-precharged tensile test conditions. Both applied coatings considerably improved the hydrogen embrittlement resistance of the investigated steel in terms of decreasing the HEI values related to the changes in the yield stress, ultimate tensile strength, and reduction of area. In contrast, the hydrogenation of both coated systems had detrimental effects on the value of total elongation, which resulted in an increase in the corresponding HEI value. This behavior was likely related to the earlier onset of necking during tensile straining due to strain localizations induced by the coatings’ surface imperfections. The findings from fractographic observations indicated that both studied coatings acted like protective barriers against hydrogen permeation. However, the surface quality in terms of pores and other superficial defects in the considered coatings remains a challenging issue.

1. Introduction

Global warming is a serious climate change due to emissions of greenhouse gases, especially carbon dioxide and methane, originating from a huge variety of anthropogenic activities, especially those related to industry and agriculture. The growth of both the world economy and human population causes a continuously increasing demand for energy production, mainly from fossil fuels, inevitably increasing the overall carbon footprint on Earth. In order to achieve both the sustainable socioeconomic development and the protection of the global environment, several efforts are being put into research and innovation aimed at the decarbonization of the economy using renewable energy sources, e.g., solar, wind, hydropower, biomass, etc. [1,2,3]. The electricity produced from these renewables can either be distributed in an energy network or further used for electrolytic water splitting to produce so-called “green hydrogen”, i.e., an ecological, carbon-free energy carrier that has been receiving increasing attention in recent years [4,5,6,7,8,9,10]. The gradual transformation of the current fossil-fuel-driven industry to a hydrogen economy brings new challenges related to the reliability and safety of hydrogen production, storage, and distribution [8,9,10,11,12,13,14,15,16].

In the case of any applications of structural metallic materials exposed to hydrogen, one has to take into account their specific resistance against degradation by hydrogen embrittlement (HE). This type of material deterioration is related to the environmentally induced degradation of the deformation properties of metallic materials by the action of free atomic hydrogen capable of diffusion into the metal lattice even at room temperature [17,18,19,20]. This issue is highly challenging because the idea of blending natural gas with hydrogen in transportation pipeline infrastructure is increasingly coming to the forefront [13,16,21,22,23,24,25,26,27]. It is generally accepted that although many HE mechanisms have been suggested and described, e.g., hydrogen-enhanced decohesion (HEDE), hydrogen-enhanced localized plasticity (HELP), adsorption-induced dislocation emission (AIDE), and hydrogen-enhanced strain-induced vacancies (HESIV) [18,19,20,21,22,23,24,25,26,27], there is no single universal mechanism that can explain all forms of hydrogen-related degradation of constructional steels and alloys.

With regard to hydrogen effects on mechanical behavior, a crucial role influencing a material’s sensitivity to HE is played by “hydrogen trapping” at various microstructural and substructural sites such as precipitate/matrix interfaces, inclusion/matrix interfaces, vacancies, dislocations, substitution atoms, etc., which exhibit internal stresses in their lattice neighborhoods [18,19,20,28,29,30,31,32]. Therefore, even metallic materials that are chemically the same may exhibit rather different HE resistances related to various processing and operational conditions, i.e., specific material states having significant effects on their resulting microstructures and properties. Some researchers divide HE into reversible and irreversible categories [33,34,35]. The reversibility of HE is understood as the restoration of metal plasticity as a result of hydrogen desorption from the metal during aging at room temperature or as a result of annealing or tempering in air or in a vacuum. If there are phenomena associated with HE that cannot be eliminated, then the HE is called irreversible. Such irreversible phenomena include bubbles, flocks, and cracks in the heat-affected zone during welding [34,35,36].

In general, it is well known that metallic materials with a body-centered cubic (BCC) crystal structure possess a higher HE susceptibility due to their lower hydrogen solubility and higher hydrogen diffusivity compared with the metals and alloys with a face-centered cubic (FCC) crystal structure showing the opposite hydrogen characteristics, i.e., higher hydrogen solubility and lower hydrogen diffusivity [37,38,39,40]. Nevertheless, for economic reasons, long-distance natural gas transportation pipelines are constructed from lower-cost BCC-structured carbon steels or low-alloy steels [41,42,43,44,45,46,47,48]. Thus, the feasibility of blending natural gas with hydrogen will strongly depend on the HE resistance and hydrogen compatibility of these materials. One of the options for suppressing the HE of metallic materials is the functional modification of their surfaces, e.g., by creating protective coatings, layers, gradient structures, etc. Promising surface barriers against hydrogen permeation include various ceramics (e.g., carbides, oxides, nitrides), graphene, polymers, and other types of coatings [49,50]. Some metal-based materials such as Pt, Cu, and so-called electroless nickel (chemically Ni–P) can be used as the protective layers for reducing the penetration of hydrogen in steel and other alloys [20,51]. Recently, Samanta et al. [52] reported their results about the development of an amorphous Ni–P coating over API X70 steel for hydrogen barrier application. It was found that the amorphous Ni–P coating displayed a slower and delayed hydrogen permeation compared with crystalline Ni-electroplated steel at similar coating thickness levels. Excellent resistance to hydrogen permeation as well as corrosion in a chloride environment has been attributed to the amorphous structure of the Ni–P coating [52]. Biggio et al. [53] compiled a comprehensive review on Ni–P coatings as hydrogen permeation barriers. Their work emphasized the key advantages of the potential application of electroless Ni–P coatings in existing pipelines, such as the simplicity of production and the possibility of achieving a homogeneous coating, regardless of the geometry of the substrate. However, it has also been admitted that further research would be necessary in order to optimize the performance of an electroless Ni–P coating with a view toward future applications in hydrogen distribution lines [53]. The early work of Chen and Wu [54] studied the hydrogen diffusion through copper-plated AISI 4140 steels. The study revealed that copper plating led to lower values of the hydrogen permeation rate and effective diffusivity due to its low hydrogen absorption rate. The study of Michler and Naumann [55] investigated a huge variety of various coatings including an electroplated Cu and electroless Ni–P coating in terms of possible HE reduction. Their results indicated that none of the coatings significantly improved the tensile ductility in the hydrogen atmosphere compared to the untreated specimens.

Our present work focuses on two approaches aimed at increasing HE resistance of X42 pipeline steel via the application of non-electrochemically deposited Cu- and Ni–P-based coatings with anticipated protective barrier effects against HE.

2. Materials and Methods

Traditional X42 grade ferritic–pearlitic pipeline steel was used in the current study as the experimental material. The chemical composition of the as-received hot-rolled pipe (114.3 mm outer diameter and 13.49 mm wall thickness) is given in

Table 1. Chemical composition of investigated heat of X42 grade pipeline steel [wt.%].

C	N	Mn	Si	P	S	Cu	Cr	Mo	V	Ni	Al	Sn	Fe
0.16	0.009	0.51	0.24	0.014	0.01	0.19	0.09	0.02	0.007	0.08	0.027	0.012	Balance

.

For performing room-temperature tensile tests, cylindrical tensile specimens with a standard sanded surface roughness of 0.6 µm were prepared by conventional lath machining (Figure 1).

Three material systems, namely the uncoated steel (X42) and two coated steels, specifically with either a copper-based coating (X42_Cu) or a nickel–phosphorus-based coating (X42_Ni-P) were investigated. Electroless plating of clean, degreased tensile test specimens with copper was conducted by immersing them into a solution of H₂O (1000 mL) with CuSO₄ (50 g) and concentrated H₂SO₄ (50 g) for 60 s. Afterwards, the prepared tensile test specimens with the Cu-based coating were rinsed with water and dried within a stream of warm air. The average thickness of a prepared Cu-based coating was 1.7 µm. Non-electrochemical deposition of clean, degreased tensile test specimens with the Ni–P-based coating was carried out by their immersion into a warm solution of H₂O (1000 mL) with (C₃H₅O₃)₂Ni (50 g) and NaH₂PO₂. H₂O (15 g) constantly kept at 90 °C for 1.5 h. Afterwards, the prepared tensile test specimens with the Ni–P-based coating were rinsed with water and finally dried in a stream of warm air. The average thickness of a prepared Ni–P-based coating was 8.8 µm.

The resistance of individual material systems against HE was examined by conducting conventional uniaxial tensile tests of prepared tensile test specimens in an initial non-hydrogenated material condition, and separately in an electrochemically hydrogen-precharged material condition. Thus, prior to tensile testing, electrochemical hydrogen precharging of tensile test specimens of individual material systems, i.e., the uncoated X42 material (X42), the coated X42 material with the Cu-based coating (X42_Cu) and the coated X42 material with the Ni–P-based coating (X42_Ni-P), was carried out at room temperature employing a potentiostat/galvanostat model 173 (Princeton Applied Research, Oak Ridge, TN, USA). The electrolytic solution used contained 1 M HCl and 0.1 N N₂H₆SO₄. The cathodic hydrogen precharging was performed at a current density of 200 A/m². After 24 h of hydrogen charging, the hydrogen-precharged tensile specimens were stored in a thermal insulating flask containing liquid nitrogen prior to subsequent ex situ tensile testing. Room-temperature tensile testing was carried out using the universal testing machine TIRATEST 2300 (TIRA GmbH, Schalkau, Germany) at a cross-head speed of 0.2 mm/min in conformity with standard ISO 6892-1:2019 [56].

The determination of HE resistance was performed according to the following procedure. For all studied material systems, three individual test specimens were tensile tested for both the non-hydrogenated and hydrogen-precharged material conditions. After performing room-temperature tensile tests, average values of yield stress (Re—here estimated as 0.2% proof stress, i.e., Rp_0.2), ultimate tensile strength (Rm), total elongation (A), and reduction of area (Z) were evaluated for both the non-hydrogenated and hydrogen-precharged material states. It is generally accepted that the hydrogen embrittlement reduces the mechanical properties, including the ductility, strength and toughness of susceptible metals and alloys [20,34]. Thus, it is possible to evaluate the hydrogen embrittlement effects using the hydrogen embrittlement index determined from relative changes in individual mechanical properties according to the following equation:

$$\mathrm{H}\mathrm{E}\mathrm{I}\mathrm{X}={\displaystyle \frac{{\mathrm{X}}_0-{\mathrm{X}}_{\mathrm{H}}}{{\mathrm{X}}_0}} ·100\mathrm{\%}$$

(1)

where X₀ is the average value of an individual mechanical property, e.g., Re, Rm, A, or Z of broken tensile test specimens in a non-hydrogenated material state and X_H is the average value of the same mechanical property of broken tensile test specimens in a hydrogen-precharged material state. The microstructural observations of the studied material systems were performed using a light-optical microscope (LOM) OLYMPUS GX71 (Olympus Corporation, Tokyo, Japan) and microstructural and fracture surface observations of fractured tensile test specimens were carried out employing a scanning electron microscope (SEM) JEOL JSM-7000F (Jeol Ltd., Tokyo, Japan) with the energy-dispersive X-ray (EDX) analyzer INCA X-sight model 7557 (Oxford Instruments, Abingdon, Oxfordshire, UK). Standard metallographic procedures (i.e., wet grinding, polishing and etching) were used for the preparation of metallographic samples. The etching was performed in 3% Nital solution (3% HNO₃ in CH₃CH₂OH). Quantitative microstructural characterization (i.e., mean grain sizes of microstructural constituents and their area fractions) was performed using software ImageJ (version 1.46, National Institutes of Health, Bethesda, MD, USA). The mean grain sizes of ferrite and pearlite were calculated as the Feret mean diameter [57]. Theoretical prediction and experimental determination of the phase composition of the studied material systems were performed using thermodynamic software Thermo-Calc, version S (Thermo-Calc AB, Solna, Sweden), employing thermodynamic database TCFE6 and X-ray diffraction (XRD) using a Philips X’Pert Pro diffractometer (Panalytical B.V., Almelo, The Netherlands) in Bragg–Brentano geometry with Co-Kα radiation, respectively. The recorded XRD patterns were evaluated for the phase identification of the individual phases using XPert HighScore Plus software (Version: 2.0, PANalytcal B.V., Almelo, The Netherlands) using the PDF-2 database (The International Centre for Diffraction Data, Philadelphia, PA, USA). The expected phases, i.e., ferrite, Fe₃C carbide and copper, were identified on recorded XRD patterns by matching with reference patterns ICDD 00-006-0696, ICDD 01-072-1110 and ICDD 00-004-0836, respectively. This refinement was focused purely on the qualitative identification of the present phases. Morphological observations and elemental chemical microanalyses of coated test specimens were performed by employing a scanning electron microscope Tescan Vega-3 LMU (TESCAN Brno, s.r.o., Brno, Czech Republic) and an EDX spectrometer Bruker XFlash Detector 410-M (Bruker Nano GmbH, Berlin, Germany), respectively.

3. Results and Discussion

3.1. Microstructure and Phase Analyses

Figure 2 shows the overall and detailed microstructures of studied X42 pipeline steel in rolling direction.

The light-optical micrographs (Figure 2a,b) show that the microstructure of the investigated material is composed of a polygonal grain structure containing two microstructural constituents, namely the ferrite (light grains) and pearlite (dark grains). According to image analyses performed, the average grain sizes in terms of Feret diameter are 23.9 µm and 19.4 µm for ferrite and pearlite, respectively. The area fractions of individual microstructural constituents are 81.6% ferrite and 18.4% pearlite. A detailed view of the pearlitic lamellar structure is shown in Figure 2c.

The material phase composition was experimentally evaluated by XRD crystallographic phase analysis and theoretically computed by thermodynamic calculation (Figure 3).

It has been indicated by XRD measurement that the studied material is formed by a BCC-structured ferritic matrix and the secondary precipitation of orthorhombic Fe₃C carbide (Figure 3a). Thermodynamic calculation (Figure 3b) shows the phase composition of studied steel depending on temperature. In the detailed view (Figure 3c), it can be seen that besides Fe₃C carbide, additional minor phases such as AlN, MnS, and elementary Cu precipitates with very low phase fractions are also computationally predicted to be stable in the studied material. However, the amount of these minor phases is far below the XRD detection limit.

Figure 4 shows scanning-electron micrographs of the studied non-electrochemically coated material systems with corresponding EDX chemical microanalyses. The results of elemental chemical microanalyses performed indicated a major presence of copper (87.3 at.% Cu) and a certain amount of iron (12.7 at.% Fe) by analyzing the Cu-based coating, whereas the Ni–P-based coating contained a considerable amount of phosphorus (18.6 at.% P) besides the major content of nickel (81.4 at.% Ni).

The iron content indicated by the EDX analysis of the Cu-based coating (Figure 4a) originates from the iron-based solid solution matrix of the X42 steel substrate beneath the coating due to its very low thickness at around 1.7 µm. On the other hand, the phosphorus content within the Ni–P-based coating (Figure 4b) is an actual part of the coating prepared from the nickel and phosphorus, containing ingredients specified in Section 2.

Both the coated material systems (X42_Cu and X42_Ni-P) were also subjected to XRD crystallographic phase analyses (see Figure 5).

The results of XRD crystallographic phase analyses show that the Cu-based coating is formed of pure crystalline FCC copper, whereas the Ni–P-based coating is of an amorphous nature as also widely reported in numerous sources in the literature, e.g., in [52,58,59,60]. The identified BCC ferrite reflections originate from the iron-based solid solution matrix of the X42 steel substrate beneath both coatings.

3.2. Tensile Properties and HE Resistance

Figure 6 shows representative engineering stress–strain curves for the studied material systems (i.e., X42, X42_Cu, X42_Ni-P) in non-hydrogenated and hydrogen-precharged material conditions.

It is clearly visible that all studied material systems exhibit pronounced yielding behavior, including the so-called upper yield stress (Re_H), lower yield stress (Re_L) and Lüders bands. The observed plastic instabilities are generally ascribed to overcoming the pinning forces of interstitial atoms (i.e., carbon and nitrogen) in so-called Cottrell atmospheres around moving dislocations at the onset of plastic deformation [61]. Figure 7 depicts the graphical summary of average values of all mechanical properties for investigated material systems, evaluated from the performed room-temperature tensile tests. Figure 7a summarizes the average strength properties, whereas Figure 7b summarizes average deformation properties.

The average values of mechanical properties for both the non-hydrogenated and hydrogen-precharged material systems (Figure 7) were used for the evaluation of the hydrogen embrittlement index according to Equation (1). The corresponding results are summarized in Figure 8.

From Figure 8, it is clear that the applied coatings improved the hydrogen embrittlement resistance of the investigated steel in terms of the decreasing of HEI values related to the changes in Re, Rm, and Z. For the uncoated X42 steel, both strength properties (Re and Rm) experienced a decrease due to hydrogen embrittlement resulting in positive values of HEI_Re and HEI_Rm, whereas the application of both coatings (Cu and Ni–P) caused increased values of strength properties of hydrogenated specimens which consequently resulted in negative HEI_Re and HEI_Rm values. A similar improvement of hydrogen embrittlement resistance for both coated material systems was observed in terms of increasing Z values of hydrogenated material states which finally resulted in a decrease in HEI_Z values. In contrast, the hydrogenation of both coated systems had decreasing effects on corresponding A values which resulted in increasing HEI_A values. The observed contradiction between the HEI_A and HEI_Z values represents rather non-typical behavior and cannot be fully understood at this point of investigation. One of the possible explanations may be related to the structure of the coatings themselves. As already shown in Figure 4, the applied coatings do not possess completely smooth and defect-free surface morphologies. On the contrary, there are various visible superficial imperfections that may serve as preferential degradation locations during hydrogen charging and afterwards as failure nucleation sites during tensile loading. Although the used coatings resulted in notable the improvement of Z values of hydrogenated tensile test specimens, the A values experienced just the opposite tendency, i.e., a clear decrease in A values of hydrogenated tensile test specimens. Consequently, in comparison with the uncoated steel, the HEI_Z values for coated material systems possess a clear decrease, whereas the HEI_A values experience a clear increase. These results may likely be ascribed to the surface degradation of coated tensile test specimens during the performed electrochemical hydrogenation, which affects the resulting deformation behavior. Indeed, Zhou et al. [62] and Popov [63] have reported that cathodic charging initiated hydrogen blistering and hydrogen-induced cracking in Ni–P coatings. Moreover, Gathimba and Kitane [64] reported that the surface degradation of tensile test specimens strongly deteriorated ductility in terms of the total elongation values of structural steels due to the earlier onset of necking during tensile straining. In fact, both parameters, i.e., total elongation (A) and reduction of area (Z), represent the measure of ductility (i.e., plasticity). Thus, both of these measures of ductility are applicable and useful for the characterization of the material deformation ability. Both A and Z are usually called deformation properties as also presented in Figure 7b. Which of those deformation properties can be more relevant to material selection in terms of a measure of ductility often depends on the material application. In addition, A is gauge length dependent, while this issue is not associated with Z. Due to the ease of practical measurement, A and Rm are useful for confirming the metallic materials’ conformity with corresponding material standards, whereas Z is more relevant for the characterization of failure resistance as it is directly related to the deformation required to produce fracture. Thus, the limiting factor in the design of a component is often more related to Z, especially in applications where the fracture resistance of a material is more relevant than its withstanding large dimensional changes. In the case of pipeline infrastructure, where fatigue resistance is a very important material property due to cyclic loading caused by gas pressure fluctuations, the Z parameter would be a highly relevant measure of ductility. Thus, regardless of the observed opposite tendencies of Z vs. A behaviors, the obtained results indicated that the applied coatings acted like a hydrogen permeation barrier suppressing the hydrogen entrance into the steel substrate. This idea has been supported by fractographic observations of broken tensile specimens in the subsequent Section.

3.3. Fractographic Characterization

Fracture surfaces of the individual studied material systems (uncoated X42 steel, X42 steel with a Cu-based coating and X42 steel with a Ni–P-based coating) in the selected material conditions with respect to hydrogen-precharging are shown in Figure 9.

The fracture surface of the uncoated and non-hydrogenated X42 steel tensile specimen after the room-temperature tensile test shows a completely ductile dimple fracture micro-mechanism (Figure 9a). In contrast, the uncoated and hydrogen-precharged steel shows a clear transition to a mixed fracture micro-mechanism consisting of both the ductile dimple tearing and brittle transgranular cleavage, exhibiting a typical hydrogen-related “fish-eye” morphological appearance with central non-metallic inclusion (Figure 9b). On the other hand, both coated material systems (X42_Cu and X42_Ni-P) with hydrogen precharging exhibit a fully ductile dimple tearing fracture micro-mechanism (Figure 9c,d). The results of fractographic analyses showed that both coated material systems exposed to hydrogen precharging exhibited similar ductile dimple fracture behavior like the original non-hydrogenated X42 base material without a coating. This observation fairly supported the results of room-temperature tensile tests carried out in non-hydrogenated and hydrogen-precharged material conditions. Thus, obtained findings of the present investigation clearly indicated that the applied electroless plating of copper- and nickel–phosphorus-based coatings may significantly reduce the susceptibility of ferritic–pearlitic pipeline steel to hydrogen embrittlement. According to our observations, both coatings withstood the applied electrolytic hydrogen charging prior to subsequent tensile tests of the coated tensile specimens. However, during the tensile tests, both coatings gradually peeled off from the tensile specimens due to the differing deformation ability between the surface coatings and the steel body of the tested tensile specimens. Thus, the bonding strength between the coatings and the steel matrix during tensile loading was rather poor. Figure 10 shows a profile view on the studied coated material systems after their electrolytic hydrogen charging and tensile testing.

These observations clearly indicated decohesion at the coating/matrix interfaces in both studied material systems. Thus, there is a need for further continuation of research efforts aimed at coating/matrix bonding strength improvements. However, in terms of current hydrogen embrittlement research, the obtained findings indicated that both studied coatings acted like protective barriers against hydrogen permeation into the studied steel during its electrolytic hydrogen charging. Moreover, further research on these material systems is necessary with respect to designing more focused experiments closer simulating real operational conditions of gas pipeline infrastructure (e.g., performing gaseous hydrogen charging) and studying the durability of the coatings in long-term conditions, taking into account complex degradation phenomena (e.g., corrosion, wear) besides the considered hydrogen-related material deterioration.

4. Conclusions

In this work, the hydrogen embrittlement resistance of X42 grade ferritic–pearlitic pipeline steel with non-electrochemically plated copper- and nickel–phosphorus-based coatings was investigated. The obtained results can be summarized in the following conclusions:

• The electroless plating procedures used resulted in chemically stable copper- and nickel–phosphorus-based coatings on the surface of the X42 steel base material. The Cu-based coating had a fully crystalline lattice structure, whereas the Ni–P-based coating was of an amorphous nature.
• The room-temperature tensile properties of the X42 material in the non-hydrogenated material condition were unaffected by the coatings used. The thickness of the studied coatings did not play a role in the view of the HE resistance of the material coating systems studied under the used electrolytic hydrogenation conditions.
• Both applied coatings considerably improved the HEI values related to the changes in the yield stress, ultimate tensile strength, and reduction of area. On the contrary, the HEI value related to total elongation deteriorated, which was ascribed to the earlier onset of necking during tensile straining due to observed surface imperfections within the coatings.
• Based on the obtained results of tensile tests and related fractographic observations, it is concluded that the studied Cu- and Ni–P-based coatings served as hydrogen permeation barriers, reducing the hydrogen migration into the steel matrix beneath the coatings. Nevertheless, further research efforts are needed in terms of the surface quality improvement of the coatings under consideration.