Effect of immersion corrosion time on the corrosion performance of welded joints of bimetallic composite pipes

Abstract

In this study, the corrosion behaviour of X65MS/ N08825 composite pipe welded joints during immersion was investigated using an electrochemical workstation and scanning electrochemical microscopy techniques. When the immersion time was more than 24 h, significant corrosion craters appeared in the weld area, with a maximum depth of 7.05 μm, and the corresponding self-corrosion potentials and self-corrosion current densities were −781 mV and 11.82 μA/cm². When the immersion time was extended to 48 h, the depth of the corrosion crater further increased to 15.23 μm, and the self-corrosion potential and self-corrosion current density were −816 mV and 23.73 μA/cm², respectively.

Keywords

bimetallic composite tubes welded joints immersion corrosion electrochemical

Introduction

The continuous exploitation of deep-well oil and gas resources has led to the emergence of fields with high concentrations of Cl⁻, CO₂, sulfide, and other corrosive media. Among these, carbon dioxide (CO₂) and hydrogen sulfide (H₂S) are the primary agents responsible for the corrosion of oil pipelines.^1–4 In the development of high-acid and high-sulfur gas fields, the pressure and temperature along the wellbore direction gradually decrease, causing the solubility of sulfur in hydrogen sulfide to diminish. This results in the continuous precipitation of sulfur. The precipitated sulfur not only undergoes hydrolysis reactions with water but also combines with H₂S, HS⁻, Cl⁻, and other substances to form monomorphic sulfur deposits. These deposits trigger undesirable phenomena, such as corrosive perforation, oil and gas leakage, and other issues.^5,6

Replacing all gathering pipelines with materials that have corrosion-resistant properties would result in a significant waste. To address the issues of corrosion resistance and cost, carbon steel/nickel-based alloy composite pipes, consisting of an outer layer of carbon steel and an inner layer of a corrosion-resistant alloy, meet the highest standards for corrosion resistance, durability, and cost-effectiveness.^7–10 The X65/N08825 bimetallic composite pipe, with N08825 as the inner liner, is a nickel-iron-chromium alloy with excellent corrosion resistance. Consequently, bimetallic composite pipes are widely used in the petroleum, chemical, aerospace, and other industrial fields.^11–13 It is important to note that the fusion welding of such bimetallic composite pipes often results in defects such as segregation of the weld metal, formation of secondary phases, grain expansion, and cracking. These defects can significantly affect the corrosion and corrosion protection properties of the weld.^14,15 Wang, Zhang, and Li et al.^16–18 investigated the corrosion behaviour of X65/Incoloy825 bimetallic composite pipe ring welds in H₂S environments. Their studies found varying degrees of corrosion for the metals at different corrosion times.

In this paper, we investigate the corrosion behaviour of X65/N08825 bimetallic composite pipe welded joints in a sulfur-containing CO₂ corrosive medium and evaluate the changes in their corrosion resistance over time. This study is of great significance for engineering applications and provides a theoretical basis for corrosion protection measures and pipeline maintenance, thereby extending the service life of the composite pipe and ensuring the safe operation of the oil and gas system.

Experimental materials and methods

Experimental materials

The size of the X65MS/N08825 composite pipe is 812.8 mm (22.25 + 2.5) mm. The base and lining materials are X65MS steel pipe and N08825 nickel-based alloy pipe, respectively. The schematic diagram of the composite tube is shown in Figure 1. The composite pipe is produced and supplied by Xi'an Xiangyang Aerospace Materials Co. Table 1 shows the chemical composition of the substrate and liner as measured by X-ray fluorescence spectroscopy (XRF) analysis.

Table 1.

Composition of bimetallic composite pipe.

Element	C	Si	Mn	P	S	Ni	Cr	Cu	Mo	Nb	Ti	Fe
X65MS	0.07	0.3	1.5	–	–	0.004	0.2	–	0.3	0.06	0.02	Bal.
N08825	0.022	0.14	0.47	0.015	0.007	41.14	22.07	1.47	3.13	–	–	–

The ring weld of X65MS/N08825 composite pipe is filled with ERNiCrMo-3 wire of 0.9 mm diameter. The welding parameters are shown in Table 2.

Table 2.

Bimetallic composite pipe welding process parameters.

Weld level	Welding current		Input voltage(V)	Welding speed (mm/min)	Maximum line energy (KJ/mm)
Weld level	Polarities	Peak current (A)	Input voltage(V)	Welding speed (mm/min)	Maximum line energy (KJ/mm)
Brazing	DCEN	140–190	9–16	10–150	1.20
Hot welding	DCEN	140–200	9–16	100–180	1.40
Padding	DCEN	220–330	9–16	100–200	1.90
Tack weld	DCEN	180–220	9–16	80–120	1.50

In order to simulate the service environment containing Cl⁻, CO₂, and sulfides, a mixture of 0.5 mol/L NaCl + 0.5 mol/L NaHCO₃+ 0.25 mol/L Na₂SO₃ was prepared by borrowing the method of Rihan,¹⁹ and the pH of the solution was adjusted to 4 using HCl. The electrode surface of the electrochemical test specimen measured 10 mm ×10 mm and the portion of the specimen other than the electrode surface was shielded with epoxy resin.

Immersion corrosion

Environmental corrosion was simulated according to the ASTM G 31 standard for Metal Immersion Corrosion Testing. Prior to the experiment, the specimens were ground and polished. The experimental corrosion solution consisted of a 3.5 wt% NaCl solution, and the immersion times were 0, 4, 8, 12, 24, and 48 h. Post-corrosion morphology and elemental distribution were characterized using optical microscopy, scanning electron microscopy (SEM), and energy-dispersive X-ray spectroscopy (EDS). A white light interferometer was used to characterize the corrosion morphology in 3D and to measure the size of the corrosion pits at the interface.

Galvanic corrosion

The corrosion resistance of the welded joints at different immersion times was characterized using a Gamry Reference 3000 electrochemical workstation. The experimental solution was a 3.5% NaCl solution, and the experimental temperature was room temperature (25°C). The test specimen had a surface area of 1 cm² and was evaluated using a traditional three-electrode system. The auxiliary electrode was a platinum electrode, the reference electrode was a saturated calomel electrode, and the working electrode was the test specimen. The open-circuit potential was first measured until the curve levelled off, and then the polarization curve was determined. The scanning speed of the polarization test was 0.5 mV/s from the cathode to the anode, with a scanning range of −0.5 V to 0.1 V. To ensure data accuracy, three electrochemical corrosion tests were conducted at each corresponding corrosion time.

Scanning electrochemical corrosion

A CHI 900C scanning electrochemical microscope was used to characterize the corrosion of the surface micro-regions of the welded joints under the same immersion time. The experiments were conducted using a four-electrode system in a 3.5 wt% NaCl corrosion solution. The probe electrode was a platinum ultramicroelectrode, the reference electrode was a saturated calomel electrode, the auxiliary electrode was a 99.99% platinum sheet, and the working electrode was the test sample. The micro-region scanning range was 100 μm along the X direction.

Results and discussion

Microstructure

After the welded joint was corroded with aqua regia, the morphology observed under the scanning electron microscope is shown in Figure 2. Figure 2(b) indicates that the microstructure of the lining metal and the weld is primarily columnar, dendritic, and equiaxed. No coarse-grained zone was observed in the heat-affected zone (HAZ), and the fusion zone is clearly distinguishable. In Figure 2(c), a significant amount of second-phase precipitation is found in the weld, primarily distributed at the grain boundaries.

From Figure 3, it is evident that there are two main forms of second-phase precipitation in the weld. One form is less abundant and has a regular quadrilateral shape, while the other form is more abundant and has an irregular shape. According to the EDS element analysis presented in Table 3, the regular tetragonal precipitates contain a large amount of Nb, with minimal amounts of Ni, Mo, and Cr elements, similar to MC-type NbC primary carbides. The irregularly shaped precipitates have higher Ni and Cr contents, and lower contents of other elements. The ratio of Ni, Cr, and Fe to Mo, Nb, and Ti (66.49%:35.51%≈2:1) is approximately 2:1, suggesting that the irregular phase is Laves (Ni, Cr, Fe)₂(Nb, Mo, Ti).^20,21

Table 3.

EDS results for different types of second phase microstructures (wt.%).

Element	Cr	Mo	Ni	Nb	Ti	Fe	Total
Spectrum 1	8.06	8.40	20.36	57.52	0.21	5.45	100
Spectrum 2	16.98	17.65	44.31	17.52	0.34	5.20	100

Immersion experiment

Corrosion macro-morphology

Figure 4 shows the macro-photomorphology of the welded joints at different immersion times. As can be seen from the figure, the surface of X65MS steel has been completely covered by black corrosion products, making it difficult to distinguish the degree of corrosion. With the increase in immersion time, the degree of corrosion of the weld metal and the liner metal N08825 increases, and the number of corrosion products and corrosion pits also increases. Compared to 4 h and 8 h, the corrosion at 12 h is the most severe. Although the degree of corrosion continues to increase at 24 h and 48 h, the increase is not as significant, indicating that the rate of corrosion slows down after 24 h. Notably, the interface between X65MS steel and the weld and liner metal is more distinct from 4 h to 12 h. When the corrosion time exceeds 24 h, the fusion interface between X65MS steel and the weld and liner metal becomes wider and more diffuse.

Corrosion microscopy

Figure 5 shows the macro-morphology of the weld and X65MS steel corrosion. From the figure, it can be seen that at 4 h of corrosion, the weld area does not exhibit obvious corrosion pits, and the surface remains smooth and intact. In contrast, the surface of X65MS steel is covered with corrosion pits of varying sizes, indicating that the corrosive agent reacts with the X65MS steel surface, forming corrosion products. EDS elemental analysis and XRD physical analysis (Figure 6) reveal that the main corrosion products are FeCO₃ and FeS, which is consistent with findings in the literature.^22,23 As corrosion continues, a dense layer of oxide or other corrosion products forms. However, as corrosion progresses, this dense layer breaks up or falls off, leading to more extensive surface damage. Pitting was observed on the weld surface after 24 h, indicating that the weld metal begins to show corrosion behaviour after 24 h of immersion. When the corrosion time reaches 48 h, the size of the corrosion pits becomes larger, and the weld metal on the left side of the fusion line is aggressively corroded, with the corrosion area expanding to the left. This further indicates that the fusion line at the dissimilar metal interface is a corrosion-prone area.

Figure 7 shows the macro-morphology of corrosion of welded joint lining metal N08825 and X65MS steel under different immersion times. As can be seen from the figure, the degree of corrosion of X65 MS and the trend in Figure 7(e) is consistent with the increase in corrosion time, the more serious the corrosion, the larger the size of the corrosion pits. 48 h of the corrosion products of the X65 MS steel EDS, the content of the C element compared to the increase in Fe elements decreased, indicating that the longer the corrosion time, the corrosion products of the more FeCO₃.

The study of corrosion pit depth is crucial for understanding material corrosion performance. Figure 5 reveals significant corrosion pits in the weld after a 24-h immersion. Utilizing white light interference, the 3D morphology and depth of these pits were meticulously characterized and quantified, as illustrated in Figure 8. Figure 8 shows the appropriate x and y paths for scanning the corrosion pits and obtaining the corrosion pit contour morphology and it can be seen that the maximum depth of the corrosion pits is 7.05 μm after 24 h and 15.23 μm after 48 h.

Polarization curves and impedance spectra

Figure 9 shows the polarization curves and impedance plots of the welded joints of bimetallic composite tubes at different immersion times. From Figure 9(a), it can be seen that the self-corrosion potential of the welded joints without immersion is the highest, and this value gradually decreases with increasing immersion time. The corrosion current densities of the welded joints were obtained by fitting the polarization curves, as shown in Table 4. At 0 h, the self-corrosion current density was 2.44 μA/cm². When the immersion time reached 8 h, the corrosion current density increased to 7.66 μA/cm², which is approximately three times higher than the uncorroded condition. At 48 h of immersion, the corrosion current density further increased to 23.73 μA/cm², which is about ten times higher than the uncorroded condition. Overall, the results indicate that the corrosion rate of the joints increases with increasing immersion time.

Table 4.

Self-corrosion potential and self-corrosion current density of welded joints under different immersion times.

Time(h)	E_corr(mV)	I_corr(μA/cm²)
0	−663	2.44
4	−709	3.62
8	−745	5.66
12	−765	10.62
24	−781	11.82
48	−816	23.73

As shown in Figure 9(b), the capacitive arc radius obtained from the corrosion of the welded joints of bimetallic composite pipes in the simulated actual gas transmission environment becomes smaller with the increase of the corrosion time, indicating that the corrosion-resistant impedance of the welded joints of the bimetallic composite pipelines decreases and the corrosion increases, which is by the law reflected by the polarisation curve.

Scanning electrochemistry of the three-phase region of welded joints

Figure 10 shows the scanning electrochemical microscope (SECM) plots of X65MS steel at different immersion times. As can be seen from the figure, there are no visible electrochemically active spots in the X65MS steel when tested directly without immersion. With the increase in immersion time, the background current of the X65MS steel gradually increases. Notably, at 12 h of immersion, the background current increases significantly. Under the high background currents at 24 h and 48 h, several more prominent current peaks were observed in the detected area, indicating the presence of electrochemically active points. The appearance of these electrochemically active points suggests the onset of microscale etching. When the immersion time reaches 48 h, the peak current of the electrochemically active points in the detected region decreases, which is attributed to the reduced reaction rate of the local dissolution reaction at these points.²⁴

Figure 11 shows the scanned electrochemical properties of the liner metal N08825 at different immersion times. The measured background currents show very little change in the current of the N08825 liner, both before and after immersion, highlighting its excellent corrosion resistance. In particular, from 4 to 48 h of measurements, the background currents showed almost no significant change and the test surfaces lacked electrochemically active spots. These observations confirm the high corrosion resistance of the lining metal N08825, especially the Ni and Cr elements.

Figure 12 shows the scanning electrochemical microscope (SECM) diagrams of the weld metal at different immersion times. From the figure, it can be seen that the background current of the weld metal without immersion corrosion and at 4 h and 8 h is relatively low, with little difference between these times. At 12 h, the background current of the weld metal increased slightly, and electrochemically active spots appeared in the measured area, indicating that pitting corrosion may have occurred on the surface of the weld metal at this immersion time. After 12 h, no electrochemically active points were observed in the measured area. This may be due to a decrease in the reaction rate of the local dissolution reaction at the electrochemically active points, suggesting that the weld metal exhibits self-passivation behaviour under the applied self-corrosion potential, which inhibits the onset and development of pitting corrosion. With increasing immersion time, the weld metal showed a significant increase in background current after 12 h of immersion, indicating a decrease in corrosion resistance. This is consistent with the results of the polarization curves and the electrochemical impedance spectroscopy (EIS) curves measured in the previous section. Overall, the weld metal demonstrates excellent corrosion resistance compared to X65 steel, which is mainly attributed to the excellent corrosion resistance of the nickel element in the ERNiCrMo-3 electrode.

Corrosion mechanism

The corrosion mechanism of composite pipe welded joints is shown in Figure 13. The corrosion mechanism of X65 steel is shown in 13a1. In the initial stage, the Fe²⁺

content in the solution is low, and its corrosion products are mainly generated by solid-phase reactions (1) and (2) to produce FeCO₃ and iron sulfides.²⁵

Fe + {CO}_{2} + H_{2} O \to {FeCO}_{3} + 2 H^{+} + 2 e^{-}

(1)

Fe + H_{2} S \to {FeS}_{1 - x} + {xHS}^{-} + (2 - x) H^{+} + (2 - 2 x) e^{-}

(2)

As the corrosion time increases, the Fe²⁺ content rises and reaches the saturated FeCO₃ and FeS state, and the crystallized FeCO₃ and FeS nucleate and grow up through a precipitation mechanism with equations as in (3) and (4).²⁶

{Fe}^{2 +} + {CO}^{2 -} \leftrightarrow {FeCO}_{3}

(3)

{Fe}^{2 +} + {HS}^{-} \leftrightarrow FeS + H^{+}

(4)

The initially formed corrosion scale is loose and porous (Figure 13a2), and the corrosive ions such as Cl⁻, HCO³⁻, and CO3²⁻ in the corrosive agent pass through the corrosion film layer and react with the substrate in contact. Due to the high permeability of Cl⁻ and its strong interaction with metal cations, Cl⁻ will gather on the surface of the metal, forming an active spot, and corrosion generally starts from the active spot.²⁷ Bubble-shaped corrosion holes and crevices can not only serve as corrosive medium ions to the inward diffusion of the channel but also as dissolved metal ions to the outward diffusion of the channel, thereby accelerating the corrosion rate.

The corrosion mechanism of the weld metal and the nickel-based alloy of the liner is approximately the same, as shown in Figure 13(b). In the corrosive medium, nickel and chromium and other elements in nickel-based alloys will form a layer of passivation on the metal surface (Cr₂O₃, NiO, etc.), which protects the weld alloy and liner metal surface from further oxidation and corrosion. With the increase of corrosion time, Cl⁻ in the corrosion medium gradually destroyed the Cr₂O₃ and NiO film layer, so that the metal is exposed to the corrosion medium, the metal and the corrosion medium in the S²⁻ or HS²⁻ reaction to generate metal sulfide, at the same time, the passivation film in the S²⁻ maybe with the help of the vacancy migration diffusion to the inside of the passivation film, which further accelerates the corrosion of the weld metal. It is worth noting that the corrosion craters in the weld and liner metal appear around the secondary phase, which is mainly due to the depletion of chromium and nickel around the secondary phase. At the same time, the Laves phase is molybdenum-rich, which surely leads to a molybdenum deficiency in the area around it as well. The addition of molybdenum improves the passivation ability of the alloy in reducing media and increases the pitting resistance of the alloy. When both Cr and Mo are deficient, this leads to preferential pitting around the Laves phase. This is in agreement with the findings of Ryan and Sung.^28,29

Conclusion

By studying the corrosion behaviour of X65MS/N08825 bimetallic composite pipe welded joints under different immersion times, the following conclusions were obtained:

The corrosion of the welded joints became more severe with increasing immersion time. The weld corrosion current density was 2.44 μA/cm² at 0 h, 11.82 μA/cm² at 24 h, and when the corrosion time was increased to 48 h, the self-corrosion current density was 23.73 μA/cm².

The maximum depth of the corrosion pits was 7.05 μm after 24 h and 15.23 μm after 48 h, respectively.

Through scanning electrochemistry, it was found that the corrosion of X65MS was the most serious in the three-phase region of the welded joints, in which there was a significant electrochemically active point at 24 h of immersion, and the peak current of the electrochemically active point decreased significantly when it reached 48 h.

Corrosion pits in welds and nickel-based alloys are preferentially formed between dendrites and around secondary phases, and the size of the pits increases as the corrosion time increases.

Footnotes

Data availability

All data and models generated or used during the study appear in the submitted article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Mei Yang

References

Mehta

Jha

Liang

. Corrosion of nickel-based alloys fabricated through additive manufacturing: a review. Progr Add Manufactur 2022; 7: 1257–1273.

Wang

, et al. Research progress on corrosion of oil and gas field gathering pipeline in H₂S-CO²-cl−system. IOP Confere Ser: Earth Environ Sci 2020; 555: 012046.

Song P, Wang W and Jia X. Corrosion study of 80S steel under the coexistence of CO₂ and H₂S. Metals 2022; 12: 1923.

Lazareva

Owen

Vargas

, et al. Investigation of the evolution of an iron carbonate layer and its effect on localized corrosion of X65 carbon steel in CO₂ corrosion environments. Corros Sci 2021; 192: 109849.

Shuxin

Faqin

Xianming

, et al. Failure analysis of the leakage in girth weld of bimetal composite pipe. Eng Failure Anal 2023; 143: 106917.

Lei

Feng

Zhang

, et al. Impact of reversed austenite on the pitting corrosion behavior of super 13Cr martensitic stainless steel. Electr Acta 2016; 191: 640–650.

Jing

Han

. Effect of welding on the corrosion behavior of X65/inconel 625 in simulated solution. Weld World 2018; 62: 363–375.

Abioye

Farayibi

Mccartney

, et al. Effect of carbide dissolution on the corrosion performance of tungsten carbide reinforced inconel 625 wire laser coating. J Mater Process Technol 2016; 231: 89–99.

Wang

Lan

Lei

. Effect of welding method on microstructure and mechanical properties of L360QS/N08825 composite pipe welded joint. Trans Indian Inst Metals 2020; 73: 629–643.

10.

Wang

Lei

B-B

Wang

, et al. Investigations on the crack formation and propagation in the dissimilar pipe welds involving L360QS and N08825. Eng Failure Anal 2015; 58: 56–63.

11.

Wang

Zhang

, et al. Influence of electromagnetic stirring (EMS) on the microstructure and mechanical property of Incoloy825 superalloy. J Manufactu Process 2017; 26: 364–371.

12.

Liu

Feng

Luo

, et al. Effect of special grain boundary on hot corrosion path in Incoloy825 alloy. J Mater Sci 2022; 57: 15560–15580.

13.

Sayyar

Shamanian

Niroumand

, et al. EBSD Observations of microstructural features and mechanical assessment of INCOLOY 825 alloy/AISI 321 stainless steel dissimilar welds. J Manufactur Process 2020; 60: 86–95.

14.

Kangazian

Shamanian

. Micro-texture and corrosion behavior of dissimilar joints of UNS S32750 stainless steel/UNS N08825 Ni-based superalloy. Mater Character 2019; 155: 109802.

15.

Zhang

Han

, et al. Insights into the intergranular corrosion of overlay welded joints of X65-Inconel 625 clad pipe and its relationship to damage penetration. Corros Sci 2019; 160. DOI:https://doi.org/10.1016/j.corsci.2019.108169.

16.

Wang

Zhang

Liu

, et al. Corrosion damage behavior of X65/incoloy 825 welded bimetallic composite pipe in H₂S environment. Front Mater Sci 2021; 8. DOI: https://doi.org/10.3389/fmats.2021.709101

17.

Zhang

Zeng

. Research on TIG welding gap corrosion resistance of X52/825 metallurgical clad pipein H₂S/CO₂ environment. Anti-Corros Method Mater 2019; 66: 412–417.

18.

Liu

, et al. Crevice corrosion of N80 carbon steel in CO²⁻saturated environment containing acetic acid. Corros Sci 2016; 112: 426–437.

19.

Rihan

Zafar

Al-Hadhrami

. A novel emulsion flow loop for investigating the corrosion of X65 steel in emulsions with H2S/CO2. J Mater Eng Perform 2016; 25: 3065–3073.

20.

Cieslak

Headley

Romig

, et al. A melting and solidification study of alloy 625, metall. Trans A 1988; 19: 2319–2331.

21.

Gao

DY-T

Bin

Yuan-jun

, et al. Research progress in evolution of precipitated phases in inconel 625 superalloy. J Mater Eng 2020; 48: 13–22.

22.

Jia

Yang

, et al. CO₂ Corrosion behavior of X70 steel under typical gas–liquid intermittent flow. Metals 2023; 13: 1239.

23.

Barker

Burkle

Charpentier

, et al. A review of iron carbonate (FeCO₃) formation in the oil and gas industry. Corros Sci 2018; 142: 312–341.

24.

Shehzad

Arslan

Rehman

, et al. Corrosion behavior of copper, aluminium, and stainless steel 316L in chicken fat oil based biodiesel-diesel blends. Sustain Energy Technol Assess 2023; 56: 103089.

25.

Gong

Yang

, et al. Corrosion of pipeline steel weld metals containing alloying elements in CO₂/H₂S environment. Gas Sci Eng 2023; 109: 104846.

26.

Cole

Corrigan

Sim

, et al.

Corrosion of pipelines used for CO₂ transport in CCS: is it a real problem?

Int J Greenh Gas Control 2011; 5: 749–756.

27.

Zhang

Zeng

Xiao

, et al. Effect of cl⁻ accumulation on corrosion behavior of steels in H₂S/CO₂ methyldiethanolamine (MDEA) gas sweetening aqueous solution. J Nat Gas Sci Eng 2016; 30: 444–454.

28.

Ryan

Williams

Chater

, et al. Why stainless steel corrodes. Nature 2002; 415: 770–774.

29.

Kim

Hong

M-s

. Effect of metallurgical factors on the pitting corrosion behavior of super austenitic stainless steel weld in an acidic chloride environment. J Mater Res 2017; 32: 1343–1350.