Passivation and Corrosion Behavior of Modified S13Cr Stainless Steel in Ultra-high Temperature Geothermal Fluid

This research aims to investigate the passivation and corrosion behaviour of modified S13Cr stainless steel (SS) in ultra-high temperature geothermal fluids. In this study, S13Cr SS before and after modified were both immersed in a simulated geothermal fluid environment with a temperature of 210°C and CO2 pressure of 3 MPa for 120 h. The results show that the modified S13Cr SS had smaller grain size and lower reverse austenite content, and exhibited higher transpassive potential and lower passive current density in the ultra-high temperature environment. After 120 h of immersion, the passivation film of the modified 13Cr SS was completely dissolved, and a corrosion product film mainly composed of FeCO3 and FeCr2O4 formed with localized corrosion occurring. Moreover, a Ni-rich barrier layer formed at the interface between the inner layer of the product film and the substrate, which hindered the penetration of the corrosive medium. Additionally, the residual MoO2 in the product film played a stabilizing role. Overall, the corrosion resistance of the modified S13Cr SS in ultra-high temperature geothermal fluids is improved.


Introduction
Currently, renewable geothermal resources have shown great potential for application due to its wide distribution, low CO2 emissions and other characteristics [1] [2].However, the aggressive geothermal extraction environments with high temperature (>150 ℃), high mineralization (up to 155000 ppm Cl -) and CO2-containing acidic gases greatly enhance the corrosion risk of downhole materials [1][3] [4].At the same time, S13Cr SS is widely used in high-temperature oil and gas well casings in CO2containing environments due to its excellent resistance to CO2 corrosion, mechanical properties, and relatively lower Cr and Ni alloy element content compared to austenitic stainless steel [3] [5].Consequently, the passivation and corrosion behavior of S13Cr SS in high-temperature and highpressure production fluid environments have become a current research focus.It is well known that the excellent corrosion resistance of SS is mainly attributed to the formation of a nanoscale passive film on the surface, which provides a protective barrier for the substrate and inhibits direct contact between the corrosive medium and the substrate [6] [7].However, the protective properties of the passive film is greatly influenced by the environmental temperature.Studies have shown that in a CO2-Cl -environment at 90°C, the main component of the passivation film on 13Cr SS is Cr2O3, while at 150°C, the film contains both Cr2O3 and FeCO3 [8].Further investigations have reported that the surface film on S13Cr SS is composed of Cr2O3 and Cr(OH)3 under the conditions of 180 ℃ and 3.8 MPa CO2 partial pressure.However, the film exhibits increased defect density and reduced stability at this condition [9].Limited research has been conducted on temperatures above 200°C, with Yue et al. [10] finding that at 200°C, the passivation film is completely ineffective, and a corrosion product film composed mainly of FeCr2O4 and FeCO3 provides protection to the substrate.The above research indicates that S13Cr SS exhibits good applicability in pure CO2 environments below 200°C.However, its applicability significantly decreases in ultra-high temperature environments (>210°C).Thus, it is necessary to take corresponding measures to improve the applicability of S13Cr SS in ultra-high temperature environments.Currently, there are two main approaches for improving the corrosion resistance of materials.The first approach involves adjusting the heat treatment process to improve the uniformity of the structure.However, due to the influence of the structure and chemical composition, the improvement in corrosion resistance through this method is limited.The second approach is to increase the content of corrosionresistant alloying elements such as Cr, Ni, and Mo in the steel to achieve modification.S13Cr SS is an example of this, as it is based on 13Cr SS with reduced carbon content and the addition of 2% Mo and 5% Ni, which significantly improves the material's corrosion resistance [5] [11].However, there is still limited research on the corrosion behavior of 13Cr SS in ultra-high temperature environments, resulting in a lack of clarity regarding the mechanisms and effects of Ni and Mo elements in improving the corrosion resistance of stainless steel.Therefore, in order to improve the corrosion resistance of S13Cr SS in ultra-high temperature geothermal fluid environments and clarify the changes in its passivation and corrosion behaviour, this study modified the S13Cr SS by increasing the content of Ni and Mo elements.Microstructure characterization, in-situ electrochemical testing, and surface analysis were used to systematically investigate the electrochemical behaviour and corrosion product film characteristics of the modified S13Cr SS at 210 ℃.The mechanism for the improved corrosion resistance of the modified S13Cr SS was proposed, providing theoretical support for the development of new high-temperature corrosionresistant pipe materials.

Materials
The test materials were selected S13Cr SS and modified S13Cr SS (M-S13Cr), whose chemical compositions were shown in Table 1.The materials were processed into corrosion samples with dimensions of 40 mm×13 mm×3 mm for corrosion simulation tests and cylindrical specimens with a diameter of 1 cm for electrochemical tests by wire cutting machine.Before the test, the surface of the samples was smoothed using SiC sandpaper with a roughness of 120, 360, 800 grit, then dripped in alcohol and acetone, dried with cold air and exposed to the air for 12 h to fully passivate the surface.Then, the weight before the test was weighed and recorded, and finally placed in a dryer for future use.The test solution was a simulated geothermal fluid prepared with deionized water and analytical pure chemicals, with soluble ion contents of 100.7 g/L Na + +K + , 0.5 g/L Ca 2+ , 155 g/L Cl -, 0.3 g/L SO4 The specimens after grinding and polishing were eroded using aqua regia (HCl: HNO3=3:1), and the microstructure of S13Cr and M-S13Cr was observed under a metallographic microscope, and then conduct EBSD testing on the vibration polished sample with an accelerated voltage of 23 kV and a scanning step of 0.125 µm.

Electrochemical test
The high-temperature and high-pressure (HTHP) electrochemical tests were conducted using a Gamary 1010E electrochemical workstation in a 3 L HTHP electrochemical autoclave equipped with a threeelectrode system.The working electrodes were cylindrical specimens of S13Cr and M-S13Cr with an area of 0.785 cm 2 after sealing.The reference electrode was an external pressure-balanced Ag/AgCl high-temperature reference electrode, and the counter electrode was a 2 cm× 2 cm platinum sheet.The experiment temperature was 210 ℃.After the open circuit potential (OCP) stabilized, electrochemical impedance spectroscopy (EIS) and cyclic polarization curve measurements were performed in sequence with an amplitude of ±5 mV.The frequency range for EIS measurements was 10 -2 to 10 6 Hz, and the cyclic polarization test scans from an initial potential of -150 mV (vs OCP) to a final potential of +1500 mV, with a scan rate of 1 mV/s and a return scan current density of 5 mA/cm 2 .All electrode potentials measured by the high-temperature reference electrode in this study were converted into standard hydrogen electrodes (SHE) according to equation (1) [4][9]: ) Where ESHE was the potential (V) relative to the standard hydrogen electrode, and Eobs was the measured potential relative to the high-temperature reference electrode.T was the test temperature (℃), and t was room temperature (25 ℃).

Corrosion simulation test
Corrosion simulation tests were conducted in a 3 L HTHP autoclave.Prior to the test, a set of samples was fixed on a polyetheretherketone (PEEK) fixture inside the autoclave.Then, 1.5 L of simulated geothermal fluid, which had been deoxygenated with high-purity N2 for 12 h, was poured into the autoclave.The autoclave was immediately sealed, and another 2 h of deoxygenation with high-purity N2 was carried out to remove any air introduced during the loading process.Subsequently, heating was started, and when the temperature inside the autoclave reached 60 ℃, 3 MPa of high-purity CO2 was introduced into the autoclave.The experiment lasted for 120 h.After the completion of the test period, the samples were taken out from the autoclave, and corrosion products were removed using an acid washing solution (prepared by mixing 1 L of hydrochloric acid, 20 g of antimony trioxide, and 50 g of tin chloride) and weighed.Additionally, the surface profile was measured using a three-dimensional profilometer.Then, the average corrosion rate (CR, mm/a) and local corrosion rate (Cpit, mm/a) were calculated separately according to equations ( 2) and (3).
Where ΔW was the weight loss of the sample before and after the test, g; A was the exposed area of the sample, cm 2 ; ρ was the material density, g/cm 3 ; t was the test period, h; h was the depth of the corrosion pit, μm.

Characterization of corrosion products
The surface and cross-sectional microstructures of the corrosion products on the sample surface were observed using a scanning electron microscope (SEM).The elemental distribution of the corroded samples was analyzed using an energy-dispersive X-ray spectrometer (EDS).The phase composition of the corrosion products was analyzed using a copper target X-ray diffractometer (XRD) operating at 40 kV and 44 mA, as well as an Mg target X-ray photoelectron spectrometer (XPS) with hv=1253.6 eV.In the XPS spectrum analysis, residual carbon correction was performed with the C1s peak at 284.8 eV.

The microstructure of S13Cr SS before and after modification
Figure 1 shows detailed information about the microstructure of S13Cr SS before and after modification.It can be observed that the microstructure of S13Cr SS and M-S13Cr SS was both lath martensite with a small amount of reverse austenite present at grain boundaries and between lath bundles.The difference is that the grain size of S13Cr SS was significantly reduced after modification, which increases the density of grain boundaries and accelerates the formation of passivation film [12]- [14].In addition, the phase distribution diagram (Figure 1b and 1e) shows that the distribution characteristics of reverse austenite changed from large particle-like to dispersed distribution after modification, with the content increased from 0.15% to 0.37%.This consumes more dislocation energy, and therefore reduced the dislocation density, which is beneficial for improving the compactness of the passivation film [15]- [17].
In order to further clarify the microstructural differences of S13Cr SS before and after modification, the distribution of grain boundaries in the two materials (figure 1c and 1f) was statistically analyzed.The results shows that the proportions of low angle grain boundaries in S13Cr SS and M-S13Cr SS were 79.1% and 82.4% respectively, indicating a higher proportion of low angle grain boundaries in the M-13Cr SS.Studies have shown that atoms at high angle grain boundaries are arranged disorderly and have higher energy levels, while the energy levels of low angle grain boundaries are relatively lower, discontinuing the continuity of high angle grain boundaries, which can effectively prevent corrosion along grain boundaries [18].

The electrochemical corrosion performance of S13Cr SS before and after modification
As shown in Figure 2, the cyclic polarization curves obtained for the two materials in the ultra-high temperature geothermal fluid environment exhibited similar characteristics, and Table 2 presents the results of electrochemical analysis.It can be seen that the pitting potential (Epit) of M-S13Cr SS was slightly higher than that of S13Cr SS, and there were significant differences in their corrosion potential (Ecorr) and passive current density (ip).The Ecorr of S13Cr SS increased from -530.21 mV before modification to -511.66 mV, while the ip decreased from 2523.54 μA/cm 2 to 886.10 μA/cm 2 .This indicates that the M-S13Cr SS has a better quality of the passive film, improved stability and corrosion resistance, as well as a relatively lower corrosion thermodynamic tendency.3 shows the EIS results (Nyquist and Bode plots) of S13Cr SS before and after modification in a simulated ultra-high temperature geothermal fluid environment.The Nyquist diagram shows that both S13Cr SS and M-S13Cr SS are composed of two semicircular capacitance circuits throughout the entire frequency range, indicating the same corrosion mechanism.However, the diameter of the capacitance circuit in the low-frequency region of M-S13Cr SS is larger than that of S13Cr SS, and there is a certain correlation between the diameter of the capacitance circuit and the corrosion resistance of stainless steel.An increase in diameter means an improvement in corrosion resistance.In addition, the increase in lowfrequency amplitude in the Bode plot shown in Figure 3b indicates an increase in charge transfer resistance, which is consistent with the results of the Nyquist plot.At the same time, the phase angle size is below -60° throughout the entire frequency range, and the phase angle size of S13Cr SS in the low-frequency region is lower than M-S13Cr, indicating that the characteristics of the capacitor are not ideal, and the decrease in phase angle is related to the gradual dissolution of the surface passivation film [19] [20].Furthermore, the impedance spectra data were fitted and analyzed using the equivalent circuit model shown in Figure 3a.The equivalent circuit consists of solution resistance (Rs), film capacitance (Qf), film resistance (Rf), double-layer capacitance (Qdl), and charge transfer resistance (Rct).The electrochemical parameters obtained from the fitting based on the equivalent circuit are shown in Table 3.It can be observed that the film capacitance and double-layer capacitance of the M-S13Cr SS decreased, while the charge transfer resistance increased.It's suggested that the charge transfer process on the surface of the M-S13Cr SS was inhibited, and the defect density within the passive film was reduced.This is consistent with the results of cyclic polarization test.
Table 3. Analysis results of electrochemical impedance spectra.[21,22] and EIS fitting results (Table 3), the passivation film formed on the sample surface is approximately regarded as a capacitor, and the thickness of the passivation film also can be regarded as the distance between the capacitor plates.According to equation ( 4), the corresponding passivation film thickness (dFilm) can be calculated.

RS(Ω•cm
Where ε0 is the vacuum dielectric constant; ε is the relative dielectric constant; AC is the effective surface area of the sample, and C is the capacitance of passivation film.Due to ε0, ε and AC are constants in this system, it can be inferred that the thickness of the passive film formed on the surface of M-S13Cr SS is greater than that of S13Cr SS and the occurrence of this phenomenon is related to the microstructure and compositional changes of the two materials.

The macroscopic morphology and corrosion rate of S13Cr SS and M-S13Cr SS after corrosion
Figure 4 shows the macroscopic morphologies and three-dimensional morphologies of the two materials after being immersed in a simulated ultra-high temperature geothermal fluid environment for 120 h.
From Figure 4a and d, it can be seen that the surface of the two materials was covered with a layer of gray-black corrosion product film, on which a large number of irregularly distributed bright yellow rhombic crystals were deposited.After removing the corrosion product, it's found that the metallic luster on the surfaces of the sample disappeared and turned grayish-white.Further three-dimensional contour measurements of the samples after removing the corrosion product are shown in Figure 4c and 4f.It can be seen that both surfaces of the two materials have experienced varying degrees of pitting.The number and depth of surface pits on S13Cr SS were higher than on M-S13Cr SS, indicating that the M-S13Cr SS exhibits improved resistance to pitting corrosion.Figure 5 shows the CR and Cpit of the two materials.It can be seen that the CR and Cpit of S13Cr SS were 2.1831mm/a and 2.8434 mm/a respectively, while M-S13Cr SS was 1.1754 mm/a and 2.0887 mm/a, both lower than S13Cr SS, which matches the morphology of the corrosion sample.And it's further indicated that the M-S13Cr SS has higher corrosion resistance.

The characteristics of product film of S13Cr SS and M-S13Cr SS after corrosion
After immersion in a simulated high-temperature geothermal fluid environment for 120 h, the surface microstructures of the two materials are shown in Figure 6.It can be observed that the surface microstructures of both materials were relatively smooth, and there were randomly distributed larger crystals on the entire surface (Figure 6a and 6c).Upon further magnification of the regions marked by the red boxes, as shown in the Figure 6, the SEM images (Figure 6b and 6d) show that the two materials have a small amount of smaller particles attached to the inner product film, which exhibit significant differences in morphology compared to the large crystalline deposits on the surface.In order to determine the elemental composition of different particles, EDS spectra were used for elemental analysis.
It can be seen that the Fe/O atomic ratio in the large crystals was close to 1:3, indicating typical CO2 corrosion products [21][24].On the other hand, the small particles show a high percentage of Fe atoms, suggesting deposition of partially corroded steel substrate on the surface.In addition, local damage to the product film was observed on the surface of S13Cr SS, indicating slightly lower stability of the product film formed on the surface of S13Cr SS compared to M-S13Cr SS. Figure 7 shows the cross-sectional microstructure and EDS line scan results of the surface film for the two materials in this environment.It can be seen that the passive films on the surfaces of the two materials completely dissolved, and a layer of continuous and evenly distributed corrosion product film with good integrity and thickness formed on the surface.Further magnification of the cross-sectional morphology of the product film reveals that it consists of two phases.According to the study by Yue et al. [10], the darker phase is the corrosion product containing Cr, while the other phase is the residual reversed austenite after the substrate corrosion.In addition, the product film of M-S13Cr SS exhibits different layers and shows obvious stratification.The EDS line scan results also show that there is a significant enrichment of Ni elements at the interface between the inner layer of the product film and the substrate of M-S13Cr SS, occupying a large number of vacancies, which hinders the diffusion of the corrosive medium to the substrate.Currently, research results have shown that FeCr2O4 is one of the thermodynamically more stable corrosion products in the Fe-Cr-Cl --CO2-H2O system, when the environmental temperature is above 200℃, and it mainly forms through equations ( 5)~( 7) [10].In the initial stage of corrosion, the local dissolution of the passive film on the material surface occurred, and Fe 2+ gradually occupies the Cr 3+ sites in the spinel structure, which damaged the protection of the passive film and provided more channels for ions to penetrate through the passive film to the substrate.With the continuous progress of corrosion, the passive film eventually disappeared completely, and the corrosion product covered the whole substrate, replacing the role of the passive film in protection.Additionally, a barrier layer formed by the enrichment of a small amount of Ni element and residual reversed austenite in the inner layer of the product film in the steel substrate, as well as the presence of residual MoO2 in the product film, both contribute to the stability of the corrosion product film.Under the combined effect of these factors, the corrosion resistance of M-S13Cr stainless steel in ultra-high temperature geothermal fluids is enhanced.

Conclusions
This study compares the microstructure and electrochemical behaviour of S13Cr SS before and after modification, as well as the characteristics of the corrosion product film in an ultra-high temperature geothermal fluid environment.The focus is on the differences in passivation and corrosion behaviour of S13Cr SS before and after modification, leading to the following conclusions: (1) The modified S13Cr SS has a reduced grain size and an increased content of reverse austenite in the steel.The distribution characteristics changes from coarse-grained to a dispersed distribution, improving the uniformity of the microstructure.
(2) The quality of the passivation film of modified S13Cr SS is improved, resulting in a lower passivation current density and an increase in charge transfer resistance within the passivation film, which hinders the charge transfer process and further reduces the material's corrosion susceptibility.
(3) The passivation film of modified S13Cr SS is completely dissolved in an ultra-high temperature geothermal fluid environment, forming a corrosion product film composed of FeCr2O4, Cr(OH)3, Cr2O3, molybdate and reverse austenite.However, a small amount of Ni elements dissolved in the steel matrix form a barrier layer in the inner layer of the product film, and the residual MoO2 in the product film stabilizes the product film.Under the combined action of both, it enhances the protective effect of the product film, reduces the corrosion rate of the material, and lowers the tendency for pitting corrosion. 87600

Figure 2 .
Figure 2. Cyclic polarization curves of S13Cr SS and M-S13Cr SS exposed to geothermal solutions.

Figure 3 .
Figure 3. Electrochemical impedance spectra of S13Cr and M-S13Cr exposed to geothermal Solutions: (a) Nyquist plots and electric equivalent circuit; (b) Bode plots.Furthermore, the impedance spectra data were fitted and analyzed using the equivalent circuit model shown in Figure 3a.The equivalent circuit consists of solution resistance (Rs), film capacitance (Qf), film resistance (Rf), double-layer capacitance (Qdl), and charge transfer resistance (Rct).The electrochemical parameters obtained from the fitting based on the equivalent circuit are shown in Table3.It can be observed that the film capacitance and double-layer capacitance of the M-S13Cr SS decreased, while the charge transfer resistance increased.It's suggested that the charge transfer process on the surface of the M-S13Cr SS was inhibited, and the defect density within the passive film was reduced.This is consistent with the results of cyclic polarization test.Table3.Analysis results of electrochemical impedance spectra.

Figure 4 .
Figure 4. Macroscopic morphologies and three-dimensional morphologies of S13Cr SS (left) and M-S13Cr SS (right) exposed to the geothermal solutions: (a)(d) before removal; (b)(e) after removal; (c)(f) three-dimensional morphology.Figure5shows the CR and Cpit of the two materials.It can be seen that the CR and Cpit of S13Cr SS were 2.1831mm/a and 2.8434 mm/a respectively, while M-S13Cr SS was 1.1754 mm/a and 2.0887 mm/a, both lower than S13Cr SS, which matches the morphology of the corrosion sample.And it's further indicated that the M-S13Cr SS has higher corrosion resistance.

Figure 5 .
Figure 5. Corrosion rates of S13Cr SS and M-S13Cr SS exposed to geothermal solutions at 210 ℃.

Figure 6 .
Figure 6.Microscopic morphology of surface corrosion products of S13Cr and M-S13C exposed to geothermal solutions (left: 100X; right: 2000X): (a)(b) S13Cr SS; (c)(d) M-S13Cr SS.Figure7shows the cross-sectional microstructure and EDS line scan results of the surface film for the two materials in this environment.It can be seen that the passive films on the surfaces of the two materials completely dissolved, and a layer of continuous and evenly distributed corrosion product film with good integrity and thickness formed on the surface.Further magnification of the cross-sectional morphology of the product film reveals that it consists of two phases.According to the study by Yue et al.[10], the darker phase is the corrosion product containing Cr, while the other phase is the residual reversed austenite after the substrate corrosion.In addition, the product film of M-S13Cr SS exhibits different layers and shows obvious stratification.The EDS line scan results also show that there is a significant enrichment of Ni elements at the interface between the inner layer of the product film and the substrate of M-S13Cr SS, occupying a large number of vacancies, which hinders the diffusion of the corrosive medium to the substrate.

Figure 7 .
Figure 7. Cross-sectional images(left) and EDS line scanning results(right) of the corrosion products formed on the surface of S13Cr SS and M-S13Cr SS exposed to geothermal solutions at 210 ℃ for 120 h: (a)(b) S13Cr; (c)(d) M-S13Cr.To characterize the composition of the corrosion product films on the surfaces of the two materials, high-resolution X-ray photoelectron spectroscopy (XPS) analysis was conducted, and the results are shown in Figure 8.It can be observed that the peaks in the Fe 2p3/2, Cr 2p3/2 and Mo 3d spectra of the corrosion products on the surfaces of the two materials mostly appeared at similar binding energies, indicating that the products have the same valence states.As shown in Figure 9a, the Fe 2p3/2 spectrum can be deconvoluted into four peaks, corresponding to FeCO3 (712 eV, 715.6 eV) and FeCr2O4 (710.3 eV, 713.8 eV) [25][26].Similarly, the Cr 2p3/2 spectrum in Fig. 9b also consists of four peaks, representing Cr(OH)3 (577.3 eV), Cr2O3 (576.4 eV, 579.6 eV) and FeCr2O4 (578.3 eV) [4][10].In figure 9c, there were slight differences in the Mo 3d spectra of the two materials.Besides the common presence of MoO4 2-(232.6 eV, 235.8 eV), the spectrum of M-S13Cr SS also contains incompletely reacted MoO2 (229.1 eV, 232.3 eV), which plays a certain stabilizing role in the formed film [15][27].

Table 2 .
Analysis results of cyclic polarization curve.