Study on the Corrosion Behavior of Austenitic Steel HR3C in Supercritical Carbon Dioxide at 550 and 600 °C

Abstract

The corrosion behavior of austenitic steel HR3C in supercritical CO₂ at 550–600 °C under 25 MPa for 1000 h was investigated. The corrosion kinetics of HR3C were evaluated using weight change measurements. The microstructure and phase composition of HR3C were studied via scanning electron microscopy, X-ray diffraction, X-ray photoelectron spectroscopy, and secondary ion mass spectroscopy. Weight gain data showed that the HR3C exhibited excellent corrosion resistance and that the corrosion kinetics followed a near-parabolic law. The surface of the sample is composed of fine granular oxides, with the main elements including C, O, Cr, Fe and Ni. The oxide phase analysis indicated that protective Cr₂O₃ formed, and a small amount of Fe₂O₃ was also detected. Carbon enrichment was observed on the surface of the outmost layer and the interface of the oxide layer and substrate. The corrosion mechanism and carbon diffusion process are furthermore discussed.

1. Introduction

When the temperature of carbon dioxide reaches 31.1 °C and the pressure reaches 7.38 MPa, it enters a supercritical state. Supercritical carbon dioxide is characterized by low viscosity and high density, which make it an ideal heat exchange medium in thermal cycles. Ultra-high-parameter carbon dioxide coal-fired power generation technology utilizes the supercritical CO₂ Brayton cycle. By absorbing the chemical energy released during combustion, the system generates high-temperature, high-pressure CO₂. Compared to steam cycles with the same parameters, this technology can improve the efficiency of power generation by more than 3% and achieve thermoelectric decoupling [1]. However, compared to traditional steam boilers, supercritical CO₂ boilers remain in the early stages of development in terms of material performance, design, and manufacturing standards. There is still insufficient data available on the corrosion resistance behavior of conventional power plant materials in supercritical CO₂ environments.

The corrosion behavior of boiler materials in a supercritical CO₂ environment is a critical factor influencing operational safety. On one hand, materials undergo oxidation reactions in this environment, leading to the formation of protective oxide films. On the other hand, carbonization reactions may also occur, which also can compromise the creep strength and other mechanical properties of the materials [2]. The pressure-bearing equipment of the supercritical carbon dioxide Brayton cycle system will be subjected to prolonged exposure to high-temperature and high-pressure CO₂. Once the oxide film is removed, it may lead to blockage, overheating, and the eventual rupture of the boiler tubes, which can be attributed to a reduction or interruption in steam flow. Therefore, during the development and research of the S-CO₂ Brayton cycle system, the corrosion resistance of boiler materials has emerged as a critical issue requiring focused attention.

Due to the insufficient strength and limited corrosion resistance of carbon steel, it is not considered a suitable material for superheaters and reheaters in supercritical boilers. A more appropriate and high-performance alternative is Austenitic steel. Austenitic steels are extensively utilized as a structural material for high-temperature superheaters and reheaters in conventional coal-fired power plants, owing to their superior thermal conductivity, excellent high-temperature mechanical properties and corrosion resistance [3]. Relevant studies have been carried out on the corrosion of metals in supercritical CO₂ environments. Cao et al. [4] have studied the corrosion of 316SS and 310SS and Alloy 800H in supercritical CO₂. Alloy 800H exhibited the best corrosion resistance compared with 310SS and 316SS. Oxide spallation occurred in 316 stainless steel. Furukawa et al. [5] investigated the corrosion behavior of 12% chromium martensitic steel in a S-CO₂ environment. The scale formed on martensitic steels is mainly composed of the iron oxides and a complex oxide of iron and chromium. Tan et al. [6] carried out corrosion tests on austenitic and ferritic-martensitic steels exposed to supercritical carbon dioxide. It was found that Alloy 800H had an oxidation resistance superior to that of AL-6XN. The ferritic–martensitic steels were less corrosion-resistant than the austenitic steels. Experimental research was conducted on 316 stainless steel in a supercritical carbon dioxide environment [7]. It was found that 316 stainless steel simultaneously underwent oxidation and carburization. He et al. [8] investigated the corrosion behavior of an alumina-forming austenitic steel exposed to supercritical carbon dioxide at 450–650 °C and 20 MPa. At low temperatures or under short exposure times, the oxide scale was predominantly composed of thin and continuous layers of Al₂O₃ and (Cr,Mn)₃O₄. With an increase in temperature and the duration of exposure, the continuity of the Al₂O₃ scale was compromised, resulting in the formation of a multilayer structure. Corrosion mechanisms have been proposed to elucidate the formation of duplex oxide growth and the occurrence of carburization beneath the oxide scale [9,10,11].

Austenitic stainless steel HR3C, which exhibits excellent creep strength and high resistance to steam oxidation, is extensively utilized in ultra-supercritical coal-fired power plants [12]. This study selects austenitic stainless steel HR3C as the research subject and performs corrosion experiments in a supercritical carbon dioxide environment at temperatures ranging from 550 °C and 600 °C under a pressure of 25 MPa. The corrosion behavior of HR3C under these conditions is investigated to provide reference data for its potential application in advanced coal-fired power generation systems.

2. Materials and Methods

Commercial-grade austenitic stainless steel HR3C was used in the experiment. The chemical composition of HR3C, as provided by the manufacturer, is presented in

Table 1. The chemical composition of HR3C (wt%).

C	N	P	Cr	S	Nb	Mn	Ni	Si	Fe
0.06	0.24	0.012	24.63	0.001	0.49	1.24	20.29	0.39	Bal.

. The material was machined into specimens measuring 20 mm × 10 mm × 2 mm using a wire-cutting machine. The six surfaces of each specimen were ground with SiC sandpapers of grit sizes 200#, 400#, 600#, 800#, and 1000#, followed by ultrasonic cleaning. The sample was cleaned using alcohol. The ultrasonic cleaning was conducted at room temperature for a duration of 5 min. Ultra-pure CO₂ gas (purity ≥ 99.999%) was employed in the experiment. Figure 1 shows the high-temperature corrosion test system. The specimens were placed on ceramic supports. To obtain corrosion weight gain data, the mass of each specimen before and after the experiment was measured using an electronic balance with a precision of 0.01 mg. In each experimental group, three parallel samples were employed. The experiments were carried out with periodic interruptions at 200-h intervals, and data were collected at 200 h, 400 h, 600 h, 800 h, and 1000 h, respectively. The rate of temperature increase is approximately 3 °C per minute. The cooling process is natural cooling after stopping the CO₂ supply and turning off the heating power supply. To further investigate the depth distribution of different layer structures and compositional variations, depth profiling and three-dimensional elemental imaging were conducted using secondary ion mass spectrometry (SIMS) equipped with cesium sputtering technology. The surface morphology of the samples was examined using a JEOL JSM-7200F scanning electron microscope (SEM) (Tokyo, Japan) using LEO 1530 field-emission (Carl Zeiss AG, Oberkochen, Germany) at 10.00 kV. The elemental composition and content of the oxides were analyzed via energy-dispersive X-ray spectroscopy (EDS). The phase composition of the oxide film was characterized via SmartLab SE X-ray diffraction (XRD) (Rigaku, Tokyo, Japan) using copper radiation (λ = 1.542 $\dot{\mathrm{A}}$) and Thermo ESCALAB 250Xi X-ray photoelectron spectroscopy (XPS) (Waltham, MA, USA). A TOF-SIMS (IONTOF V) instrument was employed to perform depth profiling and a three-dimensional imaging analysis of the oxide film, thereby acquiring detailed information on the depth distribution of different layer structures and compositional variations.

3. Results and Discussion

3.1. Weight Gain

Figure 2 presents the relationship curve between the weight gain of HR3C and the exposure time in supercritical CO₂ at 550 and 600 °C and a pressure of 25 MPa. The weight gain data and exposure duration were fitted according to Equation (1):

$$∆w=k_p{t}^n$$

(1)

with $∆w$, weight gain, in mg/dm²; $k_p$, the parabolic constant for oxidation in mg/dm²·h; and t, oxidation time, in h. n is the time exponent. As shown in Figure 2, the time exponents at 550 °C and 600 °C are 0.65 and 0.54, respectively. The corrosion kinetics at 600 °C approximately follow a parabolic rate law, indicating that the corrosion process is governed by ion diffusion. The oxidation kinetics at 550 °C deviate from the parabolic law. The weight gain after 1000 h of corrosion at 550 °C and 600 °C is 1.53279 mg/dm² and 2.3122 mg/dm², respectively. With an increase in temperature, the corrosion rate of HR3C increases by a factor of 1.5. The oxide phase formed on 9–12Cr steel in supercritical carbon dioxide mainly consists of Fe₂O₃, Fe₃O₄ and Fe-Cr spinel [13]. Compared to the weight gain of 9–12Cr steel exposure in supercritical carbon dioxide environments, the oxidation rate is two orders of magnitude higher than that of HR3C. HR3C exhibits superior corrosion resistance.

3.2. Surface Morphology and Elemental Composition of the Corrosion Layer

Figure 3, Figure 4, Figure 5 and Figure 6 present the surface morphology and elemental distribution of HR3C after 200 and 1000 h of corrosion in supercritical CO₂ at 550–600 °C. As indicated by the surface morphology, a fine and uniformly distributed layer of flaky oxides is observed on the sample surface after 200 and 1000 h at 550 °C. At 600 °C, nano-sized granular oxides are formed on the surface, with an increase in particle size observed as the time increases. The oxide film remains compact overall, with no signs of spallation. The diffusion path of metal ions affects the morphology of the surface oxide film. The formation of flaky oxides may be primarily caused by the diffusion along dislocation. The outward cation diffusion along the grain boundaries results in the formation of granular oxides.

Table 2. The atomic percent contents (at %).

Element	550 °C-200 h	550 °C-1000 h	600 °C-200 h	600 °C-1000 h
C	10.85	6.50	10.24	9.03
O	30.22	37.39	28.70	38.17
Cr	17.47	19.70	16.85	21.36
Fe	30.64	27.24	32.29	22.44
Ni	10.81	9.14	11.93	7.42

shows the primary compositional elements of the surface corrosion layer at 550–600 °C for different time. The EDS scanning results reveal that the surface oxides primarily consist of C, O, Cr, Fe, and Ni. The surface scanning results further demonstrate that these five elements are uniformly distributed across the sample surface. Based on the elemental composition and content distribution, it can be inferred that carbon is enriched in the outermost layer of the oxide film.

3.3. Phase Composition of the Corrosion Layer

Figure 7 presents the phase composition of the corrosion products formed on the surface of the HR3C steel after oxidation for 200 and 1000 h in a supercritical CO₂ environment at temperatures ranging from 550 to 600 °C. XRD analysis reveals that the primary oxide component is Cr₂O₃. Furthermore, the presence of substrate peaks indicates that the oxide film developed on the sample surface is relatively thin. The corrosion rate of HR3C in supercritical CO₂ is significantly reduced due to the slow ion diffusion within the Cr₂O₃ protective layer.

The chemical states of the constituent elements in the oxide film formed on HR3C after oxidation for 200 h and 1000 h in a supercritical CO₂ environment at 550–600 °C were analyzed using X-ray photoelectron spectroscopy (XPS). The corresponding results are presented in Figure 8, Figure 9, Figure 10 and Figure 11.

Table 3. The corresponding binding energy of the detected element.

Main Peak	550 °C-200 h	550 °C-1000 h	600 °C-200 h	600 °C-1000 h
O 1 s	530.76	530.73	530.43	530.36
Fe 2 p	---	710.3	710.6	710.8
Cr 2 p	576.37	576.69	576.26	576.32

presents the binding energies corresponding to the various elements. The narrow-scan XPS spectrum of Cr 2p_3/2 reveals that the Cr 2p_3/2 peak is located at 576.4 ± 0.3 eV, which corresponds to the binding energy of Cr₂O₃ [14,15]. The Fe 2p_3/2 peak is located at 710.5 ± 0.3 eV, which corresponds to Fe₂O₃ as the predominant oxide phase [16]. The narrow-scan O 1s spectrum exhibits binding energies at 530.5 ± 0.3 eV and 531.3 ± 0.3 eV, which correspond to O²⁻ and OH⁻ species, respectively [17]. Based on the XRD and XPS results, it can be inferred that after 200 h of oxidation at 550 °C, the oxide layer is predominantly composed of Cr₂O₃. Following 1000 h of oxidation, the oxide layer consists primarily of both Cr₂O₃ and Fe₂O₃. After oxidation for 200 and 1000 h at 600 °C, the oxide layers contain both Cr₂O₃ and Fe₂O₃. Cox et al. [18] calculated the diffusion rates of iron and chromium ions within the Cr₂O₃ layer, and the results indicate that the ion diffusion coefficients follow the order D_Fe > D_Ni > D_Cr. Consequently, Fe, Ni, and Cr were detected on the surface of the oxide film. OH⁻ was also detected. It is speculated that the following situation is more likely to occur. Although the carbon dioxide used in the experiment was high-purity, an extremely small amount of water inevitably mixed with carbon dioxide.

3.4. Corrosion Mechanism

According to the XRD and XPS results, it can be inferred that the oxide formed by HR3C in the supercritical CO₂ environment is predominantly Cr₂O₃. It is generally accepted that when the chromium content in an alloy exceeds 20%, sufficient chromium diffuses to the alloy surface to form a continuous, chromium-rich oxide layer [19]. Figure 12 presents the thickness distribution of the oxide film obtained through SIMS analysis. As both the temperature and exposure time increase, the oxide film thickness also increases. Along the thickness direction of the oxide film, the primary constituents are Cr and O, although Fe, Ni, and C are also present. The formation of Fe₂O₃ and Cr₂O₃ can be described by the following reactions [1,20,21]:

$$2\mathrm{F}\mathrm{e}+3\mathrm{C}{\mathrm{O}}_2\left(\mathrm{g}\right)={\mathrm{F}\mathrm{e}}_2{\mathrm{O}}_3+3\mathrm{C}\mathrm{O}\left(\mathrm{g}\right)$$

(2)

$$2\mathrm{C}\mathrm{r}+3\mathrm{C}{\mathrm{O}}_2\left(\mathrm{g}\right)={\mathrm{C}\mathrm{r}}_2{\mathrm{O}}_3+3\mathrm{C}\mathrm{O}\left(\mathrm{g}\right)$$

(3)

According to the Boudouard reaction [22], the carbon monoxide generated by reaction Equations (2) and (3) may lead to the formation of elemental carbon, as illustrated in the following reaction equation:

$$2\mathrm{C}\mathrm{O}\left(\mathrm{g}\right)=\mathrm{C}+\mathrm{C}{\mathrm{O}}_2\left(\mathrm{g}\right)$$

(4)

Figure 12 illustrates the enrichment of elemental carbon at the outermost layer of the oxide film. The formation of elemental carbon is related to Reactions (2)–(4). Reactions (2) and (3) primarily take place at the oxide film/environment interface, which contributes to the growth of the oxide film. Figure 13, Figure 14 and Figure 15 depict the three-dimensional elemental distribution across the cross-section of HR3C after corrosion in supercritical CO₂ at 550–600 °C for various exposure durations. Based on Figure 12, Figure 13, Figure 14 and Figure 15, it can be observed that, in addition to its enrichment at the outermost layer of the oxide film, a minor portion of carbon diffuses to the oxide film/matrix interface, with carbon also being detected within the matrix. It was also observed that carbon was distributed throughout the thickness of the oxide film formed on 316 stainless steel [2]. The enrichment of carbon at the oxide film/matrix interface and within the metal matrix may be associated with the inward diffusion of CO or CO₂ through the oxide film. The diffusion of carbon into the metal matrix is highly likely to result in the formation of carbides involving chromium and other alloying elements.

4. Conclusions

The oxidation behavior of HR3C in supercritical carbon dioxide at 600 °C approximately follows a parabolic rate law, whereas at 550 °C, the oxidation kinetics deviate from this parabolic relationship. HR3C exhibits a relatively low oxidation weight gain, indicating superior oxidation resistance. The oxide layer formed on the surface of HR3C is predominantly composed of Cr₂O₃, with a minor amount of Fe₂O₃ also detected. As a protective oxide layer, Cr₂O₃ effectively reduces the ionic diffusion rate. This is the main reason why HR3C exhibits excellent corrosion resistance. Carbon is enriched at the interface between the oxide film and the supercritical CO₂ environment, and a limited amount of carbon is also observed at the oxide film/matrix interface and within the matrix. It is possible that the outward diffusion of metal ions occurs and that these react with carbon dioxide at the oxide film/environment interface. In addition, the diffusion of carbon into the metal matrix results in the formation of carbides.