Introduction

Carbon capture, use, and storage (CCUS) is a research area aimed at reducing CO2 emissions and combating climate change. In the CCUS process (Carbon Capture, Use and Storage) [1; 2], CO2 emitted into the atmosphere from various sources such as industrial enterprises, power plants, and motor vehicles is captured and subsequently stored in underground formations [3; 4]. This allows for the isolation of carbon dioxide from the atmosphere and prevents its negative impact on the climate.

Currently, there is a challenge in selecting the material design for infrastructure objects used for CO2 transport and injection under increased environmental and economic risks [5; 6]. The selection of material should strike a balance between effectiveness in preventing intensive carbon dioxide corrosion [7] and the cost of the final solution [8].

A review of open sources on the topic of material selection for CCUS facilities [3; 9] has shown that carbon steels are the most economically advantageous materials for pipeline construction [10]. However, their application necessitates increased attention to the problem of carbon dioxide corrosion [11], which intensifies in wet CO2 flows [12]. Additionally, there is no consensus in the literature regarding the influence of steel microstructure and thermomechanical treatment modes on the mechanism and kinetics of corrosion processes [13 – 16]. This is because conditions can vary from one facility to another in terms of mineralization, component composition of environments, partial pressure, and temperature. Consequently, corrosion product deposits of varying morphology form on the steel surface, which can either be protective or accelerate the corrosion process. Therefore, the selection of materials must consider the operational conditions and actual environments at the facility. In this context, the goal of the present work is to investigate the influence of microstructural state features on the corrosion rate of low-alloy mild steels for CO2 transport and injection, based on the conditions of a specific facility.

Materials and method

To recreate the conditions of actual facilities, various bench setups are used, such as closed-loop flow stands or autoclave setups that replicate conditions of high pressure and temperature. In this work, an autoclave setup with a 3-liter capacity vessel was used. Currently, there are no standard methodologies for conducting corrosion tests in autoclaves, so the methodology used in this experiment is described below. The test conditions were selected based on the situation at the actual facility.

The essence of the corrosion testing method is to determine the mass loss of samples during their exposure to the corrosive environment. In the gravimetric method, the corrosion rate is determined by the mass parameter ρ, expressed in g/m2·h:

where m1 is the mass of the sample before testing, g; m2 is the mass of the sample after testing, g; S is the surface area of the sample, m2; τ is the duration of the test, h.

If the change in mass is directly proportional to the depth of corrosion penetration under uniform corrosion conditions, the corrosion rate is recalculated into the depth parameter v, expressed in mm/year. This parameter characterizes the uniform corrosion loss (thinning) of the sample per unit time:

where v is the depth parameter of the corrosion rate, mm/year; 8760 is the number of hours in a year; 7.85 is the density of the investigated steel, g/cm3.

The gravimetric method allows for highly accurate results, as weighing is one of the most precise operations in quantitative analysis. The confidence interval for determining measurement error was 95 %.

Before testing, the samples were degreased using an ultrasonic cleaner in acetone, air-dried, and placed on a suspension in the autoclave, which was then sealed and deaerated for 60 min with an inert gas flow rate of 100 ml/min per liter of capacity. The deaerated test solution was then saturated with CO2 . The total pressure in the autoclave was 6 MPa, with a partial pressure of CO2 at 0.3 MPa. The temperature was maintained constant at 25 °C. The test duration was 96 h. For corrosion tests, three samples were selected from each steel.

The test solution consisted of 133 g/l CaCl2 , 31 g/l MgCl2 , 0.5 g/l NaHCO3 , 0.1 g/l Na2SO4 in distilled water. The total mineralization was 164.6 g/l. This composition corresponds to the fluid composition in the well being injected with CO2 into underground formations. After testing, the samples were washed and air-dried, then cleaned of corrosion products using an eraser. The investigated steels correspond to the strength class K52, with their chemical composition presented in Table 1.

The microstructure of the steels was evaluated using a Reichert-Jung MEF3A optical microscope. To perform a quantitative assessment of the phase components, etching was selected to provide a contrasting image of the phases without revealing grain boundaries for subsequent automatic recognition by the Thixomet Pro image analyzer [17]. For this purpose, etching with a 4 % alcoholic solution of picric acid was used to reveal pearlite. For overall microstructure identification, a 3 % nitric acid solution in alcohol was used.

The traditional method for grain size assessment is described in GOST 5639 – 82, which, however, does not allow for the assessment of the heterogeneity of the grain structure. Nowadays, many methods for evaluating the grain structure of a material are known [18; 19]. For instance, in [20], a method for calculating the grain size variation factor FZ is proposed by the formula

where fmax is the fraction of the grain occupying the maximum area on the cross-section, %; Zmax is the grade of the grain occupying the maximum area on the cross-section; fi is the fraction of the grain with a specific grade, %; Zi is the grade of the grain.

Results of the quantitative assessmentof structural components and corrosion testing

The materials investigated in this study were obtained from various domestic metallurgical enterprises using a technology that includes converter smelting, ladle treatment, continuous casting, followed by thermomechanical processing. The microstructures of the steels were examined in cross-section (Fig. 1).

The microstructure of steel 1 is characterized as ferritic-bainitic. The pearlite content in this steel is the lowest among all the steels studied, at 2.20 vol. %, and the grain structure (by the grain number occupying more than 10 % of the cross-section area) is evaluated as G10 (30 %) and G11 (28 %). The microstructure of steel 2 is characterized as ferritic-pearlitic with predominantly non-polygonal ferrite, with the grain structure evaluated as G9 (13 %) and G10 (21 %). The pearlite content in this steel is 3.57 vol. %. The microstructure of steel 3 is also characterized as ferritic-pearlitic. The structure is evaluated as G11 (32 %) and G12 (32 %), and the pearlite content is 6.72 vol. %. The distribution of pearlite colonies is uneven, appearing as separate clusters with coarse morphology, similar to steel 2 (Fig. 2). The microstructure of steel 4 is characterized as ferritic-pearlitic-bainitic, with a pearlite content of 7.51 vol. %. The distribution of pearlite in this steel is fairly uniform, and the structure is evaluated as G9 (14 %), G10 (28 %), G11 (29 %), and G12 (18 %). The microstructure of steel 5 is characterized as ferritic-pearlitic. This steel has the highest pearlite content, 13.80 vol. %, after etching with picric acid. The structure is evaluated as G10 (21 %), G11 (32 %), and G12 (25 %).

Results and discussion

In ferritic-pearlitic steels, pearlite consists of a lamellar structure composed of alternating layers of ferrite and iron carbide (cementite). When exposed to aggressive media containing CO2 , the ferrite in the pearlite becomes the anode and dissolves quickly, while the cementite remains intact (Fig. 4). The carbon content is directly proportional to the pearlite content. The results of the corrosion tests shown in Fig. 3, a demonstrate that as the pearlite fraction in the microstructure increases, the overall corrosion resistance of the steels decreases.

The study [21] indicates that the most vulnerable areas are the boundaries between ferrite and cementite, where corrosion processes initially begin. It appears that increasing the number of pearlite colony areas leads to an increase in the number of vulnerable ferrite-cementite interphase boundaries and ferrite plates, which subsequently become anodes and dissolve quickly.

It is also worth noting the positive effect of copper on corrosion resistance – increasing the copper content from ~0.12 to 0.50 wt. % led to a reduction in the corrosion rate from 0.237 to 0.167 mm/year (Fig. 3, b). The beneficial effect of copper on the corrosion properties of steels is noted in many studies. For instance, in [23], it was found that the introduction of a small amount of copper (usually up to 1 %) significantly enhances the corrosion resistance of low-carbon steels. Moreover, in high-strength low-alloy steels, copper is an alloying element that provides high toughness and good weldability. However, despite the advantages of adding copper to steel composition and the corrosion characteristics of ferritic-pearlitic microstructures demonstrated in this work, low-alloy mild steels should be used cautiously for the material design of CO2 transport and injection infrastructure objects, and only with strict control over the presence of free water in the transported gas – its amount should be minimal. Otherwise, corrosion-resistant materials must be used.

Based on the evaluation of the microstructural parameters of the investigated steels presented in Table 2, it can be concluded that grain size does not have a determining influence on the corrosion resistance of the investigated steels (Fig. 3, c). However, the grain size variation, assessed by equation (3), significantly affects corrosion resistance (Fig. 3, d). Therefore, it can be concluded that the higher the uniformity of the structure (and, accordingly, the value of the FZ factor), the higher the corrosion resistance of the material.

Conclusions

A study was conducted on the influence of microstructural state features on the corrosion rate of low-alloy mild steels for CO2 transport and injection (CCUS). It was shown that in ferritic-pearlitic steels, ferrite is primarily susceptible to corrosion processes, becoming an anode and dissolving quickly within pearlite. As a result, increasing the number of pearlite colony areas in the microstructure leads to a decrease in the overall corrosion resistance of the steel. The positive effect of copper on corrosion resistance was established – increasing the copper content in the investigated compositions from 0.12 to 0.50 wt. % led to a 1.5-fold reduction in the corrosion rate. It was noted that corrosion resistance is significantly influenced not by grain size, but primarily by the heterogeneity of the structure.