Study on the passivation and corrosion behaviors of steel bars in the cements made from high-belite calcium sulfoaluminate (HB-CSA) clinker

Abstract

This study evaluates the corrosion protection performance of high-belite calcium sulfoaluminate (HB-CSA) cement and granulated blast furnace slag-calcium sulfoaluminate (G-CSA) cement for embedded steel reinforcement under chloride exposure. Steel passivation and corrosion behavior were investigated in simulated pore solutions under chloride-free and chloride-containing conditions, and the long-term corrosion performance of steel bars embedded in cement mortars exposed to 3.5% NaCl solution was further evaluated. The results showed that steel bars immersed in the chloride-free HB-CSA simulated pore solution developed a stable passive state and formed a protective CaCO₃ layer on their surfaces, providing superior corrosion resistance compared with ordinary Portland cement (OPC) and conventional CSA cement. In chloride-containing simulated pore solutions, the corrosion resistance of HB-CSA decreased; however, the corrosion degree remained lower than that in conventional CSA cement, while OPC exhibited the best protection performance. In contrast, steel bars immersed in the G-CSA simulated pore solution experienced more severe corrosion because of the low alkalinity and high sulfate content of the pore solution. For mortar specimens exposed to a 3.5% NaCl solution, both HB-CSA and G-CSA mortars exhibited lower steel corrosion than OPC mortar owing to their dense microstructures and enhanced resistance to chloride ingress, with G-CSA providing the best long-term protection. These findings demonstrate that the corrosion protection performance of HB-CSA-based cement systems strongly depends on the exposure environment and that improving matrix impermeability is an effective strategy for enhancing the durability of reinforced concrete under chloride attack.

1 Introduction

Ordinary Portland cement (OPC) has been recognized as the most widely used binder in infrastructure construction for the past 200 years (Chen et al., 2025). This extensive use has led to significant carbon dioxide (CO₂) emissions associated with its production. Statistics indicate that the cement industry is responsible for 6%–7% of global CO₂ emissions (Nejad et al., 2025). The industry faces increasing pressure to pursue sustainable development and explore alternative binders to OPC (Deng et al., 2025). In addition, OPC is susceptible to corrosion from sulfates and other substances (Lothenbach et al., 2010), especially in marine and high-salinity environments (Ramezanianpour et al., 2021; Zhou et al., 2021). These conditions significantly compromise the durability of OPC-based concrete structures, resulting in substantial annual economic losses (Val and Stewart, 2003).

Calcium sulfoaluminate (CSA) cement is a type of cement developed in the 1970s, primarily composed of ye’elimite, anhydrite, and belite (Bescher and Kim, 2019). The kiln temperature needed to produce CSA is approximately 150 °C–200 °C lower than that for OPC, resulting in a 20%–30% reduction in grinding energy and CO₂ emissions (Liu Y. et al., 2025). Consequently, CSA is a promising alternative to OPC (Srivastava et al., 2021). In addition, CSA cement exhibits advantageous engineering properties for a range of applications. Its rapid setting and hardening make it ideal for urgent repairs, ceiling applications, and soil stabilization (Pooni et al., 2020; Więckowski, 2020). Importantly, its excellent resistance to sulfate and acid attacks makes it especially suitable for marine engineering, urban sewage systems, and other projects requiring high durability (Bakera and Alexander, 2024; Tan et al., 2020). Additionally, CSA cement can be used for shrinkage compensation and self-leveling due to its slight expansibility (Sirtoli et al., 2019).

Despite its benefits, CSA has notable disadvantages. Due to the fact that the main mineral in CSA is ye’elimite, its production relies on high-grade bauxite, which restricts large-scale production (Gartner and Sui, 2018). Furthermore, the strength of CSA may deteriorate over time due to AFt-related expansion (Chaunsali and Mondal, 2015; Ogawa and Roy, 1982). Increasing the belite content has been proposed as an effective strategy to address these limitations, but the low hydration activity of belite significantly reduces the early strength of the cement (Galluccio et al., 2019). In recent years, a new high-belite calcium sulfoaluminate cement (HB-CSA) clinker has been developed to overcome the adverse effects of high belite content in CSA on cement properties (Li et al., 2014). Due to the formation of belite with high activity, the level of belite in HB-CSA clinker can exceed 50% (Li et al., 2014). Therefore, the HB-CSA can be produced using solid waste as a raw material (Su et al., 2025). Using this clinker, HB-CSA cement has been developed to significantly enhance the early strength of CSA with high belite content (Zhang et al., 2022). Additionally, it was used as an activator to increase blast slag activity and to propose a high-performance cement, named granulated blast furnace slag-calcium sulfoaluminate (G-CSA) cement (Sun et al., 2022). These innovative cements are characterized by high strength, low CO₂ emissions, and excellent durability (Huo et al., 2024).

However, both HB-CSA and G-CSA cements inherit the low-alkali properties of CSA, with pH values around 12 or below (Zhang, 2022), whereas the pH of OPC is above 13 (Revertegat et al., 1992). Low alkalinity is generally considered detrimental to the protection of steel reinforcement against corrosion (Elsner and Rossi, 1995). Therefore, problems may arise when using cement prepared from CSA clinker in reinforced concrete (Morozov et al., 2022). Nonetheless, the corrosion of steel bars embedded in cement is a complex process, influenced not only by the alkalinity of the cement but also by the impermeability of the matrix, the presence of other ions in the pore solution, and the overall durability of the cement (Tang et al., 2023; Wang et al., 2017; Song, 2012). Recent studies have also reported the corrosion behavior of steels in chloride and sulfate co-existing pore solutions with different pH values (Liu et al., 2024). Therefore, evaluating the protective effect of cements made from HB-CSA clinker on steel corrosion based solely on pH is insufficient. It is important to more thoroughly assess the protective effect of cements made from HB-CSA clinker before its use in reinforced concrete structures.

This study investigates the passivation and corrosion behavior of steel bars in cements made from HB-CSA clinker, including both HB-CSA and G-CSA. First, the simulated pore solutions were prepared to match the true pore-solution compositions of HB-CSA and G-CSA cements. The steel bars were immersed in simulated pore solutions with and without chloride ions, and a series of electrochemical and microscopic tests was conducted to investigate their passivation or corrosion behavior in these environments. Finally, the corrosion behavior of steel bars embedded in cement mortar in chloride-ion solution was also investigated using electrochemical and microscopic tests to comprehensively evaluate the protective effect of HB-CSA and G-CSA cements on steel bar corrosion in chloride environments.

2 Experiments

2.1 Materials and specimen preparation

HB-CSA and G-CSA cements supplied by Tangshan Polar Bear Building Materials Co., Ltd. Were used to study the corrosion of steel bars in the cements made from HB-CSA clinker. For comparison, CSA and OPC produced by Jidong Cement Co., Ltd. Were also included. The chemical compositions of these four cements are shown in Table 1. Steel bars for corrosion testing were cut from ribbed steel bars (HPB 235, in accordance with GB/T 1499.2–2018). The chemical composition of the steel bar was analyzed using energy-dispersive X-ray spectroscopy, and the results are shown in Table 2.

2.2 Cement pore solution extraction, analysis and preparation of simulated pore solutions

Cylindrical cement paste specimens (Φ50 mm × 60 mm) were prepared with a water-to-binder ratio of 0.41. After casting, all specimens were sealed with plastic film to prevent moisture loss and cured at 20 °C ± 2 °C and over 95% relative humidity (RH). It has been reported that the cement paste at 7 days can reach at least 80% of its 28-day strength (Vinoth et al., 2018), and that the of the pore solution chemistry remains relatively stable from 7 to 180 days for all four types of cement (Park et al., 2021). Therefore, the pore solution was extracted at 7 days to prevent ongoing hydration from reducing the amount available for extraction.

Pore solution extraction was performed using a high-pressure extraction device following the method proposed by Barneyback et al. (Barneyback and Diamond, 1981). The specimen-loaded cell was placed in a WAW100 universal testing machine, and pore solution was extracted under compressive loading at a rate of approximately 1.5 kN/s until a maximum load of 1,000 kN was reached. The extraction process lasted approximately 10 min, ensuring sufficient pore liquid recovery.

The extracted solutions were filtered through a 0.45 μm membrane filter to remove suspended particles. The pH was measured using a pH meter (Mettler InPro 4800i/SG/120) with three-point calibration. Subsequently, the extracted pore solution was diluted with 78% HNO₃. Ionic concentrations were determined using an inductively coupled plasma optical emission spectrometer (ICP-OES, Thermo Fisher iCAP PRO). The concentration of SO₄^2- ions in the solution was also tested using an ion chromatograph (ICS, Thermo Fisher ICS5000+). The measured pore solution compositions are summarized in Table 3. Simulated pore solutions were then prepared using chemically pure NaOH, KOH, Na₂SO₄, CaSO₄, K₂SO₄, Na₂SiO₃, and NaAlO₂. The concentrations were adjusted to match the measured pore solution chemistry shown in Table 3, thereby ensuring chemical equivalence between real and simulated environments.

2.3 Test method

2.3.1 Passivation and corrosion behaviors of steel bars in the simulated pore solutions without and with chloride ions

The 8-mm-diameter steel bars were cut into 5-mm-long specimens. For each cut steel bar, a copper wire was welded to one end, with the other end designated as the working surface. The remaining part was coated with epoxy resin, leaving only the working surface exposed. The working surface was gradually polished using silicon carbide (SiC) paper of 800, 1,200, 2000, 3,000, and 5,000 grit with distilled water as a lubricant, then polished with a cloth dipped in a 0.25 μm diamond solution. After polishing, the electrode was ultrasonically cleaned in anhydrous ethanol for 5 min, rinsed with deionized water, then dried with filter paper and placed in a dryer.

The polished and dried steel bars were immersed in the cement simulated pore solution for 14 days, with four replicate specimens prepared for each solution. The first specimen was measured for the open circuit potential (OCP) at 0 h, 2 h, 4 h, 6 h, 8 h, 10 h, 12 h, 1 day, 3 days, 7 days, 10 days, and 14 days. After the OCP stabilizes, electrochemical impedance spectroscopy (EIS) and linear polarization resistance (LPR) measurements were performed on the same electrode system. After 14 days of immersion, the second steel bar was removed from the solutions and used to test the surface morphology and composition of the surface layer.

The third and fourth steel bars were continuously immersed in the same simulated pore solutions, and NaCl was added to increase the solution’s NaCl content to 0.2%. The third and fourth steel bars were immersed in the simulated pore solutions containing chloride ions for another 14 days. Afterwards, both specimens were removed from the solution. The third steel bar was used for real-time EIS and LPR testing, and the fourth for surface morphology and composition testing.

2.3.1.1 Electrochemical measurements

All electrochemical measurements were performed at room temperature (25 °C ± 5 °C) using a three-electrode system connected to an electrochemical workstation (CHI660E, CH Instruments, Inc., China). The electrodes used in the experiments included a steel bar as the working electrode, a saturated calomel electrode as the reference electrode, and a platinum (Pt) electrode as the counter electrode. The corrosion potential (E_corr), polarization resistance (R_P), and corrosion current (i_corr) of the steel bar were measured using the LPR method (ASTM G102-89, 2015). During the LPR testing, the potential was set to ±10 mV relative to the open-circuit potential, with a sweep rate of 0.617 mV/s. Additionally, EIS was performed to evaluate the protective performance of the cement-simulated pore solution on steel rebars (Garcés et al., 2005). EIS measurements covered a frequency range from 100 kHz to 10 mHz, with an amplitude of 10 mV RMS. Throughout the process, Nyquist and Bode plots were recorded to provide further insight into the observed electrochemical behavior of the steel bars. All electrochemical tests were performed in triplicate, and the results are presented as average values.

2.3.1.2 Analysis of microstructure and chemical composition of the steel bar surface

The surface structure and composition of the outer layer of the steel bars were examined to evaluate passive film formation and the extent of corrosion. The surface morphology of the steel bars was analyzed using environmental scanning electron microscopy (SEM) with a Quanta 450 FEG, in conjunction with energy-dispersive X-ray spectroscopy, operating at an accelerating voltage of 5–20 kV. The oxide chemistry was characterized through X-ray photoelectron spectroscopy (XPS) using an ESCALAB 250Xi spectrometer. The elemental composition of the surface layer at different depths was measured with a Thermo K-alpha X-ray photoelectron spectrometer, utilizing argon ion (Ar+) sputtering at a rate of 0.1 nm/s. The sputtering durations were 0 s, 10 s, 20 s, 40 s, 60 s, and 100 s, corresponding to detection depths of 0 nm, 1 nm, 2 nm, 4 nm, 6 nm, and 10 nm, respectively. The X-ray source employed monochromatic Al Kα at 15 kV and 25 W, with a photoelectron take-off angle of 45° and an incident beam diameter of 500 μm.

2.3.2 The corrosion behavior of steel bars embedded in cement mortar exposed to chloride ion solutions

Prismatic mortar specimens (40 mm × 40 mm × 160 mm) were prepared according to the mix proportions shown in Table 4. Each specimen contained one concentrically embedded 120 mm-long steel bar, with a minimum cement mortar protective layer of 20 mm. Before embedding, the steel bars were treated to remove rust. Then, the steel bar was dried in a vacuum dryer, weighed, and recorded (m₁). A copper wire was welded to one end face of the steel bar, and both ends of the steel bar were wrapped with insulating tape and sealed with epoxy resin. A 100 mm working face was left exposed in the middle of the steel bar electrode within the cement mortar. The steel bar and the prepared steel bar-cement mortar specimen are shown in Figure 1.

FIGURE 1

The prepared steel bar-cement mortar specimens were immersed in clean water to cure for 28 days. After curing, the specimens were immersed in a 3.5% NaCl solution for an additional 180 days. During the immersion, the OCP (E_corr) of the steel bars embedded in the specimen was measured every 24 h. After 180 days of immersion, the specimens were removed from the solution, and the steel bars were extracted by breaking the specimens. The fragments of the broken matrix were then used for mercury intrusion porosimeter testing.

The extracted steel bars were divided into two groups. One group was cleaned of surface impurities using anhydrous ethanol in an ultrasonic cleaner, after which its surface morphology was analyzed by SEM. The other group was first cleaned with a 10% sulfuric acid solution, then rinsed with clean water and neutralized with lime water. Finally, the surface of these steel bars was thoroughly washed with water. The washed steel bars were placed in a dry container for 4 h, then weighed to the nearest 0.001 g to record their mass (m₂). The corrosion weight rate (L_w) of the steel bars was then calculated according to Equation 1.

3 Results and discussion

3.1 Passivation behavior of the steel bars in the simulated pore solution without chloride ions

Figure 2 illustrates the changes in E_corr, R_p, and i_corr for steel bars in simulated pore solutions of different cements over 14 days. E_corr is commonly used to evaluate passivation behavior of steel bars, whereas R_p and i_corr provide a more accurate insights into the corrosion severity. At the initial immersion stage, the E_corr of steel rebars immersed in simulated solutions of OPC, CSA, HB-CSA, and G-CSA are all below −400 mV. The E_corr of steel bars in simulated HB-CSA and CSA pore solution shows a similar trend. In both cases, there is an initial increase in E_corr, followed by a decrease within the first 24 h of immersion. This is then followed by a gradual increase that stabilizes after 14 days. After this period, the E_corr of steel bars in simulated HB-CSA pore solution stabilizes at approximately −240 mV, whereas in simulated CSA pore solution it stabilizes at approximately −280 mV. For the steel bar immersed in simulated OPC pore solution, the E_corr increases steadily during the first 3 days, and reaches a stable value of about −250 mV after 14 days. Conversely, the E_corr of steel bar immersed in simulated G-CSA solution decreases rapidly at the early immersion stage and then stabilizes at approximately −650 mV after 14 days.

FIGURE 2

According to the standard of ASTM C876 and previously reported experimental and theoretical data (Liu G. et al., 2025), the correlation between corrosion potential and corrosion probability of steel bar is clarified as follows: E_corr > −250 mV indicates passivation, with a less than 10% chance of corrosion, whereas E_corr < −365 mV indicates rebar corrosion, with a>90% chance of corrosion. The corrosion potential criteria from ASTM C876 were used as a qualitative reference. Since a saturated calomel electrode (SCE) was used, all potentials were interpreted relative to SCE without conversion, and conclusions were based on comparative trends rather than absolute threshold values. According to this criterion, a high-quality passivation film has formed on the surface of steel bars in simulated HB-CSA and OPC pore solutions. By comparison, the steel bars immersed in simulated CSA pore solution may have undergone some corrosion, while those immersed in simulated G-CSA pore solution have experienced severe corrosion.

The evolution trend of R_p and i_corr of steel bars immersed in various simulated cement pore solutions is consistent with that of E_corr. Generally, a higher R_p or a lower i_corr indicates a better state of passivation. After 14 days of immersion, the steel bar immersed in simulated HB-CSA pore solution exhibits the highest R_p of 305 kΩ cm² and the lowest i_corr of 0.02 μA/cm² among the four cement types, indicating optimal passivation. In contrast, the R_p and i_corr of steel bars immersed in simulated pore solutions of OPC and CSA are approximately 50 kΩ cm² and 0.8 μA/cm², respectively. Moreover, for the steel bar immersed in simulated G-CSA pore solution, R_p remains zero, while i_corr reaches 37 μA/cm² after 14 days, indicating that G-CSA has no passivation effect on the steel bar and that the steel bar experiences more severe corrosion. Based on the results of E_corr, R_p, and i_corr, the passivation effectiveness of steel bars immersed in different simulated cement pore solutions can be ranked from best to worst as follows: HB-CSA, OPC, CSA, and G-CSA.

Figure 3 displays the electrochemical impedance spectroscopy (EIS) results of steel bars after 14 days of immersion in different simulated pore solutions, including typical Nyquist and Bode plots. For steel bar subjected to alkaline pore solution environments, the diameter of the capacitive semicircle in Nyquist plots serves as a critical indicator for evaluating surface passivation quality, where a larger semicircle diameter corresponds to a denser passivation film and better corrosion resistance.

FIGURE 3

After 14 days of immersion, the capacitive reactance radius of the steel bar in simulated HB-CSA pore solution is greater than that of steel bars in other simulated pore solutions. This finding suggests that HB-CSA positively affects the passivation of the steel bar. Additionally, as expected, the capacitive reactance radius of the steel bar in simulated G-CSA pore solution is significantly smaller than in other pore solutions, indicating weaker corrosion resistance for the steel bar. Both the LPR and EIS results confirm that G-CSA has a poor ability to passivate the steel bar. Conversely, although the alkalinity of HB-CSA is very low, it provides strong protection against steel bar passivation, and the steel bar has been fully passivated. The strong protective performance of HB-CSA can also be further understood by analyzing the elements and compounds present on the rebar surface.

Figure 4 displays SEM images and EDS results of the surface of steel bar immersed in various simulated pore solutions for 14 days. The surface of the steel bar immersed in simulated OPC pore solution shows a very dense surface with randomly distributed, very small microvoids. Meanwhile, EDS results indicate that the surface oxygen content is very low, suggesting that the steel has not undergone significant corrosion. This can be attributed to the high alkalinity of OPC.

FIGURE 4

For steel bars immersed in simulated HB-CSA pore solution, numerous granular products are observed, forming a dense shell. According to the EDS data, these hydration products are likely CaCO₃, which may have formed after carbonation of ettringite (Chen et al., 2020). These CaCO₃ coatings encapsulate the steel, preventing the steel bar from contacting the external environment and thus reducing corrosion, which may explain the excellent protective performance of HB-CSA with lower pH value.

For the steel bar in simulated CSA pore solution, although some CaCO₃ is observed, it is insufficient to form a protective shell, providing limited protection to the steel bar. For steel bars in simulated G-CSA pore solution, the surface film is damaged, indicating significant corrosion has occurred. The corrosion products appear as snowflake-like structures. Along with EDS results showing high sulfur content in the surface film, which may cause more severe corrosion of steel bars. Therefore, the poor protective effect of G-CSA on steel corrosion can be attributed to its low alkalinity and the coupling effect of the sulfur element in the pore solution.

Figure 5 shows the XPS Fe 2p spectra of steel rebar immersed in simulated pore solutions for 14 days. Three characteristic spectral components were identified in the Fe 2p spectra: Femet at 706.5 and 720.1 eV, Fe oxide hydroxide (FeOOH) at 711.0, 719.9, and 725.5 eV, and Fe oxide (FeO) at 709.6, 723, and 729.4 eV.

FIGURE 5

In simulated HB-CSA pore solution, the peak intensities of iron oxides (FeO and FeOOH) were substantially higher than those of metallic Fe at a depth range of 0–4 nm. With increasing probing depth, the peak intensities of the Fe oxides decrease, while the Femet peak increases. At depths of 6 nm and 10 nm, the Fe oxide peaks become nearly undetectable, indicating that the oxide film does not exceed a thickness of 6 nm. The Fe 2p spectra for the steel bar immersed in simulated OPC pore solution closely resembles those in HB-CSA solution, suggesting that its oxide film is also likely less than 6 nm thick. For the steel bar immersed in simulated CSA solution, a high-intensity Fe oxide (FeO) peak is observed at depths of 6 nm and 10 nm, but its intensity remains lower than that of the Femet peak. In contrast, for the steel bar in simulated G-CSA pore solution, the Fe oxides (FeO and FeOOH) maintain higher peak intensities than Femet even at a depth of 10 nm, with the Femet peak being nearly invisible. This indicates that G-CSA is only moderately effective in preventing steel corrosion.

3.2 Corrosion behavior of steel bars in simulated pore solution containing chloride ions

After 14 days of passivation in simulated pore solutions, NaCl was added to each solution to achieve a NaCl concentration of 0.2%. The steel bars were then immersed in the simulated pore solution with NaCl for an additional 14 days and subjected to electrochemical testing. Figure 6 shows the evolution of E_corr, R_p, and i_corr of steel bars exposed to simulated cement pore solutions containing 0.2 wt% NaCl over the 14-day immersion period.

FIGURE 6

Initially, steel bars immersed in simulated pore solutions of HB-CSA, OPC, and CSA exhibit relatively high E_corr due to the formation of a passivation layer on the steel bar surface. However, the introduction of chloride ions caused a sharp decline in E_corr within the first day. After 3 days, the E_corr gradually reached a stable state for all groups. At the end of the 14-day period, the E_corr of the steel bars immersed in simulated pore solutions of OPC, CSA, and HB-CSA was around −500 mV, indicating that chloride ions had induced noticeable corrosion in these environments. In contrast, the E_corr of the steel bar in simulated G-CSA pore solution decreased from the initial −600 mV to below −700 mV, suggesting a substantially higher corrosion tendency under chloride exposure.

The R_b of steel bars in different simulated pore solutions followed a trend similar to that observed for E_corr. Since the R_p of all specimens become very low after 14 days of immersion, the protective performance of the different simulated pore solutions against chloride-induced corrosion could not be clearly distinguished based solely on R_p.

In comparison, the difference in i_corr of steel bars immersed in different simulated pore solutions under a chloride ion environment is large enough. Initially, the i_corr of steel bars immersed in simulated pore solutions of HB-CSA, OPC and CSA were less than 2.5 μA/cm² since the passivation forms on the surface of these steel bars, while the steel bar in simulated G-CSA pore solution exceeded 15 μA/cm². After 1 day of immersion, the i_corr of all specimens increased rapidly, indicating accelerated corrosion. By 3 days, the i_corr approached a steady state. After 14 days of immersion, the steel bars immersed in simulated OPC pore solution exhibited the lowest i_corr, indicating a low level of corrosion. In contrast, the i_corr of steel bars immersed in simulated HB-CSA and CSA pore solutions were both around 7.5 μA/cm², suggesting that the anti-rust effect of HB-CSA was diminished in a chloride-ion environment. The steel bar immersed in simulated G·CSA pore solutions display the highest i_corr, approximately 17.5 μA/cm², indicating the most severe corrosion.

Figure 7 presents the Nyquist and Bode plots of steel bars immersed in simulated pore solutions containing 0.2% NaCl for 14 days. After 14 days of immersion, the capacitance arc radius of steel bars in different cement-simulated pore solutions increased in the following order: G-CSA < CSA < HB-CSA < OPC. The change in capacitance arc radius further proves that in the presence of chloride ions, HB-CSA has a weaker effect on inhibiting steel corrosion than OPC, and the corrosion of steel bars in the G-CSA simulated pore solution became increasingly severe.

FIGURE 7

Figure 8 displays SEM images and EDS results of steel bar surfaces after 14 days of immersion in simulated pore solutions containing 0.2% NaCl. After immersion in simulated HB-CSA pore solution, the surface of steel bar remained relatively smooth, and some CaCO₃ can still be detected. However, it is clear that CaCO₃ cannot effectively protect the steel bar in a chloride environment, as the surface oxygen content increases significantly.

FIGURE 8

In simulated CSA pore solution, more pronounced corrosion features are observed, with evident rust products distributed on the steel surface. In contrast, the surface of steel bar immersed in simulated OPC pore solution appears relatively smoother, and the extent of corrosion is lower compared with that in other simulated pore solutions under chloride exposure.

The most severe surface corrosion is observed on the steel bar immersed in simulated G-CSA pore solution, where obvious cracks are visible at 5,000× magnification. Furthermore, EDS analysis reveals a relatively high sulfur content in the surface corrosion layer. The presence of sulfur-containing species is likely to promote the formation of a loose and porous rust structure, which facilitates chloride ingress and thereby accelerates steel corrosion.

Figure 9 shows the XPS Fe 2p spectra of steel rebar immersed in simulated pore solutions containing chloride ions for 14 days, analyzed as a function of depth. A certain amount of Femet residue was detected on the outermost layer of steel bars immersed in simulated pore solutions of HB-CSA and OPC, while only FeOOH and Fe₃O₄ were found on the surface of steel bars in simulated CSA and G-CSA pore solutions, indicating that the surface corrosion of steel bars in simulated HB-CSA and OPC pore solutions is lighter than that of steel bars in immersed in simulated CSA and G-CSA pore solution.

FIGURE 9

As depth increases, these intensities of the Femet peak rise, indicating that the corrosion of steel bars is weakening. In addition, at this stage, the FeCl₂ and Fe₃O₄ appear in the surface layer of steel bars, and the corresponding curve peak is 713.9 eV and 724.4 eV, respectively. Fe₃O₄ was found at a depth of 10n m in steel bars, but the Fe₃O₄ content varied across different simulated pore solutions. The proportion of FeOOH and Fe₃O₄ in simulated pore solutions of OPC and HB-CSA is relatively low, while the proportion is relatively high in G-CSA and CSA. Therefore, in chloride-containing environments, although HB-CSA has a slightly weaker protective effect on steel bars than OPC, it is better than CSA, whereas G-CSA has a poorer protective effect, leading to increasingly severe steel corrosion under chloride-rich conditions (Liu et al., 2024).

3.3 Corrosion behavior of the steel bars embedded in the cement mortar immersed in the chloride ion solution

Figure 10 illustrates the changes in E_corr for steel bars embedded in various cement mortar samples, which were immersed in a 3.5% NaCl solution over a period of 180 days. At the initial stage of immersion, the E_corr of steel bar embedded in OPC mortar was relatively high. The results showed that during the 28 days of specimen curing, a relatively stable passive film was formed on the steel surface embedded in OPC mortar, whereas the steel bars embedded in the other three cement systems either failed to form a stable passive film or experienced partial depassivation and initial corrosion. At this point, the E_corr of the steel bars embedded in G-CSA mortar is higher than that in HB-CSA, indicating that G-CSA may provide a somewhat higher initial passivation level than HB-CSA in real mortar environments, which is inconsistent with the results obtained from simulated pore solution tests.

FIGURE 10

As immersion time increases, the E_corr of steel bar embedded in OPC mortar decreases rapidly, whereas the E_corr of steel bars embedded in HB-CSA, CSA, and G-CSA changes little, with the E_corr of the steel bars embedded in G-CSA matrix even showing a slight increase. After 28 days of immersion, the E_corr of steel bars in all four cement types gradually decreased following a similar trend. Ultimately, after 180 days of immersion, the Ecorr of steel bars in the G·CSA mortar stabilizes at approximately −650 mV, which remains slightly more positive than those in the other three cement systems.

Figure 11 shows the pore structure of four types of cement mortar immersed in a 3.5% NaCl solution for 180 days Table 5 presents the porosity and average pore diameter derived from the experimental results. Among the four types of cement, G-CSA mortar exhibited the lowest total porosity and a pore size distribution dominated by fine pores, accompanied by a reduced proportion of large capillary pores. Although OPC mortar showed lower overall porosity than both CSA and HB-CSA mortars, it contains a relatively higher fraction of large pores (>100 μm), which may facilitate localized transport pathways. The lower porosity and smaller pore size make G-CSA more impermeable, while CSA exhibits higher porosity and thus lower impermeability. When the matrix is highly impermeable, harmful substances such as water and chloride ions cannot penetrate the steel bar surface, thereby preventing corrosion. Therefore, the low alkalinity of G-CSA exacerbates corrosion for steel bars immersed in simulated G-CSA pore solution. The high impermeability of G-CSA mortar, which results from its lower porosity, leads to the highest E_corr for the embedded steel bar. Consequently, the degree of corrosion in this steel bar is the lowest compared to those embedded in the other three types of cement mortars.

FIGURE 11

Figure 12 shows the SEM images of the steel bar surface embedded in cement mortar after 180 days of immersion in a 3.5% NaCl solution. It can be observed from the images that the surface of steel bars embedded in G-CSA and OPC mortars are the smoothest. However, at 5,000× magnification, microcracks can be observed on the steel bar embedded in G-CSA. Significant rust spots were present on the surface of the steel bars embedded in HB-CSA and CSA. In addition, the granular CaCO₃ can still be observed on the surface of the steel bars embedded in HB-CSA mortar. However, the penetration of external CO₂ into the matrix is relatively slow due to the high impermeability of HB-CSA. These CaCO₃ particles may have later formed on the surface. Therefore, the protective effect of CaCO₃ on steel bar may be realized only after some delay. This can also explain why the E_corr of the steel bar embedded in HB-CSA increases 60 days prior to immersion in NaCl solution.

FIGURE 12

Table 6 illustrates the corrosion rate of steel bars embedded in cement mortar after 180 days of immersion in a 3.5% NaCl solution. The corrosion rate of steel bar embedded in G-CSA mortar was the lowest, at only 0.32%. This rate is 48.4% lower than that of steel bar embedded in OPC mortar. Steel bar embedded in HB-CSA mortar showed the second-lowest corrosion rate. In contrast, the highest corrosion rate is found in steel bars embedded in CSA mortar. In summary, G-CSA and HB-CSA mortars offer excellent protection against corrosion for steel bars exposed to a chloride ion environment for a long time. This protective effect is attributed to the dense matrix structure of these two types of cement mortar, which results in greater impermeability.

4 Conclusion

References

• 1

  ASTM G102-89 (2015). Standard Practice for Calculation of Corrosion Rates and Related Information from Electrochemical Measurements. West Conshohocken: American Society for Testing and Materials, 7.

  • Google Scholar
  • View reference in article

• 2

  BakeraA. T.AlexanderM. G. (2024). In situ concrete sewer performance: comparison of Portland cement, calcium sulfoaluminate cement, and calcium aluminate cement. Mater. Struct.57 (5), 114. 10.1617/s11527-024-02390-2

  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  BarneybackR. S.DiamondS. (1981). Expression and analysis of pore fluids from hardened cement pastes and mortars. Cem. Concr. Res.11 (2), 279–285. 10.1016/0008-8846(81)90069-7

  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  BescherE.KimJ. (2019). “Belitic calcium sulfoaluminate cement: history, chemistry, performance, and use in the United States,” in Proceedings of the 1st International Conference on Innovation in Low-Carbon Cement and Concrete Technology, 24–26.

  • Google Scholar
  • View reference in article

• 5

  ChaunsaliP.MondalP. (2015). Influence of calcium sulfoaluminate (CSA) cement content on expansion and hydration behavior of various ordinary Portland cement‐CSA blends. J. Am. Ceram. Soc.98 (8), 2617–2624. 10.1111/jace.13645

  • CrossRef
  • Google Scholar
  • View reference in article

• 6

  ChenB.HorgniesM.HuetB.MorinV.JohannesK.KuznikF. (2020). Comparative kinetics study on carbonation of ettringite and meta-ettringite based materials. Cem. Concr. Res.137, 106209. 10.1016/j.cemconres.2020.106209

  • CrossRef
  • Google Scholar
  • View reference in article

• 7

  ChenH.LiN.UnluerC.ChenP.ZhangZ. H. (2025). 200 years of Portland cement: nacltechnological advancements and sustainability challenges. J. Clean. Prod.486, 144500. 10.1016/j.jclepro.2024.144500

  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  DengZ.ZhuB.DavisS. J.CiaisP.GuanD.GongP.et al (2025). Global carbon emissions and decarbonization in 2024. Nat. Rev. Earth Environ.6 (4), 231–233. 10.1038/s43017-025-00658-x

  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  ElsnerB.RossiA. M. (1995). Effect of pH on electrochemical behaviour and passive film composition of stainless steels. Mater. Sci. Forum192, 225–236. 10.4028/www.scientific.net/msf.192-194.225

  • CrossRef
  • Google Scholar
  • View reference in article

• 10

  GalluccioS.BeirauT.PöllmannH. (2019). Maximization of the reuse of industrial residues for the production of eco-friendly CSA-belite clinker. Constr. Build. Mater.208, 250–257. 10.1016/j.conbuildmat.2019.02.148

  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  GarcésP.AndradeM. C.SaezA.AlonsoM. C. (2005). Corrosion of reinforcing steel in neutral and acid solutions simulating the electrolytic environments in the micropores of concrete in the propagation period. Corros. Sci.47 (2), 289–306. 10.1016/j.corsci.2004.06.004

  • CrossRef
  • Google Scholar
  • View reference in article

• 12

  GartnerE.SuiT. B. (2018). Alternative cements clinkers. Cem. Concr. Res.114, 27–39. 10.1016/j.cemconres.2017.02.002

  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  HuoG.JiangX.SunX.LiH.ShiH. (2024). Performance of high-belite calcium sulfoaluminate cement subjected to hydrochloric acid and sulfuric acid. Front. Mater.10, 1282919. 10.3389/fmats.2023.1282919

  • CrossRef
  • Google Scholar
  • View reference in article

• 14

  LiJ.MaB.ZhouC. Y.YangY. (2014). Study on mechanism of chemical activation for minerals of high belite-calcium sulfoaluminate clinker. J. Sustain. Cement-Based Mater.3 (1), 13–23. 10.1080/21650373.2013.843476

  • CrossRef
  • Google Scholar
  • View reference in article

• 15

  LiuG.LiJ.ZhangY. (2024). Corrosion of carbon steels subjected to chloride and sulfate in simulated concrete pore solutions with different pH. Constr. Build. Mater., 440: 137445, 10.1016/j.conbuildmat.2024.137445

  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  LiuG.LiC.ZhangY. (2025). Study on the correlation between corrosion potential and polarization resistance of steel reinforcement. Electrochimica Acta522, 145957. 10.1016/j.electacta.2025.145957

  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  LiuY.TangZ.ZhouX.GuoD.WangY. (2025). Pore characteristics and properties of calcium sulfoaluminate (CSA) foam concrete: effects of foam content and water-to-binder ratio. Case Stud. Constr. Mater.24, e05708. 10.1016/j.cscm.2025.e05708

  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  LothenbachB.BaryB.Le BescopP.ThomasS.LeterrierL. N. (2010). Sulfate ingress in Portland cement. Cem. Concr. Res.40 (8), 1211–1225. 10.1016/j.cemconres.2010.04.004

  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  MorozovY.PellegriniS.MontemorM. F. (2022). Sulfate-induced corrosion of carbon steel reinforcement in fresh calcium sulfoaluminate binders. Corrosion78 (9), 813–824. 10.5006/3910

  • CrossRef
  • Google Scholar
  • View reference in article

• 20

  NejadB. M.EnferadiS.AndrewR. (2025). A comprehensive analysis of process-related CO2 emissions from Iran's cement industry. Clean. Environ. Syst.16, 100251. 10.1016/j.cesys.2024.100251

  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  OgawaK.RoyD. M. (1982). C₄A₃ hydration, ettringite formation, and its expansion mechanism: III. Effect of CaO, NaOH and NaCl conclusions. Cem. Concr. Res.12, 247–256. 10.1016/0008-8846(82)90011-4

  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  ParkS.JeongY.MoonJ.LeeN. (2021). Hydration characteristics of calcium sulfoaluminate (CSA) cement/portland cement blended pastes. J. Build. Eng.34, 101880. 10.1016/j.jobe.2020.101880

  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  PooniJ.RobertD.GiustozziF.SetungeS.XieY. M.XiaJ. (2020). Novel use of calcium sulfoaluminate (CSA) cement for treating problematic soils. Constr. Build. Mater.260, 120433. 10.1016/j.conbuildmat.2020.120433

  • CrossRef
  • Google Scholar
  • View reference in article

• 24

  RamezanianpourA. A.PourkhorshidiA. R.SobhaniJ.MoodiF. (2021). Durability of concrete containing blended cements in harsh marine environments: 18 years exposure study. Constr. Build. Mater.299, 123863. 10.1016/j.conbuildmat.2021.123863

  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  RevertegatE.RichetC.GegoutP. (1992). Effect of pH on the durability of cement pastes. Cem. Concr. Res.22 (2-3), 259–272. 10.1016/0008-8846(92)90064-3

  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  SirtoliD.WyrzykowskiM.RivaP.TortelliS.MarchiM.LuraP. (2019). Shrinkage and creep of high-performance concrete based on calcium sulfoaluminate cement. Cem. Concr. Compos.98, 61–73. 10.1016/j.cemconcomp.2019.02.006

  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  SongF. M. (2012). Predicting the chemistry, corrosion potential and corrosion rate in a crevice formed between substrate steel and a disbonded permeable coating with a mouth. Corros. Sci.55, 107–115. 10.1016/j.corsci.2011.10.013

  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  SrivastavaA.KumarR.LakhaniR. (2021). “Low energy/low carbon eco-cementitious binders as an alternative to ordinary Portland cement,” in Handbook of Smart Materials, Technologies, and Devices: Applications of Industry 4.0 (Cham: Springer International Publishing), 1–23.

  • Google Scholar
  • View reference in article

• 29

  SuW.PiZ.NieS.LiH.XuM.ZhouJ. (2025). Influence and mechanism of anhydrite on the sulfuric acid resistance of full-solid waste high-belite calcium sulfoaluminate cement. Constr. Build. Mater.484, 141816. 10.1016/j.conbuildmat.2025.141816

  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  SunZ.NieS.ZhouJ.LiH.ChenZ.XuM.et al (2022). Hydration mechanism of calcium sulfoaluminate-activated supersulfated cement. J. Clean. Prod.333, 130094. 10.1016/j.jclepro.2021.130094

  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  TanB.OkoronkwoM. U.KumarA.MaH. (2020). Durability of calcium sulfoaluminate cement concrete. J. Zhejiang University-SCIENCE A21 (2), 118–128. 10.1631/jzus.a1900588

  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  TangH.QianJ.QinJ.LiZ.JiaX.YueY.et al (2023). Electrochemical behavior of steel bars in magnesium phosphate cement. Cem. Concr. Res.169, 107167. 10.1016/j.cemconres.2023.107167

  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  ValD. V.StewartM. G. (2003). Life-cycle cost analysis of reinforced concrete structures in marine environments. Struct. Safety25 (4), 343–362. 10.1016/s0167-4730(03)00014-6

  • CrossRef
  • Google Scholar
  • View reference in article

• 34

  VinothG.MoonS. W.MoonJ.KuT. (2018). Early strength development in cement-treated sand using low-carbon rapid-hardening cements. Soils Found.58 (5), 1200–1211. 10.1016/j.sandf.2018.07.001

  • CrossRef
  • Google Scholar
  • View reference in article

• 35

  WangW.ChenH.LiX.ZhuZ. (2017). Corrosion behavior of steel bars immersed in simulated pore solutions of alkali-activated slag mortar. Constr. Build. Mater.143, 289–297. 10.1016/j.conbuildmat.2017.03.132

  • CrossRef
  • Google Scholar
  • View reference in article

• 36

  WięckowskiA. (2020). Automating CSA cement-based reinforced monolithic ceiling construction. Automation Concr. Struct.111, 103051. 10.1016/j.autcon.2019.103051

  • CrossRef
  • Google Scholar
  • View reference in article

• 37

  ZhangN. (2022). Time-Varying Law of Interfacial Bonding Performance Between FRP-reinforcement and Low Alkali corrosion-resistant Cement Concrete. Tianjin, China: Hebei University of Technology, 14.

  • Google Scholar
  • View reference in article

• 38

  ZhangW.NieS.XuM.ZhouJ.LiH. (2022). Research progress on activation of high belite calcium sulphoaluminate cement. Bull. Chin. Ceram. Soc.41 (9), 2979–2992.

  • Google Scholar
  • View reference in article

• 39

  ZhouJ.YinS.FuQ.WangQ.HuangQ.WangJ. (2021). Microbial-induced concrete corrosion under high-salt conditions: microbial community composition and environmental multivariate association analysis. Int. Biodeterior. Biodegrad.164, 105287. 10.1016/j.ibiod.2021.105287

  • CrossRef
  • Google Scholar
  • View reference in article