Analysis of the Stability and Reactivity of Carbonated Steel Slag Powder as a Supplementary Cementitious Material

Abstract

In order to address the poor volume stability and low reactivity of steel slag powder (SS) as a supplementary cementitious material (SCM), this study investigates a microbial-assisted carbonation method for its enhancement. Using untreated SS as a control, we compared the performance and microstructure of carbonated steel slag powder (CSS) and bio-mineralized steel slag powder (BSS). Results indicate that, compared to CSS, BSS exhibits a more significant reduction in the content of f-CaO and f-MgO (from 6.25% and 3.19% to 0.8% and 1.36%, respectively) and a greater improvement in 7-day and 28-day activity indices (from 59% and 72% to 78% and 87%), leading to markedly enhanced volume stability and reactivity. Calculations show that each ton of BSS can sequester 114.2 kg of CO₂, and it achieves a cement replacement ratio exceeding 30%. The utilization of BSS as an SCM not only addresses the inherent technical challenges of steel slag powder but also creates dual environmental benefits through emission reduction and active carbon sequestration, demonstrating significant potential for advancing the low-carbon transition in the construction materials industry.

1. Introduction

Steel slag, a primary solid waste generated during steelmaking, has caused severe environmental and resource issues due to its massive stockpiling. Currently, the utilization rate of steel slag exceeds 90% in developed regions such as Europe and the United States, whereas China’s utilization rate remains around 30% [1,2,3], indicating an urgent need to enhance its resource efficiency. Benefiting from a mineral composition similar to cement, utilizing steel slag as a supplementary cementitious material to partially replace cement represents an effective pathway for its large-scale utilization [4,5,6]. This approach not only reduces cement consumption and associated carbon emissions but also delivers significant economic and environmental benefits. However, the poor volume stability and low reactivity of steel slag remain critical bottlenecks restricting the broader application of this technology. On one hand, the presence of f-CaO and f-MgO in steel slag adversely affects its use in construction materials. The hydration of f-CaO and f-MgO can form Ca(OH)₂ and Mg(OH)₂, causing significant expansion of the solid phase of the steel slag. Apart from that, the content of highly reactive silicate minerals (such as C₃S) in steel slag is much lower than that in Portland cement, and its C₃S structure is denser, resulting in a slower hydration rate. Thus, improving the volume stability and enhancing the reactivity of steel slag are crucial prerequisites for its safe and resourceful utilization.

Numerous studies have shown that carbonation is an effective method to improve the volume stability and reactivity of steel slag [7,8]. As early as 1990, Seiftriz first proposed mineral carbonation: the reaction of metal oxides with CO₂ to form insoluble solid carbonates [9]. Natural minerals rich in alkali metals such as calcium and magnesium can react with dissolved or ionized CO₂ to form carbonate precipitates. Since alkaline oxides (e.g., CaO, MgO) account for over 60% of steel slag composition, these oxides can capture and sequester CO₂, converting it into stable carbonates. Thus, steel slag exhibits significant carbonation potential. The carbonation reaction of steel slag can be represented by Equations (1)–(8):

$$CaO+{CO}_2\to {CaCO}_3$$

(1)

$$MgO+{CO}_2\to {MgCO}_3$$

(2)

$$1/2\left[2CaO·{SiO}_2\right]+{CO}_2\to 1/2{SiO}_2+{CaCO}_3$$

(3)

$$1/3\left[3CaO·{SiO}_2\right]+{CO}_2\to 1/3{SiO}_2+{CaCO}_3$$

(4)

$$f-CaO+H_{2}O\to {Ca\left(OH\right)}_2$$

(5)

$${Ca\left(OH\right)}_2+{CO}_2\to {CaCO}_3$$

(6)

$$f-MgO+H_{2}O\to {Mg\left(OH\right)}_2$$

(7)

$${Mg\left(OH\right)}_2+{CO}_2\to {MgCO}_3$$

(8)

From the above reaction equation, it can be seen that the carbonation reaction can convert f-CaO and f-MgO in steel slag into stable carbonate precipitates, thereby improving the volumetric stability of the steel slag [10]. Additionally, due to the enhanced reactivity of steel slag after carbonation, when used as an SCM, it can accelerate the heat release rate and increase the total heat of hydration in cement paste, thereby promoting the hydration of cement clinker and ultimately improving the mechanical properties of cement mortar at different curing ages [11,12,13]. This enhancement effect primarily stems from two key components generated after the carbonation of steel slag: CaCO₃ and amorphous SiO₂. Among them, CaCO₃ functions in the cement system through a triple mechanism: (1) filling the pores within the cement-based materials; (2) providing nucleation sites for hydration products; and (3) reacting chemically with aluminum-containing mineral phases in the cement [14]. Meanwhile, amorphous SiO₂, with its high pozzolanic activity, can undergo a secondary reaction with Ca(OH)₂ produced during cement hydration, generating additional C-S-H gel, thereby further optimizing the microstructure and mechanical properties of the cement-based materials [15,16].

However, the CaCO₃ formed during carbonation creates a dense “passivation layer” on the steel slag surface, which severely hinders the combination of Ca²⁺ and carbonate ions, leading to rapid reaction slow down or even cessation. Meanwhile, the extensive conversion of Ca²⁺ into CaCO₃ precipitates results in insufficient ion supply in later stages, further limiting carbonation progress. As a result, conventional carbonation is often incomplete, with core regions of the slag remaining largely unreacted [17,18]. Coupled with the slow diffusion of CO₂ within the slag and its low solubility in water, the carbonation degree typically remains below 60%. To enhance carbonation efficiency and better improve the volume stability and reactivity of steel slag, researchers have proposed a novel microbial-assisted carbonation method [19,20,21]. Microbial enzymes significantly promote CO₂ absorption and conversion, accelerate CO₃²⁻ formation, and increase CaCO₃ nucleation sites, thereby overcoming the kinetic limitations and passivation layer issues associated with traditional carbonation. Furthermore, microbial metabolism releases OH⁻, maintaining an alkaline environment conducive to CaCO₃ deposition [22,23]. Based on a synthesis of existing literature, microbial action has been demonstrated to significantly enhance the carbonation efficiency of steel slag: Bacillus mucilaginous can reduce the apparent activation energy of hard-burned lime and periclase, thereby promoting the reaction of f-CaO and f-MgO [24]; microbial mineralization can improve the carbonation reactivity of γ-C₂S [25]; and carbonic anhydrase-producing bacteria can capture CO₂ and accelerate its reaction with Ca²⁺ [26]. Building upon these findings, this study employs microbial-assisted carbonation to improve the volume stability and reactivity of steel slag powder. Experimental results show that, compared to carbonated steel slag powder, the microbial-mineralized steel slag powder exhibits not only significantly higher hydration activity but also superior volume stability, laying a foundation for its use as a SCM.

2. Materials and Methods

2.1. Materials

The cement used in this experiment is Type P·I 52.5 Portland cement, sourced from Zhucheng 97 Building Materials Co., Ltd., Weifang, China. The main chemical composition of the cement is shown in

Table 1. Major chemical composition of cement (%).

CaO	SiO2	Al2O3	Fe2O3	MgO	MnO	Na2O	P2O5	SO3	LOI
58.04	22.01	5.31	2.82	5.93	0.05	0.37	0.09	4.11	1.65

. The physical properties of the cement used in this experiment were tested, and the results are presented in

Table 2. Physical properties of cement.

SSA m2/Kg	Density g/cm2	W/C %	ST /Min		Flexural Strength/Mpa		Compressive Strength/Mpa
			IST	FST	3 d	28 d	3 d	28 d
390	3.15	36	182	250	6.1	7.3	26	54.9

. All indicators meet the national standard requirements for the main performance of universal Portland cement. Figure 1 shows the particle size distribution curve of the cement. It can be observed that the cement particle sizes are mainly concentrated in the range of 0.1 to 1 μm.

The steel slag powder used in this study is derived from converter steel slag and provided by Henan Wuhu Environmental Protection Technology Co., Ltd., Zhengzhou, China. The particle size distribution of the steel slag powder is shown in Figure 2. The majority of particles fall within the 0.1–1 μm and 10–100 μm ranges. Particle size significantly influences the CO₂ diffusion rate, total absorption rate, and consequently, the overall carbonation reaction kinetics. Smaller particles provide a larger specific surface area, leading to a faster CO₂ absorption rate and higher hydration activity [27]. Therefore, the fraction in the 0.1–1 μm range exhibits relatively rapid carbonation and high hydration activity, whereas the 10–100 μm fraction shows slower carbonation kinetics and lower reactivity. Figure 3 and

Table 3. Main chemical composition of steel slag powder (%).

CaO	SiO2	Al2O3	Fe2O3	MgO	MnO	Na2O	P2O5	SO3	V2O5	K2O
32.8	13.4	6.26	26.4	8.28	4.2	0.25	0.88	1.1	0.23	0.19

present the main mineral composition and chemical composition of the steel slag powder, respectively. Three types of samples were prepared from the steel slag powder used in this study using different treatment methods: SS (raw steel slag powder), CSS (carbonated steel slag powder), and BSS (microbial-mineralized steel slag powder). This study focuses solely on analyzing the volume stability and reactivity of steel slag powder as an SCM, while the production and grinding processes of the steel slag powder are not discussed herein.

This study utilized Bacillus mucilaginous, a bacterium capable of secreting carbonic anhydrase (CA). The wet carbonation of steel slag primarily involves three stages: dissolution of CO₂, leaching of Ca²⁺/Mg²⁺, and precipitation of carbonates. The CA produced by this bacterium not only catalyzes CO₂ hydration but also enhances ion leaching, thereby accelerating carbonate precipitation. The enzymatic catalysis of CO₂ hydration proceeds as follows [22]:

(1)

Zn²⁺ in the enzyme’s active site facilitates the deprotonation of H₂O, forming E·ZnOH⁻;

(2)

E·ZnOH⁻ performs a nucleophilic attack on CO₂ bound in the hydrophobic pocket, forming E·ZnHCO₃⁻;

(3)

HCO₃⁻ in E·ZnHCO₃⁻ is displaced by a water molecule, yielding E·ZnH₂O and HCO₃⁻;

(4)

HCO₃⁻ is converted into CO₃²⁻ under the action of E·ZnOH⁻;

(5)

Ca²⁺ is adsorbed onto the negatively charged bacterial cells;

(6)

The bacterial cells serve as nucleation sites, promoting the deposition of mineralization products.

A schematic diagram illustrating the acceleration of steel slag carbonation by Bacillus mucilaginosus is shown below (Figure 4).

Figure 4. Schematic diagram of the accelerated carbonation process of steel slag by CA-type microorganisms.

Figure 4. Schematic diagram of the accelerated carbonation process of steel slag by CA-type microorganisms.

Studies have shown that a single CA molecule can catalyze the hydration of 1.4 × 10⁶ CO₂ molecules per second to form H₂CO₃. Compared to conventional wet carbonation, the CO₂ hydration reaction rate can be increased by 10⁷ times with the involvement of CA [28]. The presence of CA accelerates the CO₂ hydration reaction, enhances the conversion to CO₃²⁻ in the solution, and leverages the nucleation sites and cation enrichment effect provided by microbial cells to overcome the limitations of traditional carbonation where CaCO₃ precipitation forms a passivating layer on the steel slag surface. This enables sustained progression of subsequent carbonation reactions, thereby facilitating faster and more extensive formation of CaCO₃ precipitates [19].

The schematic diagrams of the carbonation and microbial mineralization processes for the steel slag powder are shown in Figure 5 and Figure 6, respectively. A 7.5% Bacillus mucilaginosus solution was first prepared, with the bacterial dosage set at 1.5% of the steel slag powder mass, maintaining a constant mass ratio of microbial solution to steel slag powder at 0.2 [29]. Specifically, 3.75 g of Bacillus mucilaginosus was dissolved in 50 mL of deionized water. The mixture was stirred uniformly and then incubated in a microbial shaker at 30 °C and 180 rpm for 24 h to facilitate spore germination. After this incubation period, the pH of the microbial solution was measured using pH test paper and found to be approximately 6.5. The carbonation experiment was conducted in a cylindrical reactor. CO₂ gas with a concentration exceeding 99% and a pressure of 0.3 MPa was introduced continuously into the reactor at a constant flow rate of 0.1 L/min, regulated by a mass flow controller. Prior to CO₂ introduction, 50 mL of the microbial solution was sprayed onto 250 g of steel slag powder and mixed thoroughly to provide the necessary moisture for the reaction. The carbonation reaction proceeded at 20 ± 2 °C for 24 h. The carbonation process for the control steel slag powder sample followed the same procedure described above, except that no microbial solution was added. Instead, 50 mL of deionized water was sprayed onto 250 g of steel slag powder before the reaction to maintain the same moisture condition.

2.2. Methods

2.2.1. Comparison of the Characteristics of Steel Slag Powder Under Different Treatments

To analyze the main mineral compositions of the three differently treated steel slag powder samples, this study employed X-ray diffraction (XRD). The samples were first dried in a vacuum oven at 60 °C for 24 h. Prior to testing, 20 wt% α-Al₂O₃ was added as an internal standard and thoroughly mixed for quantitative XRD analysis. Thermogravimetric analysis (TG-DTG) was used to determine mass loss under a nitrogen atmosphere, with the temperature ranging from room temperature to 1000 °C. Fourier transform infrared spectroscopy (FTIR) was applied to identify functional groups, scanning wavenumbers from 400 cm⁻¹ to 4000 cm⁻¹. The micro-morphologies of the samples were examined using field emission scanning electron microscopy (SEM).

2.2.2. Calculation of CaCO₃ Crystal Size in Steel Slag Powder with Different Treatments

The Scherrer formula, in conjunction with MDI Jade 9 software, was employed to calculate the crystallite size of CaCO₃ in the three differently treated steel slag powder samples. To determine the average crystallite size of CaCO₃, the crystallite sizes corresponding to five distinct diffraction peaks from each sample group were calculated individually and then averaged. The specific calculation method is shown in Formula (1).

$$D={\displaystyle \frac{k\lambda }{{\beta }_{1/2}\cos \theta }}$$

(9)

In the formula, D represents the crystallite size of CaCO₃, nm; k is the Scherrer constant with a value of 0.89; λ is the wavelength of the X-ray radiation used in the experiment, which is 1.5406 Å; β_1/2 denotes the full width at half maximum (FWHM) of the selected diffraction peak in radians; and θ is the diffraction angle corresponding to the selected peak in degrees.

2.2.3. The Effect of Differently Treated Steel Slag Powder on Cement Hydration Products

To investigate the influence of the three differently treated steel slag powder samples on the hydration products of cement, composite cement pastes were prepared with a water-to-binder ratio of 0.35 by thoroughly mixing 70 wt% cement and 30 wt% steel slag powder. These pastes were designated as SS cement paste, CSS cement paste, and BSS cement paste, respectively. The three types of cement pastes were cured at a temperature of 20 ± 2 °C and a relative humidity of over 95% for 7 days and 28 days, respectively. Subsequently, the cement pastes were crushed, ground into powder, and dried in a vacuum oven at 60 °C for 24 h. Prior to testing, 20 wt% α-Al₂O₃ was added as an internal standard. The hydration products of the cement pastes at 7 days and 28 days were then analyzed using X-ray diffraction.

2.2.4. Method for Detecting the Content of f-CaO and f-MgO in Steel Slag Powder

In this section, the contents of both f-CaO and f-MgO were determined by titration three times, and the average value of the three experimental results was taken as the final data for this study. According to GB/T 38216.3-2023, the content of f-CaO in steel slag powder is determined using the EDTA titration method and thermogravimetric analysis. The specific operational procedures are as follows (Figure 7).

Figure 7. Method for detecting the combined amount of f-CaO and Ca(OH)₂ in steel slag powder.

Figure 7. Method for detecting the combined amount of f-CaO and Ca(OH)₂ in steel slag powder.

The total mass fraction of f-CaO and Ca(OH)₂ can be calculated using Formula (2).

$$w_1={\displaystyle \frac{T_{CaO}\times V}{m\times 1000}}\times 100\%$$

(10)

$$T_{CaO}=c\hspace{0.17em}\left(\mathrm{EDTA}\right)\times 56.08$$

(11)

where w₁ represents the combined mass fraction of f-CaO and Ca(OH)₂, %; T_CaO is the mass of CaO equivalent to 1 mL of the EDTA standard titration solution, mg/mL; c (EDTA) is the concentration of the EDTA standard titration solution, 0.015 mol/L; 56.08 is the molar mass of CaO, g/mol. V₁ is the volume of the EDTA standard titration solution consumed during titration, mL; m is the mass of the sample, g; 1000 is the unit conversion factor.

The Ca(OH)₂ content is determined using thermogravimetric analysis. The percentage mass loss represented by the weight loss step between 400 °C and 550 °C on the thermogravimetric curve is denoted as w₂. The Ca(OH)₂ content in the steel slag can be calculated using Formula (4), denoted as w₃:

$$w_3=4.1111\times 0.7567\times w_2$$

(12)

where w₃ is the mass fraction of Ca(OH)₂ (calculated as CaO), %; 4.1111 is the ratio of the molecular weight of Ca(OH)₂ to that of H₂O; 0.7567 is the ratio of the molecular weight of CaO to that of Ca(OH)₂; w₂ is the mass fraction of H₂O released from the decomposition of Ca(OH)₂, %.

The f-CaO content in the steel slag can be calculated using Formula (5):

$$w=w_1-w_3$$

(13)

where w₁ is the combined mass fraction of f-CaO and Ca(OH)₂, %; w₃ is the mass fraction of Ca(OH)₂ (calculated as CaO), %.

The content of f-MgO in steel slag powder is determined using the EDTA titration method. The specific operational procedure is as follows (Figure 8).

Figure 8. Detection method of f-MgO content in steel slag powder.

Figure 8. Detection method of f-MgO content in steel slag powder.

The content of f-MgO can be calculated using Formula (6):

$$w_{MgO}={\displaystyle \frac{T_{MgO}\times \left(V_2-V_1\right)}{m\times 1000}}\times 100\%$$

(14)

$$T_{MgO}=c\hspace{0.17em}\left(\mathrm{EDTA}\right)\times 40$$

(15)

where T_MgO is the mass of MgO equivalent to 1 mL of the EDTA standard titration solution, mg/mL; c (EDTA) is the concentration of the EDTA standard titration solution, 0.015 mol/L;40 is the molar mass of MgO, g/mol; m is the mass of the sample, g; V₂ is the total volume of the EDTA solution consumed by calcium and magnesium, mL; V₁ is the volume of the EDTA standard solution consumed during titration for determining the Ca(OH)₂ content, mL.

2.2.5. Method for Testing the Volume Stability of Cement

According to GB/T 1346-2024, the soundness of cement paste is tested using the boiling method. A cement paste with a water-to-binder ratio of 0.35 is prepared. Three different treated steel slag powder samples are used to replace 30% of the cement content, respectively, to prepare composite cement pastes, while a pure cement paste with the same water-to-binder ratio is prepared as the control group. The different types of cement pastes are each placed into two Le Chatelier molds, and the specimens are immediately moved to a curing environment with a temperature of 20 ± 2 °C and a relative humidity of not less than 95% for 24 ± 2 h. After curing, the specimens are removed, and the initial expansion value (A) of each specimen is measured using the Le Chatelier apparatus, accurate to 0.5 mm. The specimens are then placed in a boiling tank, heated to boiling within 30 ± 5 min, and maintained at a constant boil for 180 ± 30 min. After boiling, the final expansion value (C) of each specimen is measured using the Le Chatelier apparatus, accurate to 0.5 mm. The average difference between the expansion values after boiling (C-A) for the two specimens is calculated, with the result accurate to 0.5 mm. If the average value does not exceed 5.0 mm and the difference between the two expansion values is less than 3.0 mm, the soundness of the cement paste is considered qualified; otherwise, it is deemed unqualified. Each sample was tested in triplicate, and the final expansion value data are presented as the mean ± standard deviation in the Discussion section. A schematic diagram of the boiling method for testing the volume soundness of cement paste is shown below (Figure 9).

Figure 9. Schematic diagram of the cement soundness testing process by boiling method.

Figure 9. Schematic diagram of the cement soundness testing process by boiling method.

2.2.6. Calculation of CO₂ Absorption Rate and CaCO₃ Formation

The CO₂ absorption rate and CaCO₃ formation amount are calculated based on the mass loss of the steel slag powder caused by the endothermic decomposition of CaCO₃ within the temperature range of 520–780 °C [30,31]. The specific formulas are as follows:

$$C{O}_2 \mathit{uptake}={\displaystyle \frac{W_{520}-W_{780}}{W_{780}}}\times 100\%$$

(16)

$$\mathit{Quantity} \mathit{of} {\mathit{CaCO}}_3=\left(W_{780}-W_{520}\right)\times {\displaystyle \frac{100}{44}}$$

(17)

W₅₂₀ and W₇₈₀ represent the weights of steel slag powder at 520 °C and 780 °C, respectively.

2.2.7. Method for Testing the Activity of Steel Slag Powder

According to GB/T 20491-2017, cement mortar specimens of size 40 mm × 40 mm × 160 mm were prepared using Chinese ISO standard sand sourced from Xiamen ISO Standard Sand Co., Ltd., Xiamen, China. Cement mortar was prepared with a water-cement ratio of 0.5 and a cement-sand ratio of 1:3. Composite cement mortar was produced by replacing 30% of the Portland cement with each of the three differently treated steel slag powder samples. The specimens were cured under conditions of 20 ± 2 °C temperature and over 95% relative humidity, and the compressive strength of the cement mortar was tested at 7 days and 28 days. Each sample was tested in triplicate, and the final strength data are presented as the mean ± standard deviation in the Discussion section. The activity index of the steel slag powder was calculated using Formula (10) as follows [32]:

$$R=\left(C_1/C_0\right)\times 100\%$$

(18)

where R is the activity index; C₀ is the average compressive strength of the pure cement mortar, MPa; C₁ is the average compressive strength of the mortar with 30% steel slag powder replacing cement, MPa. The activity index of the steel slag powder must exceed 65% at 7 days and 80% at 28 days.

3. Discussion

3.1. The Effect of Aifferent Treatment Methods on the Properties of Steel Slag Powder

Figure 10 shows the XRD patterns of three differently treated steel slag powder samples. Strong Ca(OH)₂ diffraction peaks can be observed in the SS samples. Since steel slag powder possesses self-cementing properties, the formation of Ca(OH)₂ in SS is likely attributed to the hydration of f-CaO in a humid environment [33]. The presence of Ca(OH)₂ is the primary cause of volume expansion in steel slag powder, indicating its poor volume stability at this stage. Correspondingly, strong diffraction peaks of CaCO₃ and amorphous SiO₂ gel were observed in both CSS and BSS, indicating that after carbonation and mineralization reactions, most of the silicate minerals and free oxides in the steel slag powder had been transformed into stable carbonate precipitates and amorphous SiO₂ gel. The presence of these two reaction products significantly enhanced the volume stability and reactivity of CSS and BSS.

Based on Formula (1) and analysis using MDI jade 9 software, the crystallite sizes of CaCO₃ in the three differently treated steel slag powder samples were calculated, as shown in

Table 4. Parameters of Scherrer equation for CaCO₃ crystallite size in steel slag powder samples under different treatments.

	Parameter	2θ/°	cos θ	β1/2	D/nm
Sample
SS		22.936	0.318	0.980	4.4
		29.491	0.141	0.967	10.055
		39.464	0.292	0.941	4.99
		47.596	0.154	0.915	9.923
		48.598	0.170	0.911	8.853
CSS		26.646	0.086	0.973	16.384
		29.507	0.152	0.967	9.328
		39.597	0.319	0.941	4.567
		43.489	0.32	0.929	4.612
		48.638	0.153	0.911	9.836
BSS		23.079	0.98	0.346	4.403
		29.476	0.967	0.141	10.055
		39.473	0.941	0.173	8.423
		47.561	0.915	0.238	6.296
		48.568	0.912	0.221	6.802

. The data represent calculation results at different diffraction angles. The average crystallite sizes of CaCO₃ in the three samples were subsequently calculated, and the results are presented in Figure 11. The average CaCO₃ crystallite sizes in the three steel slag powder samples were 7.64 nm (SS), 8.95 nm (CSS), and 7.20 nm (BSS). This indicates that the presence of Bacillus mucilaginosus influences the nucleation and growth processes of CaCO₃. In other words, Bacillus mucilaginosus not only accelerates the deposition of CaCO₃ but also affects its crystallite size [34,35].

Quantitative analysis of the main components in the three differently treated steel slag powder samples was conducted using the XRD-Rietveld method, while the contents of f-CaO and f-MgO were determined by titration. The results are shown in Figure 12. The contents of free oxides and silicates decreased in both CSS and BSS, with new carbonates and amorphous phases formed. In CSS, the f-CaO and f-MgO contents decreased from 6.25% and 3.19% to 2.75% and 2.1%, respectively, while in BSS, they further decreased to 0.8% and 1.36%. According to the specifications of GB/T 20491 “Steel slag powder for cement and concrete,” the f-CaO content should be below 4%. Therefore, both CSS and BSS used in this experiment meet the national standard requirements. The lower degree of conversion of f-CaO and f-MgO in CSS explains why the volume stability of CSS cement paste is inferior to that of BSS cement paste. It is worth noting that under the action of microorganisms, f-CaO in the steel slag powder was almost completely consumed, while only about 60% of f-MgO reacted, indicating that f-MgO is more difficult to react, i.e., the apparent activation energy of f-MgO is higher than that of f-CaO [36]. The BSS has the lowest free oxide content and generates more CaCO₃ and amorphous SiO₂ gel. This indicates that microbial mineralization promotes the transformation of more calcium-containing mineral phases in the steel slag powder, leading to a more thorough carbonation reaction. The CaCO₃ contents in the three differently treated steel slag powder samples were 16.38% (SS), 17.95% (CSS), and 21.02% (BSS), respectively. The amorphous SiO₂ content also increased from 13.4% (SS) to 14.3% (CSS) and 16.39% (BSS). CaCO₃ can act as a seed crystal for silicate phases, promoting crystal growth [37], while amorphous SiO₂ gel can participate in cement hydration and pozzolanic reactions [38]. Both contribute to enhancing the reactivity of steel slag powder as an SCM.

As shown in Figure 13, the micromorphology of the three differently treated steel slag powder samples was analyzed, and the internal elements were scanned using an energy dispersive spectrometer. It can be observed that the morphology of the steel slag powder changed before and after modification, primarily due to alterations in its chemical composition. The surface of the unmodified SS sample appeared relatively smooth. In contrast, after carbonation or microbial mineralization treatment, the surface of the steel slag powder was eroded, with numerous fine products adhering to it, ultimately leading to an increase in surface roughness. For CSS and BSS, the changes in chemical composition were mainly attributed to the intrusion of CO₂ during carbonation and the reaction of calcium-containing mineral phases with CO₂. Furthermore, compared to SS and CSS, the C content in BSS increased significantly, and the elements C, Ca, and O exhibited a high degree of overlap near Area 3, indicating the formation of CaCO₃ from the reaction between CO₂ and Ca²⁺. This suggests that a larger amount of CaCO₃ was generated inside BSS through microbial mineralization, which is consistent with the XRD analysis results.

Figure 14 shows the TG and DTG curves of three steel slag powder samples subjected to different treatments. As shown in the TG curves, the most significant weight loss for SS occurs in the temperature range of 400–550 °C. Within this range, the mass loss of steel slag powder is primarily due to the endothermic decomposition of Ca(OH)₂ [39,40]. The mass loss rates of Ca(OH)₂ for the three steel slag powders are 2.56% (SS), 0.26% (CSS), and 0.1% (BSS), respectively. The minimal mass loss of CSS and BSS indicates that most of the Ca(OH)₂ in the steel slag powder has been converted into carbonate precipitates during the modification process. Meanwhile, in the temperature range of 580–720 °C, all three samples exhibit significant mass loss, which is attributed to the decomposition of CaCO₃ [41,42]. The mass loss rates of CaCO₃ for the three steel slag powders are 8.03% (SS), 8.73% (CSS), and 10.25% (BSS), respectively, indicating that carbonation and microbial mineralization promote the formation of a considerable amount of CaCO₃ in the steel slag powder. Generally, the absorption and exothermic peaks of pure CaCO₃ mainly occur between 800 and 900 °C. However, when the CaCO₃ crystals contain atoms such as Mg or Fe, the decomposition temperature may decrease [29]. Similarly, the DTG curves also show that SS exhibits the most significant weight loss in the range of 400–500 °C, while BSS shows the most significant weight loss in the range of 700–800 °C. The weight losses in these two temperature ranges are caused by the endothermic decomposition of Ca(OH)₂ and CaCO₃, respectively. The TG and DTG curves corroborate each other. According to Equations (8) and (9), the CO₂ absorption rate and CaCO₃ formation amount were calculated. After 24 h of carbonation reaction, the CO₂ absorption rates of CSS and BSS reached 9.57% and 11.42%, respectively, indicating that microbial participation enhances the carbon sequestration capacity of the steel slag powder and that the carbonation rate of BSS is higher than that of CSS. Correspondingly, the CaCO₃ contents in CSS and BSS are 17.95% and 21.02%, respectively. This further demonstrates that the carbonation reaction proceeds more thoroughly under microbial action, thereby promoting the formation of more CaCO₃. The thermogravimetric analysis results are consistent with the XRD results, both supporting the above conclusions. This conclusion aligns with the findings of Liu et al. [8].

Figure 15 shows the FTIR spectra of the three differently treated steel slag powder samples. The characteristic peak of steel slag powder near 3440 cm⁻¹ can be attributed to the O-H stretching vibration of Ca(OH)₂ and the bound water in hydration products [43]. The characteristic bands of CaCO₃ in steel slag powder include the asymmetric stretching vibration band of C-O at 1420 cm⁻¹, the out-of-plane bending vibration band of C-O at 872 cm⁻¹, and the in-plane bending vibration of C-O at 712 cm⁻¹ [44,45]. The spectra of CSS and BSS are essentially identical, which is consistent with the XRD results, indicating that microbial participation does not alter the crystal form of the carbonation products. However, in the C-O out-of-plane bending vibration band at 872 cm⁻¹, significant differences are observed between CSS and BSS. The absorption peak of BSS is higher and sharper, suggesting higher crystallinity and better symmetry of the mineralization products. The larger absorption peak area of BSS indicates a higher CaCO₃ content compared to CSS, which aligns with the XRD and TG analysis results. The FTIR results demonstrate that the presence of microorganisms enhances the crystallinity of the reaction products. Microorganisms can serve as nucleation sites, promoting the growth and crystallization of CaCO₃ [21], thereby resulting in mineralization products with improved symmetry and crystallinity.

3.2. The Impact of Differently Treated Steel Slag Powder on the Hydration Products of Cement Paste

The reason why steel slag powder treated by carbonation or microbial mineralization can be used as an SCM lies not only in its improved volume stability but, more importantly, in its enhanced hydration reactivity. This enhancement fundamentally stems from the generation of substantial CaCO₃ and amorphous SiO₂ gel during the carbonation or mineralization process. CaCO₃ can promote the hydration reaction of Portland cement, primarily due to its nucleation effect, filling effect, and chemical reactions with C₃A [46]. The nucleation effect of CaCO₃ essentially involves its role as highly dispersed fine particles that provide abundant and easily accessible nucleation sites for cement hydration products. During the early stages of hydration, hydration products preferentially nucleate and grow on the surfaces of dispersed CaCO₃ particles, preventing excessive concentration of hydration products around the original cement particles. This significantly accelerates the hydration process and optimizes the microstructure of the cement paste [47,48]. The filling effect of CaCO₃ is a physical optimization process. The fine particles of CaCO₃ fill the voids between cement particles, which not only optimizes the particle size distribution of the cement and reduces the porosity of the cement paste but also refines the pore size of the cement, enhancing its density and thereby improving its strength and durability [49,50,51]. The chemical role of CaCO₃ in cement hydration mainly refers to its reaction with C₃A to form monocarbonate (Mc), hemicarbonate (Hc), and stabilize ettringite (AFt) structures, among others [52,53,54]. The amorphous silica gel exhibits high pozzolanic reactivity, enabling it to participate in a secondary reaction with calcium hydroxide (Ca(OH)₂) from cement hydration and produce additional C-S-H gel [55,56]. After Mc forms in the cement, it can act as a nucleation site, which in turn accelerates the hydration of C₃S and C₃A, producing more AFt and C-S-H gel, which are the main contributors to cement strength [57,58]. The smaller the CaCO₃ crystallites, the more pronounced their promoting effect on cement hydration. The fundamental reason is that smaller crystallites have a larger specific surface area. The surface area of cement particles themselves is limited. Smaller CaCO₃ crystallites distributed in the cement paste provide abundant nucleation sites for hydration products (mainly C-S-H gel and Ca(OH)₂). Additionally, smaller CaCO₃ crystallites have higher reactivity, enabling them to consume water more rapidly and form an early skeleton, thereby contributing to the development of early strength and further promoting the reaction process of the entire cement hydration system [59,60]. Figure 16 shows the XRD patterns of the hydration products of cement pastes containing different treated steel slag powder samples at 7 days and 28 days.

As shown in Figure 16a, at 7 days, cement hydration produces a large amount of Ca(OH)₂, and diffraction peaks of C₂S and C₃S are still observed in the patterns, indicating that hydration is not yet complete at this stage. The hydration products of the three cement pastes are essentially the same, mainly composed of Ca(OH)₂, C₂S, C₃S, CaCO₃, Ca₂FeO₅, Ca₂(Al,Fe)₃(SiO₄)₃(OH), AFt, C-S-H gel, and carboaluminates, among other new products. However, the diffraction peak intensity of CaCO₃ in the CSS cement paste is higher than that in the SS cement paste, and this phenomenon is also notably observed in the BSS cement paste. This is because the steel slag powder forms more CaCO₃ internally after carbonation and mineralization, and CaCO₃ can act as a crystal nucleus to promote the cement hydration process [61]. The appearance of new products such as Ca₂FeO₅ and Ca₂(Al,Fe)₃(SiO₄)₃(OH) is due to the reaction of Ca(OH)₂ with oxides like SiO₂, Al₂O₃, and FeO, forming new phases of metal oxides. AFt, as one of the important early hydration products of ordinary Portland cement, is formed by the reaction of C₃A in the cement with gypsum (CaSO₄·2H₂O). In addition to the C-S-H gel intrinsically formed by cement hydration, the amorphous silica gel in the steel slag powder reacts with Ca(OH)₂ to form additional C-S-H gel. Meanwhile, diffraction peaks of Mc and Hc are observed in all three cement paste samples, resulting from the chemical reaction between CaCO₃ in the steel slag powder and C₃A in the cement paste [62,63]. Due to the higher content of CaCO₃ and amorphous SiO₂ gel in BSS, the intensities of the aforementioned phases vary to some extent. As shown in Figure 16b, at 28 days, as the hydration reaction continues, the diffraction peak intensities of C₂S and C₃S decrease, while the diffraction peak intensity of Ca(OH)₂ increases compared to that at 7 days. The diffraction peaks of Mc are stronger, indicating the formation of more Mc. Hc, as an unstable transitional phase, has mostly converted into Mc at this stage, a phenomenon more evident in the CSS and BSS cement pastes. Mc precipitates as fine flakes or gel, filling the gaps between cement particles and the pores of larger hydration products (such as C-S-H gel and AFt). This “filling effect” reduces the porosity of the cement paste, making the structure denser and significantly improving the early and later strength of the cement. In the later stages of hydration, as the dissolved gypsum (CaSO₄·2H₂O) in the cement paste is depleted, the diffraction peaks of AFt weaken and are replaced by diffraction peaks of monosulfoaluminate (AFm). After 28 days of hydration, the diffraction peak intensity of CaCO₃ decreases due to its continuous reactions with Ca(OH)₂, C₃A, and metal oxides [64,65]. This phenomenon is particularly noticeable in the SS and CSS cement pastes. Both CSS and BSS participate in and enhance the degree of cement hydration. For the BSS cement paste, the enhancing effect is strengthened due to the presence of mineralization products—specifically, more CaCO₃ with smaller crystal sizes and amorphous silica gel.

3.3. Study on the Effect of Steel Slag Powder with Different Treatments on the Volume Stability of Cement

Figure 17 shows the results of volume soundness testing for different types of cement paste using the boiling method. Since SS contains unstable substances such as f-CaO and f-MgO, its us as an SCM typically leads to volume expansion. When SS is used as an SCM, the average expansion value of two cement specimens after boiling is 6.49 mm, indicating poor volume soundness of the cement paste, which does not comply with national standards. Due to the carbonztiong process, in which f-CaO and f-MgO are converted into stable carbonate precipitates, the average expansion value of two cement specimens is significantly reduced to 3.64 mm when CSS is used as an SCM, indicating improved volume soundness. Microbial participation in carbonation further promotes the conversion of more silicate minerals and free oxides into stable carbonate precipitates, reducing the expansion value to only about 1 mm. As a result, the volume soundness of the cement paste is further enhanced and becomes very close to that of pure cement, which is consistent with the XRD test results of the steel slag power. This demonstrates that when BSS is used as an SCM, the volume soundness index of the cement paste meets the required standards.

Figure 17. Volume stability of different types of cement slurry.

Figure 17. Volume stability of different types of cement slurry.

3.4. Comparison of the Activity of Steel Slag Powder Under Different Treatments

When steel slag powder is used as an SCM, it must not only exhibit good volume stability but also provide strength equivalent to that of cement, meaning the steel slag powder must possess high reactivity. Figure 18 shows the flexural and compressive strengths of cement mortar prepared by replacing 30% of cement with three differently treated steel slag powder samples. It can be observed that, compared to pure cement mortar, the flexural and compressive strengths of the cement mortar decrease to some extent when steel slag powder replaces part of the cement. However, compared to SS, the carbonated steel slag powder enhances the strength of the cement mortar. This strengthening effect is primarily attributed to the CaCO₃ and amorphous SiO₂ gel generated during the carbonation process, which participate in the cement hydration reaction. On one hand, CaCO₃ directly acts as an active component to promote hydration; on the other hand, in later reactions, CaCO₃ reacts with aluminum-containing minerals in the cement to form Hc and Mc, among other products. Additionally, the SiO₂ gel reacts with Ca(OH)₂ to produce more C-S-H gel, both of which contribute to improving the strength of the cement paste. It is worth noting that BSS demonstrates a superior strengthening effect, which is closely related to its higher CaCO₃ content and finer crystallite size. The larger specific surface area and higher reactivity enable it to more effectively promote the cement hydration process, thereby achieving higher strength than the other two sample groups.

The activity indices of three differently treated steel slag powder samples were calculated according to Formula (10), and the results are shown in Figure 19. The 7-day and 28-day activity indices of SS are 59% and 72%, respectively, both of which do not meet the national standards. Carbonation treatment not only improves the volume stability of the steel slag powder but also enhances its hydration activity, with the 7-day and 28-day activity indices of CSS increasing to 67% and 78%, respectively. BSS demonstrates even better modification effects, with 7-day and 28-day activity indices of 78% and 87%, respectively, indicating that microbial involvement in the carbonation reaction further enhances the activity of the steel slag powder. BSS participates in and promotes the hydration process of the cement paste, generating more hydration products that contribute to the strength of the cement, which is consistent with the XRD study results of the cement paste.

3.5. Environmental Sustainability of BSS Cement Paste

According to Equation (8) and TG analysis results, the CO₂ absorption rate of 250 g BSS in this study reached 11.42%. Based on this calculation, one ton of BSS can sequester 114.2 kg of CO₂, indicating that BSS possesses significant CO₂ sequestration potential and is an ideal carbon fixation material. When BSS was used to replace 30% of cement in producing composite cementitious materials, tests on soundness and activity showed that the expansion value of the BSS-cement paste after boiling was below 5 mm—very close to that of pure cement paste—confirming satisfactory volume soundness. Additionally, the 28-day activity index of BSS reached 87%, meeting the national standard. It can thus be concluded that BSS can replace at least 30% of cement. Producing one ton of cement emits approximately 0.6–0.9 tons of CO₂. Replacing 30% of cement with BSS implies a direct 30% reduction in cement production emissions. This creates dual environmental benefits of “emission reduction + active sequestration,” which are difficult to achieve with conventional low-carbon cementitious materials alone. It is believed that with ongoing advances in carbonation technology and processes, the cement replacement rate of BSS could reach 40–50% in the future.

4. Conclusions

Through this research, the following conclusions can be drawn:

(1)

When steel slag powder is used as an SCM, it generally suffers from poor volume stability and low reactivity. Carbonation treatment can promote the transformation of silicate minerals and free oxides in steel slag powder, thereby improving its volume stability and enhancing its reactivity. Additionally, introducing microorganisms during carbonation can further accelerate the above transformation process, promoting the formation of more stable carbonates from silicate minerals and free oxides, thus improving the cementitious properties and long-term stability of steel slag powder. In this study, the f-CaO and f-MgO contents of BSS were reduced to 0.8% and 1.36%, respectively, reaching safe levels. The activity indices of BSS at 7 days and 28 days also significantly improved compared to SS, reaching 78% and 87%, respectively.

(2)

There are differences in the chemical composition and microscopic morphology between CSS and BSS. Due to the promoting effect of microorganisms, BSS exhibits a higher content of mineralized CaCO₃, smaller crystal grain size, and higher crystallinity of biogenic CaCO₃, which is more conducive to promoting the hydration reaction of cement.

(3)

The fundamental reason why steel slag powder promotes cement hydration is that its carbonation and mineralization products participate in the cement hydration process. In the early hydration stage, CaCO₃ on the surface of steel slag powder acts as a nucleation site, promoting cement hydration and improving the early strength of cement. In the later hydration stage, CaCO₃ reacts with C₃A and Ca(OH)₂ to form new products such as Hc and Mc. Simultaneously, the pozzolanically active amorphous SiO₂ gel also participates in the hydration process, generating additional C-S-H gel, which densifies the cement paste structure and enhances compressive strength.

(4)

This study demonstrates that each ton of BSS can sequester 114.2 kg of CO₂ and achieve a cement replacement rate exceeding 30%. The dual mechanism of substitution and sequestration establishes BSS as a pivotal pathway toward enhancing sustainability and environmental stewardship in the construction materials sector.