1. Introduction

Eutrophication remains a major environmental concern in aquatic systems, primarily driven by anthropogenic nutrient inputs to coastal environments (Dai et al., 2023;Malone and Newton, 2020). Excessive nitrogen and phosphorus inputs increase nutrient loading in coastal environments, accelerating the eutrophication process. In particular, internal phosphorus release from sediment plays a critical role in sustaining eutrophication (Kong et al., 2023). The accumulation of organic matter in sediment enhances this internal loading and contributes to hypoxic conditions in bottom water, even after algal blooms have subsided (Dai et al., 2023). Therefore, immobilizing nutrients within sediment is essential for mitigating their release into overlying water.

Steel slag, a byproduct of the steel manufacturing process, has attracted attention as a cost-effective material for environmental remediation due to its high adsorption capacity and alkaline nature. It is primarily composed of various metal oxides, including calcium oxide (CaO), iron oxides (Fe₂O₃), silica oxide (SiO₂), aluminium oxide (Al₂O₃), and manganese oxide (MnO). The steel slag matrix consists mainly of calcium silicate and dicalcium ferrite phases, with embedded RO phases (FeO-MnO) (Ragipani et al., 2019). Previous study demonstrated that steel slag can effectively remove contaminants such as Ni²⁺, H₂PO₄⁻, and NH₄⁺ through mechanisms including adsorption, precipitation, and ligand exchange (Kadirova et al., 2015). Furthermore, steel slag has been shown to raise soil pH, reduce methane emissions, and enhance crop yield (Wang et al., 2018). Upon contact with water, CaO hydrolyzes to form Ca²⁺ and OH⁻, increasing the pH of the surrounding environment (Barca et al., 2012;Vu et al., 2021). Sediment pH strongly influences phosphorus speciation and binding, as phosphorus can associate with Ca, Fe/Mn, and Al compounds (Barca et al., 2012;Wang et al., 2021;Zhao et al., 2022). Moreover, steel slag exhibits a rough surface and a high specific surface area, which increases with decreasing particle size (Ragipani et al., 2019;Vu et al., 2021;Xue et al., 2009). These properties enhance adsorption and precipitation processes, improving pollutant removal efficiency (Shi et al., 2022;Wu et al., 2024).

Previous studies have often focused on chemically or thermally modified steel slag to enhance its adsorption performance. For instance, Jha et al. (2008) reported improved removal of PO₄³⁻ (40.6%) and NH₄⁺ (7.2%) using modified slag. Similarly, Touch et al. (2019) observed reductions in NH₄⁺ and PO₄³⁻ concentrations in overlying water when using slag-mixed sediment in SMFC. Thermally treated steel slag (1-2 mm) exhibited high removal efficiencies for total nitrogen (9.67%) and total phosphorus (72.76%) in constructed wetlands (Wang et al., 2021), while finely ground slag (0.15 mm) achieved up to 90.6% removal of hydroxyethylidene diphosphonic acid (Wu et al., 2024). These studies highlight the effectiveness of modified steel slag but also indicate limitations due to increased cost and complexity for large-scale applications.

The dissolution behavior of steel slag plays a key role in pollutant removal and can be described using the shrinking core model (Huang et al., 2024;Ragipani et al., 2019). In this process, Ca²⁺ diffuses from the interior of the slag matrix and reacts with oxyanions such as HCO₃⁻ and CO₃²⁻, forming precipitates on the particle surface (Luo and He, 2021). The extent of Ca²⁺ release is strongly influenced by particle size, with finer particles exhibiting greater surface area and higher dissolution rates (Ragipani et al., 2019). In contrast, coarser particles expose more RO phases (FeO-MnO), which can reduce calcium extraction efficiency. This phenomenon is purely geometric, where the presence of a higher Ca-silicate surface area in finer slags results in the higher release of Ca²⁺ (Ragipani et al., 2019). Meanwhile, coarse slag particles exposed many new RO globules (FeO-MnO) as Ca-silicate dissolved, resulting in lower Ca extraction efficiency. Calcium dissolution rate affects the pH in the sediment matrix (Barca et al., 2012) and directly influences the behavior of pollutant with respect to slag particles, where it can be adsorbed, precipitated, or exchange ions or ligands. Hence, steel slag particle size significantly impacts its capability to remove pollutants by diffusion of metal ions from its matrix.

Although previous studies have demonstrated the effectiveness of modified steel slag, the use of untreated slag remains more practical for real-world applications due to its lower cost and simpler processing requirements. Since particle size directly influences surface area and dissolution behavior (Vu et al., 2021;Xue et al., 2009), it is an important factor controlling pollutant removal efficiency. Using steel slag byproducts to remove pollutants from sediment will support the recycling of steel slag waste while simultaneously remediating the sediment. This promotes a sustainable approach within the circular economy by treating waste using other waste products. Therefore, this study investigates the effect of steel slag particle size, without pretreatment, on the remediation of contaminated coastal sediments.

2. Materials and methods

2.1 Sediment and steel slag preparation

The sediment used in this study was collected from Jukdo Island, Republic of Korea. The collected sediment was stored in a 13 L high-density polyethylene (HDPE) container and sealed to prevent moisture loss prior to use. Shell fragments and other impurities were removed by sieving the sediment through a 1 mm mesh. The sediment was then thoroughly homogenized before the experiment. Steel slag was supplied by Changshin Industrial Development Co., Ltd. (Republic of Korea). Prior to use, the slag was washed with distilled water for 30 minutes to remove excess surface CaO (Kim et al., 2024), followed by drying in an oven at 100 °C for 12 h. After drying, the slag was pulverized and sieved into three particle size ranges: < 0.6 mm, 1.0-5.0 mm, and > 5.0 mm. The chemical composition of the steel slag used is presented in Fig. 1.

2.2 Experimental setup

Four experimental conditions were prepared: control (no steel slag), < 0.6 mm, 1.0-5.0 mm, and > 5.0 mm. Each condition was set up in 1 L HDPE bottles. The sediment was mixed thoroughly with the respective steel slag and placed into the bottles to a depth of 10 cm. Distilled water was then gently added to form an overlying water layer of approximately 6 cm above the sediment surface. During the experimental period, distilled water was periodically added to compensate for evaporation losses. All experimental setups were maintained under controlled laboratory conditions for 40 days. Previous studies reported incubation periods ranging from 7 to 680 days for evaluating the effects of steel slag on sediment (Kim et al., 2012;Kim et al., 2020;León-Romero et al., 2018;Touch et al., 2019;Liyun et al., 2017;Zhao et al., 2025b). The 40-day incubation period adopted in this study was considered sufficient to allow steel slag dissolution and subsequent sediment conditioning processes to occur. Physicochemical changes require time to develop and stabilize before measurable differences between treatments can be observed. The significant variations observed after 40 days indicate that the incubation period was adequate to capture the influence of steel slag particle sizes on sediment remediation performance. No external aeration or mixing was applied during the incubation period.

2.3 Analytical methods

Sediment pH and oxidation-reduction potential (ORP) were measured using a pH/ion meter (LAQUA F-73, HORIBA, Japan). For nutrient analysis, sediment pore water was extracted by centrifugation. Approximately 50 mL of sediments were transferred into conical tubes and centrifuged at 4000 rpm for 20 minutes (VARISPIN 4, CRYSTE). The supernatant was then filtered through a 0.45 µm syringe filter (Hyundai/Micro, Japan) to remove suspended particles. The concentrations of chemical oxygen demand (COD), ammonia (NH₃-N), nitrate (NO₃-N), nitrite (NO₂-N), and phosphate (PO₄-P) were determined using a spectrophotometer (DR3900, HACH, USA) following standard Hach methods. The sediment remains were dried in the oven at 100 °C for 6 h and then pyrolyzed at 600 °C for 4 h to investigate water content (WC) and loss on ignition (LOI) using equations (1) and (2), respectively.

2.4 Statistical analysis

All experimental data were analyzed using R software (version 4.4.3). One-way analysis of variance (ANOVA) followed by Tukey’s post hoc test was employed to assess the effects of steel slag particle size on sediment physicochemical properties and nutrient concentrations. A p value of < 0.05 indicates a statistical difference.

3. Results and Discussion

3.1 Steel slag effects on pH and ORP

Fig. 2(a) and 2(b) present the pH values of the overlying water and sediment pore water at the end of the experiment. In the overlying water, the highest pH was observed in control (pH 7.9), followed by > 5.0 mm (pH 7.8), 1.0-5.0 mm (pH 7.6), and < 0.6 mm (pH 7.5) cases. In contrast, the sediment pore water exhibited the highest pH in < 0.6 mm case (pH 8.0), followed by 1.0-5.0 mm (pH 7.8), > 5.0 mm (pH 7.6), and control (pH 7.5). Overall, the sediment pH increased with decreasing steel slag particle size. This increase in pH can be attributed to the hydrolysis of free CaO (f-CaO) and MgO present in the steel slag, as shown in equations (3) and (4) (Yan et al., 2000):

Smaller slag particles provide a larger specific surface area, enhancing the dissolution of Ca²⁺ and Mg²⁺ ions. As a result, finer particles promote greater alkalinity in the sediment pore water, as pH correlates with calcium extraction efficiency (Said et al., 2015). The sediment pH increased from pH 7.6 in the > 5.0 mm case to pH 8.0 in the < 0.6 mm case. Ragipani et al. (2019) found that steel slag leaches less Ca- and more of Mg- and Fe-rich phases as the leaching progression continues. In addition to hydroxide generation, the released Ca²⁺ ions can react with bicarbonate (HCO₃⁻) to form calcium carbonate (CaCO₃), accompanied by the production of CO₂(5a). The dissolved CO₂ subsequently forms carbonic acid (H₂CO₃), which dissociates into H⁺ and carbonate species (5b)-(5c). Furthermore, hydrolysis of multivalent metal ions such as Fe³⁺ and Al³⁺ also contributes to proton generation (6)-(8). These processes increase the H⁺ concentration in the pore water, creating a concentration gradient that drives diffusion of protons upward into the overlying water.

This mechanism explains the contrasting pH trends observed between overlying water and sediment pore water (Fig. 2(a) and 2(b)). While the sediment becomes more alkaline due to CaO hydrolysis, proton diffusion from the sediment to the overlying water results in a relative decrease in pH in the water column. The effect is more pronounced for smaller particle sizes, which exhibit higher dissolution rates.

The ORP measurement of the sediment reflects the aggregation of all redox couples in the sediment (Miao et al., 2006). The ORP of the system is closely linked to pH through the Nernst equation (9):

where E = ORP, E₀ = standard electrode potential, R = universal gas constant, T = temperature, F = Faraday constant, and [H⁺] = hydrogen ion concentration.

As shown in Fig. 2(c) and 2(d), the ORP in the overlying water increased with decreasing particle size, reaching the highest value in the < 0.6 mm case (181.0 mV). In contrast, sediment ORP decreased with increasing particle size, with the lowest value observed in the 1.0-5.0 mm case (-167.0 mV).

This inverse relationship between pH and ORP is consistent with the Nernst equation, where an increase in pH leads to a shift toward a more negative redox potential. The ORP values shift to more positive in the overlying water as the pH decreases with steel slag size (Fig. 2(a) and 2(c)). On the contrary, the ORP in sediment shifts to more negative values as the pH increases with decreasing steel slag size (Fig. 2(b) and 2(d)). Additionally, the dissolution of the slag matrix exposes RO phases (FeO-MnO), releasing metal ions into the sediment (Ragipani et al., 2019). These metal species can undergo reduction reactions, such as the conversion of ferric iron (Fe³⁺) to ferrous iron (Fe²⁺), further lowering the ORP (Miao et al., 2006). Ragipani et al. (2019) found that more RO globules were released with larger steel slag particles, which could lead to more negative ORP in the sediment of 1.0-5.0 mm and > 5.0 mm. The anaerobic degradation of organic matter also consumes electron acceptors, such as Mn⁴⁺, Fe³⁺, and SO₄²⁻, producing reduced species (Mn²⁺, Fe²⁺, S²⁻), which contribute to the development of strongly reducing conditions (Touch et al., 2017). Overall, the results indicate that steel slag particle size significantly influences both pH and ORP through enhanced metal dissolution and redox reactions. Smaller particles increase sediment alkalinity, while simultaneously promoting reducing conditions within the sediment matrix.

3.2 Nutrient removal by steel slag

Figure 3(a) shows the dissolved inorganic nitrogen (DIN) concentrations in sediment pore water. The lowest DIN concentration was observed in the 1.0-5.0 mm case (0.9 mg L⁻¹), followed by the > 5.0 mm (1.2 mg L⁻¹), while both the < 0.6 mm and control cases exhibited similar values (1.3 mg L⁻¹). In this study, DIN primarily consisted of NH₃-N. The degradation of organic matter releases inorganic nitrogen into the sediment pore water in the form of ammonium (NH₄⁺) and NO₃⁻ (Langenfeld et al., 2021). Under anaerobic conditions, nitrogen removal occurs through denitrification and anaerobic ammonia oxidation (anammox), producing nitrogen gas (N₂) (Dai et al., 2023). Steel slag has been reported to support the growth of denitrifying bacteria, such as Sulfurimonas paralvinellae, in marine environments (Ogawa et al., 2020). The denitrification process involves the sequential reduction of NO₃⁻ to N₂ via intermediate species (Takai et al., 2006) with the help of enzymes (Black et al., 2016) (10)-(13). The equilibrium between NH₃ and NH₄⁺ is strongly dependent on pH (14) (Langenfeld et al., 2021):

At higher pH, the equilibrium shifts toward NH₃ formation, whereas lower pH favors NH₄⁺. In this study, the increase in sediment pH due to steel slag dissolution likely influenced this equilibrium and enhanced nitrogen removal pathways. In addition, the negatively charged surface of steel slag under alkaline conditions promotes the adsorption of cationic species such as NH₄⁺ (Cha et al., 2006;Ramakrishna and Viraghavan, 1997;Wu et al., 2024). Larger particle sizes expose more of Mg-rich phases following Ca-silicate dissolution (Ragipani et al., 2019), which can further contribute to NH₄⁺ adsorption (Kadirova et al., 2015). This explains why the 1.0-5.0 mm case exhibited lower DIN concentrations compared to the < 0.6 mm case (Fig. 3(a)). Moreover, the 1.0-5.0 mm steel slag provides a favorable balance between surface area and pore structure, enhancing intra-particle diffusion and adsorption capacity. Qiu et al. (2015) reported that 1 mm steel slag has a higher N₂-BET surface (0.46 m² g⁻¹) than coarser particles, which supports greater NH₄⁺ adsorption. Therefore, the 1.0-5.0 mm particle size appears to be optimal for DIN removal due to the combined effects of adsorption and microbial processes.

Fig. 3(b) shows the phosphate concentration in the sediment pore water. In the control, phosphate increased from the initial concentration (0.51 mg L⁻¹) to 0.70 mg L⁻¹, indicating phosphate release under reducing conditions. In anoxic and strongly reducing sediments, Fe-P compounds become thermodynamically unstable, leading to the release of phosphate into the pore water (Boynton et al., 2018;Dai et al., 2023). In contrast, the addition of steel slag significantly reduced phosphate concentrations, with the lowest value observed in the < 0.6 mm (0.4 mg L⁻¹). This reduction can be attributed to both adsorption and precipitation mechanisms. Dissolution of steel slag through adsorbed H⁺ releases multivalent metal cations such as Ca²⁺, Fe²⁺/Fe³⁺, and Al³⁺, which provide positively charged sites for phosphate adsorption (Kaplan Bekaroglu et al., 2021;Ragipani et al., 2019).

Finer steel slag particles exhibit a larger specific surface area, increasing the number of available adsorption sites (Vu et al., 2021). Additionally, grinding reduces crystallinity and enhances the reactivity of slag particles (Zhao et al., 2025a). Multivalent metal cations released to the slag surface induced the release of hydroxyl radicals (Kaplan Bekaroglu et al., 2021), leading to subsequent hydroxylation and metal hydrolysis into Si-OH and Al-OH (Wang et al., 2021). Phosphate can be immobilized through ligand exchange with surface hydroxyl groups (Wang et al., 2021) (15)-(16) and through precipitation as Ca-P (17), particularly under alkaline conditions (Jha et al., 2008;Vu et al., 2021):

Therefore, the superior phosphate removal observed in the < 0.6 mm case is primarily attributed to enhanced surface area, increased metal ion release, and favorable precipitation conditions.

Fig. 4 shows the COD concentrations in the sediment pore water. The lowest COD concentration was observed in the < 0.6 mm case (141.7 mg L⁻¹), followed by the 1.0-5.0 mm (150.0 mg L⁻¹), control (157.5 mg L⁻¹), and > 5.0 mm (171.7 mg L⁻¹) cases. The absence of excess H⁺ and OH⁻ ions binding to steel slag allows more organic matter to be absorbed, favorably between an initial pH of 6 and 8 (Van Tuyen et al., 2022). While the initial pH of the sediment is also neutral, this study shows that the introduction of < 0.6 mm steel slag particles slightly promotes COD removal. The reduction in COD can be attributed to both biological and physicochemical processes. Steel slag provides a suitable surface for microbial colonization (Ogawa et al., 2020), particularly for bacteria involved in organic matter degradation. For instance, marine Bacteroidetes, Flavimarina sp., are known to degrade complex organic compounds, contributing to COD reduction (Dutschei et al., 2023). Finer steel slag particles offer a larger surface area (Vu et al., 2021) for microbial attachment, enhancing biodegradation processes. In addition, metal cations released from steel slag can promote flocculation of organic matter by reducing electrostatic repulsion or forming cross-linked structures (Kim et al., 2024). This facilitates the removal of organic matter from the sediment pore water. Overall, the results indicate finer steel slag particles (< 0.6 mm) are more effective for COD reduction due to enhanced microbial activity and increased metal-mediated flocculation.

3.3 Steel slag influence on WC and LOI

Fig. 5 shows the changes in (a) WC and (b) LOI for all experimental conditions. Water content represents the combined contribution of pore water, mineral-bound water, and surface-adsorbed water (Woo et al., 2025). After 40 days, WC decreased slightly from the initial value (50.7%) to approximately 47-48% across all cases. This reduction can be attributed to sediment consolidation and pore structure collapse over time (Nuttle et al., 1990). However, no significant differences in WC were observed between the control and steel slag-treated cases, which is in contrast to the water content observed from the study of Woo et al. (2025). This suggests that the addition of steel slag at the applied dosage (2.5% by dry weight of sediment) had a negligible effect on sediment water retention properties. LOI, commonly used as an indicator of organic matter content in sediments (TC WI:2003 (E) 2003) remained unchanged at approximately 9% after the experiment (Fig. 5(b)). This indicates that the addition of steel slag did not significantly alter the organic matter content of the sediment. As steel slag is primarily composed of inorganic metal oxides, its incorporation does not contribute to the organic fraction of the sediment. Overall, the results demonstrate that steel slag addition, regardless of particle size, has minimal influence on the physical properties of sediment, such as water content and organic matter.

3.4 Relationship among pH, ORP, nutrient removal, WC and LOI

Steel slag dissolution primarily released alkalinity into the sediment through hydrolysis of alkaline minerals such as CaO and MgO. The observed changes in pH, ORP, and nutrient concentrations suggest that these parameters were closely interconnected through steel slag-induced sediment conditioning. Finer steel slag particles increased sediment pH, promoting phosphate immobilization through adsorption and precipitation reactions with metals released from the steel slag, thereby lowering phosphate concentrations. In contrast, steel slag addition reduced sediment ORP, creating a more reducing environment that may influence microbial nutrient transformation processes. The combined effects of altered pH and redox conditions may have affected nitrogen cycling pathways, including denitrification and anammox, contributing to the observed reduction in DIN concentrations. Therefore, phosphate removal appeared to be primarily governed by steel slag-induced physicochemical processes, whereas DIN removal was likely associated with biologically mediated processes influenced by changes in sediment redox conditions. Although significant differences were observed in pH, ORP, and nutrient concentrations, steel slag addition did not substantially affect the physical properties of the sediment, as indicated by the relatively stable WC and LOI values among treatments. This suggests that nutrient removal was predominantly controlled by changes in sediment geochemistry rather than alterations in sediment physical characteristics.

4. Conclusion

This study investigated the effect of steel slag particle size on the remediation of contaminated coastal sediment without pretreatment. The results demonstrate that steel slag dissolution significantly alters sediment geochemical conditions by increasing pH and modifying redox potential through metal hydrolysis and subsequent reactions. Smaller steel slag particles enhanced alkalinity due to greater dissolution of CaO, resulting in elevated sediment pH. DIN removal was most effective in the 1.0-5.0 mm case (0.9 mg L⁻¹) under the conditions tested, likely due to a balance between adsorption capacity and microbial activity. In contrast, phosphate removal was most effective in the < 0.6 mm case (0.62 mg L⁻¹), driven by increased surface area, enhanced metal ion release, and favorable precipitation conditions. This distinction suggests that nitrogen removal is primarily governed by adsorption and biological processes, whereas phosphate removal is dominated by precipitation and surface reactions. COD was also lowest in the < 0.6 mm (142 mg L⁻¹), indicating enhanced organic matter removal through microbial degradation and metal-induced flocculation. In contrast, steel slag addition had minimal impact on sediment physical properties, with no significant changes observed in WC and LOI. Overall, finer steel slag particles (< 0.6 mm) and intermediate particle sizes (1.0-5.0 mm) demonstrate distinct advantages for nutrient removal. These findings highlight the importance of particle size optimization and support the application of untreated steel slag as a cost-effective and sustainable material for sediment remediation.