Unlocking the Detoxification of Phenanthrene from Water Using Alkali-Activated Slag Mortar

Abstract

Low-cost and high-performance materials or techniques that could synergistically remove phenanthrene (PHE) in a simple manner were highly desired. Herein, we reported an alkali-activated slag (AAS) that proved applicable in both construction and environmental protection efforts. AAS was synthesized by mixing ground granulated blast furnace slag (GGBFS) and an alkaline solution. The prepared AAS mortar achieved the highest mechanical strength when using an alkaline activator with a Na₂O concentration of 8% by slag weight. Moreover, AAS exhibited excellent sorption performance towards PHE, with the highest sorption performance reaching 44.0 mg/g, which was much higher than that of GGBFS. Sorption of PHE reached equilibrium within approximately 120 h and fit well with the pseudo-second-order model. Furthermore, the primary sorption mechanisms for PHE on AAS were attributed to cation-π interactions, hydrogen bonding, and flocculation. The strategy of using AAS not only met the requirements for high-performance and low-cost materials but also addressed the challenging issues of developing an all-in-one treatment for PHE pollutants, which was of great significance to wastewater purification.

1. Introduction

Portland cement is one of the most widely used binders in global construction, but its production has significant environmental impacts [1,2]. The manufacture of Portland cement contributes approximately 6 to 7% of total global CO₂ emissions and consumes large amounts of raw materials like limestone and clay, as well as energy [3,4]. Due to these environmental concerns, finding alternative binders has become crucial.

Recently, alkali-activated materials (AAMs) have emerged as potential substitutes, attracting global scientific interest [5,6]. These binders are synthesized by mixing aluminosilicate-rich materials (e.g., fly ash, blast furnace slag, metakaolinite, or pozzolana) with an alkaline solution [7,8]. Ground granulated blast furnace slag (GGBFS), a by-product of steel manufacturing, is a popular precursor to AAM synthesis [9]. Compared to conventional materials from Portland cement, alkali-activated slag (AAS) offers higher early mechanical strength, lower hydration heat, better chemical and thermal resistance, but faces challenges like high drying shrinkage and rapid setting [10,11]. Additionally, the main hydration product of AAS has been proven to be calcium silicate hydrate (C-S-H) with a much lower CaO/SiO₂ ratio compared to that of Portland cement [12].

Water quality has been continuously declining due to rapid population growth, urbanization, industrialization, and other environmental challenges [13]. Phenanthrene (PHE), a polycyclic aromatic hydrocarbon (PAH), is primarily released into the environment through processes like fossil fuel combustion and oil refining [14]. The US Environmental Protection Agency has classified PHE as a priority pollutant [15]. Consequently, various methods for removing PHE from water have been extensively researched and proposed.

Various technologies have been proposed for the removal of organic contaminants from water, including chemical precipitation, ion-exchange, sorption, membrane filtration, and coagulation–flocculation methods [16]. Sorption, in particular, stands out as a highly effective method owing to its high efficiency, low energy consumption, and simplicity. Sorbents commonly used for this purpose include activated carbon, clay minerals, and organic–inorganic hybrid materials. By-products and residual wastes from steel manufacturing, such as blast fuel cokes, steel slag, and fly ash, have attracted attention as potential sorbent materials [17,18]. However, raw by-products like steel slag and fly ash generally exhibit low sorption capacities for PHE, limiting their effectiveness in pollutant removal [17]. Therefore, it is essential to activate these by-products to enhance their sorption capacities. Activation processes significantly improve the physical and chemical properties of these materials, making them more suitable for use as sorbents in water treatment applications.

Despite the potential of AAS as a sorbent for PHE removal, this application has received relatively less attention in the past. Given the promising characteristics of AAS, including its higher early mechanical strength, lower hydration heat, and better chemical and thermal resistance compared to traditional Portland cement, exploring its efficacy in PHE sorption is deemed crucial. Understanding the mechanisms and efficiency of PHE sorption using alkali-activated slag provides valuable insights and contributes to the development of more effective and sustainable water treatment solutions. Therefore, further research is necessary to investigate the sorption behavior of PHE using alkali-activated slag and to optimize its application in water purification processes.

In this study, the aim is to demonstrate the applicability of AAS produced from GGBFS as a wastewater filter to reduce PHE content and achieve acceptable ion concentrations suitable for water reuse. Additionally, different ratios of Na₂O are investigated to optimize the mechanical strength of AAS. Moreover, this study elucidates the effects of the equilibrium time, PHE concentration, pH, and coexistence of hydrophobic organic carbons (HOCs) on PHE sorption.

2. Materials and Methods

2.1. Materials

Ground granulated blast furnace slag (GGBFS) with a density of 2.9 g/cm³ and Blaine fineness of 435 m²/kg was collected from Viet Nam. The chemical properties of GGBFS were analyzed by using an X-ray fluorescence (Thermo Scientific, ARL ADVANT’X, Waltham, MA, USA) instrument and are listed in

Table 1. Chemical composition of ground granulated blast furnace slag.

Component	wt%
CaO	42.4
SiO2	33.5
Al2O3	11.3
MgO	8.5
SO3	1.9
TiO2	0.7
Na2O	0.4
K2O	0.4
Fe2O3	0.4
MnO	0.2
ZrO2	0.1
SrO	0.1
Cl	0.1

. Sodium hydroxide (NaOH) solution and water glass (8.2% Na₂O, 26.4% SiO₂, and 65.4% H₂O by weight) were mixed together to make an activator with a silicate modulus of 1 (M_s = 1). River sand from a local supplier was used as a fine aggregate, with a size of less than 5 mm. Acetone (99%), n-hexane (99.7%), and hydrochloric acid (HCl, 37%) were purchased from Daejung (Daejeon, Republic of Korea). Sodium azide (NaN₃, 98%) was supplied by Junsei (Tokyo, Japan). Phenanthrene (PHE), pyrene (PYR), naphthalene (NAP), diisobutyl phthalate (DiBP), and dibutyl phthalate (DBP) were purchased from Sigma-Aldrich (St. Louis, MO, USA). A polyethylene (PE) sheet of 51 µm thick for use as a passive sampler was obtained from Husky Plastics (Teaneck, NJ, USA). The real water samples were collected from Andong Lake, the Nakdong River, industrial wastewater, Busan groundwater, and seawater in the Republic of Korea.

2.2. Sample Preparation

In mortar proportioning, the alkaline activator was chosen based on Na₂O percentages of 4%, 6%, 8%, and 10% by slag weight (abbreviated as AAS 4%, AAS 6%, AAS 8%, and AAS 10%, respectively). The ratios of fine aggregate to GGBFS and water to solid were maintained at 2.75 and 0.45, respectively. The raw materials and activator were mixed at a stirring rate of 400 rpm for 15 min to obtain the slurry. Subsequently, the mixed slurry was poured into steel triple prismatic molds (4 × 4 × 16 cm) for flexural and compressive strength testing. The AAS mortar mixture was prepared using a universal mixer. After casting into the molds, the fresh mortar was left to cure for 24 h before demolding. The mortars were then stored under ambient conditions (at a temperature of 25 ± 2 °C and a relative humidity of 90%) for the next 27 d until testing. The AAS preparation procedure is shown in Scheme 1.

2.3. Sorption Experiment

Analyzing phenanthrene (PHE) in aqueous solutions is challenging due to its low solubility. To address this, polyethylene (PE) passive samplers were utilized for assessing PHE sorption. It has been found that the equilibrium time between polyethylene and water for PHE is less than 10 h [19]. Therefore, a sorption experiment was conducted for more than 10 h to ensure equilibrium between PE and water. PE strips measuring 2.5 cm × 5.0 cm were cleaned by immersing them in hexane for 24 h, followed by acetone for 30 min, and then rinsed in deionized water for another 30 min. After cleaning, the PE strips were dried with lab tissue and stored at 4 °C until they were ready for use.

For the sorption kinetics experiment, the AAS using 8% of Na₂O was introduced to 200 mL of a PHE solution with a concentration of 0.9 mg/L, with the pH set at 5.5. To prevent microbial activity, NaN₃ was added to achieve a final concentration of 100 mg/L. The PE was then placed in the solution. At specified time intervals, the PE samplers were retrieved, wiped with lab tissue, and extracted with 20 mL of hexane for 24 h. To explore the rate-controlling step in sorption processes, the kinetic data were analyzed using the pseudo-first-order (PFO) and pseudo-second-order (PSO) models [20]. The equations are expressed as follows:

Pseudo-first-order:

$$\ln \left(Q_e-Q_t\right)={\ln Q}_e-K_{1}t.$$

(1)

Pseudo-second-order:

$${\displaystyle \frac{t}{Q_t}}={\displaystyle \frac{1}{K_2{Q}_e^2}}+{\displaystyle \frac{t}{Q_e}}$$

(2)

where Q_t (mg/g) and Q_e (mg/g) represent the amounts of PHE adsorbed at time t (h) and at equilibrium, respectively; K₁ is the PFO constant (1/min); and K₂ is the PSO rate constant (g/mg/min).

For the sorption equilibrium study, PHE concentrations ranged from 0.05 to 0.9 mg/L. The solid/liquid ratio was fixed at 0.1 g/L. For the study of the effect of pH on PHE removal efficiency, the pH was adjusted from 3 to 9 using HCl and NaOH solutions. In cases of the coexistence of hydrophobic organic carbons, a concentration of 0.05 mg/L was used for each of the following: PHE, PYR, NAP, DiBP, and DBP. After 240 h, the PE strips were retrieved and extracted with hexane. The PHE sorption isotherm data were fitted with the Langmuir and Freundlich sorption models. Each equation is expressed as follows:

Langmuir equation:

$$Q_e={\displaystyle \frac{Q_m{K}_L{C}_e}{1+K_L{C}_e}}$$

(3)

Freundlich equation:

$$Q_e=K_F{C}_e^{{\displaystyle \frac{1}{n_F}}}$$

(4)

where Q_e (mg/g) represents the amount of PHE adsorbed at equilibrium, calculated from Equation (1), C_e (mL/g) denotes the equilibrium concentration in the PE, Q_m stands for the theoretical maximum sorption capacity of PHE in mg/g, K_L is the Langmuir equilibrium constant, relating to the free energy of exchange in L/mg, and K_F ((mg/g) (L/mg)^1/n) and n_F are empirical constants of the Freundlich model, while n_S is a constant related to heterogeneous systems [21].

To evaluate the effect of real water on PHE sorption, 0.05 mg/L of PHE was spiked into uncontaminated lake water, river water, groundwater, and seawater. The solid/liquid ratio was fixed at 1.0 g/L. The concentration of PHE in the solution before and after sorption was determined based on the PHE extracted from the PE passive sampler at 240 h. The extracted samples were concentrated under a nitrogen gas flow before analysis.

The sorption capacity of PHE (Q_e, mg/g), distribution coefficient (K_d, mL/g), and removal efficiency (R %) were obtained from the experimental data and calculated using Equations (1)–(3).

$$Q_e={\displaystyle \frac{M^{PE}\left(C_o^{PE}-C_e^{PE}\right)-V^w\frac{C_e^{PE}}{K_{PE}}}{M^s}}$$

(5)

$$K_d={\displaystyle \frac{K_{PE}}{C_e^{PE}}}\times {\displaystyle \frac{M^{PE}\left(C_o^{PE}-C_e^{PE}\right)-V^w\frac{C_e^{PE}}{K_{PE}}}{M^s}}$$

(6)

$$R\left(\mathrm{\%}\right)={\displaystyle \frac{C_o^{PE}-C_e^{PE}}{C_o^{PE}}}\times 100$$

(7)

where $M^{PE}$ (g) and $M^s$ (g) represent the masses of PE and AAS, respectively. The initial PHE concentration in the PE and the PHE concentration in the PE at equilibrium are denoted by $C_o^{PE}$ (mg/g) and $C_e^{PE}$ (mg/g), respectively. $V^w$ (mL) is the solution volume, and $K_{PE}$ (mL/g) is the PE–water partition coefficient for HOCs. The K_PE values of PHE, PYR, NAP, DiBP, and DBP were 4.04, 4.89, 3.23, 4.01, and 4.13, respectively, as obtained from the references [22,23].

2.4. Sample Testing and Characterization

The flexural and compressive strength were tested at the age of 28 days according to ASTM C348 and C349 [24,25]. The debris was chosen from broken specimens and used for scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS) analysis. Hitachi S-4200 Scanning Electron Microscopes (Hitachi, Tokyo, Japan) were used for this analysis. For sorption and XRD analysis, blast furnace slag was uniformly mixed with alkaline activator with the same Na₂O concentration as mortar mixture. After 28 d of curing, the hardened paste specimen was ground into powder with a particle size less than 0.63 mm to be used for X-ray powder diffraction (XRD) analysis. XRD patterns were collected on a 10 to 70° 2θ scale by using a Siemens D-500 Diffractometer (Siemens, Munich, Germany).

The concentrations of PHE, PYR, NAP, DiBP, and DBP were measured using gas chromatography–mass spectrometry (GC-MS) on an Agilent 7890A gas chromatograph. The GC-MS data and chromatograph of determination of target compounds are shown in Figure S1 and Table S1. The MS library from National Institute of Standards and Technology (NIST) was used to identify the target components. An Agilent HP-5 column with an internal diameter of 0.25 mm, a film thickness of 0.25 µm, and a stationary phase composed of 5% phenyl and 95% dimethylpolysiloxane was used. The temperature program for the oven commenced at 50 °C (held for 3 min), followed by an increase at a rate of 10 °C/min until reaching 300 °C, which was maintained for 5 min. The total GC running time was 33 min. The interface temperature was maintained at 285 °C. Helium gas flow rates were set at 1 mL/min. The injection and ion source temperatures were maintained at 250 °C and 230 °C, respectively. The scan mass ranged from 80 to 400 amu. The injection volume was 1 μL in splitless mode. The detection limit of phenanthrene was determined to be 0.387 µg/L, following the detection limit method [26].

3. Results and Discussions

3.1. Characterization of Ground Granulated Blast Furnace Slag (GGBFS) and Alkali-Activated Slag (AAS)

3.1.1. Mechanical Strength

The data for the mechanical strengths of AAS mortars with four various Na₂O concentrations are depicted in Figure 1. It is worth mentioning that the mortar achieved the highest mechanical strength at a Na₂O concentration of 8% by slag weight. In terms of compressive strength, the strength gains were recorded to be 20.7 and 42.6% when raising the Na₂O content from 4 to 6% and from 6 to 8%, respectively. However, the strength decreased by 13.8% with the Na₂O content beyond 8%. In comparison with the compressive strength, the flexural strength exhibited negligible change when increasing the Na₂O content from 4 to 10% but also showed the highest value at a Na₂O concentration of 8%. The strength gain could be a result of the higher activation of alkaline solution with a higher concentration of Na₂O. Consequently, the AAS mortar mixture with a Na₂O content of 8% was chosen for the subsequent analysis.

3.1.2. X-ray Diffraction Analysis

Figure 1b depicts X-ray diffractions of raw GGBFS and AAS paste material with a Na₂O concentration of 8% after 28 d of curing. The XRD pattern of GGBFS exhibited a typical hump at approximately 23–35° 2θ, which was reported to represent the amorphous aluminosilicates phase [5]. Additionally, gypsum CaSO₄·2H₂O was also detected in the crystalline phase composition of GGBFS. It is noted that the phase composition was observed to show a significant transformation when GGBFS was activated by the alkaline solution. Tobermorite (Ca₅Si₆O₁₆(OH)₂), fukalite (Ca₄Si₂O₆(CO₃)(OH)₂), and calcium silicate hydrate (1.5CaO·SiO₂·xH₂O) appeared to be the dominant crystalline phases of hydrated AAS. In addition, an amorphous hump at 23–35° 2θ for aluminosiliates of GGBFS was detected and became a smaller hump at approximately 28–32° 2θ in XRD patterns, indicating the formation of C-S-H in hydrated AAS [27].

3.1.3. Scanning Electron Microscope Analysis

Figure 2 presents SEM images of GGBFS and AAS mortars with a Na₂O concentration of 8%. As shown in Figure 2a, GGBFS particles undergo a remarkable alteration due to the alkaline activation process. The granulate GGBFS particles reacted with the alkaline solution to form AAS, which appeared to have a dense structure (Figure 2b). According to previous studies, the dominant hydration product of AAS is seen to be calcium silicate hydrate xCaO·ySiO₂·zH₂O (C-S-H) with a lower CaO/SiO₂ ratio compared with that of Portland cement [28]. The EDS analysis results demonstrated the formation of C-S-H in AAS due to the dominant presence of Si element with a low content of Ca element on the surface of specimens.

3.1.4. Nitrogen Sorption–Desorption Isotherm

The N₂ sorption–desorption curves of GGBFS were found to fit type II, which resulted from the nonporous structure, whereas type I was found to be appropriate for describing the sorption–desorption plots of AAS, revealing its microporous and mesoporous characteristics (Figure 3a) [29]. After alkali activation, the specific surface area (SSA) of AAS increased from 0.53 to 61.59 m²/g (

Table 2. Specific surface area and pore volume of the synthesized sorbents.

Sample	BET Surface Area (m2/g)	Pore Volume (cm3/g)	Pore Size (Å)
GGBFS	0.53	0.0065	14.128
AAS	61.59	0.15	97.4

). The increase in SSA of AAS was attributed to the formation of calcium silicate hydrate minerals [30]. Additionally, the pore diameters of AAS exhibited a pore size distribution mainly ranging from 1 to 10 nm (Figure 3b), clearly indicating mesoporous characteristics.

3.2. Phenanthrene Sorption Performance

3.2.1. Sorption Kinetics

The effect of contact time on PHE sorption is illustrated in Figure 4a. The results demonstrated a rapid sorption of PHE by AAS, with more than 80% of PHE being removed from the solution within the first 12 h. The rapid sorption phase indicated a high affinity between PHE molecules and the active sites on the AAS surface. As the contact time progressed, the rate of PHE removal gradually decreased, eventually reaching equilibrium after approximately 60 h. This equilibrium point suggests that the sorption sites on AAS became saturated or that a dynamic balance was established between the sorbed and free PHE molecules.

In contrast, GGBFS displayed a significantly slower sorption process. It required about 120 h to reach sorption equilibrium, which is twice as long as that for AAS. Additionally, GGBFS exhibited a much lower sorption capacity (R, %) compared to AAS. The extended time required for GGBFS to achieve equilibrium indicates a lower rate of PHE uptake and potentially fewer or less accessible active sorption sites compared to AAS. The difference in the sorption performance between AAS and GGBFS can be attributed to several factors. The activation process for AAS likely enhanced the surface area, porosity, and chemical functionality of the GGBFS, making AAS more effective in capturing PHE molecules. Conversely, GGBFS may lack the optimized structural and chemical properties needed for efficient sorption.

These findings underscore the importance of material activation in enhancing sorption capabilities and highlight AAS as a superior sorbent for PHE removal in terms of both speed and capacity. The rapid and high-efficiency removal of PHE by AAS suggests a promising candidate for practical applications in water treatment, where quick and effective contaminant removal is crucial. Further investigations into the underlying mechanisms of PHE sorption in AAS could provide deeper insights into optimizing the material for even better performance.

To explore the rate-controlling step in sorption processes, the kinetic data were analyzed using the pseudo-first-order (PFO) and pseudo-second-order (PSO) models (Equations (1) and (2)). The calculated parameters of both kinetic models are presented in

Table 3. Kinetics parameters for PHE sorption on GGBFS and AAS.

Material	Qe,exp (mg/g)	Pseudo-First-Order Model			Pseudo-Second-Order Model
		K1 (1/min)	Qe,cal (mg/g)	r2	K2 (g/mg/min)	Qe,cal (mg/g)	r2
BFS	2.36 ± 0.58	0.007	1.833	0.771	0.064	2.450	0.974
AAS	15.2 ± 0.11	0.123	0.531	0.530	0.247	15.2	0.957

. The fitting of the PFO model (Figure 4b) was poor, indicating that this model did not accurately describe the PHE sorption kinetics. In contrast, the PSO model exhibited a much higher correlation with the observed PHE uptake behavior, as shown in Figure 4c. The higher r² of the PSO model suggested this chemisorption. This is consistent with the findings of previous studies, which also indicated that the PSO model is more appropriate for describing the sorption kinetics of organic pollutants like PHE [20,31,32].

3.2.2. Sorption Equilibrium of PHE Using GGBFS and AAS

To evaluate the maximum PHE sorption capacity of both BFS and AAS, detailed isotherm experiments were conducted at a controlled temperature of 23 ± 1 °C. These experiments aimed to establish the relationship between the equilibrium concentration of PHE in the solution and the amount of PHE sorbed by the materials, ultimately determining the maximum sorption capacity (Q_m) of each sorbent.

The sorption equilibriums in Figure 4d illustrate an increase in the equilibrium sorption amount with higher concentrations of PHE, indicating the progressive occupation of sorption sites on the sorbents as the availability of PHE in the solution rises. The data from these experiments demonstrated that the sorption capacity of AAS was significantly higher than that of raw GGBFS. The PHE sorption isotherm data were fitted with the Langmuir and Freundlich sorption models, described as Equations (3) and (4). The fitting results of the sorption model for the adsorbents are listed in

Table 4. Isotherm model parameters for the sorption of PHE.

Materials	Langmuir			Freundlich
	Qm (mg/g)	KL (L/mg)	r2	KF ((mg/g) · (L/mg)1/n)	nF	r2
BFS	3.79	4.691	0.990	3.436	0.435	0.922
AAS	44.00	7.823	0.959	51.61	0.456	0.983

.

The experimental data for ground granulated blast furnace slag (BFS) demonstrated a strong agreement with the Langmuir sorption isotherm model, yielding high correlation coefficients (r² > 0.995), as detailed in Table 4. This indicates that BFS exhibits monolayer sorption properties, where PHE molecules are adsorbed onto a single layer on the sorbent surface [21]. In contrast, alkali-activated slag (AAS) showed a better fit with the Freundlich sorption model, suggesting a multilayer sorption process on a heterogeneous surface [21]. The highest sorption capacity observed for PHE onto AAS from experimental data was 44.0 mg/g. Comparative analysis of sorption capacities, summarized in

Table 5. Comparisons of the PHE sorption capacities of various sorbents.

Adsorbent	Qm (mg/g)	Ref.
Alkali-activated slag	44.0	This study
Smectite	55.0	[33]
Magnesite–talc mixture	~60	[34]
Loess	0.81	[35]
Waste brick	0.03	[36]
Steel slag	41	[17]
Magnetic graphene	43.00	[37]
Alkali-activated biochar	38.82	[38]

, highlights AAS as an efficient sorbent for PHE removal from water. AAS exhibited superior performance compared to various other sorbent materials tested, confirming its efficacy in capturing PHE molecules from aqueous solutions.

3.2.3. Effect of pH

At various pH values, both GGBFS and AAS exhibited slightly fluctuating removal efficiencies and distribution coefficients for PHE, as depicted in Figure 5a. This stability suggested that PHE, being chemically stable due to its aromatic ring structure containing C=C bonds, did not undergo significant changes in hydrophobicity or transform into negatively charged species as pH levels increased [39]. The sorption of PHE onto AAS remained relatively consistent across a broad pH range, indicating that the extent of PHE sorption by AAS was similar under different environmental conditions. This observation underscores the robustness and reliability of AAS as a sorbent for PHE, offering consistent performance regardless of varying pH levels. These findings underscore the environmental suitability of AAS for water treatment, particularly in scenarios with pH fluctuations. The enhanced sorption properties of PHE make it a practical option for the removal of persistent organic pollutants from water.

3.2.4. Effect of Competitive Hydrophobic Organic Carbons (HOCs)

The removal efficiency (R, %) and distribution coefficient (K_d, mL/g) of PHE remained consistently high, exceeding 93% and 1.3 × 10⁴ mL/g, respectively, even in the presence of four other HOCs, as illustrated in Figure 5b. This indicated the remarkable selectivity of AAS towards PHE. Furthermore, the R (%) and K_d values followed the order: PHE ≈ PYR ≈ NAP > DBP > DiBP. The reduced sorption of DBP and DiBP, compared to PHE, can be attributed to their larger molecular size and longer chains, which hinder their diffusion into the pores of AAS. This hindrance limits their interaction with the sorbent surface, thereby reducing their sorption capacity [40].

The adaptability of AAS for removing PHE from complex water matrices was evaluated, as depicted in Figure 5c. The experiments included groundwater and seawater, both of which posed challenges for PHE removal by AAS, resulting in reduced removal efficiencies of 62.5% and 66.8%, respectively (see

Table 6. Main characteristics of the real water.

	Tap Water	Lake Water	River Water	Wastewater	Ground Water	Seawater
pH	7.15	6.54	7.03	7.62	6.91	8.02
TOC (mg/L)	-	13.5	18.1	10.2	15.6	3.1
Na	55.41	26.49	0.26	605	9.83	468
K	4.26	12.87	0.38	25.3	1.78	10.2
Mg	16.96	7.28	0.17	194.6	4.3	53.1
Ca	40.70	53.36	0.07	252	4.01	10.3
Cl	0.1	21.98	1.48	15.3	19.5	605.14
SO4	13.0	15.78	0.22	1.3	5.88	28.4
HCO3	-	62.24	0.11	2.6	3.86	5.39
Br	-	-	0.96	-	-	0.84
F-	0.05	-	-	1.8	-	-

). These lower efficiencies were attributed to the complex compositions and varying conditions of these environmental matrices. Despite these challenges, AAS consistently demonstrated effective PHE removal, with removal efficiencies (R%) consistently exceeding 70% across all tested water matrices. These results implied the robust performance and potential versatility of AAS as a reclaimable sorbent for remediating PHE-contaminated water. The ability of AAS to maintain relatively high PHE removal efficiencies in diverse water matrices highlights its suitability for practical applications in environmental remediation, offering a promising solution for mitigating PHE pollution in both natural and industrial settings.

The sorption of PHE using AAS was attributed to three main mechanisms. Firstly, AAS contains Ca²⁺ and Mg²⁺ ions (Table 1), which interacted with the benzene ring of PHE through cation-π interactions [41]. Secondly, the surface of AAS exhibited a negative charge in water at pH 5.5 (pH_pzc = 6.8), allowing it to form a thin layer of water film and a surface microlayer that allowed the formation of a hydrogen bond with the benzene ring of PHE [23]. Thirdly, external surface sorption and flocculation may occur due to the presence of C-S-H minerals in AAS. C-S-H was insoluble in water and initially separated out as colloidal particles, gradually condensing into a gel structure. This gel played a role in the flocculation and subsequent removal of PHE from water [42]. These mechanisms collectively contribute to the effective sorption of PHE by AAS, making it a promising material for applications in the removal of PHE from contaminated water.

4. Conclusions

Efficient and cost-effective methods to remove phenanthrene (PHE) are highly sought-after. This study highlighted an AAS as a dual-purpose material in construction and environmental protection. Synthesized from GGBFS and an alkaline solution, AAS achieved peak mechanical strength with 8% Na₂O by slag weight. The sorption performance of PHE was well described using the Langmuir sorption model. It exhibited superior PHE sorption (up to 44.0 mg/g), which was significantly higher than that of GGBFS (3.79 mg/g). The sorption kinetics of PHE reached equilibrium in 120 h, and the pseudo-second-order model was well fitted with the experimental data. The pH had a negligible effect on PHE sorption due to the stable structure of the aromatic ring. Furthermore, AAS exhibited high selectivity (K_d = 1.3 × 10⁴ mL/g) for PHE in the presence of various HOCs and water systems, suggesting its practical applicability for PHE sorption at various environment. Mechanisms included cation-π interaction, hydrogen bonding, and flocculation, addressing PHE pollutant challenges in wastewater treatment. With its high mechanical strength and sorption capacity, AAS could have practical applicability in both construction and environmental protection.