Phosphate Recovery from Wastewater Using Red Mud-Modified Biochar Beads: Performance and Mechanism Study

Abstract

In this study, red mud (RM) was utilized as an iron and aluminum source, and reed biomass served as a carbon precursor to prepare red mud-modified biochar beads (RM/CSBC) via the gel-calcination method. Under a pyrolysis temperature of 900 °C and an RM/biomass dosage of 3 g each, RM/CSBC exhibited an optimal balance between adsorption performance and cost. Within typical pH range of 6–9 in wastewater, RM/CSBC maintained effective adsorption performance, while metal ion leaching (Fe ≤ 0.3 mg·L⁻¹, Al ≤ 0.2 mg·L⁻¹) complied with Class II surface water standards in China. Kinetic data were well fitted by the pseudo second-order model, supported by the Elovich model, indicating the involvement of both chemical and physical adsorption mechanisms. Isotherm results showed that the Langmuir model provided the best fit, indicating monolayer adsorption, with a maximum capacity of 85.16 mg·g⁻¹ at 25 °C. XPS analysis revealed the formation of AlPO₄ and FePO₄ precipitates, confirming chemical precipitation as a key mechanism, along with electrostatic attraction and physical sorption. This study highlights the feasibility of RM/CSBC as an efficient and low-cost phosphate adsorbent and provides a theoretical basis for phosphorus removal and recovery from wastewater using waste-derived materials.

1. Introduction

Phosphorus is an essential element for life and plays a critical role in biogeochemical cycles and human activities. As a non-renewable resource, phosphorus is primarily stored in the Earth’s crust in the form of phosphate rock [1]. Approximately 20 million tons of phosphate rock are mined globally each year, and at the current rate of consumption, the Earth’s phosphorus reserves are expected to be depleted within the next century [2,3]. The widespread use of phosphate fertilizers has led to excessive phosphorus entering water environments through rainfall runoff, accelerating eutrophication in water bodies [4]. Currently, the contradiction between the phosphorus crisis and excess phosphorus is a significant constraint on the sustainable development of society. Therefore, it is crucial to develop sustainable materials and efficient technologies for the removal and recovery of phosphate from wastewater.

Common phosphorus recovery technologies include enhanced biological phosphorus removal (EBPR), chemical precipitation, reverse osmosis, and adsorption methods [5,6,7,8]. While EBPR is highly sensitive to operational conditions and temperature, chemical precipitation produces large volumes of sludge, and reverse osmosis is energy-intensive and costly [9]. In contrast, adsorption stands out for its simplicity, low cost, high efficiency, and scalability in practical applications. Therefore, adsorption is widely recognized as a promising strategy [10,11]. The selection of an ideal adsorbent is critical for successful application of adsorption method. Materials such as zeolite, calcite, activated carbon (AC), and biochar (BC) have attracted considerable attention due to their high specific surface area [12]. Among them, BC, a carbon-rich product derived from the pyrolysis of biomass under anaerobic conditions, possesses abundant oxygen-containing functional groups, high porosity, a large specific surface area, and a stable carbon backbone structure, making it suitable for the removal of water pollutants [13]. Compared with AC, BC possesses more abundant surface functional groups, which facilitate complexation or loading with metal ions for further functionalization [14]. In addition to these structural advantages, BC can be produced from low-cost and renewable biomass residues at significantly lower production costs than AC [15,16]. From an environmental perspective, BC production generally requires less cumulative energy and generates fewer greenhouse gas emissions [17]. In terms of performance, although AC typically has higher surface area and faster adsorption kinetics, appropriately modified BC can achieve comparable or even superior contaminant removal while being more sustainable for large-scale wastewater treatment [18]. A concise comparison of BC and AC in terms of characteristics, cost, environmental impact, and performance is presented in

Table 1. Comparison between BC and AC in key aspects relevant to wastewater treatment.

Aspect	BC	AC	References
Characteristics	Derived from renewable biomass; meso–macroporous structure; abundant oxygen-containing functional groups; stable carbon backbone	Produced from biomass or fossil precursors; predominantly microporous; very high specific surface area	[13,14]
Cost	~USD 50–200 ton−1; some cases ~USD 1.06 kg−1	~USD 1000–3000 ton−1; some cases ~USD 1.34 kg−1	[15,16]
Environmental impact	CED: 20.3 MJ kg−1; GHG emissions: 1.53 kg CO2-eq kg−1; Meta-analysis: CED 6.1 MJ kg−1, GHG −0.9 kg CO2-eq kg−1	CED: 119.5 MJ kg−1; GHG emissions: 8.96 kg CO2-eq kg−1; Meta-analysis: CED 97 MJ kg−1, GHG 6.6 kg CO2-eq kg−1	[17]
Performance	Comparable or superior removal for certain pollutants after modification; e.g., granular BC removed COD-T at 0.41 kg m−3 d−1 vs. 0.24 kg m−3 d−1 for GAC in high-COD wastewater	Generally faster kinetics and higher capacity for a wide range of pollutants	[18]

.

However, biochar is typically electronegative, limiting its affinity for phosphate anions [7]. The use of metal elements such as Fe, Al, Ca, and Mg to modify biochar can enhance its stability and improve its phosphate removal capacity [19,20]. For example, Zheng et al. [21] prepared Al-EBC and Fe-EBC by modifying bamboo or hickory chips with Al and Fe salts, respectively, achieving phosphate removal rates of 68% and 94%, primarily through electrostatic interactions. However, the high cost remains a challenge. Red mud (RM), a byproduct generated during alumina production via the Bayer process, is primarily composed of Fe₂O₃, Al₂O₃, CaO, and SiO₂ [22]. The large-scale production and disposal of RM have severely impacted the surrounding ecological environment, making its valorization essential [23]. Compared to synthetic Fe/Al salts or other industrial byproducts such as steel slag and sewage sludge, RM offers significant economic and environmental advantages [24,25]. It is extremely low in cost, rich in iron and aluminum oxides with relatively high purity, and contains low levels of toxic and hazardous elements, thereby minimizing the risk of secondary pollution [26]. Due to the availability of metal oxides in RM, co-pyrolysis of RM with biomass to produce magnetic adsorbents offers a way to increase its value. Kang et al. [27] found that gases produced during biomass pyrolysis can reduce Fe₂O₃ in RM, increasing its paramagnetism, making the material easier to separate magnetically. Additionally, the porous structure of biochar helps better disperse the metal particles in RM. Yang et al. [28] prepared functional composite biochar (RM-BC) from RM and walnut shells, and found that RM-BC prepared at 320 °C with a 1:1 mass ratio had a high phosphate adsorption capacity of 15.48 mg·g⁻¹, more than twice that of the original biochar.

It is worth noting that biochar in its raw form is a powdered solid, which makes solid–liquid separation difficult after phosphate adsorption, potentially causing secondary pollution. In continuous flow systems, it is prone to blocking fixed beds, hindering its regeneration and recycling. However, most existing studies have focused on powdered red mud–biochar composites, which pose challenges for solid–liquid separation, regeneration, and phosphate reuse [29,30,31]. Chitosan, a hydrophilic biopolymer with hydroxyl and amino groups, has crosslinking properties and can encapsulate powdered materials in the form of beads, improving their size and settling properties [32]. Biochar can act as a foaming agent in these beads, enhancing their porosity, and also serve as a carrier to improve the dispersion and stability of metals [33]. Although the phosphate adsorption performance of red mud–biochar composites has been reported, studies focusing on their structured bead formation for efficient phosphate recovery and controlled nutrient release remain limited [29,30,31]. Therefore, this study aims to prepare red mud-modified biochar beads (RM/CSBC) with high phosphate adsorption capacity through optimization of preparation processes and microsphere design. The study will evaluate the effects of different amounts of red mud and biomass, pyrolysis temperature, contact time, and reaction temperature on phosphate adsorption. Additionally, the adsorption kinetics, isotherms, and thermodynamics will be explored, and the phosphate removal mechanism will be further elucidated using XRD, BET, and XPS analysis. Finally, phosphate desorption experiments will be conducted to further evaluate the agricultural application potential of the beads after phosphate recovery.

2. Materials and Methods

2.1. Materials and Reagents

The red mud was obtained from an alumina refining enterprise in Shandong Province, China. The biomass material was derived from discarded reeds collected from the Chanba Ecological Wetland in Xi’an, Shaanxi Province, China. Reed is abundant in wetland areas of Shaanxi Province. Its moderate density enables good compatibility during the co-blending process with red mud, facilitating uniform mixing and promoting effective crosslinking and bead formation. Prior to the experiment, the red mud samples were dried at 100 °C to a constant weight and then ground in a mortar. The reed was crushed using a high-speed continuous crusher. Both materials were sieved through a 200-mesh screen to control the particle size to less than 1 mm. The resulting powder samples were dried and sealed for storage. X-ray fluorescence (XRF) spectroscopy was used to determine the elemental composition of red mud, and its major constituent elements are shown in

Table 2. Chemical Composition of Red Mud.

Element	Oxide Form	Content (wt%)
Fe	Fe2O3	33.41
Al	Al2O3	22.52
Si	SiO2	16.16
Na	Na2O	8.64
Ti	TiO2	5.50
Ca	CaO	3.32
K	K2O	0.19

. The primary components include Fe₂O₃ and Al₂O₃, among other metal oxides, with a very low content of heavy metal oxides. Therefore, the environmental risk of using this red mud is minimal. All chemical reagents (such as Na₂HPO₄, NaH₂PO₄, HCl, NaOH, CH₃COOH, and NH₄Cl) were analytical grade and purchased from China National Pharmaceutical Group Corporation.

2.2. Preparation of Red Mud-Modified Biochar Beads (RM/CSBC)

Gel calcination was selected as the synthesis strategy because it enables the formation of uniform and structurally stable microspheres. The gelation process promotes effective crosslinking between chitosan, red mud, and biomass particles, enhancing the sphericity and mechanical integrity of the beads. During calcination, the gel-derived network helps improve the dispersion and stability of metal active sites while also facilitating the development of a porous structure with high surface area. These features contribute to the improved adsorption performance and reusability of the final biochar beads.

A 100 mL 2% (v/v) acetic acid solution was prepared, and 2 g of chitosan (CS) powder was added. The mixture was stirred magnetically at 60 °C for 1 h until the chitosan dissolved completely, forming a chitosan solution. Then, 3 g each of red mud and reed biomass (mixed in a 1:1 ratio) were added to the solution and stirred continuously for 0.5 h to achieve homogeneous dispersion. The mixture was dripped into 125 mL of 2 mol·L⁻¹ NaOH solution using a micropipette to form spherical gel beads. After curing at room temperature for 12 h, the beads were washed three times with deionized water until the wash solution became neutral, removing residual ions. The washed beads were then dried in a forced air drying oven. The dried beads were placed in a tube furnace, and nitrogen gas was passed through for 30 min to create an inert atmosphere. The temperature was then increased at a rate of 8 °C·min⁻¹ to 900 °C, where the pyrolysis reaction was maintained for 2 h. After cooling to room temperature, the biochar beads were collected and named RM/CSBC.

In the optimization of bead preparation conditions, the main factors affecting the adsorption properties of RM/CSBC were varied while keeping other experimental parameters constant. These factors included pyrolysis temperature (400 °C, 600 °C, 800 °C, 900 °C, 1000 °C, 1100 °C) and the mass ratio of raw materials (m(RM):m(BM) = 3:0, 0:3, 2:1, 2:2, 2:3, 2:4, 3:1, 3:2, 3:3, 3:4). The optimal adsorbent was selected for subsequent batch adsorption experiments.

2.3. Phosphate Adsorption Experiments

The pH of actual industrial wastewater typically ranges from 6 to 9, while phosphate-containing industrial effluents generally exhibit pH values between 6 and 7 [34]. In this study, a phosphate solution with an initial pH of 6.5 was prepared to simulate PO₄³⁻ contamination under realistic wastewater conditions. A stock solution of phosphate (300 mg·L⁻¹) was prepared by dissolving 0.7123 g NaH₂PO₄ and 0.5358 g Na₂HPO₄ in deionized water, and adjusting the pH to 6.5 using 1 mol·L⁻¹ HCl. In batch adsorption experiments, based on the previous literature and preliminary results, the RM/CSBC dose was set to 1.0 ± 0.01 g·L⁻¹. Specifically, 0.08 g of prepared RM/CSBC was added to 80 mL of 300 mg·L⁻¹ phosphate solution. The mixture was then continuously agitated in a thermostatic water bath shaker at 25 °C and 180 rpm for 21 h. After the reaction, the supernatant was collected, and the phosphate concentration in the solution was determined using the ammonium molybdate spectrophotometric method [35]. The adsorption capacity was calculated and the adsorption performance of different materials was compared.

The equilibrium adsorption capacity and the adsorption capacity at time t can be calculated using Equations (1) and (2) [12]:

$${\mathrm{q}}_{\mathrm{e}}={\displaystyle \frac{\left({\mathrm{c}}_0-{\mathrm{c}}_{\mathrm{e}}\right)\mathrm{V}}{\mathrm{m}}}$$

(1)

$${\mathrm{q}}_{\mathrm{t}}={\displaystyle \frac{\left({\mathrm{c}}_0-{\mathrm{c}}_{\mathrm{t}}\right)\mathrm{V}}{\mathrm{m}}}$$

(2)

where q_e (mg·g⁻¹) is the equilibrium adsorption capacity, q_t (mg·g⁻¹) is the adsorption capacity at time t, C₀ (mg·L⁻¹) is the initial PO₄³⁻ concentration, C_e (mg·L⁻¹) is the equilibrium concentration of PO₄³⁻, C_t (mg·L⁻¹) is the concentration of PO₄³⁻ at time t, V (L) is the volume of the solution, and m (g) is the mass of the adsorbent.

2.3.1. Adsorption Kinetics

To minimize the impact of sampling on the experimental system, 0.16 g of RM/CSBC was added to 160 mL of phosphate solution, and the mixture was continuously shaken in a thermostatic water bath shaker set at 25 °C with a shaking speed of 180 rpm. At the start of contact times of 0.25, 0.75, 1.5, 2, 3, 5, 8, 14, 18, 24, 36, and 48 h, 0.1 mL of the phosphate solution was periodically withdrawn using a syringe with a 0.45 μm filter membrane. The phosphate concentration was measured using the ammonium molybdate spectrophotometric method to obtain a series of data points and analyze the adsorption performance.

The experiments were all conducted in triplicate. Based on the adsorption results, the pseudo first-order kinetic model, pseudo second-order kinetic model, and Elovich model were used for fitting, with the calculations shown in Equations (3)–(5), respectively [36]. The adsorption results at different reaction times were fitted to preliminarily reveal the adsorption mechanism.

$${\mathrm{q}}_{\mathrm{t}}={\mathrm{q}}_{\mathrm{e}}\left(1-{\mathrm{e}}^{-{\mathrm{k}}_1\mathrm{t}}\right)$$

(3)

$${\mathrm{q}}_{\mathrm{t}}={\displaystyle \frac{{\mathrm{k}}_2{\mathrm{q}}_{\mathrm{e}}^2\mathrm{t}}{1+{\mathrm{k}}_2{\mathrm{q}}_{\mathrm{e}}\mathrm{t}}}$$

(4)

$${\mathrm{q}}_{\mathrm{t}}={\displaystyle \frac{1}{\mathsf{\beta }}}\ln \left(\mathsf{\alpha }\mathsf{\beta }\right)+{\displaystyle \frac{1}{\mathsf{\beta }}}\ln \left(\mathrm{t}\right)$$

(5)

where q_t (mg·g⁻¹) is the adsorption capacity at time t; q_e (mg·g⁻¹) is the equilibrium adsorption capacity; k₁ (min⁻¹) is the pseudo first-order adsorption rate constant; t (h) is the reaction time; k₂ (g·mg⁻¹·min⁻¹) is the pseudo second-order adsorption rate constant; β (g·mg⁻¹) is the desorption constant; and α (mg·g⁻¹·min⁻¹) is the initial adsorption rate.

2.3.2. Adsorption Isotherm

A total of 0.08 g of RM/CSBC was added to 80 mL of phosphate solution, with initial phosphate concentrations set at 37.5, 75, 150, 225, 300, 375, and 450 mg·L⁻¹. Four temperature gradients of 15 °C, 25 °C, 35 °C, and 45 °C were applied, and the mixture was continuously shaken for 24 h in a thermostatic water bath shaker. After the reaction, the supernatant was collected to determine the concentration of PO₄³⁻ and analyze the adsorption results.

Langmuir and Freundlich adsorption isotherm models were used for fitting, with calculations shown in Equations (6) and (7), respectively [37]. The maximum adsorption capacity was determined through model fitting.

$${\mathrm{q}}_{\mathrm{e}}={\displaystyle \frac{{\mathrm{q}}_{\mathrm{m}}{\mathrm{K}}_{\mathrm{L}}{\mathrm{c}}_{\mathrm{e}}}{1+{\mathrm{K}}_{\mathrm{L}}{\mathrm{c}}_{\mathrm{e}}}}$$

(6)

$${\mathrm{q}}_{\mathrm{e}}={\mathrm{K}}_{\mathrm{F}}{\mathrm{c}}_{\mathrm{e}}^{{\displaystyle \frac{1}{\mathrm{n}}}}$$

(7)

where q_m (mg·g⁻¹) is the theoretical maximum adsorption capacity of the adsorbent as fitted by the Langmuir isotherm model; K_L (L·mg⁻¹) is the Langmuir isotherm constant; K_F (mg^(1−1/n) L^1/n·g⁻¹) is the Freundlich isotherm constant; and n is the Freundlich linear constant.

2.3.3. Adsorption Thermodynamics

The isotherm models of RM/CSBC at the above four temperatures of 15 °C, 25 °C, 35 °C, and 45 °C were fitted, and the obtained values were substituted into Equations (8)–(10) to calculate the thermodynamics of phosphate adsorption by RM/CSBC [38]. The Gibbs free energy change (∆G/J·mol⁻¹·K⁻¹), enthalpy change (∆H/J·mol⁻¹), and entropy change (∆S/J·mol⁻¹) during the adsorption process were determined, providing insight into whether the adsorption of phosphate is endothermic or exothermic, spontaneous or non-spontaneous, and offering evidence for the adsorption mechanism.

$$\ln {\mathrm{K}}_{\mathrm{C}}={\displaystyle \frac{-\mathsf{\Delta }\mathrm{H}}{\mathrm{R}\mathrm{T}}}+{\displaystyle \frac{\mathsf{\Delta }\mathrm{S}}{\mathrm{R}}}$$

(8)

$$\mathsf{\Delta }\mathrm{G}=\mathsf{\Delta }\mathrm{H}-\mathrm{T}\mathsf{\Delta }\mathrm{S}$$

(9)

$${\mathrm{K}}_{\mathrm{C}}=55.5\times \mathrm{M}\times {\mathrm{K}}_{\mathrm{L}}\times 1000$$

(10)

where R (8.314 × 10⁻³ kJ·mol⁻¹·K⁻¹) is the gas constant; T (K) is the temperature; K_C is the dimensionless standard equilibrium constant, which can be derived from the Langmuir constant (K_L, L·mg⁻¹) or the Freundlich constant (K_L, (mg·g⁻¹)/(mg·L⁻¹)^1/n); M (g·mol⁻¹) is the molar mass of the adsorbate molecule; 55.5 (mol·L⁻¹) is the standard molar concentration of water.

2.3.4. Effect of Solution pH

The effect of pH on adsorption was studied by adjusting the phosphate solution pH to a range of 2–12 using HCl or NaOH, with 11 equal intervals. The adsorption capacity of RM/CSBC at different pH values was compared. After the reaction, the supernatant was analyzed for PO₄³⁻ concentration, and the total iron (TFe) and total aluminum (TAl) concentrations in the leachate were measured by ICP-OES.

2.3.5. Effect of Co-Existing Ions

To investigate the effects of co-existing ions on phosphate adsorption by RM/CSBC, typical cations (Na⁺, K⁺, Ca²⁺, Mg²⁺, Zn²⁺) and anions (Cl⁻, SO₄²⁻, NO₃⁻, CO₃²⁻, HCO₃⁻) commonly found in wastewater were selected. Each ion was added individually to a 300 mg·L⁻¹ phosphate solution at a concentration of 2 mmol·L⁻¹ using their respective salts: NaCl, KCl, CaCl₂, MgCl₂, ZnCl₂, Na₂SO₄, NaNO₃, Na₂CO₃, and NaHCO₃. The solution pH was adjusted to 6.5, and RM/CSBC dosage was maintained at 1.0 ± 0.01 g·L⁻¹. The mixtures were shaken at 25 °C and 180 rpm for 21 h. The phosphate concentrations in the supernatants were measured, and the adsorption capacities in the presence of various ions were compared to assess the ion interference on adsorption performance.

2.4. Characterization

To further investigate the mechanisms underlying phosphate adsorption by RM/CSBC, X-ray diffraction (XRD) analyses were carried out to determine the crystalline phases of RM/CSBC before and after phosphate adsorption. XRD patterns were obtained using a Bruker D8 Advance diffractometer (Bruker AXS, Karlsruhe, Germany) with Cu Kα radiation (λ = 1.5406 Å), operating at 40 kV and 40 mA. The scans were performed over a 2θ range of 10–80° at a step size of 0.02° and a scan speed of 2° min⁻¹. The obtained diffraction peaks were identified by comparison with standard reference patterns from the JCPDS database. To assess changes in surface area and porosity upon phosphate adsorption, BET surface area analysis was performed using nitrogen adsorption–desorption isotherms at 77 K. Samples were pre-treated by vacuum drying at 200 °C for 4 h. The specific surface area was calculated using the BET model, and pore size distribution in the mesoporous range was determined using the BJH algorithm. In addition, X-ray photoelectron spectroscopy (XPS) analyses were performed on RM/CSBC before and after phosphate adsorption (denoted as RM/CSBC-P). XPS measurements were conducted using a Thermo Fisher ESCALAB Xi+ spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) equipped with a dual anode Al/Mg X-ray source operating at 400 W. The binding energy scale was calibrated using the C 1s peak at 284.8 eV. Samples were thoroughly ground and pressed into pellets before analysis. High-resolution spectra of C 1s, O 1s, Al 2p, Fe 2p, and P 2p were acquired, and peak fitting was performed using Avantage software (version 5.9) to determine the chemical states and surface compositions of the elements.

2.5. Phosphate Release Experiments

To evaluate the nutrient phosphate release performance of RM/CSBC-P, desorption experiments were conducted. Deionized water was adjusted to pH 4.0, 7.0, and 9.0 using HCl or NaOH. Then, 0.16 g of phosphate-saturated and dried RM/CSBC-P was added to 80 mL of each pH-adjusted solution, and the mixtures were shaken at 25 °C and 120 rpm. At predetermined intervals, approximately 1 mL of the solution was withdrawn using a syringe, filtered through a syringe filter, and transferred into a 2 mL centrifuge tube for phosphate concentration analysis. This setup simulated phosphate release from the material in the absence of plant uptake, providing an estimate of the equilibrium release time.

Subsequently, to simulate phosphate release under continuous uptake conditions (e.g., by plants in soil), fresh deionized water was again adjusted to pH 4.0, 7.0, and 9.0 using HCl or NaOH. Then, 0.16 g of phosphate-saturated and dried RM/CSBC-P was added to 80 mL of each pH-adjusted solution and shaken at 25 °C and 120 rpm. Based on the previously determined equilibrium release time, a sampling cycle was established. After each cycle, phosphate concentration was measured, and the RM/CSBC-P beads were removed and transferred into a newly prepared 80 mL solution of the same pH. This procedure was repeated until the cumulative phosphate release approached equilibrium.

3. Results and Discussion

3.1. Preparation Conditions of RM/CSBC

3.1.1. Effect of Mass Ratio of Red Mud and Biochar

The effects of different red mud and biomass dosages on phosphate adsorption of RM/CSBC are presented in Figure 1. Figure 1a shows the influence of mass ratio of red mud and biomass under pyrolysis conditions at 400 °C. With red mud dosage held constant, increasing biomass dosage significantly enhanced phosphate adsorption capacity of RM/CSBC. This is attributed to the formation of a carbon skeleton with a rich porous structure during the pyrolysis of biomass, which provides more surface area and active sites for adsorption, thereby improving phosphate adsorption [19]. Similarly, increasing red mud dosage while keeping biomass dosage constant also improved phosphate adsorption capacity. The metal ions in red mud, including Fe³⁺ and Al³⁺, provide favorable sites for chemical precipitation reactions, forming stable metal phosphate precipitates that further promote phosphate removal [29]. Therefore, when the dosages of red mud and biomass were 3 g and 4 g, respectively, the phosphate adsorption capacity of RM/CSBC reached its maximum value of 26.29 mg·g⁻¹.

Figure 1b shows the phosphate adsorption capacity of RM/CSBC at a pyrolysis temperature of 900 °C. Without red mud, the phosphate adsorption capacity was only 6.77 mg·g⁻¹, due to the absence of metal ions from red mud that could contribute to chemical precipitation, with adsorption limited to physical adsorption only [31]. Without biomass, the adsorption capacity remained low, at only 6.28 mg·g⁻¹. However, when red mud dosage was 3 g, as biomass dosage increased from 3 g to 4 g, the adsorption capacity of beads significantly increased, reaching approximately 45.00 mg·g⁻¹, indicating the important role of biomass. In the absence of biomass, the beads lacked a loose, porous structure, which hindered the proper dispersion of metal ions and prevented the full exposure of active sites needed for efficient phosphate adsorption [39]. Notably, the phosphate adsorption capacity of RM/CSBC was lower when the dosages of red mud and biomass were 3 g and 1 g, respectively, compared to when the dosages were 2 g and 3 g. This suggests that the reed carbon skeleton of biomass effectively exposes active adsorption sites of beads, enhancing phosphate adsorption [9]. Therefore, considering both adsorption performance and preparation costs, a red mud dosage of 3 g and biomass dosage of 3 g were selected as the optimal conditions for subsequent batch adsorption experiments.

3.1.2. Effect of Pyrolysis Temperature

Pyrolysis temperature is critically associated with phosphate adsorption capacity of biochar, as it influences both pore structure characteristics and quantity of surface functional groups. As shown in Figure 2, as pyrolysis temperature increased from 600 °C to 900 °C, phosphate adsorption capacity of RM/CSBC continuously improved across various red mud and biomass mass ratios. This suggests that elevated pyrolysis temperatures facilitate phosphate adsorption process, likely due to high-temperature promoting development of biochar’s microstructure, increasing specific surface area and exposing more active adsorption sites [40]. Furthermore, higher pyrolysis temperatures enhance concentration of metal elements such as iron, aluminum, and calcium in biochar, strengthening electrostatic attraction and surface chemical precipitation, which further enhances phosphate adsorption [41]. However, when pyrolysis temperature increased from 1000 °C to 1100 °C, phosphate adsorption capacity of RM/CSBC prepared with 3 g red mud to 3 g biomass ratio decreased from 42.88 mg·g⁻¹ to 38.97 mg·g⁻¹, and for 3 g red mud to 4 g biomass ratio, adsorption capacity decreased more significantly, from 45.80 mg·g⁻¹ to 37.02 mg·g⁻¹. This can be attributed to excessively high pyrolysis temperature causing degradation of biomass skeleton and collapse of pore structure, reducing number of adsorption sites and decreasing material’s surface reactivity [42]. Additionally, this increase in temperature leads to thermal deactivation of key functional groups, such as hydroxyl and carboxyl groups, further diminishing its adsorption capability. Consequently, optimal pyrolysis temperature positively impacts phosphate adsorption capacity of RM/CSBC, and RM/CSBC prepared at 900 °C is selected for subsequent batch adsorption experiments.

3.2. Phosphate Adsorption Performance

3.2.1. Adsorption Kinetics Analysis

Figure 3 shows the phosphate adsorption capacity of RM/CSBC at different contact times for kinetics analysis. Initially, RM/CSBC rapidly adsorbed phosphate, mainly due to abundant metal active sites on the bead surface, generating electrostatic attraction with phosphate ions [43]. In the first 2 h, phosphate adsorption increased quickly, reaching 11.65 mg·g⁻¹, then the rate slowed, and the capacity stabilized after 21 h, reaching about 41 mg·g⁻¹. The adsorption process is divided into three stages: the rapid adsorption phase, dominated by physical adsorption, where phosphate ions bind to active sites on the biochar surface via diffusion, driven by the concentration gradient; the slow adsorption phase, where physical adsorption weakens and ion equilibrium is reached; and the chemical adsorption phase, where the mechanism shifts to chemical adsorption, with phosphate ions diffusing and binding within the particles, mainly influenced by chemical bonding forces, leading to stable binding with the biochar surface [44].

The phosphate adsorption kinetics of RM/CSBC were fitted using the pseudo first-order kinetic model, pseudo second-order kinetic model, and Elovich model, with the results shown in

Table 3. Parameters of kinetic models for phosphate adsorption on RM/CSBC.

Kinetic Model	Parameter	RM/CSBC
PFO model	qe/mg·g−1	41.35
	k1/min−1	0.1310
	R2	0.9580
PSO model	qe/mg·g−1	49.16
	k2/g·mg−1·min−1	0.0030
	R2	0.9666
Elovich model	α/mg·g−1·min−1	12.0390
	β/g·mg−1	0.0875
	R2	0.9661

. Among them, the pseudo second-order and Elovich models showed relatively higher agreement with the experimental data, with R² values of 0.9666 and 0.9661, respectively. Although the fitting was not perfect, the deviation may be attributed to the non-uniform distribution of pore structures and surface functional groups within the material, which introduces a certain degree of non-ideal adsorption behavior. These results suggest that both physical and chemical adsorption processes are involved in the phosphate removal by RM/CSBC [45].

3.2.2. Adsorption Isotherm Analysis

Figure 4 shows the phosphate adsorption isotherm analysis of RM/CSBC at different temperatures. As the temperature increased, the phosphate adsorption capacity of RM/CSBC gradually increased, indicating that higher temperatures promote phosphate adsorption by RM/CSBC. At lower initial phosphate concentrations, there are many unoccupied active adsorption sites on the surface of RM/CSBC, creating a strong affinity with phosphate ions, causing the phosphate adsorption to increase rapidly with higher initial phosphate concentrations. However, as the initial phosphate concentration reaches a certain level, the active sites are increasingly occupied by phosphate molecules, and the probability of unadsorbed phosphate ions interacting with the active sites decreases, leading to a gradual decrease in the adsorption rate.

The fitting results of adsorption isotherm models are shown in

Table 4. Parameters of isotherm models for phosphate adsorption on RM/CSBC at different temperatures.

Isotherm Model	Parameter	15 °C	25 °C	35 °C	45 °C
Langmuir model	qm/mg·g−1	72.59	85.16	86.84	93.54
	KL/L·mg−1	0.0048	0.0053	0.0058	0.0064
	R2	0.9976	0.9947	0.9983	0.9942
Freundlich model	KF	1.7158	1.5269	2.7139	3.2356
	1/n	0.5614	0.5951	0.5271	0.5157
	R2	0.9890	0.9944	0.9931	0.9907

. It can be seen that Langmuir adsorption isotherm model provided a better fit with higher R² values, which were 0.9976, 0.9947, 0.9983, and 0.9942 at 15 °C, 25 °C, 35 °C, and 45 °C, respectively, indicating that phosphate adsorption on RM/CSBC is mainly monolayer adsorption [41]. As the environmental temperature increased from 15 °C to 45 °C, the maximum adsorption capacity of RM/CSBC, obtained by fitting the Langmuir model, increased from 72.59 mg·g⁻¹ to 93.54 mg·g⁻¹. The higher temperature promotes the exposure of active sites and chemical reactions during the adsorption process, which is favorable for phosphate adsorption. The 1/n value derived from the Freundlich model can be used as an indicator of surface heterogeneity or exchange intensity [46]. Since all 1/n values are less than 1, it indicates that phosphate adsorption on RM/CSBC is favorable.

3.2.3. Adsorption Thermodynamics Analysis

The thermodynamic fitting curve of phosphate adsorption on RM/CSBC and the thermodynamic model fitting results are shown in

Table 5. Thermodynamic parameters for phosphate adsorption on RM/CSBC at different temperatures.

T/K	ΔG/kJ·mol−1	ΔH/kJ·mol−1	ΔS/J·mol−1·K−1	R2
288.15	−24.28	7.26	109.47	0.9989
298.15	−25.38
308.15	−26.47
318.15	−27.57

and Figure 5, respectively. From the data in the table, it can be observed that as the temperature increased from 288.15 K to 318.15 K, the Gibbs free energy change (ΔG) decreased from −24.2839 kJ·mol⁻¹ to −27.5681 kJ·mol⁻¹, indicating that the adsorption process is spontaneous across the entire temperature range. Furthermore, as the temperature increased, the spontaneity enhanced, suggesting that higher temperatures favor the adsorption and enrichment of phosphate on the RM/CSBC surface. The enthalpy change (ΔH) was positive, indicating that the adsorption process is endothermic, meaning that higher temperatures facilitate the adsorption process. The entropy change (ΔS) was also positive, suggesting that during the adsorption process, the degree of disorder at the solid–liquid interface increased, and the system’s disorder tendency grew [47]. This may be due to phosphate ions detaching from their hydration shell in the solution and attaching to the adsorbent surface, thus increasing the molecular freedom and disorder at the solid–liquid interface. This further supports the spontaneity and endothermic nature of the adsorption process.

3.2.4. Effect of Solution pH

The solution pH not only affects the dissociation form of phosphate but also regulates the surface charge of RM/CSBC and the release behavior of metal elements in the material, thereby influencing its adsorption performance. As shown in Figure 6a, RM/CSBC exhibited the highest phosphate adsorption capacity at pH 2, reaching 64.83 mg·g⁻¹. This is primarily attributed to the free Fe³⁺ and Al³⁺ ions in the solution under strong acidic conditions, which directly form insoluble aluminum and iron phosphates with phosphate ions [41]. Additionally, protonation of surface hydroxyl functional groups forms positively charged ≡Fe/Al-OH₂⁺ sites, resulting in stronger electrostatic attraction with H₂PO₄⁻ ions [48]. Furthermore, as observed from the phosphate distribution (Figure 6b), at this pH, phosphate primarily exists as H₂PO₄⁻, which, due to its lower charge and smaller steric hindrance, more readily binds to the surface cationic sites through electrostatic interactions [49].

To further analyze the leaching of iron and aluminum from red mud, the concentrations of TFe and TAl in the solution at different initial pH values were measured. As shown in Figure 7, the release of metal ions from the RM/CSBC surface was minimal across the pH range of 2–12, with a slight increase observed at pH < 4. When the initial pH ranged from 5 to 9, the concentrations of Fe and Al leached were below 0.15 mg·L⁻¹, which is below the Class II limit values of Surface Water Environmental Quality Standards (Fe ≤ 0.3 mg·L⁻¹, Al ≤ 0.2 mg·L⁻¹), indicating the material’s good stability. When the pH exceeded 9, the release of metal ions slightly increased, with some existing as Fe(OH)₄⁻ and Al(OH)₄⁻ complexes, which have weaker phosphate precipitation effects. Additionally, the positive charge on the biochar surface decreased as the pH increased, leading to a weakened electrostatic attraction, which accounts for the reduced phosphate adsorption capacity of RM/CSBC at higher pH values [50]. On the other hand, under alkaline conditions, OH⁻ ions adsorbed on the biochar surface can electrostatically repel phosphate ions, while phosphate gradually transitions to HPO₄²⁻ and PO₄³⁻ forms, thereby increasing the electrostatic repulsion with the material surface [51]. This pH dependent adsorption behavior indicates that RM/CSBC removes phosphate mainly through chemical precipitation and electrostatic attraction under acidic and neutral conditions.

3.2.5. Effect of Co-Existing Ions

Figure 8 shows the effects of various co-existing ions on the phosphate adsorption performance of RM/CSBC. The results indicate that Na⁺ and K⁺ had minimal influence on the adsorption capacity, whereas Ca²⁺, Mg²⁺, and Zn²⁺ significantly enhanced phosphate adsorption by RM/CSBC. In particular, the highest adsorption capacity of 64.83 mg·g⁻¹ was observed in the presence of Zn²⁺. This enhancement can be attributed to the formation of insoluble metal phosphate precipitates between the divalent metal ions and PO₄³⁻, thereby improving phosphate removal efficiency [13]. As for co-existing anions, NO₃⁻, SO₄²⁻, and Cl⁻ exerted only slight inhibitory effects on phosphate adsorption, which is consistent with the findings reported by Ren et al. [52]. In contrast, CO₃²⁻ and HCO₃⁻ exhibited significant suppression. On the one hand, their hydrolysis generates OH⁻ ions, increasing the solution pH and causing the RM/CSBC surface to acquire a negative charge, which weakens the electrostatic attraction toward phosphate anions [53]. On the other hand, CO₃²⁻ and HCO₃⁻ compete with phosphate for adsorption sites and tend to form carbonate complexes with Fe³⁺ or Al³⁺, thereby reducing the availability of metal ions for phosphate precipitation and further decreasing the adsorption capacity [36]. In summary, Ca²⁺, Mg²⁺, and Zn²⁺ facilitate phosphate removal, whereas carbonate-type anions exert adverse effects on the adsorption process. These findings should be considered in practical wastewater treatment applications.

3.3. Phosphate Recovery Mechanisms

As shown in Figure 9, both RM/CSBC and RM/CSBC-P display characteristic diffraction peaks of Al₂O₃ and Fe₃O₄, indicating that the main crystalline phases of the composite remain unchanged after phosphate adsorption [53]. However, the peak intensities of Al₂O₃ and Fe₃O₄ in RM/CSBC-P show a slight decrease, suggesting partial participation of Fe/Al species in the adsorption process. No new crystalline phases of AlPO₄ or FePO₄ are detected, implying that the phosphate species formed during adsorption are predominantly in an amorphous state or exist as a poorly crystalline coating on the RM/CSBC surface.

Figure 10a shows that the adsorption–desorption isotherm of RM/CSBC follows a typical type IV curve with an H3-type hysteresis loop, indicating that the material’s pore structure is primarily mesoporous, with irregular plate-like or slit-shaped channels. This structure contributes to a high specific surface area, providing ample space for phosphate adsorption. The hysteresis loop suggests complex capillary condensation within the pores, which enhances the retention and binding of phosphate ions on the pore walls, improving the overall adsorption capacity. Figure 10b presents the pore size distribution of RM/CSBC, revealing a distinct peak in the small pore size range, indicating a non-uniform distribution with dominant micropores and small mesopores. Micropores offer abundant adsorption sites for phosphates, especially for small ions like PO₄³⁻, while mesopores facilitate rapid diffusion and adsorption by providing low-resistance channels. The synergy between micropores and mesopores enhances both the adsorption capacity and kinetics, making RM/CSBC effective for efficient phosphate removal from wastewater.

To further investigate the mechanism of phosphate adsorption by RM/CSBC, XPS analysis was performed on RM/CSBC before and after adsorption (RM/CSBC-P), as shown in Figure 11. In the P 2p spectra (Figure 11a), two main peaks appeared in RM/CSBC-P: 133.58 eV corresponding to PO₄³⁻, and 132.81 eV associated with H₂PO₄⁻ and HPO₄²⁻, indicating that RM/CSBC successfully adsorbed phosphate. In the C 1s spectra (Figure 11b), the peaks of C=O/C-O and C-C/C=C in RM/CSBC-P shifted, suggesting that functional groups on the biochar surface interacted with the phosphate [54]. In the O 1s spectra (Figure 11c), P-O peak appeared after adsorption, and the M-O peak shifted, indicating that phosphate interacted with the metal oxides on the RM/CSBC surface through chemical bonding, forming Al-O-P and Fe-O-P [55]. In addition to electrostatic interactions, chemical precipitation is also an important mechanism for phosphate removal. In the Al 2p spectra (Figure 11d), after phosphate adsorption by RM/CSBC, the Al-O peak shifted from 73.71 eV to 74.19 eV, and the Al-OH peak shifted from 71.32 eV to 71.63 eV. This indicates the formation of aluminum-phosphate complexes, suggesting a chemical adsorption process between aluminum oxide and phosphate ions [56]. From the Fe 2p spectra (Figure 11e), it can be observed that after phosphate adsorption (RM/CSBC-P), the content of Fe²⁺ decreased, and Fe³⁺ increased, with a shift in the peaks. This suggests that during the high-temperature pyrolysis of biochar, reducing gases were released, leading to the formation of Fe²⁺. After phosphate adsorption, Fe²⁺ likely participated in the adsorption process and was oxidized to Fe³⁺, forming insoluble iron phosphate precipitates, thus aiding phosphate removal [57,58].

3.4. Phosphate Slow-Release Performance of RM/CSBC-P

Investigating the phosphate release behavior of RM/CSBC after phosphate adsorption is essential for evaluating its potential as a slow-release fertilizer in agricultural applications. Therefore, phosphate desorption experiments were conducted to systematically assess the phosphate release performance of RM/CSBC-P under different pH conditions, simulating its behavior in acidic, neutral, and alkaline soil environments. As shown in Figure 12a, under fixed solution conditions that simulate nutrient-saturated soils without plant uptake, P release from RM/CSBC-P gradually increased with time and approached equilibrium after approximately 8 h. Among the three pH levels, acidic conditions (pH 4) led to the highest and fastest release, reaching 8.26 mg·L⁻¹, likely due to enhanced dissolution of phosphate-containing compounds in low pH. In comparison, P release under neutral and alkaline conditions was slower and less extensive. Figure 12b shows the accumulated P release in periodically renewed solutions, mimicking soil systems where phosphorus is continuously absorbed by plants. In this dynamic environment, RM/CSBC-P exhibited sustained release under all pH conditions. After nine cycles, the cumulative P release reached 12.44, 10.57, and 11.49 mg·L⁻¹ at pH 4, 7, and 9, respectively. These results indicate that RM/CSBC-P can maintain long-term phosphate release performance, particularly under acidic and alkaline conditions, making it suitable for application in various soil environments.

4. Conclusions

In this study, red mud and reed biomass were used to synthesize red mud-modified biochar beads (RM/CSBC) with three-dimensional porous structure via gel calcination method for phosphate recovery. When pyrolysis temperature was set to 900 °C and red mud and biomass dosages were both 3 g, the material achieved the optimal balance between adsorption capacity and economic cost. Phosphate adsorption was better fitted by the pseudo second-order kinetic model, indicating that chemical adsorption was the primary mechanism, with a capacity of 40.93 mg·g⁻¹. Langmuir isotherm model provided a better fit, indicating that the adsorption was primarily monolayer adsorption, with a maximum capacity of 85.16 mg·g⁻¹ at 25 °C. Thermodynamic analysis indicated that the adsorption was endothermic and spontaneous. XPS and pH analysis revealed that the adsorption mechanism involved chemical precipitation, electrostatic attraction, and physical adsorption. Co-existing ion experiments showed that Ca²⁺, Mg²⁺, and Zn²⁺ facilitated phosphate adsorption, while CO₃²⁻ and HCO₃⁻ exerted inhibitory effects. Desorption experiments demonstrated that phosphate adsorbed by RM/CSBC exhibited a slow-release behavior, indicating its potential as a phosphorus fertilizer. The findings suggest that RM/CSBC is a promising material for efficient phosphate recovery from wastewater, with potential for practical application. Future work will focus on evaluating its phosphate removal performance in actual wastewater and conducting pot experiments to assess its effectiveness as a slow-release fertilizer in soil environments.