Lanthanum-modified biochar for the selective and efficient adsorption of ammonium nitrogen in wastewater

Abstract

Biochar has been widely applied in wastewater treatment and agricultural applications; however, its practical utilization is limited by poor selectivity and insufficient adsorption capacity. In this study, La₂O₃-modified biochars were prepared from rapeseed straw (RS), corn straw (BC), rice husk (RH), and cotton stalk (CS) via a sol–gel modification method followed by pyrolysis at 500 °C under oxygen-limited conditions. The adsorption performance, selectivity, and agricultural applicability of the prepared materials for NH₄⁺ recovery were systematically investigated. The modified biochars exhibited increased surface area, pore volume, and microporosity compared with the pristine biochars. RS-La and BC-La exhibited the highest NH₄⁺ adsorption capacities, reaching 43.37 and 42.23 mg g⁻¹ under optimal pH conditions, respectively. Langmuir fitting showed excellent agreement with experimental data (R² > 0.99), with maximum monolayer adsorption capacities of 46.11, 47.93, 45.42, and 43.32 mg g⁻¹ for RS-La, BC-La, RH-La, and CS-La, respectively. All La-modified biochars demonstrated strong NH₄⁺ selectivity, whereas Cl⁻ adsorption remained negligible (0.3–0.6 mg g⁻¹). This selectivity was attributed to electrostatic attraction and ion exchange between NH₄⁺ and La–OH groups, hydrogen bonding with surface functional groups, and weak Cl⁻ interaction with La sites. The adsorption kinetics were satisfactorily described by both PFO and PSO models. Preliminary pot experiments indicated the potential of NH₄⁺-loaded biochar for nitrogen recovery and agricultural reuse.

La₂O₃-modified biochar with abundant porous structures and La–O active sites was fabricated by sol–gel pyrolysis for enhanced selective NH₄⁺ adsorption.

1. Introduction

Wastewater containing ammonium chloride (NH₄Cl) is extensively generated from the excessive application of ammonium-based fertiliser,¹ the leachate from livestock manure,² pharmaceutical manufacturing³ and municipal sewage treatment.⁴ The improper discharge of NH₄Cl-rich wastewater elevates the nitrogen loading in receiving waters, which accelerates surface water eutrophication, results in groundwater contamination⁵ and triggers a cascading degradation of aquatic ecosystems⁶ while wasting a recoverable resource.⁷ The coexisting chloride ions (Cl⁻) in NH₄Cl-rich wastewater can also compromise soil ecological integrity⁸ and exacerbate soil salinisation, which represent latent threats to agricultural sustainability.⁹

Conventional wastewater treatment technologies encompass chemical precipitation,¹⁰ flotation,¹¹ oxidation–reduction,¹² membrane filtration,¹³ reverse osmosis, biological treatment/biosorption, ion exchange,¹⁴ and adsorption. These approaches differ substantially in their treatment efficiency, operational complexity, energy consumption, and potential for resource recovery. Chemical precipitation and oxidation–reduction processes are effective for specific contaminants but may require substantial chemical inputs and generate secondary residues, whereas membrane-based processes, including membrane filtration and reverse osmosis, can achieve high separation efficiencies but are often associated with membrane fouling, relatively high energy consumption, and concentrated waste streams. Biological treatment and biosorption offer environmentally compatible alternatives, although their performance may be sensitive to operating conditions and wastewater composition. In contrast, adsorption has attracted considerable attention because of its operational simplicity, relatively low energy demand, broad applicability, and potential for pollutant or nutrient recovery.¹⁵ Among the available adsorbents, carbon-based materials are particularly attractive owing to their tunable porous structures, diverse surface functionalities, and good chemical stability. Biochar, in particular, provides the additional advantage of converting agricultural residues into value-added adsorbents. Common adsorbents for ammonium removal include biochar,¹⁶ bentonite,¹⁷ adsorption resins,¹⁸ and zeolites.¹⁹ Biochar exhibits considerable potential owing to its well-developed porous architecture,²⁰ high specific surface area, and abundant surface functional groups.²¹ However, pristine biochar typically exhibits limited adsorption capacity²² and selectivity toward NH₄⁺, especially in the presence of competing ions such as Cl⁻.²³

Surface modification has proven an effective approach to addressing these limitations and enhancing the adsorption performance of biochar.²⁴ Acid treatment efficiently removes mineral impurities and ash while introducing oxygen-rich functional groups.²⁵ Alkali activation increases the specific surface area and porosity while metal-ion modification enhances uptake of the target ion through electrostatic attraction and coordination interactions.²⁶ Biochar composited with metal oxides can form inner-sphere complexes with pollutants via synergistic interactions between the metal oxide and oxygen-rich functional groups on the biochar surface, improving the adsorption stability.²⁷ More broadly, recent studies on advanced ion-separation materials have demonstrated that selective ion capture is strongly influenced by interfacial chemistry, local charge distribution, pore architecture, and structure–transport relationships. These findings highlight the importance of rational interface engineering for improving ion selectivity rather than relying solely on adsorption capacity.^28,29 Rare-earth elements have been garnering increased attention for their potential application to surface modification because of their unique electronic configurations, high cation exchange capacities and strong complexation affinity with nitrogenous species.³⁰ For example, lanthanum (La) possesses Lewis acid sites in its oxide form (La₂O₃) that exhibit a high specificity towards NH₄⁺, making it an ideal modifier for functionalising biochar.³¹ Despite previous reports on La-modified biochars for nutrient adsorption, several important issues remain insufficiently resolved. Most earlier studies have primarily emphasized adsorption-capacity enhancement, whereas the selective recognition of NH₄⁺ in the presence of accompanying anions such as Cl⁻ has received much less attention. In addition, systematic comparison of different agricultural-residue-derived biochars under an identical La₂O₃ modification strategy remains limited, making it difficult to clarify feedstock-dependent structure–performance relationships. Moreover, the interfacial roles of La–O species, oxygen-containing functional groups, surface charge, and electronic redistribution in governing NH₄⁺ selectivity have not been comprehensively elucidated.

Accordingly, the present study differs from previous La-modified biochar investigations in four main aspects: (i) four representative agricultural residues were comparatively evaluated under the same sol–gel La₂O₃ modification route; (ii) the study focused not only on adsorption capacity but also on the selective adsorption of NH₄⁺ over Cl⁻ in NH₄Cl-containing systems; (iii) adsorption experiments were integrated with FTIR, XPS, zeta-potential analysis, coexisting-ion tests, and DFT calculations to elucidate the interfacial origin of NH₄⁺ selectivity; and (iv) the potential reuse of NH₄⁺-loaded biochar as a nutrient carrier was further explored, linking wastewater purification with nitrogen-resource recovery.

The sol–gel process is a low-temperature and wet chemical technique for synthesising stable colloidal systems that can facilitate the uniform dispersion of metal-oxide nanoparticles onto a biochar matrix.³² Compared with conventional impregnation methods, this approach effectively prevents La₂O₃ agglomeration and enhances interfacial interactions between La species and functional groups on the biochar surface,³³ which increases the dispersion of active sites and the stability of the structure.³⁴

The objective of this study is to develop a high-efficiency adsorbent for the selective removal of NH₄⁺ from NH₄Cl-rich wastewater, realize high-value resource utilization of waste materials, effectively alleviate the pressure of solid-waste disposal, and establish a green-circular utilization pattern for agricultural wastes, so as to meet the requirements for sustainable resource utilization and development.

2. Materials and methods

2.1. Materials

Analytical-grade lanthanum nitrate (La(NO₃)₃), potassium hydroxide (KOH), ammonium chloride (NH₄Cl) and hydrochloric acid (HCl) were purchased from Shanghai Macklin Biochemical Co., Ltd (Shanghai, China). Ultrapure water was utilised for all experimental procedures. To serve as the feedstocks, rapeseed straw, maize straw, rice husks and cotton stalks were collected from Nanjing (Jiangsu Province), Shandong Province, Nanjing (Jiangsu Province) and Xinjiang Uygur Autonomous Region, China, respectively.

2.2. Sample preparation

To prepare the La-modified biochar samples, each feedstock was thoroughly washed with deionized water to remove surface contaminants, oven-dried at 70 °C for 8 h, mechanically crushed, and sieved through an 80-mesh screen. For each batch, 20.0 g of dried feedstock was immersed in 200 mL of 1.0 mol L⁻¹ KOH solution for 8 h, corresponding to a solid-to-liquid ratio of 1 : 10 (g : mL). The KOH-treated feedstock was subsequently rinsed repeatedly with deionized water until the leachate reached approximately neutral pH and was then dried at 105 °C for 12 h.

For the KOH-pretreated, La-free control (K-BC), 20.0 g of the same dried feedstock was subjected to the identical KOH pretreatment described above, including immersion in 200 mL of 1.0 mol L⁻¹ KOH for 8 h, washing with deionized water until approximately neutral pH, and drying at 105 °C for 12 h. The K-BC sample did not undergo the subsequent La-containing sol–gel modification and was directly pyrolyzed at 500 °C for 2 h under the same oxygen-limited conditions as the La-modified samples.

For the subsequent sol–gel modification, La(NO₃)₃·6H₂O and citric acid monohydrate were used at a mass ratio of 2.2 : 1. The amount of La precursor was adjusted to provide a nominal La dosage of 3 wt% relative to the initial dry feedstock mass. The precursor solution was maintained at 90 °C under continuous magnetic stirring. The dried KOH-treated feedstock was then introduced into the precursor solution, and aqueous ammonia was added dropwise to adjust the pH to 7.0. The resulting suspension was subjected to intermittent ultrasonication every 30 min to minimize particle agglomeration until a gel-like consistency was obtained.

The resulting gel was oven-dried at 80 °C, ground, and subsequently pyrolyzed in a muffle furnace at 500 °C for 2 h under oxygen-limited conditions, followed by natural cooling to room temperature inside the furnace.³⁵ The resulting La-modified biochars derived from rapeseed straw, maize straw, rice husk, and cotton stalk were denoted as RS-La, BC-La, RH-La, and CS-La, respectively.³⁶

2.3. Characterisation

The specific surface area and pore structure of the samples were analysed by using a Brunauer–Emmett–Teller (BET) surface area analyser (ASAP 2460, Micromeritics, USA), N₂ adsorption–desorption measurements were performed at 77 K. The Barrett–Joyner–Halenda (BJH) method was employed to determine the pore-size distribution based on the desorption isotherm. The surface morphology was examined by field-emission scanning electron microscopy (FE-SEM) (Gala 3 XM, Tescan, Czech Republic) at an accelerating voltage of 10 kV. Microstructural features and elemental mapping were further investigated by transmission electron microscopy (TEM) (Tecnai G2 F20, FEI, USA) operated at 200 kV. Crystallographic structures were identified by X-ray diffraction (XRD) (SmartLab 3 kW, Rigaku, Japan) using Cu Kα radiation (λ = 1.5418 Å) over a scanning range of 2θ = 10°–80° and a scanning rate of 20° min⁻¹. FTIR spectra were recorded over the range of 4000–400 cm⁻¹ at a spectral resolution of 4 cm⁻¹ using 32 scans. The surface elemental composition and chemical states were analyzed by X-ray photoelectron spectroscopy (XPS) using a K-Alpha spectrometer (Thermo Fisher Scientific, USA). The actual La contents of the modified biochars were quantified by inductively coupled plasma optical emission spectrometry (ICP-OES; Agilent 5800, Agilent Technologies, USA) after acid digestion.

2.4. Adsorption experiments

0.1 g of modified-biochar sample was added into 100 mL ammonium–chloride-based solution with an initial NH₄⁺ concentration of 100 mg L⁻¹. The mixture was shaken at 200 rpm for 24 h in a thermostatic water bath maintained at 25 °C. The supernatant was filtered through a 0.45 µm syringe filter prior to detection. The residual NH₄⁺ concentration was measured using Nessler's reagent, and the adsorption capacity was subsequently calculated. Dosage-dependent NH₄⁺ adsorption experiments were conducted using adsorbent dosages of 0.1, 0.2, 0.3, 0.5, 0.7, and 1.0 g L⁻¹. Adsorption isotherm tests were carried out under initial NH₄Cl concentrations ranging from 30 to 300 mg L⁻¹, and experimental data were fitted by isotherm models. Adsorption-kinetic models were investigated over sampling intervals of 0–1080 min. Single-factor experiments were conducted to analyse the influences of pH (4.0–9.0) and coexisting ions (0.05 mol L⁻¹ anions including NO₃⁻, PO₄³⁻ and cations including Na⁺, K⁺, Mg²⁺, Ca²⁺) on NH₄⁺ adsorption. All experiments were performed in at least three parallel replicates to guarantee the validity and reliability of experimental results.

2.5. Regeneration experiments

Regeneration experiments were conducted to evaluate the reusability of BC-La. After NH₄⁺ adsorption, the spent BC-La was regenerated using 2.0 mol L⁻¹ KCl solution at 25 °C under continuous shaking. After desorption, the adsorbent was separated from the solution, repeatedly washed with deionized water until approximately neutral pH, dried, and subsequently reused in the next adsorption cycle under the same conditions as the initial adsorption experiment. Five consecutive adsorption–desorption cycles were performed.

2.6. Data processing

The adsorption capacity (q_e) of a biochar sample was calculated as follows:

1

where q_e is the amount of NH₄⁺ adsorbed (mg g⁻¹); C₀ and C_e are the initial and equilibrium concentrations, respectively, of NH₄⁺ (mg L⁻¹); V is the volume of the solution (L) and M is the mass of the biochar sample (g).

The NH₄⁺ adsorption behaviour was elucidated by fitting the kinetic data to the pseudo-first-order (PFO) and pseudo-second-order (PSO) adsorption models, which are respectively given below:

ln(q_e − q_t) = ln q_e − k₁t 2

3

where q_e and q_t are the NH₄⁺ adsorption capacities of a biochar sample at equilibrium and at time t (mg g⁻¹), respectively; k₁ is the constant of the PFO adsorption model and k₂ is the constant of the PSO adsorption model.

Ion adsorption equilibria can, in principle, be described using classical density functional theory (cDFT), which provides a rigorous statistical-mechanical framework for characterizing interfacial ion distributions and adsorption thermodynamics.^37,38 However, cDFT requires detailed descriptions of intermolecular interactions and interfacial structures and is therefore computationally demanding. Considering the experimental scope of the present study, the equilibrium adsorption data were instead correlated using the empirical Langmuir and Freundlich isotherm models.

The corresponding equations are given below:

4

5

where q_e is the NH₄⁺ adsorption capacity (mg g⁻¹), C_e is the NH₄⁺ concentration in the solution at equilibrium (mg L⁻¹), q_m is the maximum NH₄⁺ adsorption capacity (mg g⁻¹) and K_L and K_F are the adsorption constants of the Langmuir and Freundlich models, respectively.

2.7. DFT calculation

All spin-polarized density functional theory (DFT) calculations were performed using the Vienna Ab initio Simulation Package (VASP) within the projector augmented-wave (PAW) framework. The Perdew–Burke–Ernzerhof (PBE) functional within the generalized gradient approximation was employed to describe exchange–correlation interactions, and the DFT-D3 correction was included to account for dispersion interactions. A plane-wave energy cutoff of 450 eV was used. The electronic convergence criterion was set to 1 × 10⁻⁵ eV, and all structures were optimized until the residual force on each atom was less than 0.02 eV Å⁻¹. A 3 × 2 × 1 Monkhorst–Pack k-point mesh was employed.

The biochar model was represented by a 2 × 5 graphene-based supercell containing surface oxygen species, with a total of 44 atoms. The La₂O₃ model was constructed from a 2 × 2 supercell exposing the (100) surface. A BC/La₂O₃(100) heterointerface model containing 84 atoms was further constructed to represent the La-modified biochar. No atoms were fixed during structural optimization. A vacuum layer of at least 15 Å was introduced perpendicular to the surface to minimize interactions between periodic images.

For the adsorption calculations, the adsorption configurations shown in the corresponding structural models were used as the initial configurations for NH₄⁺, Cl⁻, and K⁺. No systematic screening of multiple adsorption sites was performed; therefore, the calculated adsorption energies were used primarily for qualitative comparison of the interfacial interactions under the selected representative configurations rather than for identifying the global minimum-energy adsorption states.

The adsorption energy was calculated as follows:

E_ads = E(*adsorbate) − E(*) − E(absorbate) 6

where E(*adsorbate), E(*), and E(absorbate) are the total energies of the optimized surface–adsorbate system, the clean surface, and the isolated adsorbate species, respectively. A more negative E_ads indicates stronger adsorption. For the charged adsorbates NH₄⁺, Cl⁻, and K⁺, a consistent neutralized approximation was adopted for both the adsorption systems and the corresponding isolated reference species to minimize artificial electrostatic interactions and reference-energy uncertainties associated with charged species under periodic boundary conditions. Accordingly, the calculated adsorption energies were interpreted primarily as qualitative indicators of the relative adsorption strengths among the investigated species.

3. Results and discussion

3.1. Characterisation

The pristine biochar samples (Fig. 1a–d) exhibited a typical porous structure with unevenly distributed pores across the surface and interior and with a relatively smooth lamellar or skeletal morphology.³⁹ The limited number of pores and the heterogeneous pore-size distribution reduce the surface area and available active sites, which would degrade the adsorption capacity.⁴⁰ By contrast, the La-modified biochar samples exhibited notable differences in morphology (Fig. 1e–h). Numerous particulate deposits were uniformly attached to the biochar matrix and pores, which confirms the successful loading of La species and substantial surface modification.⁴¹ The accumulation of the nanoparticles on the rough porous surface formed a disordered and hierarchical porous structure that enhanced exposure of La-based active sites while providing additional continuous mass-transfer pathways for NH₄⁺ diffusion and capture. SEM images and energy-dispersive X-ray spectroscopy (EDS) analysis of the La-modified biochar sample indicated that they primarily comprised La K and O with traces of other elements, which demonstrates the high chemical purity of the samples. La₂O₃ and oxygen-containing functional groups were uniformly distributed on the surfaces of RS-La and BC-La, which would facilitate the adsorption of NH₄⁺. Observations with TEM (Fig. 1i and j) revealed that the nanoparticles were uniformly dispersed on external surfaces and within internal pores. They exhibited near-spherical or irregular polyhedral shapes with a relatively narrow size distribution and negligible aggregation,⁴² which improved accessibility to active sites. HRTEM images (Fig. 1k) clearly showed lattice fringes with an interplanar spacing of approximately 0.312 nm, which corresponds to the (101) plane of La₂O₃ and confirms that La species predominantly existed as a crystalline oxide.⁴³ The selected area electron diffraction (SAED) pattern (Fig. 1l) showed multiple concentric diffraction rings indexed to several La₂O₃ crystal planes, which was consistent with XRD results and indicates the formation of highly crystalline La₂O₃ nanoparticles.⁴⁴ Such polycrystalline nanostructures expose crystal facets with multiple orientations, which provides diverse active sites that facilitate selective NH₄⁺ adsorption.

Fig. 1. SEM images of (a–d) pristine and (e–h) La-modified biochar samples. TEM images of BC-La at (i) 500 nm and (j) 100 nm magnifications. (k) HRTEM image of BC-La. (l) SAED pattern of BC-La.

XRD patterns of all La-modified biochar samples exhibited similar phase compositions (Fig. 2). Broad reflections at 2θ ≈ 24° and 43° corresponded to the (002) and (100) planes of turbostratic carbon, respectively, which indicates that a disordered carbon structure was retained after the low-temperature pyrolysis. Distinct diffraction peaks at 27.4°, 28.1°, 30.0°, 39.6°, 46.4°, 48.8°, and 55.4° were assigned to hexagonal La₂O₃ (JCPDS No. 83-1344), confirming the successful formation of crystalline La₂O₃ during the sol–gel-assisted pyrolysis process. The relatively broad diffraction profiles suggest that nanocrystalline domains were constrained within the porous biochar matrix, which maximised the exposure of active sites on the surface.⁴⁵

Fig. 2. XRD patterns of La-modified biochar samples.

The FTIR spectra revealed pronounced changes in the surface functional groups of the La-modified biochar samples before and after NH₄⁺ adsorption (Fig. 3a–d). Prior to adsorption, a broad band was centred at ∼3400 cm⁻¹, which corresponded to the stretching vibration of surface hydroxyl groups (–OH) and hydrogen-bonded water and indicates strong hydrophilicity and abundant active sites. The band at 1600–1650 cm⁻¹ was attributed to the bending vibration of adsorbed water (H–O–H), which partially overlapped with vibrations of carboxyl (–COOH) or conjugated carbonyl (C O) groups. The peaks at 1000–1100 cm⁻¹ were attributed to C–O and C–O–C stretching vibrations, which confirms the presence of oxygen-containing functional groups on the biochar surface. After NH₄⁺ adsorption, the –OH band at ∼3400 cm⁻¹ exhibited noticeable changes in intensity and width, suggesting the involvement of surface hydroxyl groups in the adsorption process, particularly through hydrogen-bonding interactions. Meanwhile, the slight shift of the band at 1600–1650 cm⁻¹ indicates changes in the local environment of carboxyl/carbonyl groups and adsorbed water, suggesting that oxygen-containing functional groups also participate in NH₄⁺ adsorption. Meanwhile, the band at 1600–1650 cm⁻¹ showed a slight shift and reduced intensity, which indicates alterations in the vibrational environment of carboxyl groups and/or adsorbed water molecules and further supports the presence of electrostatic attraction or weak complexation between NH₄⁺ and oxygen-containing functional groups. No distinct new vibrational bands assignable to NH₄⁺ were detected at 1000–1200 cm⁻¹, which can be attributed to the intrinsically weak infrared activity of NH₄⁺ and the overlap of its characteristic vibrations with the intense C–O stretching bands of the biochar matrix.

Fig. 3. FTIR spectra of La-modified biochar samples before and after NH₄⁺ adsorption: (a) RS-La, (b) BC-La, (c) CS-La and (d) RH-La.

This is consistent with previous FTIR studies on NH₄⁺ adsorption by carbon-based materials. Overall, the FTIR analysis indicates that the La modification generated a composite reactive interface dominated by La–OH species and oxygen-containing functional groups.⁴⁶ These spectral changes suggest that NH₄⁺ adsorption involves specific interfacial interactions, including ion exchange, electrostatic attraction, and hydrogen bonding, rather than simple physical pore filling alone.⁴⁷ He wide peaks of –OH stretching vibration at 3000–3500 cm⁻¹ before and after adsorption shifted and their intensity changed. The characteristic peaks of –COOH and C O near 1700 cm⁻¹ also shifted, indicating that intermolecular hydrogen bonds were formed between the oxygen-containing functional groups such as –OH, –COOH, and C O on the material surface and the adsorbate. Hydrogen bonds, as secondary weak interactions, can further enhance the interfacial affinity and improve the stability of the adsorption system.

3.2. Spectral analysis

Fig. 4 presents the nitrogen adsorption–desorption isotherms and pore-size distribution curves of pristine BC and La-modified BC. The pore-size distribution curves in Fig. 4 correspond to Type-IV isotherms with H3-type hysteresis loops, demonstrating that the materials are predominantly composed of mesoporous structures. Table S1 summarises the surface parameters of BC and La-BC calculated via the BJH method. Notably, small-sized mesopores are widely distributed within the as-prepared materials. Compared with pristine BC, La-BC possesses a larger specific surface area and pore volume, which indicates abundant active adsorption sites facilitating the adsorption procedure.⁴⁸

Fig. 4. (a) N₂ adsorption–desorption isotherms measured at 77 K and (b) pore-size distributions of BC before and after La modification via the sol–gel process.

The XPS survey spectra (Fig. 5a) of La-modified biochar samples before NH₄⁺ adsorption indicated the presence of La, O, C and N on the sample surface, which confirmed the successful incorporation of La species and the coexistence of nitrogen-containing functional groups. After NH₄⁺ adsorption, both the La 3d_5/2 and La 3d_3/2 peaks shifted toward lower binding energies (Fig. 5b), indicating adsorption-induced redistribution of the local electron density around the La centers. This binding-energy shift suggests that La-containing surface sites were involved in the interfacial interaction with NH₄⁺. Importantly, this shift does not imply that La³⁺ becomes an electron-rich center; rather, La³⁺ retains its electrophilic Lewis-acid character, while the surrounding oxygen atoms act as relatively electron-rich sites. The observed shift therefore reflects local electronic polarization and charge redistribution within the La–O interfacial environment during NH₄⁺ adsorption. The O 1s spectrum before NH₄⁺ adsorption (Fig. 5c) can be deconvoluted into two main components: a lower binding energy peak at ∼531.3 eV associated with lattice oxygen, metal–oxygen bonds and carbonyl oxygen; and a higher binding energy peak at ∼532.5 eV attributed to surface hydroxyl groups. Following NH₄⁺ adsorption, slight shifts in the lower-energy component and a pronounced increase in the higher-energy component were observed, which indicates the involvement of surface –OH groups and O–La species in NH₄⁺ immobilisation that was most likely via hydrogen bonding. The N 1s spectrum (Fig. 5d) before NH₄⁺ adsorption comprised pyridinic N (398.5 eV), pyrrolic N (∼400.0 eV) and graphitic N (∼401.2 eV). After NH₄⁺ adsorption, both the binding energies and relative peak areas changed with more pronounced variations observed for pyridinic and pyrrolic N. These results suggest that nitrogen-containing functional groups provided charged sites that enhance electrostatic attraction and facilitated hydrogen bonding with NH₄⁺. Collectively, the negative shift of the La 3d peaks, redistribution of O 1s components and adsorption-induced changes in N 1s spectra demonstrate that NH₄⁺ adsorption was governed by synergistic interactions. After adsorption, the peak positions, peak shapes and intensities of the characteristic peaks of La 3d_3/2 and La3 d_5/2 all changed significantly, indicating that the chemical environment of La element was significantly altered due to coordination. Meanwhile, the binding energies of pyridinic N, pyrrolic N, and graphitic N in the N 1s spectrum, as well as C–O and C O species in the O 1s spectrum, exhibited noticeable shifts after NH₄⁺ adsorption, indicating changes in the local electronic environments of nitrogen- and oxygen-containing surface groups. These changes suggest that such functional groups participate in the interfacial interactions involving La-containing sites and NH₄⁺. Rather than providing direct evidence for the formation of specific stable La–N coordination bonds, the XPS results support adsorption-induced electronic redistribution and interfacial polarization involving La–O species and surface functional groups. La centres modulated the local electronic structure and created polarised adsorption domains⁴⁸ while oxygen- and nitrogen-containing functional groups jointly stabilised NH₄⁺ through electrostatic attraction and hydrogen bonding.

Fig. 5. XPS spectra of La-modified biochar samples before and after NH₄⁺ adsorption: (a) survey spectra, (b) La 3d spectra, (c) N 1s spectra and (d) O 1s spectra.

3.3. Adsorption performance

3.3.1. Adsorption capacity

As shown in Fig. 6, both pristine biochar and La-modified biochar exhibited a distinct rapid adsorption–equilibrium process for NH₄⁺ adsorption; however, La modification significantly enhanced the adsorption performance. For pristine biochar (Fig. 6a), NH₄⁺ adsorption mainly occurred within the first 60 min and gradually reached equilibrium thereafter. Among the tested biochars, BC exhibited the highest adsorption capacity, reaching approximately 7.3 mg g⁻¹ at 180 min, followed by CS (approximately 6.4 mg g⁻¹), RS (approximately 4.8 mg g⁻¹), and RH (approximately 3.7 mg g⁻¹), indicating that the physicochemical properties of the biochar precursors played a crucial role in determining NH₄⁺ adsorption performance. After La modification (Fig. 6b), the NH₄⁺ adsorption capacities of all biochars were remarkably improved, with equilibrium adsorption capacities ranging from approximately 34 to 44 mg g⁻¹. Among them, BC-La and RS-La exhibited the highest adsorption capacities, reaching 42.23 and 43.37 mg g⁻¹, respectively, which were approximately 5.6 times higher than that of pristine BC. This result demonstrates that La loading effectively introduced additional adsorption active sites and promoted NH₄⁺ immobilization.

Fig. 6. (a) NH₄⁺ adsorption on pristine biochar; (b) NH₄⁺ adsorption on modified biochar; (c) Cl⁻ adsorption on modified biochar; (d) pH effect on NH₄⁺ adsorption.

In contrast, the adsorption capacity of La-modified biochars toward Cl⁻ was relatively low (Fig. 6c), with equilibrium adsorption capacities only ranging from 0.39 to 0.52 mg g⁻¹, which were substantially lower than those for NH₄⁺. This finding indicates that La-modified biochars possess excellent selective adsorption ability toward NH₄⁺ rather than non-selective ion adsorption. The effect of pH on NH₄⁺ adsorption (Fig. 6d) further demonstrated that adsorption performance was strongly dependent on solution pH. All materials exhibited the highest adsorption capacities within the pH range of 6–7, and BC-La achieved the maximum adsorption capacity of approximately 42.23 mg g⁻¹ at pH 7. Under acidic conditions, excessive H⁺ ions competed with NH₄⁺ for adsorption sites, thereby inhibiting NH₄⁺ immobilization. In contrast, under alkaline conditions, the conversion of NH₄⁺ to NH₃ and the competitive adsorption of OH⁻ ions resulted in decreased adsorption efficiency. Overall, La modification significantly enhanced the adsorption capacity and selectivity of biochar toward NH₄⁺ by introducing La-based active sites, optimizing surface charge properties, and strengthening ion-exchange interactions, demonstrating its great potential for nitrogen-containing wastewater treatment and nitrogen resource recovery applications.

As shown in Fig. 7, the effects of adsorbent dosage, initial NH₄⁺ concentration, surface charge properties, and coexisting ions on the adsorption performance of La-modified biochar were systematically investigated. With increasing adsorbent dosage from 0.1 to 1.0 g L⁻¹, the NH₄⁺ removal efficiency significantly increased, whereas the adsorption capacity per unit mass (q_e) gradually decreased (Fig. 7a). Among the tested samples, BC-La and RS-La exhibited superior removal efficiencies, reaching approximately 90% and 88% at an adsorbent dosage of 1.0 g L⁻¹, respectively, while their corresponding q_e values stabilized within the range of 8–12 mg g⁻¹. This phenomenon suggests that increasing the adsorbent dosage enhances the overall removal efficiency by providing more available adsorption sites, but simultaneously decreases the utilization efficiency of adsorption sites due to the lower NH₄⁺ loading per unit mass of adsorbent.

Fig. 7. (a) Effect of solid–liquid ratio; (b) different initial concentrations; (c) zeta-potential (d) effect of coexisting ions on NH₄⁺ adsorption.

The initial NH₄⁺ concentration exerted a significant influence on the adsorption behavior (Fig. 7b). Increasing the initial NH₄⁺ concentration promoted the adsorption capacity of La-modified biochars, with BC-La achieving the maximum q_e of approximately 45 mg g⁻¹ at an initial concentration of 300 mg L⁻¹. However, the removal efficiency decreased from approximately 80–90% to 40–50% with increasing NH₄⁺ concentration, which can be attributed to the gradual saturation of adsorption sites and the limited availability of active binding sites under high NH₄⁺ loading conditions.

The surface charge characteristics of BC-La were further evaluated by zeta-potential analysis (Fig. 7c). The point of zero charge pH_pzc of BC-La was approximately 3.72. When the solution pH exceeded the pH_pzc, deprotonation of acidic oxygen-containing surface groups became increasingly significant. Carboxyl groups (–COOH) could dissociate to form –COO⁻, while hydroxyl groups could generate negatively charged –O⁻ species. Surface La–OH groups may also undergo deprotonation to form La–O⁻ sites. The accumulation of these negatively charged species accounts for the progressively negative zeta-potential of BC-La with increasing pH, thereby promoting the electrostatic attraction of NH₄⁺ and contributing to its adsorption.

The effects of coexisting ions on NH₄⁺ adsorption are presented in Fig. 7d. The presence of NO₃⁻ caused only a minor influence on NH₄⁺ adsorption, whereas PO₄³⁻ and Ca²⁺ significantly inhibited the adsorption performance, decreasing the adsorption capacity to approximately 17 and 19 mg g⁻¹, respectively. The competitive effects of coexisting cations, particularly K⁺, Ca²⁺, and Mg²⁺, indicate competition for negatively charged surface sites and suggest that ion exchange may contribute to NH₄⁺ adsorption. However, ion exchange is considered one of several contributing mechanisms rather than the sole adsorption pathway.

The physicochemical characteristics and NH₄⁺ adsorption performance of KOH-treated biochar (K-BC) were further evaluated. As shown in Fig. S3, the FTIR spectrum of K-BC exhibited characteristic bands associated with –OH, C O, –COO⁻, and C–O groups, indicating that KOH treatment modified the surface oxygen-containing functional groups. The XRD pattern further revealed structural changes after KOH treatment. The NH₄⁺ adsorption capacity of K-BC increased rapidly during the initial adsorption stage and gradually approached equilibrium, reaching approximately 15 mg g⁻¹. The adsorption performance was also pH-dependent, with the highest capacity observed around pH 7. These results indicate that KOH pretreatment itself contributes to NH₄⁺ adsorption through surface activation and modification of oxygen-containing functional groups. However, the adsorption capacity of K-BC remained substantially lower than that of BC-La, which reached approximately 43–48 mg g⁻¹ under comparable adsorption conditions. Therefore, the enhanced adsorption performance of BC-La cannot be attributed solely to KOH pretreatment, and the introduction of La-containing interfacial sites provides an additional contribution to NH₄⁺ adsorption. Overall, La modification significantly enhanced the NH₄⁺ adsorption capacity and selectivity of biochar. The combined experimental results indicate that KOH pretreatment contributes to surface activation, whereas the subsequent introduction of La-containing interfacial sites further improves NH₄⁺ adsorption and ion selectivity. These findings demonstrate the potential of BC-La for nitrogen removal and resource recovery from complex aqueous environments.

3.3.2. Regeneration and reusability

The regeneration performance of BC-La was evaluated over five consecutive adsorption–desorption cycles. As shown in Fig. S1, the NH₄⁺ adsorption capacity gradually decreased from 43.73 mg g⁻¹ in the first cycle to 41.64, 38.28, 35.42, and 32.84 mg g⁻¹ in the second to fifth cycles, respectively. After five cycles, BC-La retained approximately 75.1% of its initial adsorption capacity. The gradual decrease in adsorption performance may be attributed to incomplete desorption of strongly retained NH₄⁺, partial occupation or blockage of active sites, and changes in the accessibility of surface functional groups during repeated regeneration. Nevertheless, the material maintained a considerable adsorption capacity after repeated use, indicating reasonable regeneration potential for NH₄⁺ removal.

3.3.3. Adsorption kinetics

As shown in Fig. 8, the pseudo-first-order (PFO) and pseudo-second-order (PSO) kinetic models, together with the Langmuir and Freundlich isotherm models, were employed to characterize the adsorption kinetics and equilibrium behavior of NH₄⁺ on the La-modified biochars. Further adsorption isotherm analyses are presented in Fig. 8c and d. With increasing equilibrium concentration (C_e), the adsorption capacities of all La-modified biochars gradually increased and subsequently approached a plateau, indicating progressive saturation of the available adsorption sites. The Langmuir model provided a better fit to the experimental equilibrium data than the Freundlich model, indicating that the adsorption behavior within the investigated concentration range can be effectively described by a finite-capacity, monolayer-type adsorption model. Importantly, the superior Langmuir fitting should not be interpreted as evidence of a microscopically homogeneous surface. BC-La contains chemically heterogeneous adsorption environments, including La–O species, oxygen-containing functional groups, structural defects, and carbonaceous domains. Therefore, the Langmuir model is used here as an empirical description of the equilibrium adsorption behavior rather than as direct proof of uniform adsorption sites. According to the Langmuir fitting results, the maximum adsorption capacities of the tested materials were approximately 43–48 mg g⁻¹, with RS-La and BC-La exhibiting relatively high adsorption capacities, consistent with the experimental trend. The kinetic fitting results revealed that the pseudo-second-order (PSO) model provided a better description of the experimental data than the PFO model, indicating that the adsorption kinetics of NH₄⁺ on the La-modified biochars were better represented by the PSO model. However, kinetic fitting alone cannot conclusively identify the adsorption mechanism. Therefore, the adsorption mechanism was further evaluated by integrating the kinetic results with FTIR, XPS, zeta-potential analysis, coexisting-ion experiments, and DFT calculations.

Fig. 8. Fitting the experimental results of the La-modified biochar samples to the (a) PFO and (b) PSO kinetic models and to the (c) Langmuir and (d) Freundlich isotherm models.

Overall, La modification significantly enhanced the NH₄⁺ adsorption capacity of biochar. The PSO kinetic model and Langmuir isotherm model provided satisfactory descriptions of the adsorption kinetics and equilibrium behavior, respectively. The adsorption mechanism was instead interpreted on the basis of complementary spectroscopic, surface-charge, competitive-ion, and DFT evidence, which collectively suggests contributions from electrostatic attraction, ion exchange, hydrogen bonding, and La-associated interfacial interactions. The Langmuir maximum adsorption capacity of BC-La was 47.93 mg NH₄⁺ g⁻¹. For comparison, HNO₃/NaOH-modified corncob biochar, wood biochar, rice-husk biochar, and H₂O₂-aged peanut-shell biochar showed reported ammonium adsorption capacities corresponding to approximately 29.06, 57.39, 51.17, and 158.50 mg NH₄⁺ g⁻¹, respectively, after conversion from an NH₄⁺–N basis. A NaOH-modified oat-husk biochar showed a reported maximum ammonium adsorption capacity of 59.61 mg g⁻¹; because the mass basis was not explicitly converted here, the value is retained as originally reported. These comparisons indicate that BC-La exhibits a competitive adsorption capacity within the broad range reported for biochar-based ammonium adsorbents, although direct comparison should be interpreted cautiously because of differences in feedstock, modification method, solution composition, pH, temperature, and adsorption conditions. These findings indicate that La-modified biochar possesses high NH₄⁺ adsorption capacity, stable adsorption performance, and great potential for efficient nitrogen removal and resource recovery from aqueous environments.

3.3.4. Pot experiment

Fig. S2 presents the preliminary plant-growth responses under the different treatments. Compared with the untreated control, biochar-based amendments generally improved plant growth. The La₂O₃-modified biochar treatment showed a favorable effect on root development and maintained overall plant growth, suggesting that NH₄⁺-loaded BC-La has potential as a nutrient carrier. However, its effects on shoot and total biomass were not consistently superior to those of conventional fertilizer treatment. Therefore, the pot experiment is interpreted as preliminary evidence for the agricultural reuse potential of NH₄⁺-loaded biochar rather than as proof of superior fertilization performance. Further long-term studies are required to evaluate nutrient-release behavior, nitrogen-use efficiency, crop productivity, and possible La-related effects in soil–plant systems.

3.3.5. DFT calculations

In the above differential charge density maps, yellow represents regions of electron accumulation, while cyan represents regions of electron depletion; the isosurface level for all maps has been set to 0.001 e per bohr³ to ensure that the magnitude of electronic redistribution is comparable across different systems.

To further elucidate the preferential adsorption mechanism of NH₄⁺ on La₂O₃-modified biochar in the NH₄Cl system, the interfacial interaction behaviors of pristine biochar and the La₂O₃-modified biochar heterojunction models were systematically investigated, as shown in Fig. 9a and b.

Fig. 9. DFT-calculated adsorption configurations and charge-density-difference plots. (a) Structural model of pristine BC and (b) structural model of BC-La. (c–e) Initial adsorption configuration, optimized adsorption structure with adsorption energy, and charge-density-difference plot for Cl⁻ adsorption on BC, respectively. (f–h) Corresponding results for NH₄⁺ adsorption on BC. (i–k) Corresponding results for Cl⁻ adsorption on BC-La. (l–n) Corresponding results for NH₄⁺ adsorption on BC-La. (o) Optimized adsorption configuration of K⁺ on BC-La with an adsorption energy of −4.24 eV.

After La₂O₃ modification, the biochar surface transformed from an interface dominated by carbon-based oxygen-containing functional groups into a composite heterogeneous interface consisting of C–O, La–O species, and surface defect sites. Owing to its relatively large ionic radius, high coordination number, and strong Lewis acidity, La³⁺ can form multi-coordinated La–O polycentric structures with surrounding oxygen atoms, thereby stabilizing surface active sites. On the one hand, the strong ionic character of La–O bonds generates pronounced charge separation on the modified surface, where La sites act as electrophilic centers, while O sites serve as electron-rich centers and hydrogen-bond acceptors. On the other hand, the strong complexation capability of La³⁺ enhances the local trapping of ions, leading to substantially reduced adsorption energies for both NH₄⁺ and Cl⁻ on the modified surface. As illustrated in Fig. 9d, g, j and m, the adsorption energy of Cl⁻ on pristine biochar was only −0.73 eV, whereas it decreased to −4.55 eV after La₂O₃ modification. Similarly, the adsorption energy of NH₄⁺ decreased from −2.77 eV on pristine biochar to −5.60 eV on the modified surface. Notably, the adsorption energy of NH₄⁺ remained 1.05 eV lower than that of Cl⁻ after modification, indicating that La₂O₃ modification significantly enhanced the overall interfacial affinity toward ions while maintaining a stronger interaction with NH₄⁺.

Bader charge analysis further revealed distinct electronic regulation effects toward NH₄⁺ and Cl⁻ on different surfaces. As shown in Fig. 9e and h, Cl gained 0.27 e⁻, whereas the NH₄⁺ moiety lost 0.70 e⁻ on pristine biochar. This result suggests that the interaction between Cl⁻ and pristine biochar is relatively weak, mainly arising from van der Waals interactions, defect-induced electrostatic attraction, and local electronic polarization. Although Cl⁻ gained 0.27 e⁻, the relatively small adsorption energy indicates that such electron transfer is insufficient to induce strong coordination or chemisorption. In contrast, NH₄⁺ lost 0.70 e⁻ upon adsorption on pristine biochar, reflecting a stronger interaction with the surface, which is consistent with its more negative adsorption energy (−2.77 eV) and indicates the formation of a relatively stable adsorption configuration.

Following La₂O₃ modification, the electron-transfer behavior of Cl⁻ changed significantly, shifting from gaining 0.27 e⁻ on pristine biochar to losing 0.34 e⁻ on the modified surface. This finding indicates that La₂O₃ modification substantially altered the interfacial charge distribution, causing Cl⁻ to exhibit electron depletion rather than electron accumulation near the modified surface. Combined with the differential charge density distribution shown in Fig. 9k, pronounced electron redistribution can be observed at the Cl⁻/La₂O₃-modified interface, where both electron accumulation and depletion regions coexist, suggesting strong polarization induced by the La–O polar structure. However, the electron loss of Cl⁻ implies a weakening of its anionic electron cloud during adsorption, which is unfavorable for maintaining a stable anionic state at the surface. Therefore, although Cl⁻ exhibits a relatively strong adsorption energy (−4.55 eV) on the La₂O₃-modified surface, this interaction is mainly attributed to polarization induction and local complexation by the highly polar interface rather than stable and selective retention.

For NH₄⁺, the amount of electron loss decreased from 0.70 e⁻ on pristine biochar to 0.27 e⁻ after La₂O₃ modification, whereas the adsorption energy became substantially more negative, decreasing from −2.77 eV to −5.60 eV. This phenomenon indicates that the enhanced adsorption of NH₄⁺ on the modified surface is not governed by increased electron transfer but rather originates from the strong electrostatic stabilization, hydrogen-bond interactions, and ion–dipole interactions provided by La–O active sites. As shown in Fig. 9n, distinct regions of electron accumulation and electron depletion are observed at the NH₄⁺/La₂O₃-modified interface, indicating significant interfacial electron-density redistribution during adsorption. The electron depletion around NH₄⁺ and the corresponding electron accumulation near surface O and La–O sites collectively demonstrate strong polarization effects and interfacial charge coupling. Because NH₄⁺ loses only 0.27 e⁻ on the modified surface, its positive charge characteristic is largely preserved, which favors the formation of a more stable adsorption environment. In addition to electrostatic stabilization by the La–O polar structure, the hydrogen atoms in the N–H bonds of NH₄⁺ can form N–H⋯O hydrogen bonds with surface oxygen sites. Consequently, La₂O₃ modification reduces charge dissipation during NH₄⁺ adsorption, enabling NH₄⁺ to maintain its cationic character while being stabilized through hydrogen-bonding and ion–dipole interactions. Therefore, the selectivity of La₂O₃-modified biochar toward NH₄⁺ is not solely determined by adsorption energy but is collectively governed by enhanced adsorption stability, more favorable hydrogen-bonding and electrostatic interactions, and the electronic destabilization of Cl⁻ at the interface.

To further evaluate the competitive adsorption of coexisting cations, K⁺ was selected as a representative competing ion because of its similar monovalent positive charge to NH₄⁺. The calculated adsorption energy of K⁺ on the BC-La surface was −4.24 eV, which was less negative than that of NH₄⁺ (−5.60 eV). This difference indicates that although K⁺ can interact with and compete for negatively charged surface sites, NH₄⁺ is more strongly stabilized at the La-modified interface. The stronger affinity toward NH₄⁺ is consistent with its additional ability to form directional N–H⋯O hydrogen bonds and ion–dipole interactions with La–O-containing surface sites, interactions that are not available to K⁺.

3.3.6. Adsorption mechanism

The incorporation of La₂O₃ effectively reconstructs the active surface sites of biochar. In pristine biochar, NH₄⁺ adsorption primarily relies on oxygen-containing functional groups, defect sites, and the carbonaceous π-electron system, resulting in relatively limited adsorption modes and binding strength. Following La₂O₃ modification, highly polar La–O dual-center active structures are generated on the biochar surface. Specifically, La³⁺ acts as a high-coordination Lewis acid site, which effectively regulates interfacial electron distribution and enhances interfacial polarization effects. Meanwhile, oxygen atoms serve as electron-rich sites and hydrogen-bond acceptors, providing multiple stable binding sites for NH₄⁺ and thereby significantly improving the selective recognition and immobilization of NH₄⁺. In addition, La₂O₃ modification promotes interfacial charge separation and strengthens local electric field effects. Differential charge density analyses reveal pronounced electron redistribution among La₂O₃, the biochar matrix, and the adsorbate during the adsorption process, leading to the formation of a stable polar interface. This enhanced interfacial polarization reinforces the electrostatic attraction and orientation-induced interactions toward NH₄⁺, resulting in a substantial decrease in the adsorption energy from −2.77 eV to −5.60 eV and markedly improving adsorption stability. La₂O₃ modification alters the interfacial electron-transfer behavior of NH₄⁺ and Cl⁻. The electron loss of NH₄⁺ is significantly reduced, and its adsorption is primarily stabilized through hydrogen bonding, electrostatic interactions, and ion–dipole interactions mediated by La–O active sites. In contrast, Cl⁻ undergoes a transition from an electron-accepting species to an electron-donating species, accompanied by depletion of its electron density, which renders its interfacial retention less favorable. Consequently, La₂O₃ modification enhances the preferential adsorption of NH₄⁺. NH₄⁺ can be stabilized through N–H⋯O hydrogen bonding, electrostatic attraction, and ion–dipole interactions, whereas Cl⁻ interacts more weakly with La-containing surface sites, making its retention less favorable. More broadly, recent studies on advanced ion-separation materials have shown that ion selectivity is strongly influenced by interfacial chemistry, local charge distribution, pore architecture, and structure–transport relationships. These observations are consistent with the broader principle that rational interface engineering can enhance ion selectivity by regulating specific ion–surface interactions rather than adsorption capacity alone.^49,50 The DFT results further support the preferential adsorption of NH₄⁺ over competing monovalent cations. The adsorption energy of K⁺ on BC-La was calculated to be −4.24 eV, which was less negative than that of NH₄⁺ (−5.60 eV), indicating a stronger intrinsic interfacial affinity of BC-La toward NH₄⁺. Taken together, the adsorption mechanism should not be assigned on the basis of a single characterization technique. FTIR supports the participation of oxygen-containing functional groups and hydrogen bonding; XPS reveals adsorption-induced changes in the local electronic environments of La-, O-, and N-containing species; zeta-potential measurements support electrostatic attraction; coexisting-ion experiments suggest a contribution from ion competition and ion exchange; and DFT calculations further demonstrate favorable N–H⋯O hydrogen bonding, electrostatic stabilization, and ion–dipole interactions at the La–O-modified interface. Therefore, NH₄⁺ adsorption on BC-La is attributed to the combined contribution of multiple interfacial interactions rather than a single adsorption mechanism (Fig. 10).

Fig. 10. Schematic representation of surface complexation and ion-exchange mechanisms governing NH₄⁺ adsorption by La-modified biochar.

4. Conclusions

La-modified biochar samples were successfully synthesised using different agricultural residues as feedstock, and their adsorption of NH₄⁺ from NH₄Cl-rich wastewater was systematically evaluated. The synthetic strategy of combining the sol–gel process with low-temperature pyrolysis resulted in the uniform dispersion of crystalline La₂O₃ nanoparticles within the biochar matrix and the formation of abundant La–OH active sites, which was accompanied by a hierarchical micro- and mesoporous structure. Compared with pristine biochar samples, the La-modified biochar samples showed notable increases in the specific surface area, pore volume and microporosity, which enhanced the accessibility to active sites and diffusion efficiency. In adsorption experiments, the La-modified biochar samples exhibited high selectivity, rapid kinetics and a considerable adsorption capacity for NH₄⁺. This study provides a green and feasible technical pathway for the efficient purification of nitrogen-containing wastewater and the cyclic reuse of nitrogen resources, which aligns with the dual demands of water pollution remediation and sustainable resource utilization. Future research will involve exploring the potential integration as a nutrient carrier in soil amendment and slow-release fertiliser systems.

Author contributions

Qi Wang: writing – original draft, validation, software, investigation, formal analysis, data curation, conceptualization. Li Wang: writing – review & editing, supervision, software, formal analysis, data curation. Enliang Ren: validation, supervision, methodology, investigation. Peihui Sun: resources, investigation, data curation. Entong Zhu: investigation, data curation. Qiaolin Hu: investigation, data curation. Yunbo Zhao: methodology, investigation. Zhongyi Qu: writing – review & editing, supervision, resources, project administration, methodology, funding acquisition.

Conflicts of interest

No potential conflicts of interest was reported by the author(s).

Data availability

The data that support the findings of this study, including material characterization data, adsorption modelling simulation data, soil experimental data, and crop growth data, are available within the supplementary information (SI) of this article. Raw experimental data are available on reasonable request from the corresponding author. Supplementary information: regeneration data, pot-experiment results, KOH-treated biochar characterization, BET and pore-structure data, kinetic and isotherm parameters, plant biomass and soil properties, comparison with other biochar-based adsorbents, and ICP-determined La contents. See DOI: https://doi.org/10.1039/d6ra07475h.