Molecular-level EDTA-2Na modification of zeolite–biochar composites for efficient heavy metal adsorption: Preparation, mechanism and adsorption kinetics analysis

2.1. Materials and preparation of samples

The primary chemicals used in this work included zinc chloride (ZnCl₂, 99%) and copper chloride (CuCl₂, 98%) purchased from Aladdin Reagent Co., Ltd. (China); concentrated hydrochloric acid (HCl), nitric acid (HNO₃), sodium hydroxide (NaOH, 95%), and sodium chloride (NaCl, 99%) obtained from Sinopharm Chemical Reagent Co., Ltd. (China); and potassium bromide (KBr) supplied by Energy Chemical Co., Ltd. (China). All reagents were of analytical reagent (AR) grade and used without further purification. Deionized water was prepared using a laboratory-built purification system. Experimental deionized water was produced by the laboratory’s purification system. Rice husk powder, sourced from a local agricultural market, was thoroughly washed and dried. The dried material was pyrolyzed at 500 °C for 4 h under a nitrogen atmosphere in a tubular furnace with a heating rate of 10 °C·min⁻¹. After cooling to room temperature, the biochar (BC) was ground, passed through a 100-mesh sieve, and stored in sealed bags for later use.

The zeolite used in this study was a synthetic zeolite (particle size <10 μm) purchased from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China) and used without further purification. For surface modification, zeolite was individually treated with 0.1 mol·L⁻¹ HCl, KOH, or NaCl solutions at a solid-to-liquid ratio of 1:20. The suspensions were stirred at room temperature for 6 h and left to settle for 24 h. After washing until neutral pH, the materials were dried at 110 °C and ground. The modified zeolites were designated Z-HCl, Z-KOH, and Z-NaCl. Composite materials were then prepared by combining zeolite (either unmodified or HCl-modified) with biochar at mass ratios of 1:1, 2:1, 3:1, and 4:1. Two approaches were used: (i) direct mixing followed by stirring and drying, resulting in composites ZB1–ZB4 (unmodified) and ZBH1–ZBH4 (HCl-modified); and (ii) co-pyrolysis of zeolite and rice husk to obtain ZBG1–ZBG4 (unmodified) and ZBHG1–ZBHG4 (HCl-modified). Detailed information on sample nomenclature and composition ratios is provided in Table 1.

The sixteen composite samples were characterized using Fourier Transform Infrared Spectroscopy (FT-IR, Nicolet-6700, USA) with KBr pellet method at a 1:200 sample-to-KBr ratio, resolution 4 cm⁻¹, scan range 4000–500 cm⁻¹; Brunauer–Emmett–Teller (BET) surface analysis using Micromeritics ASAP-2460 (USA) with nitrogen adsorption at 77 K after degassing samples at 100 °C for 4 h; X-ray diffraction (XRD, Bruker D8 ADVANCE, Germany) at 2θ = 10–90°; and scanning electron microscopy (SEM, JEOL JEM-6510, Japan) at 200 kV. For adsorption experiments, 50 mL of Cu(II) or Zn(II) solutions at predetermined concentrations were placed in centrifuge tubes and different amounts of zeolite–biochar composites were added. The mixtures were shaken at room temperature for a set time, then centrifuged at 4000 rpm for 10 min. The supernatant was filtered through a 0.22 μm membrane and analyzed using an inductively coupled plasma mass spectrometer (ICP-7510, Japan) to determine Cu(II) and Zn(II) concentrations.

2.2. Preparation of EDTA-2Na-modified zeolite-biochar composites

Preliminary pretreatment experiments were conducted on zeolite–biochar composites subjected to different activation methods (HCl, KOH, and NaCl; designated Z-HCl, Z-KOH, and Z-NaCl, respectively; see Table S1 in the Supporting Information for additional details. The experimental validation of this result is presented in the following section. Through comparison of their adsorption-related properties, the ZBH2 and ZBHG1 composites were selected as representative samples for subsequent functionalization with disodium ethylenediaminetetraacetate (EDTA-2Na), aiming to further enhance their heavy metal adsorption performance, which exhibited superior adsorption potential, were selected for functionalization using disodium ethylenediaminetetraacetate (EDTA-2Na). EDTA-2Na solutions with concentrations of 0.05 and 0.1 mol·L⁻¹ were prepared, and the composites were added at a weight-to-volume ratio of 1:20 (1 g sample in 20 mL solution). The reaction mixtures were maintained at 60 °C in a thermostatic bath with constant stirring for 4 h, followed by standing for 2 h to ensure sufficient interaction between the EDTA-2Na functional groups (–COOH and –NH₂) and the surface metal sites of the zeolite–biochar matrix.

To remove unreacted EDTA-2Na and prevent contamination in subsequent adsorption experiments, the suspensions were vacuum-filtered and the solids were thoroughly washed with deionized water until neutral pH was achieved. The filtered composites were then dried at 110 °C to a constant weight. The resulting functionalized materials were labeled as ZBHE1 and ZBHE2 for ZBH2 treated with 0.05 and 0.1 M EDTA-2Na, respectively, and ZBHGE1 and ZBHGE2 for ZBHG1 treated under the same conditions. The ZBHE series corresponds to composites prepared by physical mixing (zeolite-to-biochar mass ratio 2:1), while the ZBHGE series corresponds to composites prepared via high-temperature co-pyrolysis (zeolite-to-biochar mass ratio 1:1).

The “molecular-level modification” was performed through chemical coordination of EDTA-2Na molecules with exposed metal sites on the composite surface, rather than by physical adsorption or simple impregnation. The reaction was carried out under mildly acidic conditions (pH of 4–6) to promote stable coordination, as confirmed by preliminary pH optimization. At lower pH, excessive protonation of carboxyl and amino groups reduces their availability for coordination, while at higher pH, Cu(II) and Zn(II) ions may undergo hydrolysis or precipitation, both of which decrease functionalization efficiency; thus, mildly acidic conditions provide a balance for effective chelation. The functionalized composites were subsequently applied in adsorption experiments to evaluate their performance toward Cu(II) and Zn(II) ions.

2.3. Adsorption kinetic models

Batch adsorption experiments were conducted to investigate the kinetics of Cu(II) and Zn(II) removal by the zeolite–biochar composites. Nineteen portions of 50 mL Cu(II) and Zn(II) solutions with predetermined concentrations were separately transferred into centrifuge tubes, followed by the addition of composites at different zeolite-to-biochar mass ratios. The mixtures were shaken at room temperature for 0–60 min, centrifuged at 4000 rpm for 10 min, and filtered through 0.22 μm membranes. The residual metal concentrations in the filtrates were measured using an inductively coupled plasma mass spectrometer (ICP-7510, Japan). All experiments were performed in triplicate.

The adsorption process was analyzed using first- and second-order adsorption kinetics as well as intraparticle diffusion kinetic models to identify the rate-limiting steps and clarify the governing mechanisms. The pseudo-first-order kinetic model is based on the assumption that the adsorption rate is proportional to the availability of unoccupied surface active sites. It is commonly applied to describe the initial stage of adsorption. The corresponding mathematical expression is given in Equation (1). The pseudo-second-order kinetic model is considered to be governed by chemical interactions, with the adsorption speed depending on the square of the remaining uptake potential (i.e., the gap between the uptake capacity at equilibrium and the quantity taken up at a given time). It is therefore regarded as more suitable for describing the overall adsorption behavior, and its linearized form is presented in Equation (2). The intraparticle diffusion approach is applied to analyze how adsorbates move within the pores of the adsorbent. Its potential role as the step controlling the overall rate is evaluated by examining the linearity and intercept of the fitted curve, as presented in Equation (3). A comparative analysis of the three kinetic models was carried out to thoroughly assess the material’s removal efficiency and the dominant mechanisms governing the adsorption process for the target contaminant.

Where Q_e (mg·g⁻¹) is the adsorption capacity at equilibrium, K₁ (min⁻¹) is the rate constant of the pseudo-first-order kinetic model, Q_t (mg·g⁻¹) is the amount of adsorbate adsorbed at time t, K₂ (g·mg⁻¹·min⁻¹) is the rate constant of the pseudo-second-order kinetic model, and K_D (mg·g⁻¹·min^−1/2) denotes the constant governing diffusion within the adsorbent pores.

2.4. Adsorption isotherm models

In this study, batch adsorption data were interpreted using the Langmuir and Freundlich isotherms to explore the adsorption behavior and fundamental mechanisms of the adsorbents for copper(Ⅱ) and zinc(Ⅱ). The Langmuir model assumes adsorption occurs in a single layer on a uniform surface, with identical binding sites and minimal interaction among adsorbed molecules. Its mathematical expression is provided in Equation (4). This model is often used to determine the adsorbent’s maximum uptake and to examine when the active binding sites reach saturation. In contrast, the Freundlich equation provides an experimental-based description for adsorption occurring in multiple layers on non-uniform surfaces. It reflects the adsorption affinity and intensity between the adsorbate and the adsorbent, and the corresponding equation is provided in Equation (5). By analyzing the goodness of fit for both models, it is possible to assess if the adsorption proceeds predominantly as monolayer or layered adsorption.

where C_e  (mg·L⁻¹) refers to the amount of adsorbate left dissolved in the solution once equilibrium is reached, Q_e (mg·g⁻¹) indicates the quantity of adsorbate absorbed by each gram of adsorbent when equilibrium is achieved,  Q_m (mg·g⁻¹) represents the theoretical maximum binding amount, K_L (L·mg⁻¹)  is the Langmuir constant related to the affinity of binding sites, K_Fis the Freundlich constant indicative of adsorption capacity, and 1/n is surface non-uniformity factor or observed strength parameter.

All kinetic and isotherm parameters were derived by fitting experimental data using Origin 2022 software with nonlinear regression, and the goodness of fit was evaluated via R² values. This approach ensures methodological transparency and reproducibility while avoiding unnecessary elaboration of standard theoretical formulas.

3.1. Material characterization

3.1.1. SEM Analysis

The microstructural morphologies of pristine zeolite, acid-treated zeolite, and biochar are displayed in Figures 1(a-c). Pristine zeolite exhibited rough and uneven surfaces, indicative of a well-developed network of micro- and mesopores. After HCl treatment, the zeolite surface appeared smoother with notable changes in pore structure, which is attributed to the removal of surface impurities or blockages and the exposure of additional active adsorption sites. Biochar retained a fibrous and layered morphology. In the zeolite–biochar composites (Figures 1d-g), zeolite particles were observed to be either attached to the biochar surface or embedded within the matrix, maintaining the fibrous structure of biochar. High-temperature treatment enhanced interfacial adhesion, producing a more compact structure with reduced interparticle voids, which facilitates the integration of the high adsorption capacity of zeolite with the porous biochar architecture [27].

Close

Figure 1.

SEM images of zeolite–biochar composite materials: (a) Z; (b) ZH; (c) BC; (d) ZB; (e) ZBG; (f) ZBH; (g) ZBHG.

Export to PPT

SEM images of EDTA-2Na-functionalized composites are shown in Figures 2(a-b). In unmodified composites, particle aggregation was observed on the biochar surface, likely due to carbon microcrystalline structures formed during pyrolysis or incomplete dispersion of zeolite particles during preparation. Following EDTA-2Na modification, particle stacking on the biochar surface was reduced, and the overall morphology became more uniformly rough. This change is consistent with chemical functionalization by EDTA-2Na, which can coordinate with metal ions on the biochar surface through its carboxyl and amino groups. Consequently, the surface chemical environment of the composites was optimized, leading to more homogeneous particle dispersion. Additional carboxyl and amino groups introduced by EDTA-2Na treatment are expected to enhance metal ion adsorption. The synergistic combination of biochar porosity and zeolite active sites contributes to improved overall adsorption performance.

Close

Figure 2.

SEM images of EDTA-2Na-modified zeolite–biochar composite environmental materials: (a) ZBHE; (b) ZBHGE.

Export to PPT

3.1.2. FT-IR analysis

FT-IR spectra of the zeolite–biochar composites before and after EDTA-2Na functionalization are shown in Figure 3. In the unmodified composites, stretching vibrations of silanol groups (Si–OH) appeared at 3445 cm⁻¹, while bending vibrations of Si/Al–O bonds were observed at 1654 cm⁻¹. Peaks at 1013 cm⁻¹ were assigned to Si–O–Si stretching, and those at 673 cm⁻¹ and 554 cm⁻¹ corresponded to Si/Al–O bending and extension [28].

Close

Figure 3.

FT-IR spectra of EDTA-2Na-modified zeolite–biochar composite materia: (a) without high-temperature modification; (b) after high-temperature modification.

Export to PPT

After EDTA-2Na modification, the band at 1640 cm⁻¹, corresponding to asymmetric stretching of carboxylate groups (–COO⁻) or aromatic C=C vibrations, was intensified, and the –OH stretching region around 3445 cm⁻¹ was slightly broadened. Minor shifts in Si–O–Si and Si–O–Al vibrations (1000–1100 cm⁻¹) were also observed. The main spectral changes confirm the incorporation of EDTA-derived functional groups, including the intensified 1640 cm⁻¹ band and the broadening of the –OH stretching region, indicating the successful introduction of carboxyl and hydroxyl groups onto the composite surface.

3.2. Adsorption performance analysis

Among the pretreated samples, HCl modification led to the most significant enhancement in surface reactivity and adsorption capacity, as confirmed by FT-IR and BET analyses (see Figures 4 and 5 and Table S1) in Supporting Information), and HCl modification increased the zeolite surface area to 29.55 m²·g⁻¹, nearly 15 times that of the unmodified sample. The acid dissolves channel impurities and reduces pore blockage, while small H⁺ ions penetrate the framework to enlarge pores, enhancing surface area and adsorption capacity for heavy metal ions [29]. Consequently, the ZBH2 and ZBHG1 composites were selected as the most promising precursors for subsequent EDTA-2Na functionalization (see Figure S1 and Tables S2 and S3 in Supporting Information). The adsorption performance of the resulting EDTA-2Na-modified zeolite–biochar composites was then systematically evaluated for Cu(II) and Zn(II) removal from aqueous solution. Experimental parameters, including adsorbent dosage, solution pH, and contact time, were optimized. The results indicated that an adsorbent dosage of 5 g·L⁻¹ (see Figure S2 in Supporting Information), a solution pH of 5 (see Figures S3 and S4 in Supporting Information), and a contact time of less than 60 min (see Figure S5 in Supporting Information) were optimal for achieving equilibrium adsorption.

Close

Figure 4.

(a) FT-IR and (b) XRD spectra of Z, Z-HCl, Z-KOH, and Z-NaCl.

Export to PPT

Close

Figure 5.

Infrared spectra of zeolite-biochar composites: (a) FTIR spectrum of the sample without acid modification; (b) FTIR spectrum of the sample after acid activation.

Export to PPT

The improved adsorption performance can be attributed to molecular-level modification, in which EDTA-2Na molecules chemically coordinate with exposed metal sites on the composite surface. This is considered “molecular-level” because the functional groups are incorporated through chemical bonds at the scale of individual molecules, rather than by physical adsorption or simple surface coating [30]. Thermodynamic analysis indicates that the adsorption involves both physical and chemical interactions: ΔH values between 21 and 80 kJ·mol⁻¹ (see Table S4 and S5) suggest contributions from chemisorption and physisorption, with chemisorption being enhanced by the functional groups introduced through molecular-level modification [31].

Under these conditions, the EDTA-2Na-functionalized composite ZBHE2 achieved maximum adsorption capacities of 65.67 mg·g⁻¹ for Cu(II) and 22.61 mg·g⁻¹ for Zn(II) (Figures 6a-b). The superior adsorption of Cu(II) over Zn(II) can be attributed to its smaller hydrated ionic radius and stronger coordination with surface carboxyl and hydroxyl groups, consistent with previous studies [32]. Comparatively, the unmodified composites showed lower adsorption capacities (e.g., ZBH2 adsorbed 59.43 mg·g⁻¹ Cu(II) and 21.31 mg·g⁻¹ Zn(II)), indicating that EDTA-2Na functionalization enhanced metal ion uptake.

Close

Figure 6.

Effects of different adsorbents on the adsorption performance of (a) copper(II) and (b) zinc(II). The bars represent the measured adsorption capacities under the specified experimental conditions.

Export to PPT

These results demonstrate that EDTA-2Na modification improves the adsorption performance of zeolite–biochar composites. The enhanced uptake is consistent with structural and surface property changes observed in SEM, FT-IR, and BET analyses, providing a mechanistic basis for the observed improvements.

3.3. Adsorption kinetics analysis

To elucidate the adsorption mechanism, the kinetic data of Cu(II) and Zn(II) uptake by ZBHE2 were fitted to pseudo-first-order, pseudo-second-order, and intraparticle diffusion models (Table 3, Figure 7). The experimental data were better described by the pseudo-second-order kinetic model than by the pseudo-first-order model, as indicated by higher correlation coefficients (0.998 for Cu(II) and 0.991 for Zn(II)) and calculated equilibrium capacities (Q_e,cal) that were closer to the experimental values (Q_e,exp). The higher correlation coefficient (R²) of the pseudo-second-order model indicates that the adsorption of Cu²⁺ and Zn²⁺ onto ZBHG1 is mainly controlled by a chemisorption process [33].

Close

Figure 7.

Linear fitting curves of kinetic models for the adsorption of copper(Ⅱ) and zinc(Ⅱ) by ZBHE2: for copper(Ⅱ): (a) pseudo-first-order kinetic model; (b) pseudo-second-order kinetic model; (c) intraparticle diffusion model; for zinc(Ⅱ): (d) pseudo-first-order kinetic model; (e) pseudo-second-order kinetic model; (f) intraparticle diffusion model.

Export to PPT

The higher K₂ values for Cu(II) compared with Zn(II) suggest stronger coordination between Cu(II) and the surface active sites, consistent with its smaller hydrated ionic radius and higher complex stability with EDTA groups. In contrast, Zn(II) exhibits slightly slower kinetics due to weaker surface interactions. These observations link the adsorption rate differences to the intrinsic properties of the ions and the functionalized adsorbent.

Intraparticle diffusion analysis revealed a two-stage process for both ions. The initial stage (Stage I) is dominated by rapid surface or film diffusion, as indicated by higher diffusion rate constants (K_D1 = 8.27 for Cu(II), 3.37 for Zn(II)), while the later stage (Stage II) shows slower intraparticle diffusion (K_D2 = 0.48 for Cu(II), 0.32 for Zn(II)), indicating that intraparticle diffusion gradually becomes rate-limiting. Non-zero intercepts in the plots suggest the influence of boundary layer effects and surface heterogeneity, highlighting that multiple mechanisms contribute to the overall uptake.

Overall, the kinetic analysis demonstrates that adsorption is governed by a combination of chemisorption, surface interactions, and intraparticle diffusion. Cu(II) shows faster adsorption than Zn(II), consistent with its stronger coordination and surface affinity, providing a mechanistic link between the kinetic parameters and the structural and chemical characteristics of the EDTA-2Na-modified zeolite–biochar composite.

3.4. Adsorption isotherm analysis

The adsorption isotherms of Cu(II) and Zn(II) on ZBHE2 were fitted using both Langmuir and Freundlich models (Figure 8, Table 4). While both models adequately described the adsorption behavior, a better correlation (R² ≈ 0.98) was obtained with the Freundlich model, implying that adsorption occurs on a non-uniform surface via multilayer coverage. The Freundlich constants (K_F and 1/n) further reveal that adsorption is favorable (1/n between 0 and 1) and that Cu(II) ions exhibit slightly stronger binding affinity to EDTA-functionalized sites than Zn(II), consistent with the surface functional groups identified by FT-IR and the pore structure observed from BET analysis.

Close

Figure 8.

Linear fitting plots of Langmuir and Freundlich isotherm models for copper(II) and zinc(II) adsorption by ZBHE2: (a) Langmuir and (b) Freundlich; for zinc(II): (c) Langmuir and (d) Freundlich.

Export to PPT

Kinetic and isotherm analyses suggest that both chemisorption and physisorption contribute to metal ion uptake. Multilayer adsorption and surface heterogeneity indicate that physisorption occurs at less accessible sites, while chemical interactions with surface functional groups also play an important role. The adsorbent’s internal volume is not uniformly accessible; sites near the surface reach saturation earlier, highlighting potential diffusion limitations under practical conditions. Additionally, the increase in maximum adsorption capacity (Q_m) with temperature indicates an endothermic process, which can enhance both ion–surface interactions and mobility within the pores, reflecting the combined contributions of chemical and physical adsorption.

While classical models (pseudo-second-order kinetics, Langmuir, and Freundlich isotherms) adequately describe the equilibrium behavior, the study did not apply more advanced or hybrid models—such as Temkin, Redlich–Peterson, or statistical physics-based models—which could better capture the combined effects of chemisorption, multilayer adsorption, and ion exchange. Discussing these models could provide a more comprehensive mechanistic understanding and may be considered in future research.

3.5. Adsorption performance and mechanistic discussion

The EDTA-2Na-modified zeolite–biochar composite exhibits maximum adsorption capacities of 65.67 mg·g⁻¹ for Cu(II) and 22.61 mg·g⁻¹ for Zn(II), which are comparable to or higher than values reported for similar EDTA-modified adsorbents. For example, EDTA-chitosan has been reported with Cu(II) adsorption capacities of 42.4 mg·g⁻¹ [34], and the EDTA-modified silica/carbon materials (EDCMS) reach approximately 44 mg·g⁻¹ [32]. This comparison indicates that the composite exhibits competitive adsorption performance, particularly considering cost-effectiveness and environmental stability.

FTIR and SEM analyses confirm the successful grafting of carboxyl and amino groups from EDTA-2Na onto the zeolite–biochar surface. BET results reveal increased pore accessibility despite a modest decrease in surface area, suggesting that chemical functionalization, rather than surface area alone, contributes to adsorption capacity. Pseudo-second-order kinetics and shifts in FTIR peaks after adsorption indicate that metal–ligand coordination occurs alongside other interactions, while enhanced surface exposure and roughness improve access to active sites. Following modification, particle aggregation decreased, likely due to chelation of residual surface metals or impurities by –COOH and –NH₂ groups, further optimizing the pore structure.

The modified composite shows higher uptake of Cu²⁺ than Zn²⁺, likely because of the smaller hydrated radius of Cu²⁺ and its stronger coordination with oxygen-containing functional groups. Adsorption behavior follows the Freundlich model, indicating multilayer uptake on heterogeneous surfaces, with KF peaking at 308 K [35]. EDTA-2Na was selected over H4EDTA or other chelating agents because its disodium salt form exhibits superior water solubility and readily dissociates into active carboxylate groups under mild conditions, enabling efficient grafting onto the zeolite–biochar surface without harsh pH adjustment or additional catalysts. This ensures reproducible and stable metal chelation during adsorption.

Although unmodified zeolite and biochar were not directly tested, previous reports indicate lower adsorption capacities for Cu(II) and Zn(II), as no additional chelating groups are present to facilitate metal binding. This provides a reasonable comparative baseline. The study used single-metal solutions; practical conditions such as coexisting ions, organic matter, or wastewater matrices were not investigated. Similarly, reusability and regeneration performance were not addressed in this work but will be systematically evaluated in future studies.