Experimental Study and DFT Calculation Analysis of Pb(II) and Cd(II) Adsorption from Water by MgCl₂‑Modified Kiwifruit Pomace Biochar

Abstract

MgCl₂-modified kiwifruit pomace biochar (MgBC) was synthesized and applied for Pb­(II) and Cd­(II) removal from water. The material exhibited a mesoporous structure (71.88 m²/g) and achieved high adsorption capacities of 604.70 mg/g for Pb­(II) and 304.98 mg/g for Cd­(II), with Langmuir maxima of 763.82 and 532.80 mg/g, respectively. In binary-metal systems, MgBC showed preferential Pb­(II) uptake, and Cd­(II) coexistence unexpectedly enhanced Pb­(II) adsorption. Density functional theory (DFT) and molecular dynamics (MD) simulations revealed three adsorption configurations: direct surface bonding (MgBC-Pb/Cd), hydroxyl-mediated interactions (MgBC-OPb/OCd) and defect-site occupation (MgBCPb/Cd). Adsorption energies followed MgBC-Pb (−2.93 kcal/mol) < MgBC-OPb (−2.57 kcal/mol) < MgBC-Cd (−1.09 kcal/mol) < MgBC-OCd (−0.08 kcal/mol) ≪ MgBCCd (298 kcal/mol) and MgBCPb (408 kcal/mol), which confirms chemisorption by direct and hydroxyl-assisted pathways as the dominant mechanisms with Pb­(II) being thermodynamically more favorable. Projected density of states (PDOS) analyses corroborated the adsorption capacity: MgBC-Pb > MgBC-OPb > MgBC-Cd > MgBC-OCd ≫ MgBCPb and MgBCCd. This work provides atomic-scale mechanistic insights and a sustainable strategy for converting agricultural waste into an effective biochar adsorbent for heavy metal removal.

1. Introduction

The rapid pace of industrialization and urbanization has led to increasingly severe heavy metal contamination in aquatic environments. , Among various pollutants, cadmium (Cd) and lead (Pb) are of particular concern due to their high toxicity, persistence and tendency to bioaccumulate. − These metals are primarily discharged from non-ferrous smelting, battery manufacturing, electroplating and chemical industries. In water, they exist predominantly as divalent cations (Cd²⁺ and Pb²⁺) and pose irreversible risks to ecosystems and human health. Cd and Pb cannot be metabolized or excreted once accumulated in the human body, which will lead to lifelong and irreversible health consequences. , Consequently, the treatment of Cd-laden and Pb-laden wastewater has attracted extensive research attention.

A variety of treatment technologies have been developed for heavy metal removal, including biological treatment, membrane separation, electrochemical methods, coagulation-flocculation and adsorption. − Among these, adsorption stands out due to its high efficiency, low cost, simple operation and environmental compatibility. , In recent years, biochar produced from biomass under oxygen-limited pyrolysis has emerged as a promising adsorbent for Cd and Pb removal owing to its renewable feedstock availability, cost-effectiveness and alignment with national carbon neutrality goals. − However, pristine biochar generally suffers from a limited specific surface area, scarce surface functional groups, moderate adsorption capacity and poor selectivity. To overcome these limitations, surface modification has been widely adopted to enhance both the selectivity and capacity of biochar toward target pollutants. −

Considerable efforts have been devoted to developing modified biochar for heavy metal remediation. For instance, loading Mg²⁺ Ca²⁺, K⁺ and Na⁺ into raw biochar can enhance ion exchange and mineralization performance, and this has been proven effective for improving Cd­(II) and Pb­(II) adsorption. In particular, Mg has become an effective and ideal modifier due to its abundant reserves and low toxicity. Most researchers have studied the preparation and adsorption properties of Mg-modified biochar from different biomass sources. , Wu et al. fabricated a three-dimensional manganese oxide-modified biochar porous hydrogel (MBCG), achieving maximum adsorption capacities of 84.76 mg/g for Cd and 70.90 mg/g for Pb. Qiao et al. modified chicken manure-derived biochar with potassium ferrate (K₂FeO₄) to obtain potassium ferrate-modified biochar (KFBC), which showed maximum adsorption capacities of 141.24 mg/g for Cd and 487.45 mg/g for Pb, significantly higher than those of the unmodified biochar. Li et al. prepared a Mg-modified rice straw biochar (MgBC) through chemical impregnation, achieving 92.68 mg/g for Cd and 128.06 mg/g for Pb. Zhang et al. reported that prolonging pyrolysis time transformed surface MgCl₂ into MgO on Mg-modified soybean straw biochar, which exhibited low adsorption energies (E _ads) for Cd­(II) and Pb­(II), thereby significantly enhancing both sorption capacity and selectivity.

Despite these reports have systematically summarized the global research trends in biochar modification, , established a ternary design framework (pore-surface-electronic properties) for next-generation biochar materials, and provided in-depth insights into the synthesis, characterization and mechanistic understanding of biochar-based adsorbents for heavy metal remediation. , The application of kiwifruit pomace as a precursor for Mg-modified biochar remain largely unexplored. Additionally, the atomic-scale mechanistic understanding of selective Pb/Cd adsorption on MgO surfaces is still lacking.

Moreover, while previous studies have attempted to elucidate the adsorption mechanisms of heavy metals onto modified biochar using conventional characterization techniques and empirical isotherm models, ,, the fundamental origin of selective affinity remains poorly understood, such as why Mg-modified biochar often exhibit higher affinity for Pb over Cd. Theoretical approaches such as density functional theory (DFT) and molecular dynamics (MD) simulations offer powerful tools to resolve this gap. − These methods can probe electronic-level interactions, including charge transfer, orbital hybridization, and changes in structural and electronic properties during adsorption, providing robust evidence to elucidate adsorbent-adsorbate interactions at the atomic scale.

To date, few studies have reported the preparation of Mg-modified biochar from kiwifruit residue, and its selective adsorption mechanism for Pb/Cd has not been revealed through DFT calculations. In this work, we selected kiwifruit pomace as the biomass precursor and magnesium chloride (MgCl₂) as the modifying agent to prepare MgCl₂-modified biochar (MgBC) by slow pyrolysis at 500 °C. The physicochemical properties of MgBC were systematically characterized using scanning electron microscopy (SEM), Brunauer–Emmett–Teller (BET), X-ray diffraction (XRD), Fourier-transform infrared (FTIR) and X-ray photoelectron spectroscopy (XPS). Batch adsorption experiments were conducted to evaluate its performance for Cd­(II) and Pb­(II) removal from simulated wastewater, and the adsorption behavior was analyzed using isotherm, thermodynamic, kinetic, and intraparticle diffusion models. Furthermore, DFT calculations and MD simulations were employed to reveal the electronic-scale mechanisms, including adsorption energy, highest occupied molecular orbital/lowest-unoccupied molecular orbital (HOMO/LUMO) orbital distributions and projected density of states (PDOS), which provides a comprehensive understanding of the interfacial interactions between MgBC and heavy metal ions.

2. Materials and Methods

2.1. Materials

Kiwifruit pomace, the solid residue generated from whole-fruit wine fermentation, was obtained from Liupanshui Liangdu Kiwifruit Industry Co., Ltd. The pomace was dried to constant weight in an electric thermostatic blast drying oven at 80 °C, crushed and sieved through a 60-mesh screen to obtain a fine powder, which was then stored in sealed bags for subsequent use.

All chemical reagents used in this work, including cadmium chloride, lead nitrate, magnesium chloride, hydrochloric acid and sodium hydroxide, were of analytical reagent grade. Deionized water was used throughout all experiments. The main instruments employed included an iCE 3500 atomic absorption spectrophotometer, a ZHP-250E intelligent constant-temperature oscillating incubator, a TE-0912 box-type resistance furnace, a ME 104 electronic balance, a PHS-3E pH meter, a 101 electric thermostatic blast drying oven, a Sigma 300 scanning electron microscope (SEM), a Micromeritics ASAP 2460 fully automatic specific surface area and porosity analyzer, a DX-2500 X-ray diffractometer (XRD), an INVENIO-S Fourier transform infrared (FTIR) spectrometer, a Thermo Scientific K-Alpha X-ray photoelectron spectrometer (XPS) and a STA449 F5 simultaneous thermal analyzer (NETZSCH, Germany).

2.2. Preparation of Biochar and Modified Biochar Material

The pristine biochar (BC) was prepared by pyrolyzing the pretreated kiwifruit pomace. Briefly, the pomace powder was placed in a porcelain crucible, compacted, wrapped in aluminum foil, and transferred to a muffle furnace. The temperature was ramped to 500 °C at a heating rate of 8 °C/min and held for 120 min. After natural cooling to room temperature, the resulting biochar was ground and sieved through a 100-mesh screen to obtain BC, which was stored in a sealed bag inside a desiccator.

For the preparation of magnesium-modified biochar (MgBC), 20 g of kiwifruit pomace powder was added to 200 mL of 1 mol/L MgCl₂ solution. The mixture was magnetically stirred for 60 min and then dried at 60 °C. The Mg-impregnated pomace was subsequently pyrolyzed under the same conditions as described for BC. After cooling to room temperature, the modified biochar was thoroughly rinsed with ultrapure water until the filtrate reached neutral pH, and then collected by vacuum filtration. The product was dried at 105 °C to constant weight, ground, and sieved through a 100-mesh screen to obtain MgBC. Both BC and MgBC samples were stored in sealed bags in a desiccator for further characterization and use.

The synthesis conditions of MgBC (500 °C, 1.0 mol/L MgCl₂) were selected based on literature precedents. Pyrolysis temperatures of 400–600 °C are generally optimal for biochar production, as lower temperatures cause incomplete carbonization. While higher temperatures may induce pore collapse and metal oxide sintering. Similarly, MgCl₂ concentrations of 0.5–2.0 mol/L are commonly used, with 1.0 mol/L being the most reported for balancing Mg loading and porous structure preservation. Our following content of characterization results (BET: 71.88 m²/g; Mg loading confirmed by XRD, FTIR and XPS) confirm that the selected conditions are appropriate for producing a high-performance adsorbent.

2.3. Adsorption Experiment

Batch adsorption experiments were conducted in 100 mL Erlenmeyer flasks by mixing MgBC with Cd­(II) and Pb­(II) solutions. The mixtures were shaken in an intelligent constant-temperature oscillating incubator at 180 rpm. The effects of various experimental parameters on adsorption performance were systematically investigated, including: Adsorbent type (BC vs MgBC), MgBC dosage (0.010–0.070 g), Initial solution pH (2–7), Adsorption temperature (15–35 °C), Initial Cd­(II) concentration (100–500 mg/L) and Pb­(II) concentration (200–600 mg/L), contact time (5–1440 min), competitive adsorption in mixed Cd­(II)/Pb­(II) solutions at varying concentrations. Unless otherwise specified, the initial concentrations were 200 mg/L for Cd­(II) and 400 mg/L for Pb­(II).

After the adsorption experiments, the mixtures were allowed to stand. The supernatants were filtered through 0.45 μm polyether sulfone (PES) membrane filters. The concentrations of Cd­(II) and Pb­(II) were determined using an atomic absorption spectrophotometer. The removal efficiency was calculated using eq (), and the adsorption capacity was calculated using eq (). ,

$$W=\frac{(C_0-C_X)}{C_0}\times 100\%$$ 1

$$Q=\frac{(C_0-C_X)\times \mathrm{V}}{M}$$ 2

where W is the removal efficiency of Cd­(II) and Pb­(II) (%); C ₀ is the initial concentration of Cd­(II) or Pb­(II) before adsorption (mg/L); C _x is the concentration of Cd­(II) or Pb­(II) after adsorption (mg/L); Q is the adsorption capacity of Cd­(II) or Pb­(II) (mg/g); M is the mass of the adsorbent (g); V is the volume of the solution (L).

2.4. DFT Calculation MD Simulations

To investigate the adsorption mechanism in detail, we performed theoretical modeling and DFT calculations for the adsorption of Pb­(II) and Cd­(II) ions onto MgBC using the Gaussian 09 software package. Three small MgBC-(Pb/Cd) models were constructed to thoroughly elucidate the interfacial properties among biochar, MgO clusters and Pb­(II)/Cd­(II). The HOMO/LUMO distributions and projected density of states (PDOS) of MgBC-(Pb/Cd) models were theoretical calculated and investigated. As well as MD simulation of the interfacial adsorption were predicted at the Materials Studio (MS) package. Single Pb­(II) or Cd­(II) ions were adsorbed onto the MgO (100) 3 × 3 × 1 crystal surface in the three different configurations described above and optimized. The three optimized (Pb/Cd)/MgO systems were then subjected to MD simulations to obtain reasonable relative configurations and minimum adsorption energies (E _ads).

3. Results and Discussion

3.1. Biochar Characterization

3.1.1. SEM Analysis

The surface morphologies of raw kiwifruit pomace, BC, MgBC, and MgBC after Cd and Pb adsorption were examined by scanning electron microscopy (SEM), and the images are presented in Figure . After pyrolysis, the BC (Figure (a)) began to develop some pores, although they remained relatively sparse and smooth-walled. In contrast, MgBC (Figure (b)) displayed an abundant, honeycomb-like porous structure with significantly increased surface roughness and irregularity. This morphological evolution suggests that MgCl₂ modification plays a critical role in pore development. The enhancement is likely attributable to two factors: (i) Mg²⁺ promotes the catalytic decomposition of cellulose and lignin during pyrolysis, acting as a pore-forming agent that facilitates surface dehydration and carbonization. (ii) the decomposition of MgCl₂ releases volatile species that further etch the carbon matrix, creating additional pores. These structural improvements substantially increase the specific surface area and expose more active sites for subsequent adsorption. After Cd and Pb adsorption (Figure (c,d)) irregular granular precipitates were observed on the MgBC surface, which were identified as metal oxides and hydroxides, indicating successful metal uptake.

1.

(a–d) SEM images of original kiwi fruit pomace, BC, MgBC, and MgBC after adsorption of Cd and Pb.

The Energy Dispersive Spectroscopy (EDS) results shown in Figure (a,b), BC was mainly composed of C, O and K elements, along with small amounts of N and Ca. After MgCl₂ modification, MgBC was primarily composed of C, O and Mg elements, with a decrease in C content and an increase in O content, accompanied by the appearance of Mg. These results indicate that MgCl₂ was successfully loaded onto MgBC.

2.

(a–d) (a, b) the EDS spectra of BC and MgBC. (c) XRD patterns of BC and MgBC. (d) FTIR spectra of BC, MgBC, and MgBC after adsorption.

3.1.2. BET Analysis

The textural properties of BC and MgBC are summarized in Table . Compared with BC, MgBC exhibited a 27.98-fold increase in specific surface area, a 108.31-fold increase in total pore volume, and a 3.87-fold increase in average pore diameter. This dramatic improvement is attributed to the decomposition of MgCl₂ during pyrolysis, which releases corrosive species that etch the biochar skeleton. While simultaneously generating MgO nanoparticles that are embedded onto the carbon surface, thereby creating a highly porous architecture. The pores were predominantly mesoporous.

1. Specific Surface Area and Porosity of Biochar.

biochar	BET surface area (m2/g)	total pore volume (cm3/g)	average pore diameter (nm)
BC	2.5693	0.0026	4.0495
MgBC	71.8775	0.2816	15.6727

The N₂ adsorption-desorption isotherms (Figure S1) further corroborate these findings. The BC isotherm showed a gentle rise without a distinct capillary condensation step, indicating a poorly developed pore structure. In contrast, the MgBC isotherm exhibited a gradual increase at low relative pressures (P/P ₀ < 0.1), a moderate increase in the medium-to-high pressure region (P/P ₀ ≈ 0.4-0.8) and a sharp upturn near saturation (P/P ₀ > 0.8), characteristic of a Type IV isotherm with a distinct H₃ hysteresis loop. This behavior is typical of mesoporous materials and reflects a combination of monolayer adsorption and capillary condensation. The pore size distribution further confirmed the presence of abundant and irregular mesopores, with the majority centered around 3.87 nm. Given that the atomic radii of Cd and Pb are 0.154 and 0.175 nm, respectively, this mesoporous structure is geometrically favorable for the diffusion and accommodation of Cd­(II) and Pb­(II) ions.

3.1.3. XRD Analysis

The crystalline structures of BC and MgBC were analyzed by X-ray diffraction (XRD), and the patterns are shown in Figure (c). BC exhibited a broad diffraction peak at 2θ = 20°–30°, characteristic of amorphous carbon derived from the pyrolysis of cellulose, indicating the formation of a stable, partially graphitized carbon structure. For MgBC, a series of additional diffraction peaks corresponding to Mg-containing phases were observed, confirming the successful loading of magnesium species onto the biochar. These peaks were primarily assigned to crystalline MgO and Mg­(OH)₂, consistent with the decomposition and subsequent hydration of MgCl₂ during pyrolysis and cooling.

3.1.4. FTIR Analysis

The surface functional groups of BC, MgBC, and MgBC after adsorption were characterized by Fourier transform infrared (FTIR) spectroscopy and the spectra are presented in Figure (d). Several characteristic bands were observed: a broad peak at the 3440 cm^‑ attributed to O–H stretching vibrations, a peak at 1606 cm^–1 representing CC skeletal vibrations and CO stretching, a peak at 1381 cm^–1 attributed to C–H bending vibrations, a peak at 2923 cm^–1 assigned to asymmetric C–H stretching in aliphatic methylene groups. Compared with BC, a weak band at 3696 cm^–1 corresponding to free hydroxyl groups and a distinct band at 453 cm^–1 attributed to Mg–O metal–oxygen vibrations appeared in MgBC, further confirming the successful loading of MgO. This is consistent with the XRD results.

After Cd and Pb adsorption, the intensity of the O–H peak at 3440 cm^–1 decreased significantly. It indicated that hydroxyl and carbonyl groups participated in the coordination and binding of metal ions, serving as primary active sites for heavy metal uptake. Additionally, the Mg–O peak exhibited a slight shift after adsorption, suggesting that MgO moieties may interact with cationic heavy metals by electrostatic attraction, or that Mg²⁺ may undergo ion exchange with Cd­(II) and Pb­(II). These observations highlight the dual role of oxygen-containing functional groups and MgO sites in the adsorption process.

3.1.5. XPS Analysis

The surface elemental composition and chemical states were investigated by X-ray photoelectron spectroscopy (XPS), and the survey spectra are shown in Figure (a). The XPS survey spectra revealed that BC was predominantly composed of C and O elements. Upon MgCl₂ modification, a distinct Mg peak emerged in MgBC, confirming the successful loading of magnesium species. After Cd and Pb adsorption, the Mg 1s peak intensity of MgBC decreased markedly, providing direct evidence for the involvement of Mg²⁺ in heavy metal adsorption via an ion exchange mechanism. Meanwhile, characteristic Cd 3d peaks appeared in the MgBC sample following Cd adsorption, Pb 4f peaks appeared following Pb adsorption, and both Cd and Pb peaks were detected in the co-adsorption system. These XPS observations are in good agreement with the experimental results.

3.

XPS spectra of BC, MgBC and MgBC after adsorption of Cd and Pb.

The Mg 1s XPS spectra (Figure (b)) revealed that a distinct Mg 1s peak emerged in MgBC at a binding energy of 1304.78 eV, which was absent in pristine BC. This observation indicates that Mg was successfully loaded onto the biochar surface primarily as MgO, which subsequently underwent hydration to form Mg­(OH)₂ species. The C 1s spectra (Figure (c)) showed that the π–π* satellite peak disappeared after Cd and Pb adsorption, suggesting that the aromatic π-electron system of the biochar acted as an electron donor (π-donor) during the adsorption process, facilitating cation-π interactions.

The high-resolution Cd 3d and Pb 4f spectra (Figure (d)) provided further insights into the adsorption behavior. In the co-adsorption system, the Cd 3d_5/2 and Cd 3d_3/2 peaks shifted to 406.26 and 413.14 eV, respectively, which corresponds to a binding energy shift of approximately 0.37 eV relative to the single-metal system. Similarly, the Pb 4f_5/2 and Pb 4f_3/2 peaks shifted to 139.32 and 144.16 eV, with a shift of about 0.5 eV. These shifts suggest changes in the local chemical environment and electron density of the adsorbed metal ions, likely arising from competitive interactions between Cd and Pb at surface binding sites or the formation of stronger chemical bonds with the biochar surface. The more pronounced shift for Pb indicates a stronger interaction with MgBC, consistent with its higher adsorption affinity observed in the binary-metal experiments.

3.1.6. Thermogravimetric Analysis

Thermogravimetric analysis (Figure S2) revealed that both BC and MgBC underwent marginal mass loss below 150 °C, attributable to the desorption of physisorbed moisture and volatile impurities. Above this temperature, the thermal degradation profiles diverged markedly. BC exhibited superior thermal stability, characterized by a gradual and continuous mass decline up to 800 °C, indicative of a robust carbonaceous framework undergoing only sluggish oxidative decomposition. Conversely, MgBC displayed substantially inferior thermal stability, with pronounced mass loss throughout the heating regime. In the low-temperature interval (150–350 °C), a conspicuous weight-loss shoulder emerged at ∼330 °C, reducing the residual mass to ∼83%, which is ascribed to the dehydroxylation of Mg­(OH)₂ and the volatilization of labile organic constituents. In the high-temperature region (350-800 °C), the material sustained rapid mass depletion, ultimately retaining only ∼63% at 800 °C. These observations suggest that the incorporation of magnesium augmented the volatile fraction at elevated temperatures, implicating extensive chemical restructuring during pyrolysis that culminated in a markedly diminished yield of carbonaceous residue.

3.2. Adsorption of Cd­(II) and Pb­(II) by Kiwifruit Pomace-Derived Biochar

3.2.1. Comparison of BC and MgBC Adsorption Performance

The adsorption capacities of pristine BC and MgBC toward Cd­(II) and Pb­(II) were compared under identical conditions, and the results are presented in Figure . BC exhibited limited adsorption capacities of only 15.08 mg/g for Cd and 33.40 mg/g for Pb. In contrast, MgBC achieved remarkably higher capacities of 303.95 mg/g for Cd and 604.56 mg/g for Pb, representing 20.16- and 18.10-fold increases, respectively. This dramatic enhancement can be attributed to several synergistic factors. First, the pristine BC possesses a poorly developed pore structure and scarce surface-active sites, which severely restrict its adsorption performance. After MgCl₂ modification, MgBC develops a loose, highly porous architecture with abundant mesopores, which not only provide large surface area but also serve as fast diffusion channels for metal ions, allowing them to rapidly access internal active sites. In contrast to micropores that are susceptible to pore blockage by hydroxide precipitates, mesopores offer sufficient space to accommodate such precipitates and maintain unobstructed ion transport. Furthermore, the loading of MgO and Mg­(OH)₂ on to biochar surface introduces additional adsorption pathways, including ion exchange between Mg²⁺ and heavy metal cations, as well as surface complexation. These combined effects endow MgBC with superior adsorption affinity and capacity. Owing to its outstanding performance, MgBC was selected for all subsequent adsorption experiments.

4.

Effects of BC and MgBC on the adsorption of Cd and Pb.

3.2.2. Effect of Adsorbent Dosage

The influence of MgBC dosage on Cd­(II) and Pb­(II) removal was investigated over the range of 0.010–0.070 g, and the results are shown in Figure S3. As the dosage increased, the removal efficiencies of both metals exhibited a rising trend, reaching a plateau when the dosage exceeded 0.030 g. In contrast, the adsorption capacity (mg/g) showed a continuous decline with increasing dosage. This opposing behavior can be explained as follows: at higher dosages, more adsorbent particles are available, providing a greater total number of adsorption sites and thus enhanced contact opportunities between metal ions and the biochar surface, which promotes removal efficiency. However, as the dosage increases further, the amount of metal ions adsorbed per unit mass of adsorbent decreases because a significant portion of the sites remains unsaturated, leading to a reduced adsorption capacity. To achieve a balance between high removal efficiency and material economy, an optimal MgBC dosage of 0.030 g was selected for subsequent experiments. Under this condition, the removal efficiency and adsorption capacity for Cd­(II) reached 91.35% and 304.48 mg/g, respectively, while those for Pb­(II) were 90.71% and 604.70 mg/g, respectively.

3.2.3. Effect of Initial pH

The effect of initial solution pH on the adsorption of Cd­(II) and Pb­(II) by MgBC was examined over the range of pH 2–7, and the results are presented in Figure S4. The removal efficiency and adsorption capacity of both metals increased progressively with rising pH. It should be noted that Pb­(OH)₂ precipitation occurred when the pH exceeded 7. Therefore, the pH range was limited to ≤7 to avoid confounding effects from precipitation.

The pH-dependent adsorption behavior (Figure S4) provides insights into the surface charge of MgBC. Removal efficiency was low at pH < 3, increased at pH 3–5, and plateaued at pH 5–7. Surface functional groups are protonated at acidic pH, causing electrostatic repulsion of metal cations. Deprotonation occurs as pH increases, creating negatively charged sites that attract cations. The inflection points at pH ∼ 4–5 likely corresponds to the pHZPC of MgBC. These findings confirm that electrostatic attraction contributes to adsorption alongside surface complexation and ion exchange. Notably, the equilibrium pH of the Cd­(II) and Pb­(II) solutions after biochar addition were measured to be 5.88 and 5.07, respectively, both falling within the favorable pH range. Considering both adsorption performance and economic practicality, no pH adjustment was applied in the subsequent adsorption experiments.

3.2.4. Effect of Initial Concentration on Cd and Pb Adsorption

As shown in Figure S5, with increasing initial concentrations of Cd²⁺ and Pb²⁺, the removal efficiencies of both metal ions exhibited a continuous decreasing trend, while the unit adsorption capacity increased progressively. At lower initial concentrations, abundant active sites were available on the adsorbent surface, allowing sufficient capture of heavy metal ions and resulting in higher removal efficiencies. As the number of metal ions in solution increased, the active sites were gradually occupied, leading to intensified site competition and a sustained decline in removal efficiency.

Simultaneously, the increased concentration gradient between the solid and liquid phases enhanced the mass transfer driving force, promoting greater binding of metal ions to the adsorbent surface and continuously increasing the unit adsorption capacity. The growth rate of adsorption capacity slowed in the later stage of the curve, indicating that the adsorption gradually approached saturation. Based on these results, Cd at 200 mg/L and Pb at 400 mg/L exhibited relatively high removal efficiencies and adsorption capacities, and were therefore selected as the initial concentrations for subsequent experiments.

To clearly demonstrate the advantages of MgBC, a comprehensive comparison was made with previously reported biochar adsorbents used for Pb­(II) and Cd­(II) removal (Table S1). Several key observations can be discussion, including superior adsorption capacity, feedstock novelty and mild operational conditions. Firstly, for Superior adsorption capacity, our MgBC exhibits Pb­(II) (604.70 mg/g) and Cd­(II) (304.98 mg/g) adsorption capacities that are significantly higher than those of most reported adsorbents. For example, the Pb­(II) adsorption capacity of MgBC exceeds that of magnesium-rich nitrogen-self-doped keratin-based biochar (271.55 mg/g). The Cd­(II) adsorption capacity of MgBC (304.98 mg/g) is also substantially higher than that of chitosan-modified kiwi branch biochar (126.58 mg/g). While pristine biochars generally exhibit low adsorption capacities (e.g., rice husk biochar: 38.69 mg/g for Pb, 9.22 mg/g for Cd; corn straw biochar: 81.47 mg/g for Pb, 17.59 mg/g for Cd), modification significantly enhances the adsorption performance, as demonstrated by our MgBC. Further, our feedstock (kiwifruit pomace) novelty. To the best of our knowledge, kiwifruit pomace has not been previously reported as a feedstock for MgCl₂-modified biochar in heavy metal removal, highlighting the novelty and sustainable waste valorization aspect of our work. In addition, for the mild operational conditions, MgBC achieves its high adsorption capacities under mild conditions (25 °C, almost no need pH adjustment), which are more practical for real-world applications compared with some studies that require strict pH control or higher temperatures. Moreover, for mechanistic depth, unlike most previous studies that rely solely on empirical isotherm and kinetic models, our work additionally employs DFT calculations and MD simulations to reveal the atomic-scale adsorption mechanism, providing a deeper understanding of the selective adsorption behavior. Collectively, these comparisons clearly demonstrate that MgBC derived from kiwifruit pomace represents a promising and competitive adsorbent for heavy metal removal, offering both high performance and practical applicability.

3.3. Adsorption Isotherms and Thermodynamic Analysis

To elucidate the adsorption behavior and the underlying interactions between MgBC and metal ions, the equilibrium adsorption data for Cd­(II) and Pb­(II) at 15, 25, and 35 °C were fitted using the Langmuir (eq ), Freundlich (eq ), Sips (eq ) and Redlich-Peterson (eq ) isotherm models. The fitting curves are presented in Figures S6–S7 and the corresponding parameters are summarized in Tables S2–S3.

$$Q_{\mathrm{e}}=\frac{Q_{\max }{k}_{\mathrm{L}}{C}_{\mathrm{e}}}{1+k_{\mathrm{L}}{C}_{\mathrm{e}}}$$ 3

$$Q_{\mathrm{e}}=k_{\mathrm{F}}{C}_{\mathrm{e}}^{1/n}$$ 4

$$Q_{\mathrm{e}}=\frac{Q_{\max }{(k_{\mathrm{S}}{C}_{\mathrm{e}})}^{\mathrm{m}}}{1+{(k_{\mathrm{S}}{C}_{\mathrm{e}})}^{\mathrm{m}}}$$ 5

$$Q_{\mathrm{e}}=\frac{k_{\mathrm{RP}}{C}_e}{1+{\alpha }_{\mathrm{RP}}{C}_{\mathrm{e}}^{\beta }}$$ 6

where Q _e (mg/g) is the equilibrium adsorption capacity, Q _max (mg/g) is the maximum adsorption capacity, C _e (mg/L) is the equilibrium metal concentration, k _L is the Langmuir constant related to binding affinity, k _F is the Freundlich adsorption constant. 1/n is a constant reflecting the adsorption affinity, it is generally considered that the adsorption is unfavorable when 1/n > 1. k _S is the Sips equilibrium constant representing adsorption affinity (L/mg)/^1/n ; m is the Sips surface heterogeneity index (dimensionless); k _RP is the Redlich-Peterson constant (L/g); α_RP is the Redlich-Peterson constant (L/mg)^1/β; and β is the Redlich–Peterson heterogeneity exponent (dimensionless).

As shown in Table S2, both models provided satisfactory fits to the experimental data, with all correlation coefficients (R ²) exceeding 0.90. Notably, the Langmuir model yielded higher R ² values than the Freundlich model for both Cd­(II) and Pb­(II) at all tested temperatures, indicating that the adsorption process is better described by monolayer coverage on a homogeneous surface. This suggests that the adsorption of Cd­(II) and Pb­(II) onto MgBC occurs primarily through specific binding sites with uniform energy distribution, rather than multilayer adsorption on heterogeneous surfaces. The 1/n values derived from the Freundlich model were all less than 1 for both metals across the temperature range, further confirming the favorable nature of the adsorption process. The theoretical maximum adsorption capacities (Q _max) of MgBC for Cd­(II) and Pb­(II) at 25 °C were determined to be 532.80 mg/g and 763.82 mg/g, respectively, demonstrating the superior adsorption potential of MgBC, particularly for Pb­(II).

The fitting results presented in Figure S7 and Table S3 demonstrate that both the Sips and Redlich-Peterson models provide adequate descriptions of the adsorption process. For Cd, the two models exhibited comparable goodness of fit at 15 °C. However, the Redlich-Peterson model showed superior performance with increasing temperature. The Redlich-Peterson model consistently outperformed the Sips model across the entire temperature range for Pb, indicating that the Redlich-Peterson isotherm is more appropriate for characterizing this adsorption system. The Sips model further revealed that the saturated adsorption capacity for Pb exceeded that for Cd, and the theoretical capacities for both metals increased with elevating temperature, corroborating the endothermic nature of the adsorption process. The β exponent obtained from the Redlich-Peterson isotherm model indicated a heterogeneous energy distribution of adsorption sites for both Cd and Pb at low temperatures. As the temperature rose, the β value for Cd approached unity, signifying a transition toward monolayer homogeneous adsorption. Conversely, the Pb²⁺ system yielded β > 1, suggesting that the adsorption mechanism for Pb is more intricate and deviates from simple monolayer adsorption.

To further investigate the thermodynamic driving forces of the adsorption process, the Gibbs free energy change (ΔG), enthalpy change (ΔH), and entropy change (ΔS) were calculated using the following eqs (), () and ()­

$$\mathrm{\Delta }G=-\mathit{RT}\ln K_{\mathrm{d}}$$ 7

$$\ln K_{\mathrm{d}}=\frac{\mathrm{\Delta }S}{\mathrm{R}}-\frac{\mathrm{\Delta }H}{\mathit{RT}}$$ 8

$$K_{\mathrm{d}}=\frac{Q_{\mathrm{e}}}{C_{\mathrm{e}}}$$ 9

where R (8.314 J/mol·K) is the universal gas constant, T (K) is the absolute temperature, K _d (L/g) is the equilibrium constant.

The calculated thermodynamic parameters are summarized in Table S4. The negative ΔG values at 288, 298, and 308 K for both Cd­(II) and Pb­(II) confirm that the adsorption processes are spontaneous. Moreover, the increasingly negative ΔG values with rising temperature indicate that higher temperatures favor the adsorption, consistent with the endothermic nature of the process. The positive ΔH values (38.28 kJ/mol for Cd and 14.62 kJ/mol for Pb) provide direct evidence that the adsorption reactions are endothermic, which aligns with the enhanced adsorption performance observed at elevated temperatures. The positive ΔS values (0.1529 kJ/(mol·K) for Cd and 0.0709 kJ/(mol·K) for Pb) suggest an increase in randomness at the solid-liquid interface during adsorption, which may be attributed to the displacement of water molecules from the biochar surface by metal ions, leading to a net gain in degrees of freedom. Notably, the ΔH value for Pb is considerably lower than that for Cd, indicating that Pb adsorption is less energy-demanding and thus thermodynamically more favorable, which is consistent with the higher affinity and capacity of MgBC toward Pb­(II). Overall, the adsorption of Cd­(II) and Pb­(II) onto MgBC is a spontaneous, endothermic, and entropy-driven process.

3.4. Adsorption Kinetics

To investigate the rate-controlling steps and the underlying adsorption mechanism, the experimental kinetic data were analyzed using three widely applied models: the pseudo-first-order (eq ), pseudo-second-order (eq ) and intraparticle diffusion (eq ) kinetic models. The fitting results are presented in Figure S8, and the corresponding parameters are summarized in Tables S5 and S6.

$$Q_t=Q_{\mathrm{e}}(1-{\mathrm{e}}^{-k_{1}t})$$ 10

$$Q_t=\frac{k_2{Q}_{\mathrm{e}}^{2}t}{1+k_2{Q}_{\mathrm{e}}t}$$ 11

$$Q_t=k_{\mathrm{id}}{t}^{0.5}+A$$ 12

where Q_t (mg/g) and Q _e (mg/g) are the adsorption capacities at time t and at equilibrium, respectively; k ₁ (min^–1) and k ₂ (g/(mg·min)) are the rate constants of the pseudo-first-order and pseudo-second-order models, respectively; k _id (mg/(g·min^0.5)) is the intraparticle diffusion rate constant; A is a constant related to the boundary layer thickness.

As shown in Table S5, the pseudo-second-order model yielded significantly higher correlation coefficients (R ² = 0.9902 for Cd and 0.9917 for Pb) compared with the pseudo-first-order model (R ² = 0.9478 for Cd and 0.9606 for Pb), indicating that the pseudo-second-order model provides a more accurate description of the adsorption kinetics for both metals. Furthermore, the equilibrium adsorption capacities (Q _e) calculated from the pseudo-second-order model were 304.87 mg/g for Cd and 622.92 mg/g for Pb, which are in excellent agreement with the experimental values of 311.61 mg/g and 620.48 mg/g, respectively. The strong consistency between the calculated and experimental Q _e values, together with the superior R ² values, suggests that the adsorption of Cd­(II) and Pb­(II) onto MgBC is predominantly governed by chemisorption, involving valence forces through electron sharing or exchange between the adsorbent surface and metal ions.

To further identify the rate-limiting steps, the intraparticle diffusion model was applied. As shown in Figure S8­(b), the plots of Q_t versus t ^0.5 for both Cd and Pb exhibited multilinearity, indicating that the adsorption process proceeds through three distinct stages: the first stage (0–30 min), the second stage (30–210 min), and the third stage (210–1440 min). Notably, none of the linear segments passed through the origin, suggesting that intraparticle diffusion is not the sole rate-controlling step and that multiple mechanisms operate simultaneously.

The kinetic parameters for each stage are summarized in Table S6. The first stage, characterized by the highest rate constant (k _id1), corresponds to external surface diffusion, during which metal ions are rapidly transported from the bulk solution to the outer surface of MgBC. This stage is fast and completes within a short time, as reflected by the steep slope. The second stage (k _id2) represents intraparticle diffusion, during which metal ions gradually penetrate the interior of the biochar through its mesoporous structure. The adsorption rate decreases in this stage because the available active sites are progressively occupied, leading to a reduced concentration gradient. The third stage (k _id3) is the micropore diffusion or equilibrium stage, where the remaining active sites are scarce and the diffusion of metal ions becomes extremely slow, eventually reaching a plateau. The decreasing trend of k _id (k _id1 > k _id2 > k _id3), along with the increasing trend of the boundary layer constant A (C ₃ > C ₂ > C ₁), further confirms that the adsorption process transitions from surface diffusion to intraparticle diffusion and finally to equilibrium. Overall, these results demonstrate that the adsorption of Cd­(II) and Pb­(II) onto MgBC is governed by the combined effects of external surface diffusion and intraparticle diffusion, with chemisorption serving as the predominant mechanism.

3.5. Competitive Adsorption in Mixed Cd­(II) and Pb­(II) Systems

To evaluate the selectivity and competitive behavior of MgBC toward coexisting heavy metals, batch adsorption experiments were conducted in mixed solutions containing varying concentrations of Cd­(II) and Pb­(II). The results are summarized in Table S7.

In all mixed systems, the adsorption capacity of MgBC for Pb­(II) was consistently and substantially higher than that for Cd­(II), indicating a pronounced preferential affinity of MgBC toward Pb­(II). This selectivity may be attributed to the differences in ionic radius, charge density, and hydrolysis behavior between the two metal ions. Pb²⁺ possesses a larger ionic radius and higher polarizability, which facilitates stronger interactions with the surface functional groups and MgO sites on MgBC. Notably, the presence of Cd­(II) in the mixed solution did not suppress Pb­(II) adsorption; on the contrary, the removal efficiency of Pb­(II) remained consistently above 95% across all tested conditions, suggesting that Cd­(II) may exert a synergistic or non-competitive effect on Pb­(II) uptake. This phenomenon could be explained by the fact that Pb­(II) occupies the most favorable binding sites preferentially, while Cd­(II) may bind to secondary sites or even facilitate Pb­(II) adsorption by modifying the surface charge or inducing structural changes that enhance Pb­(II) accessibility.

In contrast, the adsorption of Cd­(II) was significantly inhibited by the presence of Pb­(II). As the Pb­(II) concentration increased while the Cd­(II) concentration was held constant, both the adsorption capacity and removal efficiency of Cd­(II) decreased markedly. This competitive suppression indicates that Pb­(II) outcompetes Cd­(II) for the limited number of high-affinity binding sites on MgBC, consistent with the stronger affinity and higher binding energy of Pb­(II) toward the adsorbent surface.

The enhanced Pb­(II) uptake in the presence of Cd­(II) may be attributed to three possible mechanisms. First, Cd­(II) adsorption may modify the surface charge of MgBC, creating a more negatively charged surface that enhances electrostatic attraction for Pb²⁺. Second, the coexistence of Cd²⁺ increases ionic strength, compressing the electrical double layer and reducing electrostatic repulsion. Third, pre-adsorbed Cd­(II) may act as bridging sites to form ternary surface complexes, facilitating Pb­(II) binding. Among these, surface charge modification and ternary complex formation are considered the most plausible, consistent with the higher affinity of Pb­(II) toward MgBC. Further investigation is needed to confirm the exact mechanism. Overall, the competitive adsorption experiments demonstrate that MgBC exhibits significantly higher adsorption capacity and stronger binding affinity for Pb­(II) than for Cd­(II) in binary-metal systems. This selectivity is of practical significance for the treatment of wastewater containing multiple heavy metal pollutants, as it allows preferential removal of the more toxic or regulated contaminant (Pb) even in the presence of competing ions.

In real wastewater, coexisting cations (Na⁺, K⁺, Ca²⁺, Mg²⁺) may compete with Pb­(II) and Cd­(II) for adsorption sites on MgBC. The competitive effect should follow Ca²⁺ > Mg²⁺ > K⁺ > Na⁺, based on ionic charge, hydration energy and surface affinity. Pb²⁺ maintains stronger binding to surface sites even in the presence of competing cations due to its higher polarizability. Whereas Cd²⁺ is more susceptible to displacement. Higher ionic strength may reduce electrostatic attraction, but since adsorption is primarily governed by chemisorption, this effect is minor. MgBC is expected to maintain high Pb­(II) performance in complex wastewater, while Cd­(II) may be more sensitive to competing ions. In addition, the selective adsorption behavior of MgBC toward Pb­(II) over Cd­(II) in binary-metal systems is consistent with recent findings on modified biochars. The superior affinity for Pb­(II) is attributed to its smaller hydrated radius, lower hydration energy, and stronger complexation capability compared with Cd­(II). This mechanistic interpretation is consistent with our experimental observations and DFT results. Furthermore, it has been reported that Pb²⁺ adsorption predominantly relies on inner-sphere complexation and precipitation, while Cd²⁺ is mainly removed through outer-sphere complexation and electrostatic attraction. This mechanistic distinction explains the higher affinity and selectivity of MgBC toward Pb­(II) over Cd­(II), providing theoretical guidance for designing selective adsorbents for practical wastewater treatment.

3.6. Proposed Adsorption Mechanism

Based on the comprehensive characterization and experimental results discussed above, the adsorption of Cd­(II) and Pb­(II) onto MgBC is proposed to involve multiple synergistic mechanisms, as illustrated in Figure S9. These mechanisms include: (i) Physical adsorption via the mesoporous structure. (ii) Ion exchange between Mg²⁺ and metal cations, supported by the decreased Mg 1s XPS peak. (iii) Surface complexation with oxygen-containing functional groups (−OH, CO), as evidenced by FTIR peak changes. (iv) π–π interaction between the aromatic carbon structure and metal ions, confirmed by the disappearance of the π–π* satellite peak in XPS C 1s. (v) Precipitation of metal hydroxides/carbonates at near-neutral pH, supported by SEM observations.

Among these, chemisorption (ion exchange and surface complexation) and π–π interactions are dominant, while physical adsorption and precipitation play supplementary roles. The superior affinity of MgBC toward Pb­(II) over Cd­(II) is attributed to the stronger complexation ability and more favorable ion exchange energetics of Pb²⁺ with the MgO/Mg­(OH)₂ and oxygen-containing functional groups. According to the Hard-Soft Acid–Base (HSAB) theory, Pb²⁺ as a softer acid exhibits stronger affinity toward CO groups and π-electrons, whereas Cd²⁺ as an intermediate acid prefers −OH groups. This distinction partially explains the higher adsorption capacity and selectivity of MgBC toward Pb­(II) over Cd­(II), which is consistent with the DFT calculation and experimental results.

3.7. DFT Calculation and MD Simulation for Adsorption Mechanism

A deeper understanding of the adsorption mechanism of Pb­(II) and Cd­(II) on surface of MgBC is particularly important for improving water pollution treatment. However, the adsorption mechanism is not yet fully understood. To investigate the adsorption mechanism in detail, we performed theoretical modeling and DFT calculations and for the adsorption of Pb­(II) and Cd­(II) ions onto MgBC, as well as molecular dynamics (MD) simulation of the interfacial adsorption. Since the adsorption efficiency increases significantly when MgO is loaded onto the biochar surface, it is inferred that the adsorption mainly comes from the MgO surface within MgBC. Meanwhile, the MgO surface was selected as the DFT model because it is the most thermodynamically stable MgO facet. Although XRD and XPS confirmed the coexistence of MgO and Mg­(OH)₂ on MgBC, the surface oxygen sites on MgO(100) and Mg­(OH)₂ are chemically similar as coordination centers for metal cations. Additionally, the MgO surface is partially hydrated under experimental pH conditions, forming surface OH groups that resemble Mg­(OH)₂. MgO(100) is a reasonable proxy for the actual adsorption interface. Moreover, this well-defined periodic model enables direct comparison with prior computational work while avoiding the complexity of modeling Mg­(OH)₂ or MgO/Mg­(OH)₂ heterojunctions. Therefore, we constructed a primitive (Pb/Cd)/MgO system.

Generally, Pb­(II) or Cd­(II) can be adsorbed onto the MgO surface in three main forms. The first is chemisorption where the Pb­(II) or Cd­(II) ion bonds with oxygen from a hydroxyl group on the MgO surface, displacing a hydrogen ion; this is designated as MgBC-OPb or MgBC-OCd. The second is direct chemisorption of the Pb­(II) or Cd­(II) ion onto the MgO surface, designated as MgBC-Pb or MgBC-Cd. The third involves Pb­(II) or Cd­(II) ions occupying crystal defect sites on the MgO surface, designated as MgBCPb or MgBCCd. First, single Pb­(II) or Cd­(II) ions were adsorbed onto the MgO (100) 3 × 3 × 1 crystal surface in the three different configurations described above and optimized. The three optimized (Pb/Cd)/MgO systems were then subjected to molecular MD simulations to obtain reasonable relative configurations and minimum adsorption energies (E _ads). E _ads was calculated using the following equation: E _ads = E _(Pb/Cd)/MgO – (E _Pb/Cd + E _MgO). Here, E _Pb/Cd is the energy of the Pb­(II) or Cd­(II) ion, E _MgO is the energy of the MgO crystal, and E _(Pb/Cd)/MgO is the total energy of the (Pb/Cd)/MgO system. Figure shows the MD simulation results for Pb­(II) or Cd­(II) ions adsorbed on the MgO crystal surface over a 0.2 ps period. The configurations of Pb or Cd adsorbed on the MgO surface for initial, optimized and after MD simulation are shown in Figure S10. Table collects the adsorption energies (kcal/mol), O···Pb/Cd distances, Mg···Pb/Cd distances, and O···H distances (Å) for all (Pb/Cd)/MgO systems. Among them, after optimization and DM simulation for MgBC-Pb and MgBC-Cd, it was found that Pb­(II) and Cd­(II) preferentially coordinate with surface Mg sites on MgO, in contrast to the conventional O-coordination paradigm. The adsorption energies for MgBC-OPb, MgBC-Pb, MgBCPb, MgBC-OCd, MgBC-Cd, and MgBCCd were -2.573, -2.931, 408.492, −0.078, −1.089 and 298.461 kcal/mol, respectively. The calculated results indicate that the adsorption energies for MgBC-OPb, MgBC-Pb, MgBC-OCd, and MgBC-Cd are all less than 0. Therefore, among the three adsorption configurations, spontaneous chemisorption of Pb­(II) or Cd­(II) ions onto the MgO surface and chemisorption by bonding with oxygen from surface hydroxyl groups are the dominant adsorption mechanisms. Furthermore, direct chemisorption of Pb­(II) or Cd­(II) ions onto the MgO surface is more favorable than the hydroxyl-mediated chemisorption. The adsorption type where Pb­(II) or Cd­(II) ions occupy crystal defect sites on the MgO surface has a very low probability of occurrence. An important finding is that the adsorption energies of MgBC-OPb and MgBC-Pb are both lower (more negative) than those of MgBC-OCd and MgBC-Cd, indicating that MgBC adsorbs Pb­(II) ions more readily than Cd­(II) ions. The overall order of adsorption capacity is MgBC-Pb > MgBC-OPb > MgBC-Cd > MgBC-OCd ≫ MgBCPb ≈MgBCCd. This theoretical simulation is in good agreement with the experimental results.

5.

Optimized configurations of Pb or Cd adsorbed on the MgO surface. The dashed lines indicate the key O···Pb/Cd distances, Mg···Pb/Cd and O···H distances.

2. Pb/Cd Adsorbed on the MgBC-Terminated Surface, Together with the Corresponding Adsorption Energy (kcal/mol) and O···Pb/Cd, Mg···Pb/C and O···H Distances (Å) of the Investigated Molecules.

molecules	adsorption energy (kcal/mol)	O···Pb/Cd, Mg···Pb/Cd distances (Å)	O···H distances (Å)
MgBC-OPb	–2.573	3.685	12.056
MgBC-Pb	–2.931	3.372	-
MgBCPb	408.492	-	-
MgBC-OCd	–0.078	2.506	6.400
MgBC-Cd	–1.089	2.810	-
MgBCCd	298.461	-	-

Meanwhile, to gain a deeper understanding of the nature of MgO loading on the biochar surface and the adsorption of Pb­(II) or Cd­(II) ions at the MgO interface, a small MgBC-(Pb/Cd) system was constructed (Figure ). To thoroughly elucidate the interfacial properties among biochar, MgO clusters and Pb­(II)/Cd­(II) ions, theoretical simulations and investigations were also conducted on the biochar/MgO cluster/(Pb/Cd) system. Most literature indicates that biochar components possess a graphene-like structure. Therefore, we theoretically simulated the loading of MgO clusters on a graphene surface. Through interface adsorption simulations, it was found that MgO clusters cannot be directly adsorbed onto the carbon surface; instead, they are loaded by bonding with oxygen atoms from hydroxyl groups abundant on the carbon surface. This theoretical simulation also corroborates the XPS experimental results showing a significantly increased O 1s peak with little change in peak intensity for MgBC and MgBC-Pb/Cd. First, a single MgO cluster was adsorbed onto the graphene (100) 5 × 5 × 1 surface and optimized. Subsequently, Pb­(II) and Cd­(II) ions were adsorbed onto the MgO cluster in the three adsorption configurations described above, followed by re-optimization and calculation of their HOMO and LUMO orbital distributions. The results are shown in Figure . The calculations revealed that the HOMO and LUMO orbitals are uniformly distributed across the entire adsorbed MgBC-(Pb/Cd) system, especially for MgBC-Pb and MgBC-Cd. This demonstrates that in these adsorption systems, chemical bonds are formed between the MgO clusters and biochar by oxygen bridges, as well as different chemical bonds between the Pb­(II)/Cd­(II) ions and the MgO clusters. The theoretical simulations of the HOMO and LUMO orbitals of the small MgBC-(Pb/Cd) system are also in good agreement with the experimental results.

6.

Distribution of HOMO and LUMO orbitals of MgBC-Pb/MgBC-Cd compounds.

Furthermore, the total density of states (TDOS) of the small MgBC-(Pb/Cd) system, along with the projected density of states (PDOS) of Pb­(II)/Cd­(II) ions and MgO clusters, were simulated to explain the adsorption nature of MgO loading on the biochar surface and the adsorption of Pb­(II) or Cd­(II) ions at the MgO interface. As shown in Figure , the green line represents the TDOS, the black line represents the PDOS of the MgO cluster, and the red line represents the PDOS of Pb­(II) and Cd­(II) ions. The results revealed that the bonding energies (bonding active region) of the three MgBC-(Pb/Cd) systems are primarily in the range of −10.0 to 0.0 eV. For MgBC-OPb, the Pb ion exhibits a distinct peak at approximately −2.0 eV (1.6 states/eV). For MgBC-Pb, the Pb ion shows a distinct peak at around −4.5 eV (2.2 states/eV). For MgBCPb, only a very small peak appears at about −2.2 eV (0.3 states/eV). The order of peak heights is MgBC-Pb > MgBC-OPb > MgBCPb. For MgBC-OCd, the Cd ion displays a relatively weak peak at approximately −3.5 eV (0.5 states/eV). For MgBC-Cd, the Cd ion shows a relatively weak peak at around −5.5 eV (0.7 states/eV). For MgBCCd, the Cd ion exhibits almost no peak in the range of 0.0 eV to −10.0 eV. Similarly, the order of peak heights is MgBC-Cd > MgBC-OCd > MgBCCd. The overall order of peak heights is MgBC-Pb > MgBC-OPb > MgBC-Cd > MgBC-OCd > MgBCPb > MgBCCd.

7.

Calculated total densities of states (TDOS) and projected densities of states (PDOS) of MgBC-Pb/MgBC-Cd compounds. TDOS (green), PDOS for MgO (black), PDOS for Pb/Cd (red).

Through theoretical calculations, the order of adsorption capacity of MgBC for Pb­(II) and Cd­(II) ions via different mechanisms is as follows: direct chemisorption of Pb­(II) ions onto the MgO surface > chemisorption of Pb­(II) ions by displacement of hydrogen ions from surface hydroxyl groups on MgO > direct chemisorption of Cd­(II) ions onto the MgO surface > chemisorption of Cd­(II) ions by displacement of hydrogen ions from surface hydroxyl groups on MgO > adsorption of Pb­(II) or Cd­(II) ions occupying crystal defect sites on the MgO surface, that is MgBC-Pb > MgBC-OPb > MgBC-Cd > MgBC-OCd > MgBCPb > MgBCCd. The theoretical simulations of this small system are also consistent with the periodic theoretical simulations and experimental results.

It is worth noting that the combined computational and experimental findings overturn the widely held view that Pb­(II) and Cd­(II) preferentially coordinate with surface oxygen atoms on MgO. Rather, they demonstrate that both metal ions exhibit a stronger propensity to interact directly with surface Mg sites by electronic coupling, leading to the formation of Mg–Pb/Cd metallic bonds, especially Pb­(II), in contrast to the conventional O-coordination paradigm. This theoretical study clearly explains the adsorption mechanism of Pb­(II) and Cd­(II) ions by MgBC, which provides a new atomic scale theoretical basis for designing highly selective adsorbents in further improving water pollution treatment experiments in the future.

4. Conclusions

In summary, MgBC was successfully synthesized from waste kiwifruit pomace and demonstrated as an efficient adsorbent for the removal of Cd­(II) and Pb­(II) from aqueous solutions. Comprehensive characterization confirmed the successful loading of MgO and Mg­(OH)₂ onto the biochar matrix, which endowed MgBC with a well-developed mesoporous structure, contributing to its excellent adsorption performance. Experiments revealed high uptake capacities for Pb­(II) (604.70 mg/g) and Cd­(II) (304.98 mg/g), with Langmuir maximum capacities of 763.82 mg/g and 532.80 mg/g, respectively. In binary-metal systems, MgBC exhibited preferential affinity for Pb­(II), and the presence of Cd­(II) was found to synergistically enhance Pb­(II) adsorption. DFT calculations and MD simulations were performed to elucidate the underlying mechanisms. Three adsorption configurations were identified: direct surface bonding, hydroxyl-mediated interactions and defect-site occupation. The calculated adsorption energies followed the order: MgBC-Pb (−2.93 kcal/mol) < MgBC-OPb (−2.57 kcal/mol) < MgBC-Cd (−1.09 kcal/mol) < MgBC-OCd (−0.08 kcal/mol) ≪ MgBCCd (298 kcal/mol) < MgBCPb (408 kcal/mol), confirming that chemisorption via direct and hydroxyl-assisted pathways dominates the adsorption process, with Pb­(II) being more favorably adsorbed than Cd­(II). These findings were further corroborated by HOMO/LUMO and PDOS analyses, which consistently supported the same affinity sequence. Overall, this work not only provides a comprehensive theoretical foundation for understanding the new atomic scale adsorption behavior of heavy metals on MgBC, but also presents a sustainable strategy for the valorization of agricultural waste (kiwifruit pomace) into a high-performance biochar adsorbent, offering a promising solution for heavy metal remediation in wastewater treatment.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.6c07858.

• Illustrations of N₂ adsorption-desorption isotherms and pore size distribution of BC and MgBC (Figure S1) and TGA analysis curves of BC and MgBC (Figure S2); illustrations of effects of MgBC dosage, initial pH, and initial concentration on Cd and Pb adsorption (Figures S3–S5); illustrations of Langmuir and Freundlich, sips isotherm and Redlich-Peterson isotherm fitting for the adsorption of Cd and Pb by MgBC (Figures S6–S7); kinetic adsorption fitting and Intraparticle diffusion model for the adsorption of Cd and Pb on MgBC (Figure S8); mechanism of Cd and Pb adsorption onto MgBC (Figure S9) and the configurations of Pb or Cd adsorbed on the MgO surface for initial structure, optimized structure and structure after molecular dynamics simulation (Figure S10); comparison of Pb­(II) and Cd­(II) adsorption capacities of various modified biochar adsorbents (Table S1), Langmuir and Freundlich, Sips and Redlich-Peterson, thermodynamic, kinetic model, intraparticle diffusion model parameters of isotherm adsorption models for Cd and Pb on MgBC (Tables S2–S6); adsorption performance of MgBC in a mixed solution of Cd and Pb (Table S7) (PDF)

The authors declare no competing financial interest.