Development and Characterization of Biomass-Derived Carbons for the Removal of Cu²⁺ and Pb²⁺ from Aqueous Solutions

1. Introduction

Heavy metals' increasing contamination of water bodies poses a serious environmental threat, particularly due to the toxic nature of metals such as copper (Cu²⁺) and lead (Pb²⁺). These metals, commonly found in wastewater and leach liquor streams from mining, metallurgy, battery manufacturing, electronics production, and industrial discharge, are non-biodegradable, persistent, and harmful to both the environment and human health [1,2]. Elevated levels of Cu²⁺ and Pb²⁺ in water sources can lead to bioaccumulation in living organisms, resulting in severe health issues such as neurological disorders, kidney damage, and other chronic conditions [2,3,4].

Moreover, the global rise in electronic waste (e-waste) has become a critical environmental challenge, with waste printed circuit boards (PCBs) representing a significant portion due to their complex composition and the hazardous substances they contain [5]. PCBs are integral components in electronic devices such as computers, smartphones, and televisions, and they are often rich in valuable and toxic metals including Cu²⁺ and Pb²⁺ [6,7,8]. Improper disposal or recycling of PCBs can result in the release of these toxic metals into the environment, especially into soil and water systems, through leachates from landfills or unsuitable industrial processing, creating significant environmental and public health risks [9,10]. Therefore, removing these metals from wastewater and PCB leachates is crucial for protecting the environment and recovering valuable resources.

Among the various treatment methods for heavy metal removal, adsorption has gained widespread recognition due to its high efficiency, simplicity, and ability to remove a wide range of pollutants [11,12]. Activated carbon is widely recognized as a highly effective adsorbent for removing heavy metals from aqueous solutions owing to its well-developed porosity, extensive surface area, diverse surface chemistry properties, strong surface reactivity, and significant adsorptive capacity [2,13]. However, traditional activated carbon is often derived from non-renewable sources such as coal, making its production costly and environmentally unsustainable.

In recent years, the focus has shifted towards the use of biomass-derived activated carbons as an eco-friendly and cost-effective alternative, including orange peels [3], date palm leaflets and seeds [14,15], pinewood sawdust [16], and pistachio green hulls [17] among others for the removal of Cu²⁺ and Pb²⁺ from aqueous solutions. Carbonization and activation processes can convert biomass materials into highly porous carbons with tailored surface properties suitable for adsorption applications. Moreover, using biomass waste as a precursor for activated carbon production aligns with the principles of waste valorization and circular economy, making it an environmentally responsible approach [18].

The adsorption capacity of activated carbons has been greatly influenced by the type of raw materials, the methods used in their production, and the specific conditions during fabrication [19]. The carbonization process is typically performed prior to precursor activation, where the material is exposed to pyrolysis, a thermal treatment that enhances its carbon content [20]. During pyrolysis, moisture, hydrogen, volatiles and aromatics with low molecular weight are released, forming a stable carbon-rich structure [21]. The carbonaceous material obtained from pyrolysis normally has limited porosity as most pores are occupied by tar-like substances [21]. Therefore, an activation process is necessary to enhance and develop the specific and desired carbon properties. Activation treatment is applied physically or chemically to improve porosity and develop oxygen-containing functional groups on the surface of activated carbons [22]. KOH, ZnCl₂ and H₃PO₄ as chemical activating agents and CO₂ and steam as physical activating agents can be employed to activate the carbonaceous materials. Comparing physical and chemical activation techniques for preparing activated carbon and assessing their adsorption performance, chemical activation was found to yield superior results, particularly in terms of total pore volume, specific surface area, and porosity enhancement [23].

This work is focused on the preparation of carbon-based adsorbents from olive stones (OS) and almond shells (AS) as low-cost, eco-friendly, and available biomass precursors in order to analyze and evaluate their future potential for the removal of Cu²⁺ and Pb²⁺ from wastewater or leach liquor streams. Utilizing biomass materials as precursors for carbon production promotes sustainability while addressing the dual environmental challenges of e-waste management and wastewater treatment. The influence of the type of activating agent (KOH, H₃PO₄, or ZnCl₂) and the carbon/activating agent (C/A) ratio were examined to achieve better adsorption characteristics for each individual metal. A detailed characterization of the biomass-derived carbons was carried out to provide insights into the physical and chemical properties of the adsorbents, such as surface area, pore structure, functional groups, and crystallinity, which are critical for understanding their adsorption behavior.

2. Materials and Methods

2.1. Materials

Olive stones and almond shells, provided by Fertínez Company (Jaén, Spain) and Biogramasa Company (Granada, Spain), respectively, were utilized as biomass precursors for fabricating carbons. Potassium hydroxide (KOH 85%, Panreac), zinc chloride (ZnCl₂ 98%, Scharlau) and phosphoric acid (H₃PO₄ 85%, Panreac) were used as activating agents. Hydrochloric acid (HCl 37%, Panreac) and sodium hydroxide (NaOH 98%, Sigma Aldrich) were employed to eliminate the excess of activating agents after activation and to determine the point of zero charge of carbons. Cu²⁺ and Pb²⁺ aqueous solutions were prepared by dissolving copper nitrate trihydrate (Cu(NO₃)₂·3H₂O 99%, Fluka) and lead nitrate (Pb(NO₃)₂ 99.5%, Scharlau), respectively, in double distilled water. All chemical reagents were of analytical grade.

2.2. Fabrication of Biomass-Derived Carbons

A total of 14 carbon materials were produced using olive stones and almond shells under various conditions outlined in Table 1. A specific mass of the precursor (200 g of olive stones and 140 g of almond shells) was first placed in an alumina recipient and subjected to pyrolysis inside an oven (Nabertherm GmbH- Germany) under a flow of N₂ (34 L h^-1) with a temperature ramp of 5°C min^-1 until reaching a temperature of 600°C, which was kept for 1 h.

KOH, ZnCl₂ and H₃PO₄ were then used as activating agents to activate the materials obtained from the carbonization process. The KOH and ZnCl₂-activated carbons were produced by mixing the biochar with the corresponding quantities of KOH and ZnCl₂ through milling, achieving C/A ratios of 1:2 and 1:4 (w/w). Subsequently, the mixtures obtained were introduced into the oven and pyrolyzed at 850°C for 2 h with the same temperature ramp and N₂ flow rate as on the carbonization process (5°C min^-1 and 34 L h^-1, respectively). The materials, once cooled, were rinsed with 0.5 M HCl aqueous solutions, followed by double deionized water until a constant pH was achieved. To fabricate H₃PO₄-activated carbons, a certain volume of 85% H₃PO₄ solution was mixed with distilled water to obtain the corresponding C/A ratios (1:2 and 1:4 (w/w)). Then, a specific mass of pre-milled carbonized materials was put in contact with the diluted H₃PO₄ solution with a solid/liquid ratio of 100 g L^-1 and agitated at room temperature for 6 h. The materials obtained were then pyrolyzed in the oven using the procedure previously described for KOH and ZnCl₂ activation. After cooling, H₃PO₄-activated carbons were washed using double distilled water until constant pH. After the washing stage, all the biomass-based carbons prepared were dried at 105°C for 24 h.

2.3. Characterization of Biomass-Derived Carbons

The carbons prepared in this study were characterized through various approaches that enable the analysis of metal adsorption data. BET surface area and porosity determination was carried out using a volumetric sorption analyzer (Micromeritics ASAP 2020), analyzing N₂ adsorption-desorption measurements at -196.15°C, together with CO₂ adsorption data at 0°C. Previously, the carbons were degassed under vacuum at 300°C for 8 h, employing a temperature ramp of 10°C min^-1. The specific surface area was determined from N₂ and CO₂ adsorption isotherms using the BET equation. The total pore volume was obtained based on the quantity of N₂ adsorbed at a relative pressure close to saturation (P/P₀ ≈ 0.99), ensuring a representative measure of the entire porosity network. To thoroughly characterize the porosity, the pore size distribution and total pore volume were further refined using the two-dimensional non-local density functional theory (2D-NLDFT) model, analyzing both CO₂ and N₂ adsorption data simultaneously. This model was selected for its precision in differentiating micropores (≤ 2 nm) from mesopores (2–50 nm), leveraging a two-dimensional adsorption framework to account for surface heterogeneity and pore connectivity effects. The micropore volume was then calculated by subtracting the combined mesopore and macropore (≥ 50 nm) volumes from the total pore volume.

Scanning electron microscopy combined with energy-dispersive X-ray spectroscopy (SEM-EDX) was employed to investigate the carbons' morphological characteristics and surface chemical composition. SEM imaging was carried out using a high-resolution field emission scanning electron microscope (FESEM, Ultra-Plus Zeiss), enabling detailed surface analysis. Elemental mapping and point analysis were performed at multiple locations across the carbon surface to ensure homogeneity in the composition profiles.

The samples were exposed to a high-precision CHNS elemental analyzer (Thermo Flashsmart) to determine the carbon (C), hydrogen (H), nitrogen (N), sulfur (S), and oxygen (O) (calculated by difference) percentages. The analysis was conducted in a dynamic flash combustion mode, where the samples were introduced into a high-temperature reactor in the presence of pure oxygen.

X-ray diffraction (XRD) technique was employed to characterize the crystallinity of the samples, and the diffraction measurements were performed using an Empyrean1 diffractometer from the Panalytical brand. X-rays were generated from a sealed tube with a Cu anode (λ(K_α1) = 1.5406 Å), and the radiation emitted by the sample was obtained using a solid-state "PIXcel3D" detector. The electron diffraction patterns were collected in the angular range of 5° < 2Ɵ < 80° with a step size of 0.04° and a counting time of 1.75 s per step. The HighScore Plus Software Version 3.0d was used for data processing.

Fourier-transform infrared (FTIR) spectroscopy (VARIAN FTIR 670 spectrometer) was used to characterize the chemical bonds and surface functional groups generated on the surface of carbons. The sample pellets were prepared by mixing carbon with potassium bromide (KBr) using a carbon/KBr ratio of 0.4:600 (mg/mg), and the FTIR spectra were obtained within a wide spectral range of 400 – 4000 cm^-1, allowing for the analysis of a variety of chemical bonds.

To determine the point of zero charge (pH_pzc) of the carbons, several flasks were filled with aqueous solutions (10 mL) of 0.01 M NaCl, adjusted to various pH levels (ranging from 2 to 12) by adding varying concentrations of NaOH and HCl (0.1–2 M). After setting the desired pH, 10 mg of the sample was introduced into each flask and shaken at 350 rpm and 25°C using an orbital air bath shaker (VWR Cienytech) for 48 h to achieve adsorption equilibrium. The final pH was then recorded, and the pH_pzc was identified as the intersection of the curve plotting ∆pH (initial pH – final pH) versus initial pH, with the x-axis representing initial pH.

2.4. Adsorption Experiments

Metal adsorption experiments were carried out in batch mode to assess the capability of the prepared carbon-based adsorbents to remove Cu²⁺ and Pb²⁺ ions from aqueous solutions by analyzing their metal adsorption capacity and efficiency. Stock solutions of Cu²⁺ and Pb²⁺ (1000 mg L^-1) were first prepared by dissolving precise amounts of Cu(NO₃)₂.3H₂O and Pb(NO₃)₂ in double distilled water. Then, they were subjected to dilution to obtain the working solutions at specified concentrations. The sorption experiments were conducted in an orbital air bath shaker using 25 mL of the metal solution at initial concentrations of 100 or 500 mg L^-1, which was mixed with the corresponding amount of the adsorbent at adsorption doses of 2 or 5 g L^-1 and maintained at natural pH, an agitation rate of 350 rpm and 25°C for 24 h. After reaching adsorption equilibrium, the carbon materials were separated from solutions by filtration, and the metal concentration of filtrates was determined through flame atomic absorption spectrometry (FAAS, GBC Scientific). All the experiments were carried out in triplicate to guarantee the consistency of adsorption data, and the mean value was calculated for data analysis. Adsorption capacity and efficiency at equilibrium of carbons prepared were calculated using Equations (1) and (2), respectively.

$${\mathrm{q}}_{\mathrm{e}}=\left({\displaystyle \frac{{\mathrm{C}}_{\mathrm{i}}-{\mathrm{C}}_{\mathrm{e}}}{\mathrm{m}}}\right)\times \mathrm{V}$$

(1)

$$\mathrm{A}\mathrm{E}\left(\mathrm{\%}\right)=\left({\displaystyle \frac{{\mathrm{C}}_{\mathrm{i}}-{\mathrm{C}}_{\mathrm{e}}}{{\mathrm{C}}_{\mathrm{i}}}}\right)\times 100$$

(2)

where q_e represents the amount of metal adsorbed at equilibrium (mg g^-1); C_i is the initial metal concentration at t (time) = 0 (mg L^-1); C_e is the final concentration of metals in solution at equilibrium (mg L^-1); V is the volume of metal solution (L); m represents the mass of adsorbent used (g) and AE denotes the adsorption efficiency of the adsorbents at equilibrium (%).

2.5. Statistical Analysis

As mentioned earlier, all adsorption experiments were performed in triplicate, and the results were averaged. The presence of significant differences in the values of adsorption capacity and efficiency of carbons for Cu²⁺ and Pb²⁺ ions were evaluated using one-way analysis of variance (ANOVA), depending on the existence of variance equality and applying the Tukey or T3 Dunnett tests. All statistical analyses were carried out at the 95% confidence level using IBM SPSS Statistics 25 software.

3. Results and Discussion

3.1. Carbon Characterization

The carbons' physical, chemical, and structural characteristics were analyzed through different techniques. The surface morphological features of the carbons were assessed using SEM, and the corresponding images are shown in Figure 1. These figures illustrate the surface texture and porosity variations of non-activated and activated carbons produced using various activating agents at C/A ratios of 1:2 (w/w) and 1:4 (w/w). The results demonstrate the contrasting morphological characteristics and pore structures after activating carbons, especially for the KOH-activated ones, indicating specific modifications in the external surface and the presence of small-sized pores. Hence, a higher degree of porosity and cavities was observed for activated carbons than for non-activated ones. Moreover, the carbons activated with KOH exhibited a more porous texture than the H₃PO₄ and ZnCl₂-activated carbons with scattered cracks generated on their surface, which could result in an enhancement in the number of active surface sites and efficient adsorption of adsorbate molecules [24]. The KOH-activated carbons displayed a significantly porous structure and altered morphology, featuring a monolithic framework and expanded carbon matrix lattices, which could be driven by the mobility and reactivity of metallic potassium (K), as evidenced by the SEM images shown in Figure 1 and Figure 2 (images B and C). The carbons activated with H₃PO₄ exhibited a uniform structure with certain surface modifications, characterized by the formation of small porous regions and randomly distributed pores and craters, which could be due to the diffusion and reaction of H₃PO₄ during thermochemical activation process at 850°C, as depicted in Figure 1 and Figure 2 (images D and E) [25]. Some primary pores and an altered porous framework were also identified for ZnCl₂-activated carbons (Figure 1 and Figure 2 (images F and G)). All the activated carbons are characterized by the presence of mesopores and a significant proportion of micropores embedded within their surface structure, demonstrating a clear alignment with prior research findings [26,27,28].

Moreover, the EDX analysis offers the corresponding surface chemical composition, as presented in Table 2. The findings reveal minor quantities of K, P, Zn, and Cl on the surface of KOH, H₃PO₄, and ZnCl₂- activated carbons, respectively, indicating the effective incorporation of these compounds during the fabrication process. KOH-activated carbons were found to have the lowest content of C and the highest content of O, especially for the carbon fabricated from olive stones with a C/A ratio of 1:4 (w/w) (OSACK2, 53.79% and 35.11% of C and O, respectively). This spatial distribution implies the establishment of chemical bonds, particularly carbon- and oxygen-containing functional groups, with the aromatic structure of the carbon, highlighting the effective integration of these elements at the molecular level [29].

Based on the N₂ adsorption-desorption isotherms at -196.15°C, shown in Figure 3, certain differences in the shape and magnitude of the isotherms can be observed depending on the activating agent used to produce the carbon adsorbents. Regarding N₂ adsorption capacity, activated carbons treated with KOH demonstrated a high amount of N₂ adsorbed on the surface, indicating that these materials possess the largest surface area. In all cases, the N₂ adsorption-desorption isotherms fall under type I, which suggests a significant presence of microporosity in the solids. The isotherms of the KOH-activated carbons display a slightly different shape compared to non-activated ones and those activated with other agents, particularly showing a broader knee. This suggests that the microporosity is generated by micropores in the highest range of pore diameter (supramicroporosity), while the carbons activated with H₃PO₄ and ZnCl₂ appear to have smaller pores. Moreover, the use of olive stone as a precursor tends to generate a narrower knee, especially when using a 1:2 (w/w) C/A ratio.

On the other hand, the plateau in the isotherms for KOH-activated carbons does not reach a constant value, and a slight increase in the amount of N₂ adsorbed could indicate the presence of a certain degree of mesoporosity.

Regarding the amount of activating agent used during the fabrication process, it is evident that the isotherms for KOH-activated carbons change noticeably as the quantity of KOH is increased, with the knee extending over a larger relative pressure range and an increase in the magnitude of adsorbed nitrogen. Activated carbons treated with ZnCl₂ exhibited a similar trend to that observed for KOH-activated carbons, with an increase in the amount of N₂ adsorbed, whereas no influence was observed for H₃PO₄-activated carbons.

CO₂ adsorption isotherms at 0°C were obtained to supplement the conclusions drawn from the N₂ adsorption studies at -196.15°C. The experimental data (Figure 4) show that KOH-activated carbons achieve higher CO₂ adsorption values, although this difference is mainly observed in the high-pressure range. The presence of ultramicropores is associated with the amount of CO₂ adsorbed [30], but preferably at relative pressures below 0.02. Carbons activated with H₃PO₄ exhibit similar isotherms, suggesting a similar microporosity. For carbons activated with ZnCl₂, a substantial increase in the amount of CO₂ adsorbed is observed as the C/A ratio increases, matching the behavior seen in the N₂ adsorption studies.

As previously discussed, the experimental data obtained for N₂ and CO₂ adsorption were used to generate the information presented in Table 3. As mentioned earlier, the use of KOH tends to produce carbons with significantly greater surface areas (determined using N₂), which increase as more KOH is used in the activation process for both biomass precursors. This is consistent with the well-known ability of KOH activation to develop a highly porous network, resulting in substantial increases in both microporosity and mesoporosity. Considering the effect of the activating agent on surface area magnitude, the trend observed was KOH > H₃PO₄ > ZnCl₂ for almond shell-based carbons and KOH > ZnCl₂ > H₃PO₄ for olive stone-based carbons. While KOH and ZnCl₂ allow the modification of the surface area value when the C/A ratio is varied, H₃PO₄ does not show that influence. H₃PO₄ and ZnCl₂-activated carbons yielded moderate increases in surface area, promoting micropore formation without significantly enlarging mesopores.

Moreover, total pore volume and pore size provide further insight into the structural transformations induced by activation. The KOH-activated carbons had the highest total pore volumes compared to non-activated, H₃PO₄ and ZnCl₂-activated carbons (Table 3). This trend highlights that KOH activation increases surface area and substantially amplifies the overall pore volume, enhancing the material’s capacity to retain adsorbates. The greatest average pore diameter was obtained for KOH-treated samples in the range of 4.59 to 5.34 nm, indicating the introduction of wider mesopores alongside micropores. This bimodal pore size distribution is advantageous for applications requiring both high surface area for adsorption and mesoporous channels for rapid molecular transport.

XRD analysis was employed to provide a more detailed understanding of the changes in the crystalline structure of biomass-derived carbon adsorbents, and the corresponding patterns are presented in Figure 5. From the figure, two prominent diffraction peaks can be observed at around 23-25° and 43.6° for non-activated carbons and H₃PO₄ and ZnCl₂-activated carbons from both precursors, representing the existence of the (002) and (100) crystal planes of the graphite layer, respectively, caused by weekly crystalline phases [31]. The peak observed at around 23-25° confirms the hexagonal and developed aromatic crystallite characteristics, whereas the broad and weak band at about 43.6° represents the microcrystalline and amorphous nature of the carbons with a very low degree of graphitization [26,32]. However, different patterns were detected for KOH-activated carbons, especially for the carbons activated with a C/A ratio of 1:4 (w/w). The (002) diffraction plane characteristic of graphite is missing in all KOH-activated carbons, which may be due to the structural disintegration of the carbon matrix, potentially leading to the disruption of the graphite-like ordering [33]. On the other hand, compared to the other carbons, the intensity of the (100) plane peak (43.8°) was significantly reduced for KOH-activated carbons, particularly for those with a 1:4 (w/w) ratio, which almost disappeared, signifying a diminishing degree of graphitization order within the samples [34]. In addition, the diffraction peak observed at roughly 21.5° is likely associated with cellulose. In comparison, the peak at 26.3° is due to the presence of crystalline inorganic phases, which can be associated with the trace elements found within the KOH-activated carbons [34]. The sharp peaks observed in KOH-activated carbons with a ratio of 1:4 (w/w) may be ascribed to the formation of additional chemical species within the interlayers of the collapsed hexagonal graphite (carbon) framework [33].

FTIR analysis was carried out to provide insight into the surface chemistry and functional groups present on the surface of carbons prepared in this study. The FTIR spectra of samples exhibited several characteristic peaks, indicating the presence of a variety of functional groups, as depicted in Figure 6. The peak observed at around 3433 cm⁻¹ is typically attributed to the O-H stretching vibrations of hydroxyl groups, including those associated with alcohols, phenols, and carboxylic acids [2,35]. Non-activated and KOH-activated carbons exhibited a broader intensity at this peak compared to H₃PO₄ and ZnCl₂-activated carbons, suggesting a significant concentration of O–H groups due to their alkaline nature. This observation aligns with the pH_pzc data of the carbons, as presented in Table 4, corroborating the consistency between the spectral findings and the surface charge characteristics of the materials. The band located at about 3136-3149 cm⁻¹ can be assigned to the N–H stretching mode observed in compounds such as amines or to the O–H bond [33,36]. Moreover, the dual peaks detected at 2920 cm⁻¹ and 2852-2862 cm⁻¹ correspond to the stretching vibrations associated with aliphatic C–H bonds [37], whereas the peaks obtained within the spectral region of 1711-1733 cm⁻¹ are indicative of the C=O stretching vibrations, which are characteristic of carbonyl functionalities found in compounds such as esters and carboxylic acids [33]. The transmittance bands at roughly 1595-1637 cm⁻¹ can be ascribed to the in-plane deformation vibrations of the N=H bond or the C=C stretching vibrations, particularly those inherent to aromatic rings [34]. A sharp peak was detected at 1404 cm⁻¹, which correspond to bending vibrations of alkyl C–H bonds, reflecting the presence of methyl (–CH₃) or methylene (–CH₂–) groups [38]. The prominent and well-defined band at 1099 cm⁻¹ observed in non-activated, KOH- and H₃PO₄-activated carbons contrasted with the relatively weak and broad peak detected at 1124 cm⁻¹ in ZnCl₂-activated carbons that can be attributed to C–O stretching vibrations, typically associated with alcohols, phenols, ethers, or esters [39,40]. Furthermore, the peaks identified at 796 cm⁻¹ and 470 cm⁻¹ in non-activated, KOH- and H₃PO₄-activated carbons are likely attributed to out-of-plane bending vibrations of aromatic C–H bonds [33]. The peaks appearing at 1099 cm^-1 and 470 cm^-1 in carbons activated with H₃PO₄ can also be assigned to the stretching vibrations of the O–P–O and P=O bands, respectively [41]. In contrast, the characteristics peak observed at 592 cm⁻¹ for ZnCl₂-activated carbons can be ascribed to Zn–O stretching vibrations [42].

Point of zero charge (pH_pzc) is a key property in understanding how carbon-based materials interact with different ions in aqueous solutions, which reflects the acidity or basicity of the material's surface. The pH_pzc is the pH at which the material's surface has a neutral net charge. If the pH of the solution is below the pH_pzc, the surface of the carbon material will be positively charged. On the contrary, if the pH is above the pH_pzc, the surface will be negatively charged [43]. The point of zero charge of the carbons prepared are given in Table 4. As observed, non-activated and KOH-activated carbons possess pH_pzc values within a range of 7.3-8.4, indicating that their surface is in slightly basic conditions. This could be due to the introduction of electron-donating groups such as amine or hydroxyl groups on the carbon surface [44]. In contrast, the carbons activated with H₃PO₄ exhibited the lowest pH_pzc values (2.9-3.5), suggesting that phosphorus modification greatly increases the surface acidity. ZnCl₂-activated carbons showed intermediate pH_pzc values (5.5-6.2), confirming that zinc modification makes the surface slightly acidic but not as much as phosphorus, as zinc oxide is amphoteric and can behave as both an acid and a base [45].

The CHNS elemental analysis provides valuable insights into the changes in elemental composition following chemical activation of olive stone and almond shell-derived carbons, and the corresponding results are given in Table 5. These changes are critical as they influence the surface chemistry and the potential applications of the activated carbons. As observed in Table 5, the non-activated carbons (OSC and ASC) exhibited the highest carbon contents, 83.36%, and 83.14%, respectively, with comparable hydrogen, nitrogen, and oxygen levels. These high carbon percentages reflect the inherent carbon-rich nature of the biochar produced from biomass prior to activation. The KOH activation induced a pronounced carbon content reduction alongside a significant oxygen level increase, a trend consistently observed across both OSACK and ASACK series samples. For instance, within the OSACK series, the carbon content declined markedly from 83.36% in OSC to 66.44% in OSACK1 and further to 55.02% in OSACK2, while oxygen levels increased from 14.40% in OSC to 32.02% and 43.64% in OSACK1 and OSACK2, respectively. Similar patterns were observed in the ASACK series, where carbon content in ASACK1 and ASACK2 was reduced to 66.48% and 56.72%, with corresponding oxygen amounts rising to 32.23% and 41.75%. The substantial decrease in carbon content and concurrent oxygen increase, which is consistent with the findings obtained from the EDX analysis (Table 2), suggests the incorporation of oxygen-containing functional groups, likely as a result of KOH strong oxidizing effect, which enhances pore development and functional group introduction [46]. Activation with H₃PO₄ also reduced the carbon content (70.60%-74.18%), albeit not as drastically as KOH activation, and introduced a moderate increase in oxygen amount (23.54%-27.20%). The less pronounced carbon depletion and oxygenation in H₃PO₄-activated carbons suggest that H₃PO₄ is milder compared to KOH, promoting the formation of phosphorus-containing functional groups that enhance surface polarity without significant carbon loss [47]. The ZnCl₂-activated carbons demonstrated the highest carbon contents among the activated samples, containing 80.65%-82.12%. This indicates that ZnCl₂ is less aggressive in depleting carbon compared to KOH and H₃PO₄, likely because ZnCl₂ primarily acts as a dehydrating agent, promoting carbonization and cross-linking without extensive oxygen incorporation [48]. Correspondingly, the relatively low oxygen contents (15.57%-17.41%) were obtained for ZnCl₂-activated carbons. In addition, the significantly low hydrogen and nitrogen contents could be associated with promoting carbon and oxygen-dominant structures, possibly through gasification reactions that remove other elements [49].

3.2. Cu²⁺ Adsorption Performance of Biomass-Derived Carbons

The Cu²⁺ adsorption efficiency (AE) and capacity (q_e​) of non-activated and activated carbons were assessed under varying conditions, specifically, at two Cu²⁺ concentrations (100 and 500 mg L⁻¹) and two adsorbent dosages (2 and 5 g L⁻¹), to identify the most effective adsorbent for the removal of Cu²⁺ ions from aqueous solutions (Table 6). Variations in adsorption performance across carbons activated with KOH, H₃PO₄ and ZnCl₂ reveal the influence of activation method and surface characteristics acquired on Cu²⁺ removal efficiency. As observed, the non-activated carbons (OSC and ASC) displayed relatively lower AE and q_e values than KOH and H₃PO₄-activated carbons under all conditions essayed. At a low Cu²⁺ concentration (100 mg L^-1) with an adsorbent dosage of 2 g L^-1, OSC and ASC achieved moderate AE values of 29.92% and 43.25%, with q_e​ values of 14.75 mg g^-1 and 21.32 mg g^-1, respectively. However, increasing the adsorbent dosage to 5 g L^-1 at 100 mg L^-1, AE slightly increased (33.44% for OSC and 51.40% for ASC), but significantly lower q_e​ values (7.21 mg g^-1 and 11.10 mg g^-1), a similar trend to previous studies [50]. At 500 mg L^-1, both OSC and ASC showed a drop in AE (12.81% and 13.06% at 2 g L^-1; 8.80% and 15.33% at 5 g L^-1) with correspondingly lower q_e values, indicating that non-activated carbons struggle with efficient Cu²⁺ removal at high initial concentrations, likely due to limited active sites [51]. ASC's superior Cu²⁺ adsorption efficiency compared to OSC, particularly at the lower concentration of 100 mg L⁻¹, may be ascribed to its relatively larger average pore size (Table 3).

KOH-activated carbons were found to possess superior adsorption characteristics among all samples, especially for the carbons fabricated with a C/A ratio of 1:4 (w/w) (OSACK2 and ASACK2) with the highest AE and q_e​, underscoring the effectiveness of KOH activation in enhancing Cu²⁺ adsorption potential. At 100 mg L^-1 Cu²⁺ concentration and 2 g L^-1 dosage, OSACK2 and ASACK2 achieved notable AEs of 86.29% and 84.63% with elevated q_e​ values of 42.52 mg g^-1 and 41.69 mg g^-1, respectively. Increasing the dosage to 5 g L^-1 further raised the AE to near-complete levels (99.86% and 98.18%), although the q_e​ values decreased to 21.58 mg g^-1 and 21.20 mg g^-1. At the higher Cu²⁺ concentration (500 mg L^-1), KOH-activated carbons maintained effective performance, particularly for OSACK2 and ASACK2, which exhibited AEs of 26.88% and 24.92% at 2 g L^-1, with q_e values of 66.62 mg g^-1 and 62.33 mg g^-1, respectively. The adsorption efficiencies nearly doubled when the dosage increased to 5 g L^-1 (48.65% and 45.15% for OSACK2 and ASACK2, respectively), with the high associated q_e​ values of 49.40 mg g^-1 and 45.83 mg g^-1, respectively, indicating that KOH activation effectively creates high-energy adsorption sites that accommodate greater Cu²⁺ loadings. The findings also demonstrated that the C/A ratio profoundly affects the structural and chemical characteristics of the carbon activated by KOH, thereby influencing its Cu²⁺ adsorption performance. With increasing the C/A ratio from 1:2 (w/w) to 1:4 (w/w), the amount of copper adsorbed increased, especially evident under two specific conditions: 100 mg L^-1 with 2 g L^-1 adsorbent dose and 500 mg L^-1 with 5 g L^-1 adsorbent dose, as confirmed by the analysis of significant differences. As shown in Table 3, the surface area and total pore volume of KOH-activated carbon increased substantially with a higher C/A ratio for both precursors, as the activating agent helped to create a more porous framework by etching and expanding existing pores. This high porosity facilitates Cu²⁺ ion diffusion into the carbon matrix, thus improving the adsorption capacity [52]. Furthermore, the C/A ratio directly impacted the diversity of functional groups on the carbon surface, which plays a critical role in Cu²⁺ binding through chemisorption mechanisms [24]. EDX and CHNS elemental analyses (Table 2 and Table 5) revealed a considerably high oxygen content on the surface of KOH-activated carbons, particularly in the olive stone-derived activated carbon produced at a 1:4 (w/w) ratio. This oxygen concentration could be associated with a significant presence of oxygen-containing functional groups, which exhibit a strong affinity for metal ions, enhancing adsorption through ion exchange and surface complexation mechanisms [53].

The carbons activated with H₃PO₄ showed moderate adsorption performance, with AE and q_e​ values falling between non-activated and KOH-activated carbons. For instance, at 100 mg L^-1 Cu²⁺ concentration and 2 g L^-1 dosage, OSACP2 and ASACP1 recorded AE values of 38.97% and 44.12%, with q_e values of 19.17 mg g^-1 and 21.73 mg g^-1, respectively. When the dosage increased to 5 g L^-1, AE improved to 50.05% and 58.77%, although q_e​ decreased to 10.33 mg g^-1 and 12.13 mg g^-1, respectively. A similar trend was identified using higher Cu²⁺ concentration, achieving lower AE and q_e​ values. This implies that while H₃PO₄ activation moderately enhances adsorption sites, it is less effective than KOH at producing the high-affinity sites required for larger Cu²⁺ loadings. The almond shell-derived activated carbon fabricated at a 1:2 (w/w) ratio (ASACP1) demonstrated slightly more favorable adsorption performance among the H₃PO₄-activated carbons, especially at a Cu²⁺ concentration of 100 mg L^-1 and an adsorbent dose of 5 g L^-1, which may be attributed to its relatively higher surface area and pore volume (Table 3). Moreover, increasing the C/A ratio from 1:2 (w/w) to 1:4 (w/w) did not result in a substantial enhancement in adsorption capability for either precursor, suggesting no significant differences in adsorption efficacy between these two ratios, as outlined in Table 6. This outcome may be attributed to the lack of significant surface chemistry and pore structure alterations despite the increased C/A ratio.

ZnCl₂-activated carbons had the lowest AE and q_e​ values among the non-activated and activated carbons. For example, at 100 mg L^-1 Cu²⁺ concentration and 2 g L^-1 dosage, OSACZ1 and OSACZ2 achieved AEs of only 23.70% and 22.56%, with corresponding q_e​ values of 12.33 mg g^-1 and 11.80 mg g^-1, respectively. A similar behavior was observed when the dosage increased to 5 g L^-1, where AE slightly improved but q_e​ remained low, indicating limited adsorption efficiency. At 500 mg L^-1, ZnCl₂-activated samples displayed poor adsorption performance (e.g., 7.73% and 4.32% AE for OSACZ1 and OSACZ2, respectively, using an adsorbent dosage of 2 g L^-1), highlighting ZnCl₂ activation's limitations in creating high-capacity adsorption sites for copper. This result suggests that ZnCl₂ may primarily enhance microporosity rather than the specific types of active sites that favor Cu²⁺ ion adsorption.

In general, KOH-activated carbons (especially OSACK2 and ASACK2) demonstrated the highest Cu²⁺ adsorption efficiency and capacity, making them suitable for high-performance Cu²⁺ adsorption applications. This suggests that KOH activation produces a more suitable porous structure and surface chemistry that enhances interactions with Cu²⁺ ions. H₃PO₄ activation resulted in moderate improvements in adsorption, suitable for applications with lower concentration requirements, while ZnCl₂ activation was less effective for Cu²⁺ adsorption under the tested conditions. These results highlight the importance of activation method selection for optimizing carbon materials for specific heavy metal adsorption applications.

Based on previous analysis about the influence of average pore size upon Cu²⁺ adsorption in the different carbon-based adsorbents and considering possible size exclusion phenomena, an evaluation of the role of each pore type (total, mesopore and micropore) was carried out observing a clear impact of mesopore volume upon adsorption efficiency and capacity for all the adsorbents obtained. Figure 7 shows examples of this type of influence, and it is possible to observe that an increase in mesopore volume in the adsorbent enhances Cu²⁺ adsorption. This behavior is not observed for micropore volume, indicating that pore size plays an important role and also concluding that part of the specific surface area of adsorbent is not available for metal adsorption.

3.3. Pb²⁺ Adsorption Performance of Biomass-Derived Carbons

As in the previous section, adsorption efficiency (AE, %) and capacity (q_e, mg g^-1) of biomass-derived carbons were evaluated for removing Pb²⁺ ions from aqueous solutions under different concentrations (100 mg L^-1 and 500 mg L^-1) and adsorbent dosages (2 g L^-1 and 5 g L^-1), and the corresponding data are presented in Table 7. At 100 mg L^-1 Pb²⁺ initial concentration and 2 g L^-1 adsorbent dosage, KOH-activated carbons displayed the highest adsorption efficiency (AE=100%) and adsorption capacity (q_e=52.25-52.28 mg g^-1), showing their potential for high removal rates at low dosages. This performance indicates that these activated carbons possess a high surface area and effective pore distribution for Pb²⁺ adsorption. Other carbons, like non-activated (OSC and ASC) and H₃PO₄-activated carbons (OSACP and ASACP series), also exhibited high AE values of 62.58-75.56% and 73.04-79.28%, respectively, but with slightly lower q_e values (34.55-41.45 mg g^-1 and 37.83-41.18 mg g^-1, respectively). This demonstrates that while these carbons are effective for Pb²⁺ removal, their adsorption capacity is lower than those of KOH-activated carbons, possibly due to a less favorable pore structure (in agreement with data in Figure 8) or fewer available functional groups.

Increasing the adsorbent dosage to 5 g L^-1 at the same Pb²⁺ concentration decreased q_e across all samples. Apart from KOH-activated carbons (OSACK and ASACK series), ASC was found to be the only carbon with complete adsorption level (AE=100%). In contrast, a slight increase in adsorption efficiency was observed for other carbons (73.28%, 83.82-89.65%, and 68.05-78.90% for OSC, H₃PO₄-activated carbons, and ZnCl₂-activated carbons, respectively).

At the highest concentration (500 mg L^-1) and lower adsorbent dosage (2 g L^-1), the carbons activated with KOH at a C/A ratio of 1:4 (w/w) (OSACK2 and ASACK2) showed notable adsorption performance with AE values of 75.14% and 72.83%, respectively, and the highest q_e values (186.16 and 182.69 mg g^-1). This indicates a strong capacity for Pb²⁺ adsorption at higher initial metal concentrations, likely facilitated by the adsorbent’s high surface area and well-developed porosity. OSACK1 and ASACK1 also demonstrated good performance with AE values around 56% and q_e values near 140 mg g^-1, suggesting that they effectively accommodate higher Pb²⁺ loads even at moderate dosages. In comparison, the non-activated carbons and carbons activated with H₃PO₄ and ZnCl₂ had lower AE values (27.02-33.35%, 34.20-39.40%, and 20.54-26.88%, respectively) and q_e values ranging from 71.21 to 88.06 mg g^-1, 85.45 to 97.94 mg g^-1, and 51.77 to 67.59 mg g^-1, respectively, indicating reduced effectiveness at high Pb²⁺ concentrations. This likely points to limitations in available surface area or pore structure for these specific carbons.

At an increased dosage of 5 g L^-1 and 500 mg L^-1 Pb²⁺ concentration, OSACK2 and ASACK2 again reached maximum AE values (100%) with high q_e values of around 101 mg g^-1. These results suggest that these carbons exhibit strong metal uptake potential regardless of adsorbent dosage, making them highly efficient in Pb²⁺ removal under varying operational conditions. The improvement in AE across all carbons such as OSACK1 (97.61%), ASACK1 (93.58%), and OSACP2 (62.84%) reflects the advantage of higher adsorbent dosage in enhancing adsorption efficiency for systems with high Pb²⁺ concentrations. The enhanced AE and q_e values of KOH-activated carbons compared to other carbons can be attributed to their higher surface area, pore volume, and pore size, as indicated in Table 3 and supported by Figure 8, as well as the presence of more surface functionalities, particularly oxygen-containing functional groups, that facilitate Pb²⁺ adsorption. The findings further revealed that the carbon materials synthesized in this study exhibited a significantly enhanced capacity for Pb²⁺ ion recovery compared to Cu²⁺, implying a pronounced selectivity for Pb²⁺ over Cu²⁺ in these materials. This behavior may be attributed to the larger ionic radius of Pb²⁺ compared to Cu²⁺, which results in a stronger attraction and binding affinity of Pb⁺² ions to the functional groups present on the surface of carbons, especially oxygen-containing functional groups such as hydroxyl (OH) and carboxyl (COOH) groups, due to intensified electrostatic interactions [54,55].

The data also indicate that base and chemical activation are effective strategies for enhancing adsorption performance. However, the optimal conditions for maximum Pb²⁺ removal vary across carbon types, likely due to specific interactions between Pb²⁺ ions and carbon surface properties. The high efficiency of OSACK and ASACK series carbons, especially those activated with a 1:4 (w/w) ratio, highlights their potential as viable options for Pb²⁺ removal applications, particularly at high pollutant concentrations and moderate dosages.

4. Conclusions

In this study, biomass-based carbon adsorbents were produced from olive stones and almond shells using various activating agents such as KOH, H₃PO₄, and ZnCl₂. The activation method influenced the physical and chemical properties of the biomass-derived carbons, leading to differences in surface area, pore structure, and functional group distributions that impact their adsorption capabilities. Compared to H₃PO₄- and ZnCl₂-activated carbons with a high degree of microporosity, activation with KOH yielded carbons with highly developed micro and mesoporous structures, greater surface area and pore size, and more introduction of oxygen-containing functional groups, exhibiting significant potential for the removal of heavy metal ions, specifically Cu²⁺ and Pb²⁺, from aqueous solutions. KOH-activated carbons prepared using a C/A ratio of 1:4 (w/w) from both precursors (OSACK2 and ASACK2) were found to have the highest adsorption performance for the removal of Cu²⁺ and Pb²⁺, achieving adsorption efficiencies of 84.63-86.29% (41.69-42.52 mg g^-1) at 100 mg L^-1 and 2 g L^-1 dosage and 100% (101.41-101.68 mg g^-1) at 500 mg L^-1 and 5 g L^-1 dosage, respectively. Moreover, the carbons activated with H₃PO₄ and ZnCl₂ did not significantly influence adsorption, likely due to their specific texture, surface properties, and adsorption potential. The results also indicated that the carbons fabricated in this work have a significantly higher capacity for Pb²⁺ adsorption than for Cu²⁺ adsorption. This research aims to contribute to sustainable e-waste and wastewater management by developing biomass-derived activated carbons as cost-effective, renewable adsorbents for heavy metal removal. The study offers an environmentally friendly solution to the growing problem of heavy metal pollution, specifically targeting wastewater and leach liquors from waste PCB processing. By valorizing biomass waste into high-performance adsorbents, the research aligns with circular economy principles, reduces the environmental footprint of both e-waste and wastewater treatment, and provides a scalable method for mitigating the adverse effects of toxic metal contamination.