Sustainable Removal of Heavy Metals Using Activated Carbon Produced from Durian Skin: Experiments and Advanced Modelling

Abstract

This study reports a mechanistic analysis of Pb²⁺ and Cu²⁺ adsorption on activated carbon (AC) obtained from durian skin. Pb²⁺ and Cu²⁺ adsorption tests were carried out at 303–323 K and pH 5.5 to interpret the heavy metal—durian skin AC systems. A monolayer model was applied to simulate experimental isotherms and calculate steric and energy parameters. The results indicated that the removal of these target pollutants was a multi-ionic process. The saturation adsorption capacities of this AC improved with increasing aqueous solution temperature, ranging from 81 to 139 mg/g for Pb²⁺ and from 95 to 180 mg/g for Cu²⁺ under the tested operating conditions. The calculated interaction energies indicated a physisorption mechanism where oxygenated functional groups played a relevant role. Durian skin AC can be used as an alternative adsorbent for purifying wastewater and industrial streams polluted by heavy metals.

1. Introduction

The literature indicates that several industrial sectors (e.g., tanneries, paper, and metal plating) impact the environment because of improper disposal of residual effluents containing hazardous heavy metals [1,2,3,4,5,6]. These contaminants have been identified in water resources (e.g., groundwater, creeks, wetlands, lakes, and rivers) and, consequently, generate toxicological risks to living organisms and humans [1,2,3,4,5,6]. Copper (Cu²⁺), lead (Pb²⁺), chromium, mercury, and cadmium have been associated with water pollution, and their slow natural degradation intensifies the environmental damage they cause [7]. Therefore, the development and application of purification methods to treat wastewater and industrial streams containing heavy metals are important for protecting water resources. Note that Cu²⁺ and Pb²⁺ are priority contaminants and are frequently employed as representative target ions in adsorption studies. These cations interact with oxygen-containing functional groups on the surfaces of carbon-based materials, which are model pollutants for evaluating both adsorbent performance and physicochemical interpretation of metal–adsorbent interactions.

The literature review indicates that different technologies can be applied to depollute water containing heavy metal ions, including coagulation, ion exchange, membrane separation and flotation [8,9,10]. These technologies have techno-economic disadvantages, such as the generation of secondary pollution, low pollutant removal rates, and high operational costs [11,12,13,14]. Adsorption is an effective purification method for reducing the concentrations of heavy metals in the environment because this separation process offers simple operation, and the adsorbent can be regenerated and recycled, minimizing waste generation during the treatment process [15]. However, the minimization of water treatment costs to remove heavy metals requires the identification of competitive adsorbents. Activated carbon (AC) is one of the best materials for adsorbing pesticides [16,17,18], dyes [19,20,21], pharmaceuticals [22,23,24], and heavy metals, including Pb²⁺ and Cu²⁺ [25,26,27]. The removal properties of AC can be optimized using different feedstocks, activation routes, and technologies to obtain an adsorbent with improved adsorption capacity to purify contaminated water.

On the other hand, the analysis of the adsorption mechanism of water pollutants in AC systems provides important data for the process design and optimization of water cleaning systems. Classical adsorption models can be applied to interpret experimental isotherms generated during the removal of water pollutants under different operating conditions. For instance, Marcela et al. [27] used the Dubinin–Radushkevich, Freundlich, and Langmuir models to understand the adsorption mechanisms of Cu²⁺ and Pb²⁺ on AC fibers. However, these conventional models provide limited information for analyzing adsorption mechanisms due to their fundamental and theoretical limitations. In contrast, statistical physics models are a better alternative for studying adsorption data via the calculation of physical parameters for understanding the corresponding separation mechanisms of heavy metals using various adsorbents of different origins [28,29,30]. The statistical physics modeling of heavy metal adsorption reported in previous studies concluded that a monolayer process can occur via the contribution of one or more adsorbent functional groups [28,29,30]. These active sites usually correspond to the oxygenated functionalities common to AC surfaces. Therefore, the objective of this paper was to analyze the adsorption mechanisms of Cu²⁺ and Pb²⁺ ions on AC prepared from durian skin using a monolayer statistical physics model. The main physicochemical parameters associated with the adsorption mechanisms of these heavy metals were calculated and discussed. The novelty of this study lies in the application of advanced adsorption models to rationalize the steric and energy parameters associated with heavy metal adsorption on the AC surface. The results showed that a physisorption multi-ionic mechanism governed the removal of these heavy metal ions using durian skin AC.

2. Materials and Methods

2.1. Materials

Durian skin was the AC precursor and collected from several supermarkets in Surabaya, Indonesia. Potassium hydroxide (KOH, ACS reagent, >85%, Sigma-Aldrich, Singapore) was used for the chemical pretreatment of this precursor. Lead chloride (PbCl₂, Sigma-Aldrich, Singapore) and copper chloride (CuCl₂, Sigma-Aldrich, Singapore) were utilized to prepare the heavy metal aqueous solutions. Deionized water was also used to prepare Pb²⁺ and Cu²⁺ solutions and to wash the AC prior to the removal experiments. Nitrogen (N₂) and carbon dioxide (CO₂) gases (supplied by Aneka Gas, Surabaya, Indonesia) were utilized during carbonization and activation of the adsorbent.

2.2. AC Preparation from Durian Skin and Its Characterization

AC was prepared from durian skin through sequential chemical and physical activation. The collected durian skin was crushed to approximately 1 × 1 cm, dried, and pulverized using a hammer mill to obtain particles of approximately 425 µm (equivalent to 40 mesh). The durian skin powder was treated with 2 N KOH for 24 h and then dried at 105 °C for 12 h. The KOH-treated material was carbonized in a pyrolysis reactor at 500 °C for 2 h, with a heating rate of 10 °C/min and an N₂ flow rate of 3 L/min. Physical activation of the char was carried out at 800 °C for 1 h under CO₂ flow. After activation, the reactor was cooled to room temperature under a flow of N₂ at 3 L/min to prevent further oxidation of the AC surface. The final adsorbent was washed with deionized water prior to its use in the adsorption experiments. AC was characterized by Fourier transform infrared (FTIR) spectroscopy, powder X-ray diffraction (XRD), scanning electron microscopy (SEM), and N₂ physisorption analysis. These techniques were used to evaluate the surface functional groups, crystallographic features, morphology, and textural properties of the adsorbent and to associate the main adsorbent properties with heavy metal adsorption. Additional details of the characterization procedures are provided in the Supporting Information.

2.3. Studies of Pb²⁺ and Cu²⁺ Adsorption on Durian Skin AC

Adsorption isotherms were measured at 303, 313, and 323 K, with an initial solution pH of 5.5. The pH of the Pb²⁺ and Cu²⁺ solutions was adjusted before the adsorption experiments using dilute HCl and/or dilute NaOH, and the solutions were measured with a pH meter. The pH was checked again after adsorption, and only a slight variation was observed. Pb²⁺ and Cu²⁺ aqueous solutions with an initial concentration (Co) of 250 mg/L were prepared by dissolving PbCl₂ and CuCl₂ salts (Sigma Aldrich, Singapore) in deionized water. No acidification or other solubilization treatment was used during the solution preparation. Before generating the adsorption isotherms, preliminary screening experiments were conducted to identify the most suitable operating conditions for the equilibrium study. Preliminary contact-time tests were performed to determine that the time required to reach adsorption equilibrium was 24 h. Therefore, the adsorption equilibrium studies were carried out with 100 mL of heavy metal solutions varying AC amounts (0.01–0.1 g) under continuous agitation at 200 rpm for 24 h in a temperature-regulated water bath. The heavy metal equilibrium concentration (Ce, mg/L) was quantified at the end of the removal experiments after separating AC by centrifugation to calculate the corresponding adsorption capacity (Qe, mg/g) via the material balance. Atomic absorption spectroscopy was utilized to quantify the concentrations of both pollutants in the aqueous solutions.

2.4. Models for the Mechanistic Analysis of Heavy Metal Adsorption

Three models were adopted to explain the Cu²⁺ and Pb²⁺ adsorption mechanisms on durian skin AC: two advanced statistical physics models and the classical Langmuir model.

(a) Langmuir model: This model considers Pb²⁺ and Cu²⁺ adsorption as a monolayer process on the AC surface, and the classical hypothesis establishes that a single active site can accept only one heavy metal ion. The Langmuir model can be expressed as follows [31]:

$$Q_e={\displaystyle \frac{q_m{K_{L}C}_e}{1+{K_{L}C}_e}}$$

(1)

where q_m (mg/g) is the monolayer adsorption capacity and K_L (L/mg) is the equilibrium constant for the heavy metal—AC system.

(b) Monolayer model with one active site: This model assumes that Pb²⁺ and Cu²⁺ ions can be removed from an aqueous solution via the contribution of one type of AC functional group, forming a single layer during the adsorption process. However, this model differs from the Langmuir hypothesis in that each AC active site can adsorb a non-fixed number (n_i) of heavy metal cations, depending on the operating conditions. A single adsorption energy (ΔE, kJ/mol) can be estimated for Pb²⁺ and Cu²⁺ interactions involved in the monolayer formation on the adsorbent surface. This advanced monolayer model is given by [32]

$$Q_e={\displaystyle \frac{n_{i}D}{1+{\left(\frac{C_{\frac{1}{2}}}{C_e}\right)}^{n_i}}}$$

(2)

where C_1/2 (mg/L) is the half-saturation concentration and D (mg/g) is the functional group density of AC. The saturation adsorption capacity (Q_m, mg/g) of durian skin AC to remove Pb²⁺ and Cu²⁺ can be obtained from: Q_m = n_i ⋅ D.

(c) Monolayer model with two active sites: In this model, it is assumed that two AC active sites, adsorbing n₁ and n₂ ions, are involved in heavy metal removal. This adsorption mechanism involves two interaction energies, ΔE₁ (kJ/mol) and ΔE₂ (kJ/mol), for each type of adsorption site. The two-energy adsorption model is given by [33]:

$$Q_e={\displaystyle \frac{n_{i1}{D}_1}{1+{\left(\frac{C_1}{C_e}\right)}^{n_{i1}}}}+{\displaystyle \frac{n_{i2}{D}_2}{1+{\left(\frac{C_2}{C_e}\right)}^{n_{i2}}}}$$

(3)

where D₁ and D₂ (mg/g) are the AC active site densities, and C_i (I = 1, 2) is the half-saturation concentration for the ith AC active site. These models were compared, and their parameters were calculated via nonlinear regression of experimental isotherms.

3. Results and Discussion

3.1. Durian Skin AC Characterization

N₂ adsorption–desorption isotherms of durian skin AC are shown in Figure 1. The isotherm of raw durian skin (Figure 1a) shows a Type II-like pattern, characteristic of nonporous or weakly porous solids, and is consistent with its low BET surface area of 3 m²/g. After chemical and physical activation, the isotherm of durian skin AC (Figure 1b) changed markedly, exhibiting a combined Type I/Type IV pattern. The sharp N₂ uptake at low relative pressure indicates the presence of micropores, whereas the uptake at intermediate-to-high relative pressure reflects the contribution of mesopores formed during activation. This result confirms that the activation procedure developed the precursor’s porous texture, resulting in a BET surface area of 1090 m²/g. After heavy metal adsorption, the isotherm of metal-loaded AC (Figure 1c) retained the same general shape as that of raw AC, indicating that the overall porous framework was preserved. However, the total adsorbed N₂ volume decreased significantly, as reflected by the reduction in BET surface area to 560 m²/g. The raw durian skin exhibited a total pore volume of 0.011 cm³/g and an average pore size of ~13.2 nm, which is consistent with its Type II-like isotherm. After chemical and physical activation, the total pore volume increased to 0.767 cm³/g, while the average pore size was 2.67 nm. These values are consistent with the mixed Type I/Type IV isotherm profile and indicate the development of a microporous AC with a mesoporous contribution. After heavy metal adsorption, the total pore volume decreased to 0.394 cm³/g, whereas the average pore size changed only slightly to 2.86 nm. These results suggest that Pb²⁺ and Cu²⁺ adsorption mainly reduced the accessible pore volume through partial pore filling and/or blockage of adsorption sites inside the carbon matrix. SEM images of durian skin precursor and AC are shown in Figure 2. Overall, the AC surface morphology was not altered after heavy metal adsorption. XRD patterns of the raw durian skin and AC are shown in Figure 3. The raw precursor has an amorphous structure, whereas the diffraction pattern of AC exhibited the characteristic broad diffraction peaks of disordered carbonaceous materials at approximately 2θ = 24° and 42°, which can be assigned to the (002) and (100)/(101) reflections, respectively [33]. After heavy metal adsorption, the same broad peaks were still observed in the XRD pattern of the metal-loaded AC, albeit with lower intensity. This result indicates that the overall carbon framework was preserved after adsorption, while partial attenuation of the diffraction signal may be associated with heavy metal loading on the adsorbent surface.

Figure 1. N₂ adsorption isotherms of (a) raw durian skin, (b) durian skin AC, and (c) durian skin AC loaded with heavy metals.

Figure 2. SEM images of (a) durian skin, (b) durian skin AC, and (c) durian skin AC after adsorption of heavy metals.

Figure 3. XRD patterns of (a) durian skin, (b) durian skin AC, and (c) durian skin AC after heavy metal adsorption.

FTIR spectrum of durian skin AC confirmed the presence of oxygen-containing surface functionalities that are relevant to heavy metal adsorption; see Figure 4. After Pb²⁺ and Cu²⁺ removal, slight variations in the absorption band of O–H at 3200 cm⁻¹ and in the oxygen-related bands near 1574, 1445, and 1124 cm⁻¹ were identified. These spectral changes are consistent with the participation of hydroxyl, carbonyl/carboxylate, and C–O surface groups in metal binding. Furthermore, the absorption bands identified in the 530–470 cm⁻¹ region also indicate metal–oxygen interactions. FTIR results are consistent with the proposed mechanism in which Pb²⁺ and Cu²⁺ adsorption occurs through interactions with oxygenated functional groups located on the AC surface.

Figure 4. FTIR spectra of durian skin AC before and after heavy metal adsorption.

3.2. Pb²⁺ and Cu²⁺ Adsorption on Durian Skin AC

Figure 5 reports the experimental adsorption isotherms of Pb²⁺ and Cu²⁺ on durian skin AC at pH 5.5 and 30–50 °C. Overall, an increase in the adsorption temperature favored the removal of both cations using durian skin AC, indicating an endothermic process. The isotherms were fitted to the three models to perform the adsorption mechanism analysis. The calculated R² values for the Langmuir model were lower than those obtained with the two statistical physics models. However, the trends in the calculated parameters of the monolayer model with one active site were reasonable and easier to analyze than those of the model with two active sites. For instance, some C_i values of the monolayer model with two sites were very high and inconsistent with the observed experimental equilibrium concentrations. For simplicity, a monolayer model with one active site was selected to study the adsorption mechanism of both heavy metal ions. All model parameters are summarized in Table 1, and Figure 5 provides the model fitting performance. R² values of this model were 0.998 and 0.997 for Pb²⁺ and Cu²⁺ isotherms, respectively.

Figure 5. Pb²⁺ and Cu²⁺ adsorption isotherms of durian skin AC at 30–50 °C and pH 5.5, and their correlation with a monolayer model with one active site.

Table 1. Calculated parameters of the one-site monolayer model for Pb²⁺ and Cu²⁺ adsorption on durian skin AC.

	Systems
	Pb2+–Durian Skin AC			Cu2+–Durian Skin AC
T (K)	303	313	323	303	313	323
Number of ions adsorbed per functional group, ni	2.29	1.11	1.29	2.86	1.14	0.99
Concentration of functional groups, D (mg/g)	35.78	117.4	108.27	33.23	127.58	185.62
Saturation adsorption capacity, Qm (mg/g)	81.93	130.31	139.66	95.03	145.44	180.79

3.3. Analysis of Steric Adsorption Parameters for the Removal of Pb² and Cu²⁺ on Durian Skin AC

Calculated n_i values were 2.29, 1.11 and 1.29 and 2.86, 1.14, and 0.99 for Pb²⁺-AC and Cu²⁺-AC systems at 303, 313 and 323 K, respectively. Note that three scenarios can be concluded from these values: (a) n_i ≤ 0.5: heavy metal ions were adsorbed on durian skin AC surface via the participation of two or more active sites; (b) n_i ≥ 1: one active site of durian skin AC surface can participate in the heavy metal removal, where the adsorption mechanism can be mono- or multi-ionic; and (c) 0.5 ≤ n_i ≤ 1: the adsorption of heavy metal cations on durian skin AC surface occurred via the presence of a multi-interaction and multi-ionic process. Overall, all n_i values were ≥1, suggesting that one to three ions were adsorbed by each active site of AC. Therefore, Cu²⁺ and Pb²⁺ were removed from the aqueous solution via a multi-cationic mechanism. The estimated n_i values for both adsorption systems were similar, indicating that the AC active sites showed a similar affinity for removing both pollutants. The trend of calculated n_i values versus removal temperature is shown in Figure 6. Thermal agitation reduced the capability of the AC functional groups to interact with several heavy metal ions during the adsorption process. Notably, there is a deviation in the number of ions per site at 313 K, which may be caused by a change in the adsorbent site.

Figure 6. Adsorption temperature versus n_i parameter for Pb²⁺ and Cu²⁺ removal from aqueous solutions using durian skin AC.

Figure 7 reports the trend of adsorption temperature versus the calculated D parameter. In general, the change in solution temperature increased the concentration of AC functional groups that participated in the adsorption of the tested heavy metal ions. This behavior indicated that active sites from the internal pore structure of AC participated more actively in removing the tested pollutants as the aqueous solution temperature increased. Note that the change in density of receptors at 313 K is related to changes in the adsorption sites.

Figure 7. Adsorption temperature versus D parameter for Pb²⁺ and Cu²⁺ removal from aqueous solutions using durian skin AC (T = 303, 313, and 323 K, pH 5.5).

The Q_m results for both systems are shown in Figure 8. Calculated Q_m values increased with the aqueous solution temperature. This trend can be beneficial for industrial processes that generate wastewater at high temperatures, in which durian skin AC displays the best performance. This improvement in AC adsorption properties can be associated with better heavy metal mass transfer in the aqueous solution, favoring cation interactions with the AC active sites [34]. The same behavior was observed for the depollution of these heavy metals using other materials [34]. Q_m values were 81, 130, and 139 mg/g for Pb²⁺ and 95, 145, and 180 mg/g for Cu²⁺ at 303, 313, and 323 K, respectively, using durian skin AC. This indicates that the performance of the AC for Cu²⁺ removal was higher than that for Pb²⁺ adsorption. Table 2 lists a comparison of the heavy metal removal performance of durian skin AC with that reported for other materials. This comparison confirms that durian skin AC offers advantages for the treatment of water polluted with Cu²⁺ and Pb²⁺. The deviation in adsorption capacity is probably a consequence of changes in both the number of molecules per site and the receptor site density.

Figure 8. Adsorption temperature versus Q_m parameter for Pb²⁺ and Cu²⁺ removal from aqueous solutions using durian skin AC (T = 303, 313, and 323 K, pH 5.5).

Table 2. Survey of heavy metal adsorption capacities of different materials reported in the literature.

Heavy Metal	Adsorbent	Qm (mg/g)	Reference
Pb2+	Activated carbon	139	This work
	Carboxymethyl chitosan hybrid montmorillonite	87.95	[35]
	Orange peel	90	[36]
	Tea waste	65	[37]
	Bentonite	51	[38]
Cu2+	Activated carbon	180	This work
	OH-activated Douglas Fir biochar	29	[39]
	Amino-functionalized magnetic nanoparticles	28	[40]
	Amine-functionalized mesoporous silica	15	[41]
	Tree fern	11.7	[42]

Finally, the interaction energies of Cu²⁺ and Pb²⁺ on the tested AC samples were calculated using the following expression:

ΔE = RT ⋅ ln (S/C_1/2)

(4)

where R (J/K⋅mol) is the ideal gas constant, and S is the heavy metal solubility (mg/L). The adsorption energies varied from 18.12 to 22.10 kJ/mol for both adsorption systems. These values suggest that both systems implied the presence of physisorption to remove heavy metals from water using the tested AC. As stated, the oxygenated functionalities of durian skin AC participate in the heavy metal adsorption mechanism.

4. Conclusions

Activated carbon was prepared from durian skin biomass via chemical and physical activation, and the adsorbent was evaluated to remove Pb²⁺ and Cu²⁺ from aqueous solutions under selected laboratory conditions. The equilibrium data of heavy metal adsorption on durian skin activated carbon were better described by the monolayer statistical physics model with one active site, allowing the estimation of steric parameters and interaction energies associated with heavy metal adsorption. The results indicated that Pb²⁺ and Cu²⁺ adsorption on durian skin activated carbon proceeded through multi-cationic and endothermic interactions, in which oxygen-containing surface functionalities may play an important role. The adsorption capacities increased with temperature, reaching 139 mg/g for Pb²⁺ and 180 mg/g for Cu²⁺ under the tested conditions. These findings demonstrate the potential of durian skin-derived activated carbon as a promising adsorbent for heavy metal removal. However, regeneration and reusability, matrix effects, and removal performance in real wastewater should be evaluated to determine the limitations and capabilities of this activated carbon in water purification.