Open Access Article
Ahlem Sarra Saadi*ab,
Djihane Slimane Ben Ali
ab and
Salim Bousbacd
aDepartment of Process Engineering, Faculty of Technology, University 20-August-1955, Skikda, Algeria. E-mail: as.saadi@univ-skikda.dz; sara.ahlem21@gmail.com
bLRPCSI- Laboratoire de Recherche sur la Physico-Chimie des Surfaces et Interfaces, Université du 20 Août 1955, BP. 26, Route d’El-Hadaiek, Skikda, Algeria
cDépartement de Génie des Procédés, Faculté de Génie des Procédés, Université Salah Boubnider Constantine 3, Algeria
dLaboratoire Médicament et Développement Durable ReMeDD, Université Salah Boubnider Constantine 3, Algeria
First published on 6th November 2025
This study reports the preparation and characterization of a novel biosorbent obtained from oak pericarp for the removal of rhodamine B (RhB) dye from aqueous solutions. Comprehensive analyses (FTIR, SEM-EDX, TGA, BET, XRD, and pHpzc) confirmed the presence of diverse functional groups and a heterogeneous surface morphology contributing to adsorption. Batch experiments were performed under varying operational conditions (pH, contact time, adsorbent dosage, ionic strength, temperature, and initial concentration), with optimal adsorption achieved at pH 2, using 0.05 g L−1 of biosorbent and an equilibrium time of 300 min. Adsorption equilibrium data were best fitted by the Langmuir–Freundlich isotherm, suggesting heterogeneous and multilayer adsorption, with a maximum capacity of 160.809 mg g−1 at 20 °C. Kinetic modeling indicated that the pseudo-second-order model best described the process, pointing to chemisorption as the dominant mechanism. Thermodynamic results confirmed the spontaneous and endothermic nature of RhB adsorption. Furthermore, regeneration tests revealed that the biosorbent retained high efficiency over multiple cycles, demonstrating its potential as a sustainable, low-cost, and environmentally friendly material for wastewater treatment.
Among these pollutants, rhodamine B (RhB) stands out as a cationic xanthene dye known for its stability, vivid color, and environmental persistence. Its presence in water bodies not only reduces photosynthetic activity by limiting light penetration but also poses health hazards, as it is potentially toxic, carcinogenic, and mutagenic.3,4 Dyes are generally classified into categories such as reactive, direct, basic, disperse, acid, vat, and azo dyes, each exhibiting distinct chemical structures and environmental impacts.5
Conventional water treatment methods—including coagulation–flocculation, ion exchange, advanced oxidation processes, photocatalysis, sonochemical treatments, membrane separation, and reverse osmosis—have been widely applied. However, they often suffer from high operational costs, incomplete removal, or secondary pollution issues.6,7 In contrast, adsorption, particularly using biosorbents derived from agricultural or plant waste, has emerged as an attractive, environmentally friendly, and cost-effective alternative. Such biosorbents typically exhibit abundant functional groups (hydroxyl, carboxyl) and porous structures that facilitate the effective removal of organic dyes.8,9
Recently, attention has shifted toward sustainable, plant-based biosorbents, which present a green and nature-based solution for dye removal. These biomass-derived adsorbents offer several advantages, including low cost, abundance, ease of regeneration, high affinity for diverse pollutants, and minimal risk of secondary contamination.10,11
Nonetheless, one of the main limitations in conventional adsorbents is the difficulty of separating them from treated water. Applications of such biosorbents for RhB removal have been demonstrated using a variety of plant-derived materials for instance: banana peel powder, which displayed removal efficiencies up to 81% and followed Langmuir isotherm behavior;12 macroalgal biomass, modified or native, achieving capacities over 100 mg g−1 under acidic conditions;13 rice husk biochar, reaching removal efficiencies near 98% under optimized conditions, with monolayer physisorption behavior and feasible regeneration.14
These findings underscore the potential of plant-waste-based biosorbents, but they also highlight the need for new materials with optimized properties tailored for RhB adsorption, particularly in terms of adsorption capacity, selectivity, and reusability.
Oak acorn pericarp, an abundant lignocellulosic by-product in Algeria, has not yet been extensively explored as a biosorbent. Its heterogeneous surface structure and rich functional groups make it a promising candidate for dye removal. In this study, a biosorbent derived from oak acorn pericarp was prepared through a simple process and characterized using FTIR, SEM-EDX, BET, TGA, XRD, and pHpzc analyses to elucidate its physicochemical properties. The adsorption performance toward rhodamine B was evaluated under varying conditions of pH, contact time, adsorbent dosage, and temperature. Kinetic, isotherm, and thermodynamic analyses were conducted to clarify the adsorption mechanism, while reusability tests assessed its potential for practical wastewater treatment applications.
The concentration of rhodamine B before and after contact with the oak acorn pericarp was measured using a UV-visible spectrophotometer (SHIMADZU UV-1900i). All aqueous solutions were prepared using deionized (DI) water.
After washing, the pericarp was dried in an oven at 378 K for 24 hours. The dried biomass was then ground and sieved to obtain particles with sizes of ≤500 μm, ≤180 μm, and ≤63 μm.
FTIR analysis (Bruker) was performed to identify the functional groups present on the surface of the biosorbent both before and after the adsorption of rhodamine B dye. The infrared spectra were recorded in the range of 400 to 4000 cm−1, allowing for the detection of characteristic vibrational bands associated with key functional moieties involved in the adsorption mechanism.
SEM-EDX analysis (Quattro S) was employed to examine the surface morphology and elemental composition of the biosorbent. Micrographs were taken before and after the biosorption process to evaluate surface texture, porosity, and any structural changes, while the EDX detector provided qualitative and semi-quantitative information on the elemental distribution on the biosorbent surface.
Brunauer–Emmett–Teller (BET) surface area analysis (Quantachrome Autosorb-iQ3) was conducted to determine the specific surface area, pore volume, and pore size distribution of the biosorbent. This analysis was critical for assessing the porous nature and the physical adsorption potential of the material.
Thermogravimetric analysis (TGA, METTLER TOLEDO) was carried out to evaluate the thermal stability and decomposition behavior of the biosorbent. The analysis was performed under a nitrogen atmosphere, with a heating rate of 10 °C min−1 over a temperature range of 20 to 1000 °C. The TGA profile provided insights into moisture content, volatile matter release, and the thermal degradation profile of the biomass.
A series of 50 mL NaCl solutions with initial pH values ranging from 2 to 12 were prepared by adjusting the pH with either 0.1 M Hydrochloric acid (HCl) or 0.1 M Sodium hydroxide (NaOH). Subsequently, 0.1 g of the oak acorn pericarp biosorbent was added to each solution. The suspensions were stirred continuously at 250 rpm and kept at ambient temperature (approximately 25 °C) for 48 hours to reach equilibrium.
After equilibration, the oak acorn pericarp biosorbent was removed by filtration or decantation, and the final pH of each solution was recorded. The difference between the final and initial pH values (ΔpH = pHfinal − pHinitial) was plotted as a function of the initial pH. The pH at which ΔpH equals zero corresponds to the point of zero charge (pHpzc) of the biosorbent.15
For each experiment, a fixed amount of oak acorn pericarp biosorbent (50 mg) with a particle size of ≤63 μm was added to 100 mL Erlenmeyer flasks containing 50 mL of RhB dye solution at an initial concentration of 120 mg L−1. The flasks were sealed with aluminum foil to prevent contamination or evaporation and placed on a magnetic stirrer (DIAB MS-H-S) operating at 300 rpm. Agitation was continued until equilibrium was reached.
The influence of initial solution pH on dye adsorption was studied over a pH range of 2 to 12. pH adjustments were made using 0.1 M HCl or 0.1 M NaOH as required. The contact time between the biosorbent and the dye solution was varied from 5 to 300 minutes to examine adsorption kinetics. Additionally, the effect of the initial dye concentration was evaluated by preparing RhB solutions in the range of 10 to 1000 mg L−1. The influence of salt ionic strength on the biosorption capacity of the oak acorn pericarp was examined by adding different concentrations of sodium chloride (NaCl) and potassium chloride (KCl) to the rhodamine B (RhB) dye solution.
The effect of temperature on RhB adsorption was evaluated by conducting a series of batch experiments at an initial dye concentration of 120 mg L−1 over a temperature range of 20 to 60 °C. These experiments were designed to assess the thermal dependency of the adsorption process and to enable thermodynamic analysis.
All adsorption experiments were carried out in triplicate under identical conditions to ensure reproducibility. The mean values are presented with corresponding standard deviation error bars in all figures, confirming the statistical reliability of the obtained data (Fig. 1).
Following each adsorption test, the mixture was centrifuged at 4000 rpm for 15 minutes using a centrifuge (ONiLAB DM0412). The supernatant was carefully decanted and analyzed to determine the residual dye concentration using a double-beam UV-visible spectrophotometer (SHIMADZU UV-1900i) at the maximum absorbance wavelength of RhB (554 nm).
The amount of dye adsorbed at equilibrium (qe, in mg g−1) and the percentage removal efficiency (R%) were calculated using the following equations:
![]() | (1) |
![]() | (2) |
Adsorption equilibrium data were fitted to several isotherm models, including Langmuir, Freundlich, Temkin, and Langmuir–Freundlich. Kinetic studies were analyzed using pseudo-first-order and pseudo-second-order models. Thermodynamic parameters were evaluated using the Van't Hoff equation to determine changes in Gibbs free energy (ΔG°), enthalpy (ΔH°), and entropy (ΔS°) associated with the adsorption process.
In this study, ethanol was used as an efficient regenerant, where the dye-loaded biosorbent was dried, immersed in ethanol under continuous stirring to promote desorption, rinsed with distilled water to remove residual dye and solvent, and subsequently dried for reuse. This regeneration cycle was repeated to ensure stability and reusability. Ethanol-based regeneration proved effective in restoring the adsorption performance of the biosorbent, thereby extending its service life and offering a cost-effective and sustainable approach for dye removal from contaminated water.18
O stretching), 1250 cm−1 (C–O–C stretching), and 1048 cm−1 (C–O–H bending).19 A band at 1620 cm−1 corresponds to C
C stretching in aromatic compounds, while the absorption at 2919 cm−1 is assigned to symmetric C–H stretching and CH2 vibrations. The peak at 894 cm−1 is associated with C–H deformation in cellulose, a typical feature of the lignocellulosic matrix20
After RhB adsorption, several peaks exhibited slight shifts, including 1620 to 1613 cm−1, 2368 to 2360 cm−1, 2919 to 2912 cm−1, and 3324 to 3318 cm−1. Additionally, a decrease in the intensity of the –OH and C
O bands was observed, indicating their active participation in RhB binding. These spectral changes confirm the involvement of O–H, C
O, and C–H functional groups in the adsorption process and highlight their role in the interaction between RhB dye molecules and the oak acorn pericarp biosorbent.
![]() | ||
| Fig. 3 SEM and EDX of oak pericarp biosorbent before (a and a′) and after (b and b′) RhB dye adsorption. | ||
After contact with RhB dye, notable morphological changes were observed, as shown in Fig. 3(b) (1500× magnification).22 The biosorbent surface became less rough and exhibited reduced heterogeneity, suggesting that the dye molecules occupied and partially blocked the pores.
EDX analysis further revealed significant changes in the elemental composition of the oak pericarp after RhB adsorption. Prior to adsorption, the biosorbent was predominantly composed of carbon, oxygen, and iron, with smaller amounts of aluminum, silicon, and sulfur. Following adsorption, variations in peak intensities were observed, indicating the attachment of dye molecules and possible ion exchange processes at the biosorbent surface.23
The detection of zinc in the EDX spectra after adsorption may result from residual ZnCl2 used during the activation step. Despite extensive washing, trace amounts of zinc can remain weakly bound within the carbon matrix, which may explain its appearance in the spectrum.
![]() | ||
| Fig. 4 Thermogravimetric analysis of oak pericarp biosorbent before (a) and after (b) RhB dye adsorption. | ||
The third stage, occurring between 400 °C and 1000 °C, is attributed to the decomposition of lignin, resulting in a mass loss of 26.98%.24 The decomposition temperatures and relative weight losses are influenced by the intrinsic composition of the material and any chemical treatments applied during preparation.
Furthermore, the high thermal stability observed in the TGA curve supports the amorphous and thermally resistant nature of the biosorbent revealed by XRD analysis, confirming its structural robustness and suitability for adsorption applications under variable environmental conditions.
![]() | ||
| Fig. 5 Nitrogen adsorption – desorption isotherms (a), pore volume distributions (b) of the oak pericarp biosorbent. | ||
The isotherm curve displays a distinct H1-type hysteresis loop, indicating the presence of mesoporous structures within the biosorbent.25 Such pore geometry is advantageous for adsorption processes, as it provides accessible channels for pollutant molecules. The BJH analysis, based on the desorption branch of the nitrogen isotherm, confirmed the mesoporous nature of the oak pericarp.25
Before RhB dye adsorption, the oak pericarp exhibited a specific surface area of 45.416 m2 g−1. After adsorption, the surface area increased markedly to 319.003 m2 g−1. This substantial increase suggests that the adsorption of RhB involves electrostatic interactions and the activation of new surface sites, enhancing the biosorbent's capacity to remove dye molecules from aqueous solutions.22
After adsorption, no significant shift in the diffraction peak positions was observed, indicating that the overall crystalline framework of the biosorbent remained intact. However, a slight decrease in peak intensity can be noticed, which may be attributed to the surface coverage of the biosorbent by the adsorbed metal ions and possible partial filling of the porous sites.27 This behavior suggests that the adsorption process mainly occurs on the surface functional groups and pores of the material without altering its bulk crystallinity.
The XRD results are consistent with the FTIR and BET analyses, confirming that the amorphous structure corresponds to a disordered carbon framework enriched with surface functional groups. These findings are consistent with previous studies reporting that heavy metal ion adsorption onto amorphous carbon-based or polymer-modified composites does not cause major structural transformations but rather induces minor intensity variations due to surface interactions.28,29
At pH values below 4.50, the adsorbent surface carries a positive charge, which leads to electrostatic repulsion between cationic species and the positively charged functional groups, thereby reducing adsorption efficiency. In contrast, at pH values above 4.50, the surface becomes negatively charged, creating favorable electrostatic conditions for the adsorption of cationic species.
Therefore, maintaining the solution pH above the pHpzc promotes a negatively charged adsorbent surface, enhancing electrostatic attraction and maximizing cation removal efficiency.31
In this study, adsorption experiments were conducted at pH 2, which is below the pHpzc value. Under these conditions, the surface of the biosorbent becomes positively charged, favoring the electrostatic attraction of the negatively charged rhodamine B dye molecules. This explains the high adsorption efficiency observed at low pH values.
Furthermore, understanding the pHpzc provides valuable insights for optimizing the adsorption process at a larger scale, allowing for better control of pH conditions to maximize pollutant removal efficiency and ensure consistent performance in industrial applications.
The point of zero charge (pHpzc) of the biosorbent was determined to be 4.5. At pH values lower than the pHpzc, the biosorbent surface is positively charged, whereas at pH values higher than the pHpzc, it becomes negatively charged. RhB has a pKa value of 3.7, meaning that below pKa it is primarily in molecular (cationic/monomeric) form, while above pKa it predominantly exists in ionic or dimeric forms.
Experimental results (Fig. 8) show that RhB removal efficiency reached its maximum (R ≈ 96.52%) at an initial pH of 2 and gradually decreased with increasing pH, reaching a minimum (R ≈ 32.93%) at pH 12. The decrease in removal efficiency at higher pH values can be attributed to the deprotonation of oxygen-containing functional groups (–COOH and –OH) on the biosorbent surface, resulting in negative charges that cause electrostatic repulsion with RhB cations when pH > pHpzc.23
![]() | ||
| Fig. 8 Effect of pH on the removal of RhB dye by oak pericarp biosorbent (C0 = 120 mg L−1; T = 20 °C; t = 300 min). | ||
32Was higher than at other pH values. This suggests that electrostatic interactions were not the sole driving force for biosorption. At low pH, RhB exists mainly in monomeric form, which has a smaller molecular size than the dimeric form, allowing it to diffuse more easily into the micropores of the biosorbent.30 Similar observations have been reported by Deshpande & Kumar (2002),32 and Gad & El-Sayed (2009).33
The results (Fig. 9) show that the dye removal efficiency (R%) increased sharply from 77.79% to 99.47% as the biosorbent mass increased from 10 mg to 250 mg. This enhancement can be attributed to the greater number of available active sites, which facilitated increased dye adsorption. However, the adsorption capacity (qe) exhibited the opposite trend, decreasing with higher biosorbent doses. This inverse relationship is commonly attributed to the aggregation of biosorbent particles at higher dosages, which reduces the available surface area and leads to unsaturated adsorption sites.
![]() | ||
| Fig. 9 Effect of adsorbent dose on the removal of RhB dye using oak pericarp biosorbent (C0 = 120 mg L−1; T = 20 °C; t = 300 min). | ||
Beyond a dose of 50 mg, the increase in removal efficiency became negligible, suggesting that most accessible sites had already been occupied. Therefore, 50 mg was considered the optimum biosorbent dose for RhB removal and was selected for subsequent experiments.34
![]() | ||
| Fig. 10 Effect of contact time on the adsorption capacity of RhB dye at various initial concentrations using oak pericarp biosorbent. | ||
The results indicate that adsorption capacity increases with both contact time and initial dye concentration. The removal process started rapidly due to the abundance of available biosorption sites, but progressively slowed down as these sites became saturated.35 The equilibrium time was reached at approximately 120 minutes for all tested concentrations. Therefore, the 5 hours contact time used in this study was more than sufficient to ensure equilibrium, and the optimal adsorption period was determined to be around 300 minutes.
Moreover, as the initial RhB concentration increased from 10 to 120 mg L−1, the equilibrium adsorption capacity rose from 23.96 to 96 mg g−1, indicating the strong dependence of biosorption efficiency on the initial dye concentration.36
To investigate this effect, adsorption experiments were conducted by varying the concentrations of NaCl and CaCl2 from 0.1 to 1.0 mol L−1 under otherwise identical conditions.
As shown in Fig. 11, the results reveal no significant change in RhB removal efficiency with increasing ionic strength for either salt. This observation suggests that electrostatic interactions are not the dominant driving force in the adsorption process, and that hydrophobic–hydrophobic interactions, if present, are likely weak or masked by other adsorption mechanisms. Similar trends have been reported in previous studies, further supporting the conclusion that ionic strength has a negligible effect on the RhB–adsorbent system under the tested conditions.36
The observed enhancement in adsorption efficiency with increasing temperature suggests that the process is endothermic in nature. Higher temperatures may enhance the mobility of RhB molecules and increase the porosity or flexibility of the biosorbent structure, thereby facilitating the diffusion of dye molecules into internal adsorption sites. This thermal activation may also overcome energy barriers associated with the interaction between dye molecules and the active functional groups of the biosorbent (Fig. 12). The high thermal stability of oak pericarp confirms its suitability for dye removal in variable environmental conditions.19
The Langmuir isotherm describes monolayer adsorption onto a homogeneous surface with identical binding energies.39 Its non-linear form is expressed as follows (Langmuir, 1917):
![]() | (3) |
![]() | (4) |
The Freundlich isotherm (Freundlich, 1907) is expressed as:
| qe = KfCe1/n | (5) |
The Temkin isotherm, which accounts for adsorbate–adsorbent interactions, is represented in its non-linear form as:
![]() | (6) |
The Langmuir–Freundlich isotherm was introduced to describe adsorption on heterogeneous surfaces.41 This model illustrates the distribution of adsorption sites on the adsorbent surface.42 At low adsorbate concentrations, it reduces to the Freundlich isotherm, whereas at high concentrations it approaches the Langmuir isotherm.41 The non-linear form of the Langmuir–Freundlich equation is expressed as:
![]() | (7) |
| Models | Isotherm parameters | |||
|---|---|---|---|---|
| Langmuir model | Qmax | KL | R2 | RL |
| 160.809 ± 13.852 | 0.167 ± 0.034 | 0.988 | 0.0473 | |
| Freundlich model | Kf | 1/n | R2 | |
| 30.101 ± 2.195 | 0.510 ± 0.122 | 0.993 | ||
| Temkin model | Kt | Bt | R2 | |
| 2.786 ± 0.071 | 1.254 ± 0.011 | 0.925 | ||
| Langmuir–Freundlich model | qMLF | KLF | R2 | |
| 27.006 ± 8.427 × 10−16 | 0.167 ± 5.265 × 10−18 | 1.000 | ||
The fitting plots of these isotherm models are shown in Fig. 13, while the calculated parameters for the Langmuir (qm, KL), Freundlich (1/n, Kf), Temkin (Kt, bt) and Langmuir–Freundlich (qMLF, KLF) models, along with their respective R2 values for RhB dye adsorption, are summarized in 15.
![]() | ||
| Fig. 13 Adsorption isotherms of RhB dye on oak pericarp biosorbent at 20 °C, fitted with the Langmuir, Freundlich, Temkin and Langmuir–Freundlich models. | ||
Table 1 presents the adsorption constants obtained from the Langmuir, Freundlich, Temkin, and Langmuir–Freundlich isotherms, along with their corresponding correlation coefficients (R2). The analysis indicates that the Langmuir–Freundlich model provides the best fit to the experimental data (R2 = 1.000), confirming the heterogeneous nature of the adsorbent surface and the occurrence of multilayer adsorption. The Freundlich model also demonstrates a strong correlation (R2 = 0.993), further supporting the assumption of surface heterogeneity and the presence of adsorption sites with different energies. Notably, the Freundlich constant (n > 1) confirms the feasibility and favorable nature of the adsorption process. The Langmuir model, with a relatively high correlation coefficient (R2 = 0.988), suggests that part of the adsorption occurs as a monolayer on homogeneous sites. In contrast, the Temkin model exhibits the lowest correlation (R2 = 0.925), implying that the assumption of a linear decrease in adsorption energy with surface coverage is less applicable in this case. Moreover, for both Langmuir and Freundlich models, the RL and 1/n values fall within the range of (0–1), indicating that the adsorption of rhodamine B onto biosorbent is favorable. Overall, these findings highlight the complexity of surface interactions and confirm the predominance of heterogeneous and multilayer adsorption mechanisms (Fig. 14).
![]() | ||
| Fig. 14 Pseudo-first-order kinetic plot for the adsorption of RhB dye onto oak pericarp biosorbent at 20 °C. | ||
The pseudo-first-order model proposed by Lagergren is expressed by the following equation:
Ln(qe − qt) = ln qe + K1t
| (8) |
| C0 (mg L−1) | qe (exp) (mg g−1) | Pseudo-first-order model | Pseudo-second-order model | ||||
|---|---|---|---|---|---|---|---|
| k1 (min−1) | qe (cal) (mg g−1) | R2 | k2 (g mg−1 min−1) | qe (cal) (mg g−1) | R2 | ||
| 10 | 9.913 | −0.00006575 | 3.562 | 0.915 | 0.0109624 | 9.779 | 0.995 |
| 30 | 25.661 | −0.0001185 | 7.990 | 0.897 | 0.00481782 | 27.034 | 0.997 |
| 60 | 42.830 | −0.00035825 | 40.150 | 0.917 | 0.00229635 | 47.573 | 0.994 |
| 120 | 95.938 | −0.00030425 | 19.592 | 0.977 | 0.00185442 | 100.603 | 0.998 |
As shown in Table 2, the experimental qe values deviate significantly from the calculated ones, suggesting that the pseudo-first-order kinetic model does not adequately describe the adsorption of RhB dye onto oak pericarp biosorbent.36
The pseudo-second-order kinetic model can be expressed as:
![]() | (9) |
![]() | ||
| Fig. 15 Pseudo-second-order kinetic plot for the adsorption of RhB dye onto oak pericarp biosorbent at 20 °C. | ||
Table 2 summarizes the kinetic parameters derived from the pseudo-second-order model at different initial concentrations. The experimental data showed an excellent fit to this model, as evidenced by the high correlation coefficient (R2 > 0.998) obtained for RhB dye. Moreover, the adsorption capacity predicted by the model (qe, cal) was in close agreement with the experimentally measured values (qe, exp) across the four concentrations tested. These findings confirm that the pseudo-second-order model provides a reliable description of the adsorption kinetics of RhB dye onto oak pericarp biosorbent.
The main thermodynamic parameters, namely the standard Gibbs free energy change (ΔG°), the standard enthalpy change (ΔH°), and the standard entropy change (ΔS°), provide valuable insights into the feasibility, spontaneity, and nature of adsorption.44
The standard Gibbs free energy change (ΔG°, kJ mol−1) was determined according to eqn (10):
ΔG° = −RTln Kc
| (10) |
![]() | (11) |
The interrelationship between ΔG°, ΔH°, and ΔS° is expressed as:
| ΔG° = ΔH° − TΔS° | (12) |
![]() | (13) |
Table 3 summarizes the thermodynamic parameters determined in this work. The negative ΔG° values indicate that the adsorption of dye onto the biosorbent occurs spontaneously and is energetically favorable. The positive value of ΔH° demonstrates that the process is endothermic, and its relatively high magnitude suggests that the adsorption occurs mainly through chemisorption. Furthermore, the positive ΔS° values indicate an increase in entropy at the solid–solution interface with rising temperature, thereby supporting the enhanced feasibility of adsorption at elevated temperatures.45
| Temperature (K) | Thermodynamic parameters | |||
|---|---|---|---|---|
| ΔG° (kJ mol−1) | ΔH° (kJ mol−1) | ΔS° (kJ mol−1 K−1) | R2 | |
| 293 | −0.511 | |||
| 303 | −0.972 | |||
| 313 | −2.726 | 41.087 | 150.994 | 0.994 |
| 323 | −3.448 | |||
Despite this decrease, the biosorbent demonstrated easy recovery and relatively high regeneration efficiency, confirming its potential for repeated use. These findings suggest that the biosorbent could be considered an economically viable option for real wastewater treatment applications. Moreover, the spent biosorbent was collected for further investigation in future research.22
Overall, the promising reusability and regeneration performance of the biosorbent not only enhance its economic feasibility but also contribute to sustainable environmental management. By minimizing sorbent consumption and reducing secondary waste generation, the biosorbent represents a potential eco-friendly solution for large-scale wastewater treatment.
O), and aromatic groups, suggesting that hydrogen bonding and π–π interactions play a significant role in dye adsorption. In addition, at acidic pH values (below the pHpzc), the biosorbent surface becomes positively charged, favoring electrostatic attraction between the cationic RhB molecules and negatively charged sites formed through functional group dissociation. The heterogeneous porous structure observed by SEM further enhances dye diffusion and surface accessibility. Overall, the adsorption process is controlled by multiple mechanisms, including electrostatic attraction, hydrogen bonding, and π–π stacking interactions between the aromatic structures of RhB and lignin components of the biosorbent, leading to efficient and stable dye removal.
The adsorption behavior was best described by the Langmuir–Freundlich isotherm, indicating heterogeneous multilayer adsorption, while kinetic modeling highlighted the predominance of a pseudo-second-order mechanism, suggesting chemisorption. Thermodynamic evaluations confirmed that the process was feasible, spontaneous, and endothermic. Importantly, reusability tests demonstrated that the oak pericarp biosorbent maintained substantial adsorption capacity over several regeneration cycles.
The conclusion was rewritten to be more concise, summarizing the main quantitative results, such as the maximum adsorption capacity (160.809 mg g−1), pseudo-second-order kinetics, and Langmuir–Freundlich model fitting.
Overall, the findings highlight oak acorn pericarp as a low-cost, eco-friendly, and sustainable biosorbent with strong potential for wastewater treatment applications. Future work could focus on scaling up the process, testing performance with real industrial effluents, and exploring surface modifications to further enhance adsorption capacity and selectivity.
| This journal is © The Royal Society of Chemistry 2025 |