Doped Biopolymer Composites for Engineered Water Purification: Mechanistic Insights into Multi-Ion Adsorption

Abstract

The presence of heavy metals and fluorides (F^–) in wastewater poses serious environmental and public health challenges. Therefore, in this study, curcumin-doped chitosan-activated carbon (CsAC) biopolymer composites were synthesized and evaluated for the removal of As­(III), Cr­(VI), and F^– ions. The composite containing 5% curcumin (CsACCur5%) exhibited superior adsorption performance, while comprehensive characterization using BET, SEM, FT-IR, XPS, and XRD confirmed successful curcumin incorporation and effective contaminant adsorption. Moreover, CsACCur5% exhibited >95% removal of Cr­(VI) across a wide pH range and 75.8% and 71.8% removal of As­(III) and F^–, respectively, using 0.5 g/L at pH 5. This composite can remove up to 21.81 mg/g Cr­(VI), 0.88 mg/g As­(III), and 44.8 mg/g F^– under optimal conditions. Profound adsorption analyses through three kinetic models and ten isotherm models indicated endothermic and spontaneous adsorption mechanisms. Furthermore, swelling and stability assessments demonstrated reduced hydrophilicity and high reusability, maintaining performance through ten regeneration cycles. Overall, the use of olive-stone-derived AC combined with chitosan and curcumin biomaterials promotes sustainability, resulting in enhanced adsorption efficiency and structural stability of CsACCur, underscoring its potential for sustainable wastewater treatment.

1. Introduction

Water pollution poses a major global challenge that threatens human health and the environment, reducing access to clean water as a result of climate change and the growing need for agricultural production, , especially in countries like India, Bangladesh, China, Vietnam, and Mexico. Chromium (Cr) is one of the most common heavy metals found in waterbodies, especially groundwater. It is a fact that it can be appeared both through natural and anthropogenic activities. The United States Environmental Protection Agency (EPA) has a standard drinking water limit of 0.1 mg/L for total Cr, as the corresponding World Health Organization (WHO) limit is 50 μg/L. Recently, the European Commission will eventually even lower the WHO limit to 0.025 mg/L, which should be further implemented by the next decade..

Moreover, exposure to elevated concentrations of arsenic (As), a highly toxic pollutant, can cause both immediate and long-term health issues, including nausea, cardiovascular diseases, cancer, and, in severe cases, death. , Water supplies usually comprise arsenic, the most common of which are AsO₃ ^3– [As­(III)] and AsO₄ ^3–, where As­(III) is ∼10 times more harmful. For safety reasons, the permissible concentration of As set by WHO is 10 μg/L in drinking water. However, there are countries such as Denmark and Ireland where the relative limits are even lower, i.e., 0.005 and 0.0075 mg/L, respectively. The corresponding arsenic concentrations found in groundwater and surface water are between 100 and 5,000 μg/L.

Groundwater pollution by fluoride (F^–) is recognized as a significant global issue, reporting high levels in countries such as Argentina, Canada, Chile, Hungary, Italy, Pakistan, China, India, Mexico, South Africa, Tanzania, Uganda, etc. Consumption of water contaminated with F^– ions can cause many health problems, related to teeth as well as the skeleton (e.g., osteoporosis and arthritis), and for this reason, WHO suggests a concentration of 0.8–1.5 mg/L in potable water.

A key challenge associated with heavy metals, unlike organic contaminants, lies in their persistence within the environment, as they do not undergo significant modification or decay via natural processes, thus heightening the potential danger they present to ecosystems that are continually exposed to these dangerous elements. Various remediation techniques including adsorption and coagulation/flocculation, as well as ion exchange and chemical precipitation, and membrane filtration , have been utilized to mitigate the harmful effects of heavy metals and fluoride ions.

Adsorption is mainly advantageous because it is cost-effective and selective in removing existing high concentrations of heavy metals from water. Among the adsorbent materials that have been studied are metal oxide nanoparticles, grafted chitosan, graphene oxide nanocomposites, biochar, etc. Moreover, activated carbon (AC)-based adsorbents are highly efficient, have a large surface area and adsorption capacity, and require short contact times.

On the other hand, chitosan (Cs), which is an alternative biopolymer-based adsorbent, has recently attracted much interest. , The −NH₂ and −OH groups of Cs (Figure a) can work as adsorption sites, rendering it a superior adsorbent. However, the use of raw chitosan in batch or column experiments is not widely used due to several significant processing problems, such as poor mechanical properties, easy dissolution in acidic solutions, large swelling ratios, etc., and for these reasons, chitosan can be modified or combined with activated carbon to obtain certain desired properties. The use of typical chitosan-based biosorbents is also limited as they are difficult to separate and recover without the use of high-speed centrifugation or filtration, and thus, it is important to find environmentally friendly raw chitosan biosorbents with superior separation properties and better adsorption capacity. ,

1.

Structure of (a) chitosan (Cs) and (b) curcumin (Cur).

Furthermore, curcumin (Cur) (Figure b) or diferuloylmethane is a naturally occurring hydrophobic low-molecular polyphenolic material obtained from the plant Curcuma longa L. Among other applications, it is used as a reducing and conjugating agent for metal oxide nanoparticles. Recent researches have revealed that encapsulating curcumin in nanoparticles improves its capacity to absorb metal ions as a chelating agent. It was selected over other bioderived polyphenols because it provides a combination of redox-active, metal-chelating, and H-bonding properties through its β-diketone and phenolic hydroxyl units. For instance, tannic acid or humic acid possesses phenolic or carboxylic groups, but tannic acid’s large structure can lead to steric hindrance, , while humic substances are complex compounds. Thus, these can lead to inconsistent performance, making them less efficient for the simultaneous capture of oxyanions and metalloid species. Polyaniline contributes through pH-dependent electrostatic interactions and limited redox behavior but lacks the multifunctional chelating and hydrogen-bonding capabilities offered by Cur.

In the present study, initially, Cs and AC, two of the most researched adsorbent materials, are combined as a continuation of a previous study for the adsorption of As­(III), Cr­(VI), and F^– from water bodies. Several derivatives with various composition ratios, including CsAC1:1, CsAC1:2, and CsAC2:1, were synthesized. To the resulting optimal CsAC ratio, curcumin is added at two different percentages (CsACCur5% and CsACCur10%) and the possible optimization of the properties of the initially produced material by the addition of curcumin was studied. The innovation of this work is the synthesis of a complex adsorbent for the treatment of wastewater containing heavy metals and anions under the guise of a sustainable and effective approach. Furthermore, according to the literature, there has been insufficient research on both the adsorption capacity of curcumin and its combination with Cs or AC, thus providing an assessment of the potential of this blend of materials through this study. In addition, the introduction of curcumin further improves biocompatibility, reduces toxicity, and has anti-infective, immunomodulatory, and anti-inflammatory properties, potentially granting the material disinfectant capabilities. Its combination with Cs and olive-stone-derived AC further promotes sustainability and green principles. The adsorption effectiveness of the composites was assessed by investigating the influence of pH, kinetics, adsorbent dosage, and initial ion concentration. Characterization of the materials’ morphology and structure was carried out using FTIR, SEM, XPS, BET, and XRD techniques. Furthermore, kinetic and isotherm models, along with thermodynamic analyses, were employed to better understand and interpret the adsorption behavior.

2. Materials and Methods

2.1. Materials

Stock solutions (100 mg/L) of As­(III), Cr­(VI), and F^– were prepared by adding a specific amount of AsNaO₂, K₂Cr₂O₇, and NaF (Sigma-Aldrich, Darmstadt, Germany) in DI H₂O and stored. The stock solutions were then diluted to the required concentrations to be used in the experiments. Chitosan (310–375 kDa, DDA > 75%), curcumin (>70%) (Glentham Life Sciences, U.K), glutaraldehyde of 50 wt % in H₂O, acetic acid (≥99%) (Fisher Chemicals, Hampton, New Hampshire, U.S.A.), sodium tripolyphosphate (Na₅O₁₀P₃) (Alfa Aesar, Thermo Fisher Scientific Inc., Massachusetts, U.S.A.) as cross-linker, and activated carbon (AC) were used for the synthesis of the materials. Finally, NaOH ACS reagent, ≥97.0%, pellets, or HCl 37% (Panreac, AppliChem, Barcelona, Spain) diluted solutions were used for pH adjustments. Olive stones (OS) of the Koroneiki variety from the Arkadia Region of Peloponnese were kindly provided by the Arkas olive oil company.

2.2. Synthesis of AC

AC was produced using olive stones (OS) as food waste raw material (derived from the Koroneiki variety from the Arkadia Region of Peloponnese, Greece), which were collected after the milling process. The OS were crushed, sieved to a size of <1 mm, and washed with 10% diluted H₂SO₄ to eliminate inorganic components. The clean sample was impregnated with ZnCl₂ as an activating agent for 24 h in a water bath at 333 K (the ZnCl₂/OS ratio was 1:2 , ). The impregnated sample was dried at 393 K for 1 day before being carbonized at 793 K for 60 min under a N₂ flow of 100 ccm/min and a heating rate of 20 °C per minute. To get rid of impurities and reduce the quantity of ash in AC, the latter was treated with 0.1 N HCl. Then, AC from OS was washed with DI H₂O to the point pH became neutral and dried for 48 h at 120 °C. According to N₂ isotherm (Figure S1a), the following physicochemical properties were obtained: SA_BET: 321 m²/g, DFT pore size: 2.35 nm, V _CUM: 0.377 cm³/g, V _micro: 0.061 cm³/g.

2.3. Synthesis of Composites

2.3.1. CsAc

Regarding the synthesis procedure of chitosan/activated carbon derivatives (CsAC), 1 g of Cs was dissolved in 100 mL of aqueous CH₃COOH solution (2% v/v), based on a previous study. Then, AC was added in various weight proportions regarding Cs; hence, 3 different CsAC solutions were obtained (1:1, 1:2, and 2:1), abbreviated hereafter as CsAC1:1, CsAC1:2, and CsAC2:1. The mixtures were sonicated for 1 h in order to disperse AC. Glutaraldehyde (GLA) and sodium tripolyphosphate (TPP) solution were then poured in each one solution and subsequently stirred again. The resulting gels were placed in the freezer for 1 day and freeze-dried at −104 °C for 2 days. Freeze-drying was used in this study to prevent shrinkage, agglomeration, or densification of the polymeric matrix, which would reduce the surface area and limit adsorption sites. , Finally, the aerogels were ground to powder and treated with a H₂O/MeOH (9:1) mixture in a Soxhlet apparatus for 1 day.

2.3.2. CsAccur

After the adsorptive performance of the aforementioned materials was evaluated, as shown in the following results, the best ratio was found to be 2:1 (CsAC2:1). The latter was modified by the addition of curcumin. Specifically, in a 1% w/v Cs solution, the suitable amount of AC was incorporated, and the mixture was stirred at 50 °C. A Cur suspension was made after dissolution of Cur in acetone at 2 different amounts of 5% and 10% (by wt % to Cs), labeled as CsACCur5% and CsACCur10%, respectively. Then, Cur solution was added dropwise in CsAC suspension, which was then heated at 50 °C under vigorous stirring until acetone was evaporated. To improve the dispersion of the materials, ultrasound treatment for 60 min was followed. Following this, GLA/TPP solution was added, and the resulting gels were stored in the freezer for 1 day. These were then freeze-dried at −104 °C for 2 days and then grounded and cleaned from unreacted impurities via a Soxhlet apparatus using a H₂O/MeOH (9:1) mixture for 1 day.

In Figure , the production procedure of CsAC and CsACCur derivatives is illustrated.

2.

Schematic presentation of the production procedure of CsAC and CsACCur derivatives.

2.4. Analytical Determinations

2.4.1. Cr­(VI) Residual Concentration

The determination of the residual concentration of Cr­(VI) was done photometrically, via the 1,5-diphenylcarbazide photometric method, based on the reaction of acidified samples of Cr­(VI) by H₂SO₄ (ACS reagent, 95.0–98.0%) with 0.5% w/v 1,5-diphenylcarbazide, diluted in pure acetone ≥99.9%. After 15 min of waiting time for the reaction to occur, the absorbance at 540 nm was determined by an ultraviolet–visible (UV–vis) spectrophotometer, leading to the calculation of the residual concentration. With this method, even the minimum detectable concentration of 0.005 mg/L as Cr­(VI) is detected.

2.4.2. As­(III) Residual Concentration

Considering the remaining As­(III) concentration, atomic absorption spectroscopy (Varian Zeeman AA240Z with GTA 120; Hansen Way Palo Alto, CA, USA) was utilized.

2.4.3. F^–Residual Concentration

For the calculation of the remainder F^– concentration, the SPADNS photometric method was employed. SPADNS was mixed with zirconyl chloride (99.99%) and HCl, resulting in a Zr-SPADNS mix solution. Thus, the Zr-SPADNS reagent was mixed with each sample, and the relative absorbance was determined at 570 nm using a UV–vis spectrophotometer.

2.5. Batch Adsorption Experiments

Batch adsorption studies were performed to determine the efficiency of chitosan-, activated carbon-, and curcumin-based composites in removing targeted ions. In summary, 0.5 g/L of the adsorbents (CsAC and CsACCur) was introduced into 10 mL solutions containing As­(III), Cr­(VI), or F^– ions, with initial concentrations of 0.1 mg/L, 0.5 mg/L, and 5 mg/L, respectively, in Falcon tubes. The experiments aimed to optimize both the pH (3.0, 5.0, 7.0, and 9.0 ± 0.1) and the Cs-to-AC ratio. The mixtures were shaken at 80 rpm for 1 day at 293 K with a Trayster overhead shaker and a Loopster rotator. Subsequent trials were conducted using the most efficient materials under optimal pH conditions while varying one factor at a timesuch as adsorbent dosage, initial ion concentration, contact duration, and temperaturekeeping the others constant. After the adsorption process, the solutions were passed through a 0.45 μm nylon cellulose membrane filter for separation.

The percentage removal (R (%)) of As­(III), Cr­(VI), or F^– was determined from the subsequent equation (eq ):

$$R(\mathrm{\%})=\left(\frac{C_0-C_{\mathrm{f}}}{C_0}\right)\times 100\mathrm{\%}$$ 1

where C ₀ and C _f (mg/L) are the initial and final concentrations.

The relative adsorption capacity of each adsorbent (Q _e) (mg/g) was calculated from the following equation (eq ):

$$Q_{\mathrm{e}}=\frac{(C_0-C_{\mathrm{e}})\times V}{m}$$ 2

where C ₀ and C _e (mg/L) are the initial and equilibrium concentrations, respectively. V (L) is the volume of the solution, and m (g) is the mass of the sorbent used.

2.5.1. Adsorption Kinetics

One of the key parameters influencing adsorption efficiency is the contact time, which is explained through adsorption kinetics. To further understand the adsorption mechanism of applied adsorbents for As­(III), Cr­(VI), or F ^– removal, the pseudo-first-order (PFO) (eq ), pseudo-second-order (PSO) (eq ), intraparticle diffusion (IPD) (eq ), and Elovich (eq ) kinetics models were applied to fit the kinetic data.

PFO kinetic model:

$$Q_t=Q_{\mathrm{e}}(1-{\mathrm{e}}^{-k_{1}t})$$ 3

PSO kinetic model:

$$Q_t=\frac{k_2{Q}_{\mathrm{e}}^{2}t}{1+k_2{Q}_{\mathrm{e}}t}$$ 4

where Q _e and Q _t denote the adsorption capacities (mg/g) of As­(III), Cr­(VI), or F^– at equilibrium and at time t, respectively. k ₁ (L/min) and k ₂ [g/(mg/min)] signify the PFO and PSO reaction rate constants, respectively.

Intraparticle diffusion kinetic model:

$$Q_t=K_{\mathrm{I}\mathrm{P}\mathrm{D}}{t}^{0.5}+C$$ 5

K _IPD represents the rate constant (mg/g min^0.5) and the thickness of the boundary layer constant C (mg/g).

Elovich kinetic model:

$$Q_t=\left(\frac{1}{b}\right)\ln (ab)+\left(\frac{1}{b}\right)\ln t$$ 6

where Q _t is the amount of pollutant adsorbed at time t, a is the initial adsorption rate, and b is the Elovich constant.

2.5.2. Adsorption Isotherms

Adsorption isotherms were determined by changing the initial concentration of each of ion across three different temperatures (θ = 30, 40, 50 °C, as 293, 303, 313, and 323 K) over a 24 h contact period. The reason for choosing the above temperatures with a fixed temperature increase of 10 °C instead of arbitrary temperature increases was to enable a meaningful comparison of the adsorption process within the framework of thermodynamics. To investigate both the adsorption equilibrium state of the pollutants and the adsorption characteristics on the surface of chitosan/activated carbon/curcumin derivatives, the Langmuir, Freundlich, Langmuir–Freundlich, Dubinin–Radushkevich, Hill, Redlich–Peterson, Sips, Toth, Fritz–Schluender, and Temkin isothermal models were evaluated. Their equations (eqs to , respectively) are as follows:

$$Q_{\mathrm{e}}=\frac{Q_{\mathrm{m}}{K}_L{C}_{\mathrm{e}}}{1+K_L{C}_{\mathrm{e}}}$$ 7

Q _m (mg/g) is the maximum amount of pollutant adsorbed and K _L (L/mg) is the Langmuir constant.

$$Q_{\mathrm{e}}=K_F{C}_{\mathrm{e}}^{1/n}$$ 8

K _F (mg^1–1/n ·L^1/n /g) is the Freundlich constant and n (dimensionless) is the constant depicting the adsorption intensity.

$$Q_{\mathrm{e}}=\frac{Q_{\mathrm{m}}{K}_{\mathrm{L}\mathrm{F}}{C}_{\mathrm{e}}^{1/b}}{1+K_{\mathrm{L}\mathrm{F}}{C}_{\mathrm{e}}^{1/b}}$$ 9

K _LF (L·mg^–1/b) is the L–F constant and b (dimensionless) is the L–F heterogeneity constant.

$$Q_{\mathrm{e}}=Q_{\mathrm{m}}{\mathrm{e}}^{-K_{\mathrm{D}\mathrm{R}}(RT\ln (1+\left(\frac{1}{C_{\mathrm{e}}}\right)\left)\right)}$$ 10

K _DR (mol²/kJ²) is the mean free energy change constant.

$$Q_{\mathrm{e}}=\frac{Q_{\mathrm{m}}{C_{\mathrm{e}}}^n}{{(K_H+C_{\mathrm{e}})}^{n_H}}$$ 11

K _H (mg/L) is the Hill constant and n _H (dimensionless) is the Hill exponent.

$$Q_{\mathrm{e}}=\frac{K_{\mathrm{R}\mathrm{P}}{C}_{\mathrm{e}}}{1+a{C_{\mathrm{e}}}^{\beta }}$$ 12

K _RP (L/g) and α (mg/g) are Redlich–Peterson constants and β is the Redlich–Peterson exponent.

$$Q_{\mathrm{e}}=\frac{Q_{\mathrm{m}}{K}_S{C_{\mathrm{e}}}^{n_s}}{1+K_S{C_{\mathrm{e}}}^{n_s}}$$ 13

K _S (L/mg) is the Sips’ constant and n_s is the Sips’ exponent.

$$Q_{\mathrm{e}}=\frac{K_T{C}_{\mathrm{e}}}{{(a_T+C_{\mathrm{e}})}^{1/z}}$$ 14

K _T (L/mg) and a _T (mg/L) are the Toth constants and z (dimensionless) is the Toth exponent.

$$Q_{\mathrm{e}}=Q_{\mathrm{m}}{K}_1{C_{\mathrm{e}}}^{m/(1+K_2{C_{\mathrm{e}}}^n)}$$ 15

K ₁ (L/mg) and K ₂ (L/mg) are Fritz-Schluender constants and m (dimensionless) and n (dimensionless) are Fritz–Schluender exponents.

$$Q_{\mathrm{e}}=\frac{RT}{b}\ln (A{C}_{\mathrm{e}})$$ 16

b (J/mol) and A (L/g) are Temkin constants.

2.6. Thermodynamics

The influence of temperature on the adsorption behavior of As­(III), Cr­(VI), and F^– onto the chitosan/activated carbon/curcumin composite surface was examined at 293–323 K under optimal dosage, pH, and contact time conditions. A thermodynamic analysis was carried out to elucidate the characteristics of the adsorption mechanism by examining variations in Gibbs free energy (ΔG ⁰), enthalpy (ΔH ⁰), and entropy (ΔS ⁰). Accordingly, eqs – were applied to determine these thermodynamic parameters.

$$K_{\mathrm{c}}=\frac{C_{\mathrm{s}}}{C_{\mathrm{e}}}$$ 17

$$\mathrm{\Delta }{G}^{\mathrm{°}}=-RT\ln (K_{\mathrm{c}})$$ 18

T (K) is the temperature and R (8.314 J/mol·K) is the universal gas constant.

$$\mathrm{\Delta }{G}^{\mathrm{°}}=\mathrm{\Delta }{H}^{\mathrm{°}}-T\mathrm{\Delta }{S}^{\mathrm{°}}$$ 19

ΔG ⁰ (kJ/mol) is the Gibbs free energy change, ΔH ⁰ (kJ/mol) is the enthalpy change, and ΔS ⁰ (kJ/mol·K) is the entropy change.

$$\ln (K_{\mathrm{c}})=(-\frac{\mathrm{\Delta }{H}^{\mathrm{°}}}{R\mathrm{{\rm T}}})+\frac{\mathrm{\Delta }{S}^{\mathrm{°}}}{R}$$ 20

K _c (mL/g) is the thermodynamic constant.

2.7. Regeneration and Reuse Study

For regeneration-desorption experiments, an exact amount of the optimum obtained material(s) for As­(III), Cr­(VI), and F^– removal was treated by a 0.01 M NaOH solution, as repeatedly found in the literature. , Subsequently, the used material was washed until neutralization of the pH so that the pollutants were extracted from the adsorbent. Following this, the used materials were recharged with As­(III), Cr­(VI), and F^–, and the procedure was repeated up to 10 cycles to evaluate the effectiveness and durability of the material.

2.8. Swelling

Swelling behavior was examined at 2 different pH values (5.0 and 7.0 ± 0.1), corresponding to the optimal pH values identified from adsorption experiments. Each adsorbent sample (0.05 g) was immersed in DI H₂O for 24 h to allow for swelling. The pH of the medium was regulated with 10^–2 mol/L HCl. At specific time intervals, the swollen particles were collected, blotted, and weighed. The swelling percentage (SP, %) was then determined according to eq . M _s (g) is the mass of the swollen sample at time t (min), and M ₀ (g) is the sample’s original mass prior to swelling.

$$\mathrm{S}\mathrm{P}=\left(\frac{M_0-{\mathrm{M}}_{\mathrm{s}}}{M_0}\right)\times 100\mathrm{\%}$$ 21

2.9. Stability

Stability of the adsorbents, especially at extreme pH levels, is critical for efficient adsorption and regeneration procedures. To evaluate pH stability, CsACCur5% was added in multiple aqueous suspensions at different pH (2.0–10.0 ± 0.1) for 1 day. The resulting mass was then dried and measured.

2.10. Characterization Techniques

A variety of analytical methods were used to study the surface morphology, chemical bonding, and functional groups of the prepared materials. The surface area and porosity were measured by using Brunauer–Emmett–Teller (BET) analysis and NOVA Win software (Quantachrome Instruments, version 11.02). Scanning electron microscopy (SEM) was used to evaluate surface morphology utilizing JEOL JSM-6390 LV equipment. Fourier transform infrared spectroscopy (FTIR; PerkinElmer, New York, NY) was used to analyze the functional groups. Crystallinity was determined using X-ray diffraction (XRD) on a Rigaku MiniFlex II diffractometer as well as a Bruker D8 FOCUS for phase analysis. For XPS, Kratos data were collected using a Kratos Axis SUPRA with monochromated Al Kα (1486.69 eV) X-rays at 15 mA emission, 15 kV HT (225 W), and an analysis area of 700 × 400 μm. A BE of 932.6 eV for the Cu 2p3/2 line of metallic copper was obtained by calibrating the device to gold metal Au 4f (83.95 eV) and adjusting for dispersion. At 10 eV pass energy, the Ag 3d5/2-line fwhm was 0.54 eV. Monochromatic Al Kα X-rays have a source resolution of about 0.3 eV. Using the Fermi edge of the valence band for metallic silver, the instrumental resolution was found to be 0.29 eV at a 10 eV pass energy. Using the charge correction technique, the instrument resolution was found to be less than 1.33 eV fwhm on the PTFE. High-resolution spectra of the elements of interest were produced separately using a pass energy of 40 eV, a step size of 0.1 eV, and a dwell period of 280 ms, resulting in a line width of about 0.7 eV for Au 4f7/2. The survey spectrum was acquired using a pass energy of 80 eV. An electron flood gun was used to neutralize charges with a filament current of 0.4 A, a charge balance of 4 V, and a filament bias of 5 V. Successful neutralization was determined by studying the 1s C area, which yielded a strong peak with no lower BE structure. Spectra were charged and adjusted to the lowest BE component of the C 1s spectrum at 284.8 eV, which corresponds to the C–C/C–H bonds. All data were acquired at a base pressure of less than 9 × 10^–9 Torr and a room temperature of 294 K. The sample was placed on the stage before being introduced into the vacuum chamber by pushing the powder into copper tape. XPS analysis was performed at the EPSRC UK National Facility.

3. Results and Discussion

3.1. Effect of Initial Solution pH

Determining the optimal pH is crucial in the adsorption process, as the pH significantly influences the surface charge of the adsorbent and the speciation of the adsorbate. In this study, the impact of pH on the removal efficiency of As­(III), Cr­(VI), and F^– was investigated by varying the solution pH from 3.0 ± 0.1 to 9.0 ± 0.1. The experiments were conducted at room temperature using 0.5 g/L adsorbent and C ₀ of 0.1, 0.5, and 5.0 ppm for As­(III), Cr­(VI), and F^–, respectively.

Apart from the effect of pH, the effect of the ratio between chitosan and activated carbon was also studied in the first phase. Thus, three different CsAC ratios, i.e., 1:1, 1:2, and 2:1, were studied. As shown in Figure , the optimal ratio, at all applied pH values and for all the tested pollutants, was found to be 2:1 (CsAC2:1), meaning that the addition of chitosan in a greater proportion enhances the effectiveness of the adsorbent in the removal of As­(III), Cr­(VI), and F^–. In a previous study of our team, a corresponding conclusion was drawn when applying these materials to the removal of pharmaceuticals. For this reason, this ratio is chosen for the synthesis of the new material presented in this study with the further addition of curcumin (see section 2.2.2), in two different percentages, i.e., CsACCur5% and CsACCur10%. In addition, it is worth noting that in the case of Cr­(VI), all of the applied materials were effective, generally presenting a removal greater than 95%, regardless of the pH value. The common value for water, i.e., pH 7.0 ± 0.1, was selected for further experiments, providing a complete Cr­(VI) removal, especially when CsACCur5% is used (Figure a). Regarding As­(III) and F^– removal, the optimum pH was found to be 5.0 ± 0.1. Specifically, as shown in Figure b, 75.8% of As­(III) was reached by CsACCur5%, leading to a residual concentration of around 24 μg/L. Moreover, CsACCur5% by 71.8% removal, was also optimum for F^– (Figure c), and indeed, its residual concentration reached the ideal value within the permissible limit, i.e., 1.41 mg/L. The second more efficient adsorbent was found to be CsACCur10%, followed by CsAC2:1, and for this reason, these three materials are chosen for further evaluation.

3.

Effect of initial solution pH for the adsorption of (a) Cr­(VI), (b) As­(III), and (c) F^– onto net chitosan, chitosan/activated carbon derivatives, and chitosan/activated carbon/curcumin derivatives (C ₀ 0.1 mg/L for As­(III), 0.5 mg/L for Cr­(VI), and 5 mg/L for F^–; dose 0.5 g/L; 293 K; 24 h contact time). (d) pH_pzc of optimum CsAC2:1, CsACCur5%, and CsACCur10%, by using the pH drift method according to Tai et al. .

Additionally, as illustrated in Figure d, the point of zero charge (pH_pzc), according to Tai et al.’s pH drift method, was determined for optimum CsAC2:1, CsACCur5%, and CsACCur10% materials and found to be 6.58, 6.18, and 6.51, respectively. Hence, for pH values smaller than pH_pzc, the surface of the adsorbents is in its cationic form, and for pH values greater than pH_pzc, it converts negative. Thus, at pH 5.0, which is the optimal pH for the removal of As­(III) and F^–, these materials are positively charged and interact electrostatically with fluoride ions and the small fraction of H₂AsO₃ ^–, leading to attraction. For Cr­(VI), which seems to be less pH-dependent, under acidic conditions, the presented negative anions HCrO₄ ^– may be attracted onto the surface of the adsorbent via electrostatic interactions. On the other hand, at higher pH values, a repulsion occurs due to the rivalry between Cr₂O₄ ^2– and OH^– anions.

3.2. Effect of Dosage

The effect of the adsorbent’s dosage at 293 K, for 24 h, ranging from 0.3 to 1.0 g/L on As­(III) and F^– removal at pH 5.0 and from 0.1 to 0.5 g/L (the range is lower, as assumed from Figure a, that by 0.5 g/L, the removal was 100%) on Cr­(VI) removal at pH 7.0, is shown in Figure . It is evident that by increasing the adsorbent’s quantity, the removal (%) of As­(III) (Figure a), Cr­(VI) (Figure b), as well as F^– (Figure c) increased. Thus, we accomplished a removal percentage of more than 88% (achieving the permissible limit) for As­(III) and around 80% (∼1.0 mg/L residual concentration) in the case of F^– by using 0.8 g/L of CSACCur5%. However, for the adsorption of Cr­(VI), it is noticed that even 0.1 g/L of CSACCur5% can offer almost 80% removal, indicating that it is an effective adsorbent for all the examined pollutants. However, the dosage of 0.5 g/L is chosen for further evaluation of the adsorbents’ performance as it is effective in all cases. The enhanced removal efficiency observed with higher adsorbent quantity can be attributed to the expanded surface area and the increased availability of active functional sites for adsorption.

4.

Effect of dosage for the adsorption of (a) As­(III), (b) Cr­(VI), and (c) F^– onto optimum CsAC2:1, CsACCur5%, and CsACCur10% derivatives (C ₀ 0.1 mg/L for As­(III) (at pH 5.0 ± 0.1), 0.5 mg/L for Cr­(VI) (at pH 7.0 ± 0.1), and 5 mg/L for F^– (at pH 5.0 ± 0.1); 293 K; 24 h contact time).

3.3. Effect of Contact TimeKinetics

Figure illustrates the influence of contact time on the adsorption of As­(III), Cr­(VI), and F^– onto CsAC2:1, CsACCur5%, and CsACCur10% composites over a period ranging from 5 min to 24 h. As shown, the removal efficiency of Cr­(VI) increases rapidly at the first 0–30’, reaching about 75% removal (Figure b) using CsACCur5%. Between 30 and 120 min, the adsorption rate gradually rises from 75% to 95%, after which equilibrium is attained, indicating saturation of the Cr­(VI) adsorption. Moreover, for the adsorption of As­(III) (Figure a) and F^– (Figure c), similarly, after 120 min, the equilibrium is achieved (71% for As­(III) and 70% for F^–), but a satisfactory removal is observed after 60 min, presenting approximately 64% As­(III) removal and 57% F^– removal, by applying the optimum CsACCur5% adsorbent.

5.

Effect of contact time for the adsorption of (a) As­(III), (b) Cr­(VI), and (c) F^– onto optimum CsAC2:1, CsACCur5%, and CsACCur10% derivatives (C ₀ 0.1 mg/L for As­(III) (at pH 5.0 ± 0.1), 0.5 mg/L for Cr­(VI) (at pH 7.0 ± 0.1), and 5 mg/L for F^– (at pH 5.0 ± 0.1); dose 0.5 g/L; 293 K).

The abundance of exposed adsorption sites of CsACCur5% accounts for the initial fast removal (%) (30–60 min). Nonetheless, as reaction time rises, these active spots progressively begin to be occupied, resulting in reduced removal rates due to the effect of mass transfer and diffusion within the particles. Subsequently, an optimal adsorption period of 2 h is selected for the rest of the experiments.

The applicability of 4 nonlinear kinetic models was investigated, displaying the best fit of each material in Figure , and the obtained kinetic data are shown in Table . In the case of the heavy metals, As (III) (Figure a) and Cr (VI) (Figure b), the best model describing the adsorption process is PSO, followed by PFO. This indicates that chemisorption is the driving force for adsorption, including chemical coordination through sharing or exchanging electrons between the adsorbent and heavy metals. The material with the highest removal efficiency in both cases (CsACCur5%) also exhibits the highest k ₂, meaning that it also has the fastest kinetics, reaching equilibrium in a shorter time. Regarding F^– removal (Figure c), the Elovich kinetic model has the highest value of R ², followed by PSO. Generally, this model describes adsorption on energetically heterogeneous surfaces and systems in which the activation energy barrier varies with surface area. This finding suggests that F^– uptake may occur in several consecutive or overlapping phases, such as external or intraparticle diffusion before chemisorption to active sites. Thus, the adsorption kinetics may be influenced by a combination of transport phenomena and heterogeneous surface interactions, resulting in a more complex overall mechanism. Lastly, this time, it is observed that by increasing Cur% in the adsorbents, the adsorption kinetics decreases as it can be seen by the lower K values.

6.

Best-fitted kinetic models for (a) As­(III), (b) Cr­(VI), and (c) F^– onto optimum CsAC2:1, CsACCur5%, and CsACCur10% derivatives (C ₀ 0.1 mg/L for As­(III) (at pH 5.0 ± 0.1), 0.5 mg/L for Cr­(VI) (at pH 7.0 ± 0.1), and 5 mg/L for F^– (at pH 5.0 ± 0.1); dose 0.5 g/L; 293 K).

1. Kinetic Parameters for As­(III), Cr­(VI), and F^– Adsorption onto Optimum CsAC2:1, CsACCur5%, and CsACCur10% Derivatives.

	PFO		PSO		IPD		Elovich
	k 1 (1/min)	R 2	k 2 (L/mg·min)	R 2	K IPD (mg/g·min0.5)	R 2	K (mg/g·s)	R 2
	As(III) removal
CsAC2:1	0.033	0.974	0.026	0.974	0.01	0.890	0. 27	0.956
CsACCur5%	0.040	0.992	0.028	0.997	0.01	0.867	0.028	0.969
CsACCur10%	0.029	0.983	0.021	0.992	0.01	0.936	0.026	0.987
	Cr(VI) removal
CsAC2:1	0.047	0.984	0.061	0.997	0.062	0.796	0.196	0.972
CsACCur5%	0.067	0.968	0.082	0.972	0.060	0.710	0.283	0.926
CsACCur10%	0.030	0.964	0.029	0.980	0.071	0.794	0.070	0.917
	F – removal
CsAC2:1	0.045	0.939	0.011	0.978	0.376	0.906	0.849	0.995
CsACCur5%	0.061	0.957	0.011	0.988	0.509	0.847	0.778	0.993
CsACCur10%	0.043	0.949	0.008	0.980	0.499	0.891	0.710	0.989

3.4. Adsorption Isotherms

The study of adsorption equilibrium data from CsAC2:1, CsACCur5%, and CsACCur10% application was performed using 10 isotherm models for As­(III), Cr­(VI), or F^–. The purpose of this study is to have a better understanding of the adsorption process by finding the best fit. In each case, the best-fitted isotherm model is presented in Figure a-c and the relative parameters are shown in Table . For As­(III) removal, the best fit for all the materials studied is achieved by the Sips model, followed in proximity by Toth and Hill models. The Sips model is a modified Freundlich model which resolves the problem for high pollutant concentrations. Thus, it can be concluded that the adsorption occurs in multilayers with heterogeneous surfaces due to the nonuniform distribution of functional groups.

7.

Best-fitted isotherm models for the adsorption of (a) As­(III), (b) Cr­(VI), and (c) F^– onto optimum CsAC2:1, CsACCur5%, and CsACCur10% derivatives (for As­(III) at pH 5.0 ± 0.1, 90 min contact time; for Cr­(VI) at pH 7.0 ± 0.1, 45 min contact time; and for F^– at pH 5.0 ± 0.1, 120 min contact time; dose 0.5 g/L; 293 K).

2. Isotherm Parameters for As­(III), Cr­(VI), and F^– Adsorption onto Optimum CsAC2:1, CsACCur5%, and CsACCur10% Derivatives.

As(III) removal		CsAC2:1	CsACCur5%	CsACCur10%
Langmuir
Q m (mg/g)		0.33	0.88	0.63
R 2		0.963	0.993	0.982
Freundlich
K F (mg1–1/n L1/n/g)		0.28	0.53	0.34
R 2		0.985	0.967	0.968
Langmuir–Freundlich
Q m (mg/g)		0.57	0.93	0.75
R 2		0.990	0.993	0.982
Sips
Q m (mg/g)		0.41	0.91	0.68
R 2		0.993	0.993	0.987
Hill
Q m (mg/g)		0.57	0.93	0.73
R 2		0.990	0.993	0.982
Toth
Q m (mg/g)		0.10	0.82	0.52
R 2		0.982	0.993	0.987
Redlich–Peterson
K RP (L/g)		3.87	10.70	8.07
R 2		0.986	0.962	0.982
Fritz–Schluender
Q m (mg/g)		0.50	0.25	0.16
R 2		0.985	0.990	0.974
Dubinin–Radushkevich
Q m (mg/g)		0.26	0.72	0.52
R 2		0.795	0.928	0.913
Temkin
A (L/g)		0.21	0.23	0.16
R 2		0.957	0.918	0.920

Cr(VI) removal		CsAC2:1	CsACCur5%	CsACCur10%
Langmuir
Qm (mg/g)		0.97	0.96	0.95
R 2		0.980	0.929	0.963
Freundlich
K F (mg1–1/n L1/n/g)		0.86	0.78	0.61
R 2		0.909	0.977	0.978
Langmuir–Freundlich
Q m (mg/g)		0.96	21.81	1.91
R 2		0.978	0.985	0.989
Sips
Q m (mg/g)		0.977	5.95	1.30
R 2		0.978	0.975	0.979
Hill
Q m (mg/g)		0.97	5.45	1.91
R 2		0.978	0.975	0.979
Toth
Q m (mg/g)		1.10	10.54	1.30
R 2		0.983	0.975	0.978
Redlich–Peterson
K RP (L/g)		8.77	42.30	8.76
R 2		0.982	0.975	0.978
Fritz–Schluender
Q m (mg/g)		2.16	1.14	0.79
R 2		0.978	0.968	0.977
Dubinin–Radushkevich
Q m (mg/g)		0.95	0.87	0.76
R 2		0.981	0.918	0.924
Temkin
A (L/g)		313.07	19363.77	44.73
R 2		0.951	0.899	0.962

F– removal		CsAC2:1	CsACCur5%	CsACCur10%
Langmuir
	Qm (mg/g)	32.93	44.81	33.33
	R 2	0.970	0.990	0.990
Freundlich
K F (mg1–1/n L1/n/g)		3.28	6.64	5.81
R 2		0.939	0.952	0.956
Langmuir–Freundlich
Q m (mg/g)		27.87	40.82	33.09
R 2		0.965	0.989	0.987
Sips
Q m (mg/g)		31.34	43.50	33.70
R 2		0.962	0.988	0.987
Hill
Q m (mg/g)		27.91	40.83	33.09
R 2		0.965	0.989	0.987
Toth
Q m (mg/g)		21.04	45.07	35.49
R 2		0.989	0.994	0.989
Redlich-Peterson
K RP (L/g)		1.36	3.85	3.64
R 2		0.984	0.994	0.989
Fritz–Schluender
Q m (mg/g)		1.40	2.19	2.12
R 2		0.979	0.997	0.993
Dubinin–Radushkevich
Q m (mg/g)		23.17	33.73	26.60
R 2		0.920	0.933	0.901
Temkin
A (L/g)		1.06	1.76	1.95
R 2		0.927	0.960	0.970

In the case of Cr­(VI), the Toth isotherm model fits best, which suggests heterogeneous surfaces and a monolayer adsorption character. On the other hand, when Cur is introduced, the adsorbents adhere to the Fritz–Schluender model, which proposes combined monolayer–multilayer adsorption or higher heterogeneity. This modification would most likely take place through the addition of curcumin’s aromatic and polar functional groups (e.g., −OH, diketones), which alter surface interactions and offer multimechanism pathways such as H-bonding, reduction, or complexation.

For F^–, the shift from Toth for CsAC to Langmuir–Freundlich for CsACCur adsorbents points to deep differences in surface heterogeneity and adsorption mechanisms with Cur addition. CsAC obeys the Toth model, which describes monolayer adsorption on energetically heterogeneous surfaces typical of CsAC systems in which irregular pore structures and diverse functional groups create nonuniform binding sites. The Cur addition leads to the incorporation of polar and aromatic functional groups on the adsorbents’ surfaces, mainly due to β-diketone coordination, thus increasing surface complexity and enabling multimechanistic interactions. Such modification led the CsACCur adsorbents to fit better to Langmuir–Freundlich, a hybrid model which involves monolayer adsorption (Langmuir) and multilayer/heterogeneous adsorption (Freundlich). This change underscores the role of Cur in creating adsorption pathways other than monolayer coverage that are more appropriately described by multiparameter models. These results mirror broader trends under which surface modification with functional additives demands multiparameter isotherm models to provide an effective description of adsorption processes.

Overall, F^– demonstrated the greatest Q _m (44.8 mg/g) over Cr­(VI) (21.81 mg/g) and As­(III) (0.88 mg/g). This could be attributed to the small radii and highest charge density of fluoride ions. Smaller species with lower charge-based repulsion or steric hindrance, such as F^–, with the smallest effective radius and the highest charge, are more likely to diffuse into narrow pores than specific larger heavy metal ions. HCrO₄ ^– may diffuse more easily than the neutral H₃AsO₃ molecule, which is bulkier and lacks the electrostatic attraction that facilitates adsorption.

3.5. ThermodynamicsEffect of Temperature

Figure displays the effect of temperature on the adsorption of As­(III), Cr­(VI), or F^– onto CsAC2:1, CsACCur5%, and CsACCur10%. It is evident that by increasing the temperature, it significantly enhances both As­(III) and F^– adsorption, as the removal efficiencies increased from 60.6% and 67.4% to 90.3% (Figure a) and 79.1% (Figure c), respectively, when the temperature increases from 20 to 50 °C in 2 h by the best CsACCur5% adsorbent. Similar results were also reported on other adsorbents. However, Cr­(VI) shows only a slight increase from 81.6 to 87.6% (Figure b).

8.

Effect of temperature on the adsorption of (a) As­(III), (b) Cr­(VI), and (c) F^– onto optimum CsAC2:1, CsACCur5%, and CsACCur10% derivatives (C ₀ 0.1 mg/L for As­(III) (at pH 5.0 ± 0.1, 90 min contact time), 0.5 mg/L for Cr­(VI) (at pH 7.0 ± 0.1, 45 min contact time), and 5 mg/L for F^– (at pH 5.0 ± 0.1, 120 min contact time); dose 0.5 g/L).

According to thermodynamics, as tabulated in Table , in all cases, the ΔG ⁰ values were calculated to be negative; thus, the adsorption of As­(III), Cr­(VI), and F^– appears to be spontaneous for optimum CsAC2:1, CsACCur5%, and CsACCur10% adsorbents in the present study. Moreover, the positive ΔH ⁰ values for As­(III), Cr­(VI), and F^– were found to be 58.411, 10.305, and 20.453 kJ/mol, respectively, indicative of optimum CsACCur5%, indicating that the adsorption process was endothermic in nature. Moreover, when the ΔS ⁰ value is greater than zero, it reflects an increase in the disorder of the adsorbed molecules at the solid–liquid interface during adsorption.

3. Thermodynamic Parameters for the Adsorption of As­(III), Cr­(VI), and F^– onto Optimum CsAC2:1, CsACCur5%, and CsACCur10% Derivatives.

		ΔG 0 (kJ/mol)
		Temperature (K)
		303	313	323	ΔH 0 (kJ/mol)	ΔS 0 (kJ/mol·K)	R 2
As(III)	CsAC2:1	–0.774	–2.129	–3.484	40.280	0.1355	0.9755
	CsACCur5%	–2.394	–4.400	–6.407	58.411	0.2007	0.9990
	CsACCur10%	–1.809	–3.863	–5.917	60.437	0.2054	0.9956
Cr(VI)	CsAC2:1	–3.616	–3.948	–4.280	6.444	0.0332	0.9278
	CsACCur5%	–4.726	–5.222	–5.718	10.305	0.0496	0.9903
	CsACCur10%	–3.896	–4.592	–5.287	17.169	0.0695	0.9929
F –	CsAC2:1	–0.537	–1.279	–2.021	21.944	0.0742	0.9657
	CsACCur5%	–2.724	–3.489	–4.254	20.453	0.0765	0.9615
	CsACCur10%	–2.129	–2.893	–3.656	20.992	0.0763	0.9585

3.6. Characterization of the Composite before and after Adsorption

Table displays the calculated values of the BET surface area (S _BET), Barrett–Joyner–Halenda (BJH) average pore diameter, and total pore volume. In terms of specific surface area (S _BET), the materials follow the order CsAC2:1 (19.53 m²/g) < CsACCur5% (24.53 m²/g) < CsACCur10% (29.38 m²/g), suggesting that the encapsulation of curcumin slightly enlarges the surface area. S _BET of CsACCur10% is approximately 50% higher than that of the composite without curcumin. However, the addition of curcumin results in a minor reduction in total pore volume (by about 5–10%), likely due to partial pore blockage. However, this does not hinder the efficiency of the adsorbents, based on the adsorptive evaluation, due to the significantly increased S _BET and the introduction of additional functional units from curcumin.

4. Physical Properties of the Optimal Materials.

Physical properties	CsAC2:1	CsACCur5%	CsACCur10%
BET Surface area, S BET (m2/g)	19.53	24.53	29.38
Pore size distribution, PSD (Å)	38.57	35.34	32.08
Total pore volume, V T (cm3/g)	0.061	0.058	0.056

Scanning Electron Microscopy (SEM) was employed to examine the surface morphology of the synthesized composites before and after the adsorption of Cr­(VI), As­(III), and fluoride ions, as shown in Figure (a–l). The surface of AC obtained by ZnCl₂ activation (Figure S1b) appears to be granular with dense pores of different sizes. This morphology has also been observed for other AC materials, where the same activating agent was employed. The CsAC2:1 composite (Figure a) displays an irregular, porous structure with a relatively rough surface, characteristic of chitosan–activated carbon blends. The encapsulation of curcumin into the CsAC matrix results in notable morphological changes. In CsACCur5% and CsACCur10% (Figure b, c), a more compact and slightly smoother surface is observed, suggesting partial surface coating or pore occupation by curcumin molecules, consistent with earlier reports on curcumin-loaded biopolymer systems. , Following the adsorption of Cr­(VI), distinct surface modifications are visible. In the Cr­(VI)-loaded CsAC2:1 (Figure d), the surface appears denser with granular deposits, likely due to the binding of Cr­(VI) species. Similar, yet more pronounced, features are observed in CsACCur5%_Cr­(VI) and CsACCur10%_Cr­(VI) (Figure e, f), where particle agglomeration and film-like deposits suggest enhanced metal uptake, potentially through complexation with curcumin’s β-diketone groups. , Upon exposure to As­(III), the pristine composite (Figure g) exhibits minor textural changes, whereas the curcumin-containing composites (Figure h, i) reveal more evident surface alterations and particulate accumulation. This suggests an improved affinity of the modified materials for arsenic species, likely attributed to additional functional groups provided by curcumin, enhancing binding via hydrogen bonding or coordination. In the case of fluoride adsorption, the surfaces of CsAC2:1_F^–, CsACCur5%_F^–, and CsACCur10%_F^– (Figure j–l) exhibit roughness and pore surface changes. These alterations indicate fluoride ion interactions with hydroxyl and amine groups on the surface, with a more pronounced effect in curcumin-modified materials, suggesting a synergistic enhancement in binding capacity through hydrogen bonding. Overall, SEM analysis confirms the morphological evolution of the composites after curcumin modification and metal ion adsorption, supporting the functional improvements observed in their adsorption behavior.

9.

SEM images of CsAC2:1 (a), CsACCur5% (b), and CsACCur10% (c) adsorbents, before and after Cr­(VI) (d)–(f), As­(III) (g)–(i), and F^– (j)–(l) adsorption, respectively.

All synthesized materials were characterized by using FTIR spectroscopy, as presented in Figure (a–d). The spectrum of the CsAC2:1 composite displays a broad band centered at ∼3260 cm^– 1 due to the overlapping stretching vibrations of −NH and −OH groups present in chitosan. A band near 2800 cm^– 1 corresponds to aliphatic C–H stretching. The peaks at ∼1650 cm^–1 and ∼1580 cm^–1 are assigned to the amide I and amide II modes, respectively. A band at ∼1370 cm^– 1 is indicative of C–N stretching, while a sharp band at ∼1020 cm^– 1 is attributed to the skeletal C–O–C of Cs. The presence of AC in the composite is confirmed by the appearance of bands corresponding to O–H (∼3300 cm^– 1), aromatic CC (∼1590 cm^– 1), and C–O stretching (∼1020 cm^– 1), consistent with the spectrum of AC (Figure S1c) and with previous reports on functionalized carbon materials. For the curcumin-modified composites CsACCur5% and CsACCur10%, the characteristic hydroxyl stretching band of curcumin at ∼3300 cm^– 1 becomes broader and shifts slightly toward higher wavenumbers with increasing curcumin content, indicating enhanced hydrogen bonding interactions between curcumin and the CsAC matrix. In the CsACCur10% sample, the β-diketone CO stretch of curcumin is observed at ∼1620 cm^– 1, along with the aromatic CC stretching near 1600 cm^–1. The increased intensity of the C–O–C band at 1020 cm^– 1 further confirms the successful incorporation of curcumin into the composite structure. Thus, curcumin is stabilized on the surface of the adsorbent through H-bonding and hydrophobic interactions, which comes in agreement with SEM results. FTIR spectra recorded after the adsorption of fluoride (F^–), chromium (Cr­(VI)), and arsenic (As­(III)) are shown in Figure (b–d). The spectrum of the pristine CsAC2:1 composite exhibits all characteristic bands as described above. Upon fluoride adsorption, the O–H/N–H stretching band becomes broader and slightly red-shifted, indicating the formation of H-bonds between fluoride ions and surface hydroxyl and amine groups. In the chromium-loaded samples, a noticeable decrease in the intensity and a slight shift of the amide I band are observed, suggesting coordination of Cr­(VI) ions with nitrogen or carbonyl donor atoms, likely via inner-sphere complexation mechanisms. For arsenic adsorption, shifts and perturbations in the O–H and C–O–C regions are evident, along with the introduction of a weak band at ∼860 cm^–1, which can be attributed to As–O vibrational modes, confirming surface interaction with arsenic oxyanions. In the spectra of CsACCur5% and CsACCur10% post-adsorption (Figure c, d), further broadening of the O–H/N–H bands is observed in all cases, suggesting strong hydrogen bonding with adsorbed species. Additionally, shifts in the amide I and II regions are consistently noted, indicative of coordination interactions between metal ions and functional groups within the curcumin-modified matrix, particularly involving the β-diketone moiety. These spectral changes collectively highlight the critical roles played by hydroxyl, amine, and carbonyl functional groups in the adsorption mechanism and support the hypothesis of a synergistic effect among chitosan, activated carbon, and curcumin in enabling efficient multicontaminant removal from aqueous solutions.

10.

FTIR spectra of CsAC2:1, CsACCur5%, and CsACCur10% adsorbents, (a) before adsorption and (b), (c), and (d) post-adsorption.

The synthesized composites were characterized by XRD, and the results are presented in Figure . All samples display a diffraction peak around 2θ ≈ 20°, corresponding to the (110) plane of Cs crystallites. For CsACCur5% and CsACCur10%, this peak shifts toward higher 2θ values, suggesting a decrease in interplanar spacing, which is attributed to H-bonding interactions between Cs and curcumin molecules. Furthermore, the presence of activated carbon is confirmed by characteristic peaks at 2θ ∼ 23° and ∼ 43°, corresponding to the (002) and (100) planes, respectively. These peaks are present in the AC diffractogram (Figure S1d) and are indicative of disordered graphite-like structures, in agreement with literature reports.

11.

XRD spectra of CsAC2:1, CsACCur5%, and CsACCur10% adsorbents.

XPS was then employed to analyze the chemical surface structure of the optimum CsACCur5% adsorbent. From the survey spectrum, Figure , C, O, N, and P are detected. Using the appropriate relative sensitivity factors (RSFs), the atomic percentage of the elements was calculated from the area of the most prominent peaks after utilizing a Shirley background. The results are listed in Figure .

12.

XPS survey spectrum with the observed peaks and the atom percentage for CsACCur5%.

To gain more information about the chemical speciation of the observed elements, high-resolution scans were recorded. In Figure a, the C 1s region is shown, where a broad peak can be seen. Fitting this region according to ref reveals 4 components at 284.8, 286.3, 287.8, and 288.8 eV attributed to C–C/C–H (36.7%), C–O/C–N (40.5%), CO (18.9%), and O–CO (3.9%), respectively. For the fitting, the fwhm of the peaks was contained to be the same among them, their positions were allowed to fluctuate within 0.2 eV, and the shape used was GL (30). A similar fitting procedure was used for all the deconvoluted spectra. For the O 1s region, Figure b, fitting revealed two peaks at 533.4 and 531.9 eV, attributed to aromatic C–O (84.6%) and CO (15.4%). For the N 1s region, Figure c, two components are distinguished at 400.1 and 402.6 corresponding to N–C (73.4%) and C–NC (26.6%), respectively, and there is no evidence of NO _x species. For P 2p, Figure d, one peak is observed, and the two spin-orbit splitting components are deconvoluted for clarity. The 2p3/2 component is at 133.8 eV, which is indicative of phosphates.

13.

XPS survey spectrum of the (a) C 1s region, (b) O 1s region, (c) N 1s region, and (d) P 2p region and the deconvolution of the spectra for CsACCur5%.

3.7. Swelling

The swelling behavior of the composites was evaluated under the optimal conditions derived from adsorption experiments, in particular at pH 5.0 ± 0.1 for As­(III) and F^– and pH 7.0 ± 0.1 for Cr­(VI). As illustrated in Figure , CsAC exhibited the highest swelling percentages, approximately 1300% at pH 5.0 ± 0.1 and 1050% at pH 7.0 ± 0.1. The incorporation of curcumin significantly reduced the swelling capacity, likely due to its hydrophobic nature. Curcumin interacts with the material surface through hydrogen bonding with chitosan and combined hydrogen-bonding/hydrophobic interactions with activated carbon, thereby limiting the swelling behavior of CsAC. , The aforementioned findings come in agreement with other similar studies. , Another important observation is that when the adsorbents are in cation forms (pH 5.0 ± 0.1), the swelling capacity is much higher than in their anion forms (pH 7.0 ± 0.1). This can be owing to electrostatic repulsion, as well as the increased H-bonding due to the excess amount of H⁺ in the solutions.

14.

Swelling percentage of CsAC, CsACCur5%, and CsACCur10% at pH 5.0 ± 0.1 and 7.0 ± 0.1.

To conclude, based on the analysis of BET and swelling results, it was found that curcumin (i) can partially block some pores, (ii) reduces the swelling % of the material, and (iii) introduces new functional groups, which increase the boundary layer effect shown by the nonzero intercept in the IPD model. As a result, both surface and intraparticle transports control the overall process.

3.8. Stability

Figure a exhibits the stability of CsACCur5% at different pH values. The material is highly stable in the range of pH 5.0–7.0 ± 0.1, which was optimal in the adsorption studies, having <1% mass loss. Even at extreme pH values, it retains its stability and structural integrity, losing 2.38% of its initial weight at pH 2.0 ± 0.1 and 3.67% at pH 10.0 ± 0.1. In Figure b, the stability of the aforementioned material was examined for 10 consecutive cycles at pH 10, which was used as an eluant for the regeneration studies. There is a significant decrease of mass in the first 7 cycles, where almost 12% of it was lost. However, it was then stabilized and kept almost constant until the 10th cycle, with no more than 13% of its initial weight. Previous studies of our group showed that this weight loss can be mainly attributed to Cur that is not well attached to the CsAC surface.

15.

Stability of CsACCur5% (a) at various pH values and (b) for ten cycles at pH 10.0 ± 0.1.

3.9. Regeneration and Reuse

The regeneration analysis of the best adsorbent CsACCur5% for all three target pollutants was conducted to determine the number of times the adsorbent may be reused. The adsorption was done under optimum circumstances (0.5 g/L CsACCur5% was mixed with 0.1 mg/L for As­(III) at pH 5.0 ± 0.1, 0.5 mg/L for Cr­(VI) at pH 7.0 ± 0.1, and 5 mg/L for F^– at pH 5.0 ± 0.1, for 90 min of contact time). Ten cycles were performed, and after each cycle, the utilized adsorbent was rinsed with 0.01 M NaOH to ensure desorption. This desorbed adsorbent was then rinsed with DI H₂O, and this helped to remove any remaining molecules from the pollutant, allowing the reuse of CsACCur5%. The variation in the adsorption efficiencies of CsACCur5% with each desorption–resorption cycle is depicted in Figure . From the diagram, it can be concluded that the adsorption efficiency shows a gradual decrease with successive cycles. The adsorption efficiency decreased from 67.7%, 99.6%, 66.8% in the first cycle to 49.2%, 65.4%, and 51.3% in the 10th cycle for As­(III), Cr­(VI), and F^–, respectively. Therefore, the decrease in the efficiency of these levels, i.e., 27%, 34%, and 39% for As­(III), Cr­(VI), and F^–, respectively, was probably due to the morphological degradation of the adsorbent surface with successive adsorption–desorption cycles. This could also be explained considering both the stability test presented in Figure , where there was a significant decrease in mass in the first 7 cycles, and the desorption studies, where after the sixth cycle, the desorption efficiency decreases significantly for all of the pollutants. It is therefore not a coincidence that up to 7 cycles the decrease was only 15–20%, and after the seventh cycle, this decrease appeared more pronounced.

16.

Desorption and recyclability studies for ten cycles after regeneration of CsACCur5% for (a) As­(III), (b) Cr­(VI), and (c) F^– by using 0.01 M NaOH for 1.5 h.

3.10. Competitive Adsorption, Real Wastewater

It is well-known that multiple anions generally coexist in real water and compete with each other for adsorption sites. The latter is mainly determined by the relative ion concentration and its affinity for the adsorbent, which is fundamentally connected to the ion radius and charge. For this reason, experiments were conducted with the simultaneous presence of Cr­(VI), As­(III), and F^– in pairs, as well as all 3 ions together, and the results are presented in Figure a–c. As can be seen, Cr­(VI) and As­(III) relative removal was by the simultaneous presence of other anions, while F^– removal was significantly hindered. In particular, Cr­(VI) was increased to >98% in all cases, while As­(III) surpassed 80% removal efficiency. In Cr­(VI)/As­(III) systems, a synergistic effect was obvious. On the other hand, F^– performance dropped from 71% to less than 60% in all cases, hinting an antagonistic effect with the other two ions.

17.

Effects of coexisting ions on the removal of (a) Cr­(VI), (b) As­(III), (c) F^– (using 1 mg/L Cr­(VI), 0.05 mg/L As­(III), and 2 mg/L F^– at pH 5 and 120 min contact time), and (d) real wastewater treatment.

Additionally, experiments using real textile effluent were kindly provided by a local textile dyeing industry (Thessaloniki, Greece). The physicochemical characteristics of the effluent used are pH 8.3 ± 0.1, a chemical oxygen demand (COD) of 1019 ± 90 mg/L, total dissolved solids (TDS) of 830 ± 25 mg/L, a turbidity of 120 ± 10 NTU, and a conductivity of 9341 ± 125 μS/cm. As observed in Figure d, CsACCur5% achieves a fast removal of 40% within 15 min, which gradually reaches 53% within 120 min. This result highlights the successful applicability of the particular adsorbent in both organic and inorganic pollutants, as it contains both functional groups and hydrophobic regions.

3.11. Adsorption Mechanism

Based on the characterizations and batch experiment data, the mechanism of Cr­(VI) removal by CsACCur5% was proposed, as illustrated in Figure . The process involves three main steps: (1) electrostatic attraction, where negatively charged Cr­(VI) ions are drawn to the positively charged surface of CsACCur5%; (2) reduction of the adsorbed Cr­(VI) to Cr­(III) under acidic conditions, facilitated by π-electrons; and (3) complexation of Cr­(III) with the hydroxyl and carboxyl functional groups present on CsACCur5%.

18.

Proposed removal mechanism of Cr­(VI), As­(III), and F^– by CsACCur5%.

Regarding As­(III), batch experiments showed that the driving force for adsorption is chemisorption, possibly through H-bonding and a small contribution from electrostatic interactions. At pH 5, As­(III) exists mostly as H₃AsO₃ and a small fraction as H₂AsO₃ ^– . Also, other weaker interactions are present that include ion pair formation, as well as metal chelation due to the addition of Cur. In addition, isotherm studies revealed that adsorption occurs in multilayers.

In the case of F^–, the experiments demonstrated that the adsorption efficiency was strongly influenced by pH, with acidic environments promoting greater removal of the contaminant. At these pH values, the adsorbents get positively charged and can interact mainly with fluoride ions through electrostatic attraction between F^– and NH₂ ⁺, COOH₂ ⁺, OH⁺, as well as ion exchange and ion pair interaction. Also, other minor interactions are present, such as H-bonding and ligand exchange with curcumin and phosphate units of adsorbent.

3.12. Comparison with Literature

In Table , the performance of the optimized nanocomposite CsACCur5% synthesized in this study is compared with that of similar chitosan-based materials from previous studies in the recent literature. The comparative results display that CsACCur5% shows a good performance for the treatment of Cr­(VI) and F^– in contaminated water, providing high removal rates, using a low dose of the adsorbent. Nevertheless, the performance toward As­(III) is significantly lower than that reported in the literature for chitosan-based materials, such as Fe₃O₄/Cs (21.04 mg/g) or GO/goethite-Cs (368.38 mg/g). This is likely due to As­(III) being predominantly neutral at the optimal pH, thereby hindering electrostatic interactions, especially with the protonated amine groups of chitosan. It is important to note that the initial objective of this investigation was to develop a chitosan-based adsorbent that is biocompatible and effective toward various pollutants; this was the reason that not only As­(III) was studied here but also Cr­(VI). Also, this study is focused holistically on the synthesis and application of those materials and not specifically focused on capacity.

5. Comparison of the Synthesized Adsorbents in the Current Study with Recent Literature.

Adsorbent	C 0 (mg/L)	Dosage (g/L)	pH	Removal (%)	Q m (mg/g)	ref.
As(III) removal
(net-CS)-g-NVCL/DMAAm hydrogels are	0.013	40.0	7.6	46.0	0.002
Fe3O4/CS Nanocomposites	10.0	0.5	4.0	96.7	21.04
Cs functionalized iron nanosheet	40.0	0.4	7.0	85.0	108.60
Mg-Si-La@AC	0.10	1.5	7.0	93.0	0.32
Fe3O4/MgO/AC	6.0	0.13	7.0	96.6	1.16
MIL-88B/MnO2/GAC	10.0	0.4	2.0	88.0	15.13
GO/goethite-CS	50.0	1.0	6.0	98.0	368.38
UiO-66-NH2-Cl@PCB	1.0	1.0	7.0	92	58.09
CsACCur5%	0.10	0.5	5.0	75.8	0.88	This study
Cr(VI) removal
SA/PVA/CS (SPC) composite	10.0	2.0	3.0	89.2	37.74
CS-TFT	100.0	0.2	3.0	90.0	265.96
GAC	1000.0	10	3.5	-	45.20
TAC/Fe	50.0	0.6	2.0	99.6	68.49
Fe3O4@AC	100.0	2	2.0	86.0	68.60
Fe3O4@CS-APTMS	200.0	1.0	2.0	80.0	269.54
chitosan-allophane	70.0	0.6	3.0	98.6	112.17
MWCNT–COS	1.0	0.5	3.0	100.0	367.60
CsACCur5%	0.5	0.5	7.0	100.0	21.81	This study
F – removal
calcium-phosphate-chitosan	5.0	10.0	5.2	53.5	132.25
(net-CS)-g-NVCL/DMAAm hydrogels are	0.4	40.0	7.6	14.0	0.15
AC-Al(OH)3	-	1.0	6.15	99.0	20.00
PAC	1.0	0.2	-	61.0	12.50
Al-MOFs	50.0	1.0	6.0–7.0	99.6	37.20
D-Fe3O4-NC	10.0	0.2	7.0	95.0	-
CFeAlB	30.0	1.0	3.5	96.6	71.9
BTCA-Zr	2.0	0.2	2.0	96.9	86.0
CsACCur5%	5.0	0.5	5.0	71.8	44.80	This study

Overall, CsACCur5% has various benefits as an adsorbent, including high adsorption capacity for Cr­(VI) and F^–, due to the abundance of functional groups and sustainable, low-cost synthesis utilizing olive stones. It also has the potential to be selective for many pollutants, including both inorganic and organic contaminants, since it contains abundant hydrophilic and hydrophobic regions. This was validated by the successful application of this material in the purification of real wastewater from an industrial textile effluent. However, limitations include lower stability at extreme pH due to partial curcumin leaching. This could be fixed by various methodologies that include grafting on biochar via silanization, metal-organic frameworks using triethylamine or even 3D materials by employing ferrocyanide and pyrolysis. Another innovative approach is also to form metal-curcumin oxide complexes.

Despite these shortcomings, the composite is a promising, low-cost, and adaptable alternative for multicontaminant removal, with room for improvement. Therefore, CsACCur5% proves to be an optimal choice due to its ability to remove various pollutants, thus making it a useful adsorbent material for wastewater treatment.

4. Conclusions

The primary goal and novelty of this study were the development of a specialized adsorbent designed for the efficient removal of various hazardous pollutants, including heavy metals and anions, from wastewater, aiming for a sustainable and effective treatment strategy. Therefore, the CsACCur adsorbents were created by combining initially activated carbon originating from olive stones, which are food wastes, with chitosan in a 2:1 optimal ratio and thereafter by doping curcumin in their structure. Hence, the 5 and 10% curcumin-doped composite biopolymers were examined for the effective adsorption of As­(III), Cr­(VI), and F^– from contaminated water. CsAC and CsACCur materials characterization by BET, SEM, XRD, XPS, and FTIR verified a porous surface with enhanced physicochemical properties, as well as confirmed the deposition of Cur in the structure.

Experimental results indicated that CsACCur5% was the most effective adsorbent for all target pollutants, achieving nearly complete removal of Cr­(VI) across all tested pH values, along with the adsorption of 75.8% As­(III) and 71.8% F^– at pH 5.0 ± 0.1, using 0.5 g/L adsorbent at 293 K over 24 h. Increasing the adsorbent dosage further enhanced the removal of As­(III) and F^–, allowing the concentrations to reach the permissible limits set by the WHO.

Four kinetic models were tested to the obtained data, and it was found that in the case of the heavy metal removal, i.e., As­(III) and Cr­(VI), the best model describing the adsorption process is PSO, indicating that chemisorption is the driving force for adsorption. On the other hand, regarding F^– Elovich kinetic model fitted the results better, assuming again chemisorption. In addition, 10 isothermal models were applied.

For As­(III) removal, the Sips model provided the best fit, indicating heterogeneous multilayer adsorption. In the case of Cr­(VI), the Toth isotherm model was the most suitable, suggesting monolayer adsorption on heterogeneous surfaces. For F^–, the shift from the Toth model for CsAC adsorbents to the Langmuir–Freundlich model for CsACCur adsorbents reflects changes in surface heterogeneity and adsorption mechanisms with the incorporation of curcumin.

Thermodynamic analysis showed negative ΔG ⁰ values for As­(III), Cr­(VI), and F^–, indicating that the adsorption process is spontaneous. Positive ΔH ⁰ values (58.411, 10.305, and 20.453 kJ/mol for As­(III), Cr­(VI), and F^–, respectively, using optimal CsACCur5%) suggest that the adsorption is endothermic. Regeneration and reuse of CsACCur5% up to ten cycles confirmed the effective performance of this composite for all three pollutants, eventually up to seven cycles, while the stability test showed a significant mass loss in the first 7 cycles, where almost 12% was lost. Finally, it was hypothesized that the addition of Cur reduced the swelling capacity of CsAC, possibly due to the hydrophobic properties of Cur.

Data will be made available on request.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.iecr.5c02848.

• N₂ isotherm of AC, the SEM image of AC, FTIR of AC, XRD of AC (Figure S1) (PDF)

K.N.M. and A.K.T.: Writingreview and editing, Writingoriginal draft, Methodology, Investigation. K.L., C.D., and M.A.I.: Writingoriginal draft, Methodology, Investigation. S.N., I.A.K., and G.Z.K.: Writingreview and editing, Writingoriginal draft, Supervision, Methodology, Investigation.

The open access publishing of this article is financially supported by HEAL-Link.

The authors declare no competing financial interest.