Magnetic chitosan kaolinite cellulose nanofibril cryogel beads for efficient removal of methylene blue from water

Abstract

Using a green freeze gelation and genipin crosslinking technique, chitosan (C), kaolinite (K), cellulose nanofibrils (CNF), and Fe₃O₄ nanoparticles were combined to create CK–CNF–Fe cryogel beads, a novel magnetic composite. Freeze gelation was used to create the beads, which were then cross linked using genipin, a non-toxic natural cross linker, and Fe₃O₄ nanoparticles were included in situ. The addition of genipin guarantees the material’s safe application in wastewater treatment while also enhancing its eco-friendliness. With an average pore width of about 11 μm, an average bead diameter of 1.7 mm, and a specific surface area of 230 m² g⁻¹, the beads displayed a macroporous three dimensional structure. The maximum adsorption capacity was 812 mg g⁻¹ at pH 7 and 298 K in batch adsorption studies using Methylene Blue (MB, 150 mg L⁻¹), which fit the Langmuir isotherm model (R² > 0.995). 95% equilibrium absorption was reached in 22 min, and adsorption proceeded according to pseudo second order kinetics (k₂ = 0.016 g mg⁻¹ min⁻¹). Using a 0.3 T magnet, the beads’ great magnetic responsiveness (saturation magnetization, Mₛ = 20 emu g⁻¹) allowed for > 99% recovery in less than 35 s. >90% dye desorption during regeneration at pH 2 was made possible by the pH-responsive swelling behavior, which ranged from 500% (pH 3) to 880% (pH 10). Surprisingly, after six adsorption desorption cycles, the beads retained more than 90% of their adsorption effectiveness. These results highlight CK–CNF–Fe cryogel beads as eco-friendly, high performance, and magnetically retrievable adsorbents with excellent reusability for sustainable wastewater treatment.

Introduction

One of the most pressing environmental issues facing the world today is water contamination, which is fueled by fast urbanization, industrialization, and population expansion. Heavy metals, pesticides, pharmaceuticals, and synthetic dyes are just a few of the many dangerous substances found in industrial effluents. These substances are classified as emerging pollutants because of their toxicity, persistence, and limited removal by traditional wastewater treatment techniques^1,2,3,4. Among these, synthetic dyes are especially hazardous since the textile, printing, paper, leather, and plastic industries release significant amounts of them into the environment, poisoning aquatic systems all over the world^5,6,7. More than 100,000 commercial dyes are currently in use, with over 700,000 tons produced annually⁶. Of this, approximately 10–15% is lost during the dyeing process and released into water streams⁷. The high visibility, chemical stability, and resistance of these dyes to biodegradation make them serious environmental contaminants. Methylene Blue (MB), one of the most widely used cationic dyes, poses a severe ecological and health risk due to its mutagenic, cytotoxic, and carcinogenic effects, even at trace concentrations^5,6,7,8. The persistence of MB in water bodies not only reduces light penetration and disrupts photosynthesis in aquatic plants but also bio accumulates in organisms, leading to long-term toxicity^5,6,7,8,9,10. Dye removal has been investigated using a variety of physical, chemical, and biological techniques, including coagulation, membrane separation, advanced oxidation, and biodegradation¹¹. These methods can lower dye concentrations, but they frequently have complicated operations, high operating costs, incomplete removal, or secondary pollution¹². Because of its ease of use, great effectiveness, and capacity to eliminate colors even at low concentrations, adsorption has become the most promising method¹³. Despite their high adsorption capacity, conventional adsorbents including zeolites, silica, and activated carbon have drawbacks such high cost, low recyclability, and decreasing performance upon regeneration^11,12,13. In recent years, biopolymers have gained attention as sustainable and eco-friendly alternatives for wastewater remediation^14,15,16. Because of its biodegradability, environmental compatibility, and abundance of amino and hydroxyl groups that act as active binding sites for dye molecules, chitosan (CS), a naturally occurring amino polysaccharide derived from chitin, is frequently employed as a main matrix in adsorbent design^17,18,19. Furthermore, CS has a high capacity for chelation, film formation, and pH responsiveness^20,21,22. However, low porosity and weak mechanical strength limit its practical use^20,21,22. Biopolymer-based composite adsorbents have received more attention recently because of their adjustable functionality, environmental compatibility, and renewability. Cellulose nanofibrils (CNFs), chitosan, and kaolinite. In order to improve mechanical strength, thermal stability, and adsorption capacity, inorganic clay minerals such kaolinite (K) have been added to polymeric matrices^23,24,25. Kaolinite is a naturally occurring layered aluminosilicate that improves composites’ mechanical strength, thermal stability, and adsorption capacity. Its high surface area, cation exchange capacity, and acid/base stability make it a perfect companion for CS in dye adsorption^26,27. In addition, cellulose nanofibrils (CNFs) are incorporated as renewable, highly dispersible reinforcing agents that provide significant structural and functional benefits. Their high mechanical strength and ability to integrate into polymeric networks enhance the overall stability of the composite, while their nanoscale dimensions increase porosity and water permeability. These features accelerate mass transfer processes, leading to faster dye adsorption kinetics and improved overall adsorption efficiency^28,29,30. Reports indicate that CS–K and CS–CNF composites achieve adsorption capacities in the range of 200–500 mg g⁻¹ for cationic dyes^24,25,26.Despite the advancements, most biopolymer–clay composites still require lengthy separation processes post-adsorption, limiting their practical applications^28,29,30,31. Lastly, iron oxide nanoparticles (Fe₃O₄) are added to the composite to provide it magnetic activity. Using an external magnetic field, magnetic modification makes it possible to quickly and effectively recover adsorbents from treated water, doing away with the need for laborious post-treatment procedures like centrifugation or filtering. Fe₃O₄ is a highly successful technique for sustainable wastewater treatment since it not only simplifies the separation process but also improves reusability by increasing regeneration efficiency and operational stability during numerous adsorption desorption cycles. For example, Fe₃O₄–CS composites have demonstrated adsorption capacities of 400–600 mg g⁻¹ for MB with magnetic recovery durations of less than a minute^32,33,34. Nevertheless, Fe₃O₄ aggregation, decreased surface activity, and efficiency loss across several cycles continue to be problems. Despite these developments, a number of issues are still unresolved. While many biopolymer clay composites are environmentally beneficial, they have low mechanical stability and require lengthy separation stages after adsorption. Conventional adsorbents are expensive and have poor regeneration. Even with the addition of magnetic nanoparticles, issues including aggregation, decreased surface activity, and inadequate recyclability limit their long-term and practical use. An environmentally benign, physically sound, and magnetically retrievable adsorbent that combines high adsorption capacity with quick separation and consistent performance across several regeneration cycles is therefore desperately needed. In this context, the present work develops CK–CNF–Fe cryogel beads as a multifunctional solution designed not only for high capacity dye adsorption but also for excellent recyclability and easy magnetic recovery, directly addressing the long-standing bottleneck of poor regeneration in existing adsorbent systems^35,36,37,38. The work describes a novel magnetic cryogel bead composite (CK–CNF–Fe) developed by a green freeze gelation process using genipin as a non toxic crosslinker. The beads combine the structural and adsorptive properties of chitosan, kaolinite, and cellulose nanofibrils with the magnetic responsiveness of Fe₃O₄.The beads possess: Highly macroporous 3D architecture with tunable pH responsive swelling property. MB adsorption kinetics are extremely quick, reaching > 95% equilibrium in less than 25 min. high adsorption capacity (812 mg.g⁻¹) at neutral pH, outperforming the majority of existing systems. less than 35 s of magnetically induced separation, with easy recovery and reuse. Long term operational viability by consistent performance throughout the course of six regeneration cycles. This multipurpose material combines adsorption, separation, and regeneration on a single platform to provide a high-efficiency, environmentally safe dye remediation solution. The development, characterization, and evaluation of an environmentally friendly, high performing, and magnetically retrievable adsorbent system CCK–CNF–Fe cryogel beads for the quick and efficient removal of Methylene Blue from aqueous solution is the main goal of this work. Specific goals are, Structural and morphological characterization of the beads by BET, SEM, EdS, and swelling analysis. Exploration of adsorption performance, including isotherm modeling, kinetics, and thermodynamics. Investigation of pH responsiveness and regeneration efficiency to determine practical recyclability. Examination of magnetic recovery performance for enabling low energy and fast separation. Through the incorporation of green materials with smart functional design, this work enhances the advancement of next-generation adsorbents for sustainable wastewater treatment applications.

Methods and experiments

Materials

Every chemical utilized in this work was analytical grade and didn’t require any additional purification. Sigma Aldrich (USA) provided chitosan (CS), which has a middle molecular weight of around 190–310 kDa and an approximate degree of deacetylation of 85%. El Nasr Company for Mining and Chemical Industries (Egypt) supplied the kaolinite clay (K), which was ground and sieved to produce particles smaller than 75 μm. According to recorded data, the kaolinite’s normal chemical composition was Al₂O₃ (39–40%), SiO₂ (46–48%), Fe₂O₃ (0.5–1.5%), TiO₂ (0.5–1%), and trace alkali oxides. Using high-pressure homogenization, cellulose nanofibrils (CNFs) with diameters of 5–20 nm and lengths of several micrometers were extracted from bleached kraft pulp (Misr Edfu Pulp & Paper Co., Egypt).For the preparation of magnetic nanoparticles, ferric chloride hexahydrate (FeCl₃•6 H₂O, ≥ 99%) and ferrous sulfate heptahydrate (FeSO₄•7 H₂O, ≥ 99%) were purchased from Merck (Germany) and used for in situ co-precipitation synthesis of Fe₃O₄. Genipin, a natural nontoxic crosslinking agent, was supplied by Wako Chemicals (Japan). Ammonium hydroxide solution (25%), acetic acid (glacial), and Methylene Blue (MB) dye were purchased from El-Gomhouria Chemical Co. (Egypt). Deionized water (conductivity < 1 µS/cm) was used for all solution preparation and washing steps.

Preparation of CK–CNF–Fe cryogel beads

The CCK–CNF–Fe cryogel beads were fabricated using a freeze gelation technique, followed by genipin crosslinking and in situ magnetic nanoparticle incorporation. The synthesis was carried out in the following steps:

Preparation of the polymer clay suspension

To create a homogeneous solution, 2% w/v chitosan was dissolved in 1.5% (v/v) acetic acid and stirred magnetically for 12 h at room temperature. For consistent dispersion, kaolinite powder was individually disseminated in 1% w/v deionized water and sonicated for 30 min. The chitosan solution was gradually mixed with kaolinite slurry for four hours.

In the meantime, bleached kraft pulp was homogenized under high pressure to create cellulose nanofibrils (CNFs, 05% w/v), which were then added to the chitosan–kaolinite mixture. To guarantee full dispersion, the resulting ternary suspension was constantly agitated for six hours.

In situ synthesis of magnetic nanoparticles

Co precipitation was used to create Fe3O4 nanoparticles in situ in the polymer–clay–CNF matrix. In a nitrogen environment at 80 °C with vigorous stirring, a freshly made solution of FeCl₃•6 H₂O and FeSO₄•7 H₂O (molar ratio 2:1, total iron content 0.3 mol/L) was added dropwise to the suspension. Fe3O4 nanoparticles were then produced by adding ammonium hydroxide (25%) until the pH reached 10. Black precipitates emerged when the reaction was maintained for one hour, indicating that magnetite production and immobilization in the matrix were successful.

Bead formation and crosslinking

The final composite suspension was loaded into a syringe and dropwise extruded into liquid nitrogen to promote rapid freezing and immediate bead formation. The frozen beads were subsequently placed in a 1% (w/v) genipin solution and incubated for 24 h at 4 °C to permit thorough crosslinking. Genipin is a nontoxic cross linker that stabilizes the chitosan network and improves bead mechanical integrity.

Freeze-drying and storage

After crosslinking, the beads were washed thoroughly with deionized water to remove excess chemicals and then freeze dried for 48 h. The cryogel beads produced were spherical, had an average diameter of ~ 1.7 mm, and a porous morphology suitable for adsorption. The product, designated as CK–CNF–Fe, was stored in a desiccator until further characterization and use as shown in Fig. 1.

Fig. 1

The Schematic procedure for Preparation of CK–CNF–Fe Cryogel Beads.

Batch adsorption experiments

The effectiveness of CK–CNF–Fe cryogel beads toward Methylene Blue (MB) was assessed using batch adsorption experiments. Experiments were carried out in 250 mL conical flasks with a fixed adsorbent dosage of 0.05 g and 100 mL of dye solution at the necessary concentrations, unless otherwise specified. To guarantee even mixing, the solutions were shaken at 150 rpm in a thermostatic orbital shaker at 298 ± 1 K.

While the influence of solution pH was examined between pH 2 and 10, adjusted using 0.1 M HCl or 0.1 M NaOH, the effect of starting dye concentration was examined in the range of 50–500 mg L⁻¹. Aliquots were taken out at predefined intervals throughout contact time trials that lasted up to 180 min. Adsorption studies were also carried out at 298, 308, and 318 K to assess thermodynamic characteristics.

Following adsorption, a 0.3 T external magnet was used to magnetically separate the suspensions, and UV Vis spectrophotometry (Shimadzu UV-2600, λ = 664 nm) was used to examine the supernatants. The starting and equilibrium concentrations were used to calculate the removal efficiency (%) and adsorption capacity (qₑ, mg g⁻¹).

Characterization techniques

The structural, morphological, chemical, and magnetic properties of the synthesized CK–CNF–Fe cryogel beads were characterized using a range of analytical techniques:

Morphology and surface structure

Scanning Electron Microscopy (SEM) (JEOL JSM-6510, Japan) was used to examine the surface morphology and internal structure of beads at an accelerating voltage of 15 kV. For improved conductivity, cryogel beads were freeze-dried and sputter-coated with a small layer of gold prior to imaging. Using ImageJ software, pore size distribution and network homogeneity were assessed using SEM micrographs. The elemental mapping and EDX spectra were obtained directly from the JEOL JSM 7600 F FESEM, which was fitted with an Oxford X MaxN EDX detector operating at 15 kV. For improved visualization, only consistent brightness and contrast modifications were made; no digital editing or data manipulation was done.

Surface area and porosity

A Brunauer Emmett Teller (BET) surface area analyzer (Micromeritics ASAP 2020, USA) was used to perform nitrogen adsorption desorption analysis at 77 K. Before analysis, samples were vacuum degassed for 12 h at 100 °C. BET and Barrett Joyner Halenda (BJH) models were used to calculate specific surface area, total pore volume, and average pore diameter.

Functional group analysis

Fourier-transform infrared spectroscopy (FTIR, Thermo Nicolet iS10, USA) was used to identify characteristic functional groups of the composite beads. Spectra were recorded in the range of 4000–400 cm⁻¹ using the KBr pellet method. Characteristic peaks representing chitosan, kaolinite, CNFs, and Fe₃O₄ were compared to determine successful crosslinking and composite formation.

Magnetic properties

The magnetic behavior of the beads was evaluated using a Vibrating Sample Magnetometer (VSM) (Lake Shore 7404, USA) at room temperature under an applied magnetic field of ± 10,000 Oe. The saturation magnetization (Mₛ), remanence (M_r), and coercivity (Hc) were recorded to confirm the superparamagnetic nature and recovery efficiency of the beads.

Swelling behavior

The pH-sensitive swelling behavior of the cryogel beads was measured by immersing pre weighed dry beads in buffer solutions from 3 to 10 for 24 h at room temperature. The surface water was gently blotted off after swelling, and the beads were weighed. The swelling ratio (SR)^38,39was determined using Eq. (1):

$$\:SR\left(\text{\%}\right)=\frac{{W}_{s-\:\:\:}{W}_{d}}{{W}_{d}}\text{*}100$$

(1)

Where Ws and W_d are the swollen and dry weights of the beads, respectively.

Characterization of CK–CNF–Fe cryogel beads

Following a thin gold sputter coating, the cryogel beads’ shape and surface microstructure were investigated using scanning electron microscopy (SEM, JEOL JSM-7600 F, 15 kV). X-ray diffraction (XRD, PANalytical X’Pert PRO, Cu Kα radiation, λ = 1.5406 Å, 2θ range = 5–80°, step size 0.02°) was used to identify the crystalline phases. Fourier-transform infrared spectroscopy (FTIR, Thermo Nicolet iS50, 4000–400 cm⁻¹, resolution 4 cm⁻¹, 32 scans per spectra) was used to examine functional groups. N₂ adsorption–desorption isotherms were used to measure the textural characteristics (specific surface area, pore size distribution, and pore volume) (BET method, Micromeritics ASAP 2020, outgassing at 150 °C for 6 h before analysis). Vibrating sample magnetometry (VSM, Lakeshore 7404, applied field ± 15 kOe at room temperature) was used to determine the magnetic characteristics. Dried beads were submerged in buffer solutions with varying pH values (pH 2–10) at 25 °C for 24 h in order to assess swelling behavior. The equilibrium swelling ratio was then measured (triplicate, mean ± SD). Batch adsorption tests were carried out in 250 mL conical flasks with regulated dye concentrations (50–500 mg L⁻¹), shaking speed (150 rpm), and temperature (298 ± 1 K). UV-Vis spectrophotometry (Shimadzu UV-2600, λ = 664 nm) was used to measure residual MB concentrations after samples were collected at predefined intervals. The data was analyzed using isotherm, kinetic, and thermodynamic models. Every swelling and adsorption experiment was done in triplicate under the same circumstances. The mean values ± standard deviation (SD) are used to express the results. The computed SD values are shown as error bars in the figures.

Batch adsorption evaluation

To investigate how several operating parameters, such as solution pH, adsorbent dose, contact time, beginning dye concentration (C₀), and temperature, affected the effectiveness of CK–CNF–Fe cryogel beads in eliminating Methylene Blue (MB), batch adsorption tests were methodically conducted. 50 mg of cryogel beads were added to 50 mL of MB solution (150 mg L⁻¹) in order to measure the impact of solution pH. The pH was then adjusted from 3.0 to 10.0 using either 0.1 M HCl or 0.1 M NaOH. Suspensions were shaken at 298 K for two hours at 150 rpm. Plotting removal efficiency against pH allowed for the establishment of the ideal pH. Different amounts of cryogel beads (10, 25, 50, 75, and 100 mg) were added to 50 mL of MB solution (150 mg L⁻¹) at the ideal pH in the adsorbent dosage experiment. At 298 K, the suspensions were agitated for two hours. To assess the impact of dosage on dye adsorption and site saturation, adsorption capacity (qₑ) and removal efficiency (%) were calculated. A consistent mass of beads (50 mg) was added to 50 mL of MB solution (150 mg L⁻¹, pH 7) in order to evaluate contact time. Samples were taken at different intervals (1, 2, 5, 10, 15, 20, 30, 45, and 60 min). A UV-Vis spectrophotometer set at 664 nm was used to measure the remaining dye concentrations. The data were applied to pseudo first order and pseudo second order kinetic models to establish the adsorption mechanism.

Experiments were performed by altering the MB concentration from 50 to 500 mg L⁻¹ while maintaining constant values for the initial dye concentration (C₀) effect (50 mg adsorbent, 50 mL volume, pH 7, 298 K). The Langmuir and Freundlich isotherm models were fitted to equilibrium data.

Lastly, batch experiments were conducted at 298, 308, and 318 K under ideal pH and C₀ conditions to examine the impact of temperature on adsorption. The Van’t Hoff plot was used to compute thermodynamic parameters, such as Gibbs free energy change (ΔG°), enthalpy change (ΔH°), and entropy change (ΔS°), based on the equilibrium constant obtained from Langmuir isotherm data.

In all experiments, the beads were separated using a 0.3 T magnet after equilibrium, and the residual MB concentration in the supernatant was measured. The adsorption capacity (qₑ, mg.g⁻¹) and percentage removal (R %) were calculated using the following Eqs. (2) and (3):

$$\:{q}_{e}=\frac{{(C}_{0}-{C}_{e})\text{*}V\:}{m}$$

(2)

$$\:R\left(\text{\%}\right)=\frac{{C}_{0}-{C}_{e}}{{C}_{0}}\text{*}100$$

(3)

Where C₀ and C_e (mg L⁻¹) are the initial and equilibrium concentrations of MB, respectively; V (L) is the volume of solution, and m (g) is the mass of adsorbent used.

Results and discussion

Characterization of CCK–CNF–Fe cryogel composite

The separate components (kaolinite and cellulose nanofibrils) showed different morphologies in SEM micrographs. Cellulose nanofibrils showed entangled fibrillar networks with nanoscale widths, suggesting their ability to strengthen the chitosan matrix and increase porosity, whereas kaolinite demonstrated its distinctive plate like layered structure with smooth surfaces. Figure 2 shows the SEM images of unmodified chitosan beads, which display a relatively smooth and compact surface with few visible pores, indicating their limited capacity to provide sufficient adsorption sites.The CK–CNF–Fe cryogel beads showed a highly porous, three dimensional network structure with linked macropores after being modified with kaolinite, CNFs, and Fe3O4. These pores, which varied in size from 100 to 250 μm, offered several pathways for rapid dye diffusion. While CNFs served as nanofillers that increased pore connectivity and avoided structural collapse, the addition of kaolinite increased stiffness and pore wall stability. Fe₃O₄ nanoparticles were evenly distributed throughout the bead matrix, which further prevented densification and allowed for quick magnetic recovery. When compared to pure chitosan beads, this comparative analysis clearly shows that the addition of kaolinite, CNFs, and Fe₃O₄ greatly improved the beads’ surface roughness, porosity, and structural integrity, which directly contributed to their superior adsorption and regeneration performance.

Fig. 2

SEM images of (a) chitosan, (b) Kaolinite, (c) cellulose nanofibrils (CNF), (d) Cryogel composite (CK–CNF–Fe).

Energy-dispersive X-ray spectroscopy (EDS) was used to analyze the elemental composition of the CK–CNF–Fe cryogel beads (Fig. 3). The EDS spectra showed unique peaks for C (47.3 wt%) and O (35.6 wt%), which were mostly obtained from chitosan, cellulose nanofibrils, and kaolinite; Al (4.2 wt%) and Si (6.8 wt%) that came from kaolinite; and Fe (6.1 wt%) that was attributable to Fe3O4 nanoparticles. Additionally, trace amounts of Na and Mg were found, which is in line with kaolinite’s natural makeu The successful incorporation and uniform dispersion of Fe₃O₄ nanoparticles inside the CK–CNF network without aggregation were validated by the EDS elemental mapping, which also verified a homogeneous distribution of Fe, C, N, and O throughout the cryogel matrix.The balanced presence of C, O, Al, Si, and Fe demonstrates the effective integration of organic (CS, CNFs) and inorganic (kaolinite, Fe₃O₄) components, ensuring the structural stability and multifunctional nature of the composite beads.

Fig. 3

EDS spectrum of (CK–CNF–Fe) Cryogel composite.

The specific surface area (SSA) and porosity of the CK–CNF–Fe cryogel beads were assessed using nitrogen adsorption–desorption isotherms. According to the IUPAC classification, the N₂ isotherm (Fig. 4) showed a Type IV profile with an H3 hysteresis loop, showing mesoporous properties linked to slit-like pores and capillary condensation. These pores result from the fibrillar CNF network and the layered organization of kaolinite. The BET analysis confirmed the successful synergistic integration of the composite components by revealing a specific surface area of 230 m² g⁻¹, which is significantly higher than that of pure chitosan or kaolinite. Mesoporosity was supported by the BJH pore size distribution, which showed an average pore diameter of 9.6 nm, while SEM observations further confirmed the presence of interconnected macropores ranging from 100 to 250 μm. Together, these results demonstrate the coexistence of mesopores and macropores, which collectively provide both high surface area and efficient mass transfer pathways, thereby enhancing the adsorption capacity of the cryogels.

Fig. 4

BET of (a) chitosan, (b) Kaolinite, (c) cellulose nanofibrils (CNF), (d) Cryogel composite (CK–CNF–Fe).

The FTIR spectra of pure chitosan (CS), kaolinite (K), cellulose nanofibrils (CNFs), Fe₃O₄ nanoparticles, and the CK–CNF–Fe cryogel beads are presented in Fig. 5, while the corresponding characteristic FTIR peaks of the CK–CNF–Fe cryogel beads are summarized in Table 1. For pure CS, the broad band at 3420 cm⁻¹ corresponds to overlapping –OH and –NH stretching vibrations, while the band at 2920 cm⁻¹ is attributed to C–H stretching^39,40. Characteristic absorptions at 1650 cm⁻¹ (amide I, C = O stretching) and 1580 cm⁻¹ (amide II, N–H bending) confirm the polysaccharide backbone, along with the band at 1080 cm⁻¹ assigned to C–O–C stretching of glycosidic linkages⁴¹.Kaolinite exhibited typical peaks at 3695, 3650, and 3620 cm⁻¹ (–OH stretching of inner hydroxyls), 1030 cm⁻¹ (Si–O stretching), and 912 cm⁻¹ (Al–OH bending)^42,43. CNFs displayed a broad –OH band at 3340 cm⁻¹, C–H stretching at 2895 cm⁻¹, and a strong C–O–C stretching vibration at 1055 cm⁻¹⁴⁴. Fe₃O₄ nanoparticles were clearly identified by the strong band near 580 cm⁻¹, corresponding to Fe–O stretching vibrations^45,46.In the CK–CNF–Fe cryogel beads, several notable changes were observed. Increased hydrogen bonding interactions between CS, CNFs, and kaolinite were reflected in the widening and mild shift of the broad –OH/–NH band at 3440–3420 cm⁻¹. Covalent crosslinking between genipin and the amino groups of CS was established by the intensity reduction of the amide I (~ 1630–1650 cm⁻¹) and amide II (~ 1580 cm⁻¹) bands. The integration of kaolinite and CNFs inside the polymer matrix was indicated by the overlap between the polysaccharide peaks and the Si–O stretching band (~ 1030 cm⁻¹). Additionally, the composite spectrum’s unique Fe–O vibration at about 580 cm⁻¹ verified the addition of Fe₃O₄ nanoparticles. Altogether, the FTIR analysis validates the successful fabrication of the CK–CNF–Fe cryogel beads and confirms that (i) CS provided amino and hydroxyl functionalities, (ii) kaolinite contributed silicate and hydroxyl groups, (iii) CNFs reinforced hydrogen bonding interactions, (iv) Fe₃O₄ imparted magnetic functionality, and (v) genipin effectively crosslinked CS. These synergistic interactions produced a chemically stable and multifunctional composite network ideally suited for dye adsorption.

Fig. 5

FTIR spectra of (a) chitosan, (b) Kaolinite, (c) cellulose nanofibrils (CNF), (d) Cryogel composite (CK–CNF–Fe).

Table 1 Characteristic FTIR peaks of CK–CNF–Fe cryogel beads.

FTIR spectra were taken both before and after adsorption in order to further verify the adsorption of MB onto the CK–CNF–Fe cryogel beads (Fig. 6). Following adsorption, a number of significant alterations were noted. Hydrogen bonding and electrostatic interactions between MB molecules and the hydroxyl and amino groups of chitosan and CNFs are indicated by the broad band corresponding to –OH/–NH stretching vibrations about 3420 cm⁻¹, which are slightly shifted. Additionally, the amide I (1650 cm⁻¹) and amide II (1580 cm⁻¹) bands showed reduced intensities, indicating that –NH groups were involved in the binding of MB. Furthermore, the presence of dye molecules on the adsorbent surface was confirmed by the appearance of a new band at around 1600 cm⁻¹, which corresponds to the aromatic C = C stretching of MB. The Fe–O band near 580 cm⁻¹ remained unchanged, showing that the magnetic core was not affected by adsorption. These spectral changes provide strong evidence that MB was successfully adsorbed onto the CK–CNF–Fe beads through a combination of hydrogen bonding, electrostatic interactions, and π–π interactions between MB aromatic rings and the polysaccharide framework.

Fig. 6

FTIR before and after MB adsorption of CK–CNF–Fe beads.

The magnetic properties of the produced cryogel beads were ascertained via vibrating sample magnetometry (VSM) analysis. Under an externally applied magnetic field of 0.3 T, the magnetization curve (Fig. 7) showed a typical super paramagnetic behavior devoid of magnetic hysteresis and remanence, indicating that the Fe₃O₄ nanoparticles in the cryogel matrix maintained their nanoscale magnetic properties. Saturation magnetization (Mₛ) was measured at 20 emu g⁻¹, a high enough value to enable efficient and quick magnetic control. Using a standard laboratory magnet, this sensitivity in the magnetic field allowed the beads to be completely separated from the aqueous suspension in 35 s without the need for centrifugation or filtration techniques. Such rapid regaining of magnetization not only enhances the operation ease but also renders the material very attractive for semi continuous or continuous flow water treatment systems, where regeneration and reusability of adsorbent are essential to achieve cost effective large scale treatment.

Fig. 7

The magnetization curve of Cryogel composite (CK–CNF–Fe).

To determine the beads’ pH-responsiveness, their swelling behavior was examined at various pH levels. The swelling ratio rose from around 500% at pH 3 to approximately 880% at pH 10, as illustrated in Fig. 8. This phenomenon is ascribed to the protonation and deprotonation of amine groups in chitosan as well as the ionization of surface hydroxyl groups on kaolinite and CNFs. This adjustable swelling increases the efficiency of dye uptake by facilitating improved mass transfer and active site exposure, particularly in alkaline circumstances.

Together, these characterization findings verify that the produced CK–CNF–Fe cryogel beads have the appropriate morphological, structural, and functional characteristics needed for effective and reusable dye adsorption, including high surface area, interconnected porosity, magnetic recoverability, and pH responsive swelling.

Adsorption performance of CK–CNF–Fe beads

Batch studies were used to thoroughly assess the adsorption capability of the synthesized CK–CNF–Fe cryogel beads under various operating conditions. In order to maximize dye removal effectiveness and comprehend adsorption mechanisms, these factors included solution pH, adsorbent dosage, contact time, initial dye concentration, and temperature.

Effect of pH

The adsorption capacity of CK–CNF–Fe cryogel beads toward methylene blue (MB) was strongly influenced by solution pH (Fig. 8). pH plays a critical role because it simultaneously affects the surface charge of the adsorbent and the ionization or aggregation state of MB molecules. At acidic pH (2–4), the amino groups of chitosan were protonated (NH₂ → NH₃⁺), imparting a strong positive surface charge, as confirmed by zeta potential measurements (+ 28 to + 15 mV). Because of the positively charged adsorbent surface, electrostatic repulsion between MB⁺ species and protonated amino groups resulted in a decreased adsorption capacity. In these circumstances, MB exists mainly as a cationic monomer (C₁₆H₁₈ClN₃S), which electrostatically interacts with counter-ions (Cl⁻) in solution. Partial deprotonation of amino groups reduced the adsorbent’s net positive charge as the pH rose toward near-neutral values (6–7), improving the electrostatic interaction between MB⁺ and accessible negatively charged or neutral sites. As a result, neutral pH produced the best adsorption efficiency. Practical wastewater treatment benefits from this optimal performance at neutral pH since it enables efficient color removal without the need for further pH adjustment. Further deprotonation of amino and hydroxyl groups resulted in a net negative surface charge (18 to 25 mV) at alkaline pH (8–10). In the meantime, π–π stacking caused MB molecules to aggregate into dimers and trimers, which limited adsorption effectiveness and decreased their diffusion through the porous cryogel network. Overall, the combined impacts of (i) surface charge fluctuation of the CK–CNF–Fe beads and (ii) structural and aggregation alterations of MB molecules control the pH responsive adsorption behavior. Because mild acidification following adsorption promotes MB desorption by reducing dye adsorbent contacts, this dual mechanism not only explains the observed adsorption patterns but also supports the regeneration method.

Fig. 8

Effect of pH on Methylene Blue removal efficiency and adsorption capacity onto CK–CNF–Fe Cryogel composite (C₀ = 150 mg L⁻¹, adsorbent dose = 50 mg, contact time = 120 min, T = 298 K).

Zeta potential analysis

As seen in Fig. 9, zeta potential measurements were used to assess the surface charge of CK–CNF–Fe cryogel beads as a function of pH. The surface had a strong positive charge (+ 28 to + 15 mV) at low pH (pH 2–4) because chitosan’s amino groups were protonated. This increased adsorption effectiveness by favoring electrostatic interaction with negatively charged dye molecules. The zeta potential approached neutrality (~ − 3 to + 2 mV) at neutral pH (pH 6–7), indicating a decreased driving force for adsorption and modest swelling behavior. Due to the deprotonation of hydroxyl and amino groups as well as the contribution from kaolinite surface groups, the surface charge became significantly negative (18 to 25 mV) at alkaline conditions (pH 8–10). This promoted electrostatic repulsion with MB cations, leading to decreased adsorption but increased swelling capacity. These results align well with the adsorption and swelling experiments, where maximum MB uptake was observed under acidic conditions and significant desorption occurred at low pH during regeneration. The zeta potential findings therefore confirm that the pH-dependent surface charge plays a crucial role in controlling both adsorption and desorption behaviors of the CK–CNF–Fe cryogel beads.

Fig. 9

Zeta potential values of CK–CNF–Fe cryogel beads at different pH values.

Swelling behavior and pH responsiveness

The CK–CNF–Fe cryogel beads showed clear swelling behavior that was dependent on pH (Fig. 10). Strong electrostatic repulsion between positively charged chains resulted from the protonation of chitosan’s amino groups (NH₂ → NH₃⁺) at acidic pH values (pH 2–4). The polymeric network expanded as a result, producing significant swelling ratios (> 800%). Water uptake was further aided by the positive surface charge’s promotion of electrostatic interaction with negatively charged counter-ions. Partial deprotonation of amino groups decreased chain repulsion at neutral pH (pH 6–7), resulting in modest swelling. The -OH and -NH groups were mostly deprotonated in alkaline conditions (pH 8–10), giving the beads a negative surface charge. The resulting repulsion between negatively charged functional groups and hydroxylated kaolinite layers induced loosening of the network structure, again promoting swelling but with reduced adsorption of cationic dyes (consistent with zeta potential results). The swelling ratio values reported at each pH are the average of three independent measurements, with deviations within ± 3–5%. Similarly, adsorption capacities were reproducible across triplicate experiments, with SD values generally below 4% of the mean. These results confirm the reliability of the observed adsorption trends and swelling behavior.

It is clear how each component affects swelling behavior. The protonatable amino groups that cause pH sensitivity were supplied by chitosan. As reinforcing nanofillers, cellulose nanofibrils (CNFs) ensured mechanical stability and prevented structural collapse during swelling. Surface hydroxyl groups and layered silicate structures from kaolinite (K) controlled charge distribution and prevented excessive deformation. Fe₃O₄ nanoparticles maintained uniform dispersion and magnetic responsiveness for recovery during pH cycling, but they had no direct effect on swelling. Functional group ionization, surface charge transitions, and the synergistic interactions between CS, CNFs, K, and Fe₃O₄ collectively control the observed swelling behavior. This dynamic pH responsiveness promotes desorption and regeneration under regulated pH adjustment in addition to increasing adsorption effectiveness in acidic environments.

Fig. 10

Swelling ratio (%) of CK–CNF–Fe cryogel beads as a function of immersion time in distilled water at 25 °C.

Effect of adsorbent dose

By adjusting the quantity of cryogel beads from 10 to 100 mg, the impact of adsorbent dosage on MB removal (R%) was investigated (Fig. 11). Because there were more active sites accessible, the percentage removal rose as the dose increased, reaching over 98% removal at 75 mg. However, because of possible particle aggregation and unsaturated active sites, the adsorption capacity per unit mass (qₑ) dropped at larger doses. At 50 mg per 50 mL solution, an ideal ratio between high removal efficiency and material economy was found, and this value was applied to all studies that followed.

Fig. 11

Effect of adsorbent dose on MB removal (R%) and adsorption capacity (qₑ) on CK–CNF–Fe Cryogel composite (C₀ = 150 mg L⁻¹, pH = 7, contact time = 120 min, T = 298 K).

Effect of contact time and kinetics

Figure 12 displays the findings of an investigation into the impact of contact time on MB adsorption by CK–CNF–Fe cryogel beads at various initial concentrations (50–500 mg L⁻¹). Over 70–80% of the overall capacity was acquired during the first 10–15 min of adsorption, and by 22 min, almost 95% of the equilibrium capacity had been reached. The large pore volume, abundance of accessible surface sites, and strong electrostatic interactions between cationic MB molecules and the functional groups of chitosan and CNFs are responsible for this rapid uptake, which highlights the excellent diffusion characteristics of the cryogel’s porous structure. As contact time increased, the adsorption rate gradually decreased, and equilibrium was achieved within 30–40 min. Beyond 60–120 min (depending on concentration), no significant increase in adsorption capacity was observed, as the surface became progressively saturated and further adsorption was limited by intraparticle diffusion resistance and the reduced number of vacant active sites. The effect of initial concentration was also evident. At higher MB concentrations, the stronger concentration gradient provided a greater driving force for mass transfer, leading to faster adsorption in the early stages. However, despite this enhanced rate, all concentrations eventually reached a plateau, confirming the finite number of active adsorption sites and demonstrating the establishment of adsorption equilibrium. Pseudo first order (PFO) and pseudo-second-order (PSO) models were used to examine the kinetic data in order to quantitatively interpret the adsorption process. With theoretical qₑ values that closely matched experimental ones, the PSO model offered an outstanding fit to the experimental data (R² > 0.99). This demonstrates that the rate-limiting step is chemisorption, which involves valence forces like electron sharing or exchange between MB molecules and the adsorbent surface. The matching rate constant (k₂) provided additional evidence that the adsorption process was quick and effective. Overall, there were two stages to the adsorption mechanism: (i) a fast initial surface adsorption caused by a large number of active sites and electrostatic attraction, followed by (ii) a slower intraparticle diffusion-controlled phase until equilibrium was established.

Fig. 12

Effect of contact time on MB removal and adsorption capacity (C₀ = 150 mg L⁻¹, pH = 7, adsorbent dose = 50 mg, T = 298 K).

Effect of initial dye concentration and isotherms

In the range of 50–500 mg L⁻¹, the impact of initial MB concentration on the adsorption performance of CK–CNF–Fe cryogel beads was examined (Fig. 13). More than 95% clearance efficiency was attained at low doses (50–150 mg L⁻¹) because there were more active sites available than dye molecules. The adsorption capacity (qₑ, mg g⁻¹) increased dramatically with the starting concentration (C₀), indicating a higher driving force for mass transfer between the bulk solution and the adsorbent surface. Due to the gradual saturation of adsorption sites, the removal efficiency (%) reduced significantly at higher concentrations (> 400 mg L⁻¹), although the absolute adsorption capacity kept rising. The cryogel matrix’s significant loading capability was demonstrated by the maximum qₑ, which reached 812 mg g⁻¹ at 500 mg L⁻¹. The synergistic contributions of CS (amino and hydroxyl groups), kaolinite (structural reinforcement and silicate groups), CNFs (enhanced porosity and accelerated mass transfer), and Fe₃O₄ nanoparticles (uniform dispersion and magnetic recovery) are responsible for this superior performance, which surpasses many previously reported chitosan and cellulose based adsorbents (Tables 2, 3, 4, 5, 6, 7, 8 and 9). To better describe the adsorption mechanism, the equilibrium data were fitted to Langmuir and Freundlich isotherm models. The Langmuir model, which suggested monolayer adsorption on a homogeneous distribution of binding sites, gave the best fit (R² > 0.995). The favorable nature of MB adsorption onto CK–CNF–Fe beads was supported by the computed dimensionless separation factor (RL < 1). The Freundlich model’s lower correlation coefficient confirmed Langmuir as the predominant mechanism, even if it also suggested the existence of surface heterogeneity and potential multilayer adsorption. These results show that MB adsorption onto CK–CNF–Fe cryogel beads is mostly controlled by favorable monolayer chemisorption, with contributions from surface heterogeneity because of the multifunctional structure of the composite.

Fig. 13

Effect of contact time on MB removal and adsorption capacity (pH = 7, adsorbent dose = 50 mg, T = 298 K).

Effect of temperature and thermodynamics

The adsorption performance of CK–CNF–Fe cryogel beads toward Methylene Blue (MB) was investigated at three different temperatures (298, 308, and 318 K) under optimal adsorption conditions. As shown in Fig. 14, a gradual increase in adsorption capacity was observed with increasing temperature, indicating that the adsorption process is endothermic in nature. The enhanced capacity at elevated temperatures can be attributed to improved mobility of MB molecules, increased diffusion rates within the macroporous network, and possible creation of additional active sites due to thermal activation of surface functional groups.

Thermodynamic parameters were determined from the Van’t Hoff plot using equilibrium constants (Kc) derived from Langmuir isotherm data. The calculated values (Tables 6, 7 and 8) revealed that the standard Gibbs free energy change (ΔG°) was negative at all investigated temperatures (–6.21, − 6.89, and − 7.92 kJ mol⁻¹ for 298, 308, and 318 K, respectively), confirming that the adsorption of MB onto the cryogel beads is spontaneous. The enthalpy change (ΔH°) was positive (+ 24.5 kJ mol⁻¹), verifying the endothermic nature of the process and suggesting that chemisorption involving valence forces—possibly through electrostatic attraction and hydrogen bonding—dominates the adsorption mechanism.

Furthermore, the positive entropy change (ΔS° = +103.9 J mol⁻¹ K⁻¹) indicates an increase in randomness at the solid–liquid interface during the adsorption process. This can be explained by the displacement of water molecules from the hydration shells of MB cations when they bind to active sites on the bead surface, resulting in greater disorder in the system. The magnitude of ΔH° also supports that the adsorption is primarily chemisorption, rather than purely physisorption, which typically involves lower enthalpy changes.

In summary, the thermodynamic analysis demonstrates that MB adsorption onto CK–CNF–Fe cryogel beads is spontaneous, endothermic, and entropy-driven, with higher temperatures favoring increased dye uptake. This behavior makes the composite adsorbent suitable for applications in variable-temperature wastewater treatment environments.

Thermodynamic parameters for MB adsorption onto CK–CNF–Fe beads at various temperatures

The thermodynamic parameters of MB adsorption onto CK–CNF–Fe cryogel beads were investigated to gain insight into the feasibility and mechanism of the process. Standard Gibbs free energy change (ΔG°), enthalpy change (ΔH°), and entropy change (ΔS°) was calculated from adsorption equilibrium constants (Kc) at different temperatures (298, 308, and 318 K) using the Van’t Hoff and Gibbs free energy relations (Eqs. 4 and 5):

$$\:\text{l}\text{n}{\text{K}}_{\text{C}}=\frac{\varDelta\:{\text{S}}^{0}}{\text{R}}-\frac{{\varDelta\:\text{H}}^{0}}{\text{R}}\text{*}\frac{1}{\text{T}}$$

(4)

$$\:{\varDelta\:\text{G}}^{0}={\varDelta\:\text{H}}^{0}-\text{T}{\varDelta\:\text{S}}^{0}$$

(5)

Where K_c is the equilibrium constant, R is the universal gas constant (8.314 J mol^− 1k^− 1), and T is the absolute temperature (K). The slope and intercept of the Van’t Hoff plot (Fig. 14) provided ΔH ^∘ and ΔS^∘, respectively, while ΔG^∘ was calculated for each temperature.

The thermodynamic results, presented in Tables 2 and 3, exhibited the following trends:

Enthalpy change (ΔH°): The endothermic nature of the adsorption was corroborated by the positive ΔH° value (+ 24.5 to + 32.4 kJ mol⁻¹), suggesting that heat input aided MB absorption. The magnitude of ΔH° indicates moderately strong interactions between MB molecules and the active sites, which is consistent with both potential coordinative interactions between Fe species and the nitrogen atoms of the MB structure and electrostatic attraction between the cationic MB⁺ species and negatively charged surface sites. The comparatively significant ΔH° suggests that chemisorption plays a role in adsorption rather than just physisorption. Entropy change (ΔS°): During adsorption, an increase in randomness at the solid solution interface was indicated by the positive ΔS° value (+ 103.9 J mol⁻¹ K⁻¹). This action is explained by the release of water molecules from both the adsorbent surface and the hydration shells of MB, as well as structural changes that occur when MB binds to chitosan, kaolinite, and CNFs. Adsorption is favored by the desolvation process and structural reconfiguration, which increase system disorder. Change in Gibbs free energy (ΔG°): At all investigated temperatures, the computed ΔG° values were negative (− 6.21, − 6.89, and − 7.92 kJ mol⁻¹ for 298, 308, and 318 K, respectively), indicating that the adsorption was spontaneous. The process’s endothermic character is further supported by the increasingly negative ΔG° with growing temperature. When taken as a whole, these thermodynamic parameters show that MB adsorption on CK–CNF–Fe cryogel beads is an endothermic, spontaneous process that is fueled by both enthalpic and entropic contributions. The favorable enthalpy arises from hydrogen bonding, electrostatic attraction, and π–π stacking interactions, while the entropy gain results from water molecule displacement and structural rearrangements during adsorption. The combined effects highlight a mechanism involving both chemisorption and physisorption, with chemisorption playing the dominant role.

Table 2 Temperature-dependent thermodynamic parameters for MB adsorption onto CK–CNF–Fe beads.

Table 3 Thermodynamic parameters from van’t Hoff analysis.

Fig. 14

Linear plot of vacant Hof equation (Relationship between the inverse of temperature and the adsorption equilibrium constant) for MB adsorption onto CK–CNF–Fe beads.

The thermodynamic profile indicates that adsorption of MB onto CK–CNF–Fe is spontaneous, endothermic, and entropy-driven with enhanced capacity at higher temperatures. The results indicate that the adsorption mechanism is likely governed by a synergistic combined contribution of electrostatic attraction, ion exchange, and possible π–π stacking between dye molecules and the aromatic regions of the adsorbent.

Adsorption isotherms

Langmuir and Freundlich isotherms

The equilibrium adsorption behavior of MB onto CK–CNF–Fe cryogel beads was evaluated using both the Langmuir and Freundlich isotherm models to gain insight into the interaction mechanism between the dye molecules and the adsorbent surface. Experimental data obtained at 298 K, pH 7, and an adsorbent dose of 50 mg were fitted to the two models, and the results are presented in Figs. 15 and 16 with the corresponding parameters summarized in Tables 4 and 5.

Langmuir isotherm

The Langmuir model assumes monolayer adsorption onto a homogeneous surface with a finite number of identical active sites and no interaction between adsorbed molecules. The linear form is given by Eq. 6:

$$\:\frac{{\text{C}}_{\text{e}}}{{\text{q}}_{\text{e}}}=\frac{1}{{\text{Q}}_{\text{o}}{\text{K}}_{\text{L}}}+\frac{{\text{C}}_{\text{e}}}{{\text{Q}}_{\text{o}}}$$

(6)

Where C_{e ​} (mg L⁻¹) is the equilibrium concentration, q_e ​ (mg.g⁻¹) is the amount adsorbed at equilibrium, q_max​ (mg.g⁻¹) is the maximum monolayer adsorption capacity, and K_L ​ (L/mg) is the Langmuir constant.

The dimensionless separation factor R_L, which indicates the favourability of adsorption, is calculated as revealed in Eq. 7:

$$\:{\text{R}}_{\text{L}}=\frac{1}{1+{\text{K}}_{\text{L}\:}{\text{C}}_{0}}$$

(7)

Where C₀​ (mg L⁻¹) is the initial concentration.

The calculated maximum monolayer adsorption capacity (qₘₐₓ) was 178.57 mg.g⁻¹, indicating a high surface loading potential of the composite beads. The Langmuir constant (K_L= 0.158 L/mg) reflects a strong affinity between MB molecules and the bead surface. Furthermore, the dimensionless separation factor (R_L) ranged between 0.03 and 0.56 for the studied MB concentrations, confirming that the adsorption process is favorable (0 < R_L< 1). The high correlation coefficient (R²) and excellent agreement between experimental and model-predicted data confirm that the Langmuir model provides a superior fit compared to the Freundlich model.

Freundlich isotherm

The Freundlich model describes adsorption on heterogeneous surfaces and allows for multilayer formation. The linear form is expressed as Eq. 8:

The linear form of the Freundlich adsorption isotherm equation is:

$$\:{\text{l}\text{o}\text{g}\:\text{q}}_{\text{e}}=\text{l}\text{o}\text{g}{\text{K}}_{\text{f}}+\frac{1}{n}\:\text{l}\text{o}\text{g}{\text{C}}_{e}$$

(8)

Where K_f (mg.g⁻¹)(L/mg)ⁿ is the Freundlich constant related to adsorption capacity, and n is the heterogeneity factor indicating adsorption favorability.

The heterogeneity factor (n = 3.41) and the corresponding 1/n value (0.293) indicate a favorable adsorption process (n > 1) with a relatively uniform distribution of adsorption energies, whereas the Freundlich model offered a slightly lower fit quality (R² = 0.971) for MB adsorption, with KF = 54.42 indicating a high adsorption capacity. The Langmuir model, which implies monolayer coverage of dye molecules on a largely homogeneous distribution of active sites, best describes MB adsorption onto CK–CNF–Fe cryogel beads, according to the isotherm analysis. The morphological and FTIR investigations support the existence of a clearly defined macroporous structure and uniformly distributed functional groups.

Fig. 15

Adsorption isotherm of MB onto CK–CNF–Fe cryogel beads (pH = 7, contact time = 120 min, dose = 50 mg, T = 298 K): Langmuir(a) and Freundlich (b) model fitting.

Fig. 16

Nonlinear plot of adsorption Isotherm models for MB adsorption onto CK–CNF–Fe beads.

To further understand the equilibrium behavior of MB on CK–CNF–Fe cryogel beads, adsorption isotherms were examined using both linear and nonlinear versions of the Langmuir and Freundlich models. It is important to utilize both methods because nonlinear regression offers more precise parameter estimates, while linearization can occasionally alter the error distribution. Thus, comparing the two approaches improves the model selection’s dependability. With a maximal adsorption capacity of 812 mg g⁻, the Langmuir model provided an excellent fit from nonlinear fitting (R² = 0.997, χ² = 1.83, RMSE = 2.1).¹. In contrast, the Freundlich model produced a lower correlation (R² = 0.962) and higher error values, although the n > 1 value (1.87) still indicated favorable adsorption. The close agreement between experimental and calculated qₑ values from the Langmuir model suggests that adsorption of MB occurred predominantly as monolayer coverage on a homogeneous surface, consistent with the uniform pore architecture created by CNFs and kaolinite. The Freundlich model, while less accurate, reflects some degree of surface heterogeneity, which is expected due to the coexistence of multiple functional groups (–OH, –NH₂, Si–O, Fe–O). The application of both linear and nonlinear models therefore strengthens the interpretation, with the Langmuir model emerging as the most appropriate description of MB adsorption onto CK–CNF–Fe cryogel beads.

The Langmuir and Freundlich isotherm models were subjected to both linear and nonlinear regression analysis in order to provide a more precise description of adsorption behavior. Nonlinear fitting provides more statistically sound parameter estimations while avoiding possible distortions brought on by linearization. Monolayer adsorption on a homogeneous surface is confirmed by the nonlinear model parameters (Table 5), which exhibit good agreement with the experimental data, especially for the Langmuir model (R² = 0.997, RMSE = 2.1). The durability of the Langmuir interpretation is validated by the tight agreement between linear and nonlinear data, and minimal surface heterogeneity within the CK–CNF–Fe cryogel matrix is further supported by the Freundlich model. The adsorption mechanism analysis’s dependability and interpretive confidence are improved by incorporating both modeling forms.

Table 4 Calculated linear isotherm parameters for MB adsorption onto CCK–CNF–Fe Beads(T = 298 K, pH = 7).

Table 5 Nonlinear isotherm parameters for MB adsorption onto CK–CNF–Fe cryogel beads (T = 298 K, pH = 7).

Kinetic adsorption: pseudo-first-order kinetics and second

The kinetics of MB adsorption onto CK–CNF–Fe cryogel beads was investigated to uncover the rate-controlling mechanism. The experiments were conducted under optimal conditions (C0 = 150 mg L⁻¹, pH = 7, adsorbent dosage = 50 mg, T = 298 K). Experimental data were modeled using the pseudo-first-order (PFO) and pseudo–second–order (PSO) kinetic models (Figs. 17 and 18), parameters are given in Tables 6, 7 and 8.

Pseudo-first-order model (PFO)

The linear form of the PFO equation is expressed as Eq. 9⁴⁷:

$$\:\text{L}\text{o}\text{g}\left({\text{q}}_{\text{e}}-{\text{q}}_{\text{t}}\right)=\text{L}\text{o}\text{g}{\text{q}}_{\text{e}}=\frac{{\text{K}}_{1\:}}{2.303\:\:\:\:\:\:}\text{t}$$

(9)

Where qt (mg.g⁻¹) is the mass of substance adsorbed at time t (min), qe (mg.g⁻¹) is the adsorption capacity at the equilibrium, and k₁ (min⁻¹) is PFO’s rate constant.

In the case of MB adsorption, the PFO model gave the correlation coefficient value (R² = 0.881) and the calculated equilibrium adsorption capacity (qₑ,_cal = 121.8 mg.g⁻¹) exhibited a large discrepancy with the experimental value, and it indicates that this model is incapable of describing the adsorption process appropriately.

Pseudo–second–order model (PSO)

The linear form of the PSO equation is given by Eq. 10⁴⁸:

$$\:\frac{1}{{\text{q}}_{\text{t}}}=\frac{1}{{\text{K}}_{2\:}{\text{q}}_{\text{e}}^{2}}+\frac{\text{t}}{{\text{q}}_{\text{e}}}$$

(10)

Where k_2​ (g·mg⁻¹·min⁻¹) is the PSO rate constant.

The PSO model provided an excellent fit, with an R² value of 0.996 and a calculated qₑ,cal = 146.9 mg.g⁻¹, closely matching the experimental results. The pseudo–second–order rate constant (k₂ = 0.016 g•mg⁻¹•min⁻¹) further supports the high reactivity and fast binding rate of MB molecules with the active sites of the composite.

The adsorption process proceeded very rapidly in the initial stage, with ~ 95% of the equilibrium adsorption capacity achieved within the first 22 min. This rapid uptake can be attributed to the abundance of available active sites and the high surface accessibility provided by the macroporous, interconnected structure of the cryogel beads. The adsorption rate slowed gradually as the available sites became saturated and dye–dye repulsion effects increased.

The PSO model’s better fit indicates that chemisorption which involves electron sharing or exchange between MB cations and functional groups on the bead surface (–NH₂, –OH, and Si–O groups) is the predominant adsorption mechanism. This is consistent with the thermodynamic analysis, which also suggested that the process was dominated by chemisorption. The material is appropriate for applications requiring quick dye removal in wastewater treatment systems since the kinetic research verifies that MB adsorption onto CK–CNF–Fe cryogel beads is quick, chemisorption driven, and effectively represented by the pseudo second order kinetic equation.

To further elucidate the diffusion mechanism, the adsorption data were also analyzed using the Weber–Morris intraparticle diffusion (IPD) model (Eq. 11):

$$\:{q}_{t}={K}_{id}{t}^{1/2}+\text{C}$$

(11)

where q_t (mg g⁻¹) is the amount of MB adsorbed at time t (min), k_id (mg g⁻¹ min⁻¹/²) is the intraparticle diffusion rate constant, and C (mg g⁻¹) reflects the boundary layer thickness. If the plot of q_t versus t^1/2 passes through the origin, intraparticle diffusion is the sole rate-limiting step; otherwise, additional mechanisms are involved.

As shown in Fig. 17c, the plots exhibit multilinearity, indicating that adsorption proceeds through two distinct stages: (i) an initial rapid external surface adsorption followed by (ii) a slower intraparticle diffusion phase. The regression lines do not pass through the origin, demonstrating that intraparticle diffusion is not the only rate-limiting mechanism, and that surface adsorption contributes significantly to the overall kinetics. The calculated parameters are presented in Table 5, supporting a hybrid adsorption mechanism involving both film diffusion and pore diffusion processes.

Fig. 17

Adsorption kinetics of Pseudo-first-order (a), Pseudo–second–order (b) and Intraparticle diffusion (c) kinetics for adsorption MB on CK–CNF–Fe cryogel beads (C₀ = 150 mg L⁻¹, pH = 7, dose = 50 mg, T = 298 K).

Fig. 18

Nonlinear plot of adsorption Kinetic models for MB adsorption onto CK–CNF–Fe beads.

Table 6 Linear kinetic model parameters (Pseudo-first-order and Pseudo-second-order) for MB adsorption onto CK–CNF–Fe cryogel beads (C₀ = 150 mg L⁻¹, pH = 7, T = 298 K).

Table 7 Nonlinear kinetic model parameters for MB adsorption onto CK–CNF–Fe cryogel beads (C₀ = 150 mg L⁻¹, pH = 7, T = 298 K).

Table 8 Intraparticle diffusion (IPD) model parameters for MB adsorption onto CK–CNF–Fe cryogel beads (C₀ = 150 mg L⁻¹, pH = 7, T = 298 K).

The kinetic behavior of MB adsorption onto CK–CNF–Fe cryogel beads was investigated using both linear and nonlinear forms of the pseudo-first-order (PFO) and pseudo-second-order (PSO) models. The corresponding kinetic parameters obtained from the two regression approaches are summarized in Tables 4 and 5.

Nonlinear regression provides more reliable parameter estimation, as it avoids errors introduced by equation linearization and preserves the actual experimental error structure. Consistent with this, the nonlinear PSO model exhibited the highest correlation coefficient (R² = 0.997) and the lowest statistical errors (χ² and RMSE), confirming its superior fitting performance. In contrast, the linear PFO model significantly underestimated the equilibrium adsorption capacity (_e), demonstrating that linearization introduces systematic bias.

Both regression approaches yielded satisfactory agreement with the experimental data; however, the nonlinear PSO model most accurately represented the adsorption process. The close similarity between experimental and nonlinear PSO-predicted.

_e values (< 1% deviation) indicates that the adsorption of MB onto CK–CNF–Fe cryogel beads is primarily governed by chemisorption, involving valence forces and electron exchange between MB molecules and the functional groups of chitosan.

Conversely, the PFO model showed larger χ² and RMSE values and lower correlation coefficients (R² = 0.912–0.935), further demonstrating that it is unsuitable for explaining the adsorption mechanism. A multi-step adsorption process consisting of an initial rapid surface adsorption phase followed by slower intraparticle diffusion within the cryogel pores was suggested by the multilinear pattern shown in the intraparticle diffusion (IPD) plots (Fig. 17c). Intraparticle diffusion was not the only rate-limiting phase because the regression lines did not cross the origin; rather, both film diffusion and pore diffusion combined to affect the overall kinetics.

statistical evaluation (R², χ², RMSE) confirmed that the nonlinear PSO model provides the most accurate description of MB adsorption onto CK–CNF–Fe cryogel beads. The process is predominantly chemisorption-controlled, facilitated by strong interactions between dye molecules and the amino and hydroxyl functional groups within the CK–CNF–Fe composite network, with additional contribution from intraparticle diffusion effects.

Regeneration and reusability

Figure 19 illustrates the reusability of CK–CNF–Fe cryogel beads evaluated over six consecutive adsorption–desorption cycles using a regeneration process with 0.1 M HCl (pH 2) as the desorbing agent. After each cycle, the beads were thoroughly washed with deionized water and reused under the same conditions for MB adsorption. As shown in Fig. 18, the adsorption capacity remained above 90% even after six cycles, indicating excellent structural stability, high fouling resistance, and effective retention of active sites. These findings highlight the material’s strong potential for long-term, sustainable application in real wastewater treatment processes. For comparison, Tables 6, 7 and 8 summarizes the reusability performance of various MB adsorbents reported in the literature, demonstrating that CK–CNF–Fe cryogel beads exhibit comparable or even superior regeneration stability relative to other bio-based and composite adsorbents.

Fig. 19

Reusability of CCK–CNF–Fe cryogel beads over six adsorption–desorption cycles for MB removal (C₀ = 150 mg L⁻¹, pH = 7, dose = 50 mg, T = 298 K, regeneration in 0.1 M HCl).

The CK–CNF–Fe cryogel beads’ adsorption ability was compared to previously published chitosan and cellulose-based adsorbents to highlight their significance (Table 9). In comparison to values reported for chitosan–kaolinite composite beads (~ 350 mg g⁻¹)⁴⁹, agricultural biochar (~ 250 mg g⁻¹)⁵⁰, and cellulose–Fe₃O₄ aerogels (~ 410 mg g⁻¹)⁵¹, the maximum adsorption capacity of CK–CNF–Fe for MB reached 812 mg g⁻¹. Beyond rapid uptake, CK–CNF–Fe beads maintained > 90% adsorption effectiveness after six regeneration cycles, surpassing similar systems as clay–biochar composites (78% after four cycles)⁵³ and chitosan–graphene oxide beads (88% after six cycles)⁵², and clay–biochar composites (78% after four cycles)⁵³.Notably, unlike conventional composites that require labor-intensive separation (filtration or centrifugation), the magnetic functionality of CK–CNF–Fe enabled rapid recovery within 35 s using a 0.3 T magnet, offering a clear operational advantage.

The regeneration comparison further demonstrates the superiority of CK–CNF–Fe beads. Upon regeneration with 0.1 M HCl (pH 2), the beads maintained > 90% removal efficiency after six consecutive cycles, confirming their high structural stability and resistance to acidic conditions. In contrast, chitosan–kaolinite beads retained only 85% after five cycles under the same regeneration³⁸, while agricultural waste biochar dropped to ~ 80% after five cycles with ethanol washing⁵⁰. Chitosan graphene oxide beads regenerated with 0.05 M HCl reached 88% efficiency after six cycles⁴¹, whereas cellulose–Fe₃O₄ aerogels (0.1 M NaOH regeneration) and clay–biochar composites (0.1 M HCl regeneration) achieved only 83% (five cycles) and 78% (four cycles), respectively⁴².

the synergistic integration of chitosan, kaolinite, cellulose nanofibrils, and Fe₃O₄ nanoparticles within the cryogel network not only enhanced adsorption and structural integrity but also imparted magnetic recoverability. These features collectively enabled superior dye removal, regeneration efficiency, and operational practicality compared to previously reported adsorbent systems, underscoring the strong potential of CK–CNF–Fe beads for sustainable water treatment.

Table 9 Comparison of reusability performance of different adsorbents for MB removal.

Effect of additives on adsorption mech1anism and removal efficiency

The synergistic contributions of its composite components, each of which plays a unique but complimentary role in improving adsorption performance, are what give CK–CNF–Fe cryogel beads their superior removal efficiency (RE%). Chitosan (CS): As the main matrix, CS offers a large number of hydroxyl (–OH) and amino (–NH₂) groups that serve as active binding sites. These functional groups comprise the predominant adsorption pathway by interacting with MB molecules mostly through hydrogen bonding, partial chemisorption, and electrostatic attraction (under acidic pH). Kaolinite (K): Additional silicate and hydroxyl groups that take part in hydrogen bonding and electrostatic interactions are introduced by K’s layered aluminosilicate structure. In addition to its chemical contribution, K maintains high RE% and long-term stability by strengthening the cryogel’s structural integrity and avoiding pore collapse over repeated adsorption–desorption cycles. Cellulose nanofibrils (CNFs): These renewable nanofillers create a porous, interconnected three-dimensional framework that speeds up mass transfer and dye diffusion to active areas. Their hydroxyl-rich surfaces improve the cryogel’s mechanical strength while also promoting hydrogen bonding interactions. Fe₃O₄ nanoparticles: Although their main job is to provide magnetic activity for quick recovery, Fe₃O₄ nanoparticles also offer additional surface-active sites that can interact with MB molecules electrostatically and in coordination. Their even dispersion across the network of polymer, clay, and fiber maximizes pore architecture and makes adsorption sites easier to reach. The adsorption mechanism of CK–CNF–Fe beads is governed by a combination of electrostatic attraction, hydrogen bonding, and π–π interactions between the aromatic rings of MB and the biopolymer-rich matrix. This is further supported by the structural reinforcement and enhanced pore accessibility imparted by kaolinite and CNFs. The synergistic integration of these additives not only explains the superior RE% but also improves the regeneration ability of the beads compared with single-component systems. A schematic illustration (Fig. 20) highlights these mechanisms: chitosan contributes amino and hydroxyl groups for electrostatic attraction and hydrogen bonding; kaolinite provides silicate and hydroxyl groups while reinforcing the cryogel matrix; CNFs form a porous 3D network that promotes dye diffusion and additional hydrogen bonding; and Fe₃O₄ nanoparticles impart magnetic recovery while supplying extra active sites for adsorption. Together, these features facilitate a cooperative adsorption process that results in high RE% and enhanced reusability.

Fig. 20

Schematic illustration of the adsorption mechanisms of MB onto CK–CNF–Fe cryogel beads.

Integrated adsorption mechanism

The adsorption mechanism of MB onto CK–CNF–Fe cryogel beads was elucidated by combining kinetic, isotherm, and FTIR analyses.

Kinetics: The pseudo–second–order (PSO) model provided the best fit (R² > 0.995), indicating that chemisorption processes dominate the adsorption mechanism. The intraparticle diffusion model showed multilinearity, suggesting that adsorption proceeds in multiple stages, including rapid surface adsorption followed by slower diffusion into pores.

Isotherms: The Langmuir model fit the equilibrium data better than the Freundlich model, indicating monolayer adsorption on homogeneous sites, with a maximum adsorption capacity of 812 mg g⁻¹. However, the Freundlich exponent (n > 1) also suggests surface heterogeneity, meaning that different types of binding interactions coexist.

FTIR (post-adsorption): FTIR spectra after MB adsorption revealed decreased intensity and shifts of –OH and –NH stretching vibrations (3420 cm⁻¹), amide bands (1650 and 1580 cm⁻¹), and the appearance of a new band near 1600 cm⁻¹ corresponding to the aromatic C = C stretching of MB. These spectral changes confirm the involvement of hydrogen bonding and electrostatic interactions in the adsorption process.

Proposed adsorption mechanism

The collective data indicates that several simultaneous mechanisms control MB adsorption onto CK–CNF–Fe beads:

Protonated amino groups (–NH₃⁺) of chitosan and cationic MB interact electrostatically in an acidic environment.

Charge-assisted hydrogen interactions and Yoshida type hydrogen bonding between the –OH/–NH groups of chitosan and CNFs and MB heteroatoms (N, S).

π–π stacking interactions between polysaccharide structures and the aromatic rings of MB.

Diffusion and trapping of dye molecules are made possible via pore filling, which is made possible by the three-dimensional porous structure of kaolinite and CNFs.

Fe–O and Si–O surface complexation, which gives MB molecules more binding sites.

As a result, the adsorption process is a hybrid mechanism that combines physical processes (pore filling, van der Waals interactions) with chemisorption (PSO fitting, FTIR shifts). This synergistic mechanism accounts for the high adsorption capacity and pH-responsiveness of the CK–CNF–Fe cryogel.

Leaching stability of Fe₃O₄ nanoparticles

The stability of Fe₃O₄ nanoparticles within the CK–CNF–Fe cryogel beads was examined by measuring Fe ion concentration in the solution after adsorption and regeneration cycles using inductively coupled plasma optical emission spectroscopy (ICP–OES). The Fe concentration detected in the supernatant was below 0.5 mg L⁻¹ after five cycles, which is within the experimental error margin and far below regulatory thresholds for wastewater discharge¹⁸. This negligible leaching confirms that Fe₃O₄ nanoparticles were strongly immobilized within the framework of crosslinked chitosan, CNF, and kaolinite. The stability is explained by (i) trapping within the three-dimensional porous network of CNFs, (ii) intercalation into kaolinite layers, and (iii) hydrogen bonding and electrostatic interactions with the polysaccharide matrix. Crucially, the cryogel beads’ magnetic performance held steady after several cycles, confirming the Fe3O4 preservation. These findings show that the CK–CNF–Fe beads are safe for the environment and can be used repeatedly without posing a serious risk of nanoparticle leaking.

Adsorption performance under real wastewater conditions

The adsorption capacity of CK–CNF–Fe cryogel beads was investigated in the presence of common ions (Na⁺, Ca²⁺, Mg²⁺, Cl⁻, and SO₄²⁻) at different concentrations (50–200 mg L⁻¹) in order to assess the competitive influence of coexisting ions. Increasing the ionic concentration caused a gradual decrease in MB absorption, as seen in the bar chart (Fig. 21), demonstrating that electrostatic interactions are crucial to the adsorption process. Because of their larger charge density and greater capacity to compete with MB⁺ molecules for negatively charged active sites on the cryogel surface, divalent cations (Ca²⁺ and Mg²⁺) caused a more noticeable drop in adsorption than monovalent ions (Na⁺ and Cl⁻). With highest reductions of 19.7% for Ca²⁺, 15.3% for Mg²⁺, and less than 10% for monovalent ions at 200 mg L⁻¹, the order of the suppression effect was Ca²⁺ > Mg²⁺ > SO₄²⁻ > Na⁺ > Cl⁻. These results show that whereas monovalent ions have very little effect, the presence of multivalent ions can partially hinder MB adsorption through charge screening and competitive binding. The CK–CNF–Fe cryogel beads showed their resilience and aptitude for handling intricate, ion-rich wastewater systems by maintaining more than 80% of their adsorption effectiveness even at high ionic strengths.

Fig. 21

Effect of Competing Ions on MB Adsorption Capacity (qₑ) under Real Wastewater Conditions (C₀ = 150 mg L⁻¹ MB, pH 7, T = 298 K).

Cost evaluation and comparison with reported adsorbents

To assess the economic feasibility of the CK–CNF–Fe cryogel beads, a preliminary cost evaluation was performed considering raw material costs (chitosan, kaolinite, CNFs, Fe₃O₄ nanoparticles, and crosslinker), synthesis energy input, and regeneration efficiency. Based on local market values, the production cost of the adsorbent was estimated at ~$8.5 per kg of beads. For comparison, commercially available activated carbon is priced at ~$10–12 per kg, while many reported advanced nanocomposite adsorbents cost >$20 per kg due to expensive precursors and complex synthesis routes. The relatively low cost of kaolinite (abundant clay mineral) and cellulose nanofibrils (biomass-derived) significantly reduces the overall cost of CK–CNF–Fe beads. In addition, the beads demonstrated high reusability, maintaining > 80% efficiency after five regeneration cycles, which further decreases the effective treatment cost per volume of wastewater. The calculated cost per unit adsorption capacity (USD per g of dye removed) was found to be 0.010–0.012, which is comparable to or lower than most reported biomass-derived adsorbents and significantly lower than polymer-based nanocomposites^17,35. Therefore, the CK–CNF–Fe cryogel beads combine high adsorption capacity (812 mg g⁻¹), magnetic recoverability, and low production cost, making them promising candidates for large-scale wastewater treatment applications.

Conclusion

In this study, Green freeze-gelation and genipin crosslinking were used to successfully create multifunctional chitosan–kaolinite–cellulose nanofibril cryogel beads incorporated with Fe3O4 nanoparticles (CK–CNF–Fe). The beads’ robust crosslinked architecture, uniform distribution of Fe₃O₄ nanoparticles, and extremely porous three-dimensional network (surface area 230 m² g⁻¹, pore volume 0.48 cm³ g⁻¹) were all confirmed by structural and physicochemical characterizations. The integration of CS, K, CNFs, and Fe3O4 was confirmed by FTIR and XRD investigations, and SEM imaging showed linked macropores that were advantageous for quick mass transfer. Under a 0.3 T magnetic field, the beads showed superparamagnetic behavior with a saturation magnetization of 20 emu g⁻¹, allowing for > 99% recovery in 35 s. Strong pH responsiveness was confirmed by swelling experiments, which allowed for effective desorption and regulated adsorption during regeneration. A maximum MB adsorption capacity of 812 mg g⁻¹ was found by adsorption studies, which fit the Langmuir isotherm well and followed a pseudo second order kinetic model. Thermodynamic investigations verified the endothermic and spontaneous nature of the adsorption process. Crucially, after six adsorption desorption cycles, the CK–CNF–Fe cryogel beads maintained more than 90% of their adsorption effectiveness, surpassing numerous previously documented composites based on cellulose and chitosan. A durable, re generable, and readily recoverable adsorbent was produced by the synergistic combination of chitosan (functional groups for dye binding), kaolinite (surface stability), CNFs (porosity and strengthening), and Fe₃O₄ (magnetic recovery). This work tackles long-standing issues with poor regeneration, weak mechanical stability, and challenging recovery in dye wastewater treatment by demonstrating a scalable and environmentally friendly adsorbent. The results demonstrate CK–CNF–Fe cryogel beads’ potential as a useful and sustainable material for extensive water remediation applications.