Opuntia ficus-indica cladode powder was prepared following the method described by Djobbi et al. [25]. After removing the spines, the cladodes were carefully washed with ordinary tap water and rinsed with distilled water before being cut into small, 4 cm cubes. The cubes were dried in an oven at 333 K for 48 h and subsequently ground into a fine powder (40–80 μm) using a laboratory mill (Fig. 1). To remove hemicellulose, lignin, and excess minerals like whewellite and CaCO₃ from the cladode surface, the powder was treated with a 15% aqueous sodium hydroxide solution. The suspension was subjected to a 4-h stirring period to facilitate thorough mixing, subsequently filtered, and then rinsed with distilled water until the washings exhibited a neutral pH. The chemically treated powder was dried overnight in an electric oven at 333 K, then ground and stored in clean, dry glass vials for future use (Fig. 1). The untreated material was designated as OFIC, while the sodium hydroxide-treated material was labeled OFIC-NaOH.

To prepare Fe(OH)_x@Cellulose, 0.25 g samples of OFIC were stirred in 500 mL of aqueous FeCl₃ solutions (1 g·L⁻¹) at a 1:2 mass ratio for 2 h at room temperature. The resulting solutions were filtered and washed with water until the pH reached neutrality. Analysis of the wash water revealed that 87.7% of the iron (0.151 g) was removed, indicating a retention of 12.3% (0.021 g) on the cellulose. The chemically treated powders, Fe(OH)_x@Cellulose and OFIC-NaOH, were dried in an oven at 333 K for 4 h, and their final weights were recorded (Fig. 1).

The crystalline structure of the raw and treated materials was investigated using X-ray diffraction (XRD) on a Bruker D8 diffractometer with Cu Kα₁/Kα₂ radiation (λ₁ = 0.1540 nm, λ₂ = 0.1544 nm). Data were collected from 5° to 60° using a Malvern Panalytical X’pert3 Powder diffractometer (Almelo, The Netherlands), equipped with a flat rotating sample holder and a 1D-PIXcell detector. Fourier transform infrared spectroscopy (FTIR) analysis was performed using a Bruker Tensor 27 spectrophotometer (Billerica, MA, USA). KBr pellets were prepared for FTIR measurements, which were carried out within the 400–4000 cm⁻¹ spectral range. This technique allowed for the investigation of vibrational properties and chemical bond information present in the materials. Subsequent analysis of the iron content in the metallic material was carried out by X-ray fluorescence spectrometry “Thermo Scientific Niton FXL model 959 (Waltham, MA, USA)”. The microstructure and morphology of samples were analyzed using Scanning Electron Microscopy (SEM) at 3–5 kV acceleration voltage. Thermogravimetric analysis (TGA) was conducted using a TGA 400 thermogravimetric analyser from Perkin Elmer (PerkinElmer, Shelton, SC, USA) under an airflow in the temperature range of 50°C–800°C at a heating rate of 10°C/min. The ash content was determined from the residue at 800°C, and the moisture content was calculated from the weight loss at 150°C.

The following equation was used to determine the equilibrium adsorption capacity Qe (mg/g) of dyes onto OFIC and Fe(OH)_x@Cellulose [27,30]. (1)Qe=C0−Cem ∗ Vwhere C₀ (mg/L) is the initial dye concentration, C_e (mg/L) is the equilibrium dye concentration, V (L) is the solution volume and m (g) is the adsorbent mass.

To optimize dye removal processes, a fundamental understanding of dye-adsorbent interactions is crucial. Adsorption isotherms provide valuable insights into this relationship. The models applied to describe the dye adsorption onto the biosorbent (Fe(OH)_x@Cellulose) involve the Langmuir, Freundlich, and Temkin equations [3,31]. Langmuir’s model describes adsorption as a process in which the molecules form a single layer on the surface of the adsorbent. It requires that the adsorbent has a fixed number of adsorption sites, each possessing identical binding energies for the molecules, while neglecting any interaction among the adsorbed species. The Langmuir isotherm can be illustrated by a linear equation: (2)CeQe=1Qmax×KL+CeQmaxwhere Qmax (mg/g) is the monolayer saturation adsorption capacity and KL (L/mg) is the Langmuir constant linked to adsorption ability. The essential characteristics of the Langmuir isotherm can be expressed by a unitless constant called the separation factor (RL) also known as the equilibrium parameter, defined by Islam and Mostafa [32], as following: (3)RL=11+KL×Ciusing the Langmuir constant (K) and the initial concentration (C₀), we can determine the adsorption behavior. If the calculated R_L value is greater than 1, the adsorption is unfavorable and linear. If it’s between 0 and 1, it’s favorable. And if it’s zero, the adsorption is irreversible.

In contrast to Langmuir’s assumption of a constant adsorption enthalpy, Freundlich proposes a logarithmic relationship between enthalpy and surface coverage expressed as follows: (4)Ln Qe=LnKF+1nLnCe

n and KF (mg/g) are the Freundlich parameters.

Unlike Langmuir and Freundlich, the Temkin model accounts for changes in adsorption energy due to surface heterogeneity or intermolecular interactions. This relationship is expressed linearly as follows: (5)Qe=B LnKT+B LnCe

The Temkin model includes the parameters, B (J/mol), representing the heat of adsorption, and K (L/g), representing the equilibrium constant at the highest binding energy.

To elucidate the adsorption mechanisms of NR and CR by Fe(OH)_x@Cellulose, the pseudo-first-order, pseudo-second-order, and Elovich kinetic models were employed [33].

The X-ray diffractograms of the raw material (OFIC) and the two modified materials (OFIC-NaOH and Fe(OH)_x@Cellulose) are depicted in Fig. 2.

The X-ray powder diffractogram (Fig. 2) of the fibers obtained through alkali treatment (Fig. 2b) shows three resolved peaks at approximately 2θ = 14.8°, 21.2°, and 29.3°. These peaks correspond to the (11¯0), (110), and (200) planes, characteristic of a cellulose I-type crystalline structure [25,35]. The prominent crystalline peaks indicative of calcium oxalate monohydrate (whewellite) at 14°–15° and 24°–25° (2θ), along with those at 29°–30°, 39°–40°, and 45°–50° (2θ), characteristic of calcium carbonate (CaCO₃), observed in the diffractogram of the raw sample, disappear in diffractograms Fig. 2b,c. This indicates the removal of carbonates and whewellites following the alkaline iron(III) chloride treatments [24,25]. The XRD analysis (Fig. 2) revealed that the cellulose obtained from iron-treated biomass (Fe(OH)_x@Cellulose) displayed two diffraction peaks at 2θ = 14.8° and 21.2°, characteristic of cellulose I-type. A comparison of the diffractograms, using HighScore software and the ICDD PDF4+ database, shows that the cellulose is of type Iβ. However, following iron treatment, the cellulose fibers exhibit a higher amorphous content. Additionally, some whewellite peaks that remain at very low intensities in the OFIC-NaOH spectrum are nearly eliminated in the Fe(OH)_x@Cellulose spectrum.

As depicted in Fig. 3, the treated biomass displayed comparable spectral features to the native biomass, suggesting that no chemical alterations occurred during the treatment process. FTIR spectra of OFIC, OFIC-NaOH, and Fe(OH)_x@Cellulose all displayed broad bands in the 3000–3500 cm⁻¹ region. These bands are characteristic of O-H stretching vibrations and are indicate hydrogen bonding interactions within the molecular structures of all three samples [25,28,36,37]. The “fingerprint” region (1800–400 cm⁻¹) (Fig. 3) showed distinct bands at 1438, 1158, 1107, and 890 cm⁻¹, characteristic of native cellulose Iβ [36,38]. An absorption band at 1650 cm⁻¹ suggests the presence of deprotonated carboxylate groups, as evidenced by the asymmetric stretching of the C=O double bond [39]. The broad bands observed around 1030 and 1060 cm⁻¹ can be attributed to polysaccharides’ C-O-C or -OH vibrations [40]. The attenuation of the intensity of bands observed at 1630, 1320, and 778 cm⁻¹, which correspond to asymmetric and symmetric carboxylate stretching and OCO deformation modes, respectively, suggests the removal of calcium oxalate monohydrate (whewellite) by FeCl₃ and alkaline treatments of OFIC. The observed bands near 1150 and 1160 cm⁻¹ are assigned to antisymmetric stretching vibrations of the C-O-C and vibrations of C-O glycosidic bond, respectively [41,42].

X-ray fluorescence spectroscopy of the Fe(OH)_x@Cellulose treated biomass material revealed a 18.2% iron content, indicating significant iron incorporation. This directly impacts the material’s adsorption capabilities. The substantial integration rate increases the material’s active surface area, providing more binding sites for dye molecules. Furthermore, the presence of iron modifies the cellulose’s physicochemical properties, enhancing interaction with adsorbable pollutants [43,44].

The morphology of the neat and modified fibers was studied using scanning electron microscopy (SEM). Fig. 4 illustrates SEM micrographs of OFIC, alkaline-treated OFIC, and Fe(OH)_x@Cellulose. The raw OFIC surface presents a smooth, compact appearance with minimal visible channels [45]. The OFIC-NaOH and Fe(OH)_x@Cellulose samples (Fig. 4c,e) displayed smoother morphological structures with a greater degree of softness compared to OFIC. The fibrous components observed in both the raw material and the treated biomasses were of similar dimensions, ranging from 2 to 11 μm. The modified biomass adsorbents exhibited distinct surface features, characterized by rough surfaces and numerous extended pores, indicating the creation of larger exposure zones and channels for pollutant uptake [46,47].

The raw biomass displayed whewellite crystals, ranging in size from 0.3 to 1 μm, distributed across its entire surface (Fig. 4b). Calcium oxalate crystals were not discernible on the surface of either OFIC-NaOH (Fig. 4d) or Fe(OH)_x@Cellulose (Fig. 4f) [24]. The microcrystals observed on the surface of Fe(OH)_x@Cellulose (Fig. 4f) are likely attributed to the formation of magnetite [48,49].

Reaching equilibrium in adsorption systems is essential, and contact time is a crucial parameter that plays a key role in defining the rate at which this equilibrium is established.

To determine the optimal contact time, the evolution of dye removal rates was studied. The curves presented in Fig. 6 demonstrate that both dyes exhibited enhanced adsorption by the treated cladode powder compared to the untreated material. Investigating the adsorption rate variation over time revealed that NR adsorption onto Fe(OH)_x@Cellulose increased gradually (Fig. 6a), reaching a maximum of over 70% after two hours. The adsorption kinetic is quite rapid, reaching more than 65% within the first 20 min. Subsequently, the rate slowed significantly, suggesting saturation of the adsorbent’s surface sites. Continued, although slower, adsorption likely occurred through diffusion of the dye into the adsorbent’s pores. The treated material Fe(OH)_x@Cellulose demonstrates exceptional CR dye removal efficiency (Fig. 6b). Rapid initial adsorption is observed with a 93.7% removal rate within 5 min, highlighting its swift initial action. Moreover, sustained high performance is evident, as it maintains a 94.15% removal rate after one hour, signifying consistent and effective dye removal. The high polarity of CR probably explains the different optimal contact times due to increased interactions with polar materials [47].

Experimental adsorption data for NR and CR dyes onto Fe(OH)_x@Cellulose were analyzed using the Langmuir, Freundlich, and Temkin isotherm models. The results indicated that the adsorption of both NR and CR is best described by the Langmuir model, suggesting homogeneous adsorption sites and monolayer coverage. This conclusion was supported by high R² values of 0.980 for NR and 0.973 for CR. These findings provide valuable insights into the adsorption mechanisms and can inform the design of more efficient adsorption processes. Thermodynamic parameters obtained from the isotherm models are tabulated in Table 2.

The high regression coefficients confirm that the Langmuir model accurately represents the experimental isotherms (Figs. 7 and 8), suggesting monolayer adsorption within the interlayer pores. The Langmuir constant and monolayer capacity values effectively characterize the adsorption behavior of the dyes.

The adsorption of NR and CR onto Fe(OH)_x@Cellulose highlights distinct mechanisms. NR, with its amine (-NH₂) and azo (-N=N-) groups, exhibits an adsorption mechanism based on strong hydrogen bonding with the hydroxyl (-OH) groups of cellulose. However, the formation of rigid complexes limits its adsorption capacity (Q_m= 40.55 mg/g). CR, with its sulfonate groups (-SO3−) and aromatic rings, exhibits an adsorption mechanism dominated by hydrophobic interactions and water release, explaining its high adsorption capacity (Q_m= 89.29 mg/g).

The separation factor R_L was calculated for various dye concentrations and is presented in Table 3. As R_L values for both CR and NR adsorption lie between 0 and 1, the adsorption process is favorable, indicating that Fe(OH)_x@Cellulose is an effective bio-sorbent for removing these dyes from aqueous solution [25,54]. A comparison of the dye adsorption capacities of OFIC and OFIC-Cell with those found in similar research is summarized in Table 4.

The kinetic study of NR and CR dye adsorption onto biomass was investigated. Pseudo-first-order, pseudo-second-order, and Elovich models were applied (Figs. 9 and 10). The obtained results indicated that the adsorption kinetics followed a pseudo-second-order model. The correlation coefficients for this model were high, reaching 0.999 and 0.989 for NR and CR adsorption onto Fe(OH)_x@Cellulose, respectively.

The calculated parameters for the different models are listed in Table 5. The Elovich model plot revealed a relatively high correlation coefficient for NR adsorption but a lower one for CR adsorption. This finding aligns with the ΔH values, suggesting physisorption for CR dye [6,59].

Temperature is a critical factor affecting the adsorption process, as it directly influences the rate of molecule adsorption. The adsorption of NR and CR dyes by treated Fe(OH)_x@Cellulose biomass was studied over wide temperature range from 25°C to 80°C (298 to 353 K). While NR removal rates exceeded 74% at 298 K, CR dye removal rates increased from 41% to 54% as the temperature increased from 305 to 348 K. These contrasting trends suggest distinct adsorption mechanisms for the two dyes (Fig. 11). Notably, NR removal efficiency decreased with rising temperatures, indicating an exothermic adsorption process. Conversely, the increase in CR removal rate with temperature points to an endothermic adsorption mechanism [32,60].

The impact of temperature on dye adsorption by Fe(OH)_x@Cellulose was investigated through thermodynamic analysis. The thermodynamic parameters ΔH°, ΔS° and ΔG° were calculated and grouped in Table 6. For CR, the ΔG° values were negative across the entire temperature range, indicating a spontaneous adsorption process. In contrast, for NR, the ΔG° was negative up 303 K, and became positive above this temperature, suggesting that adsorption becomes less favorable as temperature increases. A linear relationship between ln K_F and 1/T (Fig. 12) allowed for the determination of ΔH° and ΔS°. The calculated adsorption enthalpies for NR and CR were −67.153 and 3994 kJ·mol⁻¹, respectively. These results suggest that the adsorption of NR is exothermic, whereas that of CR is endothermic. The adsorption behavior of NR and CR on Fe(OH)_x@Cellulose higlights the complex interplay of intermolecular forces and thermodynamic parameters. NR’s strong affinity to Fe(OH)_x@Cellulose, as evidenced by its negative adsorption enthalpy, is attributed to favorable hydrogen bonding between their respective functional groups, specifically the hydroxyl groups (-OH) of cellulose and the amine (-NH₂) and azo (-N=N-) groups of NR. However, the adsorption process is somewhat hindered by a decrease in system disorder, as indicated by the negative entropy value (−220.32 J·mol⁻¹·K⁻¹). Conversely, CR’s adsorption, although endothermic due to factors like electrostatic repulsion between the sulfonate groups (-SO3−) of CR and the slightly negative cellulose surface, and desolvation, is ultimately promoted by an increase in system disorder (positive entropy of 26.18 J·mol⁻¹·K⁻¹), likely caused by the release of water molecules and hydrophobic interactions involving the aromatic rings of CR and cellulose [61,62].