Green chitosan–starch–polyvinyl alcohol bio-composite hydrogel for zirconium capture from zircon raffinate Authors: G. A. Dakroury, Sh. M. Abdo, E. A. A. El Shazly Published: 2026 ABSTRACT A novel polymer-based sorbent comprising three blended components was synthesized via aqueous mixing and ultrasonication for the efficient removal of Zr(IV) from highly acidic solutions. Characterization by FT-IR, SEM, and surface area analysis confirmed active functional groups and a mesoporous structure (16.704 m2 g⁻1, 3 nm pores). Batch studies revealed that maximum uptake (102 mg g⁻1) occurred at pH 1.25 after 120 min at 20 °C, increasing to 188 mg g⁻1 with higher initial concentrations (50–750 mg L−1). Uptake decreased at elevated temperatures (293–323 K), indicating exothermic adsorption. Kinetic analysis was consistent with adsorption involving surface interactions and was best described by the pseudo-second-order and Elovich models, while equilibrium data were better fitted by the Freundlich model, suggesting a heterogeneous adsorption surface. Thermodynamic parameters (ΔG = –18.84 to –19.60 kJ mol−1; ΔH = − 11.35 kJ mol−1; ΔS = 25.57 J mol−1 K−1) confirmed a spontaneous and exothermic process. The sorbent remained stable in acidic media up to 2 M H2SO4 and was effectively regenerated, with H2SO4 achieving the highest desorption efficiency. These findings suggest that the prepared material shows potential as an effective sorbent for Zr(IV) recovery from acidic process streams.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13065-026-01841-x. Introduction Zirconium is a transition metal derived mainly from the mineral zircon and is represented by the chemical symbol Zr with atomic number 40. Owing to its low thermal neutron capture cross-Section (0.185 barns) [1], high corrosion resistance, and excellent mechanical stability at elevated temperatures, zirconium plays a significant role in nuclear technology and nuclear chemistry [2]. It is widely used in nuclear reactors, particularly in fuel cladding, core structural components, and chemical reprocessing systems [3]. The growing global demand for nuclear energy has increased the need for efficient and sustainable methods for the extraction, purification, and recovery of zirconium from various industrial process streams [4, 5]. During the hydrometallurgical processing of zircon-bearing minerals, Zr(IV) is typically leached in acidic media and subsequently separated using solvent extraction systems. Although solvent extraction is widely applied on an industrial scale, it generates an aqueous waste stream known as zircon raffinate [5], which still contains considerable concentrations of zirconium along with other metal ions. This raffinate represents not only a loss of valuable zirconium but also an environmental concern due to its high acidity and metal content [6]. Therefore, recovering Zr(IV) from raffinate is essential for improving process efficiency, minimizing waste generation, and supporting circular economy strategies in nuclear materials management [7]. Despite its effectiveness, solvent extraction suffers from several limitations, including high organic solvent consumption, emulsion formation, poor selectivity in complex matrices, and environmental and safety concerns [8, 9]. These drawbacks have motivated researchers to explore alternative, cleaner, and more cost-effective separation technologies. Among these, adsorption using polymer-based sorbents has emerged as a promising approach owing to its operational simplicity, environmental compatibility, and reusability [5, 10]. Biopolymer-based adsorbents, particularly those incorporating chitosan and starch, have attracted considerable attention due to their abundance, biodegradability, and excellent metal-binding capabilities [11]. Chitosan (CS), derived from chitin, contains amino and hydroxyl functional groups that readily interact with metal ions through chelation and electrostatic attraction. Its high affinity for multivalent cations and ease of modification make it a promising material for the development of advanced sorbents [12]. Starch (St), another naturally abundant biopolymer, is cost-effective and contains numerous hydroxyl groups that enhance hydrophilicity and provide additional sorption sites when blended with other polymers. However, starch exhibits relatively poor mechanical properties, particularly low strength and flexibility, due to its semicrystalline structure [13]. Polyvinyl alcohol (PVA) is commonly incorporated into biopolymer matrices to improve film-forming ability, mechanical strength, and thermal stability, as well as to support the formation of three-dimensional polymeric networks [14] suitable for metal sorption under harsh conditions such as acidic leachates. The combination of chitosan, starch, and PVA provides a synergistic composite material with improved chemical stability, high porosity, abundant functional groups, and good resistance to acidic conditions. These properties make the composite highly suitable for zirconium recovery from raffinate solutions. However, despite the extensive use of biopolymers for heavy-metal recovery, studies focusing specifically on Zr(IV) sorption from zircon raffinate using chitosan–starch–PVA composites remain very limited. Moreover, the behavior of such composites in highly acidic and metal-rich environments typical of nuclear raffinate streams is still insufficiently understood. A quantitative comparison of the maximum Zr(IV) adsorption capacity of the present chitosan–starch–polyvinyl alcohol (CSP) hydrogel with previously reported sorbents is presented in Table S1. Although some reported materials exhibit higher adsorption capacities (e.g., MCCh-N, 180 mg g⁻1), the CSP hydrogel demonstrates a favorable combination of acid stability (up to 2 M H2SO4), applicability to real zircon raffinate solutions, and relatively high adsorption capacity (102–188 mg g⁻1) under strongly acidic conditions that were not evaluated for many previously reported sorbents. In this work, a novel chitosan–starch–polyvinyl alcohol (CSP) composite hydrogel was prepared and evaluated for the adsorption of Zr(IV) from zircon raffinate. The effects of sorption parameters, including pH, contact time, volume-to-mass ratio, and initial concentration, were systematically investigated to optimize the recovery efficiency. The adsorption mechanism was further evaluated using kinetic, isotherm, and thermodynamic models. The findings provide insight into the potential of CSP biopolymer composites as environmentally friendly, cost-effective, and efficient materials for zirconium recovery from acidic process streams, thereby contributing to the sustainable management of nuclear process residues and the development of advanced separation technologies. Materials and methods Materials and reagents Most chemicals used in this study were of analytical grade and used as received without further purification. Low-molecular-weight chitosan (75–85% degree of deacetylation) was obtained from Sigma-Aldrich (USA). Acetic acid (CH3COOH) was supplied by ADWIC (Egypt). Polyvinyl alcohol (PVA) with an average molecular weight of 88.000 g mol−1 was purchased from Acros Organics. The zirconium raffinate solution was provided by the Nuclear Materials Authority; it was a waste stream generated after the solvent extraction of zirconium using trioctylamine (TOA). The elemental composition of the zircon raffinate was analyzed by inductively coupled plasma optical emission spectrometry (ICP-OES). The major elements detected (in mg L⁻1) were Zr (999.8), Ti (52.8), Hf (22.6), P (4.85), K (2.67), Ce (1.66), and Al (0.97). The free acidity of the raffinate was determined to be 1.8 M H2SO4. Hydrochloric acid (HCl) and Sodium hydroxide (NaOH), obtained from El-Nasr Co. (ADWIC Egypt) were used for pH adjustment. Double-distilled water was employed for the preparation of all solutions throughout the experimental work. Suggested mechanism for CSP bio-composite hydrogel preparation To prepare the CSP bio-composite, 0.5 g of chitosan was dissolved in 100 mL of 1% (V/V) glacial acetic acid and magnetically stirred until a clear and uniform solution was obtained. This step increases chitosan solubility and prepares – NH3+ groups for ionic interactions1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{Chitosan-}}{\mathrm{NH}}_{{2}} + {\text{acetic acid}} \to {\mathrm{Chitosan-}} {\mathrm{NH}}_{3}^{ + } $$\end{document} In a separate beaker, 2g of starch and 3g of PVA were dispersed in 100 mL of distilled water and heated with continuous stirring until both polymers were completely dissolved and formed a viscous solution. Then, 1 g of citric acid was added as a crosslinking agent to the warm PVA–starch solution. Citric acid acts as a multifunctional crosslinker through hydrogen bonding:2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{l}{\text{Citric acid-}}{\text{COOH }} + {\text{ HO-}}{\mathrm{PVA}} \\ \hfil \quad \to {\text{Citric acid-}}{\mathrm{COOH-}}\mathrm{-}\mathrm{-}{\mathrm{HO=}}{\mathrm{PVA}} \end{array}$$\end{document}3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{l}{\text{Citric acid-}}{\mathrm{COOH}} + {\mathrm{HO-}}{\mathrm{starch}} \\ \hfil \quad \to {\text{Citric acid-}}{\mathrm{COOH-}}\mathrm{-}\mathrm{-}{\mathrm{HO-}} {\mathrm{starch}} \end{array}$$\end{document} Or esterification:4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{Citric acid-}}{\text{COOH }} + {\mathrm{HO-}}{\mathrm{PVA}} \to {\text{Citric acid-}}{\mathrm{COO-}}{\mathrm{PVA}} + {\mathrm{H}}_{{2}} {\mathrm{O}} $$\end{document}5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{Citric acid-}}{\mathrm{COOH}} + {\mathrm{HO-}}{\text{starch }} \to {\text{Citric acid-}}{\mathrm{COO-}}{\mathrm{starch}} + {\mathrm{H}}_{{2}} {\mathrm{O}} $$\end{document} The chitosan solution and the citric acid/PVA/starch solution were then combined and magnetically stirred for 2 h to ensure homogeneity. This is followed by resting the mixture for 24 h to remove air bubbles and allow complete interaction between components. An electrostatic attraction occurs between -NH3+ and ester -COO group. This interaction forms ionic crosslinking points represented as:6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{l}{\text{Citric acid-}}{\mathrm{COO-}}{\mathrm{starch}} + {\mathrm{Chitosan-}}{\mathrm{NH}}_{3}^{ + } \\ \hfil \quad \to {\text{Citric acid-}}{\mathrm{starch-}}{\mathrm{COO}}^{ - } \ldots .{}^{ + }{\mathrm{NH}}_{{3}}{\mathrm{-Chitosan}} \end{array}$$\end{document}7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{or}}\quad {\mathrm{OH-}}{\mathrm{starch}} + {\mathrm{Chitosan-}}{\mathrm{NH}}_{3}^{ + } \to {\mathrm{starch-}}{\mathrm{O}}^{ - } \ldots .{}^{ + }{\mathrm{NH}}_{{3}}{\mathrm{-Chitosan}} $$\end{document}8\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{l}{\text{Citric acid-}}{\mathrm{COO-}}{\text{PVA }} + {\text{ Chitosan-}}{\mathrm{NH}}_{3}^{ + } \\ \hfil \quad \to {\text{Citric acid-}}{\mathrm{PVA}}{\mathrm{-COO}}^{ - } \ldots .^{ + } {\mathrm{NH}}_{{3}}{\mathrm{-Chitosan}} \end{array}$$\end{document}9\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{or}}\quad {\mathrm{OH-}}{\mathrm{PVA}} + {\mathrm{Chitosan-}}{\mathrm{NH}}_{3}^{ + } \to {\mathrm{PVA-}}{\mathrm{O}}^{ - } \ldots .^{ + } {\mathrm{NH}}_{{3}}{\mathrm{-Chitosan}} $$\end{document} Furthermore, hydrogen bonding occurs between -OH groups of chitosan and the -OH groups of starch or PVA.10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{OH-}}{\mathrm{starch}} + {\mathrm{Chitosan-}}{\mathrm{OH}} \to {\mathrm{starch-}}{\mathrm{OH}} \ldots .{\mathrm{HO-}}{\mathrm{Chitosan}} $$\end{document}11\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{OH-}}{\mathrm{PVA}} + {\mathrm{Chitosan-}}{\mathrm{OH}} \to {\mathrm{PVA-}}{\mathrm{OH}} \ldots .{\mathrm{HO-}}{\mathrm{Chitosan}} $$\end{document} These interactions promote tight chain alignment and facilitate the formation of a cohesive and stable composite structure. Afterward, the resulting blend was subjected to Ultrasonic treatment for 30 min to enhance polymer dispersion network formation. The final mixture was dried at 50°C until a solid/film composite was obtained, the removal of water brings the polymer chains closer together, strengthening the hydrogen-bonded network and reinforcing physical crosslinking. The dried composite was ground into a fine powder for subsequent sorption experiments. Figure 1 represents a schematic diagram for preparation and sorption process of Zr(IV) onto CSP bio-composite hydrogel. And the mechanism for preparation represented in Fig. 1b.Fig. 1a Schematic diagram for CSP bio-composite hydrogel preparation and sorption of Zr (IV) from Zircon raffinate solution. b Suggested mechanism for preparation a Schematic diagram for CSP bio-composite hydrogel preparation and sorption of Zr (IV) from Zircon raffinate solution. b Suggested mechanism for preparation Instrument DAIHAN Scientific ultrasonic cleaner (model WUC-AOIH, 60 Hz, 230 V; Korea) was employed to ensure thorough homogenization of the hydrogel mixture. Fourier transform infrared (FT-IR) spectra were obtained using the KBr pellet technique on a PerkinElmer 1600 spectrophotometer over the range of 400–4000 cm⁻1. The surface morphology and elemental composition of the prepared composite were analyzed using scanning electron microscopy (SEM) coupled with energy-dispersive X-ray spectroscopy (EDX) and electron backscatter diffraction (EBSD) on an FEI Quanta FEG-250 instrument. The nitrogen adsorption–desorption measurements were carried out using a Quantachrome NOVA BET surface area analyzer (Nova Win software, version 11.04). Prior to analysis, the samples were degassed under vacuum to remove adsorbed moisture and gases. The Brunauer–Emmett–Teller (BET) method was applied to determine the specific surface area, while the pore size distribution was obtained from the desorption branch using the BJH model. Zirconium concentration in solution were quantified spectrophotometrically using the alizarin red method. Adsorption studies The uptake behavior of Zr(IV) ions on the synthesized CSP bio-composite hydrogel is investigated by examining the influence of key operational parameters. The effects of solution pH, contact time, initial zirconium concentration, adsorbent dose (V/m), and temperature are systematically assessed. All sorption experiments were conducted in triplicate to ensure reproducibility. The experimental conditions applied for the Zr(IV) adsorption studies are summarized in S2. A known mass of CSP bio-composite hydrogel (0.05) was placed in a measured volume of raffinate solution (20 mL) and agitated until equilibrium was reached. Then the solid phase was separated using Whatman No.40 filter paper. The efficiency of Zr(IV) retained by the sorbent was determined using Eq. (12) where Co is the initial concentration of Zr(IV) in raffinate and Cf is its final concentration in raffinate. While the sorption capacity of zirconium per gram, qe (mg g−1) is calculated using Eq. (13)12\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ Sorption efficiency \left( {A\% } \right) = \left( {\frac{{C_{0} - C_{f} }}{{C_{0} }}} \right) \times 100 $$\end{document}13\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ q_{e} = \left( {C_{o} - C_{f} } \right)\frac{V}{m} $$\end{document}where V is the mass of the solution (20 mL) and m is the mass of sorbent (0.05 g). Kinetic and isotherm studies To elucidate the mechanism and rate of Zr(IV) sorption onto CSP bio-composite hydrogel, several kinetic models are applied, including pseudo first order (PFO), pseudo second order (PSO), and pseudo nth order (PNO), and Elovich. These models help identify the controlling steps and evaluate the chemisorption or diffusion-dominated behavior of the process. To investigate the equilibrium sorption behavior, four isotherm models; Langmuir, Freundlich, Dubinin-Redushkevich (D-R) and Temkin were applied to describe the interaction of Zr(IV) ions with the functional groups of CSP bio-composite hydrogel. S3 summarizes the nonlinear equation form of the applied models and the description for each model. Error function Model performance was assessed using the coefficient of determination (R2) and the non-linear Chi-square error (χ2) and residual error. The most suitable model is indicated by a higher R2 along with a lower χ2 value, reflecting the closest agreement between the experimental and predicted data. Equations (14, 15) clarify their calculations where: qexp is the amount of Zr(IV) sorbed onto the CSP hydrogel that experimentally determined (mg g−1), qcalc. is the theoretically predicted sorbed amount (mg g−1). qmean is the mean value of qexp (mg. g−1) values of Zr(IV), n is the experimental data points number.14\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ R^{2} = 1 - \mathop \sum \limits_{i = 1}^{n} \left( {q_{calc.} - q_{\exp .} } \right)^{2} /\mathop \sum \limits_{i = 1}^{n} \left( {q_{calc.} - q_{mean} } \right)^{2} $$\end{document}15\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ \chi^{2} = \sum {\frac{{\left( {q_{calc.} - q_{\exp ..} } \right)^{2} }}{{q_{calc.} }}} $$\end{document} Thermodynamic parameters Thermodynamic variables were calculated to assess the spontaneity of the sorption process and its nature. Equations 16–19 represent the evaluation of thermodynamic key parameters. Gibbs free energy (ΔG) provides insight into the spontaneity of the adsorption process; negative ΔG values indicate a spontaneous reaction, whereas positive values denote non-spontaneity. The Enthalpy Change (ΔH) reflects whether heat is absorbed or released during sorption, with positive ΔH signifying an endothermic process and negative ΔH indicating an exothermic one. Entropy change (ΔS) describes the degree of disorder at the solid–liquid interface, where positive ΔS values imply increased randomness upon adsorption and negative values indicate a decrease in system disorder.16\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta G= \Delta H-T\Delta S$$\end{document}17\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta G=-RT\mathrm{ln}{K}_{d}$$\end{document}18\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{ln}{K}_{d}= \frac{\Delta S}{R }+ \frac{-\Delta H}{R}. \frac{1}{T}$$\end{document}19\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${K}_{d}= \frac{{q}_{e}}{{C}_{e}}$$\end{document} Desorption studies Desorption experiments were conducted to assess the regeneration capability of the hydrogel Zr(IV)-loaded hydrogels were washed, separated, and treated with different eluents. The Zr(IV) concentration in the desorbed solution was analyzed, and desorption efficiency was calculated using:20\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ Desorption \% = \frac{{C_{aq} }}{{C_{s} }}\% $$\end{document}where: Caq is the concentration of Zr(IV) in the aqueous phase and Cs is the concentration of Zr(IV) retained in the CSP bio-composite hydrogel. Results and discussion Characterization of CSP hydrogel FT-IR analysis The FTIR spectra of CSP bio-composite hydrogel before and after Zr(IV) loading from Zircon raffinate in Fig. 2a investigates characteristic spectral changes confirming the interaction of Zr(IV) ions with the functional groups -OH, -NH2, and carbonyl containing groups of the polymeric matrix (chitosan, starch, and PVA). This agrees with the strong affinity of Zr(IV) for oxygen donor ligands. The broad absorption band around 3300 cm−1 corresponds to overlapping of -OH groups of PVA, starch and -NH2 of chitosan functional group. This confirms presence of a dense hydrogen-bonding network formed during composite preparation. The decrease in intensity of this band after Zr(IV) loading indicates possible interaction of Zr(IV) with -OH and/or -NH2 groups [15]. The peaks at 2920–2850 cm−1 before and after loading indicates C-H stretching bands. A band in the range 1720–1730 cm−1 indicates C = O stretching of ester group This supports the formation mechanism described in (Eqs.4–5). The peak near 1580–1630 cm−1 corresponds to asymmetric stretching of -COO− and amid bending from chitosan that support electrostatic interaction between NH3+ and citrate -COO− in preparation mechanism Eq. (6) and (8). The strong peak at 1200 cm−1 of CSP bio-composite hydrogel before loading related to C-O-stretching band of polysaccharide and ester C-O of citric cross linking shifted to 1000 cm−1 with decrease in intensity after Zr(IV) loading [16]. A noticeable decrease in the intensities of OH, NH2 bands at 1650 and 1550 cm−1, respectively after loading confirm the Zr(IV) sorption onto CSP bio-composite beads. The appearance of the shoulder at 650 cm−1 is attributed to the formation of Zr-O bands and deformation of polysaccharide skeleton [16]. These findings support the assumption for inner sphere complexing mechanism for Zr(IV) sorption onto CSP bio-composite hydrogel.Fig. 2a FTIR analysis b BET isotherm c SEM d TGA-DTA of CSP bio-composite hydrogel before Zr(IV) loading from Zirconium raffinate a FTIR analysis b BET isotherm c SEM d TGA-DTA of CSP bio-composite hydrogel before Zr(IV) loading from Zirconium raffinate BET surface area Figure 2b represents BET isotherm for CSP bio-composite hydrogel with adsorption and desorption branches. The gradual increase in adsorbed volume with relative pressure (P/P0) and the presence of distinct hysteresis loop corresponds to Type IV isotherm behavior that indicates that the material likely exhibits mesoporous structure. Based on the shape of the adsorption–desorption loop in Fig. 2b, the hysteresis most closely resembles H3 -type pointed to non-rigid slit- shaped pores characteristics of polymers. This result confirms that the hydrogel possesses a heterogeneous pore structure suitable for adsorption of metal ions. Table 1 summarizes the estimated surface parameters of CSP bio-composite hydrogel. The nitrogen adsorption–desorption analysis revealed that the prepared CSP bio-composite hydrogel exhibits BET surface area 16.704 m2/g, indicating the presence of accessible adsorption sites suitable for zirconium uptake. The BET constant reflects a moderate adsorbate–surface interaction. The average pore diameter is 3 nm confirm its mesopore structure. Such pore dimensions are suitable for diffusion of Zr(IV) ions where the ionic radius and the hydration radius require pore openings in the mesoporous range. Thus CSP bio-composite hydrogel is suitable for efficient Zr(IV) sorption.Table 1Estimated surface parameters of CSP bio-composite hydrogel at 20°CParameterValueBET surface area16.704 m2/gC- constant18.114Total pore volume0.0124 cm3/gAverage pore diameter3 nm Table 1Estimated surface parameters of CSP bio-composite hydrogel at 20°CParameterValueBET surface area16.704 m2/gC- constant18.114Total pore volume0.0124 cm3/gAverage pore diameter3 nm Estimated surface parameters of CSP bio-composite hydrogel at 20°C Surface morphology Figure 2c reveals a rough and irregular surface with a heterogeneous texture of CSP bio-composite hydrogel. The roughness is indicative of porosity and interstitial voids, which is favorable for adsorption applications because it increases surface area and provides more active sites for sorption. The scale bar indicates 50 µm, showing that pores and surface irregularities are in the micrometer range suggest efficient diffusion pathways for metal ions. Thermal analysis The thermal behavior of the CSP bio-composite hydrogel was investigated in Fig. 2d using simultaneous DTA–TGA analysis over the temperature range up to ~ 600 oC. The TGA curve exhibits multi-step weight loss, while the DTA profile shows corresponding exothermic and endothermic thermal peaks, confirming the composite nature and thermal stability of the hydrogel. For the 1st stage (30−150°C), a slight weight loss (11.988%) is observed in the TGA curve, accompanied by a weak exothermic DTA peak due to physical relaxation and restructuring as water desorbs, which stabilizes local hydrogen bonds before the onset of chemical degradation. Although moisture loss is typically associated with endothermic dehydration, in polysaccharide-based composites the removal of loosely bound and structurally entrapped water can be coupled with rearrangement of hydrogen-bonded networks and relaxation of polymer chains, yielding a net exothermic signal. Similar exothermic low-temperature events have been reported in chitosan-based systems where early structural reorganization occurs during moisture release prior to the major degradation steps [17]. The 2nd stage (200−350°C), a pronounced weight loss (71.81%) occurs in this region, accompanied with strong endothermic DTA peak centered around ~ 250–300 °C. This stage is mainly associated with the thermal degradation of the polymer backbone; decomposition of starch glycosidic bonds, deacetylated chitosan chain scission, and partial dehydration and degradation of PVA chains [18]. The 3rd stage (400–500°C) shows a further gradual weight loss (13.72%) is detected, accompanied by a clear endothermic DTA peak around ~ 480–520°C. This endothermic behavior is attributed to further decomposition of thermally stable polymer fragments and carbonaceous residue after the main degradation. This high-temperature endothermic event is attributed to the further decomposition of thermally stable polymer fragments and carbonaceous residue formed after the main degradation [19]. EDX analysis The elemental composition of the CSP bio-composite hydrogel before and after Zr(IV) loading was investigated in Fig. 3a, b respectively using EDX analysis to confirm metal uptake and elucidate the sorption mechanism. Before Zr(IV) sorption, the EDX spectrum of CSP hydrogel shows the presence of C, O, and N as the dominant elements, with weight percentages of 38.5 wt.% C, 51.3 wt.% O, and 10.2 wt.% N. These elements originate from the polymeric backbone of chitosan, starch, and PVA, which are rich in hydroxyl (-OH) and amine (-NH2) functional groups. The absence of any metal-related peaks confirms the purity of the prepared bio-composite. After Zr(IV) sorption, a clear Zr signal appears, with a high content of 21.6 wt.%, directly confirming the successful sorption of Zr(IV) ions onto the CSP bio-composite hydrogel. In addition, minor amounts of Na, Cl, and S are detected. Sodium may originate from residual neutralization steps or ionic exchange processes, while sulfur and chloride are attributed to sulfate and chloride counter-ions associated with the Zr(IV) precursor solution, further supporting the ion-exchange and complexation mechanism. These findings strongly support the suitability of the CSP bio-composite hydrogel as an effective sorbent for recovery of Zr(IV) in acidic media.Fig. 3EDX analysis of CSP bio-composite hydrogel a-before and b-after Zr(IV) loading EDX analysis of CSP bio-composite hydrogel a-before and b-after Zr(IV) loading Sorption studies pH effect Figure 4a shows that the sorption of Zr(IV) onto CSP bio-composite hydrogel is strongly influenced by the pH change of the solution. As the pH of the solution increases, the sorbed amount of Zr(IV) increases. At low pH, the affinity of the active sites toward Zr(IV) ions is reduced due to the high concentration of H+ which competes Zr(IV) sorption and the presence of [Zr (H2O)8] complex species at strongly acidic conditions and electric repulsion available protonated active sites [20]. Increasing the pH decreases the degree of protonation, consequently, reduces repulsion and enhances the accessibility of binding sites and Zr(IV) uptake. The precipitation curve (Fig. 4c) confirms that at pH 1.25, less than 1% of Zr(IV) is lost due to hydrolysis or colloid formation in the absence of sorbent, ensuring that the measured removal is entirely attributable to adsorption onto the CSP hydrogel Fig. 4b, the species distribution diagram determined via Hydra/Medusa chemical equilibrium software [21], indicates the formation of hydrolysis polymeric species as Zr (OH)3+ and Zr (OH)22+ which has stronger affinity towards oxygenated active sites and an increase in sorption efficiency. After pH 1.25, the sorption gradually approaches a plateau due to saturation of the available active sites, and the sorption process reached equilibrium. Moreover, Zr(IV) tends to form neutral or colloidal hydroxide species at higher pH which impedes the sorption process. Therefore, maintaining the pH within optimal range is essential to ensure the dominant mechanism remains surface complex rather than precipitation. The sorbed amount at this pH reached 102 mg g−1. Figure 4d depicts the chemical stability of the prepared hydrogel in hydrochloric acid solutions within a molar range 0.1–3 M H2SO4. The weight loss (or solubility percentage) remains essentially zero till 2 M H2SO4 indicates that the composite network structure remains stable. A slight increase in solubility was observed at 2.5 M H2SO4, consistent with the early stages of acid-catalyzed hydrolysis of glycosidic and related acetal/ether linkages in the polysaccharide backbone. Under sufficiently strong acidic conditions, protonation and cleavage of glycosidic bonds begin to occur, leading to partial swelling and minor polymer dissolution without catastrophic breakdown of the composite matrix [22]. A more noticeable solubilization appears at 3 M H2SO4 due to strong protonation and disruption of hydrogen bonded domains, hydrolysis of carbohydrates linkages, and increased polymer chain mobility. Overall, the chemical stability confirms that the composite is highly stable in the working acidity range crucial for working with acidic zircon raffinate solutions and a practical advantage for application in hydrometallurgical effluent treatment.Fig. 4a Effect of pH on the sorption of Zr(IV) onto CSP bio-composite hydrogel b Speciation diagram stability of Zr(IV) c Precipitation curve of Zr(IV) d Chemical stability of CSP bio-composite hydrogel. (Error bars represent ± one standard deviation (n = 5)) a Effect of pH on the sorption of Zr(IV) onto CSP bio-composite hydrogel b Speciation diagram stability of Zr(IV) c Precipitation curve of Zr(IV) d Chemical stability of CSP bio-composite hydrogel. (Error bars represent ± one standard deviation (n = 5)) Time effect and kinetic study Figure 5a represents the impact of time on the uptake of Zr(IV) onto CSP bio-composite hydrogel. The adsorption process exhibited an initial rapid uptake followed by a slower approach to equilibrium. In the first 20–40 min, the sorbed amount increases sharply reaching ~ 60–80 mg g−1. This rapid stage is mainly attributed to the abundance of vacant active sites on the composite surface. Beyond ~ 40–60 min, the sorption rate begins to slow significantly due to occupation of active sites [5]. The equilibrium was reached within 120 min with an equilibrium sorbed amount 102 mg g−1. Increasing contact time beyond this point results have no remarkable sorption uptake confirming the composite has reached its maximum uptake under the given condition [23].Fig. 5a Effect of time on the sorption of Zr(IV) onto CSP bio-composite hydrogel b non-linear kinetic modeling fitting of quasi-1st, quasi 2nd and Elovich modelling c Effect of Concentration on the sorption of Zr(IV) onto CSP bio-composite hydrogel d Non-linear Isotherm modeling fitting of Langmuir, Freundlich, Temkin and D-R modelling (Error bars represent ± one standard deviation (n = 5)) a Effect of time on the sorption of Zr(IV) onto CSP bio-composite hydrogel b non-linear kinetic modeling fitting of quasi-1st, quasi 2nd and Elovich modelling c Effect of Concentration on the sorption of Zr(IV) onto CSP bio-composite hydrogel d Non-linear Isotherm modeling fitting of Langmuir, Freundlich, Temkin and D-R modelling (Error bars represent ± one standard deviation (n = 5)) The sorption kinetics of Zr(IV) onto the CSP bio-composite hydrogel (Fig. 5b) were evaluated and the calculated parameters were summarized in Table 2.Table 2Kinetic parameters for Zr(IV) sorbed onto CSP bio-composite hydrogelParametersKinetic modelZr(IV)Pseudo-1st – order (PFO)qt (mg g−1) (calculated)102.639 ± 7.323K1 (min−1)0.3212 ± 0.00437R20.96164χ235.63225Pseudo-2nd – order (PSO)qt (mg g−1) (calculated)124.793 ± 5.652K2 (g mg−1 min−1)2.8035 × 10–4 ± 5.063 × 10–5R20.977χ220.0845qt (mg g−1) (experiment)102Elovichα6.6695 ± 1.09221β0.03237 ± 0.00281R20.977χ220.667The highest R2 and lowest χ2 are in italics Table 2Kinetic parameters for Zr(IV) sorbed onto CSP bio-composite hydrogelParametersKinetic modelZr(IV)Pseudo-1st – order (PFO)qt (mg g−1) (calculated)102.639 ± 7.323K1 (min−1)0.3212 ± 0.00437R20.96164χ235.63225Pseudo-2nd – order (PSO)qt (mg g−1) (calculated)124.793 ± 5.652K2 (g mg−1 min−1)2.8035 × 10–4 ± 5.063 × 10–5R20.977χ220.0845qt (mg g−1) (experiment)102Elovichα6.6695 ± 1.09221β0.03237 ± 0.00281R20.977χ220.667The highest R2 and lowest χ2 are in italics Kinetic parameters for Zr(IV) sorbed onto CSP bio-composite hydrogel The highest R2 and lowest χ2 are in italics The PFO model showed an acceptable correlation (R2 = 0.962) The qt calculated (102.64 mg g−1) is relatively close to qt experimental. However, the large (χ2 = 35.63) indicates that the model does not sufficiently represent the sorption mechanism. The PSO model was selected as the most appropriate kinetic description because it produced the highest coefficient of determination (R2 = 0.977) and least reduced chi-square (χ2 = 20.084). Although, PSO model slightly overestimates the ultimate capacity (qt calculated = 124.8 mg g−1 and qt experimental = 102 mg g−1), its behavior is consistent with rate control by chemical sorption processes (electron sharing/coordination) between Zr(IV) and functional groups (-NH2, -OH) of the hydrogel [24]. The Elovich model also performed well (R2 = 0.977; χ2 = 20.667), comparable to PSO. This model is typically associated with activated chemisorption on heterogeneous surface. And the parameters (α = 6.67, β = 0.032) indicate progressive surface coverage, where the sorption energy decreases as the hydrogel surface becomes saturated [25]. Among the tested models, the Elovich and PSO equations gave the highest R2 values (0.977). While these fits are consistent with chemisorption-controlled processes, they do not independently prove a specific mechanism. The Elovich model, which assumes heterogeneous surface binding energies, provided a good description without overestimating the equilibrium capacity, and is therefore used here as a useful empirical descriptor for engineering design. Concentration effect and Isotherm model The effect of initial Zr(IV) concentration on the adsorption performance of CSP bio-composite hydrogel is illustrated in Fig. 5c. As the initial concentration increases from 50 to 750 mg/L, the sorbed amount of qe increases from 18.8 mg g−1 to188.43 mg g−1. This is due to the increase in driving force for mass transfer at higher Co values. When the concentration gradient between the solution and the sorbent surface becomes larger, Zr(IV) ions diffuse rapidly through the boundary layer and into the hydrogel network and the available active sites (NH2, OH, C-O….) within the CSP bio-composite. The slight decrease in qe at initial concentration 1000 mg/L, this is due to the saturation of available binding sites and the occupancy of the active sites [26]. Figure 5d shows the evaluation of the fitness of the isotherm model Langmuir, Freundlich, Dubinin-Radushkevich (D-R) and Temkin models. The estimated parameters were listed in Table 3. Among the four models, the Freundlich model yielded the best fit (R2 = 0.938, χ2 = 0.0278), suggesting that the sorbent surface is heterogeneous in terms of binding energy. This is expected for a ternary polymer composite. However, good fit for the Freundlich model does not by itself prove multilayer adsorption; it simply indicates that a simple Langmuir (homogeneous) description is insufficient [27]. The lower R2 of Langmuir compared to Freundlich suggests the assumption of a homogeneous surface and monolayer coverage is less suitable for this system. The Temkin model produced R2, χ2 values close to Freundlich indicate that the heat of adsorption decreases linearly with the surface coverage. The Temkin constant b further supports the role of moderate interactions between Zr(IV) ions and the hydrogel functional groups (-NH2 and -OH). This confirms that the sorption process is influenced both by binding energy distribution and by sorbate–sorbent electrostatic interactions.Table 3Adsorption isotherm parameters for Z r(IV) sorbed onto CSP hydrogel sorbentParametersAdsorption isotherm modelsZr(IV)Langmuir modelqmL (mmol g−1)2.031 ± 0.17364b3.280 ± 1.142R20.8997χ20.0453Freundlich modeln3.81 ± 0.5409Kf (mmol n−1g−1.L−n)1.337 ± 0.07123R20.938χ20.0278D-R modelqmD-R1.796 ± 0.1573βD-R3.6076 × 10–8 ± 1.1844 × 10–8R20.835χ20.0745Temkin modelk90.145 ± 45.0122b3.12 ± 0.363R20.938χ20.0296The highest R2 and lowest χ2 are in italics Table 3Adsorption isotherm parameters for Z r(IV) sorbed onto CSP hydrogel sorbentParametersAdsorption isotherm modelsZr(IV)Langmuir modelqmL (mmol g−1)2.031 ± 0.17364b3.280 ± 1.142R20.8997χ20.0453Freundlich modeln3.81 ± 0.5409Kf (mmol n−1g−1.L−n)1.337 ± 0.07123R20.938χ20.0278D-R modelqmD-R1.796 ± 0.1573βD-R3.6076 × 10–8 ± 1.1844 × 10–8R20.835χ20.0745Temkin modelk90.145 ± 45.0122b3.12 ± 0.363R20.938χ20.0296The highest R2 and lowest χ2 are in italics Adsorption isotherm parameters for Z r(IV) sorbed onto CSP hydrogel sorbent The highest R2 and lowest χ2 are in italics The D-R isotherm yielded the lowest correlation (R2 = 0.835) and the highest χ2 value (0.0745), indicating a poorer fit. This suggests that pore-filling is not the dominant mechanism. The D–R monolayer capacity qmD-R = 1.796 mmol g−1 is comparable to the Langmuir value, but the very small βD-R value (3.6076 × 10–8 mol2 kJ⁻2) indicates low mean adsorption energy, consistent with ion exchange and surface-complexation rather than true micropore filling. Effect of temperature and thermodynamic study The effect of temperature on the sorption performance provides important insight into the nature of the interaction between Zr(IV) ions and the functional groups of CSP bio-composite. As shown in Fig. 6a the sorption process is exothermic in nature and thermodynamically favored at low temperatures. This behavior could be due to the slight decrease in protonation at higher temperature, which reduces the number of positively charged –NH3+ groups capable of attracting and binding Zr(IV) which further decreases sorption [28].Fig. 6a Effect of temperature b V/m ratio on the sorption of Zr(IV) onto CSP bio-composite hydrogel c Thermodynamic study (Error bars represent ± one standard deviation (n = 5)) a Effect of temperature b V/m ratio on the sorption of Zr(IV) onto CSP bio-composite hydrogel c Thermodynamic study (Error bars represent ± one standard deviation (n = 5)) To confirm this behavior, the standard thermodynamic parameters ΔG, ΔH, and ΔS were calculated using the Van’t Hoff equation, and the obtained values are summarized in Table 4 from the slope and intercept of Fig. 6c. The negative values of ΔG indicate that the adsorption process is spontaneous within the investigated temperature range. The negative ΔH value (− 11.35 kJ mol⁻1) confirms the exothermic nature of the adsorption process, which is consistent with the experimentally observed decrease in adsorption capacity with increasing temperature. This behavior suggests that lower temperatures favor the interaction between Zr(IV) ions and the active functional groups of the CSP bio-composite hydrogel. The positive ΔS value indicates a slight increase in randomness at the solid–solution interface during adsorption, which may be attributed to the release of water molecules from the hydration sphere of Zr( IV) during surface complexation [29]. Overall, the thermodynamic results support a spontaneous and exothermic adsorption process favored at lower temperatures. The obtained thermodynamic behavior is consistent with adsorption involving specific interactions such as surface complexation or inner-sphere coordination rather than weak physical adsorption. Similar thermodynamic trends have been reported for metal-ion adsorption onto polymer-based sorbents [26].Table 4Thermodynamic parameters for the sorption of Zr(IV) onto CSP bio-composite hydrogelTemperatureThermodynamic parameters∆ HkJmol−1∆SJmol−1 K−1∆GkJmol−1293− 11.3525.57− 18.84313− 19.34323− 19.60 Table 4Thermodynamic parameters for the sorption of Zr(IV) onto CSP bio-composite hydrogelTemperatureThermodynamic parameters∆ HkJmol−1∆SJmol−1 K−1∆GkJmol−1293− 11.3525.57− 18.84313− 19.34323− 19.60 Thermodynamic parameters for the sorption of Zr(IV) onto CSP bio-composite hydrogel Effect of volume/mass ratio The impact of the volume/mass ratio (V/m) is a key operational parameter affecting the sorption process and the availability of active functional groups on the sorbent. As shown in Fig. 6b, the sorbed amount increases as the V/m ratio increases. This is because the increasing in the V/m ratio means that a given volume of solution contains more Zr(IV) ions relative to the limited number of available active sites [30]. This enhances the driving force for mass transfer and allows a larger fraction of the sorption sites to become occupied, thereby increasing the adsorption capacity. The observed increase in q with higher V/m ratios suggests that the adsorbent possesses a sufficiently high number of accessible active sites capable of binding additional Zr(IV) ions when more ions are available in the solution. Desorption and regeneration The desorption results presented in Fig. 7a were used to evaluate the most effective desorbing agent for the recovery of sorbed Zr(IV). Among the tested eluents, sulfuric acid (H2SO4) exhibited the highest desorption efficiency, indicating its effectiveness as a stripping agent for Zr(IV) recovery from the loaded adsorbent [5]. The higher stability constant of the Zr(IV)–sulfate complex suggests stronger interactions between Zr(IV) and SO42− ions compared with Cl⁻ ions [31]. In contrast, HNO3 exhibited moderate desorption efficiency, while deionized water showed negligible desorption performance, confirming its poor ability to disrupt the interactions between Zr(IV) ions and the sorbent surface [32]. The desorbed amount increased markedly with increasing H2SO4 concentration. At low acid concentrations, desorption efficiency remained limited because the proton concentration was insufficient to effectively disrupt the strong interactions between Zr(IV) ions and the functional groups on the sorbent surface. Zr(IV) ions are believed to interact strongly with oxygen-containing functional groups through electrostatic attraction and surface complexation; therefore, stronger acidic conditions are required to weaken these interactions. As the H2SO4 concentration increased, the availability of H⁺ ions also increased, promoting competition between protons and Zr(IV) ions for the active adsorption sites. This facilitated the release of Zr(IV) ions from the sorbent surface into the solution, resulting in a significant increase in desorption efficiency. The highest desorption percentage was obtained at the maximum tested H2SO4 concentration, confirming the effectiveness of sulfuric acid in regenerating the adsorbent.Fig. 7Desorption efficiencies of Zr(IV) loaded onto CSP Bio-composite hydrogel a using different eluents b using different concentrations of H2SO4 c reusability (error bars represent ± one standard deviation (n = 5)) Desorption efficiencies of Zr(IV) loaded onto CSP Bio-composite hydrogel a using different eluents b using different concentrations of H2SO4 c reusability (error bars represent ± one standard deviation (n = 5)) Figure 7b shows that once the H2SO4 concentration reached 0.2 M, the desorption efficiency became nearly constant, suggesting that equilibrium conditions had been achieved. Increasing the acid concentration beyond this value did not significantly enhance metal release, indicating that the desorption process was governed mainly by proton exchange and disruption of surface–metal interactions rather than diffusion limitations. The reusability of the CSP hydrogel was evaluated over five consecutive adsorption–desorption cycles using 0.5 M H2SO4 as the eluent (Fig. 7c and Table 5). After the first cycle, the desorption efficiency reached 94.2 ± 1.5%, while the adsorption capacity decreased from 102 to 97.5 mg g⁻1, corresponding to a loss of only 4.4%. After five cycles, the adsorption capacity remained at 86.3 mg g⁻1 (84.6% of the initial value), indicating good reusability with only gradual performance loss.Table 5Reusability of CSP hydrogel over five consecutive adsorption–desorption cycles: adsorption capacity (mg g⁻1), desorption efficiency (%), and retained capacity relative to cycle 1 (%)CycleAdsorption capacity (mg g⁻1)Desorption efficiency (%)Retainedcapacity(%)1102.0 ± 2.194.2 ± 1.5100299.3 ± 1.892.8 ± 1.297.4395.6 ± 2.091.0 ± 1.493.7491.2 ± 1.989.5 ± 1.389.4586.3 ± 2.287.1 ± 1.684.6 Table 5Reusability of CSP hydrogel over five consecutive adsorption–desorption cycles: adsorption capacity (mg g⁻1), desorption efficiency (%), and retained capacity relative to cycle 1 (%)CycleAdsorption capacity (mg g⁻1)Desorption efficiency (%)Retainedcapacity(%)1102.0 ± 2.194.2 ± 1.5100299.3 ± 1.892.8 ± 1.297.4395.6 ± 2.091.0 ± 1.493.7491.2 ± 1.989.5 ± 1.389.4586.3 ± 2.287.1 ± 1.684.6 Reusability of CSP hydrogel over five consecutive adsorption–desorption cycles: adsorption capacity (mg g⁻1), desorption efficiency (%), and retained capacity relative to cycle 1 (%) Selectivity of CSP hydrogel for Zr(IV) in real raffinate Because all sorption experiments were performed directly using real zircon raffinate (composition given in Sect. "Materials and reagents"), the selectivity of the CSP hydrogel can be directly evaluated from concentration changes. of coexisting ions. Table 6 presents the initial and final concentrations of the major elements present in the raffinate after treatment under the standard conditions (pH 1.25, 120 min, 20 °C, V/m = 400 mL g⁻1).Table 6The initial and final concentrations of the major elements present in the raffinate after sorption onto CSP bio-composite hydrogelElementInitial concentrationFinal concentrationRemoval efficiency%Al0.97390.36366462.6P4.8460520.12835297.4K2.6724720.79914470.1Ti52.809210.39040499.3Zr999.808618881.2Ce1.657880.6226662.4Hf22.6136421.48335 Table 6The initial and final concentrations of the major elements present in the raffinate after sorption onto CSP bio-composite hydrogelElementInitial concentrationFinal concentrationRemoval efficiency%Al0.97390.36366462.6P4.8460520.12835297.4K2.6724720.79914470.1Ti52.809210.39040499.3Zr999.808618881.2Ce1.657880.6226662.4Hf22.6136421.48335 The initial and final concentrations of the major elements present in the raffinate after sorption onto CSP bio-composite hydrogel The CSP hydrogel removed 81.2% of Zr(IV), confirming its high affinity. However, Hf, a chemically similar element that acts as a neutron poison in nuclear applications – showed only 5% removal (uptake 0.45 mg g⁻1). This demonstrates that the CSP hydrogel suggest that the CSP hydrogel has good selectivity toward Zr from Hf under strongly acidic conditions, which is an important requirement for producing nuclear-grade zirconium. In contrast, the poor separation between Zr and Ti suggests that a pre-removal step for Ti may be necessary if high-purity Zr is desired. Other coexisting ions (Al, Ce, Th, K, P) showed low uptake values (≤ 1.9 mg g⁻1). Overall, the CSP hydrogel exhibits favorable selectivity for Zr(IV) over Hf, but significant competition from Ti(IV) must be considered when applying this sorbent to Ti-bearing raffinates. Conclusion The polymer-based material prepared in this study demonstrated high efficiency for the removal of Zr(IV) from strongly acidic process solutions. Characterization confirmed that the material contains active functional groups and a mesoporous structure that supported metal uptake. Sorption studies demonstrated that Zr(IV)) removal is strongly affected by pH, contact time, initial concentration, temperature, and liquid-to-solid ratio. The best performance was obtained at pH 1.25, where an uptake of 102 mg g−1 was reached within 120 min. The decrease in adsorption capacity at higher temperatures, together with the calculated thermodynamic parameters, confirmed that the process is spontaneous and exothermic. Kinetic results indicated that chemisorption is the controlling mechanism, while equilibrium data were best described by the Freundlich model, confirming a heterogeneous surface with multilayer interaction. The material also showed excellent stability in strong acids and could be regenerated effectively, especially using sulfate solutions, which achieved the highest desorption efficiency. Overall, the prepared sorbent is a promising, low-cost option for recovering Zr(IV) from harsh acidic streams and can contribute to more sustainable metal recovery processes. Supplementary Information Supplementary Material 1 Supplementary Material 1