Introduction

The discharge of radioactive waste from the nuclear fuel cycle and other nuclear applications in industry, medicine, and agriculture poses significant environmental concerns due to its toxicological and radiological influences on human health and the ecosystem (Ojovan et al. 2019). Strontium ions are toxic radionuclides in liquid radioactive wastes produced during nuclear fission (Mishra and Singh 1995). Strontium-90 is a major byproduct of uranium fission in nuclear reactors and can undergo beta decay to yield Yttrium-90 through its beta decay. Then, yttrium-90 can be regularly produced from a purified strontium-90 solution. However, yttrium-90 requires a high level of purity and is free of strontium-90 and different metal ions to achieve a substantial yield. Thus, the purification and separation of yttrium-90 from strontium-90 is noteworthy (Chakravarty and Dash 2012; Pichestapong et al. 2016). Strontium isotopes include strontium-89, with a half-life of 50.4 days, and strontium-90, with a half-life of 28 years. Strontium-90 exhibits a chemical resemblance to calcium. Consequently, when it enters the human body, it can substitute calcium, leading to anemia, leukemia, genetic alternations, and numerous diseases (Skupiński and Solecki 2014).

Several methods have been investigated for the removal of strontium from radioactive waste, such as chemical precipitation (Park 2010), solvent extraction (Okugaki et al. 2009), membrane separation (Bae et al. 2005), biological processes (Sawaki et al. 2010), and adsorption (Missana and Garcı 2007). Among the aforementioned methods, adsorption is generally considered the most practical and simple method for removing strontium ions from radioactive waste. Various adsorbents have been studied for their effectiveness in removing contaminants from water. These include nanomaterials (‏ Kasap et al. 2011), clay (Dyer et al. 1999), hydroxyapatite (Wang and Ye et al. 2008), phosphates (Kulyukhin et al. 2005; Ivanets et al. 2014), Sodium iron titanate (NaFeTiO) (Amesh et al. 2020), graphene oxide (Xing et al. 2019), magnetic chitosan beads (Chen and Wang 2012), sodium cobalt hexacyanoferrate-encapsulating alginate beads/CNTs (Vipin et al. 2014), graphene oxide–hydroxyapatite nanocomposites (GO–Hap) (Wen et al. 2014), natural clinoptilolite (Yang et al. 2008), nano tafla nanocomposite (Abu Elgoud et al. 2022b), titanate nanofibers (Wang et al. 2009), carboxymethylated cellulose (Shawabkeh et al. 2002), chitosan-fuller’s earth beads (Hasan et al. 2019), magnetic chitosan microsphere (Yin et al. 2017), silica mesoporous SBA-15 (Zhang et al. 2015), bayberry tannin (BT)–reduced graphene oxide (rGO)sponges (rGO/BT sponges) (Deng et al. 2016), magnetite nanoparticles (Fe3O4) loaded (rGO) composite (Minitha et al. 2018), TiO2-pillared zeolite (Basuki et al. 2021), and magnetic clinoptilolite/CoFe2O4 composites (Huang et al. 2017).

Chitosan is a linear polysaccharide composed of deacetylated blocks (β-(1 → 4)-linked D-glucosamine) and acetylated blocks (N-acetyl-D-glucosamine) irregularly located along the polymer chain. It is derived from alkali deacetylated chitin obtained from crabs, prawns, lobsters, and krill. Chitosan is known for its superior chemical properties and the presence of reactive amino and hydroxy groups. Chitosan’s distinctive physicochemical properties render it an optimal candidate for various applications, like bioadhesive, biocompatible, and biodegradable products. Therefore, it is well-suited for use in diverse fields, such as agricultural, food, environmental remediation, and medicine. Additionally, GO is a single layer of carbon material with a thickness of one C-atom. It contains numerous oxygen-containing functional groups, including hydroxy (-OH) and carbonyl (C–O–C) groups, on both sides, as well as carboxyl (COOH) groups on the edges. To construct efficient composites for various applications, several studies have investigated the use of GO and CS. El Rouby et al. (2018) employed GO and the functionalized GO-chitosan composite as adsorbents to remove Sr(II). They found that the maximum saturation capacity for Sr(II) was approximately 140 and 179.6 mg/g for GO and GO-chitosan composite, respectively. Zhang et al. (2018) prepared chitosan/magnetite-graphene oxide (CS/MGO) composites via a simple chemical route. Initially, they created an MGO hybrid by loading Fe3O4 particles onto the GO surface. Subsequently, CS was assembled onto MGO layers to form CS/MGO composites for the adsorption of Cr(VI). In a study by Luo et al. (2021), polydopamine (PDA) modified GO/carboxymethyl chitosan (DGO/CMC) hybrid fume was manufactured through sol–gel, self-immobilization, and lyophilization. The PDA on the GO nanosheet surface linked CMC chains through H-bonding, resulting in the construction of 3D porous aerogels. According to the Langmuir isotherm model, the prepared aerogel exhibited maximum removal capacities of 170.3, 186.8, and 312.8 mg/g for Cu2+, Ni2+, and Pb2+ ions, respectively. Chen et al. (2013) fabricated 3D network structure GO–CS hydrogels through the self-crosslinking among GO sheets and CS chains to adsorb various organic and inorganic pollutants. Wu et al. (2020) manufactured CS/GO composite aerogel microspheres using CO2 supercritical drying for bilirubin removal.

Furthermore, Vo et al. (2020) synthesized an eco-friendly GO–CS hybrid hydrogel column (GOCSCHC) and studied its use as a wide-spectrum adsorbent for removing cationic and anionic dyes from wastewater. Mohamadi et al. (2019) prepared a GO-CS composite by mixing a GO and chitosan solution, precipitating it in an alkali solution, and finally cross-linking it with glutaraldehyde to eliminate reactive blue 19 from water solutions. Omidi and kakanejadifard (2018) cross-linked GO/CS hydrogel using glutaraldehyde to adsorb cationic and anionic dyes from an aqueous environment. In a related study, Sabzevari et al. (2018) cross-linked GO with CS to create a composite (GO-CS) for the adsorption of methylene blue (MB). The characterization results indicated that cross-linking occurs between the -NH2 groups of CS and the –COOH groups of GO. Gong et al. (2019) improved the mechanical properties of CS aerogels (CSAs) by incorporating GO as a filler. Different doses of GO were added, and the porous CSAs/GO composite aerogels were structured using the freeze-drying technique. Yang et al. (2010) fabricated nanocomposites of CS and GO by self-immobilization in an aqueous solution. They observed that GO was suspended on a molecular scale in the CS solution and that interactions occurred between the CS matrix and GO sheets, which enhanced the mechanical properties of the composite. Nath et al. (2018) synthesized a pH-sensitive hydrogel composed of CS and GO through physical cross-linking for wound healing applications.

In the present study, graphene oxide nanosheets decorated with chitosan nanoparticles (GO-CSNP) were employed as an effective adsorbent for the elimination of strontium ions from aqueous solutions. The synthesis and characterization of the adsorbent were confirmed by scanning electron microscopy (SEM), X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FT-IR), thermogravimetric analysis (TGA), and Raman spectroscopy. The influence of different parameters on Sr(II) removal efficiency was evaluated in this study. In addition, the selectivity of GO-CSNP composite for the removal of strontium ions was investigated, particularly under optimal conditions, even in the presence of interfering ions, such as Cd(II), Mo(VI), Y(III), and Cs(I), as well as a mixture of (Sr(II) and Y(III)) at varying concentrations.

Experimental

Materials and chemicals

The chemicals used in this work were of analytical grade and applied without further treatment. Precisely, the following materials were utilized: graphite (200 mesh, 99.99%, Alpha Aesar), H2SO4 (95–97%, Riedel-de Haën), KMnO4 (99%, Long live), chitosan (Sigma-Aldrich, molecular weight = 310,000–375000 Da; degree of deacetylation ≥ 75%), tetraethyl orthosilicate (TEOS, 99%, Sigma-Aldrich), ethanol absolute (Sigma-Aldrich), strontium chloride hexahydrate (SrCl2· 6H2O, Sigma-Aldrich), yttrium oxide (Y2O3, Sigma-Aldrich), nitric acid (HNO3, Sigma-Aldrich), hydrogen peroxide (H2O2, 36%, Pharaohs Trading and Import, Egypt), and hydrochloric acid (HCl, 30%, El Salam for Chemical Industries, Egypt).

Preparation of graphene oxide and GO-CSNP composite

The GO was synthesized using a method similar to our previous work (Abu Elgoud et al. 2022a). The hybrid GO-CSNP composite was synthesized via a straightforward one-pot mixing technique of GO and CS, which were then linked with TEOS. To prepare the CS suspension, 4.0 g of CS was dissolved in 1000 mL of 2.0% acetic acid aqueous solution using a hot plate stirrer. Subsequently, 0.15 g of GO was introduced to the CS suspension and stirred vigorously to obtain a uniform dispersion of GO and CS (Solution A). In a separate beaker, a solution of TEOS in ethanol was prepared (Solution B). Subsequently, solution B was carefully added to the suspension Solution A. The resulting mixture was kept stirred for 48.0 h at 60 °C. The resulting solid material was isolated by filtration, washed with water, and stored for further use. Scheme 1 presents a schematic explanation of the preparation of the GO-CSNP composite.

Scheme 1.
scheme 1

Schematic presentation of GO-CSNP composite preparation method

Full size image

Characterizations (see supplementary materials).

Batch adsorption studies

A stock solution of Sr(II) was prepared by dissolving 1.0 g/L of SrCl2· 6H2O in distilled water to a specific quantity. The adsorption tests were carried out by shaking 4.0 mg of GO-CSNP composite with (5.0 mL, 100 mg/L) of Sr(II) using a shaker bath (G.F.L. 1083, Germany) at 25 ℃. After adsorption, the GO-CSNP composite was isolated from the treated solution by centrifugation at 5500 ± 5 rpm for 5 min. Each practical experiment was replicated at least three times, and the results were averaged. The initial concentration of strontium before (Co) and after sorption (Ce) was estimated using an atomic absorption spectrophotometer. The concentrations of Y(III) in the various samples were measured by the Arsenazo III method at pH 3.5 in a buffer solution with a wavelength of 650 nm (Marczenko 1976). The removal rate (R %) of Sr(II) and the sorption capacity (qo) of chitosan-decorated graphene oxide were calculated using the provided equations.

$$R \%=\left(\frac{{(\text{C}}_{\text{o}}-{\text{C}}_{\text{e}}}{{\text{C}}_{\text{o}}}\right) \times 100$$
(1)
$${\text{q}}_{{\text{o}}} = ({\text{C}}_{{\text{o}}} - {\text{C}}_{{\text{e}}} )\left( {\frac{{\text{V}}}{{\text{m}}}} \right) \left[ {{\text{mg}}/{\text{g}}} \right]$$
(2)

The distribution coefficient (Kd) of metal ions was calculated following the relation:

$${\text{k}}_{{\text{d}}} = \frac{{{\text{C}}_{{\text{o}}} {\text{ }} - {\text{ C}}_{{\text{e}}} }}{{{\text{C}}_{{\text{e}}} }}\frac{{\text{V}}}{{\text{m}}}$$
(3)
$${\text{k}}_{{\text{d}}} { = }\frac{{{\text{q}}_{{\text{e}}} }}{{{\text{C}}_{{\text{e}}} }}$$
(4)

where V is the volume (L) and m is the weight (g) of the GO-CSNP composite, Co and Ce are the initial and equilibrium concentrations (mg/L) of Sr(II), respectively.

The separation factor (SF) between metal ions was calculated according to the relation:

$$SF=\frac{{K}_{d1}}{{K}_{d2}}$$
(5)

where Kd1 and Kd2 indicate the Sr(II) and other metal ions distribution ratios, respectively.

The effect of interfering ions was carried out using a 5.0 mL solution containing 100.0 mg/L of the Cd(II), Mo(VI), Sr(II), and Cs(I), each, along with 21.0 mg of chitosan-decorated graphene oxide at pH 6.0. After the equilibrium time (1.0 min), the GO-CSNP composite was separated from the solution by centrifugation, and the mixture solution was analyzed by ICP-OES (Shimadzu Sequential Type, Kyoto, Japan) to determine the initial and residual concentration of Cd2+, Mo5+, Sr2+, and Cs+. Selectivity experiments were conducted by shaking 4.0 mg of GO-CSNP composite with 5.0 mL of the Sr(II) and Y(III) solutions at different concentration ratios for 1.0 min at pH = 4.0. After reaching equilibrium, the initial and remaining concentrations of Sr2+ and Y3+ were determined to evaluate the feasibility of their separation. The Y3+ concentrations were determined using a UV–visible double-beam spectrophotometer (model UV-160A, Japan).

Results and discussion

Characterization

SEM and TEM

The GO-CSNP composite exhibits a layered morphological structure, as evidenced by the SEM image (Fig. 1a–b). This observation was further corroborated by TEM images of GO-CSNP composite captured from different areas at 2.0 μm, as shown in Fig. 1c–d. Moreover, these images indicated the presence of uniformly distributed CS nanoparticles on the surface of the GO sheets, as shown in Fig. 1c–d. The CS nanoparticles were observed to be adhesive to the GO surfaces, with no particles were segregated away from the GO nanolayers. This unique coverage of the GO surface with CS spots is due to the presence of oxygen-functional groups (-OH and C–O–C) homogenously located at the GO layer surface, which contribute to the modification step. This can lead to uniform capture of CS on the surface of GO nanosheets, as illustrated in Fig. 1c–d. Following the adsorption of Sr(II), the layered structure of the GO-CSNP-Sr composite is not as significantly altered as that of the original GO-CS composite, as presented in Fig. 1e–f. The TEM images of the GO-CSNP-Sr composite exhibit greater detail. The CS nano spots tend to diffuse and aggregate over the GO layer, as shown in Fig. 1g–h. This observation may be due to the adsorption of Sr(II), which may result in cross-linking between two or more CS nanoparticles, leading to diffusion and aggregation.

Fig. 1
figure 1

SEM images of GO-CSNP composite (a–b) and after (e–f) adsorption of Sr(II), and TEM images of GO-CSNP composite before (c–d) and after (g–h) adsorption of Sr(II)

Full size image

TEM-EDS and mapping

The TEM-EDS and mapping are efficient tools for determining the elemental analysis of the analyzed materials and the distribution of each element in a sample, as presented in Fig. 2. The TEM-EDS of the GO-CSNP composite revealed the presence of C (GO and CS), N (CS), O (GO, CS, and TEOS) and Si (TEOS), as demonstrated in Fig. 2a. Moreover, the TEM-mapping of GO-CSNP composite indicated that these elements were homogeneously distributed throughout the analyzed area, as seen in Fig. 2b–f. After treating the Sr(II) aqueous solution with the GO-CSNP composite, the TEM-EDS data showed the presence of Sr(II) ion beside the elements identified in the GO-CSNP composite analysis data, as presented in Fig. 2g.

Fig. 2
figure 2

TEM EDS (a) and mapping (b–f) of GO-CSNP composite and TEM EDS (g) of GO-CSNP-Sr

Full size image

FTIR

FTIR is a valuable tool for obtaining information about the surface chemistry of the assessed materials. The FT-IR spectrum of chitosan (Fig. 3a) exhibits its characteristic peaks, which provide insight into the chemical composition of the material. As previously noted, the bands at 3351 and 3286 cm−1, which correspond to the stretching vibration of (O–H) of adsorbed water and the symmetric and asymmetric stretching vibration of the N–H bond in the NH2 group of the chitosan chain, respectively (Siripatrawan and Harte 2010). The peak observed at 2868 cm−1 is atibuted to the C–H stretching of the aliphatic CH2 group. The peaks at 1639 cm−1 and 1549 cm−1 are corresponding to the vibrational stretching modes of the C = O group and the scissoring motion of the –NH2 group, respectively (Brugnerotto et al. 2001). The peaks at 1417 cm−1 and 1373 cm−1 are attributed to the vibrational modes of symmetric deformation of the CH2 group (Sanuja et al. 2014). The peak at 1317 cm−1 corresponds to the amide III group resulting from the deformation combination of the NH group and CN stretching vibration and the C–O bond present at the 1021 cm−1 peak. After modifying the GO with CS, the FTIR spectra of the GO-CSNP composite differ significantly from the GO spectrum and are highly similar to the CS spectrum, with a notable shift in the CS peaks. These observations suggest a successful modification step. After the adsorption process, the GO-CSNP-Sr complex exhibited a reduction in the N–H peak intensity at 1528 cm−1 (Rahmani et al. 2017) and the emergence of a new peak at 550 cm−1, which is related to the stretching of the Sr–O bond.

Fig. 3
figure 3

a FTIR spectra of GO, CS, GO-CSNP, and GO-CSNP-Sr, b XRD pattern of GO and GO-CSNP, c Raman spectra of GO and GO-CSNP, and d TGA of GO, GO-CSNP, and GO-CSNP-Sr

Full size image

XRD

Figure 3b shows the XRD diffraction pattern of GO and GO-CSNP composite. The XRD diffractogram of the GO is typical, as obtained in our previous study (Abu Elgoud et al. 2022a), with a characteristic intense sharp peak at 2θ ≈ 9.2°. Further, chitosan possesses two main peaks at 2θ ≈ 10.0° and 20.0° (Wang et al. 2016; Onem and Nadaroglu 2014). The GO-CSNP composite was formed by immobilizing CS nanoparticles onto the GO nanosheets. The XRD diffractogram of the GO-CSNP composite exhibited a notable difference from the CS-XRD pattern and was almost identical to that of the GO, where the GO-CSNP composite exhibits an intense sharp peak at 2θ ≈ 7.0°. Notably, the modification with chitosan increased the interlayer space between the nanosheet from (d = 0.93 nm, GO) to (d = 1.26 nm, GO-CSNP composite). This behavior may improve the movement of metal aqueous solution between the GO-CSNP composite layers. Hence, the time required to reach equilibrium will be reduced, and the removal efficiency will be enhanced.

Raman

The Raman spectra of GO and GO-CSNP composite are presented in Fig. 3c. The GO showed a characteristic Raman peak at 1598 cm−1 (G-band), which relates to sp2 hybridization carbon atoms and D-band at 1352 cm−1, indicating disorder that may result from certain defects, such as vacancies due to sp3 hybrid carbon atoms. The defect ratio of the material is determined by the intensity ratio of the two bands (ID/Ig). The (ID/Ig) ratio for GO was 1.34, which was highly reduced to 0.54 for the GO-CSNP composite. This behavior may be due to the modification step, which can repair most defects on the GO basal plane. This is also evident from the TEM images (Fig. 1c-d), showing a uniform distribution of CS nano-spots throughout the GO layer.

TGA

The thermogravimetric analysis follows the decrease of the analyzed material weight with further temperature increase. The TGA of GO, GO-CSNP, and GO-CSNP-Sr composites were shown in Fig. 3d. The GO showed five degradation stages; the first stage (28–88 °C, 19.6%) was due to the release of physically adsorbed water, and the second stage (88–158 °C, 4.07%) was due to the liberation of interlayer water molecules. The pyrolysis of various oxygenated functional groups resulted in the essential decomposition stage (20.50%) observed in the range of (158–215 °C). The 4th stage (215–320 °C, 8.40%) is due to the degradation of carboxyl groups and, finally, the decomposition of the carbon skeleton of GO (5.30%) above 320 °C.

The GO can be modified with CS, resulting in CS molecules adhering to the basal plane of the GO, as shown in TEM images in Fig. 1c–d. This implies that the O-functional groups on the GO basal plane contribute to the functionalization process (Abu Elgoud et al. 2022a). This may result in a reduction in the amount of water adsorbed on the surface, which may be released at a high-temperature range (77–128 °C, 11.83%) compared to pure GO. Moreover, the CS nano-gel retains a high amount of interlayer water, reaching two folds amount retained by pure GO, and can be released at a higher temperature range (128–185 °C) than pure GO. In addition, the GO-CSNP composite showed the main pyrolysis stage at (185–336 °C, 32.65%), nearly 1.5 folds, and at high temperatures related to the GO. This behavior may be due to the decomposition of the CS backbone chain and different function groups of GO. Subsequently, the pyrolysis rate (10.46%) exhibited a gradual decline as the carbonization temperature range increased from 336 to 788 °C. When GO-CSNP and Sr are mixed, the Sr ion strongly interacts with the GO-CSNP, minimizing interlayer water content in the GO-CS-Sr (4.5%) than the GO-CSNP composite (9.4%). Moreover, the primary decomposition stage was shown at the heating range of (182–303 °C, 28.0%), i.e., with lower thermal stability. Furthermore, the following pyrolysis stages exhibited a higher degradation rate (two times) than observed for the GO-CSNP composite. This observation may be attributed to the conversion of adsorbed Sr ions to SrO at elevated temperatures, which catalytically degrades the GO-CSNP composite and increases its thermal decomposition (El Rouby et al. 2018).

Adsorption investigation

Effect of contact time

The removal rates of Sr2+ ions employing GO and GO-CSNP composite as adsorbents were followed at various contact times, as presented in Fig. 4a. Besides, the experimental results showed that about 69.0% of Sr2+ ions were captured by the GO-CSNP composite. In comparison, the removal rate of Sr2+ ions by GO is almost 52.89% and the equilibrium was reached for both GO and GO-CSNP composites within one minute. Moreover, the GO-CSNP composite was further used to study the effect of shaking time on the %R for the Y3+ ion, as presented in Fig. 4a. We observed that the equilibrium was rapidly reached and the %R related to Y3+ about (1.0 min). This rapid equilibrium time may be due to the unique layered structure of the as-prepared GO-CSNP composite and the homogenous covering of the GO layer with CS nano-spots, as shown in Fig. 1. This behavior can present the NH2 groups at a favorable location for effective interaction with the Sr2+ ions.

Fig. 4
figure 4

Effect of a contact time on the removal rate of Sr2+ using GO and GO-CSNP and Y3+ using GO-CSNP, b 2nd order kinetic model of Sr2+ and Y3+ ([M] = 100.0 mg L−1, Dose = 4.0 mg, V = 5.0 mL, pH = 6.0, T = 25 ºC) from aqueous media

Full size image

Due to the rapid equilibrium, the pseudo-first-order kinetic model was neglected. Instead, we focused on the parameters obtained from the linear plot of the pseudo-second-order kinetic model (Fig. 4b), which are listed in Table 1, showing that the correlation coefficient R2 = 0.999, close to the unit. Moreover, the experimental adsorption capacity value (qe (Exp) = 86.613 mg/g) was highly close to its calculated value (qe (Cal) = 86.613 mg/g), indicating that the PSO accurately described the adsorption kinetic process. Chemisorption occurs through H-ion exchange or electron contribution among the pollutants and amine groups of the GO-CSNP composite.

Table 1 Pseudo-second-order kinetic model parameters for the removal of Sr2+ ([Sr2+] = 100.0 mg L−1, Dose = 4.0 mg, V= 5.0 mL, pH = 6.0, T = 25 ºC) from aqueous media
Full size table

Effect of solution pH

The adsorption process is highly influenced by the pH value of the aqueous solutions, reflecting that pH not affecting only the active functional groups on the adsorbent surface but also the adsorbed species. Therefore, it is essential to study the impact of the solution pH on the adsorption performance of Sr2+ and Y3+ by applying GO-CSNP adsorbent, as presented in Fig. 5a. Moreover, the adsorption rate of GO-CS linearly increases towards Sr2+ (5.6–69%) and Y3+ (0.1–23%) in pH range (1.0–6.0). Then, the removal rate of Sr2+ remained constant with a further increase of the solution pH value to 10. This finding can be explained as follows: the pH points of zero charges (pHPZC) of GO-CSNP (Fig. 5b) demonstrated that the GO-CSNP composite was positively charged at pH < 6, while the adsorbent active sites can be negatively charged at pH > 6. This is due to the CS containing a high density of amino (−NH2) groups responsible for the adsorption of Sr2+. At a low pH value, the concentration of H+-ion is very high, which may affect the protonation of the adsorbing (−NH3+) and result in repulsion with the positive Sr2+ ion. This leads to a highly reduced adsorption performance. With a further increase in the pH value to pH ~ 6, the binding sites on the surface of GO-CSNP adsorbent become more negative, which was suitable for effective interaction with Sr2+ ions.

Fig. 5
figure 5

a Effect of the initial solution pH on the removal rate of Sr2+ and Y3+ from aqueous media, b pHZPC of the GO-CSNP composite of Sr2+ (t = 1.0 min, [Sr2+] = 100.0 mg L−1, Dose = 4.0 mg, V = 5.0 mL, T = 25 ºC) from aqueous media

Full size image

Effect of initial metal ion concentration

The impact of initial Sr2+ ions concentration on its adsorption effectiveness using GO-CSNP composite is illustrated in Fig. 6a. The removal rate decreased from 70.02 to 31.22% using Sr2+ ions concentration range of 100.0 to 1000.0 mg/L. This decrease is attributed to the fact that, at low Sr2+ ions concentration, the number of available adsorbent binding sites is sufficient to achieve a suitable adsorption rate. With a further increase in the concentration of Sr2+, the number of active adsorbent sites is lower than the number of Sr2+ ions, resulting in a lower adsorption rate. Subsequently, several adsorption isotherm models were used to describe how the metal ion interacts with the functional groups on the adsorbent surface. The linear plots of the used isotherms were presented in Fig. 6b–e, and different factors were measured and listed in Table 2. It was found that the correlation coefficient (R2 = 0.993) related to the Langmuir model recorded the highest value among all the used isotherms. This indicated that the Sr2+ is favorable to adsorb as a monolayer on the GO-CSNP composite binding sites with a saturated adsorption capacity of 473.93 mg/g. The sorption capacities of different adsorbents for Sr2+ from several investigations are presented in Table 3. The sorption capacity of GO-CSNP composite for Sr2+ in this study was higher than that of other adsorbent materials.

Fig. 6
figure 6

a Effect of initial metal ion concentration on the removal rate of Sr2+; b Linear Langmuir isotherm plots; c Linear Freundlich isotherm plots; d Linear Temkin isotherm plots; e Linear Sips isotherm plots for the sorption of Sr2+ from aqueous media, t = 1.0 min, Dose = 4.0 mg, V = 5 mL, pH = 6.0, T = 25 ºC using GO-CSNP composite

Full size image
Table 2 Results of the linear regression of the experimental data to various isotherms for Sr2+ sorbed onto the chitosan decorated graphene oxide (GO-CSNP)
Full size table
Table 3 Comparison of Sr2+ adsorption in various types of adsorbents
Full size table

Effect of adsorbent dose

The capture rate of Sr2+ was detected using various doses of GO-CSNP composite (0.004–0.008 g), as shown in Fig. 7a. We observed that the adsorption affinity of Sr2+ linearly ascending (69–87%) with further increase in the adsorbent dose in the studied range. This is attributed to the increase in GO-CSNP, which will provide extra active sites for efficient interaction with Sr2+. This result shows excellent dispersion of the adsorbent using a metal ion solution.

Fig. 7
figure 7

Effects of a adsorbent dosage (t = 1.0 min, [Sr2+] = 100 mg L−1, V = 5 mL, pH = 6.0, T = 25 ºC), b temperature and c thermodynamic parameters (t = 1.0 min, [Sr2+] = 100 mg L−1, Dose = 8.0 mg, V = 5 mL, pH = 6.0), d The sorption capacity of Sr2+ and Y3+ in the different concentration ratios at pH = 4.0, and e Effect of interfering ions on the removal percent of Sr2+ from aqueous solution

Full size image

Effect of temperature

The influence of temperature on the Sr2+ ions removal rate by GO-CSNP composite was investigated in the temperature range of 25.0–65.0  °C. As illustrated in Fig. 7b, the removal rate of Sr2+ is enhanced by an increase in the temperature in the used range. The favorability of Sr2+ ions adsorption at high temperatures may be referred to as the reduction of the solution viscosity, the motion of the Sr2+ ions improved, and the easily reachable of the Sr2+ ions from the bulk of solution to the binding sites on the adsorption. Finally, this can enhance the effectiveness of adsorption (Egbosiuba et al. 2021). To further evaluate the thermodynamic factors of Sr2+ ions captured by GO-CSNP, the linear thermodynamic plot and calculated factors are presented in Fig. 7c and Table 4, respectively, according to the following equations (Khalil et al. 2022).

$$ {\Delta G}^\circ { } = { } - {\text{ RT ln Kd}},$$
(6)
$$\Delta {\text{G}}^\circ = \Delta {\text{H}}^\circ - {\text{T}} \Delta {\text{S}}^\circ$$
(7)
$$- {\text{RT}} {\text{ln}} {\text{Kd}} = \Delta {\text{H}}^\circ - {\text{T}} \Delta {\text{S}}^\circ$$
(8)
$${\text{Ln}} {\text{Kd}} = \left( {\Delta {\text{S}}^\circ / {\text{R}}} \right) + \left( { - \Delta {\text{H}}^\circ / {\text{R}}} \right) 1/{\text{T}}$$
(9)
Table 4 Thermodynamic parameters for removal of Sr2+ ions
Full size table

The − ΔG° values at various temperatures indicate the spontaneous adsorption of Sr2+ on the active sites of GO-CSNP adsorbent. Furthermore, an increase in the temperature results in increasing the more negative values of ΔG°. The results suggest that the adsorption process can be improved at higher temperatures. In addition, the positive value of ΔH° indicated the endothermic nature of the adsorption reaction. Moreover, the ΔS° positive values illustrated the enhancement of the randomness at the Sr2+-GO-CSNP solution interface.

Separation of Sr 2+ from Y 3+

Radioactive yttrium (Y-90) is a carrier-free radionuclide commonly used for therapeutic purposes  (IAEA 2009). The radionuclides of Y-90 are produced from the decay of the long-lived fission product parent 90Sr (T1/2: 28.8 years). In this concern, relatively increased amounts of the parent 90Sr radionuclides are used to produce the daughter, short-lived radionuclides 90Y (T1/2:2.67 days). Therefore, the effect of the mass ratio of Sr to Y on the adsorption capacity of the GO-CSNP composite was investigated. In this respect, two different concentration ratios of 100 mg/L Sr2+ to 100 mg/L Y3+, and 1000 mg/L Sr2+ to 100 mg/L Y3+, in solution at pH = 4.0 and mixed for 1.0 min. It was found that, when using a similar concentration of Sr2+ and Y3+ (100 mg/L) in solution, the adsorption capacity of the GO-CSNP composite towards Sr2+ and Y3+ was close to each other, as presented in Fig. 7d. While using a high concentration ratio of Sr2+  = 1000 mg/L and Y3+  = 100 mg/L, the GO-CSNP composite showed that the adsorption capacity towards Sr2+ is five folds higher than that for the adsorption capacity of Y3+, as shown in Fig. 7d. Further work is under investigation to assess the possibility of using GO-CSNP composite to be used as a suitable generator for high-purity yttrium from strontium.

Effect of interfering ions on the removal of Sr 2+

The removal activity of the GO-CSNP composite, in terms of distribution coefficient Kd (L/g), was performed for the removal of Sr2+ from a waste solution containing Cd2+, Mo6+, and Cs+ ions, as presented in Fig. 7e. The experiment involved shaking 21.0 mg of the GO-CSNP composite with 5.0 mL of a solution containing 100 mg/L of each ion for a contact time of 1.0 min at a pH of 6.0. The experimental obtaining indicated that the GO-CSNP composite could remove (74.25% of Sr2+, 67.50% of Mo+6, 63.16% of Cd2+, and 50.82% of Cs+. Further, the separation factor values between Sr2+ and Cd2+, Mo6+ and Cs+ are 1.70, 1.40, and 1.60, respectively, see Table 5. These results indicated that the GO-CSNP composite tends to remove multi-ions that interfere with the same solution, making the prepared GO-CSNP composite suitable for practical applications.

Table 5 Effect of interfering ions on the removal of Sr2+ by using GO-CSNP composite
Full size table

Desorption of Sr (II) from loaded GO-CSNP composite

In this section, we study the desorption of Sr2+ from the loaded composite. The GO-CSNP-Sr was treated with several reagents to achieve this target (Table 6). All the used reagents were unsuccessful in the complete regeneration of the GO-CSNP-Sr complex, as seen in Table 6. This may be due to the strong interaction between GO-CSNP and Sr. According to TEM images in Fig. 1g–h, the interaction between the GO-CSNP and Sr alters the morphology of the CS nanoparticles into the fused structure.

Table 6 Desorption of Sr ions from GO-CSNP-Sr complex using different reagents
Full size table

Conclusions

Chitosan-decorated graphene oxide (GO-CSNP) composite was synthesized using a simple method. The as-prepared composite (GO-CSNP) was characterized using SEM, FTIR, XRD, Raman, TGA, and EDX. The kinetics results confirmed that the sorption process followed the pseudo-second-order model. The Langmuir isotherm indicated that the GO-CSNP composite is a suitable sorbent with a maximum sorption capacity of 473.93 mg/g for Sr(II). The experimental results reported that the GO-CSNP composite can effectively remove Sr(II) with high removal efficiency from waste solutions containing multielement. Moreover, the GO-CSNP composite has a sorption capacity for Sr(II) of 319.0 mg/g and Y(III) of 61.61 mg/g at a high concentration ratio (Sr: Y = 10: 1). These findings suggest the potential for separating Sr(II) and Y(III) under optimal conditions.