Introduction

Nuclear energy accounts for a significant proportion of the global energy supply. By 2023, 440 nuclear power reactors would have contributed approximately 10% of global electricity supply, with some advanced economies accounting for more than 70%1,2. However, the enormous threat posed by potential nuclear leaks and contamination to the global ecosystem cannot be overlooked3,4,5. Threatened by occasionally happened nuclear accidents, researchers have paid more attention on works for rapid and effective detection and adsorption of nuclear waste elements such as UO22+, Sr2+, Am3+, and I− in recent years6,7,8,9,10. The chemical sensors for UO22+, Sr2+, and I− utilized optical or electrical signals as readout, have achieved developments on sensitivity and selectivity, which is important for nuclear accident warning and subsequent disposal11,12,13,14. Among numerous nuclear waste elements, strontium-90 (90Sr) exhibits a long half-life period of 29.1 years, indicating radiation hazards for hundreds of years15,16. In fact, 90Sr is one of the most significant residual radioactivity sources from the Chernobyl nuclear accident, with radioactivity levels exceeding 103 Bq/m2 in an area of 30,000 km2 surrounding the reactor core17,18,19,20. In the Fukushima nuclear accident, the radioactivity generated by 90Sr was enormous, reaching up to 106 Bq/L in the treated radioactive water tanks21. Furthermore, strontium, owing to its chemical similarity to calcium, is easily absorbed by human bones, causing bone cancer, myeloma, or tumors in nearby soft tissue22,23. Therefore, researchers worldwide are paying significant attention to the detection of strontium in nuclear waste and the environment.

For long-term Sr2+ detection, methods such as inductively coupled plasma atomic emission spectrometry (ICP–AES)24, atomic absorption spectrometry (AAS)25, X–ray fluorescence spectroscopy (XRF)26, and potentiometric methods27,28,29 have been developed. However, the sophistication and complexity of these measuring methods limit their applicability for in-situ and real-time monitoring. To address this limitation, various chemosensors for Sr2+ detection have been developed via fluorescent, electrochemical, and colorimetric methods30,31,32,33. Nonetheless, the amount of strontium in nuclear waste has been traced to be far below the detection limits of current strontium-ion chemosensors. For example, the ion concentration of Sr2+ in the treated radioactive water tanks of Fukushima is calculated as 0.4 fM to 2.0 nM based on its radioisotope concentration21, whereas most detection limits are greater than 10 nM (Supplementary Table 1). Furthermore, the presence of numerous interfering ions and impurities poses significant challenges in terms of selectivity for complex natural environmental monitoring. In a recent study, a guanine quadruplex-based DNA biosensor was designed, and it demonstrated remarkable selectivity with a detection limit of 2.11 nM6. However, there is still potential for improvement in the sensing limit and response time, which are currently set at 15 min, to meet the practical need to detect environmental strontium.

Ion-selective optical sensors, also known as optodes, are widely known for their fast response and high selectivity. The optodes operate on an ion-exchange principle, in which the exchange of target cations and protons (or positively charged indicators) causes a significant optical signal change34,35. Owing to their precision and selectivity, ion-selective optodes have been successfully utilized to detect complex samples such as blood, serum, seawater, and mineral-water36,37,38. Among various samples, alkali and alkaline earth metals, including Na+, K+, Ca2+, and Li+ are commonly used as model ions38,39,40,41. However, there have been limited reports of optodes designed specifically for strontium detection, primarily owing to the scarcity of specific strontium capture molecules. Strontium-selective electrodes, which are the counterparts of optodes, have employed macrocyclic compounds such as crown ethers and nitrogen-containing heterocycles for strontium combination29,42,43. However, the detection limits in these cases exceed 75 nM, which is significantly higher than the concentration range of 0.4 fM to 2.0 nM of Sr2+ discovered in the Fukushima nuclear wastewater tanks, limiting the ability of electrodes to detect trace amounts of strontium.

This study presents an ultra-sensitive strontium-selective optical sensor based on the N,N,N′,N′,N″,N″-Hexacyclohexyl-4,4′,4″-propylidynetris(3-oxabutyramide) complex as the specific Sr2+ recognition ligand (Sr2+-ligand) for Sr2+ binding (Fig. 1a). The Sr2+-ligand, along with other sensing components, has been integrated into a nanoparticle sensing system designated as the Sr2+-nanosensor, demonstrating a fast response and a low detection limit of 0.5 nM for strontium ions. The stability constant calculation reveals a high binding affinity for Sr2+ and Sr2+-ligand. Furthermore, comprehensive investigations were conducted to demonstrate the coordination behavior between the strontium and oxygen atoms from the amide and alkoxy groups in the Sr2+-ligand. The findings of this study provide valuable insights for the design of Sr2+-ligand-based strontium detection methodologies as well as a promising sensing method for detecting trace amounts of environmental radioactive strontium pollution.

Fig. 1: Mechanism and sensing response of Sr2+-nanosensor.
figure 1

a Schematic diagram for sensing mechanism of Sr2+-nanosensor. b Absorbance spectra of the Sr2+-nanosensor to Sr2+ from 0.1 nM to 0.1 M in Tris-HCl solution (pH 7.0). c Photographs of the color change of Sr2+-nanosensor suspension towards Sr2+ at varying concentrations, with the numbers indicating the log values of the Sr2+ concentration. d Fitting of the absorbance at 658 nm with Sr2+ concentrations from 0.1 nM to 0.1 M. e Fitting for LOD calculation. Error bars are the standard error (SE) across replicates, n = 3.

Full size image

Results

Fabrication of Sr2+-nanosensor

As matrix material, a brush block copolymer (PS-PEO) with hydrophobic polystyrene (PS) and hydrophilic poly(ethylene oxide) (PEO) chains were used, as was the plasticiser bis(2-ethylhexyl) sebacate (DOS). The sensing Sr2+-nanosensor was prepared via a precipitation method (Supplementary Fig. 1)44. Lipophilicity allowed hydrophobic sensing compounds, including an indicator (Chromoionophore I, CHI), an ion-exchanger (sodium tetrakis-[3,5-bis(trifluoromethyl)-phenyl] borate, NaTFPB), and the Sr2+-ligand to be embedded in the nanoparticles. The PS-PEO utilized in this study possesses PS and PEO chains, facilitating the encapsulation of lipophilic sensing compounds within the nanosensor and the distribution of nanosensor in aqueous environments.

The chemical structures of sensing compounds and the sensing mechanism are presented in Fig. 1a. Previous work shows that the lipophilicity of the sensing molecule enables it stay in the organic phase with little leakage to aqueous phase45. The immersing of the sensing films in 10 mM Sr2+ solution for 12 h also proved the stability of sensing molecules in organic phase with ignorable release of the sensing molecules (Supplementary Fig. 2). The optical properties of Sr2+-ligand remain unchanged when in contact with Sr2+. To maintain charge conservation in the Sr2+-nanosensor, Sr2+-ligand would extract Sr2+ from an aqueous solution, causing protons to be released from the sensing phase of the Sr2+-nanosensor. This process would cause a change in the protonation degree of the indicator, which is easily detectable using optical readout methods. The Sr2+-nanosensor exhibits a uniform diameter of 15.3 nm as determined via transmission electron microscopy (Supplementary Fig. 3) and a hydrodynamic size of 25.2 ± 1.2 nm with a small particle dispersion index of 0.10 (Supplementary Fig. 4).

Absorbance responses to Sr2+ of Sr2+-nanosensor

The absorbance spectra of the Sr2+-nanosensor change at 658 and 530 nm (Fig. 1b), corresponding to the protonated and deprotonated forms of the indicator, respectively. As the concentration of strontium ions increased, a noticeable color transition from purple-red to blue was observed (Fig. 1c). The color response curve based on the hue value of the Sr2+-nanosensor is presented in Supplementary Fig. 5. The absorbance peak at 658 nm is used to characterize Sr2+ concentration because that it is less affected by reverse changing peak at 541 nm and it contains maximum absorbance change scale as the Sr2+ concentration varied from 0.1 nM to 0.1 M. The Sr2+-nanosensor generated an S-shaped calibration curve at 658 nm for Sr2+ concentrations from 0.1 nM to 0.1 M at pH 7.0 (Fig. 1d). Notably, the limit of detection (LOD) of the Sr2+-nanosensor was determined to be 0.5 nM using the previously reported method, in which the LOD was defined as a concentration that causes a 3% change in the initial signal readout (Fig. 1e)46.

The absorbance response was pH-dependent, which was caused by the proton-involved sensing process (Supplementary Fig. 6). Thus, the detection of Sr2+ by Sr2+-nanosensor should be operated under buffered solution to maintain a stable pH value. The temperature caused little interference to the sensing results from 20 °C to 40 °C, indicating a good thermal stability of the nanosensor (Supplementary Fig. 7). Ion-selective optodes are known to be versatile because of the multiple choices of indicators. The nanosensor could also been designed in fluorescence mode for Sr2+ detection by using a fluorescence indicator, Chromoionophore III (Supplementary Fig. 8). Besides, the sensing results were stable with different batches of Sr2+-nanosensor (Supplementary Fig. 9). Three different batch of nanosensor caused little sensing deviations. We also evaluated the effect of dilution on the nanosensor response and only rare change of the signal was observed with a constant 10 mM Sr2+ background (Supplementary Fig. 10).

The selectivity of the Sr2+-nanosensor is crucial because natural samples are often complex and contain multiple interfering ions. In this study, Sr2+-nanosensor was used to detect 31 common cations and anions, including Ba2+, Co2+, Cr3+, Ca2+, Fe3+, K+, Li+, Mg2+, Mn2+, Na+, Ni+, Cs+, Zn2+, NH4+, CNS−, OH−, SO42−, CO32−, HCO3−, HPO42−, Cl−, NO3−, Br−, Ac−, SO32−, MnO42−, HPO4−, HSO3−, NO2−, and F− (Fig. 2a). At the same concentration of 1 μM, the interfering ions only barely affect the signal change of the Sr2+-nanosensor. Although calcium and barium ions exhibit chemical properties similar to those of strontium ions, the absorbance response of strontium was discovered to be 2.39 and 2.50 times greater than that of calcium and barium, respectively. The Sr2+-ligand has been utilized to detect lithium ions47, but when exposed to 100 times and even 1000 times more lithium ions, the Sr2+-nanosensor responded similarly to strontium ions, indicating a stronger affinity for Sr2+ than for Li+ (Fig. 2b). Figure 1c demonstrates the rapid optical response of the Sr2+-nanosensor, with the signals stabilizing in less than 10 s following stirring. Long-term functionality tests revealed that the Sr2+-nanosensor responded well to Sr2+ even after following 40 d of storage in the dark, indicating exceptional stability and shelf life (Supplementary Fig. 11).

Fig. 2: Sensing properties of Sr2+-nanosensor.
figure 2

a Individual ion of 1 μM in causing the absorbance change. A0 and A represent the absorbance of the Sr2+-nanosensor before and after contact with ion, respectively. b Absorbance responses at 658 nm to Sr2+ and Li+. Equation S8 is used to calculate the protonated degree, represented as 1-α. c Time evolution of the absorbance change at 658 nm with various Sr2+ concentrations. d A comparison of Sr2+ concentrations before (c0) and after (c) sensing process. e Calibration curves for Sr2+-nanosensor with and without Sr2+-ligand. f Measurement of Sr2+ concentrations in natural water samples. The detection results obtained via ICP–MS are also shown. The data are shown as the mean ± standard deviation (n = 4). g A comparison of the applicable Sr2+ detection ranges for currently available chemical methods. The methods denoted by the reference numbers are given in Supplementary Table 2. The dashed red line represents the maximum concentration of 90Sr in the Fukushima water tanks. Error bars are the standard error (SE) across replicates, n = 3.

Full size image

Following Sr2+-nanosensor detection, the aqueous solution of Sr2+ samples was collected and analyzed via ICP–MS (Fig. 2d). The Sr2+ samples here ranged from 10−8 M to 10−1 M, as Sr2+ concentrations below 10−8 M were insufficient for ICP–MS to produce precise results. Polyvinyl chloride (PVC) films containing an equivalent amount of sensing components as those in the Sr2+-nanosensor were utilized to easily separate the organic sensing phase from the aqueous sample phase (Supplementary Fig. 12). At low concentrations (less than 10−7 M), the Sr2+ concentration decreased significantly following the sensing process, whereas the concentration change was negligible at higher concentrations (greater than 10−6 M). This phenomenon could elucidate the wide sensing range and high sensitivity in the ultra-low concentration of this Sr2+-ligand-based sensing method. At low Sr2+ levels, the Sr2+-ligand can concentrate the majority of Sr2+ from the aqueous to the organic sensing phase, resulting in a highly sensitive signal response. Meanwhile, at high concentrations, the amount of Sr2+ in the aqueous phase remains almost constant, resulting in a wide-range signal response.

The stability constant (or overall complex formation constant, β) is a key physical parameter to describe the affinity of the Sr2+-ligand to strontium ion48,49. The definition of β is described by Eq. (1), where the L is the Sr2+-ligand and LpSrns2+ represents the adduct of Sr2+-ligand (L) and Sr2+ ions, respectively:

$$\beta=\frac{\left[{{{\rm{L}}}}_{{{\rm{p}}}}{{{\rm{Sr}}}}_{{{{\rm{ns}}}}}^{2+}\right]}{{[{{\rm{L}}}]}^{{{\rm{p}}}}[{{{\rm{Sr}}}}_{{{{\rm{ns}}}}}^{2+}]}$$
(1)

To determine the stability constant of strontium ion and Sr2+-ligand, a previously reported calculation method was used to compare the response of sensors with and without Sr2+-ligand50. The method for β value calculation was represented in Supplementary Methods. In the absence of Sr2+-ligand, the absorbance spectra change only at high Sr2+ concentrations, indicating that Sr2+ phase transfer at low concentration from the aqueous sample to the nanospheres is not favored (Supplementary Fig. 13). Figure 2e depicts the protonation degree responses of Sr2+-nanosensor for Sr2+ with and without Sr2+-ligand. The Sr2+-nanosensor exhibits a log β value of 6.0 ± 0.1, which is at least 2 orders higher than the previously reported Sr2+-sensing ligands, including 18-crown-6 (3.4), 21-crown-7 (1.8) and 15-crown-5 (2.6)51,52, indicating a high affinity between the Sr2+-ligand and strontium ion. The high affinity of Sr2+-ligand to Sr2+ promotes the ion exchange process, in which the combination of Sr2+-ligand and Sr2+ forces the deprotonation of the indicators. Thus, low concentrations of strontium ions would cause obvious signal changes of the indicators, resulting a lower detection limit. These findings suggest that the Sr2+-nanosensor has potential applications in environmental samples.

Application of Sr2+-nanosensor in real samples

The Sr2+-nanosensor was used to measure strontium ion concentrations in mineral water and nuclear wastewater, demonstrating its practical application potential. As a control, the concentration of Sr2+ in the samples was determined using ICP–MS. The measured Sr2+ concentration levels were found to be similar to those obtained using the ICP–MS method (Fig. 2f). Preliminary experiments have demonstrated the potential for strontium detection in polluted environmental samples. In comparison to currently available chemical methods for strontium ion detection (Fig. 2g), the Sr2+-nanosensor sensing method demonstrated an ultralow detection limit of 0.5 nM and a rapid response time of less than 10 s. Furthermore, the sensing range is wide, spanning from 0.5 nM to 0.1 M, covering most Sr2+ concentrations in real samples such as seawater (90.1 µM) and drinking water set by the U.S. Environmental Protection Agency (USEPA 3.4 nM)6. Therefore, this detection method is capable of monitoring changes in environmental strontium concentrations caused by nuclear waste or leakage.

Mechanism for high affinity and selectivity

To better understand the high affinity and selectivity of the Sr2+-nanosensor for strontium, a detailed investigation of the interaction mechanism between the Sr2+-ligand and strontium ions was conducted. X-ray photoelectron spectroscopy (XPS) was used to investigate the binding of strontium ions to the Sr2+-ligand. The mixture of NaTFPB and Sr2+-ligand was analyzed before and after interaction with strontium. To avoid interference from free strontium, the sensing compounds were dissolved in dichloromethane, and the strontium ions were dispersed in water. To remove free strontium, the organic phase was separated and washed with deionised water after being mixed, standing, and layering. After interacting with strontium, the Na 1s peak at 1090.05 eV vanished, while the Sr 3d peak appeared at 120.02 eV, indicating an ion-exchange process has happened (Fig. 3a). Furthermore, after the interaction of Sr2+-ligand and SrCl2, the O 1s peaks of Sr2+-ligand shifted from 532.90 and 531.24 eV to 530.23 and 529.17 eV (Fig. 3b), while the Sr 3d peaks of SrCl2 decreased from 135.98 and 134.23 eV to 131.91 and 130.15 eV (Fig. 3c), demonstrating a chemical interaction between the strontium atoms in SrCl2 and the oxygen atoms in the Sr2+-ligand. An extended X-ray absorption fine structure (EXAFS) analysis was performed to investigate the strontium coordination profiles in the Sr2+-ligand. Figure 3d depicts the k-edge spectrum collected at 20 °C. Figure 3e depicts the Fourier transforms of the Sr k-edge EXAFS data (R space) for Sr2+-ligand solid samples following interaction with SrCl2. Strontium has three first-shell oxygen atoms at 2.43 Å and three second-shell oxygen atoms at 2.56 Å, indicating that Sr coordinates with six O atoms in the Sr2+-ligand molecule. The fitting agreed well with the experimental data (Supplementary Table 3). To correlate these Fourier transform peaks with k-space data, the wavelet transform was used (Fig. 3f). The intensity is highest at (3.7 Å–1; 1.8 Å), corresponding to the oxygen atoms surrounding the central strontium atom. These results illustrate that strontium coordinates with six oxygen atoms in the Sr2+-ligand, forming a Sr–O₆ configuration.

Fig. 3: Binding mechanism of Sr2+-ligand with Sr2+ ions.
figure 3

a XPS analysis of Sr2+-ligand before and after interaction with Sr2+ ions and the SrCl2 solid. b High-resolution XPS spectra of O 1s in Sr2+-ligand with and without Sr2+ ions. c High-resolution XPS spectra of Sr 3d in SrCl2 before and after interaction with Sr2+-ligand. EXAFS analysis of the k-space spectrum (d), the R-space spectrum (e), and the wavelet transform analysis for the interaction of Sr2+-ligand and Sr2+ ions (f). The x-axis of the R-space spectrum represents the apparent distance, which is approximately 0.5 Å shorter than the actual distance due to phase shift.

Full size image

Density functional theory (DFT) calculations were used to investigate the Sr2+-ligand binding mechanism with Sr2+ ions. The calculated electron density of the Sr atom demonstrates that Sr coordinates with amide and alkoxy oxygen atoms from the Sr2+-ligand (Fig. 4a). An overlap between the Sr 3d orbital and the O 2p orbital around the Fermi level was calculated, confirming the coordination interaction between Sr2+ and the Sr2+-ligand (Fig. 4b and Supplementary Fig. 14). The atomic coordinates of the optimized computational models have been provided in Supplementary Data 1. Furthermore, interactions between the Sr2+-ligand and other alkali and alkaline earth metal ions, including Na+, Ca2+, Li+, and Ba2+, were investigated. As shown in Fig. 4c–f, the adsorption energies of the Sr2+-ligand with Na+, Ca2+, Li+, and Ba2+ are −0.691, −0.993, −1.256, and −1.891 eV, respectively, indicating weaker interactions than Sr2+(−2.036 eV). The results indicate that the Sr2+-ligand is a more appropriate candidate for Sr2+ ions, which could effectively stabilize the complex structure.

Fig. 4: Calculated density functional theory of the Sr2+-ligand with Sr2+ ions.
figure 4

a Electron density isosurfaces in the Sr2+-ligand containing Sr2+ ions. b Partial density-of-states of Sr 3d and O 2p of the Sr2+-ligand bound to Sr2+ ions. The dash line represents the Fermi level. c–f Electron density isosurfaces of the Sr2+-ligand with Li+, Ca2+, Na+, and Ba2+. The yellow and cyan isosurfaces represent charge accumulation and depletion, respectively. The calculation used the first-principles calculation based on the density functional theory calculations with generalized gradient approximation using the Perdew-Burke-Ernzerhof formulation in VASP software.

Full size image

Discussion

Overall, this study developed a optical Sr2+ sensing methodology based on the binding of Sr2+ to a specific Sr2+-ligand. This detection method achieves excellent detection performance with an ultralow detection limit of 0.5 nM, a high selectivity against 31 common ions, and a wide response range of 0.5 nM to 0.1 M, rendering it one of the best-performing Sr2+ detection methods. The nanoscale sensing matrix, with a diameter of 20 nm, exhibits a short response time of less than 10 s, making it ideal for on-site detection. The Sr2+-ligand molecule was proven to exhibit high affinity for strontium ions, and its binding mechanism to Sr2+ was investigated via EXAFS and XPS analyses. The Sr2+-ligand binds strontium by forming six coordination bonds between strontium ions and oxygen atoms in amide and alkoxy groups, forming a stable six-coordinated complex with a binding stability constant (log β 6.0) at least two orders higher than that of traditional Sr2+-sensing ligands. In a preliminary test, the Sr2+-nanosensor successfully detected Sr2+ in simulated nuclear wastewater and mineral water. The Sr2+-nanosensor holds great potential for in-situ detection devices such as strips and chips due to its noticeable color changes that are visible to the naked eye upon interaction with Sr2+. Overall, the proposed detection, which is distinguished by its ultra-high sensitivity, high selectivity, broad applicability, and high reliability, exhibits great promise for protecting individuals from the dangers of radioactive pollution.

Methods

Reagents

Poly(styrene)-graft-poly(ethylene oxide) (PS−PEO, product ID P15020A-SEOcomb, Mw/Mn = 1.5) was purchased from the Polymer Source, Ltd., and used as received. Chromoionophore I, Chromoionophore III, sodium tetrakis-[3,5-bis(trifluoromethyl)-phenyl] borate (NaTFPB), bis(2-ethylhexyl) sebacate (DOS), tetrahydrofuran (THF), and N, N, N′, N′, N″, N″-Hexacyclohexyl-4,4′,4″-propylidynetris(3-oxabutyramide) (Sr2+-ligand), were obtained from Sigma−Aldrich. SrCl2 and all other metal chlorides were purchased from Xi Long Co. Deionized water was prepared using a Millipore Milli-Q water purification system with a resistivity higher than 18.25 MΩ. All chemicals used in this study were analytical reagents and they were used directly without further purification.

Fabrication of the Sr2+-nanosensor

For the Sr2+-nanosensor, first, a cocktail was prepared by dissolving 1.30 mg of Sr2+-ligand, 0.66 mg of NaTFPB, 0.28 mg of chromoionophore I, 10.00 mg of PS−PEO, and 8.00 mg of DOS in 2.00 mL of THF to form a homogeneous solution. Then 400 μL of the cocktail was quickly injected into 3 mL of deionized water on a vertex spinning at 1000 rpm. The resulting mixture was blown by compressed air to remove THF, forming a concentrated nanosensor suspension. The resulting suspension of Sr2+-nanosensor was then mixed with Tris-HCl buffer for subsequent detection. Similarly, the fluorescence nanosensor was fabricated from a THF cocktail containing 2.01 mg of Sr2+-ligand, 0.67 mg of NaTFPB, 0.30 mg of chromoionophore III, 10.00 mg of PS−PEO, and 8.00 mg of DOS.

Fabrication of the sensing films for Sr2+

Typically, 1.20 mg of NaTFPB, 0.50 mg of chromoionophore I, 2.40 mg of Sr2+-ligand, 50.00 mg of PVC, and 100.00 mg of DOS were dissolved in 2.00 mL of THF. Next, 0.10 mL of a THF solution was drop-cast onto glass slides and left in the dark for several hours for the THF to evaporate.

Detection selectivity assay

The selectivity of Sr2+-nanosensor towards other cations was analyzed by assessing their ability to alter the absorbance value of Sr2+-nanosensor at 658 nm. The Sr2+-nanosensor, each containing 10−6 M of BaCl2, CoCl2, CrCl3, CaCl2, FeCl3, KCl, LiCl, MgCl2, MnCl2, NaCl, NiCl, CsCl, ZnCl2, NH4Cl, NaCNS, NaOH, Na2SO4, Na2CO3, NaHCO3, Na2HPO4, NaNO3, NaBr, NaAc, Na2SO3, Na2MnO4, NaHPO4, NaHSO3, NaNO2, and NaF underwent testing in buffered solutions (pH 7.0).

Detection of Sr2+ concentration after sensing process

The sensing films were immersed in 1 mL of solutions of varying Sr2+ concentrations. After the color change of films, the supernatant solutions were collected for ICP−MS detection.

Detection of Sr2+ ions in a natural aqueous environment

Drinking water (mineral water) was purchased from the Coconut Palm Group and Ginten Industrial Co., Ltd. and used without pre-treatment. Simulated nuclear wastewater was prepared according to a previous report6 (Supplementary Table 2). An aliquot (100 µL) of each water sample was added to the buffered nanosensor mixture (900 µL). The Sr2+ concentration standard curve was used to calculate the concentration of Sr2+ ions in each sample.

Instrumentation and measurements

Hydrodynamic size distribution was measured by the particle size analyzer Zetasizer Nano ZS (Malvern Inc.). TEM images were recorded on a transmission electron microscope (HT-7700, Hitachi) at an acceleration voltage of 100 kV. Absorbance spectra were recorded on ultraviolet-visible (UV-Vis) absorption spectrometer (Mapada). Fluorescence detection was carried out on an FS5 spectrofluorometer (Edinburgh Instruments). Disposable PMMA cuvettes were used as the sample holders. The XPS spectra were obtained on an Axis Supra spectrometer (Kratos). EXAFS analysis was performed at the e BL14W1 substation of the Shanghai Synchrotron Radiation Facility53,54. The obtained XAFS data was processed in Athena (version 0.9.26) for background, pre-edge line, and post-edge line calibrations. Then Fourier transformed fitting was carried out in Artemis (version 0.9.26). The k2 weighting, k-range of 2– ~8 Å−1 and R range of 1–2.5 Å were used for the fitting of Sample.

Calculations method

The first-principles55,56,57 were employed to perform all density functional theory (DFT) calculations within the generalized gradient approximation (GGA) using the Perdew-Burke-Ernzerhof (PBE) formulation58,59. The projected augmented wave (PAW) potentials60,61 was applied to describe the ionic cores and take valence electrons into account using a plane wave basis set with a kinetic energy cutoff of 450 eV. Partial occupancies of the Kohn−Sham orbitals were allowed using the Gaussian smearing method and a width of 0.05 eV. The electronic energy was considered self-consistent when the energy change was smaller than 10−5 eV. A geometry optimization was considered convergent when the energy change was smaller than 0.05 eV Å−1. The vacuum spacing in a direction perpendicular to the plane of the structure is 20 Å for the surfaces. The Brillouin zone integration is performed using 2 × 2 × 1 Monkhorst-Pack k-point sampling for a structure. Finally, the adsorption energies (Eads) were calculated as Eads = Ead/sub -Ead -Esub, where Ead/sub, Ead, and Esub are the total energies of the optimized adsorbate/substrate system, the adsorbate in the structure, and the clean substrate, respectively.