Highly Selective and Sensitive Ratiometric Detection of Sn2+ Ions Using NIR-Excited Rhodamine-B-Linked Upconversion Nanophosphors.

Detection of Sn2+ ions in environmental and biological samples is essential owing to the toxicological risk posed by excess use tin worldwide. Herein, we have designed a nanoprobe involving upconversion nanophosphors linked with a rhodamine-based fluorophore, which is selectively sensitive to the presence of Sn2+ ions. Upon excitation with near-infrared (NIR) light, the green emission of the nanophosphor is reabsorbed by the fluorophore with an efficiency that varies directly with the concentration of the Sn2+ ions. We have explored this NIR-excited fluorescence resonance energy transfer (FRET) process for the quantitative and ratiometric detection of Sn2+ ions in an aqueous phase. We have observed an excellent linear correlation between the ratiometric emission signal variation and the Sn2+ ion concentration in the lower micromolar range. The detection limit of Sn2+ ions observed using our FRET-based nanoprobe is about 10 times lower than that observed using other colorimetric or fluorescence-based techniques. Due to the minimal autofluorescence and great penetration depth of NIR light, this method is ideally suited for the selective and ultrasensitive detection of Sn2+ ions in complex biological or environmental samples.

Introduction

Tin is one of the most widely utilized heavy metals in industry. It is used in both organic and inorganic forms, with applications as diverse as antifouling paints, agrochemicals, poly(vinyl chloride) (PVC) stabilizers, catalysts, biocides, etc. Owing to such widespread use, over the last few decades, an excessive amount of inorganic and organic tin has been released into the environment, as a result of which it is found in large amounts in the air, water, and soil.[1−4] When these metal ions reach the human body through the food chain, they have a major detrimental influence on health and induce a variety of ailments.[5−8] Tin doses of more than 130 mg/kg have been seen to accumulate in the kidneys, bones, and spleen.[9−11] Tin(II) ions can induce diarrhea and DNA damage in the respiratory, reproductive, nervous, and digestive systems. Nausea, vomiting, and upper respiratory tract discomfort are all symptoms of moderate tin poisoning.[12−16] Acute tin poisoning can result in permanent renal tubule damage and various neurological alterations leading to disorientation, confusion, and memory loss, as well as severe epileptic seizures.[17−19] Therefore, methods of monitoring tin ion concentrations in environmental and biological samples are essential.

A number of analytical methods have been used for the detection/screening of Sn2+ ions, which include direct ion-concentration measurements such as atomic absorption spectroscopy and microwave-induced atomic emission spectrometry, or indirect, probe-based techniques such as electrochemical, colorimetric, and fluorometric assays. The direct methods, although provide good selectivity and sensitivity of detection, are expensive and tedious, which makes them unsuitable for real-time and in situ analysis.[20−23] Electrochemical methods, which rely on differences in the redox potential of ions, have poor selectivity and reproducibility. Certain colorimetry and fluorometry-based analytical chemosensors, using organic dyes and nanoparticles, have been developed for the selective detection of tin ions.[24−26] These sensors have good selectivity but show poor sensitivity and reproducibility, thus providing limited detection efficiency. Also, these sensors are usually excited by high-energy light within the UV–vis region, making them less likely to be utilized in the biological system due to the short penetration depth of UV–vis light. Autofluorescence and a poor signal-to-noise ratio are two apparent issues that degrade sensing performance and photo-oxidize sensing probes.[27−29] As a result, the development of advanced sensing probes activated by NIR light is urgently required.

The sensitivity and depth profiling of these optical sensors can be improved with the use of near-infrared (NIR) light as the primary excitation source. The NIR light is better suited as an excitation source because it penetrates deeper into the tissue and causes less harm to biological material than UV light.[30−35] Recently, lanthanide-doped upconverting nanophosphors (UCNPs) have attracted significant interest in biological, analytical, and optoelectronic applications because of their ability to convert NIR excitation light into shorter wavelength light. In addition, they have several other appealing optical and chemical features, including high Stokes shift, low toxicity, weak autofluorescence backgrounds, and resistance to photobleaching.[36−39] UCNPs constitute a dilute guest–host system, in which lanthanide ions are doped as a guest in an appropriate dielectric host lattice.[40−43] UCNPs absorb two or more photons sequentially, resulting in the emission of a single high-energy (low-wavelength) photon.[44−47] Several studies have shown that these characteristics make UCNPs an excellent choice for use as an energy donor in nanoprobes based on fluorescence resonance energy transfer (FRET).[48−50] So far, numerous sensors have been designed based on FRET to detect DNA, metal ions, and small molecules, in which UCNPs transmit energy to other chromophores, resulting in measurable changes in emission intensity/pattern.

Herein, we have designed an UCNP-based chemosensor involving an Sn2+-sensitive fluorescent probe (rhodamine-B derivative, or RBD) covalently linked to poly(acrylic acid) (PAA)-coated UCNPs (UCNP@PAA) and investigated its application in the ultrasensitive detection of Sn2+ ions. Both UV–vis absorbance and fluorescence emission properties of RBD are highly sensitive toward Sn2+ concentration in the solution. The UCNPs act as an energy donor that converts incident NIR light (of wavelength 980 nm) into visible light for exciting RBD (energy acceptor). We have probed in detail the FRET process between the UCNPs (characterized by the attenuation in their green emission peaks) and attached RBD molecules (characterized by concurrent enhancement in their red emission peak) as a function of the concentration of Sn2+ ions in the solution. Such FRET process allows ratiometric detection of the analyte, which enhances the reproducibility and sensitivity of analysis when compared to that obtained using traditional, single-mode analytical methods. Therefore, this work combines the advantages of the high selectivity of tin-ion detection using an organic probe (RBD) with the high sensitivity and reproducibility offered by FRET-based ratiometric detection involving UCNPs as an energy donor and RBD as an energy acceptor and the high tissue penetration of incident NIR light for potentially background-free detection in biological specimens. Based on the observed data, we have carried out a ratiometric analysis of the intensities of these emission bands and plotted the ratios against the concentration of Sn2+ ions. Based on the data obtained, we compared the detection limit of Sn2+ ion sensing obtained with our technique with those of some other published reports.

Experimental Section

Materials

All of the chemicals were utilized without additional purification. Y(NO3)3·6H2O (99.99%), Yb(NO3)3·6H2O (99.99%), Er(NO3)3·6H2O (99.99%), diethylene glycol, poly(acrylic acid) (PAA), and ethylenediamine (reagent, 99%) were purchased from Sigma-Aldrich. Oleic acid (90%, technical grade), 1-octadecene (90%, technical grade), and hydrochloric acid (HCl, analytical reagent 35–38%) were purchased from Alfa Aesar. Rhodamine-B (RhB, 95%) and hydrazine hydrate 80 wt% solution in H2O were purchased from Loba Chemie Pvt. Ltd. Ammonium fluoride (NH4F, 99%), sodium hydroxide (NaOH, 97%), cyclohexane (C6H12 99%), ethanol (C2H5OH, analytical reagent 99%), and methanol (CH4OH, analytical reagent 99%) were purchased from Spectrochem Pvt. Ltd. Double distilled water was utilized to prepare all of the aqueous solutions. Aqueous solutions of Sn2+, Hg2+, Co2+, Mg2+, Ni2+, Ca2+, Zn2+, Co2+, Cd2+, Na+, K+, and Cs+ were made from the corresponding halide salts.

Synthesis of Upconverting Nanophosphors (NaYF4:Yb3+/Er3+)

Uniform UCNPs (NaYF4:Yb3+/Er3+) capped with oleic acid (OA) were prepared via solvothermal synthesis.[51] In a typical synthesis, 0.795 mmol of Y(NO3)3·6H2O, 0.20 mmol of Yb(NO3)3·6H2O, and 0.005 mmol of Er(NO3)3·6H2O were taken in a 100 mL round-bottom flask. Then, 15 mL of octadecene and 5 mL of OA were added. A homogeneous solution was formed upon heating to 160 °C for 30 min, after which the solution was cooled to room temperature. Then, 10 mL of the methanol solution containing 0.148 g of NH4F and 0.1 g of NaOH was added to the flask and heated to 60–80 °C to remove methanol from the reaction. The solution was then degassed at 100 °C for 10 min and heated under N2 protection to 300 °C. The solution was naturally cooled to room temperature after being held at 300 °C for 1.5 h. The product NaYF4:Yb3+/Er3+ nanophosphors were precipitated with ethanol from the solution and washed and centrifuged using ethanol and cyclohexane to collect the nanoparticles.

Synthesis of PAA-Coated Upconverting Nanophosphors

The ligand-exchange approach was used to coat the surface of nanophosphors with poly(acrylic acid) (PAA).[52] In a three-necked flask, 0.5 g of PAA was added to 10 mL of diethylene glycol (DEG), and the mixture was vigorously stirred at 110 °C to obtain a clear solution. One hundred mg of UCNPs was dispersed in 5 mL of cyclohexane and slowly added to the reaction mixture. Then, the above reaction mixture was refluxed at 150 °C for 150 min in the presence of nitrogen. After refluxing, the resulting mixture was heated to 240 °C for 30 min to remove the cyclohexane. After cooling the solution to room temperature, ethanol was added to form a precipitate. The PAA-coated nanophosphors were collected by centrifugation and washed three times with ethanol/water (1:1 v/v).

Synthesis of Rhodamine-B Derivative (RBD)

The derivative of rhodamine-B was synthesized by a simple reaction. First, 0.5 g of rhodamine-B was dissolved in 15 mL of ethanol. Then, ethylenediamine (2 mL, excess) was added dropwise to the solution with vigorous stirring, and the mixture was refluxed for 24 h at 90 °C. The resulting solution was pale orange in color. The mixture was then allowed to cool to room temperature before extracting the solvent under decreased pressure using a rotatory evaporator. Twenty-five and 15 mL of 0.1 M of HCl and 1 M of NaOH solutions, respectively, were prepared. The as-prepared HCl solution was added dropwise to the mixture until gas production was stopped to obtain a red solution. Then, the NaOH solution was progressively added and agitated until the pH of the solution reached around 10. To eliminate the remaining impurities, the precipitate was centrifuged and rinsed five times with distilled water. The resulting material was dried under a vacuum at 60 °C for 8 h to obtain a cream-colored solid with excellent yield.

Synthesis of UCNP@PAA-RBD

In 10 mL of double distilled water, 50 mg of PAA-coated UCNPs were taken and sonicated for 5 min. Another solution of 50 mg of RBD was prepared in 1 mL of water, followed by the addition of 100 μL of ethanol. Then, the RBD solution was added to the UCNP dispersion and stirred for 24 h at room temperature. In the reaction mixture, the free carboxylic groups of PAA-coated UCNPs reacted with ethanol to form the corresponding polyacrylate ester. Then, this ester reacted with the free amine groups of RBD to form an amide linkage between the UCNP and RBD. The RBD-modified UCNPs (UCNP@PAA-RBD) were collected after centrifugation and washed three times with ethanol.

Detection of Tin Ions

In deionized water, stock solutions of Sn2+ and other metal ions (0.1 mM) were prepared. A stock solution of RBD (3 mg/mL) and UCNP@PAA-RBD (3 mg/mL) nanophosphors were prepared in distilled water. The selectivity of metal ions was determined as follows. First, we took 100 μL of the RBD stock solution in 3 mL of water and added suitable volumes of metal ions, including tin ions, to it. Then, UV–vis and fluorescence spectra of the resulting solutions were recorded. For the detection, the same amount of RBD stock solution in 3 mL of water was taken. Then, different amounts of Sn2+ ions (in μM) were added using a micropipette, followed by the recording of UV–vis and fluorescence spectra. For the FRET-based detection, 100 μL of the stock solution of UCNP@PAA-RBD was added to 3 mL of water, and the upconverting photoluminescence spectra were recorded with different concentrations of Sn2+ ions (in ppm) under the NIR light (980 nm) excitation.

Characterization

The nanophosphors were subjected to morphological characterization by field emission scanning electron microscopy (FESEM) using scanning transmission electron microscopy (STEM), MERLIN Zeiss-Germany. For that, the suspension of nanophosphors was deposited on a copper TEM grid with a carbon film (TED Pella). To confirm the exact shape and size of the nanoparticles, transmission electron microscopy (TEM) was carried out using a TECNAI G2-30 U TWIN (FEI, Eindhoven, Netherlands) instrument operated with an accelerated voltage of 300 kV. The average hydrodynamic diameter of the nanophosphors was measured by dynamic light scattering (DLS), using a NANO-ZS series Malvern Zetasizer instrument. Powder X-ray diffraction measurement was done to analyze the phase composition and crystalline nature of nanophosphors using a Bruker D8 Discover X-ray spectrometer, which utilizes Cu Kα radiation (λ = 1.54060 Å) over the 2θ range at the rate of 2.58/min. FT-IR spectra were taken from the range of 4000 to 400 cm–1, where dried and powdered nanophosphors were mixed with KBr and the mixture was passed into a pellet for analysis using a PerkinElmer RX1 spectrometer. The absorbance and fluorescence spectra were observed using a Shimadzu UV-1601 spectrophotometer (Shimadzu, Kyoto, Japan) and a Cary Eclipse fluorescence spectrometer (Varian, Palo Alto, CA), respectively. An upconversion fluorescence spectrometer (Quanta Master, Model QM-8450-11), attached with an external 980 nm tunable diode laser, was used to acquire the upconversion luminescence emission spectra.

Results and Discussion

We first synthesized hexagonal (β-phase) UCNPs (NaYF4:Yb3+/Er3+) via the solvothermal method, with NaYF4 serving as the host matrix and dopant ions Yb3+ and Er3+ as sensitizer and activator, respectively. These UCNPs were hydrophobic and well dispersed in cyclohexane. PAA coating was done on the surface of UCNPs to make them hydrophilic using the ligand-exchange approach, which not only results in their optimal aqueous dispersion but also provides free functional groups for further conjugation reactions. The fabrication of the Sn2+-sensitive FRET-based probe is enabled by the covalent linkage between PAA-modified UCNPs and ethylenediamine-tagged derivative of rhodamine-B (RBD). This linkage during simple mixing results in the strong binding of the acceptor (RBD) to the donor (UCNP) surface. The RBD molecule, whether free or bound to the UCNPs, undergoes a structural transition from a closed-ring (nonfluorescent) to an open-ring (fluorescent) configuration as a result of the coordination with Sn2+ ions. As the NIR-excited green emission peak of the UCNP overlaps with the absorption spectrum of the open-ring (Sn2+-coordinated) form of RBD, it is evident that with the increasing Sn2+ concentration, FRET from UCNP (donor) to open-ring RBD (acceptor) is enhanced, leading to the simultaneous decrease and increase in the emission peaks of UCNP (green emission) and RBD, respectively. This facilitates the ratiometric and quantitative detection of Sn2+ ions via the FRET process under NIR excitation, as shown schematically in Figure .

Figure 1

Schematic illustration of the synthetic procedure of UCNP@PAA-RBD and the proposed FRET-based sensing mechanism of UCNP@PAA-RBD with Sn2+ ions.

The structural and morphological characterization of nanophosphors was done by utilizing FESEM and TEM. The low- and high-resolution FESEM images showed that UCNPs were synthesized with uniform sizes and had a hexagonal morphology (Figure A). From the low- and high-resolution TEM micrographs (Figure B), it is evident that UCNPs are hexagonal, with size in the range of 150–250 nm, thus validating the FESEM data. The average size and polydispersity index (PDI) of the UCNPs, as evident from DLS measurements, were measured to be 186.5 nm and 0.184, respectively (Figure S1). The X-ray diffraction (XRD) patterns of UCNP, UCNP@PAA, and UCNP@PAA-RBD are shown in Figure C. The XRD pattern of all UCNP samples was in good agreement with the standard database (JCPDS 16-0334),[53] validating the pure hexagonal phase of the UCNPs. No additional peaks were found, indicating the purity of the prepared product. The sharp peaks seen in the XRD pattern indicated the crystalline nature of the synthesized material. The sharpness of peaks of UCNP@PAA and UCNP@PAA-RBD was reduced because of the coating of PAA on the nanophosphors. The existence of Na, Y, F, Yb, and Er in the prepared samples was confirmed by the energy-dispersive X-ray (EDX) spectrum (Figure S2).

Figure 2

Low- and high-resolution (A) FESEM and (B) TEM images of UCNPs. (C) XRD patterns of the UCNP, UCNP@PAA, and UCNP@PAA-RBD, along with the standard (JCPDS No. 16-0334).

Figure shows the Fourier transform infrared (FT-IR) spectra of UCNP, UCNP@PAA, and UCNP@PAA-RBD recorded in the 500–4000 cm–1 range. The FT-IR spectrum of the synthesized UCNPs shows four notable peaks at 1462, 1612, 2854, and 2925 cm–1. The methylene (−CH2−) asymmetric and symmetric stretching vibrations of OA in the long alkyl chain were attributed to the bands at 2925 and 2854 cm–1. Moreover, asymmetric and symmetric stretching vibrations of the −COO– on the surface of the UCNPs resulted in two bands centered at 1612 and 1462 cm–1, respectively. After PAA coating on the surface of nanophosphors, the asymmetric stretching vibrations of the −CO and −CO2 groups of the PAA ligand at 1560 and 1734 cm–1, respectively, were observed in the spectrum. The symmetric stretching vibrations of CH2 caused another peak in the higher-wavenumber region, i.e., 2926 cm–1 (νCH2). In addition, the stretching vibration of the hydroxyl (−OH) group caused a wide singlet at 3443 cm–1. The results showed that PAA successfully converted the hydrophobic surface of nanoparticles to hydrophilic. Also, three new peaks were observed at 1688, 1220, and 1115 cm–1, which corresponded to the stretching frequencies of −NHCO, −C–N, and −CO groups. The peak of the −OH group in UCNP@PAA disappeared, indicating its utility in forming a covalent linkage with RBD. This was further confirmed by the absence of the −NH2 peak (of free RBD) in the UCNP@PAA-RBD conjugate.

Figure 3

FT-IR spectra of UCNP, UCNP@PAA, UCNP@PAA-RBD, and free RBD.

The synthesized UCNPs emitted characteristic emission bands at 407 nm (blue), 522 nm, 542 nm (both green), and 655 nm (red), under the NIR laser excitation of 980 nm (Figure A). These bands corresponded to Er3+ ion transitions 2H9/2 → 4I15/2, 2H11/2 → 4I15/2, 4S3/2 → 4I15/2, and 4F9/2 → 4I15/2, as already reported in several publications.[54−56] The structural and photoluminescence characteristics of RBD were also studied. The 1H NMR and 13C NMR spectra (Figure S3) of RBD showed that the compound was properly synthesized from rhodamine-B. The excitation and emission spectra of RBD in the presence of Sn2+ ions are shown in Figure B. It is evident from the graph that pure RBD (without added Sn2+) displayed no apparent absorption/excitation. However, a broader, distinctive excitation band ranging from 495 to 590 nm (λEm = 579 nm), with a peak at 554 nm, was seen after the addition of Sn2+ ions, which overlapped well with the two green upconverting emission bands of UCNPs (at 522 and 542 nm). This data clearly shows the overlap between the green emission bands of the UCNP and the absorption/excitation band of the open-ring RBD, indicating the feasibility of energy transfer from UCNP to the open-ring RBD. Finally, the emission band of the open-ring RBD is observed with a maximum at around 579 nm (λEx = 525 nm).

Figure 4

(A) Photoluminescence spectrum of UCNPs under NIR (980 nm) laser excitation. (B) Excitation and emission spectra of RBD, with and without added Sn2+ ions.

Figure A shows the UV–vis absorption spectra of free RBD treated with various metal ions, including Sn2+ ion. A noticeable enhancement in the absorption spectra of RBD (centered at 554 nm) is found only upon treatment with Sn2+ ion. This observation is visibly evidenced by the change in the probe’s color from colorless to magenta upon treatment with Sn2+ ions only (Figure B). This observation confirms the RBD’s preferential affinity for Sn2+ ions over other metal ions. The Sn2+ ions can aid in the hydrolysis of RBD’s α-amino acid ester ring, resulting in the formation of Sn-α-amino acid chelate due to the ring opening of RBD, as schematically illustrated earlier in Figure .

Figure 5

(A) UV–vis absorption spectra of RBD upon the addition of various metal ions (Ca2+, Mg2+, Fe2+, Fe3+, Co2+, Ni2+, Cu2+, Zn2+, Cd2+, Pb2+, Hg2+, Sn2+, Cs+, Na+, K+) in distilled water. (B) Visible appearance of the solutions upon the addition of various metal ions, including Sn2+.

The UV–vis titration spectra of the RBD solution (100 μL of stock in 3 mL of water) after the addition of increasing amounts of Sn2+ ions (in the range of 0–50 μM) are shown in Figure A. The intensity of the absorption band centered at 554 nm grew noticeably with the linear increase in Sn2+ concentration. A calibration curve of the relative absorbance intensity of RBD as a function of added Sn2+ ion concentration is depicted in Figure B. It can be seen that the relative intensity increased linearly with the concentration, and showed an R2 value of 0.9979. The detection limit of RBD toward Sn2+, which was calculated as thrice the deviation of the blank signal (3σ),[57] was found to be 0.15 μM.

Figure 6

(A) Absorption changes of RBD with increasing amounts of Sn2+ ions. Inset: absorbance of RBD at 554 nm as a function of Sn2+ concentration. (B) Calibration plot for the relative absorption intensity at 554 nm of RBD as a function of Sn2+ concentration.

The fluorescence spectrum of RBD toward various metal ions is shown in Figure S4A. Similar to the UV–vis spectral data, significant enhancement in the fluorescence intensity (at 579 nm) was observed after treatment of RBD with Sn2+ ions only, thus validating the preferential affinity of the dye for Sn2+ ions over other metal ions. In the fluorescence titration spectra (Figure A) of the RBD solution (100 μL of stock in 3 mL of water), a considerable increase in fluorescence intensity at 579 nm was recorded upon treatment with different concentrations of Sn2+ ions. The fluorescence intensities (at 579 nm) of RBD as a function of Sn2+ ion concentrations are shown in Figure A (inset). This intensity was found to be linearly dependent on the concentration of Sn2+ ions. The calibration curve with different concentrations of Sn2+ displayed a colinear relationship (F/F0 = 0.0212CSn – 0.0152, R2 = 0.9935), as shown in Figure B. The detection limit (3σ) by utilizing fluorescence techniques was calculated to be 0.29 μM, indicating that we could detect ecologically relevant quantities of Sn2+ utilizing this approach. The time-dependent response of RBD toward Sn2+ ions was also examined by fluorescence spectroscopy. The response of RBD (100 μL of stock in 3 mL of water) to Sn2+ was quick, as shown in Figure S4B, and the peak signal was reached in approximately 40 s. This analysis demonstrated that the nanoprobe responded very fast to Sn2+ ions and might be conveniently used to monitor and evaluate Sn2+ levels.

Figure 7

(A) Change in the fluorescence intensity of RBD with the increasing concentration of Sn2+ ions (λEx = 525 nm). Inset: fluorescence intensity at 579 nm as a function of Sn2+ ion concentration. (B) Relative fluorescence intensity at 579 nm of RBD as a function of Sn2+ ion concentrations.

Furthermore, we repeated the RBD-mediated Sn2+ detection experiment using UV–vis absorption and fluorescence measurements in a buffer with pH = 7. The data for absorption and fluorescence measurements are provided in Figures S5A,B and S6A,B, respectively. It is evident that the data obtained by carrying out the optical measurements in water are very similar to that obtained in the buffer solution of pH 7.

We next investigated whether the detection sensitivity of Sn2+ ions can still be improved with the help of UCNPs that allows FRET-based ratiometric analysis under excitation of deep-tissue-penetrating NIR light. First, the upconverting emission spectra of UCNP@PAA-RBD were analyzed using various concentrations of Sn2+ ions under NIR laser (980 nm) excitation. As mentioned previously, the green emission peak of UCNP and the absorption peak of the RBD–Sn2+ combination overlaps, leading to fluorescence resonance energy transfer (FRET) from the UCNPs to the RBD–Sn2+ complex under NIR excitation. As illustrated in Figure A, UCNP@PAA-RBD displays both green emission peaks and red emission peaks in the absence of Sn2+ ions. Next, when the Sn2+ ions are added to UCNP@PAA-RBD in increasing concentrations, the intensity of green emission peaks (at 522 and 542 nm) steadily decreases, while an additional, broader emission appears with a peak at 579 nm, which can be attributed to the attached RBD–Sn2+ complex. The energy transfer from the UCNPs to the RBD–Sn2+ complex is confirmed by a continuous drop in peaks of green emission and the appearance of a wide band at around 579 nm with the increasing concentration of Sn2+ ions, whereas the red emission of UCNPs at 655 nm remains almost consistent. The red emission intensity (at 655 nm) of UCNPs can be utilized as an internal reference since RBD does not absorb at this wavelength.

Figure 8

(A) Emission spectra of UCNP@PAA-RBD in the presence of the increasing concentration of Sn2+ ions upon 980 nm excitation. Inset: emission at 579 nm as a result of FRET upon the addition of Sn2+ ions. (B) Variation in normalized GRE and RRE at different concentrations of Sn2+ ions.

At different concentrations of Sn2+ ions, the green-to-red emission (GRE) ratios (I542/I655), as well as the RBD-to-red emission (RRE) ratio (I579/I655), are shown in Figure S7. Figure B shows the variation in both normalized GRE and RRE ratios at various concentrations of Sn2+ ions. The UCNP emission/RBD absorption overlap and the consequent FRET are supported by the changing trends of both GRE and RRE as a function of added Sn2+ concentration, with UCNP as the energy donor and Sn2+ ion-coordinated RBD as the energy acceptor. High R2 linear fits are obtained for both the GRE (0.9925) and RRE (99 809) forms, demonstrating very precise and reliable Sn2+ sensing. In the presence of different Sn2+ concentrations (0–50 μM), the ratio of I542/I655 and I579/I655 varies from 4.9 to 1.0 and 0.01 to 0.24, respectively, exhibiting strong linear correlation (I542/I655 = −0.0776x + 4.7569, R2 = 0.9925 and I579/I655 = 0.0046x + 0.0175, R2 = 0.9980, where x is the concentration of Sn2+ ions in μM).

The slope of the calibration curves determines the sensitivity of the Sn2+ measurement. The GRE and RRE are found to have sensitivities of 0.0776 and 0.0046 per unit change in concentration (in μM), respectively. This ratiometric analysis can be utilized to calculate the Sn2+ ion concentration because free RBD coordinated with Sn2+ ion does not show any fluorescence under the 980 nm excitation. The RBD emission intensity response to Sn2+ ion concentration has a strong linear coefficient (R2) up to 0.99809, implying excellent quantification capabilities. The detection limit (3σ) by utilizing this FRET-based upconverting nanophosphor photoluminescence spectroscopy is found to be as low as 10 nM (0.01092 μM), which is more than 10-fold better than that observed using UV–vis and fluorescence techniques for free RBD treated with Sn2+ ions. The determined detection limit is far lower than the World Health Organization (WHO) approved Sn2+ contamination levels of 2.1 × 10–6 and 8.4 × 10–4 M for drinking water and other products, respectively.[58] When compared to some other reported probes, the UCNP@PAA-RBD nanoprobe shows a better detection limit for Sn2+ ions, as shown in Table .

Table 1

Comparison of Detection Limits for Sn2+ Ion Using UCNP@PAA-RBD with Other Reported Probes

S. No.	probe name	media	method	detection limit (μM)	reference
1	SEPTD	DMF	absorption/emission	0.17	(59)
2	2CND	water	emission	0.41	(60)
3	copolyimides	methanol	quenching	1.9	(61)
4	carbazole-containing diarylethene	methanol	emission	1.9	(62)
5	rhodamine-B with tert-butyl carbazate group (R2)	ethanol/water	emission	0.46	(63)
6	Rh-ED	ethanol/water	emission	0.16	(64)
7	diamine Schiff base ligand (L)	DMSO/water	absorption/emission	0.31	(24)
8	CK	PBD/ethanol	emission	0.11	(65)
9	GO-CeM	HAc-NaAc/TMB	colorimetry	5.58	(66)
10	UCNP@PAA-RBD	water	FRET	0.01	this work

The ratio of the emission intensity at 579 nm (I579) to the intensity at 542 nm (I542) and vice versa, against the Sn2+ concentration range from 0 to 50 μM, are shown in Figure S8A,C, respectively, which shows an exponential variation in the ratios. However, the plots of logarithmic emission intensity ratios for the same against the concentration of Sn2+ ions were found to have a good linear relationship (Figure S8B,D). In the range of 0–50 μM, R2 is 0.9936, which was the linear correlation coefficient. Figure A shows the FRET process using the nanoprobe using an even more subtle variation in the Sn2+ ion concentration (1–5 μM). As we observe from the graph, the FRET resulting in the variation of such low analyte concentration is visually noticeable from the spectra; even 1 μM of Sn2+ ions alter the green emission peak because of the efficient FRET process in the nanoprobe. The normalized GRE and RRE ratio at different concentrations of Sn2+ (0–5 μM) is shown in Figure B. Here, we also obtain high R2 linear fits for both GRE and RRE, which is similar to the previous data, again confirming the high sensitivity toward Sn2+ ions as shown in Figure S9. The ratios of I542/I655 and I579/I655 for the concentration ranges from 0 to 5 μM exhibit a strong linear correlation (I542/I655 = −0.1514x + 5.0246, R2 = 0.9935 and I579/I655 = 0.0156x + 0.0121, R2 = 0.9962, where x is the concentration of Sn2+ ions in μM). The detection limit (3σ) from this low-concentration data was observed to be 10 nM (0.010248 μM), which correlates with the data taken for the concentration range of 0–50 μM for Sn2+ ions. Further, the actual and logarithmic ratios of emission intensity at 579 nm (I579) to intensity at 542 nm (I542) and vice versa, against Sn2+ concentration ranges from 0 to 5 μM, are shown in Figure S10, which shows an exponential correlation of the actual ratios and a linear correlation for the logarithmic ratios with analyte concentration. The pH-dependent response of UCNP@PAA-RBD toward Sn2+ detection has also been examined. The upconverting emission spectra of UCNP@PAA-RBD, with and without Sn2+ ions, in the pH range of 4–9 are recorded and the GRE and RRE responses are analyzed. As illustrated in Figure S11A,B, the variation in GRE and RRE before the addition of Sn2+ ions is negligible, but in the acidic and alkaline pH, the green emission intensity of UCNP is still less compared to that in neutral pH. After the addition of Sn2+ ions (10 μM), the efficient energy transfer has been observed from the graph (the GRE values decrease and RRE values increase in all pH values), indicating that pH has little impact on the FRET process and detection analysis. These results further confirm the successful transfer of energy between the UCNP and RBD even at a low analyte concentration and in a wider pH range. Therefore, these nanophosphor-based probes can be used for the ultrasensitive detection of Sn2+ ions in various samples, following NIR excitation.

Figure 9

(A) Upconversion emission spectra of UCNP@PAA-RBD upon addition of Sn2+ ions below 5 μM. (B) Variation in normalized GRE and RRE at the Sn2+ ion concentration of 0–5 μM.

Conclusions

In summary, we have developed polymer-modified and RBD-functionalized upconverting nanophosphors for the FRET-based ultrasensitive detection of Sn2+ ions. Here, UCNPs acted as the energy donor and convert the NIR (980 nm) into visible light, which was absorbed by the RBD molecules only in the presence of Sn2+ ions. This nanoprobe exhibited a fast response to Sn2+ ions, a low autofluorescence background, and strong selectivity. With the addition of Sn2+ ions, the nanoprobe generated a clear color shift from colorless to magenta, which could be seen with the naked eye, accompanied by an increase in fluorescence emission. More importantly, when excited by NIR (980 nm) light, the intensity of green upconversion emissions reduced progressively, while a new emission peak at 579 nm developed, which grew with increasing Sn2+ ion concentration, corresponding to the efficient FRET from UCNP to the RBD–Sn2+ complex. The ratiometric approach provided an efficient and reliable response for the detection of Sn2+ with a detection limit of 0.01 μM, which is a tenth of that of pure RBD and is lower than the WHO-approved contamination levels of Sn2+ ions. This demonstrates that the nanophosphors have high sensitivity and selectivity toward Sn2+ ions and can be used to detect under the excitation of NIR light.