Smart excimer fluorescence probe for visual detection, cell imaging and extraction of Hg²⁺

Abstract

Smart pyrene-based simple fluorescent probes 2 and 4 were designed, synthesized and characterized by different spectroscopic methods. The photophysical properties of the probes and their affinity towards different metal ions in phosphate buffer were investigated. Upon selective interaction with Hg²⁺, the molecular probes showed enhanced static excimer emission at 506 nm along with a naked-eye detectable chromo- and fluoro-genic response. Probe 2 sensitively showed a high limit of detection (3.4 pM) for Hg²⁺ in the solution. The pyrene silicate derivative, 4, was utilized to detect and extract Hg²⁺ in solution as well as in the solid state. The data obtained from NMR and ESI-MS spectroscopy supported the postulate that the mode of interaction of the probe with Hg²⁺ involves the N and O atoms of the –C=N and –OH functional groups to complex Hg²⁺ in 2 : 1 stoichiometry. Moreover, probe 2 exhibited excellent selectivity for Hg²⁺ in protein medium (BSA/HSA) and was used to detect Hg²⁺ in live HeLa cells, on test paper strips and in real contaminated water samples.

---

Introduction

Contemporary scientific research has placed much interest in developing smart organic scaffolds/fluorescent probes to detect biologically and ecologically important metal ions and anions through different photophysical mechanisms.¹ Among the various detection methods, fluorescent chemosensors exhibiting ion-induced changes in their fluorescence behavior are attractive and advantageous in terms of sensitivity, selectivity, and response time.² Moreover, the development of sensitive and reliable molecular scaffolds and/or fluorescence “turn-on” sensors for heavy and transition metal ions (HTMs) is challenging due to the fluorescence quenching effect.³ Furthermore, studies related to typical photophysical mechanisms of newly designed chemosensors, such as photoinduced electron transfer (PET), internal charge transfer (ICT), chelation enhanced fluorescence (CHEF), aggregation-induced emission (AIE), excimer, and C=N isomerization,^4–8 are a continuing endeavor.

Recently, the scientific community has found that contamination due to heavy and transition metals (HTMs) has resulted in serious environmental and health problems. In particular, mercury (Hg²⁺) is recognized as a detrimental neurological toxin, which is widely distributed in the environment by various natural processes, industrial releases and anthropogenic activities. The bioaccumulation of such toxic materials in living tissues of humans and animals via the food chain causes mercury poisoning, serious neural disorders and diseases like Minamata.^9,10 Environmental Protection Agency (EPA, USA) has set the maximum allowable level of inorganic mercury in drinking water at 2 ppb.¹¹

In this direction, great efforts have been made to design efficient fluorescent molecular sensors^12–16 to detect Hg²⁺ either by fluorescence enhancement or quenching and color changes.¹⁷ However, the numbers of good chemosensors for detecting Hg²⁺ through enhanced “turn-on” emission are limited in number because Hg²⁺ is known to induce fluorescence quenching due to the spin–orbit coupling effect/electron transfer mechanism and is spectroscopically and magnetically silent (closed-shell d¹⁰ configuration).^18–20 Moreover, the poor aqueous medium compatibility of receptors¹⁹ also limits the application of such systems as good analytical tools and/or sensitive sensors for Hg²⁺.^21–27 Thus, the selective recognition and separation of traces of mercury in the environment and in real samples has great significance and is worth investigating. Therefore, the development of cost-effective, sensitive and facile new synthetic small organic molecular scaffolds/chemosensors that are compatible with environmental and biological milieus, as well as functional in an aqueous or partially aqueous medium, in the solid state and/or at the liquid–solid interface are important goals to provide an instant optical feedback and online detection without involving sophisticated instrumentation.

Keeping these perspectives in mind, our ongoing research is currently focused on the development of some efficient fluorescent organic scaffolds/motifs to sensitively detect cations, anions, and biomolecules in different media.⁸ Through this contribution, we present the design, synthesis and potential application of two very simple pyrene-based smart fluorescent probes, 2 and 4, for the detection and extraction of Hg²⁺ in a partial aqueous medium and on solid support surfaces. We assumed that the donor sites of the probes, available in the form of N and O atoms, would show high affinity for the soft closed-shell cation Hg²⁺ and form stable complexes in the medium.²⁸ Moreover, the introduction of an ethanolamine moiety would help to generate a stable minimum energy configuration to promote the excimer mechanism.^10,11 To accomplish the objective of solid state chemosensing as well as mercury extraction from solution, we utilized a methodology employing a conventional silica gel polymer material. The high fluorescence “turn-on” chemodetection sensitivity displayed by the solid chemosensory material 4 is interesting and provided an additional way to extract and detect Hg²⁺ in different media. Moreover, the sensitivity of the probe to selectively detect Hg²⁺ in a real water sample, on cellulose paper strips, in protein medium (BSA and human HSA) as well as in live HeLa cells through an excellent fluorescence turn-on response suggested their promising prospect in environmental and biological science.

Experimental

Synthesis of probes and general experimental procedures

Data related to synthesis and characterization of probes and general experiment methodology are provided in the ESI.†

Results and discussion

Probe design, photophysical behavior and metal ion selectivity

Recently, some modified fluoroionophores immobilized on the surface of alumina or silica and nano-materials have been explored to detect HTMs.⁸ Interestingly, the use of silica gel offers an amorphous inorganic polymer surface with an abundance of siloxane (Si–O–Si) and silanol (Si–OH) functionalities in the bulk and on the surface, respectively. The silanol functional groups present on the surface may be easily explored to introduce chemical modifications, tether groups and incorporate modified fluoroionophores to reversibly trap and/or extract metal ions. Consequently, the variations in the specific optical behavior in a particular sensing event are easy to follow by different spectroscopic methods. The signaling fluorophore unit such as pyrene is known for its good quantum yield and chemical stability.⁹ It shows characteristic monomer emission at 370–410 nm and typical excimer emission between 470 and 520 nm due to the formation of an excited state dimer at a low electronic energy state.¹⁰ The excimer emission of pyrene is tunable and can be easily utilized to sensitively detect guest species¹¹ like a molecular OFF–ON switch.²⁰

Scheme 1 describes the synthetic route for molecular probes 2 and 4. Pyrene carboxaldehyde and ethanolamine were refluxed in ethanol to obtain a solid yellow colored compound 2 in good yield (78%). Compound 2 in the presence of diisopropylethylamine was reacted with 3-chloropropyltriethoxysilane in dry dichloromethane to obtain compound 3 in 72% yield. Compound 3 and silica gel (particle size: 75–150 μm; average pore diameter: 25–30 Å) were stirred overnight at room temperature in dry dichloromethane to obtain probe 4 as an orange colored solid in 68% yield. The isolated compounds were characterized by different spectroscopic techniques and the data are included as ESI (Fig. S1–S8†).

Scheme 1 Preparation of probes 2 and 4: (i) ethanolamine/ethanol/reflux, (ii) (3-chloropropyl)triethoxysilane/DIPEA/DCM, and (iii) silica gel/dry DCM.

The photophysical behavior of probe 2 was examined through absorption and emission spectroscopy at room temperature in phosphate buffer (10 mM pH 7.0; 10% aqueous ACN). The absorption spectrum of 2 (10 μM) displayed low and high energy absorption bands at 357 nm (ε = 7.26 × 10⁴ M⁻¹ cm⁻¹) and 284 nm (ε = 6.69 × 10⁴ M⁻¹ cm⁻¹) respectively, along with shoulders at 386, 342 and 273 nm. Upon excitation at 345 nm, 2 (10 μM) displayed typical pyrene monomer emissions at 375, 393 and 414 nm (Φ₂ = 0.004) (Fig. 1). The affinity of probe 2 toward different metal ions (2.0 equiv.), such as Na⁺, K⁺, Ca²⁺, Mg²⁺, Co²⁺, Ni²⁺, Cu²⁺, Zn²⁺, Al³⁺, Fe²⁺, Fe³⁺, Pb²⁺, Ag⁺, Cd²⁺, and Hg²⁺, (as their nitrate salts) was examined in phosphate buffer. Notably, the absorption maxima of 2 upon interaction with Hg²⁺ and Cu²⁺ ions showed a considerable bathochromic shift (∼78 nm) and a new electronic transition band appeared at ∼435 nm (2 + Hg²⁺; ε = 4.07 × 10⁴; 2 + Cu²⁺; ε = 1.19 × 10⁴ M⁻¹ cm⁻¹) (Fig. 1, inset), and a naked eye detectable green color appeared in the medium (Fig. 2). Similarly, upon interaction with Hg²⁺ and Cu²⁺, probe 2 exhibited an enhanced excimer emission with “turn-on” at ∼506 nm (red shift, ∼113 nm) in which the relative emission intensity was enhanced by ∼10 fold (Φ_2+Hg²⁺ = 0.039) with a ∼1.5 fold (Φ_2+Cu²⁺ = 0.086) increase in the quantum yield (Fig. 1). The color of the solution changed from a fluorescent blue to a naked-eye detectable bright fluorescent cyan color under UV light (at 365 nm) (Fig. 2). The other tested metal ions caused an insignificant change in the absorption and emission spectra of the probe.

Fig. 1 Change in (a) absorption and (b) emission spectra of 2 (10 μM) upon interaction with different metal ions (2.0 equiv.) in phosphate buffer (10 mM pH 7.0; 10% aqueous ACN, v/v).

Fig. 2 Colorimetric and fluorogenic response of 2 (10 μM) upon interaction with different metal ions (2.0 equiv.) in phosphate buffer (10 mM pH 7.0; 10% aqueous ACN v/v).

Furthermore, interference studies were performed with competitive metal ions to understand the high affinity of 2 for Hg²⁺. Interestingly, upon a sequential addition of the tested metal ions (in excess, 20 equiv.) to a solution of the probable complex of 2 + Hg²⁺ or the reverse, i.e., addition of Hg²⁺ to a solution of 2 + Mⁿ⁺ ions (Fig. S9†), insignificant changes were observed in the relative emission intensity and the color of the complex 2 + Hg²⁺ (Fig. S10†). Moreover, the reversible mode of complexation could be realized by the addition of a strong chelating reagent (EDTA) to a solution of the probable complex, 2 − Hg²⁺. Notably, the observed revived emission was very close to that of the monomer emission of 2 due to the strong binding affinity of EDTA for Hg²⁺.¹⁰ Reversibly, upon the addition of Hg²⁺ to a solution of probe 2 containing EDTA, no excimer emission was observed. This process was repeated successfully ten times and in every cycle probe 2 behaved almost consistently (Fig. S11†). Thus, it is noteworthy to mention that upon complexation of 2 with Hg²⁺ and/or Cu²⁺ through the coordination sites available in the form of O and N atoms of the hydroxyl and aldimine functional groups and the formation of a stable dimeric species in the medium, two pyrene units acquired a stacked conformation due to enhanced π → π electrostatic interaction.²⁹

Moreover, it is well known that if a fluorescent probe and/or its complex, upon excitation at their emission maxima do not show a complete spectral overlap, the observed enhanced fluorescence may be assigned to a static excimer emission.¹⁰ Therefore, to ensure that the observed enhanced emission upon the complexation of probe 2 with Hg²⁺ is due to either static or dynamic excimer formation, the fluorescence excitation spectra were acquired (Fig. S12†). Probe 2 and its probable complex 2 − Hg²⁺ were excited with emission corresponding to the observed maxima of their monomer and excimer centered at 397 and 506 nm, respectively. Interestingly, the probable complex 2 − Hg²⁺ displayed new emission bands at ∼445 and 558 nm, while no such emission was observed when probe 2 was excited at 397 and 506 nm, respectively. Thus, the observations clearly supported the presence of a static excimer emission, which is well induced by the Hg²⁺ in the present sensing event.

Absorption and emission titration studies were performed to understand the binding affinity of 2 (10 μM) with Hg²⁺. Upon gradual addition of Hg²⁺ (0–2 equiv.) to a solution of 2, the absorption bands centered at 357, 284 and 248 nm reduced progressively and new bands appeared at 435 (ε = 4.18 × 10⁴ M⁻¹ cm⁻¹), 301 (ε = 3.5 × 10⁴ M⁻¹ cm⁻¹), 258 (ε = 3.82 × 10⁴ M⁻¹ cm⁻¹) and 232 (ε = 1.01 × 10⁴ M⁻¹ cm⁻¹) nm (Fig. 3). The formation of multiple isosbestic points at 372, 316, 287, 267, 254 and 237 nm suggested the probable existence of more than one species in the medium, i.e., a dimeric complex, e.g. complex 2 + Hg²⁺, and also a geometrical change in the structure of the probe from E to Z isomer. Similarly, the acquired emission titration spectra of 2 displayed a gradual decrease in the monomer emission (at 393 nm), while the intensity of a new excimer emission was enhanced (∼5 fold) continuously at 506 nm (red shift of ∼113 nm) (Fig. 3). The estimated quantum yield was found to increase by about ∼10 fold (Φ_2+Hg²⁺ = 0.039). Job's plot analysis consistently revealed a 2 : 1 binding stoichiometry for the interaction between 2 and Hg²⁺ ions (Fig. S13†), for which the association constants were estimated by the B–H method³⁰ and were found to be K_abs = 7.29 × 10³ and K_em = 2.28 × 10⁴/M^1/2 (Fig. 3, insets).

Fig. 3 Change in (a) absorption and (b) emission titration spectra of 2 (10 μM) upon addition of Hg²⁺ (0–2 equiv.) in phosphate buffer (10 mM pH 7.0; 10% aqueous ACN, v/v). Benesi–Hildebrand plots based on (c) absorption and (d) emission titration data.

The pH dependence of probe 2

The pH dependent photophysical behaviors of probe 2 and its probable complex 2 + Hg²⁺ were examined in phosphate buffer (Fig. 4 and S14†). The optical behavior of probe 2 was found to be insensitive in the pH range from 5.5 to 14. However, in an acidic medium (pH < 5.5) the absorption spectra showed a red shift of ∼79 nm and a new band appeared at 437 nm (Fig. S14a†); however, the observed enhanced emission was very close to the excimer emission (Fig. 4 and S14b†). Moreover, the ratio of monomer to excimer emission (I_E/I_M) as a function of pH markedly decreased from pH 6, while almost negligible change occurred at very high and low pH values. It is noteworthy to mention that the hydrophobic association between the alkyl chains plays an important role in determining the overall electrostatic interaction and self-aggregation.³¹ Thus, due to variable interaction, the possibility for the existence of different types of excimers in the medium cannot be ignored. In addition, the pH-dependent photophysical response of the complex 2 + Hg²⁺ was examined and found to be almost constant in the pH range of 4–9. At high pH (≥9), an abrupt decrease in the relative emission intensity could be ascribed to the dissociation of complex 2 + Hg²⁺.²⁸ Thus, the fluorescence response of 2 can be utilized in the pH range of 5.25–8.93, within which most of the biological samples can be tested.

Fig. 4 pH-emission plot exhibiting change in relative fluorescence intensity of probe 2 and 2 + Hg²⁺ at different pH in phosphate buffer (10 mM; 10% aqueous ACN). Inset: shows change in ration of monomer and excimer emission at different pH.

Limit of detection of probe 2

To estimate the limit of detection (LOD) for probe 2, a calibration curve was constructed according to a previously reported method.²⁸ The almost linear calibration curve gave a standard deviation (σ) of 0.23172 (Fig. 5a). Similarly, the slope of the fluorescence plot obtained from the change in relative fluorescence intensities ΔI(I − I₀) with respect to different concentrations of Hg²⁺ (10⁻⁴ μM) gave a calibration sensitivity (m) of 20.633 (Fig. 5). The limit of detection (LOD) was estimated using eqn (3) and was found to be 3.4 pM, which is good and comparable to other reported methods.²⁸

Fig. 5 (a) Calibration curve of relative emission intensities and different concentrations of probe 2 and (b) calibration sensitivity (m) plot for 2 + Hg²⁺.

Mechanism of complexation between 2 and Hg²⁺

To gain an insight into the mode of interaction, ¹H NMR, FT-IR and HRMS spectra of probe 2 and the complex 2 + Hg²⁺ were obtained. The ¹HNMR spectrum of 2 showed resonances corresponding to pyrene ring protons as a multiplet at δ 8.36–8.08 ppm (H4, H6, H7, H8 and H10, H11, H13), whereas the doublets appearing at δ 9.12 and 8.56 ppm are attributable to H3 and H14 protons, respectively (Fig. 6, S1†). Similarly, the resonances appearing as a singlet and doublet at δ 3.79, 3.87 and 9.34 ppm were assigned to H3′, H4′ (–CH₂) and aldimine (–CH=N), H1′ protons, respectively. The resonance corresponding to the OH group proton appeared at δ 4.73 ppm. In contrast, the ¹HNMR spectrum of the probable complex 2 + Hg²⁺ exhibited a significant downfield shift corresponding to the resonances of H1′ (Δδ = 1.46), H3 (Δδ = 0.47), H14 (Δδ = 0.43), pyrene unit (Δδ = 0.56 ppm) and H3′ (Δδ = 0.18) protons. Similarly, the –OH proton resonances shifted downfield and became broadened to appear at δ 4.79 ppm (Δδ = 0.57 ppm), while the H4′ proton shifted upfield (Δδ = 0.04 ppm) to appear at δ 3.83 ppm (Fig. S15–S17†). Moreover, the FT-IR spectrum of 2 showed characteristic stretching vibration bands at 3439, 2943, 1631, and 1077 cm⁻¹ corresponding to –OH, –CH, –C=N, and –CO functions, respectively (Fig. S18†). Upon interaction with Hg²⁺, the OH stretching vibration disappeared almost completely and the –C=N stretching vibration underwent a blue shift and appeared at 1617 cm⁻¹. In the mass (HRMS) spectra, the molecular ion peak at m/z⁺ 273.1156 (calc. 273.1153) for probe 2 appeared at m/z⁺ 870.8308 (calc. 870.8306) upon complexation with Hg²⁺ (Fig. S5 and S19†). Thus, the observed spectroscopic data clearly supported the interaction of probe 2 with Hg²⁺ through the O and N atoms of the hydroxyl and aldimine functions in 2 : 1 stoichiometry, along with a geometrical change in the structure of probe 2 from E to Z isomer³² as predicted in Scheme 2.

Fig. 6 Change in ¹H NMR spectra of 2 (c = 1.02 × 10⁻² M) upon addition of Hg²⁺ (0.5, 1 and 1.5 equiv.) in DMSO-d₆.

Scheme 2 Proposed mechanism of interaction between 2 and Hg²⁺.

Quantum chemical calculations

To have a better understanding of the binding mode and optical response of 2 toward Hg²⁺, energy optimization and frequency calculations were performed on 2 and its complex 2 + Hg²⁺. Quantum chemical calculations were performed at the B3LYP density functional theory (DFT) level using B3LYP/6-31G(d,p) for 2 and B3LYP/LANL2DZ for the 2 + Hg²⁺ complex. All the frequency calculations and geometry optimizations were performed in the Gaussian 03 program suite³³ to make sure that the optimized structure is a real minimum. The DFT optimized structure revealed two bond lengths between Hg²⁺ and the N atom of aldimine function at 2.45 Å and 2.65 Å, while the bond length between Hg²⁺ and the O atom of hydroxyl function was 1.92 Å. The optimized structure of the 2 − Hg²⁺ complex (Fig. 7) suggested that Hg²⁺ coordinated with the N and O donor atoms of 2 in an octahedral geometry.

Fig. 7 DFT optimized minimum energy structure of 2 and 2 − Hg²⁺ complex.

Optoelectronic behavior of probe 4

To accomplish the goal of solid state chemosensing and extraction of mercury in real samples, a reusable sensory material was successfully prepared by immobilizing modified probe 2 onto the surface of mesoporous silica gel by the sol–gel method.³⁴ Due to the strong affinity between the functional triethoxysilyl and silanol units, the immobilization of probe 2 on the silica gel surface occurred successfully through silyl-etheral (–Si–O–Si–) linkages.³⁵ The yellow-orange colored mesoporous solid chemosensing probe 4 was characterized by FT-IR, AFM and SEM spectral data. The FT-IR spectrum (Fig. S8†) of 4 with respect to free silica gel exhibited aliphatic and aromatic (C–H and C=C) stretching vibration bands at 3167, 2925, and 1444 cm⁻¹, while the broad stretching vibration band due to OH function at 3444 cm⁻¹ disappeared completely. The appearance of typical –Si–O–Si– and –C=N stretching vibration bands at 1089 and 1644 cm⁻¹, respectively, provided solid evidence for the attachment of 2 onto the mesoporous silica surface.

To evaluate the aggregation and its effect on the optical behavior of the probe, a solution of 4 in methanol was subjected to scanning electron microscopy (SEM) and atomic force microscopy (AFM) before and after the addition of Hg²⁺ ions. The SEM images, as shown in Fig. 8a and b (Fig. S20†), suggest an inconsistent pattern of silica gel particles with an average diameter of 20 μm. This could probably be attributed to the formation of different types of aggregates on the surface. Moreover, when the images of the Hg²⁺-treated probe were acquired, the average particle size was substantially reduced to ∼10 μm (Fig. 8b). This suggested that Hg²⁺ induced enhanced aggregation and roughness in comparison to the unmodified silica-gel polymeric material (Fig. 8a, b and S21†). Thus, the change in the macroscopic properties due to morphological variation resulted in changes in the color of the silicate material from an orange red to yellow-green color, which also indicated the attachment of the probe to the surface of silica gel (Fig. 9). Moreover, the shape and size of these aggregates are in good agreement with the data obtained from AFM measurements, in which the average root mean square roughness of 4 and the peak to valley height area were found to be 2.077 and 1.507 nm, respectively (Fig. 8). The mechanism of aggregate build-up has been deduced by considering the dominance of the solid state interactions in the aggregated state, which was further evidenced by the powder XRD pattern (Fig. 8c). Thus, the modified and aggregated rough surfaces of 4 make it a suitable potential solid material for the detection and extraction of Hg²⁺ from contaminated solutions and the environment.

Fig. 8 SEM images of 4 (a (top)) exhibiting crystalline, cylindrical, and spherical morphology with a particle size of 20 μm and (b (top)) Hg²⁺-treated aggregated silica gel particles with a particle size of 10 μm. AFM 3D images show surface roughness of silica-gel particle 4 before (a (bottom)) and after (b (bottom)) Hg²⁺ treatment. (c) Powder XRD pattern for 4 and 4 + Hg²⁺.

Fig. 9 Images for (A) colorimetric (B) fluorogenic color change of 2 and 4 upon interaction with Hg²⁺ in solid state with plausible mode of interaction.

Selectivity of probe 4 for metal ions

The optoelectronic behavior of a sol solution of 4 was examined in the absence and presence of tested metal ions in phosphate buffer (10 mM pH 7.0; 10% aqueous ACN). Interestingly, the optoelectronic behavior of 4 was found to be almost consistent with that of probe 2. The characteristic low and high energy π → π* electronic transition bands appeared at 393 nm (ε = 27 000 M⁻¹ cm⁻¹), 360 nm (ε = 35 400 M⁻¹ cm⁻¹) and 286 nm (ε = 52 000 M⁻¹ cm⁻¹) (Fig. 10). Upon excitation at 360 nm, probe 4 displayed a broad weak monomer emission at 390 and 413 nm. Subsequently, we examined the affinity of 4 for tested metal ions (2.0 equiv.). Probe 4 has high sensitivity and selectivity for Hg²⁺, in which the absorption band centered at 286 nm exhibited a large hyperchromic shift, while the band centered at 360 nm displayed a blue-shift of ∼3 nm (λ_max = 357 nm; ε = 41 600 M⁻¹ cm⁻¹). Similarly, upon interaction with Hg²⁺, probe 4 (at λ_ex = 360 nm) exhibited enhanced excimer emission at 502 nm (Fig. 10). In addition, the other tested metal ions failed to induce any significant change in the optical behavior of the probe. Moreover, the competitive metal ion interference studies also suggested high selectivity of 4 for Hg²⁺ ions (Fig. S22†). The Job's plot analysis consistently revealed a 2 : 1 stoichiometry for the interaction between 4 and Hg²⁺ ions, for which binding constants were estimated from the absorption and emission titration experiment data, and were found to be K_ass.(abs) = 3.60 × 10⁶/M^1/2 (Fig. S23c†) and K_ass.(em) = 2.5 × 10⁵/M^1/2, respectively (Fig. S23d†).

Fig. 10 Change in absorption (inset) and emission spectra of sol solution of 4 (5 μM) upon interaction with different metal ions (2.0 equiv.) in phosphate buffer (10 mM pH 7.0; 10% aqueous ACN, v/v). Change in solid state emission spectra of 4 upon interaction with Hg²⁺ and EDTA.

Next, we tried to acquire the emission spectra of 4 in the solid state. In the solid state, probe 4 showed a broad weak monomer emission at 385 and 410 nm upon excitation at λ_ex = 345 nm. Upon interaction with Hg²⁺, probe 4 showed a noticeably enhanced emission at ∼510–540 nm and the color of the silica derived probe changed from a naked-eye detectable orange red to yellow-green color (Fig. 10). In addition, when 4 was treated with EDTA solution, the monomer emission was revived. This process was repeated five times and the intensity was found to be almost constant. Thus, it is noteworthy to mention that the solid material 4 may be further practically explored as a reusable solid fluorescent chemosensor/optoelectronic material to detect and extract Hg²⁺ in the environment.

Affinity of probe 2 to detect Hg²⁺ in biological medium

The potential applicability of probe 2 to detect Hg²⁺ in a biological medium, such as blood plasma protein BSA (bovine serum albumin) and HSA (human serum albumin), was also studied. We previously observed²⁸ that Hg²⁺ quenches (∼83%) the typical inherent emission of BSA (λ_em = ∼345 nm) (Fig. S24†). Therefore, to understand the possibilities of nonbonding interaction, we first investigated the photophysical behavior of 2 in the presence of BSA and HSA in phosphate buffer. Interestingly, the emission behavior of 2 under 345 nm excitation remained independent of the concentration of BSA/HSA (Fig. S25a†), even at high concentrations of 0–2 μM. Thus, the data suggest that the photophysical behavior of probe 2 in protein medium does not get affected due to possible H-bonding and/or electrostatic interactions as well as due to the interaction of different types of electrolytes present in blood serum. Therefore, we examined the affinity of 2 for the detection of Hg²⁺ in a protein medium. It is important to mention that in the presence of BSA and HSA (2 μM BSA/HSA, phosphate buffer pH 7.0), probe 2 (10 μM) upon interaction with Hg²⁺ (2.0 equiv.) showed considerably enhanced (∼75%) emission at ∼506 nm (Fig. 11, and S25b†), and the readily detectable naked-eye observable weak fluorescent blue color of the solution (turn-off) changed to a bright fluorescent blue-green color (turn-on) (Fig. 11b inset). The usability of 2 to determine Hg²⁺ in biological samples was assessed by an emission titration experiment. The emission titration pattern displayed by 2 in human blood serum was found to be consistent with the behavior observed in an aqueous solution (Fig. 11b). In human blood serum, the limit of detection for Hg²⁺ was estimated to be above and was found to be 0.02 μM (4.31 ppb) (Fig. S26†). Thus, it is worth mentioning that probe 2 can be utilized as a potential chemosensor to detect Hg²⁺ in biological samples.

Fig. 11 Changes in emission (a) and titration (b) spectra of 2 (10 μM probe; 2 μM HSA, pH 7.0) with Hg²⁺ (0–2 equiv.) in phosphate buffer (10 mM pH 7.0; 10% aqueous ACN, v/v). Inset: color change after the addition of Hg²⁺ in 2 containing human blood serum.

Detection of Hg²⁺ in living cells. The good response of probe 2 for the smart detection of Hg²⁺ in a protein medium prompted us to study the feasibility of 2 to detect Hg²⁺ in vivo in live cells through confocal fluorescence microscopy. According to a previous study,³⁶ HeLa cells were first incubated directly with different concentrations (10, 20, 30, 40, and 50 μM) of 2, and after proper washing (5 min, two times) with phosphate buffer (1× PBS saline, pH 7.4), the cells were visualized under a microscope. The observed bright blue colored fluorescence indicated the excellent cell permeability of 2 in HeLa cells (Fig. 12). Moreover, to understand the cell viability and cytotoxic tolerance of HeLa cells, MTT assay was performed³⁶ corresponding to probe 2 and its complex 2 + Hg²⁺. The MTT assay suggested that more than 50% of the cells were viable up to the concentration of 30 μM of probe 2 and its complex 2 + Hg²⁺. Therefore, the of concentration 30 μM of probe 2 and 2 + Hg²⁺ was chosen as the optimum concentration to incubate HeLa cells separately in the dark. The confocal microscopic images of the HeLa cells (at λ_ex = 380 nm) illustrated the intense enhanced green fluorescence for 2 + Hg²⁺ (Fig. 12, images D, E and F). Thus, the cell imaging experiment suggests the potential application of probe 2 to detect Hg²⁺ in live cells.

Fig. 12 Confocal fluorescence images showing the localization of 2 (blue in panel B) and 2 − Hg²⁺ (green in panel E) in the cytoplasm of live HeLa cells. DIC images of HeLa cells (panel A and D). Merged fluorescence images of probe 2 and 2 + Hg²⁺ (panel C and F). MTT assay histogram for (a) probe 2 and (b) 2 + HgNO₃.

Analytical applications

Detection of Hg²⁺ on cellulose strip kit. To further confirm the analytical use of probe 2, a paper strip test was performed. Small cellulose paper strips (Whatman™) containing different concentrations of probe 2 (5, 3, and 1 μM) were prepared (1.5 × 2.0 cm²) in 10% aqueous ACN and dried in air. Aqueous Hg(NO₃)₂ solutions of three different concentrations (1 × 10⁻⁵, 1 × 10⁻⁷, and 1 × 10⁻⁹ M) were prepared and dried test paper strips were dipped into them for 10 min. The observed fluorescent green color of the air dried paper strips under UV light (at 365 nm) demonstrated the potential application of probe 2 to detect Hg²⁺ on the solid surface (Fig. 13).

Fig. 13 Fluorescent test paper strips of probe 2 (a) 5 μM, (b) 3 μM, (c) 1 μM before (blue) and after addition of Hg²⁺ (green) at concentrations of (a) 1 × 10⁻⁵, (b) 1 × 10⁻⁷, and (c) 1 × 10⁻⁹ M under UV light at 365 nm.

Detection of Hg²⁺ in solid state and in real contaminated water samples. To validate the practical analytical utility of probes 2 and 4 to determine the Hg²⁺ concentration in real contaminated water samples, we first quantified the fluorescence of probe 2 and 4 (10 μM) in the presence of various concentrations of Hg²⁺ ions (0–5 μM), and the corresponding calibration plots were prepared as the standard curves (Fig. 14a and b). Considering the possible interference that other components may cause in real samples, the level of Hg²⁺ in real water samples was determined using the standard addition method.²⁸ Prior to real sample detection, when probe 2 and 4 were added directly to the water samples, no significant fluorescence enhancement occurred. However, when the emission spectra of treated contaminated water samples were acquired by this method, the recovery of Hg²⁺ with respect to the standard calibration curve was excellent. Notably, we could quantify Hg²⁺ contamination in real water samples in the range of 116–92% (Tables 1 and 2). Moreover, the color of the solutions also changed from fluorescent blue to blue-green, which was readily visual to the naked eye (Fig. 14a and b inset).

Fig. 14 Calibration sensitivity plot of probe (a) 2 and (b) 4 toward Hg²⁺. Insets: change in color of probe 2 and 4 after treating with Hg²⁺ solution (A = 0.05, B = 0.5, and C = 5 μM) in phosphate buffer (10 mM; pH 7.0; 10% aqueous ACN).

Table 1 Estimation of Hg²⁺ in water samples with probe 2

Sample concentration (μM)		Hg^2+ recovered	% Recovery of Hg^2+ from the sample
Ganga river water	0	Not detected	—
	0.05	0.054 ± 0.014	108
	0.5	0.49 ± 0.021	98
	5	4.72 ± 0.18	94
Tap water	0	Not detected	—
	0.05	0.052 ± 0.014	104
	0.5	0.49 ± 0.021	98
	5	4.89 ± 0.22	98

---

Table 2 Estimation of Hg²⁺ in water samples with probe 4

Sample concentration (μM)		Hg^2+ recovered	% Recovery of Hg^2+ from the sample
Ganga river water	0	Not detected	—
	0.05	0.058 ± 0.018	116
	0.5	0.48 ± 0.025	96
	5	4.82 ± 0.21	96
Tap water	0	Not detected	—
	0.05	0.056 ± 0.018	112
	0.5	0.46 ± 0.024	92
	5	4.88 ± 0.22	98

---

Conclusion

In conclusion, a novel pyrene-based fluorescent probe was designed and synthesized to detect Hg²⁺ sensitively and selectively in partial aqueous medium and biological medium as well as in solid state through a fluorescence turn-on response. The interaction of Hg²⁺ with aldimine and hydroxy fragments results in excimer formation, where the two pyrene rings acquire stacked arrangements to enhance the π–π interaction, and consequently an enhanced emission was observed. Moreover, the non-fluorescent solution changed to a fluorescent blue green, which was readily visual to the naked eyes. The application of the probe to detect Hg²⁺ in human blood serum, live cells, and a test paper strip as well as in real water samples suggested the promising prospect of Hg²⁺ for sensing in biological and environmental media. The present methodology also provides a strategy to detect and extract traces of Hg²⁺ in the solid state by the use of reusable mesoporous supramolecular aggregates.

Conflict of interest

Authors declare no competing financial interest.

Acknowledgements

The authors are thankful to the Council of Scientific and Industrial Research (CSIR), New Delhi for financial support (CSIR 02(0199/14/EMR-II)) and fellowships (to SSR, RA and PS) to carry out research work. Authors are also thankful to SAIF CDRI, Lucknow for providing ESI-MS spectral data.

References

1. J. Wu, W. Liu, J. Ge, H. Zhang and P. Wang, Chem. Soc. Rev., 2011, 40, 3483–3495.
2. A. K. Dwivedi, G. Saikia and P. K. Iyer, J. Mater. Chem., 2011, 21, 2502–2507.
3. X. Lou, D. Ou, Q. Li and Z. Li, Chem. Commun., 2012, 48, 8462–8477.
4. G. Aragay, J. Pons and A. Merkoci, Chem. Rev., 2011, 111, 3433–3458.
5. M. Sauer, Angew. Chem., Int. Ed., 2003, 42, 1790–1793.
6. K. Hanaoka, Y. Muramatsu, Y. Urano, T. Terai and T. Nagano, Chem.–Eur. J., 2010, 16, 568–572.
7. M. Mameli, M. C. Aragoni, M. Arca, C. Caltagirone, F. Demartin, G. Farruggia, G. D. Filippo, F. A. Devillanova, A. Garau, F. Isaia, V. Lippolis, S. Murgia, L. Prodi, A. Pintus and N. Zaccheroni, Chem.–Eur. J., 2010, 16, 919–930.
8. J. Yin, Y. Hu and J. Yoon, Chem. Soc. Rev., 2015, 44, 4619–4644.
9. A. W. Czarnik, ACS symposium series 538, Washington, D.C., 1992, p. 235.
10. A. P. deSilva, D. B. Fox, A. J. M. Huxley and T. S. Moody, Coord. Chem. Rev., 2000, 205, 41–57.
11. E. M. Nolan and S. J. Lippard, Chem. Rev., 2008, 108, 3443–3480.
12. X. Chen, T. Pradhan, F. Wang, J. S. Kim and J. Yoon, Chem. Rev., 2012, 112, 1910–1956.
13. J. Wang and X. Qian, Chem. Commun., 2006, 109–111.
14. S. Yoon, A. E. Albers, A. P. Wong and C. Chang, J. Am. Chem. Soc., 2005, 127, 16030–16031.
15. A. Ono and H. Togashi, Angew. Chem., Int. Ed., 2004, 43, 4300–4302.
16. P. Chen and C. A. He, J. Am. Chem. Soc., 2004, 126, 728–729.
17. M. Hollenstein, C. Hipolito, C. Lam, D. Dietrich and D. M. Perrin, Angew. Chem., Int. Ed., 2008, 47, 4346–4350.
18. T. J. Dickerson, N. N. Reed, J. J. LaClair and K. D. Janda, J. Am. Chem. Soc., 2004, 126, 16582–16586.
19. J. Y. Kwon, Y. J. Jang, Y. J. Lee, K. M. Kim, M. S. Seo, W. Namand and J. Yoon, J. Am. Chem. Soc., 2005, 127, 10107–10111.
20. G.-K. Li, Z.-X. Xu, C.-F. Chen and Z. T. Huang, Chem. Commun., 2008, 1774–1776.
21. M. J. V. Ros-Lis, D. Marcos, R. Martinez-Manez, K. Rurack and J. Soto, Angew. Chem., Int. Ed., 2005, 44, 4405–4407.
22. J. Ding, H. Li, C. Wang, J. Yang, Y. Xie, Q. Peng, Q. Li and Z. Li, ACS Appl. Mater. Interfaces, 2015, 7, 11369–11376.
23. Y.-R. Zhang, Q.-R. Wang, P. Su, F. Zhao, J. Huanga and B.-X. Zhao, RSC Adv., 2015, 5, 20634–20638.
24. Y. Xua, Z. Jiang, Y. Xiao, T.-T. Zhang, J.-Y. Miao and B.-X. Zhao, Anal. Chim. Acta, 2014, 807, 126–134.
25. W.-Y. Liu, S.-L. Shen, H.-Y. Li, J.-Y. Miao and B.-X. Zhao, Anal. Chim. Acta, 2013, 791, 65–71.
26. Y. Zhang, Y. Yang, J. Hao, C. Yin, F. Huo, J. Chao and D. Liu, Spectrochim. Acta, Part A, 2014, 132, 27–31.
27. F.-J. Huo, Y.-Q. Sun, J. Su, Y.-T. Yang, C.-X. Yin and J.-B. Chao, Org. Lett., 2010, 12, 4756–4759.
28. P. Srivastava, S. S. Razi, R. Ali, R. C. Gupta, S. S. Yadav, G. Narayan and A. Misra, Anal. Chem., 2014, 86, 8693–8699.
29. M. Shahid, S. S. Razi, P. Srivastava, R. Ali, B. Maiti and A. Misra, Tetrahedron, 2012, 44, 9076–9084.
30. H. A. Benesi and J. H. Hildebrand, J. Am. Chem. Soc., 1949, 71, 2703–2707.
31. T. Joy Kunjappu and P. Somasundara, Langmuir, 1995, 11, 428–432.
32. Y. Saito, K. Matsumoto, Y. Takeuchi, S. S. Bag, S. Kodate, T. Morii and I. Saito, Tetrahedron Lett., 2009, 50, 1403–1406.
33. M. J. Frisch, G. W. Truncks and H. B. Schlegel, GAUSSIAN 03, Revision, E.01, Gaussian Inc., Wallingford CT, 2007.
34. Z. Sun, D. Guo, L. Zhang, H. Li, B. Yang and S. Yan, J. Mater. Chem. B, 2015, 3, 3201–3210.
35. Y. Chen, C. L. Droumaguet, K. Li, W. E. Cotham, N. Lee, M. Walla and Q. Wang, J. Am. Soc. Mass Spectrom., 2010, 21, 403–410.
36. P. Srivastava, S. S. Razi, R. Ali, S. Srivastav, S. Patnaik, S. Srikrishna and A. Misra, Biosens. Bioelectron., 2015, 69, 179–185.

---

Footnote

† Electronic supplementary information (ESI) available: Synthesis, experimental details, ¹HNMR, ¹³CNMR, FT-IR, ESI-MS, HRMS, UV-Vis and fluorescence spectra. See DOI: 10.1039/c5ra13021b

---