Aggregation-Induced Emission-Active Hydrazide-Based Probe: Selective Sensing of Al3+, HF2 -, and Nitro Explosives.

(E)-N'-((2-Hydroxynaphthalen-1-yl)methylene)picolinohydrazide (H-PNAP) shows aggregation-induced emission (AIE) strictly in a 90% water/MeOH (v/v) mixture at 540 nm, and the solid-state emission is blue-shifted to 509 nm upon excitation at 400 nm. The AIE activity of H-PNAP is selectively quenched by 2,4,6-trinitrophenol (TNP) and 2,4-dinitrophenol (DNP) out of different nitroaromatic compounds with a limit of detection (LOD) of 7.79 × 10-7 and 9.08 × 10-7 M, respectively. The probe is nonemissive in aqueous medium (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, HEPES buffer, pH 7.2); however, it shows a strong emission to Al3+ (λem, 490 nm) in the presence of 17 other biological metal ions, and the LOD is 2.09 nM which is far below the WHO recommended value (7.41 mM). The emission of the [Al(PNAP)(NO3)2] complex is quenched by HF2 - (F- and PO4 3- are weak quencher), and the LOD is as low as 15 nM. The probable mechanism of the sensing feature of the probe has been authenticated by 1H nuclear magnetic resonance titration, mass spectrometry, Fourier transform infrared spectroscopy, Benesi-Hildebrand plot, and Job's plot in each case. The probe has some practical applications such as recovery of Al3+ from the drinking water sample, construction of the INHIBIT logic gate, and detection kits for Al3+ and TNP/DNP by simple paper test strips. The probe, H-PNAP, has successfully been applied to the detection of intracellular Al3+ and HF2 - ions in the human breast cancer cell, MDA-MB-468.

Introduction

Heavier cations (higher than Zn2+) are commonly toxic and some lower atomic weight cations like Be2+ and Al3+ are not welcome by human cells.[1] Besides, the concentration of ions has imperative impact to regulate the biochemical activity and the rates of the reactions in living cells. Even essential ions in higher or lower concentration than that of biological concentration are fatal to human health.[2] Therefore, accurate and precise measurement of ions at a very low concentration level is important in chemistry and biology. One of the important metals, aluminum (Al), has huge applications in the society because of noncorrosive nature, good conductivity, use in making cooking utensils, potable water supplies, cosmetics, and so forth.[3,4] Thus, Al enters into human body easily and may be responsible for neurotoxicity, bone deformation, disruption of reproduction, and so forth.[5] This is one of the reasons for the detection of Al3+ in drinking water, food, beverages, and so forth. Al(III), a p-block element, does not form colored complexes, and its measurement needs special design of chemosensors.

In recent times, the design of sensors (ions or small molecules) and their application in live cells for the detection of ions/molecules have received active interest from chemistry, physics, chemical engineering, electrical and electronic engineering, and many other branches. On considering the social and biological importance of Al(III), some examples of fluorescent sensors have been reported for Al3+ detection with the Schiff base scaffold of 4-aminoantipyrine derivatives,[6] rhodamine B ligands,[7] coumarinyl derivatives,[8] benzophenone,[9] chromone,[10] diphenyl pyrrol derivatives,[11] pyrazoles,[12] quinolones,[13] fluorenes,[14] benzimidazoles,[15] and so forth. Besides, a limited number of sensors exhibit aggregation-induced emission (AIE), and their emissivity, in some cases, has been influenced by Al3+ ions.[16] Literature review shows that functionalized naphthyl, polyaromatic hydrocarbon (PAH) scaffolds, phenolphthalein-based probe, metal–organic-frameworks (MOFs),[17] and so forth are specifically important for ion-sensing research. Hydrazide function has been commonly used in the detection of ions.[18] Furthermore, picolinohydrazide  attached with naphthyl scaffold is a well-known fluorophore for its selective and sensitive ion-sensing activity.[19] In this work, we have designed hitherto an unknown probe (E)-N′-((2-hydroxynaphthalen-1-yl)methylene)picolinohydrazide (H-PNAP), which serves as highly sensitive for the detection of Al(III) at nM concentration (limit of detection, LOD, 2.09 nM).

Anions are also useful to maintain fluid ionic pressure, nerve impulse, ion transportation, and so forth; however, some of the anions are toxic to human health and aquatic environment, such as HF2–, CN–, AsO43–, AsO33–(AsO2–), F–, NO3–, NO2–, and so forth. Over the last two decades, several efforts have been devoted to design several receptors for anionic species.[20] There are two approaches to design anion sensors—N-heterocyclic fluorogenic species which on protonation may enhance emission which could capable to form hydrogen bond with anions and binding strength may regulate emission efficiency; and/or emissive metal complexes may chemically prefer specific anion to eject M from the sensor cavity and emission intensity is quenched or wavelength is shifted. These signaling processes have been used for quantitative identification of anions. Herein, the emissive Al3+–H-PNAP complex serves as a sensitive fluorogenic detector of HF2–.

The probe, H-PNAP, undergoes aggregation and shows AIE. The AIE property of naphthyl–picolinyl hydrazide (H-PNAP) shows an amazing sensitivity to detect nitro-explosives, mainly 2,4,6-trinitrophenol (TNP) and 2,4-dinitrophenol (DNP). The nitro-explosives are used in terroristic activities and become a concern to the Crime Bureau of Intelligence and Home Department of the government.[21−23] Although many other methods of detection of nitro-explosives such as, police dog detection, plasma desorption mass spectrometry, ion migration spectrum, energy-dispersive X-ray diffraction, surface-enhanced Raman spectroscopy, or additional imaging techniques are known,[24−29] none of them are economically viable. However, fluorescence-based explosive detection is commercially realistic, of high-speed, and portable.[30] Thus, the probe, H-PNAP, is useful for Al3+, HF2–, and nitrophenol detection. The probe, H-PNAP, is characterized by single-crystal X-ray diffraction measurement along with other spectroscopic data.

Results and Discussion

Synthesis and Formulation of the Probe

The condensation of picolinohydrazide and 2-hydroxy-1-naphthaldehyde in dry MeOH has synthesized the probe H-PNAP (Scheme S1), and the 1H nuclear magnetic resonance (NMR) spectrum (dimethylsulfoxide, DMSO-d6) shows the characteristic peaks at 12.73 (s, 1H, OH), 12.38 (s, 1H, NH), and 9.62 (s, 1H, imine-H) which supports the structure of the probe (Figure S1). The 13C NMR (in DMSO-d6) spectrum also supports the presence of signals at 160.49 (C=O), 158.63 (C=N), and others (Figure S2). Electrospray ionization mass spectrometry (ESI-MS) spectrum also shows the ion peak at 292.1084 which resembles the formation of [H-PNAP + H]+ and at 314.0985 for [H-PNAP + Na+] (base peak) (Figure S3). The IR spectrum shows ν(C=N) at 1620 cm–1, ν(C=O) at 1661 cm–1, ν(NH) at 3191 cm–1, and ν(phenolic-OH) at 3482 cm–1 (Figure S4). The yellow X-ray diffractable crystal of H-PNAP belongs to the monoclinic system of space group P21/n (Figure ). The reaction of the methanol solution of Al(NO3)3 with H-PNAP separates the orange-colored [Al(PNAP)(NO3)2] which shows the characteristic 1H NMR (δ, ppm) (500 MHz, DMSO-d6) singlet peak at 12.388 ppm for −NH and CH=N at 9.624 ppm along with other peaks (Figure S5); the ESI-MS peak for the HPNAP–Al3+ complex appears at 442.1086 (calculated mass of [Al(PNAP)(NO3)2], 442.06) (Figure S6) and the IR spectrum (Figure S7) shows ν(NH), 3110; ν(C=O), 1662; ν(C=N), 1612 cm–1; and ν(NO3), 1388 cm–1.

Figure 1

Molecular structure of the probe, H-PNAP.

The picolinoyl hydrazide and naphthyl moieties are linked by an imine (C=N) bond. An intramolecular hydrogen bond is present between N(1) and H(5) with a bond distance of 2.22 Å and an intermolecular H bond is also present between O(1) and H(15) with a bond distance of 2.1 Å (Figure ). The details about the crystal structure, important distance, and bond angles are incorporated in Tables S1 and S2 in the Supporting Information. In a single unit, there is a water molecule, and this water molecule can form layer-to-layer intermolecular H-bonding interactions (Figure a). Wave-like supramolecular aggregation is formed by weak π...π interactions of the probe (Figure b).

Figure 2

(a) Layer-to-layer H-bonding interactions with water molecules and (b) wave-like supramolecular aggregation of the probe.

Photophysics and Application of AIE Property

The UV–vis spectrum of H-PNAP in methanol shows absorption bands at 326 and 366 nm, which are assigned to πnaphthyl → πpicolinohydrazide transitions that have been supported by the time-dependent density functional theory (TDDFT)/conductor-like polarizable continuum model (CPCM) method of computation. In solid state, these bands are red-shifted by 25–40 nm. The excitation of the probe in crystal state at 400 nm exhibits high-intense green fluorescence (λem, 509 nm) (Figure ). The fluorescence spectrum of the probe, H-PNAP, in amorphous phase is also recorded (Figure S8), and it exhibits a strong emission, but weaker than the emission in the crystalline state. The rapport of these observations is that the compound has crystallization-induced emission enhancement property.[31] Restriction of different physical processes like rotation, vibration, and isomerization in the solid state helps the emission process, whereas the solution phase of H-PNAP in pure MeOH and pure aqueous medium is nonemissive. In a pure solvent solution, the excited-state proton transfer (ESIPT), twisting, and photoinduced electron transfer (PET) processes (Scheme ) may be active to inhibit the energy emission program.

Figure 3

Solid-state fluorescence spectrum of H-PNAP on excitation by 400 nm; inset: images of H-PNAP crystals (a) under normal light and (b) under UV (λ, 365 nm) light.

Scheme 1

Deactivation Processes of Excited H-PNAP by PET, Twisting, and ESIPT

Although the pure methanol solution of H-PNAP is nonemissive, addition of water to this solution shows an amazing experience and becomes strongly emissive in a 90% binary mixture of H2O–MeOH (v/v) solution (Figure ). From 0 to 83% water/MeOH binary mixture of the probe, no emission is perceived, and AIE starts at 84% water/MeOH mixture and increases gradually up to 90% water content (Figures a and S9). Figure a shows that with an increase in the percentage of water in MeOH (from 84 to 90% water/MeOH), the emission band is shifted bathochromically (509–540 nm). This red shift in emission maxima arises because of J-type aggregation by the packing between monomers. Kasha’s exciton theory announced that the J aggregation of an analyte should red shift the emission band compared to monomer emission; this is because of transition to a lower excitonic state.[32] This AIE is maximum at 90% water content, and after this the emission intensity decreases, which indicates that on further addition of water (>90%) the aggregates may be ruptured that causes a decrease in emission intensity. This indicates that the probe (H-PNAP) is AIE-active. PAHs such as naphthalene, anthracene, acridine, pyrene, and so forth aggregate in solution upon changing the polarity of the solution and generally quench the emission process.[16] Tang et al., in 2001, reported a contradictive observation where aggregation assisted the enhancement of emission.[33] For the biosensing and imaging applications, the AIE-active fluorogenic probes have been exploited to function as novel optical materials.[16] The AIE effect may be because of the high rigidity of the probe in aggregated state which may cause restriction of intramolecular rotation.[34−36] The change of the absorption spectral feature is also distinctive upon dilution of the MeOH solution of H-PNAP with water (Figure b). Not only emission spectra but absorption spectra (Figure b) also show red-shifting upon change in the percentage of water content to the MeOH solution. In a pure MeOH solution of H-PNAP, the absorbance bands appear at 326 and 366 nm, and upon addition of water to this solution, a new band appears at 401 nm, which is also an indication of J-aggregates.[37]

Figure 4

Spectral change of H-PNAP (50 μM) with an increase in the percentage of water in MeOH; (a) fluorescence (λex, 400 nm), (b) absorbance, and (c) image of AIE of H-PNAP under UV irradiation (λ, 365 nm).

The aggregation has been supported by dynamic light scattering (DLS) and field emission scanning electron microscopy (FESEM) data (vide infra). Some morphological and structural changes are obtained by FESEM (Figure ) with the dilution of the probe H-PNAP from 0 to 90% water in MeOH. The probe in the pure methanol solution is a fibrous-shaped nanomaterial (26 ± 2.1 nm), whereas in the aggregated phase (90% water content), it has slipped bark (130 ± 7.4 nm)-like geometry. Again, changes in particle size are observed on changing the percentage of water content, that is, from 0 to 100% (v/v) MeOH–water by DLS measurements. For 0, 30, 90, and 100% water content in the binary mixture of the probe, the average diameters of the probe increase and are 30, 178, 391 and 305 nm, respectively (Figure S10). At 90% water content, the average diameter is maximum; this may be because of the reason of the highest number of excimers aggregated, and on increasing the dilution to 90–100% by water, there is a decrease in particle size, which can be because of the hydrolysis of the aggregates via the solvation process.

Figure 5

FESEM images: (a) H-PNAP in pure MeOH (26 ± 2.1 nm) and (b) H-PNAP in 90% water/MeOH (v/v) (130 ± 7.4 nm). (Scale bar = 200 nm).

The emission of the aggregated phase is supported by optical fluorescence microscopic images (Figure ). In pure methanol, H-PNAP does not show any emission with UV light excitation, but in the presence of 90% water content in MeOH, the probe, H-PNAP, shows a strong green emission. The fluorescence properties in the aggregated state are recognized for the fabrication of organic light-emitting diode material.[38]

Figure 6

Optical fluorescence microscopic images (solid state) of (A) H-PNAP (50 μM) in MeOH and (B) H-PNAP (50 μM) in 90% water/MeOH with UV light excitation.

Lifetime (τ) and quantum yield (φ) information also support longer stability and the highest φ of H-PNAP in the 90% water/MeOH mixture (τ, 1.01 ns; φ, 0.496) than pure water (τ, 0.98 ns; φ, 0.253), pure MeOH (τ, 0.0547 ns; φ, 0.0023), and 30% water/MeOH medium (τ, 0.113 ns; φ, 0.018) (Figure S11 and Table S3). The quantum yield of H-PNAP is 215 times increased, from 0.0023 to 0.496, by changing the percentage of water content in the binary mixture (H2O/MeOH) from 0 to 90%. From this observation, we found that with an increase in the percentage of water, the quantum yield (φ) of H-PNAP increases up to 90% water content, and then at 100% water content it decreases, which are supportive with the photophysical properties of the reported J-type aggregation.[39,40]

Nitroaromatics Sensing

The reversibility of the highly emissive aggregated probe, H-PNAP (90% water/MeOH), is tested by various nitroaromatic compounds (NACs) like 4-nitrophenol (4-NP), TNP, DNP, 4-chloronitrobenzene (Cl-NB), 1,3-dinitrobenzene (DNB), O-nitrophenol (O-NP), 3,5-dinitrobenzoic acid, 4-nitrotoluene, 4-nitrobenzoic acid, 4-nitroacetanilide, and 5-nitro salicylic acid. High-intensity fluorescence emission at 540 nm for the aggregated H-PNAP (90% water/MeOH) is selectively quenched in the presence of TNP and DNP only, out of the 11 NACs (Figure S12), and this may be because of the interaction of NACs (TNP and DNP) with the aggregated probe and can hamper the aggregation phenomena (Scheme ). NACs are electron-deficient, and the probe H-PNAP is electron-rich; hence, electron transfer may initiate from the excited H-PNAP to the analyte (TNP/DNP) and hence quenching. The quenching of fluorescence of the aggregated probe by NAC explosives may arise through the PET mechanism (Scheme S2).[41] The quenching efficiency (%) of the aggregated fluorescence (λem, 540 nm; v/v, 9:1 H2O/MeOH) of H-PNAP (50 μM) upon addition of different NACs with different concentrations is also checked, and it shows that among all the NACs, only TNP and DNP show the highest quenching efficiency around 95 and 92%, respectively (Figure ).

Scheme 2

Insertion of TNP/DNP to the Aggregated Probe H-PNAP

Figure 7

Quenching efficiency (%) of the excimer fluorescence of H-PNAP (50 μM) in 9:1 H2O/MeOH upon addition of different NACs at λem, 540 nm with λex, 400 nm.

Upon gradual addition of the TNP/DNP solution (0.01 mL × 10–3 M) to aggregated H-PNAP (90% water/MeOH), the emission intensity gradually decreases (Figure S13). The detection limits (3σ method) are 0.779 μM (TNP) and 0.908 μM (DNP) (Figures S14 and S15) and the quenching constants are 2.1 × 104 M–1 (TNP) and 3.2 × 104 M–1 (DNP) (Figures S16 and S17). There are some reports of MOFs,[42] metal complexes,[43] and simple molecules[35,44−46] that serve as chemosensors for the detection of TNP and DNP within the detection range 0.23–2.25 μM (Table S4), and the present result also appears in this detection range. H-PNAP in 90% water/MeOH solution is emissive in the wide pH range (2–12), and TNP/DNP can quench this emission in this varied pH range also (Figure S18).

The UV–vis absorption spectral change of H-PNAP (50 μM) in the AIE phase (90% water content) in the presence of TNP/DNP (2 equiv) (Figure S19) may also be explained on considering the formation of charge-transfer complex by the strong interaction of the electron-deficient TNP/DNP and π electron-rich naphthyl fluorophore.[47,48]

The fluorescence intensity ratio (F0/F) has been plotted against the concentration of TNP and DNP, and the nonlinear Stern–Volmer plots (Figure ) are obtained which may be because of static and dynamic quenching. To understand the insight mechanism, the fluorescence lifetime measurement of the sensor in the presence and absence of the quencher is carried out. The static quenching appears because of the binding of the sensor with the quencher at the ground state, and the dynamic quenching exists by diffusion-controlled collisions between the excited sensor and the quencher.[35] The fluorescence lifetime decay of H-PNAP (90% water/MeOH) remains intact both for TNP and DNP with different concentrations (5–100 μM) (Figures S20 and S21, Table S5), which demonstrates that the fluorescence quenching of the aggregated state occurs through a static mechanism by TNP and DNP, and an upward deviation may arise because of the incredible quenching ability at higher concentrations of TNP and DNP.

Figure 8

Plot of F0/F against the concentration of (a) TNP and (b) DNP.

Ion-Sensing Activity of the Probe

In a pure aqueous medium (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid––HEPES buffer, pH 7.2), the probe H-PNAP does not show any emission behavior upon excitation at 400 nm. Addition of biologically active cations viz., Na+, K+, Ca2+, Mg2+, Mn2+, Fe2+, Zn2+, Cd2+, Cr3+, Co2+, Ni2+, Pd2+, Hg2+, Cu2+, Ba2+, and Pb2+ in the aqueous solution of H-PNAP again has no significant change in the emission pattern and intensity. Upon addition of Al3+ in the pure aqueous medium (HEPES buffer, pH 7.2) of H-PNAP, the emission is increased nearly by 500-fold at 490 nm (Figure ). In pure water solution, the absorption peaks at 326, 366, and 440 nm of H-PNAP are observed. With an incremental addition of Al3+ in the pure aqueous medium (HEPES buffer, pH 7.2), the probe shows a gradual decrease in the intensity of absorption bands at 326 and 366 nm with the generation of a new band at 406 nm, forming an isosbestic point at 388 nm (Figure S22). This observation may be because of the existence of intramolecular charge-transfer transition through chelation.[49] The absorbance spectra are unchanged until the stoichiometry satisfies 1:1 molar ratio between H-PNAP and Al3+ and no longer changes with further addition of Al3+ in the solution of the probe.

Figure 9

Change in the emission spectrum of H-PNAP (50 μM) upon gradual addition of different metal ions (50 μM each) in pure water (HEPES buffer, pH 7.2); λex, 400 nm; inset: zooming image for H-PNAP and H-PNAP with other metal ions.

Naphthalene is a considerably potential fluorogenic agent, but the probe H-PNAP is nonemissive in a pure MeOH or a pure aqueous medium. The probe may form an intramolecular hydrogen bond and ESIPT which may be the reasons for nonradiative energy transfer of the excited state along with PET, twisting the process of quenching. Upon the addition of Al3+ out of a large number of biologically active cations viz. Na+, K+, Ca2+, Mg2+, Mn2+, Fe2+, Zn2+, Cd2+, Cr3+, Co2+, Ni2+, Pd2+, Hg2+, Cu2+, Ba2+, and Pb2+, the enhancement of emission may be explained by restricting ESIPT, PET, twisting, and inclusion of CHEF (Scheme S3).[50] The interaction of H-PNAP with Al3+ is also investigated by fluorimetric titration (Figure ), and the LOD for Al3+ is 2.09 nM (3σ method, Figure S23). Benesi–Hildebrand equation, [(Fmax – F0)/(F – F0)] versus 1/[Al3+], has been used (Figure S24) to calculate the binding constant, Kd (4.6 × 104 M–1). Use of naphthyl-appended fluorogenic scaffold for identification of very small quantity of Al3+ ions is known in the literature[51−54] (Table S6), and the present work reports the lowest LOD among them.

Figure 10

Change in emission spectra of H-PNAP (50 μM) upon gradual addition of Al3+ ions (0–50 μM) in pure water (HEPES buffer, pH 7.2) (λex, 400 nm); inset: emission under UV light (λ, 365 nm) for H-PNAP in the absence and presence of Al3+ (50 μM).

The stability of the excited probe and its aluminum complex, [Al(PNAP)(NO3)2], has been checked by fluorescence lifetime measurements. Both the probe and its complex show biexponential fluorescence decay profile with average lifetimes of 1.00 and 1.73 ns, respectively (Figure , Table S7). The higher stability of the complex on comparing with the free probe accounts for the direct influence of Al3+ in the electronic structure and energy of the molecular levels of H-PNAP. This conjecture is supported by DFT computation (vide infra).

Figure 11

Fluorescence lifetime plot of H-PNAP and [Al(PNAP)(NO3)2] in aqueous medium.

The selective detection of Al3+ by the probe has been appropriated by the intervention of various cations (Figure S25), and it shows that no significant interference is observed. H-PNAP is nonemissive in the wide range of pH (2–12) and the [Al(PNAP)(NO3)2] emission is untroubled in the pH range 2–11 (Figure S26). The low-intensity emission of the [Al(PNAP)(NO3)2] complex at a higher pH (pH, 12) may be because of the dissociation of the complex and the formation of Al(OH)3/Al(OH)2–.

The stoichiometry of the complexation of H-PNAP with Al3+ is obtained from the spectral titration, Job’s plot (Figure S27), which supports the formation of the 1:1 complex. Mass spectral analysis strongly supplements the formation of the 1:1 H-PNAP–Al3+ complex (Figure S6) with the molecular ion peak at m/z = 442.1086 ([Al(PNAP)(NO3)2], calculated mass = 442.06). The 1H NMR titration experiment (Figure S28) displays that upon gradual addition of Al3+, the peak at 12.73 ppm for phenolic OH gradually decreases, and at 1:1 mol ratio it is vanished, but the hydrazide NH peak remains intact. The mechanism for Al3+ sensing can be explained by switching OFF the PET and ESIPT processes of free H-PNAP along with switching ON the CHEF process (Scheme S3) on complexation. In the probe, the charge transfer from pyridyl N to the naphthyl ring at the excited state (PET ON) and the ESIPT of H-bonded six-member ring may be the reasons for the nonemissive feature. Upon binding with Al3+, the two above possibilities to the loss of energy are arrested, and CHEF is now ON and PET is OFF along with ESIPT being OFF, which may be accompanied by the remarkable fluorescence emission at 490 nm. The DLS measurement is done for the 1:1 complex for H-PNAP + Al3+, and it is observed that almost no significant difference in average particle size is observed (H-PNAP, 305 nm; H-PNAP + Al3+, 309 nm) (Figure S29). The probe senses Al3+ ion in pure water and also some other solvents like MeOH, EtOH, and CH3CN (emission intensity is lower compared to water) (Figure S30), but out of these solvents, water is green; so, we have performed the ion sensitivity in an aqueous medium. The single probe shows emission at two different wavelengths, one for the AIE emission at 540 nm in a 90% water/MeOH mixture and the other for the [Al(PNAP)(NO3)2] complex at 490 nm in a pure aqueous medium. Binding of a metal ion to organic fluorogen affects the energy of the states (S0, S1, ... S), and hence the emission energy, polarity, rigidity, relaxation, and so forth of the emitter have changed. Aggregation is normally a physical process and Al3+ binding is a chemical process; so, the effects of these processes are certainly different.

According to Pearson’s HSAB theory, Al3+ is a hard acid, and the hard base centers like O (phenolato, carboxylate, etc.) and N (amine, imine, azo, etc.) may prefer to bind Al3+. The probe H-PNAP has three potential monoanionic donor centers: O, N, O. Hence, according to the HSAB theory, the probe may favor to coordinate Al3+. There are as many as 25 hard acid centers (H+, Li+, Na+, K+, Be2+, Mg2+, Ca2+, Sr2+, Sn2+, Al3+, Ga3+, In3+, Cr3+, Co3+, Fe3+, Ir3+, La3+, Si4+, Ti4+, Zr4+, Th4+,U4+, VO2+ , and UO22+), whereas the stability of the complex depends on the strength of the binding interaction, charge, chelate ring size, and so forth under experimental conditions. Till date, no theoretical principle has developed to tune the fine experimental preference of donor centers to the metal ions. However, the fluorescence intensity is one of the finest experimental evidences to the preferential binding of the ligand to the metal-ion center. The probe, H-PNAP, is nonemissive in a pure aqueous medium, but becomes highly emissive in the presence of Al3+ (λem, 490 nm) out of 17 other biologically important metal ions. The selective sensing of Al3+ is unclear, and it may be explained considering the CHEF process; the hydrated Al3+ ion can fit to the cavity of H-PNAP nicely, but other ions may not show such an effect. The binding cavity of the O,N,O pocket may be of such a size to fit nicely on binding Al3+ strongly than other ions, and the resulting rigidity imposed on the motif has compelled the excited state to deactivate the radiative process. Other metal ions may have some interaction with H-PNAP, but excited state may dissociate and becomes nonfluorescent. This may be the reason for the other metal ions do not induce fluorescence.

Anion Sensitivity of the [Al(PNAP)(NO3)2] Complex

The [Al(PNAP)(NO3)2] complex is highly emissive; hence, the reversibility of emission is checked by adding various anions like S2O32–, SCN–, PO43–, H2PO4–, HPO42–, I–, OAc–, ClO4–, SO42–, HSO4–, Cl–, F–, HF2–, NO3–, Br–, NO2–, CN–, N3–, AsO43–, and AsO2–. In pure water (pH 7.2, HEPES buffer), high-intensity fluorescence emission at 490 nm for [Al(PNAP)(NO3)2] selectively quenches in the presence of HF2–. F– and PO43– also show quenching of emission but much lower than that of HF2– (Figure S31). Hence, the in situ-generated [Al(PNAP)(NO3)2] complex may act as a secondary sensor to bifluoride ion (Scheme ). There are very rare reports of bifluoride ion sensors till date (Table S8).[55−57] Hence, our probe is helpful in HF2– detection which is very much needed in the present situation. For understanding the privileged mechanism of HF2– quenching interaction with [Al(PNAP)(NO3)2], the spectrofluorimetric titration has been performed by a gradual addition of HF2– solution (3 μM each) to the in situ-generated emissive 1:1 [Al(PNAP)(NO3)2] complex in aqueous solution (pH 7.2, HEPES buffer), and it is observed that the emission intensity gradually decreases (Figure S32).

Scheme 3

Ion Sensitivity of H-PNAP in Aqueous Medium

The UV–vis titration is also performed to authenticate the binding of HF2– with the [Al(PNAP)(NO3)2] complex (Figure S33). It shows that upon incremental addition of HF2– to the emissive complex [Al(PNAP)(NO3)2], the intensity of the band decreases with the generation of new bands. This observation may account for the chemical interaction of HF2– with the complex, [Al(PNAP)(NO3)2], and finally dechelate Al3+. HF2– interacts with the [Al(PNAP)(NO3)2] complex, and to prove the type of interaction, ESI-MS spectra were also collected by the addition of HF2– to the in situ-generated solution phase of the [Al(PNAP)(NO3)2] complex; the mass spectrum displays the intense peak at 292.0915 which corresponds to the free ligand, and after zooming the spectrum from 0 to 280, we get some peaks for various aluminum salts (Figure S34). Few other chelating reagents like the disodium salts of ethylenediaminetetraacetate (Na2EDTA), 8-hydroxyquinoline (Hoxin), dimethylglyoxime (H2DMG), 1, 3 diaminoguanidine (1,3 AGUA), and ethylene glycol (EGOH) have been added and the emission intensity is checked (Figure S35). It is observed that the high emission of [Al(PNAP)(NO3)2] also remains insensitive to them with a small decrement to Na2EDTA and Hoxin. Thus, selective quenching of an intense emission of [Al(PNAP)(NO3)2] by HF2– may be very useful for the identification of toxic bifluoride.

The proposed mechanism for the selective sensing of HF2– may be explained by Scheme S4. It shows that free pyridine N may interact through a hydrogen bond to bring HF2– that becomes closer to the Al3+ center in the complex, which may be responsible for the possible chemical interaction with Al3+ to get removed from the coordination site as the fluoro complex of Al(III). Hence quenches the emission of the [Al(PNAP)(NO3)2] complex. The LOD for HF2– (3σ method) is 15 nM (Figure S36). The binding constant Kd (HF2–), 1.3 × 104 M–1, is calculated by using the Benesi–Hildebrand plot (Figure S37).

Theoretical Evaluation of Spectroscopic Data

DFT computation is managed to obtain the optimized structure of H-PNAP and Al3+–H-PNAP [Al(PNAP)(OH)2] as a model complex (Figure S38). The single crystal X-ray crystallographic parameters are used for H-PNAP. The metric parameters obtained from the crystal structure and the theoretically determined structure are in good agreement with the X-ray structure, bond length (Å), calc. (expt): C=O 1.24 (1.23), C=N 1.30 (1.28), C–N 1.37 (1.34) N–N 1.37 (1.37) C–O 1.36 (1.34). For the interpretation of the theoretical facets of the observed spectroscopic responses of H-PNAP toward Al3+, TDDFT is also performed. The frontier molecular orbital images with the energy and orbital contributions of H-PNAP (Table S9, Figure S39) and [Al(PNAP)(OH)2] (Table S10, Figure S40) are shown. The possible electronic transitions for the probe and its Al complex are interpreted by TDDFT (Tables S11 and S12). Some selected bond lengths of the [Al(PNAP)(OH)2] complex obtained from the optimized structure are: bond length (Å): Al–OH (1.74), C=O (1.27), Al–N (2.06), Al–O (1.87), and Al–O (of C=O, 2.1). In the UV–vis spectrum in pure aqueous media, the probe shows two sharp bands at 326 and 366 nm which correspond to the highest occupied molecular orbital (HOMO) →lowest occupied molecular orbital (LUMO) + 1 and HOMO → LUMO transitions, respectively (obtained from DFT calculations), and also a broad band at 440 nm (Figure a); for its Al complex, a new band appears at 406 nm in the UV–vis spectra (Figure S22), which may be because of the HOMO → LUMO transition. The HOMO–LUMO gap in H-PNAP is 3.43 eV, and upon coordination with Al3+, this gap changes to 2.99 eV, which supports the shifting of λmax in the UV–vis absorption spectrum, and is represented by the schematic MO diagram (Figure b).

Figure 12

Probable electronic transitions and relationship between UV–vis experimental measurement and theoretical (TDDFT) calculation of (a) H-PNAP and (b) [Al(PNAP)(OH)2].

Applications: Recovery Study for Al3+ from Municipal Supplied Water Sample

To evaluate the reliability of the fluorescence detection method for Al3+, a recovery study of Al3+ from drinking water sample (Kolkata Municipal Supplied water in Jadavpur University Campus) has been performed by a standard addition method. The fluorescence emission intensity is measured at 490 nm upon excitation at 400 nm for the calibration plot by varying the concentration of Al3+ ion from 1 to 10 μM by maintaining the concentration of probe at 10 μM in pure H2O (HEPES buffer, pH 7.2) (Figure S41). From the calibration plot, an unknown concentration of Al3+ is measured (Table ). A satisfactory result for the recovery of Al3+ from drinking water sample is obtained. This testing has exhibited a pleasing, rapid, and inexpensive application of the probe, H-PNAP, for the Al3+ ion for on-demand water analysis.

Table 1

Recovery of Al3+ Ions in Drinking Water Samples Using the Probe H-PNAP

sample	added Al3+ (μM)	emission intensity, 490 nm (a.u.)	experimentally Found Al3+ (μM)	recovery (%)
drinking water	2	91	1.69	84.5
	4	157	3.69	92.2
	6	209	5.22	87.0

Paper Strip Detection Kits

The probe shows distinct emission in an aggregated state (9:1 H2O/MeOH) and in the presence of Al3+ (in pure water) at 540 and 490 nm, respectively. A potential application may be assisted by this single probe as an accessible transportable tool for sensing TNP/DNP and Al3+. We have prepared a test paper strip by dip-coating a solution of H-PNAP (10 μM) in 90% water in MeOH onto a Whatman 41 filter paper. The dried paper strip is soaked in 30 μM solution of TNP/DNP, dried, and then images are taken under a UV lamp at 365 nm. The green emission of the aggregated probe quenches in the presence of TNP/DNP (Figure ). For Al3+ sensing in paper strip, we prepared test paper strips by dip-coating a solution of H-PNAP (10 μM) in pure water on a Whatmann 41 filter paper, and then this paper strip was dipped in 10 μM of Al3+ solution, dried, and images were taken under a UV lamp (365 nm). Again, this Al3+-coated paper was dipped in a 20 μM solution of HF2–, and images were taken after drying (Figure ). In the presence of Al3+, the probe shows blue emission, but upon treatment with HF2–, this high emission is quenched.

Figure 13

Detection of TNP/DNP by the test strip method under UV lamp (365 nm): probe in pure MeOH (a), 9:1 H2O/MeOH (b), in the presence of DNP (c), and TNP (d). Detection of Al3+ and HF2– by the test strip method under UV lamp (365 nm): probe in H2O (e), in the presence of Al3+ (f) and Al3+ + HF2– (g).

Application in Logic Gate

A single molecule that has light sensitivity is very much beneficial for the construction of integrated logic gates like full-adder, half-adder, full-subtractor, half-subtractor, and INHIBIT.[58] The probe in pure methanol has no emission, but in 9:1 H2O/MeOH medium it has generated a new emission band at 540 nm. This highly emissive aggregated probe selectively quenches in the presence of TNP/DNP. Hence, the probe undergoes turn-on emission in the 9:1 H2O/MeOH medium and turns off in the presence of TNP or DNP. Therefore, with the two chemical inputs 9:1 H2O/MeOH and TNP/DNP, H-PNAP can develop a logic gate (INHIBIT). Not only can the probe construct an INHIBIT logic gate for inputs such as 9:1 H2O/MeOH and TNP/DNP, but it can also build another logic gate by the chemical inputs Al3+ and HF2– because in the presence of Al3+ in pure water the probe shows high emission at 490 nm, and this emission is selectively quenched by HF2– (Scheme ).

Scheme 4

Molecular Logic Gate (INHIBIT) Construction by Al3+–HF2– and 9:1 H2O/MeOH–(TNP/DNP)

Living Cell Imaging

The in vitro cytotoxicity of the probe, H-PNAP, is estimated for checking the biocompatibility of WI38 cell line. The cells were treated with five different concentrations (20, 40, 60, 80, and 100 μM/mL) of ligand for 24 h, followed by MTT assay (Figure ). It is observed that the ligand exhibited no significant toxicities even at the highest concentration of 100 μM. Therefore, the ligand shows good biocompatibility and is beneficial for biological applications.

Figure 14

Cell survivability study of WI38 cells exposed to H-PNAP.

The fluorescence microscopic study is performed to envisage the cellular uptake of the probe H-PNAP (5 μM) and Al3+ salt (10 μM). A prominent blue fluorescent signal is observed under the microscope. After the addition of HF2– salt (10 μM), the fluorescent signal immediately disappears (Figure ). From this observation, it is concluded that the cells readily uptake the probe H-PNAP and the Al3+ salt which results in a blue fluorescent signal, and the signal immediately quenches after the addition of the HF2– salt. Hence, H-PNAP is not only an ion sensor, but has multiple applications and has successfully been applied in the detection of intracellular Al3+ and HF2– in the MDA-MB-468 cell and shows better efficiency with respect to other reported ligands (Table S13).[59−69]

Figure 15

Bright-field, fluorescence, and merged microscopic images of untreated MDA-MB-468 (control); cells treated with H-PNAP (5 μM) + Al3+ (10 μM) and with H-PNAP (5 μM) + Al3+ (10 μM) + HF2– (10 μM).

Conclusions

An AIE-active fluorescence probe, (E)-N′-((2-hydroxynaphthalen-1-yl)methylene)picolinohydrazide (H-PNAP), has been characterized and used for sensing of NACs (TNP and DNP with LOD ∼0.8 and 0.9 μM, respectively) in a 90% water/MeOH medium. The aqueous solution of the probe is highly selective and sensitive to Al3+ (LOD, 2.09 nM) which serves as a secondary sensor to HF2– (LOD, 15 nM). For practical applicability of the probe (H-PNAP), a recovery study of Al3+ from drinking water sample has been done. H-PNAP builds the INHIBIT logic gate with two chemical inputs Al3+, HF2– and 90% H2O/MeOH, TNP/DNP. Moreover, H-PNAP has the capability to be visible, practical, and speedy to monitor Al3+ and nitroexplosives (TNP/DNP) by simple paper test strips. The probe H-PNAP is used to detect Al3+, and the resulting emissive complex [Al(PNAP)(NO3)2] identifies HF2– ions in live cells (MDA-MB-468).

Experimental Section

Materials and Methods

All organic and inorganic chemicals were purchased from Merck except 2-picolinic acid (High-Media) and used without further purification. Water for aqueous solutions was obtained from Millipore water (Milli-Q). Elemental analyses (C, H and N) were performed using a PerkinElmer 2400 Series-II CHN analyzer, USA elemental analyzer. The fluorescence spectra were recorded using a PerkinElmer spectrofluorimeter model LS55, UV–vis spectra were obtained from PerkinElmer Lambda 25 spectrophotometer, and the time-resolved single-photon counting measurements were carried out by using time-correlated single-photon counting setup from HORIBA Jobin-Yvon; Fourier transform infrared (FT-IR) spectra were recorded from a PerkinElmer LX-1 FT-IR spectrophotometer (KBr disk, 4000–400 cm–1). All the required NMR spectra were obtained from a Bruker (AC) 500 MHz FT-NMR spectrometer using trimethylsilane (TMS) as an internal standard. ESI mass spectra were recorded from a Water HRMS model XEVO-G2QTOF#YCA351 spectrometer. All of the measurements were carried out at room temperature.

Synthesis of H-PNAP

Picolinohydrazide was prepared from 2-picolinic acid following the reported procedure.[70] Picolinohydrazide (137 mg, 1.0 mmol) was added to 2-hydroxy-1-naphthaldehyde (172 mg, 1.0 mmol) in dry MeOH (15 mL) and stirred for 6 h, and a greenish yellow solution was obtained. Then, this solution was kept in air and allowed to evaporate slowly. After few days (2 weeks), yellow crystals of H-PNAP were obtained. Then, the crystals were collected from the solvent and dried in open air (yield, 89%), mp 186 °C (Scheme S1). Microanalytical data: C17H13N3O2 calcd (found): C, 70.09 (69.55); H, 4.50 (4.54); N, 14.42 (13.85) %. 1H NMR (500 MHz, DMSO-d6) (δ, ppm): 12.73 (s, 1H, −OH), 12.387 (s, 1H, −NH), 9.628 (s, 1H, imine-H), 8.573–8.582 (d, 1H), 7.976–8.010 (t, 2H), 7.899–7.917 (dd, 1H), 7.740–7.758 (d, 1H), 7.699–7.715 (d, 1H), 7.507–7.531 (dd, 1H), 7.404–7.434 (t, 1H), 7.205–7.234 (t, 1H), 7.034–7.052 (d, 1H) (Figure S1); 13C NMR (in DMSO-d6) (δ, ppm): 160.49, 158.63, 149.41, 149.15, 149.04, 138.65, 133.34, 132.33, 129.45, 128.27, 128.24, 127.49, 124.06, 123.30, 120.99, 119.43, 108.99 (Figure S2); The ESI-MS peak for H-PNAP at 292.1084 (calculated mass for H-PNAP + H+, 292.11) corresponds to [M + H+] and the base peak at 314.0985 for [M + Na+] (Figure S3); FT-IR for H-PNAP: ν 3482 cm–1 (OH), ν 3191 cm–1 (NH), ν 1661 cm–1(C=O), ν 1620 cm–1 (C=N) (Figure S4).

Single-Crystal X-ray Crystallographic Measurement of H-PNAP

Slow evaporation of MeOH solution had a crystallized yellow prism (0.145 × 0.089 × 0.045 mm3) of H-PNAP. The crystal data and data collection (Table S1) were carried out using an X-ray diffraction instrument of Bruker Smart APEX II CCD Area Detector at 296(2) K. Graphite-monochromatized Mo Kα radiation of wavelength 0.71073 Å was used in a fine-focus sealed tube in the hkl range −6 ≤ h ≤ 6; −29 ≤ k ≤ 29; −17 ≤ l ≤ 17 and of angular range 1.796 ≤ θ ≤ 27.101° for data collection. The intensity was corrected, and empirical absorption corrections were considered for Lorentz and polarization effects under the condition of I > 2σ(I). The direct method using successive Fourier and difference Fourier syntheses was employed for the structure solution of the crystal. All the nonhydrogen atoms were refined anisotropically and the hydrogen atoms were refined by a riding model. The hydrogen atoms were static geometrically. All these calculations were carried out using ORTEP-32,[71] SHELXL-97,[72] and PLATON-99,[73] programs. The crystal data of H-PNAP have been deposited to Cambridge Crystallographic Data Centre with CCDC number 1873384.

Theoretical Computation

For the optimization of the ligand and all the complexes, DFT/B3LYP method by Gaussian 09 software was used.[74] The basis set used for the probe H-PNAP was 6-311+G(d) and that for its Al3+ complex was LANL2DZ. Vibrational frequency calculations were performed for confirmation that the optimized geometries represent the local minima, and these only yielded positive eigenvalues. TDDFT was also performed by the use of CPCM;[75−77] from this, theoretical UV–vis spectral transitions are observed. The fraction of contributions of various groups in each molecular orbital was calculated by carrying out GAUSSSUM.[78]