Introduction

Much attention has been focused in recent years on the development of new chemical sensors for the recognition of heavy metal ions as well as anionic species [1]. Metal ions play a vital role in a wide range of chemical and biological processes [2], yet some of them are potentially toxic. It is indeed important to detect metal ions for applications in waste management and environmental toxicology [2]. Although zinc is only moderately abundant in nature, ranking twenty third of the elements in the Earth’s crust, it is, however, a omnipresent and indispensable element in the human body and the second most abundant transition metal after iron [3]. In total, the adult human body contains 2–3 g of zinc [4]. For a long period of time, zinc has been known as an essential trace element, acting as a structural component of many protein scaffolds (e.g., carbonic anhydrase and zinc finger protein) [3, 5]. A considerable amount of zinc at a concentration of several milli-molar is accumulated in the presynaptic neurons of the brain and subsequently released when the neurons are active [6]. Thus, the detection and separation of zinc in a wide range of concentrations from nano-to the milli-molar scale is of immense interest and significance [7]. Furthermore, the greatest and most important challenge for the detection of zinc appears from the interference of other transition metal ions, in particular cadmium ion. Even though a number of techniques like atomic absorption spectroscopy, inductively-coupled plasma mass spectrometry and inductively coupled plasma atomic emission spectroscopy are available for the detection of these metal ions in food samples, they require tedious procedures and are also very expensive [8]. Optical signals based on changes in absorbance or fluorescence is the most frequently applied technique because of the simplicity of the experimental methods. A great number of fluorescent sensors have been designed to detect different kinds of metal ions. For the fluorescent detection of ions; fluorescence enhancement “turn-on” is preferable to fluorescence quenching “turn-off”, because the former reduces the chance of false positive data by other fluorescent quenchers existing in samples. [9] Design of a fluorescent probe is generally based on intramolecular charge transfer (ICT) [10], photoinduced electron transfer (PET) [11], chelation-enhanced fluorescence (CHEF) [12], metal − ligand charge transfer (MLCT) [13], excimer/exciplex formation [14], imine isomerisation [15], intermolecular hydrogen bonding [16], excited-state intramolecular proton transfer [17], displacement approach [18], and fluorescence resonance energy transfer (FRET) [19]. In CHEF process, the electron pair is shared by coordination to a cation. As a result, the redox potential of the receptor is raised so that the highest occupied molecular orbital (HOMO) of the receptor becomes lower in energy than that of the fluorophore [20].

In recent time several chemosenors for zinc ions are reported in the literature [2123]. Hsieh et al. reported A simple Schiff base type fluorescent receptor which exhibits an ‘off–on-type’ mode with high selectivity in the presence of Zn2+ ion through the formation of a rigid structure [24]. Another Schiff’s base chemosensor is reported by Li et al. which displays specific recognition to Zn2+, and especially avoids the interference of Cd2+ when tested against a range of physiological and environmentally relevant metal ions [25]. Recently Kim et al. reported a simple receptor which could simultaneously detect three biologically important metal ions (Zn2+ Fe2+ and Cu2+) in aqueous solution. The sensor could function as a “turn-on” fluorescence receptor only to Zn2+ ions and could be successfully applied to the detection of intracellular Zn2+ ion [26].

In this paper, we present N-Furfurylsalicylaldimine (HL) as a highly selective fluorescent sensor for Zn2+ ion in CH3OH-H2O mixture by chelation-enhanced fluorescence (CHEF). The fluorescence properties and high selectivity of the ligand for Zn2+ ion over other possible competitive cations were investigated in detail. The X-ray single crystal structure of the Zn(II) complex with the ligand is reported.

Experimental

Materials and Physical Measurements

The ligand N-Furfurylsalicylaldimine is prepared as per the literature method [27]. Furfurylamine (Sigma Aldrich), salisaldehyde (Merck, India) and zinc acetate dihydrate (Merck, India) have been used as received. Solution of HL is prepared in methanol medium. Metal salts such as nitrate of Na+, Ag+, Ca2+, Sr2+, Ba2+, Co2+, Ni2+, Cu2+, Zn2+, Pb2+, Cd2+ and chloride salt of Hg2+ were purchased from Merck, India and have been used as received. Metal salt solutions were prepared in a phosphate buffer (0.1 M) with KH2PO4 (0.1 M) and Na2HPO4 (0.1 M) solution (pH 7; MeOH:H2O = 9:1, v/v). Buffer capsules were purchased from Merck India. Absorption and fluorescence spectra are recorded in Shimadzu 1601 spectrophotometer and Hitachi F-7000 spectrofluorimeter respectively.

Determination of the Binding Constant

The binding constant was calculated based on the titration curve of the HL with metal ions. Binding constant was determined by a nonlinear least squares fit of the data with the following equation as referenced elsewhere [30].

$$ \mathrm{y}=\mathrm{x}/2*\mathrm{a}*\mathrm{b}*{\left(1-\mathrm{x}\right)}^2+\mathrm{x}*\mathrm{b}/2 $$

Where ‘x’ is (I-Io)/(Imax-Io), ‘y’ is the concentration of metal ion, ‘a’ is the binding constant, and ‘b’ is the concentration of sample. Here (I-I0)/(Imax-I0) is the fluorescence intensity ratio.

Determination of Fluorescence Quantum Yield

Here, the quantum yield φ was measured by using the following equation,

$$ {\varphi}_{\mathrm{x}}={\varphi}_{\mathrm{s}}\left(\ {\mathrm{F}}_{\mathrm{x}}/{\mathrm{F}}_{\mathrm{s}}\right)\left(\ {\mathrm{A}}_{\mathrm{s}}/{\mathrm{A}}_{\mathrm{x}}\right)\left({{\mathrm{n}}_{\mathrm{x}}}^2/{{\mathrm{n}}_{\mathrm{s}}}^2\right) $$

Where, ‘x’ and ‘s’ indicate the unknown and standard solution respectively, φ = quantum yield,

F = area under the emission curve, A = absorbance at the excitation wave length, n = index of refraction of the solvent. Here φ measurements were performed using anthracene in ethanol as standard [φ = 0.27] (error ~ 10 %) [31].

Synthesis of [Zn(L)2]

Zinc(II) acetate.dihydrate (0.2915 g, 1 mmol) was dissolved in 10 ml methanol in a round-bottom flask. Methanolic solution (10 ml) of HL (0.2010 g, 1 mmol) was added drop-wise with constant stirring for about 20 min at room temperature. After that the whole reaction mixture was refluxed for 1 h to get a greenish-yellow solution. The solution was cooled to room temperature and filtered. The greenish-yellow filtrate was kept in air at room temperature for crystallization. After 2 weeks, pale yellow block shaped single crystals of diffraction quality were obtained on slow evaporation of the filtrate. Yield 70 %.

X-Ray Crystallography

Data collections of 1 was carried out at 120(2) K on an Oxford Diffraction Gemini Ultra diffractometer. Cell refinement, indexing and scaling of the data sets were done with CrysAlisPro package, Version 1.171.35.10 [32]. The structures were solved by using the olex2.solve solution program [33] using the charge flipping algorithm and refined by the full matrix least-squares method based on F2 with all observed reflections [34]. The crystallographic details are listed in table 1.

Results and Discussion

Emission Studies

At first we have investigated the fluorescence properties of the ligand HL and its fluorescence response to various metal cations in methanol-water solvent. The ligand itself shows negligible fluorescence at 443 nm upon excitation at 330 nm, however, fluorescence enhancement is observed when Zn2+ ion is added to the methanolic solution of the ligand. The native fluorescence of the ligand is minimal, but when Zn2+ is added incrementally, fluorescence increases and reaches a maximum on addition of 300 micro litres Zn2+ solution (concentration, 30 μM) at 443 nm (Fig. 1). The interaction of the ligand with Zn2+ ion resulted in fluorescence enhancement, which is attributed to a chelation-enhanced fluorescence (CHEF) effect. We believe that the zinc ion is effectively acting as an electrophile which attacks the electron-rich hydroxyl oxygen atom of the HL to form a rigid metal complex. The quantum yield of ligand HL remarkably changed from 0.003 to 0.321 on the formation of the complex with Zn2+ metal ion, which indicates 107 folds enhancement in quantum yield.

Fig. 1
figure 1

Fluorescence spectra of HL in the buffer solution (MeOH:H2O = 9 : 1, pH = 7) in the presence of increasing concentration of Zn(NO3)2. [HL] = 2.0 × 10−5 M and [Zn2+] = 2.0 × 10−4 M, Excitation at λ = 330 nm. Inset: The fluorescence intensity at 443 nm vs [Zn2+] μM

Ion selectivity is an important property of the fluorescence probes. Thus, we evaluated the fluorescent response of HL with different metal ions, including Ag+, Na+, Ca2+, Ba2+, Sr2+, Pb2+, Hg2+, Co2+, Ni2+, Cu2+ and Cd2+ in buffer solution and found the perfect selectivity for Zn2+ with a considerable signal output (Fig. 2). It is worth noting, that the ligand distinguishes Zn2+ from Cd2+, which might cause interference since both metal cations have very similar properties [28].

Fig. 2
figure 2

Fluorescence intensity changes (I − I0) of free HL (2 × 10−5 M) at 443 nm in the buffer solution (MeOH:H2 = 9 : 1, pH = 7) upon addition of various metal ions (2 × 10−4 M). I and I0 denote fluorescence intensity of HL in the presence and absence of metal ions. Excitation wavelength: 330 nm

The existence of other transition metal ions may affect the detection ability of HL. The Zn2+ selectivity and the fluorescence behaviour of HL remain uninterrupted with the presence of many other metal ions. Compared with some available Zn2+ sensors, which exhibit some enhancement of the fluorescence for Cd2+ [28], the ligand HL has a selective response to Zn2+ without the interference of Cd2+ in the buffered solution, whereas, the existence of copper obviously quenched the fluorescence intensity [28, 29]. (Fig. 3)

Fig. 3
figure 3

Fluorescence intensity profile of HL + Zn 2+and in the presence of various cations (7 equiv. Zn2+ and 3 equiv. other metal cations) in buffer at pH 7 at room temperature (excitation wave length 330 nm)

We have studied the binding ability of HL towards Zn2+ by fluorescence titration. A Job’s plot indicates the 1:2 coordination stoichiometry of Zn2+ with HL (Fig. 4), exhibiting a binding constant a = 4.7 × 1010 M−2, (R 2 = 0.9783) (Fig. 5) from the theoretical nonlinear square fit of experimental data to a 1:2 binding model. The obtained data indicate a typical CHEF effect, originating from the coordination of the de-protonated ligand (L) through the imine nitrogen and de-protonated phenolate oxygen atom (Scheme 1). The detection limit of the HL towards Zn2+ was found to be about 3.2 μM based on fluorescence titration experiments.

Fig. 4
figure 4

Job’s plot for determining the stoichiometry of HL and Zn2+. The total concentration ([L] + [Zn2+]) was 2.0 × 10−5 M. λex = 330 nm

Fig. 5
figure 5

Plot of [Zn2+]M−1 vs (I-I0)/(Imax-I0) for determination of binding constant. The red line is the nonlinear fitting curve obtained assuming a 2:1 association between HL and Zn2+. λex = 330 nm, [HL] = 2.0 × 10−5 M

Scheme 1
scheme 1

Formation of the complex i.e. [Zn(L)2] (1)

X-ray Crystallography

The crystal structure of the Zn2+ complex of the HL shows 1:2 binding and a distorted tetrahedral geometry around Zn2+ is identified. As seen in Fig. 6, an imine nitrogen atom (N9) from the receptor is at the apex of the tetrahedral zinc (Zn11) and the phenolate oxygen atom (O1) of the same receptor molecule is also coordinated to Zn(II). An imine nitrogen (N91) and a phenolate oxygen (O11) from another receptor moleculecompletes tetrahedral coordination. Thus the two receptor molecules are chelated with zinc atom. The N-Zn-N and O-Zn-O bond angles are 121.29° and 123.04° respectively. The bond lengths and the bond angles are listed in Table 2.

Fig. 6
figure 6

X-ray structure of [Zn(L2)] (1)

UV–Vis Spectroscopic Studies

We have also studied the change in absorption behaviour of the ligand upon binding with Zn2+ ion. As the concentration of Zn2+ was increased, the intensity of the absorption bands produced by the Zn2+-free ligand at 317 and 256 nm gradually decreased, accompanied by new absorption bands at 238, 272 and 363 nm produced by the Zn2+-bound ligand appearing at longer wavelength with gradual increase in the intensity (Fig. 7). The spectra obtained during the stepwise addition showed four isosbestic points indicates a clean conversion of HL into its corresponding Zn2+ complex. In addition Cu2+, Co2+ and Ni2+ show increased absorption bands at 238, 272 and 363 nm but the intensity is less than Zn2+ indicates that the ligand is sensitive for Zn2+ which is shown in bar graph representation (Fig. 8).

Fig. 7
figure 7

Absorption spectra of receptor in methanol with the increase of Zn(NO3)2. Inset: The absorbance at 363 nm vs [Zn2+] μM

Fig. 8
figure 8

Bar diagram represents the absorbance value of compound ([HL] = 2 × 10−5(M)) in presence of different metal ions ([M2+] = 2 × 10−4(M)) in H2O:MeOH (1:9; v/v, pH = 7) solution at 363 nm

Conclusion

In this paper we have studied a sensor for Zn2+ ion. The thorough investigation of the fluorescence sensing and binding properties of the ligand HL displays high selectivity luminescent probe for Zn2+. The increase in emission in the presence of Zn2+ ion is accounted for by the formation of a rigid metal-ligand complex. The X-ray crystal structure reveals that the Zn(II) complex is mononuclear with distorted tetrahedral geometry. An 107-fold increase in quantum yield is identified upon coordination of HL with Zn(II) ion is attributed as Zn2+-selective chelation-enhanced fluorescence response.