Introduction

The fluorescent method, due to its high selectivity and sensitivity, rapidity and convenient operation, has been employed in the detection of many kinds of analytes including various ions (Carter et al. 2014; Fu et al. 2019; Manjare et al. 2014; Öksüza et al. 2019; Özyol et al. 2018; Saçmaci et al. 2017), biothiols (Ren et al. 2018; Wang et al. 2018a, b), saccharide (Hosseinzadeh et al. 2015; Wang et al. 2015a, b) and protein (Chen et al. 2015; Huang et al. 2017). Aluminum, as the most abundant metal with its metallic forms in environment, has been widely used in human production activities including water treatment, package of food and medicines, and the production of light alloys (Kumawat et al. 2016; Shyamal et al. 2016). However, it is able to accumulate in environment and could affect human health through the biologic chain via water and organisms, and high levels of Al3+ may induce many physiological disorder and neurologic diseases, such as the Alzheimer’s disease (AD), Parkinson’s disease and so on (Good et al. 1992; Okem et al. 2015; Walton 2007). Moreover, the limit of concentration of Al3+ defined by the World Health Organization (WHO) in drinking water is 7.41 μM (Han et al. 2012). Hence, it is essential to develop highly selective and sensitive chemosensors for detecting the concentration of Al3+, to protect the ecological environment and health.

Over the past few decades, many researchers had put their best efforts into developing chemosensors for metal ions, and a lot of Al3+ chemosensors had been reported (Balakrishnan et al. 2017; Dai et al. 2018; Gan et al. 2017; Kang et al. 2016; Lim et al. 2018; Maniyazagan et al. 2018; Singh et al. 2017; Wang et al. 2017a, b, 2018a, b, c; Yıldız et al. 2017; Zeng et al. 2018; Zhang et al. 2018; Zhu et al. 2016). However, there are some obstacles to designing an excellent fluorescent chemosensor for Al3+. One is that Al3+ has drawbacks in its weak ability in coordination and strong ability in hydration; the other is that many reported probes suffer from shortcomings including multi-steps synthesis, being disturbed by other trivalent metal ions (Fe3+ and Cr3+) and insolubility in water which further limits their practical applicability in real samples. More importantly, compared with the fluorescent turn-on (off) probes, the development of ratiometric probe is more appealing because it can eliminate the interference caused by the physical or chemical method through the ratio of two intensities of absorption or emission wavelength (Gupta et al. 2017; Manna et al. 2017; Naskar et al. 2018; Qin and Yang 2015a, b). Hence, the development of ratiometric probes for Al3+ is of great importance for the application in complex samples.

Quinoline-based derivatives have proved to be a popular fluorophore in construction chemosensors due to their structural diversity, favorable photophysical properties and potent binding affinities to many metal ions (Li et al. 2016; Qin and Yang 2015a, b; Roy and Rajak 2017; Singh et al. 2018; Wang et al. 2015a, b). However, only a few of them are ratiometric probes for the detection of Al3+ (Roy et al. 2016; Zhu et al. 2016). In addition, Al3+, as a hard acid, prefers a hard-base coordination sphere containing N and O as the binding site (Fan et al. 2014). Taking the above statements into consideration, we designed a novel quinoline-based fluorescent probe N′-(2, 4-dihydroxybenzylidene)-2-(4-(2-oxo-2-(quinolin-8-ylamino)ethyl) piperazin-1-yl)acetohydrazide (BQPA), which showed a significant fluorescence turn-on (300-fold) and ratiometric absorbance for the determination of Al3+ in ethanol–water (1: 9, v/v) medium. Furthermore, BQPA was evaluated for the detection of Al3 in real samples.

Experimental

Materials

All chemicals and reagents were purchased from commercial sources. The fluorescent spectra were recorded at room temperature on a Perkin Elmer LS55 fluorescence spectrometer. UV–Vis absorption spectra were obtained using a Pgeneral TU-1901 UV–Vis spectrophotometer. 1H NMR and 13C NMR spectra were recorded on a Bruk AV-600 spectrometer, respectively. Mass spectra were measured on a Waters Xevo UPLC/G2-SQ Tof MS spectrometer.

Preparation of stock solution

A stock solution of BQPA (10 μM) was prepared with the solution of ethanol–water (1: 9, v/v).

Stock solutions (10 mM) of the cationic salts (Na+, K+, Ca2+, Mg2+, Ba2+, Cr3+, Mn2+, Fe2+, Fe3+, Cu2+, Ag+, Co2+, Ni2+, Zn2+, Cd2+, Al3+ and Pb2+) were prepared with ultrapure water, respectively. For spectrum measurement, the test solutions were prepared by adding a certain amount of stock solution using a pipette into the BQPA stock solution. During the measurement of fluorescent spectrum, the excitation was set at 350 nm, emission wavelength was recorded in the range of 300–600 nm, and emission intensity was recorded at 420 nm for the fluorescence titration, competitive experiments and Job’s plot. The excitation and emission slit widths were 10 nm and 10 nm, respectively.

Al3+ standard solution (10 mM) was prepared by dissolving Al(NO3)3·9H2O (0.5 mmol) in ultrapure water (50 mL).

Fluorescence and colorimetric measurements

Al(NO3)3·9H2O (0.5 mmol) was dissolved in ultrapure water (50 mL). 2–50 μL of the Al3+ solution (10 mM) was transferred to BQPA solution (10 μM) prepared above, respectively. After mixing them for a few seconds, fluorescence spectra and UV–Vis absorption spectra were taken at room temperature, respectively. The color changes of BQPA (10 μM) were obtained in the presence of Al3+ ions (5 equiv.) in ethanol–water (1:9, v/v) under UV light of 365 nm.

Determination of binding constant and detection limit

According to the fluorescence intensity data, the binding constant of BQPA with Al3+ was calculated based on the modified Benesi–Hildebrand equation (Kadar et al. 2005), where, Fmax,F and Fmin are the fluorescence intensities of BQPA in the presence of Al3+ at saturation, at an intermediate Al3+ concentration, and absence of Al3+, respectively. K is the stability constant.

$$ \frac{1}{{F - F_{ \text{min} } }} = \frac{1}{{K(F_{ \text{max} } - F_{ \text{min} } )[{\text{Al}}^{{ 3 { + }}} ]}} - \frac{1}{{F_{ \text{max} } - F_{ \text{min} } }}. $$
(1)

The limit of detection (LOD) of Al3+ was calculated on the basis of 3σ/S according to the fluorescence changes, where σ is the standard deviation of the blank solution and S is the slope of the calibration curve (Liu et al. 2016).

The limit of quantification (LOQ) of Al3+ was calculated on the basis of 10 σ/S according to the fluorescence changes, where σ is the standard deviation of the blank solution and S is the slope of the calibration curve (Chitnis and Akhlaghi 2008).

The quantification limit for Al3+ in water samples was spiked with standard Al3+ ions at different concentration levels, then diluted within working linear range, and analyzed with the method proposed under optimized conditions.

NMR titration

In three NMR tubes, BQPA (5 mg) dissolved in DMSO (0.5 mL) was added. Then different equivalents (0, 0.5 and 1 equiv.) of Al(NO3)3 in DMSO (0.5 mL) were added separately to the corresponding tube, and 1H NMR spectra were measured in turn at room temperature.

Synthesis of compound BQPA

Compound BQPA was synthesized according to the synthetic route outlined in Scheme 1. The intermediate compounds 1–4 were prepared by the reported literature (Shao 2010; Wang et al. 2017a, b; Li et al. 2018).

Scheme 1
scheme 1

The synthetic route of BQPA

Full size image

A mixture of compound 4 (102 mg, 0.3 mmol) and 2, 4-hydroxybenzaldehyde (0.4 mmol) dissolved in ethanol (20 mL) was refluxed for 2 h (monitored by TLC) and then cooled to room temperature. The solvent was removed under reduced pressure, and the crude substance was further purified by column chromatography using CH3OH /CH2Cl2 (v/v, 1/30) as eluent to get light yellow crystal BQPA (30 mg, 0.06 mmol), yield: 21.6%, m.p: 290–292 °C. 1H NMR (600 MHz, DMSO-d6) (Fig. S1): δ (ppm) 11.41 (s, 1H), 11.38 (s, 1H), 11.28 (s, 1H), 9.94 (s, 1H), 8.95 (m, 1H), 8.65 (dd, J1 = 7.2 Hz, J2 = 1.2 Hz, 1H), 8.41 (m, 2H), 7.67(d, J = 7.8 Hz, 1H), 7.58–7.62 (m, 2H), 7.26 (d, J = 8.4 Hz, 1H), 6.31–6.36 (m, 2H), 3.34 (s, 2H), 3.19 (s, 2H), 2.69 (s, 8H). 13C NMR (151 MHz, DMSO-d6) (Fig. S2): δ (ppm) 169.11, 165.44, 161.34, 160.00, 149.72, 149.24, 138.49, 137.15, 134.61, 131.93, 128.44, 127.59, 122.76, 122.04, 115.94, 110.65, 107.97, 102.89, 62.25, 60.66, 53.38, 52.92. HRMS m/z (TOF MS ES+) (Fig. S3): calcd for C24H27N6O4: 463.2094 [M+H]+, found: 463.2097.

Results and discussion

Fluorescence spectra response of BQPA to ions

The selectivity of BQPA to various metal ions was examined in ethanol/water (1: 9, v/v). As shown in Fig. 1, there was almost no change in the fluorescence spectrum between the free BQPA and the mixture of BQPA with the tested metal ions (Na+, K+, Ca2+, Mg2+, Ba2+, Cr3+, Mn2+, Fe2+, Fe3+, Cu2+, Ag+, Co2+, Ni2+, Zn2+, Cd2+ and Pb2+). However, a significant enhancement (300-fold) in fluorescence intensity was observed upon the addition of Al3+ centered at 420 nm, and the color of the BQPA solution changed from colorless into blue under the irradiation of 365 nm UV-lamp. This result could be attributed to the inhibition of the C=N bond rotation and chelation-enhanced fluorescence (CHEF) (Nie et al. 2017; Pang et al. 2018).

Fig. 1
figure 1

Fluorescence spectra of BQPA (10 μM) with different metal ions (50 μM) in ethanol/water (1:9, v/v)

Full size image

To further investigate the binding property of probe BQPA, fluorescence titration of probe BQPA with Al3+ was performed (Fig. 2). Upon excitation at 350 nm, probe BQPA alone exhibited a negligible emission at 420 nm. However, with the addition of Al3+ (0–5.0 equiv.) to a solution of probe BQPA in ethanol/water (1:9, v/v), a gradual increase in emission intensity of BQPA at 420 nm was observed and then almost reached a plateau when the addition of Al3+ was 20 μM (2 equiv.), indicating a stable complex formation between the probe BQPA and Al3+ (Wen and Fan 2016). Moreover, the maximal emission intensity of BQPA and the concentration of Al3+ (varying from 2 to 12 μM) showed a good linear relationship (y = 143.0x − 37.0578) (Fig. S4), and the detection limit of BQPA for Al3+ was calculated as 4.15 × 10−8 M (41.5 nM) (Liu et al. 2016), which met the requirement defined by the US Environmental Protection Agency (maximum allowed contamination of Al3+ is 7.4 μM in drinking water) (Han et al. 2012). Moreover, the limit of quantification (LOQ) of BQPA for Al3+ was calculated as 1.38 × 10−7 M (0.138 μM). These results indicated that BQPA could be used for the detection of Al3+ in drinking water.

Fig. 2
figure 2

Fluorescence emission spectra of BQPA in ethanol/water (1: 9, v/v) solution upon the addition of Al3+ (0–5.0 equiv.) with an excitation of 350 nm. Inset shows the fluorescence change at 420 nm with the addition of Al3+

Full size image

To further visualize the fluorescence color changes with the addition of Al3+ to BQPA, we drew coordinates of BQPA and the BQPA–Al3+ complexes in the CIE diagram, as shown in Fig. 3. Upon the addition of various concentrations Al3+, the coordinate changed from (0.25, 0.34) to (0.16, 0.04), which indicated the change of color from light blue to dark blue.

Fig. 3
figure 3

The distribution of BQPA and BQPA with different equiv. Al3+ (0.1, 0.5, 5.0) in CIE

Full size image

To investigate the anti-interference from the co-existing metals during the detection of Al3+, competitive experiments in the presence of other metal ions (Na+, K+, Ca2+, Mg2+, Ba2+, Cr3+, Mn2+, Fe2+, Fe3+, Cu2+, Ag+, Co2+, Ni2+, Zn2+, Cd2+ and Pb2+) were conducted. As shown in Fig. 4, the fluorescence intensity of the BQPA–Al3+ complex had no obvious variation upon the addition of competitive metal ions except Cu2+, which caused the fluorescence of the BQPA–Al3+ complex to be completely quenched due to its paramagnetic property reported by other researchers (More and Shankarling 2017).

Fig. 4
figure 4

Competitive selectivity of compound BQPA toward Al3+ over other metal ions (5 equiv.) in ethanol/water (1: 9, v/v)

Full size image

UV–Vis response of BQPA toward Al3+

According to the specific fluorescence responses of BQPA to Al3+ among the tested cations, the UV–Vis spectrum of BQPA (10 μM) were measured in the absence and presence of Al3+ (50 μM) in ethanol/H2O (1/9, v/v) solution, respectively (Fig. 5). BQPA alone showed the maximum absorption centered at 295 nm and 317 nm, respectively. Importantly, a significant redshift (from 295 to 305 nm) was observed upon the addition of Al3+. This result could be attributed to the enhancement of the conjugate system after the complexation of BQPA with Al3+ (Zhou et al. 2018).

Fig. 5
figure 5

UV–Vis absorption spectra of BQPA (10 μM) in the absence (black curve) and presence (red curve) of 5 equiv. of Al3 in ethanol/water (1: 9, v/v)

Full size image

Moreover, UV–Vis titration experiments were further carried out to investigate the binding properties of BQPA with Al3+ (Fig. 6). On gradual addition of Al3+ (0–2 equiv.) to the solution of BQPA (Fig. 6a), an isosbestic point at 304 nm was observed, indicating the formation of the BQPA–Al3+ adduct. In addition, the absorbance intensity ratios of the BQPA at 317 nm and 295 nm (A317/A295) decreased gradually on the addition of Al3+ (Fig. 6b), and a good linear relationship (y = − 0.2858 + 1.1745) could be seen between the absorbance intensity ratios (A317/A295) and the amount of Al3+ in the range of 4–9 μM (Fig. S5), indicating the practicability of the probe BQPA with the ability of ratiometric absorbance detection toward Al3+. The detection limit and quantification limit were calculated to be as low as 31 nM and 103 nM, respectively, which were lower than the ones calculated above based on fluorescence titration. According to the above results, as for the probe BQPA, the ratiometric absorbance detection was more sensitivity than that by fluorescence intensity in the detection of Al3+.

Fig. 6
figure 6

a UV–Vis absorption spectra of BQPA (10 μM) in ethanol/water (1:9, v/v) upon the addition of Al3+ (0–2 equiv.); b the scatter plot of absorbance ratios (A317/A295) and Al3+ concentration

Full size image

Analysis sensing mechanism of BQPA for Al3+

Confirmation of binding stoichiometry

To determine the binding stoichiometry of BQPA with Al3+, both mole–ratio plot and Job’s plot were measured. According to the results of fluorescence titration (Fig. 2, insert) and UV–Vis titration (Fig. 6b) of probe BQPA with Al3+, both the fluorescence intensity and absorbance ratios (A317/A295) almost remained constant when the addition of Al3+ was less than 1.5 equiv., which could be used as a preliminary proof for the 1:1 stoichiometry between BQPA and Al3+. Moreover, the 1:1 binding stoichiometry was further identified by the Job’s plot on the basis of its fluorescence intensity at 420 nm (Fig. 7). The maximum fluorescence intensity was seen when the mole fraction of Al3+ was around 0.5, confirming a 1:1 stoichiometry for BQPA–Al3+ complexes in ethanol/water (1:9, v/v) solution.

Fig. 7
figure 7

Job’s plot for determining the stoichiometry between BQPA and Al3+ ethanol/water (1:9, v/v) solution; the total concentration of BQPA and Al3+ was 10 μM)

Full size image

Furthermore, HRMS of BQPA in the presence of Al3+ was measured and the result is shown in Fig. 8. The peaks at 487.1680 were attributed to [BQPA − 2H+ + Al3+]+ (Calcd. m/z 487.1674), which was another credible evidence for the 1:1 binding stoichiometry of BQPA with Al3+ resulting from Job’s plot analysis. So, the association constant of BQPA and Al3+ was calculated as 7.48 × 105 M−1 (Fig. S6) according to the nonlinear curve fitting of the absorbance titration data, which is higher than the calculated 7.4 × 104 M−1 based on the fluorescence titration data (Fig. S7).

Fig. 8
figure 8

ESI-MS spectrum of BQPA (50 μM) on addition of 5 equiv. of Al3+ in CH3CH2OH

Full size image

H NMR titration

To get insight into the exact binding mode of BQPA–Al3+, 1H NMR titration experiments were measured in DMSO-d6. As shown in Fig. 9, on addition of Al3+, the protons of hydroxyl (Hb) and amide (Hd) groups of BQPA disappeared, indicating deprotonation during the process of coordination of Al3+ with the oxygen atom of the hydroxyl and the nitrogen atom of amide. Moreover, the protons of methylene (He and Hh) and the protons (Hf and Hg) of the piperazine ring of BQPA were all significantly shifted downfield to 3.49 ppm, indicating that the two nitrogen atoms acted as the binding sites involved in the coordination with Al3+.

Fig. 9
figure 9

1H NMR spectra of BQPA with Al3+ in DMSO-d6

Full size image

FT-IR measurement

To obtain more details for the binding site of BQPA with Al3+, the FT-IR spectra of free BQPA and complex BQPA–Al3+ were measured (Fig. S8). BQPA exhibited strong absorbance at 3260 cm−1 and 3240 cm−1, assignable to the stretching vibration of –OH and –NH, respectively. However, all of them disappeared in the spectra of complex BQPA–Al3+, and two wide peaks at 3371 cm−1 and 3146 cm−1 indicated the –OH and –NH stretching vibrations, respectively. Moreover, the peaks at 1686 cm−1 and 1506 cm−1, which belonged to the C=O and C–N stretching vibration of BQPA, shifted to the 1594 cm−1 and 1373 cm−1, respectively. The above results indicated the coordination of BQPA with Al3+. In addition, the comparison of probe BQPA with the reported chemosensors is summarized in Table 1. Compared with the reported probes, the advantages of probe BQPA was its lower detection limit (nM level) and high sensitivity through fluorescent signal response (300-fold), but its insolubility in neat water was its shortcoming, which might to some extent limit its application in environment and in vivo.

Table 1 Comparison of BQPA with previously reported probes for Al 3+ ions
Full size table

Hence, according to the analysis of the experimental results mentioned above, a feasible bonding mode between BQPA and Al3+ was proposed (Scheme 2).

Scheme 2
scheme 2

Proposed binding mode of BQPA with Al3+

Full size image

The effect of pH on the fluorescence of BQPA

The effect of pH ranging from 2.0 to 12.0 on the emission intensity (λem = 420 nm) of BQPA in the absence and presence of Al3+ was investigated (Fig. 10). The probe BQPA had almost no fluorescence emission in the wide pH range 2.0–12.0, but upon addition of Al3+ into BQPA at different pH conditions, obvious fluorescence enhancement was observed from pH 4–6, indicating that BQPA could be a good probe for Al3+ detection in acidic medium.

Fig. 10
figure 10

The variation in fluorescence intensity with the pH of BQPA in the presence of Al3+ (5.0 eq.) at 420 nm

Full size image

To understand the effect of pH on the detection of BQPA for Al3+ in acidic conditions, we detected the variation of pH with increase in the concentration of Al3+, and the results were recorded by the double coordinate graph according to the fluorescence intensity (λem = 420 nm) and pH with various concentrations of Al3+ in ethanol–water (1:9, v/v) (Fig. 11). Within the 0–1.0 equiv. of Al3+ content, both pH and fluorescence intensity (λem = 420 nm) had good linear relationship with the concentration of Al3+, which indicated that BQPA could detect Al3+ qualitatively and quantitatively by pH and fluorescence intensity under acidic conditions.

Fig. 11
figure 11

The relationship of fluorescence intensity (black, λem = 420 nm) and pH (red) with different concentrations of Al3+ in ethanol/water (1:9, v/v)

Full size image

Reversibility of BQPA for Al3+

To evaluate its practical application, the reversibility of BQPA was necessarily investigated in ethanol/water (1:9, v/v) solution by adding Na2EDTA, which is a good chelating agent, to Al3+. On addition of Al3+ to the solution of BQPA, both the UV–Vis absorption spectra (Fig. S9) and fluorescence spectra (Fig. S10) showed significant changes compared with the corresponding spectrum of BQPA itself. However, after addition of EDTA to the solutions of BQPA–Al3+, the UV–Vis absorption spectra and fluorescence spectra of the solution of BQPA–Al3+ showed much more similarity to that of BQPA in the absence of Al3+, indicating the recovery of BQPA. This result was also supported by the bonding constant of BQPA with Al3+ calculated as 7.4 × 104 M−1, which was far lower than that of EDTA with Al3+ calculated as 1.99 × 1016 M−1.

Application of BQPA to water samples

To explore the practical application of BQPA for the detection of aluminum ions, detailed experiments were carried out for the determination of Al3+ in real water samples collected from our university campus (Table 2). The results showed that BQPA had high accuracy for the practical application of aluminum ions in water. In addition, to investigate about the co-existence of Cu2+ during the detection of Al3+ in real water samples, different concentrations of Cu2+ were added and the corresponding fluorescence spectrum tested (Fig. S11–13). The results showed that when the concentration of Cu2+ was lower than 0.2 μM, the accuracy in the detection of Al3+ was almost unaffected. This result indicated that BQPA could be used for Al3+ detection in water samples even with the co-existence of trace Cu2+.

Table 2 Determination of BQPA for Al3+ in tap water and ultrapure water
Full size table

Conclusions

In summary, a novel fluorescent chemosensor BQPA was synthesized and characterized based on quinoline derivative. It showed significant fluorescence enhancement (300-fold) accompanied by color change from colorless to blue, and ratiometric absorbance for the highly sensitive detection of Al3+ in ethanol/water (1:9, v/v). The 1:1 stoichiometry of the BQPA–Al3+ complexes was determined, and the detection limits of probe BQPA for Al3+ was 31 nM, which was sensitive enough to detect Al3+ in water samples.