Introduction

Metal ions play an important role in numerous physiological processes and have a vital important effect on environment (Grabchev and Guittonneau 2006). The concentration of metal ions required by the organism needs to be maintained or below a certain concentration. Once overdosed, it will be detrimental to a living creature (Jang et al. 2013). The damage caused by heavy metal ions to the environment is a serious problem that must be effectively resolved to remedy the polluted environment as well (Yayintas et al. 2007). Many metal ions, such as Zn2+, within a reasonable range, are essential element in man and take an extraordinary important part in a series of physiological processes. For example, human growth and development, reproductive inheritance, immunity, endocrine, and so on are all related with Zn2+ (Martin and Tortell 2008). However, once the concentration reaches an amount that above the standard, it becomes toxic to aquatic organisms, man, and the natural environment (Mahapatra et al. 2011). Cd2+ is known as an extremely poisonous heavy metal ion (Xiao-Liang et al. 2008). When Cd2+ accumulates in environment, it can be enriched in the organism, causing chronic poisoning through the food chain. After absorbed, it forms cadmium–sulfur protein in the body and selectively accumulates in the liver and kidney. It can also cause the metabolism of bones to be blocked, resulting in a series of symptoms such as osteoporosis, atrophy, and deformation (Guo et al. 2010). Therefore, it is urgent to detect and remove toxic metal ions.

There are a number of methods to monitor metal ions. Among them, instrumental analysis methods are the most accurate and reliable methods. Typical instrumental methods are atomic absorption spectroscopy (AAS) (Ghaedi et al. 2007; Mahmoud et al. 2010), ion-coupled plasma mass spectrometry (ICP-MS) (Ferrarello et al. 2002; Su et al. 2010), atomic fluorescence spectrometry (AFS) (Fang et al. 1992; Trujillo et al. 2013), and high-performance liquid chromatography (HPLC) (An-Na et al. 2004; Ying et al. 2011). However, a majority of these instruments are often expensive and they need to be maintained by professional personnel. Furthermore, sometimes the cost of using instrumental analysis is too high (Wang et al. 2012b). Considering this occasion, reagents that can effectively detect heavy metal ions should be designed.

Among the various optical reagents and techniques, fluorescent Chemosensors exhibit many merits, including but not limited to high sensitivity, operation convenient, and response swift (Kyung-Sik et al. 2008). Several studies have been developed on the synthesis and applications of fluorescent chemosensors. Wang and his group (Wang et al. 2015) have designed a novel phenol-based BODIPY chemosensor to monitor Fe3+. Their chemosensor shows obvious fluorescence enhancement and significant blue-shifts when Fe3+ is present. However, the probe exhibited weak ability to remove metal ions. Wang’s group have done much excellent work on sensing of metal ions and cell imaging (Zhou et al. 2018; Liu et al. 2018, 2019; Zheng et al. 2017; Luo et al. 2016). For example, they have designed a reversible Schiff’s base fluorescence probe QC for Al3+ based on quinoline derivative, which can not only detect trace Al3+ in tap water and human serum, but also use for fluorescence imaging of Al3+ in Hela cells. In our previous work, a new chemosensor for detection of Fe3+ and Hg2+ with highly selectivity was designed and synthesized (Jia and Cheng 2016). Fe3+ and Hg2+ can be detected as low as 0.5 mM with the chemosensor. However, the synthesizing procedure is a little bit complicated. Asiri et al. (2018) have designed a highly sensitivity turn-on fluorescent chemosensor for the detection of Al3+ ions. However, it is not widely used in real life. Shiraishi et al. (2007) have synthesized a new simple-structured fluorescent chemosensor which is based on a diethylenetriamine bearing two end quinolone fragments. It could be used in quantitative detection of Zn2+.

Usually, small molecular sensors can detect metal ions sensitively, but cannot effectively remove metals ions. This problem can be solved by converting small molecules into macromolecules. Moreover, after the small chemical sensor is polymerized into a conjugated polymer, metal ions can be easily removed by absorption and centrifugation. Because of the so-called ‘‘molecule wire’’ effect, they often show high sensitivity due to the fluorescence enhancement. Sun et al. (2018) have also reported a new sensor based on BODIPY. They used the small molecular sensors that were changed obviously in the presence of Hg2+ to synthesize macromolecular fluorescent chemosensors. The minimum detection limits of the macromolecular reached 0.15 mM. However, the synthetic routes are not simple enough. Zhang et al. (2016) synthesized a new fluorescence chemosensor through Suzuki coupling reaction using 5-(N-Ethylcarbazol-3-yl)thiophene-2-carbaldehyde (ECTC). This compound strongly quenches fluorescence by Fe3+ ions. Under optimal conditions, the fluorescence intensity of ECTC is quenched when interacting with Fe3+ ion. The limit of detection is below 0.1 μm. A new turn-on chemosensor with a bis-quinoline unit for selective detection of Cd2+ has been reported by Hrishikesan et al. (2016). Combined with a stoichiometric ratio of 1:1, the association constant was calculated to be 1.68 × 105 mol L−1.

In this work, a small molecular fluorescent chemosensor TPB and macromolecular fluorescent chemosensor PTPB have been designed and synthesized via Suzuki coupling reaction. They exhibit high selectivity and sensitivity to Cd2+ and Zn2+ ions. Obvious fluorescence shifts can be observed when contact with Cd2+ and Zn2+ for PTPB. Compared with these highly sensitive fluorescent Chemosensors (Aragoni et al. 2013; Ponnuvel et al. 2016; Sarkar et al. 2015) that detect the same ions, our fluorescent chemosensors are less sensitive than them, but the synthesis methods and procedures of our fluorescent chemosensors are simpler and more convenient. Furthermore, PTPB can be used to absorb ions for its excellent character than these probes (Hennrich et al. 1999; Sil et al. 2016; Wang et al. 2012a).

Experimental

Materials

Benzene-1,3,5-triyltriboronic acid (BTA), 2-bromo-1,10-phenanthroline (BPT), [1,1′-bis(diphenylphosphino)ferrocene]dichloropalladium(II) [Pd(dppf)Cl2], and 3,8-dibromo-1,10-phenanthroline (DBPT) were all purchased from J&K China Chemical Ltd.

Tetrahydrofuran (THF), N, N-dimethylformamide (DMF), and glacial acetic acid were all purchased from Aladdin Chemical Co. Ltd. All the metal salts (Cu2+, Hg2+, Ni2+, Mn2+, Co2+, Cd2+, Pb2+, Fe2+, Ag+, Mg2+, Ca2+, Zn2+, Fe3+, Cr3+, Na+, K+, and Al3+) were purchased from Aladdin Chemical Co. Ltd. Deionized water was used by the experiments.

Synthesis of TPB and PTPB

In this work, a small molecular sensor TPB and a macromolecular sensor PTPB were designed and synthesized via Suzuki reaction. The synthetic routes are described below and are reflected in Scheme 1.

Scheme 1
scheme 1

Synthetic routes of TPB and PTPB

Full size image

TPB was synthesized using the following procedure. At room temperature, BTA (0.0131 g) and 2-bromo-1, 10-phenanthroline (0.020 g) were dissolved in 6 mL THF and then ultrasound for 5 min. The mixture was stirred for 1 h under nitrogen atmosphere and aqueous K2CO3 solution (6 mL) was then added. When the temperature reaches 65 °C, Pd (dppf) Cl2 (0.0020 g) in 3 mL of THF was then added in it, and the reaction system was continuously stirred for 24 h. The solid–liquid mixture was centrifuged to get a crude product. The mixed solution of ethyl acetate and petroleum ether (EA: PE = 2: 1) and deionized water were used consequently to wash the crude product to obtain a pale yellow solid (0.007 g, 11%). The small molecule fluorescent probe is synthesized and the yield was found only 11% around. MS of TPB calculated for C42H24N6 [M]+ 612.206, found 612.1103. The mass spectrum can be seen in Fig. 1a. 1H NMR analysis are indicating in the following (DMSO, 400 MHz): δ (ppm): 9.39–9.00 (d, 6H), 8.79–8.36 (d, 9H), 8.12–7.62 (d, 9H). Figure 2a reflects the 1H NMR spectra of TPB.

Fig. 1
figure 1

MS spectrum of TPB and PTPB

Full size image
Fig. 2
figure 2

1H NMR and FTIR spectra of TPB and PTPB

Full size image

Using the analogous method what has been mentioned above, the brown solid PTPB was acquired (0.011 g, 22%). MS of PTPB is 2200. The mass spectrum can be seen in Fig. 1d. 1H NMR (DMSO, 400 MHz): δ (ppm): 9.12 (s, 6H), 8.82 (s, 9H), and 7.99 (s, 9H). Figure 2b reflects the 1H NMR spectra of PTPB.

Characterization

1H-NMR data were obtained by AVANCE III 600 MHz (Bruker). FTIR (4000–500 cm−1) was operated on the Nicolet NEXUS 470 spectrometer. Fluorescence spectrum was shown on Shimadzu F-4500 fluorescence spectrophotometer. For UV–Vis measurement, large numbers of samples were studied on a UV-5900PC spectrophotometer (METASH).

Results and discussion

In our work, the solvent, catalyst, and feed ratio have been found that have influence on the synthesis results when synthesizing small molecule and macromolecular chemosensors. Based on this situation, we have studied these influencing factors to explore the best reaction conditions.

Solvent effect

To explore the effect of solvent on the results, toluene, dimethyl formamide, diethylene dioxide, and water were selected besides tetrahydrofuran/water (3:2) solvent mixtures. Ensure that the other conditions in reaction remain the same, only change the solvent. The Suzuki coupling reaction conducted in tetrahydrofuran/water co-solvent systems generally give expected product. According to the results in Table 1, tetrahydrofuran/H2O (3:2) is found to be the best solvent.

Table 1 Solvent effect on coupling reaction
Full size table

Catalyst and feed ratio influence

The other conditions in coupling reactions remain the same, but using the different catalyst to investigate the catalyst influence. According to the reaction exploring, the results changed when using the different kinds of catalysts. The best reaction result was obtained when using the Pd (dppf) Cl2 as the catalyst.

When synthesizing a macromolecular network, from the effect of feed ratio investigations, all polymer compounds show their molecular weights already reached macromolecule levels, although the degree of polymerization is not too high. It can be found that the molecular weight of the polymer compound PTPB and the number of the fluorescent units could be altered when regulating the feed ratio of DBPT and BTA. The observation of BTA:DBPT = 1.05:1 proved to be the maximum molecular weight. We also found that when the molecular weight reaches 3000 and above, the macromolecule is only slightly soluble in DMF and insoluble in other solvents. The results are given in Tables 2 and 3.

Table 2 Catalyst influence on coupling reaction
Full size table
Table 3 Molecular weight of polymers
Full size table

Structural characterization of small molecular and macromolecular chemosensors

The FTIR spectra of TPB and PTPB are revealed in Fig. 1c. The peak of the –C–H at 3062 cm−1 and the vibration of the benzene ring of 1400–1650 cm−1 in the TPB could be easily seen. The peak at 3059 cm−1 of –C–H in PTPB could also be observed. And the –C=N– peaks could be seen at 1593 cm−1 in spectra of TPB and 1592 cm−1 in spectra of PTPB, respectively. From the FTIR spectra of the compounds, we can infer that both compounds have been synthesized for the characteristic peaks shown as expected.

Response of TPB and PTPB to pH

The Britton–Robinson buffer solution was prepared when performing the pH response of sensors test. TPB was dissolved in a DMF-buffer mixed solution at first (DMF/buffer, v/v = 1/1). Then, measuring the small molecular sensor TPB’s absorbance at different pH values, the pH stability of the TPB is shown in Fig. 3. Figure 3 indicates that in the pH range from 1.89 to 10.38, the small molecular sensor TPB did not have any obvious red shift. However, when the pH reached 10.88, obvious red shift could be seen, showing its stability in a relatively wide pH range.

Fig. 3
figure 3

UV–Vis spectra of TPB (30 µM) and PTPB (30 µM) recorded at different pH values

Full size image

The macromolecular sensor PTPB was also performed to pH-response studies. Unlike its low-molecular-weight precursors TPB, in the pH range of 1.89–11.98, the macromolecular sensor did not display any obvious change, but show obvious change when the pH reached 12.00, indicating its availability over a wider pH range.

Selectivity of chemosensors to metal ions

To examine the obtained compounds’ sensing properties, a series of metal ions have been used by observing changes in fluorescent properties. Different metal ion solutions of Cu2+, Hg2+, Ni2+, Mn2+, Co2+, Cd2+, Pb2+, Fe2+, Ag+, Mg2+, Ca2+, Zn2+, Fe3+, Cr3+, Na+, K+, and Al3+ were added to the TPB buffer solutions. And the fluorescence and UV–Vis results were measured, respectively. The results are indicated in Figs. 4 and 5. Fluorescence and UV–visible spectra reflected that TPB were sensitive to Zn2+ and Cd2+ ions.

Fig. 4
figure 4

Fluorescent spectra of sensors TPB (a, 20 µM) and PTPB (b, 10 µM) of the DMF solutions in the existence of different metal ions Cu2+, Hg2+, Ni2+, Mn2+, Co2+, Cd2+, Pb2+, Fe2+, Ag+, Mg2+, Ca2+, Zn2+, Fe3+, Cr3+, Na+, K+, and Al3+ (10 equiv, the excited light is λex = 325 nm)

Full size image
Fig. 5
figure 5

UV–Vis spectra of sensors TPB and PTPB when in contact with different metals in DMF/water (v/v = 1/1) solution (5 µM, Cu2+, Hg2+, Ni2+, Mn2+, Co2+, Cd2+, Pb2+, Fe2+, Ag+, Mg2+, Ca2+, Zn2+, Fe3+, Cr3+, Na+, K+, and Al3+)

Full size image

Figure 4 shows the fluorescence changes of the sensor TPB when added with different ions, excited with λex = 325 nm. What we can observe is that except zinc and cadmium ions, the existence of other metal ions did not cause obviously changes in the TPB emission profile. After adding Zn2+ and Cd2+, the fluorescence of TPB clearly quenched and its intensity decreased nearly twofold for Zn2+ and Cd2+.

Figure 5 reveals the UV–visible spectra of TPB and PTPB. When ions were added to TPB, Zn2+ and Cd2+ ions caused strong fluorescence change. As we can see, the fluorescence increased when the TPB solutions reacted with Zn2+ and Cd2+ ions, compared to that with other ions. The absorption maximum of TPB with Zn2+ red shifted from 284 to 291 nm. And as for Cd2+, the absorption maximum of TPB red shifted from 284 to 288 nm. It is worth emphasizing that the sensor and Zn2+ and Cd2+ can achieve instantaneous effects. As soon as the ions added, the fluorescence effect changes immediately. Therefore, using the fluorescence change of TPB, we can identify Zn2+ and Cd2+ efficiently.

As shown in Fig. 4, the macromolecular chemosensor for Zn2+ and Cd2+ ions recognition is studied. Fluorescence and UV–Vis results indicate that PTPB is sensitive to Zn2+ and Cd2+, and other ions can be added without clear intensity change. It exhibits the similar performance as that of the precursor small molecular sensor, but is more sensitive than the small molecule because of the wavelength shift.

Figure 4 shows the fluorescence intensity changes of the macromolecular chemosensor PTPB upon the addition of different metal ions. The 388 nm peak was found to be the PTPB’s emission maximum (λex = 325 nm). Upon the addition of Zn2+ and Cd2+, the fluorescence enhanced approximately fivefold for Zn2+ and for Cd2+, and we can hardly see apparent changes when PTPB is in touch with the other ions: Cu2+, Hg2+, Ni2+, Mn2+, Co2+, Pb2+, Fe2+, Ag+, Mg2+, Ca2+, Fe3+, Cr3+, Na+, K+, and Al3+. Figure 4 indicates that the emission peak of PTPB was red shifted at a new wavelength, 418 nm. Therefore, PTPB showed the similar selectivity to Zn2+ and Cd2+ as TPB.

To better study the ion recognition ability of the macromolecular sensor PTPB, the UV–visible spectra of PTPB when in contact with different ions were measured (Fig. 5). As the figure shows, the absorption maximum of macromolecular PTPB was found red shift from 275 to 281 nm with Zn2+ and from 275 to 279 nm with Cd2+.

To better explore the small molecular and macromolecular sensors fluorescence selectivity, different ions were added to the TPB and PTPB, and the photos were taken under natural light and UV light (Fig. 6). As we can see, upon addition Zn2+ and Cd2+, fluorescence intensity altered and the color of the TPB and PTPB were also changed.

Fig. 6
figure 6

Images of PTPB (10 µM) and TPB (10 µM) with different metal ions in DMF under natural light and UV light

Full size image

Chemosensors TPB and PTPB’s sensitivity to Zn2+ and Cd2+

The fluorescence spectra of the two sensors were measured when adjusted the amount of zinc ions and cadmium ions to obtain their detection limit. As shown in Fig. 7, TPB is a fluorescence-quenched sensor and PTPB is a fluorescence-enhanced sensor, and the fluorescence intensity of TPB decreased and PTPB increased with the increase in Zn2+ and Cd2+ concentration gradually.

Fig. 7
figure 7

Two sensors’ fluorescence changes (DMF-buffer solution) when react with unequal amount of Cd2+ and Zn2+ from 0.01 to 10 equiv for TPB and from 0.1 to 10 equiv for PTPB, and λex is 325 nm. The concentration of TPB and PTPB is both 10 μM

Full size image

Concentration-dependent fluorescence signaling of chemical sensors for Zn2+ and Cd2+ is obtained from Fig. 7 and plotted in Fig. 8.

Fig. 8
figure 8

Concentration-dependent fluorescence signaling of Zn2+ and Cd2+ by chemosensors, TPB and PTPB toward Zn2+ and Cd2+ using the fluorescence spectroscopy technique

Full size image

In general, the following equation is used to acquire the detection limit (LOD):

$$ {\text{LOD }} = { 3}\alpha /K. $$
(1)

In this formula, α is the blank sensor solution’s standard deviation; Κ is the calibration curve’s slope. From what have been reflected in Figs. 7 and 8, the LOD of TPB towards Cd2+ ion and Zn2+ ion were found to be 0.475 μM and 1.53 μM, respectively, and those of PTPB towards Cd2+ ion and Zn2+ ion were 0.424 μM and 1.27 μM. It can be seen from the results that compared with the small molecular chemosensor, the macromolecular chemosensor is more sensitive to Zn2+ and Cd2+, and the limit of detection for Zn2+ and Cd2+ is significantly improved.

In our previous work, we also discovered that sensors with two or more fluorophore units in one molecule have a stronger signal than those with one fluorophore unit in one molecule, whether quenching or enhancing fluorescence (Jia and Cheng 2016). We analyzed this phenomenon and inferred that it because of the so-called ‘‘molecule wire’’ effect, the macromolecular chemosensors often show higher sensitivity and lower limit of detection because of the amplified fluorescence signals. In this work, the macromolecular sensor is a little bit more sensitive than its corresponding small molecules.

The analysis of Job’s plot for fluorescent sensors

Figure 9 shows the binding ratio studied by Job’s plot. The total concentration is always maintained at 10 μM, changing the concentration of metal ions and sensors continuously. Using the difference of fluorescence intensity (ΔF = F0 − F) between the theoretical value (F0) and the real-time value (F) as the ordinate, and the value of the ion concentration compared to the total concentration as the abscissa, we can acquired the Job’s plots.

Fig. 9
figure 9

The analysis of Job’s plots for PTPB and TPB after the addition of Zn2+ and Cd2+

Full size image

To know more about the binding characteristics, the binding ratio of chemical sensors was studied by Job’s plots. As indicated in Fig. 9, the binding ratio of PTPB and TPB to Cd2+ and Zn2+ was both 3:1 (ion/sensor). According to plot of Job’s, there may be three binding sites in PTPB and TPB for Cd2+ and Zn2+ (three 1, 10-phenanthroline).

Mechanism and reversibility analysis for fluorescent sensors

To better understand the binding mechanism, the infrared spectra (IR) and mass spectrometry (MS) were studied of chemosensors and ions. Figure 10 shows the IR spectra of the TPB and PTPB combined with Zn2+ and Cd2+. Clear changes could be seen upon addition Zn2+ and Cd2+. The peaks at 1585 cm−1 of TPB and PTPB, 1512 cm−1 of TPB, and 1430 cm−1 of PTPB have shifted and the intensity also enhanced, showing that –C=N– and –C=C– changed after combining ions. Obvious peak shape change could also be observed. The shift of the –C–H– peaks could be seen from 845 to 941 cm−1, 740 to 680 cm−1 and 825 cm−1 for TPB. As for PTPB, the –C–H– peaks were found shifted from 895 cm−1 and 692 cm−1 to 955 cm−1 and 823 cm−1. All these IR wavenumber offsets indicated that the N atom might be the combined site.

Fig. 10
figure 10

FTIR spectra of TPB and PTPB after adding Zn2+ and Cd2+

Full size image

As shown in Fig. 11, mass spectra of TPB with Zn2+ and Cd2+ were used to analyse the binding mechanism. The molecular weight was calculated to be 456.22 ([TPB + 3Zn2++4OH−+2H2O], m/z = 456.22) after TPB combining with Zn2+, measured to be 456.98. And the molecular weight for TPB with Cd2+ was calculated to be 363.48 ([TPB + 3Cd2++2NO3−+OH−], m/z = 363.48), found 363.73. The mass spectra also demonstrate that the TPB was chemical combine with ions and the binding ratio is 3:1.

Fig. 11
figure 11

Mass spectra of TPB after binding with Zn2+ (zinc acetate) and Cd2+ (cadmium nitrate)

Full size image

Sensors TPB and PTPB are different in their appearance and physical properties, but both of them can detect Zn2+ and Cd2+ ions efficiently. The possible reason for this phenomenon might be because all sensors have –C=N–. When –C=N– is bonded in a cyclic form, the isomerization of C=N is inhibited, resulting in fluorescence recovery of the molecule (Wang and Wu 1995; Yang et al. 1996).

After adding Zn2+ ion and Cd2+ ion, the fluorescence altered and the peak shift can be explained by the intramolecular charge transfer (ICT) mechanism (Thakare et al. 2017). The electron acceptor acts as a part of the recognition group and when the recognition group and the analyte are combined, the push–pull ability of the electron-withdrawing group as the recognition group changes, and the π electronic structure of the whole system is redistributed, resulting in absorption spectrum and emission spectrum change, the spectrum red shift.

To better explore the fluorescent sensors’ properties, FTIR spectra was studied to test the reversibility for the coordination of fluorescence chemosensors with Zn2+ and Cd2+. Edetate disodium (EDTA) was added separately to the solution containing the chemosensor PTPB with zinc and PTPB with cadmium. Figure 12 shows the IR spectra after the samples were dried. Significant changes at –C=N– and –C=C– vibration could be seen, the shift of the –C–H– peaks also can be observed. All the data proved that EDTA was unable to disrupt the complexation of PTPB with Zn2+ and Cd2+ ions, showing the irreversibility of the chemosensors when combining with ions.

Fig. 12
figure 12

FTIR spectra of the EDTA with PTPB containing zinc salt and PTPB with cadmium salt for reversibility test

Full size image

Conclusions

In conclusion, two new sensors, TPB and PTPB, with different fluorophore units in one molecule have been successfully synthesized for the first time and their notable sensing with Zn2+ and Cd2+ ions were discussed by the fluorescence techniques. Both the two sensors reflected high sensitivity when in contact with Zn2+ and Cd2+ in a wide pH range. Also, the PTPB was more sensitive and higher selective than TPB. The sensing ability to Zn2+ and Cd2+ ions was obviously increased because of the conjugation mechanism of molecular wires.