Introduction

Development of chemosensors for sensing and recognition of environmentally and biologically important heavy and transition metal ions, for example, Hg2+, Cu2+, Fe3+ and Cr3+, have attracted considerable attention of current researchers [1–9]. Fluorescent chemosensors provide several advantages over other analytical methods, such as high sensitivity, specificity, convenience, real time monitoring with fast response times, and low cost [10]. During the past decades, increasing research interests have been devoted to the fluorescence sensing of heavy and transition metal ions [11].

Because rhodamine dyes have many advantages such as low cost, long-wave length absorption/emission, high molar absorption coefficient, high quantum yield, and photostability [12–15]. They have been widely used as a molecular platform for the design of new spectroscopic probes [3, 10]. Moreover, in the absence of cations, these rhodamine-based chemosensors exist in a spirocyclic form, which is colorless and non-fluorescent. The addition of a specific metal ion leads to spirocycle unit opening via coordination or irreversible chemical reaction, resulting in the appearance of a pink color or orange fluorescence [3, 6, 15]. Thereby, the rhodamine fluorophore can be an ideal framework to construct off–on system for the specific metal ion [8]. Exploiting this idea in 1997, Czarnik et al. first reported rhodamine based chemosensor [16]. Following this work, the spirolactam-ring opening phenomenon has been utilized by many groups and a number of reports have appeared in literature for the detection of various ions based on rhodamine as signalling moiety [4, 8, 12, 17].

Up to now, variety of chemosensors for metal ions have been fabricated from small organic molecules based rhodamine derivataves [3]. However, most of these small molecular chemosensors typically exhibit poor water solubility, which partially limits their practical applications in diverse fields [4, 11]. This inconvenience can be overcome by using hydrophilic copolymers that also contain small amounts of the lipophilic organic receptors [18]. At present, there are some fluorescent polymeric sensors with different macromolecular structure for the detection of metal cations and protons in the environment [19–23].

Earlier we have developed polymeric sensors based on rhodamine in “off-on” mode for the detection of metal cations in aqueous solution [24–26]. As part of our continuous work on polymeric chemosensors based on polyving akohol grafting the rhodamine derivative [24], we prepared severally two water-soluble polymers, polyving akohol grafting N-mono-maleic acid amide-N′-rhodamine B hydrazide (PVA-MRBH) and polyving akohol grafting N-mono-succinic acid amide-N′-rhodamine 6G hydrazide (PVA-SR6GH), (Scheme 1). Unexpectedly, we found that the derivatives display obvious differences in fluorescence sensory pattern, metal cation species and color changes.

Scheme 1
scheme 1

Synthesis of MRBH, SR6GH, PVA-MRBH and PVA-SR6GH

Full size image

Experimental

As shown in Scheme 1, rhodamine containing copolymers PVA-MRBH and PVA-SR6GH were easily obtained by using postfunctionalization strategy. Firstly, the acylation reaction of maleic anhydride (MAH) with rhodamine B hydrazide (RBH) or of succinic anhydride (SAH) with rhodamine 6G hydrazide (R6GH) were carried out [27–29]. Then, the rhodamine moieties were linked to polyving akohol by esterification reaction between MRBHCl or SR6GHCl and the -OH functional group of PVA after transforming to acyl chloride from carboxylic acid. Although increased one step reaction using the esterification by acyl chloride with –OH of PVA, the yield was raised [30, 31]. Polyving akohol (PVA) is a polymeric material available in the market, which has good solubility in water. It can be modified due to the presence of abundant OH functional groups in the backbone [30]. Rhodamine B and rhodamine 6G belong to homolog, they have analogous fluorescent property [13].

Materials

PVA (Anhui Wanwei updated High-tech material industry company limited, Hefei, China) used in this study had a degree of polymerization of 1700 with a saponification value of 99 % with an average molecular weight of 75 000, and dried at 40 °C for 24 h before use. Rhodamine B, rhodamine 6G, maleic anhydride (MAH) and succinic anhydride (SAH) were purchased from Sigma-Aldrich Trading Co. Ltd. (Shanghai, China); Hydrazine hydrate, thionyl chloride and dimethyl sulfoxide (DMSO) were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China) and used without further purification; Anhydrous methanol and ethanol, acetic ether, dichloromethane, petroleum ether, N,N-dimethylformamide (DMF) and pyridine were obtained from commercial suppliers. Rhodamine B hydrazide (RBH) and rhodamine 6G hydrazide (R6GH) were prepared according to the literature method [32–35].

Instrumentation

The 1H NMR spectra were measured on a DRX 400 Bruker spectrometer (AVANCE AV 400, Bruker corporation, Switzerland) at 298 K in CDCl3 or D2O with TMS as internal standard. FTIR spectra were recorded on a Nicolet Neus 8700 FTIR spectrophotometer (Thermo Scientific Instrument Co. U.S.A) with KBr compressing tablet. Elemental analyses (C, H and N) were carried out on a VarioELIII analyzer (Elementar corporation, Germany) for MRBH and SR6GH. All pH measurements were made with a Model pHS-3C pH meter (Shanghai, China). Fluorescence spectra were acquired on a RF5301PC fluorescence spectrophotometer (Shimadzu Corporation, Japan).

Synthesis

Synthesis of N-Mono-Maleic Acid Amide-N′-Rhodamine B Hydrazide (MRBH) and N-Mono-Succinic Acid Amide-N′-Rhodamine 6G Hydrazide (SR6GH)

N-mono-maleic acid amide-N′-rhodamine B hydrazide (MRBH) were prepared according to literature protocol [3, 27, 28]. Briefly, rhodamine B hydrazide (RBH) (0.4566 g, 1.0 mmol) was dissolved in dichloromethane (30 mL), to which a solution of maleic anhydride (MAH) (0.1975 g, 1.0 mmol) in dichloromethane (10 mL) was added dropwise. The resulting solution was stirred at 50 °C for 4 h, and then the solvent was removed under reduced pressure to get a violet-red residue, which was purified by silica-gel column chromatography with ethyl acetate-petroleum ether (bp 60–90 °C) (1:9, v/v) as eluent, affording 0.1956 g of MRBH (Yield: 42.77 %). FTIR of MRBH: 3508.46, 3441.66 (s, ν-NH2); 2924.76 (s, ν-CH3), 2850.16(s, ν-CH2); 1700.31(s, νc=o); 1630.29 (s, νAr C=C); 1573.04 (m, νAr C=C), 1463.32, 1382.13 (m, νAr C=C). 1H NMR (400 MHz, CDCl3, 298 K) of MRBH: 7.83 (d, Ar-H), 7.46 (m, C = C-H), 7.39 (d, Ar-H), 7.25 (d, Ar-H), 7.24(d, Ar-H), 6.97 (s, C = C-H), 6.84(d, Ar-H), 6.19 (d, Ar-H), 6.13 (d, Ar-H), 3.39 (n, −CH2-), 1.13 (m, −CH3). Elemental analysis, calcd. for C32H34N4O5: C 69.30, H 6.18, N 10.10 %; found, C 69.41, H 6.32, N 9.72 %.

The synthesis of N-mono-succinic acid amide-N′-rhodamine 6G hydrazide (SR6GH) was similar to MRBH (Yield: 37.0 %). FTIR of SR6GH: 3423.11 (s, ν-NH2); 2973.04, 2931.28 (s, ν-CH3), 2877.15(s, ν-CH2); 1690.90(s, νc=o); 1632.13 (s, νAr C=C); 1514.58 (m, νAr C=C), 1424.88, 1383.12 (m, νAr C=C). 1H NMR (400 MHz, CDCl3, 298 K) of SR6GH: 7.88–7.89 (d, Ar-H), 7.37–7.88 (m, Ar-H), 6.96–6.99 (d, Ar-H), 6.43 (s, Ar-H, −CO-NH-N), 6.21 (s, Ar-H), 3.63–3.68 (m, Ar-NH-CH2-), 3.14–3.19 (m, Ar-NH-CH 2-), 2.55–2.59 (d, Ar-H), 6.13 (m, −CH 2–COOH), 3.39 (m, −CH 2-CO-NH-), 1.89(m, −CH2),1.14–1.29 (m, −CH3). Elemental analysis, calcd. for C30H32N4O5: C 68.17, H 6.10, N 10.60 %; found, C 68.21, H 6.22, N 9.98 %.

Synthesis of PVA-MRBH and PVA-SR6GH

Addition of 1 drop of N,N-dimethylformamide (DMF), 1 mL of dichloromethane, 0.5546 g (1 mmol) of MRBH, 0.18 mL of SOCl2 (1.5 mmol) in three-necked flask of 100 mL, the mixture was magnetic stirred for 12 h at room temperature and then was reacted under reflux for 2 h. The residual SOCl2 was removed by the reduced pressure distillation, affording pink solid of acyl chloride, N-mono-maleic acyl chloride amide-N′- rhodamine B hydrazide (MRBHCl) [28].

PVA solution in DMSO was prepared by adding 3.312 g of purified PVA powder to 50 mL of DMSO and heating this mixture to 80 °C with continuous mechanical stirring until a clear solution was obtained. 0.2210 g (0.3856 mmol) of MRBHCl was dissolved in 5 mL dimethyl sulfoxide (DMSO) and added the solution to PVA solution of DMSO by dropping funnel under agitating at 80 °C. The mixture was reacted in an oil bath at 80 °C by mechanical stirring for 5 h. While cooled to room temperature, the orange solution was allowed to precipitate in excess anhydrous methanol to provide a pink deposit. After vacuum filtration, the solid was washed with ethanol until the solvent was not fluorescent. The powder was then put in a Soxhlet extractor and extracted with ethanol and chloroform for at least 12 h, respectively, to ensure that there was noncovalently bounded RBH or MRBH in PVA. The desired product was ultimately synthesized via vacuum drying (Yield: 83.51 %), as illustrated in Scheme 1. [20, 36]. FTIR of PVA-MRBH (KBr), cm−1: 3444.83(ν PVA-OH); 2909.05–2951.1, 2843.57(ν-CH2-,-CH-); 1645.45(νC=O); 1469.91, 1329.47, 1232.92, 1145.14(νaromaticC=C); 1099.06(νC-O); 1423.82 (δC-H). 1H NMR (400 MHz, D2O, 298 K) of PVA-MRBH: 7.94 (d, Ar-H), 7.46 (d, C = C-H), 7.31 (m, CO-NH-), 7.24 (d, Ar-H), 7.2–7.1 (d, Ar-H), 6.97, 6.90 (d, C = C-H, Ar-H), 6.19 (s, Ar-H), 6.13 (d, Ar-H), 4.65, 4.52, 4.45 (d, −OH), 4.51 (d, −CH-MRBH), 3.89 (m, −CH-OOCCH3), 3.41 (m, −OOCCH3), 3.32 (d, −CH-OH), 2.51, 2.50, 2.498 (m, −CH 2 -CH3), 1.59–1.37 (m, −CH2-), 1.07, 1.06, 1.04 (m, −CH2-CH 3).

The preparation of PVA-SR6GH was similar to PVA-MRBH (Yield: 87.27 %). FTIR of PVA-SR6GH (KBr), cm−1: 3444.76 (ν PVA-OH); 2920.46, 2846.32 (ν-CH2-,-CH-); 1632.13 (νC=O); 1462.00, 1387.46, 1134.11 (νaromaticC=C); 1099.00 (νC-O); 1414.05 (δC-H). 1H NMR (400 MHz, D2O, 298 K) of PVA- SR6GH: 1.04–1.08 (m, −CH2-CH 3), 1.60–1.38 (m, −CH2-), 1.99 (s, −Φ-CH3), 2.50, (m, −CH2-), 3.18 (s, CH3CH2-NH-Φ), 3.32 (d, −CH-OH), 3.41 (m, −OOCCH3), −CH-OOCCH3), 4.66, 4.53, 4.46 (d, −OH), 6.31 (s,–Φ-H), 6.53–6.59 (s,–Φ-H), 7.07–7.10 (s,–Φ-H), 7.29 (s,–Φ-H), 7.30 (m, CO-NH-), 7.46–7.51 (s,–Φ-H), 7.97–8.01 (s,–Φ-H).

Preparation and Fluorescence Intensity of PVA-MRBH and PVA-SR6GH Aqueous Solution

PVA-MRBH and PVA-SR6GH, like PVA, were hard to dissolve in water at room temperature, so the polymer solutions of desired concentration were prepared by dissolving a known amount of PVA-MRBH or PVA-SR6GH in deionized water with gentle stirring at 80 °C, and were kept for 2 h to ensure homogenization [29]. For fluorescence emission measurements, a 10 × 10 mm quartz cell was used for detection. The effect of the metal cations on fluorescence intensity was examined by adding a few microlitre of stock solution of the metal cations to a known volume of the polymer solution (2.00 mL). The addition was limited to 0.10 mL, so that the dilution of the polymer solution remained insignificant [37]. The excitation and the emission slit widths were 10 nm and 5 nm, respectively, excitation wavelength was 500 nm, scanning range were from 520 to 650 nm, scanning speed was medium, and testing temperature was at 25 °C.

The detection limit was calculated with the equation: detection limit = 3S/ρ, where S is the standard deviation of blank measurements and ρ is the slope between relative intensity versus sample concentration [18, 30].

Results and Discussion

Effect of Polymer Concentration on Fluorescence Intensity

Figure 1 was fluorescent emission spectra of PVA-MRBH in different polymer concentration under neutral conditions (λex = 500 nm). Inset was effect of polymer concentration on fluorescence intensity. It can be seen that the higher the polymer concentration was, the stronger the fluorescence intensity was. Between fluorescence intensity (I) and polymer concentration ([P]) shown good line relationship (R = 0.9855).

Fig. 1
figure 1

Fluorescence emission spectra of PVA-MRBH in different polymer concentration (Inset Effect of polymer concentration on fluorescence intensity, λex = 500 nm)

Full size image
$$ I=3.329+6.176\left[P\right] $$
(1)

This indicated that rhodamine derivatives underwent equilibrium between spirocyclic (nonfluorescence) and ring opened (fluorescence) forms in the aqueous solution. Under certain pH, the contents of either the ring-opened or ring closed structures were increased with the increasing of polymer concentration [12, 38].

Time-Dependence of PVA-MRBH–Cu2+ Complex

The reaction between PVA-MRBH and Cu2+ ions was found to be instantaneous due to the presence of active phenyl hydrazide group but for time taken for complete reaction, which was studied by keeping all other reaction parameters constant. In a standard cuvette, 5 μL of 0.1 mol/L Cu2+ solution was added to 2 mL of 8.5 mg/mL PVA-MRBH ([Cu2+] = 2.5 × 10−4 mol/L) and the fluorescence emission (λex = 500 nm) values were measured and plotted as function of time [39]. As shown in Fig. 2, the time dependence of the response of PVA-MRBH to Cu2+ ions was investigated. It could be seen that the fluorescence signal of the PVA-MRBH with Cu2+ ion remarkably decreased for a few minutes, and leveled off as the time continues. The fluorescence intensity of PVA-MRBH with Cu2+ reached its minimum value at about 10 min, after which the fluorescence intensity remained almost constant [26, 40–44].

Fig. 2
figure 2

Effect of time on fluorescence intensity of PVA-MRBH (Cp = 8.5 mg/mL, [Cu2+] = 2.5 × 10−4 mol/L, λex = 500 nm)

Full size image

The Sensitivity of PVA-MRBH for Cu2+ and Fe3+

The fluorogenic sensing behaviors of PVA-MRBH aqueous solutions were depicted in Fig. 3 (Cp = 8.5 mg/mL, λex = 500 nm) [18]. Unlike most of the spirocycle RBH derivatives, when the polymer concentration was high, for example, 8.5 mg/mL, the aqueous solution of PVA-MRBH was rose pink and exhibited certain fluorescence property in neutral water, implying that there were a some of acid amides form. When the concentration of Cu2+ and Fe3+ added to the aqueous solution of PVA-MRBH were reached to 2.0 × 10−3 and 2.5 × 10−3 mol/L, the fluorescence was quenched and were reduced to a 0.58-fold and 0.60-fold, respectively (Fig. 3). The fluorescence turn-off were further supported by the observation that the emission color of the probe solution turned from pale pink to deep purple for Cu2+, and from pale pink with orange red to orange for Fe3+, respectively (Fig. 4) [22, 31, 44, 45].

Fig. 3
figure 3

The effect of Cu2+ or Fe3+ ions concentration ([Cu2+] or [Fe3+]) on relative fluorescence intensity (I/I0) of PVA-MRBH (Cp = 8.5 mg/mL, λex = 500 nm)

Full size image
Fig. 4
figure 4

Photographs recorded under visible light for aqueous solutions of PVA-MRBH (Cp = 8.5 mg/mL) in the presence of different metal ions concentration ((a) [Cu2+] = 0–5.0 × 10−4 mol/L, (b) [Fe3+] = 0–7.50 × 10−3 mol/L)

Full size image

Relative fluorescence intensities (I/I0) at 583 nm obtained from Fig. S11 were plotted vs. Cu2+ or Fe3+concentration and linear spectrofluoro-metric responses for Cu2+ or Fe3+ concentration were obtained with very good regression coefficients as R = 0.9740 and R = 0.9834, respectively. These linear responses could be used for detection of Cu2+ or Fe3+ concentration using the following equations for concentration range between 0.5 × 10−4−2.0 × 10−4 mol/L (a) and 0–5.0 × 10−4 mol/L (b) (Eqs. 2 and 3, respectively):

$$ I/{I}_0=0.8103-7.787\times {10}^2\left[C{u}^{2+}\right] $$
(2)
$$ I/{I}_0=0.9909-5.357\times {10}^2\left[F{e}^{3+}\right] $$
(3)

where I is the emission intensity of tested sample at 583 nm and I0 is the emission intensity of metal free polymer solution [31, 45, 46].

The detection limits, which were calculated as three times the standard deviation of the background noise from the calibration curve, for the determination of Cu2+ and Fe3+ ions in the same medium were found to be 3.85 × 10−6 and 8.40 × 10−7 mol/L [18, 23, 31, 42].

The Sensitivity of PVA-SR6GH for Cu2+, Fe3+, Cr3+ or Hg2+

Figure 5 presented the sensibility of the PVA-SR6GH (Cp = 1.0 mg/mL, λex = 500 nm) in aqueous solution at different concentrations of Cu2+, Fe3+, Cr3+ or Hg2+ ions. The increase of the fluorescence intensity occured after the addition of Cu2+, Fe3+, Cr3+ or Hg2+ ions in the concentration range from Cu2+, Fe3+, Cr3+ or Hg2+-free solution to 2.0 × 10−3, 1.5 × 10−4, 2.0 × 10−4 or 4.0 × 10−4 mol/L, respectively. The fluorescence intensity value remained constant above certain concentration. The relative fluorescence intensity enhanced nearly 1.64, 1.55, 1.60, or 4.14 times, respectively. However, we attempted addition of Cu2+, Fe3+, Cr3+ or Hg2+ ions to aqueous solution of the PVA-SR6GH and did not find any significant changes in the relative fluorescence intensity [18, 22, 31, 44, 45].

Fig. 5
figure 5

The effects of Cu2+, Fe3+, Cr3+ or Hg2+ ions concentration ([Cu2+], [Fe3+], [Cr3+] or [Hg2+]) on relative fluorescence intensity (I/I0) of PVA-SR6GH (Cp = 1.0 mg/mL; λex = 500 nm; [Mn+] was [Cu2+], [Fe3+], [Cr3+], or [Hg2+])

Full size image

Figure S12 showed the comparative response of PVA-SR6GH to Cu2+, Fe3+, Cr3+ or Hg2+ in aqueous solution. The dependence of the relative fluorescence intensity (I/I0) versus the concentration of Cu2+, Fe3+, Cr3+ or Hg2+ ([Mn+]) in the certain concentration range (0 to 2.0 × 10−4, 0 to 3.0 × 10−5, 0 to 5.0 × 10−5 or 0 to 7.5 × 10−5 mol/L) exhibited quite good linear correlation, which were described by Eqs. 4–7 with the correlation coefficient 0.9827, 0.9906, 0.9929 or 0.9968, respectively [31, 45, 46].

$$ I/{I}_0=1.0098+2.269\times {10}^3\left[C{u}^{2+}\right] $$
(4)
$$ I/{I}_0=1.0281+1.5908\times {10}^4\left[F{e}^{3+}\right] $$
(5)
$$ I/{I}_0=0.9944+1.0094\times {10}^4\left[C{r}^{3+}\right] $$
(6)
$$ I/{I}_0=1.0014+2.4423\times {10}^4\left[H{g}^{2+}\right] $$
(7)

The detection limits for Cu2+, Fe3+, Cr3+ and Hg2+ions in the same medium were found to be 6.61 × 10−11, 7.51 × 10−11, 4.45 × 10−11 and 1.23 × 10−12 mol/L [18, 23, 31, 42].

As mentioned previously, the color reaction of PVA-MRBH with Cu2+ or Fe3+ were attributed to the ring-opening of the spirolactam structure promoted by Cu2+ or Fe3+ complexation. However, the reaction system showed fluorescence quenching, which was rather different from that of the common rhodamine spirolactam derivatives [12, 13, 17] and PVA-SR6GH. Sun et al. [3] have investigated this unusual reaction mechanism by a comparative study on N-mono-maleic acid amide-N′-rhodamine B hydrazide (MRBH) and model compound N-acryloyl rhodamine B hydrazide (ARB). They believd that the extra carboxyl group in CARB played a crucial role in the color-on reaction, and without it the reaction could not occur. According to MRBH and Cu2+ been formed complex with 1:2 stoichiometery, they conjectured that the two Cu2+ ions in the complex may play different roles: one induces the opening of the spirocyclic structure and the other quenches the fluorescence of the xanthene moiety. We further found that it is the single and double bonds in linkers that affects the sensing mode. It’s derivatives can be enhanced for the former and quenching for the latter by metal ions.

Selectivity and Competitiveness of PVA-MRBH and PVA-SR6GH

Relative fluorescence intensity has been used as a quantitative measure of the effects of metal cations (including in Ag+, Ba2+, Cd2+, Co2+, Cr3+, Cu2+, Fe2+, Fe3+, Hg2+, K+, La3+, Mg2+, Na+, Ni2+, Pb2+, and Zn2+ cations, metal cations concentration was 2.50 × 10−4 mol/L, polymer concentration was 8.5 and 1.0 mg/mL for PVA-MRBH and PVA-SR6GH) on relative fluorescence intensity [30]. The changes in the relative fluorescence intensity (I/I0) of PVA-MRBH and PVA-SR6GH induced by the metal cations were investigated and presented in Figs. 6 and 7. As seen from Fig. 6, the addition of metal cations led to a decrease or increase of the fluorescence intensity for the polymer system, which was different for each metal cation. The PVA-MRBH has no response for fluorescence spectra upon the addition of Cd2+. There were small fluorescence enhancement for Cr3+ and Hg2+. The fluorescence quenching effects were observed in the presence of most of these metal cations, but the highest for Cu2+, then the Fe3+ ions. As seen from Fig. 7, there were small changes of fluorescence intensity for PVA-SR6GH adding Ag+, Ba2+, Cd2+, Co2+, Fe2+, K+, La3+, Mg2+, Na+, Ni2+, Pb2+, and Zn2+ cations, but there were highest fluorescence enhancement for PVA-SR6GH upon addition of Cu2+, Fe3+, Cr3+ and Hg2+ [30, 31, 44, 46, 47].

Fig. 6
figure 6

Fluorescence response of PVA-MRBH (8.5 mg/mL) to 2.50 × 10−4 mol/L of Cu2+ or Fe3+ and other metal ions (the red bar portion) and to the mixture of 2.50 × 10−4 mol/L of other metal ions with 2.50 × 10−4 mol/L Cu2+ (the green bar portion) or Fe3+ (the blue bar portion)

Full size image
Fig. 7
figure 7

Fluorescence response of PVA-SR6GH (1.0 mg/mL) to 5.0 × 10−4 mol/L of Cu2+, Fe3+, Cr3+ or Hg2+ and other metal ions (the red bar portion) and to the mixture of 5.0 × 10−4 mol/L of other metal ions with 5.0 × 10−4 mol/L Cu2+ (the blue bar portion), Fe3+ (the cyan bar portion) Cr3+ (the green bar portion) or Hg2+ (the magenta bar portion)

Full size image

The competitive experiments were conducted by adding Cu2+or Fe3+ ions (2.5 × 10−4 mol/L) to the solution of PVA-MRBH (Cp = 8.5 mg/mL) in the presence of 2.5 × 10−4 mol/L of other metal ions (Ag+, Ba2+, Cd2+, Co2+, Cr3+, Fe2+, Hg2+, K+, La3+, Mg2+, Na+, Ni2+, Pb2+ and Zn2+, see Fig. 6). The results further revealed that for PVA-MRBH, other metal ions, except for Cr3+, Hg2+, Fe3+ and Zn2+ ions, did not interfere with Cu2+-induced fluorescence quenching. Ag+, Cu2+, La3+, and Zn2+ ions can slightly interfere with the fluorescence quenching of PVA-MRBH moieties by Fe3+ ions.

The competitive experiments were also carried out by adding Cu2+, Fe3+, Cr3+ or Hg2+ ions (5.0 × 10−5 mol/L) to the solution of PVA-SR6GH (Cp = 1.0 mg/mL) in the presence of 5.0 × 10−4 mol/L of other metal ions (Ag+, Ba2+, Cd2+, Co2+, Fe2+, K+, La3+, Mg2+, Na+, Ni2+, Pb2+ and Zn2+) as shown in Fig. 7. As shown in Fig. 7a, most cations, such as Ag+, Co2+, K+ and Pb2+, had negligible influence on Cr3+. However, Fe3+ ion led to a significant fluorescence enhancement for the solution of the PVA-SR6GH and Cr3+, while the Cu2+ ion led to a significant fluorescence quenching of the solution of the PVA-SR6GH and Cr3+. From Fig. 7a, it also can be seen that among a series of cations in PVA-SR6GH aqueous solution, including Ag+, Ba2+, Cd2+, Co2+, Cr3+, Cu2+, Fe2+, Fe3+, Hg2+, K+, La3+, Mg2+, Na+, Ni2+, Pb2+, and Zn2+ cations, adding of Cu2+, only Hg2+ made a considerable fluorescence enhancement, which interfered with Cu2+-induced fluorescence enhancement. As shown in Fig. 7b, the Fe3+-induced luminescence enhancement was not obviously affected in the presence of environmentally relevant alkali, alkaline-earth metals as well as other cations mentioned above, except for Hg2+, Cu2+and Fe2+. It could also be seen from Fig. 7b that besides Ag+, Cu2+, Fe2+ and Fe3+, other ions all had not seriously interfered on Hg2+ [22, 31, 47].

The color and fuorescence changes of PVA-MRBH (Cp = 8.5 mg/mL) or PVA-SR6GH (Cp = 1.0 mg/mL) upon the addition of various cations (Ag+, Ba2+, Cd2+, Co2+, Cr3+, Cu2+, Fe2+, Fe3+, Hg2+, K+, La3+, Mg2+, Na+, Ni2+, Pb2+ and Zn2+, 2.50 × 10−4 mol/L) are shown in Fig. 8. Figure 8 (Top (a), under visible light) showed that when Cu2+ or Fe3+ ions were added into the aqueous solution of PVA-MRBH, the dramatic color of the solution changes occured from slight pink to magenta or orange red, respectively. Moreover, we also noted that the rate of chromogenic reaction for Cu2+ was faster than for Fe3+ ions. Under the same conditions, upon additions of other ions including Ag+, Ba2+, Cd2+, Co2+, Cr3+, Fe2+, Hg2+, K+, La3+, Mg2+, Na+, Ni2+, Pb2+ and Zn2+ resulted in small or no obvious color changes. Under UV light at 365 nm (Fig. 8, Top (b)), the colors of the aqueous solution of PVA-MRBH were changed from weak yellow to dark upon the addition of Cu2+or Fe3+ ions, which could be ascribed to the fluorescence quenching of PVA-MRBH by Cu2+or Fe3+ ions.

Fig. 8
figure 8

Images of color reactions of PVA-MRBH (8.50 mg/mL) and PVA-SR6GH (1.0 mg/mL) with various ions (Ag+, Ba2+, Cd2+, Co2+, Cr3+, Cu2+, Fe2+, Fe3+, Hg2+, K+, La3+, Mg2+, Na+, Ni2+, Pb2+, and Zn2+ cations at the same concentration of 2.50 × 10−4 mol/L). Reactions were performed at room temperature for 10 min in aqueous solution. a under visible light; b UV light at 365 nm

Full size image

Figure 8 (Bottom) showed that when the metal cations mentioned above were added into the solution of PVA-SR6GH, the color changes of the solution occured from colorless to yellow for Fe3+. Although Cr3+, Cu2+and Hg2+ could make fluorescence of PVA-SR6GH enhancing, there were no color changes while added these three cations [18, 21, 22, 31, 44, 46, 48].

Conclusions

In summary, a simple and low-cost post-functionalization strategy was adopted to prepare two fluorescent polymeric chemosensors, PVA-MRBH and PVA-SR6GH, by covalent coupling of fluorescent molecular MRBH and SR6GH to water-soluble polyving akohols (PVA). It was found that although there were only a difference in single and double in the linkers, they possessed wholly diverse properties in fluorescent sensory pattern, metal cation species and color changes in aqueous solution. PVA-MRBH could sense Cu2+ and Fe3+ metal cations with fluorescence quenching pattern, while PVA-SR6GH could respond Cr3+, Cu2+, Fe3+ and Hg2+ metal cations with fluorescence enhancements. Moreover, PVA-MRBH and PVA-SR6GH had favorable colorimetric properties. When titration of Cu2+ and Fe3+ into PVA-MRBH, the change of clear color occured from rose pink to amaranth and orange, respectively. Upon the addition of Cr3+, Cu2+, Fe3+ and Hg2+ into the aqueous solution of PVA-SR6GH, only Fe3+ could make the color of the solution changing from colorless to yellow.