Turn-On Fluorescent Chemosensor for Hg²⁺ Based on Multivalent Rhodamine Ligands

Abstract

Rhodamine-based fluorescent chemosensors 1 and 2 exhibit selective fluorescence enhancement to Fe³⁺ and Hg²⁺ over other metal ions at 580 nm in CH₃CN/H₂O (3/1, v/v) solution. Bis(rhodamine) chemosensor 1, under optimized conditions (CH₃CN/HEPES buffer (0.02 M, pH = 7.0) (95/5, v/v)), shows a high selectivity and sensitivity to Hg²⁺, with a linear working range of 0–50 μM, a wide pH span of 4–10, and a detection limit of 0.4 μM Hg²⁺.

1. Introduction

Mercury is considered to be a highly dangerous element by the United States Environmental Protection Agency due to its special properties, such as migration through cell membranes and bioaccumulation within living tissues [1,2]. Therefore, there is a high demand for the determination of the Hg²⁺ ion in environmental analysis.

In recent years, rhodamine-based fluorescent chemosensors have received considerable attention for the detection of Hg²⁺[3–14], Cu²⁺[15–17], Pb²⁺[18], Cr³⁺[19], and Fe³⁺[20], because their special structural properties provide an ideal mode to construct off-on fluorescent switch chemosensors. Rhodamine having a spirolactam structure is non-fluorescent, whereas ring-opening of the spirolactam gives rise to a strong fluorescence emission. Moreover, they have a longer emission wavelength (about 550 nm), which is often preferred to serve as reporting group for analytes to avoid the influence of the background fluorescence (below 500 nm) [21–23]. However, most of them have shortcomings in practical application, such as cross-sensitivities toward other metal cations, low water solubility, a narrow pH span, and delayed response, etc. Accordingly, quantitative practical Hg²⁺ detection requires a linear fluorescence response, uniform fluorescence output at a broad pH range, compatibility with aqueous medium, high selectivity, sensitivity, and a fast response, while easy synthetic procedures for the sensors are of utmost importance.

This study deals with new rhodamine-based CHEF (chelation-enhanced fluorescence) chemosensors 1 and 2 (Chart 1) for the detection of Hg²⁺ ions showing that, compared to related rhodamine-based chemosensors, small structural changes give rise to improved selectivity and sensitivity. Chemosensor 1 is a bis(rhodamine) in which the two units are connected via amide groups meta substituted to a benzene ring. In order to study the influence of a third functionalized rhodamine on the Hg²⁺ complexation, tris(rhodamine) chemosensor 2 was prepared and evaluated for comparison.

2. Results and Discussion

Rhodamine derivatives 1 and 2, possessing two or three rhodamine moieties, respectively, were prepared by reacting rhodamine B hydrazide (4) with isophthaloyl dichloride (3) or benzene-1,3,5-tricarbonyl trichloride (5) in THF as a solvent (Scheme 1). The formation of 1 and 2 followed from the ¹H NMR spectra as the doublets at 6.42 and 6.46 ppm in rhodamine B hydrazide (4) shifted to 6.61–6.75 ppm and 6.59–6.76 ppm as multiplets for 1 and 2, respectively. In the ESI-MS mass spectra the [M+H] peaks were found at m/z 1043.5 and 1525.7 for 1 and 2, respectively.

The perchlorate salts of Na⁺, K⁺, Pb²⁺, Co²⁺, Cd²⁺, Cs⁺, Ag⁺, Cu²⁺, Mg²⁺, Zn²⁺, Hg²⁺, Fe²⁺, and Fe³⁺ ions were used to evaluate the metal ion binding properties of chemosensors 1 and 2 in CH₃CN/H₂O (3/1, v/v). The fluorescence spectra were obtained by excitation of the rhodamine fluorophore at 510 nm. Among these metal ions (80 equiv), chemosensors 1 and 2 both showed large chelation enhanced fluorescence (CHEF) effects with Hg²⁺, Fe³⁺ and smaller CHEF effects with Cu²⁺ (Figure 1). The addition of 400 μM (80 equiv) of Fe³⁺ and Hg²⁺ immediately yielded a pink solution with a absorption signal at 561 nm [24] and a strong fluorescence signal at 580 and 590 nm, respectively (Figure 2). For chemosensor 1, there was 35-fold enhancement with Fe³⁺ and 84-fold enhancement with Hg²⁺, while chemosensor 2 yielded a 27-fold enhancement with Fe³⁺ and 33-fold with Hg²⁺. The results can be attributed to a similar binding behavior of 1 and 2 both containing rhodamine moieties. In addition, a very weak fluorescence signal for free 1 and 2 was observed at 580 and 590 nm, respectively, upon excitation at 510 nm, confirming the presence of a ring-closed spirolactam structure, whereas with the addition of Fe³⁺ or Hg²⁺ ions, ring-opening of the spirolactam occurs and gives rise to a strong fluorescence emission at 580 and 590 nm, respectively. Though Cu²⁺ gave a small color change and a very small fluorescence enhancement, the spectroscopy and interaction of chemosensors 1 and 2 with Cu²⁺ are completely different from those of 1 and 2 with Fe³⁺ and Hg²⁺ as recently reported by others [20].

To investigate the binding mode and the affinity, Job’s plots (see Figure S1) were determined and fluorescence titration experiments were carried out for chemosensors 1 and 2 with Fe³⁺ and Hg²⁺ (Figure 3). The Job’s plots show that in all cases 1:1 complexes were formed. The resulting titrations also fitted to a 1:1 binding model, and the association constant (K_s) can be gained using Equation 1 [25,26].

$$\frac{(I-I_0)}{I_0}=\frac{\alpha \times [\text{M}]}{(1/K_s)+[\text{M}]}$$

(1)

Where I₀ is the fluorescence intensity of the chemosensors 1 and 2 in the absence of metal ions and I is the fluorescence intensity upon the addition of metal ions. α is the maximum specific binding, [M] is the metal ion concentration, K_s is the association constant.

The association constants for 1 with Fe³⁺ and Hg²⁺ were found to be 7.99 × 10³ M⁻¹ and 8.62 × 10³ M⁻¹, while those for 2 with Fe³⁺ and Hg²⁺ were 6.18 × 10³ M⁻¹ and 6.08 × 10³ M⁻¹, respectively, which are close to those of a related bis(rhodamine) chemosensor [20]. The K_s values of chemosensors 1 and 2 only marginally differ, those of tris(rhodamine) 2 even being slightly smaller than those of bis(rhodamine) 1. In addition to the Job’s plot determination, this also demonstrates that two rhodamines are sufficient for optimal metal ion binding. In chemosensor 1, two carbonyl oxygens as well as two amide oxygens can provide a stable binding pocket for metal ions.

To obtain a high selectivity and sensitivity for Hg²⁺ under aqueous conditions, HEPES buffer (0.02 M, pH = 7.0), MES buffer (0.01 M, pH = 7.0), PES buffer (0.01 M, pH = 7.0), and Tris HCl buffer (0.01 M, pH = 7.0) were used, respectively (Figure 4a). The fluorescence intensities upon addition of 40 equiv of Fe³⁺, Hg²⁺, and Cu²⁺ ions show the effect of the different buffer systems. In this case, HEPES effectively inhibits the interference of Fe³⁺ and Cu²⁺ ions during the detection of Hg²⁺.

The fraction of HEPES buffer used in CH₃CN played an important role in the affinity of 1 toward Hg²⁺. Because of the strong interaction between buffer anions and Fe³⁺, a small fraction of HEPES buffer in organic solvent was already beneficial to inhibit the binding of Fe³⁺. However, a high fraction of HEPES buffer caused a decrease of the fluorescence emission for 1·Hg²⁺ and an increase of that of the complex with Cu²⁺, which has to be avoided for Hg²⁺ analysis. To determine the optimal analysis condition, 5 μM chemosensor 1 in CH₃CN containing different fractions (5%, 10%, 15%, and 25% (v/v)) of 0.02 M HEPES buffer at pH 7.0 were used for the detection of Fe³⁺, Hg²⁺, and Cu²⁺ (Figure 4b). No significant fluorescence enhancement could be observed at 580 nm for Fe³⁺ compared to that of Hg²⁺ and Cu²⁺ at the same concentration. The results already show that 5% HEPES/CH₃CN was already sufficient for efficient monitoring of Hg²⁺.

Fluorescence titrations of Hg²⁺ by 1 were performed, under the optimized conditions of CH₃CN/HEPES buffer (0.02 M, pH = 7.0) (95/5, v/v) (Figure 5). For the quantitative detection of Hg²⁺ ions, under the optimized conditions, a calibration curve was generated by determining the fluorescence intensity of 1 (5 μM) at 580 nm upon addition of Hg²⁺ ions with different concentrations, ranging from 0 to 50 μM [27]. Figure 5 exhibits over the entire Hg²⁺ concentration range an almost perfect linearity (I₅₈₀ = 1.43 × [Hg²⁺] + 9.34, R² = 0.9962) between the fluorescence intensity of 1 and the Hg²⁺ concentration, indicating a linear detection range for Hg²⁺ determination. The detection limit, defined as three times the standard deviation of the blank signals [28], was found to be 0.4 μM from 10 blank solutions. In addition, Figure 6 shows that chemosensor 1 detected Hg²⁺ ions with high selectivity under these conditions.

For practical applicability of this new chemosensor, a proper pH range of 4–10 was determined. Figure 7a shows variations of the fluorescence intensity of 1 with pH in the absence and presence of the Hg²⁺ ion in CH₃CN/H₂O solution (95/5, v/v). In this region, free 1 has a weak fluorescence emission due to the presence of the ring-closed spirolactam structure, while addition of the Hg²⁺ ion leads to ring-opening of the spirolactam ring, resulting in a remarkable increase of the fluorescence.

The ring-opening of the spirolactam in chemosensor 1 produced a time-dependent dosimetric response, controlled by the reaction kinetics. Under the optimized conditions less than 4 min were required to complete the reaction (Figure 7b).

Chemosensor 1 in CH₃CN/HEPES buffer (0.02 M, pH = 7.0) (95/5, v/v) has a low detection limit (0.4 μM), a large linear detection range (0–50 μM), a wide pH span (4–10), and a rapid response time (4 min), exhibiting higher sensitivity and selectivity than most other previously reported rhodamine-based chemosensors [3–14].

3. Experimental Section

3.1. General

Absolute acetonitrile of analytical grade and deionized water were used throughout the experiments. All chemicals needed for the synthesis were purchased from known suppliers and used without further purification. The known rhodamine B hydrazide (4) was prepared according to a literature procedure [29]. The metal ion solutions were prepared from their analytical grade perchlorate salts. HEPES buffer, MES buffer, PES buffer, Tris HCl buffer solutions and different pH solutions were prepared using proper amounts of HEPES, MES, PES, Tris, 1.0 M HCl, and 1.0 M NaOH (all of analytical grade) under adjustment by a pH meter.

3.2. Equipment

Absorption spectra were determined on a Perkin Elmer Lambda 850 UV-vis spectrophotometer. Fluorescence spectroscopy measurements were performed on a Perkin Elmer LS55 spectrofluorimeter equipped with a xenon discharge lamp and using 1 cm quartz cells. All pH measurements were made with a Mettler Toledo SevenEasy pH meter. ¹H NMR and ¹³C NMR spectra were recorded on a Varian Unity INOVA (300 MHz) spectrometer in CDCl₃. ¹H NMR (300 MHz) and ¹³C NMR (75 MHz) chemical shift values are reported as δ using the residual solvent signal as an internal standard. Electrospray Ionization (positive mode) mass spectra were recorded on a WATERS LCT mass spectrometer.

3.3. Synthesis of 1 and 2

3.3.1. General Procedure for the Synthesis of 1 and 2

To a solution of rhodamine B hydrazide (4) and triethylamine in THF a solution of isophthaloyl dichloride (3) or benzene-1,3,5-tricarbonyl trichloride (5) in THF was added dropwise at 0 °C. The reaction mixture was brought to room temperature in 1 h, followed by stirring overnight at room temperature. The solvent was evaporated and the residue was dissolved in dichloromethane (50 mL), washed with 10% NaHCO₃ solution (3 × 50 mL) and water (3 × 50 mL). The organic layer was concentrated under reduced pressure to afford crude products 1 or 2.

Chemosensor 1 was synthesized starting from rhodamine B hydrazide (4) (1.3 g, 2.8 mmol), isophthaloyl dichloride (3) (0.29 g, 1.4 mmol) in THF (5 mL) and triethylamine (0.3 g, 3.0 mmol) in THF (70 mL). The crude product was recrystallized from a mixture of diethyl ether and dichloromethane (3:1) to afford the pure product (0.77 g, 52%) as a solid. mp 184 °C–186 °C. ¹H NMR: δ 1.02–1.22 (m, 24H, CH₃), 3.17–3.41 (m, 16H, CH₂), 6.22–6.40 (m, 8H, ArH), 6.61–6.75 (m, 4H, ArH). 7.15 (d, J = 6.0 Hz, 2H, ArH), 7.39–7.47 (m, 2H, ArH), 7.47–7.56 (m, 4H, ArH), 7.61–7.69 (m, 2H, ArH), 7.96 (d, J = 6.0 Hz, 2H, ArH). ¹³C NMR: δ 12.6, 44.3, 97.7, 98.0, 107.9, 123.5, 124.2, 128.3, 129.3, 131.0, 133.1, 149.0, 153.7. ESI MS: m/z 1043.6, calculated: 1043.5 for [M+H]⁺.

Chemosensor 2 was prepared starting from rhodamine B hydrazide (4) (1.5 g, 3.3 mmol), benzene-1,3,5-tricarbonyl trichloride (5) (0.28 g, 1.1 mmol) in THF (5 mL) and triethylamine (0.35 g, 3.5 mmol) in THF (70 mL). The crude product was recrystallized from a mixture of diethyl ether and dichloromethane (3:1) to afford the pure product (1.02 g, 61%) as a solid. mp 204 °C–206 °C. ¹H NMR: δ 1.05–1.24 (m, 36H, CH₃), 3.19–3.39 (m, 24H, CH₂), 6.21–6.41(m, 12H, ArH), 6.59–6.76 (m, 6H, ArH), 7.12 (d, J = 6.0 Hz, 3H, ArH), 7.38–7.56 (m, 6H, ArH), 7.88–8.01 (m, 6H, ArH). ¹³C NMR: δ 12.6, 44.2, 98.0, 99.9, 103.9, 107.9, 123.5, 124.2, 127.9, 129.2, 129.8, 132.9, 149.0, 153.6. ESI MS: m/z 1525.7, calculated: 1525.7 for [M+H]⁺.

3.4. Absorption and Fluorescence Measurements

Absorption and fluorescence titrations were performed in a 1 cm quartz cell by addition of small aliquots of metal ion work solutions to a 3 mL solution of proper amounts of 1 and 2 in CH₃CN/H₂O (3/1, v/v) and a 3 mL solution of a proper amount of 1 in CH₃CN/HEPES buffer (0.02 M, pH = 7.0) (95/5, v/v). After thorough mixing, the solutions were allowed to stand at ambient temperature for 5 min, whereupon absorption or fluorescence spectra were recorded. Both the excitation and emission slits were 5 nm.

To determine the optimal conditions for Hg²⁺ detection, small aliquots of Hg²⁺, Cu²⁺, and Fe³⁺ work solutions were respectively added into 5 μM chemosensor 1 solutions which contained different fractions (5%, 10%, 15%, and 20%) of HEPES buffer (0.02 M, pH = 7.0) in CH₃CN and mixed in a 1 cm quartz cell for 5 min. Then the fluorescence measurement was performed at ex/em = 510/580 nm.

4. Conclusions

In conclusion, the rhodamine derivatives 1 and 2 are very good fluorescent chemosensors, with a good selectivity toward Hg²⁺ and Fe³⁺ over other competitive ions in CH₃CN/H₂O (3/1, v/v). For practical Hg²⁺ detection, the experimental conditions were optimized to CH₃CN/HEPES buffer (0.02 M, pH = 7.0) (95/5, v/v). Under these conditions, the fluorimetric quantification of Hg²⁺ by 1 was satisfactory in a linear working range of 0–50 μM, with a detection limit of 0.4 μM Hg²⁺ and a pH span of 4–10. These very good features make chemosensor 1 very promising for practical applications.

Supplementary Information

ijms-13-16822-s001.pdf