Remarkable role of positional isomers in the design of sensors for the ratiometric detection of copper and mercury ions in water

Abstract

Cation sensing properties of the three positional isomers of rhodamine based sensors (1–3) are studied in water. The sensors differ only in the position of pyridine's nitrogen. The chemosensor 1, with pyridine nitrogen at ortho-position, showed a selective colorimetric detection of Cu(II) ions in water, at physiological pH 7.4 and also in medium containing BSA (bovine serum albumin) and blood serum. Notably the compound 2 and 3, with pyridine end located at meta- and para-positions did not show any color change with Cu(II) ions, although both the compounds showed turn-on change both in color and fluorescence with Hg(II) ions specifically. All the probes showed ratiometric changes with the specific metal ions. The changing position of nitrogen also changed the complexation pattern of the sensors with the metal ions. Probe 1 showed 2 : 1 complexation with Cu(II), whereas 2 and 3 showed 1 : 1 complexation with Hg(II) ions. The mechanism investigation showed that the change in color upon addition of metal ions is due to the ring-opening of the spirolactam ring of the probes. Cu(II) interacted with ligand 1 through a three-point interaction mode comprising carbonyl oxygen, amido nitrogen and pyridine nitrogen end. But in case of 2 and 3, Hg²⁺ only interacted through pyridine nitrogen ends. Quantitative estimation of Cu²⁺ and Hg²⁺ in complex biological media such as bovine albumin protein (BSA) and human blood serum were performed using these sensors. Rapid on-site detection as well as discrimination of these toxic ions was demonstrated using easily prepared portable test-strips.

---

Introduction

The detection of transition metal ions has gained much importance in recent years due to their widespread applicability in various fields and also in terms of their hazard as environment pollutants.^1–5 Among the various transition metals, some of them are crucial for living organisms. Copper is one such important transition metal ion, as it is the third most abundant transition metal ion in the human body. It plays important roles in various biological processes. It is essential for functioning of various metalloenzymes, including superoxide dismutase, cytochrome c oxidase, and tyrosinase.^6,7 But the amount of Cu²⁺ ion required for all the physiological functions is very limited and an excess of this ion becomes highly toxic.^8,9 Thus the permitted limit of copper in drinking water is only <1.3 ppm according to the United States Environmental Protection Agency (EPA).¹⁰ An excess of copper in the human body causes gastrointestinal disorders and several neurodegenerative diseases such as Alzheimer's disease, Wilson's disease and Menkes' kinky hair syndrome etc.^11,12

Mercury ion is another highly toxic transition metal ion. It is carcinogenic and highly toxic.¹ The standard for the maximum allowable level of inorganic Hg in drinking water is only 2 ppb according to the United States Environmental Protection Agency (EPA).^11,12

Rhodamine based sensors have gained much popularity in the last few years due to their strikingly different properties in the closed spirolactam ring and open-ring form. In the closed spirolactam form, the central quaternary carbon of lactone ring breaks the π-conjugation of xanthene moiety and renders nearly colorless and non-fluorescent entity. On the other hand, the opening of this spirolactam ring extended the π-conjugation which leads enhancement of color and fluorescence emission in the visible region. Exploiting this idea in 1997, Czarnik et al. first reported Rhodamine based chemosensor.¹³ 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.^14,15

Many rhodamine based sensors have been reported for the detection of the Cu²⁺ ion^13,16–24 as well as for the Hg²⁺ ion.^{14,15,25–33} Some of the reports have shown the detection both Cu²⁺ and Hg²⁺ ion.^34–36 But most of the reports have shown the detection only in aqueous–organic mixture. So far very few instances are known which report the sensing in complete water or in aqueous–organic medium with water as a major solvent.^37–42 The response of rhodamine based sensors are known to depend upon the choice of solvent system and nature of spacer between binding site and signalling unit.^14,15 But till date no effort has been made to investigate the effect of different positional isomers on metal ion selective binding ability of rhodamine based sensors.

Earlier we have developed sensors and functional assemblies for the detection of various analytes.^43–47 In this work we have investigated the sensing properties of the three positional isomers based on rhodamine as signalling moiety (1–3). The three isomers differ in position of nitrogen end of the pyridine moiety. The probe 1 has been reported earlier for the detection Hg²⁺ and Cu²⁺ ion in mixtures of different aqueous–organic medium.^48,49 However, no effort has been made to explore the properties of this sensor in pure water and also in biological samples. The properties of isomeric 2 and 3 are however not known in any medium as they have not even been investigated earlier. Notably by changing the position of the nitrogen atom in the sensors, it has completely changed the sensing properties of these probes. The sensor 1, containing pyridine nitrogen at the ortho-position showed a visible colorimetric and ratiometric detection of Cu²⁺ ion in natural water as well as in buffer solution at physiological pH 7.4. In contrast, the other two probes 2 and 3 possessing pyridine nitrogen at meta- and para-position respectively did not show any evidence of interaction with Cu²⁺ ion. However, both of these compounds showed selective interaction with Hg²⁺ ion in water. Both 2 and 3 showed visible colorimetric and ratiometric change with the Hg²⁺ ion. Also these two sensors showed ‘turn-on’ detection of the Hg²⁺ ion under fluorescence spectroscopy.

Results and discussion

Synthesis

All the three sensors were prepared by simple strategy as shown in Scheme 1. Rhodamine hydrazone was synthesized from rhodamine-6G, which was further coupled with the pyridine-aldehydes (ortho-, meta- and para-) to get the compounds 1, 2 and 3 respectively.

Scheme 1 Synthetic route to 1, 2 and 3.

Colorimetric and UV-vis spectral response

The sensing properties of all the three sensors towards various transition metal ions were investigated in water. The solutions of all the three probes were colorless in water to begin with.

Sensing properties of 1

Probe 1 showed a prominent color change from colorless to bright pink selectively upon addition of Cu²⁺ (inset of Fig. 1).

Fig. 1 Absorption spectral changes of 1 (10 μM) in water upon addition of 2 equiv. of different salts of Ag⁺, Cd²⁺, Co²⁺, Cu²⁺, Fe²⁺, Fe³⁺, Hg²⁺, Mg²⁺, Mn²⁺, Ni²⁺, Pb²⁺ and Zn²⁺ [inset: color change of sensor 1 (50 μM) in water upon the addition of Cu²⁺ (2 equiv.)].

Addition of other cations (Ag⁺, Cd²⁺, Co²⁺, Fe²⁺, Fe³⁺, Hg²⁺, Mg²⁺, Mn²⁺, Ni²⁺, Pb²⁺ and Zn²⁺) did not induce any change in visible color. The UV-vis absorption spectra of 1 were recorded individually upon addition of various cations. The absorption spectrum of 1 showed two bands with the absorption maxima at 244 and 310 nm. The addition of Cu²⁺ ion to the solution of 1 resulted in the appearance of a new band with the absorption maximum at 530 nm with a concomitant decrease in the absorbance at 310 nm (Fig. 1). The addition of other cations however, did not show any such change in the absorption spectra. Further the UV-vis titration was performed upon gradual addition of Cu²⁺ ion. The titration showed progressive increases in the absorption maxima at 530 nm with concomitant decreases at 310 nm (Fig. 2). The titration curve showed two isosbestic points at 340 and 255 nm, indicating the existence of an equilibrium between 1 and 1-Cu²⁺ complex (Fig. 2). Further when the absorbance ratio at 529 and 308 nm was plotted with the addition of Cu²⁺ ion, a linear response with the added Cu²⁺ ion was observed (inset of Fig. 2). This showed that the probe 1 can be used as a ratiometric sensor for the detection of Cu²⁺ ion in pure water.

Fig. 2 UV-vis titration of 1 (10 μM) in water with Cu²⁺ (0 to 30 μM) [inset: ratiometric plot with added equiv. of Cu²⁺ at the absorbance ratio of 529 and 308 nm].

Interference and reversibility

The detection of Cu²⁺ ion was also checked in presence of the other ions. Cu²⁺ ions (2 equiv.) was added to a solution of the sensor 1 (10 μM) in water in the presence of excess of other cations (10 equiv.). No interference was observed in the Cu²⁺ ion detection due to the presence of other cations. The addition of Cu²⁺ resulted in a >40-fold increase in the absorbance even in the presence of excess of other cations (Fig. S1, ESI†). The reversible nature of binding of the Cu²⁺ ion was also checked using EDTA. First Cu²⁺ ion (5 equiv.) was added to 1 and to that the same equivalent of EDTA was added, which completely revived the original spectrum of 1 (Fig. S2, ESI†). This was repeated several times and each time the revival of the original spectrum of the sensor was observed after the EDTA addition. This showed that the single solution of the probe can be used multiple times for the detection of Cu²⁺ ion.

Further sensing property of 1 towards various cations was investigated at physiological pH 7.4 (Fig. S3, ESI†). At this pH also, the probe 1 remained colorless with the two bands with absorption maxima at 241 and 310 nm. Upon addition of Cu²⁺ ion, the probe 1 showed the emergence of a new band at 530 nm as was observed in pure water. The selectivity of the sensor was checked by adding various other cations to the probe solution. It did not show any change with other cations, which indicated that it remains selective towards the Cu²⁺ ion. Further the plot of absorbance at 530 nm with various cations showed that the addition of Cu²⁺ ion resulted in a 5-fold increase in the absorbance compared to other cations (Fig. S3b, ESI†). The titration was performed with Cu²⁺ ion which showed a linear increase in the absorbance at 530 nm against the added Cu²⁺ ion (Fig. S3d, ESI†).

Thus the sensor 1 showed the ‘turn-on’ detection of Cu²⁺ in water without any interference from the other cations.

Sensing properties of 2 and 3

After checking the sensing property of sensor 1, we investigated the sensing properties of sensor 2 and 3 in water. Notably none of these two sensors showed any change with the Cu²⁺ ion. But both 2 and 3 showed perceptible change with the immediate addition of Hg²⁺ ion. Both the sensors showed immediate visual color change from colorless to pink upon addition of Hg²⁺ (Fig. 3). The change was monitored by UV-vis spectroscopy and the addition of Hg²⁺ resulted into the emergence of a new band at 532 nm absorption maximum (Fig. 3).

Fig. 3 Absorption spectral changes of sensor (a) 2 (10 μM) and (b) 3 (10 μM) in water upon addition of 5 equiv. of different salts of Ag⁺, Cd²⁺, Co²⁺, Cu²⁺, Fe²⁺, Fe³⁺, Hg²⁺, Mg²⁺, Mn²⁺, Ni²⁺, Pb²⁺ and Zn²⁺. [Inset: color change of the sensor 2 or 3 (50 μM) in water upon the addition of Hg²⁺ (5 equiv.)].

The selectivity was checked with the addition of other cations. Importantly none of the two probes showed any change with the addition of other cations (Fig. 3). Both the probes showed >10-fold enhancement in the absorbance at 532 nm with the addition of Hg²⁺ compared to the addition of other cations. Further the titration of both the sensors 2 and 3 were performed with the gradual addition of Hg²⁺ ion. Both 2 and 3 maintained isosbestic points in titration with the Hg²⁺ ions (Fig. S4, ESI†). This showed the formation of equilibrium between each of 2 or 3 and their corresponding metal complexes. The titration resulted in the gradual increase in absorbance at 532 nm with the concomitant decrease at ∼310 nm. Both sensors showed ratiometric changes with the added Hg²⁺ ion (Fig. S4, ESI†).

Interference and reversibility

The interferences of other cations were checked for the detection of Hg²⁺ ion. Hg²⁺ (4 equiv.) were added to the sensor solutions (both 2 and 3) in the presence of excess of other cations (12 equiv.). None of the other cations showed any interference towards the detection of Hg²⁺ ion (Fig. S5a, ESI†).

Further the applicability of the sensor for the detection of Hg²⁺ multiple times was checked using EDTA. Both the sensors showed complete reversal of the probe's absorbance with the addition of EDTA and showed that the sensors (2 and 3) can be used multiple times for the detection of the Hg²⁺ ion (Fig. S5b, ESI†).

pH dependence of metal ion sensing of the probes

The pH dependence of all the sensors in aqueous media was also checked by UV-vis spectroscopy. First Cu²⁺ ion was added to the buffered solution of the 1 at different pH values. The results indicated that the binding of 1 with the Cu²⁺ occurred effectively above pH 5 (Fig. S6a, ESI†). Below pH 5, no change was observed in the absorption spectra due to the addition of Cu²⁺ ion.

Further the pH dependence of 2 and 3 were also checked for the detection of Hg²⁺ ion. Both the sensors showed effective detection of Hg²⁺ ion above pH 5. But above pH 7, the change in the absorbance due to the addition of Hg²⁺ was very little (Fig. S6b and c, ESI†). Thus the effective range of pH for the Hg²⁺ ion detection using 2 and 3 is from pH 5 to pH 7.

Emission spectral response

All the three probes showed very weak fluorescence response in pure water.

Emission spectral response of 1

The emission spectrum of sensor 1 was recorded upon excitation at 520 nm. It showed a very weak fluorescence emission (Fig. S7, ESI†). To the probe solution, various cations were added and the emission spectra were recorded. The addition of Cu²⁺ ion resulted in a complete quenching of the emission intensity of 1. The changes in the fluorescence emission intensity of 1 due to the addition of other cations were however, negligible compared to the Cu²⁺ ion. This indicated that sensor 1 may be also used for the detection of Cu²⁺ ion selectively by emission spectroscopy.

Emission spectral response of 2 and 3

The emission spectral responses of 2 and 3 were also checked upon addition of various cations. Both 2 and 3 showed the immediate visual change from no color to bright yellow fluorescence under a UV-lamp upon addition of Hg²⁺ ion (Fig. 4). The emission spectral responses were recorded by fluorescence spectroscopy. Both of the sensors were excited at 520 nm. The addition of Hg²⁺ to 2 resulted in >50-fold enhancement of fluorescence intensity at 560 nm with a quantum yield 0.22.⁴⁹ However with 3, the addition of Hg²⁺ showed similar enhancement in the fluorescence intensity at 560 nm (Fig. 4) with a quantum yield of 0.25.

Fig. 4 Fluorescence emission spectra of the sensor (a) 2 (5 μM) or (b) 3 (5 μM) in water (λ_ex = 520 nm) in the presence of various cations Ag⁺, Cd²⁺, Co²⁺, Cu²⁺, Fe²⁺, Fe³⁺, Hg²⁺, Mg²⁺, Mn²⁺, Ni²⁺, Pb²⁺ and Zn²⁺ (5 equiv.). [Inset: change in the fluorescence under UV-lamp (365 nm) with the addition of Hg²⁺ ion (5 equiv.)].

The addition of other cations did not show any change in the fluorescence intensity. The titrations were performed with the gradual addition of Hg²⁺ ion to both 2 and 3. The addition of Hg²⁺ resulted in an increase in the emission intensity at 560 nm (Fig. S8, ESI†). Thus both 2 and 3 can be used as turn-on sensors for the detection of Hg²⁺ ion in water.

Detection limit determination

The detection limits for the metal ions were estimated.⁵³ The detection limit of Cu²⁺ ion was determined from the UV-vis titration of sensor 1. It showed a detection limit of ∼12.6 ppb. The detection limit of Hg²⁺ ion was determined from the emission titration spectra. Sensor 2 and 3 showed detection limit of ∼9.4 and ∼4 ppb respectively for Hg²⁺ ion.

Stoichiometry determination and binding constant

The stoichiometries of interaction of all the three sensors with the chosen metal ion were investigated by Job plot analysis. The Job plot for 1 with Cu²⁺ ion showed evidence of 2 : 1 interaction between 1 and Cu²⁺ (Fig. S9a, ESI†). The binding constant was calculated with the help of Benesi–Hildebrand equation for 2 : 1 stoichiometry (Fig. S9b†). 1 showed a very strong binding with Cu²⁺ with a binding constant of log K = 8.54 ± 0.01.

Further we determined the stoichiometry of interaction of both 2 and 3 with the Hg²⁺ ion. Both the probes showed 1 : 1 stoichiometry with Hg²⁺ ion (Fig. S10a, ESI†). The binding constants were calculated for both 1 and 2 with Hg²⁺ ion using Benesi–Hildebrand equation for 1 : 1 stoichiometry (Fig. S10b, ESI†). Both the probes 2 and 3 showed good binding towards the Hg²⁺ ion with binding constant of log K = 3.7 ± 0.01 and 3.91 ± 0.01 respectively.

Furthermore we have also calculated the selectivity coefficients of all the three probes towards the metal ions in terms of the relative enhancement in absorbance of probe (Table 1, and Fig. S11, ESI†).

Table 1 Selectivity coefficients of the probes for various cations^a

Metal ion	Compound 1	Compound 2	Compound 3
a Based on the relative enhancement in absorbance of probe upon addition of metal ions.
Ag^+	5.55	9.31	6.09
Cd^2+	2.63	7.59	5.19
Co^2+	7.81	8.70	10.72
Cu^2+	100.00	8.70	8.47
Hg^2+	7.07	100.00	100.00
Mg^2+	5.75	4.90	3.11
Mn^2+	5.55	7.32	5.96
Ni^2+	8.14	8.57	4.10
Pb^2+	7.64	7.96	8.07
Zn^2+	8.19	4.91	10.46
Fe^2+	7.65	6.23	5.89
Fe^3+	6.82	6.92	5.83

---

Mechanism investigation: effect of different positional isomers

The interaction of rhodamine based sensors with metal ions either result into ion-induced spirolactam ring opening or ion-catalyzed hydrolysis reaction. The above mentioned EDTA recovery experiment clearly indicated that observed change is due to reversible spirolactam ring opening and not due to ion induced hydrolysis. This was further evident from the ESI mass spectra of probes in presence of corresponding metal ions. ESI mass spectra of 1 + Cu²⁺ showed peak at 580.5 corresponding to 1 : 2 metal complex of 1 with Cu²⁺ (Fig. S12, ESI†). Similarly ESI mass spectra of 2 + Hg²⁺ and 3 + Hg²⁺ also exhibited sharp peaks with proper isotopic distribution corresponding to 1 : 1 complexes (Fig. S13 and 14, ESI†).

To further examine the mode of interaction between the probe and the metal ion, we have performed ¹H NMR titration of the sensors with the metal ions. The titration of 1 could not be performed with Cu²⁺ ion, as the addition of Cu²⁺ ion to the solution of 1 completely quenched all the peaks, due to the paramagnetic nature of the Cu²⁺ ion.

Further both 2 and 3 showed similar interactions and stoichiometry with Hg²⁺ ion. Taking into account of this we performed ¹H NMR titration of 2 with Hg²⁺ ion to explore the mechanism of complexation (Fig. 5). Upon addition of Hg²⁺ to the probe solution, the pyridine protons (H_b, H_c, H_d and H_e) experienced greater extent of downfield shift compared to the protons present near the vicinity of carbonyl group (H_f). This indicated that the Hg²⁺ interacted with probes through pyridine nitrogen end rather than carbonyl of spirolactam ring. The resonance signal of the imine proton (H_j) showed a downfield shift of 0.06 ppm and broadened. This indicates the involvement of imine proton in opening of the spirolactam ring.

Fig. 5 ¹H NMR titration of 2 (8 mM) in DMSO d₆ with Hg²⁺ [0, 0.25, 0.5, 0.75 and 1 equiv. (1–5)].

To confirm this further, we have recorded the FT-IR spectra of 2 both in absence and presence of Hg²⁺ ion. The addition of Hg²⁺ did not show any change in carbonyl (C=O) stretching frequency upon complexation with Hg²⁺ (Fig. S15, ESI†). The non-involvement of carbonyl oxygen in coordination with Hg²⁺ has also been reported by Duan et al.⁴⁸

Furthermore for Cu²⁺ ion to interact with rhodamine based probes, three-point interaction mode has been reported⁵⁴ comprising the oxygen atom of carbonyl, nitrogen atom of amido, and another coordinating atom of adjacent ligand. Czarnik et al. have shown the interaction of Cu²⁺ towards the carbonyl oxygen and amido nitrogen of the rhodamine based probe.¹³ In addition Zhang et al. and Lee et al. have independently demonstrated that the third coordination site required for the binding with Cu²⁺ in rhodamine type probes can be pyridine or picolyl nitrogen.^54,55

Now, as the compound 1 possesses a rigid C=N bond, the tridentate ligand model may not be favorable for the binding with Cu²⁺. That could be the reason for Cu²⁺ induced 1 : 2 complexation which enhances the flexibility in the system to achieve appropriate geometry. To check the involvement of carbonyl oxygen (C=O) in the complexation, the FT-IR spectra were recorded for both 1 and 1-Cu²⁺. The addition of Cu²⁺ resulted into ∼49 cm⁻¹ shift in C=O stretching frequency (Fig. S16, ESI†). This was further confirmed by the synthesizing analogous molecule 4, where the instead of pyridine, phenyl ring was substituted (see ESI† for details). This compound did not show any change with Cu²⁺ ions. This confirmed the involvement of pyridine's nitrogen in interaction with Cu²⁺ ion.

Apart from the experimental evidences, DFT calculations were performed with all the probes using B3LYP/6-31G* basis set. All possible configurations of the probes were made and screened for the low energy conformers by semi-empirical PM3 method. Then most stable conformers were selected by B3LYP/6-31G* method. The distance between carbonyl oxygen and pyridine nitrogen ends were found to be 2.85, 3.84, 4.64 Å for probes 1, 2, and 3 respectively (Fig. 6). The presence of a rigid backbone with distant pyridine ends made probes 2 and 3 incapable of providing suitable geometry for Cu²⁺ even with 1 : 2 complexation. Instead they can easily form linear complex with Hg²⁺ through pyridine ends. Thus in 1, the pyridine nitrogen which is in close vicinity with the carbonyl oxygen, enables it to form more flexible 1 : 2 complexes with Cu²⁺ to fulfill the required tridentate mode of ligand geometry. This was further supported from energy minimized structures (using B3LYP/LANL2DZ method) of the corresponding metal complexes (Fig. S17, ESI†). The structures of metal complexes showed that Cu²⁺ form complex with 1 comprising carbonyl oxygen, amido nitrogen and pyridine nitrogen end. However, with 2 and 3, Hg²⁺ could only interact through pyridine's nitrogen ends.

Fig. 6 DFT optimized structure of 1, 2 and 3 using B3LYP functional and 6-31G* basis set.

Based on the above a plausible mechanism of interaction may be proposed as shown in Scheme 2.

Scheme 2 Sensing chemistry of the three probes (1–3) in pure water.

Detection of ions in real life water samples

We further checked the applicability of these sensors for the detection of the Cu²⁺ and Hg²⁺ ions in the real life water samples without any prior purification.

We have employed these probes to detect Cu²⁺ or Hg²⁺ ions from diverse sources such as tap water, sea water and swimming pool. Emission spectra of probes 1 and 3 were recorded immediately upon addition of different amount of metal ions to these water samples. Addition of Cu²⁺ induced significant quenching in emission intensity of 1 in all the three water samples. Emission intensity varied linearly with the added Cu²⁺ ion (0 to 150 ppb), which made the system suitable for the estimation of Cu²⁺ in real life water samples (Fig. S18, ESI†). As both 2 and 3 showed similar changes with Hg²⁺ ion, we herein chose 3 as it showed greater increase in emission intensity and better quantum yield. The addition of Hg²⁺ resulted in increase in the emission intensity of 3 as expected. In all the three water samples significant fluorescence changes were observed with a good linear response (Fig. S19, ESI†) upon addition of Hg²⁺ (0 to 40 ppb). Significantly low level of metal ion could be detected using these probes which ensured the effective analysis of Cu²⁺ or Hg²⁺ in natural water sources without any potential interference. The estimated detection limits have been tabulated in Table 2. This shows that the probes may indeed be used for the detection of above mentioned metal ions in real water samples.

Table 2 Detection limit of sensors (1 and 3) for Cu²⁺ and Hg²⁺ in various water sources^a

	System	Linear response (r^2)	Detection limit (in ppb)
a Sources.b Bangalore city water supply.c From a nearby swimming pool.d From Arabian sea.
1 with Cu^2+	In tap water^b	0.979	6.7 ± 0.02
	In pool water^c	0.973	4.0 ± 0.04
	In sea water^d	0.981	5.8 ± 0.03
3 with Hg^2+	In tap water^b	0.996	5.2 ± 0.03
	In pool water^c	0.980	5.6 ± 0.02
	In sea water^d	0.998	2.5 ± 0.03

---

Fast track detection of Cu(II) and Hg(II) using test strips

It is a great challenge to ensure purity of the drinking water and consumable food stuffs in remote places where laboratory facilities are not available. Therefore portable test strips were prepared for rapid on-site detection of Cu²⁺ and Hg²⁺ ions, as they did not require any sophisticated facilities. For this purpose filter-paper strips were first soaked in CHCl₃ solution of sensors (1 and 3) and then dried in air. After that they were dipped in the corresponding metal ion solutions (1 equiv.) in water. A distinct visible color change was observed immediately upon dipping the test strips in the metal ion solutions (Fig. 7). A notable change was also observed under UV light. Presence of Cu²⁺ selectively changed the test-strips color to pink in case of 1 whereas; Hg²⁺ did the same when it was soaked with 3. Therefore this newly developed sensor systems provide an alternative method to confirm the nature (whether it is due to Cu²⁺ or Hg²⁺ contamination) as well as extent of metal ion induced toxicity in natural water sources. This makes the probes quite useful for rapid on-site detection of metal ions.

Fig. 7 Photographs of the test strips made from 1 and 3 (100 μM) for the detection of Cu²⁺ and Hg²⁺ in water both under day light and under UV lamp (365 nm).

A comparison of the applicability and analytical part of the present probes with some of the previous reports in terms of their water solubility, detection limit and specific application is shown in Tables 3 and 4 below. Thus these probes can detect both the metals ions Cu²⁺ and Hg²⁺ in water. Accordingly we have used the present probes to check their applicability for the detection of above ions in certain biological media.

Table 3 A comparison of properties of various rhodamine based probes for the detection of Cu²⁺ ion

Ref.	Medium	Detection limit (nM)	Application probed
a J. Am. Chem. Soc., 1997, 119, 7386.b Org. Lett., 2006, 8, 2863.c Org. Lett., 2009, 11, 4442.d Org. Lett., 2010, 12, 3852.e Analyst, 2009, 134, 1826.f Org. Biomol. Chem., 2010, 8, 5277.g Sens. Actuators, B, 2009, 141, 506.h Chem.–Eur. J., 2008, 14, 6892.i Sens. Actuators, B, 2009, 135, 625.j Present work.
^a	HEPES–CH3CN (4 : 1), Incubation time = 2 min	—	—
^b	Tris–CH3CN (1 : 1)	25	—
^c	HEPES–CH3CN (6 : 4)	—	—
^c	HEPES–CH3CN (1 : 1)	—	—
^d	PBS buffer, Incubation time = 10 min	37	Cell imaging
^e	MeOH–Water (8 : 2)	3	—
^f	EtOH–water (8 : 1)	150	—
^g	CH3CN–water (3 : 7), Incubation time = 1 min	50	Cell imaging
^h	CH3CN–water (3 : 7)	40	In tap water
^i	Water, Incubation time = 7 min	200	—
Probe 1^j	Water (DMSO < 0.5%)	63	In real life water samples and also using test strips

---

Table 4 A comparison of properties of various rhodamine based probes for the detection of Hg²⁺ ion

Ref.	Medium	Detection limit (nM)	Application probed
a Org. Lett., 2010, 12, 476.b J. Org. Chem., 2008, 73, 8571.c Chem. Commun., 2009, 4417.d Org. Biomol. Chem., 2010, 8, 4143.e J. Org. Chem., 2009, 74, 2167.f Org. Biomol. Chem., 2010, 8, 3618.g Org. Biomol. Chem., 2010, 8, 4819.h Org. Lett., 2006, 8, 859.i Anal. Chim. Acta, 2010, 663, 85.j Present work.
^a	EtOH–water (1 : 1)	4.5	Cell imaging
^b	CH3CN	—	—
^c	20% EtOH–H2O	—	Cell imaging
^d	MeCN–water (95 : 5)	10	Cell imaging
^e	MeCN–water (15 : 85)	—	—
^f	2.5% CH3CN in water	27.5	Cell imaging
^g	Aqueous ethanol	1.7	Cell imaging
^h	Aqueous solution of pH 3.4	100	—
^i	EtOH–water (1 : 1)	40	Detection shown in tap and river water samples
Probe 2^j	Water (DMSO 0.5%)	47	Detection shown in real life water samples and also using test strips
Probe 3^j		20

---

Response in different biological media

In order to investigate the applicability of the probes towards the detection of metal ions in biological media the titrations were performed in various biological samples. The serum albumins (human, bovine) are known to have specific metal ion binding sites, and act as a metal ion transport protein in blood plasma.⁵⁶ In order to investigate the sensitivity of the probes towards Cu²⁺ and Hg²⁺ ions in presence of bovine serum albumin (BSA), the titrations of 1 and 3 were performed with Cu²⁺ and Hg²⁺ respectively in BSA.

Even in this highly concentrated BSA medium (0.1 mg mL⁻¹), the sensor 1 showed effective detection of Cu²⁺ (Fig. S20a, ESI†). A linear change in the absorbance at 562 nm was observed upon the addition of Cu²⁺ (Fig. S20b, ESI†). Similarly, titration of 3 with Hg²⁺ in presence of BSA exhibited “turn-on” response. A linear change in emission at 560 nm band clearly indicated high sensitivity of the probe towards Hg²⁺ even in presence of BSA (Fig. S21, ESI†). On the other hand a complex system, human blood serum contains different electrolytes, antibodies, antigens, and hormones. We further investigated the efficiency of the probes towards the detection of Cu²⁺ and Hg²⁺ ions in high serum condition (10%). In this condition also, the sensitivity of the probes found equally good to detect metal ions (Fig. S22, ESI†). The control experiments were also performed to ensure that the probe molecules did not interact with the BSA or the components of human blood serum.

Conclusions

In conclusion, we have investigated the sensing properties of three rhodamine based positional isomers (1–3) in pure water. The three sensors differ only in the position of pyridine nitrogen. The extensive studies showed that the variation only in the position of the pyridine nitrogen end can significantly change the sensing properties of molecules in terms of selectivity towards the metal ions. Sensor 1, which contains the pyridine nitrogen at the ortho-position showed selective sensing toward Cu²⁺ ion in 1 : 2 fashion, whereas 2 and 3 specifically interact with Hg²⁺ in 1 : 1 stoichiometry. In all the cases the spectral changes were observed due to reversible spirolactam ring opening. Compound 1 was found to interact with Cu²⁺ through a well known tridentate ligand model comprising carbonyl oxygen, amido nitrogen and pyridine nitrogen. However, in 2 and 3 pyridine nitrogen ends were only involved in linear complexation with Hg²⁺. The stoichiometry and mode of interaction were found to be dependent on both the flexibility of ligands and nature of the metal ions investigated. The sensitivity of metal ion recognition by probe 1 and 3 was also investigated in the presence of excess of plasma protein (BSA) and human blood serum. The detection as well as discrimination of these (Cu²⁺ and Hg²⁺) toxic metal ions was also achieved in different natural water sources at ppb level. In addition we have also demonstrated the rapid on-site detection of these ions using portable test strips.

Experimental

Materials

All solvents and reagents were purified and dried by usual methods. All starting materials were obtained from the best known commercial suppliers and used as received. The stock solutions of the compounds (1–3) were made in dimethyl sulphoxide (DMSO) and the final concentration of DMSO in all the studies were less than 0.5%.

Instrumentation

¹H and ¹³C NMR were measured on a 400 MHz/100 MHz Bruker Advance DRX 400 spectrometer. HRMS analyses were performed with Q-TOF YA263 high resolution instrument. IR spectra were recorded on Perkin Elmer FT-IR spectrum BX. UV-vis absorption spectra were obtained on a Shimadzu UV-2100 spectrophotometer. Fluorescence spectra were recorded on a Cary-Eclipse spectrofluoriphotometer.

The fluorescence quantum yield was calculated by using rhodamine 6G (Φ = 0.94 in EtOH) as a reference.⁵⁰ And the quantum yield is calculated using the equation

Φ_unk = Φ_std[(I_unk/A_unk)/(I_std/A_std)](η_unk/η_std)²

where, Φ_unk and Φ_std are the radiative quantum yields of the sample and standard, I_unk and I_std are the integrated emission intensities of the corrected spectra for the sample and standard, A_unk and A_std are the absorbances of the sample and standard at the excitation wavelength, and η_unk and η_std are the indices of refraction of the sample and standard solutions, respectively.

Synthesis of rhodamine hydrazone (4). Rhodamine-6G hydrazone (4) was prepared according to literature protocol.^51,52 In a 50 mL round bottomed flask, rhodamine-6G (5) (0.48 g, 1 mmol) was dissolved in 15 mL ethanol. To that 1.5 mL (excess) hydrazine monohydrate (85%) was added drop-wise with vigorous stirring at room temperature. After the addition, the stirred mixture was refluxed for 2 h, and then cooled overnight. The resulting precipitate was filtered and washed 3 times with 10 mL EtOH–water. After drying under vacuum, the reaction afforded rhodamine-6G hydrazone. Yield 0.37 g, 80%; IR (neat, cm⁻¹) 3370.6, 2921.5, 1621.4, 1516.6, 1270.3, 1203.6, 1017.8, 742.4; ¹H NMR (400 MHz, CDCl₃) δ (ppm) 1.32 (t, J = 7.1 Hz, 6H), 1.92 (s, 6H), 3.22 (q, J = 6.9 Hz, 4H), 3.56 (br, 4H), 6.26 (s, 2H), 6.39 (s, 2H), 7.06 (t, J = 3.3 Hz, 1H), 7.45 (t, J = 4.6 Hz, 2H), 7.96 (t, J = 3.2 Hz, 1H); HRMS m/z calcd for C₂₆H₂₈N₄O₂ (M + Na)⁺ 451.2110, found 451.2106.

Synthesis of 1, 2 and 3. Rhodamine hydrazone (47 mg, 0.11 mmol) was mixed with the corresponding aldehydes 1, 2 and 3 (12 mg, 0.11 mmol) in 4 mL of methanol. To each of these, 2–3 drops of glacial acetic acid was added and stirred at rt for 4–5 h. After completion of the reaction, the precipitate was filtered and washed with methanol.

Compound 1. White solid; (yield: 35 mg, 70%); IR (neat, cm⁻¹); 3381.1, 2968.9, 1619.0, 1517.0, 1199.7, 742.2, 688.4; ¹H NMR (400 MHz, CDCl₃) δ (ppm) 1.30 (t, J = 7.2 Hz, 6H), 1.87 (s, 6H), 3.20 (q, J = 7.2 Hz, 4H), 6.35 (s, 2H), 6.41 (s, 2H), 7.05 (q, J = 7.6 Hz, 1H), 7.12 (t, J = 6.8 Hz, 1H), 7.46–7.48 (m, 2H), 7.60 (t, J = 7.6 Hz, 1H), 7.99–8.04 (m, 2H), 8.21 (s, 1H), 8.45 (s, 1H); ¹³C NMR (100 MHz, CDCl₃) δ (ppm): 14.7, 16.6, 38.2, 65.6, 97.0, 105.7, 117.9, 120.5, 123.5, 123.6, 127.2, 127.6, 128.2, 133.8, 136.0, 145.3, 147.5, 148.9, 151.0, 152.7, 154.3, 165.5; HRMS m/z calcd for C₃₂H₃₁N₅O₂ (M + H)⁺ 518.2556, found 518.2558; anal. calcd for C₃₂H₃₁N₅O₂: C, 74.25; H, 6.04; N, 13.53%. Found: C, 73.94; H, 6.11; N, 13.53%.

Compound 2. White solid; (yield: 38 mg, 75%); IR (neat, cm⁻¹); 3365.9, 2917.5, 1698.0, 1522.7, 1269.6, 1217.5, 1014.1, 747.9; ¹H NMR (400 MHz, CDCl₃) δ (ppm): 1.32 (t, J = 7.1 Hz, 6H), 1.87 (s, 6H), 3.22 (q, J = 6.9 Hz, 4H), 3.49 (br, 2H), 6.31 (s, 2H), 6.41 (s, 2H), 7.09 (q, J = 3 Hz, 1H), 7.18 (q, J = 4.9 Hz, 1H), 7.50 (t, J = 3.7 Hz, 2H), 7.96 (d, J = 7.9 Hz, 1H), 8.01–8.04 (m, 1H), 8.45 (d, J = 4.7 Hz, 1H), 8.52 (s, 1H), 8.55 (s, 1H); ¹³C NMR (100 MHz, CDCl₃) δ (ppm): 14.6, 16.6, 38.2, 66.0, 95.6, 105.9, 117.8, 123.2, 123.4, 123.8, 127.5, 128.3, 128.7, 131.0, 133.3, 133.6, 143.2, 147.5, 149.5, 150.2, 151.3, 151.8, 165.1; HRMS m/z calcd for C₃₂H₃₁N₅O₂ (M + Na)⁺ 540.2375, found 540.2375; anal. calcd for C₃₂H₃₁N₅O₂: C, 74.25; H, 6.04; N, 13.53%. Found: C, 74.54; H, 6.01; N, 13.49%.

Compound 3. White solid; (yield: 37 mg, 74%); IR (neat, cm⁻¹); 3298.0, 2927.1, 1647.6, 1526.0, 1272.3, 1201.4, 1018.4, 750.3; ¹H NMR (400 MHz, CDCl₃) δ (ppm): 1.31 (t, J = 7.1 Hz, 6H), 1.87 (s, 6H), 3.22 (q, J = 8 Hz, 4H), 3.50 (d, J = 4 Hz, 2H), 6.29 (s, 2H), 6.41 (s, 2H), 7.09 (d, J = 8 Hz, 1H), 7.37 (d, J = 5.6 Hz, 2H), 7.51 (t, J = 3, 5.4 Hz, 2H), 8.03 (d, J = 7.8 Hz, 1H) 8.46 (s, 1H), 8.49 (d, J = 5.2 Hz, 2H); ¹³C NMR (100 MHz, CDCl₃) δ (ppm): 14.7, 16.6, 38.3, 66.1, 96.6, 105.9, 118.0, 121.1, 123.5, 123.9, 127.6, 128.4, 128.6, 133.8, 142.4, 143.4, 147.6, 149.8, 151.4, 151.8, 165.3; HRMS m/z calcd for C₃₂H₃₁N₅O₂ (M + H)⁺ 518.2556, found 518.2554; anal. calcd for C₃₂H₃₁N₅O₂: C, 74.25; H, 6.04; N, 13.53%. Found: C, 74.51; H, 6.04; N, 13.52%.

Notes and references

1. C. Orvig and M. J. Abrams, Chem. Rev., 1999, 99, 2201.
2. R. McRae, P. Bagchi, S. Sumalekshmy and C. J. Fahrni, Chem. Rev., 2009, 109, 4780.
3. E. L. Que, D. W. Domaille and C. J. Chang, Chem. Rev., 2008, 108, 1517.
4. C. Bargossi, M. C. Fiorini, M. Montalti, L. Prodi and N. Zaccheroni, Coord. Chem. Rev., 2000, 208, 17.
5. L. Prodi, F. Bolletta, M. Montalti and N. Zaccheroni, Coord. Chem. Rev., 2000, 205, 59.
6. Y.-M. Kuo, B. Zhou, D. Cosco and J. Gitschier, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 6836.
7. E. L. Que, D. W. Domaille and C. J. Chang, Chem. Rev., 2008, 108, 1517.
8. X. Wu, D. Sinani, H. Kim and J. Lee, J. Biol. Chem., 2009, 284, 4112.
9. J. Heredia, M. Crooks and Z. Zhu, J. Biol. Chem., 2001, 276, 8793.
10. http://www.epa.gov/safewater/mcl.html#mcls.
11. E. S. Forzani, H. Zhang, W. Chen and N. Tao, Environ. Sci. Technol., 2005, 39, 1257.
12. E. L. Frank, M. P. Hughes, D. D. Bankson and W. L. Roberts, Clin. Chem., 2001, 47, 1109.
13. V. Dujols, F. Ford and A. W. Czarnik, J. Am. Chem. Soc., 1997, 119, 7386.
14. X. Chen, T. Pradhan, F. Wang, J. S. Kim and J. Yoon, Chem. Rev., 2012, 112, 1910 and references therein.
15. H. N. Kim, M. H. Lee, H. J. Kim, J. S. Kim and J. Yoon, Chem. Soc. Rev., 2008, 37, 1465 and references therein.
16. Y. Xiang, A. Tong, P. Jin and J. Yong, Org. Lett., 2006, 8, 2863.
17. Y. Zhou, F. Wang, Y. Kim, S. Kim and J. Yoon, Org. Lett., 2009, 11, 4442.
18. J. F. Zhang, Y. Zhou, J. Yoon, Y. Kim, S. J. Kim and J. S. Kim, Org. Lett., 2010, 12, 3852.
19. F. Yu, W. Zhang, P. Li, Y. Xing, L. Tong, J. Ma and B. Tang, Analyst, 2009, 134, 1826.
20. C. Yu, J. Zhang, R. Wang and L. Chen, Org. Biomol. Chem., 2010, 8, 5277.
21. X. Zeng, L. Dong, C. Wu, L. Mu, S.-F. Xue and Z. Tao, Sens. Actuators, B, 2009, 141, 506.
22. M. Yu, M. Shi, Z. Chen, F. Li, X. Li, Y. Gao, J. Xu, H. Yang, Z. Zhou, T. Yi and C. Huang, Chem.–Eur. J., 2008, 14, 6892.
23. M. Zhao, X.-F. Yang, S. He and L. Wang, Sens. Actuators, B, 2009, 135, 625.
24. Y.-B. Ruan, C. Li, J. Tang and J. Xie, Chem. Commun., 2010, 46, 9220.
25. J. Du, J. Fan, X. Peng, P. Sun, J. Wang, H. Li and S. Sun, Org. Lett., 2010, 12, 476.
26. Y. Shiraishi, S. Sumiya, Y. Kohno and T. Hirai, J. Org. Chem., 2008, 73, 8571.
27. A. Jana, J. S. Kim, H. S. Jung and P. K. Bharadwaj, Chem. Commun., 2009, 4417.
28. Y. Zhao, Y. Sun, X. Lv, Y. Liu, M. Chen and W. Guo, Org. Biomol. Chem., 2010, 8, 4143.
29. J. Huang, Y. Xu and X. Qian, J. Org. Chem., 2009, 74, 2167.
30. W. Lin, X. Cao, Y. Ding, L. Yuan and Q. Yu, Org. Biomol. Chem., 2010, 8, 3618.
31. Y. Zhou, X.-Y. You, Y. Fang, J.-Y. Li, K. Liu and C. Yao, Org. Biomol. Chem., 2010, 8, 4819.
32. H. Zheng, Z.-H. Qian, L. Xu, F.-F. Yuan, L.-D. Lan and J.-G. Xu, Org. Lett., 2006, 8, 859.
33. Q.-J. Ma, X.-B. Zhang, X.-H. Zhao, Z. Jin, G.-J. Mao, G.-L. Shen and R.-Q. Yu, Anal. Chim. Acta, 2010, 663, 85.
34. M. Suresh, A. Shrivastav, S. Mishra, E. Suresh and A. Das, Org. Lett., 2008, 10, 3013.
35. L. Tang, F. Li, M. Liu and R. Nandhakumar, Spectrochim. Acta, Part A, 2011, 78, 1168.
36. M. Suresh, S. Mishra, S. K. Mishra, E. Suresh, A. K. Mandal, A. Shrivastav and A. Das, Org. Lett., 2009, 11, 2740.
37. W. Lin, X. Cao, Y. Ding, L. Yuan and Q. Yu, Org. Biomol. Chem., 2010, 8, 3618.
38. H. N. Kim, S.-W. Nam, K. M. K. Swamy, Y. Jin, X. Chen, Y. Kim, S.-J. Kim, S. Park and J. Yoon, Analyst, 2011, 136, 1339.
39. D. Wu, W. Huang, Z. Lin, C. Duan, C. He, S. Wu and D. Wang, Inorg. Chem., 2008, 47, 7190.
40. S. Bhattacharya and M. Thomas, Tetrahedron Lett., 2000, 44, 10313.
41. G. Fang, M. Xu, F. Zeng and S. Wu, Langmuir, 2010, 26, 17764.
42. X. Chen, S.-W. Nam, M. J. Jou, Y. Kim, S.-J. Kim, S. Park and J. Yoon, Org. Lett., 2008, 10, 5235.
43. N. Kumari, N. Dey, S. Jha and S. Bhattacharya, ACS Appl. Mater. Interfaces, 2013, 5, 2438.
44. N. Kumari, S. Jha and S. Bhattacharya, J. Org. Chem., 2011, 76, 8215.
45. N. Kumari, S. Jha and S. Bhattacharya, Chem.–Asian J., 2012, 7, 2805.
46. S. Bhattacharya and A. Gulyani, Chem. Commun., 2003, 1158.
47. S. Bhattacharya, K. Snehalatha and V. P. Kumar, J. Org. Chem., 2003, 68, 2741.
48. D. Wu, W. Huang, C. Duan, Z. Lin and Q. Meng, Inorg. Chem., 2007, 46, 1538.
49. A. Sikdar, S. S. Panja, P. Biswas and S. Roy, J. Fluoresc., 2012, 22, 443.
50. W. Huang, C. Song, C. He, G. Lv, X. Hu, X. Zhu and C. Duan, Inorg. Chem., 2009, 48(5061), 19.
51. Y.-K. Yang, K.-J. Yook and J. Tae, J. Am. Chem. Soc., 2005, 127, 16760.
52. X.-F. Yang, X.-Q. Guo and Y.-B. Zhao, Talanta, 2002, 57, 883.
53. E. Ballesteros, D. Moreno, T. Gomez, T. Rodrıguez, J. Rojo, M. Garcıa-Valverde and T. Torroba, Org. Lett., 2009, 11, 1269.
54. Y. Zhao, X.-B. Zhang, Z.-X. Han, L. Qiao, C.-Y. Li, L.-X. Jian, G.-L. Shen and R.-Q. Yu, Anal. Chem., 2009, 81, 7022.
55. H. S. Jung, P. S. Kwon, J. W. Lee, J. Kim II, C. S. Hong, J. W. Kim, S. Yan, J. Y. Lee, J. H. Lee, T. Joo and J. S. Kim, J. Am. Chem. Soc., 2009, 9, 2008.
56. R. A. Løvstad, BioMetals, 2004, 17, 111.

---

Footnote

† Electronic supplementary information (ESI) available: Synthetic procedures, experimental details, additional spectroscopic data. See DOI: 10.1039/c3ra45054f

---