A highly selective and dual responsive test paper sensor of Hg²⁺/Cr³⁺ for naked eye detection in neutral water

Abstract

A highly selective and sensitive colorimetric and fluorogenic sensor (L₁) for Hg²⁺/Cr³⁺ is reported. This reagent (L₁) was synthesized by reacting 4-((4 (dimethylamino)phenyl)diazenyl)benzene-1-sulfonyl chloride, which has a dimethylaminophenyldiazenyl fragment as a photoactive signalling unit, with 2,2′-(3,3′-azanediylbis(propane-3,1-diyl))diisoindoline-1,3-dione as the receptor fragment. The reagent was characterized by standard analytical and spectroscopic techniques. Electronic spectral studies revealed that the reagent was selective for Hg²⁺ and Cr³⁺ in the presence of all other metal ions of Group 1A, IIA and all other common transition metal ions. On binding of L₁ to the Hg²⁺ or Cr³⁺ centres, a new intense absorption band with a λ_max of 509 nm appeared with associated changes in the visually detectable solution colour from yellow to red. Fluorescence spectral studies revealed a significant enhancement in the emission intensities upon coordination to Hg²⁺ or Cr³⁺ without any change in the emission wavelength. This could be explained by the efficient interruption of the photo induced electron transfer signalling mechanism involving an unshared pair of electrons from the central tertiary amine centre. An easy to prepare paper test kit, which was obtained by soaking the filter paper in a dichloromethane solution of L₁, presents an approach that could be successfully used in the detection of Hg²⁺ or Cr³⁺ ions present in neutral aqueous media. This indicates the potential application of this dip strip type sensor for the detection of Hg²⁺ and Cr³⁺ in neutral aqueous environments without any spectroscopic instrumentation. Importantly, this reagent binds specifically to Cr³⁺ in the presence of an excess of iodide ions, which act as a masking agent for Hg²⁺. To the best of our knowledge, there are very few examples of detection limits lower than the present test strip for Hg²⁺ in the literature, while, for Cr³⁺, no such report is available.

---

Introduction

Certain transition metal ions are of grave concern not only for chemists, biologists and environmentalists, but also for the general population due to their deleterious influences on living systems, human heath and the environment.¹ Thus, considerable efforts have been devoted to the development of easy-to-use methodologies for the detection and sensing of these specific transition metal ions in aqueous environments.² Unfortunately, many industrial detection methods rely on time consuming and sophisticated analytical techniques, such as atomic absorption spectroscopy, inductively coupled plasma mass spectroscopy, ion chromatography, etc., which are generally not very conducive for an “in-field” application.³ Thus, it is almost essential to develop sensitive, rapid and simple to use methodologies for the recognition and estimation of these toxic transition metal ions present in trace amounts.

Among various toxic transition metal ions, Cr³⁺ and Hg²⁺ are important due to their biological significance as well as their adverse influence on various living organisms.⁴ The trivalent form of chromium is an essential nutrient for humans and its deficiency causes disturbances in glucose levels and lipid metabolism.⁵ On the other hand, a build up of Cr³⁺ beyond the threshold value in any living organism has a detrimental influence. Various industrial or/and agricultural activities that add to the build up of Cr³⁺ in the environment is a matter of concern.⁵ Among other transition metal ions, the toxic influences of Hg²⁺ on the human population is well documented.^6–8 Hg²⁺ is known to be involved in biological processes, such as the transmission of nerve impulses and the regulation of critical cell activities.⁶ Furthermore, Hg²⁺ present in water and soil is known to bioaccumulate in the human body and other higher organisms through propagation in the food chain.⁷ These accumulations and exposure to an excess of Hg²⁺ leads to DNA damage, mitosis impairment and nervous system defects.⁸ These collectively have led to an urgency for the development of an efficient and appropriate chemosensor for Cr³⁺ and Hg²⁺ ions that are capable of detecting these ions in biological and environmental samples in the presence of other interfering cations. Over the last few decades considerable attention have been paid to the development of reversible optical chemosensors that are able to detect the presence of these ions.⁹ In this regards, three different approaches have generally been adopted to convert the binding event into a measurable feedback signal. These involve use of either a fluorescent/colorimetric group as a reporter functionality that responds through changes in its emission/optical spectral pattern, or a reagent that allows the binding process to be probed through changes in the NMR spectral pattern or redox active groups that allow monitoring through changes in the redox potentials. Among these approaches, reagents that allow the binding process to be probed by monitoring changes in the optical feedback through a “turn-on” response, rather than a turn-off response are preferred for the design of an efficient sensor for practical in the field applications. The Cr³⁺ ion, through its paramagnetic influence and the Hg²⁺ ion, through its efficient spin–orbit coupling, are commonly known to effectively quench the receptor fluorescence. To date, reports on fluorescence-based sensors with a turn-on response for Cr³⁺ are very limited in the literature due to the lack of availability of an appropriate selective ligand system. More importantly, there are very few examples of receptor molecules that are able to convert the binding event into more than one type of read-out signal, e.g. fluorescence and optical spectral changes. Among these, colorimetric receptors that allow the visual detection of Cr³⁺ and Hg²⁺ are of special significance.¹⁰ For simplicity, convenience and to achieve a low-cost alternative, the development of a colorimetric test kit that allows the semi-quantitative detection of target cations using a standard colour comparator is in high demand.

In this article we have presented a rational strategy for the development of a selective chromo-sensor (L₁) for the detection of Cr³⁺ and Hg²⁺ with a remarkable colour variation in a CH₃CN medium along with a turn-on response for all other common alkali and transition metal ions. More importantly, this reagent could be used for the development of an easy-to-use paper test strip for the semi-quantitative colorimetric detection of these two ions present in neutral aqueous media with concentrations as low as 10 ppm. Notably, this reagent binds specifically to Cr³⁺ in the presence of an excess of iodide ions, which act as a masking agent for Hg²⁺.

Results and discussion

Synthesis

Intermediate ligand L (Scheme 1) was synthesized following a reported procedure by reacting bis(3-aminopropyl)amine and phthalic anhydride in an appropriate ratio and was achieved in a reasonably good yield.¹¹ Isolated compound L was further allowed to react with 4-(dimethylamino)phenyl)diazenyl)benzene-1-sulfonyl chloride to obtain the desired compound, L₁.¹¹ These receptors were characterised using various analytical and spectroscopic techniques and the data thus obtained matched well with that reported earlier for these two ligands.¹¹

Scheme 1

UV–vis spectroscopy

In this study, the N,N'-dimethyl azo group was introduced to bis(phthalimidyl propyl)amine (L) to obtain the host molecule, L₁. The UV absorption spectra of L and L₁ (2.0 × 10⁻⁵ M) were measured in a CH₃CN solution. The assignments of the UV–vis spectral bands have been reported earlier.¹¹ Compound L₁ exhibits an absorption maxima at 440 nm; while, for L, the adsoprtion maxima is 290 nm (Fig. 1A). For L, the absorption maxima at ∼290 nm is predominantly a π–π*-based intramolecular charge transfer (ICT) process involving the donor N_Amine atom and the acceptor phthalimide moiety. For L₁, the band at 440 nm is predominantly an ICT-based process involving the N,N'-dimethylamine as the donor functionality and the N=N functionality as the acceptor. The appreciably higher extinction coefficient (8.3 × 10⁴ mol⁻¹ L cm⁻¹) for L₁, as compared to that of L (5.7 × 10³ mol⁻¹ L cm⁻¹) is presumably due to the higher value of the transition dipole moment associated with this CT transition for L₁.¹¹

Fig. 1 (A) The absorbance spectra of L and L₁ (2.0 × 10⁻⁵ M) in a CH₃CN medium; (B) the absorbance spectra of L₁ (2.0 × 10⁻⁵ M) in different solvents with varying polarities.

The ICT nature of this characteristic band of L₁ was further confirmed by the bathochromic shifts of this absorption band with a gradual increase in the solvent polarity (Fig. 1B). The energies of the electronic transitions, which are associated with the ICT absorption band, are known to be appreciably affected by a change in the medium polarity and the acceptor number present in the solvent.¹² It is generally argued that this effect grossly depends on the extent to which the lone pair of electrons at the terminus of the N,N-dimethylamine group can actually interact with the Lewis acidity of the solvent molecules, which is generally determined by the respective acceptor numbers of the solvent.¹² This results in a red-shifted absorption band maxima with an increase in the solvent polarity.

A detailed analysis of the UV–vis absorption spectra of L₁ in the absence and presence of group IA, group IIA and all other common transition metal ions revealed that this reagent binds preferentially to Cr³⁺ and Hg²⁺. The binding is associated with a significant change in the absorption spectral pattern and a visually detectable colour change in acetonitrile (Fig. 2C). This reagent was also found to bind to Nd³⁺, among other lanthanide ions in acetonitrile, and this result has been reported in our previous article.¹¹ No change in the electronic spectral pattern could be detected in the presence of externally added Li⁺, Na⁺, K⁺, Cs⁺, Mg²⁺, Ca²⁺, Ba²⁺, Sr²⁺, Co³⁺, Ni²⁺, Cu²⁺, Fe²⁺, Fe³⁺, Cd²⁺, Al³⁺ and Ag⁺. For Zn²⁺ (Fig. 2), an insignificant shift in the absorption spectra was observed. This tends to suggest a preferential and appreciable binding of L₁ to Cr³⁺ and Hg²⁺ and a weak binding to Zn²⁺ (Fig. 2). However, one cannot rule out the possibility of a very weak binding to any one of the other metal ions, where the binding may be too weak to perturb the energies of the frontier orbitals and, thus, the energies for the spectral transition. The complexation ability of L₁ with Cr³⁺ and Hg²⁺ was investigated by systematic UV–vis absorption titrations for varying concentrations of Cr³⁺ and Hg²⁺ as their perchlorate salts. During the titration process, the effective concentration of L₁ remained unaltered. Notably, upon addition of increasing amounts of Cr³⁺ or Hg²⁺, the characteristic strong absorption band at 440 nm for L₁ was found to decrease with a simultaneous growth of an absorption band with a maxima at 509 nm along with an isosbestic point at 464 nm (Fig. 3). Spectral changes were associated with a visually detectable colour change from yellow to red (Fig. 2C). However, no apparent visual colour change occurred for L upon addition of Cr³⁺ and Hg²⁺ in CH₃CN as the red-shifted absorption band for L upon binding to Cr³⁺ or Hg²⁺ is in the UV region (Electronic Supplementary information, ESI†). The red-shift of the absorption band for L confirmed the binding of Cr³⁺ and Hg²⁺ to the N-tertiary amine of the propyl chain, together with the carbonyl group of the phthalimide ring. Such a binding mode is consistent with a previous literature report.¹³ Presumably, L₁ binds to Cr³⁺ and Hg²⁺ using the same binding site as that of L and this was confirmed by detailed ¹H NMR studies and is discussed later.

Fig. 2 (A) The absorbance spectra of the chemosensor (L₁) (2.0 × 10⁻⁵ M) in the absence and presence of 10 mole equivalent of various metal ions, such as Cr³⁺ , Hg²⁺, Zn²⁺, Cu²⁺, Cd²⁺, Ni²⁺, Co³⁺, Ca²⁺, Fe²⁺, Al³⁺ Mg²⁺, Na⁺, Ag⁺, Sr²⁺, Ba²⁺, Sr²⁺, Li⁺ and Cs⁺ in CH₃CN; (B) a bar diagram showing the change in the absorbance for L₁ (2.0 × 10⁻⁵ M) in the presence of the above mentioned metal ions (2 × 10⁻⁴ M) in CH₃CN at a λ_max of 509 nm; (C) the colour variation of L₁ (20 μM) in the presence of different metal ion (0.4 mM) in CH₃CN.

Fig. 3 The electronic spectra of L₁ (2.0 × 10⁻⁵ M) in the presence of varying concentrations of (A) Cr³⁺ (0 − 5.56 × 10⁻⁴ M) and (B) Hg²⁺ (0 − 4.60 × 10⁻⁴ M). Insets: the corresponding titration profiles for varying [Cr³⁺] or [Hg²⁺] (the ΔA at 509 nm vs. [Cr³⁺] or [Hg²⁺]) confirms the 1 : 1 binding stoichiometry.

The spectrophotometric titrations with varying [Cr³⁺] and [Hg²⁺] are shown in Fig. 3A and 3B. The sigmodal curves were obtained by plotting the change in the absorbance intensity at 509 nm as a function of [Mⁿ⁺] (where Mⁿ⁺ is Hg²⁺ or Cr³⁺) and confirmed the 1 : 1 binding stoichiometry of L₁·Cr³⁺ and L₁·Hg²⁺. The binding stoichiometry was further confirmed by a non-linear least squares plot (ESI†) using eqn (4).¹⁴ The respective association constants (K_a) for the binding of L₁ to Cr³⁺ and Hg²⁺ were 6.54 × 10⁴ M⁻¹ and 4.31 × 10⁴ M⁻¹, respectively, at 25 °C in CH₃CN. The N,N'-dimethylamine functionality in L₁ acts as a donor group, while the diazo and the phthalimide fragment act as acceptor groups with poor donating abilities. The addition of metal ions (Cr³⁺ and Hg²⁺) may result in the binding of the respective metal ion to the central tertiary amine along with the carbonyl “O_Phthalimide” atom and this consequently results in the transformation of the donor-π-donor structure to a donor-π-acceptor structure (Scheme 2). This is expected to favour the photoinduced intramolecular charge transfer process with an associated red-shift (70 nm) in the absorption maxima of L₁.

Scheme 2

Luminescence study

The emission spectrum of L₁ (2.0 × 10⁻⁵ M) was recorded in CH₃CN in the absence and presence of various metal ions upon excitation at 464 nm. However, the emission spectrum of compound L₁ does not exhibit any obvious emission signal when excited at 464 nm. In the absence of any emission band, the quenching process was ascribe to the photo induced electron transfer (PET) process involving the lone pair of electrons from the tertiary amine functionality coupled with a photoactive signalling unit in L₁.¹¹ However, a significant enhancement in the emission intensity (a 22 fold increase for Cr³⁺ and a 20 fold increase for Hg²⁺) at 575 nm was observed upon binding of the reagent, L₁, with Cr³⁺ or Hg²⁺ in a CH₃CN solution (Fig. 4A). For all other cations, such as Li⁺, Na⁺, K⁺, Cs⁺, Ca²⁺, Mg²⁺, Sr²⁺, Ba²⁺, Zn²⁺, Cu²⁺, Cd²⁺, Ni²⁺, Co³⁺, Fe²⁺ and Ag⁺, no detectable increase in the emission intensity for L₁ in the CH₃CN was observed. Generally, receptors that show a “switch-on” rather than a “switch-off ” fluorescence response upon binding to an analyte is more advantageous, as this adds to the ease of the detection process, as well as permitting a lower detection limit with an improved signal-to-noise ratio. The enhancement in the emission intensity without any change in the excitation and emission wavelength suggests a potential PET-based process as the signal transduction mechanism.

Fig. 4 (A) The emission spectra for the chemosensor (L₁) (2.0 × 10⁻⁵ M) in the absence and presence of 10 mole equivalent of various metal ions, such as Cr³⁺ , Hg²⁺, Zn²⁺, Cu²⁺, Cd²⁺, Ni²⁺, Co³⁺, Ca²⁺, Fe²⁺, Al³⁺ Mg²⁺, Na⁺, Ag⁺, Sr²⁺, Ba²⁺, Sr²⁺, Li⁺ and Cs⁺ in CH₃CN; (B) the luminescence response of L₁ (2.0 × 10⁻⁵ M) to the above representative metal ions (2 × 10⁻⁴ M) in CH₃CN. The emission intensity was monitored at 575 nm.

On binding to Cr³⁺ or Hg²⁺, the lone pair of electrons on the tertiary N_amine atom becomes involved in the coordination and effectively interrupts the PET-based luminescence quenching process. This results in a switch-on luminescence response. Furthermore, the rigidity imparted to the receptor, L₁, on coordination to the Cr³⁺ or Hg²⁺ may also contribute to its fluorescence enhancement by inhibiting non-radiative decay pathways through the restriction of vibrational and rotational relaxation deactivation processes. A systematic analysis (Benesi–Hilbebrand plot) of the luminescent titration profiles (Fig. 5) supports the formation of a 1 : 1 molecular assembly, while the association constant values (K_a) of 5.71 × 10⁴ M⁻¹ (for Cr³⁺) and 3.88 × 10⁴ M⁻¹ (for Hg²⁺) were obtained. These values are in close proximity to the values obtained from the spectrophotometric titrations.

Fig. 5 The changes in the emission spectra of L₁ (2.0 × 10⁻⁵ M) in the presence of varying (A) [Cr³⁺] (0–4.0 × 10⁻⁴ M) and (B) [Hg²⁺] (0–5.0 × 10⁻⁴ M). Insets: Benesi–Hildebrand plots for the respective titrations with Cr³⁺ and Hg²⁺ showing a 1 : 1 binding stoichiometry.

In the presence of excess of [Bu₄N]I, the characteristic red colour of the L₁·Hg²⁺ solution or the electronic spectra, with an absorbance maxima at 509 nm, was found to disappear, while the spectra of the free receptor (L₁) with λ_max at 440 nm was restored (Fig. 6A). The luminescence intensity of the L₁·Hg²⁺ solution with λ_max at 575 nm was also found to decrease with an increasing [I⁻] and, eventually, the emission band disappeared (Fig. 6D). These collectively show that Hg²⁺ sensing by L₁ is a reversible process. However, L₁·Cr³⁺ in solution did not exhibit any significant change in colour or luminescence intensity in the presence of externally added excesses of [Bu₄N]I. This is further supported by the fact that the addition of Cr(ClO₄)₃ to the solution with L₁·Hg²⁺ + excess [Bu₄N]I induced a visually detectable change in the solution colour from yellow to red and a simultaneous new absorption spectral band for L₁·Hg²⁺ at 509 nm with a concomitant decrease in the absorbance at 440 nm (Fig. 3 and 6A). A separate study also confirmed that the presence of [Bu₄N]I had no effect on the absorption and emission profile of the free dye. Changes in both the absorption and emission spectral pattern of L₁·Hg²⁺ in the presence of excess iodide suggest the formation of a stable iodide complex of Hg²⁺, for example HgI₄²⁻.^9f,15

Fig. 6 (A) The absorbance spectra of (i) L₁ (20 μM) + Hg²⁺ (80 μM) (ii) L₁ + Hg²⁺ in the presence of [Bu₄N]I (5 mM) and (iii) L₁ + Hg²⁺ in the presence of [Bu₄N]I (5 mM) + Cr³⁺ (80 μM) in an acetonitrile solution; (B) the bar diagram represents the absorbance intensities at 509 nm of the respective solutions containing (1) L₁ + Hg²⁺, (2) L₁ + Hg²⁺ + [Bu₄N]I, (3) L₁ + Hg²⁺ + [Bu₄N]I + Cr³⁺ and (4) L₁ + Cr³⁺ + [Bu₄N]I; (C) shows the colour variation of the solution of L₁ in the presence of different ions and reagent: (1) L₁ (20 μM) only, (2) L₁ in the presence of Hg²⁺ (80 μM), (3) L₁ in the presence of Hg²⁺ and [Bu₄N]I (5 mM), (4) L₁ in the presence of Hg²⁺, [Bu₄N]I and Cr³⁺ (80 μM); (D) the bar diagram represents the emission intensity of the solution containing (1) only L₁, (2) L₁ + Hg²⁺, (3) L₁ + Cr³⁺ , (4) L₁ + Hg²⁺ [Bu₄N]I and (5) L₁ + [Bu₄N]I + Cr³⁺ at 575 nm in acetonitrile.

The receptor, L₁, failed to bind Cr³⁺ and Hg²⁺ in the aqueous medium due to the high and deleterious solvation energies of the respective cations. Thus, to avoid the competing solvation effect of water, we explored the possibility of using this reagent for the development of a paper test strip, which could be used for the semi-quantitative colorimetric detection of Cr³⁺ and Hg²⁺ present in neutral aqueous environments by the naked eye and, thus, appropriate for practical in the field applications.

Test paper detection

The possibility of using chemosensor L₁ in the development of paper test strips was examined. Paper test strips were prepared by immersing the filter papers (Whatman filter paper No. 42) in a dichloromethane solution of L₁ (0.002 M), which was then air dried. These test strips, containing L₁, were then immersed in neutral aqueous solutions of the respective metal ions with varying concentrations (400 ppm to 10 ppm) for about 2–3 min and then air dried. The colour of these test strips was compared with one that was dipped only in a dichloromethane solution of L₁. The colour of these strips turned red when dipped either in Cr³⁺ or Hg²⁺ solutions (Fig. 7) and the colour of the respective strip became lighter as the concentration of the respective metal ion in the aqueous solution decreased. This clearly revealed that these strips containing L₁ could distinguish a concentration gradient of Cr³⁺ or Hg²⁺ in an aqueous solution through gradual changes in the visually detectable colour of the strips. The lowest concentration limit of the metal ions that could be detected from neutral aqueous solutions using these strips is 10 ppm. The change in colour seems to be more prominent for Hg²⁺. This can be rationalized if one considers the higher solvation energy of the Cr³⁺ ion (H_hyd = 4639 kJ mol⁻¹) over that of the Hg²⁺ ion (H_hyd = 1824 kJ mol⁻¹) in an aqueous medium. Thus, these test strips can be utilized for semi-quantitative in the field detection of these two ions present in neutral aqueous environments through a simple colour comparison performed by the naked eye using a dip-stick type test strip containing L₁ and different standard solutions of the respective metal ions. Such examples of test strips for Hg²⁺ are very limited in the contemporary literature.¹⁶ To the best of our knowledge, the example of a test strip with such a detection limit is rare. However, for the Cr³⁺ ion, no such report is available in the literature.

Fig. 7 The variation in the colour of the test strip for L₁ after dipping in aqueous solutions with varying concentrations of Hg²⁺ and Cr³⁺. A solution of L₁ with a concentration of 0.002 M was used to develop the strip.

The preferential formation of HgI₄²⁻ in the presence of an excess of KI was successfully exploited for the development of a specific test paper sensor for Cr³⁺ in a pure aqueous medium.

The red coulored test strip, containing L₁·Hg²⁺ (80 ppm of Hg²⁺) was developed following the procedure mentioned above and air dried. This was then dipped into an aqueous solution of KI (10.0 mM). The red colour of the strip disappeared and the original yellow colour of the test strip was restored (Fig. 8). However, a similar experiment with the red coloured test strip containing L₁·Cr³⁺ (prepared using 100 ppm of Cr³⁺) did not show any visually detectable change in colour when dipped in an aqueous solution of KI (10.0 mM) and the colour remained comparable to the one shown in Fig. 7 for 100 ppm of Cr³⁺.

Fig. 8 The variation in the colour of the test strip for L₁ only (1), L₁ after dipping in an aqueous solution of Hg²⁺ (100 ppm) (2), and after dipping in an aqueous solution of KI (10 mM) (3).

¹HNMR analysis

The metal ion binding affinity of the receptor, L₁, was examined by detailed ¹H NMR spectroscopy. The relative affinity of Hg²⁺ for L₁ was also evident in the observed spectral shift of the ¹H NMR spectra for L₁ in the absence and presence of comparable concentrations of Hg²⁺ (Fig. 9). An upfield shift in the ¹H NMR signals for most of the protons of L₁ was observed. Such an upfield shift for most of the proton signals is not an unusual phenomenon and not very uncommon in the literature.^9a,14 For L₁, upfield shifts of 0.0395 ppm (for H2), 0.0613 ppm (for H1) and 0.045 ppm (for H3) were observed, while a downfield shift of 0.36 ppm was noted for H6 upon binding to Hg²⁺. Noticeable upfield shifts in the ¹H NMR signals were also observed for most of the aromatic protons, except H7. An upfield shift of 0.0631 ppm was observed for H8 and H10 on binding to Hg²⁺. The signals for H4 and H5 became broad upon binding of the C=O group to Hg²⁺ with an associated upfield shift of ∼0.1 ppm, while for H7, a downfield shift of 0.376 ppm was noticed. The ¹H NMR spectra of the two model receptors, L and L₂, were also recorded in the absence and presence of Hg²⁺ for the unambiguous assignment of the probable binding mode between L₁ and Hg²⁺ (ESI†). Prominent upfield shifts for the aliphatic, as well as the aromatic protons for L, were observed when the spectra were recorded in the presence of Hg²⁺. Thus, the nature of the shifts in the ¹H NMR spectra for L and L₁ is similar, which signifies a similar binding mode for L and L₁ towards Hg²⁺, which proceeds through one N_Amine and two O_Phthalimide centres. The shifts for the ¹H NMR signals of H3^/, H4^/ and H5^/ of L₂ upon addition of an excess (25 mole equivalents) of Hg²⁺ were only within ∼0.0004 ppm, a value that is within the limit of experimental error; while that for H2^/ was 0.0032 ppm. The shift in the Δδ value for H6^/ is 0.008 ppm. These minor shifts in the δ values for L₂ (ESI†), as compared to the shift observed for L and L₁ upon binding to Hg²⁺, tend to nullify the possibilty of coordination either through the aromatic part or the –NMe₂ fragment of L₂. The similarities in the chemical shift patterns for L and L₁ confirm that the N_Amine and the two O_Phthalimide centres of L and L₁ are actually involved in the coordination to the Hg²⁺ centre. As expected, no significant ¹H NMR spectral change for L₁ was observed in the presence of Cr³⁺.

Fig. 9 The ¹H NMR spectra of L₁ in the absence and presence of Hg²⁺ in CD₃CN, (A) the change in the aromatic region, (B) the change in the aliphatic region, and (C) represents the probable binding mode of L₁ with the metal ions (Hg²⁺, Cr³⁺) and the molecular structures for L and L₂.

Experimental section

Materials and methods

Phthalic anhydride, bis(3-aminopropyl)amine-1,4′-sulfonyl chloride, 4-(dimethylaminobenzene)azo benzene, the percholarate salts of the different cations (Cr³⁺, Hg²⁺, Zn²⁺, Cu²⁺, Cd²⁺, Ni²⁺, Co³⁺, Ca²⁺, Fe²⁺, Al³⁺ Mg²⁺, Na⁺, Ag⁺, Sr²⁺, Ba²⁺, Sr²⁺, Li⁺, Cs⁺etc) were obtained from Sigma–Aldrich and used as received. All the other reagents used were procured from S. D Fine Chemicals, India. Acetonitrile was procured from Merck, India and all solvents were dried and distilled prior to use.

Instrumentation

The electronic spectra were recorded using a Cary 500 Scan UV–vis–NIR spectrometer, while the steady-state luminescence spectra were recorded on a HORIBA JOBIN YVON spectrofluorometer . ¹H-NMR spectra were recorded on a Bruker 200 MHz, 500 MHz FT NMR (Model: Avance-DPX 200) using CD₃CN as the solvent and tetramethylsilane (TMS) as an internal standard. The FTIR spectra were recorded using a Perkin--Elmer Spectra GX 2000 spectrometer.

Synthesis of bis(phthalimidylpropyl)amine (L). The known procedure was adopted for the synthesis of L.¹¹ Phthalic anhydride (4.74 g, 32 mmol) was added to a solution of bis(3-aminopropyl)amine-1 (2.0 g, 15.24 mmol) in 150 ml of acetonitrile as the solvent. The reaction mixture was stirred under reflux for 2 days. Then, the solvent was evaporated off and 200 ml of ethanol was added to the solid residue. After stirring for 5 h, the white precipitates were filtered, collected and dried over P₂O₅ to give the desired compound in pure form and was used for reaction without any further purification (Yield: 4.15 g; 70%). ¹H NMR (200 MHz, CDCl₃, 25 °C, TMS); δ (ppm): 7.82–7.78 (m, 4H, ArH), 7.71–7.66 (m, 4H, ArH) , 3.77 (t, J = 6.6 Hz, 4H, nCH₂), 2.78 (t, J = 6.6 Hz, 4H, –CH₂), 1.99–1.92 (m, 4H, –CH₂). IR (KBr): ν_max/cm⁻¹ = 3461, 3031, 2936, 2816, 1771, 1711, 1640, 1398, 1364, 1044, 719, 530. ESI-MS (m⁺/z): 392.16 [M + H]⁺ (90%). Elemental analysis for C₂₂H₂₁N₃O₄ (391.15): calculated, C 61.51, H 5.41, N 10.74%; found, C 61.6, H 5.3, N 10.7%.

Synthesis of L₁. This reagent was also synthesized following the reported procedure with slight modification in the methodology for the purification process.¹¹ Bis(phthalimidylpropyl)amine (L; 0.5 g, 1.278 mmol) and K₂CO₃ (0.265 g; 1.917 mmol) were added to 100 ml dry acetonitrile. The reaction mixture was refluxed for 2 h and then 4-(dimethylamino)phenyl)diazenyl)benzene-1-sulfonyl chloride (0.516 g; 1.6 mmol), dissolved in 40 ml dry acetonitrile, was added in a dropwise manner to the reaction mixture. After the complete addition, the mixture was refluxed for 24 h under nitrogen gas and the solution colour changed from red to orange. The reaction mixture was filtered and the solvent from the filtrate was removed under reduced pressure. Then, the desired compound was extracted thrice in chloroform (3 × 40 ml), while the undesired impurities in the aqueous layer were discarded. The organic phase was collected and dried over anhydrous sodium sulphate and a light orange oily residue was obtained upon removal of the solvent under vacuum. This oily residue was sonicated for 10 min with n-hexane, while an orange solid precipitated. This was collected by filtration and dried to give L₁ (Yield: 0.55 g; 63%) in pure form. ¹H NMR (500 MHz, CD₃CN, 25 °C, TMS); δ (ppm): 7.86 (d, J = 9 Hz, 2H; ArH), 7.79 (m, 4H, ArH), 7.77–7.77 (m, 8H, ArH), 6.83 (d, J = 9 Hz, 2H, ArH), 3.60 (t, J = 7 Hz, 4H, –CH₂), 3.23 (t, J = 7 Hz, 4H, –CH₂), 3.09 (s, 6H, –CH₃), 1.91–1.85 (m, 4H, –CH₂), IR (KBr); ν_max/cm⁻¹: 3460, 3069, 2926, 1770, 1711, 1601, 1521, 1366, 1332, 1132, 712, 594. ESI-MS (m⁺/z): 679.09 (M + H)⁺ (5%), 701.03 (M + Na)⁺ (100%), C₃₆H₃₄N₆O₆S (678.23). Elemental analysis: calculated, C 63.70, H 5.05, N 12.38%; found, C 63.8, H 5.1, N 12.3%.

Synthesis of L₂. 4-(4-dimethylamino-phenylazo)benzene sulphonyl chloride (100 mg, 0.31 mM) was added to 15 ml of pyrrolidine (in excess) under an inert atmosphere and heated to reflux for 24 h. Then, the reaction mixture was evaporated to obtain the dry solid. To get L₂ in a pure form the crude product was recrystallised from methanol in 85% yield. ¹H NMR (500 MHz, CD₃OD, 25 °C, TMS); δ (ppm): 7.93 (d, J = 8.4 Hz, Ar–H5^/; ArH), 7.87–7.8 (m, 4H, Ar–H3^/, H4^/) 6.83 (d, J = 9 Hz, ArH2^/), 3.32–3.27 (t, 4H, H6^/(–CH₂)), 3.1 (s, H1^/ (–NMe₂) 2.03–1.96 (m, 4H, H7^/); ν_max/cm⁻¹: 3454, 3062, 2920, 1771, 1711, 1601, 1521, 1366, 1332, 1128, 710, 594. ESI-MS (m⁺/z): 359.18 (M + H)⁺ (10%), Elemental analysis: calculated, C 60.3, H 6.2, N 15.6%; found, C 60.4, H 6.1, N 15.5%.

Spectrophotometric titration

A 1.00 × 10⁻⁴ M solution of L₁ in acetonitrile was prepared and stored in the dark. This solution was used for all spectroscopic studies after appropriate dilution. 1.0 × 10⁻⁴ M solutions of the perchlorate salt of the respective cations were prepared in pre-dried and distilled acetonitrile and were stored under an inert atmosphere. Solutions of L₁ were further diluted for spectroscopic titrations and the effective final concentration was adjusted to 2.0 × 10⁻⁵ M, while the final analyte (cation) concentration for titration was varied from 0 to 6.0 × 10⁻⁴ M.

The equilibrium of complexation between L₁ and Mⁿ⁺ was evaluated using the non-linear least squares method (eqn (4)–(6)) for the following reaction:

L₁ + nMⁿ⁺ → [Mⁿ⁺]_nL₁ (a)

The stability constant of the complex could be define as:

K_n = [M_nL₁]/[L₁][M]ⁿ (1)

where, Mⁿ⁺ stands for Cr³⁺ and Hg²⁺. Neither L₁, Cr³⁺ or Hg²⁺ has an absorption maxima above 440 nm. Thus, the new absorbance maxima that developed at around 509 nm upon addition of Cr³⁺ or Hg²⁺ was attributed to the formation of a coordination complex (eqn (a)). We have selected these two wavelengths for our studies. Respective molar extinction coefficient of L₁ at 440 and 509 nm are ε_{L1}440 and ε_{L1}509:

A₄₄₀ = ε_{L1}440 × l ×[L₁] + ε_{MnL1}440 × l × [ML₁] (2)

A₅₀₈ = ε_{L1}509 × l × [L₁] + ε_{MnL1}509 × l × [ML₁] (3)

ε_{ML1}440 and ε_{ML1}509 are the molar extinction coefficients of [Mⁿ⁺]L₁ at 440 and 509 nm, respectively. For an optical path length (l) of 1 cm, as used in the present study, the ratio A₄₄₀/A₅₀₉ can be expressed by eqn (4):

A₄₄₀/A₅₀₉= (K_nε_{MnL1}440[M]ⁿ⁺ + ε_{L1}440)/(K_nε_{MnL1}509[M]ⁿ⁺ + ε_{L1}509) (4)

The overall concentration of the cation is C_M and thus:

C_M = [M] + n[M_nL] (5)

Thus, [M] = C_M − n[M_nL₁] = (C_M − nA₄₄₀)/ε_{MnL1}440 (6)

The absorbance data were analysed using a non-linear least squares method.¹⁷ Appropriate substitutions and approximations were used for the evaluation of the stoichiometry and affinity constant (n and K_n, respectively). The analysis provided the binding stoichiometry of the complex formed between L₁ and Cr³⁺ or Hg²⁺ (ESI†) and the association constant for the respective metal ion.

Luminescence titration

A stock solution of L₁ with an initial concentration of 1.0 × 10⁻⁴ M in CH₃CN was prepared and stored in the dark. For all measurements, a λ_Ext = 464 nm was used (with excitation and emission slit widths of 5 nm). This stock solution was used for recording the changes in the emission spectral pattern for L₁ in the presence of various externally added cations (Cr³⁺, Hg²⁺, Zn²⁺, Cu²⁺, Cd²⁺, Ni²⁺, Co³⁺, Ca²⁺, Fe²⁺, Al³⁺ Mg²⁺, Na⁺, Ag⁺, Sr²⁺, Ba²⁺, Sr²⁺, Li⁺ and Cs⁺) and the effective concentration of L₁ was maintained at 2.0 × 10⁻⁵ M. Two different luminescence titrations were carried out using L₁ with varying concentrations of the perchlorate salt of Cr³⁺ (0–4.0 × 10⁻⁴ M) and Hg²⁺ (0–5.0 × 10⁻⁴ M) in CH₃CN. The affinity constant and 1 : 1 binding stoichiometry for the respective metal ions were calculated using the Benesi–Hildebrand method and related expression as shown in eqn (7).^10e,18

where F₀ is the emission intensity of L₁ , F is the emission intensity obtained with externally added metal ions, K is the association constant (M⁻¹) and [Mⁿ⁺] is the concentration of the externally added metal ion. As shown in Fig. 5 (C) and (D), the plot of 1/(F − F₀) against 1/[Mⁿ⁺] shows a linear relationship, indicating that L₁ indeed associates with Mⁿ⁺ (Cr³⁺ or Hg²⁺) in a 1 : 1 stoichiometry.

Conclusions

Experimental studies revealed that a molecular receptor, L₁, binds only to Cr³⁺ and Hg²⁺ among other biologically relevant common alkali and transition metal ions in acetonitrile. The binding of Cr³⁺ and Hg²⁺ to L₁ gave rise to the visually detectable change in colour from yellow to red. The receptor, L₁, was also found to act as a “turn-on” fluorescence sensor for these two ions upon formation of L₁·Cr³⁺ or L₁·Hg²⁺ in CH₃CN. This chelation enhanced fluorescence response was attributed to the interruption of the PET process, which was otherwise operational for the free receptor molecule (L₁). The probable binding mode for the receptor (L₁) to the metal ions was established through detailed ¹H NMR studies of L₁ and the related model receptors (L and L₂). A test strip from readily available filter papers was developed using the reagent, L₁. This reagent was used for the semi-quantitative in the field visual detection of these two ions present in neutral aqueous medium. The lowest concentration limit of the metal ions that could be detected using these test strips was 10 ppm. More interestingly, in the presence of an excess of KI, these paper test strips could be used for the specific recognition of Cr³⁺ in an aqueous solution that contained both Hg²⁺ and Cr³⁺. Preferential binding of Hg²⁺ to I⁻ allowed the specific binding of the reagent, L₁, to Cr³⁺. To the best of our knowledge, for Hg²⁺, there are only a few previous examples where detection limits lower than the present test strip are reported, while for the Cr³⁺ ion no such report is available in the literature.

Acknowledgements

DST and CSIR (India) have supported this work. P. D and A. G. Acknowledge CSIR, New Delhi for Sr. research fellowship.

References

1. (a) For selected examples, see: J. J. R. Frausto da Silva and R. J. P. Williams, The Biological chemistry of elements: The Inorganic Chemistry of Life, Oxford University Press, Oxford, 1993; (b) M. Olivares and M. Gonzalez, Am. J. Clin, Nutr, 1998, 67, 952S; (c) M. C. Linder and M. Hazegh-Azam, Am. J. Clin. Nutr., 1996, 63, 797S; (d) Bioinorganic chemistry, ed. I. Bertini, H. B. Gray, S. J. Lippard and J. S. Valentine, University science book, Mill Valley, CA. 1994; (e) Functional Synthetic Receptors ed. T. Schraderr and A. D. Hamilton, Wiley-VCH: Weinheim, Germany, 2005; (f) J. P. Desergne and A. W. Czarnik, Chemosensors of ion and molecule recognition; Kluwer: Dordrecht, 1997.
2. (a) E. M. Nolan and S. J. Lippard, Chem. Rev., 2008, 108, 3443; (b) Z. Yang, C. Yan, Y. Chen, C. Zhu, C. Zhu, C. Zhang, X. Dong, W. Yang, Z. Guo, Y. Lu and W. He, Dalton Trans, 2011, 40, 2173; (c) D. W. Domaille, E. L. Que and C. J. Chang, Nat. Chem. Biol., 2008, 4, 168; (d) B. Valeur and I. Leary, Coord. Chem. Rev., 2000, 205, 3; (e) J. W. Lee, H. S. Jung, P. S. Kwon, J. W. Kim, R. A. Bartsch, Y. Kim, S. J. Kim and J. S. Kim, Org. Lett., 2008, 10, 3801; (f) A. P. De Silva, H. Q. N. Gunaratne, T. Gunnlaugsson, A. J. M. Huxley, C. P. McCoy, J. T. Rademacher and T. E. Rice, Chem. Rev., 1997, 97, 1515; (g) T. W. Bell and N. M. Hext, Chem. Soc. Rev., 2004, 33, 589; (h) M. Suresh, S. Mishra, S. K. Mishra, E. Suresh, A. K. Mondal, A. Shrivastav and A. Das, Org. Lett., 2009, 11, 2740.
3. (a) P. M. Kopittke, C. J. Asher and N. W. Menizes, Plant Soil, 2007, 292, 283; (b) M. B. Jones and T. C. Hutchinson, New Phytol., 1988, 108, 451; (c) B. J. Alloway, E.Steinnes, Cadmium in Soils and Plant, Kluwer Academic Publishers, Boston, 1999; (d) G. F. Nordberg, R. F. M. Herber and L. Alessio, Cadmium in the Human Environment, Oxford University Press, Oxford, 1992.
4. (a) O. Malm, Environ. Res., 1998, 77, 73; (b) Mercury Update Impact of Fish Advisories: EPA Fact Sheet, EPA-823-F-01-011, EPA Office of Water, Washington, DC, 2001; (c) P. Chu and D. B. Porcella, Water, Air, Soil Pollut., 1995, 80, 135; (d) A. Renzoni, F. Zino and E. Fraanchi, Environ. Res., 1998, 77, 68; (e) T. W. Clarkson, L. Magoos and G. J. N. Myers, N. Engl. J. Med., 2003, 349, 1731; (f) M. Suresh, A. K. Mondal, S. Saha, E. Suresh, A. Mondali, R. D. Liddo, P. P. Parnigotto and A. Das, Org. Lett., 2010, 12, 5406.
5. M. Zayed and T. Norman, Plant Soil, 2003, 249, 139.
6. (a) H. H. Harris, I. J. Pickering and G. N. George, Science, 2003, 301, 1203; (b) M. Harada, Crit. Rev. Toxicol., 1995, 25, 1; (c) G. E. McKeown- Eyssen, J. Ruedy and A. Neims, Am. J. Epidemol., 1983, 118, 470; (d) P. Grandjean, P. Weihe, R. F. White and F. Debes, Environ. Res., 1998, 77, 165.
7. M. Nendza, T. Herbst, C. Kussatz and A. Gies, Chemosphere, 1997, 35, 1875.
8. (a) P. B. Tchounwou, W. K. Ayensu, N. Ninashvili and D. Sutton, Environ. Toxicol., 2003, 18, 149; (b) M. Farina, F. A. Soares, A. Feoli, C. Roehring, A. M. Brusque, L. Rotta, M. L. Perry, D.O. Souja and J. B. T. Rocha, Nutrition, 2003, 19, 531.
9. (a) P. Mahato, A. Ghosh, S. Saha, S. Mishra, S. K. Mishra and A. Das, Inorg. Chem., 2010, 49, 11485; (b) M. Suresh, A. Shrivastava, S. Mishra, E. Suresh and A. Das, Org. Lett., 2008, 10, 3013; (c) M. Suresh, S. K. Mishra, S. Mishra and A. Das, Chem. Commun., 2009, 2496; (d) J. Li, J. Yao and W. Zhong, Chem. Commun., 2009, 4962; (e) D. Wu, A. B. Descalzo, F. Weik, F. Emmerling, Z. Shen, X. Z. You and K. Rurack, Angew. Chem., Int. Ed., 2008, 47, 193; (f) H. Zheng, Z-H. Qian, L. Xu, F-F. Yuan, L-D. Lan and J-G. Xu, Org. Lett., 2006, 8, 859; (g) X. Guo, X. Qian and L. Jia, J. Am. Chem. Soc., 2004, 126, 2272; (h) J. Mao, L. Wang, W. Dou, X. Tang, Y. Yan and W. Liu, Org. Lett., 2007, 9, 4567; (i) M. Sarkar, S. Banthia and A. Samanta, Tetrahedron Lett., 2006, 47, 7575; (j) Z. Zhou, M. Yu, H. Yang, K. Huang, F. Li, T. Yi and C. Huang, Chem. Commun., 2008, 3387; (k) C. S. Lim, D. W. Kang, Y. S. Tian, J. H. Han, H. L. Hwang and B. R. Cho, Chem. Commun., 2010, 46, 2388.
10. (a) M. Suresh, A. Ghosh and A. Das, Chem. Commun., 2008, 3906; (b) I-T. Ho, G. H. Lee and W. S. Chung, J. Org. Chem., 2007, 72, 2434; (c) Y. F. Cheng, D. T. Zhao, M. Zhang, Z-Q. Liu, Y. F. Zhou, T. M. Shu, F. Y. Li, T. Yi and C-H. Haung, Tetrahedron Lett., 2006, 47, 6413; (d) J. Huang, Y. Xu and X. Qian, J. Org. Chem., 2009, 74, 2167; (e) Y. Shiraishi, S. Sumiya, Y. Kohono and T. Hirai, J. Org. Chem., 2008, 73, 8571; (f) B. N. Ahamed, M. Arunachalam and P. Ghosh, Inorg. Chem., 2010, 49, 4447; (g) B. N. Ahamed and P. Ghosh, Inorg. Chim. Acta, 2011, 372, 100.
11. P. Das, A. Ghosh and A. Das, Inorg. Chem., 2010, 49, 6909.
12. (a) P. Kar, S. Verma, A. Sen, A. Das, B. Ganguly and H. N. Ghosh, Inorg. Chem., 2010, 49, 4167; (b) N. R. M. Simpson, M. D. Ward, A. F. Morales and F. Barigelletti, J. Chem. Soc., Dalton Trans., 2002, 2449; (c) M. D. Ward, Coord. Chem. Rev., 2006, 250, 3128.
13. (a) Y. Lingappa, S. Rao, S. R. V. Ravikumar and P. A. Rao, Radiat. Eff. Defects Solids, 2007, 162, 11; (b) G. Narain and P. Kla, Aust. J. Chem., 1967, 20, 227.
14. (a) Z. Liu, L. Jiang, Z. Lianga and Y. Gaoa, Tetrahedron, 2006, 62, 3214; (b) B. Valeur, J. Pouget and J. Bouson, J. Phys. Chem., 1992, 96, 6545.
15. S. K. Ko, Y.K. Yang, J. S. Tae and I. J. Sin, J. Am. Chem. Soc., 2006, 128, 14150.
16. (a) M. H. Lee, S. J. Lee, J. H. Jung, H. Lim and J. S. Kim, Tett. Lett., 2007, 120, 87; (b) E. Coronado, J. R. Galán-Mascaro, C. M. Gastaldo, E. Palomares, J. R. Durrant, R. Viler, M. Gratzel and M. K. Nazeeruddin, J. Am. Chem. Soc., 2005, 127, 12351; (c) D. Wu, W. Huang, Z. Lin, C. Duan, C. He, S. Wu and D. Wang, Inorg. Chem., 2008, 47, 7190; (d) R. Shunmugam, G. J. Gabriel, C. E. Smith, K. A. Aamer and G. N. Tew, Chem.–Eur. J., 2008, 14, 3904.
17. (a) Y. Shraishi, H. Maehara, K. Ishizumi and T. Hirai, Org. Lett., 2007, 9, 3125; (b) A. K. Mondal, M. Suresh, E. Suresh, S. K. Mishra, S. Mishra and A. Das, Sens. Actuators, B, 2010, 145, 32.
18. (a) H. A. Benesi and J. H. Hildebrand, J. Am. Chem. Soc., 1949, 71, 2703; (b) C. Yang, L. Liu, T. W. Mu and Q.-X. Guo, Anal. Sci., 2000, 16, 537; (c) M. I. Rodriguez-Caceres, R. A. Agbaria and I. M. Warner, J. Fluoresc., 2005, 15, 185.

---

Footnote

† Electronic Supplementary Information (ESI) available: characterisation of L, L₁ and L₂, ¹H NMR scanning of L and L₂ with Hg²⁺ , non-linear least squares method from UV–vis titration UV spectra of L in the presence of Cr³⁺ and Hg²⁺]. See DOI: 10.1039/c2ra00788f/

---