A carbazole based “Turn on” fluorescent sensor for selective detection of Hg²⁺ in an aqueous medium

Abstract

We developed a multimode Hg²⁺ selective sensor system based on carbazole-barbituric acid conjugate (CBA). Its colorimetric and fluorescent behavior for Hg²⁺ ions in aqueous medium was investigated. Upon Hg²⁺ binding, CBA exhibited significant change in the absorption spectrum along with an enhanced emission at 593 nm. This allows the selective and sensitive detection of Hg²⁺ without any interference from other metal ions.

---

Heavy metals, which are toxic to the environment and ecosystem, have been dangerously accumulating in the biosphere as a result of human activity. They contaminate the food chain and cause persistent damage to living beings. Hence, investigations on the mechanism of accumulation and transmission of toxic metals in the environment and precautionary measures thereby are active areas of research these days. With the advent of sensitive optical imaging tools that are capable of assaying heavy and transition metal ions in biological samples, there is a renewed interest in the development of selective and sensitive fluorescent chemosensors for the rapid identification of the toxic heavy metal¹ ions, such as the Pb²⁺, Cd²⁺, and Hg²⁺. Among these, mercury contamination which is found to be widespread, originates from a variety of natural as well as anthropogenic sources including volcanic eruptions, incineration of urban solid wastes, mining and improper management of e-wastes. Mercury that ends up in the marine environment, by bacterial action gets converted into organic mercury (methylmercury), which is lipophilic and easily enters the food chain and thus accumulates in higher organisms, in particular large fishes. Mercury poisoning through food-chain and high dose exposure, can result in several diseases, including acrodynia, Hunter-Russell syndrome, and Minamata disease.² Studies carried out in patients inflicted with mercury poisoning, have shown that mercury concentrates more in blood and brain, as seen in the case of certain patients with Alzheimer's disease.³ The maximum allowed level of mercury(II) ions in potable water prescribed by EPA is 2 ppb.⁴ Detection and quantification of mercury is thus necessary for monitoring and preventing the contamination of the environment and the living world.

Scheme 1 Binding mechanism of CBA towards Hg²⁺.

Fluorescent chemosensors of metal ions consist of a fluorophore moiety and a receptor moiety. The fluorescence emission gets altered in terms of energy or intensity as a result of metal ion binding. Depending on the design of the chemosensor system, such changes in the fluorescence may be due to photoinduced electron/charge transfer, energy transfer, excimer/exciplex formation or dissociation or charge induced changes in the polarity of the environment etc.⁵ The commonly used fluorophores are 1,8-napthalimide, coumarin, pyrene, anthracene, BODIPY, squaraine, xanthanes, cyanine, rhodamine, fluorescein etc.^6–8 As fluorescent chemosensors are hydrophobic in nature, most fluorescent chemosensors require organic solvents for proper operation due to low solubility in aqueous medium. In recent years, hydrophilic biomolecules such as amino acids, peptides, and DNA have been used as receptors in the design of fluorescent chemosensors. These molecules and supramolecular assemblies of natural origin have potent binding affinities to specific metal ions, higher solubility in aqueous and other polar solvents and biological compatibility required for in vivo or in vitro tissue imaging applications. Such chemosensors showed sensitive responses to heavy metal ions in aqueous solutions.⁹ For practical applications, two major requirements to avoid filter effects are, having a long wavelength emission maximum along with an enhanced emission intensity in the presence of the metal ion. A majority of fluorescent chemosensors show quenching of fluorescence due to the heavy atom effect. In 2013 S. Li and co-workers¹⁰ reported a barbituric acid derivative (AnB) having anthracene as the fluorophore, that can selectively detect mercury cation, by spectrofluorometric method via the formation of Hg²⁺–AnB coordination polymers. We report for the first time a cost effective carbazole based chemosensor (CBA) which can detect Hg²⁺ ions with high degree of sensitivity and selectivity in aqueous methanolic medium, showing over 100 fold enhancement in emission (Scheme 1).

The carbazole barbituric acid derivative CBA was synthesized according to a reported procedure by Knoevenagel condensation between 9-octyl-9H-carbazole-3-carbaldehyde (1) and barbituric acid (2) in ethanol (Scheme 2).¹¹ The product was filtered and purified to get a yield of 85%. The structure of compound CBA was confirmed by ¹H and ¹³C-NMR spectroscopy and mass spectrometric data (S1, S2 and S3, ESI†)

Scheme 2 Synthesis of CBA.

The absorption spectrum of CBA is characteristic of a π-conjugated donor–acceptor molecule with a prominent charge transfer band at 438 nm (ε_max, 3.16 × 10⁴ cm⁻¹ M⁻¹) in methanol and is sensitive to solvent polarity. The molecule showed a weak emission with λ_max = 521 nm in methanol. In order to assure an aqueous medium for the metal binding studies, a methanol water mixture of 1 : 1 (v/v) ratio was used. It was observed that as the water content increased, it resulted in the precipitation of the dye. The variation in the aqueous content during the addition of metal ions was kept at 0.19%. Blank experiments showed no change in the absorption and emission properties of the dye in this range. The absorption maximum observed for the dye in this solvent composition was at 452 nm. The observed emission was very weak (Φ_f = 0.0016) and broad, with a maximum at 533 nm (S4, ESI†).

The interaction of CBA with an aqueous solution of mercury(II) acetate was investigated by spectrophotometric and spectrofluorometric titrations in aqueous methanol (Fig. 1). During the photometric titration of mercury ions with CBA, the light yellow solution turned colourless along with a blue-shift of the long-wavelength absorption maximum from 452 nm to 438 nm. The isosbestic point at 408 nm indicates formation of a complex with Hg²⁺ ions which is in equilibrium with the free ligand. As it is evident from the plot of absorbance at 452 nm vs. number of molar equivalents of Hg²⁺, no further change in the absorbance was observed after the addition of about 0.7 equivalence of Hg²⁺ (Fig. 1(a) inset). Since this value is less than one, it can be assumed that a 2 : 1 complex between CBA and Hg²⁺ i.e., CBA–Hg–CBA by replacing two acetate anions is formed. The blue-shift in the absorption spectrum is indicative of the formation of an aggregate of CBA or due to a reduction in the electron affinity of the pyrimidine moiety upon complexation with Hg²⁺ ions. We further investigated the emission spectral characteristics and its response to Hg²⁺ in aqueous methanol solutions. Fluorescence spectra were recorded by exciting optically matched solutions of CBA (1.0 × 10⁻⁵ M) at the isosbestic point (408 nm). This was done to rule out fluorescence intensity variation due to changes in the absorbance at the excitation wavelength. With increasing concentration of Hg²⁺ ions the weak fluorescence of CBA was found to increase along with a 60 nm red shift in the emission maximum to 593 nm (Fig. 1(b)). The intensity variation at 593 nm was plotted against the number of equivalents of Hg²⁺ ions (inset in Fig. 1(b)). The dramatic 150 fold enhancement in intensity at 593 nm has reached a plateau after the addition of about 0.7 equivalents of Hg²⁺ ions. Fluorescence quantum yield of the complex under this condition was determined as 0.195 (ESI† section 1.4.1) which is 122 times higher than free CBA. This result agrees with the outcome of spectrophotometric titration and suggests that CBA shows a very strong affinity to Hg²⁺ ions and is also indicative of a complex formation equilibrium involving two CBA molecules and one Hg²⁺ ion with a 2 : 1 binding stoichiometry. The colorimetric response as well as the fluorescence response under UV illumination of CBA to Hg²⁺ in 1 : 1 methanol water solution, is visible even with the naked eye as showed in the Fig. 2(a) and (b), respectively.

Fig. 1 (a) Evolution of absorption spectra of CBA in the presence of increasing concentration of an aqueous solution of Hg(OAc)₂ in 1 : 1 MeOH H₂O ([CBA] = 30 μM, [Hg(OAc)₂] = 0–19 μM). (b) Evolution of fluorescence spectra of CBA in the presence of increasing concentration of an aqueous solution of Hg(OAc)₂ in 1 : 1 MeOH H₂O ([CBA] = 10 μM, [Hg(OAc)₂] = 0–8 μM, λ_ex = 408 nm). Arrows indicate the direction change in the absorbance or fluorescence intensity.

Fig. 2 Photographs of CBA and CBA in the presence of Hg²⁺ ions under (a) normal light and (b) UV 365 nm light.

In a related work, Li and co-workers have identified formation of a coordination polymer of Hg²⁺ ions with an anthracene-barbituric acid conjugate. According to this report, the polymer formation involved deprotonation of N–H protons of the barbituric acid moiety and the interaction was proposed to have a 1 : 1 stoichiometry.¹⁰ Our study of spectrophotometric titration data by continuous variation of mole fraction of Hg²⁺ in comparison to CBA and the Job plot thus obtained, fitted very well with a model corresponding to a 2 : 1 CBA–Hg²⁺–CBA complex (S5, ESI†). This result indicates a mechanism that involves deprotonation of NH protons of the barbituric acid moiety. The complex formation could be a stepwise formation of the 1 : 1 complex CBA–Hg²⁺–OAc followed by the formation of the 2 : 1 complex CBA–Hg²⁺–CBA. In this mechanism two possibilities exists. In the first one the two binding interactions occur with comparable rate constants. In the second mode one of the binding interactions is very fast compared to the second binding interaction. However, the observation of the well-focused isosbestic point at 408 nm with a very small standard deviation (0.003 absorbance units) suggests the second mechanism. Here an apparent equilibrium between CBA and CBA–Hg²⁺–CBA is involved and the formation of the 1 : 1 complex is not observable by the steady state UV-Vis spectroscopy. In order to add further insight into the structure of the complex formed between the CBA and Hg²⁺ ions and the complexation mechanism, we have recorded ¹H-NMR spectrum in dmso-d₆ in the presence of increasing concentration of Hg(II) acetate. Fig. 3 shows ¹H-NMR spectrum of CBA and mixtures of CBA and Hg(II) acetate at different ratios from 0 to 0.5 equivalents of Hg²⁺. The spectrum of free CBA is characterized by two distinct peaks corresponding to the two NH protons at 11.15 ppm and 11.27 ppm. Analysis of the spectra evolved in the presence of Hg²⁺ acetate shows a new resonance peak at 12 ppm, 11.5 ppm and 1.95 ppm. The peak at 12 ppm is assigned to the –OH proton of acetic acid which is indicative of a mechanism of Hg²⁺ binding involving the deprotonation of NH of the barbituric acid moiety releasing a molecule of acetic acid. Moreover, the binding event led to deprotonation of N–H protons on CBA as indicated by the decrease in the peak area of two N–H protons and a proportional increase in the area of the peak corresponding to the –OH proton of acetic acid. A careful look at these spectra reveals a small peak appear at 11.5 ppm after the addition of Hg²⁺ acetate and this may be due to the 1 : 1 CBA–Hg²⁺–OAc complex. This is observed probably due to the relatively high concentrations used for the NMR experiments or the second reaction may be slow enough leaving some 1 : 1 complex to be observed under the time scale of the NMR experiment. When 0.5 equivalents of Hg(OAc)₂ was added, the two NH protons completely vanished and the CBA–Hg²⁺–CBA complex precipitated which is undoubtedly seen in ¹H-NMR, i.e. only acetic acid peaks and solvents peaks are seen in the resulting spectrum. ¹H NMR of the precipitate was taken in THF-d₈ was indicative of a symmetrical structure with one free NH proton (Fig. S6, ESI†). Further corroborative evidence for the mercury complex was observed in the MALDI-TOF mass spectrum which showed a mass peak at m/z = 1082.0839 corresponding to [2CBA + Hg + 2H + 2Na − 2H] fragment, where the calculated molecular weight is 1082.3584 and 1066.1122 corresponding to [2CBA + Hg + MeOH − 2H] fragment, where the calculated molecular weight is 1066.3917 (Fig. S7 and Table S1, ESI†). The results of the ¹H-NMR titration and MALDI-TOF results both suggest a 2 : 1 binding stoichiometry of CBA with Hg²⁺. Such mercuration by deprotonation and insertion between T–T base pairs has observed in DNA double helices with a linear T–Hg–T geometry.¹² Similarly in thiamine–graphene conjugates a mercury induced deprotonation and formation of a 2 : 1 complex with a bend structure has been reported.¹³

Fig. 3 The ¹H-NMR spectra of the CBA and CBA in the presence of 0.13, 0.25, 0.5 equivalents of Hg²⁺ acetate (400 MHz, dmso-d₆).

The observation of fluorescence enhancement is contrary to what is generally observed. Usually heavy metals quench the fluorescence of the molecules due to heavy atom effect. A plausible reason for the observed enhanced emission could be the formation of an aggregate or the complex having a linear or bend geometry (Fig. S8, ESI†). Such geometries can lead to close interaction between two carbazole moieties. Further, association of non-polar n-octyl chains assists in bringing two carbazole moieties closer favouring aggregation induced emission.

Recently, Tang and co-workers¹⁴ and Park and co-workers¹⁵ have individually presented that some flexible molecular systems with weak emission are capable of showing strong fluorescence at high concentration due to aggregation. This phenomenon finds applications in numerous designs of fluorescent sensors and solid state lighting devices.^7a To establish aggregation induced emission enhancement (AIEE) of a compound, its fluorescent behaviour is studied with a poor solvent added to a solution of the compound.¹⁶ As CBA is insoluble in water, increasing the water fraction in the mixed solvent can change its existing form from a solution state in pure MeOH to aggregated particles in mixtures with high water content. The emission spectra of CBA in the MeOH/H₂O mixtures with different water contents are shown in Fig. 4. Fig. 4(a) is the absorption spectral changes observed as a function of percentage of water content. Here the absorption spectrum initially showed a red shift to 450 nm, typical for a charge transfer band that responds to increase in polarity. However addition of water above 50% lead to a blue shift to 435 nm for 70% of water and higher water content made the spectrum more flat and finally the dye precipitated. The fluorescence spectral change was more dramatic and the fluorescence emission showed a sudden enhancement in the intensity in a 60% water methanol mixture along with a red shift to 600 nm (Fig. 4(b) inset). This enhancement in luminescence is due to the aggregation effect. Thus, it is clear that aggregation can lead to such enhancement in emission in CBA. Such stacking between two CBA units is possible in the CBA–Hg²⁺–CBA complex. More importantly, the enhancement of fluorescence intensity at 593 nm of CBA corresponds to the concentration of Hg²⁺ in a linear manner (linearly dependent coefficient: R² = 0.994). This indicate that the CBA can be used to detect and quantify Hg²⁺ with a detection limit of 8.89 nM (Fig. S9, ESI†). The association constant K_a was determined by non-linear least square fit to a model described for 2 : 1 complexation stoichiometry (section 1.5 & Fig. S10, ESI†). A remarkably high value of 8.73 × 10¹³ M⁻² (R² = 0.996) is obtained which show the very high affinity of CBA to Hg²⁺ ions.

Fig. 4 (a) Changes in the absorption spectra and (b) emission spectra of CBA in MeOH/H₂O mixture as a function of percentage of solvent composition (λ_ex = 408 nm) (inset: fluorescence intensity @ 604 nm vs. % water).

For any chemosensor system, reusability is a desired feature. To verify this, CBA–Hg²⁺–CBA solution was subsequently treated with excess cysteamine hydrochloride, which holds a thiol group which is a strong ligand to Hg²⁺ ions. The strong fluorescence of the CBA–Hg²⁺–CBA was almost quenched and the absorption spectrum showed a red shift indicating a decomplexation releasing free CBA thereby, establishing CBA as a reversible chemosensor (S11, ESI†).

To evaluate the selectivity of the fluorescent probe CBA towards Hg²⁺ ions, control experiments were performed, in which the influence of the addition of acetate salts of Na⁺, Mg²⁺, Ca²⁺, Cd²⁺, Zn²⁺, Ni²⁺, Ba²⁺, Cu²⁺, Pb²⁺, Co²⁺, and Mn²⁺ on the absorption and emission properties of CBA was studied (S12, ESI†). No change in fluorescence intensity was observed for a metal ion concentration up to 35 μM except for the Hg²⁺ salts. The chloride, sulfate and perchlorate salts of mercury were also used and showed similar enhancement in fluorescence. The bar-diagram in Fig. 5 illustrates the selectivity feature of CBA in the detection of Hg²⁺ ions.

Fig. 5 The bar diagram showing the selectivity of CBA in the detection of Hg²⁺ ions.

In conclusion, we present a new simple colorimetric and fluorescent sensor for Hg²⁺ ions. The sensor CBA can be easily synthesized using inexpensive chemicals with a high yield. The remarkable binding affinity for Hg²⁺ ions (K_a = 8.73 × 10¹³ M⁻²) with a lower detection limit of 8.9 nM makes CBA a selective and sensitive chemosensor with great potential to be used as a probe in fluorescence imaging and as a reusable probe in devices for online monitoring of mercury pollution.

Acknowledgements

This work was supported by Cochin University of Science and technology (CUSAT) and University Grants Commission (UGC), Government of India. We thank STIC & DST-SAIF, CUSAT and Dr C. T. Aravind Kumar, School of Environmental Sciences, Mahatma Gandhi University, Kottayam, Kerala for analytical facilities.

References

1. (a) N. M. M. Moura, C. Núñez, S. M. Santos, M. A. F. Faustino, J. A. S. Cavaleiro, M. G. P. M. S. Neves, J. L. Capelo and C. Lodeiro, Inorg. Chem., 2014, 53, 6149; (b) C. Y. Li, Y. Zhoua, Y. F. Li, X. F. Konga, C. X. Zoua and C. Wenga, Anal. Chim. Acta, 2013, 774, 79.
2. M. Harada, Minamata Disease: Methylmercury Poisoning in Japan Caused by Environmental Pollution, Crit. Rev. Toxicol., 1995, 25, 1.
3. S. Hardy and P. Jones, Capillary electrophoresis determination of methylmercury in fish and crab meat after extraction as the dithizonesulphonate complex, J. Chromatogr. A, 1997, 333.
4. (a) Guidelines for drinking-water quality, ed. M. Sheffer, World Health Organization, Geneva, 3rd edn, 2008; (b) US Environmental Protection Agency, EPA Office of Water, Washington, DC, http://www.epa.gov/ogwdw000/contaminants/index.html#mcls.
5. (a) E. M. Nolan and S. Lippard, Chem. Rev., 2008, 108, 3443; (b) H. N. Kim, W. X. Ren, J. S. Kim and J. Y. Yoon, Chem. Soc. Rev., 2012, 41, 3210; (c) S. Karuppannan and J. C. Chambron, Chem.–Asian J., 2011, 6, 964; (d) Q. Zhao, F. Li and C. Huang, Chem. Soc. Rev., 2010, 39, 3007.
6. (a) M. Tian, L. Liu, Y. Li, R. Hu, T. Liu, H. Liu, S. Wang and Y. Li, Chem. Commun., 2014, 50, 2055; (b) K. Bera, A. K. Das, M. Nag and S. Basak, Anal. Chem., 2014, 86, 2740; (c) V. Bhalla, M. Kumar, P. R. Sharma and T. Kaur, Inorg. Chem., 2012, 51, 2150.
7. (a) M. Sanjoy and T. Pakkirisamy, Chem. Commun., 2013, 49, 7292; (b) M. H. Yang, P. Thirupathi and K. H. Lee, Org. Lett., 2011, 13, 5028.
8. (a) D. Wu, W. Huang, C. Duan, Z. Lin and Q. Meng, Inorg. Chem., 2007, 46, 1538; (b) M. H. Lee, S. W. Lee, S. H. Kim, C. Kang and J. S. Kim, Org. Lett., 2009, 11, 2101; (c) X. Chen, X. Meng, S. Wang, Y. Cai, Y. Wu, Y. Feng, M. Zhua and Q. Guob, Dalton Trans., 2013, 42, 14819; (d) M. H. Lee, B. K. Cho, J. Yoon and J. S. Kim, Org. Lett., 2007, 9, 4515.
9. L. N. Neupane, J. Y. Park, J. H. Park and K. H. Lee, Org. Lett., 2013, 15, 254.
10. J. Wang, L. Zhang, Q. Qi, S. Li and Y. Jianga, Anal. Methods, 2013, 5, 608.
11. M. A. Tehfe, F. Dumur, B. Graff, F. M. Savary, D. Gigmes, J. P. Fouassier and J. Lalevee, Polym. Chem., 2013, 4, 3866.
12. (a) A. Ono, H. Torigoe, Y. Tanaka and I. Okamoto, Chem. Soc. Rev., 2011, 40, 5855; (b) T. Marino, J. Mol. Model., 2014, 20, 2303.
13. D. Dinda, B. K. Shaw and S. K. Saha, ACS Appl. Mater. Interfaces, 2015, 7, 14743.
14. (a) J. Luo, Z. Xie, J. W. Y. Lam, L. Cheng, H. Chen, C. Qin, H. S. Kwok, X. Zhan, Y. Liu, D. Zhu and B. Z. Tang, Chem. Commun., 2001, 1740; (b) Y. Hong, J. W. Y. Lam and B. Z. Tang, Chem. Commun., 2009, 4332; (c) J. Shi, N. Chang, C. Li, J. Mei, C. Deng, X. Luo, Z. Liu, Z. Bo, Y. Q. Dong and B. Z. Tang, Chem. Commun., 2012, 48, 10675.
15. (a) B. K. An, S. K. Kwon, S. D. Jung and S. Y. Park, J. Am. Chem. Soc., 2002, 124, 14410; (b) B.-K. An, S.-K. Kwon and S. Y. Park, Angew. Chem., Int. Ed., 2007, 46, 1978.
16. (a) X. Zhang, Z. Chi, J. Zhang, H. Li, B. Xu, X. Li, S. Liu, Y. Zhang and J. Xu, J. Phys. Chem. B, 2011, 115, 7606; (b) G. Singh, V. Bhalla and M. Kumar, Phys. Chem. Chem. Phys., 2015, 17, 22079.

---

Footnote

† Electronic supplementary information (ESI) available: Synthetic details, UV-Vis and fluorescence spectra, binding constant calculation, Job's plot etc. See DOI: 10.1039/c5ra27530j

---