Low Molecular Weight Fluorescent Probes (LMFPs) to Detect the Group 12 Metal Triad

Fluorescence sensing, of d-block elements such as Cu2+, Fe3+, Fe2+, Cd2+, Hg2+, and Zn2+ has significantly increased since the beginning of the 21st century. These particular metal ions play essential roles in biological, industrial, and environmental applications, therefore, there has been a drive to measure, detect, and remediate these metal ions. We have chosen to highlight the low molecular weight fluorescent probes (LMFPs) that undergo an optical response upon coordination with the group 12 triad (Zn2+, Cd2+, and Hg2+), as these metals have similar chemical characteristics but behave differently in the environment.


Introduction
Sensors are used in every aspect of modern life; for example, many modern households have a carbon monoxide sensor or a smoke detector installed. In industry, fiber-optic sensors are used for monitoring process variables, such as temperature, pressure flow, and the level of a liquid. The environment requires sensors to monitor toxic gases and air pollution, and in clinical applications for the detection of medical conditions, i.e., Type 1 diabetes.
The field of sensor technology has rapidly been expanding over the last several decades, and the field can be categorized into two broad areas (1) chemical sensors and (2) biosensors. Chemical sensors measure, monitor, and identify chemical compounds, whereas biosensors combine a biological component, for example, microorganisms, tissue, cells, organelles, nucleic acids, enzymes, or antibodies [1]. Both types of sensors have the same purpose; to measure an analyte, either qualitatively or quantitatively. Some excellent reviews have covered recent advances in biosensors, and these have been omitted in this article [2][3][4][5][6][7][8][9][10][11][12][13][14].
The word sensor has become more ambiguous over time, and there is no universal definition of the term chemical sensor; the term is highly debated in the scientific community. Therefore, there has been much confusion over the two broad definitions of "chemical sensors" and "biosensor." One definition states that "chemical sensors are miniaturized devices which can deliver real-time and online information in the presence of a specific compound or ions in complex samples" [15]. This definition itself is not concise, as the term device is also unclear. Interestingly, the International Union of Pure and Applied Chemistry (IUPAC) does not have a specific definition of "chemical sensor" defined in the Compendium of Chemical Terminology: "Gold Book" [16]. The term "biosensor" is also broad, but the IUPAC does define a biosensor "as a device that uses specific biochemical reactions mediated by isolated enzymes, immune systems, tissues, organelles or whole cells to detect chemical compounds usually by electrical, thermal or optical signals" [16].
The salicylaldehyde motif is a popular organic functional group, often incorporated into molecular probes [130], as it readily undergoes Schiff base reactions, and therefore, introduces the Lewis basic sites to bind metal ions in a stable chelating ring system. Compound 1, a bis(bidentate) iminophenol ligand was synthesized by Chakraborty et al. (Figure 1A) [131], through the condensation of salicaldehyde and 4-aminobenzylamine in ethanol. The X-ray crystal structure shows that two resonance-assisted hydrogen bonding (RAHB) interactions are present. Therefore, in order for compound 1 to act as a molecular probe, Et3N was added to the DMF solution to disrupt the intramolecular hydrogen bonding interaction to produce the chelating motif. Upon the addition of Zn 2+ ions the absorbance wavelength, initially seen at 321 nm, bathochromically shifted to 406 nm. Unfortunately, the X-ray structure of the zinc complex could not be solved as the crystals diffracted weakly, but the analogous Co 2+ complex was solved, showing a 2:2 binding ratio in the solid state. Based on the Co 2+ structure (a tetrahedral arrangement around the Co 2+ ions) and the Zn 2+ ions UV-Vis titration data, the authors speculated that a 1:1 binding stoichiometry occurs in the solution phase. Their UV-Vis does show a pseudo-isosbestic point at 365 nm, but yet the binding isotherm saturates upon the addition of two equivalences of Zn 2+ ions. Therefore, the author's interpretation of a 1:1 binding is misleading. No calculated binding constants were reported, presumably due to a number of different species existing in solution, rendering it difficult to calculate a precise binding constant, and their claim of 1:1 binding mode is questionable. The free ligands (1) are essentially nonfluorescent (φ = 0.025 in DMF) but on addition of Zn 2+ ions, a new emission band is observed at 527 nm (λex = 406 nm), Figures 1B and 1C. This increase did not occur in the presence of the other metals studied (Fe 3+ , Co 2+ , Cu 2+ , Ni 2+ , Cr 2+ Mn 2+ , Cd 2+ , Hg 2+ , and Pb 2+ ) and the authors attributed this to a CHEF fluorescence mechanism. Unfortunately, no fluorescence binding isotherms were reported and, upon closer inspection of their titration data, the fluorescence spectra seemed to increase continuously. This could be due to an aggregated fluorescence enhancement, as it is reasonable to assume that other polymeric species are present in solution, and not only in the discrete 2:2 model, which has been suggested from their solid-state studies. This might explain why the binding constants were not calculated, as they could not be easily modeled. Adapted from [131], with permission from Elsevier. Adapted from [131], with permission from Elsevier. Zhang, Zhao, and Zhou prepared chemosensor 2 (Scheme 1), by incorporating a binding motif to the dye [132]. The optical studies were carried out in a CH 3 CN-HEPES buffer (6:4 v/v; pH 7.4). On adding the 20 equivalents of metal salts (Na + , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , Pb 2+ , Cd 2+ , Ag + , Hg 2+ , Fe 2+ , Mg 2+ , and Cr 3+ ), negligible emission enhancement was seen, except for in Zn 2+ ions. The emission spectra of LMFP 2 was very weak (φ = 0.06), and was attributed to the PET quenching by the lone pair of electrons residing on the nitrogen atom on the imine group. A 13-fold increase in the fluorescence intensity was seen at 556 nm (λ ex = 397 nm) when 20 equivalents of Zn 2+ ions were added. The binding constant K a for a 1:1 binding model was calculated to be approximately 3 × 10 3 M −1 . The stoichiometry was supported by a Job's plot analysis. As Zn 2+ was the only species that produced a fluorescent signature, the authors investigated the biological imaging of compoun 2, with an array of cell lines. These included adenocarcinomic human alveolar basal epithelial cells (A549), bronchial epithelial cell (BEAS-2B), Chinese hamster ovary (CHO), Hela, and human liver cancer cells (HEP G2). Compound 2 showed different fluorescence emission colors in the cell nucleus and in the cytoplasm. Zhang, Zhao, and Zhou prepared chemosensor 2 (Scheme 1), by incorporating a binding motif to the dye [132]. The optical studies were carried out in a CH3CN-HEPES buffer (6:4 v/v; pH 7.4). On adding the 20 equivalents of metal salts (Na + , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , Pb 2+ , Cd 2+ , Ag + , Hg 2+ , Fe 2+ , Mg 2+ , and Cr 3+ ), negligible emission enhancement was seen, except for in Zn 2+ ions. The emission spectra of LMFP 2 was very weak (φ = 0.06), and was attributed to the PET quenching by the lone pair of electrons residing on the nitrogen atom on the imine group. A 13-fold increase in the fluorescence intensity was seen at 556 nm (λex = 397 nm) when 20 equivalents of Zn 2+ ions were added. The binding constant Ka for a 1:1 binding model was calculated to be approximately 3 × 10 3 M −1 . The stoichiometry was supported by a Job's plot analysis. As Zn 2+ was the only species that produced a fluorescent signature, the authors investigated the biological imaging of compoun 2, with an array of cell lines. These included adenocarcinomic human alveolar basal epithelial cells (A549), bronchial epithelial cell (BEAS-2B), Chinese hamster ovary (CHO), Hela, and human liver cancer cells (HEP G2). Compound 2 showed different fluorescence emission colors in the cell nucleus and in the cytoplasm. Scheme 1. The proposed binding motif between compound 2 and Zn 2+ ions. The binding environment was also confirmed by 1 H NMR spectroscopy.
It is well-known that UV-visible irradiation can induce isomerization of organic and inorganic compounds [108,133]. This switching mechanism has been extensively used in molecular probe design [134]. The spyiropyran functional group is known to undergo a reversible heterolytic cleavage of the spiro C-O bond. A metastable merocyanine is then formed upon cis-trans isomerization. Giordani et al. synthesized a photochromic compound 3 containing a metal binding site, Figure 2A [135]. They utilized a methyl pyridinyl group, often used in zinc selective chromophores and incorporated the pyridinyl group onto the spirochromene indolic nitrogen atom. The stability of the open-closed form of the spiropyrene was affected by the electron withdrawing or electron donating groups attached to it, thus, influencing the feasibility of the open and closed form. Electron donating groups favor the closed system, whereas electron withdrawing groups favor the open form. Therefore, the Giordani team incorporated the electron withdrawing NO2 group into their molecular probe design, to favor the open form and, thus, making the Lewis basic sites available to coordinate to a metal center. Many metal perchlorate salts were added to a 1 × 10 −4 M CH3CN solution of probe 3. The Zn(ClO4)2 salt produced a distinctive color at 504 nm, from which the binding isotherm was obtained, Figure 2B, although these data points were not used to calculate the binding constant. A modified Benesi-Hilderbrand equation was used but the regression line showed curvature so the calculated binding constant (K = 1.6 × 10 4 M −1 ) should be viewed with a little skepticism. Upon the irradiation of the solution, the absorbance band disappeared. The switching was maintained over 100 second period. 1 H NMR studies showed that the trans configuration was sustained when the probe coordinated to the metal ion, confirmed by 1 H NMR, a large coupling constant was calculated (J = 16 Hz).
It is well-known that UV-visible irradiation can induce isomerization of organic and inorganic compounds [108,133]. This switching mechanism has been extensively used in molecular probe design [134]. The spyiropyran functional group is known to undergo a reversible heterolytic cleavage of the spiro C-O bond. A metastable merocyanine is then formed upon cis-trans isomerization. Giordani et al. synthesized a photochromic compound 3 containing a metal binding site, Figure 2A [135]. They utilized a methyl pyridinyl group, often used in zinc selective chromophores and incorporated the pyridinyl group onto the spirochromene indolic nitrogen atom. The stability of the open-closed form of the spiropyrene was affected by the electron withdrawing or electron donating groups attached to it, thus, influencing the feasibility of the open and closed form. Electron donating groups favor the closed system, whereas electron withdrawing groups favor the open form. Therefore, the Giordani team incorporated the electron withdrawing NO 2 group into their molecular probe design, to favor the open form and, thus, making the Lewis basic sites available to coordinate to a metal center. Many metal perchlorate salts were added to a 1 × 10 −4 M CH 3 CN solution of probe 3. The Zn(ClO 4 ) 2 salt produced a distinctive color at 504 nm, from which the binding isotherm was obtained, Figure 2B, although these data points were not used to calculate the binding constant. A modified Benesi-Hilderbrand equation was used but the regression line showed curvature so the calculated binding constant (K = 1.6 × 10 4 M −1 ) should be viewed with a little skepticism. Upon the irradiation of the solution, the absorbance band disappeared. The switching was maintained over 100 second period. 1 H NMR studies showed that the trans configuration was sustained when the probe coordinated to the metal ion, confirmed by 1 H NMR, a large coupling constant was calculated (J = 16 Hz).  (insert). Unfortunately, the calculated binding constant was not calculated using the non-linear regression presented. Adapted from [135], with permission from Elsevier.
Another tridentate coordination moiety was incorporated onto a 1,8 naphthyridine scaffold where a protected L-proline was attached to position two on the naphthlyridine ring system, chemosensor 4 [136]. Upon the addition of various metal salts (Li + , K + Cd 2+ , Mn 2+ , Fe 3+ , Cu 2+ , Mg 2+ , Fe 2+ , Ni 2+ , Co 2+ , Hg 2+ , Pb 2+ , Ag + , and Zn 2+ ) in MeOH, only the Zn 2+ ions showed a significant fluorescence enhancement by a factor of 2.5, whereas all other metal ions quenched emission. The authors suggested that the CHEF mechanism was the reason for the fluorescence enhancement observed. The fluorescence signal was saturated after the addition of one equivalent of Zn 2+ ions and the calculated Ka was approximately 6 × 10 4 M −1 . The coordination environment of compound 4 and the Zn 2+ complex was investigated by IR spectroscopy and TD-DFT. The natural bond orbital (NBO) calculations suggested that the zinc metal ion preferred to coordinate with the two nitrogen atoms and the oxygen atom (the negative charge density on N(2), N(3), and O (2), were calculated to be −0.403, −0.63, and −0.65, respectively), the final coordination site was occupied by an MeOH molecule. Based on these theoretical calculations and, the IR study, they suggested that the Zn 2+ ions had a tetrahedral geometry.
The naphthalene moiety has been used by Kim to prepare two very similar chemosensors, compounds 5 and 6 [137,138]. The hydrazide functional group was utilized, as it readily undergoes a Schiff base condensation reaction, to prepare a tridentate binding site (shown in red). The binding group is shown to be ideal for Zn 2+ ions and the compounds coordinated in a 2:1 fashion. In both examples, the fluorescent mechanism was due to inhibition of C=N isomerization and ESIPT, upon the addition of Zn 2+ ions, which drastically increased the emission intensity, via a CHEF mechanism. The calculated β values were 1 × 10 9 M −2 in a DMSO-buffer and DMF-buffer for compounds 5 and 6, respectively. Unfortunately, the calculated binding constant was not calculated using the non-linear regression presented. Adapted from [135], with permission from Elsevier.
Another tridentate coordination moiety was incorporated onto a 1,8 naphthyridine scaffold where a protected L-proline was attached to position two on the naphthlyridine ring system, chemosensor 4 [136]. Upon the addition of various metal salts (Li + , K + Cd 2+ , Mn 2+ , Fe 3+ , Cu 2+ , Mg 2+ , Fe 2+ , Ni 2+ , Co 2+ , Hg 2+ , Pb 2+ , Ag + , and Zn 2+ ) in MeOH, only the Zn 2+ ions showed a significant fluorescence enhancement by a factor of 2.5, whereas all other metal ions quenched emission. The authors suggested that the CHEF mechanism was the reason for the fluorescence enhancement observed. The fluorescence signal was saturated after the addition of one equivalent of Zn 2+ ions and the calculated K a was approximately 6 × 10 4 M −1 . The coordination environment of compound 4 and the Zn 2+ complex was investigated by IR spectroscopy and TD-DFT. The natural bond orbital (NBO) calculations suggested that the zinc metal ion preferred to coordinate with the two nitrogen atoms and the oxygen atom (the negative charge density on N(2), N(3), and O(2), were calculated to be −0.403, −0.63, and −0.65, respectively), the final coordination site was occupied by an MeOH molecule. Based on these theoretical calculations and, the IR study, they suggested that the Zn 2+ ions had a tetrahedral geometry. . Unfortunately, the calculated binding constant was not calculated using the non-linear regression presented. Adapted from [135], with permission from Elsevier.
Another tridentate coordination moiety was incorporated onto a 1,8 naphthyridine scaffold where a protected L-proline was attached to position two on the naphthlyridine ring system, chemosensor 4 [136]. Upon the addition of various metal salts (Li + , K + Cd 2+ , Mn 2+ , Fe 3+ , Cu 2+ , Mg 2+ , Fe 2+ , Ni 2+ , Co 2+ , Hg 2+ , Pb 2+ , Ag + , and Zn 2+ ) in MeOH, only the Zn 2+ ions showed a significant fluorescence enhancement by a factor of 2.5, whereas all other metal ions quenched emission. The authors suggested that the CHEF mechanism was the reason for the fluorescence enhancement observed. The fluorescence signal was saturated after the addition of one equivalent of Zn 2+ ions and the calculated Ka was approximately 6 × 10 4 M −1 . The coordination environment of compound 4 and the Zn 2+ complex was investigated by IR spectroscopy and TD-DFT. The natural bond orbital (NBO) calculations suggested that the zinc metal ion preferred to coordinate with the two nitrogen atoms and the oxygen atom (the negative charge density on N(2), N(3), and O(2), were calculated to be −0.403, −0.63, and −0.65, respectively), the final coordination site was occupied by an MeOH molecule. Based on these theoretical calculations and, the IR study, they suggested that the Zn 2+ ions had a tetrahedral geometry.
The naphthalene moiety has been used by Kim to prepare two very similar chemosensors, compounds 5 and 6 [137,138]. The hydrazide functional group was utilized, as it readily undergoes a Schiff base condensation reaction, to prepare a tridentate binding site (shown in red). The binding group is shown to be ideal for Zn 2+ ions and the compounds coordinated in a 2:1 fashion. In both examples, the fluorescent mechanism was due to inhibition of C=N isomerization and ESIPT, upon the addition of Zn 2+ ions, which drastically increased the emission intensity, via a CHEF mechanism. The calculated β values were 1 × 10 9 M −2 in a DMSO-buffer and DMF-buffer for compounds 5 and 6, respectively.
The naphthalene moiety has been used by Kim to prepare two very similar chemosensors, compounds 5 and 6 [137,138]. The hydrazide functional group was utilized, as it readily undergoes a Schiff base condensation reaction, to prepare a tridentate binding site (shown in red). The binding group is shown to be ideal for Zn 2+ ions and the compounds coordinated in a 2:1 fashion. In both examples, the fluorescent mechanism was due to inhibition of C=N isomerization and ESIPT, upon the addition of Zn 2+ ions, which drastically increased the emission intensity, via a CHEF mechanism. The calculated β values were 1 × 10 9 M −2 in a DMSO-buffer and DMF-buffer for compounds 5 and 6, respectively. The chemosensor alone had hydrogen bonding donor and acceptor groups. Kim et al. also investigated the anion binding ability of compound 6 in DMF [138]. Upon the addition of AcO − and F − , to ligand 6, a 15 nm red shift was observed. The AcO − anion showed the greatest hyperchromic shift. This was attributed to a stable chelating binding mode between compound 6 and the AcO − anion. The authors proposed a binding stoichiometry of 2:1, whereby, the AcO − molecules were bound on the same side of the ligand, although no evidence was provided to support their claim.
A similar binding motif was used as a tridentate binding environment by Erdemir et al. who prepared a triphenylamine derivative 7 [139]. The absorbance spectrum showed subtle wavelength changes which the authors claimed to represent a 1:1 binding stoichiometry with a binding constant of 6 × 10 6 in CH3CN. It is reasonable to assume that a 1:1 complex would exist in solution. If this were true, the Benesi-Hildebrand plot would be a straight line, yet the data reported shows a curvature in the double reciprocal plot, indicative of stoichiometries other than a 1:1. The emission spectrum showed a distinctive enhancement upon the addition of Cd 2+ and Zn 2+ , when a 1 μM CH3CN solution was excited at 400 nm.
Interestingly, the λem for Cd 2+ and Zn 2+ was seen at 595 nm and 620 nm, respectively, a large enough difference to discriminate between these two metals. The optical responses between Cd 2+ and Zn 2+ ions are usually difficult to differentiate, due to the similarities of these ions. The authors argue that the difference in the metals radius (88 and 109 pm for Zn 2+ and Cd 2+ , respectively) was enough to influence the binding energy in the maximum wavelength seen. The fluorescence mechanism of compound 7 was due to the C=N isomerization which was inhibited upon the coordination of metal ions, thus, the intensity enhancement was a consequence of CHEF. The fluorescence binding isotherms showed that the onset of curvature occurred at two equivalents of metal ion, suggesting that the binding constant was not 1:1.
Lippard has made a career by designing and synthesizing molecules to bind Zn 2+ ions selectively. Over the years, his group has incorporated the dipicoylamine functional group on to different organic scaffolds [140][141][142][143][144], including the rhodamine dyes 8a-c, Figure 3 [145]. The rhodamine dye itself has been used extensively as an off-on sensor to detect a number of metal ions [146]. These molecules are colorless in organic solvents (DMSO, acetone, MeOH, and CH3CN) but under simulated physiological conditions (50 mM PIPES, 100 mM KCl pH 7.0), the spirolactam opens upon the addition of ZnCl2. The absorption band was observed at 569 nm, and the emission wavelength was The chemosensor alone had hydrogen bonding donor and acceptor groups. Kim et al. also investigated the anion binding ability of compound 6 in DMF [138]. Upon the addition of AcO − and F − , to ligand 6, a 15 nm red shift was observed. The AcO − anion showed the greatest hyperchromic shift. This was attributed to a stable chelating binding mode between compound 6 and the AcO − anion. The authors proposed a binding stoichiometry of 2:1, whereby, the AcO − molecules were bound on the same side of the ligand, although no evidence was provided to support their claim.
A similar binding motif was used as a tridentate binding environment by Erdemir et al. who prepared a triphenylamine derivative 7 [139]. The absorbance spectrum showed subtle wavelength changes which the authors claimed to represent a 1:1 binding stoichiometry with a binding constant of 6 × 10 6 in CH 3 CN. It is reasonable to assume that a 1:1 complex would exist in solution. If this were true, the Benesi-Hildebrand plot would be a straight line, yet the data reported shows a curvature in the double reciprocal plot, indicative of stoichiometries other than a 1:1. The emission spectrum showed a distinctive enhancement upon the addition of Cd 2+ and Zn 2+ , when a 1 µM CH 3 CN solution was excited at 400 nm.
Interestingly, the λ em for Cd 2+ and Zn 2+ was seen at 595 nm and 620 nm, respectively, a large enough difference to discriminate between these two metals. The optical responses between Cd 2+ and Zn 2+ ions are usually difficult to differentiate, due to the similarities of these ions. The authors argue that the difference in the metals radius (88 and 109 pm for Zn 2+ and Cd 2+ , respectively) was enough to influence the binding energy in the maximum wavelength seen. The fluorescence mechanism of compound 7 was due to the C=N isomerization which was inhibited upon the coordination of metal ions, thus, the intensity enhancement was a consequence of CHEF. The fluorescence binding isotherms showed that the onset of curvature occurred at two equivalents of metal ion, suggesting that the binding constant was not 1:1. The chemosensor alone had hydrogen bonding donor and acceptor groups. Kim et al. also investigated the anion binding ability of compound 6 in DMF [138]. Upon the addition of AcO − and F − , to ligand 6, a 15 nm red shift was observed. The AcO − anion showed the greatest hyperchromic shift. This was attributed to a stable chelating binding mode between compound 6 and the AcO − anion. The authors proposed a binding stoichiometry of 2:1, whereby, the AcO − molecules were bound on the same side of the ligand, although no evidence was provided to support their claim.
A similar binding motif was used as a tridentate binding environment by Erdemir et al. who prepared a triphenylamine derivative 7 [139]. The absorbance spectrum showed subtle wavelength changes which the authors claimed to represent a 1:1 binding stoichiometry with a binding constant of 6 × 10 6 in CH3CN. It is reasonable to assume that a 1:1 complex would exist in solution. If this were true, the Benesi-Hildebrand plot would be a straight line, yet the data reported shows a curvature in the double reciprocal plot, indicative of stoichiometries other than a 1:1. The emission spectrum showed a distinctive enhancement upon the addition of Cd 2+ and Zn 2+ , when a 1 μM CH3CN solution was excited at 400 nm.
Interestingly, the λem for Cd 2+ and Zn 2+ was seen at 595 nm and 620 nm, respectively, a large enough difference to discriminate between these two metals. The optical responses between Cd 2+ and Zn 2+ ions are usually difficult to differentiate, due to the similarities of these ions. The authors argue that the difference in the metals radius (88 and 109 pm for Zn 2+ and Cd 2+ , respectively) was enough to influence the binding energy in the maximum wavelength seen. The fluorescence mechanism of compound 7 was due to the C=N isomerization which was inhibited upon the coordination of metal ions, thus, the intensity enhancement was a consequence of CHEF. The fluorescence binding isotherms showed that the onset of curvature occurred at two equivalents of metal ion, suggesting that the binding constant was not 1:1.
Lippard has made a career by designing and synthesizing molecules to bind Zn 2+ ions selectively. Over the years, his group has incorporated the dipicoylamine functional group on to different organic scaffolds [140][141][142][143][144], including the rhodamine dyes 8a-c, Figure 3 [145]. The rhodamine dye itself has been used extensively as an off-on sensor to detect a number of metal ions [146]. These molecules are colorless in organic solvents (DMSO, acetone, MeOH, and CH3CN) but under simulated physiological conditions (50 mM PIPES, 100 mM KCl pH 7.0), the spirolactam opens upon the addition of ZnCl2. The absorption band was observed at 569 nm, and the emission wavelength was Lippard has made a career by designing and synthesizing molecules to bind Zn 2+ ions selectively. Over the years, his group has incorporated the dipicoylamine functional group on to different organic scaffolds [140][141][142][143][144], including the rhodamine dyes 8a-c, Figure 3 [145]. The rhodamine dye itself has been used extensively as an off-on sensor to detect a number of metal ions [146]. These molecules are colorless in organic solvents (DMSO, acetone, MeOH, and CH 3 CN) but under simulated physiological conditions (50 mM PIPES, 100 mM KCl pH 7.0), the spirolactam opens upon the addition of ZnCl 2 . The absorption band was observed at 569 nm, and the emission wavelength was seen at 569 nm. Once the concentration of Zn 2+ ions was greater than 1 mM, no further optical response occurred. The φ yield was calculated to be 0.22 for the zinc complex, compared to 0.001 for the molecular probe 8a. The Zn 2+ ions could be removed from the chelating motif by the addition of N,N,N ,N -tetrakis(2-pyridinylmethyl)-1,2 ethanediamine (TPEN) or ethylenediaminetetra acetic acid (EDTA). Compounds 8b and 8c exhibited little zinc binding via the lactone ring. It was highly probable that Zn 2+ ions did coordinate with the DPA group, but the Zn 2+ ions were too far away to participate in the coordination of the Lewis basic sites, in the lactone ring. Therefore, no emission signal was observed. This suggested that the lactone group did partake in the binding of the Zn 2+ ion with 8a, but no evidence was offered to substantiate this. The other biological relevant ions (Na + , K + , Ca 2+, and Mg 2+ ) did not generate the equivalent optical response. The authors calculated the K t value to be 73 µM, which took into account the zinc dissociation constant and the lactam formation. Despite this calculated value being at the higher concentration range, in comparison to other dipicolylamine molecular probes, it was still in the biological sensing range for neurochemical applications [147]. The ability for probe 8a to track Zn 2+ ions in Hela cells was also investigated and successfully showed cell permeability.
Chemosensors 2019, 7, x 9 of 50 seen at 569 nm. Once the concentration of Zn 2+ ions was greater than 1 mM, no further optical response occurred. The φ yield was calculated to be 0.22 for the zinc complex, compared to 0.001 for the molecular probe 8a. The Zn 2+ ions could be removed from the chelating motif by the addition of N, N N' N'-tetrakis(2-pyridinylmethyl)-1,2 ethanediamine (TPEN) or ethylenediaminetetra acetic acid (EDTA). Compounds 8b and 8c exhibited little zinc binding via the lactone ring. It was highly probable that Zn 2+ ions did coordinate with the DPA group, but the Zn 2+ ions were too far away to participate in the coordination of the Lewis basic sites, in the lactone ring. Therefore, no emission signal was observed. This suggested that the lactone group did partake in the binding of the Zn 2+ ion with 8a, but no evidence was offered to substantiate this. The other biological relevant ions (Na + , K + , Ca 2+, and Mg 2+ ) did not generate the equivalent optical response. The authors calculated the Kt value to be 73 μM, which took into account the zinc dissociation constant and the lactam formation. Despite this calculated value being at the higher concentration range, in comparison to other dipicolylamine molecular probes, it was still in the biological sensing range for neurochemical applications [147]. The ability for probe 8a to track Zn 2+ ions in Hela cells was also investigated and successfully showed cell permeability. Dipodal receptors have been used extensively in molecular recognition [148]. By incorporating pendent side arms with the appropriate functional groups to bind a guest species, the thermodynamic properties of the host-guest complex can be significantly enhanced, as it wraps itself around the guest. Kim et al. prepared dipodal chemosensor 9 from a piperazine group with two quinoline derivatives attached [149]. Many dipodal receptors are symmetrical but the group chose to synthesize an asymmetric system. On the addition of metal salts (Na + , K + , Mg 2+ , Ca 2+ , Al 3+ , Ga 3+ , Cr 3+ , Mn 2+ , Fe 2+ , Fe 3+ , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , and Pb 2+ ), significant optical changes were seen. The Zn 2+ ions produced a significant intense broad band centered around 500 nm, Figure 4B. The authors claimed that the fluorescence signal was turned off, due to the PET quenching mechanism, which was subsequently inhibited by the coordination of the Zn 2+ ions.
On closer inspection, the broad band was more reminiscent of an excimer band ( Figure 4B). It seemed that, in solution, the two naphthalene units in the dipodal system came closer together to produce this band. It just so happened that the Zn 2+ ion was the correct size to bring the units together and maximize the excimer signal. Molecular modeling calculations, however, showed that the two fluorophores were significantly further apart, suggesting that this band was not an intramolecular excimer formation. An excimer band might be formed between molecules at concentrations greater than 1 μM [38] and these studies were at 5 μM. Intermolecular excimer formation could have been ruled out by dilution experiments and intramolecular excimer formation could have been ruled out by a model compound with a single fluorophore. Other metals also showed broad bands but of Dipodal receptors have been used extensively in molecular recognition [148]. By incorporating pendent side arms with the appropriate functional groups to bind a guest species, the thermodynamic properties of the host-guest complex can be significantly enhanced, as it wraps itself around the guest. Kim et al. prepared dipodal chemosensor 9 from a piperazine group with two quinoline derivatives attached [149]. Many dipodal receptors are symmetrical but the group chose to synthesize an asymmetric system. On the addition of metal salts (Na + , K + , Mg 2+ , Ca 2+ , Al 3+ , Ga 3+ , Cr 3+ , Mn 2+ , Fe 2+ , Fe 3+ , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , and Pb 2+ ), significant optical changes were seen. The Zn 2+ ions produced a significant intense broad band centered around 500 nm, Figure 4B. The authors claimed that the fluorescence signal was turned off, due to the PET quenching mechanism, which was subsequently inhibited by the coordination of the Zn 2+ ions.
chemistry. The studies were in 10 mM bis-tris buffer at pH 7, but at this pH there was an equilibrium between H2S and HS − , and it was not until the pH reached 13 that the sulfide could exist in solution.
There is no doubt that the authors showed that upon the addition of Na2S to their receptor, drastic spectral changes were observed ( Figure 4C), but the Ksp for ZnS was approximately 2 × 10 −25 M -1 [150]. Thus, the onset of ZnS precipitation drove the equilibrium towards the formation of the salt under their conditions. Photochromic sensors have been used in many applications. The development of ratiometric fluorescent sensors utilizing a UV-Vis switch to sense metal ions is a unique way to prepare logic gates, in which a unimolecular fluorescent probe was designed to bind a metal that could be removed by a competing ligand, such as EDTA. Photoisomerization by UV-Vis could act as an input and the fluorescence emission could act as the output. Pu et al. synthesized two similar molecular switches compounds 10 and 11, which were utilized as a logic circuit based on the truth table that could be constructed from four different inputs and one output (Table 1 and Figure 5) [151,152]. On closer inspection, the broad band was more reminiscent of an excimer band ( Figure 4B). It seemed that, in solution, the two naphthalene units in the dipodal system came closer together to produce this band. It just so happened that the Zn 2+ ion was the correct size to bring the units together and maximize the excimer signal. Molecular modeling calculations, however, showed that the two fluorophores were significantly further apart, suggesting that this band was not an intramolecular excimer formation. An excimer band might be formed between molecules at concentrations greater than 1 µM [38] and these studies were at 5 µM. Intermolecular excimer formation could have been ruled out by dilution experiments and intramolecular excimer formation could have been ruled out by a model compound with a single fluorophore. Other metals also showed broad bands but of significantly lower intensity. The emission bands were difficult to assess as the band for Zn 2+ ions saturated the other metals' optical responses. Additionally, it was claimed that the Zn 2+ molecular probe was reversible by adding S 2− , yet the titration conditions did not seem to be conducive to sulfide chemistry. The studies were in 10 mM bis-tris buffer at pH 7, but at this pH there was an equilibrium between H 2 S and HS − , and it was not until the pH reached 13 that the sulfide could exist in solution.
There is no doubt that the authors showed that upon the addition of Na 2 S to their receptor, drastic spectral changes were observed ( Figure 4C), but the K sp for ZnS was approximately 2 × 10 −25 M −1 [150]. Thus, the onset of ZnS precipitation drove the equilibrium towards the formation of the salt under their conditions. Photochromic sensors have been used in many applications. The development of ratiometric fluorescent sensors utilizing a UV-Vis switch to sense metal ions is a unique way to prepare logic gates, in which a unimolecular fluorescent probe was designed to bind a metal that could be removed by a competing ligand, such as EDTA. Photoisomerization by UV-Vis could act as an input and the fluorescence emission could act as the output. Pu et al. synthesized two similar molecular switches compounds 10 and 11, which were utilized as a logic circuit based on the truth table that could be constructed from four different inputs and one output (Table 1 and Figure 5) [151,152]. fluorescent sensors utilizing a UV-Vis switch to sense metal ions is a unique way to prepare logic gates, in which a unimolecular fluorescent probe was designed to bind a metal that could be removed by a competing ligand, such as EDTA. Photoisomerization by UV-Vis could act as an input and the fluorescence emission could act as the output. Pu et al. synthesized two similar molecular switches compounds 10 and 11, which were utilized as a logic circuit based on the truth table that could be constructed from four different inputs and one output (Table 1 and Figure 5) [151,152].
The absorption band assigned to the open isomers is seen at 243 nm and a new band at 534 nm is seen upon irradiation in a 2.0 × 10 −5 M solution of CH 3 CN. The purple color is a consequence of the rigid conjugation of alternated double bonds, across the backbone of the organic molecule, when the molecule is in its closed state. This was shown to be reversible and was supported by the difference in quantum yields (φ) of 0.03 and 0.17 for the open and closed states, respectively. The spectroscopic signatures were different for the ring opening and closing, and upon interaction with Zn 2+ by EDTA, dual-controlled-switching behavior of compound 10 and 11 was shown, Scheme 2. The absorption band assigned to the open isomers is seen at 243 nm and a new band at 534 nm is seen upon irradiation in a 2.0 × 10 −5 M solution of CH3CN. The purple color is a consequence of the rigid conjugation of alternated double bonds, across the backbone of the organic molecule, when the molecule is in its closed state. This was shown to be reversible and was supported by the difference in quantum yields (φ) of 0.03 and 0.17 for the open and closed states, respectively. The spectroscopic signatures were different for the ring opening and closing, and upon interaction with Zn 2+ by EDTA, dual-controlled-switching behavior of compound 10 and 11 was shown, Scheme 2.
* when the emission band at 515 nm is seven times larger than the initial band at 421 nm, the output is "on", this is assigned a "1". Otherwise a "0" is assigned, i.e., the "off" state.
Chemosensors 2019, 7, x 12 of 50 * when the emission band at 515 nm is seven times larger than the initial band at 421 nm, the output is "on", this is assigned a "1". Otherwise a "0" is assigned, i.e., the "off" state. Zhu et al. systematically changed the linker groups and fluorophores to improve the fluorescent output [153]. Zhu employed the PET switch mechanism to quench the fluorescence signal in a number of chemical fluorophores, which include the di(2-dicolyl) amino group (DPA) [154][155][156]. This functional group is known to bind Zn 2+ ions and has been attached to anthracene [157], fluorescein [158], and rhodamine [159] groups, directly or via short linker groups, such as the methylene group. Zhu's team utilized the triazole functionality, to "click" the DPA binding site. Their early molecular probes relied on fluorophores that were excited around 400 nm, but this visible wavelength region was not conducive if the fluorophore was to be used in fluorescence microscopy for cell imaging, whereby, fluorophores that emit in the near IR region were more desirable. Therefore, they prepared molecular probes 12 to 14 that used the visible region to excite the fluorophore. All of their compounds were evaluated as zinc sensors and studied for their cell imaging capabilities, and all but one was shown to report zinc ion concentration at 50 μM in the Hela cells, by exciting the fluorophores at either 405 (compound 12 and compound 13) or 488 nm (compound 14). The anilino-DPA group Zhu et al. systematically changed the linker groups and fluorophores to improve the fluorescent output [153]. Zhu employed the PET switch mechanism to quench the fluorescence signal in a number of chemical fluorophores, which include the di(2-dicolyl) amino group (DPA) [154][155][156]. This functional group is known to bind Zn 2+ ions and has been attached to anthracene [157], fluorescein [158], and rhodamine [159] groups, directly or via short linker groups, such as the methylene group. Zhu's team utilized the triazole functionality, to "click" the DPA binding site. Their early molecular probes relied on fluorophores that were excited around 400 nm, but this visible wavelength region was not conducive if the fluorophore was to be used in fluorescence microscopy for cell imaging, whereby, fluorophores that emit in the near IR region were more desirable. Therefore, they prepared molecular probes 12 to 14 that used the visible region to excite the fluorophore. All of their compounds were evaluated as zinc sensors and studied for their cell imaging capabilities, and all but one was shown to report zinc ion concentration at 50 µM in the Hela cells, by exciting the fluorophores at either 405 (compound 12 and compound 13) or 488 nm (compound 14). The anilino-DPA group was found to be the key to shift the excitation wavelength for the fluorophores and selectively coordinate the Zn 2+ ions. was found to be the key to shift the excitation wavelength for the fluorophores and selectively coordinate the Zn 2+ ions.
Another dipodal system was developed by Yang et al. who prepared a mixed fluorophore receptor (coumarin-naphthalene) which contained a very elaborate chelating motif, compound 15, Scheme 3 [160]. The absorbance spectra showed a hypochromic shift at 335 nm and a hyperchromic shift at 430 nm, through a pseudo-isosbestic point at 307 nm. The binding isotherm was sigmoidal, which is indicative of cooperative binding. The fluorescence spectrum in EtOH (50 μM) showed a very strong emission band at 498 nm (λex = 430 nm). There was a preference to the Zn 2+ ions, in comparison to the other metal ions studied (Al 3+ , Ba 2+ , Ca 2+ , Cu 2+ , Hg 2+ Co 2+ , K + , Mg 2+ , Mn 2+ , Na + , Cr 3+ , Li + , Pb 2+ , Cd 2+ , Ag + and Fe 3+ ), however, competition experiments showed that both Cu 2+ and Fe 3+ significantly quenched the fluorescence intensity, when added to a [Zn (15)] 2+ solution. The binding constant was calculated to Ka = 2.0 × 10 4 M −1 , using the Benesi-Hildebrand method and the stoichiometry was confirmed by a Job's plot. The fluorescence signal was off in the initial state, as a consequence of the C=N isomerization and PET quenching. On addition of the Zn 2+ ions, the fluorescence was turned on via the CHEF mechanism. The limit of detection (LoD) was calculated to be 10 −7 M. Another dual fluorescence mechanism used in sensor technology is the PET and ICT quenching approach, which is inhibited by the addition of Zn 2+ ions. Molecular probe 16, Scheme 4, was synthesized by Wang et al., [161] based on a (1,4 diketo-3,6-diphenyl-pyrrolo [3,4-c]pyrrole derivative, which is known to be a brilliant red and strongly fluorescent dye. Additionally, this organic fluorophore is stable to heat with high photostability and shows a large Stokes shift, which makes it an excellent choice in sensor design. Compound 16 shows a weak emission band at 630 nm (φ = 0.07), in the absence of Zn 2+ ions, but upon the addition of Zn 2+ ions to compound 16, a blue shift to 560 nm (φ = 0.85) was observed; no other metal ion showed this enhancement. The optical studies (absorbance and fluorescence), and the NMR studies showed that no further optical or structural changes Another dipodal system was developed by Yang et al. who prepared a mixed fluorophore receptor (coumarin-naphthalene) which contained a very elaborate chelating motif, compound 15, Scheme 3 [160]. The absorbance spectra showed a hypochromic shift at 335 nm and a hyperchromic shift at 430 nm, through a pseudo-isosbestic point at 307 nm. The binding isotherm was sigmoidal, which is indicative of cooperative binding. The fluorescence spectrum in EtOH (50 µM) showed a very strong emission band at 498 nm (λ ex = 430 nm). There was a preference to the Zn 2+ ions, in comparison to the other metal ions studied (Al 3+ , Ba 2+ , Ca 2+ , Cu 2+ , Hg 2+ Co 2+ , K + , Mg 2+ , Mn 2+ , Na + , Cr 3+ , Li + , Pb 2+ , Cd 2+ , Ag + and Fe 3+ ), however, competition experiments showed that both Cu 2+ and Fe 3+ significantly quenched the fluorescence intensity, when added to a [Zn (15)] 2+ solution. The binding constant was calculated to K a = 2.0 × 10 4 M −1 , using the Benesi-Hildebrand method and the stoichiometry was confirmed by a Job's plot. The fluorescence signal was off in the initial state, as a consequence of the C=N isomerization and PET quenching. On addition of the Zn 2+ ions, the fluorescence was turned on via the CHEF mechanism. The limit of detection (LoD) was calculated to be 10 −7 M.
Chemosensors 2019, 7, x 13 of 50 was found to be the key to shift the excitation wavelength for the fluorophores and selectively coordinate the Zn 2+ ions.
Another dipodal system was developed by Yang et al. who prepared a mixed fluorophore receptor (coumarin-naphthalene) which contained a very elaborate chelating motif, compound 15, Scheme 3 [160]. The absorbance spectra showed a hypochromic shift at 335 nm and a hyperchromic shift at 430 nm, through a pseudo-isosbestic point at 307 nm. The binding isotherm was sigmoidal, which is indicative of cooperative binding. The fluorescence spectrum in EtOH (50 μM) showed a very strong emission band at 498 nm (λex = 430 nm). There was a preference to the Zn 2+ ions, in comparison to the other metal ions studied (Al 3+ , Ba 2+ , Ca 2+ , Cu 2+ , Hg 2+ Co 2+ , K + , Mg 2+ , Mn 2+ , Na + , Cr 3+ , Li + , Pb 2+ , Cd 2+ , Ag + and Fe 3+ ), however, competition experiments showed that both Cu 2+ and Fe 3+ significantly quenched the fluorescence intensity, when added to a [Zn (15)] 2+ solution. The binding constant was calculated to Ka = 2.0 × 10 4 M −1 , using the Benesi-Hildebrand method and the stoichiometry was confirmed by a Job's plot. The fluorescence signal was off in the initial state, as a consequence of the C=N isomerization and PET quenching. On addition of the Zn 2+ ions, the fluorescence was turned on via the CHEF mechanism. The limit of detection (LoD) was calculated to be 10 −7 M. Another dual fluorescence mechanism used in sensor technology is the PET and ICT quenching approach, which is inhibited by the addition of Zn 2+ ions. Molecular probe 16, Scheme 4, was synthesized by Wang et al., [161] based on a (1,4 diketo-3,6-diphenyl-pyrrolo [3,4-c]pyrrole derivative, which is known to be a brilliant red and strongly fluorescent dye. Additionally, this organic fluorophore is stable to heat with high photostability and shows a large Stokes shift, which makes it an excellent choice in sensor design. Compound 16 shows a weak emission band at 630 nm (φ = 0.07), in the absence of Zn 2+ ions, but upon the addition of Zn 2+ ions to compound 16, a blue shift to 560 nm (φ = 0.85) was observed; no other metal ion showed this enhancement. The optical studies (absorbance and fluorescence), and the NMR studies showed that no further optical or structural changes Another dual fluorescence mechanism used in sensor technology is the PET and ICT quenching approach, which is inhibited by the addition of Zn 2+ ions. Molecular probe 16, Scheme 4, was synthesized by Wang et al. [161] based on a (1,4 diketo-3,6-diphenyl-pyrrolo [3,4-c]pyrrole derivative, which is known to be a brilliant red and strongly fluorescent dye. Additionally, this organic fluorophore is stable to heat with high photostability and shows a large Stokes shift, which makes it an excellent choice in sensor design. Compound 16 shows a weak emission band at 630 nm (φ = 0.07), in the absence of Zn 2+ ions, but upon the addition of Zn 2+ ions to compound 16, a blue shift to 560 nm (φ = 0.85) was observed; no other metal ion showed this enhancement. The optical studies (absorbance and fluorescence), and the NMR studies showed that no further optical or structural changes occurred after the addition of two equivalents of Zn 2+ ions. A binding constant of 4.69 × 10 7 M −2 was determined using the fluorescence data and the stoichiometry was confirmed as a 2:1 metal:probe complex formation, through MS studies. Throughout this section there have been examples of molecular probes for the detection of Zn 2+ ions, but very few papers also discuss the counter-ion effect in their systems. Zinc complexes (in particular ZnCl2) also play an important role, as a moderate strength Lewis acid, in Friedel-Crafts acylations. Our own group has prepared pyrene-based molecular probes 17 and 18 [162]. Compound 17 was shown to simultaneously bind the cation and anion, in a self-assembled process. Self-assembly occurred in the presence of ZnX2 salts (X = F − , Cl − , Br − , I − , BF4 − , CH3CO2 − , H2PO4 − , ClO4 − , CN − , NO3 − ) and ZnSO4. The optical response was due to a strong excimer band observed at 410 nm in the presence of ZnCl2 in acetonitrile. DFT calculations supported the proposed structure (Scheme 5). The intensity of the excimer signal was dependent on the proximity of the two pyrene units. In absence of the Zn 2+ ion, the excimer signal was not observed, as was confirmed by the analogous spectroscopic studies with tetrabutylammonium salts, i.e., the metal was required for the selfassembly process to occur. This suggests that the Zn 2+ halide was templating the pyrene units. The geometry around the Zn 2+ ion was tetrahedral and the pyrene moieties adopted a syn configuration. The fluorescence titrations were in good agreement with the NMR studies, which showed a 2:1 binding mode calculated by non-linear regression with a calculated binding constant (K21) of 1.8 × 10 6 M −2 for ZnCl2. This example was a good example of size-selective self-assembled recognition and was carried out at a low concentration, to rule out intermolecular excimer formation. ions, but very few papers also discuss the counter-ion effect in their systems. Zinc complexes (in particular ZnCl 2 ) also play an important role, as a moderate strength Lewis acid, in Friedel-Crafts acylations. Our own group has prepared pyrene-based molecular probes 17 and 18 [162]. Compound 17 was shown to simultaneously bind the cation and anion, in a self-assembled process. Self-assembly occurred in the presence of ZnX 2 salts ( and ZnSO 4 . The optical response was due to a strong excimer band observed at 410 nm in the presence of ZnCl 2 in acetonitrile. DFT calculations supported the proposed structure (Scheme 5). The intensity of the excimer signal was dependent on the proximity of the two pyrene units. In absence of the Zn 2+ ion, the excimer signal was not observed, as was confirmed by the analogous spectroscopic studies with tetrabutylammonium salts, i.e., the metal was required for the self-assembly process to occur. This suggests that the Zn 2+ halide was templating the pyrene units. The geometry around the Zn 2+ ion was tetrahedral and the pyrene moieties adopted a syn configuration. The fluorescence titrations were in good agreement with the NMR studies, which showed a 2:1 binding mode calculated by non-linear regression with a calculated binding constant (K 21 ) of 1.8 × 10 6 M −2 for ZnCl 2 . This example was a good example of size-selective self-assembled recognition and was carried out at a low concentration, to rule out intermolecular excimer formation.

Scheme 5.
A simultaneous ion pair recognition event. Adapted from [160] with permission from The Royal Society of Chemistry.
The next step our group took was to tether two of the pyrene units, shown in compound 17, onto an organic backbone, to prepare a dipodal receptor. Three molecular probes containing an amide functional group and a triazole moiety was attached to a benzene scaffold (18a-c ortho, meta and para respectively) [163].
Fluorescence analysis of compounds 18a and 18b with different metal ions (Zn 2+ , Mg 2+ , Cd 2+ , Al 3+ , Hg 2+ , Ni 2+ , Cu 2+ , Fe 2+ , Fe 3+ and Na + ) as either their chloride or perchlorate salts were initially tested, with the addition of 10 equivalents of metal salts in CH3CN and excited at 325 nm. The fluorescence signal of the free receptors showed the typical monomer bands at 380 and 400 nm, characteristic of the unique electronic transitions in the pyrene-derived compounds, which were assigned to the π-π* electron transitions. A very broad band at 495 nm was also seen, which was assigned to the intramolecular excimer formation, upon excitation at 325 nm. On inspection of the fluorescence spectrum, quenching was observed for all metals except Zn 2+ (and to a lesser degree Cd 2+ and Mg 2+ ), and a new band was seen to increase at 400 nm. It was evident that the Zn 2+ salts showed significantly greater fluorescence at 400 nm for 18b than for 18a, for which a small increase in intensity was observed at 400 nm. The fluorescence intensity at 400 nm for 18a was a consequence of the two pyrene units in 18a adopting an anti-orientation, which gave a greater pyrene-pyrene separation than was available to 18b. It was shown that the binding affinity for Zn(ClO4)2 was a magnitude greater for 18a than for 18b, even though the fluorescence change at 400 nm was not observed to be as dramatic. In Scheme 5. A simultaneous ion pair recognition event. Adapted from [160] with permission from The Royal Society of Chemistry. The next step our group took was to tether two of the pyrene units, shown in compound 17, onto an organic backbone, to prepare a dipodal receptor. Three molecular probes containing an amide functional group and a triazole moiety was attached to a benzene scaffold (18a-c ortho, meta and para respectively) [163]. The next step our group took was to tether two of the pyrene units, shown in compound 17, onto an organic backbone, to prepare a dipodal receptor. Three molecular probes containing an amide functional group and a triazole moiety was attached to a benzene scaffold (18a-c ortho, meta and para respectively) [163].
Fluorescence analysis of compounds 18a and 18b with different metal ions (Zn 2+ , Mg 2+ , Cd 2+ , Al 3+ , Hg 2+ , Ni 2+ , Cu 2+ , Fe 2+ , Fe 3+ and Na + ) as either their chloride or perchlorate salts were initially tested, with the addition of 10 equivalents of metal salts in CH3CN and excited at 325 nm. The fluorescence signal of the free receptors showed the typical monomer bands at 380 and 400 nm, characteristic of the unique electronic transitions in the pyrene-derived compounds, which were assigned to the π-π* electron transitions. A very broad band at 495 nm was also seen, which was assigned to the intramolecular excimer formation, upon excitation at 325 nm. On inspection of the fluorescence spectrum, quenching was observed for all metals except Zn 2+ (and to a lesser degree Cd 2+ and Mg 2+ ), and a new band was seen to increase at 400 nm. It was evident that the Zn 2+ salts showed significantly greater fluorescence at 400 nm for 18b than for 18a, for which a small increase in intensity was observed at 400 nm. The fluorescence intensity at 400 nm for 18a was a consequence of the two pyrene units in 18a adopting an anti-orientation, which gave a greater pyrene-pyrene separation than was available to 18b. It was shown that the binding affinity for Zn(ClO4)2 was a magnitude greater for 18a than for 18b, even though the fluorescence change at 400 nm was not observed to be as dramatic. In Fluorescence analysis of compounds 18a and 18b with different metal ions (Zn 2+ , Mg 2+ , Cd 2+ , Al 3+ , Hg 2+ , Ni 2+ , Cu 2+ , Fe 2+ , Fe 3+ and Na + ) as either their chloride or perchlorate salts were initially tested, with the addition of 10 equivalents of metal salts in CH 3 CN and excited at 325 nm. The fluorescence signal of the free receptors showed the typical monomer bands at 380 and 400 nm, characteristic of the unique electronic transitions in the pyrene-derived compounds, which were assigned to the π-π* electron transitions. A very broad band at 495 nm was also seen, which was assigned to the intramolecular excimer formation, upon excitation at 325 nm. On inspection of the fluorescence spectrum, quenching was observed for all metals except Zn 2+ (and to a lesser degree Cd 2+ and Mg 2+ ), and a new band was seen to increase at 400 nm. It was evident that the Zn 2+ salts showed significantly greater fluorescence at 400 nm for 18b than for 18a, for which a small increase in intensity was observed at 400 nm. The fluorescence intensity at 400 nm for 18a was a consequence of the two pyrene units in 18a adopting an anti-orientation, which gave a greater pyrene-pyrene separation than was available to 18b. It was shown that the binding affinity for Zn(ClO 4 ) 2 was a magnitude greater for 18a than for 18b, even though the fluorescence change at 400 nm was not observed to be as dramatic. In fact this trend was seen for all the other analytes studied, we argued that the more favorable chelating motif was the reason for the observed calculated binding constant. To understand the binding mode, extensive IR and molecular modeling studies were carried out. It is known that significant infrared spectral changes occur when perchlorate coordinates with a metal ion through different modes [164]. Determination of the changing symmetry of the ClO 4 − ion, upon binding, from T d (ionic) to C 3v (unidentate) and C 2v (bidentate or bridging), could help identify if the binding mode proposed model was consistent with IR data reported. Molecular mechanics calculations for 18a and 18b and their corresponding Zn(ClO 4 ) 2 complexes were carried out ( Figure 6). Both of the free receptors showed very different π stacking interactions. The ortho-isomer, 18a, showed that the aromatic scaffold of the molecular probes participates in π-stacking and one of the pyrene arms seemed to intercalate and form an intramolecular sandwich interaction Figure 6A. This intercalation was not observed in the meta-isomer, 18b, which showed the π-π distance between the two pyrene units to be 4.77 Å. The distance was measured between the pyrene centroids ( Figure 6C) and the CH hydrogens on the triazole rings orientated towards the center of the molecular probes. This is in excellent agreement with the interactions expected to generate the excimer band seen in the fluorescence spectrum. Both of the Zn 2+ complexes showed drastically different optimized geometries. The Zn 2+ ion was bound in a tetrahedral arrangement, coordinated to the oxygen atom of the carbonyl functional group, and a nitrogen atom in the triazole ring, forming a seven-membered chelating ring. The distance between the two pyrene units in [Zn(18a)] 2+ was 10.8 Å, whereas the distance between the two pyrene units in [Zn(18a)] 2+ was 5.5 Å. This was in excellent agreement with the fluorescence data in which a broad excimer band was seen at 400 nm for the 18b complex; the same band was significantly less intense for the 16a complex.
Chemosensors 2019, 7, x 16 of 50 fact this trend was seen for all the other analytes studied, we argued that the more favorable chelating motif was the reason for the observed calculated binding constant. To understand the binding mode, extensive IR and molecular modeling studies were carried out. It is known that significant infrared spectral changes occur when perchlorate coordinates with a metal ion through different modes [164]. Determination of the changing symmetry of the ClO4 -ion, upon binding, from Td (ionic) to C3v (unidentate) and C2v (bidentate or bridging), could help identify if the binding mode proposed model was consistent with IR data reported. Molecular mechanics calculations for 18a and 18b and their corresponding Zn(ClO4)2 complexes were carried out ( Figure 6). Both of the free receptors showed very different π stacking interactions. The ortho-isomer, 18a, showed that the aromatic scaffold of the molecular probes participates in π-stacking and one of the pyrene arms seemed to intercalate and form an intramolecular sandwich interaction Figure 6A. This intercalation was not observed in the meta-isomer, 18b, which showed the π-π distance between the two pyrene units to be 4.77 Å. The distance was measured between the pyrene centroids ( Figure 6C) and the CH hydrogens on the triazole rings orientated towards the center of the molecular probes. This is in excellent agreement with the interactions expected to generate the excimer band seen in the fluorescence spectrum. Both of the Zn 2+ complexes showed drastically different optimized geometries. The Zn 2+ ion was bound in a tetrahedral arrangement, coordinated to the oxygen atom of the carbonyl functional group, and a nitrogen atom in the triazole ring, forming a seven-membered chelating ring. The distance between the two pyrene units in [Zn(18a)] 2+ was 10.8 Å, whereas the distance between the two pyrene units in [Zn(18a)] 2+ was 5.5 Å. This was in excellent agreement with the fluorescence data in which a broad excimer band was seen at 400 nm for the 18b complex; the same band was significantly less intense for the 16a complex.

Cadmium Probes
As cadmium has essentially no biological use and the bioaccumulation of Cd 2+ ions in the environment is a real threat, there is an increasing need to remove cadmium from wastewater from an industrial process. Cadmium has been ranked as one of the top 10 priority lists from the top 275 hazardous materials that have been identified by The Comprehensive Environmental Response, Compensation, and Liability Act in the USA [165], therefore, detection and remediation of Cd 2+ ions in the environment are imperative. A review by El-Safty outlined the importance of cadmium in the environment, discussed common techniques to remediate Cd 2+ ions, and highlighted the chemosensors used for cadmium monitoring [166]. The chemistry and structural similarities of cadmium is similar to zinc. Both the tetrahedral and octahedral forms are known. The type of Lewis basic atom plays a role in the geometry. For example, Parkin et al. showed that Zn 2+ ions preferred a tetrahedral environment when the donor atom was sulfur, whereas the Cd 2+ complex showed a six-coordinate environment [167]. Selenolato complexes are known to coordinate to Zn 2+ and Cd 2+ in a distorted tetrahedral environment [168]. A review published by Sigel and Martin state that the Cd 2+ ion could be found in 4, 5, and 6 coordination environments, 20%, 8%, and 56% of the time, respectively [169].
Therefore, it is not surprising that the LMFP reported for the detection of Cd 2+ ions is similar to the probes for zinc ions. Additionally, the fluorescence mechanism is alike. For example, Pu et al. synthesized a closely related molecule to probes 10 and 11 and subtly changed their binding motif to prepare a sensor for Cd 2+ ions [170].
A rhodamine-based conjugated dyad probe (19) has been synthesized by Stalin and coworkers (Scheme 6) [171]. The authors claimed that the FRET mechanism between the rhodamine and pyridine conjugated dyad is responsible for the emission observed upon the coordination of Cd 2+ ions in the organic-aqueous system (CH 3 CN:HEPES buffer 2:8, v/v, pH 7.2, 10 mM). However, upon inspection of their molecular probe, it is difficult to justify the authors claim. For FRET to occur, several criteria must be met: (i) the fluorescence emission spectrum of the donor molecule must overlap the absorption or excitation spectrum of the acceptor chromophore, (ii) the two fluorophores (donor and acceptor) must be in close proximity to one another (typically 1 to 10 nm), (iii) the transition dipole orientations of the donor and acceptor must be approximately parallel to each other, and (iv) the fluorescence lifetime of the donor molecule must be of sufficient duration to allow the FRET to occur. The pyridine unit alone is not usually considered a fluorophore. It is more reasonable to assume that the mechanism is an off-on fluorescence mechanism, based on the spirolactam ring opening. However, the authors did show high selectivity and sensitivity towards Cd 2+ ions in the presence of other competing metal ions, which was indicated by the appearance of a new fluorescence band at 590 nm, due to Cd 2+ coordination. Spectroscopic studies were used to determine a LoD of 1 × 10 −8 M, which is well below the WHO acceptable limit of 1.85 mM for Cd 2+ ions in drinking water. Both the association constant and binding ratio were determined from the absorption spectra. The absorption titration was used to calculate an association constant of 4 × 10 4 M −1 and, from the Job's plot data, the binding ratio was found to be 1:1. In the presence of sulfide ions CdS was formed, which resulted in decomplexation and the disappearance of the emission band observed at 590 nm. The results described above suggest that the sensor was recycled during the detection of S 2− ions.

Cadmium Probes
As cadmium has essentially no biological use and the bioaccumulation of Cd 2+ ions in the environment is a real threat, there is an increasing need to remove cadmium from wastewater from an industrial process. Cadmium has been ranked as one of the top 10 priority lists from the top 275 hazardous materials that have been identified by The Comprehensive Environmental Response, Compensation, and Liability Act in the USA [165], therefore, detection and remediation of Cd 2+ ions in the environment are imperative. A review by El-Safty outlined the importance of cadmium in the environment, discussed common techniques to remediate Cd 2+ ions, and highlighted the chemosensors used for cadmium monitoring [166]. The chemistry and structural similarities of cadmium is similar to zinc. Both the tetrahedral and octahedral forms are known. The type of Lewis basic atom plays a role in the geometry. For example, Parkin et al. showed that Zn 2+ ions preferred a tetrahedral environment when the donor atom was sulfur, whereas the Cd 2+ complex showed a six-coordinate environment [167]. Selenolato complexes are known to coordinate to Zn 2+ and Cd 2+ in a distorted tetrahedral environment [168]. A review published by Sigel and Martin state that the Cd 2+ ion could be found in 4, 5, and 6 coordination environments, 20%, 8%, and 56% of the time, respectively [169].
Therefore, it is not surprising that the LMFP reported for the detection of Cd 2+ ions is similar to the probes for zinc ions. Additionally, the fluorescence mechanism is alike. For example, Pu et al. synthesized a closely related molecule to probes 10 and 11 and subtly changed their binding motif to prepare a sensor for Cd 2+ ions [170].
A rhodamine-based conjugated dyad probe (19) has been synthesized by Stalin and coworkers (Scheme 6) [171]. The authors claimed that the FRET mechanism between the rhodamine and pyridine conjugated dyad is responsible for the emission observed upon the coordination of Cd 2+ ions in the organic-aqueous system (CH3CN:HEPES buffer 2:8, v/v, pH 7.2, 10 mM). However, upon inspection of their molecular probe, it is difficult to justify the authors claim. For FRET to occur, several criteria must be met: (i) the fluorescence emission spectrum of the donor molecule must overlap the absorption or excitation spectrum of the acceptor chromophore, (ii) the two fluorophores (donor and acceptor) must be in close proximity to one another (typically 1 to 10 nm), (iii) the transition dipole orientations of the donor and acceptor must be approximately parallel to each other, and (iv) the fluorescence lifetime of the donor molecule must be of sufficient duration to allow the FRET to occur. The pyridine unit alone is not usually considered a fluorophore. It is more reasonable to assume that the mechanism is an off-on fluorescence mechanism, based on the spirolactam ring opening. However, the authors did show high selectivity and sensitivity towards Cd 2+ ions in the presence of other competing metal ions, which was indicated by the appearance of a new fluorescence band at 590 nm, due to Cd 2+ coordination. Spectroscopic studies were used to determine a LoD of 1 × 10 −8 M, which is well below the WHO acceptable limit of 1.85 mM for Cd 2+ ions in drinking water. Both the association constant and binding ratio were determined from the absorption spectra. The absorption titration was used to calculate an association constant of 4 × 10 4 M −1 and, from the Job's plot data, the binding ratio was found to be 1:1. In the presence of sulfide ions CdS was formed, which resulted in decomplexation and the disappearance of the emission band observed at 590 nm. The results described above suggest that the sensor was recycled during the detection of S 2− ions.  A FRET molecular probe incorporating rhodamine was synthesized by Goswami et al. Chemosensor 20 was shown to be selective for Cd 2+ ions, with the sensor acting as an 'off-on' FRET sensor capable of detecting Cd 2+ ions in a mixed organic-aqueous solvent system. Additionally, a distinctive color change was seen, typical of rhodamine dyes. A quinoline-benzothiazole functional group (FRET donor) was tethered to the rhodamine molecule (FRET acceptor), via a nonconjugated spacer, Scheme 7 [172]. As the emission spectra of the quinoline benzothiazole group overlapped significantly with the absorption spectra of the ring-opened rhodamine B dye, it was possible to observe the transfer of energy from donor to acceptor, once the sensor interacted with the Cd 2+ ions. Both the quantum yield and FRET efficiency of the sensor were calculated from fluorescence studies carried out in MeOH/H 2 O (1:4, v/v, pH 7.2, 10mM HEPES buffer solution) and were found to be φ = 0.37 and 48%, respectively. These experimental conditions were optimized for the water content. They reported that when the percentage of water was increased above 80%, the intensity ratio (I 585 /I 470 ) of the emission band decreased significantly, whereas less water content showed modest changes in the emission intensity. From the fluorescence titration, Aich and coworkers found that the lowest concentration of Cd 2+ ions detectable by the sensor was 3 × 10 −7 M and the association constant (determined using a nonlinear least-squares fit) was found to be 1 × 10 5 M −1 . The other metal ions studied (K + , Ca 2+ , Ni 2+ , Mn 2+ , Zn 2+ , Cu 2+ , Fe 3+ , Cr 3+ , Hg 2+ , and Pb 2+ ) did not show the same optical response. The 1:1 binding mode was confirmed both in solution and in the solid state. The X-ray crystal structure confirmed that the Cd 2+ binds in an octahedral environment with five of the donor groups from compound 20 and a chloride ion in the remaining coordination site. A FRET molecular probe incorporating rhodamine was synthesized by Goswami et al. Chemosensor 20 was shown to be selective for Cd 2+ ions, with the sensor acting as an 'off-on' FRET sensor capable of detecting Cd 2+ ions in a mixed organic-aqueous solvent system. Additionally, a distinctive color change was seen, typical of rhodamine dyes. A quinoline-benzothiazole functional group (FRET donor) was tethered to the rhodamine molecule (FRET acceptor), via a nonconjugated spacer, Scheme 7 [172]. As the emission spectra of the quinoline benzothiazole group overlapped significantly with the absorption spectra of the ring-opened rhodamine B dye, it was possible to observe the transfer of energy from donor to acceptor, once the sensor interacted with the Cd 2+ ions. Both the quantum yield and FRET efficiency of the sensor were calculated from fluorescence studies carried out in MeOH/H2O (1:4, v/v, pH 7.2, 10mM HEPES buffer solution) and were found to be φ = 0.37 and 48%, respectively. These experimental conditions were optimized for the water content. They reported that when the percentage of water was increased above 80%, the intensity ratio (I585/I470) of the emission band decreased significantly, whereas less water content showed modest changes in the emission intensity. From the fluorescence titration, Aich and coworkers found that the lowest concentration of Cd 2+ ions detectable by the sensor was 3 × 10 −7 M and the association constant (determined using a nonlinear least-squares fit) was found to be 1 × 10 5 M −1 . The other metal ions studied (K + , Ca 2+ , Ni 2+ , Mn 2+ , Zn 2+ , Cu 2+ , Fe 3+ , Cr 3+ , Hg 2+ , and Pb 2+ ) did not show the same optical response. The 1:1 binding mode was confirmed both in solution and in the solid state. The X-ray crystal structure confirmed that the Cd 2+ binds in an octahedral environment with five of the donor groups from compound 20 and a chloride ion in the remaining coordination site. Another popular fluorescent mechanism that is being used in molecular probe design is aggregation induced emission (AIE), a similar mechanism to excimer formation. Wang et al. employed this by tethering a hydrophobic tetraphenylethene (TPE) moiety (hydrophobic part) onto a β-cyclodextrin (hydrophilic part), via a click reaction to give compound 21. Therefore, by varying the ratios of water and organic solvent the mechanism could be turned on and off. The triazole moiety was used as the Lewis basic site to coordinate the Cd 2+ ions [171]. In solutions containing ≤50% water, Another popular fluorescent mechanism that is being used in molecular probe design is aggregation induced emission (AIE), a similar mechanism to excimer formation. Wang et al. employed this by tethering a hydrophobic tetraphenylethene (TPE) moiety (hydrophobic part) onto a β-cyclodextrin (hydrophilic part), via a click reaction to give compound 21. Therefore, by varying the ratios of water and organic solvent the mechanism could be turned on and off. The triazole moiety was used as the Lewis basic site to coordinate the Cd 2+ ions [171]. In solutions containing ≤50% water, the sensor gave no significant emission when the probe was excited at 330 nm. However, upon the addition of Cd 2+ ions, a noticeable emission band emerged around 476 nm. This could be attributed to the overlap of the tetraphenylethene groups as a result of the 2:1 coordination between the sensor and Cd 2+ ions. This same behavior was observed as the water content of the solvent system was increased up to 98%. In higher concentrations of water, the solution of the sensor becomes strongly emissive at 476 nm, and this emission was visible with the naked eye, Figure 8  Another popular fluorescent mechanism that is being used in molecular probe design is aggregation induced emission (AIE), a similar mechanism to excimer formation. Wang et al. employed this by tethering a hydrophobic tetraphenylethene (TPE) moiety (hydrophobic part) onto a β-cyclodextrin (hydrophilic part), via a click reaction to give compound 21. Therefore, by varying the ratios of water and organic solvent the mechanism could be turned on and off. The triazole moiety was used as the Lewis basic site to coordinate the Cd 2+ ions [171]. In solutions containing ≤50% water, the sensor gave no significant emission when the probe was excited at 330 nm. However, upon the addition of Cd 2+ ions, a noticeable emission band emerged around 476 nm. This could be attributed to the overlap of the tetraphenylethene groups as a result of the 2:1 coordination between the sensor and Cd 2+ ions. This same behavior was observed as the water content of the solvent system was increased up to 98%. In higher concentrations of water, the solution of the sensor becomes strongly emissive at 476 nm, and this emission was visible with the naked eye, Figure 8. This strong emission was due to the formation of aggregates in solution, owing to the decreased solubility of the sensor.   . Aggregation could be seen by the naked-eye once the water content was greater than 90%. Adapted from [173] with permission from The Royal Society of Chemistry.
Many LMFPs have been adsorbed (physisorption or chemisorption) onto surfaces to improve sensitivity. This is done to increase the number of binding sites, by incorporating a number of the desired molecular probes onto a single platform. Commonly used scaffolds are mesoporous silica materials. These materials are rigid and porous and have been used to support a number of molecular sensors. An advantage of these materials is that they are optically transparent in the visiable range, therefore, the signal generated is genuinely from the receptor and the analyte. These types of materials have been used in sensor applications and for remediation purposes [119,129,174,175]. A reusable fluorescent sensor capable of detecting Cd 2+ , Hg 2+ , and Pb 2+ ions in aqueous solutions was synthesized by Zhu et al. by physidorption, of a 2,2-dipicoylamine modified naphthalimide fluorophore, 22, onto the surface of a silica microsphere [176]. It is inherently hydrophilic and capable of metal detection and adsorption (Scheme 8), as indicated by increases in its fluorescence intensity. Control studies indicated that, in the absence of the silylpropylamine linkers on the microspheres, no adsorption of metal ions had occurred. Under optimal conditions (low concentration of metal ions and 22 at pH 4.5-10), compound 22 had an adsorptivity greater than 95%. In acidic media (pH < 4.5), the adsorption capability of the sensor was greatly diminished, which allowed the desorption of metal ions from the sensor. Presumably, the nitrogen atoms become protonated, removing the ability of the molecular probe to form a Lewis acid-base adduct, before the acidic medium removes the metal ions. This allows microspheres coated with 22 to be recycled multiple times, by washing them with an acidic solution, then readjusting the pH by adding base. Fluorescence studies determined that the LoD of 22 was in the nM concentration range with an association constant of 2.3 ×10 7 M −1 for Cd 2+ ions.
Interestingly, this molecular probe was able to remove Cd 2+ (and Hg 2+ and Pb 2+ ) from tap water, lab wastewater, and simulated biological fluid. Samples were spiked with 2.0 mg·L −1 Cd 2+ salts and 99%, 98%, and 96% of the metal was recovered from tap water, wastewater, and biological fluid, respectively, indicating that the molecular probe is a good candidate for cadmium remediation, but it remains to be seen if the probe can selectively remove metals from a mixture of metal ions in the samples. fluorophore, 22, onto the surface of a silica microsphere [176]. It is inherently hydrophilic and capable of metal detection and adsorption (Scheme 8), as indicated by increases in its fluorescence intensity. Control studies indicated that, in the absence of the silylpropylamine linkers on the microspheres, no adsorption of metal ions had occurred. Under optimal conditions (low concentration of metal ions and 22 at pH 4.5-10), compound 22 had an adsorptivity greater than 95%. In acidic media (pH < 4.5), the adsorption capability of the sensor was greatly diminished, which allowed the desorption of metal ions from the sensor. Presumably, the nitrogen atoms become protonated, removing the ability of the molecular probe to form a Lewis acid-base adduct, before the acidic medium removes the metal ions. This allows microspheres coated with 22 to be recycled multiple times, by washing them with an acidic solution, then readjusting the pH by adding base. Fluorescence studies determined that the LoD of 22 was in the nM concentration range with an association constant of 2.3 ×10 7 M −1 for Cd 2+ ions.
Interestingly, this molecular probe was able to remove Cd 2+ (and Hg 2+ and Pb 2+ ) from tap water, lab wastewater, and simulated biological fluid. Samples were spiked with 2.0 mgL −1 Cd 2+ salts and 99%, 98%, and 96% of the metal was recovered from tap water, wastewater, and biological fluid, respectively, indicating that the molecular probe is a good candidate for cadmium remediation, but it remains to be seen if the probe can selectively remove metals from a mixture of metal ions in the samples. Lv and Shen synthesized a ratiometric porphyrin-based fluorescent probe (23) for detecting Cd 2+ ions in aqueous systems from pH 6.5 to 10 [177]. The incorporation of sulfonic groups enhances the probe's water solubility of the porphyrin ring system. The 2,2'-dipyridylamine (DPA) possesses remarkable binding ability towards Cd 2+ ions; therefore two DPA groups were attached to coordinate the Cd 2+ ions. The porphyrin fluorophore serves as an ideal signaling unit owing to its large conjugated π plane. Upon the coordination of DPA to Cd 2+ ions, a ratiometric fluorescent change to Scheme 8. Chemosensor 22, attached to silica, is able to undergo a fluorescence turn on, upon the addition of Cd 2+ , Hg 2+ , and Pb 2+ . The removal of the metal was achieved by the addition of acid.
Lv and Shen synthesized a ratiometric porphyrin-based fluorescent probe (23) for detecting Cd 2+ ions in aqueous systems from pH 6.5 to 10 [177]. The incorporation of sulfonic groups enhances the probe's water solubility of the porphyrin ring system. The 2,2 -dipyridylamine (DPA) possesses remarkable binding ability towards Cd 2+ ions; therefore two DPA groups were attached to coordinate the Cd 2+ ions. The porphyrin fluorophore serves as an ideal signaling unit owing to its large conjugated π plane. Upon the coordination of DPA to Cd 2+ ions, a ratiometric fluorescent change to the emission spectra of the probe was observed. The fluorescence mechanism is based on an intramolecular charge transfer (ICT). Compound 23 shows a good solubility in aqueous solutions without the addition of an organic solvent. Spectroscopic studies were carried out in HEPES buffer solution (pH 7.4). Upon the addition of Cd 2+ ions to a solution of the probe, the emission wavelength shifted from 653 to 611 nm (φ = 0.15). The noticeable hypsochromic shift of 42 nm was indicative of a reduction in the electron donating ability of the DPA moieties, upon Cd 2+ ions coordination, as the Cd 2+ ion is electron-withdrawing. Moreover, as the Cd 2+ ions are coordinated to the DPA molecule through the nitrogen atom, the porphyrin ring system is perturbed subtly, therefore, decreasing the conjugation, and shifting the wavelength in the blue direction. Direct fluorimetric titration as a function of Cd 2+ ion concentration was used to calculate a dissociation constant (K d ) value of 31 µM, and a Job's plot showed a 1:2 binding ratio between the sensor and Cd 2+ ions. Often signaling systems that depend on fluorescence changes (increases or decreases) at a fixed wavelength, suffer from uncertainty in absolute fluorescence measurements; however, a ratiometric fluorescence response enhances the reliability of the signaling from a quantitative viewpoint. Through linear regression of the calibration plot ([F 611 /F 653 ] vs [Cd 2+ ]), the detection limit was calculated to be 0.032 µM. Furthermore, the Cd 2+ ions could be removed from the DPA groups by the addition of EDTA. As compound 23 showed good aqueous solubility studies in Hela cells and A549 (adenocarcinomic human alveolar basal epithelial) cells, the two cancer lines remained 87% viable, implying that the sensor exhibited minimal cytotoxicity and is likely safe for use in bioimaging.
conjugation, and shifting the wavelength in the blue direction. Direct fluorimetric titration as a function of Cd 2+ ion concentration was used to calculate a dissociation constant (Kd) value of 31 μM, and a Job's plot showed a 1:2 binding ratio between the sensor and Cd 2+ ions. Often signaling systems that depend on fluorescence changes (increases or decreases) at a fixed wavelength, suffer from uncertainty in absolute fluorescence measurements; however, a ratiometric fluorescence response enhances the reliability of the signaling from a quantitative viewpoint. Through linear regression of the calibration plot ([F611/F653] vs [Cd 2+ ]), the detection limit was calculated to be 0.032 μM. Furthermore, the Cd 2+ ions could be removed from the DPA groups by the addition of EDTA. As compound 23 showed good aqueous solubility studies in Hela cells and A549 (adenocarcinomic human alveolar basal epithelial) cells, the two cancer lines remained 87% viable, implying that the sensor exhibited minimal cytotoxicity and is likely safe for use in bioimaging.

Siva and Chellapa synthesized a ratiometric benzimidazole tagged coumarin probe 24.
Compound 24 could selectively detect Cd 2+ and F − ions in a CH3CN/H2O (1:9, v/v) HEPES-buffered solutions at pH 7.52, Scheme 9 [178]. The UV-Vis absorption spectrum showed three distinct bands at 292, 342, and 401 nm. The addition of Cd 2+ ions led to a hyperchromic shift at 292 and 343 nm, and a hypochromic shift at 401 nm, through an isosbestic point at 367 nm, indicative of metal complexes in solution. Upon exciting probe 24 at 340 nm, two distinct emission bands at 410 and 530 nm were observed. The addition of Cd 2+ ions led to a blue emission band at 410 nm and an increase in emission intensity, at 530 nm, through an isoemissive point 518 nm. From these spectral changes the association constant was calculated to be 3.5 × 10 3 M −1 , using a modified Benesi−Hildebrand method and the LoD was determined to be 1.5 × 10 −10 M. The addition of the Cd 2+ ions inhibited the intermolecular charge transfer (ICT) between the lone pair on the secondary amine in the imidazole ring system and the carbonyl group on the coumarin unit. It is reasonable that the molecular probe would chelate the metal ions in a tridentate fashion, via the three Lewis basic sites (Scheme 9, red bonds). Additionally, the authors carried out an extensive anion study which showed very little spectral changes for a number of tetrabutylammonium salts (TBAX, where X = F − , Cl − , Br − , I − , OAc − , Siva and Chellapa synthesized a ratiometric benzimidazole tagged coumarin probe 24. Compound 24 could selectively detect Cd 2+ and F − ions in a CH 3 CN/H 2 O (1:9, v/v) HEPES-buffered solutions at pH 7.52, Scheme 9 [178]. The UV-Vis absorption spectrum showed three distinct bands at 292, 342, and 401 nm. The addition of Cd 2+ ions led to a hyperchromic shift at 292 and 343 nm, and a hypochromic shift at 401 nm, through an isosbestic point at 367 nm, indicative of metal complexes in solution. Upon exciting probe 24 at 340 nm, two distinct emission bands at 410 and 530 nm were observed. The addition of Cd 2+ ions led to a blue emission band at 410 nm and an increase in emission intensity, at 530 nm, through an isoemissive point 518 nm. From these spectral changes the association constant was calculated to be 3.5 × 10 3 M −1 , using a modified Benesi−Hildebrand method and the LoD was determined to be 1.5 × 10 −10 M. The addition of the Cd 2+ ions inhibited the intermolecular charge transfer (ICT) between the lone pair on the secondary amine in the imidazole ring system and the carbonyl group on the coumarin unit. It is reasonable that the molecular probe would chelate the metal ions in a tridentate fashion, via the three Lewis basic sites (Scheme 9, red bonds). Additionally, the authors carried out an extensive anion study which showed very little spectral changes for a number of tetrabutylammonium salts (TBAX, where X = F − , Cl − , Br − , I − , OAc − , NO 3 − , H 3 PO 4 − , CN − ). Fluoride was the only anion to produce a small changes in the fluorescence spectrum, which was also shown to inhibit the ICT mechanism, by forming a hydrogen bonding interaction with the NH group in the benzimidazole group. Fluoride was the only anion to produce a small changes in the fluorescence spectrum, which was also shown to inhibit the ICT mechanism, by forming a hydrogen bonding interaction with the NH group in the benzimidazole group. Another LMFP coumarin dyad, 25, was prepared by Kumar and Ahmed [179]. The molecular probe served as, both, a colorimetric sensor and a fluorescent sensor, through chelation-enhanced fluorescence (CHEF), Figure 9. Upon the addition of Cd 2+ ions in a HEPES-buffered solution (20 mM, Scheme 9. Proposed binding of molecular probe 24 and Cd 2+ ions. Another LMFP coumarin dyad, 25, was prepared by Kumar and Ahmed [179]. The molecular probe served as, both, a colorimetric sensor and a fluorescent sensor, through chelation-enhanced fluorescence (CHEF), Figure 9. Upon the addition of Cd 2+ ions in a HEPES-buffered solution (20 mM, CH 3 CN:H 2 O, 3:7, v/v, pH 7.0) of probe 25, the solution changed from yellow to colorless, whereas there were no color changes observed upon the addition of the other metal ions studied (Na + , K + , Ba 2+ , Ca 2+ , Mg 2+ , Cr 3+ , Mn 2+ , Fe 3+ , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , Pb 2+ , Ag + , Al 3+ , Hg 2+ , Hg + , and Sn 2+ ). On the addition of Cd 2+ ions to a solution of 25, there was a blue shift in the absorbance band from 387 nm to 370 nm, whereas the addition of other metal ions resulted in no notable changes to the absorbance spectra. Titration studies were carried out to further study the interactions between 25 and Cd 2+ ions. The gradual changes in the absorbance bands suggests that a new species was produced during the titration. A Job's plot was used to verify the formation of a 1:1 complex between the sensor and cadmium ions and a Benesi-Hilderbrand plot was used to calculate an association constant of 1.8 × 10 6 M −1 . Furthermore, the LoD was calculated to be 6.3 × 10 −8 M. The fluorescence data of the sensor with different metal ions was analogous to the absorbance data, as there were only notable spectroscopic changes in the presence of Cd 2+ ions. Under the same solution conditions as the absorbance experiments, 25 was non-fluorescent. However, in the presence of Cd 2+ ions, a gradual increase in the emission band at 495 nm was observed and the absolute fluorescence quantum yield (φ) increased by a factor of nine, upon coordination with the Cd 2+ ions. A fluorescence lifetime decay was observed from 0.217 to 1.97 ns, after the addition of Cd 2+ ions, indicative of the strong coordination between the sensor and Cd 2+ ions. In an attempt to prepare a workable probe, test strips were prepared, which changed color upon the immersion into a Cd 2+ solution (5 × 10 −4 M), Figure 9.  Many molecular probes designed to bind cadmium can also bind zinc ions and it is often difficult to distinguish between the two. The binding ratio between the metal and the probe can be enough to discriminate between two metal ions. Interestingly, the authors used their probe to monitor the amount of Cd 2+ in lipstick, hair dye, and makeup cream as Cd 2+ is a common ion found in the preparation of these cosmetics. Spectrofluorometry and AAS analysis showed the amount of Cd 2+ ions in the samples to be very similar ( Table 2). Many molecular probes designed to bind cadmium can also bind zinc ions and it is often difficult to distinguish between the two. The binding ratio between the metal and the probe can be enough to discriminate between two metal ions. For example, LMWP 26 was synthesized by Rajak et al. Scheme 10 [180]. Fluorene-pyridine were coupled together via an imine bond. Compound 26 served as both a colorimetric sensor, producing noticeable color changes in the presence of both zinc and cadmium, and a turn-on fluorescence sensor. As Cd 2+ ions were added to a CH 3 CN:H 2 O solution (8:2, v/v, 10 mM HEPES, pH 7.4) of 26, the solution changed from pale yellow to deep orange, whereas a deep pink color was observed in the presence of Zn 2+ ions. The UV-Vis titration data showed the emergence of several new bands, upon the addition of Cd 2+ ions to a solution of the sensor, with a noticeable difference to that of the Zn 2+ ion titration. Fluorescence titrations further highlighted the probe's selectivity for Cd 2+ and Zn 2+ ions.
Decomplexation of the [CdCl (26)] complex and subsequent emission turn off could be accomplished with the addition of S 2− ions to a solution of the complex. The fluorescence turn-off also served as a means of distinguishing between Zn 2+ and Cd 2+ ions, as decomplexation of the Zn 2+ ion's complex did not occur in the presence of S 2− ions. A Job's plot verified the binding stoichiometry of the sensor for Cd 2+ ions as 1:1 with an association constant of 6.4 × 10 4 M −1 , calculated from a Benesi-Hildebrand plot using the data from UV-Vis titrations. The weak fluorescence band seen around 430 nm is due to the lone pair on the nitrogen atom on the imine group quenching the intensity. However, on chelation of Cd 2+ ions with the sensor, the PET process is prevented and the molecule becomes less flexible, as the free rotation of the azomethine carbon is also restricted. The coordination environment was further validated by studying the 1 H NMR titration of Cd 2+ ions into a solution of the sensor which showed a clear disappearance of the phenolic proton and a downfield shift of the azomethine proton. From the fluorescence titration, an LoD of 0.53 µM for Cd 2+ ions was calculated and a fluorescence quantum yield of 0.996 was observed, whereas the fluorescence quantum yield of 26 was calculated to be 0.042. As Zn 2+ and Cd 2+ produced different optical inputs with one output, a binary logic gate function could be produced Table 3). This selective behavior mimics an OR binary logic gate ( Figure 10).
Chemosensors 2019, 7, x 24 of 50 quantum yield of 26 was calculated to be 0.042. As Zn 2+ and Cd 2+ produced different optical inputs with one output, a binary logic gate function could be produced Table 3). This selective behavior mimics an OR binary logic gate ( Figure 10).     A simple cleft-shaped fluorescent chemosensor, 27, was prepared by Qian and Yang, Scheme 11 [181]. Originally prepared to bind group 1 and group 2 metal ions, NH4 + ions and heavy metal salts in methanol and at high concentrations (1.9 × 10 −2 M) by Maass and Vögtle [182][183][184], the molecular cleft did not discriminate between different metal ions. The fluorescence spectra for 10 μM of compound 27 (20 mM Tris-HCl, pH 7.4) with Cd 2+ ions showed a significant fluorescence band at 417 nm. The other metals studied did not show any response (Ba 2+ , Mn 2+ , Hg 2+ , Ni 2+ , Ca 2+ , Cu 2+ , Co 2+ , Pb 2+ , Mg 2+ , Zn 2+ , Cd 2+ , Fe 2+ , Fe 3+ , Cr 3+ , Ag + , Li + , Na + , and K + ). The PET mechanism quenched the initial fluorescence intensity, via the lone pairs on the nitrogen atom. On the addition of Cd 2+ ions, the flexible molecular cleft became rigid and the fluorescence increased, due to CHEF, and the π-π* transitions gave rise to the emission band observed. The association constant calculated by the nonlinear least square analysis was found to be 5.3 × 10 7 M −1 . Figure 10. Logic schemes prepared from the binary inputs/outputs in the Truth tables [180].  A simple cleft-shaped fluorescent chemosensor, 27, was prepared by Qian and Yang, Scheme 11 [181]. Originally prepared to bind group 1 and group 2 metal ions, NH 4 + ions and heavy metal salts in methanol and at high concentrations (1.9 × 10 −2 M) by Maass and Vögtle [182][183][184], the molecular cleft did not discriminate between different metal ions. The fluorescence spectra for 10 µM of compound 27 (20 mM Tris-HCl, pH 7.4) with Cd 2+ ions showed a significant fluorescence band at 417 nm. The other metals studied did not show any response (Ba 2+ , Mn 2+ , Hg 2+ , Ni 2+ , Ca 2+ , Cu 2+ , Co 2+ , Pb 2+ , Mg 2+ , Zn 2+ , Cd 2+ , Fe 2+ , Fe 3+ , Cr 3+ , Ag + , Li + , Na + , and K + ). The PET mechanism quenched the initial fluorescence intensity, via the lone pairs on the nitrogen atom. On the addition of Cd 2+ ions, the flexible molecular cleft became rigid and the fluorescence increased, due to CHEF, and the π-π* transitions gave rise to the emission band observed. The association constant calculated by the nonlinear least square analysis was found to be 5.3 × 10 7 M −1 . Liu and Wang took advantage of the encapsulating and adsorbing capabilities of cationic surfactant hexadecyl trimethyl ammonium bromide (CTAB) and the selectivity of thymine-squaraine based fluorescent chemosensor 28, to develop a system capable of sensing and capturing Cd 2+ ions, Figure 11 [185]. The thymine moiety of 28 efficiently bound to Cd 2+ ions, while the squaraine moiety served as a signaling unit whose spectroscopic performances could be tuned by changing its solution environments. The CTAB micelles could induce aggregation of the squaraine dyes, reducing the likelihood of any intermolecular electron transfer. On the addition of Cd 2+ ions, the fluorescence intensity of compound 28 increased. Optimal conditions for the system were observed when the concentration of CTAB was less than 1 mM. At higher concentrations, the CTAB micelles made it difficult for the Cd 2+ ions to access the thymine moiety via an adsorption process into the cell. Under these conditions, a 5-fold increase in the emission band at 660 nm was observed for compound 28 in the presence of Cd 2+ ions. There was no notable emission change observed for any other metal ions that were studied (K + , Zn 2+ , Li + , Fe 3+ , Hg 2+ , Fe 2+ , Ag + Co 2+ Na + Ca 2+ , and Cu 2+ ). The sensor was also found to be very sensitive, demonstrating an LoD in the nM range. The binding constant was determined using a Benesi-Hildebrand plot and found to be 2 × 10 3 M −1 . Additionally, a Job's plot Liu and Wang took advantage of the encapsulating and adsorbing capabilities of cationic surfactant hexadecyl trimethyl ammonium bromide (CTAB) and the selectivity of thymine-squaraine based fluorescent chemosensor 28, to develop a system capable of sensing and capturing Cd 2+ ions, Figure 11 [185]. The thymine moiety of 28 efficiently bound to Cd 2+ ions, while the squaraine moiety served as a signaling unit whose spectroscopic performances could be tuned by changing its solution environments. The CTAB micelles could induce aggregation of the squaraine dyes, reducing the likelihood of any intermolecular electron transfer. On the addition of Cd 2+ ions, the fluorescence intensity of compound 28 increased. Optimal conditions for the system were observed when the concentration of CTAB was less than 1 mM. At higher concentrations, the CTAB micelles made it difficult for the Cd 2+ ions to access the thymine moiety via an adsorption process into the cell. Under these conditions, a 5-fold increase in the emission band at 660 nm was observed for compound 28 in the presence of Cd 2+ ions. There was no notable emission change observed for any other metal ions that were studied (K + , Zn 2+ , Li + , Fe 3+ , Hg 2+ , Fe 2+ , Ag + Co 2+ Na + Ca 2+ , and Cu 2+ ). The sensor was also found to be very sensitive, demonstrating an LoD in the nM range. The binding constant was determined using a Benesi-Hildebrand plot and found to be 2 × 10 3 M −1 . Additionally, a Job's plot revealed that the binding stoichiometry between the squaraine moiety and Cd 2+ ions was 1:1. Competition studies did not highlight any notable inhibition, due to the presence of other metal ions. However, the presence of some anionic and non-ionic surfactants (sodium dodecyl sulfate (SDS), Dodecylbenzene sulfonic acid sodiym salt (SDBS), and Triton X-100) did diminish the probe's ability to detect Cd 2+ ions. The micelle was found to be stable to photobleaching, as decomplexation did not occur when the soft material was exposed to constant irradiation for 6 h. Two turn-"on" fluorescent sensors compounds 29a and 29b were synthesized by Xu and coworkers, Scheme 12 [186]. Both compounds could bind both Zn 2+ and Cd 2+ ions in a CH3OH/H2O solution (1:1, v/v, Tris, 10 mmol⋅L −1 , pH 7.4). These molecular probes displayed a weak fluorescence emission band, however, in the presence of Zn 2+ and Cd 2+ ions, an eight-fold increase in the emission intensity, due to CHEF mechanism, was observed at 430 nm. Data obtained from the fluorescence titrations and Job's plots, suggested a 1:1 complex formation. This was supported by NMR studies. The calculated association constants of compounds 29a and 29b to Cd 2+ ions were 1.5 × 10 7 M −1 and 1.0 × 10 6 M −1 , respectively. Competition studies were carried out using other metal ions but there were no notable interferences observed in the fluorescence spectra. There were however interferences from biologically relevant anions, namely adenosine triphosphate (ATP), adenosine diphosphate (ADP), and other phosphate ions. Complete quenching was observed when phosphate ions were introduced to a solution of the complex demonstrating the complex's ability to serve as an on-off sensor for phosphate ions. Figure 11. The proposed intermolecular interaction between compound 28 and the Cd 2+ ions encapsulated in the micelle. Adapted and reprinted from [183], with permission from Elsevier. Two turn-"on" fluorescent sensors compounds 29a and 29b were synthesized by Xu and coworkers, Scheme 12 [186]. Both compounds could bind both Zn 2+ and Cd 2+ ions in a CH 3 OH/H 2 O solution (1:1, v/v, Tris, 10 mmol·L −1 , pH 7.4). These molecular probes displayed a weak fluorescence emission band, however, in the presence of Zn 2+ and Cd 2+ ions, an eight-fold increase in the emission intensity, due to CHEF mechanism, was observed at 430 nm. Data obtained from the fluorescence titrations and Job's plots, suggested a 1:1 complex formation. This was supported by NMR studies. The calculated association constants of compounds 29a and 29b to Cd 2+ ions were 1.5 × 10 7 M −1 and 1.0 × 10 6 M −1 , respectively. Competition studies were carried out using other metal ions but there were no notable interferences observed in the fluorescence spectra. There were however interferences from biologically relevant anions, namely adenosine triphosphate (ATP), adenosine diphosphate (ADP), and other phosphate ions. Complete quenching was observed when phosphate ions were introduced to a solution of the complex demonstrating the complex's ability to serve as an on-off sensor for phosphate ions. Scheme 12. The proposed interaction between the molecular sensors and the metal ions. The binding cavity size was ideal to chelate Cd 2+ or Zn 2+ ions. When the phosphate anion was added, the fluorescence intensity was switched off.
Kim and Harrison synthesized a tripodal receptor containing quinoline and pyridyl aminophenol moieties (compound 30) [187]. It was shown to prefer to coordinate with Zn 2+ ions in CH3CN solution but had a preference for Cd 2+ ions in aqueous solutions. Optical investigations were carried out in the presence of 19 different nitrate metal salts (Na + , K + , Zn 2+ , Cd 2+ , Mg 2+ , Ca 2+ , Al 3+ , Ga 3+ , In 3+ , Pb + , Hg 2+ , Cr 3+ , Mn 2+ , Fe 2+ , Fe 3+ , Co 2+ , Ni 2+ , Cu 2+ , Ag + ) at 10 μM in CH3CN solution. The addition of Zn 2+ ions lead to a large fluorescence enhancement (φ = 0.001 to φ = 0.24) at 510 nm. The only other metal ion that produced an emission intensity increase was Cd 2+ but to a lesser extent, approximately one third of the magnitude seen for Zn 2+ (φ= 0.039). The other metals did not lead to an increase in fluorescent intensity, but several did interfere with the probe's ability to bind to Zn 2+ . Similar optical investigations were undertaken in a CH3CN:H2O (1:1, v/v) HEPES-buffered solution (10 mM, pH 7.4). The fluorescence enhancement seen in the presence of Cd 2+ remained the same, but the fluorescence enhancement seen in the presence of Zn 2+ was only 10% of that seen in CH3CN. Other metal ions did not lead to an increase in intensity, but Zn 2+ , Cr 3+ , Cu 2+ , Co 2+ , Ni 2+ , Hg 2+ interfered with Cd 2+ binding. The differences in fluorescence intensities between the two solvent systems suggested that the coordination environments for Cd 2+ ions were similar in both. The Zn 2+ ion could adopt an equilibrium between tetrahedral and octahedral geometries. The binding environment for [Cd(NO3)2 (30)] was confirmed by solid-state studies ( Figure 12) and supported by NMR studies. The weak fluorescence seen for 30 was a consequence of PET quenching, which was inhibited upon the coordination of Zn 2+ ions. The binding constants, in CH3CN were calculated to be 1.4 × 10 6 M −1 and 5.6 × 10 5 M −1 , for Zn 2+ , and Cd 2+ respectively, but it is unclear how these binding constants were obtained, as no binding isotherms were reported. It was also claimed that 30 bound Cd 2+ ions more favorably than Zn 2+ ions in an organic-aqueous system. No binding constants were reported to back up this claim, only a single fluorescence spectrum showed that the intensity was greater for Cd 2+ ions. Scheme 12. The proposed interaction between the molecular sensors and the metal ions. The binding cavity size was ideal to chelate Cd 2+ or Zn 2+ ions. When the phosphate anion was added, the fluorescence intensity was switched off.
Kim and Harrison synthesized a tripodal receptor containing quinoline and pyridyl aminophenol moieties (compound 30) [187]. It was shown to prefer to coordinate with Zn 2+ ions in CH 3 CN solution but had a preference for Cd 2+ ions in aqueous solutions. Optical investigations were carried out in the presence of 19 different nitrate metal salts (Na + , K + , Zn 2+ , Cd 2+ , Mg 2+ , Ca 2+ , Al 3+ , Ga 3+ , In 3+ , Pb + , Hg 2+ , Cr 3+ , Mn 2+ , Fe 2+ , Fe 3+ , Co 2+ , Ni 2+ , Cu 2+ , Ag + ) at 10 µM in CH 3 CN solution. The addition of Zn 2+ ions lead to a large fluorescence enhancement (φ = 0.001 to φ = 0.24) at 510 nm. The only other metal ion that produced an emission intensity increase was Cd 2+ but to a lesser extent, approximately one third of the magnitude seen for Zn 2+ (φ= 0.039). The other metals did not lead to an increase in fluorescent intensity, but several did interfere with the probe's ability to bind to Zn 2+ . Similar optical investigations were undertaken in a CH 3 CN:H 2 O (1:1, v/v) HEPES-buffered solution (10 mM, pH 7.4). The fluorescence enhancement seen in the presence of Cd 2+ remained the same, but the fluorescence enhancement seen in the presence of Zn 2+ was only 10% of that seen in CH 3 CN. Other metal ions did not lead to an increase in intensity, but Zn 2+ , Cr 3+ , Cu 2+ , Co 2+ , Ni 2+ , Hg 2+ interfered with Cd 2+ binding. The differences in fluorescence intensities between the two solvent systems suggested that the coordination environments for Cd 2+ ions were similar in both. The Zn 2+ ion could adopt an equilibrium between tetrahedral and octahedral geometries. The binding environment for [Cd(NO 3 ) 2 (30)] was confirmed by solid-state studies ( Figure 12) and supported by NMR studies. The weak fluorescence seen for 30 was a consequence of PET quenching, which was inhibited upon the coordination of Zn 2+ ions. The binding constants, in CH 3 CN were calculated to be 1.4 × 10 6 M −1 and 5.6 × 10 5 M −1 , for Zn 2+ , and Cd 2+ respectively, but it is unclear how these binding constants were obtained, as no binding isotherms were reported. It was also claimed that 30 bound Cd 2+ ions more favorably than Zn 2+ ions in an organic-aqueous system. No binding constants were reported to back up this claim, only a single fluorescence spectrum showed that the intensity was greater for Cd 2+ ions. Hou and Wang developed a phenylene-acetylene based compound 31 that was capable of selectively detecting Fe 2+ by colorimetric methods and Cd 2+ and Zn 2+ via ratiometric fluorescent detection [188]. Optical studies were carried out in the presence of 23 different chloride or perchlorate Hou and Wang developed a phenylene-acetylene based compound 31 that was capable of selectively detecting Fe 2+ by colorimetric methods and Cd 2+ and Zn 2+ via ratiometric fluorescent detection [188]. Optical studies were carried out in the presence of 23 different chloride or perchlorate metal salts (Li + , Ba 2+ , Ca 2+ , K + , Li+, Mg 2+ , Mn 2+ , Na + , Cr 3+ , Au 3+ , Zr 4+ , Rh 3+ , Pb 2+ , Ag + , Co 2+ , Cu 2+ , Ni 2+ , Hg 2+ , Cd 2+ , Zn 2+ , Fe 2+ , Fe 3+ , Pt 2+ ). In an EtOH:H 2 O (10 µM, 1:1, v/v) solution, 31 showed two π-π * absorption bands at 286 nm and 327 nm, and the addition of Fe 2+ ion led to the appearance of a new peak at 573 nm, with a corresponding color change from colorless to purple-red. The other metals did not produce any significant changes in the absorption spectra, and their presence did not interfere with the probe's response to Fe 2+ . Job's plot and ESI-MS data indicated that the 31:Fe 2+ complex had a 2:1 binding stoichiometry and the association constant was calculated to be 3.4 × 10 7 M −1 . The limit of detection for the method was found to be 2.0 × 10 −8 M. When excited at 340 nm, 31 showed an emission at 408 nm and the addition of Cd 2+ ions led to a decrease in the intensity of the emission at 408 nm, the appearance of a new emission at 475 nm, and a change in fluorescent color from blue to cyan. The addition of Zn 2+ ions led to a decrease in the intensity of the emission at 408 nm, the appearance of a new emission at 498 nm, and a change in fluorescent color from blue to yellow-green. No other metals showed a significant response, and only Fe 2+ interfered with the probe's response to Cd 2+ and Zn 2+ . The ratios of the two emissions (I 475 /I 408 and I 498 /I 408 ) could be used to calculate the LoD for Cd 2+ and Zn 2+ of 2.6 × 10 −8 M and 2.0 × 10 −9 M, respectively. Job's plot and ESI-MS data indicated that the 31:Cd 2+ complex had a 1:1 binding stoichiometry and the association constant was calculated to be 0.2 × 10 6 M −1 . The 31:Zn 2+ complex had a 2:1 binding stoichiometry, and an association constant of 0.9 × 10 6 M −1 . 1 H NMR data indicated that the metals bound to the nitrogen atoms; the proposed sensing mechanism for the probe toward Fe 2+ , Cd 2+ , and Zn 2+ is presented in Scheme 13.

Mercury Probes
Out of the three metal ions in group 12, mercury in the form of organic mercury is the most toxic. Therefore, designing molecular chemosensors for mercury ions is a very important endeavor, especially in aqueous systems, as certain biological species readily convert metallic mercury into organic mercury. There have been many different types of fluorescent sensors for mercury, recent examples include quantum dots [189], biosensors [8], carbon dots [190], carbon nanoparticles [191], gold and silver nanoparticles [192,193], but the focus in this final section is on LMFP sensors.
Highly selective off-on optical sensor 32 was synthesized by Ahmed et al. for the detection of mercuric ions in an organic-aqueous, HEPES-buffered (CH3CN:H2O, 1:2, v/v, pH 7.2) system [194], Compound 32 drastically changed color from yellow (340 nm) to purple (550 nm), upon the addition Scheme 13. The binding modes of probe 31 and the different metal ions.

Mercury Probes
Out of the three metal ions in group 12, mercury in the form of organic mercury is the most toxic. Therefore, designing molecular chemosensors for mercury ions is a very important endeavor, especially in aqueous systems, as certain biological species readily convert metallic mercury into organic mercury. There have been many different types of fluorescent sensors for mercury, recent examples include quantum dots [189], biosensors [8], carbon dots [190], carbon nanoparticles [191], gold and silver nanoparticles [192,193], but the focus in this final section is on LMFP sensors.
Highly selective off-on optical sensor 32 was synthesized by Ahmed et al. for the detection of mercuric ions in an organic-aqueous, HEPES-buffered (CH 3 CN:H 2 O, 1:2, v/v, pH 7.2) system [194], Compound 32 drastically changed color from yellow (340 nm) to purple (550 nm), upon the addition of Hg 2+ ions and was also found to be highly fluorescent, with an emission band at 430 nm, when the coordination complex was formed in situ. The weak fluorescence emission of 32 was a consequence of PET quenching which was inhibited upon by the coordination of the Hg 2+ ions. The binding constant was calculated to be 1.9 × 10 3 M −1 using the Benesi-Hildebrand linear regression analysis, which showed a straight line. However on closer inspection of the regression line, the data presented was not a Benesi-Hildebrand plot. The authors plotted an intensity ratio versus concentration of Hg 2+ ions, therefore, the binding constant reported should be viewed with a little skepticism. The coordination environment was confirmed by IR and NMR studies, as compound 32 had distinct functional groups (carbonyl and hydrazide) that could act as IR labels, the stretching frequencies shifted when coordinated to Hg 2+ ion, the tridentate binding motif was further supported by DFT calculations. The Hg 2+ ion was removed from the binding site by the addition of KI ions to precipitate HgI 2 , which turned off the emission signal. Therefore, this "off-on-off" approach could be translated into binary code to build an INHIBIT logic gate (a combination function consisting of and AND gate plus one other event, in this example a NOT event), Figure 13.  [194].
Aggregation-induced emission (AIE) has become a popular mechanism exploited for sensors often used in chemodosimeter and chemoreactand type molecular probes. A recent review covers the use of chemodosimeters (non-reversible covalent bonds and chemoreactands with reversible covalent bonds), whereby, the optical properties were different from the bound and unbound chemical adducts [53]. Chatterjee and coworkers have prepared chemodosimeter 33 to detect organic and inorganic Hg 2+ species, Figure 14 [193]. The probe functions based on a mercury ion-promoted  [194].
Aggregation-induced emission (AIE) has become a popular mechanism exploited for sensors often used in chemodosimeter and chemoreactand type molecular probes. A recent review covers the use of chemodosimeters (non-reversible covalent bonds and chemoreactands with reversible covalent bonds), whereby, the optical properties were different from the bound and unbound chemical adducts [53]. Chatterjee and coworkers have prepared chemodosimeter 33 to detect organic and inorganic Hg 2+ species, Figure 14 [193]. The probe functions based on a mercury ion-promoted transmetalation reaction between tetraphenylethylene mono-boronic acid 33 and Hg 2+ ions. For aggregation to occur, the correct working concentration needed to be established, which was found to be 30 µM (H 2 O-THF (90:10)). On the addition of HgCl 2 to probe 33, transmetalation occurred to form compound 33a. As a consequence of the chemical transformation, the planarity of 33a-HgCl was restored and the onset of AIE occurred. The AIE aggregation was supported by transmission electron microscopy (TEM) and dynamic light scattering (DLS) experiments, which showed average nanoaggregates of 36 nm but, after the addition of HgCl 2 to the 30 µM solution of 33, these grew to 690 nm. No other metal was shown to transmetalate compound 33, therefore, the molecular chemodosimeter was very specific for Hg 2+ ions. Interestingly, methylmercury ion was also shown to undergo transmetalation, forming 33b, which also aggregated and turned on emission. As methylmercury ion readily bioaccumulates in fish, the chemodosimeter was tested using zebrafish embryos. In the presence of methylmercury ion, fluorescence microscopic images showed that a distinctive blue glow occurred. This elegant experiment showed that transmetallation of the C-B bond to C-Hg bond, and aggregation can occur in vivo, making this a highly attractive approach to monitor organomercury species, in higher order fish species. Another chemodosimeter, has been developed by Ou et al. [196] who employed a protectiondeprotection strategy to monitor mercury. This approach takes advantage of the distinctly different electronic properties of two different molecules. For example, protected aldehydes and ketones can be deprotected with Hg 2+ salts. As no other metal ion causes a selective cleavage under these conditions, the molecular probes are highly selective for Hg 2+ ions. Thioacetal-modified pyrene sensor 34 was prepared and was shown to be cleaved by Hg 2+ ions, to form pyrenecarboxaldehyde 34a, Scheme 14. Compound 34 undergoes an off-on fluorescence response when Hg 2+ ions are added. The Hg 2+ ions deprotect the thioacetals to form the aldehyde functional group, confirmed by 1 H NMR as the aldehyde proton appeared at 10.82 ppm. The UV-Vis and fluorescence experiments were used to Another chemodosimeter, has been developed by Ou et al. [196] who employed a protection-deprotection strategy to monitor mercury. This approach takes advantage of the distinctly different electronic properties of two different molecules. For example, protected aldehydes and ketones can be deprotected with Hg 2+ salts. As no other metal ion causes a selective cleavage under these conditions, the molecular probes are highly selective for Hg 2+ ions. Thioacetal-modified pyrene sensor 34 was prepared and was shown to be cleaved by Hg 2+ ions, to form pyrenecarboxaldehyde 34a, Scheme 14. Compound 34 undergoes an off-on fluorescence response when Hg 2+ ions are added. The Hg 2+ ions deprotect the thioacetals to form the aldehyde functional group, confirmed by 1 H NMR as the aldehyde proton appeared at 10.82 ppm. The UV-Vis and fluorescence experiments were used to verify that one equivalent of Hg 2+ ions were needed to fully convert 34 to 34a.
Hg 2+ ion and CH3Hg + . Reprinted (adapted) with permission from [195]. (A)Chemical Scheme, (B) proposed fluorescence aggregation Another chemodosimeter, has been developed by Ou et al. [196] who employed a protectiondeprotection strategy to monitor mercury. This approach takes advantage of the distinctly different electronic properties of two different molecules. For example, protected aldehydes and ketones can be deprotected with Hg 2+ salts. As no other metal ion causes a selective cleavage under these conditions, the molecular probes are highly selective for Hg 2+ ions. Thioacetal-modified pyrene sensor 34 was prepared and was shown to be cleaved by Hg 2+ ions, to form pyrenecarboxaldehyde 34a, Scheme 14. Compound 34 undergoes an off-on fluorescence response when Hg 2+ ions are added. The Hg 2+ ions deprotect the thioacetals to form the aldehyde functional group, confirmed by 1 H NMR as the aldehyde proton appeared at 10.82 ppm. The UV-Vis and fluorescence experiments were used to verify that one equivalent of Hg 2+ ions were needed to fully convert 34 to 34a.

Scheme 14. Chemodosimeter 34.
The pyridine ring is a versatile moiety and is often found in molecular receptors. Tetrahydrodibenzo phenanthridines, contain a pyridine functional group with five fused sixmembered ring systems with connectivity, shown in Figure 15. This arrangement of ring systems are found in natural products and biologically active compounds [197,198]. Two bis(dithioester) substituted phenanthridines, compound 35 and compound 36, were synthesized by Sundaramurthy et al., who showed that they could coordinate with Hg 2+ ions [199]. Mono-sulfur derivative 35 demonstrated a strong fluorescence turn on signal in an aqueous solution, in the presence of Hg 2+ ions, whereas tetra-sulfur derivative 36 showed a strong fluorescence quenching The pyridine ring is a versatile moiety and is often found in molecular receptors. Tetrahydrodibenzo phenanthridines, contain a pyridine functional group with five fused six-membered ring systems with connectivity, shown in Figure 15. This arrangement of ring systems are found in natural products and biologically active compounds [197,198]. Two bis(dithioester) substituted phenanthridines, compound 35 and compound 36, were synthesized by Sundaramurthy et al., who showed that they could coordinate with Hg 2+ ions [199]. Mono-sulfur derivative 35 demonstrated a strong fluorescence turn on signal in an aqueous solution, in the presence of Hg 2+ ions, whereas tetra-sulfur derivative 36 showed a strong fluorescence quenching towards Hg 2+ ions. The binding stoichiometry of both sensors toward Hg 2+ ions was 1:1, according to the UV-Vis absorption spectra, fluorescence measurements, and computation (DFT) studies. The LoD for the mono-sulfur derivative was found to be as low as 0.91 nM for Hg 2+ ions, while the bis(dithioester) derivative demonstrated an LoD of 0.04 nM. The spectroscopic changes were also seen in paper-based test strips as seen in Figure 15. towards Hg 2+ ions. The binding stoichiometry of both sensors toward Hg 2+ ions was 1:1, according to the UV-Vis absorption spectra, fluorescence measurements, and computation (DFT) studies. The LoD for the mono-sulfur derivative was found to be as low as 0.91 nM for Hg 2+ ions, while the bis(dithioester) derivative demonstrated an LoD of 0.04 nM. The spectroscopic changes were also seen in paper-based test strips as seen in Figure 15. Adapted from [199] with permission from the Centre National de la Recherche Scientifique (CNRS) and the Royal Society of Chemistry.
Anthracene-based, reusable fluorescent sensor 37 was synthesized via a one-pot reaction imine condensation by Misra et al. [200]. As anthracene is hydrophobic, it was expected that compound 37 would aggregate when the solvent mixture had a high content of water. By changing the THF:H2O ratio, it was found that the optimal working condition required 70% water. A strong emission band around 518 nm was observed (20 μM) with a calculated φ of 0.614. As the water content was increased, the molecules came closer together to restrict intramolecular rotation around the imine bond and, therefore, increased the fluorescence emissions, Scheme 15. In solutions where the water content was low, very weak emission bands were observed in the fluorescence spectra (φ = 0.002). When Hg(NO3)2 was added to the aggregates, significant optical quenching was seen in the fluorescence emission,  [199] with permission from the Centre National de la Recherche Scientifique (CNRS) and the Royal Society of Chemistry.
Anthracene-based, reusable fluorescent sensor 37 was synthesized via a one-pot reaction imine condensation by Misra et al. [200]. As anthracene is hydrophobic, it was expected that compound 37 would aggregate when the solvent mixture had a high content of water. By changing the THF:H 2 O ratio, it was found that the optimal working condition required 70% water. A strong emission band around 518 nm was observed (20 µM) with a calculated φ of 0.614. As the water content was increased, the molecules came closer together to restrict intramolecular rotation around the imine bond and, therefore, increased the fluorescence emissions, Scheme 15. In solutions where the water content was low, very weak emission bands were observed in the fluorescence spectra (φ = 0.002). When Hg(NO 3 ) 2 was added to the aggregates, significant optical quenching was seen in the fluorescence emission, due to the heavy atom effect. This was a consequence of perturbating the intersystem crossing from the first singlet state (S 1 ) to the lowest triplet state (T 1 ), thereby, increasing the spin-orbit coupling from the aromatic scaffold of the molecular probe to the external heavy atom [201,202], upon the coordination of the Hg 2+ ions to 37, through chelation. Therefore, the Stern-Volmer analysis showed the quenching constant (K sv ) to be 2.5 × 10 5 M −1 and the LoD was found to be in the ppb concentration range (3 ppb). Moreover, analysis of municipal water and river water from the local area, using probe 37 detected Hg 2+ ions. The rhodamine fluorophore has been used as a molecular probe for a variety of metal ions over the years and xanthene derivatives have previously been highlighted in this review. Compound 38 was shown to distinguish between Hg 2+ and Cu 2+ ions. It was prepared by Gao and Zheng who showed that either in the absence or presence of UV light, a distinction could be made between Hg 2+ and Cu 2+ ions, scheme 16 [203]. Optical studies were carried out in the presence of 16 different metal nitrate salts (Hg 2+ , Cu 2+ , Ag + , K + , Al 3+ , Ca 2+ , Na + , Ba 2+ , Co 2+ , Cd 2+ , Fe 3+ , Mg 2+ , Mn 2+ , Zn 2+ , Ni 2+ , Pb 2+ ) in a IPA:H2O solvent (25 μM, 8:1, v/v). Probe 38 was initially colorless, indicative of the closed lactam moiety. However, on the addition of one equivalent of Hg 2+ ions, a new band at 557 nm was observed (colorless to pink), indicating that the spirolactam ring was in the open form, which is very common for these molecular probes. Compound 38 shows a very weak emission intensity, but an emission band was seen at 578 nm, when Hg 2+ ions were added to the solution (λex = 550 nm). This phenomenon was not seen in the presence of the other metal cations. Copper(II) is well-known to quench the fluorescence emission, therefore, it was not surprising that an emission signal was not seen on the addition of Cu 2+ ions, but on the addition of Hg 2+ ions, a new absorption band was seen at 559 nm and a visible color change from colorless to purple was observed. The coexistence of Hg 2+ and Cu 2+ in the solution showed a decrease in the emission intensity of the band at 559 nm, indicating that the presence of Hg 2+ interfered with Cu 2+ detection. One way to deliberately make their mixed system monitor one of the ions over the other, is to take advantage of solubility products. Bromide ions were used to precipitate HgBr2 (Ksp = 3 × 10 −7 ), enabling the detection of Cu 2+ ions. The rhodamine fluorophore has been used as a molecular probe for a variety of metal ions over the years and xanthene derivatives have previously been highlighted in this review. Compound 38 was shown to distinguish between Hg 2+ and Cu 2+ ions. It was prepared by Gao and Zheng who showed that either in the absence or presence of UV light, a distinction could be made between Hg 2+ and Cu 2+ ions, Scheme 16 [203]. Optical studies were carried out in the presence of 16 different metal nitrate salts (Hg 2+ , Cu 2+ , Ag + , K + , Al 3+ , Ca 2+ , Na + , Ba 2+ , Co 2+ , Cd 2+ , Fe 3+ , Mg 2+ , Mn 2+ , Zn 2+ , Ni 2+ , Pb 2+ ) in a IPA:H 2 O solvent (25 µM, 8:1, v/v). Probe 38 was initially colorless, indicative of the closed lactam moiety. However, on the addition of one equivalent of Hg 2+ ions, a new band at 557 nm was observed (colorless to pink), indicating that the spirolactam ring was in the open form, which is very common for these molecular probes. Compound 38 shows a very weak emission intensity, but an emission band was seen at 578 nm, when Hg 2+ ions were added to the solution (λ ex = 550 nm). This phenomenon was not seen in the presence of the other metal cations. Copper(II) is well-known to quench the fluorescence emission, therefore, it was not surprising that an emission signal was not seen on the addition of Cu 2+ ions, but on the addition of Hg 2+ ions, a new absorption band was seen at 559 nm and a visible color change from colorless to purple was observed. The coexistence of Hg 2+ and Cu 2+ in the solution showed a decrease in the emission intensity of the band at 559 nm, indicating that the presence of Hg 2+ interfered with Cu 2+ detection. One way to deliberately make their mixed system monitor one of the ions over the other, is to take advantage of solubility products. Bromide ions were used to precipitate HgBr 2 (K sp = 3 × 10 −7 ), enabling the detection of Cu 2+ ions. fluorescence emission, therefore, it was not surprising that an emission signal was not seen on the addition of Cu 2+ ions, but on the addition of Hg 2+ ions, a new absorption band was seen at 559 nm and a visible color change from colorless to purple was observed. The coexistence of Hg 2+ and Cu 2+ in the solution showed a decrease in the emission intensity of the band at 559 nm, indicating that the presence of Hg 2+ interfered with Cu 2+ detection. One way to deliberately make their mixed system monitor one of the ions over the other, is to take advantage of solubility products. Bromide ions were used to precipitate HgBr2 (Ksp = 3 × 10 −7 ), enabling the detection of Cu 2+ ions. As mercury is classified as a "soft" metal, mercury ions forms many stable ionic compounds that are insoluble [204,205]. Thermodynamic stability is further enhanced when "soft" Lewis bases are present in the receptor, for example sulfur atoms [206]. The carbonyl group (C=O) in the lactam ring in rhodamine probes could be readily converted to the analogous thio (C=S) group, by using Scheme 16. The coordination of 38 to Hg 2+ ions.
As mercury is classified as a "soft" metal, mercury ions forms many stable ionic compounds that are insoluble [204,205]. Thermodynamic stability is further enhanced when "soft" Lewis bases are present in the receptor, for example sulfur atoms [206]. The carbonyl group (C=O) in the lactam ring in rhodamine probes could be readily converted to the analogous thio (C=S) group, by using Lawesson's reagent. Zhou, Hu and Yao synthesized thiooxo-rhodamine B hydrazine compounds 39 and 40, which contain a number of sulfur atoms to help the molecular probe to selectively coordinate to Hg 2+ over other metal ions [203]. Both molecular probes are colorless but addition of Hg 2+ ions leads to a significant increase in absorption at 560 nm, due to the spirolactam ring-opening. No optical changes were seen in the presence of the other metals (Na + , K + , Mg 2+ , Ca 2+ , Al 3+ , Cr 3+ , Ba 2+ , Hg 2+ , Mn 2+ , Pb 2+ , Cd 2+ , Ni 2+ , Zn 2+ , Co 2+ , Fe 3+ ) in a EtOH/HEPES solution (5 µM 1:1, v/v pH 7.4) but Ag + and Cu 2+ ions increased in intensity at 560 nm. Probe 40 selectively detected Hg 2+ ions via an "off-on" response in solution. The Job's plot and UV-Vis titration data indicated that the stoichiometry between 40 and Hg 2+ ions was 1:2. This is unsurprising as there is free rotation around the bisthiophene bond, so the two rhodamine units could adapt an anti-conformation with the Hg 2+ ions bound in each coordination environment. By comparison, compound 39 bound the Hg 2+ ions in a 1:1 ratio. The association constants for 39 and 40 with mercury were calculated using the Benesi-Hildebrand method and was found to be 4 × 10 5 M −1 and 1.3 × 10 11 M −2 , respectively. Interestingly, the fluorescence mechanism for the two molecular probes was very different. Molecular probe 39 was shown to exhibit a FRET mechanism, due to overlap between the emission spectra of the bisthiophene donor and the absorption spectra of the rhodamine B acceptor. Compound 40 did not show FRET and the fluorescence emission was due to the ring-opening of the spriolactam. The authors also demonstrated that compounds 39 and 40 could monitor the levels of Hg 2+ ions in municipal water. Lawesson's reagent. Zhou, Hu and Yao synthesized thiooxo-rhodamine B hydrazine compounds 39 and 40, which contain a number of sulfur atoms to help the molecular probe to selectively coordinate to Hg 2+ over other metal ions [203]. Both molecular probes are colorless but addition of Hg 2+ ions leads to a significant increase in absorption at 560 nm, due to the spirolactam ring-opening. No optical changes were seen in the presence of the other metals (Na + , K + , Mg 2+ , Ca 2+ , Al 3+ , Cr 3+ , Ba 2+ , Hg 2+ , Mn 2+ , Pb 2+ , Cd 2+ , Ni 2+ , Zn 2+ , Co 2+ , Fe 3+ ) in a EtOH/HEPES solution (5 μM 1:1, v/v pH 7.4) but Ag + and Cu 2+ ions increased in intensity at 560 nm. Probe 40 selectively detected Hg 2+ ions via an "off-on" response in solution. The Job's plot and UV-Vis titration data indicated that the stoichiometry between 40 and Hg 2+ ions was 1:2. This is unsurprising as there is free rotation around the bisthiophene bond, so the two rhodamine units could adapt an anti-conformation with the Hg 2+ ions bound in each coordination environment. By comparison, compound 39 bound the Hg 2+ ions in a 1:1 ratio. The association constants for 39 and 40 with mercury were calculated using the Benesi-Hildebrand method and was found to be 4 × 10 5 M −1 and 1.3 × 10 11 M −2 , respectively. Interestingly, the fluorescence mechanism for the two molecular probes was very different. Molecular probe 39 was shown to exhibit a FRET mechanism, due to overlap between the emission spectra of the bisthiophene donor and the absorption spectra of the rhodamine B acceptor. Compound 40 did not show FRET and the fluorescence emission was due to the ring-opening of the spriolactam. The authors also demonstrated that compounds 39 and 40 could monitor the levels of Hg 2+ ions in municipal water.
4,4-Difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY) possesses unique spectroscopic signatures and the BODIPY moiety has been used extensively in molecular probe design [207][208][209]. Qiu, Zhang, and Yin synthesized a BODIPY-based probe with a dipyridylamino (DPA) receptor (compound 41), for which the optical response was different for Cu 2+ , Hg 2+ , and Pb 2+ ions [210]. On the addition of the Cu 2+ , Hg 2+ , and Pb 2+ ions, the optical response was different for each metal ion. Optical studies were carried out in the presence of a number of different perchlorate metal salts (Ag + , Ba 2+ , Ca 2+ , Cd 2+ , Co 2+ , Cu 2+ , Fe 2+ , Hg 2+ , K + , Mg 2+ , Na + , Ni 2+ , Pb 2+ , Zn 2+ ) in a 2.0 μM CH3CN solution. The presence of Hg 2+ and Pb 2+ in the solution, lead to a blue shift in the maximum absorbance peak from 595 nm to 572 nm and a visible color change from blue to purple. The presence of Cu 2+ ions in the solution led to a blue shift in the maximum absorbance from 595 nm to 512 nm, and a visible color change from blue to fuchsia. Compound 41 showed a single weak emission, when it was excited at 365 nm. The presence of Hg 2+ and Pb 2+ ions led to a new emission band at 587 nm and a change from dark red fluorescence to orange fluorescence. The presence of Cu 2+ ions led to an emission band at 545 nm and a change from dark red fluorescence to green fluorescence. Competition studies have showed that the presence of Cu 2+ ions in the solution quenched the Hg 2+ and Pb 2+ ions fluorescent responses. The presence of other metal ions did not influence the Cu 2+ response. Both Hg 2+ and Pb 2+ ions could be removed from the binding site of 41, when EDTA was added to the solution. A Job's plot and ESI-MS data indicated a 1:1 binding stoichiometry, and 1 H NMR data indicated that the 4,4-Difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY) possesses unique spectroscopic signatures and the BODIPY moiety has been used extensively in molecular probe design [207][208][209]. Qiu, Zhang, and Yin synthesized a BODIPY-based probe with a dipyridylamino (DPA) receptor (compound 41), for which the optical response was different for Cu 2+ , Hg 2+ , and Pb 2+ ions [210]. On the addition of the Cu 2+ , Hg 2+ , and Pb 2+ ions, the optical response was different for each metal ion. Optical studies were carried out in the presence of a number of different perchlorate metal salts (Ag + , Ba 2+ , Ca 2+ , Cd 2+ , Co 2+ , Cu 2+ , Fe 2+ , Hg 2+ , K + , Mg 2+ , Na + , Ni 2+ , Pb 2+ , Zn 2+ ) in a 2.0 µM CH 3 CN solution. The presence of Hg 2+ and Pb 2+ in the solution, lead to a blue shift in the maximum absorbance peak from 595 nm to 572 nm and a visible color change from blue to purple. The presence of Cu 2+ ions in the solution led to a blue shift in the maximum absorbance from 595 nm to 512 nm, and a visible color change from blue to fuchsia. Compound 41 showed a single weak emission, when it was excited at 365 nm. The presence of Hg 2+ and Pb 2+ ions led to a new emission band at 587 nm and a change from dark red fluorescence to orange fluorescence. The presence of Cu 2+ ions led to an emission band at 545 nm and a change from dark red fluorescence to green fluorescence. Competition studies have showed that the presence of Cu 2+ ions in the solution quenched the Hg 2+ and Pb 2+ ions fluorescent responses. The presence of other metal ions did not influence the Cu 2+ response. Both Hg 2+ and Pb 2+ ions could be removed from the binding site of 41, when EDTA was added to the solution. A Job's plot and ESI-MS data indicated a 1:1 binding stoichiometry, and 1 H NMR data indicated that the probe's fluorescent and colorimetric responses were due to π delocalization between the N-pyridyl and BODIPY groups, caused by the coordination of the metal ions to the BODIPY and DPA groups. As Hg 2+ , Pb 2+ and Cu 2+ produced different emission intensities it was possible to construct OR, NOT, and AND logic circuits, where the inputs were the metal ions and the output was the emission intensity at 587 nm. If the emission intensity was greater than a chosen arbitrary unit (150 au), then "1" was assigned for (i) both Hg 2+ and Pb 2+ , (ii) Hg 2+ , or (iii) Pb 2+ on their own were present, meaning that the gate was opened. If, Cu 2+ was present alone, with Hg 2+ or Pb 2+ or both Hg 2+ and Pb 2+ , it led to a situation whereby the intensity was lower than 150 au, therefore a "0" was assigned, meaning the gate was closed. Therefore, the combinations of the three different metal ions gave rise to different logic gates (OR, NOT, and AND), Figure 16. a situation whereby the intensity was lower than 150 au, therefore a "0" was assigned, meaning the gate was closed. Therefore, the combinations of the three different metal ions gave rise to different logic gates (OR, NOT, and AND), Figure 16. Gao and Chen synthesized a Schiff-base probe with coumarin and benzimidazole fluorophores (42), which was capable of simultaneously detecting Hg 2+ and Cu 2+ ions, by two different mechanisms [211]. Compound 42 produced a fluorescence emission enhancement, when bound to Hg 2+ ions, whereas, the Cu 2+ ions quenched the emission band. A weak fluorescence intensity at 426 nm (φ = 0.009) was seen for 42 in a 10 μM HEPES:DMSO (20 mM, 9:1, v/v) solution. A significant enhancement was seen in the presence of Hg 2+ ions (φ = 0.152), whereas in the presence of Cu 2+ ions, a red shift to Gao and Chen synthesized a Schiff-base probe with coumarin and benzimidazole fluorophores (42), which was capable of simultaneously detecting Hg 2+ and Cu 2+ ions, by two different mechanisms [211]. Compound 42 produced a fluorescence emission enhancement, when bound to Hg 2+ ions, whereas, the Cu 2+ ions quenched the emission band. A weak fluorescence intensity at 426 nm (φ = 0.009) was seen for 42 in a 10 µM HEPES:DMSO (20 mM, 9:1, v/v) solution. A significant enhancement was seen in the presence of Hg 2+ ions (φ = 0.152), whereas in the presence of Cu 2+ ions, a red shift to 448 nm was seen, but the emission band was reduced in intensity (φ = 0.005). Job's plot analysis and fluorescence titration data both indicated a 1:1 binding ratio for both the Cu 2+ and Hg 2+ complexes. The structure of the complexes was supported by HRMS, IR, and NMR studies. The Hg 2+ ion bound in a tridentate fashion, but the molecular probes underwent hydrolysis via a mercury-promoted hydrolysis (Scheme 17), so the fluorescence emission was due to the coumarin-carboxaldehyde being formed. Fluorescence titration data were used to determine the probe's sensitivity to Hg 2+ and Cu 2+ ions, and the Benesi-Hildebrand equation used to calculate the association constants for the probe towards Cu 2+ ions and Hg 2+ ions, gave 7 × 10 4 M −1 and 9 × 10 4 M −1 , respectively. Another example of molecular chemodosimeter that can selectively monitor Hg 2+ ions is compound 43 [212]. A solution of compound 43 in 10 μM HEPES (10 mM, pH 7) had a maximum absorption band at 315 nm, and when excited at 371 nm, a weak emission band was observed at 450 nm. Interestingly, the authors failed to mention why the excitation wavelength used in their studies was so far removed from the maximum absorption band. An increase in absorption and emission intensity was seen when Hg 2+ ions were added to 43. The fluorescence quantum yield of 43 was calculated with respect to rhodamine B and found to be φ = 0.027. The addition of one equivalence of Hg 2+ led to a ten-fold increase in fluorescent intensity (φ = 0.22), which was accompanied by a change in fluorescence from colorless to blue. No other metal cation showed the same response (Zn 2+ , Hg 2+ , Na + , Pb 2+ , Co 2+ , K + , Ni 2+ , Cu 2+ , Al 3+ , Mn 2+ , Cr 3+ , Mg 2+ , Fe 3+ , Cd 2+ ). ESI-MS and NMR studies showed that the fluorescence emission was due to the cleavage of the carbonothioate, via a Hg 2+ -promoted hydrolysis, thereby, forming 7-hydroxy-4-methylcoumarin dye, which was highly fluorescent. This "simple" molecule was used to monitor deliberately spiked tap water with Hg 2+ , to show that it could be used in a real-world application.
The imine functional group can undergo hydrolysis under acidic conditions [213]. This is a wellknown and thoroughly investigated mechanism, whereby the rate determine step is the initial nucleophilic addition to the carbon-nitrogen double bond. It has also been shown that metal ions can accelerate the hydrolysis mechanism via a substrate-metal association, whereby the metal ion is coordinated to the nitrogen atom, enhancing the substrate for nucleophilic attack [214]. Compound 44 was prepared by Duan et al. [215]. It was shown that compound 44 works as a turnon sensor through Hg 2+ promoted hydrolysis, in which the imine is cleaved to form the coumarin aldehyde fluorophore and a 5-aminoisophthalic acid methyl ester. The cleavage did not occur in the presence of other metal ions (Na +, K + , Mg 2+ , Ca 2+ , Cd 2+ , Mn 2+ , Fe 2+ , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , Ag + , Pb 2+ , Hg 2+ ) Scheme 17. The coordination and subsequent hydrolysis of molecular probe 42.
Another example of molecular chemodosimeter that can selectively monitor Hg 2+ ions is compound 43 [212]. A solution of compound 43 in 10 µM HEPES (10 mM, pH 7) had a maximum absorption band at 315 nm, and when excited at 371 nm, a weak emission band was observed at 450 nm. Interestingly, the authors failed to mention why the excitation wavelength used in their studies was so far removed from the maximum absorption band. An increase in absorption and emission intensity was seen when Hg 2+ ions were added to 43. The fluorescence quantum yield of 43 was calculated with respect to rhodamine B and found to be φ = 0.027. The addition of one equivalence of Hg 2+ led to a ten-fold increase in fluorescent intensity (φ = 0.22), which was accompanied by a change in fluorescence from colorless to blue. No other metal cation showed the same response (Zn 2+ , Hg 2+ , Na + , Pb 2+ , Co 2+ , K + , Ni 2+ , Cu 2+ , Al 3+ , Mn 2+ , Cr 3+ , Mg 2+ , Fe 3+ , Cd 2+ ). ESI-MS and NMR studies showed that the fluorescence emission was due to the cleavage of the carbonothioate, via a Hg 2+ -promoted hydrolysis, thereby, forming 7-hydroxy-4-methylcoumarin dye, which was highly fluorescent. This "simple" molecule was used to monitor deliberately spiked tap water with Hg 2+ , to show that it could be used in a real-world application. Another example of molecular chemodosimeter that can selectively monitor Hg 2+ ions is compound 43 [212]. A solution of compound 43 in 10 μM HEPES (10 mM, pH 7) had a maximum absorption band at 315 nm, and when excited at 371 nm, a weak emission band was observed at 450 nm. Interestingly, the authors failed to mention why the excitation wavelength used in their studies was so far removed from the maximum absorption band. An increase in absorption and emission intensity was seen when Hg 2+ ions were added to 43. The fluorescence quantum yield of 43 was calculated with respect to rhodamine B and found to be φ = 0.027. The addition of one equivalence of Hg 2+ led to a ten-fold increase in fluorescent intensity (φ = 0.22), which was accompanied by a change in fluorescence from colorless to blue. No other metal cation showed the same response (Zn 2+ , Hg 2+ , Na + , Pb 2+ , Co 2+ , K + , Ni 2+ , Cu 2+ , Al 3+ , Mn 2+ , Cr 3+ , Mg 2+ , Fe 3+ , Cd 2+ ). ESI-MS and NMR studies showed that the fluorescence emission was due to the cleavage of the carbonothioate, via a Hg 2+ -promoted hydrolysis, thereby, forming 7-hydroxy-4-methylcoumarin dye, which was highly fluorescent. This "simple" molecule was used to monitor deliberately spiked tap water with Hg 2+ , to show that it could be used in a real-world application.
The imine functional group can undergo hydrolysis under acidic conditions [213]. This is a wellknown and thoroughly investigated mechanism, whereby the rate determine step is the initial nucleophilic addition to the carbon-nitrogen double bond. It has also been shown that metal ions can accelerate the hydrolysis mechanism via a substrate-metal association, whereby the metal ion is coordinated to the nitrogen atom, enhancing the substrate for nucleophilic attack [214]. Compound 44 was prepared by Duan et al. [215]. It was shown that compound 44 works as a turnon sensor through Hg 2+ promoted hydrolysis, in which the imine is cleaved to form the coumarin aldehyde fluorophore and a 5-aminoisophthalic acid methyl ester. The cleavage did not occur in the presence of other metal ions (Na +, K + , Mg 2+ , Ca 2+ , Cd 2+ , Mn 2+ , Fe 2+ , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , Ag + , Pb 2+ , Hg 2+ ) The imine functional group can undergo hydrolysis under acidic conditions [213]. This is a well-known and thoroughly investigated mechanism, whereby the rate determine step is the initial nucleophilic addition to the carbon-nitrogen double bond. It has also been shown that metal ions can accelerate the hydrolysis mechanism via a substrate-metal association, whereby the metal ion is coordinated to the nitrogen atom, enhancing the substrate for nucleophilic attack [214]. Compound 44 was prepared by Duan et al. [215]. It was shown that compound 44 works as a turn-on sensor through Hg 2+ promoted hydrolysis, in which the imine is cleaved to form the coumarin aldehyde fluorophore and a 5-aminoisophthalic acid methyl ester. The cleavage did not occur in the presence of other metal ions (Na +, K + , Mg 2+ , Ca 2+ , Cd 2+ , Mn 2+ , Fe 2+ , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , Ag + , Pb 2+ , Hg 2+ ) in a 2 × 10 −6 M CH 3 CN:H 2 O 8:2 (v/v) solution at physiological pH (0.1 M KClO 4 buffer, pH 7.3). Compound 44 displayed a weak fluorescence signal at 530 nm. The lone pair on the nitrogen atom PET, quenched the emission intensity. On the addition of Hg 2+ ions, a 20-fold increase in emission was seen at 490 nm. The emission at 490 nm was assigned to the coumarin aldehyde fluorophore, and the ratio between the intensity at 490 nm and 530 nm (I 490 /I 530 ) was used to detect Hg 2+ ions, via a ratiometric approach. The LoD of Hg 2+ was calculated to be in the nanomolar concentration range. Fluorescence was significantly turned on as 44 broke down into two separate molecules, one of which, the coumarin aldehyde is highly fluorescent. It was reasoned that 44 could be used to monitor Hg 2+ ions in the human breast adenocarcinoma cell (MCF-7) line, which was observed when the cells were incubated with Hg 2+ ions.
Chemosensors 2019, 7, x 37 of 50 ratiometric approach. The LoD of Hg 2+ was calculated to be in the nanomolar concentration range.
Fluorescence was significantly turned on as 44 broke down into two separate molecules, one of which, the coumarin aldehyde is highly fluorescent. It was reasoned that 44 could be used to monitor Hg 2+ ions in the human breast adenocarcinoma cell (MCF-7) line, which was observed when the cells were incubated with Hg 2+ ions.

Multivariate Analysis
Molecular recognition between low molecular weight fluorescent probes (LMFPs) and metal cations has dominated the field for many years. Sensor arrays, using statistical analysis and patternrecognition algorithms are now becoming the main approach to measure and monitor a number of analytes in a single experiment, and a recent review written by Jolliffe, highlights some recent examples in fluorescent sensing arrays [216]. Here, we give a couple of recent examples on this new sensing approach, to highlight the direction of research in the field.
Minami et al. used commercially available dyes (alizarin red, bromopyrogallol red, and pyrogallol red) that contained the catechol moiety [217]. The catechol functional groups reacted with 3-nitrophenylbornoic acid (3-NPBA) to form a dynamic covalent bond. The metal ion competed with the boronic acid for the catechol moiety; thus, upon coordination of the metal ion, 3-NPBA was displaced. The metal-dye complex generated different colors, depending on the dye and the metal, Scheme 18.

Multivariate Analysis
Molecular recognition between low molecular weight fluorescent probes (LMFPs) and metal cations has dominated the field for many years. Sensor arrays, using statistical analysis and pattern-recognition algorithms are now becoming the main approach to measure and monitor a number of analytes in a single experiment, and a recent review written by Jolliffe, highlights some recent examples in fluorescent sensing arrays [216]. Here, we give a couple of recent examples on this new sensing approach, to highlight the direction of research in the field.
Minami et al. used commercially available dyes (alizarin red, bromopyrogallol red, and pyrogallol red) that contained the catechol moiety [217]. The catechol functional groups reacted with 3-nitrophenylbornoic acid (3-NPBA) to form a dynamic covalent bond. The metal ion competed with the boronic acid for the catechol moiety; thus, upon coordination of the metal ion, 3-NPBA was displaced. The metal-dye complex generated different colors, depending on the dye and the metal, Scheme 18.
The coordination of the catechol dyes and 3-NPBA resulted in blue shifts in the absorption spectra, and the addition of metal ions led to a red shift, due to the displacement of 3-NPBA and the formation of a dye-metal complex. FAB-MS and titration isotherms indicated that both the dye:3-NPBA and dye:metal complexes had a 1:1 stoichiometry. Extensive univariant work gave an understanding of the interaction and the K assoc for selected metal ions (Cu 2+ , Ni 2+ and Zn 2+ ), which were in the range of 5 × 10 2 M −1 to > 10 6 M −1 . The array utilized a 384 well microplate for high throughput screening. The response of the sensor array in the presence of 11 metals was investigated (50 mM HEPES buffer solution containing 10 mM NaCl interferant, 40 µM catechol dye, 4 mM 3-NPBA, and 10 µM metals) and was capable of discriminating between Ni 2+ , Cu 2+ , Zn 2+ , Cd 2+ , Co 2+ , Fe 2+ , Hg 2+ , Al 3+ , Pb 2+ , Ga 3+ , and Ca 2+ . LDA analysis indicated that 95.2% of variance was sufficient to discriminate between the different metals, and the cross-validation method indicated a 100% correct classification within each cluster, Figure 17A. A potential semi-quantitative analysis to determine the concentration of the metals was demonstrated using a mixture of Cu 2+ , Zn 2+ , and Ni 2+ , and a regression analysis using a support vector machine (SVM) algorithm ( Figure 17B). The results of the semi-quantitative assay, where the three different metals were tested at a concentration range of 0-3.7 ppm with 20 repetitions at each concentration, showed that a plot of the actual vs. predicted concentrations was highly accurate. The concentration range over which the results were obtained (Ni 2+ < 30 ppm, Cu 2+ < 2.5 ppm, Zn 2+ < 6 ppm), the method's potential for high-throughput screening, and the fact that the assay gave no response in the presence of other metals common in environmental water sources, such as K + , Na + , Mg 2+ , indicated that the assay developed by Sasaki and coworkers could be a useful tool for real time testing, in environmental water sources.

Multivariate Analysis
Molecular recognition between low molecular weight fluorescent probes (LMFPs) and metal cations has dominated the field for many years. Sensor arrays, using statistical analysis and patternrecognition algorithms are now becoming the main approach to measure and monitor a number of analytes in a single experiment, and a recent review written by Jolliffe, highlights some recent examples in fluorescent sensing arrays [216]. Here, we give a couple of recent examples on this new sensing approach, to highlight the direction of research in the field.
Minami et al. used commercially available dyes (alizarin red, bromopyrogallol red, and pyrogallol red) that contained the catechol moiety [217]. The catechol functional groups reacted with 3-nitrophenylbornoic acid (3-NPBA) to form a dynamic covalent bond. The metal ion competed with the boronic acid for the catechol moiety; thus, upon coordination of the metal ion, 3-NPBA was displaced. The metal-dye complex generated different colors, depending on the dye and the metal, Scheme 18. Scheme 18. The displacement approach by which the dyes formed a dynamic bond with the boronic acid derivative which was then displaced upon the coordination of the metal ion.
The coordination of the catechol dyes and 3-NPBA resulted in blue shifts in the absorption spectra, and the addition of metal ions led to a red shift, due to the displacement of 3-NPBA and the Scheme 18. The displacement approach by which the dyes formed a dynamic bond with the boronic acid derivative which was then displaced upon the coordination of the metal ion. formation of a dye-metal complex. FAB-MS and titration isotherms indicated that both the dye:3-NPBA and dye:metal complexes had a 1:1 stoichiometry. Extensive univariant work gave an understanding of the interaction and the Kassoc for selected metal ions (Cu 2+ , Ni 2+ and Zn 2+ ), which were in the range of 5 × 10 2 M -1 to > 10 6 M -1 . The array utilized a 384 well microplate for high throughput screening. The response of the sensor array in the presence of 11 metals was investigated (50 mM HEPES buffer solution containing 10 mM NaCl interferant, 40 μM catechol dye, 4 mM 3-NPBA, and 10 μM metals) and was capable of discriminating between Ni 2+ , Cu 2+ , Zn 2+ , Cd 2+ , Co 2+ , Fe 2+ , Hg 2+ , Al 3+ , Pb 2+ , Ga 3+ , and Ca 2+ . LDA analysis indicated that 95.2% of variance was sufficient to discriminate between the different metals, and the cross-validation method indicated a 100% correct classification within each cluster, Figure 17A. A potential semi-quantitative analysis to determine the concentration of the metals was demonstrated using a mixture of Cu 2+ , Zn 2+ , and Ni 2+ , and a regression analysis using a support vector machine (SVM) algorithm ( Figure 17B). The results of the semiquantitative assay, where the three different metals were tested at a concentration range of 0-3.7 ppm with 20 repetitions at each concentration, showed that a plot of the actual vs. predicted concentrations was highly accurate. The concentration range over which the results were obtained (Ni 2+ < 30 ppm, Cu 2+ < 2.5 ppm, Zn 2+ < 6 ppm), the method's potential for high-throughput screening, and the fact that the assay gave no response in the presence of other metals common in environmental water sources, such as K + , Na + , Mg 2+ , indicated that the assay developed by Sasaki and coworkers could be a useful tool for real time testing, in environmental water sources. The cross-validation routine shows 100% correct classification LDA results of the semi-quantitative assay, toward three different metal ions (Cu 2+ ,Zn 2+ and Ni 2+ ); concentration range of 0-3.7 ppm.%. Adapted from [218] with permission from The Royal Society of Chemistry.
Our group has carried out multivariate pattern recognition approaches to the binding between coumarin-enamine probes (compounds 45a-e) and ten divalent metal ions [219]. The coordination of the metal ions with compounds 45b-e, through the nitrogen atom, showed optical responses in a univariant approach, but the data were limited in terms of discrimination. To capture this wealth of information and to be able to carry out the discrimination of metal cations as proposed, a multivariate approach was adopted. A multimode microwell plate reader was utilized, allowing for rapid and automated acquisition of multivariate data. The sensor array comprised compounds 45b to 45f. Compound 45a did not possess a binding site on the aromatic ring system, so it was used as a The cross-validation routine shows 100% correct classification LDA results of the semi-quantitative assay, toward three different metal ions (Cu 2+ , Zn 2+ and Ni 2+ ); concentration range of 0-3.7 ppm.%. Adapted from [218] with permission from The Royal Society of Chemistry.
Our group has carried out multivariate pattern recognition approaches to the binding between coumarin-enamine probes (compounds 45a-e) and ten divalent metal ions [219]. The coordination of the metal ions with compounds 45b-e, through the nitrogen atom, showed optical responses in a univariant approach, but the data were limited in terms of discrimination. To capture this wealth of information and to be able to carry out the discrimination of metal cations as proposed, a multivariate approach was adopted. A multimode microwell plate reader was utilized, allowing for rapid and automated acquisition of multivariate data. The sensor array comprised compounds 45b to 45f. Compound 45a did not possess a binding site on the aromatic ring system, so it was used as a reference. The absorbance response of this array of four sensors at 330, 380, 400, and 430 nm, and fluorescence intensity at 330/450 nm, 330/528 nm, 330/580 nm, 380/450 nm, 380/528 nm, and 380/450 nm (λex/λem) were determined concurrently. The sensing array was exposed to a panel of analytes comprising the chloride salts of-Ca 2+ , Cd 2+ , Co 2+ , Cu 2+ , Fe 2+ , Hg 2+ , Mg 2+ , Ni 2+ , Pb 2+ , and Zn 2+ , and all measurements were conducted in the same solvent (DMSO) as the univariate analysis for consistency. The sensor array was exposed to each metal multiple times, generating a cluster of 18 replicates for each analyte. This approach generated a very large multivariate data set (4 × sensors by 10 × analytes by 18 × replicates = 720 data points, each described by nine instrumental variables). Linear discriminant analysis (LDA) was used to organize the high-dimensionality data and to extract the most relevant information, ultimately providing the information necessary to classify the ten analytes considered.
The well-established LDA algorithm computed a linear combination of the original variables that maximized the separation between multiple analytes, while minimizing the separation between replicate measurements of the same analyte [218]. In doing so, the algorithm generated a new set of variables, called factors, which were returned in order of decreasing relative information content. Data reduction was then accomplished by retaining only the first two or three LDA factors and dropping the higher-order ones. This produced a transformed dataset, in which each point was associated with a pair or a triplet of numbers, respectively, typically referred to as the factor scores of those points. These factor scores could be used as coordinates in plotting the transformed points in either a two-or three-dimensional plot. The absorbance response of this array of four sensors at 330, 380, 400, and 430 nm, and fluorescence intensity at 330/450 nm, 330/528 nm, 330/580 nm, 380/450 nm, 380/528 nm, and 380/450 nm (λ ex /λ em ) were determined concurrently. The sensing array was exposed to a panel of analytes comprising the chloride salts of-Ca 2+ , Cd 2+ , Co 2+ , Cu 2+ , Fe 2+ , Hg 2+ , Mg 2+ , Ni 2+ , Pb 2+ , and Zn 2+ , and all measurements were conducted in the same solvent (DMSO) as the univariate analysis for consistency. The sensor array was exposed to each metal multiple times, generating a cluster of 18 replicates for each analyte. This approach generated a very large multivariate data set (4 × sensors by 10 × analytes by 18 × replicates = 720 data points, each described by nine instrumental variables). Linear discriminant analysis (LDA) was used to organize the high-dimensionality data and to extract the most relevant information, ultimately providing the information necessary to classify the ten analytes considered.
The well-established LDA algorithm computed a linear combination of the original variables that maximized the separation between multiple analytes, while minimizing the separation between replicate measurements of the same analyte [218]. In doing so, the algorithm generated a new set of variables, called factors, which were returned in order of decreasing relative information content. Data reduction was then accomplished by retaining only the first two or three LDA factors and dropping the higher-order ones. This produced a transformed dataset, in which each point was associated with a pair or a triplet of numbers, respectively, typically referred to as the factor scores of those points. These factor scores could be used as coordinates in plotting the transformed points in either a two-or three-dimensional plot.
It was shown that, the first two factors obtained from the LDA analysis, out of the ten present in the system, accounted for 80.7% of the total information content of the original system. A two-dimensional score plot (Figure 18) was obtained that showed clear clustering of the data-replicate measurements of samples of each divalent metal ion were classified as similar and grouped together by the LDA analysis. In short, the pattern-based approach was capable of differentiating almost all ten divalent metal cations, using the additional information collected through patterning and multivariate analysis. However, this analysis failed to definitively separate the ten metal ions, this was due to similarities in geometry that was adopted upon the binding of the molecular probe.
variables, called factors, which were returned in order of decreasing relative information content. Data reduction was then accomplished by retaining only the first two or three LDA factors and dropping the higher-order ones. This produced a transformed dataset, in which each point was associated with a pair or a triplet of numbers, respectively, typically referred to as the factor scores of those points. These factor scores could be used as coordinates in plotting the transformed points in either a two-or three-dimensional plot.

(A)
(B) Figure 18. (A) Two-dimensional LDA score plot for the analysis of probes 45a-e binding 10 divalent metal chlorides. (B) A three-dimensional LDA score plot captured 90% of all information available from the pattern recognition system, significantly improving on the separation of difficult cases that were imperfectly discriminated, using the two-dimensional results. Confidence ellipsoids were drawn at 95% confidence. Adapted from [218] with permission from The Royal Society of Chemistry. A three-dimensional LDA score plot captured 90% of all information available from the pattern recognition system, significantly improving on the separation of difficult cases that were imperfectly discriminated, using the two-dimensional results. Confidence ellipsoids were drawn at 95% confidence. Adapted from [218] with permission from The Royal Society of Chemistry.

Conclusions and Future Prospects
The chemical sensors that have been highlighted throughout this review showcase recent low molecular weight fluorescent probes designed to target the group 12 triad (Zn 2+ , Cd 2+ , and Hg 2+ ). During the last few years there has been a great deal of work on the design, synthesis, and detection of these metal ions. Much of the work presented here is focused on the traditional approach of sensor design, where the spectroscopic response is univariant in nature i.e., one LMFP targets one metal. As small molecule recognition becomes more popular, it is becoming more challenging to prepare novel compounds, but as computer power improves significantly, monitoring and discriminating metal ions in a complex environment seems to be the direction in which the sensor community is moving. Groups still publish their findings on LMFP and, whilst writing this review and taking a closer look at the data, it is clear that there are some general omissions or misleading statements. One of the biggest challenges is to work in 100% aqueous system, however, the systems are often in a mixture of organic-aqueous systems, and more often than not, only a fraction is water. Nevertheless, it has been claimed that a particular probe works in water. Additionally, there is often little or no discussion on the aqueous chemistry of the metal ion which influences the coordination. Moreover, many reports often omit a serious discussion on speciation, in their particular system, claiming that only a 1:1 probe-metal exists, when the data often suggest that other species might exist. More often than not, the counter-anion is omitted and authors make the assumption that it plays no role in binding, but only exists to balance the charge, when many counter ions can form bridging species. This is particularly important if a system is to work in an aqueous system, as anions might not exist in their conjugate base form, depending on the pH, which might influence the speciation. Additionally, care must be taken in reporting equilibrium constants using old linear regression methods, for example, the Benesi-Hilderbrand double reciprocal plot and Job's plot analysis. Even though they are easy to use they are significantly over used and misinterpreted. For example, subtle curvature in the Benesi-Hilderbrand plot is often mistaken as indicative of a 1:1 binding, but in fact it more often than not implies a more complicated equilibria occurring in the system. There are many problems associated with these regression calculations. Computer power has significantly improved over the years and software now exists for non-linear regression analysis to calculate more precise binding constants, including those involving complex equilibria.
Author Contributions: The authors comprise a graduate student (A.D.J.), an undergraduate student (R.M.C.) and the principle investigator (K.J.W.). Each student contributed significantly to the review. All three authors extensively carried out the literature search and wrote specific sections. A.D.J. wrote the section pertaining to the cadmium section, R.M.C. wrote the section on mercury, and K.J.W. put together the remaining sections and looked after the overall management of the manuscript.
Funding: This research received no external funding.