A hexa-quinoline based C 3 -symmetric chemosensor for dual sensing of zinc(II) and PPi in an aqueous medium via chelation induced “OFF–ON–OFF” emission

A quinoline-based C 3 -symmetric fluorescent probe ( 1 ), N , N ′, N ′′-((2,4,6-trimethylbenzene-1,3,5-triyl)tris(methylene))tris(1-(quinolin-2-yl)- N -(quinolin-2-ylmethyl)methanamine), has been developed which can selectively detect Zn 2+ without the interference of Cd 2+ via significant enhancement in emission intensity (fluorescence “turn-ON”) associated with distinct fluorescence colour changes and very low detection limits (35.60 × 10 −9 M in acetonitrile and 29.45 × 10 −8 M in 50% aqueous buffer (10 mM HEPES, pH = 7.4) acetonitrile media). Importantly, this sensor is operative with a broad pH window (pH 4–10). The sensing phenomenon has been duly studied through UV-vis, steady-state, and time-resolved fluorescence spectroscopic methods indicating 1 : 3 stoichiometric binding between 1 and Zn 2+ which is further corroborated by 1 H NMR studies. Density functional theoretical (DFT) calculations provide the optimized molecular geometry and properties of the zinc complex, 1 [Zn(ClO 4 )] 33+


Designing aspect of 1
Being ions of similar sizes (radius of Zn 2+ is only 21 pm shorter than that of Cd 2+ ) 30 it is quite difficult to discriminate Zn 2+ and Cd 2+ using some chelating fluorophoric systems.
Thus the choice of the platform, chelating unit and the sensing system should be cleverly designed to obtain selectivity for a particular metal analyte over its common interferrants.
In this context, C 3 -symmetric ligands, 52a-c which have the potential to develop a suitable receptor to meet the required coordination environment of metal ions, are of real importance.52d Our group also reported a selective Zn(II) sensor where two quinoline units were used to chelate with Zn 2+ . 45Keeping this in mind, an easily synthesizable ligand, 1 (Chart 1), has been designed where six quinoline moieties have been incorporated into an arene platform to enhance the aqueous solubility of the sensor molecule as well as to create a sterically crowded environment, expecting that it could provide better fitting toward Zn 2+ than the closely related analyte Cd 2+ .Furthermore, the sensor with six flexible side arms is expected to have a non-luminescent nature due to better non-radiative decay.However, the coordination of multiple metal centres with the multiple side arms is expected to rigidify the overall system effectively, which might result in a drastic enhancement in the fluorescence output.[54][55][56][57][58][59][60][61][62][63][64][65][66][67][68][69] Chart 1 Chemical structure of the ligand 1.

Zn 2+ sensing studies in acetonitrile
A detailed study about the photophysical properties of 1 is conducted using absorbance (UV-visible), photoluminescence (PL), and time-resolved spectroscopic methods.A prominent signature of the quinoline unit is observed in the absorption spectrum of 1 which contains absorption bands at ∼210 (ε = 111 420 M −1 cm −1 ), ∼240 (ε = 116 552 M −1 cm −1 ), ∼303 (ε = 26 854 M −1 cm −1 ) and ∼315 (ε = 23 287 M −1 cm −1 ) nm in acetonitrile at 25 °C (Fig. S9a, ESI †).Among these, the bands at 210 and 240 nm might originate from the intra-ligand (IL) π-π* electronic transitions while the bands at 303 and 315 nm may arise from the n-π* transitions i.e. the electronic transitions from nonbonding orbitals on the nitrogen atoms (amine and quinoline) to ligand π* orbitals. 70,71Initially the ligand is very weakly emissive; illumination at the 315 nm wavelength results in a broad centered weak emission band at 430 nm resulting in a light green photoluminescence (Fig. S9a, ESI †).The weak emission may be the result of inner quenching of fluorescence due to the contribution of non-bonding electrons.In quinoline-based molecules like 1, the n-π* singlet and π-π* singlet transitions are energetically not very far.Hence, though the π-π* singlet transition occurred directly upon excitation, it might have a large propensity to transfer to the n-π* singlet.And once the π-π* singlet → nπ* singlet transfer takes place, it is consequently followed by an inter-system crossing (ISC) from n-π* singlet → n-π* triplet with a considerable ease. 70,71Thus, the n-π* triplet state makes these molecules very weakly luminescent.To check whether the ligand is selective toward any metal ion, metal binding properties of 1 are explored in acetonitrile.The changes in its spectral behavior in the presence of different metal ions are studied using the acetonitrile solutions of their corresponding perchlorate salts.The addition of 10 equiv. of various metal ions except Zn 2+ to 6.5 × 10 −5 M solution of 1 in acetonitrile does not result in any remarkable changes in the absorption spectrum, whereas, in the presence of Zn 2+ , an increase in the absorbance values at 303 and 315 nm is observed with the formation of a clear isosbestic point at 290 nm (Fig. S9b, ESI †).
A huge change is noticed in the emission profile of the ligand (8.5 × 10 −6 M in acetonitrile) upon the addition of 10 equiv. of Zn 2+ (Fig. 1).The very weak broad centered emission band of the ligand is changed to a sharp peak with 63-fold enhancement in emission intensity.The emission maximum is blue-shifted to 400 nm associated with a fluorescence color change from light green to blue (Fig. 1).However, other metal ions (10 equiv.)do not result in any significant change in the emission spectrum.The increase in fluorescence intensity in the presence of zinc may be the outcome of the reduced flexibility of the six quinoline arms upon chelation with Zn 2+ which tends to increase the fluorescence quantum efficiency.Coordination with a metal, on the other hand, stabilizes the non-bonding electrons over nitrogens and as a result the level of the n-π* singlet state might have increased remarkably. 70This in turn could affect the π-π* singlet → nπ* singlet transition (which occurs easily in the free ligand making the molecule very weakly emissive) to occur and thus the molecule could emit more easily from the π-π* singlet state (i.e.highly fluorescent).Besides changing the absorption and emission spectral behavior, the addition of Zn 2+ also brings about an enhancement in the quantum yield value of 1.In the case of a free ligand, the value is calculated as 0.0041 considering anthracene as the standard, 71 while it increases to 0.25 upon the addition of Zn 2+ (Fig. S10, ESI †).The initial very small quantum yield value could easily be correlated with the flexible structure of the ligand which allows the non-radiative decay through rotational and vibrational pathways and hence lowers the emission quantum yield.Coordination with Zn 2+ via quinoline side arms, as mentioned previously, decreases this probability and thus might be responsible for a higher quantum yield in the adduct.
To understand the Zn 2+ binding properties of 1 better, absorption and emission titration experiments are carried out.Incremental addition of Zn 2+ in 6.5 × 10 −5 M solution of 1 in acetonitrile promotes an increase in the absorbance at 303 and 315 nm with the formation of a clear isosbestic point at 290 nm (Fig. 2a).In PL titration, the intensity of the emission band gradually increased in the presence of an increasing amount of Zn 2+ , whereas the broad emission band at 430 nm blue-shifts to give a sharp peak at 400 nm (Fig. 2b).In both the cases (UV-vis and PL titration), the spectral changes are ceased after the addition of 3 equiv.of Zn 2+ indicating 1 : 3 stoichiometric binding between 1 and Zn 2+ .This binding stoichiometry is further confirmed by Job plot analysis which shows an inflection point at 0.33 (Fig. S11a, ESI †).The association constant for Zn 2+ binding by 1 is calculated as K Zn1 = 1.23 × 10 5 , K Zn2 = 9.19 × 10 4 , and K Zn3 = 6.46 × 10 4 M −1 according to the procedure reported in the literature (Fig. S11b, ESI †), whereas the lower limit of the detection of Zn 2+ by 1 is found to be 35.60 × 10 −9 M (Fig. 3a) using the calibration curve of change in emission intensity (I − I 0 ) versus the concentration of Zn 2+ .Thus, the sensitivity of 1 appears to be better than that of many of the reported sensors as enlisted in Table S2, ESI.† To check whether the ligand is selective for Zn 2+ a selectivity study is carried out.The addition of Zn 2+ in the presence of an excess amount of other metal ions like Mn 2+ , Mg 2+ , Cr 3+ , Cu 2+ , Cd 2+ , Ag + , Hg 2+ , Al 3+ , Pb 2+ , Fe 2+ , Ni 2+ and Co 2+ (10 equiv.each) results in the same emission enhancement as observed in the case of only Zn 2+ (Fig. 3b).
Thus, the high association constant values of Zn 2+ binding by 1 as well as its low limit of detection for Zn 2+ makes 1 a selective luminescent sensor for Zn 2+ even in the presence of a large excess of other competitive metal ions in acetonitrile.

Time-resolved spectroscopic study
The inference drawn from steady-state spectroscopic experiments is further supported by the results of the fluorescence lifetime study.The lifetime (τ) is calculated for free ligand (1) and the ligand in presence of various metal ions using time-correlated singlephoton count (TCSPC) experiment (Table S3 and Fig. S12, ESI †).The decay pattern of 1 is shown in Fig. 5a where the lifetime (τ) is found to be 1.48 ns.The decay pattern as well as the lifetime remains unaffected in the presence of various metal ions except Zn 2+ .
However, the addition of Zn 2+ brings a noticeable change in the fluorescence decay pattern as demonstrated in Fig. 4a and b which is associated with an increase in the lifetime value to 3.24 ns.Such a small increase in the τ value could be justified from the change in the values of two parameters, the emissive rate of the fluorophore (Γ) and its rate of non-radiative decay to S 0 (k nr ), which in this case are acting in opposite directions. 72,73The first one i.e. the radiative decay rate of 1 is increased upon Zn 2+ coordination as explained before and hence, leads to a lowering of the fluorescence lifetime.Chelation with Zn 2+ , on the other hand, suppresses non-radiative decay pathways by rigidifying the flexible side arms which in turn results in an increase in the τ value of the system.These two opposing factors try to compensate each other and as a result the τ value could not be changed too much.However, the net increase in the lifetime value indicates that the second factor is dominating over the first one in this case.
The observed bi-exponential decay in the presence of Zn 2+ might be the outcome of the co-existence of two different species: one is the free ligand (1) with a shorter lifetime and the other is the Zn 2+ adduct of 1 with a comparatively longer lifetime.Upon incremental addition of Zn 2+ the contribution from the latter is increased and as a result a gradual increase in the lifetime is observed which is stopped after the addition of 3 equiv.of Zn 2+ (Fig. 4b).This again supports the 1 : 3 binding stoichiometry between the host and guest as concluded from the steady state photophysical studies.

H nuclear magnetic resonance spectroscopy and electrospray ionization mass spectrometry
From the previous section it is obvious that ligand 1 selectively binds the Zn 2+ ion with 1 : 3 host-guest stoichiometry but the mode of interaction remains unrevealed.To know about this, a 1 H NMR titration experiment is carried out for 1 with Zn 2+ in CD 3 CN (Fig. .This approximated structure which is based on the 1 H NMR and ESI-MS studies agrees with all the solution state experimental outcomes as well as the previous reports. 74According to this, a plausible mechanism is demonstrated in Scheme 2. Density functional theoretical (DFT) studies are performed to establish the proposed structure of the Zn 2+ complex.

Theoretical calculations for 1 and its Zn 2+ complex
To investigate the stability and the structure-property relationship of the proposed zinc complex, DFT study is carried out on ground (S 0 ) and excited states (S 1 ) of both 1 and 1[Zn(ClO 4 )] 3

3+
. The hybrid B3LYP functional 75 is used in all cases as integrated in the Gaussian 09 package, 76 mixing the exact Hartree-Fock-type exchange with the Becke's exchange functional 77 and that proposed by Lee-Yang-Parr for the correlation contribution. 78The 6-31G(d) 79 basis set is used for C, N and H atoms and Zn is treated with LanL2DZ.The integral equation formalism variant of a polarizable continuum model (IEF-PCM) 80 is used to address the effect of acetonitrile (Fig. 5, Fig. indicating that the addition of Zn 2+ does not lead to a remarkable shift in the absorption spectrum and only an increase in absorbance is observed in the presence of Zn 2+ .The theoretically determined UV-vis spectra of the ligand as well as the proposed zinc complex match nicely with their corresponding absorption spectra which are observed in acetonitrile (Fig. S17 and Table S5, ESI †).

Zn 2+ sensing by 1 in an aqueous medium
To check whether the ligand is capable to detect Zn 2+ in an aqueous environment absorption and emission spectroscopic studies of 1 are carried out in an aqueous buffer (10 mM HEPES, pH 7.4)/CH 3 CN (1 : 1 v/v) solvent mixture with various metal ions (e.g., Mn 2+ , Mg 2+ , Cr 3+ , Cu 2+ , Cd 2+ , Ag + , Hg 2+ , Al 3+ , Pb 2+ , Fe 2+ , Ni 2+ , Co 2+ and Zn 2+ ).Two sharp absorption bands are observed at 303 nm (ε = 8580 M −1 cm −1 ) and 315 nm (ε = 8411 M −1 cm −1 ) in the UV-vis spectrum of the ligand, while the excitation at 315 nm results in an emission band at 430 nm with a small peak at 410 nm (Fig. 6a).Interestingly, the emission intensity of 1 in aqueous buffer (10 mM HEPES, pH = 7.4) media is greater than the value observed in pure acetonitrile.The reason might be the hydrogen bonding interaction between quinoline nitrogens and water which decreases the electron density from nitrogens and consequently reduces the effect of the n-π* transition which was responsible for weakening the emission intensity in acetonitrile. 70,71g. 6  The addition of an aqueous buffer solution of Zn 2+ (10 equiv.) to the ligand leads to a luminescence color change from green to blue while the emission intensity increases up to 4.76 fold (Fig. 6b).The emission maximum shifts to a comparatively higher energy region to give two new peaks at 384 nm and 405 nm.The increase in the emission intensity in the presence of Zn 2+ may be attributed to the rigidification of the side arms which leads an increase in the quantum efficiency.The presence of metal ions other than Zn 2+ does not affect the emission profile significantly.For better understanding the zinc binding properties of 1 in aqueous buffer (10 mM HEPES, pH = 7.4) media, absorption and emission titration experiments are performed by the gradual addition of Zn 2+ into an aqueous buffer (10 mM HEPES, pH 7.4)/acetonitrile (1 : 1 v/v) solution of 1. Upon the addition of an increasing amount of Zn 2+ , a gradual increase in the absorbance of the bands at 303 nm and 315 nm is observed with the appearance of a single isosbestic point at 280 nm (Fig. S18, ESI †).In an emission titration experiment, the addition of Zn 2+ results in a blue shift of the emission bands from 430 and 410 nm to 384 and 405 nm, respectively, with a steady increase in the emission intensity.The saturation point is reached at 3 equiv.Zn 2+ concentration which along with the corresponding Job plot analysis reveals 1 : 3 stoichiometric binding between 1 and Zn 2+ (Fig. 7a and b).The binding constant between 1 and Zn 2+ is found to be K Zn1 = 1.27 × 10 4 , K Zn2 = 9.55 ×

Effect of pH on Zn 2+ sensing by 1.
Since pH is one of the major factors governing the practical application of a sensor in the sensing of environmental as well as biological samples, the effect of pH upon the Zn 2+ sensing properties of 1 is duly investigated.For all the above mentioned sensing studies in an aqueous buffer (10 mM HEPES, pH = 7.4) solvent, the pH of the solution remains the same before and after the addition of Zn 2+ .This justifies the fact that the changes observed in the spectral properties of 1 upon the addition of Zn 2+ are exclusively due to its coordination with Zn 2+ but not for the effect of pH.Changing the pH from 4 to 10 does not lead to any significant change in the emission intensity value of the free ligand while the extent of emission enhancement in the presence of 3 equiv.Zn 2+ is reduced at a highly acidic or basic pH i.e. it stops beyond pH 4 on acidic side and pH 10 on basic side (Fig. 7d).The decrease in the sensing efficiency of 1 at highly acidic pH might be the result of protonation on quinoline nitrogens, which traps the nonbonding electron pairs on N, through which it could coordinate with the metal.Though the PET from nitrogen centers is stopped the occurrence of non-radiative decay is not inhibited because of the flexibility in the side arm in the absence of Zn 2+ .On the other hand, at a highly basic pH, a competition starts between the ligand and OH − ion for binding with Zn 2+ which lowers the sensitivity of 1 toward Zn 2+ . 18Thus, it can be concluded from the overall experimental results that 1 can be used as a selective fluorescent sensor for Zn 2+ in acetonitrile as well as in 50% aqueous buffer (10 mM HEPES, pH = 7.4) acetonitrile system.

Single crystal X-ray structural analysis of the Zn 2+ complex
To get the solid state structural evidence of Zn 2+ binding with 1, X-ray structural analysis is performed with the single crystals of the resulting zinc complex obtained from the slow evaporation of its DMF/methanol/water (3 : 1 : 1) solution.The complex crystallizes in the P2 1 space group where the asymmetric unit contains the ligand, three Zn 2+ ions, three ClO 4 − , three NO 3 − groups and the solvent molecules (Table S8, ESI †).
Interestingly, the 1 : 3 ligand-metal stoichiometry which has already been observed in detailed solution state studies is further assisted by the single crystal X-ray structure.As depicted in Fig. S21, † every Zn 2+ ion is chelated with 1via two quinoline nitrogens and one secondary nitrogen atom.One nitrate bridges two Zn 2+ centers leading to a different coordination number and geometry around them i.e. one is pentacoordinated while the other one is tetracoordinated.The remaining zinc centre binds with two oxygen atoms from the solvent molecules to adopt a pentacoordinated geometry.
Anion sensing by the trinuclear Zn 2+ complex.High aqueous solubility of the trinuclear Zn 2+ complex and presence of NO 3 − in the coordination sphere insists us to check whether it can be replaced by other anions.To see this, the change in the emission spectral behavior of the complex is monitored in the presence of various anions as their sodium salts in a 70% aqueous buffer (10 HEPES, pH = 7. , AMP and ADP does not cause any appreciable change in the emission spectrum of the complex while in the presence of 5 equiv. of PPi the emission intensity is decreased drastically (Fig. 8).Excess ATP (20 equiv.)also reduces the intensity but to a small extent.However, the quenching of emission intensity in the presence of PPi is not perturbed even in the presence of excess ATP in the medium.Careful analysis of the single crystal X-ray structure of the Zn 2+ complex, described above, reveals that two Zn 2+ centres are separated by 4.639 Å which matches very well with that of the length of a PPi anion.This analogy suggests the probable reason behind the selectivity toward PPi, which is, the best fitting of the guest between two Zn 2+ centres.For other structurally similar phosphate analogues this fitting does not occur efficiently resulting in a silent behavior of the trinuclear Zn 2+ complex towards these anions.To know the PPi binding better, the emission titration experiment of the trinuclear Zn 2+ complex with PPi is carried out.It shows a gradual decrease in emission intensity upon incremental addition of PPi (Fig. 9a) where the changes are ceased beyond the addition of 2 equiv. of the anion (Fig. 9b).The results along with the PL Job plot (Fig. S22a, ESI †) suggest 1 : 2 host-guest stoichiometry while the detection limit is calculated to be 45.37 × 10 −9 M (Fig. 9c) and the related association constants are determined by a nonlinear curve fitting method as 5.03 × 10 5 and 1.84 × 10 5 M −1 (Fig. S22b, ESI †).The values suggest a very high sensitivity as well as a significant binding efficiency of the trinuclear Zn 2+ complex toward PPi which is good enough when compared with those of previously reported PPi sensors (Table S10, ESI †).The selectivity study shows that the quenching of the fluorescence upon the addition of a PPi anion remains unperturbed even in the presence of other competitive anions (Fig. 9d).Based on the experimental outcomes, pieces of spectroscopic evidence (Fig. S23, ESI †), and previous literature reports, 19,24 a probable binding mode is demonstrated in Scheme S1, ESI, † where two PPi anions form an adduct with the trinuclear Zn 2+ complex.The NO 3 − anion which was bridged between two Zn 2+ ions is replaced by one PPi while the other PPi coordinates with the third Zn 2+ .Thus, as depicted in Scheme S1, ESI, † every Zn 2+ ion adopts fivecoordination mode.The quenching of the emission intensity of the trinuclear Zn 2+ complex in the presence of PPi could be attributed to the weakening of the N → Zn bond upon the coordination of the Zn 2+ ions with a PPi anion. 19,24This effectively increases the density of the non-bonding electrons over quinoline nitrogens, which, as discussed in the previous section, favors the ISC process from n-π* singlet → n-π* triplet and as a result the luminescence of the system is quenched.Thus, the decrease of the emission intensity upon the addition of a very small amount (only 2 equiv.) of PPi makes the trinuclear Zn 2+ complex a suitable "turn-OFF" fluorescence sensor for PPi in 70% aqueous buffer (10 mM HEPES, pH = 7.4) acetonitrile media.where P(t) is decay, i is the number of discrete emissive species, B is the baseline correction, α i is the pre-exponential factor, and τ i is the excited state lifetime associated with the i th component.In the case of multi-exponential decays the following equation was used to calculate an average lifetime: where a i is the contribution of the i th decay component, and a i = α i /Σα i .
X-ray crystallographic refinement details for 1 and Zn 2+ complex.In each case, a diffractable size crystal was collected from the mother liquor, dipped in paratone oil, and then it was cemented on the tip of a glass fiber using an epoxy resin.The intensity data of the crystals were collected using Mo Kα (λ = 0.7107 Å) radiation on a Bruker SMART APEX diffractometer, equipped with a CCD area detector at 100 K and 106 K for 1 and its tri-nuclear Zn 2+ complex, respectively.Data integration and reduction were processed by the SAINT 82a software.Empirical absorption correction to the collected reflections was done by applying SADABS.82b The structures were solved using SHELXTL 83 and was refined on F2 by the full-matrix least-squares technique using the SHELXL-97 84 program package.PLATON-97 85 and MERCURY 3.8 86 were used to generate graphics.Some of the carbon and nitrogen atoms of one quinoline ring in 1 and the trinuclear Zn 2+ complex are highly disordered.Some other disordered solvent molecules are removed using the PLATON/SQUEEZE program.The occupancy factors of the disordered atoms are refined using the FVAR command of the SHELXTL program and are isotropically refined.
Though good crystals have been selected and the data have been collected at 150 K, the crystals did not show diffraction beyond the theta max 20.39 and 20.81 for 1 and trinuclear Zn 2+ complex, respectively, even after several data collections.
CCDC 1583646 and 1583647 † contain the supplementary crystallographic data for this paper.were added to it.The reaction mixture was refluxed and stirred for 12 h.After that the remaining acetonitrile was evaporated and the resulting residue was extracted in dichloromethane (DCM).The DCM part was evaporated to get an ash-colored crude product which was further purified by column chromatography with silica gel of a 60-120 mesh size using chloroform/methanol as an eluent.

Fig. 4
Fig. 4 Time-resolved luminescence decays of 1 (22.5 × 10 −6 M; λ ex = 340 nm and λ em = 410 nm) (a) in the presence of various metal ions as their perchlorate salts and (b) upon the addition of an increasing amount of Zn 2+ in acetonitrile at room temperature.

Fig. 5
Fig. 5 Frontier molecular orbitals of 1 and 1[Zn(ClO 4 )] 3 3+ with their corresponding energy gaps as calculated from DFT B3LYP/6-31G(d) and the LanL2DZ mixed basis set.The 6-31G(d) basis set is used for H, C, N, O, and Cl atoms; Zn is treated with the LanL2DZ computational level using the IEFPCM model for acetonitrile [isovalue = 0.02].
2+  into 1 results in a gradual downfield shift of almost all quinoline protons as well as H b and H c , which are adjacent to the secondary N while the peak position corresponding to methyl protons (1.5 ppm) in the central arene ring (H a ) remains unaltered.However, the overall changes stopped after the addition of 3 equiv.of Zn 2+ .This leads us to conclude that the quinoline nitrogens as well as the linker N simultaneously act as the donor site for chelation with the Zn 2+ ion.Besides this, in the ESI-MS of the isolated Zn 2+ complex of 1, a peak is observed at 517.07 m/z which could be assigned for 1[Zn(ClO 4 )] 3 3+ with m/z 517.06 (Fig.S14, ESI †).The distribution patterns of the species matches well with the corresponding theoretically calculated distribution patterns.Based on these, a possible structure of the zinc complex is outlined where three zinc ions are bound with the ligand and each of them coordinates with three nitrogens and one ClO 4 −

S15-
S17 and Tables S4-S7, ESI †).The results reveal that, in acetonitrile the three side arms of the ligand are arranged in a propeller-like shape where three quinoline units are present on one side of the central arene ring while the other three are on the opposite side.As depicted in Fig.5, upon the addition of Zn 2+ the two quinoline rings in the same arm come closer to coordinate with Zn 2+ resulting in a distorted tetrahedral geometry around it.Every Zn 2+ ion is surrounded by two quinoline nitrogens, one secondary N and the O in a perchlorate anion.The corresponding bond lengths and bond angles are listed in TableS4, ESI.† From the energy level diagram shown in Fig.5, it is obvious that the chelation of the ligand with Zn 2+ does not induce a huge change in the value of energy difference between the HOMO and LUMO, instead a slight decrease in the same is observed.This agrees with the solution state photo-physical experimental finding Fig. 6 (a) Absorption and emission profiles of 1 and (b) Emission spectrum of 1 in the presence of 10 equiv.various metal ions in an aqueous buffer (10 mM HEPES, pH 7.4)/acetonitrile (1 : 1 v/v).

10 3
and K Zn3 = 6.89 × 10 3 M −1 (Fig.7c) and the calibration curve of change in emission intensity (I − I 0 ) versus the concentration of Zn 2+ results in the detection limit 29.45 × 10 −8 M (Fig.S19a, ESI †).The selectivity of the ligand toward Zn 2+ is also studied by monitoring the change in the emission spectrum of 1 upon the addition of 3 equiv.Zn 2+ in the presence of various metal ions (10 equiv.each) in an aqueous buffer (10 mM HEPES, pH 7.4)/acetonitrile (1 : 1 v/v) solution.The result shows almost the same emission spectral changes in each case as observed with only Zn 2+ in the absence of other metal ions (Fig.S19b, ESI †).However, no remarkable change is observed in the fluorescence decay profile of 1 upon the addition of an incremental concentration of Zn 2+ in its aqueous buffer (10 mM HEPES, pH 7.4)/acetonitrile (1 : 1 v/v) solution (Fig.S20, ESI †).This indirectly suggests that the extent of the increase in the lifetime by reducing the probability of non-radiative decay via rigidification of the flexible ligand upon zinc coordination might be just comparable to the effect of lowering the τ value due to the increase in the emissive rate of the fluorophore as explained before.Thus the two opposite factors completely compensate each other resulting in no change in the lifetime value.
Calculation of association constants.The 1 : 3 association constants were determined using a nonlinear least-squares analysis of I versus c M using the deduced equation: 81a where β 21 = K 11 K 21 , β 31 = K 11 K 21 K 31 , and [M] ≈ c M are the concentrations of Zn 2+ ions, I 0 or I is integrated emission in the absence or presence of Zn 2+ .Φ 1 is approximately 0.09, the quantum yield of the 1 : 1 1-Zn 2+ complex; Φ 2 is approximately 0.185, the quantum yield of the 1 : 2 1-Zn 2+ complex.The 1 : 2 association constants between the trinuclear Zn 2+ complex and PPi are determined by using the equation: 81b where [G] ≈ the concentration of the trinuclear Zn 2+ complex Φ 1 is approximately 0.185, the quantum yield of the 1 : 1 Zn 2+ complex-PPi species; Φ 2 is approximately 0.09, the quantum yield of the 1 : 2 Zn 2+ complex-PPi species.Calculation of detection limit.Detection limits (DL) were calculated using the following equation: DL = (3 × SD)/slope where SD corresponds to the standard deviation of the blank sample, measured using 15 consecutive scans of the blank sample.The slope is obtained from the linear fit plot of PL intensity changes versus the concentration of Zn 2+ added.The SD values of ligand 1 were 1000.46 in acetonitrile and 995.59 in an aqueous buffer (10 mM HEPES, pH 7.4)/acetonitrile (1 : 1 v/v).In the case of a trinuclear Zn 2+ complex the value was 930.99.Calculation of excited-state lifetimes.The following equation was used to analyze the time-resolved emission decays: