Crystallographic Aspects, Photophysical Properties, and Theoretical Survey of Tetrachlorometallates of Group 12 Metals [Zn(II), Cd(II), and Hg(II)] with a Triply Protonated 2,4,6-Tris(2-pyridyl)-1,3,5-triazine Ligand

Zn(II) (complex 1), Cd(II) (complex 2), and Hg(II) (complex 3) complexes have been synthesized using a triply protonated tptz (H3tptz3+) ligand and characterized mainly by single-crystal X-ray analysis. The general formula of all of the complexes is (H3tptz)3+·Cl–·[MCl4]2–·nH2O (where n = 1, 1.5, and 1.5 for complexes 1, 2, and 3, respectively). The crystallographic analysis reveals that the anion···π, anion···π+, and several hydrogen bonding interactions play a fundamental role in the stabilization of the self-assembled architectures that in turn help to enhance the dimensionality of all of the complexes. In addition, Hirshfeld surfaces and fingerprint plots have been deployed here to visualize the similarities and differences in hydrogen bonding interactions in 1–3, which are very important in forming supramolecular architectures. A density functional theory (DFT) study has been used to analyze and rationalize the supramolecular interactions by using molecular electrostatic potential (MEP) surfaces and combined QTAIM/NCI plots. Then, the device parameters for the complexes (1–3) have been thoroughly investigated by fabricating a Schottky barrier diode (SBD) on an indium tin oxide (ITO) substrate. It has been observed that the device made from complex 2 is superior to those from complexes 1 and 3, which has been explained in terms of band gaps, differences in the electronegativities of the central metal atoms, and the better supramolecular interactions involved. Finally, theoretical calculations have also been performed to analyze the experimental differences in band gaps as well as electrical conductivities observed for all of the complexes. Henceforth, the present work combined supramolecular, photophysical, and theoretical studies regarding group 12 metals in a single frame.


■ INTRODUCTION
Recent years have witnessed an immense growth in crystal engineering propelled by numerous noncovalent interactions. 1,2 The primary objective of the crystal engineering is to achieve the preferred structural topology that requires judicious choice of ligands and metal precursors. 3,4 Based on crystallographic data available in the structural database, new supramolecular synthons have been planned and synthesized accordingly. In general, either cationic or anionic part of the complex interacts with oppositely charged counterions by exploiting several noncovalent interactions including π···π, cation···π, anion···π, hydrogen bonding interactions, etc. 5−7 Second-sphere coordination binding occurs using noncovalent interactions of chemical species to the first-sphere coordination (covalently bonded) of a transition metal complex. In current research, supramolecular interactions directed by organic cations have arrested interest due to their unique structural features related to stacking interactions. 8,9 The second-sphere coordination strategy can form a binding site for anion recognition selectively and effectively. 10,11 In this regard, some organic−inorganic hybrid complexes can be prepared by the second-sphere coordination strategies involving organic cations and anionic metal complexes. 2,4,6-Tris(2-pyridyl)-1,3,5triazine (tptz) or its structural analogues have been well recognized as neutral binding motifs and can be used as primary as well as auxiliary ligands in contemporary metallosupramolecular chemistry. 12−16 Group 12 metal ions in the periodic table have a d 10 electronic configuration and do not exhibit any crystal field stabilization, and hence, their geometrical preference can be guided by steric factors mainly. As both "π" and "anion" are electron-rich species, their interactions have been unacknowledged for a long period. In 2002, Alkorta et al., Deyàet al., and Mascal et al. confirmed theoretically the presence of favorable noncovalent interactions between electronic-deficient aromatic rings and anions. 17−19 From the crystallographic database, it was found that only anion···π interaction was not able to ensure structural assembly, but an orchestrated interplay between either anion···π and π···π or anion···π and hydrogen bonding (classical, nonclassical, or both) interactions was energetically favorable to construct desired supramolecular architectures. 6−8 Metal−organic hybrids with essentially high conducting properties have grown to be an important research topic owing to their potential applications in electronic devices. Designing such flexible metal−organic hybrids of organic and inorganic building blocks into stable structural integrity plays a significant role in electron transportation, which in turn displays a wide spectrum of applications as a Schottky barrier diode (SBD) in electronics and optoelectronic devices. 20,21 An SBD is broadly used in photovoltaics, 22 batteries, 23,24 sensors, 25 supercapacitors, 26 and transistors. 27 Despite these advantages, application of most of the metal−organic hybrids is not up to the mark because of incompetent electron transportation (lower conductivity lesser than 10 −10 S/cm) between insulated organic ligands and non-redox metal ions. 28,29 Thus, exploring the role of metals in metal−organic hybrids with enhanced conductivity has become a trendy research topic for investigators. However, there are other device parameters like a short lifetime, high barrier potential and transit time, low effective mobility and carrier concentration, etc., that must be confirmed before fabricating the final device. The flexibility of devices depends on the assortment of metal ions and the binding functionality of the ligands. Therefore, before preparing metal−organic hybrids, one has to understand the influence of weak interactions in metal− organic frameworks (MOFs) for easy electron tunneling. Even weak interactions can increase electrical conductivity by several orders (by electronic delocalization), and linkers with electrondonating or -withdrawing groups convey more interactions for making electronically conductive arrangements. 30 By precisely controlling these supramolecular interactions, a chemist can impart new functions such as hydrophobicity, high thermal stability, charge carrier density, and charge mobility related to these materials. Li et al. and Dutta et al. successfully fabricated SBD devices with d 10 metal ions (Zn 2+ , Cd 2+ ) by using carboxylate linkers. 31,32 Considering the above facts, we report here the syntheses, Xray crystal structures, and supramolecular aspects of Zn (1), Cd (2), and Hg (3) complexes derived from a triply protonated 2,4,6-tris(2-pyridyl)-1,3,5-triazine ligand along with their photophysical properties.
In this present work, we have strategically planned to restrict the ligand "tptz" not to participate as a chelating ligand to form a metal complex but should act as an organic cation. In this regard, we have protonated all of the peripheral pyridine nitrogen (to restrict the neutral coordination in the first coordination sphere), maintaining the proper acidic pH of the reaction medium. Taking into consideration our previous works, 6,9 where the title ligand was monoprotonated (pH was adjusted to 4), herein, we regulated the pH to 0 (i.e., 1 M acidity) for further protonation of the ligand as our requirement. To stabilize this almost planar large-size cation (with more than one positive charge), our intention was to generate a large-size anion preferably a tetrahedral-like tetrahalometallate having more than one negative charge.
In this regard, we have chosen group 12 metal ions (Zn 2+ , Cd 2+ , and Hg 2+ ) that are able to form tetrachlorometallates [MCl 4 ] 2− (M = Zn, Cd, Hg) in the presence of dilute HCl (Scheme S1). Herein, the supramolecular interactions between the organic salt (tptzH 3 ) 3+ and the tetrachlorometallate having the general formula (H 3 tptz) 3+ ·Cl − ·[MCl 4 ] 2− ·nH 2 O are described. The geometry of the central metal in the complexes is distorted to be tetrahedral. Here, the tptz ligand is of our prime interest due to its large π-conjugation, and we want to utilize these π-based supramolecular interactions to execute extended architectures. Hence, in this communication, we have found hydrogen bonding and anion···π interactions in all these complexes. The interesting water−anion clusters have been explored for the title complexes (1−3) in this work. Structural analysis supported by Hirshfeld surfaces and fingerprint plots has also been displayed to gain additional insights into the hydrogen bonding interactions for all these complexes. The structure-directing role of anion−π/π + and hydrogen bonding interactions has been analyzed by DFT calculations. Under illumination conditions, the magnitudes of the photophysical properties of the complexes improve remarkably, though the improvement differs from complex to complex, which has also been analyzed theoretically. To our knowledge, it is the first report covering three group 12 metals that encompass both supramolecular and photophysical studies with proper theoretical justification.

■ EXPERIMENTAL SECTION
The details of starting materials and physical measurements have been incorporated in the Supporting Information.

Synthesis of [H 3 tptz]Cl·[HgCl 4 ]·1.5H 2 O (Complex 3).
Complex 3 was obtained via the same procedure as that for complex 1 using an aqueous solution of HgCl 2 (0.271 g, 1 mmol) instead of ZnCl 2 . The yellow crystals of 3 suitable for X-ray analysis were collected after one month with a yield of 67%. Anal. calcd for C 18 Figure  S2. All these complexes are isostructural and bear monoclinic structures with the P2 1 /n space group. All of the complexes contain a tetrachlorometallate MCl 4 2− (where M = Zn, Cd, and Hg) unit, a noncoordinated chloride ion, and a triply protonated organic moiety (H 3 tptz 3+ ) to balance the entire charge. The product stoichiometry is thus given by the overall charge balance with H 3 tptz 3+ /MCl 4 2− /Cl − = 1:1:1. The phase purity of all of the complexes has been confirmed by the powder X-ray diffraction (PXRD) pattern analysis ( Figure S3). All three complexes have been hydrated to different extents. Complex 1 bears one H 2 O molecule, whereas complexes 2 and 3 bear one and a half (1.5) H 2 O molecules per asymmetric unit. Thermogravimetric (TG) analysis was also carried out for all three complexes (1−3) to confirm the number of water molecules present in the complexes ( Figure S4). Using the water molecules, complex 1 forms a one-dimensional (1D) water−anion cluster, whereas complexes 2 and 3 form similar types of both 1D and two-dimensional (2D) water−anion clusters.
In complex 1, the noncoordinated water molecule acts as a double donor to Cl(1) and Cl(4) of two different ZnCl 4 2− ions at angles of 166 and 179°, respectively, to produce a 1D zigzag chain along the (010) direction ( Figure S5). In complex 2, two different noncoordinated water molecules are engaged with CdCl 4 2− ions in producing two different 1D water−anion clusters (Figure 1a,b) that in turn make an esthetically beautiful 2D extended architecture having an R 8 8 (24) ring motif in the "bc"-plane ( Figure 1c).
A similar type of water−anion cluster ( Figure 1) has also been observed for complex 3, where HgCl 4 2− ions are hydrogen-bonded with two different noncoordinated water molecules present in it ( Figure S6).
In our previous studies, 6,9 we were able to synthesize Ni(II) complexes with the same title ligand in its monoprotonated form (Htptz + ), and in all these complexes, the metal ion was directly coordinated to Htptz + and several supramolecular interactions are responsible to stabilize the structures. But numerous attempts to prepare single crystals (X-ray quality) with group 12 elements with the same protonated ligand in different solvents proved unsuccessful. We then strategically modified the synthetic protocol using further the protonated (here, triply) tptz ligand. Tetrahedral MCl 4 2− ions (where M = Zn, Cd, and Hg) have got some stabilities from a stereoelectronic point of view (as all M 2+ is the d 10 system and devoid of any crystal field stabilization energies). With this contention, we paved our work in further protonation of the tptz ligand in relatively higher acidic conditions (here, we used 1 M acidity) to enhance the cationic charge and forcefully convinced it to interact with tetrachlorometallate ions. In this regard, our aim was to identify the role of supramolecular interactions in stabilization of the crystal structure in the solid state. The asymmetric units of complexes consist of M 2+ cations (M = Zn, Cd and Hg) coordinated to four chlorido ligands in tetrahedral fashion with ∠ClMCl ranging from 103.44 (2) (Table S4). The local tetrahedral M 2+ tetrachloro coordination sphere and noncoordinated chloride ions have participated in hydrogen bonding interactions and helped to generate higher dimensionality of the crystal structure in the solid state.
In the triply protonated organic moiety ( Figure S7), the major coordination site is restricted due to two different intramolecular hydrogen bonding interactions (N1−H1···N2 and N6−H6···N2). The moderate coordination site is restricted through another intramolecular hydrogen bonding interaction (N5−H5···N4), but there are no such interactions in the minor coordination site.
On comprehensive analysis, it was exhibited that the 1D chain (shown in Figure 2) further extended its dimensionality (2D) through two different anion···π + interactions, where the noncoordinated chloride ion (Cl5) simultaneously interacted with Cg(1) and Cg (3) [where Cg(1) and Cg(3) are the centroids of N1C1C2C3C4C5 and N5C9C13C12C11C10 rings, respectively] of the triply protonated moiety (H 3 tptz 3+ ), as depicted in Figure S8. The same also occurred for complexes 2 and 3.
The water−anion clusters play an important role to stabilize various supramolecular architectures for all three complexes in the solid state. In complex 1, triply protonated organic moieties are almost parallel to each other with an intervening water−anion cluster ( Figure S5), which connects these organic moieties through strong N5−H5···O1 interactions (1.94 Å) to ensure a 1D tape-like architecture ( Figure S9) in the "ab"plane. Another interesting phenomenon is that complexes 2 and 3 also exhibit architectures (Figures S10 and S11) similar to that of complex 1 ( Figure S9), where triply protonated organic moieties are stitched with the water−anion cluster (as depicted in Figures 1a and S6a for complexes 2 and 3, respectively).
Besides several hydrogen bonding interactions, all four chloride ions attached with group 12 metal ions have engaged in anion···π and anion···π + interactions. Anion···π interactions are mediated by the middle triazine ring, whereas protonated peripheral pyridine rings of H 3 tptz 3+ are responsible for anion···π + interactions. When the aromatic π systems adopt positive charges, the charges are often observed to take part in anion···π + interactions to strengthen the binding ability than Inorganic Chemistry pubs.acs.org/IC Article that of conventional anion···π interactions. But, in this case, the anion···π interaction is stronger than the anion···π + interaction. It was evident that generally, the central triazine ring participates in the anion···π interaction if the triazine ring is attached to some electron-withdrawing groups. Here, in our case, the central triazine ring is attached to three protonated pyridine rings that in turn make the central triazine ring sufficiently electron-deficient and dictate to take part in the strong anion···π interaction.
Hirshfeld Surface Analysis. The Hirshfeld surfaces of the three complexes (1−3) have been mapped over d norm , d i , d e , shape index, curvedness, and fragment patches (shown in Figures S15−S17). From the shape index surface, one can get the information about each donor−acceptor pair, whereas the curvedness surface measures how much shape successfully splits the surfaces into a set of patches. The fragment patches correspond to the environment of the nearest neighbor on the surface depending on the proximity of the adjacent molecules in the form of color patches. The surfaces were enabled transparent to visualize the molecular structures in a similar Here, "hν" represents the photon energy, "α" is the absorption coefficient, and "n" is a coefficient for direct band-to-band transition having a value of 1/2. By plotting (αhν) 2 vs hν and extrapolating the linear line to the x-axis, the band gaps were calculated. The "E g " was calculated to be 3.87, 3.77, and 3.96 eV for complexes 1, 2, and 3, respectively. E g values suggest that only changing the metal ion has no such broad impact in altering the band structure of the complexes. The insets of the figures represent the ultraviolet−visible (UV−vis) absorption spectra of the complexes where no such distinct changes have been noticed for a particular complex.

Fabrication of Device and Electrical Properties.
To measure the electrical properties related to SBD parameters, current−voltage I−V measurement was executed. Here, two contacts were made, one was from the ITO substrate (bottom contact) and another was from the deposited part (top contact), where graphite was used as the contact metal (ITO/ complex/graphite). The details of thin-film fabrication from the precursor complexes are given in the Supporting Information. Experiments were performed with a highly sophisticated I−V analyzer (Keithley 4200) under dark and light conditions in the voltage range of ±1 V. Figure 6a−c shows I−V characteristics of complexes 1, 2, and 3, respectively. It is obvious that the I−V nature of complex 2 displays enhanced nonlinear rectifying performance than complexes 1 and 3. The nonlinearity of the I−V curve specifies that for all of the complexes (1−3), the conduction mechanism is non-ohmic, and rectifying nature indicates the Schottky barrier diode (SBD) characteristics of the devices. The conductivity of the complexes at room temperature under dark and light conditions has been measured. In the presence of light, the conductivity increased from 7.20 to 15.64 S/m for complex 1, 10.72 to 22.35 S/m for complex 2, and 7.89 to 13.5 S/m for complex 3. In all of the cases, conductivity was found to be increased in the presence of light and the higher conductivity of complex 2 than those of 1 and 3 indicates the significant photoresponse behavior of device 2. As the complexes show a wide band gap, thermionic emission theory could be applied to justify the electrical properties. 36 All of the I−V's and their relative parameters are quantitatively examined and authenticated by Cheung's equation. 37 where I 0 is the saturation current was calculated from the intercept of ln(I) at V = 0 and can be simplified as follows where the terms q, K, T, V, A, A*, and η stand for the electronic charge, Boltzmann constant, temperature (in kelvin), forward bias voltage, effective diode area, Richardson constant, and ideality factor, respectively. In this case, the effective diode area is maintained as 8.0 × 10 −2 cm 2 , and for all the SBDs, the Richardson constant is considered as 32 A K −2 cm −2 . Now, forward I−V characteristics can also be expressed by the following equation.
Here, IR s symbolizes a drop in voltage diagonally the series resistance and it is determined by the following equation. Here, complex 2 is comparatively ideal for SBD. Figure S25 represents the H(I) vs I plot for all devices under dark and light conditions. The H(I) vs I plot gives a straight line where the intercept along the y-axis becomes equal to ηϕ B , Inorganic Chemistry pubs.acs.org/IC Article and the slope gives R s . Now, by using the measured η values, potential barrier heights (Φ B ) for respective devices were calculated. All of the device-related parameters including potential height (ϕ B ), series resistance (R s ), and ideality factor (η) are listed in Tables S6 and S7.
To quantify a detailed study of charge transport phenomena at the interface of the MS junction, space-charge-limited current (SCLC) theory is introduced. Using this theorem, several SBD parameters like carrier mobility (μ eff ), carrier concentration (N), diffusion coefficient (D), diffusion length (L D ), and transient response time (τ) have been calculated. To acquire a significant outcome with respect to the carrier transport phenomenon at the interface of the MS junction, log(I) vs log(V) plots were drawn for the forward bias ( Figure  7). This plot is compared with the power law (I ∝ V m ) to locate the charge transfer zone in the MS junction. The slope value (m) fences the boundary in between ohmic and SCLC regions.
For the ohmic nature, the m value will be in the range of less than or equal to 1. Further greater than 1 or equal to 2 (m ≥ 2) indicates a space-charge-limited current (SCLC) region. 38 Figure 7a−c represents log(I) vs log(V) plots of devices 1, 2, and 3, respectively, under dark and illumination conditions. The log(I)−log(V) plot illustrates two distinct linear regions (I and II) of different slopes, hence concluding that two different conduction mechanisms are being operated in the MS junction. At low potential region I, current becomes proportional to the applied voltage (I ∝ V), and hence, the device exhibits an ohmic nature in this zone. Region I attributes thermionic emission where bulk-generated electrons are responsible to produce current than the injected free carriers from the metal. 39,40 At a higher potential, the log(I)−log(V) plot obeys the simple power law (I ∝ V m ), and it is separated as region II. In region II, the current becomes proportional to the square of the applied voltage (I ∝ V 2 ), specifying that the current is run by space-charge-limited current (SCLC) obtained from the discrete trapping level. 41,42 From the I vs V 2 graph ( Figure S26) and using the Mott− Gurney equation, 43 carrier mobility (μ eff ) was evaluated. where the terms "J", "ε 0 ", "ε r ", and "d" stand for the current density, permittivity of free space, relative dielectric constant of complexes, and film thickness. Thicknesses of the films were about 2 μm for all of the devices, which have been measured by a surface profilometer (Bruker Contour GT; noncontacting mode). Before this measurement, dielectric studies were made for all of the complexes. Dielectric constants for all of the complexes have been calculated from the plots of capacitance vs frequency ( Figure S27). Capacitance was measured by varying the frequency between 1 and 10 kHz, under a constant bias of 1 V. By using the conventional equation, 44 relative dielectric constants for complexes 1, 2, and 3 were measured and found 6.23, 2.27, and 4.13, respectively. The relatively lower dielectric constant of complex 2 is an indication of its potential use as an optoelectronic device to reduce power loss. Transit time (τ) of the charge carriers is another important parameter to investigate charge transport across the junction. From the slope of the forward I−V curve (Figure 6), τ was calculated using the following equation. 45

Inorganic Chemistry pubs.acs.org/IC Article
The details of these parameters are presented in Table S7. The higher effective mobility value of device 2 confirms fast charge transport of the material through the MS junction compared to 1 and 3. The mobility enhancement of the carrier due to light illumination indicates the good photoresponsivity of the device. The results demonstrate that the charge transport property of device 2 (made from complex 2) is better than those of devices 1 and 3. All of these SBD-related parameters imply the ultimate supremacy of complex 2. DFT Studies. Theoretical Study for Noncovalent Interactions. As detailed above, all complexes reported herein exhibit structure-directing anion−π and hydrogen bonding interactions between the protonated organic ligand, counterions, and water molecules. The DFT study is devoted to analyze and compare the interactions in complexes 1−3. First, we have computed the MEP surfaces of a neutral model of each complex that consists of tricationicH 3 tptz 3+ with the hydrogen-bonded Cl − and one anion−π bonded [MCl 4 ] 2− as counterions and the hydrogen-bonded water molecule. The geometries used for the MEP calculations were extracted directly from the crystallographic coordinates, since we are interested in evaluating the interactions, as they stand in the solid state. The MEP surfaces for these models of complexes 1−3 are represented in Figure 8, evidencing that the MEP maxima are located at the H-atoms of the water molecule, ranging from 69 to 75 kcal/mol.
The MEP values over the aromatic rings of the ligand are also large and positive, ranging from 30 to 56 kcal/mol, thus revealing the existence of an extended π−acidic surface. The MEP minima are located at the [MCl 4 ] 2− counterions, ranging from −75 to −84 kcal/mol. The MEP is also large and negative at the hydrogen-bonded Cl atoms, ranging from −49 to −52 kcal/mol. We have also analyzed the energies of the anion−π and anion−π + interactions using DFT calculations and characterized the interactions using a combination of the QTAIM and NCI plot analyses, since they are very convenient to reveal noncovalent interactions in real space. We have first analyzed the energies of the neutral assemblies used to construct the MEPs shown in Figure 8 with the Cl − anion (anion−π + ). That is, the energies summarized in Figure 9 were calculated as dimers, where one monomer is the chloride anion and the other one is the [(H 3 tptz)(Cl)(MCl 4 )(H 2 O)] assembly. The anion−π energies are very large and negative due to the large electrostatic attraction between the anion and the protonated pyridine ring. The stronger energy is found for complex 1 (ΔE 1 = −43.1 kcal/mol), while for complexes 2 and 3, it is almost identical, likely due to the similar orientation of the [MCl 4 ] 2− anion at the opposite side of the Cl···π interaction. The combined QTAIM/NCI plot analyses for the assemblies of [(H 3 tptz)(Cl)(MCl 4 )(H 2 O)] interacting with Cl − are represented in Figure 9, showing that each Cl − anion is connected to the aromatic ligand via two bond critical points (CPs, red spheres) and bond paths (dashed bonds), thus confirming the existence of the anion−π + interactions. The anion−π nature of the interaction is further confirmed by the size and shape of the reduced density gradient (RDG)  Inorganic Chemistry pubs.acs.org/IC Article isosurfaces, embracing the region between the π-system and the chlorine atom. We have also computed the energies of the hydrogen bonds by using the values of the potential energy density (V r ) at the bond CPs. It can be observed that in all complexes, the chloride is connected to the NH groups via bond CPs and bond paths and is further characterized by small blue isosurfaces. The water molecule is also connected to one NH bond via a bond CP, bond path, and blue isosurface. Moreover, in all of the complexes (1−3), a secondary CH···O (water) interaction is also established. The energies of the hydrogen bonds are indicated in red adjacent to the bond CPs. In complexes 2 and 3, the hydrogen bond energies are almost identical, with a total of 10.2 kcal/mol for both NH···Cl hydrogen bonds and 5.4 and 5.6 kcal/mol for the NH···O(water). For complex 1, the NH···Cl hydrogen bonds are slightly weaker (9.5 kcal/mol in total) and the NH···O(water) hydrogen bond is stronger (6.3 kcal/mol). The secondary CH···O(water) contacts are significantly weaker (0.6 kcal/mol for 1 and 0.9 kcal/mol for 2 and 3) in line with the small and green NCI plot isosurface.
We have also evaluated the anion−π energies for the assemblies of [(H 3 tptz)(Cl)(MCl 4 )(H 2 O)] with the [MCl 4 ] 2− anions, as detailed in Figure 10. The strongest interaction is observed again for complex 1 (ΔE 4 = −54.9 kcal/mol), while for complexes 2 and 3, the energies are ΔE 4 = −42.7 kcal/mol and ΔE 6 = −44.0 kcal/mol, respectively. This is likely due to the different binding modes observed in 1, where the two Cl atoms are pointing to the acidic π-surface of the ligand, while in complexes 2 and 3, only one Cl atom of the anion points to the central ring. In the latter complexes, only one bond CP and bond path connect the anion to the N atom of the ligand. However, the shape and extension of the NCI plot isosurface confirm the anion−π nature of this interaction, though the QTAIM analysis only reveals a single Cl···N contact. In the case of complex 1, two bond CPs and bond paths connect the anion to the ligand, one to the central triazine ring, and the other one to one protonated pyridine ring. The interaction energies obtained for the anion−π complexes involving the [MCl 4 ] 2− anions ( Figure 10) are stronger than those involving the Cl − anion (Figure 9) due to their monoanionic nature and consequently weaker electrostatic attraction.
Theoretical Study for Photophysical Properties. The solid-state structures of the 1, 2, and 3 crystals were modeled using TD-DFT methodology and choosing the experimental crystal lattices as a starting point for optimizing the atomic position. Standard band theory was used to determine the adjustment degree between the experimental and theoretical data. Figure 11 shows the band diagram between the crystals in  Inorganic Chemistry pubs.acs.org/IC Article the study. All systems present direct band gap behavior at point D of the Brillouin zone and around 3.50 eV. This is in reasonable agreement with the experimental values that range from 3.77 to 3.96 eV, thus evidencing that the TD-DFT method slightly underestimates the band gap values. Nevertheless, the theoretical calculations confirm the semiconductor type and low influence of the metal atom. Figure 12 shows the partial density of states for the three complexes. The plots represent the contribution of [MCl 4 ] 2− (M = Zn, Cd, Hg) and chloride anions and the tptz ligand separately. It is revealed that the p-character of the anions mainly dominates the valence band, and the p-character of the tptz ligands mainly dominates the conduction bands. Interestingly, the d-orbitals of the metal centers do not participate either in the conduction bands or in the valence bands, in line with the negligible effect of the group 12 element on the band gap.
This fact is also verified by calculating the frontier molecular orbitals in all of the systems that show the total absence of density at the [MCl 4 ] 2− anions in both frontier orbitals. Figure  13 shows the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) of complex 2 as representative crystal structures, since these are almost identical in the three complexes. The HOMO plot in Figure 13 shows the contribution of the p-orbital at the chloride anion, while the LUMO is a π-antibonding orbital localized at the tptz ligand. Theoretically, it is possible to know the frequency dependence of an incident photon in a material with the calculation of the dielectric function ε(ω) (see Figure  14). 47−52 We have chosen a photon energy range of 0−12 eV to calculate the optical response using the band structure. Figure 14 shows that complexes 1−3 have similar dielectric constant diagrams concerning the intensity of their maxima, in agreement with the similar experimental behavior of complexes 1−3. We have also measured the optical conductivity of the material (σ(ω)), since it is helpful to analyze how the conductivity of the material changes upon illumination. Moreover, the photoconductivity and hence electrical conductivity of materials increase because of absorbing photons. 47−52 As a rough approximation, it is possible to relate the optical conductivity of the material with the electrical conductivity measured experimentally, verifying whether any change in its electronic configurations occurs when the material absorbs a photon. 53−56 These facts can be related to the increased conductivity of these materials when illuminated. However, the results shown in Figure 14 do not explain the fact that complex 2 is superior to 1 and 3 in terms of optical properties. A likely explanation for this is likely related to the fabrication of the device like differences in the junctions rather than differences in the photophysical properties of the compounds.
Recently, a number of efforts have been explored to induce conductivity via through-bond, extended conjugation, and through-space involving weak interaction strategies. 57−59 In this work, the charge transportation occurs through the combination of these three strategies. The extended con-   3). Here, such striking differences in device properties among the complexes can majorly be attributed to two factors: (i) the increased electron density of the protonated organic moiety (H 3 tptz 3+ ) by partial electron transfer from MCl 4 2− and Cl − anions involved in anion−π/ anion−π + interactions and (ii) the possibility of effective charge transport through the strong hydrogen bonding interactions between the MCl 4 2− anions and noncoordinated water molecules.
The electrical conductivity depends not only on the ease of electron transportation through several intermolecular (covalent and noncovalent interactions) interactions but also on the band gap between the valence and conduction bands of the materials. As the electronegativity values of Zn and Cd (in Pauling scale 1.65 and 1.69, respectively) are more or less similar, it was expected that both complexes exhibit the same electrical conductance properties, but the experimental result reveals that complex 2 is somewhat higher conducting than complex 1. This may be due to the combined effect of a lower band gap (3.77 eV) and better electron hopping through the 2D water−anion cluster (that is absent in complex 1) of complex 2. In the case of complex 3, though it adopts a similar interaction network (including 2D water−anion cluster) as complex 2, it interestingly exhibits relatively lower electrical conductivity than other complexes. This observation can be explained by the higher band gap (3.96 eV) arising from the exceptionally higher electronegativity of Hg (2.00 in the Pauling scale because of both "f" contraction and relativistic effects) in complex 3.

■ CONCLUSIONS
In summary, a series of tetrachlorometallates including Zn(II) (complex 1), Cd(II) (complex 2), and Hg(II) (complex 3) have been synthesized with the help of a triply protonated tptz (H 3 tptz 3+ ) ligand and characterized by single-crystal X-ray analysis. The crystallographic analysis reveals that anion···π, anion···π + , and several hydrogen bonding interactions stabilize the self-assembled structures, which in turn are responsible for the expansion of dimensionality for all of the complexes. The Hirshfeld surfaces and the related 2D fingerprint plots of 1−3 structures facilitate a comparison of hydrogen bonding interactions significantly. Besides, we have been able to fabricate SBD devices with the d 10 metal series. The devices made from complexes 1, 2, and 3 exhibit non-ohmic rectifying nature under dark and illumination conditions. The measured device parameters, e.g., conductivity, effective mobility, transit time, carrier concentration, diffusion coefficient, diffusion length, and ideality factor, authenticate that complex 2 made with Cd 2+ is more ideal as an SBD than complex 1 and 3.
A DFT study has been used to disclose the interplay among these interactions and compare them toward the stability of the complexes. Furthermore, the interactions were also rationalized and analyzed by using MEP surfaces as well as by the combined QTAIM/NCI plots. The electrical properties of all of the complexes were analyzed theoretically. But the superiority of complex 2 as an SBD was critically explained by the band gap, differences in electronegativities of the central metal atoms, and better supramolecular interactions in it. We envision that the light-harvesting nature of these complexes could be exploited for optical as well as photovoltaic applications, which are the main focus of our ongoing investigations.

Accession Codes
CCDC 2223679−2223681 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by