Crystal structures of three mercury(II) complexes [HgCl2 L] where L is a bidentate chiral imine ligand

Three complexes synthesized by coordination of chiral imines to HgCl2 have been characterized, in which the tetrahedral HgII centre has a geometry strongly distorted towards the disphenoidal geometry.


Chemical context
The coordination geometry for Hg II is very versatile, in particular because the available coordination numbers for this 5d 10 metal ion cover a large range, from 2 (e.g. Moreno-Alcá ntar et al., 2013) to 10 (Williams et al., 2009). In the case of tetracoordinated Hg II complexes, the possible geometry extends from square planar, similar to d 8 transition metals, to tetrahedral, as for d 7 transition metals. Intermediate situations resulting from a distortion of the tetrahedral geometry are, however, the most common. The disphenoidal arrangement, also known as a seesaw geometry, is frequently observed in mononuclear Hg II complexes bearing non-sterically demanding ligands with significant -donating ability. This geometry, resulting from the formal distortion T d ! C 2v , may be regarded as derived from a trigonal bipyramid, with an unoccupied site in the equatorial plane (e.g. Bell et al., 1988;Wang et al., 2005). Much less frequently observed is the symmetry distortion T d ! C 3v , for which one axial site of the trigonal bipyramid is vacant (e.g. Adams et al., 1970).
Within this class of complexes, the coordination of the HgCl 2 molecule to a Schiff base is of interest, especially if the donor atoms from the ligand form a bite angle on the metal. Since this angle is generally less than 90 , a substantial distortion of the T d geometry is expected, which could modulate intermolecular interactions in the crystal.
We gained experience in the synthesis of such ligands via sustainable processes, using solvent-free one-pot reactions between a chiral amine and an aldehyde, providing that at least one reactant is liquid at room temperature. Three Schiff bases in this series, synthesized from 2-pyridinecarboxaldehyde, have been coordinated to HgCl 2 , and we now report the crystal structures of the resulting complexes. The main purpose of the X-ray characterization is to assess the consequence of the N-Hg-N bite angle on the coordination geometry. Moreover, the synthetic chemistry of Hg II compounds is still topical, mainly due to their potential applications as electroluminescent devices (Fan et al., 2009), sensors (Zhou et al., 2010), fluorescent lamps, batteries and preservatives in wood-pulp industry, etc. The interference of this metal in biological systems, mainly by targeting and eventually inactivating thio-containing enzymes, also requires a better understanding of its coordinative properties (Shettihalli & Gummadi, 2013).

Structural commentary
The first imine, L 1 , was obtained by condensation between 2-pyridinecarboxaldehyde and (S)-(À)-1-phenylethylamine, and coordination to HgCl 2 afforded complex (I), [HgCl 2 L 1 ]. The monoclinic unit cell contains four molecules per asymmetric unit (Fig. 1), each one displaying a slightly different conformation for the ligand. The imine bond is coplanar with the pyridine ring in all independent molecules, favoring the coordination of both N donors of L 1 to the metal. However, the phenyl ring has a degree of free rotation, generating four conformers: the observed dihedral angles between the pyridine and phenyl rings in complexes built on Hg1, Hg2, Hg3 and Hg4, are 71.1 (6), 78.0 (5), 82.3 (4) and 86.3 (6) , respectively. These angles thus span a quite broad range of ca 15 , which could account for the Z 0 = 4 character of the crystal.
Regarding the coordination geometry, the four complexes present an arrangement intermediate between tetrahedral and disphenoidal. The N-Hg-N bite angles formed by the Schiff base range from 69.7 (5) to 71.3 (5) , confirming the rigid character of this part of L 1 . In contrast, Cl-Hg-Cl angles are found in a larger range, from 116.0 (2) to 126.78 (17) ( Table 1). The coordination is however far from the idealized C 2v -disphenoidal or C 3v -trigonal pyramid arrangements.
Ligand L 2 was obtained using (S)-(À)-1-(4-methylphenyl)ethylamine for the Schiff condensation, and complex (II), [HgCl 2 L 2 ] crystallized in the triclinic system, with two independent molecules in the asymmetric unit (Fig. 2). Although the relative position of these molecules emulates a non-crystallographic inversion centre, the structure was refined in space group P1 on the basis of the chiral nature of (II). The correctness of this choice was confirmed by the refinement of the Flack parameter (see Refinement section). Geometric features related to the conformation for L 2 and to its The asymmetric unit for complex (I), with displacement ellipsoids at the 30% probability level. The labels for C and N atoms in molecules Hg2, Hg3 and Hg4 are as in molecule Hg1, but increased by 20, 40 and 60, respectively. Table 1 Comparison of key conformation parameters ( ) for compounds (I), (II) and (III). Compound Notes: (a) dihedral angle between aromatic rings in the ligand L; (b) N-Hg-N angle.

Figure 2
The asymmetric unit for complex (II), with displacement ellipsoids at the 30% probability level. coordination geometry are compiled in Table 1, for comparison purposes. As expected, only small differences between (I) and (II) are observed. The most significant difference is for the bent conformation of the ligand, since L 1 seems to be more flexible than L 2 . This difference could be sufficient to produce a symmetry reduction from P2 1 to P1, accompanied by the halving of independent conformers in the crystals, from Z 0 = 4 to Z 0 = 2. The third imine, L 3 , was obtained by condensation between 2-pyridinecarboxaldehyde and (1S,2S,3S,5R)-(+)-isopinocampheylamine. The complex formed upon coordination to HgCl 2 , (III), crystallizes with two molecules in the asymmetric unit ( Fig. 3), which have very similar conformations: the r.m.s.d. for a fit between the independent molecules is 0.47 Å (Macrae et al., 2008). As for (II), the independent molecules are related by a non-crystallographic inversion centre, at least until chiral centres are considered. The bite angle formed by L 3 is comparable to that formed by L 1 or L 2 (Table 1). However, in the case of (III), the Cl-Hg-Cl angles are larger and, as a consequence, the tetrahedral coordination geometry in that case is more distorted towards the C 2v -disphenoidal geometry, compared to (I) and (II). No robust correlations between N-Hg-N and Cl-Hg-Cl angles were found after mining the CSD for tetracoordinated Hg II complexes, making a rationalization on distortion trends in these complexes difficult to draw.
The point of interest is that in all cases, the dimeric species are formed through a non-crystallographic inversion centre, if chiral centres in ligands L 1-3 are ignored. Since the chiral nature of the complexes forces them to crystallize in a Sohncke space group, the stabilization of the crystal structures through the formation of such pseudo-centrosymmetric dimers is possible only if Z 0 > 1, as observed. On the other hand, it appears that the coordination geometry in the reported complexes is far enough from a disphenoidal geometry in order to promote dimerization. Indeed, the idealized C 2v -disphenoidal coordination would prevent the formation of the (-Cl) 2 bridge, since in that case the metalÁ Á Ámetal separation would become too short.

Database survey
The crystal structures of L 1-3 remain unknown, presumably because these compounds are obtained as oils at room temperature. However, L 1 has been widely used as a ligand for coordination chemistry. The current release of the CSD (Version 5.36 with all updates; Groom & Allen, 2014) reports complexes with numerous transition metals, for example Mn II , Zn II , Ni II , Co II and Co III (Howson et al., 2011), Cu II (Min et al., 2010), Pd II (Mishnev et al., 2000), and Rh III (Carmona et al., 1999). Nevertheless, no crystal structures have been deposited for Hg II complexes. An Hg II complex bearing a non-chiral Schiff base close to L 1 has been published (Kim & Kang, 2010). There are no structures including ligands L 2 or L 3 deposited in the CSD.

Biological activity
The antimicrobial activity of the complexes (I)-(III) was evaluated against Gram positive (Staphylococcus aureus) and Gram negative (E. coli, Pseudomonas aeruginosa) bacteria, and yeast (Candida albicans). All complexes were found to possess noteworthy antimicrobial activity (see supporting information). Among the compounds analyzed, (I) and (III) show high antimicrobial activity against all strains assessed. In general, all complexes tested displayed antifungal activity against the strains of C. albicans.

Synthesis and crystallization
Caution!! Any mercury compound poses potential health risks, and appropriate safety precautions along with disposal procedures must be taken in handling the complexes here reported. HgCl 2 sublimes to emit highly poisonous fumes, and must be handled only by trained persons, under appropriate conditions.
spectroscopic techniques (see supporting information) and were used without further purification. Synthesis of complexes. A solution of the chiral imine L 1-3 (0.35 mmol) in methanol (20 ml) was treated with HgCl 2 (0.1 g, 0.35 mmol) with stirring at room temperature for 1 h. The solid obtained was filtered out and dried in vacuo, and then dissolved in dichloromethane. The resulting solution was slowly evaporated in a non-controlled atmosphere, and after a few days, colourless crystals of complexes (I)-(III) were collected, with yields of 81, 75, and 77%, respectively. Spectroscopic data are available from the supporting information.

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 2. In the case of the triclinic crystal (II), the refined model contains a pseudo-inversion centre, at a confidence level of 95%. However, Wilson statistics, h|E 2 À 1|i = 0.726, point to the space group P1. This is confirmed by the optical activity measured for (II), and the convergence of the Flack parameter to the expected value. For (II), diffraction data for two crystals from different synthesis were collected, giving the same space group and final model. The best data set has been retained. However, due to strong correlations between parameters of p-tolyl groups in the independent molecules, these groups were restrained to have the same geometry, with effective standard deviations of 0.02 and 0.04 Å for the 1,2-and 1,3-distances, respectively (SAME command in SHELXL; Sheldrick, 2015). In all structures, H atoms were placed in idealized positions and refined in the riding approximation, with C-H distances constrained to 0.93 (aromatic CH), 0.96 (methyl CH 3 ), 0.97 (methylene CH 2 ) or 0.98 Å (methine CH). Isotropic displacement parameters for H atoms were calculated as U iso (H) = xU eq (carrier C), with x = 1.5 (methyl groups) or 1.2 (other H atoms).
[α] D 25 = +22.7 (c=1, CHCl 3 ). Biological activity of complexes: The antimicrobial activity of the Hg(II)-complexes (I-III) was evaluated against Gram positive (Staphylococcus aureus) and Gram negative (E. coli and Pseudomonas aeruginosa) bacteria and yeast (Candida albicans). The antimicrobial activity were assessed by measuring the Inhibitory zone diameters with the Disk Diffusion Test. We used disk of Amikacin 30 µg, Chloramphenicol 30 µg, Cefepime 30 µg and Fluconazole 25 µg (BD) used for in vitro susceptibility testing by the agar disk diffusion test procedure of bacterial and fungal pathogens as antimicrobial control (see Table at the end of this section). According to the results, all complexes were found to possess noteworthy antimicrobial activity. Among the compounds analyzed, (I) and (III) show high antimicrobial activity against all strains assessed, mainly Gram positive bacteria and fungi. In general all complexes tested displayed antifungal activity against the strains of Candida albicans.