(Tris{2-[(5-chloro-2-oxidobenzylidene-κO)amino-κN]ethyl}amine-κN)ytterbium(III): crystal structure and Hirshfeld surface analysis

The title compound features an amine-N-capped octahedral coordination geometry for YbIII defined by an N4O3 donor set. The packing features supramolecular layers sustained by C—H⋯O, C—H⋯π(aryl) and C—Cl⋯π(aryl) interactions.


Chemical context
Despite being less studied than transition metal complexes, the crystal chemistry of lanthanides is rich and diverse as their complexes can display various coordination numbers and geometries that are not readily predicted (Salehzadeh et al., 2010). In recent years, interest in the coordination chemistry of lanthanides has increased owing, for example, to the molecular dynamics exhibited by their complexes (Pedersen et al., 2014). Further, lanthanide-based luminescent compounds are potentially useful materials for the fabrication of organic lightemitting devices (OLEDs) due to their ability to exhibit sharp emission bands, high colour purity and long-lived emission states (Ahmed et al., 2016;Bü nzli et al., 2015). In the context of the present report, tripodal and tri-anionic heptadentate ligands, having tertiary-amine, three neutral imine and three phenolate donors, leading to a potential N 4 O 3 donor set, are of interest in coordination chemistry due to the cavity size they define and owing to the relative rigidity of the ligand. Being large and having seven potential donor atoms, these ligands are capable of coordinating lanthanides even if the atomic sizes of lanthanides are greater in comparison to their transition metal counterparts (Yang et al., 1995). Indeed, from the literature, several tripodal lanthanide complexes have been successfully characterized and described (Liu et al., 1992;Bernhardt et al., 2001;Kanesato et al., 2001aKanesato et al., ,b, 2004Hu et al., 2015). As part of an on-going study, the crystal and molecular structures of the title heptadentate Yb III complex (I) is described herein along with an analysis of its Hirshfeld surface. ISSN 2056-9890

Structural commentary
The molecular structure of (I) is shown in Fig. 1 and selected geometric parameters are collected in Table 1. The tris{[5chlorosalicylidene)amino]ethyl}amine) tri-anion coordinates in a heptadentate mode, utilizing the three phenolate-oxygen, three imine-nitrogen and tertiary amine-nitrogen atoms. The coordination geometry is based on a amine-N-capped octahedron with the amine-nitrogen atom capping the triangular face defined by the three imine-nitrogen atoms. Supporting this assignment are the observations that the Yb-O bond lengths are significantly shorter than the Yb-N(imine) bonds which, in turn, are shorter than the Yb-N(amine) bond length, Table 1. The dihedral angle between the phenolate-O 3 and imine-N 3 faces is 3.86 (6) , indicating a parallel disposition. There is a range in the Yb-O bond lengths, i.e. >0.02 Å , with the shortest Yb-O1 bond being trans to the most loosely bound imine-N4 atom, and the longest Yb-O2 bond being trans to most tightly held imine-N2 atom, Table 1. The sixmembered chelate rings adopt different conformations. Thus, the O1-chelate ring is essentially planar (r.m.s. deviation of the six fitted atoms = 0.0377 Å ), with the maximum deviation of 0.0672 (14) Å being for atom O1. The O2-chelate ring is considerably less planar (r.m.s. deviation = 0.0551 Å ) for the non-Yb atoms with the maximum deviation being 0.0850 (16) Å for the C12 atom, with the Yb atom, the flap atom in an envelope conformation, lying 0.762 (3) Å out of the least-squares plane defined by the remaining atoms. An intermediate geometry is found for the O3-chelate ring, also an envelope, with the Yb flap atom lying 0.524 (3) Å out of the plane of the remaining atoms (r.m.s. deviation = 0.0376 Å ). When viewed down the amine-N-Yb vector, the ethylene bridges are in the same orientation as are the benzene rings, and the organic residues splay outwards to define the shape of a capped cone.

Supramolecular features
In the absence of conventional hydrogen bonding, the packing in (I) is sustained by a range of weak interactions, Table 2. Aryl-C-HÁ Á ÁO and imine-C-HÁ Á ÁO interactions assemble molecules into supramolecular helical chains propagating along the b axis, Fig. 2a. These are reinforced by methylene-C-HÁ Á Á(aryl) contacts, also shown in Fig. 2a. Chains are connected into supramolecular layers in the bc plane by endon C-ClÁ Á Á(aryl) interactions, Fig. 2b. Layers stack along the a axis with no directional interactions between them, Fig. 2c (Spek, 2009).

Analysis of the Hirshfeld surfaces
The close contacts present in the crystal of (I) were studied by mapping the Hirshfeld surface over the d norm contact distances within the range of À0.18 to 1.65 Å through calculation of the internal (d i ) and external (d e ) distances of a particular Molecular structure of (I), showing the atom-labelling scheme and displacement ellipsoids at the 70% probability level.  Hydrogen-bond geometry (Å , ).
Hirshfeld surface point to its nearest nucleus (McKinnon et al., 2007;Spackman & Jayatilaka, 2009), while the two-dimensional fingerprint plots of the corresponding close contacts were produced through the plot of d e vs d i (Spackman & McKinnon, 2002). All analyses were performed using Crystal Explorer (Wolff et al., 2012). The distances involving hydrogen atoms were normalized to the standard neutron-diffraction bond lengths. The Hirshfeld surface analysis was performed on (I) in order to gain better understanding, on a quantitative basis, of the different close intermolecular contacts. The percentage contributions to the overall Hirshfeld surface are summarized in Table 3. Fig. 3a shows a butterfish-like two-dimensional fingerprint plot for (I). In general, the diffuse region at the top-right corner of the plot may indicate relatively low packing efficiency which could be due to the absence of hydrogen bonding. A detailed analysis of the decomposed fingerprint plots reveals that close contacts resulting from CÁ Á ÁH/HÁ Á ÁC as well as OÁ Á ÁH/HÁ Á ÁO contacts are evident. These constitute ca 26 and 5%, respectively, of the overall interactions in the crystal, with d e + d i distances of $2.54 Å and $2.43 Å , respectively, Fig. 3b and 3c. As seen from the images of Fig. 4, these interactions connect adjacent molecules in such a way that the molecules are arranged in a shape complementary array with an overall packing index of 69.6%. The end-on C-ClÁ Á Á(aryl) contacts appear as a focused area in the middle of the fingerprint plot delineated into ClÁ Á ÁC/ CÁ Á ÁCl contacts, Fig. 3d. Finally, the ClÁ Á ÁH/HÁ Á ÁCl inter-  actions appear as the third most dominant interaction, right after HÁ Á ÁH and CÁ Á ÁH/HÁ Á ÁC, with an overall contribution of ca 24% to the Hirshfeld surface, despite the fact that these interactions are considered weak with contact distances greater than the sum of van der Waals radii. However, as seen from Fig. 3e, these interactions are responsible for the appearance of the tails of the 'butterfish-shape'.

Database survey
The structure of the unsubstituted Yb III complex has been the subject of two independent determinations (Bernhardt et al., 2001;Kanesato et al., 2004). The molecule exhibits a very similar coordination geometry except that it adopts crystallographic threefold symmetry. The Yb-O, Yb-N(imine) and  Table 3 Percentage contribution of the different intermolecular contacts to the Hirshfeld surface of (I).    Yb-N(amine) bond lengths, taken from Kanesato et al. (2004), are 2.168 (5), 2.433 (6) and 2.700 (7) Å , respectively, and follow the same trends as in (I), Table 1, and indeed are equal within experimental error. A wider range of lanthanide (Ln) structures are available for comparison where the ligand is identical to that in (I), and each conforms to the conformation found in (I), approximating threefold symmetry. These fall into two crystal symmetries, i.e. P2 1 /n, for Ln = Sm (Kanesato et al., 2001a), Tb (Hu et al., 2015) and Er (Pedersen et al., 2014) [approximate reduced-cell parameters: a = 12.6 Å , b = 15.1 Å , c = 15.3 Å and = 110 ] and P2 1 /c, for Ln = Gd (Kanesato et al., 2001b) and Yb in (I) [approximate reduced-cell parameters: a = 10.1 Å , b = 13.2 Å , c = 21.3 Å and = 101 ], having no obvious trends across the series. Salient bond-length data are collated in Table 4 for the five structures. As expected from the influence of the lanthanide contraction across the series Ln = Sm, Gd, Tb, Er and Yb, there is a systematic reduction in the Ln-O and Ln-N(imine) bond lengths. The only anomalous parameter might be the length of the Gd-N(amine) bond, i.e. 2.737 (8) Å , the relatively high standard uncertainty value notwithstanding.

Synthesis and crystallization
The Schiff base ligand, tris{[(5-chlorosalicylidene)amino]eth-yl}amine (Kanesato et al., 2000;0.56 g, 1 mmol) and triethylamine (0.14 ml, 1.0 mmol) were taken in absolute ethanol (25 ml) and refluxed for 1 h. An ethanolic solution (15 ml) of ytterbium(III) chloride hexahydrate (Sigma-Aldrich; 0.39 g, 1 mmol) was added to the mixture which was refluxed for 2 h and filtered. The filtrate was evaporated until a precipitate was obtained. The precipitate was recrystallized from ethanol solution and the by-product, triethylammonium chloride, was removed through filtration. Yellow needles of the title complex suitable for X-ray crystallographic studies were obtained from the slow evaporation of the filtrate. Yield

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 5. The carbon-bound H-atoms were placed in calculated positions (C-H = 0.95-0.99 Å ) and were included in the refinement in the riding-model approximation, with U iso (H) set to 1.2U eq (C). Owing to poor agreement, one reflection, i.e. (7 0 4), was omitted from the final cycles of refinement. The maximum and minimum residual electron density peaks of 1.65 and 0.45 e Å À3 , respectively, were located 0.84 and 1.51 Å from the Yb and N2 atoms, respectively.

(Tris{2-[(5-chloro-2-oxidobenzylidene-κO)amino-κN]ethyl}amine-κN)ytterbium(III)
Crystal data Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.