Crystal structure of bis(bis{μ3-3-methyl-3-[(4-nitro-2-oxidobenzylidene)amino]propane-1,3-diolato}tris[chlorido(dimethyl sulfoxide)iron(III)]) dimethyl sulfoxide heptasolvate dihydrate

The title compound is based on a trinuclear {Fe3(μ-O)4} core with an angular arrangement of the FeIII ions that can be explained by geometrical restrictions of two bulky ligands each coordinated to all the metal centres.

The title compound, [Fe 3 (C 11 H 11 N 2 O 5 ) 2 Cl 3 (C 2 H 6 OS) 3 ] 2 Á7C 2 H 6 OSÁ2H 2 O, was isolated accidentally from an Fe 0 -NiCl 2 Á6H 2 O-H 3 L-TEA-DMSO system [where H 3 L is the product of the condensation between p-nitrosalicylaldehyde and 2-amino-2-methylpropane-1,3-diol and dimethyl sulfoxide (DMSO), and TEA is triethylamine]. The structure is based on a trinuclear {Fe 3 (-O) 4 } core, with an angular arrangement of the Fe III ions that can be explained by the geometrical restrictions of two bulky ligands, each coordinating to all of the metal cations.

Chemical context
Almost 30% of GDP (gross domestic product) is generated through catalysis, which explains the ongoing interest in the development of compounds with potential as new efficient catalysts. Polynuclear associates have been found to be cofactors of many enzymes and catalysts for various processes (Buchwalter et al., 2015). In this work, we present the synthesis of a new trinuclear Fe III complex obtained accidentally while exploring the Fe 0 -NiCl 2 Á6H 2 O-H 3 L-TEA-DMSO system (TEA is triethylamine and DMSO is dimethyl sulfoxide). We did not investigate this complex for any catalytic activity, although it has a hypothetical practical interest because it was obtained in facile way from commercially abundant air-stable non-hazardous materials and consists of redoxactive metal atoms and ligands. The synthesis is based on the self-assembling paradigm, in particular on direct synthesis (Garnovskii et al., 1999); the metal ions and ligands are allowed to choose the most favourable charge and coordination modes and do not require specific synthetic manipulations and laboratory equipment. However, under these conditions we cannot predict the structure of the final molecule that will be obtained. Earlier, our group has shown the successful application of this approach for obtaining novel monometallic [either polynuclear, as in Babich & Kokozay (1997), or mixed valence, as in Kovbasyuk et al. (1997)], heterobimetallic [either polynuclear, as in Kovbasyuk et al. (1998), Vassilyeva et al. (1997) and Nikitina et al. (2008) or polymeric, as in Nesterova et al. (2004Nesterova et al. ( , 2005Nesterova et al. ( , 2008] and heterotrimetallic [as in Nesterov et al. (2011)] complexes.

Structural commentary
The molecular complex [Fe III 3 L 2 Cl 3 (DMSO) 3 ] 2 Á7DMSOÁ2H 2 O is based on a trinuclear {Fe 3 (-O) 4 } core with an angular ISSN 2056-9890 arrangement of the metal cations [the FeÁ Á ÁFeÁ Á ÁFe angle is 104.70 (4) ], linked pairwise by two -O bridges from the fully deprotonated Schiff base ligand (Fig. 1). The structure can also be viewed as a combination of two {Fe III L} blocks joined through a central Fe III ion via alkoxy bridges and completed by chloride ligands and solvent molecules (DMSO and water).
The {Fe(-O) 2 Fe} fragments are almost perpendicular [angle between planes = 96.4 (1) ]. Both Schiff base ligands reveal a 3.2211 coordination mode (Coxall et al., 2000).  (5) ]. It should be noted that the coordination environments of the terminal metal cations are not equivalent. The N atom occupies an axial position at atom Fe3, but an equatorial one at atom Fe1, assuming the chloride ligand is in an axial position in both polyhedra, due to an antiparallel arrangement of the two Schiff base ligands, which is also favourable for an intramolecular stacking interaction between the benzene rings [intercentroid distance = 4.034 (4) Å , plane-to-centroid distance = 3.505 (7) Å , centroid displacement = 2.00 (1) Å and angle between planes = 7.8 (2) ]. The weak intramolecular attractive interaction C23-H23CÁ Á ÁO12 (HÁ Á ÁO = 2.43 Å ) stabilizes the orientation of adjacent DMSO ligands.

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 2. All H atoms were placed in idealized positions (C-H = 0.95-0.99 Å and O-H = 0.87 Å ) and constrained to ride on their parent atoms, with U iso (H) = 1.5U eq (C,O) for water molecules and methyl groups, and 1.2U eq (C) otherwise. Two of the non-coordinating DMSO solvent molecules were disordered, each over two sites. The refined occupancy factors for the S6A/S6B disordered DMSO molecule converged to 0.745:0.255. For the S7 disordered molecule, the occupancy factors were fixed at 0.50:0.50 due to symmetry restrictions; two sites of this molecule are located in neighbouring asymmetric parts of the unit cells and are connected by the symmetry transformation (Àx + 1, Ày + 2, Crystal packing diagram showing the presence of supramolecular four-membered hydrogen-bonded rings aggregating two water molecules with two uncoordinated DMSO molecules. Hydrogen bonds are denoted with dashed lines and H atoms have been omited for clarity.  (Agilent, 2012); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL2014 (Sheldrick, 2015); molecular graphics: OLEX2 (Dolomanov et al., 2009); software used to prepare material for publication: OLEX2 (Dolomanov et al., 2009). 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.