Crystal structure of diammonium bis[tris(oxamide dioxime-κ2 N,N′)nickel(II)] bis[tris(oxalato-κ2 O,O′)chromate(III)] 6.76-hydrate

In the structure of the title compound, [Ni(C2H6N4O2)3]2+ cations and [Cr(C2O4)3]3– anions are ordered alternately into negatively charged hydrogen-bonded pillars running parallel to the a axis. These pillars delimit channels accommodating the charge-balancing NH4 + cations as well as the water molecules of crystallization.


Chemical context
Tris(oxalato)metallate(III) complex anions, [M III (C 2 O 4 ) 3 ] 3-, are versatile building blocks for the design of new moleculebased materials with interesting magnetic, electrical and optical properties (Coronado et al., 2000). Through coordination bonds with a variety of metallic ions, these anions can act as ligands, forming various one-, two-and three-dimensional polymeric networks (Pardo et al., 2012;Decurtins et al., 1998). Moreover, in the presence of appropriate hydrogen-donor groups, they can act as hydrogen-bond acceptors resulting in a multitude of hydrogen-bonded networks (Kenfack Tsobnang et al., 2014;Muzioł et al., 2011;Zhuge et al., 2010;Borel et al., 2009). When [M III (C 2 O 4 ) 3 ] 3anions are combined with triply charged tris-bidentate complex cations of D 3 symmetry in which the ligating atoms are all bonded to H atoms or OH groups, they build infinite neutral pillars of alternating complex cations and anions that leave channels in the structure (Bé lombé et al., 2009;Hua et al., 2001;Kuroda, 1991). If functional species (such as spin-crossover or photochromic complexes) are inserted into such voids, interesting properties of the resulting material can be expected, similar to what has been achieved with oxalate-based two-dimensional polymeric networks (Clemente-Leó n et al., 2011). A convenient way of ISSN 2056-9890 forcing additional species into the channels would be by designing compounds with charged, instead of neutral, pillars. In this way, the charge-balancing species could only reside in the channels. This strategy proved successful by combining tris(oxalato)chromate(III) anions, [Cr(C 2 O 4 ) 3 ] 3-, with tris-(oxamide dioxime)nickel(II) cations, [Ni(C 2 H 6 N 4 O 2 ) 3 ] 2+ , the charge-balancing species being K + and H 3 O + (Mbiangué et al., 2012). An attempt to insert NH 4 + (a proton carrier) into such channels led to (NH 4 ) 2 [Ni(C 2 H 6 N 4 O 2 ) 3 ] 2 [Cr(C 2 O 4 ) 3 ] 2 Á-6.76H 2 O (I). Herein, we report its structure.

Structural commentary
The structure of (I) is made up of infinite negatively charged pillars of alternating [Ni(C 2 H 6 N 4 O 2 ) 3 ] 2+ cations and [Cr(C 2 O 4 ) 3 ] 3anions. The pillars run parallel to [100] and delimit channels containing the charge-compensating cations, NH 4 + , as well as the water molecules of crystallization (Figs. 1,2). The molecular components of the asymmetric unit are depicted in Fig. 3. For each metal, two crystallographically independent sites (Ni1 and Ni2 and Cr1 and Cr2, respectively) are present. All of these sites are coordinated in the form of distorted octahedra by six imino N atoms from three bidentate oxamide dioxime ligands (for the nickel sites) and six O atoms from three bidentate oxalate ligands (for the chromium sites). The resulting complexes are chiral. Within a pillar, all the metallic sites have the same chirality, either Á or Ã. Thus, each pillar is chiral but related to another pillar in the crystal through an inversion center. The Ni-N bond lengths range from 2.051 (3) to 2.097 (3) Å and the Cr-O bond lengths, from 1.947 (3) to 1.983 (3) Å (Table 1). Within a pillar, the Ni1Á Á ÁCr1 distances alternate between 4.8897 (8) and 4.9170 (8) Å , and the Ni2Á Á ÁCr2 distances between 4.8743 (7) and 4.9323 (7) Å .

Figure 2
A view along [100] of the crystal packing of (I), illustrating the orientation of the complex ions in an eclipsed configuration within each pillar as well as the channels between the pillars. Table 1 Selected bond lengths (Å ).

Database survey
A search of the Cambridge Structural Database (CSD version 5.41, August 2020 update; Groom et al., 2016) for tris-bidentate transition metal complexes with five membered chelate rings and only N-donor atoms gave 5914 hits. A search for similar complexes but with O-donor atoms gave 1009 hits. A combined search with the two previous queries gave 77 hits. A close examination of the latter structures revealed that only four of them contain hydrogen-bonded pillars of alternating cations and anions with D 3 symmetry. Their CSD refcodes are RUPGEP (Bé lombé et al., 2009), IFOCEL and IFOCIP (Hua et al., 2001), and SOZFIW (Kuroda, 1991) absent from the CSD, was reported a few years ago (Mbiangué et al., 2012).

Synthesis and crystallization
The two precursor salts, (NH 4 (Bailar & Jones, 1939) and [Ni(C 2 2008), were synthesized as described in the literature. The title compound was prepared as follows: finely powdered [Ni(C 2 H 6 N 4 O 2 ) 3 ]SO 4 Á5H 2 O (0.18 g, 0.30 mmol) was added in successive small portions to an aqueous solution (20 ml) of (NH 4 ) 3 [Cr(C 2 O 4 ) 3 ]Á3H 2 O (0.13 g, 0.31 mmol) acidified with two drops of sulfuric acid. The resulting violet mixture was stirred at room temperature (303 K) for 45 min and then filtered. The filtrate was left for evaporation. After one day, violet single crystals were harvested. Upon drying, these crystals lost their brightness, suggesting a possible dehydration.

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 3. All hydrogen atoms of the cationic complex were located in difference-Fourier maps but were finally placed in geometrically idealized positions with U iso (H) = 1.2U eq (N) and U iso (H) = 1.5U eq (O). The assignment of water O atoms and ammonium N atoms was not straightforward. Hence, ten isolated peaks with significant electron densities (between 3.46 and 9.45 e À Å À3 ) were first modeled as N atoms. Their site occupancies were subsequently refined freely. Two of these ten N atoms then had site occupancies inferior but close to unity (0.98 and 0.99). Finally, taking into consideration the electroneutrality of the crystal, the assignment of the aforementioned two N atoms (labeled as N1H and N2H) was assumed to be correct and their site occupancies were fixed to 1. The remainder of the alleged N atoms were finally treated as water O atoms. The site occupancies of these O atoms were fixed to 1 for one of them (O1W) and refined to 0.797 (12), 0.840 (11), 0.835 (11), 0.692 (14), 0.878 (14), 0.909 (13) and 0.812 (14), for the seven others (O2W-O8W). The hydrogen atoms of the ammonium ions and water molecules could not be found in difference-Fourier maps, but they were included in the final formula.  (Rigaku OD, 2015); cell refinement: CrysAlis PRO (Rigaku OD, 2015); data reduction: CrysAlis PRO (Rigaku OD, 2015); program(s) used to solve structure: SHELXT (Sheldrick, 2015a); program(s) used to refine structure: SHELXL2018/3 (Sheldrick, 2015b); molecular graphics: ORTEP-3 for Windows (Farrugia, 2012) and DIAMOND (Brandenburg & Putz, 2018); software used to prepare material for publication: WinGX (Farrugia, 2012) and publCIF (Westrip, 2010).

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.