Structural diversity in copper(I) iodide complexes with 6-thioxopiperidin-2-one, piperidine-2,6-dithione and isoindoline-1,3-dithione ligands

Five copper(I) iodide coordination compounds were synthesized and characterized by single-crystal X-ray diffraction measurements; the resulting structures display a diverse array of structural features.


Chemical context
Copper (I) iodide compounds have been of interest for the past 50 years because of their diverse structural (Peng et al., 2010) and spectroscopic properties (Ford et al., 1999;Hardt & Pierre, 1973). In particular, Cu I complexes range from simple Cu 2 I 2 L 2 dimers (L = Lewis basic ligands) to complex threedimensional coordination polymers (Peng et al., 2010). Traditionally, soft Lewis basic donors such as thiols or phosphines have been used as ligands to the Cu I centers. We were interested in exploring the structures of Cu I coordination complexes with three ligands, piperidine-2,6-dithione (SNS), isoindoline-1,3-dithione (SNS6), and 6-thioxopiperidin-2-one A molecular drawing of I with 50% probability ellipsoids. Dotted lines are used to indicate hydrogen-bonding interactions. All H atoms bound to C atoms are omitted. [Symmetry code: (i) Àx + 1, Ày + 1, Àz.]

Figure 3
A molecular drawing of II shown with 50% probability ellipsoids. Dotted lines are used to indicate hydrogen-bonding interactions. All H atoms bound to C atoms are omitted. [Symmetry code: (i) Àx + 1, y, Àz + 1 2 .]

Figure 1
Diagrams of the three ligands used in the preparation of structures I-V.

Figure 6
A molecular drawing of III viewed along the crystallographic b axis with 50% probability ellipsoids. Dotted lines are used to indicate hydrogenbonding interactions. All H atoms bound to C atoms are omitted.

Figure 5
A molecular drawing of III's Cu 3 I 3 fused rings viewed along the crystallographic a axis with 50% probability ellipsoids. All H atoms bound to C atoms are omitted.
In contrast to the monodentate SNO ligands in III, which only permit polymer propagation in III through the 3 -I À atoms, the bidentate SNS ligand facilitates polymer propagation in IV. This results in the formation of rings formed by four [Cu( 2 -I) 2 ( 2 -SNS)] units. The propagation of these rings in the (001) crystallographic plane results in a mesh-like sheet structure, and the layering of these sheets perpendicularly to the (001) plane results in the presence of sizable solventaccessible voids (ca 200 Å 3 ) in the structure (Fig. 9). These voids are filled with a combination of acetonitrile and dichloromethane in an approximately 2:1 ratio; however, these solvent species were positionally disordered and the PLATON SQUEEZE routine (Spek, 2015) was required to model the diffuse electron density from the solvent species in these voids (see Refinement section).
Complex V also crystallizes as two-dimensional polymeric sheets with the symmetry-independent unit formula [Cu( 2 -I)( 2 -SNS6)] (Fig. 10). The Cu center is coordinated by two 2 -I À atoms and two thione moieties of the 2 -SNS6 ligands and has a distorted tetrahedral geometry [I-Cu-I and I-

Figure 7
A molecular drawing of the repeat unit of IV shown with 50% probability ellipsoids.

Figure 8
A molecular drawing of IV with the full coordination spheres of the Cu centers shown with selected atom labels. All atoms are shown with 50% probability ellipsoids; dotted lines are used to indicate hydrogen-bonding interactions. All H atoms bound to C atoms are omitted. [Symmetry codes: (i) Àx, Ày, 1 À z; (ii) 1 À x, 1 À y, 1 À z; (iii) 1 À x, 1 À y, 1 À z; (iv) À1 + x, y, z]

Figure 9
A molecular drawing of IV viewed along the [101] crystallographic direction with 50% probability ellipsoids. All H atoms bound to C atoms are omitted.

Figure 10
A molecular drawing of the symmetry-independent portion of V with the full coordination sphere of the Cu center shown. All atoms are shown with 50% probability ellipsoids; dotted lines are used to indicate hydrogen-bonding interactions. All H atoms bound to C atoms are omitted. [Symmetry codes: (i) x, 1 + y, z; (ii) x, 1 À y, 1 2 + z.] Cu-S angles between 100.30 (6) and 120.16 (7) ]. Whereas the two S-Cu distances are almost identical, the two Cu-I distances are quite different (Table 1).
The polymeric sheet propagates parallel to the (100) crystallographic plane. The 2 -I À atoms bridge two Cu centers and form Cu-I zigzag chains that propagate parallel to the [010] crystallographic direction. Similarly to IV, the 2 -SNS6 ligands participate in the polymer propagation in V by bridging two Cu atoms and connecting the Cu-I chains and are generated by the c glide plane (Fig. 11). Among the five structures discussed, V is the only non-centrosymmetric structure. This results in a packing motif with a polar arrangement of SNS6 ligands on one side of the inorganic sheets, which results in a smaller spacing between the inorganic layers [7.598 (3) Å , see Fig. 12] in V than in III [14.134 (5) Å , see Fig. 13].

Supramolecular features
Among the five structures reported in this work, III, IV, and V crystallize as polymeric sheets; their extended structural characteristics are described above. In addition to the polymeric structural features in III, IV, and V, there are also several types of intermolecular interactions present in each of the five structures that are relevant to a description of their supramolecular architectures.
All structures except III display non-classical (e.g., H-atom acceptors that are not N, O or Cl) hydrogen-bonding interactions between the N-H of the SNO/SNS/SNS6 ligands and the 2 -I À / 3 -I À atoms. According to our statistical analysis of A molecular drawing of V viewed along the crystallographic b axis with 50% probability ellipsoids with emphasis on the weak N-HÁ Á ÁI interactions (dotted lines). All H atoms bound to C atoms are omitted.

Figure 12
A molecular drawing of V viewed along the crystallographic c axis with 50% probability ellipsoids. Dotted lines are used to indicate hydrogenbonding interactions. All H atoms bound to C atoms are omitted.

Figure 13
A molecular drawing of III viewed along the crystallographic c axis with 50% probability ellipsoids. Dotted lines are used to indicate hydrogenbonding interactions. All H atoms bound to C atoms are omitted.
reported to the CSD, their DÁ Á ÁA distances range from 3.15 to 4.12 Å with a mean DÁ Á ÁA distance of 3.69 (13) Å . The DÁ Á ÁA distances in structures I, II, IV, and V are typical for these types of interactions (Table 2). For structures I and II, the N-HÁ Á ÁI interaction is intramolecular. For IV, there are two symmetry-independent hydrogen-bonding interactions, which is expected given that the structure contains two symmetryindependent SNS ligands. The first, between atoms N1-H1Á Á ÁI1 ii [symmetry code: (ii) Àx + 1, Ày + 1, Àz + 1], is a stronger interaction; the second is between atoms N2 iv -H2 iv Á Á ÁI1 [symmetry code: (iv) x + 1, y, z] and is a weaker interaction (Table 2). Both interactions form S(6) hydrogenbonding motifs (Etter et al. 1990), which provide some rigidity to the mesh-like sheet of the polymer.
Structure III is unique among all the structures discussed in this work as it is the only structure to exhibit classical hydrogen-bonding interactions. There are two identical hydrogen bonds per SNO ligand, with the N-H serving as an H-bond donor and the O atom serving as an H-bond acceptor [N1-H1Á Á ÁO1 iii and N1 iii -H iii Á Á ÁO1; symmetry code: (iii) Àx + 1, Ày + 1, Àz + 2]. These hydrogen bonds are relatively strong (Table 2) and form R 2 2 (8) motifs between the stacked [Cu 3 I 3 ] n polymeric layers. Their presence leads to an extended three-dimensional framework structure, where the propagation of the [Cu 3 I 3 ] n polymeric sheets accounts for two dimensions and the connection of those sheets through the hydrogen-bonding interactions provides the third (Fig. 13).
Structure V has two distinct types of intermolecular interactions. First, there is the non-classical hydrogen-bonding interaction between the N-H of the SNS6 ligand and the symmetry-equivalent 2 -I À atoms [N1-H1Á Á ÁI1 ii; symmetry code: (ii) x, 1 À y, À 1 2 + z] within the same polymeric sheet. This interaction forms R 2 2 (6) motifs that are of typical strength (see Figs. 10 and 11; Table 2). In addition to the non-classical hydrogen-bonding interactions, there are alsostacking interactions between SNS6 ligands within the same polymeric sheet due to the presence of the extended system in the SNS6 ligand backbone. These interactions, formed by the overlap between the five-membered rings with atoms C1-C2-C7-C8-N1 (R 5 ) and the phenyl rings with atoms C2 i -C3 i -C4 i -C5 i -C6 i -C7 i (R 6 ) [symmetry code: (i) x, 1 + y, z], is of moderate strength [plane R 5 to R 6 centroid distance: 3.369 (5) Å ; R 5 to R 6 centroid offset distance: 1.165 (14) Å ]. Thesestacking interactions, in tandem with the increased size of the SNS6 ligand relative to the SNS/SNO ligands, results in a tightly packed two-dimensional sheet (packing coefficient: 71.8%), which prevents the formation of the more mesh-like structure seen in IV (packing coefficient: 54.1%) (Kitaigorodskii, 1973).

Database survey
All searches in the Cambridge Structural Database (Version 5.41, latest update May 2020; Groom et al. 2016) were performed with moderate search criteria (for structures I and II: no errors or ions, not polymeric, only single crystal structures; for structures III, IV, and V: no errors or ions, only single crystal structures. The surveys of the database for each individual structure are described below. I: A search for Cu 2 ( 2 -I) 2 dimers with two neutral ligands binding with one nitrogen and one sulfur atom resulted in 17 matches. Only one had a homometallic [Cu( 2 -I) 2 (S)(N)] 2 type structure where the S and N donors were part of monodentate ligands, which indicates that the coordination environment in I is a relatively unusual one. This structure, bis[( 2 -iodo)(acetonitrile)(triphenylthiophosphorane)copper(I)] (refcode: OCALOT; Lobana et al., 2001), has similar Cu-S and Cu-N distances and a slightly longer Cu-I distance. However, OCALOT has a dramatically longer CuÁ Á ÁCu distance [3.4141 (16) Å ] than that in I (Table 1). This elongation is likely due to the larger steric requirements of the SPPh 3 sulfur donor ligand in OCALOT.
An additional, broader search for all non-polymeric Cu 4 ( 3 -I) tetrahedra yielded 130 results for Cu 4 ( 3 -I) 4 (L) 4 (L = N, S, P, I, O, As) tetrahedra with L as a neutral ligand. All of the resulting structures had identical first coordination spheres for each of the Cu centers [e.g., Cu 4 ( 3 -I) 4 (L) 4 , rather than the Cu 4 ( 3 -I) 4 (L) 2 (L 0 ) 2 in II]. To the best of our knowledge, II is the first reported instance of a non-polymeric Cu 4 ( 3 -I) 4 tetrahedron with N and (non-bridging) S ligands.
V: A search for structures with Cu-X zigzag chains that did not contain the Cu 2 ( 2 -X) 2 rhombus afforded 112 matches. Among these, 56 were for polymeric homometallic structures and three of these [refcodes: AFUDUA (Caradoc-Davies et al., 2002), CIQQOL (Musina et al., 2017), and FIWWAK (Cingolani et al., 2005)] contained one-dimensional Cu-I À zigzag chains. All three structures contain tetrahedral Cu I centers coordinated by the two 2 -I atoms and two neutral donor ligands (binding with sulfur and nitrogen for AFUNDA, arsenic for CIQQOL, and FIWWAK). These structures have similar geometries to that of V except for the Cu-ligand distances.
Unless otherwise specified, all reactions were performed at room temperature under a dry N 2 atmosphere using standard glovebox methods.
I was prepared by combining 10 ml of a clear yellow solution of 6-thioxopiperidin-2-one (0.500 mmol) in dichloromethane with 10 mL of a colorless solution of CuI (0.502 mmol) in acetonitrile. Upon combination, the solution turned a bright-orange color. Vapor diffusion of the orange solution with diethyl ether afforded large, yellow, blockshaped crystals of I after three days.
Two by-products were also obtained from the reaction of 6-thioxopiperidin-2-one and CuI. The first (II) were small, yellow, plate-shaped crystals that co-crystallized with the larger yellow block-shaped crystals of I. The second byproduct formed after exposing the initial orange solution from the reaction of 6-thioxopiperidin-2-one and CuI to air, and allowing that solution to slowly evaporate in air for approximately one week. After this time, small, red-orange crystals of III were obtained.
IV was prepared by layering 10 mL of a clear yellow solution of piperidine-2,6-dithione (1.01 mmol) in dichloromethane over 10 mL of a colorless solution of CuI (1.00 mmol) in acetonitrile. Dark-red crystals of IV were obtained after one week.
Black, needle-shaped crystals of V were obtained in a similar manner to IV, with the exception that 1.00 mmol of isoindoline-1,3-dithione was used instead of piperidine-2,6dithione.

Refinement
For structure I, the diffraction data were consistent with the space groups P1 and P1; the E-statistics were consistent for the centrosymmetric space group P1 and were used to make the final space-group determination. For structures II-V, a combination of the systematic absences in the diffraction data and the E-statistics were used to assign the centrosymmetric space groups C2/c (II), Pbcn (III), P2 1 /c (IV) and the noncentrosymmetric space group Cc (V).
The structures were solved via intrinsic phasing and refined by least-squares refinement on F 2 followed by difference-Fourier synthesis. All non-hydrogen atoms were refined with anisotropic displacement parameters. Unless otherwise specified, all hydrogen atoms were included in the final structure-factor calculation at idealized positions and were allowed to ride on the neighboring atoms with relative isotropic displacement coefficients.
The coordinates of the H atoms bound to N atoms in structures I, II, and III were refined freely with a distance restraint for each N-H distance.
In structure IV, there were three partially occupied solvent molecules of dichloromethane and one partially occupied molecule of acetonitrile present in the asymmetric unit. A significant amount of time was invested in identifying and refining the disordered molecules. Bond-length restraints were applied to model the molecules, but the resulting isotropic displacement coefficients suggested the molecules were mobile. In addition, the refinement was computationally unstable. The SQUEEZE option (Spek, 2015) of the PLATON software suite (Spek, 2020) was used to correct the diffraction data for diffuse scattering effects and to identify the solvent molecule. PLATON calculated the upper limit of volume that can be occupied by the solvent in the unit cell to be 615 Å 3 . This solvent-accessible volume is comprised of two smaller (ca 115 Å 3 ) and two larger (ca 196 Å 3 ) solventaccessible voids and is 27% of the unit-cell volume. The program calculated 155 electrons in the unit cell for the diffuse species. This corresponds to approximately one molecule of dichloromethane (42 electrons) that is 50% occupied and one molecule of acetonitrile (22 electrons) in the asymmetric unit. It is very likely that the solvent molecules are disordered over several positions. All derived results in Tables 1 and 2 are based on the known contents. No data are given for the diffusely scattering species.
Crystal data, data collection and structure refinement details are summarized in Table 3.   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.

catena-Poly[[(µ-6-sulfanylidenepiperidin-2-one-κ 2 O:S)copper(I)]-µ 3 -iodido] (III)
Crystal data where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max = 0.001 Δρ max = 1.31 e Å −3 Δρ min = −1.03 e Å −3 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.

Poly[[(piperidine-2,6-dithione-κS)copper(I)]-µ 3 -iodido] (IV)
Crystal data where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max = 0.001 Δρ max = 1.58 e Å −3 Δρ min = −0.48 e Å −3 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.

Poly[[(µ-isoindoline-1,3-dithione-κ 2 S:S)copper(I)]-µ 3 -iodido] (V)
Crystal data  (12) 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.