Different packing motifs mediated by weak interactions and polymorphism in the crystal structures of five 2-(benzylidene)benzosuberone derivatives

The title compounds, which differ in the substituent at the 4-position of the pendant benzene ring, show different packing motifs mediated by weak C—H⋯X (X = O or N) interactions. One of them is a polymorph of a known structure.


Structural commentary
The molecular structures of (I)-(V) are shown in Figs The molecular structure of (I), showing 50% probability displacement ellipsoids.
observed. Otherwise, the geometrical data for (I)-(V) are unexceptional and similar to those for related compounds (Dimmock et al., 1999(Dimmock et al., , 2002. It may be noted that a polymorph of (I) [Cambridge Structural Database (CSD; Groom et al., 2016) refcode VENQUA; Dimmock et al., 1999] has been reported in the same space group, i.e. P2 1 /c; VENQUA was recrystallized from methanol solution rather than ethanol for (I). The bond lengths and angles in (I) and VENQUA are very similar, although there is a $10 difference in the dihedral angle between the benzene rings [value for VENQUA = 35.88 (11) , calculated with PLATON (Spek, 2009)]; for an overlay plot of (I) and VENQUA, see the supporting information.

Supramolecular features
There are obviously no classical hydrogen bonds in these structures and, in each case, just one C-H group can be identified as the donor for a weak hydrogen bond with atom O1 as the acceptor in (I)-(IV) and atom N1 in (V); geometrical data for these interactions are listed in Tables 1-5 and illustrated in Figs. 7-11. All the structures also feature weak C-HÁ Á Á interactions with either the fused or pendant benzene rings as acceptors, but (II) and (III) are the only structures to display weak aromaticstacking, in both cases between inversion-related C13-C18 rings. For (II), the centroid-centroid separation is 3.8414 (7) Å and the slippage is 1.72 Å ; equivalent data for (III) are 3.9475 (7) and 1.89 Å , respectively.
The packing motifs for the extended structures of (I) and

Table 7
Summary of the C-HÁ Á ÁO and C-HÁ Á ÁN hydrogen bonds and packing motifs for 2-(benzylidene)benzosuberone derivatives. The packing motifs for (II) and (IV) feature inversion dimers. In (II), C18-H18 (meta to the 4-substituent) is the donor group and R 2 2 (14) loops arise. In this motif, C12-H12 is 'sandwiched' between the donor and acceptor and the H12Á Á ÁO1 separation of 2.60 Å (see Fig. 8) is borderline to be regarded as a directional bond. The donor group in (IV) is C10-H10 in the fused benzene ring, which generates an R 2 2 (10) loop. The only possible interaction involving the Cl atom is a long contact from C8-H8, with HÁ Á ÁCl = 2.93 Å . The presence of the cyano group in (V) allows for the formation of pairwise C-HÁ Á ÁN hydrogen bonds and an R 2 2 (10) graph-set motif arises; the shortest HÁ Á ÁO contact in (V) is 2.72 Å .
Rather than the C(8) chains arising from C15-H15Á Á ÁO1 hydrogen bonds seen in (I), the packing for VENQUA (see above) features inversion dimers built from pairwise C10-H10Á Á ÁO1 interactions, which are very similar to those seen in 4-chloro derivative (IV) in the present study. It may be noted that the density of VENQUA ( = 1.368 Mg m À3 ) is significantly greater than that of (I) ( = 1.284 Mg m À3 ), suggesting that the former might be the more stable polymorph if the 'rational packing rule' (Kitaigorodskii, 1961) is applicable in this case.
In order to gain further insight into these different packing motifs, the Hirshfeld surfaces and fingerprint plots for (I) and VENQUA were calculated using CrystalExplorer (Turner et al., 2017), following the approach recently described by Tan et al. (2019). The Hirshfeld surface for (I) (see supporting information) shows the expected large red spots (close contacts) in the vicinity of H15 and O1 corresponding to the C15-H15Á Á ÁO1 interaction noted above, but there is little if any evidence of close contacts in the vicinity of H19A and H19C corresponding to the C-HÁ Á Á contacts listed in Table 1. The surface for VENQUA (see supporting informa-tion) shows red spots in the vicinity of H10 and O1 corresponding to the C10-H10Á Á ÁO1 hydrogen bond and H2A (our numbering scheme) corresponding to a C3-H2AÁ Á Á interaction (HÁ Á Á = 2.69 Å ) to the centroid of the C6-C11 benzene ring, but there are also probably spurious features close to H8 and H17 corresponding to a short HÁ Á ÁH contact of 2.07 Å between these atoms, which possibly arose because the H atoms of the C19 methyl group in VENQUA were geometrically placed and not treated using a rotating-group model. Notwithstanding this, the fingerprint plots for (I) and VENQUA (see supporting information) decomposed into the different percentage contact types (Table 6) are almost identical; HÁ Á ÁH (van der Waals) contacts dominate both structures, followed by CÁ Á ÁH/HÁ Á ÁC and then OÁ Á ÁH/HÁ Á ÁO. The percentage contributions of the other contact types are negligible.

Database survey
A survey of the Cambridge Structural Database (CSD; Groom et al., 2016) revealed 167 structures incorporating a 1-benzosuberone fragment but only 20 hits when an exocyclic C C double bond at the 2-position was added to the search structure. The key papers reporting the structures of closely related, differently substituted, 2-benzylidene-1-benzosuberones are Dimmock et al. (1999Dimmock et al. ( , 2002. The hydrogen-bond data for (I)-(V) and the 12 structures reported in the two papers by Dimmock et al. are summarized in Table 7. The most frequently observed motif is the centrosymmetric R 2 2 (10) loop involving C10-H10 as the donor group, but there are many others involving different C-H groups as donor and we see no obvious connection to the nature and position of the substituent(s) on the remote benzene ring. There are no structures in which the fused and pendant benzene rings tend towards being perpendicular (dihedral angle > 60 ).
The fact that (I) and VENQUA have similar conformations but distinct packing motifs mediated by different C-HÁ Á ÁO interactions to the same acceptor O atom may be compared with the fascinating recent survey of weak-interaction polymorphs by Lo Presti (2018). He concluded that weak hydrogen bonds and solvent effects may play an important kinetic role in promoting polymorph formation (after all, something has to favour a situation where the lowest-energy packing motif is not adopted) but they do not play a dominant energetic role in polymorph formation and that the overall energy balance between dispersive (attractive) and repulsive interactions is the most important consideration.

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 8. All H atoms were located geometrically (C-H = 0.95-0.99 Å ) and refined as riding atoms, with U iso (H) = 1.2U eq (C) or 1.5U eq (methyl C). The methyl groups in (I) and (II) were allowed to rotate, but not to tip, to best fit the electron density.

sup-1
Acta Cryst. For all structures, data collection: CrysAlis PRO (Rigaku, 2017); cell refinement: CrysAlis PRO (Rigaku, 2017); data reduction: CrysAlis PRO (Rigaku, 2017); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL2014 (Sheldrick, 2015); molecular graphics: ORTEP-3 (Farrugia, 2012); software used to prepare material for publication: 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.

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.

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.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å 2 )
x y z U iso */U eq C1 −0.05094 (13) 0.41786 (13) 0.37210 (9) 0.0206 (2) (15)  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.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å 2 )
x y z U iso */U eq C1 0.2086 (2)  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.