Crystal structure and Hirshfeld surface analysis study of (E)-1-(4-chlorophenyl)-N-(4-ferrocenylphenyl)methanimine

The unsubstituted cyclopentadienyl ring is rotationally disordered while the other Cp ring and its substituent are close to coplanar. In the crystal, the molecules pack in ‘bilayers’ parallel to the ab plane with the ferrocenyl groups on the outer faces and the substituents directed towards the regions between them. The ferrocenyl groups are linked by C—H⋯π(ring) interactions.

In a continuation of our research towards the synthesis of ferrocene-derived Schiff bases, we have been using 4-ferrocenyl aniline as an intermediate in the synthesis of new heterocyclic systems and have studied the condensation reactions between 4-ferrocenyl aniline and 4-chlorobenzaldehyde. The title compound (I) was obtained and characterized by single crystal X-ray diffraction techniques as well as by Hirshfeld surface analysis.

Structural commentary
4-Ferrocenyl aniline was synthesized according to a reported procedure (Hu et al., 2001;Ali et al., 2013) and single crystals of its condensation product with 4-chlorobenzaldehyde were obtained by recrystallization from methanol ( Fig. 1). Bond distances and angles are in the expected ranges and agree well with values observed for similar compounds (see e.g. Kumar et al., 2020;Shabbir et al., 2017;Toro et al., 2018). The unsubstituted cyclopentadienyl ring, C1-C5, was found to be rotationally disordered, with a refined occupancy of 0.666 (7) for the major moiety. The two Cp rings are not quite parallel as there is a 2.7 (5) dihedral angle between them. The substituted cyclopentadienyl ring, C6-C10, is nearly coplanar with the phenyl-1-(4-chlorophenyl)methanimine substituent. The Cp ring is inclined by 16.8 (2) with respect to the C11-C16 phenylene ring. The imine fragment is essentially coplanar with the chlorophenyl unit, with an r.m.s. deviation from planarity of only 0.05 Å . The dihedral angle between the phenylene ring and the plane of the 4-chlorophenyl-methanimine unit, N1/C17-C23, is 9.23 (10) . This renders the entire molecule, with the exception of the Fe atom and the unsubstituted Cp ring, mostly flat.

Supramolecular features
In the crystal, molecules are arranged in double layers perpendicular to the c axis with alternating ferrocenyl and Schiff base segments, with the ferrocenyl groups facing towards the outside of each layer and bordering the ferrocene moieties of the neighboring layer, and the phenyl-1-(4chlorophenyl)methanimine substituents at the center of the double layers with the substituents from both sides of the layer interdigitating with each other ( Detail of the intermolecular interactions. C-HÁ Á ÁCl hydrogen bonds and C-HÁ Á Á(ring) interactions are depicted, respectively, by green and orange dashed lines. Non-interacting H atoms are omitted for clarity.

Figure 3
Packing viewed along the b-axis direction with intermolecular interactions depicted as in Fig. 2. Non-interacting H atoms are omitted for clarity.

Figure 1
The asymmetric unit of the title compound with the atom-numbering scheme. Displacement ellipsoids are drawn at the 50% probability level. Only the major orientation of the disordered cyclopentadienyl ring is shown.

Hirshfeld surface analysis
In order to visualize the intermolecular interactions in the crystal of the title compound, a Hirshfeld surface (HS) analysis (Hirshfeld, 1977) was carried out using Crystal Explorer 17.5 (Turner et al., 2017). In the HS plotted over d norm (Fig. 4), the white surface indicates contacts with distances equal to the sum of van der Waals radii, and the red and blue colors indicate distances shorter (in close contact) or longer (distinct contact) than the van der Waals radii, respectively (Venkatesan et al., 2016). The bright-red spots indicate their roles as the respective donors and/or acceptors. The blue regions indicate positive electrostatic potentials (hydrogen-bond donors), while the red regions indicate negative electrostatic potentials (hydrogen-bond acceptors).
The shape-index of the HS is a tool to visualizestacking by the presence of adjacent red and blue triangles; the absence of adjacent red and/or blue triangles, Fig. 5, indicates that there are nointeractions. The overall two-dimensional fingerprint plot is shown in Fig. 6a, and those delineated into HÁ Á ÁH, HÁ Á ÁC/CÁ Á ÁH, HÁ Á ÁCl/ClÁ Á ÁH, HÁ Á ÁN/NÁ Á ÁH, CÁ Á ÁC, CÁ Á ÁN/NÁ Á ÁC and ClÁ Á ÁCl contacts (McKinnon et al., 2007) are illustrated in Fig. 6b-h, respectively, together with their relative contributions to the Hirshfeld surface. The most important interaction is HÁ Á ÁH, contributing 46.1% to the overall crystal packing, which is reflected in Fig. 6b as widely scattered points of high density due to the large hydrogen content of the molecule. The presence of C-HÁ Á Á interactions, as described in the Supramolecular features section, is indicated by pairs of characteristic wings in the fingerprint plot representing HÁ Á ÁC/ Jerry P. Jasinski tribute 3 of 5

Figure 4
View of the three-dimensional Hirshfeld surface of the title compound, plotted over d norm in the range À0.1325 to 1.1632 a.u. The red dots indicate the C-HÁ Á Á(ring) interactions involving the ferrocene and the C18-C23 ring.

Figure 5
Hirshfeld surface of the title compound plotted over shape-index.

Synthesis and crystallization
4-Ferrocenyl aniline was synthesized according to a reported procedure (Hu et al., 2001;Ali et al., 2013). In a 250 mL round-bottom flask, 1.0 mmol of 4-ferrocenyl aniline in 15 mL of dried methanol was mixed with an equimolar amount of 4chlorophenyl aldehyde in 15 mL of dried methanol. The mixture was agitated under reflux, the progress of the reaction was monitored by TLC, and the desired product was formed within 6 h. The solvent was removed under vacuum and the solid that was obtained was recrystallized from methanol (yield: 87%) to yield brown crystals, m.p. 210 K. 1 H NMR (300 MHz, CDCl 3 ) 4.08 (s, 5H, Cp C 5 H 5 ); 4.36 (t, 2H, Cp

Refinement
Crystal, data collection and refinement details are presented in Table 1. Analysis of 1284 reflections having I/(I) > 15 and chosen from the full data set with CELL_NOW (Sheldrick, 2008) showed the crystal to be either split or non-merohedrally twinned. The top choice of unit cell had parameters a = 7.662, b = 10.009, c = 45.974 Å , = 90.05, = 90.21, = 89.97 (unrefined) with a second component (14%) rotated 180 about the b axis. To eliminate possible bias, the raw data were processed as triclinic using the multi-component version of SAINT (Bruker, 2020) under control of the two-component orientation file generated by CELL_NOW, leading to an orthorhombic cell within experimental error and a twin matrix of: À0.99988 À 0.00291 À 0.00258 À 0.00684 0.99978 0.00453 0.09083 0.09422 À 0.99967, thus indicating presence of two separate domains not related by twinning ('split crystal'). The data were corrected for absorption using TWINABS (Sheldrick, 2009), which was also used to extract a single-component reflection file from the two-component intensity data, which was used to determine the space group and solve the structure. The resulting space group of Pbca required transformation of the original cell by the matrix: 0 1 0 1 0 0 0 0 À1. Trial final refinements with the single-component reflection file and with the complete two-component data showed the former to be more satisfactory on the basis of a lower values for R1 and su's on derived parameters as well as smaller residual features about the Fe atom.
H atoms attached to carbon were placed in calculated positions (C-H = 0.95-1.00 Å ). All were included as riding contributions with isotropic displacement parameters 1.2-1.5 times those of the parent atoms. The unsubstituted cyclopentadienyl ring is rotationally disordered over two sets of sites with the two components refined as rigid pentagons (AFIX 56 constraint of SHELXL). ADPs of equivalent major and minor disordered C atoms were constrained to be identical. The occupancy ratio for the two orientations refined to a 0.666 (7)/0.334 (7)

Special details
Experimental. The diffraction data were obtained from 7 sets of frames, each of width 0.5° in ω, collected with scan parameters determined by the "strategy" routine in APEX3. The scan time was 40 sec/frame. Analysis of 1284 reflections having I/σ(I) > 15 and chosen from the full data set with CELL_NOW (Sheldrick, 2008) showed the crystal to nonmerohedrally twinned. The top choice of unit cell had parameters a = 7.662, b = 10.009, c = 45.974 Å, α = 90.05, β = 90.21, γ = 89.97° (unrefined) with a second component (14%) rotated 180° about the b-axis. To eliminate possible bias, the raw data were processed as triclinic using the multi-component version of SAINT (Bruker, 2020) under control of the two-component orientation file generated by CELL_NOW leading to an orthorhombic cell within experinental error and a twin matrix of: -0.99988 -0.00291 -0.00258 -0.00684 0.99978 0.00453 0.09083 0.09422 -0.99967. 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. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger. H-atoms attached to carbon were placed in calculated positions (C-H = 0.95 -1.00 Å). All were included as riding contributions with isotropic displacement parameters 1.2 -1.5 times those of the attached atoms. The C1···C5 ring is rotationally disordered over two orientations in a 0.666 (7)/0.334 (7) ratio. The two components were refined as rigid pentagons. Trial refinements with the single-component reflection file extracted from the full data set with TWINABS and with the complete two-component data showed the former to be more satisfactory on the basis of a lower values for R1 and su's on derived parameters as well as smaller residual features abot the Fe atom.