Crystal structures of bis(phenoxy)silicon phthalocyanines: increasing π–π interactions, solubility and disorder and no halogen bonding observed

We report the syntheses and characterization of three solution-processable phenoxy silicon phthalocyanines (SiPcs). The π–π interactions between the aromatic SiPc cores were studied. In all three cases, the solubility of the molecules was increased by the addition of phenoxy groups while maintaining π–π interactions between the aromatic SiPc groups.


Chemical Context
Organic photovoltaic (OPV) devices represent an emerging technology with immense potential for inexpensive solar energy generation. The majority of these prototypes depend on fullerenes as acceptor molecules that are problematic due to their high manufacturing cost, low photovoltage generation and poor photochemical stability (Li et al., 2014;Eftaiha et al., 2014). Recently, examples have emerged where fullerene-free materials are being implemented into OPV devices reaching overall efficiencies of 5-7% (Li et al., 2014;Eftaiha et al., 2014;Cnops et al., 2014;Zhang et al., 2013). Among these emerging materials are the family of silicon phthalocyanines (SiPcs).
Metalphthalocyanines (MPcs) are composed of a nitrogenlinked tetrameric diiminoisoindoline conjugated macrocycle that chelate a metal or metalloid through two covalent bonds and two coordination bonds (see Scheme 1). The resulting molecules are highly stable materials that have been used for a variety of applications including dyes and pigments for decades. Silicon phthalocyanines (SiPcs) are characterized by having an additional two axial bonds that are perpendicular to the SiPc macrocycle. These axial groups can serve as chemical handles for the functionalization of the base SiPc molecule. Such functionalizational groups can impart solubility as well as change the solid-state arrangement.
Honda et al. and our group have studied highly soluble tri-nhexyl-silyl-SiPc [(3HS) 2 -SiPc] as ternary additives in bulk heterojunction (BHJ) OPV devices (Lessard et al., 2014;Honda et al., 2011Honda et al., , 2009). Our hypothesis was that the high solubility was also combined with a high tendency to crystallize into the solid state with high levels of order. As part of that study, (3HS) 2 -SiPc and an analog bis(3-pentadecylphenoxy)-SiPc [(PDP)2-SiPc] were found to have very fewinteractions between the aromatic SiPc core due to the large alkyl substituents (Lessard et al., 2014). Our group recently reported that simple phenoxylation chemistry can be employed to enhance theinteractions present with the solid-state arrangement of the SiPc molecules, resulting in improved efficiency of planar heterojunction (PHJ) OPV devices Lessard, Grant et al., 2015). Our work on boron subphthalocyanines (BsubPcs) has also illustrated that a meta-methyl phenoxy group is a carbonefficient method for significantly increasing the solubility of BsubPcs (Paton et al., 2012), a characteristic that is necessary for solution-processed OPVs and other characterization techniques. In addition, 3-iodo-phenoxy-BsubPc was found to exhibit halogen bonding between the iodo group and the BsubPc macrocycle and therefore resulting in a well-defined solid-state arrangement. The sum of these observations therefore lead our group to focus on the synthesis of soluble solution-processable phenoxy SiPcs that may have varying degrees of carbon-efficient solubilities and tendencies to crystallize with high order into the solid state. We therefore have synthesized three new derivatives: bis(3-methylphen-oxy)silicon phthalocyanine [(3MP) 2 -SiPc], bis(2-sec-butylphenoxy)silicon phthalocyanine [(2secBP) 2 -SiPc] and bis(3iodophenoxy)silicon phthalocyanine [(3IP) 2 -SiPc] (Fig. 1). We wished to investigate whether a 1-and 4-carbon solubilizing group would both enable solubility and facilitate moreinteractions between the aromatic SiPc units compared to (3HS) 2 -SiPc and also to probe whether halogen bonding would be present in crystals grown of (3IP) 2 -SiPc (Virdo et al., 2013).
Single crystals of (3MP) 2 -SiPc, (3IP) 2 -SiPc and (2secBP) 2 -SiPc were grown by slow diffusion of heptane into THF and were characterized by single crystal X-ray diffraction. (3MP) 2 -SiPc was also grown by slow diffusion of pentane into benzene and evaporation form chloroform, resulting in identical crystals as identified by X-ray crystallography. Fig. 2  An optical microscope image of (3MP) 2 -SiPc grown by slow diffusion of heptane into THF.

Figure 3
Hirshfeld surface analysis of (3MP) 2 -SiPc mapped with (a) d norm and (b) shape index. Red spots on the d norm surface indicate contacts at distances closer than the sum of the corresponding van der Waals radii. Significant interactions between (3MP) 2 -SiPc are outlined by the dashed black circle.

Figure 4
Hirshfeld surface analysis of (2secBP) 2 -SiPc mapped with (a) d norm and (b) shape index. Red spots on the d norm surface indicate contacts at distances closer than the sum of the corresponding van der Waals radii. Significantinteractions between (2secBP) 2 -SiPc are outlined by the dashed black circle.

Figure 5
Hirshfeld surface analysis of (3IP) 2 -SiPc mapped with (a) d norm and (b) shape index. Red spots on the d norm surface indicate contacts at distances closer than the sum of the corresponding van der Waals radii. Significant interactions between (3IP) 2 -SiPc are outlined by the dashed black circle.

Structural commentary
Of note at the structural level, when considering the three reported structures, is the relatively higher disorder observed for (2secBP) 2 -SiPc in the solid state (as indicted by the size of the ellipsoids, Fig. 1) compared to that of (3MP) 2 -SiPc, (3IP) 2 -SiPc and other known bis-phenoxy-SiPc structures (Lessard, Grant et al., 2015). This is consistent with the very high solubility observed for (2secBP) 2 -SiPc and in contrast to the low disorder observed for the also highly soluble (3HS) 2 -SiPc) (Lessard et al., 2014).

Supermolecular Features
The crystal structures were studied using Hirshfeld surface (HS) analysis (Spackman & Jayatilaka, 2009). All three crystals were mapped using (a) d norm and (b) shape index in Fig. 3 for (3MP) 2 -SiPc, Fig. 4 for (3IP) 2 -SiPc and Fig. 5 for (2secBP) 2 -SiPc. In all three figures, the regions shaded in red correspond to the contacts at distances shorter than the sum of the van der Waals radii while the white to blue are for the distances longer than the sum of the van der Waals radii. In each crystal, the close contacts (and their symmetry equivalents) are readily identified on these maps and in all three cases they are different. For example for (3MP) 2 -SiPc ( Fig. 3) one of the hydrogen atoms (H39C) of the 3-methyl group on the phenoxy group experiences a contact of a distance of 2.341 Å (C39-H39CÁ Á ÁH3A-C3; Table 1). It is interesting to note that for (3IP) 2 -SiPc, the iodo group does not have any significant interactions with adjacent molecules (Fig. 2a). These observations are not consistent with our previous observations for various halo-phenoxy-BsubPcs such as 3-iodo-phenoxy BsubPc (Virdo et al., 2013). The shape index (Fig. 3b, 4b, 5b) is based on the two local principal curvatures of the HS, with concave regions shaded in red and convex regions shaded in blue (Spackman & Jayatilaka, 2009). Again, these plots illustrate the difference in the solid-state arrangement between all three molecules (Fig. 3b, 4b, 5b). Unfortunately, similarly to previously reported carbazole derivatives (Rozycka-Sokolowska et al., 2015), these plots do not generate further insight into theinteractions between molecules due to their relatively large distances of 3.5-4.0 Å .
Being interested in the stacking between aromatic macrocycles, we previously established (Lessard, Grant et al., 2015) criteria to compare theinteractions between neighboring Pc molecules for single crystals of SiPcs. Following these established criteria, theinteractions of (3MP) 2 -SiPc were identified and compared to previously published phenoxy SiPcs (Table 2). Fig. 6a illustrates the packing of (3MP) 2 -SiPc crystals which is very similar to the packing of previously reported bis(3,4,5-trifluorophenoxy) SiPc [(345F) 2 -SiPc; Lessard, Grant et al., 2015]. For example, both molecules experience a complete isoindoline stacking where the shortest molecular distances between isoindoline groups of (3MP) 2 -SiPc and (345FP) 2 -SiPc were determined to be 3.655 and 3.580 Å , respectively. In addition, the (3MP) 2 -SiPc exhibits a slip angle of 22.33/22.53 with a slight offset of 0.21 between the aromatic planes while (345F) 2 -SiPc has a less significant slip angle of 18.90 and exactly parallel (0 between planes) interacting isoindoline groups (Fig. 6b Table 1 Comparison of contacts (Å ) less than the sum of the van der Waals radii for various meta-functional bis(meta-functional phenoxy) silicon phthalocyanines.
These results indicate that (3MP) 2 -SiPc has similar interactions to (345F) 2 -SiPc, which represents significant increases ininteraction between SiPc groups compared to the starting Cl 2 -SiPc molecule. (3IP) 2 -SiPc and (2secBP) 2 -SiPc on the other hand exhibit a parallel stacking of two of the peripheral aromatic groups. Of the SiPcs similar to (35F) 2 -SiPc and (246F) 2 -SiPc Lessard, Grant et al., 2015), for example, (3IP) 2 -SiPc experienced a similar stacking to (246F) 2 -SiPc (Lessard, Grant et al., 2015), both having a parallel stacking of two of the peripheral aromatic units of the SiPc chromophore, with very similar inter-ring distances of 3.716 and 3.860 Å , respectively, suggesting similar strength ininteractions between neighboring molecules for both (3IP) 2 -SiPc and (246F) 2 -SiPc (Fig. 6, Table 2). (3IP) 2 -SiPc has a slip angle of 17.55/14.60 with 10.99 between the aromatic planes while (246F) 2 -SiPc has a more significant slip angle of 30.08 and completely parallel (0 between planes) and interacting aromatic groups (Fig. 6, Table 2). (2secBP) 2 -SiPc has a unique two-dimensional stacking where two peripheral aromatic groups will stack with an adjacent SiPc molecule and one of the same peripheral aromatic groups along with a third one will stack in a similar fashion but at 90 from the first interaction ( Fig. 6c, Table 2). In both cases a relatively large inter-ring distance of 3.947 Å was observed, suggesting a weakinteractions between neighboring (2secBP) 2 -SiPcs (Fig. 6, Table 2). This weak interaction is not a surprise due to the additional solubilizing groups (sec-butyl) which space out the molecules and increase the size of the unit cell.

Synthesis of silicon phthalocyanine derivatives
The synthesis of (3MP) 2 -SiPc, (3IP) 2 -SiPcs and (2secBP) 2 -SiPcs were performed following the general procedure used to synthesize F 10 -SiPcÁ . For example, the synthesis of (3MP) 2 -SiPc was performed in a roundbottom flask equipped with a condenser and nitrogen purge, which was filled with a 10:1 molar excess of m-cresol (2.3g, 21 mol) to Cl 2 -SiPc (1.3g, 2.1 mol) in chlorobenzene (100 ml). The mixture was stirred and heated to 388 K overnight and cooled to room temperature. The product was then obtained by precipitation into isopropanol and filtered. The product was then dried in a vacuum oven overnight. Yield: 1.3g (80.2 mol%). DART Mass spectroscopy: calculated mass: 755.234, obtained mass: 755.236. (3IP) 2 -SiPcs and (2secBP) 2 -SiPcs were synthesized under similar conditions and crystals were again obtained by slow diffusion of heptane into a THF solution.

Refinement
Crystal data collection and structure refinement details are summarized in Table 3. H atoms were placed in calculated positions C-H = 0.94-0.98 Å and included in a riding-motion approximation with U iso (H) = 1.2U eq (C) or 1.5U eq (C methyl ).
In (3MP) 2 -Si there appears to be pseudosymmetry with an approximate centre of symmetry. The c-glide reflections are weak but present and the P2 1 /c structure refines only to ca R1 = 10% compared to 4.4% for the P2 1 structure. The crystal is an inversion twin with a ratio of components of 0.51 (4):0.49 (4).
During the refinement of (2secBP) 2 -SiPc, electron density peaks were located that were believed to be highly disordered solvent molecules (possibly pentane/dichloromethane). Attempts made to model the solvent molecule were not successful. The SQUEEZE option (Spek, 2015) in PLATON (Spek, 2009) indicated there was a large solvent cavity 367 A 3 . In the final cycles of refinement, this contribution (99 electrons) to the electron density was removed from the observed data. The density, the F(000) value, the molecular weight and the formula are given without taking into account the results obtained with SQUEEZE. Similar treatments of disordered  Table 2 Summary of single-crystal X-ray diffraction data (Å , ).

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. 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.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å 2 )
x y z U iso */U eq I1 0.58373 (2) 0.18313 (2) 0.00675 (4) 0.03731 (10)  where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max < 0.001 Δρ max = 0.40 e Å −3 Δρ min = −0.36 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. 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.