2-{3-Cyano-4-[2-(4-diethylamino-2-hydroxyphenyl)ethenyl]-5,5-dimethyl-2,5-dihydrofuran-2-ylidene}malononitrile acetone 0.25-solvate

In the title compound, C22H22N4O2·0.25C3H6O, the disordered acetone molecule lies with partial occupancy about the 2 axis. The molecule of the malononitrile derivative is essentially planar excluding the methyl groups, with the largest deviation from the mean plane through the non-H atoms being 0.1955 (13) Å. Two rotamers with different orientations of the benzene ring are observed in the ratio of 0.919 (2):0.081 (2), and as a result the OH group is disordered over two sets of sites. In the crystal, the molecules form ribbons along (101) utilizing a strong O—H⋯N(cyano) hydrogen bond. Interleaving of the nearly planar ribbons is provided by the twofold disordered acetone molecule through C—H⋯O interactions.


Comment
Organic donor-π-acceptor (D-π-A) molecules show much promise due to their potential application in areas such as photonics, optical power limiting and optical data storage (Dalton, 2004;Ma et al., 2002). These molecules are typically push-pull conjugated systems that can be modified by altering either the donor, acceptor or conjugated interconnect moieties. Consequently, there are typically a number of options available to iteratively improve the overall molecular response and stability of such compounds, especially when applied to their use in second-order nonlinear optics. For example, a successful approach to optimizing their second-order nonlinear optical (NLO) response is based on tuning the ground-state polarization -and hence the degree of bond-length alternation -through modification of the end groups and the spacer (Marder et al., 1997). Furthermore, thermal and photochemical stability can be improved through the use of ring-locked spacer units between the donor-acceptor moieties. However, in order to be successfully deployed in devices the chromophores need to be embedded into a polymer and their dipoles aligned in a non-centrosymmetric fashion using a process known as poling. This can be difficult to achieve as NLO chromophores embedded in polymer matrices have a tendency to aggregate due to their large dipole moments. As a result there is often a need to find expedient methods to minimize aggregation in NLO chromophores and these include the use of bulky fluorinated and non-fluorinated pendant groups as well as the use of hydrogen-bonding substituents to control the molecular interactions.
Studies of hydrogen bonds connecting organic and organic-inorganic compounds have long been a topic of intense research in crystal engineering because this allows not only for a rational approach to bottom-up construction but hydrogen bonds also effectively regulate the molecular architecture (Li et al., 2007). With this in mind, and in line with our on-going work on the development of novel organic NLO compounds, we sought a straightforward route to a D-π-A molecule containing a hydroxyl substituent to allow us to study its impact on crystal packing, as well as providing a potentially reactive site for future modifications. Consequently, we prepared the title compound 3 using the method outlined in Fig. 1. This involved the condensation of 5-diethylaminosalicylaldehyde (1) with 2-(3-cyano-4,5,5-trimethyl-5H-furan-2-ylidene)-malononitrile 2 and provided the title compound in an 80% yield. (N1,C1,C2,C3,N2,C6) is planar but twisted by 5.46 (6)° with respect to the `CDFP′ group as been found in previous studies e.g. 5.69 (17)° in compound NOJKUT (Gainsford et al., 2008). The entire `backbone′ including the hydroxyl atom O2 and C2 but excluding the ethyl and methyl atoms (C8,C9,C19-C22) and the dicyano groups can be considered essentially planar with average mean plane deviation 0.019 (1) Å and maximum deviation 0.027 (1) Å for C11. This is in marked contrast to the related benzoyloxy structure KARXAE (Wu et al., 2012) ((4-(2(2-Benzoyloxy)-4-(diethylamino)phenyl)vinyl)-3-cyano-5,5 -dimethylfuran-2(5H)-ylidene) malonitrile), where the CDFP and (diethylamino)phenyl groups make interplanar angles of ~10°. The difference probably relates to the different crystal packing arrangements, as noted below, but this twist also alleviates potential close contacts between the methylene group of the benzoyloxy group and the pendant nearest ethyl group. The pendant ethyl groups in KARXAE are also in the opposite configurations, with the nearest ethyl group pointing away from the benzoyloxy moiety. The acetone molecule is disordered around a 2 fold axis, and the final concentration was decided by a thermal parameter comparison with the other atoms and confirmed by a stable final refinement.
One other related structure, without the ortho oxygen substituent at C15, is NUGNUZ (Li et al., 2009). Here the backbone is twisted along its length, with ~7° between CDFP and the polyene atoms, and a further ~ 8° between the latter and the (diethylamino)phenyl group. This twisting is probably driven by the intermolecular hydrogen bonding interactions which involve one of the polyene H atoms (to the terminal alcohol O) and a phenyl C-H···N(cyano) contact.
The crystal packing can be described as interleaved ribbons of molecules, approximately in the 1,0,1 direction, formed by the major almost in-plane hydrogen bonding (entry 1, Table 1 and Figure 3). The alternate ribbon molecule planes make a dihedral angle of ~15°. The interplanar interactions are provided by weak C-H···O(acetone) interactions (entries 2 and 3, Table 1). Given this weak interaction (and the estimated 0.25 concentration of the acetone), it is not surprising that rotational conformers are present. By contrast the packing in related structure KARXAE is a traditional herringbone pattern, with ~39° between the molecular planes; here the single weak C-H···N(cyano) and (methylene)C-H···π are sufficient for the crystal packing stability.

Refinement
The molecule is rotationally disordered about the C12-C13 bond in the ratio of 0.919 (2):0.081 (2) as determined by refining the O2-H atoms over two positions with identical thermal parameters (EADP). Four further restraints were applied to the minor rotamer atoms with C18-O2B and O2B-H2B fixed at 1.340 (5) and 0.84 (1) Å respectively, and two antibumping restraints tied to the equivalent major rotamer distances (using SADI). The final residual difference density is consistent with this ~8% presence for the remaining atoms in the rotated group; their refinement is impractical and would add nothing to the final conclusions. Four reflections affected by the backstop and 16 others, which were clearly outlier data (mostly at low angle) with Δ(F 2 )/e.s.d. > 5.0, were omitted from the refinements (using OMIT). The methyl and other H atoms were refined with U iso 1.5 and 1.2 times respectively that of the U eq of their parent atom. The hydroxyl hydrogen on major rotamer O2A was located on a difference Fourier map and its position refined. The hydroxyl hydrogen bound to the (8%) O2B atom was located via an HFIX 147 tetrahedral position refinement and refined as noted above. All H atoms bound to carbon were constrained to their expected geometries (C-H 0.95, 0.98 and 0.99 Å).

Figure 1
Chemical synthesis of the title compound 3.

Figure 2
Structure of the asymmetric unit (Farrugia, 1997) showing the atom labelling scheme and displacement ellipsoids drawn at the 30% probability level. Only the major rotamer atom (O2A) is shown (see text).   -1/2 -y, -1/2 + z (ii) 1/2 + x, -1/2 -y, 1/2 + z. 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 > σ(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.