6-[3-(2,4-Dimethylanilino)-2-hydroxypropoxy]-1,8-dihydroxy-3-methyl-9,10-dihydroanthracene-9,10-dione

In the title compound, C26H25NO6, the anthraquinone ring system forms a dihedral angle of 15.5 (1)° with the benzene ring of the dimethylaniline group. Intramolecular O—H⋯O hydrogen bonding is observed between the carbonyl and two hydroxyl groups. The molecules are linked into a ribbon-like structure along the [100] direction by O—H⋯N and C—H⋯O hydrogen bonds. The crystal used was twinned via a 180° rotation about [100]. The ratio of the two twin components is 0.947 (1):0.053 (1).

In the title compound, C 26 H 25 NO 6 , the anthraquinone ring system forms a dihedral angle of 15.5 (1) with the benzene ring of the dimethylaniline group. Intramolecular O-HÁ Á ÁO hydrogen bonding is observed between the carbonyl and two hydroxyl groups. The molecules are linked into a ribbon-like structure along the [100] direction by O-HÁ Á ÁN and C-HÁ Á ÁO hydrogen bonds. The crystal used was twinned via a 180 rotation about [100]. The ratio of the two twin components is 0.947 (1):0.053 (1).

Comment
Emodin and its derivatives have been found to possess diverse biological properties, such as antimicrobial, antiviral, antitumor, anti-inflammatory, anti-oxidant, immunosuppressive, anti-ulcerogenic, fungicidal and chemopreventive activities (Wang & Xu, 2005;Teich et al., 2004;Srinivas et al., 2003). As part of our ongoing research on emodin derivatives (Wang & Xu, 2005;Wang et al., 2006), we report here the crystal structure of the title compound.
The molecular structure of the title compound is illustrated in Fig. 1. The anthraquinone ring system is essentially planar and it forms a dihedral angle of 15.5 (1)° with the benzene ring of the dimethylaniline group. There are two intramolecular O-H···O hydrogen-bonding interactions between the carbonyl and two hydroxy groups (Fig. 1).
The molecules are linked into a ribbon-like structure along the [100] by intermolecular O-H···N and C-H···O hydrogen bonds (Table 1 and Fig.2).

Experimental
A mixture of emodin (10 mmol) and epichlorohydrin (421 mmol, 33 ml) was stirred under reflux in a solution of potassium hydroxide (10 mmol) in water (3 ml) until the disappearance of the starting material, as evidenced by thin-layer chromatography (about 4 h). After the reaction was over, the solvent was removed in vacuo and the residue was partitioned between chloroform (50 ml) and distilled water (20 ml). The organic phase was washed with water (15 ml) and brine (15 ml), and dried over anhydrous sodium sulfate. The solvent was removed to give the key intermediate, 1,8-Dihydroxy-3-methyl-6-(oxiran-2-ylmethoxy)-9,10-dihydroanthracene-9,10-dione (Wang et al., 2006) as a yellow oil, which was purified by flash chromatography (silica gel, petroleum ether-acetone 3:1). To a solution of above intermediate (0.326 g, 1 mmol) in chloroform was added 2,4-dimethylaniline (1.1 mmol). The mixture was refluxed with stirring and monitored by TLC until the reaction was completed. The crude product was purified by column chromatography (petroleum ether-acetone 3:1) to afford the title compound, which was dissolved in methanol (15 ml) and kept at room temperature for 15 d to get yellow single crystals.

Refinement
All H atoms were placed in geometrically calculated positions and refined using a riding model, with O-H = 0.82 Å, N-H = 0.86 Å and C-H = 0.93-0.98 Å. The U iso values were set at 1.2 to 1.5 (hydroxyl and methyl) times the U eq (carrier atom).
The components of the U ij parameters in the direction of the C17-O6 bond were restrained to be equal. The highest residual density peak is located 0.65 Å from atom H17. Attempts to refine this peak as a disordered O6 atom, say O6A, resulted in a very short C17-O6A distance (1.14 Å) and hence the original model was retained. The crystal used was twinned via a

Special details
Geometry. All e.s. 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 Rfactors(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.