3-(1-Methylpyrrolidin-2-ylidene)-3H-indole sesquihydrate

The asymmetric unit of the title compound, C13H14N2·1.5H2O, contains two similar molecules of 3-(1-methylpyrrolidin-2-ylidene)-3H-indole, (I), and three water molecules. (I) is the product of reacting indole with 1-methylpyrrolidin-2-one in the presence of phosphorus oxychloride. Both organic molecules are almost completely planar; the maximum distances above and below the least-squares plane through all the atoms of molecule 1 are 0.050 (8) and −0.045 (8) Å, respectively, and the deviations for molecule 2 are 0.096 (8) and −0.059 (8) Å, respectively. In the crystal, the two crystallographically different molecules alternate in π-stacked columns [centroid–centroid distances = 3.729 (5) and 3.858 (5) Å], which are linked by O—H⋯N hydrogen bonds to a network of hydrogen-bonded water molecules. O—H⋯O interactions are also present.


Comment
Vilsmeier reactions of indole using a tertiary amide in combination with phosphorus oxychloride give rise to 3-acylindoles corresponding to the acyl residue of the amide. However, when 1-methylpyrrolidin-2-one is used as the amide component, et al., 1964). We have discussed this structure and studied its strong basicity (Harris & Joule, 1978) and other reactions (Bishop et al., 1981. Bishop et al., 1982. Al-Khawaja et al., 1984. In order to further understand the chemical reactivity of (1) we felt it was important to crystallographically establish the planar structure, predicted by resonance contributor (2) The asymmetric unit of (1) contains two similar molecules of (1), together with three water molecules. The two molecules are essentially planar; the maximum distances above and below the least squares plane through all the atoms of molecule 1 are 0.050 (8) and -0.045 (8) Å for atoms C12 and C11, respectively; the maximum distances above and below the least squares plane through all the atoms of molecule 2 are 0.096 (8) and -0.059 (8) Å, for atoms C24 and C18, respectively.
The two ring systems linked by a double bond in each molecule of (1), are essentially coplanar, with dihedral angles of 0.7 (2) between the atoms C1-C8/N1 and C9-C12/N2 and 2.7 (2)° between the atoms C14-C21/N3 and C22-C25/N4. The conjugation between the two nitrogen atoms, as illustrated by (2) is reflected in the bond lengths: in particular, the C8-N1 double bond at 1.311 (8) Å is long for a double bond between carbon and nitrogen, and the C9-N2 single bond at 1.316 (8) Å is correspondingly short, and almost identical in length to that of the nitrogen-carbon double bond; for molecule 2, the respective corresponding distances are 1.320 (8) and 1.322 (8) Å, for C21-N3 and C22-N4. One intriguing aspect of the crystal packing is shown in Figure 2. The two crystallographically different molecules lie one above the other and are parallel to one another, with a dihedral angle between the molecules of 0.6 (1) °. The molecules stack in such a way as to locate the positively charged end of the resonating system (cf. (2)) over the negatively charged end of the system in the second molecule. Each molecule forms π-stacking interactions with the crystallographically different molecule above and below it, to form columns along a ( Figure 2); the perpendicular distance between the ring N2/C9-C12 in molecule 1 to the ring C14-C19 in molecule 2 is 3.404 Å, with a centroid to centroid distance of 3.729 (5) Å (symmetry equivalent x, y, z), whilst the perpendicular distance of the N2/C9-C12 ring in molecule 1 to the C14-C19 in molecule 2 on the other side of molecule 1 is 3.461 Å, with a centroid to centroid distance of 3.858 (5) Å (symmetry equivalent 1 + x, y, z). Between the π-stack columns of molecules 1 and 2 is a network of H-bonded water molecules, which link the columns together via hydrogen bonds between O2S and O3S, and N1 and N3, respectively (see Table 1).

Experimental
To 1-methyl-2-pyrrolidinone (4 mL, 0.04 mol) cooled in an ice bath was added phosphorous oxychloride (4.08 g, 0.026 mol) with stirring during 30 min. The temperature did not exceed 288 K. The mixture was stirred for an additional 10 min, and then a solution of indole (2.80 g, 0.024 mol) in 1-methyl-2-pyrrolidinone (4 mL) was added slowly during 1 h. The temperature rose to 318 K and a solid separated. The mixture was heated at 353 K for 3 h, and then mixed with water sup-2 (100 mL). The clear solution was made basic by the addition of sodium hydroxide (6 g) in water (30 mL) causing a solid to separate. The solid was filtered off and washed with water. Recrystallization from ethanol-water afforded 3-(1-methylpyrrolidin-2-ylidene)-3H-indole (4.21 g, 90 percent), m.p. 491-492 K.

Refinement
H atoms bonded to the C atoms were fixed geometrically and treated as riding with C-H = 0.95Å (aromatic), 0.98Å (methyl) and 0.99Å (methylene), with Uiso(H) = 1.2 times those of the parent atoms for the aromatic and methylene H atoms and Uiso(H) = 1.5 times those of the parent atoms for the methyl H atoms. Restraints were applied to the geometry of the water molecules and to the anisotropic thermal parameters of the non-H atoms. The crystal diffracted very weakly, so the data were cut at 1.1 Å resolution. Fig. 1. Plot showing the two crystallographically independent molecules of (1), with 50% probability ellipsoids.

Special details
Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s 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 > 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.   (2)  supplementary materials sup-9