Abstract

The structure of Poly[(acetone-O)-3-((3,4-dimethoxyphenyl)(4-hydroxy-2-oxo-2H-chromen-3-yl)methyl)-(2-oxo-2H-chromen-4-olate)sodium] was determined by X-ray crystallography. The compound crystallizes in an orthorhombic system and was characterized thus P     , (2) Å, (3) Å, Å. , (10) Å3. The crystal structure was solved by direct methods and refined by full-matrix least-squares on to final values of and .

Biscoumarin derivatives possess anticoagulant, spasmolytic, bacteriostatic, and rodenticidal activities. Some of them can be used as herbicides. By chemical modifications (different substituents on the aromatic ring) it is possible to obtain a compound with good biological activity, but with lower toxicity and fewer side effects.

The title compound was synthesized from 3,3′-[(3,4-dimethoxyphenyl)-methylidene]-bis(4-hydroxy-2H-chromen-2-one) and water solution of sodium hydroxide at a molar ratio. This compound showed an effect on HIV replication in acutely infected cells by microtiter infection assay. The same substance demonstrated no impact on early stages of HIV-1 replication cycle [1]. The transformation of the compound to sodium salt was a stage for synthesizing complex compounds with lanthanides.

We only succeeded in growing colourless thin needles for single-crystal X-ray diffraction analysis by slow evaporation of an ethanol/acetone solution. Crystallographic data collected at room temperature with an Enraf-Nonius KappaCCD diffractometer using graphite monochromated Mo-   (  Å) radiation were therefore of limited diffraction quality (Table 1). The solid state structure of the molecule was nonetheless investigated satisfactorily from a chemical/crystallographical point of view.

Table 1 
Crystal data and structure refinement for Compound 1.

Crystal unit-cell and orientation parameters were determined by the DENZO [1] auto indexing procedure, as implemented in the data collection monitoring program COLLECT [2]. Intensities recorded up to a diffraction angle, , of 22.1° were also integrated by DENZO, scaled, and then reduced using SCALEPACK-HKL2000 [2], after postrefinement of the unit-cell parameters and absorption correction based on symmetry-equivalent and repeated reflections. The structure was solved by direct methods using SIR97 [3], and all of the nonhydrogen atoms were refined anisotropically by full-matrix least-squares on using SHELXL97 [4]. All hydrogen atoms were located in difference electron-density maps, but refined as riding, with C–H = 0.93, 0.96, 0.97, and 0.98 Å for the aromatic, methyl, and methyne H atoms, respectively, O–H = 0.82 Å for hydroxyl H atoms, and with (H) (C) or 1.5 (methyl C). Crystallographic data and details of the data collection and structure refinements are listed in Table 1. The observed anisotropic thermal parameters, the calculated structure factors, and full lists of the bond distances, bond angles, torsion angles, and intermolecular H-bond interactions are given as supplementary material (Tables 2, 3, 4, 5, 6, 7, and 8). The bond lengths and bond angles are all within the expected ranges.

Table 2 
Atomic coordinates and equivalent isotropic displacement parameters (A2 ) for Compound 1. U(eq) is defined as one third of the trace of the orthogonalized tensor.
Table 3 
Bond lengths (Å) for Compound 1.
Table 4 
Bond angles (°) for Compound 1.
Table 5 
Anisotropic displacement parameters (A2 103) for Compound 1. The anisotropic displacement factor exponent takes the form: .
Table 6 
Hydrogen coordinates ( 104) and isotropic displacement parameters (Å2 103) for Compound 1.
Table 7 
Torsion angles (°) for Compound 1.
Table 8 
Hydrogen bonds for Compound 1 (Å and °).

The X-ray crystal structure of is formally ionic, containing an anionic biscoumarin consisting in two 4-hydroxycoumarin moieties (one of it with a deprotonated hydroxyl group) linked through a methylene bridge on which one hydrogen has been replaced by a dimethoxyphenyl residue, Na+ ions, and an acetone molecule. However, the structure could be better described as basic fragments of formula forming polymeric chains along the a axis with a Na1 Na1i separation of 9.967(2) Å (Figure 2(a)). These chains are interconnected through aromatic - stacking interactions involving the methoxyphenyl group and one coumarin group at position , , , with Cg1 Cg5(Cg1 = centroid of the O1–C2–C3–C4–C4A–C8A six-membered ring and Cg5 = centroid of the C11–C16 six-membered ring)distance of 3.634 Å, and C–H weak interactions (Table 8), generating a two-dimensional layer architecture parallel to the crystallographic ab plane (Figure 2(b)), and placing the Na1ii at distances from Na1 or Na1i in the range of 7.426–9.143 Å [symmetry codes: (i) , , ; (ii) , , ]. The layered assembly is merely consolidated in the third-dimension by even weaker C–H interactions (e.g., methyl groups of acetone ligands and coumarin moieties from adjacent layers) (Figure 2(c)).


Figure 1 
(a)
(a)
(b)
(b)
(c)
(c)
Figure 2 
(a) View of the polymeric chain propagating along the direction. (b) Bidimensional array parallel to the plane. (c) View of the crystal structure down the axis.

Figure 3 
A view of the coordination sphere around the Na+ ion in compound 1 with 30% probability displacement ellipsoids is displayed with the numbering scheme. The complete coordination of the Na atom is shown. Symmetry-related atoms are shown in transparency with symmetry codes: i: , , , ii: , , .

The Na1 cation is coordinated by six O atoms in a distorted octahedral geometry, with five oxygen atoms (O2, O2′, O4i, O5i, O3ii) from three molecules of biscoumarin and one O7 from an acetone molecule. The first biscoumarin molecule chelates to the sodium atom through the two atoms from the oxo units, the second molecule through two O atoms from the two methoxy groups of the 3,4-dimethoxyphenyl moiety whereas the third molecule through the hydroxyl atom O3ii. The six oxygen atoms are at distances from the cation in the range of 2.267(7)–2.774(6) Å, the longest distance being observed with the hydroxyl oxygen atom.

Unlike nonionic structures of biscoumarin compounds [5–7] for which the two 4-hydroxycoumarin moieties are also intramolecularly hydrogen bonded between hydroxyls and carbonyls, the coumarin residues here are arranged in such a way that hydroxyl O3 and O3′ atoms are brought close enough to form an intramolecular hydrogen bond. This feature seems characteristic of biscoumarin structures with a deprotonated hydroxyl since it was previously noted with the following salt structure, C5H12N+ C29H23 [8]. Limited crystallographic data resolution and long hydroxyl C–O bond lengths >1.3 Å are in favour of a 50/50 donor/attractor character in both residues. For the sake of the model refinement, O3 has been chosen to act as the donor (Table 8). Otherwise, the geometric parameters of the biscoumarin agree with closely related structures [5–8]. All of the twelve non-H atoms of the coumarin rings are essentially coplanar, with r.m.s deviation of 0.032 and 0.056 Å, respectively. The plane of the dimethoxyphenyl ring is inclined at angles of 78.49(19)° and 67.17(18)° to the coumarin moieties. The dihedral angle between the two coumarin moieties is 58.65(16)°. The orientations of the coumarins about C1 may be described in terms of the torsion angles C3–C1–C3′–C4′ of −72.1(10)°, and C4–C3–C1–C3′ of 83.2(9)°. The bond angles C3′–C1–C3, 115.1(6), C3–C1–C16, 113.7(6), and C3′–C1–C16, 115.4(6)° at C1 are also widened in comparison with standard tetrahedral values. Steric crowding around this atom may be invoked to explain this feature, as well as in the case of the C1–C16 distance of 1.541(11) Å, longer as expected than an unstrained Csp2–Car bond [5]. The exocyclic bond angles at C3 [C2–C3–C1, 114.9(7)°, and C4–C3–C1, 124.4(7)°] and those at C3′ [C2′–C3′–C1, 116.8(6)° and C4′–C3′–C1, 122.4(7)°] do not differ very significantly (9.5 and 5.6°, resp.) in comparison with dicoumarols [5].

Acknowledgments

Financial support from the Ministry of Education and Science-Sofia, Bulgaria through Project No. DO 02-129/2008 is acknowledged.