Sergey Z Vatsadze1, Marina A Manaenkova1, Evgeny V Vasilev1, Nikolai U Venskovsky2, Victor N Khrustalev2. 1. Department of Chemistry, Moscow State University, 1 Leninskie Gory, Moscow 119991, Russian Federation. 2. Inorganic Chemistry Department, Faculty of Science, Peoples' Friendship University of Russia (RUDN University), 6 Miklukho-Maklay St., Moscow 117198, Russian Federation.
Abstract
Compounds (I), C9H10N2O4, (II), C11H12N2O4, and (III), C14H12N2O4·C3H7NO represent 9,9-disubstituted-3,7-di-aza-bicyclo-[3.3.1]nonane-2,4,6,8-tetra-one deriv-atives with very similar mol-ecular geometries for the bicyclic framework: the dihedral angle between the planes of the imide groups is 74.87 (6), 73.86 (3) and 74.83 (6)° in (I)-(III), respectively. The dimethyl derivative (I) is positioned on a crystallographic twofold axis and its overall geometry deviates only slightly from idealized C2v symmetry. The spiro-cyclo-pentane derivative (II) and the phen-yl/methyl analog (III) retain only inter-nal Cs symmetry, which in the case of (II) coincides with crystallographic mirror symmetry. The cyclo-pentane moiety in (II) adopts an envelope conformation, with the spiro C atom deviating from the mean plane of the rest of the ring by 0.548 (2) Å. In compound (III), an N-H⋯O hydrogen bond is formed with the di-methyl-formamide solvent mol-ecule. In the crystal, both (I) and (II) form similar zigzag hydrogen-bonded ribbons through double inter-molecular N-H⋯O hydrogen bonds. However, whereas in (I) the ribbons are formed by two trans-arranged O=C-N-H amide fragments, the amide fragments are cis-positioned in (II). The formation of ribbons in (III) is apparently disrupted by participation of one of its N-H groups in hydrogen bonding with the solvent mol-ecule. As a result, the mol-ecules of (III) form zigzag chains rather than the ribbons through inter-molecular N-H⋯O hydrogen bonds. The crystal of (I) was a pseudo-merohedral twin.
Compounds (I), C9H10N2O4, (II), C11H12N2O4, and (III), C14H12N2O4·C3H7NO represent 9,9-disubstituted-3,7-di-aza-bicyclo-[3.3.1]nonane-2,4,6,8-tetra-one deriv-atives with very similar mol-ecular geometries for the bicyclic framework: the dihedral angle between the planes of the imide groups is 74.87 (6), 73.86 (3) and 74.83 (6)° in (I)-(III), respectively. The dimethyl derivative (I) is positioned on a crystallographic twofold axis and its overall geometry deviates only slightly from idealized C2v symmetry. The spiro-cyclo-pentane derivative (II) and the phen-yl/methyl analog (III) retain only inter-nalCs symmetry, which in the case of (II) coincides with crystallographic mirror symmetry. The cyclo-pentane moiety in (II) adopts an envelope conformation, with the spiro C atom deviating from the mean plane of the rest of the ring by 0.548 (2) Å. In compound (III), an N-H⋯O hydrogen bond is formed with the di-methyl-formamide solvent mol-ecule. In the crystal, both (I) and (II) form similar zigzag hydrogen-bonded ribbons through double inter-molecular N-H⋯O hydrogen bonds. However, whereas in (I) the ribbons are formed by two trans-arranged O=C-N-H amide fragments, the amide fragments are cis-positioned in (II). The formation of ribbons in (III) is apparently disrupted by participation of one of its N-H groups in hydrogen bonding with the solvent mol-ecule. As a result, the mol-ecules of (III) form zigzag chains rather than the ribbons through inter-molecular N-H⋯O hydrogen bonds. The crystal of (I) was a pseudo-merohedral twin.
Diazabicyclononane-tetraones are used in the synthesis of the sparteine subgroup of lupine alcaloids (Norcross et al., 2008 ▸) and are precursors in obtaining 3,7-diazabicyclo[3.3.1]nonanes which have been studied in computer models as serine protease inhibitors (Vatsadze et al., 2016 ▸). They also have value as building blocks in the design of other biologically active compounds (Kudryavtsev et al., 2014 ▸), and in the synthesis of imaging agents for positron emission tomography (Medved’ko et al., 2016 ▸). In addition, they are good chelating ligands for 3d transition metals (Vatsadze et al., 2005 ▸) including Cu (Vatsadze et al., 2014 ▸).However, the crystal structures of this class of compounds have not been adequately characterized so far, as shown by a small number (eight) of similar structures found in the Cambridge Structural Database (CSD; Groom et al., 2016 ▸). Moreover, their ability to form different supramolecular structures depending on the substituents at the 9-position in the heterocycle, which we report in this work, has not been not reported before. A search in the CSD for the substructure 3,7-diaza-2,4,6,8-tetraoxobicyclo[3.3.1]nonane yielded eight hits. Although there is a similarity in chemical structure of known related compounds (Horlein et al., 1981 ▸; Norcross et al., 2008 ▸), their supramolecular features are significantly different because of the impact of substituents and solvatation.In this work, we have synthesized three 9,9-disubstituted-3,7-diazabicyclo[3.3.1]nonane-2,4,6,8-tetraones and show how groups bound to C9 as well as the presence of solvate molecules affect their ability to form different hydrogen-bonding systems.
Structural commentary
Compounds (I), C9H10N2O4, (II), C11H12N2O4, and (III), C14H12N2O4·C3H7NO represent 9,9-disubstituted-3,7-diazabicyclo[3.3.1]nonane-2,4,6,8-tetraone derivatives and have very similar molecular geometries (Figs. 1 ▸–3 ▸
▸). In general, the 3,7-diazabicyclo[3.3.1]nonane-2,4,6,8-tetraone skeleton exhibits idealized C
2v (mm2) symmetry. The molecule of (I), containing two 9-methyl substituents, occupies a special position on a twofold axis [C
2 (2)], and its geometry deviates only slightly from the perfectly symmetrical C
2v. As a result of the presence of spiro-9-cyclopentane [in the case of (II)] and 9-phenyl and 9-methyl [in the case of (III)] substituents, the overall symmetry of these molecules decreases to C
s (m). However, in the crystal, the intrinsic C
s symmetry remains only for the molecule of (II), which occupies a special position on a mirror plane. Compound (III) crystallizes as a dimethyl formamide monosolvate, with the main molecule occupying a general position.
Figure 1
The molecular structure of (I). Displacement ellipsoids are shown at the 50% probability level. H atoms are presented as small spheres of arbitrary radius. [Symmetry code: (A) 1 + x, y, −z + .]
Figure 2
The molecular structure of (II). Displacement ellipsoids are shown at the 50% probability level. H atoms are presented as small spheres of arbitrary radius. [Symmetry code: (A) x, − y, z.]
Figure 3
The molecular structure of (III)·DMF. Displacement ellipsoids are shown at the 50% probability level. H atoms are presented as small spheres of arbitrary radius. Dashed line indicates the intramolecular N—H⋯O hydrogen bond.
The two imide fragments in the molecules of (I)–(III) are almost planar (r.m.s. deviations are 0.013, 0.009 and 0.009/0.036 Å, respectively). The dihedral angles between the imide planes are 74.87 (6), 73.86 (3) and 74.83 (6)° for (I)–(III), respectively. Moreover, the four carbonyl carbon atoms in (I)–(III) are each coplanar with r.m.s. deviations of 0.018, 0.000, and 0.031 Å, respectively; the bridged carbon atom lies by 1.854 (3), 1.846 (1), and 1.858 (2) Å, respectively, above this plane in (I)–(III). The cyclopentane substituent in (II) adopts an envelope conformation, with the C6 spiro-carbon atom deviating from the mean plane through the other ring atoms by 0.548 (2) Å.Importantly, in (III) the main molecule forms a strong N7—H7⋯O5 hydrogen bond with the dimethyl formamide solvate molecule (Table 3 ▸, Fig. 3 ▸).
Table 3
Hydrogen-bond geometry (Å, °) for (III)
D—H⋯A
D—H
H⋯A
D⋯A
D—H⋯A
N3—H3⋯O1i
0.861 (19)
2.12 (2)
2.9650 (19)
168.3 (18)
N7—H7⋯O5
0.90 (2)
1.86 (2)
2.7682 (19)
178.6 (18)
Symmetry code: (i) .
Supramolecular features
In general, any compound of type (I)–(III) could form up to six intermolecular hydrogen bonds utilizing two hydrogen-bond donor NH groups and four hydrogen-bond acceptor carbonyl oxygen atoms. In the literature, even the unsubstituted analogue (refcode GOHHER; Norcross et al., 2008 ▸) shows only four intermolecular hydrogen bonds involving both imide fragments of bispidintetraone with the formation of an infinite three-dimensional hydrogen-bonded network. If one of the nitrogen atoms is alkylated (for example, refcode BAHFIZ; Horlein et al., 1981 ▸), the other one is involved in the formation of a doubly hydrogen-bonded dimer. When both nitrogen atoms are functionalized [refcodes JIMWUY (Hametner et al., 2007 ▸), NAWLIH (Mereiter et al., 2014 ▸), NAWLON et al., 2014 ▸), PILXAK (Hametner et al., 2007 ▸), XAZGAH (Blakemore, et al., 2005 ▸)], no hydrogen-bonds are observed.Despite the geometrical similarity of compounds (I)-(III), they form different supramolecular structures in the solid state. Thus, in the crystals of (I) and (II), the molecules form the zigzag hydrogen-bonded ribbons by double N—H⋯O hydrogen bonds (Tables 1 ▸ and 2 ▸, Figs. 4 ▸ and 5 ▸). The hydrogen-bonded ribbons in (I) and (II) are distinguished by the binding sites of the 3,7-diazabicyclo[3.3.1]nonane-2,4,6,8-tetraone skeleton. According to symmetry, the ribbons in (I) are formed by the two trans-arranged O=C—N—H amide fragments, whereas the binding O=C—N—H amide fragments in (II) are cis disposed. As one of the two NH groups in (III) is bonded to the dimethyl formamide solvate molecule, the N—H⋯O hydrogen bonds form the zigzag chains rather than ribbons (Table 3 ▸, Fig. 6 ▸).
Table 1
Hydrogen-bond geometry (Å, °) for (I)
D—H⋯A
D—H
H⋯A
D⋯A
D—H⋯A
N3—H3⋯O2i
0.90 (3)
2.01 (3)
2.906 (2)
173 (3)
Symmetry code: (i) .
Table 2
Hydrogen-bond geometry (Å, °) for (II)
D—H⋯A
D—H
H⋯A
D⋯A
D—H⋯A
N3—H3⋯O1i
0.855 (14)
2.021 (14)
2.8718 (11)
173.7 (13)
Symmetry code: (i) .
Figure 4
The crystal structure of (I), demonstrating the H-bonded zigzag-like ribbons propagating toward [001]. Dashed lines indicate the intermolecular N—H⋯O hydrogen bonds.
Figure 5
The crystal structure of (II), demonstrating the H-bonded zigzag-like ribbons propagating toward [010]. Dashed lines indicate the intermolecular N—H⋯O hydrogen bonds.
Figure 6
The crystal structure of (III)·DMF, demonstrating the H-bonded zigzag-like chains propagating toward [100]. Dashed lines indicate the intermolecular N—H⋯O hydrogen bonds.
Synthesis and crystallization
The title compounds (I)–(III) were synthesized (Fig. 7 ▸) according to the procedure described earlier (Schon et al., 1998 ▸).
Figure 7
Synthesis of (I)–(III) from 2-cyanoacetamide and ketones.
Dinitrile subproducts were obtained by adding 2-cyanoacetamide to the corresponding ketone [(I) – acetone, (II) – acetophenone, (III) – cyclopentanone] in ethanol at room temperature. Then, the dinitriles were heated to 393–413 K upon stirring in an acidic medium to complete dissolving. After 10–15 min, the mixture was poured into ice–water. The precipitated tetraoxo-compounds were filtered off by suction, recrystallized from ethanol solution and finally dried. Single crystals suitable for X-ray diffraction study were obtained by recrystallization of the crude products from DMF solution.
Refinement
Crystal data, data collection and structure refinement details are summarized in Table 4 ▸. The hydrogen atoms of the amino groups were localized in the difference-Fourier maps and refined isotropically with fixed displacement parameters [U
iso(H) = 1.2U
eq(N)]. The other hydrogen atoms were placed in calculated positions with C—H = 0.95–1.00 Å and refined in the riding/rotating model with fixed isotropic displacement parameters [U
iso(H) = 1.5U
eq(C) for the CH3-groups and 1.2U
eq(C) for the other groups]. The crystal of (I) was a pseudo-merohedral twin. The twin matrix is ( 0 0 0 0 0.775 0 1), and BASF = 0.180 (1).
Table 4
Experimental details
(I)
(II)
(III)
Crystal data
Chemical formula
C9H10N2O4
C11H12N2O4
C14H12N2O4·C3H7NO
Mr
210.19
236.23
345.35
Crystal system, space group
Monoclinic, C2/c
Orthorhombic, Pnma
Orthorhombic, Pbca
Temperature (K)
100
120
120
a, b, c (Å)
11.4321 (17), 6.6263 (10), 12.4819 (19)
12.8058 (6), 11.4850 (6), 6.9058 (3)
7.7876 (5), 19.4656 (12), 21.7879 (13)
α, β, γ (°)
90, 110.788 (3), 90
90, 90, 90
90, 90, 90
V (Å3)
884.0 (2)
1015.67 (8)
3302.8 (4)
Z
4
4
8
Radiation type
Mo Kα
Mo Kα
Mo Kα
μ (mm−1)
0.13
0.12
0.10
Crystal size (mm)
0.30 × 0.20 × 0.15
0.30 × 0.20 × 0.20
0.22 × 0.20 × 0.18
Data collection
Diffractometer
Bruker SMART 1K CCD
Bruker SMART 1K CCD
Bruker SMART 1K CCD
Absorption correction
Multi-scan (SADABS; Sheldrick, 2003 ▸)
Multi-scan (SADABS; Sheldrick, 2003 ▸)
Multi-scan (SADABS; Sheldrick, 2003 ▸)
Tmin, Tmax
0.950, 0.970
0.960, 0.970
0.970, 0.975
No. of measured, independent and observed [I > 2σ(I)] reflections
4993, 1289, 1165
15297, 2181, 1782
41691, 5056, 3210
Rint
0.027
0.031
0.090
(sin θ/λ)max (Å−1)
0.703
0.802
0.716
Refinement
R[F2 > 2σ(F2)], wR(F2), S
0.062, 0.184, 1.06
0.043, 0.119, 1.03
0.051, 0.124, 1.01
No. of reflections
1289
2181
5056
No. of parameters
74
85
235
H-atom treatment
H atoms treated by a mixture of independent and constrained refinement
H atoms treated by a mixture of independent and constrained refinement
H atoms treated by a mixture of independent and constrained refinement
Crystal structure: contains datablock(s) global, I, II, III. DOI: 10.1107/S2056989017009458/ld2140sup1.cifStructure factors: contains datablock(s) I. DOI: 10.1107/S2056989017009458/ld2140Isup2.hklStructure factors: contains datablock(s) II. DOI: 10.1107/S2056989017009458/ld2140IIsup3.hklStructure factors: contains datablock(s) III. DOI: 10.1107/S2056989017009458/ld2140IIIsup4.hklClick here for additional data file.Supporting information file. DOI: 10.1107/S2056989017009458/ld2140Isup5.cmlClick here for additional data file.Supporting information file. DOI: 10.1107/S2056989017009458/ld2140IIsup6.cmlClick here for additional data file.Supporting information file. DOI: 10.1107/S2056989017009458/ld2140IIIsup7.cmlCCDC references: 1558317, 1558316, 1558315Additional supporting information: crystallographic information; 3D view; checkCIF report
Primary atom site location: difference Fourier map
Least-squares matrix: full
Secondary atom site location: difference Fourier map
R[F2 > 2σ(F2)] = 0.062
Hydrogen site location: mixed
wR(F2) = 0.184
H atoms treated by a mixture of independent and constrained refinement
S = 1.06
w = 1/[σ2(Fo2) + (0.0846P)2 + 3.9571P] where P = (Fo2 + 2Fc2)/3
1289 reflections
(Δ/σ)max < 0.001
74 parameters
Δρmax = 0.55 e Å−3
0 restraints
Δρmin = −0.54 e Å−3
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.
Primary atom site location: difference Fourier map
Least-squares matrix: full
Secondary atom site location: difference Fourier map
R[F2 > 2σ(F2)] = 0.043
Hydrogen site location: mixed
wR(F2) = 0.119
H atoms treated by a mixture of independent and constrained refinement
S = 1.03
w = 1/[σ2(Fo2) + (0.0631P)2 + 0.338P] where P = (Fo2 + 2Fc2)/3
2181 reflections
(Δ/σ)max < 0.001
85 parameters
Δρmax = 0.42 e Å−3
0 restraints
Δρmin = −0.23 e Å−3
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.
Primary atom site location: difference Fourier map
Least-squares matrix: full
Secondary atom site location: difference Fourier map
R[F2 > 2σ(F2)] = 0.051
Hydrogen site location: mixed
wR(F2) = 0.124
H atoms treated by a mixture of independent and constrained refinement
S = 1.01
w = 1/[σ2(Fo2) + (0.0448P)2 + 1.1286P] where P = (Fo2 + 2Fc2)/3
5056 reflections
(Δ/σ)max < 0.001
235 parameters
Δρmax = 0.33 e Å−3
0 restraints
Δρmin = −0.26 e Å−3
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.