Polymorphs of gabapentin

Gabapentin [or 1-(amino­meth­yl)cyclo­hexaneacetic acid], C₉H₁₇NO₂, exists as a zwitterion [1-(ammonio­meth­yl)cyclo­hexaneacetate] in the solid state. The crystal structures and bonding networks of two new monoclinic polymorphs ([beta]-gabapentin and [gamma]-gabapentin) are studied and compared with a previously reported gabapentin polymorph [[alpha]-gabapentin: Ibers (2001). Acta Cryst. C57, 641-643]. All three polymorphs have extensive networks of hydrogen bonds between the NH₃⁺ and COO^- groups of neighbouring mol­ecules. In [beta]-gabapentin, there is an additional weak intra­molecular hydrogen bond.

S1. Comment

Gabapentin (C₉H₁₇NO₂) is structurally related to the neurotransmitter γ-aminobutyric acid (GABA), which has been widely studied for its significant inhibitory action in the central nervous system (Bowery, 1993). There have also been studies on the cocrystallization of GABA with various carboxylic acids as a means of possibly improving the effectiveness of GABA (Wenger & Bernstein, 2006).

In recent years, there has been intense interest in the polymorphs of gabapentin and in the synthesis of gabapentin analogues. Gabapentin was originally used as an antiepileptic drug, but its applications have been extended to the treatment of neuropathic pain (Magnus, 1999). One form of gabapentin (α-gabapentin) crystallizes as a zwitterion in space group P2₁/c [a = 5.8759 (6) Å, b = 6.9189 (7) Å, c = 22.262 (2) Å, β = 90.080 (2)° and V = 905.173 ų; Ibers, 2001]. Note that α-gabapentin corresponds to form II of Lladó et al. (2003) and to the form of gabapentin found in the commercially available Pfizer pharmaceutical, Neurontin.

The structures of a gabapentin monohydrate (Ibers, 2001), several gabapentin derivatives (Ananda et al., 2003), peptides incorporating gabapentin (Vasudev et al., 2007) and a related 1-(aminomethyl) structure called pregabalin (Venu et al., 2007) have also been reported. The crystal structures of two previously unreported gabapentin polymorphs, β-gabapentin, (I) (Fig. 1), and γ-gabapentin, (II) (Fig. 2), are presented here. X-ray powder diffraction (XRPD) studies show that β-gabapentin corresponds to form III in of Pesachovich et al. (2001) and γ-gabapentin corresponds to form IV of Satyanarayana et al. (2004).

As in the case of α-gabapentin, β- and γ-gabapentin also exist as zwitterions. In all three forms the cyclohexane ring has a chair conformation. However, as shown in Figs. 3–5, each of the three polymorphs has different orientations of the NH₃⁺ and COO^- groups. Significantly, in β-gabapentin, the NH₃⁺ group is in an equatorial position and the COO^- group in an axial position, while in the α- and γ-gabapentin polymorphs, these groups are in the opposite positions. This difference is further indicated by the torsion angle C3—C1—C5—C7 (Table 2). There is an interesting difference in the rotation of the amine group, which is in an unusual eclipsed conformation for the α form and in a normal staggered conformation for the β and γ forms. These differences are quantified by the torsion angles O1—C2—C3—C1, C2—C3—C1—C4 and C3—C1—C4—N1 (Table 2), showing the different angles and directions in which the NH₃⁺ and COO^- groups are twisted (see Figs. 3–5). The torsion angles in Table 2 indicate that the largest conformational difference is due to the twisting of the COO^- group.

In β- and γ-gabapentin, there is extensive hydrogen bonding between the NH₃⁺ and COO^- groups of neighbouring molecules. The NH₃⁺ group hydrogen bonds to three COO^- groups, each from a different neighbouring molecule. The geometry of hydrogen-bonding interactions for the three polymorphs of gabapentin are compared in Table 3. There is a significant difference in the hydrogen-bonding networks of α- and β-gabapentin, and this is primarily due to the presence of an intramolecular hydrogen bond in β-gabapentin [N1(H15)···O1 = 2.940 (2) Å]. Atom O1 acts as a trifurcated hydrogen-bond acceptor in β-gabapentin as opposed to the bifurcated acceptor in α-gabapentin. There is a corresponding lengthening of the N1(H15)···O1ⁱ hydrogen bond between two neighbouring molecules [i.e. N1(H15)···O1ⁱ = 2.783 (2) and 3.024 (2) Å for α- and β-gabapentin, respectively; symmetry codes: (i) -x + 1, -y + 1, -z + 1; -x, -y, -z + 1, respectively].

As with α-gabapentin, there is no intramolecular hydrogen bond in γ-gabapentin. The N···O hydrogen-bond distances are very similar in α- and γ-gabapentin [average N1···O = 2.763 and 2.771 Å for α- and γ-gabapentin, respectively]. In both the α and γ polymorphs, atom O1 acts as a bifurcated hydrogen-bond acceptor. The differences in the packing of gabapentin in the three polymorphs is evident from Fig. 6, which shows their hydrogen-bonding networks and the graph sets (Bernstein et al., 1995) of these networks. The hydrogen-bonded chains and rings are shown along the b axis of the α polymorph (Fig. 6a) and along the c axes of the β and γ polymorphs (Figs. 6b and 6c). There are four graph sets common to each polymorph, i.e. two unitary-level graph sets, C¹₁(7) and R²₂(14), and two secondary-level graph sets, C²₂(14) and C²₂(6). The centrosymmetric hydrogen-bonded dimers shown in Figs. 3–5 all have the R²₂(14) graph set. Only the γ polymorph has the unitary-level S(7) graph set for the intramolecular interaction.

A comparison of the densities (1.257, 1.247 and 1.216 kg m^-3) and packing efficiencies (71.3, 70.5 and 68.7%) for the three gabapentin polymorphs shows that the molecules are most efficiently packed in α-gabapentin and suggests, as shown by preliminary differential scanning calorimetry (DSC) sudies, that it is the thermodynamically most stable form. Hot stage microscopy and DSC (see supplementary material, Figs. 7 and 8) were used to measure the melting points of the three polymorphs, which occur over a broad temperature range. The melting points (peak positions and range) are 434.0 (428–439), 439.28 (428–441) and 436.80 (423–441) K for the α-, β- and γ-gabapentin forms, respectively. However, DSC studies also show that possible phase transitions occur at temperatures between 358 and 368 K in both the β (small endotherm) and the γ polymorphs (small exotherm), implying that the final melting points refer to only one thermodynamically stable polymorph. Further investigations of these phase transformations are currently being undertaken (Levendis & Reece, 2007). A fourth polymorph has been reported in the patent literature, referred to as form I by Lladó et al. (2003). However single crystals of this polymorph have not yet been isolated.

Related literature

For related literature, see: Ananda et al. (2003); Bowery (1993); Ibers (2001); Levendis & Reece (2007); Lladó et al. (2003); Magnus (1999); Pesachovich et al. (2001); Satyanarayana et al. (2004); Vasudev et al. (2007); Venu et al. (2007); Wenger & Bernstein (2006).

S2. Experimental

Gabapentin purchased from Sigma–Aldrich was found by XRPD to correspond to the α polymorph. The β-gabapentin was prepared by dissolving the commercially available gabapentin (Sigma–Aldrich) in 96% ethanol until the solution was saturated. The solution was then heated at 333 K for 48 h. Needle-like crystals of β-gabapentin were grown from the solution at 333 K. In the preparation of γ-gabapentin, α-gabapentin was dissolved in 96% ethanol at 333 K. The solution was then cooled to room temperature. Plate-like crystals of γ-gabapentin formed after 3 d of slow evaporation (see supplementary material for photographs of the β and γ crystals; Fig. 7). In each case, crystals suitable for single-crystal X-ray diffraction were selected directly from the samples. A comparison of the experimental and calculated XRPD of each of the bulk samples confirmed that the bulk solid was, in fact, a pure form of the polymorph studied by single-crystal X-ray diffraction.

S3. Refinement

H atoms were positioned geometrically and allowed to ride on their respective parent atoms.

Computing details

For both compounds, data collection: APEX2 (Bruker, 2005); cell refinement: SAINT-NT (Bruker, 2005); data reduction: SAINT-NT (Bruker, 2005); program(s) used to solve structure: SHELXTL (Bruker, 1999); program(s) used to refine structure: SHELXTL (Bruker, 1999); molecular graphics: SHELXTL (Bruker, 1999) and DIAMOND (Brandenburg, 2004); software used to prepare material for publication: SHELXTL (Bruker, 1999), ORTEP-3 (Farrugia, 1997) and PLATON (Spek, 2003).

Figures

	Fig. 1. The atomic numbering scheme of β-gabapentin. Displacement ellipsoids are drawn at the 50% probability level.
	Fig. 2. The atomic numbering scheme of γ-gabapentin. Displacement ellipsoids are drawn at the 50% probability level.
	Fig. 3. A view of the R^22(14) centrosymmetric hydrogen-bonded dimer in α-gabapentin.
	Fig. 4. A view of the R^22(14) centrosymmetric hydrogen-bonded dimer in β-gabapentin.
	Fig. 5. A view of the R^22(14) centrosymmetric hydrogen-bonded dimer in γ-gabapentin.
	Fig. 6. The hydrogen-bonding networks and associated unitary- and secondary-level graph sets for α-, β- and γ-gabapentin. The α-form (left) is viewed down the a axis. The β (centre) and γ (right) forms are viewed down the b axis.

(I) 1-(ammoniomethyl)cycloheaxaneacetate

Crystal data

C9H17NO2	F(000) = 376
Mr = 171.24	Dx = 1.247 Mg m^−3
Monoclinic, P21/c	Melting point: 439 K
Hall symbol: -P 2ybc	Mo Kα radiation, λ = 0.71073 Å
a = 14.5376 (16) Å	Cell parameters from 868 reflections
b = 6.6329 (6) Å	θ = 2.9–23.7°
c = 9.8343 (9) Å	µ = 0.09 mm^−1
β = 105.922 (5)°	T = 173 K
V = 911.91 (15) Å^3	Needle, colourless
Z = 4	0.58 × 0.10 × 0.02 mm

Data collection

Bruker APEXII CCD detector diffractometer	1168 reflections with I > 2σ(I)
Radiation source: fine-focus sealed tube	Rint = 0.084
Graphite monochromator	θmax = 28.0°, θmin = 1.5°
ω scans	h = −18→19
5118 measured reflections	k = −8→8
2195 independent reflections	l = −12→8

Refinement

Refinement on F^2	Primary atom site location: structure-invariant direct methods
Least-squares matrix: full	Secondary atom site location: difference Fourier map
R[F^2 > 2σ(F^2)] = 0.046	Hydrogen site location: inferred from neighbouring sites
wR(F^2) = 0.101	H-atom parameters constrained
S = 0.87	w = 1/[σ^2(Fo^2) + (0.0322P)^2] where P = (Fo^2 + 2Fc^2)/3
2195 reflections	(Δ/σ)max < 0.001
109 parameters	Δρmax = 0.20 e Å^−3
0 restraints	Δρmin = −0.28 e Å^−3

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 F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > σ(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq
C2	0.10437 (13)	0.2698 (3)	0.47078 (19)	0.0237 (4)
C3	0.17104 (12)	0.1189 (3)	0.56450 (18)	0.0210 (4)
H1	0.1329	0.0230	0.6039	0.025*
H2	0.2148	0.1912	0.6446	0.025*
C4	0.16677 (12)	−0.0743 (3)	0.34216 (18)	0.0233 (4)
H3	0.1456	0.0435	0.2796	0.028*
H4	0.2053	−0.1615	0.2971	0.028*
C1	0.23067 (12)	0.0001 (3)	0.48329 (18)	0.0191 (4)
C5	0.28098 (12)	−0.1785 (3)	0.57268 (18)	0.0227 (4)
H7	0.2325	−0.2632	0.5990	0.027*
H8	0.3124	−0.2621	0.5150	0.027*
C7	0.35531 (13)	−0.1127 (3)	0.70615 (19)	0.0276 (5)
H9	0.3237	−0.0369	0.7675	0.033*
H10	0.3861	−0.2331	0.7588	0.033*
C8	0.43075 (13)	0.0192 (3)	0.6708 (2)	0.0365 (5)
H11	0.4678	−0.0620	0.6198	0.044*
H12	0.4755	0.0685	0.7595	0.044*
C9	0.38609 (12)	0.1983 (3)	0.5799 (2)	0.0318 (5)
H13	0.4365	0.2720	0.5501	0.038*
H14	0.3585	0.2915	0.6369	0.038*
C6	0.30796 (12)	0.1340 (3)	0.44909 (19)	0.0256 (4)
H5	0.3377	0.0596	0.3848	0.031*
H6	0.2771	0.2561	0.3987	0.031*
N1	0.08083 (10)	−0.1887 (2)	0.35151 (15)	0.0231 (4)
H15	0.0438	−0.1090	0.3904	0.035*
H16	0.0994	−0.2998	0.4065	0.035*
H17	0.0467	−0.2274	0.2633	0.035*
O1	0.01899 (8)	0.21326 (19)	0.41666 (13)	0.0272 (3)
O2	0.13809 (10)	0.4345 (2)	0.44863 (17)	0.0460 (4)

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
C2	0.0269 (10)	0.0189 (11)	0.0256 (10)	0.0027 (9)	0.0076 (9)	−0.0044 (9)
C3	0.0203 (9)	0.0222 (10)	0.0199 (9)	−0.0010 (8)	0.0042 (8)	−0.0024 (8)
C4	0.0236 (9)	0.0254 (11)	0.0221 (10)	−0.0021 (8)	0.0080 (9)	−0.0004 (9)
C1	0.0204 (9)	0.0184 (10)	0.0187 (9)	−0.0006 (8)	0.0056 (8)	0.0014 (8)
C5	0.0228 (9)	0.0210 (11)	0.0250 (10)	0.0011 (8)	0.0075 (8)	0.0027 (9)
C7	0.0232 (9)	0.0288 (12)	0.0285 (11)	0.0020 (8)	0.0030 (9)	0.0054 (9)
C8	0.0238 (10)	0.0419 (14)	0.0393 (13)	−0.0034 (10)	0.0010 (10)	0.0051 (11)
C9	0.0261 (10)	0.0347 (13)	0.0341 (11)	−0.0093 (9)	0.0071 (10)	0.0037 (10)
C6	0.0241 (10)	0.0275 (11)	0.0266 (10)	−0.0017 (8)	0.0093 (9)	0.0034 (9)
N1	0.0223 (8)	0.0224 (9)	0.0238 (8)	−0.0001 (7)	0.0048 (7)	−0.0028 (7)
O1	0.0211 (6)	0.0296 (8)	0.0282 (7)	0.0020 (6)	0.0023 (6)	0.0032 (6)
O2	0.0380 (8)	0.0197 (8)	0.0761 (12)	−0.0020 (7)	0.0085 (8)	0.0084 (8)

Geometric parameters (Å, º)

C2—O2	1.241 (2)	C7—C8	1.516 (2)
C2—O1	1.266 (2)	C7—H9	0.9900
C2—C3	1.517 (2)	C7—H10	0.9900
C3—C1	1.546 (2)	C8—C9	1.521 (3)
C3—H1	0.9900	C8—H11	0.9900
C3—H2	0.9900	C8—H12	0.9900
C4—N1	1.486 (2)	C9—C6	1.525 (2)
C4—C1	1.525 (2)	C9—H13	0.9900
C4—H3	0.9900	C9—H14	0.9900
C4—H4	0.9900	C6—H5	0.9900
C1—C5	1.536 (2)	C6—H6	0.9900
C1—C6	1.540 (2)	N1—H15	0.9100
C5—C7	1.518 (2)	N1—H16	0.9100
C5—H7	0.9900	N1—H17	0.9100
C5—H8	0.9900
O2—C2—O1	125.16 (17)	C5—C7—H9	109.5
O2—C2—C3	118.05 (16)	C8—C7—H10	109.5
O1—C2—C3	116.72 (16)	C5—C7—H10	109.5
C2—C3—C1	112.20 (14)	H9—C7—H10	108.1
C2—C3—H1	109.2	C7—C8—C9	111.49 (15)
C1—C3—H1	109.2	C7—C8—H11	109.3
C2—C3—H2	109.2	C9—C8—H11	109.3
C1—C3—H2	109.2	C7—C8—H12	109.3
H1—C3—H2	107.9	C9—C8—H12	109.3
N1—C4—C1	114.93 (13)	H11—C8—H12	108.0
N1—C4—H3	108.5	C8—C9—C6	112.13 (16)
C1—C4—H3	108.5	C8—C9—H13	109.2
N1—C4—H4	108.5	C6—C9—H13	109.2
C1—C4—H4	108.5	C8—C9—H14	109.2
H3—C4—H4	107.5	C6—C9—H14	109.2
C4—C1—C5	110.30 (15)	H13—C9—H14	107.9
C4—C1—C6	106.66 (13)	C9—C6—C1	113.34 (15)
C5—C1—C6	107.97 (14)	C9—C6—H5	108.9
C4—C1—C3	110.32 (14)	C1—C6—H5	108.9
C5—C1—C3	110.13 (14)	C9—C6—H6	108.9
C6—C1—C3	111.38 (15)	C1—C6—H6	108.9
C7—C5—C1	112.80 (15)	H5—C6—H6	107.7
C7—C5—H7	109.0	C4—N1—H15	109.5
C1—C5—H7	109.0	C4—N1—H16	109.5
C7—C5—H8	109.0	H15—N1—H16	109.5
C1—C5—H8	109.0	C4—N1—H17	109.5
H7—C5—H8	107.8	H15—N1—H17	109.5
C8—C7—C5	110.86 (15)	H16—N1—H17	109.5
C8—C7—H9	109.5
O2—C2—C3—C1	−81.3 (2)	C6—C1—C5—C7	56.37 (18)
O1—C2—C3—C1	95.97 (18)	C3—C1—C5—C7	−65.44 (18)
N1—C4—C1—C5	69.42 (18)	C1—C5—C7—C8	−58.3 (2)
N1—C4—C1—C6	−173.57 (14)	C5—C7—C8—C9	54.7 (2)
N1—C4—C1—C3	−52.5 (2)	C7—C8—C9—C6	−52.5 (2)
C2—C3—C1—C4	−46.4 (2)	C8—C9—C6—C1	53.3 (2)
C2—C3—C1—C5	−168.37 (14)	C4—C1—C6—C9	−172.22 (15)
C2—C3—C1—C6	71.87 (18)	C5—C1—C6—C9	−53.7 (2)
C4—C1—C5—C7	172.56 (13)	C3—C1—C6—C9	67.35 (19)

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
N1—H15···O1	0.91	2.19	2.9398 (19)	139
N1—H15···O1^i	0.91	2.42	3.0242 (18)	124
N1—H16···O2^ii	0.91	1.86	2.723 (2)	157
N1—H17···O1^iii	0.91	1.81	2.7174 (19)	176

Symmetry codes: (i) −x, −y, −z+1; (ii) x, y−1, z; (iii) −x, y−1/2, −z+1/2.

(II) 1-(ammoniomethyl)cycloheaxaneacetate

Crystal data

C9H17NO2	F(000) = 752
Mr = 171.24	Dx = 1.216 Mg m^−3
Monoclinic, C2/c	Melting point: 437 K
Hall symbol: -C 2yc	Mo Kα radiation, λ = 0.71073 Å
a = 30.5452 (11) Å	Cell parameters from 2596 reflections
b = 5.9268 (2) Å	θ = 2.8–27.9°
c = 10.8841 (4) Å	µ = 0.09 mm^−1
β = 108.316 (2)°	T = 173 K
V = 1870.58 (12) Å^3	Plate, colourless
Z = 8	0.58 × 0.49 × 0.11 mm

Data collection

Bruker APEXII CCD detector diffractometer	1608 reflections with I > 2σ(I)
Radiation source: fine-focus sealed tube	Rint = 0.031
Graphite monochromator	θmax = 27.0°, θmin = 1.4°
ω scans	h = −38→37
7746 measured reflections	k = −7→7
2042 independent reflections	l = −13→13

Refinement

Refinement on F^2	Primary atom site location: structure-invariant direct methods
Least-squares matrix: full	Secondary atom site location: difference Fourier map
R[F^2 > 2σ(F^2)] = 0.046	Hydrogen site location: inferred from neighbouring sites
wR(F^2) = 0.125	H-atom parameters constrained
S = 1.06	w = 1/[σ^2(Fo^2) + (0.0542P)^2 + 1.349P] where P = (Fo^2 + 2Fc^2)/3
2042 reflections	(Δ/σ)max = 0.002
110 parameters	Δρmax = 0.24 e Å^−3
0 restraints	Δρmin = −0.17 e Å^−3

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 F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > σ(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq
C1	0.36190 (5)	0.2176 (3)	0.72157 (14)	0.0310 (3)
C2	0.30226 (5)	0.2639 (3)	0.48949 (14)	0.0311 (3)
C3	0.32851 (6)	0.1096 (3)	0.59961 (14)	0.0334 (4)
H3	0.3463	0.0014	0.5648	0.040*
H4	0.3056	0.0210	0.6265	0.040*
C4	0.33683 (5)	0.3708 (3)	0.79133 (15)	0.0331 (4)
H1	0.3587	0.4123	0.8767	0.040*
H2	0.3279	0.5115	0.7406	0.040*
C5	0.39772 (6)	0.3621 (3)	0.68455 (18)	0.0433 (4)
H5	0.4101	0.2736	0.6257	0.052*
H6	0.3821	0.4966	0.6365	0.052*
C6	0.38743 (6)	0.0247 (3)	0.81054 (16)	0.0389 (4)
H13	0.3651	−0.0573	0.8432	0.047*
H14	0.3991	−0.0827	0.7586	0.047*
C7	0.43791 (7)	0.4397 (4)	0.8004 (2)	0.0576 (5)
H7	0.4604	0.5230	0.7691	0.069*
H8	0.4263	0.5439	0.8542	0.069*
C8	0.46176 (7)	0.2409 (4)	0.8823 (2)	0.0649 (6)
H9	0.4763	0.1451	0.8314	0.078*
H10	0.4864	0.2968	0.9592	0.078*
C9	0.42760 (7)	0.1015 (4)	0.92549 (18)	0.0530 (5)
H11	0.4157	0.1925	0.9845	0.064*
H12	0.4435	−0.0322	0.9739	0.064*
N1	0.29507 (4)	0.2717 (2)	0.81185 (12)	0.0332 (3)
H15	0.2711	0.2793	0.7369	0.050*
H16	0.2879	0.3501	0.8748	0.050*
H17	0.3006	0.1249	0.8364	0.050*
O1	0.29105 (4)	0.1805 (2)	0.37705 (10)	0.0381 (3)
O2	0.29272 (4)	0.4596 (2)	0.51366 (11)	0.0418 (3)

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
C1	0.0362 (8)	0.0254 (7)	0.0307 (7)	0.0041 (6)	0.0093 (6)	0.0027 (6)
C2	0.0329 (8)	0.0308 (8)	0.0302 (7)	0.0003 (6)	0.0108 (6)	0.0034 (6)
C3	0.0449 (9)	0.0270 (8)	0.0289 (8)	0.0046 (6)	0.0122 (7)	0.0005 (6)
C4	0.0383 (8)	0.0261 (8)	0.0332 (8)	0.0011 (6)	0.0087 (6)	−0.0032 (6)
C5	0.0442 (9)	0.0379 (9)	0.0499 (10)	0.0016 (7)	0.0180 (8)	0.0069 (8)
C6	0.0438 (9)	0.0322 (8)	0.0385 (9)	0.0086 (7)	0.0099 (7)	0.0056 (7)
C7	0.0399 (10)	0.0500 (12)	0.0771 (14)	−0.0055 (9)	0.0102 (9)	0.0043 (11)
C8	0.0391 (10)	0.0626 (14)	0.0783 (15)	0.0056 (9)	−0.0026 (10)	0.0015 (12)
C9	0.0513 (11)	0.0503 (11)	0.0461 (11)	0.0142 (9)	−0.0006 (8)	0.0079 (9)
N1	0.0364 (7)	0.0308 (7)	0.0295 (6)	0.0048 (5)	0.0061 (5)	−0.0038 (5)
O1	0.0444 (7)	0.0380 (6)	0.0287 (6)	0.0000 (5)	0.0071 (5)	0.0001 (5)
O2	0.0570 (7)	0.0319 (6)	0.0369 (6)	0.0118 (5)	0.0154 (5)	0.0056 (5)

Geometric parameters (Å, º)

C1—C4	1.534 (2)	C6—C9	1.521 (2)
C1—C3	1.537 (2)	C6—H13	0.9900
C1—C5	1.539 (2)	C6—H14	0.9900
C1—C6	1.542 (2)	C7—C8	1.518 (3)
C2—O2	1.2440 (19)	C7—H7	0.9900
C2—O1	1.2632 (19)	C7—H8	0.9900
C2—C3	1.520 (2)	C8—C9	1.517 (3)
C3—H3	0.9900	C8—H9	0.9900
C3—H4	0.9900	C8—H10	0.9900
C4—N1	1.484 (2)	C9—H11	0.9900
C4—H1	0.9900	C9—H12	0.9900
C4—H2	0.9900	N1—H15	0.9100
C5—C7	1.528 (3)	N1—H16	0.9100
C5—H5	0.9900	N1—H17	0.9100
C5—H6	0.9900
C4—C1—C3	112.05 (12)	C1—C6—H13	108.7
C4—C1—C5	107.87 (13)	C9—C6—H14	108.7
C3—C1—C5	109.82 (13)	C1—C6—H14	108.7
C4—C1—C6	111.14 (13)	H13—C6—H14	107.6
C3—C1—C6	107.51 (13)	C8—C7—C5	111.32 (17)
C5—C1—C6	108.42 (13)	C8—C7—H7	109.4
O2—C2—O1	124.01 (14)	C5—C7—H7	109.4
O2—C2—C3	119.65 (13)	C8—C7—H8	109.4
O1—C2—C3	116.34 (13)	C5—C7—H8	109.4
C2—C3—C1	118.21 (13)	H7—C7—H8	108.0
C2—C3—H3	107.8	C9—C8—C7	110.78 (16)
C1—C3—H3	107.8	C9—C8—H9	109.5
C2—C3—H4	107.8	C7—C8—H9	109.5
C1—C3—H4	107.8	C9—C8—H10	109.5
H3—C3—H4	107.1	C7—C8—H10	109.5
N1—C4—C1	115.27 (13)	H9—C8—H10	108.1
N1—C4—H1	108.5	C8—C9—C6	111.33 (16)
C1—C4—H1	108.5	C8—C9—H11	109.4
N1—C4—H2	108.5	C6—C9—H11	109.4
C1—C4—H2	108.5	C8—C9—H12	109.4
H1—C4—H2	107.5	C6—C9—H12	109.4
C7—C5—C1	113.85 (15)	H11—C9—H12	108.0
C7—C5—H5	108.8	C4—N1—H15	109.5
C1—C5—H5	108.8	C4—N1—H16	109.5
C7—C5—H6	108.8	H15—N1—H16	109.5
C1—C5—H6	108.8	C4—N1—H17	109.5
H5—C5—H6	107.7	H15—N1—H17	109.5
C9—C6—C1	114.37 (14)	H16—N1—H17	109.5
C9—C6—H13	108.7
O2—C2—C3—C1	−29.1 (2)	C3—C1—C5—C7	−168.79 (15)
O1—C2—C3—C1	151.56 (14)	C6—C1—C5—C7	−51.60 (19)
C4—C1—C3—C2	63.47 (18)	C4—C1—C6—C9	−66.85 (19)
C5—C1—C3—C2	−56.38 (18)	C3—C1—C6—C9	170.20 (14)
C6—C1—C3—C2	−174.14 (14)	C5—C1—C6—C9	51.54 (19)
C3—C1—C4—N1	47.74 (17)	C1—C5—C7—C8	55.5 (2)
C5—C1—C4—N1	168.72 (13)	C5—C7—C8—C9	−55.8 (3)
C6—C1—C4—N1	−72.56 (17)	C7—C8—C9—C6	55.4 (2)
C4—C1—C5—C7	68.85 (18)	C1—C6—C9—C8	−55.0 (2)

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
N1—H15···O1^i	0.91	1.93	2.7980 (16)	159
N1—H16···O2^ii	0.91	1.85	2.7324 (17)	161
N1—H17···O1^iii	0.91	1.91	2.7851 (18)	161

Symmetry codes: (i) −x+1/2, −y+1/2, −z+1; (ii) x, −y+1, z+1/2; (iii) x, −y, z+1/2.