[Journal logo]

Volume 68 
Part 8 
Pages o298-o301  
August 2012  

Received 2 May 2012
Accepted 4 July 2012
Online 13 July 2012

Enantiomerically pure (1S,5R) and racemic 3-(1-benzothiophen-2-yl)-8-azoniabicyclo[3.2.1]oct-2-ene acetate

aNeuroSearch A/S, Pederstrupvej 93, 2750 Ballerup, Denmark, and bDepartment of Physics, Chemistry and Pharmacy, University of Southern Denmark, Campusvej 55, 5230 Odense, Denmark
Correspondence e-mail: adb@chem.sdu.dk

The title compound, C15H16NS+·C2H3O2-, has been crystallized as both a pure enantiomer (1S,5R) and a racemate. The racemate crystallizes in the space group Cc, with molecules of opposite handedness related to each other by the action of the c-glide. The enantiomer is essentially isostructural with the racemate, except that the glide symmetry is violated by interchange of CH and CH2 groups within the seven-membered ring. The space-group symmetry is reduced to P1 with two molecules in the asymmetric unit. The enantiomer structure shows disorder of the thiophene ring for one of the molecules in the asymmetric unit. The major component of the disorder has the thiophene ring in the same position as in the racemate, but generates a higher-energy molecular conformation. The minor disorder component has different intermolecular interactions but retains a more stable molecular conformation.

Comment

The title compound shows a pharmacological profile as a triple monoamine re-uptake inhibitor, altering the level of activity of the monoamine neurotransmitters serotonin, noradrenaline and dopamine, for the treatment of neuropathic pain and other disorders of the central nervous system (Peters et al., 2006[Peters, D., Brown, D. T., Egestad, B., Dam, E., Jones, D. S., Frøstrup, B., Nielsen, E. O., Olsen, G. M. & Redrobe, J. P. (2006). PCT Int. Appl. WO/2006064031.]). The chirality of the molecule depends on the position of the double bond relative to the NH2+ bridge in the bicyclic ring (see Scheme). For a given orientation of the seven-membered ring, the NH2+ bridge can lie either above or below the ring. In the solid state (with restricted rotation about the C-C bond between the thiophene and bicyclic rings), there is also the possibility for the S atom to lie either adjacent to the double bond (cis) or opposite it (trans).

The racemate, (I)[link] (Fig. 1[link]), and enantiomer, (II)[link] (Fig. 2[link]), are essentially isostructural. The primitive triclinic unit cell used to describe the structure of the enantiomer is transformed to the C-centred monoclinic cell of the racemate by the transformation matrix [[011/0{\overline 1}1/100]]. Racemate (I)[link] crystallizes in the space group Cc, with molecules of opposite handedness related to each other by the action of the c-glide. The thiophene ring adopts the cis orientation with respect to the double bond of the bicyclic ring, and there is no indication of any disorder. The molecules form hydrogen-bonded chains along the c axis, with the acetate anions linking between the NH2+ groups (Table 1[link] and Fig. 3[link]).

[Scheme 1]

Initially, we also solved the structure of enantiomer (II)[link] in the space group Cc, with the implication that racemization had occurred at some stage in the synthesis or crystallization process. However, chiral high-performance liquid chromatography (chiral HPLC) clearly showed that the sample was a single enantiomer, causing us to re-examine the diffraction data. Integration of the data for (II)[link] in the reported primitive cell gave Rint = 0.019 for 5001 unique reflections. In the C-centred monoclinic setting used for the racemate, the Rint value increased to 0.104 for 1972 unique data, and 163 reflections violate the systematic absence conditions for the c-glide at the 3[sigma](I) level. Thus, application of the primitive setting for (II)[link] is supported by the diffraction data.

The structure of enantiomer (II)[link] in the space group P1 contains two molecules in the asymmetric unit. The c-glide present in the structure of (I)[link] is violated in (II)[link] only by the interchange of the CH and CH2 groups in one of the independent molecules. One of the two independent molecules is ordered, while the other displays disorder for the thiophene ring. The ordered molecule has the S atom in the cis orientation, while the disordered molecule has a cis-trans ratio of 0.161 (3):0.839 (3). This result was consistent in several crystals examined. The implication is that the orientation of the thiophene group is governed principally by the intermolecular interactions, rather than any intramolecular factor. The thiophene rings adopt (principally) the same orientation within the structures of (I)[link] and (II)[link], but the interchange of the CH and CH2 groups in one of the two independent molecules of enantiomer (II)[link] results in the trans conformation for the majority component. The principal intermolecular interaction influenced by the disorder involves a neighbouring benzene ring (C1A-C6A) of the thiophene group (Fig. 4[link]). The majority trans component makes a C-H...[pi] contact (H7BA...Cgiii = 2.77 Å; Cg is the centroid of the C1A-C6A ring), while the minority cis component makes an S...[pi] contact [S1C...Cgiii = 3.159 (1) Å; symmetry code: (iii) x - 1, y - 1, z].

The fact that disorder is observed in (II)[link] suggests that there must be some intramolecular preference for the cis conformation. Calculations for an isolated molecule using density functional theory (DFT) methods support this interpretation (see Supplementary materials for optimized molecular structures). Geometry optimization of an isolated molecule starting from the trans conformation causes the thiophene ring to rotate away from the plane containing the CH group by ca 31°. This alleviates a short H...H contact (H7BA...H10B = 2.29 Å) that exists in the coplanar arrangement observed in the crystal structure. By contrast, optimization of the cis conformation causes essentially no change from the conformation observed in the crystal structure. Thus, the coplanar trans conformation of the major disorder component in (II)[link] is less favourable than the cis conformation in terms of the intramolecular energy, but this is outweighed by more favourable intermolecular interactions for the trans conformation in the crystal structure. In some molecules, the intermolecular preference for the trans conformation is overcome by the intramolecular preference for the cis arrangement.

Isostructural enantiomer/racemate pairs represent a special circumstance with regard to Wallach's rule (Brock et al., 1991[Brock, C. P., Schweizer, W. B. & Dunitz, J. D. (1991). J. Am. Chem. Soc. 113, 9811-9820.]). In accordance with expectation, the calculated density of racemate (I)[link] is found to be marginally higher than that of enantiomer (II)[link] (1.304 versus 1.302 Mg m-3), but the difference is barely significant and no firm conclusions can be drawn. Since the crystal structures of the enantiomer and racemate are essentially identical, their simulated powder X-ray diffraction (PXRD) patterns are also essentially identical, and any distinction between the pure enantiomer and the racemate cannot be expected to be made reliably by PXRD analysis. The enantiomer and racemate would also be expected to form solid solutions, which prevents optical resolution by crystallization. A similar case has been reported recently for citalopram oxalate (Lopez de Diego et al., 2011[Lopez de Diego, H., Bond, A. D. & Dancer, R. J. (2011). Chirality, 23, 408-416.]). It is perhaps interesting that the closest comparable compound in the Cambridge Structural Database (Version 5.33; Allen, 2002[Allen, F. H. (2002). Acta Cryst. B58, 380-388.]) also exhibits pseudosymmetry, viz. (1R)-2-(pyrimidin-5-yl)-8-azoniabicyclo[3.2.1]oct-2-ene oxalate propan-2-ol solvate (Gundisch et al., 2001[Gundisch, D., Harms, K., Schwarz, S., Seitz, G., Stubbs, M. T. & Wegge, T. (2001). Bioorg. Med. Chem. 9, 2683-2691.]), which crystallizes in the space group P21, but clearly approximates P21/c.

[Figure 1]
Figure 1
The molecular structure of racemate (I)[link], with displacement ellipsoids drawn at the 50% probability level.
[Figure 2]
Figure 2
The two molecules in the asymmetric unit of enantiomer (II)[link], with displacement ellipsoids drawn at the 50% probability level. Two orientations of the thiophene ring are present for one of the two molecules. For clarity, the acetate anions are not shown.
[Figure 3]
Figure 3
The hydrogen-bonded (pale solid lines) chain along the c axis in racemate (I)[link]. In enantiomer (II)[link], these chains lie along the a axis. [Symmetry code: (i) x, -y + 1, z - [{1\over 2}].]
[Figure 4]
Figure 4
The intermolecular interactions for the two disorder components in enantiomer (II)[link]. For the trans component, a C-H...[pi] contact is highlighted. For the cis component, an S...[pi] contact is highlighted. [Symmetry code: (iii) x - 1, y - 1, z.]

Experimental

The title compound was synthesized in high enantiomeric purity (ee 98.9%) according to the method of Malmgren et al. (2011[Malmgren, H., Cotton, H., Frøstrup, B., Jones, D. S., Loke, M.-L., Peters, D., Schultz, S., Sølver, E., Thomsen, T. & Wennerberg, J. (2011). Org. Process Res. Dev. 15, 408-412.]). The crystallization conditions are also described in Malmgren et al. (2011[Malmgren, H., Cotton, H., Frøstrup, B., Jones, D. S., Loke, M.-L., Peters, D., Schultz, S., Sølver, E., Thomsen, T. & Wennerberg, J. (2011). Org. Process Res. Dev. 15, 408-412.]).

Racemate (I)[link]

Crystal data
  • C15H16NS+·C2H3O2-

  • Mr = 301.39

  • Monoclinic, C c

  • a = 13.316 (3) Å

  • b = 14.195 (3) Å

  • c = 9.020 (2) Å

  • [beta] = 115.775 (11)°

  • V = 1535.3 (6) Å3

  • Z = 4

  • Mo K[alpha] radiation

  • [mu] = 0.22 mm-1

  • T = 180 K

  • 0.20 × 0.10 × 0.07 mm

Data collection
  • Bruker Nonius X8 APEXII CCD area-detector diffractometer

  • Absorption correction: multi-scan (SADABS; Sheldrick, 2003[Sheldrick, G. M. (2003). SADABS. Bruker AXS Inc., Madison, Wisconsin, USA.]) Tmin = 0.830, Tmax = 0.985

  • 6763 measured reflections

  • 2761 independent reflections

  • 2318 reflections with I > 2[sigma](I)

  • Rint = 0.033

Refinement
  • R[F2 > 2[sigma](F2)] = 0.036

  • wR(F2) = 0.084

  • S = 1.03

  • 2761 reflections

  • 191 parameters

  • 2 restraints

  • H-atom parameters constrained

  • [Delta][rho]max = 0.38 e Å-3

  • [Delta][rho]min = -0.22 e Å-3

  • Absolute structure: Flack (1983[Flack, H. D. (1983). Acta Cryst. A39, 876-881.]), with 731 Friedel pairs

  • Flack parameter: 0.08 (6)

Table 1
Hydrogen-bond geometry (Å, °) for racemate (I)[link]

D-H...A D-H H...A D...A D-H...A
N1-H1A...O2 0.92 1.74 2.659 (2) 175
N1-H1B...O1i 0.92 1.84 2.748 (3) 168
N1-H1B...O2i 0.92 2.50 3.125 (2) 125
Symmetry code: (i) [x, -y+1, z-{\script{1\over 2}}].

Enantiomer (II)[link]

Crystal data
  • C15H16NS+·C2H3O2-

  • Mr = 301.39

  • Triclinic, P 1

  • a = 9.0244 (2) Å

  • b = 9.7278 (2) Å

  • c = 9.7492 (2) Å

  • [alpha] = 92.9699 (10)°

  • [beta] = 107.4568 (10)°

  • [gamma] = 107.5510 (9)°

  • V = 768.85 (3) Å3

  • Z = 2

  • Mo K[alpha] radiation

  • [mu] = 0.21 mm-1

  • T = 180 K

  • 0.30 × 0.20 × 0.10 mm

Data collection
  • Bruker-Nonius X8 APEXII CCD area-detector diffractometer

  • Absorption correction: multi-scan (SADABS; Sheldrick, 2003[Sheldrick, G. M. (2003). SADABS. Bruker AXS Inc., Madison, Wisconsin, USA.]) Tmin = 0.686, Tmax = 0.978

  • 11591 measured reflections

  • 5001 independent reflections

  • 4845 reflections with I > 2[sigma](I)

  • Rint = 0.019

Refinement
  • R[F2 > 2[sigma](F2)] = 0.028

  • wR(F2) = 0.074

  • S = 1.03

  • 5001 reflections

  • 392 parameters

  • 15 restraints

  • H-atom parameters constrained

  • [Delta][rho]max = 0.28 e Å-3

  • [Delta][rho]min = -0.30 e Å-3

  • Absolute structure: Flack (1983[Flack, H. D. (1983). Acta Cryst. A39, 876-881.]), with 1495 Friedel pairs

  • Flack parameter: -0.04 (4)

Table 2
Hydrogen-bond geometry (Å, °) for enantiomer (II)[link]

D-H...A D-H H...A D...A D-H...A
N1A-H1A2...O2A 0.92 1.75 2.664 (2) 177
N1A-H1A1...O1Bii 0.92 1.85 2.7550 (18) 168
N1B-H1B2...O1Aiii 0.92 1.83 2.7346 (19) 169
N1B-H1B1...O2Biii 0.92 1.75 2.663 (2) 173
Symmetry codes: (ii) x-1, y, z; (iii) x-1, y-1, z.

H atoms bound to C atoms were placed in idealized positions, with C-H = 1.00 (Csp3-H), 0.99 (CH2), 0.98 (CH3) or 0.95 Å (Csp2-H), and refined as riding, with Uiso(H) = 1.2 or 1.5Ueq(C). The H atoms of the NH2+ group were visible in difference Fourier maps in both structures, but were placed geometrically (N-H = 0.92 Å) and refined as riding [Uiso(H) = 1.2Ueq(N)] for the final refinements. In both cases, the absolute structure was established reliably by refinement of the Flack parameter. For enantiomer (II)[link], the thiophene ring was modelled in cis and trans orientations with respect to the position of the C9B=C10B double bond by splitting atoms C7B/C7C and S1B/S1C. The C6B-S1C, C8B-S1C, C1B-S1B and C8B-S1B bonds were restrained to a common refined value with an s.u. of 0.02 Å, and the C6B-C7B, C7B-C8B, C1B-C7C and C7C-C8B bonds were restrained to a second common refined value with an s.u. of 0.02 Å. The five atoms of each thiophene ring were restrained to lie in a common plane, with an s.u. of 0.01 Å. The DFT calculations were carried out using the DMol3 module (Delley, 1990[Delley, B. (1990). J. Chem. Phys. 92, 508-517.]) in Materials Studio (Accelrys, 2011[Accelrys (2011). Materials Studio. Accelrys Inc., San Diego, California, USA.]), employing the B3LYP function with the DNP 4.4 (double numerical plus d-functions plus polarization) basis set.

For both compounds, data collection: APEX2 (Bruker-Nonius, 2004[Bruker-Nonius (2004). APEX2. Bruker-Nonius BV, Delft, The Netherlands.]); cell refinement: SAINT (Bruker, 2003[Bruker (2003). SAINT. Bruker AXS Inc., Madison, Wisconsin, USA.]); data reduction: SAINT; program(s) used to solve structure: SHELXTL (Sheldrick, 2008[Sheldrick, G. M. (2008). Acta Cryst. A64, 112-122.]); program(s) used to refine structure: SHELXTL; molecular graphics: SHELXTL; software used to prepare material for publication: SHELXTL.


Supplementary data for this paper are available from the IUCr electronic archives (Reference: EG3093 ). Services for accessing these data are described at the back of the journal.


Acknowledgements

The authors are grateful to the Danish Natural Sciences Research Council and the Carlsberg Foundation for provision of the X-ray equipment, and to the Lundbeck Foundation for provision of the Materials Studio software.

References

Accelrys (2011). Materials Studio. Accelrys Inc., San Diego, California, USA.
Allen, F. H. (2002). Acta Cryst. B58, 380-388.  [ISI] [CrossRef] [details]
Brock, C. P., Schweizer, W. B. & Dunitz, J. D. (1991). J. Am. Chem. Soc. 113, 9811-9820.  [CrossRef] [ChemPort] [ISI]
Bruker (2003). SAINT. Bruker AXS Inc., Madison, Wisconsin, USA.
Bruker-Nonius (2004). APEX2. Bruker-Nonius BV, Delft, The Netherlands.
Delley, B. (1990). J. Chem. Phys. 92, 508-517.  [CrossRef] [ChemPort]
Flack, H. D. (1983). Acta Cryst. A39, 876-881.  [CrossRef] [details]
Gundisch, D., Harms, K., Schwarz, S., Seitz, G., Stubbs, M. T. & Wegge, T. (2001). Bioorg. Med. Chem. 9, 2683-2691.  [PubMed] [ChemPort]
Lopez de Diego, H., Bond, A. D. & Dancer, R. J. (2011). Chirality, 23, 408-416.  [ISI] [ChemPort] [PubMed]
Malmgren, H., Cotton, H., Frøstrup, B., Jones, D. S., Loke, M.-L., Peters, D., Schultz, S., Sølver, E., Thomsen, T. & Wennerberg, J. (2011). Org. Process Res. Dev. 15, 408-412.  [CrossRef] [ChemPort]
Peters, D., Brown, D. T., Egestad, B., Dam, E., Jones, D. S., Frøstrup, B., Nielsen, E. O., Olsen, G. M. & Redrobe, J. P. (2006). PCT Int. Appl. WO/2006064031.
Sheldrick, G. M. (2003). SADABS. Bruker AXS Inc., Madison, Wisconsin, USA.
Sheldrick, G. M. (2008). Acta Cryst. A64, 112-122.  [CrossRef] [details]


Acta Cryst (2012). C68, o298-o301   [ doi:10.1107/S0108270112030569 ]