Powder study of (R)-1-phenylethylammonium (R)-2-phenylbutyrate form 2

Solid-State Research Group, Department of Pharmaceutical Sciences, University of Strathclyde, 27 Taylor Street, Glasgow G4 0NR, Scotland, ISIS Facility, Rutherford Appleton Laboratory, Chilton, Didcot, Oxon OX11 0QX, England, Christopher Ingold Laboratory, Department of Chemistry, University College London, 20 Gordon Street, London WC1H 0AJ, England, and Department of Chemical Engineering, University College London, Torrington Place, London WC1E 7JE, England

The crystal structure of a new polymorph of the title compound, C 8 H 12 N + ÁC 10 H 11 O 2 À , was solved by simulated annealing from laboratory X-ray powder diffraction data, collected at 295 K. Subsequent Rietveld refinement using data collected to 1.54 Å resolution yielded an R wp of 0.029. The compound crystallized with one (R)-1-phenylethylammonium cation and one (R)-2-phenylbutyrate anion in the asymmetric unit.

Comment
The structure of the title compound, (I), was first reported by Brianso (1978), hereafter referred to as form 1. Crystallization from ethanol yielded a second polymorph, which is reported here.
The crystal structure of the new form (form 2) was solved by simulated annealing using laboratory capillary X-ray powder diffraction data. The compound crystallizes in the orthorhombic space group P2 1 2 1 2 1 with one (R)-1-phenylethylammonium cation and one (R)-2-phenylbutyrate anion in the asymmetric unit (Fig. 1).
The ion pairs in this new polymorph pack to form a hydrogen-bonded ladder parallel to the a axis (Fig. 2). Each ladder consists of R 3 4 (10) (Etter, 1990) hydrogen-bonded rings comprising four alternating ammonium and carboxylate groups linked by N-HÁ Á ÁO C contacts (Table 1). O1 forms a bifurcated hydrogen bond to H1NB and H1NC, while O2 forms just one hydrogen bond to H1NA. All strong hydrogenbond donors and acceptors are satisfied.
Form 1 shows the same hydrogen-bonded ring motif, albeit with different packing of the ladders in the crystal structure. This arises because the orientation of the 2-phenylbutyrate ion with respect to the 1-phenylethylammonium ion is different. Also, the conformation of the terminal methyl group differs between the two forms -adopting a gauche conformation with respect to the carboxylate group in form 2, while in form 1 the groups are in an anti conformation.
The sample was loaded into a 0.7 mm borosilicate glass capillary and rotated throughout the data collection to minimize preferred orientation effects. Data were collected using a variable count time (VCT) scheme in which the step time is increased with 2 (Shankland et al., 1997;Hill & Madsen, 2002).  Table 1 Hydrogen-bond geometry (Å , ).  The asymmetric unit of (I), with the atom-numbering scheme. Displacement spheres are shown at the 50% probability level (Bruker, 2000). The dashed line indicates a hydrogen bond.

Figure 2
The hydrogen-bonded ladder motif observed in form 2. Atoms not directly involved in hydrogen-bond contacts have been omitted for clarity.

Figure 3
Final observed (points), calculated (line) and difference [(y obs À y calc )/ (y obs )] profiles for the Rietveld refinement of the title compound.
The diffraction pattern indexed to a monoclinic cell [M(19) = 30.7, F(19) = 63.0; DICVOL91; Boultif & Louë r, 1991] and the space group P2 1 2 1 2 1 was assigned from volume considerations and a statistical consideration of the systematic absences (Markvardsen et al., 2001). The data set was background subtracted and truncated to 59.5 2 for Pawley fitting (Pawley, 1981; 2 Pawley = 6.10) and the structure solved using the simulated annealing (SA) global optimization procedure, described previously (David et al., 1998), that is now implemented in the DASH computer program . The SA structure solution used 311 reflections and involved the optimization of two fragments totaling 14 degrees of freedom (six positional and orientational for each fragment present in the asymmetric unit plus a torsion angle for each fragment). All degrees of freedom were assigned random values at the start of the simulated annealing. The best SA solution had a favourable 2 SA / 2 Pawley ratio of 3.41 and a chemically reasonable packing arrangement, with no significant misfit to the diffraction data.
The solved structure was then refined against the data in the range 6-59.7 2 using a restrained Rietveld (1969) method as implemented in TOPAS (Coelho, 2003), with R wp falling to 0.029 during the refinement. All atomic positions (including H atoms) for the structure of (I) were refined, subject to a series of restraints on bond lengths, bond angles and planarity. The refined C-H distances were 0.949 (5)-0.973 (5) Å . U iso values for H atoms were constrained to equal 0.076 Å 2 .
The restraints were set such that bonds and angles did not deviate more than 0.01 Å and 0.8 , respectively, from their initial values during the refinement. Atoms C16, C15, C14, C13, C18, C17, H16, H15, H14, H18 and H17 (phenylethylammonium) and atoms C5, C6, C7, C8, C9, C10, H6, H7, H8, H9 and H10 (phenylbutyrate) were restrained to lie in respective planar groups. A spherical harmonics (fourth order) correction of intensities for preferred orientation was applied in the final refinement (Jä rvinen, 1993). The refined final spherical harmonics coefficients were consistent with mild preferred orientation effects in the sample. The observed and calculated diffraction patterns for the refined crystal structure are shown in Fig. 3.