Synthesis, resolution and crystal structures of two enantiomeric rhodamine derivatives

The synthesis of rac-6′-bromo-3′-diethylamino-3H-spiro[2-benzofuran-1,9′-xanthen]-3-one and its resolution into separate enantiomers is described. The structures of the racemate and of the individual enantiomers were determined and showed differing degrees of folding of the xanthene portion, which were attributed primarily to packing interactions. The supramolecular features of the three structures show significant differences.


Chemical context
The compounds synthesized here are part of ongoing work to form chiral sensors based on the supramolecular interactions of chiral rhodamine derivatives with analytes. Enantiomeric sensing is critical for the efficient and safe formation of chiral pharmaceuticals (LaPlante et al., 2011) since enantiomers may have vastly different biological effects including toxicity (Reist et al., 1998). Most current methods for the detection of enantiomeric purity involve chromatographic techniques that require costly instrumentation (Wang et al., 2006). Chiral supramolecular sensors offer an inexpensive alternative Jo et al., 2014;Zhang et al., 2014;Yu & Pu, 2015). Supramolecular sensors, such as modified rhodamine derivatives, have garnered recent interest as sensors with biological applications (Pak et al., 2015;You et al., 2015). Additionally, recent work has shown that rhodamine B can function as a sensor differentiating between diastereomers (Shimizu & Stephenson, 2010). Herein, we report the synthesis, resolution and structures of two asymmetric rhodamine derivatives 4 and 5 which are being investigated for potential as chiral sensors.

Structural commentary
In general terms, the structures of 3-5 are similar to those of other rhodamine derivatives that have been reported in that the xanthene portion is modestly folded along the OÁ Á ÁC axis of the central ring and the benzofuranone unit is nearly perpendicular to the mean plane of the xanthene unit. Of note in the present work is the variation in the fold of the xanthene portion which is largest in 3, distinctly smaller in 4 and smallest in 5 but with a significant difference in this angle between the two independent molecules (see the first four entries in Table 1Fig. 1). We attribute these differences to the different packing modes for the three structures. In 3 (Fig. 2), the molecules form zigzag stacks with each pair of adjacent molecules related by centers of symmetry. This leads to pairwise H17CÁ Á ÁC4 separations of 3.04 Å which are only 0.14 Å less than the sum of the van der Waals radii. Were the xanthene portions flatter, these would develop into significant intermolecular contacts. With 4 and 5 (Figs. 3 and 4) in the noncentrosymmetric space group P2 1 2 1 2 1 , this stacking is no longer possible and while in 4 there is a van der Waals contact  Table 1 Dihedral angles ( ) in selected rhodamine derivatives.. R 1 -R 6 positions are defined in Fig. 1.   Okada (1996); (o) Wang et al. (1990); (p) Miao et al. (1996).

Figure 1
Key for Table 1.

Figure 2
Perspective view of 3, with the atom-numbering scheme and 50% probability displacement ellipsoids. of 2.90 Å between H17A and C4 i [symmetry code: (i) 3 2 À x, 1 À y, À 1 2 + z] which could be lessened by a greater folding, this is opposed by a H2Á Á ÁH19 ii [symmetry code: (ii) À 1 2 + x, 3 2 À y, 1 À z] separation of 2.48 (4) Å which is only 0.08 Å greater than the sum of the van der Waals radii. In the case of 5, the C8-C13 ring experiences the opposing contacts H40BÁ Á ÁC13 (2.79 Å ) and H41B iii Á Á ÁC11 [2.79 Å ; symmetry code: (iii) 1 + x, y, z], both of which are 0.11 Å less than a van der Waals contact and serve to hold this ring in position in the packing. On the other side of this xanthene moiety there is a Br1Á Á ÁO6 iv [symmetry code: (iv) 1 2 + x, 3 2 À y, 1 À z] contact of 3.251 (3) Å which is 0.12 Å less than a van der Waals contact and imparts more of a twist than a simple fold to this portion. This can be seen from the dihedral angle of 5.7 (2) between the C1-C6 ring and the C1/C6/C7/O1 plane. For the second molecule, there are no short intermolecular contacts with either side of the xanthene moiety to influence its conformation.  Table 2. These include two sets of pairwise C-HÁ Á ÁO hydrogen bonds, two additional sets of C-HÁ Á ÁO hydrogen bonds and a set of C-HÁ Á Á(ring) interactions. The C14Á Á ÁH14BÁ Á ÁO2 i and C19-H19Á Á ÁO1 iii interactions bind the molecules into stacks along the a-axis direction while the C16-H16AÁ Á ÁO3 i and C17-H17AÁ Á Á(ring) iv interactions tie the stacks together (Fig. 6). Intermolecular interactions are much fewer in the crystal of 4 with C14-H14BÁ Á ÁO3 v and C20-H20Á Á ÁO3 vi hydrogen bonds (Table 3) forming zigzag chains (Fig. 7) running approximately along the c-axis direction and arranged to form rectangular channels along the a-axis direction (Fig. 8)  Perspective view of 4, with the atom-numbering scheme and 50% probability displacement ellipsoids.

Figure 6
Packing of 3, viewed along the a-axis direction, with the color code for C-HÁ Á ÁO interactions as in Fig. 5.

Figure 8
Packing of 4, viewed along the a-axis direction, with C-HÁ Á ÁO hydrogen bonds shown as dotted lines.

Database survey
There are 71 structures of rhodamine derivatives in the literature, although many are considerably more substituted than 4 and 5 and include a variety of fused-ring systems. Table 1 lists, in addition to those reported here, 20 other structures which are most nearly comparable to those of this work. In all of these, the lactone ring (ring 1, Fig. 1) is nearly perpendicular to the mean plane of the central pyran ring (ring 2, Fig. 1) with dihedral angles ranging from 87.08 (13) to 90.0 (2) and with three structures having the lactone ring on a crystallographic mirror (Table 1). In all cases, the xanthene moiety is folded across the OÁ Á ÁC axis, with the majority having a dihedral angle between rings 3 and 4 ( Fig. 1) in the range 2.42 (3)-7.36 (5) , but there are six having angles up to 17.5 (5) ( Table 1). In this latter group, those with the largest angles involve a twist of the xanthene moiety as well as a fold, and this is seen in both symmetrically and unsymmetrically substituted examples. Inspection of intermolecular contact calculations indicates that the largest dihedral angles correlate with intermolecular contacts at or somewhat less than the sums of the relevant van der Waals radii.

Synthesis and crystallization
As outlined in the scheme, compound 1 (2.00 g, 6.73 mmol) was mixed with compound 2 (1.10g, 6.35 mmol) in 16 mL of methylsulfonic acid. The mixture was stirred and heated for 1 h at 373 K. The cooled solution was poured over ice and then extracted with dicholoromethane. A crude yield of the race-mate 3 was obtained. A portion of the crude product (1.343 g) was purified on a flash column with 15% ethyl acetate in hexanes followed by 100% ethyl acetate. The fractions containing the product were combined and left in a beaker covered with a tissue and the solvent was allowed to evaporate slowly. After about two weeks, the purified racemate yielded a mixture of long needle-shaped as well as plate-shaped crystals (0.293 g, 0.651 mmol, 21.8% yield). Thin layer chromatography demonstrated that both crystal shapes were the desired product (racemate 3), but only the needles provided a wellrefined structure. The melting point range was found to be 420.6-428.9 K for the needles and 415.9-429.8 K for the plates. An NMR spectrum of compound 3 was also obtained (Figs. S1 and S2).
To separate the enantiomers a mobile phase of 70% hexanes, 29.97% ethanol and 0.03% diethylamine was used. A 4 mg mL À1 solution of the racemic bromo-rhodamine derivative, 3 was dissolved in the mobile phase. A two-pump system, both Shimadzu LC-20AD pumps, was utilized for moving the mobile phase through the column. Pump A pumped hexanes and Pump B pumped the mixture of 95% ethanol and 0.5% diethylamine at a flow rate of 3.0 mL min À1 for a total of 16 minutes. The sample was placed in a Shimadzu SIL-20AC autosampler, which injected 400 mL of the sample into the mobile phase. A Shimadzu CTO-20A oven, set at 298 k, held the ChiralPak AD-H column whose stationary phase is amylose tris (3,5-dimethylphenylcarbamate) coated on 5 mm silica-gel. The compounds were eluted and then detected with a Shimadzu SPD-20A UV photodiode array detector with a deuterium lamp set at 230 nm. Each enantiomer was collected  with a Shimadzu FRC-10A fraction collector. One enantiomer (4) elutes from 11.6-12.8 minutes, and the other (5) elutes from 13.4-14.8 minutes using the method described above. Slow evaporation of the solutions of the pure enantiomers at room temperature afforded X-ray quality crystals over 1-5 days.

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 5. In 3 and 5, H atoms attached to carbon were placed in calculated positions (C-H = 0.95-0.99 Å ) and included as riding contributions with isotropic displacement parameters 1.2-1.5 times those of the attached atoms. In 4, the methyl group H atoms were placed in calculated positions as in 3 and 5 (due to poor geometry resulting from individual refinement) while the remainder were refined.   For all compounds, data collection: APEX3 (Bruker, 2016); cell refinement: SAINT (Bruker, 2016); data reduction: SAINT (Bruker, 2016); program(s) used to solve structure: SHELXT (Sheldrick, 2015a); program(s) used to refine structure: SHELXL2014 (Sheldrick, 2015b); molecular graphics: DIAMOND (Brandenburg & Putz, 2012); software used to prepare material for publication: SHELXTL (Sheldrick, 2008).  Hydrogen site location: inferred from neighbouring sites H-atom parameters constrained

(3) rac-6′-Bromo-3′-diethylamino-3H-spiro[2-benzofuran-1,9′-xanthen]-3-one
where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max < 0.001 Δρ max = 1.01 e Å −3 Δρ min = −0.98 e Å −3 Special details 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. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger. Hatoms attached to carbon were placed in calculated positions (C-H = 0.95 -0.99 Å) and included as riding contributions with isotropic displacement parameters 1.2 -1.5 times those of the attached atoms.

Special details
Experimental. The diffraction data were obtained from 3 sets of 400 frames, each of width 0.5° in ω, colllected at φ = 0.00, 90.00 and 180.00° and 2 sets of 800 frames, each of width 0.45° in φ, collected at ω = -30.00 and 210.00°. The scan time was 20 sec/frame. 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. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.

Special details
Experimental. The diffraction data were collected in three sets of 363 frames (0.5° width in ω) at φ = 0, 120 and 240°. A scan time of 60 sec/frame was used. 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.

sup-13
Acta Cryst. (2017). E73, 327-333 Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger. Hatoms attached to carbon were placed in calculated positions (C-H = 0.95 -0.99 Å). All were included as riding contributions with isotropic displacement parameters 1.2 -1.5 times those of the attached atoms.