2,2′-(Ethane-1,2-di­yl)bis­[2-(5-bromo­thio­phen-2-yl)-1,3-dioxolane] at 100 K refined using a multipolar atom model

The title compound, C₁₆H₁₆Br₂O₄S₂, which is a precursor for the synthesis of oligothio­phenes and their substituted homologues, was synthesized and its X-ray crystal structure determined at 100 K. The experimental electron-density parameters for the available atom types were transferred from the ELMAM2 database. The compound lies about an inversion centre, which coincides with the mid-point of a C—C bond. The mol­ecules in the crystal are linked by several types of weak inter­actions; the largest contact surfaces are for H...H and H...Br.

S1. Comment

In the recent past, oliogothiophenes and their substituted homologues have attracted the attention of the scientific community as they are a promising class of organic semiconductor materials with applications in the production of organic field-effect transistors and as electronic devices (Roncali, 1992). These π-conjugated materials have found important applications, for example, in organic solar cells (Ma et al., 2008; Gao et al., 2008; Cao et al., 2009; Rousseau et al., 2010) or OLEDs [organic light-emitting diode?] (Mazzeo et al., 2003; Li et al., 2005). Compared to non-organic semiconductors, they offer many advantages such as tunable electronic properties, especially the band gap, by chemical modifications and substitutions. The title compound has been used as a precursor for these materials (Kreyes, Amirkhani et al., 2010, Kreyes, Ellinger et al., 2010). The reactive C—Br bonds allow for further functionalizations and polymer chain extension. Therefore a detailed knowledge of the structural parameters, stereochemistry, planarity and the mutual arrangement of the molecule is required to better understand the structure–property relationships. In addition, knowledge of the intra- and intermolecular interactions is crucial to rationally carry out the substitutions. The structure of the title compound was briefly reported earlier at 223 (2) K by Ellinger et al. (2007). The structure is here redetermined at low temperature, refined using a multipolar atom model and described in greater detail.

The commonly used spherical atom approximation (IAM, Independent Atom Model) does not give all the information about the intermolecular interactions and is likely to produce severe systematic errors in the refined atomic parameters (Ruysink & Vos, 1974). Experimental electron-density analysis is carried out by X-ray diffraction of mono-crystals at ultra high resolution d_min ≈ 0.5 Å (Coppens, 1998). A difficulty in crystallography is the separation of the anisotropic atomic mean square displacements from the static molecular electron distribution (Hirshfeld, 1976). Proper experimental deconvolution requires diffraction data measured at ultra high resolution. However, effective thermal displacement deconvolution and meaningful electron-density distributions can be achieved even at lower resolutions by transferring the parameters from an electron-density database (Pichon-Pesme et al., 1995; Jelsch et al., 1998; Dittrich et al., 2004, 2005, 2007). Transferring electron-density parameters is comparable to constraining the deformation parameters at their most likely values. The transferability of atomic electron densities was tested the first time by Brock et al. (1991) who applied atomic charge density parameters from an accurate low-temperature study of perylene to diffraction data collected at several temperatures on naphthalene and anthracene crystals.

The ELMAM database (Zarychta et al., 2007) has been extended to ELMAM2 (Domagała et al., 2011) from protein atom types to common organic molecules and is based on optimal local coordinate systems (Domagała & Jelsch, 2008). An automatic transfer procedure of the ELMAM2 database is now available in the MoPro software (Guillot et al., 2001; Jelsch et al., 2005). The different atom types in a molecule are recognized according to the nature and the number of their neighbours. For most atoms, only the first shell of neighbours is analysed, while for hydrogen and oxygen atoms, the second and third shells are investigated, respectively (Domagała & Jelsch, 2008; Domagała et al., 2011).

Using the transferability principle, a multipolar model is applied for the molecule (I) and only the structural parameters (scale factor, atomic coordinates and thermal parameters) are refined. The Fourier residual maps are improved notably on the covalent bonds due to the proper electron- density modeling. The root-mean-square residual density is reduced from 0.090 (IAM) to 0.076 e Å^-3 in the transferred model. In addition to the detailed structural description, the redetermined structure has also significantly better refinement statistics than the previous one. Using a I/s(I) > 2 cutoff, the crystallographic wR2(F) factors are indeed 4.1% and 3.0% for the IAM and the transferred-multipolar structures, respectively.

There is one half of the molecule in the asymmetric unit and four molecules in the unit cell. The two symmetry-equivalent half molecules are linked by an inversion centre in the middle of the C8—C8a bond (Fig. 1).

The molecular assembly is built on five different types of interactions. Dimers of molecules, related by an inversion centre, are built by two very weak hydrogen bonds between C8—H8A and O2 (Table 1). There is a sulfur–hydrogen H3 short contact at 2.985 Å in a dimer of molecules related by b-axis translation.

Two different neighbouring molecules are in van der Waals interaction with the bromine atom. The bromine atom makes a contact with the O1 oxygen atom of the dioxalane ring of the adjacent molecule at a distance of 3.161 (1) Å, which is shorter than the sum of the two van der Waals radii (3.37 Å). An even shorter bromine–oxygen contact (d = 3.0 Å) occurred for instance in the structure of protein human aldose reductase complexed with an inhibitor (Muzet et al., 2003). The C1—Br1···O1 angle equal to 168.7 (1)° (Fig. 2) is almost linear, which is consistent with the value found in the review by Auffinger et al. (2004) of halogen···oxygen short interactions in biomolecules. Nyburg & Faerman (1985) revised the van der Waals radii for several atoms bonded to carbon atoms in molecular crystals; the proposed radius for bromine is 1.84 Å in the equatorial directions and 1.54 Å in the polar direction opposite to the C—Br bond. The direction-dependent effective van der Waals radius is related to the anisotropic electron density of the bromine atom. Bromine and more generally halogen atoms (such as I, Cl or F) show a torus of electron accumulation in the equatorial region while the polar region is electron depleted (Fig. 3). Halogen atoms X can therefore form electrostatically favourable interactions with oxygen atoms when the angle C—X···O is not far from being flat. This characteristic charge density distribution also has consequences in the stereochemistry of halogen···halogen interactions (Bui et al., 2009). In the present structure, the bromine atom also makes a weak interaction with the H7A atom of another molecule at a distance of 3.002 Å. The H7A···Br1···O1 angle is 57.5°.

In addition, two different hydrogen–hydrogen interactions are found to contribute to the formation of the crystal packing. The H2 atom makes a contact with the H8B atom of an adjacent molecule at a distance of 2.316 Å (Fig. 2). The H6B atom makes a comparatively shorter interaction with H6B of a neighbouring molecule at a distance of 2.248 Å through an inversion centre (Fig. 2). Hydrogen–hydrogen bonding has been shown to be a stabilizing interaction in molecules and crystals (Matta et al., 2003).

To analyse quantitatively the intermolecular contacts in the title compound, a Hirshfeld surface analysis was performed with CrystalExplorer (Spackman & Jayatilaka, 2009). The analysis reveals that H···H interactions (31.1%) and Br···H (25.1%) are the most prevalent interactions. The next major crystal packing interactions are S···H (14.4%), C···H (12.5%), O···H (10.3%) and Br···O (3.8%).

The thiophene ring has a planar configuration whereas the dioxalane ring is not planar. The thiophene ring and the O2—C1—O1 plane in the dioxalane rings are oriented at an angle of 55.1° from each other (angle between normal vectors). The two dioxalane rings adopt an anti conformation due to the intramolecular inversion centre. As viewed along the c axis, the molecules are stacked over each other and form two kinds of channels of different size. In the largest channel, two bromine atoms of opposite molecules are pointing towards each other at a distance of 4.113 (4) Å.

When interatomic distances are compared between the spherical and the transferred models, most of the covalent bond lengths (between non-H atoms) are comparable within one or two times their standard uncertainties. However, there is an exception for the C5—O2 bond length, which decreases from 1.410 (2) to 1.405 (1) Å after transfer and subsequent structure refinement. The same trend is also observed for all other C—O bonds of the dioxalane group. This shortening of oxygen-containing covalent bonds can be explained by the fact that the modelling of oxygen electron lone pairs has an effect on the coordinates of the O atoms. When the spherical atom model (IAM) is used, the oxygen atom is slightly displaced towards the middle of the lone pairs. The transfer procedure, followed by the refinement of the structural parameters, leads to removal of this bias on the oxygen atom coordinates, thus shortening the covalent bonds in which they are involved. This is confirmed by the values of the equivalent B_iso factor of O1 and O2, which also decrease slightly, upon transfer, by about 0.06 Ų, which is above the standard uncertainty on B_iso parameters (~0.02 Ų). These observations clearly support the motive behind the transfer of electron-density parameters as it gives a better structural model, not biased by the non-modelled deformation electron density.

In the comparison of the two structures, the H—X distances in the multipolar atom model are also elongated to the standard neutron diffraction values (Allen, 1986). These structural modifications have repercussions on some of the intermolecular contacts. For instance, the distance between H6B atoms of two symmetry-related molecules is decreased significantly upon transfer from 2.305 to 2.248 Å. There are two C8—H8B···O2 interactions in a dimer of molecules (symmetry -X+³/₂;-Y- 1/2; -Z+1) in the crystal structure which can be considered as a very weak hydrogen bond. The H···O distance is 2.57 Å and the C—H···O angle is 140.7°. These values are slightly different: (2.67 Å and 142.0°) in the IAM structure. H8B also forms weak intramolecular contacts with atoms O1 and O2. In the H8B···O1 interaction which generates a five-atom cycle, the HO distance is 2.69 Å but the C—H···O angle of 96.2° is unfavourable for a hydrogen bond. These values were equal to 2.66 Å and 99.1° in the IAM structure. The different positioning of the hydrogen atoms in the IAM structure results in significantly altered geometric parameters of the hydrogen bonds, compared to the transferred models.

The transfer of the multipolar parameters significantly decreases the residual Fourier electron density. The maximum, minimum and r.m.s. values for the spherical atom model are 0.43, -0.68 and 0.090 e Å^-3, respectively. The corresponding values for the transferred multipolar atom model are decreased to 0.32, -0.40 and 0.076 e Å^-3, respectively (Fig. 4). The electron-density parameters also allowed calculation of the dipole moment of the molecule whose value turns out to be 4.80 Debye.

Stevens & Coppens (1976) have introduced a suitability factor S for the multipolar atom model which is based on the observation that the improvement in the refinement statistics is mainly due to a better description of the valence electron density. The suitability factor S of a compound is equal to the following ratio:

S = V / (Σ n²_core )

where V is the unit-cell volume and n_core is the number of core electrons for the given atom types. The denominator is a measure of the core electron scattering of the unit cell. The suitability factor was calculated to be 0.235 for compound (I). This low value is due to the bromine atom in the chemical formula. After the database transfer of compound (I), the difference ΔR(F) between the spherical atom model (2.7%) and the transferred model (2.2%) is 0.50. As illustrated by Dittrich et al. (2007), the lower the suitability factor, the lower is the expected ΔR(F).

A rigid bond test analysis shows that the r.m.s. difference of U_ij ellipsoids along the covalent bonds shows a slight improvement, with ΔZ = 0.0019 Ų for the IAM model and ΔZ = 0.0018 Ų for the transferred model. The magnitude of the thermal displacement parameters also generally decreases upon the transfer: U_eq(multipolar) » U_eq(IAM) * 0.96.

Related literature

For related literature, see: Allen (1986); Auffinger et al. (2004); Blessing (1987); Brock et al. (1991); Bui et al. (2009); Cao et al. (2009); Coppens (1998); Dittrich et al. (2004, 2005, 2007); Domagała & Jelsch (2008); Domagała et al. (2011); Dominiak et al. (2007); Ellinger et al. (2007); Gao et al. (2008); Guillot et al. (2001); Hirshfeld (1976); Jelsch et al. (1998, 2005); Kreyes, Amirkhani, Lieberwirth, Mauer, Laquai, Landfester & Ziener (2010); Kreyes, Ellinger, Landfester, Defaux, Ivanov, Elschner, Meyer-Friedrichsen & Ziener (2010); Li et al. (2005); Ma et al. (2008); Matta et al. (2003); Mazzeo et al. (2003); Muzet et al. (2003); Nyburg & Faerman (1985); Pichon-Pesme, Lecomte & Lachekar (1995); Roncali (1992); Rousseau et al. (2010); Ruysink & Vos (1974); Spackman & Jayatilaka (2009); Stevens & Coppens (1976); Volkov et al. (2004, 2007); Zarychta et al. (2007).

S2. Experimental

For the synthesis of the title compound, the same procedure as that reported by Ellinger et al. (2007) was adopted, in toluene solvent (see Scheme). 1,4-Bis(5-bromothien-2-yl)butane-1,4-dione (1.0 g, 2.5 mmol) was dissolved in hot toluene (50 ml). After complete dissolution of the diketone, p-toluenesulfonic acid (p-TosH; 200 mg, 1.1 mmol) and ethylene glycol (10 ml, 179.3 mmol) were added. The mixture was then stirred and heated at 388 K for 48 h using a Dean–Stark trap. After cooling, saturated aqueous NaHCO₃ was added. The organic phase was separated and the aqueous phase was extracted with toluene three times. The combined organic phases were dried over anhydrous MgSO₄, filtered, evaporated to dryness and purified by fractional recrystallization using cyclohexane to give the title product (yield 48%). Crystals used for analysis were grown by slow evaporation from a chloroform solution at room temperature. ¹H NMR (200 MHz, CDCl₃): δ 6.9 (d, 2H, J = 3.8 Hz), 6.74 (d, 2H, J = 3.8 Hz), 3.98 (m, 8H, OCH₂CH₂O), 2.08 (s, 4H).

S3. Refinement

The space group was found to be C2/c and the reflections, including Friedel pairs, were merged with SORTAV (Blessing, 1987) before final refinement. Scale factors, atomic positions and thermal displacement parameters were refined using MoPro software (Jelsch et al., 2005) until convergence. The least-squares refinement was based on |F|².

Initially, a conventional spherical atom model was applied. Electron-density parameters were then transferred from the ELMAM2 library (Domagała et al., 2011) for all the atoms, except C5, and were subsequently kept fixed. The C5 chemical atom type was not available in the ELMAM2 library and was modelled as atom C444 (c1-oCo-c2) in the UBDB theoretical database (Volkov et al., 2004, 2007; Dominiak et al. 2007). With the electron-density library transfer, the same structural parameters were refined but a multipolar charged atom model was applied. The molecule was set electrically neutral after library transfer. A view of the transferred deformation electron density is shown in Fig. 5.

The H—X distances were constrained to the standard values in the neutron diffraction studies (Allen, 1986): 1.092 Å for CH₂ and 1.083 Å for aromatic C—H groups. Riding constraints on H-atom isotropic thermal displacement parameters were applied: U_iso(H) = 1.2U_eq(X), where X is the neighbour carbon atom. The refinements were carried out using all reflections. The ELMAM2 refinement shows a slight improvement in the statistical indexes when compared to the spherical atom refinement. With a I/σ_I > 2 cutoff, the crystallographic factors are reduced from 2.69 to 2.19% for R (F), and from 4.09 to 3.01% for wR² (F).

Computing details

Figures

	Fig. 1. The molecular configuration and atom-numbering scheme for the title compound. Displacement ellipsoids are drawn at the 50% probability level and H atoms are not labelled. [Symmetry code: (i) please supply]
	Fig. 2. The crystal packing of (I), viewed along the b axis, showing different intermolecular interactions (thin lines). [Please indicate the origin (O) and provide symmetry codes where appropriate]
	Fig. 3. Deformation electron density transferred from the ELMAM2 database. Contour: ±0.05 e Å^-3. In the electronic version of the paper, blue continuous line: positive, red dashed lines: negative, and yellow dashed lines: zero contours.
	Fig. 4. A Fourier map of residual density of the thiophene ring of the molecule, shown (a) for the spherical atom model and (b) for the transferred multipolar atom model.

2,2'-(ethane-1,2-diyl)bis[2-(5-bromothiophen-2-yl)-1,3-dioxolane]

Crystal data

C16H16Br2O4S2	F(000) = 984
Mr = 496.23	Dx = 1.899 Mg m^−3
Monoclinic, C2/c	Mo Kα radiation, λ = 0.71073 Å
Hall symbol: -C 2yc	Cell parameters from 1726 reflections
a = 19.255 (1) Å	θ = 3.7–25.3°
b = 5.7804 (4) Å	µ = 4.93 mm^−1
c = 16.9328 (6) Å	T = 100 K
β = 112.845 (4)°	Cubic, yellow
V = 1736.6 (3) Å^3	0.37 × 0.20 × 0.20 mm
Z = 4

Data collection

Bruker APEXII CCD detector diffractometer	1518 independent reflections
Radiation source: fine-focus sealed tube	1478 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.043
ω scans	θmax = 25.2°, θmin = 3.7°
Absorption correction: multi-scan (Blessing, 1995)	h = −22→21
Tmin = 0.329, Tmax = 0.346	k = 0→6
1565 measured reflections	l = 0→20

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)] = 0.021	Hydrogen site location: inferred from neighbouring sites
wR(F^2) = 0.053	H-atom parameters constrained
S = 1.04	w = 1/[σ^2(Fo^2) + (0.027P)^2 + 3.4P] where P = (Fo^2 + 2Fc^2)/3
1518 reflections	(Δ/σ)max = 0.002
109 parameters	Δρmax = 0.33 e Å^−3
0 restraints	Δρmin = −0.40 e Å^−3

Special details

Refinement. Refinement of F² against reflections. The threshold expression of F² > 2sigma(F²) is used for calculating R-factors(gt) 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
BR	0.61274 (1)	0.49392 (3)	0.77161 (1)	0.02877 (6)
S	0.61813 (2)	0.38176 (7)	0.59408 (3)	0.02090 (8)
O1	0.60278 (7)	0.2111 (2)	0.42402 (8)	0.0236 (2)
O2	0.64198 (7)	−0.1534 (2)	0.46632 (8)	0.0240 (2)
C1	0.63676 (9)	0.2993 (3)	0.6976 (1)	0.0209 (3)
C2	0.6695 (1)	0.0889 (3)	0.7174 (1)	0.0247 (3)
H2	0.68553	0.00678	0.77946	0.02963*
C3	0.6802 (1)	−0.0114 (3)	0.6458 (1)	0.0225 (4)
H3	0.70565	−0.17937	0.64749	0.02699*
C4	0.65489 (9)	0.1275 (3)	0.5748 (1)	0.0181 (3)
C5	0.65931 (9)	0.0798 (3)	0.4889 (1)	0.0184 (3)
C6	0.5390 (1)	0.0633 (5)	0.3843 (1)	0.0349 (4)
H6A	0.48919	0.13415	0.39132	0.04194*
H6B	0.52753	0.04031	0.31633	0.04194*
C7	0.56214 (10)	−0.1648 (3)	0.4322 (1)	0.0262 (3)
H7A	0.54176	−0.31165	0.38869	0.03143*
H7B	0.54093	−0.17780	0.48320	0.03143*
C8	0.73659 (9)	0.1265 (3)	0.4878 (1)	0.0211 (3)
H8A	0.73362	0.09413	0.42307	0.02536*
H8B	0.77670	0.00687	0.53245	0.02536*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
BR	0.0305 (2)	0.0368 (2)	0.0250 (2)	0.00077 (7)	0.0174 (1)	−0.00272 (6)
S	0.0207 (2)	0.0242 (2)	0.0201 (2)	0.0029 (2)	0.0104 (2)	0.0026 (2)
O1	0.0197 (6)	0.0296 (7)	0.0198 (6)	−0.0018 (5)	0.0058 (5)	0.0019 (5)
O2	0.0183 (6)	0.0239 (7)	0.0310 (7)	−0.0047 (5)	0.0107 (5)	−0.0071 (5)
C1	0.0203 (8)	0.0280 (9)	0.0169 (8)	−0.0022 (7)	0.0101 (6)	−0.0012 (7)
C2	0.0298 (9)	0.0256 (10)	0.0175 (8)	−0.0026 (8)	0.0078 (7)	0.0015 (7)
C3	0.026 (1)	0.023 (1)	0.017 (1)	−0.0002 (6)	0.0067 (8)	0.0006 (6)
C4	0.0153 (7)	0.0210 (8)	0.0196 (8)	−0.0001 (6)	0.0085 (6)	−0.0004 (6)
C5	0.0149 (8)	0.0237 (9)	0.0178 (8)	−0.0031 (7)	0.0076 (7)	−0.0030 (7)
C6	0.023 (1)	0.048 (1)	0.026 (1)	−0.0087 (9)	0.0013 (8)	0.0013 (10)
C7	0.0193 (8)	0.0336 (10)	0.0255 (9)	−0.0082 (7)	0.0086 (7)	−0.0074 (8)
C8	0.0160 (8)	0.0248 (9)	0.0250 (9)	−0.0034 (7)	0.0106 (7)	−0.0041 (7)

Geometric parameters (Å, º)

BR—C1	1.871 (2)	C3—H3	1.083
S—C1	1.715 (2)	C4—C5	1.514 (2)
S—C4	1.717 (2)	C5—C8	1.519 (2)
O1—C5	1.428 (2)	C6—C7	1.521 (3)
O1—C6	1.431 (2)	C6—H6B	1.092
O2—C5	1.405 (2)	C6—H6A	1.092
O2—C7	1.418 (2)	C7—H7B	1.092
C1—C2	1.351 (3)	C7—H7A	1.092
C2—C3	1.430 (3)	C8—H8A	1.092
C2—H2	1.083	C8—H8B	1.092
C3—C4	1.368 (3)
BR—C1—S	120.6 (1)	C1—C2—H2	124.4
BR—C1—C2	126.0 (2)	C2—C3—C4	112.8 (2)
S—C1—C2	113.3 (2)	C2—C3—H3	123.6
S—C4—C5	121.3 (1)	H2—C2—C3	124.4
S—C4—C3	111.6 (2)	C3—C4—C5	127.1 (2)
O1—C5—O2	105.7 (2)	H3—C3—C4	123.6
O1—C5—C4	109.4 (1)	C4—C5—C8	113.5 (1)
O1—C5—C8	110.7 (1)	C5—O2—C7	105.1 (1)
O1—C6—C7	104.3 (2)	C5—O1—C6	107.6 (2)
O1—C6—H6B	110.8	C5—C8—H8A	108.2
O1—C6—H6A	110.8	C5—C8—H8B	108.2
O2—C5—C4	109.9 (1)	C6—C7—H7B	111.1
O2—C5—C8	107.3 (1)	C6—C7—H7A	111.1
O2—C7—C6	102.9 (2)	H6A—C6—C7	110.8
O2—C7—H7B	111.1	H6A—C6—H6B	109.5
O2—C7—H7A	111.1	H6B—C6—C7	110.8
C1—S—C4	91.08 (8)	H7A—C7—H7B	109.5
C1—C2—C3	111.2 (2)	H8A—C8—H8B	109.5
BR—C1—S—C4	179.71 (8)	C1—C2—C3—C4	−0.2 (2)
BR—C1—C2—C3	−179.6 (1)	C1—C2—C3—H3	179.8
BR—C1—C2—H2	0.4	C2—C1—S—C4	0.09 (13)
S—C1—C2—C3	0.02 (13)	C2—C3—C4—C5	178.1 (1)
S—C1—C2—H2	−180.0	H2—C2—C3—C4	179.8
S—C4—C5—O2	−141.0 (1)	H2—C2—C3—H3	−0.2
S—C4—C5—O1	−25.3 (1)	C3—C4—C5—C8	−78.8 (1)
S—C4—C5—C8	98.8 (1)	H3—C3—C4—C5	−1.9
S—C4—C3—C2	0.2 (1)	C4—C5—O2—C7	80.8 (1)
S—C4—C3—H3	−179.8	C4—C5—O1—C6	−96.5 (1)
O1—C5—O2—C7	−37.1 (1)	C4—C5—C8—H8A	−177.7
O1—C5—C4—C3	157.0 (1)	C4—C5—C8—H8B	63.8
O1—C5—C8—H8A	−54.3	C5—O2—C7—C6	36.7 (1)
O1—C5—C8—H8B	−172.8	C5—O2—C7—H7B	−82.3
O1—C6—C7—O2	−22.9 (2)	C5—O2—C7—H7A	155.6
O1—C6—C7—H7B	96.1	C5—O1—C6—C7	1.0 (1)
O1—C6—C7—H7A	−141.8	C5—O1—C6—H6B	−118.2
O2—C5—O1—C6	21.7 (1)	C5—O1—C6—H6A	120.2
O2—C5—C4—C3	41.4 (2)	C6—O1—C5—C8	137.6 (2)
O2—C5—C8—H8A	60.6	H6A—C6—C7—H7B	−23.1
O2—C5—C8—H8B	−57.9	H6A—C6—C7—H7A	99.0
O2—C7—C6—H6B	96.3	H6B—C6—C7—H7B	−144.7
O2—C7—C6—H6A	−142.0	H6B—C6—C7—H7A	−22.6
C1—S—C4—C5	−178.1 (1)	C7—O2—C5—C8	−155.3 (1)
C1—S—C4—C3	−0.2 (1)

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
C8—H8B···O2^i	1.09	2.57	3.484 (2)	141

Symmetry code: (i) −x+3/2, −y−1/2, −z+1.