Neutron powder diffraction study of perdeuterodimethyl sulfone

The crystal structure of perdeuterodimethyl sulfone, (CD₃)₂SO₂ or C₂D₆O₂S, has been refined at 4.5 K against high-resolution neutron powder diffraction data. The structure determined previously by Sands [Z. Kristallogr. (1963), 119, 245-251] at ambient temperature is shown to remain down to liquid helium temperature, and at 4.5 K the S-C and S-O bond distances are 1.441 (2) and 1.760 (2) Å, respectively. The mol­ecules are distorted tetra­hedra with C_2v point symmetry (crystallographic symmetry m2m for S and m for C, O and one D atom) and are linked through a network of weak hydrogen bonds in the C-centred ortho­rhom­bic structure.

S1. Comment

Dimethyl sulfone is a member of a family compounds with the general formula Y₂XZ₂ (Y = H, F or CH₃; X = O, S or Se; Z = O or S) in which the molecules may be characterized in terms of distorted tetrahedra with C_2v symmetry. In view of recent interest in sulfone reactivity and chemical bonding (Dobado et al., 1999), and structural analyses of co-ordinated sulfone ligands compared with the free molecules (Dikarev et al., 2003), it is surprising that the fully refined low-temperature structure of dimethyl sulfone, (I), has not been reported. The heavy-atom structure of (I) has been known for some time (Sands, 1963), determined from single-crystal X-ray diffraction data at ambient temperature. The structure is orthorhombic, space group Cmcm, with bond lengths determined to a precision of some 0.02 Å and bond angles to 1°. A more recent investigation (Langs et al., 1970) reports the structure, including H-atom positions, in space group Amma, and although there are inconsistencies in the reported co-ordinates according to the Cambridge Structural Database (CSD, Version 5.28 of January 2007; Allen, 2002), the structure is essentially in agreement with that proposed by Sands. Dimethyl sulfone has a relatively high melting point (382 K) in comparison with the sulfuryl halides, such as Cl₂SO₂ and F₂SO₂, which are similar in structure and molecular weight, and early reports had suggested properties associated with plastic crystalline behaviour. Heat capacity measurements (Clever & Westrum, 1970) found no corroborative evidence supporting plastic behaviour in (I) or associated structural phase transitions between 5 K and the melting point. The present study, using modern high-resolution neutron powder diffraction techniques, aims to provide a more accurate reference structure at low temperature.

The molecular structure of (I) at 4.5 K is presented in Fig. 1 and plots of the Rietveld refinement results are shown in Fig. 2. The derived molecular dimensions are given in Table 1 and in general there is good agreement with results from density functional theory calculations (Dobado et al., 1999), molecular mechanics calculations (Allinger & Fan, 1993) and electron diffraction (Hargittai & Hargittai, 1974). The main exception is the O—S—O bond angle, which is some 2° smaller in the crystal structure, although good agreement is maintained for the C—S—C bond angle. The S—C bond length is also shorter by some 0.2 Å in the 4.5 K structure, which is a small but still significant discrepancy.

The difference in molecular conformation between the crystal and gas-phase structures is presumably a consequence of weak hydrogen bonding in the former. The effect of these non-bonded interactions was also reported by Machida et al. (1979) following analysis of the vibrational spectra recorded on both normal and perdeuterodimethyl sulfone. Fig. 3 shows the hydrogen bonding in (I) at 4.5 K and geometric details are given in Table 2. The C—S—C plane of each molecule is aligned parallel to the crystallographic c axis and these sheets are cross-linked by hydrogen bonds. Similar behaviour has been reported in the corresponding selenone structure (Dikarev et al., 2003), which also shows hydrogen bonding in the solid state.

In the crystal structure, the S atom lies on a site of symmetry m2m, and the C and O atoms, along with one D atom, lie on mirror planes.

Related literature

For related literature, see: Allen (2002); Allinger & Fan (1993); Brochier (1977); Clever & Westrum (1970); Dikarev et al. (2003); Dobado et al. (1999); Hargittai & Hargittai (1974); Langs et al. (1970); Larson & Von Dreele (1994); Machida et al. (1979); Sands (1963).

S2. Experimental

(CD₃)₂SO₂ (1 g) (99 atom% D; Sigma Aldrich Chemical Ltd.) was sealed in an 8 mm diameter vanadium sample can and loaded into a standard vanadium-tailed 'orange' cryostat (Brochier, 1977). Data were recorded using the high-resolution powder diffractometer (HRPD) at the ISIS Facility, Rutherford Appleton Laboratory, UK, at backscattering 2θ = 168° over the time-of-flight range 30–130 ms, corresponding to a d-spacing range of some 0.6–2.6 Å. The sample has a waxy appearance and data were recorded at 293 K for a brief period of 15 µAh (ca 25 min) to check first the crystallinity of the powder. [Refined lattice constants at 293 K are a = 7.3276 (1) Å, b = 8.0182 (1) Å, c = 7.3457 (1) Å and V = 431.60 (1) ų.] The sample was then cooled to 4.5 K and measurements made for a period of 200 µAh (ca 6 h).

S3. Refinement

The diffraction data were analysed using the Rietveld method implemented in the GSAS program suite (Larson & Von Dreele, 1994). The peak shape was modelled using the type 3 time-of-flight profile function comprising the convolution of back-to-back exponentials with a pseudo-Voigt function. The background was modelled using a shifted Chebyshev function comprising ten terms. The structure was refined without the use of constraints and using anisotropic displacement parameters for all atoms. Parameters for the starting model were taken from the ambient-temperature structure reported by Sands (1963), including estimated positions for the D atoms assuming a C—D bond length of 1.095 Å.

Computing details

Data collection: ISIS instrument control program (ICP); cell refinement: GSAS (Larson & Von Dreele, 1994); data reduction: Standard HRPD normalisation routines; program(s) used to solve structure: Please provide information; program(s) used to refine structure: GSAS; molecular graphics: DIAMOND (Brandenburg & Putz, 2004); software used to prepare material for publication: GSAS.

Figures

	Fig. 1. The molecular structure of (I) at 4.5 K, showing the atom-numbering scheme for the asymmetric unit. Displacement ellipsoids are drawn at the 50% probability level. Unlabelled atoms are related to labelled atoms by the symmetry operator ? [Please complete]
	Fig. 2. The final Rietveld plot of (I) at 4.5 K, showing the calculated (line), observed (open circles) and difference (lower) profiles. Vertical bar markers indicate calculated Bragg peak positions. The equivalent d-spacing range corresponds to 0.68–2.38 Å (0.73–1.04 Å inset).
	Fig. 3. The hydrogen bonding (dashed lines) in the crystal structure of (I), viewed along the crystallographic a direction.

Perdeuterodimethyl sulfone

Crystal data

C2D6O2S	Z = 4
Mr = 100.15	Dx = 1.639 Mg m^−3
Orthorhombic, Cmcm	Melting point: 382 K
Hall symbol: -C 2c 2	Time-of-flight neutron radiation, λ = 1.24-5.36 Å
a = 7.12899 (2) Å	T = 5 K
b = 7.93194 (3) Å	Particle morphology: irregular powder
c = 7.17658 (2) Å	white
V = 405.81 (1) Å^3	cylinder, 25 × 8 mm

Data collection

HRPD diffractometer	Scan method: tof 30-130 ms
Specimen mounting: standard cylindrical vanadium sample holder

Refinement

Least-squares matrix: full	Profile function: GSAS TOF Profile function number 3
Rp = 0.027	48 parameters
Rwp = 0.037	0 restraints
Rexp = 0.019	0 constraints
R(F^2) = 0.10726	Weighting scheme based on measured s.u.'s
χ^2 = 3.920	(Δ/σ)max = 0.01
6240 data points	Background function: Shifted Chebyshev function (ten terms)
Excluded region(s): None	Preferred orientation correction: None

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

	x	y	z	Uiso*/Ueq
C1	0.0	0.28136 (13)	0.44279 (15)	0.00963
D1	0.0	0.20423 (15)	0.56675 (18)	0.02758
D2	0.12576 (10)	0.36063 (10)	0.43668 (12)	0.02347
O1	0.17219 (15)	0.04929 (15)	0.25	0.01115
S1	0.0	0.1443 (4)	0.25	0.00779

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
C1	0.0121 (6)	0.0070 (5)	0.0097 (6)	0.0	0.0	−0.0014 (6)
D1	0.0421 (7)	0.0228 (6)	0.0178 (5)	0.0	0.0	0.0018 (6)
D2	0.0194 (4)	0.0160 (4)	0.0350 (5)	−0.0058 (4)	0.0013 (4)	−0.0065 (5)
O1	0.0081 (6)	0.0098 (6)	0.0156 (6)	0.0031 (4)	0.0	0.0
S1	0.0059 (16)	0.0019 (15)	0.0155 (17)	0.0	0.0	0.0

Geometric parameters (Å, º)

C1—D1	1.0796 (13)	D2—D1	1.7928 (13)
C1—D2	1.0959 (8)	D2—D2^i	1.7931 (14)
C1—D2^i	1.0959 (8)	D2—S1	2.354 (2)
C1—S1	1.760 (2)	O1—S1	1.4405 (19)
D1—D2	1.7928 (13)	O1—O1	2.4551 (15)
D1—S1	2.3223 (14)
D1—C1—D2	110.98 (9)	C1—S1—C1	103.69 (16)
D1—C1—D2^i	110.98 (9)	C1—S1—O1	108.86 (4)
D1—C1—S1	107.33 (12)	C1—S1—O1^i	108.86 (4)
D2—C1—D2^i	109.79 (12)	C1—S1—O1	108.86 (4)
D2—C1—S1	108.84 (9)	C1—S1—O1^i	108.86 (4)
D2^i—C1—S1	108.84 (9)	O1—S1—O1	116.9 (2)

Symmetry code: (i) −x, y, z.

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
C1—D1···O1	1.08 (1)	2.70 (1)	3.6395 (14)	146 (1)
C1—D2···O1	1.10 (1)	2.47 (1)	3.4485 (13)	148 (1)