Rotationally disordered phase of 1,3-dibromo-5-iodo-2,4,6-trimethyl­benzene at 293 K

In the crystal state at room temperature, the mol­ecule of dibromo­iodo­mesitylene (1,3-dibromo-5-iodo-2,4,6-trimethyl­benzene), C₉H₉Br₂I, is prone to strong disorder, apparently involving only the three halogen sites (occupied identically by 66.7% Br and 33.3% I). This disorder, of the rotational type according to previously published NMR measurements, corresponds to fast 2π/3 stochastic in-plane reorientations of the whole mol­ecule between three discernable locations. This kind of rotational disorder can be revealed for the first time by diffractometry thanks to the C_2v idealized mol­ecular symmetry of the title compound, although it has been indirectly suspected at room temperature in other trihalogeno­mesityl­enes of similar crystal packing but of D_3h mol­ecular symmetry. The average endocyclic angles facing the Br/I sites and the methyl groups are 124.14 (6) and 115.85 (2)°, respectively. The angle between the normal to the aromatic ring and the normal to the (100) plane is 4.1°. TLS analysis indicates that only the aromatic ring and the methyl groups behave as a rigid body with respect to the thermal librations.

S1. Comment

Most benzene compounds hexasubstituted by halogens, methyl groups or other small radicals crystallize at room temperature in the monoclinic space group P2₁/n, Z = 2, according to the hexachlorobenzene structure type (Brown & Strydom, 1974; Reddy et al., 2006). Partial compilations of previous contributions can be found in Kitaigorodsky (1973), Tazi et al. (1995), Brock & Fu (1997) and Reddy et al. (2006). The packing of these compounds is dominated by intermolecular π–π interactions and the molecules are arranged in stacks (at distances typically ranging from 3.8 to 4.3 Å) parallel to the monoclinic unique axis. In addition, they form corrugated (110) molecular planes, with the halogen–halogen interactions within these planes turning out to be weaker than the π–π inter-stack forces. Another crystal packing is observed, too, at room temperature, especially if the molecules display a threefold axis of symmetry: they are stacked into a triclinic unit cell, space group P1, Z = 2, deriving from the hexamethylbenzene structure type (Hamilton et al., 1969; Le Maguères et al., 2001). This is the case for the trihalogenomesitylene series (1,3,5-trihalogeno-2,4,6-trimethylbenzene; Table 1), of which only trichloromesitylene (TCM) (Tazi et al., 1995; Hernandez et al., 2006), tribromomesitylene (TBM) (Meinnel et al., 2000; Bosch & Barnes, 2002) and triiodomesitylene (TIM) (Boudjada et al., 2001, 2002; Bosch & Barnes, 2002) have been characterized so far. The methyl groups in these compounds experience relatively small hindering potentials and interesting tunnelling properties have been reported (Meinnel et al., 1992, 1995, 2000; Boudjada et al., 2002). What happens if one of the halogen atoms is replaced by another atom, breaking the idealized D_3h molecular symmetry of these molecules? If it is an H atom, as encountered for dibromomesitylene (DBM) (Hernandez et al., 2003), the ordered low-temperature phase (the phase is disordered above ~297 K) is monoclinic, space group P2₁/n, Z = 4. Unexpectedly, however, the methyl group located between the two Br atoms is a quasi-free rotor with H atoms highly delocalized in a sixfold potential, whereas the rotation of the other two methyl groups is quasi-forbidden (Meinnel et al., 1995; Plazanet et al., 2002). In order to establish for the trihalogenomesitylene series the impact of small changes on the molecular symmetry, we report here the crystal structure of 1,3-dibromo-5-iodo-2,4,6-trimethylbenzene, (I) (dibromoiodomesitylene, DBIM), obtained by single-crystal X-ray diffraction at 293 K. This compound can indeed be viewed as a TBM with one Br atom substituted by I, or alternatively as a TIM with two I atoms substituted by Br, the idealized symmetry of the isolated molecule (ignoring H atoms) decreasing in both cases from D_3h (-6m2) to C_2v (mm2).

The best refinement with 54 geometric soft restraints (see Refinement) led to the molecular structure shown in Fig. 1. The crystallographically inequivalent halogen sites are identically constituted by 67% Br and 33% I (see Refinement), the other atoms apparently not being affected by any kind of disorder. NMR measurements at room temperature showed a long time ago that in this family of compounds - at least for the hydrogenated materials - there is evidence for a dynamic reorientation of the molecules by jumps of 2mπ/6 within their plane [Please define m], at a frequency in the MHz range (Eveno & Meinnel, 1966). The disorder in (I) therefore corresponds to fast 2π/3 stochastic in-plane reorientations of the whole molecule between three discernable locations, inducing the apparent overlap of Br and I atoms, according to a 2/3:1/3 ratio. This novel feature has been characterized by diffractometry thanks to the idealized (instantaneous) C_2v molecular symmetry of the isolated molecule (ignoring H atoms), which is time-averaged towards an apparent idealized D_3h symmetry by means of the diffraction probe. The same kind of rotational disorder, although suspected at room temperature in TCM (Tazi et al., 1995; Hernandez et al., 2006), TBM and TIM, has not been directly proved so far, due to the ternary symmetry of these latter molecules.

In (I), significant deformation of the angles of the benzene ring from a regular D_6h (6/mmm) hexagon is observed. The average endocyclic angle is 124.14 (6)° around the C atoms linked to Br/I sites, and an average value of 115.85 (2)° is observed for the C atoms bonded to the methyl groups, making the aromatic ring a distorted hexagon fulfilling approximately the D_{3 h} symmetry. This deformation is close to that found in TCM, TBM and TIM compounds at the same temperature and is in agreement with systematic trends establishing that the endocyclic angle facing the C atom bearing the most electronegative substituent is enlarged (Domenicano et al., 1975). After the restrained refinement, the mean values for the C_ar—Br and C_ar—I distances are 1.907 (9) and 2.106 (6) Å, respectively.

The crystal packing of (I) can be described as a stacking of (100) molecular layers at x/a ~1/4 and 3/4, forming zigzag molecular columns propagating along the a axis (Figs. 2 and 3). The angle between the normal to the aromatic plane and the normal to the (100) plane is 4.1°. Within each column, the arrangement is approximately `antiferroelectric': a given molecule is sandwiched between two molecules at ~4.06 Å generated by inversion centres and belonging to adjacent layers. A Br/I site is more or less directly below the methyl groups and vice versa. The symmetry centres are located on the a axis, but the centres of mass of the molecules are shifted away from the latter axis by 0.46 Å (Fig. 3), which explains the slightly zigzag shape of the molecular rows. Within each (100) layer, one probe molecule is surrounded by six neighbours in such a way that characteristic triangular halogen···halogen and Me···Me intermolecular contacts are formed, the three corresponding contact distances being very similar (e.g. mean value of 3.61 Å for I···I). For molecules in different layers, the shortest intermolecular contact distances, compared with those between molecules within the same layer, are significantly shortened for C_Me···C_Me contacts, almost unchanged for C_Me···halogen contacts and significantly increased for halogen···halogen contacts. The same structural topology is encountered at room temperature in TBM (Meinnel et al., 2000) and TIM (Boudjada et al., 2001), but also in TCM below 160 K (lowest temperature phase, or phase IV; Hernandez et al., 2006).

The C atoms of the ring display U_eq displacement parameters significantly lower (mean 0.039 Ų) than those of the substituted atoms (mean values of 0.057, 0.048 and 0.059 Ų for Br, I and C_Me, respectively). In order to clarify this point, we have performed a conventional TLS analysis using the CRYSTALS program (Betteridge et al., 2003). The overall rigid-body motion tensors T, L and S (Schomaker & Trueblood, 1968) were least-squares fitted to the individual anisotropic displacement parameters. At first, we included only the C atoms in the rigid-body treatment, and the reliability factor for the U values was R = 0.054, revealing the rigid character of the body constituted by the aromatic ring and the methyl groups. Clearly, the translation and screw tensors are negligible, the thermal motion of the aforementioned rigid body being essentially of librational origin. The eigenvalues of the libration tensor are L₁₁ = 9°² (in-plane libration), and L₂₂ = 19°² and L₃₃ = 22°² (both corresponding to off-plane librations). In the second step of the TLS analysis, we added the Br and I atoms, i.e. we took into account all the non-H atoms. The reliability factor for the U values increased substantially to R = 0.131, indicating that, relative to the aromatic ring and methyl groups, the halogen substituents do not behave rigidly.

S2. Experimental

Compound (I) was synthesized at 353 K as follows. A mixture of sulfuric acid (20 ml), nitric acid (30 ml) and acetic acid (150 ml) was added dropwise into a balloon flask containing dibromomesitylene (0.07 mol), iodine (0.04 mol) and acetic acid (200 ml). After 1 h of reaction, a powder of dibromoiodomesitylene was obtained. Subsequent recrystallization of a saturated solution in chloroform yielded pure colourless crystals of (I) suitable for X-ray analysis. High-resolution mass spectrometry (electron energy 70 eV, temperature source 418 K) indicated m/z = 401.8114 versus a theoretical value of 401.81157. NMR decoupled spectra for protons were recorded for solutions of (I) and reveal that only one material is present. Solid-state C¹³ NMR spectra carried out with a rotation at the magic angle (speed 6000 Hz) demonstrate that there is only one component at each molecular site within the entire crystal.

S3. Refinement

The triclinic unit-cell setting, although not conventional, was chosen for the sake of comparison with the previously published isomorphous compounds TCM, TBM and TIM (Table 1). For the sake of stability, the refinement was undertaken by initially fixing the occupancies of the Br and I atoms to 0.667 and 1/3, respectively. These values are in agreement with the compound stoichiometry and with the above-mentioned dynamic disorder, and they were checked once convergence was reached (see below). Due to the disorder, 54 geometric soft restraints were applied in order to maintain a chemically reasonable molecule, according to the distances and angles calculated for the isolated molecule by means of density functional theory quantum chemistry calculations (MPW1PW91 function, LanL2DZ basis set; Software and reference?): C_r—Br = 1.920 (5), C_r—I = 2.120 (5), C_r—C_r = 1.40 (1) and C_r—C_Me = 1.50 (1) Å, C_r—C_r—Br(I) = 117.0 (3), C_r—C_r—C_r = 124.0 (1) or 116.0 (1)° in front of a halogen site or methyl group, respectively, and C_r—C_r—C_Me = 122.0 (4)°, with planarity = 0.02 Å for all non-H atoms. The final refinement cycles included the atomic coordinates and the anisotropic displacement parameters for all non-H atoms. Methyl H atoms, not detectable in the difference Fourier map, were generated geometrically, with their positions riding on their parent C atoms and their isotropic displacement parameters set to one least-squares parameter. Starting from the final structure model, refinement of the Br and I occupancies for each halogen site (plus the scale factor) led to the following values: 0.671 (3)/0.3351 (16), 0.663 (3)/0.3309 (17) and 0.666 (3)/0.3327 (16) for sites 1, 3 and 5, respectively, without changing the agreement factors at all (i.e. not significantly different from the reported model with 2/3:1/3 disorder).

Computing details

Data collection: COLLECT (Nonius, 2001); cell refinement: DIRAX (Duisenberg et al., 2003); data reduction: EVALCCD (Duisenberg et al., 2003); program(s) used to solve structure: SIR97 (Altomare et al., 1999); program(s) used to refine structure: CRYSTALS (Betteridge et al., 2003); molecular graphics: CAMERON (Watkin et al., 1996); software used to prepare material for publication: CRYSTALS (Betteridge et al., 2003).

Figures

	Fig. 1. View of the molecule of DBIM, (I), at 293 K, showing the atom-numbering scheme. Displacement ellipsoids are drawn at the 50% probability level and H atoms are shown as small spheres of arbitrary radii.
	Fig. 2. The crystal packing of (I) at 293 K, along the b axis. H atoms have been omitted for the sake of clarity.
	Fig. 3. The crystal packing of (I) at 293 K, along the a axis. H atoms have been omitted for the sake of clarity. Notice within the (100) planes the pseudo-hexagonal environment around a given molecule.

1,3-dibromo-5-iodo-2,4,6-trimethylbenzene

Crystal data

C9H9Br2I	Z = 2
Mr = 403.84	F(000) = 371.964
Triclinic, P1	Dx = 2.459 Mg m^−3
a = 7.895 (1) Å	Mo Kα radiation, λ = 0.71073 Å
b = 9.3180 (9) Å	Cell parameters from 10298 reflections
c = 9.3130 (11) Å	θ = 2.9–30.0°
α = 60.187 (8)°	µ = 10.21 mm^−1
β = 72.998 (9)°	T = 293 K
γ = 95.10 (1)°	Plate, colourless
V = 545.33 (13) Å^3	0.29 × 0.28 × 0.18 mm

Data collection

Nonius KappaCCD area-detector diffractometer	2995 independent reflections
Radiation source: X-ray tube	1629 reflections with I > 3σ(I)
Horizontally mounted graphite crystal monochromator	Rint = 0.045
ω/2θ scans	θmax = 30.0°, θmin = 3.6°
Absorption correction: multi-scan (SADABS; Sheldrick, 1996)	h = −10→9
Tmin = 0.06, Tmax = 0.16	k = −11→13
5479 measured reflections	l = −13→13

Refinement

Refinement on F	H-atom parameters constrained
Least-squares matrix: full	Method, part 1, Chebychev polynomial (Watkin, 1994, Prince, 1982) [weight] = 1.0/[A0*T0(x) + A1*T1(x) ··· + An-1]*Tn-1(x)], where Ai are the Chebychev coefficients listed below and x = F /Fmax. Method = Robust Weighting (Prince, 1982) W = [weight] * [1-(deltaF/6*sigmaF)^2]^2 Ai are: 1.38 0.469 1.07
R[F^2 > 2σ(F^2)] = 0.040	(Δ/σ)max = 0.000206
wR(F^2) = 0.044	Δρmax = 1.21 e Å^−3
S = 1.18	Δρmin = −1.01 e Å^−3
1629 reflections	Extinction correction: Larson (1970), Eq. 22
138 parameters	Extinction coefficient: 12 (3)
54 restraints

Special details

Experimental. The structure was successfully solved using direct methods (SIR97 program, Altomare et al., 1999) anticipating the rotational 120°-jumps of the whole molecule within its plane. A pseudo-atom of niobium (41 electrons) was used for mimicing each disordered halogen site constituted by 67% of bromine and 33% of iodine.

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

	x	y	z	Uiso*/Ueq	Occ. (<1)
Br1	0.7753 (12)	0.4473 (5)	−0.1329 (7)	0.0590	0.6670
I1	0.7722 (13)	0.4701 (6)	−0.1390 (7)	0.0466	0.3330
Br3	0.7668 (15)	−0.0906 (7)	−0.2533 (7)	0.0565	0.6670
I3	0.7631 (19)	−0.1002 (8)	−0.2681 (9)	0.0511	0.3330
Br5	0.7121 (13)	−0.2268 (6)	0.4226 (6)	0.0564	0.6670
I5	0.7119 (15)	−0.2428 (6)	0.4473 (7)	0.0454	0.3330
C12	0.7765 (11)	0.2802 (8)	−0.3524 (8)	0.0559
C14	0.7322 (12)	−0.3195 (8)	0.1431 (9)	0.0599
C16	0.7395 (12)	0.1641 (9)	0.2504 (9)	0.0616
C1	0.7589 (8)	0.2097 (4)	−0.0471 (5)	0.0392
C2	0.7632 (8)	0.1578 (5)	−0.1642 (4)	0.0399
C3	0.7569 (8)	−0.0131 (5)	−0.0980 (5)	0.0407
C4	0.7405 (8)	−0.1324 (4)	0.0760 (5)	0.0388
C5	0.7371 (8)	−0.0693 (4)	0.1823 (4)	0.0382
C6	0.7465 (9)	0.1014 (5)	0.1267 (5)	0.0394
H121	0.7789	0.3967	−0.3735	0.21 (3)*
H122	0.8923	0.2885	−0.4404	0.21 (3)*
H123	0.6673	0.2369	−0.3683	0.21 (3)*
H141	0.7342	−0.3394	0.0466	0.21 (3)*
H142	0.8417	−0.3448	0.1741	0.21 (3)*
H143	0.6164	−0.3971	0.2522	0.21 (3)*
H161	0.7304	0.0664	0.3693	0.21 (3)*
H162	0.8543	0.2588	0.1956	0.21 (3)*
H163	0.6293	0.2083	0.2677	0.21 (3)*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
Br1	0.094 (2)	0.0204 (9)	0.0652 (15)	0.0239 (9)	−0.0396 (14)	−0.0192 (8)
I1	0.072 (2)	0.0275 (15)	0.0497 (17)	0.0201 (16)	−0.0293 (15)	−0.0239 (14)
Br3	0.0792 (19)	0.0605 (14)	0.0499 (14)	0.0163 (12)	−0.0248 (13)	−0.0445 (12)
I3	0.074 (2)	0.0557 (17)	0.0447 (16)	0.0202 (16)	−0.0276 (15)	−0.0393 (13)
Br5	0.0742 (17)	0.0561 (16)	0.0270 (11)	0.0144 (14)	−0.0272 (12)	−0.0116 (10)
I5	0.070 (2)	0.0328 (11)	0.0270 (14)	0.0140 (11)	−0.0253 (15)	−0.0085 (10)
C12	0.070 (5)	0.050 (4)	0.042 (4)	0.019 (4)	−0.025 (4)	−0.019 (3)
C14	0.079 (6)	0.037 (4)	0.060 (5)	0.013 (4)	−0.028 (4)	−0.025 (3)
C16	0.082 (6)	0.067 (5)	0.052 (4)	0.027 (4)	−0.031 (4)	−0.040 (4)
C1	0.038 (3)	0.037 (3)	0.045 (3)	0.006 (3)	−0.017 (3)	−0.024 (3)
C2	0.031 (3)	0.041 (3)	0.038 (3)	0.009 (3)	−0.012 (3)	−0.016 (3)
C3	0.037 (3)	0.044 (3)	0.039 (3)	0.007 (3)	−0.010 (3)	−0.025 (3)
C4	0.033 (3)	0.035 (3)	0.040 (3)	0.008 (3)	−0.010 (2)	−0.018 (3)
C5	0.038 (3)	0.041 (3)	0.035 (3)	0.008 (3)	−0.013 (3)	−0.022 (3)
C6	0.041 (3)	0.042 (3)	0.039 (3)	0.013 (3)	−0.017 (3)	−0.023 (3)

Geometric parameters (Å, º)

Br1—C1	1.920 (4)	C3—C4	1.392 (5)
I1—C1	2.114 (5)	C4—C5	1.376 (5)
Br3—C3	1.898 (4)	C5—C6	1.393 (5)
I3—C3	2.100 (5)	C12—H121	1.000
Br5—C5	1.904 (4)	C12—H122	1.000
I5—C5	2.105 (5)	C12—H123	1.000
C12—C2	1.505 (6)	C14—H141	1.000
C14—C4	1.519 (6)	C14—H142	1.000
C16—C6	1.520 (7)	C14—H143	1.000
C1—C2	1.389 (5)	C16—H161	1.000
C1—C6	1.384 (5)	C16—H162	1.000
C2—C3	1.382 (5)	C16—H163	1.000
Br1—C1—C2	118.04 (19)	C16—C6—C5	122.1 (2)
I1—C1—C2	118.28 (18)	C16—C6—C1	122.0 (2)
Br1—C1—C6	117.90 (19)	C5—C6—C1	115.87 (9)
I1—C1—C6	117.67 (18)	C2—C12—H121	109.3
C2—C1—C6	124.05 (9)	C2—C12—H122	109.5
C12—C2—C1	122.3 (2)	H121—C12—H122	109.5
C12—C2—C3	121.9 (2)	C2—C12—H123	109.6
C1—C2—C3	115.82 (9)	H121—C12—H123	109.5
Br3—C3—C2	117.8 (2)	H122—C12—H123	109.5
C2—C3—I3	117.8 (2)	C4—C14—H141	109.4
Br3—C3—C4	118.0 (2)	C4—C14—H142	109.1
C2—C3—C4	124.19 (9)	H141—C14—H142	109.5
I3—C3—C4	118.0 (2)	C4—C14—H143	109.9
C14—C4—C3	122.3 (2)	H141—C14—H143	109.5
C14—C4—C5	121.9 (2)	H142—C14—H143	109.5
C3—C4—C5	115.85 (9)	C6—C16—H161	109.4
Br5—C5—C4	117.82 (18)	C6—C16—H162	109.4
I5—C5—C4	117.87 (18)	H161—C16—H162	109.5
Br5—C5—C6	117.99 (19)	C6—C16—H163	109.6
I5—C5—C6	117.94 (18)	H161—C16—H163	109.5
C4—C5—C6	124.17 (9)	H162—C16—H163	109.5