Ab initio structure determination of m-toluidine by powder X-ray diffraction

The powder X-ray diffraction pattern of the crystalline phase of m-toluidine has been recorded with a sensitive curved detector (CPS120) at 150 K. The structure has been solved by real-space methods (simulated annealing) followed by Rietveld refinements with phenyl rings as rigid bodies and with soft constraints on bond lengths for peripheral atoms. The cell is monoclinic with space group P2₁/c and Z = 8. Equivalent molecules form chains along c. The crystalline cohesion is achieved by N—H...N hydrogen bonds between neighbouring chains of non-equivalent molecules and by van der Waals interactions of neighbouring chains of equivalent molecules. The hydrogen-bonding network has been confirmed by lattice-energy minimization. Anisotropic strain effects of the cell have been calculated. The directions of the minimal strains correspond to the directions of the hydrogen bonds. An explanation of the difficulty to crystallize the metastable phase is given.

S1. Experimental

Liquid m-toluidine is commercially available (Aldrich Chemical Company) with a purity rate of 99%. Contacts with air or light easily induce an oxidation which gives rise to a typical brown-coloured compound. Furtherpurifications are necessary and they consist of two stages removal of organic and inorganic impurities. The compound is then distilled under low pressure, in order to get a colourless compound. A cold-nitrogen blower (Cryostream Oxford Cryosystem) allowed to access to low temperatures with temperature fluctuations within 0.5 K and was used for the in situ crystallization according to the following procedure. The liquid m-toluidine is quenched at 130 K by laying the cold stream on the capillary. After the temperature stabilization, the sample is heated until 220 K with an heating rate of 6 K/min. It is kept at this temperature during about 15 h in order to get a complete crystallization of the powder. Finally, the sample is cooled at the experiment temperature.

S2. Refinement

To determine the lattice parameters of m-toluidine at 150 K, the automatic powder indexing program N-TREOR (Altomare et al., 2000) implanted on EXPO (Altomare et al., 1999) has first been used. With the 25 first reflections of the pattern, no solution was found, suspecting the possibility of several phases in the powder. With the program WinPlotr (Roisnel & Rodriguez-Carvajal, 2002), the profiles of the 38 reflections with a 2-theta angle lower than 36 degrees were individually refined with a pseudo-Voigt function in order to obtain their 2? positions. The 2-theta angles of the 15 more intense reflections are introduced in the program TREOR (Werner et al., 1985) and these lines are completely indexed with a monoclinic cell with the following parameters: a = 24.815, b = 5.799, c = 8.743 A, beta = 100.03 degrees, V = 1238.8 A3. The calculated figures of merit are: M(15) = 31, F(15) = 43. (0.038, 94) (de Wolff, 1968; Smith & Snyder, 1979). It is to notice that the relation between the lattice parameters 2a cos(beta) + c = 0 is nearly verified leading to a pseudo-B-faced orthorhombic cell. As a consequence, the two reflections in the monoclinic cell (h, k, l) and (-h-l, k, l), with h and l > 0, have approximately the same d-spacing. The powder diffraction pattern from 12 to 72 degrees 2-theta, was subsequently refined with this cell and resolution constraints in the monoclinic space group P2/m (a space group without systematic extinctions) using the "profile matching" option [Le Bail et al. (1988), profile refinement] of the program FullProf (Rodriguez-Carvajal, 2001). Five lines with low intensity are not indexed with the previous cell and they can correspond to a second phase due to an impurity (Fig. 1). With the program TREOR, it is possible to find a monoclinic cell with these five reflections, but, it seems more realistic to remove regions of the pattern corresponding to these reflections (about 0.20 degree 2-theta around the maximum of the peaks) to avoid to introduce this cell in the refinements. A possibility of chemical for this impurity is the m-nitrotoluene resulting from the oxidation of the amine group of m-toluidine.

Computing details

Data collection: PEAKOC; program(s) used to solve structure: FOX; program(s) used to refine structure: FULLPROF; molecular graphics: ORTEP-3; software used to prepare material for publication: SHELX97 and PARST.

Figures

Figure 1: The 4 more intense Bragg reflections due to an impurity. Dots correspond to observed data, the upper solid line is the calculated profile for m-toluidine and the lower line is the difference between observed and calculated pattern. The fine vertical bars indicate the position of Bragg peaks for m-toluidine and the thick ones for the impurity. Figure 2: Final Rietveld plot of m-toluidine at 150 K. Observed data points are indicated by dots, the best-fit profile (upper trace) and the difference pattern (lower trace) are solid lines. The vertical bars correspond to the position of Bragg peaks. Figure 3: Atomic numbering and molecular structure of m-toluidine at 150 K. The dashed bond represents the hydrogen bond between the two molecules of the asymmetric unit. Figure 4: Projection of the unit cell of m-toluidine along b. For the sake of clarity, H atoms have been omitted. Dashed lines correspond to hydrogen bonds and the arrows show the direction of the N—H···N bonds. Figure 5: Calculated strain in the monoclinic plane of m-toluidine at 150 K.

3-methylbenzenamine

Crystal data

C7H9N	F(000) = 464
Mr = 107.16	Dx = 1.142 Mg m^−3
Monoclinic, P21/C	Melting point: 232 K
Hall symbol: -P 2ybc	Cu Kα radiation, λ = 1.540560 Å
a = 24.8727 (6) Å	µ = 0.52 mm^−1
b = 5.80729 (17) Å	T = 150 K
c = 8.7615 (3) Å	white
β = 100.0624 (17)°	cylinder, diameter 0.7 mm, ? × ? × ? mm
V = 1246.07 (7) Å^3	Specimen preparation: Prepared at 220 K K
Z = 8

Data collection

Enraf Nonius FR 590 diffractometer	Data collection mode: transmission
Radiation source: Sealed X-ray tube	Scan method: Stationary detector
Quartz (Curved Monochromator) monochromator	Absorption correction: for a cylinder mounted on the ϕ axis ?
Specimen mounting: Lindemann Glass Capillary	Tmin = ?, Tmax = ?

Refinement

Rp = 3.499	Profile function: Pseudo-Voigt
Rwp = 5.044	92 parameters
Rexp = 1.293	8 restraints
RBragg = 4.039
χ^2 = NOT FOUND	Background function: linear interpolation between 34 points
3946 data points	Preferred orientation correction: March-Dollase function, axis [0,0,1], G1=0.914(3), G2=0
Excluded region(s): 0.29 to 12.0 : bad calibration of the CPS120 90.0 to 120.0 : peak intensities too low 18.55 to 18.95, 19.50 to 19.80, 21.10 to 21.45 and 23.20 to 23.60 : presence of peaks due to an impurity

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

	x	y	z	Uiso*/Ueq
C11	0.9127 (3)	−0.0803 (8)	0.3811 (6)	0.026 (2)*
C12	0.8827 (3)	−0.2258 (8)	0.4613 (6)	0.026 (2)*
C13	0.8320 (3)	−0.1557 (8)	0.4916 (6)	0.026 (2)*
C14	0.8112 (3)	0.0600 (8)	0.4420 (6)	0.026 (2)*
C15	0.8412 (3)	0.2057 (8)	0.3620 (6)	0.026 (2)*
C16	0.8919 (3)	0.1355 (8)	0.3315 (6)	0.026 (2)*
C17	0.9646 (4)	−0.2014 (19)	0.3508 (18)	0.026 (2)*
N1	0.7997 (5)	−0.2785 (13)	0.5758 (14)	0.026 (2)*
H12	0.8973	−0.3773	0.4961	0.026 (2)*
H14	0.7756	0.1092	0.4634	0.026 (2)*
H15	0.8266	0.3571	0.3271	0.026 (2)*
H16	0.9130	0.2379	0.2752	0.026 (2)*
H171	0.9885	−0.2163	0.4503	0.026 (2)*
H172	0.9803	−0.1000	0.2794	0.026 (2)*
H173	0.9532	−0.3490	0.3032	0.026 (2)*
HN11	0.769 (3)	−0.244 (16)	0.636 (10)	0.026 (2)*
HN12	0.815 (3)	−0.426 (7)	0.541 (8)	0.026 (2)*
C21	0.5874 (3)	−0.0995 (8)	0.4826 (6)	0.034 (2)*
C22	0.6227 (3)	−0.2046 (8)	0.6037 (6)	0.034 (2)*
C23	0.6707 (3)	−0.0933 (8)	0.6718 (6)	0.034 (2)*
C24	0.6833 (3)	0.1231 (8)	0.6186 (6)	0.034 (2)*
C25	0.6479 (3)	0.2284 (8)	0.4975 (6)	0.034 (2)*
C26	0.6000 (3)	0.1171 (8)	0.4296 (6)	0.034 (2)*
C27	0.5337 (4)	−0.1961 (19)	0.3934 (15)	0.034 (2)*
N2	0.7026 (3)	−0.231 (2)	0.7874 (9)	0.034 (2)*
H22	0.6139	−0.3568	0.6410	0.034 (2)*
H24	0.7169	0.2013	0.6664	0.034 (2)*
H25	0.6567	0.3804	0.4602	0.034 (2)*
H26	0.5751	0.1910	0.3444	0.034 (2)*
H271	0.5325	−0.3586	0.4227	0.034 (2)*
H272	0.5346	−0.1754	0.2838	0.034 (2)*
H273	0.5044	−0.1071	0.4275	0.034 (2)*
HN21	0.692 (3)	−0.248 (15)	0.893 (4)	0.034 (2)*
HN22	0.742 (10)	−0.215 (17)	0.782 (11)	0.034 (2)*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
?	?	?	?	?	?	?

Geometric parameters (Å, º)

C17—C11	1.534 (14)	C27—C21	1.531 (12)
C17—H171	0.970	C27—H271	0.980
C17—H172	0.987	C27—H272	0.977
C17—H173	0.974	C27—H273	0.982
N1—C13	1.380 (13)	N2—C23	1.420 (11)
N1—HN11	1.01 (8)	N2—HN21	1.01 (5)
N1—HN12	1.01 (6)	N2—HN22	1.01 (3)
C11—C17—H171	106.7	H271—C27—H273	111.7
C11—C17—H172	105.8	H272—C27—H273	112.4
H171—C17—H172	112.1	C23—N2—HN21	120 (3)
C11—C17—H173	106.5	C23—N2—HN22	110 (3)
H171—C17—H173	113.2	HN21—N2—HN22	118 (4)
H172—C17—H173	111.9	C12—C11—C17	110.5 (6)
C13—N1—HN11	137 (4)	C16—C11—C17	129.4 (4)
C13—N1—HN12	89 (4)	N1—C13—C12	125.6 (8)
HN11—N1—HN12	133 (6)	N1—C13—C14	114.3 (5)
C21—C27—H271	106.5	C22—C21—C27	127.2 (7)
C21—C27—H272	106.9	C26—C21—C27	112.8 (6)
H271—C27—H272	112.6	C24—C23—N2	128.1 (6)
C21—C27—H273	106.3	C22—C23—N2	111.9 (6)
C17—C11—C12—C13	−175.6 (7)	C27—C21—C22—C23	179.2 (8)
C17—C11—C16—C15	174.7 (8)	C27—C21—C26—C25	−179.4 (7)
C11—C12—C13—N1	−176.7 (8)	C21—C22—C23—N2	−177.1 (7)
N1—C13—C14—C15	177.0 (7)	N2—C23—C24—C25	176.7 (7)
HN11—N1—C13—C12	158 (7)	HN21—N2—C23—C22	−76 (5)
HN11—N1—C13—C14	−19 (7)	HN21—N2—C23—C24	107 (5)
HN12—N1—C13—C12	−30 (4)	HN22—N2—C23—C22	140 (5)
HN12—N1—C13—C14	153 (4)	HN22—N2—C23—C24	−36 (5)

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
N1—HN11···N2	1.01 (8)	2.30 (8)	3.302 (15)	170 (6)
N2—HN22···N1^i	1.01 (3)	2.71 (3)	3.176 (13)	108 (3)

Symmetry code: (i) x, −y−1/2, z+1/2.