p-Phenyl­azoaniline hydro­chloride from powder data: protonation site and UV-visible spectra

The crystal structure of the title compound, (4-amino­phenyl)phenyl­diazenium chloride, C₁₂H₁₂N₃⁺·Cl^-, was determined from X-ray laboratory powder data, and the protonation on the azo group was confirmed by the neutron powder data. The cations form stacks along [100], while the chloride anions form hydrogen bonds to all three H atoms attached to N atoms. The absorption maximum of the crystalline salt is shifted bathochromically by 850 cm^-1, compared with that in ethanol solution.

S1. Comment

Aromatic azo compounds containing the amino group may be protonated either on the amino group or on the azo group, to give a tautomeric mixture (Cox & Buncel, 1975). In polar solvents capable of effective hydrogen-bond formation, the equilibrium is shifted towards the ammonium form, whereas in the solvents of lower polarity the azonium form prevails, because the dipole moment of ammonium cation is more than twice as large as that of the azonium cation (Liwo et al., 1994). The UV–visible spectra of solutions contain two peaks centered at 325 and 500 nm, attributable to the ammonium and azonium forms, respectively (Kuroda et al., 1980). Surprisingly, in the solid state most of these salts exist in azonium form, although ordinarily the crystal surrounding shifts tautomeric equilibria in the same way as the extremely polar solvents.

The cation of the title compound, (I), is closely planar, two phenyl rings form a 9.8 (5)° dihedral angle. Large s.u.'s for bond lengths and angles preclude their comparison to 2,4-diaminoazobenzene hydrochloride (Moreiras et al., 1981). The crystal packing is illustrated in Fig. 1; the dotted lines present the hydrogen bonds (Table 1). The neighboring molecules within the stack are related by inversion centers with interplanar distances of 3.34 (1) and 3.48 (1) Å. The bond-restrained Rietveld refinement of (I) on neutron diffraction data led to R_B = 0.14 and 0.18 for azonium and ammonium cations, respectively. \sch

The UV–visible absorption spectrum of ethanol solution of (I) and the reflection spectra of solid samples prepared by grinding (I) with BaSO₄ are presented in Fig. 2. When a dry sample was ground, the reflection spectrum exhibits the same intensity ratio of the long-wavelength (azonium) and short-wavelength (ammonium) absorption maxima as in the solution, thus the azonium form prevails in this specimen. However, if a sample is ground in the presence of 1 M HCL, the intensity of the azonium band increases and the ammonium maximum becomes diffuse and indistinct indicating that the tautomeric shift towards the azonium form occurs. This finding is in line with the well known fact that the salts of many p-aminoazo compounds can be prepared as the metastable yellow phases, which rapidly undergo into the dark-red crystalline form (Bershtein & Ginzburg, 1972). In the presence of HCL solution, the recrystallization of dispersed (I) can occur, and a stable crystalline azonium form results from the amorphous ammonium form. There are two most probable reasons why the crystal packing stabilizes the azonium form. First, the azonium cations allow a more uniform distribution of chloride anions in the unit cell, and the closest Cl···Cl contacts in (I) are 5.865 (6) Å, whereas in the structures of chlorides of simple aniline derivatives all anions are localized in the vicinity of NH₃⁺ groups, and multiple Cl···Cl contacts shorter than 5.2 Å are present (Colapietro et al., 1981; Ploug-Soerensen & Andersen, 1985; Linden et al., 1995). Second, the charge in the azonium cation is delocalized over the azo group and the amino-substituted ring, and such delocalization can be stabilized by the intermolecular π–π interactions within the stacks, whereas in the ammonium form the charge is localized on the NH₃⁺ group.

The main maximum at 19160 cm⁻¹ in the reflection spectrum has a shoulder at 24200 cm⁻¹, which can be attributed to Davydov splitting. When transferring from solution to solid state, azonium and ammonium bands demonstrate opposite wavelength shifts: bathochromic and hypsochromic, respectively. In solutions, both bands demonstrate blue shifts when polarity of the solvent increases, thus the red shift of the azonium band can be attributed only to the splitting of the energy levels of cations into the zones due to the π-π interactions within the stacks.

S2. Experimental

The title compound was prepared dissolving p-phenylazoaniline (Solvent Yellow 1) in hot 1 M HCl. The UV-visible spectra (absorption and reflection) were recorded on a Specord M-40 spectrophotometer (Carl Zeiss, Jena).

S3. Refinement

The monoclinic cell dimensions were determined with TREOR90 (Werner et al., 1985) and refined to M₂₀ = 31 and F₃₀ = 69 (0.009, 45) using the first 60 peaks from the X-ray pattern. Because of reflections overlapping, the space groups P2₁/n and P2₁/c were indistinguishable on the basis of systematic extinctions, and the final choice was made at the stage of structure determination. The neutron-diffraction experiment was carried out at the high-resolution multi-counter powder diffractometer G4.2 situated at the Orphee reactor of the Leon Brillouin Laboratory. The position and orientation of the cation was determined with the grid search procedure (Chernyshev & Schenk, 1998) on neutron data, the initial molecular model was built with MOPAC7 (Stewart, 1993). The position of the chlorine atom was determined by grid search on the X-ray data, with fixed position and orientation of the cation. Because of high background, the signal-to-noise ratio in the neutron diffraction pattern was significantly worse than that obtained in the X-ray experiment, thus the details of refinement on the neutron data are not discussed here. The X-ray diffraction profile and the difference between the measured and calculated profiles after the Rietveld redfinement are shown in Fig. 3, final R_B = 0.068. The chlorine atom was refined anisotropically, the C and N atoms isotropically, and the planarity of phenyl rings was constrained. H atoms were placed in geometrically calculated positions, with a common isotropic displacement parameter U_iso fixed at 0.076 Ų. The anisotropy of diffraction-line broadening was approximated by a quartic form in hkl (Popa, 1998). Final (Δ/σ)_max value is relatively high due to a correlation between the line-broadening parameters, whereas for the atomic positional parameters these values do not exceed 0.01.

Computing details

Data collection: home-written program; cell refinement: LSPAID (Visser, 1986); program(s) used to solve structure: MRIA (Zlokazov & Chernyshev, 1992); program(s) used to refine structure: MRIA; molecular graphics: PLUTON (Spek, 1992); software used to prepare material for publication: PARST (Nardelli, 1983).

Figures

	Fig. 1. Packing diagram showing atom numbering scheme (ellipsoids and spheres represent the 50% probability level).
	Fig. 2. UV-visible spectra of (I): ethanol solution (dotted line), solid grinded with BaSO4 without (dashed line) and with (solid line) 1 M HCl.
	Fig. 3. The Rietveld plot for (I), showing the observed and difference profiles. The reflection positions are shown above the difference profile.

4-Phenylazoaniline hydrochloride

Crystal data

C12H12N3^+·Cl^−	Z = 4
Mr = 233.70	F(000) = 488
Monoclinic, P21/c	Dx = 1.369 Mg m^−3
a = 7.400 (3) Å	Cu Kα radiation, λ = 1.5418 Å
b = 18.511 (5) Å	T = 295 K
c = 8.920 (3) Å	Particle morphology: plates
β = 111.91 (2)°	black
V = 1133.7 (9) Å^3	flat_sheet, 25 × 25 mm

Data collection

DRON-3M (Burevestnik, Russia) horizontal diffractometer	Data collection mode: reflection
Radiation source: X-ray sealed tube	Scan method: step
Ni filtered monochromator	2θmin = 8.40°, 2θmax = 70.00°, 2θstep = 0.02°
Specimen mounting: pressed in the specimen holder

Refinement

Refinement on Inet	109 parameters
Least-squares matrix: full with fixed elements per cycle	20 restraints
Rp = 0.030	0 constraints
Rwp = 0.040	H-atom parameters not refined
Rexp = 0.023	Weighting scheme based on measured s.u.'s
χ^2 = 3.062	(Δ/σ)max = 0.06
3201 data points	Background function: Chebyshev polynomial up to the 5th order
Excluded region(s): 6.00 - 8.38	Preferred orientation correction: March-Dollase (Dollase, 1986) along [100], G1 = 0.7759(6)
Profile function: split-type pseudo-Voigt

Special details

Experimental. specimen was rotated in its plane

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

	x	y	z	Uiso*/Ueq
Cl1	0.2501 (4)	0.1475 (2)	0.2491 (3)	0.074 (2)
N1	0.683 (2)	0.6999 (8)	0.884 (2)	0.086 (6)*
N2	0.220 (2)	0.5154 (8)	0.422 (2)	0.085 (6)*
N3	0.161 (2)	0.4527 (7)	0.427 (2)	0.077 (6)*
C1	0.590 (2)	0.6516 (9)	0.775 (2)	0.078 (6)*
C2	0.525 (2)	0.5840 (9)	0.820 (2)	0.046 (6)*
C3	0.415 (2)	0.5393 (8)	0.710 (2)	0.081 (7)*
C4	0.340 (2)	0.5535 (8)	0.540 (2)	0.029 (6)*
C5	0.389 (2)	0.6239 (5)	0.493 (2)	0.038 (6)*
C6	0.508 (2)	0.6687 (9)	0.603 (2)	0.103 (8)*
C7	0.023 (2)	0.4153 (9)	0.294 (2)	0.089 (9)*
C8	−0.052 (3)	0.449 (1)	0.136 (2)	0.16 (1)*
C9	−0.160 (3)	0.4044 (9)	0.011 (2)	0.108 (8)*
C10	−0.191 (2)	0.3317 (8)	0.030 (2)	0.042 (6)*
C11	−0.129 (2)	0.3024 (7)	0.184 (2)	0.082 (7)*
C12	−0.012 (2)	0.3424 (9)	0.316 (1)	0.063 (6)*
H2	0.568	0.570	0.946	0.076*
H3	0.379	0.488	0.750	0.076*
H5	0.328	0.640	0.368	0.076*
H6	0.545	0.720	0.563	0.076*
H8	−0.026	0.505	0.118	0.076*
H9	−0.224	0.427	−0.109	0.076*
H10	−0.262	0.298	−0.075	0.076*
H11	−0.173	0.248	0.201	0.076*
H12	0.052	0.318	0.434	0.076*
H13	0.217	0.426	0.536	0.076*
H14	0.729	0.748	0.852	0.076*
H15	0.741	0.688	1.006	0.076*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
Cl1	0.115 (3)	0.038 (3)	0.040 (3)	0.033 (9)	−0.012 (5)	0.016 (9)

Geometric parameters (Å, º)

N1—C1	1.32 (2)	C5—C6	1.33 (2)
N1—H14	1.03	C5—H5	1.08
N1—H15	1.03	C6—H6	1.08
N2—N3	1.25 (2)	C7—C8	1.45 (2)
N2—C4	1.31 (2)	C7—C12	1.40 (2)
N3—C7	1.42 (2)	C8—C9	1.38 (2)
N3—H13	1.03	C8—H8	1.08
C1—C2	1.45 (2)	C9—C10	1.38 (2)
C1—C6	1.45 (2)	C9—H9	1.08
C2—C3	1.31 (2)	C10—C11	1.39 (2)
C2—H2	1.08	C10—H10	1.08
C3—C4	1.43 (2)	C11—C12	1.39 (2)
C3—H3	1.08	C11—H11	1.08
C4—C5	1.45 (2)	C12—H12	1.08
H14—N1—H15	116	C1—C6—C5	123 (1)
C1—N1—H15	121	C5—C6—H6	119
C1—N1—H14	122	C1—C6—H6	119
N3—N2—C4	128 (1)	N3—C7—C12	118 (1)
N2—N3—H13	118	N3—C7—C8	120 (1)
N2—N3—C7	125 (1)	C8—C7—C12	122 (1)
C7—N3—H13	118	C7—C8—H8	123
N1—C1—C6	122 (1)	C7—C8—C9	115 (2)
N1—C1—C2	121 (2)	C9—C8—H8	123
C2—C1—C6	115 (1)	C8—C9—H9	118
C1—C2—H2	120	C8—C9—C10	124 (2)
C1—C2—C3	121 (2)	C10—C9—H9	118
C3—C2—H2	120	C9—C10—H10	120
C2—C3—H3	118	C9—C10—C11	119 (1)
C2—C3—C4	125 (1)	C11—C10—H10	120
C4—C3—H3	118	C10—C11—H11	120
N2—C4—C3	130 (1)	C10—C11—C12	120 (1)
C3—C4—C5	115 (1)	C12—C11—H11	120
N2—C4—C5	115 (1)	C7—C12—C11	119 (1)
C4—C5—H5	120	C11—C12—H12	121
C4—C5—C6	121 (1)	C7—C12—H12	121
C6—C5—H5	120

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
N1—H14···Cl1^i	1.03	2.17	3.18 (2)	166
N1—H15···Cl1^ii	1.03	2.29	3.27 (2)	158
N3—H13···Cl1^iii	1.03	2.27	3.27 (2)	163

Symmetry codes: (i) −x+1, −y+1, −z+1; (ii) −x+1, y+1/2, −z+3/2; (iii) x, −y+1/2, z+1/2.