Selective Reduction of an Aromatic Nitro Group
421
2 [6], the cinnoline derivative 3 [7], and the desired azoxy compound 4
(Scheme 1). The reaction proved to be somewhat temperamental, giving vary-
ing yields of these compounds. Such poor reproducibility is not unusual for
heterogeneous reactions. The best result obtained with respect to the azoxy
species 4 was 12% of 2, 28% of 3 and 56% of 4. Tedious work-up involving
distillation and chromatography was necessary for separating this product mix-
ture. In contrast to this, reduction of 1 according to the protocol by Ren et al.
[7] (sodium borohydride in the presence of substoichiometric amounts of bis-
muth powder in methanol) gave 4 in 70% yield after simple recrystallisation
(Scheme 1).
The identity of the product was confirmed by a single-crystal X-ray structure
analysis. 4 adopts the (Z)-configuration (Fig. 1).
Two individual molecules are present whose bond parameters are almost iden-
tical. The asymmetric unit contains 1.5 molecules. Structural parameters are un-
exceptional and compare well to those obtained for related species such as
azoxybenzene [8] and 4,40-azoxydianisole [9]. The N–N distance is 126.9(2)
and 128.4(3) pm corresponding to a double bond. The azoxy N–O distance has
a value of 126.05(18) and 126.4(3) pm, which is slightly, but significantly, longer
than the N–O distances of the nitro groups which range from 121.99(19) to
122.48(19) pm.
Experimental
1
NMR: Bruker Avance DRX 500 (500.13 MHz, H, ext. TMS 125.04MHz, 13C, ext. TMS). MS: VG
Autospec (70 eV). Elemental analyses agreed favourably with the calculated values.
1-Amino-2-(2-amino-2-methylpropyl)-benzene (2), 3,3-Dimethyl-3,4-dihydrocinnoline
(3, C10H16N2) and 2,20-Di-(2-nitro-2-methylpropyl)-azoxybenzene (4, C20H24N4O5)
by Reduction of 1-Nitro-2-(2-nitro-2-methylpropyl)-benzene (1, C10H12N2O4)
with Zinc in Hydrochloric Acid
A mixture of 40cm3 6 N hydrochloric acid, 2.0 g 1 [5] (8.92 mmol) and 8.0 g Zn powder (122 mmol)
was heated to reflux for 1 h. The mixture was allowed to cool to room temperature. Undissolved Zn
was removed by decanting and washed with 20cm3 of water. After alkalisation by addition to an
excess of 2 N NaOH solution the mixture was extracted with 3Â 40 cm3 of diethyl ether. The combined
extracts were dried with Na2CO3. The solvent was removed in vacuo. 1-Amino-2-(2-amino-2-methyl-
propyl)-benzene (2) and 3,3-dimethyl-3,4-dihydrocinnoline (3) were distilled off at 0.1 Pa at a bath
temperature of 60ꢁC. The distillate was separated chromatographically using a silica gel column:
417 mg of 2 (28%) was eluted with CH2Cl2 and 165mg of 3 (12%) with ethyl acetate. Recrystallisation
of the distillation residue from ethanol afforded 2.01g of 4 (56%). In other runs the yield of 4 varied
and was as low as 10%.
2: The compounds was characterised by comparison with an authentic sample prepared by the
established route [6].
3: 1H NMR (CDCl3): ꢀ ¼ 1.33 (s, 6H, 2 Me), 2.51 (s, 2H, CH2), 7.04 (d, 1H, J ¼ 7.5 Hz, CH), 7.35
(t, 1H, J ¼ 7.5 Hz, CH), 7.42 (t, 1H, J ¼ 7.5 Hz, CH), 7.94 (d, 1H, J ¼ 7.5 Hz, CH) ppm; 13C NMR
(CDCl3): ꢀ ¼ 24.7 (Me), 31.9 (CH2), 60.4 (CMe2), 121.8 (ring C), 127.9 (ring C), 128.0 (ring C), 128.8
(ring C), 131.4 (ring C), 141.2 (ring C) ppm; EI-MS: m=z (%) ¼ 160 [Mþ ] (11), 132 [Mþ –N2] (40),
117 [Mþ –N2–Me].