ABSTRACT
Abstract The reductive alkylation reaction occurring during the hydrogenation of nitrobenzene in a 1-hexanol solvent is not between the aniline formed and the 1hexanol but is between 1-hexanol and a surface species retained by the catalyst. This surface species is not formed during the aniline 1-hexanol reaction. There is also inhibition of the reaction between aniline and 1-hexanol. Introduction The reduction of nitrobenzene to aniline is a major industrial process at the heart of the production of polyurethanes, and it is also often used as a marker reaction to compare activities of catalysts [1,2]. It can be performed over a variety of catalysts and in a variety of solvents. As well as its main use in polyurethanes, aniline is used in a wide range of industries such as dyes, agrochemicals, by further reaction and functionalisation. Reductive alkylation is one such way of functionalising aromatic amines [3, 4]. The reaction usually takes place between an amine and a ketone, aldehyde or alcohol. However it is possible to reductively alkylate direct from the nitro precursor to the amine and in this way remove a processing step. In this study we examined the reductive alkylation of nitrobenzene and aniline by 1hexanol. Experimental Section The catalyst used throughout this study was prepared by impregnation. To a slurry of silica (M5 Cab-O-Sil, Surface Area 200 m2g-1) in water sufficient copper nitrate solution was added to give a loading of 8.6 % w/w Cu. The resulting suspension was dried at 353 K until a free flowing powder was obtained. Copper metal surface area was determined by nitrous oxide decomposition. A sample of catalyst (0.2 g) was reduced by heating to 563 K under a flow of 10 % H2/N2 (50 cm3min-1) at a heating rate of 3 deg.min-1. The catalyst was then held at this temperature for 1 h before the gas flow was switched to helium. After 0.5 h the catalyst was cooled in to 333 K and a flow of 5 %N2O/He (50 cm3min-1) passed over the sample for 0.25 h to surface oxidise the copper. At the end of this period the flow was switched to 10 % H2/N2 (50 cm3min-1) and the sample heated at a heating rate of 3 deg.min-1. The hydrogen up-take was quantified, from this a
surface area could be determined [5, 6]. The metal surface area calculated was 5.47 m2g-1, giving a dispersion of 9.5% and a metal particle size of 10.7 nm. The reductive alkylation reactions were performed in a 1 atm stirred tank reactor with hydrogen sparged (200 cm3min-1) through the solution. The catalyst (0.5 g) was reduced in situ, by heating under a flow of 10 % H2/N2 at 3 Kmin-1 from ambient to 563 K and holding at this temperature for 1 h. During this procedure the catalyst was supported on a glass sinter and the gas flow was through the catalyst bed. After 1 h in 10 % H2/N2 the gas flow was switched to helium and the temperature held for another 0.5 h. The catalyst was then cooled to room temperature and the reactor inverted to transfer the catalyst into the stirred tank reactor under a flow of helium. The reaction mix of 80 cm3 of 0.1 M nitrobenzene (or aniline) in 1-hexanol was added and the system heated to the reaction temperature (383 K). The reaction was initiated by switching the gas flow from helium to hydrogen. The reaction solution was stirred using a magnetic stirrer and the absence of diffusion control was determined by a variety of means [7]. Samples (0.3 cm3) were removed from the reactor at regular intervals and were analysed by gas chromatography. Table 1 Nitrobenzene hydrogenation in 1-hexanol at 383 K.
