In this paper an alternative anodising procedure, that confers a good base for painting and adhesive bonding of aluminium, is envisaged to replace the chromic acid anodising The substrate consisting of Al 2024-T3 is immersed in a cerium solution in order to create a cerium-containing film that covers the copper rich precipitates, followed by anodising in a boric acid-sodium borate/sulphuric acid solution. The results showed that this procedure confers good corrosion protection to the substrate and can potentially constitute an alternative to the Cr-VI based treatment. Introduction Painting and adhesive bonding of aluminium are essential technologies in many industrial sectors (e.g. aerospace industry) and each process depends on the generation of a strong and stable interface between an organic layer and the surface consisting of naturally formed metal oxide. Pre-treatments on the substrates, which leave the metal with an oxide structure that is stable and compatible with the organic layer, are vital to achieve the desirable levels of strength and reliable long-term performance and have been used for decades. However, they involve a range of environmentally objectionable chemicals, such as solvents and chromates, thus needing to be replaced. Several attempts have been made to replace the Cr (VI) – based treatments. In the present work the traditional chromic acid anodising was replaced by a boric acid-sodium borate/sulphuric acid bath. The alloy was pre-treated in a cerium nitrate solution, in order to form a film that covers the copper rich precipitates of the alloy, which impair good anodising. Experimental Aluminium alloy 2024-T3 coupons were used. The specimens were degreased with tricloroethylene (4 min), followed by etching in 3 g.l NaOH solution (8 min) and desmutting in 50% v/v HNO3 solution (30 sec). The anodising bath (ABS) consisted of a mixture of H2SO4 (15%) with a solution containing 0.5M H3BO3 and 0.05 M Na2B4O7.10H2O, in the proportion 70/30 (v/v). The optimised temperature and current density were respectively 40oC and 2 A.dm and the anodising time was 30 minutes. The choice of 40oC came from the fact that lower temperatures lead to higher coating thickness, which should be deleterious for fatigue resistance. The anodising was performed on as-etched surfaces and on surfaces also pre-treated by immersion in 0.01M Ce(NO3)3 solutions for 4 hours. After anodising the specimens were sealed in boiling distilled water for 30 minutes. However, before sealing the specimens were observed on a field emission scanning electron microscope (SEM), on the surface and in a section almost perpendicular to the surface. This latter observation also allowed the determination of the coating thickness. Since the fatigue is an important property of these materials, fatigue tests were carried out on anodised specimens. For comparison, fatigue resistance was also obtained on as received and chromic acid anodised specimens. The alloy was submitted to tensile-tensile fatigue tests in air and at the environmental temperature, in order to obtain S-N plots. The tests were carried out according to ASTM E 466-82 [1], using an INSTRON 8502 servo-hydraulic machine. The corrosion behaviour was studied in 3% NaCl solution by Electrochemical Impedance Spectroscopy, using a Solartron 1250 frequency response analyser connected to the cell via a Solartron 1286 electrochemical interface. Results and Discussion Figure 1a shows a SEM photograph of the transverse section of the specimen anodised in the ABS. It allows the determination of the coating thickness that is approximately 2.5 μm. Higher magnification micrograph (160 000 ×, Fig. 1b) shows in more detail the structure of this coating. Linear continuous pores perpendicular to the specimen surface, as those reported in the literature for anodised aluminium, cannot be seen. The coatings have a grain like structure with the grains separated by pores. The grains seem to have grown preferentially in one direction, revealing a lamellar shape. Coatings formed after pre-treatment in cerium solutions were also observed by SEM. The structures observed were similar to those previously reported, apart a finer structure with smaller grains. The presence of a barrier layer in contact with the metallic surface could not be detected in both cases, which indicates that it should be very thin. a) b) Figure 1. Micrographs of transverse sections of anodised Al 2024-T3 in ABS: a) 20 000×.; b)160 000 × The impedance results for 1 day and 7 days immersion in 3% NaCl solution for ABS anodising are shown in Fig. 2a. The impedance values do not change significantly with time, which reveals a good corrosion performance of the specimen. As a plateau is not usually observed at low frequencies for anodised samples that show even pitting or pronounced corrosion, practical criteria have been established to assess the corrosion behaviour of this material [2-4]. Most of them are based on the value of the impedance in the low frequency region. Thus, according to Mansfeld and Kendig [2], a damage function, D, which is defined as