The effect of ageing temperature on the fracture toughness of an 8090 Al-Li alloy

Abstract

It is well established that during plastic deformation of AI-Li alloys, the coherent 6' precipitates are sheared by dislocations. The resultant planar slip associated with 6' shearing leads to heavy localized dislocation pile-ups at grain boundaries. Suresh et al. [1] suggested that a major contribution to the fracture process in peak-aged and overaged alloys arises from intergranular failure promoted by microvoids formed around grain boundary precipitates. This process in enhanced by slip planarity, and Duva et al. [2] proposed that slip becomes more localized as the size of the coherent ageing precipitate increases. The size and distribution of the 6' phase in the AI-Li alloy 8090 will depend on the ageing temperature selected, and in the present work the fracture toughness of this alloy was investigated in specimens aged to the same hardness level at progressively lower ageing temperatures. A 90 mm thick 8090 plate supplied by Messrs Alcan Plate Ltd had the composition shown in Table I. It had been solution-treated at 545 °C before being cold-water quenched and stretched 7% before ageing. Samples 10 mm × 15 mm x 15 mm were cut from the mid-section of the plate and ageing was carried out at 130 and 150 °C in an air furnace and at 170 and 190 °C in an oil bath. The average value of five hardness indentations was recorded, and the surface of each specimen was ground before testing in order to remove any lithium-depleted layer. The object was to examine the variation of toughness with ageing temperature in material of constant hardness. The hardness selected was that achieved in 32 h at 170 °C, which is commonly commercially employed. The ageing times to achieve this hardness were 320 h at 130 °C, 78 h at 150 °C, 32 h at 170 °C and 8.3 h at 190 °C. Both K~ and crack opening displacement (COD) tests were conducted, and toughness specimens were also aged for 8 and 24 h at 170 °C, so that the variation of toughness with ageing time at constant temperature could be examined. K,~ tests were conducted to American Society for Testing and Materials standards on compact tension (CT) specimens of 20 mm thickness, and COD tests on specimens of 6.35 mm thickness. All specimens were machined so that they were strained in the longitudinal direction of the original plate, with the notch in the transverse direction. The change of hardness with ageing time at the four temperatures is shown in Fig. la to d. As expected, peak hardness was achieved at increasingly longer ageing times for increasingly lower ageing temperatures. There was also the expected tendency for the peak hardness to be greater the lower the ageing temperature was. ~ though the duration of ageing at 130 °C was insufficient to achieve a peak value, it is noteworthy that plateaux are present in the curve, which may reflect the formation of precursor phases to the 6' precipitate. These data appear to be consistent with those reported by Firrao et aL [3] for a ternary A1-Li-Cu, who similarly reported a maximum hardness achieved after ageing at 150 °C in comparison with the other ageing temperatures studied. It is obvious, however, that ageing to peak hardness at 150 °C would not be commercially viable in view of the protracted time involved. However, ageing at a temperature between 150 and 170 °C may produce a superior combination of strength and toughness within an economically viable ageing time. The values of KIo or K o obtained are shown in Table II. The COD values obtained on specimens after the different heat treatments are shown in Fig. 2, and it is apparent that the specimens aged at lower temperatures and for shorter times have higher toughnesses. From both sets of data it is apparent that the greatest toughness change occurs between ageing temperatures of 150 and 170 °C, which again suggests that an ageing treatment between these two temperatures would produce optimum properties. The materials showing higher toughnesses correspond to ageing treatments that lead to smaller 6' precipitate sizes. This in turn will result in a less planar slip distribution [2], and hence greater plasticity. It may thus be concluded that, first, greater peak hardnesses are achieved at lower ageing temperatures. Secondly, the form of the ageing curve at 130 °C suggests that other metastable transition phases precede the formation of the 6' precipitate. Thirdly, as the ageing time is increased (at a given temperature) the toughness decreases. Fourthly, if specimens are aged to the same hardness at different ageing temperatures, then the toughness (as measured by both Kic and COD) will be higher at