Experimental investigation of tension and compression creep-ageing behaviour of AA2050 with different initial tempers

Abstract

Creep-ageing behaviour of aluminium alloy 2050 with different initial tempers (T34, T84 and as-quenched) has been experimentally investigated under both tension and compression creep-ageing conditions, with different stress levels at 155 °C for 18 h. Corresponding strengthening phenomena have been studied by interrupted creep-ageing tests and subsequent tensile tests. Moreover, the microstructures of some selected specimens after creep-ageing tests have been examined by transmission electron microscopy (TEM) and the precipitation process has been analysed. It has been found that creep strains under tensile stresses are much larger than those under compressive stresses during the tests. A new “double primary creep feature” has been observed in both the as-quenched alloys and the pre-stretched/natural-aged (T34) alloys, in which an intermediate inverse creep stage with an increasing creep strain rate locates between the initial primary+transient steady-state creep stages and the second primary+second steady-state creep stages. While for the alloy with peak-aged initial temper (T84), typical primary and steady-state secondary creep stages are observed during tension creep-ageing tests and little creep strain occurs under compressive stresses of 150 and 175 MPa. The mechanisms for these phenomena are discussed in terms of microstructural interactions among the changing dislocations, solute-matrix bonding and precipitates, and their effects on the creep resistance of the alloy during creep-ageing tests are analysed.

Introduction

Creep age forming (CAF) is a process to deform aluminium alloys by keeping an external stress onto the material at an artificial ageing temperature for a specific time, thus simultaneously achieving high mechanical properties by age hardening and plastically forming the part within its elastic limit by creep deformation [1], [2]. It is an efficient way to manufacture extra-large light weight and high performance panel products in aerospace industry. Aluminium alloy 2050 (AA2050) is a third generation aluminium–copper–lithium (Al–Cu–Li) alloy which can achieve excellent strength-ductility performances by artificial ageing, leading to a great potential in aerospace applications [3], [4]. In order to manufacture AA2050 components with ideal mechanical properties through a CAF process, it is important to investigate the creep-ageing behaviour of the alloy. In a CAF process, a thick aluminium plate workpiece is mainly subjected to bending, in which one side of the material is in tension and the other side is in compression [1]. The study of the tension and compression creep-ageing behaviour of the alloy therefore becomes more important for CAF applications.

Although AA2050 is a recently-developed material [4] and has not been studied as thoroughly as some other aluminium alloys, its ageing behaviour can be referenced by the widely published studies of other Al–Cu–Li alloys with similar compositions [5], [6]. Generally, its main strengthening contribution comes from precipitation hardening [7]. The precipitates during the artificial ageing process can be summarised as GP zones, θ′ (Al₂Cu), δ′ (Al₃Li), S′ (Al₂CuMg) and T₁ (Al₂CuLi), among which T₁ is mostly regarded as a main strengthening phase [8], [9]. Kumar et al. [8] have investigated the ageing process of Al–Cu–Li alloys with initial tempers of T3 and T4 and concluded that both precipitates from natural ageing and dislocations from pre-stretch in the initial temper would influence the subsequent ageing process. In addition, it has been reported that dislocations induced by the pre-stretch facilitate the nucleation of T₁ precipitates, thus accelerating ageing progress and promoting strength [10], [11]. By contrast, little research has been published on the creep behaviour of Al–Cu–Li alloys. Kazanjian et al. [12] have compared the creep curve of a Al–3.5Cu–1.0Li alloy with some other Al–Cu–Mg alloys and showed that all these alloys exhibit the typical two-stage (primary stage and steady secondary stage) creep behaviour and the Al–Cu–Li alloy presents a higher creep resistance than others. Nevertheless, the study has only considered the alloy with T8 initial temper and the creep behaviour of the alloy with other initial tempers has not been studied. As other initial tempers can be more important for the CAF process, such as the T34 initial temper of AA2050 which is solution heat-treated, pre-stretched and then naturally-aged, the effects of different initial tempers on the creep-ageing behaviour which concerns both ageing and creep processes, therefore, necessitate further investigations.

In addition, both tensile and compressive stresses will appear in the material during the CAF process, but most of the studies now consider only the creep-ageing behaviour in tensile stress conditions [13], [14]. The stress directions have been demonstrated to affect the orientation of precipitates during the ageing process of some aluminium alloys [15], [16]. Eto et al. [15] and Zhu et al. [17], [18] have found that the main precipitates in Al–Cu alloys will adjust to be parallel with the direction of a tensile stress, while with a compressive stress, the preferential orientation tends to be perpendicular. However, creep tests of the material during these conditions were not included in their studies. Heimerl et al. [19] have reported a much larger creep strain in tension than in compression of the 2024-T3 (Al–Cu–Mg) alloy at elevated temperatures from 150 °C to 200 °C. The probable reasons have been proposed as both the pre-stretch and ageing effects. Moreover, Zhang et al. [20] and Przydatek [21] have observed the asymmetric tensile and compressive creep behaviour in Al–Si–Cu and Al–Cu–Mg alloys respectively and they both attributed the reason to the easier cavity nucleation on grain boundaries in tension creep conditions. As the precipitates in Al–Cu–Li alloys, such as θ′ and T₁, have negative precipitate-matrix misfits which are believed to be the reason for the stress-orientation phenomena [15], [16], the tension and compression creep-ageing behaviour of such alloy may also be different. These phenomena require further investigations, so as to achieve the potential CAF applications of AA2050.

This study, for the first time, investigates and compares the creep-ageing behaviour of AA2050 with different initial tempers under both tensile and compressive stresses for their CAF applications. To investigate the interactions between the ageing and creep processes during creep-ageing tests, three initial tempers with different pre-stretch and initial ageing (natural-aged and peak-aged) conditions were studied. The results were analysed in detail by combining together the creep-ageing behaviour and the corresponding microstructural evolution observed in the experiments.