A study of the composition dependence of the rapid hardening phenomenon in Al–Cu–Mg alloys using diffusion couples

https://doi.org/10.1016/j.msea.2012.03.043Get rights and content

Abstract

A diffusion couple approach is demonstrated for the combinatorial study of the compositional dependence of the rapid hardening phenomena in Al–Cu–Mg alloys. A series of Al–Cu/Al–Cu–Mg and Al–Mg/Al–Cu–Mg diffusion couples have been successfully fabricated to contain a gradient in Mg or Cu concentration, respectively. The effects of both total solute content and Cu:Mg ratio of the rapid hardening phenomena are considered. The rapid hardening response has been monitored by measuring the hardness profile across the diffusion gradients before and after artificial ageing. The composition profiles were quantitatively verified by electron probe microanalysis (EPMA) and the microstructure of the ternary end members has been characterised using atom probe tomography (APT). It is demonstrated that a critical Cu content exists, above which the rapid hardening phenomena diminishes. A change in the chemistry of the Cu–Mg clusters that form, and which are thought to determine the rapid hardening increment, also occurs with changes in the bulk alloy chemistry. The results suggest that Cu–Mg clusters rich in Mg have greater strengthening potency and this is consistent with recent suggestions by Marceau et al. The need for the development of combinatorial approaches to characterisation (microstructural and mechanical) for the full utilisation of such approaches in physical metallurgy is emphasised.

Highlights

► Diffusion couples are used for a combinatorial study of age hardening in Al–Cu–Mg alloys. ► A critical Cu content is identified above which the rapid hardening is diminished. ► Cu–Mg clusters rich in Mg appear to have the higher strengthening potency.

Introduction

An understanding of phase transformations in metal alloys requires an understanding of the effects of both temperature and composition. Whereas experimental studies of the effect of temperature are relatively straightforward, studies of the effect of composition are more time consuming and costly since they require the fabrication of different alloy compositions. This can be difficult if the transformations are particularly sensitive to small compositional variations or if small variations in composition separate different regimes of behaviour. Indeed a combinatorial approach to such a problem that allows rapid screening of a large number of alloy compositions, would be a welcome tool in the field of physical metallurgy. In solid-state chemistry, diffusion couples are used for this purpose in phase diagram determination, e.g. [1], but their use in physical metallurgy has so far been rather limited. They have been used to determine the boundaries separating different modes of ferrite growth in steels [2], [3], [4], [5], [6] and in studies of solute effects on recrystallisation [7]. These examples illustrate the advantages of combinatorial approaches. An area in physical metallurgy where diffusion couples could be used quite effectively is in studies of the competition of phase formation during precipitation, e.g. age-hardenable aluminium alloys. In this class of important engineering materials, subtle variations in chemistry can strongly influence the decomposition of the supersaturated solid solution and result in the development of different precipitate phases that can significantly modify the mechanical properties of the material.

A particularly interesting example is the case of Al–Cu–Mg alloys. Certain compositions that lie in the α (FCC Al) + S (Al2CuMg) phase field of the ternary phase diagram (Fig. 1) can exhibit a remarkable rapid hardness increase during artificial ageing at elevated temperatures [8], [9]. This rapid hardening phenomenon (RHP) occurs within the first 60 s of ageing and may provide more than 50% of the total hardness increment possible from ageing. For this reason, and because these alloys form the compositional basis of one of the two main classes of Al alloys used in aircraft construction, there has been much research interest in the fundamental origins of the RHP [10], [11], [12], [13]. There are two aspects to this problem. The first is an understanding of the microstructural features that form from the quenched solid solution that cause the remarkable hardening and the second is a quantitative understanding of their interactions with dislocations and how they provide the strengthening increment observed.

Atom probe tomography (APT) has contributed greatly to our understanding of the first of these questions. It now appears clear that the RHP is due to the formation of a very fine and uniform distribution of Cu and Mg containing clusters less than 1 nm in size and that Cu–Mg clusters rich in Mg seem to exhibit a high strengthening potency [12], [14]. The strength of the individual clusters has been estimated for one of the compositions investigated [13] but a quantitative rationalisation of the cluster strength in terms of dislocation/cluster interaction is lacking.

However, the RHP is not observed in all alloy compositions. An understanding of the range of alloy compositions that may exhibit the RHP and whether a critical Cu:Mg ratio and/or a dependence on the total solute content (and onset driving force for decomposition of the solid solution) is required, is currently not known. This is necessary for efforts both for understanding the fundamental origin of the RHP and for the development of new alloys that exploit the RHP. This is the context for and objective of the current study.

In this experimental contribution we address the question of the composition dependence of the RHP using a combinatorial approach. Diffusion couples containing gradients in Cu or Mg are fabricated and used to sample composition space for the RHP in alloys containing three different total solute contents. This approach allows us to rapidly screen a large number of alloy compositions, between the end-member compositions of the diffusion couples, for the RHP. The ternary end members of the diffusion couples have also been studied using APT to aid in the interpretation.

In Section 2, we outline in detail the approach used to generate the diffusion couples, the choice of heat treatment conditions, the measurements of the composition profiles within the diffusion couples and the use of Vickers hardness and APT as characterisation tools for the RHP. The results of subjecting the diffusion couples to standard solution heat treatment, water quenching and then artificial ageing, to test for the presence of the RHP, are described in Section 3 along with the APT results of selected ternary end members of the diffusion couples. In Section 4, we discuss the relative effects of Cu and Mg on the RHP in Al–Cu–Mg alloys as well as the implications of Cu–Mg cluster chemistry on the hardening response observed experimentally. Some perspectives on the usefulness of combinatorial approaches in physical metallurgy are presented in this section and concluding remarks are reserved for Section 5.

Section snippets

Alloy composition selection and preparation

Nine alloy compositions were selected for the fabrication of six different diffusion couples (discussed in Section 2.2). The nominal and actual (measured by ICP) alloy compositions prepared are listed in Table 1.The actual compositions are in good agreement with the nominally designed compositions in all cases except for the final alloy that contained ∼1.7Mg (wt.%) instead of the targeted ∼2.3Mg (wt.%). The alloys were prepared from 99.99% pure elements and melting occurred in a graphite

Hardness and composition measurements

Plots of the hardness variation across the diffusion couples in the solution treated and quenched state (AQ) and after ageing for 5 min at 200 °C are shown in Fig. 3, Fig. 4, Fig. 5 for the low, medium and high total solute containing couples, respectively. In each figure, a diffusion couple containing (a) a gradient in Mg at constant Cu concentration and (b) a gradient in Cu at constant Mg concentration, is shown. The range of composition space sampled by each couple is indicated on the phase

Discussion

In this contribution we have introduced the use of diffusion couples for studying the composition dependence of age hardening reactions in Al–Cu–Mg alloys. As shown by Fig. 6b, they are particularly useful for identifying critical compositions separating different regimes of behaviour, in this case a critical Cu content above which continued additions are detrimental to the rapid hardening increment. One could imagine such approaches being used for studies of competition in phase formation, the

Conclusions

A series of Al–Cu/Al–Cu–Mg and Al–Mg/Al–Cu–Mg diffusion couples have been successfully fabricated to contain a gradient in Mg and Cu concentration, respectively, whilst also covering a range of total solute contents but with approximately constant Cu:Mg atomic ratio. By combinatorial measurement of the hardness of the diffusion couples as a function of composition, in the AQ state and after ageing for 5 min at 200 °C, it is demonstrated that above a critical Cu content the rapid hardening

Acknowledgements

This work is supported by the Australian Research Council (ARC) through the ARC Centre of Excellence for Design in Light Metals. CRH gratefully acknowledges the support of the ARC through the award of a Future Fellowship and Prof. Alexis Deschamps (SIMAP, INPG) and Prof. Chad Sinclair (UBC) for stimulating discussions. The authors are also grateful for scientific and technical input and support from the Australian Microscopy & Microanalysis Research Facility (AMMRF) nodes at The University of

References (24)

  • A.A. Kodentsov et al.

    J. Alloys Compd.

    (2001)
  • A. Borgenstam et al.

    Acta Mater.

    (2000)
  • C.R. Hutchinson et al.

    Scripta Mater.

    (2004)
  • S.P. Ringer et al.

    Acta Mater.

    (1997)
  • R.K.W. Marceau et al.

    Acta Mater.

    (2010)
  • A. Deschamps et al.

    Acta Mater.

    (2011)
  • R.K.W. Marceau et al.

    Acta Mater.

    (2010)
  • M.P. Moody et al.

    Ultramicroscopy

    (2009)
  • P. Bas et al.

    Appl. Surf. Sci.

    (1995)
  • R.K.W. Marceau et al.

    Ultramicroscopy

    (2011)
  • A. Phillion et al.

    Metall. Mater. Trans. A

    (2004)
  • A. Borgenstam et al.
  • Cited by (37)

    • Making sustainable aluminum by recycling scrap: The science of “dirty” alloys

      2022, Progress in Materials Science
      Citation Excerpt :

      When joining aluminum alloys in the solid state, a difficulty is the presence of the oxide layer, which prevents good metallurgical bonding. However, interdiffusion can still be achieved by simple hot pressing [593]. Due to time limitations, large gradients cannot be obtained in this way, and extending the gradient is hard due to the brittleness of the joining interface.

    • Effects of axial cold-compression on microstructure uniformity and mechanical property enhancement of large 2219 Al–Cu alloy rings

      2020, Materials Science and Engineering: A
      Citation Excerpt :

      After solution treatment, the area fraction of Al2Cu particles also decreased, indicating that the content of Cu increased in Al matrix. This ulteriorly causes an increase in the precipitation rate during the aging heat treatment [12]. The effects of axial cold-compression deformation on the microstructure of large 2219 Al–Cu alloy rings, including grain-refining efficiency, uniformity of grain structures, precipitation behaviors, and mechanical property variations (especially in the radial direction) have not been studied extensively.

    View all citing articles on Scopus
    1

    Present address: Max-Planck Institut für Eisenforschung, Max-Planck-Str. 1, D-40237 Düsseldorf, Germany.

    View full text