High temperature deformation and microstructural instability in AZ31 magnesium alloy

https://doi.org/10.1016/j.msea.2013.01.060Get rights and content

Abstract

The high temperature deformation behaviour of AZ31 magnesium alloy is analysed by comparing numerous investigations, including work by these authors and by other researchers. Three main deformation mechanisms are observed, i.e grain boundary sliding, solute drag creep and climb-controlled dislocation creep. A combined set of constitutive equations, which takes into account the concurring effect of these different deformation mechanisms, is proposed. Grain boundary sliding is observed to cause a superplastic behaviour in fine-grained materials, but grain growth due to excessively prolonged high temperature exposure invariably results in a transition to either viscous glide or dislocation climb as a rate-controlling mechanism. On the basis of these considerations, the differences observed by testing the same material under constant strain rate or by strain rate change experiments are rationalised by quantifying the effect of static and dynamic grain growth and dynamic recrystallisation. This procedure provides a unitary description of the high temperature deformation of AZ31 in a wide range of strain rates and temperatures.

Introduction

Magnesium alloys fabricated by mechanical forming, are of considerable interest because they provide higher ductility and specific strength than castings. However, it is necessary to develop processes with competitive productivity and performance to ensure a profitable use of these alloys. High temperature forming (300–450 °C) is especially important both as a primary process (rolling, extrusion) to produce long products for secondary manufacturing and as a direct secondary process such as forging. High temperature plastic deformation of materials and creep both occur at temperatures above approximately 0.5 T/Tm where Tm is the absolute melting point. Constitutive equations, which have been developed for predicting the behaviour of materials, are essential for design and control. Generally, the creep rate or strain rate, ε̇, has been observed to be related to the absolute temperature, T, and the stress, σ, by the following equation:ε̇=k1exp(Q/RT)(σ/E)nwhere k1 is a function of structure, E is Young's modulus, n is the stress exponent, Q is the activation energy for plastic flow and R is the universal gas constant. Although creep is usually correctly described by a diffusion-controlled dislocation creep equation, other mechanisms may be important. Grain boundary sliding (GBS) and diffusional flow may in fact dominate deformation especially in fine-grained materials. Each of these mechanisms is considered to operate independently of the others and has a particular dependence on creep rate with stress, temperature and grain size. Over a certain temperature and strain rate range, when two mechanisms operate in parallel, their contribution to the total strain will be additive, and if one is much faster than the other, it will dominate the deformation process. By contrast, when the mechanisms operate in sequence, the slowest will become rate-controlling. However, great attention should also be paid to the influence of the microstructure on the creep mechanism because the microstructure of the AZ31 alloy is prone to evolve dynamically at high temperatures. As is well-known, grain size is of central importance for the grain boundary sliding mechanism. Del Valle et al. [1], [2], [3], for example, analysed the behaviour of the AZ31 alloy with a grain size of 17 μm, observing that the low thermal stability of this material leads to noticeable grain growth during tensile tests, and a subsequent increase in flow stress, which affects the GBS mechanism. In such circumstances, the analysis of the deformation mechanism using measurements of the maximum stress could lead, erroneously, to the conclusion that a creep mechanism occurs involving an n=3 exponent, interpreted as evidence of viscous glide-controlled deformation. Therefore, the experimental technique used is of great importance when trying to study the deformation mechanisms in this alloy. While strain rate change (SRC) experiments clearly highlight the existence of a well-defined regime of GBS-controlled deformation where n=2 [1], [2], [3], in the case of constant strain rate (CSR) or constant stress (CS, i.e. creep) tests the stress exponent is sometimes higher than 3 [4]. In addition, the microstructure could evolve by dynamic recrystallisation (DRX) which is the mechanism that in most systems has the major role in controlling the evolution of the structure during high temperature straining. Therefore, DRX operates as a dynamic grain refiner which opposes the effect of dynamic grain growth. A necessary condition for DRX is the formation of a dislocation substructure; therefore DRX is mainly operative at high strain rates where the deformation mechanism involved is dislocation creep.

The aim of the present paper is, on one hand, to quantify the effect of grain growth and DRX on the high temperature mechanical properties. On the other hand, an extensive recompilation of literature data regarding creep in the AZ31 alloy has allowed a set of creep laws to be introduced, which, in addition to the microstructure evolution equations, are suitable for describing the entire range of the strain rate–stress–temperature map in AZ31 alloy, providing a unitary model.

Section snippets

Analysis of mechanical data at high temperatures

In this section we review creep power laws in order to characterise the different deformation mechanisms that occur in the AZ31 alloy, i.e. the constitutive laws (Eqs. (2), (3), (4), (5), (6)) for creep corresponding to the three modes of deformation found for this material.

The GBS-controlled deformation is usually described by one of the following equations, i.e.ε̇=k2(DL/d2)(σ/E)2andε̇=k3(Dgb/d3)(σ/E)2with b=3.21×10−10 m (the expressions for bulk and grain boundary diffusivities, DL and Dgb

Grain growth in static annealing (SGG)

High temperature processing, such as rolling or extrusion, usually brings about the partial or complete dynamic recrystallisation of the alloy. Subsequent annealing leads to static recrystallisation (SRX) of the deformed structure, and/or to static grain growth (SGG). Yang et al. studied this process by annealing, at 493, 503 and 513 K, different samples previously deformed in compression up to ε=0.3 or 1.2 [50]. Grain growth was observed in all the investigated conditions, irrespective of the

Microstructural instability during CSR and SRC experiments

The constitutive model described above contains two parameters which are specific expressions of the microstructure, namely the amount of Al in solid solution (through the α parameter) and the grain size. While the amount of Al in solid solution in AZ31 is substantially constant at above 473 K, the grain size can exhibit important variations due to the combined effect of DRX and DGG. Fig. 13 shows the variation in the grain size as measured by del Valle et al. [3] during the continuous test

Conclusions

The high temperature deformation of an AZ31 magnesium alloy was analysed by comparing the results of numerous investigations. Three main deformation mechanisms were observed:

  • 1.

    grain boundary sliding, which causes a superplastic behaviour in fine-grained materials; grain growth due to prolonged high temperature exposure leads to a reduction in the creep rate and eventually to the transition to either viscous glide or dislocation climb as the rate-controlling mechanism;

  • 2.

    deformation controlled by

Acknowledgements

Financial support from CICYT, Spain, Project MAT2012-38962 is gratefully acknowledged.

References (67)

  • J.A. del Valle et al.

    Scripta Mater.

    (2007)
  • J.A. del Valle et al.

    Mater. Sci. Eng. A

    (2007)
  • S. Spigarelli et al.

    Mater. Sci. Eng. A

    (2011)
  • H. Somekawa et al.

    Mater. Sci. Eng. A

    (2005)
  • H.K. Lin et al.

    Mater. Sci. Eng. A

    (2005)
  • S.H. Kang et al.

    J. Mater. Process. Technol.

    (2008)
  • S.B. Yi et al.

    Mater. Sci. Eng. A

    (2006)
  • D.L. Yin et al.

    Mater. Lett.

    (2005)
  • W.J. Kim et al.

    J. Alloys Compd.

    (2008)
  • S. Spigarelli et al.

    Scr. Mater.

    (2009)
  • J.A. del Valle et al.

    Mater. Lett.

    (2008)
  • H. Somekawa et al.

    Scr. Mater.

    (2003)
  • H. Somekawa et al.

    Mater. Sci. Eng. A

    (2003)
  • H. Watanabe et al.

    Mater. Sci. Eng. A

    (2008)
  • K. Kitazono et al.

    Scr. Mater.

    (2001)
  • H. Watanabe et al.

    Int. J. Plasticity

    (2001)
  • X. Wu et al.

    Scr. Mater.

    (2002)
  • H.Y. Wu et al.

    Mater. Sci. Eng. A

    (2011)
  • M.R.R. Panicker et al.

    Mater. Sci. Eng. A

    (2011)
  • H. Takuda et al.

    J. Mater. Process. Technnol.

    (2005)
  • J.A. del Valle et al.

    Mater. Sci. Eng. A

    (2008)
  • H. Somekawa et al.

    Mater. Sci. Eng. A

    (2005)
  • W.J. Kim et al.

    Acta Mater.

    (2001)
  • F.A. Mohamed

    Scr. Metall.

    (1978)
  • S.S. Vagarali et al.

    Acta Metall.

    (1981)
  • J.A. del Valle et al.

    Acta Mater.

    (2007)
  • S. Spigarelli et al.

    Mater. Sci. Eng. A

    (2010)
  • X. Yang et al.

    Mater. Sci. Eng. A

    (2012)
  • S. Abdessameud et al.

    Trans. Nonferrous Met. Soc. China

    (2010)
  • Q. Miao et al.

    J. Alloys Compd.

    (2010)
  • S. Alsagabi et al.

    Mater. Sci. Eng. A

    (2012)
  • H.-K. Kim et al.

    Mater. Sci. Eng. A

    (2004)
  • M. Wang et al.

    Comput. Mater. Sci.

    (2009)
  • Cited by (42)

    • Analysis of the creep behavior of fine-grained AZ31 magnesium alloy

      2020, Materials Science and Engineering: A
      Citation Excerpt :

      Nevertheless, a careful review shows that analyses of the superplastic behavior of this material have focused almost exclusively on the early stages of deformation and there have been no attempts to provide a comprehensive examination of the effect of concurrent grain growth. A recent report suggests that grain growth may play a key role in the high temperature deformation of the AZ31 alloy [16] but a more detailed analysis is necessary to fully document the creep properties. In addition, magnesium alloys processed by ECAP are generally tested using tensile testing whereas many of the early experimental results on conventional superplastic alloys were obtained using double-shear testing in which the application of a constant load is equivalent to a constant stress [17–19].

    • Superplasticity in a commercially extruded ZK30 magnesium alloy

      2018, Materials Science and Engineering: A
    • Threshold creep behaviour of an aged Mg–Zn–Ca alloy

      2017, Materials Science and Engineering: A
    View all citing articles on Scopus
    View full text