Elsevier

Materials Science and Engineering: A

Volume 582, 10 October 2013, Pages 147-154
Materials Science and Engineering: A

The superposition of strengthening contributions in engineering alloys

https://doi.org/10.1016/j.msea.2013.06.032Get rights and content

Abstract

In engineering alloys strengthened by a combination of solid solution, precipitation and work hardening, plasticity is controlled by the interaction between dislocations and different sets of obstacles on the glide plane. How to add the strengthening contributions of these obstacles to predict yield stresses and work hardening is an important question. To answer this question, a 2D areal glide model based on the method of Nogaret and Rodney that simulates the glide of a single dislocation through an array of randomly distributed point obstacles was used. The effect of one, two and three sets of obstacles was studied with their strengths varying from weak (Φc=180°) to strong (Φc=0°). The results of these simulations were used for the first time to develop a mathematical expression for the exponent of the addition law τq=τ1q+τ2q. An extension of this addition law to three sets of obstacles was also derived. These new laws were then applied to predict the work hardening response of an Al–Mg–Sc alloy.

Introduction

In metallic alloys, the yield stress and work hardening are governed by the interaction between mobile dislocations and obstacles. Common obstacles include solute atoms, solute clusters, precipitates and forest dislocations. The appropriate way to add the strengths of multiple obstacle to determine the macroscopic flow stress is complex owing to the statistical nature by which a dislocation samples the distribution of obstacle in the glide plane [1].

The simplest approach to addition of flow stress components can be traced back to the case of grain size strengthening examined by Cottrell [2]. Cottrell assumed that the contributions from the intrinsic resistance to dislocation motion τ1 and the long range stress contribution from grain boundaries τ2 could be added in a linear mannerτ=τ1+τ2.

Subsequently, Koppenall and Kuhlmann-Wilsdorf [3] considered the superposition of flow stress components for the case of irradiated copper single crystals. They suggested that the strengthening contributions from point defects generated from irradiation damage and the pre-existing dislocations could be added geometrically (a formulation often referred as Pythagorean superposition)τ2=τ12+τ22.

On the other hand, Labusch [4] proposed a statistical theory for strengthening of solid solutions that suggested yet another way to superimpose two strengthening contributions, i.e.τ3/2=τ13/2+τ23/2.

Given the statistical nature of the problem, Foreman and Makin [5] were the first to use areal glide simulations to examine the motion of a dislocation across a glide plane populated with either one or two sets of obstacles. Based on their results, they found that the Pythagorean superposition, Eq. (2) gave good agreement with exceptions (e.g. when a few strong obstacles are added to many weak obstacles); a conclusion supported by the early review from Brown and Ham [1].

The seminal review of Kocks, Argon and Asbhy [6] re-examined the question of flow stress superposition laws. General qualitative guidance was provided to the reader according to what they refer to as “increasing structural scale”. They examined a number of scenarios , for example: (i) linear superposition of lattice resistance and the resistance due to strong obstacles such as forest dislocations or non-shearable precipitates, (ii) linear superposition of long range barriers (e.g. grain boundaries) and “small scale glide resistance” (lattice resistance, discrete obstacles such as small precipitates or forest dislocations) and (iii) the case of two sets of discrete obstacles of identical strength, but with different densities where the density of obstacles is simply summed, i.e. the Pythagorean addition law applies. Clearly, while these general guidelines are valuable, they leave many scenarios where ambiguity exists in which value of the superposition exponent to use, i.e. 1, 3/2 or 2.

In his review of precipitation hardening, Ardell [7], following Niete et al. [8], suggested that the superposition law could be generalized asτq=τ1q+τ2qwhere the exponent q is a parameter varying between 1 and 2. Ardell discussed specific examples were q could have values other than 1, 3/2 or 2. The more recent work of Bütter and Nembach [9], Neite et al. [8] and Zhu et al. [10] considered q as a parameter fit to simulation results without any explicit guidance on how its value should physically related to obstacle strength or density. The analysis of experimental data consequently, has employed empirically fit values for q [11], [12].

Recently, there has been a renewed interest in this question. Dong et al. [13] conducted areal glide simulations to examine in detail the situation for combinations of extremely weak obstacle (breaking angles larger than 176°). In that case, it was found that the Pythagorean addition was appropriate. However, they also did note that when the density of the weakest set of obstacles is less than 0.06%, this result breaks down. In these conditions, q was found to be a function of the density. It is worth noting, however, that this is a rather restrictive regime which is not of practical significance for engineering alloys.

As noted above, given only approximate theoretical recommendations, experimental studies have proposed empirical fits to select the best value for q [11], [12], [14], [15]. Alternatively, single values of q (q=1 or q=2) have been chosen to fit a wide range of data [16], [17], [18]. The goal of this paper is to use results from areal glide simulations to derive a mathematical expression which could be used to determine the addition law parameter, q, for a wide range of obstacle strengths and densities. Due to the nature of the areal glide model, this expression would be valid for alloys featuring obstacles of which dimensions are small compared to their average spacing and thus could be reasonably modeled as point obstacles. Further, the question of how to sum flow stresses when more than two sets of obstacles is considered.

To simulate the contribution from different sets of obstacles to the critical resolved shear stress (CRSS), a 2D areal glide model based on the method proposed by Nogaret and Rodney has been implemented [19]. The model is first applied to simulate a single set of randomly distributed obstacles and compared with the results from the classic “circle rolling” method first used by Foreman and Makin [20]. Next, the case of two sets of randomly distributed obstacles is examined and an expression for q proposed. Then, the case of three sets of obstacles is considered and one approach to implement the addition law is proposed. Finally, the developed formulae are applied to experimental data for aged Al–Mg–Sc specimens [12] where three sets of obstacles contribute to the strength.

Section snippets

Computational model

The glide of a single dislocation through an array of randomly distributed obstacles is simulated with a two-dimensional areal glide model similar to the one developed by Nogaret and Rodney [19]. This model differs from the “circle rolling” method previously used to simulate the glide of a single dislocation through an array of point obstacles. It assumes a constant dislocation line tension with T=μb2 and models obstacles as points characterized by their breaking angle Φc. Breaking angles vary

Simulations of dislocation glide through a single set of obstacles

For a single set of obstacles having a breaking angle Φc distributed in the glide plane, Foreman and Makin [5] proposed an empirical law for the normalized CRSSτ=0.956cos(Φc2)3/2(118cos(Φc2)2).

Owing to the difference between the “circle rolling” method used by Foreman and Makin and the algorithm used here, it was decided to check the applicability of Eq. (11). Fig. 3 shows that the present simulations match very well with Eq. (11) for weak obstacles (Φc>60°) but that the results diverge for

Addition law for two sets of obstacles

In many practical situations, moving dislocations interact with more than one type of obstacles. In the following, a normalized version of the commonly used addition law (Eq. 4) is used. If N1 and N2 are respectively the number of obstacles of types 1 and 2 (N=N1+N2), from Eqs. (6), (7), the normalized critical resolved shear stresses are thenτ1=τ1b2TN1andτ2=τ2b2TN2.Using the densities of obstacles, X1=N1/N and X2=N2/N, the normalized addition law isτq=(τ1X1)q+(τ2X2)q.

Simulations were

Addition law for three sets of obstacles

There are a number of practical scenarios where one must consider three sets of obstacles. For example, in the next section the case of an Al–Mg–Sc alloy will be considered where solute, precipitates and forest dislocations all act as obstacles. Extending the addition law from two to three (or more) sets of obstacles is non-trivial. Thus the understanding of how these obstacles contribute to the CRSS requires simulations which, to the best of our knowledge, have never been done before. Similar

An example of application of the three obstacle addition law for an Al–Mg–Sc alloy

To test the methodology developed above, the experimental stress–strain response of a precipitation hardened Al–Mg–Sc has been re-examined using the proposed addition law. This particular alloy is a Al-2.8 wt% Mg-0.16 wt% Sc alloy that was originally studied by Fazeli et al. [12]. Strengthening contributions in this alloy comes from solid solution strengthening due to Mg (σsol), Al3Sc precipitates (σppt) and work hardening due to the accumulation of dislocations (σdis) [24], [25], [26].

In the

Conclusion

Areal glide simulations have been used to study the motion of a dislocation across a glide plane populated with randomly distributed obstacles of different strengths. These results were utilized to derive a new expression for the exponent q in the addition law for two sets of obstacles. This expression is an analytical function of only the breaking angle of the respective sets. The addition law was then extended to the case of three sets of obstacles. These new laws were then used to re-examine

Acknowledgement

The authors would like to gratefully acknowledge the financial support from Natural Sciences and Engineering Research Council of Canada (NSERC). A. de Vaucorbeil would like to acknowledge Henry Proudhon and the École Normale Supérieure de Cachan for their support during the final year of his Master's program.

References (29)

  • A.W. Zhu et al.

    Acta Mater.

    (1999)
  • F. Fazeli et al.

    Acta Mater.

    (2008)
  • J. da Costa Teixeira et al.

    Acta Mater.

    (2009)
  • G. Fribourg et al.

    Acta Mater.

    (2011)
  • U.F. Kocks et al.

    Prog. Mat. Sci.

    (2003)
  • G.P.M. Leyson et al.

    Acta Mater.

    (2012)
  • E.A. Marquis et al.

    Acta Mater.

    (2003)
  • L.M. Brown et al.
  • A.H. Cottrell

    Dislocations and Plastic Flow in Crystals

    (1953)
  • T.J. Koppenaal et al.

    Appl. Phys. Lett.

    (1964)
  • R. Labusch

    Phys. Stat. Sol. (b)

    (1970)
  • A.J.E. Foreman et al.

    Can. J. Phys.

    (1967)
  • U.F. Kocks et al.

    Prog. Mat. Sci.

    (1975)
  • A. Ardell

    Metall. Trans. A

    (1985)
  • Cited by (86)

    View all citing articles on Scopus
    View full text