Micromechanical modelling of fracture toughness in overaged 7000 alloy forgings

https://doi.org/10.1016/j.msea.2006.07.018Get rights and content

Abstract

In this article, a multiple micromechanisms-based model that quantitatively relates the plane-strain fracture toughness, KIC, of overaged 7000 alloy forgings to their microstructural attributes, fracture surface morphology and basic tensile properties is developed. To verify the proposed model, extensive microstructural and fractographic analyses along with mechanical tests are carried out using three industrially produced alloys with different contents of Fe and Si impurities. The fracture mechanisms are identified and individual contributions to the overall fracture are quantitatively assessed. The fracture toughness is then calculated using the experimentally obtained average values of relevant mechanical properties, area fractions of main fracture modes and microstructural parameters such as volume fraction of coarse intermetallic (IM) particles, their size and spacing, density of intragranular precipitates, number and width of the precipitate-free zones (PFZ). The proposed model correctly predicts the effect of individual microstructural parameters on the overall fracture behaviour.

Introduction

High-strength 7000 series aluminium alloys based on the Al–Zn–Cu–Mg system are used in many structural applications where a good compromise between strength and damage tolerance is required [1], [2], [3]. Since in that case the fracture toughness is a critical parameter, an understanding of the deformation and fracture mechanisms can help to improve the fracture resistance of these alloys. The fracture properties could be strongly affected by a number of microstructural features, flow strength, strain-hardening rate and slip character [2], [3]. The individual features, including coarse intermetallic (IM) phases, dispersoids, aging precipitates and grain structure, play a fundamental role in determination of the fracture modes and toughness levels of precipitation-hardened 7000 alloy products [3], [4], [5]. It is now recognised that their complex microstructure contributes to the phenomenon of multiple micromechanisms in fracture processes of wrought 7000 alloys [1], [3], [6], [7]. The three major fracture modes are coarse voiding at IM particles, ductile intergranular fracture along high-angle grain boundaries and transgranular fracture.

The first steps in the sequence of events leading to the overall fracture are the nucleation, growth and coalescence of voids at the coarse (0.5–10 μm) IM particles. Because they are brittle and have weak interfaces, they either fracture or separate from the matrix when the local normal stress exceeds a critical value [8]. As a result, they provide preferential crack paths ahead of a crack loaded to develop high stress intensity. The presence of these preferential crack paths reduces the energy needed to propagate the advancing crack and leads to lower toughness. The remaining fracture path is a combination of intergranular fracture and microvoid-induced transgranular fracture.

Since the lower toughness and resistance to fatigue crack propagation are related to an increasing number of coarse particles, there is a strong motivation to minimise the volume fraction of both soluble and insoluble IM phases. However, the removal of insoluble phase particles usually containing the Fe and Si impurities could be achieved only by a reduction of the content of impurities. The reduction is limited by the costs and availability of high-purity materials. Consequently, all commercial alloys contain significant amounts of Fe and Si, which react with Al and alloying elements to form a significant number of phases [9], [10].

Relatively small dispersoids (0.05–0.5 μm) used for suppressing recrystallisation and grain growth are usually not detrimental to alloy toughness [3], [11], but they may affect toughness indirectly by altering the degree of recrystallisation that occurs during the solution treatment [1], [3], [12]. Due to the high angle of the grain boundaries created during recrystallisation and the presence of relatively coarse incoherent η precipitates formed during quenching from the solution treatment temperature, the transfer of slip across these boundaries is difficult. This, together with the low yield stress in the precipitate-free zone (PFZ) adjacent to the grain boundaries, leads to low plastic energy dissipation for intergranular crack nucleation and propagation.

In solution treated and quenched 7000 alloys, an appropriate aging treatment leads to formation of fine η and η′ precipitates that harden the material. The interaction of dislocations with these precipitates uniformly distributed within the grains controls the transgranular fracture, whose relative contribution to the overall fracture depends on the state of precipitation. Hence, the fracture toughness of 7000 alloys is determined by the actual fraction of these micromechanisms in the overall fracture, which are in turn controlled by the microstructure. Micromechanical models, giving a direct relationship between the fracture behaviour and microstructure descriptors allow testing different microstructures in order to find the suitable microstructures for the required fracture properties. Some investigations have been carried out in order to account for and predict the effects of individual parameters on fracture toughness in plane-strain conditions [2], [6], [13]. However, the majority of the existing models generally consider only a selection of the microstructural and mechanical parameters that are known to be important. Furthermore, these models have been developed by making appropriate simplifications. They assume that the microstructure is isotropic, the grain-size distribution is uniform and coarse IM particles are not present [2], [5], [6].

Thus, the Garrett and Knott model based on the experimental observations of Hahn and Rosenfield [14], which considers the effects of strain-hardening coefficient and yield stress on toughness, is expressed as:KIC2CEεc*σyn21ν2,where C is a constant, εc* the critical crack tip strain at which unstable propagation occurs, n the work-hardening exponent, ν the Poisson ratio, σy and E are the yield stress and Young's modulus, respectively, and KIC is the plane-strain fracture toughness. It is taken that εc* is a function of the volume fraction and distribution of the particles at whose location voids initiate [13]. In that way consideration of transgranular fracture characterised by microvoid initiation, growth and coalescence has also been partially introduced into analysis. Several other models also assume that a single transgranular fracture micromechanism is operative [5], [15]. In addition, some attempts have been made to model the overall fracture process when the fracture is completely intergranular [2] or where both ductile intergranular and transgranular fracture modes are operative [6]. These models also describe the fracture toughness of alloys in absence of the IM particles. Only in an earlier Hahn and Rosenfield model, presented in a number of fracture toughness studies [2], [13], the effect of the volume fraction of coarse particles for constant yield strength and constant particle size was considered. Assuming that unstable crack propagation starts when crack tip opening reaches the length of the unbroken ligaments separating cracked particles, they gave the following equation:KIC2σyEπ21/3D1/2fv1/6,where D is the diameter of the coarse particles and fv is their volume fraction, while the rest of the terms were already defined. Their experimental data, however, did not confirm the predictions of that model, significantly underestimating the toughness.

All these results confirm that the modelling of the fracture toughness in high-strength 7000 alloys requires incorporation of the multiple-fracture micromechanisms for toughness predictions. Based on this argument and experimental observations as well, the following quantitative relationship between the microstructural parameters of the partially recrystallised 7050 alloy and the plane-strain fracture toughness has recently been proposed by Gokhale et al. [6]:[KIC]2=exp[ap(cosθpPV)n]F1+F2exparcosθrVV[Dr+2Δ]DrΔm,

where VV is the degree of recrystallisation, Dr the average size of the recrystallised grains, Δ the average thickness of the recrystallised regions and PV is the total surface area of the constituent particles per unit volume, while the anisotropy of the high-angle grain boundaries and of the constituent particles is characterised by 〈cos θr〉 and 〈cos θp〉, respectively. The parameters F1 and F2 are directly related to the fracture toughness of completely intergranular and transgranular fractures in the absence of the constituent particles. They suggest that overall fracture toughness level may be expressed by the contributions of individual fracture modes. They have experimentally measured the fracture surface area fractions, AA, generated from various operative fracture mechanisms, and established a simple fracture criterion: failure occurs when the sum of the area fractions of the surface of fracture induced by the three micromechanisms equals to 1. However, although their data were in reasonable agreement with the model proposed for the fracture toughness of the thin hot-rolled plates, its applicability is limited.

One reason for this limitation is a relatively low recrystallisation level, typical for commercial thick-plate products [16]. Moreover, in heavy plates and forgings the microstructure will exhibit only a limited amount of mechanical working and the microstructure of the component will often resemble that of the as-cast material [17]. Because of the importance of such products produced for an aircraft for instance, modelling of fracture toughness requires improved prediction of fracture resistance for high-strength aluminium alloys products of varying thickness.

Detailed quantitative microstructural and fractographic investigation, together with mechanical tests, are performed in order to develop a model capable of predicting the plain-strain fracture toughness of overaged 7000 alloy (high-zinc variant) forgings with different microstructures controlled by the variations in composition. The effects of, and synergy between, the attributes associated with the coarse IM particles and the state of precipitation within the actual commercial microstructures have been considered. The major fracture modes have been identified and their area fractions as a function of microstructural state determined. The obtained data are then used to model the toughness variation with basic tensile properties and microstructural parameters. The existing models reviewed above are integrated into the current analysis to the extent to which it is possible.

Section snippets

Experimental procedure

The material used in this research was the hot-forged 50-mm-thick pancake-type plates of three industrially produced alloys, having the compositions given in Table 1. Apart from the Zn contents, the only significant difference in composition between the alloys is the total (Fe + Si) content, ranging from 0.23 to 0.37 mass%. All plates were supplied in T73 temper. After solution treatment (1 h at 460 °C) and water quenching, the materials were aged for 5 h at 100 °C plus 5 h at 160 °C.

The differences in

Microstructures

Optical microscopy revealed a deformed dendrite cell structure for all forgings. Due to the appreciable amounts of Cr, Mn and Zr as dispersoid-forming elements, their hot-working sequence to a thick-plate product takes place below the temperature of recrystallisation. Fig. 1a illustrates a typical microstructure, showing elongated dendrite arm boundaries with a significant number of coarse IM particles and segregation of dispersoids and aging precipitates in Al-rich matrix.

Irregularly shaped

Conclusions

A quantitative microstructural and fractographic analysis showed that the rupture mode of the industrially produced 7000 alloy (high-zinc variant) forgings in overaged condition is a complex phenomenon, consisting of coarse voiding at large IM particles, ductile intergranular fracture and microvoid-induced transgranular fracture. The amount, size and spatial distribution of the coarse Fe- and Si-containing particles had a strong influence on the crack propagation modes and toughness level. The

Acknowledgements

This work was financially supported by the Ministry of Science and Environmental Protection of the Republic of Serbia through the project no. 144027. All SEM and TEM experiments were performed at the Institute of Materials Science & Technology and USTEM, Technical University of Vienna. The authors gratefully acknowledge Professors H.P. Degischer and G. Rumplmair for their kind help.

References (18)

  • D. Dumont et al.

    Mater. Sci. Eng. A

    (2003)
  • T. Pardoen et al.

    J. Mech. Phys. Solids

    (2003)
  • N.U. Deshpande et al.

    Metall. Mater. Trans. A

    (1998)
  • B. Morere et al.

    Metall. Mater. Trans. A

    (2000)
  • R.C. Dorward et al.

    Metall. Mater. Trans. A

    (1995)
  • N. Kamp et al.

    Metall. Mater. Trans. A

    (2002)
  • A.M. Gokhale et al.

    Metall. Mater. Trans. A

    (1998)
  • J.T. Staley

    Aluminium

    (1979)
  • X.M. Li et al.

    Mater. Sci. Technol.

    (2001)
There are more references available in the full text version of this article.

Cited by (29)

  • Effect of three-dimensional deformation at different temperatures on microstructure, strength, fracture toughness and corrosion resistance of 7A85 aluminum alloy

    2022, Journal of Alloys and Compounds
    Citation Excerpt :

    The highest fracture toughness was obtained for the CD sample, which was improved by 21.4% as compared to that of the UD sample. The fracture toughness of the alloy is mainly affected by the second-phase particles and the strength difference between the matrix and the grain boundary [9,26,27]. The same aging temper under the five deformation processes led to little difference in the size of their MPs and GBPs, so the main influence on the strength difference between the matrix and the grain boundary lies in the grain boundary PFZ.

  • Investigation on strength, toughness and microstructure of cryogenically-deformed 7A85 aluminum alloy under various aging tempers

    2022, Materials Characterization
    Citation Excerpt :

    The pinning difference between the matrix and grain boundaries is reduced, which in turn causes an even distribution of stress inside the alloy during tensile tests, which improves the EL. Studies have shown that the fracture mechanism after aging treatment depends on the strength difference between the matrix and grain boundaries [5,27]. For aluminum alloys, the main influences on the matrix and grain boundary strengths are the size and distribution of the MPs and GBPs and the PFZ width at the grain boundaries [27,28].

  • Effect of heat treatment on stress corrosion cracking, fracture toughness and strength of 7085 aluminum alloy

    2014, Transactions of Nonferrous Metals Society of China (English Edition)
  • STEM-HAADF tomography investigation of grain boundary precipitates in Al-Cu-Mg alloy

    2011, Materials Letters
    Citation Excerpt :

    As for a GB segment as illustrated in Fig. 1a, these parameters were determined by the following methods. The first one is suggested to tilt the GB surface to be parallel to the incident beam (edge-on) firstly in a transmission electron microscope (TEM) (Fig. 1b) and determine the average values of GBP size, center to center distance of GBPs according to the statistics results, then estimate the number of GBPs per unit GB area and area fraction of GB covered by GBPs using mathematical approximation methods [8,14–16]. However, due to the factor of sample thickness, while keeping GB edge-on GBPs in different depth will more or less overlap with each other especially when they are high in density and small in size.

View all citing articles on Scopus
View full text