Effects of temperature and strain rate on critical damage value of

AZ80 magnesium alloy

XIA Yu-feng( ), QUAN Guo-zheng( ), ZHOU Jie( )

School of Materials Science and Engineering, Chongqing University, Chongqing 400044, China
Received 23 September 2009; accepted 30 January 2010

Abstract: A series of AZ80 billets were compressed with 60% height reduction on hot process simulator at 250, 300, 350, 400
under strain rates of 0.01, 0.1, 1 and 10 s−1. In order to predict the occurrence of surface fracture, the values of the Cockcroft-Latham
equation were calculated by the corresponding finite element numerical algorithm developed. A concept about damage incremental
ratio in plastic deformation was defined as the ratio of damage increment at one step to the accumulated value. A method of finding
the intersection of incremental ratio varying curve and simulation step axis was brought forward to make the fracture step certain.
Then, the effects of temperature and strain rate on critical damage value were achieved. The results show that the critical damage
value is not a constant but changes in a range of 0.021 8−0.378 0. It decreases significantly with the increase of strain rate at a certain
temperature. While under a certain strain rate, the critical damage value has little change with the increase of temperature.
Key words: AZ80 magnesium alloy; critical damage value; compaction; finite element analysis

constant of material like yield stressor stress limit.

1 Introduction Cockcroft and Latham hasn’t expounded whether the
critical damage value depends on the temperature and
An important concern in metal-forming processes is strain rate[5−9]. In this study, for AZ80 magnesium alloy,
whether the desired deformation can be accomplished the relations between critical damage value and
without fracture of the workpiece. In industrial practice, temperature and strain rate were analyzed based on a
however, the empirical know-how of the designer is series of compressing tests data and damage data by
decisive for the fracture-free quality of the products, but numerical computation.
often requires very costly trial-and-error[1−2]. Thus,
there is a critical need for predicting and preventing 2 Experimental
fracture, which is a major feature of the forming
processes and the quality of the products. The chemical compositions of AZ80 magnesium
Most bulk metal-forming processes may be limited alloy were 8.90% Al, 0.53% Zn, 0.2% Mn, 0.008% Si,
by ductile fracture e.g. the occurrence of internal or 0.004% Fe, 0.008% Cu, 0.008% Ni (mass fraction).
surface fracture in the workpiece. If it was possible to A computer-controlled, servo-hydraulic Gleeble
predict the conditions of the deforming workpiece which 1500 machine was used for compression testing. It can
lead to fracture, it may be feasible to choose appropriate be programmed to simulate both thermal and mechanical
process conditions and to modify the forming processes industrial processing variables in a wide range of
in order to produce sound products[3−4]. hot-deformation conditions[10]. The specimens were
Historically, ductile-fracture criteria are based on resistance heated to the deformation temperature at a
experimental work that utilizes a deformation process heating rate of 1 /s and held at that temperature for 180
related to actual industrial applications. So far, the s. The homogenized ingot was scalped to diameter of 10
fracture criterion such as Cockcroft-Latham is suited for mm and height of 12 mm with grooves on both sides
tenacity fracture in bulk metal-forming simulation. filled with machine oil mingled using graphite powder as
Usually, the critical damage value is considered as an lubrication during isothermal hot compression tests at

Foundation item: Projects(CSTC, cstc2009aa3012-1) supported by Science and Technology Committee of Chongqing, China; Project(20100470813)
supported by China Postdoctoral Science Foundation; Project(2010011511) supported by Sharing Fund of Chongqing University’s
Large-scale Equipment
Corresponding author: XIA Yu-feng; Tel: +86-23-65103065; E-mail: [email protected]
 XIA Yu-feng, et al/Trans. Nonferrous Met. Soc. China 20(2010) s580−s583 s581
250, 300, 350, 400 ˚C and strain rates of 0.01, 0.1, 1 and damage theory, developed a damage computation
10 s−1. module which has been applied successfully to a variety
of loading situations[12−14]. It has been used to assess
3 Analysis of damage in deformation the degree of working imposed on magnesium alloy in
such processes as extrusion, rolling and upsetting by
3.1 Stress strain behavior tools. The damage in plastic deformation is defined by
The true compressive stress strain curves of AZ80 Cockcroft and Latham as an amount of work that the
magnesium alloy are shown in Fig.1. The flow stress as ratio of maximum tensile stress σT to effective stress σ
well as the shape of the flow curves is sensitively carries out through the applied equivalent strain ε in a
dependent on temperature and strain rate[11]. For all of metal-working process, i.e.
specimens, after initial yielding, the flow stress decreases εf σT
monotonically with different softening rates. For a C=∫ dε (1)
0 σ
specific strain rate, the flow stress decreases markedly
with the increase of temperature, while for a specific where ε f is the total equivalent strain at the end of
temperature, the flow stress increases markedly with the forming process. The magnitude of C cannot exceed a
increase of strain rate. Only for strain rate of 10 s−1, maximum value Cmax (critical damage value) to failure.
fracture is seen to occur after a dynamic softening. By comparing C with Cmax, the risk of material failure
For the simulation of plastic deformation process, during processing is assessed. In order to calculate C by
the material model for billet can be defined by inputting FE simulation, Eq.(1) has to be converted to an
the true stress strain curve data (as Fig.1). The yield appropriate discrete expression convenient for FE code:
stress of AZ80 magnesium alloy under uniaxial
εf σ T dε tf σ T
tf
σ ε&∆t
conditions as a function of strain ( ε ), strain rate ( ε& ), C=∫ dt = ∫ ε&dt ≅ ∑ T (2)
and temperature (T) can also be considered as flow stress.
0 σ dt 0 σ 0 σ
The metal starts flowing or deforming plastically when
the applied stress reaches the value of yield stress. where ε& is the equivalent strain rate as Eq.(1)
calculated from the individual principal strain-rate
3.2 Cockcroft and Latham damage components, and ∆t is the variable time increment used
Cockcroft and Latham, based on cumulative in the FE analysis.

Fig.1 True stress strain curves of AZ81 Mg alloy at different strain rates and temperatures: (a) ε& =0.01 s−1; (b) ε& =0.1 s−1; (c)
−1 −1
ε& =1 s ; (d) ε& =10 s
s582 XIA Yu-feng, et al/Trans. Nonferrous Met. Soc. China 20(2010) s580−s583
According to cumulative damage theory, the damage value), is dependent on the same material
damage value a moment ago is less than that a moment parameters that forming limits are dependent on.
later during a compressing process. Therefore, the Metallurgical properties such as the microstructure, alloy
maximum value will appear at the last simulation step, constants, grain size and grain form, and non-metallic
but it does not mean fracture step. Fig.2 shows the inclusion content, whilst having a small effect on the
damage distribution at the last step (height reduction of strength and the hardness, have a significant effect on the
60%) at 300 and strain rate of 0.01 s−1. From the critical damage value[15]. In order to predict the
simulation results, it can be seen that the maximum occurrence of surface fracture, the value of the
damage value appears in the region of upsetting drum, Cockcroft-Latham equation expressed by means of Eq.(1)
while the minimal value appears in the middle region. is calculated at the integration point inside the elements.
Furthermore, a concept about incremental ratio of
Cockcroft-Latham damage in plastic deformation (Rstep)
is brought out and defined as the ratio of the damage
increment at one step (∆D) to the accumulated value
(Dacc):

Rstep = ∆D / Dacc (3)

Fig.3 shows this incremental ratio varying during

the upsetting processes at different temperatures and
different strain rates. It can be seen that Rstep decreases to
Fig.2 Damage distribution at last step (height reduction of 60%) the trough point rapidly, then it has a slight increase,
at 300 and strain rates of 0.01 s−1 soon after which it decreases to zero gradually. The zero
point has a tolerances of 0−0.03. In this study, the zero
3.3 Determination of critical damage value of AZ80 point is assumed as the fracture time, and the other
The Cockcroft-Latham constant, C max (critical horizontal curve segment is considered as the dynamic

Fig.3 Incremental ratio varying of Cockcroft-Latham damage during compressing process at different temperatures and different
strain rates: (a) ε& =0.001 s−1; (d) ε& =0.1 s−1; (c) ε& =1 s−1; (d) ε& =10 s−1
 XIA Yu-feng, et al/Trans. Nonferrous Met. Soc. China 20(2010) s580−s583 s583
softening due to an increase of dislocation density and changes in a range of 0.0218−0.378. It decreases
the dislocation multiplication rate. significantly with strain rate increasing at a certain
To find the fracture time (step), the point arrays temperature. While under a certain strain rate, the critical
after step 40 are picked out from each incremental ratio damage value has little change with temperature
varying curve and fitted linearly (as shown in Fig.3). The increasing. In a word, it is more sensitive to strain rate.
intersection of line fitted and horizontal axis is obtained,
and it is made certain as the fracture step. Then, the References
simulation with only 60 steps is continued to run until
the fracture. The maximum cumulative damage value at [1] SOWERBY R, CHANDRASEKARAN N. The prediction of damage
last step or the critical damage value is computed. By accumulation when upsetting AISI 1045 steel specimens, based on
McClintock's model [J]. Materials Science and Engineering, 1986,
this way, the critical damage value of AZ80 magnesium
79(1): 27−35.
alloy can be achieved. Fig.4 shows the effects of
[2] DORUM C, HOPPERSTAD O S, BERSTAD T. Numerical
temperature and strain rate on critical damage value of modelling of magnesium die-castings using stochastic fracture
AZ80 magnesium alloy. It can be seen that the critical parameters [J]. Engineering Fracture Mechanics, 2009, 76(14):
damage value is not a constant but changes in a range of 2232−2248.
0.021 8−0.378 0. Further, the critical damage value [3] KO D C, KIM B Y, CHOI J C. Prediction of surface-fracture
decreases significantly with strain rate increasing at a initiation in the axisymmetric extrusion and simple upsetting of an
certain temperature. While under a certain strain rate, the aluminum alloy [J]. Journal of Materials Processing Technology,
1996, 62(1/3): 166−174.
critical damage value has little change with the increase
[4] PEREIRA I M, RUBIM G, ACSELRAD O. Comparison of the
of temperature. It can be seen that the critical damage
experimental and the numerically predicted mechanical damage in
value depends on the temperature and strain rate, and it is the sheet forming of steel [J]. Journal of Materials Processing
more sensitive to strain rate. Technology, 2008, 203(1/3): 13−18.
[5] QUAN Guo-zheng, TONG Ying, CHEN Bin. A constitutive
description for drawing limit of magnesium alloy tube based on
continuum damage mechanics [J]. Materials Science Forum, 2009,
610/613: 951−954.
[6] TAI Wei-hua, YANG Bing-xian. A damage criterion for ductile
fracture and its modification applied to sheet metal forming [J].
Journal of Mechanical Working Technology, 1987, 15(1): 319−333.
[7] DEY S, BORVIK T, HOPPERSTAD O S. On the influence of
fracture criterion in projectile impact of steel plates [J].
Computational Materials Science, 2006, 38(1): 176−191.
[8] KOMORI K. Effect of ductile fracture criteria on chevron crack
formation and evolution in drawing [J]. International Journal of
Mechanical Sciences, 2003, 45(1): 141−160.
[9] LADEMO O G, ENGLER O, KELLER S. Identification and
validation of constitutive model and fracture criterion for Al-Mg-Si
alloy with application to sheet forming [J]. Materials and Design,
2009, 30(8): 3005−3019.
[10] ZHANG Xian-hong, RUAN Xue-yu, OSAKADA K. Forgeability of
Fig.4 Effects of temperature and strain rate on critical damage
AZ31B magnesium alloy in warm forging [J]. Transactions of
value of AZ80 magnesium alloy
Nonferrous Metals Society of China, 2003, 13(3): 632−635.
[11] XU Wen-chen, SHAH De-bin, YANG Guo-ping. Flow behavior and
4 Conclusions microstructure evolution during hot compression of TA15 titanium
alloy [J]. Transactions of Nonferrous Metals Society of China, 2006,
1) The true compressive stress strain curves of 16(s3): s2066−s2071.
AZ80 magnesium alloy deformed at four temperatures [12] COCKCROFT M G, LATHAM D J. Ductility and workability of
under four strain rates are obtained. The flow stress as materials [J]. J Inst Met, 1968, 96: 33−39.
[13] WARD M J, MILLER B C, DAVEY K. Simulation of a multi-stage
well as the shape of the flow curves is sensitively
railway wheel and tyre forming process [J]. Journal of Materials
dependent on temperature and strain rate.
Processing Technology, 1998, 80/81(3): 206−212.
2) According to cumulative damage theory, the [14] FAN W F, LI J H. An investigation on the damage of AISI-1045 and
maximum value appears at the last simulation step, but it AISI-1025 steels in fine-blanking with negative clearance [J].
does not mean fracture step. Thus, a method of finding Materials Science and Engineering A, 2009, 499(1/2): 248−251.
the intersection of incremental ratio varying curve and [15] XUE Liang. Damage accumulation and fracture initiation in
simulation step axis is brought forward to make the uncracked ductile solids subject to triaxial loading [J]. International
fracture step certain. Journal of Solids and Structures, 2007, 44(16): 5163−5181.
(Edited by ZHAO Jun)
3) The critical damage value is not a constant but