Accepted Manuscript

Technical report

Investigation of tension and compression behavior of AZ80 magnesium alloy

F. Fereshteh-Saniee, Kh. Fallah-Nejad, A. Sh. Beheshtiha, H. Badnava

PII: S0261-3069(13)00288-4

DOI: http://dx.doi.org/10.1016/j.matdes.2013.03.080

Reference: JMAD 5309

To appear in: Materials and Design

Received Date: 25 December 2012

Accepted Date: 24 March 2013

Please cite this article as: Fereshteh-Saniee, F., Fallah-Nejad, Kh., Sh. Beheshtiha, A., Badnava, H., Investigation

of tension and compression behavior of AZ80 magnesium alloy, Materials and Design (2013), doi: http://dx.doi.org/

10.1016/j.matdes.2013.03.080

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers

we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and

review of the resulting proof before it is published in its final form. Please note that during the production process

errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Investigation of tension and compression behavior of AZ80 magnesium

alloy

F. Fereshteh-Saniee, Kh. Fallah-Nejad*, A. Sh. Beheshtiha, H. Badnava Mechanical Engineering Department, Bu-Ali Sina University, Hamedan, Iran

Abstract

Magnesium alloy deformation should be carried out at high temperature and it is essential

to investigate the deformation behavior of these alloys at high temperature. In this paper,

practical tests are conducted on AZ80 alloy which includes tension and compression tests

at high temperature and different strain rates. As this alloy is sensitive to temperature and

strain rate, tension tests are difficult to carry out. The traditional compression test should

be conducted in zero-friction condition but such a condition is impossible to prepare.

Therefore, bulge correction factor and numerical correction factor are used to eliminate

the friction effect which exists between the surfaces. Tensile tests are also carried out on

the standard specimens at high temperature and different strain rates. The effect of

necking phenomenon is corrected using Bridgman correction factor and is simulated

using finite element software. Also T – shape compression test is used to valuate friction

parameter at high temperatures.

Keywords: AZ80-magnesium alloy; compression test ; Tension tests ;T shape test; Finite Element

simulation

1. Introduction

In many industries, engineers tend to produce structures with a high strength and low

weight. For this reason, nowadays magnesium alloys are very attractive for many

researchers and are extensively used in many important industries such as automotive and

aerospace [1]. In addition, magnesium alloys have other unique properties which make

them different in comparison with other metals. Some of these properties are high

corrosion resistance, welding ability and good recyclability. AZ80 is one of the most

important magnesium alloys and is recommended as a good alloy for forging in ASTM

B80 - 09 Standard Specification for Magnesium-Alloy Sand Castings. But there is no

complete research about its properties until today.

Many parameters can be found from the tension curves such as elasticity module,

yield stress, tensile strength and strain parameters. To achieve this aim, it is necessary to

investigate the mechanical and metallurgical phenomena which happen in a tension test.

Strain hardening and strain softening phenomena should be considered during the test.

Many researchers have studied magnesium alloys properties during the last decades.

Fereshteh-Saniee et al. presented a material model for stress-strain curve behavior of

some magnesium AZ series which was able to predict the behavior of those alloys

correctly [2]. Mathis et al. investigated the deformation behavior of 91, AE42 and

AS21 magnesium alloys in a wide temperature range and different strain rates. The effect

of strain hardening coefficient on stress was investigated as a criterion of strain hardening

and strain softening. Their tests also showed that strain hardening decreases with

increasing the temperature and stress [3, 4]. Sivapragash et al. studied the deformation

and fracture behavior of ZE41A at different temperatures and strain rates using tension

test. This investigation an analytical model was proposed to predict deformation behavior

at various temperatures and strain rates [5]. Wen et al. investigated the mechanical

behavior of AZ31 and found that “temperature activated plates” cause strain rate

sensitivity. They also found that the reason behind the asymmetry in value of the yield

stress in tension and especially compression in low temperature was twinning [6]. Abedi

et al. investigated tension and deformation behavior of AZ31 at high temperature range

and at strain rate of 0.001s-1

[7]. Masoudpanah and Mahmudi investigated tension

deformation and micro structure properties of AZ31 alloy after extrusion process and

ECAP process. They showed that the ECAP specimens have lower yield stress and higher

formability in comparison with the extrusion specimens [8]. Jayamathy et al. studied the

effect of reinforcement on compression deformation and impact reaction in AZ92 alloy.

They investigated the effect of “Sic” in compression deformation and energy absorption

ability in this alloy [9]. Helis et al. did many researches to improve the microstructure of

the AZ31 magnesium alloy during axial compression deformation in high temperature

[10].

There are some studies about tension process simulation when the necking

phenomenon occurs. Ponthot carried out a finite element simulation using an

imperfection with light slip which was placed in the middle of specimen. The result of

this simulation was in good agreement with experimental results [11].

Strain rate and temperature are two important parameters to predict the mechanical

behavior of materials. Song et al. investigated the compression properties of three AM20,

AM50 and AM60 alloys in high strain rates. Results showed that strain rate sensitivity

increases by increasing the strain rate [12]. Palumbo et al. experimentally and

numerically studied AZ31 magnesium alloy at high temperature and steady strain rate.

They proposed an equation between strain rate and the griper displacement [13].

Anbuselvan and Ramanathan investigated deformation of ZE41A alloy using

compression tests in high temperatures and different strain rates. Their results showed

that the optimized deformation parameters of this alloy are 400°C temperature and 0.1S -1

strain rate [14]. Raghu et al. also used a ring test to investigate friction coefficient in hot

deformation of the ZM-21 magnesium alloy. This investigation is practiced at various

temperatures and by different lubricants [15]. Narayanasamy et al. studied buckling

deformation of magnesium alloys. they proposed a model to determine the buckling

radius. They showed that buckling radius depended on geometric parameters of primal

specimen [16].

In this paper T shape method is used to investigation of temperature and strain rate

effects on friction factor. Compression tests are conducted to investigate behavior of

AZ80 at 250°C and 300° C temperature and 0.001s-1

, 0.01s-1

and 0.1s-1

strain rates. Also,

tension tests are carried out at 250° C, 275° C, 300° C, 325° C and 350°C temperature

and 0.0005s-1

, 0.001s-1

and 0.005s-1

strain rates. It is obvious that when a metal forming

process is performed, appropriate temperatures and strain-rates should be chosen to

prepare enough strain for that process. The temperatures and strain-rates in this research

are chosen so that the results are useful for other metal forming and simulation processes

such as deep drawing. Although the behavior of this magnesium alloy specially

mechanical behavior have not yet investigated, but it can be found from the other

literatures that almost all researchers who have studied the other magnesium alloys have

practiced their investigations in same temperatures and strain rates ranges approximately

or their tests have conducted in ranges to be able to prepare same strains as this research

[6, 10, 17, 18]. Bulge and numerical correction factor are used to correct the results of

compression tests. The effects of strain rate and temperature on the bulge and numerical

correction factors are studied. Also, T shape test is used to evaluate friction in hot

deformation. Bridgman correction factor is used to correct the tension tests results.

Tension and compression processes are simulated using finite element analysis. Some

microstructure properties of AZ80 are investigated utilizing optical microscope images of

compression tests and SEM images of tension tests.

2. Friction

Two friction laws are used to compute contact friction which exists between specimen

and gadgets. Colomb law is an important model to describe friction. This law is shown in

Eq. 1. p (1)

is the friction stress, is the friction module and p is the vertical compression on

contact surface. This law is suitable for low compression deformation like 1.5o

p

,

where o is the flow stress of specimen. The second law is the shear friction law which is

shown in Eq. 2.

3o

m (2)

which m is the friction factor. In this paper only shear friction law is used. Friction test

should be conducted under conditions which are almost resembling the reality. Therefore,

a suitable friction test should have a high surface compression and high ((new surface

production)). T shape compression test is used to valuate friction parameter.

3. T shape method

3. 1. T shape specimens’ properties

AZ80 magnesium alloy is produced by casting. AZ80 alloy compositions are shown in

Table 1. Microstructure image of this alloy prior to deformation which is taken by optical

microscope is shown in Figure 1. Cylindrical specimens which have a diameter of 6mm

and a length of 9mm are used for T shape tests and compression tests. These specimens

are produced by machining. A schematic image of T shape compression test is shown in

Figure 2. According to the ASTM A681 - 08 which is Standard Specification for Tool

Steels Alloy, L6 steel is used to make casts. The first segment is a cylinder which is

29mm in diameter and a V shape groove is prepared for it. The second segment is also a

cylinder which is used for loading procedure. The maximum displacement for each T

shape test is 4mm. The tests are conducted at 250 C and 300 C temperatures and

10.001s , 10.01s and 10.1s strain rates without any lubricant. To ensure accurate results

each test is carried out several times.

3. 2. Results of T shape test

At the end of each test, the deformed sample was taken out of the electrical furnace and

the necessary measurements were done. The microstructure of the alloy changed due to

friction which is shown in Figure 3. This figure illustrates that how the flow pattern is

affected by the geometry of T-shape test that described before. Pattern of grains which

are situated near the surfaces are affected by friction and the length of these grains has

increased while this change cannot be seen for the grains which are far from the

surfaces.The elongated grains are in the vicinity of edge radii of the grooved die. A point

is specified in calibration curves based on finite element simulation using extruded length

and the applied force on each specimen. The friction factor of specimen is defined by this

point. Calibration curves for various temperatures are shown in Figure 4. The average of

friction factors for various lubrication conditions based on experiment results is shown in

Table 2. It can be seen that MoS2 always has the lowest friction and dry condition always

has the highest friction. Finite element simulation is performed using friction factor

obtained from calibration curves to verify the accuracy of results. Using numerical

simulation of T-shape friction test were first carried out for various values of shear

friction factor. In this step, results of the compression tests were used as input data to

define material behavior. This process has been performed to obtain different sets of

calibration curves in terms of forming load and height of extruded part. For each set of

conditions, the test was conducted thrice. The height of extruded part and reaction force

of the sample is affected by the friction. Hence, for higher frictions, this height becomes

smaller. Also, in order to produce the same height for extruded part, higher applied force

is required. It is obvious from Fig. 4 that for a greater strain rate, the necessary forming

load becomes larger, although the slope of the linear part of the load curve and the

friction factor decrease. By increasing the forming load, strain rate icreases. It is

reasonable because at higher strain rates, the material possesses greater flow

stresses.From Figure 5 it can be seen that simulation results are in good agreement with

experimental results. In order to verify the obtained results, numerical simulations of T-

shape tests have conducted based on compression tests flow curves and shear friction

factors which are determined by calibration curves.

4. Compression and tension tests

4. 1. Bulge and numerical correction factors

Bulge correction factor is a method to eliminate the effect of friction and determination of

material flow stress in compression tests. The schematic image of the test before and after

compression load is shown in figure 6. It is obvious that buckling is caused by friction

between specimen and cast surface. Therefore, a correction factor is used to evaluate the

true stress.

The average stress in the midsection ( ave ) and corrected flow stress ( f ) are computed

from Eq.3 and Eq.4.

22/ ( / 2)ave L d (3)

f aveC (4)

L is the deformation force and C can be computed from Eq.5. using analytical method

which is based on stress distribution in the mid section. 1

21 ln 1

2

R aC

a R

(5)

R is the bulge section radius and can be evaluated by Eq. 6. 2 2

2 1

2 1

( )

4( )

h d dR

d d

(6)

Saniee et al. carried out some compression tests and corrected stress-strain curves [19].

They proposed numerical correction factor using numerical simulations which was in

better agreement with experimental results than bulge correction factor. Bulge correction

factor is just based on geometric parameters and other effective parameters are not

considered. Saniee et al. proposed numerical correction factor which considers the

friction effect. Bulge correction factor always increases when strain rises but numerical

correction factor decreases in high strains which is a result of increase in the amount of

friction. Experiments carried out on thin and plumb specimens at room temperature to

verify his results.

4. 2. Compression tests

Mechanical properties and stress-strain behavior of AZ80 magnesium alloy are

investigated using compression tests. Stress-strain curves are obtained at 250oC and

300oC temperatures and 10.001s , 10.01s and

10.1s strain rates. Before carrying each

test, specimens are kept at mentioned temperatures for 5 minutes and then chilled with

cold water. To evaluate the bulge correction factor it is necessary to measure changes of

diameter in various strains and each test is carried out in several stages. For this reason,

various geometrical parameters of the deformed sample should be measured for each

value of axial strain, each flow curve of individual alloys were gained using 8 samples,

each sample for a specific value of the axial strain. To ensure accurate results each test is

carried out several times.

4. 3. The results of compression tests

Stress-strain curves of compression tests are shown in Figure 7. At the beginning of

curves it can be seen that by strain increase, the amount of stress starts to go up. Stress

then decreases and becomes steady at the end. At first, the material has a strain hardening

behavior and then has strain softening behavior. This trend can be seen in the other

magnesium alloys such as ZE41, AZ31, AM20, AM50 and AM60 [6, 12]. The

compression strength of AZ80 is higher in comparison with AZ31 alloy at distinct hot

temperature and strain rate. The shape of stress–strain curves is considered to contain

some information related to the mechanisms of hot deformation. Such flow pattern is

characteristic for hot working accompanied by dynamic recrystallization [14]. From

Figure 7 we observed that the flow stresses are sensitively dependent on the strain rate

and temperature. This is conformed by earlier results published in literatures [14, 22]. At

the beginning of the curve, there is not much difference between bulge correction factor

and numerical correction factor and both of them are in a good agreement with the results

[18]. The results which are corrected by numerical correction factor are situated in upper

position. The force-displacement curves of two specimens are shown in Fig 8. It is

obvious that the results of numerical correction factor especially at high strains are more

acceptable as was shown in previous literatures [19]. For low frictions and reductions in

height, two correction factor methods provide reasonable flow curves and load curves.

Since, in bulge correction factor method, there is no consideration to parameters which

can be affected by temperature variation, such as geometry and material properties of the

sample and friction coefficient which is depend on the correction factor, where the

friction coefficient and/or the plastic true strain induced in the sample are quite high, the

results of the bulge correction factor method are overestimate in comparison with

experimental data of [19].

Two important and effective parameters in metal forming are temperature and strain rate.

The effect of strain rate which is corrected by the bulge and numerical correction factors

at 250oC and 300oC are shown in Figure 9. At a distinctive temperature, at low strains,

bulge correction factor does not have a remarkable change when strain rate increases. But

at higher strains, bulge correction factor always rises because of elimination of friction

effect. In fact, bulge correction factor is only based on geometric parameters. Strain rate,

temperature and friction affect each other. These effects are considered in the

simulations. In all temperatures, numerical correction factor does not have remarkable

change when the strain rate increases at low strains. Numerical correction factor

decreases at high strains as was anticipated. The reason of this decrease is the increment

of friction effect. However, in high strains the rate of increment of numerical correction

factor declines when strain rate increases. In fact, the effect of friction on deformation

force falls by increase of the strain rate. The results of bulge correction factor variations

at room temperature are in good agreement with [19].

Strain rate sensitivity parameter is defined by Eq. 7.

ln

lnm

(7)

Stress logarithm versus strain logarithm is shown in Figure 10. At a distinctive

temperature, strain rate sensitivity increases when strain goes up. It can be found that

stress is depended on temperature and strain rate. At distinctive strain rate, stress declines

when temperature increases. when the deformation temperature decreases while the strain

rate is constant, or the strain rate increases while the deformation temperature is kept

unchanged, the level of the flow curve decreased. The reason is that low strain rates and

high temperatures provide longer time for energy accumulation and higher mobility at

boundaries for the nucleation and increment of dynamic recrystallized grains and

dislocation annihilation [10].

4. 4. Tension tests

Tension specimen is 4mm in diameter and 20mm in gage length. The geometry of this

specimen is according to ASTM E8 / E8M - 11 which is Standard Test Methods for

Tension Testing of Metallic Materials. Tests carried out at

250 C , 275 C , 300 C , 325 C and 350 C temperatures and 10.005S , 10.001S and 10.0005S

strain rates. To obtain accurate results, each test is carried out several times. True stress

and strain can be found by engineering stress and strain (Eq. (8, 9)), if deformation be

considered continuous and homogeneous.

(1 )t e e (8)

ln(1 )t e (9)

Which t , e , t and e are true stress, engineering stress, true strain and engineering

strain respectively. As mentioned above, these equations can be used until the

deformation is uniform. Therefore, when the necking phenomenon occurs these equations

can not be used to evaluate true stress and true strain. After necking true stress and true

strain should be computed by Eq. (10, 11).

t

P

A (10)

ln( )it

A

A (11)

which A and iA are the moment and initial sections respectively. In this paper Eq. (8, 9)

are used to evaluation the true stress-strain curves until necking phenomenon occurs and

after necking Eq. (10, 11) are used to compute stress-strain curves by measuring the A.

Schematic curves of true and engineering stress-strain are shown in Figure 11. Also this

figure clarify the differences between true and engineering stresses after necking

phenomenon.

There is a three dimensional stress in necking section. Stress distribution in necking

section is shown in Figure 12. Because of the complicated state of stress in the necked

section, reduction values of area are dependent on specimen geometry and deformation

behavior. So they should not be assumed as true material properties. Hence, stress should

be corrected.

In necking section, stress should be computed by Eq. 12 [20].

corr r hyd (12)

corr is the corrected stress, r is the redial stress and hyd is the hydro static stress.

corr also can be computed by Eq. 13 [19].

corr aveB (13)

B is the Bridgman correction factor and ave is the average in smaller section. B can be

evaluated by Eq. 14 [21].

1

2(1 ) (1 )

2

BR a

Lna R

(14)

a is the radius of necked section and R is the radius of necked section profile. R can be

computed by Eq. 13. 2 2( )

4( )

L r aR

r a

(15)

L is the length of necked section and r is the radius of cylindrical section which is not

necked.

In Figure 12 L is corr which is the corrected stress [20].

True stress-strain curves of tension tests which are corrected by Bridgman correction

factor are shown in Figure 13. At a distinctive temperature, when strain rate increases, the

strength of alloy rises and elongation declines, but at a distinctive strain rate, when

temperature increases, the strength of alloy rises and elongation increases. At the

distinctive temperature stress increases and elongation declines when strain rate goes up

and at the distinctive strain rate, stress decreases and elongation increases when

temperature rises. The main reasons of these changes are chiefly recrystallizations and

slip mechanism which exist in magnesium alloys during the hot deformations and have

investigated and clarified in earlier results published in literatures [21]. The behavior of

stress-strain curves of this alloy is similar to the other magnesium alloys such as AZ31 or

AJ50 which are investigated in previous literatures [6, 4]. The tensile strength of AZ31

alloy is higher than the Az80 alloy at distinct hot temperature and strain rate. By contrast

the elongation of AZ80 is higher in comparison with AZ31. When recrystallization

occurs, new recrystallized grains are stress free and so with increasing the

recrystallization phenomenon, the strength of alloy decreases. These increases and

decreases are not in regular ranges quantitatively. The changes of fracture stress at

different temperatures and strain rates are shown in Figure 14. The decrease percentage

of fracture stress when a strain rate decreases from 0.005 s-1

to 0.0005 s-1

at 250° C is

9.1%, at 275° C is 7.4%, at 300° C is 9.4% and at 350° C is 5.4%. The changes of

fracture strain at different temperatures and strain rates are shown in Figure 15. The

increase percentage of fracture strain when a strain rate decreases from 0.005 s-1

to

0.0005 s-1

,

at 250° C is 25%, at 275° C is 8.5%, at 300° C is 18.7% and at 350° C is

19.6%.

Awareness and prediction of phenomena which occur during the tensile test is important.

One of the most important of these phenomena is necking which is an important

parameter in analyzing and investigation of data. LS-DYNA software is used to finite

element simulation and the PIECEWISE-LINEAR-PLASTICITY is chosen as material

model. This simulation is based on Ponthot simulation by use of an imperfection in the

middle of the specimen. This imperfection was a light slip which was created in the

simulated specimen [11]. The result of this simulation was in good agreement with

experimental results. Simulation model and force-displacement curves of simulation and

experiment at 300oC and 10.005S are shown in Figure 16.

4. 5. Investigation of fracture section of tension test

There are two kinds of fracture mechanism in magnesium alloys [22]:

1- Ductile mechanism which is because of fine microscopic voids.

2- Brittle mechanism which is because of microscopic and macroscopic cracks.

In AZ80 alloy the main reason of fracture is the existence of voids. In 100m scale, these

fine voids can be seen which are randomly distributed (Figure 17). These dense voids

which are signs of a ductile fracture of magnesium alloys at high temperature cause to

decrease of fracture strain [22]. Although it is difficult to find the center or the core of

these voids but it has found that the cogestion of them is around the solid particles which

are distributed in the alloy context. These particles are usually aluminum compounds.

The Centralism of microscopic voids around a dispersed particle and other second-phase

particles, occurs when the elastic energy in the particle exceeds the surface energy of the

newly formed voids surfaces. While this is a necessary condition, it must be aided by a

stress of the matrix-second-phase particle interface which is higher than the interfacial

strength. When interface stress reaches to a critical value, voids are anticipated to become

condensed. At the final stage usually the combination of fine voids can lead to generate a

shallow depth.

5. Conclusions

Tension and compression Experiment tests at various temperatures and strain rates are

carried out on AZ80 alloy. The results can be mentioned as below:

1- From T shape results It can be seen that MoS2 always has the lowest friction and

dry condition always has the highest friction. simulation results are in good

agreement with experimental results.

2- It found from Tshape method that for a greater strain rate, the necessary forming

load becomes larger, although the slope of the linear part of the load curve and the

friction factor decrease. Increase in the forming load with any growth in strain

rate is reasonable because at higher strain rates, the material possesses greater

flow stresses.

3- For T-shape tests, the height of extruded part and reaction force of the sample is

affected by the friction. Hence, for higher frictions, this height becomes smaller.

Also, in order to produce the same height for extruded part, higher applied force is

required.

4- At low strains, there is no discernable change in numerical correction factor. But

numerical correction factor grows when strain increases and reaches to its

maximum value at lowest strain rate. In the other word at median strains,

numerical correction factor declines when strain rate increases. But at high

strains, numerical correction factor decreases because of rising of friction effect.

5- Bulge correction factor always increases by the strain rising and has the lower

accuracy in compartion of the numerical correction factor. The main reason

6- Force-displacement curve is in good agreement with experimental results when

the numerical correction factor is used. In fact the results are more acceptable

when numerical correction factor is used in comparison with bulge correction

factor. The main reason of this difference is that the bulge correction factor does

not consider parameters such as geometry, material properties and friction factor.

All of these parameters can be changed by temperature variation.

7- In both tension and compression tests, at a distinctive temperature, stress

increases and elongation declines when strain rate goes up and at a distinctive

strain rate stress decreases and elongation rises when temperature goes up. The

main reasons of these changes are chiefly recrystallizations and slip mechanism

which exist in magnesium alloys during the hot deformations. The recrystallized

grains of alloy have no stress at first so it leads to decrease of alloy strength. Two

parameters which are effective in recrystallization phenomenon are time and

temperature, when temperature increases the number of grains which are

recrystallized rises. On the other hand when the strain rate goes up more grains

have enough time to be recrystallized. That’s why recristallization increases when

temperature rises and strain rate declines.

8- It is obvious that tension process of this alloy is a rate sensitive process like the

other alloys of magnesium but it can be found from tension curves that in 0250 C

and 275oC , rate sensitivity is almost equal in all strains but in higher

temperatures, rate sensitivity increases when strain goes up.

9- SEM micrographs of deformed AZ80 samples showed that the fracture of this

alloy is ductile. Fine voids and depth are apparent reasons of ductile

fracture.These voids form and centralize around the particles which are mostly

aluminum compounds and the main reason of this phenomenon is that the elastic

energy of these particles is higher than the surface energy of voids which are

formed.

References

[1] Caceres CH. Transient environmental effects of light alloy substitutions in transport vehicles.

Mater Des 2009; 30:2813 –22.

[2] Fereshteh-Saniee F, Barati F, Badnava H, Fallah Nejad KH. An exponential material model

for prediction of the flow curves of several AZ series magnesium alloys in tension and

compression. Mater Des 2012;35:1–11.

[3] Mathis K, Trojonava Z, lukac P. Hardening and softening in deformed magnesium alloys.

Mater Sci Eng A 2002;324:141–4.

[4] Trojanova Z, Kral R, Chatey A. Deformation behavior of an AJ50 magnesium alloy at

elevated temperatures. Mater Sci Eng A 2007;462:202–4.

[5] Sivapragash M, Lakshminarayanan PR, Karthikeyan R, Hanumantha M, Bhatt RR. Hot

deformation behavior of ZE41A magnesium alloy. Mater Des 2008;29:860–7.

[6] Wen TC, Na S, Lu W, Yong CZ, Chi WF, Hong C. Effect of temperature on mechanical

behavior of AZ31 magnesium alloy. Trans Nonferrous Met Soc China 2007;17:41–5.

[7] Abedi HR, Zarei-Hanzaki A, Fatemi-Varzaneh SM, Roostaei AA. The semi-solid tensile

deformation behavior of wrought AZ31 magnesium alloy. Mater Des 2010;31:4386–91.

[8] Masoudpanah SM, Mahmudi R, Roostaei AA. The microstructure, tensile, and shear

deformation behavior of an AZ31 magnesium alloy after extrusion and equal channel angular

pressing. Mater Des 2010;31:3512–6.

[9] Jayamathy M, Kailas SV, Kumar K, Seshan S, Srivatsan TS. The compressive deformation

and impact response of a magnesium alloy: influence of reinforcement. Mater. Sci. Eng. A 393

2005;393:27–35.

[10] Helis L, Okayasu K, Fukutomi H. Microstructure evolution and texture development during

high-temperature uniaxial compression of magnesium alloy AZ31 Mater. Sci. Eng. A 393

2006;430:98–103.

[11] Ponthot JP. Unified stress update algorithms for the numerical simulation of large

deformation elasto-plastic and elasto-viscoplastic processes. International Journal of Plasticity

2002;18:91–126.

[12] Song WQ, Beggs P, Easton M. Compressive strain-rate sensitivity of magnesium–aluminum

die casting alloys. Mater Des 2009;30:642–7.

[13] Palumbo G, Sorgente D, Tricarico L. A numerical and experimental investigation of AZ31

formability at elevated temperatures using a constant strain rate test. Mater Des 2010;31:1308–9.

[14] Anbuselvan S, Ramanathan S. Hot deformation and processing maps of extruded ZE41A

magnesium alloy. Mater Des 2010;31:2319–23.

[15] Raghu T, Sivakesavam O, Saha GG. Friction characteristics of magnesium alloy ZM-21. J

Mater Process Technol 1994;42:349–60.

[16] Narayanasamy R, Sathiyanarayanana S, Ponalagusamy R. A study on barrelling in

magnesium alloy solid cylinders during cold upset forming. J. of Mater. Proces. Tech

2000;101:64–9.

[17] Al-Samman T, Li X, Ghosh Chowdhury S. Orientation dependent slip and twinning during

compression and tension of strongly textured magnesium AZ31 alloy. Mater Sci Eng A 2010;527:

3450–3463.

[18] Ji Y H, Park J J, Ghosh Chowdhury S. Formability of magnesium AZ31 sheet in the

incremental forming at warm temperature. J Mater Process Technol 2008;201: 354–358.

[19] Fereshteh-Saniee F, Fatehi-Sichan F. An investigation on determination of flow curves at

room temperature and under forming conditions. J Mater Process Technol 2006;177:478–82.

[20] E. M. Mielnik, "Metalworking science and engineering"; McGraw Hill International Book

Company;1991.

[21] Sheng ZQ, Shivpuri R. Modeling flow stress of magnesium alloys at elevated temperature.

Mater Sci Eng A 2006;419:202–7.

[22] T.S. Srivatsan, Satish Vasudevan, M. Petraroli. The tensile deformation and fracture

behavior of a magnesium alloy. Journal of Alloys and Compounds 2008; 461:154-9.

List of figures

Fig. 1. Microstructure image of AZ80 alloy

Fig. 2. Schematic image of T shape compression test

Fig. 3. Change of crystallization in T shape test due to friction effect

Fig. 4. Calibration curves of AZ80 for various strain rates (a, b, c) at 250oC and (d, e, f) at 300oC

Fig. 5. Comparison between simulation and experiment results (a) at 250oC (b) at 300oC

Fig. 6. Schematic image of the test before and after applying compression

Fig. 7. Stress-strain curves of compression test (a) 250 C temperature (b) 300 C temperature

Fig. 8. Comparison between bulge correction factor, numerical correction factor and experiment results

Fig. 9. The effect of strain rate on correction factors (a, c) 250 C (b, d) 300 C

Fig. 10. Strain rate sensitivity (a) 10.0005S (b) 300 C .

Fig. 11. Schematic curves of true and engineering stress strain curves.

Fig. 12. Distribution of axial stress in necked section

Fig. 13. Stress strain curves of tension tests at various strain rates (a) 250oC , (b) 275oC , (c)300oC , (d)

325oC and (e) f

Fig.14. The changes of f at various temperatures and strain rates

Fig. 15. The changes of f at various temperatures and strain rates

Fig. 16. Simulated model at 300oC temperature and 10.005S strain rate (a) experiment specimen (b)

simulated specimen (c) force-displacement curve

Fig. 17. SEM images of fracture surface of AZ80 alloy at 350oC temperature and 10.0005S strain rate.

List of tables

Table 1: material composition of AZ80 alloy Table 2: The result of T shape tests

Fig. 1. Microstructure image of AZ80 alloy

Fig. 2. Schematic image of T shape compression test

Fig. 3. Change of crystallization in T shape test due to friction effect

Fig. 4. Calibration curves of AZ80 for various strain rates (a, b, c) at 250oC and (d, e, f) at 300oC

Fig. 5. Comparison between simulation and experiment results (a) at 250oC (b) at 300oC

Fig. 6. Schematic image of the test before and after applying compression

Fig. 7. Stress-strain curves of compression test (a) 250 C temperature (b) 300 C temperature

Fig. 8. Comparison between bulge correction factor, numerical correction factor and experiment results

Fig. 9. The effect of strain rate on correction factors (a, c) 250 C (b, d) 300 C

Fig. 10. Strain rate sensitivity (a) 10.0005S

(b) 300 C .

Fig. 11. Schematic curves of true and engineering stress strain curves.

Fig. 12. Distribution of axial stress in necked section

Fig. 13. Stress strain curves of tension tests at various strain rates (a) 250oC , (b) 275oC , (c) 300oC ,

(d) 325oC and (e) f

Fi.14. The changes of f at various temperatures and strain rates

Fig. 15. The changes of f at various temperatures and strain rates

Fig. 16. Simulated model at 300oC temperature and 10.005S

strain rate (a) experiment specimen (b)

simulated specimen (c) force-displacement curve

Fig. 17. SEM images of fracture surface of AZ80 alloy at 350oC temperature and

10.0005S strain rate.

Table 1: material composition of AZ80 alloy

Mg Al Zn Mn Si Ni

Bal. 7.83 0.46 0.25 0.03 0.001

Table 2: The result of T shape tests

M Strain

rate s-1

Temperature

(ºC)

0.7 0.001

250 0.65 0.01

0.61 0.1

0.71 0.001

300 0.67 0.01

0.63 0.1

We investigated tension and compression behavior of AZ80 alloy at various

temperatures and strain rates.

Bulge and numerical correction factor are used to correct the results of

compression tests.

Bridgman correction factor is used to correct the tension tests results.

T shape method is used to investigation of temperature and strain rate effects on

friction factor.