Home Physical Sciences Mathematical modeling of AZ30 magnesium alloys at high temperature using the ring compression test and genetic algorithm method
Article Open Access

Mathematical modeling of AZ30 magnesium alloys at high temperature using the ring compression test and genetic algorithm method

  • Farzan Barati EMAIL logo , Mona Esfandiari , Sajjad Babaei , Zahra Hoseini-Tabar and Aida Atarod
Published/Copyright: September 30, 2021

Abstract

In the present investigation, a new method is proposed to study the AZ30 flow curve at elevated temperatures and various strain rate. Experiments were carried out with the goal of obtaining flow curve of AZ30 at three different temperature and strain rates by using the ring test method. The presented work aims to develop a model using genetic algorithm for AZ30 flow stress prediction during different test conditions. The Santam machine was implicated that was able to perform experiments by controlling both the position and load modes. At each temperature and strain rate the isothermal test was performed respectively. In the present investigation for three varios temperatures and strain rates, 54 ring compression tests were carried out with different levels of reduction in height. Then each specimen was water cooled quickly to investigate the microstructure of AZ30 magnesium alloy by using optical microscope. The model used in the present study was able to predict the flow curve with an 2.3% accuracy. This model has excellent potential to be employed in various industry applications.

1 Introduction

Because of the high ratio of strength to weight and proper heat dissipation, damping, applications of Mg and its alloys have been used extensively at several important industries. Magnesium has very low ductility property at room temperature because of its HCP structure and low stacking fault energy [1, 2, 3, 4]. Magnesium alloys are the lightest structural alloy in nature, therefore these alloys are used widely at mobile device cases. Analyzing the hot deformation behavior of these alloys needs describing the mechanical behavior changes under an exerted load at each forming condition. This important issue should be expressed in terms of a constitutive equation which can relate stress state and strain state to their related conditions. At this time, the amount of as cast magnesium products is definitely more than wrought product amounts. However, wrought magnesium alloys may have more development potential with the higher strength, improved ductility and other mechanical property advantages. Due to very bad formability and negligible ductility at room temperature, the elevated forming technology should be employed [5, 6, 7, 8, 9]. Jia Zhou et al. investigate the compressive behavior of at the temperature range of 573 K to 673 K and led to increase of grain size with the increasing the strain rate and decreasing of temperature. They also showed that evolution of the grain size is more influenced by the temperatures rather than the strain rate [11].

In the case of magnesium and its related alloys investigations, very little works were performed to predict the flow stress considering it's forming condition. Only few research teams studied the mechanical behavior of magnesium alloys to hot deformation using constitutive equations [12].

In the present study, The hot compression ring test was employed for the as-cast AZ30 magnesium alloys, then by employing finite-element analyses numerical calibration curves and numerical sigmoid curves were obtained. By using these numerical curves flow curve of as-cast AZ30 magnesium alloy was obtained. For the AZ30 floe stress prediction, a two part model using Zener-Hollomon was introduced. This model can predict flow stress of magnesium alloys precisely.

2 Methodology

2.1 Material

The studied material was as-cast AZ30. The chemical composition used in the present work had the following (wt.%): Al = 3.02%, Zn = 0.45%, Mn = 0.15%, Si = 0.18% mg matrix.

2.2 Isothermal ring compression tests

The ring test can be used for determining the friction factor between die and the sample as well as flow behavior of the samples. The friction factor leads to increase or decrease in internal diameter based on amount of friction factor but external diameter increases for each value of friction factor. The ring test is capable of determining the flow curve as well as the friction factor at the same time. This advantages lead to low costs and time of experiments.

Isothermal ring compression tests were carried out implicating a 15 KK Santam machine that was equipped by an electrical furnace involving a thermocouple. Ring samples were machined precisely with an initial height (h0), inner diameter (di) and diameter (d0) of 3, 4.5 and 9 mm. The experimental samples were compressed more than 55% height reduction during isothermal conditions in the three different temperatures of from 473 to 513 K and strain rate range from of 0.025 s−1, 0.0025 s−1 to 0.00025 s−1. The ring compression tests were conducted under dry interfacial condition. To find out the stress curve of as-cast AZ 30 magnesium alloys, samples were compressed at 6 different amount of height level decrease. Finally, at least six different points were obtained for the stress curve at each test conditions. The 150 KN Santam machine is illustrated in the Figure 1. The analytical calibration curve as well as the numerical one should be calculated precisely. The analytical calibration curve is shown at Figure 2 [13].

Figure 1 The 150 KN Santam testing machine used in this research.
Figure 1

The 150 KN Santam testing machine used in this research.

Figure 2 The calibration curve for calculating the f [13].
Figure 2

The calibration curve for calculating the f [13].

To determine the flow stress, various numerical sigmoid curves should be obtained by implicating the finite-element method procedure. The numerical calibration curve should be provided for measuring the friction factor (f). The results indicate that the flow curves of as-cast AZ30 indicates DRX phenomenon that lead to increase the flow stress to a peak, and then decreases sharply to a steady state.

The initial sigmoid curves were achieved in analytical form as shown at Figure 3 and any effects of mechanical properties were not considered on the analytical sigmoid curve shapes. But material property affects on sigmoid curve, therefore numerical sigmoid curve were obtained using finite-element analysis. FE simulations were shown in Figure 3 [13]. Each specimen after deformation was water quenched, then polished carefully and etched for investigation of microstructure of as-cast AZ30 magnesium alloys.

Figure 3 Analytical sigmoid curves for measuring the compressive stress using the ring compression test [13].
Figure 3

Analytical sigmoid curves for measuring the compressive stress using the ring compression test [13].

Figure 4 Determining the friction factor as well as the flow curve of the presented model.
Figure 4

Determining the friction factor as well as the flow curve of the presented model.

3 Results and discussions

3.1 Flow stress curve of AZ30 magnesium alloy

The AZ30 specimens with do, di, and h0 of 9mm, 4.5mm and 3mm were examined at temperature range of 473 to 493K and strain rate ranging from 0.025 s−1 to 0.00025 s−1. Isothermal ring test were conducted for all of the samples. At each deformation condition the ring compression test carried out at all the different height reductions using hydraulic testing machine equipped with the thermocouple. During the experiments, The thermocouple is connected to the specimen at each time in order to obtain testing temperature very carefully. After using numerical calibration curves obtained from finite-element procedure according Figure 3 at each testing temperature and strain rates, friction factors were calculated. Results of friction factor showed that by decreasing the strain rates and increasing the temperature, friction factor increased. This can be resulted of increasing the local welding between die and samples. The maximum amount of friction factor from the calibration curve was 0.36 at strain rate of 0.00025 s−1 and temperature of 513 K and the minimum amount of that was 0.27 that was observed at the strain rate of 0.025 s−1 and temperature of 473 K.

Results from flow stress shows that in the initial stage of the forming process in as-cast AZ30magnesium alloy, the stress sharply reached to its maximum amount approximately near strain about 0.2. This could be interpreted as the dominance of work hardening at this stage. After this stage the flow stress sharply decreases due to the dominance of DRX phenomenon. Finally, the flow stress exhibits steady-state region that could be as a result of the equilibrium between DRX and work hardening and the same results obtained by other researchers that verified the experimental results of this study [14, 15, 16].

3.2 Material model for as-cast AZ30 magnesium alloy

Manufacturing the products with proper mechanical, physical, and chemical properties can be achieved by using the implication of hot forming process. A lot of factors can affect on material flow such as strain, strain rate, temperature, and other forming conditions can affect the flow behavior of the materials. It is worthy to know that it is very difficult to understand their effects due to their very complicated interactions with each other's.

As stated before, the flow stress of magnesium alloy increases sharply to its maximum amount because the hot deformation properties for magnesium alloys. After this stage, the stress state decreases with the same rate after peak stress due to DRX phenomenon. The steady stress state is the final step. Both strain rate and temperature affect on flow stress.

There is a specific relationship among flow stress, strain rate, and temperature that was introduced by the Zener-Hollomon parameter. This parameter can successfully express the interaction between these parameters to each other.

from ring compression test result, the material constants can be prperly determined. The Zener–Hollomon parameter can be experessed as bellow [17, 18].

Z=ε˙exp(QdRT)

Where Q, R, and T are the deformation activation energy, the universal gas constant, and the deformation temperature respectively, The activation energy can be calculated as below:

Q=RnTp

Tp is known as the slope of Ln(sinh(ασ)) against 1000/T at the same strain and strain rate and n is known as the slope of Ln(sinh(ασ)) against Ln(ε˙) when the deformation temperature is considered to be constant. These slopes are shown at Figure 5 and Figure 6 at strain equal 0.2 and different forming condition.

Figure 5 Plotting Ln(sinh(ασ)) against 1000/Tp at the same strain rates and different temperatures and strain = 0.2.
Figure 5

Plotting Ln(sinh(ασ)) against 1000/Tp at the same strain rates and different temperatures and strain = 0.2.

Figure 6 Plotting Ln(sinh(ασ)) versus 



σ=kεmsinh−1[ (A1z)n ]{ 1+exp( −A2δ(ε−εp)−exp[−A3δ(ε−εp)} ] 
\sigma  = k{\varepsilon ^m}\, {\sinh }^{ - 1} \left[ {{{\left( {{A_1}z} \right)}^n}} \right]\left\{ {1 + \exp \left( { - {A_2}\delta \left( {\varepsilon  - {\varepsilon _p}} \right) - \exp [ - {A_3}\delta \left( {\varepsilon  - {\varepsilon _p}} \right)\} } \right]} \right.


 at constant temperature and different strain rates condition at the maximum amount of strain related to stress.
Figure 6

Plotting Ln(sinh(ασ)) versus σ=kεmsinh1[ (A1z)n ]{ 1+exp( A2δ(εεp)exp[A3δ(εεp)} ] at constant temperature and different strain rates condition at the maximum amount of strain related to stress.

Figure 7 Predicted model and experimental data at temperature of 473 K and different forming rates.
Figure 7

Predicted model and experimental data at temperature of 473 K and different forming rates.

For prediction the compressive flow curve of AZ30 magnesium alloys a two part material was introduced by equation 3.

δ(εεp)=εεpforε>εpδ(εεp)=0forε<εp
Ln(ε˙)

Where ɛp is the strain corresponding with the maximum amount of stress and A1, A2, A3 and n are material constants. By using genetic algorithm method these coefficients can be determined as shown in Table 1. As shown at Figure 8–9 the output of this new model results show excellent agreements with the experimental ones. The average error is about 2% and at the worst state the maximum difference between predicted data and experimental data exceed only 5%.

Table 1

Coefficients predicted model for as-cast AZ30 magnesium alloys obtained from genetic algorithm.

K(Mpa) m A1 N A2 A3 Qoptimum (kJ/mol)
98.7 0.44 3.2×10−12 0.32 2.408 2.2×10−1 165
Figure 8 Predicted and experimental data at temperature of 493 K and different forming rateconditions.
Figure 8

Predicted and experimental data at temperature of 493 K and different forming rateconditions.

Figure 9 Predicted and experimental data at temperature of 493 K and different strain rates.
Figure 9

Predicted and experimental data at temperature of 493 K and different strain rates.

3.3 Microstructure observation

The samples were water quenched quickly for microstructural study. Then samples were polished very precisely to be prepared for etching process. The etching decomposition was the combination of picric acid, ethanol, and pure water. After etching the specimens an optical microscope was implicated. As shown at Figure 10, the AZ30 magnesium with average grain size of 64.7 μm are encompassed by the grain boundary phase chain because solidification process occurs and Mg17Al12 phase is visible clearly. Figure 11 shows the AZ30 microstructure at T equals 493K with a minimum forming rate of at a strain of 0.3 and the mean grain size is about 14.6 μm. Observations were conducted on the cross-sections compression plane.

Figure 10 Optical microscope image of as-cast AZ30 after casting at room temperature.
Figure 10

Optical microscope image of as-cast AZ30 after casting at room temperature.

Figure 11 Optical microscope image of as-cast AZ30 at strain of 0.3 and temperature of 493 K and forming rate of 0.00025 s−1.
Figure 11

Optical microscope image of as-cast AZ30 at strain of 0.3 and temperature of 493 K and forming rate of 0.00025 s−1.

Figure 12 shows the microstructure o deformed AZ30 magnesium alloy at a strain of 0.3, 493 K and the maximum studied deformation rate and the mean grain size is nearly 8.7 μm. By comparing Figures 11 and 12 it is obvious that the increasing of the forming rate leads to the average grain size decreases.

Figure 12 Optical microscope image of as-cast AZ30 magnesium alloy at strain of 0.3 and temperature of 493 K and strain rate of 0.025 s−1.
Figure 12

Optical microscope image of as-cast AZ30 magnesium alloy at strain of 0.3 and temperature of 493 K and strain rate of 0.025 s−1.

4 Conclusions

The present study aimed to identify the forming rates and forming temperature effects on flow curve of AZ 30 magnesium by employing the ring compression tests. Also a new model was introduced for predicting of the flow curves. In addition, the microstructural study of AZ30 alloy were studied by implicating the optical microscope and the following interesting results were obtained:

  1. Increasing the forming temperature as well as decreasing the forming rate leads to increase the amount friction factor coefficient due to the local welding between the die and the specimen.

  2. By increasing the temperature flow stress decreases but by increasing the strain rate the flow stress increases.

  3. A two-part model presented in this research work can predict the flow stress curve very precisely. The results were validated by the experiments and very proper agreement between these results were seen. The maximum amount of error was about 5 percent but the mean average error was about 2 percent.

  4. Increasing the strain rate led to increasing the average grain size.

  5. After peak stress dynamic recrystalization and increasing the number of twins were observed clearly.

  6. The maximum amount of stress was seen at the strain near 0.2.

  7. AZ30 grains were encompassed by the grain boundary phase chain Mg17Al12 because of the occurrence of the solidification process

  1. Funding information:

    The authors state no funding involved.

  2. Author contributions:

    All authors have accepted responsibility for the entire content of this manuscript and approved its submission.

  3. Conflict of interest:

    The authors state no conflict of interest.

References

[1] Zhao Z, Chen Q, Chao H, Huang S. Microstructural evolution and tensile mechanical properties of thixoforged ZK60-Y magnesium alloys produced by two different routes, Mater Design. 2010;(30):1906–16.10.1016/j.matdes.2009.10.056Search in Google Scholar

[2] Anbuselvan S, Ramanathan S. Hot deformation and processing maps of extruded ZE41A magnesium alloy. Materials Design. 2010;(30):2309–23.10.1016/j.matdes.2009.12.008Search in Google Scholar

[3] Foley DC, Al-Maharbi M, Hartwig KT, Karaman I, Kecskesc LJ, Mathaudhu SN. Grain refinement vs. crystallographic texture: Mechanical anisotropy in a magnesium alloy. Scripta Mater. 2011;64:193–6.10.1016/j.scriptamat.2010.09.042Search in Google Scholar

[4] Palumbo G, Sorgente D, Tricarico L. A numerical and experimental investigation of AZ30 formability at elevated temperatures using a constant strain rate test. Mater Design. 2010;(30):1308–16.10.1016/j.matdes.2009.09.009Search in Google Scholar

[5] Zhu T, Chen ZW, Gao W. Dissolution of Eutectic b-Mg17Al12 Phase in Magnesium AZ91 Cast Alloy at Temperatures Close to Eutectic Temperature. J Mater Eng Perfor. 2010;(19):860–7.10.1007/s11665-009-9539-ySearch in Google Scholar

[6] Ming-long M, Xing-gang L, Yong-jun L, Lan-qiang H, Kuil Z, Xian-wen W, Li-fang C. Establishment and application of flow stress models of Mg-Y-MM-Zr alloy. Nonferrous Met. Soc. China. 2011;(21):857–62.10.1016/S1003-6326(11)60793-0Search in Google Scholar

[7] Liu Y, Wu X. A Microstructure Study on an AZ30 Magnesium Alloy Tube after Hot Metal Gas Forming Process. Journal of Materials Engineering and Performance. 2007; (16):354–9.10.1007/s11665-007-9062-ySearch in Google Scholar

[8] Guedes CF, José MA, De Sá C. A proposal to deal with contact and friction by blending mesh free and finite element methods in forming processes, Int J Mater Form. 2008;(1):177–88.10.1007/s12289-008-0382-8Search in Google Scholar

[9] Lin XZ, Chen DL. Strain Hardening and Strain-Rate Sensitivity of an Extruded Magnesium Alloy. J Mater Eng Perfor. 2008;17:894–901.10.1007/s11665-008-9247-zSearch in Google Scholar

[10] Abu-Farha FK, Khraisheh MK, Mechanical Characteristics of Superplastic Deformation of AZ30Magnesium Alloy. Journal of Materials Engineering and Performance., 2007; (16): 192–9.10.1007/s11665-007-9031-5Search in Google Scholar

[11] Zhou J, Li L, Liu B, Liu Y. Deformation Microstructures of Mg-3Al-1Zn Magnesium Alloy Compressed Over Wide Regions of Temperature and Strain Rate, Journal of Materials Engineering and Performance. 2011 (20): 133–8.10.1007/s11665-010-9646-9Search in Google Scholar

[12] Fereshteh-Saniee F, Barati F, Badnava H, Falah-nejad K. . J Mater Design. 2012;(72):1–11.10.1016/j.matdes.2011.08.036Search in Google Scholar

[13] Mielnik EM. Metalworking Science and Engineering. McGraw – Hill, Inc. 1991:177–98.Search in Google Scholar

[14] Yanga YQ, Li BC, Zhang ZM. Flow stress of wrought magnesium alloys during hot compression deformation at medium and high temperatures. Mater Sci Eng A. 2009;(499):238–41.10.1016/j.msea.2007.11.106Search in Google Scholar

[15] Cheng-wen T, Shan-na X, Lu W, Zhi-yong C, Fu-chi W, Hong-nian C et al. Deformation behavior of AZ30 magnesium alloy at different strain rates and temperatures. Trans. Nonferrous Met. Soc. China. 2007;(17):347–352.Search in Google Scholar

[16] Ti S, Kun Y, Wen-xian L, Ri-chu W, Xiao-yan W, Zhi-yong C. Hot-compression constitutive relation of as-cast AZ30 magnesium alloy. Nonferrous Met. Soc. China, 2007; (17): 336–41.Search in Google Scholar

[17] Helis L, Okayasu K, Fukutomi H. Microstructure evolution and texture development during high-temperature uniaxial compression of magnesium alloy AZ30. Mater Sci Eng A. 2006;430:98–103.10.1016/j.msea.2006.04.125Search in Google Scholar

[18] Maksouda IA, Ahmed H, Rodel J. Investigation of the effect of strain rate and temperature on the deformability and microstructure evolution of AZ30 magnesium alloy. Mater Sci Eng A. 2009;504(30):40–8.10.1016/j.msea.2008.10.033Search in Google Scholar

Received: 2021-05-14
Accepted: 2021-07-05
Published Online: 2021-09-30

© 2021 Farzan Barati et al., published by De Gruyter

This work is licensed under the Creative Commons Attribution 4.0 International License.

Articles in the same Issue

  1. Research Articles
  2. The elastic-plastic properties of an anti-icing coating on an aluminum alloy: Experimental and numerical approach
  3. Optimization of recycled slag-fresh flux mixture based upon weld bead quality for submerged arc welding of stainless steel
  4. Design and optimization of differential capacitive micro accelerometer for vibration measurement
  5. Mechanical performance of abrasive sandpaper made with palm kernel shells and coconut shells
  6. Experimental investigation of WEDM process through integrated desirability and machine learning technique on implant material
  7. Mechanical properties and microstructural characteristics of rotating arc-gas metal arc welded carbon steel joints
  8. Assessment of cement replacement with fine recycled rubber particles in sustainable cementitious composites
  9. Structural response and sensitivity analysis of granular and asphaltic overlayment track considering linear viscoelastic behavior of asphalt
  10. Unmanned aerial vehicle evasion manoeuvres from enemy aircraft attack
  11. Effect of corrosion on surface degradation of galvanized steel in poultry dung, pig dung and urea solutions using rice straw as an inhibitor
  12. Mathematical modeling of AZ30 magnesium alloys at high temperature using the ring compression test and genetic algorithm method
  13. Study on hot deformation behavior and workability of stir-cast Al6063-6wt.% steelp based composites
  14. The effects of processing parameters on the formation of oxide layers in aluminium alloys using plasma electrolytic oxidation technique
  15. Behavior of green reactive powder mortar reinforced with steel fibers
  16. On the hygrothermal properties of sandcrete blocks produced with sawdust as partial replacement of sand
  17. Mechanical behavior of thin-walled steel under hard contact with rigid seabed rock: Theoretical contact approach and nonlinear FE calculation
  18. Mechanical properties and microstructural characteristics of rotary friction welded dissimilar joints of rolled homogeneous armor steel and medium carbon steel
  19. Studies of carboxylated nitrile butadiene rubber/butyl reclaimed rubber (XNBR/BRR) blends for shoe soles application
  20. Mechanical properties of wire arc additive manufactured carbon steel cylindrical component made by gas metal arc welding process
  21. Synthesis and mechanical characterization of Si3N4 reinforced copper-tin matrix composites
  22. Analysis of plated-hull structure strength against hydrostatic and hydrodynamic loads: A case study of 600 TEU container ships
  23. Mechanical performance investigation of lignocellulosic coconut and pomegranate / LDPE biocomposite green materials
  24. Special Issue MICAP-2021
  25. Double hydrothermal synthesis of iron oxide/silver oxide nanocomposites with antibacterial activity**
  26. Enhanced photocatalytic activity of TiO2-CdS composite nanofibers under sunlight irradiation**
  27. Structural properties of CoxCu1−xFe2O4 solid solution**
  28. Green-synthesis of Ag2O nanoparticles for antimicrobial assays**
  29. Effect of current density on the porous silicon preparation as gas sensors**
  30. A mechanochemical preparation, properties and kinetic study of kaolin–N, P fertilizers for agricultural applications**
  31. Impact strength of surface treated SS316L wires reinforced PMMA**
  32. Computational studies on electronic and optical properties of dopamine derivatives structure: A DFT study**
  33. Multilayer coating effects on the thermal conductivity of tools using an electric furnace technique**
  34. The positron and mechanical parameters of a cold-worked aluminum alloy (3004) Using PALT, PADBT and HV**
  35. Effect of thermal annealing on the structural and optical properties of TiO2 nanostructures**
  36. Improvement of forging die life by failure mechanism analysis**
Downloaded on 23.1.2026 from https://www.degruyterbrill.com/document/doi/10.1515/jmbm-2021-0011/html
Scroll to top button