Microstructural Evolution and Mechanical Properties of 6061 Aluminum Alloy Processed with Equal Channel Angular Pressing (ECAP) and Multi-Axial Compression (MAC)

Abstract

In this study, two of the severe plastic deformation (SPD) techniques, equal channel angular pressing (ECAP) and multi-axial compression (MAC) have been successfully applied to the Al-6061 alloy and the development of microstructure, texture and mechanical properties have been studied. It was found that the MEM (MAC+ECAP+MAC) and MEE (MAC+ECAP+ECAP) can effectively refine grains and hinder the movement of dislocations. The mechanisms of continuous dynamic recrystallization operating during severe deformation are discussed in the detail. The tensile strength and the macro-hardness of alloy after MEM and MEE deformation was increased. Refined grain and dislocation aggregation were mainly dedicated to the improvement of alloy mechanical properties.

Fig. 2 The mold and schematic illustrations of ECAP and MAC.

1. Introduction

Aluminum alloys, like Al-6061, are the most widely used materials today in almost entire range of industries¹⁾ because they have excellent chemical, physical, mechanical and processing properties, such as low density, acceptable strength, good ductility, excellent oxidation resistance both at room and elevated temperatures, good formability.^2,3) At elevated temperatures, the Al-6061 alloy can withstand large degree of deformation, and therefore usually they are used to produce high performance products with fine microstructures. At present, about two-thirds of extruded aluminum materials are produced with 6000 series alloys.⁴⁾ These processed 6000 series aluminum alloys are widely used in vehicle industry, subway production, air conditioner production and even some engine parts production, such as vehicle body frames, some key components of trains.⁵⁾ It is well known that grain refinement process is a skill to control and regulate mechanical properties of polycrystalline metallic materials. Severe plastic deformation (SPD) techniques is a well-known approach which can effectively modify grain dimension and shape of different materials.^6,7) SPD has been achieved through ECAP (Equal Channel Angular Pressing),⁸⁾ ARB (Accumulative Roll Bonding),⁹⁾ HPT (High Pressure Torsion),¹⁰⁾ and MAC (Multi Axial Compression), also widely known as Multi-Directional Forging (MDF).¹¹⁾ From practical point of view, it has been widely accepted that using SPD to bulk billets appears to be the most promising processing route for the manufacture of nanostructure Al alloys.²⁾ Among those SPD processes ECAP has been developed most intensively and is almost the most effective way to produce large quantity materials with ultra-fine grains (UFG).¹²⁾ ECAP is a processing technique whereby an intense plastic strain is imposed on the sample by repeatedly pressing the sample in a special die which consists of two channels of equal cross-section. The two channels intersect at an angle ϕ ranging from 90° to 157°.¹³⁾ After ECAP processing, significant grain refinement together with increased dislocations lead to a significant enhancement in the strength of the alloys. Therefore, a large number of studies on the development of UFG materials on Al alloy via ECAP have been published recently.^14,15) Ehab A et al.¹⁶⁾ reported that after ECAP, the grains of AA1050, 5083, 6082 and 7010 alloys were much refined. M. Suresh et al.¹⁷⁾ revealed that ECAP could increase the hardness, yield strength, and ultimate tensile strength of AA2195 alloy with the increase of ECAP passes. It was found that dislocation strengthening,¹⁸⁾ fine grain strengthening¹⁹⁾ and precipitation enhancement^20,21) can effectively work together to the strengthening of the materials underwent ECAP. MAC is another attractive approach to obtain fine-grained bulk alloys due to the simplicity of the procedure.^22–24) For example, Mohammad Reza Jandaghi et al.²⁵⁾ applied MAC process to AA5056 Aluminum alloy. Their work revealed that a grain refinement and strength enhancement of the alloy was realized. Cleber Granato de Faria et al.²⁶⁾ combined ECAP and MAC to investigate 1070 aluminum alloy, and revealed that with aluminum alloy which experienced 1 pass of ECAP, underwent MAC the aluminum alloy could be softened and became much more ductile. The above findings showed both ECAP and MDF can effectively refine alloy grains and improve alloy mechanical properties. In practice point of view, ECAP and MDF do not change the dimensions of the initial material, and thus are used as large quantities production of UFG.²⁷⁾ However, the research about the combining effect on materials microstructural modification of ECAP and MDF are barely reported. Especially, no details of the microstructural evolution and mechanical properties were report by now. Therefore, the present work provides the microstructure evolution and mechanical properties of Al 6061 alloy after processed with ECAP and MAC.

2. Experiment Procedures

A commercial 6061 aluminum alloy (Al–Mg–Si alloy) was used in this work. The chemical composition of the 6061 aluminum alloy is listed in Table 1. The dimensions of the used cubic block samples are 10 × 10 × 20 mm³. Figure 1 displays the average grain size of Al-6061 alloy after annealing before deformation. The cubic samples were firstly heated to 350°C and kept for on that temperature 4 h. Then ECAP and MAC were performed. The experiments were carried out at room temperature. Figure 2 shows the mold structures of ECAP and MAC. The experiment was carried out at room temperature. The extrusion speed is 35 mm/min. The single strain of MAC is 20%. The mold was lubricated with a mixture of molybdenum disulfide and engine oil to reduce the large friction generated by the ECAP and MAC processes. In this research, ECAP and MAC were combined by different methods. The routes are MAC+ECAP+ECAP (MEE) and MAC+ECAP+MAC (MEM), respectively. During the ECAP process, the route was B_C (90° rotation clockwise after each pass), which is the most efficient processing route for grain refinement reported by many researchers.²⁸⁾ During the MAC process, after the initial forging the sample is rotated by 90° and then placed in a mold for compression, so that the X, Y, and Z sides all have certain strains.

Table 1 Chemical composition (in mass%) of the Al-6061 alloy.

Fig. 1

The microstructure of annealed and undeformed Al-6061 alloy.

Fig. 2

The mold and schematic illustrations of ECAP and MAC.

For the study of the phase composition and the microstructure in the Al 6061 alloys, The XRD experiments were performed on D/max-2200 diffraction instrument manufactured by Rigaku Company, tested by CuKα with wavelength λ = 0.1514416 nm. In this study, we choose the c plane of the samples for XRD measurement and microstructure observations. The grain structure of the Al-6061 alloy was studied by scanning electron microscopy (SEM) of VEGA 3 SBH manufactured by TE-SCAN Corporation. Electron back-scattered diffraction (EBSD) was used for the determination of the average grain size. The HKL CHANNELS software was used to performed EBSD data visualization and post processing. The EBSD experiments were performed in the NOVA Nano SEM450 manufactured by FEI. The TEM test was performed on the FEI tencnai tf30. The experiments and mechanical property tests were performed in the SHIMADZU tensile testing machine. The tensile test direction was perpendicular to final deformation direction. The Vickers hardness (HV) was determined after extrusion, ECAP and MAC process. The hardness of the samples was determined as an average of at least eight measurements taken from left to right.

3. Results and Discussion

3.1 Phase composition of the samples processed by ECAP and MDF

The phase composition for the extruded and compressed Al-6061 alloy was studied by XRD. Figure 3 shows the X-ray diffraction patterns obtained from the Al-6061 alloy processed by ECAP, MAC, MEM and MEE. The constituting phases and the relative amount of these phases could be distinguished by comparing the ratio of the intensities. We can clearly observe that Mg₂Si showed up in all deformation routes, which is consistent with the reports of Nejadseyfi²⁹⁾ et al. It can be concluded that the combining of ECAP and MAC did not change the precipitates in Al-6061. Moreover, the diffraction intensities of (111) and (200) were relatively high in all deformation routes. We attribute this phenomenon to the occurrence of slip. As the deformation continues, slip deformation process will occur and the crystal slip system begin to operate, leading to the change of (111) and (200) peaks. More detailed content will be exhibited in the follow-up.

Fig. 3

XRD patterns of the Al-6061 alloy processed by (a) ECAP (b) MAC (c) MEM (d) MEE.

3.2 Effect of ECAP and MAC on the grain size

The EBSD orientation map in Fig. 4 shows the microstructure of the Al-6061 alloy processed by ECAP, MAC, MEM and MEE. The different colors represent different grain size in the diagram. The blue indicates the smallest grains while the red represents the largest grains. The average grain sizes of alloy after deformation are listed in Fig. 5. The grain size obtained from single ECAP and MAC was 1.45 µm and 1.67 µm. After MEM and MEE process, the grain size was 0.85 µm and 0.96 µm. From the above, we can conclude that MEM and MEE process can effectively refine grains. The grains of the sample processed by MEM and MEE were much finer and more uniform than single ECAP and MAC.

Fig. 4

The grain size of Al-6061 alloy processed by (a) 3 passes of ECAP (b) 3 passes of MAC (c) MEM (d) MEE.

Fig. 5

The grain size chart.

Low angle grain boundaries (LAGB) and High angle grain boundaries (HAGB) will make a contribution in material strengthening.^30,31) Latest research reported that LAGB can either act as dislocation nucleation site or as “tunable” barrier that allows selective transmission of dislocations, similar to the behavior of HAGB.³²⁾ In order to better analyze the structure evolution and mechanical properties of alloy, the samples were tested by EBSD. Figure 6 displays the misorientation angle distribution profiles, respectively: (a and a1) ECAP, (b and b1) MAC, (c and c1) MEM, (d and d1) MEE. The green area represents the low-angle grain boundary and the black area represents the high-angle grain boundary. Figure 6(a) indicates that ECAP displaying a predominance of grain disorientions in the range of 2° to 5° (0.118). Comparing with the 3-passes ECAP, the LAGB of three passes of MAC increased (0.59). To the MEE and MEM process, the number of LAGB in MEE process was higher than MEM process.

Fig. 6

Grain misorientation distributions of Al-6061 alloy (a and a1) 3 passes of ECAP, (b and b1) 3 passes of MAC, (c and c1) MEM, (d and d1) MEE.

The grain boundary angle is closely related to dynamic recrystallization. To achieve more information on the DRX mechanisms of the alloy, the EBSD study is conducted. Figure 7 shows the recrystallization distribution of Al-6061 alloy for the conditions: (a and a1) ECAP, (b and b1) MAC, (c and c1) MEM, (d and d1) MEE. The blue represents recrystallized grains. In this area, grain re-nucleats and grows up.³³⁾ The red represents deformed grains. The grain in these areas has a tendency to be stretched. The substructured grains are represented by the yellow region. From the picture, it can be concluded that the number of recrystallized grains in 3 passes of ECAP was the largest. In the case of 3 passes of MAC, the number of the deformed grains was the biggest. Meantime, it should be noted that the number of substructured grains was the smallest through MAC process. It is worth mentioning that under MAC-processed condition, the sample undergoes repeated compression with change in the direction of applied strain (i.e., x → y → z → x …) at each step.^34,35) Considering various factors such as deformation temperature and strain rate, the energy for dynamic recrystallization supplied by MAC is not as much as ECAP. Therefore, the MAC process has the maximum deformed grains and minimum recrystallized grains. As shown in Fig. 7(b), banded arrays composed of new grains and dislocation boundaries with moderate and high misorientations were frequently found. The characteristics of the band, which are the microshear bands (MSB),³⁶⁾ are described in detail elsewhere.^37,38) By comparing Fig. 7(c) and (d), it’s easy to find that the amount of recrystallized grains in MEM process is higher than MEE process. Both of them have similar effect on dynamic recrystallization.

Fig. 7

Recrystallization distribution of Al-6061 alloy (a) 3 passes of ECAP (b) 3 passes of MAC (c) MEM (d) MEE.

In order to understand the relationship of crystal orientation, the Inverse Pole Figure (IPF) of ECAP, MAC, MEM and MEE is given in Fig. 8. After 3-passes ECAP, the grain orientation preferred to be ⟨111⟩ in the X direction. The similar phenomenon also found in MAC and MEE process. However, the grain orientation in the Y and Z direction is not obvious. By comparing the IPF of MEM and MEE process, it can be concluded that the grain orientations of MEM is more obvious than MEE. Such a phenomenon will have a considerable effect on mechanical properties.³⁹⁾

Fig. 8

IPF colour images of ECAP, MAC, MEM, MEE samples respectively.

From the above research, it’s easy to find that different arrangement of ECAP and MAC will lead to different microstructure. For further analysis the grain refinement and the change of secondary-phase particles in MEM and MEE process, TEM images, SAED patterns and EDS analysis of MEM are given in Fig. 9. After MEM process, the average grain of alloy decreased to 700 nm. The SAED patterns was a series of concentric rings of different radii and it demonstrated that Al-6061 alloy belongs to polycrystal. The EDS results of point 1 and 2 combined with XRD patterns (Fig. 3) indicate that the particle phase was still Mg₂Si phase. The process of MEM didn’t change the composition of second phase.

Fig. 9

(a) Bright field images and (b) corresponding SAED of Al-6061 alloy processed by MEM and (c) the corresponding EDS maps of point 1 and 2.

For better analysis of grain refinement mechanism, Fig. 10 shows the typical TEM micrographs of the MEM processed Al-6061 alloy, with special attention paid to the dislocation. The average diameter of Mg₂Si phase particles in MEM process are 240 nm. As shown in the Fig. 10(b), the sub-grain can be easily found. It is worth noting that a high density of dislocations are accumulated in these deformed regions. The rearrangement of these dislocations will cause dynamic recovery and DRX during MEM.⁴⁰⁾ It was already reported that these irregular dislocation tangles and the organized substructures will lead to accumulative action of the strengthening mechanisms, which have a significant strength enhancement in alloy.⁴¹⁾

Fig. 10

The TEM images showing the microstructures of Al-6061 alloy processed by MEM.

In order to compare with MEM process, the TEM images, SAED patterns and EDS analysis of MEE process are given in Fig. 11. Similar to the process of MEM, the alloy still belongs to polycrystal. It is worth noting that, unlike the MEM, the peak of Fe elements is relatively high in MEE process. The reason might be attributed to the β-FeSiAl₃ and (FeMn)₃Si₂Al₁₅ phases in Al-6061 alloy.⁴²⁾

Fig. 11

(a) Bright field images and (b) corresponding SAED of Al-6061 alloy processed by MEE and (c) the corresponding EDS maps of point 1 and 2.

Figure 12 shows the TEM images of the MEE process. It can be clearly seen that there is a lot of tangled dislocations are stacked together, similar to the MEM process. Meanwhile, in the figure, we note that the stripe patterns with small misorientations are parallel to the wall and separated by high-density dislocations. The deformed grains have a tendency to be elongated. As already confirmed in previous studies by several authors,^43,44) the extrusion process leads to the elongation of the equiaxed dendritic microstructure along the extrusion direction. During the process of MEE, the alloy experienced two extrusions, this may affect the shape of the grain to some extent. Comparing MEM and MEE process, it can be concluded that the commonality of both is that they can effectively refine grains, form subgrains and dislocation walls.

Fig. 12

The TEM images showing the microstructures of Al-6061 alloy processed by MEE.

3.3 Mechanical properties

The Vicker’s macro-hardness values of the Al-6061 alloy, subjecting to MEM and MEE, are displayed in Fig. 13. For a better contrast, the undeformed alloy and previous work⁴⁵⁾ are also given in Fig. 13. After the process of MEM and MEE, the macro-hardness increased severely, compared to the undeformed alloy (48 HV), which can be attributed to the grain size refinement. It is worth noting that the macro-hardness of MEM-ed alloy is higher than MEE-ed alloy. This phenomenon has been related to the different microstructural aspects already described in the previous section (Fig. 5).

Fig. 13

Macro-hardness of the alloy.

The tensile properties of Al-6061 alloy are showed in Fig. 14. The tensile strength of specimens which are annealed and undeformed is 144 MPa, the elongation is 16%. The tensile strength of MEM and MEE-processed specimens reached the 399.362 MPa and 386.325 MPa, respectively. It can be concluded that the process of MEM and MEE can significantly improve the strength of the Al-6061 alloy. Since the specimens have the same chemical composition, the observed differences in the mechanical properties are considered to be mainly due to the processing methods. As we described above in Fig. 10 and Fig. 12, the MEM and MEE can effectively refine the grains and hinder the movement of dislocation, which is very suitable to alloy strengthening.

Fig. 14

Tensile test curves for MEM, MEE and unreformed Al-6061 alloy, at the room temperature.

It was reported that the large number of voids on the fracture surface are often linked to the ductile failure.⁴⁶⁾ Further analysis of the tensile fracture will help at better understanding of the observed differences in strength between the MEM and MEE. As shown in Fig. 15, the dimple of MEE processed alloy is more obvious and homogeneous than MEM processed alloy, which indicate that the MEE processed alloy have a better plasticity. The results are consistent with the mechanical properties we obtained.

Fig. 15

SEM images of the tensile fracture of A1-6061 alloy processed by (a) MEM (b) MEE.

4. Conclusions

The microstructure and mechanical behavior of Al-6061 alloy processed by MEE and MEM have been examined. The following conclusions can be drawn:

Compared to ECAP and MAC, the newly-designed routes (MEM and MEE), have refined grains more effectively to the nanometer scale. The effect of grain refinement processed by MEM is the best. The recrystallized grains which weakened the preferred orientation of crystals are easy to be formed in MEM process. Moreover, grain orientations of MEM is more obvious than MEE.

The process of MEM and MEE can hinder the movement of dislocations. There are many dislocation walls in the MEM-ed and MEE-ed alloy. After MEE process, the deformed grains have a tendency to be elongated. This phenomenon indicates that the deformation route has an important influence on the grain shape of the alloy.

Due to the grain refinement, the macro-hardness of MEM-ed and MEE-ed alloy increase severely. The tensile strength of alloy is 2.5 times higher than undeformed one, but the elongation decreased a little. Comparing with tensile fracture of alloy, the MEE processed alloy have a better plasticity.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (grant numbers 51571086 and 51271073) and partial financial supports come from the Natural Science Foundation of Henan Polytechnic University (grant numbers B2010-20 and B2019-42).

REFERENCES

• 1)   A.S.  Khan and  C.S.  Meredith: Int. J. Plast. 26 (2010) 189–203.
• 2)   I.  Sabirov,  M.Y.  Murashkin and  R.Z.  Valiev: Mater. Sci. Eng. A 560 (2013) 1–24.
• 3)   H.M.  Flower: Int. Mater. Rev. 37 (1992) 196–196.
• 4)   A.J.  Knowles,  X.  Jiang,  M.  Galano and  F.  Audebert: J. Alloy. Compd. 615 (2014) S401–S405.
• 5)   R.E.  Sanders: JOM 53(2) (2001) 21–25.
• 6)   K.H.  Yang: J. Plast. Eng. 58 (2006) 33–39.
• 7)   R.Z.  Valiev,  R.K.  Islamgaliev and  I.V.  Alexandrov: Prog. Mater. Sci. 45 (2000) 103–189.
• 8)   R.Z.  Valiev and  T.G.  Langdon: Prog. Mater. Sci. 51 (2006) 881–981.
• 9)   Y.  Saito,  N.  Tsuji,  H.  Utsunomiya,  T.  Sakai and  R.G.  Hong: Scr. Mater. 39 (1998) 1221–1227.
• 10)   A.P.  Zhilyaev and  T.G.  Langdon: Prog. Mater. Sci. 53 (2008) 893–979.
• 11)   C.G.  de Faria,  N.G.S.  Almeida,  M.T.P.  Aguilar and  P.R.  Cetlin: Mater. Lett. 174 (2016) 153–156.
• 12)   Y.G.  Jin,  H.M.  Baek,  K.H.  Sun,  Y.T.  Im and  B.C.  Jeon: J. Mater. Process. Technol. 212 (2012) 848–855.
• 13)   K.  Nakashima,  Z.  Horita,  M.  Nemoto and  T.G.  Langdon: Acta Mater. 46 (1998) 1589–1599.
• 14)   A.P.  Zhilyaev,  D.L.  Swisher,  K.  Oh-Ishi,  T.G.  Langdon and  T.R.  Mcnelley: Mater. Sci. Eng. A 429 (2006) 137–148.
• 15)   M.A.  Afifi,  Y.C.  Wang,  P.H.R.  Pereira,  Y.  Huang,  Y.  Wang,  X.  Cheng,  S.  Li and  T.G.  Langdon: J. Alloy. Compd. 769 (2018) 631–639.
• 16)   E.A.  El-Danaf: Mater. Des. 32 (2011) 3838–3853.
• 17)   M.  Suresh,  A.  Sharma,  A.  More,  R.  Kalsar,  A.  Bisht,  N.  Nayan and  S.  Suwas: J. Alloy. Compd. 785 (2019) 972–983.
• 18)   L.  Tang,  X.  Peng,  J.  Huang,  A.  Ma,  Y.  Deng and  G.  Xu: Mater. Sci. Eng. A 754 (2018) 295–308.
• 19)   S.R.  Kumar,  V.H.  Sankar,  S.  Arunvinthan and  S.  Nadarajapillai: Mater. Today Proceedings 5 (2018) 21482–21488.
• 20)   H.  Jia,  R.  Bjørge,  K.  Marthinsen,  R.H.  Mathiesen and  Y.  Li: Mater. Sci. Eng. A 703 (2017) 304–313.
• 21)   M.  Gazizov and  R.  Kaibyshev: Mater. Sci. Eng. A 588 (2013) 65–75.
• 22)   M.G.  Jiang,  H.  Yan and  R.S.  Chen: Mater. Des. 87 (2015) 891–900.
• 23)   A.  Heczel,  F.  Akbaripanah,  M.A.  Salevati,  R.  Mahmudi,  Á.  Vida and  J.  Gubicza: J. Alloy. Compd. 763 (2018) 629–637.
• 24)   K.B.  Nie,  K.K.  Deng,  X.J.  Wang,  T.  Wang and  K.  Wu: Mater. Charact. 124 (2017) 14–24.
• 25)   M.R.  Jandaghi,  H.  Pouraliakbar,  G.  Khalaj and  M.  Shirazi: Mater. Sci. Eng. A 657 (2016) 431–440.
• 26)   C.G.  de Faria,  N.G.  e Silva Almeida,  F.  de Castro Bubani,  K.  Balzuweit,  M.T.P.  Aguilar and  P.R.  Cetlin: Mater. Lett. 227 (2018) 149–153.
• 27)   T.  Tański,  P.  Snopiński,  K.  Prusik and  M.  Sroka: Mater. Charact. 133 (2017) 185–195.
• 28)   G.  Dyakonov,  S.  Mironov,  I.  Semenova,  R.  Valiev and  S.  Semiatin: Mater. Sci. Eng. A 701 (2017) 289–301.
• 29)   O.  Nejadseyfi,  A.  Shokuhfar,  A.  Dabiri and  A.  Azimi: J. Alloy. Compd. 648 (2015) 912–918.
• 30)  L. Priester: Grain Boundaries and Crystalline Plasticity, (Wiley, Hoboken, 2013) pp. 60–66.
• 31)   L.  Li,  Z.  Zhang,  J.  Tan,  C.  Jiang,  R.  Qu,  P.  Zhang,  J.  Yang and  Z.  Zhang: Sci. Rep. 5 (2015) 15631.
• 32)   S.  Chen and  Q.  Yu: Scr. Mater. 163 (2019) 148–151.
• 33)   X.  Zhang,  X.  Zhou,  J.-O.  Nilsson,  Z.  Dong and  C.  Cai: Corros. Sci. 144 (2018) 163–171.
• 34)   T.  Sakai,  H.  Miura,  A.  Goloborodko and  O.  Sitdikov: Acta Mater. 57 (2009) 153–162.
• 35)   O.  Sitdikov,  T.  Sakai,  A.  Goloborodko,  H.  Miura and  R.  Kaibyshev: Mater. Trans. 45 (2004) 2232–2238.
• 36)   O.  Sitdikov,  T.  Sakai,  A.  Goloborodko,  H.  Miura and  R.  Kaibyshev: Philos. Mag. 85 (2005) 1159–1175.
• 37)   C.  Kobayashi,  T.  Sakai,  A.  Belyakov and  H.  Miura: Philos. Mag. Lett. 87 (2007) 751–766.
• 38)   T.  Sakai,  A.  Belyakov and  H.  Miura: Metall. Mater. Trans. A 39 (2008) 2206–2214.
• 39)   Y.  Zhao,  M.N.  Polyakov,  M.  Mecklenburg,  M.E.  Kassner and  A.M.  Hodge: Scr. Mater. 89 (2014) 49–52.
• 40)   Y.  Zhou,  Z.  Chen,  J.  Ji and  Z.  Sun: J. Alloy. Compd. 770 (2019) 540–548.
• 41)   M.  Ciemiorek,  W.  Chrominski,  L.  Olejnik and  M.  Lewandowska: Mater. Des. 130 (2017) 392–402.
• 42)   H.  Wang,  W.  Sun and  Y.  Xing: Phys. Procedia 50 (2013) 68–75.
• 43)   M.  Hirano,  M.  Yamasaki,  K.  Hagihara,  K.  Higashida and  Y.  Kawamura: Mater. Trans. 51 (2010) 1640–1647.
• 44)   Y.M.  Zhu,  A.J.  Morton and  J.F.  Nie: Acta Mater. 58 (2010) 2936–2947.
• 45)   N.  Panwar and  A.  Chauhan: J. Mater. Res. Technol. 8 (2019) 660–669.
• 46)   J.  Wu,  Q.  Shi,  H.C.  Ma and  Y.L.  Chiu: Mater. Sci. Eng. A 669 (2016) 417–427.