- Email: [email protected]
B4C composite during compression at elevated temperature
Author’s Accepted Manuscript Hot deformation behavior and microstructural evolution of particulate-reinforced AA6061/B4C composite during compression at elevated temperature Kaikai Wang, Xiaopei Li, Qiulin Li, Guogang Shu, Guoyi Tang www.elsevier.com/locate/msea
PII: DOI: Reference:
S0921-5093(17)30304-0 http://dx.doi.org/10.1016/j.msea.2017.03.013 MSA34796
To appear in: Materials Science & Engineering A Received date: 27 October 2016 Revised date: 2 March 2017 Accepted date: 3 March 2017 Cite this article as: Kaikai Wang, Xiaopei Li, Qiulin Li, Guogang Shu and Guoyi Tang, Hot deformation behavior and microstructural evolution of particulatereinforced AA6061/B4C composite during compression at elevated temperature, Materials Science & Engineering A, http://dx.doi.org/10.1016/j.msea.2017.03.013 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 galley proof before it is published in its final citable 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.
Hot deformation behavior and microstructural evolution of particulatereinforced AA6061/B4C composite during compression at elevated temperature Kaikai Wanga,b, Xiaopei Lia, Qiulin Lia,b,1*, Guogang Shuc, Guoyi Tanga,* a
Advanced Materials Institute, Graduate School at Shenzhen, Tsinghua University, Shenzhen 518055, PR China b
Shenzhen Engineering Laboratory of Nuclear Materials and Service Safety, Shenzhen, 518055, PR China c
Joint Laboratory of Nuclear Materials and Service Safety, Shenzhen, 518055, PR China Abstract The flow stress behavior of AA6061/B4C composites has been researched by compression test using Gleeble-3800 thermal simulator, in the temperature ranging from 633-783K and strain rate ranging from 0.001-1s-1. Typical true stress-true strain curves showed that the peak stress levels decreased with the rising of temperature but increased with the rising of strain rates. The combined effects of temperature and strain rate on deformation were analyzed by constitutive equation which containing the Zener-Hollomon parameter (Z) in hyperbolic sine function. The effects of Z values on dynamic softening and associated microstructural evolution during hot deformation were investigated by electron back scattered diffraction technique (EBSD). It was found that with the decrease of Z values, local strain induced by deformation was released and the grain size of aluminum matrix increased gradually, which indicated that the main softening mechanism of **
Corresponding author: Tel/fax: +86 755 2603 6224 Corresponding author. Tel/fax: +86 755 2603 6752. E-mail address: [email protected] (G. Tang) 1
[email protected] (Q. Li)
AA6061/B4C composites was dynamic recrystallization (DRX), and the lower the Z value was, the easier the DRX occurred. Keywords Metal matrix composites; hot compression; flow stress behavior; constitutive equation; dynamic recrystallization 1. Introduction Aluminum alloy 6061 is widely used as a base matrix for Metal matrix composites (MMCs) due to its perfect machinability and good wear and mechanical properties. And boron carbide (B4C) is one of the promising reinforcement particle in Al matrix composites for its attractive properties, such as super high hardness, low density and good thermal and chemical stability, making the AA6061/B4C MMCs popular in automotive and aerospace industry, and especially in nuclear industry [1-3]. For instance, it is often used as a neutron absorption material for storage and transportation applications of spent fuel [4]. Traditionally, Al/ B4C MMCs are manufactured by powder metallurgy (P/M). However, P/M method is still far from being used in industrial application for its high cost and limited dimension of the finished product. Among all the manufacturing methods, liquid-stirring casting, an efficient and cost-saving method, which introduces reinforced particles into molten metal through a vortex, is ideal for industrial production of Al/ B4C MMCs and is employed in the present study [5-7]. Additionally, it is commonly accepted that AA6061/B4C composite often needs secondary processing such as hot extrusion and rolling before it is used as structural or functional material in industry. However, owing to the super high hardness of reinforcement B4C particles in the aluminum matrix, hot working and machinability of the AA6061/B4C composite is much harder than matrix aluminum alloy. So, in the light of the above analysis, the studies on flow stress behavior and the corresponding microstructure evolution during hot deformation process becomes rather important. In order to figure
out the deformation process of AA6061/B4C composites and develop a processing methods such as hot rolling and hot extrusion, it is necessary to study the flow stress behavior, deformation mechanism and microstructure evolution under different processing conditions [8, 9]. It is well known that some dynamic behaviors such as work hardening (WH), dynamic recrystallization (DRX) and dynamic recovery (DRV) often occur during the hot deformation of alloys or metallic composites, accompanied by mechanical properties and microstructure evolution of material [10-13]. Dynamic recrystallization (DRX) affects the microstructure evolution of matrix during hot deformation, which can influence the properties of the composites to some great extent [14]. Thus, it is necessary to study the DRX of Al matrix composite during hot deformation. For particle reinforced Al matrix composites, particles will significantly affect the recrystallization behavior of metal matrix [15]. Ganesan et al. [16] investigated the development of processing maps for 6061Al/15% SiCp composite material by hot compression test. they found the microstructure evaluation displays particle fracture, debonding, adiabatic shear band formation and matrix cracking which led to flow instability of the composites, and they also identified the super plastic deformation and dynamic recrystallization zones, corresponding to optimum working regions. Generally, DRX depends sensitively on the deformation temperature and strain rate. The combined effects of strain rate and deformation temperature on hot deformation and DRX behavior can be represented by a Zenerhollomon parameter (Z) [17-19]. In order to study the DRX behavior of AA6061/B4C through Z values, it is necessary to explore the relationship between Z values and the corresponding microstructures, which will provide valuable instructions for the control of structure and performance of AA6061/B4C in industrial manufacturing [20] Based on the above statement, in this paper, the flow stress behavior of AA6061/B4C is analyzed by typical true stress-true strain curves and constitutive equation with hot deformation activation energy. Microstructure evolution under different deformation
conditions is studied using EBSD. The relationship between Z values and dynamic softening mechanism is also discussed. 2. Experimental In this paper, AA6061/B4C composites containing 31 vol.% B4C particles were prepared by the stable liquid-stirring process, using commercially available pure aluminum, Al-Ti and Al-Si master alloy as raw materials, and B4C powders with grain sizes of 15-50 μm (23.2 μm, average grain size) as reinforcement. Stirring casting ingot B4C/6061, with a size of 65mm in diameter and 130mm in height, were homogenized at 698 K for 12 h and furnace cooled to room temperature. The ingots prepared for compression test were machined to cylindrical specimens with 10mm in diameter and 15mm in height. Compression tests were carried out on Gleeble-3800 thermal simulator at the temperatures of 633 K, 663 K, 693 K, 723 K, 753 K and 783 K, at the strain rate of 0.001 s-1, 0.01 s-1, 0.1 s-1 and 1 s-1 respectively. During compression tests, samples were heated to target temperature and homogenized for 3 minutes in order to obtain a stable and uniform condition for deformation. All samples were compressed to the true stain of -0.7 (55% reduction), and quenched in water immediately after compression. All deformed specimens for microstructural observations were sectioned along the compression axis direction, followed by mechanical grinding and ion beam cross section polisher (IB-09020CP, JOEL) successively. Electron back scattered diffraction (EBSD) analyses were performed using the scanning electron microscopy (SEM, TESCAN MIRA 3 LMH), step size between points on the scan grid was set to 1μm. 3. Results and discussion 3.1 Flow stress behavior Actually, the fabrication parameters and interaction between B4C and matrix have great effects on the mechanical of composites, and these effects are directly revealed in strain-stress curves.
Yu Li et al. [21] fabricated the Tri-modal Al-B4C-Al3Ti composites and investigated the effect of Ti content and stirring time on microstructure and mechanical behavior of Al- B4C composites. They researched the interface reaction between Al, B4C and Ti and found that, higher Ti levels resulted in higher volume fraction and particle size of Al3Ti denser TiB2 interface and even more uniform spatial distribution of B4C. A prolonged stirring time led to copious coarse TiB2 crystals at the expense of Al3Ti particles and more uniform particle spatial distribution was achieved. During plastic deformation, these coarse TiB2 crystals sticking out of B4C surface can carry extra applied load. As a result, their tensile strength showed an inclined leaf-shape curve. Jiyun Zheng et al [22] discussed the microstructure evolution and interface reaction of 15wt% B4C/Al composites with titanium addition during liquidstirring process. They found that The dispersion of B4C particles changes from clusters to discrete distribution as time increases, which attributes to improved wetting through interfacial reaction in addition to high speed stirring. They also pointed that interface reaction produced Al3BC, TiB2, and AlB2 by B4C erosion and Al3Ti decomposition; AlB2 only precipitated in matrix after long time stirring. The growth of TiB2 transformed from a fine layer to discretely coarse crystals on the B4C surface. Similar discussion of interaction in Al alloy/Ti2AlC composites system and detailed chemical mechanism was studied by Liangfa Hu et al [23], which could give some instruction for further investigation of B4C /Al composites system.
Hot deformation theory indicated that flow stress is the most basic parameter to characterize plastic deformation properties of metals and alloys. The stress determines the size of load and energy needed during plastic deformation of materials [24]. From compression test, a series of typical true stress-true strain curves of the AA6061/B4C composites at various deformation conditions are presented in Fig. 1. It can be easily found that the deformation temperatures and strain rates have obvious effects on the flow stress behavior. As shown in the flow stress curves, the true stress increases rapidly because of the work hardening with the increase of true strain, and then holds to a steady value or decreases slightly after the true stress reaches the peak value as a result of dynamic softening. Apart from this, Fig. 1 also shows that the peak values of stresses fluctuated at different deformation conditions. It decreases with the increase of deformation temperatures, while it increases with the increase of strain rates at a given temperature. Obviously, lower temperatures lead to more drastic work hardening, as the temperature increases, the work hardening is decreases. From the test conducted by B. L. Zhang et al [25], we also found the similar changing trend of flow stress in AA6061/SiCp MMCs, and they pointed that The flow stresses of the AA6061/SiCp composite were significantly higher than those of the unreinforced alloy at the lower deformation temperature. These curves in present work also reveal that the AA6061/B4C composite is a positive strain rate sensitive material. Current research shows that the deformation at elevated temperature is a competing process of work hardening and dynamic softening [26-29], and dynamic recovery and dynamic recrystallization are the two main softening mechanisms in hot deformation of particle reinforced Aluminum matrix composites. In the present study, the dynamic softening seems more distinguished at higher temperature and lower strain rate. At the beginning of the deformation,
the work hardening exceeds the dynamic softening, the dislocation density increased dramatically under the external force, leading to the rapid increase of true stress at the initial stage of deformation. As the compression proceeding, the strengthening effect of B4C particles is receded, dynamic softening occurs simultaneously and it counteracts or partially offsets the effect of work hardening, which makes the true stresses hold unchanged or decreased slightly after reaching the peak values. [30,31]. However, different mechanisms can result in similar flow stress behavior [32] for instance, flow softening can be attributed to DRX, temperature rise of the specimens and texture evolution, or grain size increase during the deformation process. It is difficult to define the accurate deformation mechanism only by the typical true stress-true strain curves, so comprehensive microstructure evolution analysis will be discussed to explain the specific process and softening mechanism of deformation. Besides of the above discussion, it is necessary for us to compare the particle reinforced composite deformation behavior with non-reinforced matrix alloy, which is helpful for us to figure out the effects of particles on deformation Actually, the flow stress for the composite is higher than that of the alloy, especially at the low deformation temperature, compared to the study conducted by H. R. Ezatpour et al [24]. They researched the work softening mechanisms and texture in 6061 aluminum alloy during hot deformation. Their experiment showed that the flow stress increased with increasing strain rate and decreasing deformation temperature and dynamic recovery and recrystallization only occurred at low Z values. And it has been explained by Humphreys et al [33]. During deformation, the strengthening of composites is caused by the generation of the vacancies and dislocations in the matrix around the particles, owing to incompatibility between the deforming matrix and the non-deformable particles, dislocations accumulate around the particles, resulting in a higher flow stress in the composite compared with the pure alloy.
Traditionally DRX is associated with hot working of high stacking fault energy (SFE) metals which exhibit flow softening after reaching a critical strain. Flow curves eventually reach a steady state at large true strain. As shown in Fig.1.a~d, at lower strain rates oscillations in the flow curves are observed after reaching a steady state. Some other Detailed analysis will be discussed in next section.
Fig.1.True stress-true stain curve of AA6061/B4C composites during hot compression deformation: (a) ε=1s-1; (b) ε=0.1s-1; (c) ε=0.01s-1; (d) ε=0.001s1
3.2 kinetic analysis 3.2.1 constitutive equation As for hot deformation behavior of metallic materials and metal matrix composites, it is commonly accepted that the relationship between the strain rate, the peak stress (or the steady stress) and deformation temperature can be described by the empirical function using Arrhenius type equation[34-36]as follows： A
(1)
where F(σ) is a function of applied stress with the following three possible forms： =
if
= =
if
(2)
for all values of σ
Where σ is the flow stress (Mpa) for a given strain, T is the absolute deformation temperature (K), R is the gas constant(8.314Jmol-1K-1), Q is the activation energy of deformation (Jmol-1) and A,β,α(=β/ n’), n, n’ are material constants. For the low stress level and high stress level, substituting suitable F(σ) function into Eq.(2) leads to Eq.(3)and Eq.(4), respectively [37,38]. ε=A1σn'
ασ<0.8 (3),
ε=A2eβσ
ασ>1.2 (4),
Where A1 and A2 are material constants. By taking the logarithm for both sides of Eq. (3) and Eq. (4) and substitute the peak stress, the relationship between the flow stress and strain rate (lnσlnε and σ-lnε) are shown in Fig.2(a) and Fig.2(b). The flow stress can be
estimated by a group of parallel straight lines in different hot compression conditions. The material constants n’ and β can be obtained from the slope of every single line in the lnσ-lnε and σ-lnε plots by linear regression fit method. The mean value of n’ and β can be calculated as 9.331801 and 0.158482 Mpa-1 respectively, then α=β/ n’=0.016983.
Fig2. Evaluating the value of (a).n’ by fitting lnσ-lnε, (b).β by fitting σ-lnε, (c).n by fitting ln[sinh(ασ)]-lnε, (d).Q by fitting ln[sinh(ασ)]-1/T
Eq. (1) can be expressed in terms of temperature compensated stain rate factor the Zener-Hollomon parameter Z, which is defined by Z=εeQ/RT=A[sinh(ασ)]n
(5),
Where Q is the hot deformation activation energy. Similarly, for all the stress level, the value of n in Eq. (4) can be derived from the slopes of lines representative of ln[sinh(ασ)]-lnε by taking the logarithm, n and Q can be calculated as 6.93861 and 151.634 kJmol-1, respectively. The details of all material constants evaluated in the present study are given in Table 1. Table 1 Material constants and Q of AA6061/B4C in the hyperbolic sine equation Q/kJmol-1 151.634
α/Mpa-1 0.016983
A/s-1 1.94658*109
n 6.93861
The constitutive equation can be described by hyperbolic sine equation as follow：
{
[
]
}
{( [(
) )
]
}
3.2.2 hot deformation activation energy Generally, the hot deformation activation energy can qualitatively reflect the energy barrier to dislocation motion for metals and alloys during hot deformation. A higher value of activation energy signifies the existence of higher dragging forces to the movement of the dislocation in hot deformation [39]. The hot working behavior of unreinforced Al6061 alloys has been extensively studied. Table 2 shows a comparison of Q values for 6061 alloys under different conditions and B4C /6061 composites from the present work and other investigation.
Table2 Comparison of activation energy Q values for different conditions of 6061 Al alloys and B4C /6061 composites Material B4C /6061（stirring casting, Hot extruded） 6061
Q/kJmol-1 152 344
Reference * [44]
6061 6061 * for present work
274 205
[24] [40]
Obviously, the activation energy of the 6061 alloy is larger than that of the AA6061/B4C composite（151kJmol-1）In the current compression test,. This may be attributed to the reinforced B4C particles. It will be discussed in more detail in Section 3.3.2.
Fig3.Relationship between ln[sinh(ασ)] and lnZ
3.3 microstructural evolution 3.3.1 Grain size
Fig 4 IPF map for initial homogenized AA6061/B4C composite Fig.4 is the inverse pole figure (IPF) map with grain boundaries for homogenized AA6061/B4C. It can be seen that the initial sample, which has been treated by homogenization, consist of uniform equiaxed grains with average diameter of 21.65μm. Many researches have reported that dynamic recrystallization occurs in aluminum alloy during hot deformation, and the dynamic recrystallization behavior can be described by Zener-hollomon parameter (Z), which corresponds to the flow behavior and microstructure after deformation [26,28,41]. Z parameter is also called as a temperature-compensate strain
rate, which could represent the combined effects of strain rate and temperature on hot deformation [42,43]. Z values obtained from Eq. (5) and average grain size of the specimens under different deformation conditions are shown in Table 3. Table 3 Z values and grain sizes under different deformation conditions lnZ1 lnZ2 lnZ3 lnZ4 lnZ5 lnZ6
Temperature/K Temperature 753 480 723 450 693 420 723 450 753 480 723 450
Strain rate 0.001 0.001 0.001 0.01 0.1 0.1
lnZ/s-1 17.3132 18.31822 19.41025 20.6208 21.91837 22.92339
Grain size/μm 14.29 13.38 10.9 10.28 9.99 7.98
The grain sizes correspond to different Z values are shown in Fig.5. As can be seen, the lower the Z is, the bigger the grain sizes are. DRX is a beneficial process in hot deformation since it only gives stable flow and good workability to the materials by simultaneously softening.
Fig.5 Average grain size under different Z values
Fig.6 the inverse pole figure (IPF) map with grain boundary and grain size distribution maps of specimens deformed under different Z values: (a) lnZ=17.3132; (b) lnZ=18.31822; (c) lnZ=19.41025; (d) lnZ=20.6208; (e) lnZ=21.71284; (f) lnZ=22.92339. Fig.6 shows the microstructure evolution of compressed samples. Microstructures are given in the form of inverse pole figure (IPF) map embed with grain size distribution map. as shown in Fig.6e and Fig.6f, the samples with higher Z value, a lot of fine grains can be observed while many large grains still remain. The large grains are non-recrystallized, most of the grains are newly formed nuclei with little grain size. As can be seen in the grain size distribution maps, both curves have only one peak around 5μm, indicating that the DRX does not finish in the extruded composites. H. R. Ezatpour et al [24] investigated the 6061 alloy deformation and dynamic softening mechanism they found dynamic recovery and
recrystallization only occurred at low Z values. With decreasing Z value, the subgrain size increased and the dislocation density decreased, which is consistent with data presented in present study. H. E. Hu et al [44] studied the hot processing maps for 6061 aluminum alloy, their conclusion showed that dynamic recovery and dynamic recrystallization play the role at the same time during high temperature deformation of 6061 aluminium alloy. High deformation temperature and low strain rate benefit dynamic recrystallization and hot workability. Compared with the difference of DRX between the alloys without particles and the AA6061/B4C composites, we easily found that particles promote DRX. For the Fig. 6(a) and (b) samples with lower Z values, obvious grain growth process has happened. Most of the grains are equiaxial with relatively larger grain size. Different from the samples with higher Z values, their grain size distribution curves have another peak in the range of 20~30μm, which is very close to their original average grain size (21.65μm), other than the peak around 5μm mentioned above. That is to say, these samples are kept at different stages of recrystallization after compression tests. 3.3.2 Local misorientation Up to now, the PSN caused by particle deformation zones (PDZ) formed during deformation has been extensively studied [13,45,46]. As discussed in section 3.1, The presence of hard B4C particles in aluminum matrix during hot deformation causes the interface to crack and debond since the matrix undergoes plastic flow while the extremely hard particles do not deform. matrix around the particles undergoes a greater deformation owing to the existence of particles. the mismatch between a non-deforming particle and drastic deforming metal matrix during deformation leads to the enforced strain gradient in the matrix near a particle. The strain gradient creates a region of high dislocation density and large orientation gradient, which is called PDZ [47]. PDZ can extend to a distance of about a diameter of the particle from the surface of the particle and it may be misoriented by tens of degrees from adjacent matrix. Thus, the PDZ can
be considered as an ideal site for development of a recrystallization nucleus, and particles can promote the nucleation of DRX. Fig.7 shows the local misorientation map of compressed sample with different Z values. In these maps, the local misorientations are interpreted in terms of density of geometrically necessary dislocations to represent the local strain value. Meanwhile, the thin black lines indicate boundaries of misorientation between 2° and 15°, and the thick black lines indicate boundaries above 15°. As can be seen, local misorientation of the samples with higher Z values are larger than the lower ones. For the higher Z values samples (Fig.7 (e) and (f)), there are many large deformed grains, which are connected with local stress concentration, as indicated by the red arrows. Besides, there are many nuclei with small local misorientation in the vicinity of large deformed grains, forming a necklace structure. All of these features indicate the occurrence of particle stimulated nucleation (PSN). It has been reported that B4C particle favor the particle stimulated nucleation（PSN）by nucleation of new grains in the deformation zones, the particles decreased the potential barrier of DRX nucleation, makes composites hot deformation more easier than 6061 alloy[48]. While as the decrease of Z value, most of the large deformed grains are consumed by DRX, accompanied by nucleation and nuclei growth. Eventually, for the Fig. 7(a) sample, a homogeneous grain structure with small local misorientation is obtained. Therefore, in the present work, the growth of grain size and the decrease of local strain and LAB are linked closely with the decreasing Z values. Such evolution indicates the occurrence of partial or fully DRX. It can be seen from the distribution map of grain size and local strain that the lower the Z value is, the easier the DRX happens, in other words, higher deformation temperature and lower stain rate can lead to thoroughly DRX. As presented in Fig. 7(a) and (b)（lower Z values）, the thoroughly DRX can release the
local stress, so it also reduces the oscillation of flow stress curves, which is a supplementary conclusion for flow stress behavior.
Fig.7 the local misorientation maps with grain boundary of specimens deformed under different Z values: (a) lnZ=17.3132; (b) lnZ=18.31822; (c) lnZ=19.41025; (d) lnZ=20.6208; (e) lnZ=21.71284; (f) lnZ=22.92339. 4. Conclusions Flow stress behavior of AA6061/B4C composites and dynamic softening mechanism in hot deformation were investigated by hot compression test under different strain rates and temperatures. Following conclusions can be reached: (1) Typical true stress-true strain curves demonstrate that flow stress increases with the increase of strain rate and the decrease of temperature. The true stress increases rapidly with the increase of true strain on account of the work hardening, and then holds to a steady value or decreases slightly after
the true stress reaching the peak value due to the dynamic softening mechanism. (2) Flow stress behavior of AA6061/B4C are described by constitutive equation in hyperbolic sine function with the hot deformation activation energy 151.634kJ/mol. The combined effects of strain rate and temperature can be characterized by a Zener-hollomon parameter. The bigger the Z value is, the higher the flow stress. (3) EBSD analysis of local stress and grain size evolution under different Z values shows that the main softening mechanism of B4C/6061 in hot deformation was identified as DRX. Those specimens deformed under lower Z values has bigger grain size and lower local stress, as a result of grain growth and stress relief during DRX process. The lower the Z value is, the easier the DRX occurs. Acknowledgements This work was financially supported by Shenzhen Development and Reform Commission Engineering Laboratory Project (NO.2015-1033) and Shenzhen Technology Research and Development Plan Technology Innovation Project (NO.CXZZ20140702113545562). This work made use of the Shenzhen Engineering Laboratory of Nuclear Materials and Service Safety. References [1] L. Chen, Y. Yao. Acta Metallurgica Sinica-English Letters, 27 (2014) 762-774. [2] D.C. Halverson, A.J. Pyzik, I.A. Aksay, W.E. Snowden, Journal of the American Ceramic Society, 72 (1989) 775-780. [3] P. Zhang, Y. Li, W. Wang, Z. Gao, B. Wang, Journal of Nuclear Materials, 437 (2013) 350-358. [4] F. Thevenot, Journal of the European Ceramic Society, 6 (1990) 205-225.
[5] J.M. Torralba, C.E. da Costa, F. Velasco, Journal of Materials Processing Technology, 133 (2003) 203-206. [6] D.J. Lloyd, International Materials Reviews, 39 (1994) 1-23. [7] F. Thevenot, J. Cueilleron, Analusis, 5 (1977) 105-121. [8] H.J. McQueen, A. Shen, P. Leo, B. Cerri, AL-ZN-MG for extrusion - hot workability, 2008. [9] Y. Li, Q. Li, W. Liu, G. Shu, Journal of Alloys and Compounds, 684 (2016) 496-503. [10] H.J. McQueen, M.E. Kassner, Scripta Materialia, 51 (2004) 461-465. [11] T. Nishizawa, I. Ohnuma, K. Ishida, Materials Transactions Jim, 38 (1997) 950-956. [12] Y.S. Li, Y. Zhang, N.R. Tao, K. Lu, Acta Materialia, 57 (2009) 761-772. [13] R.D. Doherty, D.A. Hughes, F.J. Humphreys, J.J. Jonas, D.J. Jensen, M.E. Kassner, W.E. King, T.R. McNelley, H.J. McQueen, A.D. Rollett, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 238 (1997) 219-274. [14] Y.W. Wu, K. Wu, K.K. Deng, K.B. Nie, X.J. Wang, X.S. Hu, M.Y. Zheng. Journal of Alloys and Compounds 506(2) (2010) 688-692. [15] H. Chang, X. Wang, X. Hu, Y. Wang, K. Nie, K. Wu. Rare Metal Materials and Engineering 43(8) (2014) 1821-1825. [16] G. Ganesan, K. Raghukandan, R. Karthikeyan, B.C. Pai. Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing 369(1-2) (2004) 230-235. [17] N. Ravichandran, Metallurgical Transactions a-Physical Metallurgy and Materials Science, 22 (1991) 2339-2348. [18] S. Gourdet, F. Montheillet, Acta Materialia, 51 (2003) 2685-2699.
[19] H.J. McQueen, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 387 (2004) 203-208. [20] Y. Deng, Z. Yin, J. Huang, Materials Science and Engineering aStructural Materials Properties Microstructure and Processing, 528 (2011) 1780-1786. [21] Y. Li, Q. Li, W. Liu, G. Shu. Journal of Alloys and Compounds 684 (2016) 496-503. [22] J. Zheng, Q. Li, W. Liu, G. Shu. Journal of Composite Materials 50(27) (2016) 3843-3852. [23] L. Hu, M. O'Neil, V. Erturun, R. Benitez, G. Proust, I. Karaman, M. Radovic. Scientific Reports 6 (2016). [24] H.R. Ezatpour, M.H. Sabzevar, S.A. Sajjadi, Y. Huang. Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing 606 (2014) 240-247. [25] B.L. Zhang, M.S. Maclean, T.N. Baker. Materials Science and Technology 16(7-8) (2000) 897-902. [26] N. Srinivasan, Y. Prasad, P.R. Rao, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 476 (2008) 146-156. [27] N. Jin, H. Zhang, Y. Han, W. Wu, J. Chen, Materials Characterization, 60 (2009) 530-536. [28] X.Y. Liu, Q.L. Pan, Y.B. He, W.B. Li, W.J. Liang, Z.M. Yin, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 500 (2009) 150-154. [29] H. Zhang, L.X. Li, D. Yuan, D.S. Peng, Materials Characterization, 58 (2007) 168-173.
[30] X. Xiaoxin, P. Sakaris, H.J. McQueen. Materials Science and Technology 10(6) (1994) 487-96. [31] S.L. Semiatin, M.R. Staker, J.J. Jonas. Acta Metallurgica 32(9) (1984) 1347-1354. [32] L.X. Li, Y. Lou, L.B. Yang, D.S. Peng, K.P. Rao, Materials & Design, 23 (2002) 451-457. [33] F.J. Humphreys, W.S. Miller, M.R. Djazeb. Materials Science and Technology 6(11) (1990) 1157-1166. [34] H. Shi, A.J. McLaren, C.M. Sellars, R. Shahani, R. Bolingbroke, Materials Science and Technology, 13 (1997) 210-216. [35] H.J. McQueen, N.D. Ryan, Materials Science and Engineering aStructural Materials Properties Microstructure and Processing, 322 (2002) 43-63. [36] H. Li, H. Wang, M. Zeng, X. Liang, H. Liu, Composites Science and Technology, 71 (2011) 925-930. [37] E. Cerri, S. Spigarelli, E. Evangelista, P. Cavaliere, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 324 (2002) 157-161. [38] S. Gourdet, F. Montheillet, Materials Science and Engineering aStructural Materials Properties Microstructure and Processing, 283 (2000) 274-288. [39] D. Sang, R. Fu, Y. Li. Metals 6(3) (2016). [40] E.M. Herba, H.J. McQueen. Materials Science and Technology 14(9-10) (1998) 1057-1064. [41] H.J. McQueen, E. Fry, P. Cavaliere, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 324 (2002) 157-161.
[42] C.I. Chang, C.J. Lee, J.C. Huang, Scripta Materialia, 51 (2004) 509-514. [43] X.Y. Yang, H.M. Miura, T. Sakai, Materials Transactions, 44 (2003) 197-203. [44] H.E. Hu, X.Y. Wang, L. Deng. Materials Science and Technology 30(11) (2014) 1321-1327. [45] B. Inem. Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing 197(1) (1995) 91-95. [46] X.J. Wang, K. Wu, H.F. Zhang, W.X. Huang, H. Chang, W.M. Gan, M.Y. Zheng, D.L. Peng. Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing 465(1-2) (2007) 78-84. [47] F.J. Humphreys, P.N. Kalu. Acta Metallurgica 35(12) (1987) 2815-2829. [48] A. Belyakov, H. Miura, T. Sakai, Materials Science and Engineering aStructural Materials Properties Microstructure and Processing, 255 (1998) 139-147.
PII: DOI: Reference:
S0921-5093(17)30304-0 http://dx.doi.org/10.1016/j.msea.2017.03.013 MSA34796
To appear in: Materials Science & Engineering A Received date: 27 October 2016 Revised date: 2 March 2017 Accepted date: 3 March 2017 Cite this article as: Kaikai Wang, Xiaopei Li, Qiulin Li, Guogang Shu and Guoyi Tang, Hot deformation behavior and microstructural evolution of particulatereinforced AA6061/B4C composite during compression at elevated temperature, Materials Science & Engineering A, http://dx.doi.org/10.1016/j.msea.2017.03.013 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 galley proof before it is published in its final citable 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.
Hot deformation behavior and microstructural evolution of particulatereinforced AA6061/B4C composite during compression at elevated temperature Kaikai Wanga,b, Xiaopei Lia, Qiulin Lia,b,1*, Guogang Shuc, Guoyi Tanga,* a
Advanced Materials Institute, Graduate School at Shenzhen, Tsinghua University, Shenzhen 518055, PR China b
Shenzhen Engineering Laboratory of Nuclear Materials and Service Safety, Shenzhen, 518055, PR China c
Joint Laboratory of Nuclear Materials and Service Safety, Shenzhen, 518055, PR China Abstract The flow stress behavior of AA6061/B4C composites has been researched by compression test using Gleeble-3800 thermal simulator, in the temperature ranging from 633-783K and strain rate ranging from 0.001-1s-1. Typical true stress-true strain curves showed that the peak stress levels decreased with the rising of temperature but increased with the rising of strain rates. The combined effects of temperature and strain rate on deformation were analyzed by constitutive equation which containing the Zener-Hollomon parameter (Z) in hyperbolic sine function. The effects of Z values on dynamic softening and associated microstructural evolution during hot deformation were investigated by electron back scattered diffraction technique (EBSD). It was found that with the decrease of Z values, local strain induced by deformation was released and the grain size of aluminum matrix increased gradually, which indicated that the main softening mechanism of **
Corresponding author: Tel/fax: +86 755 2603 6224 Corresponding author. Tel/fax: +86 755 2603 6752. E-mail address: [email protected] (G. Tang) 1
[email protected] (Q. Li)
AA6061/B4C composites was dynamic recrystallization (DRX), and the lower the Z value was, the easier the DRX occurred. Keywords Metal matrix composites; hot compression; flow stress behavior; constitutive equation; dynamic recrystallization 1. Introduction Aluminum alloy 6061 is widely used as a base matrix for Metal matrix composites (MMCs) due to its perfect machinability and good wear and mechanical properties. And boron carbide (B4C) is one of the promising reinforcement particle in Al matrix composites for its attractive properties, such as super high hardness, low density and good thermal and chemical stability, making the AA6061/B4C MMCs popular in automotive and aerospace industry, and especially in nuclear industry [1-3]. For instance, it is often used as a neutron absorption material for storage and transportation applications of spent fuel [4]. Traditionally, Al/ B4C MMCs are manufactured by powder metallurgy (P/M). However, P/M method is still far from being used in industrial application for its high cost and limited dimension of the finished product. Among all the manufacturing methods, liquid-stirring casting, an efficient and cost-saving method, which introduces reinforced particles into molten metal through a vortex, is ideal for industrial production of Al/ B4C MMCs and is employed in the present study [5-7]. Additionally, it is commonly accepted that AA6061/B4C composite often needs secondary processing such as hot extrusion and rolling before it is used as structural or functional material in industry. However, owing to the super high hardness of reinforcement B4C particles in the aluminum matrix, hot working and machinability of the AA6061/B4C composite is much harder than matrix aluminum alloy. So, in the light of the above analysis, the studies on flow stress behavior and the corresponding microstructure evolution during hot deformation process becomes rather important. In order to figure
out the deformation process of AA6061/B4C composites and develop a processing methods such as hot rolling and hot extrusion, it is necessary to study the flow stress behavior, deformation mechanism and microstructure evolution under different processing conditions [8, 9]. It is well known that some dynamic behaviors such as work hardening (WH), dynamic recrystallization (DRX) and dynamic recovery (DRV) often occur during the hot deformation of alloys or metallic composites, accompanied by mechanical properties and microstructure evolution of material [10-13]. Dynamic recrystallization (DRX) affects the microstructure evolution of matrix during hot deformation, which can influence the properties of the composites to some great extent [14]. Thus, it is necessary to study the DRX of Al matrix composite during hot deformation. For particle reinforced Al matrix composites, particles will significantly affect the recrystallization behavior of metal matrix [15]. Ganesan et al. [16] investigated the development of processing maps for 6061Al/15% SiCp composite material by hot compression test. they found the microstructure evaluation displays particle fracture, debonding, adiabatic shear band formation and matrix cracking which led to flow instability of the composites, and they also identified the super plastic deformation and dynamic recrystallization zones, corresponding to optimum working regions. Generally, DRX depends sensitively on the deformation temperature and strain rate. The combined effects of strain rate and deformation temperature on hot deformation and DRX behavior can be represented by a Zenerhollomon parameter (Z) [17-19]. In order to study the DRX behavior of AA6061/B4C through Z values, it is necessary to explore the relationship between Z values and the corresponding microstructures, which will provide valuable instructions for the control of structure and performance of AA6061/B4C in industrial manufacturing [20] Based on the above statement, in this paper, the flow stress behavior of AA6061/B4C is analyzed by typical true stress-true strain curves and constitutive equation with hot deformation activation energy. Microstructure evolution under different deformation
conditions is studied using EBSD. The relationship between Z values and dynamic softening mechanism is also discussed. 2. Experimental In this paper, AA6061/B4C composites containing 31 vol.% B4C particles were prepared by the stable liquid-stirring process, using commercially available pure aluminum, Al-Ti and Al-Si master alloy as raw materials, and B4C powders with grain sizes of 15-50 μm (23.2 μm, average grain size) as reinforcement. Stirring casting ingot B4C/6061, with a size of 65mm in diameter and 130mm in height, were homogenized at 698 K for 12 h and furnace cooled to room temperature. The ingots prepared for compression test were machined to cylindrical specimens with 10mm in diameter and 15mm in height. Compression tests were carried out on Gleeble-3800 thermal simulator at the temperatures of 633 K, 663 K, 693 K, 723 K, 753 K and 783 K, at the strain rate of 0.001 s-1, 0.01 s-1, 0.1 s-1 and 1 s-1 respectively. During compression tests, samples were heated to target temperature and homogenized for 3 minutes in order to obtain a stable and uniform condition for deformation. All samples were compressed to the true stain of -0.7 (55% reduction), and quenched in water immediately after compression. All deformed specimens for microstructural observations were sectioned along the compression axis direction, followed by mechanical grinding and ion beam cross section polisher (IB-09020CP, JOEL) successively. Electron back scattered diffraction (EBSD) analyses were performed using the scanning electron microscopy (SEM, TESCAN MIRA 3 LMH), step size between points on the scan grid was set to 1μm. 3. Results and discussion 3.1 Flow stress behavior Actually, the fabrication parameters and interaction between B4C and matrix have great effects on the mechanical of composites, and these effects are directly revealed in strain-stress curves.
Yu Li et al. [21] fabricated the Tri-modal Al-B4C-Al3Ti composites and investigated the effect of Ti content and stirring time on microstructure and mechanical behavior of Al- B4C composites. They researched the interface reaction between Al, B4C and Ti and found that, higher Ti levels resulted in higher volume fraction and particle size of Al3Ti denser TiB2 interface and even more uniform spatial distribution of B4C. A prolonged stirring time led to copious coarse TiB2 crystals at the expense of Al3Ti particles and more uniform particle spatial distribution was achieved. During plastic deformation, these coarse TiB2 crystals sticking out of B4C surface can carry extra applied load. As a result, their tensile strength showed an inclined leaf-shape curve. Jiyun Zheng et al [22] discussed the microstructure evolution and interface reaction of 15wt% B4C/Al composites with titanium addition during liquidstirring process. They found that The dispersion of B4C particles changes from clusters to discrete distribution as time increases, which attributes to improved wetting through interfacial reaction in addition to high speed stirring. They also pointed that interface reaction produced Al3BC, TiB2, and AlB2 by B4C erosion and Al3Ti decomposition; AlB2 only precipitated in matrix after long time stirring. The growth of TiB2 transformed from a fine layer to discretely coarse crystals on the B4C surface. Similar discussion of interaction in Al alloy/Ti2AlC composites system and detailed chemical mechanism was studied by Liangfa Hu et al [23], which could give some instruction for further investigation of B4C /Al composites system.
Hot deformation theory indicated that flow stress is the most basic parameter to characterize plastic deformation properties of metals and alloys. The stress determines the size of load and energy needed during plastic deformation of materials [24]. From compression test, a series of typical true stress-true strain curves of the AA6061/B4C composites at various deformation conditions are presented in Fig. 1. It can be easily found that the deformation temperatures and strain rates have obvious effects on the flow stress behavior. As shown in the flow stress curves, the true stress increases rapidly because of the work hardening with the increase of true strain, and then holds to a steady value or decreases slightly after the true stress reaches the peak value as a result of dynamic softening. Apart from this, Fig. 1 also shows that the peak values of stresses fluctuated at different deformation conditions. It decreases with the increase of deformation temperatures, while it increases with the increase of strain rates at a given temperature. Obviously, lower temperatures lead to more drastic work hardening, as the temperature increases, the work hardening is decreases. From the test conducted by B. L. Zhang et al [25], we also found the similar changing trend of flow stress in AA6061/SiCp MMCs, and they pointed that The flow stresses of the AA6061/SiCp composite were significantly higher than those of the unreinforced alloy at the lower deformation temperature. These curves in present work also reveal that the AA6061/B4C composite is a positive strain rate sensitive material. Current research shows that the deformation at elevated temperature is a competing process of work hardening and dynamic softening [26-29], and dynamic recovery and dynamic recrystallization are the two main softening mechanisms in hot deformation of particle reinforced Aluminum matrix composites. In the present study, the dynamic softening seems more distinguished at higher temperature and lower strain rate. At the beginning of the deformation,
the work hardening exceeds the dynamic softening, the dislocation density increased dramatically under the external force, leading to the rapid increase of true stress at the initial stage of deformation. As the compression proceeding, the strengthening effect of B4C particles is receded, dynamic softening occurs simultaneously and it counteracts or partially offsets the effect of work hardening, which makes the true stresses hold unchanged or decreased slightly after reaching the peak values. [30,31]. However, different mechanisms can result in similar flow stress behavior [32] for instance, flow softening can be attributed to DRX, temperature rise of the specimens and texture evolution, or grain size increase during the deformation process. It is difficult to define the accurate deformation mechanism only by the typical true stress-true strain curves, so comprehensive microstructure evolution analysis will be discussed to explain the specific process and softening mechanism of deformation. Besides of the above discussion, it is necessary for us to compare the particle reinforced composite deformation behavior with non-reinforced matrix alloy, which is helpful for us to figure out the effects of particles on deformation Actually, the flow stress for the composite is higher than that of the alloy, especially at the low deformation temperature, compared to the study conducted by H. R. Ezatpour et al [24]. They researched the work softening mechanisms and texture in 6061 aluminum alloy during hot deformation. Their experiment showed that the flow stress increased with increasing strain rate and decreasing deformation temperature and dynamic recovery and recrystallization only occurred at low Z values. And it has been explained by Humphreys et al [33]. During deformation, the strengthening of composites is caused by the generation of the vacancies and dislocations in the matrix around the particles, owing to incompatibility between the deforming matrix and the non-deformable particles, dislocations accumulate around the particles, resulting in a higher flow stress in the composite compared with the pure alloy.
Traditionally DRX is associated with hot working of high stacking fault energy (SFE) metals which exhibit flow softening after reaching a critical strain. Flow curves eventually reach a steady state at large true strain. As shown in Fig.1.a~d, at lower strain rates oscillations in the flow curves are observed after reaching a steady state. Some other Detailed analysis will be discussed in next section.
Fig.1.True stress-true stain curve of AA6061/B4C composites during hot compression deformation: (a) ε=1s-1; (b) ε=0.1s-1; (c) ε=0.01s-1; (d) ε=0.001s1
3.2 kinetic analysis 3.2.1 constitutive equation As for hot deformation behavior of metallic materials and metal matrix composites, it is commonly accepted that the relationship between the strain rate, the peak stress (or the steady stress) and deformation temperature can be described by the empirical function using Arrhenius type equation[34-36]as follows： A
(1)
where F(σ) is a function of applied stress with the following three possible forms： =
if
= =
if
(2)
for all values of σ
Where σ is the flow stress (Mpa) for a given strain, T is the absolute deformation temperature (K), R is the gas constant(8.314Jmol-1K-1), Q is the activation energy of deformation (Jmol-1) and A,β,α(=β/ n’), n, n’ are material constants. For the low stress level and high stress level, substituting suitable F(σ) function into Eq.(2) leads to Eq.(3)and Eq.(4), respectively [37,38]. ε=A1σn'
ασ<0.8 (3),
ε=A2eβσ
ασ>1.2 (4),
Where A1 and A2 are material constants. By taking the logarithm for both sides of Eq. (3) and Eq. (4) and substitute the peak stress, the relationship between the flow stress and strain rate (lnσlnε and σ-lnε) are shown in Fig.2(a) and Fig.2(b). The flow stress can be
estimated by a group of parallel straight lines in different hot compression conditions. The material constants n’ and β can be obtained from the slope of every single line in the lnσ-lnε and σ-lnε plots by linear regression fit method. The mean value of n’ and β can be calculated as 9.331801 and 0.158482 Mpa-1 respectively, then α=β/ n’=0.016983.
Fig2. Evaluating the value of (a).n’ by fitting lnσ-lnε, (b).β by fitting σ-lnε, (c).n by fitting ln[sinh(ασ)]-lnε, (d).Q by fitting ln[sinh(ασ)]-1/T
Eq. (1) can be expressed in terms of temperature compensated stain rate factor the Zener-Hollomon parameter Z, which is defined by Z=εeQ/RT=A[sinh(ασ)]n
(5),
Where Q is the hot deformation activation energy. Similarly, for all the stress level, the value of n in Eq. (4) can be derived from the slopes of lines representative of ln[sinh(ασ)]-lnε by taking the logarithm, n and Q can be calculated as 6.93861 and 151.634 kJmol-1, respectively. The details of all material constants evaluated in the present study are given in Table 1. Table 1 Material constants and Q of AA6061/B4C in the hyperbolic sine equation Q/kJmol-1 151.634
α/Mpa-1 0.016983
A/s-1 1.94658*109
n 6.93861
The constitutive equation can be described by hyperbolic sine equation as follow：
{
[
]
}
{( [(
) )
]
}
3.2.2 hot deformation activation energy Generally, the hot deformation activation energy can qualitatively reflect the energy barrier to dislocation motion for metals and alloys during hot deformation. A higher value of activation energy signifies the existence of higher dragging forces to the movement of the dislocation in hot deformation [39]. The hot working behavior of unreinforced Al6061 alloys has been extensively studied. Table 2 shows a comparison of Q values for 6061 alloys under different conditions and B4C /6061 composites from the present work and other investigation.
Table2 Comparison of activation energy Q values for different conditions of 6061 Al alloys and B4C /6061 composites Material B4C /6061（stirring casting, Hot extruded） 6061
Q/kJmol-1 152 344
Reference * [44]
6061 6061 * for present work
274 205
[24] [40]
Obviously, the activation energy of the 6061 alloy is larger than that of the AA6061/B4C composite（151kJmol-1）In the current compression test,. This may be attributed to the reinforced B4C particles. It will be discussed in more detail in Section 3.3.2.
Fig3.Relationship between ln[sinh(ασ)] and lnZ
3.3 microstructural evolution 3.3.1 Grain size
Fig 4 IPF map for initial homogenized AA6061/B4C composite Fig.4 is the inverse pole figure (IPF) map with grain boundaries for homogenized AA6061/B4C. It can be seen that the initial sample, which has been treated by homogenization, consist of uniform equiaxed grains with average diameter of 21.65μm. Many researches have reported that dynamic recrystallization occurs in aluminum alloy during hot deformation, and the dynamic recrystallization behavior can be described by Zener-hollomon parameter (Z), which corresponds to the flow behavior and microstructure after deformation [26,28,41]. Z parameter is also called as a temperature-compensate strain
rate, which could represent the combined effects of strain rate and temperature on hot deformation [42,43]. Z values obtained from Eq. (5) and average grain size of the specimens under different deformation conditions are shown in Table 3. Table 3 Z values and grain sizes under different deformation conditions lnZ1 lnZ2 lnZ3 lnZ4 lnZ5 lnZ6
Temperature/K Temperature 753 480 723 450 693 420 723 450 753 480 723 450
Strain rate 0.001 0.001 0.001 0.01 0.1 0.1
lnZ/s-1 17.3132 18.31822 19.41025 20.6208 21.91837 22.92339
Grain size/μm 14.29 13.38 10.9 10.28 9.99 7.98
The grain sizes correspond to different Z values are shown in Fig.5. As can be seen, the lower the Z is, the bigger the grain sizes are. DRX is a beneficial process in hot deformation since it only gives stable flow and good workability to the materials by simultaneously softening.
Fig.5 Average grain size under different Z values
Fig.6 the inverse pole figure (IPF) map with grain boundary and grain size distribution maps of specimens deformed under different Z values: (a) lnZ=17.3132; (b) lnZ=18.31822; (c) lnZ=19.41025; (d) lnZ=20.6208; (e) lnZ=21.71284; (f) lnZ=22.92339. Fig.6 shows the microstructure evolution of compressed samples. Microstructures are given in the form of inverse pole figure (IPF) map embed with grain size distribution map. as shown in Fig.6e and Fig.6f, the samples with higher Z value, a lot of fine grains can be observed while many large grains still remain. The large grains are non-recrystallized, most of the grains are newly formed nuclei with little grain size. As can be seen in the grain size distribution maps, both curves have only one peak around 5μm, indicating that the DRX does not finish in the extruded composites. H. R. Ezatpour et al [24] investigated the 6061 alloy deformation and dynamic softening mechanism they found dynamic recovery and
recrystallization only occurred at low Z values. With decreasing Z value, the subgrain size increased and the dislocation density decreased, which is consistent with data presented in present study. H. E. Hu et al [44] studied the hot processing maps for 6061 aluminum alloy, their conclusion showed that dynamic recovery and dynamic recrystallization play the role at the same time during high temperature deformation of 6061 aluminium alloy. High deformation temperature and low strain rate benefit dynamic recrystallization and hot workability. Compared with the difference of DRX between the alloys without particles and the AA6061/B4C composites, we easily found that particles promote DRX. For the Fig. 6(a) and (b) samples with lower Z values, obvious grain growth process has happened. Most of the grains are equiaxial with relatively larger grain size. Different from the samples with higher Z values, their grain size distribution curves have another peak in the range of 20~30μm, which is very close to their original average grain size (21.65μm), other than the peak around 5μm mentioned above. That is to say, these samples are kept at different stages of recrystallization after compression tests. 3.3.2 Local misorientation Up to now, the PSN caused by particle deformation zones (PDZ) formed during deformation has been extensively studied [13,45,46]. As discussed in section 3.1, The presence of hard B4C particles in aluminum matrix during hot deformation causes the interface to crack and debond since the matrix undergoes plastic flow while the extremely hard particles do not deform. matrix around the particles undergoes a greater deformation owing to the existence of particles. the mismatch between a non-deforming particle and drastic deforming metal matrix during deformation leads to the enforced strain gradient in the matrix near a particle. The strain gradient creates a region of high dislocation density and large orientation gradient, which is called PDZ [47]. PDZ can extend to a distance of about a diameter of the particle from the surface of the particle and it may be misoriented by tens of degrees from adjacent matrix. Thus, the PDZ can
be considered as an ideal site for development of a recrystallization nucleus, and particles can promote the nucleation of DRX. Fig.7 shows the local misorientation map of compressed sample with different Z values. In these maps, the local misorientations are interpreted in terms of density of geometrically necessary dislocations to represent the local strain value. Meanwhile, the thin black lines indicate boundaries of misorientation between 2° and 15°, and the thick black lines indicate boundaries above 15°. As can be seen, local misorientation of the samples with higher Z values are larger than the lower ones. For the higher Z values samples (Fig.7 (e) and (f)), there are many large deformed grains, which are connected with local stress concentration, as indicated by the red arrows. Besides, there are many nuclei with small local misorientation in the vicinity of large deformed grains, forming a necklace structure. All of these features indicate the occurrence of particle stimulated nucleation (PSN). It has been reported that B4C particle favor the particle stimulated nucleation（PSN）by nucleation of new grains in the deformation zones, the particles decreased the potential barrier of DRX nucleation, makes composites hot deformation more easier than 6061 alloy[48]. While as the decrease of Z value, most of the large deformed grains are consumed by DRX, accompanied by nucleation and nuclei growth. Eventually, for the Fig. 7(a) sample, a homogeneous grain structure with small local misorientation is obtained. Therefore, in the present work, the growth of grain size and the decrease of local strain and LAB are linked closely with the decreasing Z values. Such evolution indicates the occurrence of partial or fully DRX. It can be seen from the distribution map of grain size and local strain that the lower the Z value is, the easier the DRX happens, in other words, higher deformation temperature and lower stain rate can lead to thoroughly DRX. As presented in Fig. 7(a) and (b)（lower Z values）, the thoroughly DRX can release the
local stress, so it also reduces the oscillation of flow stress curves, which is a supplementary conclusion for flow stress behavior.
Fig.7 the local misorientation maps with grain boundary of specimens deformed under different Z values: (a) lnZ=17.3132; (b) lnZ=18.31822; (c) lnZ=19.41025; (d) lnZ=20.6208; (e) lnZ=21.71284; (f) lnZ=22.92339. 4. Conclusions Flow stress behavior of AA6061/B4C composites and dynamic softening mechanism in hot deformation were investigated by hot compression test under different strain rates and temperatures. Following conclusions can be reached: (1) Typical true stress-true strain curves demonstrate that flow stress increases with the increase of strain rate and the decrease of temperature. The true stress increases rapidly with the increase of true strain on account of the work hardening, and then holds to a steady value or decreases slightly after
the true stress reaching the peak value due to the dynamic softening mechanism. (2) Flow stress behavior of AA6061/B4C are described by constitutive equation in hyperbolic sine function with the hot deformation activation energy 151.634kJ/mol. The combined effects of strain rate and temperature can be characterized by a Zener-hollomon parameter. The bigger the Z value is, the higher the flow stress. (3) EBSD analysis of local stress and grain size evolution under different Z values shows that the main softening mechanism of B4C/6061 in hot deformation was identified as DRX. Those specimens deformed under lower Z values has bigger grain size and lower local stress, as a result of grain growth and stress relief during DRX process. The lower the Z value is, the easier the DRX occurs. Acknowledgements This work was financially supported by Shenzhen Development and Reform Commission Engineering Laboratory Project (NO.2015-1033) and Shenzhen Technology Research and Development Plan Technology Innovation Project (NO.CXZZ20140702113545562). This work made use of the Shenzhen Engineering Laboratory of Nuclear Materials and Service Safety. References [1] L. Chen, Y. Yao. Acta Metallurgica Sinica-English Letters, 27 (2014) 762-774. [2] D.C. Halverson, A.J. Pyzik, I.A. Aksay, W.E. Snowden, Journal of the American Ceramic Society, 72 (1989) 775-780. [3] P. Zhang, Y. Li, W. Wang, Z. Gao, B. Wang, Journal of Nuclear Materials, 437 (2013) 350-358. [4] F. Thevenot, Journal of the European Ceramic Society, 6 (1990) 205-225.
[5] J.M. Torralba, C.E. da Costa, F. Velasco, Journal of Materials Processing Technology, 133 (2003) 203-206. [6] D.J. Lloyd, International Materials Reviews, 39 (1994) 1-23. [7] F. Thevenot, J. Cueilleron, Analusis, 5 (1977) 105-121. [8] H.J. McQueen, A. Shen, P. Leo, B. Cerri, AL-ZN-MG for extrusion - hot workability, 2008. [9] Y. Li, Q. Li, W. Liu, G. Shu, Journal of Alloys and Compounds, 684 (2016) 496-503. [10] H.J. McQueen, M.E. Kassner, Scripta Materialia, 51 (2004) 461-465. [11] T. Nishizawa, I. Ohnuma, K. Ishida, Materials Transactions Jim, 38 (1997) 950-956. [12] Y.S. Li, Y. Zhang, N.R. Tao, K. Lu, Acta Materialia, 57 (2009) 761-772. [13] R.D. Doherty, D.A. Hughes, F.J. Humphreys, J.J. Jonas, D.J. Jensen, M.E. Kassner, W.E. King, T.R. McNelley, H.J. McQueen, A.D. Rollett, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 238 (1997) 219-274. [14] Y.W. Wu, K. Wu, K.K. Deng, K.B. Nie, X.J. Wang, X.S. Hu, M.Y. Zheng. Journal of Alloys and Compounds 506(2) (2010) 688-692. [15] H. Chang, X. Wang, X. Hu, Y. Wang, K. Nie, K. Wu. Rare Metal Materials and Engineering 43(8) (2014) 1821-1825. [16] G. Ganesan, K. Raghukandan, R. Karthikeyan, B.C. Pai. Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing 369(1-2) (2004) 230-235. [17] N. Ravichandran, Metallurgical Transactions a-Physical Metallurgy and Materials Science, 22 (1991) 2339-2348. [18] S. Gourdet, F. Montheillet, Acta Materialia, 51 (2003) 2685-2699.
[19] H.J. McQueen, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 387 (2004) 203-208. [20] Y. Deng, Z. Yin, J. Huang, Materials Science and Engineering aStructural Materials Properties Microstructure and Processing, 528 (2011) 1780-1786. [21] Y. Li, Q. Li, W. Liu, G. Shu. Journal of Alloys and Compounds 684 (2016) 496-503. [22] J. Zheng, Q. Li, W. Liu, G. Shu. Journal of Composite Materials 50(27) (2016) 3843-3852. [23] L. Hu, M. O'Neil, V. Erturun, R. Benitez, G. Proust, I. Karaman, M. Radovic. Scientific Reports 6 (2016). [24] H.R. Ezatpour, M.H. Sabzevar, S.A. Sajjadi, Y. Huang. Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing 606 (2014) 240-247. [25] B.L. Zhang, M.S. Maclean, T.N. Baker. Materials Science and Technology 16(7-8) (2000) 897-902. [26] N. Srinivasan, Y. Prasad, P.R. Rao, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 476 (2008) 146-156. [27] N. Jin, H. Zhang, Y. Han, W. Wu, J. Chen, Materials Characterization, 60 (2009) 530-536. [28] X.Y. Liu, Q.L. Pan, Y.B. He, W.B. Li, W.J. Liang, Z.M. Yin, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 500 (2009) 150-154. [29] H. Zhang, L.X. Li, D. Yuan, D.S. Peng, Materials Characterization, 58 (2007) 168-173.
[30] X. Xiaoxin, P. Sakaris, H.J. McQueen. Materials Science and Technology 10(6) (1994) 487-96. [31] S.L. Semiatin, M.R. Staker, J.J. Jonas. Acta Metallurgica 32(9) (1984) 1347-1354. [32] L.X. Li, Y. Lou, L.B. Yang, D.S. Peng, K.P. Rao, Materials & Design, 23 (2002) 451-457. [33] F.J. Humphreys, W.S. Miller, M.R. Djazeb. Materials Science and Technology 6(11) (1990) 1157-1166. [34] H. Shi, A.J. McLaren, C.M. Sellars, R. Shahani, R. Bolingbroke, Materials Science and Technology, 13 (1997) 210-216. [35] H.J. McQueen, N.D. Ryan, Materials Science and Engineering aStructural Materials Properties Microstructure and Processing, 322 (2002) 43-63. [36] H. Li, H. Wang, M. Zeng, X. Liang, H. Liu, Composites Science and Technology, 71 (2011) 925-930. [37] E. Cerri, S. Spigarelli, E. Evangelista, P. Cavaliere, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 324 (2002) 157-161. [38] S. Gourdet, F. Montheillet, Materials Science and Engineering aStructural Materials Properties Microstructure and Processing, 283 (2000) 274-288. [39] D. Sang, R. Fu, Y. Li. Metals 6(3) (2016). [40] E.M. Herba, H.J. McQueen. Materials Science and Technology 14(9-10) (1998) 1057-1064. [41] H.J. McQueen, E. Fry, P. Cavaliere, Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing, 324 (2002) 157-161.
[42] C.I. Chang, C.J. Lee, J.C. Huang, Scripta Materialia, 51 (2004) 509-514. [43] X.Y. Yang, H.M. Miura, T. Sakai, Materials Transactions, 44 (2003) 197-203. [44] H.E. Hu, X.Y. Wang, L. Deng. Materials Science and Technology 30(11) (2014) 1321-1327. [45] B. Inem. Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing 197(1) (1995) 91-95. [46] X.J. Wang, K. Wu, H.F. Zhang, W.X. Huang, H. Chang, W.M. Gan, M.Y. Zheng, D.L. Peng. Materials Science and Engineering a-Structural Materials Properties Microstructure and Processing 465(1-2) (2007) 78-84. [47] F.J. Humphreys, P.N. Kalu. Acta Metallurgica 35(12) (1987) 2815-2829. [48] A. Belyakov, H. Miura, T. Sakai, Materials Science and Engineering aStructural Materials Properties Microstructure and Processing, 255 (1998) 139-147.