Functionally graded aluminum foam consisting of dissimilar aluminum alloys fabricated by sintering and dissolution process

Author’s Accepted Manuscript Functionally graded aluminum foam consisting of dissimilar aluminum alloys fabricated by sintering and dissolution process Yoshihiko Hangai, Tomoaki Morita, Takao Utsunomiya www.elsevier.com/locate/msea

PII: DOI: Reference:

S0921-5093(17)30528-2 http://dx.doi.org/10.1016/j.msea.2017.04.070 MSA34971

To appear in: Materials Science & Engineering A Received date: 10 March 2017 Revised date: 15 April 2017 Accepted date: 17 April 2017 Cite this article as: Yoshihiko Hangai, Tomoaki Morita and Takao Utsunomiya, Functionally graded aluminum foam consisting of dissimilar aluminum alloys fabricated by sintering and dissolution process, Materials Science & Engineering A, http://dx.doi.org/10.1016/j.msea.2017.04.070 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.

Functionally graded aluminum foam consisting of dissimilar aluminum alloys fabricated by sintering and dissolution process a*

a

Yoshihiko Hangai , Tomoaki Morita , Takao Utsunomiyab a

Faculty of Science and Technology, Gunma University, Kiryu 376-8515, Japan

b

Department of Mechanical Engineering, Shibaura Institute of Technology, Tokyo

135-8548, Japan *

Corresponding author. Tel: +81-277-30-1554. E-mail: [email protected]

Abstract Functionally graded (FG) aluminum (Al) foam, which consists of multilayers of different Al foams, is expected to exhibit higher functionality than ordinary uniform Al foam. In this study, uniform Al foams and two kinds of two-layered FG Al foams with different types of Al were fabricated by a sintering and dissolution process. From X-ray computed tomography (CT) inspection of the obtained foams, it was confirmed that NaCl was completely removed from the foams by dissolution. In addition, the FG Al foams in each layer had almost constant porosity (NaCl volume fraction, Vf) with seamless bonding between the layers. From the static compression tests of uniform foams, it was shown that the compression properties can be controlled by varying the type of Al, which is a similar tendency to the mechanical properties of the bulk materials. In addition, the compression properties can be controlled by varying Vf, regardless of the type of Al. From the static compression tests of FG Al foams, the foams exhibited multiple compression properties corresponding to the deformation of each layer for various Vf and different types of Al, which were similar to those of the corresponding uniform foams. In addition, the width of the plateau regions of FG Al foams can be controlled by controlling the height ratio between the layers. The advantage of varying the type of Al is that the mechanical properties of foams can be 1

controlled without changing their geometric structures. Therefore, FG Al foams with various Vf and types of Al are expected to enable the optimum design of foams used for structural materials.

Keywords : cellular materials; powder metallurgy; sintering; X-ray computed tomography; functionally graded materials

1. Introduction Aluminum (Al) foam is a lightweight Al exhibiting high impact energy absorption and good sound insulation[1, 2]. Functionally graded (FG) Al foam, which consists of multilayers of Al foams with different properties, is expected to exhibit higher functionality than ordinary uniform Al foam[3-6]. Until now, the fabrication of FG Al foams has been conducted by space holder routes[7, 8], an investment casting route using polyurethane foam precursors[9], a precursor foaming route[10-12] and a melt foaming route[13]. These FG Al foams were fabricated with varying density (porosity). In our previous studies, FG Al foams consisting of multilayers of Al foams with different types of Al were fabricated by the precursor foaming route[14-16]. From the compression tests of the FG Al foams, it was found that the deformation behavior and plateau stress can be controlled, i.e., they started to deform from the low-strength Al foam layer exhibiting low plateau stress, followed by the high-strength Al foam layer exhibiting high plateau stress. However, control of the pore structures during the foaming of FG Al foam consisting of different types of Al is difficult because the foaming temperature and foaming time are different for different types of Al owing to the different melting points. Zhao and Sun developed a sintering and dissolution process that can easily control the pore size and porosity[17, 18]. In this process, a mixture of Al powder as the base 2

material and sodium chloride (NaCl) powder as space holders is sintered, which is followed by leaching in water to dissolve the NaCl, resulting in Al foam. Hangai et al demonstrated that FG Al foam can be fabricated by the sintering and dissolution process changing the porosity[19, 20] and the amount of remaining NaCl[21]. In this study, uniform Al foams and two kinds of two-layered FG Al foams with different types of Al were fabricated by the sintering and dissolution process. Uniform Al foams were fabricated using pure Al, ADC12 Al-Si-Cu alloy and AC4CH Al-Si-Mg alloy. One of the FG Al foams consisted of a pure Al foam layer and an ADC12 foam layer (pure Al / ADC12 FG foam), which had almost constant porosity throughout the entire foam. The other FG foam consisted of a pure Al foam layer and an AC4CH foam layer (pure Al / AC4CH FG foam). In the pure Al / AC4CH FG foam, the porosity and the height ratio between the layers were also varied. X-ray computed tomography (CT) inspections of the fabricated foams were conducted to observe the porosity distributions in the FG foams and to confirm that NaCl was completely removed from the foams by dissolution. In addition, static compression tests were performed on the foams, and through the comparison of the test results for the uniform foams and FG foams, the compression properties of the FG foams are discussed.

2. Experimental procedure 2.1 Fabrication process Fig. 1 shows a schematic illustration of the fabrication process of the FG foam with different types of Al by the sintering and dissolution process. As shown in Fig. 1(a) and (b), pure Al powder with a particle diameter of approximately 20 μm was thoroughly mixed with NaCl powder with a particle diameter ranging from 425 μm to 500 μm (Mixture I), and ADC12 Al alloy powder with a particle diameter under 26 μm or AC4CH Al alloy powder with a particle diameter of approximately 25 μm was 3

thoroughly mixed with NaCl powder with the same particle diameter as in Mixture I (Mixture II). Fig. 2(a)-(d) show scanning electron microscope (SEM) images of pure Al, ADC12, AC4CH and NaCl powder particles, respectively. Table 1 shows chemical compositions of ADC12 and AC4CH. Next, as shown in Fig. 1(c), Mixture II was first placed in a graphite die, then Mixture I was placed in the die on top of Mixture II. Then, the layered mixture was sintered by spark plasma sintering (SPS) using an SPS-1050 spark plasma sintering system (Sumitomo Coal Mining Co. Ltd.). The SPS conditions were determined in accordance with previous studies, in which pure Al foam was fabricated by the sintering and dissolution process using SPS[22, 23], and the liquidus and solidus temperatures of the Al alloys[24]. The temperature was increased from room temperature to 743 K or 753 K in 5 min, then from 743 K or 753 K to 793 K in 2 min, then held at 793 K for 3 min before furnace cooling. The pressure during the SPS was 50 MPa. Next, as shown in Fig. 1(d), the as sintered mixture was placed in water to dissolve the NaCl. Finally, as shown in Fig. 1(e), cylindrical two-layered FG foam compression test specimens with a diameter of 20 mm and a height of 10 mm were obtained. Uniform pure Al, ADC12 and AC4CH foam compression test specimens were also fabricated only from Mixture I or II under the same SPS conditions.

2.2 X-ray CT inspection The obtained compression test specimens were subjected to nondestructive X-ray CT inspections by a cone-type, SMX-225CT microfocus X-ray CT system (Shimadzu Corporation). The X-ray tube voltage and current were 113 kV and 30 μA, respectively.

2.3 Compression tests The obtained specimens were subjected to static compression tests at a strain rate of 4

3.3 × 10-3 s-1 in accordance with Japanese Industrial Standards JIS H 7902[25] using an Autograph AG-100kNG universal testing machine (Shimadzu Corporation). The deformation behavior during the compression tests was recorded by digital video camera. The nominal stress and strain during the compression tests were obtained by dividing the load by the initial area and by dividing the displacement by the initial height, respectively. The load and displacement were recorded by a universal testing machine during the compression tests. 3. Results and discussion 3.1 Uniform foams Fig. 3(a)-(c) show the compression test specimens of uniform pure Al, ADC12 and AC4CH foams fabricated with a NaCl volume fraction of Vf = 70%. No collapse of the foam was observed for all three foams after the dissolution of NaCl. Namely, the Al particles in the cell walls of the foams were well sintered for all types of Al. Fig. 3(d)-(f) show cross-sectional X-ray CT images taken around the middle height of the foams shown in Fig. 3(a)-(c), respectively. White regions indicate cell walls and black regions indicate pores. It was found that the pores were distributed almost homogeneously in the foams, regardless of the type of Al. Also, it was confirmed that the NaCl was completely removed from the foams by dissolution. If some NaCl had remained in the foams, gray cloudlike regions would have been observed in the foams[21, 26]. Moreover, foams containing NaCl exhibit different compression properties from those with no NaCl[21, 26]. The complete dissolution of the NaCl was also confirmed from the weights of the foam before and after the dissolution process. Fig. 4(a)-(c) show the deformation behavior of each uniform foam with Vf = 70% during the compression tests. It was found that the foams exhibited ductile deformation behavior, and no collapse of the foams was observed regardless of the type of Al. Fig. 4(d) shows the relationship between the stress σ and strain ε corresponding to the 5

deformation behavior shown in Fig. 4(a)-(c). First, an elastic region appeared, then a plateau region with an almost constant σ (plateau stress[25], σpl) appeared for a wide range of ε, and finally a densification region appeared. These σ–ε curves are consistent with those of ordinary foams fabricated by the sintering and dissolution process[17, 23] and other fabrication processes[1, 27]. Insufficient sintering of the Al particles induces the brittle fracture of foams, resulting in an abrupt decrease in σ during a compression test[22]. Therefore, it was confirmed that the Al particles were well sintered under the SPS conditions used in this study. The ADC12 foam exhibited the highest σpl, followed by the AC4CH foam then the pure Al foam. The tensile 0.2% proof stresses of bulk pure Al, ADC12 Al alloy and AC4CH Al alloy are 15–20 MPa, 185 MPa and 95 MPa, respectively[24]. Therefore, this tendency of σpl for the uniform foams is similar to the mechanical properties of bulk Al. Fig. 4(e) shows the relationship between σ and the cumulative energy absorption per unit volume EV up to a specific σ. The low strength pure Al foam absorbed a larger amount of energy at a low σ than the high-strength Al alloy foams, while the high-strength ADC12 foam absorbed a larger amount of energy at a high σ than the low-strength pure Al and AC4CH foams. Therefore, it is expected that multiple compression properties can be realized within a single foam by combining these foams with different compression properties.

3.2 Pure Al / ADC12 FG foams Fig. 5(a) shows a representative compression test specimen of pure Al / ADC12 FG foam with Vf = 70%. The upper and lower layers were pure Al foam and ADC12 foam, respectively, and the height ratio between them was 1:1. The FG foams were well sintered, although the different types of Al, which had different melting points, were simultaneously sintered. No separation between the two layers was observed for any of the specimens. The compression direction was parallel to the h direction, where h is the 6

height from the bottom of the specimen as shown in Fig. 5(a). Fig. 5(b) and (c) show cross-sectional X-ray CT images of the pure Al and ADC12 foam layers of the specimen in Fig. 5(a) perpendicular to the h direction, respectively. The pores were distributed homogeneously in the foam. Also, it was confirmed that NaCl was completely removed from the foams. The complete dissolution of the NaCl was also confirmed by the weight of the foam after the dissolution process. Fig. 5(d) shows a cross-sectional X-ray CT image of the specimen in Fig. 5(a) parallel to the h direction. The boundary between the pure Al foam layer and the ADC12 foam layer, whose location is shown by a white arrow, cannot be observed. Fig. 5(e) shows the relationship between the porosity p and the height of the specimen h, which was evaluated from the X-ray CT images shown in Fig. 5(b) and (c). The figure shows that an FG foam with an almost constant porosity throughout the foam was obtained. A similar tendency was observed for all the FG foams fabricated in this study. Namely, it was confirmed that FG foams with two seamlessly bonded layers having constant porosities and different types of Al were obtained. Fig. 6(a) shows the deformation behavior of the pure Al / ADC12 FG foam shown in Fig. 5 during the static compression tests. The black lines at the middle height of the specimen indicate the initial boundary between the pure Al foam layer and the ADC12 foam layer. It was found that the upper pure Al foam layer first deformed. Thereafter, the lower ADC12 foam layer deformed along with the densification of the upper pure Al foam layer. Namely, the deformation behavior of the foam can be controlled by varying the type of Al, while the porosity was constant throughout the entire specimen. Fig. 6(b) shows the σ–ε curve corresponding to the deformation behavior shown in Fig. 6(a) along with the σ–ε curves of uniform pure Al and ADC12 foams modified from those shown in Fig. 4(d). The modified σ–ε curves were obtained in accordance with previous studies[15] to enable a simple comparison between the σ–ε curves of the FG foam and 7

the uniform foams. Several assumptions were made in modifying the σ–ε curves of the uniform foams. I: The deformation of the FG foam at a low ε only occurred in the pure Al foam layer. The elastic deformation of the ADC12 foam layer was neglected. II: The deformation of the FG foam at a high ε only occurred in the ADC12 foam layer. The densification of the pure Al foam layer was neglected. III: The ε value of the uniform ADC12 foam was offset to that at the beginning of the deformation of the ADC12 foam layer, which was determined from the deformation behavior of the FG foam. n The nominal strain of an FG foam 𝜀FG is generally defined as n 𝜀FG =

𝑥 ℎ0

where ℎ0 is the initial height of the FG foam and x is the displacement during the compression tests. From assumption I, the real strain applied to the pure Al foam layer modified of the FG foam 𝜀FG (pure Al) should be modified 𝜀FG (pure Al) =

𝑥 ℎ(pure Al)

=

ℎ0 ℎ(pure Al)

n × 𝜀FG

where ℎ(pure Al) is the initial height of the pure Al foam layer of the FG foam. Namely, the σ–ε curves of the FG foam at a low ε became ℎ0 /ℎ(pure Al) times wider in the ε direction. Here, instead of modifying the σ–ε curves of the FG foam, the nominal strain of the uniform pure Al foam was multiplied by ℎ(pure Al) /ℎ0 to modify the σ–ε curves of the uniform pure Al foam. From assumption II, the nominal strain of the uniform ADC12 foam was multiplied by the height ratio of the ADC12 foam layer in the FG foam ℎ(ADC12) /ℎ0 . In Fig. 6, ℎ(pure Al) /ℎ0 and ℎ(ADC12) /ℎ0 were 1/2. Furthermore, from assumption III, the strain was offset for the uniform ADC12 foam. From these modifications, it was found that two plateau regions appeared for the FG foam and each 8

plateau region exhibited similar curves to the modified σ–ε curves of the uniform foams with almost the same σpl. Table 2 shows the σpl for the three FG foams along with those of the corresponding uniform foams. The σpl values of the uniform foams were average stresses at ε of 20– 30% in accordance with reference[25]. The σpl values of the FG foams in each plateau region were obtained in accordance with previous studies[15]. Briefly, as described above, the ε value of each uniform foam was modified for easy comparison with the FG foam in each plateau region. The σpl values of the FG foam were obtained from the average stress corresponding to the modified strain of the uniform foam in the range originally 20–30%. Differences between the values of σpl in the first and second plateau regions were clearly observed. In addition, almost the same value of σpl as for the corresponding uniform foam was achieved in each plateau region. Fig. 6(c) shows the σ–EV curve of the FG foam along with the modified σ–EV curves of the uniform pure Al and ADC12 foams obtained from the modified σ–ε curves in Fig. 6(b). At a low σ, the pure Al foam layer of the FG foam mainly deformed and absorbed energy, therefore, the σ–EV curve of the FG foam at a low σ exhibited a similar tendency to that of the uniform pure Al foam. As σ increased, the ADC12 foam layer started to deform and mainly absorbed energy. Therefore, the σ–EV curve of the FG foam diverged from that of the uniform pure Al foam and exhibited a similar tendency to that of the uniform ADC12 foam. These results show that the FG foams with different types of Al exhibited multiple mechanical properties corresponding to the deformation of each layer, similar to the corresponding uniform foams. In addition, it was possible to compare the mechanical properties between FG foams and uniform foams by adopting some simple assumptions.

3.3 Pure Al / AC4CH FG foams 9

Although the ADC12 foam with Vf = 70% exhibited superior compression properties to the AC4CH foam with Vf = 70%, as shown in Fig. 4(d), AC4CH Al alloy exhibits superior corrosion resistance to ADC12 Al alloy. Fig. 7 shows the σpl value of each uniform foam with various Vf. It was found that AC4CH foams with Vf = 60% and 65% exhibited a higher σpl than that of ADC12 foam with Vf = 70%. It is expected that varying both Vf and the type of Al in FG foams will further expand their functionality. In addition, it is expected that the width of each plateau region of FG foams, which is related to the amount of energy absorption, can be controlled by varying the height ratio between the layers. In this section, the mechanical properties of pure Al / AC4CH FG foams with various Vf and height ratios between the layers, as shown in Table 3, were investigated. Fig. 8(a) shows a compression test specimen (Sample II). The upper and lower layers are pure Al foam and AC4CH foam, respectively. The FG foams were well sintered and no separation of pure Al and AC4CH foams was observed for all of the samples. The compression direction was parallel to h. Fig. 8(b) shows a cross-sectional X-ray CT image of the specimen in Fig. 8(a) parallel to the h direction. The boundary between the pure Al foam layer and the AC4CH foam layer, shown by a white arrow, can be clearly observed owing to the differences in Vf, but they were seamlessly bonded. Fig. 8(c) shows the relationship between the porosity p and the height of the specimen h, which indicates that an FG foam with almost constant porosity in each layer and different porosities in the two layers was obtained. Namely, it was confirmed that FG foams with seamlessly bonded layers having various values of Vf and types of Al were obtained. Fig. 9 and 10 show the σ–ε and σ–EV curves of the FG foam for samples I and II, respectively, along with the corresponding modified curves of the uniform foams. It was shown from the σ–ε curves that each layer exhibited similar compression properties to the corresponding uniform foams regardless of the value of Vf for each layer and the 10

height ratio. In addition, by controlling the height ratio, a wide first plateau region and a narrow second plateau region appeared for Sample I. In contrast, a narrow first plateau region and a wide second plateau region appeared for Sample II. According to the σ–EV curves, a larger amount of EV can be absorbed at a low σ in Sample I than in Sample II because a wide first plateau region was exhibited in Sample I. In contrast, EV for Sample II became higher than that for Sample I as σ increased because a wide second plateau region was exhibited in Sample II. From these results, it was found that the width of each plateau region and the resulting EV can be controlled by controlling the height ratio between the layers in the FG foams. Fig. 11 shows the σpl for the first and second plateau regions of Samples I – IV and those of the corresponding uniform foams. It was found that σpl for each plateau region can be controlled by controlling Vf. In addition, the σpl values for the first and second plateau regions are almost the same, corresponding to σpl for the uniform foams, regardless of Vf. Furthermore, by comparing Samples I and II, it can be seen that the variation of the height ratio has little effect on the σpl value of each layer. From these results, the FG foams exhibited multiple compression properties, which can be widely and precisely controlled by varying Vf, the type of Al and the height ratio between the layers while considering various other factors such as weight, geometric structures (density, pore size and pore distribution of the foams) and corrosion resistance. It is expected that FG foams consisting of multilayers exhibiting various properties can be realized by stacking several foams while controlling Vf, the type of Al and the height ratio. Such precise control of the properties layer by layer can be easily realized by applying the sintering and dissolution process.

4. Conclusion In this study, uniform Al foams and two-layered pure Al / ADC12 and pure Al / 11

AC4CH FG foams were fabricated by a sintering and dissolution process and their compression properties were investigated. From the experimental results, the following conclusions were obtained. (1) Uniform ADC12 foam exhibited the highest σpl, followed by the uniform AC4CH foam then the pure Al foam, for the same Vf, which is a similar tendency to the mechanical properties of the bulk materials. (2) Uniform pure Al foam absorbed the largest amount of energy at a low σ, whereas uniform ADC12 foam absorbed the largest amount of energy at a high σ. Namely, the compression properties can be controlled by varying the type of Al. (3) The compression properties of uniform foams can also be controlled by varying Vf, regardless of the type of Al. (4) Each layer of the FG foams had an almost constant Vf with seamless bonding between each layer. (5) The deformation behavior of FG foams can be controlled by varying Vf and the type of Al in each layer. (6) FG foam exhibited multiple compression properties corresponding to the deformation of each layer upon varying Vf and the type of Al, which were similar to those of the corresponding uniform foams. In addition, the width of the plateau regions can be controlled by controlling the height ratio between the layers. (7) The advantage of varying the type of Al is that the mechanical properties of foams can be controlled without changing their geometric structures. FG foams are expected to exhibit multiple properties in a single foam. Therefore, FG foams with various Vf and types of Al are expected to enable the optimum design of foams used for structural materials.

12

Acknowledgments The authors are grateful to Toyo Aluminium K.K., Japan, for providing the AC4CH alloy powder. Furthermore, the authors are grateful to The Salt Industry Center of Japan for providing the NaCl powder. This work was partly financially supported by JSPS KAKENHI Grant-in-Aid for Scientific Research (B) and grants from the Salt Science Research Foundation (No. 1616).

References [1] L.J. Gibson, Mechanical behavior of metallic foams, Annu. Rev. Mater. Sci. 30 (2000) 191-227. [2] J. Banhart, Manufacture, characterisation and application of cellular metals and metal foams, Prog. Mater. Sci. 46(6) (2001) 559-632. [3] G.Y. Sun, G.Y. Li, S.J. Hou, S.W. Zhou, W. Li, Q. Li, Crashworthiness design for functionally graded foam-filled thin-walled structures, Mater. Sci. Eng. A 527(7-8) (2010) 1911-1919. [4] M.S. Attia, S.A. Meguid, H. Nouraei, Nonlinear finite element analysis of the crush behaviour of functionally graded foam-filled columns, Finite Elem. Anal. Des. 61 (2012) 50-59. [5] X. Zhang, H. Zhang, Optimal design of functionally graded foam material under impact loading, Int. J. Mech. Sci. 68 (2013) 199-211. [6] Y. Zhang, M.H. Lu, G.Y. Sun, G.Y. Li, Q. Li, On functionally graded composite structures for crashworthiness, Compos. Struct. 132 (2015) 393-405. [7] A. Pollien, Y. Conde, L. Pambaguian, A. Mortensen, Graded open-cell aluminium foam core sandwich beams, Mater. Sci. Eng. A 404(1-2) (2005) 9-18. [8] A. Hassani, A. Habibolahzadeh, H. Bafti, Production of graded aluminum foams via powder space holder technique, Mater. Des. 40 (2012) 510-515. [9] A.H. Brothers, D.C. Dunand, Density-graded cellular aluminum, Adv. Eng. Mater. 8(9) (2006) 805-809. [10] R. Suzuki, K. Kitazono, Effect of graded pore distribution on thermal insulation of metal foam, J.

13

Japan Inst. Metals 72(9) (2008) 758-762. [11] Y. Hangai, K. Takahashi, T. Utsunomiya, S. Kitahara, O. Kuwazuru, N. Yoshikawa, Fabrication of functionally graded aluminum foam using aluminum alloy die castings by friction stir processing, Mater. Sci. Eng. A 534 (2012) 716-719. [12] Y. Hangai, K. Takahashi, R. Yamaguchi, T. Utsunomiya, S. Kitahara, O. Kuwazuru, N. Yoshikawa, Nondestructive observation of pore structure deformation behavior of functionally graded aluminum foam by X-ray computed tomography, Mater. Sci. Eng. A 556 (2012) 678-684. [13] S.-Y. He, Y. Zhang, G. Dai, J.-Q. Jiang, Preparation of density-graded aluminum foam, Mater. Sci. Eng. A 618 (2014) 496-499. [14] Y. Hangai, Y. Oba, S. Koyama, T. Utsunomiya, Fabrication of A1050-A6061 functionally graded aluminum foam by friction stir processing route, Metall. Mater. Trans. A 42(12) (2011) 3585-3589. [15] Y. Hangai, K. Saito, T. Utsunomiya, O. Kuwazuru, N. Yoshikawa, Fabrication and compression properties of functionally graded foam with uniform pore structures consisting of dissimilar A1050 and A6061 aluminum alloys, Mater. Sci. Eng. A 613 (2014) 163-170. [16] Y. Hangai, N. Kubota, T. Utsunomiya, H. Kawashima, O. Kuwazuru, N. Yoshikawa, Drop weight impact behavior of functionally graded aluminum foam consisting of A1050 and A6061 aluminum alloys, Mater. Sci. Eng. A 639(0) (2015) 597-603. [17] Y.Y. Zhao, D.X. Sun, A novel sintering-dissolution process for manufacturing Al foams, Scr. Mater. 44(1) (2001) 105-110. [18] D.X. Sun, Y.Y. Zhao, Static and dynamic energy absorption of Al foams produced by the sintering and dissolution process, Metall. Mater. Trans. B 34(1) (2003) 69-74. [19] Y. Hangai, T. Morita, S. Koyama, O. Kuwazuru, N. Yoshikawa, Functionally graded aluminum foam fabricated by friction powder sintering process with traversing tool, J. Mater. Eng. Perform. 25(9) (2016) 3691-3696. [20] Y. Hangai, A. Ishihara, T. Utsunomiya, O. Kuwazuru, N. Yoshikawa, Fabrication of two layers porous aluminum varying porosity by friction powder sintering process and its compression properties, J.

14

Jpn. Inst. Met. Mater. 80(6) (2016) 390-393. [21] Y. Hangai, K. Zushida, O. Kuwazuru, N. Yoshikawa, Functionally graded Al foam fabricated by sintering and dissolution process with remaining spacers, Mater. Trans. 57(5) (2016) 748-750. [22] M. Hakamada, Y. Yamada, T. Nomura, Y.Q. Chen, H. Kusuda, M. Mabuchi, Fabrication of porous aluminum by spacer method consisting of spark plasma sintering and sodium chloride dissolution, Mater. Trans. 46(12) (2005) 2624-2628. [23] M. Hakamada, T. Kuromura, Y. Chino, Y. Yamada, Y. Chen, H. Kusuda, M. Mabuchi, Monotonic and cyclic compressive properties of porous aluminum fabricated by spacer method, Mater. Sci. Eng. A 459(1-2) (2007) 286-293. [24] The-Japan-Institute-of-Light-Metals, Structures and properties of aluminum, The Japan Institute of Light Metals Tokyo, 1991. [25] JIS-H-7902, Method for compressive test of porous metals, Japanese Standards Association2016. [26] Y. Hangai, H. Yoshida, R. Yamaguchi, O. Kuwazuru, N. Yoshikawa, Relationship between amount of residual NaCl and compressive properties of porous Al/NaCl composites fabricated by sintering and dissolution process, Mater. Trans. 54(5) (2013) 850-853. [27] T. Miyoshi, M. Itoh, S. Akiyama, A. Kitahara, ALPORAS aluminum foam: Production process, properties, and applications, Adv. Eng. Mater. 2(4) (2000) 179-183.

15

(a)

(b) Pure Al and NaCl

Pure Al powder

Al alloy powder

NaCl powder

Mixture I

Base materials (c)

Al alloy and NaCl

Mixture II Mixing

(d)

(e)

Mixture I Mixture II

Spark plasma sintering

Leaching

FG Al foam

Fig. 1 Schematic illustration of fabrication of two-layered FG foam with differnt types of Al by sintering and dissolution process.

16

(a)

(b)

50 µm

50 µm (d)

(c)

50 µm

500 µm

Fig. 2 SEM images of powder particles: (a) pure Al, (b) ADC12 Al, (c) AC4CH Al and (d) NaCl.

17

Cross-sectional X-ray CT

Compression test specimen

Pure Al

Fig. 3

(a)

ADC12

AC4CH

(b)

10 mm

10 mm

(d)

(c)

(e)

10 mm

10 mm

(f)

10 mm

10 mm

Compression test specimens of uniform foams with Vf = 70% and

cross-sectional X-ray CT images taken around the middle height of the specimens.

18

(a)

(b)

(c)

20 mm Compression stress，σ/MPa

(d)

100 Uniform ADC12

80

Uniform AC4CH

Uniform pure Al

60 40 20 0 0

0.1

0.2

0.3 0.4 Compression strain，ε

0.5

0.6

0.7

Absorbed energy, Ev/MJ・m-3

(e) 20 Uniform ADC12 15

Uniform AC4CH

Uniform pure Al 10 5 0 0

10

20 30 40 Compression stress，σ/MPa

50

60

Fig. 4 Deformation behavior of (a) pure Al, (b) ADC12 and (c) AC4CH, and (d) σ–ε and (e) σ–EV curves of uniform foams with Vf = 70%.

19

(e) 10

(d)

Pure Al

Height，h/mm

(a)

h

ADC12

10 mm

10 mm

10 mm

8 6

4 2 0 40 70 100 Porosity，p(%)

Fig. 5

Pure Al / ADC12 FG foam with Vf = 70% : (a) compression test specimen (the

compression direction was parallel to h), (b) X-ray CT image of pure Al foam layer perpendicular to h direction, (c) X-ray CT image of ADC12 foam layer perpendicular to h direction, (d) X-ray CT image of compression test specimen parallel to h direction, (e) porosity distribution in h direction.

20

(a)

ε=0

ε = 0.1

ε = 0.2

ε = 0.3

ε = 0.4

ε = 0.5

ε = 0.6

Pure Al ADC12 20 mm

Compression stress，σ/MPa

(b)

100 Pure Al / ADC12 FG

80

Uniform pure Al (modified)

Uniform ADC12 (modified)

60 40

2nd plateau 1st plateau

20 0 0

Absorbed energy, Ev/MJ・m-3

(c)

0.1

0.3 0.4 Compression strain，ε

0.5

0.6

0.7

20 Pure Al / ADC12 FG 15

Uniform pure Al (modified)

Uniform ADC12 (modified) 10 5 0 0

Fig. 6

0.2

10

20 30 40 Compression stress，σ/MPa

50

60

(a) Deformation behavior and (b) σ–ε and (c) σ–EV curves of pure Al / ADC12

FG foam with Vf = 70%.

21

Plateau stress，σpl/MPa

50 ADC12 AC4CH Pure Al

40 30

20 10 0 55

60

65

70

75

80

85

Volume fraction of NaCl，Vf（%）

Fig. 7 Relationship between Vf and σpl for uniform foams.

22

(c) 10

Pure Al (b)

Height，h/mm

(a)

h

AC4CH

10 mm

8 6

4 2 0 40 60 80 100 Porosity，p(%)

Fig. 8

Pure Al / AC4CH FG foam (Sample II) : (a) compression test specimen, (b)

X-ray CT image of compression test specimen parallel to h direction, (c) porosity distribution in the h direction.

23

Compression stress， σ/MPa

(a)

100

Pure Al / AC4CH FG Uniform pure Al (modified) Uniform AC4CH (modified) 2nd plateau

80

60 40

20

1st plateau

0

0

(b) Absorbed energy, Ev/MJ・m-3

20

0.1

0.2

0.3 0.4 0.5 Compression strain，ε

0.6

0.7

Pure Al / AC4CH FG Uniform pure Al (modified) Uniform AC4CH (modified)

15 10 5 0

0

10

20 30 40 Compression stress，σ/MPa

50

60

(a) σ–ε and (b) σ–EV curves of pure Al / AC4CH FG foam (Sample I) along

Fig. 9

with those of corresponding uniform foams.

Compression stress， σ/MPa

(a)

100

Pure Al / AC4CH FG Uniform pure Al (modified) Uniform AC4CH (modified)

80

60 40

2nd plateau 1st plateau

20 0

0

Absorbed energy, Ev/MJ・m-3

(b)

20

0.1

0.2

0.3 0.4 0.5 Compression strain，ε

0.6

0.7

Pure Al / AC4CH FG Uniform pure Al (modified) Uniform AC4CH (modified)

15 10 5 0

0

Fig. 10

10

20 30 40 Compression stress，σ/MPa

50

60

(a) σ–ε and (b) σ–EV curves of pure Al / AC4CH FG foam (Sample II) along

with those of corresponding uniform foams.

24

Plateau stress，σpl/MPa

50

■ FG (1st plateau) ● FG (2nd plateau) □ Uniform pure Al ○ Uniform AC4CH

40 30 2nd plateau

20 10

1st plateau

0 0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5 I

Fig. 11

II

III

IV

σpl for Samples I – IV and corresponding uniform foams.

25

Table 1 Chemical compositions of ADC12 and AC4CH (mass%). Si

Fe

Mn

Mg

ADC12 9.6-12.0 <1.3 1.5-3.5 <0.5

<0.3

AC4CH 6.5-7.5 <0.2

Table 2

Cu

<0.2

Zn

Ni

Sn

<1.0 <0.5

Al

<0.2 Bal.

<0.1 0.25-0.45 <0.1 <0.05 <0.05 Bal.

σpl values of first and second plateau regions of three FG foams and

corresponding uniform foams.

FG (1st plateau) σpl

Uniform pure Al

(MPa)

FG (2nd plateau)

Sample A

Sample B

Sample C

10.4

11.2

11.3

10.8 24.7

Uniform ADC12

24.8

24.8

24.8

Table 3 Overview of pure Al / AC4CH FG foams with various Vf and height ratios between the layers.

Vf

Sample

Height ratio

No.

pure Al

AC4CH

pure Al : AC4CH

I

80%

60%

1:1

II

80%

60%

3:7

III

75%

65%

3:7

IV

70%

70%

1:1

26