2018 Volume 59 Issue 11 Pages 1854-1859
Aluminum (Al) foams are expected as components of vehicles and construction materials. Among the several routes for fabricating Al foams, a precursor foaming route has been commonly used to produce Al foam. In this study, optical heating, which can directly heat a precursor, was employed through light-transmitting materials of glass and sapphire to obtain ADC12 (Al–Si–Cu alloy) foam. Fundamental investigations on the forming of the ADC12 foam using a light-transmitting material as a die, which can transmit light during foaming, were conducted. From the free foaming of the ADC12 precursor by optical heating through the glass, through the sapphire and without those materials, it was found that, although a slight energy loss was observed, the ADC12 precursor can foam by optical heating through the light-transmitting materials. In addition, it was indicated that a similar foaming behavior was observed with the generally used electric furnace except that the foaming time was much shorter by the optical heating. Furthermore, from the foaming of the ADC12 precursor by optical heating through the light-transmitting materials, which restrict upward expansion, a flat ADC12 foam can be obtained. The pore structures similar to those of the free foaming were observed. In this forming process, cracks were observed in the case of the glass during the cooling of the ADC12 foam, which was not observed in the case of the sapphire. Therefore, it was indicated that sapphire can be used as the die for the forming of the ADC12 foam during the foaming.
Lightweight materials are expected as components of vehicles and construction materials. Aluminum (Al) foams are candidates for these materials owing to their ultralight weight, good energy absorption properties and sound insulation properties.1) Among the several routes for fabricating Al foams, a precursor foaming route has been commonly used to produce Al foam.2–4) In this process, a blowing agent powder is mixed into the solid Al by the powder metallurgy route,5) by the accumulative roll-bonding (ARB) route6) or by the friction stir welding (FSW) route.7) This solid Al containing blowing agent powder is called precursor. By heat treatment of the precursor around the melting point of the Al matrix, gases that are generated by the decomposition of the blowing agent expand the precursor and result in the Al foam.
In this study, we focused on the optical heating for the foaming of the precursor. Generally, the foaming of the precursor is conducted in an electric furnace,5,7) which is atmosphere heating. Namely, the precursor is heated by the heated air around the precursor. In contrast, the precursor can be directly heated by optical heating without heating the surrounding air. Therefore, it is expected that the precursor can foam in a short processing time in an environmentally friendly process. In addition, the precursor can be heated by light passing through light-transmitting materials such as glass and sapphire. Therefore, it is expected that light-transmitting materials can be used as a die for the forming of the Al foam during the foaming of the precursor by irradiating the light through the die. There are some investigations related to optical heating for the foaming of the precursor such as infrared lamp6,8,9) and concentrated solar energy.10) However, they are used as a heating device equivalent to an electric furnace, and a steel die is generally used. In contrast, there are few papers related to the use of light-transmitting materials as the die for the forming of Al foam during the foaming.
In this study, first, the foaming behavior of the precursor in the generally used electric furnace was directly observed for the purpose of obtaining the precise foaming temperature of the Al alloy precursor used in this study. In addition, the foaming behavior of the precursor of the Al foam observed by optical heating was compared with that observed with the generally used electric furnace to confirm that there is little difference between them. Second, the precursor was free foamed by the optical heating through the light-transmitting materials and without them to investigate whether the use of light-transmitting materials during the optical heating has a negative effect on the foaming behavior of the precursor and pore structures of the obtained Al foams. Finally, the upward expansion was restricted by the light-transmitting materials during the foaming and flat Al foams were obtained. The pore structures of the obtained Al foam were observed by X-ray computed tomography (X-ray CT) to confirm whether the pore structures were retained during the forming of the Al foam. From these experiments, a fundamental investigation related to the use of light-transmitting materials as the die for forming Al foam by the optical heating of the precursor through the die during the foaming was conducted.
Figure 1 shows the setup for the direct observation of the foaming in the electric furnace. A tube was set in a hole previously made in the wall of the electric furnace. The foaming behavior was recorded with a video camera through the tube. The precursor was set on a ceramic plate. A steel spacer was placed under the ceramic plate to adjust the height of the precursor to allow the precursor to be observed through the tube. The specification of the rate of the temperature increase of the atmosphere in the electric furnace used in this study was approximately 50 K/min. The temperature of the precursor during the foaming was directly measured using a K-type thermocouple placed in a hole of the precursor previously drilled at midheight and 3 mm inside from the side surface of the precursor.
Schematic illustration of direct observation of foaming of precursor in an electric furnace.
ADC12 (Al–Si–Cu) alloy (equivalent to A383.0 Al alloy) precursors were prepared in accordance with Ref. 11), 12). Solidus and liquidus temperatures of ADC12 are 788 and 853 K, respectively.13) Titanium (II) hydride (TiH2, <45 µm, 1 mass%) powder as a blowing agent and α-alumina (Al2O3, ∼1 µm, 5 mass%) powder as a stabilization agent were used. The precursors with dimensions of 15 × 15 × 6 mm3 were obtained.
2.3 Setup for free foaming of precursor by optical heating through light-transmitting materialsFigure 2 shows the setup of the free foaming of the precursor by optical heating through light-transmitting materials. The precursor was set on the ceramic honeycomb. The light-transmitting material was set 25 mm above the surface of the ceramic honeycomb by using two square steel pipe spacers to prevent contact with the foamed precursor during the foaming. The temperature was measured at the center of the precursor using a K-type thermocouple. The thermocouple was inserted into the hole previously drilled at a 3 mm depth at the bottom center of the precursor.
Schematic illustration of free foaming of ADC12 foam precursor by optical heating through a glass or sapphire.
Figure 3 shows the setup of the foaming of the precursor by optical heating through a light-transmitting material that restricts the upward expansion of the precursor. The precursor was set on the ceramic honeycomb. The light-transmitting material was set 10 mm above the surface of the ceramic honeycomb by using two square steel spacers to allow contact with the foamed precursor during the foaming. The same two square steel spacers were placed on the edge of the light-transmitting material to prevent the light-transmitting material from moving upward owing to the foaming force during the foaming. The upward expansion of the foamed precursor is expected to be restricted by the light-transmitting material, and the upper part of the foam will become flat owing to the smooth surface of the light-transmitting material. The temperature was measured at the center of the precursor using a K-type thermocouple. The thermocouple was inserted into the hole previously drilled at a 3 mm depth at the bottom center of the precursor.
Schematic illustration of foaming of ADC12 foam precursor by optical heating through a glass or sapphire die with upward expansion restricted by the die.
The distance from the surface of the precursor to the lamp was set at 40 mm throughout the experiments. Four halogen lamps were used. The voltage and current applied to the four lamps were 200 V and 10 A, respectively, so as not to exceed the specification of the optical heating equipment used in this study and for the precursor temperature to exceed 903 K. The lamps were turned off when the precursor temperature reached 903 K, then, cooled to room temperature.
A BOROFLOAT® glass (borosilicate float glass, Edmund Optics Inc.) of round shape with 50 mm diameter and 3.3 mm thickness and a sapphire (Edmund Optics Inc.) of round shape with 50.8 mm diameter and 3.15 mm thickness were used as the light-transmitting materials.
2.6 Pore structures observation by X-ray CT inspectionThe pore structures of the foamed samples were nondestructively observed by X-ray CT using an SMX-225CT microfocus X-ray CT system (SHIMADZU Corporation) at room temperature. The X-ray source was tungsten. Cone-type CT was employed. The X-ray tube voltage and current were 80 kV and 30 µA, respectively.
Figure 4(a) shows the relationship between the temperature of the precursor T and the foaming time t, along with the solidus and liquidus temperatures of ADC12. t = 0 was defined as the time when the temperature of the electric furnace started to increase. Figure 4(b) shows an enlargement of the T–t relationship in the region of dashed box A in Fig. 4(a). The rate of temperature increase of the precursor decreased at around the midpoint between the solidus and liquidus temperatures owing to the latent heat of melting when the solid ADC12 became liquid ADC12. After the temperature reached the liquidus temperature, it increased rapidly. Figures 4(c)–(g) show the sequential foaming behavior extracted from the video, corresponding to (c)–(g) in Fig. 4(b). As shown in Fig. 4(c), no foaming was observed from the surface when the temperature of the precursor reached the solidus temperature of the ADC12. As the rate of the temperature increase decreased at around the midpoint temperature in the solid-liquid coexistent region and the temperature became almost constant, slight foaming of the precursor began and then gradually proceeded, as shown in Fig. 4(d). After the temperature of the precursor reached the liquidus temperature, the precursor foamed vigorously for a short time, as shown in Figs. 4(e)–(g). The foaming was stopped (Fig. 4(g)) when the gases generated by the decomposition of the blowing agent started to be released from the surface of the precursor and pores began to burst. Further foaming may cause the generated pores to coalesce and rise upward, and drainage may occur at the bottom of the precursor. From these results, it was found that the foaming of the precursor mainly occurred after the temperature of the precursor exceeded the liquidus temperature. These foaming experiments were conducted four times, and all the samples exhibited similar tendencies. These foaming behaviors were consistent with the results for 6063 Al–Mg–Si alloy and A4042 Al–Si alloy.14)
(a) Relationship between temperature of the precursor T and foaming time t. (b) Enlarged T–t relationship in area of dashed box A in (a). (c)–(g) Sequential foaming behavior corresponding to (c)–(g) in (b).
Figure 5 shows the foamed samples along with their cross-sectional X-ray CT images and porosities obtained by Archimedes’ principle. The foaming was stopped by removing the precursors from the electric furnace at T = 793 K, 843 K, 863 K, 903 K and 953 K to clearly observe their pore structures during the foaming. Note that these samples were foamed from different ADC12 precursors fabricated as described in section 2.2. The above temperatures were around and above the solidus and liquidus temperatures in accordance with the results shown in Fig. 4. The X-ray CT images correspond to vertical cross sections, i.e., the vertically downward direction was the gravitational direction during the foaming. The white and black regions indicate ADC12 and pores, respectively. The foaming behaviors of these samples were similar to those indicated in Figs. 4(c)–(g) observed using a video camera. As shown in Fig. 5(a), no foaming was observed at the surface of the sample immediately after the temperature of the precursor reached the solidus temperature of the ADC12, but a small number of pores were observed as cracks in the X-ray CT image, as shown in Fig. 5(b). As shown in Figs. 5(c) and (d), foaming started in the entire precursor at temperatures within the solid-liquid coexistent region, but many of the pores were cracklike. As shown in Figs. 5(e)–(j), as the temperature reached and exceeded the liquidus temperature, the pores rapidly increased in size because the Al matrix was easily deformed by the weak foaming force and the pores tended to become round owing to surface tension. These results indicated that although the foaming of the precursor started after the temperature of the precursor exceeded the solidus temperature, the temperature of the precursor must be raised to above the liquidus temperature of the ADC12 to obtain high porosity with round pores in the ADC12 foam. Note that the ADC12 foam shown in Fig. 5(i) has a through-hole on the upper surface, which is considered to be where a pore burst. Therefore, in the experiments on optical heating described below, the foaming was stopped at the precursor temperature of 903 K.
Foamed samples along with their cross-sectional X-ray CT images and porosities p (%). The foaming was stopped at certain temperatures during the foaming.
Figures 6–8 show the foaming behavior of the precursor by optical heating without a light-transmitting material, through glass and through sapphire, respectively. Figures 6(a), 7(a) and 8(a) show the initial precursor before the start of foaming; Figs. 6(b), 7(b) and 8(b) show the precursor during the foaming between the solidus and liquidus temperatures described below; and Figs. 6(c), 7(c) and 8(c) show the foaming of the ADC12 foam just before the halogen lamp was turned off and the foaming stopped. The porosities p (%) of the ADC12 foam obtained by Archimedes’ principle were 77.0%, 74.4% and 73.3%, respectively, which were similar for each precursor. It was shown that ADC12 foam with similar porosity to that obtained using the electric furnace can be obtained by optical heating. In addition, the precursor can be similarly foamed by optical heating through both the light-transmitting materials and without a light-transmitting material.
(a)–(c) Foaming behavior of precursor by optical heating without light-transmitting materials, and (d) cross-sectional X-ray CT image of obtained ADC12 foam in vertical direction (p = 77.0%).
(a)–(c) Foaming behavior of precursor by optical heating through glass, and (d) cross-sectional X-ray CT image of obtained ADC12 foam in vertical direction (p = 74.4%).
(a)–(c) Foaming behavior of precursor by optical heating through sapphire, and (d) cross-sectional X-ray CT image of obtained ADC12 foam in vertical direction (p = 73.3%).
Figure 9 shows the relationship between the temperature of the precursor T and the foaming time t, during the foaming of the precursor by optical heating. For the foaming without a light-transmitting material, T increased until it exceeded the solidus temperature, after which it was almost constant. The foaming started after T exceeded the solidus temperature, then foaming gradually proceeded as shown in Fig. 6(b). Then, T increased to around the liquidus temperature. This rapid increase in T induced the vigorous foaming of the precursor as shown in Fig. 6(c). The foaming behavior and T–t relationship obtained by the optical heating are consistent with those of the foaming in the electric furnace, as described in section 3.1, except for the short foaming time in the case of optical heating. In addition, similar tendencies can be observed for the T–t relationship and foaming behavior of the precursors with and without using light-transmitting materials. The only difference between them is that the rate of increase of T for the T–t relationship was lower using the light-transmitting materials than without using a light-transmitting material. This is considered to be due to the small amount of reflection and absorption of the light by the light-transmitting materials. Therefore, the foaming of the ADC12 foam was delayed for about 1 min when using the light-transmitting materials. Furthermore, similar tendencies can be observed for the T–t relationship and foaming behavior of precursors for the use of glass and sapphire.
Relationship between foaming time t and temperature of precursor, T.
Figures 6(d), 7(d) and 8(d) show the cross-sectional X-ray CT images of the obtained ADC12 foams in the vertical direction. The pore structures are homogeneously distributed, which indicated that little drainage and little coalescence of pores occurred during the foaming owing to the rapid heating and rapid cooling during the optical heating. The foaming of the precursor by optical heating through the glass, through the sapphire and without a light-transmitting material was conducted twice, and similar behaviors were observed.
From these results, although a slight energy loss occurred in the case of optical heating through the light-transmitting materials, the precursor can be foamed by optical heating through the light-transmitting materials. Note that the porosity of the ADC12 foam that was foamed in the electric furnace was lower than that of the foam subjected to optical heating. This is considered to be due to the slow heating and cooling of the atmosphere in the electric furnace, which induced pore collapse. In contrast, the optical heating enabled rapid foaming and cooling before the pores collapsed. However, it is necessary to investigate the effect of the heating rate of the precursor on the pore structures in a future study.
3.3 Forming of ADC12 foam using light-transmitting materials as die during foaming of precursorFigures 10 and 11 respectively show the foaming behavior of the precursor induced by optical heating with glass and sapphire dies restricting the upward expansion. Figures 10(a) and 11(a) show the initial precursor before the start of foaming. There is a gap between the precursor and each light-transmitting material. As the foaming started, the foamed precursor came in contact with each light-transmitting material as shown in Figs. 10(b) and 11(b). Figures 10(c) and 11(c) show the ADC12 foam just before the halogen lamp was turned off and the foaming stopped. Although expansion in the vertical direction was prevented by the light-transmitting material, expansion occurred in the horizontal direction. The ADC12 foam cooled while remaining in contact with the light-transmitting material. A flat ADC12 foam was thus obtained for both types of light-transmitting material. However, cracks occurred in the glass during the cooling after foaming, which is considered to be due to the difference in the coefficient of linear expansion between the glass and the Al alloy along with the low heat resistance of the glass. In addition, in some cases, the ADC12 foam and glass bonded and the ADC12 foam could not be peeled off from the glass. This is considered to be due to the strong bonding between the new surface of Al and the glass when the oxide film surface of the precursor broke upon coming in contact the glass. In contrast, although the ADC12 foam was in tight contact with the sapphire in a small region, it could be peeled off by pulling the ADC12 foam and no cracks occurred on the sapphire because it is generally difficult for sapphire and Al alloy to bond. It is considered that coating the sapphire with a small quantity of releasing agent before foaming may prevent the ADC12 foam from coming in tight contact with the sapphire.
Foaming behavior of precursor by optical heating with upward expansion restricted by a glass die.
Foaming behavior of precursor by optical heating with upward expansion restricted by a sapphire die.
Figure 12 shows photographs, cross-sectional X-ray CT images in the vertical direction and porosities p (%) obtained by Archimedes’ principle for the flat ADC12 foams. It was shown that a flat surface can be observed where the ADC12 foam was in contact with each light-transmitting material. In addition, as shown in Figs. 12(c) and (e), pore structures similar to those obtained by free foaming shown in Figs. 6(d), 7(d) and 8(d) can be observed. There were no thick dense parts at the surface caused by the collapse of the pores due to the contact with the light-transmitting materials, and the restriction of the expansion was observed. Furthermore, the flat ADC12 foams had almost the same porosities as the ADC12 foams obtained by free foaming. These results show that the forming of the ADC12 foam during foaming had little effect on the pore structures of the obtained ADC12 foams. These results also show that glass is not suitable whereas sapphire is suitable for use as a die in the forming of ADC12 foam during foaming.
Flat ADC12 foams along with their cross-sectional X-ray CT images and porosities p (%).
In this study, optical heating, which can directly heat a precursor, was employed for the foaming of the precursor to obtain ADC12 foam. In addition, light-transmitting materials, which can transmit light during the foaming, were used for the forming of the ADC12 foam. The experimental results led to the following conclusions.
The authors thank Mr. T. Ishida, Mr. S. Ishida and Mr. T. Nagata, PVR, for their great support in conducting optical heating.
