Foaming Behavior and Pore Structure Evolution of Foamed Aluminum under the Extrusion Constraint

-is study evaluates foaming behavior and pore structure evolution of aluminum foam by cross-sectional image processing, which was prepared by the powder compact melting (PCM) method under the constraint. -e results revealed that the heating time and the heating rate are the key factors affecting the foaming behavior and pore structure of aluminum foam. -e thermal decomposition of the foaming agent and aluminum melting behavior affect the foam expansion under different heating times and heating rates.-e evolution of the pore structure during the foaming process includes pore formation, small-sized spherical pores, large-sized spherical pores, polygonal pores, merging, and collapse. Due to the limitation of the pipe wall, the maximum expansion height and porosity are constrained, and the macrostructure of aluminum foam is improved.


Introduction
As the representative of metal foam materials, aluminum foam has the advantages of low density, strong impact absorption, high temperature resistance, flame retardancy, corrosion resistance, sound insulation and noise reduction, low thermal conductivity, excellent weather resistance, and good electromagnetic shielding performance. Aluminum foam products have the advantages of convenient processing and installation, high molding precision, surface coating, reusability, recycling, and so on, and have many potential applications. erefore, they have attracted wide attention in science and industry [1][2][3][4][5][6][7][8].
With the establishment of the research and development center for aluminum foam by Ethyl Corporation in the 1960s, aluminum foam has been successfully applied in sound absorption, noise reduction, and collision prevention including carriages and containers, bumpers and silencers, thermal insulation walls and fire doors, and energy-saving movable houses [9][10][11]. However, as structural materials with most application potential, relatively speaking, it has not made much progress. e aluminum foam-filled pipe exhibits lightweight characteristics and high specific stiffness, as well as excellent vibration damping and damping properties. e filling structure solves the issue of low strength when completely utilizing the characteristics of aluminum foam materials [12][13][14][15], considerably extending the applications of aluminum foam. e aluminum foam-filled hollow pipe prepared by the powder compact melting (PCM) extrusion method exhibits advantages of a simple process, facile operation, and facile adaptability to irregular parts with different structures; hence, aluminum foam has become attractive for research both domestically and internationally [7,[16][17][18][19][20]. Internationally, the process has developed rapidly and entered the stage of large-scale industrial production. On the contrary, in domestic countries, although some progress has been made, the preparation process is still the most basic problem. In [21,22], precursor samples have been typically prepared by cold pressing, hot pressing, and extrusion. e results revealed that a more dense, uniform precursor structure is prepared by the extrusion method. In the study, aluminum foam inside the pipe was prepared by the PCM method by extrusion, and the foaming behavior and pore structure evolution of aluminum foam in the hollow pipe were investigated to clarify the relation variation and formation mechanism among the process conditions, sample temperature, foaming behavior, and pore structure, which laid a foundation for the further development and application of aluminum foam.

Preparation of the Foamable Precursors.
e first step of the phase-change material method is to produce foamable precursors. It mainly includes three steps: (1) powder mixing: the two powders were evenly mixed in a four-head planetary ball miller, that is, Al (99.5 wt.%, technically pure (99.9%), and 300 mesh) and powdery foaming agent TiH 2 (0.5 wt.%, 99.6%, and 300 mesh). (2) Cold pressing: the mixture was cold pressed into a cylindrical preform of 934 mm and 40 mm by a servo hydraulic press at 300 MPa so that the density of the precursor is greater than 90% of the bulk density. (3) Hot extrusion: the preform was placed in the corresponding mold and heated to 500°C and kept for 25 min. Under the pressure of 250 MPa applied by the hydraulic press, the cylindrical preform was transformed into a cylindrical belt with a size of Φ10 mm and 300 mm and was called a foamable precursor. en, the small granulates of Φ10 mm and 50 mm were cut by a wire electrical discharge machining (WEDM) system (Figure 1(a)).

Production of the Aluminum Foam-Filled Stainless Pipe.
e second step of the PCM method is foaming the foamable precursor. e constrained foaming method was employed for the foaming of the precursor samples in the hollow pipe. (4) Heating the pipe: the small precursor (Figure 1(a)) was put in the hollow pipe at 760∼800°C for 6∼7 min in a horizontal resistance furnace (Figure 1(b)). (5) Foaming: in decomposition of TiH 2 , the precursor foams to fill the pipe. (6) Accelerated cooling: it was taken out and rapidly cooled using an atomizing spray. e effects of the foaming process and foaming of the same sample at different heating rates (furnace temperatures) were investigated. e precursor samples were put into a hollow steel pipe, heated, and foamed at a specific furnace temperature. In order to compare the foaming characteristics and filling compounding of foamed aluminum in the steel pipe with different initial cross-section ratios, the hollow pipes used have two diameters. Table 1 summarizes the detailed process plan.

Experimental Results
Two thermocouples were used to measure the furnace temperature and the temperature of the sample, and the temperature and time were recorded by a data acquisition instrument and a computer (Figure 1(b)). e pictures were taken by the digital camera, and the foaming process of the sample was recorded in real-time penetrating quartz glass, and then the expansion height (and expansion height ratio) was calculated. Expansion height (H) is the height of the maximum expansion of each foaming time sample; expansion height ratio (H r ) is the ratio of the maximum expansion height of the sample to the original diameter of the sample during the foaming time.

Expansion and Foaming Behavior of the Sample at Different Furnace
Temperatures. Figure 2 shows the variation in the temperature and expansion height of the sample in pipe A as a function of the heating time at furnace temperatures of 760°C and 800°C (referred to as the temperature riseexpansion curves). Figure 3 shows the temperature riseexpansion curves of the sample in pipe B at furnace temperatures of 740°C and 760°C. Figure 4 shows the ten stages of foaming behavior of the sample in pipe A at a furnace temperature of 800°C, where (a) not foaming; (b)-(d) early expansion; (e) and (f ) pore growth; (g) filled the cross section of the steel pipe; (h) expanding to the maximum point; and (i) and (j) the pores merge and collapse.

Evolution of the Sample Pore Structure at Different Furnace Temperatures.
According to the heating curves, five to six heating time points are selected from each furnace temperature, and five to six substantially identical samples are heated to the corresponding time points, rapidly removed, cooled, cut, and photographed. Figure 5 shows the pore structure evolution of the sample in pipe A at a furnace temperature of 800°C. For the samples in pipe A obtained at different furnace temperatures, Figures 6 and 7 summarize their variations in the cross-sectional photographs, the corresponding binarization-treated diagram, porosity P 0 , average equivalent-circle diameter D a , and average circularity C a at different temperatures. e situations in pipe B are similar to these and will not be described again.

Effects of Heating Time (Sample Temperature).
For the experiments with the aluminum foam-filled composite pipe, the complete filling of a steel pipe constitutes a condition that must be satisfied. erefore, the samples listed in Table 2 are all the samples that have achieved the full filling condition (sample temperatures >700°C). Although there is no significant difference in the expansion and foaming behavior of each sample, the porosity and pore structure still change with temperature. According to Figures 2 and 3, the relationship between the expansion height ratio and temperature of the sample is plotted and is shown in Figure 8. e combination of Figures 2-5 revealed that the sample temperature increases with the extension of the heating time (when the furnace temperature is constant, the two values correspond to each other) and comprises three stages: temperature rise, melting plateau, and second temperature rise. e expansion curve comprises several stages, including zero expansion (before the temperature plateau); slow expansion (plateau zone); rapid expansion (after the temperature plateau); maximum value; and collapse. e pore structure in the foaming process has experienced a stage evolution process, including pore formation, small-sized spherical pores, large-sized spherical pores, polygonal pores, merger, and collapse.
It can be seen from Figures 6 and 7 that when the foaming process is constrained, the pore structure continues to change with increasing temperature after the foam completely fills the steel pipe. erefore, the porosity has experienced the process of rising; maximum value; and reduction. However, the pore size shows two processes: regular growth of a single pore and merge growth. e average diameter is constantly increasing, the circularity of the pores tends to decrease, but the change rule is unclear. During the process of merger and growth, serious deterioration occurred.
Compared with the results reported previously [21], during the foaming process under the in-pipe constraint, the heating rate of the sample is greater than that of free foaming, and the sample temperature can reach or exceed the furnace temperature, probably as the heat released during the melting of the sample cannot dissipate on time.
Owing to the limitation of the pipe diameter, the maximum expansion height and porosity are constrained, while the pore size, shape uniformity, and circularity of pores are better than those during free foaming. In addition, the transverse expansion process is observed after the steel pipe is filled with the sample. erefore, after full filling, the porosity continues to increase. e combination of Figures 6-8 indicated that, in the case of constrained foaming, the foaming effect with a porosity range of 70%-79% and different pore structures can be obtained at a temperature range of 700°C to 750°C, where the foaming effect is optimum for a certain temperature. e range of foaming temperature, the best foaming temperature, the    In-pipe constrained foaming and free foaming are similar [21]. Different furnace temperatures result in different heating curves of samples. e higher the furnace temperature, the earlier the moment to enter the plateau and the shorter the plateau length. On the contrary, the lower the furnace temperature, the later the time to enter the plateau and the longer the plateau length; however, the difference in the plateau length is less than that of free foaming. At each furnace temperature, the sample can fill with steel pipes. For the same steel pipe, the expansion rate is the same, and the temperature of the sample filled with the pipe is basically the same. e higher the furnace temperature, the larger the  maximum porosity of the sample. ere is little difference in porosity, and the maximum porosity temperature is basically the same. e collapse temperature of the sample at different furnace temperatures is basically the same. e heating rate of the in-pipe constrained foaming is significantly greater than that of the free foaming method [21] at the same furnace temperature, but the porosity at the same sample temperature is comparable to or slightly less than that of free foaming. e heating rate is greater than that of free foaming, possibly related to the homogenizing effect and heat insulation of the steel pipe, and the slightly lower porosity may be related to the inhibition effect of the steel pipe wall. Due to the influence of the steel pipe on the heating rate, the furnace temperature is no longer the only factor affecting the heating rate. On the contrary, the steel pipe also exhibits a certain inhibition on the decomposition and hydrogen evolution of TiH 2 .
erefore, the effect of furnace temperature on the expansion and foaming characteristic parameters during the in-pipe constrained foaming process decreases, but the effect of the sample temperature is more significant. e in-pipe constrained foaming process can achieve foaming at lower furnace temperatures. e temperature at which the sample begins to collapse at different furnace temperatures is basically consistent under the in-pipe constraint, and it is basically the same as that in the free foaming process. e result indicated that the sample temperature and viscosity play essential roles during the collapse. e comparison between pipe B and pipe A indicated that the lower section ratio of pipe B makes the sample completely fill the steel pipe earlier, indicative of a better filling effect and a lower temperature for complete filling. However, due to the great resistance caused by the pipe wall, the maximum porosity is also low.  Figure 5: Expansion and foaming process of aluminum foam and its pore structure evolution.

Effects of the Heating Rate on the Pore Structure.
With different furnace temperatures, Figures 6 and 7 show the section photographs, corresponding to the binarizationtreated images, pore structure parameters, and its distribution of the sample in pipe A at different sample temperatures.
As can be observed in Figure 9 and Table 3, the heating rate significantly affects the porosity and equivalent diameter. e general trend is that high heating rates lead to high porosity and a large equivalent diameter, but the difference is smaller than that during the free foaming process [21]. However, pipe B exhibits a different situation. e porosity at 740°C is greater than that at 760°C, but the equivalent diameter is lower possibly because the smaller diameter of pipe B exerts a higher impact on the foaming process. e effect of the heating rate on the average circularity is marginal; hence, the range and maximum value of circularity at different heating rates (furnace temperatures) are almost identical. As can be observed from the comparison between pipes B and A, under the same conditions, pipe B shows lower porosity and lower equivalent diameter, which may be due to the greater resistance of the pipe wall caused by the lower diameter of pipe B.

Analysis of the Mechanism.
In the preparation of aluminum foam by the PCM method, the main foaming mechanism includes the simultaneous melting of aluminum and the decomposition of foaming agent TiH 2 . e higher the pyroconductivity of the matrix material and pipe is, the faster the melting temperature of the matrix material is reached at earlier time. e decomposition of TiH 2 can form hydrogen gas and corresponds to endogenous foaming. e foam volume changes with time, and the whole foaming   process is a dynamic nonequilibrium process, including formation and termination. e factors that may lead to volume changes are summarized. e possible factors affecting the volume increase/loss of foamed aluminum are listed in Table 4. Figures 2 and 3 show the foaming behavior, the expansion and foaming process, and the pore structure evolution of the aluminum foam sample prepared by the PCM method, respectively. From the point of view of physical and chemical processes, the foaming process includes the decomposition of the TiH 2 foaming agent, the accumulation and nucleation of gas molecules, and the growth, variation, and collapse of pores. TiH 2 will decompose and release gas when heated to a certain temperature. Hydrogen decomposed by TiH 2 is dissolved in the aluminum liquid in an atomic form at first. After reaching saturation, hydrogen will evolve as a gas and then nucleate, agglomerate, and grow in the matrix material. erefore, pores are formed in the interior of the matrix material. Calculations and experiments indicated that loose TiH 2 can spontaneously decompose when T s is greater than 450°C [12]. With increasing temperature, the decomposition rate of TiH 2 increases correspondingly. erefore, the decomposition amount of TiH 2 increases with time and temperature. After reaching a certain temperature and time, the TiH 2 content decreases; hence, the decomposition rate of TiH 2 gradually decreases until the decomposition is terminated. Within the experimental temperature and time range, TiH 2 was not completely decomposed [12].
As hydrogen gas forms pores and pores grow in liquid metal, the pressures experienced include air pressure, metal static pressure, the additional pressure generated by surface tension, and the resistance caused by the viscosity of liquid metal. erefore, the nucleation and growth of pores must comply with the following equation [24]: where P is the pressure inside hydrogen gas, Pa; P 1 is the air pressure, constant, Pa; ρ is the density of liquid, kg/m 3 ; g is the acceleration of gravity, m/s 2 ; h is the length of liquid above the pores, m; σ is the liquid-gas surface tension (energy), N/m; r is the pore diameter, m; and F is the viscous force of liquid metal. e force originates from the viscosity. erefore, it exhibits a clear impact at low temperatures, and its effect rapidly decreases with increasing temperature.
Meanwhile, the growth of gas inside pores also conforms to the state equation of an ideal gas at a constant temperature: where P is the internal pressure of hydrogen gas, Pa; V is the volume of hydrogen gas inside foams, m 3 ; n is the moles of hydrogen gas, mol; R is the gas constant; and T is the temperature, K. e volume variation during foaming can be considered as the sum of volume gain (VG) and volumetric loss (VL), which can be expressed as follows: where V HP denotes the expanded gas volume caused by hydrogen gas (HP) produced by the decomposition of TiH 2 ; V GE denotes the expansion caused by the rising temperature of the sample (T s ); V HD represents volume loss caused by hydrogen diffusion and overflow; and V BR indicates the volume reduction loss caused by surface bubble fracture.
Combined with the data summarized in Table 3, the sample temperature affects the foaming process by affecting the decomposition rate on the TiH 2 foaming agent, the viscosity of the sample, and the expansion of gas. TiH 2 begins to decompose at temperatures greater than 400°C and rapidly decomposes at temperatures greater than 600°C. As the temperature is less than the melting point (660°C), the sample is still in the solid state, so it is unlikely to expand and foam. As the temperature reaches the melting point, the sample begins to melt, and its viscosity decreases. As can be observed from formula (1), if the hydrogen pressure is greater than the external pressure, the sample begins to expand, and foaming is observed; however, within the plateau zone of the temperature curve of the sample (heating time >100 s), the temperature changes slightly, and the viscosity slowly increases due to endothermic characteristics of the melting process. Hence, the expansion caused by foaming is unclear. As the temperature exceeds the plateau, the temperature of the sample increases rapidly with the extension of heating time; the sample viscosity decreased rapidly, and the decomposition rate of the foaming agent TiH 2 increased. Despite hydrogen evolution and bubble collapse, it still meets the requirements of accelerating expansion, so the porosity increases rapidly. When the temperature rises to a certain value (700°C), V HP caused by the TiH 2 decomposition considerably decreases, and the viscosity of the sample drastically decreases, leading to considerable gas evolution as well as the discharge of liquid between the bubbles [23,25,26]; hence, bubbles start to merge and disappear. Currently, the sample begins to collapse, the volume decreases, and the foaming process is over. On the contrary, high temperature also indicates the extension of the heating time, which is beneficial for the decomposition of TiH 2 , and strengthening of the growth and merging of pores, even the gas overflow. All in all, these behaviors have a complex influence on the foaming behavior and pore structure. e higher the furnace temperature, the higher the heating rate of the sample and the shorter the time needed to reach the same temperature. Hence, V HD caused by the hydrogen evolution during the TiH 2 decomposition decreases. At the same time, it reduces the probability of bubble burst and merger, that is, it improves the utilization ratio. erefore, the system can achieve a high expanded volume.

Advances in Materials Science and Engineering
At a low heating rate, V HD increases, expansion volume decreases, and porosity decreases due to hydrogen evolution during the TiH 2 decomposition.

Conclusions
It is obvious from the results presented that the heating time and the heating rate are the principal factors that affect the foaming behavior and pore structure of aluminum foam.
Owing to the limitation of the pipe diameter, the maximum expansion height and porosity are constrained, while the pore size, shape uniformity, and circularity of pores are better than those during free foaming. Under the experimental conditions, the sample can foam and completely fill the steel pipe within the furnace temperature range of 740°C-800°C; the aluminum foam with a porosity of 70%-79%, an equivalent-circle diameter of 2.1 mm-3.6 mm, and a circularity of 0.73-0.84 can be obtained. e maximum expansion temperature (full filling of the pipe) is between 730°C and 740°C, so the optimum temperature for producing aluminum foam is 730°C and 740°C.
In the preparation of the aluminum foam by using the PCM method, the decomposition of hydride (TiH 2 ) to generate hydrogen gas is the main foaming mechanism. e entire foaming process is a dynamic nonequilibrium process comprising the formation and collapse of foams. e process is mainly influenced by four parameters, namely, V HP , V GE , V HD , and V BR , among which the volume change of foam is more influenced by H 2 released during the decomposition process of TiH 2 . Meanwhile, the possibility of merger and collapse is reduced, which indicates high utilization efficiency and large expansion volume.

Data Availability
e data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that they have no conflicts of interest. Table 4: Possible mechanisms and their contribution to the volume gain/loss during foaming [23].

Foaming and expansion
Influencing factors Volume gain (VG) Gas expansion caused by hydrogen gas produced by TiH 2 decomposition (HP) Gas expansion (GE) corresponded to the temperature rise of the sample (T s ) Volume loss (VL) Volume loss during the diffusion and overflow process of hydrogen gas (HD) Volume loss during the fracture of surface bubbles (BR) Advances in Materials Science and Engineering 9