New Insights into Fabrication of Al-Based Foam with Homogeneous Small Pore-Structure Using MgCO₃/Zn Composite Powder as a Foaming Agent

Abstract

Due to its excellent mechanical properties and ultra-lightweight, Al-based foam with homogeneous small pore-structures has wide applicational prospects in many industrial fields. However, during the foaming process of molten Al, it is difficult to manipulate the pore structures of the Al-based foam by means of the ALPORAS^© production route due to the violent gas-releasing performance of TiH₂ as a traditional foaming agent. Herein, we developed the melt-foaming route, that is, using MgCO₃/Zn composite powder as a foaming agent instead of TiH₂, the Al-based foam with homogeneous small pore-structures (average diameter was about 1 mm) was prepared successfully. Meanwhile, the decomposition model of the MgCO₃/Zn composite powder was proposed and further verified experimentally. The decomposition kinetics of the MgCO₃/Zn composite powder was also analyzed. Our findings not only shed light on the practical manufacturing of Al-based foam with homogeneous small pore-structures, but provide an insightful improvement for melt-foaming approaches.

1. Introduction

Al-based foams have a great potential to be utilized in the fields of automobile, aerospace, shipment, railway, and civil construction due to their excellent properties [1,2,3,4]. In particular, due to the unique combination of low density, high specific strength, and stiffness, Al-based foams can be applied as bearing structural materials for front side beam, engine support of automobiles, and so on [5,6,7]. However, the mechanical properties of Al-based foams can be influenced by many factors. Among them, pore-structures such as pore size, pore shape, and distribution, etc. play a crucial role [8,9]. Previous research has demonstrated that Al-based foam with small pore size not only has more reliable mechanical response, but also improved mechanical properties [10,11]. Therefore, it is very urgent to develop approaches to fabricate Al-based foams with homogeneous small pore-structures.

Generally, it has been verified that TiH₂ powder has been widely used as a foaming agent in the industrial manufacturing route of Al-based foam (ALPORAS^©) due to its friendly practical advantages such as easy-keeping, sufficient gas-production, and excellent dispersion in molten Al, etc. [12,13,14,15]. However, it has been found that TiH₂ powder dispersing in the molten Al exhibits over-activity and strong reaction kinetics, resulting in an unstable melt composite during the foaming process [16]. Therefore, it is difficult to obtain an Al-based foam with a homogeneous small pore-structure using TiH₂ as the foaming agent. In past decades, metal carbonates [17,18] (e.g., CaCO₃, BaCO₃, and MgCO₃) have attracted much attention as foaming agents due to their smooth thermal decomposition reactions combined with the thickening effect. However, the limited foaming draw force generated from the decomposition of CaCO₃ and BaCO₃ is insufficient at high temperatures, and it is still very difficult to manipulate the pore-structure during the preparation process of Al-based foam. Decomposition of MgCO₃ can also produce CO₂ to draw the melt to grow up when dispersed in the melt. Some studies have used MgCO₃ powder as a foaming agent [19]. Unfortunately, MgCO₃ has a relatively lower density (<1.73 g cm⁻³) than molten Al due to its usual existence in the form of hydrates, leading to its unsatisfactory dispersion in the melt and final inhomogeneous pore-structures of Al-based foam [20]. In order to overcome this problem, MgCO₃ was coated with silica. However, the foaming experimental results have shown that the decomposition rate of modified MgCO₃ particles is very low, which decreased the foaming force [21]. However, the addition of Cu cannot subsequently melt in the Al melt because it has a higher melting point (1358 K) than Al (933 K), resulting in an increase in the foaming resistance, which further impedes its practical uses. In contrast, Zn not only has a low melting point (693 K) and infinite mutual solubility with Al, but has excellent intrinsic plasticity due to its hexagonal close-packed structure (HCP). Therefore, Zn is easily applied as a carrier material and has no impact on the foaming process during the decomposition of MgCO₃.

In this work, we attempted to fabricate Al-based foams with homogeneous small pore-structures using the MgCO₃/Zn composite powder as the foaming agent. In addition, the decomposition kinetics of the MgCO₃/Zn composite powder were also investigated based on our proposed decomposition model.

2. Materials and Methods

2.1. Materials

As-received MgCO₃ powder (purity >99.5 wt%, 40 µm in diameter) and Zn powder (purity >99.5 wt%, 80 µm in diameter) were purchased from Alfa Aesar, Haverhill, MA, USA and then fully mixed in accordance with the mass ratio of 1:4. Then, the mixture powder was preheated at 573 K for 1 h to ensure dehydration. After that, the mixture of two powders were ground thoroughly for 2 h in order to make sure that all the MgCO₃ particles were embedded in the Zn powder particles before finally forming a composite powder. Additionally, Ca particles (purity > 99.9%, 50 µm in diameter, Beijing Chemical Industry Group Co. Ltd., Beijing, China) were selected as the thickening agent and a commercial pure Al ingot (purity > 99.5 wt%, Beijing Lichengxin Metallic Materials Company, Beijing, China) was prepared.

2.2. Decomposition Experiments of MgCO₃/Zn Composite Powder

The decomposition experiments of the MgCO₃/Zn composite powder were carried out on a specially designed apparatus for which a diagram is presented in Figure 1. Two decomposition measurements, namely constant temperature decomposition experiment and continuous temperature decomposition experiment, were conducted. In the constant temperature decomposition experiment, the foaming agent with a weight of 1 g was put into the furnace, and then heated to the temperature of 1173 K and kept constant. Therefore, the maximum decomposition gas released from the foaming agent was obtained. Additionally, in the continuous temperature decomposition experiment, 1 g of foaming agent was also heated with a linear-rising temperature from 293 K to 1173 K with a heating speed of 30 K min⁻¹ in order to clarify the maximum temperature range of the foaming agent decomposition. These experiments were also conducted using TiH₂ powder for comparison.

The actual decomposition rate of MgCO₃ was also determined by the mass loss method, which was characterized by using a differential thermal analysis (DTA, Netzsch STA 449F, Selb, Bavaria, Germany) at a rate of 20 K min⁻¹. The experimental temperature was up to 1173 K and finally, the thermal weight loss curves were obtained.

2.3. Fabrication of Al-Based Foams

The melt-foaming method based on thee ALPORAS© route was applied to fabricate Al-based foams and the specific preparation procedure has been elaborated in detail previously [22,23]. Briefly, the bulk Al with a predetermined mass was cut from a pure Al ingot, and initially melted in a stainless-steel crucible at 973 K, followed by the addition of 2.0 wt% Ca particles into the melt to increase its viscosity. Finally, the thickened Al melt was foamed after adding the foaming agent, the MgCO₃/Zn composite powder (2.5 wt%), at the foaming temperature of about 953 K. The pore structures were adjusted by regulating the stirring foaming time and the holding foaming time with an impellor rotation speed of about 1000 rpm. In our case, the stirring foaming time and the holding foaming time was 90 s and 72 s, respectively. As a result, Al-based foams with a homogeneous small pore-structure were obtained. In addition, Al-based foams using TiH₂ (adding 1.5 wt%) as a foaming agent were also fabricated via same route for comparison.

3. Results and Discussion

3.1. Decomposition Performance of the MgCO₃/Zn Composite Powder

The constant temperature decomposition experiment of the MgCO₃/Zn composite powder is shown in Figure 2a, which demonstrates that the MgCO₃/Zn composite powder exhibited a smoother and moderate gas-releasing performance compared with the TiH₂ powder while being heated at 1173 K. Specifically, the MgCO₃/Zn composite powder decomposed correspondingly after being heated, followed by the constant increase (76.2 L mg⁻¹) until it ceased in six minutes. In contrast, the decomposition of the TiH₂ powder started after being heated for about 2 min later at same temperature, and then stopped at about ten minutes after being decomposed severely. Obviously, the decomposition of the TiH₂ powder released H₂ gas with a total of 253.3 L/mg in the final. As shown in Figure 2b, the MgCO₃/Zn composite powder also revealed different decomposition kinetics with the TiH₂ powder, where the maximum gas-released volume was 175 L at about 673 K. With the increase in temperature, the decomposition of the MgCO₃/Zn composite powder was nearly finished at about over 900 K. In comparison, H₂ gas was generated initially at 483 K after the TiH₂ powder was heated and kept constant, where the maximum gas volume obtained was over 250 L.

3.2. Decomposition Kinetics of the MgCO₃/Zn Composite Powder

(1) Decomposition model

Previous research has presented a decomposition model of CaCO₃ as a viscosity-controlling agent used in the Al alloy foaming process, which has been verified to fit well with the experimental data [24]. According to the experimental results, where CaCO₃ particles could be heated to 1173 K within 0.2 ms in the melt, it was assumed that the MgCO₃/Zn composite powder was heated to the predetermined temperature at once as it was added into the molten Al due to very strong convective heat transaction combined with the stirring effects. Therefore, a decomposition model of the MgCO₃/Zn composite powder particle is proposed in Figure 3. In the model, specifically due to the lower melting point than MgCO₃, the outer Zn melts to form the Zn layer and then diffuses to the Al melt, forming the Al diffusion layer. Meanwhile, MgCO₃, as a core surrounded by the above two layers, decomposes to produce CO₂ and MgO, then finally, three layers are formed. Therefore, the thermal decomposition of the MgCO₃/Zn composite powder is controlled by many processes such as heat transfer, chemical reaction, CO₂ diffusion in the MgO layer and the Al and Zn fusion layers. The decomposition is accelerated by convective flow moments, mechanical stirring effects, and solid–liquid–gas composite interaction, etc. Meanwhile, in our case, the wrapping of MgCO₃ particles by high viscosity melt also increased the diffusion resistance of CO₂ and reduced the diffusion rate.

(2) Calculated kinetics of reaction between MgCO₃ and Al melt

During the foaming process, assuming that the MgCO₃/Zn composite powder particle is heated to predetermined temperature at once due to ignoring the thermal resistance in this process, Zn surrounded by the composite powder is melted first due to its low melt point (692 K) compared with Al (971 K). Therefore, the decomposition speed of MgCO₃ in the melt can be expressed as:

$$v_{dec}=KA\left[P\left(T\right)-P_{C{O}_2}\right]$$

(1)

where K is the constant; A is the interface area; and T is the reactive temperature. P(T), $P_{C{O}_2}$ are the partial pressure at a predetermined temperature and a partial pressure of CO₂ in the equilibrium status, respectively. Among which, P(T) can be presented as:

$$P\left(T\right)=\exp \left[17.74-0.001087T+0.332lnT-\frac{22020}{T}\right]$$

(2)

In our case, during the foaming process, the bubbles in the molten Al can be grown up under the driving force, which comes from the incremental gas produced by MgCO₃ decomposition. Meanwhile, the internal pressure increase of the bubble is also resisted by several forces including the melt pressure, the additional pressure caused by the static pressure of atmosphere and surface tension, etc. Herein, $P_{Zn}$ can be regarded as zero. When the driving force and resistant forces are balanced, the bubble in the melt remains stable and this balance can be written as:

$$P_{C{O}_2}=\frac{2\sigma }{r}+P_0+P_{Al}+P_{MgO}$$

(3)

where $\frac{2\sigma }{r}$, and P₀, P_Al, and $P_{MgO}$ are the surface tension, the static pressure of atmosphere, and the partial pressure of MgO at the given temperature, respectively.

Herein, $d{V}_{dec}=\delta v_{dec}dt$ and δ is a constant. Therefore, based on the assumption that MgCO₃ is decomposed at the interface between MgCO₃ and the molten Al, the relationship between decomposition rate (V_dec) and time (t) is established after integration as:

$$1-{(1-V_{dec})}^{\frac{1}{3}}=\frac{M\left[P\left(T\right)-\frac{2\sigma }{r}-P_0-P_{Al}-P_{MgO}\right]Kt}{r_p\rho }$$

(4)

among which, M, $r_b$, and ρ are the molar mass, the radius, and the density of the MgCO₃ particle, respectively. For more derivation details, please see [23,24].

According to Equation (4), it can be concluded that the decomposition rate has an inverse proportion with the size and the density of MgCO₃ particle, and has a proportion with the pore size of the Al-based foam and decomposition time (t).

(3) Experimental verification

As shown in Figure 4, the theoretical values of the MgCO₃/Zn decomposition rate in molten Al are in agreement with the experimental values, which were measured by the DTA experiments, indicating that the model nearly reflects the actual decomposition rate of the MgCO₃/Zn composite powder.

3.3. The Characterization of Pore-Structure

Porosity and pore size are two of the most important characters of metallic foams. The porosity, Pr, is defined as the pore fraction of a foam sample by using the following equation [25,26]:

$$Pr(\%)=1-\frac{M}{V{\rho }_s}\times 100\%$$

(5)

where M, V, and ${\rho }_s$ are the mass, the volume of the foam, and the density of the matrix, respectively.

The porosity (Pr) of the prepared Al-based foams in our case was 82.4% after measurement and then calculated according to Equation (5). The pore-structure of the cross-section was observed by a digital stereo optical microscope (DSM, Vikeshow Technology Inc., Shenzhen, Guangdong, China), which is presented in Figure 5. In general, as shown in Figure 5a, the Al-based foams, which were fabricated by using the MgCO₃/Zn composite powder as the foaming agent, exhibited homogeneous pore-distribution. It was revealed that the MgCO₃/Zn composite powder had good dispersion in the melt under complex interaction between the stirring and thermal convection due to its similar density with the melt. The enlargement of the pore-structure morphology is inset in the figure, indicating that a large number of spherical and near spherical pores were distributed, where the diameters were less than about 1 mm. In contrast, as shown in Figure 5b, the pore-structure of the Al-based foam fabricated by using TiH₂ displayed more differences. The inset in Figure 5b further exhibited polygon pore shapes and larger pore size (average equivalent diameter was about 2.8 mm) than that of the sample using the MgCO₃/Zn composite powder as the foaming agent, which also suggests that the MgCO₃/Zn composite powder can be used as a suitable foaming agent for Al-based foams with small pore size and spherical pore shapes.

4. Conclusions

Based on the melt-foaming method, an Al-based foam with homogeneous small pore-structures was fabricated successfully using MgCO₃/Zn composite powder as the foaming agent. The decomposition model was also proposed and verified, which demonstrated that the MgCO₃/Zn composite powder is practical for use compared to TiH₂. Furthermore, the decomposition kinetics of the MgCO₃/Zn composite powder revealed that the characteristics of the gas released were smoother, and the gas production temperature range was matched during the melt foaming process. The pores of the fabricated Al-based foams were distributed homogeneously in the foams, where the diameters were less than about 1 mm, heralding further satisfactory mechanical properties.

Our findings not only shed light on the practical manufacturing of Al-based foam with homogeneous small pore-structures, but provide new insightful improvement for melt-foaming approaches.