Strengthening of Al and Al-Mg alloy wires by melt inoculation with Al/MgB₂ nanocomposite

1 Introduction

One of the main challenges that the aerospace industry confronts has been the formulation of novel materials with low fabrication cost and more energy-efficient [1]. In particular, one means of increasing energy consumption efficiency is to lower the overall weight of the aircraft, or, more specifically, certain parts of the supporting systems such as the electrical wiring of the aerospace structure.

Particularly, magnesium diboride (MgB₂) has a density of 2.57 g/cm³, close to the density of aluminum, which is 2.7 g/cm³, which becomes superconducting at a critical temperature of 39 K [2–5]. Because of its low density and high hardness MgB₂ turns into a particularly interesting diboride to deserves further studies. Another characteristic is that the crystal structure of MgB₂ is an hp3 hexagonal structure similar to AlB₂, which is used in aluminum grain refiners [6]. Based on the successful addition of other diboride nanocomposites [7], the present research explores the feasibility of reinforcing aluminum and its alloys with this promising diboride, which was nanosized to avoid excessive impact on the final electrical resistivity of the reinforced metal.

2 Experimental procedure

In a prior research at our institution we demonstrated how a Pulverisette P4^™ Vario-planetary high energy ball milling (Pulverisette 4, Fritsch GmbH, Idar-Oberstein, Germany) could reducing diboride particle sizes by fragmentation to submicron scale [8]. For the present research, the MgB₂ powder (AlfaAesar, Ward Haverhill, MA, USA) was mixed with high purity aluminum powder (~44 μm) (Acros Organics, Morris Plains, NJ, USA) in the aforementioned ball milling unit operated at 1000 rpm, in four intervals of 15 min separated by 5 min of rest, to form the diboride/Al nanocomposite pellets. The resulting mixture of 90%Al-10%MgB₂ was sintered at 260°C for 30 min in a reduced pressure atmosphere (~4 kPa) in order to enhance the aluminum/diboride interface. All samples were characterized by X-ray diffraction (XRD) using a Siemens^® D500 diffractometer operated with Cu K_α radiation (λ=0.154178 nm). This procedure helped to: a) estimate the nanoparticle size of MgB₂ upon ball milling, using Scherrer’s equation [9]; and b) to analyze the XRD patterns of the composite pellets so as to determine whether unintended phases formed upon milling.

Commercially pure aluminum (99.5%) was then melted at 760°C in a graphite crucible in an electrical resistance furnace, The melt was inoculated with the sintered pellets, and mechanically stirred to improve the nanocomposite pellet distribution as its aluminum constituent melted. The targeted levels of MgB₂ in the final material were: 0, 0.5, 1, and 2 wt.%. The treated aluminum containing the MgB₂ nanoparticles was poured into a cylindrical mold to produce 6 mm diameter ingots. A similar procedure was used to produce an Al-1%Mg Alloy by diluting a 10 wt.% Mg master alloy with pure aluminum. The molten alloy was subjected to the same treatment as the pure aluminum melt.

Then, the ingots were cold-formed to reduce the diameter to 2.7 mm. The wires were fully annealed at 400°C for 5 h to warranty additional cold rolling to reduce the diameter to almost 1.4 mm. Another full annealing at 400°C for 5 h allowed reducing the diameter to 1 mm. Finally, the wires were cold-drawn to remove the excess material and obtain a uniform 1 mm diameter wire.

The microstructures of all specimens, i.e. as-received powder, pellets and wires (at different stages of the manufacturing process), were observed using a Nikon^® Epiphot 200 optical microscope as well as a JEOL-7000 scanning electron microscope (SEM). An Instron^®, model 5944, low force universal testing machine allowed testing of the wires following ASTM E8/E8m-09 standards under tension at room temperature [10].

3 Results and analysis

3.1 Production of the MgB₂ nanoparticles

As mentioned, a high energy ball milling allowed the MgB₂ particle size to be readily reduced. To determine the most efficient milling time we monitored the particle reduction between successive milling stages. The analysis showed that after 5 h of milling the fragmented particle size remained constant. Figure 1 shows the X-ray diffraction (XRD) pattern of MgB₂ specimens as a function of milling time.

The Scherrer’s equation allowed estimating the average particle size of the MgB₂ upon milling using the three largest peaks of each XRD, i.e. at 2θ of 33.6°, 42.6° and 60.1°. Figure 2 presents these particles size as a function of ball milling hours.

Figure 1:

XRD patterns of MgB₂ specimens at different milling hours.

Figure 2:

Average MgB₂ particle size vs. hours of ball milling.

The secondary electron images in Figure 3A and B correspond to clusters of MgB₂ particles without milling and after 5 h of milling, respectively; the reduction of the diboride particle size becomes apparent. We selected the MgB₂ particles with 5 milling hours with approximately ~17 nm for the ensuing experiments.

Figure 3:

Secondary electron images of MgB₂ particles: (A) As-received MgB₂ (without ball milling); and (B) clusters of MgB₂ particles after 5 milling hours.

3.2 Fabrication of MgB₂/Al nanocomposite pellets

In the same ball milling unit discussed before the nanosized MgB₂ particles were mixed with pure aluminum powder at 1000 rpm for 1 h to form MgB₂/Al nanocomposite pellets, which were then sintered at 260°C for half hour in a reduced vacuum atmosphere. Figure 4 shows the XRD pattern of MgB₂/Al pellets. One can observe that no other additional phases were obtained in the pellet fabrication, at least in amounts above the XRD detection level. Images of the MgB₂/Al nanocomposite pellets microstructure permitted to corroborate that the MgB₂ nanoparticles were well embedded into the Al matrix of the pellets, as shown in Figure 5.

Figure 4:

XRD pattern of MgB₂/Al nanocomposite pellet. The inset corresponds to the largest MgB₂ peak in the square.

Figure 5:

Optical micrographs of the nanocomposite pellets: (A) MgB₂/Al nanocomposite at low magnification; (B) MgB₂/Al nanocomposite at higher magnification.

3.3 Electrical conductivity

The results of the wire electrical characterization, performed at temperatures from 0° to 100°C, are displayed in Figures 6 and 7. They reveal the MgB₂/Al nanocomposite addition lowered the electrical conductivity of the wires. Figure 7 appears to demonstrate that there is a significant change in the conductivity of the wires with Mg content. Morever and naturally, the resistivity increased almost linearly with temperature.

Figure 6:

Effect of the MgB₂ nanoparticles amount and temperature on the electrical conductivity of aluminum wires (measured as percent of IACS).

Figure 7:

Effect of the MgB₂ nanoparticles amount and temperature on the electrical conductivity of Al – 1 wt.%Mg wires (measured as percent of IACS).

3.4 Tensile tests

Figure 8 reveals how the ultimate tensile strength of the wires increases notably as the amount of diboride nanoparticles added increases. The tensile strengths achieved in the aluminum wires is much higher than the strength normally reported for pure aluminum, i.e. 70 MPa [11].

Figure 8:

Measured ultimate tensile strength of aluminum wire samples as a function of the amount of diboride added.

4 Discussion

A multiple linear regression study was carried out to verify that statistical significance of the effect of the addition of nanoparticles on the electrical conductivity and ultimate tensile strength of the wires. Equations 1 and 2, respectively, are the resulting descriptive models of the electrical conductivity with respect to the amount of nanoparticles added and the temperature for pure Al and the Al-Mg alloy. The nomenclature used in the equations is as follows: %IACS=percent of International Annealed Copper Standard; T=temperature (°C); %MgB₂=weight percent of MgB₂; and UTS=ultimate tensile strength (MPa).

The resulting analysis of variance (ANOVA) in Table 1 displays the fitting parameters and p-values for the aforementioned models. It is apparent that the electrical conductivity of both types of wires, i.e. made of pure Al and Al-Mg alloy, is well described by the linear regression equations 1 and 2. The values of the coefficients of multiple determinations, R², for both models are very high, i.e. 88.23% and 89.73%, respectively, demonstrating that both models explain well the variability of the response data. For Equation 1, the p-value calculated is 0.007, which proves that increasing the addition of MgB₂/Al nanocomposite pellets had a strong effect on the wire electrical conductivity. For Equation 2, the p-value of 0.000 and the statistically significant relevance of the (%MgB₂)² parameter demonstrate how a larger amount of MgB₂/Al nanocomposite pellets can decidedly affect the electrical conductivity of the Al-Mg alloy wires.

Equations 3 and 4 correspond to the resulting multiple regression models of the wires ultimate tensile strength (UTS) with respect to the amount of nanoparticles added to both wires. Once again the models have high R² values: 81.8% and 85.74%, respectively. The resulting ANOVA in Table 2 further displays the fitting parameters and p-values for both models, i.e. both materials. For Equation 3, the p-value of 0.002 proves that increasing the amount of MgB₂/Al nanocomposite pellets strengthen significantly the Al wires. Also, for the Equation 4, the p-value calculated is 0.003, which implies that more MgB₂ nanoparticles also help in the ultimate strength of the Al-Mg alloy wires.

As observed in Table 1 and Figure 6, the electrical conductivity of the wires treated with the MgB₂/Al nanocomposites was not linearly affected with the rising diboride content. Only at high addition amounts, i.e. above 1 wt.% MgB₂, are there noticeable increments of resistivity while at the same time those additions are beneficial as strengtheners of the wires.

According to Figure 7, the electrical conductivity of the wires Al-1%Mg remained almost constant, with a value over 63 IACS% at 20°C [11]. We attributed this result to Mg low resistivity reported as 4.45 μΩ·cm [12]. However, when MgB₂ levels are raised, the nanoparticles appeared to increase the electrical conductivity of the wires much more than that of aluminum wires. In this respect, a prior research by Calderón et al. [13] demonstrated that AlB₂ can dissolve Mg atoms present in the surrounding aluminum matrix in an AlB₂/Al composite. A ternary AlMgB₂ forms as a consequence. One can assume that this diffusion could occur in the MgB₂-reinforced wires motivating Al to substitute Mg atoms in the diboride causing a slight increasing of magnesium levels around each particle. This would also affect adversely the conductivity of the wires, explaining the larger increase in resistivity measured.

Jointly Table 2 and Figure 8 reveal that the tensile strengths of the wires have effectively increased as a function of the amount of MgB₂ added. Such strengthening could have been attributed to finer grains resulting from heterogeneous nucleation on the diborides (through a Hall-Petch effect), as demonstrated by prior research on other boride-metal systems [14]. However, qualitative data obtained from optical microscopy observations of the treated wires, indicated that no significant grain refinement resulted; hence, the nanoparticles did not serve as catalytic substrates for solid Al formation.

Existing models accounting for the interaction between dislocations are based on various descriptions of particle geometry and the spatial distribution of small particles [15]. It is well known, that when an isotropic distribution of nanoparticles is achieved, the composites are additionally expected to be Orowan-strengthened [16]. Succinctly, the interaction between the dislocation and the particle contribute to raise the mechanical resistance of the matrix, in our case aluminum or aluminum-magnesium alloy, because the energy necessary to travel through the gliding plane of the dislocations were the particles are located also increases [17]. Similar effects are reported in metals strengthened with SiC, Al₂O₃, Y₂O₃, and SiO₂ nanoparticles as well as carbon nanotubes [16]. In closing, in the present research it is of particular relevance to note that the measured tensile strengths of the treated wires became much higher that the strength normally reported for pure aluminum, i.e. 70 MPa [11].

5 Conclusions

The completion of the present research allows for several relevant conclusions to be stated:

• MgB₂ nanoparticles were successfully produced and used in the fabrication of MgB₂/aluminum nanocomposite pellets via high energy ball milling.

• Molten aluminum and Al-1wt.%Mg alloy were inoculated with the MgB₂/Al pellets to produce wires by cold-forming.

• Electrical characterization of the both types of wires demonstrated that increasing levels of MgB₂ nanoparticles decreased the Al wires and Al-1wt.%Mg wires electrical conductivity. These results were substantiated via statistical analysis.

• Ultimate tensile strength of the inoculated wires are significantly higher with respect to untreated wires of both pure aluminum and the studied Al-Mg alloy.