The effect of precipitation of metastable phases on the thermophysical and mechanical properties of the EN AW-6082 alloy

Uroš Stamenković*, Svetlana Ivanov, Ivana Marković, Ljubiša Balanović, Milan Gorgievski

University of Belgrade, Technical Faculty in Bor, Vojske Jugoslavije 12, Bor, Serbia

(*Corresponding author: ustamenkovic@tfbor.bg.ac.rs)

1. INTRODUCTIONTOP

Aluminium alloys from 6000 series are widely used in many applications due to excellent properties which can be additionally improved by precipitation hardening (Marioara et al., 2001; Marioara et al., 2003; Birol, 2006; Marioara et al., 2006; Abid et al., 2010; Birol, 2013; Prabhu, 2017; Acosta and Veleva, 2018). Alloys from this series are often used for production of heat sinks due to good electrical conductivity, high strength and good thermal properties (Karabay, 2006; Birol, 2013; Cui et al., 2014; Choi et al., 2014; Choi et al., 2015; Kim et al., 2018). The EN AW-6082 aluminium alloy is a good candidate for this application too. In the literature many research groups have investigated EN AW-6082 alloy and its properties during different parameters of heat. The increase in mechanical properties during different thermal treatments was studied by other authors (Gupta et al., 2001; Marioara et al., 2003; Birol, 2006; Abid et al., 2010; Birol, 2013). Electrical conductivity change during aging treatment was studied by Birol (2013), Karabay (2006), and Cui et al. (2014). Prabhu (2017) reported the change in grain structure of the aged samples that provided the increase in hardness values of the 6082 aluminium alloy. Marioara et al. (2001), Marioara et al. (2003) and Marioara et al. (2006) followed the precipitation sequence by TEM analysis. The change in thermal properties was investigated by Choi et al. (2014), Choi et al. (2015), Zhang et al. (2016), Vishwakarma et al. (2017) and Kim et al. (2018). Kim et al. (2018) confirmed the highest thermal diffusivity at 250 °C for as-quenched samples. Choi et al. (2014) showed an increase in thermal conductivity and diffusivity of aged samples. Zhang et al. (2016) also proved that the thermal properties are highly influenced by the microstructure of the Al-Mg-Si-Cu alloy. Vishwakarma et al. (2017) defined the optimization model for obtaining the highest thermal properties for EN AW-6082 Al alloy. In review of the literature present, it can be concluded that aging process is defined primarily by aging temperature and time. The researches for EN AW-6082 alloy are based more on isothermal aging (Marioara et al., 2001; Marioara et al., 2003; Marioara et al., 2006; Abid et al., 2010; Cui et al., 2014) rather than isochronal aging. Therefore, our aim was to investigate the influence of isochronal aging on thermophysical and mechanical properties of EN AW-6082. Optimal temperatures for isochronal annealing were defined by DSC analysis and thermal investigation. Our interests were focused on defining the isochronal aging regime that included short aging periods on higher temperatures which would try to yield the hardness values that are dictated by the European and American standards for these types of alloys and tempers (in this case T6 temper).

2. MATERIALS AND METHODSTOP

For this experimental investigation an EN AW-6082 alloy was chosen. Extruded rectangular bars from “AlCu metali d.o.o.” company were delivered in peak aged condition. Optical emission spectrometer “Belec Compact Port” was used for obtaining the chemical composition of the alloy, given in Table 1. O-temper was achieved by annealing the delivered samples at 550 °C for 6 hours in the electric resistance furnace Heraeus K-1150/2. After that, super saturated solid solution (α_ssss) was obtained by heat treatment at 550 °C for 1 hour and quenching in ice water. Quenched samples were subjected to isochronal aging at different temperatures, ranging from 160 °C to 330 °C for 30 minutes and 60 minutes. Mechanical and physical properties of the aged samples were compared to the quenched sample (noted as QS in Figures). The quenched sample was subjected to DSC analysis on SDT Q600 (TA Instruments) simultaneous DSC/TGA analyzer with heating rate of 10 °C·min⁻¹ up to 600 °C. For investigating the thermophysical properties, Flash method was used with DXF 500 thermal analyzer manufactured by TA Instruments. Quenched samples were investigated by heating at the rate of 10 °C·min⁻¹ in a nitrogen atmosphere. Electrical conductivity was measured using the electrical conductivity tester “Sigmatest 2.063”. VEB Leipzig Vickers hardness tester with a 10 kg load and a 15 s dwelling time was used for hardness measurements. Additionally, microhardness values were measured using a PMT-3 Vickers microhardness tester using 100 gf loads with load duration of 15 s. The ASTM E384 standard was followed during both the hardness and microhardness measurements (ASTM E384-17, 2017).

TESCAN Vega 3 LMU scanning electron microscope equipped with an X act EDS detector by Oxford Instruments was used for metallographic phase investigations and distribution of phases.

3. RESULTS AND DISCUSSIONTOP

3.1. DSC analysisTOP

The DSC thermogram of the quenched EN AW-6082 sample is shown in Fig. 1. The thermogram reveals the largely accepted precipitation sequence in Al-Mg-Si alloys which is well established to be SSSS (super saturated solid solution)→ atomic clusters→ G.P. zones→ β’’→ β’→ Si (if there is excess silicon in alloys composition)→ β (Zhen et al., 1997; Edwards et al., 1998; Gupta et al., 2001; Vedani et al., 2007; Shang et al., 2011; Birol, 2013; Kim et al., 2018). It can be seen from Fig. 1 that five exothermic peaks appear in Fig. 1. Some of these peaks are less pronounced and this can be attributed to the low calorimetric effect (Zhen et al., 1997). The first exothermic peak appears at around 85 °C and it is associated with the Mg:Si clustering (Edwards et al., 1998; Birol, 2006). The two exothermic peaks with their maximums situated at 254 °C and 303 °C represent the formation of metastable β’’ and β’ precipitates, respectively (Zhen et al., 1997; Edwards et al., 1998; Birol, 2006; Vedani et al., 2007; Abid et al., 2010; Shang et al., 2011; Birol, 2013; Choi et al., 2015; Kim et al., 2018). In the literature many researchers study the effect of metastable β’’ phase on different properties. According to Marioara et al. (2001), it’s a needle-like phase with monoclinic lattice with the unit cell parameters, a = 15.16 Å, b = 4.05 Å, c = 6.74 Å, β = 105.3° that precipitates along the <100> Al direction. Also, depending on the aging parameters, β’’ phase is often very finely dispersed throughout the structure, reaching the 100.000 – 200.000 particles per square micrometer (Marioara et al., 2003). The precipitation of the excess Si and the formation of the equilibrium β phase is represented by the last two exothermic peaks appearing at 377 °C and 456 °C (Birol, 2013).

3.2. Investigation of mechanical propertiesTOP

Figure 2 shows the development of absolute hardness change as a function of annealing temperature at two different aging times. Hardness gradually increases with aging temperature and it reaches maximum value at 230 °C, when aged for 30 minutes. This peak is in agreement with the DSC analysis as it represents the precipitation of the metastable β’’ phase, which is mainly responsible for the high hardness values (Marioara et al., 2001; Gupta et al., 2001; Abid et al., 2010; Birol, 2013). In comparison with the hardness of 71 HV10 for the quenched sample (circled out in the graph), the peak aged sample has a hardness value of 113 HV10 when aged for 30 minutes and 109 HV10 when aged for 60 minutes. That’s a 59.15% increase for the sample aged at 230 °C for 30 minutes, and a 53.52% increase for the sample aged at 220 °C for 60 minutes. Peak aged samples, at both aging times, reached the hardness values that are recommended by the standard EN 755-2:2016 (EN 755-2, 2016). As aging temperature increases the hardness values decrease due to the transformation of the metastable β’’ hardening phase. The formation of β’ metastable phases can increase the hardness values but only by a small amount, as seen in Fig. 2 at the temperatures between 260–300 °C. By comparing the two graphs presented in Fig. 2, it can be seen that aging for longer time shifts the peak for maximum hardness to the lower temperatures. All the other peaks, after the peak aged state, are more pronounced when aged for a shorter time.

The same conclusions were also drawn by measuring the microhardness of the investigated samples, as shown in Fig. 3. Peak microhardness due to precipitation of β’’ phase was achieved at 230 °C when aged for 30 minutes with a 44.6% increase in microhardness value, in relation to the quenched sample, from 94 HV0.1 to 136 HV0.1. When aged for 60 minutes at 220 °C there was a 38.29% increase in microhardness value, again in relation to the quenched sample, from 94 HV0.1 to 130 HV0.1. After the aging at higher temperature than 230 °C the microhardness values rapidly decrease due to overaging.

3.3. Investigation of electrical conductivityTOP

Figure 4 shows the effect of aging temperature on the electrical conductivity values of the investigated alloy. In Al-Mg-Si alloys there is a strong connection between electrical conductivity and precipitation, so measuring this property can be well used for investigation of the precipitation processes (Karabay, 2006). There is a sudden drop in electrical conductivity for aged samples compared to the quenched one. The electrical conductivity for the quenched sample is 28 MS/m and it decreases after aging at 160 °C. This drop can be explained via the formation of clusters, closely spaced GP zones and fine early-stage precipitates, which both carry the strong electron scattering effect, mentioned by some researchers (Edwards et al., 1998; Cui et al., 2014; Prabhu, 2017). After the initial drop, there is a gradual increase in electrical conductivity values as aging continues, and clusters and early-stage precipitates coarsen due to diffusion process. After aging at higher temperatures than 210 °C for both aging times, samples exceed the electrical conductivity values of the quenched one. As the matrix gets less saturated (loses all the quenched-in elements) during the aging treatment, as a result of the formation of metastable β’’ and β’ phases, the electrical conductivity increases (Karabay, 2006; Cui et al., 2014). It can be observed that the 60 minutes aging curve is slightly higher than the 30 minutes aging curve. This can be associated with the easier formation of pre-β’’ and β’’ phases due to the prolonged aging time, as confirmed by Cui et al. (2014).

3.4. Investigation of thermal propertiesTOP

Thermal diffusivity, conductivity and specific heat of the quenched samples were investigated during the continuous heating in Figs. 5–7, respectively. It is expected that the thermal diffusivity and conductivity decrease with the increase in temperature due to heat vibrations of the atoms and the interruption of electron movement (Tritt, 2004). However, there are increases in these properties during the continuous heating during which the precipitation sequence takes place (Kim et al., 2018). In the 200–220 °C temperature range the first peak appears for all measured thermal properties due to formation of pre-β’’ phase. As the temperature rises, precipitation continues and the values of thermal diffusivity, conductivity and heat capacity rise. Two more peaks appear at around 280 °C and 320 °C due to formation of the β’’ and β’ phases, respectively. Thermal properties have highest values at these peaks due to the reduction of alloying elements in the solid solution caused by the precipitation (Kim et al., 2018; Choi et al., 2019). The peaks for maximal values of hardness and electrical conductivity partially coincide with the peaks obtained for thermal properties. However, the obtained curves for thermal properties shift to the right side of the diagram due to continuous heating and measuring equipment. The specific heat of the investigated alloy increases monotonically with temperature; however, it is not as much influenced by the precipitation of phases that appear in the investigated temperature range comparatively to thermal diffusivity and conductivity measurements (Choi et al., 2015).

3.5. Microstructural investigationTOP

In order to complete the investigation in more detail and study the structural change of the aging process, SEM-EDS analyses were performed. Aged samples with minimal and maximal hardness values were chosen for microstructural investigation i.e. samples aged at 160 °C and 220/230 °C for 30 minutes and 60 minutes. Figure 8 shows the microstructures of the aged samples in different aging conditions. It can be seen that the amount of metastable phases increases with an increase in temperature and aging time.

To further investigate the aforementioned phases, EDS analysis was performed on the sample with the highest hardness value (after aging at 230 °C for 30 minutes), as presented in Fig. 9 and Table 2. In the aged samples, almost the entire structure is covered with finely dispersed particles of the metastable β’’ phase, represented by the spectrum 5 in Fig. 9. Some authors suggested that the ideal ratio of the magnesium and silicon in β’’ phase should be 5 to 6 (Marioara et al., 2003; Marioara et al., 2006).

Our result doesn’t confirm this suggestion probably due to excess Si; still it can be assumed that β’’ phase does appear as finely dispersed. Ultra-fine particles of this phase are very hard to detect at these magnifications; however hardness and microhardness results obtained for the sample aged at 230 °C for 30 minutes indirectly show its existence. In the investigated sample, there is a phase that appears light grey to white in colour. This phase is the quaternary AlMnFeSi phase, represented by spectrums 1 and 3. There is also an additional phase that appears as dark particles throughout the structure. This phase is also a quaternary phase but based on Al, Mg, Mn, Si as presented on spectrums 2 and 4.

4. CONCLUSIONSTOP

Achievements in this paper can be summarized as follows:

ACKNOWLEDGMENTSTOP

The research results were developed with the assistance of the Ministry of Education, Science and Technological Development of the Republic of Serbia under the projects OI 172037 and TR34003.

REFERENCESTOP