Introduction

The high-strength 7xxx series are aluminium alloys from the Al–Zn–Mg–Cu alloy system. They are heat treatable alloys characterised by high-strength, limited fatigue strength, stress corrosion resistance and excellent wear resistance [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16]. They are widely used as structural materials in aerospace, rail, automotive and other industries due to their high-strength-to-mass ratio achieved by heat treatment [1, 2, 5, 6, 9,10,11,12,13, 15,16,17]. EN AW-7175 alloy is one of the newer variants from the Al–Zn–Mg–Cu alloy system based on the alloy EN AW-7075 [8]. The precipitations during solidification/cooling of the 7xxx alloy system are important and complex. Studies by numerous researchers [1,2,3,4,5,6,7,8,9,10, 12,13,14,15,16, 18,19,20,21] have identified phases that precipitate (in addition to α-Al) during the solidification/cooling process of the Al–Zn–Mg–Cu alloy system. They are shown in Table 1. In these studies, it was also shown that the stoichiometry of the M/η, S, T, V, and Z phases varies greatly and depends on the Zn/Mg ratio and the cooling rates, which is why the composition of the phases is usually given with minimum and maximum values. Previous studies [1, 8, 15] also indicated that when the Zn/Mg ratio is below 2.2, the main strengthening phase is T, while when the ratio is above 2.2, the main strengthening phase is η/M. Lim et al. [8] investigated the Zn/Mg ratio in the alloy EN AW-7175 and found that a lower (Cu + Mg) content and a higher Zn/Mg ratio (3.0) compared to a higher (Cu + Mg) content and a lower Zn/Mg ratio (2.0) resulted in a lower solvus of the S phase and thus a coarse S phase.

Table 1 List of phases precipitated during the solidification and cooling process of the Al-Zn-Cu-Mg alloy system
Full size table

In general, 7xxx series alloys do not provide their best as-cast mechanical properties and performance due to inherent microstructural deficiencies such as coarse intermetallic phases, micro and macro segregation, coarse grains, and hot cracking [9, 17]. Hot tearing could be overcome during solidification by two methods: Refining the secondary dendrite arm spacing (SDAS) of the primary α-Al phase and/or by changing the morphology of this phase to a non-dendritic phase [9]. This is one of the reasons why the addition of La has been investigated, as previous studies [22,23,24,25] have reported that it promotes the refinement of the SDAS of the primary α-Al phase/grains. And it is important to first understand the effects of La additions on the solidification process and as-cast phase evolution to successfully achieve optimal temperatures for further heat treatment to achieve the desired mechanical properties. Previous studies [22,23,24,25,26,27,28,29,30] have also shown that the addition of La increases mechanical properties and corrosion resistance.

Based on the above findings, we decided to investigate the influence of La additions on the solidification and phase evolution of the alloy EN AW-7175 in the as-cast state. Thermodynamic calculations based on the CALPHAD method were carried out, using equilibrium calculations and solidification simulations according to Scheil. In addition, DSC (differential scanning calorimetry) analyses were carried out to evaluate the influence of the La additives on the solidification process, and microstructural analyses were carried out using SEM (scanning electron microscopy) to evaluate the microstructural components of the investigated as-cast samples and the influence of the La additives on the phase evolution. The alloy EN AW-7175 (sample 0) served as a reference and the other samples had elevated La additions (0.05, 0.1, 0.14 and 0.17 mass%—samples 1, 2, 3 and 4 respectively).

Experimental work

Since the aim of the study was to investigate the solidification and phase evolution of the alloy EN AW-7175 in the as-cast condition, DSC analyses and metallography were carried out after the preparation of the alloys. CALPHAD calculations were also carried out to evaluate the thermodynamically stable phases of the samples studied and the influence of La addition on phase evolution.

Materials

For each batch, 500 g of a commercial EN AW-7175 alloy was melted. The alloys were produced in an electric resistance furnace with a graphite crucible covered with a calcium silicate (Ca2O4Si) to prevent heat loss and ensure stability of the melt. The melt was heated to 750 °C, which was also the pouring temperature for each batch. After the addition of La (99.99 mass%), the melt was skimmed off and poured into permanent moulds. The temperature of the melt was controlled with a K-type thermocouple. As the aim was to ensure that the metallurgical process parameters were the same for each batch, sample 0 (reference) was also remelted and cast.

The chemical composition (Table 2) was measured with OES ARL iSpark 8860.

Table 2 Chemical composition of the investigated alloys in mass%
Full size table

Since the amount of Cu, Mg and Zn and the ratio Zn/Mg are important, they were also calculated and are shown in Table 3. The amounts and the ratio are almost the same for all samples examined. Since the Zn/Mg ratio is above 2.2, according to the literature [1, 8, 15] the main strengthening phase should be η/M.

Table 3 Amount of Cu + Mg, Zn + Mg + Cu in mass% and ratio Zn/Mg calculated from mass%
Full size table

Calculation of phase diagrams (CALPHAD)

The commercial Thermo-Calc software version 2023a was used for the CALPHAD modelling. The Thermo-Calc Software TCAL8 Aluminium-based Alloys database was selected from which the thermodynamic data for the calculations were obtained. A one axis type of calculation was chosen for the equilibrium calculations, from which plots were obtained showing the amount of thermodynamically stable phases in the alloys studied. Since, solidification and phase evolution were studied in the as-cast state, the determination of solidification under non-equilibrium conditions was calculated using the classical model of Scheil [31, 32] (Scheil-Gulliver solidification simulation). The model assumes that the diffusion of all elements in the liquid phase is infinitely fast (complete mixing in the liquid), that there is no diffusion in the solid (the diffusion of all elements in the solid phases is zero) and that the interface between the liquid and the solid is in complete thermodynamic equilibrium. It is a well-known model and is often used in studies [33,34,35,36,37] on the solidification of metals and alloys.

Differential scanning calorimetry (DSC)

For the DSC analysis, samples with dimensions of approximately 4 × 3.5 × 1.5 mm (± 0.2 mm) and a mass between 30 and 50 mg were prepared from the produced alloys. The solidification process was investigated with the NETZSCH DSC 404 F3 Pegasus device. The samples were heated to 720 °C at a heating rate of 10 K min−1 and cooled to room temperature at a cooling rate of 10 K min−1, with the furnace under Ar protective atmosphere throughout the analysis. Empty DSC crucibles made of Al2O3 were used as reference. Only the cooling curves were used for further analysis, as solidification was investigated. DSC analysis was used to determine the characteristic temperatures of the solidification process, such as the liquidus, solidus and phase precipitation/solidification start temperatures. All peaks on the DSC cooling curves were determined with NETZSCH Proteus® software using the tangent method. To ensure that we determined the exact temperatures for shallower sections, in our case for the Al13Fe4 and Mg2Si/LaSi2 phases (as the amount of these phases was small), the first derivative was calculated and the inflexion points on the first derivative curve were used to determine the characteristic temperatures for the mentioned phases, even if they are visible on the DSC cooling curve. But to be on the safe side, we check them again with the derivative. DSC analysis has been shown to provide very accurate results when investigating the solidification process of metals and alloys, as reported in previous studies [38,39,40,41].

Metallography

The samples for metallography were ground and polished. Metallographic analysis was performed using the ThermoFisher Scientific Quattro S FEG-SEM (field emission gun scanning electron microscopy). The latter was used for phase analysis with the analyser EDS (energy-dispersive X-ray spectroscopy) Ultim® Max Oxford Instruments, where element mapping, chemical composition of the phases and line scanning were performed.

Results and discussion

The results of the CALPAHD and DSC analysis shown are in the temperature range between 400 and 720 °C, as the solidification process and phase development were investigated in the cast state. Sample 0 is used as a reference sample without La addition.

CALPHAD analysis

Based on the results of the calculated thermodynamically stable equilibrium phases for the samples studied, there were minor differences, we have only drawn diagrams for samples 0, 2 and 4 (Fig. 1) as they are the most representative. All results concerning solidus, liquidus and precipitation temperatures are shown in Table 4. In addition, the mass fractions of the phases stable at 400 °C in the samples studied (Table 5) have been shown.

Fig. 1
figure 1

Mass fraction of all thermodynamically stable equilibrium phases as a function of temperature for sample 0 (a) and (b—magnification of a), sample 2 (c) and (d—magnification of c) and sample 4 (e) and (f—magnification of e)

Full size image
Table 4 Liquidus temperatures of the equilibrium phases in the temperature range between 400 °C and 720 °C
Full size table
Table 5 Mass fraction of the thermodynamically stable equilibrium phases at 400 °C in all investigated samples
Full size table

Since the effect of La addition on solidification was investigated, we first looked at the liquidus and solidus temperatures. For sample 0 the liquidus temperature is 663.6 °C, for sample 1 663.4 °C, for sample 2 663.8 °C, for sample 3 665.8 °C and for sample 4 663.8 °C. On the other hand, the solidus temperature for sample 0 is 504.5 °C, for sample 1 510.8 °C, for sample 2 506.7 °C, for sample 3 509.0 °C and for sample 4 509.2 °C. As can be seen from the calculated results, there are minor changes in the liquidus and solidus temperatures, with the former remaining almost unchanged and the latter increasing slightly compared to reference sample 0 (6.3, 2.2, 4.5 and 4.7 °C for samples 1, 2, 3 and 4, respectively). Meanwhile, the solidification interval (TL–TS) is slightly reduced by the addition of La compared to reference sample 0 (the maximum difference is 6.5 °C in the case of sample 1, the others are smaller), but no significant differences can be observed.

In the following table (Table 4), the liquidus temperatures of the thermodynamically stable equilibrium phases are compiled for all samples investigated. According to the calculations, three new phases are formed by the addition of La to the alloy EN AW-7175: LaSi2, Al11La3 and La2Zn11. Compared to the reference sample (0), the addition of La affects the liquidus temperature of Mg2Si, Al18Mg3Cr, Al7Cu2Fe and the T-Phase. However, in the case of Mg2Si (up to 0.14 mass% La) phase and T-Phase it is obvious that with increasing La addition, the liquidus temperature of the mentioned phases decreases. In the case of Al18Mg3Cr and Al7Cu2Fe (up to 0.14 mass% La), however, the liquidus temperature also increases with increasing La content.

Table 5 shows the mass fraction of the stable phases at 400 °C for all samples investigated. The main difference with the addition of La is the mass fraction of the S-phase, which increases slightly with the addition of La, and the amount of the T-phase, which decreases with increasing La addition and is no longer present in sample 4. As for the La phases, only the La2Zn17 phase is stable at 400 °C. According to calculations, La begins to form the La2Zn17 phase (cyan-coloured curve in Fig. 1) from LaSi2 (lime-coloured curve in Fig. 1) and Al11La3 (pink-coloured curve in Fig. 1), which also has the lowest liquidus temperature (Table 4) of all La phases formed in the samples studied.

Since, the as-cast samples were used for microscopic analysis to evaluate phase evolution at a cooling rate of 10 K min−1, Scheil solidification simulations were also performed. In this case, plots showing the mass fractions of the phases during the Scheil solidification simulations (Fig. 2) were prepared for samples 0, 2 and 4. All results concerning the solidus temperatures are shown in Table 6. In addition, we have also listed the mass fraction of the phases in the samples after solidification (Table 7).

Fig. 2
figure 2

Mass fraction of the phases solidified during the Scheil solidification calculation sample 0 (a) and (b—magnification of a), sample 2 (c) and (d—magnification of c) and sample 4 (e) and (f—magnification of e)

Full size image
Table 6 Liquidus temperatures of the phases in the Scheil solidification simulation for the investigated samples
Full size table
Table 7 Mass fraction of the phases at solidus temperatures based on the Scheil solidification simulation of the investigated samples
Full size table

Since, the Scheil model simulates solidification, the solidus temperatures are given for the samples studied, as the liquidus temperatures are the same as in the equilibrium calculations. The solidus temperature for sample 0 is 460.3 °C, for sample 1 461.3 °C, for sample 2 461.3 °C, for sample 3 461.3 °C and for sample 4 468.8 °C. The results show that the addition of La does not change the solidification interval significantly (it only changes by 1 °C compared to sample 0). Only sample 4 shows a slightly shorter solidification interval (8.5 °C compared to sample 0).

Based on the Scheil solidification calculations in terms of the liquidus temperatures of the phases (Table 6), the addition of La during solidification forms two new phases (LaSi2 and Al11La3—the latter only in sample 4). Compared to the reference sample (0), the addition of La influences the liquidus temperatures of Mg2Si. The higher the La content, the lower the liquidus temperature of Mg2Si. For Mg2Si, the trend is the same as in the equilibrium calculations (Table 4). Another difference is in sample 4, where 0.17 mass% La has an effect on the S-phase and the V-phase, which are no longer present during solidification but form the M-phase. This differs from the equilibrium calculations (Table 4) where the M-phase and the V-phase were not present. The Al7Cu2Fe, Al15Si2(Fe,Cr)3 and Al18Mg3Cr phases are also absent during solidification compared to the results of the equilibrium calculations (Tables 4 and 5). On the other hand, the Al13Fe4 phase appears during solidification.

Since, different phases solidify during solidification of the investigated samples based on Scheil calculations compared to equilibrium calculations, a table (Table 7) was also prepared showing the mass fractions of the phases at solidus temperatures of the investigated samples. There is an obvious trend in terms of the amount of La added: the more La is added to the samples, the less T-phase is present in the samples. It is also clear from the calculations that LaSi2 is the only La phase present after solidification, except in sample 4 where Al11La3 is also present. When we compare the results of the La phases with the results of the equilibrium calculations, in the case of the Scheil solidification simulations, no La2Zn17 phase is present after the solidus temperature, which was present in the equilibrium calculations. Moreover, at higher La addition, the S-phase (above 0.14 mass% La) and the V-phase (above 0.10 mass% La) are no longer present during solidification. On the other hand, the M-phase starts to solidify at higher La addition (above 0.14 mass%).

Thus, if we compare the results of two different types of calculations and first consider the La-based phases, it is obvious that after solidification there should be a La2Zn17 phase in the samples with La addition, based on equilibrium, but a LaSi2 phase based on the Scheil solidification calculations. This is confirmed by further analysis (DSC and metallography) to determine which La-based phase is actually present in the samples. The other main difference concerns the T-phase and the S-phase. The results of the equilibrium calculations show that there should be more S-phase in the samples after solidification, but the Scheil calculations show the opposite, that the amount of T-phase should be higher. And the last main difference concerns the Fe-based phase. The results of the equilibrium calculations show that the Al7Cu2Fe phase should be present, while the Scheil calculations show that the Al13Fe4 phase should be present. Since, these phases are well-known and have already been described and characterised by other researchers [1, 2, 5, 7,8,9, 13, 21], we will only confirm which phases are present in the investigated samples after solidification in the as-cast state by further analysis.

DSC analysis

In the DSC analysis, only the most meaningful diagrams are shown, just as in the CALPHAD calculations, i.e., the DSC cooling curves for samples 0, 2 and 4. However, in the following table (Table 8) we have compiled all the characteristic temperatures obtained from the DSC analysis. In the reference sample (Fig. 3a), an additional peak was determined at 620.1 °C, which belongs to the liquidus temperature of α-Al and was no longer visible when La was added.

Table 8 Characteristic temperatures in °C obtained from the DSC cooling curves of all samples investigated
Full size table
Fig. 3
figure 3

DSC cooling curves for sample 0 (a), sample 2 (b) and sample 4 (c)

Full size image

It is evident from the results that the addition of La does not drastically change the solidification interval (ΔT—Table 8), it remains almost unchanged. However, the addition of La lowers the liquidus temperature of samples 1, 2, 3 and 4 by 0.7, 1.7, 2.8 and 2.2 °C, respectively, compared to the reference sample (0). The results show that the liquidus temperatures decrease up to an addition of 0.14 mass% La and then increase again (sample 4). The same is true for the solidus temperature (1.8, 0.7, 2.7 and 0.9 °C for samples 1, 2, 3 and 4, respectively), but in this case there is no logical trend as the results vary greatly with the addition of La. On the other hand, the liquidus temperature of the Al13Fe4 phase starts to decrease with the addition of La (1.2, 3.5, 5.6 and 4.6 °C for samples 1, 2, 3 and 4, respectively). In this case, the trend is the same as for the liquidus temperature of the alloy. Moreover, the liquidus temperature increases for the Mg2Si phase (1.8, 12.4, 15.9 and 29.4 °C for samples 1, 2, 3 and 4 respectively). For the Mg2Si phase, there is a trend that the liquidus temperature of the Mg2Si phase increases with increasing La content. The reason for this is that when La is added, a new phase LaSi2 begins to form, which fundamentally changes the Mg2Si phase. This trend has already been demonstrated in previous studies [42,43,44]. In sample 2, only a few residues of the Mg2Si phase are present, the rest being LaSi2 phase, which is confirmed by further microstructural analyses. So, in sample 2, both phases solidify simultaneously. And in the samples with a La addition of more than 0.1 mass%, no Mg2Si phase was observed in the microstructure in the as-cast state. This means that the liquidus temperature determined in samples 3 and 4 (Table 8) is the liquidus temperature of LaSi2. This is also consistent with the CALPHAD calculations (Tables 4 and 6), which indicate that with the addition of La, the liquidus of the LaSi2 phase begins to increase and the liquidus of the Mg2Si phase begins to decrease. Furthermore, the mass fraction of the Mg2Si phase decreases and the mass fraction of LaSi2 increases with the addition of La (Table 7). In contrast, the liquidus temperature of the eutectic phase (α-Al + σ[Mg(Zn, Cu, Al)2]) is not affected. Overall, the addition of La has no influence on the solidification interval, but on the liquidus temperature of the Mg2Si (and LaSi2—in samples 3 and 4) and Al13Fe4 phase.

Since, we can also determine the energies released during cooling, we compile them in the following table (Table 9). It is obvious that the addition of La in sample 1 has no influence on the solidification energy. However, in samples 2 and 3, the solidification energy increases, while the addition of 0.17 mass% La decreases the solidification energy. The same trend is observed for the solidification energies of the eutectic α-Al + σ[Mg(Zn, Cu, Al)2] phase. As for the energies, the addition of 0.05 mass% La does not seem to affect them, but 0.1 and 0.14 mass% increase them and 0.17 mass% decrease them.

Table 9 Exothermic energies released on cooling of all the samples studied, for the solidification of the matrix and the eutectic α(Al) + σ[Mg(Zn, Cu, Al)2] phase
Full size table

Based on the results of the DSC analysis, we can confirm that the solidus temperatures determined with the Scheil solidification calculations are more accurate than those determined with the equilibrium calculations. But this was somehow to be expected, since the cooling rate during the DSC analysis was 10 K min−1. The liquidus temperatures are lower than the calculated ones, but this was also to be expected due to the undercooling during the DSC analysis. However, from the shape of the peaks of the Al13Fe4 and the Mg2Si/LaSi2 phases, we can see that with the addition of La, the first peak (Al13Fe4 phase) slowly becomes flatter, and the second peak (Mg2Si/LaSi2 phase) slowly becomes more pronounced (Fig. 3). This is related to the solidification of the La-based phase, more specifically the La2Si phase.

Metallography

Metallographic analysis was carried out to confirm the presence of phases in the microstructure in the as-cast state, as determined by CALPHAD calculations and DSC analyses. For this purpose, an EDS analysis of the phases was present was performed and the results are given in at%. In addition, EDS element mapping was carried out to clarify which phases were present, as the La phases were generally observed above the eutectic phase and the amounts of Zn, Mg and Cu were elevated, as shown. The stoichiometry of the phases differs because they are metastable non-equilibrium phases in the microstructure in the as-cast state. In addition, we also measured the crystal grain size. For sample 0, the average grain size was 114.2 µm, for sample 2 116.9 µm and for sample 4 119.3 µm. The results show that the grain size does not decrease due to the addition of La, but just the opposite. This is probably due to the fact that with the addition of La, a La-based phase (Al20Cr2La) starts to form, the size of which is too large to successfully apply grain refinement as intended (and as is the case with Al3Ti) when the phase first starts to solidify. This is also consistent with the literature [22,23,24] where no Cr content was present and the addition of La refined the grains. But unfortunately, the Al45Cr7 phase was not found in reference sample 0 because it solidifies first, and the size of the phase is too small to be successfully identified with the SEM analysis.

Figure 4 shows the EDS analysis of sample 0, where the eutectic phase was analysed. Based on the literature [1,2,3,4,5,6,7,8,9,10, 12,13,14,15,16, 18,19,20,21] and the phase composition in at%, this is the α(Al) + σ[Mg(Zn, Cu, Al)2] eutectic phase (Fig. 4—Spectrum 3), and we also analysed only the σ[Mg(Zn, Cu, Al)2] phase (Fig. 4—Spectrum 1, 2 and 4). In addition, the matrix was also analysed (Fig. 4—Spectrum 5).

Fig. 4
figure 4

The analysed areas (marked as spectrum in the table) with EDS are with the corresponding chemical composition in at% for sample 0

Full size image

To further confirm that this is indeed a eutectic α(Al) + σ[Mg(Zn, Cu, Al)2], we also made EDS element mapping images (Fig. 5). It can be seen that there are increased amounts of Mg, Zn and Cu in the phase (in the white areas of the phase) and Al in the darker areas (Fig. 5). This means that it is a eutectic phase, which is the last to solidify (last peak on the DSC curves during cooling—Fig. 3).

Fig. 5
figure 5

EDS element mapping images for sample 0

Full size image

To confirm other phases, present in the microstructure in the as-cast state, we analysed different areas of sample 0 (Fig. 6). Here we can see that the Al13Fe4 phase (Spectrum 6—Fig. 6) and the Mg2Si phase (Spectrum 8—Fig. 6) are present. However, as increased amounts of Zn, Mg and Cu are also present, we have also created EDS element mapping images to confirm these phases (Fig. 7). Meanwhile, we also found the eutectic α(Al) + σ[Mg(Zn, Cu, Al)2] phase in this area (Spectrum 7—Fig. 6), which we have already analysed (Figs. 4 and 5). An additional EDS analysis of the matrix was performed (Spectrum 9—Fig. 6).

Fig. 6
figure 6

The analysed areas (marked as spectrum in the table) with EDS are indicated with the corresponding chemical composition in at% for sample 0

Full size image
Fig. 7
figure 7

EDS element mapping images for sample 0

Full size image

EDS element mapping images of sample 0 (Fig. 7) were made in the same area as in Fig. 6 to confirm that there are three distinct phases: eutectic α(Al) + σ[Mg(Zn, Cu, Al)2], Mg2Si and Al13Fe4 in the microstructure of reference sample 0 in the as-cast state. To illustrate that it is indeed the Al13Fe4 phase (Spectrum 6—Fig. 6), the amounts of Fe and Al are increased at the same location (Fig. 7). The same applies to the Mg2Si phase (Spectrum 8—Fig. 6), as the amounts of Mg and Si are increased in the same place (Fig. 7).

Figure 8 shows the EDS analysis of sample 2, where all phases present in the microstructure in the as-cast state were analysed. Based on the previous results and phase composition Fig. 8—Spectrum 2 and 4 the eutectic phase is α(Al) + σ[Mg(Zn, Cu, Al)2], but in spectrum 4 there is a slightly increased La addition, which is further clarified with EDS element mapping images. The next phase is a La-based phase with increased Si content (probably LaSi2) and the presence of Mg (Fig. 8—Spectrum 3), but as the phase is above the eutectic phase, the content of other elements is also increased. Therefore, additional EDS element mapping was carried out to confirm the LaSi2 phase with increased Mg content. Another phase present is Al13Fe4 (Fig. 8—Spectrum 5). Another new phase besides LaSi2 appears in the microstructure with the addition of La and this phase is sharp-edged (Fig. 8—Spectrum 6 and 7). This phase has increased amounts of La, Cr and Ti, among other elements present in the matrix (Al, Zn, Mg and Cu). Based on at%, the stoichiometry of this phase cannot be determined, but based on the literature [45, 46] it could be a La-based phase with increased Cr content (probably Al20Cr2La, as no other phase with the combination of these chemical elements is reported in the literature), as, based on CALPHAD calculations (Tables 4 and 6), the Al45Cr7 phase starts to solidify first (it seems that it is the nucleus for the growth of the Al20Cr2La phase), the phase is a sharp-edged phase like Al45Cr7 [47,48,49] and is located in the α-Al crystal grains, which means that it solidifies first. But as mentioned earlier, the Al45Cr7 phase was not found in sample 0 because it is too small to be successfully identified by SEM analysis. To confirm this phase and the LaSi2 phase, we performed an EDS element mapping (Fig. 9). In addition, the matrix was also analysed (Fig. 8—Spectrum 8).

Fig. 8
figure 8

The analysed areas (marked as spectrum in the table) with EDS are indicated with the corresponding chemical composition in at% for sample 2

Full size image
Fig. 9
figure 9

EDS element mapping images for sample 2

Full size image

To confirm that La-based phases are present in the microstructure, Fig. 9 shows the EDS element mapping of the same location as in Fig. 8. The first is LaSi2, needle shape phase, as seen in Fig. 9, in which increased amounts of La and Si are present. So basically, Fig. 8—Spectrum 3 is the LaSi2 phase, but it still contains some amounts of Mg because La is starting to modify the Mg2Si phase, as mentioned earlier. Another phase based on La is the sharp-edged phase, which according to EDS is probably the Al20Cr2La phase and has an increased amount of Ti, as shown earlier (Fig. 8—Spectrum 6 and 7). All other phases present are eutectic α(Al) + σ[Mg(Zn, Cu, Al)2] and Al13Fe4.

The microstructure of sample 4 was also examined using the EDS analysis (Fig. 10). In this case, the phases present in the microstructure in the as-cast state are the same as for sample 2. Based on the previous results and the phase composition in at% (Fig. 10), Spectrum 1 is the eutectic phase α(Al) + σ[Mg(Zn, Cu, Al)2]. The next phase present is LaSi2 with some amounts of Mg (Fig. 10—Spectrum 2, 3 and 4), but as the phase is above the eutectic phase, the amount of other elements (Al, Zn, Cu, Mg and Ni) was again increased, just as in sample 2. The sharp-edged Al20Cr2La phase is also present (Fig. 10—Spectrum 5 and 6), again with increased amounts of other elements (Al, Zn, Mg, Cu and V). The matrix was also analysed (Fig. 10—Spectrum 7).

Fig. 10
figure 10

The analysed areas (marked as spectrum in the table) with EDS are indicated with the corresponding chemical composition in at% for sample 4

Full size image

EDS element mapping was also performed for the same area as in Fig. 10 (Fig. 11). The results show that two La-based phases are present in the microstructure. The first is needle-like LaSi2. Here, it is clear that the phase shown in Fig. 10—Spectrum 2, 3 and 4—is indeed LaSi2, as no Mg is present and the amounts of La and Si are increased. This also confirms the previous assumptions made when discussing the DSC results. In the samples above 0.1 mass % La, no Mg2Si phase is present anymore and is completely replaced by the LaSi2 phase. Furthermore, from the EDS element mapping, we can confirm that the sharp-edged Al20Cr2La phase (Fig. 10—Spectrum 5 and 6) is also present, as increased amounts of La and Cr are present. It also appears that higher amounts of Ti are dissolved in this phase, which is also consistent with the assumptions made earlier, as it is known that Ti is added to Al alloys (as grain refinement—Al3Ti, etc.) to reduce the size of α-Al grains as it nucleates prior to solidification of α-Al. This means that this phase (Al20Cr2La) starts to solidify first and based on the CALPHAD calculations, the first phase to solidify is Al45Cr7. So, this means that the addition of La not only changes the Mg2Si phase, but also the Cr-based phase mentioned earlier. No Al13Fe4 phase was found in this region. To confirm these results, we performed another EDS element mapping analysis (Fig. 12) to show that the mentioned phase is still present in the microstructure.

Fig. 11
figure 11

EDS element mapping images for sample 4

Full size image
Fig. 12
figure 12

EDS element mapping images for sample 4

Full size image

As mentioned earlier, Fig. 12 EDS element mapping images were taken only to confirm that the Al13Fe4 phase is still present in the microstructure of sample 4, as it was not present in Figs. 10 and 11. From the following results (Fig. 12), we can see that the Al13Fe4 phase is still present, while increased amounts of Fe and Al are present in the same phase. Here we also see the eutectic α(Al) + σ[Mg(Zn, Cu, Al)2] phase and some tiny needles of the LaSi2 phase. In sample 4, on the other hand, there was no Mg2Si phase.

To clarify the observation regarding the La phases: This is only a first insight into the phase stoichiometry of the La-based phases and needs to be further investigated with TEM and XRD analyses, but the phase stoichiometry is not crucial for the effects on solidification.

Conclusions

As far as solidification is concerned, the CALPHAD results show that the addition of La has minimal effect on the liquidus temperature. The solidus temperature increases slightly in the equilibrium calculations, while the solidification interval decreases slightly compared to the reference sample 0, although no significant differences are observed.

Similarly, the Scheil simulations show increased solidus temperatures, with smaller differences for the first three samples compared to the equilibrium temperatures. The addition of La does not significantly change the solidification, apart from a slight shrinkage of the solidification interval in sample 4.

In comparing the phases, we have noted some differences between the two types of calculations. Concerning the La-based phases, the equilibrium calculations predict the presence of a La2Zn17 phase after solidification in samples with La addition, while the Scheil solidification calculations suggest a LaSi2 phase. Another notable difference concerns the T-phase and the S-phase. Equilibrium calculations suggest a higher presence of S-phase after solidification, while Scheil calculations suggest a higher amount of T-phase. Finally, there is a discrepancy in the Fe-based phase. Equilibrium calculations predict the presence of the Al7Cu2Fe phase, while Scheil calculations suggest the presence of the Al13Fe4 phase.

The DSC results support the CALPHAD calculations and show that the addition of La does not drastically affect the liquidus and solidus temperatures. However, the addition of La lowers the liquidus temperature. The solidus temperatures also decrease with the addition of La, but the results show that there is no logical trend. The solidification interval remains largely unchanged. The liquidus temperature of the Al13Fe4 phase decreases with the addition of La, while the liquidus temperature of the Mg2Si phase increases. This is due to the formation of a new phase, LaSi2, which fundamentally changes the Mg2Si phase. In sample 2, only remnants of the Mg2Si phase are present, while the majority consists of the LaSi2 phase. Samples with more than 0.1 mass% of La do not show any Mg2Si phase in the microstructure in the as-cast state. The addition of La has no effect on the liquidus temperature of the eutectic phase (α-Al + σ[Mg(Zn, Cu, Al)2]). Another La-based phase, Al20Cr2La, starts to form with the addition of La. It modifies the Al45Cr7 phase and solidifies first. Interestingly, the grain size does not decrease with the addition of La, but rather increases. This is probably because the formation of the La-based phase (Al20Cr2La) occurs earlier, and its size is too large for effective grain refinement as intended.