Modeling and Experimental Results of Selected Lightweight Complex Concentrated Alloys, before and after Heat Treatment

Lightweight complex concentrated alloys (LWCCA), composed of elements with low density, have become a great area of interest due to the high demand in a large number of applications. Previous research on LWCCAs was focused on high entropy multicomponent alloy systems that provide low density and high capability of solid solution formation. Present research introduces two alloy systems (Al-Cu-Si-Zn-Mg and Al-Mn-Zn-Mg-Si) that contain readily available and inexpensive starting materials and have potential for solid solution formation structures. For the selection of appropriate compositions, authors applied semi-empirical criteria and optimization software. Specialized modeling software (MatCalc) was used to determine probable alloy structures by CALPHAD, non-equilibrium solidification and kinetic simulations. The selected alloys were prepared in an induction furnace. Specimens were heat treated to provide stable structures. Physicochemical, microstructural, and mechanical characterization was performed for the selected alloy compositions. Modeling and experimental results indicated solid solution-based structures in the as-cast and heat-treated samples. Several intermetallic phases were present at higher concentrations than in the conventional alloys. Alloys presented a brittle structure with compression strength of 486–618 MPa and hardness of 268–283 HV. The potential for uniform intermetallic phase distribution in the selected alloys makes them good candidates for applications were low weight and high resistance is required.


Introduction
Complex concentrated alloys (CCA) represent a new group of materials, with complex microstructures, comprising of a large number of elements [1]. CCAs extend the established limits of the high entropy alloys (HEAs) concept by allowing a smaller number of elements in the composition and in various proportions, that can deliver mixed structures containing solid solution and well dispersed intermetallic phases. CCAs generally have higher configurational entropy than conventional alloys and are capable of obtaining complex, disordered compositionally complex solid solution structures. CCAs are distinct from conventional alloys in that they are not based on a single major element and

Materials and Methods
In order to reach the desired physical and mechanical properties, it is very important to find a suitable composition of the alloy. The properties of CCAs are specifically influenced by the constitutive elements. A vast majority of lightweight high-entropy alloys contain elements like Al, Si, Mg, Ti, and B in order to reduce alloy density. In order to avoid alloy brittleness due to poor machinability, elements like Cu, Zn, and Mn were selected as most appropriate. Although, these elements are not known as having low density, they are able to form stable solid solutions with lighter elements. Thus, Mn favour formation of complex solid solutions and Cu induces high machinability of the material by increasing deformability.
Selection of most appropriate compositions was performed by the optimization of semi-empirical criteria. The semi-empirical criteria are defined by the following relations: a.
The entropy of mixing (∆S mix ) is calculated using Boltzmann's equation where R is the gas constant (8.314 J/mol·K) and c i is the molar fraction for element i, b.
The enthalpy of mixing (∆H mix ) is calculated with a formula derived from the Miedema macroscopic model [27] ∆H mix = 4c i c j × ∆H ij (2) where ∆H ij is the binary enthalpy of mixing for elements i and j c.
The atomic size difference (δ) was defined by [14] δ = 100 × c i × 1 − r i r 2 (3) where r i is the atomic radius of element i, r is the average atomic radius. d.
The derived parameter Ω, which includes the influence of both ∆S mix and ∆H mix [15], was calculated with where T m is the melting temperature calculated with T m = c i × T mi , and T mi is the melting temperature of element i. e.
The formula for electronegativity difference after Allen (∆χ) was deduced as follows [17] ∆χ = 100 × c i × 1 − χ i χ 2 (5) where χ i is the Pauling electronegativity of element i and χ is the average electronegativity. f.
The valence electron concentration (VEC) is used for determining the type of solid solution formed in the alloy. VEC is calculated with an equation based on element concentration and individual VEC [19] VEC = c i × VEC i (6) where VEC i is the valence electron concentration of element i. g.
The geometrical parameter (Λ) that defines the ration between mixing entropy and atomic size difference of the alloy [21] Λ = ∆S mix /δ 2 (7) The properties for the elements that enter the calculations were acquired from valuable sources, suggested by the authors that originally developed the criteria, such as: atomic radius [28], Allen electronegativity [29,30], melting temperatures [31], and pair mixing enthalpy [32].
The optimization of the alloy compositions was performed by means of the multi-objective optimization module from MATLAB software (version 6.02, MatCalc Engineering GmbH, Vienna, Austria).
MatCalc Pro edition, version 6.02, was used for multi-component phase equilibrium and thermodynamics by CALPHAD method, as well as non-equilibrium solidification and multi-phase precipitation kinetics.
The optimal alloy compositions were prepared in an induction furnace type Linn MFG-30 (Linn High Therm GmbH, Eschenfelden, Germany) with inert atmosphere and cast in a copper mould. Technical purity raw materials of Al, Cu, Si, Zn, Mg, and Mn were used in the experimental trials. A 350 g charge for each alloy was placed in an alumina-based crucible. The alloys were melted in the induction furnace and cast in a copper mould, under protective atmosphere. The resulted as-cast alloys were annealed in an electrical furnace, LHT 04/17 Nabertherm GMBH (Lilienthal, Germany), under protective atmosphere (Ar). The heat treatment stage was conducted at 400 • C for 20 h, with slow cooling rate of 3 • C/min. Samples were taken before and after the heat treatment process for chemical, structural, and mechanical analyses.
The chemical composition of the alloys was determined by inductively coupled plasma spectrometry (ICP-OES) using an Agilent 725 spectrometer (Santa Clara, CA, USA). Samples from various places of the alloy ingot were investigated. Optical microscopy investigation was performed with a Zeiss Axio Scope A1m Imager microscope (Jena, Germany) with bright field, dark field, DIC and polarization capabilities, and high contrast EC Epiplan 109/509/1009 lenses. Image analyzer software ImageJ vers. 1.53e, developed by W. Rasband, U. S. National Institutes of Health, Bethesda, MD, USA, was used to count porosity and shrinkage in the specimens. Samples were previously etched with a Keller type solution (prepared on-site) to enhance the visibility of the grains and the grain boundaries. Scanning electron microscopy (SEM) was performed with a FEI Quanta 3D FEG microscope (FEI Europe B.V., Eindhoven, Netherlands), operating at 20-30 kV, equipped with an energy dispersive X-ray spectrometer (EDS). The phase configuration was analyzed by X-ray diffractometry (XRD) (FEI Europe B.V., Eindhoven, Netherlands). Data acquisition was performed on BRUKER D8 ADVANCE diffractometer (Bruker Corporation, Billerica-MA, USA), using DIFFRAC plus XRD Commender (Bruker AXS) software (version 2018, Bruker Corporation, Billerica-MA, USA), Bragg-Brentano diffraction method, Θ -Θ coupled in vertical configuration, with the following parameters: CuKα radiation, 2Θ Region: 20-1240, 2Θ Step: 0.020, Time/step: 8.7 s/step, rotation speed 15 rot/min, Cukβ radiation was removed with SOL X detector (Bruker Corporation, Billerica, MA, USA). The resulting data was processed using Bruker ® Diffracplus EVA Release 2018 software to search the database ICDD ® Powder Diffraction File (PDF4+, 2019 edition) and the full pattern matching (FPM) module of the same software package. The semi-quantitative evaluation was performed by the RIR method (reference intensity ratio) of the identified phase concentrations.
Compression strength for both alloys was determined with a LBG testing machine, model TC-100 (LBG testing equipment srl, Azzanos, Paolo, Italy), with max load of 100 kN. The testing samples were alloy bars with 6 mm in diameter and 6 mm length. The compression speed was 6 mm/min. Vickers microhardness of the samples was measured at room temperature using microindenter attachment (Anton Paar MHT10, Anton Paar GmbH, GRAZ, Austria), with an applied load of 2 N and slope of 0.6 N/s. Ten measurements were made for each sample to determine the average values.

Criteria Calculation
The criteria calculations were performed for the Al-Cu-Si-Zn-Mg and Al-Mn-Zn-Mg-Si systems (Table 1) with the variation of each element composition in a relevant interval, appropriate to the high entropy concept. Each element from the alloy composition was varied separately from 0.5 to 2.5 molar concentrations, according to the high entropy alloy conventions (1 is the value for equal composition). The influence of element composition on the main parameters is illustrated in Figures 1 and 2. The optimal ranges for solid solution formation, as presented in the introduction section, are: Ω > 1.1, ∆χ Allen < 6% and Λ > 24 J/mol·K. The atomic size mismatch was not represented as is included in the Λ parameter. The range limit is shown on diagrams with a dotted line. The results show that, for a Al-Cu-Si-Zn-Mg system, Mg and Si are decreasing the probability for solid solution formation, while Al, Cu, and Zn have a positive effect. However, Cu and Zn are rather heavy elements and increase the alloy density considerably, which has to be taken into account in final alloy selection. The Al-Mn-Zn-Mg-Si system capability to form solid solutions is strongly improved by the increase of Al concentration. Mn and Zn have also a positive influence on the alloy structure, but less important than Al. Another contradiction is shown here between Ω and Λ parameter, for the Mn and Mg influence. As expected, Si is not a good solid solution former, similar to the effect on Al-Cu-Si-Zn-Mg system. studied for practical use. The influence of the composition elements on the structural behavior is important in material design for determining compositional areas of interest that can be attractive for different applications. Those areas can then be studied further for the development of optimal compositions with targeted properties.
Taking data acquired in the present research work into consideration, a general approach for the selection of optimal compositions with high probability in the formation of solid solution structures can be performed by the means of special algorithms of global multi-objective optimization process.   In order to select most appropriate compositions for both systems, the equations defined by the established criteria entered a multi-objective optimization process. There are several methods that can be applied to present case scenario for multi-objective optimization. The Pareto front method was selected for increased accuracy. The optimization problem was set for finding the minimum value for the electronegativity difference (Equation (5)) and alloy density (Equation (8)), and the maximum value for the derived parameter Ω (Equation (4)) and geometrical parameter Λ (Equation (7)). The density of the alloy was calculated with formula 1 = × 100 for different applications. Those areas can then be studied further for the development of optimal compositions with targeted properties.
Taking data acquired in the present research work into consideration, a general approach for the selection of optimal compositions with high probability in the formation of solid solution structures can be performed by the means of special algorithms of global multi-objective optimization process.
In order to select most appropriate compositions for both systems, the equations defined by the established criteria entered a multi-objective optimization process. There are several methods that can be applied to present case scenario for multi-objective optimization. The Pareto front method was selected for increased accuracy. The optimization problem was set for finding the minimum value for the electronegativity difference (Equation (5)) and alloy density (Equation (8)), and the maximum value for the derived parameter Ω (Equation (4)) and geometrical parameter Λ (Equation (7)). The density of the alloy was calculated with formula where ρ is the alloy density, ρ i is the component metal density, and w i is the concentration in weight percent (wt.%) of the component metal i.
The problem setup contained the constraints for optimal values, which correspond to the limits for each parameter presented earlier in the introduction section: Λ > 0. 24, Ω > 1.1 and ∆χ Allen < 6. The genetic algorithm solver was used with following options: double vector population, feasible population for creation function, adaptive feasibility for the mutation function, and single point crossover function.
Several optimized compositions were determined (Tables 2 and 3) and were illustrated in Figures 3 and 4. For the Al-Cu-Si-Zn-Mg system, several alloy compositions are complying with the solid solution formation criteria. Even if the electronegativity criterion is respected only by two compositions, the rest of the alloys are not far away from the solid solution formation limit. The lowest density was shown by the Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 alloy. The optimal compositions for the Al-Mn-Zn-Mg-Si system present limited availability for solid solution formation structures. Only three out of seven compositions are placed well in the required intervals. Due to the targeted application that originated the research study, an alloy with density close to 3 g/cm 3 was desired. The Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 has the lowest density and presents similar criteria values with the other low-density alloys.

CALPHAD Modeling
The CALPHAD method is using thermodynamic and kinetic data and equations to determine Gibbs free energies and diffusion mobility characteristics of a system [22]. Phase stability parameters and phase composition can be determined function of alloy composition and system temperature. Recent improvements of alloys databases allow for the obtaining of pertinent results when using CALPHAD based software.      Figure 5a) shows two different regions concerning the formation of hard intermetallic phases. Zn 2 Mg and Al 2 Cu based phases are in higher proportion bellow 316 • C and decrease significantly above this temperature. Presence of these compounds is determined by the high proportions of Zn, Mg, and Cu. At higher temperatures the The BCC-A2 and FCC-A1 disordered phases form at aprox. 488 • C through an invariant reaction from the liquid state. These structures were mentioned before in Cu and Si containing multicomponent alloys [33]. In the 316-488 • C range the alloy structure is composed mainly of disordered FFC-A1, BCC-A2 phases and multicomponent intermetallic Q-phase, which suggests that the alloy composition is favouring the complex structures at this temperature level. Modeling results show a near-equimolar composition of Al, Zn and Cu in the BCC-A2 phase and higher concentrations of Mg and Si in the Q-phase. The Al 2 Cu phase is also forming with a steady increase in concentration, in this temperature range. The BCC-A2 and Q-phase structures become unstable at approx. 316 • C participating in a second invariant transformation with the formation of stable Zn 2 Mg and Si-A4 phases. A substantial increase in the Al 2 Cu phase is also presented after this critical transformation. The ductile FCC-A1 phase is predominant all across the temperature range. For the Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloy (Figure 5b), two intermetallic phases of alpha-Al 9 Mn 2 Si and Mg 2 Si are shown to have great stability at lower and higher temperatures and form early during the solidification process. Same as in the previous alloy structure, the FCC_A1 phase is predominant at all temperatures, but starts forming after the intermetallic phases.

CALPHAD Modeling
The CALPHAD method is using thermodynamic and kinetic data and equations to determine Gibbs free energies and diffusion mobility characteristics of a system [22]. Phase stability parameters and phase composition can be determined function of alloy composition and system temperature. Recent improvements of alloys databases allow for the obtaining of pertinent results when using CALPHAD based software.
The resulted Al3.4Cu0.5Si0.2Zn0.5Mg0.2 phase diagram ( Figure 5a) shows two different regions concerning the formation of hard intermetallic phases. Zn2Mg and Al2Cu based phases are in higher proportion bellow 316 °C and decrease significantly above this temperature. Presence of these compounds is determined by the high proportions of Zn, Mg, and Cu. At higher temperatures the The BCC-A2 and FCC-A1 disordered phases form at aprox. 488 °C through an invariant reaction from the liquid state. These structures were mentioned before in Cu and Si containing multicomponent alloys [33]. In the 316-488 °C range the alloy structure is composed mainly of disordered FFC-A1, BCC-A2 phases and multicomponent intermetallic Q-phase, which suggests that the alloy composition is favouring the complex structures at this temperature level. Modeling results show a near-equimolar composition of Al, Zn and Cu in the BCC-A2 phase and higher concentrations of Mg and Si in the Q-phase. The Al2Cu phase is also forming with a steady increase in concentration, in this temperature range. The BCC-A2 and Q-phase structures become unstable at approx. 316 °C participating in a second invariant transformation with the formation of stable Zn2Mg and Si-A4 phases. A substantial increase in the Al2Cu phase is also presented after this critical transformation. The ductile FCC-A1 phase is predominant all across the temperature range. For the Al3Mn0.2Zn0.3Mg0.7Si0.8 alloy (Figure 5b), two intermetallic phases of alpha-Al9Mn2Si and Mg2Si are shown to have great stability at lower and higher temperatures and form early during the solidification process. Same as in the previous alloy structure, the FCC_A1 phase is predominant at all temperatures, but starts forming after the intermetallic phases.  The phase evolution function of element concentration was also calculated for both alloy systems. In the Al-Cu-Si-Zn-Mg system ( Figure 6), Cu content is critical for the formation of the Al2Cu phase, with a peak at 40 wt.% were solid solutions are completely supressed. A predominant BCC solid solution structure is found when Cu concentration is between 50 and 60 wt.%. The main influence of Zn content is shown by the transition between the FCC and BCC at higher Zn concentrations. The total proportion of the intermetallic based phases remains constant with the increase of Zn content, with the transition between the formation of Al2Cu and Zn2Mg phases. The phase evolution function of element concentration was also calculated for both alloy systems. In the Al-Cu-Si-Zn-Mg system ( Figure 6), Cu content is critical for the formation of the Al 2 Cu phase, with a peak at 40 wt.% were solid solutions are completely supressed. A predominant BCC solid solution structure is found when Cu concentration is between 50 and 60 wt.%. The main influence of Zn content is shown by the transition between the FCC and BCC at higher Zn concentrations. The total proportion of the intermetallic based phases remains constant with the increase of Zn content, with the transition between the formation of Al 2 Cu and Zn 2 Mg phases.
systems. In the Al-Cu-Si-Zn-Mg system (Figure 6), Cu content is critical for the formation of the Al2Cu phase, with a peak at 40 wt.% were solid solutions are completely supressed. A predominant BCC solid solution structure is found when Cu concentration is between 50 and 60 wt.%. The main influence of Zn content is shown by the transition between the FCC and BCC at higher Zn concentrations. The total proportion of the intermetallic based phases remains constant with the increase of Zn content, with the transition between the formation of Al2Cu and Zn2Mg phases. Higher concentrations of Mg have a strong influence over the formation of intermetallic based phases, mainly the Zn 2 Mg intermetallic compound, which becomes very stable over 8 wt.% Mg. Because most of the Mg is taken for the formation of Zn 2 Mg compound, the Si content does not influence the concentration of intermetallic based phases. The proportion of the FCC solid solution decreases with the increase of the Si content, which is typical for the Al-Si alloys.
The Al-Mn-Zn-Mg-Si alloy system ( Figure 7) is mainly defined by the presence of Mn, which has an important contribution to the formation of complex compound-based phases. Thus, the increase of the Mn content induces a significant increase in the Al 9 Mn 2 Si type-complex compound phase, which reaches a peak of 0.7 phase fraction at just 20 wt.% Mn, after which decreases abruptly at the exchange of the AlMnSi phase. Meanwhile, the proportion of FCC and BCC phases suffer a major decrease, with under 0.1 phase fraction at 20 wt.% Mn. Zn influence on the phase stability is similar to the Al-Cu-Si-Zn-Mg system, with the difference consisting in the formation of Mg 2 Si phase instead of Al 2 Cu phase. At lower Zn concentrations, Mg 2 Si is more stable than Zn 2 Mg. The FCC-A1 phase is decreasing strongly at the exchange of the BCC phase formation and the Al 9 Mn 2 Si type phase is dominant at up to 55 wt.% Zn. As in the Al-Cu-Si-Zn-Mg system, the increase in Mg concentration determines a high increase on the intermetallic compounds proportion, thus the Mg 2 Si compound is predominant after 20 wt.% Mg. As expected, the influence of Si content in the phase evolution of Al-Mn-Zn-Mg-Si structure is very similar to the Al-Cu-Si-Zn-Mg system.
The phases formed during non-equilibrium solidification, modeled with Scheil-Gulliver method, are shown in Figure 8 increase of the Mn content induces a significant increase in the Al9Mn2Si type-complex compound phase, which reaches a peak of 0.7 phase fraction at just 20 wt.% Mn, after which decreases abruptly at the exchange of the AlMnSi phase. Meanwhile, the proportion of FCC and BCC phases suffer a major decrease, with under 0.1 phase fraction at 20 wt.% Mn. Zn influence on the phase stability is similar to the Al-Cu-Si-Zn-Mg system, with the difference consisting in the formation of Mg2Si phase instead of Al2Cu phase. At lower Zn concentrations, Mg2Si is more stable than Zn2Mg. The FCC-A1 phase is decreasing strongly at the exchange of the BCC phase formation and the Al9Mn2Si type phase is dominant at up to 55 wt.% Zn. As in the Al-Cu-Si-Zn-Mg system, the increase in Mg concentration determines a high increase on the intermetallic compounds proportion, thus the Mg2Si compound is predominant after 20 wt.% Mg. As expected, the influence of Si content in the phase evolution of Al-Mn-Zn-Mg-Si structure is very similar to the Al-Cu-Si-Zn-Mg system. The phases formed during non-equilibrium solidification, modeled with Scheil-Gulliver method, are shown in Figure 8. The behavior of the phases with termination S describes the cumulative solidification patterns, containing equilibrium and non-equilibrium values. The equilibrium and non-equilibrium solidification are similar for the Al3Mn0.2Zn0.3Mg0.7Si0.8 alloy (Figure  8a), while in the Al3.4Cu0.5Si0.2Zn0.5Mg0.2 alloy (Figure 8b) small differences are revealed between solidification temperatures. Obviously, the Al3.4Cu0.5Si0.2Zn0.5Mg0.2 alloy presents a pattern characteristic to castable aluminum alloys, having also a smaller solidification temperature than Al3Mn0.2Zn0.3Mg0.7Si0.8 ( Figure 9). FCC-A1 type solid solution solidifies first in the Al3.4Cu0.5Si0.2Zn0.5Mg0.2 alloy, while the solidification of Al3Mn0.2Zn0.3Mg0.7Si0.8 alloy is mostly defined by the behavior of Al9Mn2Si and Mg2Si intermetallic phases. The AlMnSi phase is unstable at lower temperatures and transforms into Al9Mn2Si at later stages.

Kinetics Simulation
The heat treatment of Al3.4Cu0.5Si0.2Zn0.5Mg0.2 and Al3Mn0.2Zn0.3Mg0.7Si0.8 alloys was studied by the identification of hard phase precipitation in the FCC_A1 matrix. It can be seen from Figure 10 that the Al2Cu precipitate has the highest fraction, while the precipitate radius is small and uniform across the annealing interval. The Q-phase is also shown in the Al3.4Cu0.5Si0.2Zn0.5Mg0.2 precipitation diagram, but in lower proportion. The precipitation kinetics of the Al3Mn0.2Zn0.3Mg0.7Si0.8 alloy (Figure 11) shows a close competition between the Al9Mn2Si-type and Mg2Si compounds. Nevertheless, the size of the complex compound is significantly smaller than that of the Mg2Si. The Si_A4 phase is most Al3.4Cu0.5Si0.2Zn0.5Mg0.2 alloy, while the solidification of Al3Mn0.2Zn0.3Mg0.7Si0.8 alloy is mostly defined by the behavior of Al9Mn2Si and Mg2Si intermetallic phases. The AlMnSi phase is unstable at lower temperatures and transforms into Al9Mn2Si at later stages.

Kinetics Simulation
The heat treatment of Al3.4Cu0.5Si0.2Zn0.5Mg0.2 and Al3Mn0.2Zn0.3Mg0.7Si0.8 alloys was studied by the identification of hard phase precipitation in the FCC_A1 matrix. It can be seen from Figure 10 that the Al2Cu precipitate has the highest fraction, while the precipitate radius is small and uniform across the annealing interval. The Q-phase is also shown in the Al3.4Cu0.5Si0.2Zn0.5Mg0.2 precipitation diagram, but in lower proportion. The precipitation kinetics of the Al3Mn0.2Zn0.3Mg0.7Si0.8 alloy (Figure 11)

Kinetics Simulation
The heat treatment of Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 and Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloys was studied by the identification of hard phase precipitation in the FCC_A1 matrix. It can be seen from Figure 10 that the Al 2 Cu precipitate has the highest fraction, while the precipitate radius is small and uniform across the annealing interval. The Q-phase is also shown in the Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 precipitation diagram, but in lower proportion. The precipitation kinetics of the Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloy ( Figure 11) shows a close competition between the Al 9 Mn 2 Si-type and Mg 2 Si compounds. Nevertheless, the size of the complex compound is significantly smaller than that of the Mg 2 Si. The Si_A4 phase is most likely found as eutectic in both alloys. Even if the AlMnSi precipitate is showing on the diagram, appears to be a nonequilibrium phase at 400 • C and is not stable during the heat treatment stage.

Experimental Results
The experimental chemical composition of the alloys is presented in Table 4. The resulted alloy specimens have each element composition within a maximum of 2 wt.% interval from the nominal values. Due to the high percentage of the elements, a variation of 2 wt.% in composition has little influence on the structural behavior of the alloy, and the analyses implied in the paper has the meaning in offering a mainly qualitative and not standardized information related to the studied alloys characteristics.
Optical analyses of the alloy samples (Figures 12 and 13) revealed significantly different morphologies between as cast and heat-treated states. Optical microscopy results for the as cast Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 alloys showed a fine and uniform dispersed dendritic structure (Figure 12a). Several phases are found in the interdendritic area, including occasional eutectic formations. The Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 heat treated alloy (Figure 12b) shows a dendritic structure with several micropores, mainly placed in the interdendritic area. In average, less than 5 vol % porosity was found in the specimens. The dendrite size is significantly larger than in the as-cast alloy, with well-defined interdendritic phases. Eutectic formations were also identified in the heat-treated sample. The as-cast and heat treated Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 samples (Figure 13a,b) show a similar fine dendritic structure.        Microstructural analyses through scanning electron microscopy ( Figures 14 and 15), shows large difference between the as-cast and heat-treated alloys. The as-cast Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 (Figure 14a) is formed of multiple phases of similar size and uniform distribution in the material. After heat treatment the alloy exhibits a large dendrite structure and well-defined interdendritic eutectic (Figure 14b). The composition of Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 phase structures, for as-cast and heat-treated states are presented in Table 5. Results for EDS analyses of Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 alloy reveals high concentration of Al and Zn in the dendritic area (DR). Cu is found mainly in the interdendritic area (ID), forming well defined phases with Al, Zn, and Mg (ID1, ID2, ID3, ID4, and ID5). The eutectic formation investigated with EDS-mapping (Figures 16 and 17), is composed of Al and Zn based constituents. Mg and Si are found concentrated in common regions, most likely forming intermetallic compounds.     Table 5).

Experimental Results
The interdendritic area indicates differences in terms of phase size and distribution, both samples revealing multiple phases. Phase composition determined by EDS analysis (Table 6) showed Al-Mn and Mg-Si based dendritic formations. Three separate interdendritic phases, with various concentrations of the elements (ID1, ID2, and ID3), were distinguished in the as-cast sample. Al is present in larger quantity in the interdendritic area forming a continuous phase with Zn. An occasional phase composed of Al, Mn, Si, and Zn was also identified in the as-cast sample. Si was also identified segregated in the interdendritic area. Similar compositions were found in the heattreated sample, setting aside a minor phase with large Zn concentration. The EDS-mapping results (Figures 18 and 19) confirm the previous findings. The well-dispersed hard phases are composed mostly of Si and Mg. Si segregations appear as long branches in certain regions. Al and Mn are found highly concentrated in the dendritic area.   (Table 5).      The interdendritic area indicates differences in terms of phase size and distribution, both samples revealing multiple phases. Phase composition determined by EDS analysis (Table 6) showed Al-Mn and Mg-Si based dendritic formations. Three separate interdendritic phases, with various concentrations of the elements (ID1, ID2, and ID3), were distinguished in the as-cast sample. Al is present in larger quantity in the interdendritic area forming a continuous phase with Zn. An occasional phase composed of Al, Mn, Si, and Zn was also identified in the as-cast sample. Si was also identified segregated in the interdendritic area. Similar compositions were found in the heat-treated sample, setting aside a minor phase with large Zn concentration. The EDS-mapping results (Figures 18 and 19) confirm the previous findings. The well-dispersed hard phases are composed mostly of Si and Mg. Si segregations appear as long branches in certain regions. Al and Mn are found highly concentrated in the dendritic area.      The mechanical testing results for both alloys in as-cast and heat treated states are presented in Table 7, Figures 22 and 23. The Al3.4Cu0.5Si0.2Zn0.5Mg0.2 alloy presented a slightly higher resistance (590 MPa in as-cast and 618 MPa in heat treated state) than the Al3Mn0.2Zn0.3Mg0.7Si0.8 alloy (486 MPa in ascast and 507 MPa in heat treated state).   The mechanical testing results for both alloys in as-cast and heat treated states are presented in Table 7, Figures 22 and 23. The Al3.4Cu0.5Si0.2Zn0.5Mg0.2 alloy presented a slightly higher resistance (590 MPa in as-cast and 618 MPa in heat treated state) than the Al3Mn0.2Zn0.3Mg0.7Si0.8 alloy (486 MPa in ascast and 507 MPa in heat treated state).  The mechanical testing results for both alloys in as-cast and heat treated states are presented in Table 7     Both alloys showed a brittle behavior with elongation under 3%. The microhardness tests showed high values in comparison with the conventional aluminum alloys (A357.0 and A5083). The as-cast samples presented a fine and well distributed phase structure, which allowed for multiple determinations with similar values, averaged at 268 HV for Al3.4Cu0.5Si0.2Zn0.5Mg0.2 alloy and at 277 HV for the Al3Mn0.2Zn0.3Mg0.7Si0.8 alloy. However, the annealed structures presented large and welldefined phases, so microhardness indentations revealed individual values for each distinct phase. The rounded proportional average of the determined values for the annealed alloys were: 274 HV for Al3.4Cu0.5Si0.2Zn0.5Mg0.2 and 283 HV for Al3Mn0.2Zn0.3Mg0.7Si0.8. The results showed similar microhardness values between the two alloy states.

Discussion
The thermodynamic and Hume-Rothery rules for the formation of solid solution structures offer a good estimate for the selection of most appropriate compositions, but cannot fully predict real life    Both alloys showed a brittle behavior with elongation under 3%. The microhardness tests showed high values in comparison with the conventional aluminum alloys (A357.0 and A5083). The as-cast samples presented a fine and well distributed phase structure, which allowed for multiple determinations with similar values, averaged at 268 HV for Al3.4Cu0.5Si0.2Zn0.5Mg0.2 alloy and at 277 HV for the Al3Mn0.2Zn0.3Mg0.7Si0.8 alloy. However, the annealed structures presented large and welldefined phases, so microhardness indentations revealed individual values for each distinct phase. The rounded proportional average of the determined values for the annealed alloys were: 274 HV for Al3.4Cu0.5Si0.2Zn0.5Mg0.2 and 283 HV for Al3Mn0.2Zn0.3Mg0.7Si0.8. The results showed similar microhardness values between the two alloy states.

Discussion
The thermodynamic and Hume-Rothery rules for the formation of solid solution structures offer a good estimate for the selection of most appropriate compositions, but cannot fully predict real life

Discussion
The thermodynamic and Hume-Rothery rules for the formation of solid solution structures offer a good estimate for the selection of most appropriate compositions, but cannot fully predict real life experimental results. The targeted alloys were intentionally annealed to eliminate inherent out of equilibrium structures that may appear after the casting process.
Selection criteria applied to the Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 alloy showed optimal values for most of the parameters. The electronegativity difference was the only parameter that was above the recommended limit. By comparison, the selected Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloy presented most of the criteria parameters out of the recommended limits, which suggests a structure with a higher percentage of intermetallic compounds. For both alloys, the experimental findings matched the criteria calculations and showed a high percentage of the FCC-Al phase and presence of intermetallic compounds. For the Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloy was found a significantly larger number of intermetallic phases. CALPHAD method produced similar phase configurations with those obtained in the experimental results, with slight changes for low proportion phases. Thus, Si suggested by MatCalc was not found in the Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 heat treated sample, instead Mg 2 Zn 11 was detected at X-ray analysis. The Q-phase was determined as a stable phase in the studied alloy and was not indicated by the analyses of the experimental samples. Thermodynamic calculations showed presence of Zn 2 Mg which were not detected in the Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 heat treated sample. The Al 9 Mn 2 Si phase shown in MatCalc simulation is very similar to the experimental Al 4.01 MnSi 0.74 . It is known that the stoichiometry of the Al 4.01 MnSi 0.74 phase is similar to the stoichiometry of the α-Al 9 Mn 2 Si phase, thus indicating a similar crystal structure [34]. The proportion of phases was also well represented by the CALPHAD method.
There is also a good representation of the solidification behavior of the selected alloys obtained with the Scheil-Gulliver method. The resulted diagram for Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 alloy (Figure 8a) showed that the first phase to form is FCC-A1, which is indicated as a dendrite formation in the optical and SEM results. The Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloy has a more complex structure with two dendrite formations (Mg 2 Si and Al 9 Mn 2 Si), which are also indicated by the solidification simulation diagram (Figure 8b).
Precipitation kinetics was applied to determine the structural changes in the as-cast structure over the annealing process. MatCalc software was setup to predict the nucleation and growth of the precipitated intermetallic phases during the heat treatment process, starting with the as-cast structure. The simulation results for Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 alloy indicated a high proportion of the Al 2 Cu precipitate, which was maintained at the same value from the as-cast structure. The results were verified by X-ray analyses. The number density of Al 2 Cu decreases from as-cast to heat treated state, which remains consistent with the increase in the precipitate size, shown both by the simulation and SEM results. However, the size of the precipitate is much larger in the experimental specimens. The simulation results for the Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 shows important discrepancies against the experimental findings. The Mg 2 Si precipitate is in much larger proportion than the Al 9 Mn 2 Si precipitate in the MatCalc simulation, which is in contradiction with the X-ray analyses results. Still, there are similarities regarding the high stability of the Mg 2 Si precipitate number density and size, which can be observed from the simulation and microstructural characterization images.
The SEM-EDS and XRD results show a larger number of phases for the as-cast specimens. The complex M phase in Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 and complex phase Al 10 (Mn 0.58 Zn 0.24 Si 0.18 ) 3 in Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 are no longer present in the annealed state. This illustrates the metastable structure that is present in the as-cast samples and that slow diffusion processes are taking place in complex concentrated alloys. There is also a large difference between the behavior of the selected alloys during the annealing process. The Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 alloy suffers large transformations in phase size and distribution, similar to the castable aluminum alloys, while the Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 presents small changes in structural configuration, similar to the wrought aluminum alloys.
The mechanical properties of the alloys showed relatively high values for compression yield strength and hardness, in comparison with conventional aluminum alloys. This was expected due to the higher content of intermetallic phases. Unfortunately, the stress strain curves showed a brittle behavior that is detrimental for practical use. Even if the structural change is significant after annealing, with a significantly larger grain size, the Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 retains similar values for the mechanical properties. This aspect could be analysed further in future studies. The Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloy presents no significant changes in grain size after annealing (typical to wrought alloys) and consequently the mechanical properties present similar values for both alloy states.
Overall, the tools used for the prediction of the lightweight CCAs structures were in good agreement with the experimental findings. Small differences shown between simulation and experimental results indicate inherent errors that may appear when calculating multicomponent alloys with a large proportion of elements. Elemental diffusion plays an important role in the evolution of the alloys structure, leading to the formation of complex phases, which are hard to be determined by conventional analysis methods and also difficult to be simulated by present software.
The selected compositions show a good potential for obtaining practical, low cost LWCCA, with improved properties. The substantial increase in the alloy's elements composition does not show dramatical changes in the structural behaviors. The intermetallic phases induced by the higher composition of the reactive elements, typical for LWCCA, are not necessarily detrimental to the final properties. A uniform distribution and reduced size of the precipitates can lead to a stable structure with high mechanical properties at higher operation temperatures. However, future studies are recommended for further compositional tuning and customised heat treatment processes to obtain optimal configurations for targeted applications.

Conclusions
The selection and analyses of low weight complex concentrated alloys, with common and less expensive elements, before and after heat treatment, were presented in the paper. The selection of alloys with low density and high solid solution content from Al-Cu-Si-Zn-Mg and Al-Mn-Zn-Mg-Si systems was achieved by the optimization of previously established semiempirical criteria. Results showed that a high aluminum content determined better solid solution forming ability for both alloy systems. While Cu, Zn, and Mn have a beneficial influence on the criteria parameters, the increase in element content affects the density of the alloy. A balance between density and criteria optimal values, modeled with an optimization software, was applied to offer practical solutions in the selection process.
The selected Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 and Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloys were analysed for structural behavior by CALPHAD, solidification and kinetics simulations. Phase diagrams showed the formation of a multiphase structure with preponderant solid solutions. Intermetallic compound-based phases were found at lower concentrations but still significantly higher than in the conventional aluminum alloys. Two invariant transformations were indicated for Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 alloy, with a BCC-A2 phase stable between 316-488 • C. The influence of the elements on the structural evolution were similar with the results obtained from semiempirical criteria calculations. The non-equilibrium solidification simulation discussed the order of which the phases are solidifying in normal casting conditions. The soft FCC phase is stabilizing first in the Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 alloy, while intermetallic compound phases Mg 2 Si and Al 9 Mn 2 Si are forming first in the Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloy. The diffusion driven precipitation simulation offered indications on the formation of intermetallic precipitates in the solid solution matrix, showing that the Al 2 Cu and Mg 2 Si phases will be stabilizing at highest concentration levels for the Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 and Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloys, respectively.
The experimental findings, provided by optical, SEM-EDS, and XRD analyses, indicated that both alloys presented complex structures, containing mostly solid solutions, but also a significant amount of intermetallic phases. The Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 presented a refined structure in the as-cast state with a predominant Al solid solution phase. The Al 2 Cu and Mg-Zn intermetallics were also present in the alloy structure and continued to remain at a high percentage in the heat-treated samples. A dendritic structure based on FCC-Al phase was indicated in the Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 alloy. The Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloy showed a refined structure that was maintained also after the annealing process. This time the dendritic formations were indicated to be intermetallic compound phases: Mg 2 Si and Al 4.01 MnSi 0.74 . Between the as-cast and heat-treated state, a transformation of less stable Al 10 (Mn 0.58 Zn 0.24 Si 0.18 ) 3 into Al 4.01 MnSi 0.74 was revealed, which lead to a significant increase in the concentration of the Al 4.01 MnSi 0.74 compound (32 wt.%). The mechanical characterisation of the alloys revealed a high compression strength for both alloys (aprox 600 Mpa for Al 3.4 Cu 0.5 Si 0.2 Zn 0.5 Mg 0.2 and aprox. 500 MPa for Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 ) and brittle behavior. The microhardness was found in the range of hard aluminum alloys (250-300 HV).
The comparison of the experimental results with the criteria and simulation results showed good representations. There were small differences in phase composition and precipitation behavior between the CALPHAD equilibrium diagrams for both alloys, kinetics simulation, and experimental results for the Al 3 Mn 0.2 Zn 0.3 Mg 0.7 Si 0.8 alloy. These relative inconsistencies could be resolved in the future development of the software by improving the alloys database and the diffusion simulation, which are difficult to achieve for multicomponent alloys with reactive elements.
The intention of this study was to provide a path in the selection of multicomponent CCAs, containing less expensive raw materials, for lightweight applications, using tools that are readily available. The optimization criteria can be easily modified to fit required properties. The alloying examples presented in the paper were studied from several angles to give an indication of the expected structural behavior in this type of alloys.