Chemical composition of the 354 alloys used in this study.
The present study focused on the tensile properties at ambient and high temperatures of alloy 354 without and with the addition of zirconium. Tensile tests were performed on alloy samples submitted to various aging treatments, with the aim of understanding the effects of the addition made on the tensile properties of the alloy. Zirconium reacts only with Ti, Si, and Al in the alloys examined to form the phases (Al,Si)2(Zr,Ti) and (Al,Si)3(Zr,Ti). Testing at 25°C reveals that the minimum and maximum quality index values, 259 and 459 MPa, are observed for the as-cast and solution heat-treated conditions, respectively. The yield strength shows a maximum of 345 MPa and a minimum of 80 MPa within the whole range of aging treatments applied. The ultimate tensile and yield strength values obtained at room temperature for T5-treated samples stabilized at 250°C for 200 h are comparable to those of T6-treated samples stabilized under the same conditions, and higher in the case of elevated-temperature (250°C) tensile testing. Coarsening of the strengthening precipitates following such prolonged exposure at 250°C led to noticeable reduction in the strength values, particularly the yield strength, and a remarkable increase in the ductility values.
- aluminum alloys
- thermal exposure
- tensile testing
The 354 alloy belongs to the Al-Si-Cu-Mg system similar to B319 alloy that is widely used for automotive engine blocks . The high silicon content in the 354 alloy improves the alloy castability whereas the presence of Cu and Mg noticeably enhances the yield strength (YS) and the ultimate tensile strength (UTS) of the 354 alloy due to the formation of intermetallic phases, mainly Al2Cu or eutectic Al + Al2Cu, and Mg2Si precipitates [2, 3]. However, segregation behavior of Cu may lead to incipient melting during solution treatment which will apparently reduce the alloy strength . Addition of Mg has a strong affinity to react with Sr, leading to the formation of a complex Mg2SrAl4Si3 intermetallic phase, and hence reducing the effectiveness of Sr as a Si modifying agent . In the absence of Cu, high Fe and Mg contents lead to the formation of π-FeMg3Si6Al8 phase which is difficult to dissolve during the solution treatment process [6, 7]. In the quaternary Al-Si-Cu-Mg alloy system, Q-phase (Al4Mg8Cu2Si6) can coexist with the Al2Cu, Mg2Si, and Si phases depending on the levels of Cu, Mg, and Si [8, 9, 10, 11]. The different factors that may influence the mechanical behavior of cast aluminum alloys are schematically represented in Figure 1 .
Zirconium may be added to Al alloys in order to refine the grain structure due to the presence of fine coherent dispersoids (mainly Al3Zr) which obstruct dislocation motion and in turn, enhance the elevated temperature mechanical properties of aluminum alloys . In order to increase the volume fraction of Al3Zr precipitates and based on the phase diagram of Al-Zr, the concentration of Zr in the alloys investigated in this study was kept at around 0.3 wt.% .
The main purpose of solution heat treatment is to obtain a supersaturated solid solution at high temperatures (below the eutectic temperature). As a result, a homogeneous supersaturated solid solution (SSSS) will form through dissolving the precipitated phases during the solidification process, such as β-Mg2Si, θ-Al2Cu, Q-Al5Cu2Mg8Si6, π-Al9FeMg3Si5 and β-Al5FeSi phases. The β-Mg2Si and θ-Al2Cu phases can be easily dissolved when the optimum solution heat treatment temperature and time are employed. The solution treatment temperature is determined according to the alloy composition and solid solubility limit; however, it must be lower than the melting point of the phases that exist in the as-cast structure to avoid incipient melting of these phases [15, 16].
Following the empirically developed concept of quality index proposed by Drouzy et al. [17, 18] Cáceres proposed a mathematical model emphasizing the significance of the quality index as follows [17, 19, 20]:
where quality index Q can be calculated using the relative quality index (q), strain-hardening exponent (n), and the strength coefficient (K).
The present study was undertaken to explore the effect of Zr addition and aging conditions of the as cast tensile bars on:
Characterizing the microstructural features of the investigated alloys,
Exploring the tensile properties at both ambient and elevated temperatures, and
Correlating the tensile properties to the microstructural features to establish the strengthening or softening mechanisms responsible for the observed properties.
It should be noted here that the term “temperature” applies to aging temperatures as well as testing temperature.
2. Experimental procedure
Alloy 354 modified with 200 ppm of strontium (using Al-10% Sr master alloy) and grain refined using 0.20 wt.%Ti (Al-5%Ti-1%B) was used as the base alloy (alloy A). To this alloy, 0.3%Zr in the form of Al-25wt.%Zr master alloy was added (alloy B). The chemical compositions of both alloys are listed in Table 1. Figure 2 shows the microstructure of the as-received base alloy ingots. Melting and casting procedures were carried out as described elsewhere.
To prepare test bars for the tensile tests, three samples for chemical analysis were also taken at the time of the casting; this was done at the beginning, in the middle, and at the end of the casting process to ascertain the exact chemical composition of each alloy. The experimental work was divided into two stages: Stage I in which the 354 alloy (alloy A) was used, and Stage II where the 354 alloy with 0.3%Zr (alloy B) was used. In Stage I, the melt temperature was kept around 750°C, whereas in Stage II, the melt temperature was superheated to 800°C, to ensure the complete decomposition of the Al-25%Zr master alloy used.
2.1 Stage I-alloy A
Tensile bars were solution heat treated at 495°C for 8 h, followed by quenching in warm water at 60°C, after which artificial aging was applied according to the plan listed in Table 2. After aging, the test bars were allowed to cool naturally at room temperature (25°C). All of the samples, whether as-cast, solution heat-treated, or aged, were tested to the point of fracture using an MTS servo-hydraulic mechanical testing machine at a strain rate of 4 × 10−4 s−1.
|Alloy code||Element (wt.%)|
|Temperature (°C)||Aging time (h) and aging condition codes|
The yield strength (YS) was calculated according to the standard 0.2% offset strain, and the fracture elongation was calculated as the percent elongation (%El) over 50 mm gauge length, as recorded by the extensometer. The ultimate tensile strength (UTS) was also obtained from the data acquisition system of the MTS machine. The average %El, YS, or UTS values obtained from the five samples tested per condition were considered to be the values representing that specific condition. An extensometer, or strain gage was used in the tests to measure the extent of deformation in the samples.
Samples for metallography were sectioned from the tensile-tested bars of all the alloys studied, about 10 mm below the fracture surface. The percentage porosity and eutectic Si-particle characteristics were measured and quantified using an optical microscope linked to a Clemex image analysis system. The microstructures of the polished sample surfaces were examined using an Olympus PMG3 optical microscope. Phase identification was carried out using electron probe microanalysis (EPMA) in conjunction with wavelength dispersive spectroscopic (WDS) analysis, using a JEOL*JXA-889001WD/ED combined microanalyzer operating at 20 kV and 30 nA, where the electron beam size was ~2 μm.
Mapping of certain specific areas of the polished sample surfaces was also carried out where required, so as to show the distribution of different elements within the phases. The fracture surfaces of tensile-tested samples were also examined using the same SEM, employing the backscattered electron (BSE) detector and EDS system. The fracture behavior was analyzed using the backscattered electron (BSE) images obtained, and analysis of the EDS spectra of phases observed on the fracture surface. Differential scanning calorimetry (DSC) was used to characterize the sequence of reactions occurring during the heating and/or cooling cycles of an alloy sample during a DSC scan which continuously changes with the increasing or decreasing temperature cycle to produce peaks according to the two expected reactions:
Phase formation → heat emission → exothermic peak
Phase dissolution → heat absorption → endothermic peak
2.2 Stage II-alloy B
For the high temperature tensile tests, samples from selected conditions were tested to fracture using an Instron Universal mechanical testing machine at a strain rate of 4 × 10−4 s−1. The heating furnace installed on the testing machine is an electrical resistance, forced-air box type, having the dimensions 30 × 43 × 30 cm. The yield strength (YS) was calculated according to the standard 0.2% offset strain, and the fracture elongation was calculated as the percent elongation (%El) over the 25.4 mm gauge length as recorded by the extensometer. The ultimate tensile strength (UTS) was obtained from the data acquisition system of the universal machine. In order to reach and stabilize the intended test temperature during the tests, at the time that the samples were mounted in the tensile machine, the furnace was already pre-set at the required temperature; also, these samples were kept mounted in the furnace of the tensile testing machine for 30 min before the start of every test.
3. Results and discussion
3.1 Stage I-alloy A
Figure 3 shows the macrostructure revealing the grain size for alloy A, about 200 μm. A complete modification of the silicon particles in the microstructure of alloy A in the as-cast condition can be seen in Figure 4(a). From Figure 4(a) and (b), solution heat treatment has changed the morphology of the silicon particles from faceted to globular. As a consequence of solution heat treatment, there may also be observed a reduction in the number of silicon particles and a reduction in the density of the silicon phase, due to the diffusion of silicon into the aluminum matrix. The white arrows in Figure 4(a) show the rounded shape of the dendrites with grain refining , whereas Figure 4(b) reveals the dissolution of the Al2Cu phase observed in Figure 4(a)—circled.
Zhu and Liu  proposed a model of the granulation of unmodified eutectic Si composed of three major stages during heat treatment: (i) the mass transport of solute, (ii) a discontinuous phase fragmentation, and lastly (iii) spheroidization. During heat treatment, the silicon atoms in the matrix at the Si particle tips diffuse to locations on the curved surfaces of the particles, leading to the dissolution of eutectic silicon at the tips. This transport of silicon atoms ultimately causes the fragmentation and spheroidization of eutectic silicon which is important from strength point of view compared to Si particles with sharp edges which act as sites for stress concentration.
The values of secondary dendrite arm spacing (SDAS), porosity, modification level, and grain size for both the as-cast (AC) and solution heat-treated (SHT) condition are listed in Tables 3 and 4. As can be seen, SHT resulted in (i) no noticeable change in both the SDAS and grain size, (ii) a significant decrease in the particle density due to coarsening of the eutectic Si particles, and (iii) almost complete solubility of Al2Cu in the aluminum matrix. Since the solutionizing temperature was well below the incipient melting temperature, tensile test bars revealed negligible change in the amount of porosity, i.e., no incipient melting.
|Alloy code condition||Area (%)||Particle length (μm)||Roundness ratio (%)||Aspect ratio||Density (particles/mm2)|
Figure 5 illustrates the effect of aging treatment on the alloy strength parameters. The main observations inferred from this figure can be summarized as follows:
Solution heat treatment and artificial aging at 190°C for 2 h or at 155°C for 100 h resulted in an increase in the alloy strength by ~64% over its as-cast strength.
Aging at 155 or 170°C for a long period of time offered maximum resistance to softening.
The greatest decrease in tensile strength occurred at 240°C (312 MPa at 2 h to 240 MPa at 100 h). Similarly, a significant decrease in strength took place upon aging at 190°C for a lengthy time (from 382 MPa at 2 h to 314 MPa at 100 h) indicating the end of peak-aging or the commencement of over aging.
The greatest reduction in the alloy UTS and YS levels resulted when the tensile bars were aged at a temperature as high as 350°C even for a short period of 2 h.
In comparison to the ascendant and steady strength curves corresponding to aging temperatures of 155°C and 170°C, fluctuations in the strength curves were observed at aging temperatures of 190°C and over, similar to that reported by Reif  where a similar alloy was used and an ascendant monotonic strength curve was observed at an aging temperature of 155°C.
Although the highest ductility values were obtained after 2 h aging at 350°C (~5%), none of the aging conditions reached the higher ductility values exhibited by the solution heat-treated condition (~6.5%). This observation may be considered evidence that the mechanical behavior displayed by alloy A is common to that of the Al-Si-Cu-Mg alloys whose strength is obtained at the expense of ductility [24, 25].
In order to analyze the alloy quality by means of the Quality Index charts, the as-cast and the solution heat treated conditions plus aging conditions at 155°C, 190°C, and 350°C for aging times in the range of 2–100 h were used. From a previous study , K was calculated as 500 MPa.
The plastic strain and the quality index (Q ) both exhibit a great improvement following solution heat treatment. The fact that plastic deformation (q) was about 0.31 in the solution-treated condition means that the alloy reached 31% of its maximum quality index value (Q ). The importance of q is that it shows how much a sample is away from its maximum possible ductility q = 1 and indicates that it would be possible to control the microstructure, for example by reducing the SDAS, or the porosity, or intermetallic level to enhance the alloy ductility and hence, the quality index, Q. When the ductility increases sharply from the as-cast to the solution heat treated condition, such changes can be related to the spheroidization of silicon particles and to the uniformity of the microstructure in the solution heat-treated condition, as shown in Figure 6(a).
From the data presented in Figure 6(b)–(d), it is evident that the change in crystallographic structure of Al2Cu phase from G-P zones (155°C) to a metastable phase (190°C) to a stable phase (350°C) is the main parameter controlling the alloy performance quality. As can be seen, at each aging temperature, all points fall within a narrow circle due the progress in the formation of the precipitated phase. The broken lines in these figures show the change in the Q-level as a function of aging temperature. The width of the circle deceased from 175 MPa (155°C) to 75 MPa (190°C) to 25 MPa (350°C), representing the hardening and softening behavior of the alloy as a function of the aging temperature and time . Using aging times of 2 and 100 h as reference points, the Q , UTS and %El values are presented in Table 5. As can be seen, the Q values after 2 h are more-or-less same over such a large range of aging temperatures, due to the variation in both UTS and %El. However, aging for 100 h revealed highest value at 190°C compared to 155°C (under aging) and 350°C (over aging). The Q values for test bars aged at 350°C for 100 h is the same due to the balance between UTS and %El.
|Aging temperature (°C)||Q (MPa)||UTS (MPa)||%El||Aging time (h)|
3.2 Stage II-alloy B
The heat treatment procedures followed for the alloy B are listed in Table 6. The same treatments were applied for both 25°C and 250°C tensile testing.
|Heat treatment procedures and parameters|
|Heat treatment||Solution treatment||Quenching||Aging|
|SHT *||495°C for 5 h||Warm water (60°C)||NA|
|T5 temper||N/A||N/A||180°C for 8 h|
|T6 temper||495°C for 5 h||Warm water (60°C)||180°C for 8 h|
Figure 7  shows the DSC heating curves of the alloys in the as-cast and SHT conditions, where three explicit peaks could be detected and coded 1, 2, 3. Considering the main parameter is the precipitation of Al2Cu phase particles, thus the height of peak number 1 following SHT compared to that in the as-cast condition plays a crucial role in controlling the alloy performance after aging. In addition, it is an indication of the effectiveness of the SHT process in dissolving the initial Al2Cu phase. In Figure 7, peak # 1 after solutionizing is more or less negligible due to dissolution of most of the Al2Cu phase, as shown in Figure 4(b).
The principle phases seen in alloy B are demonstrated in the optical as well as backscattered (BSE) images displayed in Figure 8(a) and (b) , respectively. Figure 8(a) exhibits α-Al dendrites separated by eutectic silicon colonies. The phases observed in Figure 8(b) were identified using EDS analysis and reference to the results of Hernandez-Sandoval  and Garza-Elizondo . Selective EDS spectra identifying these phases are displayed in Figure 8(c) through Figure 8(d). The existence of Al2Cu phase in the block-like form may be attributed to the presence of Sr. in the alloy which leads to segregation of copper to localized areas . The platelets of the Fe-rich β-Al5FeSi phase are easily recognized in the BSE image, surrounded by the blocky Al2Cu particles. The Mg-rich Q-phase (Al5Cu2Mg8Si6) is found growing out of the Al2Cu phase as seen in the BSE image. The absence of coarse Al3Zr precipitates  may be related to superheating that led to considerable dissolution of the Al3Zr phase from the master alloy during the melting process. As a result, the coarse Zr-containing phases are rarely detected since Al3Zr particles act as nucleation spots for these coarse phases. According to Garza-Elizondo , coarse Zr-rich particles may be nucleated on the undissolved Al3Zr particles provided by the master alloy, i.e., Al-15 wt.%Zr. In the present study, superheating the melt to 800°C would significantly reduce the numbers of Al3Zr particles in the matrix. The predicted fine zirconium trialuminide (Al3Zr) dispersoids that may be present on a nanoscale would require a high magnification BSE image to be detected.
Figure 9(a) shows a bright-field (BF) TEM image obtained in a T6-treated sample of alloy B with the electron beam parallel to the  zone axis. This figure shows a high density of uniformly distributed needle-like precipitates which are oriented along <110> family of directions and aligned along the  planes. The length of these precipitates ranges from 50 to 150 nm, close to the reported size of θ′-Al2Cu plates (50–100 nm long) [32, 33]. Figure 9(b) displays the associated selected area electron diffraction (SAED) pattern obtained from Figure 9(a). The observable discrete diffraction maxima for the precipitates in SAED pattern indicate the presence of θ′-Al2Cu, where the streaks most probably result from the presence of fine S′-Al2CuMg particles. Computer simulation studies [34, 35, 36, 37] on the S′-phase reflections show that they are hidden within the streaks of θ′.
In the present work, Figure 10(a), the addition of ~0.3 wt.%Zr to the 354-type Al-Si-Cu-Mg cast alloy in the as-cast condition improves the ambient-temperature (25°C) strength values of the Zr-free 354 alloy (alloy A), by ~26 MPa (UTS) and 40 MPa (YS), respectively. Following SHT, the UTS and ductility values remained almost constant at ~300 MPa and ~ 6.3%, respectively, while the yield strength increased by ~33 MPa compared to alloy A. It is believed that the improved strength values of alloy B emphasizes the role of Zr addition in enhancing the ambient-temperature tensile properties through the formation of fine secondary strengthening precipitates (Al3Zr) as reported by many authors [14, 38, 39, 40]. The fact that UTS and YS in the T5 and T6 conditions are very close may be attributed to the strengthening effect of the fine dispersoids, which precipitate during the artificial aging stage of the T5, and T6 treatments as reported in Figure 9.
Tensile testing at 250°C, endured a significant softening due to the possible coarsening of the strengthening precipitates (Al2Cu) that existed during tensile testing at room temperature (Figure 11). In addition, the T5 heat treatment did not improve the elevated-temperature strength values of the as-cast alloys but reduced the alloy ductility by ~50%. However, application of the T6 heat treatment noticeably enhanced the strength values of the as-cast condition from about 175 to 225 MPa. Another parameter to consider is the effect of thermal stability. In the present work, some tensile samples were stabilized at 250°C (following the T5 and T6 aging treatment) for a lengthy period of time, i.e., 100 and 200 h. As can be seen from Figure 10(c), the stabilized T5-treated alloy B samples exhibit better strength values (UTS and YS) than those obtained in the stabilized T6-treated condition. However, the ductility values obtained after stabilization of the T5-treated samples are dramatically lower in comparison. Figure 11(a) shows Al2Cu particle size and distribution in a T6 sample stabilized at 250°C for 200 h, whereas Figure 11(b) is the corresponding EDS spectrum.
A detailed investigation of the fracture surfaces of tensile bars of alloy B were examined in the T6-treated condition, before and after stabilization for 200 h at 250°C. The T6-temper treatment was selected due to its wide use in the automotive industry. The BSE image shown in Figure 12(a)  shows the fracture surface of the tensile-tested alloy in the T6-treated condition. The fracture surface has a dimpled-structure throughout, which indicates the ductile nature of the fracture mode. In addition, the BSE image exhibits the precipitation of Alx(Zr,Ti)Si compound, in the form of star-like shape, as confirmed by the associated EDS spectrum in Figure 12(b). Also, cracks can be spotted in various particles of this compound, as indicated by the arrows. The higher magnification BSE image shown in Figure 11(c) reveals a cracked Alx(Zr,Ti)Si phase particle.
Figure 13(a)  shows the fracture surface of the T6-treated B alloy tested at 250°C after stabilization for 200 h at the testing temperature. The dimple structure is coarser compared to that before stabilization at 250°C. This observation would explain the improved ductility of the alloy due to the softening behavior associated with the prolonged elevated-temperature exposure at 250°C. Coarsened precipitates appear in the interiors of the dimples, as indicated by the oval contours in Figure 13(a). The BSE image and the EDS spectrum shown in Figure 13(b) and (c), respectively, confirm the presence of Alx(Zr,Ti)Si phase particles.
Based on an analysis of the results presented in this article, the following conclusions may be made:
For the base 354 alloy A, solution heat treatment and artificial aging at 190°C for 2 h or at 155°C for 100 h resulted in an increase in the alloy strength by ~64% over its as-cast strength. Aging at 155 or 170°C for a long period of time offered maximum resistance to softening.
The Zr-rich intermetallic phases appear in two different forms, namely (Al,Si)2(Zr,Ti) in block-like form, and containing high level of silicon, and (Al,Si)3(Zr,Ti) in needle-like form, containing high level of aluminum.
Quality index (Q) charts constructed for alloy 354 characterize the tensile properties in terms of the heat treatment conditions applied. Minimum and maximum Q values, i.e., 259 and 459 MPa, are observed for the as-cast and solution heat-treated conditions, respectively; the yield strength shows a maximum of 345 MPa and a minimum of 80 MPa within the range of aging treatments applied.
DSC runs carried out on alloy B (354 alloy + 0.3%Zr) revealed peak patterns which included differences in peak heights—which reflected the amount of the precipitated phase, and shifts in the transformation temperature.
Melt superheating at 800°C is beneficial in terms of reducing the amount of coarse Zr-rich phases in the alloy structure, as it provides efficient dissolution of the Al3Zr phase from the master alloy during the melting process. Coarse Zr-containing phases are rarely observed due to the limited number of Al3Zr particles available to act as nucleation sites for these coarse phases.
TEM investigations confirm that the investigated alloys are strengthened primarily by θ-Al2Cu and S-Al2CuMg precipitates and their precursors, in addition to a secondary strengthening effect by precipitates in the form of Alx(Zr,Ti)Si which form following the addition of Zr.
Prolonged exposure at 250°C, resulting in coarsening of the strengthening precipitates, causes noticeable reduction in strength values, particularly the yield strength (cf. 160 and 325 MPa), and a remarkable increase in the ductility values (cf. 6.3 and 1.1%).
The strength values (UTS and YS) obtained at room temperature for the stabilized T5-treated alloy samples are comparable to those of the stabilized T6-treated condition, and higher in the case of elevated-temperature tensile testing.
The fracture surface of the T6-treated alloy B after stabilization for 1 h at 250°C reveals a dimpled-structure throughout, indicating the ductile nature of the fracture mode.
The Alx(Zr,Ti)1-xSi complex compound is observed with star-like and blocky morphologies, with cracks appearing in various particles of this compound. By increasing the stabilization time up to 200 h, coarser and deeper dimples are formed, highlighting the improved ductility of the alloy due to the softening behavior associated with the prolonged exposure at 250°C.