A Quantitative Phase Analysis by Neutron Diffraction of Conventional and Advanced Aluminum Alloys Thermally Conditioned for Elevated-Temperature Applications

As the issue of climate change becomes more prevalent, engineers have focused on developing lightweight Al alloys capable of increasing the power density of powertrains. The characterization of these alloys has been focused on mechanical properties and less on the fundamental response of microstructures to achieve these properties. Therefore, this study assesses the quality of the microstructure of two high-temperature Al alloys (A356 + 3.5RE and Al-8Ce-10Mg), comparing them to T6 A356. These alloys underwent thermal conditioning at 250 and 300 °C for 200 h. Time-of-flight neutron diffraction experiments were performed before and after conditioning. The phase evolution was quantified using Rietveld refinement. It was found that the Si phase grows significantly (13–24%) in T6 A356, A356 + 3.5RE, and T6 A356 + 3.5RE alloys, which is typically correlated with a reduction in mechanical properties. Subjecting the A356 3.5RE alloy to a T6 heat treatment stabilizes the orthorhombic Al4Ce3Si6 and monoclinic β-Al5FeSi phases, making them resistant to thermal conditioning. These two phases are known for enhancing mechanical properties. Additionally, the T6 treatment reduced the vol.% of the cubic Al20CeTi2 and hexagonal ᴨ-Al9FeSi3Mg5 phases by 13% and 23%, respectively. These phases have detrimental mechanical properties. The Al-8Ce-10Mg alloy cubic β-Al3Mg2 phase showed significant growth (82–101%) in response to conditioning, while the orthorhombic Al11Ce3 phase remained stable. The growth of the beta phase is known to decrease the mechanical properties of this alloy. These efforts give valuable insight into how these alloys will perform and evolve in demanding high-temperature environments.


Introduction
Advanced aluminum (Al) alloys capable of operating in elevated-temperature environments, a subject of ongoing research and development in the transportation industry, pose a significant design challenge for material engineers.Conventional Al alloys are renowned for their high strength-to-weight ratio, low density, excellent creep resistance, and good castability, making them a compelling choice for various applications [1][2][3][4].Al-Si conventional alloys, such as A356 and B319, have emerged as key players in the transportation industry, particularly in the aerospace and automotive sectors, by providing a solution to reduce weight and combat climate change by reducing greenhouse gas emissions [3, [5][6][7][8].These clear benefits have encouraged the use of Al-Si-based alloys for geometrically complex cast automotive components, such as transmission cases, engine blocks, and turbines.However, the automotive industry's needs are evolving.Internal combustion engines are being phased out because they emit significant amounts of greenhouse gas emissions, contributing to climate change.One of the most promising technologies to replace internal combustion engines is replacing their fuel source with hydrogen.Hydrogen-fueled internal combustion engines share similar manufacturing and infrastructure requirements but with the added benefit of producing H 2 O as a combustion byproduct rather than harmful greenhouse gases.However, the ignition point of hydrogen (587 • C) is significantly higher than gasoline (260 • C).Additionally, hydrogen internal combustion engines require a compression ratio closer to the lower end of diesel engines, much higher than conventional spark-ignition ones.This additional temperature and pressure pose a challenging material problem [9,10].Most engine blocks today are made up of Al alloys for the benefits previously stated.Moreover, Al alloys are some of the most resistant materials to gaseous H 2 embrittlement, which would be good for such applications.However, in the conversion to hydrogen internal combustion engines, a significant problem arises from the microstructural thermal instability of conventional Al alloys.
Fundamentally, Al-Si-based alloys rely on intermetallics such as cubic Mg 2 Si and Si phases for their strengthening benefits [3,[11][12][13].Heat treatments can manipulate the volume percentages (vol.%) and morphologies of these phases for additional strengthening benefits, one of the most common being a T6 condition (solutionizing + peak aging) [14].However, this manipulated microstructure also limits the ability of Al-Si-based alloys to maintain their mechanical properties in elevated-temperature applications.In operating environments above 200 • C, Mg 2 Si and Si phases have limited coarsening resistance [11,15]; therefore, most internal combustion engines today have an operating temperature limit under 200 • C [16].Once these phases begin to coarsen, the alloy loses its dispersoid and solid solution strengthening effects as Mg and Si migrate from the solid solution and coarsen the Si and Mg 2 Si phases.The coarsening is typically subtle but can be significantly detrimental to mechanical properties.For example, Stroh et al. [17] tested a T6 A356 alloy in tension at 250 • C and found that the alloy exhibited an ultimate tensile strength (UTS) and yield strength (YS) of just 66 and 63 MPa, respectively.When the testing temperature was increased to 300 • C, the UTS and YS were reduced by over 50% to 31 and 25 MPa, respectively.At ambient temperature, an A356 alloy with a T6 heat treatment typically exhibited a UTS and YS of 234 and 165 MPa, respectively [18].This inability of conventional Al alloys to perform adequately when exposed to elevated temperatures severely limits their applicability for hydrogen internal combustion engines; the high ignition point of H 2 will most certainly lead to a significant increase in operating temperatures above 200 • C.
Therefore, over the last decade, material scientists have found that some of the research in the 20th century on alloying Al with rare-earth (RE) elements can help with the issue of insufficient thermal stability in Al alloys.For example, in 1999, Belov et al. [19] conducted elevated-temperature tensile tests on several Al alloys with cerium (Ce, a light rare-earth metal) and nickel (Ni) additions.Targeting 350 • C, they found that an Al-12Ce-5Ni (wt.%) alloy had a 75% increase in UTS over an A339 (Al-12.0Si-2.3Cu-1.0Mg in wt.%), which was conventionally used as a piston alloy then.The enhanced thermal stability was attributed to the thermally stable binary eutectics (orthorhombic Al 4 Ce, now known as Al11Ce 3 and Al 3 Ni).These findings have been the basis of most modern-day research focused on alloying Al with rare-earth (RE) elements [1,7,[20][21][22][23][24][25][26][27][28][29][30].For example, Aghaie et al. [31] found that adding 0.1 wt.% Ce to a B319 alloy increased the material's UTS and YS at 250 • C by 7% and 14%, respectively.The increases in UTS and YS at 250 • C were attributed to the formation of Al 3 Ce 4 Si 6 and AlCeSi 2 phases.The added Ce chemical also refined the eutectic Si phase, which helped enhance the UTS and YS at 250 • C.
Such benefits are also found with much higher additions of RE elements in conventional Al alloys.For example, Stroh et al. [1,17], in a two-part study, investigated the alloy A356 with 3.5% RE mischmetal (~50% cerium, 26% lanthanum, 16 wt.% neodymium, and 6 wt.% praseodymium) additions into an A356 alloy.In the first study, the authors of [17] found that casting A356 with 3.5 wt.% additions of RE and subjecting the alloy to a T6 heat treatment resulted in a 133% increase in YS at 250 • C. Additionally, the modified alloy experienced a 158% increase in YS at 300 • C over the conventional T6 A356 alloy.
The increased mechanical properties were attributed to the thermally stable orthorhombic AlSiRE and cubic Al 20 Ti 2 RE phases as well as the spheroidization of the Si phase after heat treatment.The second study [1] took the same A356 alloy with 3.5 wt.% RE mischmetal and refined the ratio of magnesium-manganese (Mg-Mn), two elements already present in A356.The Mg content decreased from 0.49 to 0.25 wt.% while the Mn content increased from 0.10 to 0.41 wt.%.This compositional modification refined the harmful iron-bearing intermetallics, resulting in increases in the UTS, YS, and modulus of elasticity of 9%, 14%, and 10%, respectively, at 250 • C compared to the unmodified A356 + 3.5% RE.One of the more profound insights of this second study is how Ce interacts in complex Al alloy systems.
Studies like these are prompting research into utilizing the Al-Ce alloying system as a base for next-generation powertrain alloys [5,7,18,23,32,33].In the last 5 years, extensive research has gone into the fundamentals of the Al-Ce binary system to understand the platform from which a new alloying system can be structured.The solidification characteristics and phase analysis of hypoeutectic [20,25,34], eutectic [25,[35][36][37], and hypereutectic [20,25,36,38,39] binary alloy compositions revealed that the orthorhombic Al 11 Ce 3 phase has excellent castability and thermal stability up to 500 • C. In a hypereutectic composition, primary Al 11 Ce 3 formation occurs.If the composition of Ce exceeds 16 wt.%, the primary phase begins to crack due to the large coefficient of thermal expansion discrepancy [33].Therefore, most studies focus on the near-eutectic composition to avoid any possible decrease in mechanical properties.These eutectic alloys have a lamellae eutectic Al-Al 11 Ce 3 phase that can retain up to 80% of its hardness when exposed to 500 • C for 168 h [24,35,37].For comparison, the Al-Si eutectic composition only retains ~50% of its hardness when exposed to the same experiment [37].Weiss et al. [40] took the Al-Ce binary system further and tested two ternary Al-Ce alloys with magnesium (Mg) additions, each with 8 wt.% Ce but varying Mg content (7 vs. 10 wt.%).Both alloys were tested at ambient temperature and 260 • C. The alloy containing 7 wt.% Mg exhibited a UTS and YS of 195 and 151 MPa, respectively.The alloy with 10 wt.% Mg exhibited a UTS and YS of 227 and 186 MPa, respectively.When tested at 260 • C, the 7 wt.% Mg alloy retained 69% of its UTS and 80% of its YS.The 10 wt.% Mg alloy retained 60% of its UTS and 70% of its YS.An important point to note is that these alloys were not conditioned (i.e., exposed to 260 • C for an extended period) before testing.
Thermal conditioning (sometimes attached to terms like 'over-aging') is crucial in the elevated-temperature testing of Al alloys.If an alloy is not thermally conditioned, the microstructure is not stabilized.Testing an instability in a microstructure leads to mechanical properties that do not accurately represent the alloy at the intended operating temperature.This effect becomes even more critical when the application requires longterm exposure to elevated temperatures.Studies have shown that thermally conditional Al-Si-based alloys between 15 min and 200 h can lead to a 20-50% decrease in mechanical properties compared to non-conditioned alloys [2,41].
The investigation into the microstructures of these alloys before and after thermal conditioning has been a sparse area in research.Even studies that do promote the concept of thermal conditioning tend to focus more on the mechanical properties and typically only assess the microstructure via surface techniques such as optical microscopy, scanning electron microscopy (SEM), X-ray diffraction (XRD), and electron backscatter diffraction (EBSD).Although these techniques are conventional in the materials engineering world, their assessment of quantitatively analyzing bulk phases of a material can be skewed by surface preparation, texture within the material, and the statistical reliability of the area chosen to present the microstructure by the researcher.These drawbacks can be significantly improved using time-of-flight neutron diffraction for bulk phase analysis.Utilizing time-of-flight neutron diffraction allows for the account of how the texture of each phase, even in small amounts, contributes to changes in intensity peaks [42].This account of texture results in a much more accurate Rietveld refinement and a much more accurate bulk phase analysis [43].
Therefore, this study aims to take the most promising advanced Al alloys (A356 + 3.5RE and Al-8Ce-10Mg) and compare them to a conventional T6 A356 via a quantitative phase analysis (QPA) using time-of-flight neutron diffraction.This QPA study will identify the growth and dissolution effects that thermal conditioning has on stabilizing the microstructure of these alloys at 250 and 300 • C, two desirable temperatures sought after for operating temperatures of hydrogen internal combustion engines.

Materials and Methods
This section provides relevant details on the materials, sample manufacturing parameters, neutron diffraction experimentation specifics, Rietveld analysis (using MAUD, version 2.9993), and ThermoCalc TM software packages (2024a, Thermo-Calc Software, Solna, Sweden) used to characterize the alloys' microstructures.These details make this study reproducible for others interested in using it to expand the concepts of the characterization of other materials for engineering purposes.

Materials and Sample Preparation
This study investigated three alloys of interest: A356, A356 + 3.5RE, and Al-8Ce-10Mg.These alloys were cast at Eck Industries in Manitowoc, WI, USA.A356 ingots were used to cast the A356 samples and fabricate the A356 + 3.5% RE alloy.The 356 + 3.5% RE alloy was fabricated by introducing the RE mischmetal to an A356 melt before mixing with an impeller at 250 RPM, similar to Stroh et al. in [17].The alloy was degassed with argon for 20 min before being poured into an ASTM permanent tensile mould preheated to 400 • C. The Al-8Ce-10Mg alloy was cast from prepared ingots that were melted, degassed, and poured, replicating the same process as the A356 and A356 + 3.5RE alloys.The composition of each alloy is shown in Table 1.All three alloys were cast as tensile bars at Eck Industries and provided to the High-Performance Powertrain Materials (HPPM) laboratory at the University of British Columbia.At the HPPM laboratory, one A356 tensile bar and one A356 + 3.5RE tensile bar were T6 heat treated.The T6 heat treatments consisted of solution annealing at 538 • C (1000 • F) for 8 h, quenching in water heated to 78 • C (172 • F), held at ambient temperature for 12 h, and then aged at 154 • C (310 • F) for 4 h, after which the alloys were cooled naturally to room temperature.
At the HPPM laboratory, all as-cast (AC) and T6 alloys were machined into cylindrical samples 10 mm in diameter and 12 mm tall from the gauge section of the tensile bars.One sample from each alloy was thermally conditioned at 250 • C, while another was conditioned at 300 • C for 200 h to stabilize the microstructure at these respective temperatures.The cylinders were then sent for QPA at the Los Alamos Neutron Science Center (LANSCE) in Los Alamos, NM, USA.

Neutron Diffraction
The neutron diffraction experiment was conducted at a short-pulsed spallation neutron source at the LANSCE [44], utilizing the High-Pressure-Preferred Orientation (HIPPO) neutron time-of-flight diffractometer [45,46].HIPPO utilizes 1200 3 He detector tubes arranged on 45 panels, covering 51.7% of 4π by conducting three scans per sample (rotating the sample in between scans about the vertical axis by 0 • , 67.5 • , and 90 • ) [46,47]; the latter was employed for this study.Samples were glued on Cd-wrapped sample holders (shielding diffraction from the holder material) and measured for 15 min at 100 µA proton beam current (thus adjusting for possible proton beam fluctuations) per rotation angle.The raw data were processed using the Material Analysis Using Diffraction (MAUD) software (version 2.9993) and the E-WIMV algorithm, which utilized a resolution of 7.5 • to derive the orientation distribution function from the diffraction data and calculate pole figures [45,48].Both texture and phase fraction refinements were conducted to collect an accurate QPA of each material and thermal condition.Figure 1 below shows that the alloys exhibited extremely weak textures (E-WIMV, resolution 7.5 • ); therefore, random texture was assumed for the phase analysis.The absence of texture shows that reflections do not vary significantly; therefore, the phase analysis can be conducted with high levels of accuracy.
Materials 2024, 17, x FOR PEER REVIEW 5 of 20 tubes arranged on 45 panels, covering 51.7% of 4π by conducting three scans per sample (rotating the sample in between scans about the vertical axis by 0°, 67.5°, and 90°) [46,47]; the latter was employed for this study.Samples were glued on Cd-wrapped sample holders (shielding diffraction from the holder material) and measured for 15 min at 100 µA proton beam current (thus adjusting for possible proton beam fluctuations) per rotation angle.The raw data were processed using the Material Analysis Using Diffraction (MAUD) software (version 2.9993) and the E-WIMV algorithm, which utilized a resolution of 7.5° to derive the orientation distribution function from the diffraction data and calculate pole figures [45,48].Both texture and phase fraction refinements were conducted to collect an accurate QPA of each material and thermal condition.Figure 1 below shows that the alloys exhibited extremely weak textures (E-WIMV, resolution 7.5°); therefore, random texture was assumed for the phase analysis.The absence of texture shows that reflections do not vary significantly; therefore, the phase analysis can be conducted with high levels of accuracy.

Thermodynamic Modelling in ThermoCalc TM
This study utilized ThermoCalc TM equipped with the TCAL9 Al database to determine the phase evolution of all three alloys (A356, A356 + 3.5RE, and Al-8Ce-10Mg) during solidification.Equilibrium and Scheil (nonequilibrium) simulations were performed to understand the effect of solidification rates on the microstructure of each alloy.The equilibrium solidification simulations assume thermal equilibrium at any temperature without considering the effect of time.They provided valuable insight into each alloy composition's anticipated heat treatment (T6) or thermal conditioning outcomes.The Scheil simulations assume the complete mixing of the liquid, the equilibrium is at the phase boundary between the solid and liquid phases, and there is no back diffusion from the solid to the liquid phases.Scheil solidification simulation explains how nonequilibrium cooling can trap solute in the matrix, creating supersaturation.This information is valuable to this study because when heat treatments or thermal conditioning is applied, these concentrated solute atoms (either in the matrix or in the intermetallics) migrate and alter phase amounts, which is the focus of characterization in this study.

Thermodynamic Simulations
The Scheil (nonequilibrium) solidification diagrams for the A356 composition from Table 1 are shown in Figure 2. The Scheil diagram shows that the phases of significant volume percentages are Al and Si.In minor quantities, these are the Mg2Si, ᴨ-Al9FeMg5Si3,

Thermodynamic Modelling in ThermoCalc TM
This study utilized ThermoCalc TM equipped with the TCAL9 Al database to determine the phase evolution of all three alloys (A356, A356 + 3.5RE, and Al-8Ce-10Mg) during solidification.Equilibrium and Scheil (nonequilibrium) simulations were performed to understand the effect of solidification rates on the microstructure of each alloy.The equilibrium solidification simulations assume thermal equilibrium at any temperature without considering the effect of time.They provided valuable insight into each alloy composition's anticipated heat treatment (T6) or thermal conditioning outcomes.The Scheil simulations assume the complete mixing of the liquid, the equilibrium is at the phase boundary between the solid and liquid phases, and there is no back diffusion from the solid to the liquid phases.Scheil solidification simulation explains how nonequilibrium cooling can trap solute in the matrix, creating supersaturation.This information is valuable to this study because when heat treatments or thermal conditioning is applied, these concentrated solute atoms (either in the matrix or in the intermetallics) migrate and alter phase amounts, which is the focus of characterization in this study.

Thermodynamic Simulations
The Scheil (nonequilibrium) solidification diagrams for the A356 composition from Table 1 are    The equilibrium diagram for the A356 composition is shown in Figure 3. Figure 3 shows that ThermoCalc TM predicts the Si phase will grow from ~6.4 to 7.4 vol.% after the solidus temperature.The equilibrium diagram also predicts that Si, Fe, Mg, and Al begin to precipitate out, starting at ~520 °C, to form the ᴨ-Al9FeMg5Si3 phase.This phase then dissolves at 182 °C and is replaced by the precipitation of the Mg2Si phase.These phases were also reported by Stroh et al. [17,18] and Sims [49] and were in good agreement with the equilibrium simulation as opposed to the Scheil after heat treatment.The Scheil and equilibrium ThermoCalc TM simulations also predict the precipitation of Al9Fe2Si2, Al3Ti, and Al15Si2Mn4 at low volume percentages (<0.5%).These phases are rarely reported as significant in the published literature on A356.The equilibrium diagram for the A356 composition is shown in Figure 3. Figure 3 shows that ThermoCalc TM predicts the Si phase will grow from ~6.4 to 7.4 vol.% after the solidus temperature.The equilibrium diagram also predicts that Si, Fe, Mg, and Al begin to precipitate out, starting at ~520 • C, to form the Π-Al 9 FeMg 5 Si 3 phase.This phase then dissolves at 182 • C and is replaced by the precipitation of the Mg 2 Si phase.These phases were also reported by Stroh et al. [17,18] and Sims [49] and were in good agreement with the equilibrium simulation as opposed to the Scheil after heat treatment.The Scheil and equilibrium ThermoCalc TM simulations also predict the precipitation of Al 9 Fe 2 Si 2 , Al 3 Ti, and Al 15 Si 2 Mn 4 at low volume percentages (<0.5%).These phases are rarely reported as significant in the published literature on A356.
Al3Ti, Al9Fe2Si2, and Al15Si2Mn4 phases.At the solidus temperature (shown in the Scheil diagram as 558 °C), excessive amounts of Mg and Si are trapped in a solid solution of the matrix (0.8 and 1.3 wt.%, respectively) due to nonequilibrium cooling.The equilibrium diagram for the A356 composition is shown in Figure 3. Figure 3 shows that ThermoCalc TM predicts the Si phase will grow from ~6.4 to 7.4 vol.% after the solidus temperature.The equilibrium diagram also predicts that Si, Fe, Mg, and Al begin to precipitate out, starting at ~520 °C, to form the ᴨ-Al9FeMg5Si3 phase.This phase then dissolves at 182 °C and is replaced by the precipitation of the Mg2Si phase.These phases were also reported by Stroh et al. [17,18] and Sims [49] and were in good agreement with the equilibrium simulation as opposed to the Scheil after heat treatment.The Scheil and equilibrium ThermoCalc TM simulations also predict the precipitation of Al9Fe2Si2, Al3Ti, and Al15Si2Mn4 at low volume percentages (<0.5%).These phases are rarely reported as significant in the published literature on A356.The thermal treatments of interest in this study are the T6 and conditioning at 250 and 300 • C. Therefore, it is important to highlight what the equilibrium solidification diagrams predict will happen to the microstructure at these temperatures of interest.Figure 3 shows the equilibrium diagram with these temperatures highlighted.As expected, the T6 heat treatment (540 • C solutionizing, quenching, followed by aging at 154 • C) represented in the ThermoCalc TM equilibrium diagram shows that the solutionizing should result in the complete dissolution of the Mg 2 Si, Al 3 Ti, and Π-Al 9 FeMg 5 Si 3 phases and the partial dissolution of the Si phase.After this, the aging should precipitate out the Mg 2 Si phase (which is well known [11,17,18,49]) and the Al 3 Ti phase (Ti is typically used for grain refinement purposes [14,50], and this phase is not typically reported in the microstructure).The Al 9 Fe 2 Si 2 phase is also expected to precipitate out of the matrix according to ThermoCalc TM , which is much more favourable for mechanical properties than the Π Fe-containing phase [1].The transition from the Al 9 Fe 2 Si 2 phase to the Π-Al 9 FeMg 5 Si 3 phase is expected to happen around 182 • C at the expense of some Mg 2 Si.Stroh et al. [33] confirmed the presence of the Π-Al 9 FeMg 5 Si 3 phase at approximately 1.2 vol.% after a T6 heat treatment.This volume percentage was estimated by calculating the area of scanning electron microscopy micrographs using ImageJ and correlated well with the ThermoCalc TM prediction of 1.14 vol.% between 182 and 400 • C.An Al matrix and a Si eutectic phase were also discussed in the study by Stroh et al., but specific values of their volume percentages were not given.
The equilibrium diagrams in Figures 2 and 3 show that these phases (Mg 2 Si and Si) will partially dissolve into the matrix or form other phases when thermal conditioning is applied at 250 or 300 °C.The Mg 2 Si and Si phases are heavily relied upon for their strengthening benefits to the A356 alloy [14,49,51].However, excess Si is likely trapped in the matrix as a solid solution due to the nonequilibrium cooling.When conditioning is applied, Si will migrate out of the solution, resulting in the increases in the Mg 2 Si and Si phases [11,49].
Adding 3.5% RE to the A356 alloy constitutes a more complex microstructure.Figure 4   Following the principle shown in Figure 3, taking a more in-depth look at the equilibrium diagram for phases at temperatures of interest is paramount.Figure 5 gives a closer look at the aging (154 °C), solutionizing (540 °C), and thermal conditioning (250 and 300 °C) temperatures and the phases predicted.ThermoCalc TM indicates the same transi- Comparatively, the equilibrium diagram is shown in Figure 4.The equilibrium simulations predict the Si phase will grow from ~5.5 to 7.0 vol.% after the solidus temperature to 100 • C.These phases were also reported by Stroh et al. [17,18] and Sims [49] and were in good agreement with the equilibrium simulation as opposed to the Scheil after heat treatment.The Stroh et al. [17] reported no presence of the LaSi 2 and Al 11 RE 3 .This was attributed to the large presence of La in the AlSiRE [1].The AlSiRE phase presented in [1] and [14] is characterized as similar to the τ1 phase presented in [5] and [21].These phases could be any combination of the three AlCeSi phases presented by the ThermoCalc TM simulations.Aghaie et al. [31], in a B319 + Ce alloy, found the presence of both AlCeSi 2 and Al 4 Ce 3 Si 6 .In the equilibrium solidification diagram, AlCeSi transitions to AlCeSi 2 at 611 • C. Shortly after this, the AlCeSi 2 phase transitions to Al 4 Ce 3 Si 6 at 598 • C. Based on the EDS compositional analysis in [1], it is likely that the Al 4 Ce 3 Si 6 is the τ1 phase highlighted within these studies.The high solubility of La in this phase, combined with its high thermal stability, suggests that the Al 11 RE 3 phase ThermoCalc TM predicts will also be absent in the microstructure of the as-cast and T6 alloy.It was reported that in the as-cast state, the A356 + 3.5RE alloy consisted of 4.37% AlSiRE, 2.00% Mg 2 Si, 1.85% Π-Al 9 FeMg 5 Si 3 , and 0.70% Al 20 Ti 2 RE (all in vol.%).The Al 20 Ti 2 RE phase is not predicted in any ThermoCalc TM simulation; however, it is well documented in most Al-RE alloys containing Ti as a grain refiner [52][53][54][55].
Following the principle shown in Figure 3, taking a more in-depth look at the equilibrium diagram for phases at temperatures of interest is paramount.Figure 5   Following the principle shown in Figure 3, taking a more in-depth look at the eq librium diagram for phases at temperatures of interest is paramount.Figure 5 give closer look at the aging (154 °C), solutionizing (540 °C), and thermal conditioning (250 a 300 °C) temperatures and the phases predicted.ThermoCalc TM indicates the same tran tions for Mg2Si, ᴨ-Al9FeMg5Si3, Si, and Al3Ti as in regular A356 (Figure 3).It is also simi that the Al15Si2Mn4 phase is present in the equilibrium diagram but not reported in any the literature for this alloy.The main difference in the equilibrium solidification of A356 and A356 + 3.5RE is the Al 4 Ce 3 Si 6 phase (2 vol.%, thermally stable up to 550 • C), the Al 11 RE 3 phase, and the LaSi 2 phase.The LaSi 2 phase is shown to transition to the Al 11 RE 3 phase between 175 and 100 • C. The vol.% of the Si phase also increases in this temperature range.However, as Stroh et al. [17] report, the formation of these phases is unlikely as there is expected to be a significant amount of La in the solid solution of the Al 4 Ce 3 Si 6 phase.
The last alloy of interest is Al-8Ce-10Mg (detailed composition in Table 1).The Scheil solidification diagrams of this composition are shown in Figure 6 The main difference in the equilibrium solidification of A356 and A356 + 3.5 RE is Al4Ce3Si6 phase (2 vol.%, thermally stable up to 550 °C), the Al11RE3 phase, and the L phase.The LaSi2 phase is shown to transition to the Al11RE3 phase between 175 and °C.The vol.% of the Si phase also increases in this temperature range.However, as St et al. [17] report, the formation of these phases is unlikely as there is expected to b significant amount of La in the solid solution of the Al4Ce3Si6 phase.
The last alloy of interest is Al-8Ce-10Mg (detailed composition in Table 1).The Sch solidification diagrams of this composition are shown in Figure 6.The simulation revea the presence of the following phases during solidification: Al, Al11Ce3, β-Al3M Al13CeMg6, Mg2Si, AlCeSi, and Al3Ti.The latter three phases are in minor quantities (< vol.%) while the prior four are shown in significant quantities.Of the significant pha Al begins to precipitate at 587 °C.Shortly thereafter, Al11Ce3 begins to precipitate at °C.At the solidus temperature (446 °C), ThermoCalc TM predicts ~16.7 wt.% Mg is trapp in the solid solution of the matrix.There was no heat treatment used on the Al-8Ce-10Mg alloy in this study.Howev 250 and 300 °C thermal conditioning was applied on separate samples for 200 h.Therefo similar to the other alloys discussed, it is important to closely examine the equilibri ThermoCalc TM diagram at these respective temperatures.Figure 7 shows the equilibri diagram between 200 and 350 °C.The diagram shows that as a result of conditioning, β-AlMg phase is expected to dissolve into the matrix, creating Al with Mg in a solid so tion.However, nonequilibrium cooling results in a significant amount of Mg be trapped in a solid solution within the matrix [49].Therefore, the β-Al3Mg2 phase is pected to increase due to the Mg in solid solution having a propensity to migrate out a stabilize the β-AlMg phase.This β-Al3Mg2 phase is proven to be detrimental to mechan In no other study were the ternary Al 13 CeMg 6 and other minor phases found in the microstructure of Al-8Ce-10Mg.
There was no heat treatment used on the Al-8Ce-10Mg alloy in this study.However, 250 and 300 • C thermal conditioning was applied on separate samples for 200 h.Therefore, similar to the other alloys discussed, it is important to closely examine the equilibrium ThermoCalc TM diagram at these respective temperatures.Figure 7 shows the equilibrium diagram between 200 and 350 • C. The diagram shows that as a result of conditioning, the β-AlMg phase is expected to dissolve into the matrix, creating Al with Mg in a solid solution.However, nonequilibrium cooling results in a significant amount of Mg being trapped in a solid solution within the matrix [49].Therefore, the β-Al 3 Mg 2 phase is expected to increase due to the Mg in solid solution having a propensity to migrate out and stabilize the β-AlMg phase.This β-Al 3 Mg 2 phase is proven to be detrimental to mechanical properties [38,49].These publications do not include specific vol.% of these phases of significance before or after thermal conditioning, making the QPA focus on this study all the more relevant and necessary.

Quantitative Phase Analysis in MAUD
To understand the effects of thermal conditioning for specific operating temperatu on the microstructures of T6 A356, A356 + 3.5 RE, T6 A356 + 3.5RE, and Al-8Ce-20Mg neutron diffraction experiment was carried out using the HIPPO beam at LANSCE a the respective software for Rietveld refinement of diffraction data (MAUD 2.9993).Figu 8 shows the degree of fit targeted in this study.It shows the Rietveld refinement of the (0° rotation) panels of the HIPPO detectors of the T6 A356 alloy.The MAUD setup each alloy was guided by [43,45,58,59] to ensure a thorough and accurate QPA was co ducted since a limited amount of available MAUD studies do not work with multi-pha materials.Phases and their crystalline structures used were CIF files gathered from t crystallography open database [60] and the inorganic crystal structure database, exce the Al11Ce3 phase.Since an appropriate Al11Ce3 CIF file was not available, one was creat using the parameters from [61].
Specifics of the fitting parameters for each MAUD simulation are shown in Table Each simulation had a d-spacing between 0.5 and 3.0 Å.There was an aim to fit sigm values (Rwp/Rexp) close to 1 and the Rwp values being well below 10%, which are the typi indicators for well-fit Rietveld refinements [43,58,[62][63][64].
Table 3 through 5 show the QPA for each alloy and their respective thermal con tioning: Table 3 shows the results of the T6 A356 alloy, Table 4 shows the results of the cast and T6 A356 + 3.5RE alloy, and Table 5 shows the results of the Al-8Ce-10Mg alloy For the T6 A356 alloy, all the phases found in the ThermoCalc TM simulations (Figu 1 and 2

Quantitative Phase Analysis in MAUD
To understand the effects of thermal conditioning for specific operating temperatures on the microstructures of T6 A356, A356 + 3.5RE, T6 A356 + 3.5RE, and Al-8Ce-20Mg, a neutron diffraction experiment was carried out using the HIPPO beam at LANSCE and the respective software for Rietveld refinement of diffraction data (MAUD 2.9993).Figure 8 shows the degree of fit targeted in this study.It shows the Rietveld refinement of the 90 (0 • rotation) panels of the HIPPO detectors of the T6 A356 alloy.The MAUD setup for each alloy was guided by [43,45,58,59] to ensure a thorough and accurate QPA was conducted since a limited amount of available MAUD studies do not work with multiphase materials.Phases and their crystalline structures used were CIF files gathered from the crystallography open database [60] and the inorganic crystal structure database, except the Al 11 Ce 3 phase.Since an appropriate Al 11 Ce 3 CIF file was not available, one was created using the parameters from [61].
Specifics of the fitting parameters for each MAUD simulation are shown in Table 2.Each simulation had a d-spacing between 0.5 and 3.0 Å.There was an aim to fit sigma values (R wp /R exp ) close to 1 and the R wp values being well below 10%, which are the typical indicators for well-fit Rietveld refinements [43,58,[62][63][64].After thermal conditioning at 250 °C, the alloy experienced an increase in the Si phase (7 to 8 vol.%).This increase comes at the expense of a decrease in every other phase (Al, ᴨ-Al9FeSi3Mg5, Mg2Si, and β-Al5FeSi).The phases in the T6 A356 alloy responded similarly  Table 3 through 5 show the QPA for each alloy and their respective thermal conditioning: Table 3 shows the results of the T6 A356 alloy, Table 4 shows the results of the as-cast and T6 A356 + 3.5RE alloy, and Table 5 shows the results of the Al-8Ce-10Mg alloy.For the T6 A356 alloy, all the phases found in the ThermoCalc TM simulations (Figures 1 and 2) were included in the MAUD simulation for the Rietveld refinement.Al 3 Ti, Al 9 Fe 2 Si 2 , and Al 15 Si 2 Mn 4 were quickly excluded during the refinement.Stroh et al. [17] identified the β-Al 5 FeSi phase using SEM and EDS.Therefore, this phase was also included in the Rietveld refinement.Table 3 shows that the β-Al 5 FeSi phase existed in the microstructure of the T6 A356 alloy at low volume percentages (0.5 vol.%).The Al, Si, Π-Al 9 FeSi 3 Mg 5 , and Mg 2 Si phases were all found in volume percentages of 89.1, 7.0, 3.2, and 0.2, respectively.
After thermal conditioning at 250 • C, the alloy experienced an increase in the Si phase (7 to 8 vol.%).This increase comes at the expense of a decrease in every other phase (Al, Π-Al 9 FeSi 3 Mg 5 , Mg 2 Si, and β-Al 5 FeSi).The phases in the T6 A356 alloy responded similarly to both 250 and 300 • C conditioning.Most of the microstructure stabilizes except for Mg 2 Si.Thermally conditioned at 250 • C, the Mg 2 Si phase reduced in volume percentage down 23.37% (from 0.23 to 0.17).However, conditioning at 300 • C slightly reduces Mg 2 Si (0.23 to 0.21).This will be further analyzed later in Section 4 of this paper.
Similar to the discussion on the T6 A356 alloy, the ThermoCalc TM simulation of the composition of the A356 + 3.5RE phases was used for the MAUD Rietveld refinement.The following phases were not found in any significant quantities: AlCeSi, AlCeSi 2 , Al 3 Ti, Mg 2 Si, Al 11 RE 3 , Al 9 Fe 2 Si 2 , Al 15 Si 2 Mn 4 , and LaSi 2 .Also similar to the T6 A356 was the β-Al 5 FeSi phase, identified by Stroh et al. in the A356 + 3.5RE alloy [1,17].In the same studies, there was also confirmation of an Al-RE-Ti phase identified as Al 20 Ti 2 RE.This phase is reported in another study [52,53] as Al 20 Ti 2 Ce.A CIF file for the Al 20 Ti 2 Ce phase was used in the Rietveld analysis of this alloy.Table 4 shows the results of the MAUD QPA of the as-cast and T6 A356 + 3.5RE alloy.The phase vol.% changes (indicated by the percentages within the brackets) are comparisons between the conditioned and the original unconditioned sample.For example, A356 + 3.5RE 250 • C-200 h values are compared to the unconditioned A356 + 3.5RE phase vol.% values, and the difference is reported in brackets under the conditioned phase value.Similarly, the percent change in the T6 A356 + 3.5RE 250 • C-200 h alloy would be the unconditioned T6 A356 reference sample.This point avoids confusion on which percent change each thermally conditioned alloy references.
Table 4 shows that the Si phase of the A356 + 3.5RE alloy in the as-cast and T6 states both exhibit increases in vol.% with applied thermal conditioning.The Al 4 Ce 3 Si 6 phase is relatively stable at these temperatures, except for the presence of this phase in the as-cast A356 + 3.5RE 250 • C-200 h alloy, being lower than average (2.07 vol.%).This discrepancy is likely a result of a variation in composition (discussed in more detail in Section 4).The Al 20 Ti 2 Ce and β-Al 5 FeSi phases were also relatively stable when subjected to thermal conditioning.However, after a T6 heat treatment, the Al 20 Ti 2 Ce phase decreased in volume while the β-Al 5 FeSi phase increased.The Π-Al 9 FeSi 3 Mg 5 phase had a small decrease in volume with increasing thermal conditioning temperatures.This phase also decreased in volume in response to the T6 heat treatment.
Table 5 shows the QPA results from the Al-8Ce-10Mg alloy before and after thermal conditioning.It shows the main trend of an increase in β-Al 3 Mg 2 'pooling' with thermal conditioning at 250 and 300 • C, similar to chapter 5 in Sims [49].This increase was likely a coarsening of this phase at the expense of Mg coming out of the solid solution of the matrix, as shown by a slight decrease in the Al phase.The Al 11 Ce 3 phase also shows a slight decrease in vol.%.However, this phase is quite stable at these thermal conditioning temperatures.Therefore, the difference can be attributed to the slight compositional differences in the material.

T6 A356
The results from Table 3 are shown in a graphical format in Figure 9 to visualize the vol.% changes in each phase.The Π-Al 9 FeSi 3 Mg 5 phase was found in much higher volume percentages than predicted in ThermoCalc TM (~3.0 vs. ~1.0).Stroh et al. [14] were among the few attempts to try and quantify the vol.% in a T6 A356 alloy (same composition and heat treatment used in this study).Using SEM micrographs and ImageJ threshold processing, study [14] quantified the Π-Al 9 FeSi 3 Mg 5 phase to be ~1.2 vol.%, in good correlation with the ThermoCalc TM result.However, as mentioned previously, this processing method can be skewed by aspect ratios of phases and relies heavily on large areas to achieve statistical reliability.Also, determining the proper contrast to distinguish the difference between the Π-Al 9 FeSi 3 Mg 5 phase and the Si phase using ImageJ can be difficult and subjective.This may lead researchers to favour the ThermoCalc TM values for good agreement and verification.Time-of-flight neutron diffraction, however, is a much more reliable technique for the QPA of bulk phase compositions [43,62].Therefore, the elevated vol.% of the Π-Al 9 FeSi 3 Mg 5 phase is likely a more accurate representation of the presence of this phase in the microstructure than the previously used surface techniques.

PEER REVIEW
13 of 20 slight decrease in vol.%.However, this phase is quite stable at these thermal conditioning temperatures.Therefore, the difference can be attributed to the slight compositional differences in the material.

T6 A356
The results from Table 3 are shown in a graphical format in Figure 9 to visualize the vol.% changes in each phase.The ᴨ-Al9FeSi3Mg5 phase was found in much higher volume percentages than predicted in ThermoCalc TM (~3.0 vs. ~1.0).Stroh et al. [14] were among the few attempts to try and quantify the vol.% in a T6 A356 alloy (same composition and heat treatment used in this study).Using SEM micrographs and ImageJ threshold processing, study [14] quantified the ᴨ-Al9FeSi3Mg5 phase to be ~1.2 vol.%, in good correlation with the ThermoCalc TM result.However, as mentioned previously, this processing method can be skewed by aspect ratios of phases and relies heavily on large areas to achieve statistical reliability.Also, determining the proper contrast to distinguish the difference between the ᴨ-Al9FeSi3Mg5 phase and the Si phase using ImageJ can be difficult and subjective.This may lead researchers to favour the ThermoCalc TM values for good agreement and verification.Time-of-flight neutron diffraction, however, is a much more reliable technique for the QPA of bulk phase compositions [43,62].Therefore, the elevated vol.% of the ᴨ-Al9FeSi3Mg5 phase is likely a more accurate representation of the presence of this phase in the microstructure than the previously used surface techniques.The QPA results in Figure 9 also show a distinct relationship between the ᴨ-Al9FeSi3Mg5 and Mg2Si phase.When the ᴨ-Al9FeSi3Mg5 phase of the T6 A356 alloy de- The QPA results in Figure 9 also show a distinct relationship between the Π-Al 9 FeSi 3 Mg 5 and Mg 2 Si phase.When the Π-Al 9 FeSi 3 Mg 5 phase of the T6 A356 alloy decreases in vol.% after 250 • C conditioning, the Mg 2 Si phase does the same.However, this is not a result of thermal conditioning but rather a slight difference in composition between alloys.Wang et al. [65,66] and Taylor et al. [67] identified that the Π-Al 9 FeSi 3 Mg 5 phase is quite thermally stable when the Mg content in an alloy is above 0.5 wt.% (the Mg content in Table 1 for A356 is 0.49 wt.%).However, after thermal conditioning at 250 • C for 200 h, the Π-Al 9 FeSi 3 Mg 5 phase in the T6 A356 alloy decreases from 3.17 to 2.96 vol.%.This is also accompanied by a decrease in Mg 2 Si from 0.23 to 0.17.This decrease in Mg 2 Si can be explained by slight compositional differences between alloys.Taylor et al. [67] show that a slight decrease in Mg content below 0.5 wt.% of the alloys would decrease the vol.%Mg 2 Si and Π-Al 9 FeSi 3 Mg 5 , similar to what is shown in Figure 9.
The two phases affected by thermal conditioning are the Si and β-Al 5 FeSi.The β-Al 5 FeSi phase decreases from 0.51 vol.% to 0.22 vol.% when conditioning at either temperature.This partial dissolution is consistent with the findings of Wang et al. [65,66].Opposite the β-Al 5 FeSi phase partial dissolution is the growing response of the Si phase to thermal conditioning.The refinement of the Si phase after a T6 heat treatment coupled with the solid solution of Si in the Al matrix is well known as a significant contributor to the strength of the A356 alloy.However, during thermal conditioning at 250 and 300 • C, a significant amount of the Si in solid solution is removed, resulting in an extensive increase in this phase (~13% increase).This growth is a significant contributor to the reduction in tensile strength that Sims [49] and Stroh et al. reported [17].

As-Cast and T6 A356 + 3.5RE
The QPA results for the (as-cast and T6) A356 + 3.5RE alloy are shown in Figure 10.These results are in agreement with Stroh et al.'s paper on this alloy [17] concerning the Mg 2 Si phase.The Rietveld fitting of the diffraction pattern revealed no detectable Mg 2 Si within the alloy.This suggests that the tensile strength of this alloy is independent of Mg 2 Si, unlike its compositionally simpler counterpart, A356.The phases revealed via Rietveld analysis were Al, Si, Al 4 Ce 3 Si 6 Al 20 Ti 2 Ce, Π-Al 9 FeSi 3 Mg 5 , and β-Al 5 FeSi.After conditioning the alloy at 250 • C for 200 h, the Al 4 Ce 3 Si 6 phases decreased in vol.% by 34.7.This decrease in vol.% increases the Si phase and slightly increases the Al 20 Ti 2 Ce phase.The Si from the Al 4 Ce 3 Si 6 and excess trapped in the matrix contribute to the growth of the Si phase.The growth of the Al 20 Ti 2 Ce phase is a result of sharing common elements (Ce and Nd [17]) with the Al 4 Ce 3 Si 6 phase.During solidification, the Al 20 Ti 2 Ce phase is known to precipitate early as a primary phase [68], similar to Al 4 Ce 3 Si 6 .As a result, they compete for similar elements and nonequilibrium cooling results in both these phases forming in a metastable condition.Thermal conditioning at 250 • C causes the excess Ti from the matrix and Ce and Nd from Al 4 Ce 3 Si 6 to stabilize the Al 20 Ti 2 Ce phase, resulting in the growth shown A356 + 3.5RE in Figure 10.The Π and β Fe-containing phases remain thermally stable at 250 • C.
At 300 • C conditioning, the Π-Al 9 FeSi 3 Mg 5 phase partially dissolves, and the Al 4 Ce 3 Si 6 and β-Al 5 FeSi return to similar vol.% as the as-cast values.The Al 20 Ti 2 Ce and Si phases remain at a similar vol.% as the 250 • C conditioned samples.Increases in the vol.% of the Al 4 Ce 3 Si 6 and β-Al 5 FeSi phases and decreases in the Π-Al 9 FeSi 3 Mg 5 phase are typically beneficial for enhancing the mechanical properties of alloys [1,31,69].However, the increase in the Si and Al 20 Ti 2 Ce phases is shown to be detrimental to mechanical properties [17,53].At 250 • C in the as-cast state, the A356 + 3.5RE alloy performs similarly to the T6 A356 alloy, showing no mechanical benefit [18] with this microstructural evolution.However, at 300 • C, the as-cast A356 + 3.5RE alloy significantly outperforms the T6 A356 alloy [18], showing that the enhancing benefits of the increase in the Al 4 Ce 3 Si 6 phase and the decrease in Π-Al 9 FeSi 3 Mg 5 outweigh the detrimental effect of Si and Al 20 Ti 2 Ce phase increases.
to precipitate early as a primary phase [68], similar to Al4Ce3Si6.As a result, they compete for similar elements and nonequilibrium cooling results in both these phases forming in a metastable condition.Thermal conditioning at 250 °C causes the excess Ti from the matrix and Ce and Nd from Al4Ce3Si6 to stabilize the Al20Ti2Ce phase, resulting in the growth shown A356 + 3.5RE in Figure 10.The ᴨ and β Fe-containing phases remain thermally stable at 250 °C.At 300 °C conditioning, the ᴨ-Al9FeSi3Mg5 phase partially dissolves, and the Al4Ce3Si6 and β-Al5FeSi return to similar vol.% as the as-cast values.The Al20Ti2Ce and Si phases remain at a similar vol.% as the 250 °C conditioned samples.Increases in the vol.% of the Al4Ce3Si6 and β-Al5FeSi phases and decreases in the ᴨ-Al9FeSi3Mg5 phase are typically beneficial for enhancing the mechanical properties of alloys [1,31,69].However, the increase After the T6 heat treatment of the A356 + 3.5RE alloy, the vol.% of the β-Al 5 FeSi increased and the Π-Al 9 FeSi 3 Mg 5 phase decreased.This transition stabilized the β-Al 5 FeSi phase at ~0.7 vol.%, which was then unaffected by thermal conditioning.The Π-Al 9 FeSi 3 Mg 5 phase was thermally unstable after the T6 heat treatment and linearly decreased with increasing thermal conditioning (2.25 vol.% to 1.99 vol.% at 250 • C then 1.86 vol.% at 300 • C).This lack of thermal stability of the Π-Al 9 FeSi 3 Mg 5 can be linked to Tayor et al.'s [67] findings regarding the limited mobility of Mg.Stroh et al. [1,17] found Mg was present in the Al 20 Ti 2 RE, AlSiRE, and β-Al 5 FeSi phases, which are thermally stable and unaffected by thermal conditioning.Therefore, the Mg within these phases cannot migrate and stabilize the Π-Al 9 FeSi 3 Mg 5 phase [67]; hence, its vol.% decreases when thermal conditioning is applied to the alloy.
The Al 20 Ti 2 Ce phase also decreases in volume after the T6 heat treatment by 13% (from 3.3 to 2.9 vol.%).There is no knowledge of the stability of the Al 20 Ti 2 Ce phase above 400 • C.However, in this instance, solution annealing at 538 • C for 8 h results in the partial dissolution of this phase with the benefit of increasing the more favourable Al 4 Ce 3 Si 6 phase.After a T6 heat treatment, the Al 20 Ti 2 Ce phase is relatively unaffected by thermal conditioning (i.e., thermally stabilized) unlike the as-cast state of the alloy, which showed that thermal conditioning caused slight growth of this phase.This phase's reduction and stabilization, the decreased vol.% of the Π-Al 9 FeSi 3 Mg 5 phase, and the increased vol.% of the β-Al 5 FeSi phase are large contributors to explain why the mechanical properties Stroh et al. [17,18] reported at 250 and 300 • C are significantly better than the as-cast state of the same alloy.
The one constant between the as-cast and T6 A356 + 3.5RE alloys is the growth of the Si phase when the alloys are subjected to thermal conditioning.The Si phase of the T6 A356 alloy also experienced a similar vol.% increase when subjected to thermal conditioning.However, the A356 + 3.5RE and T6 A356 + 3.5RE both had better tensile properties than the T6 A356 alloy [17], despite this similar Si growth.The refinement of this Si phase from an irregular blocky intermetallic to spheroidized particles after a T6 heat treatment is well known to enhance the room temperature strength of A356-based alloys [14,18].However, the growth of this phase above 200 • C limits its potential for automotive powertrain applications.The A356 + 3.5RE suggests that both scenarios combine the best approaches, utilizing the refined Si phase for room-temperature strength and the RE intermetallics for elevated-temperature benefits.These RE intermetallics, in addition to the Fe-and Si-containing phases, respond well to a conventional T6 heat treatment, which ultimately leads to enhanced tensile strength [1,17], even after the microstructure stabilizes thermal conditioning.The results from this QPA study provide a complete analysis of how the fundamental microstructural evolution results in these elevated tensile properties.

As-Cast Al-8Ce-10Mg
The results of the Al-8Ce-10Mg alloy MAUD QPA are shown in Figure 11.In the as-cast state, there is an excessive amount of Mg in the matrix as a solid solution, identifiable by the high vol.% of Al and low amounts of β-Al 3 Mg 2 .The Scheil solidification in ThermoCalc TM (Figure 6) predicted the excess Mg in the solid solution to be ~16 wt.%.When 250 • C thermal conditioning is applied, the β-Al 3 Mg 2 phase grows by 82% from 8.0 to 14.6 vol.%.The growth of the β-Al 3 Mg 2 phase is due to the high amount of Mg in the alloy.Golovin et al. [70] show that alloys with an excess of 8 wt.% Mg have an absence of internal friction that inhibits the migration of Mg, which is the case in this alloy.This explains why thermal conditioning results in the extensive growth of the β-Al 3 Mg 2 phase, shown in Figure 11.Sims [49] correlates the increase in the β-Al 3 Mg 2 phase to a decrease in mechanical properties.In the same study, Sims observes the β-Al 3 Mg 2 to dissolve back into the matrix around 400 • C after extensive thermal conditioning (>500 h).The 300 • C-200 h thermal conditioning targeted in this study shows that the β-AlMg phase continues to grow at 300 • C up to 16.1 vol.% (doubled from the as-cast state).

PEER REVIEW
16 of 20 in Figure 11.Sims [49] correlates the increase in the β-Al3Mg2 phase to a decrease in mechanical properties.In the same study, Sims observes the β-Al3Mg2 to dissolve back into the matrix around 400 °C after extensive thermal conditioning (>500 h).The 300 °C-200 h thermal conditioning targeted in this study shows that the β-AlMg phase continues to grow at 300 °C up to 16.1 vol.% (doubled from the as-cast state).The other significant phase identified in the Rietveld analysis is the Al11Ce3 phase.This phase appears relatively stable in both this study and the one by Sims [49].The small volume percentage discrepancy of the Al11Ce3 phase can be attributed to slight compositional variances.The other phases predicted by ThermoCalc TM (Al13CeMg6, Mg2Si, Al13Fe4, Al6Mn, and T-phase) were not identified in the MAUD Rietveld analysis.The microstructural evolution of this alloy suggests that the 250 to 300 °C range is not an ideal operating temperature for this alloy.It is not until 400 °C that the microstructure stabilization becomes beneficial for mechanical properties [49].

Conclusions
This study aimed to conduct an in-depth quantitative phase analysis via the time-offlight neutron diffraction of a conventional T6 A356 alloy and advanced A356 + 3.5 RE and The other significant phase identified in the Rietveld analysis is the Al 11 Ce 3 phase.This phase appears relatively stable in both this study and the one by Sims [49].The small volume percentage discrepancy of the Al 11 Ce 3 phase can be attributed to slight compositional variances.The other phases predicted by ThermoCalc TM (Al 13 CeMg 6 , Mg 2 Si, Al 13 Fe 4 , Al 6 Mn, and T-phase) were not identified in the MAUD Rietveld analysis.The microstructural evolution of this alloy suggests that the 250 to 300 • C range is not an ideal operating temperature for this alloy.It is not until 400 • C that the microstructure stabilization becomes beneficial for mechanical properties [49].

Conclusions
This study aimed to conduct an in-depth quantitative phase analysis via the time-offlight neutron diffraction of a conventional T6 A356 alloy and advanced A356 + 3.5RE and Al-8Ce-10Mg alloys.These alloys were subjected to 250 and 300 • C thermal conditioning to understand how the phases of all three alloys stabilized at these temperatures.
• The T6 A356 alloy contained the following phases, confirmed via Rietveld analysis: cu- bic Al, Si, Mg 2 Si, Π-Al 9 FeSi 3 Mg 5 , β-Al 5 FeSi.Conditioning at 250 and 300 • C decreases the β-Al 5 FeSi phase from 0.51 to 0.22 vol.%.The Si phase grows significantly by ~13% when subjected to either condition temperature.The volume of other phases remained relatively stable when subjected to thermal conditioning.The growth of the Si phase significantly weakens this alloy and makes it appealing for this temperature region.C, respectively.The volume of the Si phase within the alloy increases similarly to the T6 A356 and the as-cast A356 + 3.5RE alloys.All these refinements in the phases explain why the T6 A356 + 3.5RE alloy significantly outperforms the T6 A356 alloy in the desirable temperature region in other studies.• The phases of the Al-8Ce-10Mg alloy in the as-cast state consist of Al (with Mg in solid solution), β-Al 3 Mg 2 , and Al 11 Ce 3 .The Al 11 Ce 3 phase changes negligibly in response to thermal conditioning at 250 and 300 • C. The β-AlMg phase grows by 82% at 250 • C, and its volume doubles at 300 • C.This growth results from the alloy having little internal resistance to the migration of Mg from the solid solution with the matrix, evidenced by the decrease in the Al phase in response to the thermal conditions.The benefits from the stable Al 11 Ce 3 are negatively affected by the increase in β-Al 3 Mg 2 at 250 and 300 • C, making utilizing the alloy at this temperature region unfavourable.The full benefits of the stable Al 11 Ce 3 phase will only be obtained at the temperature point where it is more favourable for the β-Al 3 Mg 2 to dissolve back into the matrix.
As Al alloys become a more prominent solution to problems associated with elevatedtemperature applications, it is paramount that they be characterized in a manner that suits the application conditions.Not thermally conditioning alloys before testing them at elevated temperatures results in misrepresenting the alloy's microstructure at these temperatures.An unrepresentative microstructure leads to inaccurate tensile data, which can lead to much more significant engineering issues.This study indicates that thermal conditioning is important to establish quality microstructures, especially when developing novel Al alloys for high-temperature applications.

Figure 1 .
Figure 1.Pole figures of the texture in the FCC-Al matrix of each alloy were assessed in this study before thermal conditioning.

Figure 1 .
Figure 1.Pole figures of the texture in the FCC-Al matrix of each alloy were assessed in this study before thermal conditioning.
shown in Figure 2. The Scheil diagram shows that the phases of significant volume percentages are Al and Si.In minor quantities, these are the Mg 2 Si, Π-Al 9 FeMg 5 Si 3 , Al 3 Ti, Al 9 Fe 2 Si 2 , and Al 15 Si 2 Mn 4 phases.At the solidus temperature (shown in the Scheil diagram as 558 • C), excessive amounts of Mg and Si are trapped in a solid solution of the matrix (0.8 and 1.3 wt.%, respectively) due to nonequilibrium cooling.
Al3Ti, Al9Fe2Si2, and Al15Si2Mn4 phases.At the solidus temperature (shown in the Scheil diagram as 558 °C), excessive amounts of Mg and Si are trapped in a solid solution of the matrix (0.8 and 1.3 wt.%, respectively) due to nonequilibrium cooling.

Figure 3 .
Figure 3. Equilibrium solidification of the A356 alloy composition highlights the thermal conditioning specific to this study.
shows both the equilibrium and Scheil solidification diagrams.The addition of RE mischmetal results in five additional phases compared to A356: AlCeSi, AlCeSi 2 , Al 4 Ce 3 Si 6 , LaSi 2 , and Al 11 RE 3 .The solidus temperature (shown in the Scheil diagrams) is 557 • C, similar to the T6 A356 shown in Figure 2. Similar amounts of Mg and Si (0.8 and 1.3 wt.%, respectively) are trapped in a solid matrix solution compared to A356.Materials 2024, 17, x FOR PEER REVIEW 8 of 20 simulation; however, it is well documented in most Al-RE alloys containing Ti as a grain refiner [52-55].
gives a closer look at the aging (154 • C), solutionizing (540 • C), and thermal conditioning (250 and 300 • C) temperatures and the phases predicted.ThermoCalc TM indicates the same transitions for Mg 2 Si, Π-Al 9 FeMg 5 Si 3 , Si, and Al 3 Ti as in regular A356 (Figure3).It is also similar that the Al 15 Si 2 Mn 4 phase is present in the equilibrium diagram but not reported in any of the literature for this alloy.

Figure 5 .
Figure 5. Equilibrium solidification of the A356+ 3.5RE alloy composition highlights the therm conditioning specific to this study.

Figure 5 .
Figure 5. Equilibrium solidification of the A356+ 3.5RE alloy composition highlights the thermal conditioning specific to this study.
. The simulation revealed the presence of the following phases during solidification: Al, Al 11 Ce 3 , β-Al 3 Mg 2 , Al 13 CeMg 6 , Mg 2 Si, AlCeSi, and Al 3 Ti.The latter three phases are in minor quantities (<0.6 vol.%) while the prior four are shown in significant quantities.Of the significant phases, Al begins to precipitate at 587 • C. Shortly thereafter, Al 11 Ce 3 begins to precipitate at 578 • C. At the solidus temperature (446 • C), ThermoCalc TM predicts ~16.7 wt.% Mg is trapped in the solid solution of the matrix.terials 2024, 17, x FOR PEER REVIEW 9 o

Figure 6 .
Figure 6.Scheil solidification of the Al-8Ce-10Mg alloy composition using ThermoCalc TM .The alternative equilibrium diagram, showing what potentially would happen after the solidus, is shown in Figure 7. ThermoCalc TM predicts that a ternary Al-Ce-Mg phase (at 332 • C) separates into β-Al 3 Mg 2 and Al 11 Ce 3 .In all the literature on the Al-8Ce-10Mg alloy, SEM and EDS confirm the presence of Al, Al 11 Ce 3 , and β-Al 3 Mg 2 [18,40,49,52,56,57].In no other study were the ternary Al 13 CeMg 6 and other minor phases found in the microstructure of Al-8Ce-10Mg.There was no heat treatment used on the Al-8Ce-10Mg alloy in this study.However, 250 and 300 • C thermal conditioning was applied on separate samples for 200 h.Therefore, similar to the other alloys discussed, it is important to closely examine the equilibrium ThermoCalc TM diagram at these respective temperatures.Figure7shows the equilibrium diagram between 200 and 350 • C. The diagram shows that as a result of conditioning, the β-AlMg phase is expected to dissolve into the matrix, creating Al with Mg in a solid solution.However, nonequilibrium cooling results in a significant amount of Mg being trapped in a solid solution within the matrix[49].Therefore, the β-Al 3 Mg 2 phase is expected to increase due to the Mg in solid solution having a propensity to migrate out and stabilize the β-AlMg
) were included in the MAUD simulation for the Rietveld refinement.Al3 Al9Fe2Si2, and Al15Si2Mn4 were quickly excluded during the refinement.Stroh et al. [

Figure 7 .
Figure 7. ThermoCalc TM of equilibrium solidification of the Al-8Ce-10Mg alloy composition highlighting the thermal conditioning at 250 and 300 • C.

Table 2 .
MAUD parameters and fitting evaluation of the T6 A356 alloy.

Table 2 .
MAUD parameters and fitting evaluation of the T6 A356 alloy.

Table 5 .
QPA results for the as-cast and thermally conditioned Al-8Ce-10Mg • The A356 + 3.5RE alloy responded to 250 • C conditioning with an increase in the Si, Π-Al 9 FeSi 3 Mg 5 , and Al 20 Ti 2 Ce phases, causing a reduction in Al 4 Ce 3 Si 6 .When subjected to 300 • C thermal conditioning, the Al 4 Ce 3 Si 6 and β-Al 5 FeSi stabilize, the Si and Al 20 Ti 2 Ce phases increase similarly to the 250 • C condition, and the Π-Al 9 FeSi 3 Mg 5 partially dissolves.The benefits of increasing the Al 4 Ce 3 Si 6 and decreasing the Π-Al 9 FeSi 3 Mg 5 are beneficial for mechanical properties.However, the growth effects of Si and Al 20 Ti 2 Ce are typically detrimental to mechanical properties.• When subjecting the A356 + 3.5RE alloy to the same T6 heat treatment as the A356 alloy, the Al 4 Ce 3 Si 6 phase stabilizes, the Al 20 Ti 2 Ce decreases by ~13% from 3.3 to 2.9 vol.%, and the volume of β-Al 5 FeSi doubles from 0.35 to 0.7 vol.%.These phases become thermally stable and have negligible responses to thermal conditioning at 250 and 300 • C. The Π-Al 9 FeSi 3 Mg 5 phase decreases by 23% from 2.91 to 2.25 vol.%.This phase is thermally unstable due to the lack of Mg mobility and decreases by ~12% and 18% in response to thermal conditions at 250 and 300