ARL 24-15.pdf

ARL 24-15P

TITLE OF INVENTION: Flux Growth and Nucleation Casting Technique

EXECUTIVE SUMMARY OVERVIEW:

(The purpose of this overview is to provide a brief summarized overview of the invention as described in this invention disclosure document and a direct explanation on what is new about the invention and information about potential military and commercial uses)

1. What is this Invention?

This invention, Flux Growth Nucleation Casting, is a novel casting technique designed to enhance the mechanical properties of metal alloys through the controlled nucleation and growth of high-temperature secondary phases that form as the first to solidify out of the melt, e.g. primaries. This method significantly influences and imparts control on the volume fraction and morphology of these strengthening phases to target improvements in alloy strength, wear resistance, elastic modulus, and potentially other physical properties.

2. What is the basic premise of the design?

The Flux Growth and Nucleation Casting technique operates by holding the prior fully molten alloy isothermally at a temperature below that required for the formation of high melting point secondary phases, e.g. primaries followed by returning the melt to a higher temperature to facilitate suitable fluidity and good castability. This technique can be applied to alloys systems with two characteristics, 1) a system with secondary phase with a melt temperature 110-310% of the matrix melt temperature, or 2) a system with a secondary phase that has a very high enthalpy of formation when compared to the matrix phase. The initial melt temperature, the temperature of the isothermal hold, and the duration of the isothermal hold can all be varied to achieve the desired final microstructure.

3. What is new about this Invention?

The Flux Growth and Nucleation Casting technique is new in that it relies only on the temperature profile and history of the molten alloy in the casting process to generate and control the volume fraction and morphology of primaries in the alloy, unlike traditional casting or rheocasting which requires both the cooling of the alloy to a temperature where liquid and solid phases are stable and mechanical stirring to impart control on the volume fraction and morphology of the strengthening phases.

4. Has this Invention been verified?

The Flux Growth and Nucleation Casting technique has been verified in multiple physical castings, casting forms, isothermal holds, and alloy compositions.

5. Identify Potential Military Uses of the Invention:

Military uses of the invention can include any structural components that can be produced through casting and require the unique physical properties imparted through the control of the strengthening phases volume fraction and morphology. Due to the high temperature stability of the strengthening phases and enhancement of physical properties of the alloys utilized in this casting technique there is potential use for propulsion and high power density engine components that experience high thermal and mechanical stresses and are suitable for manufacture via casting. Another potential use of the invention is for low density alloy plates that would utilize the unique size and distribution of strengthening phases through the casting technique to impart oriented or enhanced ballistic and high strain rate stress and shock resilience.

6. Identify Potential Commercial Uses of the Invention:

Similar to military applications, due to the high temperature stability of the strengthening phases and enhancement of physical properties of the alloys utilized in this casting technique there is potential use for propulsion and high-power density engine components that experience high thermal and mechanical stresses and are suitable for manufacture via casting. Other commercial applications of the invention include any structural components that require low density and high specific stiffness, strength, or wear resistance and that are suitable for manufacture via casting.

.Conversation held discussing the possibility of controlling the morphology of the reinforcement phase in an experimental Al-Ce-Cu-Ni-Mn-Mg multicomponent alloy (“Alloy One”) reinforced by second phase particles known as the Rhombicuboctahedral (“RCO”) phase. Differential Scanning Calorimetry (DSC) results were considered which showed the RCO phase in Alloy One melting at a considerably higher temperature than that of aluminum, allowing for the possibility of controlling nucleation of the reinforcement phase via isothermal holding of the alloy in the melt. A plan was developed to hold the alloy in the melt for 30 minutes at several temperatures lower than the casting temperature of the alloy.

Initial Vacuum Induction Melting (VIM) casting trials of Alloy One utilizing the isothermal hold were conducted. Charges of master alloy were heated to 1000 oC to dissolve any second phase reinforcement particles present before decreasing temperature and holding the alloy in the melt at either 650 oC, 700 oC, or 750 oC for 30 minutes. After the isothermal hold, the temperature was increased to 800 oC and the alloy was cast.

Initial SEM examination of the microstructure of the alloys held at 650, 700, and 750 oC revealed differing size, volume fraction and distribution of reinforcement particles with changing isothermal hold temperature. It was found that with increasing hold temperature, the number of reinforcement particles present decreased and the size of the particles/particle clusters increased. The alloy held at 650 oC had the smallest, best distributed reinforcement particles. These results demonstrated that isothermal holding can be used to control intermetallic size and distribution.

The 650 oC and 750 oC casting trials were repeated with a second RCO forming composition (“Alloy Two”). Again, charges of master alloy were heated to 1000 oC and held for five minutes, then the temperature of the melt was decreased, and the alloy was held at either 650 or 750 oC for 30 minutes, after which the temperature was increased to 800 oC and the alloy was cast.

SEM characterization of Alloy Two, again showing a difference in the size and amount of the RCO phase present in the alloy, with more RCO present in the alloy held at 650 oC as compared to the alloy held at 750 oC.

The 650 oC casting trial of Alloy Three with Scanning Electron Microscopy (SEM) imaging that shows a consistent nucleation rate and particle size along the length of the cast rod.

The casting procedure for Alloy One described above was disclosed; initial SEM micrographs showing the change in intermetallic size and distribution with changing hold temperature were shared.

The casting procedure and SEM and initial microstructural characterization for Alloy Two were disclosed, including micrographs and quantitative image analysis for area fraction of RCO phase present and average size of RCO particles.

Casting procedures and SEM and Energy-Dispersive X-ray Spectroscopy (EDS) characterization for a third RCO forming alloy (“Alloy Three") and paths to improving these alloys were discussed.

Past: The past use for this system has been to modify the morphology, volume fraction, and particle size of a primary strengthening phase in an Al-Ce-Cu-Ni- Mn-Mg-Zn alloy for design and optimization of the alloys composition and microstructure.

Present: The invention is currently being used to hold particles morphology constant while exploring modifications to an alloy’s matrix independent of the composition of the nucleated strengthening particle. Additional exploration into other alloy systems with high temperature intermetallics is currently underway.

Future: This technology can be applied more directly to the Al alloy compositions as shown in the invention disclosure to modify the size, volume fraction, and morphology of the predominant strengthening phase. Looking forward, the same technique can be applied to other alloy systems. The control of primary forming intermetallics in any metal system can have a massive impact on material properties, especially those for high temperature and wear resistance applications.

Abstract

The Flux Growth Nucleation Casting technique is a novel casting technique designed to enhance the mechanical properties of metal alloys by controlling the nucleation and growth of second phase particles, specifically high-temperature phases that form as the first phase to solidify out of the melt, i.e. primaries. This method significantly influences the volume fraction and morphology of these strengthening phases, targeting improvements in strength, wear resistance, and elastic modulus. The technique operates by holding the alloy isothermally in the melt at a temperature below that required for the formation of high melting point secondary phases. This encourages the nucleation and growth of these phases in the liquid state, followed by a return to a higher temperature to facilitate fluidity and good castability. Based on thermodynamic relationships the technique can be applied to alloy systems with two characteristics, 1) a system with a secondary phase with a melt temperature 110-310% of the matrix melt temperature, or 2) a system with a secondary phase that has a very high enthalpy of formation when compared to the matrix phase. The initial melt temperature (to dissolve all secondary phases), the temperature of the isothermal hold, and the duration of the isothermal hold can all be varied to achieve the desired final microstructure. These temperatures can be determined using differential scanning calorimetry to target phase nucleation temperatures.

Invention Summary The disclosed invention, Flux Growth Nucleation Casting (FGNC), is a novel casting technique designed to enhance the mechanical properties of metal alloys by controlling the nucleation and growth of second phase particles, specifically high-temperature phases that form as the first phase to solidify out of the melt, i.e. primaries. This method significantly influences the volume fraction and morphology of these strengthening phases, targeting improvements in strength, wear resistance, and elastic modulus. Unlike traditional metal casting or Rheocasting which is a technique for producing a partially solid alloy by cooling of the alloy into the liquid + solid range while mechanically stirring to ensure that the initial solid particles are broken up into near-spherical forms before casting, FGNC operates by holding the alloy isothermally in the melt at a temperature below that required for the formation of high melting point secondary phases. This encourages the nucleation and growth of these phases in the liquid state, followed by a return to a higher temperature to facilitate fluidity and good castability.

FGNC, which draws parallels with the flux growth method used in single crystal growth, is demonstrated in the alloy system Al-Ce-Cu-Ni-Mn-Mg + Zn, which forms a secondary phase (Al25.3Ce3Cu3.6Ni3.1Mn) known as the Rhombicuboctahedral (“RCO”) phase that has a melting point of ~973 °C, which is 313 °C higher than pure Al. By adjusting the isothermal hold temperature to 650, 700, and 750 °C, variations in particle number, size, and morphology are observed. Additionally, a simplified composition that showed a dependence on cooling rate for RCO particle formation in a wedge mold was homogenized by using the FGNC casting technique. Based on thermodynamic relationships the FGNC technique can be applied to alloy systems with two characteristics, 1) a system with a secondary phase with a melt temperature 110-310% of the matrix melt temperature, or 2) a system with a secondary phase that has a very high enthalpy of formation when compared to the matrix phase. The initial melt temperature (to dissolve all secondary phases), the temperature of the isothermal hold, and the duration of the isothermal hold can all be varied to achieve the desired final microstructure, and while examples primarily focus on aluminum alloys, FGNC is applicable to other metallic alloys meeting these criteria.

Example 1 The FGNC technique was first used for an Al-base multicomponent eutectic-like alloy with the composition Al-7.49Ce-8.34Cu-3.4Ni-1.27Mn-1.76Mg-4.33Zn (all compositions in this document in weight percent unless otherwise specified). This alloy, henceforth known as “Alloy One”, is strengthened by a phase known as the Rhombicuboctahedron (RCO) phase which may form in several morphologies, the most desirable of which are polyhedral particles or large clusters of polyhedral-like particles.

DSC results for Alloy One are shown in Figure 1. As can be seen in the DSC results, the melting temperature of the RCO phase is approximately 973 oC, considerably higher than that of the aluminum matrix which is less than 660 °C for pure aluminum due to excess solid solution elements. The high melting temperature allows for the RCO phase to nucleate and grow within the liquid aluminum matrix. It was desired to determine if the morphology of the RCO phase could be controlled by altering the parameters of the casting process.

Informed by the DSC results, a three-step casting process was devised in which charges of master alloy were first heated to 1000 °C and held for five minutes to dissolve any RCO phase already present, then cooled and held isothermally at 650, 700 or 750 °C for 30 minutes as illustrated in Figure 2.

Figure 1: DSC data for RCO single crystals up to (a) 1050 °C and (b) 1100 °C at 10 °C/min showing the onset of melting at an average temperature of 973 °C.

Figure 2: Flowchart showing the isothermal hold steps for casting the RCO containing alloys to control nucleation and growth of the primary forming phase.

At the end of the 30-minute hold, the melt temperature was increased to 800 °C to ensure adequate fluidity and the alloy was cast into a rod mold. The dimensions of the mold used for these experiments are shown in Figure 3. For each rod, samples were sectioned from the top, bottom, and middle of the casting as shown in Figure 3, allowing for comparison of the microstructure in different regions of the casting.

Figure 4 shows Backscattered Electron (BSE) SEM images of Alloy One samples prepared with isothermal holds at 650, 700, and 750 °C, respectively. BSE imaging provides contrast based on the atomic number of elements present in an alloy, with phases containing higher atomic number elements appearing brighter. As can be seen in Figure 4, Alloy One is composed of large primary-like RCO particles (bright contrast) with polyhedral, polyhedral-cluster or dendritic-like morphology and smaller eutectic-like intermetallics dispersed within an aluminum matrix. Visual examination of Figure 4 suggests a trend of increasing RCO size and decreasing amount of RCO present with increasing isothermal hold temperature; these trends are confirmed by quantitative image analysis.

It was also observed that the microstructure of the alloy was identical at the bottom of

Figure 3: Image of casting mold used for initial Flux Growth Nucleation Casting (FGNC) experiments on Alloy 1 and schematic image showing where “top”, “middle” and “bottom” samples were sectioned from castings.

Figure 4: SEM backscattered electron (BSE) micrographs showing the microstructure of an Al-Ce- Cu-Ni-Mn-Mg-Zn alloy (“Alloy One”) created by isothermally holding in the melt at a) 650, b) 700 or c) 750 oC for 30 minutes prior to casting at 800 oC. As can be seen in the micrographs, the size of the Rhombicuboctahedral (RCO) reinforcement particles (brighter phase) increases with increasing isothermal hold temperature, and the area percentage covered by such particles decreases.

the casting, in the middle, and at the top of the casting, despite the differing cooling rates in these areas, suggesting that the isothermal treatment rendered the microstructure of this alloy independent of cooling rate. Quantitative image analysis supports this conclusion, as the average particle size and area fraction for the top, bottom, and middle of the castings are within the error for all three samples. This is an important result as cooling-rate independent microstructures are highly desired for commercial casting applications.

A sample of alloy one was also prepared without an isothermal hold (“no-hold”) and was used as a control to assess the effectiveness of isothermal holding. The control sample was prepared from the same master alloy as the samples described above. The casting procedure consisted of first heating to 1000 °C and holding for 5 minutes to dissolve any pre-existing RCO, then immediately cooling to 800 °C and casting into the same rod mold used for the isothermal holding trials. Figure 5 shows BSE SEM images of the top, middle, and bottom of the no-hold alloy.

The microstructure of this alloy is not as homogeneous as that of the samples subjected to a 30-minute isothermal hold. In particular, the morphology of the primary-like RCO particles varies with height in the cast rod – the RCO appears smaller and more polyhedral at the top of the sample and becomes larger and more dendritic in form moving towards the bottom of the sample. Quantitative image analysis in Figure 6 shows the average particle size increases from 297.93 um3 to 320.59 um3 moving from the top of the sample to the bottom, an increase of approximately 7.5%. The area fraction occupied by particles is more homogeneous, being 13.37% and the top of the casting and 11.44% at the bottom of the casting. The FGNC technique can also be used with a longer isothermal hold. Based on kinetic principles it may be predicted that a longer hold will result in more coarsening of the primary RCO giving a larger average particle size irrespective of hold temperature. Thus, the time of the isothermal hold provides another potential variable for microstructure control.

Figure 6 provides a summary of the average particle size and the area fraction occupied by particles for the top, bottom, and middle sections of the castings held at 650,700, and 750 °C. The average particle size of RCO particles in the sample held at 650 °C (taking the middle section as representative) was 301.42 µm2 compared to 402.83 µm2 for the sample held at 750 °C, hence the size of RCO particles increased by roughly 30% with increasing the isothermal hold temperature from 650 to 750 °C. The average area fraction occupied by particles also decreased from 16.79% for 650 °C to 8.61% for 750 °C.

Figure 5: BSE-SEM micrographs showing the microstructure of Alloy One created without an isothermal hold, showing samples taken from a) the bottom b) the middle and

c) the top of the casting.

Example 2 A second Al-Ce-Cu-Ni-Mn-Mg alloy (“Alloy Two”) which was leaner in all alloying elements than Alloy One was cast using the same FGNC technique; performing the isothermal holds at 650 °C and 750 °C. The microstructure of this alloy is shown in Figure 7. Due to the leaner alloying addition content, Alloy Two formed considerably less primary-like RCO than did Alloy One. The percentage area in SEM micrographs was found to decrease from 1.07% to 0.46% in the sample held at 650 °C vs the sample held at 750 °C. Unlike Alloy One, RCO size was found to decrease with increasing isothermal hold temperature.

Figure 6: Summary of a) average particle size for RCO particles and b) area fraction of SEM images occupied by RCO particles for the top, bottom, and middle sections of castings of Alloy One created using the FGNC technique with isothermal holds at 650, 700, and 750 °C.

Example 3 A further composition of “Alloy Three” with the composition Al-7.90Ce-5.70Cu-2.70Ni- 1.20Mn-1.00Mg was cast into a wedge mold. The casting procedure and details of cooling rate vs distance to the tip can be seen in the technical report published by the Army Research Laboratory (ARL-TN-1092). This casting had a plate sectioned from the center and optically imaged at 10x. Image analysis of RCO grains showed that at fast cooling rates from 5.25 to 4.25 °C/s a small number of RCO particles are observed in the microstructure. A transition region from 4.25 to 2.75 °C/s shows a steady increase in the number of RCO particles where finally a near steady state formation of many particles is observed from 2.75 to 1.5 °C/s.

Figure 7: SEM backscattered electron (BSE) micrographs showing the microstructure of an Al-Ce- Cu-Ni-Mn-Mg alloy (“alloy two”) created by isothermally holding in the melt at a) 650, or b) 750 oC for 30 minutes prior to casting at 800 oC. More RCO phase is present in the sample held at 650 oC as opposed to the sample held at 750 oC.

Figure 8 exemplifies the dependence on cooling rate for the formation of a primary phase in the RCO containing alloys. The same alloy composition was then remelted using the FGNC method holding at 650 °C. Figure 9 shows the microstructure of Alloy Three after casting using the FGNC method. For the top, middle, and bottom section no large dendritic growth of primary RCO grains can be observed which was observed in Alloy One with no hold or an isothermal hold at 750 °C.

Image analysis of the Alloy Three FGNC’s microstructure can be seen in Figure 10.

Sample’s area fractions of RCO particles are equivalent to ~4 °C/s in the wedge mold which matches the estimated cooling rate for the Cu cylinder mold based on the Chvorinov's rule calculation of ~196 s for the sample to solidify (assuming weighted averages for liquid density and heat capacity). There is expected to be a gradient in the cooling rate of the cast rod that is not observed in the microstructure, i.e. FGNC on the alloy reduced cooling rate dependence and improved homogeneity.

Figure 8: The (a) wedge casting of Alloy Three with (b) showing the change in area percent and number of particles per area for RCO particles vs the cooling rate.

Figure 9: SEM backscattered electron (BSE) micrographs showing the microstructure of Alloy Three created with a 650 °C isothermal hold for 30 minutes, showing samples taken from a) the bottom b) the middle and c) the top of the casting.

Example 4 The optimization of the isothermal hold temperature in the FGNC technique is crucial for controlling the nucleation and growth of secondary phases like the RCO phase.

Differential Scanning Calorimetry (DSC) serves as a vital tool in determining the precise temperatures at which these phases nucleate, allowing for adjustments in the FGNC process parameters. This example illustrates how both cooling rate and alloy composition can influence the nucleation temperature of the RCO phase, thereby altering the optimal isothermal hold temperature.

Alloy Four, whose composition is Al-6.89Ce-7.76Cu-2.30Ni-1.09Mn-1.17Mg was subjected to DSC analysis to observe the effect of different cooling rates on the nucleation temperature of the RCO phase. DSC cooling curves were obtained at rates of 3.5, 5, 7.5, 10, 15, 20 °C/min (see Figure 11a-b). In the area normalized curves, the position of the slope change accounting for RCO solidification shifts to lower temperatures as the cooling rate increases. When the cooling rate was increased to 20 °C/min, the nucleation peak shifted to a lower temperature of about 690 °C. This shift indicates that higher cooling rates increase the degree of undercooling required for nucleation, thereby delaying the onset of RCO phase formation.

The delayed nucleation at higher cooling rates suggests that the target isothermal hold temperature for the FGNC process must be adjusted accordingly. To validate this, Alloy Four samples were cooled to isothermal hold temperatures based on the DSC results corresponding to each cooling rate.

Figure 10: Summary of a) average particle size for RCO particles and b) area fraction of SEM images occupied by RCO particles for the top, bottom, and middle sections of castings of Alloy Three created using the FGNC technique with an isothermal hold at 650 °C.

Figure 12a shows the microstructure of a sample cooled at a slow rate of approximately 3.8°C/min, compared Figure 12b that was cooled at approximately 26 °C/min approaching the isothermal hold temperature (650 °C). Note that the faster cooling reaches the nucleation temperature later than the slower cooling rate, giving less time for particles to grow and change morphologies.

Alloy composition plays a significant role in determining the nucleation temperature of secondary phases. To demonstrate this effect on the FGNC process, a modified version of Alloy Four was prepared by adding Zn (designated as "Alloy Five"). The exact composition of Alloy Five is represented as Al-6.82Ce-7.44Cu-2.26Ni-0.96Mn-1.24Mg- 3.38Zn (wt%). DSC analysis of Alloy Five is presented in Figure 13 comparing the Alloy Four DSC at 20 °C/min curve to the Alloy Five 20 °C/min curve. The addition of Zn altered the thermodynamics of phase formation in the alloy, resulting in a shift of the nucleation temperature of the RCO phase to a lower temperature compared to Alloy Four.

a b

Figure 11: a) DSC solidification curves normalized to the area under each curve with varying cooling rates showing the large Al peak and the smaller slope change in the 650-725 °C range for RCO solidification. b) Section of the DSC curve showing the suppression of RCO crystallization as the cooling rate increases, which changes the target isothermal hold temperature.

Figure 12: a) Back Scatter Scanning Electron Microscopy images of Alloy Four that have undergone the FGC process with cooling rates a) 3.8 °C/min and b) 26 °C/min when cooling down to the isothermal hold temperature.

This shift can be attributed to the effect of the added Zn, which modifies the liquidus temperature and the enthalpy of formation for the RCO phase observed in a larger undercooling requirement for 20 °C/min cooling rates.

By incorporating element Zn into the alloy, the target isothermal hold temperature for the FGNC process can be adjusted. Figure 12a shows the microstructure of a sample cooled at a slow rate

This example underscores the importance of using DSC analysis to optimize the FGNC process parameters. Both the cooling rate and alloy composition significantly influence the nucleation temperature of strengthening phases like the RCO phase. By understanding these effects, the isothermal hold temperature can be accurately targeted to achieve the desired microstructure. Adjusting the cooling rate and composition provides flexibility in processing conditions and broadens the applicability of the FGNC technique to a wider range of alloy.

Figure 13: a) DSC solidification curves normalized to the area under each curve for Alloy Four and Five showing the effect of Zn addition. b) Section of the DSC curve showing the increase in nucleation temperature with Zn additions, which changes the target isothermal hold temperature.