Aluminum alloy with additions of scandium, zirconium and erbium

Abstract

An aluminum alloy including additions of scandium, zirconium, erbium and, optionally, silicon.

Claims

1. A method for forming an aluminum alloy comprising the steps of: forming a molten mass of aluminum comprising additions of scandium, zirconium, erbium and, optionally, silicon; cooling said molten mass to form a solid mass; during a first heat treating step, maintaining said solid mass at a temperature ranging from about 275 to about 325 C. for a first predetermined amount of time; and after said first heat treating step, maintaining said solid mass at a temperature ranging from about 375 to about 425 C. for a second predetermined amount of time.

2. The method of claim 1 wherein said first predetermined amount of time is about 2 to about 8 hours.

3. The method of claim 1 wherein said second predetermined amount of time is about 4 to about 12 hours.

4. The method of claim 1 wherein said first predetermined amount of time is about 2 to about 8 hours and said second predetermined amount of time is about 4 to about 12 hours.

5. The method of claim 1 wherein said molten mass consists essentially of said aluminum, said scandium, said zirconium, said erbium and, optionally, said silicon.

6. The method of claim 5 wherein iron is present in said molten mass as an impurity.

7. The method of claim 6 wherein said iron is present at a concentration of at most about 0.0025 at. %.

8. The method of claim 1 wherein: said scandium comprises at most about 0.1 at. % of said molten mass; said zirconium comprises at most about 0.1 at. % of said molten mass; said erbium comprises at most about 0.05 at. % of said molten mass; and said silicon comprises from 0 to about 0.1 at. % of said molten mass.

9. The method of claim 1 wherein: said scandium comprises at most about 0.08 at. % of said molten mass; said zirconium comprises at most about 0.08 at. % of said molten mass; and said erbium comprises at most about 0.04 at. % of said molten mass.

10. The method of claim 9 wherein said molten mass is substantially free of said silicon.

11. The method of claim 1 wherein: said scandium comprises at most about 0.06 at. % of said molten mass; said zirconium comprises at most about 0.06 at. % of said molten mass; and said erbium comprises at most about 0.02 at. % of said molten mass.

12. The method of claim 11 wherein said molten mass is substantially free of said silicon.

13. The method of claim 1 wherein: said scandium comprises at most about 0.08 at. % of said molten mass; said zirconium comprises at most about 0.08 at. % of said molten mass; said erbium comprises at most about 0.04 at. % of said molten mass; and said silicon comprises at most about 0.08 at. % of said molten mass.

14. The method of claim 1 wherein: said scandium comprises at most about 0.06 at. % of said molten mass; said zirconium comprises at most about 0.06 at. % of said molten mass; said erbium comprises at most about 0.02 at. % of said molten mass; and said silicon comprises at most about 0.04 at. % of said molten mass.

15. The method of claim 1 wherein said molten mass consists essentially of said aluminum, said scandium, said zirconium, said erbium, and said silicon.

16. The method of claim 1 further comprising the step of, prior to said first heat treating step, homogenizing said solid mass at a temperature of about 600 to about 660 C. for about 1 to about 20 hours.

17. A method for forming an aluminum alloy comprising the steps of: forming a molten mass consisting of: a non-zero quantity of scandium, present at a concentration of at most about 0.1 at. %; a non-zero quantity of zirconium, present at a concentration of at most about 0.1 at. %; a non-zero quantity of erbium, present at a concentration of at most about 0.05 at. %; from 0 to about 0.1 at. % silicon; and aluminum, forming substantially the balance of said molten mass; cooling said molten mass to form a solid mass; during a first heat treating step, maintaining said solid mass at a temperature ranging from about 275 to about 325 C. for about 2 to about 8 hours; and after said first heat treating step, maintaining said solid mass at a temperature ranging from about 375 to about 425 C. for about 4 to about 12 hours.

18. The method of claim 17 wherein iron is present in said molten mass as an impurity.

19. The method of claim 17 wherein the content of silicon in said molten mass is at least about 0.02 at. %.

20. The method of claim 17 wherein: the content of scandium in said molten mass is at most about 0.08 at. %; the content of zirconium in said molten mass is at most about 0.08 at. %; the content of erbium in said molten mass is at most about 0.04 at. %; and the content of silicon in said molten mass is about 0 at. %.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIGS. 1A and 1B are scanning electron microscope (SEM) micrographs of as-homogenized microstructures in Al-0.06 Zr-0.06 Sc (FIG. 1A) and Al-0.06 Zr-0.05 Sc-0.01 Er (FIG. 1B) (all compositions are given herein in atomic percent);

(2) FIGS. 2A and 2B are graphical illustrations of the evolution of the Vickers microhardness (FIG. 2A) and electrical conductivity (FIG. 2B) during isochronal aging in stages of 25 C. h.sup.1 for Al-0.06 Zr-0.06 Sc, Al-0.06 Zr-0.05 Sc-0.01 Er and Al-0.06 Zr-0.04 Sc-0.02 Er;

(3) FIGS. 3A and 3B are graphical illustrations of concentration profiles across the matrix/precipitate interface following isochronal aging to 450 C. in stages of 25 C. h.sup.1 for Al-0.06 Zr-0.06 Sc (FIG. 3A) and Al-0.06 Zr-0.04 Sc-0.02 Er (FIG. 3B), which were obtained using 3-D atom-probe tomography (APT);

(4) FIGS. 4A and 4B are graphical illustrations of the evolution of the Vickers microhardness (FIG. 4A) and electrical conductivity (FIG. 4B) during isothermal aging at 400 C. for Al-0.06 Zr-0.06 Sc, Al-0.06 Zr-0.05 Sc-0.01 Er and Al-0.06 Zr-0.04 Sc-0.02 Er;

(5) FIGS. 5A and 5B are graphical illustrations of concentration profiles across the matrix/precipitate interface for Al-0.06 Zr-0.04 Sc-0.02 Er samples aged isothermally at 400 C. for 0.5 h (FIG. 5A) and 64 days (FIG. 5B), which were obtained using 3-D APT;

(6) FIGS. 6A and 6B are graphical illustrations of the temporal evolution of the Vickers microhardness (FIG. 6A) and electrical conductivity (FIG. 6B) during isothermal aging at 400 C. for Al-0.06 Zr-0.06 Sc, Al-0.06 Zr-0.05 Sc-0.01 Er and Al-0.06 Zr-0.04 Sc-0.02 Er previously aged 24 hours at 300 C.;

(7) FIGS. 7A-7H depicts optical and SEM micrographs of Al-0.06 Zr-0.06 Sc-0.04 Si and Al-0.06 Zr-(0.05 Sc-0.01 Er)-0.04 Si after heat treatment;

(8) FIGS. 8A and 8B are graphical illustrations of average concentration profiles across the matrix/precipitate interface after a two-stage peak-aging treatment (4 h at 300 C. followed by 8 h at 425 C.) for Al-0.06 Zr-0.06 Sc-0.04 Si (FIG. 8A) and Al-0.06 Zr-(0.05 Sc-0.01 Er)-0.04 Si (FIG. 8B), which were obtained using 3-D APT;

(9) FIG. 9 is a double logarithmic plot of minimum creep rate versus applied stress for compressive creep experiments at 400 C. for Al-0.06 Zr-0.06 Sc-0.04 Si and Al-0.06 Zr-(0.05 Sc-0.01 Er)-0.04 Si after heat treatment; and

(10) FIG. 10 is a double logarithmic plot of minimum creep rate versus applied stress for compressive creep experiments at 400 C. for Al-0.06 Zr-(0.05 Sc-0.01 Er)-0.04 Si (a) after a two-stage peak-aging treatment (4 h/300 C. and 8 h/425 C.) and (b) after subsequent exposure at 400 C. for 325 h at applied stresses ranging from 6 to 8.5 MPa.

DETAILED DESCRIPTION

(11) It has now been discovered that the substitution of some scandium with the lower-cost rare earth element erbium may be effective in maintaining high-temperature strength, and improving the creep resistance, of aluminum-scandium-zirconium alloys at temperatures as high as 400 C.

(12) In a first aspect, the disclosed aluminum alloy may include aluminum with additions of scandium, zirconium and erbium.

(13) In one particular implementation of the first aspect, the disclosed aluminum alloy may include at most about 0.1 at. % scandium, at most about 0.1 at. % zirconium and at most about 0.05 at. % erbium, with the balance of the alloy being substantially aluminum.

(14) In another particular implementation of the first aspect, the disclosed aluminum alloy may include at most about 0.08 at. % scandium, at most about 0.08 at. % zirconium and at most about 0.04 at. % erbium, with the balance of the alloy being substantially aluminum.

(15) In yet another particular implementation of the first aspect, the disclosed aluminum alloy may include at most about 0.06 at. % scandium, at most about 0.06 at. % zirconium and at most about 0.02 at. % erbium, with the balance of the alloy being substantially aluminum.

(16) Those skilled in the art will appreciate that the disclosed aluminum alloys may include trace amounts of impurities, such as iron and silicon, without departing from the scope of the present disclosure. For example, iron and silicon may be present in the disclosed aluminum alloys in amounts below 0.0025 and 0.005 at. %, respectively.

(17) Without being limited to any particular theory, it is believed that the addition of scandium to aluminum leads to the precipitation of a strengthening Al.sub.3Sc phase in the form of numerous coherent precipitates. The Al.sub.3Sc phase is rendered coarsening resistant by the addition of zirconium, which precipitates to form an Al.sub.3(Sc,Zr) outer shell on the Al.sub.3Sc precipitate core. The addition of erbium substitutes for some of the scandium in the precipitate, while also increasing the precipitate's lattice parameter mismatch with the aluminum matrix, thereby improving creep properties at high temperatures.

(18) It has also been discovered that the presence of silicon in the disclosed aluminum alloy may accelerate the precipitation kinetics of scandium. Therefore, silicon may be intentionally added to the disclosed aluminum alloy to minimize the amount of heat treating, and hence energy cost and use of furnaces, required to achieve peak strength from Al.sub.3Sc (L1.sub.2) precipitates.

(19) Therefore, in another aspect, the disclosed aluminum alloy may include aluminum with additions of scandium, zirconium, erbium and silicon.

(20) In one particular implementation of the second aspect, the disclosed aluminum alloy may include at most about 0.1 at. % scandium, at most about 0.1 at. % zirconium, at most about 0.05 at. % erbium and at most about 0.1 at. % silicon, with the balance of the alloy being substantially aluminum.

(21) In another particular implementation of the second aspect, the disclosed aluminum alloy may include at most about 0.08 at. % scandium, at most about 0.08 at. % zirconium, at most about 0.04 at. % erbium and at most about 0.08 at. % silicon, with the balance of the alloy being substantially aluminum.

(22) In yet another particular implementation of the second aspect, the disclosed aluminum alloy may include at most about 0.06 at. % scandium, at most about 0.06 at. % zirconium, at most about 0.02 at. % erbium and at most about 0.04 at. % silicon, with the balance of the alloy being substantially aluminum.

EXAMPLES

Alloys 1-3

Alloy Compositions and Processing

(23) A ternary and two quaternary alloys were cast with nominal compositions, in atomic percent (at. %), of Al-0.06 Zr-0.06 Sc (Alloy 1) (comparative example), Al-0.06 Zr-0.05 Sc-0.01 Er (Alloy 2) and Al-0.06 Zr-0.04 Sc-0.02 Er (Alloy 3). The compositions of Alloys 1-3 in the as-cast state, as measured by direct current plasma emission spectroscopy (DCPMS) (ATI Wah Chang, Albany, Oreg.) and 3-D local-electrode atom-probe (LEAP) tomography, are provided in Table 1. The silicon and iron content of the alloys was less than the 0.005 and 0.0025 at. % detection limits, respectively, of the DCPMS technique.

(24) TABLE-US-00001 TABLE 1 Measured Composition Measured Composition (DCPMS) (3-D LEAP) Alloy Zr Sc Er Zr Sc Er 1 0.052 0.067 0.0256 0.0685 2 0.035 0.047 0.01 0.0198 0.0476 0.0038 3 0.035 0.042 0.019 0.02 0.0394 0.0046

(25) The alloys were dilution cast from 99.999 at. % pure Al (Alfa Aesar, Ward Hill, Mass.) and Al-0.9 at. % Sc, Al-0.6 at. % Zr and Al-1.15 at. % Er master alloys. The AlSc and AlZr master alloys were themselves dilution cast from commercial Al-1.3 at. % Sc (Ashurst Technology, Ltd., Baltimore, Md.) and Al-3 at. % Zr (KB Alloys, Reading, Pa.) master alloys. The AlEr master alloy was prepared by melting 99.999 at. % pure Al with 99.99 at. % Er (StanfordMaterials Corporation, Aliso Viejo, Calif.) using non-consumable electrode arc-melting in a gettered purified-argon atmosphere (Atlantic Equipment Engineers, Bergenfield, N.J.). To create the final dilute alloys, the master alloys and 99.999 at. % pure Al were melted in flowing argon in zirconia-coated alumina crucibles in a resistively heated furnace at 850 C. The master alloys were preheated to 640 C. to accelerate solute dissolution and minimize solute losses from the melt. The melt was held in a resistively heated furnace for 7 min at 850 C., stirred vigorously, and then cast into a graphite mold preheated to 200 C. During solidification, the mold was chilled by placing it on an ice-cooled copper platen to encourage directional solidification and discourage the formation of shrinkage cavities.

(26) The castings were homogenized in air at 640 C. for 72 h and then water quenched to ambient temperature.

(27) Three separate aging studies were conducted: (i) isochronal aging in stages of 25 C. h.sup.1 for temperatures from 100 to 600 C.; (ii) isothermal aging at 400 C. for times ranging from 0.5 min to 256 days (8 months); and (iii) two-stage isothermal aging consisting of a first heat treatment at 300 C. for 24 h followed by aging at 400 C. for times ranging from 0.5 h to 64 days. Molten salt (NaNO.sub.2NaNO.sub.3KNO.sub.3) baths were used for aging durations less than 0.5 h to ensure rapid heat transfer, while longer aging experiments were performed in air.

Analytical Techniques

(28) The homogenized microstructure of unetched samples polished to a 1 m surface finish was imaged by SEM using a Hitachi S3400N-II microscope, equipped with an Oxford Instruments INCAx-act detector for energy-dispersive X-ray spectroscopy (EDS). The precipitate morphology was studied using a Hitachi 8100 transmission electron microscope at 200 kV. TEM foils were prepared by grinding aged specimens to a thickness of 100-200 m, from which 3 mm diameter disks were punched. These disks were thinned by twin-jet electropolishing at about 20 V DC using a Struers TenuPol-5 with a 10 vol. % solution of perchloric acid in methanol at 40 C.

(29) Precipitation in these alloys was monitored by Vickers microhardness and electrical conductivity measurements. Vickers microhardness measurements were performed on a Duramin-5 microhardness tester (Struers) using a 200 g load applied for 5 s on samples polished to a 1 m surface finish. Fifteen indentations were made per specimen across several grains. Electrical conductivity measurements were performed using a Sigmatest 2.069 eddy current instrument (Foerster Instruments, Pittsburgh, Pa.) at frequencies of 120, 240, 480 and 960 kHz.

(30) Specimens for three-dimensional local-electrode atom-probe (3-D LEAP) tomography were prepared by cutting blanks with a diamond saw to approximate dimensions of 0.35 by 0.35 by 10 mm.sup.3. These were electropolished at 8-20 V DC using a solution of 10% perchloric acid in acetic acid, followed by a solution of 2% perchloric acid in butoxyethanol at room temperature. Pulsed-laser 3-D atom-probe tomography was performed with a LEAP 4000X Si X tomograph (Cameca, Madison, Wis.) at a specimen temperature of 35 K, employing focused picosecond UV laser pulses (wavelength=355 nm) with a laser beam waist of less than 5 mm at the e.sup.2 diameter. A laser energy of 0.075 nJ per pulse, a pulse repetition rate of 250 kHz, and an evaporation rate of 0.04 ions per pulse were used. 3-D LEAP tomographic data were analyzed with the software program IVAS 3.4.1 (Cameca). The matrix/precipitate heterophase interfaces were delineated with Sc isoconcentration surfaces, and compositional information was obtained with the proximity histogram methodology. The measurement errors for all quantities were calculated based on counting statistics and standard error propagation techniques.

As-Homogenized Microstructural Analysis

(31) The homogenized microstructure of the alloys consists of columnar grains with diameters of the order of 1-2 mm. SEM shows the presence of intragranular Al.sub.3Zr flakes in all alloys, which are retained from the melt due to incomplete dissolution of the AlZr master alloy (FIG. 1A). The approximate composition of the flakes was obtained by semi-quantitative EDS, i.e. without rigorous calibration, which confirms the Al.sub.3Zr stoichiometry, and reveals neither Er nor Sc in the flakes. The differences between the nominal and measured Zr concentrations of the alloys in Table 1 are believed to be a result of these Zr-rich flakes, which are not uniformly distributed in the alloys, and may have been excluded from the 300 mm.sup.3 of material used for DCPMS. No Al.sub.3Zr flakes were present in the small analysis volume of the 3-D LEAP tomo graphic reconstructions, and therefore the average of the measured Zr concentrations from the 3-D LEAP tomographic datasets of each alloy (Table 1) shows the Zr available in the matrix for precipitation during aging.

(32) In the Er-containing alloys, intergranular Al.sub.3Er (L1.sub.2) primary precipitates were detected, and contained neither Zr nor Sc, as confirmed by EDS (FIG. 1B). Primary precipitation in these alloys decreases strength by depleting the matrix of solute and, when excessive, can result in grain refinement, reducing the resistance to diffusional creep. The formation of primary precipitates in the homogenized samples indicates that the Er-containing alloys exceeded their solubility limit during solidification and homogenization. The addition of Sc and Zr has thus decreased the 0.046 at. % solubility of Er in binary AlEr. The analysis volume of the 3-D LEAP tomography technique is too small to detect intergranular Al.sub.3Er, as was the case for the Al.sub.3Zr flakes. The 3-D LEAP-tomographic measured compositions of Er of 0.00460.0004 and 0.00380.0004 at. % for Al-0.06 Zr-0.04 Sc-0.02 Er and Al-0.06 Zr-0.05 Sc-0.01 Er, are well below the nominal values of 0.02 and 0.01 at. % Er, respectively (Table 1). Only a fraction of the Er added to the alloys is available for nanoscale precipitation.

(33) Previous research on arc-melted Al-0.06 Zr-0.06 Sc and Al-0.1 Zr-0.1 Sc at. % alloys revealed microsegregation of both Sc and Zr in the as-cast condition using linear composition profiles obtained employing quantitative electronprobe microanalysis (EPMA). The first solid to form in dilute AlZrSc alloys is enriched in Zr, resulting in a microstructure consisting of Zr-enriched dendrites surrounded by Sc-enriched interdendritic regions. The as-cast Al-0.06 Zr-0.06 Sc at. % alloy in the previous work showed a Zr enrichment of about 0.04 at. % Zr and a Sc depletion of about 0.01 at. % in the dendrites with respect to the average alloy composition, while the interdendritic region was depleted by about 0.04 at. % Zr and enriched by about 0.02 at. % Sc. Microsegregation is expected in the present alloys, though to a lesser extent than in the previous Al-0.06 Zr-0.06 Sc and Al-0.1 Zr-0.1 Sc alloys, because the incomplete dissolution of the AlZr master alloy diminishes the effective Zr alloy concentration to 0.02-0.03 at. % (Table 1).

(34) The degree of solute microsegregation in the present research is also diminished by homogenization at 640 C. for 72 h, which was not performed in prior work on Al-0.06 Zr-0.06 Sc due to concerns about primary precipitation of Al.sub.3Zr. In a similar study on Al-0.06 Sc, the microsegregation of Sc was completely eliminated by homogenization at 640 C. for 28 h. Given that the diffusivity of Zr in Al, 1.010.sup.15 m.sup.2 s.sup.1, is significantly smaller than that of Sc in Al, 6.710.sup.14 m.sup.2 s.sup.1, at 640 C., homogenization of Zr requires heat-treatment durations that are impractically long.

(35) In summary, the effective Zr and Er concentrations of the alloys are believed to be smaller than their nominal values due to incomplete dissolution of the AlZr master alloy, and the formation of intergranular primary Al.sub.3Er (L1.sub.2) precipitates. For simplicity, the nominal compositions are used herein to label the alloys.

Isochronal Aging

(36) The precipitation behavior of Alloys 1-3 during isochronal aging in stages of 25 C. h.sup.1 is shown in FIG. 2, as monitored by Vickers microhardness and electrical conductivity. In Alloy 1 (Al-0.06 Zr-0.06 Sc), precipitation commences at 300 C., as reflected by a sharp increase in the microhardness and electrical conductivity. The microhardness peaks for the first time at 350 C. and achieves a value of 5825 MPa, before decreasing to 54316 MPa at 400 C. The microhardness increases again at 425 C., achieving a second peak of 59716 MPa at 450 C. The electrical conductivity increases continuously from 300 to 375 C., before reaching a plateau at values of 33.940.09 and 33.990.09 MS m.sup.1 for 375 and 400 C. At 425 C., the electrical conductivity increases to 34.750.10 MS m.sup.1, reaching a peak of 34.920.11 MS m.sup.1 at 450 C. Above 450 C., both microhardness and electrical conductivity decrease quickly due to precipitate dissolution.

(37) The first peak in the microhardness of Alloy 1 at 325 C. occurs at the same temperature as the peak microhardness in recent studies of Al-0.06 Sc and Al-0.1 Sc alloys aged isochronally for 3 h for every 25 C. increase. As such, the first peak in the microhardness we observe can be attributed to the precipitation of Al.sub.3Sc. The second peak in the microhardness at 450 C. occurs at the same temperature as was previously found to produce a peak in the microhardness of an Al-0.1 Zr alloy aged isochronally for 3 h for every 25 C. increase. The peak microhardness in an Al-0.06 Zr alloy was found to occur at 475 C. for samples aged isochronally for 3 h for every 25 C. increase. The second peak in the microhardness is thus due to precipitation of Zr from the matrix. Previously studied Al-0.06 Zr-0.06 Sc and Al-0.1 Zr-0.1 Sc alloys aged isochronally for 3 h for every 25 C. increase were found to have only one peak in the microhardness, occurring at 400 C. The detection of only one peak in the microhardness was probably due to the smaller temporal resolution used in the previous studies, compared to the isochronal aging of 1 h for every 25 C. employed for Alloys 1-3.

(38) The peak microhardness of the Er-containing alloys (Alloys 2 and 3) is smaller than that observed in Alloy 1. These results are consistent with isochronal microhardness results from Al-0.12 Sc and Al-0.9 Sc-0.03 Er alloys, where it was reasoned that the decrease in strength with the addition of Er was a result of solute consumption by primary precipitates, such as those in FIG. 1A. Nanoscale precipitation in the Er-containing alloys, as evidenced by increases in microhardness and conductivity, begins at temperatures as low as 200 C. The microhardness values of the Er-containing alloys achieve a plateau between 325 and 450 C. Beyond 450 C., both microhardness and electrical conductivity decrease rapidly due to precipitate dissolution, as observed in Al-0.06 Zr-0.06 Sc. The electrical conductivity of homogenized Al-0.06 Zr-0.06 Sc of 31.50.2 MS m.sup.1 is significantly smaller than the values of 32.60.2 and 33.00.2 MS m.sup.1 for Al-0.06 Zr-0.05 Sc-0.01 Er (Alloy 2) and Al-0.06 Zr-0.04 Sc-0.02 Er (Alloy 3), respectively. This is a result of primary precipitation of Al.sub.3Er (L1.sub.2) in the Er-containing alloys, which deprives the matrix of solute and increases the electrical conductivity.

(39) The nanostructures of Al-0.06 Zr-0.06 Sc and Al-0.06 Zr-0.04 Sc-0.02 Er aged isochronally to peak strength at 450 C., and obtained from 3-D LEAP tomography. The Al-0.06 Zr-0.06 Sc alloy, has a number density of precipitates, N.sub.v, of 2.10.210.sup.22 m.sup.3, with an average radius, <R>, of 3.10.4 nm, and a volume fraction, , of 0.2510.002%. The number density in Al-0.06 Zr-0.04 Sc-0.02 Er is smaller, 8.61.510.sup.21 m.sup.3, with average radius and volume fraction values of 3.40.6 nm and 0.1570.003%, respectively. The number density and volume fraction of precipitates are smaller in the Er-containing alloy because the matrix solute supersaturation is smaller due to primary precipitation of Er during solidification and homogenization (FIG. 1). The concentration profiles across the matrix/precipitate interface obtained from the 3-D LEAP tomographic results are displayed in FIG. 3. As anticipated, the precipitates in Al-0.06 Zr-0.06 Sc consist of a Sc-enriched core surrounded by a Zr-enriched shell, with an average precipitate composition of 71.950.10 at. % Al, 5.420.05 at. % Zr and 22.630.09 at. % Sc. The precipitates in Al-0.06 Zr-0.04 Sc-0.02 Er consist of an Er-enriched core surrounded by a Sc-enriched inner shell and a Zr-enriched outer shell, with an average precipitate composition of 73.270.15 at. % Al, 5.010.07 at. % Zr, 18.960.13 at. % Sc and 2.750.05 at. % Er.

Isothermal Aging at 400 C.

(40) The precipitation behavior of the alloys during isothermal aging at 400 C. for aging times from 0.5 min to 256 days, as monitored by Vickers microhardness and electrical conductivity, is displayed in FIG. 4. The Vickers microhardness of Alloy 1 (Al-0.06 Zr-0.06 Sc) does not increase significantly over the full range of aging times, which is surprising given the strengths achieved by isochronal aging (see FIG. 2). The electrical conductivity of Alloy 1 remains unchanged over the first 0.5 h of aging at 400 C., before increasing steadily over the subsequent 64 days. Small strengths in dilute AlSc alloys with Sc concentrations of 0.06-0.07 at. % have been observed previously to be a result of inadequate solute supersaturation, resulting in a small number density of larger precipitates, which do not strengthen the material significantly. The precipitates, which have large radii, of the order of 50 nm, have a non-equilibrium lobed-cuboidal morphology. This morphology is believed to be due to growth instabilities that accommodate the anisotropy of the elastic constants of the matrix and the precipitates.

(41) The microhardness values of the two Er-containing alloys, Alloys 2 and 3, during isothermal aging at 400 C. are comparable over the full range of aging times. Both alloys exhibit a microhardness increase after 0.5 min, with a concomitant increase in the electrical conductivity. After 0.5 h of aging, the microhardness values of Alloys 1 and 2 are 42212 and 41411 MPa, respectively. This is in dramatic contrast to the Er-free alloy (Alloy 1), whose microhardness does not increase beyond the homogenized value of 19914 MPa after 0.5 h, and achieves a peak microhardness of only 2433 MPa after 8 days at 400 C. By contrast, the microhardness of Alloy 2 peaks at a value of 46115 MPa after 2 days, and diminishes slightly to 43821 MPa after 64 days of aging at 400 C. Alloy 3 has a maximum microhardness of 45111 MPa after 1 day of aging, and has the same microhardness, within uncertainty, of 44821 MPa after 64 days at 400 C. The microhardness values of Alloys 2 and 3 decrease for aging times of 128 and 256 days due to precipitate coarsening. The electrical conductivities of Alloys 2 and 3 increase steadily over the first 1-2 days, as precipitation proceeds. Between 2 and 64 days, the electrical conductivities of both alloys achieve plateaus, indicating that the majority of the available solute has precipitated out of solution. The electrical conductivities of Alloys 2 and 3 increase slightly after 128 and 256 days of aging, as the alloys continue to slowly approach equilibrium.

(42) The nanostructures of Alloy 3 aged isothermally for 0.5 h and 64 days at 400 C. were compared employing 3-D LEAP tomography. From the 3-D LEAP tomographic images, and the associated concentration profiles (FIG. 5), it is clear that the precipitates consist of an Er-enriched core surrounded by a Sc-enriched shell after 0.5 h of aging. After 0.5 h of aging, Alloy 3 has a number density of precipitates of 5.41.710.sup.21 m.sup.3, with an average radius of 3.70.3 nm, and a volume fraction of 0.1440.006%. The number density of 6.11.910.sup.21 m.sup.3 and the radius of 3.80.4 nm are unchanged, within uncertainty, after 64 days at 400 C., although the volume fraction increases to 0.2070.007%.

(43) After 0.5 h of aging at 400 C., the precipitates in Alloy 3 consist of an Er-enriched core surrounded by a Sc-enriched shell structure with an average precipitate composition of 73.020.20 at. % Al, 0.640.04 at. % Zr, 22.250.19 at. % Sc and 4.080.09 at. % Er at. %. The average precipitate composition after 64 days at 400 C., 70.460.22 at. % Al, 6.550.12 at. % Zr, 19.750.19 at. % Sc, 3.240.09 at. % Er, reflects the precipitation of the Zr-enriched outer shell, which renders the precipitates coarsening resistant. The matrix is depleted of Sc and Zr as precipitation proceeds, as evidenced by decreases in the Zr concentration from 16714 to 3515 at. ppm, and in Sc from 706 to 256 at. ppm between 0.5 h and 64 days.

(44) The precipitation behavior of Alloys 1-3 exhibits three distinct stages of development at 400 C., as shown in FIG. 4. In the Er-containing alloys, a short incubation period of 0.5 min is followed by a rapid increase in the microhardness and electrical conductivity over the first hour, associated with the precipitation of Er and Sc, which is followed by a slower increase in conductivity due to the precipitation of Zr. In Alloy 1, the incubation period of 0.5 h is followed by a rapid increase in the electrical conductivity from 0.5 to 24 h as Sc precipitates from solution, followed by a slow second increase in the conductivity due to precipitation of Zr.

Two-Stage Isothermal Aging

(45) A two-stage heat treatment was performed: (i) to improve the microhardness of Alloy 1 at 400 C.; and (ii) to optimize the nanostructure, and hence the microhardness, of Alloys 2 and 3.

(46) The first stage of the heat treatment was performed at 300 C. for 24 h. The objective of this first stage is to precipitate the Er and Sc atoms from solid solution at a temperature as low as practical, maximizing the solute supersaturation, and hence the number density of precipitates. Zr is essentially immobile in Al at 300 C. over a period of 24 h, with a root-mean-square (RMS) diffusion distance of 1.5 nm, as compared to RMS diffusion distances of 56 and 372186 nm for Sc and Er, respectively.

(47) The second stage of the heat treatment, designed to precipitate Zr, was performed at 400 C. for aging times ranging from 0.5 h to 64 days. At 400 C., the Zr RMS diffusion distance after 24 h is 64 nm, comparable to the Sc RMS diffusion distance of 56 nm in 24 h at 300 C. The precipitation response during the second stage, as monitored by the Vickers microhardness and electrical conductivity, is shown in FIG. 6.

(48) The microhardness of Alloy 1 following the two-stage 300/400 C. heat treatment (FIG. 6), is significantly improved compared to the values measured for the single isothermal aging at 400 C. (FIG. 4). After 24 h at 300 C., the microhardness of Alloy 1 is 5237 MPa, compared to 2363 MPa after 24 h at 400 C. (FIG. 4). The aging treatment at 300 C. provides sufficient solute supersaturation to precipitate a significant number density (10.sup.21-10.sup.22 m.sup.3), of spheroidal precipitates, such as those obtained during isochronal aging. Following a second heat treatment of 8 h at 400 C., the microhardness achieves a maximum value of 56114 MPa, and decreases only slightly to 53331 MPa after 64 days at 400 C.

(49) The Er-containing alloys (Alloys 2 and 3) achieve peak microhardness after 8 h of aging at 400 C., with values of 50711 and 48911 MPa for Alloys 2 and 3, respectively. These peak values are larger than those achieved in single-stage isothermal aging at 400 C. (46115 and 45111 MPa). The Er-containing alloys (Alloys 2 and 3) that underwent two-stage aging experience only a slight decrease in microhardness after 64 days at 400 C., from 50711 to 46423 MPa for Alloy 2, and from 48911 to 45819 MPa for Alloy 3.

(50) Thus, Zr and Er are effective replacements for Sc in AlSc systems, accounting for 331% of the total precipitate solute content in Al-0.06 Zr-0.04 Sc-0.02 Er aged at 400 C. for 64 days. The addition of Er to the AlScZr system was found to result in the formation of coherent, spheroidal, L12-ordered precipitates with a nanostructure consisting of an Er-enriched core surrounded by a Sc-enriched inner shell and a Zr-enriched outer shell were formed. This core/double-shell structure is formed upon aging as solute elements precipitate sequentially according to their diffusivities, where D.sub.Er>D.sub.Sc>D.sub.Zr. The core/double-shell structure remains coarsening resistant for at least 64 days at 400 C.

Alloys 4 and 5

Alloy Compositions and Processing

(51) Two alloys were prepared with nominal compositions, in atomic percent (at. %), of Al-0.06 Zr-0.06 Sc-0.04 Si (Alloy 4) (comparative example) and Al-0.06 Zr-(0.05 Sc-0.01 Er)-0.04 Si (Alloy 5). Alloys 4 and 5 were inductively-melted to a temperature of 900 C. from 99.99 at. % pure Al, 99.995 at. % Si, and Al-0.96 at. % Sc, Al-3 at. % Zr and Al-78 at. % Er master alloys. The two alloys were cast into a cast-iron mold preheated to 200 C. The compositions of Alloys 4 and 5 in the as-cast state, as measured using direct current plasma emission spectroscopy (DCPMS) and three dimensional local-electrode atom-probe (3-D LEAP) tomography are given in Table 2. The impurity iron content of Alloys 4 and 5 was 0.006 at. %.

(52) TABLE-US-00002 TABLE 2 Measured Composition Measured Composition (DCPMS) (3-D LEAP) Alloy Si Zr Sc Er Si Zr Sc Er 4 0.036 0.062 0.059 0.0211 0.0441 0.0583 5 0.033 0.056 0.046 0.011 0.0347 0.0412 0.0434 0.0044

(53) The cast alloys were homogenized in air at 640 C. for 72 h and then water quenched to ambient temperature. A two-stage aging treatment of 4 h at 300 C. followed by 8 h at 425 C. was employed to achieve peak strength and coarsening resistance, as explained above. The second stage temperature of 425 C. was selected so that the final aging temperature was higher than the creep testing temperature of 400 C.

Microstructure Observations

(54) The microstructures of samples polished to a 1 m surface finish were imaged by SEM using a Hitachi S3400N-II microscope, equipped with an Oxford Instruments INCAx-act detector for energy-dispersive x-ray spectroscopy (EDS). Polished specimens were then etched for 30 s using Keller's reagent to reveal their grain boundaries. Vickers microhardness measurements were performed on a Duramin-5 microhardness tester (Struers) using a 200 g load applied for 5 s on samples polished to a 1 m surface finish. Fifteen indentations were made per specimen across several grains.

(55) Specimens for three-dimensional local-electrode atom-probe (3-D LEAP) tomography were prepared by cutting blanks with a diamond saw to dimensions of 0.350.3510 mm.sup.3. These were electropolished at 8-20 Vdc using a solution of 10% perchloric acid in acetic acid, followed by a solution of 2% perchloric acid in butoxyethanol at room temperature. Pulsed-voltage 3-D atom-probe tomography (APT) was performed with a LEAP 4000X Si X tomograph (Cameca, Madison, Wis.) at a specimen temperature of 35 K, employing a pulse repetition rate of 250 kHz, a pulse fraction of 20%, and an evaporation rate of 0.04 ions per pulse. 3-D LEAP tomo graphic data were analyzed with the software program IVAS 3.4.1 (Cameca). The matrix/precipitate heterophase interfaces were delineated with Al isoconcentration surfaces, and compositional profiles were obtained with the proximity histogram (proxigram) methodology. The measurement errors for all quantities were calculated based on counting statistics and standard error propagation techniques.

(56) Previous attempts to measure Si concentrations in Al by 3-D LEAP tomography have resulted in measured values that are smaller than both the expected nominal value, and the value measured by DCPMS. For the 3-D LEAP tomographic operating conditions we employed, Si evaporates exclusively as .sup.28Si.sup.2+, whose peak in the mass spectrum lies in the decay tail of the .sup.27Al.sup.2+ peak, further reducing the accuracy of the concentration measurement. The Si.sup.2+ concentration is measured to be less than both the nominal and DCPMS measured values (Table 2).

Creep Experiments

(57) Constant load compressive creep experiments were performed at 4001 C. on cylindrical samples with a diameter of 10 mm and a height of 20 mm. The samples were heated in a three-zone furnace, and the temperature was verified by a thermocouple placed within 1 cm of the specimen. The samples were placed between boron nitride-lubricated alumina platens and subjected to uniaxial compression by Ni superalloy rams in a compression creep frame using dead loads. Sample displacement was monitored with a linear variable displacement transducer with a resolution of 6 m, resulting in a minimum measurable strain increment of 310.sup.4. When a measurable steady-state displacement rate was achieved for a suitable duration, the applied load was increased. Thus, a single specimen yielded minimum creep rates for a series of increasing stress levels, at the end of which the strain did not exceed 11%. Strain rates at a given load were obtained by measuring the slope of the strain versus time plot, in the secondary, or steady-state, creep regime.

Microstructure

(58) The microstructures of the peak-aged Er-free (Alloy 4) and Er-containing (Alloy 5) alloys are displayed in FIGS. 7a and 7b, respectively. The grains in both alloys are elongated radially along the cooling direction, with smaller grains at the center of the billet, as expected for cast alloys. Alloy 5 has smaller grains than Alloy 4, with a larger grain density of 2.10.2 compared to 0.50.1 grains mm.sup.2, as determined by counting grains in the billet cross-sections. The finer grain structure in Alloy 5 is due to intergranular Al.sub.3Er precipitates with trace amounts of Sc and Zr, with diameters of about 2 m, visible in FIG. 7C, and with compositions verified by semi-quantitative EDS. These particles inhibit grain growth after solidification and/or during homogenization. Such primary precipitates were not observed in Alloy 4, indicating that the solubility limit of Alloy 5 was exceeded during solidification and heat-treatment. The addition of Sc and Zr has thus significantly decreased the 0.046 at. % solubility of Er in a binary AlEr alloy. The Er concentration, as measured by 3-D LEAP tomography in the matrix of the peak-aged Er-containing alloy (Alloy 5) is 0.00440.0005 at. %.

(59) Thus, less than half of the nominal value of 0.01 at. % Er is available for nanoscale precipitates formed on aging, while the remainder is present in the coarser primary Al.sub.3Er precipitates. Alloy 5 also contains submicron intragranular Al.sub.3Er precipitates, FIG. 7C, which is probably a result of microsegregation during solidification. The first solid to form in dilute AlZrScEr alloys is enriched in Zr, resulting in a microstructure consisting of Zr-enriched dendrites surrounded by Sc and Er-enriched interdendritic regions.

(60) In summary, the presence of Al.sub.3Er primary precipitates refines the grain size and reduces the effective Er concentration available for strengthening nanoscale precipitation. In the following, the nominal compositions are used to label the alloys.

Nanostructure of Peak-Aged Alloys

(61) The nanostructures of Alloys 4 and 5, after aging isothermally for 4 h at 300 C. and 8 h at 425 C., were compared employing 3-D LEAP tomography. The spheroidal precipitates in the Er-free alloy (Alloy 4) consist of a Sc-enriched core surrounded by a Zr-enriched shell, as shown in FIG. 8. The precipitates have an average radius of 2.40.5 nm, a number density of 2.50.510.sup.22 m.sup.3 and a volume fraction of 0.2590.007%. The spheroidal precipitates in the Er-containing alloy (Alloy 5) consist of a core enriched in both Er and Sc surrounded by a Zr-enriched shell, with an average radius, <R>, of 2.30.5 nm, a number density, N.sub.v, of 2.00.310.sup.22 m.sup.3, and a volume fraction of 0.2800.006%. Silicon partitions to the precipitate phase and shows no preference for the precipitate core or shell in either alloy.

(62) The precipitate and matrix compositions of the two alloys demonstrate that all alloying additions (Si, Zr, Sc and Er) partition to the precipitate phase. The matrix of the Er-containing alloy (Alloy 5) is more depleted of solute, with a composition of 10712 at. ppm Zr, 324 at. ppm Sc and 74 at. ppm Er, than that of the Er-free alloy (Alloy 4), with a composition of 15328 at. ppm Zr, 8914 at. ppm Sc.

Peak-Aged Condition

(63) The as-cast microhardness values of Alloys 4 and 5 are 2564 and 2708 MPa, respectively. These microhardness values are larger than those of previous as-cast dilute AlScX alloys, with comparable solute contents, of 210-240 MPa. The larger microhardness values may be evidence of early-stage clustering or precipitation, possibly as a result of the addition of Si, which accelerates precipitate nucleation in an Al-0.06 Zr-0.06 Sc at. % alloy aged at 300 C. After homogenization and peak-aging, the microhardness values of the present alloys increase to 62710 and 60620 MPa, respectively.

(64) FIG. 9 displays the minimum compressive strain rate versus uniaxial compressive stress at 400 C. for Alloys 4 and 5 tested in the peak-aged condition. The apparent stress exponent for dislocation climb-controlled creep for Alloy 4 (measured over the range 7-13 MPa) is 161, which is significantly greater than that of 4.4 expected for Al. Larger than expected stress exponents were previously measured in other AlSc-based alloys and are indicative of a threshold stress for creep, below which dislocation creep is not measurable in laboratory time frames.

(65) The microstructures of Alloys 4 and 5 following creep testing at 400 C. are displayed in FIGS. 7D and 7E, respectively. After creep at 400 C., the grains in Alloy 4 (FIG. 7D) appear unchanged with 0.60.1 grains mm.sup.2, compared to the 0.50.1 grains mm.sup.2 before creep (FIG. 7A). The grains in Alloy 5 following creep (FIG. 7E) have undergone recrystallization, resulting in an increase in the grain density to 3.60.2 grains mm.sup.2 from the pre-creep value of 2.10.2 (FIG. 7B). The intergranular Al.sub.3Er precipitates remain following creep (FIG. 7F).

Over-Aged Condition

(66) To collect more data in the diffusional creep regime of Alloy 5, a second series of creep experiments was performed at 400 C. on another peak-aged sample, beginning at a lower applied stress of 6 MPa. Compressive creep data were collected over 325 h for four stresses ranging from 6-8.5 MPa, which yielded a nearly constant strain rate of 1.20.210.sup.8 s.sup.1, where the error is the standard deviation of the four resulting strain rates. A constant strain rate for increasing applied stress is indicative of an evolving microstructure, that is, grain growth during the creep test. Since the rate of diffusional creep at a given stress decreases with increasing grain size, grain growth can account for the nearly constant strain rate measured between 6 and 8.5 MPa.

(67) The applied stress was then removed, and the sample was held in the creep frame for 48 h at 400 C. to allow for a full recovery of the dislocation microstructure. Creep testing of the sample, by then at 400 C. for 373 h (15.5 days), and labeled in the following as over-aged, was then resumed, beginning at a stress of about 6 MPa and lasting 672 hours (28 days), most of it spent below 13 MPa. The results of this series of tests on the over-aged sample are displayed in FIG. 10, and compared to those obtained for the peak-aged alloy. For all measured stresses, the creep rates of the over-aged Er-containing alloy (Alloy 5) are lower than in the peak-aged condition, in some cases by about three orders of magnitude. In the dislocation creep regime at high stresses (14-18 MPa), an apparent stress exponent of 292 is again indicative of a threshold stress, which is determined to be 13.91.6 MPa. In the diffusional creep regime at low stresses (6-11 MPa), the apparent stress exponent is 2.50.2, and the threshold stress is 4.50.8 MPa. A transition region between diffusional and dislocation creep between 11 and 13 MPa is observed, which was not present in the peak-aged sample.

(68) The microstructure of the over-aged alloy after a total of 1045 h (43.5 days) in the creep frame at 400 C., is shown in FIG. 7G. There is evidence of void-formation at the grain boundaries, and of significant coarsening of the intragranular Al.sub.3Er precipitates as compared to the peak-aged state, FIG. 7B. The formation of voids may be due to tensile stresses developing perpendicular to the applied compressive load, resulting from slight barreling of the sample during compressive creep testing. It is likely that these voids formed after considerable strain had accumulated in the sample, and they may thus affect the last few creep data points measured at the highest stresses, resulting in higher than expected strain rates. The over-aged sample exhibits a microhardness of 43610 MPa, following 1075 h of creep at 400 C., which is, as anticipated, below the peak-aged value of 60620 MPa.

(69) The grains are slightly larger in the Er-containing alloy (Alloy 5) that was exposed for 1045 h at 400 C., with a larger grain density of 3.10.2 grains mm.sup.2, as compared to the 3.60.2 grains mm.sup.2 from the Er-containing sample exposed for 123 h. 3-D LEAP tomographic analysis of the crept material revealed a number density of precipitates of 2110.sup.21 m.sup.3, where the high degree of error is because only five precipitates were detected in a 50 million atom dataset, all of which were only partially bound by the tip volume. Given the poor precipitate statistics, detailed compositional and structural analyses were not possible, though the precipitate radius was estimated by eye from the 3-D LEAP tomographic reconstruction to be 5-10 nm. Assuming that the volume fraction of precipitates is constant for the peak-aged and overaged sample, and using the measured number density of 2110.sup.21 m.sup.3, a radius of 6-9 nm is calculated for the spheroidal precipitates, in good agreement with the above estimate.

(70) Accordingly, the disclosed aluminum alloys having additions of scandium, zirconium, erbium and, optionally, silicon, exhibit good mechanical strength and creep resistance at elevated temperatures.

(71) Although various aspects of the disclosed aluminum alloy and method have been shown and described, modifications may occur to those skilled in the art upon reading the specification. The present application includes such modifications and is limited only by the scope of the claims.