Age-Induced Precipitation and Hardening Behavior of Ni 3 Al Intermetallic Alloys Containing Vanadium

L12-type Ni3Al alloys containing vanadium are potential candidates for solid-solution and age-hardenable alloy systems, according to the Ni3Al-Ni3V pseudo-binary phase diagram. Therefore, herein, variations in the microstructure and mechanical properties of Ni-13 at.% Al-12 at.% V-50 ppm B alloy during isothermal aging were investigated. Alloy specimens were solution-treated at 1323 K for 48 h, quenched in water, and aged at 1073 K to 1173 K. The quenched specimens exhibited a single phase of Ni3Al (L12 structure derived from Al (fcc) structure), while in the aged specimens, numerous fine disk-shaped precipitates identified as Ni3V (D022 structure from orthorhombic structure) were formed on {001} planes of the Ni3Al matrix. The size of the disk-shaped Ni3V precipitates increased gradually with increasing aging period. The hardness and strength of the specimens increased initially during aging at 1073 K to 1173 K, reached a maximum, followed by a subsequent decrease. The age-hardening behavior observed for the specimens can be explained in terms of precipitation of the fine disk-shaped Ni3V precipitates in the Ni3Al matrix. Furthermore, the peak-aged specimens exhibited an increase in yield strength with increasing testing temperature, similar to other L12-type intermetallic alloys.


Introduction
Ni 3 Al-based alloys offer several advantages such as high melting point, excellent mechanical properties at high temperatures, high thermal conductivity, good oxidization resistance, and low density, which are prerequisites for the vehicle components used in high-temperature applications such as engines and turbine blades.With the development of novel high-temperature materials, Ni 3 Al intermetallic compounds with the L1 2 structure derived from the A1 (fcc) structure play an important role as the primary phase for single-phase Ni 3 Al-based alloys, [1][2][3][4][5][6][7][8][9][10][11] Ni 3 Al/Ni 3 V for dual two-phase alloys [12][13][14][15], and the dispersive strengthening phase for Ni-based superalloys [16][17][18][19][20].However, Ni 3 Al-based alloys suffer from poor ductility at ambient temperature and therefore, fundamental and practical efforts to improve not only the strength at low and high temperatures but also the intrinsic deformability of the Ni 3 Al phase have been of significant interest.
In order to provide an excellent combination of mechanical properties of Ni 3 Al-based alloys, the Ni 3 Al phase should be modified by alloying and microstructural optimization.For example, a drastic increase in the ductility of Ni 3 Al obtained by boron doping has been reported [21][22][23][24].Further, there have been numerous investigations on Ni 3 Al-based alloys strengthened by solid-solution hardening via the elemental addition of Ti, Cr, Nb, and Ta [25][26][27].There have also been studies on alloys strengthened by the dispersion of soft precipitates of a Ni solid solution [28,29], and hard particles such as ceramics and oxides [30][31][32].However, to meet the current requirement of high-performance components, further enhancement of the mechanical, and physical properties of Ni 3 Al-based alloys is crucial.
Alloys based on the Ni 3 Al-Ni 3 V pseudo-binary alloy system should be potential solid-solution and aged-hardenable alloys, because vanadium dissolves well in the L1 2 Ni 3 Al phase and largely supersaturates at a eutectoid temperature, as shown in Figure 1 [33].It is also expected that age-induced hardening due to the precipitation of hard Ni 3 V with the D0 22 structure can improve the strength without significant degradation of other properties, such as thermal conductivity and environmental resistance [34,35].The mechanical properties of Ni 3 Al alloys were improved by adding 1-2 at.% of vanadium [26,27]; however, there is no study on the mechanical properties of Ni 3 Al alloys containing a large amount of vanadium.Further, there have only been few reported attempts to demonstrate the age-hardening behavior in the Ni 3 Al-Ni 3 V system.Hence, in this study, we strengthened Ni 3 Al alloys through the fine dispersion of age-induced precipitates, as well as through the dissolution of solutes in the matrix, by which the dislocation motion was hindered even at high temperatures.We then examined variations in the microstructure and mechanical properties during isothermal aging for a Ni 3 Al alloy containing a large amount of vanadium.
Metals 2019, 9, x FOR PEER REVIEW 2 of 11 been studies on alloys strengthened by the dispersion of soft precipitates of a Ni solid solution [28,29], and hard particles such as ceramics and oxides [30][31][32].However, to meet the current requirement of high-performance components, further enhancement of the mechanical, and physical properties of Ni3Al-based alloys is crucial.Alloys based on the Ni3Al-Ni3V pseudo-binary alloy system should be potential solid-solution and aged-hardenable alloys, because vanadium dissolves well in the L12 Ni3Al phase and largely supersaturates at a eutectoid temperature, as shown in Figure 1 [33].It is also expected that age-induced hardening due to the precipitation of hard Ni3V with the D022 structure can improve the strength without significant degradation of other properties, such as thermal conductivity and environmental resistance [34,35].The mechanical properties of Ni3Al alloys were improved by adding 1-2 at.% of vanadium [26,27]; however, there is no study on the mechanical properties of Ni3Al alloys containing a large amount of vanadium.Further, there have only been few reported attempts to demonstrate the age-hardening behavior in the Ni3Al-Ni3V system.Hence, in this study, we strengthened Ni3Al alloys through the fine dispersion of age-induced precipitates, as well as through the dissolution of solutes in the matrix, by which the dislocation motion was hindered even at high temperatures.We then examined variations in the microstructure and mechanical properties during isothermal aging for a Ni3Al alloy containing a large amount of vanadium.

Materials and Methods
Based on the phase diagram of the Ni3Al-Ni3V pseudo-binary alloy system (Figure 1; [33]), single-phase L12 Ni3Al alloys with the nominal composition of Ni-25 at.%Al-50 wt.ppm B and Ni-13 at.%Al-12 at.%V-50 wt.ppm B were prepared.Here, a small amount (50 wt.ppm) of B was doped on both alloys in order to improve the ductility of the parent Ni3Al phase through the suppression of intergranular fracture [21][22][23][24].Hereafter, these alloys are denoted as Ni3Al-0V and Ni3Al-12V alloys, respectively.Alloy button ingots with a diameter of 30 mm or 50 mm were fabricated by arc-melting in argon atmosphere using a non-consumable tungsten electrode on a copper hearth.Pure nickel grains (99.99 wt.% in purity), aluminum tips (99.9 wt.%), vanadium grains (99.9 wt.%), and boron powder (99.9 wt.%) were used as raw materials.The button ingots were melted repeatedly at least four times to ensure complete chemical homogeneity, and were then cut into sample pieces using an electro-discharge machine.After the contaminated layer was removed from the specimen surface by mechanical grinding, the pieces were encapsulated in a quartz tube, in which the pressure was maintained to be less than 2.0 × 10 −3 Pa.Subsequently, the encapsulated specimens were heat-treated at 1323 K for 48 h and quenched in water, to obtain single-phase L12 Ni3Al with and without

Materials and Methods
Based on the phase diagram of the Ni 3 Al-Ni 3 V pseudo-binary alloy system (Figure 1; [33]), single-phase L1 2 Ni 3 Al alloys with the nominal composition of Ni-25 at.%Al-50 wt.ppm B and Ni-13 at.%Al-12 at.%V-50 wt.ppm B were prepared.Here, a small amount (50 wt.ppm) of B was doped on both alloys in order to improve the ductility of the parent Ni 3 Al phase through the suppression of intergranular fracture [21][22][23][24].Hereafter, these alloys are denoted as Ni 3 Al-0V and Ni 3 Al-12V alloys, respectively.Alloy button ingots with a diameter of 30 mm or 50 mm were fabricated by arc-melting in argon atmosphere using a non-consumable tungsten electrode on a copper hearth.Pure nickel grains (99.99 wt.% in purity), aluminum tips (99.9 wt.%), vanadium grains (99.9 wt.%), and boron powder (99.9 wt.%) were used as raw materials.The button ingots were melted repeatedly at least four times to ensure complete chemical homogeneity, and were then cut into sample pieces using an electro-discharge machine.After the contaminated layer was removed from the specimen surface by mechanical grinding, the pieces were encapsulated in a quartz tube, in which the pressure was maintained to be less than 2.0 × 10 −3 Pa.Subsequently, the encapsulated specimens were heat-treated at 1323 K for 48 h and quenched in water, to obtain single-phase L1 2 Ni 3 Al with and without vanadium.Some Ni 3 Al-12V alloy specimens were encapsulated again under a vacuum of less than 2.0 × 10 −3 Pa in a quartz tube, and then isothermally aged at 1073 K, 1123 K, and 1173 K for various durations from 1 h to 240 h, followed by quenching in water.
The microstructures of the specimens were investigated by conventional field-emission scanning electron microscopy (FESEM) using a JSM-7001F (JOEL Inc., Tokyo, Japan)" and transmission electron microscopy (TEM) using a JEM-2000FX (JOEL Inc., Tokyo, Japan) operated at an accelerating voltage of 200 kV.Specimens for TEM observations were first mechanically abraded to less than 30 µm in thickness using SiC paper, and then polished by low-angle ion milling using a PIPS (Gatan Inc., Tokyo, Japan) with an argon ion beam accelerated at 3.5 kV.Hardness measurements were carried out using a Vickers microhardness tester HM-101 (Mitsutoyo Inc., Tokyo, Japan) under an applied load of 4.9 N for a dwell time of 10 s.The average hardness was obtained from more than 10 indentations.Tensile tests were performed using a universal tensile testing machine TENSIRON UTA-10 KN (A&D Co. Ltd., Tokyo, Japan) at temperatures ranging between room temperature (298 K) and 1073 K with an initial strain rate of 1.67 × 10 −4 s −1 under a vacuum of less than 2.0 × 10 −3 Pa.For the tensile tests, dog-bone-shaped specimens with a cross-section of 2 mm × 1.5 mm and gauge length of 10 mm were used.

Microstructure
Figure 2 shows typical FESEM and bright-field (BF) images and a corresponding selected-area electron diffraction (SAED) pattern of the Ni 3 Al-12V alloy specimen quenched from 1323 K.The specimens obtained through casting, annealing, and then quenching comprise relatively coarse grains with sizes ranging from 100 µm to 1 mm, as observed in Figure 2a.There are no contrasts from second phases in the matrix phase (Figure 2a,b), although few strain contrasts from dislocations are observed in Figure 2b.The SAED pattern (Figure 2c), which was taken from the matrix phase, indicates that all the fundamental and superlattice reflections can be indexed to the L1 2 structure of the Ni 3 Al phase (lattice parameter, a = 0.356 nm), as illustrated schematically in Figure 2d.Thus, we confirmed that the Ni 3 Al-12V alloy specimen quenched from 1323 K comprised a single phase of L1 2 Ni 3 Al containing vanadium element.This indicates that the solubility limit of vanadium for the L1 2 Ni 3 Al phase at 1323 K (Figure 1) should be more than 12 at.%. vanadium.Some Ni3Al-12V alloy specimens were encapsulated again under a vacuum of less than 2.0 × 10 −3 Pa in a quartz tube, and then isothermally aged at 1073 K, 1123 K, and 1173 K for various durations from 1 h to 240 h, followed by quenching in water.
The microstructures of the specimens were investigated by conventional field-emission scanning electron microscopy (FESEM) using a JSM-7001F (JOEL Inc., Tokyo, Japan),, and transmission electron microscopy (TEM) using a JEM-2000FX (JOEL Inc., Tokyo, Japan) operated at an accelerating voltage of 200 kV.Specimens for TEM observations were first mechanically abraded to less than 30 μm in thickness using SiC paper, and then polished by low-angle ion milling using a PIPS (Gatan Inc., Tokyo, Japan) with an argon ion beam accelerated at 3.5 kV.Hardness measurements were carried out using a Vickers microhardness tester HM-101 (Mitsutoyo Inc., Tokyo, Japan) under an applied load of 4.9 N for a dwell time of 10 s.The average hardness was obtained from more than 10 indentations.Tensile tests were performed using a universal tensile testing machine TENSIRON UTA-10 KN (A&D Co. Ltd., Tokyo, Japan) at temperatures ranging between room temperature (298 K) and 1073 K with an initial strain rate of 1.67 × 10 −4 s −1 under a vacuum of less than 2.0 × 10 −3 Pa.For the tensile tests, dog-bone-shaped specimens with a cross-section of 2 mm × 1.5 mm and gauge length of 10 mm were used.

Microstructure
Figure 2 shows typical FESEM and bright-field (BF) images and a corresponding selected-area electron diffraction (SAED) pattern of the Ni3Al-12V alloy specimen quenched from 1323 K.The specimens obtained through casting, annealing, and then quenching comprise relatively coarse grains with sizes ranging from 100 μm to 1 mm, as observed in Figure 2a.There are no contrasts from second phases in the matrix phase (Figure 2a,b), although few strain contrasts from dislocations are observed in Figure 2b.The SAED pattern (Figure 2c), which was taken from the matrix phase, indicates that all the fundamental and superlattice reflections can be indexed to the L12 structure of the Ni3Al phase (lattice parameter, a = 0.356 nm), as illustrated schematically in Figure 2d.Thus, we confirmed that the Ni3Al-12V alloy specimen quenched from 1323 K comprised a single phase of L12 Ni3Al containing vanadium element.This indicates that the solubility limit of vanadium for the L12 Ni3Al phase at 1323 K (Figure 1) should be more than 12 at.%.  Figure 3a shows a BF image of the Ni 3 Al-12V alloy specimen isothermally aged at 1123 K for 4 h.The incident beam direction is close to the [001] m zone axis (the subscript "m" denotes the L1 2 Ni 3 Al matrix phase).We can see dark contrasts, with a length of approximately 1 µm, extended preferentially along the [100] m and [010] m directions, as indicated by the arrows in Figure 3a. Figure 3b shows a SAED pattern taken from the extended dark contrasts (dotted circle in Figure 3a).The SAED pattern was identified by a superposition of reflections from the L1 2 Ni 3 Al matrix shown in Figure 2c, and the Ni 3 V with a D0 22 structure from the orthorhombic structure with lattice parameters a = 0.355 nm and c = 0.719 nm, as marked by the diamond shapes in Figure 3c.Thus, the precipitates introduced by aging at 1123 K were identified as D0 22 Ni 3 V, which was predicted from the phase diagram of Figure 3a shows a BF image of the Ni3Al-12V alloy specimen isothermally aged at 1123 K for 4 h.The incident beam direction is close to the [001]m zone axis (the subscript "m" denotes the L12 Ni3Al matrix phase).We can see dark contrasts, with a length of approximately 1 μm, extended preferentially along the [100]m and [010]m directions, as indicated by the arrows in Figure 3a. Figure 3b shows a SAED pattern taken from the extended dark contrasts (dotted circle in Figure 3a).The SAED pattern was identified by a superposition of reflections from the L12 Ni3Al matrix shown in Figure 2c, and the Ni3V with a D022 structure from the orthorhombic structure with lattice parameters a = 0.355 nm and c = 0.719 nm, as marked by the diamond shapes in Figure 3c.Thus, the precipitates introduced by aging at 1123 K were identified as D022 Ni3V, which was predicted from the phase diagram of Figure 1.The orientation relationship between the Ni3Al matrix and Ni3V precipitates was [001]m//[001]p, (100)m//(100)p (the subscript "p" denotes the precipitate phase of Ni3V with a D022 structure).Figure 4 shows the BF images of the Ni3Al-12V alloy specimen isothermally aged at 1123 K for 16 h and 48 h.The incident beam direction is close to the [001]m zone axis.We can see dark contrasts from Ni3V precipitates, which extend preferentially along the [100]m and [010]m directions, similar to those in Figure 3a.There are two shapes of contrasts of the Ni3V precipitates; i.e., extended elliptical (indicated by arrows in Figure 4b) and almost circular (dotted circle) ones.This suggests that the D022 Ni3V precipitates in the L12 Ni3Al matrix should be dispersed preferentially along (001)m with a disk-like shape (Figure 5).The size and aspect ratio of the Ni3V precipitates appear to increase with aging time.The diameter is approximately 1.5 μm for 16 h of aging and 2 μm for 48 h of aging; both exhibit a similar thickness of 0.5 μm.Here, the misfit strains along the a-and c-axes on the matrix plane with the precipitates are estimated to be approximately 0.3% and 0.8%, respectively, because the lattice parameters of L12 Ni3Al and D022 Ni3V in the two-phase region have been reported as a = 0.356 nm, and a = 0.355 nm, and c = 0.719 nm respectively [36].Therefore, the disk-like two-dimensional growth of D022 Ni3V precipitates, which have coherent broad faces on the (001)m planes containing the a-and b-axes of the matrix, should be caused by the minimization of the interface energy between the L12 Ni3Al matrix and the D022 Ni3V precipitates [37].4b) and almost circular (dotted circle) ones.This suggests that the D0 22 Ni 3 V precipitates in the L1 2 Ni 3 Al matrix should be dispersed preferentially along (001) m with a disk-like shape (Figure 5).The size and aspect ratio of the Ni 3 V precipitates appear to increase with aging time.The diameter is approximately 1.5 µm for 16 h of aging and 2 µm for 48 h of aging; both exhibit a similar thickness of 0.5 µm.Here, the misfit strains along the aand c-axes on the matrix plane with the precipitates are estimated to be approximately 0.3% and 0.8%, respectively, because the lattice parameters of L1 2 Ni 3 Al and D0 22 Ni 3 V in the two-phase region have been reported as a = 0.356 nm, and a = 0.355 nm, and c = 0.719 nm respectively [36].Therefore, the disk-like two-dimensional growth of D0 22 Ni 3 V precipitates, which have coherent broad faces on the (001) m planes containing the aand b-axes of the matrix, should be caused by the minimization of the interface energy between the L1 2 Ni 3 Al matrix and the D0 22 Ni 3 V precipitates [37].

Mechanical Properties
Figure 6 shows the variation in the Vickers hardness of the Ni3Al-12V alloy specimens during isothermal aging at 1073 K, 1123 K, and 1173 K.The hardness for all three aging temperatures increases initially, reaches a maximum value, and then decreases for the specimens that underwent prolonged aging.The value of the maximum hardness decreases with increasing aging temperature, while the period to reach the maximum hardness is shortened.
The age-hardening behavior might be common with some conventional age-hardenable Al, Cu, and Fe alloy systems [38][39][40][41], and can be, in this case, explained by the precipitation of D022 Ni3V.In the initial stage of aging, the hardness steadily increased with the nucleation and growth of D022 Ni3V precipitates in the L12 Ni3Al matrix phase.The maximum hardness is associated with the maximum number of D022 Ni3V precipitates.This implies that the continuous nucleation of D022 Ni3V precipitates is saturated, due to the depletion of super-saturated solid-solute vanadium atoms in the matrix.The hardness decreases with further aging due to an increase in the distance between the D022 Ni3V precipitates, which might be because the precipitates undergo Ostwald ripening, like other age-hardenable alloys [38,39,41].

Mechanical Properties
Figure 6 shows the variation in the Vickers hardness of the Ni3Al-12V alloy specimens during isothermal aging at 1073 K, 1123 K, and 1173 K.The hardness for all three aging temperatures increases initially, reaches a maximum value, and then decreases for the specimens that underwent prolonged aging.The value of the maximum hardness decreases with increasing aging temperature, while the period to reach the maximum hardness is shortened.
The age-hardening behavior might be common with some conventional age-hardenable Al, Cu, and Fe alloy systems [38][39][40][41], and can be, in this case, explained by the precipitation of D022 Ni3V.In the initial stage of aging, the hardness steadily increased with the nucleation and growth of D022 Ni3V precipitates in the L12 Ni3Al matrix phase.The maximum hardness is associated with the maximum number of D022 Ni3V precipitates.This implies that the continuous nucleation of D022 Ni3V precipitates is saturated, due to the depletion of super-saturated solid-solute vanadium atoms in the matrix.The hardness decreases with further aging due to an increase in the distance between the D022 Ni3V precipitates, which might be because the precipitates undergo Ostwald ripening, like other age-hardenable alloys [38,39,41].

Mechanical Properties
Figure 6 shows the variation in the Vickers hardness of the Ni 3 Al-12V alloy specimens during isothermal aging at 1073 K, 1123 K, and 1173 K.The hardness for all three aging temperatures increases initially, reaches a maximum value, and then decreases for the specimens that underwent prolonged aging.The value of the maximum hardness decreases with increasing aging temperature, while the period to reach the maximum hardness is shortened.
The age-hardening behavior might be common with some conventional age-hardenable Al, Cu, and Fe alloy systems [38][39][40][41], and can be, in this case, explained by the precipitation of D0 22 Ni 3 V.In the initial stage of aging, the hardness steadily increased with the nucleation and growth of D0 22 Ni 3 V precipitates in the L1 2 Ni 3 Al matrix phase.The maximum hardness is associated with the maximum number of D0 22 Ni 3 V precipitates.This implies that the continuous nucleation of D0 22 Ni 3 V precipitates is saturated, due to the depletion of super-saturated solid-solute vanadium atoms in the matrix.The hardness decreases with further aging due to an increase in the distance between the D0 22 Ni 3 V precipitates, which might be because the precipitates undergo Ostwald ripening, like other age-hardenable alloys [38,39,41].Figure 7 shows the typical nominal stress-strain curves at 298 K for the Ni3Al-12V alloy specimens quenched and peak-aged at 1123 K for 16 h, together with the quenched Ni3Al-0V alloy specimen for comparison.The quenched Ni3Al-0V alloy yields with mild plastic elongation, which is consistent with previous reports [2,7,10].On the other hand, both, the quenched and aged Ni3Al-12V alloy specimens show a greater yield strength than the quenched Ni3Al-0V alloy specimen, followed by extensive work hardening accompanied by a larger fracture strain.The enhancement in yield strength for both specimens should be attributed primarily to solid-solution hardening by the vanadium atom dissolved in the L12 Ni3Al phase.Mishima et al. reported an increased rate of change in solid-solution hardening per atomic percent, (dσ/dc), for vanadium [27].However, the detailed influence of a large amount of vanadium dissolving in the Ni3Al phase on the mechanical properties is still unclear.We are systematically investigating this and will report it in the future.The yield strength of the peak-aged Ni3Al-12V alloy specimen seemed to be slightly greater by 15 MPa than that of the quenched Ni3Al-12V alloy specimen.This must be due to the age-induced precipitates of D022 Ni3V in the L12 Ni3Al matrix phase.Nevertheless, yield strengthening by the age-induced precipitates should not be as significant as strengthening by the solid-solution effect of vanadium atoms.Figure 7 shows the typical nominal stress-strain curves at 298 K for the Ni3Al-12V alloy specimens quenched and peak-aged at 1123 K for 16 h, together with the quenched Ni3Al-0V alloy specimen for comparison.The quenched Ni3Al-0V alloy yields with mild plastic elongation, which is consistent with previous reports [2,7,10].On the other hand, both, the quenched and aged Ni3Al-12V alloy specimens show a greater yield strength than the quenched Ni3Al-0V alloy specimen, followed by extensive work hardening accompanied by a larger fracture strain.The enhancement in yield strength for both specimens should be attributed primarily to solid-solution hardening by the vanadium atom dissolved in the L12 Ni3Al phase.Mishima et al. reported an increased rate of change in solid-solution hardening per atomic percent, (dσ/dc), for vanadium [27].However, the detailed influence of a large amount of vanadium dissolving in the Ni3Al phase on the mechanical properties is still unclear.We are systematically investigating this and will report it in the future.The yield strength of the peak-aged Ni3Al-12V alloy specimen seemed to be slightly greater by 15 MPa than that of the quenched Ni3Al-12V alloy specimen.This must be due to the age-induced precipitates of D022 Ni3V in the L12 Ni3Al matrix phase.Nevertheless, yield strengthening by the age-induced precipitates should not be as significant as strengthening by the solid-solution effect of vanadium atoms.8 shows the typical nominal stress-strain curves at elevated temperatures from 298 K to 1073 K for the Ni 3 Al-12V alloy specimens peak-aged at 1123 K for 16 h.The 0.2% yield strength is shown by arrows in Figure 8. Figure 9 summarizes the yield strength (0.2% proof stress), ultimate tensile strength, and fractural strain for the Ni 3 Al-12V alloy specimens quenched and peak-aged at 1123 K for 16 h as a function of the testing temperature.In comparison to the stoichiometric Ni-25 at.%Al binary alloy [10], the yield strengths of the quenched and peak-aged Ni 3 Al-12V alloy specimens at 298 K are higher, as described above.Notably, the yield strength of both specimens steadily increases with the testing temperature to 1073 K, indicating an anomalous temperature dependency of yield strength, similar to other L1 2 intermetallic compounds [1,2,42], while that of the stoichiometric Ni 3 Al binary alloy specimen shows a peak at around 800 K.The peak temperature at which the maximum flow strength is observed has been suggested to be associated with the transition of the slip system with temperature [3,43].Accordingly, the result observed in Figure 9a indicates that the transition temperature of the slip system from <110>{111} to <110>{100} is raised due to the effect of the vanadium atom dissolving in the Ni 3 Al matrix phase.
Metals 2019, 9, x FOR PEER REVIEW 7 of 11 Figure 8 shows the typical nominal stress-strain curves at elevated temperatures from 298 K to 1073 K for the Ni3Al-12V alloy specimens peak-aged at 1123 K for 16 h.The 0.2% yield strength is shown by arrows in Figure 8. Figure 9 summarizes the yield strength (0.2% proof stress), ultimate tensile strength, and fractural strain for the Ni3Al-12V alloy specimens quenched and peak-aged at 1123 K for 16 h as a function of the testing temperature.In comparison to the stoichiometric Ni-25 at.%Al binary alloy [10], the yield strengths of the quenched and peak-aged Ni3Al-12V alloy specimens at 298 K are higher, as described above.Notably, the yield strength of both specimens steadily increases with the testing temperature to 1073 K, indicating an anomalous temperature dependency of yield strength, similar to other L12 intermetallic compounds [1,2,42], while that of the stoichiometric Ni3Al binary alloy specimen shows a peak at around 800 K.The peak temperature at which the maximum flow strength is observed has been suggested to be associated with the transition of the slip system with temperature [3,43].Accordingly, the result observed in Figure 9a indicates that the transition temperature of the slip system from <110>{111} to <110>{100} is raised due to the effect of the vanadium atom dissolving in the Ni3Al matrix phase.Figure 8 shows the typical nominal stress-strain curves at elevated temperatures from 298 K to 1073 K for the Ni3Al-12V alloy specimens peak-aged at 1123 K for 16 h.The 0.2% yield strength is shown by arrows in Figure 8. Figure 9 summarizes the yield strength (0.2% proof stress), ultimate tensile strength, and fractural strain for the Ni3Al-12V alloy specimens quenched and peak-aged at 1123 K for 16 h as a function of the testing temperature.In comparison to the stoichiometric Ni-25 at.%Al binary alloy [10], the yield strengths of the quenched and peak-aged Ni3Al-12V alloy specimens at 298 K are higher, as described above.Notably, the yield strength of both specimens steadily increases with the testing temperature to 1073 K, indicating an anomalous temperature dependency of yield strength, similar to other L12 intermetallic compounds [1,2,42], while that of the stoichiometric Ni3Al binary alloy specimen shows a peak at around 800 K.The peak temperature at which the maximum flow strength is observed has been suggested to be associated with the transition of the slip system with temperature [3,43].Accordingly, the result observed in Figure 9a indicates that the transition temperature of the slip system from <110>{111} to <110>{100} is raised due to the effect of the vanadium atom dissolving in the Ni3Al matrix phase.The tensile strength and fracture stain for the quenched and peak-aged Ni 3 Al-12V alloy specimens were maximum around 673 K, and then decreased monotonically with increasing testing temperature.However, the fracture strain at the highest testing temperature of 1073 K was still above 8% for the peak-aged Ni 3 Al-12V alloy specimens, which is a significant improvement compared with the quenched Ni 3 Al-12V alloy.Figure 10 shows the FESEM fractographs of the peak-aged Ni 3 Al-12V alloy specimen tensile-tested at 298 K and 973 K, together with that of the quenched Ni 3 Al-0V tensile-tested at 298 K for comparison.For the former, most areas of the fracture surfaces show a ductile-like dimple pattern (Figure 10a,b).On the other hand, the latter shows a brittle grain boundary facet pattern (Figure 10c), as has been observed in previous studies [2,21,22].The enhancement in the tensile strength as well as the fracture strain might be attributed to the enhanced grain boundary strength by the V solutes largely dissolving in the Ni 3 Al phase via suppressing intergranular fracture.A similar phenomenon has been observed in Ni 3 Al alloys containing a large amount of Mn [44,45] and Fe [46], and Ni 3 Si containing a large amount of Ti [47].Moreover, the decrease in both, the tensile strength and fracture strain with increasing testing temperature for the peak-aged Ni 3 Al-12V alloy specimen, may be associated with the intrinsic temperature dependence of the Ni 3 Al phase matrix, e.g., the transition of the activated slip system [3,43].The tensile strength and fracture stain for the quenched and peak-aged Ni3Al-12V alloy specimens were maximum around 673 K, and then decreased monotonically with increasing testing temperature.However, the fracture strain at the highest testing temperature of 1073 K was still above 8% for the peak-aged Ni3Al-12V alloy specimens, which is a significant improvement compared with the quenched Ni3Al-12V alloy.Figure 10 shows the FESEM fractographs of the peak-aged Ni3Al-12V alloy specimen tensile-tested at 298 K and 973 K, together with that of the quenched Ni3Al-0V tensile-tested at 298 K for comparison.For the former, most areas of the fracture surfaces show a ductile-like dimple pattern (Figure 10a,b).On the other hand, the latter shows a brittle grain boundary facet pattern (Figure 10c), as has been observed in previous studies [2,21,22].The enhancement in the tensile strength as well as the fracture strain might be attributed to the enhanced grain boundary strength by the V solutes largely dissolving in the Ni3Al phase via suppressing intergranular fracture.A similar phenomenon has been observed in Ni3Al alloys containing a large amount of Mn [44,45] and Fe [46], and Ni3Si containing a large amount of Ti [47].Moreover, the decrease in both, the tensile strength and fracture strain with increasing testing temperature for the peak-aged Ni3Al-12V alloy specimen, may be associated with the intrinsic temperature dependence of the Ni3Al phase matrix, e.g., the transition of the activated slip system [3,43].When comparing the tensile data for the quenched and peak-aged Ni3Al-12V alloy specimens, the yield strength, ultimate tensile strength, and fracture strain of the aged specimen tended to be superior to those of the quenched specimen, rendering them favorable for application as high-temperature materials.The enhancement in the yield strength and ultimate tensile strength can be explained by age-induced hardening by the D022 Ni3V precipitates.However, the reason why both the ultimate-tensile strength and fracture strain of the Ni3Al-12V alloy specimens were improved by aging remains unclear at this stage and is scope for future work.

Conclusions
In this study, we investigated the variations in the microstructure and mechanical properties during aging for L12 Ni3Al alloy containing a large amount of vanadium (nominal composition of Ni-13 at.%Al-12 at.%V-50 wt.ppm B), which was solution-treated at 1323 K for 48 h in vacuum and then quenched in water.The specimens quenched after solution treatment exhibited a single phase of L12 Ni3Al.For the specimens aged at 1123 K after 4 h, a number of fine disk-shaped precipitates were found on {001} planes of the L12 Ni3Al matrix.The disk-shaped precipitates were identified as D022 Ni3V.The size and aspect ratio of the disk-shaped Ni3V precipitates increased gradually with increasing aging time.The Vickers hardness of the quenched specimen was approximately 280 Hv.During aging at 1073 K to 1173 K, the hardness of the specimens increased and reached a maximum, then decreased.The age-hardening behavior observed for the Ni-13 at.%Al-12 at.%V alloy can be explained in terms of precipitation of the fine disk-shaped Ni3V phase in the Ni3Al matrix.The mechanical properties, such as yield strength, ultimate tensile strength, and fracture strain of the When comparing the tensile data for the quenched and peak-aged Ni 3 Al-12V alloy specimens, the yield strength, ultimate tensile strength, and fracture strain of the aged specimen tended to be superior to those of the quenched specimen, rendering them favorable for application as high-temperature materials.The enhancement in the yield strength and ultimate tensile strength can be explained by age-induced hardening by the D0 22 Ni 3 V precipitates.However, the reason why both the ultimate-tensile strength and fracture strain of the Ni 3 Al-12V alloy specimens were improved by aging remains unclear at this stage and is scope for future work.

Conclusions
In this study, we investigated the variations in the microstructure and mechanical properties during aging for L1 2 Ni 3 Al alloy containing a large amount of vanadium (nominal composition of Ni-13 at.%Al-12 at.%V-50 wt.ppm B), which was solution-treated at 1323 K for 48 h in vacuum and then quenched in water.The specimens quenched after solution treatment exhibited a single phase of L1 2 Ni 3 Al.For the specimens aged at 1123 K after 4 h, a number of fine disk-shaped precipitates were found on {001} planes of the L1 2 Ni 3 Al matrix.The disk-shaped precipitates were identified as D0 22 Ni 3 V.The size and aspect ratio of the disk-shaped Ni 3 V precipitates increased gradually with increasing aging time.The Vickers hardness of the quenched specimen was approximately 280 Hv.During aging at 1073 K to 1173 K, the hardness of the specimens increased and reached a maximum, then decreased.The age-hardening behavior observed for the Ni-13 at.%Al-12 at.%V alloy can be explained in terms of precipitation of the fine disk-shaped Ni 3 V phase in the Ni 3 Al matrix.The mechanical properties, such as yield strength, ultimate tensile strength, and fracture strain of the

Figure 2 .
Figure 2. (a) Field-emission scanning electron microscopy (FESEM) image and (b) bright-field (BF) transmission electron microscope (TEM) image of the Ni3Al-12V alloy quenched from 1323 K. (c) A selected-area electron diffraction (SAED) pattern taken from the matrix.(d) Schematic illustration of the SAED pattern, viewed from the [001] zone axis of the L12 structure.The fundamental and superlattice reflections from the L12 structure are marked by open and filled circles, respectively.

Figure 2 .
Figure 2. (a) Field-emission scanning electron microscopy (FESEM) image and (b) bright-field (BF) transmission electron microscope (TEM) image of the Ni 3 Al-12V alloy quenched from 1323 K. (c) A selected-area electron diffraction (SAED) pattern taken from the matrix.(d) Schematic illustration of the SAED pattern, viewed from the [001] zone axis of the L1 2 structure.The fundamental and superlattice reflections from the L1 2 structure are marked by open and filled circles, respectively.
Figure3ashows a BF image of the Ni 3 Al-12V alloy specimen isothermally aged at 1123 K for 4 h.The incident beam direction is close to the [001] m zone axis (the subscript "m" denotes the L1 2 Ni 3 Al matrix phase).We can see dark contrasts, with a length of approximately 1 µm, extended preferentially along the [100] m and [010] m directions, as indicated by the arrows in Figure3a.Figure3bshows a SAED pattern taken from the extended dark contrasts (dotted circle in Figure3a).The SAED pattern was identified by a superposition of reflections from the L1 2 Ni 3 Al matrix shown in Figure2c, and the Ni 3 V with a D0 22 structure from the orthorhombic structure with lattice parameters a = 0.355 nm and c = 0.719 nm, as marked by the diamond shapes in Figure3c.Thus, the precipitates introduced by aging at 1123 K were identified as D0 22 Ni 3 V, which was predicted from the phase diagram of Figure1.The orientation relationship between the Ni 3 Al matrix and Ni 3 V precipitates was [001] m //[001] p , (100) m //(100) p (the subscript "p" denotes the precipitate phase of Ni 3 V with a D0 22 structure).

Figure 3 .
Figure 3. (a) Bright-field (BF)-TEM image of the Ni3Al-12V alloy aged at 1023 K for 4 h, viewed from the [001] zone axis of the matrix with a L12 structure.(b) SAED pattern taken from the area with a dark elliptical contrast as marked by the dotted circle in (a).(c) Schematic illustration of the SAED pattern viewed from the [100] zone axis for D022 structure [15].The fundamental and superlattice reflections from the D022 structure are marked by open and filled diamonds, respectively.

Figure 3 .
Figure 3. (a) Bright-field (BF)-TEM image of the Ni 3 Al-12V alloy aged at 1023 K for 4 h, viewed from the [001] zone axis of the matrix with a L1 2 structure.(b) SAED pattern taken from the area with a dark elliptical contrast as marked by the dotted circle in (a).(c) Schematic illustration of the SAED pattern viewed from the [100] zone axis for D0 22 structure [15].The fundamental and superlattice reflections from the D0 22 structure are marked by open and filled diamonds, respectively.

Figure 4
Figure4shows the BF images of the Ni 3 Al-12V alloy specimen isothermally aged at 1123 K for 16 h and 48 h.The incident beam direction is close to the [001] m zone axis.We can see dark contrasts from Ni 3 V precipitates, which extend preferentially along the [100] m and [010] m directions, similar to those in Figure3a.There are two shapes of contrasts of the Ni 3 V precipitates; i.e., extended elliptical (indicated by arrows in Figure4b) and almost circular (dotted circle) ones.This suggests that the D0 22 Ni 3 V precipitates in the L1 2 Ni 3 Al matrix should be dispersed preferentially along (001) m with a disk-like shape (Figure5).The size and aspect ratio of the Ni 3 V precipitates appear to increase with aging time.The diameter is approximately 1.5 µm for 16 h of aging and 2 µm for 48 h of aging; both exhibit a similar thickness of 0.5 µm.Here, the misfit strains along the aand c-axes on the matrix plane with the precipitates are estimated to be approximately 0.3% and 0.8%, respectively, because the lattice parameters of L1 2 Ni 3 Al and D0 22 Ni 3 V in the two-phase region have been reported as a = 0.356 nm, and a = 0.355 nm, and c = 0.719 nm respectively[36].Therefore, the disk-like two-dimensional growth of D0 22 Ni 3 V precipitates, which have coherent broad faces on the (001) m planes containing the aand b-axes of the matrix, should be caused by the minimization of the interface energy between the L1 2 Ni 3 Al matrix and the D0 22 Ni 3 V precipitates[37].

Figure 4 .
Figure 4. BF-TEM images of the Ni3Al-12V alloy aged at 1023 K for (a) 16 h (peak-hardened condition) and (b) 48 h, viewed from the [001] zone axis of the matrix with the L12 structure.

Figure 4 .
Figure 4. BF-TEM images of the Ni 3 Al-12V alloy aged at 1023 K for (a) 16 h (peak-hardened condition) and (b) 48 h, viewed from the [001] zone axis of the matrix with the L1 2 structure.

Figure 4 .
Figure 4. BF-TEM images of the Ni3Al-12V alloy aged at 1023 K for (a) 16 h (peak-hardened condition) and (b) 48 h, viewed from the [001] zone axis of the matrix with the L12 structure.

Figure 6 .
Figure 6.Variation in Vickers hardness with aging time for the Ni3Al-12V alloy specimens aged at 1073 K, 1123 K, and 1173 K.The measured hardness of the as-quenched specimen is 278 Hv.

Figure 6 .
Figure 6.Variation in Vickers hardness with aging time for the Ni 3 Al-12V alloy specimens aged at 1073 K, 1123 K, and 1173 K.The measured hardness of the as-quenched specimen is 278 Hv.

Figure 7
Figure7shows the typical nominal stress-strain curves at 298 K for the Ni 3 Al-12V alloy specimens quenched and peak-aged at 1123 K for 16 h, together with the quenched Ni 3 Al-0V alloy specimen for comparison.The quenched Ni 3 Al-0V alloy yields with mild plastic elongation, which is consistent with previous reports[2,7,10].On the other hand, both, the quenched and aged Ni 3 Al-12V alloy specimens show a greater yield strength than the quenched Ni 3 Al-0V alloy specimen, followed by extensive work hardening accompanied by a larger fracture strain.The enhancement in yield strength for both specimens should be attributed primarily to solid-solution hardening by the vanadium atom dissolved in the L1 2 Ni 3 Al phase.Mishima et al. reported an increased rate of change in solid-solution hardening per atomic percent, (dσ/dc), for vanadium[27].However, the detailed influence of a large amount of vanadium dissolving in the Ni 3 Al phase on the mechanical properties is still unclear.We are systematically investigating this and will report it in the future.The yield strength of the peak-aged Ni 3 Al-12V alloy specimen seemed to be slightly greater by 15 MPa than that of the quenched Ni 3 Al-12V alloy specimen.This must be due to the age-induced precipitates of D0 22 Ni 3 V in the L1 2 Ni 3 Al matrix phase.Nevertheless, yield strengthening by the age-induced precipitates should not be as significant as strengthening by the solid-solution effect of vanadium atoms.

Figure 6 .
Figure 6.Variation in Vickers hardness with aging time for the Ni3Al-12V alloy specimens aged at 1073 K, 1123 K, and 1173 K.The measured hardness of the as-quenched specimen is 278 Hv.

Figure
Figure8shows the typical nominal stress-strain curves at elevated temperatures from 298 K to 1073 K for the Ni 3 Al-12V alloy specimens peak-aged at 1123 K for 16 h.The 0.2% yield strength is shown by arrows in Figure8.Figure9summarizes the yield strength (0.2% proof stress), ultimate tensile strength, and fractural strain for the Ni 3 Al-12V alloy specimens quenched and peak-aged at 1123 K for 16 h as a function of the testing temperature.In comparison to the stoichiometric Ni-25 at.%Al binary alloy[10], the yield strengths of the quenched and peak-aged Ni 3 Al-12V alloy specimens at 298 K are higher, as described above.Notably, the yield strength of both specimens steadily increases with the testing temperature to 1073 K, indicating an anomalous temperature dependency of yield strength, similar to other L1 2 intermetallic compounds[1,2,42], while that of the stoichiometric Ni 3 Al binary alloy specimen shows a peak at around 800 K.The peak temperature at which the maximum flow strength is observed has been suggested to be associated with the transition of the slip system with temperature[3,43].Accordingly, the result observed in Figure9aindicates that the transition temperature of the slip system from <110>{111} to <110>{100} is raised due to the effect of the vanadium atom dissolving in the Ni 3 Al matrix phase.

Figure 8 .
Figure 8. Nominal stress versus nominal strain curves for the Ni3Al-12V alloy specimens peak-aged at 1123 K for 16 h, measured at temperatures ranging from 298 K to 1073 K.

Figure 9 .
Figure 9. Variations in (a) yield strength (0.2% proof stress), (b) ultimate tensile strength, and (c) fracture strain for the Ni3Al-12V alloy specimens quenched (Q.) and peak-aged (P.A.) at 1123 K for 16 h as a function of the testing temperature.

Figure 8 .
Figure 8. Nominal stress versus nominal strain curves for the Ni 3 Al-12V alloy specimens peak-aged at 1123 K for 16 h, measured at temperatures ranging from 298 K to 1073 K.

Figure 8 .
Figure 8. Nominal stress versus nominal strain curves for the Ni3Al-12V alloy specimens peak-aged at 1123 K for 16 h, measured at temperatures ranging from 298 K to 1073 K.

Figure 9 .
Figure 9. Variations in (a) yield strength (0.2% proof stress), (b) ultimate tensile strength, and (c) fracture strain for the Ni3Al-12V alloy specimens quenched (Q.) and peak-aged (P.A.) at 1123 K for 16 h as a function of the testing temperature.

Figure 9 .
Figure 9. Variations in (a) yield strength (0.2% proof stress), (b) ultimate tensile strength, and (c) fracture strain for the Ni 3 Al-12V alloy specimens quenched (Q.) and peak-aged (P.A.) at 1123 K for 16 h as a function of the testing temperature.

Figure 10 .
Figure 10.FESEM fractographs for the peak-aged Ni 3 Al-12V alloy specimens tensile-tested at (a) 298 K and (b) 973 K, together with (c) that for the Ni 3 Al-0V alloy specimens tensile-tested at 298 K.