Pulse Plasma Sintering of NiAl-Al2O3 Composite Powder Produced by Mechanical Alloying with Contribution of Nanometric Al2O3 Powder

NiAl-Al2O3 composites, fabricated from the prepared composite powders by mechanical alloying and then consolidated by pulse plasma sintering, were presented. The use of nanometric alumina powder for reinforcement of a synthetized intermetallic matrix was the innovative concept of this work. Moreover, this is the first reported attempt to use the Pulse Plasma Sintering (PPS) method to consolidate composite powder with the contribution of nanometric alumina powder. The composite powders consisting of the intermetallic phase NiAl and Al2O3 were prepared by mechanical alloying from powder mixtures containing Ni-50at.%Al with the contribution of 10 wt.% or 20 wt.% nanometric aluminum oxide. A nanocrystalline NiAl matrix was formed, with uniformly distributed Al2O3 inclusions as reinforcement. The PPS method successfully consolidated NiAl-Al2O3 composite powders with limited grain growth in the NiAl matrix. The appropriate sintering temperature for composite powder was selected based on analysis of the grain growth and hardness of Al2O3 subjected to PPS consolidation at various temperatures. As a result of these tests, sintering of the NiAl-Al2O3 powders was carried out at temperatures of 1200 °C, 1300 °C, and 1400 °C. The microstructure and properties of the initial powders, composite powders, and consolidated bulk composite materials were characterized by SEM, EDS, XRD, density, and hardness measurements. The hardness of the ultrafine-grained NiAl-Al2O3 composites obtained via PPS depends on the Al2O3 content in the composite, as well as the sintering temperature applied. The highest values of the hardness of the composites were obtained after sintering at the lowest temperature (1200 °C), reaching 7.2 ± 0.29 GPa and 8.4 ± 0.07 GPa for 10 wt.% Al2O3 and 20 wt.% Al2O3, respectively, and exceeding the hardness values reported in the literature. From a technological point of view, the possibility to use sintering temperatures as low as 1200 °C is crucial for the production of fully dense, ultrafine-grained composites with high hardness.


Introduction
Ceramics are combined with metals and intermetallic materials to produce composites. Various metals are used, such as V, Mo, Cu, Ni, Ti, or Fe [1][2][3][4][5][6][7], along with intermetallics such as NiAl [8,9] or Ni 3 Al [10]. Such materials have been developed because of their desirable properties. Improving the fracture toughness of the composites with the ceramic matrix is crucial [11,12]. Furthermore, composites of intermetallic-ceramic systems can offer excellent combinations of properties, e.g., low density and relatively high hardness and strength. The presence of nanocrystalline or ultrafine structure in composites may further enhance their desired properties [13,14] and their size was in the range of 100 ± 20 nm. In the experiments, nickel and aluminum were used as metallic components. Nickel powder (ABCR GmbH & Co.KG, Karlsruhe, Germany) was characterized by 3-7 µm average particle size. The aluminum powder (ABCR GmbH & Co.KG, Karlsruhe, Germany) was characterized by an average particle size of 44 µm. The powders used for grinding were characterized by purity at the level of 99.99%-aluminum oxide, 99.9%-nickel, and 99.7%-aluminum. The data above on the characteristics of the initial powders are based on the manufacturers' data. Nevertheless, these characteristics required confirmation through experimental investigation.

Preparation of Samples
The first stage of the work was mechanical alloying to obtain a composite powder based on nickel, aluminum, and aluminum oxide. The primary materials for the milling processes were powder blends containing Ni-50at.%Al with the addition of 10 wt.% and 20 wt.% aluminum oxide. The milling processes were performed on a SPEX 8000D highenergy shaker ball mill (SPEX ® SamplePrep, Metuchen, NJ, USA) with an 8:1 ball-topowder weight ratio. The milling processes and powder sampling were carried out under an argon atmosphere.
In the subsequent steps, samples were fabricated using the PPS technique. The PPS method is also identified as a field-assisted sintering technology (FAST) [23]-a process using the activation of an electric field generated by the flow of an external electric current during sintering [23]. During the PPS process, the high-current forcing pulses come from the discharge of the capacitor bank with a capacity of~200 µF. This forcing causes a rapid increase in the temperature of the sintered powder. This heating method, which is carried out with simultaneous pressing under pressure, allows the ignition temperature of the SHS reaction to be quickly reached, simultaneously, in the entire volume of the synthesized powders [24]. The heating of the powder occurs with Joule's heat, which is given off during the passage of current through the consolidated powder, and spark discharges (plasma generation) in the spaces between the powder grains. During the PPS process, four operations can be distinguished, including preparing the working atmosphere, setting the press pressure, sintering, and chamber discharge [24,25]. The first step in the PPS method involves the preparation of the working atmosphere, which involves achieving the required vacuum level (low vacuum 1 Pa, high vacuum 10 −3 Pa). Then, a pressure force starting from 20 MPa and ending at 80 MPa is applied to the punches in which the die is placed. In this way, the pressing pressure during the process is established. Current pulses with specific parameters flow through the powder in the sintering stage. The process parameters were developed based on previous experimental work. At the last stage of the PPS process, the chamber is discharged, i.e., atmospheric pressure is established within it. Table 1 presents the PPS process parameters for the samples produced. Samples containing pure alumina were sintered at six different temperatures: 1000 • C, 1100 • C, 1200 • C, 1300 • C, 1400 • C, and 1500 • C. These studies aimed to determine the appropriate sintering temperature for samples prepared from ultrafine-grained alumina powder. Subsequently, based on these tests, sintering of the NiAl-Al 2 O 3 powders was carried out at temperatures of 1200 • C, 1300 • C, and 1400 • C.

Research Techniques
In the research, different techniques were used to determine the initial powders' characteristics, and to investigate the powders after mechanical alloying, as well as the specimens fabricated by pulse plasma sintering.
According to the European Standard EN623-2 [27], the Archimedes method was applied to determine the selected properties-such as the density, open porosity, and soaking-of the samples obtained via PPS.
To characterize the phase composition, a Rigaku MiniFlex II diffractometer (Rigaku Corporation, Tokyo, Japan) with a CuK radiation wavelength of l = 1.54178 Å was used, set to the following recording parameters: current 15 mA, voltage 30 kV, step D2θ 0.05 • , counting time 3 s, angular range 2θ 23-120 • . The Williamson-Hall method was employed to estimate the mean crystallite size. The instrumental broadening was subtracted from the experimental breadth to obtain the physical broadening of each diffraction line.
Observations of the morphology of raw Al 2 O 3 , Ni, Al, NiAl-10%Al 2 O 3 , and NiAl-20%Al 2 O 3 powders, along with the microstructure of the produced bulk specimens, were performed using a JEOL JSM-6610 scanning electron microscope (Tokyo, Japan) equipped with secondary electron (SE) and backscattered electron (BSE) detectors. A voltage of 15 kV was applied throughout the observations. Surface microanalysis was conducted using an X-Max-type energy-dispersive X-ray spectrometer (EDS, Oxford, UK) to determine the elemental concentration in the fabricated composites after MA and after consolidation by PPS.
A quantitative microstructural description was carried out with a MicroMeter v.086b computer image analyzer. The raw Al 2 O 3 powder and Al 2 O 3 grains in the bulk samples of pure Al 2 O 3 obtained at different temperatures were investigated [28,29]. A quantitative description of the microstructure of the samples was carried out based on SEM images of randomly determined areas of the fracture of the samples. This technique provides the ability to obtain knowledge of the actual size of alumina in the specimen. Furthermore, based on stereological analysis, the following shape parameters of the Al 2 O 3 grains were determined: convexity (W = p/p c ), a coefficient describing the surface development (R = p/p c ·d 2 ), and elongation (α = d max /d 2 ), where p is the perimeter of the grain (µm) and p c is the Cauchy perimeter (µm), d 2 is the diameter of the circle of the same surface as the surface of the analyzed grain (µm), and d max is the maximum diameter of grain projection (µm) [28,29].
The Vickers method was used to determine the hardness of the bulk samples. Hardness was measured on the polished sample surface. The used load was 10 kg, and the holding time was 10 s. In the experiment, an HVS-30T hardness tester (Huatec Group Corporation, Beijing, China) was used. For individual specimens, at least 15 measurements were performed. The indentation sizes were estimated using a Nikon Eclipse LV15ON light microscope (Tokyo, Japan). Figure 1 shows the morphology and a particle size distribution histogram of the alumina powder. Microscopic observations reveal that the Al 2 O 3 powder used is spherical ( Figure 1). The observations showed that alumina tends to form agglomerates. Data from the literature [30] show that nanometric powders, which characterize a high surface-areato-single-particle volume ratio, can form agglomerates. Analysis of the obtained histogram shows that the alumina particle size distribution is unimodal (Figure 1). The average particle size was determined to be 0.140 µm, with a standard deviation equal to 0.06 µm. The density measured using a helium pycnometer for alumina was 3.9828 g/cm 3 . The density value obtained is close to the density reported by the manufacturer.

Description of the Initial Powders
( Figure 1). The observations showed that alumina tends to form agglomerates. Data from the literature [30] show that nanometric powders, which characterize a high surface-areato-single-particle volume ratio, can form agglomerates. Analysis of the obtained histogram shows that the alumina particle size distribution is unimodal (Figure 1). The average particle size was determined to be 0.140 µm, with a standard deviation equal to 0.06 µm. The density measured using a helium pycnometer for alumina was 3.9828 g/cm 3 . The density value obtained is close to the density reported by the manufacturer.     Figure 3 shows the XRD patterns of the (Ni-50at.%Al) + 10wt.%Al2O3 sample after various milling times. In the early stage of mechanical alloying, a new phase was developed, corresponding to the appearance of new peaks in the XRD spectra for the two  Figure 3 shows the XRD patterns of the (Ni-50at.%Al) + 10wt.%Al 2 O 3 sample after various milling times. In the early stage of mechanical alloying, a new phase was developed, corresponding to the appearance of new peaks in the XRD spectra for the two h-milled powders; these peaks correspond to a NiAl intermetallic phase. At the same time, the intensity of the Ni and Al peaks decreased significantly, and these peaks vanished with increasing milling time. The observed phase development (reaction between Ni and Al, and creation of the NiAl phase) is analogous to that previously described for mechanical alloying of a Ni-50at.%Al powder mixture carried out using the same ball mill [31]. The diffraction peaks of alumina remain constantly present in the XRD patterns. The phase composition of the milling product after 12 h is the NiAl intermetallic phase and Al 2 O 3 . For the Ni-50at.%Al-20wt.%Al 2 O 3 sample, a similar phase evolution was observed; however, it occurred more slowly, and the process required 15 h of milling to be completed. The influence of the amount of reinforcing phase on the rate of the phase evolution and NiAl phase formation during mechanical alloying was reported recently [17]. In the case of ball milling of Ni-Al-B powder mixtures, it was found that with the increase in the amount of boron in the mixture, the creation of the NiAl phase required a longer milling time [17].   Figure 4 shows the SEM images of the composite powder particles after the mechanical alloying: (a) NiAl-10%Al2O3 powder; (b) NiAl-20%Al2O3 powder. The observation reveals that the composite powders are characterized by the tendency to create spherical agglomerates with sizes ranging from below 1 µm up to 20 µm. The density measured using a helium pycnometer for NiAl-10%Al2O3 powder was 5.4571 As milling time increased, the NiAl diffraction peaks broadened ( Figure 3). This widening of the peaks was due to the decrease in the size of the NiAl crystallites, and the increased lattice microstrains in this intermetallic phase [32]. In the final milling product, the mean crystallite size of the NiAl phase was 14 nm and 11 nm for the samples containing 10% and 20% Al 2 O 3 , respectively. The mean crystal size was estimated by the Williamson-Hall method; in this method, it is assumed that the physical broadening of diffraction lines (β) is the sum of broadening resulting from the small size of crystallites (below 0.1 µm) (β c ) and the presence of lattice microstrains (β s ) [32]. In the Williamson-Hall method, considering the width of peaks as a function of 2θ allows one to estimate the mean crystallite size and the mean lattice microstrains separately. For the Williamson-Hall analysis, the instrumental broadening was subtracted from the experimental breadth to obtain the physical broadening of each diffraction line. Figure 4 shows the SEM images of the composite powder particles after the mechanical alloying: (a) NiAl-10%Al 2 O 3 powder; (b) NiAl-20%Al 2 O 3 powder. The observation reveals that the composite powders are characterized by the tendency to create spherical agglomerates with sizes ranging from below 1 µm up to 20 µm. The density measured using a helium pycnometer for NiAl-10%Al 2 O 3 powder was 5.4571 g/cm 3 , while for the NiAl-20%Al 2 O 3 powder it was 5.1858 g/cm 3 .  The EDS maps of the mechanical alloying products are shown in Figure 5. The presence of three elements-aluminum, oxygen, and nickel-was found. Furthermore, the EDS of the areas studied revealed that NiAl-10%Al2O3 powder ( Figure 5a  The EDS maps of the mechanical alloying products are shown in Figure 5. The presence of three elements-aluminum, oxygen, and nickel-was found. Furthermore, the EDS of the areas studied revealed that NiAl-10%Al 2 O 3 powder (Figure 5a The EDS maps of the mechanical alloying products are shown in Figure 5. The presence of three elements-aluminum, oxygen, and nickel-was found. Furthermore, the EDS of the areas studied revealed that NiAl-10%Al2O3 powder (Figure 5a) contained 15.69 at.% oxygen, 42.96 at.% aluminum, and 41.35 at.% nickel. In the case of NiAl-20%Al2O3 powder (Figure 5b), it was found that the powder contained 25.70 at.% oxygen, 43.29 at.% aluminum, and 31.01 at.% nickel.

Characterization of Al2O3 Samples after PPS Consolidation
The X-ray phase analysis of the alumina powder consolidated at various temperatures confirmed the presence of only the Al2O3 phase in the material. Figure 6 presents XRD patterns of the samples consolidated at 1000 °C and 1500 °C as examples.

Characterization of Al 2 O 3 Samples after PPS Consolidation
The X-ray phase analysis of the alumina powder consolidated at various temperatures confirmed the presence of only the Al 2 O 3 phase in the material. Selected properties of the Al2O3 samples after PPS consolidation are described in Table 2. The results obtained showed that the samples sintered at a temperature of 1000 °C had a relative density of 74.39 ± 2.45%, while the density of the Al2O3 samples sintered at 1100 °C was equal to 89.64 ± 1.35%. The sintered samples at temperatures from 1200 °C to 1500 °C were characterized by a relative density close to 100%.   Selected properties of the Al 2 O 3 samples after PPS consolidation are described in Table 2. The results obtained showed that the samples sintered at a temperature of 1000 • C had a relative density of 74.39 ± 2.45%, while the density of the Al 2 O 3 samples sintered at 1100 • C was equal to 89.64 ± 1.35%. The sintered samples at temperatures from 1200 • C to 1500 • C were characterized by a relative density close to 100%. Table 2 does not include the open porosity values for sintered samples at temperatures from 1200 • C to 1500 • C, because the open porosity and soaking are negligible for samples with very high densities (>99%). This is consistent with sintering theory, because there should be no open pores at high relative densities; therefore, the absorbency would be zero.  Figure 7 shows examples of SEM observation of the fractures of the Al 2 O 3 samples sintered at different temperatures by the PPS method, together with histograms showing the size distribution of the Al 2 O 3 grains in the tested samples. Histograms revealed that the alumina samples manufactured by PPS at 1000 • C were characterized by the smallest average grain size. The results showed that the samples sintered at 1000 • C had an average Al 2 O 3 grain size of 0.15 ± 0.06 µm. For these specimens, Al 2 O 3 grains were in the size range of 0.02 to 0.46 µm. It was found that sintering at 1000 • C does not cause grain growth of the original powder but, on the other hand, does not allow good-quality compacting of the samples. For the samples consolidated at 1100 • C, a non-significant increase in the size of Al 2 O 3 grains was observed. For these samples, an average grain size of Al 2 O 3 of 0.19 ± 0.08 µm was noted. In the case of the samples sintered at temperatures of 1000 • C and 1100 • C, weak consolidation of the samples was observed.
For the samples manufactured by PPS at 1200 • C, the average Al 2 O 3 size was 0.48 ± 0.17 µm. For the samples sintered at 1200 • C, an over-threefold increase in the size of the Al 2 O 3 grains was observed compared to the original powder size (0.14 ± 0.06 µm). For the samples obtained at 1300 • C, the average grain size was 1.27 ± 0.59 µm. Sintering at 1300 • C caused the Al 2 O 3 grains to grow ninefold compared to raw powder, while sintering at a temperature of 1400 • C resulted in a 12-fold increase in Al 2 O 3 grains compared to with raw alumina. For the samples obtained at 1400 • C, the grain size ranged from 0.02 µm to 5.99 µm, and the average size was 1.71 ± 0.85 µm. The highest growth of Al 2 O 3 grains was observed with the starting powder for the samples sintered at a temperature of 1500 • C. Histograms showed that the average grain size of Al 2 (Figure 7f) showed a significant effect of the temperature on the grain growth of Al 2 O 3 in the samples during the PPS process; this was confirmed by analysis of histograms. In all of these samples, complete sintering of the alumina grains was observed; moreover, this was confirmed by the high density results ( Table 2).
The parameters characterizing the shape factors of alumina grains in all samples are shown in Table 3. Based on the obtained results, it can be concluded that a similar shape characterizes the Al 2 O 3 grains sintered at 1000 • C and 1100 • C. This is evidenced by the similar values of parameters, i.e., convexity, the curvature of grain boundaries, and elongation. Moreover, the average Al 2 O 3 grain size values are close to the starting size of the alumina powder. Analysis of parameters such as convexity, the curvature of grain boundaries, and elongation showed that grains in all samples were similar.
A high sintering temperature is essential to obtaining good-quality samples; on the other hand, the use of a high sintering temperature causes grain growth. Accordingly, the applied temperature should be high enough to allow good sintering, but at the same time, it must not be so high that the grains do not grow excessively. First, experiments for pure Al 2 O 3 were carried out in the research, enabling the selection of the appropriate temperature for NiAl-Al 2 O 3 sintering. For further research, the temperature of 1500 • C was not selected, despite the density of samples being close to 100% due to the excessive growth of Al 2 O 3 grains.    Figure 8 shows the obtained hardness values for the alumina samples. The values obtained varied from 13 GPa to 23 GPa, and were correlated with the sintering temperatures applied during their preparation. Consequently, the samples sintered at 1000 • C were characterized by the lowest hardness values, with a hardness equal to 13.00 ± 0.38 GPa, while the samples sintered at the highest temperature of 1500 • C were characterized by the highest hardness values, amounting to 23.00 ± 0.97 GPa.
from the micrometric powders studied in our previous work [21]. While the maximal hardness attained previously for Al2O3 sintered at the highest temperature of 1500 °C amounted to 15.30 ± 0.87 GPa [21], the corresponding sample of ultrafine-grained Al2O3 ceramics in the present study (sample sintered at 1500 °C) exceeded this value by 67% Moreover, the achieved results correlate well with the results presented in the literature regarding dense Al2O3 ceramics. In the work carried out by Yuan et al. [33], oscillatory pressure sintering (OPS) and hot pressing at various temperatures in the range from 1100 °C to 1500 °C were used to produce high-density Al2O3 ceramics. The demonstrated hardness-temperature correlation was analogous to that observed in our study, as the hardness increased with increasing temperature, which was associated with higher compaction quality. Regarding the higher sintering temperatures, the hardness of samples obtained via the OPS method, with a maximum value at 1500 °C of approximately 24 GPa, stands in good correlation with the results achieved in our study. In the case of the lower sintering temperatures, the materials analyzed were characterized by lower hardness, with values below 10 GPa for 1100 °C. In comparison, the hardness of the samples made by the hot-pressing method was generally lower throughout the entire range [33].
Furthermore, in the study by Sun et al., where a combination of uniaxial pressing and pressureless sintering was employed to produce Al2O3 ceramics from nanospheres, the hardness of the ceramics sintered at 1550 °C amounted to 23.00 ± 1.4 GPa-a value very similar to our results. The relationship between hardness and sintering temperature was similarly observed, attributed to the observed correlation between the densification of the material and the size of the grains [34].
Our results showed that hardness of the specimens produced from the pure nanometric Al2O3 powder, as expected, was significantly higher than that of the NiAl-Al2O3 composites (results presented in Section 3.3).  The observed correlation between the hardness and the sintering temperature of Al 2 O 3 ceramics confirmed the increased compaction quality of samples sintered at higher temperatures. This effect is not clearly visible in the density values of materials, which are similarly high for sintering at 1300, 1400, and 1500 • C; however, it is visible during SEM observations of the fracture surface of samples. For the sample sintered at 1500 • C-in which the largest grain size was observed and, at the same time, the highest hardness was achieved-good compaction of grains prevented its pulling out (Figure 7f). It should be noted that the grains of Al 2 O 3 grow to more than 18 times the size of the initial Al 2 O 3 grains. However, it can be concluded that the increase in temperature, together with the better developed interfaces between grains, results in high hardness.
The obtained hardness values exceeded those for the similarly formed Al 2 O 3 ceramics from the micrometric powders studied in our previous work [21]. While the maximal hardness attained previously for Al 2 O 3 sintered at the highest temperature of 1500 • C amounted to 15.30 ± 0.87 GPa [21], the corresponding sample of ultrafine-grained Al 2 O 3 ceramics in the present study (sample sintered at 1500 • C) exceeded this value by 67% Moreover, the achieved results correlate well with the results presented in the literature regarding dense Al 2 O 3 ceramics. In the work carried out by Yuan et al. [33], oscillatory pressure sintering (OPS) and hot pressing at various temperatures in the range from 1100 • C to 1500 • C were used to produce high-density Al 2 O 3 ceramics. The demonstrated hardnesstemperature correlation was analogous to that observed in our study, as the hardness increased with increasing temperature, which was associated with higher compaction quality. Regarding the higher sintering temperatures, the hardness of samples obtained via the OPS method, with a maximum value at 1500 • C of approximately 24 GPa, stands in good correlation with the results achieved in our study. In the case of the lower sintering temperatures, the materials analyzed were characterized by lower hardness, with values below 10 GPa for 1100 • C. In comparison, the hardness of the samples made by the hotpressing method was generally lower throughout the entire range [33].
Furthermore, in the study by Sun et al., where a combination of uniaxial pressing and pressureless sintering was employed to produce Al 2 O 3 ceramics from nanospheres, the hardness of the ceramics sintered at 1550 • C amounted to 23.00 ± 1.4 GPa-a value very similar to our results. The relationship between hardness and sintering temperature was similarly observed, attributed to the observed correlation between the densification of the material and the size of the grains [34].
Our results showed that hardness of the specimens produced from the pure nanometric Al 2 O 3 powder, as expected, was significantly higher than that of the NiAl-Al 2 O 3 composites (results presented in Section 3.3).

Characterization of NiAl-Al2O3 Samples after PPS Consolidation
Figures 9 and 10 show the XRD patterns of the NiAl-Al2O3 composite powders consolidated at various temperatures. Only NiAl and Al2O3 diffraction peaks are present in the patterns, showing that no new phases are formed during the PPS process. The width of the NiAl diffraction patterns is smaller than the breadth of the NiAl peaks in the powders before the consolidation. The revealed sharpened peaks indicate grain growth in the NiAl phase during the PPS process. The mean crystallite size of the NiAl phase estimated by the Williamson-Hall method is in the range of 98-130 nm, and increases with the increase in the consolidation temperature. Thus, the estimated grain size is just above 100 nm, so it is at the limit of the applicability of the Williamson-Hall method, and may be affected by error. It can be concluded that the NiAl intermetallic phase in the consolidated samples has an ultrafine grain size. The SEM images obtained in the BSE (backscattered electron) mode presented in Figures 11 and 12 show the microstructure of sintered NiAl-Al2O3 composites with different alumina phase contents (10%- Figure 11; 20%- Figure 12) and different temperatures of the sintering process; red-marked areas correspond to magnified areas. The Al2O3 phase is revealed as dark grey areas in BSE mode, while the NiAl matrix is enrichment or lack of the intermetallic phase, regardless of the content of the alumina phase and the temperature of sintering. The composite with the highest average Al2O3 inclusions was obtained during sintering at a temperature of 1400 °C for the sample with 20% alumina content. Results of investigation revealed a lack of cracks and pores in the microstructure of the samples, independent of Al2O3 phase content, confirming good quality of the consolidation process performed. Moreover, the density of samples reached 100%. To reveal the chemical composition of the composite samples, energy-dispersive Xray spectroscopy (JEOL Ltd., Tokyo, Japan) was applied. Maps of the elemental distribution of the NiAl-10%Al2O3 and NiAl-20%Al2O3 composites were produced-the maps obtained from the surface of the NiAl-10%Al2O3 sample are shown in Figure 13, while the maps for NiAl-20%Al2O3 are presented in Figure 14. The results of the EDS analysis confirmed the chemical composition of the light and dark areas observed in the SEM images as a NiAl matrix and Al2O3, respectively. Contamination of the composites by iron was shown by the EDS results for the composites, regardless of the content of the alumina phase. The presence of iron is the result of contamination of the powders from steel milling tools, which is typical for mechanical alloying processes [16]. The SEM images obtained in the BSE (backscattered electron) mode presented in Figures 11 and 12 show the microstructure of sintered NiAl-Al 2 O 3 composites with different alumina phase contents (10%- Figure 11; 20%- Figure 12) and different temperatures of the sintering process; red-marked areas correspond to magnified areas. The Al 2 O 3 phase is revealed as dark grey areas in BSE mode, while the NiAl matrix is shown as light gray areas. Observation revealed Al 2 O 3 inclusions in the NiAl matrix, which were fairly uniformly distributed throughout the volume of the composite samples. A similar structure was obtained by consolidation of milled Ni-Al-B powders; in this case, composites with boron particles uniformly distributed in the nanocrystalline NiAl matrix were produced [17]. The analysis of the SEM results did not reveal areas characterized by enrichment or lack of the intermetallic phase, regardless of the content of the alumina phase and the temperature of sintering. The composite with the highest average Al 2 O 3 inclusions was obtained during sintering at a temperature of 1400 • C for the sample with 20% alumina content. Results of investigation revealed a lack of cracks and pores in the microstructure of the samples, independent of Al 2 O 3 phase content, confirming good quality of the consolidation process performed. Moreover, the density of samples reached 100%.     To reveal the chemical composition of the composite samples, energy-dispersive X-ray spectroscopy (JEOL Ltd., Tokyo, Japan) was applied. Maps of the elemental distribution of the NiAl-10%Al 2 O 3 and NiAl-20%Al 2 O 3 composites were produced-the maps obtained from the surface of the NiAl-10%Al 2 O 3 sample are shown in Figure 13, while the maps for NiAl-20%Al 2 O 3 are presented in Figure 14. The results of the EDS analysis confirmed the chemical composition of the light and dark areas observed in the SEM images as a NiAl matrix and Al 2 O 3 , respectively. Contamination of the composites by iron was shown by the EDS results for the composites, regardless of the content of the alumina phase. The presence of iron is the result of contamination of the powders from steel milling tools, which is typical for mechanical alloying processes [16].  The hardness measurements for NiAl-Al2O3 composites indicated that the hardness strongly depends on the Al2O3 content in the composite, as well as on the sintering temperature applied. Figure 15 shows hardness values for thr NiAl-10%Al2O3 and NiAl-20%Al2O3 samples. The hardness of the composites containing 20 wt.% Al2O3 was in the range of 7.0-8.4 GPa, while for the composites with 10 wt.% Al2O3 the hardness values varied from  The hardness measurements for NiAl-Al2O3 composites indicated that the hardness strongly depends on the Al2O3 content in the composite, as well as on the sintering temperature applied. Figure 15 shows hardness values for thr NiAl-10%Al2O3 and NiAl-20%Al2O3 samples. The hardness of the composites containing 20 wt.% Al2O3 was in the range of 7.0-8.4 GPa, while for the composites with 10 wt.% Al2O3 the hardness values varied from  However, regardless of Al2O3 content in the structure, the specimens sintered at the lowest temperature (1200 °C) were characterized by the highest hardness in both series, reaching 7.2 ± 0.29 GPa and 8.4 ± 0.07 GPa for 10 wt.% Al2O3 and 20 wt.% Al2O3, respectively. The hardness variation by the applied sintering temperature was lower for the samples with 20 wt.% Al2O3 content than for the samples with a lower amount of Al2O3.
Moreover, the hardness of NiAl-Al2O3 composites decreased with increasing sintering temperature, while the opposite was the case for sintered Al2O3 samples. This phenomenon could be explained by various structures of the compared samples, the developed interfaces, and/or the growth of NiAl and Al2O3 grains.
The intermetallic matrix in all samples (1200 °C, 1300 °C, 1400 °C) has an ultrafine grain size (98 to 130 nm); however, the growth of NiAl grains is correlated with the temperature of sintering. At the same time, the growth of Al2O3 grains is more intensive, as was expected after grain growth investigation of pure Al2O3 powder consolidated by PPS. SEM observations revealed the growth of Al2O3 inclusions with the increase in the temperature of sintering (Figures 11 and 12). These changes of both matrix and reinforcement grains contribute to the hardness. However, the new interfaces between NiAl and Al2O3 grains appeared in composites. Moreover, Al2O3 inclusions are separated by NiAl matrix grains (Figures 13 and 14); thus, we can predict the effect of hardening by Al2O3 inclusions. With smaller inclusions, higher hardening should be expected. The contribution of the microstructure of composites to crack propagation should be also considered; investigations in this area are in progress.
The hardness values obtained for lower Al2O3 content (10 wt.%) were practically identical to those presented in the previous research conducted by the authors of this work [21]. The composite with 20 wt.% Al2O3 after sintering at 1400 °C reached the highest value compared to the corresponding sample from the previous work [19]. The hardness values of the NiAl-Al2O3 composites with 10 wt.% and 20 wt.% Al2O3 content sintered at 1400 °C were the same, at 5.8 GPa [21]; however, in the previous study, micrometric alumina was used as the initial powder [21].
Regarding the material analyzed by Albiter et al. [35], the hardness of NiAl produced by a combination of mechanical alloying with vacuum hot pressing at 1200 °C and 1500 °C was equal to 5.1 GPa and 5.0 GPa, respectively. Similar results were obtained by Biranvand et al. [36], where the hardness of NiAl fabricated by combining mechanical alloying and spark plasma sintering at 1000 °C reached 5.63 ± 0.07 GPa. The hardness of nanocrystalline NiAl produced by mechanical alloying followed by high-pressure hot- However, regardless of Al 2 O 3 content in the structure, the specimens sintered at the lowest temperature (1200 • C) were characterized by the highest hardness in both series, reaching 7.2 ± 0.29 GPa and 8.4 ± 0.07 GPa for 10 wt.% Al 2 O 3 and 20 wt.% Al 2 O 3 , respectively. The hardness variation by the applied sintering temperature was lower for the samples with 20 wt.% Al 2 O 3 content than for the samples with a lower amount of Al 2 O 3 .
Moreover, the hardness of NiAl-Al 2 O 3 composites decreased with increasing sintering temperature, while the opposite was the case for sintered Al 2 O 3 samples. This phenomenon could be explained by various structures of the compared samples, the developed interfaces, and/or the growth of NiAl and Al 2 O 3 grains.
The intermetallic matrix in all samples (1200 • C, 1300 • C, 1400 • C) has an ultrafine grain size (98 to 130 nm); however, the growth of NiAl grains is correlated with the temperature of sintering. At the same time, the growth of Al 2 O 3 grains is more intensive, as was expected after grain growth investigation of pure Al 2 O 3 powder consolidated by PPS. SEM observations revealed the growth of Al 2 O 3 inclusions with the increase in the temperature of sintering (Figures 11 and 12). These changes of both matrix and reinforcement grains contribute to the hardness. However, the new interfaces between NiAl and Al 2 O 3 grains appeared in composites. Moreover, Al 2 O 3 inclusions are separated by NiAl matrix grains (Figures 13 and 14); thus, we can predict the effect of hardening by Al 2 O 3 inclusions. With smaller inclusions, higher hardening should be expected. The contribution of the microstructure of composites to crack propagation should be also considered; investigations in this area are in progress.
The hardness values obtained for lower Al 2 O 3 content (10 wt.%) were practically identical to those presented in the previous research conducted by the authors of this work [21]. The composite with 20 wt.% Al 2 O 3 after sintering at 1400 • C reached the highest value compared to the corresponding sample from the previous work [19]. The hardness values of the NiAl-Al 2 O 3 composites with 10 wt.% and 20 wt.% Al 2 O 3 content sintered at 1400 • C were the same, at 5.8 GPa [21]; however, in the previous study, micrometric alumina was used as the initial powder [21].
Regarding the material analyzed by Albiter et al. [35], the hardness of NiAl produced by a combination of mechanical alloying with vacuum hot pressing at 1200 • C and 1500 • C was equal to 5.1 GPa and 5.0 GPa, respectively. Similar results were obtained by Biranvand et al. [36], where the hardness of NiAl fabricated by combining mechanical alloying and spark plasma sintering at 1000 • C reached 5.63 ± 0.07 GPa. The hardness of nanocrystalline NiAl produced by mechanical alloying followed by high-pressure hot-pressing consolidation was 9.53 GPa [31]. According to Kaliński et al. [37], the hardness of NiAl samples obtained by a combination of mechanical alloying and hot pressing amounted merely to 3.08 ± 0.10 GPa for those sintered at 1400 • C; in the same study, however, it was confirmed that the addition of Al 2 O 3 to the structure could improve the mechanical characteristics of the resulting composite, including hardness. The hardness of the analogically prepared NiAl with the addition of 20 vol.% Al 2 O 3 was found to be equal to 4.29 ± 0.5 GPa [37] A similar relationship was observed in research carried out by Zhu et al. [38], where the addition of Al 2 O 3 caused an increase in the microhardness of the material up to a value reached for~10 wt.% Al 2 O 3 in the composite. The decrease in hardness with further increasing Al 2 O 3 content was associated with an increase in the material's porosity and the appearance of a brittle Ni 2 Al phase in the structure [38]. The favorable effect of Al 2 O 3 on the hardness of the NiAl matrix composite was also demonstrated in the study conducted by Udhayabanu et al. [39], where the composite with 10 vol.% Al 2 O 3 was obtained by a combination of reactive milling and subsequent SPS. The composite hardness value of 7.6 GPa was found to result from the presence of nanometric particles of Al 2 O 3 [9]. The hardness of the composites with a similar structure, but containing boron instead of alumina distributed in the NiAl matrix, ranged from 10.3 GPa to 12.6 GPa, and increased with increasing boron content [17].
Analyzing the available scientific literature, it should be noted that the hardness of NiAl observed in various studies exhibits considerable discrepancies depending on the starting materials applied, the manufacturing method, and/or the sintering temperature.
Bulk NiAl-Al 2 O 3 composites obtained in the presented research exceeded the hardness values reported in the literature and in the authors' previous paper [21]. The following factors can be held responsible for these results: the contribution of nanometric initial Al 2 O 3 powders in the MA process and, consequently, the presence of the ultrafine-grained structures in the composites after sintering, along with the high level of compaction of the obtained materials.

Conclusions
To the best of our knowledge, this paper represents the first time NiAl-Al 2 O 3 composite powders have been consolidated by the pulse plasma sintering method after mechanical alloying.
Final NiAl-Al 2 O 3 composites were obtained from composite powders that were produced by mechanical alloying of powder blends containing Ni-50at.%Al, with the contribution of 10 wt.% and 20 wt.% nanometric aluminum oxide powder. Nanometric initial Al 2 O 3 powder was used for the reinforcement of synthetized intermetallic matrix powder in MA processes; consequently, the presence of ultrafine-grained structures of composites after PPS sintering enhanced the hardness.
Grain growth and hardness of bulk Al 2 O 3 after PPS consolidation at various temperatures were analyzed; these studies aimed to determine the appropriate sintering temperature for the composite powders. Subsequently, based on these tests, sintering of the NiAl-Al 2 O 3 powders was carried out at temperatures of 1200 • C, 1300 • C, and 1400 • C.
Characterization of the morphology and size of initial powders, composite powders and, finally, the grains in samples after PPS consolidations, were carried out in order to describe the correlation between the sintering temperature, structure, and hardness of composites. The hardness of the NiAl-Al 2 O 3 composites depends on the Al 2 O 3 content in the material and the sintering temperature applied. The samples sintered at the lowest temperature (1200 • C) were characterized by the highest hardness, reaching 7.2 ± 0.29 GPa and 8.4 ± 0.07 GPa for the composites containing 10 wt.% Al 2 O 3 and 20 wt.% Al 2 O 3 , respectively. The hardness of the ultrafine-grained NiAl intermetallic matrix composites with Al 2 O 3 inclusions exceeded the hardness values reported in the literature. The contribution of the intial nanometric Al 2 O 3 powders in the MA process, and the consequent presence of the ultrafine-grained structures in the composites after sintering and the high level of compaction of the obtained materials, as well as interfaces between NiAl and Al 2 O 3 , are responsible for the high hardness of the bulk composites. The increase in the sintering temperature does not improve the compaction of the composite powders; moreover, it causes the grain growth of Al 2 O 3 inclusions to activate.
From a technological point of view, the possibility of using a sintering temperature as low as 1200 • C is crucial in order to produce fully dense, ultrafine-grained composites with high hardness.
Investigations of correlations between the microstructures of NiAl-Al 2 O 3 composites and mechanical properties, crack propagation, and fracture toughness are in progress.