Characterization of Al2O3 Samples and NiAl–Al2O3 Composite Consolidated by Pulse Plasma Sintering

The paper describes an investigation of Al2O3 samples and NiAl–Al2O3 composites consolidated by pulse plasma sintering (PPS). In the experiment, several methods were used to determine the properties and microstructure of the raw Al2O3 powder, NiAl–Al2O3 powder after mechanical alloying, and samples obtained via the PPS. The microstructural investigation of the alumina and composite properties involves scanning electron microscopy (SEM) analysis and X-ray diffraction (XRD). The relative densities were investigated with helium pycnometer and Archimedes method measurements. Microhardness analysis with fracture toughness (KIC) measures was applied to estimate the mechanical properties of the investigated materials. Using the PPS technique allows the production of bulk Al2O3 samples and intermetallic ceramic composites from the NiAl–Al2O3 system. To produce by PPS method the NiAl–Al2O3 bulk materials initially, the composite powder NiAl–Al2O3 was obtained by mechanical alloying. As initial powders, Ni, Al, and Al2O3 were used. After the PPS process, the final composite materials consist of two phases: Al2O3 located within the NiAl matrix. The intermetallic ceramic composites have relative densities: for composites with 10 wt.% Al2O3 97.9% and samples containing 20 wt.% Al2O3 close to 100%. The hardness of both composites is equal to 5.8 GPa. Moreover, after PPS consolidation, NiAl–Al2O3 composites were characterized by high plasticity. The presented results are promising for the subsequent study of consolidation composite NiAl–Al2O3 powder with various initial contributions of ceramics (Al2O3) and a mixture of intermetallic–ceramic composite powders with the addition of ceramics to fabricate composites with complex microstructures and properties. In composites with complex microstructures that belong to the new class of composites, in particular, the synergistic effect of various mechanisms of improving the fracture toughness will be operated.


Introduction
Ceramic matrix composites are an important group of composites developed over many years. Ceramic-metal composites belong to this group of materials. New methods of fabrication and new types of these materials have been elaborated on. Metals such as Mo, V, Al, Ni, Cu, or Ti are often introduced into the ceramic matrix [1][2][3][4][5][6][7]. Metal particles located in the ceramic matrix interact with propagating cracks and cause deflection of the crack, bridging or stopping the cracks. As a consequence, the increasing fracture toughness of the composites has been observed [8][9][10][11][12]. However, it is not the only metal that is so active in improving the fracture behavior of brittle ceramic or intermetallic matrix composites. Intermetallic phases and other compounds are also regarded as reinforcement of composites [13][14][15][16][17][18][19]. Particular attention has been paid to intermetallic matrix composite particle-reinforced intermetallic compounds or ceramics. For example, in work [18], Al 2 O 3 particle reinforced TiAl composites were reaction-synthesized from a powder mixture of Ti, Al, TiO 2 , and Nb 2 O 5 by the hot-pressing method. In the TiAl matrix, Ti 3 Al, NbAl 3 phases, and fine Al 2 O 3 particles were found. The composites' bending strength and fracture toughness reach the values of 398.5 MPa and 6.99 MPa m 1/2 , respectively [18]. Other examples of investigation of composites of the Al 2 O 3 -TiAl system can be found, for example, in papers [16,17,19].
Another interesting system is Al 2 O 3 -NiAl [20][21][22] or NiAl/TiC-Al 2 O 3 [23]. Combining these ceramics with intermetallic phases allows producing composites for hightemperature and chemical-resistant applications. Moreover, the introduction of the intermetallic phase into the ceramic matrix and ceramics into the intermetallic matrix can efficiently improve the fracture toughness of composite materials of intermetallic-ceramics systems. Because of that, the NiAl-Al 2 O 3 system has been intensively investigated, especially the composites in which NiAl is a matrix and alumina is a reinforcing phase [22,[24][25][26][27]. Abe and Ohwa, in their work [27] prepared NiAl-Al 2 O 3 composites by the pressureless sintering of the powders synthesized via chemical precipitation route. The dominant formation of NiAl 2 O 4 was observed in composites. The compressive stress for an oxidized NiAl-Al 2 O 3 composite was equal to 127 MPa, and the improved fracture toughness was equal to 6.2 MPa m 1/2 [27].
Primarily, the reinforcing phase is prepared separately before the composite fabrication. Then the consolidation of mixed powders is provided to produce bulk composite materials. For example, at work [25], dense Al 2 O 3 -NiAl composites were prepared by hot pressing. Al 2 O 3 and NiAl powders were used as starting powders. However, ceramic matrix and intermetallic matrix composites can also be achieved not by consolidating the blended powders of matrix and reinforcement but by consolidating previously prepared composite powders.
There are different methods for producing composite powders. In methods called in situ, compounds are created by chemical reactions. As a result, very fine reinforcement particles and their homogeneity distribution in the matrix can be obtained in the composite powder. For example, Beyhaghi et al. [22,24] produced nanocomposite powders NiAl-Al 2 O 3 via the mechano-synthesis route. As initial substrates, Ni, NiO, and Al powders were used. As a result of in situ synthesis, nano-crystallites of NiAl and Al 2 O 3 were obtained. In another work [21], NiAl-Al 2 O 3 composites were sintered in-situ from Al powder and preoxidized Ni particles by aluminothermic reactions. Mechanical alloying (MA) is a popular method of synthesis materials [28][29][30][31]. This method successfully produces intermetallics and intermetallic matrix composites [28,[32][33][34]. There are other techniques of intermetallics synthesis, such as other methods of powder metallurgy, self-propagating high-temperature synthesis (SHS), or rapid solidification [35][36][37]. However, the synthesis of intermetallic materials requires special conditions because of the restrictive stoichiometry of this compound and the complex crystal structure. In mechanical alloying, high-energy ball milling is involved in the synthesis of the intermetallic material and, during this, the chemical reaction and phase transformation have been complied with. The parameters of the mechanical alloying process and used reactants control the final product.
In the present paper, NiAl-Al 2 O 3 composite powder was produced and then consolidated. To form the intermetallic phase, Ni and Al powders were used. The process of milling (MA) began with metals and added ceramic powder (Al 2 O 3 ). This allowed us to obtain a uniform distribution of constituencies in the composite powder and then keeping it in the bulk composite obtained by consolidating them. During the MA process, the Ni with Al will constitute the NiAl with trapped inside Al 2 O 3 particles. After the consolidation of such composite powder in bulk material, it should be the positive result of improving the fracture toughness. Significantly, the redirecting of crack propagation by Al 2 O 3 particles was expected. Moreover, the composite powder NiAl-Al 2 O 3 can also be mixed with ceramic powder and consolidated to obtain the final ceramic-intermetallic bulk composite. The microstructure of such forming composite will be complex and consists of a ceramic matrix distributed in its intermetallic phase with ceramic particles trapped inside. Such a method of producing the ceramic matrix composites with the contribution of intermetallic phase with the ceramic particles inside is not commonly presented in the literature. In the scope of own research, such composites are included. However, initially, the composite NiAl-Al 2 O 3 powder fabrication by the proposed MA method must be elaborated and characterized and the following consolidation process. As a method of composite consolidation, pulse plasma sintering (PPS) was selected.
In this technique, the material was heated by electric pulses generated periodically by a discharged capacitor battery and, at the same time, subjected to uniaxial pressing. The application of a capacitor battery as a source of electrical energy allows the electric pulses to be produced periodically over several hundred microseconds and a current intensity of about 100 kA [38,39]. Through the short duration of the electric pulse related to the time interval between the individual pulses, the temperature achieved during the pulse is higher than that stabilized during the traditional sintering method [38,39]. The temperature of the specimen and its heating rate are regulated by controlling the energy dissipated through the electric pulses, including adjusting the intervals between consecutive pulses. The choice of this method was based upon the advantage of short sintering time, which gives the possibility to rapidly sinter bulk materials and avoid the crystal coarsening [39]. In the experimental work, Al 2 O 3 powder was also consolidated by the PPS method. This research aims to recognize the PPS method as a possible technique to produce bulk composite materials from ceramic powder and a mixture of composite powder NiAl-Al 2 O 3 .
The characterization of pure Al 2 O 3 powder and NiAl-Al 2 O 3 composite powder before and after the PPS consolidation is presented. These experiments will be treated as the beginning of further research on the fabrication of ceramic-intermetallic composites using composite NiAl-Al 2 O 3 powder.

Initial Powders and Preparing the Composite Powder
In this experiment, the following powders were used: α-Al 2 O 3 powder (MARTOXID ® MR-5, Martinswerk GmbH, Bergheim, German) with an average particle size range of 0.3-6 µm, Ni powder (ABCR GmbH and Co.KG, Karlsruhe, German) with an average particle size ranged from 3 µm to 7 µm and Al powder (ABCR GmbH and Co.KG, Karlsruhe, German) with average particle size equal 44 µm. Characterization of the raw powders was performed based on data contributed by the manufacturer. The powders used for milling were Ni (99.9% purity), Al (99.7% purity), and Al 2 O 3 (99.98% purity).
The first stage of the research was producing composite powders based on Al 2 O 3 , Ni, and Al due to mechanical alloying. For the milling processes we used powder blends of Ni-50at.%Al with the addition of 10 wt.% and 20 wt.% of Al 2 O 3 . The milling processes were carried out in a high-energy SPEX 8000 D shaker ball mill (SPEX ® SamplePrep, Metuchen, NJ, USA). The ball-to-powder weight ratio was 10:1. The milling processes and sampling of powders were conducted under the protective atmosphere of Ar.

Pulse Plasma Sintering (PPS) Process
The PPS method was used to produce the bulk specimens from the prepared composite powder. Table 1 gives the PPS process parameters. In Figure 1, the equipment used to produce composites by the PPS method is shown. The powder is loaded into the matrix and heating by the heat generated during the electric pulses. After the process, the disc samples were obtained. In the experiment, samples containing 100% Al 2 O 3 were sintered at various temperatures (1000 • C, 1100 • C, 1200 • C, 1300 • C, 1400 • C, 1500 • C) in order to choose the proper temperature for sintering. The following temperatures in the PPS process were determined base on the required temperature of Al 2 O 3 sintering, which is generally estimated with melting temperature and is close to 1450 • C.
On the other hand, as mentioned in the Introduction, the temperature of consolidation achieved in the PPS process is higher than in the traditional sintering method [38,39]. Because of that, 1500 • C was the higher applied temperature of the process. Moreover, to control the progress of sintering in the PPS method, experiments were undertaken with the proposed range of temperature, starting at 1000 • C. On the other hand, as mentioned in the Introduction, the temperature of consolidation achieved in the PPS process is higher than in the traditional sintering method [38,39]. Because of that, 1500 °C was the higher applied temperature of the process. Moreover, to control the progress of sintering in the PPS method, experiments were undertaken with the proposed range of temperature, starting at 1000 °C.  An exemplary record of changes in temperature, shrinkage rate, and pressing pressure during the PPS sintering process for the Al2O3 sample and composite sample is presented in Figure 2. The displacement of the stamp measured by a laser extensometer is used to estimate the consolidation progress, which is called shrinkage.
Based on the analysis of Figure 2a, it can be concluded that for the Al2O3 powder during the PPS process at the temperature of about 1200 °C, the powder only expands. In this temperature range (20-1200 °C), no shrinkage of Al2O3 powder was observed. Above the temperature of 1200 °C, sample shrinkage begins, i.e., the consolidation stage during PPS. At the same time, when the start of shrinkage of the sample, the pressure was increased to 80 MPa (green line in the diagram). Based on the graph obtained, it can be concluded that during 3 min of being at the target temperature (1400 °C), the Al2O3 sample An exemplary record of changes in temperature, shrinkage rate, and pressing pressure during the PPS sintering process for the Al 2 O 3 sample and composite sample is presented in Figure 2. The displacement of the stamp measured by a laser extensometer is used to estimate the consolidation progress, which is called shrinkage.
Based on the analysis of Figure 2a, it can be concluded that for the Al 2 O 3 powder during the PPS process at the temperature of about 1200 • C, the powder only expands. In this temperature range (20-1200 • C), no shrinkage of Al 2 O 3 powder was observed. Above the temperature of 1200 • C, sample shrinkage begins, i.e., the consolidation stage during PPS. At the same time, when the start of shrinkage of the sample, the pressure was increased to 80 MPa (green line in the diagram). Based on the graph obtained, it can be concluded that during 3 min of being at the target temperature (1400 • C), the Al 2 O 3 sample is finally consolidated. This is confirmed by a slight shrinkage visible in the chart (Figure 2a-blue line). In NiAl-Al 2 O 3 powder, it was found to shrink after the first pulse ( Figure 2b). The obtained diagram found that more significant NiAl-Al 2 O 3 powder shrinkage occurs after exceeding 500 • C and lasts up to 1200 • C (Figure 2b-blue line). Another shrinkage observed at 1300 • C is related to the increase in pressure to 80 MPa. At the target temperature, which is 1400 • C, no shrinkage of the NiAl-Al 2 O 3 sample was observed. The process flows for the samples produced by the PPS method presented in Figure 2 are representative. The shrinkage characteristics during the process shown in Figure 2a,b have been different. This is because, in the research, we used two other materials with different thermal expansion coefficients.
is finally consolidated. This is confirmed by a slight shrinkage visible in the chart ( Figure  2a-blue line). In NiAl-Al2O3 powder, it was found to shrink after the first pulse ( Figure  2b). The obtained diagram found that more significant NiAl-Al2O3 powder shrinkage occurs after exceeding 500 °C and lasts up to 1200 °C (Figure 2b-blue line). Another shrinkage observed at 1300 °C is related to the increase in pressure to 80 MPa. At the target temperature, which is 1400 °C, no shrinkage of the NiAl-Al2O3 sample was observed. The process flows for the samples produced by the PPS method presented in Figure 2 are representative. The shrinkage characteristics during the process shown in Figure 2a,b have been different. This is because, in the research, we used two other materials with different thermal expansion coefficients.

Experimental Techniques
In the experiment, several methods were used to determine the properties and microstructure of the raw powder, powder after mechanical alloying, and samples obtained via the PPS. The phase composition and the structure of the powders after different milling times and powders after consolidation by the PPS technique were investigated by Xray diffraction (XRD). The examination was accomplished by a Rigaku Miniflex II X-ray diffractometer (Rigaku Corporation, Tokyo, Japan) with Cu Kα (λ = 1.54178 Å, 15 mA, 30 kV) at a step size 0.05° with radiation in the 2θ range from 23° to 123°. For assessing the mean crystallite size, the Williamson-Hall method was employed (the instrumental broadening was subtracted from the experimental breadth to obtain the physical broadening of each diffraction line).

Experimental Techniques
In the experiment, several methods were used to determine the properties and microstructure of the raw powder, powder after mechanical alloying, and samples obtained via the PPS. The phase composition and the structure of the powders after different milling times and powders after consolidation by the PPS technique were investigated by X-ray diffraction (XRD). The examination was accomplished by a Rigaku Miniflex II X-ray diffractometer (Rigaku Corporation, Tokyo, Japan) with Cu Kα (λ = 1.54178 Å, 15 mA, 30 kV) at a step size 0.05 • with radiation in the 2θ range from 23 • to 123 • . For assessing the mean crystallite size, the Williamson-Hall method was employed (the instrumental broadening was subtracted from the experimental breadth to obtain the physical broadening of each diffraction line).
To establish the level of sintering compaction, the relative density was determined using the density of the powders designated using a helium pycnometer (AccuPyc 1340 II by Micromeritics, Norcross, GA, USA). The Archimedes method was applied to calculate the apparent and relative density, open porosity, and absorptivity of samples prepared by the PPS technique. According to the European Standard ISO 18754:2013 (EN), Archimedes' method was measurement [40].
The hardness of the prepared samples was measured by the Vickers method on the polished sample surface under a load of 20 kG with a 10-s holding time. The hardness tester HVS-30T (Huatec Group Corporation, Beijing, China) was used to determine the hardness. For each sample, at least 15 measurements were made. The corresponding indentation sizes were determined using diagonals, measured using a light microscope Nikon Eclipse LV15ON (Nikon Corporation, Tokyo, Japan). Based on the length of cracks propagating from the corner of the hardness indentation, the material's fracture toughness (K IC ) was determined. In this investigation, a Vickers hardness indenter was applied to propagate the median cracks on the surface.
Observations of the microstructure and morphology of source Al 2 O 3 , Ni, and Al powders, composite powder after mechanical alloying, and fabricated bulk samples were carried out using a JEOL JSM-6610 scanning electron microscope-SEM (JEOL, Tokyo, Japan). Before observation, the samples were carbon-coated using the Quorum Q150T ESS coating system. The observation was performed using a secondary electron detector (SE) and a back-scattered electron (BSE) detector. A voltage of 15 kV was applied during the observations. Surface microanalysis of the chemical composition was performed using an X-Max type energy-dispersive X-ray spectrometer (EDS, Hitachi High-Tech Corp., Oxford, UK) to determine the elemental distribution in the obtained powder particles of the MA products and composites after sintering via PPS.
The changes in the size distribution of the raw Al 2 O 3 powder and Al 2 O 3 grains in the bulk samples of pure alumina obtained at different temperatures were examined using a stereological analysis. A quantitative description of the microstructure of the specimens was carried out based on scanning electron microscopy (SEM) images of randomly chosen areas on the fracture of samples. The quantitative description was carried out using a MicroMeter v.086b computer image analyzer [41,42].  Figure 4 shows a histogram of the particle size distribution of the Al2O3 powder. Based on the histogram, it can be assumed that the Al2O3 distribution is almost unimodal, with particles ranging from 0.02 µm to 5.30 µm in size. The results obtained showed that the average particle size of the alumina was about 0.39 µm. The results of the Al2O3 particle size distribution analysis are consistent with the data provided by the manufacturer.  Figure 4 shows a histogram of the particle size distribution of the Al 2 O 3 powder. Based on the histogram, it can be assumed that the Al 2 O 3 distribution is almost unimodal, with particles ranging from 0.02 µm to 5.30 µm in size. The results obtained showed that the average particle size of the alumina was about 0.39 µm. The results of the Al 2 O 3 particle size distribution analysis are consistent with the data provided by the manufacturer.  Figure 4 shows a histogram of the particle size distribution of the Al2O3 powder. Based on the histogram, it can be assumed that the Al2O3 distribution is almost unimodal, with particles ranging from 0.02 µm to 5.30 µm in size. The results obtained showed that the average particle size of the alumina was about 0.39 µm. The results of the Al2O3 particle size distribution analysis are consistent with the data provided by the manufacturer. The density of Al2O3 was equal to 3.93 g/cm 3, which corresponds to the value given by the manufacturer.

NiAl-Al2O3 Composite Powder Characterization
The powders' density was determined, and for the NiAl-10%Al2O3 powder, it was 5.561 g/cm 3 , while for the NiAl-20%Al2O3, it was equal to 5.236 g/cm 3 . As shown in Figure The density of Al 2 O 3 was equal to 3.93 g/cm 3, which corresponds to the value given by the manufacturer.

NiAl-Al 2 O 3 Composite Powder Characterization
The powders' density was determined, and for the NiAl-10%Al 2 O 3 powder, it was 5.561 g/cm 3 , while for the NiAl-20%Al 2 O 3, it was equal to 5.236 g/cm 3 . As shown in Figure 5, the composite powders make spheroidal agglomerates. Fine (below 1 µm) as well as large agglomerates up to 50 µm were observed.   Figure 6 shows EDS maps of the MA product. Besides nickel, aluminum, and oxygen, the presence of iron was detected. Contamination of powders by Fe from steel milling tools is commonly observed in mechanical alloying processes [28].  Figure 6 shows EDS maps of the MA product. Besides nickel, aluminum, and oxygen, the presence of iron was detected. Contamination of powders by Fe from steel milling tools is commonly observed in mechanical alloying processes [28].  Figure 6 shows EDS maps of the MA product. Besides nickel, aluminum, and oxygen, the presence of iron was detected. Contamination of powders by Fe from steel milling tools is commonly observed in mechanical alloying processes [28]. Phase development in the Ni50Al50-Al2O3 powder mixtures during mechanical alloying can be examined based on the XRD patterns of the powders after various milling times. Figure 7 displays the patterns for the milled Ni-50at.%Al-10wt.%Al2O3 sample. It can be seen that after two h of mechanical alloying, a new phase was formed, which is demonstrated by the appearance of novel peaks in the XRD pattern. These peaks have been assigned to a NiAl phase. At the same time, the intensity of Ni and Al peaks decreased significantly. In the pattern for the three h-milled powder mixtures, all Al and Ni peaks vanished. This indicates that all Ni reacted with Al, creating a NiAl intermetallic phase, at least partially ordered. The observed reaction between Ni and Al and phase development during the formation of the NiAl phase was analogous to those described earlier for mechanical alloying of Ni-50at.%Al powder mixture performed in the same kind  Figure 7 displays the patterns for the milled Ni-50at.%Al-10wt.%Al 2 O 3 sample. It can be seen that after two h of mechanical alloying, a new phase was formed, which is demonstrated by the appearance of novel peaks in the XRD pattern. These peaks have been assigned to a NiAl phase. At the same time, the intensity of Ni and Al peaks decreased significantly. In the pattern for the three h-milled powder mixtures, all Al and Ni peaks vanished. This indicates that all Ni reacted with Al, creating a NiAl intermetallic phase, at least partially ordered. The observed reaction between Ni and Al and phase development during the formation of the NiAl phase was analogous to those described earlier for mechanical alloying of Ni-50at.%Al powder mixture performed in the same kind of mill [43]. The diffraction peaks of Al 2 O 3 were present in the discussed XRD patterns all while. The phase composition of the powder mixture did not change for longer milling times. For the Ni-50at.%Al-20wt.%Al 2 O 3 sample, the same phase evolution was observed.
It can be seen in Figure 7 that the NiAl diffraction peaks broadened with the milling time extension. This broadening was due to the reduction in the crystallite size of the NiAl phase and the increase in lattice microstrains in this phase [44]. Since some of the NiAl peaks overlap with the Al 2 O 3 peaks, for the analysis of the peaks' width and Williamson-Hall calculations the peaks were separated by fitting and deconvolution. The estimated mean crystallite size of the NiAl phase in the final milling product was 14 nm and 12 nm for the sample containing 10% and 20% of Al 2 O 3 , respectively.
The produced powders have a composite structure with Al 2 O 3 particles distributed in the nanocrystalline NiAl intermetallic matrix. Interestingly, literature data show that Krasnowski et al. obtained a similar structure for NiAl-B powder [32]. Krasnowski et al. conducted the process of mechanical alloying for Ni, Al, and B powders. They managed to manufacture a powder with a composite structure, in which fine B particles were homogeneously distributed in a nanocrystalline NiAl matrix [32].
for the sample containing 10% and 20% of Al2O3, respectively.
The produced powders have a composite structure with Al2O3 particles distributed in the nanocrystalline NiAl intermetallic matrix. Interestingly, literature data show that Krasnowski et al. obtained a similar structure for NiAl-B powder [32]. Krasnowski et al. conducted the process of mechanical alloying for Ni, Al, and B powders. They managed to manufacture a powder with a composite structure, in which fine B particles were homogeneously distributed in a nanocrystalline NiAl matrix [32].

Characterization of Al 2 O 3 Powder Compacted by PPS
The PPS consolidation method was firstly applied for pouring Al 2 O 3 powder at the range of temperature from 1000 • C up to 1500 • C to describe the sintering process of ceramic powder in the function of temperature. The density, hardness, and size of ceramic grains in particular were examined in the function of sintering temperature. The results of these experiments were used to select a proper temperature of PPS consolidation for the composite NiAl 2 O 3 -Al 2 O 3 powder. The crucial was to sinter the composite powder and not allow for grain growth of Al 2 O 3 particles.
In the XRD patterns of the Al 2 O 3 powder after consolidation by PPS at various temperatures, only the peaks of Al 2 O 3 are visible. Figure 8 shows the pattern of the sample sintered at 1400 • C as an example. grains in particular were examined in the function of sintering temperature. The results of these experiments were used to select a proper temperature of PPS consolidation for the composite NiAl2O3-Al2O3 powder. The crucial was to sinter the composite powder and not allow for grain growth of Al2O3 particles.
In the XRD patterns of the Al2O3 powder after consolidation by PPS at various temperatures, only the peaks of Al2O3 are visible. Figure 8 shows the pattern of the sample sintered at 1400 °C as an example. The Vickers hardness measured for compacted samples was used to compare the results and estimate fracture toughness ( Table 2). It was found that the Al2O3 composite samples sintered in the higher temperature (1500 °C) exhibited hardness equal to 15.3 ± 0.87 GPa. The hardness value determined for the samples sintered at 1400 °C amounted to 15.1 ± 0.8 GPa. It may be noted that the hardness values achieved for both samples are similar (values are in the range of measured errors). The relative density of both specimens remained relatively high. Samples sintered at 1400 °C were characterized by a relative density around 97% of the theoretical density, while samples sintered at 1500 °C presented a relative density close to 100% of the theoretical density. The measured values of densities for samples are shown in Table 3. The difference in density between the samples appeared to have no significant effect on the hardness results. For Al2O3 samples sintered at lower temperatures, the quality of the consolidation process was too poor, which made hardness testing impossible.
Vickers indentation fracture toughness measurements determined the fracture toughness (KIC-critical stress intensity factor). For a thorough comparative analysis, several equations, summarized in Table 2, have been used to calculate KIC values. While the Niihara and Anstis equations apply to the median type of crack, the Lankford equation can be used for any kind of crack. Unified symbol designations were used in all equations applied to determine KIC coefficient values. Therefore "E" corresponds to Young's modulus, "HV" to hardness, and "F" to the load applied, "a" is half-length of the diagonal of the Vickers indentation, and through "c" the crack length from the center of the indentation to the crack tip is given.
The results of the indentation fracture toughness analysis are presented in Figure 9. The critical stress intensity factor (KIC) values depend strongly on the equation applied for the calculations. However, it is noticeable that the general tendency among the KIC values of the examined series remains constant. Regardless of the equation used for the accounting, higher fracture toughness was characterized by samples sintered at a lower temperature. The KIC values for the specimen sintered at 1400 °C were in the range 4.53-11.81 The Vickers hardness measured for compacted samples was used to compare the results and estimate fracture toughness ( Table 2). It was found that the Al 2 O 3 composite samples sintered in the higher temperature (1500 • C) exhibited hardness equal to 15.3 ± 0.87 GPa. The hardness value determined for the samples sintered at 1400 • C amounted to 15.1 ± 0.8 GPa. It may be noted that the hardness values achieved for both samples are similar (values are in the range of measured errors). The relative density of both specimens remained relatively high. Samples sintered at 1400 • C were characterized by a relative density around 97% of the theoretical density, while samples sintered at 1500 • C presented a relative density close to 100% of the theoretical density. The measured values of densities for samples are shown in Table 3. The difference in density between the samples appeared to have no significant effect on the hardness results. For Al 2 O 3 samples sintered at lower temperatures, the quality of the consolidation process was too poor, which made hardness testing impossible.
Vickers indentation fracture toughness measurements determined the fracture toughness (K IC -critical stress intensity factor). For a thorough comparative analysis, several equations, summarized in Table 2, have been used to calculate K IC values. While the Niihara and Anstis equations apply to the median type of crack, the Lankford equation can be used for any kind of crack. Unified symbol designations were used in all equations applied to determine K IC coefficient values. Therefore "E" corresponds to Young's modulus, "HV" to hardness, and "F" to the load applied, "a" is half-length of the diagonal of the Vickers indentation, and through "c" the crack length from the center of the indentation to the crack tip is given.
The results of the indentation fracture toughness analysis are presented in Figure 9. The critical stress intensity factor (K IC ) values depend strongly on the equation applied for the calculations. However, it is noticeable that the general tendency among the K IC values of the examined series remains constant. Regardless of the equation used for the accounting, higher fracture toughness was characterized by samples sintered at a lower temperature. The K IC values for the specimen sintered at 1400 • C were in the range 4.53-11.81 MPa·m 0.5 , while the values for the sample sintered at 1500 • C varied from 3.65 to 9.78 MPa·m 0.5 .
Regarding samples made from Al 2 O 3 ceramics, Maiti et al. [45], in their study, achieved comparable hardness values ranging from 9-13.2 GPa, depending on the sintering time. The samples were made by solid-state sintering after initial uniaxial pressing with polyvinyl alcohol (PVA). They were characterized by lower relative density values than the samples in this paper (96-98%) [45]. Hardness values similar to those presented in this study were also obtained by Ouyang et al., which were uniaxially pressed spheroidal Al 2 O 3 powder sintered at 1550 • C. Depending on the holding time at the sintering temperature, the hardness of the samples was in the range of 16-18 GPa, with relative densities not exceeding 96% of the theoretical density [46].
In the case of literature reports on the hardness of Al 2 O 3 ceramics prepared by the field-assisted sintering technique/spark plasma sintering (FAST/SPS), the hardness values obtained are significantly higher at similar densities. In the work of Xu et al., where Al 2 O 3 was sintered with the current-assisted sintering (ACS) method, which is a modification of the classical spark plasma sintering (SPS), at a pressure of 30 MPa, the hardness values obtained at 1400 • C and 1500 • C were 19.24 GPa and 17.76 GPa, respectively [47]. Similar values were also brought in the work of Yuan et al., where oscillatory pressure sintering was used to obtain samples with densities above 95% and hardness in the range 17-23 GPa [48].
Due to the different methods for determining fracture toughness and, in the case of the Vickers indentation test, the high dependence of the results obtained on the choice of the equation, it is difficult to compare the data available in the literature regarding fracture toughness. However, it is worth mentioning that the fracture toughness determined in the work of Maiti et al. [45] with comparable hardness values was lower than in the present study. The value calculated with the Anstis equation was in the range 5.   The results of the helium pycnometer measurements and Archimedes measurements were presented in Table 3. It was observed that with the increasing temperature of the sintering process, the relative density of the samples increased. Analysis of the results obtained based on the Archimedes method indicated that the lowest relative density 66.58% values were observed for samples prepared at the lowest temperature of sintering (1000 °C). The highest open porosity value also characterized the same sample among all examined ones. On the other hand, the highest values were observed for the samples prepared at 1500 °C. They were close to the total density (99.99%). The literature reports the results of PPS sintering of intermetallic and ceramics materials where very dense structures are obtained [55]. In our own (not published) preliminary experiments of ceramics The results of the helium pycnometer measurements and Archimedes measurements were presented in Table 3. It was observed that with the increasing temperature of the sintering process, the relative density of the samples increased. Analysis of the results obtained based on the Archimedes method indicated that the lowest relative density 66.58% values were observed for samples prepared at the lowest temperature of sintering (1000 • C). The highest open porosity value also characterized the same sample among all examined ones. On the other hand, the highest values were observed for the samples prepared at 1500 • C. They were close to the total density (99.99%). The literature reports the results of PPS sintering of intermetallic and ceramics materials where very dense structures are obtained [55]. In our own (not published) preliminary experiments of ceramics consolidation by PPS, the high density of samples was noticed too.
Porosity and water absorption of the produced samples decreased with increasing density according to the well-known rule [56]. Table 3 does not show the results of open porosity and water absorption for the specimen sintered at 1500 • C because this material had open porosity close to zero and consequently did not exhibit any water absorption. The SEM images ( Figure 10) obtained in secondary electron mode show that the samples' characteristic areas of Al 2 O 3 fractures of the samples sintered at different temperatures. The observations were carried out at fracture sites. SEM results revealed that the Al 2 O 3 samples sintered at relatively low temperatures ranging from 1000 • C to 1300 • C are characterized by regular-shaped alumina particles with fractures (Figure 10a-d). In SEM images, the boundaries of all grains of the ceramic are visible, which confirms the intergranular fracturing and means that in this range of temperatures, the sintering process is not sufficient. This observation can be related to density measurements, which indicate that the lowest density is achieved during the sintering at temperatures 1000-1300 • C (Table 3).
Moreover, the histograms shown in Figure 11a-d revealed a high percentage contribution of small (0.02-0.38 µm) grains in the composites sintered at temperatures 1000 • C-1300 • C. Small grains in the bulk ceramics confirm no sufficient sintering process in this range of temperatures. SEM observation of the fractures of the Al 2 O 3 samples sintered at temperatures 1400 • C (Figure 10e) and 1500 • C (Figure 10f) revealed complete sintering of the alumina grains, which is confirmed by the high-density results (Table 3) and low percentage contribution of alumina grains with a size corresponding to the size of the initial powder (Figure 11e,f). bution of small (0.02-0.38 µm) grains in the composites sintered at temperatures 1000 °C -1300 °C. Small grains in the bulk ceramics confirm no sufficient sintering process in this range of temperatures. SEM observation of the fractures of the Al2O3 samples sintered at temperatures 1400 °C (Figure 10e) and 1500 °C (Figure 10f) revealed complete sintering of the alumina grains, which is confirmed by the high-density results (Table 3) and low percentage contribution of alumina grains with a size corresponding to the size of the initial powder (Figure 11e-f). The next step was performed a stereological analysis based on the observation of fracture to determine the effect of the PPS process temperature on Al2O3 grain growth. The histograms of Al2O3 grain distribution in individual sinters ( Figure 11) indicate that, along with the increasing temperature of the PPS process, the percentage amount of the larger particles in the microstructure increases. Consideration of the results described above related to the Al2O3 powder compacted by PPS allows us to select the temperature of 1400 °C as the sintering temperature for the NiAl-Al2O3 composite powders. Although the density of compacted Al2O3 was only 97% at this temperature, the hardness was as high as for the sample compacted at 1500 °C, and the high KIC value was obtained. Moreover, by choosing the temperature of 1400 °C decided that the ceramic grains did not grow for 1500 °C. Figure 12 presents examples of SEM images of the NiAl-10%Al2O3 and NiAl-20%Al2O3 bulk specimens. The observation demonstrated that dark areas in the micro- Consideration of the results described above related to the Al 2 O 3 powder compacted by PPS allows us to select the temperature of 1400 • C as the sintering temperature for the NiAl-Al 2 O 3 composite powders. Although the density of compacted Al 2 O 3 was only 97% at this temperature, the hardness was as high as for the sample compacted at 1500 • C, and the high K IC value was obtained. Moreover, by choosing the temperature of 1400 • C decided that the ceramic grains did not grow for 1500 • C. Figure 12 presents examples of SEM images of the NiAl-10%Al 2 O 3 and NiAl-20%Al 2 O 3 bulk specimens. The observation demonstrated that dark areas in the microstructure stand for the Al 2 O 3 , while the brightly grey areas correspond to the NiAl matrix. It was found that the surface of the polished samples was free of cracks and pores, which evidences the good quality of consolidation of specimens. SEM observations also showed that the samples had a homogeneous microstructure. Al 2 O 3 inclusions are present in the NiAl matrix.

Characterization of NiAl-Al 2 O 3 Powder Compacted by PPS
Based on the measurements from the pycnometer, it was found that a relative density of 97.9% characterized the samples containing 10 wt.% Al 2 O 3 while the sample containing 20 wt.% Al 2 O 3 was characterized by a relative density close to 100%. Based on density measurements by pycnometer, it was found that the relative density increases with the increase of Al 2 O 3 content in the sinter. Subsequently, energy-dispersive X-ray spectroscopy (JEOL Ltd., Tokyo, Japan) was conducted to obtain the elemental distribution maps of the prepared NiAl-10%Al2O3 and NiAl-20%Al2O3 samples. The maps obtained are exhibited in Figure 13. The chemical element distribution maps revealed a non-uniform presence of aluminum, nickel, oxygen, and iron. Contamination of powders by iron from steel milling tools is generally recognized in mechanical alloying processes [28].  Figure 14 presents the XRD patterns of the composite NiAl-Al2O3 powders after consolidation. Only the diffraction peaks of NiAl and Al2O3 are present, which shows that no phase changes occurred during the PPS process. The NiAl diffraction profiles' widths are smaller than the NiAl peaks' breadths in the powders. This sharpening of the peaks indi- Subsequently, energy-dispersive X-ray spectroscopy (JEOL Ltd., Tokyo, Japan) was conducted to obtain the elemental distribution maps of the prepared NiAl-10%Al 2 O 3 and NiAl-20%Al 2 O 3 samples. The maps obtained are exhibited in Figure 13. The chemical element distribution maps revealed a non-uniform presence of aluminum, nickel, oxygen, and iron. Contamination of powders by iron from steel milling tools is generally recognized in mechanical alloying processes [28]. Subsequently, energy-dispersive X-ray spectroscopy (JEOL Ltd., Tokyo, Japan) was conducted to obtain the elemental distribution maps of the prepared NiAl-10%Al2O3 and NiAl-20%Al2O3 samples. The maps obtained are exhibited in Figure 13. The chemical element distribution maps revealed a non-uniform presence of aluminum, nickel, oxygen, and iron. Contamination of powders by iron from steel milling tools is generally recognized in mechanical alloying processes [28].  Figure 14 presents the XRD patterns of the composite NiAl-Al2O3 powders after consolidation. Only the diffraction peaks of NiAl and Al2O3 are present, which shows that no phase changes occurred during the PPS process. The NiAl diffraction profiles' widths are smaller than the NiAl peaks' breadths in the powders. This sharpening of the peaks indi-  Figure 14 presents the XRD patterns of the composite NiAl-Al 2 O 3 powders after consolidation. Only the diffraction peaks of NiAl and Al 2 O 3 are present, which shows that no phase changes occurred during the PPS process. The NiAl diffraction profiles' widths are smaller than the NiAl peaks' breadths in the powders. This sharpening of the peaks indicates grain growth in the NiAl phase during consolidation. The mean crystallite size estimated by the Williamson-Hall method was 112 nm and 123 nm for the sample containing 10% and 20% of Al 2 O 3, respectively. These values exceed 100 nm, so they are beyond the applicability limit of the Williamson-Hall method and may be affected by an error. However, it can be concluded that the NiAl phase has a submicrometer grain size. For the hardness measurements of NiAl-Al2O3 composite samples sintered at 1400 °C, the hardness values achieved were naturally significantly lower than those for pure Al2O3 ceramics; however, for both the composite with lower (10 wt.% Al2O3) and higher (20 wt.% Al2O3) Al2O3 content, the HV values obtained were practically identical. They were equal to 5.8 ± 0.08 GPa and 5.8 ± 0.09 GPa, respectively. Interestingly, regarding the composite specimens in the research conducted by Marek Krasnowski et al. [32] on NiAl-B composites with nanocrystalline intermetallic matrix fabricated through mechanical alloying and consolidation, hardness results ranging from 10.3 GPa to 12.6 GPa were achieved. The lower hardness values observed in the present experiment are probably related to the growth of grains in the NiAl phase compared to the nanocrystalline grains observed in [32]. In the aforementioned study [32], the nanocrystalline structure was observed in the samples after consolidation. Another research on nanocrystalline NiAl carried out by the same research team showed that the presence of a nanocrystalline phase significantly exceeds the hardness compared to microcrystalline NiAl alloys. The hardness obtained for nanocrystalline NiAl amounted to 9.53 GPa [43]. According to literature data, the addition of Al2O3 to NiAl favorably affects the hardness of the material. Michalski et al. [55], in their work investigating the effect of Al2O3 on NiAl fabricated by the PPS method with the participation of the SHS reaction, reported a noticeable increase in hardness with increasing Al2O3 content. In the aforementioned work [55], the hardness of AlNi-Al2O3 composite with 38 vol.% Al2O3 was about 6.08 GPa, while the hardness of NiAl produced with the same process was equal to 4.22 GPa.
Due to their high elasticity, the composite specimens did not fracture under the indentor impact, which can be seen in the exemplary indentations presented in Figure 15. Therefore, it was not possible to determine KIC values for these specimens using the method applied in this research. For the hardness measurements of NiAl-Al 2 O 3 composite samples sintered at 1400 • C, the hardness values achieved were naturally significantly lower than those for pure Al 2 O 3 ceramics; however, for both the composite with lower (10 wt.% Al 2 O 3 ) and higher (20 wt.% Al 2 O 3 ) Al 2 O 3 content, the HV values obtained were practically identical. They were equal to 5.8 ± 0.08 GPa and 5.8 ± 0.09 GPa, respectively. Interestingly, regarding the composite specimens in the research conducted by Marek Krasnowski et al. [32] on NiAl-B composites with nanocrystalline intermetallic matrix fabricated through mechanical alloying and consolidation, hardness results ranging from 10.3 GPa to 12.6 GPa were achieved. The lower hardness values observed in the present experiment are probably related to the growth of grains in the NiAl phase compared to the nanocrystalline grains observed in [32]. In the aforementioned study [32], the nanocrystalline structure was observed in the samples after consolidation. Another research on nanocrystalline NiAl carried out by the same research team showed that the presence of a nanocrystalline phase significantly exceeds the hardness compared to microcrystalline NiAl alloys. The hardness obtained for nanocrystalline NiAl amounted to 9.53 GPa [43]. According to literature data, the addition of Al 2 O 3 to NiAl favorably affects the hardness of the material. Michalski et al. [55], in their work investigating the effect of Al 2 O 3 on NiAl fabricated by the PPS method with the participation of the SHS reaction, reported a noticeable increase in hardness with increasing Al 2 O 3 content. In the aforementioned work [55], the hardness of AlNi-Al 2 O 3 composite with 38 vol.% Al 2 O 3 was about 6.08 GPa, while the hardness of NiAl produced with the same process was equal to 4.22 GPa.
Due to their high elasticity, the composite specimens did not fracture under the indentor impact, which can be seen in the exemplary indentations presented in Figure 15. Therefore, it was not possible to determine K IC values for these specimens using the method applied in this research.

Conclusions
This work aimed to recognize the PPS method as a possible technique to produce bulk composite materials from NiAl-Al2O3 composite powder. The composite powder consisted of intermetallic NiAl, and Al2O3 was prepared by mechanical alloying. As an initial powder Ni, Al and Al2O3 were used.
As a preliminary study, the Al2O3 was consolidated by PPS and characterized. These experiments gave an important input in order to describe the sintering process in function of temperature. Mainly based on these results, the sintering temperature of the composite powders, which did not cause intensive grain growth, was estimated.
The results obtained revealed that the PPS method allows for the consolidation of ceramic and composite powders. In consolidated NiAl-Al2O3 powder, the bulk materials are generally set up in two phases: Al2O3 is located within the NiAl matrix. The intermetallic ceramic composites have relative densities: for composites with 10 wt.% Al2O3 97.9% and samples containing 20 wt.% Al2O3 close to 100%. The hardness of both composites was equal to 5.8 GPa. Moreover, after PPS consolidation, NiAl-Al2O3 composites are characterized by high plasticity.
The results obtained are promising for further study of the consolidation of composite NiAl-Al2O3 powder with various initial contributions of ceramics (Al2O3) to prepare composite powder by mechanical alloying. Moreover, PPS could be applied to the consolidation of a mixture of intermetallic-ceramic composite powders with the addition of ceramics. This last option, when an amount of 50 or more % of Al2O3 will be mixed with NiAl-Al2O3 composite powder, can lead to the production of a new class of ceramic-intermetallic composites, with complex structure, i.e., Al2O3 ceramic matrix with areas of NiAl with fine particles of Al2O3 inside. Such composites should have attractive properties as well. The active contribution and synergistic effect of elements of the structure on the crack propagation and finally on improving the fracture toughness of the new composite are expected. Our work in this area is in progress.

Conclusions
This work aimed to recognize the PPS method as a possible technique to produce bulk composite materials from NiAl-Al 2 O 3 composite powder. The composite powder consisted of intermetallic NiAl, and Al 2 O 3 was prepared by mechanical alloying. As an initial powder Ni, Al and Al 2 O 3 were used.
As a preliminary study, the Al 2 O 3 was consolidated by PPS and characterized. These experiments gave an important input in order to describe the sintering process in function of temperature. Mainly based on these results, the sintering temperature of the composite powders, which did not cause intensive grain growth, was estimated.
The results obtained revealed that the PPS method allows for the consolidation of ceramic and composite powders. In consolidated NiAl-Al 2 O 3 powder, the bulk materials are generally set up in two phases: Al 2 O 3 is located within the NiAl matrix. The intermetallic ceramic composites have relative densities: for composites with 10 wt.% Al 2 O 3 97.9% and samples containing 20 wt.% Al 2 O 3 close to 100%. The hardness of both composites was equal to 5.8 GPa. Moreover, after PPS consolidation, NiAl-Al 2 O 3 composites are characterized by high plasticity.
The results obtained are promising for further study of the consolidation of composite NiAl-Al 2 O 3 powder with various initial contributions of ceramics (Al 2 O 3 ) to prepare composite powder by mechanical alloying. Moreover, PPS could be applied to the consolidation of a mixture of intermetallic-ceramic composite powders with the addition of ceramics. This last option, when an amount of 50 or more % of Al 2 O 3 will be mixed with NiAl-Al 2 O 3 composite powder, can lead to the production of a new class of ceramic-intermetallic composites, with complex structure, i.e., Al 2 O 3 ceramic matrix with areas of NiAl with fine particles of Al 2 O 3 inside. Such composites should have attractive properties as well. The active contribution and synergistic effect of elements of the structure on the crack propagation and finally on improving the fracture toughness of the new composite are expected. Our work in this area is in progress.